<%BANNER%>

Development of the MCNPX Depletion Capability

Permanent Link: http://ufdc.ufl.edu/UFE0021946/00001

Material Information

Title: Development of the MCNPX Depletion Capability A Monte Carlo Linked Depletion Method that Automates the Coupling between MCNPX and Cinder90 for High Fidelity Burnup Calculations
Physical Description: 1 online resource (188 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: afci, burnup, carlo, depletion, gnep, mcnp, mcnpx, monte, reactor, transmutation
Nuclear and Radiological Engineering -- Dissertations, Academic -- UF
Genre: Nuclear Engineering Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Monte Carlo-linked depletion methods have gained recent interest due to the ability to more accurately model complex 3-dimesional geometries and better track the evolution of temporal nuclide inventory by simulating the actual physical process utilizing continuous energy coefficients. The integration of CINDER90 into the MCNPX Monte Carlo radiation transport code provides a high-fidelity completely self-contained Monte-Carlo-linked depletion capability in a well established, widely accepted Monte Carlo radiation transport code that is compatible with most nuclear criticality (KCODE) particle tracking features in MCNPX. MCNPX depletion tracks all necessary reaction rates and follows as many isotopes as cross section data permits in order to achieve a highly accurate temporal nuclide inventory solution. This work chronicles relevant nuclear history, surveys current methodologies of depletion theory, details the methodology in applied MCNPX and provides benchmark results for three independent OECD/NEA benchmarks. Relevant nuclear history, from the Oklo reactor two billion years ago to the current major United States nuclear fuel cycle development programs, is addressed in order to supply the motivation for the development of this technology. A survey of current reaction rate and temporal nuclide inventory techniques is then provided to offer justification for the depletion strategy applied within MCNPX. The MCNPX depletion strategy is then dissected and each code feature is detailed chronicling the methodology development from the original linking of MONTEBURNS and MCNP to the most recent public release of the integrated capability (MCNPX 2.6.F). Calculation results of the OECD/ NEA Phase IB benchmark, H. B. Robinson benchmark and OECD/ NEA Phase IVB are then provided. The acceptable results of these calculations offer sufficient confidence in the predictive capability of the MCNPX depletion method. This capability sets up a significant foundation, in a well established and supported radiation transport code, for further development of a Monte Carlo-linked depletion methodology which is essential to the future development of advanced reactor technologies that exceed the limitations of current deterministic based methods.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Anghaie, Samim.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021946:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021946/00001

Material Information

Title: Development of the MCNPX Depletion Capability A Monte Carlo Linked Depletion Method that Automates the Coupling between MCNPX and Cinder90 for High Fidelity Burnup Calculations
Physical Description: 1 online resource (188 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: afci, burnup, carlo, depletion, gnep, mcnp, mcnpx, monte, reactor, transmutation
Nuclear and Radiological Engineering -- Dissertations, Academic -- UF
Genre: Nuclear Engineering Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Monte Carlo-linked depletion methods have gained recent interest due to the ability to more accurately model complex 3-dimesional geometries and better track the evolution of temporal nuclide inventory by simulating the actual physical process utilizing continuous energy coefficients. The integration of CINDER90 into the MCNPX Monte Carlo radiation transport code provides a high-fidelity completely self-contained Monte-Carlo-linked depletion capability in a well established, widely accepted Monte Carlo radiation transport code that is compatible with most nuclear criticality (KCODE) particle tracking features in MCNPX. MCNPX depletion tracks all necessary reaction rates and follows as many isotopes as cross section data permits in order to achieve a highly accurate temporal nuclide inventory solution. This work chronicles relevant nuclear history, surveys current methodologies of depletion theory, details the methodology in applied MCNPX and provides benchmark results for three independent OECD/NEA benchmarks. Relevant nuclear history, from the Oklo reactor two billion years ago to the current major United States nuclear fuel cycle development programs, is addressed in order to supply the motivation for the development of this technology. A survey of current reaction rate and temporal nuclide inventory techniques is then provided to offer justification for the depletion strategy applied within MCNPX. The MCNPX depletion strategy is then dissected and each code feature is detailed chronicling the methodology development from the original linking of MONTEBURNS and MCNP to the most recent public release of the integrated capability (MCNPX 2.6.F). Calculation results of the OECD/ NEA Phase IB benchmark, H. B. Robinson benchmark and OECD/ NEA Phase IVB are then provided. The acceptable results of these calculations offer sufficient confidence in the predictive capability of the MCNPX depletion method. This capability sets up a significant foundation, in a well established and supported radiation transport code, for further development of a Monte Carlo-linked depletion methodology which is essential to the future development of advanced reactor technologies that exceed the limitations of current deterministic based methods.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Anghaie, Samim.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021946:00001


