UFDC Home  myUFDC Home  Help 



Full Text  
xml version 1.0 encoding UTF8 REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchemainstance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd INGEST IEID E20101111_AAAABG INGEST_TIME 20101111T13:17:55Z 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 SHA1 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 ZeroD 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 TimeDependent 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 CELL2 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 31 M O X fuel com position. .......................................................................... .....................89 32 B orated w after com position. ...................................................................... ...................90 33 Emitted and recoverable energy per fission ofU235 ........ .......................................97 34 Prompt fission energy release data available in MCNPX .............................................98 35 Available actinide fission yield distributions in CINDER90................. ..................102 41 OM IT keyw ord variable tasks. .......................... ......... .................................................117 42 Options invoked for different B2 values. ................................... ........... ................... 118 43 Options invoked for different B2 values. ................................... ........... ................... 119 44 Available options for the B3 value of the BOPT keyword. .................. .................119 45 M ATM OD keyword variable values. ........................................ ......................... 121 51 Examined bench ark cases. ..... ........................... ......................................... 129 52 Model description for Phase IB Benchmark*........................ ... .................130 53 Fuel composition for Phase IB Benchmark (density =10.045 g/cm3). ..........................131 54 Clad composition for Phase IB Benchmark (6.44 g/ cm3)............................................132 55 Borated water composition for Phase IB Benchmark (0.7569 g/ cm3)............................132 56 Operating history data for Phase IB Benchmark. ................................. .................132 57 Operating power (MW) for each case for Phase IB Benchmark ...................................133 58 Initial Phase IB Benchmark results and comparison of Case A (27.35 MWD/MTU). ...133 59 Initial Phase IB Benchmark results and comparison of Case B (37.12 MWD/MTU).....134 510 Initial Phase IB Benchmark results and comparison of Case C (44.34 MWD/MTU).....135 511 MCNPXc Phase IB Benchmark results and comparison of Case A (27.35 M W D /M TU ). .............. ... ........ ....................................... ........ .. ......... 137 512 MCNPXc Phase IB Benchmark results and comparison of Case B (37.12 M W D /M TU ). .............. ... ........ ....................................... ........ .. ....... .. 138 513 MCNPXc Phase IB Benchmark results and comparison of Case C (44.34 M W D /M TU ). .............. ... ........ ....................................... ........ .. ......... 139 514 MONTEBURNS Phase IB Benchmark results and comparison of Case A (27.35 M W D/M TU). .................. ........ ................................ ........ .. ....... .. 144 515 MONTEBURNS Phase IB Benchmark results and comparison of Case A (37.38 M W D/M TU). .................. ........ ................................ ........ .. ....... .. 144 516 Fuel assembly design data for H. B. Robinson Benchmark. .........................................147 517 Operating data for H. B. Robinson fuel assembly. .................................. ............... 148 519 The H. B. Robinson Benchmark results using MCNPX compared with SCALE...........149 520 The H. B. Robinson Benchmark results using MCNPXc compared with SCALE. ........150 521 Fuel assembly design data for OECD/NEA Burnup Credit Phase IVB Benchmark.......152 522 Initial MOX fuel composition for Case A for Phase IVB Benchmark..........................153 523 Initial MOX fuel composition for Case B for Phase IVB Benchmark ..........................154 524 Borated water composition for Phase IVB Benchmark (660ppm boron, density 0 .72 4 5 g/cm 3)............................................................................................. . 154 525 Zicaloy2 composition used for clad and guide tube material for Phase IVB B enchm ark (density 5.8736 g/cm 3)......................................... ............................. 156 526 Operating data for Phase IVB Benchmark.................... .....................156 527 EOC 1 results for Case A of the Phase IVB Benchmark ............................................... 157 528 EOC 2 results for Case A of the Phase IVB Benchmark ............................................. 158 529 EOC 3 results for Case A of the Phase IVB Benchmark ............................................. 159 530 Five year cooling time results for Case A of the Phase IVB Benchmark....................... 160 531 EOC 1 results for Case B of the Phase IVB Benchmark. .........................................161 532 EOC 2 results for Case B of the Phase IVB Benchmark........ ............ ................. 162 533 EOC 3 results for Case A of the Phase IVB Benchmark ............................................ 163 534 Five year cooling time results for Case B of the Phase IVB Benchmark........................164 LIST OF FIGURES Figure page 21 L inked D epletion Process. ........................................................................ ...................46 22 Flux depression in the neighborhood of a resonance at Eo. .......................................... 51 23 Flux depression in the neighborhood of a resonance at Eo. ...........................................53 24 Flow diagram for the Monte Carlo particle transport process.........................................56 31 M onteCarlolinked depletion process model......................................... ............... 72 32 Uranium235 fission product yield distribution ...........................................................75 33 Fission Yield Distributions for Various Plutonium and Uranium Isotopes.81 ...................76 34 Original fission product array containing 167 fission products.................. ...............77 35 Resulting timedependent 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 36 Percent difference of certain fission products from MONTEBURNS using MCNPX Tier1 fission products with and without cross section models.................. ...............78 37 Percent difference of certain fission products from MONTEBURNS using MCNPX Tier2 and Tier3 fission products with crosssection models........................................79 38 Enhanced fission product array containing 220 isotopes............................. .............79 39 Nuclides created from a decay or nuclear reaction process and generated for particle transport by the isotope generator algorithm .3........... ............................................ 80 310 Examined infinitely reflected pin cell geometry ............................. ....8............89 311 Percent difference of certain reference actinides between the OLD MCNPX method and the M ONTEBURN S m ethod. ............................................ ............................. 91 312 Percent difference of certain reference actinides between the NEW MCNPX method and the M ONTEBURN S method. ............................................ ............................. 92 313 Percent difference of certain reference fission products between the NEW MCNPX method and the MONTEBURNS method. ............................................. ............... 92 314 Tim edependent eigenvalue calculations ........................................ ....................... 94 315 Fission yield selection process diagram ............................................... ......... ...... 104 316 MONTEBURNS cross section averaging technique ..................................................... 106 317 CELL2 PredictorCorrector m ethod .................................... ............................. ........ 107 318 Time dependant neutron multiplication for the predictor corrector method tests. ..........109 319 Burn steps implemented to achieve 2191 day burn ...................................................110 320 Time dependent neutron multiplication utilizing 5 burn steps. ............. ..................111 321 Time dependent neutron multiplication utilizing 10 burn steps. ..................................11 322 Percent Difference in keff When Compared with the NoPC 30Step Case ...................... 112 41 B urn card input structure. ..................................................................... ..................... 115 42 BURN card referencing burn materials ml, m3 and m4 .............................................117 43 E xam ple collision rate outputs ........................................................................... ... .... 123 44 Example burnup sum m ary table. ............................................ ............................. 124 45 Example individual material burnup data. ............................ ....................................124 46 Example cropped actinide output at two separate timesteps and fission product output for one time step. ................................... .. .... ........ ...............125 47 Cropped example of print table 220............ .......................................... .................126 51 Infinitely reflected pin cell geom etry........................................ ........................... 130 52 Infinitely reflected pin cell geom etry........................................ ........................... 146 53 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 Carlolinked depletion methods have gained recent interest due to the ability to more accurately model complex 3dimesional 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 highfidelity completely selfcontained MonteCarlolinked 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 selfsustaining 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 selfsustaining 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 selfsustaining 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 selfsustaining chain reaction.1 As the uranium rich deposits burned fissile uranium235, fission products and other transmuted actinides were generated. In some nuclear reactions, fertile actinides consumed radiation resulting in new fissionable actinides that along with uranium235 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 selfsustaining chain reaction. The system would then cease to maintain selfsustaining 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 selfsustaining 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 U235 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 selfsustaining chain reaction. On December 2, 1942, implementing a delicate balance of positive and negative reactivity, based on proper modeling and simulation, the first controlled selfsustaining 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 selfsustained 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 ofU235 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 subcooled and contains homogeneously dispersed boron poison in order to maintain criticality. The system utilizes an indirect dualcycle 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 U235 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 selfsustaining 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 wellinformed 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 moreinformed 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 longterm 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 nextgeneration reactors with the goal of (1) reducing volume of highlevel waste; (2) reducing longlived 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, carbonfree 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 nonproliferation and recycling nuclear fuel to reduce the nation's nuclear waste. This closed fuel cycle requires the development of technologies that enable consumption of longlived radioactive waste, and recycling of useful fissile isotopes.9'19 The following include the major objectives of the GNEP program: 1. New proliferationresistant 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 proliferationresistance 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 longterm 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 halflives 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 shortlived fission products. The shortlived 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 Pu239 as compared to the uranium isotopes, and faster reactor response, from the reduced delayed neutron fraction of Pu239, 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 multirecycling 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 fissiontocapture 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 longlived 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 2dimensional 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 nonpeaceful 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 proliferationresistant 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 nondestructiveanalysis (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 timedependant 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.2534 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 MonteCarlolinked depletion codes exist including ALEPH, BURNCALC, MCODE, MCREBUS, MCWO, MCB, MCMGBURN, MOCUP, MONTEBURNS, MVP and RACER.3749 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 multigroup 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 selfcontained, eliminate excess toplevel 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 highfidelity 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 easytouse 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 statereaction calculator to the CINDER90 number density calculator.50, 51 MCNPX will be used to calculate steadystate reaction rates and normalization parameters while CINDER90 will be used to calculate the timedependent 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 timedependent 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 easytoimplement 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 IB;52 12. H. B. Robinson Infinitely Reflected Pressurized Water Reactor Fuel Assembly Calculation;53 13. OECD/NEA Bumup Credit Calculational Criticality Benchmark Phase IVB.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 U235 depletes as a result of the fission process, fission products are generated. Because these newly created fission products posses capture reaction probabilities that rival U235 for neutron capture, the fission products, combined with the depletion of U235, reduce the propensity of the system to fission and maintain a steadystate chain reaction. In order to maintain a self sustaining steadystate 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 steadystate selfsustaining 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 timedependent 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 21 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 (21) dt kam dN dNm time dependant change in nuclide m dt Nm (t)Jm = destruction of nuclide m V Nk (t)Ykm = creation of nuclide m via other nuclides in the system kam Y = production of nuclide m via an external source Solving for the timedependent change of a specified nuclide also involves knowledge of the timedependent change of each contributing nuclide. Therefore the solution of each nuclides timedependent 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, Ykm clarifies the nature of the nonlinearity of equation 21: im = Am + f cm, (E)(r,E,t)dE (22a) r 7k>, = Lkm+k I JYkmr(E)+k,r(E)()(r,E,t)dE (22b) 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 timedependent flux. Unfortunately, the time dependent flux is also dependent upon the timedependent nuclide inventory. To make equation 21 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 timedependent 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 21. 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 Xe135 and Sm149 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 21 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 timestep, assuming a constant transmutation rate for a given timestep is valid.56 ZeroDimension Depletion Calculation The depletion equation requires knowledge of reaction rates, decay coefficients and isotope concentrations at a specific timestep in order to solve for the isotope concentrations at the subsequent timesteps. 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 21 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 23:51' 5 6.242*ld8(P) (23) = 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 24: Power= fI(r)QYfdV (24) 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 21 may be recast into the following form d n (r, t)= Bk (r,t),, (r,t) (25) dt k where: rLkmAk +J kmrk,ro(r,t) for m k Bmk (r, t)= (rr = k (26) ^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) (27) where N is a column vector of nuclide densities and B is the matrix of elements calculated of equation 26. The function exp(Bt) is a matrix exponential function, and is therefore calculated from the following expansion: exp(Bt) (Bt) (28) 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 halflives of each isotope tracked during the depletion process.59 Certain isotopes possess halflives 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 (29) n=o where c" is generated by use of the following recursive relationship c, = x, (0) (210a) n+1 t N C ac (210b) n+lJ In this solution method a, is an element in the matrix that is the firstorder 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 29 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 halflives. 