This item has the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101111_AAAABG INGEST_TIME 2010-11-11T13:17:55Z PACKAGE UFE0021946_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 50961 DFID F20101111_AAAWAE ORIGIN DEPOSITOR PATH fensin_m_Page_161.pro GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
226bcccde184c060d34fea3dd109019f
SHA-1
952cba753b33ece021acb28e394e4aefeea71a9b
21658 F20101111_AAAURH fensin_m_Page_038.QC.jpg
bf75c9acdcafaaa6bf6c89aecff4ac97
5cae07c31e28f9faaac90eda40c212dadc469b54
1053954 F20101111_AAAVUJ fensin_m_Page_152.tif
a3b34dfa3e0c86f652f6821c9787ed63
7a567cef0c561e23c3a9f99989b4bee0322e146e
F20101111_AAAVTV fensin_m_Page_134.tif
c63ec1dc3495e4d4e8bd687f1df48ffe
cdb2a812c0a8f3d462256541e09781410df338d4
50949 F20101111_AAAWAF fensin_m_Page_162.pro
8c7a107816debf5f73f23e0023abf31d
13f870c7be48e4e504dafcbe4bec0142037a9194
57165 F20101111_AAAURI fensin_m_Page_106.jpg
c34a513b13d6aee2f54af8df5abc332e
98055ff094cfe5c7b08e08160bbbe50b31f4caa7
25271604 F20101111_AAAVUK fensin_m_Page_155.tif
d6812c161a6b658fbe2e5d734cf70e83
741aaad28682851b30314b710a2ebcf25a72ab61
F20101111_AAAVTW fensin_m_Page_135.tif
a5efdd95d5e277e4cd255861a3d335d6
a1d5a64c12df0b785183eb5ed5b1af15eeefbd21
50978 F20101111_AAAWAG fensin_m_Page_163.pro
4d8c4d03dc5f215a2ab6c762be83e96e
573928654d99daa45c6287363c1396817a3e0add
F20101111_AAAVUL fensin_m_Page_156.tif
6c13bdc053a062e25de0ae865f10efb4
70f4c3ad43534feb76a448b497ebbddb4df27748
F20101111_AAAVTX fensin_m_Page_136.tif
e97e6002b245a4d49db32992f9359c7b
1d44e3df95b7991b1c773f77a49c159c459ac0d1
F20101111_AAAVVA fensin_m_Page_176.tif
c7d38ed057f75b034c869cced5a1fc1b
1a8a43156addcdfd032a3c84de1ea77b56203951
2251 F20101111_AAAURJ fensin_m_Page_115.txt
521b46fad36f81e7ded2d9228c3faa4b
3af3c5a51fcaf750d20895a69b32bd58eb5e4ef0
54298 F20101111_AAAWAH fensin_m_Page_165.pro
16cccad9a0cfb9a7f3724db1f5f02c83
cf487061132c8022f34628ae9bf567ab96dc61a8
F20101111_AAAVUM fensin_m_Page_157.tif
9e29569bd9164cfe608b12d7a46ed275
2a4a4be867bc6eb8cd1c6b97dfc0637687b927ee
F20101111_AAAVTY fensin_m_Page_137.tif
535aa8f64d9b9eb6ca6f3b7d3370c99f
a77df10e3b2907da6b45182e9c2de1db2aecfec8
F20101111_AAAVVB fensin_m_Page_177.tif
2b5fb45c47151cf14eb7b093e5cd53f9
cec5c60e59aad83afd38dd53755ddbffb80cab11
122992 F20101111_AAAURK fensin_m_Page_050.jp2
9012ca38f3ff73419afc0bb04fc1168a
fd141a79615a4c4048c8038e3ba83349fdc8e0ed
42262 F20101111_AAAWAI fensin_m_Page_166.pro
1996be4f722391cf3116c108e25b25c3
9f2e379ea0f7d2b6085c6fff7efb45e5e3e39f6b
F20101111_AAAVUN fensin_m_Page_158.tif
46d62eacd2a7be1c0776e0d5726509dc
763693ce085b4cb8626283472e635df07f370288
F20101111_AAAVTZ fensin_m_Page_138.tif
1598439a00c75cb27ba2a4294eb83950
0deca518531447e50f76f01119c9823814976458
6883 F20101111_AAAUSA fensin_m_Page_128thm.jpg
0b88dea64a4c79153bbe96dc2e997ea8
b57b0e74d5890df9f5e037c2622bdf361cfea7ab
F20101111_AAAVVC fensin_m_Page_178.tif
6adcda8acca757c3acd8f4db103bc29e
715091f7b24ab5b55dbaa8023a986b0ca725aceb
5421 F20101111_AAAURL fensin_m_Page_072thm.jpg
66c8675eb90fd549dfb54a84dcfe762b
2170f9c361424d0bf2c1789f12c9f86a747697ad
50837 F20101111_AAAWAJ fensin_m_Page_167.pro
4384c9c5ecedc3bc2a2cf6830cf37a7d
46c88104021d83f0db30f3edf757dd4ea6f4d1e5
F20101111_AAAVUO fensin_m_Page_159.tif
2f101779bac1bd5293178bbb8ef357b3
2f91a5035910433a68ed88f3c40c711a7f621ac1
24818 F20101111_AAAUSB fensin_m_Page_016.QC.jpg
144173734ad6206f7b054c449424bca9
861e71297ef9f8e0721fd97d95805f82395ef2a7
F20101111_AAAVVD fensin_m_Page_179.tif
addc24fff05fb5c395c0b3eb88182eb1
8c0fcda03b36a6a7f5ea9842c4fd0a1459946586
2813 F20101111_AAAURM fensin_m_Page_168.txt
7f2c77717946670e86a2be076151e0a1
be6bb1b14e7431b2798c0d8072a1763657c0aef0
63560 F20101111_AAAWAK fensin_m_Page_168.pro
b2c0decdf215ef48274ae2afcb1e279f
46c0f2f98c4aa06b213ecc203826efde1929c460
F20101111_AAAVUP fensin_m_Page_160.tif
203d475ca84cc6ff76bb98795e65bd10
e4d291427d63703363e9e6ab26ac9f7be225b67b
121818 F20101111_AAAURN fensin_m_Page_066.jp2
10467b842477c227227e8c991bf892db
eb8c6b9e86efdbe8b97d8effceaea3fa753a6d98
58266 F20101111_AAAWAL fensin_m_Page_169.pro
a42626192fd8e3ddc5df052175eeac91
4dba63c8e8d6078f205001cb8e3b21e21f148829
F20101111_AAAVUQ fensin_m_Page_161.tif
a7fd49fe54bb3b74f1694fb5775f58e9
a4c999c889383604a18e0ae63be293880be86afd
53544 F20101111_AAAUSC fensin_m_Page_016.pro
6dde042b38fd8ffefec0265a6b090edb
32ef6b32e547d4135242497ab0b2cc1c93643cba
F20101111_AAAVVE fensin_m_Page_180.tif
e26c90d8627338a3a4d462470aa5b356
d158b9dae47377b83cbbb4fa42930328fae22a3f
80290 F20101111_AAAURO fensin_m_Page_100.jpg
a3ed2ea154f5be1a3fdb88d036bea6e2
f770922b5f298fed28a642511d2ffd3caeb233cb
2748 F20101111_AAAWBA fensin_m_Page_008.txt
f582af0db97555ce209dacc616e6a125
49ce852898d7b4856e01b7ebc826b77794799577
52573 F20101111_AAAWAM fensin_m_Page_172.pro
2641e997cbe64bb7f06c117b456edf9d
98529fcbb5545f1a8a8bb29dec4121492d1aef28
F20101111_AAAVUR fensin_m_Page_162.tif
061f84613616aac92ac0195ccf5d9080
727273a34e3b6bfe4efa4dad15592bf5dc63e869
2509 F20101111_AAAUSD fensin_m_Page_133.txt
400ecb05046529dfce743c6942a37a1b
50df2caf7c1993a48ae9df0f5c2053d0cabc1613
F20101111_AAAVVF fensin_m_Page_181.tif
d81232b16052ded6dffe8ef0a926472b
84678dcac916efd64d47b6e1842cb2fc70ebe768
45436 F20101111_AAAURP fensin_m_Page_109.pro
c92f2b0e6ee558238847e003a9f25eb7
1b011e09159294cfc6805b745b5ef32d7febd316
3037 F20101111_AAAWBB fensin_m_Page_010.txt
1d6b45705c5b63cf94b979a18809df73
d770ef7c2ac40a22cb8fbefdb910598a31bd0b66
57958 F20101111_AAAWAN fensin_m_Page_175.pro
4f5d866295450a3600591b59716118ef
dc0b07c8ff99c2683db409a3c0130a6aa6f4acbd
F20101111_AAAVUS fensin_m_Page_163.tif
3c333d1db5dbf9bba92baaf943b57bac
00071c5dea0aee2f68575c86d3e7188dc3f1e0e9
81609 F20101111_AAAUSE fensin_m_Page_025.jpg
eff36c1444541fdda4ccd7246b4ccaf4
1f48f632f1476d070cdd2d73aff09701c2f37a1d
F20101111_AAAVVG fensin_m_Page_182.tif
2f75e38b95bfda07d9631856a6c5d91e
b6ca2fc86f8188856e69e9cc46a685657400b304
F20101111_AAAURQ fensin_m_Page_174.tif
b5ceb06a875ce60bd6b21940252d8b6e
4cfadbf0f744e3a2b2454fbc05a4553331e0bc02
2082 F20101111_AAAWBC fensin_m_Page_011.txt
85d3beaad459a2bd3d0335fe77ec1286
5146d71d15a1de8acd93aaf39176a34d274d42b3
F20101111_AAAVUT fensin_m_Page_164.tif
a0d913104e7d31d277c2c32fc1d95fad
05dd438c68038834892348a63022f9058dcb909d
120060 F20101111_AAAUSF fensin_m_Page_093.jp2
e6d6b1d8003c5346fd6d017332b4dd57
96280a698142d5d70719cc04e56a558293638431
F20101111_AAAVVH fensin_m_Page_183.tif
3b6cf177a8ae296c9a430d15b5e966dd
0f5216c4a22d6411045c886a79bd27ce0826b75f
F20101111_AAAURR fensin_m_Page_165.tif
d109bf58a94daa78092b8372f3882e2a
376ed66eed680a6d5585556b7885891e91e8be8e
2072 F20101111_AAAWBD fensin_m_Page_012.txt
39132d8a8e4180ef13ad98190bb9aa04
49c3b32059eeb345fc7280c43fa719fe48dd36ed
16460 F20101111_AAAWAO fensin_m_Page_176.pro
0a23ab4d7f6a907a2eb38ac8a6c44ff9
857f698b5eb791d569d0085118c8b336dce35364
1717 F20101111_AAAUSG fensin_m_Page_188.txt
c4bf806a0cec5fc70019e633673f3826
acf0201ace346df4e2793ef1cdb7305ae323834f
F20101111_AAAVVI fensin_m_Page_185.tif
877e9dcd92b80acee8231ede31bbad8e
6ee6ecd83b7a95f6d32b750384a5cf63eb003e54
80867 F20101111_AAAURS fensin_m_Page_103.jpg
543aa8773ab18cd6e33cf7366e0453de
3a1cddcb663220077b870b98f80d29112a7ec444
F20101111_AAAVUU fensin_m_Page_166.tif
0c27af97ad6f9189051ad8ec0a453d87
76df3fecc46239a002f2462a4549ac20080be15a
941 F20101111_AAAWBE fensin_m_Page_013.txt
d191f66e63439af88f8edcaa0eb29b6a
2e15ff92f5f4ccd3cb71ab6baacf26ce8a880a4e
39268 F20101111_AAAWAP fensin_m_Page_177.pro
591f05c8922d1ef6d2bae4678ea6f90d
fbb73a5d7227fc864b237b29383c4343feb4fdf9
26258 F20101111_AAAUSH fensin_m_Page_142.QC.jpg
c4268cd8d7726d71021a040caba8ffd0
9a6dd67ba5a3418cf7fb6b18d86a48b093bbacc6
F20101111_AAAVVJ fensin_m_Page_186.tif
b329c85b139f6e1b8ff377c41e262939
fdb710b974b69c13144b9de85a57e87f0dfcdf69
61110 F20101111_AAAURT fensin_m_Page_121.pro
3de64e9f537d1f472fc2663f8832e070
25ac4343d236086ae54166161fab22016dd1521a
F20101111_AAAVUV fensin_m_Page_167.tif
23357fe3c70bfc0133d38132694fd94e
e2ad2afefd4757203d2f03d40690bd1e2fa977b4
2128 F20101111_AAAWBF fensin_m_Page_014.txt
8ed4080f090cb6ab46b9f1da4a588877
fc57efe5582a9cd3a15aa55e5e446f7f72c61313
49846 F20101111_AAAWAQ fensin_m_Page_178.pro
e234e5abce0ef6512106c176e2c628d9
309aa3b53bc590cc83cb2e0c402ddb7851fe64c2
47787 F20101111_AAAUSI fensin_m_Page_039.pro
c034edf36fc1ae2126340f1ab925445f
15a8308a29b5e224b7019d88decbef49d5732d5e
F20101111_AAAVVK fensin_m_Page_187.tif
13dd5575f321ee4dcc6a7519a9fe77d3
5764ac2f6e7f38e2ee90d71f3d73ba29f5b15048
122855 F20101111_AAAURU fensin_m_Page_134.jp2
807292b564a9f2b022f8016b953ae7a9
1e8ed18366ba891bcd89a9796dd3304a10b8f1d0
F20101111_AAAVUW fensin_m_Page_168.tif
6d17545775053ffb754d662a8ae3f374
d133cd9c12ec9f07a0cca88afe932f54ae162b9b
2102 F20101111_AAAWBG fensin_m_Page_016.txt
96dc0f1c68cb62c6df1af83a1e324eec
582ebe6cf33e3025b03c6a9f07f17f819e0b68c6
56823 F20101111_AAAWAR fensin_m_Page_179.pro
56bc94247a672521904c91f135efde92
d1785b98771571095f2f53398e8a9ab7786a0ac5
52732 F20101111_AAAVWA fensin_m_Page_020.pro
208e9ef7749cbe813655db9ab8284083
e42b1a499d4d0e4001716d7692b2c7bc8c34e8d0
1296 F20101111_AAAUSJ fensin_m_Page_092.txt
5732d1c147152ca89f3fdfec9d12503c
387750419a56eb68511942b82170fc39366a1970
F20101111_AAAVVL fensin_m_Page_188.tif
2bf262785077dca67ca4bce44a0bf051
e81bf8c6675d7fb4f0d20a3bdc92edec532b91db
67105 F20101111_AAAURV fensin_m_Page_039.jpg
e7cfc287ac554b9c3d2cd521052b17d4
96ba1e22d7c58ffdf57d241c179cc79f2677490c
F20101111_AAAVUX fensin_m_Page_170.tif
d62dfa063fc7371833f41d7f6f27e098
fbbf9c48834c6b5123c16920649aceec63182f32
2029 F20101111_AAAWBH fensin_m_Page_017.txt
911c6c449f9c7d94427d8bc0d0229924
317e973878b85a5a65163abdb8d7421374f82992
59206 F20101111_AAAWAS fensin_m_Page_181.pro
11290cd2fa5a644f0fa0d2351c871b48
d9d208b4e56acb64679aca7dd037580cabd93a77
53391 F20101111_AAAVWB fensin_m_Page_021.pro
5a02aeaef4a83654e554c7023525a2f0
2535663c4b9ca2584aa045a6ff679eb4f99405af
F20101111_AAAUSK fensin_m_Page_079.tif
2496397d7930bd61806ec0c94e0f979d
f264e8185042cca64142c44eb4c7fa5ec10ce21b
10903 F20101111_AAAVVM fensin_m_Page_001.pro
fe2afcb7433fc0edc30a8ca5072da4e1
aa1fedc5951867a7c7f5828efc9d44baecca1ca0
17892 F20101111_AAAURW fensin_m_Page_162.QC.jpg
620af1321b063e2852e9601e614c1b74
8767b8272e8f4dde0e2ae91717b5b0c2a47882f5
F20101111_AAAVUY fensin_m_Page_171.tif
76d756b227f592edd1f0fb3f5d9de99f
c7b218ad57ae29614d6c09bbeb64f883726f700e
2107 F20101111_AAAWBI fensin_m_Page_018.txt
72e8c0b2ab321ce643a6ce58b8cb18a3
702f0cdc9bf1b05025475539ba8cb9973ab56acc
57365 F20101111_AAAWAT fensin_m_Page_183.pro
886e7b55ac1a52bff77b631b8461fc87
1adad06db9f84a723de26a56fdfc089473826233
55948 F20101111_AAAVWC fensin_m_Page_022.pro
3a7ee4d27b35c8220f577ea6a9d5d1e9
df51ade43757b692f1b3a6d65d4e2184eb0be539
965 F20101111_AAAVVN fensin_m_Page_002.pro
c87d06c7af9601fd5f613fcb6a571d3c
d55a9f9d4c3f8493ab2e26ef5a8c737a33ac1893
76511 F20101111_AAAUTA fensin_m_Page_114.jpg
57e8028da53d49906160c9f496995024
4892c92953c12ef8977140c78aa181a0b8a65745
6896 F20101111_AAAUSL fensin_m_Page_016thm.jpg
31ce72ef4c2595f651ec0127a04fba4d
839ebdd912c2647ce234ec52c37e288d84d4d24e
121967 F20101111_AAAURX fensin_m_Page_121.jp2
ba7c50d6c1001dc69e1a3fc47f200a77
028c0eb24db29adbb86fcd9a6abea9f881ed67fe
F20101111_AAAVUZ fensin_m_Page_175.tif
7922af7cc7ad9d8f84da194bc9bc291e
a487d1da53df66ce79b66e05be854b3d8c9adc9c
2237 F20101111_AAAWBJ fensin_m_Page_019.txt
6daf28f191022b13ff8725166df2afbd
9a86880c22f010bd89006f4e20e10dbed6258cab
70486 F20101111_AAAWAU fensin_m_Page_185.pro
1c0bcd84ccaea95e890917bb1914b5a6
9c3e9ade9be78946de6760dd4b40f512b0da5ef3
54638 F20101111_AAAVWD fensin_m_Page_023.pro
be439627d643ffcd2aaa53b7ba845a14
70ed2e0f3affda5dc15f290e30db46025dc4b9f1
1275 F20101111_AAAVVO fensin_m_Page_003.pro
c47a79baf5d75614ce5f85de2f0e6e5b
c52cb57dba160d5a59c345f0cb46c66fadd6b5bb
2262 F20101111_AAAUTB fensin_m_Page_175.txt
5bcf4c7dc59c703d7728e8076316fe83
58bc8e1ea351b84523f5ddafd2a5e1c97667178f
105 F20101111_AAAUSM fensin_m_Page_003.txt
33b6636b4b6f9184facb4279738adb1d
8f3e193b5bcbab8a3ed88f5c56107473434f27e6
2165 F20101111_AAAURY fensin_m_Page_171.txt
22bec9490bcb0b6162b6552f2edf4a2e
fc45b79b39fd0be462b865d985b5b76be004c4ec
2076 F20101111_AAAWBK fensin_m_Page_020.txt
b63d28c2de0cef11777b95aea5525e3c
8ffbdf15480a0ae03fc0f288220dd57a2b86ac83
63242 F20101111_AAAWAV fensin_m_Page_186.pro
08a42b3ca195126fdea7e268a4dbae08
0243cbeea01c6301d2bc05fbb6e5b4b8b2e44f8e
56234 F20101111_AAAVWE fensin_m_Page_024.pro
2be25e0ca0c12496a1556540f0a227d0
3c803bd7d99b7f295160ea85ae96457c42ef56a5
57554 F20101111_AAAVVP fensin_m_Page_004.pro
a601cc360f5a40743e60baaa6f337f2e
248377e6e80f809a91b04a66b218b2249e17b8c8
F20101111_AAAUTC fensin_m_Page_133.tif
9afd24e220ba9be2245bb4fb04bf7494
3368b96165df29144a86a64de1fd23af34a1aba9
60341 F20101111_AAAUSN fensin_m_Page_180.pro
44941e62a26b3bc7bf38b6359a66241c
8c3cb70121dd4e8eaf2459fd107ffc6f21ca4106
16386 F20101111_AAAURZ fensin_m_Page_057.QC.jpg
0be69b9fcf853308273cb8c92e08a4f4
f6f100fd9d18119146aa731423dcc8a9d6fd8704
2173 F20101111_AAAWBL fensin_m_Page_021.txt
c232d0cd4f7722f3efc8d4c26a1cd730
46769866be3f63d6fd026deafe23e53535bf1102
38444 F20101111_AAAWAW fensin_m_Page_187.pro
578d79fbe566b7aaf5ac076ad60adfc9
c7c2cfae135219b4a76c8e993f7c3d857b486936
3682 F20101111_AAAVVQ fensin_m_Page_005.pro
4222cf84e8880d0dbe85d4edef3605c5
fc471238b297b7b35435ec7b1563050427edbd00
2300 F20101111_AAAUSO fensin_m_Page_152.txt
9dee45e32547833e2f1da222a4675878
fedc2aa0db31f159517928df76f978bf6fee0c70
2154 F20101111_AAAWCA fensin_m_Page_041.txt
b7b1d010c551b5ece9453d31de5e9128
d430080e95d0ddeb2def7d558a54ae5ae96586f3
2377 F20101111_AAAWBM fensin_m_Page_022.txt
0e239180dce61d94787e6194b07f2840
84ce623c7f1dc25510377840ce4e948402d6c4f7
42418 F20101111_AAAWAX fensin_m_Page_188.pro
71c9511e4cf43e69809bf152916143f1
5e5839ff3b38dc68c63d16f8fb16fd3e09f8c080
58453 F20101111_AAAVWF fensin_m_Page_025.pro
779693b590a8d1342c7377e9fc0764a4
bd33dc83b5fc4ee9ed135333d3f2aae6d5381f98
89183 F20101111_AAAVVR fensin_m_Page_007.pro
02ce8763460ef9d4db65a615629fea16
9c3487481046f82e54306716c8f88faa9af872a8
2089 F20101111_AAAUTD fensin_m_Page_026.txt
9a442a8298fa52fdb60041d3a5d78273
176f363e6e4705bae0706e98d7f97a86887adc90
119649 F20101111_AAAUSP fensin_m_Page_044.jp2
793f3e4ab356b49243dd6bbda943a618
7b30f8bca6b158251d9c562e7fe8c3b22d290b2b
2244 F20101111_AAAWCB fensin_m_Page_044.txt
338cde4e32e3ca12b909251a215e4deb
04fd223974c7f39d8872beb9e415ca431bb8d619
2146 F20101111_AAAWBN fensin_m_Page_023.txt
816dfd7d6c91d1b7a484f611c297017b
c2a3799b499aadfbdf01ade253e0cc404eb941e4
591 F20101111_AAAWAY fensin_m_Page_001.txt
101148bf790d9e00e008403e58633c97
bb7893cd5f2fab2d43aaf8d504840272750b443c
53195 F20101111_AAAVWG fensin_m_Page_026.pro
3fc4ba50f4e43b64c38aa7d4d2eb3fe5
877ad88e554671bd84407f6f7eb0c3750de866b9
75693 F20101111_AAAVVS fensin_m_Page_010.pro
1a6a9cafaee562c7293a1b9d32a2add6
861156c4ee715cc3c1c45b1f7c21a9905d3d6711
F20101111_AAAUTE fensin_m_Page_052.tif
7339d7002a14231221c43d7edb59f3a4
33af9f7ff67a4ae2ca4402ef785d58265136e97d
93 F20101111_AAAUSQ fensin_m_Page_002.txt
ceeba278ac1a689d456fd86bb87be152
5665531e4cd20dbe6bbb2ae7ec4f400d5edf5e19
1412 F20101111_AAAWCC fensin_m_Page_046.txt
290e0aaca2b908c3311f2f38f725c882
c1751d5058f1a3dd572548c896dbfc0b56f06ffc
2206 F20101111_AAAWBO fensin_m_Page_024.txt
837bba37ecc9b7d865b9a8e07aacae80
da28327bd9be92dd4dd4e27efafae59013f40f2a
3676 F20101111_AAAWAZ fensin_m_Page_007.txt
634e18e6fb0ce50f1341c51d64f94d3e
c0cb2c8dd004dd4512b3ab61b178e093407bf8ba
55273 F20101111_AAAVWH fensin_m_Page_027.pro
cc31e18345d63d2c62642c91ade2eba2
f658787ecf97ff19281efdbefb368e44d942a857
52183 F20101111_AAAVVT fensin_m_Page_011.pro
3bc8c5c662f193720fe100611f141c22
17a4356251e192eb9932a4d0c6233ed46e3338c2
16754 F20101111_AAAUTF fensin_m_Page_110.QC.jpg
47c01b4c01625d5f3e171cc07a84c034
fede2eb76be8f41d82c076b442727b2164b9ce01
19693 F20101111_AAAUSR fensin_m_Page_072.QC.jpg
bf27fc21bfb0208340f4600be9165d9e
b6ac48425deb22d823bdef22bf1e586825c4db30
F20101111_AAAWCD fensin_m_Page_049.txt
f413ddf2136e5fef1887780b7145e1c1
205d103df0eafbc4980338a73a338171a1f37d23
51998 F20101111_AAAVWI fensin_m_Page_028.pro
7c8f532056454bdd36cfa994a558b832
780af0d4015ba957163aa8975a4bb07493680691
23745 F20101111_AAAVVU fensin_m_Page_013.pro
75966bbd20b1beb4cd5afc429c4c101b
0466e635e77e7922c6a741aab4f56332a2c5d3d0
25959 F20101111_AAAUTG fensin_m_Page_009.QC.jpg
058251f0dd72d3a26fc30c4a226bdb9f
fcb15277803e7453ab17b9f0271c6f9fc43f9fc5
107315 F20101111_AAAUSS fensin_m_Page_144.jp2
3cf367972996f45806358bbda4c74cb3
8c7bc8681d7b9e42448c3fcee8c18c91a7b68b89
2221 F20101111_AAAWCE fensin_m_Page_050.txt
24effde98b421eb626973bde3af42f9d
15d5cef647bc2932ee8172ecb39e1cebee65c6e6
2293 F20101111_AAAWBP fensin_m_Page_025.txt
025b0f67abc11b654f47c54f89225c77
9367b433ace43cfd4da8ad3c228f497016b9e431
55798 F20101111_AAAVWJ fensin_m_Page_029.pro
f302c51e6e1af37e1d52e9127f573988
a0ddebdee6edecfb75a1cbd9e0559c21e6ef15dc
51708 F20101111_AAAVVV fensin_m_Page_014.pro
2be850f11307b8a3dbfb71baba4058dc
733e518d492d5bfed7ac65802cdf78c1e80fa7fb
123159 F20101111_AAAUTH fensin_m_Page_025.jp2
8dd30e426be2c8fd09f3dfd755cf53bc
a339648aa4e637f3179f714fec3ec5ba54ef8661
19260 F20101111_AAAUST fensin_m_Page_094.QC.jpg
a0c4ab6e6715d367a9d55016bbda7b63
8dd30f1d1a71cceb54c6563fc33b16dbe432db0a
1661 F20101111_AAAWCF fensin_m_Page_051.txt
b609416a22f9922a7f07e5a918389c2f
e19cc9a3d36836e41fbfcb70640308ef2dd61a0a
2047 F20101111_AAAWBQ fensin_m_Page_028.txt
721a2fe01b8ed7281cf6e5e8289bd9ab
f8f375777dfc1e084473decfe0a149bd51acb092
45154 F20101111_AAAVWK fensin_m_Page_030.pro
2b4f7f123dc6a32fa4147d41b184c58b
1a66829c14d4773ad7e0d2e6d6f54f514ddcf125
55192 F20101111_AAAVVW fensin_m_Page_015.pro
8765db265d1996d169bec9cc7d697dc2
5222deaee7f928182ccdd4910c07f0d8275b7279
24074 F20101111_AAAUTI fensin_m_Page_104.QC.jpg
885ecb3392998beb1139b0f92b3e8249
602caf16af97198c7005f3802125de26ffd408cd
F20101111_AAAUSU fensin_m_Page_024.tif
7b05710d50697c75c8d2f2643a9e0e63
aaac623f30b22e8204be57280c930285e0b07c7a
2133 F20101111_AAAWCG fensin_m_Page_052.txt
bf2c37c8f803c35b614f13284a06546a
5fcd1c04b6b1b15e2c1878dff4622b388221bd10
2232 F20101111_AAAWBR fensin_m_Page_029.txt
fa6d5591525e385cc3766ea9b467b14f
72446f76372a5156f3a6fac3bc4fcd131f7b05e7
56778 F20101111_AAAVXA fensin_m_Page_050.pro
b9551e3e592d06f26186134c3fab2e74
507162bce8de473fb18aebb7c3479d539f27f432
39228 F20101111_AAAVWL fensin_m_Page_031.pro
59dba86822cfbe5296ebd7a83a9f6ba8
529bfe2623cba8ae127c03a39554d0b3c2cc2d5b
51573 F20101111_AAAVVX fensin_m_Page_017.pro
49ed88c556410a9110e1278df48fb670
e66f8cd595715dacebd9cffc0f6d71fb14952bb4
57457 F20101111_AAAUTJ fensin_m_Page_098.pro
df8ab39d836562e80d5d421c71f3302d
a129d659338a10689eb8fc201dd36bb467937a29
6251 F20101111_AAAUSV fensin_m_Page_112thm.jpg
d7debfadd3e772097fbe3f5f34b3622c
0787e5b1f56dda8d31602017cf271a8a3afdfbfc
F20101111_AAAWCH fensin_m_Page_054.txt
85f25aeba08f712893c61d39cf87162f
f9e2944926e073927212be6e9b5f4f63a157b655
2013 F20101111_AAAWBS fensin_m_Page_030.txt
77b74054960f83bdd24cde0abee6fd6c
7aa83fac81167d4ff6e9e4dbaddc8519c92e64d8
37388 F20101111_AAAVXB fensin_m_Page_051.pro
5197851236a7f30ec698d974716d6550
42fd9aed477be886f3138d952409e78ab55c5fc7
57690 F20101111_AAAVWM fensin_m_Page_032.pro
27ff07e2eda422d35c891f1ba4ed01f0
c4b0c4b6fef69c7d28cbfb16939b303f374b0c40
53505 F20101111_AAAVVY fensin_m_Page_018.pro
19afc1b7fd7a421ee0b9e8cd406024bc
98692793da14331544a31bcefe1df20a9dcec799
6974 F20101111_AAAUTK fensin_m_Page_143thm.jpg
e9ed7df6b9271afa85620f8fbff87680
5503184b3e0bc63f293edbb1f9b34b14f5d8f66d
F20101111_AAAUSW fensin_m_Page_047.tif
9385a678d7b90e378f9ce30b1fa55fd3
8ed1dbf7424bda8d71ea6d76a9babdd721c392fb
2109 F20101111_AAAWCI fensin_m_Page_055.txt
1c43eff91ea82ca6a20c16ed4ea6f7d4
b47aaa86b0db4626a33146b21f2041d1f9f76019
2308 F20101111_AAAWBT fensin_m_Page_032.txt
b22d35a800b0d3da86d42685555b3f94
634f8d638b05c16c0dd682414db709bb0a9c6781
44042 F20101111_AAAVXC fensin_m_Page_053.pro
f963cdd1e8283d1da6ad0f47d741b4e4
e96cae623d8f455e601e7cbd888be1845c584546
46362 F20101111_AAAVWN fensin_m_Page_033.pro
ec91fd0c76199842783aa3fa0d7ab8f9
66481943f011f5d9ea82a88697677c14e1254ed9
56499 F20101111_AAAVVZ fensin_m_Page_019.pro
5fc096fe2038aa4e8b732ad1de6f3dc3
0e749f6835b0522f80e684832c1d09c4bb3a37c0
F20101111_AAAUTL fensin_m_Page_169.tif
7b0a46db647181e1fddab39eb8b6ae62
abb7519822da2217d6c9888b3bf677f651862003
55209 F20101111_AAAUSX fensin_m_Page_174.pro
eabebb08068c48f3f82cdd301682c6f4
552dc31c38fd238ce6c43f815c0f4baeec8cf6d1
2678 F20101111_AAAUUA fensin_m_Page_009.txt
1d32e96df28990a436d0e2aace9e0e44
5b2c0af76c7879535b393481b20f61366cc73fcf
1710 F20101111_AAAWCJ fensin_m_Page_056.txt
7c88ef8ad88de77414f47ecf1ec83168
6d49fdf1c122490f61b58a83867a72f1b37afc96
2096 F20101111_AAAWBU fensin_m_Page_033.txt
0b29d3033274c5e5f2539020103655c9
2e98a034358a9e87d96f3fc6cad6f2b5c3a73766
55285 F20101111_AAAVXD fensin_m_Page_054.pro
bb35f547b3b1544f6a6c5d54baf3f2a6
5496718b99c5c840e2b314145f20a11d3587a8dd
52467 F20101111_AAAVWO fensin_m_Page_034.pro
b1417b8e9f56bc98e9e97a8a2cb73162
cb90603f57a9c38bb2a2e77857e4638518f016b4
73357 F20101111_AAAUTM fensin_m_Page_063.jpg
5b9e5abf092f1b0d9f3212a4aeaf91d0
47f0d37dd690bc22c8d8e3a9556faf5a4a051033
1051952 F20101111_AAAUSY fensin_m_Page_064.jp2
5663998ee64fc30ad6aa55a7d56bf443
f30ae2f44a26314a8cf4253bb25631166094138b
118742 F20101111_AAAUUB fensin_m_Page_047.jp2
392512a457f65dea6b2fd53521be9fc0
98e9f7c156fe3cb5000fd6c1ff382c0ec3d9228b
1875 F20101111_AAAWCK fensin_m_Page_057.txt
80cdc80c5d91f3b0ce11b03a86ead1bf
2f130fb0a3360e8c0a00320dc1a33cfc31ee293a
2235 F20101111_AAAWBV fensin_m_Page_035.txt
dc1bd8eb4b12e19eb93e4ed4bde5c3bf
a6fa4b4f2946ad57e2c0ffdf15cde43e36e6ebc0
48398 F20101111_AAAVXE fensin_m_Page_055.pro
70e45d17d40afdaababf0f7b2430f6ff
12f572e133afd6e8cd3b09c8ba2a324fc3b2879e
56494 F20101111_AAAVWP fensin_m_Page_035.pro
e76d3641957d99f1df048df84910c902
d2f2f7103e8cd7666da703a3caee8821b30f460d
58060 F20101111_AAAUTN fensin_m_Page_170.pro
b9c5c554124a31aeb7312cecea98eac2
7fe776be3d0657e65405ff85876746901adb735c
6815 F20101111_AAAUSZ fensin_m_Page_088thm.jpg
0cc4581ad31e7783f70f709df806b36a
ff289b6dd4110ac6589bb6be6c251d4c32f79a79
19403 F20101111_AAAUUC fensin_m_Page_051.QC.jpg
81a7a1b7a796bbc86e8bb30db4791632
eca6be8080216f30999121fdddc00eeb958d1ef1
2002 F20101111_AAAWCL fensin_m_Page_058.txt
1f3e1e9347d30fa402bd8bca49bc2176
d1537f095b1d8e580898c6470323290e3f6adf90
2151 F20101111_AAAWBW fensin_m_Page_036.txt
d55f9448f6e42d19f1fb897cc95b793e
7f5343c8b2b456b760176940025af93eb194834a
29120 F20101111_AAAVXF fensin_m_Page_056.pro
3e62bb13ced82cc29d7634443ab38e21
809b9df9be655b6bb8615a993e039a8ee0a46d8a
39297 F20101111_AAAVWQ fensin_m_Page_036.pro
ac545990733b6991dffce343d1ba02dc
7040ccb31ad79ce2b04afe0548e7cd30e933bb74
7055 F20101111_AAAUTO fensin_m_Page_065thm.jpg
4f258c2d649992766d47ddb1487b3e3a
fe20eeb17a1a81cbfd1054b0c7b6ffa1adb037e8
F20101111_AAAUUD fensin_m_Page_103.tif
9c99e9e44b8e1c14772ab6de3324b0f1
a49a23a167bc1b7b1f32054b0b296a2b4f4cfe12
2097 F20101111_AAAWDA fensin_m_Page_078.txt
ab28fd1fcb597a40c0b43f0b5603872f
f557bbab257fddd76af22dd7f27a458932496b10
1623 F20101111_AAAWCM fensin_m_Page_059.txt
f94a0bd29f4ea8e03234b1fe34b63dc0
85d003f0837ba7b1b94b2eb62e788832614a849f
2077 F20101111_AAAWBX fensin_m_Page_038.txt
35a8978486b01d92fa88dc5146dc3eec
87be2228c878d1f53b4e6b52a29e39dcc9041d3f
57764 F20101111_AAAVWR fensin_m_Page_040.pro
061816c9046fb7734ae9dc506faa4c39
438259a34977fdef824e626450076415b1e8c12c
24901 F20101111_AAAUTP fensin_m_Page_028.QC.jpg
9650dccd69c02fe96320f66899aaa2dc
a5ff78a831aaf818b07445f63e764f26565150c4
2954 F20101111_AAAWDB fensin_m_Page_079.txt
3c59d52eba6d29ff8bafec18012787a3
d95782eef929cf1752cf8f57d55869b91e1b0580
F20101111_AAAWCN fensin_m_Page_061.txt
9af2d1a65bbc316358da416fee7dd50c
b580fde74318d611e7d6d2761ea09ad31eb595ed
2081 F20101111_AAAWBY fensin_m_Page_039.txt
b8c32c4ecf2a76ec05db1946f38fa8ca
df82666a3e3db38e34517af549f6ceaddb16503c
36817 F20101111_AAAVXG fensin_m_Page_057.pro
9903b4c70677b1d52fa66c598e28d17c
736de78435a2c158121ad6483eb34a84ba563469
51921 F20101111_AAAVWS fensin_m_Page_041.pro
2e024fb8fa72a3222efaf53439488be8
740773a19c7db4697ae2446319a91ac1f9e6761a
5995 F20101111_AAAUTQ fensin_m_Page_083thm.jpg
836d1b57eabe46413905bccbdf9e428f
1cb436dc9ecdd06b3046242b26e94c494c3645bb
2184 F20101111_AAAUUE fensin_m_Page_101.txt
a5a482327d8d3091e7771d14c079f3ef
1055dafc17f62d5b501bd85610e5a0935f112f15
2196 F20101111_AAAWDC fensin_m_Page_080.txt
aeb46e20464d25b65d19f23446c14891
7a55cecfdab492c441c8441924a9bd8c62900599
2021 F20101111_AAAWCO fensin_m_Page_062.txt
ec60f19deb719f263e4f1387fa2c95ec
94227687a2b69bb30fa2e9a4054b5703b4190180
2260 F20101111_AAAWBZ fensin_m_Page_040.txt
85f81783614abca9a6084b1fd3f5da0f
cafe422755531e09678c2e8592d99e214dedc379
40886 F20101111_AAAVXH fensin_m_Page_058.pro
22d6ffe51c2d6b821ebea61d5daf058a
2be955d32bb6e00e941dce655ad8c347aa9a8152
45788 F20101111_AAAVWT fensin_m_Page_042.pro
721725b2aa2d552ba3dfbcecb8678a3e
44820f699c3420d44a273c8b3c6cfd83d94c868a
2684 F20101111_AAAUTR fensin_m_Page_163.txt
9baceb61c99b65099bf84be1f2863e39
bddea7dda428325dcb36b749319b1b430af772f5
25142 F20101111_AAAVAA fensin_m_Page_153.QC.jpg
bc88190b324843c7bf5850568739b14d
93ede0b7aa3610082484df4fa3334174c462b39c
6973 F20101111_AAAUUF fensin_m_Page_020thm.jpg
a36a1f9f113dea59e3cfe676dd8a6528
21c550081ccf3c524bbae54f202d6d80852f3ac0
2130 F20101111_AAAWDD fensin_m_Page_081.txt
545216e2563a7a54c93036bd31833093
2a5c75de4cecd188dc3e7e227bc418d86d703fd0
2156 F20101111_AAAWCP fensin_m_Page_063.txt
50fc536bd44170287741549d5c331b16
b6fd8160a4916aaaca6b0e66c2297423413cfed9
31071 F20101111_AAAVXI fensin_m_Page_059.pro
95cd9661d4dab4cdd8a40610f31200e5
58e4e50159f6d62de5fda5d4a6d1bc4d6082e85c
41279 F20101111_AAAVWU fensin_m_Page_043.pro
524e860b08c6ec212f7e475d24d8a5a4
b12878eb9c169c766d1689d2376742413691a7f4
25370 F20101111_AAAUTS fensin_m_Page_141.QC.jpg
ce0029f4f9839f769f1ee828d0506414
e1ff5a69dcc917f033639b794269ca993d875307
4016 F20101111_AAAVAB fensin_m_Page_006.txt
3ee762cb756aa741240efd8002f70435
a067ebe74da5be39b54a676ca2257cea3004c134
514298 F20101111_AAAUUG fensin_m_Page_155.jp2
78c65617622ed9809570bb3c5ec097f8
f6f741a872f8e636bf365505c7c5c95e1c89e491
1090 F20101111_AAAWDE fensin_m_Page_084.txt
734dbfd6c12d67b9a800b821f6a7e3ce
7b5830db47738580bc04dbfc765a449f1b6d8096
47515 F20101111_AAAVXJ fensin_m_Page_061.pro
dbce3736bb8c817c809a7d83a7251edf
7c69f80580d7b63643450537202436f68fe18dca
56729 F20101111_AAAVWV fensin_m_Page_044.pro
f9907a5f0ac38e6c0b64b692307dd534
2934a70ce5c4596c4c8f1f79970978f7df14b451
74648 F20101111_AAAUTT fensin_m_Page_099.jpg
c7532b412471813fc9754618b6304317
62c60946d0bf9a906cd5e1ecfba9541a542659ef
F20101111_AAAVAC fensin_m_Page_074.tif
9ae14761aaa44aa4dbe0d7c08160f0c1
a75e59bbc863e5ca3abc8b7e75c3b7059ccd31aa
7122 F20101111_AAAUUH fensin_m_Page_023thm.jpg
27c2671f9d948f8181ed126eb34faf09
cd86faf8ec0f7d14b52183324f5cfaf03012e8a5
1870 F20101111_AAAWDF fensin_m_Page_085.txt
01786dd3c48ebbba67c119beac3d20f9
2a2b4a31ba9bca7b7bdc79e29e06c1f1ae6d1213
2142 F20101111_AAAWCQ fensin_m_Page_064.txt
850690ddd29a118ce9cbb3139313fd44
1ccd218aa7ba53929baa30a9859e10d50625f6fb
44980 F20101111_AAAVXK fensin_m_Page_062.pro
ad46fe568906b9e518a4b2498ede3259
96649b96ae89609172fbb77daa83bcb9ad7bff10
55424 F20101111_AAAVWW fensin_m_Page_045.pro
50ee4a3b6fad5abf4cc87acf8bcd1b8d
0d576e5cb07004380943ac96c474ce8ccf52e3e9
4345 F20101111_AAAUTU fensin_m_Page_157thm.jpg
735f0251fc7995cb1ce59a58db73e0f5
33a45bcd64b5f5f96c6579c1c532f4b7ffb9edd0
F20101111_AAAVAD fensin_m_Page_081.tif
8306c11794f8c77c48551d81dc49dc51
8b772b7611f13cc2d4a6895b8ed0ace198a712e0
69771 F20101111_AAAUUI fensin_m_Page_089.jpg
ef8a5db66923b1874ff7eb74fae192e9
4579ca7b2a893d8c512bc3842589332fb1ae0694
2144 F20101111_AAAWDG fensin_m_Page_086.txt
5e71423f3fe068b5e0b4f4b624ca9714
01d43f2fd3098deaf7d05e63c4d1b1af75ceafd4
2227 F20101111_AAAWCR fensin_m_Page_065.txt
3bcbeca4938b08fd86cd8f383fb1c555
1fda77d085cedc23ec1238603edb68d1d48264d2
46430 F20101111_AAAVYA fensin_m_Page_083.pro
dbd0d6a35dc01dfdd2c0afeec1ce1240
f194cb418321b71257bce85a06dd844bd024fb89
54570 F20101111_AAAVXL fensin_m_Page_064.pro
44dc089af8393f84ac6080a0b92ab84c
b6fabd37b9e8229511ff6abccccdca61087f6630
56003 F20101111_AAAVWX fensin_m_Page_047.pro
76daa3dca36f9b9fad91e9ccde8b92d0
a5fbdf5d7c9e099fb4316b3aafb8b2fce9281b54
6749 F20101111_AAAUTV fensin_m_Page_123thm.jpg
5e5134d71db10402fda7cd7f9d5498e7
598ce03a200d3517303ea21bd903b7bf946951fc
112676 F20101111_AAAVAE fensin_m_Page_185.jpg
107fb9c7919046c6308054f2720bad88
6ec970f8d5748baabdd0033d2d67bf9054e11fbf
19972 F20101111_AAAUUJ fensin_m_Page_082.QC.jpg
c605ee0e971fe5de0b27586e9bd5f351
67f9e4bb8fa1555c8e027bf48101c5dd7f870c8b
2198 F20101111_AAAWDH fensin_m_Page_087.txt
9caa1d532ca0e03c49dbcc0e709c0bd2
77b6b7987cf3cd6eaf59fc0287311156ef9b849a
2103 F20101111_AAAWCS fensin_m_Page_068.txt
c4649825315e8aed7ec39097417cafff
f803c2038fb24c4c5a83fdb707663172a23da2b8
22540 F20101111_AAAVYB fensin_m_Page_084.pro
7d701e355591aee51311fa8fce63cc6e
38fc3bc92ba781fc09d5bac38215272d65b3b37f
56344 F20101111_AAAVXM fensin_m_Page_066.pro
e611a9333f92d55a3814da159639700a
ab52aa80738635d066260b5aeed48e24778c88ed
51368 F20101111_AAAVWY fensin_m_Page_048.pro
0ccf0cce160f5a8f83ef5547faba7ea9
de878c66e82f9bbf88af6e7ef41f5342987d1c14
26776 F20101111_AAAUTW fensin_m_Page_025.QC.jpg
959fefd7829e56385735bbab93538021
6da601aebaf93f013687c9f4bb6500c95c09a5be
F20101111_AAAVAF fensin_m_Page_046.tif
6bd2fad45d0c32042684cca765861ea5
486891ad6c9f06ee06eeae48e89de727e5c7634d
3696 F20101111_AAAUUK fensin_m_Page_102thm.jpg
3437bc37d4fa4c7463887adbd0b18d59
d47bd388a099d457e885dacfde0e911021543b6d
F20101111_AAAWDI fensin_m_Page_090.txt
9551a8b857b160244a33be1b04ae15e9
5b99ec584b3a487179325fcb68fa6b15b8ace5e0
2098 F20101111_AAAWCT fensin_m_Page_069.txt
8f96ff626372792ed7761e55e3671343
357e1da766bbc49912ca49ec6062ff0ae338fc01
52343 F20101111_AAAVYC fensin_m_Page_086.pro
7489f683b6879506a4d513776bfaa032
d6cd27953e102bc0194412f9accfbc422337a85a
54612 F20101111_AAAVXN fensin_m_Page_067.pro
fb7fb677a47badd05f3ce41d66689870
ed139a50cc5d8a60b0b096b56e3c693df0da382b
47597 F20101111_AAAVWZ fensin_m_Page_049.pro
5320288851dcadf800d9d23245552bb3
f7438f9781564a07d69da7b1f78924fd9a307639
75699 F20101111_AAAUTX fensin_m_Page_028.jpg
7111bae93698c8e9ffd1be64e5b5b440
fc0ae1d003919ae28a2adaf69e515de5a84cd67f
80254 F20101111_AAAVAG fensin_m_Page_123.jpg
e393f3173581697c6b10e13b97691505
1d783e196e04f5450d7f3338a728444f34d53097
119994 F20101111_AAAUVA fensin_m_Page_171.jp2
74212941b86a56889ccbe1a06ded2877
2a81e028d9e2669cf28a967729ac8e5c75c51c8a
119474 F20101111_AAAUUL fensin_m_Page_136.jp2
4666995c703d6a37275023f223008bcf
6aa75cafd0277f1522d34ee626397512e367f8a8
2285 F20101111_AAAWDJ fensin_m_Page_096.txt
bfe6c603b467cf0cd71be5bdb78928d4
c8d03dbf57e8c83215c2328059273a7b062da0f1
2195 F20101111_AAAWCU fensin_m_Page_071.txt
09812cae5adf7315d9847ff749939d42
745ce2903ab73c27b5754e81b5c8f1b0d34e1e8f
56284 F20101111_AAAVYD fensin_m_Page_087.pro
0e562e2984557accb84c53cb8384f739
2896e97d03189ddf95901a74578e7a9e8b6bcfb7
53243 F20101111_AAAVXO fensin_m_Page_069.pro
a08c2f4e7d25fc773446e58bb256b26e
63c4364d3194f3e3433ab6d4f5d506cb7514c7a9
6793 F20101111_AAAUTY fensin_m_Page_014thm.jpg
68430db0ee8c7ec5028799c885c78eb1
0f27d5d7421abbbbd668e9b7207b6b38e2b063df
55814 F20101111_AAAVAH fensin_m_Page_105.pro
f5331e8c425d34691dca0f7077802d41
00a8da7db054c7cdff95b3e1fd62382ac3e7409d
2210 F20101111_AAAUVB fensin_m_Page_066.txt
7e2ad80403594cdef175db699e30a575
57db74126f14bcf1fd7c6a9ea444364e2f86835b
6315 F20101111_AAAUUM fensin_m_Page_098thm.jpg
70b3f8b84256a87bf8980ee225ead59b
78a6bf98184d32854a324cde5ea23ebbdfeaccbb
2231 F20101111_AAAWDK fensin_m_Page_097.txt
d655709f79c55ae9c42df58b950625fb
030bab8b5175d4c2c90e1f210339549cfafc9dfe
1697 F20101111_AAAWCV fensin_m_Page_072.txt
e3921c53e3bd5b59199537dae0c4a1cc
30346eb14be43668fd6846fb3f851e38d230b234
54873 F20101111_AAAVYE fensin_m_Page_088.pro
46a47fbdb9ea25b6fad7250d8b0e864e
2f274dfb7771bcd8295b4ef8264da95ab200ba15
31044 F20101111_AAAVXP fensin_m_Page_070.pro
d7e027fdab124d4d3f84f87e7907438d
6a5c9167182d3e38d24161ba7945335699d37010
2155 F20101111_AAAUTZ fensin_m_Page_088.txt
d96006a37f16529307a691e651c0177a
826b29a195b33ece9d178291695136c3e2526c76
54102 F20101111_AAAVAI fensin_m_Page_081.pro
9a1926715cba432e1da79cd83f21ecde
84e744f73611698c66be628e6bc485e6a36c516e
7495 F20101111_AAAUVC fensin_m_Page_180thm.jpg
da341fc1b1ca11246200ace0af2f638e
dfced1efb474baaecbf343d23a36a21b84e00bf0
6860 F20101111_AAAUUN fensin_m_Page_173thm.jpg
743ecdf2bcad7de306c30dcc49f1d575
5ce1b6d3524046d51992ec3d6897caec050bffe8
2576 F20101111_AAAWDL fensin_m_Page_098.txt
eb776909820f849d121250441379e2bc
8a4ddf5721675998e63fef5213d964d4c1a43fda
2266 F20101111_AAAWCW fensin_m_Page_073.txt
32f527c1652391e22022953c5cc01328
8f9c85bb2025cc54a2292b9c1d989a05fe4038ac
41272 F20101111_AAAVYF fensin_m_Page_089.pro
89e03cb34927e0bbc75dcd423616f045
30245749158e0874c9f666203d3fdfc8a01c9f50
54410 F20101111_AAAVXQ fensin_m_Page_071.pro
1fe9d236ef5c4f0e1b5a4a33d66d3cca
223e48dc6bd29fc9617b00937d10b5e55ac5527a
2176 F20101111_AAAVAJ fensin_m_Page_045.txt
c9f9031f377714ffe4038239b8c76ba6
b9b78f21cf177780e265eb9e3895b47b3a48e4be
25382 F20101111_AAAUVD fensin_m_Page_095.QC.jpg
4e8790cd8442a8df29b7de6dd264077e
befcba7df20cdac47c19bc8fb17582704ca35ead
25163 F20101111_AAAUUO fensin_m_Page_020.QC.jpg
11b88f76ab85931a02e37c9ab209b58e
c9b7e692c05bf0df215ed3d1dd082c09925dc338
2367 F20101111_AAAWEA fensin_m_Page_119.txt
64856db0c7a37fdaa384316662c1cfca
11e3372505e6c6489751eef0f893dceb6f49cd55
2191 F20101111_AAAWDM fensin_m_Page_099.txt
9ed6960a4f163d27001b12abbeaa9819
2fe3baf611845d81a2b698549f90bcd001cc0c37
1971 F20101111_AAAWCX fensin_m_Page_075.txt
ed3da5b224c96c8a086a17235f5872b2
498e0c94764af961eb14c9711eae2bb001dec051
50044 F20101111_AAAVYG fensin_m_Page_090.pro
da3564f79ee135bed213e74f1526fba6
fa4e32ccf47eca5e7b0b93f9867ec5f608fcb8f5
32785 F20101111_AAAVXR fensin_m_Page_072.pro
08a4059c6a69b0add179d8fcb13b0d74
de9091b42dccb26d80169ed0b355cae2c5c1a15c
46007 F20101111_AAAVAK fensin_m_Page_038.pro
bcf67d82a22958299f8ecdb53b591039
20a2564cbcb54ae668958927e40bff03ecde65a5
F20101111_AAAUVE fensin_m_Page_036.tif
d9fd38c1e34c3eca003e916e9ae7503b
94cc54e0868e024a5d6f43784194b23dd42b5c35
6752 F20101111_AAAUUP fensin_m_Page_120thm.jpg
e8cffc35e80d883500f0f3751f203cdc
ca286962692d7526c6969db0a195502a32dd5d90
2104 F20101111_AAAWEB fensin_m_Page_120.txt
ee59c0505e6026cc6974336b13b21df0
54b6ac50bb2fcb9b0752879742c78361d63ce40f
2183 F20101111_AAAWDN fensin_m_Page_100.txt
e05660b555d0d21e30f7e3701047aa2f
c7c82cbfa46f7c35232d556b03fbefbf07bde363
1536 F20101111_AAAWCY fensin_m_Page_076.txt
a185a9d320b89e30c3b2ae5a61a22a54
078becb3cb783c816e6c2018021b76cc88b520ed
58116 F20101111_AAAVXS fensin_m_Page_073.pro
41286e36bcecb1dbd9e6f4db6ac32e3c
d709d6d29b58df32f2b6e6c79a17178b168059b6
74984 F20101111_AAAVAL fensin_m_Page_104.jpg
1b0978a8214a23fcf52e77c33c320713
408ab85a1aa11921d13c7691104d0248b4e6f32b
53792 F20101111_AAAUUQ fensin_m_Page_173.pro
d7fd241af70225c4bcc25b28492efa83
9bee9ab236c35dc7468468b2f1f9b3a2d4ed23eb
2402 F20101111_AAAWEC fensin_m_Page_121.txt
575f4713c4e6e878cc66828f4d952567
10f101f45306f06f61855dd18ed6b69a29d26007
2252 F20101111_AAAWDO fensin_m_Page_102.txt
1de829f6154155b2bb958518c147f267
93e2d49baf06dffc59b087ca91e3edb5b403bbdb
2921 F20101111_AAAWCZ fensin_m_Page_077.txt
694c6b9a4f7997f2eefc22a2008f1cdd
5cfbc605204fac64ec8586a4b203055613dca09c
42495 F20101111_AAAVYH fensin_m_Page_091.pro
4a85ad9c4f568e256c7ba80bfaafa035
23b6e63d9389fb999bc11eee5a5a157c821762f7
52487 F20101111_AAAVXT fensin_m_Page_074.pro
72a6fe18bff407d7ac65b6e7321f8925
078023e39fe0ad346acdb6a8cf0f8a8d735526fc
74635 F20101111_AAAVAM fensin_m_Page_012.jpg
ff81f5e7cdffc7c1fa2cf2240b376777
c2e4b86e5f912da3af2c875db526fbfd9bc8babf
11283 F20101111_AAAUVF fensin_m_Page_005.jp2
d83299d75ace3d24787fa0e1b4363507
0843b36524b9685b21e882a63000876ef464e256
106242 F20101111_AAAUUR fensin_m_Page_052.jp2
1447ce1eda55a10f443ce6ab19571907
783d4dfdc419c4ea29ae6961e88490d43a0349d9
6372 F20101111_AAAVBA fensin_m_Page_138thm.jpg
86b6795ed5b72eb47431a1e7fb9a9e15
456d299a511edd105e1829cb4c72c4858ae41121
2505 F20101111_AAAWED fensin_m_Page_123.txt
ed327692182d55ad90141a29e3d22a29
b1254038d49e160c7dc88c3cb495464170c2a0e4
1518 F20101111_AAAWDP fensin_m_Page_104.txt
a0488dd234e9ebf30071d7c167d6f5de
f0acf3038af3bb927cba00b84aa361009cfe9e94
54270 F20101111_AAAVYI fensin_m_Page_093.pro
9fa8b373e723bd9f1195f25b37516e93
fd902ae1e0f0d178e2f090c52986d3282c1251fb
39944 F20101111_AAAVXU fensin_m_Page_075.pro
9355d1cc9d2adfb69a81988c8ec0e732
2ae4e3d9273d5d49d709e6dd979e3cc629fdf430
25342 F20101111_AAAVAN fensin_m_Page_105.QC.jpg
2a89ed4b99fdbddcd6ece26990974c91
c7c6254c1b31365bd4153666295569f63d9513db
6069 F20101111_AAAUVG fensin_m_Page_038thm.jpg
a79236c1a991c52a7bcba603428971fe
c1ee18ab80a68fd4b459efc24196d6885ada323e
26803 F20101111_AAAUUS fensin_m_Page_170.QC.jpg
afa93295e5cec93f66882d8b61cad165
72d5b110866bc48dffe7f2b01cf114b789979ffb
6971 F20101111_AAAVBB fensin_m_Page_093thm.jpg
d9f079c606167d5bd386155afa7d6612
af439375035fee5ebddb8680955dff19fb0979d9
2401 F20101111_AAAWEE fensin_m_Page_124.txt
a09002a15a6b339768ce26030cf5ea74
97449a77f5af257771cbec41c97a660fd6c1dc48
F20101111_AAAWDQ fensin_m_Page_105.txt
e4ccd4186c0f426370f5f2adf08c46b0
3e95d468928dc5f6e51c220cc5df7691b83cee20
36971 F20101111_AAAVYJ fensin_m_Page_094.pro
38c2882fd7c92e9c6285300afcf3cc2b
4fc2c554df4ae2a642c91b0e2fd7ce833561aeec
66843 F20101111_AAAVXV fensin_m_Page_077.pro
68069317ac5215c51291f2f13762d1ed
5ba9489265a320db20fdb272bcd59031042b20f6
F20101111_AAAVAO fensin_m_Page_147.tif
3191a7fc9e4210e4a125726784216f83
2306b6bcf6154525be2063873cc0408f426e36af
99402 F20101111_AAAUVH fensin_m_Page_033.jp2
7e85d20ddc73878934b55ee8681f531a
4356178da763b094a4e2eae21dad7fd57333ffd8
759313 F20101111_AAAUUT fensin_m.pdf
cbb968298b8786b8d02400b8dcf2cfe8
a34f19bc3d2b387018a1a38e49b260a62d013a41
F20101111_AAAVBC fensin_m_Page_184.tif
b2fec300397d4e44ad5dda61436d3aa1
362980c3edf20d31a47f7d19035a280638ff770b
2124 F20101111_AAAWEF fensin_m_Page_125.txt
7e0a34e42007bb751c27f53eac0a09b8
df9c9bbbf39b5277f806393bfe8c643cb4eed56f
58641 F20101111_AAAVYK fensin_m_Page_096.pro
bc334e151022cabf1e453cb0a4e41719
54116339fe2762fd97f0a705acee05e5d4fdff0c
34334 F20101111_AAAVXW fensin_m_Page_078.pro
72e12e222246a47efd76318df59e81ad
49a0937e172c8a1eabf41808a5aab8817c1306e7
150 F20101111_AAAUVI fensin_m_Page_005.txt
cad0538137d68dcfcae0abb89395b9dc
b590b1153cb9a4dc79b4e1843d4a0272e1454642
6846 F20101111_AAAUUU fensin_m_Page_022thm.jpg
6be3f3449825269374623443e3f3fa67
8190803c953d6472d7ae386f59e13c5444c4d797
F20101111_AAAVBD fensin_m_Page_154.tif
4c99d8e3479f8b49318bfaf850688087
71e05f6fd39d65040c80588ab5c7cffa36966b33
2384 F20101111_AAAWEG fensin_m_Page_126.txt
f3a28c10d3e314fdec370b1a75525b2f
77dd7a7c85918a65acccec3845a48777b5e4fe01
2180 F20101111_AAAWDR fensin_m_Page_106.txt
b770e5bc5f094bb848448138d7cefd55
7ba25603ac3adf18b785c9bdbb4a22a4d9e8abe9
61862 F20101111_AAAVZA fensin_m_Page_123.pro
aa571c4d59d1f6b32c470b5d1d15aca3
2dc7ec0ebcc4fead52f660cd230c7786abae4d20
53101 F20101111_AAAVYL fensin_m_Page_101.pro
86a5b4954dec9f7e36e3ab6d5d0b82d6
6509c422b97bd456f2f974bfa9bbff292b65ddfc
56316 F20101111_AAAVXX fensin_m_Page_079.pro
e0219142f402be2d1df3732ae5024fdf
4d041e88fe7d5ce83b5ab02bdbdfccd3606580af
1961 F20101111_AAAVAP fensin_m_Page_112.txt
a191c9730c2ec8fef8f8eeeaebebc081
f2036d89f64242e96664e573de114c7308147f75
119337 F20101111_AAAUVJ fensin_m_Page_035.jp2
4b360efd3b73092568a670a1df061bf8
b7e037219078ddf56f49b989500ae240845b1f8e
63698 F20101111_AAAUUV fensin_m_Page_158.jpg
e4648a96e07387964d35110bbbe16066
d661ea18f6a64a70182c2b427cba7d0ce8c6568b
69345 F20101111_AAAVBE fensin_m_Page_008.pro
8c8929901d0a52d34bab7dc440618b6c
1c2f3eacdac00ee07a8e6fab597476fbae945e78
398 F20101111_AAAWEH fensin_m_Page_127.txt
b944a737e652639d4a6c17c4bcebd682
b64c5a5edea488da6609156f3cdf6a20de7da317
2031 F20101111_AAAWDS fensin_m_Page_107.txt
1f4b781f0e29ebb87b35588f6347d6b2
b7f394a2726c66eb27e735e1b5fc1cd9817de367
47586 F20101111_AAAVZB fensin_m_Page_124.pro
9280e496e7609ef8277fd895b09bcbfe
bac33f9e0fc0fdf31a573a37863fe22066f825ec
39453 F20101111_AAAVYM fensin_m_Page_102.pro
ad1c946ca52066b74a8eea0df0589f67
f69e81ef2d2744233e64796c94a8e555364dfe7a
40044 F20101111_AAAVXY fensin_m_Page_080.pro
efcae02c6e81622b34ace74e3075694b
f5c795e769e6165a029088a1dc53c570599da072
6262 F20101111_AAAVAQ fensin_m_Page_145thm.jpg
6ef540b8d4fa3eee4257dbc5597d92b7
c260bbb7393d749d5c2fd57bd9e836a87b921625
26045 F20101111_AAAUVK fensin_m_Page_067.QC.jpg
7b61ae036ae83f1c496f472d6ff4fe78
640ddb04bf1a775f67348a7c457c342b9165a07e
17844 F20101111_AAAUUW fensin_m_Page_158.QC.jpg
c27ede43effc450e95eaf89ca3bbeed8
9671f753360fb5204236ee7ac2b3bb627fd4cf61
804882 F20101111_AAAVBF fensin_m_Page_078.jp2
f8b0fb993f38cc0f8af5e1bf61434b5a
c2f1a1bac5bbcdc14698f9451aed7040eaf0a938
F20101111_AAAWEI fensin_m_Page_128.txt
cae1afab93bebac9bb3634d6018692b3
17898f68a1d9800a09396afae491affab39836cc
2042 F20101111_AAAWDT fensin_m_Page_109.txt
4b2fedf72b2e3759704e8d9143aac527
3d84458926a876cb801bf255239023f594af183d
50707 F20101111_AAAVZC fensin_m_Page_125.pro
bf1f480deb302bc2071f244c55534ed7
f203dd229135a1b6e33fa4d731d5acde930100d7
56663 F20101111_AAAVYN fensin_m_Page_103.pro
c6ea4a8173b3bf0d717a2486e9e2cfa1
86117653e72d91a75d8f3f1eb9faddc65d6ad83f
46669 F20101111_AAAVXZ fensin_m_Page_082.pro
a3c639bd1b1ce87818a868ba78a3d29f
82a7f41ebf90db0972ba511afecda404fcecc936
43538 F20101111_AAAUWA fensin_m_Page_085.pro
6e7920ac1e58e48bda9af4f2bd34648d
ac3d0c7eb00573e3fcbf85da930c897054a59095
6081 F20101111_AAAVAR fensin_m_Page_152thm.jpg
30c3dc7dc8ad3cf7eaa6dea5fc3c06c0
39fd459a471f8f6634865ebbd400186202ef0f7f
39068 F20101111_AAAUVL fensin_m_Page_176.jp2
7d6085dcf2cb7395fbe54426403f1c46
14354642e8883df1e0c225ed27570577b841cc0c
57962 F20101111_AAAUUX fensin_m_Page_078.jpg
c929861281ce38eb5750d71f1511bec8
ab66bd1a3dd5e62f30305ce1e68063f73fe35946
F20101111_AAAVBG fensin_m_Page_128.tif
749be708395e85a10ab7924b251270be
362de32b3885f14fd8078da048e58713b8532336
F20101111_AAAWEJ fensin_m_Page_129.txt
5433b25190791b4204ca350f34a2de96
a98bded175464b790f00142d32482e6fe061e250
2137 F20101111_AAAWDU fensin_m_Page_110.txt
87c9ca0cf90da6abf7f2cc125df16593
4b129647e7608d8f2d35f3094016b6819e56a8ad
58948 F20101111_AAAVZD fensin_m_Page_126.pro
64f6381cd78d62d5226b300b9766cb83
c3edf4b2f6d6736283dbe99d3de1d7b03607dbe6
35653 F20101111_AAAVYO fensin_m_Page_104.pro
111237d5ab36398082f61ad639b7a67d
1f6b0d10f2f0993fa7d316801c5a977ce172f9ec
6754 F20101111_AAAUWB fensin_m_Page_153thm.jpg
dcdc78d4223c67f293e4fc932a75f112
e2d92fb2e0acc97e453a07b218b8372c8a35c0ba
77487 F20101111_AAAVAS fensin_m_Page_105.jpg
5d799430b5b2a7675181172bf08ef3bf
6f8e26f7c4085856417ab40b5e261bfa2233e853
121573 F20101111_AAAUVM fensin_m_Page_032.jp2
990f2f8168902bc017964315cd808e78
14e92767286e2e0c2fd7cd51b33d8fbd6db6ac22
23228 F20101111_AAAUUY fensin_m_Page_144.QC.jpg
c757a7003c9dd9e6db0b62b7f35a091a
f4ffad7a275becf63bf75f7a2b239433859773b6
F20101111_AAAVBH fensin_m_Page_060.tif
e8bfe166bc375623142358d49b30bf00
63008166c61bb7eca23ad2b85bb11d95ba951dfc
1637 F20101111_AAAWEK fensin_m_Page_130.txt
06934c782f03653b0bbc61938e630536
e606726087eb972b22c50212595c7499b063d7bf
903 F20101111_AAAWDV fensin_m_Page_111.txt
73b0ae2dd5786a371edd1c8db037c0f9
240a2742bd6079ca3f68aa60f00637a4eb5ae415
10004 F20101111_AAAVZE fensin_m_Page_127.pro
13ca35639e4e7892fc86d7da51baacba
fc8140bf851123a48fbe9c141f248bba41b3694d
36168 F20101111_AAAVYP fensin_m_Page_106.pro
7efa2d9afaced1c6c2fae2d62ab37dcb
f6a91709b92b9e42e749460c8b49225586e500e2
60955 F20101111_AAAUWC fensin_m_Page_080.jpg
9ce226b43400a3eb4b68ac188bdacbab
3cb44273afc577c3750e00a35595fd8566a1159a
2366 F20101111_AAAVAT fensin_m_Page_181.txt
a34ca6a7fa7f634f0f235223440f226e
829217b36d12d7f477b17e9c0dfd7600eeac9c38
F20101111_AAAUVN fensin_m_Page_041.tif
c7b8c1488801739304fcf8ce422120ef
25d9bc3b8c4c2b8c781e650e2296799e9249e1a1
4484 F20101111_AAAUUZ fensin_m_Page_005.QC.jpg
2a8aad75a7703d6b72e06f5c80e1fd5a
3f3aa33938a4aa64401829a4551b3f912ba3f021
109854 F20101111_AAAVBI fensin_m_Page_108.jp2
b3654143e763cb4597fa8490f443f9c5
ca9ff1996ddaa2d80b6d4878f34645cca8e0862f
2268 F20101111_AAAWEL fensin_m_Page_131.txt
87514338a88c66c2125d21c52a217cdf
65537357cbc8c83d6beb52e0d1af8306de56828f
2220 F20101111_AAAWDW fensin_m_Page_113.txt
8b3b30c15581de6f4624630c01616c27
8dc68205db7e6c77d634e8b72354d3956fcf1f33
52794 F20101111_AAAVZF fensin_m_Page_128.pro
2c04c00ebd9aae8f5eef09cc04d9f711
a290f2eb7f91503d7718cb1813a864dde6df8c9c
33338 F20101111_AAAVYQ fensin_m_Page_107.pro
e8460c77be25a3860c6f43878deb1d8d
746f1f6ed06eb7fa487f9f5b870218f576a40e6f
2185 F20101111_AAAUWD fensin_m_Page_027.txt
5db76a856347cd6bc14ae10b44ab1a97
36b8cd47c13c52b36c2163ca4dcde8f8c1735527
96099 F20101111_AAAVAU fensin_m_Page_082.jp2
6b06b4ecd875dc1e81d10a62010e5177
b04bb4ce94fb40ea6a8a720e64953f682bbd1b97
2267 F20101111_AAAUVO fensin_m_Page_082.txt
2d9d8d21f25f5d90b125555a432831c0
e95da0f161d955b3fa6d2081335ca37aebbf8b72
2299 F20101111_AAAVBJ fensin_m_Page_183.txt
42349d5dc0ffa8bd8294b7ae5807db14
0e56d930f7ec2fb89111c36d43223a80a538d0e0
2522 F20101111_AAAWFA fensin_m_Page_153.txt
2200621db6ca20fe26c9ec246f406916
fe9df63084ef4ec00ad23dcf8596b0c55ce97edb
2403 F20101111_AAAWEM fensin_m_Page_132.txt
3b88e31661401d63e1f7d67a9c76f12a
b72ba4409711cfb870888eaa5e84fe254224e492
2118 F20101111_AAAWDX fensin_m_Page_114.txt
611c2359bd21995ffb7d84888ce83280
e166e47d8e67a4401b80cd8ccea11afe7bc5fbb1
55045 F20101111_AAAVZG fensin_m_Page_129.pro
e5f5dd53e0c9cc5275b58cd69072162c
e5df1579890de2f73c17f9310bd0eb24beffc074
38511 F20101111_AAAVYR fensin_m_Page_110.pro
ddb6edf7bffab61224ef7eb835ce6030
eabd303bce254304976ff7a0ff29cb4b3a8d3e7e
F20101111_AAAUWE fensin_m_Page_063.tif
6820d64ac1bed717a30a48a23704c313
c24c44b4bec7e3af357b0a9d1f0de022cf78d73a
31645 F20101111_AAAVAV fensin_m_Page_001.jp2
ed8a07c8c87272000b7e86d92a64444b
6d956fd13e64979ad304ea81fac34c79ed89e8ac
F20101111_AAAUVP fensin_m_Page_172.tif
8dfe816e17b12bb2c5523a9a5063bd34
3834bc1695efc4e70da57aeee1bb6a2b9aabf5dd
2172 F20101111_AAAVBK fensin_m_Page_015.txt
7c7270acff9ce1e4cd9386fa903c2eee
5e41f57222e5940bea3d3e6505debc078485b250
2480 F20101111_AAAWFB fensin_m_Page_154.txt
aed4a0558ea00246202cd4e9df170b34
ff32d03aa99b2f0a77c95adb8aeeed7a7e27836b
2263 F20101111_AAAWEN fensin_m_Page_136.txt
9032f8e95cecec8d707524a9924a5be3
8422ab9c17cfb114398b967861eac6639e27f6f3
1878 F20101111_AAAWDY fensin_m_Page_116.txt
fdc6eae07e39ae95b86fb4eea3ac2a08
93c4eba4247e5f9adc6323a50cccc4e16738e4b8
35272 F20101111_AAAVZH fensin_m_Page_130.pro
ef986068023839ea2a2cc9a9c06e4aa7
a8076dd90c50513662c1411d2fb7f16ca3cdcfbb
14509 F20101111_AAAVYS fensin_m_Page_111.pro
fd18b66b106db8da83254c85471a8099
ca196429795624da27ec1cb749588cbaa34384c3
2046 F20101111_AAAUWF fensin_m_Page_108.txt
6d246d8d53c7cc13c4a347679d6287b3
06a70ff7fa89a579a4ae69b8d2376178920e3e87
1051985 F20101111_AAAVAW fensin_m_Page_053.jp2
5e984e24a6980b3f08a6c942621ffaad
9e559d6ce23fd903ab4ad90d9c62528467b1d9fb
2216 F20101111_AAAUVQ fensin_m_Page_047.txt
163ed398ee6cebade6fd51fe07a27163
1a36a42ecba8fc8db83da64b4a7613e3a5b9a20f
1831 F20101111_AAAVBL fensin_m_Page_037.txt
110f15e4b3ddd7faa6c03df589121915
3d4397c1ad3c663758f311bc4a50e63a98b343b9
496 F20101111_AAAWFC fensin_m_Page_155.txt
feb5d2675d61973f2f36fd3c17cb5454
931f7a29e5a0af1109fdb291c97f4ad959dd8bf5
2977 F20101111_AAAWEO fensin_m_Page_137.txt
ac51fb2cfafe0a61de1cf925267e05d5
b17477c96ebcc7067d20992640aaddf071343423
2396 F20101111_AAAWDZ fensin_m_Page_117.txt
abdb82ed1130a3a6d44b4b1f3f50bb36
87d9e6bee934773a813ddf2459a68bcf45be428e
56880 F20101111_AAAVYT fensin_m_Page_113.pro
0fe5737e29b6c14a66957c672e89f7b5
8c52baf8c0ea3a502bbbebd85145dd9d00b3d43c
63987 F20101111_AAAVAX fensin_m_Page_160.jpg
6690882e5eab122693871a81a1b33923
74a3dd02a41ab91216aca7a18f0e77474666f78d
64268 F20101111_AAAUVR fensin_m_Page_083.jpg
258b364296de1de1217e07a8e48ce97e
5810732c7c4f50f96db13a05b94b42c5a584b943
114178 F20101111_AAAVCA fensin_m_Page_149.jp2
5a7b875a483c8e4447503b333f1d6f00
f375eac2de52250c7e5121afcb1a55acd5eb529a
117894 F20101111_AAAVBM fensin_m_Page_023.jp2
194dc3a3a94a48331d9ed7561c2bcee0
9fe0377bba61186bcc4d619eb39e3f02ee63a52b
2225 F20101111_AAAWFD fensin_m_Page_156.txt
4e1951d05b3c4c1b2722d7ec52921bdb
571be2728ee445fc78061ee35d062ce510fa7f1f
2405 F20101111_AAAWEP fensin_m_Page_138.txt
804be31854385a0e7c67538b547b77f5
d72cdadd9dcbc8acab03eb9dce952ce1709c368f
52424 F20101111_AAAVZI fensin_m_Page_131.pro
9205807b1132a0469ec472c1b7a6218b
d292302b2882bed9200fd9555505ca6df0ee0cad
52230 F20101111_AAAVYU fensin_m_Page_114.pro
39ee113ede205a8dfb05a623c9022217
b63c4ee3cb3f3728868ae6115fb903c1ce76cd82
1051981 F20101111_AAAVAY fensin_m_Page_010.jp2
68a73a689f45c59f6c3d93e8fb57ba9c
31873358757f6da4cdcce08e36a4462ff97ad66e
2218 F20101111_AAAUVS fensin_m_Page_103.txt
34dfd5caeeeaaca31d2225531bb21f47
3652a782676da01f0fb5f38ea8192d0f4664e0ab
2382 F20101111_AAAVCB fensin_m_Page_083.txt
36b1122d3508ce4e4b9b9a86a450f081
9d4e134ea21b83dc1319d393ca57109a9472da0f
23313 F20101111_AAAVBN fensin_m_Page_092.pro
99a24e3434924d9bd07bbbda191396fe
49c69653360cee7ba06d5296298697030a9578f1
F20101111_AAAUWG fensin_m_Page_141.tif
17218c2b13ca68f99257b17cfacd8fed
389d911feb8e3790e598b63b83b2c30faec9c68b
F20101111_AAAWFE fensin_m_Page_157.txt
0deb9d03d6cd3c1e67cf2586831a2be2
694729d567410b0462a8599ad93bd343909c7d77
F20101111_AAAWEQ fensin_m_Page_139.txt
c47fdc62cf9888ca0131cf1d1f41ca50
8aa4962f6551acc2b28a9cb55fe96165b4a6acac
52368 F20101111_AAAVZJ fensin_m_Page_133.pro
eccbbe10ce663be3ecafa8eb037fa4ac
0c39a264dea496161eae75749951dbd8bb061f87
47166 F20101111_AAAVYV fensin_m_Page_116.pro
dc2ff557a55f06e562d6a3762f6da27e
c611899a5974b52916f26a4bab3305d0e5d59c28
F20101111_AAAVAZ fensin_m_Page_178.QC.jpg
d38a38a80be20dea476b501e2eab8a61
ac4f1bc4b315352ca6d0eecb2275c9f41f49d0c4
2025 F20101111_AAAUVT fensin_m_Page_048.txt
3787d90150d2dfeb947f83b37c6196d5
4046c6db74d6effb4665440664c7c26d93393349
56102 F20101111_AAAVCC fensin_m_Page_145.pro
e6d94bad72621c02550e15c321783185
2237b12c4c85cce45b02306fb7bc46cde17433a6
26352 F20101111_AAAVBO fensin_m_Page_103.QC.jpg
25097fceb323daab0a381b3f54672db9
7e584e5ac2dc1da757bdf3e6d4b46b2ae5618028
6953 F20101111_AAAUWH fensin_m_Page_174thm.jpg
93ecafaef3910fa0115a7f7ecb4e52ff
2f3e45b80cabea1cd307727e81e7b25060f3c990
2681 F20101111_AAAWFF fensin_m_Page_158.txt
12201d0f4024e0659b3ab036c2d069ba
2194b014bc7c7158618ed326df08efefab3c3d95
1963 F20101111_AAAWER fensin_m_Page_140.txt
42630e008f98681b950d4d31be5d211f
a77586c0e62a4be631200b3eee04f0aefd8cc025
58798 F20101111_AAAVZK fensin_m_Page_134.pro
45a882cc167f9e1c376aad97f536c536
4e098626f4e2721e88f5904d715611f577aff9d1
53188 F20101111_AAAVYW fensin_m_Page_118.pro
539b921734de628508578b7070094280
52117b35e5dfe32a943411a7936a375d7399fa4e
104494 F20101111_AAAUVU fensin_m_Page_133.jp2
11c6fba43440fe512a442451763b8c55
aeab859f78c5d5526e606da412936e59640e784c
120485 F20101111_AAAVCD fensin_m_Page_087.jp2
93d4414d936c7cb4267812382dbcf5d6
4ec258f4c547483a0c8a22c70887c6de7c338d71
102156 F20101111_AAAVBP fensin_m_Page_049.jp2
4c681f4affe2b6e61156ec9409425f6d
851d2952d7b1bae81e38fcd3fe971bec934bf726
3208 F20101111_AAAUWI fensin_m_Page_164.txt
b5fc8a3dd96f5a4f1778cdf9cbb3a427
a5e51c3e0b4bbe109a0ebc22036f6e72c76cd21c
2685 F20101111_AAAWFG fensin_m_Page_159.txt
5e9478ec2018d894153be8fa5b7d06b3
0ca5aa901bc23639d049de34c5b9ae89d56003ed
56868 F20101111_AAAVZL fensin_m_Page_136.pro
df1ebcf803424de3f558665a7d4bbc3f
0bae2a344755ed2b551d08ae5c270343633db826
51396 F20101111_AAAVYX fensin_m_Page_119.pro
b941d277bd4a13d596e91860c6cd2c81
c24b6d113e441d40f4d1a7bb5cc72c94621554b5
6952 F20101111_AAAUVV fensin_m_Page_067thm.jpg
bdf244d902a4ce2915803d3495fc28e9
d462443540d527025f2d430360da67b62b586944
91270 F20101111_AAAVCE fensin_m_Page_122.jp2
dd8414070c942835c9fe6a87377ef13b
50d1ba330d2071e447efffcb0e85b98c7d7d99ea
2492 F20101111_AAAUWJ fensin_m_Page_134.txt
327a3b2f2b0360540863856203ea0b9b
c75d1d12a0dcf17b3470bfb707ab62c95de268e2
2701 F20101111_AAAWFH fensin_m_Page_160.txt
2a1a998ebed2b028c05b1b5ac87dbc51
f9c99612e2d2189fdcd6cedf815eaa7669b931c5
F20101111_AAAWES fensin_m_Page_141.txt
2ca8f973c6b4608e16ab6c8e78f32d4a
2b720fbf5e139d6eafb8fa10131ea598406edf56
64205 F20101111_AAAVZM fensin_m_Page_137.pro
44f7a7b307f818f6bcaeb01a3c20476c
6327d390e4ec134f75e0dfd66ba634bf4966b88b
52094 F20101111_AAAVYY fensin_m_Page_120.pro
90e949acf7fd8c5ddd5fd06c1d9350c4
cd1cefc6e13f8248981314435fc81c24896c5526
63337 F20101111_AAAUVW fensin_m_Page_163.jpg
12f365ebf11c91b4f09c85e7ae3eef26
1621cd6bb176fed2e70d66dbc024b893aaf78ba8
5082 F20101111_AAAVCF fensin_m_Page_124thm.jpg
0af7d680746b14d0ac46c099c2f8fc66
5b503412a435f4ae739b59d33fa3cf03edff7f4d
1706 F20101111_AAAVBQ fensin_m_Page_031.txt
a6b0d1e93394fcdf8c3e0a635419cdb5
1f2dc04d08c54cf03439ed654f354f63368bf3f2
2607 F20101111_AAAUWK fensin_m_Page_145.txt
40c6d46f818b9a7c9a839e5d407ee6de
8b205e4c2e80d4e39b810d2bb9a81dfc6eb865eb
2683 F20101111_AAAWFI fensin_m_Page_162.txt
697b5ea32672cbed56df80f311ac5aeb
069db4a57dcb6787a8afa2a1b54860f1f84c3ecc
F20101111_AAAWET fensin_m_Page_142.txt
94cc60e5ce24a0bf66f4cc34294bfd26
316316552c5ff7c992b798b8fa7a8dde236f44ed
56329 F20101111_AAAVZN fensin_m_Page_138.pro
4e6a269cbc1e014e02a1df74d3500b1e
9e2bb49cf4c2eddd31ba04d38d14444b6997266b
46287 F20101111_AAAVYZ fensin_m_Page_122.pro
2e707b8416e371bac3fd44d3909cf612
3e2ebfb35ac0dedd3373ec77af6a5fb3daf02494
7305 F20101111_AAAUVX fensin_m_Page_025thm.jpg
edbeefd5046ad15a748c7a3f77546548
a9f4771dd2068f8eae21d9edf748e37d95fc3bd4
49131 F20101111_AAAVCG fensin_m_Page_060.pro
782e06e387c3dcf48bab5b24fc63f6f4
ba51ed7e8e03d257f283ba3ce8f90c7c9e96c722
F20101111_AAAUXA fensin_m_Page_040.tif
7944ca9b638428fdf250c25df0f20483
ccb2d987be12d06bbc5828a3236728a462d5f833
29635 F20101111_AAAVBR fensin_m_Page_111.jpg
c299dc6dfc972353dea83c6075dfad4f
aaaf311a4f7a4447d72c2cd7e04bc67296b98c17
7137 F20101111_AAAUWL fensin_m_Page_073thm.jpg
e869b2602682bb01583df09d20268cee
e5c311fedf4644a2aed293fe98c172c461271bbc
2126 F20101111_AAAWFJ fensin_m_Page_165.txt
04cc9582735833b4b89ccf0cce27093c
9de624e6e92310444c3b1af89c383e1737a55242
2623 F20101111_AAAWEU fensin_m_Page_144.txt
4041bd79ffe0d040afe9ce766a3bdcf2
d9edbfb4b572063c021966ab919cd51f16132e26
49711 F20101111_AAAVZO fensin_m_Page_140.pro
2f88879ee574efc219e1cf339efa9c73
4e767fd15d3b26eddce83e590f76c5399bac0d4a
96242 F20101111_AAAUVY fensin_m_Page_146.jpg
3dfb26d223d49c97cf5a674aec615840
45b32f46dfc3ddc39d7920435324b4c98f730ba1
6869 F20101111_AAAVCH fensin_m_Page_169thm.jpg
2308b3525742f2552dec55d8d9a051e1
189fdae0aaa388eceae5f6edab397df8c2d52e2f
78653 F20101111_AAAUXB fensin_m_Page_153.jpg
a9f17f3b03a390042c57c0b488944b6a
1891c230cb23d43946c93c01e04285937832ac5c
60526 F20101111_AAAVBS fensin_m_Page_037.jpg
7fb4b23236eb5e547ffa3485513880ec
4d733f816f0393e59df3948b5c67dd51f8dd9ee3
F20101111_AAAUWM fensin_m_Page_078.QC.jpg
90291d49c97f26fc845c63ef334f5586
18e0cfaa4096a14a3eb448f88562e9ce1d3805bf
1720 F20101111_AAAWFK fensin_m_Page_166.txt
760cc04ee4d40b127909bb107e2eb526
f9ce071538a501939b618bff20168fc8e7223fc0
1544 F20101111_AAAWEV fensin_m_Page_146.txt
c7052a6998b4ecff842b69d48f2f2f29
bdd57d8178f400aa5f8640ccf3b07847bc71222b
54758 F20101111_AAAVZP fensin_m_Page_141.pro
6740615ea3dfa39ee0dcabdab92a0d53
04ed33758699861598ec2af2ef3f8db6905fb5bb
37703 F20101111_AAAUVZ fensin_m_Page_117.pro
a0754a5def9810e8e9377c7f1080b8e5
7b647ad83a5a8149f4da59337058ce62fd9cbd60
6865 F20101111_AAAVCI fensin_m_Page_148thm.jpg
6e273e584b68969e77595fabb98fe162
3d6d94342f7177e38cf99c7cba9292c2887b6fe8
1552 F20101111_AAAUXC fensin_m_Page_187.txt
8e6d9ac0d1e67224e66ef53bf63b460a
aa49586c1e3aced2672fed7870ed412e3b031042
2356 F20101111_AAAVBT fensin_m_Page_127thm.jpg
23f1e7ea21e5c8ee9b81dc357f75988b
442ea9fe4c9a06292fb1882215839b0a60a95856
93236 F20101111_AAAUWN fensin_m_Page_062.jp2
1ea2f585a386c940dfc40c8681c3a55e
6d85bb717e199b7ae27031d26edbe806986c5a65
2069 F20101111_AAAWFL fensin_m_Page_167.txt
5a2e88550789c3cbdb5fe9d82cb6e300
56030588e4eac161247ac63a19e1544e7a20040f
1332 F20101111_AAAWEW fensin_m_Page_147.txt
b36ecb06f36938026beafa7941ab1d0a
58faf252eedda33082718ca3f5ca98e3f7878ec3
55268 F20101111_AAAVZQ fensin_m_Page_142.pro
fd8925b54928f0ad02593085bf222014
a521a77796c57c8f8027a943162acf0f17d0235f
2170 F20101111_AAAVCJ fensin_m_Page_034.txt
262b925a6054f793913bd6481090f2bc
c54838e427f9832e31c247cacd228925cac53898
48122 F20101111_AAAUXD fensin_m_Page_012.pro
7df0e2e4c38f289295095051d29d3580
ecf2fcd41224d89701db2b2453bb8a142b59e6ac
54337 F20101111_AAAVBU fensin_m_Page_143.pro
cd635dbd3e9da6d378f2812218293656
5f33e4747794ad0396d6845db1b902422d9be5ae
125810 F20101111_AAAUWO fensin_m_Page_169.jp2
9541eac857e912e40657e08fc8e36ae3
65deb52436c64925c1a26516b1db2f645d36c070
1339 F20101111_AAAWGA fensin_m_Page_002thm.jpg
1a3a0a107975141efe248f0cb8489ebd
78fd506356c0d15c841b2a6fe81d8dab335f2e48
F20101111_AAAWFM fensin_m_Page_169.txt
d12ecec8a58b0c3ab0384204deeee975
4ad5965e1e7a6eb48f5c4aa55816e4228fb4856a
2174 F20101111_AAAWEX fensin_m_Page_148.txt
6be01fd433e9f13e2c66fdd2648caab1
bab68b47b0c8ff878e4ee269e0d488f9422fcf0f
57094 F20101111_AAAVZR fensin_m_Page_144.pro
0202c746889562da9a0d34ff710467e9
445c906e1e73195d5a9b7f0519d3a1156a01752f
6839 F20101111_AAAVCK fensin_m_Page_027thm.jpg
be72849ba900669bf8bb8263beeed4ed
b11affff1bf226878d0dd8b43cf7cb81433d991b
24882 F20101111_AAAUXE fensin_m_Page_148.QC.jpg
36a6c95c618173a096cdd0a70cb34e6b
1735a4998cd32440b2a35d35f1ef1987be0e1ca5
F20101111_AAAVBV fensin_m_Page_056.tif
5d14748f1620794b6f43206116c11198
dc58a4ca0721294528e7e9640d874f8114d394c9
56977 F20101111_AAAUWP fensin_m_Page_058.jpg
3f99b585e231b7b3ebfb48fa19dd6275
95d5d8fe9cb6a2b9b480bbd9219035cb3ac9ff80
3183 F20101111_AAAWGB fensin_m_Page_003.QC.jpg
da193e2b0e1625bb65d1b1fc26786337
a0fee34b106ad67335efc9e5e944e1c45ccf5ee2
2264 F20101111_AAAWFN fensin_m_Page_170.txt
757ac7545f7fbd9cc362fc234b3c7420
fc5fe83ad4e6484a19d06ed5878bc02944825ae7
2651 F20101111_AAAWEY fensin_m_Page_150.txt
d40fc323a89a6b917853dfe02b7ae2ca
0bcbd7205c9ecb648739966179c10308336915c2
37320 F20101111_AAAVZS fensin_m_Page_146.pro
9acab6c068b08f779c9f324a12bdcfcb
d6a22f4f42416c93ff2ae2f3208638eb53e7214e
F20101111_AAAVCL fensin_m_Page_084.tif
75249ac54e5a4434e6fcfa15732863e5
228d0d56e0f6f6abc7cf7f656208b479c543f7b7
F20101111_AAAUXF fensin_m_Page_018.tif
f696e7bf4f6fe31fb42967b8ca3ff775
df351ecc4d8b56b745a55353440dc132ed0366e9
91151 F20101111_AAAVBW fensin_m_Page_158.jp2
994cd519f8267ad5f32654db65d4d29c
976bb9c4ea5f6440adaa4a4f65e0e80491ac3706
2138 F20101111_AAAUWQ fensin_m_Page_067.txt
94fd88a24ba718e5e880b88c631dd9ea
ef29bb03801dcb08b3d377f6868e4ae414ad7311
1426 F20101111_AAAWGC fensin_m_Page_003thm.jpg
722982a6068468a48d87ba907d386535
946c93841001c5c64d8c48253e3dee9e33ba79cf
2064 F20101111_AAAWFO fensin_m_Page_172.txt
9c8a072ddfbc544f09cd7fd285b30957
cd66adbff60e9baf39c0eb2ac9002b6c1cf46a7d
F20101111_AAAWEZ fensin_m_Page_151.txt
9e5a7d915ec8b6e5302ad304476a276c
79d7b0a07e2927a351d9752c7b1e45d434f3dec8
32919 F20101111_AAAVZT fensin_m_Page_147.pro
5ea9aedc75437a0f55f190de0122ea01
d1fb1120f60025ba2894127aafadcbf41c604807
17699 F20101111_AAAVDA fensin_m_Page_011.QC.jpg
d7d36d6fa7e6d2c2697f4cb7c026a6d5
9e15e230e25095bd8905f7a11cdd298b5867c8ac
24299 F20101111_AAAVCM fensin_m_Page_017.QC.jpg
94be915eddd14db4979d54348d248d43
47cf9e36afe041d9c6e081e916a98770669a54da
F20101111_AAAUXG fensin_m_Page_119.tif
7a1c1f3f2e2b1ca4cda9fb1f99430482
614906fde41315f33c587e8c25333ba1fc719b3c
2400 F20101111_AAAVBX fensin_m_Page_180.txt
b0042a2c64f6b00a3ea3a0408bf6db00
3fb0b69d0ea10dd38e79990c6e7455caab93a8e3
F20101111_AAAUWR fensin_m_Page_095.txt
0350934203e2b08e9301e98f96f3763b
f1cdb418cd051075da31ca83a4016e2388de8361
27586 F20101111_AAAWGD fensin_m_Page_004.QC.jpg
fe1bb7e785504bb8c533eb94fd313041
868da4ea70b8595a790e74e1fff9f5bccb736f4e
2113 F20101111_AAAWFP fensin_m_Page_173.txt
618287f34fa6279f1eb0c284ccce40cc
076a723dbb47f2f5e0f8e4e09f47abcfbe69ad0c
52002 F20101111_AAAVZU fensin_m_Page_148.pro
9dc1effa547340e64e3bd924ddae331d
6ac75fb0b07cf6f3c4dd82bb85ac797ebe39eb9b
F20101111_AAAVDB fensin_m_Page_053.tif
456dfdaefae797d56ee120dfe8b775bb
c4dd6a465994e57e7cc1a774886a6b6755c4393b
29118 F20101111_AAAVCN fensin_m_Page_046.pro
befae59dea8441de29c84a85c45c5b9c
af82b4940e025c7f6136609e6bb26f3ce4f7f971
6214 F20101111_AAAVBY fensin_m_Page_119thm.jpg
6ec552fe1c4e33bfc4f61c47ee2044bf
25bcad5f38d8a735de5ac0057de786ac8bd84f62
26229 F20101111_AAAUWS fensin_m_Page_010.QC.jpg
6d945781bfb86695b3f237d95f4e912c
262f6afcd1fe1a69a3c61f682b571e681dda0b2d
1662 F20101111_AAAWGE fensin_m_Page_005thm.jpg
c19d0bebbbce02f92ce0cc7fac969fd6
be646e8f236dfc31fc7a733f3c58241664d41a34
2167 F20101111_AAAWFQ fensin_m_Page_174.txt
a092d6f72ee2c22a768c933d16dc69f2
8d14acdee7d1385a0ee75e012c9c259cf0b2f316
53971 F20101111_AAAVZV fensin_m_Page_151.pro
8465ce7eb699f4b2f31ed008c8c2cf17
c2447f712d71f02dbc2f118c0a23efe9d58032ba
24547 F20101111_AAAVDC fensin_m_Page_086.QC.jpg
4202c3333a3cb0f6a587d5b3a7fd473d
67866a8de0f495ce2c93f8286d73b24e03162db7
45678 F20101111_AAAVCO fensin_m_Page_115.pro
71b1b95d70e2822e6c57b8f7bcaba0ac
4bea23640cd507d14a66a453ab1177be1f10ac07
76591 F20101111_AAAUXH fensin_m_Page_095.jpg
9fc831285044368253007d259a59c446
28e5e4604259ba98308de3b09b7f2c354f570951
2139 F20101111_AAAVBZ fensin_m_Page_118.txt
723e3fcab258eae990cdca597b477666
65d3233b82b34fb917d4c4b2d67eaabc821c06bb
8423998 F20101111_AAAUWT fensin_m_Page_078.tif
2128a442ae378e624fd65deb89dfe410
41a8ad1014f7eb11b17b1365de5709fd2e880e26
6296 F20101111_AAAWGF fensin_m_Page_006thm.jpg
95fc6bf8718cad9a993a0d33129ab98e
038373ebdd9cdf30c99f719b319459ac3aa4ddff
662 F20101111_AAAWFR fensin_m_Page_176.txt
996de05f4c0b5200d52e82686053bacb
d6c5aa62e8e30c08b516d8ba2149173e0a4c5468
53071 F20101111_AAAVZW fensin_m_Page_152.pro
1c4493fb4ccc2b3bec39d5312371e5aa
4c3c25fe1f49508624f21409b51061a0c52185fd
14294 F20101111_AAAVDD fensin_m_Page_132.QC.jpg
7cfe0a37fe13c9243695c8d6f6c93417
f36602161ad2a11b69e401f6aa86c0c9145f4478
55305 F20101111_AAAVCP fensin_m_Page_171.pro
b351c573a4a59d9317351f641684f646
a426e88872d07e58b83f59a06221ea6bc6b33fa7
109796 F20101111_AAAUXI fensin_m_Page_120.jp2
46b0b3c03d770abf1b37596694954753
5390be03d20676cd89cf4294d4cda4d873849e9a
F20101111_AAAUWU fensin_m_Page_173.tif
63664cf913dfdf04ea9f24f3cdc1ed63
12539e157f4be7e1bf00f6fdef1e82ffa6dce489
20648 F20101111_AAAWGG fensin_m_Page_007.QC.jpg
05421ece57997a6ce266a79fdf832e1b
0b98eee4fa4bd0914680bd7b38c44adc514b8ecd
1900 F20101111_AAAWFS fensin_m_Page_177.txt
451c3f617bde37bf476a132416787c27
fcd72e177f654fbd9374cc7832b55ce4e3937612
57396 F20101111_AAAVZX fensin_m_Page_153.pro
1e4eb04d6964f949001cd6573ecbdbfa
6e8a40b2177fd55625c933dbacc373d838a5da3c
24177 F20101111_AAAVCQ fensin_m_Page_120.QC.jpg
75d98e0dd1f3395251ba6833d286e996
ed59f0b7206f59adf78cea9c273b92a346099ba4
22418 F20101111_AAAUXJ fensin_m_Page_145.QC.jpg
ae97ad505471777323f5c78ddaca99b3
d21800043758e7b4c50476d312b1f5aa379ff421
5549 F20101111_AAAUWV fensin_m_Page_058thm.jpg
e161566cbeb75273dd942921b017558a
4656b6474b3562a14362ac3186bf9015b1c752d3
3022 F20101111_AAAVDE fensin_m_Page_135.txt
fecef428c3b887c2a7d2e1bf54ec2149
bd2efb70eb930e4547a277d64a7343244264fa9d
5408 F20101111_AAAWGH fensin_m_Page_007thm.jpg
ea25f4d1a3efbeb9380b9ea43ce8ae7e
c513545d2a8cc7241abd6aef246eb49adef7f99c
51513 F20101111_AAAVZY fensin_m_Page_154.pro
a3c26bc78b26ff2e32678f48b3cbe559
7bcb5094e58103d965ecb7e866dffd16987a9226
F20101111_AAAUXK fensin_m_Page_101.tif
857f397366a4ec63154d07a7ce612c06
e98f08bf8e7547a40dc26ff4a320d93e2e31cefe
38183 F20101111_AAAUWW fensin_m_Page_132.pro
ddff205b4c894d72bcdc9af5d943026e
9f56e20e816e35b3db40ee7827553e014e11ce55
50873 F20101111_AAAVDF fensin_m_Page_097.pro
7365fa68fbb12dbe61cc7f3dedf0a1ea
b2902c45c7bcbf83c44044287909a514ca89fdf1
24442 F20101111_AAAWGI fensin_m_Page_008.QC.jpg
d5cc3dd387e826b8dfdef03a6cb4bab1
264423df830631cb70dd4f511a2cca7dacf68ad1
1970 F20101111_AAAWFT fensin_m_Page_178.txt
b2626de818e5b91a304c911a3b4c3d84
1f0114841ebc9d4a8c8031435e31eef1f22520d2
7934 F20101111_AAAVZZ fensin_m_Page_155.pro
719c462016a7100ee0b1e5eed7e1c0b3
ebea02a9b3b59c08e1e8d1b42c7faaa457648b9a
22595 F20101111_AAAUYA fensin_m_Page_119.QC.jpg
ff99b6ae9a2b6b14b77d025d0fc51be7
1f15ef16043a850baacbd2135055acbe709f09b8
98743 F20101111_AAAVCR fensin_m_Page_125.jp2
4599a0a18c67f11925fc1f1cd98927ce
6e258675a6abbfe2fa1f1103967d0722dc17f882
7374 F20101111_AAAUXL fensin_m_Page_004thm.jpg
2625dd6ac9e8e64857b81b56c7754f2c
c39c7fadada0acb30907952c622e74fe270362ba
4959 F20101111_AAAUWX fensin_m_Page_059thm.jpg
336ebda1d844c8301c28dca09aca7a71
45a0b92a5a9138a91ca6f94c27e0f089f3ad1041
5892 F20101111_AAAVDG fensin_m_Page_082thm.jpg
afe680d8231628de8bb82ccbb3ecb5e2
ec013a0be5015fae32166c87bdaf3690c833a88e
6041 F20101111_AAAWGJ fensin_m_Page_008thm.jpg
a7a9673011625e221d9c38135909a37b
cd558b37ef065fb025aa5ba2ae0909eb9dbb9253
2256 F20101111_AAAWFU fensin_m_Page_179.txt
ad27f81d88d72461113381ea64609218
8547de159408963507cd944b1fec79edd156de65
72378 F20101111_AAAUYB fensin_m_Page_140.jpg
2f4a07858ad928336b8bd7d8e93ce441
bd6f014d42f4dcf3c6f2597ce0fdd776a8e83572
78431 F20101111_AAAVCS fensin_m_Page_045.jpg
9ffeb693895f3085012f3357be35a617
d94e73ee787e028aca9de3877fc4135835ea8a5f
F20101111_AAAUXM fensin_m_Page_010.tif
81e7c4b43d4c6939cf30630b54d485c0
e9a4fe6e1479e592a93100f519b87c46644f9026
F20101111_AAAUWY fensin_m_Page_060.txt
d01430dd030a9e345880c109cc14dc03
8d7d3c74b3eb534972a93fde553baa1ac9093c47
F20101111_AAAVDH fensin_m_Page_042.tif
ddea1d8a76b580e5d412ff20298e0671
bea6fbcfde6d870c18ca4d86c5505c0269f1886e
6759 F20101111_AAAWGK fensin_m_Page_009thm.jpg
ecd66b4b278c82327ce203a47446db98
ab69880a8b09adbda013a839a501745604f8df28
2572 F20101111_AAAWFV fensin_m_Page_182.txt
15e00154ede788a27516ef91ec59a113
add8b565a91341d96b2b602d6d79d598bd8b1909
65347 F20101111_AAAUYC fensin_m_Page_187.jpg
4139599ecfc979ed56f5f10d80158a1f
c223a7f470c32938c15cc846b4a54e12a9c7516e
F20101111_AAAVCT fensin_m_Page_096.tif
12f7c5fd78826bd96dc3938c9a841f75
b0f19d28aa9e51c1fee00f5fd81a5445adb9e0bd
23551 F20101111_AAAUXN fensin_m_Page_006.QC.jpg
82e166d85ce6da6a768f69105ac6ba5a
2cd96231d1d805536728ccd0e0cdf8db81bd6cc3
56054 F20101111_AAAUWZ fensin_m_Page_139.pro
6cd626b621b3ec71c1d2cb98aa198a35
05abc3e8853ae8194bf05fda2cdff78faaab135a
7975 F20101111_AAAVDI fensin_m_Page_185thm.jpg
99738771d5a3d995e7a6c18986c62635
de4c1aaa5a805f44ff3aa79bca0ebc8e9aee5ba4
6907 F20101111_AAAWGL fensin_m_Page_010thm.jpg
5c8e1ff22be4b63f7bd9aa5acb135c79
50d075590f6de37c57d97e316d20bd1e95459e58
2431 F20101111_AAAWFW fensin_m_Page_184.txt
a646a6cb7737713c659e9154c5c1650b
a4575eb0a49507e8c8c49b4ba8bcb3874079443d
82086 F20101111_AAAUYD fensin_m_Page_175.jpg
13fca34d29b7b13735e71a2ba5ebf1a2
824f9a3018ebef06de31c723efd006c551b4ea97
22585 F20101111_AAAVCU fensin_m_Page_154.QC.jpg
0852a2f63077050c5d2a9436f85f36a6
6bf522ef475de31034260b447eeadb500c003843
55500 F20101111_AAAUXO fensin_m_Page_095.pro
2ac9d0ee02e84cd76790786374000a0d
904e56fe9a3e6af71670ddcdae539436022bc298
77336 F20101111_AAAVDJ fensin_m_Page_068.jpg
a643d0907de19ad7eb9d38c5a0402c86
62d905b9460cff0b01f77d9b8ea8c3d21e71a684
6857 F20101111_AAAWHA fensin_m_Page_024thm.jpg
8552e569edcf467d5d96653d3d933dfb
12583f653fcddf1c01c41f2a0e10db85e6850c5a
5019 F20101111_AAAWGM fensin_m_Page_011thm.jpg
34f6c6374ab1aa552d839f8d92ff5633
3256ba9e90fe0a5122f58e58d0de7791e214a9b5
2794 F20101111_AAAWFX fensin_m_Page_185.txt
75cc2cf9f71db2205198107c70e2931c
58e0def6efebd1ce7bd27b387aca44ab941121ad
57955 F20101111_AAAUYE fensin_m_Page_031.jpg
92870afd25981c19e049355e193d3cca
e66cb2662afad3b27600854c204cd0fad2679a72
24320 F20101111_AAAVCV fensin_m_Page_123.QC.jpg
c5ed8fa0a902a3091b9f18406853c742
ea7940af8de494b9880ea20421ba2317c108a86e
4280 F20101111_AAAUXP fensin_m_Page_132thm.jpg
a5a4ccbcc9b4c7f865ec12875f0f267e
9406f60ee565a3ab1d08509453423dfe22c95c2a
2136 F20101111_AAAVDK fensin_m_Page_093.txt
067c178cec550177c69ded7d3bc3b9d7
c76ed363687b9eb3d703b26777f9ab6a81c177cf
25228 F20101111_AAAWHB fensin_m_Page_026.QC.jpg
0971d6855bd48382aef93957c0239a52
b8587b73a8a1b0a78a6c5b4563648ba850a61eef
23051 F20101111_AAAWGN fensin_m_Page_012.QC.jpg
521ff88cf4097ddafe61f92691d16e79
3abf5356c451f5713ff03861388fe433521269a1
2517 F20101111_AAAWFY fensin_m_Page_186.txt
50fb9de4db8bcce2d6e9c71548ee57a1
6897ea6bd2ae42654fe1950e2cc247d77dd84d86
97760 F20101111_AAAUYF fensin_m_Page_006.pro
5a3d874d743442315fe56b54ada8eeb4
f64a0cf0982d44cfbe10e86f261874bb49bec676
6534 F20101111_AAAVCW fensin_m_Page_090thm.jpg
954730ee890426d457751e015152da99
eca8c570442515f692dc09da1c8502c431fcdc7c
7045 F20101111_AAAUXQ fensin_m_Page_103thm.jpg
e68b543d0869eda88b2d8ff2d4c308c5
0a0274ea2c19f759f8f406f23305d4a7027faafa
F20101111_AAAVDL fensin_m_Page_161.txt
14324a9310697ddfc740a2aab64654c0
81542bc5e5f9e5ca5ea11d8434ca4c609cab92a4
6881 F20101111_AAAWHC fensin_m_Page_026thm.jpg
39e3751aaa7a01d37926f4597e48fdcd
dfc4ec18f4f599178775925c30da04271ccae6cd
6444 F20101111_AAAWGO fensin_m_Page_012thm.jpg
78daf81098f6bd3536ae17e4c1f47598
aa20f131ba34657da011c97beee9b40ed395832d
3109 F20101111_AAAWFZ fensin_m_Page_002.QC.jpg
1ed5499303fc86241957ddf12118305f
4668fd0dd1e3a1054e6a00b86447d3cf1e93123c
F20101111_AAAUYG fensin_m_Page_008.tif
b86502b8939829e74219eb4a910a627c
26bd4f4750461b1ae8ab919eb57cd6c76a74c7ae
F20101111_AAAVCX fensin_m_Page_090.tif
ab640147bff354782d323bb489aaff16
5bc6ede115afd7ff48aa151dc887a207305cb554
25754 F20101111_AAAUXR fensin_m_Page_151.QC.jpg
cbb46194e25dacd4844d4d7722d0ab15
9c5d1ce1c2727ae4e8ae7bb043c44bafb006569e
1982 F20101111_AAAVEA fensin_m_Page_043.txt
4f22d150ba3db47ba3a960c030d0b130
302d9dd98144add8d985c04773ba179abf0f2e6b
6680 F20101111_AAAVDM fensin_m_Page_041thm.jpg
2751d9bbbd7a42d3bb8865aacf265376
d2f15a469b7f1a56807528ce0f3a67d194d0eec9
25745 F20101111_AAAWHD fensin_m_Page_027.QC.jpg
a52b7d70ab573be10369fb06e967c0ed
bb1f090d1aa4d0269bc16e98e71396621c212a92
3894 F20101111_AAAWGP fensin_m_Page_013thm.jpg
95590acecea8d2278dfccdca5fbe4315
99e42a5e79620690943c2190b741382edbbed16c
118250 F20101111_AAAUYH fensin_m_Page_137.jp2
028badf6ecaa30822a5396f67ba76b31
4b366b7c4e95c33f2c520e0ee02ed6c735e48977
2051 F20101111_AAAVCY fensin_m_Page_091.txt
a036e9ad25ce6f62062b46cd266c742a
ba6b186e14d7ff08db98aa2d3a7469f6ab4b0462
7075 F20101111_AAAUXS fensin_m_Page_170thm.jpg
036e0c75954a4eaae99ba7cfdf7f7949
c3e7d3b38ded51f8caa0c26ba029c3750f149379
81082 F20101111_AAAVEB fensin_m_Page_138.jpg
75550e1b63679884cde4bdb757fb74be
e644fa4a87fb2f5281cf58b384f000b0e9770e7b
27229 F20101111_AAAVDN fensin_m_Page_040.QC.jpg
8b6c410d3730d1ebf0200b5ea6931da2
4ce960e8096674561cbcb15c4fdecffcd55267eb
6845 F20101111_AAAWHE fensin_m_Page_028thm.jpg
0fe2a473569ac3ef3e6a88ddad2403e0
c4fbd8f013b10abac1ed01196b77b400f080d1e6
26204 F20101111_AAAWGQ fensin_m_Page_015.QC.jpg
e4246d9cdb173398e6d856442619378d
9e09343edb59fe6a608351846257ee0697e2e37f
7140 F20101111_AAAVCZ fensin_m_Page_129thm.jpg
d1286ec6cafe9deb8db334c6795e659b
0cd9b25e639734f18b2b0e944abcf0d2f651a6d1
F20101111_AAAUXT fensin_m_Page_015.tif
2da9f18b7a9886257b3c21076aafd38c
8a4acd075b0662767c0e32b40b8cadee0791c684
2679 F20101111_AAAVEC fensin_m_Page_001thm.jpg
09a3dfc0a137412b4c57c3c338971d4b
100e34404a323b33d63e6731dd202ce82501e40e
49931 F20101111_AAAVDO fensin_m_Page_052.pro
543590b4c87817bc56b009b3759a5704
0ca8e45fb367466c6a3f9a866e2a532d05ac9f8f
6859 F20101111_AAAWHF fensin_m_Page_029thm.jpg
08b8f113a82c0f12f33b11812ff70670
ee78e48cc1454f9ff37ad1ae8e6a80d3036a1ab3
6804 F20101111_AAAWGR fensin_m_Page_015thm.jpg
ddce9fdd919ad1d723160b60ed29e06e
746cf9c4bb65f751ecd3e6ce9ca6595960f41e7f
6550 F20101111_AAAUYI fensin_m_Page_108thm.jpg
3518d0a96433504ae3a43a35bec615fa
763e204937976d638b0d47b7d3f558aea950c339
64329 F20101111_AAAUXU fensin_m_Page_164.pro
fc7b48912a7f197bc4a562a83fc831dc
0952786c93571ccf51e2bc14044aedcc0131cf75
122916 F20101111_AAAVED fensin_m_Page_040.jp2
2b93f4a26ed06add61b953d7426746d8
025a690bc5d7720443a0ad48accd8986cd42324a
6874 F20101111_AAAVDP fensin_m_Page_171thm.jpg
57256b60861ebdf1b3b4c888f10286eb
82778f57b0bb9718c54c7aa61637fed403b33fdd
20537 F20101111_AAAWHG fensin_m_Page_030.QC.jpg
8172622a08c812c2adf2a70512bfb31e
88e8bd441bca91b528c249a1f7f9ce60948ebcd8
6634 F20101111_AAAWGS fensin_m_Page_017thm.jpg
56306e8e9c04c7beca69e5032fc88532
33b1cdf69bc56ee069c6a9649447b3cb184c4e9b
25630 F20101111_AAAUYJ fensin_m_Page_121.QC.jpg
2b59303b23b43bfde0aacdeb3d2c138e
105fdb6529ebe7f65a060085b2b699431e827b47
32695 F20101111_AAAUXV fensin_m_Page_076.pro
2f2de4932eb4381ca51241a964f99b56
4d5bf8ccb4eab6ca43273f182107923ad09ae2fb
24248 F20101111_AAAVEE fensin_m_Page_118.QC.jpg
eec7d00600b0ade62fb5a2380635e6b4
184a2eff0e4f6ac7fc72403c6a888a1e63e469c8
71830 F20101111_AAAVDQ fensin_m_Page_053.jpg
34ab4a455b1f0c5ef261ec1b915b518c
a83d9ba1ae4a8f0e09f13fc5a4ebfd5efcaf73c0
5634 F20101111_AAAWHH fensin_m_Page_030thm.jpg
01927209ee6d1f502934cd9d421b0942
902e79f4a2cbf6d9a5015e4dd6a463321c5e1679
6679 F20101111_AAAWGT fensin_m_Page_018thm.jpg
4edb8245fc5e45829a73c4e2219fa7db
01f39a9654acb209e66680b9eec845523ec63be3
114775 F20101111_AAAUYK fensin_m_Page_173.jp2
d6da090e3d7920ed8f3ff847561bfd2a
46abef551e95c46b539b4a0f9a710b5353484a18
78955 F20101111_AAAUXW fensin_m_Page_136.jpg
76790640ccd6f0ccf5a8eb555b7f739b
1d9ae02567e48cb44272fd9671c7828b20ae2467
2060 F20101111_AAAVEF fensin_m_Page_042.txt
070354b5e73a5e2831352a92ab0c7057
16dd1096a63aaa9cb71553bef8182c796ca97008
F20101111_AAAVDR fensin_m_Page_074.txt
b39b1221f6914e358e7dfa0c1e379bbc
5ed00d359ac927e169fdff2d3d884d2bac464d06
17627 F20101111_AAAWHI fensin_m_Page_031.QC.jpg
4a513b19e12bf07e10c6488a3e1ae56a
e24bd2ecac42d4c6fd5419393490e56029f8599a
5902 F20101111_AAAUYL fensin_m_Page_085thm.jpg
1123988ba75ce579222acf7b25758785
67ab1223d2cae10f253a62a7e4ae8c27ecfc761e
81814 F20101111_AAAUXX fensin_m_Page_137.jpg
3ffb2e755bfdcc9a51273709d37a1d5f
e230c911bedf1e1e96850bb8e4740cfacd888dc5
58433 F20101111_AAAVEG fensin_m_Page_124.jpg
f0490902dfd892c218d0e307efe266f5
9c9e1b484b3df838356de147cf8e5138b0593e37
79656 F20101111_AAAUZA fensin_m_Page_093.jpg
c8855aa145f661cfe9d87e0864271800
dcce8746485efa83a06283de58b1d9e25985aefd
5205 F20101111_AAAWHJ fensin_m_Page_031thm.jpg
c0f565c3fb05bb8c4d95e016f1b94efa
9438064ef3fb8418d8e0e442140b430fff3dd4e5
7023 F20101111_AAAWGU fensin_m_Page_019thm.jpg
ee704ffdf622f74ebbae2245b39318bc
9f674b754514998dafe950d1640e16ecb927e9d8
5430 F20101111_AAAUYM fensin_m_Page_036thm.jpg
66867f6c9a4f2b43a9adba5e80628ce8
dcd99cffedb6b32c9639d0720462bd2071920ff9
25851 F20101111_AAAUXY fensin_m_Page_029.QC.jpg
4f6772f863e675cccb223373adb63d70
2d15574cd2be64b07bd14e3458d844b70905a53c
51387 F20101111_AAAVEH fensin_m_Page_159.pro
141df405e09fcb7d5da728b446706eba
83bec62a5c7cca7184c60e4c9496eb445448897a
64966 F20101111_AAAUZB fensin_m_Page_182.pro
3355cae544f5d5202af10daac82d062a
cf1e929c5357090c6597d6633664a212df2b94bf
127680 F20101111_AAAVDS fensin_m_Page_179.jp2
113bbf46d7b4c354003afd7b5a36c0cf
d98d3a8720e396d21f38ecdf184d8040dcc81399
26688 F20101111_AAAWHK fensin_m_Page_032.QC.jpg
35098d757cfdf09822af8957301b385a
b8a951ce09f663673b68c018324e1515b75f2eea
25236 F20101111_AAAWGV fensin_m_Page_021.QC.jpg
1383b3cad09e287100969c430458b7dd
b1eaa0307ee4d54ecde09a2a761204cc4a6f6b79
55786 F20101111_AAAUYN fensin_m_Page_100.pro
1d16e5f65fb34eaf8480107264e0b441
cc9171a31604c0c67aec9400d7c1c3871f1cdeb7
53547 F20101111_AAAUXZ fensin_m_Page_068.pro
f60b4122059421b3dbf530ca364581ac
ebb26ad1d1200f7f62fa1ad4af7f945d100a4eaf
F20101111_AAAVEI fensin_m_Page_168thm.jpg
ef5fe790a9d2e422d83123b732030417
893b066316f9d63d14d0337285ff091e4e3d71b9
83179 F20101111_AAAUZC fensin_m_Page_015.jpg
7679f66ea71b45396fd00d85a64e2c3e
ff080bc5402196c2f942263af9bea5835488185e
21265 F20101111_AAAVDT fensin_m_Page_115.QC.jpg
de5a50a555c2e661d1ebca6e72654c6e
33941cfaeeed09bfd2f49763afba74fa22a10430
7172 F20101111_AAAWHL fensin_m_Page_032thm.jpg
e4bc08a91d9332d2fa6a0b68b776ee21
3076761bd66c7e6d68e546e93b211f85c0a957f7
6632 F20101111_AAAWGW fensin_m_Page_021thm.jpg
cbe882fb2b7f0f45ea80fb034a9c329a
f61921da14ac070fe0a404d45b9d7f4f659bdb40
26383 F20101111_AAAUYO fensin_m_Page_019.QC.jpg
fc55c53fe5158e481ed9db52ebf5dd66
f5ecf4a63c3aa6a704e6117601317af25065d076
24274 F20101111_AAAVEJ fensin_m_Page_074.QC.jpg
b7c75dfc827579f2d2197f73d922542a
a0df8b432e32a94a28750f25c42ef5e12c74fbf2
F20101111_AAAUZD fensin_m_Page_049thm.jpg
1311879945f19093c516b024fd08aeb7
1753ac97a7749e7e542c7ac224fd36a7a0ecb7b8
75654 F20101111_AAAVDU fensin_m_Page_167.jpg
4e1b9d1c7cd75af23fa7f278946132e4
1fa22a44f04865f1684e441a88d1fd2b5ac43f6d
20059 F20101111_AAAWIA fensin_m_Page_043.QC.jpg
d3012addc93222c39cdb1e3d1051708c
5dcfc8dca43c41adc5b05a410e95d711dbb7dffc
21094 F20101111_AAAWHM fensin_m_Page_033.QC.jpg
07f6ca9c29c8e41a910509045c6b959d
ec6e4b2b639e7cf1d0e96098c518d10f293c8071
25111 F20101111_AAAWGX fensin_m_Page_022.QC.jpg
73f8dead7a58dc44f58d8ad9b62a60b1
5b36f7865371d2e54fa8f288705dbe6d191fda0b
F20101111_AAAUYP fensin_m_Page_088.tif
8ddd93cd5cdeca481ee672c3831f6d5b
a289eb2852f10819fdc9669c58e8057035545733
110389 F20101111_AAAVEK fensin_m_Page_034.jp2
91cdf2f4ae9e4a5667819f3148b580f4
e0e0ae50d6fac6c227b56bf007b4d74a1cea8a15
71782 F20101111_AAAUZE fensin_m_Page_147.jp2
4a85b76988a9cd98b3f12fe1dde87037
1e117d1ca73eba6d8e5f55c58c3e1b36ae56d6c4
F20101111_AAAVDV fensin_m_Page_140.tif
3f9d0dadbf29e0294a346b9d3cc6c676
26ff8c198f8868703bfe0dc0ad61ee9141e2b483
25988 F20101111_AAAWIB fensin_m_Page_044.QC.jpg
8cb76dbe6320af572e56733043ae22ed
02608378337dbbcef020c75711e3e10e9c541eca
6417 F20101111_AAAWHN fensin_m_Page_033thm.jpg
2e56bd0dd2abe7145b8a9e32a5f704fa
d97b886d933afa23c26b745d786ab8ebdb94ef89
25290 F20101111_AAAWGY fensin_m_Page_023.QC.jpg
ba184d4901b34caf87e5c4f986b440e8
cfae9b4af454aa7918050ae6ba42df6010db7286
15996 F20101111_AAAUYQ fensin_m_Page_107.QC.jpg
54f7ce4115b09a58908a17de311f0115
7dcbdd6a22b1117643e3c67a6c230fbcf6e8e8b0
F20101111_AAAVEL fensin_m_Page_153.tif
704dd7d5daa38f52dc3a0cbb56887b18
bcedf318a51476a27c7e4cde434010b511b4f56c
52380 F20101111_AAAUZF fensin_m_Page_063.pro
72c35b1b0af52b12d1d0184e14b845da
9d9f14e041ce97251a00adffb2a0245d61814db5
1051974 F20101111_AAAVDW fensin_m_Page_015.jp2
3bd298756ec975240089ea439cfcd492
d66d549cc6bd4047ef78cb493e0dba66896cee79
7152 F20101111_AAAWIC fensin_m_Page_044thm.jpg
a0e0db019cccd3af78d024255c6507b7
758d505f4b99fd073985bff7db6643024ac1b900
24345 F20101111_AAAWHO fensin_m_Page_034.QC.jpg
12cf5c70aeb54cf88bb6dbe4b8377efa
fd4270f903dfb26fa9c65a2fe12e497cd2633ba3
25519 F20101111_AAAWGZ fensin_m_Page_024.QC.jpg
fb90ea4d8de267ac3208e8291f124825
79f8964a258d316a89966cfc420498be1ee39807
1051979 F20101111_AAAUYR fensin_m_Page_104.jp2
4d2326a5afc514e7a16dd1297d020b09
222cafb224e8633f83f0ffc870e2fbdb761e7fc1
2073 F20101111_AAAVFA fensin_m_Page_094.txt
11d869a7c954f22a313408331f619961
ad14dd943fbb8d5748c810bf59d35b0b3de97dce
1051905 F20101111_AAAVEM fensin_m_Page_146.jp2
7def222ebba0763afb84f5e6d7b124f0
ec8b324f2a1f4438d886d1537d8f69a2c5df015a
26134 F20101111_AAAUZG fensin_m_Page_171.QC.jpg
264cb557f79a1361f1813c2196874268
2f4edc5c108ce80221b626b6da860a62ad07de79
116195 F20101111_AAAVDX fensin_m_Page_153.jp2
82e7100ef5e45d2389a88e65d3e04e7f
04a24ef4feb9ce9fcc45461bb086bb32d42371d0
25644 F20101111_AAAWID fensin_m_Page_045.QC.jpg
cba6a13b61cbf799d5a302631ac4c3b4
6bc77adce341eadb5b654db656fa3248bdd6d449
6768 F20101111_AAAWHP fensin_m_Page_034thm.jpg
cc29984dab60a0a5dc9547e454cdee73
a2467257659edbe8d68fda41fc2850070d111927
20377 F20101111_AAAUYS fensin_m_Page_042.QC.jpg
190ed522bd7187ec7430bbc519f3355b
54358cfc29cd095cfd8027d37f6f6e0367009f9a
73546 F20101111_AAAVFB fensin_m_Page_118.jpg
0ea90d1c574d76b556858da36f165310
12706ea2269c0dc0f27adfb759ddf386cdf03fd8
95378 F20101111_AAAVEN fensin_m_Page_115.jp2
40b4474298e95904419a736a47044911
dcda6cb87965523add26afea0c5ce7ef928a8c30
F20101111_AAAUZH fensin_m_Page_142.tif
79e9b1d72ea0be00ca843e5ce95460e6
011694a59b8df07dc83dbb7a7abbada79fb5c10e
74480 F20101111_AAAVDY fensin_m_Page_086.jpg
bccf64b38ef9c2a39ecf8e4463af03df
c23ac366cd3e8374cfa809862c44f79bd6508e03
6853 F20101111_AAAWIE fensin_m_Page_045thm.jpg
33923512b34116503cd4823e7ac5a7cc
ee632f3b00b764dc58134f78bb6646b9879a5244
25862 F20101111_AAAWHQ fensin_m_Page_035.QC.jpg
26a3517d45d2057a114a5df01948a8f7
81c032742df0828eda69b86c77140c710976eea3
56925 F20101111_AAAUYT fensin_m_Page_065.pro
5f3abe98ffcb1ede9e19808a9e95ac05
91dda24b9d9bb35f5a98691c76cbefc68062ac28
26628 F20101111_AAAVFC fensin_m_Page_175.QC.jpg
db865b96250ae898645aae5926f797b5
f5f25b59b85297a62a9e35ce13af65a9c7e5952f
58372 F20101111_AAAVEO fensin_m_Page_149.pro
f5741f6aeead97a6e56d1a24a159f562
4a850175fdeba0a76c09ad9d366dc58886037957
2134 F20101111_AAAUZI fensin_m_Page_143.txt
d43c6f02b73007bd2567cbed7630b7ea
0ac5bd30fef723204d743af3c22937ad63ae8173
F20101111_AAAVDZ fensin_m_Page_112.tif
a35bdefdfa8f1bcf628971fa785eb17f
f4197342ed128df69a712c6e796c1cadb2aad5e6
21188 F20101111_AAAWIF fensin_m_Page_046.QC.jpg
59705a95246bf1812f54e691acd860f5
a05c95a8929351a1bad6964c1663c3d9bc417872
6999 F20101111_AAAWHR fensin_m_Page_035thm.jpg
c547440c74d9130708aa6dc12202bd67
b768e59408e1833e0e4519dda3f88e3dd066cc07
52797 F20101111_AAAUYU fensin_m_Page_099.pro
7d4ad9d74cf245a018d1716e0e4216da
f37b428b93d6ae3b0ad4a38ac3e5502f65d689e7
51456 F20101111_AAAVFD fensin_m_Page_108.pro
dd008ad937bc332b62f323ac91079a90
802d7573f133f98b1cc8b1d45d19412db21348a6
87108 F20101111_AAAVEP fensin_m_Page_064.jpg
55d3a8926be8f4b3a2d339b1982cf88c
653043bf7d3fb7a1278e8c00ebd9ba9acf21f0eb
6353 F20101111_AAAWIG fensin_m_Page_046thm.jpg
2e0a6e031f3ef5e853707caed4d3893f
b70cb76aa11cd4ed917c13f35e7f51bebe82c860
19030 F20101111_AAAWHS fensin_m_Page_036.QC.jpg
0bfcf274fa0afbe123d3b7ca935e3192
83e8379a5e3d81f929acc9dfbdc0c88397acf38f
21553 F20101111_AAAUYV fensin_m_Page_061.QC.jpg
def25879e490472484d6473811ff8967
a2d612ca30b717d22e53dfd67c5a94d18836e081
60624 F20101111_AAAVFE fensin_m_Page_056.jp2
0c1fd15824138ab7882ce267bb88349e
a017876995e5e743de0474440b5d981259137dd8
2193 F20101111_AAAVEQ fensin_m_Page_122.txt
8efd4f54750ab167edbda4f51cec5b65
7e4869fc9e138452686ff4f16ad9b3648f2fff9d
79027 F20101111_AAAUZJ fensin_m_Page_067.jpg
15e83260131b21e88fa7a8a2eaab82f8
e4b0b78c5cd26f5e8ec33416bcd762cbf0799b1c
26079 F20101111_AAAWIH fensin_m_Page_047.QC.jpg
7eba251d5b2bfa0b98238ba6ecc5c3e8
ad71b3f2a0d004c6a09fd4c868a773ed34ff3e0a
19430 F20101111_AAAWHT fensin_m_Page_037.QC.jpg
bb022fee249f4b758708645243109fc9
3f8f8cf3fca412d174ff28ba37888afb9a723fb3
24079 F20101111_AAAUYW fensin_m_Page_172.QC.jpg
a2d307074648198adde16f9817368055
bf7476782b5e8d7173f55905000eae5dd4ff15bd
13062 F20101111_AAAVFF fensin_m_Page_005.jpg
e5543ab031a538b5aa85652c4a40f9ea
035819198bdb11d34e4fccbd78fc1ae2303d2eed
117601 F20101111_AAAVER fensin_m_Page_126.jp2
4b65fcb4af9cdc27883ec4e712d81a54
94460e1fe77ce1bef268957eda36db18c34b2317
1991 F20101111_AAAUZK fensin_m_Page_053.txt
c0f730e581746966a2e28d5fc40a5dfd
55f2bd2e074187c7386485c214e947a21faa1b54
6922 F20101111_AAAWII fensin_m_Page_047thm.jpg
0c966256284feb4dc4f73466a07d307f
b70a3bf1b991004922a7ad31cbd9a4a36aa0509d
5842 F20101111_AAAWHU fensin_m_Page_037thm.jpg
e0aef3fa583f8aa6f9ebcccae58cfcdb
aa5fe64eee78a5bf95ca078618ab7b0b6ff7f50b
25626 F20101111_AAAUYX fensin_m_Page_149.QC.jpg
aeb4d549e036a6e88141601f36fc7b41
624b485ff4226374ec1642e861d5c0da9dc4133a
1235 F20101111_AAAVFG fensin_m_Page_070.txt
5f43f410e380d0c6d9e3fa47da22a3e7
368caa38ee2f3b0c62a7e9b36365b38a9d1bbe67
1051984 F20101111_AAAVES fensin_m_Page_008.jp2
1232a07871592e4fb2bf15959a1fff90
da99432f90e3ada8321ba27ddd0552d23bbaaea8
119127 F20101111_AAAUZL fensin_m_Page_067.jp2
aa290ebd91d8edfb11ad75a333dfdb16
a356794f9338b0c66bace8cbadd297546053dbe0
24051 F20101111_AAAWIJ fensin_m_Page_048.QC.jpg
9276159a709fa240ac317e06db3d5471
9d3b6dc1b39ccc1e8d6406a138e2c2d0fd8fcfef
24596 F20101111_AAAUYY fensin_m_Page_168.QC.jpg
3c45fb1be9bba6f895d85acbf5842f04
562032e0cfc9728822b39a270cda5ef9dd69a614
40859 F20101111_AAAVFH fensin_m_Page_112.pro
09024c7ca0b931a96a048f19c1588463
c2e8bb24ed4cd52c0b41c40e382e327815c1b156
66825 F20101111_AAAUZM fensin_m_Page_135.pro
9ea928b9244280727fdd7abf4424ceca
e2989e873b3411ce70b17ac38cebbc9022cf3e3e
6514 F20101111_AAAWIK fensin_m_Page_048thm.jpg
196d177b7e025dc3feb89d7a1d9bf420
1cda42a4b35287b7d018d320ef054dd1098c8ae8
21022 F20101111_AAAWHV fensin_m_Page_039.QC.jpg
0f4e31e2355417232c2e952a81960dec
12b1f14f505e48b2cf643171b141d7fc7521b51a
115606 F20101111_AAAUYZ fensin_m_Page_026.jp2
977ce26c43bfede40928fd746c919f70
4d020ef9d62f0ae161616798dc97dd55a96e1b2e
13326 F20101111_AAAVFI fensin_m_Page_013.QC.jpg
25842e3f05cd3295416a20c721ef9a67
542ef49f8ee887116ab0e17b0765261b381ac1db
2278 F20101111_AAAVET fensin_m_Page_004.txt
53208683e4ee0987618a16b06c97abca
2380e64e5277b3260b0afd622b12aa4b13e3081e
78640 F20101111_AAAUZN fensin_m_Page_171.jpg
c913682aa2a9bd64f2ebc05c7859005e
72145b371974b0ad4c63325071d8eeca770908a2
22246 F20101111_AAAWIL fensin_m_Page_049.QC.jpg
834415551b5f7798bac5dd19ddc1e662
1b52a171bd6435bec5d63827784bdb1a65f5da29
6211 F20101111_AAAWHW fensin_m_Page_039thm.jpg
637a3feb13974b0b20d3353adc16fbde
578782d156cbb4f29fe08517eed53466aaf919c1
64958 F20101111_AAAVFJ fensin_m_Page_188.jpg
ea84e4267c1cbea90ad9bd96868a3724
2ffdb33e0779582ddbc33617067f2207e05bb305
81187 F20101111_AAAVEU fensin_m_Page_073.jpg
da2bb48c1dc22c427495035af547e219
0547d2129b0ca9a16f169ffb063b48ae6ea9dace
91006 F20101111_AAAUZO fensin_m_Page_162.jp2
9d46c170ae14f754bf1f3f36a1406e1b
0fcfe3e51ab191acedf242257b37cece20077b6c
19363 F20101111_AAAWJA fensin_m_Page_058.QC.jpg
819fb7f8e4da92aec82b528705141ff6
b8116b67443d14a05dca8414bb954629a4ca0fd9
26425 F20101111_AAAWIM fensin_m_Page_050.QC.jpg
c12b67416b8826a5174c8ade27dd03bc
4cb908fb0ce9e57db925cb5fe6754796fa6e0bea
7243 F20101111_AAAWHX fensin_m_Page_040thm.jpg
9305875f60a1ae5557cecfe1222d21c0
e858e8b3e31f62ad34d39c0f996b92dc7f43363e
5942 F20101111_AAAVFK fensin_m_Page_043thm.jpg
4dfd4c98b4d47922fff2ba9b7470e22f
db60c6798ac8521397af4fe0937b198d979785c9
F20101111_AAAVEV fensin_m_Page_093.tif
71eac97f398b1d59c3b45d5825dce29e
25fea8c1720b012ee8fa9e532d7a93d9be608986
1051971 F20101111_AAAUZP fensin_m_Page_007.jp2
4ddf4a766bae0a8a2a365ed30392a918
87004b42afd70e04b9319ffbe329c4b1b9cf14df
15963 F20101111_AAAWJB fensin_m_Page_059.QC.jpg
b69388473b8e089a4f8c58b869381c24
005a4c44d949e8f48f9e5bbacbec70b6bbe61524
7082 F20101111_AAAWIN fensin_m_Page_050thm.jpg
20d3fcec960483aceb527ef43ddfa79d
3efd3e59af2dd3ddc72d9dbca28f050acfac9922
25154 F20101111_AAAWHY fensin_m_Page_041.QC.jpg
438ab1446812ae9d9194c61ca8aabe9b
d23f8a71c578f50fdd82f0cec577663aafc3afce
98859 F20101111_AAAVFL fensin_m_Page_182.jpg
0826b050541333eadc9c092944c59871
7759ac66e20f788165fd96f6e350ff9be6fdb10c
61116 F20101111_AAAVEW fensin_m_Page_184.pro
e82b93e8425c9035d9c8693f7a3f51f7
791141127efca47f0c9b998f2fb8b42f43024243
61409 F20101111_AAAUZQ fensin_m_Page_150.pro
4a8c3f39349da17eaa2167955e36014a
d0af22b3e40ebe4ab94e7fdfedfa02ad3119bc5d
23160 F20101111_AAAWJC fensin_m_Page_060.QC.jpg
558d26fe8ba05d201c63b3f3d0d96139
eb22dcf2b281e7ecdebbd1ddbe0f30409dbd0018
5588 F20101111_AAAWIO fensin_m_Page_051thm.jpg
b2188c2dbff8fb4f2dab9f29f10fcb42
e999075d60f10cb219368899c43a251351019c40
6530 F20101111_AAAWHZ fensin_m_Page_042thm.jpg
a14ced3c5b0b6b9aae9859da748ed278
8e3c02e46b94e54d982fc80e0648d9fc63d2f478
66002 F20101111_AAAVFM fensin_m_Page_038.jpg
30269a305fcc68a102cecee4bf7a24d0
47f3b20a51a3b631079031cbd2dcd7773ebafc27
1890 F20101111_AAAVEX fensin_m_Page_089.txt
2abddbee40db47c13664fbf7b31f590b
a991e09d8c6cb8919761ed61828d6c74f2211838
25270 F20101111_AAAUZR fensin_m_Page_018.QC.jpg
a40480c45e99d695be373376fe0d6aaa
6425d02a82fa8e58b952e8baa7a0b58210e9d687
90741 F20101111_AAAVGA fensin_m_Page_010.jpg
3230ae99d4b599979340c50e9c5347b1
6bc7b851fc865c0adde00e70797e95ff948a9fdf
6434 F20101111_AAAWJD fensin_m_Page_060thm.jpg
c9a8d61dbd8fed396e663454bf352a8a
459085ff9039805145da20906df0311e9f8d09e8
23998 F20101111_AAAWIP fensin_m_Page_052.QC.jpg
31f72704f561859d7f14c617af05895d
21b278b0b2e5c034be3010684ba863a511e38ac4
67863 F20101111_AAAVFN fensin_m_Page_009.pro
17e1fb69ce306ca8f163e36991a2b9de
2b88b527740f666979692126e81581b9c7ed980f
25166 F20101111_AAAVEY fensin_m_Page_134.QC.jpg
3a34249f2df2bcf713eb9cb5372d05b6
56da7954d0e5b70b3682fd3463fab81ad3924414
112185 F20101111_AAAUZS fensin_m_Page_074.jp2
07fb2f8bc57d2a8dd7478ed38522e10b
295eb445ad52926648b2f80950b6de5d5191f941
60408 F20101111_AAAVGB fensin_m_Page_011.jpg
906c381602f62b4ef1aa6164966bb3b2
1a04cfd9df5504f40366c9975bbd9f87e9ad16e6
6217 F20101111_AAAWJE fensin_m_Page_061thm.jpg
b693519c2edb99aec589bec56f5e2dcd
a8327f61daa22b30222f49f3a6b42036077ace75
6264 F20101111_AAAWIQ fensin_m_Page_052thm.jpg
ac1f90c5c65be4c3d730eae60c62a4da
751e3f3f5c0b12ee4cfd8fb2fe2cd40d2cbce514
73163 F20101111_AAAVFO fensin_m_Page_108.jpg
0c17f505548342794d39c925fb4aaba2
93352d06ab6f2883d91550a484247fb398927502
2674 F20101111_AAAVEZ fensin_m_Page_149.txt
bf70cd879204fcfa6ab58e48cdc11874
e0b3dae5a63915d0dcfbcff184b9fce6f7eeb29f
F20101111_AAAUZT fensin_m_Page_106.tif
0ef4446bc2c85cd506bdc964472e2ece
44279b175b619e1498999a94861504454c7114c6
40446 F20101111_AAAVGC fensin_m_Page_013.jpg
2dbd6efeb1f51acded75cfb94a4026d3
fef47481e04b6000d6cf26f1e821325402192eea
20411 F20101111_AAAWJF fensin_m_Page_062.QC.jpg
6c5744126208c160ab05e820a7ee328a
4f47bb4a8b9522457e2158fc0492518afa00db3f
22332 F20101111_AAAWIR fensin_m_Page_053.QC.jpg
840e2eda53c98b6999cd99b5086755b1
192495eef236de6781db1586c7913afcd0c7ceee
280324 F20101111_AAAVFP UFE0021946_00001.xml FULL
a4d279732c11664a9de8c2ba5087c86e
d8eb77aaa2ef952b429e23929833acae1dc3ece1
41625 F20101111_AAAUZU fensin_m_Page_037.pro
3f655f119b6691c5c7fb4e92c1b6592a
db080bf9bfa993626f78c0cff9ebdab4c5f1364f
75042 F20101111_AAAVGD fensin_m_Page_014.jpg
af0d4e4cdb3d8672f5e029dee02abc66
87a25b53163d01103f4abe986ec377b099ae9e9f
6053 F20101111_AAAWJG fensin_m_Page_062thm.jpg
c763b44a68396593345d40a9305cdcbb
5b57befb7793225d6bb1eef9e76dd1dc9f59b48c
6526 F20101111_AAAWIS fensin_m_Page_053thm.jpg
2950532b43b760dfce4f51dc93dcdd35
dbcdd9df5a1cd720b310c4c1e73c52781b91af49
24539 F20101111_AAAUZV fensin_m_Page_014.QC.jpg
e8510ddc16afcd1eda22f45905307fa2
1de493018aed91f0a2c2b5eecebba99b7ff0c26b
76557 F20101111_AAAVGE fensin_m_Page_016.jpg
2e1d4e9aa2648a857fd148800e9104bd
8491b203707fe0cbce62321ccbb11f932406ee3a
24665 F20101111_AAAWJH fensin_m_Page_063.QC.jpg
f0c0e5aff00af80f20366b1d08b1d067
e8eda32719340ecb3c57ad449d7ecbad8a0740d2
25542 F20101111_AAAWIT fensin_m_Page_054.QC.jpg
066e6c49f4c3813789a984ca6b95c35c
e9475b0a710187df3d3cc0596000b556d0f55332
78508 F20101111_AAAUZW fensin_m_Page_165.jpg
d93de250a74cc8fd43eb6b6f6ca14b9c
fa0f2426a6b53c36716f9f9e56091ebd1f79b1be
74859 F20101111_AAAVGF fensin_m_Page_017.jpg
e8c405b53271964c59aa4179cd33e641
c564de1b503b4d6163ddf3946d9e0e1ea3e7c2bd
6741 F20101111_AAAWJI fensin_m_Page_063thm.jpg
e711ad98b964c7d79ac557cf1cc046c7
9c36b330f4783a314435da7066b0e78914367be4
6848 F20101111_AAAWIU fensin_m_Page_054thm.jpg
0256741faf7b8ddd6725b03f4566da12
edc20746827b5e0c1211ff2f9547dd29160c9404
28493 F20101111_AAAVFS fensin_m_Page_001.jpg
c8dfeb7f3a9eafb408705249a981054d
84105364ecc53cb8bd1e6ceeab887cf82c906a71
8535 F20101111_AAAUZX fensin_m_Page_001.QC.jpg
1d3460dc20352b53f5c97c23f9388219
8d39c99d1d172bdcdf10ca66a13238d00f01d635
76724 F20101111_AAAVGG fensin_m_Page_018.jpg
a6815fef37b9e8f717e722daa475ff72
d614e6381651b91466bbeac451a1cd7c23e7698b
27193 F20101111_AAAWJJ fensin_m_Page_064.QC.jpg
629d0450890225111d6fac6389eea4b9
0f7afc23f3651bff3a17d92b99f56efe61bb97b0
22074 F20101111_AAAWIV fensin_m_Page_055.QC.jpg
974545d776b8959bb7cc078707a16103
7e7deed1c7b30d70b54a5eb66125eb4627d2bc1c
10077 F20101111_AAAVFT fensin_m_Page_002.jpg
24ac85bf8d85556641901a29a55dfb83
ff6d57ba32b02b2e1572a5c802b2b5d8dec87464
73655 F20101111_AAAUZY fensin_m_Page_144.jpg
fc9fb76892587b846bc0c44b36d30cd0
2a5213f011dc8c549328b7b0086ffe2dd768d493
79967 F20101111_AAAVGH fensin_m_Page_019.jpg
3c1e28b7726780170da3745c8eccfb3b
832707378582bf31041ec1794bc73e46128734cb
7500 F20101111_AAAWJK fensin_m_Page_064thm.jpg
478dca85fe21953b8d139d272fcdbe98
39ab49a09fde30997413e7f6a34ff0f28e88bf87
F20101111_AAAUZZ fensin_m_Page_001.tif
b1dc6d3b28b857a74c8180c2a8720fed
b8dc7d02dacebc2baa420e61a2b0bcebe0488a13
76402 F20101111_AAAVGI fensin_m_Page_020.jpg
bc98a80215fc35a5c930676bb57c4577
6a1a5899a8d102f478b4e472857f7f6b7473f646
26749 F20101111_AAAWJL fensin_m_Page_065.QC.jpg
834247b8a882a73d89831e2e1005c215
e2b1c8ae06dce5b1bb151d599863ec977c3256ff
6193 F20101111_AAAWIW fensin_m_Page_055thm.jpg
e780748641ba6a95a8f3f11ec72c8507
eae23d6f4638fc036c37ad6cec2b2df419898fe4
10400 F20101111_AAAVFU fensin_m_Page_003.jpg
c679a40f8b925a21206dc749e1ad9eea
3ed7d7f3d60a11d23cf39fd99dd7c4a13250674f
77294 F20101111_AAAVGJ fensin_m_Page_021.jpg
8864b2c9244989aab728fe3b1ed4bf23
836686dd5e09f65ab2798e697e620a97d3493420
19354 F20101111_AAAWKA fensin_m_Page_076.QC.jpg
6fa3d9cfa28e535e4eca883fdb3309e0
c9118765f43d3fe3804ecea802b5d5171a8c8eee
26146 F20101111_AAAWJM fensin_m_Page_066.QC.jpg
d5eb7e4a06308253dc0ba9900af35a2a
3edef2ca3ad40316a3a72e2f886bb13a1af4ccd0
14826 F20101111_AAAWIX fensin_m_Page_056.QC.jpg
b7865bdfa3c365043f4c20cdbc12783d
18283e4c5175294c1ad6586780e45fb6d23d2fe1
83335 F20101111_AAAVFV fensin_m_Page_004.jpg
de29db7925fc3015f05284658f1989ff
dff1e9d5de0cc73b58e454f7ac04d16ddb5c56e7
79417 F20101111_AAAVGK fensin_m_Page_022.jpg
b1ed722a6a72db277c9ae7686472e2e0
03fd24b937c8fcbf98131b11310e678093671afb
5664 F20101111_AAAWKB fensin_m_Page_076thm.jpg
fe3596d86a93d98cc77fce8eb8cb0977
1499d4d8b4a54916e6358df7bcc3926c784527d8
6984 F20101111_AAAWJN fensin_m_Page_066thm.jpg
1236c1ed10b32a1ad73d13499cc45a5d
174231f4d97f96818c02a07af58ea2b7f9c31b20
4503 F20101111_AAAWIY fensin_m_Page_056thm.jpg
5ed9ee587f213b7eee1d01f16deae9dc
63822809adc373b8c776877a5c1e51284610b242
91949 F20101111_AAAVFW fensin_m_Page_006.jpg
7df6b2550ced169c4f67caaf0986b6e3
69fe46af15faf0577378f5a9f6c7fce70eba0ffe
76676 F20101111_AAAVGL fensin_m_Page_023.jpg
af2416cc564b8764ad35f8d0969b85da
b77a679874affafdbb67f4f530ebbfac1d1b3a77
27799 F20101111_AAAWKC fensin_m_Page_077.QC.jpg
ab2096d985f1e56db7ce993f2bfc9719
6358a823df5b1ac5842a4fee866e76c0ffcf1554
26207 F20101111_AAAWJO fensin_m_Page_068.QC.jpg
7ab5273c9ba7ab075d0511481408a8c1
699d26205893ef6d584c8ce1ee4e005eccbf78dd
5436 F20101111_AAAWIZ fensin_m_Page_057thm.jpg
f7160995c25a7565bf7ca67caa882aa2
7849b6f5125aa8ec395b7d534f51a1e307a601b0
81641 F20101111_AAAVFX fensin_m_Page_007.jpg
aaeea7101ece0fa165ccc9e6f96d66f8
3b55e8cfe6c8975c9148a283e697e60ca77391d9
78276 F20101111_AAAVHA fensin_m_Page_044.jpg
0664bc022d92afaeffbbdc14f2fa450f
8cba03632b8197c9f1076f42f96ec3d08ba24df6
78122 F20101111_AAAVGM fensin_m_Page_024.jpg
8e3df41a23a236410bd937d941ab4245
c88707036fa6a63994f5eb21c0d8cd36407bb85c
7166 F20101111_AAAWKD fensin_m_Page_077thm.jpg
a19f0c29cd04e0517b4923c335abccc6
aed70021d1db1c0b75bf95c2509d6de0a1080cc2
6890 F20101111_AAAWJP fensin_m_Page_068thm.jpg
21cbcc7c494b43c8b9e8646f0e58614f
31f9cdb8bdc8ee1608e87116e6f1268b88d29a5f
84480 F20101111_AAAVFY fensin_m_Page_008.jpg
6e048cac5ee616bf6601a94293a3e45c
89fb5d26402568c9f23cbc50ea1272dee8612502
65829 F20101111_AAAVHB fensin_m_Page_046.jpg
4e7505e59ece6a243ebda4728d060526
3e08424cf1f791ddd3c79d3715eb57ea7872eeae
75899 F20101111_AAAVGN fensin_m_Page_026.jpg
18fef67ce3c98be229f9bd5b104d5e1e
5350bbf6f58a4622f11be2cd2025218ddbd88f53
5012 F20101111_AAAWKE fensin_m_Page_078thm.jpg
078772398876d8368bd5d38dc66886c5
16aec1295bd73774d5baab91da745fc0ace4b8b1
24444 F20101111_AAAWJQ fensin_m_Page_069.QC.jpg
19d7c9f82f6442cf6eaaec98c42ecd48
63f3c97608a8e08291f8d2ada9a030e55a1ec878
89954 F20101111_AAAVFZ fensin_m_Page_009.jpg
f56c0993d864f8d4d412250804689f94
a7c11c62e6f3481d067d50124d3efa37ecc31caf
78662 F20101111_AAAVHC fensin_m_Page_047.jpg
c39181a751402c710da3adb10cccc5d4
8a11935b02a93bf2051cda4dc4c8fbe14ff3ac33
78698 F20101111_AAAVGO fensin_m_Page_027.jpg
35db9e0014d1000b7441d88095f605bb
be245414937fbfa81ec00bbc232ccfa7772df125
19952 F20101111_AAAWKF fensin_m_Page_079.QC.jpg
ef9de413be207df3bca0d02b6dd5cd11
2a8b17a9fb7209f4aafc9401728a4e99228cd2f4
6548 F20101111_AAAWJR fensin_m_Page_069thm.jpg
924b3c77ad8f76999d13ff2cfeb70c63
dc998731582756d308efb1cac0e1187a82413a53
72449 F20101111_AAAVHD fensin_m_Page_048.jpg
dee085025ff283b1a723bec59be25a60
adec1b1da8a318371d76a35314f6ce5effee9391
78550 F20101111_AAAVGP fensin_m_Page_029.jpg
2ef217407bd32a65c3bf9278303a3a8d
09b79ead2f70129bf39e787f80b55c435ed13b4d
5385 F20101111_AAAWKG fensin_m_Page_079thm.jpg
8cfc8a9eae6d252b8fc627d855570be9
c0fed1a69ccbca83ce4b23d7cae21473a6116ccc
15991 F20101111_AAAWJS fensin_m_Page_070.QC.jpg
92356c9a64dcba77c918cfd2f78b746a
446d9210441f39b25fcf04ee9c00903af2abf141
67001 F20101111_AAAVHE fensin_m_Page_049.jpg
25721d098128446c1ba754d0cebbcd30
c4793287410891d6860930fa192007fb302d5243
64123 F20101111_AAAVGQ fensin_m_Page_030.jpg
760b1864be2a31aa5d9bafa179704c83
4253e1d0886d151b28b454ef4a2267d9f8516014
19419 F20101111_AAAWKH fensin_m_Page_080.QC.jpg
409161ce731572cb6e65294cfcad4487
792ca5ef00590b1181ab815d3bb894a01c7bfe6f
F20101111_AAAWJT fensin_m_Page_070thm.jpg
ce0d6faad22927b19030fd0b9d27afe1
5430c53b0ed5f81f7fe4a1b8b64d9ab355fb83a2
80795 F20101111_AAAVHF fensin_m_Page_050.jpg
df0c340c6387da97c6a63d7dd89b857c
79bdf263f8b53c2317a474354e2db929272adb4d
82077 F20101111_AAAVGR fensin_m_Page_032.jpg
34f066c7c0b83e9f2e4ec2f838aef9c2
f504f84f6e91403b2c085086b9f8edc49230578f
5462 F20101111_AAAWKI fensin_m_Page_080thm.jpg
5e8584518ea2a8e9a3f56260daccdd93
13910839ebc0f5722cf500353cb4ab2067212ae3
26256 F20101111_AAAWJU fensin_m_Page_071.QC.jpg
0e452aaab195b61bfe24707ace3925be
4818c5a024cfcac3153b4835477420b258522eb9
58374 F20101111_AAAVHG fensin_m_Page_051.jpg
612113d90d76748807d0f17201840765
a6f0b655f7980e7f99a00863df362b05bb4e0bb8
64255 F20101111_AAAVGS fensin_m_Page_033.jpg
4837e803167d9eab836d849f7cb64cb1
94749b649b82bf030d41be1defe77929361a907d
25220 F20101111_AAAWKJ fensin_m_Page_081.QC.jpg
2968796cfd88b3e6a46b440f01fe2a5a
bc15785065635187b5f2e41892538d534152f64e
7129 F20101111_AAAWJV fensin_m_Page_071thm.jpg
6c6defecdccf10b62b038da085a94ced
d2b0e589500332468833eb16bc5afdf53251e0d8
70131 F20101111_AAAVHH fensin_m_Page_052.jpg
79b430998d4ad85115caf29d56c2db94
595373b9347aaa3b2bcb26ab48ef8777378f6b06
73926 F20101111_AAAVGT fensin_m_Page_034.jpg
7f54924933ecde9e810c718f7df0bec4
11cc3baa8f5514efceef5e3df7d5668af7a8ba8c
F20101111_AAAWKK fensin_m_Page_081thm.jpg
8685086b900204c479791745014b66c1
96cf1fe38ffc6e4e893cf67e870a46a535760f2a
26761 F20101111_AAAWJW fensin_m_Page_073.QC.jpg
6422f229043840e34146bcf610354ccc
41e4f0f786bb2b93ab1534e232b22c3907fb43cc
77745 F20101111_AAAVHI fensin_m_Page_054.jpg
67577626f497e238f0cb52d48746887c
f1d77a941912d9d62cad3f2ae6828c802a88f30d
78689 F20101111_AAAVGU fensin_m_Page_035.jpg
cd4eea04c0b8359e442058af0f274bba
800c96c443d2c33270fbf95d4594ad3349deea27
19500 F20101111_AAAWKL fensin_m_Page_083.QC.jpg
e1a57b40bd848183702671cde18f470f
9b7ac367fee4fed8cb57ef42425026efd8baeec0
68130 F20101111_AAAVHJ fensin_m_Page_055.jpg
cb45804bf5e27fa9a2aabde9dd1cc768
597e7f508dfb8f9f2b02ac406367bf6cb7b6ab76
26039 F20101111_AAAWLA fensin_m_Page_093.QC.jpg
1be6e2e2f80302ebfbf840cbcde2a628
cd878965b40c3bc4c65f0b15a49e65e63c524313
14827 F20101111_AAAWKM fensin_m_Page_084.QC.jpg
0e69b7888a1d6f8bab4f1711199e5a38
1cc390e1ce7f56d17b1f81ae6e91d0ead347815c
6735 F20101111_AAAWJX fensin_m_Page_074thm.jpg
4fafe4264cb1977c052d8cee4d1d2b1f
0f8de89e71f12228d89cfa469ee7a77d46971989
43975 F20101111_AAAVHK fensin_m_Page_056.jpg
9d9802e9e6d8cb761b1d4bd6cf781f09
c3d9b8579135b9c5f76bdaa80d62afc59d238eb4
56994 F20101111_AAAVGV fensin_m_Page_036.jpg
de248fc06513d22d50e830940d5c27f4
f8bae41e010a8b154febf7ff7a07b40871d4cb2f
5566 F20101111_AAAWLB fensin_m_Page_094thm.jpg
372f4d15affcce53a417eeb1f943634b
2b0d45a9d8e6509319673496ff513ac01ec8004c
4625 F20101111_AAAWKN fensin_m_Page_084thm.jpg
56cc90e9c0d850a8b09729b1385899d9
459fb712c0e9684ec459b2a66c84228c9c8295cc
21065 F20101111_AAAWJY fensin_m_Page_075.QC.jpg
2fb1f5dba83bd3346b7ed54832cbf5e9
a0be9e5bdf5c9ba2255894df84d9029dbf820694
52537 F20101111_AAAVHL fensin_m_Page_057.jpg
406cfe0463ddeb12983d5e6ad9893e11
0649a26f7a111fa561afd93d80294ff842d36863
82156 F20101111_AAAVGW fensin_m_Page_040.jpg
f929b2f7925f45319597a0499eb30c7f
e6b62e6c42fe35c333208cf64cbef9537064e709
6851 F20101111_AAAWLC fensin_m_Page_095thm.jpg
29ac9b1005273129fdb756f66243fca6
9165c2cde0376023e8d065f13c9590ef8b1f8a77
20907 F20101111_AAAWKO fensin_m_Page_085.QC.jpg
62136395ce13d73ece4a7c100e46168a
4afafda5cc36a4b2d6e3ce5d97f629a64b93a5e9
6006 F20101111_AAAWJZ fensin_m_Page_075thm.jpg
0c344d0a05b7d1b4ad6308c9d61d765d
e07e6ee5afbf5721f36cd981b94896e2ad45de30
73653 F20101111_AAAVIA fensin_m_Page_079.jpg
0ce1e7da6edf5770a3702c05a58aa437
102f9ae9eae63c4ec72a38bb6036b274eb454e85
46424 F20101111_AAAVHM fensin_m_Page_059.jpg
3c1574747db6dd8fd035b6f032ede103
7e4d4cec6c4eac0d9211f1d16a7610bbae25a3e1
73622 F20101111_AAAVGX fensin_m_Page_041.jpg
f2024d3d2303567deb468a8745950d91
31f7a90428eac241405a9972d67e0ebb5ca99318
27167 F20101111_AAAWLD fensin_m_Page_096.QC.jpg
b9ad7f18b95907f557ce51eaff1e7ae8
899b42d6972e06bf3d1112fe0093dd2d07a2a458
6955 F20101111_AAAWKP fensin_m_Page_086thm.jpg
52ef0e4e9bdd8ca1ca53b38a78bc5881
5bb37038d6ea9c444869959f293f117e2e128bd0
69168 F20101111_AAAVHN fensin_m_Page_060.jpg
a6a50b2243c05623a4b288ae050f6372
aafe2db4034c848e3b9b2ba0c2c053e099811c28
64076 F20101111_AAAVGY fensin_m_Page_042.jpg
ba722f759d496c17016c5862bef827a6
091fcfe6bc16007eb2bffb2560101ecb0a068461
76768 F20101111_AAAVIB fensin_m_Page_081.jpg
c8837aa5ee11f5a8e511e84e6e8db139
541c3c7fa91fb7150223c0d31bfa85dab8cd9e0f
7302 F20101111_AAAWLE fensin_m_Page_096thm.jpg
e8210ab5b0a5236541c1a3d3eb742f55
6ab30261cd3520c64de146f279916a734777204d
25835 F20101111_AAAWKQ fensin_m_Page_087.QC.jpg
9c7a46e80d23ee07bb6e4a467746105f
06b6e9e4b8b7f412909f1fe8fc417e4d44a3e584
66197 F20101111_AAAVHO fensin_m_Page_061.jpg
6db3b4dff6d2c44cbef3a757e4210e85
1326eb2d066684558e47785f69fb177599971070
58401 F20101111_AAAVGZ fensin_m_Page_043.jpg
fb6976691816bd0e0272b050bd77d389
f13db5a3bb71b3382845f77c52f75927fde69348
63387 F20101111_AAAVIC fensin_m_Page_082.jpg
cc3bd660fe61b3159d4c81389652345c
160a2a9d7cbd1f462caec216a86c042bd27331a9
22519 F20101111_AAAWLF fensin_m_Page_097.QC.jpg
06d22a665ece9fe3ca5c43cbccdde590
a714c20db56e5e1457172e422c5aac58f5a6eb3f
7042 F20101111_AAAWKR fensin_m_Page_087thm.jpg
c3cc22994fd6032924076d5a1bcdc646
40ba3b55200956e9732233dad1e52e77947a19f6
61941 F20101111_AAAVHP fensin_m_Page_062.jpg
b37b0645a1b2895abeebb245a00a7066
0aa566f78e9c86764b658a5afcb786aaca578213
45548 F20101111_AAAVID fensin_m_Page_084.jpg
90deb6f00d35d0747c1ff86a121f935c
47b7c5beb4187ec2f717ec2adca5f3978fa4d715
6356 F20101111_AAAWLG fensin_m_Page_097thm.jpg
e7704fdda6e20cb29d5172901f0296a3
5d7153e16399a9c8f3bdbe1e712391f37e82a74b
25809 F20101111_AAAWKS fensin_m_Page_088.QC.jpg
a5ee78042122d0a39857b17603ebac80
49092da017f734657294659b8c18e611ff272146
80863 F20101111_AAAVHQ fensin_m_Page_065.jpg
e65bbdb85e5ac342011faa9ba0486396
52befda1e4b2049765277d60c81a2fc6a19cd6cb
63021 F20101111_AAAVIE fensin_m_Page_085.jpg
9de227f2eda762fa12d693842ad8c5d6
a749ec21c1aa5195b60cc389e475db38ac50bfc6
23246 F20101111_AAAWLH fensin_m_Page_098.QC.jpg
72dd0481baf641edc21caf7d3710cd38
ca9a9f7b8627170a31a7ad48ecc73367bcbe08ff
21535 F20101111_AAAWKT fensin_m_Page_089.QC.jpg
10311a444f5d997bbb42063182aa5fa9
41cbd7fc9c998812d790c1182d3bc6a0da021fe7
79098 F20101111_AAAVHR fensin_m_Page_066.jpg
7faa87e6a838e3bab913d94976edd9e6
1f96ee4452606f66644aef50d4cef5953896b227
78329 F20101111_AAAVIF fensin_m_Page_087.jpg
b376d07ed5cef4393033385ee8355963
4c3c100a3a4dcd5c099dce3c571e62046df137cd
24694 F20101111_AAAWLI fensin_m_Page_099.QC.jpg
b50ff1c6125ebf645f62df3e2fa7922e
427df108e53f71de7162084c6673077042d6f220
6109 F20101111_AAAWKU fensin_m_Page_089thm.jpg
44294739264fd21033c1a2a4bba5706d
b2b15816f6fdf28394008431b6e6f85075f33190
75632 F20101111_AAAVHS fensin_m_Page_069.jpg
ad9218d91e2513cef7e7427b349ccc9e
cec54c6fa062ae3b75f698f0729a0b23e27040e8
78195 F20101111_AAAVIG fensin_m_Page_088.jpg
aa357f398449b217ef79729eb6f0c487
5f4cae0f64b34079e620ed40a179149d1124539c
6801 F20101111_AAAWLJ fensin_m_Page_099thm.jpg
d32aefa496e09911245f96055b8659fb
5e4c0f924289a9eee4f69802966332428a5648be
23804 F20101111_AAAWKV fensin_m_Page_090.QC.jpg
314b928e77349014125031f7ae04cfa2
65a62d5e4a115c08a26db8d9613636e66a7c3459
48368 F20101111_AAAVHT fensin_m_Page_070.jpg
cecc59b6d782c443dd24d1e359bfd051
cc99d979cb140c1420d862dd7ee63d35759592fd
72166 F20101111_AAAVIH fensin_m_Page_090.jpg
fa6382d23fd5bc62500f7dc62cba2c4c
2d6b6a2b116f13c5074dd44a50ae0c9103beb1e6
26552 F20101111_AAAWLK fensin_m_Page_100.QC.jpg
412775faeb1f499d7ad08a7e48cac3a4
b6d80104e42a7205d0466396a93cc2dfc42863d8
21951 F20101111_AAAWKW fensin_m_Page_091.QC.jpg
879aedac0b50a8169d1f95d1fc6d65ad
bec0fcf72246f97f54928abf2489401c8a44d455
80545 F20101111_AAAVHU fensin_m_Page_071.jpg
0c9f620a3a5759ca01497572acb9f935
500984c4f7b956ed24e6eb20f199579db1ee42a5
69433 F20101111_AAAVII fensin_m_Page_091.jpg
5c3706af8e859efbe3518f1af0bea239
c3aae91b7c02cdffe9fff9ad301544199500b6cf
6989 F20101111_AAAWLL fensin_m_Page_100thm.jpg
c8cca71f698c47cfab8516fa4cac871c
afeffc01c5ef9f140ecbeae12720459afee680dd
6222 F20101111_AAAWKX fensin_m_Page_091thm.jpg
2f24ed0ae0d69115fc109ffe7c62bf67
63aa59c9eeedea47a05644866606509f7ff9cecb
59015 F20101111_AAAVHV fensin_m_Page_072.jpg
14693641789db8968e4a6c37d33770aa
61a72286c4bbda3bd24d19e89249737dacd9a2e6
49329 F20101111_AAAVIJ fensin_m_Page_092.jpg
1793df469680bd740d77e220cc80092f
d71fb73eaac7c165bf430912d508f279c3f850b2
22617 F20101111_AAAWMA fensin_m_Page_112.QC.jpg
bbb1a91dc21d05916a6d03fb2911b678
06b97c1d389562a22af0c38482162353d2583c98
23103 F20101111_AAAWLM fensin_m_Page_101.QC.jpg
cf9b779c26b974f79197b199f583aadd
707dd177d541dcc3b1733650c10287cfd625e325
60471 F20101111_AAAVIK fensin_m_Page_094.jpg
01878c398808614a459cf8a2a599a1dd
7e818df69defced77a2c470bcf51f83b43ead9ce
26661 F20101111_AAAWMB fensin_m_Page_113.QC.jpg
2f63e38b7c8140a8aad8da8dacf15523
4b87bf4b06eda31bde915049e9d39ba918324dcd
6301 F20101111_AAAWLN fensin_m_Page_101thm.jpg
d834db407069223589fff32c6db42b34
869ebaf88ec42e45e2ac6ba10eac1b8b4db1fba7
15578 F20101111_AAAWKY fensin_m_Page_092.QC.jpg
536fd9b251e39e33bdd5202ffadb5d65
10f12619847dcd6bfb193ccbca18b5e2f1a6d0f9
74218 F20101111_AAAVHW fensin_m_Page_074.jpg
291a30560c46a129007bcbeba3f00286
a5010a31d55b63e9a666d956f0c1b388f67e42d9
80647 F20101111_AAAVIL fensin_m_Page_096.jpg
51dd83337bf1788409da9911b8d1e3c1
2457ae4411068b5646a940b8a3dffc6bc8a97644
7099 F20101111_AAAWMC fensin_m_Page_113thm.jpg
7bf50d7a7cda70c875ca3d5f33868cca
0a352bc3cd8930e52ce64838a58a47cb02f68d98
12404 F20101111_AAAWLO fensin_m_Page_102.QC.jpg
b501cf452bed3e39b315dae6610b4a24
a4cd1aa507c74780092036be8f2170f1bdc4730e
4621 F20101111_AAAWKZ fensin_m_Page_092thm.jpg
cf52c276649bdb94b9a9111b149c82c2
cdc3187ee53146b5db1cc64f80df6c616169d382
65431 F20101111_AAAVHX fensin_m_Page_075.jpg
701fe69a64ed418f81d91b3e5a3d01bc
715b5eb01dad8904d6e86fa270b12449368fa66c
87360 F20101111_AAAVJA fensin_m_Page_121.jpg
29e0180cb058339f18486a67648c046c
c8b328710fd926a03b7977a501c054be9e914402
67889 F20101111_AAAVIM fensin_m_Page_097.jpg
10d1d6c80a036aee5d8d1bf28de6d099
36f4b1e6e5e47625e47e22f6bdfdf1532bc12c9e
24860 F20101111_AAAWMD fensin_m_Page_114.QC.jpg
7889769c5d7af72a6082793183304498
71af844305a4b7d8b240fd7f11221045882baaf0
F20101111_AAAWLP fensin_m_Page_104thm.jpg
6edfb8c76a84951f1d3541baa541d8af
1913d22bd16e4224e59a5b3cd51cd52a610663f1
61857 F20101111_AAAVHY fensin_m_Page_076.jpg
6afd517a92638d0bee83410a73035e41
2757ab1c4f141f698ecd6f5bc74d7a35ffebc9b5
61262 F20101111_AAAVJB fensin_m_Page_122.jpg
2d7552ede518f54392457be4e668da68
a9bcf3752c83d32aa06e89c6429431f31c6709a1
75666 F20101111_AAAVIN fensin_m_Page_098.jpg
7ed860a722d9fb1abe0f642a2848059c
d0203b03491c7e8c9a4eed6b7e414f25069e1412
6761 F20101111_AAAWME fensin_m_Page_114thm.jpg
1537d3564d1c3fbb5bef45e439f40817
065a99d417bc00e40476681e5f3b66c97ac970bd
6802 F20101111_AAAWLQ fensin_m_Page_105thm.jpg
3792f76bc4b3bcb3f7373fdc840a7bd3
9d9d826f352c8e9e581b7f866bc27509e4a6e7f8
95148 F20101111_AAAVHZ fensin_m_Page_077.jpg
86cea1508c6b591cc459a39f517e3591
7b3bd03f0baa88b6840e56bbc0e2e36ad47a514d
64109 F20101111_AAAVJC fensin_m_Page_125.jpg
61e9612db91b0008322a0f0395831128
6628c9464087b3c3fbe3164904923505875fbe7a
74437 F20101111_AAAVIO fensin_m_Page_101.jpg
9b90049aa2756c57cb683715f24dc62a
dd98fe32fed362dcff8cebfdf95ff4a67f5e3e6b
5924 F20101111_AAAWMF fensin_m_Page_115thm.jpg
a6c84a5775d3fdccf32468537337adfa
6a781f61ab4d7525cbad3edece8899d7f9fc7b45
17884 F20101111_AAAWLR fensin_m_Page_106.QC.jpg
1a1332afc45adb68720304bd8aab5d5c
7915489590b5f720341d2332552b52e5d0bf8364
73136 F20101111_AAAVJD fensin_m_Page_126.jpg
e92ba192c6a12882ce2c2ae80cd36862
318cf8be15dfec5f9d6cc1c472fb7605bffc079c
36845 F20101111_AAAVIP fensin_m_Page_102.jpg
eb6e55784abb7e6608d9e93c392ec0c4
493a6672e773eb6e1fbf79fcf3be52706497a518
22606 F20101111_AAAWMG fensin_m_Page_116.QC.jpg
5b54d7bde1560b40cfd610b97bf38ba0
1defd73a2b4b7fbefa96d99f529329ad04ec3928
5208 F20101111_AAAWLS fensin_m_Page_106thm.jpg
b406c45a87e502ffe366f4788324ef3b
fbe5ff1fba7d2ea556a4cc6a3a24185489824ebd
20696 F20101111_AAAVJE fensin_m_Page_127.jpg
b303e3dfc1e02bfeda6be04cd3383fc0
ab0329955a6e0a9a5ffd4b2d32167fae08284431
49746 F20101111_AAAVIQ fensin_m_Page_107.jpg
3334962b1f38f7f5a44ea9dc7456f1cc
f1e679c262317bffb43d153d7c0fe48d68dc1f32
6403 F20101111_AAAWMH fensin_m_Page_116thm.jpg
fb43167481609766f0f4bdb83a56e6b2
0673921b38a6afc24b027bff84c9d8ce528f3b35
4576 F20101111_AAAWLT fensin_m_Page_107thm.jpg
48f80e9f874e7443448554c8d5e18d07
8b7ff2ad5d6ce72e81a4982caa409e5cf14d29d8
78226 F20101111_AAAVJF fensin_m_Page_128.jpg
fe66140f2dece2cd84c29ad18ff01a6b
27b4ddcf5da498751c90b040beee919afeb501d0
72361 F20101111_AAAVIR fensin_m_Page_109.jpg
2570bc42a0863f7c41141b217cd84394
00f9e9d4343b03ea1c4f63a71a93bd6c51ec179e
16660 F20101111_AAAWMI fensin_m_Page_117.QC.jpg
29f423c8316d50b22757f56a25a1e9de
dce6f3f86989c7c9627f1542316a0b394e038ff6
24482 F20101111_AAAWLU fensin_m_Page_108.QC.jpg
f59bf0306b020b86705a2edd4eedd3fe
590c41907fd907e8401d732af2914026fe756b81
80081 F20101111_AAAVJG fensin_m_Page_129.jpg
31f702e71c10ee0e7619286d7cc9fe90
8df3be9c9d6cae28d8d23119c358d3d1cfdd6d60
56161 F20101111_AAAVIS fensin_m_Page_110.jpg
12b9b665a9462c40eb3440fa2100245e
a23f582a44784627cd039601e0901aecf22a404e
4659 F20101111_AAAWMJ fensin_m_Page_117thm.jpg
28406dc8d1045d028ee7488c84111937
750f69ac14f6aef697215641d1506c265e71e943
23567 F20101111_AAAWLV fensin_m_Page_109.QC.jpg
538f3fd1f29c913b76f03d3ffd703ecb
52b7d2e155efb58aae6d49bbcc4bbb7716e60133
60182 F20101111_AAAVJH fensin_m_Page_130.jpg
99f8d58b4c3d07bcd8137dbd014b1952
c56ce84d07f4d5ed8f99f0b3d702adbafc1910cc
71834 F20101111_AAAVIT fensin_m_Page_112.jpg
5d0de9a74f94c2e0112021af70547294
6fab1704625a67caa299c5601009d27a5e75d514
6475 F20101111_AAAWMK fensin_m_Page_118thm.jpg
8cb08c41711a51a006b0b17df83c0910
8c934a67bbd1dc13e3a7c553fe8ba02690401bfc
6458 F20101111_AAAWLW fensin_m_Page_109thm.jpg
7922e63509d1a4715dce24ad1185e47f
eab9e380bc3cdad3118be70a249f91fd80775deb
70912 F20101111_AAAVJI fensin_m_Page_131.jpg
cdf6918792c872be1b6daceafd7fb589
04a54952f5737b669c54409989ba596aa7ce6abe
81015 F20101111_AAAVIU fensin_m_Page_113.jpg
06df8395c6e734fb964580c821d0d750
c7ccf2573d739df65fddfea25bff645b0e96f5cc
6724 F20101111_AAAWML fensin_m_Page_121thm.jpg
2ba863527db20ce2424e3145deee897a
d585356d2e6a9dd7fbc12a932eb7a3132a7edce1
4586 F20101111_AAAWLX fensin_m_Page_110thm.jpg
80d90e31e944df9de00988e779058677
27da40e3c147b0085662d69819af66d84265ede9
46200 F20101111_AAAVJJ fensin_m_Page_132.jpg
fcb1385f8f68547a87f61f9db95caa57
f0838238aee2eb41af4f504042d97cc4a4ec9389
62986 F20101111_AAAVIV fensin_m_Page_115.jpg
795e00a986e95fbbad0e0f75c1b2feaf
2f76160601db0ef52b5ef11df276c5516c484bc2
22071 F20101111_AAAWNA fensin_m_Page_133.QC.jpg
cbea9bc705cb700f62bebda6e0fc4392
10a622053901207457f8ee97de5e6b6b1c28b8ee
20818 F20101111_AAAWMM fensin_m_Page_122.QC.jpg
8509513037d118bc4d0a4e365dbca280
1d1a58bdb76496f7118fb508e6f9fd54b9f0ef8d
9599 F20101111_AAAWLY fensin_m_Page_111.QC.jpg
ae4540c63ac03bcf2a2b2c1dc1ebc2b6
f944bebb97440a56569eb2ebcf9b28fbadcc9c46
74276 F20101111_AAAVJK fensin_m_Page_133.jpg
3dd7fb71ff89f875bbd5c553cb508e64
5bc1813f1b49c6b6fda2a39b27da3f04edc5ba14
67376 F20101111_AAAVIW fensin_m_Page_116.jpg
2f7b9250f5b3b43ee81e312d69fb01a6
6b70feb3d3b798434d09fc6ddcc157e6159781e0
5648 F20101111_AAAWNB fensin_m_Page_133thm.jpg
f84d942b7d84e2ed9ca8a8fe100939e1
1c3abbe668da9c836e1cfb0b2c3b35d84628d409
5857 F20101111_AAAWMN fensin_m_Page_122thm.jpg
66854b58acfdcd1a5d2d06c38bfe09fc
65c2c4bc83b4d783fb2d0f3d874529e512038972
83758 F20101111_AAAVJL fensin_m_Page_134.jpg
d873a868ce490cf22f16d38d6b813f81
4205b50c2ba724d601b56e8ee664d25c619e3ad9
6421 F20101111_AAAWNC fensin_m_Page_134thm.jpg
2f5f85d341ab51ea3034400ffebb44f2
94a2fd4136b4d56f3d36bd8bad180b546ee4e5e4
18407 F20101111_AAAWMO fensin_m_Page_124.QC.jpg
471dca836317f04cfc26bf0923201eb5
1c1319410d5c5d70fa054f45909705cd65be69f1
3236 F20101111_AAAWLZ fensin_m_Page_111thm.jpg
8bfe365432b8715297f98869fafb11f0
9235847b8e2d6edb6f157e1bbc4e3ba625276a48
61987 F20101111_AAAVKA fensin_m_Page_156.jpg
984463bc8e48234d06f0d8435a686c5e
24db1fa55fee7f1a56a7cd441623a4c31b36ba82
86345 F20101111_AAAVJM fensin_m_Page_135.jpg
bd84959a4f639041f6206e94728f6f7c
2e9a13c9a778f807c6796d54d4a51b3777dfa453
54433 F20101111_AAAVIX fensin_m_Page_117.jpg
f1602dec5b891353407e5d9cfb3a084b
fe55da6b7be96c1a1d310e4de56198dd9f2edae6
25545 F20101111_AAAWND fensin_m_Page_135.QC.jpg
42a8be6ab081385bf04f7b2c57e23f1e
8ba63f46c491fc9bdaef90c55cea2a929aff20ee
20103 F20101111_AAAWMP fensin_m_Page_125.QC.jpg
7f022c1a5a0ed88e1c86772e06f41e07
0798376190554352d208c568222e31f621c1fcb3
63245 F20101111_AAAVKB fensin_m_Page_157.jpg
495b1d2a5acbd7b72d2ddaabcf296bd1
8b81b30d304f0701e79ef659a784188e6b0f9c2d
81209 F20101111_AAAVJN fensin_m_Page_139.jpg
8778b64bdbb4a04cfbe71b355316be43
a77a5540905f058ce5626243749f3192ae118ebf
73398 F20101111_AAAVIY fensin_m_Page_119.jpg
89d7f388ad7c77865b53442685130371
e4941631596340960ffca86cd7935c9c1da5018d
6756 F20101111_AAAWNE fensin_m_Page_135thm.jpg
bd5a9ada060a173297d1028900d637ba
1c72b45606dd2c75624a172960c08d748bf6e256
5729 F20101111_AAAWMQ fensin_m_Page_125thm.jpg
6105a1014df270322f912b0a1a45aee2
2182cb50677971eba36b218335d6a9f720e75ea4
63570 F20101111_AAAVKC fensin_m_Page_159.jpg
5c70da273a3e90b5ade7e4eb58fbcd7b
97f7beaf895095a1fa1dbf3e197b2992e8485057
76403 F20101111_AAAVJO fensin_m_Page_141.jpg
aa0c7f329a3e5da1d56f686742a9f053
7cc1c82df840685b9eee1b80a8ca6de05adf1ce6
72537 F20101111_AAAVIZ fensin_m_Page_120.jpg
23dd7c256bae0344bdad8eb821337401
a21abd6462ba241850a6a9674ac27ac0d91b2fbf
25961 F20101111_AAAWNF fensin_m_Page_136.QC.jpg
be4afe8ed6642927a7756f6dffa15609
ec0669862f67de43aa8d58985913afcf08405530
23195 F20101111_AAAWMR fensin_m_Page_126.QC.jpg
aa78da333b70edf1f12e0c41dc9180f2
1a57aadb600a66107f01c299f60d7f0c5278b83a
63283 F20101111_AAAVKD fensin_m_Page_161.jpg
3ebcf72835c2afd5a9509a3738acfeaa
95adf980dcf6020fbc0670d5dc3091444e900516
78266 F20101111_AAAVJP fensin_m_Page_142.jpg
7ae0fc569b17c94dfbac29203221b9e9
00982a317cba8be0ece17af9ab3e633079795ea4
7101 F20101111_AAAWNG fensin_m_Page_136thm.jpg
0d42cf7fef5c742d46d6de94ecb57e7b
00502ed3b58c260f02177a8b9a269af613507c0f
6407 F20101111_AAAWMS fensin_m_Page_126thm.jpg
7100b9ac1348ad3fc8db0a91eee16896
9f149ad6bd99d8142e9acc8d2b13b3d79817adaa
63609 F20101111_AAAVKE fensin_m_Page_162.jpg
fd0b94bb48c486e944bee3c7a887d6de
aeda3dc7dd9eaf43c267f0b9ca0c16e85c556bda
80877 F20101111_AAAVJQ fensin_m_Page_143.jpg
917f101841a2f56e82dbf4a32fe0a8df
2ae0339cd55ad282b265e5e0751422d55053b31a
25269 F20101111_AAAWNH fensin_m_Page_137.QC.jpg
59010cf59d5e59ffa988b9dfbb48cbf5
4390ae8aa81963430cc30d7040f3626872004650
6668 F20101111_AAAWMT fensin_m_Page_127.QC.jpg
6c7a5ca6e57e5b75e6a0c2f7bead5aee
a3e2eb6db2ccf3818ea28f25e0977e2dee81118a
79686 F20101111_AAAVKF fensin_m_Page_164.jpg
85f6103de136c6344a116131b446a309
9011dbcc77cbca7d073a27a9d56fbab83a1a077c
72659 F20101111_AAAVJR fensin_m_Page_145.jpg
48592feda7984fb03f5c5fdd4630d7fd
ce36a6672a691ba4504047f78c6ee8f3bfdf5e30
6306 F20101111_AAAWNI fensin_m_Page_137thm.jpg
099759279148ac9d0a40fbd99c39ebbb
4f1d233c09c80b435257f6e19867e45f30ade53c
25384 F20101111_AAAWMU fensin_m_Page_128.QC.jpg
4a1386bf546b1f8e38ad98bb4963016f
d36169cf985c33a7fe23815ecd25d9cafb3443ba
63320 F20101111_AAAVKG fensin_m_Page_166.jpg
2fa6bdccffadc7d4c56d6e1b885e3853
b3bd96717a7a193b6245055ec3fe05217d3647d5
51019 F20101111_AAAVJS fensin_m_Page_147.jpg
34718852c6ddcb2ee6d0b2e3d5297fc2
5a47aa9066bb50d8f2743df80ee9c050cd88efc0
24597 F20101111_AAAWNJ fensin_m_Page_138.QC.jpg
e56d40c2d8519049061550267b7a49e3
f26d5d896260a9304eb0e3453de9d535348ce364
27056 F20101111_AAAWMV fensin_m_Page_129.QC.jpg
c03ca36839083d8e09a155a2e66b7c3a
4c6afc8324d1b57df08f972700568f1c7763a76a
85345 F20101111_AAAVKH fensin_m_Page_168.jpg
d04d81845ba786d9f88eb022f95ebb2c
4a561537a9234d8c2b6c47cd952318a9c9d10224
76413 F20101111_AAAVJT fensin_m_Page_148.jpg
59e8a43bcbed82a3906f693ab3e57c75
3332147b1c45064958383219bae3a4b2e481e5bc
23927 F20101111_AAAWNK fensin_m_Page_139.QC.jpg
1799b9a0eed2f8bdcd1f11bd24499468
3b4ed664a423c212a4041aee25bf5db18699f6f8
18916 F20101111_AAAWMW fensin_m_Page_130.QC.jpg
e8809356576b392845b83a79f454a40f
9f559b6664ce622f644ab09fba4a714b31eed7fe
85261 F20101111_AAAVKI fensin_m_Page_169.jpg
e89d07ec8bc0f67273ca04987926962e
9f56be42111392c4b76c55cd46c2ed503d30eb2b
82766 F20101111_AAAVJU fensin_m_Page_149.jpg
7e1d81167454b2d6e50cd3a805849c09
3f669a1c01a3a28ae64287806f47bef54c5f0b8a
F20101111_AAAWNL fensin_m_Page_139thm.jpg
cb0402e9c34c3aa8a4d37c026c61e33c
269567610ac2fe3a04338a28105d603a8b1cf8c4
5813 F20101111_AAAWMX fensin_m_Page_130thm.jpg
77bb9d6d882bf6a5a519fbe7ecf1da74
eabaeb91ce3278b5bed080f9d2dc04732af77313
81821 F20101111_AAAVKJ fensin_m_Page_170.jpg
90e749409a126db3281e9d25f664a659
8a270be9b88216c047f520897bbf42aa4d0021c9
81348 F20101111_AAAVJV fensin_m_Page_150.jpg
b8f3364b416b543b767c8a063f6da9bc
4bf1bf0cd6182821b8b9a68ba3e2362b386ae9e7
22536 F20101111_AAAWOA fensin_m_Page_152.QC.jpg
78738d485ab303710d3c7ddff42ed091
ec5feac741b56f4a06ae10d319c425af83a63c74
23947 F20101111_AAAWNM fensin_m_Page_140.QC.jpg
700952c9bbca1a31c7e8f7e65dfa46bf
e3c66a4f4800077c8a5b432e5f147510c9bdc765
23018 F20101111_AAAWMY fensin_m_Page_131.QC.jpg
5f3a98dbbd706157673061a9a51b7890
7e27c5b93512d3a56a099dde7f880b01c2f74678
74673 F20101111_AAAVKK fensin_m_Page_172.jpg
3ec3920709b6323ef34eebaca0d93d09
569d6a1041680b27216ffbbe6b622c12034812ca
77129 F20101111_AAAVJW fensin_m_Page_151.jpg
4f5a67abc85590c4edc9aa7a9752fe26
874c2911af3b063df700713ee0f98663f020e16e
6362 F20101111_AAAWOB fensin_m_Page_154thm.jpg
3dd305201694d9e6c191f420ee14f2a0
b92f2e74f21eb6aa551a5b6309899e271ce4634a
6600 F20101111_AAAWNN fensin_m_Page_140thm.jpg
58e7096b4923a96495bd092c4808623b
037708cfde94d7499257eee20c61639fd9851c0a
6454 F20101111_AAAWMZ fensin_m_Page_131thm.jpg
6886a81f0c297b0135cc090ad05eb19b
b65086cf46deb449e81f19149c4f2f37d5cb0d66
76374 F20101111_AAAVKL fensin_m_Page_173.jpg
72857df84e59d2815f7fedc19231ec0c
f2186a81dc790dc2ad5d05eec1cc785b5019e98a
70225 F20101111_AAAVJX fensin_m_Page_152.jpg
ee70da46a102ad4dbecf1de7387ae652
c958355e9548932941cb105994fbba5bc88310c8
15389 F20101111_AAAWOC fensin_m_Page_155.QC.jpg
f008dd1f052b3e4005ff0829662f9a99
62417054d64d7091242444d5f64759f4782f8534
6903 F20101111_AAAWNO fensin_m_Page_141thm.jpg
6a4c11a48df1bedb7169b5f73a774257
c929265fef0fda39b8f622a04eb0cea4457f6185
78255 F20101111_AAAVKM fensin_m_Page_174.jpg
9d8c97ac2125fdd0b700fcc67c85969e
96f3030bf09a923d4af6b986777b110bf0768101
F20101111_AAAVLA fensin_m_Page_009.jp2
4199a8e2fe45942d403fd5651e20e9a6
fd0fdecdcf4b0d74204fa73fcc3d013211b19630
4736 F20101111_AAAWOD fensin_m_Page_155thm.jpg
2dbba66d8c15bcd0691ff0e08275c531
3c3f16e68f6ef2b84bfa266c5aa77fada4316a49
7159 F20101111_AAAWNP fensin_m_Page_142thm.jpg
8c423998db8048ee42b3bb6b640a890b
5a5ffd83e72f1bd62d24d429b62c69b77f53a148
29307 F20101111_AAAVKN fensin_m_Page_176.jpg
263ed236ea922a9d87c9ee462b6df05f
7c64201f6011c8c55fe8592f2c69f1393a364c55
69537 F20101111_AAAVJY fensin_m_Page_154.jpg
2ba205a8ba5e0d6d1556a4dc398440cd
ab02a1bc9ca24e88d5f581bcd0eada1ffebb8da4
1051978 F20101111_AAAVLB fensin_m_Page_011.jp2
a1ab973479312369d45205a5f9f5ea1f
666f0e580cf0a907a8365adf52df798d5ad43da2
19364 F20101111_AAAWOE fensin_m_Page_156.QC.jpg
10312d1fc04bbc7bd0a92d427a229324
b5748d4461178b88796706db7322ecdd0318feb4
26716 F20101111_AAAWNQ fensin_m_Page_143.QC.jpg
6288956c28c47fa71c33119ed23d3998
9488f3f8e4285ca111e84ff7a588fe291a69a16a
53638 F20101111_AAAVKO fensin_m_Page_177.jpg
914ae016b09169f19493c2671a9ec4a6
e6e7766e0e8a105f8a0e5307f7f6bb2ee34c318e
44848 F20101111_AAAVJZ fensin_m_Page_155.jpg
eb68387136ca2ae31d722bdbdaa96054
8291d9b7d72db25187eea1de6685172f4aecd03e
107776 F20101111_AAAVLC fensin_m_Page_012.jp2
a18b197b33dcccffa7839d35f6bf26d5
88a6807c176b92ebc5333306b07e01af2062078e
5322 F20101111_AAAWOF fensin_m_Page_156thm.jpg
77cbfdccbd6a73f3a8a01ec9cfef63f0
64516b36ef9c070cdd781aabc73850533fd0bdc2
6234 F20101111_AAAWNR fensin_m_Page_144thm.jpg
fd81c0898392bca4f790b6bd78184c5b
5ea9c67581981051e0194864333e1e8730f0adc4
72179 F20101111_AAAVKP fensin_m_Page_178.jpg
59b1449c867c73d615065a8580077f29
8ca45356edb048fab85b4d58b9a8f0da748227bc
55454 F20101111_AAAVLD fensin_m_Page_013.jp2
a0c5e88c7adcdc57f103956020f54eb8
763caf2434da88006a9fe393671049bdc364af25
17862 F20101111_AAAWOG fensin_m_Page_157.QC.jpg
f98af3062d48926a5cec1dc9da962260
fe6c5036f6190d4eec7ab389c5c33ba673b65aba
28817 F20101111_AAAWNS fensin_m_Page_146.QC.jpg
d31e9168a892131eea15f700b0a1aa63
472864edbe78547e20183248f2b38ccd61309605
88959 F20101111_AAAVKQ fensin_m_Page_179.jpg
1dede5f17bf09e1a6e21d4d1d5ef2908
f03f3f58507b27fa112e0249d4465cd00047d07b
111748 F20101111_AAAVLE fensin_m_Page_014.jp2
793ecea8bb7453da9c19d7f06327275c
c822d19fd08631107559f30e9c88f13fd1e099a8
4465 F20101111_AAAWOH fensin_m_Page_158thm.jpg
372b665658e353b94256113d23aef7d1
63d5e52669b55636a821c07f57abb0e82a0ffbd6
7580 F20101111_AAAWNT fensin_m_Page_146thm.jpg
f41fa3da347396542d57cfc7e967347d
967e48e5483059ded75c21ab1489e495a36205c9
90529 F20101111_AAAVKR fensin_m_Page_180.jpg
be473e47d2b208b6083d04f1258bc71e
9e1dad607a3b1c8be4c17f855c4b51b42fe56a8d
114024 F20101111_AAAVLF fensin_m_Page_016.jp2
216ceeebb800c60d0c743f4254c2c6d8
89d82a67e872f71a0ed7c277368d13d85c98dc5a
17826 F20101111_AAAWOI fensin_m_Page_159.QC.jpg
27c4363c70f4663d33d5b33115039db9
0f896037411b42d7ae1a47fc7ea4dff49177a0fc
15812 F20101111_AAAWNU fensin_m_Page_147.QC.jpg
8e7a7443f2592776ca96a2536fc8eef6
15907d8b897f70417079b205d95c915adf40841a
92331 F20101111_AAAVKS fensin_m_Page_181.jpg
1a488dffee47079db16ce8a24245dec0
7df8fcadfaa3655a1bcd0f18dad7a9d77a88db4f
113563 F20101111_AAAVLG fensin_m_Page_017.jp2
faab4399ac718f173dc16d9fa6033c0f
b4648a70e6067f37ac5e7c2f4b50613bb3e2afa1
4446 F20101111_AAAWOJ fensin_m_Page_159thm.jpg
bf75f614dcbec2b694226f3fe1ab8faa
696b7277b9504166dc1df65c811db7ad7343fee3
4500 F20101111_AAAWNV fensin_m_Page_147thm.jpg
55630ff75720f01bc0f9aee4a6b22ef0
d2b9de67828bc09f35380b60e35f1a8b5441dcff
91054 F20101111_AAAVKT fensin_m_Page_183.jpg
e9521d30c26ec3b45648bce6664344ff
eeadbb76c446cc6c55f2525c839ac12eceda5429
117111 F20101111_AAAVLH fensin_m_Page_018.jp2
296d0d247cd8676260799f47552a2875
ade5fa6bb482176d4a2352e67b3543d293218c4b
17932 F20101111_AAAWOK fensin_m_Page_160.QC.jpg
32e298375b7bbf1b774388725eaa9e43
baa170e74e26227cfcc5666589e4ef6b7ad2bd06
6961 F20101111_AAAWNW fensin_m_Page_149thm.jpg
5d26d349593903dcdd990a9353bb6eba
6ed975026b6a62bcfcecc9829815554c1ff71c34
96883 F20101111_AAAVKU fensin_m_Page_184.jpg
2da399c5d205e933d3685f810dc8d5c0
92b0b0d9f80278270c63046eddfcbc68e54634a2
122771 F20101111_AAAVLI fensin_m_Page_019.jp2
75145e48c8d39df7db542876be146b43
41a5fe1278e6111b327c296345da39e640d954fb
4495 F20101111_AAAWOL fensin_m_Page_160thm.jpg
076f1aa8cafc0d1316b6847f976eb085
8d7a3c63eebaf8f5b6c181a3deccad7ea8952ebf
25893 F20101111_AAAWNX fensin_m_Page_150.QC.jpg
93b369c4fc85e9383d0a0927917d8a35
ece67264f8035ed92a93e2187bb1608871dbf192
95324 F20101111_AAAVKV fensin_m_Page_186.jpg
06685c6db610c451947e9ede326072a8
be2879d78a62ded15ed2ab2248c7c4d7335ca196
115342 F20101111_AAAVLJ fensin_m_Page_020.jp2
04a077d1e6b18e46fd99cc81b996d5ea
f22e62277fd246988631e6072e5a9be1251ae607
6809 F20101111_AAAWPA fensin_m_Page_172thm.jpg
829020cee282daae0c8c4f0b881ad834
90cb6f8d48cc340ba2cb87a34157f4949ea46a58
17839 F20101111_AAAWOM fensin_m_Page_161.QC.jpg
a43dd541934c19604caed8bebde131ce
0773b06c47599f1bab87078217fe4d4f09081c3d
F20101111_AAAWNY fensin_m_Page_150thm.jpg
a3317caa2afe4c90e398d8666da04065
87f7a1dde039465d6727083c1ff85b442a011f72
5672 F20101111_AAAVKW fensin_m_Page_002.jp2
d14e748e9f87ff8bab807637fbb1200b
a6a21012915cc675dacb87a3b893e6c5f445a2c5
117536 F20101111_AAAVLK fensin_m_Page_021.jp2
cc98820a878f44b700fe2dbd26f252cf
743ab823a2a4381b03192392d58ee1fc0c1ffbd8
24956 F20101111_AAAWPB fensin_m_Page_173.QC.jpg
15b4e1849d643ffde484daa76575485c
f678490e2c67bbddc9e9c5b45b7fcba22e8983e1
4424 F20101111_AAAWON fensin_m_Page_161thm.jpg
af33fab064653d2eaa5711606ea6fae2
a6150a337cc83941d17f09e9f37ea959d68ae4c8
7062 F20101111_AAAWNZ fensin_m_Page_151thm.jpg
76cf021a942f006f0554fe20d561dab0
7785373e4cac3afddded385a6419f6621c466db9
6054 F20101111_AAAVKX fensin_m_Page_003.jp2
44bb637fc3f458f33e1657c8ad8c7b1d
b7a8fdfe0863d1427fc593eec69a6c6c054ecd63
119949 F20101111_AAAVLL fensin_m_Page_022.jp2
ae0a67a54a3232ec3eb19168d188f00b
a64dd6aa8556d6505f9c4d5566337623ab907606
25480 F20101111_AAAWPC fensin_m_Page_174.QC.jpg
e0aeed0a0d7a4e4dead3fe7b40a7dad8
45ee709b2de63783a5acb737c81a56769f97e04d
4463 F20101111_AAAWOO fensin_m_Page_162thm.jpg
b6bdb8d7da72ed0f1641749230b805d5
07908b9a9e48d921de8d5beead1d2e864dd190d7
122608 F20101111_AAAVKY fensin_m_Page_004.jp2
09e765f72b9725e770f67a221561583f
341e69314262fd3c27e25935b94a0e853ebef963
906440 F20101111_AAAVMA fensin_m_Page_046.jp2
1fb70fb4855ba284591c1181a33fa548
7127baa016beb40403b6b07730e3aec13893af1b
119619 F20101111_AAAVLM fensin_m_Page_024.jp2
403a6177e6e0c0f5c4e9a7985192b99e
07a040031623d598900e1e87fcf072c2d27d251b
7167 F20101111_AAAWPD fensin_m_Page_175thm.jpg
e76d174e9071c269811bcd71796737f4
0cd978f714c1ba9800445a90df0d2eac970a6256
17899 F20101111_AAAWOP fensin_m_Page_163.QC.jpg
89f8463500933594e7558b1bec4ebb6c
f7f544fdd336382292b864b92fbf3facc449c963
110779 F20101111_AAAVMB fensin_m_Page_048.jp2
e391abb8938403d5fec37fb59e452753
9a37a82f28eb1b80d013c2eedadf74a74a157369
119020 F20101111_AAAVLN fensin_m_Page_027.jp2
339baa84cf7b5e7deae395f5a2b8f482
8db1bd51c2cc664e823295cec62029d8a10a820d
9883 F20101111_AAAWPE fensin_m_Page_176.QC.jpg
f28dfe4e06d00309c26fd20a6716fdf9
86ba705a3956b944cc0b6d857efcdcec88216922
4470 F20101111_AAAWOQ fensin_m_Page_163thm.jpg
d2c271f591aae13827d0eaf4552dc394
9c2e199fe173cd434098880760fc9d18eb41d9be
84692 F20101111_AAAVMC fensin_m_Page_051.jp2
29beeb608c0064490089931c95df5193
890257191a88987aff46e3b54c4c27f393c47eb4
115754 F20101111_AAAVLO fensin_m_Page_028.jp2
ad6edb1cd7b76f505167e3741e4550a5
3b79886b83bf6b3d453f91375d57719ca1a15b73
1051986 F20101111_AAAVKZ fensin_m_Page_006.jp2
ae575e17b3cd256954cd79d7eca61ebe
92fa88cd53f444056f5d8caf0195a79629d9e0eb
3137 F20101111_AAAWPF fensin_m_Page_176thm.jpg
e3dc5234f4f54d7f977a423fbff1dcf1
cf50c0d9a007171bb531a160c432a3fd673c06d3
23046 F20101111_AAAWOR fensin_m_Page_164.QC.jpg
ea4511c5be0548f10b6f9476a5d83630
a40cca0a72d4556a48bcdb10ea7f0a8de42c760d
118270 F20101111_AAAVMD fensin_m_Page_054.jp2
1a7c897f6d5128e1c2d021c54092fbb0
35d45d15c93f6ae48db4358e5cdc15aafbb15244
118904 F20101111_AAAVLP fensin_m_Page_029.jp2
b813b2d7da6b79cba8ef8e2cc60c1db0
a813b1bccb8cbe32dc6722bedc6575376d9209ce
17124 F20101111_AAAWPG fensin_m_Page_177.QC.jpg
12b5f1bd11ad7f3ea16adef7a4ddb33f
948224084fbff458829ae91289e35a258e6de9b0
F20101111_AAAWOS fensin_m_Page_164thm.jpg
fd845d145b7e0931ae36b46bc1ac9554
ef4cf8f739927e113ac17207969dad3f8a35f8eb
103348 F20101111_AAAVME fensin_m_Page_055.jp2
f5c748be9c035ad81f3270c57bc2cc19
80a55d197a9ef17fcf8ed53a0c7973cdd1e147e3
95531 F20101111_AAAVLQ fensin_m_Page_030.jp2
3c46ac72c4d36573df259c0620c6b5b0
8828a43a8705868d3c21ef78eb7e3f824ae80278
4856 F20101111_AAAWPH fensin_m_Page_177thm.jpg
c03cccc2d86e585e03b5280240e93ebf
1187891c825f588b8cc99ef69f3f638670fbb2d6
25647 F20101111_AAAWOT fensin_m_Page_165.QC.jpg
240111783c88d4aa8d9f5fe3dd97abe2
46e67d59a4178a02bf1cfb453b8426e8637b44cc
76619 F20101111_AAAVMF fensin_m_Page_057.jp2
0110a934223296570b63f81b261816ba
42094557abcd26da6055ea99de71bcf424b849a0
85295 F20101111_AAAVLR fensin_m_Page_031.jp2
712916b060eb4443532864b8b96df358
2cf9677efc3ba43898aa33e978bcd200553c9f44
6662 F20101111_AAAWPI fensin_m_Page_178thm.jpg
e00020141c1280271d1ff31d352d0883
0922fc35e16a844296cea7ae91499d39c26e946b
7067 F20101111_AAAWOU fensin_m_Page_165thm.jpg
e057b70fb117181f4a180c663c73bd94
a6a1ffd903c3c443131a94cb409813290ffb9cfc
83815 F20101111_AAAVMG fensin_m_Page_058.jp2
46c7e30e8a639d3f4988141ac1ecc3bc
d84e7a6a0986e05420cb9f28fc98874823c383bd
83472 F20101111_AAAVLS fensin_m_Page_036.jp2
651d1e858c4a01e7bc9009e2ee678d1a
e5c1548dcd003897c1905f28ba45583289beb83e
26382 F20101111_AAAWPJ fensin_m_Page_179.QC.jpg
6dc2a72aab555fe6bbadc64c6034713c
1987272ec380628f7061171d60d7b1b3b0274965
20835 F20101111_AAAWOV fensin_m_Page_166.QC.jpg
024aa501ecae8fcc7446ff972423e5af
a7c03fadf4780310d2c409dca8e6305c652d0bb7
67522 F20101111_AAAVMH fensin_m_Page_059.jp2
ae0450e3053487257950656115355091
547faa95b28806ec45aa59856d06ea51eab03502
88378 F20101111_AAAVLT fensin_m_Page_037.jp2
2fbe18620cd3f8ae26a010568de2f3bf
d9fd6fc9ec46909c99a57bbe21abbdc401a87b4a
7163 F20101111_AAAWPK fensin_m_Page_179thm.jpg
43a0a1341086bf31331661c4cb773b42
64a6f69c7b1f21ffc46415f4cd7031f9768773a8
5856 F20101111_AAAWOW fensin_m_Page_166thm.jpg
b83c136d2dedc88b5504392148ea035f
255f2eafd908c1e4b587846f48963efb8493f71b
101960 F20101111_AAAVMI fensin_m_Page_060.jp2
806cdf499d26ed5ec1cd7edcf9772664
3b1700c795d381eb2e8f22773eef5a94191d129a
98114 F20101111_AAAVLU fensin_m_Page_038.jp2
260a5254f44bde3bc38784ca38e3ad1d
6b4b8fa3f7faff569906195772002e8b745b6c5e
27600 F20101111_AAAWPL fensin_m_Page_180.QC.jpg
10d975550d54e042d4f14d9a1563ee05
a19ea1ac5fc1a89ed1357762f1471cfccb8391ff
24078 F20101111_AAAWOX fensin_m_Page_167.QC.jpg
1443cd1184d1f84a2ae1f06e604e53ca
ef893ac01d1ab4c897650578c5e1ad125afd0d0a
99118 F20101111_AAAVMJ fensin_m_Page_061.jp2
c67c95ae5ec7e63a2e91c72f79a97b62
6617d2d7cdfd31f192f8616596cebaa87f96b789
98841 F20101111_AAAVLV fensin_m_Page_039.jp2
b95349af4d906c6c5a7f9e64bcaa1cfb
f3b618f0648249a54c1dd9153047d7e36d48ecf1
5755 F20101111_AAAWQA fensin_m_Page_188thm.jpg
70e924ce167ced520973f0a455f1932a
0ee832a1a4fe53d43881bfc9755346c4ba63a4d9
27428 F20101111_AAAWPM fensin_m_Page_181.QC.jpg
bc1928e43e469bb63bd8e6937c08b8a7
4f453908e0eb63743fd274c599bc987275841c4c
6576 F20101111_AAAWOY fensin_m_Page_167thm.jpg
85f5a0b9420a8595286b5bfccd9d59bd
906bea7235f10642541722bc7edf4cfc2b9ae4f6
112156 F20101111_AAAVMK fensin_m_Page_063.jp2
3d57e0bc2f8e7e4760eee9530bc2e436
f4400322231d602e0972916d480853d0716eb540
110871 F20101111_AAAVLW fensin_m_Page_041.jp2
5dd493e86a8bb2ab7fdb27bbfd6e96c9
c4146b558bca4fb4a8f2071f8090de38229243fe
216442 F20101111_AAAWQB UFE0021946_00001.mets
5cf050bf5bffbc34ba3d65c33c55c920
83ae4b6e79cd141487bbab97a1affe500d9a259d
7248 F20101111_AAAWPN fensin_m_Page_181thm.jpg
b5e874b6b711ccf31963e1bb36b03cdb
84770513e4619fb0540ee3d8b5cb48f9f46f488b
26274 F20101111_AAAWOZ fensin_m_Page_169.QC.jpg
dc22d4712135cea2904e3b2422442f10
677a689462d09dce9618d462fe558c289b3eabbf
122516 F20101111_AAAVML fensin_m_Page_065.jp2
41131e16e9b27bef61554c54dd40f1fd
5d6af45371af43336b6c5329a7f6f716d8bd2ade
96556 F20101111_AAAVLX fensin_m_Page_042.jp2
41f2820a70aaa9db2bbd9d3ab70de7e6
2541066f315c3c7323abe177a0290b07b8958d26
29701 F20101111_AAAWPO fensin_m_Page_182.QC.jpg
594b806a1ef55ab2254e0be379cf366b
3f402e1f9e9a7c2172f631885c103b91b3e643c9
116583 F20101111_AAAVMM fensin_m_Page_068.jp2
9a58ce959010b4ef81adea348aca224d
bf4ec5c4863767e429156d39014712e548fcf9b2
86735 F20101111_AAAVLY fensin_m_Page_043.jp2
de2d780a69b8a64e0b9125a6c35b1059
3b6e689c8d1928337c984b6ad3d972c7a7df056b
94414 F20101111_AAAVNA fensin_m_Page_085.jp2
fdc646b3f243f167624b1cfb83d550c6
c08482a122ef5092bc928e3da0f7530495afdfc7
8020 F20101111_AAAWPP fensin_m_Page_182thm.jpg
d04e00ce1073518df52d525452f61393
cf2c802d2450acbebb3d2d7150ab25bb0dd3d1b2
115393 F20101111_AAAVMN fensin_m_Page_069.jp2
dc087963399db4c434d2a3d111aabd95
75b18c227fe3378e9eb1d92f45acd5e7b40193f3
118784 F20101111_AAAVLZ fensin_m_Page_045.jp2
6c0db3d87e1988b51dbaebb943bf5f77
7c97f7636724dcb8d4b02c70b5b27a48e0b567ed
111873 F20101111_AAAVNB fensin_m_Page_086.jp2
0ca904f90669180cb036318d6faa28c0
290c96f2f1ab677488bfedb0582709f899347032
26247 F20101111_AAAWPQ fensin_m_Page_183.QC.jpg
d6138eb58abd478ee438e0318de1fbdc
bfe75b8d7bfc91e602a2ea248247113b271fc615
69959 F20101111_AAAVMO fensin_m_Page_070.jp2
01af34810053b1c4f58c7803354fef12
b7898c5dbd7adf426baf112690ccfe3e52d4423a
118951 F20101111_AAAVNC fensin_m_Page_088.jp2
32ba9ed21eaa6686a11a535d7177043c
f64f833c657d5b6d512cdf67a50fce3a36e9fd78
7366 F20101111_AAAWPR fensin_m_Page_183thm.jpg
09e9faa6a7b2cd6986c08b64f10475ba
4c5e1de0cbdfc42f8c95b6717926708f776359ac
120254 F20101111_AAAVMP fensin_m_Page_071.jp2
ac03eab96ba53b7090432b229d81a614
1c73ae5c4933e5db1a2df3af6105307b56657dc4
956282 F20101111_AAAVND fensin_m_Page_089.jp2
09aff13f10f9fd8de33ded0a6c9e1407
a06ea8cee6ac7697012814ea39cd424be62393a9
27994 F20101111_AAAWPS fensin_m_Page_184.QC.jpg
0bc9b77eeb303dcc9d683308bebf92f4
35d2b728ef5e1af73549f40947dd526a4295c480
764187 F20101111_AAAVMQ fensin_m_Page_072.jp2
dfe31e64ffe816ab4a2d62c93c16b949
23f83276f6a7434bd1841c6db051a2e8c7be10c9
107631 F20101111_AAAVNE fensin_m_Page_090.jp2
2c8f51723cff063238b0dc018a86055c
a873a07be97e4744fe44edcaec1e49825d3c5d5d
7518 F20101111_AAAWPT fensin_m_Page_184thm.jpg
3337db5cecbfdf15862bfde2a41f6e8a
9c84d3df6895e3c0d1fa2bdeb50fda38a23ced48
123513 F20101111_AAAVMR fensin_m_Page_073.jp2
915e3835163fa03b891d75b876b26e96
990ab6efe798d87badb53ef15c7fd53e85891620
960924 F20101111_AAAVNF fensin_m_Page_091.jp2
d50f5fe54096b54e27fcf55cc74c6f9e
f65f34bc8c720db560c7f5f180041d5b176a0c58
32012 F20101111_AAAWPU fensin_m_Page_185.QC.jpg
a0438249b1f3a2e64c2b1c346a3a83c0
c7a020c5849ff414804ce96296bd7727f50d35d1
908804 F20101111_AAAVMS fensin_m_Page_075.jp2
494cb380acff5f967cc7705a257462fc
1474dc26b519c4c6efa9be319be90dbe1445fa06
652626 F20101111_AAAVNG fensin_m_Page_092.jp2
07f3c7465226d7bcfc1a13d6e5dc6415
17c4b303dadbc3aa4b971eed6dc05bae03f44764
27434 F20101111_AAAWPV fensin_m_Page_186.QC.jpg
8164e7720949aeb843598d71e4bc8834
d16f340e8e259eefcc5caeb1387a8f993848fb17
861752 F20101111_AAAVMT fensin_m_Page_076.jp2
ee319058635cda68b55c31beff4095bf
23bf46772ac6c5891049ef210c2cb3a5216a9e73
841528 F20101111_AAAVNH fensin_m_Page_094.jp2
f0f594d2f5f3766c9ae27e3a7016e25b
547e29b38dbcd8722a4f9bfef72b66ea50b20bb1
7388 F20101111_AAAWPW fensin_m_Page_186thm.jpg
5e8176356409b690a633f91badcb3525
0a22be0ee34c8c816b212dda08ab96bd03e87c74
1051975 F20101111_AAAVMU fensin_m_Page_077.jp2
826e1f17bfa415c5acf68b9803248b16
2b3959662edd87cee4ea044c3763877deb07893b
116584 F20101111_AAAVNI fensin_m_Page_095.jp2
5a78c30e046baad013e43b04778f2fda
f14fbdf8c1a95924111a09b4cf2961f9a3003436
18619 F20101111_AAAWPX fensin_m_Page_187.QC.jpg
9064ed6dd817a32a86e0e7bf2bdc180e
eaff773e0e7f654b9b7a61cb0124a72723c97ee0
1026988 F20101111_AAAVMV fensin_m_Page_079.jp2
bd11846608467aed72587c6f0b6ce6cf
012bcd47164ac09dd7927f0404766caf947cd436
122935 F20101111_AAAVNJ fensin_m_Page_096.jp2
cd41d0aa318c8c9151fd333bfc659914
df25d4d5d225b7f83e4069caeb98de2cf3e2b7b9
5005 F20101111_AAAWPY fensin_m_Page_187thm.jpg
b3ad5877abe3b024de1b103841d33fa7
8e3484a65eb8a2637adc54c8445e103e833e367f
824258 F20101111_AAAVMW fensin_m_Page_080.jp2
73a9beac8d7ebca07cfc5c2cfeed0bd1
6ec374e43aa7c629eedcc14b06f7e264009ebe38
101579 F20101111_AAAVNK fensin_m_Page_097.jp2
b93820be9918c9a83aa982a0d1908287
903923e0fc028222af6515a78293b33ba36b492f
21272 F20101111_AAAWPZ fensin_m_Page_188.QC.jpg
83c0fbb2846a01947eebcf2eb8a8b448
72cdcd2c92942f8756f5d08eed037dbb181e2467
115896 F20101111_AAAVMX fensin_m_Page_081.jp2
bbce90a9c10b5a43ebffcffb16241274
92639d25ba624426efac15bea42668aaec5112be
119616 F20101111_AAAVNL fensin_m_Page_098.jp2
cd8a3215c0d9831162b86c6d22ddd06e
891df96950fecff3db3bd399c295c9e2e6214255
94519 F20101111_AAAVMY fensin_m_Page_083.jp2
5c81e601b22e81b1bd325023fcb7777b
fd2dbe644c0bbf422be072efea0108ab3095cd27
100959 F20101111_AAAVOA fensin_m_Page_116.jp2
a25d2b82bf386d261a9aacb91e13c6d1
22e9716a0d0531cf9e31a7c5fd49bccaa3d5333a
112114 F20101111_AAAVNM fensin_m_Page_099.jp2
c1c25b59a0df6f00e521804ad1068ec8
1f22a7bfe0fc24a8c1bcae0538c9b08eda02e03f
63274 F20101111_AAAVMZ fensin_m_Page_084.jp2
9a89e5f45d3884774d8b35b5cd663333
44134efa0c8ca50085cd8a0405233301518ebbeb
75404 F20101111_AAAVOB fensin_m_Page_117.jp2
49636b9061d3256028309f7851b71bab
5fe36d7e9f68073cedbe6db44fcbeb902a66192e
122195 F20101111_AAAVNN fensin_m_Page_100.jp2
eabac50443706b961d001b263e2420a1
da676cbf03652a64d31813872d50d904f2c95c61
109142 F20101111_AAAVOC fensin_m_Page_118.jp2
2e9860d01dd3dd4128e677a751fe8029
bcdd974769709f9a6f0a7f5d4a55c3dbb99aee72
113184 F20101111_AAAVNO fensin_m_Page_101.jp2
03a1e9824bf21088d73a0ce3b1d98a9f
a5895a2725e57a77e9fd9e9761c544deeb93b0ba
105541 F20101111_AAAVOD fensin_m_Page_119.jp2
f8ca0d4eec763d248cf716278634d39e
a88e4c06d5358fa82f75d7af0d3d2a912e33c980
50814 F20101111_AAAVNP fensin_m_Page_102.jp2
5458fe856eba9e9a13ff512871014414
b5c80d6da9ac6cdb2fa9691eeb1176779aa45cd4
119401 F20101111_AAAVOE fensin_m_Page_123.jp2
55b8aaa012a6e3e13d8722508694e2c9
e820a2a94267923ac9eb0ee2bdc69aaf49b787fa
122503 F20101111_AAAVNQ fensin_m_Page_103.jp2
0d2ad0c09ad2e93d73535928e46ca432
59f542729ae7863baff439c856a8414488864ca2
91140 F20101111_AAAVOF fensin_m_Page_124.jp2
6df6f68a60e318f35682d7a5cdfac0b5
bd62565b3cd371f5286ed815feabf49075ba0af5
118206 F20101111_AAAVNR fensin_m_Page_105.jp2
5cdef403dd81b214687f5d9780e7376c
fab324622dfb35b626a856dc23b3928e991fc06e
24294 F20101111_AAAVOG fensin_m_Page_127.jp2
d1cfa994ec31a3c5fe958bc182f289a7
51c531dcc92b5b84237394850ac8bf627867913b
79933 F20101111_AAAVNS fensin_m_Page_106.jp2
1a1153bc4a5fea01c81aa7d3167289db
9eddfe085152d16e2ffbaf931a05adceba2eb01a
116971 F20101111_AAAVOH fensin_m_Page_128.jp2
5e3d7e33fe58fb56461990da51caab65
1699343c2affd557b951b451b7ea7ef06bf67bba
69821 F20101111_AAAVNT fensin_m_Page_107.jp2
68da13a4eeedca9d41052147abe59be6
9b28da34df1dca56eebaa0635014e854b601a286
119096 F20101111_AAAVOI fensin_m_Page_129.jp2
0ef866eb24656a2ca6ddd0236cdbc796
57918098ee1f0bccd82ab7261f18de2e24fd764e
1035673 F20101111_AAAVNU fensin_m_Page_109.jp2
af27b00994fb483534f8515efc24b1f6
d679748c1cdff843fd79a67bf59a5ab2e9d1e957
869804 F20101111_AAAVOJ fensin_m_Page_130.jp2
fbafc81892f0d25bbb34e78e694501d2
4a5517b4b4a35c2ee8279ec63f89fedde1f29867
73304 F20101111_AAAVNV fensin_m_Page_110.jp2
9c9bb3f3f59bd4f452160fafd772e0c1
a88809b6425f8228361f441678622a7fcb082f7b
105337 F20101111_AAAVOK fensin_m_Page_131.jp2
562fd4aa963195b4b2fb17a506710e82
b0fa5aa495d94f1174c290f82017fcedc3646888
361720 F20101111_AAAVNW fensin_m_Page_111.jp2
76c1120236d9818707a0111e3d374d6a
3ce2b010fb261e8bcf2c267918674a80f4ca708d
59639 F20101111_AAAVOL fensin_m_Page_132.jp2
7f336866512a243c943c457e2bb82473
8c3a816bdbb2d51c2c7560345205525dfaaf2b81
948045 F20101111_AAAVNX fensin_m_Page_112.jp2
1b37f65cb7ed56a8021e27fd19aafeaf
1de9e4fa4f4485df01ed1f1378ac7fcf7e952a80
89567 F20101111_AAAVPA fensin_m_Page_157.jp2
3377633970506b9a5e9c011b83f1598f
b03ce901800188475301c1bb892196c804697b8c
127787 F20101111_AAAVOM fensin_m_Page_135.jp2
2f4c5fa9d0a20eb91e0406d4bb6166a3
9ba13f81047ae01aaf3affdf889c5363e9910d05
122946 F20101111_AAAVNY fensin_m_Page_113.jp2
3037506139aaeab86bf9d51eb7e5a242
4fd862cf7e219cdf17ae1a4e7f240233d9d04fe2
91331 F20101111_AAAVPB fensin_m_Page_159.jp2
2c2c39c47eed86c51081c25a122a3f0f
1265d66b293a4e5db33f235216e65530648502bf
118249 F20101111_AAAVON fensin_m_Page_138.jp2
c7297903c835af0bf5b37f95be3dc121
ac8cf33fd40b67dfda5be27870409d6dc624d93c
113399 F20101111_AAAVNZ fensin_m_Page_114.jp2
0eb8a578ca6c462f987d73c1f16658e5
1b8d07d28be31615a9786fc8f69fadebce272a1b
91786 F20101111_AAAVPC fensin_m_Page_160.jp2
c5d7e8b43ee8b9e3066796f39dec6a83
6c35dd739c7153070d5aba48a9e27919420e1b1f
118471 F20101111_AAAVOO fensin_m_Page_139.jp2
4ef80530932e48faf1e8431d521494a6
02fc03c8ddfba9207c1d19694af4011c7c6f3c40
F20101111_AAAVPD fensin_m_Page_161.jp2
02c9ccf6a82c2d4b73bff160f0782032
19c8c759faf127f9ddd4477e29bd2e1a84d2583b
108667 F20101111_AAAVOP fensin_m_Page_140.jp2
4be353f9da265e4dc89f15f498ca4488
bd15422c0402e9b9dc96368bbc73faf9ac65e05f
90361 F20101111_AAAVPE fensin_m_Page_163.jp2
a547e77bf83594506aaf43f665d15de1
906e70ad7bde812a2675aa324dbdb62258accb73
116841 F20101111_AAAVOQ fensin_m_Page_141.jp2
ac32b7be80485576180d5aeec4e64122
7b6c8643a59cb647c14c91783d2003ea99647b59
116738 F20101111_AAAVPF fensin_m_Page_164.jp2
c336efd9966ca8e341c5f113c003d829
cb578c9eb38cf8f3ca55118fc6d0fc31bb321cf0
118504 F20101111_AAAVOR fensin_m_Page_142.jp2
dde2fd8811990fb8bfb87102065fb4aa
3f400c42b86248c0f6748b7e5b42a5f85c2ccbae
118672 F20101111_AAAVPG fensin_m_Page_165.jp2
c4251374bb1def543a408e1095b95182
ebe1dfde8fc41ab0b000140191d1b3e62ce713e3
120692 F20101111_AAAVOS fensin_m_Page_143.jp2
05b357606859c50577ebce8fd27afd69
4349865262825ca5c8674af91a9f41874b97dae4
94384 F20101111_AAAVPH fensin_m_Page_166.jp2
db91e3b3cf99426510bc3079207734a1
70cf5666cc57d1a80d62b7c928d69c4f2bc30daf
104582 F20101111_AAAVOT fensin_m_Page_145.jp2
2a3b4a93c5feaca7caaf9ba683c609c1
1dffdc95aa9641663d0f406d2a397bf1ad65d630
111712 F20101111_AAAVPI fensin_m_Page_167.jp2
b8695b781a5de06d31957c203c70483a
2004e7ae7967bb261405c2b78cba9d8fd378d39f
108080 F20101111_AAAVOU fensin_m_Page_148.jp2
a6435e35d9b6360cba1e74914d050d93
79668533db99279b45adf3b82ee2f8ed34f7bed1
131524 F20101111_AAAVPJ fensin_m_Page_168.jp2
418f7e7539337007737d2e6a1eb24344
048ba8f10f3615af0f2b74ee6fa16ac4e0daa4fd
122150 F20101111_AAAVOV fensin_m_Page_150.jp2
6d68cfd830fe9aa1f9570f48270160b0
8df00ad0244fe8375dcb4be5b8f033822792de28
123474 F20101111_AAAVPK fensin_m_Page_170.jp2
c238abac06c862e4f64e01245b502bb1
d21bfb32030c41f8b1401ced1fad922483548607
116331 F20101111_AAAVOW fensin_m_Page_151.jp2
16263cdfb33f6811d63f3bf77a2e7d09
8c6866ee425ed99930b798ab59fa13f7118221e6
113285 F20101111_AAAVPL fensin_m_Page_172.jp2
1442f7123d4fc83a04f531419d58fdf0
49a6928ad29a9d7eafd23b7bd2a33dff98344741
103567 F20101111_AAAVOX fensin_m_Page_152.jp2
b2f35bbe13f28f0860673499c6fb4e45
84e8c0db8bcbe0817eb517699334577841ce1046
116099 F20101111_AAAVPM fensin_m_Page_174.jp2
a8c7effd62c89ea0d757ffcc9bdedbbf
fca48a973b33c76656574245083088144d92c349
98056 F20101111_AAAVOY fensin_m_Page_154.jp2
3f51a7b6dfe17a2ac6cbb4e846a78cd3
80ca552d84d717a938f43cce50ac5991bce3b4d1
F20101111_AAAVQA fensin_m_Page_003.tif
f9dca3925b3fa6e9b6e6ab0570044dd5
f33b25cd62cffa5119defe3cce79ba31a657fed5
123755 F20101111_AAAVPN fensin_m_Page_175.jp2
b6ff0502b1a3b12f852653397f140912
81f1bc03023d70f6a622c002dea11a728625edb5
85512 F20101111_AAAVOZ fensin_m_Page_156.jp2
1b8f288d11cde7c098c1ec019c27cdb0
07f7de137378afb578696023c752f9517c0c334f
F20101111_AAAVQB fensin_m_Page_004.tif
6dbf7e81a6c56a4a8e0547c0a1ab042a
de72e862abeeb3a6377862a55fec23c447e98d31
79168 F20101111_AAAVPO fensin_m_Page_177.jp2
8803ab4fee246abeaf7a238455b3d373
c27d21313021395094498ded0e223ee57f21140b
F20101111_AAAVQC fensin_m_Page_005.tif
7f1323b2bb26c9e7d56181e090d56dd1
e330bbfc0c7faf4137a7f3efb19956e026f9aaa5
107706 F20101111_AAAVPP fensin_m_Page_178.jp2
b27696c18aa3bfa1faccf56675ca6096
dc221dc304a721b74f9f01e291268d98252eff1e
F20101111_AAAVQD fensin_m_Page_006.tif
c4bb8aaf81878b61400724ec641b45ee
e11d5a08fefa6ec3acb73c38be298979b5ce28a0
138062 F20101111_AAAVPQ fensin_m_Page_180.jp2
3f3c71fa0ce7f41598b492c245e79ed2
518ebd61e2e75b7c93da4ed013a59199f68db301
F20101111_AAAVQE fensin_m_Page_007.tif
01cedbf9af2ddbb006fdefa3f56f1dec
3c5cbeb518eae7ad43d14e503611726e96d7ee2b
136810 F20101111_AAAVPR fensin_m_Page_181.jp2
1a1eeb47247847a23140103bcb296108
d16fa22fe34773523885b232eddda4b37f3f4fdd
F20101111_AAAVQF fensin_m_Page_009.tif
2493d4a9a6eb1537c47400e49c52735f
76d3061810a041cd04637ae40141e96296461177
144818 F20101111_AAAVPS fensin_m_Page_182.jp2
ab940f43c478f8286f9f9e4d0166d812
51e77de41dae552857c30156ac1c8d2c490a0991
F20101111_AAAVQG fensin_m_Page_011.tif
413b740a50ee108c6c887fac10bea50a
443c64c80a66e24a3f08f35eca86391f83859f6f
131514 F20101111_AAAVPT fensin_m_Page_183.jp2
f78556e5a8436fdaf3d2dea09ec95494
43dbebf05526e51637d4fdf719fd3ccc49290c53
F20101111_AAAVQH fensin_m_Page_012.tif
b9f07607eb86f3e1aff3e18598c558b2
bb4ca1dae07567584be83dbc61c4ed7d26c6b64d
138031 F20101111_AAAVPU fensin_m_Page_184.jp2
64c3c0a096ef4883a7bcf807ddeebfba
522764a6d02a2a55ba16741a2111eac5f78d55e6
F20101111_AAAVQI fensin_m_Page_013.tif
486adbb668aa3f79fc19d15d4ceddac7
a5806cbd67861e8d81dff91ddf493aae36305832
151853 F20101111_AAAVPV fensin_m_Page_185.jp2
d2fc79e9b0de8880f13cbf448a8d0a38
14b07469c1ad045569767062bc5c206f7d9fb00f
F20101111_AAAVQJ fensin_m_Page_014.tif
44c9eafe3831398337642bfbafe17e07
a459f929e8a2c2af4dc92dd72889705c3bfde456
141801 F20101111_AAAVPW fensin_m_Page_186.jp2
e639e015923ba07c6f9f894f82b2269a
ff6e37140d52b4036d2357b12c9cacec8163e894
F20101111_AAAVQK fensin_m_Page_016.tif
42a3cf8794b60b152a867e92fac5561d
bceb19eb93d5f7aa819a1553c58c6737a92f7136
F20101111_AAAVQL fensin_m_Page_017.tif
fa62998f697b9282fdf41ec5f994ca8a
50bd8f33ad0e572f227d396cc083371527f3a1c4
87901 F20101111_AAAVPX fensin_m_Page_187.jp2
17e297a4d1458addbe4070232b779c8c
dce06fbb6dcda5699f4a19a78c533b0d6d49f871
F20101111_AAAVRA fensin_m_Page_034.tif
82be17952c9541325c67065382fe532a
f46c742c21c8f38c967adcc41329d1521e9120d1
F20101111_AAAVQM fensin_m_Page_019.tif
3c7df608ed95a26abe44ea97f59a8e72
a8f763cb4c589e51398b8203912ae36a14d35258
94990 F20101111_AAAVPY fensin_m_Page_188.jp2
38be32817f3b25bc7e746c104b72ff37
14cbccd7492032e424d2c4c540366362c5d7553d
F20101111_AAAVRB fensin_m_Page_035.tif
df6ba7e27009248900e8812041c327f2
765e62fbc3873b23658fd6570f6fa28f57c0f7e4
F20101111_AAAVQN fensin_m_Page_020.tif
8628a0c07d34a44e2a132df479b02f32
603ca433519d2f21ee87b72481c703a58632b6d3
F20101111_AAAVPZ fensin_m_Page_002.tif
4ce95fa45506f131e0d9a20f84c1f876
c13e243789736bbe15f44b05deeed5704f4bba55
F20101111_AAAVRC fensin_m_Page_037.tif
58091dcda87e4980003cbaee9a414fd8
b5ddc63b959f0ab5d1dc546561be811fdbc17630
F20101111_AAAVQO fensin_m_Page_021.tif
022da3367b51485dcd0dbbbd06070dd9
94f0451555d311601888f61ae0186ee5453ef109
F20101111_AAAVRD fensin_m_Page_038.tif
4f470a364e349c8a0db2999f7d1c6bc5
c1e3c40a04036fcf45f352c874630fcdb2659d7d
F20101111_AAAVQP fensin_m_Page_022.tif
3e02c975b9bcf4d6549cf1ef2a734921
13fb343b5b7786d12274ac7c68ef4d765275115e
F20101111_AAAVRE fensin_m_Page_039.tif
2032432b298ec4e1dec1d4e77b315c3a
c40ae4b659ff9b556860346f031c39bf1bb454bc
F20101111_AAAVQQ fensin_m_Page_023.tif
69a00b3d1cfc437cb07336934cb7bcd9
91bc7a5fc3224b7faef8aeb4feb636b5707c5fd9
F20101111_AAAVRF fensin_m_Page_043.tif
00238d7e12c338070fe73040d6c5c4e8
0c56dc13160b2a11d66ae780253bf4b6bbcb9fef
F20101111_AAAVQR fensin_m_Page_025.tif
c68ba18d79f930dc53bed54236598735
387d87b2240324b808246a99c5faf881e502f937
F20101111_AAAVRG fensin_m_Page_044.tif
f5b916b8ce2ed6695423d1a3dca6c124
2f7d3870103c9b0c41678b52b8c8e05b73b1645d
F20101111_AAAVQS fensin_m_Page_026.tif
0bd8c22d3a2b77be3f4c85d4f84cbddb
d7da0ab2c59ed6cb7188d23e249880022d9e8acc
F20101111_AAAVRH fensin_m_Page_045.tif
8044ec3a283544214e81651bff26ca28
7615ab55018e1947bc0b24fad53932ede9d649b3
F20101111_AAAVQT fensin_m_Page_027.tif
50c38c1308411b26685c9dc95d6f50fd
b5f967b5d2ed98c5975b311547ff98db5cd27684
F20101111_AAAVRI fensin_m_Page_048.tif
d97a1ae3e1559e3c1a1aeaeb9e1e59e6
6fd7f50ab26de7987a4af3b6279d75e41dba8f16
F20101111_AAAVQU fensin_m_Page_028.tif
df85d92b8bd43724f851945375cacde5
1a9e65108cd97ac357fa2c689ad18190f03e5a45
F20101111_AAAVRJ fensin_m_Page_049.tif
86f2323dd8ce9e44e9d841795c60783e
2f89b660f527e02efe14d3da2677ec0277fc3e75
F20101111_AAAVQV fensin_m_Page_029.tif
cd0e9f4c16377e5286eeff5471c55379
41c3ed6849e7ff650c3d9a7b404c7b457e07ac29
F20101111_AAAVRK fensin_m_Page_050.tif
bfdf2ba4f25c5e4102e9f4a533d2bf98
7ad225af92bcb518d11e1ff5f8ce304de9963c9a
F20101111_AAAVQW fensin_m_Page_030.tif
60cacb4328beb8c8f615a82c1e43c365
c855273ff415259b52d0f5328c469efe28e1d9ee
F20101111_AAAVRL fensin_m_Page_051.tif
07a1a47ead8a295103cebf5babf6e268
fbbf03184138eff7f6b943a85269a79c261497db
F20101111_AAAVQX fensin_m_Page_031.tif
570225619a7c4525b20c7dc83434b910
7dfa4126884b1c1e05becace57ec6a8cf02d5a47
F20101111_AAAVSA fensin_m_Page_071.tif
373ac7c958c0be0208a08e29be15684c
d5ff959f63d408d8e6ca94a67d123fece68dde19
F20101111_AAAVRM fensin_m_Page_054.tif
9e02d5d35bfc97c4f79b6c47881acdb3
a9696311b9319b2ad55987669020eb227ae3eb96
F20101111_AAAVQY fensin_m_Page_032.tif
109c3bd2e04bfe0e64e9aa8d98130060
ab433a9e5f7b68088a16ba145ee636b97355d975
F20101111_AAAVRN fensin_m_Page_055.tif
58fb09e83e54838ef90a6d510450fd7e
b7c9a1eaeda2bd8cb69b3e2812d17a4dcc1dbc77
F20101111_AAAVQZ fensin_m_Page_033.tif
e6142d4eba13e1dc992ef9e3db233472
5abb7c7ab2fee7d3fa7649ad40172a346f6aabbf
F20101111_AAAVSB fensin_m_Page_072.tif
3c718f69bd52e7da0128d9648822f4bd
80bbbbb9df9ccfe501460bb4ed7a94117e028446
F20101111_AAAVRO fensin_m_Page_057.tif
16f62fbe1d01bd38870e1cb57b4a20d6
d6f04a2b40f3e6992c366ac1bb5926d4a303461b
F20101111_AAAVSC fensin_m_Page_073.tif
0841267559ade606e52f5e1defe28ba4
4910302785e1059ae6065a30e03bccde88d0e0b9
F20101111_AAAVRP fensin_m_Page_058.tif
001f95acc2793abb3a9355a00a776be1
6fdab8e8781f4d5edff4ead4bf588534a2c44aea
F20101111_AAAVSD fensin_m_Page_075.tif
f705fcc88fb42e7ca434e4a8908df14c
eec0ca28a7d531a7957fcec296dd5c33e218ce9d
F20101111_AAAVRQ fensin_m_Page_059.tif
7353a407a0baed062046db3cc14356de
3482eb0a627e389e5a8f1ad1d68cc1eed0dd6dc5
F20101111_AAAVSE fensin_m_Page_076.tif
b99182b5551f32e02eb80b16c27e0e2e
a7afee21acd70751d971d82f814302e2751c92ff
F20101111_AAAVRR fensin_m_Page_061.tif
b57212030f00c242544f0e1e05b02a2a
723fe5884870ca404f52ede390efb57cbc43e4ac
F20101111_AAAVSF fensin_m_Page_077.tif
f9bf431c6e6b97a4340b81f616163e8d
16c76bd8f0ba888e87f1038a9a2b165bd934c37d
F20101111_AAAVRS fensin_m_Page_062.tif
53ed4d28e3be998e7879f1e17de3dc0e
6967f470bf4fd8a7f3a704c9dc602203002a1397
F20101111_AAAVSG fensin_m_Page_080.tif
c9455840d14cd648098f6c7c9ab7dc2f
361044b03a551635737392dba1539a3da120ff55
F20101111_AAAVRT fensin_m_Page_064.tif
093d8a78bba16983da1ef66e4da0112e
611e32196eda3437f11e32f43d8e4f8bef224dfe
F20101111_AAAVSH fensin_m_Page_082.tif
e1cf818b903b98809777fbfd244dad64
69639e307e9c71f29a4add45a0f9f2cfbdaa263b
F20101111_AAAVRU fensin_m_Page_065.tif
687b5d50548a5f9f2ba8958cbe3c02b8
ed3784ab29fd64d30e7c8aeb32389e0aba99fadc
F20101111_AAAVSI fensin_m_Page_083.tif
78cc53446c5aaf950865ff1f2001c0b6
a8bb6f69ca760ea9f6b7e6fa8bfa59dd704c22ff
F20101111_AAAVRV fensin_m_Page_066.tif
bf6a6204108a73961acf48d88e0e5210
3e1b64d619d8bf5d286bcb0b8590ff0344218f92
F20101111_AAAVSJ fensin_m_Page_085.tif
a0522cabfaba203a44a5330fa3caa661
5749fa1501cfefcb68d99cdd400c25e84e105ed2
F20101111_AAAVRW fensin_m_Page_067.tif
b2cd38e63e88b371d59ba6b00a70d334
01e168cfaea915f0bc43071a2a68ae8b3b5117ef
F20101111_AAAVSK fensin_m_Page_086.tif
7b05c89de258787092943df870781770
b3886a15febbfaf4f97e842a87b50f67301f166a
F20101111_AAAVRX fensin_m_Page_068.tif
a72e7730f700bbd57e49ab95608bf479
9852a56cf6355cff80a7234ad9cc7c87c7d3d0fe
F20101111_AAAVTA fensin_m_Page_109.tif
467bf36e8254f99299a85bd482a42f16
eba113c9ef251d6e7837a76beb259c8372ae4015
F20101111_AAAVSL fensin_m_Page_087.tif
a1e91af6239850b79d056a831a6a5f5e
b2aec02be3d34ca13b48baf54efcc529c1aa962f
F20101111_AAAVRY fensin_m_Page_069.tif
d368093bc01b120765ecaf2855453093
80d7c048ae26119950945d337ffc039021c40cb5
F20101111_AAAVTB fensin_m_Page_110.tif
cea6a392f449a2d574595a4ba57718fe
ab1c1f8d64020e12a78b9ebc6d773ac0c54a1183
F20101111_AAAVSM fensin_m_Page_089.tif
17fea13df9e7939543a86c5a9db12748
bb17d3de84d10911fc6ed998b10e94365f71d88e
F20101111_AAAVRZ fensin_m_Page_070.tif
bf180e211c328d6c1969aeb05ed8218a
7f59d7afe822df11cfde077f96d4ee0d32a256f4
F20101111_AAAVSN fensin_m_Page_091.tif
87a19339c0ef3a33c8a80a15ba3c7714
e533f73c00f5acadf0a8b95fac49d2217b233dd8
F20101111_AAAVTC fensin_m_Page_111.tif
2aed12f471e68a29e1c6a3c5e51e752f
0c5c8f12d306059c6066afe9f4095312288a6746
F20101111_AAAVSO fensin_m_Page_092.tif
61aeb60185c7bbced62713ecca69166e
9e6f31c96b50af571cc35368fffa9b4a6fa530b4
F20101111_AAAVTD fensin_m_Page_113.tif
85ada7b40f4e76ef9c07121eeb164913
b6bd8716596397ab3d4ef2a951230976d1f9987d
F20101111_AAAVSP fensin_m_Page_094.tif
fda58c39c451ebb8645551eba6e0f5c1
52bbedb62425011ab65b10009fed6b3ac8be121a
F20101111_AAAVTE fensin_m_Page_114.tif
f08ec76b60745b433df4c6f7badc8cae
23079d0b76fc1f32224ec0c01e3e3d21b4fbc4d9
F20101111_AAAVSQ fensin_m_Page_095.tif
b185bff3756cf9c1ffbe94a11fbc27d5
6d06f10cb12a032bf2029e308ee49dcaf73b0f66
F20101111_AAAVTF fensin_m_Page_115.tif
ed963ea4b903f2cf6ee3c3890f4f0324
f91a6056ad0c616bada9fc9423d740bc8ee77a8a
F20101111_AAAVSR fensin_m_Page_097.tif
96e7c6517bab8fd1bc1dc53e83776185
a1bf970c49776bbf5e9c8d43d9a2529ca1087b4c
F20101111_AAAVTG fensin_m_Page_116.tif
aa25d2eea642ba5d5d25e2c862a0a3ff
3fb8059aa0623074ac5902fae8dbdfcfef37e416
F20101111_AAAVSS fensin_m_Page_098.tif
edc25b0fed5bdb223419e6335e5adbc0
c5ac4482962b4dac00bbce4aa09b7f26eb20f1db
F20101111_AAAVTH fensin_m_Page_117.tif
dd9289bee91339783538d271fee28fd6
f96d3397601c1d0310a46d465eb799f379a08c9c
F20101111_AAAVST fensin_m_Page_099.tif
2e614590098fc3b685f4a40bc11c0b0d
da749fcdf46832bb6159b98110b7b87affe0da3e
F20101111_AAAVTI fensin_m_Page_118.tif
16f78a97a7363912f195a9e57a5173d7
2be1f288b8b8c095e6bfb30f13a9cc0ae0e98948
F20101111_AAAVSU fensin_m_Page_100.tif
2ca88900bfcf462f2c318047095c75a3
bff157136e4665a8f4dfbdff0cb0f44a7abeb29e
F20101111_AAAVTJ fensin_m_Page_120.tif
cb5baed0c9ca34e20ae4b1fc60781244
f3fa72cd4116bfffc068a21ea609d5a0c39063ad
F20101111_AAAVSV fensin_m_Page_102.tif
83f72fbe4e3694600c06a8e39c9f197f
4f9322c983195ae9bb4c25b18fc7789fd2ce9f86
F20101111_AAAVTK fensin_m_Page_121.tif
0971ac37514ff4b524553c14f746e92d
41e9510957180efb84adb0efc4f6d982bfada9fa
F20101111_AAAVSW fensin_m_Page_104.tif
57a9420babba2f898517c34546e35b99
5df361e0e08e1e11d4a392fa5f635cc2c4606894
F20101111_AAAVUA fensin_m_Page_139.tif
e82ba13daf0dc88422f413d05499e03b
85cd19b04b273d31843d7a7e1d3f7efee5c586f0
F20101111_AAAVTL fensin_m_Page_122.tif
cbf76db367ed7e346feb9ed5a46ad4c3
82856ec22cf3bc486d5f953ff74bd05f66a12a29
F20101111_AAAVSX fensin_m_Page_105.tif
9d25af8cade16b2b25131031b32c7e65
2daaf84ce0ddc73808ad80ad9ef9a55b67eea7f6
F20101111_AAAVUB fensin_m_Page_143.tif
77eccfbcbd98b93b5b1fbc7368055e24
92f85527fb1195fa3a6fbdf6b620f11ee8738a3a
F20101111_AAAVTM fensin_m_Page_123.tif
0949995f78b7252ad82499d626a55220
3b4498921ef05d57d88a700f7be0fbd2ec68cb22
F20101111_AAAVSY fensin_m_Page_107.tif
af44f010949fc8ba1a6de32ba71e0048
15fe231a75423b74e91aa9fddaccfa62b94a6b5f
F20101111_AAAVUC fensin_m_Page_144.tif
6539fd9c0d0e70b54ff49e8644d80c08
e4bce8fc594ca6dcb50c41081665b7a2faf13024
F20101111_AAAVTN fensin_m_Page_124.tif
109e78f0c4027d6b0f72f2df94d66edf
ea0e0ba386de0c3081aa826f846763ce79c6dd7c
F20101111_AAAVSZ fensin_m_Page_108.tif
e33aaac9e9652927fa6b13ad01a07762
1a20c0aa2f695bcafc1770872eede618af15fb5b
F20101111_AAAVTO fensin_m_Page_125.tif
500c0f37004dd2035b1823e6b3c649a8
38a512c74373b2741e4a298bc04594708d4d6218
F20101111_AAAVUD fensin_m_Page_145.tif
d6f2178cdbd4a12f84e6206b6cc48c9f
1ce5b857859b08bf8b72b7bbde536373f6855549
F20101111_AAAVTP fensin_m_Page_126.tif
864b85f1913884d48a4131e922838e2b
48fa886f1ccec9b3b4e58c4f760f840c323ddead
F20101111_AAAVUE fensin_m_Page_146.tif
8ade98ad88590a2102a9f1662c7ba1c3
293ad49ec66706dcbb89c674a44f3e3515a218e7
F20101111_AAAVTQ fensin_m_Page_127.tif
d46eeb923340ceaf78c61a695f61d153
e110150af43875283ea6ce29ab9d85fe5d6937d1
47127 F20101111_AAAWAA fensin_m_Page_156.pro
1eb74f8b2e440a35e3d76cb45bf6f437
1b7797a7be075eed8f9f17b3ac96501fe40f36f9
F20101111_AAAVUF fensin_m_Page_148.tif
4cb5ca04bde2c691f55be9408e2af689
385a211dd68de226ecf76d8fc8cbcf6a03a972e2
F20101111_AAAVTR fensin_m_Page_129.tif
2f47725c2694aed89291b12e151e8ba1
8a3f598a3dfdf0fc8722d601c36136de82a2a78a
50912 F20101111_AAAWAB fensin_m_Page_157.pro
069ccdb5401fedd6dcb0ef6294be9940
20fac0cdf04887282b3f7760ec535150c971cbc4
F20101111_AAAVUG fensin_m_Page_149.tif
b4973230509adbf4ac5f37078f56ce5c
a6adee220817cbe0c3f41a6555b3ebaa7d2e8db7
F20101111_AAAVTS fensin_m_Page_130.tif
6a20ad1b2fe6c494cb723332b7e3744e
b9c75ed211e4f4ded469690b698d456702ba1446
50904 F20101111_AAAWAC fensin_m_Page_158.pro
ed1ceeb02af935c17aa308874b43520e
5c1616754288a242ee0339a9c40a773001f89b80
F20101111_AAAVUH fensin_m_Page_150.tif
1154a44451f38f29f916c7864e00b39f
1ca743b70b8e1bf7578bdc995d336918a105782d
F20101111_AAAVTT fensin_m_Page_131.tif
4830ce7133f39c6b8b75dbefb4604607
746473adcf061c5becb5436897c3f3f29e5602d5
51289 F20101111_AAAWAD fensin_m_Page_160.pro
a0eea537ecef5699c98c1b37e2dbc545
d5e5cb1040854ef6bcea0ff7badd38bd5e06df62
F20101111_AAAVUI fensin_m_Page_151.tif
f463b2546b6c7fc2cd719a3adee0e9f5
a9153329d89f05441229c450cccf4eb50cb5eddb
F20101111_AAAVTU fensin_m_Page_132.tif
483d702ea4557b2d06ff113d4bae0c43
59af0a8d38b0f15ce984b9e977fe977202914326