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 GaussSiedel 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 reactorfuel 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 21 in order to calculate the timedependent nuclide densities: dN, dN = N (t) + r, + Nk (t)k, (211) dt knm where all terms are the same as equation 21 except that Ym is included to represent the average feed rate. Like equation 21, equation 211 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 timedependent 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, (i1) element, all parameters are assumed known. The entire computation is then computed from the simplified equation dN  d = + N,_ (t)y, N, (t)/, (212) 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 (213) {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 (214) Y, =0,m 1 (215) ultimately resulting in the following solution for a given linear sequence:61 nl 1 efly N 1 +N (216) = 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 timeintegrated transmutation of that nuclide for a specified time interval: (t) = IN, (t)/,dt (217) 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. 211). If the code is used by itself to calculate nuclide inventory, the user would be relying heavily upon the inherent multigroup cross sections, utilized to determine collapsed energy integrated reaction rates, in order to achieve a solution. Since CINDER90 is a zerodimensional 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 nonperiodic 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 energyintegrated 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 ofU235 atoms. As the core burns over time many neutrons are parasitically absorbed by U238 leading to production ofU239. U239 quickly decays by beta emission to Np239 and then to Pu239 which is highly fissile. At the end of the reactor core life, an immense amount of Pu239 has been created, absorbed neutrons, fissioned, and contributed to system power. The average thermal absorption cross section for Pu239 is 1011.3 b which is much higher as compared to U235 at 680.8 b.7 The average thermal fission cross section for Pu239 is 742.5 b which is also much higher as compared to U235 at 582.2b.7 capture cross section is defined as all absorption reactions not leading to fission, and therefore a thermal capturetofission ratio may be defined for a specified isotope by equation 218: af (218) f Changes in the capturefissionratio 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 timedependant 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 21. Linked Depletion Process. Since considerable changes in the isotope concentration are required in order to significantly alter the neutron energy spectrum, equation 21 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 21. The reaction rates utilized in the capture terms of equation 21, system eigenvalue, and flux normalization terms are first calculated utilizing a steadystate reaction rate calculator. Those capture reactions are then assumed constant over a specified time step and implemented in a timedependent number density calculation. The timedependant number density calculation then calculates a new set of number densities to be utilized in the subsequent steadystate 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 steadystate LBE is defined as equation 219: 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') + (219) 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 219 into a set of coupled differential equations, where the coupling of energy groups results from the outgroup scatter and fission terms: Q V V,g (F, Q) + (F), (F Q)= G I I G (220) f JdQ gg', (,F' Q)g (F, Q') + va,, (F)JdQ' Vg' (F, (') g'=1 g'=1 The angular dependence of equation 220 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 infinitedilute, 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 selfshielding. Figure 22 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 selfshielding 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 selfshielding in order to properly model the true system physics. S1/F / 4(E) Eo E Figure 22. Flux depression in the neighborhood of a resonance at E0. Casting equation 219 into the multigroup form requires that multigroup coefficients be generated. The coefficients are generated based on the premise that the multigroup 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 221: Eg1 f c(r, E)O(r, E)dE ca(r), = E (221) JD(r,E)dE The concern with this approach is that the flux, the variable that is to be to calculated in equation 220, is used to calculate the cross section which is then utilized to calculate the flux in equation 220 making the process nonlinear. To make the process linear, an approximation is made in which first a homogenous, spatiallyinfinite, 1D 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 selfshielded cross sections for a further multigroup 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 23 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/Cm244:27 Cm244 T=300 K from ENDF/BVI.O mat9637  green mcnp/Pu239:27 Pu239 T=300 KfromENDF/B6.2 mat9437 blue mcnp/U235:27 U235T=300K fom ENDF/B6.2mat9228 purple J mcnp/Nd143:27 Nd143 T=300 K fom ENDF/B6 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 23. 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 multigroup 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 uraniumoxide fuel, with zircaloy clad, surrounded by water, and possibly controlled by borncarbide or hafnium.25 26 30 32 However, those libraries are not applicable for examining exotic isotope combinations such as uraniumnitride fueled, rhenium cladded and niobiumcarbide 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, nonstochastic, 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 24. 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 onedimensional 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 222: b jf (x)dx = P{a < x < b} (222) 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 223: Jf(x)dx= 1 (223) on From this equation we can further derive the cumulative probability distribution function (CDF), F(x) in equation 224, which will be utilized to ultimately determine the actual length of particle travel: f (x)dx = F(x) = P{x'< x} (224) 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 24. 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 (225) and the associated CDF is F(x) = e x (226) Therefore, inverting the cumulative probability distribution function to solve for the mean free path results in x ln(1 ) (227) and if we choose to be uniformly distributed between 0 and 1 we can further write x = ln() (228) or in terms of track length, 1, utilizing the individual isotope number density N, and microscopic cross section total cross section, o(E),, = ln() (229) 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 ) (230) 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), (231) 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 (232) 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: m1 N m _p(E)k, < p(E)k,, < p(E)k,, (233) 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 (234) For the Monte Carlo computation the flux may therefore be approximated by: = (235) 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 (236) 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 (237) 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 (239) 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 n1 is used in the divisor to account for this underestimating. For a sample distribution, the sample variance of that distribution is calculated by equation 238: S2 (_ )2 (238) n1 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 nl nI n1 n1 n1 n x2 _2 (239) n1 The sample variance is referred to having n1 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 n1 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 238 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 ..) (240) 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 (241) 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 / = (242) 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 (243) 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) (244) ,,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) (245) eff N n ,n, (E) kf= 1 n ,,n,f (E)u,,n (E) (246) 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) (247) 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 industrybased depletion codes is to support the operation and fuel cycle design of commercial power reactors. Industrybased 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 deterministiclinked depletion methodologies.35 Although various companies, such as Westinghouse, General Electric, Areva, and Studsvick, use slightly different techniques to approximate timedependent reactor behavior, the overall methodology and processes are quite similar. First, a finefidelity deterministic calculation (using PHOENIX, TGBLA, CASMO3, etc.) examining many axial and radial slices of the reactor are completed to generate coarser 1 to 4 groupaveraged interaction rate parameters for implementation in a coarser calculation of the fullcore 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 selfshielding 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 multigroup code developed for and LWR with certain group structure, fission products, and reactions followed will not be accurate in determining individual gammaline 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. Academiabased 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 higherfidelity 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 finerenergy 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 selfshielding considerations. To eliminate dependence from this burden, burnup codes coupling the Monte Carlo reaction rate calculation have received more attention.3641'4648'75 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 boron10. Further acceleration techniques for Monte Carlo may include using multigroup Monte Carlo such as in Keno Va.76 Multigroup 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 multigroup 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 multipurpose 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 timedependent 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 multigroup or stochastic calculation (using MCNP, PHOENIX, TGBLA, CASMO3, 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 fullcore 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 microsection 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 multipurpose 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) steadystate 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 31. MCNPX runs a steadystate calculation to determine the system eigenvalue, 63group fluxes, collision rates, v, and Q values. CINDER90 then takes those MCNPXgenerated 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, timedependant 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 31. MonteCarlolinked 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: steadystate 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 halflife 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 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 steadystate 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 63group 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 63group spatially dependent flux is calculated within MCNPX and sent to CINDER90 to be match with a 63group 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/greatgranddaughter/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 twotothree 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 32 displays the fission product distribution for U235 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 33 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 32. Uranium235 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 33. 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 34, with the following fission product tiers:79 1. Tier 1. (default) Zr93, Mo95, Tc99m, Ru101, Xe131, Xe134, Cs133, Cs137, Ba 138, Pr141, Nd143, Nd145.9 2. Tier 2. Isotopes contained in the fission product array that are included in the released crosssection 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 34. 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 35 displays the difference in neutron multiplication versus time and Figure 36 and Figure 37 display the difference in endoflife 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 38.84 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 35. Resulting timedependent 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  S5.00%  S10.00%  15.00% Nuclides o Tier 1 without Models Tier 1 with Models Figure 36. Percent difference of certain fission products from MONTEBURNS using MCNPX Tier1 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 37. Percent difference of certain fission products from MONTEBURNS using MCNPX Tier2 and Tier3 fission products with crosssection 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 38. 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 1E36. 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 39. 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 steadystate 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 39 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 multiinteraction 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 TimeDependent 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 timedependant 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 postexecuting 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 (31) 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 (32) 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 (33) 1fn 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 (34) 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 1fn 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%, (35a) p, =w%, *NA *p (35b) 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, (35d) (35e) Y w%k + w%, w%k k MAk k AMk w%k *N * A Mk P,k AFk %k N P k Mk (36f) 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 (35c) 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 (37) 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 (38) 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 Ag110m, 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 (39) For example: Ag110m = 47510 47510 = 47110 + 1st metastable (310) 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 21, 22a and 22b, 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 steadystate 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 63group cross section set for all reactions available in the library file. Therefore, CINDER90 theoretically only requires a 63 group flux to match to its 63group 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 tradeoff 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 63group flux in MCNPX to a 63group 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 63group flux calculated within the steadystate Monte Carlo to a 63group 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 63group matching approach mentioned previously. The concern of the 63group matching approach was that the validity of this technique resided in the applicability of the 63group crosssection set to various designs. The 63group 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 63group cross section set within CINDER90 generated cross sections that were reliable for calculations only representative of that system flux. Therefore calculations implementing the 63group 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 multigroup 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 timedependent 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 310. Borated Water Clad Fuel Figure 310. Examined infinitely reflected pin cell geometry. Table 31. MOX fuel composition. ZA value Atom Density [atoms/cm3] 8016 4.5854e2 92235 1.4456e4 92238 1.9939e2 94238 1.1467e4 94239 1.0285e3 94240 7.9657e4 94241 3.3997e4 94242 5.6388e4 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 31. The entire fuel pin was encased within a borated water rectangle 365 cm tall with a 1.313cm length and 1.313cm width. The borated water composition is listed in Table 32. Table 32. Borated water composition. ZA value Atom Density [atoms/cm3] 1001.60c 4.7716e2 8016.60c 2.3858e2 5010.60c 3.6346e6 5011.60c 1.6226e5 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 steadystate 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 steadystate 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 S40.00% S60.00% S80.00% 100.00% 120.00% Nuclides Figure 311. 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 63group Monte Carlo flux to be matched to a generalized 63group CINER90 crosssection set. The integral continuous energy reaction rates were calculated separately for each burn material and tallied at the site of collision. Figure 312 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 Np237 drops from 70.24% to 4.85%, and the percent difference in Pu237 production drops from 95.69% to 1.02%. Due to accentuation of U237, Np237, and Pu237 differences in Figure 311, Figure 312 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, 12% improvements in actinide predictions are also witnessed in U236, U239 and Pu238. 2.00% 1.00% 0.00% 1.00% 2.00% 3.00% 4.00% 5.00% 6.00% M 0 0 Nuclides Figure 312. Percent difference of certain reference actinides between the NEW MCNPX method and the MONTEBURNS method. 2.00% 0.00% S200% 8 4.00% 2 6.00%  8.00% 10.00%  Nuclides Figure 313. 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 313 displays the difference in EOL fission product prediction when comparing the new MCNPX method to MONTEBURNS. Unfortunately, timedependent fission product buildup is a major contributor to negative reactivity trajectories, and therefore MCNPX and MONTEBURNS only predicted slightly different timedependent reactivity. Figure 314 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 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 timedependent cross section averaging technique. Figure 314 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 314. Timedependent 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 311: Pxv total = MNP x Flux Multiplier > Flux Multiplier = (311) 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 311. 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 33. Emitted and recoverable energy per fission ofU235. Form Emitted Energy (MeV) Recoverable Energy (MeV) Fission Fragments 168 168 Fission Product Decay Gammarays 8 8 Betarays 7 7 Neutrinos 12 Prompt Gamma Rays 7 7 Fission Neutrons (Kinetic 5 5 Energy) Capture Gamma Rays  312 Total 207 198207 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 33 lists the emitted and recoverable energy per fission from each micro fission event for U235.7 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 34. The sum of EFR, ENP and EGP for each actinide is the prompt fission release energy for that actinide. Table 34. 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 Qvalue 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 U235 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 312, 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) (312) 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 (313) 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 313 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 313 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 32 and Figure 33. 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 35. 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, highenergy fission yield for each fissile isotope contained in the CINDER90 data file.61 Table 35 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 energydependent 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 315. 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 energydependent fission rate of the analyzed system.8 Figure 315. 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 21 we see that the depletion equation is only linear if the depletion coefficients are constant; however, in reality the depletion coefficients are also timedependent. 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 21 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 314 illustrates how power is proportional to the macroscopic fission cross section multiplied by the flux: Power ~C, (314) 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 timedependent 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 nonlinear equations;46 74,8991 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 predictorcorrector method. 46, 74, 89,9192 First, an assumption is made about the timeintegrated 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 timedependent parameters. Predictorcorrector 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 Cell2 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 Cell2 method follows a more traditional predictorcorrector 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 316. MONTEBURNS cross section averaging technique. The MONTEBURNS cross section averaging procedure is displayed in Figure 316.74 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. CELL2 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 317. CELL2 PredictorCorrector method. The CELL2 predictorcorrector method is displayed in Figure 317.91 The CELL2 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 nopredictorcorrector method (NoPC); 2. MONTEBURNS predictorcorrector method (PC1); 3. CELL2 predictorcorrector 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 timesteps approached infinity the calculation would approach true behavior. Neutron multiplication versus time for the examined cases is displayed in Figure 318. 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 318. 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 320 and Figure 321 display the timedependent neutron multiplication versus burnup for the different predictor corrector cases. Both predictor corrector methods, PC1 and PC2, calculate similar neutron multiplications at each timestep, 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 322 displays the percent difference in neutron multiplication of each case when compared with the NoPC 30step 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 2.48E02 2.48E02 2.48E02 2.48 4.09E01 4.09E01 4.09E01 1.94 7.93E01 7.93E01 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 319. 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 2.48E02 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 320. 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 321. 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 322. Percent Difference in keff When Compared with the NoPC 30Step Case Further Commentary on both Cross Section Averaging Techniques Both methods tested were not the exact original method employed within both codes. Initially, the CELL2 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 317 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 CELL2 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 CELL2 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 timestep 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 selfcontained 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 selfcontained 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.AF.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 41 displays the input structure for the BURN card. Each input variable is then explained below Figure 41. 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 41. 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 100day bum, followed by a 15day 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 42 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 nonactinide 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 4.5854e2 92235.60c 1.4456e4 92238.60c 1.9939e2 94238.60c 1.1467e4 94239.60c 1.0285e3 94240.60c 7.9657e4 94241.60c 3.3997e4 94242.60c 5.6388e4 m2 2004 1.0 m3 40000.60c 1.0 m4 1001.60c 4.7716e2 8016.60c 2.3858e2 5010.60c 3.6346e6 5011.60c 1.6226e5 mt4 lwtr.0lt Figure 42. 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 63group flux into CINDER90, matching the 63group flux calculated from MCNPX to a 63group 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 41 contains the available options for this keyword. Table 41. 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.OE10. The A2 value sets the decay chain convergence criteria (mentioned in the Benchmarks section). The default for this value is now le10 and was le4 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 42. 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 42 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 fissionproduct 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 43 gives a description of the available actinides in each fission product tier. Table 43. Options invoked for different B2 values. Tier Fission Product Content 1 Zr93, Mo95, Tc99m, Ru101, Xe131, Xe134, Cs133, Cs137, Ba138, Pr141, Nd143, Nd145 (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 38. The B3 value of the BOPT keyword invokes the models option. Crosssection models are a heritage of the highenergy physics capabilities of MCNPX. The models are benchmarked for a variety of problems for neutron energies >150 MeV. Table 44 describes the available options for the B3 value of the BOPT keyword. Table 44. 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"* (41) 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 41 was altered to Z = I (Y,,1*w,) (42) 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 45. 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 timedependent manual nuclide concentration changes. The details of the methodology for this feature were explained in the Manual TimeDependent NMNT MTNT,NMNT KNT, NMINT NT ,NMNT C NT,' NT z . C Isotope Concentration Changes section. Table 45 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 steadystate KCODE calculation. The error associated with each collision rate is the error assuming the steadystate calculation is independent from the propagated error resulting from propagated flux and number density error from the burnup calculation. Figure 43 displays a cropped example of the format of the collision rate outputs. These outputs help the user to calculate generalized capturefission 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 43. 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 44 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 2.000E01 1.43413 1.699E+14 2.510 202.199 2.155E+01 1.550E+16 Figure 44. 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 5.015E01 0.000E+00 1 5.000E+01 5.000E+01 5.016E01 7.205E+00 2 1.000E+01 6.000E+01 5.002E01 7.205E+00 3 5.000E+02 5.600E+02 5.002E01 2.158E+01 Material #: 4 step duration time power fraction bumup (days) (days) (GWd/MTU) 0 0.000E+00 0.000E+00 4.985E01 0.000E+00 1 5.000E+01 5.000E+01 4.984E01 7.161E+00 2 1.000E+01 6.000E+01 4.998E01 7.161E+00 3 5.000E+02 5.600E+02 4.998E01 2.152E+01 Figure 45. Example individual material burnup data. If multiple materials are burned then there also exists individual material burnup data. Figure 45 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/bcm) 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 2.293E03 1.000E01 9.886E02 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 1.286E09 6.837E04 2 90232 2.394E08 2.625E15 spec.act. atom den. atom fr. mass fr. (Ci/gm) (a/bcm) 5.315E+05 8.718E15 3.832E13 3.723E13 1.097E07 1.616E13 7.100E12 6.929E12 totals 3.455E+03 2.584E+05 7.479E+01 2.275E02 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/bcm) 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 4.585E02 1.000E+00 1.000E+00 Figure 46. Example cropped actinide output at two separate timesteps and fission product output for one time step. (43) (44) 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 46 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 47 shows a cropped example of print table 220. Iburnup summary table summed over all materials print table 220 nuclides with atom fractions below 1.000E32 for a material are zeroed and deleted from print tables after t0 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/bcm) 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 4.585E03 1.000E01 9.886E02 Figure 47. 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 cooperation, 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 burnup credit as applied to criticality safety in the transportation, storage, and treatment of spent fuel for a wide range of fuel types, including urainiumoxides (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 codetocode 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 51 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 51. 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 pincell 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 fueltomoderator 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 51. Infinitely reflected pin cell geometry. Table 52. 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 51 and the fuel pin dimensions are listed in Table 52. 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 Zicaloy2 clad region. Finally encasing the entire geometry was a rectangular borated water box. The fuel, clad, and borated water concentrations are listed in Table 53, Table 54 and Table 55. 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 56 details the common operating strategies utilized for each case. Table 57 lists the operating powers utilized for each individual case in order to achieve the prescribed burnups. Table 53. Fuel composition for Phase IB Benchmark (density =10.045 g/cm3). Nuclide Number Density (atoms/bcm) 234U 6.15164E6 235U 6.89220E4 236U 3.16265E6 238U 2.17104E2 12C 9.13357E6 14N 1.04072E5 160 4.48178E2 ENDF/B VII.0 temperature dependent cross sections and scattering law data were implemented in the steady state reaction rate calculation. Each steadystate 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 54. 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 55. Borated water composition for Phase IB Benchmark (0.7569 g/ cm3). Nuclide Number Density (atoms/bcm) H 5.06153E2 160 2.53076E2 10B 2.75612E6 11B 1.11890E5 Table 56. 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 57. 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 58. Initial Phase IB Benchmark results and comparison of Case A (27.35 MWD/MTU). Isotope MCNPX Value (mg/g UO2) 234U 1.408E1 235U 7.988 236U 3.227 238U 8.375E+2 237Np 2.764E1 238Pu 9.046E2 239Pu 3.790 240Pu 1.599 241Pu 6.491E1 242Pu 2.952E1 241Am 2.274E1 243Am 4.487E2 95Mo 5.501E1 99Tc 5.397E1 133Cs 8.012E1 135Cs 3.535E1 143Nd 5.752E1 145Nd 4.886E1 147Sm 1.522E1 149Sm 1.628E3 15sSm 1.762E1 151Sm 7.070E3 152Sm 7.940E2 153Eu 7.022E2 155Gd 2.629E3 Experiment Value (mg/g UO2) 1.600E1 8.470 3.140 8.425E+2 2.680E1 1.012E1 4.264 1.719 6.812E1 2.886E1 N/A N/A N/A N/A 8.500E1 3.600E1 6.130E1 5.100E1 N/A 2.900E3 2.070E1 N/A 8.700E2 7.900E2 N/A *(Calculated/Measured1)* 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 1.330E1 to 1.750E1 7.445 to 8.661 3.128 to 3.540 8.637E+2 to 8.415E+2 2.527E1 to 3.396E1 5.721E2 to 1.083E1 3.660 to 4.690 1.573 to 8.60 5.310E1 to 7.335E1 2.000E1 to 3.192E1 2.269E1 to 2.598E1 3.480E2 to 4.672E1 5.590E1 to 5.795E1 5.648E1 to 6.904E1 6.820E1 to 8.640E1 3.728E1 to 3.959E1 6.040E1 to 6.792E1 4.984E1 to 5.151E1 1.564E1 to 1.932E1 1.626E3 to 2.900E3 1.713E1 to 2.146E1 6.376E3 tol.413E2 7.947E2 to 1.073E1 6.730E2 to 8.921E2 1.507E3 to 5.762E3 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 58, Table 59 and Table 510. 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 59. Initial Phase IB Benchmark results and comparison of Case B (37.12 MWD/MTU). MCNPX Value (mg/g UO2) 1.148E01 4.587 3.641 8.303E+02 3.980E01 1.727E01 3.849 2.057 8.379E01 5.916E01 2.853E01 1.181E01 7.066E01 6.898E01 1.022 3.793E01 6.523E01 6.112E01 1.607E01 1.653E03 2.311E01 7.397E03 9.521E02 9.940E02 4.076E03 Experiment Value (mg/g UO2) 1.400E01 5.170 3.530 8.327E+02 3.560E01 1.893E01 4.357 2.239 9.028E01 5.761E01 N/A N/A N/A N/A 1.090 4.000E01 7.160E01 6.530E01 N/A 3.000E03 2.710E01 N/A 1.040E01 1.090E01 N/A *(Calculated/Measured1)* 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 1.080E1 to 1.570E01 4.022 to 5.510 3.526 to 3.930 8.292E+02 to 8.360E+02 3.619E01 to 4.919E01 1.144E01 to 2.069E01 3.710 to 4.877 1.996 to 2.347 7.510E01 to 9.846E01 4.200E01 to 6.347E01 2.880E01 to 3.418E01 9.637E02 to 1.391E01 7.214E01 to 7.545E01 7.327E01 to 8.372E01 8.784E01 to 1.117 3.967E01 to 4.317E01 7.013E01 to 8.254E01 6.326E01 to 6.600E01 1.659E01 to 2.201E01 1.736E03 to 3.092E03 2.297E01 to 3.152E01 7.990E03 to 1.571E02 9.761E02 to 1.416E01 9.960E02 to 9.960E02 2.538E03 to 1.028E02 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 510. 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 9.896E02 1.200E01 17.53 9.030E02 to 1.440E01 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 4.794E01 4.680E01 2.44 4.327E01 to 5.934E01 238Pu 2.399E01 2.688E01 10.76 1.656E01 to 2.810E01 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 9.229E01 1.020 9.52 8.560E01 to 1.111 242Pu 8.479E01 8.401E01 0.93 5.960E01 to 9.103E01 241Am 3.071E01 N/A N/A 3.102E01 to 3.785E01 243Am 1.924E01 N/A N/A 1.634E01 to 2.316E01 95Mo 8.100E01 N/A N/A 8.092E01 to 8.742E01 99Tc 7.852E01 N/A N/A 8.449E01 to 9.861E01 133CS 1.160 1.240 6.45 9.723E01 to 1.286 135Cs 3.958E01 4.300E01 7.94 3.977E01 to 4.605E01 143Nd 6.746E01 7.630E01 11.58 7.397E01 to 8.839E01 145Nd 6.858E01 7.440E01 7.82 7.170E01 to 7.560E01 147Sm 1.598E01 N/A N/A 1.655E01 to 2.302E01 149Sm 1.686E03 4.700E03 64.13 1.842E03 to 3.286E03 15Sm 2.690E01 3.610E01 25.49 2.725E01 to 3.980E01 15Sm 7.585E03 N/A N/A 8.102E03 to 1.682E02 152Sm 1.038E01 1.210E01 14.25 1.077E01 to 1.587E01 153Eu 1.173E01 1.480E01 20.75 1.210E01 to 1.596E01 155Gd 5.046E03 N/A N/A 3.379E03 to 1.318E02 *(Calculated/Measuredl)* 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 halflive 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 511. 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 1.422E01 1.600E1 11.13 1.330E1 to 1.750E1 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 2.794E01 2.680E1 4.25 2.527E1 to 3.396E1 238Pu 9.445E02 1.012E1 6.67 5.721E2 to 1.083E1 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 6.615E01 6.812E1 2.90 5.310E1 to 7.335E1 242Pu 3.069E01 2.886E1 6.36 2.000E1 to 3.192E1 241Am 2.320E01 N/A N/A 2.269E1 to 2.598E1 243Am 4.886E02 N/A N/A 3.480E2 to 4.672E1 95Mo 5.661E01 N/A N/A 5.590E1 to 5.795E1 99Tc 5.661E01 N/A N/A 5.648E1 to 6.904E1 133Cs 8.259E01 8.500E1 2.83 6.820E1 to 8.640E1 135Cs 3.620E01 3.600E1 0.54 3.728E1 to 3.959E1 143Nd 5.840E01 6.130E1 4.73 6.040E1 to 6.792E1 145Nd 4.894E01 5.100E1 4.04 4.984E1 to 5.151E1 147Sm 1.518E01 N/A N/A 1.564E1 to 1.932E1 149Sm 1.679E03 2.900E3 42.10 1.626E3 to 2.900E3 150Sm 1.854E01 2.070E1 10.44 1.713E1 to 2.146E1 15Sm 7.401E03 N/A N/A 6.376E3 tol.413E2 152Sm 8.124E02 8.700E2 6.62 7.947E2 to 1.073E1 153Eu 7.257E02 7.900E2 8.13 6.730E2 to 8.921E2 55Gd 2.724E03 N/A N/A 1.507E3 to 5.762E3 *(Calculated/Measuredl)* 100 A further investigation was completed to set the limiting checking value for density contribution of a specific nuclide to 1E10 from the default 1E4 value in order to determine if this change actually affected the ultimate depletion outcome. Table 511, Table 512 and Table 513 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 512. 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) 1.164E01 4.822 3.619 8.303E+02 4.032E01 1.750E01 3.987 2.120 8.699E01 5.936E01 2.976E01 1.186E01 7.293E01 7.277E01 1.060 3.930E01 6.750E01 6.172E01 1.632E01 1.711E03 2.521E01 7.924E03 9.892E02 1.035E01 4.251E03 *(Calculated/Measuredl)* 100 % Error* Rang of Values from other Codes Experiment Value (mg/g UO2) 1.400E01 5.170 3.530 8.327E+02 3.560E01 1.893E01 4.357 2.239 9.028E01 5.761E01 N/A N/A N/A N/A 1.090 4.000E01 7.160E01 6.530E01 N/A 3.000E03 2.710E01 N/A 1.040E01 1.090E01 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 1.080E1 to 1.570E01 4.022 to 5.510 3.526 to 3.930 8.292E+02 to 8.360E+02 3.619E01 to 4.919E01 1.144E01 to 2.069E01 3.710 to 4.877 1.996 to 2.347 7.510E01 to 9.846E01 4.200E01 to 6.347E01 2.880E01 to 3.418E01 9.637E02 to 1.391E01 7.214E01 to 7.545E01 7.327E01 to 8.372E01 8.784E01 to 1.117 3.967E01 to 4.317E01 7.013E01 to 8.254E01 6.326E01 to 6.600E01 1.659E01 to 2.201E01 1.736E03 to 3.092E03 2.297E01 to 3.152E01 7.990E03 to 1.571E02 9.761E02 to 1.416E01 9.960E02 to 9.960E02 2.538E03 to 1.028E02 (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 513. 