DEVELOPMENT OF THE MCNPX DEPLETION CAPABILITY:
A MONTE CARLO LINKED DEPLETION METHOD THAT AUTOMATES THE
COUPLING BETWEEN MCNPX AND CINDER90 FOR HIGH FIDELITY BURNUP
CALCULATIONS



















By

MICHAEL LORNE FENSIN


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2008

































2008 Michael Lome Fensin


































To the spirit of the west endzone crew









ACKNOWLEDGMENTS

This work was initially supported by the Department of Energy, Office of Nuclear Energy,

Advanced Fuel Cycle Initiative, AFCI. The AFCI program then evolved into the Global Nuclear

Energy Partnership Program, GNEP, which then continued to support this effort. The author

would first like to acknowledge his thesis advisor, Dr. Samim Anghaie, from the University of

Florida, for his support and guidance through the process of developing an interesting problem,

and then seeing the efforts of solving that problem till conclusion. The author would then also

like to thank, Dr. Edward Dugan, Dr. Alireza Haghighat, Professor James Tulenko, and Dr.

William Vernetson for providing the author with a comprehensive education in reactor physics

and fuel depletion analysis. The author would also like to thank Dr. Wesley Bolch, Dr. David

Gilland, Dr. Travis Knight, Dr. David Hintenlang, from the University of Florida, for instilling a

fulfilling education in the areas of applied radiological engineering. The author would then like

to thank Rick Kapernick, Thomas Marcille, Dr. Dave Poston, and Dr. Pratap Sadasivan, from the

space reactor design group at Los Alamos National Laboratory, for their financial and mentor

support, giving the author a solid foundation in reactor design considerations. The author would

also like to thank Dr. Mike Miller, Dr. Peter Santi, Dr. Marytn Swinhoe and Dr. Steve Tobin,

from the reactor safeguards group at Los Alamos National Laboratory, for their guidance and

suggestions in the methodology development of certain applied capabilities. The author would

also like to thank Dr. Holly Trellue, from the reactor design group at Los Alamos National

Laboratory for her valued guidance and support from experience in developing a linked

depletion capability. Most importantly the author would like to thank Dr. Gregg Mckinney, Dr.

Michael James, Dr. Russell Johns, Dr. Joe Durkee, Dr. Joshua Finch, Denise Pelowitz, Dr.

Laurie Waters and Dr. Jay Elson, for their patience, support and guidance in implementing the

methodology, developing theory, and testing the code's capabilities. Finally, the author would









especially like to thank Dr. John S. Hendricks for his patience and guidance as a superb mentor

at Los Alamos National Laboratory.









TABLE OF CONTENTS

page

A CK N O W LED G M EN T S ................................................................. ........... ............. .....

L IS T O F T A B L E S ................................................................................. 8

LIST OF FIGURES ................................... .. .... .... ................. 10

A B S T R A C T ............ ................... ............................................................ 12

CHAPTER

1 INTRODUCTION AND BACKGROUND ........................................ ....................... 14

R actor D evelopm ent H history ....................................................................... ...................14
A advanced F uel C ycle Initiative ..................................................................... ...................19
Global Nuclear Energy Partnership ......................................................... ............... 21
A advanced Burner Reactor Cam paign.................................... ........................... ......... 23
R actor Safeguards C am paign ........................................................................................25
Motivations for a Monte Carlo Depletion Tool in MCNPX................................................27
O b j e ctiv e ............. ..... ....... ..... ............................................... ................ 2 9

2 SURVEY OF STATE OF THE ART APPROACHES TO DEPLETION ANALYSIS........32

T h e D epletion E qu action ................................ ................................................... .. 33
Zero-D im ension D epletion Calculation........................................... ........................... 35
M atrix Exponential M ethod ................................................ ................................ 37
The CINDER90 Linear Markov Chain Method...................................... .....................40
N necessity for a Linked Approach .................................. ................................. ..............44
Methods of Steady State Reaction Rate Calculation ................................... .................47
Deterministic Approach to Reaction Rate Calculations...............................................48
Monte Carlo Approach to Reaction Rate Calculations ................................................54
Current Capabilities and Methods of State of the Art Depletion Codes.............................63
Im plem entation A rchitecture............... ............ ...... ........... ................... ...............64
Nuclide Inventory Limitations/ Calculation Size Considerations ..............................68

3 METHODOLOGY DEVELOPMENT........................................................ ............... 71

A approach to Isotope T racking ........................................................................ .................. 72
Fission Product Tiers ............... ................. ....................... ........... 74
Isotope Generator Algorithm..................................... ............................. 79
Manual Time-Dependent Isotope Concentration Changes.................. ................ 81
M etastable Isotope Tracking .................................................. .............. ............... 85
Calculation of Depletion Equation Coefficients ............... .............................................86
Im portant R actions Follow ed ............................................... ............................. 87
Flux N orm alization ........................................................ ......... .. ............ 94









A utom atic Fission Y ield Selection ....................................................... ..... .......... 101
Time Dependent Cross Section Averaging Techniques ...................................................... 104
M ON TEBURN S M ethod ............................................................................ .. 106
CELL-2 M ethod ................................. .. .. ... ... .. ....... .......... 107
Prelim inary A analysis ............... .... ...................... ... ...................... ...............108
Further Commentary on both Cross Section Averaging Techniques............................112

4 IMPLEMENTATION ARCHITECTURE................................................................ 114

In p u t In te rfa c e ................................................................................................................. 1 1 5
Output Structuring .............. ..... ............................ ............ 122

5 B E N C H M A R K S .................. .............................................. .......... ...... ........ 128

OECD/NEA Bumup Credit Phase IB Benchmark Specification ...................129
OECD/NEA Bumup Credit Phase IB Results................................................. ...............133
Nuclear Data ............. ............ ..... ......... 139
Treatment of Normalization Parameters .................................................. ...............140
Inadequate Tim e Steps .......................................... .. .. .... ........ .. .... .. 140
Statistical V ariance ............................... ................. ...... ................. 141
D differences in N eutron Spectra ............................................................ ..................... 14 1
Spatial Reactivity Effects of the Modeled Geometry...........................142
Further M ONTEBURN S Comparisons..................................... ........................ 142
H. B. Robinson Benchmark Specification...................... ..... .......................... 145
H B Robinson Benchm ark Results ................... ......................................................... 149
OECD/NEA Bumup Credit Phase IVB Benchmark Specification .....................................151
OECD/NEA Bumup Credit Phase IVB Benchmark Results................. ..... .............156
Further C onsiderations.......... ........................................................................ ...............166

6 CONCLUSIONS AND SUM M ARY ...................................................... ............... 167

A ccom plishm ents .............................................................. .. .... .. ............ 16 8
A analysis of B enchm ark R esults.............................................................. ..... ............ 169
F future W ork .. ........ ........................................................... ...................................... 175

A PPEN D IX SA M PLE IN PU T ........................................................................ .................. 177

L IST O F R EFE R EN C E S ......... ........................................................................................... 179

BIO GR A PH ICA L SK ETCH ............................................................................................188











7









LIST OF TABLES


Table page

3-1 M O X fuel com position. .......................................................................... .....................89

3-2 B orated w after com position. ...................................................................... ...................90

3-3 Emitted and recoverable energy per fission ofU-235 ........ .......................................97

3-4 Prompt fission energy release data available in MCNPX .............................................98

3-5 Available actinide fission yield distributions in CINDER90................. ..................102

4-1 OM IT keyw ord variable tasks. .......................... ......... .................................................117

4-2 Options invoked for different B2 values. ................................... ........... ................... 118

4-3 Options invoked for different B2 values. ................................... ........... ................... 119

4-4 Available options for the B3 value of the BOPT keyword. .................. .................119

4-5 M ATM OD keyword variable values. ........................................ ......................... 121

5-1 Examined bench ark cases. ..... ........................... ......................................... 129

5-2 Model description for Phase IB Benchmark*........................ ... .................130

5-3 Fuel composition for Phase IB Benchmark (density =10.045 g/cm3). ..........................131

5-4 Clad composition for Phase IB Benchmark (6.44 g/ cm3)............................................132

5-5 Borated water composition for Phase IB Benchmark (0.7569 g/ cm3)............................132

5-6 Operating history data for Phase IB Benchmark. ................................. .................132

5-7 Operating power (MW) for each case for Phase IB Benchmark ...................................133

5-8 Initial Phase IB Benchmark results and comparison of Case A (27.35 MWD/MTU). ...133

5-9 Initial Phase IB Benchmark results and comparison of Case B (37.12 MWD/MTU).....134

5-10 Initial Phase IB Benchmark results and comparison of Case C (44.34 MWD/MTU).....135

5-11 MCNPXc Phase IB Benchmark results and comparison of Case A (27.35
M W D /M TU ). .............. ... ........ ....................................... ........ .. ......... 137

5-12 MCNPXc Phase IB Benchmark results and comparison of Case B (37.12
M W D /M TU ). .............. ... ........ ....................................... ........ .. ....... .. 138









5-13 MCNPXc Phase IB Benchmark results and comparison of Case C (44.34
M W D /M TU ). .............. ... ........ ....................................... ........ .. ......... 139

5-14 MONTEBURNS Phase IB Benchmark results and comparison of Case A (27.35
M W D/M TU). .................. ........ ................................ ........ .. ....... .. 144

5-15 MONTEBURNS Phase IB Benchmark results and comparison of Case A (37.38
M W D/M TU). .................. ........ ................................ ........ .. ....... .. 144

5-16 Fuel assembly design data for H. B. Robinson Benchmark. .........................................147

5-17 Operating data for H. B. Robinson fuel assembly. .................................. ............... 148

5-19 The H. B. Robinson Benchmark results using MCNPX compared with SCALE...........149

5-20 The H. B. Robinson Benchmark results using MCNPXc compared with SCALE. ........150

5-21 Fuel assembly design data for OECD/NEA Burnup Credit Phase IVB Benchmark.......152

5-22 Initial MOX fuel composition for Case A for Phase IVB Benchmark..........................153

5-23 Initial MOX fuel composition for Case B for Phase IVB Benchmark ..........................154

5-24 Borated water composition for Phase IVB Benchmark (660ppm boron, density
0 .72 4 5 g/cm 3)............................................................................................. . 154

5-25 Zicaloy-2 composition used for clad and guide tube material for Phase IVB
B enchm ark (density 5.8736 g/cm 3)......................................... ............................. 156

5-26 Operating data for Phase IVB Benchmark.................... .....................156

5-27 EOC 1 results for Case A of the Phase IVB Benchmark ............................................... 157

5-28 EOC 2 results for Case A of the Phase IVB Benchmark ............................................. 158

5-29 EOC 3 results for Case A of the Phase IVB Benchmark ............................................. 159

5-30 Five year cooling time results for Case A of the Phase IVB Benchmark....................... 160

5-31 EOC 1 results for Case B of the Phase IVB Benchmark. .........................................161

5-32 EOC 2 results for Case B of the Phase IVB Benchmark........ ............ ................. 162

5-33 EOC 3 results for Case A of the Phase IVB Benchmark ............................................ 163

5-34 Five year cooling time results for Case B of the Phase IVB Benchmark........................164









LIST OF FIGURES


Figure page

2-1 L inked D epletion Process. ........................................................................ ...................46

2-2 Flux depression in the neighborhood of a resonance at Eo. .......................................... 51

2-3 Flux depression in the neighborhood of a resonance at Eo. ...........................................53

2-4 Flow diagram for the Monte Carlo particle transport process.........................................56

3-1 M onte-Carlo-linked depletion process model......................................... ............... 72

3-2 Uranium-235 fission product yield distribution ...........................................................75

3-3 Fission Yield Distributions for Various Plutonium and Uranium Isotopes.81 ...................76

3-4 Original fission product array containing 167 fission products.................. ...............77

3-5 Resulting time-dependent neutron multiplication from implementing fission products
with and without cross section models in MCNPX as compared to the benchmarked
M O N TE B U R N S m ethod. ........................................................................ .....................78

3-6 Percent difference of certain fission products from MONTEBURNS using MCNPX
Tier-1 fission products with and without cross section models.................. ...............78

3-7 Percent difference of certain fission products from MONTEBURNS using MCNPX
Tier-2 and Tier-3 fission products with cross-section models........................................79

3-8 Enhanced fission product array containing 220 isotopes............................. .............79

3-9 Nuclides created from a decay or nuclear reaction process and generated for particle
transport by the isotope generator algorithm .3........... ............................................ 80

3-10 Examined infinitely reflected pin cell geometry ............................. ....8............89

3-11 Percent difference of certain reference actinides between the OLD MCNPX method
and the M ONTEBURN S m ethod. ............................................ ............................. 91

3-12 Percent difference of certain reference actinides between the NEW MCNPX method
and the M ONTEBURN S method. ............................................ ............................. 92

3-13 Percent difference of certain reference fission products between the NEW MCNPX
method and the MONTEBURNS method. ............................................. ............... 92

3-14 Tim e-dependent eigenvalue calculations ........................................ ....................... 94

3-15 Fission yield selection process diagram ............................................... ......... ...... 104









3-16 MONTEBURNS cross section averaging technique ..................................................... 106

3-17 CELL-2 Predictor-Corrector m ethod .................................... ............................. ........ 107

3-18 Time dependant neutron multiplication for the predictor corrector method tests. ..........109

3-19 Burn steps implemented to achieve 2191 day burn ...................................................110

3-20 Time dependent neutron multiplication utilizing 5 burn steps. ............. ..................111

3-21 Time dependent neutron multiplication utilizing 10 burn steps. ..................................11

3-22 Percent Difference in keff When Compared with the NoPC 30-Step Case ...................... 112

4-1 B urn card input structure. ..................................................................... ..................... 115

4-2 BURN card referencing burn materials ml, m3 and m4 .............................................117

4-3 E xam ple collision rate outputs ........................................................................... ... .... 123

4-4 Example burnup sum m ary table. ............................................ ............................. 124

4-5 Example individual material burnup data. ............................ ....................................124

4-6 Example cropped actinide output at two separate timesteps and fission product
output for one time step. ................................... .. .... ........ ...............125

4-7 Cropped example of print table 220............ .......................................... .................126

5-1 Infinitely reflected pin cell geom etry........................................ ........................... 130

5-2 Infinitely reflected pin cell geom etry........................................ ........................... 146

5-3 MOX fuel enrichment map for Phase IVB Benchmark.............................155









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

DEVELOPMENT OF THE MCNPX DEPLETION CAPABILITY: A MONTE CARLO
LINKED DEPLETION METHOD THAT AUTOMATES THE COUPLING BETWEEN
MCNPX AND CINDER90 FOR HIGH FIDELITY BURNUP CALCULATIONS

By

Michael Lorne Fensin

May 2008

Chair: Samim Anghaie
Major: Nuclear Engineering Sciences

Monte Carlo-linked depletion methods have gained recent interest due to the ability to

more accurately model complex 3-dimesional geometries and better track the evolution of

temporal nuclide inventory by simulating the actual physical process utilizing continuous energy

coefficients. The integration of CINDER90 into the MCNPX Monte Carlo radiation transport

code provides a high-fidelity completely self-contained Monte-Carlo-linked depletion capability

in a well established, widely accepted Monte Carlo radiation transport code that is compatible

with most nuclear criticality (KCODE) particle tracking features in MCNPX. MCNPX depletion

tracks all necessary reaction rates and follows as many isotopes as cross section data permits in

order to achieve a highly accurate temporal nuclide inventory solution.

This work chronicles relevant nuclear history, surveys current methodologies of depletion

theory, details the methodology in applied MCNPX and provides benchmark results for three

independent OECD/NEA benchmarks. Relevant nuclear history, from the Oklo reactor two

billion years ago to the current major United States nuclear fuel cycle development programs, is

addressed in order to supply the motivation for the development of this technology. A survey of

current reaction rate and temporal nuclide inventory techniques is then provided to offer









justification for the depletion strategy applied within MCNPX. The MCNPX depletion strategy

is then dissected and each code feature is detailed chronicling the methodology development

from the original linking of MONTEBURNS and MCNP to the most recent public release of the

integrated capability (MCNPX 2.6.F). Calculation results of the OECD/ NEA Phase IB

benchmark, H. B. Robinson benchmark and OECD/ NEA Phase IVB are then provided. The

acceptable results of these calculations offer sufficient confidence in the predictive capability of

the MCNPX depletion method. This capability sets up a significant foundation, in a well

established and supported radiation transport code, for further development of a Monte Carlo-

linked depletion methodology which is essential to the future development of advanced reactor

technologies that exceed the limitations of current deterministic based methods.









CHAPTER 1
INTRODUCTION AND BACKGROUND

The purpose of the development the Monte Carlo linked depletion methodology within

MCNPX is to provide a high fidelity depletion methodology available within a widely accepted

code package that is already the "gold standard" for radiation transport calculations. To

understand why development of this methodology is vital to the current nuclear community and

why the methodology must be implemented into MCNPX, involves first understanding the

history of reactor development from the first natural reactor in Oklo, Gabon, to the current major

United States fuel cycle development programs, as well as why the current United States fuel

cycle program, the Global Nuclear Energy Partnership (GNEP), is interested in such a

technology.

Reactor Development History

The first self-sustaining chain reaction achieved on earth predated the first man made self-

sustaining chain reaction by roughly two billion years. The governing process to maintain this

self-sustaining reaction required that precisely one neutron from each fission process resulted in

another fission event. A delicate balance between processes that resulted in increasing the

neutron population, positive reactivity, and processes that resulted in decreasing the neutron

population, negative reactivity, was required in order to maintain a self-sustaining reaction. In

Oklo, Gabon, considerable heat occurred in several uranium rich deposits when groundwater

seeped into the uranium deposits leading to ample moderation that sufficiently increased the

probability of fission to allow for a self-sustaining chain reaction.1 As the uranium rich deposits

burned fissile uranium-235, fission products and other transmuted actinides were generated. In

some nuclear reactions, fertile actinides consumed radiation resulting in new fissionable

actinides that along with uranium-235 contributed to maintaining the sustainability of the chain









reaction. Conversely, the production of fission products and non fissile transmuted actinides, as

well as the reduction in moderator density from heat build, contributed to decreasing the

probability of fission, ultimately hindering the sustainability of the chain reaction.2

Like a gigantic coffee pot percolating away over hundreds of thousands of years, the Oklo

reactor would first undergo a fission chain reaction until enough heat and nonfissile nuclides

were generated to hinder the self-sustaining chain reaction. The system would then cease to

maintain self-sustaining fission events resulting in a cool off in the moderator leading to an

increase in the moderator density. This subsequent moderator density increase would

correspondingly lead to enough positive reactivity to ultimately restart the chain reaction.

About 2 billion years later, scientists would later discover the fission process and

mechanisms in order to fission a nucleus and maintain a self-sustaining reaction. In 1930, W.

Bothe and H. Becker found that a highly penetrating radiation was emitted when beryllium,

boron or lithium were bombarded by alpha particles from a polonium source.3 In 1932, Irene

Currie and her husband found that protons were produced when striking hydrogen containing

substances, such as paraffin, with this newly discovered penetrating radiation. James Chadwick

later demonstrated that the evidence from these experiments was compatible with the assumption

that this new radiation consisted of an uncharged particle having approximately the same mass of

a proton.3 He named this new particle "neutron".3

In 1934, Enrico Fermi irradiated uranium with neutrons trying to produce the first

transuranic element; however, he accidentally achieved the world's first nuclear fission. In 1938,

he received the Nobel Prize in Physics, "for his demonstrations of the existence of new

radioactive elements produced by neutron irradiation, and for his related discovery of nuclear

reactions brought about by slow neutrons".3 In 1939, Hans and Strassmann later showed that the









products from neutron bombardment of uranium included elements in the medium mass region,

much lighter than the initial uranium nuclide. The presence of these medium mass region

nuclides from the bombardment process suggested that the nucleus had split during the reaction,

and the fact that the sum of these medium mass nuclides did not add to the sum of the initial

parent uranium nuclide suggested that some of the mass in the reaction was converted into

energy. L. Meitner and O. Frisch termed this process fission and also calculated the energy

released during fission of a U-235 nuclide to be 200 MeV.3

Albert Einstein then wrote his famous letter to President Franklin Roosevelt on August 2,

1939 stating "that the element uranium may be turned into a new and important source of energy

in the immediate future." That source of energy was to be utilized by the United States to create

the first atomic bomb. With theoretical models and simulation, Enrico Fermi later theorized that

the neutrons emitted in fission might induce fission reactions concluding that it should be

possible to sustain a chain reaction in uranium and that this chain reaction could be made steady

state by implementing a balance of fissionable and nonfissionable material. Fermi and Leo

Szilard further proposed placing uranium in a matrix of graphite, where graphite was used to

slow the fission neutrons to low enough energy in order to increase the probability of fission to

maintain a self-sustaining chain reaction. On December 2, 1942, implementing a delicate

balance of positive and negative reactivity, based on proper modeling and simulation, the first

controlled self-sustaining chain reaction was achieved in a squash court under the University of

Chicago's Stagg Field.3 The later implementation of this discovery was unleashed, on August 6,

1945, when Little Boy exploded over Hiroshima killing over 100,000 people, and on August 9,

1945, when Fat Man exploded over Nagasaki killing 75,000, ultimately ending the war with









Japan. Both atomics bombs were designed using the Monte Carlo modeling and simulation

method.

In a later document in 1946, Enrico Fermi published a scheme for outlining the future uses

of nuclear energy. In this document he clearly stated that controlled self-sustained fission chain

reaction energy may be utilized for power generation for the world by use of a steam cycle or

direct energy conversion process.4 By enriching the amount ofU-235 in natural uranium, the

uranium could be combined with a water coolant and poison materials in order to make a self-

sustaining power reactor system. On December 8, 1953, United States President Dwight D.

Eisenhower gave the Atoms for Peace speech to the United Nations and outlined the advantages

of commercial power generation. He later tasked the United States Navy with developing the

first commercial power station. A proposal by Duquesne Light Company was later accepted for

the Shippingport Atomic Power Station, and the first reactor went critical on December 2nd

1957.5 The Shippingport Atomic Power Station was a type of light water reactor (LWR) called a

pressurized water reactor (PWR). Another type of competing LWR design was the boiling water

reactor (BWR), which was developed by a combined effort between Argonne National

Laboratory and General Electric. The GE Valecitos BWR was the first commercial BWR to be

licensed by the United States Atomic Energy Commission.6

The PWR concept is characterized as a system in which the coolant is sub-cooled and

contains homogeneously dispersed boron poison in order to maintain criticality. The system

utilizes an indirect dual-cycle that uses a steam generator to transfer heat from the exiting

primary radioactive subcooled reactor coolant to a secondary loop containing nonradioactive

steam. This nonradioactive steam is used to spin a turbine that when combined with a generator

creates power for hundreds of thousands of homes.









The BWR concept is characterized as a system that has boiling in the reactor core, with the

bulk coolant containing no boron, utilizing a direct cycle for power conversion (the

demonstration BWR/1 plants utilized a dual cycle). Instead of a secondary loop, this type of

system exhausts high quality steam from the top of the reactor through the use of moisture

separators and dryers. This radioactive steam is fed directly into the turbines that are also

connected to generators used to make power for hundreds of thousands of homes.

The progress of these commercial power systems later led to the development of other

competing reactor systems. Due to the advantages in being able to mine large amounts of

uranium ore, and due to the thought that a country developing enrichment technology may also

be developing weapons technology, the Canadians decided to develop a system that would not

rely on enrichment technology. By taking advantage of the scattering and absorption properties

of deuterium, the Canadian deuterium uranium reactors (CANDU) were able to generate power

with natural U-235 concentrations. Other technologies were also spawned employing a variety

of coolants and moderators such as the gas cooled fast reactor (GCFR) and graphite moderated

reactors in order to achieve higher operating temperatures and enhanced thermodynamic

efficiency. Technologies were also developed to support the breeding of fissile nuclides

incorporating various operational characteristics such as the liquid metal fast breeder reactors

(LMFBR).8

Though all the vast reactor types incorporate many different materials, fuel and coolant

forms, geometry combinations and operating strategies, the basic design objectives are still very

much the same. The goal is to determine the correct balance of fissile and control material

needed in order maintain the self-sustaining chain reaction, operate the system at a prescribed

power, for a prescribed amount of time, achieving some type of operating objective while









accounting for reactivity deficits. These reactivity deficits include temperature and isotope

buildup effects. The operating objectives range from maximizing capacity factor, to minimizing

discharged wasted, to producing irradiation environments, to generating medical isotope

material.9, 10, 11, 12 In order to develop the most optimized and advanced systems for specific

operating strategies, designers constantly examine varied types of material and geometry

combinations in order to design a system that meets the operating goals at the minimum cost.

This process therefore involves simulating a model in order to investigate many varied types of

designs, with many design iterations and perturbations, until an optimum configuration is

reached.13 As the model becomes more detailed and gives better understanding of what is

actually happening, the designer is able to make a more well-informed decision on how to

proceed with designing the system. Furthermore, as the development decisions for nuclear

reactor extend beyond running the system at a prescribed power for a prescribed amount time to

minimize the nuclear waste and addressing proliferation concerns, a higher fidelity modeling

methodology is required in order to make more-informed design decisions.14, 15

Advanced Fuel Cycle Initiative

The Advanced Fuel Cycle Initiative (AFCI) was established October 1, 2002, as an

evolution of the Advanced Accelerator Applications (AAA) program. The AAA program was a

Department of Energy (DOE) program commissioned by Congress in FY2000 to construct a

national effort by DOE laboratories, industry, and universities in order to address key significant

nuclear issues facing the United States, particularly the advanced fuel cycle technologies,

actinide transmutation, and treatment of spent nuclear fuel. The mission of the program was to

develop a technology basis for the transmutation of nuclear waste and demonstrate the

practicality of the process for use in long-term nuclear waste management.16 Initially, the focus

was on the use of accelerator based systems, hence the program name AAA. In 2002, the









program was evolved into AFCI with the focus changing to chemical separations and

transmutation fuel technologies. The idea was to couple reactor use of fuel with transmutation

and separation technologies in order to improve the management of nuclear waste through fuel

recycle and transmutation.16

The mission of the AFCI program was to develop advanced fuel cycle technologies

including spent fuel treatment, advanced fuels development, and transmutation technologies for

application to current operating commercial nuclear reactors and next-generation reactors with

the goal of (1) reducing volume of high-level waste; (2) reducing long-lived and highly

radiotoxic elements; (3) reclaiming valuable energy content in spent nuclear fuel.

AFCI was part of an integrated strategy of the DOE office of Nuclear Energy, Science and

Technology, which also includes the Nuclear Power 2010 Program, the Generation IV Nuclear

Energy Systems Initiative (Generation IV), and the Nuclear Hydrogen Initiative. AFCI was

important in reestablishing the viability of civilian nuclear capabilities and facilities in the United

States by utilizing services of DOE laboratories, industry, and universities. The development of

an advanced high fidelity technology for simulating transmutation and depletion of nuclear fuel

fit nicely with the goals of AFCI, thus leading to the initial funding of the MCNPX Monte Carlo

linked depletion project.17 However, in FY2007, AFCI along with the Nuclear Power 2010

Program, the Generation IV Nuclear Energy Systems Initiative (Generation IV), and the Nuclear

Hydrogen Initiative evolved into the GNEP program with the objective of developing an

Advanced Burner Reactor (ABR) technology in order to transmutate spent nuclear fuel and

develop separations technologies in order to maximize the amount of energy extracted from

nuclear fuel, minimize nuclear waste, and mitigate concerns regarding proliferation of weapons

material from commercial nuclear fuel.18









Global Nuclear Energy Partnership

The main goal of GNEP is to advance nuclear power technologies and enhance the nation's

nuclear energy capacity while effectively addressing the challenge of nuclear waste disposal and

decreasing the risk of nuclear weapons proliferation and.18 As part of President George W.

Bush's Advanced Energy Initiative, GNEP seeks to develop worldwide consensus on enabling

expanded use of nuclear power to meet the growing energy demand while ensuring reduced

dependence on carbon based energy producing systems.18 The GNEP campaign recognizes that

a plentiful, reliable supply of energy is paramount to sustained economic growth and prosperity,

and that nuclear power is the only proven technology that can provide abundant supplies of

electricity reliably and without air pollution or emissions of greenhouse gasses. The GNEP

campaign has further recognized that the use of this economical, carbon-free technology will

reduce the nation's dependence on environment unfriendly energy producing technologies that

increase greenhouse gas emissions.18

GNEP addresses two key issues: implementing sensitive nuclear technologies in a way

that protects global security and determining proper methods of disposing of nuclear waste

safely. The proposed new nuclear energy infrastructure would implement a closed nuclear fuel

cycle that enhances energy security, while promoting non-proliferation and recycling nuclear

fuel to reduce the nation's nuclear waste. This closed fuel cycle requires the development of

technologies that enable consumption of long-lived radioactive waste, and recycling of useful

fissile isotopes.9'19

The following include the major objectives of the GNEP program:

1. New proliferation-resistant technologies to recycle spent nuclear fuel will be used
so that more energy may be recovered and total nuclear waste is reduced.

2. The latest and advanced fuel cycle strategies will be employed in order to reduce
the risk of nuclear proliferation worldwide.









3. The growth of prosperity and sustainable development around the world will be
encouraged by meeting energy capacity demands through the use of nuclear energy.

4. The nation's dependence on fossil fuels will be reduced encouraging the continual
improvement of the environment.

These objectives will be achieved by the following strategies:

1. A new generation of nuclear power plants will be developed in the United States
that employ advanced fuel burning strategies in order to minimize actinide wastes.

2. An integrated fuel separations technology and recycling capability will be designed
and deployed.

3. An aggressive strategy to manage spent nuclear fuel and nuclear waste in the
United States will be determined including permanent geologic storage at Yucca
Mountain.

4. A reliable fuel services program will be engineered in which participating nations
with secure nuclear technologies would be required to supply and receive fresh and
spent nuclear fuel.

5. Cost effective efficient power reactors will be developed and deployed.

6. Nuclear safeguards to enhance the proliferation-resistance and safety of expanded
nuclear power will be improved.

In order to achieve the GNEP objectives by implementing the mentioned strategies, a

symbiotic fuel cycle strategy must be implemented that utilizes both LWR and fast reactor (FR)

technology to maximize the amount of energy extraction from transuranics while minimizing the

amount of overall waste buildup. Spent nuclear fuel contains a plethora new actinides due to the

transmutation that occurs during the burning of the fuel. These actinides are sources of both

long-term radiological hazards, because they dominate the long term ingestion radiotoxicity of

the waste, and proliferation hazards, because they are fissionable nuclides.19 The long lived

actinides tend to have much larger half-lives than typical fission products. Therefore the task of

actinide burning is of utmost importance to addressing the long term radiological waste problem

and mitigating proliferation concern. To bum actinides effectively, the actinides must be









separated from other isotopes that compete for reactions.20, 21,22 The GNEP campaign is tasked

with developing a separations facility to exact burnable actinides from spent nuclear fuel and

separate these actinides from short-lived fission products. The short-lived fission products may

be easily stored in a geological repository as these isotopes will decay away to stability in a

reasonable amount of time.

Advanced Burner Reactor Campaign

The separated actinides must then be recycled and further incinerated in either an

accelerator or reactor. Many types of scenarios have been studied that examine this portioned

approach to actinide incineration; however, the transmutation of waste by accelerator option is

limited by cost and wasted use of extractable actinide energy. Two basic types of reactors have

been considered for further actinide burning. The extracted actinides could be fabricated into

mixed oxide (MOX) fuel bundles and placed back into a commercial LWR. The advantage to

burning in a commercial thermal spectrum reactor is that commercial reactors are already in

operation, and therefore transitioning operating strategies for accepting MOX fuel would seem to

be more viable solution than developing an entire new reactor for transmutation.12, 15, 19,21

Recycling of plutonium in thermal reactors would also help to reduce total plutonium and seek to

mitigate proliferation concern. Developing operating strategies to account for reduced poison

worth, from the higher thermal absorption cross section of Pu-239 as compared to the uranium

isotopes, and faster reactor response, from the reduced delayed neutron fraction of Pu-239, may

be complicated; however, the solutions to these strategies may be engineered and therefore does

not limit deployment.1

The limit of the effectiveness of multi-recycling of actinides in commercial LWRs is

dictated by the amount of higher actinides generated as a function of burnup. In a thermal

reactor spectrum, the higher actinides buildup due to lower fission rates (i.e. curium) in those









isotopes and higher capture rates in the thermal spectrum. The fission-to-capture ratio in thermal

reactors is not as high as in fast reactors and therefore higher actinides are generated and not

fissioned as readily as compared to fast reactors. Therefore more Am, Cm, and Cf are generated

as function of burnup in thermal reactors as compared to fast reactors.19 22 These isotopes make

spent fuel recycle challenging through increased heat generation and spontaneous neutron

emission, and further result in both criticality and radiation protection difficulties.

The fast reactor spectrum is more favorable for higher actinide transmutation because the

fission to capture ratio is much higher as compared to thermal spectrum reactors.19 Destruction

of long-lived actinides requires excess neutrons. In a fast reactor the difference between neutron

production by fission and losses by capture in structural and absorbing materials or leakage from

the core is -0.4 -0.6 neutrons per fission while in a thermal reactor this surplus is well below at

0.1 neutrons per fission.19 The cross sections in the fast spectrum favor fission rather than

capture; however, the fast spectrum cross sections are smaller in magnitude and therefore a

larger flux is required to achieve the same destruction yield as in thermal reactors. 15 Since the

principal of transmutation consists of recycling actinide wastes in reactor scenarios that favor

fission as opposed to capture, development of a fast reactor technology is integral to the

objectives of the GNEP campaign in order to maximize energy extraction from fuel and

minimize the buildup of long term actinide wastes.

GNEP will develop and demonstrate Advanced Burner Reactors (ABRs) that consume

actinide wastes while extracting their usable energy. The objective of developing these fast

spectrum reactors will be to destroy actinide wastes in commercial spent nuclear fuel from

nuclear power plants, avoiding the need to accommodate material in a geological repository for

hundreds of thousands of years while it decays. The reality of this objective is that these ABRs









will not avoid but reduce the burden of current repository demand. Significant prior United

States investment in fast reactors, such as Fast Flux Test Facility and the Experimental Breeder

Reactor, provides a valuable technology base for supporting future development of ABRs.

Though the development of ABRs requires following the isotopes in exact detail in order

understand true transmutation rates of actinide wastes, the initial design process will require

thousands of core calculations that integrate neutronic, structural, and thermal hydraulic concerns

in order to license the facility. These calculations include varied core operating and shutdown

strategies in order to demonstrate that the core may operate at a certain power, for a certain

amount of time, and shutdown safely in an accident scenario. Therefore calculations of actinide

transmutation rates need not be followed in tremendous detail in order to license the facility so as

to hamper the ability to complete the plethora of integrated system calculations. In fact,

calculations that capture more detail than what is necessary to license the facility will simply

hinder the completion of the design and possibly the building of the facility. If a 2-dimensional

diffusion theory code that runs in 5 seconds achieves an adequate result, why waste time running

a high fidelity calculation that takes a week (the fallibility in this statement may reside in the fact

that without experimental benchmarks to validate the less accurate solution method, one cannot

be certain that the approximations applied to that solution method are indeed correct). Therefore

high fidelity depletion modeling may not be of interest to the initial design of the ABRs.

However, high fidelity depletion modeling is integral to determining adequate source term

generation for the development of technologies that will be used to segregate detectable signals

into key nuclide signatures to be used for interpreting proliferation concerns.

Reactor Safeguards Campaign

The objective of reactor safeguards is to develop proliferant resistant fuel cycle

technologies and programs the help prevent misuse of civilian nuclear facilities for non-peaceful









purposes. Reactor safeguards responsibilities include accounting for special nuclear materials,

control of technology and the ability to inspect and verify compliance with international

agreements.23 Under the auspices of the International Atomic Energy Agency reactor safeguards

have been an effective deterrent against the spread of nuclear technology and materials. The role

of GNEP safeguards is to provide the opportunity to design modern safeguards directly into the

planning and building of new nuclear energy systems and fuel cycle facilities. .23 The GNEP

Safeguards campaign will also focused on improving international safeguards including the

development of future proliferation-resistant fuel cycle and reactor technologies to be integrated

world wide.23

The development of safeguards technologies involves improving the way in which

diversion of special nuclear material is detected. The source term of the special nuclear material

is the actinide wastes generated in nuclear reactors and the soon to be developed ABRs.

Therefore the safeguards community is concerned with measuring the content of key specific

nuclides within spent fuel leaving the reactor. For the GNEP campaign, the spent nuclear

material will take on many different types of combinations depending on the reactor type that

generated that specific spent fuel.

The isotope content of key specific nuclides within the spent fuel may be determined by

implementing various non-destructive-analysis (NDA) techniques involving the discrimination

of a detectable signal into the components of a specific nuclide's radiation signature.24 The

efficiency of the detection equipment to measure the concentration of a specific nuclide is

therefore dependent on the ability to segregate the detectable signal into the components that

contribute to the specific nuclide's signature and the components that do not contribute.24 Since

the detectable signal may be a conglomeration of many simultaneously contributing radiation









signatures, discrimination of the signal into the key components of a specific nuclide's signature

involves knowledge of all the contributions to the signal. From a modeling standpoint, designing

more efficient detection equipment involves explicitly following the evolutionary buildup of all

isotopes contributing to the detectable signatures. However, in order to accurately account for

the temporal evolution of the nuclide inventory of all contributors to the detectable signal of key

nuclide signatures, a method that more accurately captures the true system physics of the source

term generation becomes paramount.

MCNPX is widely used in the safeguards community for modeling all aspects of reactor

safeguards and is considered the "gold standard" in radiation transport modeling. Historically,

the MCNPX code lacked the capability to model fuel depletion and therefore could not

accurately generate the needed source term for development high fidelity detection equipment.

In the past, this source term was either generated by an externally linked technology or by an

entirely separate code of which both incorporated varied assumption that ultimately affected

calculation accuracy. A technology that calculates a high fidelity time-dependant isotope

inventory is essential for determining a sufficient source term use in a very precise calibration

and design of high fidelity radiation detection equipment. It is vital that this capability be placed

in MCNPX in order to make MCNPX, a tool already implemented by the reactor safeguards and

considered a gold standard by the community, a more complete tool for reactor safeguards

calculations.

Motivations for a Monte Carlo Depletion Tool in MCNPX

Many deterministic linked tools exist that are capable of modeling fuel depletion for

advanced reactor concepts. However, these tools employ a variety of approximations about

system physics in order to achieve a quick result.25-34 For the result to be reliable, costly

experimental benchmarking and thorough investigative analysis is required in order to determine









the limits of the modeling technology. Approximations implemented in the calculation of one

type of system are not necessarily valid for the calculation of a different system.32'34 This issue

leads to the necessity to develop and catalogue an assortment of tools employing a diverse range

of calculation assumptions.35 The selection of an adequate modeling tool therefore relies on the

ability to predetermine which assumptions are adequate for a particular calculation type and then

selecting the design tool employing those assumptions. Therefore, it is possible that false

conclusions of system behavior may be drawn from false hypotheses of system physics when

incorrect approximations are implemented leading to the inability to make adequate design

decisions.

As advanced reactor concepts challenge the accuracy of current modeling technologies, a

higher fidelity depletion calculation, which employs few, if any, assumptions of system physics,

is crucial to properly model the wide variety of advanced reactor concepts. An optimum

modeling tool would not rely upon predetermined knowledge of adequate assumptions that could

result in an incorrect calculation; the tool would eliminate system physics assumptions entirely

and preserve the true physics behavior. A Monte Carlo linked depletion tool is capable of

achieving this goal because true system physics is preserved through the simulation of the actual

complete physical process. This makes the Monte Carlo linked depletion method highly robust

and capable of analyzing a wide variety of calculations with immense confidence of achieving an

accurate result.36

A few Monte-Carlo-linked depletion codes exist including ALEPH, BURNCALC,

MCODE, MC-REBUS, MCWO, MCB, MCMG-BURN, MOCUP, MONTEBURNS, MVP and

RACER.37-49 Most of these codes are externally linked technologies and implement complicated

directory structures to integrate each externally linked code package. This results in increased









memory restrictions that ultimately constrict the maximum size of the calculation model. Some

of these efforts are also not supported, and therefore offer little if any code support for the typical

user. Some of these efforts use multi-group Monte Carlo and are therefore slave to the

calculation assumption of the deterministic methodologies.49 Most of these codes are also

focused on accelerating the calculation by minimizing the amount of isotopes and reactions

tracked in order to minimize computational expense at the cost of detail in the simulation thereby

loosing fidelity that may be of interest to the user. Since most of these codes are externally

linked technologies, in most cases, the codes only work with a minimal amount of functionality

of the codes in which they are linked. In order to truly take advantage of the robustness of the

Monte Carlo linked depletion method, the technology must be self-contained, eliminate excess

top-level processing and calculation size requirements as well as automatically determine and

track parameters important to the depletion process.

MCNPX depletion provides a vital modeling framework, within a well established,

supported reactor safeguards radiation transport code, for high-fidelity depletion calculations to

assist in the development of technologies requiring ample intricate detail in the depletion

solution such as the reactor safeguards calculations required for GNEP. Because this capability

is fully integrated into the functionality of a well established radiation transport package already

utilized by the GNEP safeguards committee, the community now has a tool that meets almost all

the functionality required to meet their current modeling needs.

Objective

The goal of this work is to develop a self contained easy-to-use Monte Carlo linked

depletion modeling framework within a well established radiation transport package that

eliminates calculation size requirements as well as automatically determines and explicitly tracks

all the important parameters for computing an accurate depletion solution. This task involves









internally linking the MCNPX steady state-reaction calculator to the CINDER90 number density

calculator.50, 51 MCNPX will be used to calculate steady-state reaction rates and normalization

parameters while CINDER90 will be used to calculate the time-dependent isotope

buildup/depletion.

The work scope may be bifurcated into two separate processes: code development and

code verification and validation. The following ten code development issues will be addressed:

1. A Monte Carlo linked depletion capability that works within the confines of the MCNPX
code package will be developed, making sure the depletion capability is compatible with
major execution features of MCNPX.

2. The input structure must reside within an MCNPX input deck. The input structure will
be simplified yet include ways of user specification for

a. time steps to report isotope and system data;
b. power lever and percent of full power level to implement at specific time steps;
c. burn materials to be depleted;
d. isotopes to omit;
e. fission products to track;
f. nuclide concentrations to manually change for different time steps;
g. ability to order output according to user specification.


3. The output will contain averaged time-dependent information for

a. neutron multiplication;
b. neutrons per fission;
c. recoverable energy per fission;
d. burnup;
e. source neutrons;
f. isotope concentrations and activity of radioactive isotopes;
g. sum of isotope concentrations and activity information.

4. For individual bum materials the code output will report

a. fission power fractions;
b. burnup;
c. collision rates;
d. isotope buildup and activity.

5. The following mechanisms for isotope tracking will include









a. tracking each material specified by the user;
b. selecting fission products to be tracked based on easy-to-implement predefined
fission product "Tier" sets;
c. following the concentrations of all the possible daughter reactions from isotopes
specified as burn materials;
d. tracking reaction rates of metastable isotopes;
e. manually adjusting bum material concentrations of specific isotopes if necessary.

6. Reaction rates to be computed for use in the depletion process will be determined, and a
method will be developed for normalizing the reaction rates based on given and
calculated system parameters.

7. A methodology will be implemented for addressing the nonlinearity in the reaction rate
utilization in the depletion equation.

8. A method will be engineered for selecting the proper fission yield to be utilized within a
specific computation.

9. A capability will be invoked to burn multiple bum materials where individual burn
material characteristics may be calculated and reported.

10. A technique will be implemented for accurately capturing the average recoverable
capture gamma energy contribution to the true energy per fission event.

Once the methodology is developed the following benchmark studies will be completed:

11. OECD/NEA Bumup Credit Calculational Criticality Benchmark Phase I-B;52

12. H. B. Robinson Infinitely Reflected Pressurized Water Reactor Fuel Assembly
Calculation;53

13. OECD/NEA Bumup Credit Calculational Criticality Benchmark Phase IV-B.54









CHAPTER 2
SURVEY OF STATE OF THE ART APPROACHES TO DEPLETION ANALYSIS

During the operation of a nuclear system, the nuclide concentration will change as isotopes

consume radiation (neutrons, protons, alphas, betas, etc.) and undergo various nuclear reactions

[(n, fission), (n, 2n), (n,p), (n,a), (n,3), etc.].2 The temporal change in isotope concentrations

results in further changes in scatter, capture, and fission events as isotopes either transmutate or

fission into new nuclides possessing different probabilities for these reactions. For example, as

U-235 depletes as a result of the fission process, fission products are generated. Because these

newly created fission products posses capture reaction probabilities that rival U-235 for neutron

capture, the fission products, combined with the depletion of U-235, reduce the propensity of the

system to fission and maintain a steady-state chain reaction. In order to maintain a self

sustaining steady-state chain reaction, more fuel than is necessary in order to maintain a steady-

state chain reaction must be loaded. The introduction of this excess fuel increases the net

multiplication capability of the system. To maintain the net neutron multiplication of the system

at a steady state, poison material is introduced to absorb the excess neutrons generated from the

excess fuel loading. As the fuel depletes and produces absorbing fission products, the

controlling material is extracted in order to maintain the delicate reactivity balance needed in

order to maintain a self sustaining chain reaction.9 Therefore it is crucial to monitor the temporal

isotopic concentration in the nuclear system because changes in this composition affect the

operating strategies that are necessary to maintain a steady-state self-sustaining reaction.

The operating strategies of a nuclear system are affected by the ability to meet limiting

values of key operational parameters. These parameters include: flux, power distribution,

reactivity, shutdown margin, etc. The limiting values of these parameters are set such that the

system be able to operate safely for a prescribed time and meet a prescribed objective. Since









these system parameters therefore limit operational characteristics of the system, it is crucial to

compute these values as each value temporally evolves drastically enough to warrant changes in

the system operation. The study of the change of these operational parameters with the time-

dependent production/ depletion of nuclei is known as depletion analysis.2

The Depletion Equation

Solving for the time-dependent change of a specified nuclide involves accounting for all

processes that lead to either the creation or destruction of that nuclide. The temporal nuclide

density may be described by the Bateman equations.55 Equation 2-1 states a simplification of a

form of the Bateman equations so as to describe the depletion of an individual nuclide:

dN, (F, t)
= -Nm (r, t), + 7m + Nk,(r,t)ykm (2-1)
dt kam

dN
dNm time dependant change in nuclide m
dt
Nm (t)Jm = destruction of nuclide m
V Nk (t)Yk-m = creation of nuclide m via other nuclides in the system
kam
Y = production of nuclide m via an external source

Solving for the time-dependent change of a specified nuclide also involves knowledge of

the time-dependent change of each contributing nuclide. Therefore the solution of each nuclides

time-dependent change involves solving a set of coupled differential equations.55 Though

solving a set of coupled linear differential equations is a straightforward mathematical process,

the coefficients in the coupled depletion equations are nonlinear therefore making the equations


unsolvable without approximation. Expanding the destruction coefficient, im, and the creation


coefficient, Yk-m clarifies the nature of the nonlinearity of equation 2-1:

im = Am + f cm, (E)(r,E,t)dE (2-2a)
r










7k->, = Lkm+k I JYkmr(E)+k,r(E)()(r,E,t)dE (2-2b)
mnk mnk r

A = destruction of nuclide m by radioactive decay

fZ Jm,r (E)((r, E, t)dE = destruction of nuclide m by transmutation reaction
r
Lkmk = creation of nuclide m by some isotope radioactively decaying to isotope m
mnk
YIkmr (E)akr (E)D(r, E, t)dE = creation of isotope m by some isotope
m~k r
transmutating to isotope m via a transmutation reaction, r.

The analysis of the temporal nuclide inventory is complicated by the fact that the temporal

nuclide inventory solution is dependent upon the time-dependent flux. Unfortunately, the time-

dependent flux is also dependent upon the time-dependent nuclide inventory. To make equation

2-1 linear and solvable, approximations must be made regarding the coefficients. The first

approximation assumes that the coefficients be made constant over a time step of interest, tacitly

implying that the time-dependent nuclide solution is separable from the spatial transmuation rate

calculation. The second approximation assumes that spatial transmutation rates may be

calculated using discrete time steps. The transmutation rates for the temporal nuclide inventory

calculation are first computed implementing a steady state reaction rate solver. Then those

transmutation rates are assumed constant and applied to the temporal nuclide inventory

calculation in equation 2-1.

The validity of the approximations is contingent on the fact that the magnitude of the

transmutation rate implies that large time durations are needed in order to initiate significant

buildup and depletion of nuclides so as to appreciably alter the spatial transmutation rate

calculation.2, 55 The definition of "large time durations" is relative and dependent upon the

builup of highly absorbing isotopes during a time duration that lead to a significant change in the

transmutation rate. For example, at the initiation of reactor startup Xe-135 and Sm-149 both









contain large absorption cross sections and builup to equilibrium values early in the reactor

operating cycle. In reactors containing burnable poisons incorporated into the fuel assembly, the

burnout of these highly absorbing isotopes during the fuel cycle also offers a significant change

in transmutation rate as the fuel becomes more reactive in the absence of poisons. Taking time

steps that exceed the time to equilibrium buildup of these isotopes affects the further trajectory of

the depletion solution because the equilibrium buildup of these highly absorbing isotopes does

greatly affect the calculation of spatial transmutation rates. This affects the accuracy of the

number densities generated from equation 2-1 for that time step which further affects the

calculation of transmutation rates for the next time step. As long as the transmutation rates do

not significantly change during a time-step, assuming a constant transmutation rate for a given

time-step is valid.56

Zero-Dimension Depletion Calculation

The depletion equation requires knowledge of reaction rates, decay coefficients and isotope

concentrations at a specific time-step in order to solve for the isotope concentrations at the

subsequent time-steps. The accuracy in the depletion calculation is highly dependent upon the

appropriateness of the coefficients implemented at given time steps in the solution method, and

therefore the amount of spatial detail followed in the depletion solution greatly affects

calculation accuracy. The depletion equations could be solved on a normalized basis, in which

power is specified in a per weight basis and isotopes concentrations are determined on a per

weight basis, if the collision rates utilized in equation 2-1 truly emulate the actual collision rates

for a given system. Since this type of calculation is solved over a normalized basis, the solution

of the coupled depletion equation does not require the coefficients to exhibit spatial dependence.

The accuracy and applicability of this solution method resides in the validity of the implemented

coefficients for a given analyzed system.57









The ORIGEN and CINDER codes, by themselves, are a family of zero spatial dimension

depletion codes and therefore the codes have no knowledge of the spatial dependence of the

transmutation rates.51'58 Dimensions of the analyzed system are approximated by specifying the

power in terms of a bases unit (i.e. power/weight). The major assumption that ORIGEN and

CINDER make is that the flux is not spatially dependent and may be calculated by equation 2-3:51' 5

6.242*ld8(P) (2-3)
= X2of]3

P = System power [MW]
x = Concentration of the ith fissile nuclide [g atom]
"f = Microscopic probability of fission of the ith fissile nuclide [b]

= Recoverable energy per fission event of the ith fissile nuclide MeV
fission
neutrons
= Total flux ne
L cm s

The calculation of power based on some spatial flux solution is described by equation 2-4:

Power= fI(r)QYfdV (2-4)
V

(.(r) neutrons
() = Spatial flux distribution neutron
Scm s
= Macroscopic probability of fission c1
cm
SdV = Differential volume element cm31


= Average recoverable energy per fission for the whole system MeV
fission

After determination of flux and system power, the solution methods of the two codes differ

in approach. The two most widely used methods for solving the coupled depletion equations

involve a simplification of either the matrix exponential method or the Markov linear chain









method. The Markov linear chain method involves uncoupling the coupled depletion equations

into a set of partial linear transmutation paths, solving for the partial buildup/depletion of each

nuclide within each linear chain, and summing the final buildups of each partial path to

determine the final nuclide buildup (CINDER90 method previously explained).51 This method

takes advantage of minimal memory requirements because the computation is broken up into

smaller memory manageable sets. The matrix exponential method tries to explicitly solve the

depletion equations by casting the solution in terms of a matrix exponential (method used in

ORIGEN and other industry based codes that evolved from the ORIGEN methodology).58'59

The main objective of both techniques is to try and determine, as accurately as possible, the

evolution of the temporal nuclide inventory solution.

Matrix Exponential Method

The underlying concept of the matrix exponential method is that the formulas for solving

normal systems of equations with constant coefficients are identical to the formulas for solving

first order differential equations with constant coefficients.60 Equation 2-1 may be recast into the

following form

d
n (r, t)= Bk (r,t),, (r,t) (2-5)
dt k

where:

rLkmAk +J kmrk,ro(r,t) for m k
Bmk (r, t)= (rr = k (2-6)
-^-m- m,r((,) form = k




This results in a set of first order homogenous ordinary differential equations with constant

coefficients. The solution of this first order differential equation is:60









N(t)= N(to)exp(Bt) (2-7)

where N is a column vector of nuclide densities and B is the matrix of elements calculated of

equation 2-6. The function exp(Bt) is a matrix exponential function, and is therefore calculated

from the following expansion:


exp(Bt) (Bt) (2-8)
m=o0 m!

If the size of the matrix is not too large and not too sparse, the main advantage of this method is

that the coupled depletion equations may be solved explicitly.59 However, the depletion

equations require knowledge of all transmuted actinides and fission products leading to a large

sparse matrix. This large sparse matrix requires an enormous amount memory in order to store

all the necessary computations. Also, the sparse matrix contains both very large values and very

small eigenvalues depending upon the half-lives of each isotope tracked during the depletion

process.59 Certain isotopes possess half-lives of a fraction of a second while others posses half-

lives of a billion years.

The issue of having a large sparse matrix with widely separated eigenvalues leads to

problems in the ability to compute the solution and therefore approximations are made in order

to generate a stable solvable matrix.60 Potentially, the size of the matrix can be as large as n2

where n is the amount of nuclides followed in the set of equations (1700 nuclides for

ORIGEN2.2). However, because the matrix is sparse, a recursive relationship may be generated

for formulating the matrix exponential function thereby eliminating the massive memory cost

associated with storing all the matrix terms of the matrix exponential function. The relationship

in ORIGEN2.2 for one nuclide is as follows:


x, (t)= c (2-9)
n=o









where c" is generated by use of the following recursive relationship


c, = x, (0) (2-10a)


n+1 t N
C ac (2-10b)
n+lJ

In this solution method a, is an element in the matrix that is the first-order rate constant for the

creation of isotope i from isotope j. This algorithm for solving for the temporal nuclide

inventory only requires the storage of a single vector, c", greatly reducing memory cost.

The main concern computing the summation in equation 2-9 is that it is necessary to

ensure precision is not lost as a result of addition and subtraction of nearly equal large numbers.

The time step for the calculation may be reduced so as to limit the size of the matrix coefficients;

then the resulting matrix may be multiplied by a factor in order to obtain the desired time step.60

Scaling the matrix calculation would be impractical for a large number of nuclides as many of

the nuclides contain very short half-lives. Therefore in order to fix the mentioned numerical

issue, ORIGEN2.2 decays analytically all short lived nuclides. A short lived nuclide is defined

such that the matrix coefficients for this nuclide, resulting from the large amount of decay, are so

large as to result in numerical issues such that the norm of the matrix exceeds the predetermined

value of the word length of the computer used for the calculations. This simplification increases

the stability of the solution method by eliminating numerical issues. This assumption is,

however, only acceptable as long as the isotope does not contain a significant long lived

precursor with a large interaction cross section.60 For short lived daughter isotopes from long

lived parent isotopes, the Gauss-Siedel iterative technique is used to accelerate the calculation of

these nuclides.









Ultimately, the amounts of isotopes that are tracked explicitly are limited by the maximum

memory requirements of the computer utilized to analyze the system. Therefore implementation

of the matrix exponential method actually results in only applying a subset of nuclides to the

explicit matrix exponential computation, and then employing approximations to follow the rest.

The user must then be careful to select a subset of nuclides that best achieves the most correct

characterization of the true interaction behavior of the analyzed system, and then the user must

implement an adequate approximation to follow the rest of the nuclides. However, further

numerical limitation still exists as a result of the amount available precesion resulting from the

applied numerical technique. If the user wishes to therefore try to solve the Bateman equations

explicitly, the matrix exponential method may be the method of choice; however, if the user

determines that a complete set of fission products and actinides is required in order to achieve a

reliable answer then the linear Markov chain method is more capable of storing the

concentrations of more nuclide sets without the same type of numerical cost.

The CINDER90 Linear Markov Chain Method

The CINDER90 code is a FORTRAN program coupled with an extensive data library used

to calculated nuclide inventory.61 The origin of the CINDER90 code may be traced back to the

original development of the CINDER code in 1960 at the Bettis Atomic Power Laboratory

(BAPL) in support of thermal reactor simulations. CINDER was utilized by both BAPL and Los

Alamos National Lab (LANL) for the design of many reactor-fuel investigations.61 Many later

versions of the CINDER code were developed with improvements in data evaluations, fission

yield sets, and enhanced decay information. Initially, CINDER required preset formation of a

consistent set of linear chains to describe the creation and transmutation paths of all investigated

cases.61 All versions of CINDER utilized decay and energy integrated reaction rate probabilities

along with fission yield information to calculate the temporal nuclide buildup and depletion.









Versions of CINDER preceding CINDER90 were limited to reactor fuel calculations involving

fission products (66
The development effort of CINDER90 arose from a need by the LANL accelerator

community to have a more complete calculator for temporal nuclide inventories.61 The library of

data in CINDER90 was developed to then include isotope decay and interaction probability data

for 3456 isotopes including, -30 fission yield sets, and yield data for 1325 fission products.5

CINDER90 is also capable of following as many reaction types as data exists within the

CINDER90 library file. The CINDER90 computation process involves utilizing linear

Markovian chains to determine the time dependent nuclide densities, solving for the independent

contributions to nuclide densities in each of a number of linear nuclide chains. Instead of

requiring preexisting transmutation path information, in CINDER90 each transmutation of each

nuclide defined by available nuclear data is followed until the path reaches a nuclide that is

deemed to be insignificant. CINDER90 implements an equation similar to equation 2-1 in order

to calculate the time-dependent nuclide densities:

dN,
dN = -N (t) + r, + Nk (t)k, (2-11)
dt knm

where all terms are the same as equation 2-1 except that Ym is included to represent the average

feed rate. Like equation 2-1, equation 2-11 relies on the assumption that the transmutation

probabilities are to remain constant for the for the time interval in which a solution is desired.61

The set differential equations utilized to solve for temporal nuclide buildup/depletion is

coupled since each equation contains time-dependent isotope density information from other

nuclides. In CINDER90, this set of coupled equations is reduced to a set of linear differential

equations using the Markov method. Linear chains are created for each isotope transmutation









path. The solutions of each linear chain determines a partial nuclide density, N,. Each

calculated partial nuclide density, N,, computed from a linear chain, is then summed to obtain

the total nuclide inventory of nuclide, N,. The differential equation governing the computation

of N(t) is therefore only coupled to any preceding elements in the sequence leading to the ith

element. For the preceding, (i-1) element, all parameters are assumed known. The entire

computation is then computed from the simplified equation

dN -
d = + N,_ (t)y,- N, (t)/, (2-12)
dt

where quantities are now indexed by the order in which they appear within a given sequence and

y,, is the transmutation probability of forming nuclide element N,. The general solution form

of a linear sequence of nuclides coupled by any sequence of absorption or decay, first derived

and implemented by the CINDER code, was:


n n- 1 n pt n -it
Nn(t)= ---k n-- + NZ n (2-13)
{Y1k m n {Jf i 8 8 mI-'8' '8
l=m i=m,)j i=m,#j

This algorithm therefore depended upon predetermined knowledge of the transmutation path in

order to properly include every No encountered within a specific path. This method further

suffered from large computational cost because the entire transmutation path was to be followed

even if the probability was low of incurring further nuclide production of a specific isotope

within a chain.5

For computational speed enhancement, simplification, and reduction of memory

requirements, the CINDER90 code employs a methodology that does not require a predefined set

of nuclides employed in a transmutation path. The methodology starts at a base set of nuclides









and follows each path independently. Tests of significance at each step of the transmutation path

are set. If the test of significance is failed, then the transmutation path is stopped and further

computation of extraneous nuclides is not incurred. This lead to the following simplifications

No = 0,m 1 (2-14)

Y, =0,m 1 (2-15)

ultimately resulting in the following solution for a given linear sequence:61


n-l 1 efly
N -1 +N (2-16)
= z=n, Q z=1, ky


The test for significance in CINDER90 involves calculating the passby quantity, P, (t).

The passby of a nuclide is the time-integrated transmutation of that nuclide for a specified time

interval:


(t) = IN, (t)/,dt (2-17)
0

The passby therefore may be thought of as the progeny of nuclide n resulting from transmutation

or decay.61

The accuracy of CINDER90 is dependent upon the data that is used in the solution of (Eq.

2-11). If the code is used by itself to calculate nuclide inventory, the user would be relying

heavily upon the inherent multi-group cross sections, utilized to determine collapsed energy

integrated reaction rates, in order to achieve a solution. Since CINDER90 is a zero-dimensional

code, similar approximations, as implemented by ORIGEN, would have to be employed in order

to determine a flux to be used in the computation of reaction rates. This flux would not

incorporate the spatial behavior of the system and therefore lead to inaccurate interaction rate









calculations. This would further propagate into inaccurate temporal nuclide inventory

computations. To resolve this issue, CINDER90 must be linked to a steady state reaction rate

calculator in order to capture the spatial resolution and possible spectral change of the time-

dependent reaction rate behavior.

Necessity for a Linked Approach

In the zero spatial dimension calculation, the flux is no longer spatially dependant and

events such as neutron leakage, scatter and absorption are not accurately depicted. The physics

of assuming spatial independence of the reaction rates is highly false because the energy

dependence of these reactions makes capture and scatter events heavily dependent upon

geometric configurations. For example, BWR reactor fuel pins are loaded at various

enrichments in combination with gadolinium burnable poisons in order to try and achieve a flat

power distribution across the bundle.62 These bundles are then strategically positioned within the

reactor core in order to further maintain a flat power distribution across the reactor core. A flat

power distribution is preferential because if the peak to average power in the bundle is too high

then fuel centerline melt, creep, and fission gas pressure constraints limit the average power the

bundle can be operated, therefore increasing the amount of needed loaded fuel bundles per cycle

and ultimately reducing fuel economy. The ability to meet this flat power distribution demand is

limited by the reactor asymmetries that force non-periodic boundary conditions on the flux

solution. For instance, neutrons, generated in fuel bundles at the core periphery, exhibit more

moderation because of the larger interaction with the surrounding water reflector. This

asymmetric interaction is further complicated by the fact that the density of the moderator

decreases as the moderating fluid heats up and boils as the fluid passes through the reactor core.

This reduction in moderation leads to less fission and more capture at the beginning of the

operating cycle. The boundary condition for the flux solution is therefore set by the asymmetric









properties of the external reflecting fluid and these imposed boundary conditions lead to a flux

gradient across the reactor core.

There exists further complicating asymmetries within the bundle itself. Burnable poisons,

large cruciform control blades, water holes, and structural guide tubes within the fuel assembly

also further contribute to flux asymmetry within the fuel assembly.62 The asymmetry in the flux

solution further propagates during fuel burnup as capture rates and thermalizations are spatially

affected as a function of the asymmetry in isotope concentrations resulting from the asymmetry

in prior reaction rates resulting from the asymmetries in the fuel assembly and core structure.

Therefore implementing a depletion model for a reactor system without accounting for spatial

dependence will lead to discrepancies in the solution as these types of spatial asymmetries in the

flux solution lead to further propagation of asymmetry in isotope buildup and depletion.

As isotopes consume radiation, undergo nuclear reaction, and transmutate into new

isotopes, the neutron energy spectrum will change due to how the neutrons will interact with the

newly created transmuted isotopes.2 Simply assuming a single energy-integrated cross section

set for an entire depletion calculation may not be a valid assumption for systems that undergo

significant spectral changes. For example, in a freshly fueled LWR, initially the majority of

energy is created by fission ofU-235 atoms. As the core burns over time many neutrons are

parasitically absorbed by U-238 leading to production ofU-239. U-239 quickly decays by beta

emission to Np-239 and then to Pu-239 which is highly fissile. At the end of the reactor core

life, an immense amount of Pu-239 has been created, absorbed neutrons, fissioned, and

contributed to system power. The average thermal absorption cross section for Pu-239 is 1011.3

b which is much higher as compared to U-235 at 680.8 b.7 The average thermal fission cross

section for Pu-239 is 742.5 b which is also much higher as compared to U-235 at 582.2b.7









capture cross section is defined as all absorption reactions not leading to fission, and therefore a

thermal capture-to-fission ratio may be defined for a specified isotope by equation 2-18:


af- (2-18)
-f

Changes in the capture-fission-ratio over time change the amount of neutrons that will be

parasitically absorbed as well as created by the fission process. These factors ultimately alter

the neutron energy spectrum and thus lead to time-dependant changes in the transmutation rates.

The process is also highly geometry dependent because the capture to fission ratios in certain

sections of a given system will change at different rates due to the energy flux witnesses in a

specific region. Inaccurate capture of these changes ultimately affects the accuracy of the

calculation; therefore it is vital for transmutation rates to be recalculated when isotope

concentration changes result in significant changes in the energy flux distribution.


Collision Densities, Eigenvalue,
Flux Normalization Parameters











Isotope Concentrations



Figure 2-1. Linked Depletion Process.









Since considerable changes in the isotope concentration are required in order to

significantly alter the neutron energy spectrum, equation 2-1 may be approximated as separable

in space and time. The equation may then be discretized in several time steps where a steady-

state recalculation of reaction rates is completed at time intervals where it is assumed that

neutron energy spectrum will be altered significantly enough to alter the trajectory of equation 2-

1. The procedure for this solution method is outlined in Figure 2-1. The reaction rates utilized in

the capture terms of equation 2-1, system eigenvalue, and flux normalization terms are first

calculated utilizing a steady-state reaction rate calculator. Those capture reactions are then

assumed constant over a specified time step and implemented in a time-dependent number

density calculation. The time-dependant number density calculation then calculates a new set of

number densities to be utilized in the subsequent steady-state reaction rate calculation. The

process repeats itself until the final step.

Methods of Steady State Reaction Rate Calculation

Steady state reaction rates may be calculated by either explicitly solving a form of the

Linear Boltzmann Equation (LBE) or determining the average interaction rate behavior of a

sample distribution by simulating the actual physical process using the Monte Carlo method.63' 64,

65 Solving the LBE involves deterministically calculating a continuous spatial reaction rate

solution by employing a variety of discretizations and approximations. Accuracy of the

calculation is highly dependant upon the validity of the discretization and approximation

techniques utilized to represent the system physics.64 Assumptions must be benchmarked against

viable experimental data in order to ensure calculation accuracy. The Monte Carlo method,

however, simulates the actual physical process through the random sampling of probabilistic

events. The method minimizes approximations applied to system coefficients and relies solely

on knowledge of the probability of particle interaction.









Due to the random sampling process, the reaction rates calculated are stochastic and

therefore the solutions are probabilistic, in contrast to deterministic, thus having an associated

statistical error.63 64,65 Reducing the stochastic error involves modeling larger sample

distributions to better account for the average behavior of particle interactions.66 The need for

larger sample sizes to secure adequate calculation precision generally causes the Monte Carlo

method to execute slower than a typical deterministic calculation. In certain modeling scenarios,

an investigator may wish to analyze a plethora of perturbations to a well understood

benchmarked system of which the approximations applied within the deterministic solution

method are valid or within the solution tolerance for analyzing the system. The deterministic

method is more suited for this type of calculation because the gain in computational performance

outweighs the accuracy penalty of which may be within the bounds of the solution tolerance. In

other modeling scenarios, an investigator may wish to analyze a system that is not well

understood and/or not benchmarked. For this type of calculation, the investigator may wish to

compromise computational performance at the expense of enhancing solution accuracy. At the

cost of computational performance, the lack of coefficient approximations, exactness in 3-

dimensional modeling, and explicitness in angular tracking gives the Monte Carlo method a

distinct advantage for this type of modeling scenario. The crux of each modeling technique will

be detailed in the subsequent sections.

Deterministic Approach to Reaction Rate Calculations

The deterministic approach to reaction rate calculations involves solving the LBE by using

a variety of assumptions to achieve a quick, reliable, spatially continuous reaction rate and

neutron multiplication solution. The steady-state LBE is defined as equation 2-19:









Q V ,(r, Q, E) + a, (F, E)VI(F, Q, E) =
I d'J dE'cr, (F, Q'- Q, E'-> E)yj(F, Q',E') + (2-19)

1 x(E) dE'v(E')f (F, E)d' (F, E')
k 4r

Generally, this equation is discretized in energy, space, and angle, or an approximation for

these dimensions is applied, in order to cast the equation in a deterministic solvable form.63 64 65

The energy variable is discretized into few defined energy groups ultimately segregating

equation 2-19 into a set of coupled differential equations, where the coupling of energy groups

results from the out-group scatter and fission terms:

Q V V,g (F, Q) + (F), (F Q)=
G I I G (2-20)
f JdQ gg', (,F' -Q)g (F, Q') + va,, (F)JdQ' Vg' (F, (-')
g'=1 g'=1



The angular dependence of equation 2-20 is accounted for by either discretizing the angular

dependence into few angular quadratures, as in the discrete ordinates method, or by assuming

isotropic scatter behavior and linear dependence of the angular flux, as in diffusion theory. 63,64,
65, 67 After angular dependence is accounted for, the examined system geometry must be meshed

into a fine grid or cast into a solvable ray tracing system. Reaction rate distributions are

calculated within each mesh or along each ray by using assuming homogenized parameters

within the mesh or along the ray and implementing the particle fluxes and currents of the

neighboring meshes or rays as the boundary conditions for that specific mesh or ray solution. 63,
64, 65

The mesh shapes or rays should be chosen so that a problem's geometry may be modeled

explicitly, and the homogenizations within each of the shapes or rays do not upset the fidelity of

the actual calculation. The ideal mesh or ray scheme would include thousands of tiny meshes or









rays so that connect in highly irregular geometries in order to explicitly model a complicated

geometry. However, thousands of meshes would require thousands of angular flux at many

angular quadratures and many energy groups which would ultimately exceed the memory of any

computer. Irregular meshing schemes are available such as in the code Atilla; however, irregular

shaped meshing schemes lead to a highly memory intensive bank of complicated algorithms

utilized to find a flux solution for each irregular mesh scheme.68 Assumptions must be employed

on acceptable mesh size and shape so as not to violate the allotted computational memory

restrictions. Therefore modeling of highly irregular complicated geometries must be

approximated by simpler less precise models that may possibly affect the accuracy of the

solution method. Ray tracing schemes such as that implemented in the method of characteristics

(MOC) are capable of better accounting for complicated geometries by setting up a scheme that

solves a solution at the explicit boundaries of a given geometry. The MOC method has been

successfully applied for modeling complicated geometries.69 However, the MOC method, like

the discrete ordinates method, relies on approximation schemes to develop proper coefficients (

i.e. interaction cross sections) in order to solve the deterministic form of the transport equation.