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) 1.006E01 3.131 3.758 8.247E+02 4.854E01 2.434E01 3.970 2.362 9.633E01 8.463E01 3.222E01 1.919E01 8.359E01 8.303E01 1.206 4.104E01 7.038E01 6.930E01 1.635E01 1.751E03 2.975E01 8.239E03 1.084E01 1.228E01 5.289E03 % Error* Rang of Values from other Codes Experiment Value (mg/g UO2) 1.200E01 3.540 3.690 8.249E+02 4.680E01 2.688E01 4.357 2.543 1.020 8.401E01 N/A N/A N/A N/A 1.240 4.300E01 7.630E01 7.440E01 N/A 4.700E03 3.610E01 N/A 1.210E01 1.480E01 N/A *(Calculated/Measured)* 100 Nuclear Data The MCNPX benchmark calculations used ENDF/B VII.0 temperature dependent cross section data for steadystate 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 9.030E02 to 1.440E01 2.389 to 3.716 3.641 to 4.030 8.234E+02 to 8.316E+02 4.327E01 to 5.934E01 1.656E01 to 2.810E01 3.659 to 4.902 2.180 to 2.661 8.560E01 to 1.111 5.960E01 to 9.103E01 3.102E01 to 3.785E01 1.634E01 to 2.316E01 8.092E01 to 8.742E01 8.449E01 to 9.861E01 9.723E01 to 1.286 3.977E01 to 4.605E01 7.397E01 to 8.839E01 7.170E01 to 7.560E01 1.655E01 to 2.302E01 1.842E03 to 3.286E03 2.725E01 to 3.980E01 8.102E03 to 1.682E02 1.077E01 to 1.587E01 1.210E01 to 1.596E01 3.379E03 to 1.318E02 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 timesteps 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 fueltomoderator 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 le10 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 514 and Table 515. 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 514. 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 1.422E01 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 2.794E01 6.65 6.11 4.25 238Pu 9.445E02 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 6.615E01 2.72 9.77 2.90 242Pu 3.069E01 6.65 12.07 6.36 133Cs 8.259E01 1.91 6.92% 2.83 135Cs 3.620E01 4.46 1.40% 0.54 143Nd 5.840E01 0.36 5.51% 4.73 15Sm 1.854E01 13.22 14.51% 10.44 152Sm 8.124E02 1.35 6.39% 6.62 153Eu 7.257E02 5.11 5.71% 8.13 *MONTEBURNS OLD = MONTEBURNS from prior thesis;75 MONTEBURNS NEW =MOTNEBURNS 2.0 with higher fidelity model ***(Calculated/Measuredl)* 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 515. 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 1.164E01 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 4.032E01 14.38 36.40 13.26 238Pu 1.750E01 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 8.699E01 4.05 21.07 3.65 242Pu 5.936E01 7.41 77.15 3.04 133CS 1.060 2.55 20.93 6.24 135Cs 3.930E01 4.79 7.42 5.17 143Nd 6.750E01 0.76 11.81 8.90 150Sm 2.521E01 8.96 30 14.72 152Sm 9.892E02 0.20 27 8.45 153Eu 1.035E01 13.17 30 8.81 *MONTEBURNS OLD = MONTEBURNS from prior thesis;75 MONTEBURNS NEW =MOTNEBURNS 2.0 with higher fidelity model ***(Calculated/Measuredl)* 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 bumup 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 52. 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/TM12667.28, 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 52. Table 516 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/TM12667.28.53 Table 516. 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 Zircaloy4 0.6502 cm 0.6934 cm Zircaloy4 0.6502 cm 0.6934 cm Zircaloy4 0.5677 cm 0.6007 cm 0.6172 cm 1.0058 cm 1.0173 cm 1.1151 cm Each steadystate 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 3dimensions, 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 AD 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 517 and Table 518 display the operating intervals, time steps used for each interval, corresponding down times, and cycle dependent soluble boron concentrations for Case AD. Table 517. 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 AB or 3637 for Cases CD. Table 518. 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/TM12667.53 The SCALE sequence used in the calculations was SAS2H/ORIGENS analysis sequence and 27group crosssection library (27BURNUPLIB) of SCALE4.53 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 520. Table 519. 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 518 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 AB was <8%; however, at higher burnups, Cases CD, 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 519, the results show significant improvement in actinide prediction for uranium and plutonium isotopes when utilizing MCNPXc. Table 520. 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 AB, the percent difference in actinide prediction from measured experiment is <4% except for Np237 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 521. 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 Zircaloy2 Guide and Instrument tube data Inner radius 0.571 cm Outer radius 0.613 cm Material Zircaloy2 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 Pu239 and large initial loading of Pu239 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 Pu239 possesses a smaller delayed neutron fraction (P = 0.0021) and shorter neutron lifetime (1 = 27as), than U235 (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 codetocode 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 codetocode comparison for MOX assembly modeling strategy. Table 522. Initial MOX fuel composition for Case A for Phase IVB Benchmark. Values Expressed in (atoms/bcm) Nuclide High Enrichment Medium Enrichment Low Enrichment 234U 2.57E07 2.64E07 2.68E07 235U 5.38E05 5.53E05 5.60E05 238U 2.12E02 2.18E02 2.21E02 238Pu 5.17E05 3.61E05 2.85E05 239Pu 1.13E03 7.87E04 6.20E04 24Pu 5.35E04 3.74E04 2.95E04 241Pu 1.94E04 1.36E04 1.07E04 242Pu 1.46E04 1.02E04 8.06E05 160 4.66E02 4.66E02 4.65E02 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 Zircaloy2 clad fuel pins along with Zircaloy2 guide tubes. The geometry specifications for both cases are given in Table 521. 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 523. Initial MOX fuel composition for Case B for Phase IVB Benchmark. Values Expressed in (atoms/bcm) Nuclide High Enrichment Medium Enrichment Low Enrichment 234U 2.69E07 2.73E07 2.75E07 235U 5.63E05 5.71E05 5.74E05 238U 2.22E02 2.25E02 2.26E02 238Pu 5.09E07 3.56E07 2.81E07 239Pu 9.49E04 6.64E04 5.23E04 240Pu 6.06E05 4.24E05 3.34E05 241Pu 3.02E06 2.11E06 1.66E06 242Pu 5.01E07 3.50E07 2.76E07 160 4.65E02 4.65E02 4.65E02 For each case, three types of plutonium concentration loadings were used for the fuel pins within each case. Table 522 lists the fuel compositions for Case A, and Table 523 lists the fuel composition for Case B. Table 524 lists the borated water composition and Table 525 lists the Zircaloy2 composition used for the fuel clad and guide tubes. Figure 53 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 526. Table 524. Borated water composition for Phase IVB Benchmark (660ppm boron, density 0.7245 g/cm3). Nuclide Number Density (atoms/bcm) H 4.8414E02 160 2.4213E02 10B 4.7896E06 11B 1.9424E05 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Figure 53. 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 525. Zicaloy2 composition used for clad and guide tube material for Phase IVB Benchmark (density 5.8736 g/cm3). Nuclide (Atoms/bcm) 50Sn 2.9656E06 52Sn 5.7189E05 53Sn 6.4848E06 54Sn 1.6142E06 54Fe 7.8122E06 56Fe 1.2245E04 57Fe 2.8278E06 58Fe 3.7633E07 90Zr 1.9889E02 91Zr 4.3373E03 92Zr 6.6297E03 94Zr 6.7186E03 96Zr 1.0824E03 Table 526. 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 527, Table 528, Table 529 and Table 530 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 531, Table 532, Table 533 and Table 534 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 527. EOC 1 results for Case A of the Phase IVB Benchmark. Nuclide MCNPX Range of Values from (atoms/bcm) Other Depletion Codes (atoms/bcm) 234U 5.8483E07 2.1279E07 to 5.8216E07 235U 4.3753E05 4.3720E05 to 4.3925E05 6U 2.6406E06 2.4858E06 to 2.6190E06 238U 2.1134E02 2.1155E02to 2.1163E02 238Pu 4.1533E05 4.0713E05 to 4.1630E05 9Pu 8.0377E04 7.9700E04 to 8.1275E04 24Pu 4.7370E04 4.7109E04 to 4.8300E04 241Pu 2.1855E04 2.1520E04 to 2.2011E04 242Pu 1.6393E04 1.3389E04 to 1.3785E04 237Np 1.6176E06 1.0400E07 to 1.6560E06 241Am 9.4895E06 8.4140E06 to 8.8762E06 243Am 1.8069E05 1.5613E05 to 1.8260E05 2Cm 9.5381E07 8.7154E07 to 9.8523E07 243Cm 1.3915E08 1.1641E08 to 1.4560E08 244Cm 5.3807E06 2.6688E06 to 3.9610E06 2Cm 2.8581E07 1.2915E07 to 1.9877E07 95Mo 1.2829E05 1.1690E05 to 1.7651E05 99Tc 2.1958E05 2.1756E05 to 2.3060E05 101Ru 2.2331E05 2.1640E05 to 2.3054E05 103Rh 2.0106E06 1.9630E05 to 2.2642E05 109Ag 5.0294E06 4.6603E06 to 5.9460E06 133CS 2.4605E05 2.3463E05 to 2.4597E05 143Nd 1.5949E05 1.5350E05 to 1.6594E05 145Nd 1.1238E05 1.1183E05 to 1.1518E05 147Sm 8.0947E07 7.8960E07 to 8.2700E07 149Sm 4.2895E07 3.7767E07 to 4.3945E07 15Sm 4.7391E06 4.7807E06 to 5.1050E06 15Sm 1.3397E06 1.3420E06 to 1.5281E06 152Sm 2.7837E06 2.7790E06 to 3.1530E06 153Eu 2.0936E06 2.0380E06 to 2.2411E06 55Gd 8.6977E09 4.3940E09 to 1.6769E08 Table 528. EOC 2 results for Case A of the Phase IVB Benchmark. Nuclide MCNPX Range of Values from (atoms/bcm) Other Depletion Codes (atoms/bcm) 234U 8.0735E07 1.7413E07 to 8.4354E07 235U 3.4593E05 3.4500E05 to 3.4893E05 236U 4.6979E06 4.4215E06 to 4.6370E06 238U 2.0896E02 2.0910E02 to 2.0927E02 238Pu 3.9782E05 3.5013E05 to 3.9340E05 239Pu 6.3728E04 6.2980E04 to 6.5524E04 24Pu 4.4837E04 4.4499E04 to 4.6350E04 241Pu 2.3731E04 2.3210E04 to 2.3979E04 242Pu 1.7027E04 1.4147E04 to 1.4982E04 237Np 3.1341E06 3.6114E07to3.2139E06 241Am 1.5907E05 1.4680E05 to 1.5982E05 243Am 2.9191E05 2.7797E05 to 3.1620E05 2Cm 2.5536E06 2.3253E06 to 2.6204E06 2Cm 7.2443E08 5.8709E08 to 7.0660E08 244Cm 1.7168E05 9.2778E06 to 1.3050E05 24Cm 1.7035E06 8.1985E07 to 1.2348E06 95Mo 2.9454E05 2.8330E05 to 3.4297E05 99Tc 4.1886E05 4.1811E05 to 4.4400E05 101Ru 4.3865E05 4.2320E05 to 4.5186E05 103Rh 4.0075E05 3.7469E05 to 4.0914E05 109Ag 9.0959E06 8.0989E06 to 1.0640E05 133CS 4.6303E05 4.4403E05 to 4.6602E05 143Nd 3.0328E05 2.9970E05 to 3.1809E05 145Nd 2.1678E05 2.1322E05 to 2.2191E05 147Sm 2.1430E06 2.4130E06 to 2.4742E06 149Sm 4.0373E07 3.5930E07 to 4.1990E07 15Sm 1.0296E05 1.0197E05 to 1.1350E05 15Sm 1.5541E06 1.5306E06 to 1.8870E06 152Sm 4.9712E06 5.0468E06 to 5.9850E06 153Eu 4.9325E06 5.0480E06 to 5.4745E06 55Gd 1.6016E08 7.9700E09 to 2.8935E08 Table 529. EOC 3 results for Case A of the Phase IVB Benchmark. Nuclide MCNPX Range of Values from (atoms/bcm) Other Depletion Codes (atoms/bcm) 234U 9.4561E07 1.4169E07 to 1.0509E06 235U 2.6715E05 2.6596E05 to 2.7129E05 236U 6.2239E06 5.8461E06 to 6.1140E06 238U 2.0648E02 2.0657E02 to 2.0678E02 238Pu 4.0397E05 2.9577E05 to 3.9688E05 239Pu 5.0933E04 5.0368E04 to 5.3435E04 24Pu 4.1054E04 4.0836E04 to 4.2870E04 241Pu 2.3785E04 2.3107E04 to 2.4058E04 242Pu 1.8086E04 1.5231E04 to 1.6558E04 237Np 4.3966E06 6.9348E07 to 4.5652E06 241Am 1.8127E05 1.7960E05 to 2.0066E05 243Am 3.6401E05 3.6793E05 to 4.2000E05 2Cm 4.3963E06 3.5650E06 to 4.0436E06 3Cm 1.5834E07 1.2551E07 to 1.4820E07 244Cm 3.1800E05 1.8435E05 to 2.4950E05 24Cm 4.1506E06 2.1928E06 to 3.2335E06 95Mo 4.3988E05 4.3990E05 to 4.9929E05 99Tc 5.9641E05 6.0068E05 to 6.3850E05 101Ru 6.4487E05 6.1990E05 to 6.6364E05 103Rh 5.3043E05 5.1041E05 to 5.5357E05 109Ag 1.2395E05 1.0524E05 to 1.4480E05 133CS 6.4706E05 6.2829E05 to 6.6172E05 143Nd 4.2327E05 4.2830E05 to 4.5385E05 145Nd 3.1270E05 3.0386E05 to 3.2104E05 147Sm 3.1679E06 4.0280E06 to 4.1579E06 149Sm 3.2126E07 3.2302E07 to 3.8520E07 15Sm 1.5873E05 1.5410E05 to 1.7660E05 15Sm 1.5666E06 1.5340E06 to 2.0110E06 152Sm 6.3075E06 6.3335E06 to 7.9580E06 153Eu 7.5412E06 8.1310E06 to 8.6655E06 55Gd 1.9499E08 1.3920E08 to 4.2299E08 Table 530. Five year cooling time results for Case A of the Phase IVB Benchmark. Nuclide MCNPX Range of Values from (atoms/bcm) Other Depletion Codes (atoms/bcm) U 2.6607E06 1.4169E07 to 2.7220E06 235U 2.6785E05 2.6668E05 to 2.7206E05 236U 6.4390E06 5.8461E06 to 6.2030E06 238U 2.0648E02 2.0657E02 to 2.0678E02 238Pu 4.3111E05 2.8432E05 to 4.1989E05 239Pu 5.1081E04 5.0542E04 to 5.3613E04 24Pu 4.1585E04 4.1217E04 to 4.3260E04 241Pu 1.8680E04 1.8167E04to 1.8891E04 242Pu 1.8086E04 1.5231E04 to 1.6558E04 237Np 4.8116E06 7.0645E07 to 4.9848E06 241Am 6.8783E05 6.6320E05 to 7.0845E05 243Am 3.6406E05 3.6793E05 to 3.9880E05 2Cm 2.5594E09 1.6887E09 to 5.0942E09 2Cm 1.4022E07 1.1143E07 to 1.2852E07 244Cm 2.6279E05 1.5224E05 to 1.9671E05 24Cm 4.1491E06 2.1928E06 to 3.2321E06 95Mo 4.9799E05 4.9929E05 to 5.0765E05 99Tc 5.9856E05 6.0280E05 to 6.4090E05 101Ru 6.4487E05 6.3251E05 to 6.6366E05 103Rh 5.6463E05 5.1041E05 to 5.8208E05 109Ag 1.2416E05 1.0524E05 to 1.4490E05 133CS 6.5257E05 6.2829E05 to 6.6172E05 43Nd 4.3169E05 4.4116E05 to 4.5385E05 145Nd 3.1278E05 3.0386E05 to 3.2104E05 147Sm 7.8905E06 9.9452E06 to 1.0734E05 149Sm 3.6852E07 3.7494E07 to 4.2796E07 15Sm 1.5873E05 1.5410E05 to 1.6679E05 15Sm 1.5196E06 1.4779E06 to 1.8916E06 152Sm 6.3090E06 6.3335E06 to 7.9580E06 153Eu 7.5868E06 8.1650E06 to 8.6716E06 55Gd 3.6011E07 3.3400E07 to 1.0081E06 Table 531. EOC 1 results for Case B of the Phase IVB Benchmark. Nuclide MCNPX Range of Values from (atoms/bcm) Other Depletion Codes (atoms/bcm) 234U 2.1859E07 2.1683E07 to 2.2681E07 235U 4.2074E05 4.1730E05 to 4.2085E05 236U 3.1619E06 3.0287E06 to 3.1900E06 238U 2.2041E02 2.2000E02 to 2.2049E02 238Pu 1.1757E06 3.7913E07 to 7.1826E07 239Pu 5.6960E04 5.6280E04 to 5.7659E04 24Pu 1.5241E04 1.4600E04 to 1.5008E04 241Pu 6.0895E05 5.6781E05 to 6.3800E05 242Pu 7.1572E06 5.3572E06 to 6.1170E06 237Np 1.5688E06 1.3364E07 to 1.6488E06 241Am 1.6417E06 1.3107E06 to 1.4360E06 243Am 9.6882E07 6.2689E07 to 8.0100E07 2Cm 1.8012E07 1.5269E07 to 1.7950E07 2Cm 2.0951E09 1.6814E09 to 2.0300E09 244Cm 2.4339E07 9.3251E08 to 1.3100E07 24Cm 1.1257E08 3.4090E09 to 4.6024E09 95Mo 1.3153E05 1.2181E05 to 1.8204E05 99Tc 2.1838E05 2.1900E05 to 2.2950E05 101Ru 2.1979E05 2.1769E05 to 2.3058E05 103Rh 2.0336E05 1.8759E05 to 2.1765E05 109Ag 4.3321E06 3.6778E06 to 5.5840E06 133CS 2.4560E05 2.3439E05 to 2.4574E05 143Nd 1.5453E05 1.5100E05 to 1.6346E05 145Nd 1.0969E05 1.1117E05 to 1.1454E05 147Sm 7.6879E07 7.6500E07 to 7.8701E07 149Sm 2.3529E07 1.9930E07 to 2.2432E07 15Sm 4.7705E06 4.9051E06 to 5.2700E06 15Sm 8.6623E07 8.8308E07 to 1.0200E06 152Sm 2.8949E06 2.9600E06 to 3.3370E06 153Eu 2.0836E06 2.0570E06 to 2.2659E06 55Gd 4.5104E09 1.8000E09 to 7.6886E09 Table 532. EOC 2 results for Case B of the Phase IVB Benchmark. Nuclide MCNPX Range of Values from (atoms/bcm) Other Depletion Codes (atoms/bcm) 234U 1.8049E07 1.7098E07 to 1.9333E07 235U 2.9106E05 2.8760E05 to 2.9257E05 236U 5.5884E06 5.3234E06 to 5.5600E06 238U 2.1773E02 2.1767E02 to 2.1800E02 238Pu 2.6720E06 2.9817E07 to 2.0587E06 239Pu 3.7031E04 3.6608E04 to 3.8300E04 24Pu 1.8694E04 1.8070E04 to 1.8624E04 241Pu 9.7981E05 9.4093E05 to 1.0080E04 242Pu 2.2218E05 1.9550E05 to 2.1840E05 237Np 3.0365E06 4.5945E07 to 3.1897E06 241Am 4.1152E06 3.6692E06 to 3.9280E06 243Am 4.3621E06 3.6612E06 to 4.7300E06 2Cm 8.9769E07 7.7886E07 to 8.9591E07 2Cm 4.3621E06 1.6565E08 to 2.0900E08 244Cm 1.8948E06 1.0277E06 to 1.4100E06 24Cm 1.4495E07 6.0866E08 to 8.6913E08 95Mo 2.9951E05 2.9300E05 to 3.5096E05 99Tc 4.1181E05 4.1730E05 to 4.3780E05 101Ru 4.3022E05 4.2699E05 to 4.5057E05 103Rh 3.5805E05 3.3166E05 to 3.7099E05 19Ag 7.7718E06 6.5223E06 to 9.7410E06 133CS 4.5646E05 4.3917E05 to 4.6063E05 43Nd 2.8262E05 2.8500E05 to 3.0200E05 145Nd 2.0952E05 2.1048E05 to 2.1882E05 147Sm 1.9461E06 2.2520E06 to 2.2924E06 149Sm 2.0460E07 1.7336E07 to 1.9700E07 15Sm 1.0170E05 1.0500E05 to 1.1600E05 15Sm 8.4275E07 8.4937E07 to 1.0900E06 152Sm 4.7605E06 4.8514E06 to 5.9030E06 153Eu 4.8992E06 5.1650E06 to 5.5763E06 55Gd 9.7191E09 3.8100E09 to 1.1850E08 Table 533. EOC 3 results for Case A of the Phase IVB Benchmark. Nuclide MCNPX Range of Values from (atoms/bcm) Other Depletion Codes (atoms/bcm) 234U 1.1520E13 1.3174E07to 1.7954E07 235U 1.8445E05 1.8119E05 to 1.8791E05 236U 7.1794E06 6.8091E06 to 7.0700E06 238U 2.1481E02 2.1467E02 to 2.1500E02 238Pu 5.5067E06 2.2123E07 to 4.7927E06 239Pu 2.5259E04 2.4767E04 to 2.7200E04 24Pu 1.8124E04 1.7470E04 to 1.8437E04 241Pu 1.0690E04 1.0323E04 to 1.0850E04 242Pu 4.3597E05 3.9815E05 to 4.3080E05 237Np 4.1982E06 8.5608E07 to 4.4424E06 241Am 4.9089E06 4.9773E06 to 5.4746E06 243Am 1.0121E05 9.3994E06 to 1.1700E05 2Cm 1.9925E06 1.5821E06 to 1.8030E06 2Cm 6.2628E08 4.8232E08 to 5.9200E08 244Cm 6.7113E06 4.1608E06 to 5.4200E06 24Cm 6.6887E07 3.2603E07 to 4.7115E07 95Mo 4.4224E05 4.5000E05 to 5.0570E05 99Tc 5.7742E05 5.9218E05 to 6.2220E05 101Ru 6.2895E05 6.2300E05 to 6.5908E05 103Rh 4.3988E05 4.2315E05 to 4.7011E05 109Ag 1.0536E05 8.5551E06 to 1.2920E05 133CS 6.2697E05 6.1230E05 to 6.4442E05 143Nd 3.7203E05 3.8700E05 to 4.0746E05 145Nd 2.9770E05 2.9641E05 to 3.1166E05 147Sm 2.7210E06 3.5820E06 to 3.6688E06 149Sm 1.2934E07 1.4821E07 to 1.6900E07 15Sm 1.5432E05 1.5880E05 to 1.7900E05 15Sm 7.8954E07 8.0820E07 to 1.1000E06 152Sm 5.6451E06 5.7314E06 to 7.4850E06 153Eu 7.1295E06 7.9849E06 to 8.6528E06 55Gd 8.0878E09 6.3000E09 to 1.5641E08 Table 534. Five year cooling time results for Case B of the Phase IVB Benchmark. Nuclide MCNPX Range of Values from (atoms/bcm) Other Depletion Codes (atoms/bcm) U 4.4485E07 1.3174E07 to 4.1870E07 235U 1.8480E05 1.8119E05to 1.8830E05 236U 7.2745E06 6.9037E06 to 7.0634E06 238U 2.1481E02 2.1467E02 to 2.1493E02 238Pu 7.2430E06 2.1267E07 to 6.3483E06 239Pu 2.5476E04 2.4983E04 to 2.7284E04 24Pu 1.8232E04 1.7540E04 to 1.8510E04 241Pu 8.3970E05 8.1013E05 to 8.5710E05 242Pu 4.3597E05 3.9817E05 to 4.3090E05 237Np 4.3965E06 8.7398E07 to 4.6418E06 241Am 2.7709E05 2.6942E05 to 2.8303E05 243Am 1.0125E05 9.3994E06 to 1.0616E05 242Cm 9.8698E10 7.5522E10 to 1.4676E09 2Cm 5.5486E08 4.2821E08 to 5.1277E08 244Cm 5.5500E06 3.4362E06 to 3.8927E06 24Cm 6.6887E07 3.2603E07 to 4.7096E07 95Mo 5.0010E05 5.0570E05 to 5.1528E05 99Tc 5.7945E05 5.9428E05 to 6.2460E05 101Ru 6.2895E05 6.2654E05 to 6.5909E05 103Rh 4.7293E05 4.2315E05 to 4.9970E05 109Ag 1.0552E05 8.5551E06 to 1.2940E05 133CS 6.3248E05 6.1230E05 to 6.4442E05 143Nd 3.8029E05 3.9608E05 to 4.0746E05 45Nd 2.9778E05 2.9642E05 to 3.1176E05 147Sm 6.8077E06 8.7798E06 to 9.5399E06 149Sm 1.7692E07 2.0516E07 to 2.2260E07 15Sm 1.5432E05 1.5880E05 to 1.7344E05 15Sm 7.7139E07 7.7864E07 to 9.9180E07 152Sm 5.6451E06 5.7314E06 to 7.4860E06 153Eu 7.1797E06 7.9849E06 to 8.7110E06 55Gd 3.3990E07 3.2182E07 to 9.2307E07 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 codecode 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 39 and 310. As a result of the heritage of the data release, MCNPX depletion expects Am242m 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 63group 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 Am243, which decays to Cm243. The nuclide Am242 decays quickly to Cm242 which then decays to Pu238. 