Within the Evaluated Nuclear Data Files (ENDF), exist infinite-dilute, energy dependant

reaction rate probability, cross section, data.70 This data functions as the coefficients for the

transport equation. The data itself are stated at many energy points and therefore is not exactly

"continuous" in energy, however since the magnitude of data points existing for a specific

reaction is considered in fine enough fidelity to assume a continuous distribution, the data is

deemed continuous. The ENDF data therefore represents a continuous function for each reaction

type that is dependent upon impingent energy, and used as the coefficients in order to solve the

transport equation. Parameters also exist for reconstructing complicated sections of the









coefficient function such as the resonances resulting from the eigenstates of the nucleus. The

data are also stated to be infinitely dilute because it is assumed that reaction probabilities do not

incur energy and spatial self-shielding. Figure 2-2 shows two separate distributions of flux in the

vicinity of a cross section resonance. As compared to a 1/E distribution, the actual flux, Q(E),

will be depressed in the neighborhood of a cross section resonance peak. This effect is termed

energy self-shielding since the large interaction probability of the resonance shields the isotope's

nuclei from neutrons around the energy EO leading to a flux depression. If the nuclei of interest

is then further contained within a lump of similar nuclei and placed in the pathway of a beam of

neutrons with energies in the vicinity of the resonance energy of those nuclei, the majority of

neutrons will only interact with nuclei at the surface of the lump. This effect is termed spatial

self shielding because the inner nuclei of that lump are spatially shielded by the nuclei at the

surface of the lump.2 In implementing a deterministic solution approach the reaction rate

coefficients must be properly adjusted to adequately account for these types of self-shielding in

order to properly model the true system physics.






S1/F





/ 4(E)




Eo E


Figure 2-2. Flux depression in the neighborhood of a resonance at E0.









Casting equation 2-19 into the multi-group form requires that multi-group coefficients be

generated. The coefficients are generated based on the premise that the multi-group coefficients

be a homogenized term that best represents the average behavior within a specified energy

group. To preserve true system physics, the energy group averaged interaction rates are

calculated dividing the energy integrated reaction rate, within a specific energy band, by the

energy integrated flux within that same energy band as shown in equation 2-21:

Eg-1
f c(r, E)O(r, E)dE

ca(r), = E- (2-21)
JD(r,E)dE


The concern with this approach is that the flux, the variable that is to be to calculated in equation

2-20, is used to calculate the cross section which is then utilized to calculate the flux in equation

2-20 making the process nonlinear. To make the process linear, an approximation is made in

which first a homogenous, spatially-infinite, 1-D dimensional calculation is completed with the

continuous energy ENDF cross sections. The flux solution from that calculation is then used to

generate collapsed group energy self-shielded cross sections for a further multi-group

calculation.63'64,65

Deciding which energy bands to use to compute the group average cross sections will

highly affect the ultimate outcome of the further calculations that implement those cross sections.

Ideally, a proper group structure, robust enough to be used for a wide class of calculations,

would account for every major resonance of every single nuclide. Figure 2-3 displays the

absorption cross sections for a few nuclides found in typical reactor calculations. Notice that

each isotope contains many different resonances therefore requiring a large amount of groups to

capture the true interaction rate behavior of the displayed isotopes. If the rest of the 389 isotopes












































I .J I


containing ENDF data are figured into the calculation, one may then see that an immense amount


of energy groups is needed in order to model each resonance. Because the size of the group


structure is one of the many parameters that dictate the needed memory allocation to solve a


specific problem, in practice, the ideal approach is not viable because it is possible that


thousands of energy groups would be necessary in order accurately capture the resonances of


each available isotope. Therefore group structures are chosen that utilize as few groups as


possible, in order to account for as many resonances as possible, to make the group structure


small enough to be practical for an actual calculation.


I'"" I" 1111 I" 11111 I 111111 I"" 1111111 I 1111 I' 11111 I I' 1111 I" 11111 I1111


Al


-- Fred mcnp/Cm-244:27 Cm-244 T=300 K from ENDF/B-VI.O mat9637

-- green mcnp/Pu-239:27 Pu-239 T=300 KfromENDF/B-6.2 mat9437

blue mcnp/U-235:27 U-235T=300K fom ENDF/B-6.2mat9228

purple J mcnp/Nd-143:27 Nd-143 T=300 K fom ENDF/B-6 mat6028

11111111 I 11111111 111111111 111111111 111111111 111111111 111111111 11111111 ...1 I .11111111 ..111111111 -I


109 8 107 106 105 104 103 102 -i1 i1 1
Energy (MeV)


Figure 2-3. Flux depression in the neighborhood of a resonance at Eo.


Accurate group structures may be generated for classes of problems. For example, the


nuclear power industry contains accurate multi-group cross section libraries for examining


-Q
104


10


102


10

ol 0
is
LlB0
0
Vo


W 1
30
10






164


155


"""" """" """"' """"' """"' """"' """"' """"' """"' """"'


=-


=-







1


i









reactors containing uranium-oxide fuel, with zircaloy clad, surrounded by water, and possibly

controlled by born-carbide or hafnium.25 26 30 32 However, those libraries are not applicable for

examining exotic isotope combinations such as uranium-nitride fueled, rhenium cladded and

niobium-carbide coated space reactor designs.71 72 Furthermore, the nuclear industry is only

successful in determining these libraries after years of comparing and tuning cross sections to

operational data.

The implementer of a deterministic method is constantly burdened by the choice of

appropriate approximations in order to achieve an accurate result. If chosen correctly the

implementer will achieve a quick, reliable, non-stochastic, spatially continuous reaction rate and

neutron multiplication solution that runs much faster than a Monte Carlo. However, the burden

of these choices and the knowledge to make proper determinations of group structure and self-

shielding accountability may be out of the experience base of the implementer or may not be

possible with current memory restrictions, in which case a more robust tool exhibiting fewer

approximations of system physics should be implemented.

Monte Carlo Approach to Reaction Rate Calculations

The Monte Carlo method uses probability theory to model a system stochastically by

simulating the actual physical process through the random sampling of events. Like

deterministic methods the Monte Carlo method seeks to solve the transport equation to determine

the average behavior of the neutron population within a system; however, unlike deterministic

methods the Monte Carlo method solves the transport equation in a probabilistic approach. The

Monte Carlo method does not require clever simplifications and approximations to determine

transport coefficients and model geometry in order to compute solutions of neutron behavior. In

certain circumstances, simplifications and approximations employed by deterministic methods

alter a calculation so significantly that accuracy of neutron behavioral characteristics is lost.









Since the Monte Carlo method does not rely on these simplifications and approximations of

system physics, the method is highly robust and capable of modeling very complex systems.36 63

Combinatorial geometry is used to build models for simulation within the Monte Carlo

process.50 Geometric shapes are represented by the combinations of basic surface types. The

geometric shapes created from the combination of these surface types are called "cells". These

cells contain homogenized information of isotope concentration and particle interaction

behaviors. Most possible shapes/cells incurred in nuclear system design may be modeled as

combinations of these basic surface types; therefore making it possible to explicitly model any

type of geometry encountered. Particle position is tracked in reference to these cells and

surfaces, and the particle track length and interaction behavior are determined from interaction

characteristics within each of the encountered cells.63'64,65

A basic diagram of the Monte Carlo process is displayed in Figure 2-4. Initially, a particle

of specified energy and direction enters a cell volume. The distance the particle travels within

that cell volume is dependent upon the probability that the particle will have an interaction

within the specified volume some distance into the volume. In the one-dimensional case, the

probability that the particle interacts within a distance between point "a" and point "b" is

determined from the representative probability density function (PDF), f(x), and calculated by

equation 2-22:

b
jf (x)dx = P{a < x < b} (2-22)
a

The probability of an event clearly may not be less than 0. If x may be any real value between oo

and -oo the distribution may be normalized resulting in equation 2-23:


Jf(x)dx= 1 (2-23)
-on










From this equation we can further derive the cumulative probability distribution function (CDF),

F(x) in equation 2-24, which will be utilized to ultimately determine the actual length of particle

travel:


f (x)dx = F(x) = P{x'< x} (2-24)


In order to then simulate the process of particle travel through the random sampling of events we

set the cumulative distribution function equal to a random number, and invert the equation in

order to compute a correlated path length of travel.

Compute distance to next Track enters volume


No
Interaction _0 to
Interaction Go to cell boundary and enter
inside cell next cell
next cell
Yes

Determine interacting isotope


Determine reaction type and
resolve residual parameters



Yes
Particle
killedN? s Determine interacting isotope -
No



Figure 2-4. Flow diagram for the Monte Carlo particle transport process

If the calculation is in terms of mean free paths and 0 < x< oo the PDF is


f(x) = ex (2-25)

and the associated CDF is

F(x) = -e x (2-26)









Therefore, inverting the cumulative probability distribution function to solve for the mean free

path results in

x ln(1 -) (2-27)

and if we choose to be uniformly distributed between 0 and 1 we can further write

x = -ln() (2-28)

or in terms of track length, 1, utilizing the individual isotope number density N, and microscopic

cross section total cross section, o(E),,

=- ln() (2-29)



If a particle then interacts within a cell volume, then the isotope that the particle interacted

with must be determined. The probability of a particle at specific energy, E, interacts with

isotope i is simply the ratio of the number density of i multiplied by the total cross section of i to

the sum of all the isotopes' number densities multiplied their associated total cross sections:

N, o(E) .,
p(E), N ) (2-30)


Then to sample the process from a random event, and assuming to be randomly distributed

between 0 and 1, isotope m is selected for interaction only if the following criterion is met:


"p(E), < i p(E), < Cp(E), (2-31)


After the interacting isotope is determined, the type of particle interaction incurred is sampled in

a similar manner. First, the probability of incurred a specific reaction, m, is computed by [41]:


p(E)k (E)k (2-32)
ra(E)k









Then to sample the process from a random event, and assuming to be randomly distributed

between 0 and 1, isotope m is selected for interaction only if the following criterion is also met:

m-1 N m
_p(E)k, < p(E)k,, < p(E)k,, (2-33)
1=1 1=1 1=1

After the interaction type is determined then subsequent residual nuclides/particles are

determined and resulting energy and angular distributions are calculated.

Particles continue to move within the system until the particle is either absorbed or leaked

from the system. A tallying estimator is used in order to retrieve particle information from the

system such as reaction rates and fluxes.63 64,65 The most basic Monte Carlo tally estimator is

the collision estimator which tallies the amount of a specific collision within a cell volume. In

order to calculate the average number of collisions in a volume, V, per unit time the following is

used equation:

c = V&c (2-34)

For the Monte Carlo computation the flux may therefore be approximated by:


= (2-35)
Vac

Where q is the average flux, &c is the average macroscopic collision interaction cross section,

and c is the mean number of collisions in a volume, V If the calculation is normalized to per

source particle then the collision estimator may be rewritten as:


= 1 1 cl (2-36)
Vyc N ,

A variation of collision estimator is the absorption estimator where the flux is now tallied not at

every collision site but at each absorption site:









1 1 a (2-37)


Both the collision and absorption estimators are dependent upon interactions taking place within

a volume.63 64,65

Due to the stochastic nature of the Monte Carlo process, all quantities computed result in a

stochastic error (standard error or relative error). The relative error is composed from the sample

variance. The actual variance of the population is



cr2 = =1 (2-39)


where [t is the actual population mean. Unfortunately, the actual population mean is almost

never known, and therefore the sample mean, x, is used to determine a sample variance. The

xi's tend to be closer to their average, x, than to the population average, /u resulting in

estimated values that are too small on average, therefore a n-1 is used in the divisor to account

for this underestimating. For a sample distribution, the sample variance of that distribution is

calculated by equation 2-38:


S2 (_ -)2 (2-38)
n-1

With further mathematical simplification we derive73

Z(x_ -)2 (x- 2 x,* + 2) 2 2 x, + x 2 Z 2 -2x nx, + n(x)2
n-l n-I n-1 n-1



n-1 n x2 _-2 (2-39)
n-1









The sample variance is referred to having n-1 degrees of freedom. This terminology results from

the fact that although the sample variance is based on the n quantities, (x, x), (x2 x),

..., (x ), these sum to 0, so specifying the values of any n-1 of the quantities determines the

remaining value.73 The Strong Law of Numbers states that if the true mean is finite then the

sample tends to the true mean as the sample size approaches infinity.66 Furthermore, examining

equation 2-38 we see that increasing the sample size, n, decreases the sample variance. The

variance of the sample mean is calculated to then be implemented in the calculation of the

relative error74

2 Y) 1
2 S2 S( 1 x 1
S2 R ..) (2-40)
n x x n 2x, n

Because the relative error implements the sample mean variance, increasing the sample size

further decreases the relative error of the calculation. A major shortcoming of the collision and

absorption estimators is that reliable estimates of the flux, and thus interaction rates, may only be

computed if many collisions/absorptions occur within tallied cell volume. For cell volumes

containing minimal amount of collisions/ absorptions the relative error for calculated quantities

will be high and thus leading to a lack of confidence in the computed quantity.

The track length estimator is another type of Monte Carlo estimator that does not incur the

mentioned shortcoming. The track length estimator utilizes the particle's average track length, 1,

within a specified cell volume to tally flux, and therefore relies of particles passing through a

specific cell volume and not if the particle happens to interact within the cell volume. The

particle density is equated to the scalar flux by

((F, E, t)dVdEdt = vN(F, E, t)dVdEdt (2-41)









Since a particle with travel a distance vdt during the time increment dt, the scalar flux may be

thought of as the total of the path lengths traveled during dt by all particles in the phase space

volume of dVdE.63 The scalar flux may therefore be defined as the total track length traversed

by particles within a specified volume


/ = (2-42)
V

where i is the average track length per unit time a particle traverses in volume, V In a Monte

Carlo simulation utilizing N particles the track length estimator of the flux may be calculated by:


^ = l (2-43)
VN="

Because the track length estimator of the flux relies only upon tracks entering a volume, more

particles contribute to the calculation of the relative error and therefore computing reaction rates

with lower relative error is easier with this flux estimator.63

Monte Carlo applied to neutron transport may make two distinct types of calculations:

fixed source and criticality. A fixed source calculation involves following a preset amount of

particles from a source, given by the user, until the particles are killed. Criticality calculations,

however, determine the effective neutron multiplication of a system by constantly changing the

geometric source distribution of neutrons for subsequent neutron cycles and following each

neutron till it is killed in an inelastic reaction, scattered below the weight threshold, or leaked

from the system and then determining if the way in which the neutron was killed resulted in

production of new neutrons. Many cycles of source distributions are analyzed in order to

converge on a source distribution that is truly representative of the analyzed system so that an

accurate characterization of the neutron multiplication may be computed.50









For a criticality the total amount of particles examined in the computation may be broken

up into the particles distributed per cycle and the total number of cycles examined. Initially the

user inputs a guess at the source distribution of particles. These particles then transport and are

eventually killed by an inelastic event, scattered below a weight threshold, or leaked from the

system. The absorption sites containing fissile nuclides that lead to fission then become the birth

sites for the next set of particles in the subsequent cycle. At each absorption site containing a

fissile nuclide, the amount of neutrons produced per fission is determined. The amount of

neutrons created per source neutron examined represents the multiplication of the system or keff.

If the system is a self sustaining critical system, for each source neutron lost from the system and

additional neutron is created for the next cycle. The multiplication of the system is a global

parameter and therefore keff is used to explain the entire global system.

Similar to the mentioned tallying process, three estimators exist for determining criticality.

Each estimator involves multiplying the flux estimator by a response function, h, to determine

the system multiplication:74

h = o~,, (E)u,, (E) (2-44)


,,n f (E) is the macroscopic fission cross section of isotope i at energy E, and u, (E) is the total

number of neutrons per fission of fissile isotope i. The collision and track length estimators are

tallied from interactions with each isotope i in all volumes n, by:74


kC (E)v (E) (2-45)
eff N n ,n, (E)


kf= 1 n ,,n,f (E)u,,n (E) (2-46)
N" n fin(









The absorption criticality estimator is computed slightly different form the absorption flux

estimator in that only absorptions in fissionable nuclides are tallied

k 1 c n (E)vn (E) (2-47)
eff N c" (E)


All three estimators (similar to the flux estimators) rely on large sample sizes and large

numbers of hits (i.e. collision/absorptions/number of entering tracks) in order to decrease the

relative error of the computation to have high confidence in the accuracy of the computed

quantity. Therefore, unfortunately, decreasing the relative error of the calculation is directly

correlated with increasing the associated computational cost of increasing the sample size to be

followed. Historically, the Monte Carlo method has been used only as a benchmarking

technique because of the computational cost associated with large sample sizes. Deterministic

codes generally calculate results much faster than Monte Carlo codes. However, as core

designers start to examine more advanced technologies that exhibit behavior that violates the

validity of the assumptions of the current deterministic methodologies, a method that captures

the true systems physics accurately becomes vital in order to effectively characterize system

properties.

Current Capabilities and Methods of State of the Art Depletion Codes

Many depletion codes exist in industry and academia for the purpose of determining

temporal nuclide inventories and reaction rate distributions.35 These codes are either externally

linked or sequential build technologies that contain a set of constraints used for determining

which depletion and transport coefficients to track and calculate. Depending upon the

investigated system, more or less detail in the solution method may be invoked in order to

achieve an acceptable characterization. The level of acceptable characterization may be

dependent on the validity of the benchmarking for the applied method or the actual level of









fidelity necessary to achieve proper knowledge for further design consideration. As many

calculations are required in order to license a reactor core, codes utilized by the nuclear power

industry for continual fuel cycle design of heavily benchmarked systems may employ many

approximations in order to achieve quick reliable results within the timeframe deadline allotted

by a particular customer.9 These assumptions are acceptable based on the premise that the

benchmarking of the system covers the design application space in which the code is

implemented. On the other hand, codes utilized for the development of radically new system

designs, incorporating a wide variety of unique material and geometry combinations that are not

heavily benchmarked, may implement a higher fidelity of detail in the solution method

employing few if any approximations.

Implementation Architecture

The main focus of most industry-based depletion codes is to support the operation and fuel

cycle design of commercial power reactors. Industry-based codes usually focus on streamline

manufacturing of heavily benchmarked systems that push the limiting design constraints to

maximize returnable profit.35 These types of systems involve many optimization iterations of

common heavily benchmarked technologies. Only broad, generalized parameters are required to

meet limiting conditions of operation (i.e. eigenvalue, power density, and linear heat generation

rate) for most commercial reactor designs. Therefore, because the calculation tools are heavily

benchmarked, the design requirements are broad, and the designer must complete the final

design analysis within a short prescribed time period, most of the current licensed industry codes

employ deterministic-linked depletion methodologies.35

Although various companies, such as Westinghouse, General Electric, Areva, and

Studsvick, use slightly different techniques to approximate time-dependent reactor behavior, the

overall methodology and processes are quite similar. First, a fine-fidelity deterministic









calculation (using PHOENIX, TGBLA, CASMO-3, etc.) examining many axial and radial slices

of the reactor are completed to generate coarser 1- to 4- group-averaged interaction rate

parameters for implementation in a coarser calculation of the full-core geometry (using ANC,

PANACEA, PRISM, etc.).35 However, due to the limitations of the deterministic method, these

methodologies are only valid within the design space that they have been benchmarked for.

Further analysis in group structure and cross section self-shielding is required in order to attempt

to apply the tools to more exotic applications. These codes choose to follow a predetermined

subset of reaction rates for the transport calculation, and further propagate a subset of fission

products to be followed during the depletion calculation. Certain reaction rates may not be of

importance to specific reactor designs or may only contribute within the noise of an acceptable

solution. Omitting these unnecessary reaction rates results in extra available memory as well as

minimizing computation cost. For most industry based codes major fission products such as the

Sm and Xe isotopes may be followed explicitly while the buildup of minor reactivity

contributors may be grouped into a lumped fission product set.2'25'29'31,32 The advantage to

fission product lumping is that fewer fission products are followed in the transport process

leading to a speedup in computation. The fission product lumps are, however, geometry and

spectrum dependant, and therefore though one type of fission product lump may be acceptable

for a certain type of system, that same fission product lump may be acceptable for a different

system. The combination of the acceptability of the transport method solution (i.e. energy group

structure, angular dependence, and geometric approximation), the reaction rates followed, and

the fission products tracked limits the application space for most industry based depletion codes.

These codes are usually only valid for the specific systems for which the codes were

originally developed. For example, a multi-group code developed for and LWR with certain









group structure, fission products, and reactions followed will not be accurate in determining

individual gamma-line signatures of fast reactor fission products for implementation in detector

signal segregation. However, that same code is very capable of determining an adequate neutron

multiplication and power shape calculation for licensing the reactor for operation.

Academia-based codes generally are developed for finer fidelity design of a robust variety

of nuclear systems that are not heavily benchmarked. These codes must implement as few

assumptions as possible in order to achieve an accurate result. The types of analysis for these

systems involve understanding more precisely the sensitivity of certain parameters within the

design making a higher-fidelity methodology, the design tool of choice. In general, deterministic

codes developed in academia implement fewer angular approximations [i.e., discrete ordinance

transport theory (XSDRNPM) versus diffusion theory], implementing a higher degree of

examined dimensions (i.e., 3D versus 2D), and/or using a finer-energy or continuous group

structure.35 Even though higher fidelity deterministic technologies exist to address concerns of

angular dependence, by nature of the process, these technologies still suffer from the burden of

proper group structure generation and flux self-shielding considerations.

To eliminate dependence from this burden, burnup codes coupling the Monte Carlo

reaction rate calculation have received more attention.36-41'46-48'75 However, due to the long

computation times required to execute a detailed high fidelity calculation, these codes have had

limiting success in implementation for commercial reactor design processes. To overcome this

issue, users would have to consider utilizing longer irradiation steps, with larger spatial zones,

following fewer nuclides, and fewer interaction rates per nuclides. An unfortunate example of

this issue is exemplified in the MCODE.39 In the MCODE only isotopes specified by the user

are followed in the transmutation process, and only (n,y) and (n,f) energy integrated reaction









rates are calculated for each nuclide. This simplification neglects key (n,a) reactions as

witnessed in the depletion of boron-10. Further acceleration techniques for Monte Carlo may

include using multi-group Monte Carlo such as in Keno Va.76 Multi-group Monte Carlo speeds

up the search algorithm when determining and energy dependent cross section at an interaction

site because far less energy groups need be queried. However, the limitation of the multi-group

approach is ultimately the validity of the group structure to a specific calculation. Implementing

approximations to a Monte Carlo linked capability, unfortunately, reduces the robustness of the

technology; therefore to take advantage of the explicitness of the Monte Carlo process,

computational cost will have to be sacrificed.

All linked depletion methodologies suffer from how the linking process affects the

functionality of both codes. For example, MONTEBURNS links MCNPX/5 with ORIGEN2.2

for high fidelity depletion calculations. MCNPX/5 is capable of tracking -393 isotopes for

particle transport and ORIGEN2.2 is capable of tracking 1700 nuclides for transmutation.

MONTEBURNS only chooses to follow a small subset of isotopes for particle transport

determined by specific reactivity criteria, and is not capable of following multiple isotope

transmutations at multiple temperatures.7 Other features within MCNPX/5 or ORIGEN2.2 are

also not compatible with the MONTEBURNS implication as the code was not developed for

these purposes. Many other linked technologies suffer from this same deficiency. All linked

methodologies suffer the added external input structure necessary to execute the coupling. For

example, the SCALE package successfully preserves the functionality of each code by

implementing each package as a separate module and using a driver code to execute link each

module however, the driver inputs are limited to certain setups and therefore though each code is

run as a separate module, the linking input still limits the user to a specific implementation.77









Specific features of code mechanics that enable the user larger flexibility might be ignored in the

linking process such as: operating in parallel execution and continuing a stopped execution.

All externally linked technologies suffer from the fact that the directory structure of the

linked package must be coordinated with the setup of each of the technologies to be linked, and

only the functionality of the linked code is supported by the developer of the linking process.

For example, ORIGEN is supported by Oak Ridge National Laboratory, MCNP is supported by

Los Alamos National Laboratory, and MCWO, a script linking MCNP and ORIGEN, is

supported by Idaho National Laboratory; however, MCWO does not support development of

ORIGEN or MCNP.41 Therefore if a user of MCWO finds a particle transport error in MCNP or

a transmutation limitation in ORIGEN, developers of MCWO are not responsible or able to

support fixes of these independent codes. Codes like the SCALE package benefit from the fact

that each code used for calculation is supported by the same organization.76 77 Therefore an

optimum multi-purpose depletion package is a code system that minimizes approximations in the

transport, depletion and linking calculations, supports development of all codes implemented in

the solution method and maintains full functionality and benefit of each code used in the

calculation.

Nuclide Inventory Limitations/ Calculation Size Considerations

The goal in designing any nuclear system is to determine the minimum amount of incurred

cost necessary in order to achieve the power and cycle requirements while still meeting safety

margin considerations.78 To determine these limiting conditions of operation only broad

generalized parameters are required (i.e., eigenvalue, power density, linear heat generation rate,

and shutdown margin). Hundreds of combinations of loading and control strategies incurring

many design iterations are examined in order to meet limiting design criteria. Therefore in order









to reduce the computational modeling expense approximations of larger system behavior must be

made.

To model an entire commercial power reactor core explicitly, capturing the exact

interaction rate behavior of every single microscopic section of every single material in the

system, would be an extremely memory restrictive. Considering that the power distribution

varies continuously in the radial and axial directions, reactor designers are forced to make

approximations in order to discretize the full depletion model so that it may be solved. Within

each discretization, reaction rates for each isotope of interested must be computed and stored for

implementation into the time-dependent isotope concentration calculator. For finer

discretizations, larger memory requirements and increased computational costs are incurred.62

For cases that are not concerned with microscopic phenomenon, instead of trying to

simulate an entire system explicitly, detail of a systems behavior may be resolved from

examining the large system as composed of average models of explicit sections of the reactor.

Many calculations of explicit sections of the larger reactor model are completed and major

interaction parameters are collapsed into an average set representing the explicit microscopic

section. Then these average sets of parameters are implemented into a courser representation of

the full system leading to a final computation that does not require as much explicit detail and

therefore reducing memory restriction and enhancing computational incentive.

Currently in the commercial power industry, reactor physics computations are bifurcated

into two separate processes. First, a multi-group or stochastic calculation (using MCNP,

PHOENIX, TGBLA, CASMO-3, etc.) is completed on a fuel bundle at certain axial zones to

generate collapsed interaction parameters. Then those generated collapsed interaction parameters

are implemented in a full-core simulator tool (using ANC, PANACEA, PRISM, etc).35 The









advantage to this process is the speed up in computation; however, the major fault is that the

legitimacy of the process now rests upon the validity of the chosen discretizations and averaged

parameters that are generated from the finer explicit micro-section calculations, and the fidelity

to determine microscopic behavior of interest is now lost in the averaging process.35

The most robust, optimum design tool must allow the user the option to examine different

sizes of possible geometric averaging techniques as well as the capability to eliminate them

entirely. Due to the robust modeling capability of combinatorial geometry, Monte Carlo codes

are capable easily modeling both intricate explicit models of the entire system as well as courser

representations implementing large blocks of common materials that incur decreased

computational expense. Since the Monte Carlo method employs few, if any, assumptions of

system physics, collapsed average interaction parameters generated by the code may be secured

to be of highest fidelity. This makes utilizing Monte Carlo for either group constant generation

or full core modeling the optimum robust linked depletion tool.









CHAPTER 3
METHODOLOGY DEVELOPMENT

A driver program is required in order to structure the execution of coupled algorithms

within coupled process. The MCNPX code serves as the driver program for the coupled

depletion process. MCNPX was chosen as the driver program because the code is already a

large scale multi-purpose particle transport package. The MCNPX code package contains -1000

subroutines with each subroutine containing hundreds of execution lines. The CINDER90

depletion algorithm is placed within the MCNPX code package, and an interface subroutine sets

CINDER90 inputs and retrieves CINDER90 outputs. The other 29 implemented subroutines

execute various assignments in order to properly average data, set parameters for implementation

in the CINDER90 depletion algorithm, reconstruct atom density information for subsequent

particle transport calculations, retrieve input information for problem set up or to construct

output. The coupled deletion process involves a Monte Carlo (MCNPX) steady-state reaction

rate calculation linked to a deterministic depletion calculation (CINDER90). The whole process

resides in roughly 30 subroutines in the MCNPX code (five new subroutines and 25 augmented

subroutines). The linking process is displayed in Figure 3-1. MCNPX runs a steady-state

calculation to determine the system eigenvalue, 63-group fluxes, collision rates, v, and Q values.

CINDER90 then takes those MCNPX-generated values and carries out the depletion calculation

to generate new number densities at the following time step. MCNPX then takes those new

number densities and generates another set of fluxes and reaction rates; the process repeats itself

until after the final time step.79 The MCNPX linked depletion architecture requires important

decisions regarding isotope tracking philosophy and methods, calculation of depletion equation

coefficients, time-dependant cross section averaging techniques, and the implementation

architecture.79 Chapter 3 details the development and implementation of these philosophies










within the MCNPX depletion methodology and offers key important considerations to account

for in the development similar technologies.


MCNPX
Steady State Eigenvalue Calculation



Yes Final Depletion Step?

No
Collision Densities, v and Q




CINDER90
Depletion Calculation


Isotone Concentrations

-- Final Output Processing


Figure 3-1. Monte-Carlo-linked depletion process model.

Approach to Isotope Tracking

MCNPX depletion burns a system by specified burn material instead of by individual cell

location.78 This technique was employed in order to take advantage of repeated structure

geometry simplifications already available in the MCNPX geometry build subroutines that help

decrease the size of necessary input in order to bum an entire system.50 The trade off in taking

advantage of this input simplification is that the user must be aware that unique materials are

required for each burn cell containing a unique power profile. However, the advantage of this

capability eases the burden in modeling complicated geometries.

MCNPX depletion requires that each burn material contain all isotopes to be generated

within that material at the beginning of the computation. This therefore requires that the code









determine all fission and activation products upfront as well as particle interaction data to be

used in the depletion process. As mentioned previously, the depletion process is bifurcated into

two separate processes: steady-state particle transport and isotope transmutation. Particle

reaction data for use in the particle transport calculation is available from direct experiment via

evaluated data library sets. These evaluated library sets include ENDF, JEF, JENDL, BROND,

CENDL, etc. Because most of the evaluated library data sets include reaction information from

experiment, ability to execute and retrieve data from experiments limits the amount of isotopes

that actually posses evaluated library transport information. For example, generating an

experiment to measure interaction rate data from a radioactive isotope with a very short half-life

may not be possible and therefore the evaluated nuclear data files will usually not contain

transport information for such an isotope. The current ENDF/B VII.0 library only contains

actual transport cross sections for 393 isotopes, and therefore, without an analytical or empirical

model, a maximum of 393 isotopes may be tracked in particle transport.80 Unfortunately, the

actual transmutation process has the possibility of generating 3400 isotopes. Though there exists

measured decay data for most of these nuclides, interaction rate data does not exist. CINDER90

is capable of tracking 3400 isotopes in the transmutation process; however, transmutation rates

are required in order to complete the calculation.61 These transmutation rates will only have

spatial dependence if determined from the steady-state reaction rate calculation. However, if

continuous energy cross section data does not exist for a specific nuclide, that nuclide cannot be

employed in the steady-state reaction rate calculation. For nuclides not containing tabulated

transport cross sections in the MCNPX data library, MCNPX does have the capability to use a

theoretical model to generate a cross section for that specific isotope; however, this capability

was benchmarked for > 150 MeV particle interactions, which is well above the energy regime for









the majority of particle interactions incurring in a reactor, and therefore not a safe technique to

be implemented in reactor calculations.79

CINDER90 does contain 63-group transmutation rate information for 3400 nuclides.61

The majority of this data, not derived from evaluated library files, was generated from analytical

cross section modeling codes or by expert best estimates. Therefore MCNPX applies a mix and

match method for transmutation rates in which if there exists and evaluated library transport

cross section for a nuclide for use in MCNPX, then the reaction rate information from that

nuclide is calculated directly within MCNPX and sent to CINDER90 as a transmutation

coefficient. For all other nuclides, a 63-group spatially dependent flux is calculated within

MCNPX and sent to CINDER90 to be match with a 63-group cross section sets available in the

CINDER90 data library file, where that cross section set most probably was derived from an

analytical model or expert best estimate, to be integrated into a transmutation coefficients for use

in the depletion equation.79 To save computation time and reduce the influx of memory-

prohibitive information, MCNPX only tracks those isotopes accounted for from the following

processes:

1. Listing the isotope on a material card.
2. Selecting an isotope from a preset fission product tier.
3. Producing the isotope from the Isotope Generator Algorithm.

CINDER90 still tracks all of the daughter/granddaughter/great-granddaughter/etc. decay

reactions for 3400 isotopes; therefore, the total isotope concentrations listed in MCNPX are the

true concentrations from the full depletion process.

Fission Product Tiers

Executing a successful depletion calculation involves accounting for the reactivity effects

of as many fission products as possible. In most cases, a fission event will result in the creation

of two-to-three fission products with a total mass that sums to less than the mass of the original










isotope. The creation of the fission product isotopes is probabilistic in nature and follows a

distribution. Figure 3-2 displays the fission product distribution for U-235 for thermal and 14

MeV fission reactions. Hundreds of fission product isotope yield combinations are possible for a

given fission event from a specific isotope. Each fissionable isotope also contains different yield

distributions. Figure 3-3 shows fission product yield distribution for certain plutonium and

uranium isotopes. CINDER90 contains energy dependent fission product yield data for 1325

different fission products.61

10
14 MeV




SO.I


0 01
Thermal




0 001


00001
70 90 110 130 150
Mass number

Figure 3-2. Uranium-235 fission product yield distribution.7

Due to the nature of the Monte Carlo process, increasing the amount of isotopes tracked in

a problem increases the associated computational cost. When a particle interacts within a

specified cell volume, a calculation must be run to sample which isotope within that material the

particle interacted with. This sampling is dependant upon individual isotope atom density,

interaction cross section, and total amount of isotopes within that given cell. Increasing the

amount of isotopes, increases the overall associated computational cost. However, not tracking

certain fission products within a calculation leads to altered neutron spectrum ultimately leading

to inadequate estimation of system reactivity and inaccurate representation of true system









physics. Therefore determining an adequate combination of fission products necessary to

sufficiently account for the system physics becomes paramount in achieving an accurate

depletion calculation.



I-

: d. 3I





,I j



1/'

19 i M i0


Mm imb.r A.

Figure 3-3. Fission Yield Distributions for Various Plutonium and Uranium Isotopes.

Many options exist for determining how the user accounts for the generation of fission

products within a calculation. The user could be forced to list, on an input deck, all the fission

products that were of interest. This would require that the user had prior knowledge of all the

fission products generated during the fission process that contained useful transport cross section

information. Inputting every single fission product of interest, in all burn material would be a

very cumbersome task. MCNPX offers the user preset fission product "tiers", which are subsets

of a preset fission product array. By using preset fission product tier sets, the cumbersome and

tedious task of cataloguing every fission product on an input file is eliminated.82 The user then

can omit certain isotopes from the fission product tier if necessary. Three tiers of fission product










content are available. Each tier corresponds only to those isotopes that are used for particle

transport in MCNPX and the nuclide densities that are reported in the following output file. In

MCNPX 2.6.A the fission product array contained 167 fission products, listed in Figure 3-4, with

the following fission product tiers:79

1. Tier 1. (default) Zr-93, Mo-95, Tc-99m, Ru-101, Xe-131, Xe-134, Cs-133, Cs-137, Ba-
138, Pr-141, Nd-143, Nd-145.9

2. Tier 2. Isotopes contained in the fission product array that are included in the released
cross-section library file (XSDIR) for MCNPX.

3. Tier 3. All isotopes contained in the fission product array.

32072,32073,32074,32076 33075 34077 34079 34080 34082,35081
36082,36083,36084,36085 36086,37085 37086,37087,38086,38088
38089,3809090 39089,9039039091,40090 40091,40092 40093,40094
40095 40096 41095,41097,42095.42096 42097.42098 42099,42100
43099 44100 44101.44102 44103.44104 44105.44106 45103.45105
46104,46105,46106,46107 46108,46110,47109,47111 48110,48111
48112 48113 48114,48116,49115,50116, 0117,50118 50119,50120
50122 50123 50124,50125 50126.51121 51123,51124 51125,51126
52122,52124 52125,52126 52128.52130 52132.53127 53129,53130
53131 53132 53133,53134 53135,54128 54129,54130,54131,54132
54133 54134 54135,54136 55133,55134 55135,55136 55137,56134
56136 56137 56138,56140,57139.57140 58140.58141 58142,58143
58144 59141,59142,59143 59145.60142 60143,60144 60145,60146
60147 60148 60150,61147 61148,61149 61151,62147 62148,62149
62150 62151 62152,62153 62154,63151,63152,63153 63154,63155
63156 63157 64152,64154,64155.64156 64157.64158 64160,65159
65160 66160,66161,66162 01003.05010,05011

Figure 3-4. Original fission product array containing 167 fission products.

Preliminary testing of a typical PWR pin cell calculation utilizing the fission product tiers

in MCNPX as compared to the benchmarked MONTEBURNS depletion code suggested that

utilizing the maximum amount fission products increased the agreement between the two

codes.81 For preliminary analysis purposes, comparing MCNPX depletion and MONTEBURNS

was deemed acceptable for determining "ballpark" estimates of appropriate functionality because

MONTEBURNS was also a Monte Carlo linked technology that has been thoroughly

benchmarked.74, 83 Figure 3-5 displays the difference in neutron multiplication versus time and

Figure 3-6 and Figure 3-7 display the difference in end-of-life fission product buildup when

comparing the varied fission product tiers to the benchmarked MONTEBURNS method.











Analysis with MCNPX 2.6.A suggested that increasing the amount of fission products followed

in the calculation, increased the agreement between MCNPX and MONTEBURNS.82 With the

release of ENDF/B VII.0, it was determined that every fission product containing transport cross

sections should be included in the fission product array, increasing the array to account for 220

fission products.84 Each fission product tier continued followed the same criteria as previously

implemented. The fission products included in the current 220 fission product array are listed in

Figure 3-8.84

1.18
1.16
1.14
1.12
S1.1
1.08
1.06
1.04
1.02
0 5 10 15 20 25 30
Burnup (GWd/MTHM)

MONTEBURNS --Tier 1 w/o models
Tier 2 w/o models -x Tier 1 w/ models
-e- Tier 2 w/ models -*-Tier 3 w/ models


Figure 3-5. Resulting time-dependent neutron multiplication from implementing fission
products with and without cross section models in MCNPX as compared to the
benchmarked MONTEBURNS method.

10.00%

S5.00%
U


SO -
S-5.00% -


S-10.00% -

-15.00%
Nuclides

o Tier 1 without Models Tier 1 with Models


Figure 3-6. Percent difference of certain fission products from MONTEBURNS using MCNPX
Tier-1 fission products with and without cross section models.














10.00%

w 5.00%

0.00%

-5.00%
2 I P, OD -W -n
-10.00%

-15.00%
Nuclides

STier 2 u Tier3



Figure 3-7. Percent difference of certain fission products from MONTEBURNS using MCNPX
Tier-2 and Tier-3 fission products with cross-section models.

31069,31071,32070,32072,32073,32074,32076,33074,33075 34074,
34076,34077,34078,34079,34080,34082,35079,35081,36078 36080,
36082.36083.36084,36085,36086,37085,37086,37087,38084 38086,
38087,38088,38089,38090,39088,39089,39090,39091,40090 40091,
40092,40093.40094,40095,40096,41093,41094,41095,41097 42092,
42094,42095,42096.42097,42098,42099.42100,43099 44096 44098.
44099,44100,44101 44102.44103,44104.44105,44106 45103 45105,
46102.46104.46105,46106,46107,46108.46110,47107 47109 47111,
48106,48108,48110 48111,48112,48113.48114,48116 49113 49115.
50112,50113.50114,50115,50116,50117.50118,50119 50120 50122,
50123,50124,50125,50126.51121,51123.51124,51125 51126 52120,
52122,52123 52124,52125,52126,52128 52130,52132 53127 53129,
53130,53131 53132,53133,53134,53135.54123,54124 54126 54128,
54129,54130,54131,54132,54133,54134.54135,54136 55133 55134,
55135,55136.55137,56130,56132,56133.56134,56135 56136 56137,
56138,56140,57138,57139,57140,5 365813 5813858139 58140 58141,
58142,58143,58144,59141.59142,59143 59145,60142 60143 60144,
60145,60146.60147,60148,60150,61147.61148,61149 61151 62144,
62147,62148,62149,62150,62151,62152.62153,62154.63151 63152,
63153,63154.63155,63156,63157,64152.64153,64154.64155 64156,
64157,64158,64160,65159,65160,66156.66158,66160.66161,66162,
66163,66164,67165,68162,68164,68166.68167,68168.68170,69169


Figure 3-8. Enhanced fission product array containing 220 isotopes.


Isotope Generator Algorithm

Storing every decay chain for every possible combination of isotope listed on a burn


material card would be extraordinarily memory prohibitive and would greatly slow down the


speed of the code. To limit memory consumption and computational cost, only the immediate


daughter products from isotopes specified within selected bur materials are tracked in MCNPX


for particle transport.85 These immediate daughter reactions are determined by the Isotope










Generator Algorithm. The algorithm places all possible daughter products, from either reaction

or decay, within a bum material card at an atom density of 1E-36.






p out


(Cn,2n) Original
n out Nudeus (n;y)
(sn)
(n) (n,d) (n,p)
t out (n,nd) (f{lp) u
d out p out ou

(nt) (nHe)
ct OUl (rn,r',' (- ,I Fe
He out (n,pd)
a neutronQ a- alpha particle
pN:-Il.n = beta minus (negative electron)
d = deLteron p' = beta plus I sit n r
t = triton = eectron capture


Figure 3-9. Nuclides created from a decay or nuclear reaction process and generated for particle
transport by the isotope generator algorithm.3

Since the isotopes tracked within the MCNPX steady-state must be determined at the

initiation of the entire calculation, specifying the isotope at a super low atom density secures the

inclusion of the isotope in the calculation while not incurring an artificial reactivity penalty.

Figure 3-9 displays the isotopes generated for a nuclide with Z > 4. The location of each

generated isotope corresponds to the position of that isotope as it would appear in the Chart of

the Nuclides.3 Each box contains information regarding the interaction or decay process that

leads to the creation of that isotope. Blank boxes represent possible decay chain isotopes

resulting from multi-interaction processes or from other particle interactions (this is a result of

the possible future application of utilizing the code to look at high energy particle interactions in

accelerators and solar flares).









Manual Time-Dependent Isotope Concentration Changes

To model real life depletion systems, isotopes may have to added or extracted in order to

meet the conditions of the operating strategies. For example, in a PWR, boron must be diluted as

fuel is depleted in order to maintain criticality. For another example, in a gaseous fuel or molten

salt reactor fission products may need to be extracted continuously in order to maintain system

criticality. Certain depletion codes only allow the user to manipulate concentrations of isotopes

used to maintain criticality in LWRs such as diluting boron concentration. MCNPX is a general

purpose modeling code used to model all types of reactors. For the modeling capability to truly

be robust, the user must be able to manually manipulate the concentration of any nuclide during

the execution of the depletion process in order to properly model the true system behavior.

MCNPX depletion process allows for the user to manually manipulate the concentration of

any isotope within any burned material at any time step. The user invokes this capability by use

of the MATMOD keyword on the BURN card (input specification, input structure, cards and

keywords are explained in the Input Interface section).86 The user is able to manipulate the

concentration of any isotope by specifying the isotopes atom density, weight density, atom

fraction or weight fraction within the bum material volume. The user is also able to able to

specify how the concentration will change during the time-dependant cross section averaging

process. The capability resides within the bum data retrieval subroutine (bummat.F) where

new atom fractions and atom densities are set for use in the next particle transport step. The

process of manually setting concentrations for isotopes therefore exists post-executing the

CINDER90 number density calculation. For particle transport, the MCNPX code requires that

isotope concentrations be ultimately be converted to atom fractions to be applied to total material

atom densities.









The process for calculating the renormalized atom fractions if the user specifies an isotope

to be altered by atom density is as follows:86

1. A loop is generated that loops through every isotope used in particle transport.

2. If an isotope is not to be manually altered that isotope's atom density is summed in
the s2 variable.

3. If the isotope is to be manually altered that isotope's atom density is summed in the
cf variable.

4. The total material atom density is then calculated by
pa = s2 + cf (3-1)

5. To calculate the individual isotope atom fractions, another loop over all the
isotopes is generated, and the individual isotope atom fractions are calculated by

AFk Pak where p, = Pa,k (3-2)
Pa k

The process for calculating the renormalized atom fractions if the user specifies an isotope to be

altered by weight density is similar to the previously mentioned process except the code first

immediately converts the inputted weight density to an atom density and then the code continues

along the same solution path as the previously mentioned process. If the user specifies an

isotope to be altered by atom fraction, the following process is utilized: 86

1. A loop is generated that loops through every isotope used in particle transport.

2. If an isotope is not to be altered that isotope's atom density is summed in the s2
variable.

3. If the isotope is to be altered that isotope's input atom fraction is immediately
saved as well as also summed into the fn variable.

4. The total material atom density is then calculated by

pA s (3-3)
1-fn
5. To calculate the individual isotope atom fractions for an isotope not manually
altered, another loop over all the isotopes is generated:









a. If a manually altered isotope is encountered, then the manually given atom
fraction is implemented and no calculation is required.

b. For the rest of the isotopes the individual isotope atom fractions are
calculated by

AFk k (3-4)
s2

If the user specifies an isotope to be altered by weight fraction, the following process is similar to

the process that implements the altered atom fraction: 86

1. A loop is generated that loops through every isotope used in particle transport.

2. For an isotope that is to be altered, the following quantities are tracked (sums for
this process include only isotopes that are to be altered):

a. The individual isotope weight percent, wj, divided by the individual isotope
mass, Mj, stored in the xj variable.

b. The sum ofxj is stored in the wf variable.

c. The sum of the wj's are also tracked and stored in fn variable.

3. If an isotope is not to be altered the following quantities are tracked (sums for this
process include only isotopes that are not to be altered):

a. The sum of the individual isotope masses multiplied by the individual
isotope atom densities, -M, pa, are stored in the s3 variable.


b. The sum of the individual isotope atom densities are then stored in the s2
variable.

4. Another loop over all the isotopes is generated.

5. The following quantity is calculated for each that isotope not to be altered (sums
encompass only isotopes not to be altered ):
pa,, *(1- fn) p,, *(1- fn) .
a. x = *(- *( is stored, where 1-fn represents the
s3 p, *M

weight percent and Mi represents the individual atomic mass for each of the
unaltered isotope.

b. The sum of the xj variable for unaltered isotopes is then stored in the s4
variable.










6. Another loop over all isotopes is generated

7. The following derivation is used in order to generate the equation for calculating
the individual atom fractions for each isotope

1 w%, = w%, (3-5a)


p, =w%, *NA *p


(3-5b)


pa' I


w%
Z +W


zw%o
a, M p NA
p %w ,,*p N%

I ,
p'I *Mf kw% *p*NA
w"" Mj p w%


x
s4+wf


p*NA *-wo


/wk a%

kMk P*N


w%,



ZMk +M,


(3-5d)


(3-5e)


Y w%k + w%, w%k
k MAk k AMk
w%k *N *
A
Mk P,k AFk

%k N P
k Mk


(3-6f)


This capability enables manual changing of nuclide concentrations during burnup

calculations. For example, boron concentrations are usually specified in parts per million weight

concentration while actinide extraction for fuel recycle may be specified in fractions per atom.


S
s4+wf


(3-5c)









The versatility of this capability allows the user flexibility in modeling most types of manual

isotope changes that exist under real operating strategies.

Metastable Isotope Tracking

The MCNPX data library requires that isotopes be listed in and referenced by a ZAID

format. The ZAID format consists of the following:

Z = atomic number
A = atomic mass
ID = identifier corresponding to the evaluation type (i.e. specific library set,
temperature, particle type, etc.)

The following equation is used for calculating the ZAID:

ZAID = (Z*1000 +A).ID (3-7)

CINDER90 uses the AZS format for listing and referencing isotopes. The AZS format consists

of the following:

Z = atomic number
A = atomic mass
S = isomeric state

The AZS format is similar to the ZAID format and calculated using the following equation:

AZS = S+10+10*z+10000*A (3-8)

Since both formats are required in order to execute both codes, an algorithm is necessary in order

to swap formats when accounting for isotopes in each code. Unfortunately, since there did not

initially exist a convention for representing isomeric states in ZAID format, transitioning

metastable isotope between each code was not trivial. In the past, users created their own

naming convention for ZAID format metastable isotopes, and as long as a tabular cross section

file was referenced in the XSDIR library file for these types of isotopes, any naming convention

could be accepted (i.e. To represent Ag-110m, ZA=47210, Z=47 and A=110+100*(metastable

state)). However, because MCNPX is now linked to the CINDER90 code which uses a different









naming convention than MCNPX, an MCNPX standard convention for naming metastable

isotopes is now required. Metastable isotopes ZA are represented by using the following

equation:

A'= (A+300) + (m*100) where m=0,1,2,3,4 (3-9)

For example:

Ag-110m = 47510 47510 = 47110 + 1st metastable (3-10)

MCNPX uses the following standard naming convention to convert ZAID format to AZS for

user in CINDER90, and the conversion takes place within the interface subroutine.87

Calculation of Depletion Equation Coefficients

There exists the probability for hundreds of different types of particle reactions within a

typical nuclear system. Examining equations 2-1, 2-2a and 2-2b, we can see that increasing the

amount of reaction rates tracked per isotope greatly increases the size of the depletion equations

for more reaction types contribute to the calculation of the creation and destruction coefficients.

Due to the high energy threshold for most of these particle reactions, for the majority of reactor

applications, only a base set of reaction actually have a significant probability of contributing to

the calculation of the destruction and creation coefficients. Therefore choosing to follow

reactions that do not contribute to the calculation restricts the available memory and diminishes

computational performance. For the computed reaction rates, the Monte Carlo steady-state

calculation computes normalized fluxes and reaction rates per source neutron. The depletion

equation requires a total magnitude value for the destruction and creation coefficients. Because

the Monte Carlo calculation only calculates these quantities relative to the particle source

strength, normalization parameters must be further calculated in order to convert the normalized

coefficients into coefficients that are set at the exact magnitude of the analyzed problem. After

the normalization parameters have been calculated for each isotope, the code must be able to









properly determine how fission products are yielded from the fission process. The culmination

of all of these mentioned procedures compose the important considerations necessary for

calculating adequate depletion coefficients for achieving an acceptable solution to the temporal

nuclide inventory.

Important Reactions Followed

Each reaction rate tracked per isotope represents memory allocation that further restricts

other aspects of an analyzed problem. CINDER90 does contain a 63-group cross section set for

all reactions available in the library file. Therefore, CINDER90 theoretically only requires a 63-

group flux to match to its 63-group cross section set in order to calculate depletion coefficients.

The accuracy of this matching process is therefore highly dependent upon the applicability of the

cross section sets to a specific design application as well as being limited by the same

approximations applied to typical deterministic calculations. Therefore calculating a continuous

energy integrated reaction rate for use in a depletion coefficient for the depletion equation

solution better preserves the true system physics represented within the coefficient.

Unfortunately, the trade-off in tracking continuous energy integrated coefficients within the

Monte Carlo calculation is that each coefficient tracked must be tracked for each isotope which

then limits the available memory for other aspects of the calculation. Initially the MCNPX

depletion code was set to only track the continuous energy integrated (n,fission) and (n,y)

reaction rates for application in the depletion solution.17 Previous analysis suggested that the

(n,y) reaction rate was the dominant contributor to the neutron destruction and therefore other

capture reaction rates were tracked only by matching a 63-group flux in MCNPX to a 63-group

cross section set in CINDER90 in order minimize necessary memory requirements. The code

therefore applied a mix and match procedure in which only the dominant contributors to the

coefficients were calculated as continuous energy integrated coefficients while the rest of the









reactions were calculated by matching a 63-group flux calculated within the steady-state Monte

Carlo to a 63-group cross section set in CINDER90. Recoverable energy per fission and

neutrons generated per fission were also tracked and calculated in continuous energy; however,

threshold reactions such as (n,3n) and (n,2n) and residual minor capture reactions such as (n,p)

and (n,a) were not tracked utilizing continuous energy cross sections. Individual threshold

reaction rates were implemented by the 63-group matching approach mentioned previously. The

concern of the 63-group matching approach was that the validity of this technique resided in the

applicability of the 63-group cross-section set to various designs.

The 63-group cross section set within the CINDER90 code was generated by collapsing a

finer cross section set over an initial arbitrary system flux. The arbitrary flux utilized to generate

the collapsed 63-group cross section set within CINDER90 generated cross sections that were

reliable for calculations only representative of that system flux. Therefore calculations

implementing the 63-group cross section set within CINDER90 would be in error if the flux of

the examined system differed from the flux utilized to generate the cross section set.

Furthermore, if the cross sections themselves were improperly collapsed to be poorly

representative of threshold reaction boundaries, the calculation would be further in error.

Therefore, MCNPX was augmented to track these threshold reactions properly in continuous

energy to eliminate the error in calculation from the multi-group approach.79' 85

A numerical benchmark calculation was executed in order to isolate whether MCNPX

burnup should track continuous energy reaction rates for (n,2n), (n,3n), (n,p) and (n,a). A

hypothetical infinitely reflected pin cell geometry was depleted using MCNPX 2.6.A

implementing only (n,y) and (n,fission) continuous energy reaction rates (OLD MCNPX method)

and then compared to MCNPX 2.6.A implementing (n,y), (n,fission), (n,2n), (n,3n), (n,p) and









(n,a) continuous energy reaction rates (NEW MCNPX method) and MONBTEBURNS. Both of

the MCNPX versions used for this test did not employ a time-dependent cross section averaging

technique such as the method employed in MONTEBURNS. MONTEBURNS was chosen as a

comparator code because of the extensive acceptable benchmarking already completed with this

depletion code.74' 83 It is true that the methods of isotope tracking and reaction rate calculation

differ slightly between MCNPX and MONTEBURNS; however, the base solutions should be

similar since MONTEBURNS has been extensively benchmarked.

The objective of these comparisons was to determine the magnitude in improvement as a

result of employing continuous energy reaction rates for minor capture and threshold reaction

events. The analyzed pin cell geometry is displayed in Figure 3-10.

Borated Water

Clad

Fuel


Figure 3-10. Examined infinitely reflected pin cell geometry.

Table 3-1. MOX fuel composition.
ZA value Atom Density
[atoms/cm3]
8016 4.5854e-2
92235 1.4456e-4
92238 1.9939e-2
94238 1.1467e-4
94239 1.0285e-3
94240 7.9657e-4
94241 3.3997e-4
94242 5.6388e-4

The pin cell geometry consisted of 3 regions: a cylindrical mixed oxide (MOX) fuel region, a

zirconium clad region, and a borated water region. The MOX cylindrical fuel dimensions were

365 cm tall and 0.4095 cm in radius, and the clad was 0.0655 cm thick and composed entirely of









natural zirconium. The fuel composition is listed in Table 3-1. The entire fuel pin was encased

within a borated water rectangle 365 cm tall with a 1.313-cm length and 1.313-cm width. The

borated water composition is listed in Table 3-2.

Table 3-2. Borated water composition.
ZA value Atom Density
[atoms/cm3]
1001.60c 4.7716e-2
8016.60c 2.3858e-2
5010.60c 3.6346e-6
5011.60c 1.6226e-5

The fuel pin was depleted at a power of 66.956 kWt. This power was approximated to be

the individual pin power for a typical 3670 MWth pressurized water reactor assuming -193 17 X

17 fuel assemblies. The geometry was burned for a total of 730.645 days (2 years). The entire

burn time was separated into 6 depletion steps: 0.645 days, 40 days, 100 days, 140 days, 200

days, and 250 days, for a total of 730.654 days. Each steady-state Monte Carlo calculation used

5000 particles per cycle, skipping the first 5 cycles, for 300 cycles. ENDF/BVI.0 cross sections

were used for each steady-state calculation. For these calculations, MONTEBURNS used

MCNPX 2.5.0 linked to ORIGEN2. Though the depletion steps were unrealistic for achieving

an acceptable depletion solution due to the expected significant change in the flux shape and

magnitude over such course depletion steps, the comparison of the methods was hypothesized to

provide similar results regardless of the implementation as long as the consistency in the

approach was preserved. Since both methods employed a linked calculation preserving spatial

dependence of the solution, incurring the exact same time steps, both methods were deemed

acceptably consistent enough to warrant fair comparison.

Preliminary results comparing the OLD MCNPX method with the MONTEBURNS

depletion code displayed discrepancies in certain actinide and fission product buildup. Figure 3-










11 displays the EOL (end of life) percent difference of certain actinides between the OLD

MCNPX method and the MONTEBURNS method. The speculated reason for the discrepancies

were lack of calculating continuous energy threshold reaction rates (reasons of secondary

concern also included limits of the matrix exponential method in ORIGEN2; however, these

effects were hypothesized to be of second order effect).


20.00%
0.00% -
cc icc |z f -? -?
-20.00% U k NY
U 0 0) co (D M o
S-40.00%

S-60.00%
S-80.00%
-100.00%
-120.00%
Nuclides




Figure 3-11. Percent difference of certain reference actinides between the OLD MCNPX method
and the MONTEBURNS method.

The NEW MCNPX method calculated continuous energy integral reaction rates for (n,

fission), (n,y), (n,2n), (n,3n), (n,p), and (n,a). These integral reaction rates were calculated during

the Monte Carlo process and fed to CINDER90 for isotope depletion. This process varied from

the OLD MCNPX method in that the majority of significant threshold reaction rates were now

calculated in continuous energy and no longer required a 63-group Monte Carlo flux to be

matched to a generalized 63-group CINER90 cross-section set.

The integral continuous energy reaction rates were calculated separately for each burn

material and tallied at the site of collision. Figure 3-12 displays the improvement in reference

actinide calculation. A drastic improvement in actinide generation is realized when using

continuous energy reaction rates for threshold reactions. The percent difference in Np-237 drops












from -70.24% to -4.85%, and the percent difference in Pu-237 production drops from -95.69% to


-1.02%. Due to accentuation of U-237, Np-237, and Pu-237 differences in Figure 3-11, Figure


3-12 does not capture the improvements in EOL calculation of other important actinides;


however, by utilizing the continuous energy reaction rates for threshold and minor capture


reactions, 1-2% improvements in actinide predictions are also witnessed in U-236, U-239 and


Pu-238.


2.00%

1.00%
0.00%

-1.00%

-2.00%

-3.00%
-4.00%

-5.00%

-6.00%


M 0 0


Nuclides



Figure 3-12. Percent difference of certain reference actinides between the NEW MCNPX
method and the MONTEBURNS method.


2.00%

0.00%

S-200%

8 -4.00%

2 -6.00% -

-8.00%

-10.00% -
Nuclides



Figure 3-13. Percent difference of certain reference fission products between the NEW MCNPX
method and the MONTEBURNS method.









Though tracking continuous energy threshold and minor capture reactions enhanced

actinide prediction capabilities, little improvement was witnessed for fission product prediction.

The NEW and OLD MCNPX methods for reaction rate tracking witnessed similar discrepancies

in fission product buildup. Figure 3-13 displays the difference in EOL fission product prediction

when comparing the new MCNPX method to MONTEBURNS. Unfortunately, time-dependent

fission product buildup is a major contributor to negative reactivity trajectories, and therefore

MCNPX and MONTEBURNS only predicted slightly different time-dependent reactivity.

Figure 3-14 displays the difference in time-dependent neutron multiplication between the

MCNPX and MONTEBURNS methods. Two major possible reasons existed for the differences

in fission product generation between both MCNPX methods and the MONTEBURNS method:

fission yield discrepancies and lack of cross section averaging technique.

CINDER90 offers the latest fission yield data available.61 This data includes fission yield

information for 35 fissionable isotopes incurring thermal, fast or high energy neutron reactions.

ORIGEN2 contains fewer fission yield data and makes approximations for yield data for

fissionable actinides not contained in an ORIGEN library file. Furthermore, not all fission

product transmutation paths are followed explicitly in the matrix exponential method in

ORIGEN2.60 Therefore, it was deemed plausible that the fission yields and subsequent

transmutation paths used in CINDER90 and ORIGEN2 differed slightly.

The versions of the MCNPX methods examined for this specific study did not implement

any type of time-dependent cross section averaging technique. Figure 3-14 displays evidence of

this hypothesis. The eigenvalue, keff, is tracked almost exactly during the short time steps at the

beginning of life (BOL) but differs in the longer time steps. Therefore, a predictor corrector

methodology was suggested to reduce the difference in isotope transmutation trajectory and











eigenvalue prediction for the two codes (this concept is explained further in the Predictor

Corrector Method section). However, because actinide production was improved by tracking

continuous energy threshold and minor capture reaction rates, these reaction rates were chosen

for inclusion in tracking for further versions of the code.85


1.180
1.160 1
1.140
1.120
1.100
1.080
1.060
1.040
1.020
0 5 10 15 20 25 30
Burnup (GWd/MTHM)

--- MONTEBURNS -m- MCNPX Old MCNPX New


Figure 3-14. Time-dependent eigenvalue calculations.

Flux Normalization

MCNPX calculates reaction rates on a per source basis; however, CINDER90 requires a

total magnitude reaction rate to deplete the system appropriately. In order to generate a total

magnitude reaction rate, MCNPX must determine the magnitude of the source and multiply this

value by the tracked reaction rates per source to calculate the total magnitude of the reaction

rates for use in the depletion solution. The source of neutrons in a multiplying system is the total

neutron production rate of that system. The total magnitude of the flux may then be calculated by

multiplying the flux per source by the production rate as in equation 3-11:

Pxv
total = MNP x Flux Multiplier -> Flux Multiplier = (3-11)
Value

v = neutrons per fission event
Qvalue = recoverable energy per fission event
P = power level










How these parameters are calculated and what regions of the calculation these parameters are

applied complicate the implementation of these parameters as normalization constants.

At each interaction site within a material that contains an actinide, each reaction mentioned

in the prior section is tallied along with:

(n, v ) = neutrons per fission event multiplied by the macroscopic
fission cross section for that material per source neutron.

(n, Qvaue f ) = recoverable energy per fission event multiplied the macroscopic
fission cross section for that material per source neutron.

Calculating neutrons per fission and recoverable energy per fission then involves dividing the

above two values by the macroscopic cross section. For a complex system of many different

fissionable actinides, in many different burnable regions, the major question, regarding the

calculation of the Qvalue and v, which now arises, is how to represent the normalization

parameters and apply them to the tallied reaction rates. Reaction rates and normalization

parameters are tallied within each bum material. If for every specific burnable region of a

system, the interaction rates were normalized explicitly by these calculated parameters within

that burnable region, it would not be possible to accurately transmute regions that did not contain

fissionable actinides. If the region did not contain fissionable actinides, the Qvalue for that

specific region would not exist forcing the normalization multiplier to be undefined.

In order to normalize the reaction rates, the code requires knowledge of the source of

neutrons. The source of neutrons is actually a global quantity because neutrons bor within one

burnable region migrate to other burnable regions for reaction. For example, consider a

geometry containing a single cylinder contained within a rectangular fluid of which posses a

reflective boundary condition at the boundary of the rectangle. At the center of the cylinder, 100

source histories are started and transported. If 60% of the histories are captured within the









cylinder and 40% leak out and interact in the surrounding material, then 60 neutrons are captured

within that cylinder or 0.6 neutrons per source are absorbed. Now let us assume that we now

have two cylinders of same dimension as the prior cylinder and separated in a manner so that the

pitch between each cylinder mimics the "pseudo" pitch implied from the prior calculation. A

rectangular fluid encases both cylinders and that fluid also posses a reflective boundary condition

at the boundary of the rectangle. For this new calculation we will still use 100 histories, but start

50 histories in each of the two spheres. Ignoring effects of statistical randomness of particle

travel, if the each cylinder employs the same material as the prior example then we would expect

that 60% of the neutrons emitted within each cylinder are then captured by each cylinder. Since

50 neutrons are emitted in each cylinder, each cylinder must then capture 30 neutrons. In the

first example, some neutrons were emitted from the cylinder into the fluid, hit the fluid

boundary, and then were scattered back into the cylinder for interaction. In the second example,

the same process happened as well as having neutrons emitted from one cylinder interacting with

the other cylinder. Because the source sites of neutrons for neutron interaction within each

cylinder may have been located within either cylinder, the neutron source is said to be global to

the entire geometry. If we now apply the same thinking to our calculation of flux normalization

parameters, these normalization parameters must also be tallied globally. This implementation

forces the normalization multiplier to be constant and representative of the entire system. For

each specified contributing material, MCNPX first tracks (n, v s ) and (n, Qvae s ) within

each specified contributing material, divides by the macroscopic fission cross section within each

material, and then volumetrically averages the quantities to generate globally averaged values for

neutrons per fission and recoverable energy per fission. These globally averaged values are then

combined to create the flux normalization multiplier in equation 3-11. The normalization









multiplier is then said to be the "global production rate" or "source" and multiplied by each

reaction rate per source to generate the total magnitude of the reaction rates used as destruction

and creation coefficients in the depletion solution. Therefore, it no longer maters if the material

that this normalization factor is applied contains actinides or not, for the same averaged

normalization constant is applied to reaction rate for each material. The normalization

parameters are recalculated and averaged at the end of each time step, so that the multiplier

dynamically adjusts to changes in temporal parameters as actinide concentrations are

created/depleted.79

Table 3-3. Emitted and recoverable energy per fission ofU-235.
Form Emitted Energy (MeV) Recoverable Energy
(MeV)
Fission Fragments 168 168
Fission Product Decay
Gamma-rays 8 8
Beta-rays 7 7
Neutrinos 12
Prompt Gamma Rays 7 7
Fission Neutrons (Kinetic 5 5
Energy)
Capture Gamma Rays -- 3-12
Total 207 198-207

The Qvalue used in the calculation is an estimate of the total recoverable energy per fission

event. The total recoverable energy per fission may be described as the sum of many mirco

events the lead to the recover of energy from a fission event. Table 3-3 lists the emitted and

recoverable energy per fission from each micro fission event for U-235.7 MCNPX inherently

contains an array possessing prompt energy per fission for certain fissionable actinides. Prompt

fission release energy is available within the evaluated data libraries, such as ENDF/B and JEFF,

for roughly 1/3 of all the actinides. In the ENDF/B based libraries, the breakdown of the data for

prompt fission energy release exists within File 1 MT 458.70 The breakdown of prompt fission









release data includes: prompt kinetic fission energy (EFR), prompt neutron emission energy

(ENP), prompt gamma energy (EGP), other delayed energy and neutrino energy contributions.70

Each actinide available within the prompt fission energy release array in MCNPX, as well as the

breakdown of prompt recoverable energy per fission if available, is listed in Table 3-4. The sum

of EFR, ENP and EGP for each actinide is the prompt fission release energy for that actinide.

Table 3-4. Prompt fission energy release data available in MCNPX.
Isotope EFR ENP EGP Total
90232 160.39000 4.41000 7.11000 171.91000
91233 163.50000 5.04000 7.03000 175.57000
92233 168.22000 4.90000 7.72000 180.84000
92234 167.10000 4.85000 7.50000 179.45000
92235 169.13000 4.91600 6.60000 180.64600
92236 167.50000 4.70000 7.30000 179.50000
92237 180.40000
92238 169.80000 4.80400 6.68000 181.28400
92239 180.40000
92240 169.57000 5.21000 6.53000 181.31000
93237 170.60000 5.94000 7.13000 183.67000
94238 173.60000 5.92000 7.13000 186.65000
94239 175.55000 6.07000 6.74100 188.36100
94240 173.70000 6.48000 6.18000 186.36000
94241 175.36000 5.99000 7.64000 188.99000
94242 174.00000 6.76000 5.22000 185.98000
94243 187.48000
95241 176.40000 6.53000 7.90000 190.83000
95242m 182.37000 4.05000 1.17210 187.59210
95243 176.30000 7.53000 6.42000 190.25000
96242 186.22000 5.20000 0.00000 191.42000
96244 178.50000 7.62000 4.37000 190.49000
Values listed are in MeV.

To properly account for all the energy per fission event the inherent prompt values must be

adjusted in order to account for delayed energy as well. A constant normalization factor, 1.111,

is multiplied by every Q-value calculated in MCNPX in order to account for delayed energy per

fission event. This constant normalization factor 1.111 was estimated based on the idea that if

multiplied by the prompt energy per fission of a U-235 fission event, the total amount of









recoverable fission, delayed and prompt, would equal 200 MeV.79 By implementing a constant

normalization factor, an approximation is made about the value of the capture gamma energy

contributing to the total recoverable energy per fission.

In a large LWR, most of the capture gamma energy is deposited in the heated volume and

thus contributes to recoverable energy; however, in a long, thin space reactor a lot of the capture

gamma energy leaks out of the system and does not contribute to recoverable energy of the

system. Therefore, the capture gamma energy contribution is highly dependent upon system

geometry and therefore the normalization constant may or may not estimate the actual Q value

properly. Underestimating the Q value leads to an overestimation of the system fluxes and thus,

an overestimation of the system collision densities. Because the equation for nuclide depletion is

highly dependent on the collision densities as displayed in equation 3-12, increasing the collision

density leads to an overestimation of the material burnup.

dN(r, t) E
dt I (r, t)E, (r, t)N, (r, t) (3-12)

In the current depletion capability, three methods are used to correct this overestimation

(input structure will be later explained in the Input Interface section):

1. Adjust the Q value multiplier (B value on the BOPT keyword; default = 1.0).
2. Modify the POWER keyword.
3. Alter the system power on the PFRAC keyword.

Unfortunately, the capture gamma contribution to the recoverable energy per fission is burnup

dependent. The following equation displays the how the capture gamma energy contribution

should be applied to the calculation of total recoverable energy per fission event:

Qrecoverable = prompt + Qdelayed + (v(E)- keff)* Qcapture Qneutnno (3-13)

Since neutrons per fission vary as a function of burnup, assuming a critical system is maintained

in the calculation, recoverable energy per fission event should also vary as a function of burnup.









In ENDF/B VII.0 only 207 out of 390 isotopes contain capture gamma release data. In

some cases, the data contain complete spectra information; however, in many cases the data are

incomplete.80 Without capture gamma photon emission spectra, employing equation 3-13 is

impossible because there is no way of determining how and where the emission energy is

deposited. The integral energy emission may be calculated from a mass defect calculation.

However, transporting a photon possessing the integral energy from the mass defect calculation

would not truly represent the physics of the system because the photon would travel farther and

deposit energy in different regions of the geometry as compared to the photons generated from

the true emission spectra. Nonetheless an initial calculation was completed to determine the

average integral capture gamma energy of all isotopes contained within the CINDER90 library

file. A mass defect calculation assuming zero energy neutron capture was completed to

determine the integral energy from a neutron capture event. The atomic mass data was taken

from the Atomic Mass Data Center (AMDC) website.88 The average integral energy from a

capture reaction with isotopes available in the CINDER library file was calculated to be 7.9 MeV

per capture. For fission products containing an MCNPX cross section and CINDER90 fission

yield information, the average integral energy from neutron capture was calculated to be 6.48

MeV. These calculated values only represent integral quantities based on 0 energy neutron

capture. Trying to track the true capture gamma energy deposition was deemed futile in

comparison to applying a delayed energy normalization constant because the true gamma ray

emission spectra were unknown. The integral energy deposition from the mass defect could be

assumed to be deposited locally; however, this assumption would violate the true physics of the

system. The implementation of this technique could theoretically at best only improve the

calculation of recoverable energy per fission by a few percent and at worst result in a more









inaccurate representation of the recoverable energy per fission as compared to the normalization

constant.

The normalization factor was thus chosen as the method for approximating the recoverable

energy per fission, and the following recommendations are suggested for trying to better account

for capture gamma energy contribution:

1. Photon emission spectra must be generated for all isotopes used in particle
transport and these photons should be transported to determine where each photon
deposits its energy, and then equation 3-13 may be used to adjust recoverable
energy per fission.

2. If the integral mass defect energy is to be employed for determining capture gamma
energy contribution, a qualified approximation must be derived for determining
how that energy is smeared within the geometry and what percent of that energy
escapes the heated volume contributing to system power.

However, since the normalization factor was chosen for implementation, the current recoverable

energy approximation will continue to slightly affect the flux normalization constant.

Automatic Fission Yield Selection

Fission products are generated as actinides consume neutrons and undergo fission events.

The fission product yield distributions from the fission events are a function of the isotope that

the causes fission event as well as the impinging neutron energy at the interaction site. For

thermal neutron interactions the fission product yield distributions follows the curves witnessed

in Figure 3-2 and Figure 3-3.

For higher energy reactions the fission yield distribution curve flattens and isotopes that

did not have a high probability of formation from a thermal reaction now have a higher

probability of formation in fast fission events. The differences in the fission products created

leads to adverse reactivity conditions; therefore accounting for the proper fission products that

are generated during a fission event is vital for calculation of reactivity and further interaction










rate distributions. Currently, the ENDF/B files offer a thermal, fast and high energy fission yield

distribution for various fissionable actinides.80

Table 3-5. Available actinide fission yield distributions in CINDER90.
Isotope Thermal Fast High Energy Spontaneous Fission
227Th x
229Th x
232Th x x
231Pa x
232U x
233 U
233U x x x
234 x x
235U x x x
235U


238U x x x
236U x x
237U
238
Sx x x
237P

Pu x
239x x x
Pu x x x
241pu x x
240
242u x x x
241Am x x x
242mAm
xAm x
243Am x
242Cm x

Am x
4Cm x
245Cm x
246Cm x x
248Cm x x
249Cf X
25Cf x
251Cf x
252Cf x
253Es x
254Es
254Fm
25Fm x
'Fm x
256Fm x


The CINDER90 code offers a thermal, fast, high-energy fission yield for each fissile

isotope contained in the CINDER90 data file.61 Table 3-5 displays the actinide containing









fission yield distributions, within certain energy bands, utilized in CINDER90.61 Since the

CINDER90 code is an energy integrated depletion code, and is incapable of determining spectral

characteristics, knowledge of the spectral characteristics must be input by the user. Because

proper fission yield selection involves predetermined knowledge of the neutron spectrum and

because this predetermined knowledge is possibly not known before the system is analyzed, a

preliminary calculation may be necessary to determine spectrum information. This preliminary

calculation contributes to the computational cost associated with the depletion process.

Selecting the adequate fission yield for a given system is dependent upon selecting the

specific yield that best represents the energy band containing the majority of the system's

missions. This selection process therefore depends upon the energy-dependent fission rates. The

MCNPX depletion code incorporates an algorithm to determine which energy band contains the

largest fission rate and then properly selects the correct fission yield.85 The MCNPX automatic

fission yield selection method is diagramed in Figure 3-15. MCNPX determines which fission

yield to use for a specified problem by calculating the integral fission rate for the defined energy

boundaries (thermal, fast, and high energy), determining which energy range contains the

majority of fissions, and then selecting the appropriate fission yield corresponding to the energy

range containing the majority of fissions. The method is versatile in that each individual burn

material may implement a separate fission yield if the calculation warrants a separate fission

yield for different geometric locations. By using this process, the end user does not have to

endure the computational cost associated with a preliminary calculation to determine the proper

fission yield to be used for the depletion process. The end user is also given confidence that the

fission yield being implemented for a specific problem is in fact the fission yield corresponding

to the energy-dependent fission rate of the analyzed system.8
























Figure 3-15. Fission yield selection process diagram

Time Dependent Cross Section Averaging Techniques

The depletion equation use interaction rates and nuclide densities to determine the time-

dependent nuclide inventory. Reexamining equation 2-1 we see that the depletion equation is

only linear if the depletion coefficients are constant; however, in reality the depletion coefficients

are also time-dependent. As the nuclide inventory changes over time, as a function of the

reaction rates, the interaction rates will also change as a function of time thus making the process

of nuclide depletion nonlinear. As discussed in Chapter 2, to make equation 2-1 linear and

solvable, the depletion coefficients are assumed constant over a given time step. Because a

certain amount of fluence is required in order to change the nuclide concentration significantly

enough to further alter the applied reaction rates, the assumption is valid for certain time

durations.2

Acceptable time durations are related to the equilibrium buildup or depletion of highly

absorbing isotopes. How the coefficients are applied within these time durations is paramount to

the accuracy of the calculation. Equation 3-14 illustrates how power is proportional to the

macroscopic fission cross section multiplied by the flux:

Power -~C, (3-14)









From a first order standpoint, as the fuel burns and the macroscopic fission cross section

decreases, the flux must increase in order to maintain constant power. Therefore if the depletion

coefficients calculated at to(initial time step) are utilized for a depletion to ta(subsequent time

step), the interaction rates implemented into the nuclide inventory calculation will be

underestimated in comparison to the average interaction rates; ultimately, leading to an under

burning of the system.

Making the depletion coefficients time-dependent in order to account for the true

interaction rate behavior makes the depletion equations nonlinear. Therefore if the coefficients

must be made constant in order to make the depletion equation solvable, then the average

interaction rates must be applied in order to attempt to adequately account for the true interaction

behavior. Unfortunately, the true average interaction rates for a given time step cannot be known

exactly, for that would involve solving the nonlinear equations exactly which is impossible

because the equations are nonlinear. However, a "pseudo" average reaction rate may be

calculated by applying appropriate assumptions related to how the interaction rate is expected to

behave. Many methods exist for approximating average coefficient behavior in the solution of

non-linear equations;46 74,89-91 however, for the method to be useful in a computationally

expensive linked Monte Carlo application, the method must implement as few extra Monte Carlo

calculations as possible. One general method for approximating this nonlinear behavior without

excessive computational expense is the predictor-corrector method. 46, 74, 89,9192 First, an

assumption is made about the time-integrated flux (setting it to a constant value) to deplete the

system over a time step and calculate a "predicted" value of the number densities, and interaction

rates. Values then are "corrected" by depleting the system again over the time step implementing

the newly calculated interaction rates. The hope of implementing such a calculation is to deplete










the system using a best representation of the average of the time-dependent parameters.

Predictor-corrector methods are defined as any type of routine in which a prediction is made

regarding behavior and then a further correction is made if that behavior is deemed to change.

Two cross section averaging techniques were investigated for inclusion in MCNPX. The

Cell-2 and MONTEBURNS cross section averaging methods were chosen based on the fact that

both codes were highly benchmarked and the methods incurred minimal additional

computational cost when applied in Monte Carlo.74' 91 The Cell-2 method follows a more

traditional predictor-corrector approach, and the MONTEBURNS method follows a slightly

different approach that still relies upon the same implied assumptions of the traditional predictor-

corrector approaches but requires less memory per calculation.

MONTEBURNS Method

Monte Carlo:
Time = t(i)
Initial collision densities/fluxes
Initial number densities (No)

CINDER90:
Depletion Calculation [t(i) t(i+1/2)]
Calculate New Number Densities


Monte Carlo:
Time = t(i+1/2)
Recalculate collision c

CINDER90:
Rerun Depletion Calculation [t(i) t(i+l)]
Use recalculated collision densities /fluxes
Calculate New Number Densities = N

For next sequence of time steps
No = N

Final Time Step

Sesg Done


Figure 3-16. MONTEBURNS cross section averaging technique.









The MONTEBURNS cross section averaging procedure is displayed in Figure 3-16.74 The

MONTEBURNS code first makes a predictor calculation of the nuclide inventory to time

t(i+1/2), based on interaction rates calculated at time, t(i). The interaction rates then are

recalculated using the nuclide inventory at t(i+1/2). It is then assumed that these newly

calculated interaction rates are the average interaction rates. Then a calculation is made over the

whole time step implementing the newly calculated average interaction rates to generate

corrected number densities. The entire process repeats until the final time step.

CELL-2 Method

Monte Carlo:
Time = t(i)
Initial collision densities/fluxes
Initial number densities = No


CINDER90:
Depletion Calculation [t(i) t(i+1)]
Calculate New Number Densities
Calculated number densities become
predictor number density = Np

Monte Carlo:
Time= t(i+l)
Recalculate collision ,/ i.i, r i,,


1
]^


CINDER90:
Rerun Depletion Calculation [t(i) t(i+l)]
Use recalculated collision densities /fluxes
Calculate New Number Densities
New number densities become corrector
number density = No

For next sequence of time steps
(NP + Nc)
2

Final Time Step No

FYes Done

Figure 3-17. CELL-2 Predictor-Corrector method.









The CELL-2 predictor-corrector method is displayed in Figure 3-17.91 The CELL-2 code

first makes a predictor calculation of the nuclide inventory to time, t(i+l), based on interaction

rates calculated at time, t(i). The interaction rates then are recalculated using the nuclide

inventory at t(i+l), and the system then is redepleted from t(i) to t(i+l) to generate new corrected

number densities. Finally, the predicted and corrected number densities are averaged to

determine the nuclide inventory at t(i+l). The entire process repeats until the final time step.

Preliminary Analysis

An infinitely reflected pin cell geometry was depleted at a power of 66.956 kWt for 2191

days (6 years), using 5000 particles per cycle for 300 cycles, skipping the first 15 cycles.92 The

three types of calculations examined were

1. MCNPX with no-predictor-corrector method (NoPC);

2. MONTEBURNS predictor-corrector method (PC1);

3. CELL-2 predictor-corrector method (PC2).

An initial test involved only burning for a 371 day period. The 371 day cycle was broken up into

6 steps for testing predictor corrector methods: 0, 0.654, 31, 91, 181, 271, and 371 days. A

MCNX calculation without implementing a predictor corrector method was run for 14 time

steps: 0, 0.645, 5.7, 16, 31, 46, 61, 76, 91, 136, 181, 271, 321 and 371 days. This case was run

as a control case, with the assumption that as the amount time-steps approached infinity the

calculation would approach true behavior. Neutron multiplication versus time for the examined

cases is displayed in Figure 3-18.

Unfortunately, it appeared that the examined cycle time was not long enough to incur a

significant difference from choosing to utilize the predictor corrector method or not, suggesting

that the 6 times steps implemented within the 371 day cycle was adequate enough to model

reaction rate behavior. A new set of tests were generated examining longer cycle times in order










to truly analyze the benefit of implementing the predictor corrector method. A 2191 day bum (3

cycles of 2 year burns) was split up into various burn steps and analyzed to determine which of

the two predictor corrector methods would behave better. NoPC was analyzed using 30, 20, 10,

and 5 time steps, whereas PC1 and PC2 were examined using 15, 10, and 5 time steps. Figure 3-

19 displays the bum steps implemented to achieve the 2191 day burn.


o 1.18

1.14
Z l 1.12
1.1
0 5 10 15
Burnup [GWD/MTU]

-- MONTEBURNS/ORIGEN -- MONTEBURNS/CINDER90
NoPC 6 steps PC1
-- PC2 -- NoPC 14 steps



Figure 3-18. Time dependant neutron multiplication for the predictor corrector method tests.

For these cases, the assumption was made that the 30 time step case was assumed to be an

infinite set of time steps. Therefore each subsequent case was compared to the 30 step case.

Figure 3-20 and Figure 3-21 display the time-dependent neutron multiplication versus burnup for

the different predictor corrector cases. Both predictor corrector methods, PC1 and PC2, calculate

similar neutron multiplications at each time-step, and the differences in actinide and fission

product production and depletion rates were also similar in both methods (>0.4% in EOL

production/depletion between both methods).

Figure 3-22 displays the percent difference in neutron multiplication of each case when

compared with the NoPC 30-step case. When either the PC 1 or PC2 5 step calculation (5 step

calculation = 5 predictor steps + 5 corrector steps = 10 total steps) was employed, there was a 3X

speedup and less than 1.55 % difference in the EOL keff. Therefore, for this specific geometry,











the conclusion may be made that both the PC1 and PC2 methods were equally acceptable


techniques for reducing the needed number of time steps to achieve an acceptable EOL keff.


Since this result suggested that both methods achieve acceptable answers, and since the


MONTEBURNS method employs less memory, because predictor number densities were not


saved during the corrector calculation, the MONTEBURNS method was chosen as the method of


choice for implementation into MCNPX.92


Burnup Steps
(GW DIM TU)
30 20 15 1
0.00E+00 0.00E+00 0.00E+00 0.00
2.48E-02 2.48E-02 2.48E-02 2.48
4.09E-01 4.09E-01 4.09E-01 1.94
7.93E-01 7.93E-01 1.18E+00 1.15
1.18E+00 1.18E+00 1.94E+00 2.11
1.56E+00 1.56E+00 6.74E+00 3.07
1.94E+00 1.94E+00 1 .35E+01 4.13
2.71E+00 4.82E+00 2.11E+01 5.19
3.86E+00 7.70E+00 2.88E+01 6.24
5.40E+00 1.15E+01 3.65E+01 7.30
6.94E+00 1.54E+01 4.42E+01 8.41
8.47E+00 2.11E+01 5.19E+01
1.00E+01 2.73E+01 5.95E+01
1.1 5E+01 3.34E+01 6.72E+01
1.39E+01 3.96E+01 7.49E+01
1.62E+01 4.57E+01 8.41E+01
1 .85E+01 5.19E+01
2.11E+01 5.95E+01
2.38E+01 6.72E+01
2.69E+01 7.49E+01
3.00E+01 8.41E+01
3.34E+01
3.73E+01
4.1 9E+01
4.69E+01
5.19E+01
5.76E+01
6.38E+01
7.03E+01
7.72E+01
8.41E+01


Figure 3-19. Bur steps implemented to achieve 2191 day burn


0
E+00
E -02
E+00
E+01
E+01
E+01
E+01
E+01
E+01
E+01
E+01


5
0.00E+00
2.48E-02
1.94E+00
2.11E+01
5.19E+01
8.41E+01













1.17

1.14 -

1.11

1.08

S1.05

1.02

0.99

0.96

fn 3 _


4
*4
'141


4


4 t
4
4


a
4 4-


0 20 40 60 80

Burnup (GWDIMTU)

26A 30 Steps PC1 5 Steps + PC2 5 Steps 26A 5 steps



Figure 3-20. Time dependent neutron multiplication utilizing 5 burn steps.


1.17

1.14

1.11

1.08

1.05

1.02

0.99

0.96

0.93


a


* t


0 20 40 60 80

Burnup (GWDIMTU)

26A 30 Steps PC1 10 Steps + PC2 10 Steps m 26A 10 steps



Figure 3-21. Time dependent neutron multiplication utilizing 10 burn steps.











3.50%
3.00%
a *
^ 2.50%
S2.00%
S1.50%
2 1.00%
a. 0.50% -
0.00%
51 551 1350 2191
Time (days)

o NoPC 20 Steps o NoPC 10 Steps m NoPC 5 steps
m PC1 15 steps B PC1 10 Steps a PC1 5 steps
o PC2 15 steps c PC2 10 Steps PC2 5 steps


Figure 3-22. Percent Difference in keff When Compared with the NoPC 30-Step Case

Further Commentary on both Cross Section Averaging Techniques

Both methods tested were not the exact original method employed within both codes.

Initially, the CELL-2 method only required a full spectrum calculation for the predictor

calculation.91 The method reuses the predicted number densities to propagate the spectrum

calculation at each time step.91 Corrected number densities are calculated using a similar

procedure as detailed in Figure 3-17 except that these new number densities are not implemented

in further propagation of the spectra calculation.91 These corrected number densities only serve

to update outputted values of what is to be believed the true number density at that time step.

Implementing the CELL-2 method in this way reduces the amount of needed spectra calculations

by half the amount. As initially implemented, the MONTEBURNS method only recalculates

spectra at the half time steps, and then uses that specta as the assumed average spectra to be

implemented over the entire time step. Implementing the MONTEBURNS method in this way

also reduces the amount of needed spectra calculations by half the amount.74 Initially, the









method was tested implementing the corrector spectra based on the premise that the corrector

spectra would be required for systems incurring significant concentration change of highly

absorbing isotopes. Later evidence from calculations with MCODE supported this premise for

BWR lattices containing heavy initial gadolinium loading.7 This evidence suggested that the

predicted spectra and corrected spectra differed significantly enough to warrant a significant

enough change in interaction rate to invalidate the applicability of the method for the desired

time step. Therefore either smaller time steps were required, therefore limiting the effectiveness

of both methods for long time steps, or the corrected number densities were required in order to

accurately propagate the temporal nuclide solution. This evidence supports that using the

corrected spectra, either from the calculated corrector number densities in the CELL-2 method or

from the recalculated spectra at the tO time steps in the MONTEBURNS methods, improves

nuclide predictability. Using the corrected spectra increase the length of acceptable time step to

a value beyond which would have been valid if only the predicted spectra were implemented;

however, the method still is dependent on the time-step choice. The corrector spectra calculation

does represent an extra calculation with extra incurred computational cost. For any depletion

system, there may exist an optimum set of time steps in which the extra corrector spectra

calculation employed for these time steps does not improve computational cost; however, these

optimum time steps are usually unknown prior to calculation. For a general purpose tool,

employing the extra corrector spectra calculation gives the user the best change for achieving

accuracy in calculation. Because the MCNPX depletion methodology paradigm was to be

applied to all types of depletion systems, possibly incurring the limiting cases given from the

literature, both methods were tested using the corrected spectra calculation, and the "qausi"

MONTEBURNS predictor corrector method was chosen for implementation in MCNPX 2.6.B.94









CHAPTER 4
IMPLEMENTATION ARCHITECTURE

The depletion capability is a modular methodology residing within a large scale MCNPX

code. As mentioned before, MCNPX is composed of -1000 subroutines. The depletion process

resides in roughly 30 subroutines in the MCNPX code. Five new self-contained subroutines

compose the majority of the actual variable averaging, outputting, and interfacing necessary for

the depletion methodology while the other 25 augmented subroutines simply allow new variables

to be inputted and set within the code package. The objective of confining the major

computations to a minimal amount of self-contained subroutines is to make the package modular

enough to be portable to earlier or different versions of MCNP. Self containment within minimal

subroutines also helps in code augmentation and testing, for if the capability is confined to a

small section of the larger code, then finding errors generated by the capability will be

centralized to a minimal amount of subroutines employed by the code. If the methodology was

not confined and self contained, debugging and coverage analysis would involve looking at

larger set of subroutines. Furthermore, without containment to a small set of subroutines,

augmentation to the methodology would result in needed coverage analysis for a larger

subroutine set of the code increasing the computational cost associated with development and

testing.

The MCNPX code requires the user to utilize input files in order to initialize variables for

computation. The input file structure includes50 (1) a title card; (2) cell cards defining a

geometry from unions and intersections of surfaces specified on the surface cards (cards

explaining certain physics options may also exist within in the cell cards section); (3) surface

cards defining base surfaces that are used to define volumes on the cell cards; (3) data cards

defining material, source and physics options available for implementation.










For each type of card defining source and physics options, keywords are available that determine

how the card is implemented. For example on the SDEF card, the keyword PAR with a particle

designator is used for determining the type of particle to be omitted from the source.50 Since the

depletion capability involves many types of options in order to burn a geometry model, the

depletion capability was setup in a similar manor as the SDEF card. The development of the

input interface and output structure spans MCNPX 2.6.A-F.79 84 86, 87,94,95 The input and output

features listed below, state the relevant capabilities of the most recent release of MCNPX 2.6.F.7

Input Interface

The depletion capability is invoked by using the BURN card. Within the BURN card exists

many different types of keyword options. Figure 4-1 displays the input structure for the BURN

card. Each input variable is then explained below Figure 4-1.

BURN TIME=Ti,T2,T, ...
PFRAC=Fi, F, F3, .
POWER=P
MAT=+/-M, +/-M2, +/-M3, .
OMIT=Ji,Ni,I11, 2,. .,J2,N2,I21,I 22,
AFMIN=A1 A2
MATVOL= V1, V2, V3, ..., Vn
MATMOD= ...
BOPT=B1,B2,B3
Figure 4-1. Burn card input structure.

The TIME keyword corresponds to the incremental time duration for each depletion step.

Each "Ti" listed in this keyword corresponds to a time duration (days) in which the system is

burned. The default for this keyword is 1 day. For example:

TIME = 10, 100, 15

The system is first burned for 10 days, followed by a 100-day bum, followed by a 15-day burn,

for a total burn of 125 days.

The PFRAC keyword corresponds to the fractional value of total system power (POWER) in

which to burn the system for the equivalent time duration. The PFRAC option only changes the









total magnitude of power used to normalize the flux, and does not account for possible

temperature changes in an actual power change. For each Ti value listed on the TIME keyword,

a corresponding Fi value listed on the PFRAC keyword must exist. For each Ti value that does

not contain a corresponding Fi value, Fi = 0 for that time step. The exception is when the PFRAC

keyword is not specified at all, and then each F = 1 for every time step.

The POWER keyword is the total system power level, P, (in megawatts). Because this value

corresponds to the total recoverable energy from the system, the value entered on this keyword

should correspond to the total recoverable thermal system power. The default for this keyword is

1MW. For example:

POWER = 100

In this case, the total recoverable thermal system power is assumed to be 100 MW.

The materials to be burned must be listed on the MAT keyword. Each "Mi" entry

corresponds to the material number listed in data cards section of the input deck. Figure 4-2

displays an example BURN card and subsequent materials cards referenced by the BURN card as

burn materials. In this example, materials ml, m3, and m4 will be burned.

Burnup is calculated for the entire system of materials listed on the MAT keyword, as well

as for each individual material listed on the MAT keyword containing a fissile actinide. Negative

material numbers signify materials that are not burned, but do contribute to the power

normalization by applying the tallied recoverable energy per fission and neutrons per fission of

those materials to the total power normalization. Actinide and non-actinide buildup/depletion

information for negative MAT numbers is not displayed in the output file because the negative MAT

numbered materials are not burned.










BURN TIME=100,70
MAT=1,3,4
POWER=1.0
PFRAC=1.0,1.0
BOPT=1.0 -12 1.0
C Material Cards
ml
8016.60c 4.5854e-2
92235.60c 1.4456e-4
92238.60c 1.9939e-2
94238.60c 1.1467e-4
94239.60c 1.0285e-3
94240.60c 7.9657e-4
94241.60c 3.3997e-4
94242.60c 5.6388e-4
m2 2004 -1.0
m3 40000.60c -1.0
m4 1001.60c 4.7716e-2
8016.60c 2.3858e-2
5010.60c 3.6346e-6
5011.60c 1.6226e-5
mt4 lwtr.0lt

Figure 4-2. BURN card referencing burn materials ml, m3 and m4.

The OMIT keyword allows the user to run a transport calculation without including the

transport data from a specified isotope. If an OMIT keyword is used, then the isotope is omitted

from the transport calculation and omitted from the output file. Reaction rates are still generated

for the isotope by importing a 63-group flux into CINDER90, matching the 63-group flux

calculated from MCNPX to a 63-group cross section set inherent within CINDER90, energy

integrating the reaction rates, and then producing a total collision rate for each collision type

tracked. The OMIT keyword has the has the format

OMIT=J1,N, I1, 112, ... J2N2, 21, 122,..

Table 4-1 contains the available options for this keyword.

Table 4-1. OMIT keyword variable tasks.
Input Variable Variable
Ji ith material for which to omit nuclides In, Ii2, etc.
Ni Number of omitted nuclides listed for the ith material
I i, Ii2, ... 1st, 2nd, etc., omitted nuclide for the ith material









If J =-1, then the omitted nuclide list is applied to all materials and J J etc., are not allowed.

There is no default for this option.

The AFMIN keyword allows the user to determine below which mass fraction an isotope

will be excluded from the transport calculation. The Al value ofAFMIN keyword signifies the

minimum atom fraction for all isotopes in which isotopes below that atom fraction will have

their atom fraction set to zero. The default for this option is 1.OE-10. The A2 value sets the decay

chain convergence criteria (mentioned in the Benchmarks section). The default for this value is

now le-10 and was le-4 for MCNPX 2.6.E and earlier. Using the new default improves nuclide

predictability as the cost of increased computation time.

The B1 value on the BOPT keyword corresponds to the recoverable energy per fission

multiplier. The default for this value is 1.0. The user may choose to adjust this value in order to

adjust the Q value to better account for the true recoverable energy in the system.

Table 4-2. Options invoked for different B2 values.
Input Value for B2 Output Inventory Ordering
1 High to low, based on mass (default)
2 High to low, based on activity
3 High to low, based on specific activity
4 Increasing zaid

The B2 value on the BOPT keyword corresponds to the selected fission products used in

the MCNPX transport process, as well as the format for the nuclide concentrations in the output

file. MCNPX reports nuclide concentrations only for isotopes that are listed on the material

cards, generated by the isotope generator algorithm, and/or selected from a fission product tier.

The B2 value represents the burn table output frequency, ordering, and content flag. Table 4-2

displays the different options invoke for different values of B2. Positive input values cause the

output to be printed to the output file only at end of an entire job. Negative input values cause

output at the end of each bum step.









Three tiers of fission-product content are available. Each tier corresponds only to those

isotopes that are used for particle transport in MCNPX and the nuclide densities, which are

reported in the output file. To select a specific tier, add zero (default), 10, or 20 to the magnitude

of the B2 input value. Table 4-3 gives a description of the available actinides in each fission

product tier.

Table 4-3. Options invoked for different B2 values.
Tier Fission Product Content
1 Zr-93, Mo-95, Tc-99m, Ru-101, Xe-131, Xe-134, Cs-133, Cs-137,
Ba-138, Pr-141, Nd-143, Nd-145 (default).
2 All fission products that posses both CINDER90 yield data and
transport cross data that are listed in the default MCNPX library file
for MCNPX 2.6.F (XSDIR file).
3 All 220 available fission products listed in Figure 3-8.

The B3 value of the BOPT keyword invokes the models option. Cross-section models are a

heritage of the high-energy physics capabilities of MCNPX. The models are benchmarked for a

variety of problems for neutron energies >150 MeV. Table 4-4 describes the available options

for the B3 value of the BOPT keyword.

Table 4-4. Available options for the B3 value of the BOPT keyword.
B3 value Available Option
-1 Receive a fatal error if cross section models are used in the problem
0 Zeros out the atom fraction of any isotope using a cross section
model (it is preferable to determine which isotopes are needed to be
omitted and to place them on the OMIT keyword for the burned
material).
1 Use cross section models in the calculation for isotopes not
containing tabular interaction rate data.

The MCNPX depletion capability tallies energy integrated reaction rates and normalization

constants within each separate bum material using the track length estimator. The track length

estimator of the flux is calculated by:

1 (1"* (4-1)
N[ V









Initially, MCNPX was setup to calculate and sum the quantity within the larger parenthesis

during each track for each individual bum material and then later divide by the total number of

histories. Due to limitations in the repeated structures mapping algorithm in MCNPX, the code

cannot always determine the total volume of a plethora of repeated cells. Excessive time is also

wasted trying to query and calculate the total volume. For simplicity, in MCNPX 2.6.D, the

tallying algorithm in equation 4-1 was altered to


Z = I (Y,,1*w,) (4-2)

Now only the particle track multiplied by the weight is tallied during each track for each

individual burn material. Once particle tracking is complete, the track length estimator is now

determined by dividing this new quantity by the sum of all volumes for a specific burn material

and the total histories. The user may then input the total burn material volume on the MATVOL

keyword. This procedure eliminates the dependency of the code to try and determine this total

volume when the code may or may not be able to determine this quantity or when the code may

require excess computation to map this value. Therefore the user is now free to take advantage

of the repeated structure geometry features, such as modeling repeated arrays of fuel pins, while

using the bumup capability.96

For burnup regions without repeated structures, MCNPX will try to calculate a volume

and, if unsuccessful, a fatal error is generated and the user is required to then specify the material

volume using a VOL card. In the case of cells in a lattice, MCNPX is only capable of

successfully accounting for the volumes of materials if each cell is used only once in the lattice.

Otherwise, if certain cells are repeated in a lattice, MCNPX is unable to determine the total

volume of everywhere the material is repeated. The MATVOL keyword is used to account for the

volume of materials listed in a repeated structure









MATVOL = Vi, V2, V3,... Vn,

where V, is the total volume of all cells [cm3] containing burn material, M,, on the MAT keyword

of the BURN card.

Table 4-5. MATMOD keyword variable values.


Keyword Value
NT
TSNT


Description
Number of time steps (1 through NT)
Time step (1..NT) for which to manually change nuclide concentrations of
material MTNT,NMNT Enter "1" for 2nd, etc. (If positive apply concentrations
discretely at Ti and Ti+1/2. If negative apply at Ti, and make linear
interpolation between Ti and Ti+l to determine the concentration at Ti+1/2. If
TS is negative at Ti and the concentrations of any of the altered isotopes at
Ti+1 is equal to the concentration set at Ti then the concentrations of the
altered isotopes will be set to the value at Ti for Ti, Ti+1/2, Ti+l. At Ti+3/2 the
isotopes will undergo a normal depletion and the concentrations will not be
set to the value at Ti+l.
Number of materials at time step "NT" that incur nuclide concentration
changes
NMth material number for which to manually change nuclides at time step
"NT". Positive value indicates atom/wt. fraction and negative value
indicates atom/gram densities.
Number of nuclides to manually change for the NMth material
1st, 2nd,.. Kth nuclide of the MTNT,NMNT th material at time step "NT" for
which a concentration will be specified. List as a ZA value.
Concentration for the Kth isotope in material MTNT,NMNT at time step "NT".
Enter positive values for atom fractions or atom densities, and enter a
negative value for wt. fractions or gram densities. See sign ofMTNT,NMNT to
specify either fraction or density


If the MATVOL keyword is utilized on the BURN card then there must exist a corresponding

MATVOL entry for every entry on the MAT keyword of the BURN card. If repeated structures are

used without a MATVOL keyword, MCNPX will incorrectly calculate the volume, if possible,

assuming that the total volume for the entire burn material is equal to the volume stated on the

VOL card and not the accumulated the sum for the amount of times the cell is repeated.

The MATMOD keyword allows for time-dependent manual nuclide concentration changes.

The details of the methodology for this feature were explained in the Manual Time-Dependent


NMNT

MTNT,NMNT


KNT, NMINT
NT ,NMNT

C NT,' NT
z .

C









Isotope Concentration Changes section. Table 4-5 describes the detail of each value within the

MATMOD keyword. The MATMOD keyword uses the following structure for describing a manual

concentration change:

MATMOD = NT TS, NM1 MT1,1 K1,1 Z C Z2 C CK'


MTiM K, Z C Z c Z C'
MT1, NM1 K1, NM Z C1 Z C Z. NMC C N"
1 ,NM1 1,NM1 1,NM1 1,NM 1,Nl 1 Nl

TSNT Ti1 MTNT,1 KNT,I Z1 C1 Z2 C2 ZK. CK
NT,1 NT,1 NT,1 NT,1 NT,1 NT,1


MTNT, KNT Z1 C1 Z2 C2 zKNT.'.NT cK..NT,'NT
'NT,NMNT NT,NMNT NT,NMNT NT,NMNT NT,NMNT NT,NMNT

When using the MATMOD keyword, if a burn material is set to have a concentration change at

T1, the atom density of that isotope at T1/2is set to the initial value specified at To. This is only

set for the initial midpoint time step, the rest of the calculation will follow the procedure

mentioned above.

Output Structuring

The MCNPX depletion capability offers many varied outputs describing the temporal

nuclide inventory process. Initially, the objective of the depletion outputs was to give the user

the maximum amount of information necessary in order to make informed system design

decisions from understanding important parameters regarding the evolution of the temporal

nuclide inventory solution. The depletion outputs may be segregated into five distinct sections.

These output sections include collision rates sent to CINDER90 for use as depletion coefficients,

generalized global system averaged depletion data, generalized individual material depletion

data, individual material nuclide data, and system average nuclide data. Each of the output

sections is composed of key important information regarding the temporal nuclide inventory.










Total continuous energy integrated collision rates within a each burn volume are calculated

in the particle transport calculation for (n,fission), (n,y), (n,2n), (n,3n), (n,p) and (n,a).

Examining these collision rates is important to understanding how the depletion system operates.

The collision rates for each burn material, tracked in particle transport to be sent to CINDER90

for use as depletion coefficients are printed at the end of each steady-state KCODE calculation.

The error associated with each collision rate is the error assuming the steady-state calculation is

independent from the propagated error resulting from propagated flux and number density error

from the burnup calculation. Figure 4-3 displays a cropped example of the format of the

collision rate outputs. These outputs help the user to calculate generalized capture-fission ratios,

as well as other inelastic reaction ratios, to determine system characteristics.

Material #: 1

no. zaid (n,garrma) error (n,fission) error (n,nu*fission) error (n,fission*Q) error
1 6012 8.19001E+05 0.5671 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000
2 6013 2.81939E+08 0.1330 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000
3 6014 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000


60 96244 8.41178E+13 0.4713 5.60552E+12 0.1094 2.09288E+13 0.1034 1.18632E+15 0.1094
61 96245 4.07591E+11 0.0850 2.84278E+12 0.0738 1.02719E+13 0.0737 5.68499E+14 0.0738

no. zaid (n,2n) error (n,3n) error (n,alpha) error (n,p) error
1 6012 0.00000E+00 0.0000 0.00000E+00 0.0000 7.08058E+05 0.6535 0.00000E+00 0.0000
2 6013 3.16027E+08 0.9128 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000
3 6014 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000


60 96244 9.03869E+09 0.6024 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000
61 96245 4.49267E+08 0.4078 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000


Figure 4-3. Example collision rate outputs.

The burnup summary table contains global system averaged burnup information and is

designated "print table 210". Each time step is listed with the corresponding time duration and

actual specified depletion time. The following system averaged quantities are also listed: the

power used for the flux normalization, keff, energy integrated flux, neutrons per fission,










recoverable energy per fission, burnup, and production rate. Figure 4-4 displays an example

burnup summary table.

Ibumup summary table by material print table 210


neutronics and bumup data

step duration time power keff flux ave. nu ave. q burnup source
(days) (days) (MW) (GWd/MTU) (nts/sec)
0 0.000E+00 0.000E+00 1.000E+00 1.54021 7.715E+14 2.452 200.979 0.000E+00 7.616E+16
1 5.000E+01 5.000E+01 1.000E+00 1.50987 7.945E+14 2.473 201.411 7.183E+00 7.664E+16
2 1.000E+01 6.000E+01 0.000E+00 1.51150 0.000E+00 2.474 201.448 7.183E+00 0.000E+00
3 5.000E+02 5.600E+02 2.000E-01 1.43413 1.699E+14 2.510 202.199 2.155E+01 1.550E+16


Figure 4-4. Example burnup summary table.


Individual Material Bumup

Material #: 1

step duration time power fraction bumup
(days) (days) (GWd/MTU)
0 0.000E+00 0.000E+00 5.015E-01 0.000E+00
1 5.000E+01 5.000E+01 5.016E-01 7.205E+00
2 1.000E+01 6.000E+01 5.002E-01 7.205E+00
3 5.000E+02 5.600E+02 5.002E-01 2.158E+01

Material #: 4

step duration time power fraction bumup
(days) (days) (GWd/MTU)
0 0.000E+00 0.000E+00 4.985E-01 0.000E+00
1 5.000E+01 5.000E+01 4.984E-01 7.161E+00
2 1.000E+01 6.000E+01 4.998E-01 7.161E+00
3 5.000E+02 5.600E+02 4.998E-01 2.152E+01

Figure 4-5. Example individual material burnup data.

If multiple materials are burned then there also exists individual material burnup data.

Figure 4-5 displays example individual material bumup data. For this example both materials 1

and 4 were burned and therefore individual bum material burnup data is available only for those

materials. The available information includes: time step, time duration, actual time, fission

power fraction and individual material burnup. The fission power fraction is calculated by










taking the ratio of the fission power in a particular bum material to the sum of all burn materials.


The calculated fission power fraction is therefore only related to missions in bum materials.


power fraction = ( V)



The individual material burnup is calculated by


Burnup = Burnup, +


POWER POWER Fraction TIME PFRAC
MTHM


nuclide data are sorted by increasing zaid for material 1 volume 3.8457E+02 (cm**3)

actinide inventory for material 1 at end of step 0, time 0.000E+00 (days), power 1.000E+00 (MW)


no. zaid mass activity
(gm) (Ci)
1 90231 0.000E+00 0.000E+00
2 90232 0.000E+00 0.000E+00


spec.act. atom den. atom fr. mass fr.
(Ci/gm) (a/b-cm)
0.000E+00 0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00 0.000E+00


6 92235 3.441E+02 0.000E+00 0.000E+00 2.293E-03 1.000E-01 9.886E-02


actinide inventory for material 1 at end of step 1, time 5.000E+01 (days), power 1.000E+00 (MW)


no. zaid mass activity
(gm) (Ci)
1 9023 1.286E-09 6.837E-04
2 90232 2.394E-08 2.625E-15


spec.act. atom den. atom fr. mass fr.
(Ci/gm) (a/b-cm)
5.315E+05 8.718E-15 3.832E-13 3.723E-13
1.097E-07 1.616E-13 7.100E-12 6.929E-12


totals 3.455E+03 2.584E+05 7.479E+01 2.275E-02 1.000E+00 1.000E+00


nonactinide inventory for material 1 at end of step 0, time 0.000E+00 (days), power 1.000E+00 (MW)


mass activity
(gm) (Ci)
0.000E+00 0.000E+00
0.000E+00 0.000E+00
0.000E+00 0.000E+00
0.000E+00 0.000E+00
4.684E+02 0.000E+00


spec.act. atom den. atom fr. mass fr.
(Ci/gm) (a/b-cm)
0.000E+00 0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00 0.000E+00
0.000E+00 0.000E+00 0.000E+00 0.000E+00
0.000E+00 4.585E-02 1.000E+00 1.000E+00


Figure 4-6. Example cropped actinide output at two separate timesteps and fission product
output for one time step.


(4-3)


(4-4)


no. zaid


6012
6013
7014
7015
8016










The time dependant isotope buildup/depletion is listed after the burnup summary

information. The isotope buildup/depletion for each individual material is given at each time

step. The information is further subdivided into actinide and nonactinide categories. Figure 4-6

displays a cropped example of the reported actinide inventory for two separate time steps. At

the end of each subdivision is an accumulation total of the isotope information for that

subdivision. Atom and weight fractions calculations are based on the fractions of that specific

subdivision. After isotope information for each individual material is given, print table 220 lists

the sum total information of all actinides and nonactinides from all materials combined for each

time step. Figure 4-7 shows a cropped example of print table 220.



Iburnup summary table summed over all materials print table 220

nuclides with atom fractions below 1.000E-32 for a material are zeroed and deleted from print tables after t-0

nuclide data are sorted by increasing zaid summed over all materials volume 7.6914E+02 (cm**3)

actinide inventory for sum of materials at end of step 0, time 0.000E+00 (days), power 1.000E+00 (MW)

no. zaid mass activity spec.act. atom den. atom fr. mass fr.
(gm) (Ci) (Ci/gm) (a/b-cm)
1 90231 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00
2 90232 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00
3 90233 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00
4 91233 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00
5 92234 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00
6 92235 6.883E+02 0.000E+00 0.000E+00 4.585E-03 1.000E-01 9.886E-02

Figure 4-7. Cropped example of print table 220.

The output is structured so as to give the user maximum amount of information in well

organized concise tables. Each concise table is located in the output in the order in which the

calculations have been made. For example, the collision rate tables always follow the KCODE

information, and the burnup summary tables are then located directly below the collision rate

tables (or at the end of the entire output if so invoked by an output option). For a large burnup









example, using many bum materials, the user will generate huge output files as a result of

reporting all of the information available in the bumup tables. However, this information is

necessary for determining how derivative quantities, such as reaction rates, contribute to the

integral solution, such as global time dependant neutron multiplication and isotope analysis.









CHAPTER 5
BENCHMARKS

The Nuclear Energy Agency (NEA) is a specific agency within the Organization for

Economic Cooperation and Development (OECD), an intergovernmental organization composed

of many industrialized countries, based in Paris, France. The mission of the NEA is to assist its

member countries in maintaining and further developing, through international co-operation, the

scientific, technological and legal bases required for the safe, environmentally friendly and

economical use of nuclear energy for peaceful purposes.97 The membership consists of 28

countries including: Australia, Austria, Belgium, Canada, Czech Republic, Denmark, Finland,

France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Japan, Luxembourg, Mexico, the

Netherlands, Norway, Portugal, Republic of Korea, Slovak Republic, Spain, Sweden,

Switzerland, Turkey, the United Kingdom and the United States.97 The NEA is further divided

into many working parties and these working parties are composed of expert working groups that

are tasked with examining various aspects of the Nuclear Fuel Cycle.97 Criticality safety aspects

of the Nuclear Fuel Cycle are examined by the Working Party on Nuclear Criticality Safety

(WPNCS), and understanding the reactivity and isotope prediction capability is the responsibility

of the Burnup Credit Criticality Safety expert group. One of the objectives of the Burnup Credit

Criticality Safety expert group is to examine burn-up credit as applied to criticality safety in the

transportation, storage, and treatment of spent fuel for a wide range of fuel types, including

urainium-oxides (UOX) and MOX fuels for PWRs, BWRs, and Vodaa Vodiannee

Energititscherski Reactors (VVER). Achieving this objective involves carrying out international

comparisons and benchmarks in order to assess the capability of code systems to accurately

predict reactivity and temporal nuclide inventories, including both code-to-code and available

experimental data comparisons.97









In order to asses the validity of the MCNPX depletion method three separate types of

OECD/NEA benchmarks were completed. Table 5-1 lists the benchmarks that were used to test

the MCNPX depletion capability. In most of the benchmarks, many combinations of operating

techniques were investigated. For the purpose of this study, only benchmarks that employ the

maximum amount of fission products following all transmuted actinides were investigated.

Therefore only a subset of the full amount of cases analyzed in the complete benchmark were

examined. However, the analyzed sections of these benchmarks do adequately test the ability of

the code to predict real core behavior as a maximum amount of fission products are generated

during real reactor behavior. It is important to realize that the specified benchmarks only truly

test the codes predictive capability for certain LWR operating strategies. However, before one

tries to validate a methodology for more complicated operating scenarios, one must be able to

first also correctively predict behavior of well benchmarked LWRs.

Table 5-1. Examined benchmark cases.
Benchmark Case Geometry Type
OECD/NEA Burnup Credit Phase IB52 PWR UO2 adjusted pin cell calculation
H. B. Robinson53 PWR U02 assembly calcation
OECD/NEA Burnup Credit Phase IVB54 LWR MOX assembly calculation


OECD/NEA Burnup Credit Phase IB Benchmark Specification

The purpose of the OECD/NEA Burnup Credit Calculational Criticality Benchmark Phase

IB is to compare computed nuclide concentrations for a simple pin-cell depletion model.52 Prior

radiochemical analysis was completed on a 14 X 14 Combustion Engineering fuel assembly in

order to determine nuclide concentrations for use in the benchmark. The simple pin cell model

incorporates actual pin dimensions from a Combustion Engineering assembly; however, the fuel

pin pitch has been adjusted in order to mimic the fuel-to-moderator ratio exhibited in the full

assembly. Further approximations applied to the analyzed geometry included modeling the fuel









pin as an infinitely reflected geometry. An infinitely reflected geometry assumes zero net

leakage at the geometry boundary. This type of boundary condition is only incurred within a flat

power distribution region of the reactor. Such a region is never incurred on a pin cell level and

therefore the approximation of infinitely reflected geometry was expected to affect the final

solution outcome.



Borated Water

Clad

Fuel

Gap






Figure 5-1. Infinitely reflected pin cell geometry.

Table 5-2. Model description for Phase IB Benchmark*.
Parameter Data
Type fuel pellet U02
Rod pitch 1.5586 cm
Clad outer diameter 1.118 cm
Clad inner diameter 0.986 cm
Fuel diameter 0.9563 cm
Active fuel length 347.2 cm
Effective fuel temperature 841 K
Clad temperature 620 K
Water temperature 558 K
*Data taken from reference.

The pin cell model was divided into four radial regions. The pin geometry is displayed in

Figure 5-1 and the fuel pin dimensions are listed in Table 5-2. Temperatures were assumed

constant throughout each region. The fuel pin was composed of a tall cylindrical fuel region

divided into ten axial regions implementing the same average temperature and volume in each









axial division. Surrounding the fuel region was a cylindrical voided gap region. Enclosing the

fuel and gap regions was a cylindrical Zicaloy-2 clad region. Finally encasing the entire

geometry was a rectangular borated water box. The fuel, clad, and borated water concentrations

are listed in Table 5-3, Table 5-4 and Table 5-5. The compositions are displayed in the same

units as listed in the benchmark report.52 Three different pin burnup cases were analyzed. For

each burnup case, four operating cycles were implemented using common burn times, down

times and operating concentrations. Table 5-6 details the common operating strategies utilized

for each case. Table 5-7 lists the operating powers utilized for each individual case in order to

achieve the prescribed burnups.

Table 5-3. Fuel composition for Phase IB Benchmark (density =10.045 g/cm3).
Nuclide Number Density
(atoms/b-cm)
234U 6.15164E-6
235U 6.89220E-4
236U 3.16265E-6
238U 2.17104E-2
12C 9.13357E-6
14N 1.04072E-5
160 4.48178E-2

ENDF/B VII.0 temperature dependent cross sections and scattering law data were

implemented in the steady state reaction rate calculation. Each steady-state reaction rate

calculation was executed using 3000 particles per cycle for 115 cycles skipping the first 5 cycles.

Ten starting source locations were equally spaced axially along the fuel pin center. The

combination of skipped cycles and source locations was hypothesized to result in an adequately

converged fission source distribution, thereby resulting in a further converged reaction rate

calculation. The amount time steps, particles per cycle, and cycles were minimized in order to

achieve a depletion solution within a minimal amount of time. Running on a linux platform

across 25 1.4 GHz Opteron processors, each case ran for -22 hours.









Table 5-4. Clad composition for Phase IB Benchmark (6.44 g/ cm3).
Nuclide Weight Fraction
54Fe 0.0275 %
56Fe 0.4477 %
57Fe 0.0105 %
58Fe 0.0143 %
90Zr 49.0926 %
91Zr 10.8251 %
92Zr 17.8204 %
94Zr 17.3218 %
96Zr 2.8501 %
112Sn 0.0145 %
114Sn 0.0101%
115Sn 0.0052 %
116Sn 0.2257 %
117Sn 0.1203 %
11Sn 0.3825 %
119Sn 0.1368 %
120Sn 0.5232 %
122Sn 0.0756 %
124Sn 0.0961%

Table 5-5. Borated water composition for Phase IB Benchmark (0.7569 g/ cm3).
Nuclide Number Density
(atoms/b-cm)
H 5.06153E-2
160 2.53076E-2
10B 2.75612E-6
11B 1.11890E-5

Table 5-6. Operating history data for Phase IB Benchmark.
Operating Cycle Bum time Burn Duration Downtime Boron
(days) (days) (days) Concentration
(ppm)
1 306.0 1, 15, 30, 40, 50, 71.0 331.0
50, 60, 60
2 381.7 1.7, 20, 35, 45, 83.1 469.7
55, 65, 75, 85
3 466.0 6, 20, 45, 65, 70, 85.0 504.1
80, 85, 95
4 461.1 6.1, 20, 45, 60, 1870.0 492.5
70, 80, 85, 95









Table 5-7. Operating power (MW) for each case for Phase IB Benchmark.
Operating Cycle Case A Case B Case C
(final burnup = (final burnup = (final burnup =
27.35 GWD/MTU) 37.12 GWD/MTU) 44.34 GWD/MTU)
1 0.038068 0.054585 0.068717
2 0.042904 0.059090 0.071786
3 0.037627 0.050434 0.057853
4 0.032172 0.041667 0.048844


OECD/NEA Burnup Credit Phase IB Results

Table 5-8. Initial Phase IB Benchmark results and comparison of Case A (27.35 MWD/MTU).


Isotope MCNPX
Value
(mg/g UO2)
234U 1.408E-1
235U 7.988
236U 3.227
238U 8.375E+2
237Np 2.764E-1
238Pu 9.046E-2
239Pu 3.790
240Pu 1.599
241Pu 6.491E-1
242Pu 2.952E-1
241Am 2.274E-1
243Am 4.487E-2
95Mo 5.501E-1
99Tc 5.397E-1
133Cs 8.012E-1
135Cs 3.535E-1
143Nd 5.752E-1
145Nd 4.886E-1
147Sm 1.522E-1
149Sm 1.628E-3
15sSm 1.762E-1
151Sm 7.070E-3
152Sm 7.940E-2
153Eu 7.022E-2
155Gd 2.629E-3


Experiment
Value
(mg/g UO2)
1.600E-1
8.470
3.140
8.425E+2
2.680E-1
1.012E-1
4.264
1.719
6.812E-1
2.886E-1
N/A
N/A
N/A
N/A
8.500E-1
3.600E-1
6.130E-1
5.100E-1
N/A
2.900E-3
2.070E-1
N/A
8.700E-2
7.900E-2
N/A


*(Calculated/Measured-1)* 100


% Error* Range of Values from other Codes


-12.00
-5.69
2.77
-0.59
3.12
-10.61
-11.12
-6.99
-4.71
2.30
N/A
N/A
N/A
N/A
-5.74
-1.80
-6.16
-4.19
N/A
-43.86
-14.86
N/A
-8.74
-11.12
N/A


1.330E-1 to 1.750E-1
7.445 to 8.661
3.128 to 3.540
8.637E+2 to 8.415E+2
2.527E-1 to 3.396E-1
5.721E-2 to 1.083E-1
3.660 to 4.690
1.573 to 8.60
5.310E-1 to 7.335E-1
2.000E-1 to 3.192E-1
2.269E-1 to 2.598E-1
3.480E-2 to 4.672E-1
5.590E-1 to 5.795E-1
5.648E-1 to 6.904E-1
6.820E-1 to 8.640E-1
3.728E-1 to 3.959E-1
6.040E-1 to 6.792E-1
4.984E-1 to 5.151E-1
1.564E-1 to 1.932E-1
1.626E-3 to 2.900E-3
1.713E-1 to 2.146E-1
6.376E-3 tol.413E-2
7.947E-2 to 1.073E-1
6.730E-2 to 8.921E-2
1.507E-3 to 5.762E-3










The Burnup Credit Phase IB benchmark contains data from a variety of calculation tools,

as well as experimental results from radiochemical analysis of the 14 X 14 Combustion

Engineering fuel assembly, for certain key actinides and fission products. The initial MCNPX

calculated nuclide concentrations for all three benchmark cases are displayed in Table 5-8, Table

5-9 and Table 5-10. For each nuclide the following quantities are listed: MCNPX calculated

value, experimentally determined value, percent error between MCNPX calculated value and

experimental value and range of calculated values from other various methodologies presented in

the benchmark.

Table 5-9. Initial Phase IB Benchmark results and comparison of Case B (37.12 MWD/MTU).


MCNPX
Value
(mg/g UO2)
1.148E-01
4.587
3.641
8.303E+02
3.980E-01
1.727E-01
3.849
2.057
8.379E-01
5.916E-01
2.853E-01
1.181E-01
7.066E-01
6.898E-01
1.022
3.793E-01
6.523E-01
6.112E-01
1.607E-01
1.653E-03
2.311E-01
7.397E-03
9.521E-02
9.940E-02
4.076E-03


Experiment


Value
(mg/g UO2)
1.400E-01
5.170
3.530
8.327E+02
3.560E-01
1.893E-01
4.357
2.239
9.028E-01
5.761E-01
N/A
N/A
N/A
N/A
1.090
4.000E-01
7.160E-01
6.530E-01
N/A
3.000E-03
2.710E-01
N/A
1.040E-01
1.090E-01
N/A


*(Calculated/Measured-1)* 100


% Error* Range of Values from other Codes


-17.96
-11.28
3.15
-0.28
11.79
-8.77
-11.66
-8.14
-7.19
2.69
N/A
N/A
N/A
N/A
-6.24
-5.17
-8.90
-6.41
N/A
-44.91
-14.72
N/A
-8.45
-8.81
N/A


1.080E-1 to 1.570E-01
4.022 to 5.510
3.526 to 3.930
8.292E+02 to 8.360E+02
3.619E-01 to 4.919E-01
1.144E-01 to 2.069E-01
3.710 to 4.877
1.996 to 2.347
7.510E-01 to 9.846E-01
4.200E-01 to 6.347E-01
2.880E-01 to 3.418E-01
9.637E-02 to 1.391E-01
7.214E-01 to 7.545E-01
7.327E-01 to 8.372E-01
8.784E-01 to 1.117
3.967E-01 to 4.317E-01
7.013E-01 to 8.254E-01
6.326E-01 to 6.600E-01
1.659E-01 to 2.201E-01
1.736E-03 to 3.092E-03
2.297E-01 to 3.152E-01
7.990E-03 to 1.571E-02
9.761E-02 to 1.416E-01
9.960E-02 to 9.960E-02
2.538E-03 to 1.028E-02


Isotope


234U
235U
236U
238U
237Np
Np
238Pu
239Pu
240Pu
241Pu
242Pu
241Am
243Am
95Mo
99Tc
133Cs
135Cs
143Nd
145Nd
147Sm
149Sm
150Sm
15Smm
152Sm
153Eu
155Gd









Table 5-10. Initial Phase IB Benchmark results and comparison of Case C (44.34 MWD/MTU).
Isotope MCNPX Experiment % Error* Range of Values from other Codes
Value Value
(mg/g UO2) (mg/g UO2)
234U 9.896E-02 1.200E-01 -17.53 9.030E-02 to 1.440E-01
235U 2.897 3.540 -18.18 2.389 to 3.716
236U 3.772 3.690 2.21 3.641 to 4.030
238U 8.243E+02 8.249E+02 -0.07 8.234E+02 to 8.316E+02
237Np 4.794E-01 4.680E-01 2.44 4.327E-01 to 5.934E-01
238Pu 2.399E-01 2.688E-01 -10.76 1.656E-01 to 2.810E-01
239Pu 3.832 4.357 -12.04 3.659 to 4.902
24Pu 2.292 2.543 -9.86 2.180 to 2.661
241Pu 9.229E-01 1.020 -9.52 8.560E-01 to 1.111
242Pu 8.479E-01 8.401E-01 0.93 5.960E-01 to 9.103E-01
241Am 3.071E-01 N/A N/A 3.102E-01 to 3.785E-01
243Am 1.924E-01 N/A N/A 1.634E-01 to 2.316E-01
95Mo 8.100E-01 N/A N/A 8.092E-01 to 8.742E-01
99Tc 7.852E-01 N/A N/A 8.449E-01 to 9.861E-01
133CS 1.160 1.240 -6.45 9.723E-01 to 1.286
135Cs 3.958E-01 4.300E-01 -7.94 3.977E-01 to 4.605E-01
143Nd 6.746E-01 7.630E-01 -11.58 7.397E-01 to 8.839E-01
145Nd 6.858E-01 7.440E-01 -7.82 7.170E-01 to 7.560E-01
147Sm 1.598E-01 N/A N/A 1.655E-01 to 2.302E-01
149Sm 1.686E-03 4.700E-03 -64.13 1.842E-03 to 3.286E-03
15Sm 2.690E-01 3.610E-01 -25.49 2.725E-01 to 3.980E-01
15Sm 7.585E-03 N/A N/A 8.102E-03 to 1.682E-02
152Sm 1.038E-01 1.210E-01 -14.25 1.077E-01 to 1.587E-01
153Eu 1.173E-01 1.480E-01 -20.75 1.210E-01 to 1.596E-01
155Gd 5.046E-03 N/A N/A 3.379E-03 to 1.318E-02
*(Calculated/Measured-l)* 100

MCNPX initially had poor nuclide prediction. The predictive capability also seemed to get

worse at higher bumups. MCNPX was able to calculate actinide prediction within the range of

other depletion codes; however, the actinide predictions were not close to the measured values

and the percent error between measured and calculated values increased with increasing bumup.

MCNPX fission product predictions were outside of the range values predicted by other

depletion codes, and the fission product predictions were also not close to the measured values.

The percent error between measured and calculated fission product values also increased with

increasing burnup. Because the MCNPX results were poorer at higher bumups, it was









hypothesized that the burnup per time step might affect the accuracy of the result. Limitation of

the bumup per time step is function of both significant changes in reaction rate over a time step,

and limitations in the actual depletion algorithm. It is true that common time steps were used for

each burnup case, and therefore the burnup step durations were unequal for all three cases.

Therefore for the higher bumup cases it is possible that the time step duration was too long and

too much bumup occurred ultimately invalidating the assumption that significant change in flux

had not occurred. However, since the predictive capability was so poor compared to a

deterministic solver, the hypothesis that the CINDER90 algorithm may be faulted was tested.

CINDER90 uses a Linear Markov Chain method in order to solve the depletion equations.

CINDER90 solves for nuclide densities by following a series of equations along the

transmutation chain.61' 50 The following procedure is used:

1. Linear chains are created for each isotope transmutation path.

2. The solutions of each linear chain determines a partial nuclide density

3. Each calculated partial nuclide density, computed from a linear chain, is then
summed to obtain the total nuclide inventory of nuclide.

The differential equation governing the computation of the ith nuclide is therefore only coupled

to preceding contributing nuclides leading to the creation of the ith nuclide. However, in order to

stop the propagation of a chain, the code determines if significant progeny will be created; and if

the current nuclide does not produce significant progeny the calculation of the chain ceases. A

series of checks exist within CINDER90 to determine if progeny will be created, passby

calculation, and these checks posses limiting checking values for determining significance. For

example, the magnitude of the nuclide density of a given isotope and the integral of the creation

coefficient must both be above a certain value or the code ceases the transmutation chain. This

would appear to affect short half-live nuclides in secular equilibrium and all resulting nuclides









within that transmutation chain. Preliminary investigation suggested that the order of such

checks and the magnitude of the limiting checking values determined how accurately the code

determined if significant progeny existed thereby ultimately affecting the solution outcome.

Table 5-11. MCNPXc Phase IB Benchmark results and comparison of Case A (27.35
MWD/MTU).
Isotope MCNPX Experiment % Error* Range of Values from other Codes
Value Value
(mg/g U02) (mg/g UO2)
234U 1.422E-01 1.600E-1 -11.13 1.330E-1 to 1.750E-1
235U 8.339 8.470 -1.54 7.445 to 8.661
236U 3.173 3.140 1.06 3.128 to 3.540
238U 8.375E+02 8.425E+2 -0.59 8.637E+2 to 8.415E+2
237Np 2.794E-01 2.680E-1 4.25 2.527E-1 to 3.396E-1
238Pu 9.445E-02 1.012E-1 -6.67 5.721E-2 to 1.083E-1
239Pu 3.891 4.264 -8.74 3.660 to 4.690
240Pu 1.640 1.719 -4.60 1.573 to 8.60
241Pu 6.615E-01 6.812E-1 -2.90 5.310E-1 to 7.335E-1
242Pu 3.069E-01 2.886E-1 6.36 2.000E-1 to 3.192E-1
241Am 2.320E-01 N/A N/A 2.269E-1 to 2.598E-1
243Am 4.886E-02 N/A N/A 3.480E-2 to 4.672E-1
95Mo 5.661E-01 N/A N/A 5.590E-1 to 5.795E-1
99Tc 5.661E-01 N/A N/A 5.648E-1 to 6.904E-1
133Cs 8.259E-01 8.500E-1 -2.83 6.820E-1 to 8.640E-1
135Cs 3.620E-01 3.600E-1 0.54 3.728E-1 to 3.959E-1
143Nd 5.840E-01 6.130E-1 -4.73 6.040E-1 to 6.792E-1
145Nd 4.894E-01 5.100E-1 -4.04 4.984E-1 to 5.151E-1
147Sm 1.518E-01 N/A N/A 1.564E-1 to 1.932E-1
149Sm 1.679E-03 2.900E-3 -42.10 1.626E-3 to 2.900E-3
150Sm 1.854E-01 2.070E-1 -10.44 1.713E-1 to 2.146E-1
15Sm 7.401E-03 N/A N/A 6.376E-3 tol.413E-2
152Sm 8.124E-02 8.700E-2 -6.62 7.947E-2 to 1.073E-1
153Eu 7.257E-02 7.900E-2 -8.13 6.730E-2 to 8.921E-2
55Gd 2.724E-03 N/A N/A 1.507E-3 to 5.762E-3
*(Calculated/Measured-l)* 100

A further investigation was completed to set the limiting checking value for density

contribution of a specific nuclide to 1E-10 from the default 1E-4 value in order to determine if

this change actually affected the ultimate depletion outcome. Table 5-11, Table 5-12 and Table

5-13 display the results augmenting this limiting checking value (MCNPXc). All of the same










conditions applied in the initial calculations were also applied for the MCNPXc cases. The

results from the MCNPXc cases show significant improvement in actinide and fission product

prediction.

Table 5-12. MCNPXc Phase IB Benchmark results and comparison of Case B (37.12
MWD/MTU).


Isotope


234U
235U
236U
238U
237Np
Np
238Pu
239Pu
240Pu
241Pu
242Pu
241Am
243Am
95Mo
99Tc
133Cs
135Cs
143Nd
145Nd
147Sm
149Sm
150Sm
151Sm
152Sm
153Eu
155Gd


MCNPXc
Value
(mg/g UO2)
1.164E-01
4.822
3.619
8.303E+02
4.032E-01
1.750E-01
3.987
2.120
8.699E-01
5.936E-01
2.976E-01
1.186E-01
7.293E-01
7.277E-01
1.060
3.930E-01
6.750E-01
6.172E-01
1.632E-01
1.711E-03
2.521E-01
7.924E-03
9.892E-02
1.035E-01
4.251E-03


*(Calculated/Measured-l)* 100


% Error* Rang of Values from other Codes


Experiment
Value
(mg/g UO2)
1.400E-01
5.170
3.530
8.327E+02
3.560E-01
1.893E-01
4.357
2.239
9.028E-01
5.761E-01
N/A
N/A
N/A
N/A
1.090
4.000E-01
7.160E-01
6.530E-01
N/A
3.000E-03
2.710E-01
N/A
1.040E-01
1.090E-01
N/A


For the most part, the results calculated by MCNPXc fell within the range of values

calculated by the other depletion codes. These calculated values, however, did not exactly match

the measured experiment values. There are many possible explanations for the discrepancies in

calculated results as compared to the experimental data. The reasons include: (1) nuclear data;


-16.88
-6.73
2.51
-0.28
13.26
-7.55
-8.50
-5.31
-3.65
3.04
N/A
N/A
N/A
N/A
-6.24
-5.17
-8.90
-6.41
N/A
-44.91
-14.72
N/A
-8.45
-8.81
N/A


1.080E-1 to 1.570E-01
4.022 to 5.510
3.526 to 3.930
8.292E+02 to 8.360E+02
3.619E-01 to 4.919E-01
1.144E-01 to 2.069E-01
3.710 to 4.877
1.996 to 2.347
7.510E-01 to 9.846E-01
4.200E-01 to 6.347E-01
2.880E-01 to 3.418E-01
9.637E-02 to 1.391E-01
7.214E-01 to 7.545E-01
7.327E-01 to 8.372E-01
8.784E-01 to 1.117
3.967E-01 to 4.317E-01
7.013E-01 to 8.254E-01
6.326E-01 to 6.600E-01
1.659E-01 to 2.201E-01
1.736E-03 to 3.092E-03
2.297E-01 to 3.152E-01
7.990E-03 to 1.571E-02
9.761E-02 to 1.416E-01
9.960E-02 to 9.960E-02
2.538E-03 to 1.028E-02










(2) treatment of normalization parameters; (3) inadequacy of time steps; (4) statistical variance

(5) differences in neutron spectra; (6) spatial reactivity effects of the modeled geometry.

Table 5-13. MCNPXc Phase IB Benchmark results and comparison of Case C (44.34
MWD/MTU).


Isotope


234U
235U
236U
238U
237Np
238Pu
239Pu
240Pu
241Pu
242Pu
241Am
243Am
95Mo
99Tc
133Cs
135Cs
143Nd
145Nd
147Sm
149Sm
150Sm
Sm
151Sm
152Sm
153Eu
155Gd


MCNPXc
Value
(mg/g UO2)
1.006E-01
3.131
3.758
8.247E+02
4.854E-01
2.434E-01
3.970
2.362
9.633E-01
8.463E-01
3.222E-01
1.919E-01
8.359E-01
8.303E-01
1.206
4.104E-01
7.038E-01
6.930E-01
1.635E-01
1.751E-03
2.975E-01
8.239E-03
1.084E-01
1.228E-01
5.289E-03


% Error* Rang of Values from other Codes


Experiment
Value
(mg/g UO2)
1.200E-01
3.540
3.690
8.249E+02
4.680E-01
2.688E-01
4.357
2.543
1.020
8.401E-01
N/A
N/A
N/A
N/A
1.240
4.300E-01
7.630E-01
7.440E-01
N/A
4.700E-03
3.610E-01
N/A
1.210E-01
1.480E-01
N/A


*(Calculated/Measured-)* 100

Nuclear Data

The MCNPX benchmark calculations used ENDF/B VII.0 temperature dependent cross

section data for steady-state particle transport. The range of values calculated from other

depletion codes used either ENDF/B IV.0 or V.0 which was the data available at the time of the

benchmark. Possible discrepancies in the implemented transport data may result in calculation


-16.14
-11.54
1.84
-0.02
3.72
-9.47
-8.89
-7.13
-5.56
0.74
N/A
N/A
N/A
N/A
-2.71
-4.56
-7.76
-6.85
N/A
-62.75
-17.59
N/A
-10.39
-17.00
N/A


9.030E-02 to 1.440E-01
2.389 to 3.716
3.641 to 4.030
8.234E+02 to 8.316E+02
4.327E-01 to 5.934E-01
1.656E-01 to 2.810E-01
3.659 to 4.902
2.180 to 2.661
8.560E-01 to 1.111
5.960E-01 to 9.103E-01
3.102E-01 to 3.785E-01
1.634E-01 to 2.316E-01
8.092E-01 to 8.742E-01
8.449E-01 to 9.861E-01
9.723E-01 to 1.286
3.977E-01 to 4.605E-01
7.397E-01 to 8.839E-01
7.170E-01 to 7.560E-01
1.655E-01 to 2.302E-01
1.842E-03 to 3.286E-03
2.725E-01 to 3.980E-01
8.102E-03 to 1.682E-02
1.077E-01 to 1.587E-01
1.210E-01 to 1.596E-01
3.379E-03 to 1.318E-02









error. CINDER90 uses ENDF/B VI.0 fission yield data. This data may differ from the fission

yield available in ENDF/B V.0 or IV.0 resulting in further error in calculation. CINDER90 also

contains transmutation cross sections for 3400 nuclides. Over 3000 of those nuclides do not

contain available ENDF data. Therefore cross section determination codes were used to

calculate the data implemented in CINDER90. The CINDER90 data may also differ from what

is used by other codes leading to further error in calculation when comparing to other tools.

Treatment of Normalization Parameters

MCNPX currently does not account for the true recoverable energy per fission. An

approximation is made in order to try and determine the delayed energy contribution. The

delayed energy contribution is unfortunately system dependent. For this modeled system,

however, the geometry modeled was not the actual experimental geometry and therefore

accounting for the true recoverable energy per fission in the adjusted geometry would not be

possible. Therefore the assumptions made about delayed energy contribution were deemed

acceptable for lack of a better approximation but known to affect the final solution outcome.

Inadequate Time Steps

The same time steps were used for each examined case. Therefore a larger amount of

burnup was witnessed per time step for the higher burnup cases. It is possible that the time steps

used for the benchmark were not short enough in order to account for significant change in

neutron spectrum. Adequate time step selection is usually a trial and error process of accounting

for significant changes in the neutron spectrum as a result of the asymptotic buildup of highly

absorbing isotopes. Therefore choosing time-steps that are too large will not properly account

for significant changes in neutron spectrum ultimately resulting in calculation error.









Statistical Variance

By using the Monte Carlo method and simulating the actual physical process, the

calculated reaction rates, used as coefficients for the depletion solution, possessed an associated

statistical error. This error propagates through the linked solution process. Therefore the

coefficients generated for each time step possessed a propagated statistical error. For rarely

sampled interactions, this error can be very large ultimately affecting the transmutation vector for

nuclides resulting from those interactions. Furthermore, the cross sections themselves also

posses a measured error associated with each value. The combination of the measured error

incurred from determining the cross section values and the stochastic error from the modeling

process ultimately influences the accuracy of the calculated results.

Differences in Neutron Spectra

The Combustion Engineering 14 X 14 fuel assembly that was used for the experimental

benchmark was burned with a critical spectrum. The modeled geometry, however, possessed a

spectrum that was either supercritical, for the beginning time steps, or subcritical, for the latter

time steps. Unfortunately, true system operating parameters were not given. For example, to

maintain criticality in a PWR, the boron concentration in the water coolant is constantly diluted

in order to subsidize the negative reactivity deficit incurred from burning fuel. For the

benchmark cases, boron concentration was assumed constant over the entire cycle. Therefore the

neutron spectrum of the modeled geometry could not possibly be made critical.

Simply adjusting the boron concentration to maintain criticality in a pin cell model would

also result in further error. Criticality is maintained in a system by balancing the leakage and

absorption. For example, the critical spectrum in a modeled geometry is preserved by

accounting for the leakage of neighboring geometries while operating the system with the true

operating conditions. Unfortunately, when modeling a subsection of a larger geometry, the









leakage into that subsection may not be known and must be approximated. Approximating a pin

cell calculation with reflective boundary conditions is a false representation of the boundary

conditions witnessed in a typical reactor. The only way to truly burn the fuel bundle at critical is

to model the entire system. This benchmark did not give specification for modeling the entire

system and therefore errors should be expected in comparing results with experiment.

Spatial Reactivity Effects of the Modeled Geometry

The modeled geometry was an infinitely reflected pin cell with adjusted fuel-to-moderator

ratio in order to try and mimic the same nuclide buildup as a 14 X14 Combustion Engineering

fuel assembly. An infinitely reflected boundary condition is only appropriate for scenarios in

which the boundary posses a true zero net leakage. For a certain section of a larger geometry,

approximating a zero net leakage boundary condition is acceptable if the section of the geometry

resides in a region possessing a flat flux or power distribution. For example, a fuel assembly that

sits in a region of the reactor where the fuel assembly is surrounded by similar fuel assemblies

exhibiting a similar power distribution. Unfortunately, this condition is never truly realized on a

pin cell level. A radial power distribution exists across the fuel assembly and therefore each pin

within the bundle bums at a different rate. By only modeling a single fuel pin to represent the

entire fuel assembly, the spatial reactivity effects are ignored that would otherwise be witnessed

in the full fuel assembly. This disregard of spatial reactivity effects ultimately affects the

accuracy of the solution outcome. Because only a pin cell is modeled, comparisons between the

calculated results and radiochemical analysis data taken from the full assembly analysis are

expected to be error.

Further MONTEBURNS Comparisons

MONTEBURNS calculations for the OECD/NEA Bumup Credit Calculational Criticality

Benchmark Phase IB had been completed as part of a prior thesis. Due to computational









limitations at the time of those calculations, a coarser representation of the benchmark was

modeled. The model implemented for that work used a single 360 cm tall fuel zone, ENDF V

cross sections, and single time steps for operating durations (the exact model description

implemented is explained in the reference). The model also lacked many of the fission products

captured by the MCNPX depletion methodology (setting the MONTEBURNS transmutation

importance fraction to 0.001). In the thesis, values were only calculated for cases A and B, and

these values were within the range of values demonstrated by other codes and MCNPX.

In order to better compare the MONTEBURNS methodology and MCNPX, a series of

MONTEBURNS 2.0 calculations were rerun using the same fidelity model as implemented for

the MCNPX depletion calculations. ENDF VII.0 temperature dependent cross sections were

implemented. The same kcode parameters, time steps, power levels, and manually set material

concentrations used in the MCNPX calculations were also used for the MONTEBURNS 2.0

calculations. For these MONTEBURNS calculations, the importance fraction was set to le-10 in

order to account for more of the fission and activation products. MONTEBURNS 2.0 was

incapable of burning the borated water region and resetting the boron concentrations at each step.

MONTEBURNS 2.0 was also incapable of modeling the same isotope at multiple temperatures;

therefore all fission products that aligned with cladding nuclides used the same temperature

dependent cross sections for the cladding temperature. These two issues are easily handled in

MCNPX, however, not treated in MONTEBURNS 2.0, and therefore differences were expected

as a result of these discrepancies. MONTEBURNS results are compared with MCNPXc results

and experimental values, for cases A and B, in Table 5-14 and Table 5-15.

The higher fidelity MONEBURNS calculations for this benchmark are in larger error than

the MCNPXc calculations. The reasons for the gross error in 238Pu is mostly due to the 242mAm









and 242Am cross section mixup. The large over prediction of nuclides is not well determined.

Increasing the amount of fission products tracked within MONTEBURNS 2.0 causes a

competition for capture and therefore reduces actinide burning and transmutation. However, the

prior thesis results show a large under prediction of nuclide concentration while the new results

show a large over prediction. These discrepancies may be related to the treatment of incomplete

cross sections, or limitations of the MONTEBURNS 2.0 tally routines, or limitations of the

ORIGEN2.2 algorithm.

Table 5-14. MONTEBURNS* Phase IB Benchmark results and comparison of Case A (27.35
MWD/MTU).
Isotope Experiment % Error % Error % Error
Value MONTEBURNS MONTEBURNS MCNPXc**
(mg/g U02) OLD** NEW**
234U 1.422E-01 -2.45 2.77 -11.13
235U 8.339 -4.32 9.06 -1.54
236U 3.173 2.09 3.07 1.06
238U 8.375E+02 -0.5 0.31 -0.59
237Np 2.794E-01 6.65 6.11 4.25
238Pu 9.445E-02 -6.12 25.03 -6.67
239Pu 3.891 -7.50 7.34 -8.74
240Pu 1.640 -2.00 10.29 -4.60
241Pu 6.615E-01 -2.72 9.77 -2.90
242Pu 3.069E-01 6.65 12.07 6.36
133Cs 8.259E-01 1.91 -6.92% -2.83
135Cs 3.620E-01 4.46 -1.40% 0.54
143Nd 5.840E-01 -0.36 -5.51% -4.73
15Sm 1.854E-01 -13.22 -14.51% -10.44
152Sm 8.124E-02 -1.35 -6.39% -6.62
153Eu 7.257E-02 5.11 -5.71% -8.13
*MONTEBURNS OLD = MONTEBURNS from prior thesis;75
MONTEBURNS NEW =MOTNEBURNS 2.0 with higher fidelity model
***(Calculated/Measured-l)* 100

The results do show major cause for concern, because a higher fidelity model should show

better results. It seems that modeling more of the available fission products actually hurt the

calculation; however, further testing is required in order to determine the root cause of the

miscalculation. New MONTEBURNS versions have been developed since MONTEBURNS 2.0









and therefore further testing will be required in order to determine the origin of this cause and

whether or not the cause is problem dependent.

Table 5-15. MONTEBURNS* Phase IB Benchmark results and comparison of Case A (37.38
MWD/MTU).
Isotope Experiment % Error % Error % Error
Value MONTEBURNS MONTEBURNS MCNPXc**
(mg/g U02) OLD** NEW**
234U 1.164E-01 -5.05 -11 -16.88
235U 4.822 -9.66 -31.71 -6.73
236U 3.619 2.68 -11.43 2.51
238U 8.303E+02 -0.28 1.25 -0.28
237Np 4.032E-01 14.38 -36.40 13.26
238Pu 1.750E-01 -3.84 -44.37 -7.55
239Pu 3.987 -7.46 3.95 -8.50
24Pu 2.120 -2.47 -18.18 -5.31
241Pu 8.699E-01 -4.05 -21.07 -3.65
242Pu 5.936E-01 7.41 -77.15 3.04
133CS 1.060 2.55 20.93 -6.24
135Cs 3.930E-01 4.79 7.42 -5.17
143Nd 6.750E-01 -0.76 11.81 -8.90
150Sm 2.521E-01 -8.96 -30 -14.72
152Sm 9.892E-02 -0.20 27 -8.45
153Eu 1.035E-01 13.17 -30 -8.81
*MONTEBURNS OLD = MONTEBURNS from prior thesis;75
MONTEBURNS NEW =MOTNEBURNS 2.0 with higher fidelity model
***(Calculated/Measured-l)* 100

H. B. Robinson Benchmark Specification

The Bumup Credit Criticality Safety expert group offers the Spent Fuel Isotopic

Composition database containing measured nuclide composition data for 14 LWRs in order to

help validate bum-up codes.97 The H. B. Robinson benchmark is one of the 14 available LWR

benchmarks. Post irradiation examination measurements exist for select fuel pins within one 15

X 15 Westinghouse fuel assembly from the H. B. Robinson Unit 2 reactor.53 For a specific fuel

pin within the assembly, radiochemical analysis was completed at several axial levels

corresponding to several different burnups. The benchmark was chosen for this study because

full assembly geometry specification was available, and the full modeled assembly was








considered to be in a core location assumed to possess a reasonably "flat enough" power

distribution to assume zero net leakage at the physical boundary of the assembly. Because zero

net leakage was assumed at the assembly boundary, it was also further assumed that the

assembly could be modeled as infinitely reflected at the geometry boundary. Because leakage

from assembly to assembly is affected by the ultimate core boundaries, power shaping and

poison positioning, this assumption is not exactly true. However, the assumption was deemed

acceptable enough to model the assembly because the power distribution across the assembly

region was assumed "flat enough".


Analyzed Fuel



Burnable Poison


Instrument Tube



Guide Tube


*O O0@00 '




Om10 ..*.. OseAO
I*E 1S^^o'j :y+~IFE~ _I
^C~lr &S$MPHJ3 j


Figure 5-2. Infinitely reflected pin cell geometry.

Oak Ridge National Laboratory had also completed a verification study of the SCALE

package using this benchmark. This verification study offered knowledge of how to setup the

benchmark for direct comparison with the SCALE package. The Oak Ridge National Laboratory

report, ORNL/TM-12667.28, was therefore used to determine the benchmark setup.53 The H. B.

Robinson Fuel assembly was modeled using eighth bundle symmetry, and took advantage of

repeated structure modeling available since MCNPX 2.6.D. The exact modeled geometry is









displayed in Figure 5-2. Table 5-16 displays the modeled assembly design data for the

benchmark. The material compositions for the fuel, clad, guide tube, and burnable poison rod

materials are specified in the Oak Ridge National Laboratory report, ORNL/TM-12667.28.53


Table 5-16. Fuel assembly design data for H. B
Parameter
Assembly general data
Designer
Lattice
Number of fuel rods
Number of guide tubes
Number of burnable poison rods
Number of instrument tubes
Assembly pitch
Assembly fuel
Fuel rod data
Type of fuel pellet
Enrichment: wt % 235U
wt % 234U
wt % 236U
Pellet density
Rod Pitch
Rod OD
Rod ID
Pellet diameter
Active fuel length
Clad temperature
Clad material
Guide tube data
Inner radius
Outer radius
Material
Instrument tube data
Inner radius
Outer radius
Material
Burnable poison rod data
Air OD
SS304 OD
Air OD
Borosilicate glass OD
Air OD
SS304 OD
*Data taken from reference53


Robinson Benchmark*.
Data

Westinghouse
15 x 15
204
8
12
1
21.50 cm
443.7 kg U

UO2
2.561
0.023
0.013
9.44 g/cm3
1.43 cm
1.0719 cm
0.9484 cm
0.9242 cm
365.76 cm
595 K
Zircaloy-4

0.6502 cm
0.6934 cm
Zircaloy-4

0.6502 cm
0.6934 cm
Zircaloy-4

0.5677 cm
0.6007 cm
0.6172 cm
1.0058 cm
1.0173 cm
1.1151 cm









Each steady-state Monte Carlo calculation was completed running 4000 histories per cycle

for 155 cycles skipping the first 25 cycles. Initial source particles were started in each of the

fissile fuel locations. Though the geometry was modeled in 3-dimensions, only 1 axial region

was utilized; therefore the model does not account for axial variation of the power distribution as

a function ofburnup. However, each fuel pin, within an eighth assembly symmetric section,

used a separate burn material in order to account for radial variation of the interaction rates.

Four different cases, corresponding to four axial locations, experiencing four separate bumups,

were examined as part of the benchmark. Cases A-D represents the different burnup cases from

the benchmark: (1) Case A = 16.02 GWD/MTU; (2) Case B = 23.8 GWD/MTU; (3) Case C =

28.47 GWD/MTU; (4) Case D = 31.66 GWD/MTU. The measured bumup given for case A did

not correspond to the final time interval specified in the Oak Ridge National Laboratory Report.

For the Case A calculation, the final operating interval was changed from 156 days to 165 days

in order to correspond to the expected bumup exhibited in the assembly.54 Table 5-17 and Table

5-18 display the operating intervals, time steps used for each interval, corresponding down times,

and cycle dependent soluble boron concentrations for Case A-D.

Table 5-17. Operating data for H. B. Robinson fuel assembly.
Cycle 1 2 3 4
Operating Interval 243.5 243.5 156 156
(days)
Interval Time Step Durations 1, 15, 20, 30, 1, 15, 20, 30, 1, 15, 20, 1, 15, 20,
(days) 37, 44, 46 37, 44, 46 30, 40, 50 30, 40, 50*
Downtime 40 64 39 -**
(days)
Average Soluble Boron 625.5 247.5 652.5 247.5
Concentration
(ppm)
* This value was set to 59 for Case A.
** This value was set either to 3936 for Cases A-B or 3637 for Cases C-D.









Table 5-18. Operating data for H. B. Robinson fuel assembly
Case Total Burnup Moderator Temperature Moderator Density Fuel Temperature
(GWD/MTU) (K) (g/cm3) (K)
A 16.02 559 0.7544 743
B 23.81 559 0.7538 830
C 28.47 576 0.7208 883
D 31.66 576 0.7135 923

H. B. Robinson Benchmark Results

The results of the benchmark were compared to the results from SCALE as specified in

ORNL/TM-12667.53 The SCALE sequence used in the calculations was SAS2H/ORIGEN-S

analysis sequence and 27-group cross-section library (27BURNUPLIB) of SCALE-4.53 Two

sets of analysis were completed. The first set used the original MCNPX, and the next set used

the modified MCNPXc. The results of the benchmark using MCNPX are displayed in Table 5-

19, and the results of the benchmark using MCNPXc are displayed in Table 5-20.



Table 5-19. The H. B. Robinson Benchmark results* using MCNPX compared with SCALE.
Case A Case B Case C Case D
16.02 GWD/MTU 23.8 GWD/MTU 28.47 GWD/MTU 31.66 GWD/MTU
Isotope MCNPX SCALE MCNPX SCALE MCNPX SCALE MCNPX SCALE
235U 0.18 0.60 -3.11 1.40 -11.8 -4.90 -9.66 0.10
236U -1.62 -1.5 -1.05 -2.20 3.72 2.20 1.18 -0.50
238U -0.12 0.10 -0.60 -0.60 0.47 0.50 -0.73 -0.20
237Np -10.57 1.50 -8.09 0.90 -14.72 -6.50 -10.69 -0.40
238Pu -6.35 7.00 -6.41 7.70 -9.22 5.30 -8.66 8.20
239Pu -3.19 -1.5 -4.75 -4.20 -5.42 -4.90 -6.52 -3.70
24Pu -1.23 5.90 -1.45 6.00 -11.03 0.50 -8.79 5.4
241Pu -8.23 6.00 -6.30 5.50 2.43 14.30 3.08 11.10
99Tc 7.74 12.40 4.20 8.60 9.58 14.60 5.53 11.70
137Cs -3.10 0.20 -3.10 -0.80 -0.38 3.90 -3.09 1.2
*Percent difference from measured experimental results [(Calculated/Measured -1)*100].

The results displayed in 5-18 exhibits the same trend in burnup and predictive capability as

the MCNPX results for the OECD/NEA Burnup Credit Phase IB. As burnup increases, nuclide

predictive capability gets worse and the calculated values seem not as close to the actual









measured calculation. For example, the calculation percent difference from experiment for most

actinides for Cases A-B was <8%; however, at higher burnups, Cases C-D, the calculation

percent difference from experiment grew to 14%. Though the SCALE values and the MCNPX

values seemed relatively close in magnitude, further calculations were also completed using

MCNPXc to see the magnitude of improvement. In Table 5-19, the results show significant

improvement in actinide prediction for uranium and plutonium isotopes when utilizing

MCNPXc.

Table 5-20. The H. B. Robinson Benchmark results* using MCNPXc compared with SCALE.
Case A Case B Case C Case D
16.02 GWD/MTU 23.8 GWD/MTU 28.47 GWD/MTU 31.66 GWD/MTU
Isotope MCNPXc SCALE MCNPXc SCALE MCNPXc SCALE MCNPXc SCALE
235U 0.47 0.60 -0.58 1.40 -8.19 -4.90 -5.35 0.10
236U -1.81 -1.5 -1.90 -2.20 2.77 2.20 0.57 -0.50
238U 0.12 0.10 -0.54 -0.60 0.53 0.50 -0.73 -0.20
237Np -6.63 1.50 -7.31 0.90 -11.21 -6.50 -7.41 -0.40
238Pu -4.125 7.00 -3.86 7.70 -3.29 5.30 -1.89 8.20
239Pu -0.005 -1.5 -0.37 -4.20 -0.38 -4.90 -0.52 -3.70
24Pu 4.013 5.90 0.59 6.00 -6.94 0.50 -3.66 5.4
241Pu 2.4 6.00 2.82 5.50 1.88 14.30 1.36 11.10
99Tc 10.815 12.40 6.76 8.60 12.13 14.60 8.49 11.70
137Cs -1.152 0.20 -1.88 -0.80 0.67 3.90 -1.81 1.2
*Percent difference from measured experimental results [(Calculated/Measured -1)*100].

The possible reasons for error in the calculation are similar to those mentioned in the

OECD/NEA Burnup Credit Phase IB benchmark. Errors in nuclear data, treatment of

normalization parameters, statistical variance in reaction rates, and not modeling a critical

spectrum all contribute to errors in the calculation. It is important to note that the same time step

durations were also used for each of the cases. Therefore each case was modeled using different

size burnup steps. The coefficients for the depletion calculation were therefore updated over

longer burnup steps for the higher burnup cases. Therefore these results also suggest that the

longer burnup step durations used for the higher burnup cases may have been too large to assume









that significant change in the flux had not occurred over the time step. As mentioned previously,

selecting adequate time steps depends upon the buildup and depletion of highly absorbing

isotopes that affect the neutron spectrum in the material. Therefore, as is displayed in the results,

the prior statement may be further translated to mean that adequate time selection is not just

related to resident time but also related to the combination of resident time and operating power

over that time, or burnup. However, for Cases A-B, the percent difference in actinide prediction

from measured experiment is <4% except for Np-237 which is superior to the SCALE

calculation. These results therefore suggest that MCNPX is very capable of modeling full

assembly behavior and accurately predicting nuclide buildup as long as adequate burnup step

durations are maintained.

Another important source of error to consider is that the measured data also possess error

as a function of the precision and accuracy of the measurement process. Unfortunately, the

measurement error for this benchmark does not exist. The Spent Fuel Isotopic Composition

database posted on the NEA website also does not contain measurement error for any of the 14

available experimental benchmarks.97 Most of the reports used to generate the database do

contain the methods of radiochemical analysis applied to extract the experimental data; however,

these reports do not contain a quantification of the error associated in obtaining the measured

data. This problem of not listing experimental error therefore must be addressed in future

benchmarks in order to understand the validity of calculated results when compared to the

measured data.

OECD/NEA Burnup Credit Phase IVB Benchmark Specification

Both the OECD/NEA Burnup Credit Phase IB benchmark and H. B. Robinson benchmark

modeled typical PWR uranium dioxide fuel operating conditions. In order to prove the

versatility of the of the MCNPX depletion methodology for different types of reactors, another









operating scenario, besides a typical PWR uranium dioxide fuel operating strategy, was

investigated. The OECD/NEA Burnup Credit Phase IVB benchmark offers a suite of benchmark

calculations for mixed oxide fuel (MOX).54 MOX assembly modeling is important to a closed

reactor fuel cycle because the recycled assembly will contain a combination of uranium and

plutonium (and possibly other actinides) at BOL for the recycled assembly.9 Calculations of the

OECD/NEA Burnup Credit Phase IVB benchmark have been completed with many different

depletion codes, and therefore an adequate comparison database is available for the benchmark.54

Table 5-21. Fuel assembly design data for OECD/NEA Burnup Credit Phase IVB Benchmark.
Parameter Data
Assembly general data
Lattice 17x 17
Number of fuel rods 264
Number of guide tubes 24
Number of instrument tubes 1
Fuel rod data
Type of fuel pellet MOX
Rod pitch 1.26 cm
Clad thickness 0.065 cm (no gap between fuel and clad)
Pellet diameter 0.410 cm
Active fuel length 365.76 cm
Fuel temperature 900 K
Clad temperature 620 K
Clad material Zircaloy-2
Guide and Instrument tube data
Inner radius 0.571 cm
Outer radius 0.613 cm
Material Zircaloy-2

Modeling a single infinitely reflected MOX fuel assembly does not adequately represent

the true reactor operating conditions of the assembly. Many neutronic characteristics of the

MOX assembly limit the assembly's placement to certain patterns within certain sections of the

reactor core. The higher absorption cross section of the plutonium isotopes, limits the negative

reactivity worth of control material, such as dissolved boron or control blades, in the presence of

the MOX assembly. The high fission cross section of Pu-239 and large initial loading of Pu-239









produces large fission peaks; therefore to accommodate this affect and flatten the power peak,

MOX assemblies are placed away from water gap regions.9 Finally, the fact that Pu-239

possesses a smaller delayed neutron fraction (P = 0.0021) and shorter neutron lifetime (1 = 27as),

than U-235 (P=0.0065, 1 = 47[ts), thus limiting controllability of kinetic response, limits the

amount of MOX assemblies placed in the reactor core.9 Countries already implementing the

MOX fuel cycle, such as France, limit the amount of MOX loading to no more than 30%, and

never load MOX assemblies surrounded by other MOX assemblies.9 However, since the H. B.

Robinson benchmark only offered code-to-code comparison with SCALE, the author decided to

use the OECD/NEA Burnup Credit Phase IVB benchmark as a comparison tool, knowing that

modeling a infinitely reflected MOX assembly would not model true reactor behavior, because

the benchmark offered data from a variety of codes for adequate code-to-code comparison for

MOX assembly modeling strategy.

Table 5-22. Initial MOX fuel composition for Case A for Phase IVB Benchmark.
Values Expressed in (atoms/b-cm)
Nuclide High Enrichment Medium Enrichment Low Enrichment
234U 2.57E-07 2.64E-07 2.68E-07
235U 5.38E-05 5.53E-05 5.60E-05
238U 2.12E-02 2.18E-02 2.21E-02
238Pu 5.17E-05 3.61E-05 2.85E-05
239Pu 1.13E-03 7.87E-04 6.20E-04
24Pu 5.35E-04 3.74E-04 2.95E-04
241Pu 1.94E-04 1.36E-04 1.07E-04
242Pu 1.46E-04 1.02E-04 8.06E-05
160 4.66E-02 4.66E-02 4.65E-02

Two types of infinitely reflected 17 X 17 MOX fuel assemblies were modeled. Each

modeled case assembly was composed of MOX fuel pellets encased in Zircaloy-2 clad fuel pins

along with Zircaloy-2 guide tubes. The geometry specifications for both cases are given in Table

5-21. The Case A model included a MOX assembly possessing plutonium nuclide

concentrations consistent with typical concentrations expected from a first recycle PWR









assembly. The Case B model included a MOX assembly possessing plutonium nuclide

concentrations consistent with typical concentrations expected from weapons disposition

material.

Table 5-23. Initial MOX fuel composition for Case B for Phase IVB Benchmark.
Values Expressed in (atoms/b-cm)
Nuclide High Enrichment Medium Enrichment Low Enrichment
234U 2.69E-07 2.73E-07 2.75E-07
235U 5.63E-05 5.71E-05 5.74E-05
238U 2.22E-02 2.25E-02 2.26E-02
238Pu 5.09E-07 3.56E-07 2.81E-07
239Pu 9.49E-04 6.64E-04 5.23E-04
240Pu 6.06E-05 4.24E-05 3.34E-05
241Pu 3.02E-06 2.11E-06 1.66E-06
242Pu 5.01E-07 3.50E-07 2.76E-07
160 4.65E-02 4.65E-02 4.65E-02

For each case, three types of plutonium concentration loadings were used for the fuel pins

within each case. Table 5-22 lists the fuel compositions for Case A, and Table 5-23 lists the fuel

composition for Case B. Table 5-24 lists the borated water composition and Table 5-25 lists the

Zircaloy-2 composition used for the fuel clad and guide tubes. Figure 5-3 displays the

distribution of the different fuel pins throughout the fuel assembly for both cases. Each steady-

state reaction rate case was completed using 4000 particles per cycle skipping the first 25 cycles

for 155 cycles. Initial source locations were placed within each fissile cell location. Case A was

burned at 17.896 MW, and Case B was burned at 17.862 MW. The operating data for the

assembly is listed in Table 5-26.

Table 5-24. Borated water composition for Phase IVB Benchmark (660ppm boron, density
0.7245 g/cm3).
Nuclide Number Density
(atoms/b-cm)
H 4.8414E-02
160 2.4213E-02
10B 4.7896E-06
11B 1.9424E-05










17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1

















Figure 5-3.


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Keys

| Low enriched MOX fuel pin

SMedium enriched MOX fuel pin

SHigh enriched MOX fuel pin

D Guide tube

S Instrument tube

MOX fuel enrichment map for Phase IVB Benchmark.












Table 5-25. Zicaloy-2 composition used for clad and guide tube material for Phase IVB
Benchmark (density 5.8736 g/cm3).
Nuclide (Atoms/b-cm)
50Sn 2.9656E-06
52Sn 5.7189E-05
53Sn 6.4848E-06
54Sn 1.6142E-06
54Fe 7.8122E-06
56Fe 1.2245E-04
57Fe 2.8278E-06
58Fe 3.7633E-07
90Zr 1.9889E-02
91Zr 4.3373E-03
92Zr 6.6297E-03
94Zr 6.7186E-03
96Zr 1.0824E-03

Table 5-26. Operating data for Phase IVB Benchmark.
Cycle 1 2 3
Operating Interval 420 420 420
(days)
Interval Time Step Durations 1, 15, 20, 30, 50, 1, 15, 20, 30, 50, 1, 15, 20, 30, 50,
(days) 60, 60, 60, 60, 64 60, 60, 60, 60, 64 60, 60, 60, 60, 64
Downtime 30 30 0, 1826.25
(days)

OECD/NEA Burnup Credit Phase IVB Benchmark Results

The OECD/NEA Burnup Credit Phase IVB benchmark offers cycle by cycle nuclide

buildup and depletion information for a plethora of actinides and fission products. Table 5-27,

Table 5-28, Table 5-29 and Table 5-30 present the MCNPX EOC 1, 2, 3 and after 5 years of

cooling results for Case A as compared to the range of values achieved from other depletion

codes in the benchmark. Table 5-31, Table 5-32, Table 5-33 and Table 5-34 present the

MCNPX EOC 1, 2, 3 and after 5 years of cooling results for Case B as compared to the range of

values achieved from other depletion codes in the benchmark. Due to the success in nuclide

prediction of MCNPXc, each of these cases were run using MCNPXc.












Table 5-27. EOC 1 results for Case A of the Phase IVB Benchmark.
Nuclide MCNPX Range of Values from
(atoms/b-cm) Other Depletion Codes
(atoms/b-cm)
234U 5.8483E-07 2.1279E-07 to 5.8216E-07
235U 4.3753E-05 4.3720E-05 to 4.3925E-05
6U 2.6406E-06 2.4858E-06 to 2.6190E-06
238U 2.1134E-02 2.1155E-02to 2.1163E-02
238Pu 4.1533E-05 4.0713E-05 to 4.1630E-05
9Pu 8.0377E-04 7.9700E-04 to 8.1275E-04
24Pu 4.7370E-04 4.7109E-04 to 4.8300E-04
241Pu 2.1855E-04 2.1520E-04 to 2.2011E-04
242Pu 1.6393E-04 1.3389E-04 to 1.3785E-04
237Np 1.6176E-06 1.0400E-07 to 1.6560E-06
241Am 9.4895E-06 8.4140E-06 to 8.8762E-06
243Am 1.8069E-05 1.5613E-05 to 1.8260E-05
2Cm 9.5381E-07 8.7154E-07 to 9.8523E-07
243Cm 1.3915E-08 1.1641E-08 to 1.4560E-08
244Cm 5.3807E-06 2.6688E-06 to 3.9610E-06
2Cm 2.8581E-07 1.2915E-07 to 1.9877E-07
95Mo 1.2829E-05 1.1690E-05 to 1.7651E-05
99Tc 2.1958E-05 2.1756E-05 to 2.3060E-05
101Ru 2.2331E-05 2.1640E-05 to 2.3054E-05
103Rh 2.0106E-06 1.9630E-05 to 2.2642E-05
109Ag 5.0294E-06 4.6603E-06 to 5.9460E-06
133CS 2.4605E-05 2.3463E-05 to 2.4597E-05
143Nd 1.5949E-05 1.5350E-05 to 1.6594E-05
145Nd 1.1238E-05 1.1183E-05 to 1.1518E-05
147Sm 8.0947E-07 7.8960E-07 to 8.2700E-07
149Sm 4.2895E-07 3.7767E-07 to 4.3945E-07
15Sm 4.7391E-06 4.7807E-06 to 5.1050E-06
15Sm 1.3397E-06 1.3420E-06 to 1.5281E-06
152Sm 2.7837E-06 2.7790E-06 to 3.1530E-06
153Eu 2.0936E-06 2.0380E-06 to 2.2411E-06
55Gd 8.6977E-09 4.3940E-09 to 1.6769E-08











Table 5-28. EOC 2 results for Case A of the Phase IVB Benchmark.
Nuclide MCNPX Range of Values from
(atoms/b-cm) Other Depletion Codes
(atoms/b-cm)
234U 8.0735E-07 1.7413E-07 to 8.4354E-07
235U 3.4593E-05 3.4500E-05 to 3.4893E-05
236U 4.6979E-06 4.4215E-06 to 4.6370E-06
238U 2.0896E-02 2.0910E-02 to 2.0927E-02
238Pu 3.9782E-05 3.5013E-05 to 3.9340E-05
239Pu 6.3728E-04 6.2980E-04 to 6.5524E-04
24Pu 4.4837E-04 4.4499E-04 to 4.6350E-04
241Pu 2.3731E-04 2.3210E-04 to 2.3979E-04
242Pu 1.7027E-04 1.4147E-04 to 1.4982E-04
237Np 3.1341E-06 3.6114E-07to3.2139E-06
241Am 1.5907E-05 1.4680E-05 to 1.5982E-05
243Am 2.9191E-05 2.7797E-05 to 3.1620E-05
2Cm 2.5536E-06 2.3253E-06 to 2.6204E-06
2Cm 7.2443E-08 5.8709E-08 to 7.0660E-08
244Cm 1.7168E-05 9.2778E-06 to 1.3050E-05
24Cm 1.7035E-06 8.1985E-07 to 1.2348E-06
95Mo 2.9454E-05 2.8330E-05 to 3.4297E-05
99Tc 4.1886E-05 4.1811E-05 to 4.4400E-05
101Ru 4.3865E-05 4.2320E-05 to 4.5186E-05
103Rh 4.0075E-05 3.7469E-05 to 4.0914E-05
109Ag 9.0959E-06 8.0989E-06 to 1.0640E-05
133CS 4.6303E-05 4.4403E-05 to 4.6602E-05
143Nd 3.0328E-05 2.9970E-05 to 3.1809E-05
145Nd 2.1678E-05 2.1322E-05 to 2.2191E-05
147Sm 2.1430E-06 2.4130E-06 to 2.4742E-06
149Sm 4.0373E-07 3.5930E-07 to 4.1990E-07
15Sm 1.0296E-05 1.0197E-05 to 1.1350E-05
15Sm 1.5541E-06 1.5306E-06 to 1.8870E-06
152Sm 4.9712E-06 5.0468E-06 to 5.9850E-06
153Eu 4.9325E-06 5.0480E-06 to 5.4745E-06
55Gd 1.6016E-08 7.9700E-09 to 2.8935E-08











Table 5-29. EOC 3 results for Case A of the Phase IVB Benchmark.
Nuclide MCNPX Range of Values from
(atoms/b-cm) Other Depletion Codes
(atoms/b-cm)
234U 9.4561E-07 1.4169E-07 to 1.0509E-06
235U 2.6715E-05 2.6596E-05 to 2.7129E-05
236U 6.2239E-06 5.8461E-06 to 6.1140E-06
238U 2.0648E-02 2.0657E-02 to 2.0678E-02
238Pu 4.0397E-05 2.9577E-05 to 3.9688E-05
239Pu 5.0933E-04 5.0368E-04 to 5.3435E-04
24Pu 4.1054E-04 4.0836E-04 to 4.2870E-04
241Pu 2.3785E-04 2.3107E-04 to 2.4058E-04
242Pu 1.8086E-04 1.5231E-04 to 1.6558E-04
237Np 4.3966E-06 6.9348E-07 to 4.5652E-06
241Am 1.8127E-05 1.7960E-05 to 2.0066E-05
243Am 3.6401E-05 3.6793E-05 to 4.2000E-05
2Cm 4.3963E-06 3.5650E-06 to 4.0436E-06
3Cm 1.5834E-07 1.2551E-07 to 1.4820E-07
244Cm 3.1800E-05 1.8435E-05 to 2.4950E-05
24Cm 4.1506E-06 2.1928E-06 to 3.2335E-06
95Mo 4.3988E-05 4.3990E-05 to 4.9929E-05
99Tc 5.9641E-05 6.0068E-05 to 6.3850E-05
101Ru 6.4487E-05 6.1990E-05 to 6.6364E-05
103Rh 5.3043E-05 5.1041E-05 to 5.5357E-05
109Ag 1.2395E-05 1.0524E-05 to 1.4480E-05
133CS 6.4706E-05 6.2829E-05 to 6.6172E-05
143Nd 4.2327E-05 4.2830E-05 to 4.5385E-05
145Nd 3.1270E-05 3.0386E-05 to 3.2104E-05
147Sm 3.1679E-06 4.0280E-06 to 4.1579E-06
149Sm 3.2126E-07 3.2302E-07 to 3.8520E-07
15Sm 1.5873E-05 1.5410E-05 to 1.7660E-05
15Sm 1.5666E-06 1.5340E-06 to 2.0110E-06
152Sm 6.3075E-06 6.3335E-06 to 7.9580E-06
153Eu 7.5412E-06 8.1310E-06 to 8.6655E-06
55Gd 1.9499E-08 1.3920E-08 to 4.2299E-08











Table 5-30. Five year cooling time results for Case A of the Phase IVB Benchmark.
Nuclide MCNPX Range of Values from
(atoms/b-cm) Other Depletion Codes
(atoms/b-cm)
U 2.6607E-06 1.4169E-07 to 2.7220E-06
235U 2.6785E-05 2.6668E-05 to 2.7206E-05
236U 6.4390E-06 5.8461E-06 to 6.2030E-06
238U 2.0648E-02 2.0657E-02 to 2.0678E-02
238Pu 4.3111E-05 2.8432E-05 to 4.1989E-05
239Pu 5.1081E-04 5.0542E-04 to 5.3613E-04
24Pu 4.1585E-04 4.1217E-04 to 4.3260E-04
241Pu 1.8680E-04 1.8167E-04to 1.8891E-04
242Pu 1.8086E-04 1.5231E-04 to 1.6558E-04
237Np 4.8116E-06 7.0645E-07 to 4.9848E-06
241Am 6.8783E-05 6.6320E-05 to 7.0845E-05
243Am 3.6406E-05 3.6793E-05 to 3.9880E-05
2Cm 2.5594E-09 1.6887E-09 to 5.0942E-09
2Cm 1.4022E-07 1.1143E-07 to 1.2852E-07
244Cm 2.6279E-05 1.5224E-05 to 1.9671E-05
24Cm 4.1491E-06 2.1928E-06 to 3.2321E-06
95Mo 4.9799E-05 4.9929E-05 to 5.0765E-05
99Tc 5.9856E-05 6.0280E-05 to 6.4090E-05
101Ru 6.4487E-05 6.3251E-05 to 6.6366E-05
103Rh 5.6463E-05 5.1041E-05 to 5.8208E-05
109Ag 1.2416E-05 1.0524E-05 to 1.4490E-05
133CS 6.5257E-05 6.2829E-05 to 6.6172E-05
43Nd 4.3169E-05 4.4116E-05 to 4.5385E-05
145Nd 3.1278E-05 3.0386E-05 to 3.2104E-05
147Sm 7.8905E-06 9.9452E-06 to 1.0734E-05
149Sm 3.6852E-07 3.7494E-07 to 4.2796E-07
15Sm 1.5873E-05 1.5410E-05 to 1.6679E-05
15Sm 1.5196E-06 1.4779E-06 to 1.8916E-06
152Sm 6.3090E-06 6.3335E-06 to 7.9580E-06
153Eu 7.5868E-06 8.1650E-06 to 8.6716E-06
55Gd 3.6011E-07 3.3400E-07 to 1.0081E-06











Table 5-31. EOC 1 results for Case B of the Phase IVB Benchmark.
Nuclide MCNPX Range of Values from
(atoms/b-cm) Other Depletion Codes
(atoms/b-cm)
234U 2.1859E-07 2.1683E-07 to 2.2681E-07
235U 4.2074E-05 4.1730E-05 to 4.2085E-05
236U 3.1619E-06 3.0287E-06 to 3.1900E-06
238U 2.2041E-02 2.2000E-02 to 2.2049E-02
238Pu 1.1757E-06 3.7913E-07 to 7.1826E-07
239Pu 5.6960E-04 5.6280E-04 to 5.7659E-04
24Pu 1.5241E-04 1.4600E-04 to 1.5008E-04
241Pu 6.0895E-05 5.6781E-05 to 6.3800E-05
242Pu 7.1572E-06 5.3572E-06 to 6.1170E-06
237Np 1.5688E-06 1.3364E-07 to 1.6488E-06
241Am 1.6417E-06 1.3107E-06 to 1.4360E-06
243Am 9.6882E-07 6.2689E-07 to 8.0100E-07
2Cm 1.8012E-07 1.5269E-07 to 1.7950E-07
2Cm 2.0951E-09 1.6814E-09 to 2.0300E-09
244Cm 2.4339E-07 9.3251E-08 to 1.3100E-07
24Cm 1.1257E-08 3.4090E-09 to 4.6024E-09
95Mo 1.3153E-05 1.2181E-05 to 1.8204E-05
99Tc 2.1838E-05 2.1900E-05 to 2.2950E-05
101Ru 2.1979E-05 2.1769E-05 to 2.3058E-05
103Rh 2.0336E-05 1.8759E-05 to 2.1765E-05
109Ag 4.3321E-06 3.6778E-06 to 5.5840E-06
133CS 2.4560E-05 2.3439E-05 to 2.4574E-05
143Nd 1.5453E-05 1.5100E-05 to 1.6346E-05
145Nd 1.0969E-05 1.1117E-05 to 1.1454E-05
147Sm 7.6879E-07 7.6500E-07 to 7.8701E-07
149Sm 2.3529E-07 1.9930E-07 to 2.2432E-07
15Sm 4.7705E-06 4.9051E-06 to 5.2700E-06
15Sm 8.6623E-07 8.8308E-07 to 1.0200E-06
152Sm 2.8949E-06 2.9600E-06 to 3.3370E-06
153Eu 2.0836E-06 2.0570E-06 to 2.2659E-06
55Gd 4.5104E-09 1.8000E-09 to 7.6886E-09











Table 5-32. EOC 2 results for Case B of the Phase IVB Benchmark.
Nuclide MCNPX Range of Values from
(atoms/b-cm) Other Depletion Codes
(atoms/b-cm)
234U 1.8049E-07 1.7098E-07 to 1.9333E-07
235U 2.9106E-05 2.8760E-05 to 2.9257E-05
236U 5.5884E-06 5.3234E-06 to 5.5600E-06
238U 2.1773E-02 2.1767E-02 to 2.1800E-02
238Pu 2.6720E-06 2.9817E-07 to 2.0587E-06
239Pu 3.7031E-04 3.6608E-04 to 3.8300E-04
24Pu 1.8694E-04 1.8070E-04 to 1.8624E-04
241Pu 9.7981E-05 9.4093E-05 to 1.0080E-04
242Pu 2.2218E-05 1.9550E-05 to 2.1840E-05
237Np 3.0365E-06 4.5945E-07 to 3.1897E-06
241Am 4.1152E-06 3.6692E-06 to 3.9280E-06
243Am 4.3621E-06 3.6612E-06 to 4.7300E-06
2Cm 8.9769E-07 7.7886E-07 to 8.9591E-07
2Cm 4.3621E-06 1.6565E-08 to 2.0900E-08
244Cm 1.8948E-06 1.0277E-06 to 1.4100E-06
24Cm 1.4495E-07 6.0866E-08 to 8.6913E-08
95Mo 2.9951E-05 2.9300E-05 to 3.5096E-05
99Tc 4.1181E-05 4.1730E-05 to 4.3780E-05
101Ru 4.3022E-05 4.2699E-05 to 4.5057E-05
103Rh 3.5805E-05 3.3166E-05 to 3.7099E-05
19Ag 7.7718E-06 6.5223E-06 to 9.7410E-06
133CS 4.5646E-05 4.3917E-05 to 4.6063E-05
43Nd 2.8262E-05 2.8500E-05 to 3.0200E-05
145Nd 2.0952E-05 2.1048E-05 to 2.1882E-05
147Sm 1.9461E-06 2.2520E-06 to 2.2924E-06
149Sm 2.0460E-07 1.7336E-07 to 1.9700E-07
15Sm 1.0170E-05 1.0500E-05 to 1.1600E-05
15Sm 8.4275E-07 8.4937E-07 to 1.0900E-06
152Sm 4.7605E-06 4.8514E-06 to 5.9030E-06
153Eu 4.8992E-06 5.1650E-06 to 5.5763E-06
55Gd 9.7191E-09 3.8100E-09 to 1.1850E-08











Table 5-33. EOC 3 results for Case A of the Phase IVB Benchmark.
Nuclide MCNPX Range of Values from
(atoms/b-cm) Other Depletion Codes
(atoms/b-cm)
234U 1.1520E-13 1.3174E-07to 1.7954E-07
235U 1.8445E-05 1.8119E-05 to 1.8791E-05
236U 7.1794E-06 6.8091E-06 to 7.0700E-06
238U 2.1481E-02 2.1467E-02 to 2.1500E-02
238Pu 5.5067E-06 2.2123E-07 to 4.7927E-06
239Pu 2.5259E-04 2.4767E-04 to 2.7200E-04
24Pu 1.8124E-04 1.7470E-04 to 1.8437E-04
241Pu 1.0690E-04 1.0323E-04 to 1.0850E-04
242Pu 4.3597E-05 3.9815E-05 to 4.3080E-05
237Np 4.1982E-06 8.5608E-07 to 4.4424E-06
241Am 4.9089E-06 4.9773E-06 to 5.4746E-06
243Am 1.0121E-05 9.3994E-06 to 1.1700E-05
2Cm 1.9925E-06 1.5821E-06 to 1.8030E-06
2Cm 6.2628E-08 4.8232E-08 to 5.9200E-08
244Cm 6.7113E-06 4.1608E-06 to 5.4200E-06
24Cm 6.6887E-07 3.2603E-07 to 4.7115E-07
95Mo 4.4224E-05 4.5000E-05 to 5.0570E-05
99Tc 5.7742E-05 5.9218E-05 to 6.2220E-05
101Ru 6.2895E-05 6.2300E-05 to 6.5908E-05
103Rh 4.3988E-05 4.2315E-05 to 4.7011E-05
109Ag 1.0536E-05 8.5551E-06 to 1.2920E-05
133CS 6.2697E-05 6.1230E-05 to 6.4442E-05
143Nd 3.7203E-05 3.8700E-05 to 4.0746E-05
145Nd 2.9770E-05 2.9641E-05 to 3.1166E-05
147Sm 2.7210E-06 3.5820E-06 to 3.6688E-06
149Sm 1.2934E-07 1.4821E-07 to 1.6900E-07
15Sm 1.5432E-05 1.5880E-05 to 1.7900E-05
15Sm 7.8954E-07 8.0820E-07 to 1.1000E-06
152Sm 5.6451E-06 5.7314E-06 to 7.4850E-06
153Eu 7.1295E-06 7.9849E-06 to 8.6528E-06
55Gd 8.0878E-09 6.3000E-09 to 1.5641E-08











Table 5-34. Five year cooling time results for Case B of the Phase IVB Benchmark.
Nuclide MCNPX Range of Values from
(atoms/b-cm) Other Depletion Codes
(atoms/b-cm)
U 4.4485E-07 1.3174E-07 to 4.1870E-07
235U 1.8480E-05 1.8119E-05to 1.8830E-05
236U 7.2745E-06 6.9037E-06 to 7.0634E-06
238U 2.1481E-02 2.1467E-02 to 2.1493E-02
238Pu 7.2430E-06 2.1267E-07 to 6.3483E-06
239Pu 2.5476E-04 2.4983E-04 to 2.7284E-04
24Pu 1.8232E-04 1.7540E-04 to 1.8510E-04
241Pu 8.3970E-05 8.1013E-05 to 8.5710E-05
242Pu 4.3597E-05 3.9817E-05 to 4.3090E-05
237Np 4.3965E-06 8.7398E-07 to 4.6418E-06
241Am 2.7709E-05 2.6942E-05 to 2.8303E-05
243Am 1.0125E-05 9.3994E-06 to 1.0616E-05
242Cm 9.8698E-10 7.5522E-10 to 1.4676E-09
2Cm 5.5486E-08 4.2821E-08 to 5.1277E-08
244Cm 5.5500E-06 3.4362E-06 to 3.8927E-06
24Cm 6.6887E-07 3.2603E-07 to 4.7096E-07
95Mo 5.0010E-05 5.0570E-05 to 5.1528E-05
99Tc 5.7945E-05 5.9428E-05 to 6.2460E-05
101Ru 6.2895E-05 6.2654E-05 to 6.5909E-05
103Rh 4.7293E-05 4.2315E-05 to 4.9970E-05
109Ag 1.0552E-05 8.5551E-06 to 1.2940E-05
133CS 6.3248E-05 6.1230E-05 to 6.4442E-05
143Nd 3.8029E-05 3.9608E-05 to 4.0746E-05
45Nd 2.9778E-05 2.9642E-05 to 3.1176E-05
147Sm 6.8077E-06 8.7798E-06 to 9.5399E-06
149Sm 1.7692E-07 2.0516E-07 to 2.2260E-07
15Sm 1.5432E-05 1.5880E-05 to 1.7344E-05
15Sm 7.7139E-07 7.7864E-07 to 9.9180E-07
152Sm 5.6451E-06 5.7314E-06 to 7.4860E-06
153Eu 7.1797E-06 7.9849E-06 to 8.7110E-06
55Gd 3.3990E-07 3.2182E-07 to 9.2307E-07

In both cases, for most actinides and fission products, values were predicted within the

range of values of the other depletion codes available in the benchmark. General trends of

buildup and depletion also fit the data from other depletion codes available in the benchmark.

However, the buildup of Cm isotopes is larger in both Cases A and B as compared to other

depletion codes. Since this benchmark is code-code comparison benchmark, deviation from









expected buildup must be the result of methodology or utilized data differences. First of all,

though metastable isotope representation is available in MCNPX, for this benchmark, metastable

isotopes were not tracked in particle transport. However, historically in the data releases for the

MCNP codes, 242mAm has been listed with a zaid of 95242 instead of the 95642 as expected from

equations 3-9 and 3-10. As a result of the heritage of the data release, MCNPX depletion

expects Am-242m to be listed as 95242; however, the cross sections for these calculations were

generated by an external source that decided to generate 242Am as 95242. The end result of this

confusion is that in particle transport reaction rates were calculated for 95242 using the 242Am

cross sections; however, these reaction rates were converted into destruction and creation

coefficients for 242mAm, and the 63-group match approach was used to generate coefficients for

242Am. The isomer and the ground state isotope do posses different reaction rate probabilities;

242mAm has a long half life (141 years) and a larger resonance integral in the harder MOX

spectrum as compared to the ground state 242Am. 242mAm either fissions, possessing a large

fission cross section, or captures a neutron to become Am-243, which decays to Cm-243. The

nuclide Am-242 decays quickly to Cm-242 which then decays to Pu-238. 242Am has a higher

capture-to-fission ratio than 242mAm and a larger radiative capture resonance integral in the

harder MOX spectrum; therefore, using the 242Am cross section instead of 242mAm will lead to

more captures resulting in greater production of Cm-243 Cm-244 and Cm-245. This hypothesis

seems to be validated by the fact that Cm-243, Cm -244 and Cm-245 seem to be predicted higher

than the range of values. Other possible hypothesized differences in calculation may result from

the CINDER90 depletion algorithm. Further working of the CINDER90 algorithm will be

required in order to validate the latter hypothesis. The results do, however, suggest good









agreement with the other available depletion codes, and do show that MCNPX is capable of

modeling MOX fuel assemblies.

Further Considerations

In the analyzed benchmarks either single pin or single assembly geometries were analyzed.

Reflective boundary conditions were assumed for each boundary of the examined geometries. In

a real reactor operating scenario, net current at the pin or assembly boundary is never zero.

Therefore in order to truly test the validity of the methodology for a certain application, a full

core analysis would be required. In order to preserve exact accuracy, a full core analysis

involves detailed initial isotope and operation data. Most commercial data, at this level of detail,

is proprietary and for the moment unavailable to the general public.

Another major issue to consider is that the NEA only has available accepted benchmark

data for roughly 14 LWRs. The necessary measured data to test a depletion methodology versus

more complicated reactor designs is not readily available or yet to be validated. Development of

time-dependent nuclear reactor experiments and the radiochemical analysis involved in exacting

isotope concentrations for analysis is expensive. Unfortunately, this concern has limited the

amount of available benchmark data for advanced reactor designs. However, as the United States

attempts to meet the initiatives of the GNEP program and develop further advanced reactor

technologies, experiments should be developed, and valuable data should be extracted, leading to

a plethora of new benchmarks to prove the viability of advanced methodologies.









CHAPTER 6
CONCLUSIONS AND SUMMARY

The MCNPX depletion capability offers complete, relatively easy-to-use depletion

calculations in a single, well established, radiation transport Monte Carlo code. The capability

links the well-established MCNPX steady-state reaction rate calculator to the CINDER90

temporal nuclide inventory calculator. MCNPX is not limited by typical approximations utilized

in deterministic methodologies. The MCNPX steady-state reaction rate calculator implements 3-

dimensional continuous energy Monte Carlo in order to determine true interaction rate behavior

for any type of material and geometry combination. The use of CINDER90 allows for the most

potential robustness in modeling capability because the Linear Markov Chain method does not

rely upon parsing a matrix into a solvable form and approximating depletion behavior of certain

nuclides. CINDER90 simply casts the set of coupled depletion equations into sets of linear

transmutation chains, solving each chain for partial nuclide densities, and then summing those

nuclide densities to determine total nuclide concentrations. This algorithm allows any type of

depletion system, fast burning or slow burning, to be modeled exactly. This work differs from

prior efforts in Monte Carlo linked depletion development in that the capability has been

incorporated into a widely used and accepted radiation transport package. Each feature

implemented in the methodology has been tested in order to provide adequate justification for

implementation instead of simply arbitrarily incorporating methodologies for the sake of linking

to independent code packages. This capability attempts to track as many nuclides as data permits

while tracking all important reaction rates, and accounting for average interaction rate behavior

in order to achieve a highly accurate depletion solution that attempts to account for true system

physics behavior.









Accomplishments


Many features have been implemented into the methodology and tested in order to enhance

the versatility of the capability as compared to other available code packages. The following

features were accomplished as part of this work:

1. A comprehensive input structure that gives the user versatility in modeling
capability was developed.

2. An organized easy to understand output structure that groups data in terms of
average system parameters and individual burn material data was provided.

3. Isotope tracking was simplified by providing predefined fission product tier sets
which contain all available fission products containing ENDF/B VII.0 transport
cross sections as well as CINDER90 fission yield information. Testing
demonstrated that implementing the maximum amount of fission products better
mimics the results of the benchmarked MONTEBURNS depletion code.

4. A mechanism was developed for automatically determining and tracking all
immediate daughter reaction isotopes.

5. A capability was created in order to allow the user to manually adjust the
concentration of any nuclide at any time step by specifying the atom fraction,
weight fraction, atom density or weight density of the nuclide to be changed.

6. A structure was engineered for including available metastable isotopes in
continuous energy reaction rate tracking and transforming that the nuclide's
information for use in the depletion calculation in CINDER90.

7. A methodology was established for tracking in continuous energy the (n, fission),
(n,y), (n,2n), (n,3n), (n,p), and (n,a) interaction rates for predefined nuclides.
Testing has demonstrated that the inclusion of these reaction rates, for a simple pin
cell calculation, demonstrates improvement in actinide predictability as compared
to simply tracking (n, fission) and (n,y).

8. A methodology was conceived for determining system averaged flux normalization
parameters used in determining global neutron source, which is also used in
calculating the total magnitude of the flux. This capability was further augmented
to work with the repeated structures feature in MCNPX.

9. An algorithm was developed to automatically determine the proper energy
dependent fission yield to be used in a given burn material by determining the
energy band containing the majority of fission reactions and using that
corresponding fission yield.









10. A technique for adjusting depletion coefficients was developed in order to account
for average behavior over a given time step by incorporating a cross section
averaging technique that utilizes spectra calculation at both the predictor and
corrector steps. Two independent methods were tested, and the results demonstrate
the MONTEBURNS "like" method achieved similar results to the CELL-2 "like"
method, and because the MONTEBURNS "like" method achieved the result
incorporating less memory usage, this method was chosen for implementation in
MCNPX.

11. The default decay chain convergence criterion for use in CINDER90 was
determined based on testing. A keyword value was implemented in order to allow
user control of this criteria

Analysis of Benchmark Results

Three different benchmarks were also completed in order to validate the current

methodology. The OECD/ NEA Phase IB benchmark results showed that the code was capable

of modeling an adjusted fuel-to-moderator ratio pin cell and achieve results within the range of

other depletion methodologies. Though the modeled geometry for this benchmark did not

account for the true spatial behavior witnessed in the experiment, comparison to experiment did

demonstrate ability to achieve adequate results. The H. B. Robinson benchmark demonstrated

ability to model real core behavior, of a typical PWR assembly, and achieve accurate results if

appropriate burnup steps were implemented. The OECD/ NEA Phase IVB benchmark results

demonstrated the ability to model MOX assemblies and achieve the similar results to other

depletion methodologies. Though these calculations only represent the ability to model a PWR

assembly, uranium-dioxide or MOX fueled, placed in an operating scenario in which the

assembly boundaries may be modeled as infinitely reflected, the results do suggest positive

evidence that the capability can model depletion systems adequately.

Differences between measured and calculated values, or code-to-code differences, were a

result of the following issues: accuracy of the applied nuclear data, treatment of normalization

parameters, inadequacy of the implemented time steps, differences in neutron spectra as









compared to a critical spectrum calculation, spatial reactivity effects of the modeled geometry,

and statistical variance of the reaction rates and rare events. ENDF/B VII.0 cross sections were

utilized in particle transport. Due to the recent availability of this data, the pedigree of the data is

not exactly known as it has not been widely tested. The cross sections implemented in the

calculation were generated utilizing an automated subroutine with resonance reconstruction

tolerances that may or may not be adequate for the applied application. The ENDF/B VII.0 data

may also not be complete and therefore may not contain cross section information for every

reaction of interest. However, MCNPX depletion assumes that if an isotope is used in particle

transport and contains an accessible cross section file, then data must exist for each reaction type.

Though the methodology only requires cross section information for (n, fission), (n,y), (n,2n),

(n,3n), (n,p), and (n,a) interactions, it may be possible that some of these reactions do not exist in

the cross section data file. Unfortunately, such an instance would result in the code assuming a

zero cross section for that reaction type even though a nonzero reaction rate might be available in

the CINDER90 library file. Most of the "other depletion code" results were previously

generated using ENDF/B IV, V, or VI data and therefore discrepancies in the ENDF/B VII.0 data

as compared to these libraries also resulted in differences in calculated values. CINDER90 also

contains transmutation cross sections for 3400 nuclides and over 3000 of those nuclides do not

contain available ENDF data. Therefore cross section determination codes were used to

calculate the data implemented in CINDER90 library file. Unfortunately, the pedigree of the

cross section determination code is version dependent and has evolved in accuracy as better

models have become available. Therefore the CINDER90 data for these nuclides may have also

differed from what is used by other codes leading to further error in calculation when comparing

to other tools. Pedigree of this implemented data will only be proven with further benchmarking









and testing. The buildup of Cm isotopes, in the Phase IVB Benchark, was larger for both cases

when compared to other depletion codes. Though metastable isotope representation was

available in MCNPX, for this benchmark, metastable isotopes were not tracked in particle

transport. The confusion in utilizing the 242Am cross section as opposed to the 242mAm resulted

in further calculation error. The isomer and the ground state isotope do posses very different

reaction rate probabilities. Ground state 242Am has a larger resonance integral over the harder

MOX spectrum than 242Am. 242Am also decays into 242Cm which then decays to 238Pu, or

fissions. The 1st excited state isomer 242mAm absorbs a neutron and either fissions or becomes

243Am which later becomes other Cm isotopes. Due to the differences in capture-to-fission ratio

between the ground and excited states, and the difference in resonance integral over the harder

MOX spectrum, the mix up in isomer representation caused more buildup of 244Cm and thus

more buildup of 245Cm. The discrepancies in 238Pu production were also probably a function of

the misrepresentation of this isomer.

MCNPX determines the global neutron source, for use as a flux normalization constant, by

calculating system averaged parameters. Unfortunately, these system average parameters only

exist for certain actinides. One example is recoverable energy per fission. Though an energy

dependent fission cross section exists for almost every actinide, prompt fission release energy

information only exists for 1/3 of all fissionable actinides. For all other actinides, 235U based

parameters are assumed in MCNPX depletion. MCNPX depletion currently does account for the

true recoverable energy per fission. An approximation is made in order to try and determine the

delayed energy contribution by multiplying the prompt fission release energy by a constant that

is assumed to account for delayed energy. The delayed energy contribution is unfortunately

system dependent. For these benchmarks, however, the geometry modeled was not the actual









experimental geometry, the full core was not modeled, and therefore accounting for the true

recoverable energy per fission in the adjusted geometry would not be possible. Also

understanding how the delayed energy is deposited involves knowing the emission spectra of

that energy and transporting that energy to where it is deposited. As mentioned previously, even

for the majority of nuclides containing ENDF/B VII.0 transport cross sections, capture gamma

photon emission spectra is unknown. An initial calculation was performed to determine the mass

defect integral energy contribution. However, a photon containing the integral energy from the

mass defect transports differently than the actual photons spectra generated from the reaction.

Therefore due to lack of photon spectra information, the applied constant was deemed most

acceptable for accounting for delayed energy contribution, and this constant ultimately affected

the solution outcome. For full core modeling at critical, the delayed energy contribution only

scales as a function of neutrons emitted per fission therefore this value only changes between

2.43-3 neutrons per fission as function of building up higher actinides, and therefore the delayed

energy contribution change should not be very significant as function of burnup for these types

of systems.

For each calculation, common time durations using varied powers were implemented in

order to burn each of the benchmarks to the required burnups. Unfortunately, this type of

burning led to burning the system with varied burnup steps. If reactivity contributing isotope

changes density too significantly during a burn step, then it is possible that the constant creation

and destruction coefficient approximation is not valid for that time step. The most probable

reason for the benchmark calculations achieving poorer predictability for higher burnups was

related to the fact that the burnup steps for the higher burnup cases may have been too long in









order to assume the constant coefficient approximation. However, the results for the lower

burnup cases did show excellent agreement in code-to-code and experimental comparison.

Each of the modeled geometries did not incorporate an approximation to account for

critical spectrum. When the pin or assembly is in the reactor core, the geometry is burned

assuming the geometry is in a critical configuration. Therefore the true leakage in and out of the

modeled benchmark geometries varies as function of burnup and is not a true reflective boundary

condition. In deterministic based codes, critical spectrum is approximated by adjusting the flux

by a normalization parameter generated from a buckling adjustment calculation. Not modeling a

geometry utilizing the critical spectra will cause the propagation of nuclide generation and

depletion to be different from what is expected from the actual geometry because the captures,

fissions, and scatters will be different than what is expected from the critical spectrum. Because

the system is not modeled accounting for the true operating behavior, the spatial reactivity

behavior will also be affected. For the Phase IB Benchmark, the adjusted-fuel-to-moderator ratio

pin cell calculation does not account for the true spatial behavior of the reactions because the pin

dimensions have been augmented in order to achieve similar global reactivity behavior.

However, the actual interaction rate behavior within the pin for each reaction type is not

preserved. Therefore the evolution of the temporal nuclide inventory will be in error as the true

reaction rates differ from the calculation. Even in the H. B. Robinson and Phase IVB

benchmarks, where a full assembly was modeled, true interaction rate behavior is not preserved

because the reflective boundary condition is not indicative of the expected leakage condition for

those assemblies during actual reactor operation. Only modeling the full reactor, with the actual

operating conditions will achieve the most accurate result.









Most codes, deterministic/Monte Carlo based, have historically had issues in calculated

237Np concentration as well as other downstream nuclides produced from the 237Np transmutation

path. As determined from the previous testing, 237Np and therefore 238Pu predictability is

improved by accounting for the energy integrated collision rate of (n,2n) reactions (i. e. 238U + n

--> 2n+ 237U; 237U --> 237Np; 237Np + n --> 23Np --> 238Pu (rare event!) or 239Pu + n --> 23Pu

+2n (very rare event!). It also possible to form 237Np from either 235U +n --> no fission --> 236U;

236U +n-->237U -->237Np (only account for 10-20% of reactions with 235U) or alpha decay from

241Am (rare event due to long half life). 238Pu is also formed from decay of 242mAm -->242Cm --

>238Pu. Where 242mAm is formed from multiple Pu isotope captures. The common theme in the

miscalculation of these isotopes is related to properly sampling rare events. Therefore why not

try a type of forced reaction type sample? In "Nonanalog Implementations of Monte Carlo

Isotope Inventory Analysis," the authors tried to develop a method of stochastically solving the

actual depletion equations (instead of just stochastically solving for coefficients to be applied for

a deterministic temporal nuclide inventory solution).99 They tried a forced collision method for

sampling rare events (rare interactions), and found that forced collision sampling of rare events

for a finite amount of histories decreased the amount of sampled non-rare events leading to an

inadequate sampling of non-rare events and ultimately affecting the transmutation path outcome.

Forced collision sampling may affect some part of the analyzed problem when an artificial bias

is introduced. They found that the only way to really cure this issue was to use more histories.

Therefore this evidence suggests that running more histories may fix the problem of sampling

rare events and therefore lead to better prediction of rare event isotopes. Further testing of a

variety of benchmarks will be required in order to truly validate this hypothesis.









Future Work

Further enhancements to the MCNPX depletion methodology will be required in order to

enhance the usefulness of the technology. The current MCNPX 2.6.F depletion method does not

currently account for critical spectrum adjustments or posses a method to incorporate thermo

hydraulic feedback, which is essential for modeling true system physics.2 Critical spectrum

may be calculated by weighting the impinging reflected leakage flux with the mean free path and

continuing to adjust the impinging flux until criticality is reached. More or less weight would be

given to a reflected neutron depending upon if the system was initially sub or super critical. This

type of calculation would involve tagging and tracking all reflected neutrons that contribute to

reactivity. However, this approximation must be thoroughly tested versus a wide variety of

benchmarks in order to prove validity of the method. Thermohydraulic feedback may be

incorporated into MCNPX by linking MCNPX to a code that takes power distribution data in

order to generate temperature distributions. Unfortunately, the calculation of thermohydraulic

feedback is an iterative process, because the effects are nonlinear and dependent upon each other,

and therefore a computationally expensive process when using Monte Carlo. However, this type

calculation is required in order to preserve true system physics. The depletion methodology also

does not propagate number density error, or possess functionality to alter the geometry during

the burnup.98 The relative error associated with the isotope concentration calculations are a

function of the relative error of the flux used to calculate the collision densities, as well as the

isotopes' relative error from the previous calculation. Knowledge of the propagation of the

number density error is required in order to understand the precision of the calculated result and

therefore must be incorporated in further versions of the capability. The evidence of the

benchmark calculations does suggest that further investigation is also required in the passby

calculation in the CINDER90 algorithm. Many tests exist for determining progeny generation









and it is possible that the ordering of these tests may affect the solution outcome. However,

these issues along with other functionality concerns will be addressed in further development of

the capability.

This capability does set up a significant foundation, in a well established and supported

radiation transport code, for further development of a Monte Carlo-linked depletion

methodology. Such a capability will be essential to the future development of advanced reactor

technologies, and detector systems to monitor those reactor technologies, that violate the

limitations of current deterministic based methods.










APPENDIX A
SAMPLE INPUT

In the following example, a 4 X 4 fuel pin array is modeled in order to demonstrate the

necessary keyword implementation for burning a repeated structure while also utilizing manual

concentration changes at specific time steps:


1 1 6.87812e-2 -1 u=2 imp:n=l vol=192.287 $fuel
3 2 4.5854e-2 1-2 u=2 imp:n=l vol=66.43 $ clad
4 3 7.1594e-2 2 u=2 imp:n=l vol=370.82 $ water
6 4 6.87812e-2 -1 u=3 imp:n=l vol=192.287 $fuel
7 5 4.5854e-2 1-2 u=3 imp:n=l vol=66.43 $ clad
8 6 7.1594e-2 2 u=3 imp:n=l vol=370.82 $ water
10 0 -3 4 -56 u=l imp:n=l lat=l fill=0:1 0:10:0
23
23



BURN TIME=50,10,500
MAT= 14
POWER= 1.0
PFRAC= 1.0 0 0.2
OMIT= 1,8,6014,7016,8018,9018,90234,91232,95240,95244
4,8,6014,7016,8018,9018,90234,91232,95240,95244
BOPT= 1.0, -4
AFMIN= le-32
MATVOL= 384.57 384.57
MATMOD= 2
1
1 -4 1 94238 le-6
2
2 -1 2 94238 le-6 94241 le-6
-4 1 94238 le-6


The following example utilizes a lattice containing universes 2 and 3 which are both

repeated twice in the lattice. Universe 2 is comprised of cells 1, 3 and 4, where cell 1 contains

material 1, and universe 3 is comprised of cells 6, 7 and 8, where cell 6 contains material 4. The

MAT keyword specifies that both materials 1 and 4 will be burned. The combination of the

TIME, POWER and PFRAC keywords specify that these materials will be burned first for 50

days at 100% of 1MW, then decayed for 10 days, and then finally burned for 500 days at 20% of

1 MW.











The BOPT keyword specifies that the following options will be invoked: Q-value

multiplier = 1.0, tier 1 fission products, output ordered by zaid, the output is printed at the end of

each kcode run and only tabular transport cross sections are utilized. Because tabular transport

cross sections do not exist for every isotope generated, and OMIT card is required in order to

omit these isotopes from the transport process. The transmutation of these isotopes is accounted

for by sending a 63-group flux from MCNPX to be matched to a 63-group cross section set

within CINDER90 which will be energy integrated to determine a total collision rate. The OMIT

card in the example omits 8 isotopes from material 1 and 8 isotopes from material 4. The

AFMIN card states that only isotopes possessing an atom fraction below le-32 will be omitted

from the transport calculation.

Since repeated structures are utilized in the example, a MATVOL keyword is required in

order to calculate the track length estimated reaction rates in each repeated structure. In this

example, since material 1 and 4 are repeated twice and each material possesses a volume of

192.287, MATVOL keyword entries of 384.57 (192.287*2) were required for each material

being burned.

A MATMOD keyword is used in order to manually change the concentration of certain

isotopes at specified time steps. In this example, manual isotope concentration changes are to be

completed at 2 time steps. At time step 1, material 4 will have the atom density of 1 isotope

changed: 94238 will be set to le-6 atoms/barn-cm. At time step 2, material 1 will have the atom

densities of 2 isotopes changed: material 1 will have isotope 94238 changed to le-6 atoms/barn-

cm and isotope 94241 will be changed to le-6 atoms/barn-cm, material 4 will have the atom

density of 1 isotope changed: 94238 will be set to le-6 atoms/barn-cm.









LIST OF REFERENCES


1. U.S. Department of Energy Office of Civilian Radioactive Waste Management, Oklo:
Natural Nuclear Reactors, November 2004,
(December 2007).

2. J. J. DUDERSTADT and L. J. HAMILTON, Nuclear Reactor Analysis, John Wiley & Sons,
Inc., New York, NY (1976).

3. E. M. BAUM, H. D. KNOX, and T. R. MILLER, "Nuclides and Isotopes: Chart of Nuclides,
Sixteenth Edition," Knolls Atomic Power Laboratory and Lockheed and Martin Corporation,
Schenectady, NY (2006).

4. E. FERMI, "The Future of Atomic Energy," Technical Information Division, Oak Ridge
Directed Operation of the Atomic Energy Commission Report: 10-11-48-850-11425, Oak
Ridge, TN (1946).

5. S. GLASSTONE and W. H. JORDAN, Nuclear Power and its Environmental Effects,
American Nuclear Society, La Grange Park, IL (1980).

6. R.T. LAHEY and F.J. MOODY, The Thermal Hydraulics of a Boiling Water Nuclear
Reactor, American Nuclear Society, La Grange Park, IL(1979).

7. J. R. LAMARSH, Introduction to Nuclear Engineering, Addison-Wesley Publishing
Company, Inc., Melano Park, CA(1983).

8. K. O. OTT, C. W. TERREL, R. C. BORG and P. J. MAUDLIN, "Integrated Fuel-Cycle
Models for Fast Breeder Reactors," Annuls ofNuclear Engineering, 8, pg. 371-392 (1981).

9. R. G COCHRAN and N. TSOULFANIDIS, The Nuclear Fuel Cycle: Analysis and
Management, American Nuclear Society, Inc., La Grange Park, IL (1999).

10. A. ROMANO, P. HEJZLAR, and N. E. TODREAS, "Fertile-Free Fast Lead-Cooled
Incinerators for Efficient Actinide Buruning," Journal of Nuclear Technology, 147, pg. 368-
386 (2004).

11. G S. CHANG "Neutronic and Burnup Characteristics of an Actinide-Reduced Plutonium
Fuel with Tungsten," Journal ofNuclear Technology, 122, pg. 43-51 (1998).

12. T. A. TAIWO, T. K. KIM, J. A. STILLMAN, R. N. HILL, M. SALVATORES, "Assessment
of a Heterogeneous PWR Assembly for Plutonium and Minor Actinide Recycle," Journal of
Nuclear Technology, 155, pg. 43-51 (1998).

13. X. JIANG and Z. XIE, "Transport-bumup Code Systems and Their Applications for IAEA
ADS Benchmark," Annuls ofNuclear Energy, 31, pg. 213-225 (2004).









14. N. MESSAOUDI and J. TOMMASI, "Fast Burner Reactor Devoted to Minor Actinide
Incineration," Journal ofNuclear Technology, 137, pg. 89-96 (2002).

15. M. DELPECH, H. GOLFIER, A. VASILE, F. VARAINE, L. BOUCHER, "Minor Actinides
in PWRs," International Congress on Advances in Nuclear Power Plants Embedded
International Topical Meeting at the 2006 ANS Annual Meeting, Reno, NV, (2006).

16. United States Department of Energy, Office of Nuclear Energy, Science and Technology
"Advanced Fuel Cycle Initiative: Status Report for FY 2005", Report to congress (2006).

17. G W. MCKINNEY and H. R. TRELLUE, "Transmutation Feature within MCNPX," Los
Alamos National Laboratory Report: LA-UR-04-1572, Los Alamos, NM (2004).

18. United States Department of Energy, Office of Nuclear Energy, Office of Fuel Cycle
Management "Global Nuclear Energy Partnership Strategic Plan", Report to congress:
GNEP-167312 (2007).

19. M. D. LOWENTHAL, "Transmutation in the Nuclear Fuel Cycle: Approaches and Impacts,"
Journal ofNuclear Technology, 138, pg. 284-299 (2002).

20. M. VISOSKY, P. HEJZLAR, M. KAZIMI, "Actinide Transmutation Using Pressurized Water
Reactors," International Congress on Advances in Nuclear Power Plants Embedded
International Topical Meeting at the 2006 ANS Annual Meeting, Reno, NV, (2006).

21. T. TAKEDA, T. YAMAMOTO, M. MIYAUCHI, "Interpretation of Actinide Transmutation,"
Progress in Nuclear Energy, 40, pg. 449-456 (2002).

22. M. SALVATORES, "Nuclear Fuel Cycle Strategies Including Partitioning and
Transmutation," Nuclear Engineering and Design, 235, pg. 805-816 (2005).

23. United States Department of Energy, Global Nuclear Energy Partnership: Improved Nuclear
Safeguards, December 2006,
(December 2007).

24. G F. KNOLL, Radiation Detection and Measurement, John Wiley and Sons, Inc., New York,
NY (1999).

25. H. C. HURIA and R. J. BUECHEL, "Recent Improvements and New Features in the
Westinghouse Lattice Physics Codes," Transactions of the American Nuclear Society, 72, pp.
369-371 (1995).

26. T. P. SHANNON, J. K. WHEELER, and G TOUVANNAS, "TGBLA/PANACEA and
CASMO/MICROBURN analyses of GE9B/GE10 Fuel in the Quad Cities," Transactions of
the American Nuclear Society, 74, pp. 287-290 (1996).









27. C. V. PARKS, "Summary Description of the SCALE Modular Code System," Martin
Marietta Energy Systems, Inc. and Oak Ridge National Laboratory Report: NUREG/CR-
5033 ORNL/CSD/TM-252, Oak Ridge, TN (1987).

28. M. J. KHAN, N. AHMAD, "Proliferation Resistance Potential and Bumup Characteristics of
an Equalibrium Core of Novel Natural Uranium Fueled Nuclear Research Reactor," Annuls
ofNuclear Energy, 32, pg. 612-620 (2005).

29. K. SUYAMA, H. MOCKIZUKI, T. KIYOSUMI, "Revised Bumup Code System SWAT:
Description and Validation Using Postirradiation Examination Data," Nuclear Technology,
138, pp. 97-108.

30. M. EDENIUS, K. EKBERG B. H. FORSSEN, and D. KNOTT, "Casmo-4, A Fuel Assembly
Burnup Program, User's Manual," Studsvik Report: SOA-95/15, Studsvik of America, Inc.,
Kjeller, Norway (1995).