242Am has a higher capturetofission 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 Cm243 Cm244 and Cm245. This hypothesis seems to be validated by the fact that Cm243, Cm 244 and Cm245 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 timedependent 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 easytouse depletion calculations in a single, well established, radiation transport Monte Carlo code. The capability links the wellestablished MCNPX steadystate reaction rate calculator to the CINDER90 temporal nuclide inventory calculator. MCNPX is not limited by typical approximations utilized in deterministic methodologies. The MCNPX steadystate 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 CELL2 "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 fueltomoderator 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, uraniumdioxide 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 codetocode 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 capturetofission 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 2.433 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 codetocode 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 adjustedfueltomoderator 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 1020% 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 nonrare events leading to an inadequate sampling of nonrare 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 Carlolinked 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 6.87812e2 1 u=2 imp:n=l vol=192.287 $fuel 3 2 4.5854e2 12 u=2 imp:n=l vol=66.43 $ clad 4 3 7.1594e2 2 u=2 imp:n=l vol=370.82 $ water 6 4 6.87812e2 1 u=3 imp:n=l vol=192.287 $fuel 7 5 4.5854e2 12 u=3 imp:n=l vol=66.43 $ clad 8 6 7.1594e2 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= le32 MATVOL= 384.57 384.57 MATMOD= 2 1 1 4 1 94238 le6 2 2 1 2 94238 le6 94241 le6 4 1 94238 le6 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: Qvalue 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 63group flux from MCNPX to be matched to a 63group 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 le32 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 le6 atoms/barncm. At time step 2, material 1 will have the atom densities of 2 isotopes changed: material 1 will have isotope 94238 changed to le6 atoms/barn cm and isotope 94241 will be changed to le6 atoms/barncm, material 4 will have the atom density of 1 isotope changed: 94238 will be set to le6 atoms/barncm. LIST OF REFERENCES 1. U.S. Department of Energy Office of Civilian Radioactive Waste Management, Oklo: Natural Nuclear Reactors, November 2004, 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: 10114885011425, 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, AddisonWesley Publishing Company, Inc., Melano Park, CA(1983). 8. K. O. OTT, C. W. TERREL, R. C. BORG and P. J. MAUDLIN, "Integrated FuelCycle Models for Fast Breeder Reactors," Annuls ofNuclear Engineering, 8, pg. 371392 (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, "FertileFree Fast LeadCooled Incinerators for Efficient Actinide Buruning," Journal of Nuclear Technology, 147, pg. 368 386 (2004). 11. G S. CHANG "Neutronic and Burnup Characteristics of an ActinideReduced Plutonium Fuel with Tungsten," Journal ofNuclear Technology, 122, pg. 4351 (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. 4351 (1998). 13. X. JIANG and Z. XIE, "Transportbumup Code Systems and Their Applications for IAEA ADS Benchmark," Annuls ofNuclear Energy, 31, pg. 213225 (2004). 14. N. MESSAOUDI and J. TOMMASI, "Fast Burner Reactor Devoted to Minor Actinide Incineration," Journal ofNuclear Technology, 137, pg. 8996 (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: LAUR041572, 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: GNEP167312 (2007). 19. M. D. LOWENTHAL, "Transmutation in the Nuclear Fuel Cycle: Approaches and Impacts," Journal ofNuclear Technology, 138, pg. 284299 (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. 449456 (2002). 22. M. SALVATORES, "Nuclear Fuel Cycle Strategies Including Partitioning and Transmutation," Nuclear Engineering and Design, 235, pg. 805816 (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. 369371 (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. 287290 (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/TM252, 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. 612620 (2005). 29. K. SUYAMA, H. MOCKIZUKI, T. KIYOSUMI, "Revised Bumup Code System SWAT: Description and Validation Using Postirradiation Examination Data," Nuclear Technology, 138, pp. 97108. 30. M. EDENIUS, K. EKBERG B. H. FORSSEN, and D. KNOTT, "Casmo4, A Fuel Assembly Burnup Program, User's Manual," Studsvik Report: SOA95/15, 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 NextGeneration Neutronics Design System," Nuclear Science and Engineering, 155, pp. 276289 (2007). 32. D. KNOTT, E. WEHLAGE, "Description of the LANCER02 Lattice Physics Code for SingleAssembly and Multibundle Analysis," Nuclear Science and Engineering, 155, pp. 331354 (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. 222233 (1998). 34. A. NUNEZCARRERA, J. L. FRANCOIS, G ESPINOSAPAREDES, "Comparison between HELIOS CriticalDepletion Calculations and a PWR Thorium Cell Burnup Benchmark," Annuls ofNuclear Energy, 31, pg. 713722 (2004). 35. M. L. FENSIN and S. ANGHAIE, "Current Methods of Depletion Analysis," Los Alamos National Laboratory Report LAUR063933, American Nuclear Society's 2006 Winter Meeting, Albuquerque, NM (2006). 36. G S. CHANG J. M. RYSKAMP, "Depletion Analysis of MixedOxide Fuel Pins in Light Water Reactors and the Advanced Test Reactor", Nuclear Technology, 129, pp. 326337 (2000). 37. W. HAECK, "An Optimum Approach to Monte Carlo BumUp," PhD thesis, Ghent University, Belgium (2007). 38. E. J. PARMA, "BURNCAL: A Nuclear Reactor Burnup Code Using MCNP Tallies," Sandia National Laboratory Report: SAND20023868, 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 "MCWOLinking 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 MonteCarlo Continuous Energy Burnup (MCB) CodeVersion 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. 127132 (2003). 44. R. L. MOORE, B. G SCHNITZLER, C. A. WEMPLE, R. S. BABCOCK, D. E. WESSOL, "MOCUP: MCNPORIGEN2 Coupled Utility Program", Idaho National Engineering Laboratory Report: INEL95/0523, 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 LAUR994999, 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. 1925 (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: ANLTDCP97492, 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, "HighFidelity 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 LACP050369, 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 LAUR933080, 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 IB 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/TM12667, Oak Ridge, TN, USA (1995). 54. G J. O'CONNOR and P. H. LIEM "Burnup Credit Criticality Benchmark PHASE IVB: 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. 407414 (1993). 56. M. YAMAMOTO, "Coarse TimeStep Integration Method for Burnup Calculation of LWR Lattice Containing GadoliniumPoisoned Rods," Journal ofNuclear Science and Technology, 22, pg.115 (1985). 57. G ILAS, B. D. MURPHY, and I. C. GAULD, "Overview of OrigenARP 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. 335352 (1983). 59. G F. THOMAS and D. H. BARBER, "Stiffness in Radioactive Decay Chains," Annuls of Nuclear Energy, 21, pg 309320 (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: ORNL5621, 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. 197210 (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. 6577 (1999). 70. H. HERMAN, "ENDF6 FORMATS MANUAL," Brookhaven National Laboratory Report: BNLNCS4494505Rev., Long Island, NY (2005). 71. M. L. FENSIN and D. I. POSTON, "Optimum Reflector Configurations for Minimizing Fission Power Peaking in a LithiumCooled, LiquidMetal Reactor with Sliding Reflectors," Los Alamos National Laboratory Report LAUR7154, 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 LAUR05 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, "MCNPA General Monte Carlo NParticle Transport Code," Los Alamos National Laboratory Report: LA13709M, 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 LA13514T, Los Alamos, NM (1998). 76. S. M. BOWMAN, M. D. DEHART, and L. M. PETRIE, "Integrated Keno Monte Carlo Transport for 3Depletion 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, Tokaimura, 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. 154157 (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: LAUR05 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/BVII.0: Next Generation Evaluated Nuclear Data Library for Nuclear Science and Technology", Nuclear Data Sheets, 107, pg. 29313060 (2006). 81. T. R. ENGLAND, B. F. RIDER, "ENDF349 evaluation and compilation of fission product yields," Los Alamos National Laboratory Report: LAUR943106, 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 LAUR057895, 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 TwoTiered System Comprising Light Water Reactors and AcceleratorDriven Systems,"Los Alamos National Laboratory Report: LA14052T, 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: LAUR067991, 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 LAUR06 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: LAUR076632, 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, 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. 636648 (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 923933 (2001). 91. C. M. KANG R. O. MOSTELLER, and S. LEVY, "Incorporation of a PredictorCorrector Depletion Capability into the CELL2 Code," Transaction of the American Nuclear Society, 45, pp. 729731 (1983). 92. M. L. FENSIN, J. S. HENDRICKS, H. R. TRELLUE, S. ANGHAIE "Incorporation of a PredictorCorrector Methodology and 1GroupReaction Rate Reporting Scheme for the MCNPX Depletion Capability," Los Alamos National Laboratory Report LAUR063925, American Nuclear Society's 2006 Winter Meeting, Albuquerque, NM (2006). 93. Z. XU, J. RHODES III, K. SMITH, N. GHEORGHIU, "MCNPX5/ORIGEN2.2/MCODE 2.2 versus CASMO5 Depletion for a Heavy GdPoisoned 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: LAUR063248, 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: LAUR074137, 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: LAUR080305, 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 Cooperation and Development: Nuclear Energy Agency, "NEA Expert Group on Bumup Credit Criticality Safety," September 2003, 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. 164179 (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 ZeroDimension 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 TimeDependent 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 CELL2 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 31 MOX fuel composition......................................................................................................89 32 Borated water composition................................................................................................90 33 Emitted and recoverable energy per fission of U235.......................................................97 34 Prompt fission energy releas e data available in MCNPX..................................................98 35 Available actinide fission yi eld distributions in CINDER90...........................................102 41 OMIT keyword variable tasks..........................................................................................117 42 Options invoked for different B2 values.........................................................................118 43 Options invoked for different B2 values.........................................................................119 44 Available options for the B3 value of the BOPT keyword..............................................119 45 MATMOD keyword variable values...............................................................................121 51 Examined benchmark cases.............................................................................................129 52 Model description for Phase IB Benchmark*..................................................................130 53 Fuel composition for Phase IB Benchmark (density =10.045 g/cm3).............................131 54 Clad composition for Phase IB Benchmark (6.44 g/ cm3)...............................................132 55 Borated water composition for Phase IB Benchmark (0.7569 g/ cm3)............................132 56 Operating history data for Phase IB Benchmark.............................................................132 57 Operating power (MW) for each case for Phase IB Benchmark.....................................133 58 Initial Phase IB Benchmark results a nd comparison of Case A (27.35 MWD/MTU)....133 59 Initial Phase IB Benchmark results a nd comparison of Case B (37.12 MWD/MTU).....134 510 Initial Phase IB Benchmark results a nd comparison of Case C (44.34 MWD/MTU).....135 511 MCNPXc Phase IB Benchmark results and comparison of Case A (27.35 MWD/MTU)....................................................................................................................137 512 MCNPXc Phase IB Benchmark results and comparison of Case B (37.12 MWD/MTU)....................................................................................................................138 PAGE 9 9 513 MCNPXc Phase IB Benchmark results and comparison of Case C (44.34 MWD/MTU)....................................................................................................................139 514 MONTEBURNS Phase IB Benchmark results and comparison of Case A (27.35 MWD/MTU)....................................................................................................................144 515 MONTEBURNS Phase IB Benchmark results and comparison of Case A (37.38 MWD/MTU)....................................................................................................................144 516 Fuel assembly design data for H. B. Robinson Benchmark............................................147 517 Operating data for H. B. Robinson fuel assembly...........................................................148 519 The H. B. Robinson Benchmark results using MCNPX compared with SCALE...........149 520 The H. B. Robinson Benchmark results using MCNPXc compared with SCALE.........150 521 Fuel assembly design data for OECD/NEA Burnup Credit Phase IVB Benchmark.......152 522 Initial MOX fuel composition for Case A for Phase IVB Benchmark............................153 523 Initial MOX fuel composition for Case B for Phase IVB Benchmark............................154 524 Borated water composition for Phase IVB Benchmark (660ppm boron, density 0.7245 g/cm3)...................................................................................................................154 525 Zicaloy2 composition used for clad and guide tube material for Phase IVB Benchmark (density 5.8736 g/cm3)..................................................................................156 526 Operating data for Phase IVB Benchmark.......................................................................156 527 EOC 1 results for Case A of the Phase IVB Benchmark.................................................157 528 EOC 2 results for Case A of the Phase IVB Benchmark.................................................158 529 EOC 3 results for Case A of the Phase IVB Benchmark.................................................159 530 Five year cooling time results fo r Case A of the Phase IVB Benchmark........................160 531 EOC 1 results for Case B of the Phase IVB Benchmark.................................................161 532 EOC 2 results for Case B of the Phase IVB Benchmark.................................................162 533 EOC 3 results for Case A of the Phase IVB Benchmark.................................................163 534 Five year cooling time results fo r Case B of the Phase IVB Benchmark........................164 PAGE 10 10 LIST OF FIGURES Figure page 21 Linked Depletion Process..................................................................................................46 22 Flux depression in the ne ighborhood of a resonance at E0................................................51 23 Flux depression in the ne ighborhood of a resonance at E0................................................53 24 Flow diagram for the Monte Ca rlo particle transport process...........................................56 31 MonteCarlolinked de pletion process model....................................................................72 32 Uranium235 fission pr oduct yield distribution.7..............................................................75 33 Fission Yield Distributions for Va rious Plutonium and Uranium Isotopes.81...................76 34 Original fission product arra y containing 167 fission products.........................................77 35 Resulting timedependent neutron multip lication from implementing fission products with and without cross section models in MCNPX as compared to the benchmarked MONTEBURNS method...................................................................................................78 36 Percent difference of certain fission products from MONTEBURNS using MCNPX Tier1 fission products with and without cross section models.........................................78 37 Percent difference of certain fission products from MONTEBURNS using MCNPX Tier2 and Tier3 fission products with crosssection models...........................................79 38 Enhanced fission product ar ray containing 220 isotopes...................................................79 39 Nuclides created from a decay or nuclear reaction process and ge nerated for particle transport by the isotope generator