31. N. SUGIMURA, A. YAMAMOTO, T. USHIO, M. MORI, M. TABUCHI, and T. ENDO,
"Neutron Transport Models of AEGIS: An Advanced Next-Generation Neutronics Design
System," Nuclear Science and Engineering, 155, pp. 276-289 (2007).

32. D. KNOTT, E. WEHLAGE, "Description of the LANCER02 Lattice Physics Code for
Single-Assembly and Multibundle Analysis," Nuclear Science and Engineering, 155, pp.
331-354 (2007).

33. K. LASSMANN, C. T. WALKER, J. VAN DE LAAR, "Extensions of the TRANSURANUS
Burnup Model to Heavy Water Reactor Conditions," Journal of Nuclear Materials, 255, pg.
222-233 (1998).

34. A. NUNEZ-CARRERA, J. L. FRANCOIS, G ESPINOSA-PAREDES, "Comparison
between HELIOS Critical-Depletion Calculations and a PWR Thorium Cell Burnup
Benchmark," Annuls ofNuclear Energy, 31, pg. 713-722 (2004).

35. M. L. FENSIN and S. ANGHAIE, "Current Methods of Depletion Analysis," Los Alamos
National Laboratory Report LA-UR-06-3933, American Nuclear Society's 2006 Winter
Meeting, Albuquerque, NM (2006).

36. G S. CHANG J. M. RYSKAMP, "Depletion Analysis of Mixed-Oxide Fuel Pins in Light
Water Reactors and the Advanced Test Reactor", Nuclear Technology, 129, pp. 326-337
(2000).

37. W. HAECK, "An Optimum Approach to Monte Carlo Bum-Up," PhD thesis, Ghent
University, Belgium (2007).

38. E. J. PARMA, "BURNCAL: A Nuclear Reactor Burnup Code Using MCNP Tallies," Sandia
National Laboratory Report: SAND2002-3868, Albuquerque, NM (2002).









39. Z. XU, "Design Strategies for Optimizing High Burnup Fuel in Pressurized Water Reactors",
PhD thesis, Massachusetts Institute of Technology, Cambridge, MA (2003).

40. N. A. HANAN, A. P. OLSON, R. B. POND, J. E. MATOS, "A Monte Carlo Burnup Code
Linking MCNP and REBUS", 1998 International Meeting on Reduced Enrichment for
Research and Test Reactors, Sao Paulo, Brazil (1998).

41. G S. CHANG "MCWO-Linking MCNP and ORIGEN2 For Fuel Bumup Analysis,"
American Nuclear Society's Topical Meeting on Monte Carlo, Chattanooga, TN (2005).

42. J. CETNAR, W. GUDOWSKI, and J. WALLENIUS, "User Manual for Monte-Carlo
Continuous Energy Burnup (MCB) Code-Version 1," draft report, Department of Nuclear
and Reactor Physics, Royal Institute of Technology, Stockholm, Sweden (2005).

43. D. LI, X. ZHONGSHENG L. SHU, "Monte Carlo Transport and Bumup Calculation,"
Annuls ofNuclear Energy, 30, pg. 127-132 (2003).

44. R. L. MOORE, B. G SCHNITZLER, C. A. WEMPLE, R. S. BABCOCK, D. E. WESSOL,
"MOCUP: MCNP-ORIGEN2 Coupled Utility Program", Idaho National Engineering
Laboratory Report: INEL-95/0523, Idaho Falls, ID (1995).

45. H. R. TRELLUE and D. I. POSTON, "User's Maunual, Version 2.0 for MONTEBURNS,
Version 5B," Los Alamos National Laboratory report LA-UR-99-4999, Los Alamos, NM
(1999).

46. M. AOYAMA, S. UCHIKAWA, K. MIKI, K. HIRAMOTO, and R. TAKEDA, "A
Conceptual Design of a Fuel Bundle for Extended Bunup in Boiling Water Reactors,"
Journal ofNuclear Technology, 64, pg. 19-25 (July 1993).

47. N. A HANAN, A. P. OLSON, R. B. POND, and J. E. MATOS, "A Monte Carlo Bumup Code
Linking MCNP and REBUS," Argonne National Laboratory Report: ANL-TD-CP-97492,
Agonne, IL (no date specified).

48. D. J. KELLY, "Depletion of a BWR Lattice Using the RACER Continuous Energy Monte
Carlo Code," International Conference of Mat u tlhe tiiL and Computation, Reactor Physics,
and Environmental Analysis, Portland, OR, (1995).

49. M. DEHART, "High-Fidelity Depletion Capabilities of the SCALE Code System Using
Triton," American Nuclear Society 's 2007 Winter Meeting, Washington, D. C. (2007).

50. D. B. PELOWITZ, "MCNPXTM User's Manual Version 2.5.0," Los Alamos National
Laboratory Report LA-CP-05-0369, Los Alamos, NM (2005).

51. W. B. WILSON, T. R. ENGLAND, E. D. ARTHUR, C. A. BEARD, C. D. BOWMAN, L. N.
ENGEL, A. GAVRON, D. C. GEORGE, L. L. DAEMEN, H. G HIGHEST, III, W. W.
KINNISON, R. J. LABAUVE, D. M. LEE, H. LICHTENSTEIN, P. W. LISOWSKI, D. W.









MUIR, A. P. MUIR, A. P. PALOUNEK, R. T. PERRY, E. J. PITCHER, R. E. PRAEL, R. J.
RUSSEL, G SANDERS, L. S. WATERS, P. G YOUNG and J. J. ZIOCK, "Accelerator
Transmutation Studies at Los Alamos with LAHET, MCNPX and CINDER90," Los Alamos
National Laboratory Report LA-UR-93-3080, Workshop on Simulation of Accelerator
Radiation Environments, Sante Fe, NM (1993).

52. M. D. DEHART, M. C. BRADY, C. V. PARKS, "OECD/NEA Burnup Credit Calculation
Criticality Benchmark Phase I-B Results," Oak Ridge National Laboratory Report: ORNL-
6901, Oak Ridge, TN (1996).

53. O. W. HERMANN, S. M. BOWMAN, M. C. BRADY, C. V. PARKS, "Validation of the
Scale System for PWR Spent Fuel Isotopic Composition Analysis," Oak Ridge National
Laboratory Report: ORNL/TM-12667, Oak Ridge, TN, USA (1995).

54. G J. O'CONNOR and P. H. LIEM "Burn-up Credit Criticality Benchmark PHASE IV-B:
Results and Analysis of MOX Fuel Depletion Calculations," Nuclear Energy Agency Report:
NEA/NSC/DOC(2003)4 (2003).

55. G F. THOMAS and D. H. BARBER, "Numerical Solution of Bateman Systems," Annuls of
Nuclear Energy, 20, pg. 407-414 (1993).

56. M. YAMAMOTO, "Coarse Time-Step Integration Method for Burnup Calculation of LWR
Lattice Containing Gadolinium-Poisoned Rods," Journal ofNuclear Science and Technology,
22, pg.1-15 (1985).

57. G ILAS, B. D. MURPHY, and I. C. GAULD, "Overview of Origen-ARP and its Applications
VVER and RBMK," American Nuclear Society's 2007 Winter Meeting, Washington, D. C.
(2007).

58. A. G CROFF, "ORIGEN2: A Versatile Computer Code for Calculating the Nuclide
Compositions and Characteristics of Nuclear Materials," Journal of Nuclear Technology, 62,
pg. 335-352 (1983).

59. G F. THOMAS and D. H. BARBER, "Stiffness in Radioactive Decay Chains," Annuls of
Nuclear Energy, 21, pg 309-320 (1994).

60. A. G CROFF, "ORIGEN2 A Revised and Updated Version of the Oak Ridge Isotope
Generation and Depletion Code", Oak Ridge National Laboratory Report: ORNL-5621, Oak
Ridge, TN (1980).

61. W. B. WILSON, V DYADECHKO, T. R. ENGLAND, P. MOLLER, H. R. TRELLUE, "A
Manual for CINDER90 Version 06.1 Codes and Data," Los Alamos National Laboratory
Draft Report, Los Alamos, NM (2006).









62. M. L. FENSIN, "Optimum Boiling Water Reactor Fuel Design Strategies to Enhance
Shutdown by the Standby Liquid Control System" Masters Thesis, University of Florida,
Gainesville, FL (2004).

63. E. E. LEWIS and W. F. MILLER, Jr., Computational Methods of Neutron Transport, Ch. 7,
American Nuclear Society, Inc., La Grange Park, IL (1993).

64. H. GREENSPAN. C. N. KELBER, D. OKRENT, Computing Methods in Reactor Physics,
Gordon and Breach Science Publishers, Inc. New York, NY (1968).

65. M. CLARK, K. F. HANSEN, Numerical Methods of Reactor Analysis, Academic Press, Inc.,
New York, NY (1964).

66. I. M. SOBOL, The Monte Carlo Method, Mir Publishers, Moscow, Russia (1984).

67. A. PAUTZ, "Fuel Assembly Calculations Using the Method of Discrete Ordinates," Nuclear
Science and Engineering, 149, pg. 197-210 (2005).

68. D. S. LUCAS, "Attila Modeling with Comparison to Data, MCNP and MCNPX for the
Advanced Test Reactor," International Congress on Advances in Nuclear Power Plants
Embedded International Topical Meeting at the 2006 ANS Annual Meeting, Reno, NV,
(2006).

69. S. G HONG and N. Z. CHO, "Method of Characteristic Direction Probabilities for
Heterogeneous Lattice Calculation," Nuclear Science and Engineering, 132, pg. 65-77
(1999).

70. H. HERMAN, "ENDF-6 FORMATS MANUAL," Brookhaven National Laboratory Report:
BNL-NCS-44945-05-Rev., Long Island, NY (2005).

71. M. L. FENSIN and D. I. POSTON, "Optimum Reflector Configurations for Minimizing
Fission Power Peaking in a Lithium-Cooled, Liquid-Metal Reactor with Sliding Reflectors,"
Los Alamos National Laboratory Report LA-UR-7154, Space Technology and Applications
International Forum 2005, Albuquerque, NM (2005).

72. M. L. FENSIN, J. O. ELLIOTT, R. J. LIPINSKI, D. I. POSTON, "Radiation Shielding
Design and Orientation Considerations for a 1 kWe Heat Pipe Cooled Reactor Utilized to
Bored Through the Ice Caps of Mars," Los Alamos National Laboratory Report LA-UR-05-
8018, Space Technology and Applications International Forum 2006, Albuquerque, NM
(2006).

73. J. L. DEVORE, Probability and Statistics for Engineers and Scientists, Brooks/Cole, Pacific
Grove, CA (2000).

74. J. F. BRIESMEISTER, "MCNP-A General Monte Carlo N-Particle Transport Code," Los
Alamos National Laboratory Report: LA-13709-M, Los Alamos, NM (2000).










75. H. R. TRELLUE, "Development of Monteburns: A Code That Links MCNP and ORIGEN2
in a Automated Fashion for Burnup Calculations" Los Alamos National Laboratory Report
LA-13514-T, Los Alamos, NM (1998).

76. S. M. BOWMAN, M. D. DEHART, and L. M. PETRIE, "Integrated Keno Monte Carlo
Transport for 3-Depletion with SCALE," The American Nuclear Society's Monte Carlo 2005
TopicalMeeting, Chattanooga, TN (2005).

77. S. M. BOWMAN, D. F. HOLLENBACH, M. D. DEHART, B. T. REARDEN, I. C. GAULD,
and S. GOLUOGLU, "SCALE 5: Powerful New Criticality Safety Analysis Tools," 7th
International Conference on Nuclear Criticality Safety, Tokai-mura, Japan (2003).

78. A. S. GERASIMOV, T. S. Zaritskaya, G V Kiselev, and L. A. Myrtsymova, "The Cost of
Transmutation of Fission Products in Nuclear Reactors," Atomic Energy, 94, pg. 154-157
(2003).

79. J. S. HENDRICKS, G W. MCKINNEY, H. R. TRELLUE, J. W. DURKEE, T. L. ROBERTS,
H. W. EGDORF, J. P. FINCH, M. L. FENSIN, M. R. JAMES, D. B. PELOWITZ, and L. S.
WATERS, "MCNPX, Version 2.6.A," Los Alamos National Laboratory Report: LA-UR-05-
8225, Los Alamos, NM (2005).

80. M. B. CHADWICK, P. OBLOZINSKY, M. HERMAN, N. M. GREENE, R. D. MCKNIGHT,
D. L. SMITH, P. G YOUNG R. E. MACFARLANE, G M. HALE, S. C. FRANKLE, A. C.
KAHLER, T. KAWANO, R. C. LITTLE, D. G MADLAND, P. MOLLER, R. D.
MOSTELLER, P. R. PAGE, P. TALOU, H. TRELLUE, M. C. WHITE, W. B. WILSON, R.
ARCILLA, C. L. DUNFORD, S. F. MUGHABGHHAB, B. PRITYCHENKO, D.
ROCHMAN, A. A. SONZOGNI, C. R. LUBITZ, T. H. TRUMBULL, J. P. WEINMAN, D.
A. BROWN, D. E. CULLEN, D. P. HEINRICHS, D. P. MCNABB, H. DERRIEN, M. E.
DUNN, N. M. LARSON, L. C. LEAL, A. D. CARLSON, R. C. BLOCK, J. B. BRIGGS, E.
T. CHENG H. C. HURIA, M. L. ZERKLE, K. S. KOZIER, A. COURCELLE, V.
PRONYAEV and S. C. VAN DER MARCK,"ENDF/B-VII.0: Next Generation Evaluated
Nuclear Data Library for Nuclear Science and Technology", Nuclear Data Sheets, 107, pg.
2931-3060 (2006).

81. T. R. ENGLAND, B. F. RIDER, "ENDF-349 evaluation and compilation of fission product
yields," Los Alamos National Laboratory Report: LA-UR-94-3106, Los Alamos, NM (1994).

82. M. L. FENSIN, J. S. HENDRICKS, G W. MCKINNEY, H. R. TRELLUE, "Advances in
Monte Carlo Depletion Capabilities for MCNPX," Los Alamos National Laboratory Report
LA-UR-05-7895, American Nuclear Society 14th Biennial Topical Meeting of the Radiation
Protection and .\niel/mg Division, Carlsbad, NM, (2006).

83. H. R. TRELLUE, "Reduction of the Radiotoxicity of Spent Nuclear Fuel Using a Two-Tiered
System Comprising Light Water Reactors and Accelerator-Driven Systems,"Los Alamos
National Laboratory Report: LA-14052-T, Los Alamos, NM (2003).










84. J. S. HENDRICKS, G W. MCKINNEY, J. W. DURKEE, J. P. FINCH, M. L. FENSIN, M. R.
JAMES, R. C. JOHNS, D. B. PELOWITZ, L. S. WATERS, F. X. GALLMEIER, "MCNPX,
Version 2.6.C," Los Alamos National Laboratory Report: LA-UR-06-7991, Los Alamos, NM
(2006).

85. M. L. FENSIN, J .S. HENDRICKS, S. ANGHAIE, "Enhanced Monte Carlo Linked
Depletion Capabilities in MCNPX," Los Alamos National Laboratory Report LA-UR-06-
0363, 2006 International Congress on Advances in Nuclear Power Plants Embedded
International Topical Meeting at the 2006 ANS Annual Meeting, Reno, NV, (2006).

86. J. S. HENDRICKS, G W. MCKINNEY, M. L. FENSIN, M. R. JAMES, R. C. JOHNS, J. W.
DURKEE, J. P. FINCH, D. B. PELOWITZ, L. S. WATERS, F. X. GALLMEIER, "MCNPX
Version 2.6.E," Los Alamos National Laboratory Report: LA-UR-07-6632, Los Alamos, NM
(2007).

87. J. S. HENDRICKS, G W. MCKINNEY, M. L. FENSIN, M. R. JAMES, R. C. JOHNS, J. W.
DURKEE, J. P. FINCH, D. B. PELOWITZ, L. S. WATERS, F. X. GALLMEIER, "MCNPX
Version 2.6.F," Los Alamos National Laboratory Report: Draft Report, Los Alamos, NM
(2007).

88. G AUDI, A.H. WAPSTRA, C. THIBAULT, J. BLACHOT and O. BERSILLON "Ame2003:
Atomic Mass Evaluation," November 2003,
(November 2007).

89. S. K. GHOSHAL, M. GUPTA, and V RAJARAMAN, "A Parallel Multistep Predictor-
Corrector Algorithm for Solving Ordinary Differential Equations," Journal of Parallel and
Distributed Computing, 6, pg. 636-648 (1989).

90. A. E. ABOANBER, "Numerical Solution of the chain model of fission product nuclides.
Effect on the reactivity of the reactor," Annuls ofNuclear Energy, 28, pg 923-933 (2001).

91. C. M. KANG R. O. MOSTELLER, and S. LEVY, "Incorporation of a Predictor-Corrector
Depletion Capability into the CELL-2 Code," Transaction of the American Nuclear Society,
45, pp. 729-731 (1983).

92. M. L. FENSIN, J. S. HENDRICKS, H. R. TRELLUE, S. ANGHAIE "Incorporation of a
Predictor-Corrector Methodology and 1-GroupReaction Rate Reporting Scheme for the
MCNPX Depletion Capability," Los Alamos National Laboratory Report LA-UR-06-3925,
American Nuclear Society's 2006 Winter Meeting, Albuquerque, NM (2006).

93. Z. XU, J. RHODES III, K. SMITH, N. GHEORGHIU, "MCNPX-5/ORIGEN-2.2/MCODE-
2.2 versus CASMO-5 Depletion for a Heavy Gd-Poisoned BWR Fuel Assembly," American
Nuclear Society 's 2007 Summer Meeting, Boston, MA (2007).









94. J. S. HENDRICKS, G W. MCKINNEY, H. R. TRELLUE, J. W. DURKEE, J. P. FINCH, M.
L. FENSIN, M. R. JAMES, D. B. PELOWITZ, L. S. WATERS, F. X. GALLMEIER J. S.
HENDRICKS, J. C. DAVID, "MCNPX, Version 2.6.B," Los Alamos National Laboratory
Report: LA-UR-06-3248, Los Alamos, NM (2006).

95. J. S. HENDRICKS, G W. MCKINNEY, M. L. FENSIN, M. R. JAMES, R. C. JOHNS, J. W.
DURKEE, J. P. FINCH, D. B. PELOWITZ, L. S. WATERS, F. X. GALLMEIER "MCNPX,
Version 2.6.D," Los Alamos National Laboratory Report: LA-UR-07-4137, Los Alamos,
NM (2007).

96. M. L. FENSIN, J. S. HENDRICKS, S. ANGHAIE, "The Enhancements and Testing of the
MCNPX Depletion Capability," Los Alamos National Laboratory Report: LA-UR-08-0305,
To be presented at: International Congress on Advances in Nuclear Power Plants Embedded
International Topical Meeting at the 2008 ANS Annual Meeting, Anaheim, CA, (2008).

97. Organisation for Economic Co-operation and Development: Nuclear Energy Agency, "NEA
Expert Group on Bum-up Credit Criticality Safety," September 2003,
(December 2007).

98. T. TAKEDA, N. HIROKAWA, and T. NODA, "Estimate of Error Propagation in Monte-
Carlo Burnup Calculations," Journal of Nuclear Science and Technology, 36, No. 9, pp. 738-
745 (1999).

99. P. PHRUKASAROJANAKUN and P. WILSON, "Nonanalog Implementations of Monte
Carlo Isotope Inventory Analysis," Nuclear Science and Engineering, 156, pp. 164-179
(2007).









BIOGRAPHICAL SKETCH

Michael Fensin was born on February 2 1980, to Dr. Matthew and Dr. Sheryl Fensin.

Michael has one brother, Jeffrey Fensin. During his career as a student at the University of

Florida's Nuclear and Radiological Engineering Department, Michael has completed various

internships. Michael got his first start, from Dr. William Vernetson, as a laboratory technician

for the University of Florida's radiochemistry laboratory and training reactor facility. Michael

then moved on to complete a summer internship in reactor engineering with Southern Nuclear

Company at the plant Vogtle site. During this internship Michael, participated in daily reactor

operation and monitoring tasks. After his summer internship, Michael returned to the University

of Florida to complete his Master of Engineering degree. During this degree, Michael also

completed a six month internship with Global Nuclear Fuels (GNF), a General Electric

Company, where he collaborated his master's degree research efforts with work completed at

GNF. At GNF, Michael developed optimum fuel bundle design strategies in order to meet the

limiting constraints of shutdown reactivity management. Michael then moved on to Los Alamos

National Laboratory where he completed research in the area of compact reactor design for space

applications. Michael then moved to the, MCNPX code development group, where he

completed the work for this dissertation. After graduation, Michael will stay at Los Alamos

National Laboratory and continue his work on advanced Monte Carlo linked depletion strategies

for applied reactor safeguards efforts.





PAGE 1

1 DEVELOPMENT OF THE MCNP X DEPLETION CAPABILITY: A MONTE CARLO LINKED DEPLETION METHOD THAT AUTOMATES THE COUPLING BETWEEN MCNPX AND CIND ER90 FOR HIGH FIDELITY BURNUP CALCULATIONS By MICHAEL LORNE FENSIN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Michael Lorne Fensin

PAGE 3

3 To the spirit of the west endzone crew

PAGE 4

4 ACKNOWLEDGMENTS This work was initially supported by the Depa rtment of Energy, Office of Nuclear Energy, Advanced Fuel Cycle Initiative, AFCI. The AFCI program then evolved into the Global Nuclear Energy Partnership Program, GNEP, which then c ontinued to support this effort. The author would first like to acknowledge his thesis advisor, Dr. Samim A nghaie, from the University of Florida, for his support and guida nce through the process of deve loping an interesting problem, and then seeing the efforts of solving that probl em till conclusion. The author would then also like to thank, Dr. Edward Dugan, Dr. Alireza Haghighat, Professor James Tulenko, and Dr. William Vernetson for providing the author with a comprehensive education in reactor physics and fuel depletion analysis. The author would also like to thank Dr. Wesley Bolch, Dr. David Gilland, Dr. Travis Knight, Dr. David Hintenlang, from the Universi ty of Florida, for instilling a fulfilling education in the areas of applied radiological engineering. The author would then like to thank Rick Kapernick, Thomas Marcille, Dr. Dave Poston, and Dr. Pratap Sadasivan, from the space reactor design group at Los Alamos Nationa l Laboratory, for their financial and mentor support, giving the author a solid foundation in re actor design consideratio ns. The author would also like to thank Dr. Mike Miller, Dr. Peter Santi, Dr. Marytn Swinhoe and Dr. Steve Tobin, from the reactor safeguards group at Los Alam os National Laboratory, for their guidance and suggestions in the methodology development of cert ain applied capabilities. The author would also like to thank Dr. Holly Trellue, from the reactor design group at Los Alamos National Laboratory for her valued guidance and suppor t from experience in developing a linked depletion capability. Most im portantly the author would like to thank Dr. Gregg Mckinney, Dr. Michael James, Dr. Russell Johns, Dr. Joe Durk ee, Dr. Joshua Finch, Denise Pelowitz, Dr. Laurie Waters and Dr. Jay Elson, for their pa tience, support and guidance in implementing the methodology, developing theory, and testing the code s capabilities. Fina lly, the author would

PAGE 5

5 especially like to thank Dr. John S. Hendricks for his patience and guidance as a superb mentor at Los Alamos National Laboratory.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES................................................................................................................ .......10 ABSTRACT....................................................................................................................... ............12 CHAPTER 1 INTRODUCTION AND BACKGROUND...........................................................................14 Reactor Development History................................................................................................14 Advanced Fuel Cycle Initiative..............................................................................................19 Global Nuclear Energy Partnership........................................................................................21 Advanced Burner Reactor Campaign..............................................................................23 Reactor Safeguards Campaign........................................................................................25 Motivations for a Monte Carlo Depletion Tool in MCNPX...................................................27 Objective...................................................................................................................... ...........29 2 SURVEY OF STATE OF THE ART A PPROACHES TO DEPLETION ANALYSIS........32 The Depletion Equation......................................................................................................... .33 Zero-Dimension Depletion Calculation..................................................................................35 Matrix Expone ntial Method............................................................................................37 The CINDER90 Linear Markov Chain Method..............................................................40 Necessity for a Linked Approach...........................................................................................44 Methods of Steady State R eaction Rate Calculation..............................................................47 Deterministic Approach to Reaction Rate Calculations..................................................48 Monte Carlo Approach to R eaction Rate Calculations...................................................54 Current Capabilities and Methods of State of the Art Depletion Codes.................................63 Implementation Architecture...........................................................................................64 Nuclide Inventory Limitations/ Ca lculation Size C onsiderations...................................68 3 METHODOLOGY DEVELOPMENT...................................................................................71 Approach to Isotope Tracking................................................................................................72 Fission Product Tiers.......................................................................................................74 Isotope Generator Algorithm...........................................................................................79 Manual Time-Dependent Isotope Concentration Changes..............................................81 Metastable Isotope Tracking...........................................................................................85 Calculation of Depletio n Equation Coefficients.....................................................................86 Important Reactions Followed........................................................................................87 Flux Normalization..........................................................................................................94

PAGE 7

7 Automatic Fission Yield Selection................................................................................101 Time Dependent Cross Section Averaging Techniques.......................................................104 MONTEBURNS Method..............................................................................................106 CELL-2 Method............................................................................................................107 Preliminary Analysis.....................................................................................................108 Further Commentary on both Cross Section Averaging Techniques............................112 4 IMPLEMENTATION ARCHITECTURE...........................................................................114 Input Interface................................................................................................................ .......115 Output Structuring............................................................................................................. ...122 5 BENCHMARKS...................................................................................................................128 OECD/NEA Burnup Credit Phase IB Benchmark Specification.........................................129 OECD/NEA Burnup Credit Phase IB Results......................................................................133 Nuclear Data..................................................................................................................139 Treatment of Normalization Parameters.......................................................................140 Inadequate Time Steps..................................................................................................140 Statistical Variance........................................................................................................141 Differences in Neutron Spectra.....................................................................................141 Spatial Reactivity Effects of the Modeled Geometry....................................................142 Further MONTEBURNS Comparisons.........................................................................142 H. B. Robinson Benchmark Specification............................................................................145 H. B. Robinson Benchmark Results.....................................................................................149 OECD/NEA Burnup Credit Phase IVB Benchmark Specification......................................151 OECD/NEA Burnup Credit Phase IVB Benchmark Results................................................156 Further Considerations......................................................................................................... .166 6 CONCLUSIONS AND SUMMARY...................................................................................167 Accomplishments..........................................................................................................168 Analysis of Benchmark Results.....................................................................................169 Future Work...................................................................................................................175 APPENDIX SAMPLE INPUT................................................................................................177 LIST OF REFERENCES.............................................................................................................179 BIOGRAPHICAL SKETCH.......................................................................................................188

PAGE 8

8 LIST OF TABLES Table page 3-1 MOX fuel composition......................................................................................................89 3-2 Borated water composition................................................................................................90 3-3 Emitted and recoverable energy per fission of U-235.......................................................97 3-4 Prompt fission energy releas e data available in MCNPX..................................................98 3-5 Available actinide fission yi eld distributions in CINDER90...........................................102 4-1 OMIT keyword variable tasks..........................................................................................117 4-2 Options invoked for different B2 values.........................................................................118 4-3 Options invoked for different B2 values.........................................................................119 4-4 Available options for the B3 value of the BOPT keyword..............................................119 4-5 MATMOD keyword variable values...............................................................................121 5-1 Examined benchmark cases.............................................................................................129 5-2 Model description for Phase IB Benchmark*..................................................................130 5-3 Fuel composition for Phase IB Benchmark (density =10.045 g/cm3).............................131 5-4 Clad composition for Phase IB Benchmark (6.44 g/ cm3)...............................................132 5-5 Borated water composition for Phase IB Benchmark (0.7569 g/ cm3)............................132 5-6 Operating history data for Phase IB Benchmark.............................................................132 5-7 Operating power (MW) for each case for Phase IB Benchmark.....................................133 5-8 Initial Phase IB Benchmark results a nd comparison of Case A (27.35 MWD/MTU)....133 5-9 Initial Phase IB Benchmark results a nd comparison of Case B (37.12 MWD/MTU).....134 5-10 Initial Phase IB Benchmark results a nd comparison of Case C (44.34 MWD/MTU).....135 5-11 MCNPXc Phase IB Benchmark results and comparison of Case A (27.35 MWD/MTU)....................................................................................................................137 5-12 MCNPXc Phase IB Benchmark results and comparison of Case B (37.12 MWD/MTU)....................................................................................................................138

PAGE 9

9 5-13 MCNPXc Phase IB Benchmark results and comparison of Case C (44.34 MWD/MTU)....................................................................................................................139 5-14 MONTEBURNS Phase IB Benchmark results and comparison of Case A (27.35 MWD/MTU)....................................................................................................................144 5-15 MONTEBURNS Phase IB Benchmark results and comparison of Case A (37.38 MWD/MTU)....................................................................................................................144 5-16 Fuel assembly design data for H. B. Robinson Benchmark............................................147 5-17 Operating data for H. B. Robinson fuel assembly...........................................................148 5-19 The H. B. Robinson Benchmark results using MCNPX compared with SCALE...........149 5-20 The H. B. Robinson Benchmark results using MCNPXc compared with SCALE.........150 5-21 Fuel assembly design data for OECD/NEA Burnup Credit Phase IVB Benchmark.......152 5-22 Initial MOX fuel composition for Case A for Phase IVB Benchmark............................153 5-23 Initial MOX fuel composition for Case B for Phase IVB Benchmark............................154 5-24 Borated water composition for Phase IVB Benchmark (660ppm boron, density 0.7245 g/cm3)...................................................................................................................154 5-25 Zicaloy-2 composition used for clad and guide tube material for Phase IVB Benchmark (density 5.8736 g/cm3)..................................................................................156 5-26 Operating data for Phase IVB Benchmark.......................................................................156 5-27 EOC 1 results for Case A of the Phase IVB Benchmark.................................................157 5-28 EOC 2 results for Case A of the Phase IVB Benchmark.................................................158 5-29 EOC 3 results for Case A of the Phase IVB Benchmark.................................................159 5-30 Five year cooling time results fo r Case A of the Phase IVB Benchmark........................160 5-31 EOC 1 results for Case B of the Phase IVB Benchmark.................................................161 5-32 EOC 2 results for Case B of the Phase IVB Benchmark.................................................162 5-33 EOC 3 results for Case A of the Phase IVB Benchmark.................................................163 5-34 Five year cooling time results fo r Case B of the Phase IVB Benchmark........................164

PAGE 10

10 LIST OF FIGURES Figure page 2-1 Linked Depletion Process..................................................................................................46 2-2 Flux depression in the ne ighborhood of a resonance at E0................................................51 2-3 Flux depression in the ne ighborhood of a resonance at E0................................................53 2-4 Flow diagram for the Monte Ca rlo particle transport process...........................................56 3-1 Monte-Carlo-linked de pletion process model....................................................................72 3-2 Uranium-235 fission pr oduct yield distribution.7..............................................................75 3-3 Fission Yield Distributions for Va rious Plutonium and Uranium Isotopes.81...................76 3-4 Original fission product arra y containing 167 fission products.........................................77 3-5 Resulting time-dependent neutron multip lication from implementing fission products with and without cross section models in MCNPX as compared to the benchmarked MONTEBURNS method...................................................................................................78 3-6 Percent difference of certain fission products from MONTEBURNS using MCNPX Tier-1 fission products with and without cross section models.........................................78 3-7 Percent difference of certain fission products from MONTEBURNS using MCNPX Tier-2 and Tier-3 fission products with cross-section models...........................................79 3-8 Enhanced fission product ar ray containing 220 isotopes...................................................79 3-9 Nuclides created from a decay or nuclear reaction process and ge nerated for particle transport by the isotope generator algorithm.3...................................................................80 3-10 Examined infinitely re flected pin cell geometry................................................................89 3-11 Percent difference of certain referen ce actinides between the OLD MCNPX method and the MONTEBURNS method......................................................................................91 3-12 Percent difference of certain referen ce actinides between the NEW MCNPX method and the MONTEBURNS method......................................................................................92 3-13 Percent difference of certain refere nce fission products between the NEW MCNPX method and the MONTEBURNS method.........................................................................92 3-14 Time-dependent eigenvalue calculations...........................................................................94 3-15 Fission yield selection process diagram...........................................................................104

PAGE 11

11 3-16 MONTEBURNS cross sec tion averaging technique.......................................................106 3-17 CELL-2 Predictor-Corrector method...............................................................................107 3-18 Time dependant neutron multiplication fo r the predictor corrector method tests...........109 3-19 Burn steps implemented to achieve 2191 day burn.........................................................110 3-20 Time dependent neutron multip lication utilizing 5 burn steps........................................111 3-21 Time dependent neutron multip lication utilizing 10 burn steps......................................111 3-22 Percent Difference in keff When Compared with the NoPC 30-Step Case......................112 4-1 Burn card input structure.................................................................................................115 4-2 BURN card referencing burn ma terials m1, m3 and m4...................................................117 4-3 Example collision rate outputs.........................................................................................123 4-4 Example burnup summary table......................................................................................124 4-5 Example individual material burnup data........................................................................124 4-6 Example cropped actinide output at two separate timesteps and fission product output for one time step...................................................................................................125 4-7 Cropped example of print table 220.................................................................................126 5-1 Infinitely reflected pin cell geometry...............................................................................130 5-2 Infinitely reflected pin cell geometry...............................................................................146 5-3 MOX fuel enrichment map for Phase IVB Benchmark...................................................155

PAGE 12

12 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy DEVELOPMENT OF THE MCNPX DEPLETION CAPABILITY: A MONTE CARLO LINKED DEPLETION METHOD THAT AUT OMATES THE COUPLING BETWEEN MCNPX AND CINDER90 FOR HIGH FIDE LITY BURNUP CALCULATIONS By Michael Lorne Fensin May 2008 Chair: Samim Anghaie Major: Nuclear Engineering Sciences Monte Carlo-linked depletion methods have gain ed recent interest due to the ability to more accurately model complex 3-dimesional ge ometries and better tr ack the evolution of temporal nuclide inventory by simulating the actual physical process utiliz ing continuous energy coefficients. The integration of CINDER90 into the MCNPX Monte Carlo radiation transport code provides a high-fidelity completely self-con tained Monte-Carlo-linked depletion capability in a well established, widely accepted Monte Carlo radiation transport code that is compatible with most nuclear criticality (KCODE) particle tracking features in MC NPX. MCNPX depletion tracks all necessary reaction rates and follows as many isotopes as cross section data permits in order to achieve a highly accurate temporal nuclide inventory solution. This work chronicles relevant nuclear hist ory, surveys current methodologies of depletion theory, details the methodology in applied MCNP X and provides benchmark results for three independent OECD/NEA benchmarks. Relevant nuclear history, from the Oklo reactor two billion years ago to the current major United Stat es nuclear fuel cycle development programs, is addressed in order to supply the motivation for the development of this technology. A survey of current reaction rate and tempor al nuclide inventory techniques is then provided to offer

PAGE 13

13 justification for the depletion strategy applied within MCNPX. The MCNPX depletion strategy is then dissected and each code feature is detailed chronicling th e methodology development from the original linking of MONTEBURNS and MCNP to the most recent public release of the integrated capability (MCNPX 2.6.F). Calculation results of the OECD/ NEA Phase IB benchmark, H. B. Robinson benchmark and OECD / NEA Phase IVB are then provided. The acceptable results of these calculations offer suffic ient confidence in the pr edictive capability of the MCNPX depletion method. This capability sets up a significant foundation, in a well established and supported radiation transport co de, for further development of a Monte Carlolinked depletion methodology which is essential to the future deve lopment of advanced reactor technologies that exceed the limitations of current deterministic based methods.

PAGE 14

14 CHAPTER 1 INTRODUCTION AND BACKGROUND The purpose of the development the Monte Carlo linked depletion methodology within MCNPX is to provide a high fidelity depleti on methodology available within a widely accepted code package that is already the gold standa rd for radiation transport calculations. To understand why development of this methodology is vital to the current nuclear community and why the methodology must be implemented into MCNPX, involves fi rst understanding the history of reactor development fr om the first natural reactor in Oklo, Gabon, to the current major United States fuel cycle development programs, as well as why the current United States fuel cycle program, the Global Nuclear Energy Partne rship (GNEP), is interested in such a technology. Reactor Development History The first self-sustaining chain reaction achieved on earth predated the first man made selfsustaining chain reaction by roughly two billion years. The governi ng process to maintain this self-sustaining reaction required that precisely one neutron from each fission process resulted in another fission event. A delicate balance betwee n processes that resulted in increasing the neutron population, positive reac tivity, and processes that resu lted in decreasing the neutron population, negative reactivity, was re quired in order to maintain a self-sustaining reaction. In Oklo, Gabon, considerable heat occurred in se veral uranium rich deposits when groundwater seeped into the uranium deposits leading to am ple moderation that sufficiently increased the probability of fission to allow for a self-sustaining chain reaction.1 As the uranium rich deposits burned fissile uranium-235, fission products and other transmuted actinides were generated. In some nuclear reactions, fertile actinides cons umed radiation result ing in new fissionable actinides that along with uranium235 contributed to maintaining the sustainability of the chain

PAGE 15

15 reaction. Conversely, the production of fission products and non fiss ile transmuted actinides, as well as the reduction in moderator density from heat build, contribute d to decreasing the probability of fission, ultimately hindering th e sustainability of the chain reaction.2 Like a gigantic coffee pot percolating away over hundreds of thousands of years, the Oklo reactor would first undergo a fi ssion chain reaction until enough h eat and nonfissile nuclides were generated to hinder the self-sustaining ch ain reaction. The system would then cease to maintain self-sustaining fission events resulting in a cool off in the moderator leading to an increase in the moderator density. This subsequent moderator density increase would correspondingly lead to enough pos itive reactivity to ultimately restart the chain reaction. About 2 billion years later, scientists would later discover the fission process and mechanisms in order to fission a nucleus and ma intain a self-sustaini ng reaction. In 1930, W. Bothe and H. Becker found that a highly pene trating radiation was emitted when beryllium, boron or lithium were bombarded by alpha particles from a polonium source.3 In 1932, Irene Currie and her husband found that protons were produced when striking hydrogen containing substances, such as paraffin, with this newly discovered penetrating radiation. James Chadwick later demonstrated that the evidence from thes e experiments was compatible with the assumption that this new radiation consiste d of an uncharged particle havi ng approximately the same mass of a proton.3 He named this new particle neutron.3 In 1934, Enrico Fermi irradiated uranium w ith neutrons trying to produce the first transuranic element; however, he accidentally achieved the world's first nuclear fission. In 1938, he received the Nobel Prize in Physics, "for his demonstrations of the existence of new radioactive elements produced by neutron irradiation, and for his related discovery of nuclear reactions brought about by slow neutrons.3 In 1939, Hans and Strassm ann later showed that the

PAGE 16

16 products from neutron bombardment of uranium included elements in the medium mass region, much lighter than the initial uranium nuclide. The presence of these medium mass region nuclides from the bombardment process suggested that the nucleus had split during the reaction, and the fact that the sum of these medium mass nuclides did not add to the sum of the initial parent uranium nuclide suggested that some of the mass in the reaction was converted into energy. L. Meitner and O. Frisch termed this process fission and also calculated the energy released during fission of a U-235 nuclide to be 200 MeV.3 Albert Einstein then wrote his famous letter to President Franklin Roosevelt on August 2, 1939 stating "that the element uranium may be turn ed into a new and important source of energy in the immediate future." That source of energy was to be utilized by the United States to create the first atomic bomb. With theoretical models and simulation, Enrico Fermi later theorized that the neutrons emitted in fission might induce fiss ion reactions concludi ng that it should be possible to sustain a chain reaction in uranium a nd that this chain reactio n could be made steady state by implementing a balance of fissionable and nonfissionable material. Fermi and Leo Szilard further proposed placing uranium in a matr ix of graphite, where graphite was used to slow the fission neutrons to low enough energy in order to increase the probability of fission to maintain a self-sustaining chain reaction. On December 2, 1942, implementing a delicate balance of positive and negative reactivity, ba sed on proper modeling and simulation, the first controlled self-sustaining chain reaction was achieve d in a squash court under the University of Chicagos Stagg Field.3 The later implementation of this discovery was unleashed, on August 6, 1945, when Little Boy exploded over Hiroshim a killing over 100,000 people, and on August 9, 1945, when Fat Man exploded over Nagasaki k illing 75,000, ultimately ending the war with

PAGE 17

17 Japan. Both atomics bombs were designed us ing the Monte Carlo modeling and simulation method. In a later document in 1946, Enrico Fermi publis hed a scheme for outlining the future uses of nuclear energy. In this document he clearly stat ed that controlled self -sustained fission chain reaction energy may be utilized for power genera tion for the world by use of a steam cycle or direct energy conversion process.4 By enriching the amount of U-235 in natural uranium, the uranium could be combined with a water coolant and poison materials in order to make a selfsustaining power reactor system. On Decembe r 8, 1953, United States President Dwight D. Eisenhower gave the Atoms for Peace speech to th e United Nations and outlined the advantages of commercial power generation. He later task ed the United States Navy with developing the first commercial power stati on. A proposal by Duquesne Light Company was later accepted for the Shippingport Atomic Power Station, and th e first reactor went critical on December 2nd 1957.5 The Shippingport Atomic Power Station was a type of light water reactor (LWR) called a pressurized water reactor (PWR). Another type of competing LWR design was the boiling water reactor (BWR), which was developed by a co mbined effort between Argonne National Laboratory and General Electric. The GE Valecitos BWR was th e first commercial BWR to be licensed by the United States Atomic Energy Commission.6 The PWR concept is characterized as a system in which the coolan t is sub-cooled and contains homogeneously disperse d boron poison in orde r to maintain criticality. The system utilizes an indirect dual-cycle that uses a st eam generator to transfer heat from the exiting primary radioactive subcooled reactor coolan t to a secondary loop containing nonradioactive steam. This nonradioactive steam is used to spin a turbine that when combined with a generator creates power for hundreds of thousands of homes.7

PAGE 18

18 The BWR concept is characterized as a system that has boiling in the reactor core, with the bulk coolant containing no bor on, utilizing a direct cycle for power conversion (the demonstration BWR/1 plants utilized a dual cycle) Instead of a seconda ry loop, this type of system exhausts high quality steam from the t op of the reactor through the use of moisture separators and dryers. This radioactive steam is fed directly into the turbines that are also connected to generators used to make power for hundreds of thousands of homes.7 The progress of these commercial power system s later led to the development of other competing reactor systems. Due to the advantages in being able to mine large amounts of uranium ore, and due to the thought that a countr y developing enrichment technology may also be developing weapons technology, the Canadians d ecided to develop a sy stem that would not rely on enrichment technology. By taking advantage of the scat tering and absorption properties of deuterium, the Canadian deut erium uranium reactors (CANDU) were able to generate power with natural U-235 concentrations Other technologies were al so spawned employing a variety of coolants and moderators such as the gas cooled fast reacto r (GCFR) and graphite moderated reactors in order to achieve higher operating temperatures and enhanced thermodynamic efficiency. Technologies were also developed to support th e breeding of fissile nuclides incorporating various operational characteristics such as the li quid metal fast breeder reactors (LMFBR).8 Though all the vast reactor types incorporate many different materials, fuel and coolant forms, geometry combinations and operating stra tegies, the basic design objectives are still very much the same. The goal is to determine the correct balance of fissile and control material needed in order maintain the self-sustaining ch ain reaction, operate the sy stem at a prescribed power, for a prescribed amount of time, achie ving some type of operating objective while

PAGE 19

19 accounting for reactivity deficits. These reactivity deficits in clude temperature and isotope buildup effects. The operating objectives range from maximizing capacity factor, to minimizing discharged wasted, to produci ng irradiation environments, to generating medical isotope material.9, 10, 11, 12 In order to develop the most optimized and advanced systems for specific operating strategies, designers constantly exam ine varied types of material and geometry combinations in order to design a system that meets the operating goals at the minimum cost. This process therefore involves simulating a model in order to investigate many varied types of designs, with many design iterations and pertur bations, until an optim um configuration is reached.13 As the model becomes more detailed and gives better unders tanding of what is actually happening, the designer is able to ma ke a more well-informed decision on how to proceed with designing the system. Furthermor e, as the development decisions for nuclear reactor extend beyond running the system at a pres cribed power for a pres cribed amount time to minimize the nuclear waste and addressing prolif eration concerns, a higher fidelity modeling methodology is required in order to make more-informed design decisions.14, 15 Advanced Fuel Cycle Initiative The Advanced Fuel Cycle Initiative (AFC I) was established October 1, 2002, as an evolution of the Advanced Accelerator Applic ations (AAA) program. The AAA program was a Department of Energy (DOE) program commi ssioned by Congress in FY2000 to construct a national effort by DOE laboratorie s, industry, and univers ities in order to a ddress key significant nuclear issues facing the United States, particul arly the advanced fuel cycle technologies, actinide transmutation, and treatm ent of spent nuclear fuel. The mission of the program was to develop a technology basis for the transmutation of nuclear waste and demonstrate the practicality of the process for use in long-term nuclear waste management.16 Initially, the focus was on the use of accelerator based systems, hence the program name AAA. In 2002, the

PAGE 20

20 program was evolved into AFCI with the fo cus changing to chemical separations and transmutation fuel technologies. The idea was to couple reactor use of fuel with transmutation and separation technologies in order to improve the management of nuclear waste through fuel recycle and transmutation.16 The mission of the AFCI program was to de velop advanced fuel cycle technologies including spent fuel treatment, advanced fuels de velopment, and transmutation technologies for application to current operating commercial nucl ear reactors and next-generation reactors with the goal of (1) reducing volume of high-leve l waste; (2) reducing long-lived and highly radiotoxic elements; (3) reclaiming valuable energy content in spent nuclear fuel. AFCI was part of an integrated strategy of the DOE office of Nuclear Energy, Science and Technology, which also includes the Nuclear Power 2010 Program, the Generation IV Nuclear Energy Systems Initiative (Generation IV), and the Nuclear Hydrogen Initiative. AFCI was important in reestablishing the viab ility of civilian nuclear capabil ities and facilities in the United States by utilizing services of DOE laboratories, industry, and unive rsities. The development of an advanced high fidelity technology for simulati ng transmutation and depletion of nuclear fuel fit nicely with the goals of AFCI, thus leadi ng to the initial funding of the MCNPX Monte Carlo linked depletion project.17 However, in FY2007, AFCI al ong with the Nuclear Power 2010 Program, the Generation IV Nuclear Energy System s Initiative (Generation IV), and the Nuclear Hydrogen Initiative evolved in to the GNEP program with th e objective of developing an Advanced Burner Reactor (ABR) technology in or der to transmutate spent nuclear fuel and develop separations technologies in order to maximize the am ount of energy extracted from nuclear fuel, minimize nuclear waste, and mitigat e concerns regarding proliferation of weapons material from commercial nuclear fuel.18

PAGE 21

21 Global Nuclear Energy Partnership The main goal of GNEP is to advance nuclear power technologies a nd enhance the nations nuclear energy capacity while effectively addressi ng the challenge of nuclear waste disposal and decreasing the risk of nuclea r weapons proliferation and.18 As part of President George W. Bush's Advanced Energy Initiative, GNEP s eeks to develop worldwide consensus on enabling expanded use of nuclear power to meet the growing energy demand while ensuring reduced dependence on carbon based energy producing systems.18 The GNEP campaign recognizes that a plentiful, reliable supply of energy is param ount to sustained economic growth and prosperity, and that nuclear power is the only proven tec hnology that can provide abundant supplies of electricity reliably and without air pollution or emissions of greenhouse gasses. The GNEP campaign has further recognized that the use of this economical, carbon-free technology will reduce the nations dependence on environment unfriendly energy produci ng technologies that increase greenhouse gas emissions.18 GNEP addresses two key issues: implementi ng sensitive nuclear technologies in a way that protects global security and determining proper methods of disposing of nuclear waste safely. The proposed new nuclear energy infrastr ucture would implement a closed nuclear fuel cycle that enhances energy secu rity, while promoting non-prolif eration and recycling nuclear fuel to reduce the nations nuclear waste. This closed fuel cycle requires the development of technologies that enable consump tion of long-lived radioactive wa ste, and recycling of useful fissile isotopes.9, 19 The following include the major obj ectives of the GNEP program: 1. New proliferation-resistant technologies to recycle spent nuclear fuel will be used so that more energy may be recovere d and total nuclear waste is reduced. 2. The latest and advanced fuel cycle strate gies will be employed in order to reduce the risk of nuclear proliferation worldwide.

PAGE 22

22 3. The growth of prosperity and sustainable development around the world will be encouraged by meeting energy capacity de mands through the use of nuclear energy. 4. The nations dependence on fossil fuels will be reduced encouraging the continual improvement of the environment. These objectives will be achieve d by the following strategies: 1. A new generation of nuclear power plants will be developed in the United States that employ advanced fuel burning strategi es in order to minimize actinide wastes. 2. An integrated fuel separations technology and recycling capabili ty will be designed and deployed. 3. An aggressive strategy to manage spent nuclear fuel and nuclear waste in the United States will be determined including permanent geologic storage at Yucca Mountain. 4. A reliable fuel services program will be engineered in which participating nations with secure nuclear technologi es would be required to supply and receive fresh and spent nuclear fuel. 5. Cost effective efficient power reactors will be developed and deployed. 6. Nuclear safeguards to enhance the prolifer ation-resistance and safety of expanded nuclear power will be improved. In order to achieve the GNEP objectives by implementing the mentioned strategies, a symbiotic fuel cycle strategy must be implemente d that utilizes both LWR and fast reactor (FR) technology to maximize the amount of energy extrac tion from transuranics while minimizing the amount of overall waste buildup. Sp ent nuclear fuel contains a plet hora new actinides due to the transmutation that occurs during the burning of the fuel. Thes e actinides are sources of both long-term radiological hazards, because they domi nate the long term ingestion radiotoxicity of the waste, and proliferat ion hazards, because they are fissionable nuclides.19 The long lived actinides tend to have much larg er half-lives than typical fissi on products. Theref ore the task of actinide burning is of utmost importance to addr essing the long term radi ological waste problem and mitigating proliferation concern. To burn actinides effectively, the actinides must be

PAGE 23

23 separated from other isotopes that compete for reactions.20, 21, 22 The GNEP campaign is tasked with developing a separations facility to exact burnable actinides from spent nuclear fuel and separate these actinides from short-lived fissi on products. The short-li ved fission products may be easily stored in a geological repository as th ese isotopes will decay away to stability in a reasonable amount of time. Advanced Burner Reactor Campaign The separated actinides must then be recycl ed and further incine rated in either an accelerator or reactor. Many types of scenarios have been studied that examine this portioned approach to actinide incineration; however, the transmutation of waste by accelerator option is limited by cost and wasted use of extractable actin ide energy. Two basic types of reactors have been considered for further actin ide burning. The extract ed actinides could be fabricated into mixed oxide (MOX) fuel bundles and placed back into a commercial LWR. The advantage to burning in a commercial thermal spectrum reactor is that commercial reactors are already in operation, and therefore transitioni ng operating strategies for accep ting MOX fuel would seem to be more viable solution th an developing an entire ne w reactor for transmutation.12, 15, 19, 21 Recycling of plutonium in thermal reactors would al so help to reduce total plutonium and seek to mitigate proliferation concern. Developing operating strategies to account for reduced poison worth, from the higher thermal absorption cross s ection of Pu-239 as compared to the uranium isotopes, and faster reactor re sponse, from the reduced delayed neutron fraction of Pu-239, may be complicated; however, the solutions to these st rategies may be engineered and therefore does not limit deployment.11 The limit of the effectiveness of multi-recycling of actinides in commercial LWRs is dictated by the amount of higher actinides gene rated as a function of burnup. In a thermal reactor spectrum, the higher actin ides buildup due to lower fission rates (i.e. curium) in those

PAGE 24

24 isotopes and higher capture rates in the thermal sp ectrum. The fission-to-capture ratio in thermal reactors is not as high as in fast reactors and therefore higher actinides are generated and not fissioned as readily as compared to fast reactors Therefore more Am, Cm, and Cf are generated as function of burnup in thermal reacto rs as compared to fast reactors.19, 22 These isotopes make spent fuel recycle challenging through increa sed heat generation and spontaneous neutron emission, and further result in both criticality and radiation protection difficulties. The fast reactor spectrum is more favorable for higher actinide transmutation because the fission to capture ratio is much higher as compared to thermal spectrum reactors.19 Destruction of long-lived actinides requires ex cess neutrons. In a fast reacto r the difference between neutron production by fission and losses by capture in struct ural and absorbing materials or leakage from the core is ~0.4 -0.6 neutrons per fission while in a thermal reactor this su rplus is well below at 0.1 neutrons per fission.19 The cross sections in the fast spectrum favor fission rather than capture; however, the fast spectrum cross sec tions are smaller in magnitude and therefore a larger flux is required to ach ieve the same destruction yield as in thermal reactors.15 Since the principal of transmutation consis ts of recycling actinide wastes in reactor scenarios that favor fission as opposed to capture, de velopment of a fast reactor technology is integral to the objectives of the GNEP campaign in order to maximize energy extraction from fuel and minimize the buildup of long term actinide wastes. GNEP will develop and demonstrate Advanced Burner Reactors (ABRs) that consume actinide wastes while extracting their usable en ergy. The objective of developing these fast spectrum reactors will be to destroy actinide wast es in commercial spent nuclear fuel from nuclear power plants, avoiding th e need to accommodate material in a geological repository for hundreds of thousands of years while it decays. Th e reality of this objectiv e is that these ABRs

PAGE 25

25 will not avoid but reduce the burden of current repository demand. Significant prior United States investment in fast reacto rs, such as Fast Flux Test Faci lity and the Experimental Breeder Reactor, provides a valuable technology base for supporting future development of ABRs. Though the development of ABRs requires follow ing the isotopes in exact detail in order understand true transmutation rates of actinide wastes, the initial desi gn process will require thousands of core calculations th at integrate neutronic, structur al, and thermal hydraulic concerns in order to license the facility. These calcula tions include varied core operating and shutdown strategies in order to demonstr ate that the core may operate at a certain power, for a certain amount of time, and shutdown safely in an acciden t scenario. Therefore ca lculations of actinide transmutation rates need not be fo llowed in tremendous detail in orde r to license the facility so as to hamper the ability to complete the plethora of integrated system calculations. In fact, calculations that capture more deta il than what is necessary to license the facility will simply hinder the completion of the design and possibly the building of the facility. If a 2-dimensional diffusion theory code that runs in 5 seconds ach ieves an adequate result, why waste time running a high fidelity calculation that take s a week (the fallibility in this statement may reside in the fact that without experimental benc hmarks to validate the less accurate solution method, one cannot be certain that the approximations applied to that solution method are indeed correct). Therefore high fidelity depletion modeling may not be of in terest to the initial design of the ABRs. However, high fidelity depletion modeling is inte gral to determining adequate source term generation for the development of technologies that will be used to segregate detectable signals into key nuclide signatures to be used fo r interpreting proliferation concerns. Reactor Safeguards Campaign The objective of reactor safeguards is to develop proliferant re sistant fuel cycle technologies and programs the help prevent misuse of civilian nuclear f acilities for non-peaceful

PAGE 26

26 purposes. Reactor safeguards responsibilities include accounting for special nuclear materials, control of technology and the ab ility to inspect and verify compliance with international agreements.23 Under the auspices of the Internationa l Atomic Energy Agency reactor safeguards have been an effective deterrent against the spre ad of nuclear technology and materials. The role of GNEP safeguards is to provide the opportunity to design modern safeguards directly into the planning and building of new nuclear ener gy systems and fuel cycle facilities. .23 The GNEP Safeguards campaign will also focused on improving international safeguards including the development of future proliferation-resistant fuel cycle and reactor technologies to be integrated world wide.23 The development of safeguards technologi es involves improving the way in which diversion of special nuclear materi al is detected. The source term of the special nuclear material is the actinide wastes generated in nuclear re actors and the soon to be developed ABRs. Therefore the safeguards community is concerne d with measuring the content of key specific nuclides within spent fuel leaving the reacto r. For the GNEP campaign, the spent nuclear material will take on many differe nt types of combinations depe nding on the reactor type that generated that specific spent fuel. The isotope content of key specific nuclides within the spent fuel may be determined by implementing various non-destructive-analysis (NDA) techniques involving the discrimination of a detectable signal into the components of a specific nuclide s radiation signature.24 The efficiency of the detection equipment to meas ure the concentration of a specific nuclide is therefore dependent on the ability to segregate the detectable signal into the components that contribute to the specific nuc lides signature and the com ponents that do not contribute.24 Since the detectable signal may be a conglomeration of many simultaneously contributing radiation

PAGE 27

27 signatures, discrimination of the signal into the key components of a specific nuclides signature involves knowledge of all the cont ributions to the signa l. From a modeling standpoint, designing more efficient detection equipment involves expl icitly following the evolutionary buildup of all isotopes contributing to the detect able signatures. However, in order to accurately account for the temporal evolution of the nucli de inventory of all contributors to the detectable signal of key nuclide signatures, a method that more accurately captures the true system physics of the source term generation becomes paramount. MCNPX is widely used in the safeguards community for modeling all aspects of reactor safeguards and is considered the gold standard in radiation transport modeling. Historically, the MCNPX code lacked the capability to m odel fuel depletion a nd therefore could not accurately generate the needed source term for development high fidelity detection equipment. In the past, this source term was either gene rated by an externally linked technology or by an entirely separate code of which both incorporat ed varied assumption that ultimately affected calculation accuracy. A technology that calcul ates a high fidelity time-dependant isotope inventory is essential for determ ining a sufficient source term us e in a very pr ecise calibration and design of high fidelity radiation detection equipmen t. It is vital that th is capability be placed in MCNPX in order to make MCNPX, a tool al ready implemented by the reactor safeguards and considered a gold standard by the community, a more complete tool for reactor safeguards calculations. Motivations for a Monte Carlo Depletion Tool in MCNPX Many deterministic linked tools exist that ar e capable of modeling fuel depletion for advanced reactor concepts. However, these tools employ a variety of approximations about system physics in order to achieve a quick result.25-34 For the result to be reliable, costly experimental benchmarking and thorough investigativ e analysis is required in order to determine

PAGE 28

28 the limits of the modeling technology. Approxima tions implemented in the calculation of one type of system are not necessarily valid for the calculation of a different system.32, 34 This issue leads to the necessity to devel op and catalogue an assortment of tools employing a diverse range of calculation assumptions.35 The selection of an adequate m odeling tool therefore relies on the ability to predetermine which assumptions are ade quate for a particular calculation type and then selecting the design tool employi ng those assumptions. Therefore, it is possible that false conclusions of system behavior may be draw n from false hypotheses of system physics when incorrect approximations are implemented leadi ng to the inability to make adequate design decisions. As advanced reactor concepts challenge the accuracy of current modeling technologies, a higher fidelity depleti on calculation, which employs few, if any, assumptions of system physics, is crucial to properly model the wide variety of advanced reactor concepts. An optimum modeling tool would not rely upon predetermined knowledge of adequate assumptions that could result in an incorrect calculat ion; the tool would eliminate sy stem physics assumptions entirely and preserve the true physics behavior. A Monte Ca rlo linked depletion to ol is capable of achieving this goal because true system physics is preserved through the simulation of the actual complete physical process. This makes the Mo nte Carlo linked depletio n method highly robust and capable of analyzing a wide variety of cal culations with immense confidence of achieving an accurate result.36 A few Monte-Carlo-linked depletion c odes exist including ALEPH, BURNCALC, MCODE, MC-REBUS, MCWO, MCB, MCMG-B URN, MOCUP, MONTEBURNS, MVP and RACER.37-49 Most of these codes are externally linked technologies and implement complicated directory structures to integrate each externally linked code package. This results in increased

PAGE 29

29 memory restrictions that ultimately constrict th e maximum size of the calculation model. Some of these efforts are also not supported, and theref ore offer little if any code support for the typical user. Some of these efforts use multi-group Monte Carlo and are th erefore slave to the calculation assumption of the deterministic methodologies.49 Most of these codes are also focused on accelerating the calculation by minimi zing the amount of isotopes and reactions tracked in order to minimize computational expense at the cost of detail in the simulation thereby loosing fidelity that may be of interest to the user. Since most of these codes are externally linked technologies, in most cas es, the codes only work with a minimal amount of functionality of the codes in which they are linked. In order to truly take advantage of the robustness of the Monte Carlo linked depletion method, the technol ogy must be self-contained, eliminate excess top-level processing and calculati on size requirements as well as automatically determine and track parameters important to the depletion process. MCNPX depletion provides a vital modeling framework, within a well established, supported reactor safeguards radiatio n transport code, for high-fideli ty depletion calculations to assist in the development of technologies requ iring ample intricate de tail in the depletion solution such as the reactor safeguards calculation s required for GNEP. Because this capability is fully integrated into the functionality of a we ll established radiation transport package already utilized by the GNEP safeguards committee, the comm unity now has a tool that meets almost all the functionality required to m eet their current modeling needs. Objective The goal of this work is to develop a se lf contained easy-to-use Monte Carlo linked depletion modeling framework within a well es tablished radiation tr ansport package that eliminates calculation size requirements as well as automatically determines and explicitly tracks all the important parameters for computing an ac curate depletion solutio n. This task involves

PAGE 30

30 internally linking the MCNPX st eady state-reaction calculator to the CINDER90 number density calculator.50, 51 MCNPX will be used to calculate st eady-state reaction rates and normalization parameters while CINDER90 will be used to calculate the time-dependent isotope buildup/depletion. The work scope may be bifurcated into two separate processes: code development and code verification and validation. The following te n code development issues will be addressed: 1. A Monte Carlo linked depletion capability that works within the confines of the MCNPX code package will be developed, making sure th e depletion capability is compatible with major execution features of MCNPX. 2. The input structure must reside within an MCNPX input deck. The input structure will be simplified yet include ways of user specification for a. time steps to report isotope and system data; b. power lever and percen t of full power level to implement at specific time steps; c. burn materials to be depleted; d. isotopes to omit; e. fission products to track; f. nuclide concentrations to manually change for different time steps; g. ability to order output accord ing to user specification. 3. The output will contain averaged ti me-dependent information for a. neutron multiplication; b. neutrons per fission; c. recoverable energy per fission; d. burnup; e. source neutrons; f. isotope concentrations and activ ity of radioactive isotopes; g. sum of isotope concentrati ons and activity information. 4. For individual burn materials the code output will report a. fission power fractions; b. burnup; c. collision rates; d. isotope buildup and activity. 5. The following mechanisms for isotope tracking will include

PAGE 31

31 a. tracking each material specified by the user; b. selecting fission products to be tracked based on easy-to-implement predefined fission product Tier sets; c. following the concentrations of all the possible daughter reactions from isotopes specified as burn materials; d. tracking reaction rates of metastable isotopes; e. manually adjusting burn material concentr ations of specific isotopes if necessary. 6. Reaction rates to be computed for use in the depletion process will be determined, and a method will be developed for normalizing the reaction rates based on given and calculated system parameters. 7. A methodology will be implemented for addre ssing the nonlinearity in the reaction rate utilization in the depletion equation. 8. A method will be engineered for selecting the pr oper fission yield to be utilized within a specific computation. 9. A capability will be invoked to burn multip le burn materials where individual burn material characteristics may be calculated and reported. 10. A technique will be implemented for accura tely capturing the average recoverable capture gamma energy contribution to the true energy per fission event. Once the methodology is developed the following benchmark studies will be completed: 11. OECD/NEA Burnup Credit Calculationa l Criticality Benchmark Phase I-B;52 12. H. B. Robinson Infinitely Reflected Pr essurized Water Reactor Fuel Assembly Calculation;53 13. OECD/NEA Burnup Credit Calculational Criticality Benchmark Phase IV-B.54

PAGE 32

32 CHAPTER 2 SURVEY OF STATE OF THE ART A PPROACHES TO DEPLETION ANALYSIS During the operation of a nuclear system, the nuclide concentration wi ll change as isotopes consume radiation (neutrons, protons, alphas, beta s, etc.) and undergo vari ous nuclear reactions [(n, fission), (n, 2n), (n,p), (n, ) (n, ), etc.].2 The temporal change in isotope concentrations results in further changes in scatter, capture, a nd fission events as isotope s either transmutate or fission into new nuclides possessing different probabilities for these reactions. For example, as U-235 depletes as a result of the fission proce ss, fission products are generated. Because these newly created fission products posses capture reac tion probabilities that rival U-235 for neutron capture, the fission products, combined with the depletion of U-235, reduce the propensity of the system to fission and maintain a steady-state ch ain reaction. In order to maintain a self sustaining steady-state chain reaction, more fuel than is necessary in order to maintain a steadystate chain reaction must be lo aded. The introduction of this excess fuel increases the net multiplication capability of the system. To mainta in the net neutron multiplication of the system at a steady state, poison material is introduced to absorb the excess neutrons generated from the excess fuel loading. As the fuel depletes and produces absorbing fission products, the controlling material is extracted in order to ma intain the delicate reactivity balance needed in order to maintain a self sustaining chain reaction.9 Therefore it is crucial to monitor the temporal isotopic concentration in the nuc lear system because changes in this composition affect the operating strategies that are necessary to main tain a steady-state self -sustaining reaction. The operating strategies of a nuclear system are affected by the ability to meet limiting values of key operational parameters. These parameters include: flux, power distribution, reactivity, shutdown margin, etc. The limiting values of these parameters are set such that the system be able to operate safely for a prescrib ed time and meet a prescribed objective. Since

PAGE 33

33 these system parameters therefore limit operational characteristics of the system, it is crucial to compute these values as each value temporally e volves drastically enough to warrant changes in the system operation. The study of the change of these operational para meters with the timedependent production/ depletion of nucle i is known as depletion analysis.2 The Depletion Equation Solving for the time-dependent change of a specified nuclide involves accounting for all processes that lead to either the creation or destruction of th at nuclide. The temporal nuclide density may be described by the Bateman equations.55 Equation 2-1 states a simplification of a form of the Bateman equations so as to desc ribe the depletion of an individual nuclide: m k m k k m m m mt r N Y t r N dt t r dN ) ( ) ( ) ( (2-1) dt dNm= time dependant change in nuclide m m mt N ) (= destruction of nuclide m m k m k kt N) (= creation of nuclide m via other nuclides in the system mY= production of nuclide m via an external source Solving for the time-dependent change of a specified nuclide also involves knowledge of the time-dependent change of each contributing nuc lide. Therefore the solution of each nuclides time-dependent change involves solving a set of coupled differential equations.55 Though solving a set of coupled linear differential equati ons is a straightforward mathematical process, the coefficients in the coupled depletion equa tions are nonlinear therefore making the equations unsolvable without approximation. E xpanding the destruction coefficient,m and the creation coefficient,m k clarifies the nature of the nonlinearity of equation 2-1: r r m m mdE t E r E ) ( ) (, (2-2a)

PAGE 34

34 k mr r k r km k m k km m kdE t E r E E Y L ) ( ) ( ) (, (2-2b) m destruction of nuclide m by radioactive decay r r mdE t E r E ) ( ) (, = destruction of nuclide m by transmutation reaction k m k kmL= creation of nuclide m by some isotope radioactively decaying to isotope m k mr r k r kmdE t E r E E Y ) ( ) ( ) (, = creation of isotope m by some isotope transmutating to isotope m via a transmutation reaction, r. The analysis of the temporal nuclide inventory is complicated by the fact that the temporal nuclide inventory solution is depe ndent upon the time-dependent fl ux. Unfortunately, the timedependent flux is also dependent upon the time-dep endent nuclide inventory. To make equation 2-1 linear and solvable, approximations must be made regarding the coefficients. The first approximation assumes that the coefficients be made constant over a time st ep of interest, tacitly implying that the time-dependent nuc lide solution is separable from the spatial transmuation rate calculation. The second approximation assumes that spatial transmutation rates may be calculated using discrete time step s. The transmutation rates fo r the temporal nuclide inventory calculation are first computed implementing a st eady state reaction rate solver. Then those transmutation rates are assumed constant a nd applied to the tem poral nuclide inventory calculation in equation 2-1. The validity of the approximations is contingent on the fact that the magnitude of the transmutation rate implies that large time dura tions are needed in order to initiate significant buildup and depletion of nuclides so as to appr eciably alter th e spatial transmutation rate calculation.2, 55 The definition of large time dura tions is relative a nd dependent upon the builup of highly absorbing isotopes during a time duration that lead to a significant change in the transmutation rate. For example, at the ini tiation of reactor start up Xe-135 and Sm-149 both

PAGE 35

35 contain large absorption cross se ctions and builup to equilibrium values early in the reactor operating cycle. In reactors c ontaining burnable poisons incorporat ed into the fuel assembly, the burnout of these highly absorbing isotopes during the fuel cycle also offers a significant change in transmutation rate as the fuel becomes more reactive in the absence of poisons. Taking time steps that exceed the time to equi librium buildup of these isotopes a ffects the further trajectory of the depletion solution because the equilibrium buildup of these highly absorbing isotopes does greatly affect the calculation of spatial transmutation rates. This affects the accuracy of the number densities generated from equation 2-1 for that time step which further affects the calculation of transmutation rates for the next time step. As long as the transmutation rates do not significantly change during a time-step, assuming a constant transmutation rate for a given time-step is valid.56 Zero-Dimension Depletion Calculation The depletion equation requires knowledge of re action rates, decay coefficients and isotope concentrations at a specific time-step in order to solve for the isotope concentrations at the subsequent time-steps. The accuracy in the de pletion calculation is highly dependent upon the appropriateness of the coefficients implemented at given time steps in the solution method, and therefore the amount of spatial detail followe d in the depletion solution greatly affects calculation accuracy. The deple tion equations could be solved on a normalized basis, in which power is specified in a per weight basis and is otopes concentrations are determined on a per weight basis, if the collision rates utilized in equation 2-1 truly emulate the actual collision rates for a given system. Since this type of calculatio n is solved over a normalized basis, the solution of the coupled depletion equation do es not require the coefficients to exhibit spatial dependence. The accuracy and applicability of this solution me thod resides in the validity of the implemented coefficients for a given analyzed system.57

PAGE 36

36 The ORIGEN and CINDER codes, by themselves are a family of zero spatial dimension depletion codes and therefore the codes have no knowledge of the spat ial dependence of the transmutation rates.51, 58 Dimensions of the analyzed syst em are approximated by specifying the power in terms of a bases unit (i.e. power/wei ght). The major assumption that ORIGEN and CINDER make is that the flux is not spatially dependent and may be calculated by equation 2-3: 51, 58 i i f i f iR X P1810 242 6 (2-3) P = System power MW f iX = Concentration of the ith fissile nuclide atom g f i = Microscopic probability of fission of the ith fissile nuclideb iR = Recoverable energy per fission event of the ith fissile nuclide fission MeV = Total flux s cm neutrons2 The calculation of power based on some spatia l flux solution is desc ribed by equation 2-4: V fdV Q r Power ) ( (2-4) ) ( r = Spatial flux distribution s cm neutrons2 f = Macroscopic probability of fission cm 1 dV = Differential volume element 3cm Q = Average recoverable energy per fission for the whole system fission MeV After determination of flux and system power, the solution methods of the two codes differ in approach. The two most widely used met hods for solving the coupled depletion equations involve a simplification of eith er the matrix exponential met hod or the Markov linear chain

PAGE 37

37 method. The Markov linear chai n method involves uncoupling the coupled depletion equations into a set of partial linear transmutation paths, solving for the partial buildup/depletion of each nuclide within each linear chain, and summing the final buildups of each partial path to determine the final nuclide buildup (CI NDER90 method previously explained).51 This method takes advantage of minimal memory requirement s because the computation is broken up into smaller memory manageable sets. The matrix exponential method tries to explicitly solve the depletion equations by casting th e solution in terms of a matr ix exponential (method used in ORIGEN and other industry based codes that evolved fr om the ORIGEN methodology).58, 59 The main objective of both techniques is to try and determine, as accurately as possible, the evolution of the temporal nuc lide inventory solution. Matrix Exponential Method The underlying concept of the matrix exponentia l method is that the formulas for solving normal systems of equations with constant coeffici ents are identical to th e formulas for solving first order differential equations with constant coefficients.60 Equation 2-1 may be recast into the following form k k mk mt r n t r B t r n dt d ) ( ) ( ) ( (2-5) where: k m for t r k m for t r Y L t r Br r m m r r k r km k km mk) ( ) ( ) (, (2-6) This results in a set of firs t order homogenous ordinary differe ntial equations with constant coefficients. The solution of this first order differential equation is:60

PAGE 38

38 ) exp( ) ( ) (0Bt t N t N (2-7) where N is a column vector of nuclide densities and B is the matrix of elements calculated of equation 2-6. The function ) exp( Bt is a matrix exponential function, and is therefore calculated from the following expansion: 0! ) ( ) exp(m mm Bt Bt (2-8) If the size of the matrix is not too large and not too sparse, the main adva ntage of this method is that the coupled depletion equatio ns may be solved explicitly.59 However, the depletion equations require knowledge of all transmuted ac tinides and fission product s leading to a large sparse matrix. This large sparse matrix requir es an enormous amount memory in order to store all the necessary computations. Also, the sparse matrix contains both very large values and very small eigenvalues depending upon the half-lives of each isotope tracked during the depletion process.59 Certain isotopes possess half-lives of a fraction of a second while others posses halflives of a billion years. The issue of having a large sparse matrix w ith widely separated eigenvalues leads to problems in the ability to compute the solution and therefore approximations are made in order to generate a stable solvable matrix.60 Potentially, the size of the matrix can be as large as n2 where n is the amount of nuclides followed in the set of equati ons (1700 nuclides for ORIGEN2.2). However, because the matrix is sp arse, a recursive relationship may be generated for formulating the matrix exponential function thereby eliminating the massive memory cost associated with storing all the matrix terms of the matrix exponential function. The relationship in ORIGEN2.2 for one nuclide is as follows: o n n i ic t x ) ( (2-9)

PAGE 39

39 where n ic is generated by use of the fo llowing recursive relationship ) 0 (0 i ix c (2-10a) N j n j ij n ic a n t c1 11 (2-10b) In this solution method ija is an element in the matrix that is the first-order rate constant for the creation of isotope i from isot ope j. This algorithm for so lving for the temporal nuclide inventory only requires the storage of a single vector,n ic greatly reducing memory cost. The main concern computing the summation in equation 2-9 is that it is necessary to ensure precision is not lost as a result of addition and subtraction of nearly equal large numbers. The time step for the calculation may be reduced so as to limit the size of the matrix coefficients; then the resulting matrix may be multiplied by a factor in order to obtain the desired time step.60 Scaling the matrix calculation w ould be impractical for a large number of nuclides as many of the nuclides contain very short half-lives. Ther efore in order to fix the mentioned numerical issue, ORIGEN2.2 decays analytic ally all short lived nuclides. A short lived nuclide is defined such that the matrix coefficients for this nuclide resulting from the larg e amount of decay, are so large as to result in numerical i ssues such that the norm of the matrix exceeds the predetermined value of the word length of the computer used fo r the calculations. This simplification increases the stability of the solution method by eliminating numerical issues. This assumption is, however, only acceptable as long as the isotope does not contain a significant long lived precursor with a large interaction cross section.60 For short lived daught er isotopes from long lived parent isotopes, the Gauss-Si edel iterative technique is used to accelerate th e calculation of these nuclides.

PAGE 40

40 Ultimately, the amounts of isotopes that are tr acked explicitly are limited by the maximum memory requirements of the computer utilized to analyze the system. Therefore implementation of the matrix exponential method actually results in only applying a subset of nuclides to the explicit matrix exponential computation, and then employing approximations to follow the rest. The user must then be careful to select a subset of nuclides that best achieves the most correct characterization of the tr ue interaction behavior of the analyz ed system, and then the user must implement an adequate approximation to follow the rest of the nuclides However, further numerical limitation still exists as a result of the amount availa ble precesion resulting from the applied numerical technique. If the user wishes to therefore try to solve the Bateman equations explicitly, the matrix exponentia l method may be the method of choice; however, if the user determines that a complete set of fission products and actinides is required in order to achieve a reliable answer then the lin ear Markov chain method is mo re capable of storing the concentrations of more nuc lide sets without the same type of numerical cost. The CINDER90 Linear Markov Chain Method The CINDER90 code is a FORTRA N program coupled with an ex tensive data library used to calculated nuclide inventory.61 The origin of the CINDER90 c ode may be traced back to the original development of the CINDER code in 1960 at the Bettis Atomic Power Laboratory (BAPL) in support of thermal reactor simulatio ns. CINDER was utilized by both BAPL and Los Alamos National Lab (LANL) for the desi gn of many reactor-fuel investigations.61 Many later versions of the CINDER code were developed with improvements in data evaluations, fission yield sets, and enhanced decay information. In itially, CINDER required preset formation of a consistent set of linear chains to describe the creation and transmut ation paths of all investigated cases.61 All versions of CINDER uti lized decay and energy integrat ed reaction rate probabilities along with fission yield information to calculate the temporal nuclide buildup and depletion.

PAGE 41

41 Versions of CINDER preceding CINDER90 were lim ited to reactor fuel calculations involving fission products (66
PAGE 42

42 path. The solutions of each linear chain determines a partial nuclide density, i Each calculated partial nuclide density, i computed from a linear chai n, is then summed to obtain the total nuclide inve ntory of nuclide, mN The differential equati on governing the computation of ) ( t is therefore only coupled to any preceding elements in the sequence leading to the ith element. For the preceding, (i-1) element, a ll parameters are assumed known. The entire computation is then computed from the simplified equation i i i i i it t Y dt d ) ( ) (1 1 (2-12) where quantities are now indexed by the order in which they app ear within a given sequence and 1 i is the transmutation probabil ity of forming nuclide element i The general solution form of a linear sequence of nuclides coupled by any sequence of absorption or decay, first derived and implemented by the CINDER code, was: n m n m k n m j n j m i j i m n m j n j m i j i n m l l m k njt jte e Y t1 1 0 ,1 ) ( (2-13) This algorithm therefore depe nded upon predetermined knowledge of the transmutation path in order to properly include every 0mencountered within a specific path. This method further suffered from large computational cost because th e entire transmutation path was to be followed even if the probability was low of incurring fu rther nuclide production of a specific isotope within a chain.51 For computational speed enhancement, si mplification, and reduction of memory requirements, the CINDER90 code employs a met hodology that does not requi re a predefined set of nuclides employed in a transmutation path. The methodology starts at a base set of nuclides

PAGE 43

43 and follows each path independently. Tests of si gnificance at each step of the transmutation path are set. If the test of significance is failed, then the transmutation path is stopped and further computation of extraneous nuclides is not incurred. This lead to the following simplifications 1 00 mm (2-14) 1 0 m Ym (2-15) ultimately resulting in the following so lution for a given linear sequence:61 1 1 1 1 0 1 1 1 11 ) (n k n j n j i j i n j n j i j i n l l m k njt jte e Y t (2-16) The test for significance in CINDER90 i nvolves calculating the passby quantity, ) (t Pn. The passby of a nuclide is the time-integrated tr ansmutation of that nuclide for a specified time interval: dt t t Pn t n n 0) ( ) ( (2-17) The passby therefore may be thought of as the pr ogeny of nuclide n resulting from transmutation or decay.61 The accuracy of CINDER90 is dependent upon the da ta that is used in the solution of (Eq. 2-11). If the code is used by itself to calculate nuclide inventory, the user would be relying heavily upon the inherent multi-group cross sections, utilized to determine collapsed energy integrated reaction rates, in order to achieve a solution. Since CINDER90 is a zero-dimensional code, similar approximations, as implemented by ORIGEN, would have to be employed in order to determine a flux to be used in the comput ation of reaction rates. This flux would not incorporate the spatial behavior of the system and therefore lead to inaccurate interaction rate

PAGE 44

44 calculations. This would further propagate into inaccurate temporal nuclide inventory computations. To resolve this issue, CINDER90 must be linked to a steady state reaction rate calculator in order to capture the spatial reso lution and possible spectral change of the timedependent reaction rate behavior. Necessity for a Linked Approach In the zero spatial dimension calculation, the flux is no longer spatially dependant and events such as neutron leakage, scatter and absorption are not accurately depicted. The physics of assuming spatial independence of the reacti on rates is highly false because the energy dependence of these reactions makes capture and scatter events h eavily dependent upon geometric configurations. For example, BWR reactor fuel pins are loaded at various enrichments in combination with gadolinium burnabl e poisons in order to try and achieve a flat power distribution across the bundle.62 These bundles are then strategically positioned within the reactor core in order to further maintain a flat power distribution across the reactor core. A flat power distribution is preferential because if the peak to averag e power in the bundle is too high then fuel centerline melt, creep, and fission gas pressure constraints limit the average power the bundle can be operated, therefore increasing the am ount of needed loaded fuel bundles per cycle and ultimately reducing fuel economy. The ability to meet this flat power distribution demand is limited by the reactor asymmetr ies that force non-periodic boundary conditions on the flux solution. For instance, neutrons, generated in fuel bundles at the core periphery, exhibit more moderation because of the larger interaction with the surrounding water reflector. This asymmetric interaction is further complicated by the fact that the density of the moderator decreases as the moderating fluid heats up and boils as the fluid passes through the reactor core. This reduction in moderation leads to less fiss ion and more capture at the beginning of the operating cycle. The boundary condition for the fl ux solution is therefore set by the asymmetric

PAGE 45

45 properties of the external reflecting fluid and these imposed boundary conditions lead to a flux gradient across the reactor core. There exists further complicating asymmetries within the bundle itself. Burnable poisons, large cruciform control blades, wa ter holes, and structural guide t ubes within the fuel assembly also further contribute to flux asym metry within the fuel assembly.62 The asymmetry in the flux solution further propagates during fuel burnup as cap ture rates and thermaliz ations are spatially affected as a function of the asymmetry in isot ope concentrations resulting from the asymmetry in prior reaction rates resulting from the asymmetries in the fuel assembly and core structure. Therefore implementing a depletion model for a reactor system without accounting for spatial dependence will lead to discrepancies in the soluti on as these types of spatial asymmetries in the flux solution lead to further propagation of as ymmetry in isotope buil dup and depletion. As isotopes consume radiation, undergo nucle ar reaction, and transmutate into new isotopes, the neutron energy spectrum will change due to how the neutrons will interact with the newly created transmuted isotopes.2 Simply assuming a single en ergy-integrated cross section set for an entire depletion calcu lation may not be a valid assu mption for systems that undergo significant spectral changes. For example, in a freshly fueled LWR, initially the majority of energy is created by fission of U-235 atoms. As the core bu rns over time many neutrons are parasitically absorbed by U238 leading to production of U-239. U-239 quickly decays by beta emission to Np-239 and then to Pu-239 which is hi ghly fissile. At the en d of the reactor core life, an immense amount of Pu-239 has been created, absorbed neutrons, fissioned, and contributed to system power. The average thermal absorption cross section for Pu-239 is 1011.3 b which is much higher as compared to U-235 at 680.8 b.7 The average thermal fission cross section for Pu-239 is 742.5 b which is also much higher as compared to U-235 at 582.2b.7

PAGE 46

46 capture cross section is defined as all absorption reactions not leading to fission, and therefore a thermal capture-to-fission ratio may be define d for a specified isotope by equation 2-18: f f a (2-18) Changes in the capture-fission-ratio over time change the amount of neutrons that will be parasitically absorbed as well as created by the fi ssion process. These factors ultimately alter the neutron energy spectrum and thus lead to timedependant changes in the transmutation rates. The process is also highly geometry dependent b ecause the capture to fission ratios in certain sections of a given system will change at differe nt rates due to the energy flux witnesses in a specific region. Inaccurate capture of these ch anges ultimately affects the accuracy of the calculation; therefore it is vita l for transmutation rates to be recalculated when isotope concentration changes result in significant changes in the energy flux distribution. Figure 2-1. Linked Depletion Process. Steady-State Reaction Rate Calculation Time-Dependant Number Density Calculation Isotope Concentrations Collision Densities, Eigenvalue, Flux Normalization Parameters

PAGE 47

47 Since considerable changes in the isotope concentration are re quired in order to significantly alter the neutron energy spectrum, equation 2-1 may be approximated as separable in space and time. The equation may then be di scretized in several tim e steps where a steadystate recalculation of reaction rates is comple ted at time intervals where it is assumed that neutron energy spectrum will be altered significantly enough to alter the trajectory of equation 21. The procedure for this solution method is outlin ed in Figure 2-1. The r eaction rates utilized in the capture terms of equation 2-1, system eige nvalue, and flux normalization terms are first calculated utilizing a steady-state reaction rate ca lculator. Those capture reactions are then assumed constant over a specified time step and implemented in a time-dependent number density calculation. The time-dependant number de nsity calculation then calculates a new set of number densities to be utilized in the subse quent steady-state reaction rate calculation. The process repeats itself until the final step. Methods of Steady State Re action Rate Calculation Steady state reaction rates may be calculated by either explicitly solving a form of the Linear Boltzmann Equation (LBE) or determining the average interaction rate behavior of a sample distribution by simulating the actual phys ical process using th e Monte Carlo method.63, 64, 65 Solving the LBE involves deterministically calculating a continuous spatial reaction rate solution by employing a variety of discretizatio ns and approximations. Accuracy of the calculation is highly dependant upon the validi ty of the discretiza tion and approximation techniques utilized to re present the system physics.64 Assumptions must be benchmarked against viable experimental data in order to ensure calculation accuracy. The Monte Carlo method, however, simulates the actual phys ical process through the random sampling of probabilistic events. The method minimizes approximations app lied to system coefficients and relies solely on knowledge of the probability of particle interaction.

PAGE 48

48 Due to the random sampling process, the re action rates calculated are stochastic and therefore the solutions are probabilistic, in contra st to deterministic, thus having an associated statistical error.63, 64, 65 Reducing the stochastic error involves modeling larger sample distributions to better account for the aver age behavior of particle interactions.66 The need for larger sample sizes to secure adequate calcu lation precision generally causes the Monte Carlo method to execute slower than a typical determinis tic calculation. In cer tain modeling scenarios, an investigator may wish to analyze a plet hora of perturbations to a well understood benchmarked system of which the approximations applied within the deterministic solution method are valid or within the solution tolerance for analyzing the system. The deterministic method is more suited for this type of calculati on because the gain in co mputational performance outweighs the accuracy penalty of which may be w ithin the bounds of the solution tolerance. In other modeling scenarios, an investigator may wish to analyze a system that is not well understood and/or not benchmarked. For this type of calculation, the investigator may wish to compromise computational performance at the e xpense of enhancing solution accuracy. At the cost of computational performance, the lack of coefficient approximations, exactness in 3dimensional modeling, and explicitness in a ngular tracking gives the Monte Carlo method a distinct advantage for this type of modeling scen ario. The crux of each modeling technique will be detailed in the subsequent sections. Deterministic Approach to Reaction Rate Calculations The deterministic approach to reaction rate calculations involves solving the LBE by using a variety of assumptions to achieve a quick, re liable, spatially conti nuous reaction rate and neutron multiplication solution. The steadystate LBE is defined as equation 2-19:

PAGE 49

49 ) ( ) ( ) ( 4 ) ( 1 ) ( ) ( ) ( ) ( ) ( 'E r d E r E v E d E k E r E E r dE d E r E r E rg f s t (2-19) Generally, this equation is discretized in en ergy, space, and angle, or an approximation for these dimensions is applied, in order to cast the equation in a deterministic solvable form. 63, 64, 65 The energy variable is discretized into fe w defined energy groups ultimately segregating equation 2-19 into a set of coupled differentia l equations, where the coupling of energy groups results from the out-group scatter and fission terms: G g g g f g G g g gg s g g t gr d r v k r r d r r r1 1 ,) ( ) ( 1 ) ( ) ( ) ( ) ( ) ( (2-20) The angular dependence of equa tion 2-20 is accounted for by e ither discretizing the angular dependence into few angular quadr atures, as in the discrete or dinates method, or by assuming isotropic scatter behavior and linear dependence of the angular flux, as in diffusion theory. 63, 64, 65, 67 After angular dependence is ac counted for, the examined system geometry must be meshed into a fine grid or cast into a solvable ra y tracing system. Reaction rate distributions are calculated within each mesh or along each ray by using assuming homogenized parameters within the mesh or along the ray and implementing the partic le fluxes and currents of the neighboring meshes or rays as the boundary cond itions for that specific mesh or ray solution. 63, 64, 65 The mesh shapes or rays should be chosen so that a problems geometry may be modeled explicitly, and the homogenizations within each of the shapes or rays do not upset the fidelity of the actual calculation. The ideal mesh or ray sc heme would include thousa nds of tiny meshes or

PAGE 50

50 rays so that connect in highly irregular geometries in order to explicitly model a complicated geometry. However, thousands of meshes wo uld require thousands of angular flux at many angular quadratures and many energy groups whic h would ultimately exceed the memory of any computer. Irregular meshing schemes are available such as in the code At illa; however, irregular shaped meshing schemes lead to a highly memo ry intensive bank of complicated algorithms utilized to find a flux solution for each irregular mesh scheme.68 Assumptions must be employed on acceptable mesh size and shape so as not to violate the allotted computational memory restrictions. Therefore mode ling of highly irregular complicated geometries must be approximated by simpler less precise models th at may possibly affect the accuracy of the solution method. Ray tracing schemes such as that implemented in the method of characteristics (MOC) are capable of better accoun ting for complicated geometries by setting up a scheme that solves a solution at the explic it boundaries of a given geometry. The MOC method has been successfully applied for modeling complicated geometries.69 However, the MOC method, like the discrete ordinates method, relies on approximation schemes to develop proper coefficients ( i.e. interaction cross sections) in order to solve the deterministic form of the transport equation. Within the Evaluated Nuclear Data Files (ENDF ), exist infinite-dilute, energy dependant reaction rate probability, cross section, data.70 This data functions as the coefficients for the transport equation. The data itself are stated at many energy points and therefore is not exactly continuous in energy, however since the magn itude of data points existing for a specific reaction is considered in fine enough fidelity to assume a continuous distribution, the data is deemed continuous. The ENDF data therefore re presents a continuous function for each reaction type that is dependent upon impinge nt energy, and used as the coef ficients in order to solve the transport equation. Parameters also exist fo r reconstructing complicat ed sections of the

PAGE 51

51 coefficient function such as the resonances resu lting from the eigenstate s of the nucleus. The data are also stated to be infinitely dilute b ecause it is assumed that reaction probabilities do not incur energy and spatial self-shiel ding. Figure 2-2 shows two separa te distributions of flux in the vicinity of a cross section res onance. As compared to a 1/E distribution, the actual flux, (E), will be depressed in the neighborhood of a cross s ection resonance peak. This effect is termed energy self-shielding since the large interaction pr obability of the resonance shields the isotopes nuclei from neutrons around the energy E0 leading to a flux depression. If the nuclei of interest is then further contained within a lump of sim ilar nuclei and placed in th e pathway of a beam of neutrons with energies in the vi cinity of the resonance energy of those nuclei, the majority of neutrons will only interact with nuclei at the surf ace of the lump. This effect is termed spatial self shielding because the inner nuclei of that lump are spatially shielded by the nuclei at the surface of the lump.2 In implementing a deterministic solution approach the reaction rate coefficients must be properly adjusted to adequa tely account for these ty pes of self-shielding in order to properly model the true system physics. Figure 2-2. Flux depression in th e neighborhood of a resonance at E0. 1/E (E) E0 E

PAGE 52

52 Casting equation 2-19 into the multi-group form requires that multi-group coefficients be generated. The coefficients are generated base d on the premise that the multi-group coefficients be a homogenized term that best represents th e average behavior within a specified energy group. To preserve true system physics, th e energy group averaged interaction rates are calculated dividing the energy integrated reactio n rate, within a specific energy band, by the energy integrated flux within that same energy band as shown in equation 2-21: 1 1) ( ) ( ) ( ) (g g g gE E E E gdE E r dE E r E r r (2-21) The concern with this approach is that the flux, th e variable that is to be to calculated in equation 2-20, is used to calculate the cross section which is then utilized to calculate the flux in equation 2-20 making the process nonlinear. To make the process linear, an approximation is made in which first a homogenous, spatiall y-infinite, 1-D dimensional calcu lation is completed with the continuous energy ENDF cross sections. The flux so lution from that calculation is then used to generate collapsed group energy self-shiel ded cross sections for a further multi-group calculation.63, 64, 65 Deciding which energy bands to use to compute the group average cross sections will highly affect the ultimate outcome of the further calculations that implement those cross sections. Ideally, a proper group structure, robust enough to be used for a wide class of calculations, would account for every major resonance of ever y single nuclide. Figure 2-3 displays the absorption cross sections for a few nuclides found in typical reactor calculations. Notice that each isotope contains many different resonances therefore requiring a large amount of groups to capture the true interaction rate be havior of the displayed isotopes. If the rest of the 389 isotopes

PAGE 53

53 containing ENDF data are figured into the calcul ation, one may then see that an immense amount of energy groups is needed in order to model each resonance. Because the size of the group structure is one of the many parameters that di ctate the needed memory allocation to solve a specific problem, in practice, the ideal appro ach is not viable becau se it is possible that thousands of energy groups would be necessary in order accurately capture the resonances of each available isotope. Therefore group structures are chosen that utilize as few groups as possible, in order to account fo r as many resonances as possibl e, to make the group structure small enough to be practical for an actual calculation. Figure 2-3. Flux depression in th e neighborhood of a resonance at E0. Accurate group structures may be generated fo r classes of problems. For example, the nuclear power industry contains accurate multigroup cross section libraries for examining

PAGE 54

54 reactors containing uranium-oxi de fuel, with zircaloy clad, surrounded by water, and possibly controlled by born-carbide or hafnium.25, 26, 30, 32 However, those libraries are not applicable for examining exotic isotope combinations such as uranium-nitride fueled, rhenium cladded and niobium-carbide coated space reactor designs.71, 72 Furthermore, the nu clear industry is only successful in determining these libraries after ye ars of comparing and tu ning cross sections to operational data. The implementer of a deterministic method is constantly burdened by the choice of appropriate approximations in order to achieve an accurate result. If chosen correctly the implementer will achieve a quick, reliable, non-stoc hastic, spatially contin uous reaction rate and neutron multiplication solution that runs much fast er than a Monte Carlo. However, the burden of these choices and the knowledge to make prope r determinations of gr oup structure and selfshielding accountability may be out of the expe rience base of the implementer or may not be possible with current memory restrictions, in which case a more robust tool exhibiting fewer approximations of system physics should be implemented. Monte Carlo Approach to Reaction Rate Calculations The Monte Carlo method uses probability theo ry to model a system stochastically by simulating the actual physical process thr ough the random sampling of events. Like deterministic methods the Monte Carlo method seek s to solve the transpor t equation to determine the average behavior of the ne utron population within a system; however, unlike deterministic methods the Monte Carlo method solves the trans port equation in a probabilistic approach. The Monte Carlo method does not requ ire clever simplifications a nd approximations to determine transport coefficients and model geometry in orde r to compute solutions of neutron behavior. In certain circumstances, simplifications and appr oximations employed by deterministic methods alter a calculation so significan tly that accuracy of neutron beha vioral characteristics is lost.

PAGE 55

55 Since the Monte Carlo method does not rely on these simplifications and approximations of system physics, the method is highly robust a nd capable of modeling very complex systems.36, 63 Combinatorial geometry is used to build models for simulation within the Monte Carlo process.50 Geometric shapes are represented by the combinations of basic surface types. The geometric shapes created from the combination of these surface types are called cells. These cells contain homogenized information of isot ope concentration and particle interaction behaviors. Most possible shapes/cells incurre d in nuclear system design may be modeled as combinations of these basic surf ace types; therefore making it po ssible to explicitly model any type of geometry encountered. Particle positi on is tracked in reference to these cells and surfaces, and the particle track length and intera ction behavior are determ ined from interaction characteristics within each of the encountered cells.63, 64, 65 A basic diagram of the Monte Carl o process is displayed in Figure 2-4. Initially, a particle of specified energy and direction enters a cell volume. The distan ce the particle travels within that cell volume is dependent upon the probability that the particle wi ll have an interaction within the specified volume some distance into the volume. In the one-dimensional case, the probability that the particle interacts within a distance between point a and point b is determined from the representative probability de nsity function (PDF), f(x), and calculated by equation 2-22: b ab x a P dx x f } { ) ( (2-22) The probability of an event clearly may not be le ss than 0. If x may be any real value between and the distribution may be normali zed resulting in equation 2-23: 1 ) ( dx x f (2-23)

PAGE 56

56 From this equation we can further derive the cu mulative probability dist ribution function (CDF), F(x) in equation 2-24, which will be utilized to ultimately determin e the actual length of particle travel: x x P x F dx x fx ) ( ) ( (2-24) In order to then simulate the pr ocess of particle trav el through the random sampling of events we set the cumulative distribution f unction equal to a random number, and invert the equation in order to compute a correlated path length of travel. Figure 2-4. Flow diagram for the M onte Carlo particle transport process If the calculation is in terms of mean free paths and 0 < x< the PDF is xe x f) ( (2-25) and the associated CDF is xe x F 1 ) ( (2-26) No Track enters volume Go to cell boundary and enter next cell Determineinteractin g isoto p e Compute distance to next Determine interacting isotope Determine reaction type and resolve residual parameters Interaction inside cell Particle killed? Yes Yes No

PAGE 57

57 Therefore, inverting the cumula tive probability distribution functi on to solve for the mean free path results in ) 1 ln( x (2-27) and if we choose to be uniformly distributed betw een 0 and 1 we can further write ) ln( x (2-28) or in terms of track length, l, utilizin g the individual isotope number density iN and microscopic cross section total cross section, t iE,) ( i t i iE N l,) ( ) ln( (2-29) If a particle then interacts with in a cell volume, then the isotope that the particle interacted with must be determined. The probability of a particle at specific energy, E, interacts with isotope i is simply the ratio of the number density of i multiplied by the total cross section of i to the sum of all the isotopes number densities mu ltiplied their associated total cross sections: i t i i t i i iE N E N E p, ,) ( ) ( ) ( (2-30) Then to sample the process from a random event, and assuming to be randomly distributed between 0 and 1, isotope m is selected for interaction only if the following criterion is met: k i i N i i k i iE p E p E p1 1 1 1) ( ) ( ) ( (2-31) After the interacting isotope is determined, the type of particle interaction incurred is sampled in a similar manner. First, the probabi lity of incurred a specific reaction, m is computed by [41]: r m k m k m kE E E p, ,) ( ) ( ) ( (2-32)

PAGE 58

58 Then to sample the process from a random event, and assuming to be randomly distributed between 0 and 1, isotope m is selected for interaction only if the following criterion is also met: m i i k N i i k m i i kE p E p E p1 1 1 1 ,) ( ) ( ) ( (2-33) After the interaction type is determined then subsequent residual nuclides/particles are determined and resulting energy and a ngular distributions are calculated. Particles continue to move within the system un til the particle is either absorbed or leaked from the system. A tallying estimator is used in order to retrieve particle information from the system such as reaction rates and fluxes.63, 64, 65 The most basic Monte Carlo tally estimator is the collision estimator which tallies the amount of a specific collision within a cell volume. In order to calculate the average number of collisions in a volume, V, per unit time the following is used equation: cV c ~ ~ (2-34) For the Monte Carlo computation the fl ux may therefore be approximated by: cV c ~ ~ (2-35) Where is the average flux, c ~ is the average macroscopic colli sion interaction cross section, and cis the mean number of collisions in a volume,V ~ If the calculation is normalized to per source particle then the collisi on estimator may be rewritten as: i i cc N V 1 ~ ~ 1 (2-36) A variation of collision estimator is the absorpti on estimator where the flux is now tallied not at every collision site but at each absorption site:

PAGE 59

59 i i aa N V 1 ~ ~ 1 (2-37) Both the collision and absorption estimators are dependent upon interactio ns taking place within a volume.63, 64, 65 Due to the stochastic nature of the Monte Carl o process, all quantities computed result in a stochastic error (standard error or relative error). The relative error is composed from the sample variance. The actual vari ance of the population is N xN i i 1 2 2 (2-39) where is the actual population mean. Unfortunate ly, the actual population mean is almost never known, and therefore the sample mean, x is used to determine a sample variance. The xis tend to be closer to their average, x than to the population average, resulting in estimated values that are too sm all on average, therefore a n-1 is used in the divisor to account for this underestimating. For a sample distribu tion, the sample variance of that distribution is calculated by equation 2-38: 12 2 n x x Si (2-38) With further mathematical simplification we derive73 1 2 1 2 1 2 12 2 2 2 2 2 2n x n nx x x n x x x x n x x x x n x xi i i i i i i 2 2 2 21 x x n n x xi i (2-39)

PAGE 60

60 The sample variance is referred to having n-1 degrees of freedom. This terminology results from the fact that although the sample va riance is based on the n quantities, x x 1, x x 2, ..., x xn, these sum to 0, so specifying the values of any n-1 of the quantities determines the remaining value.73 The Strong Law of Numbers states that if the true mean is finite then the sample tends to the true mean as the sample size approaches infinity.66 Furthermore, examining equation 2-38 we see that increasing the sample size, n, decreases the sample variance. The variance of the sample mean is calculated to then be implemented in the calculation of the relative error74 n x x n x n x n x n x x x S R n S Si i i i x x1 12 2 2 2 2 2 2 2 2 (2-40) Because the relative error implem ents the sample mean variance, increasing the sample size further decreases the relative error of the calcula tion. A major shortcoming of the collision and absorption estimators is that reliable estimates of the flux, and thus interaction rates, may only be computed if many collisions/absorptions occur within tallied cell volume. For cell volumes containing minimal amount of collisions/ absorptions the relative error fo r calculated quantities will be high and thus leading to a lack of confidence in the computed quantity. The track length estimator is another type of Monte Carlo estimator that does not incur the mentioned shortcoming. The track length estimator utilizes the particles average track length, l, within a specified cell volume to tally flux, a nd therefore relies of particles passing through a specific cell volume and not if the particle happ ens to interact within the cell volume. The particle density is equa ted to the scalar flux by dVdEdt t E r vN dVdEdt t E r ) ( ) ( (2-41)

PAGE 61

61 Since a particle with travel a distance vdt duri ng the time increment dt, the scalar flux may be thought of as the total of the path lengths trav eled during dt by all part icles in the phase space volume of dVdE.63 The scalar flux may therefore be defi ned as the total track length traversed by particles within a specified volume V l~ (2-42) where lis the average track length per unit ti me a particle traverses in volume, V ~ In a Monte Carlo simulation utilizing N particles the track le ngth estimator of the flux may be calculated by: n nl N V1 ~ 1 (2-43) Because the track length estimator of the fl ux relies only upon tracks en tering a volume, more particles contribute to the calculation of the re lative error and therefore computing reaction rates with lower relative error is easier with this flux estimator.63 Monte Carlo applied to neutr on transport may make two dis tinct types of calculations: fixed source and criticality. A fixed source ca lculation involves followi ng a preset amount of particles from a source, given by the user, until the particles are ki lled. Criticality calculations, however, determine the effective neutron multiplic ation of a system by constantly changing the geometric source distribution of neutrons for subsequent neutron cycles and following each neutron till it is killed in an inelastic reaction, scattered below the weight threshold, or leaked from the system and then determining if the way in which the neutron was killed resulted in production of new neutrons. Many cycles of s ource distributions are an alyzed in order to converge on a source distribution that is truly representative of th e analyzed system so that an accurate characterization of the neut ron multiplication may be computed.50

PAGE 62

62 For a criticality the total amount of particles examined in the computation may be broken up into the particles distributed per cycle and the total number of cycles examined. Initially the user inputs a guess at the source di stribution of particles. These particles then transport and are eventually killed by an inelastic event, scattere d below a weight threshol d, or leaked from the system. The absorption sites contai ning fissile nuclides that lead to fission then become the birth sites for the next set of particles in the subseque nt cycle. At each absorption site containing a fissile nuclide, the amount of neutrons produced per fission is determined. The amount of neutrons created per source neutron examined represents the multiplication of the system or keff. If the system is a self sustaining critical system for each source neutron lost from the system and additional neutron is created for the next cycle. The multiplication of the system is a global parameter and therefore keff is used to explain the entire global system. Similar to the mentioned tallying process, three estimators exist for determining criticality. Each estimator involves multiplying the flux estim ator by a response function, h, to determine the system multiplication:74 i n i f n iE E h) ( ) ( ~ (2-44) ) ( ~ ,Ef n i is the macroscopic fission cross secti on of isotope i at energy E, and ) (Ei is the total number of neutrons per fission of fissile isotope i. The collisi on and track length estimators are tallied from interacti ons with each isotope i in all volumes n, by:74 ni c n i n i f n i n c effE E E c N k) ( ~ ) ( ) ( ~ 1, , (2-45) i n i f n i n n l effE E l N k) ( ) ( ~ 1, (2-46)

PAGE 63

63 The absorption criticality estimator is computed slightly different form the absorption flux estimator in that only absorptions in fissionable nuclides are tallied ni a n i n i f n i n a effE E E c N k) ( ~ ) ( ) ( ~ 1, , (2-47) All three estimators (similar to the flux estimators) rely on large sample sizes and large numbers of hits (i.e. collision/ absorptions/number of entering tr acks) in order to decrease the relative error of the co mputation to have high confidence in the accuracy of the computed quantity. Therefore, unfortunate ly, decreasing the relative error of the calculation is directly correlated with increasing the associated computa tional cost of increasing the sample size to be followed. Historically, the Monte Carlo me thod has been used only as a benchmarking technique because of the computa tional cost associated with large sample sizes. Deterministic codes generally calculate results much faster than Monte Carlo codes. However, as core designers start to examine more advanced technol ogies that exhibit behavi or that violates the validity of the assumptions of the current de terministic methodologies, a method that captures the true systems physics accurately becomes vital in order to effectively characterize system properties. Current Capabilities and Methods of State of the Art Depletion Codes Many depletion codes exist in industry a nd academia for the purpose of determining temporal nuclide inventories a nd reaction rate distributions.35 These codes are either externally linked or sequential build technologies that cont ain a set of constraints used for determining which depletion and transport coefficients to track and calculate. Depending upon the investigated system, more or less detail in the solution method may be invoked in order to achieve an acceptable characterization. Th e level of acceptable characterization may be dependent on the validity of the benchmarking for the applied method or the actual level of

PAGE 64

64 fidelity necessary to achieve proper knowledge for further design consideration. As many calculations are required in order to license a reactor core, codes utilized by the nuclear power industry for continual fuel cycle design of heavily benchmarked systems may employ many approximations in order to achieve quick reliabl e results within the timeframe deadline allotted by a particular customer.9 These assumptions are acceptable based on the premise that the benchmarking of the system covers the desi gn application space in which the code is implemented. On the other hand, codes utilized for the development of radically new system designs, incorporating a wide vari ety of unique material and geometry combinations that are not heavily benchmarked, may implement a higher fidelity of detail in the solution method employing few if any approximations. Implementation Architecture The main focus of most industry-based depletio n codes is to support the operation and fuel cycle design of commercial pow er reactors. Industry-based co des usually focus on streamline manufacturing of heavily benchmarked system s that push the limiting design constraints to maximize returnable profit.35 These types of systems involve many optimization iterations of common heavily benchmarked technologies. Only br oad, generalized parameters are required to meet limiting conditions of operation (i.e. eigenv alue, power density, and linear heat generation rate) for most commercial reactor designs. Ther efore, because the calculation tools are heavily benchmarked, the design requirements are broa d, and the designer must complete the final design analysis within a short prescribed time peri od, most of the current licensed industry codes employ deterministic-linke d depletion methodologies.35 Although various companies, such as Westinghouse General Electric, Areva, and Studsvick, use slightly different techniques to ap proximate time-dependent reactor behavior, the overall methodology and processes are quite simila r. First, a fine-fidelity deterministic

PAGE 65

65 calculation (using PHOENIX, TGBLA, CASMO-3, et c.) examining many axial and radial slices of the reactor are completed to generate coarser 1to 4group-averaged interaction rate parameters for implementation in a coarser calcu lation of the full-core geometry (using ANC, PANACEA, PRISM, etc.).35 However, due to the limitation s of the deterministic method, these methodologies are only valid within the design space that they have been benchmarked for. Further analysis in group structur e and cross section self-shielding is required in order to attempt to apply the tools to more exot ic applications. These codes choose to follow a predetermined subset of reaction rates for the transport calcul ation, and further propagate a subset of fission products to be followed during the depletion calculation. Certai n reaction rates may not be of importance to specific reactor designs or may onl y contribute within the noise of an acceptable solution. Omitting these unnecessary reaction rates results in extra availa ble memory as well as minimizing computation cost. For most industry based codes major fission products such as the Sm and Xe isotopes may be followed explic itly while the buildup of minor reactivity contributors may be grouped in to a lumped fission product set.2, 25, 29, 31, 32 The advantage to fission product lumping is that fewer fission prod ucts are followed in the transport process leading to a speedup in comput ation. The fission product lump s are, however, geometry and spectrum dependant, and therefore though one ty pe of fission product lump may be acceptable for a certain type of system, that same fissi on product lump may be acceptable for a different system. The combination of the acceptability of the transport method solution (i.e. energy group structure, angular dependence, and geometric approximation), the reactio n rates followed, and the fission products tracked limits the application space for most industry based depletion codes. These codes are usually only valid for the sp ecific systems for which the codes were originally developed. For example, a multi-group code developed for and LWR with certain

PAGE 66

66 group structure, fission products, and reactions followed will not be accurate in determining individual gamma-line signatures of fast reactor fission products for implementation in detector signal segregation. However, that same code is very capable of determining an adequate neutron multiplication and power shape calculation for licensing the reactor for operation. Academia-based codes generally are developed for finer fidelity design of a robust variety of nuclear systems that are not heavily benc hmarked. These codes must implement as few assumptions as possible in order to achieve an ac curate result. The types of analysis for these systems involve understanding more precisely the se nsitivity of certain pa rameters within the design making a higher-fidelity methodology, the design tool of choice. In general, deterministic codes developed in academia implement fewer a ngular approximations [i.e ., discrete ordinance transport theory (XSDRNPM) versus diffusion theory], implementing a higher degree of examined dimensions (i.e., 3D versus 2D), and/or using a finer-ene rgy or continuous group structure.35 Even though higher fidelity deterministic technologies ex ist to address concerns of angular dependence, by nature of the process, these technologies still suffer from the burden of proper group structure generation and fl ux self-shielding considerations. To eliminate dependence from this burde n, burnup codes coupling the Monte Carlo reaction rate calculation have received more attention.36-41,46-48,75 However, due to the long computation times required to execute a detailed high fidelity calculation, these codes have had limiting success in implementation for commercial reactor design processes. To overcome this issue, users would have to consid er utilizing longer irradiation st eps, with larger spatial zones, following fewer nuclides, and fewer interaction rates per nuclides. An unfortunate example of this issue is exemplified in the MCODE.39 In the MCODE only isotopes specified by the user are followed in the transmutation process, and only (n, ) and (n,f) energy integrated reaction

PAGE 67

67 rates are calculated for each nuclide. This simplification neglects key (n, ) reactions as witnessed in the depletion of boron-10. Furt her acceleration techniques for Monte Carlo may include using multi-group Monte Carlo such as in Keno Va.76 Multi-group Monte Carlo speeds up the search algorithm when determining and ener gy dependent cross secti on at an interaction site because far less energy groups need be queri ed. However, the limitation of the multi-group approach is ultimately the validity of the group st ructure to a specific calculation. Implementing approximations to a Monte Carlo linked capability, unfortunately, reduces the robustness of the technology; therefore to take advantage of the explicitness of the Monte Carlo process, computational cost will have to be sacrificed. All linked depletion methodologies suffer from how the linking process affects the functionality of both codes. For example, MONTEBURNS links MCNPX/5 with ORIGEN2.2 for high fidelity depletion calculations. MCNPX/5 is capable of tracking ~393 isotopes for particle transport and ORIGEN2.2 is capable of tracking 1700 nuclides for transmutation. MONTEBURNS only chooses to follow a small subset of isotopes for particle transport determined by specific reactivity criteria, and is not capable of following multiple isotope transmutations at multiple temperatures.75 Other features within MCNPX/5 or ORIGEN2.2 are also not compatible with the MONTEBURNS implication as the code was not developed for these purposes. Many other linked technologies su ffer from this same deficiency. All linked methodologies suffer the added external input stru cture necessary to execute the coupling. For example, the SCALE package successfully pres erves the functionality of each code by implementing each package as a separate module and using a driver code to execute link each module however, the driver inputs are limited to certain setups a nd therefore though each code is run as a separate module, the linking input stil l limits the user to a specific implementation.77

PAGE 68

68 Specific features of code mechanic s that enable the user larger fl exibility might be ignored in the linking process such as: operating in parallel execution and con tinuing a stopped execution. All externally linked technologies suffer from the fact that th e directory structure of the linked package must be coordinated with the setu p of each of the technologies to be linked, and only the functionality of the linked code is suppo rted by the developer of the linking process. For example, ORIGEN is supported by Oak Ridge National Laboratory, MCNP is supported by Los Alamos National Laboratory, and MCWO, a script linking MC NP and ORIGEN, is supported by Idaho National Laboratory; however MCWO does not support development of ORIGEN or MCNP.41 Therefore if a user of MCWO finds a particle transport error in MCNP or a transmutation limitation in ORIGEN, developers of MCWO are not responsible or able to support fixes of these independent codes. Code s like the SCALE package benefit from the fact that each code used for calculation is supported by the same organization.76, 77 Therefore an optimum multi-purpose depletion pa ckage is a code system that minimizes approximations in the transport, depletion and linking calculations, supp orts development of all codes implemented in the solution method and maintains full functional ity and benefit of each code used in the calculation. Nuclide Inventory Limitations/ Calculation Size Considerations The goal in designing any nuclear system is to determine the minimum amount of incurred cost necessary in order to achie ve the power and cycle requireme nts while still meeting safety margin considerations.78 To determine these limiting conditions of operation only broad generalized parameters are requir ed (i.e., eigenvalue, power dens ity, linear heat generation rate, and shutdown margin). Hundreds of combinations of loading and control strategies incurring many design iterations are examined in order to m eet limiting design criteria Therefore in order

PAGE 69

69 to reduce the computational modeling expense approxi mations of larger system behavior must be made. To model an entire commercial power re actor core explicitly, capturing the exact interaction rate behavior of every single micros copic section of every si ngle material in the system, would be an extremely memory restric tive. Considering that the power distribution varies continuously in the radial and axial dir ections, reactor designers are forced to make approximations in order to discre tize the full depletion model so th at it may be solved. Within each discretization, reacti on rates for each isotope of interested must be computed and stored for implementation into the time-dependent isot ope concentration calculator. For finer discretizations, larger memory requirements a nd increased computati onal costs are incurred.62 For cases that are not con cerned with microscopic phenom enon, instead of trying to simulate an entire system explicitly, detail of a systems behavior may be resolved from examining the large system as composed of average models of explicit sections of the reactor. Many calculations of explicit sections of the larger reactor model are completed and major interaction parameters are collapsed into an average set representing the explicit microscopic section. Then these average sets of parameters are implemented into a courser representation of the full system leading to a final computation th at does not require as much explicit detail and therefore reducing memory restriction a nd enhancing computational incentive. Currently in the commercial power industry, reactor physics computations are bifurcated into two separate processes. First, a mu lti-group or stochastic calculation (using MCNP, PHOENIX, TGBLA, CASMO-3, etc.) is complete d on a fuel bundle at certain axial zones to generate collapsed interaction pa rameters. Then those generated co llapsed interaction parameters are implemented in a full-core simulator tool (using ANC, PANACEA, PRISM, etc).35 The

PAGE 70

70 advantage to this process is the speed up in co mputation; however, the major fault is that the legitimacy of the process now rests upon the validity of the chosen discre tizations and averaged parameters that are generated from the finer expl icit micro-section calculations, and the fidelity to determine microscopic behavior of intere st is now lost in the averaging process.35 The most robust, optimum design tool must a llow the user the option to examine different sizes of possible geometric averaging techniques as well as the capability to eliminate them entirely. Due to the robust modeling capability of combinatorial geometry, Monte Carlo codes are capable easily modeling both intricate explicit m odels of the entire system as well as courser representations implementing large blocks of common materials that incur decreased computational expense. Since the Monte Carlo method employs few, if any, assumptions of system physics, collapsed averag e interaction parameters generated by the code may be secured to be of highest fidelity. This makes utilizi ng Monte Carlo for either group constant generation or full core modeling the optimum robust linked depletion tool.