algorithm.3...................................................................80 310 Examined infinitely re flected pin cell geometry................................................................89 311 Percent difference of certain referen ce actinides between the OLD MCNPX method and the MONTEBURNS method......................................................................................91 312 Percent difference of certain referen ce actinides between the NEW MCNPX method and the MONTEBURNS method......................................................................................92 313 Percent difference of certain refere nce fission products between the NEW MCNPX method and the MONTEBURNS method.........................................................................92 314 Timedependent eigenvalue calculations...........................................................................94 315 Fission yield selection process diagram...........................................................................104 PAGE 11 11 316 MONTEBURNS cross sec tion averaging technique.......................................................106 317 CELL2 PredictorCorrector method...............................................................................107 318 Time dependant neutron multiplication fo r the predictor corrector method tests...........109 319 Burn steps implemented to achieve 2191 day burn.........................................................110 320 Time dependent neutron multip lication utilizing 5 burn steps........................................111 321 Time dependent neutron multip lication utilizing 10 burn steps......................................111 322 Percent Difference in keff When Compared with the NoPC 30Step Case......................112 41 Burn card input structure.................................................................................................115 42 BURN card referencing burn ma terials m1, m3 and m4...................................................117 43 Example collision rate outputs.........................................................................................123 44 Example burnup summary table......................................................................................124 45 Example individual material burnup data........................................................................124 46 Example cropped actinide output at two separate timesteps and fission product output for one time step...................................................................................................125 47 Cropped example of print table 220.................................................................................126 51 Infinitely reflected pin cell geometry...............................................................................130 52 Infinitely reflected pin cell geometry...............................................................................146 53 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 Carlolinked depletion methods have gain ed recent interest due to the ability to more accurately model complex 3dimesional 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 highfidelity completely selfcon tained MonteCarlolinked 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 selfsustaining 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 selfsustaining 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 selfsustaining 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 selfsustaining chain reaction.1 As the uranium rich deposits burned fissile uranium235, 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 selfsustaining ch ain reaction. The system would then cease to maintain selfsustaining 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 selfsustaini 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 U235 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 selfsustaining 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 selfsustaining 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 U235 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 subcooled and contains homogeneously disperse d boron poison in orde r to maintain criticality. The system utilizes an indirect dualcycle 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 U235 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 selfsustaining 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 wellinformed 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 moreinformed 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 longterm 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 nextgeneration reactors with the goal of (1) reducing volume of highleve l waste; (2) reducing longlived 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, carbonfree 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 nonprolif 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 longlived radioactive wa ste, and recycling of useful fissile isotopes.9, 19 The following include the major obj ectives of the GNEP program: 1. New proliferationresistant 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 ationresistance 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 longterm 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 halflives 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 shortlived fissi on products. The shortli 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 Pu239 as compared to the uranium isotopes, and faster reactor re sponse, from the reduced delayed neutron fraction of Pu239, 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 multirecycling 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 fissiontocapture 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 longlived 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 2dimensional 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 nonpeaceful 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 proliferationresistant 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 nondestructiveanalysis (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 timedependant 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.2534 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 MonteCarlolinked depletion c odes exist including ALEPH, BURNCALC, MCODE, MCREBUS, MCWO, MCB, MCMGB URN, MOCUP, MONTEBURNS, MVP and RACER.3749 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 multigroup 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 selfcontained, eliminate excess toplevel 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 highfideli 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 easytouse 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 statereaction calculator to the CINDER90 number density calculator.50, 51 MCNPX will be used to calculate st eadystate reaction rates and normalization parameters while CINDER90 will be used to calculate the timedependent 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 medependent 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 easytoimplement 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 IB;52 12. H. B. Robinson Infinitely Reflected Pr essurized Water Reactor Fuel Assembly Calculation;53 13. OECD/NEA Burnup Credit Calculational Criticality Benchmark Phase IVB.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 U235 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 U235 for neutron capture, the fission products, combined with the depletion of U235, reduce the propensity of the system to fission and maintain a steadystate ch ain reaction. In order to maintain a self sustaining steadystate 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 steadystate 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 timedependent 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 21 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 ) ( ) ( ) ( (21) 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 timedependent change of a specified nuclide also involves knowledge of the timedependent change of each contributing nuc lide. Therefore the solution of each nuclides timedependent 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 21: r r m m mdE t E r E ) ( ) (, (22a) PAGE 34 34 k mr r k r km k m k km m kdE t E r E E Y L ) ( ) ( ) (, (22b) 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 timedependent fl ux. Unfortunately, the timedependent flux is also dependent upon the timedep endent nuclide inventory. To make equation 21 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 timedependent 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 21. 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 Xe135 and Sm149 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 21 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 timestep, assuming a constant transmutation rate for a given timestep is valid.56 ZeroDimension Depletion Calculation The depletion equation requires knowledge of re action rates, decay coefficients and isotope concentrations at a specific timestep in order to solve for the isotope concentrations at the subsequent timesteps. 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 21 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 23: 51, 58 i i f i f iR X P1810 242 6 (23) 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 24: V fdV Q r Power ) ( (24) ) ( 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 21 may be recast into the following form k k mk mt r n t r B t r n dt d ) ( ) ( ) ( (25) 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) ( ) ( ) (, (26) 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 (27) where N is a column vector of nuclide densities and B is the matrix of elements calculated of equation 26. The function ) exp( Bt is a matrix exponential function, and is therefore calculated from the following expansion: 0! ) ( ) exp(m mm Bt Bt (28) 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 halflives of each isotope tracked during the depletion process.59 Certain isotopes possess halflives 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 ) ( (29) PAGE 39 39 where n ic is generated by use of the fo llowing recursive relationship ) 0 (0 i ix c (210a) N j n j ij n ic a n t c1 11 (210b) In this solution method ija is an element in the matrix that is the firstorder 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 29 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 halflives. 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 GaussSi 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 reactorfuel 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, (i1) 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 (212) 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 ) ( (213) 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 (214) 1 0 m Ym (215) 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 (216) The test for significance in CINDER90 i nvolves calculating the passby quantity, ) (t Pn. The passby of a nuclide is the timeintegrated tr ansmutation of that nuclide for a specified time interval: dt t t Pn t n n 0) ( ) ( (217) 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. 211). If the code is used by itself to calculate nuclide inventory, the user would be relying heavily upon the inherent multigroup cross sections, utilized to determine collapsed energy integrated reaction rates, in order to achieve a solution. Since CINDER90 is a zerodimensional 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 nonperiodic 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 ergyintegrated 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 U235 atoms. As the core bu rns over time many neutrons are parasitically absorbed by U238 leading to production of U239. U239 quickly decays by beta emission to Np239 and then to Pu239 which is hi ghly fissile. At the en d of the reactor core life, an immense amount of Pu239 has been created, absorbed neutrons, fissioned, and contributed to system power. The average thermal absorption cross section for Pu239 is 1011.3 b which is much higher as compared to U235 at 680.8 b.7 The average thermal fission cross section for Pu239 is 742.5 b which is also much higher as compared to U235 at 582.2b.7 PAGE 46 46 capture cross section is defined as all absorption reactions not leading to fission, and therefore a thermal capturetofission ratio may be define d for a specified isotope by equation 218: f f a (218) Changes in the capturefissionratio 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 21. Linked Depletion Process. SteadyState Reaction Rate Calculation TimeDependant 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 21 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 21. The r eaction rates utilized in the capture terms of equation 21, system eige nvalue, and flux normalization terms are first calculated utilizing a steadystate reaction rate ca lculator. Those capture reactions are then assumed constant over a specified time step and implemented in a timedependent number density calculation. The timedependant number de nsity calculation then calculates a new set of number densities to be utilized in the subse quent steadystate 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 219: 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 (219) 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 219 into a set of coupled differentia l equations, where the coupling of energy groups results from the outgroup 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 ) ( ) ( ) ( ) ( ) ( (220) The angular dependence of equa tion 220 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 infinitedilute, 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 selfshiel ding. Figure 22 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 selfshielding 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 selfshielding in order to properly model the true system physics. Figure 22. Flux depression in th e neighborhood of a resonance at E0. 1/E (E) E0 E PAGE 52 52 Casting equation 219 into the multigroup form requires that multigroup coefficients be generated. The coefficients are generated base d on the premise that the multigroup 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 221: 1 1) ( ) ( ) ( ) (g g g gE E E E gdE E r dE E r E r r (221) The concern with this approach is that the flux, th e variable that is to be to calculated in equation 220, is used to calculate the cross section which is then utilized to calculate the flux in equation 220 making the process nonlinear. To make the process linear, an approximation is made in which first a homogenous, spatiall yinfinite, 1D 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 selfshiel ded cross sections for a further multigroup 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 23 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 23. 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 uraniumoxi de fuel, with zircaloy clad, surrounded by water, and possibly controlled by borncarbide or hafnium.25, 26, 30, 32 However, those libraries are not applicable for examining exotic isotope combinations such as uraniumnitride fueled, rhenium cladded and niobiumcarbide 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, nonstoc 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 24. 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 onedimensional 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 222: b ab x a P dx x f } { ) ( (222) 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 223: 1 ) ( dx x f (223) PAGE 56 56 From this equation we can further derive the cu mulative probability dist ribution function (CDF), F(x) in equation 224, which will be utilized to ultimately determin e the actual length of particle travel: x x P x F dx x fx ) ( ) ( (224) 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 24. 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) ( (225) and the associated CDF is xe x F 1 ) ( (226) 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 (227) and if we choose to be uniformly distributed betw een 0 and 1 we can further write ) ln( x (228) 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( (229) 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, ,) ( ) ( ) ( (230) 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) ( ) ( ) ( (231) 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, ,) ( ) ( ) ( (232) 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 ,) ( ) ( ) ( (233) 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 ~ ~ (234) For the Monte Carlo computation the fl ux may therefore be approximated by: cV c ~ ~ (235) 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 (236) 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 (237) 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 (239) 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 n1 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 238: 12 2 n x x Si (238) 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 (239) PAGE 60 60 The sample variance is referred to having n1 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 n1 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 238 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 (240) 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 ) ( ) ( (241) 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~ (242) 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 (243) 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) ( ) ( ~ (244) ) ( ~ ,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, , (245) i n i f n i n n l effE E l N k) ( ) ( ~ 1, (246) 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, , (247) 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 industrybased depletio n codes is to support the operation and fuel cycle design of commercial pow er reactors. Industrybased 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 deterministiclinke d depletion methodologies.35 Although various companies, such as Westinghouse General Electric, Areva, and Studsvick, use slightly different techniques to ap proximate timedependent reactor behavior, the overall methodology and processes are quite simila r. First, a finefidelity deterministic PAGE 65 65 calculation (using PHOENIX, TGBLA, CASMO3, et c.) examining many axial and radial slices of the reactor are completed to generate coarser 1to 4groupaveraged interaction rate parameters for implementation in a coarser calcu lation of the fullcore 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 selfshielding 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 multigroup code developed for and LWR with certain PAGE 66 66 group structure, fission products, and reactions followed will not be accurate in determining individual gammaline 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. Academiabased 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 higherfidelity 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 finerene 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 selfshielding considerations. To eliminate dependence from this burde n, burnup codes coupling the Monte Carlo reaction rate calculation have received more attention.3641,4648,75 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 boron10. Furt her acceleration techniques for Monte Carlo may include using multigroup Monte Carlo such as in Keno Va.76 Multigroup 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 multigroup 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 multipurpose 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 timedependent 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 ltigroup or stochastic calculation (using MCNP, PHOENIX, TGBLA, CASMO3, 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 fullcore 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 microsection 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 multipurpose 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) steadystate 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 31. MCNPX runs a steadystate calculation to determine the system eigenva lue, 63group fluxes collision rates, and Q values. CINDER90 then takes those MCNPXgenerated 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, timedependant 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 31. MonteCarlolinke 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 halflife 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 steadystate 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 63group 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 63group spatially depende nt flux is calculated within MCNPX and sent to CINDER90 to be match with a 63group 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 atgranddaughter/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 twotothree 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 32 displa ys the fission product distribut ion for U235 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 33 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 32. Uranium235 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 33. 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 34, with the following fission product tiers:79 1. Tier 1. (default) Zr93, Mo95, Tc99m, Ru101, Xe131, Xe134, Cs133, Cs137, Ba138, Pr141, Nd143, Nd145.9 2. Tier 2. Isotopes contained in the fission product array that are included in the released crosssection library f ile (XSDIR) for MCNPX. 3. Tier 3. All isotopes containe d in the fission product array. Figure 34. 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 35 displays the difference in neutron multiplication versus time and Figure 36 and Figure 37 displa y the difference in endoflif 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 38.84 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 35. Resulting timedependent 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%Zr93 Mo95 Tc99 Ru101 Xe131 Xe134 Cs133 Cs137 Ba138 Pr141 Nd143 Nd145NuclidesPercent Difference Tier 1 without Models Tier 1 with Models Figure 36. Percent difference of certain fission products fr om MONTEBURNS using MCNPX Tier1 fission products with a nd without cross section models. PAGE 79 79 15.00% 10.00% 5.00% 0.00% 5.00% 10.00%Zr93 Mo95 Tc99 Ru101 Xe131 Xe134 Cs133 Cs137 Ba138 Pr141 Nd143 Nd145NuclidesPercent Difference Tier 2 Tier 3 Figure 37. Percent differen ce of certain fission products fr om MONTEBURNS using MCNPX Tier2 and Tier3 fission produc ts with crosssection models. Figure 38. 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 1E36. Figure 39. 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 steadystate 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 39 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 multiinteraction 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 TimeDependent 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 timedepe 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 postexecuting 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 (31) 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, (32) 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 (33) 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 (34) 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 1fn 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 (35a) i A i aM N w % (35b) 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, , (35c) 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, (35d) k k k i i i k k kM w M w M w % % % (35e) k a k a k A k k A k kAF N M w N M w ,* % % (36f) 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 (37) 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 (38) 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 (39) For example: Ag110m = 47510 47510 = 47110 + 1st metastable (310) 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 21, 22a and 22b, 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 steadystate 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 63group cross section set for all reactions available in the libr ary file. Therefore, CINDER90 theoretically only requires a 63group flux to match to its 63group 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 tradeoff 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 63group flux in MCNPX to a 63group 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 63group flux calcula ted within the steadystate Monte Carlo to a 63group 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 63group ma tching approach mentioned previously. The concern of the 63group matching approach was that the validity of this te chnique resided in the applicability of the 63group crosssection set to various designs. The 63group 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 63group 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 63group 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 multigroup 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 timedependent 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 310. Figure 310. Examined infinitely reflected pin cell geometry. Table 31. MOX fuel composition. ZA value Atom Density [atoms/cm3] 8016 4.5854e2 92235 1.4456e4 92238 1.9939e2 94238 1.1467e4 94239 1.0285e3 94240 7.9657e4 94241 3.3997e4 94242 5.6388e4 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 31. The entire fuel pin was encased within a borated water rectangl e 365 cm tall with a 1.313cm length and 1.313cm width. The borated water composition is listed in Table 32. Table 32. Borated water composition. ZA value Atom Density [atoms/cm3] 1001.60c 4.7716e2 8016.60c 2.3858e2 5010.60c 3.6346e6 5011.60c 1.6226e5 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 steadystate 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 steadystate 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%U235 U236 U237 U238 U239 Np237 Np239 Pu237 Pu238 Pu239 Pu240 Pu241NuclidesPercent Difference Figure 311. 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 63group Monte Carlo flux to be matched to a generalized 63group CINER90 crosssection set. The integral continuous energy reaction rates were calculated separately for each burn material and tallied at the site of collision. Figure 312 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 Np237 drops PAGE 92 92 from 70.24% to 4.85%, and the percent differe nce in Pu237 production drops from 95.69% to 1.02%. Due to accentuation of U237, Np237, and Pu237 differences in Figure 311, Figure 312 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, 12% improvements in actinide predic tions are also witne ssed in U236, U239 and Pu238. 6.00% 5.00% 4.00% 3.00% 2.00% 1.00% 0.00% 1.00% 2.00%U235 U236 U237 U238 U239 Np237 Np239 Pu237 Pu238 Pu239 Pu240 Pu241NuclidesPercent Difference Figure 312. 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%Zr93 Mo95 Tc99 Ru101 Xe131 Xe134 Cs133 Cs137 Ba138 Pr141 Nd143 Nd145NuclidesPercent Difference Figure 313. 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 313 displays the difference in EOL fission product prediction when comparing the new MCNPX method to MO NTEBURNS. Unfortunately, timedependent fission product buildup is a major contributor to negative reactivity trajec tories, and therefore MCNPX and MONTEBURNS only pr edicted slightly different timedependent reactivity. Figure 314 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 timedependent cr oss section averaging technique. Figure 314 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 314. Timedependent 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 311: value MCNPX totalQ P Multiplier Flux Multiplier Flux (311) = 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 311. 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 33. Emitted and recovera ble energy per fission of U235. Form Emitted Energy (MeV) Recoverable Energy (MeV) Fission Fragments 168 168 Fission Product Decay Gammarays 8 8 Betarays 7 7 Neutrinos 12 Prompt Gamma Rays 7 7 Fission Neutrons (Kinetic Energy) 5 5 Capture Gamma Rays 312 Total 207 198207 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 33 lists the emitted and recoverable energy per fission from each micro fission event for U235.7 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 34. The sum of EFR, ENP and EGP for each actinide is the prom pt fission release energy for that actinide. Table 34. 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 Qvalue 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 U235 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 312, 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 (312) 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 (313) 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 313 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 313 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 32 and Figure 33. 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 35. 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, highenergy fission yield for each fissile isotope contained in the CINDER90 data file.61 Table 35 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 315. 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 energydependent 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 315. 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 21 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 timedependent. 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 21 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 314 illustrates how power is proportional to the macroscopic fission cross section multiplied by the flux: fPower ~ (314) 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 timedependent 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 nonlinear equations;46, 74, 8991 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 predictorcorrector method. 46, 74, 89, 91, 92 First, an assumption is made about the timeintegrated 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 timedependent parameters. Predictorcorrector 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 Cell2 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 Cell2 method follows a more traditional predictorcorrector 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 316. MONTEBURNS cross section averaging technique. PAGE 107 107 The MONTEBURNS cross section averaging procedure is displayed in Figure 316.74 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. CELL2 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 317. CELL2 PredictorCorrector method. PAGE 108 108 The CELL2 predictorcorrector method is displayed in Figure 317.91 The CELL2 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 nopredictorc orrector method (NoPC); 2. MONTEBURNS predictorcorrector method (PC1); 3. CELL2 predictorcorrector 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 timesteps approached infinity the calculation would approach true behavior. Neut ron multiplication versus time for the examined cases is displayed in Figure 318. 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 318. 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 320 and Figure 321 display the timedepe 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 timestep 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 322 displays the percent difference in neutron multiplication of each case when compared with the NoPC 30step 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 319. Burn steps impl emented to achieve 2191 day burn 302015105 0.00E+000.00E+000.00E+000.00E+000.00E+00 2.48E022.48E022.48E022.48E022.48E02 4.09E014.09E014.09E01 1.94E+001.94E+00 7.93E017.93E011.18E+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 320. 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 321. 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 322. Percent Difference in keff When Compared with the NoPC 30Step 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 CELL2 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 317 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 CELL2 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 CELL2 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 timest 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 selfcontained 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 selfcontaine 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.AF.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 41 displays the i nput structure for the BURN card. Each input variable is then explained below Figure 41. 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 41. 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 100day burn, followed by a 15day 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 42 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 nonactinide 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 4.5854e2 92235.60c 1.4456e4 92238.60c 1.9939e2 94238.60c 1.1467e4 94239.60c 1.0285e3 94240.60c 7.9657e4 94241.60c 3.3997e4 94242.60c 5.6388e4 m2 2004 1.0 m3 40000.60c 1.0 m4 1001.60c 4.7716e2 8016.60c 2.3858e2 5010.60c 3.6346e6 5011.60c 1.6226e5 mt4 lwtr.01t Figure 42. 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 63group flux into CINDER90, matching the 63group flux calculated from MCNPX to a 63group 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 41 contains the availabl e options for this keyword. Table 41. 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 1.0E10. The A2 value sets the decay chain convergence criteria (mentioned in the Benchm arks section). The default for this value is now 1e10 and was 1e4 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 42. 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 42 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 fissionproduct 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 43 gives a descripti on of the available actinides in each fission product tier. Table 43. Options invoked for different B2 values. Tier Fission Product Content 1 Zr93, Mo95, Tc99m, Ru101, Xe131, Xe134, Cs133, Cs137, Ba138, Pr141, Nd143, Nd145 (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 38. The B3 value of the BOPT keyword invokes the models opti on. Crosssection models are a heritage of the highenergy physics capabilities of MCNPX. The models are benchmarked for a variety of problems for neutron energies >150 Me V. Table 44 describes the available options for the B3 value of the BOPT keyword. Table 44. 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 (41) 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 41 was altered to n n nw l NV 1 (42) 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 45. 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 timedependent manual nuclide concentration changes. The details of the methodology for this feature were explained in the Manual TimeDependent 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 steadyst ate KCODE calculation. The error associated with each collision rate is the error assuming the steadystate calculation is independent from the propagated error resulting from propagated flux and number densitiy error from the burnup calculation. Figure 43 displa ys a cropped example of the format of the collision rate outputs. These outputs help the us er to calculate generalized capturefission 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 43. 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 44 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 2.000E01 1.43413 1.699E+14 2.510 202.199 2.155E+01 1.550E+16 ... Figure 44. 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 5.015E01 0.000E+00 1 5.000E+01 5.000E+01 5.016E01 7.205E+00 2 1.000E+01 6.000E+01 5.002E01 7.205E+00 3 5.000E+02 5.600E+02 5.002E01 2.158E+01 Material #: 4 step duration time power fraction burnup (days) (days) (GWd/MTU) 0 0.000E+00 0.000E+00 4.985E01 0.000E+00 1 5.000E+01 5.000E+01 4.984E01 7.161E+00 2 1.000E+01 6.000E+01 4.998E01 7.161E+00 3 5.000E+02 5.600E+02 4.998E01 2.152E+01 ... Figure 45. Example indi vidual material burnup data. If multiple materials are burned then there al so exists individual material burnup data. Figure 45 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 (43) The individual material burnup is calculated by MTHM PFRAC TIME Fraction POWER POWER Burnup Burnupi* *1 (44) ... 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/bcm) 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 2.293E03 1.000E01 9.886E02 ... ... 