PAGE 71

71 CHAPTER 3 METHODOLOGY DEVELOPMENT A driver program is required in order to structure the execution of coupled algorithms within coupled process. The MCNPX code se rves as the driver program for the coupled depletion process. MCNPX was chosen as the driver program because the code is already a large scale multi-purpose particle transport package. The MCNPX code package contains ~1000 subroutines with each subroutine containing hund reds of execution lines. The CINDER90 depletion algorithm is placed with in the MCNPX code package, and an interface subroutine sets CINDER90 inputs and retrieves CINDER90 outputs The other 29 implemented subroutines execute various assignments in order to properly average data, set parameters for implementation in the CINDER90 depletion algorithm, reconstr uct atom density information for subsequent particle transport calculations, retrieve input information for problem set up or to construct output. The coupled deletion process involves a Monte Carlo (MCNPX) steady-state reaction rate calculation linked to a deterministic depletion calculation (CINDER90). The whole process resides in roughly 30 subroutines in the MCNPX code (five new subroutines and 25 augmented subroutines). The linking process is displaye d in Figure 3-1. MCNPX runs a steady-state calculation to determine the system eigenva lue, 63-group fluxes collision rates, and Q values. CINDER90 then takes those MCNPX-generated valu es and carries out the depletion calculation to generate new number densit ies at the following time step. MCNPX then takes those new number densities and generates another set of flux es and reaction rates; th e process repeats itself until after the final time step.79 The MCNPX linked depletion ar chitecture requires important decisions regarding isotope tracking philosophy and methods, calculation of depletion equation coefficients, time-dependant cross section av eraging techniques, and the implementation architecture. 79 Chapter 3 details the development a nd implementation of these philosophies

PAGE 72

72 within the MCNPX depletion methodology and offe rs key important considerations to account for in the development similar technologies. Figure 3-1. Monte-Carlo-linke d depletion process model. Approach to Isotope Tracking MCNPX depletion burns a system by specified bu rn material instead of by individual cell location.78 This technique was employed in order to take advantage of repeated structure geometry simplifications already available in the MCNPX geometry build subroutines that help decrease the size of necessary input in order to burn an entire system.50 The trade off in taking advantage of this input simplification is that th e user must be aware th at unique materials are required for each burn cell containing a unique powe r profile. However, the advantage of this capability eases the burden in modeling complicated geometries. MCNPX depletion requires that each burn mate rial contain all isotop es to be generated within that material at the beginning of the com putation. This therefore requires that the code N o CINDER90 Depletion Calculation Collision Densities, and Q Isoto p e Concentrations MCNPX Steady State Eigenvalue Calculation Final Depletion Step? Final Output Processing Yes

PAGE 73

73 determine all fission and activation products upfront as well as particle interaction data to be used in the depletion process. As mentioned prev iously, the depletion process is bifurcated into two separate processes: steadystate particle transport and isot ope transmutation. Particle reaction data for use in the particle transport calc ulation is available from direct experiment via evaluated data library sets. These evaluated library sets include ENDF, JEF, JENDL, BROND, CENDL, etc. Because most of the evaluated libra ry data sets include reaction information from experiment, ability to execute and retrieve data from experiments limits the amount of isotopes that actually posses evaluated library transpor t information. For example, generating an experiment to measure interaction rate data from a radioactive isotope with a very short half-life may not be possible and therefor e the evaluated nuclear data f iles will usually not contain transport information for such an isotope. Th e current ENDF/B VII.0 library only contains actual transport cross sec tions for 393 isotopes, and therefore, without an analytical or empirical model, a maximum of 393 isotopes may be tracked in particle transport.80 Unfortunately, the actual transmutation process has the possibility of ge nerating 3400 isotopes. Though there exists measured decay data for most of these nuclides, interaction rate data does not exist. CINDER90 is capable of tracking 3400 isot opes in the transmutation proce ss; however, transmutation rates are required in order to complete the calculation.61 These transmutation rates will only have spatial dependence if determined from the steadystate reaction rate calculation. However, if continuous energy cross section data does not exist for a specific nuclide, that nuclide cannot be employed in the steady-state reaction rate calcu lation. For nuclides not containing tabulated transport cross sections in the MCNPX data lib rary, MCNPX does have the capability to use a theoretical model to generate a cross section for that specific is otope; however, this capability was benchmarked for > 150 MeV particle interacti ons, which is well above the energy regime for

PAGE 74

74 the majority of particle interactions incurring in a reactor, and therefore not a safe technique to be implemented in reactor calculations.79 CINDER90 does contain 63-group transmutation rate information for 3400 nuclides.61 The majority of this data, not derived from evalua ted library files, was ge nerated from analytical cross section modeling codes or by expert best estimates. Therefore MCNPX applies a mix and match method for transmutation rates in which if there exists and evaluated library transport cross section for a nuclide for use in MCNPX, then the reaction rate information from that nuclide is calculated directly within MCNP X and sent to CINDER90 as a transmutation coefficient. For all other nuc lides, a 63-group spatially depende nt flux is calculated within MCNPX and sent to CINDER90 to be match with a 63-group cross section sets available in the CINDER90 data library file, wher e that cross section set most probably was derived from an analytical model or expert best estimate, to be in tegrated into a transmutation coefficients for use in the depletion equation.79 To save computation time and reduce the influx of memoryprohibitive information, MCNPX only tracks thos e isotopes accounted for from the following processes: 1. Listing the isotope on a material card. 2. Selecting an isotope from a preset fission product tier. 3. Producing the isotope from the Isotope Generator Algorithm. CINDER90 still tracks all of the daughter/granddaughter/gre at-granddaughter/etc. decay reactions for 3400 isotopes; theref ore, the total isotope concentr ations listed in MCNPX are the true concentrations from the full depletion process. Fission Product Tiers Executing a successful depleti on calculation involves accounti ng for the reactivity effects of as many fission products as possible. In most cases, a fission event will result in the creation of two-to-three fission products with a total mass th at sums to less than th e mass of the original

PAGE 75

75 isotope. The creation of the fi ssion product isotopes is probabilis tic in nature and follows a distribution. Figure 3-2 displa ys the fission product distribut ion for U-235 for thermal and 14 MeV fission reactions. Hundreds of fission product isotope yield combinations are possible for a given fission event from a specific isotope. Each fissionable isotope also c ontains different yield distributions. Figure 3-3 show s fission product yield distributi on for certain pluotonium and uranium isotopes. CINDER90 contains energy dependent fission product yield data for 1325 different fission products.61 Figure 3-2. Uranium-235 fissi on product yield distribution.7 Due to the nature of the Monte Carlo process, increasing the amount of isotopes tracked in a problem increases the associated computationa l cost. When a particle interacts within a specified cell volume, a calculation must be run to sample which isotope within that material the particle interacted with. This sampling is dependant upon individual isotope atom density, interaction cross section, and to tal amount of isotopes within that given cell. Increasing the amount of isotopes, increases the overall associat ed computational cost. However, not tracking certain fission products within a calculation lead s to altered neutron spectrum ultimately leading to inadequate estimation of system reactivity and inaccurate representation of true system

PAGE 76

76 physics. Therefore determining an adequate combination of fission products necessary to sufficiently account for the system physics becomes paramount in achieving an accurate depletion calculation. Figure 3-3. Fission Yield Di stributions for Various Plut onium and Uranium Isotopes.81 Many options exist for determining how the user accounts for the generation of fission products within a calculation. The user could be forced to list, on an input deck, all the fission products that were of interest. This would require that the us er had prior knowledge of all the fission products generated during th e fission process that contained useful transport cross section information. Inputting every single fission product of interest, in all bur n material would be a very cumbersome task. MCNPX offers the user preset fission product tie rs, which are subsets of a preset fission product array. By using preset fission product tier se ts, the cumbersome and tedious task of cataloguing ev ery fission product on an i nput file is eliminated.82 The user then can omit certain isotopes from the fission product tie r if necessary. Three tiers of fission product

PAGE 77

77 content are available. Each tier corresponds only to those isotopes that are used for particle transport in MCNPX and the nuclide densities that are reported in the following output file. In MCNPX 2.6.A the fission product array contained 167 fission products, listed in Figure 3-4, with the following fission product tiers:79 1. Tier 1. (default) Zr-93, Mo-95, Tc-99m, Ru-101, Xe-131, Xe-134, Cs-133, Cs-137, Ba138, Pr-141, Nd-143, Nd-145.9 2. Tier 2. Isotopes contained in the fission product array that are included in the released cross-section library f ile (XSDIR) for MCNPX. 3. Tier 3. All isotopes containe d in the fission product array. Figure 3-4. Original fission produc t array containing 167 fission products. Preliminary testing of a typical PWR pin cell calculation utilizing th e fission product tiers in MCNPX as compared to the benchmarked MONTEBURNS depletion code suggested that utilizing the maximum amount fission products increased the agreement between the two codes.81 For preliminary analysis purposes, co mparing MCNPX depletion and MONTEBURNS was deemed acceptable for determining ballpark estimates of appropriate functionality because MONTEBURNS was also a Monte Carlo li nked technology that has been thoroughly benchmarked.74, 83 Figure 3-5 displays the difference in neutron multiplication versus time and Figure 3-6 and Figure 3-7 displa y the difference in end-of-lif e fission product buildup when comparing the varied fission product tiers to the benchmarked MONTEBURNS method.

PAGE 78

78 Analysis with MCNPX 2.6.A suggested that in creasing the amount of fission products followed in the calculation, increased the ag reement between MCNPX and MONTEBURNS.82 With the release of ENDF/B VII.0, it was determined that every fission product containing transport cross sections should be included in the fission product array, increasing the array to account for 220 fission products.84 Each fission product tier continued fo llowed the same criteria as previously implemented. The fission products included in th e current 220 fission produc t array are listed in Figure 3-8.84 1.02 1.04 1.06 1.08 1.1 1.12 1.14 1.16 1.18 051015202530 Burnup (GWd/MTHM)keff MONTEBURNS Tier 1 w/o models Tier 2 w/o models Tier 1 w/ models Tier 2 w/ models Tier 3 w/ models Figure 3-5. Resulting time-dependent neut ron multiplication from implementing fission products with and without cross section models in MCNPX as compared to the benchmarked MONTEBURNS method. -15.00% -10.00% -5.00% 0.00% 5.00% 10.00%Zr-93 Mo-95 Tc-99 Ru-101 Xe-131 Xe-134 Cs-133 Cs-137 Ba-138 Pr-141 Nd-143 Nd-145NuclidesPercent Difference Tier 1 without Models Tier 1 with Models Figure 3-6. Percent difference of certain fission products fr om MONTEBURNS using MCNPX Tier-1 fission products with a nd without cross section models.

PAGE 79

79 -15.00% -10.00% -5.00% 0.00% 5.00% 10.00%Zr-93 Mo-95 Tc-99 Ru-101 Xe-131 Xe-134 Cs-133 Cs-137 Ba-138 Pr-141 Nd-143 Nd-145NuclidesPercent Difference Tier 2 Tier 3 Figure 3-7. Percent differen ce of certain fission products fr om MONTEBURNS using MCNPX Tier-2 and Tier-3 fission produc ts with cross-section models. Figure 3-8. Enhanced fission pr oduct array containing 220 isotopes. Isotope Generator Algorithm Storing every decay chain for every possibl e combination of isotope listed on a burn material card would be extraordinarily memory prohibitive and would gr eatly slow down the speed of the code. To limit memory consumpti on and computational cost, only the immediate daughter products from isotopes specified within selected burn material s are tracked in MCNPX for particle transport.85 These immediate daughter reacti ons are determined by the Isotope

PAGE 80

80 Generator Algorithm. The algorithm places all possi ble daughter products, from either reaction or decay, within a burn material card at an atom density of 1E-36. Figure 3-9. Nuclides created from a decay or nuc lear reaction process and generated for particle transport by the isotope generator algorithm.3 Since the isotopes tracked within the MCNPX steady-state must be determined at the initiation of the entire calculation, specifying th e isotope at a super low atom density secures the inclusion of the isotope in the calculation whil e not incurring an artificial reactivity penalty. Figure 3-9 displays the isotope s generated for a nuclide with Z > 4. The location of each generated isotope corresponds to the position of that isotope as it would a ppear in the Chart of the Nuclides.3 Each box contains information regardi ng the interaction or decay process that leads to the creation of that isotope. Blank boxes represent possible decay chain isotopes resulting from multi-interaction processes or from ot her particle interactions (this is a result of the possible future application of utilizing the code to look at high energy partic le interactions in accelerators and solar flares).

PAGE 81

81 Manual Time-Dependent Isotope Concentration Changes To model real life depletion systems, isotopes may have to added or extracted in order to meet the conditions of the operating strategies. Fo r example, in a PWR, boron must be diluted as fuel is depleted in order to maintain criticality. For another example, in a gaseous fuel or molten salt reactor fission products may need to be extracted continuously in order to maintain system criticality. Certain depletion codes only allow the user to mani pulate concentrations of isotopes used to maintain criticality in LWRs such as diluting boron concentration. MCNPX is a general purpose modeling code used to model all types of reactors. For the modeling capability to truly be robust, the user must be able to manually manipulate the concentration of any nuclide during the execution of the depletion pr ocess in order to properly mode l the true system behavior. MCNPX depletion process allows for the user to manually manipulate the concentration of any isotope within any burned ma terial at any time step. The us er invokes this capability by use of the MATMOD keyword on the BURN card (input specification, input structure, cards and keywords are explained in the Input Interface section).86 The user is able to manipulate the concentration of any isotope by specifying the isotopes atom density, weight density, atom fraction or weight fraction within the burn material volume. The user is also able to able to specify how the concentration wi ll change during the time-depe ndant cross section averaging process. The capability resides within the bur n data retrieval subrou tine (burn_mat.F) where new atom fractions and atom densities are set for use in the next particle transport step. The process of manually setting concentrations for isotopes therefore exists post-executing the CINDER90 number density calculati on. For particle transport, the MCNPX code requires that isotope concentrations be ultimately be converted to atom fractions to be applied to total material atom densities.

PAGE 82

82 The process for calculating the renormalized atom fractions if the user specifies an isotope to be altered by atom density is as follows:86 1. A loop is generated that loops through ever y isotope used in particle transport. 2. If an isotope is not to be manually altere d that isotopes atom density is summed in the s2 variable. 3. If the isotope is to be manually altered that isotopes atom density is summed in the cf variable. 4. The total material atom density is then calculated by cf sa 2 (3-1) 5. To calculate the individual isotope at om fractions, another loop over all the isotopes is generated, and the individual isotope atom fractions are calculated by a k a kAF where k k a a, (3-2) The process for calculating the renormalized atom fr actions if the user specifies an isotope to be altered by weight density is similar to the prev iously mentioned proce ss except the code first immediately converts the inputted we ight density to an atom density and then the code continues along the same solution path as the previously me ntioned process. If the user specifies an isotope to be altered by atom fractio n, the following process is utilized: 86 1. A loop is generated that loops through ever y isotope used in particle transport. 2. If an isotope is not to be altered that isotopes atom density is summed in the s2 variable. 3. If the isotope is to be altered that is otopes input atom fr action is immediately saved as well as also summed into the fn variable. 4. The total material atom density is then calculated by fn sa 1 2 (3-3) 5. To calculate the individual isotope atom fractions for an isotope not manually altered, another loop over all the isotopes is generated:

PAGE 83

83 a. If a manually altered isotope is enco untered, then the manually given atom fraction is implemented and no calculation is required. b. For the rest of the isotopes the i ndividual isotope atom fractions are calculated by 2 ) 1 ( *,s fn AFk a k (3-4) If the user specifies an isotope to be altered by weight fraction, the following process is similar to the process that implements the altered atom fraction: 86 1. A loop is generated that loops through ever y isotope used in particle transport. 2. For an isotope that is to be altered, th e following quantities are tracked (sums for this process include only isotopes that are to be altered): a. The individual isotope weight percent, wj, divided by the individual isotope mass, Mj, stored in the xj variable. b. The sum of xj is stored in the wf variable. c. The sum of the wjs are also tracked and stored in fn variable. 3. If an isotope is not to be altered the following quantitie s are tracked (sums for this process include only isotopes th at are not to be altered): a. The sum of the individual isotope masses multiplied by the individual isotope atom densities, i i a iM,*, are stored in the s3 variable. b. The sum of the individual isotope atom densities are then stored in the s2 variable. 4. Another loop over all the isotopes is generated. 5. The following quantity is calculated for each that isotope not to be altered (sums encompass only isotopes not to be altered ): a. i i i a i a i a jM fn s fn x ) 1 ( 3 ) 1 ( *, is stored, where 1-fn represents the weight percent and Mi represents the individual atomic mass for each of the unaltered isotope. b. The sum of the xj variable for unaltered isotopes is then stored in the s4 variable.

PAGE 84

84 6. Another loop over all isotopes is generated 7. The following derivation is used in order to generate the equation for calculating the individual atom fractions for each isotope i i j jw w % % 1 (3-5a) i A i aM N w % (3-5b) ki i A i i i i a j j i A i i i i a ji i i i a i i i i a j j i i i a i i i a jN w w M w N w w M M w M w M w wf s x % % % % % % % % 4, , (3-5c) i i i k k k i i k i i A i i i a k k i i A i i i a jM w M w M w w N w M w w N w wf s x % % % % % % % % 4, (3-5d) k k k i i i k k kM w M w M w % % % (3-5e) k a k a k A k k A k kAF N M w N M w ,* % % (3-6f) This capability enables ma nual changing of nuclide concentrations during burnup calculations. For example, boron concentrations are usually specified in parts per million weight concentration while actinide extraction for fuel r ecycle may be specified in fractions per atom.

PAGE 85

85 The versatility of this capability allows the us er flexibility in modeli ng most types of manual isotope changes that exist unde r real operating strategies. Metastable Isotope Tracking The MCNPX data library requires that isotopes be listed in and referenced by a ZAID format. The ZAID format consists of the following: Z = atomic number A = atomic mass ID = identifier corresponding to the evaluation type (i.e. specific library set, temperature, particle type, etc.) The following equation is used for calculating the ZAID: ZAID = (Z*1000 +A).ID (3-7) CINDER90 uses the AZS format for listing and refe rencing isotopes. The AZS format consists of the following: Z = atomic number A = atomic mass S = isomeric state The AZS format is similar to the ZAID format and calculated using the following equation: AZS = S+10+10*z+10000*A (3-8) Since both formats are required in order to execute both codes, an algorithm is necessary in order to swap formats when accounting for isotopes in each code. Unfortunately, since there did not initially exist a convention for representing isomeric states in ZAID format, transitioning metastable isotope between each code was not tr ivial. In the past, users created their own naming convention for ZAID format metastable is otopes, and as long as a tabular cross section file was referenced in the XSDIR library file for these types of isotopes, any naming convention could be accepted (i.e. To represent Ag -110m, ZA=47210, Z=47 and A=110+100*(metastable state)). However, because MC NPX is now linked to the CINDER90 code which uses a different

PAGE 86

86 naming convention than MCNPX, an MCNPX st andard convention for naming metastable isotopes is now required. Metastable isotope s ZA are represented by using the following equation: A' = (A+300) + (m*100) where m=0,1,2,3,4 (3-9) For example: Ag-110m = 47510 47510 = 47110 + 1st metastable (3-10) MCNPX uses the following standard naming conve ntion to convert ZAID format to AZS for user in CINDER90, and the conversion ta kes place within th e interface subroutine.87 Calculation of Depletion Equation Coefficients There exists the probability for hundreds of diffe rent types of particle reactions within a typical nuclear system. Examini ng equations 2-1, 2-2a and 2-2b, we can see that increasing the amount of reaction rates tracked per isotope greatl y increases the size of the depletion equations for more reaction types contribute to the calculation of the creati on and destruction coefficients. Due to the high energy threshold for most of these particle reactions, for the majority of reactor applications, only a base set of reaction actually have a significant probability of contributing to the calculation of the destruc tion and creation coefficients. Therefore choosing to follow reactions that do not contribute to the calculation restricts the available memory and diminishes computational performance. For the computed reaction rates, the Monte Carlo steady-state calculation computes normalized fluxes and reacti on rates per source neutron. The depletion equation requires a total magnitude value for the destruction and creation coefficients. Because the Monte Carlo calculation only calculates thes e quantities relative to the particle source strength, normalization parameters must be furthe r calculated in order to convert the normalized coefficients into coefficients that are set at th e exact magnitude of the analyzed problem. After the normalization parameters have been calculate d for each isotope, the code must be able to

PAGE 87

87 properly determine how fission prod ucts are yielded from the fission process. The culmination of all of these mentioned procedures compos e the important considerations necessary for calculating adequate depletion coefficients for ac hieving an acceptable solution to the temporal nuclide inventory. Important Reactions Followed Each reaction rate tracked per isotope represents memory allocation that further restricts other aspects of an analyzed problem. CINDER9 0 does contain a 63-group cross section set for all reactions available in the libr ary file. Therefore, CINDER90 theoretically only requires a 63group flux to match to its 63-group cross section set in order to calculate depletion coefficients. The accuracy of this matching process is therefor e highly dependent upon the applicability of the cross section sets to a specific design application as well as being limited by the same approximations applied to typical deterministic ca lculations. Therefore calculating a continuous energy integrated reaction rate for use in a de pletion coefficient for the depletion equation solution better preserves the true system phys ics represented within the coefficient. Unfortunately, the trade-off in tracking continuo us energy integrated coefficients within the Monte Carlo calculation is that each coefficient tracked must be tracked for each isotope which then limits the available memory for other as pects of the calculation. Initially the MCNPX depletion code was set to only track the cont inuous energy integrated (n,fission) and (n, ) reaction rates for applicati on in the depletion solution.17 Previous analysis suggested that the (n, ) reaction rate was the dominant contributor to the neutron destructio n and therefore other capture reaction rates were tracked only by matching a 63-group flux in MCNPX to a 63-group cross section set in CINDER90 in order minimize necessary memory requirements. The code therefore applied a mix and match procedure in which only the dominant contributors to the coefficients were calculated as continuous energy integrated coeffi cients while the rest of the

PAGE 88

88 reactions were calculat ed by matching a 63-group flux calcula ted within the steady-state Monte Carlo to a 63-group cross section set in CI NDER90. Recoverable energy per fission and neutrons generated per fission were also tracked and calculated in cont inuous energy; however, threshold reactions such as (n,3n) and (n,2n) a nd residual minor capture re actions such as (n,p) and (n, ) were not tracked utilizi ng continuous energy cross sect ions. Individual threshold reaction rates were implemented by the 63-group ma tching approach mentioned previously. The concern of the 63-group matching approach was that the validity of this te chnique resided in the applicability of the 63-group crosssection set to various designs. The 63-group cross section se t within the CINDER90 code was generated by collapsing a finer cross section set over an initial arbitrary sy stem flux. The arbitrary fl ux utilized to generate the collapsed 63-group cross secti on set within CINDER90 generate d cross sections that were reliable for calculations only representative of that system flux. Therefore calculations implementing the 63-group cross section set within CINDER90 would be in error if the flux of the examined system differed from the flux u tilized to generate the cross section set. Furthermore, if the cross sections themselves were improperly collapsed to be poorly representative of thre shold reaction boundaries, the calculati on would be further in error. Therefore, MCNPX was augmented to track thes e threshold reactions properly in continuous energy to eliminate the error in calculation from the multi-group approach.79, 85 A numerical benchmark calculation was execu ted in order to is olate whether MCNPX burnup should track continuous energy reaction rates for (n,2n), (n,3n), (n,p) and (n, ). A hypothetical infinitely reflected pin cell ge ometry was depleted using MCNPX 2.6.A implementing only (n, ) and (n,fission) continuous energy reaction rates (OLD MCNPX method) and then compared to MCNPX 2.6.A implementing (n, ), (n,fission), (n,2n), (n,3n), (n,p) and

PAGE 89

89 (n, ) continuous energy reaction ra tes (NEW MCNPX method) a nd MONBTEBURNS. Both of the MCNPX versions used for this test did not employ a time-dependent cross section averaging technique such as the method employed in M ONTEBURNS. MONTEBURNS was chosen as a comparator code because of the extensive accep table benchmarking already completed with this depletion code.74, 83 It is true that the methods of is otope tracking and reac tion rate calculation differ slightly between MCNP X and MONTEBURNS; however, th e base solutions should be similar since MONTEBURNS has b een extensively benchmarked. The objective of these comparisons was to de termine the magnitude in improvement as a result of employing continuous energy reaction ra tes for minor capture and threshold reaction events. The analyzed pin cell geomet ry is displayed in Figure 3-10. Figure 3-10. Examined infinitely reflected pin cell geometry. Table 3-1. MOX fuel composition. ZA value Atom Density [atoms/cm3] 8016 4.5854e-2 92235 1.4456e-4 92238 1.9939e-2 94238 1.1467e-4 94239 1.0285e-3 94240 7.9657e-4 94241 3.3997e-4 94242 5.6388e-4 The pin cell geometry consisted of 3 regions: a cylindrical mi xed oxide (MOX) fuel region, a zirconium clad region, and a bor ated water region. The MOX cyli ndrical fuel dimensions were 365 cm tall and 0.4095 cm in radius, and the clad was 0.0655 cm thick and composed entirely of Borated Water Clad Fuel

PAGE 90

90 natural zirconium. The fuel composition is listed in Table 3-1. The entire fuel pin was encased within a borated water rectangl e 365 cm tall with a 1.313-cm length and 1.313-cm width. The borated water composition is listed in Table 3-2. Table 3-2. Borated water composition. ZA value Atom Density [atoms/cm3] 1001.60c 4.7716e-2 8016.60c 2.3858e-2 5010.60c 3.6346e-6 5011.60c 1.6226e-5 The fuel pin was depleted at a power of 66. 956 kWt. This power was approximated to be the individual pin power for a typical 3670 MWth pressurized water reactor assuming ~193 17 X 17 fuel assemblies. The geometry was burned for a total of 730.645 days (2 years). The entire burn time was separated into 6 depletion step s: 0.645 days, 40 days, 100 days, 140 days, 200 days, and 250 days, for a total of 730.654 days. Each steady-state Monte Carlo calculation used 5000 particles per cycle, skipping the first 5 cycles, for 300 cycles ENDF/BVI.0 cross sections were used for each steady-state calculation. For these calculati ons, MONTEBURNS used MCNPX 2.5.0 linked to ORIGEN2. Though the depletion steps were unrealistic for achieving an acceptable depletion solution due to the expe cted significant change in the flux shape and magnitude over such course depletion steps, th e comparison of the methods was hypothesized to provide similar results regardless of the impl ementation as long as the consistency in the approach was preserved. Sin ce both methods employed a linked calculation preserving spatial dependence of the solution, incu rring the exact same time step s, both methods were deemed acceptably consistent enough to warrant fair comparison. Preliminary results comparing the OL D MCNPX method with the MONTEBURNS depletion code displayed discrepancies in cert ain actinide and fission pr oduct buildup. Figure 3-

PAGE 91

91 11 displays the EOL (end of life) percent diffe rence of certain actinides between the OLD MCNPX method and the MONTEBUR NS method. The speculated reason for the discrepancies were lack of calculating continuous energy th reshold reaction rates (reasons of secondary concern also included limits of the matrix exponential method in ORIGEN2; however, these effects were hypothesized to be of second order effect). -120.00% -100.00% -80.00% -60.00% -40.00% -20.00% 0.00% 20.00%U-235 U-236 U-237 U-238 U-239 Np-237 Np-239 Pu-237 Pu-238 Pu-239 Pu-240 Pu-241NuclidesPercent Difference Figure 3-11. Percent difference of certain refe rence actinides between the OLD MCNPX method and the MONTEBURNS method. The NEW MCNPX method calculate d continuous energy integr al reaction rates for (n, fission), (n, ), (n,2n), (n,3n), (n,p), and (n, ). These integral reaction rates were calculated during the Monte Carlo process and fed to CINDER90 for isotope depletion. This process varied from the OLD MCNPX method in that the majority of significant threshold re action rates were now calculated in continuous energy and no longer required a 63-group Monte Carlo flux to be matched to a generalized 63-group CINER90 cross-section set. The integral continuous energy reaction rates were calculated separately for each burn material and tallied at the site of collision. Figure 3-12 displays the improvement in reference actinide calculation. A drastic improvement in actinide generation is realized when using continuous energy reaction rates for threshold re actions. The percent difference in Np-237 drops

PAGE 92

92 from -70.24% to -4.85%, and the percent differe nce in Pu-237 production drops from -95.69% to -1.02%. Due to accentuation of U-237, Np-237, and Pu-237 differences in Figure 3-11, Figure 3-12 does not capture the improvements in EO L calculation of other important actinides; however, by utilizing the continuous energy react ion rates for threshold and minor capture reactions, 1-2% improvements in actinide predic tions are also witne ssed in U-236, U-239 and Pu-238. -6.00% -5.00% -4.00% -3.00% -2.00% -1.00% 0.00% 1.00% 2.00%U-235 U-236 U-237 U-238 U-239 Np-237 Np-239 Pu-237 Pu-238 Pu-239 Pu-240 Pu-241NuclidesPercent Difference Figure 3-12. Percent difference of certain reference actinides between the NEW MCNPX method and the MONTEBURNS method. -10.00% -8.00% -6.00% -4.00% -2.00% 0.00% 2.00%Zr-93 Mo-95 Tc-99 Ru-101 Xe-131 Xe-134 Cs-133 Cs-137 Ba-138 Pr-141 Nd-143 Nd-145NuclidesPercent Difference Figure 3-13. Percent difference of certain re ference fission products between the NEW MCNPX method and the MONTEBURNS method.

PAGE 93

93 Though tracking continuous energy threshold and minor capture reactions enhanced actinide prediction capabilities, li ttle improvement was witnessed for fission product prediction. The NEW and OLD MCNPX methods for reaction ra te tracking witnessed similar discrepancies in fission product buildup. Figure 3-13 displays the difference in EOL fission product prediction when comparing the new MCNPX method to MO NTEBURNS. Unfortunately, time-dependent fission product buildup is a major contributor to negative reactivity trajec tories, and therefore MCNPX and MONTEBURNS only pr edicted slightly different time-dependent reactivity. Figure 3-14 displays the difference in timedependent neutron multiplication between the MCNPX and MONTEBURNS methods. Two major possible reasons existed for the differences in fission product generation between both MCNP X methods and the MONTEBURNS method: fission yield discrepancies and lack of cross secti on averaging technique. CINDER90 offers the latest fission yield data available.61 This data includes fission yield information for 35 fissionable isotopes incurring th ermal, fast or high energy neutron reactions. ORIGEN2 contains fewer fission yield data a nd makes approximations for yield data for fissionable actinides not containe d in an ORIGEN library file. Furthermore, not all fission product transmutation paths are followed exp licitly in the matrix exponential method in ORIGEN2.60 Therefore, it was deemed plausible that the fission yiel ds and subsequent transmutation paths used in CINDER 90 and ORIGEN2 differed slightly. The versions of the MCNPX methods examined for this specific study did not implement any type of time-dependent cr oss section averaging technique. Figure 3-14 displays evidence of this hypothesis. The eigenvalue, keff, is tracked almost exactly during the short time steps at the beginning of life (BOL) but differs in the longer time steps. Therefore, a predictor corrector methodology was suggested to reduce the differen ce in isotope transmutation trajectory and

PAGE 94

94 eigenvalue prediction for the two codes (this c oncept is explained further in the Predictor Corrector Method section). However, becau se actinide production was improved by tracking continuous energy threshold and minor capture reac tion rates, these reactio n rates were chosen for inclusion in tracking for fu rther versions of the code.85 1.020 1.040 1.060 1.080 1.100 1.120 1.140 1.160 1.180 051015202530 Burnup (GWd/MTHM)keff MONTEBURNS MCNPX Old MCNPX New Figure 3-14. Time-dependent eigenvalue calculations. Flux Normalization MCNPX calculates reaction rates on a per s ource basis; however, CINDER90 requires a total magnitude reaction rate to deplete the system appropriately. In order to generate a total magnitude reaction rate, MCNPX must determine the magnitude of the source and multiply this value by the tracked reaction rates per source to calculate the total magnitude of the reaction rates for use in the depletion solution. The source of neutrons in a multiplying system is the total neutron production rate of that system. The total magnitude of the flux may then be calculated by multiplying the flux per source by the produc tion rate as in equation 3-11: value MCNPX totalQ P Multiplier Flux Multiplier Flux (3-11) = neutrons per fission event Qvalue = recoverable energy per fission event P = power level

PAGE 95

95 How these parameters are calculated and what re gions of the calculation these parameters are applied complicate the implementation of these parameters as normalization constants. At each interaction site within a material that contains an actinide, each reaction mentioned in the prior section is tallied along with: (n,f ) = neutrons per fission event multiplied by the macroscopic fission cross section for that ma terial per source neutron. (n,f valueQ *) = recoverable energy per fission event multiplied the macroscopic fission cross section for that ma terial per source neutron. Calculating neutrons per fissi on and recoverable energy per fi ssion then involves dividing the above two values by the macroscopic cross sectio n. For a complex system of many different fissionable actinides, in many different burnabl e regions, the major question, regarding the calculation of the Qvalue and which now arises, is how to represent the normalization parameters and apply them to the tallied r eaction rates. Reaction rates and normalization parameters are tallied within each burn materi al. If for every specific burnable region of a system, the interaction rates were normalized explicitly by these calculated parameters within that burnable region, it would not be possible to accurately transmute regions that did not contain fissionable actinides. If the region did not contain fissionable actinides, the Qvalue for that specific region would not exist forcing the normalization multip lier to be undefined. In order to normalize the reaction rates, th e code requires knowledge of the source of neutrons. The source of neutrons is actually a global quantity because neutrons born within one burnable region migrate to other burnable regions for reaction. For example, consider a geometry containing a single cylinder contained within a rectangular fluid of which posses a reflective boundary condition at the boundary of the rectangle. At the center of the cylinder, 100 source histories are started and transported. If 60% of the histories are captured within the

PAGE 96

96 cylinder and 40% leak out and interact in the su rrounding material, then 60 neutrons are captured within that cylinder or 0.6 neutrons per source are absorbed. Now let us assume that we now have two cylinders of same dimension as the prio r cylinder and separated in a manner so that the pitch between each cylinder mimics the pseudo pitch implied from the prior calculation. A rectangular fluid encases both cy linders and that fluid also pos ses a reflective boundary condition at the boundary of the rectangle. For this new ca lculation we will still us e 100 histories, but start 50 histories in each of the two spheres. Ignori ng effects of statistical randomness of particle travel, if the each cylinder employs the same materi al as the prior example then we would expect that 60% of the neutrons emitted within each cyli nder are then captured by each cylinder. Since 50 neutrons are emitted in each cylinder, each cyli nder must then capture 30 neutrons. In the first example, some neutrons were emitted from the cylinder into the fluid, hit the fluid boundary, and then were scattered b ack into the cylinder for interact ion. In the second example, the same process happened as well as having neutrons emitted from one cylinder interacting with the other cylinder. Because the source sites of neutrons for neutron interaction within each cylinder may have been located with in either cylinder, the neutron s ource is said to be global to the entire geometry. If we now apply the same thinking to our calculation of flux normalization parameters, these normalization parameters must al so be tallied globally. This implementation forces the normalization multiplier to be constant and representative of the entire system. For each specified contributing mate rial, MCNPX first tracks (n,f ) and (n,f valueQ *) within each specified contributing materi al, divides by the macroscopic fi ssion cross section within each material, and then volumetrically averages the quant ities to generate globally averaged values for neutrons per fission and recoverabl e energy per fission. These globa lly averaged values are then combined to create the flux normalization multiplier in equation 3-11. The normalization

PAGE 97

97 multiplier is then said to be the global production rate or source and multiplied by each reaction rate per source to generate the total ma gnitude of the reaction rates used as destruction and creation coefficients in the depletion solution. Therefore, it no longer maters if the material that this normalization factor is applied cont ains actinides or not, for the same averaged normalization constant is applied to reaction rate for each material. The normalization parameters are recalculated and averaged at the end of each time step, so that the multiplier dynamically adjusts to changes in temporal parameters as actinide concentrations are created/depleted.79 Table 3-3. Emitted and recovera ble energy per fission of U-235. Form Emitted Energy (MeV) Recoverable Energy (MeV) Fission Fragments 168 168 Fission Product Decay Gamma-rays 8 8 Beta-rays 7 7 Neutrinos 12 -Prompt Gamma Rays 7 7 Fission Neutrons (Kinetic Energy) 5 5 Capture Gamma Rays -3-12 Total 207 198-207 The Qvalue used in the calculation is an estimate of the total recoverable energy per fission event. The total recoverable energy per fission may be described as the sum of many mirco events the lead to the recover of energy from a fission event. Table 3-3 lists the emitted and recoverable energy per fission from each micro fission event for U-235.7 MCNPX inherently contains an array possessing prom pt energy per fission for certain fissionable actin ides. Prompt fission release energy is available within the eval uated data libraries, such as ENDF/B and JEFF, for roughly 1/3 of all the actinides In the ENDF/B based libraries the breakdown of the data for prompt fission energy release ex ists within File 1 MT 458.70 The breakdown of prompt fission

PAGE 98

98 release data includes: prompt kinetic fission energy (EFR), prompt neutron emission energy (ENP), prompt gamma energy (EGP), other dela yed energy and neutrino energy contributions.70 Each actinide available within th e prompt fission energy release ar ray in MCNPX, as well as the breakdown of prompt recoverable energy per fission if available, is listed in Table 3-4. The sum of EFR, ENP and EGP for each actinide is the prom pt fission release energy for that actinide. Table 3-4. Prompt fission energy re lease data available in MCNPX. Isotope EFR ENP EGP Total 90232 160.39000 4.41000 7.11000 171.91000 91233 163.50000 5.04000 7.03000 175.57000 92233 168.22000 4.90000 7.72000 180.84000 92234 167.10000 4.85000 7.50000 179.45000 92235 169.13000 4.91600 6.60000 180.64600 92236 167.50000 4.70000 7.30000 179.50000 92237 180.40000 92238 169.80000 4.80400 6.68000 181.28400 92239 180.40000 92240 169.57000 5.21000 6.53000 181.31000 93237 170.60000 5.94000 7.13000 183.67000 94238 173.60000 5.92000 7.13000 186.65000 94239 175.55000 6.07000 6.74100 188.36100 94240 173.70000 6.48000 6.18000 186.36000 94241 175.36000 5.99000 7.64000 188.99000 94242 174.00000 6.76000 5.22000 185.98000 94243 187.48000 95241 176.40000 6.53000 7.90000 190.83000 95242m 182.37000 4.05000 1.17210 187.59210 95243 176.30000 7.53000 6.42000 190.25000 96242 186.22000 5.20000 0.00000 191.42000 96244 178.50000 7.62000 4.37000 190.49000 Values listed are in MeV. To properly account for all the en ergy per fission event the inherent prompt values must be adjusted in order to account for delayed energy as well. A constant normalization factor, 1.111, is multiplied by every Q-value calc ulated in MCNPX in order to account for delayed energy per fission event. This constant normalization fact or 1.111 was estimated based on the idea that if multiplied by the prompt energy per fission of a U-235 fission event, the total amount of

PAGE 99

99 recoverable fission, delayed a nd prompt, would equal 200 MeV.79 By implementing a constant normalization factor, an approximation is made about the value of the capture gamma energy contributing to the total recove rable energy per fission. In a large LWR, most of the capture gamma energy is deposited in the heated volume and thus contributes to recoverable energy; however, in a long, thin space reactor a lot of the capture gamma energy leaks out of the system and does not contribute to rec overable energy of the system. Therefore, the capture gamma energy contribution is highly dependent upon system geometry and therefore the normalization consta nt may or may not estim ate the actual Q value properly. Underestimating the Q value leads to an overestimation of the system fluxes and thus, an overestimation of the system collision densities. Because the equation for nuclide depletion is highly dependent on the collision densities as disp layed in equation 3-12, increasing the collision density leads to an overestim ation of the material burnup. ) ( ) ( ) ( ) ( t r N t r t r dt t r dNi i E i i (3-12) In the current depletion capabi lity, three methods are used to correct this overestimation (input structure will be later explained in the Input Interface section): 1. Adjust the Q value multiplier (B1 value on the BOPT keyword; default = 1.0). 2. Modify the POWER keyword. 3. Alter the system power on the PFRAC keyword. Unfortunately, the capture gamma contribution to the recoverable en ergy per fission is burnup dependent. The following equati on displays the how the captu re gamma energy contribution should be applied to the cal culation of total recoverable energy per fission event: neutrino capture eff delayed prompt erable reQ Q k E Q Q Q ) (cov (3-13) Since neutrons per fission vary as a function of burnup, assuming a critical system is maintained in the calculation, recoverable energy per fission event should also vary as a function of burnup.

PAGE 100

100 In ENDF/B VII.0 only 207 out of 390 isotopes contain capture gamma release data. In some cases, the data contain complete spectra in formation; however, in many cases the data are incomplete.80 Without capture gamma photon emission spectra, employing equation 3-13 is impossible because there is no way of determ ining how and where the emission energy is deposited. The integral energy emission may be calculated from a mass defect calculation. However, transporting a photon possessing the integr al energy from the mass defect calculation would not truly represent the physics of the syst em because the photon would travel farther and deposit energy in different regions of the geometry as compared to the photons generated from the true emission spectra. Nonetheless an in itial calculation was completed to determine the average integral capture gamma energy of all isotopes contained within the CINDER90 library file. A mass defect calculation assuming zer o energy neutron capture was completed to determine the integral energy from a neutron ca pture event. The atomic mass data was taken from the Atomic Mass Data Center (AMDC) website.88 The average integral energy from a capture reaction with isotopes available in the CI NDER library file was calculated to be 7.9 MeV per capture. For fission products containing an MCNPX cross secti on and CINDER90 fission yield information, the average integral energy fr om neutron capture was calculated to be 6.48 MeV. These calculated values only represent integral quantities based on 0 energy neutron capture. Trying to track the true capture gamma energy deposition was deemed futile in comparison to applying a delayed energy normali zation constant because the true gamma ray emission spectra were unknown. The integral energy deposition from the mass defect could be assumed to be deposited locally; however, this assumption would violate th e true physics of the system. The implementation of this technique could theoretically at best only improve the calculation of recoverable ener gy per fission by a few percent a nd at worst result in a more

PAGE 101

101 inaccurate representation of the recoverable ener gy per fission as compared to the normalization constant. The normalization factor was thus chosen as the method for approximating the recoverable energy per fission, and the following recommendati ons are suggested for trying to better account for capture gamma energy contribution: 1. Photon emission spectra must be generate d for all isotopes used in particle transport and these photons should be tr ansported to determine where each photon deposits its energy, and then equation 3-13 may be used to adjust recoverable energy per fission. 2. If the integral mass defect energy is to be employed for determining capture gamma energy contribution, a qualified approxim ation must be derived for determining how that energy is smeared within the ge ometry and what percent of that energy escapes the heated volume contributing to system power. However, since the normalization factor was chos en for implementation, the current recoverable energy approximation will continue to slightly affect the flux normalization constant. Automatic Fission Yield Selection Fission products are generated as actinides consume neutrons and undergo fission events. The fission product yield distributio ns from the fission events are a function of the isotope that the causes fission event as well as the impinging neutron energy at the in teraction site. For thermal neutron interactions the fission product yi eld distributions follows the curves witnessed in Figure 3-2 and Figure 3-3. For higher energy reactions the fission yield di stribution curve flattens and isotopes that did not have a high probability of formati on from a thermal reaction now have a higher probability of formation in fast fission events. The differences in the fission products created leads to adverse reactivity condi tions; therefore accounting for the proper fission products that are generated during a fission event is vital for calculation of r eactivity and further interaction

PAGE 102

102 rate distributions. Currently, the ENDF/B files offer a thermal, fast and high energy fission yield distribution for various fissionable actinides.80 Table 3-5. Available actinide fission yield distributions in CINDER90. Isotope Thermal Fast High Energy Spontaneous Fission 227Th x 229Th x 232Th x x 231Pa x 232U x 233U x x x 234U x x 235U x x x 236U x x 237U x 238U x x x 237Np x x x 238Np x 238Pu x 239Pu x x x 240Pu x x x 241Pu x x 242Pu x x x 241Am x x x 242mAm x 243Am x 242Cm x 243Cm x x 244Cm x x 245Cm x 246Cm x x 248Cm x x 249Cf x 250Cf x 251Cf x 252Cf x 253Es x 254Es x 254Fm x 255Fm x 256Fm x The CINDER90 code offers a thermal, fast, high-energy fission yield for each fissile isotope contained in the CINDER90 data file.61 Table 3-5 displays the actinide containing

PAGE 103

103 fission yield distributions, within certain energy bands utilized in CINDER90.61 Since the CINDER90 code is an energy integr ated depletion code, and is inca pable of determining spectral characteristics, knowledge of the spectral charac teristics must be input by the user. Because proper fission yield selection involves predet ermined knowledge of the neutron spectrum and because this predetermined knowledge is possibl y not known before the system is analyzed, a preliminary calculation may be necessary to determine spectrum information. This preliminary calculation contributes to the co mputational cost associated w ith the depletion process. Selecting the adequate fission yield for a gi ven system is dependent upon selecting the specific yield that best represents the energy band containing the majority of the systems fissions. This selection proce ss therefore depends upon the energydependent fission rates. The MCNPX depletion code incorporates an algorith m to determine which energy band contains the largest fission rate and then properl y selects the correct fission yield.85 The MCNPX automatic fission yield selection method is diagramed in Figure 3-15. MCNPX determines which fission yield to use for a specified problem by calculating the integral fission rate for the defined energy boundaries (thermal, fast, and high energy), determining which energy range contains the majority of fissions, and then selecting the appropriate fission yield corresponding to the energy range containing the majority of fissions. The me thod is versatile in that each individual burn material may implement a separate fission yield if the calculation warrants a separate fission yield for different geometric locations. By usi ng this process, the end user does not have to endure the computational cost a ssociated with a preliminary cal culation to determine the proper fission yield to be used for the depletion process. The end user is also given confidence that the fission yield being implemented fo r a specific problem is in fact the fission yield corresponding to the energy-dependent fission rate of the analyzed system.85

PAGE 104

104 Steady State System Tracking Collision Which Energy Band? Thermal, Fast or High Energy Tally Fission Rate within Energy Band: X = X + ?SfF Is System Tracking Complete? Utilize that Fission Yield for Depletion Determine which Energy Band Contains Largest Integral Fission Rate and Select Proper Fission Yield Is this the Final Depletion Step? Finished Calculation Yes Yes No No Steady State System Tracking Collision Which Energy Band? Thermal, Fast or High Energy Tally Fission Rate within Energy Band: X = X + ?SfF Is System Tracking Complete? Utilize that Fission Yield for Depletion Determine which Energy Band Contains Largest Integral Fission Rate and Select Proper Fission Yield Is this the Final Depletion Step? Finished Calculation Yes Yes No No Figure 3-15. Fission yield se lection process diagram Time Dependent Cross Section Averaging Techniques The depletion equation use interaction rates and nuclide densities to determine the timedependent nuclide inventory. Reexamining equation 2-1 we see that the depletion equation is only linear if the depletion coeffici ents are constant; however, in re ality the depletion coefficients are also time-dependent. As the nuclide invent ory changes over time, as a function of the reaction rates, the interaction rates will also chan ge as a function of time thus making the process of nuclide depletion nonlinear. As discussed in Chapter 2, to make equation 2-1 linear and solvable, the depletion coefficients are assume d constant over a given time step. Because a certain amount of fluence is required in order to change the nuclide c oncentration significantly enough to further alter the applied reaction rate s, the assumption is valid for certain time durations.2 Acceptable time durations are related to th e equilibrium buildup or depletion of highly absorbing isotopes. How the coefficients are app lied within these time durations is paramount to the accuracy of the calculation. Equation 3-14 illustrates how power is proportional to the macroscopic fission cross section multiplied by the flux: fPower ~ (3-14)

PAGE 105

105 From a first order standpoint, as the fuel burns and the macroscopic fission cross section decreases, the flux must increase in order to main tain constant power. Th erefore if the depletion coefficients calculated at t0(initial time step) are util ized for a depletion to ta(subsequent time step), the interaction rates implemented into the nuclide inventory calculation will be underestimated in comparison to the average inte raction rates; ultimate ly, leading to an under burning of the system. Making the depletion coeffici ents time-dependent in order to account for the true interaction rate behavior makes the depletion equations nonlinear. Therefore if the coefficients must be made constant in order to make the depletion equation solvable, then the average interaction rates must be applied in order to attempt to adequately account for the true interaction behavior. Unfortunately, the true average interaction rates for a given time step cannot be known exactly, for that would involve solving the nonlinear equations exactly which is impossible because the equations are nonlinear. Howeve r, a pseudo average reaction rate may be calculated by applying appropriate assumptions relate d to how the interaction rate is expected to behave. Many methods exist for approximating aver age coefficient behavior in the solution of non-linear equations;46, 74, 89-91 however, for the method to be useful in a computationally expensive linked Monte Carlo application, the me thod must implement as few extra Monte Carlo calculations as possible. One general method for approximating th is nonlinear behavior without excessive computational expense is the predictor-corrector method. 46, 74, 89, 91, 92 First, an assumption is made about the time-integrated flux (setting it to a constant value) to deplete the system over a time step and calculate a predicted value of the number de nsities, and interaction rates. Values then are corrected by depleti ng the system again over the time step implementing the newly calculated interaction rates. The hope of implementing su ch a calculation is to deplete

PAGE 106

106 the system using a best representation of th e average of the time-dependent parameters. Predictor-corrector methods are defined as any t ype of routine in which a prediction is made regarding behavior and then a furt her correction is made if that be havior is deemed to change. Two cross section averaging techniques were investigated for inclusion in MCNPX. The Cell-2 and MONTEBURNS cross sec tion averaging methods were c hosen based on the fact that both codes were highly benchmarked and th e methods incurred minimal additional computational cost when applied in Monte Carlo.74, 91 The Cell-2 method follows a more traditional predictor-corrector approach, and the MONTEBURNS method follows a slightly different approach that still relies upon the same implied assumptions of the traditional predictorcorrector approaches but requires less memory per calculation. MONTEBURNS Method CINDER90: Rerun Depletion Calculation [t(i) t(i+1)] Use recalculated collision densities /fluxes Calculate New Number Densities = Nc For next sequence of time steps N0 = Nc Final Time Step No Yes Done CINDER90: Depletion Calculation [t(i) t(i+1/2)] Calculate New Number Densities Monte Carlo: Time = t(i) Initial collision densities/fluxes Initialnumber densities ( N 0 ) Monte Carlo: Time = t(i+1/2) Recalculate collision densities/fluxes Figure 3-16. MONTEBURNS cross section averaging technique.

PAGE 107

107 The MONTEBURNS cross section averaging procedure is displayed in Figure 3-16.74 The MONTEBURNS code first makes a predictor ca lculation of the nuclide inventory to time t(i+1/2), based on interaction rates calculated at time, t(i). The interaction rates then are recalculated using the nuclide inventory at t(i+1/2). It is then assumed that these newly calculated interaction rates are th e average interaction ra tes. Then a calculation is made over the whole time step implementing the newly calcula ted average interaction rates to generate corrected number densities. The entire process repeats until the final time step. CELL-2 Method For next sequence of time steps 20C PN N N Final Time Step No Yes Done CINDER90: Depletion Calculation [t(i) t(i+1)] Calculate New Number Densities Calculated number densities become predictor number density = Np Monte Carlo: Time = t(i) Initial collision densities/fluxes Initialnumber densities= N 0 Monte Carlo: Time = t(i+1) Recalculate collision densities/fluxes CINDER90: Rerun Depletion Calculation [t(i) t(i+1)] Use recalculated collision densities /fluxes Calculate New Number Densities New number densities become corrector number density = Nc Figure 3-17. CELL-2 Predictor-Corrector method.

PAGE 108

108 The CELL-2 predictor-corrector method is displayed in Figure 3-17.91 The CELL-2 code first makes a predictor calculation of the nuclide inventory to time, t(i+ 1), based on interaction rates calculated at time, t(i). The interaction rates then are recalculated using the nuclide inventory at t(i+1), a nd the system then is redepleted from t(i) to t(i+1) to generate new corrected number densities. Finally, th e predicted and corrected numbe r densities are averaged to determine the nuclide inventory at t(i+1). The entire process repe ats until the final time step. Preliminary Analysis An infinitely reflected pin cell geometry wa s depleted at a power of 66.956 kWt for 2191 days (6 years), using 5000 particles per cycl e for 300 cycles, skipping the first 15 cycles.92 The three types of calculations examined were 1. MCNPX with no-predictor-c orrector method (NoPC); 2. MONTEBURNS predictor-corrector method (PC1); 3. CELL-2 predictor-corrector method (PC2). An initial test involved only burning for a 371 da y period. The 371 day cycle was broken up into 6 steps for testing predictor corrector methods: 0, 0.654, 31, 91, 181, 271, and 371 days. A MCNX calculation without implementing a pred ictor corrector method was run for 14 time steps: 0, 0.645, 5.7, 16, 31, 46, 61, 76, 91, 136, 181, 271, 321 and 371 days. This case was run as a control case, with the assu mption that as the amount time-steps approached infinity the calculation would approach true behavior. Neut ron multiplication versus time for the examined cases is displayed in Figure 3-18. Unfortunately, it appeared that the examined cycle time was not long enough to incur a significant difference from choosing to utilize th e predictor corrector method or not, suggesting that the 6 times steps implemented within th e 371 day cycle was adequate enough to model reaction rate behavior. A new se t of tests were generated examining longer cycle times in order

PAGE 109

109 to truly analyze the benefit of implementing th e predictor corrector method. A 2191 day burn (3 cycles of 2 year burns) was spli t up into various burn steps and analyzed to determine which of the two predictor corrector methods would behave better. NoPC was analyzed using 30, 20, 10, and 5 time steps, whereas PC1 and PC2 were examined using 15, 10, and 5 time steps. Figure 319 displays the burn steps implemented to achieve the 2191 day burn. 1.1 1.12 1.14 1.16 1.18 051015Burnup [GWD/MTU]Neutron Multiplication MONTEBURNS/ORIGEN MONTEBURNS/CINDER90 NoPC 6 steps PC1 PC2 NoPC 14 steps Figure 3-18. Time dependant neutron multiplicati on for the predictor corrector method tests. For these cases, the assumption was made that the 30 time step case was assumed to be an infinite set of time steps. Therefore each subs equent case was compared to the 30 step case. Figure 3-20 and Figure 3-21 display the time-depe ndent neutron multiplication versus burnup for the different predictor corrector cases. Both pr edictor corrector methods, PC1 and PC2, calculate similar neutron multiplications at each time-step and the differences in actinide and fission product production and depletion rates were also similar in both methods (>0.4% in EOL production/depletion between both methods). Figure 3-22 displays the percent difference in neutron multiplication of each case when compared with the NoPC 30-step case. When eith er the PC1 or PC2 5 step calculation (5 step calculation = 5 predictor steps + 5 corrector steps = 10 total steps) was employed, there was a 3X speedup and less than 1.55 % difference in the EOL keff. Therefore, for this specific geometry,

PAGE 110

110 the conclusion may be made that both the PC 1 and PC2 methods were equally acceptable techniques for reducing the needed number of time steps to achieve an acceptable EOL keff. Since this result suggested th at both methods achieve accep table answers, and since the MONTEBURNS method employs less memory, because predictor number densities were not saved during the corrector calcu lation, the MONTEBURNS method was chosen as the method of choice for implementation into MCNPX.92 Figure 3-19. Burn steps impl emented to achieve 2191 day burn 302015105 0.00E+000.00E+000.00E+000.00E+000.00E+00 2.48E-022.48E-022.48E-022.48E-022.48E-02 4.09E-014.09E-014.09E-01 1.94E+001.94E+00 7.93E-017.93E-011.18E+001.15E+01 2.11E+01 1.18E+001.18E+00 1.94E+002.11E+015.19E+01 1.56E+001.56E+006.74E+003.07E+01 8.41E+01 1.94E+001.94E+00 1.35E+014.13E+01 2.71E+004.82E+00 2.11E+015.19E+01 3.86E+007.70E+002.88E+016.24E+01 5.40E+001.15E+013.65E+017.30E+01 6.94E+001.54E+014.42E+01 8.41E+01 8.47E+00 2.11E+015.19E+01 1.00E+012.73E+015.95E+01 1.15E+013.34E+016.72E+01 1.39E+013.96E+017.49E+01 1.62E+014.57E+01 8.41E+01 1.85E+01 5.19E+01 2.11E+01 5.95E+01 2.38E+016.72E+01 2.69E+017.49E+01 3.00E+01 8.41E+01 3.34E+01 3.73E+01 4.19E+01 4.69E+01 5.19E+01 5.76E+01 6.38E+01 7.03E+01 7.72E+01 8.41E+01 (GWD/MTU) Burnup Steps

PAGE 111

111 0.93 0.96 0.99 1.02 1.05 1.08 1.11 1.14 1.17 020406080100 Burnup (GWD/MTU)Keff 26A 30 Steps PC1 5 Steps PC2 5 Steps 26A 5 steps Figure 3-20. Time dependent neutron multiplication utilizing 5 burn steps. 0.93 0.96 0.99 1.02 1.05 1.08 1.11 1.14 1.17 020406080100 Burnup (GWD/MTU)Keff 26A 30 Steps PC1 10 Steps PC2 10 Steps 26A 10 steps Figure 3-21. Time dependent neutron multiplication utilizing 10 burn steps.

PAGE 112

112 0.00% 0.50% 1.00% 1.50% 2.00% 2.50% 3.00% 3.50% 5155113502191 Time (days)Percent Difference NoPC 20 Steps NoPC 10 Steps NoPC 5 steps PC1 15 steps PC1 10 Steps PC1 5 steps PC2 15 steps PC2 10 Steps PC2 5 steps Figure 3-22. Percent Difference in keff When Compared with the NoPC 30-Step Case Further Commentary on both Cro ss Section Averaging Techniques Both methods tested were not the exact or iginal method employed within both codes. Initially, the CELL-2 method only required a full spectrum calculation for the predictor calculation.91 The method reuses the predicted numb er densities to propagate the spectrum calculation at each time step.91 Corrected number densities are calculated using a similar procedure as detailed in Figure 3-17 except that these new number densities are not implemented in further propagation of the spectra calculation.91 These corrected number densities only serve to update outputted values of what is to be beli eved the true number dens ity at that time step. Implementing the CELL-2 method in this way reduces the amount of needed spectra calculations by half the amount. As initially implemente d, the MONTEBURNS method only recalculates spectra at the half time steps, and then uses th at specta as the assumed average spectra to be implemented over the entire time step. Implem enting the MONTEBURNS method in this way also reduces the amount of needed sp ectra calculations by half the amount.74 Initially, the

PAGE 113

113 method was tested implementing the corrector sp ectra based on the premise that the corrector spectra would be required for systems incurri ng significant concentra tion change of highly absorbing isotopes. Later evidence from calcula tions with MCODE supported this premise for BWR lattices containing heavy initial gadolinium loading.75 This evidence suggested that the predicted spectra and corrected spectra differe d significantly enough to warrant a significant enough change in interaction rate to invalidate the applicabilit y of the method for the desired time step. Therefore either smaller time steps were required, therefore li miting the effectiveness of both methods for long time steps, or the correct ed number densities were required in order to accurately propagate the temporal nuclide soluti on. This evidence supports that using the corrected spectra, either from the calculated corrector number de nsities in the CELL-2 method or from the recalculated spectra at the t0 tim e steps in the MONTEBURNS methods, improves nuclide predictability. Using the corrected spectra increase the length of acceptable time step to a value beyond which would have been valid if only the predicted spectra were implemented; however, the method still is dependent on the time-st ep choice. The corrector spectra calculation does represent an extra calculation with extra incurred computational cost. For any depletion system, there may exist an optimum set of tim e steps in which the extra corrector spectra calculation employed for these time steps does no t improve computationa l cost; however, these optimum time steps are usually unknown prior to calculation. For a general purpose tool, employing the extra corrector spectra calculation gi ves the user the best change for achieving accuracy in calculation. Because the MCNP X depletion methodology paradigm was to be applied to all types of deplet ion systems, possibly incurring the limiting cases given from the literature, both methods were tested using the corrected spectra calcul ation, and the qausi MONTEBURNS predictor corrector method was chosen for implementation in MCNPX 2.6.B.94

PAGE 114

114 CHAPTER 4 IMPLEMENTATION ARCHITECTURE The depletion capability is a modular met hodology residing within a large scale MCNPX code. As mentioned before, MCNPX is composed of ~1000 subroutines. The depletion process resides in roughly 30 subroutines in the MCNPX code. Five ne w self-contained subroutines compose the majority of the actu al variable averaging, outputti ng, and interfacing necessary for the depletion methodology while th e other 25 augmented subroutines simply allow new variables to be inputted and set within the code pack age. The objective of confining the major computations to a minimal amount of self-containe d subroutines is to make the package modular enough to be portable to earlier or different versions of MCNP. Self containment within minimal subroutines also helps in code augmentation and te sting, for if the capability is confined to a small section of the larger code, then findi ng errors generated by the capability will be centralized to a minimal amount of subroutines employed by the code. If the methodology was not confined and self contained, debugging and coverage analysis woul d involve looking at larger set of subroutines. Fu rthermore, without containment to a small set of subroutines, augmentation to the methodology would result in needed coverage anal ysis for a larger subroutine set of the code increasing the comput ational cost associated with development and testing. The MCNPX code requires the user to utilize in put files in order to initialize variables for computation. The input file structure includes50 (1) a title card; (2) cell cards defining a geometry from unions and intersections of surfaces specified on th e surface cards (cards explaining certain physics options may also exist within in the cell cards section); (3) surface cards defining base surfaces that are used to de fine volumes on the cell cards; (3) data cards defining material, source and physics op tions available fo r implementation.

PAGE 115

115 For each type of card defining source and physics options, keywords are available that determine how the card is implemented. For example on th e SDEF card, the keyword PAR with a particle designator is used for determining the type of particle to be omitted from the source.50 Since the depletion capability involves many types of op tions in order to burn a geometry model, the depletion capability was setup in a similar manor as the SDEF card. The development of the input interface and output st ructure spans MCNPX 2.6.A-F.79, 84, 86, 87, 94, 95 The input and output features listed below, state the relevant capabili ties of the most recent release of MCNPX 2.6.F.87 Input Interface The depletion capability is invoked by using the BURN card. Within the BURN card exists many different types of keyword options. Figur e 4-1 displays the i nput structure for the BURN card. Each input variable is then explained below Figure 4-1. BURN TIME=T1,T2,T3,... PFRAC=F1,F2,F3,... POWER=P MAT=+/-M1,+/-M2,+/-M3,... OMIT=J1,N1,I11,I12,...,J2,N2,I21,I22,... AFMIN=A1 A2 MATVOL= V1, V2, V3, Vn MATMOD= ... BOPT=B1,B2,B3 Figure 4-1. Burn car d input structure. The TIME keyword corresponds to the incrementa l time duration for each depletion step. Each Ti listed in this keyword corresponds to a tim e duration (days) in which the system is burned. The default for this keyword is 1 day. For example: TIME = 10, 100, 15 The system is first burned for 10 days, followed by a 100-day burn, followed by a 15-day burn, for a total burn of 125 days. The PFRAC keyword corresponds to the fractiona l value of total system power (POWER) in which to burn the system for th e equivalent time duration. The PFRAC option only changes the

PAGE 116

116 total magnitude of power used to normali ze the flux, and does not account for possible temperature changes in an actual power change. For each Ti value listed on the TIME keyword, a corresponding Fi value listed on the PFRAC keyword must exist. For each Ti value that does not contain a corresponding Fi value, Fi = 0 for that time step. The exception is when the PFRAC keyword is not specified at all, and then each Fi = 1 for every time step. The POWER keyword is the total system power level, P, (in megawatts). Because this value corresponds to the total recoverable energy from the system, the value entered on this keyword should correspond to the total recoverable thermal system power. The default for this keyword is 1MW. For example: POWER = 100 In this case, the total rec overable thermal system power is assumed to be 100 MW. The materials to be burned must be listed on the MAT keyword. Each Mi entry corresponds to the material number listed in data cards section of the input deck. Figure 4-2 displays an example BURN card and subsequent materials cards referenced by the BURN card as burn materials. In this example, materials m1, m3, and m4 will be burned. Burnup is calculated for the entire system of materials listed on the MAT keyword, as well as for each individual material listed on the MAT keyword containing a fiss ile actinide. Negative material numbers signify materials that are not burned, but do contribute to the power normalization by applying the tall ied recoverable energy per fissi on and neutrons per fission of those materials to the total power normalization. Actinide and non-actinide buildup/depletion information for negative MAT numbers is not displayed in the output file because the negative MAT numbered materials are not burned.

PAGE 117

117 BURN TIME=100,70 MAT=1,3,4 POWER=1.0 PFRAC=1.0,1.0 BOPT=1.0 -12 1.0 C Material Cards m1 8016.60c 4.5854e-2 92235.60c 1.4456e-4 92238.60c 1.9939e-2 94238.60c 1.1467e-4 94239.60c 1.0285e-3 94240.60c 7.9657e-4 94241.60c 3.3997e-4 94242.60c 5.6388e-4 m2 2004 -1.0 m3 40000.60c -1.0 m4 1001.60c 4.7716e-2 8016.60c 2.3858e-2 5010.60c 3.6346e-6 5011.60c 1.6226e-5 mt4 lwtr.01t Figure 4-2. BURN card referencing burn ma terials m1, m3 and m4. The OMIT keyword allows the user to run a tr ansport calculation without including the transport data from a specified isotope. If an OMIT keyword is used, then the isotope is omitted from the transport calculation and omitted from th e output file. Reaction rates are still generated for the isotope by importing a 63-group flux into CINDER90, matching the 63-group flux calculated from MCNPX to a 63-group cross sect ion set inherent within CINDER90, energy integrating the reaction rates, and then producing a total collision rate for each collision type tracked. The OMIT keyword has the has the format OMIT=J1,N1,I11,I12,...,J2,N2,I21,I22, ... Table 4-1 contains the availabl e options for this keyword. Table 4-1. OMIT keyword variable tasks. Input Variable Variable Ji ith material for which to omit nuclides Ii1, Ii2, etc. Ni Number of omitted nuclides listed for the ith material Ii1, Ii2, 1st, 2nd, etc., omitted nuclide for the ith material

PAGE 118

118 If J1=-1, then the omitted nuclide list is applied to all materials and J2, J3, etc., are not allowed. There is no default for this option. The AFMIN keyword allows the user to determin e below which mass fraction an isotope will be excluded from the transport calculation. The A1 value of AFMIN keyword signifies the minimum atom fraction for all isotopes in whic h isotopes below that at om fraction will have their atom fraction set to zero. The de fault for this option is 1.0E-10. The A2 value sets the decay chain convergence criteria (mentioned in the Benchm arks section). The default for this value is now 1e-10 and was 1e-4 for MCNPX 2.6.E and ear lier. Using the new default improves nuclide predictability as the cost of increased computation time. The B1 value on the BOPT keyword corresponds to the r ecoverable energy per fission multiplier. The default for this value is 1.0. The user may choose to adjust this value in order to adjust the Q value to better account for th e true recoverable energy in the system. Table 4-2. Options invoked for different B2 values. Input Value for B2 Output Inventory Ordering 1 High to low, based on mass (default) 2 High to low, based on activity 3 High to low, based on specific activity 4 Increasing zaid The B2 value on the BOPT keyword corresponds to the selected fission products used in the MCNPX transport process, as well as the form at for the nuclide concentrations in the output file. MCNPX reports nuclide concentrations only for isotopes that are listed on the material cards, generated by the isotope generator algorithm, and/or selected from a fission product tier. The B2 value represents the burn tabl e output frequency, ordering, and content flag. Table 4-2 displays the different options invoke for different values of B2. Positive input values cause the output to be printed to the output file only at end of an entire job. Negative input values cause output at the end of each burn step.

PAGE 119

119 Three tiers of fission-product content are ava ilable. Each tier co rresponds only to those isotopes that are used for particle transpor t in MCNPX and the nuclide densities, which are reported in the output file. To sele ct a specific tier, add zero (def ault), 10, or 20 to the magnitude of the B2 input value. Table 4-3 gives a descripti on of the available actinides in each fission product tier. Table 4-3. Options invoked for different B2 values. Tier Fission Product Content 1 Zr-93, Mo-95, Tc-99m, Ru-101, Xe-131, Xe-134, Cs-133, Cs-137, Ba-138, Pr-141, Nd-143, Nd-145 (default). 2 All fission products that posse s both CINDER90 yield data and transport cross data that are listed in the default MCNPX library file for MCNPX 2.6.F (XSDIR file). 3 All 220 available fission pr oducts listed in Figure 3-8. The B3 value of the BOPT keyword invokes the models opti on. Cross-section models are a heritage of the high-energy physics capabilities of MCNPX. The models are benchmarked for a variety of problems for neutron energies >150 Me V. Table 4-4 describes the available options for the B3 value of the BOPT keyword. Table 4-4. Available options for the B3 value of the BOPT keyword. B3 value Available Option -1 Receive a fatal error if cross sect ion models are used in the problem 0 Zeros out the atom fraction of any isotope using a cross section model (it is preferable to determin e which isotopes are needed to be omitted and to place them on the OMIT keyword for the burned material). 1 Use cross section models in the calculation for isotopes not containing tabular inte raction rate data. The MCNPX depletion capability tallies energy integrated reaction rates and normalization constants within each separate burn material us ing the track length estimator. The track length estimator of the flux is calculated by: n n nV w l N 1 (4-1)

PAGE 120

120 Initially, MCNPX was setup to calculate and su m the quantity within the larger parenthesis during each track for each individu al burn material and then later divide by the total number of histories. Due to limitations in the repeated structures mapping algorithm in MCNPX, the code cannot always determine the total volume of a pletho ra of repeated cells. Excessive time is also wasted trying to query and calculate the tota l volume. For simplicity, in MCNPX 2.6.D, the tallying algorithm in equa tion 4-1 was altered to n n nw l NV 1 (4-2) Now only the particle track multiplied by the weight is tallied during each track for each individual burn material. Once particle tracking is complete, th e track length estimator is now determined by dividing this new quantity by the su m of all volumes for a specific burn material and the total histories. The user may then i nput the total burn materi al volume on the MATVOL keyword. This procedure eliminates the dependenc y of the code to try and determine this total volume when the code may or may not be able to determine this quantity or when the code may require excess computation to map this value. Th erefore the user is now free to take advantage of the repeated structure geometry features, such as modeling repeat ed arrays of fuel pins, while using the burnup capability.96 For burnup regions without repeated structur es, MCNPX will try to calculate a volume and, if unsuccessful, a fatal error is generated and the user is requi red to then specify the material volume using a VOL card. In the case of cells in a lattice, MCNPX is only capable of successfully accounting for the volumes of materials if each cell is used onl y once in the lattice. Otherwise, if certain cells are repeated in a lattice, MCNPX is unable to determine the total volume of everywhere the ma terial is repeated. The MATVOL keyword is used to account for the volume of materials listed in a repeated structure

PAGE 121

121 MATVOL = V1, V2, V3, Vn, where Vi is the total volume of all cells [cm3] containing burn material, Mi, on the MAT keyword of the BURN card. Table 4-5. MATMOD keyw ord variable values. Keyword Value Description NT Number of time steps (1 through NT) TSNT Time step (1..NT) for which to manua lly change nuclide concentrations of material MTNT,NM NT Enter for 2nd, etc. (If positive apply concentrations discretely at Ti and Ti+1/2. If negative apply at Ti, and make linear interpolation between Ti and Ti+1 to determine the concentration at Ti+1/2. If TS is negative at Ti and the concentrations of an y of the altered isotopes at Ti+1 is equal to the concentration set at Ti then the concentrations of the altered isotopes will be set to the value at Ti for Ti, Ti+1/2, Ti+1. At Ti+3/2 the isotopes will undergo a normal depleti on and the concentrations will not be set to the value at Ti+1. NMNT Number of materials at time step NT that incur nuclide concentration changes MTNT,NM NT NMth material number for which to manually change nuclides at time step NT. Positive value indicates at om/wt. fraction and negative value indicates atom/gram densities. KNT,NM NT Number of nuclides to manually change for the NMth material NT NM NT, NTK NM NT,Z 1st, 2nd,.. Kth nuclide of the MTNT,NM NT th material at time step NT for which a concentration will be specified. List as a ZA value. NT NM NT, NTK NM NT,C Concentration for the Kth isotope in material MTNT,NM NT at time step NT. Enter positive values for atom fracti ons or atom densities, and enter a negative value for wt. fractions or gram densities. See sign of MTNT,NM NT to specify either fraction or density If the MATVOL keyword is utilized on the BURN card then there must exist a corresponding MATVOL entry for every entry on the MAT keyword of the BURN card. If repeated structures are used without a MATVOL keyword, MCNPX will incorrectly ca lculate the volume, if possible, assuming that the total volume for the entire bur n material is equal to the volume stated on the VOL card and not the accumulated the sum for the amount of times the cell is repeated. The MATMOD keyword allows for time-dependent manual nuclide concentration changes. The details of the methodology for this feature were explained in the Manual Time-Dependent

PAGE 122

122 Isotope Concentration Changes section. Table 45 describes the detail of each value within the MATMOD keyword. The MATMOD keyword uses the following st ructure for describing a manual concentration change: MATMOD = NT TS1 NM1 MT1,1 K1,1 1 1,1Z 1 1,1C 2 1,1Z 2 1,1C 1,1K 1,1Z 1,1K 1,1C MT1,NM 1 K1,NM 1 1 NM 1,1Z 1 NM 1,1C 2 NM 1,1Z 2 NM 1,1C 1 NM 1, 1K NM 1,Z 1 NM 1, 1K NM 1,C TSNT NMNT MTNT,1 KNT,1 1 NT,1Z 1 NT,1C 2 NT,1Z 2 NT,1C NT,1K NT,1Z NT,1K NT,1C MTNT,NM NT KNT,NM NT 1 NM NT,NTZ 1 NM NT,NTC 2 NM NT,NTZ 2 NM NT,NTC. NT NM NT, NTK NM NT,Z NT NM NT, NTK NM NT,C When using the MATMOD keyword, if a burn material is set to have a concentration change at T1, the atom density of that isotope at T1/ 2 is set to the initial value specified at T0. This is only set for the initial midpoint time step, the rest of the calculation will follow the procedure mentioned above. Output Structuring The MCNPX depletion capability offers many varied outputs describing the temporal nuclide inventory process. Initially, the objective of the depl etion outputs was to give the user the maximum amount of information necessary in order to make informed system design decisions from understanding important parameters regarding the evolut ion of the temporal nuclide inventory solution. The depl etion outputs may be segregated into five distinct sections. These output sections incl ude collision rates sent to CINDER90 for use as depletion coefficients, generalized global system averag ed depletion data, generalized individual material depletion data, individual material nuclide data, and syst em average nuclide data. Each of the output sections is composed of key important informa tion regarding the temporal nuclide inventory.

PAGE 123

123 Total continuous energy integrated collision ra tes within a each burn volume are calculated in the particle transport calculation for (n,fission), (n, ), (n,2n), (n,3n), (n,p) and (n, ). Examining these collision rates is important to un derstanding how the deplet ion system operates. The collision rates for each burn material, tracked in particle transport to be sent to CINDER90 for use as depletion coefficients are printed at the end of each steady-st ate KCODE calculation. The error associated with each collision rate is the error assuming the steady-state calculation is independent from the propagated error resulting from propagated flux and number densitiy error from the burnup calculation. Figure 4-3 displa ys a cropped example of the format of the collision rate outputs. These outputs help the us er to calculate generalized capture-fission ratios, as well as other inelastic reaction ratios, to determine system characteristics. Material #: 1 no. zaid (n,gamma) error (n,fission) error (n,nu*fission) error (n,fission*Q) error 1 6012 8.19001E+05 0.5671 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000 2 6013 2.81939E+08 0.1330 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000 3 6014 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000 ... ... 60 96244 8.41178E+13 0.4713 5.60552E+12 0.1094 2.09288E+13 0.1034 1.18632E+15 0.1094 61 96245 4.07591E+11 0.0850 2.84278E+12 0.0738 1.02719E+13 0.0737 5.68499E+14 0.0738 no. zaid (n,2n) error (n,3n) error (n,alpha) error (n,p) error 1 6012 0.00000E+00 0.0000 0.00000E+00 0.0000 7.08058E+05 0.6535 0.00000E+00 0.0000 2 6013 3.16027E+08 0.9128 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000 3 6014 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000 ... ... 60 96244 9.03869E+09 0.6024 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000 61 96245 4.49267E+08 0.4078 0.00000E+00 0.0000 0.00000E+00 0.0000 0.00000E+00 0.0000 ... Figure 4-3. Example collision rate outputs. The burnup summary table contains global sy stem averaged burnup information and is designated print table 210. E ach time step is listed with the corresponding time duration and actual specified depletion time. The following syst em averaged quantities are also listed: the power used for the flux normalization, keff, energy integrated flux, neutrons per fission,

PAGE 124

124 recoverable energy per fission, burnup, and produ ction rate. Figure 4-4 displays an example burnup summary table. 1burnup summary table by material print table 210 neutronics and burnup data step duration time power keff flux ave. nu ave. q burnup source (days) (days) (MW) (GWd/MTU) (nts/se c) 0 0.000E+00 0.000E+00 1.000E+00 1.54021 7.715E+14 2.452 200.979 0.000E+00 7.616E+16 1 5.000E+01 5.000E+01 1.000E+00 1.50987 7.945E+14 2.473 201.411 7.183E+00 7.664E+16 2 1.000E+01 6.000E+01 0.000E+00 1.51150 0.000E+00 2.474 201.448 7.183E+00 0.000E+00 3 5.000E+02 5.600E+02 2.000E-01 1.43413 1.699E+14 2.510 202.199 2.155E+01 1.550E+16 ... Figure 4-4. Example bur nup summary table. ... Individual Material Burnup Material #: 1 step duration time power fraction burnup (days) (days) (GWd/MTU) 0 0.000E+00 0.000E+00 5.015E-01 0.000E+00 1 5.000E+01 5.000E+01 5.016E-01 7.205E+00 2 1.000E+01 6.000E+01 5.002E-01 7.205E+00 3 5.000E+02 5.600E+02 5.002E-01 2.158E+01 Material #: 4 step duration time power fraction burnup (days) (days) (GWd/MTU) 0 0.000E+00 0.000E+00 4.985E-01 0.000E+00 1 5.000E+01 5.000E+01 4.984E-01 7.161E+00 2 1.000E+01 6.000E+01 4.998E-01 7.161E+00 3 5.000E+02 5.600E+02 4.998E-01 2.152E+01 ... Figure 4-5. Example indi vidual material burnup data. If multiple materials are burned then there al so exists individual material burnup data. Figure 4-5 displays example indivi dual material burnup data. Fo r this example both materials 1 and 4 were burned and therefore individual burn material burnup data is available only for those materials. The available information includes: time step, time duration, actual time, fission power fraction and individual material burnup. The fission power fract ion is calculated by

PAGE 125

125 taking the ratio of the fission power in a particular burn material to the sum of all burn materials. The calculated fission power fraction is therefor e only related to fissions in burn materials. i i f i fVQ VQ fraction power (4-3) The individual material burnup is calculated by MTHM PFRAC TIME Fraction POWER POWER Burnup Burnupi* *1 (4-4) ... nuclide data are sorted by increasing zaid for material 1 volume 3.8457E+02 (cm**3) actinide inventory for material 1 at end of step 0, time 0.000E+00 (days), power 1.000E+00 (MW) no. zaid mass activity spec.act. atom den. atom fr. mass fr. (gm) (Ci) (Ci/gm) (a/b-cm) 1 90231 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 2 90232 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 ... ... 6 92235 3.441E+02 0.000E+00 0.000E+00 2.293E-03 1.000E-01 9.886E-02 ... ... actinide inventory for material 1 at end of step 1, time 5.000E+01 (days), power 1.000E+00 (MW) no. zaid mass activity spec.act. atom den. atom fr. mass fr. (gm) (Ci) (Ci/gm) (a/b-cm) 1 9023 1.286E-09 6.837E-04 5.315E+05 8.718E-15 3.832E-13 3.723E-13 2 90232 2.394E-08 2.625E-15 1.097E-07 1.616E-13 7.100E-12 6.929E-12 ... ... totals 3.455E+03 2.584E+05 7.479E+01 2.275E-02 1.000E+00 1.000E+00 ... ... nonactinide inventory for material 1 at end of step 0, time 0.000E+00 (days), power 1.000E+00 (MW) no. zaid mass activity spec.act. atom den. atom fr. mass fr. (gm) (Ci) (Ci/gm) (a/b-cm) 1 6012 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 2 6013 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 3 7014 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 4 7015 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 5 8016 4.684E+02 0.000E+00 0.000E+00 4.585E-02 1.000E+00 1.000E+00 ... Figure 4-6. Example cropped actinide output at two separate timesteps and fission product output for one time step.

PAGE 126

126 The time dependant isotope buildup/deple tion is listed afte r the burnup summary information. The isotope buildup/depletion for e ach individual material is given at each time step. The information is further subdivided in to actinide and nonactinide categories. Figure 4-6 displays a cropped exampled of the reported ac tinide inventory for two separate time steps. At the end of each subdivision is an accumulati on total of the isotope information for that subdivision. Atom and weight fr actions calculations are based on the fractions of that specific subdivision. After isotope information for each i ndividual material is given, print table 220 lists the sum total information of all actinides and non actinides from all materials combined for each time step. Figure 4-7 shows a cropp ed example of print table 220. 1burnup summary table summed over all materials print table 220 nuclides with atom fractions below 1.000E-32 for a mate rial are zeroed and deleted from print tables after t=0 nuclide data are sorted by increasing zaid summe d over all materials volume 7.6914E+02 (cm**3) actinide inventory for sum of materials at end of step 0, time 0.000E+00 (days), power 1.000E+00 (MW) no. zaid mass activity spec.act. atom den. atom fr. mass fr. (gm) (Ci) (Ci/gm) (a/b-cm) 1 90231 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 2 90232 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 3 90233 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 4 91233 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 5 92234 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 0.000E+00 6 92235 6.883E+02 0.000E+00 0.000E+00 4.585E-03 1.000E-01 9.886E-02 ... Figure 4-7. Cropped example of print table 220. The output is structured so as to give the user maximum amount of information in well organized concise tables. Each concise table is located in the output in the order in which the calculations have been made. For example, th e collision rate tables always follow the KCODE information, and the burnup summary tables are th en located directly be low the collision rate tables (or at the end of the en tire output if so invoked by an output option). For a large burnup

PAGE 127

127 example, using many burn material s, the user will generate hu ge output files as a result of reporting all of the information available in the burnup tables. However, this information is necessary for determining how derivative quantiti es, such as reaction rates, contribute to the integral solution, such as gl obal time dependant neutron multip lication and isotope analysis.

PAGE 128

128 CHAPTER 5 BENCHMARKS The Nuclear Energy Agency (NEA) is a speci fic agency within the Organization for Economic Cooperation and Development (OECD), an intergovernmental organization composed of many industrialized countries, ba sed in Paris, France. The missi on of the NEA is to assist its member countries in maintaining and further de veloping, through interna tional co-operation, the scientific, technological and lega l bases required for the safe environmentally friendly and economical use of nuclear energy for peaceful purposes.97 The membership consists of 28 countries including: Australia, Austria, Belgiu m, Canada, Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Iceland, Irel and, Italy, Japan, Luxembourg, Mexico, the Netherlands, Norway, Portugal, Republic of Korea, Slovak Republic, Spain, Sweden, Switzerland, Turkey, the United Kingdom and the United States.97 The NEA is further divided into many working parties and these working partie s are composed of expert working groups that are tasked with examining various aspects of the Nuclear Fuel Cycle.97 Criticality safety aspects of the Nuclear Fuel Cycle are examined by th e Working Party on Nuclear Criticality Safety (WPNCS), and understanding the reactivity and isotope prediction capability is the responsibility of the Burnup Credit Criticality Safety expert gr oup. One of the objectives of the Burnup Credit Criticality Safety expert group is to examine burn-up credit as applied to criticality safety in the transportation, storage, and treatment of spent fu el for a wide range of fuel types, including urainium-oxides (UOX) and MOX fuels for PWRs, BWRs, and Vodaa Vodiannee Energititscherski Reactors (VVER). Achieving this objective involves car rying out international comparisons and benchmarks in order to assess the capability of code systems to accurately predict reactivity and temporal nuclide inventories, including bot h code-to-code and available experimental data comparisons.97

PAGE 129

129 In order to asses the validity of the MCNP X depletion method three separate types of OECD/NEA benchmarks were completed. Table 5-1 lists the benchmarks that were used to test the MCNPX depletion capability. In most of the benchmarks, many combinations of operating techniques were investigated. For the purpose of this study, only benchmarks that employ the maximum amount of fission products following all transmuted actinides were investigated. Therefore only a subset of the full amount of cas es analyzed in the complete benchmark were examined. However, the analyzed sections of th ese benchmarks do adequate ly test the ability of the code to predict real core behavior as a maximum amount of fission products are generated during real reactor behavior. It is important to realize that the specified benchmarks only truly test the codes predictive capability for certain LWR operating strategies. However, before one tries to validate a methodology for more complicated operating scenarios, one must be able to first also correctively predict beha vior of well benchmarked LWRs. Table 5-1. Examined benchmark cases. Benchmark Case Geometry Type OECD/NEA Burnup Credit Phase IB52 PWR UO2 adjusted pin cell calculation H. B. Robinson53 PWR UO2 assembly calcation OECD/NEA Burnup Credit Phase IVB54 LWR MOX assembly calculation OECD/NEA Burnup Credit Phase IB Benchmark Specification The purpose of the OECD/NEA Burnup Credit Ca lculational Criticality Benchmark Phase IB is to compare computed nuclide concentra tions for a simple pin-cell depletion model.52 Prior radiochemical analysis was completed on a 14 X 14 Combustion Engineering fuel assembly in order to determine nuclide concentrations for use in the benchmark. The simple pin cell model incorporates actual pin dimensions from a Comb ustion Engineering assembly; however, the fuel pin pitch has been adjusted in order to mimic the fuel-to-moderator ratio exhibited in the full assembly. Further approximations applied to th e analyzed geometry included modeling the fuel

PAGE 130

130 pin as an infinitely reflected geometry. An infinitely reflected geometry assumes zero net leakage at the geometry boundary. This type of boundary condition is only incurred within a flat power distribution region of the reactor. Such a region is never incurred on a pin cell level and therefore the approximation of infi nitely reflected geometry was expected to affect the final solution outcome. Figure 5-1. Infinitely reflected pin cell geometry. Table 5-2. Model descripti on for Phase IB Benchmark*. Parameter Data Type fuel pellet UO2 Rod pitch 1.5586 cm Clad outer diameter 1.118 cm Clad inner diameter 0.986 cm Fuel diameter 0.9563 cm Active fuel length 347.2 cm Effective fuel temperature 841 K Clad temperature 620 K Water temperature 558 K *Data taken from reference. The pin cell model was divided in to four radial regions. The pin geometry is displayed in Figure 5-1 and the fuel pin dimens ions are listed in Table 5-2. Temperatures were assumed constant throughout each region. The fuel pin wa s composed of a tall cylindrical fuel region divided into ten axial regions implementing th e same average temperature and volume in each BoratedWater Clad Fuel Gap

PAGE 131

131 axial division. Surrounding the fuel region was a cylin drical voided gap region. Enclosing the fuel and gap regions was a cylin drical Zicaloy-2 clad region. Finally encasing the entire geometry was a rectangular borated water box. Th e fuel, clad, and borated water concentrations are listed in Table 5-3, Table 5-4 and Table 5-5. The compositi ons are displayed in the same units as listed in the benchmark report.52 Three different pin burnup cases were analyzed. For each burnup case, four operating cycles we re implemented using common burn times, down times and operating concentrations. Table 5-6 de tails the common operating strategies utilized for each case. Table 5-7 lists the operating power s utilized for each individual case in order to achieve the prescribed burnups. Table 5-3. Fuel composition for Phas e IB Benchmark (density =10.045 g/cm3). Nuclide Number Density (atoms/b-cm) 234U 6.15164E-6 235U 6.89220E-4 236U 3.16265E-6 238U 2.17104E-2 12C 9.13357E-6 14N 1.04072E-5 16O 4.48178E-2 ENDF/B VII.0 temperature dependent cross sections and scattering law data were implemented in the steady state reaction rate ca lculation. Each stead y-state reaction rate calculation was executed using 3000 particles per cycle for 115 cycles skipping the first 5 cycles. Ten starting source locations were equally spaced axially along the fuel pin center. The combination of skipped cycles a nd source locations was hypothesized to result in an adequately converged fission source distribution, thereby resulting in a further c onverged reaction rate calculation. The amount time steps, particles per cycle, and cycles were minimized in order to achieve a depletion solution w ithin a minimal amount of time. Running on a linux platform across 25 1.4 GHz Opteron processors, each case ran for ~22 hours.

PAGE 132

132 Table 5-4. Clad composition for Phase IB Benchmark (6.44 g/ cm3). Nuclide Weight Fraction 54Fe 0.0275 % 56Fe 0.4477 % 57Fe 0.0105 % 58Fe 0.0143 % 90Zr 49.0926 % 91Zr 10.8251 % 92Zr 17.8204 % 94Zr 17.3218 % 96Zr 2.8501 % 112Sn 0.0145 % 114Sn 0.0101 % 115Sn 0.0052 % 116Sn 0.2257 % 117Sn 0.1203 % 118Sn 0.3825 % 119Sn 0.1368 % 120Sn 0.5232 % 122Sn 0.0756 % 124Sn 0.0961 % Table 5-5. Borated water composition for Phase IB Benchmark (0.7569 g/ cm3). Nuclide Number Density (atoms/b-cm) 1H 5.06153E-2 16O 2.53076E-2 10B 2.75612E-6 11B 1.11890E-5 Table 5-6. Operating history da ta for Phase IB Benchmark. Operating Cycle Burn time (days) Burn Duration (days) Downtime (days) Boron Concentration (ppm) 1 306.0 1, 15, 30, 40, 50, 50, 60, 60 71.0 331.0 2 381.7 1.7, 20, 35, 45, 55, 65, 75, 85 83.1 469.7 3 466.0 6, 20, 45, 65, 70, 80, 85, 95 85.0 504.1 4 461.1 6.1, 20, 45, 60, 70, 80, 85, 95 1870.0 492.5

PAGE 133

133 Table 5-7. Operating power (MW) for each case for Phase IB Benchmark. Operating Cycle Case A (final burnup = 27.35 GWD/MTU) Case B (final burnup = 37.12 GWD/MTU) Case C (final burnup = 44.34 GWD/MTU) 1 0.038068 0.054585 0.068717 2 0.042904 0.059090 0.071786 3 0.037627 0.050434 0.057853 4 0.032172 0.041667 0.048844 OECD/NEA Burnup Credit Phase IB Results Table 5-8. Initial Phase IB Benchmark resu lts and comparison of Case A (27.35 MWD/MTU). Isotope MCNPX Value (mg/g UO2) Experiment Value (mg/g UO2) % Error* Range of Values from other Codes 234U 1.408E-1 1.600E-1 -12.00 1.330E-1 to 1.750E-1 235U 7.988 8.470 -5.69 7.445 to 8.661 236U 3.227 3.140 2.77 3.128 to 3.540 238U 8.375E+2 8.425E+2 -0.59 8.637E+2 to 8.415E+2 237Np 2.764E-1 2.680E-1 3.12 2.527E-1 to 3.396E-1 238Pu 9.046E-2 1.012E-1 -10.61 5.721E-2 to 1.083E-1 239Pu 3.790 4.264 -11.12 3.660 to 4.690 240Pu 1.599 1.719 -6.99 1.573 to 8.60 241Pu 6.491E-1 6.812E-1 -4.71 5.310E-1 to 7.335E-1 242Pu 2.952E-1 2.886E-1 2.30 2.000E-1 to 3.192E-1 241Am 2.274E-1 N/A N/A 2.269E-1 to 2.598E-1 243Am 4.487E-2 N/A N/A 3.480E-2 to 4.672E-1 95Mo 5.501E-1 N/A N/A 5.590E-1 to 5.795E-1 99Tc 5.397E-1 N/A N/A 5.648E-1 to 6.904E-1 133Cs 8.012E-1 8.500E-1 -5.74 6.820E-1 to 8.640E-1 135Cs 3.535E-1 3.600E-1 -1.80 3.728E-1 to 3.959E-1 143Nd 5.752E-1 6.130E-1 -6.16 6.040E-1 to 6.792E-1 145Nd 4.886E-1 5.100E-1 -4.19 4.984E-1 to 5.151E-1 147Sm 1.522E-1 N/A N/A 1.564E-1 to 1.932E-1 149Sm 1.628E-3 2.900E-3 -43.86 1.626E-3 to 2.900E-3 150Sm 1.762E-1 2.070E-1 -14.86 1.713E-1 to 2.146E-1 151Sm 7.070E-3 N/A N/A 6.376E-3 to1.413E-2 152Sm 7.940E-2 8.700E-2 -8.74 7.947E-2 to 1.073E-1 153Eu 7.022E-2 7.900E-2 -11.12 6.730E-2 to 8.921E-2 155Gd 2.629E-3 N/A N/A 1.507E-3 to 5.762E-3 *(Calculated/Measured-1)*100

PAGE 134

134 The Burnup Credit Phase IB benchmark contains data from a variety of calculation tools, as well as experimental results from radi ochemical analysis of the 14 X 14 Combustion Engineering fuel assembly, for certain key act inides and fission products. The initial MCNPX calculated nuclide concentrations for all three be nchmark cases are displayed in Table 5-8, Table 5-9 and Table 5-10. For each nuclide the follo wing quantities are listed: MCNPX calculated value, experimentally determined value, per cent error between MCNPX calculated value and experimental value and range of calculated values from other va rious methodologies presented in the benchmark. Table 5-9. Initial Phase IB Benchmark resu lts and comparison of Case B (37.12 MWD/MTU). Isotope MCNPX Value (mg/g UO2) Experiment Value (mg/g UO2) % Error* Range of Values from other Codes 234U 1.148E-01 1.400E-01 -17.96 1.080E-1 to 1.570E-01 235U 4.587 5.170 -11.28 4.022 to 5.510 236U 3.641 3.530 3.15 3.526 to 3.930 238U 8.303E+02 8.327E+02 -0.28 8.292E+02 to 8.360E+02 237Np 3.980E-01 3.560E-01 11.79 3.619E-01 to 4.919E-01 238Pu 1.727E-01 1.893E-01 -8.77 1.144E-01 to 2.069E-01 239Pu 3.849 4.357 -11.66 3.710 to 4.877 240Pu 2.057 2.239 -8.14 1.996 to 2.347 241Pu 8.379E-01 9.028E-01 -7.19 7.510E-01 to 9.846E-01 242Pu 5.916E-01 5.761E-01 2.69 4.200E-01 to 6.347E-01 241Am 2.853E-01 N/A N/A 2.880E-01 to 3.418E-01 243Am 1.181E-01 N/A N/A 9.637E-02 to 1.391E-01 95Mo 7.066E-01 N/A N/A 7.214E-01 to 7.545E-01 99Tc 6.898E-01 N/A N/A 7.327E-01 to 8.372E-01 133Cs 1.022 1.090 -6.24 8.784E-01 to 1.117 135Cs 3.793E-01 4.000E-01 -5.17 3.967E-01 to 4.317E-01 143Nd 6.523E-01 7.160E-01 -8.90 7.013E-01 to 8.254E-01 145Nd 6.112E-01 6.530E-01 -6.41 6.326E-01 to 6.600E-01 147Sm 1.607E-01 N/A N/A 1.659E-01 to 2.201E-01 149Sm 1.653E-03 3.000E-03 -44.91 1.736E-03 to 3.092E-03 150Sm 2.311E-01 2.710E-01 -14.72 2.297E-01 to 3.152E-01 151Sm 7.397E-03 N/A N/A 7.990E-03 to 1.571E-02 152Sm 9.521E-02 1.040E-01 -8.45 9.761E-02 to 1.416E-01 153Eu 9.940E-02 1.090E-01 -8.81 9.960E-02 to 9.960E-02 155Gd 4.076E-03 N/A N/A 2.538E-03 to 1.028E-02 *(Calculated/Measured-1)*100

PAGE 135

135 Table 5-10. Initial Phase IB Benchmark resu lts and comparison of Case C (44.34 MWD/MTU). Isotope MCNPX Value (mg/g UO2) Experiment Value (mg/g UO2) % Error* Range of Values from other Codes 234U 9.896E-02 1.200E-01 -17.53 9.030E-02 to 1.440E-01 235U 2.897 3.540 -18.18 2.389 to 3.716 236U 3.772 3.690 2.21 3.641 to 4.030 238U 8.243E+02 8.249E+02 -0.07 8.234E+02 to 8.316E+02 237Np 4.794E-01 4.680E-01 2.44 4.327E-01 to 5.934E-01 238Pu 2.399E-01 2.688E-01 -10.76 1.656E-01 to 2.810E-01 239Pu 3.832 4.357 -12.04 3.659 to 4.902 240Pu 2.292 2.543 -9.86 2.180 to 2.661 241Pu 9.229E-01 1.020 -9.52 8.560E-01 to 1.111 242Pu 8.479E-01 8.401E-01 0.93 5.960E-01 to 9.103E-01 241Am 3.071E-01 N/A N/A 3.102E-01 to 3.785E-01 243Am 1.924E-01 N/A N/A 1.634E-01 to 2.316E-01 95Mo 8.100E-01 N/A N/A 8.092E-01 to 8.742E-01 99Tc 7.852E-01 N/A N/A 8.449E-01 to 9.861E-01 133Cs 1.160 1.240 -6.45 9.723E-01 to 1.286 135Cs 3.958E-01 4.300E-01 -7.94 3.977E-01 to 4.605E-01 143Nd 6.746E-01 7.630E-01 -11.58 7.397E-01 to 8.839E-01 145Nd 6.858E-01 7.440E-01 -7.82 7.170E-01 to 7.560E-01 147Sm 1.598E-01 N/A N/A 1.655E-01 to 2.302E-01 149Sm 1.686E-03 4.700E-03 -64.13 1.842E-03 to 3.286E-03 150Sm 2.690E-01 3.610E-01 -25.49 2.725E-01 to 3.980E-01 151Sm 7.585E-03 N/A N/A 8.102E-03 to 1.682E-02 152Sm 1.038E-01 1.210E-01 -14.25 1.077E-01 to 1.587E-01 153Eu 1.173E-01 1.480E-01 -20.75 1.210E-01 to 1.596E-01 155Gd 5.046E-03 N/A N/A 3.379E-03 to 1.318E-02 *(Calculated/Measured-1)*100 MCNPX initially had poor nuclide prediction. The predictive capability also seemed to get worse at higher burnups. MCNPX was able to calc ulate actinide prediction within the range of other depletion codes; however, the actinide pred ictions were not close to the measured values and the percent error betw een measured and calculated values increased with increasing burnup. MCNPX fission product predictions were outside of the range values predicted by other depletion codes, and the fission pr oduct predictions were also not close to the measured values. The percent error between measur ed and calculated fission product values also increased with increasing burnup. Because the MCNPX re sults were poorer at higher burnups, it was

PAGE 136

136 hypothesized that the burnup per time step might affect the accuracy of the result. Limitation of the burnup per time step is function of both significan t changes in reaction ra te over a time step, and limitations in the actual depletion algorithm. It is true that common time steps were used for each burnup case, and therefore the burnup step durations were unequal for all three cases. Therefore for the higher burnup cases it is possible that the time step duration was too long and too much burnup occurred ultimat ely invalidating the assumption th at significant change in flux had not occurred. However, since the pred ictive capability was so poor compared to a deterministic solver, the hypothesis that the CI NDER90 algorithm may be faulted was tested. CINDER90 uses a Linear Markov Chain method in order to solve the depletion equations. CINDER90 solves for nuclide densities by following a series of equations along the transmutation chain.61, 50 The following procedure is used: 1. Linear chains are created for each isotope transmutation path. 2. The solutions of each linear chain determines a partial nuclide density 3. Each calculated partial nuclide density, co mputed from a linear chain, is then summed to obtain the total nuclid e inventory of nuclide. The differential equation governing the computati on of the ith nuclide is therefore only coupled to preceding contributing nuclides leading to the cr eation of the ith nuclide. However, in order to stop the propagation of a chain, the code determines if significant progeny will be created; and if the current nuclide does not produce significant pr ogeny the calculation of the chain ceases. A series of checks exist within CINDER90 to determine if progeny will be created, passby calculation, and these checks posses limiting checki ng values for determining significance. For example, the magnitude of the nuclide density of a given isotope and the integral of the creation coefficient must both be above a ce rtain value or the code ceases the transmutation chain. This would appear to affect short ha lf-live nuclides in secular equili brium and all resulting nuclides

PAGE 137

137 within that transmutation chain. Preliminary investigation suggested th at the order of such checks and the magnitude of the limiting checking values determined how accurately the code determined if significant progeny existed thereby ultimately affecting the solution outcome. Table 5-11. MCNPXc Phase IB Benchmark results and comparison of Case A (27.35 MWD/MTU). Isotope MCNPX Value (mg/g UO2) Experiment Value (mg/g UO2) % Error* Range of Values from other Codes 234U 1.422E-01 1.600E-1 -11.13 1.330E-1 to 1.750E-1 235U 8.339 8.470 -1.54 7.445 to 8.661 236U 3.173 3.140 1.06 3.128 to 3.540 238U 8.375E+02 8.425E+2 -0.59 8.637E+2 to 8.415E+2 237Np 2.794E-01 2.680E-1 4.25 2.527E-1 to 3.396E-1 238Pu 9.445E-02 1.012E-1 -6.67 5.721E-2 to 1.083E-1 239Pu 3.891 4.264 -8.74 3.660 to 4.690 240Pu 1.640 1.719 -4.60 1.573 to 8.60 241Pu 6.615E-01 6.812E-1 -2.90 5.310E-1 to 7.335E-1 242Pu 3.069E-01 2.886E-1 6.36 2.000E-1 to 3.192E-1 241Am 2.320E-01 N/A N/A 2.269E-1 to 2.598E-1 243Am 4.886E-02 N/A N/A 3.480E-2 to 4.672E-1 95Mo 5.661E-01 N/A N/A 5.590E-1 to 5.795E-1 99Tc 5.661E-01 N/A N/A 5.648E-1 to 6.904E-1 133Cs 8.259E-01 8.500E-1 -2.83 6.820E-1 to 8.640E-1 135Cs 3.620E-01 3.600E-1 0.54 3.728E-1 to 3.959E-1 143Nd 5.840E-01 6.130E-1 -4.73 6.040E-1 to 6.792E-1 145Nd 4.894E-01 5.100E-1 -4.04 4.984E-1 to 5.151E-1 147Sm 1.518E-01 N/A N/A 1.564E-1 to 1.932E-1 149Sm 1.679E-03 2.900E-3 -42.10 1.626E-3 to 2.900E-3 150Sm 1.854E-01 2.070E-1 -10.44 1.713E-1 to 2.146E-1 151Sm 7.401E-03 N/A N/A 6.376E-3 to1.413E-2 152Sm 8.124E-02 8.700E-2 -6.62 7.947E-2 to 1.073E-1 153Eu 7.257E-02 7.900E-2 -8.13 6.730E-2 to 8.921E-2 155Gd 2.724E-03 N/A N/A 1.507E-3 to 5.762E-3 *(Calculated/Measured-1)*100 A further investigation was completed to set the limiting checking value for density contribution of a specific nuclide to 1E-10 from th e default 1E-4 value in order to determine if this change actually affected the ultimate deplet ion outcome. Table 5-11, Table 5-12 and Table 5-13 display the results augmenting this limiting checking value (MCNPXc). All of the same

PAGE 138

138 conditions applied in the initia l calculations were also applie d for the MCNPXc cases. The results from the MCNPXc cases show significan t improvement in actin ide and fission product prediction. Table 5-12. MCNPXc Phase IB Benchmark results and comparison of Case B (37.12 MWD/MTU). Isotope MCNPXc Value (mg/g UO2) Experiment Value (mg/g UO2) % Error* Rang of Values from other Codes 234U 1.164E-01 1.400E-01 -16.88 1.080E-1 to 1.570E-01 235U 4.822 5.170 -6.73 4.022 to 5.510 236U 3.619 3.530 2.51 3.526 to 3.930 238U 8.303E+02 8.327E+02 -0.28 8.292E+02 to 8.360E+02 237Np 4.032E-01 3.560E-01 13.26 3.619E-01 to 4.919E-01 238Pu 1.750E-01 1.893E-01 -7.55 1.144E-01 to 2.069E-01 239Pu 3.987 4.357 -8.50 3.710 to 4.877 240Pu 2.120 2.239 -5.31 1.996 to 2.347 241Pu 8.699E-01 9.028E-01 -3.65 7.510E-01 to 9.846E-01 242Pu 5.936E-01 5.761E-01 3.04 4.200E-01 to 6.347E-01 241Am 2.976E-01 N/A N/A 2.880E-01 to 3.418E-01 243Am 1.186E-01 N/A N/A 9.637E-02 to 1.391E-01 95Mo 7.293E-01 N/A N/A 7.214E-01 to 7.545E-01 99Tc 7.277E-01 N/A N/A 7.327E-01 to 8.372E-01 133Cs 1.060 1.090 -6.24 8.784E-01 to 1.117 135Cs 3.930E-01 4.000E-01 -5.17 3.967E-01 to 4.317E-01 143Nd 6.750E-01 7.160E-01 -8.90 7.013E-01 to 8.254E-01 145Nd 6.172E-01 6.530E-01 -6.41 6.326E-01 to 6.600E-01 147Sm 1.632E-01 N/A N/A 1.659E-01 to 2.201E-01 149Sm 1.711E-03 3.000E-03 -44.91 1.736E-03 to 3.092E-03 150Sm 2.521E-01 2.710E-01 -14.72 2.297E-01 to 3.152E-01 151Sm 7.924E-03 N/A N/A 7.990E-03 to 1.571E-02 152Sm 9.892E-02 1.040E-01 -8.45 9.761E-02 to 1.416E-01 153Eu 1.035E-01 1.090E-01 -8.81 9.960E-02 to 9.960E-02 155Gd 4.251E-03 N/A N/A 2.538E-03 to 1.028E-02 *(Calculated/Measured-1)*100 For the most part, the results calculated by MCNPXc fell within the range of values calculated by the other depletion codes. These calculated values, however did not exactly match the measured experiment values. There are many possible explanations fo r the discrepancies in calculated results as compared to the experimental data. The reasons include: (1) nuclear data;

PAGE 139

139 (2) treatment of normalization parameters; (3) inad equacy of time steps; (4) statistical variance (5) differences in neutron spectra; (6) spatial reactivity effects of the modeled geometry. Table 5-13. MCNPXc Phase IB Benchmark results and comparison of Case C (44.34 MWD/MTU). Isotope MCNPXc Value (mg/g UO2) Experiment Value (mg/g UO2) % Error* Rang of Values from other Codes 234U 1.006E-01 1.200E-01 -16.14 9.030E-02 to 1.440E-01 235U 3.131 3.540 -11.54 2.389 to 3.716 236U 3.758 3.690 1.84 3.641 to 4.030 238U 8.247E+02 8.249E+02 -0.02 8.234E+02 to 8.316E+02 237Np 4.854E-01 4.680E-01 3.72 4.327E-01 to 5.934E-01 238Pu 2.434E-01 2.688E-01 -9.47 1.656E-01 to 2.810E-01 239Pu 3.970 4.357 -8.89 3.659 to 4.902 240Pu 2.362 2.543 -7.13 2.180 to 2.661 241Pu 9.633E-01 1.020 -5.56 8.560E-01 to 1.111 242Pu 8.463E-01 8.401E-01 0.74 5.960E-01 to 9.103E-01 241Am 3.222E-01 N/A N/A 3.102E-01 to 3.785E-01 243Am 1.919E-01 N/A N/A 1.634E-01 to 2.316E-01 95Mo 8.359E-01 N/A N/A 8.092E-01 to 8.742E-01 99Tc 8.303E-01 N/A N/A 8.449E-01 to 9.861E-01 133Cs 1.206 1.240 -2.71 9.723E-01 to 1.286 135Cs 4.104E-01 4.300E-01 -4.56 3.977E-01 to 4.605E-01 143Nd 7.038E-01 7.630E-01 -7.76 7.397E-01 to 8.839E-01 145Nd 6.930E-01 7.440E-01 -6.85 7.170E-01 to 7.560E-01 147Sm 1.635E-01 N/A N/A 1.655E-01 to 2.302E-01 149Sm 1.751E-03 4.700E-03 -62.75 1.842E-03 to 3.286E-03 150Sm 2.975E-01 3.610E-01 -17.59 2.725E-01 to 3.980E-01 151Sm 8.239E-03 N/A N/A 8.102E-03 to 1.682E-02 152Sm 1.084E-01 1.210E-01 -10.39 1.077E-01 to 1.587E-01 153Eu 1.228E-01 1.480E-01 -17.00 1.210E-01 to 1.596E-01 155Gd 5.289E-03 N/A N/A 3.379E-03 to 1.318E-02 *(Calculated/Measured-1)*100 Nuclear Data The MCNPX benchmark calculations used E NDF/B VII.0 temperature dependent cross section data for steady-state part icle transport. The range of values calculated from other depletion codes used either ENDF/B IV.0 or V.0 which was the data available at the time of the benchmark. Possible discrepancies in the implem ented transport data may result in calculation

PAGE 140

140 error. CINDER90 uses ENDF/B VI.0 fission yiel d data. This data may differ from the fission yield available in ENDF/B V.0 or IV.0 resulting in further error in calculation. CINDER90 also contains transmutation cross sections for 3400 nuclides. Over 3000 of those nuclides do not contain available ENDF data. Therefore cros s section determination codes were used to calculate the data implemented in CINDER90. The CINDER90 data may also differ from what is used by other codes leading to further error in calculation when comparing to other tools. Treatment of Normalization Parameters MCNPX currently does not account for the true recoverable energy per fission. An approximation is made in order to try and de termine the delayed energy contribution. The delayed energy contribution is unfortunately system dependent For this modeled system, however, the geometry modeled was not the act ual experimental geometry and therefore accounting for the true recoverabl e energy per fission in the adjusted geometry would not be possible. Therefore the assumptions made a bout delayed energy contribution were deemed acceptable for lack of a better approximation but known to affect the final solution outcome. Inadequate Time Steps The same time steps were used for each exam ined case. Therefore a larger amount of burnup was witnessed per time step for the higher burnup cases. It is possi ble that the time steps used for the benchmark were not short enough in order to account for si gnificant change in neutron spectrum. Adequate time step selection is usually a trial and e rror process of accounting for significant changes in the neutron spectrum as a result of the asymptotic buildup of highly absorbing isotopes. Therefore choosing time-step s that are too large wi ll not properly account for significant changes in neutron spectrum u ltimately resulting in calculation error.

PAGE 141

141 Statistical Variance By using the Monte Carlo method and simula ting the actual physical process, the calculated reaction rates, used as coefficients fo r the depletion solution, possessed an associated statistical error. This error propagates thr ough the linked solution pro cess. Therefore the coefficients generated for each time step possessed a propagated statistical error. For rarely sampled interactions, this error can be very larg e ultimately affecting the transmutation vector for nuclides resulting from those interactions. Furt hermore, the cross sections themselves also posses a measured error associated with each va lue. The combination of the measured error incurred from determining the cr oss section values and the stoc hastic error from the modeling process ultimately influences the accuracy of the calculated results. Differences in Neutron Spectra The Combustion Engineering 14 X 14 fuel asse mbly that was used for the experimental benchmark was burned with a cr itical spectrum. The modeled geometry, however, possessed a spectrum that was either supercritical, for the be ginning time steps, or su bcritical, for the latter time steps. Unfortunately, true system operating parameters were not given. For example, to maintain criticality in a PWR, the boron concentra tion in the water coolant is constantly diluted in order to subsidize the negative reactivity deficit incurred from burning fuel. For the benchmark cases, boron concentration was assumed constant over the entire cycle. Therefore the neutron spectrum of the modeled geometry could not possibly be made critical. Simply adjusting the boron concentration to ma intain criticality in a pin cell model would also result in further error. Criticality is ma intained in a system by balancing the leakage and absorption. For example, the critical spect rum in a modeled geometry is preserved by accounting for the leakage of neighboring geometries while operating the system with the true operating conditions. Unfortunate ly, when modeling a subsection of a larger geometry, the

PAGE 142

142 leakage into that subsection may not be known a nd must be approximated. Approximating a pin cell calculation with reflective boundary conditions is a false representation of the boundary conditions witnessed in a typical re actor. The only way to truly bur n the fuel bundle at critical is to model the entire system. This benchmark di d not give specification for modeling the entire system and therefore errors should be expected in comparing results with experiment. Spatial Reactivity Effects of the Modeled Geometry The modeled geometry was an infinitely reflec ted pin cell with adjusted fuel-to-moderator ratio in order to try and mimic the same nuc lide buildup as a 14 X14 Combustion Engineering fuel assembly. An infinitely reflected boundary condition is only approp riate for scenarios in which the boundary posses a true zero net leakage. For a certain section of a larger geometry, approximating a zero net leakage boundary condition is acceptable if the s ection of the geometry resides in a region possessing a flat flux or power distribut ion. For example, a fuel assembly that sits in a region of the reactor where the fuel assembly is surrounded by similar fuel assemblies exhibiting a similar power distribut ion. Unfortunately, this conditi on is never truly realized on a pin cell level. A radial power distribution exists across the fuel assembly and therefore each pin within the bundle burns at a diffe rent rate. By only modeling a single fuel pin to represent the entire fuel assembly, the spatia l reactivity effects are ignored th at would otherwise be witnessed in the full fuel assembly. This disregard of spatial reactivity effects ultimately affects the accuracy of the solution outcome. Because only a pin cell is modeled, comparisons between the calculated results and radiochemical analysis data taken from the full assembly analysis are expected to be error. Further MONTEBURNS Comparisons MONTEBURNS calculations for the OECD/NEA Burnup Credit Calculational Criticality Benchmark Phase IB had been comple ted as part of a prior thesis. 75 Due to computational

PAGE 143

143 limitations at the time of those calculations, a coarser representation of the benchmark was modeled. The model implemented for that work used a single 360 cm tall fuel zone, ENDF V cross sections, and single time steps for operati ng durations (the exact model description implemented is explained in the reference). Th e model also lacked many of the fission products captured by the MCNPX depletion methodology (setting the MONTEBURNS transmutation importance fraction to 0.001). In the thesis, valu es were only calculated for cases A and B, and these values were within the range of valu es demonstrated by other codes and MCNPX. In order to better compare the MONTEBUR NS methodology and MCNPX, a series of MONTEBURNS 2.0 calculations were rerun using the same fidelity model as implemented for the MCNPX depletion calculations ENDF VII.0 temperature de pendent cross sections were implemented. The same kcode parameters, time steps, power levels, and manually set material concentrations used in the MCNPX calculatio ns were also used for the MONTEBURNS 2.0 calculations. For these MONTEBURNS calculations, the importance fraction was set to 1e-10 in order to account for more of the fission a nd activation products. MONTEBURNS 2.0 was incapable of burning the borated water region and re setting the boron concentrations at each step. MONTEBURNS 2.0 was also incapable of modeli ng the same isotope at multiple temperatures; therefore all fission products th at aligned with cladding nuclide s used the same temperature dependent cross sections for the cladding temperat ure. These two issues are easily handled in MCNPX, however, not treated in MONTEBURNS 2. 0, and therefore differences were expected as a result of these discrepancies. MONTEBUR NS results are compared with MCNPXc results and experimental values, for cases A a nd B, in Table 5-14 and Table 5-15. The higher fidelity MONEBURNS calculations for this benchmark are in larger error than the MCNPXc calculations. The reasons for the gross error in 238Pu is mostly due to the 242mAm

PAGE 144

144 and 242Am cross section mixup. The large over pred iction of nuclides is not well determined. Increasing the amount of fission products tr acked within MONTEBURNS 2.0 causes a competition for capture and therefore reduces actin ide burning and transmutation. However, the prior thesis results show a la rge under prediction of nuclide concentration while the new results show a large over prediction. These discrepancies may be related to the treatment of incomplete cross sections, or limitations of the MONTEBUR NS 2.0 tally routines, or limitations of the ORIGEN2.2 algorithm. Table 5-14. MONTEBURNS* Phase IB Benchmar k results and comparison of Case A (27.35 MWD/MTU). Isotope Experiment Value (mg/g UO2) % Error MONTEBURNS OLD** % Error MONTEBURNS NEW** % Error MCNPXc** 234U 1.422E-01 -2.45 2.77 -11.13 235U 8.339 -4.32 9.06 -1.54 236U 3.173 2.09 3.07 1.06 238U 8.375E+02 -0.5 0.31 -0.59 237Np 2.794E-01 6.65 6.11 4.25 238Pu 9.445E-02 -6.12 25.03 -6.67 239Pu 3.891 -7.50 7.34 -8.74 240Pu 1.640 -2.00 10.29 -4.60 241Pu 6.615E-01 -2.72 9.77 -2.90 242Pu 3.069E-01 6.65 12.07 6.36 133Cs 8.259E-01 1.91 -6.92% -2.83 135Cs 3.620E-01 4.46 -1.40% 0.54 143Nd 5.840E-01 -0.36 -5.51% -4.73 150Sm 1.854E-01 -13.22 -14.51% -10.44 152Sm 8.124E-02 -1.35 -6.39% -6.62 153Eu 7.257E-02 5.11 -5.71% -8.13 *MONTEBURNS OLD = MONTEBURNS from prior thesis; 75 MONTEBURNS NEW =MOTNEBURNS 2.0 with higher fidelity model ***(Calculated/Measured-1)*100 The results do show major cause for concern, because a higher fidelity model should show better results. It seems that modeling more of the available fission products ac tually hurt the calculation; however, further test ing is required in order to de termine the root cause of the miscalculation. New MONTEBURNS versions have been developed since MONTEBURNS 2.0

PAGE 145

145 and therefore further testing will be required in or der to determine the origin of this cause and whether or not the cause is problem dependent. Table 5-15. MONTEBURNS* Phase IB Benchmar k results and comparison of Case A (37.38 MWD/MTU). Isotope Experiment Value (mg/g UO2) % Error MONTEBURNS OLD** % Error MONTEBURNS NEW** % Error MCNPXc** 234U 1.164E-01 -5.05 -11 -16.88 235U 4.822 -9.66 -31.71 -6.73 236U 3.619 2.68 -11.43 2.51 238U 8.303E+02 -0.28 1.25 -0.28 237Np 4.032E-01 14.38 -36.40 13.26 238Pu 1.750E-01 -3.84 -44.37 -7.55 239Pu 3.987 -7.46 3.95 -8.50 240Pu 2.120 -2.47 -18.18 -5.31 241Pu 8.699E-01 -4.05 -21.07 -3.65 242Pu 5.936E-01 7.41 -77.15 3.04 133Cs 1.060 2.55 20.93 -6.24 135Cs 3.930E-01 4.79 7.42 -5.17 143Nd 6.750E-01 -0.76 11.81 -8.90 150Sm 2.521E-01 -8.96 -30 -14.72 152Sm 9.892E-02 -0.20 27 -8.45 153Eu 1.035E-01 13.17 -30 -8.81 *MONTEBURNS OLD = MONTEBURNS from prior thesis; 75 MONTEBURNS NEW =MOTNEBURNS 2.0 with higher fidelity model ***(Calculated/Measured-1)*100 H. B. Robinson Benchmark Specification The Burnup Credit Criticality Safety expert group offers the Spent Fuel Isotopic Composition database containing measured nuclide composition data for 14 LWRs in order to help validate burn-up codes.97 The H. B. Robinson benchmark is one of the 14 available LWR benchmarks. Post irradiation examination measurem ents exist for select fuel pins within one 15 X 15 Westinghouse fuel assembly from the H. B. Robinson Unit 2 reactor.53 For a specific fuel pin within the assembly, radiochemical analysis was completed at several axial levels corresponding to several different burnups. The benchmark was chosen for this study because full assembly geometry specification was available, and the full modeled assembly was

PAGE 146

146 considered to be in a core location assume d to possess a reasonably flat enough power distribution to assume zero net l eakage at the physical boundary of the assembly. Because zero net leakage was assumed at the assembly bounda ry, it was also further assumed that the assembly could be modeled as infinitely refl ected at the geometry boundary. Because leakage from assembly to assembly is affected by th e ultimate core boundaries, power shaping and poison positioning, this assumption is not exactly true. However, the assumption was deemed acceptable enough to model the assembly because the power distribution across the assembly region was assumed flat enough. Figure 5-2. Infinitely reflected pin cell geometry. Oak Ridge National Laboratory had also co mpleted a verification study of the SCALE package using this benchmark. This verifica tion study offered knowledge of how to setup the benchmark for direct comparison with the SCA LE package. The Oak Ri dge National Laboratory report, ORNL/TM-12667.28, was therefore us ed to determine the benchmark setup.53 The H. B. Robinson Fuel assembly was modeled using ei ghth bundle symmetry, and took advantage of repeated structure modeling available since MC NPX 2.6.D. The exact modeled geometry is Analyzed Fuel Burnable Poison Instrument Tube Guide Tube

PAGE 147

147 displayed in Figure 5-2. Ta ble 5-16 displays the modeled assembly design data for the benchmark. The material compositions for the fuel, clad, guide tube, and burnable poison rod materials are specified in the Oak Ridge National Laboratory report, ORNL/TM-12667.28.53 Table 5-16. Fuel assembly design da ta for H. B. Robinson Benchmark*. Parameter Data Assembly general data Designer Westinghouse Lattice 15 15 Number of fuel rods 204 Number of guide tubes 8 Number of burnable poison rods 12 Number of instrument tubes 1 Assembly pitch 21.50 cm Assembly fuel 443.7 kg U Fuel rod data Type of fuel pellet UO2 Enrichment: wt % 235U 2.561 wt % 234U 0.023 wt % 236U 0.013 Pellet density 9.44 g/cm3 Rod Pitch 1.43 cm Rod OD 1.0719 cm Rod ID 0.9484 cm Pellet diameter 0.9242 cm Active fuel length 365.76 cm Clad temperature 595 K Clad material Zircaloy-4 Guide tube data Inner radius 0.6502 cm Outer radius 0.6934 cm Material Zircaloy-4 Instrument tube data Inner radius 0.6502 cm Outer radius 0.6934 cm Material Zircaloy-4 Burnable poison rod data Air OD 0.5677 cm SS304 OD 0.6007 cm Air OD 0.6172 cm Borosilicate glass OD 1.0058 cm Air OD 1.0173 cm SS304 OD 1.1151 cm *Data taken from reference53

PAGE 148

148 Each steady-state Monte Carlo calculation wa s completed running 4000 histories per cycle for 155 cycles skipping the first 25 cycles. Initia l source particles were started in each of the fissile fuel locations. Though the geometry wa s modeled in 3-dimensions, only 1 axial region was utilized; therefore the model does not account for axial variati on of the power distribution as a function of burnup. However, each fuel pin, wi thin an eighth assembly symmetric section, used a separate burn material in order to account for radial variation of the interaction rates. Four different cases, corresponding to four axial locations, expe riencing four separate burnups, were examined as part of the benchmark. Case s A-D represents the different burnup cases from the benchmark: (1) Case A = 16.02 GWD/MTU; (2) Case B = 23.8 GWD/MTU; (3) Case C = 28.47 GWD/MTU; (4) Case D = 31.66 GWD/MTU. The measured burnup given for case A did not correspond to the final time interval specified in the Oak Ridge National Laboratory Report. For the Case A calculation, the final operating in terval was changed from 156 days to 165 days in order to correspond to the expect ed burnup exhibited in the assembly.54 Table 5-17 and Table 5-18 display the operating intervals, time steps used for each interval, corresponding down times, and cycle dependent soluble boron concentrations for Case A-D. Table 5-17. Operating data for H. B. Robinson fuel assembly. Cycle 1 2 3 4 Operating Interval (days) 243.5 243.5 156 156 Interval Time Step Durations (days) 1, 15, 20, 30, 37, 44, 46 1, 15, 20, 30, 37, 44, 46 1, 15, 20, 30, 40, 50 1, 15, 20, 30, 40, 50* Downtime (days) 40 64 39 --** Average Soluble Boron Concentration (ppm) 625.5 247.5 652.5 247.5 This value was set to 59 for Case A. ** This value was set either to 3936 for Cases A-B or 3637 for Cases C-D.

PAGE 149

149 Table 5-18. Operating data for H. B. Robinson fuel assembly Case Total Burnup (GWD/MTU) Moderator Temperature (K) Moderator Density (g/cm3) Fuel Temperature (K) A 16.02 559 0.7544 743 B 23.81 559 0.7538 830 C 28.47 576 0.7208 883 D 31.66 576 0.7135 923 H. B. Robinson Benchmark Results The results of the benchmark were compared to the results from SCALE as specified in ORNL/TM-12667.53 The SCALE sequence used in the calculations was SAS2H/ORIGEN-S analysis sequence and 27-group cross-sec tion library (27BURNUPLIB) of SCALE-4.53 Two sets of analysis were completed. The first set used the original MCNPX, and the next set used the modified MCNPXc. The results of the be nchmark using MCNPX are displayed in Table 519, and the results of the benchmark usi ng MCNPXc are displayed in Table 5-20. Table 5-19. The H. B. Robinson Benchmark results* using MCNPX compared with SCALE. Case A 16.02 GWD/MTU Case B 23.8 GWD/MTU Case C 28.47 GWD/MTU Case D 31.66 GWD/MTU Isotope MCNPX SCALE MCNPXSCALE MCNPXSCALE MCNPX SCALE 235U 0.18 0.60 -3.11 1.40 -11.8 -4.90 -9.66 0.10 236U -1.62 -1.5 -1.05 -2.20 3.72 2.20 1.18 -0.50 238U -0.12 0.10 -0.60 -0.60 0.47 0.50 -0.73 -0.20 237Np -10.57 1.50 -8.09 0.90 -14.72 -6.50 -10.69 -0.40 238Pu -6.35 7.00 -6.41 7.70 -9.22 5.30 -8.66 8.20 239Pu -3.19 -1.5 -4.75 -4.20 -5.42 -4.90 -6.52 -3.70 240Pu -1.23 5.90 -1.45 6.00 -11.03 0.50 -8.79 5.4 241Pu -8.23 6.00 -6.30 5.50 2.43 14.30 3.08 11.10 99Tc 7.74 12.40 4.20 8.60 9.58 14.60 5.53 11.70 137Cs -3.10 0.20 -3.10 -0.80 -0.38 3.90 -3.09 1.2 *Percent difference from measured experiment al results [(Calculat ed/Measured -1)*100]. The results displayed in 5-18 e xhibits the same trend in burnup and predictive capability as the MCNPX results for the OECD/NEA Burnup Cr edit Phase IB. As burnup increases, nuclide predictive capability gets wors e and the calculated values seem not as close to the actual

PAGE 150

150 measured calculation. For example, the calculat ion percent difference from experiment for most actinides for Cases A-B was <8%; however, at higher burnups, Cases C-D, the calculation percent difference from experiment grew to 14%. Though the SCALE values and the MCNPX values seemed relatively close in magnitude, fu rther calculations were also completed using MCNPXc to see the magnitude of improvement. In Table 5-19, the results show significant improvement in actinide prediction for uran ium and plutonium isotopes when utilizing MCNPXc. Table 5-20. The H. B. Robinson Benchmark re sults* using MCNPXc compared with SCALE. Case A 16.02 GWD/MTU Case B 23.8 GWD/MTU Case C 28.47 GWD/MTU Case D 31.66 GWD/MTU Isotope MCNPXc SCALE MCNPXcSCALEMCNPXcSCALE MCNPXc SCALE235U 0.47 0.60 -0.58 1.40 -8.19 -4.90 -5.35 0.10 236U -1.81 -1.5 -1.90 -2.20 2.77 2.20 0.57 -0.50 238U 0.12 0.10 -0.54 -0.60 0.53 0.50 -0.73 -0.20 237Np -6.63 1.50 -7.31 0.90 -11.21 -6.50 -7.41 -0.40 238Pu -4.125 7.00 -3.86 7.70 -3.29 5.30 -1.89 8.20 239Pu -0.005 -1.5 -0.37 -4.20 -0.38 -4.90 -0.52 -3.70 240Pu 4.013 5.90 0.59 6.00 -6.94 0.50 -3.66 5.4 241Pu 2.4 6.00 2.82 5.50 1.88 14.30 1.36 11.10 99Tc 10.815 12.40 6.76 8.60 12.13 14.60 8.49 11.70 137Cs -1.152 0.20 -1.88 -0.80 0.67 3.90 -1.81 1.2 *Percent difference from measured experiment al results [(Calculat ed/Measured -1)*100]. The possible reasons for error in the calcula tion are similar to those mentioned in the OECD/NEA Burnup Credit Phase IB benchmark. Errors in nuclear data, treatment of normalization parameters, statistical variance in reaction rates, and not modeling a critical spectrum all contribute to errors in the calculation. It is important to note that the same time step durations were also used for each of the cases. Therefore each case was modeled using different size burnup steps. The coefficients for the de pletion calculation were therefore updated over longer burnup steps for the higher burnup cases. Th erefore these results al so suggest that the longer burnup step durations used for the higher burnup cases may have been too large to assume

PAGE 151

151 that significant change in the flux had not occurred over the time step. As mentioned previously, selecting adequate time steps depends upon th e buildup and depletion of highly absorbing isotopes that affect the neutron spectrum in the mate rial. Therefore, as is displayed in the results, the prior statement may be furthe r translated to mean that ade quate time selection is not just related to resident time but also related to the combination of resident time and operating power over that time, or burnup. However, for Cases AB, the percent difference in actinide prediction from measured experiment is <4% except fo r Np-237 which is superior to the SCALE calculation. These results ther efore suggest that MCNPX is very capable of modeling full assembly behavior and accurately predicting nuclide buildup as long as adequate burnup step durations are maintained. Another important source of erro r to consider is that the m easured data also possess error as a function of the precision and accuracy of the measurement process. Unfortunately, the measurement error for this benchmark does not exist. The Spent Fuel Isotopic Composition database posted on the NEA website also does no t contain measurement error for any of the 14 available experimental benchmarks.97 Most of the reports used to generate the database do contain the methods of radiochemical analysis ap plied to extract the expe rimental data; however, these reports do not contain a qua ntification of the error associated in obtaining the measured data. This problem of not listing experimental error therefore must be addressed in future benchmarks in order to unders tand the validity of calculated results when compared to the measured data. OECD/NEA Burnup Credit Phase IVB Benchmark Specification Both the OECD/NEA Burnup Credit Phase IB benchmark and H. B. Robinson benchmark modeled typical PWR uranium dioxide fuel ope rating conditions. In order to prove the versatility of the of the MCNP X depletion methodology for different types of reactors, another

PAGE 152

152 operating scenario, besides a typical PWR ur anium dioxide fuel operating strategy, was investigated. The OECD/NEA Burnup Credit Phas e IVB benchmark offers a suite of benchmark calculations for mixed oxide fuel (MOX).54 MOX assembly modeling is important to a closed reactor fuel cycle because the recycled assemb ly will contain a combination of uranium and plutonium (and possibly other actinides) at BOL for the recycled assembly.9 Calculations of the OECD/NEA Burnup Credit Phase IVB benchmark have been completed with many different depletion codes, and therefore an adequate comp arison database is available for the benchmark.54 Table 5-21. Fuel assembly design data for OECD/NEA Burnup Credit Phase IVB Benchmark. Parameter Data Assembly general data Lattice 17 17 Number of fuel rods 264 Number of guide tubes 24 Number of instrument tubes 1 Fuel rod data Type of fuel pellet MOX Rod pitch 1.26 cm Clad thickness 0.065 cm (no gap between fuel and clad) Pellet diameter 0.410 cm Active fuel length 365.76 cm Fuel temperature 900 K Clad temperature 620 K Clad material Zircaloy-2 Guide and Instrument tube data Inner radius 0.571 cm Outer radius 0.613 cm Material Zircaloy-2 Modeling a single infinitely re flected MOX fuel assembly doe s not adequately represent the true reactor operating conditions of the a ssembly. Many neutronic characteristics of the MOX assembly limit the assemblys placement to certa in patterns within cer tain sections of the reactor core. The higher absorption cross sectiond of the plutonium isotopes, limits the negative reactivity worth of control material such as dissolved boron or cont rol blades, in the presence of the MOX assembly. The high fission cross secti on of Pu-239 and large in itial loading of Pu-239

PAGE 153

153 produces large fission peaks; therefore to accommoda te this affect and flatten the power peak, MOX assemblies are placed away from water gap regions.9 Finally, the fact that Pu-239 possesses a smaller delayed neutron fraction ( = 0.0021) and shorter neutron lifetime (l = 27 s), than U-235 ( =0.0065, l = 47 s), thus limiting controllability of kinetic response, limits the amount of MOX assemblies pl aced in the reactor core.9 Countries already implementing the MOX fuel cycle, such as France, limit the am ount of MOX loading to no more than 30%, and never load MOX assemblies surrounded by other MOX assemblies.9 However, since the H. B. Robinson benchmark only offered code-to-code co mparison with SCALE, the author decided to use the OECD/NEA Burnup Credit Phase IVB benchmark as a comparison tool, knowing that modeling a infinitely reflected MOX assembly w ould not model true reactor behavior, because the benchmark offered data from a variety of codes for adequate code-to-code comparison for MOX assembly modeling strategy. Table 5-22. Initial MOX fuel compositi on for Case A for Phase IVB Benchmark. Values Expressed in (atoms/b-cm) Nuclide High Enrichment Medium Enrichment Low Enrichment 234U 2.57E-07 2.64E-07 2.68E-07 235U 5.38E-05 5.53E-05 5.60E-05 238U 2.12E-02 2.18E-02 2.21E-02 238Pu 5.17E-05 3.61E-05 2.85E-05 239Pu 1.13E-03 7.87E-04 6.20E-04 240Pu 5.35E-04 3.74E-04 2.95E-04 241Pu 1.94E-04 1.36E-04 1.07E-04 242Pu 1.46E-04 1.02E-04 8.06E-05 16O 4.66E-02 4.66E-02 4.65E-02 Two types of infinitely reflected 17 X 17 M OX fuel assemblies were modeled. Each modeled case assembly was composed of MOX fuel pellets encased in Zircaloy-2 clad fuel pins along with Zircaloy-2 guide tubes. The geometry specifications for both cases are given in Table 5-21. The Case A model included a M OX assembly possessing plutonium nuclide concentrations consistent with typical concen trations expected from a first recycle PWR

PAGE 154

154 assembly. The Case B model included a MOX assembly possessing plutonium nuclide concentrations consistent with typical concen trations expected from weapons disposition material. Table 5-23. Initial MOX fuel compositi on for Case B for Phase IVB Benchmark. Values Expressed in (atoms/b-cm) Nuclide High Enrichment Medium Enrichment Low Enrichment 234U 2.69E-07 2.73E-07 2.75E-07 235U 5.63E-05 5.71E-05 5.74E-05 238U 2.22E-02 2.25E-02 2.26E-02 238Pu 5.09E-07 3.56E-07 2.81E-07 239Pu 9.49E-04 6.64E-04 5.23E-04 240Pu 6.06E-05 4.24E-05 3.34E-05 241Pu 3.02E-06 2.11E-06 1.66E-06 242Pu 5.01E-07 3.50E-07 2.76E-07 16O 4.65E-02 4.65E-02 4.65E-02 For each case, three types of plutonium concentr ation loadings were used for the fuel pins within each case. Table 5-22 lists the fuel compositions for Case A, and Table 5-23 lists the fuel composition for Case B. Table 5-24 lists the bo rated water composition a nd Table 5-25 lists the Zircaloy-2 composition used for the fuel clad and guide tube s. Figure 5-3 displays the distribution of the different fuel pins throughout the fuel assemb ly for both cases. Each steadystate reaction rate case was comp leted using 4000 particles per cycle skipping the first 25 cycles for 155 cycles. Initial so urce locations were placed within each fissile cell location. Case A was burned at 17.896 MW, and Case B was burned at 17.862 MW. The operating data for the assembly is listed in Table 5-26. Table 5-24. Borated water composition for Phase IVB Benchmark (660ppm boron, density 0.7245 g/cm3). Nuclide Number Density (atoms/b-cm) 1H 4.8414E-02 16O 2.4213E-02 10B 4.7896E-06 11B 1.9424E-05

PAGE 155

155 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 1234567891011121314151617 Guide tube Instrument tube Keys Low enriched MOX fuel pin Medium enriched MOX fuel pin High enriched MOX fuel pin Figure 5-3. MOX fuel enrichment map for Phase IVB Benchmark.

PAGE 156

156 Table 5-25. Zicaloy-2 composition used for cl ad and guide tube material for Phase IVB Benchmark (density 5.8736 g/cm3). Nuclide (Atoms/b-cm) 50Sn 2.9656E-06 52Sn 5.7189E-05 53Sn 6.4848E-06 54Sn 1.6142E-06 54Fe 7.8122E-06 56Fe 1.2245E-04 57Fe 2.8278E-06 58Fe 3.7633E-07 90Zr 1.9889E-02 91Zr 4.3373E-03 92Zr 6.6297E-03 94Zr 6.7186E-03 96Zr 1.0824E-03 Table 5-26. Operating data for Phase IVB Benchmark. Cycle 1 2 3 Operating Interval (days) 420 420 420 Interval Time Step Durations (days) 1, 15, 20, 30, 50, 60, 60, 60, 60, 64 1, 15, 20, 30, 50, 60, 60, 60, 60, 64 1, 15, 20, 30, 50, 60, 60, 60, 60, 64 Downtime (days) 30 30 0, 1826.25 OECD/NEA Burnup Credit Phas e IVB Benchmark Results The OECD/NEA Burnup Credit Phase IVB benchmark offers cycle by cycle nuclide buildup and depletion information for a plethora of actinides and fission products. Table 5-27, Table 5-28, Table 5-29 and Table 5-30 present the MCNPX EOC 1, 2, 3 and after 5 years of cooling results for Case A as compared to the range of values achieved from other depletion codes in the benchmark. Table 5-31, Table 5-32, Table 5-33 and Ta ble 5-34 present the MCNPX EOC 1, 2, 3 and after 5 ye ars of cooling results for Case B as compared to the range of values achieved from other depletion codes in the benchmark. Due to the success in nuclide prediction of MCNPXc, each of these cases were run using MCNPXc.

PAGE 157

157 Table 5-27. EOC 1 results for Case A of the Phase IVB Benchmark. Nuclide MCNPX (atoms/b-cm) Range of Values from Other Depletion Codes (atoms/b-cm) 234U 5.8483E-07 2.1279E-07 to 5.8216E-07 235U 4.3753E-05 4.3720E-05 to 4.3925E-05 236U 2.6406E-06 2.4858E-06 to 2.6190E-06 238U 2.1134E-02 2.1155E-02 to 2.1163E-02 238Pu 4.1533E-05 4.0713E-05 to 4.1630E-05 239Pu 8.0377E-04 7.9700E-04 to 8.1275E-04 240Pu 4.7370E-04 4.7109E-04 to 4.8300E-04 241Pu 2.1855E-04 2.1520E-04 to 2.2011E-04 242Pu 1.6393E-04 1.3389E-04 to 1.3785E-04 237Np 1.6176E-06 1.0400E-07 to 1.6560E-06 241Am 9.4895E-06 8.4140E-06 to 8.8762E-06 243Am 1.8069E-05 1.5613E-05 to 1.8260E-05 242Cm 9.5381E-07 8.7154E-07 to 9.8523E-07 243Cm 1.3915E-08 1.1641E-08 to 1.4560E-08 244Cm 5.3807E-06 2.6688E-06 to 3.9610E-06 245Cm 2.8581E-07 1.2915E-07 to 1.9877E-07 95Mo 1.2829E-05 1.1690E-05 to 1.7651E-05 99Tc 2.1958E-05 2.1756E-05 to 2.3060E-05 101Ru 2.2331E-05 2.1640E-05 to 2.3054E-05 103Rh 2.0106E-06 1.9630E-05 to 2.2642E-05 109Ag 5.0294E-06 4.6603E-06 to 5.9460E-06 133Cs 2.4605E-05 2.3463E-05 to 2.4597E-05 143Nd 1.5949E-05 1.5350E-05 to 1.6594E-05 145Nd 1.1238E-05 1.1183E-05 to 1.1518E-05 147Sm 8.0947E-07 7.8960E-07 to 8.2700E-07 149Sm 4.2895E-07 3.7767E-07 to 4.3945E-07 150Sm 4.7391E-06 4.7807E-06 to 5.1050E-06 151Sm 1.3397E-06 1.3420E-06 to 1.5281E-06 152Sm 2.7837E-06 2.7790E-06 to 3.1530E-06 153Eu 2.0936E-06 2.0380E-06 to 2.2411E-06 155Gd 8.6977E-09 4.3940E-09 to 1.6769E-08

PAGE 158

158 Table 5-28. EOC 2 results for Case A of the Phase IVB Benchmark. Nuclide MCNPX (atoms/b-cm) Range of Values from Other Depletion Codes (atoms/b-cm) 234U 8.0735E-07 1.7413E-07 to 8.4354E-07 235U 3.4593E-05 3.4500E-05 to 3.4893E-05 236U 4.6979E-06 4.4215E-06 to 4.6370E-06 238U 2.0896E-02 2.0910E-02 to 2.0927E-02 238Pu 3.9782E-05 3.5013E-05 to 3.9340E-05 239Pu 6.3728E-04 6.2980E-04 to 6.5524E-04 240Pu 4.4837E-04 4.4499E-04 to 4.6350E-04 241Pu 2.3731E-04 2.3210E-04 to 2.3979E-04 242Pu 1.7027E-04 1.4147E-04 to 1.4982E-04 237Np 3.1341E-06 3.6114E-07 to 3.2139E-06 241Am 1.5907E-05 1.4680E-05 to 1.5982E-05 243Am 2.9191E-05 2.7797E-05 to 3.1620E-05 242Cm 2.5536E-06 2.3253E-06 to 2.6204E-06 243Cm 7.2443E-08 5.8709E-08 to 7.0660E-08 244Cm 1.7168E-05 9.2778E-06 to 1.3050E-05 245Cm 1.7035E-06 8.1985E-07 to 1.2348E-06 95Mo 2.9454E-05 2.8330E-05 to 3.4297E-05 99Tc 4.1886E-05 4.1811E-05 to 4.4400E-05 101Ru 4.3865E-05 4.2320E-05 to 4.5186E-05 103Rh 4.0075E-05 3.7469E-05 to 4.0914E-05 109Ag 9.0959E-06 8.0989E-06 to 1.0640E-05 133Cs 4.6303E-05 4.4403E-05 to 4.6602E-05 143Nd 3.0328E-05 2.9970E-05 to 3.1809E-05 145Nd 2.1678E-05 2.1322E-05 to 2.2191E-05 147Sm 2.1430E-06 2.4130E-06 to 2.4742E-06 149Sm 4.0373E-07 3.5930E-07 to 4.1990E-07 150Sm 1.0296E-05 1.0197E-05 to 1.1350E-05 151Sm 1.5541E-06 1.5306E-06 to 1.8870E-06 152Sm 4.9712E-06 5.0468E-06 to 5.9850E-06 153Eu 4.9325E-06 5.0480E-06 to 5.4745E-06 155Gd 1.6016E-08 7.9700E-09 to 2.8935E-08

PAGE 159

159 Table 5-29. EOC 3 results for Case A of the Phase IVB Benchmark. Nuclide MCNPX (atoms/b-cm) Range of Values from Other Depletion Codes (atoms/b-cm) 234U 9.4561E-07 1.4169E-07 to 1.0509E-06 235U 2.6715E-05 2.6596E-05 to 2.7129E-05 236U 6.2239E-06 5.8461E-06 to 6.1140E-06 238U 2.0648E-02 2.0657E-02 to 2.0678E-02 238Pu 4.0397E-05 2.9577E-05 to 3.9688E-05 239Pu 5.0933E-04 5.0368E-04 to 5.3435E-04 240Pu 4.1054E-04 4.0836E-04 to 4.2870E-04 241Pu 2.3785E-04 2.3107E-04 to 2.4058E-04 242Pu 1.8086E-04 1.5231E-04 to 1.6558E-04 237Np 4.3966E-06 6.9348E-07 to 4.5652E-06 241Am 1.8127E-05 1.7960E-05 to 2.0066E-05 243Am 3.6401E-05 3.6793E-05 to 4.2000E-05 242Cm 4.3963E-06 3.5650E-06 to 4.0436E-06 243Cm 1.5834E-07 1.2551E-07 to 1.4820E-07 244Cm 3.1800E-05 1.8435E-05 to 2.4950E-05 245Cm 4.1506E-06 2.1928E-06 to 3.2335E-06 95Mo 4.3988E-05 4.3990E-05 to 4.9929E-05 99Tc 5.9641E-05 6.0068E-05 to 6.3850E-05 101Ru 6.4487E-05 6.1990E-05 to 6.6364E-05 103Rh 5.3043E-05 5.1041E-05 to 5.5357E-05 109Ag 1.2395E-05 1.0524E-05 to 1.4480E-05 133Cs 6.4706E-05 6.2829E-05 to 6.6172E-05 143Nd 4.2327E-05 4.2830E-05 to 4.5385E-05 145Nd 3.1270E-05 3.0386E-05 to 3.2104E-05 147Sm 3.1679E-06 4.0280E-06 to 4.1579E-06 149Sm 3.2126E-07 3.2302E-07 to 3.8520E-07 150Sm 1.5873E-05 1.5410E-05 to 1.7660E-05 151Sm 1.5666E-06 1.5340E-06 to 2.0110E-06 152Sm 6.3075E-06 6.3335E-06 to 7.9580E-06 153Eu 7.5412E-06 8.1310E-06 to 8.6655E-06 155Gd 1.9499E-08 1.3920E-08 to 4.2299E-08

PAGE 160

160 Table 5-30. Five year cooling time results for Case A of the Phase IVB Benchmark. Nuclide MCNPX (atoms/b-cm) Range of Values from Other Depletion Codes (atoms/b-cm) 234U 2.6607E-06 1.4169E-07 to 2.7220E-06 235U 2.6785E-05 2.6668E-05 to 2.7206E-05 236U 6.4390E-06 5.8461E-06 to 6.2030E-06 238U 2.0648E-02 2.0657E-02 to 2.0678E-02 238Pu 4.3111E-05 2.8432E-05 to 4.1989E-05 239Pu 5.1081E-04 5.0542E-04 to 5.3613E-04 240Pu 4.1585E-04 4.1217E-04 to 4.3260E-04 241Pu 1.8680E-04 1.8167E-04 to 1.8891E-04 242Pu 1.8086E-04 1.5231E-04 to 1.6558E-04 237Np 4.8116E-06 7.0645E-07 to 4.9848E-06 241Am 6.8783E-05 6.6320E-05 to 7.0845E-05 243Am 3.6406E-05 3.6793E-05 to 3.9880E-05 242Cm 2.5594E-09 1.6887E-09 to 5.0942E-09 243Cm 1.4022E-07 1.1143E-07 to 1.2852E-07 244Cm 2.6279E-05 1.5224E-05 to 1.9671E-05 245Cm 4.1491E-06 2.1928E-06 to 3.2321E-06 95Mo 4.9799E-05 4.9929E-05 to 5.0765E-05 99Tc 5.9856E-05 6.0280E-05 to 6.4090E-05 101Ru 6.4487E-05 6.3251E-05 to 6.6366E-05 103Rh 5.6463E-05 5.1041E-05 to 5.8208E-05 109Ag 1.2416E-05 1.0524E-05 to 1.4490E-05 133Cs 6.5257E-05 6.2829E-05 to 6.6172E-05 143Nd 4.3169E-05 4.4116E-05 to 4.5385E-05 145Nd 3.1278E-05 3.0386E-05 to 3.2104E-05 147Sm 7.8905E-06 9.9452E-06 to 1.0734E-05 149Sm 3.6852E-07 3.7494E-07 to 4.2796E-07 150Sm 1.5873E-05 1.5410E-05 to 1.6679E-05 151Sm 1.5196E-06 1.4779E-06 to 1.8916E-06 152Sm 6.3090E-06 6.3335E-06 to 7.9580E-06 153Eu 7.5868E-06 8.1650E-06 to 8.6716E-06 155Gd 3.6011E-07 3.3400E-07 to 1.0081E-06

PAGE 161

161 Table 5-31. EOC 1 results for Case B of the Phase IVB Benchmark. Nuclide MCNPX (atoms/b-cm) Range of Values from Other Depletion Codes (atoms/b-cm) 234U 2.1859E-07 2.1683E-07 to 2.2681E-07 235U 4.2074E-05 4.1730E-05 to 4.2085E-05 236U 3.1619E-06 3.0287E-06 to 3.1900E-06 238U 2.2041E-02 2.2000E-02 to 2.2049E-02 238Pu 1.1757E-06 3.7913E-07 to 7.1826E-07 239Pu 5.6960E-04 5.6280E-04 to 5.7659E-04 240Pu 1.5241E-04 1.4600E-04 to 1.5008E-04 241Pu 6.0895E-05 5.6781E-05 to 6.3800E-05 242Pu 7.1572E-06 5.3572E-06 to 6.1170E-06 237Np 1.5688E-06 1.3364E-07 to 1.6488E-06 241Am 1.6417E-06 1.3107E-06 to 1.4360E-06 243Am 9.6882E-07 6.2689E-07 to 8.0100E-07 242Cm 1.8012E-07 1.5269E-07 to 1.7950E-07 243Cm 2.0951E-09 1.6814E-09 to 2.0300E-09 244Cm 2.4339E-07 9.3251E-08 to 1.3100E-07 245Cm 1.1257E-08 3.4090E-09 to 4.6024E-09 95Mo 1.3153E-05 1.2181E-05 to 1.8204E-05 99Tc 2.1838E-05 2.1900E-05 to 2.2950E-05 101Ru 2.1979E-05 2.1769E-05 to 2.3058E-05 103Rh 2.0336E-05 1.8759E-05 to 2.1765E-05 109Ag 4.3321E-06 3.6778E-06 to 5.5840E-06 133Cs 2.4560E-05 2.3439E-05 to 2.4574E-05 143Nd 1.5453E-05 1.5100E-05 to 1.6346E-05 145Nd 1.0969E-05 1.1117E-05 to 1.1454E-05 147Sm 7.6879E-07 7.6500E-07 to 7.8701E-07 149Sm 2.3529E-07 1.9930E-07 to 2.2432E-07 150Sm 4.7705E-06 4.9051E-06 to 5.2700E-06 151Sm 8.6623E-07 8.8308E-07 to 1.0200E-06 152Sm 2.8949E-06 2.9600E-06 to 3.3370E-06 153Eu 2.0836E-06 2.0570E-06 to 2.2659E-06 155Gd 4.5104E-09 1.8000E-09 to 7.6886E-09

PAGE 162

162 Table 5-32. EOC 2 results for Case B of the Phase IVB Benchmark. Nuclide MCNPX (atoms/b-cm) Range of Values from Other Depletion Codes (atoms/b-cm) 234U 1.8049E-07 1.7098E-07 to 1.9333E-07 235U 2.9106E-05 2.8760E-05 to 2.9257E-05 236U 5.5884E-06 5.3234E-06 to 5.5600E-06 238U 2.1773E-02 2.1767E-02 to 2.1800E-02 238Pu 2.6720E-06 2.9817E-07 to 2.0587E-06 239Pu 3.7031E-04 3.6608E-04 to 3.8300E-04 240Pu 1.8694E-04 1.8070E-04 to 1.8624E-04 241Pu 9.7981E-05 9.4093E-05 to 1.0080E-04 242Pu 2.2218E-05 1.9550E-05 to 2.1840E-05 237Np 3.0365E-06 4.5945E-07 to 3.1897E-06 241Am 4.1152E-06 3.6692E-06 to 3.9280E-06 243Am 4.3621E-06 3.6612E-06 to 4.7300E-06 242Cm 8.9769E-07 7.7886E-07 to 8.9591E-07 243Cm 4.3621E-06 1.6565E-08 to 2.0900E-08 244Cm 1.8948E-06 1.0277E-06 to 1.4100E-06 245Cm 1.4495E-07 6.0866E-08 to 8.6913E-08 95Mo 2.9951E-05 2.9300E-05 to 3.5096E-05 99Tc 4.1181E-05 4.1730E-05 to 4.3780E-05 101Ru 4.3022E-05 4.2699E-05 to 4.5057E-05 103Rh 3.5805E-05 3.3166E-05 to 3.7099E-05 109Ag 7.7718E-06 6.5223E-06 to 9.7410E-06 133Cs 4.5646E-05 4.3917E-05 to 4.6063E-05 143Nd 2.8262E-05 2.8500E-05 to 3.0200E-05 145Nd 2.0952E-05 2.1048E-05 to 2.1882E-05 147Sm 1.9461E-06 2.2520E-06 to 2.2924E-06 149Sm 2.0460E-07 1.7336E-07 to 1.9700E-07 150Sm 1.0170E-05 1.0500E-05 to 1.1600E-05 151Sm 8.4275E-07 8.4937E-07 to 1.0900E-06 152Sm 4.7605E-06 4.8514E-06 to 5.9030E-06 153Eu 4.8992E-06 5.1650E-06 to 5.5763E-06 155Gd 9.7191E-09 3.8100E-09 to 1.1850E-08

PAGE 163

163 Table 5-33. EOC 3 results for Case A of the Phase IVB Benchmark. Nuclide MCNPX (atoms/b-cm) Range of Values from Other Depletion Codes (atoms/b-cm) 234U 1.1520E-13 1.3174E-07 to 1.7954E-07 235U 1.8445E-05 1.8119E-05 to 1.8791E-05 236U 7.1794E-06 6.8091E-06 to 7.0700E-06 238U 2.1481E-02 2.1467E-02 to 2.1500E-02 238Pu 5.5067E-06 2.2123E-07 to 4.7927E-06 239Pu 2.5259E-04 2.4767E-04 to 2.7200E-04 240Pu 1.8124E-04 1.7470E-04 to 1.8437E-04 241Pu 1.0690E-04 1.0323E-04 to 1.0850E-04 242Pu 4.3597E-05 3.9815E-05 to 4.3080E-05 237Np 4.1982E-06 8.5608E-07 to 4.4424E-06 241Am 4.9089E-06 4.9773E-06 to 5.4746E-06 243Am 1.0121E-05 9.3994E-06 to 1.1700E-05 242Cm 1.9925E-06 1.5821E-06 to 1.8030E-06 243Cm 6.2628E-08 4.8232E-08 to 5.9200E-08 244Cm 6.7113E-06 4.1608E-06 to 5.4200E-06 245Cm 6.6887E-07 3.2603E-07 to 4.7115E-07 95Mo 4.4224E-05 4.5000E-05 to 5.0570E-05 99Tc 5.7742E-05 5.9218E-05 to 6.2220E-05 101Ru 6.2895E-05 6.2300E-05 to 6.5908E-05 103Rh 4.3988E-05 4.2315E-05 to 4.7011E-05 109Ag 1.0536E-05 8.5551E-06 to 1.2920E-05 133Cs 6.2697E-05 6.1230E-05 to 6.4442E-05 143Nd 3.7203E-05 3.8700E-05 to 4.0746E-05 145Nd 2.9770E-05 2.9641E-05 to 3.1166E-05 147Sm 2.7210E-06 3.5820E-06 to 3.6688E-06 149Sm 1.2934E-07 1.4821E-07 to 1.6900E-07 150Sm 1.5432E-05 1.5880E-05 to 1.7900E-05 151Sm 7.8954E-07 8.0820E-07 to 1.1000E-06 152Sm 5.6451E-06 5.7314E-06 to 7.4850E-06 153Eu 7.1295E-06 7.9849E-06 to 8.6528E-06 155Gd 8.0878E-09 6.3000E-09 to 1.5641E-08

PAGE 164

164 Table 5-34. Five year coo ling time results for Case B of the Phase IVB Benchmark. Nuclide MCNPX (atoms/b-cm) Range of Values from Other Depletion Codes (atoms/b-cm) 234U 4.4485E-07 1.3174E-07 to 4.1870E-07 235U 1.8480E-05 1.8119E-05 to 1.8830E-05 236U 7.2745E-06 6.9037E-06 to 7.0634E-06 238U 2.1481E-02 2.1467E-02 to 2.1493E-02 238Pu 7.2430E-06 2.1267E-07 to 6.3483E-06 239Pu 2.5476E-04 2.4983E-04 to 2.7284E-04 240Pu 1.8232E-04 1.7540E-04 to 1.8510E-04 241Pu 8.3970E-05 8.1013E-05 to 8.5710E-05 242Pu 4.3597E-05 3.9817E-05 to 4.3090E-05 237Np 4.3965E-06 8.7398E-07 to 4.6418E-06 241Am 2.7709E-05 2.6942E-05 to 2.8303E-05 243Am 1.0125E-05 9.3994E-06 to 1.0616E-05 242Cm 9.8698E-10 7.5522E-10 to 1.4676E-09 243Cm 5.5486E-08 4.2821E-08 to 5.1277E-08 244Cm 5.5500E-06 3.4362E-06 to 3.8927E-06 245Cm 6.6887E-07 3.2603E-07 to 4.7096E-07 95Mo 5.0010E-05 5.0570E-05 to 5.1528E-05 99Tc 5.7945E-05 5.9428E-05 to 6.2460E-05 101Ru 6.2895E-05 6.2654E-05 to 6.5909E-05 103Rh 4.7293E-05 4.2315E-05 to 4.9970E-05 109Ag 1.0552E-05 8.5551E-06 to 1.2940E-05 133Cs 6.3248E-05 6.1230E-05 to 6.4442E-05 143Nd 3.8029E-05 3.9608E-05 to 4.0746E-05 145Nd 2.9778E-05 2.9642E-05 to 3.1176E-05 147Sm 6.8077E-06 8.7798E-06 to 9.5399E-06 149Sm 1.7692E-07 2.0516E-07 to 2.2260E-07 150Sm 1.5432E-05 1.5880E-05 to 1.7344E-05 151Sm 7.7139E-07 7.7864E-07 to 9.9180E-07 152Sm 5.6451E-06 5.7314E-06 to 7.4860E-06 153Eu 7.1797E-06 7.9849E-06 to 8.7110E-06 155Gd 3.3990E-07 3.2182E-07 to 9.2307E-07 In both cases, for most actinides and fission pr oducts, values were predicted within the range of values of the other depletion codes av ailable in the benchmark. General trends of buildup and depletion also fit the data from other depletion codes available in the benchmark. However, the buildup of Cm isotopes is larger in both Cases A and B as compared to other depletion codes. Since this benchmark is c ode-code comparison benchm ark, deviation from

PAGE 165

165 expected buildup must be the resu lt of methodology or utilized data differences. First of all, though metastable isotope representation is availabl e in MCNPX, for this benchmark, metastable isotopes were not tracked in particle transport. Ho wever, historically in the data releases for the MCNP codes, 242mAm has been listed with a zaid of 95242 instead of th e 95642 as expected from equations 3-9 and 3-10. As a result of the he ritage of the data release, MCNPX depletion expects Am-242m to be listed as 95242; however, the cross sections for these calculations were generated by an external sour ce that decided to generate 242Am as 95242. The end result of this confusion is that in particle transport reaction rates were calculated for 95242 using the 242Am cross sections; however, these reaction rates were converted into destruction and creation coefficients for 242mAm, and the 63-group match approach was used to generate coefficients for 242Am. The isomer and the ground state isotope do posses different reactio n rate probabilities; 242mAm has a long half life (141 years) and a larg er resonance integral in the harder MOX spectrum as compared to the ground state 242Am. 242mAm either fissions, possessing a large fission cross section, or captur es a neutron to become Am-243 which decays to Cm-243. The nuclide Am-242 decays quickly to Cm242 which then decays to Pu-238. 242Am has a higher capture-to-fission ratio than 242mAm and a larger radi ative capture resonan ce integral in the harder MOX spectrum; therefore, using the 242Am cross section instead of 242mAm will lead to more captures resulting in greater production of Cm-243 Cm-244 and Cm-245. This hypothesis seems to be validated by the fact that Cm-243, Cm -244 and Cm-245 seem to be predicted higher than the range of values. Other possible hypothesi zed differences in calculation may result from the CINDER90 depletion algorithm. Further working of the CINDER90 algorithm will be required in order to validate the latter hypothe sis. The results do, however, suggest good

PAGE 166

166 agreement with the other availa ble depletion codes, and do show that MCNPX is capable of modeling MOX fuel assemblies. Further Considerations In the analyzed benchmarks either single pin or single assembly geometries were analyzed. Reflective boundary conditions were assumed for each boundary of the examined geometries. In a real reactor operating scenario net current at the pin or as sembly boundary is never zero. Therefore in order to truly test the validity of the methodology fo r a certain application, a full core analysis would be required. In order to preserve exact accuracy, a full core analysis involves detailed initial isotope and operation data. Most commercial data, at this level of detail, is proprietary and for the moment una vailable to the general public. Another major issue to consider is that the NEA only has available accepted benchmark data for roughly 14 LWRs. The necessary measured data to test a depl etion methodology versus more complicated reactor designs is not readily ava ilable or yet to be validated. Development of time-dependent nuclear reactor experiments and th e radiochemical analysis involved in exacting isotope concentrations for analys is is expensive. Unfortunately, this concern has limited the amount of available benchmark data for advanced reactor designs. However, as the United States attempts to meet the initiatives of the GNEP program and develop further advanced reactor technologies, experiments should be developed, and valuable data should be extracted, leading to a plethora of new benchmarks to prove the viability of advanced methodologies.

PAGE 167

167 CHAPTER 6 CONCLUSIONS AND SUMMARY The MCNPX depletion capability offers co mplete, relatively easy-to-use depletion calculations in a single, well established, radia tion transport Monte Carlo code. The capability links the well-established MCNPX steady-stat e reaction rate calcul ator to the CINDER90 temporal nuclide inventory calculator. MCNPX is not limited by typical approximations utilized in deterministic methodologies. The MCNPX stea dy-state reaction rate cal culator implements 3dimensional continuous energy Monte Carlo in order to determine tr ue interaction rate behavior for any type of material and geometry combina tion. The use of CINDER90 allows for the most potential robustness in modeling capability b ecause the Linear Markov Chain method does not rely upon parsing a matrix into a solvable form and approximating depletio n behavior of certain nuclides. CINDER90 simply casts the set of coup led depletion equations into sets of linear transmutation chains, solving each chain for part ial nuclide densities, and then summing those nuclide densities to determine total nuclide concentrations. This algorithm allows any type of depletion system, fast burning or slow burning, to be modeled exactly. This work differs from prior efforts in Monte Carlo linked depletion development in that the capability has been incorporated into a widely used and accepted radiation transport package. Each feature implemented in the methodology has been tested in order to provide ade quate justification for implementation instead of simply arbitrarily in corporating methodologies for the sake of linking to independent code packages. This capability at tempts to track as many nuclides as data permits while tracking all important reaction rates, and a ccounting for average intera ction rate behavior in order to achieve a highly accurate depletion so lution that attempts to account for true system physics behavior.

PAGE 168

168 Accomplishments Many features have been implemented into th e methodology and tested in order to enhance the versatility of the capability as compared to other available code packages. The following features were accomplished as part of this work: 1. A comprehensive input stru cture that gives the user versatility in modeling capability was developed. 2. An organized easy to understand output st ructure that groups data in terms of average system parameters and individua l burn material data was provided. 3. Isotope tracking was simplified by providing predefined fission product tier sets which contain all available fission produc ts containing ENDF/B VII.0 transport cross sections as well as CINDER90 fission yield information. Testing demonstrated that implementing the maxi mum amount of fission products better mimics the results of the benchm arked MONTEBURNS depletion code. 4. A mechanism was developed for automa tically determining and tracking all immediate daughter reaction isotopes. 5. A capability was created in order to a llow the user to manually adjust the concentration of any nuclide at any time step by specifying the atom fraction, weight fraction, atom density or weight density of the nuclide to be changed. 6. A structure was engineered for includi ng available metastable isotopes in continuous energy reaction rate tracking and transforming that the nuclides information for use in the depl etion calculation in CINDER90. 7. A methodology was established for tracking in continuous energy the (n, fission), (n, ), (n,2n), (n,3n), (n,p), and (n, ) interaction rates for predefined nuclides. Testing has demonstrated that the inclusion of these reaction rates, for a simple pin cell calculation, demonstrates improvement in actinide predictability as compared to simply tracking (n, fission) and (n, ). 8. A methodology was conceived for determini ng system averaged flux normalization parameters used in determining global ne utron source, which is also used in calculating the total magnitude of the flux. This capability was further augmented to work with the repeated structures feature in MCNPX. 9. An algorithm was developed to automa tically determin the proper energy dependent fission yield to be used in a given burn material by determining the energy band containing the majority of fission reactions and using that corresponding fission yield.

PAGE 169

169 10. A technique for adjusting de pletion coefficients was de veloped in order to account for average behavior over a given time step by incorporating a cross section averaging technique that utilizes spectr a calculation at both the predictor and corrector steps. Two independent methods were tested, and the results demonstrate the MONTEBURNS like method achieved similar results to the CELL-2 like method, and because the MONTEBURNS like method achieved the result incorporating less memory usage, this method was chosen for implementation in MCNPX. 11. The default decay chain convergence criterion for use in CINDER90 was determined based on testing. A keyword value was implemented in order to allow user control of this criteria Analysis of Benchmark Results Three different benchmarks were also co mpleted in order to validate the current methodology. The OECD/ NEA Phase IB benchmark results showed that the code was capable of modeling an adjusted fuel-to-moderator ratio pin cell and achieve result s within the range of other depletion methodologies. Though the mode led geometry for this benchmark did not account for the true spatial behavi or witnessed in the experiment comparison to experiment did demonstrate ability to achieve adequate results. The H. B. Robinson benchmark demonstrated ability to model real core behavior, of a typi cal PWR assembly, and achieve accurate results if appropriate burnup steps were implemented. The OECD/ NEA Phase IVB benchmark results demonstrated the ability to model MOX assemb lies and achieve the similar results to other depletion methodologies. Though these calculati ons only represent the ability to model a PWR assembly, uranium-dioxide or MOX fueled, placed in an operating scenario in which the assembly boundaries may be modeled as infinitely reflected, the results do suggest positive evidence that the capability can m odel depletion systems adequately. Differences between measured and calculated values, or code-to-code differences, were a result of the following issues: accuracy of th e applied nuclear data, treatment of normalization parameters, inadequacy of the implemented tim e steps, differences in neutron spectra as

PAGE 170

170 compared to a critical spectrum calculation, spat ial reactivity effects of the modeled geometry, and statistical variance of the reaction rates and ra re events. ENDF/B VII.0 cross sections were utilized in particle transport. Du e to the recent availability of this data, the pedigree of the data is not exactly known as it has not b een widely tested. The cross sections implemented in the calculation were generated utili zing an automated subroutine w ith resonance reconstruction tolerances that may or may not be adequate for the applied application. The ENDF/B VII.0 data may also not be complete and therefore may not contain cross section information for every reaction of interest. However, MCNPX depletion a ssumes that if an isotope is used in particle transport and contains an accessible cross section f ile, then data must exist for each reaction type. Though the methodology only requires cross sec tion information for (n, fission), (n, ), (n,2n), (n,3n), (n,p), and (n, ) interactions, it may be possible that some of these reactions do not exist in the cross section data file. Unfortunately, such an instance would result in the code assuming a zero cross section for that reacti on type even though a nonzero reaction rate might be available in the CINDER90 library file. Most of the oth er depletion code resu lts were previously generated using ENDF/B IV, V, or VI data and th erefore discrepancies in the ENDF/B VII.0 data as compared to these libraries also resulted in differences in calculated values. CINDER90 also contains transmutation cross sections for 3400 nuclides and over 3000 of those nuclides do not contain available ENDF data. Therefore cros s section determination codes were used to calculate the data implemented in CINDER90 libra ry file. Unfortunately, the pedigree of the cross section determination code is version de pendent and has evolved in accuracy as better models have become available. Therefore the CINDER90 data for these nuclides may have also differed from what is used by other codes leading to further error in calc ulation when comparing to other tools. Pedigree of this implemented data will only be proven with further benchmarking

PAGE 171

171 and testing. The buildup of Cm isotopes, in the Phase IVB Benchark, was larger for both cases when compared to other depletion codes. Though metastable isot ope representation was available in MCNPX, for this benchmark, metast able isotopes were not tracked in particle transport. The confusion in utilizing the 242Am cross section as opposed to the 242mAm resulted in further calculation error. The isomer a nd the ground state isotope do posses very different reaction rate probabi lities. Ground state 242Am has a larger resonance integral over the harder MOX spectrum than 242Am. 242Am also decays into 242Cm which then decays to 238Pu, or fissions. The 1st excited state isomer 242mAm absorbs a neutron and either fissions or becomes 243Am which later becomes other Cm isotopes. Du e to the differences in capture-to-fission ratio between the ground and excited stat es, and the difference in resona nce integral over the harder MOX spectrum, the mix up in isomer representation cause d more buildup of 244Cm and thus more buildup of 245Cm. The discrepancies in 238Pu production were also probably a function of the misrepresentation of this isomer. MCNPX determines the global neutron source, for use as a flux normalization constant, by calculating system averaged parameters. Unfort unately, these system average parameters only exist for certain actinides. On e example is recoverable energy per fission. Though an energy dependent fission cross section ex ists for almost every actinide, prompt fission release energy information only exists for 1/3 of all fissi onable actinides. For all other actinides, 235U based parameters are assumed in MCNPX depletion. MCNPX depletion currently does account for the true recoverable energy per fission. An approximation is made in order to try and determine the delayed energy contribution by multiplying the prompt fission release energy by a constant that is assumed to account for delayed energy. The delayed energy contribution is unfortunately system dependent. For these benchmarks, howev er, the geometry modeled was not the actual

PAGE 172

172 experimental geometry, the full core was not modeled, and therefore accounting for the true recoverable energy per fission in the adjusted geometry woul d not be possible. Also understanding how the delayed energy is depos ited involves knowing the emission spectra of that energy and transporting that energy to where it is deposited. As mentioned previously, even for the majority of nuclides containing ENDF/B VII.0 transport cross se ctions, capture gamma photon emission spectra is unknown. An initial ca lculation was performed to determine the mass defect integral energy contributi on. However, a photon containing the integral energy from the mass defect transports differently than the actu al photons spectra generated from the reaction. Therefore due to lack of photon spectra information, the applie d constant was deemed most acceptable for accounting for delayed energy contribu tion, and this constant ultimately affected the solution outcome. For full core modeling at critical, the delayed energy contribution only scales as a function of neutrons emitted per fi ssion therefore this valu e only changes between 2.43-3 neutrons per fission as function of buildi ng up higher actinides, and therefore the delayed energy contribution change should not be very si gnificant as function of burnup for these types of systems. For each calculation, common time durations using varied powers were implemented in order to burn each of the benchmarks to the re quired burnups. Unfortunately, this type of burning led to burning the system with varied bu rnup steps. If reactivity contributing isotope changes density too significantly during a burn step, then it is po ssible that the constant creation and destruction coefficient appr oximation is not valid for that time step. The most probable reason for the benchmark calculations achieving poorer predictability for higher burnups was related to the fact th at the burnup steps for the higher burnup cases may have been too long in

PAGE 173

173 order to assume the constant coefficient approx imation. However, the results for the lower burnup cases did show excellent agreement in c ode-to-code and experimental comparison. Each of the modeled geometries did not incorporate an approximation to account for critical spectrum. When the pin or assembly is in the reactor core, the geometry is burned assuming the geometry is in a critical configurati on. Therefore the true le akage in and out of the modeled benchmark geometries varies as functi on of burnup and is not a true reflective boundary condition. In deterministic based codes, criti cal spectrum is approximated by adjusting the flux by a normalization parameter generated from a buc kling adjustment calculation. Not modeling a geometry utilizing the critical spectra will cause the propagation of nuclide generation and depletion to be different from what is expect ed from the actual geometry because the captures, fissions, and scatters will be different than what is expected from the critical spectrum. Because the system is not modeled accounting for the tr ue operating behavior, the spatial reactivity behavior will also be affected. For the Phase IB Benchmark, the adjusted-fuel-to-moderator ratio pin cell calculation does not account for the true spatial behavior of the reactions because the pin dimensions have been augmented in order to achieve similar global reactivity behavior. However, the actual interaction rate behavior within the pin for each reaction type is not preserved. Therefore the evoluti on of the temporal nuclide inventory will be in error as the true reaction rates differ from the calculation. Even in the H. B. Robinson and Phase IVB benchmarks, where a full assembly was modeled, tr ue interaction rate beha vior is not preserved because the reflective boundary condition is not i ndicative of the expected leakage condition for those assemblies during actual reactor operation. Only modeling the full r eactor, with the actual operating conditions will achieve the most accurate result.

PAGE 174

174 Most codes, deterministic/Monte Carlo based, have historically had issues in calculated 237Np concentration as well as other downstream nuclides produced from the 237Np transmutation path. As determined from the previous testing, 237Np and therefore 238Pu predictability is improved by accounting for the energy integrated collision rate of (n,2n) reactions (i. e. 238U + n --> 2n+ 237U; 237U --> 237Np; 237Np + n --> 238Np --> 238Pu (rare event!) or 239Pu + n --> 238Pu +2n (very rare event!). It also possible to form 237Np from either 235U +n --> no fission --> 236U; 236U +n-->237U -->237Np (only account for 10-20% of reactions with 235U) or alpha decay from 241Am (rare event due to long half life). 238Pu is also formed from decay of 242mAm -->242Cm ->238Pu. Where 242mAm is formed from multiple Pu isotope captures. The common theme in the miscalculation of these isotopes is related to pr operly sampling rare events. Therefore why not try a type of forced reaction type sample? In "Nonanalog Implementations of Monte Carlo Isotope Inventory Analysis," the authors tried to develop a method of stochastically solving the actual depletion equations (instead of just stochastically solving for coefficients to be applied for a deterministic temporal nuclide inventory solution).99 They tried a forced collision method for sampling rare events (rare intera ctions), and found that forced co llision sampling of rare events for a finite amount of histories decreased the am ount of sampled non-rare events leading to an inadequate sampling of non-rare events and ultim ately affecting the transm utation path outcome. Forced collision sampling may affect some part of the analyzed problem when an artificial bias is introduced. They found that the only way to rea lly cure this issue was to use more histories. Therefore this evidence suggests that running more histories ma y fix the problem of sampling rare events and therefore lead to better prediction of rare event isotopes. Further testing of a variety of benchmarks will be required in order to truly validate this hypothesis.

PAGE 175

175 Future Work Further enhancements to the MCNPX deplet ion methodology will be required in order to enhance the usefulness of the technology. The current MCNPX 2.6.F depletion method does not currently account for critical spectrum adjustme nts or posses a method to incorporate thermo hydraulic feedback, which is essential for modeling true system physics.32 Critical spectrum may be calculated by weighting the impinging reflect ed leakage flux with the mean free path and continuing to adjust the impinging flux until criticalit y is reached. More or less weight would be given to a reflected neutron depending upon if the sy stem was initially sub or super critical. This type of calculation would involve tagging and tracking all reflected neutrons that contribute to reactivity. However, this approximation must be thoroughly tested versus a wide variety of benchmarks in order to prove validity of the method. Thermohydraulic feedback may be incorporated into MCNPX by linki ng MCNPX to a code that take s power distribution data in order to generate temperature distributions. Unfortunately, the calculation of thermohydraulic feedback is an iterative process, because the effects are nonlinear and dependent upon each other, and therefore a computationally e xpensive process when using Monte Carlo. However, this type calculation is required in order to preserve true system physics. The depletion methodology also does not propagate number density error, or posse ss functionality to alter the geometry during the burnup.98 The relative error associated with the isotope concentration calculations are a function of the relative error of the flux used to calculate the co llision densities, as well as the isotopes relative error from the previous calcu lation. Knowledge of the propagation of the number density error is required in order to und erstand the precision of th e calculated result and therefore must be incorporated in further vers ions of the capability. The evidence of the benchmark calculations does suggest that further investigation is also required in the passby calculation in the CINDER90 algorithm. Many te sts exist for determining progeny generation

PAGE 176

176 and it is possible that the orde ring of these tests may affect the solution outcome. However, these issues along with other func tionality concerns will be addressed in further development of the capability. This capability does set up a significant founda tion, in a well established and supported radiation transport code, for further deve lopment of a Monte Carlo-linked depletion methodology. Such a capability will be essential to the future development of advanced reactor technologies, and detector systems to monito r those reactor technologi es, that violate the limitations of current deterministic based methods.

PAGE 177

177 APPENDIX A SAMPLE INPUT In the following example, a 4 X 4 fuel pin a rray is modeled in order to demonstrate the necessary keyword implementation for burning a re peated structure while also utilizing manual concentration changes at specific time steps: ... 1 1 6.87812e-2 -1 u=2 imp:n=1 vol=192.287 $ fuel 3 2 4.5854e-2 1 -2 u=2 imp:n=1 vol=66.43 $ clad 4 3 7.1594e-2 2 u=2 imp:n=1 vol=370.82 $ water 6 4 6.87812e-2 -1 u=3 imp:n=1 vol=192.287 $ fuel 7 5 4.5854e-2 1 -2 u=3 imp:n=1 vol=66.43 $ clad 8 6 7.1594e-2 2 u=3 imp:n=1 vol=370.82 $ water 10 0 -3 4 -5 6 u=1 imp:n=1 lat=1 fill=0:1 0:1 0:0 2 3 2 3 ... ... BURN TIME=50,10,500 MAT= 1 4 POWER= 1.0 PFRAC= 1.0 0 0.2 OMIT= 1,8,6014,7016,8018,9018,90234,91232,95240,95244 4,8,6014,7016,8018,9018,90234,91232,95240,95244 BOPT= 1.0, -4 AFMIN= 1e-32 MATVOL= 384.57 384.57 MATMOD= 2 1 1 -4 1 94238 1e-6 2 2 -1 2 94238 1e-6 94241 1e-6 -4 1 94238 1e-6 ... The following example utilizes a lattice containing universes 2 and 3 which are both repeated twice in the lattice. Universe 2 is co mprised of cells 1, 3 and 4, where cell 1 contains material 1, and universe 3 is comprised of cells 6, 7 and 8, where cell 6 contains material 4. The MAT keyword specifies that both materials 1 an d 4 will be burned. The combination of the TIME, POWER and PFRAC keywords specify that these materials will be burned first for 50 days at 100% of 1MW, then decayed for 10 days, and then finally burned for 500 days at 20% of 1 MW.

PAGE 178

178 The BOPT keyword specifies that the fo llowing options will be invoked: Q-value multiplier = 1.0, tier 1 fission products, output orde red by zaid, the output is printed at the end of each kcode run and only tabular transport cross se ctions are utilized. B ecause tabular transport cross sections do not exist for every isotope gene rated, and OMIT card is required in order to omit these isotopes from the transport process. The transmutation of thes e isotopes is accounted for by sending a 63-group flux from MCNPX to be matched to a 63-group cross section set within CINDER90 which will be energy integrated to determine a total collision rate. The OMIT card in the example omits 8 isotopes from materi al 1 and 8 isotopes from material 4. The AFMIN card states that only isotopes possessing an atom fraction below 1e-32 will be omitted from the transport calculation. Since repeated structures are utilized in the example, a MA TVOL keyword is required in order to calculate the track lengt h estimated reaction rates in each repeated structure. In this example, since material 1 and 4 are repeated twice and each material possesses a volume of 192.287, MATVOL keyword entries of 384.57 ( 192.287*2) were required for each material being burned. A MATMOD keyword is used in order to ma nually change the concentration of certain isotopes at specified time steps. In this exampl e, manual isotope concentration changes are to be completed at 2 time steps. At time step 1, material 4 will have the atom density of 1 isotope changed: 94238 will be set to 1e-6 atoms/barn-cm. At time step 2, material 1 will have the atom densities of 2 isotopes changed: material 1 will have isotope 94238 changed to 1e-6 atoms/barncm and isotope 94241 will be changed to 1e-6 at oms/barn-cm, material 4 will have the atom density of 1 isotope changed: 94238 w ill be set to 1e-6 atoms/barn-cm.

PAGE 179

179 LIST OF REFERENCES 1. U.S. Department of Energy Office of Civilian Radioactive Waste Management, Oklo: Natural Nuclear Reactors, November 2004, (December 2007). 2. J. J. DUDERSTADT and L. J. HAMILTON, Nuclear Reactor Analysis John Wiley & Sons, Inc., New York, NY (1976). 3. E. M. BAUM, H. D. KNOX, and T. R. MILLER, Nuclides and Isotopes: Chart of Nuclides, Sixteenth Edition, Knolls Atomic Power Laborat ory and Lockheed and Martin Corporation Schenectady, NY (2006). 4. E. FERMI, The Future of Atomic Energy, Technical Information Division, Oak Ridge Directed Operation of the Atomic Ener gy Commission Report: 10-11-48-850-11425, Oak Ridge, TN (1946). 5. S. GLASSTONE and W. H. JORDAN, Nuclear Power and its Environmental Effects American Nuclear Society, La Grange Park, IL (1980). 6. R.T. LAHEY and F.J. MOODY, The Thermal Hydraulics of a Boiling Water Nuclear Reactor American Nuclear Society, La Grange Park, IL(1979). 7. J. R. LAMARSH, Introduction to Nuclear Engineering Addison-Wesley Publishing Company, Inc., Melano Park, CA (1983). 8. K. O. OTT, C. W. TERREL, R. C. BORG, and P. J. MAUDLIN, Integrated Fuel-Cycle Models for Fast Breeder Reactors, Annuls of Nuclear Engineering 8, pg. 371-392 (1981). 9. R. G. COCHRAN and N. TSOULFANIDIS, The Nuclear Fuel Cycle: Analysis and Management American Nuclear Society, In c., La Grange Park, IL (1999). 10. A. ROMANO, P. HEJZLAR, and N. E. T ODREAS, Fertile-Fre e Fast Lead-Cooled Incinerators for Efficient Actinide Buruning, Journal of Nuclear Technology 147, pg. 368386 (2004). 11. G. S. CHANG, Neutronic and Burnup Charact eristics of an Actinide-Reduced Plutonium Fuel with Tungsten, Journal of Nuclear Technology 122, pg. 43-51 (1998). 12. T. A. TAIWO, T. K. KIM, J. A. STILLMAN, R. N. HILL, M. SALVATORES, Assessment of a Heterogeneous PWR Assembly for Plutonium and Minor Actinide Recycle, Journal of Nuclear Technology 155, pg. 43-51 (1998). 13. X. JIANG and Z. XIE, Transport-burnup Code Systems and Their Applications for IAEA ADS Benchmark, Annuls of Nuclear Energy 31, pg. 213-225 (2004).

PAGE 180

180 14. N. MESSAOUDI and J. TOMMASI, Fast Burn er Reactor Devoted to Minor Actinide Incineration, Journal of Nuclear Technology 137, pg. 89-96 (2002). 15. M. DELPECH, H. GOLFIER, A. VASILE, F. VARAINE, L. BOUCHER, Minor Actinides in PWRs, International Congress on Advances in Nuclear Power Plants Embedded International Topical Meeting at the 2006 ANS Annual Meeting Reno, NV, (2006). 16. United States Department of Energy, Office of Nuclear Energy, Science and Technology Advanced Fuel Cycle Initiative: Status Report for FY 2005, Report to congress (2006). 17. G. W. MCKINNEY and H. R. TRELLUE, Transmutation Feature within MCNPX, Los Alamos National Laboratory Report: LA-UR-04-1572, Los Alamos, NM (2004). 18. United States Department of Energy, Office of Nuclear Energy, Office of Fuel Cycle Management Global Nuclear Energy Partners hip Strategic Plan, Report to congress: GNEP-167312 (2007). 19. M. D. LOWENTHAL, Transmutation in the Nucl ear Fuel Cycle: Approaches and Impacts, Journal of Nuclear Technology 138, pg. 284-299 (2002). 20. M. VISOSKY, P. HEJZLAR, M. KAZIMI, Act inide Transmutation Using Pressurized Water Reactors, International Congress on Advances in Nuclear Power Plants Embedded International Topical Meeting at the 2006 ANS Annual Meeting Reno, NV, (2006). 21. T. TAKEDA, T. YAMAMOTO, M. MIYAUCHI, I nterpretation of Actin ide Transmutation, Progress in Nuclear Energy 40, pg. 449-456 (2002). 22. M. SALVATORES, Nuclear Fuel Cycle Strategies Including Partitioning and Transmutation, Nuclear Engineerin g and Design, 235, pg. 805-816 (2005). 23. United States Department of Energy, Global Nuclear Energy Partnership: Improved Nuclear Safeguards, December 2006, (December 2007). 24. G. F. KNOLL, Radiation Detection and Measurement, John Wiley and Sons, Inc., New York, NY (1999). 25. H. C. HURIA and R. J. BUECHEL, Recen t Improvements and New Features in the Westinghouse Lattice Physics Codes, Transactions of the Ameri can Nuclear Society, 72, pp. 369 (1995). 26. T. P. SHANNON, J. K. WHEELER, and G. TOUVANNAS, TGBLA/PANACEA and CASMO/MICROBURN analyses of GE9B/GE10 Fuel in the Quad Cities, Transactions of the American Nuclear Society, 74, pp. 287 (1996).

PAGE 181

181 27. C. V. PARKS, Summary Description of the SCALE Modular Code System, Martin Marietta Energy Systems, Inc. and Oak Ridge National Laborat ory Report: NUREG/CR5033 ORNL/CSD/TM-252, Oak Ridge, TN (1987). 28. M. J. KHAN, N. AHMAD, Proliferation Resist ance Potential and Bur nup Characteristics of an Equalibrium Core of Novel Natural Ur anium Fueled Nuclear Research Reactor, Annuls of Nuclear Energy 32, pg. 612-620 (2005). 29. K. SUYAMA, H. MOCKIZUKI, T. KIYOSUMI, Revised Burnup Code System SWAT: Description and Validation Using Po stirradiation Examination Data, Nuclear Technology 138, pp. 97-108. 30. M. EDENIUS, K. EKBERG, B. H. FORSSEN, and D. KNOTT, Casmo-4, A Fuel Assembly Burnup Program, Users Manual, Studsvik Repor t: SOA-95/15, Studsvik of America, Inc., Kjeller, Norway (1995). 31. N. SUGIMURA, A. YAMAMOTO, T. USHIO, M. MORI, M. TABUCHI, and T. ENDO, Neutron Transport Models of AEGIS: An Ad vanced Next-Generation Neutronics Design System, Nuclear Science and Engineering 155, pp. 276-289 (2007). 32. D. KNOTT, E. WEHLAGE, Description of the LANCER02 Lattice Physics Code for Single-Assembly and Multibundle Analysis, Nuclear Science and Engineering 155, pp. 331-354 (2007). 33. K. LASSMANN, C. T. WALKER, J. VAN DE LAAR, Extensions of the TRANSURANUS Burnup Model to Heavy Water Reactor Conditions, Journal of Nuclear Materials 255, pg. 222-233 (1998). 34. A. NUNEZ-CARRERA, J. L. FRANCOIS, G. ESPINOSA-PAREDES, Comparison between HELIOS Critical-Depletion Calcul ations and a PWR Thorium Cell Burnup Benchmark, Annuls of Nuclear Energy 31, pg. 713-722 (2004). 35. M. L. FENSIN and S. ANGHAIE, Current Me thods of Depletion Analysis, Los Alamos National Laboratory Report LA-UR-06-3933, American Nuclear Societys 2006 Winter Meeting, Albuquerque, NM (2006). 36. G. S. CHANG, J. M. RYSKAMP, Depletion An alysis of Mixed-Oxide Fuel Pins in Light Water Reactors and the Advanced Test Reactor, Nuclear Technology 129, pp. 326-337 (2000). 37. W. HAECK, An Optimum Approach to Monte Carlo Burn-Up, PhD thesis, Ghent University, Belgium (2007). 38. E. J. PARMA, BURNCAL: A Nuclear Reactor Burnup Code Using MCNP Tallies, Sandia National Laboratory Report: SAND 2002-3868, Albuquerque, NM (2002).

PAGE 182

182 39. Z. XU, Design Strategies for Optimizing High Burnup Fuel in Pressurized Water Reactors, PhD thesis, Massachusetts Institute of Technology, Cambridge, MA (2003). 40. N. A. HANAN, A. P. OLSON, R. B. POND, J. E. MATOS, A Monte Carlo Burnup Code Linking MCNP and REBUS, 1998 International Meeting on Reduced Enrichment for Research and Test Reactors Sao Paulo, Brazil (1998). 41. G. S. CHANG, MCWO-Linking MCNP and ORIGEN2 For Fuel Burnup Analysis, American Nuclear Societys T opical Meeting on Monte Carlo Chattanooga, TN (2005). 42. J. CETNAR, W. GUDOWSKI, and J. WALLEN IUS, User Manual for Monte-Carlo Continuous Energy Burnup (MCB) CodeVersion 1, draft report, Depa rtment of Nuclear and Reactor Physics, Royal Institute of Technology, Stockholm, Sweden (2005). 43. D. LI, X. ZHONGSHENG, L. SHU, Monte Carlo Transport and Burnup Calculation, Annuls of Nuclear Energy 30, pg. 127-132 (2003). 44. R. L. MOORE, B. G. SCHNITZLER, C. A. WEMPLE, R. S. BABCOCK, D. E. WESSOL, MOCUP: MCNP-ORIGEN2 Coupled Utility Program, Idaho National Engineering Laboratory Report: INEL-95/0523, Idaho Falls, ID (1995). 45. H. R. TRELLUE and D. I. POSTON, Use rs Maunual, Version 2.0 for MONTEBURNS, Version 5B, Los Alamos National Laborat ory report LA-UR-99-4999, Los Alamos, NM (1999). 46. M. AOYAMA, S. UCHIKAWA, K. MIKI, K. HIRAMOTO, and R. TAKEDA, A Conceptual Design of a Fuel Bundle for Extended Bunup in Boiling Water Reactors, Journal of Nuclear Technology 64, pg. 19-25 (July 1993). 47. N. A HANAN, A. P. OLSON, R. B. POND, and J. E. MATOS, A Monte Carlo Burnup Code Linking MCNP and REBUS, Argonne Natio nal Laboratory Report: ANL-TD-CP-97492, Agonne, IL (no date specified). 48. D. J. KELLY, Depletion of a BWR Latti ce Using the RACER Continuous Energy Monte Carlo Code, International Conference of Mathematics and Computa tion, Reactor Physics, and Environmental Analysis Portland, OR, (1995). 49. M. DEHART, High-Fidelity Depletion Capa bilities of the SCALE Code System Using Triton, American Nuclear Societys 2007 Winter Meeting Washington, D. C. (2007). 50. D. B. PELOWITZ, MCNPXTM Users Manual Version 2.5.0, Los Alamos National Laboratory Report LA-CP-05-0369, Los Alamos, NM (2005). 51. W. B. WILSON, T. R. ENGLA ND, E. D. ARTHUR, C. A. BE ARD, C. D. BOWMAN, L. N. ENGEL, A. GAVRON, D. C. GEORGE, L. L. DAEMEN, H. G. HIGHES, III, W. W. KINNISON, R. J. LABAUVE, D. M. LEE, H. LICHTENSTEIN, P. W. LISOWSKI, D. W.

PAGE 183

183 MUIR, A. P. MUIR, A. P. PALOUNEK, R. T. PERRY, E. J. PITCHER, R. E. PRAEL, R. J. RUSSEL, G. SANDERS, L. S. WATERS, P. G. YOUNG and J. J. ZIOCK, Accelerator Transmutation Studies at Los Alamos with LAHET, MCNPX and CINDER90, Los Alamos National Laboratory Report LA-UR-93-3080, Workshop on Simulation of Accelerator Radiation Environments Sante Fe, NM (1993). 52. M. D. DEHART, M. C. BRADY, C. V. PA RKS, OECD/NEA Bur nup Credit Calculation Criticality Benchmark Phase I-B Results, Oa k Ridge National Labor atory Report: ORNL6901, Oak Ridge, TN (1996). 53. O. W. HERMANN, S. M. BOWMAN, M. C. BRA DY, C. V. PARKS, Validation of the Scale System for PWR Spent Fuel Isotopic Composition Analysis, Oak Ridge National Laboratory Report: ORNL/TM-12667, Oak Ridge, TN, USA (1995). 54. G. J. OCONNOR and P. H. LIEM Burn-up Credit Criticality Benchmark PHASE IV-B: Results and Analysis of MOX Fu el Depletion Calculations, Nuclear Energy Agency Report: NEA/NSC/DOC(2003)4 (2003). 55. G. F. THOMAS and D. H. BARBER, Num erical Solution of Bateman Systems, Annuls of Nuclear Energy 20, pg. 407-414 (1993). 56. M. YAMAMOTO, Coarse Time-Step Integrati on Method for Burnup Calculation of LWR Lattice Containing Gadolinium-Poisoned Rods, Journal of Nuclear Science and Technology 22, pg.1-15 (1985). 57. G. ILAS, B. D. MURPHY, and I. C. GAULD, O verview of Origen-ARP and its Applications VVER and RBMK, American Nuclear Societ ys 2007 Winter Meeting Washington, D. C. (2007). 58. A. G. CROFF, ORIGEN2: A Versatile Com puter Code for Calculating the Nuclide Compositions and Characteristics of Nuclear Materials, Journal of Nuclear Technology 62, pg. 335-352 (1983). 59. G. F. THOMAS and D. H. BARBER, Sti ffness in Radioactive Decay Chains, Annuls of Nuclear Energy 21, pg 309-320 (1994). 60. A. G. CROFF, ORIGEN2 A Revised and Updated Version of the Oak Ridge Isotope Generation and Depletion Code, Oak Ridge National Laboratory Report: ORNL-5621, Oak Ridge, TN (1980). 61. W. B. WILSON, V. DYADECHKO, T. R. ENGLAND, P. MOLLER, H. R. TRELLUE, A Manual for CINDER90 Version 06.1 Codes and Data, Los Alamos National Laboratory Draft Report, Los Alamos, NM (2006).

PAGE 184

184 62. M. L. FENSIN, Optimum Boiling Water React or Fuel Design Strategies to Enhance Shutdown by the Standby Liquid Control System Masters Thesis, University of Florida, Gainesville, FL (2004). 63. E. E. LEWIS and W. F. MILLER, Jr., Computational Methods of Neutron Transport Ch. 7, American Nuclear Society, Inc ., La Grange Park, IL (1993). 64. H. GREENSPAN. C. N. KELBER, D. OKRENT, Computing Methods in Reactor Physics Gordon and Breach Science Publishers, Inc. New York, NY (1968). 65. M. CLARK, K. F. HANSEN, Numerical Methods of Reactor Analysis Academic Press, Inc., New York, NY (1964). 66. I. M. SOBOL, The Monte Carlo Method Mir Publishers, Moscow, Russia (1984). 67. A. PAUTZ, Fuel Assembly Calculations Using the Method of Discrete Ordinates, Nuclear Science and Engineering 149, pg. 197-210 (2005). 68. D. S. LUCAS, Attila Modeling with Comp arison to Data, MCNP and MCNPX for the Advanced Test Reactor, International Congress on Advan ces in Nuclear Power Plants Embedded International Topical Meeting at the 2006 ANS Annual Meeting Reno, NV, (2006). 69. S. G. HONG and N. Z. CHO, Method of Characteristic Direction Probabilities for Heterogeneous Lattice Calculation, Nuclear Science and Engineering 132, pg. 65-77 (1999). 70. H. HERMAN, ENDF-6 FORMATS MANUAL, Brookhaven National Laboratory Report: BNL-NCS-44945-05-Rev., Long Island, NY (2005). 71. M. L. FENSIN and D. I. POSTON, Optimum Reflector Configurat ions for Minimizing Fission Power Peaking in a Lithium-Cooled, Li quid-Metal Reactor with Sliding Reflectors, Los Alamos National Laboratory Report LA-UR-7154, Space Technology and Applications International Forum 2005 Albuquerque, NM (2005). 72. M. L. FENSIN, J. O. ELLIOTT, R. J. LIPI NSKI, D. I. POSTON, Radiation Shielding Design and Orientation Considerations for a 1 kWe Heat Pipe Cooled Reactor Utilized to Bored Through the Ice Caps of Mars, Los Alamos National Laboratory Report LA-UR-058018, Space Technology and Applications International Forum 2006 Albuquerque, NM (2006). 73. J. L. DEVORE, Probability and Statistics for Engineers and Scientists, Brooks/Cole, Pacific Grove, CA (2000). 74. J. F. BRIESMEISTER, MCNP-A General Monte Carlo N-Particle Transport Code, Los Alamos National Laboratory Report: LA-13709-M, Los Alamos, NM (2000).

PAGE 185

185 75. H. R. TRELLUE, Development of Monteburns : A Code That Links MCNP and ORIGEN2 in a Automated Fashion for Burnup Calcula tions Los Alamos Nati onal Laboratory Report LA-13514-T, Los Alamos, NM (1998). 76. S. M. BOWMAN, M. D. DEHART, and L. M. PETRIE, Integrated Keno Monte Carlo Transport for 3-Depletion with SCALE, The American Nuclear Societys Monte Carlo 2005 Topical Meeting Chattanooga, TN (2005). 77. S. M. BOWMAN, D. F. HOLLE NBACH, M. D. DEHART, B. T. REARDEN, I. C. GAULD, and S. GOLUOGLU, SCALE 5: Powerful New Criticality Safety Analysis Tools, 7th International Conference on Nuclear Criticality Safety Tokai-mura, Japan (2003). 78. A. S. GERASIMOV, T. S. Zaritskaya, G. V. Ki selev, and L. A. Myrtsymova, The Cost of Transmutation of Fission Products in Nuclear Reactors, Atomic Energy 94, pg. 154-157 (2003). 79. J. S. HENDRICKS, G. W. MCKINNEY, H. R. TRELLUE, J. W. DURKEE, T. L. ROBERTS, H. W. EGDORF, J. P. FINCH, M. L. FENSIN, M. R. JAMES, D. B. PELOWITZ, and L. S. WATERS, MCNPX, Version 2.6.A, Los Alam os National Laborator y Report: LA-UR-058225, Los Alamos, NM (2005). 80. M. B. CHADWICK, P. OBLOZINSKY, M. HE RMAN, N. M. GREENE, R. D. MCKNIGHT, D. L. SMITH, P. G. YOUNG, R. E. MACFAR LANE, G. M. HALE, S. C. FRANKLE, A. C. KAHLER, T. KAWANO, R. C. LITTLE, D. G. MADLAND, P. MOLLER, R. D. MOSTELLER, P. R. PAGE, P. TALOU, H. TR ELLUE, M. C. WHITE, W. B. WILSON, R. ARCILLA, C. L. DUNFORD, S. F. MUGHABGHHAB, B. PRITYCHENKO, D. ROCHMAN, A. A. SONZOGNI, C. R. LUBITZ, T. H. TRUMBULL, J. P. WEINMAN, D. A. BROWN, D. E. CULLEN, D. P. HEINRICH S, D. P. MCNABB, H. DERRIEN, M. E. DUNN, N. M. LARSON, L. C. LEAL, A. D. CA RLSON, R. C. BLOCK, J. B. BRIGGS, E. T. CHENG, H. C. HURIA, M. L. ZERKLE, K. S. KOZIER, A. COURCELLE, V. PRONYAEV and S. C. VAN DER MARCK,"E NDF/B-VII.0: Next Generation Evaluated Nuclear Data Library for Nuclear Science and Technology", Nuclear Data Sheets 107, pg. 2931-3060 (2006). 81. T. R. ENGLAND, B. F. RIDER, ENDF-349 evaluation and compilation of fission product yields, Los Alamos Nationa l Laboratory Report: LA-UR-943106, Los Alamos, NM (1994). 82. M. L. FENSIN, J. S. HENDRICKS, G. W. MCKINNEY, H. R. TRELLUE, "Advances in Monte Carlo Depletion Capabilities for MCNP X," Los Alamos National Laboratory Report LA-UR-05-7895, American Nuclear Society 14th Biennial Topical Meeting of the Radiation Protection and Shielding Division Carlsbad, NM, (2006). 83. H. R. TRELLUE, Reduction of the Radiotoxicity of Spent Nuclear Fuel Using a Two-Tiered System Comprising Light Water Reactors and Accelerator-Driven Systems,Los Alamos National Laboratory Report: LA-14 052-T, Los Alamos, NM (2003).

PAGE 186

186 84. J. S. HENDRICKS, G. W. MCKINNEY, J. W. DUR KEE, J. P. FINCH, M. L. FENSIN, M. R. JAMES, R. C. JOHNS, D. B. PELOWITZ, L. S. WATERS, F. X. GALLMEIER, MCNPX, Version 2.6.C, Los Alamos National Labor atory Report: LA-UR-06-7991, Los Alamos, NM (2006). 85. M. L. FENSIN, J .S. HENDRICKS, S. ANGHAIE, Enhanced Monte Carlo Linked Depletion Capabilities in MCNPX, Los Alamos National Labor atory Report LA-UR-060363, 2006 International Congress on Advances in Nuclear Power Plants Embedded International Topical Meeting at the 2006 ANS Annual Meeting Reno, NV, (2006). 86. J. S. HENDRICKS, G. W. MCKINNEY, M. L. FENSIN, M. R. JAMES, R. C. JOHNS, J. W. DURKEE, J. P. FINCH, D. B. PELOWITZ, L. S. WATERS, F. X. GALLMEIER, MCNPX Version 2.6.E, Los Alamos National Labor atory Report: LA-UR-07-6632, Los Alamos, NM (2007). 87. J. S. HENDRICKS, G. W. MCKINNEY, M. L. FENSIN, M. R. JAMES, R. C. JOHNS, J. W. DURKEE, J. P. FINCH, D. B. PELOWITZ, L. S. WATERS, F. X. GALLMEIER, MCNPX Version 2.6.F, Los Alamos National Laborat ory Report: Draft Report, Los Alamos, NM (2007). 88. G. AUDI, A.H. WAPSTRA, C. THIBAULT, J. BLACHOT and O. BERSILLON "Ame2003: Atomic Mass Evaluation," November 2003, (November 2007). 89. S. K. GHOSHAL, M. GUPTA, and V. RAJARA MAN, A Parallel Multistep PredictorCorrector Algorithm for Solving Or dinary Differential Equations, Journal of Parallel and Distributed Computing 6 pg. 636-648 (1989). 90. A. E. ABOANBER, Numerical Solution of the chain model of fission product nuclides. Effect on the reactivity of the reactor, Annuls of Nuclear Energy 28, pg 923-933 (2001). 91. C. M. KANG, R. O. MOSTELLER, and S. LEVY Incorporation of a Predictor-Corrector Depletion Capability into the CELL-2 Code, Transaction of the American Nuclear Society 45, pp. 729 (1983). 92. M. L. FENSIN, J. S. HENDRICKS, H. R. TRELLUE, S. ANGHAIE Incorporation of a Predictor-Corrector Methodology and 1-GroupR eaction Rate Reporting Scheme for the MCNPX Depletion Capability, Los Alamos National Laboratory Report LA-UR-06-3925, American Nuclear Society s 2006 Winter Meeting Albuquerque, NM (2006). 93. Z. XU, J. RHODES III, K. SMITH, N. GHEORGHIU, MCNPX-5/ORIGEN-2.2/MCODE2.2 versus CASMO-5 Depletion for a H eavy Gd-Poisoned BWR Fuel Assembly, American Nuclear Societys 2007 Summer Meeting Boston, MA (2007).

PAGE 187

187 94. J. S. HENDRICKS, G. W. MCKINNEY, H. R. TRELLUE, J. W. DURKEE, J. P. FINCH, M. L. FENSIN, M. R. JAMES, D. B. PELOWITZ L. S. WATERS, F. X. GALLMEIER J. S. HENDRICKS, J. C. DAVID, MCNPX, Versi on 2.6.B, Los Alamos National Laboratory Report: LA-UR-06-3248, Los Alamos, NM (2006). 95. J. S. HENDRICKS, G. W. MCKINNEY, M. L. FENSIN, M. R. JAMES, R. C. JOHNS, J. W. DURKEE, J. P. FINCH, D. B. PELOWITZ, L. S. WATERS, F. X. GALLMEIER MCNPX, Version 2.6.D, Los Alamos National Laborat ory Report: LA-UR-07-4137, Los Alamos, NM (2007). 96. M. L. FENSIN, J. S. HENDRICKS, S. ANGHAIE, The Enhancements and Testing of the MCNPX Depletion Capability, Los Alamos National Laboratory Report: LA-UR-08-0305, To be presented at: International Congress on Advances in Nuclear Power Plants Embedded International Topical Meeting at the 2008 ANS Annual Meeting Anaheim, CA, (2008). 97. Organisation for Economic Co -operation and Development: Nuclear Energy Agency, NEA Expert Group on Burn-up Credit Criticality Safety, September 2003, (December 2007). 98. T. TAKEDA, N. HIROKAWA, and T. NODA, E stimate of Error Propagation in MonteCarlo Burnup Calculations, Journal of Nuclear Science and Technology, 36, No. 9 pp. 738 745 (1999). 99. P. PHRUKASAROJANAKUN and P. WILSON, "Nonanalog Implementations of Monte Carlo Isotope Inventory Analysis," Nuclear Science and Engineering 156, pp. 164-179 (2007).

PAGE 188

188 BIOGRAPHICAL SKETCH Michael Fensin was born on February 2 1980, to Dr. Matthew and Dr. Sheryl Fensin. Michael has one brother, Jeffrey Fensin. During his career as a student at the University of Floridas Nuclear and Radiological Engineeri ng Department, Michael has completed various internships. Michael got his first start, from Dr. William Vernetson, as a laboratory technician for the University of Floridas radiochemistry la boratory and training reac tor facility. Michael then moved on to complete a summer internship in reactor engineering with Southern Nuclear Company at the plant Vogtle site. During this internship Michael, participated in daily reactor operation and monitoring tasks. After his summer internship, Michael returned to the University of Florida to complete his Master of Engineer ing degree. During this degree, Michael also completed a six month internship with Globa l Nuclear Fuels (GNF), a General Electric Company, where he collaborated his masters de gree research efforts with work completed at GNF. At GNF, Michael developed optimum fuel bundle design strategies in order to meet the limiting constraints of shutdown reactivity manage ment. Michael then moved on to Los Alamos National Laboratory where he completed research in the area of compact reactor design for space applications. Michael then moved to the, MCNPX code development group, where he completed the work for this dissertation. Afte r graduation, Michael wi ll stay at Los Alamos National Laboratory and continue his work on advanced Monte Carl o linked depletion strategies for applied reactor safeguards efforts.