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/bcm) 1 9023 1.286E09 6.837E04 5.315E+05 8.718E15 3.832E13 3.723E13 2 90232 2.394E08 2.625E15 1.097E07 1.616E13 7.100E12 6.929E12 ... ... totals 3.455E+03 2.584E+05 7.479E+01 2.275E02 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/bcm) 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 4.585E02 1.000E+00 1.000E+00 ... Figure 46. 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 46 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 47 shows a cropp ed example of print table 220. 1burnup summary table summed over all materials print table 220 nuclides with atom fractions below 1.000E32 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/bcm) 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 4.585E03 1.000E01 9.886E02 ... Figure 47. 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 cooperation, 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 burnup credit as applied to criticality safety in the transportation, storage, and treatment of spent fu el for a wide range of fuel types, including urainiumoxides (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 codetocode 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 51 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 51. 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 pincell 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 fueltomoderator 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 51. Infinitely reflected pin cell geometry. Table 52. 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 51 and the fuel pin dimens ions are listed in Table 52. 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 Zicaloy2 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 53, Table 54 and Table 55. 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 56 de tails the common operating strategies utilized for each case. Table 57 lists the operating power s utilized for each individual case in order to achieve the prescribed burnups. Table 53. Fuel composition for Phas e IB Benchmark (density =10.045 g/cm3). Nuclide Number Density (atoms/bcm) 234U 6.15164E6 235U 6.89220E4 236U 3.16265E6 238U 2.17104E2 12C 9.13357E6 14N 1.04072E5 16O 4.48178E2 ENDF/B VII.0 temperature dependent cross sections and scattering law data were implemented in the steady state reaction rate ca lculation. Each stead ystate 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 54. 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 55. Borated water composition for Phase IB Benchmark (0.7569 g/ cm3). Nuclide Number Density (atoms/bcm) 1H 5.06153E2 16O 2.53076E2 10B 2.75612E6 11B 1.11890E5 Table 56. 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 57. 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 58. 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 1.408E1 1.600E1 12.00 1.330E1 to 1.750E1 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 2.764E1 2.680E1 3.12 2.527E1 to 3.396E1 238Pu 9.046E2 1.012E1 10.61 5.721E2 to 1.083E1 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 6.491E1 6.812E1 4.71 5.310E1 to 7.335E1 242Pu 2.952E1 2.886E1 2.30 2.000E1 to 3.192E1 241Am 2.274E1 N/A N/A 2.269E1 to 2.598E1 243Am 4.487E2 N/A N/A 3.480E2 to 4.672E1 95Mo 5.501E1 N/A N/A 5.590E1 to 5.795E1 99Tc 5.397E1 N/A N/A 5.648E1 to 6.904E1 133Cs 8.012E1 8.500E1 5.74 6.820E1 to 8.640E1 135Cs 3.535E1 3.600E1 1.80 3.728E1 to 3.959E1 143Nd 5.752E1 6.130E1 6.16 6.040E1 to 6.792E1 145Nd 4.886E1 5.100E1 4.19 4.984E1 to 5.151E1 147Sm 1.522E1 N/A N/A 1.564E1 to 1.932E1 149Sm 1.628E3 2.900E3 43.86 1.626E3 to 2.900E3 150Sm 1.762E1 2.070E1 14.86 1.713E1 to 2.146E1 151Sm 7.070E3 N/A N/A 6.376E3 to1.413E2 152Sm 7.940E2 8.700E2 8.74 7.947E2 to 1.073E1 153Eu 7.022E2 7.900E2 11.12 6.730E2 to 8.921E2 155Gd 2.629E3 N/A N/A 1.507E3 to 5.762E3 *(Calculated/Measured1)*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 58, Table 59 and Table 510. 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 59. 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 1.148E01 1.400E01 17.96 1.080E1 to 1.570E01 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 3.980E01 3.560E01 11.79 3.619E01 to 4.919E01 238Pu 1.727E01 1.893E01 8.77 1.144E01 to 2.069E01 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 8.379E01 9.028E01 7.19 7.510E01 to 9.846E01 242Pu 5.916E01 5.761E01 2.69 4.200E01 to 6.347E01 241Am 2.853E01 N/A N/A 2.880E01 to 3.418E01 243Am 1.181E01 N/A N/A 9.637E02 to 1.391E01 95Mo 7.066E01 N/A N/A 7.214E01 to 7.545E01 99Tc 6.898E01 N/A N/A 7.327E01 to 8.372E01 133Cs 1.022 1.090 6.24 8.784E01 to 1.117 135Cs 3.793E01 4.000E01 5.17 3.967E01 to 4.317E01 143Nd 6.523E01 7.160E01 8.90 7.013E01 to 8.254E01 145Nd 6.112E01 6.530E01 6.41 6.326E01 to 6.600E01 147Sm 1.607E01 N/A N/A 1.659E01 to 2.201E01 149Sm 1.653E03 3.000E03 44.91 1.736E03 to 3.092E03 150Sm 2.311E01 2.710E01 14.72 2.297E01 to 3.152E01 151Sm 7.397E03 N/A N/A 7.990E03 to 1.571E02 152Sm 9.521E02 1.040E01 8.45 9.761E02 to 1.416E01 153Eu 9.940E02 1.090E01 8.81 9.960E02 to 9.960E02 155Gd 4.076E03 N/A N/A 2.538E03 to 1.028E02 *(Calculated/Measured1)*100 PAGE 135 135 Table 510. 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 9.896E02 1.200E01 17.53 9.030E02 to 1.440E01 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 4.794E01 4.680E01 2.44 4.327E01 to 5.934E01 238Pu 2.399E01 2.688E01 10.76 1.656E01 to 2.810E01 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 9.229E01 1.020 9.52 8.560E01 to 1.111 242Pu 8.479E01 8.401E01 0.93 5.960E01 to 9.103E01 241Am 3.071E01 N/A N/A 3.102E01 to 3.785E01 243Am 1.924E01 N/A N/A 1.634E01 to 2.316E01 95Mo 8.100E01 N/A N/A 8.092E01 to 8.742E01 99Tc 7.852E01 N/A N/A 8.449E01 to 9.861E01 133Cs 1.160 1.240 6.45 9.723E01 to 1.286 135Cs 3.958E01 4.300E01 7.94 3.977E01 to 4.605E01 143Nd 6.746E01 7.630E01 11.58 7.397E01 to 8.839E01 145Nd 6.858E01 7.440E01 7.82 7.170E01 to 7.560E01 147Sm 1.598E01 N/A N/A 1.655E01 to 2.302E01 149Sm 1.686E03 4.700E03 64.13 1.842E03 to 3.286E03 150Sm 2.690E01 3.610E01 25.49 2.725E01 to 3.980E01 151Sm 7.585E03 N/A N/A 8.102E03 to 1.682E02 152Sm 1.038E01 1.210E01 14.25 1.077E01 to 1.587E01 153Eu 1.173E01 1.480E01 20.75 1.210E01 to 1.596E01 155Gd 5.046E03 N/A N/A 3.379E03 to 1.318E02 *(Calculated/Measured1)*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 lflive 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 511. 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 1.422E01 1.600E1 11.13 1.330E1 to 1.750E1 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 2.794E01 2.680E1 4.25 2.527E1 to 3.396E1 238Pu 9.445E02 1.012E1 6.67 5.721E2 to 1.083E1 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 6.615E01 6.812E1 2.90 5.310E1 to 7.335E1 242Pu 3.069E01 2.886E1 6.36 2.000E1 to 3.192E1 241Am 2.320E01 N/A N/A 2.269E1 to 2.598E1 243Am 4.886E02 N/A N/A 3.480E2 to 4.672E1 95Mo 5.661E01 N/A N/A 5.590E1 to 5.795E1 99Tc 5.661E01 N/A N/A 5.648E1 to 6.904E1 133Cs 8.259E01 8.500E1 2.83 6.820E1 to 8.640E1 135Cs 3.620E01 3.600E1 0.54 3.728E1 to 3.959E1 143Nd 5.840E01 6.130E1 4.73 6.040E1 to 6.792E1 145Nd 4.894E01 5.100E1 4.04 4.984E1 to 5.151E1 147Sm 1.518E01 N/A N/A 1.564E1 to 1.932E1 149Sm 1.679E03 2.900E3 42.10 1.626E3 to 2.900E3 150Sm 1.854E01 2.070E1 10.44 1.713E1 to 2.146E1 151Sm 7.401E03 N/A N/A 6.376E3 to1.413E2 152Sm 8.124E02 8.700E2 6.62 7.947E2 to 1.073E1 153Eu 7.257E02 7.900E2 8.13 6.730E2 to 8.921E2 155Gd 2.724E03 N/A N/A 1.507E3 to 5.762E3 *(Calculated/Measured1)*100 A further investigation was completed to set the limiting checking value for density contribution of a specific nuclide to 1E10 from th e default 1E4 value in order to determine if this change actually affected the ultimate deplet ion outcome. Table 511, Table 512 and Table 513 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 512. 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 1.164E01 1.400E01 16.88 1.080E1 to 1.570E01 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 4.032E01 3.560E01 13.26 3.619E01 to 4.919E01 238Pu 1.750E01 1.893E01 7.55 1.144E01 to 2.069E01 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 8.699E01 9.028E01 3.65 7.510E01 to 9.846E01 242Pu 5.936E01 5.761E01 3.04 4.200E01 to 6.347E01 241Am 2.976E01 N/A N/A 2.880E01 to 3.418E01 243Am 1.186E01 N/A N/A 9.637E02 to 1.391E01 95Mo 7.293E01 N/A N/A 7.214E01 to 7.545E01 99Tc 7.277E01 N/A N/A 7.327E01 to 8.372E01 133Cs 1.060 1.090 6.24 8.784E01 to 1.117 135Cs 3.930E01 4.000E01 5.17 3.967E01 to 4.317E01 143Nd 6.750E01 7.160E01 8.90 7.013E01 to 8.254E01 145Nd 6.172E01 6.530E01 6.41 6.326E01 to 6.600E01 147Sm 1.632E01 N/A N/A 1.659E01 to 2.201E01 149Sm 1.711E03 3.000E03 44.91 1.736E03 to 3.092E03 150Sm 2.521E01 2.710E01 14.72 2.297E01 to 3.152E01 151Sm 7.924E03 N/A N/A 7.990E03 to 1.571E02 152Sm 9.892E02 1.040E01 8.45 9.761E02 to 1.416E01 153Eu 1.035E01 1.090E01 8.81 9.960E02 to 9.960E02 155Gd 4.251E03 N/A N/A 2.538E03 to 1.028E02 *(Calculated/Measured1)*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 513. 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 1.006E01 1.200E01 16.14 9.030E02 to 1.440E01 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 4.854E01 4.680E01 3.72 4.327E01 to 5.934E01 238Pu 2.434E01 2.688E01 9.47 1.656E01 to 2.810E01 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 9.633E01 1.020 5.56 8.560E01 to 1.111 242Pu 8.463E01 8.401E01 0.74 5.960E01 to 9.103E01 241Am 3.222E01 N/A N/A 3.102E01 to 3.785E01 243Am 1.919E01 N/A N/A 1.634E01 to 2.316E01 95Mo 8.359E01 N/A N/A 8.092E01 to 8.742E01 99Tc 8.303E01 N/A N/A 8.449E01 to 9.861E01 133Cs 1.206 1.240 2.71 9.723E01 to 1.286 135Cs 4.104E01 4.300E01 4.56 3.977E01 to 4.605E01 143Nd 7.038E01 7.630E01 7.76 7.397E01 to 8.839E01 145Nd 6.930E01 7.440E01 6.85 7.170E01 to 7.560E01 147Sm 1.635E01 N/A N/A 1.655E01 to 2.302E01 149Sm 1.751E03 4.700E03 62.75 1.842E03 to 3.286E03 150Sm 2.975E01 3.610E01 17.59 2.725E01 to 3.980E01 151Sm 8.239E03 N/A N/A 8.102E03 to 1.682E02 152Sm 1.084E01 1.210E01 10.39 1.077E01 to 1.587E01 153Eu 1.228E01 1.480E01 17.00 1.210E01 to 1.596E01 155Gd 5.289E03 N/A N/A 3.379E03 to 1.318E02 *(Calculated/Measured1)*100 Nuclear Data The MCNPX benchmark calculations used E NDF/B VII.0 temperature dependent cross section data for steadystate 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 timestep 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 fueltomoderator 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 1e10 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 514 and Table 515. 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 514. 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 1.422E01 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 2.794E01 6.65 6.11 4.25 238Pu 9.445E02 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 6.615E01 2.72 9.77 2.90 242Pu 3.069E01 6.65 12.07 6.36 133Cs 8.259E01 1.91 6.92% 2.83 135Cs 3.620E01 4.46 1.40% 0.54 143Nd 5.840E01 0.36 5.51% 4.73 150Sm 1.854E01 13.22 14.51% 10.44 152Sm 8.124E02 1.35 6.39% 6.62 153Eu 7.257E02 5.11 5.71% 8.13 *MONTEBURNS OLD = MONTEBURNS from prior thesis; 75 MONTEBURNS NEW =MOTNEBURNS 2.0 with higher fidelity model ***(Calculated/Measured1)*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 515. 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 1.164E01 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 4.032E01 14.38 36.40 13.26 238Pu 1.750E01 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 8.699E01 4.05 21.07 3.65 242Pu 5.936E01 7.41 77.15 3.04 133Cs 1.060 2.55 20.93 6.24 135Cs 3.930E01 4.79 7.42 5.17 143Nd 6.750E01 0.76 11.81 8.90 150Sm 2.521E01 8.96 30 14.72 152Sm 9.892E02 0.20 27 8.45 153Eu 1.035E01 13.17 30 8.81 *MONTEBURNS OLD = MONTEBURNS from prior thesis; 75 MONTEBURNS NEW =MOTNEBURNS 2.0 with higher fidelity model ***(Calculated/Measured1)*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 burnup 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 52. 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/TM12667.28, 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 52. Ta ble 516 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/TM12667.28.53 Table 516. 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 Zircaloy4 Guide tube data Inner radius 0.6502 cm Outer radius 0.6934 cm Material Zircaloy4 Instrument tube data Inner radius 0.6502 cm Outer radius 0.6934 cm Material Zircaloy4 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 steadystate 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 3dimensions, 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 AD 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 517 and Table 518 display the operating intervals, time steps used for each interval, corresponding down times, and cycle dependent soluble boron concentrations for Case AD. Table 517. 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 AB or 3637 for Cases CD. PAGE 149 149 Table 518. 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/TM12667.53 The SCALE sequence used in the calculations was SAS2H/ORIGENS analysis sequence and 27group crosssec tion library (27BURNUPLIB) of SCALE4.53 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 520. Table 519. 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 518 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 AB was <8%; however, at higher burnups, Cases CD, 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 519, the results show significant improvement in actinide prediction for uran ium and plutonium isotopes when utilizing MCNPXc. Table 520. 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 Np237 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 521. 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 Zircaloy2 Guide and Instrument tube data Inner radius 0.571 cm Outer radius 0.613 cm Material Zircaloy2 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 Pu239 and large in itial loading of Pu239 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 Pu239 possesses a smaller delayed neutron fraction ( = 0.0021) and shorter neutron lifetime (l = 27 s), than U235 ( =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 codetocode 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 codetocode comparison for MOX assembly modeling strategy. Table 522. Initial MOX fuel compositi on for Case A for Phase IVB Benchmark. Values Expressed in (atoms/bcm) Nuclide High Enrichment Medium Enrichment Low Enrichment 234U 2.57E07 2.64E07 2.68E07 235U 5.38E05 5.53E05 5.60E05 238U 2.12E02 2.18E02 2.21E02 238Pu 5.17E05 3.61E05 2.85E05 239Pu 1.13E03 7.87E04 6.20E04 240Pu 5.35E04 3.74E04 2.95E04 241Pu 1.94E04 1.36E04 1.07E04 242Pu 1.46E04 1.02E04 8.06E05 16O 4.66E02 4.66E02 4.65E02 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 Zircaloy2 clad fuel pins along with Zircaloy2 guide tubes. The geometry specifications for both cases are given in Table 521. 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 523. Initial MOX fuel compositi on for Case B for Phase IVB Benchmark. Values Expressed in (atoms/bcm) Nuclide High Enrichment Medium Enrichment Low Enrichment 234U 2.69E07 2.73E07 2.75E07 235U 5.63E05 5.71E05 5.74E05 238U 2.22E02 2.25E02 2.26E02 238Pu 5.09E07 3.56E07 2.81E07 239Pu 9.49E04 6.64E04 5.23E04 240Pu 6.06E05 4.24E05 3.34E05 241Pu 3.02E06 2.11E06 1.66E06 242Pu 5.01E07 3.50E07 2.76E07 16O 4.65E02 4.65E02 4.65E02 For each case, three types of plutonium concentr ation loadings were used for the fuel pins within each case. Table 522 lists the fuel compositions for Case A, and Table 523 lists the fuel composition for Case B. Table 524 lists the bo rated water composition a nd Table 525 lists the Zircaloy2 composition used for the fuel clad and guide tube s. Figure 53 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 526. Table 524. Borated water composition for Phase IVB Benchmark (660ppm boron, density 0.7245 g/cm3). Nuclide Number Density (atoms/bcm) 1H 4.8414E02 16O 2.4213E02 10B 4.7896E06 11B 1.9424E05 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 53. MOX fuel enrichment map for Phase IVB Benchmark. PAGE 156 156 Table 525. Zicaloy2 composition used for cl ad and guide tube material for Phase IVB Benchmark (density 5.8736 g/cm3). Nuclide (Atoms/bcm) 50Sn 2.9656E06 52Sn 5.7189E05 53Sn 6.4848E06 54Sn 1.6142E06 54Fe 7.8122E06 56Fe 1.2245E04 57Fe 2.8278E06 58Fe 3.7633E07 90Zr 1.9889E02 91Zr 4.3373E03 92Zr 6.6297E03 94Zr 6.7186E03 96Zr 1.0824E03 Table 526. 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 527, Table 528, Table 529 and Table 530 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 531, Table 532, Table 533 and Ta ble 534 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 527. EOC 1 results for Case A of the Phase IVB Benchmark. Nuclide MCNPX (atoms/bcm) Range of Values from Other Depletion Codes (atoms/bcm) 234U 5.8483E07 2.1279E07 to 5.8216E07 235U 4.3753E05 4.3720E05 to 4.3925E05 236U 2.6406E06 2.4858E06 to 2.6190E06 238U 2.1134E02 2.1155E02 to 2.1163E02 238Pu 4.1533E05 4.0713E05 to 4.1630E05 239Pu 8.0377E04 7.9700E04 to 8.1275E04 240Pu 4.7370E04 4.7109E04 to 4.8300E04 241Pu 2.1855E04 2.1520E04 to 2.2011E04 242Pu 1.6393E04 1.3389E04 to 1.3785E04 237Np 1.6176E06 1.0400E07 to 1.6560E06 241Am 9.4895E06 8.4140E06 to 8.8762E06 243Am 1.8069E05 1.5613E05 to 1.8260E05 242Cm 9.5381E07 8.7154E07 to 9.8523E07 243Cm 1.3915E08 1.1641E08 to 1.4560E08 244Cm 5.3807E06 2.6688E06 to 3.9610E06 245Cm 2.8581E07 1.2915E07 to 1.9877E07 95Mo 1.2829E05 1.1690E05 to 1.7651E05 99Tc 2.1958E05 2.1756E05 to 2.3060E05 101Ru 2.2331E05 2.1640E05 to 2.3054E05 103Rh 2.0106E06 1.9630E05 to 2.2642E05 109Ag 5.0294E06 4.6603E06 to 5.9460E06 133Cs 2.4605E05 2.3463E05 to 2.4597E05 143Nd 1.5949E05 1.5350E05 to 1.6594E05 145Nd 1.1238E05 1.1183E05 to 1.1518E05 147Sm 8.0947E07 7.8960E07 to 8.2700E07 149Sm 4.2895E07 3.7767E07 to 4.3945E07 150Sm 4.7391E06 4.7807E06 to 5.1050E06 151Sm 1.3397E06 1.3420E06 to 1.5281E06 152Sm 2.7837E06 2.7790E06 to 3.1530E06 153Eu 2.0936E06 2.0380E06 to 2.2411E06 155Gd 8.6977E09 4.3940E09 to 1.6769E08 PAGE 158 158 Table 528. EOC 2 results for Case A of the Phase IVB Benchmark. Nuclide MCNPX (atoms/bcm) Range of Values from Other Depletion Codes (atoms/bcm) 234U 8.0735E07 1.7413E07 to 8.4354E07 235U 3.4593E05 3.4500E05 to 3.4893E05 236U 4.6979E06 4.4215E06 to 4.6370E06 238U 2.0896E02 2.0910E02 to 2.0927E02 238Pu 3.9782E05 3.5013E05 to 3.9340E05 239Pu 6.3728E04 6.2980E04 to 6.5524E04 240Pu 4.4837E04 4.4499E04 to 4.6350E04 241Pu 2.3731E04 2.3210E04 to 2.3979E04 242Pu 1.7027E04 1.4147E04 to 1.4982E04 237Np 3.1341E06 3.6114E07 to 3.2139E06 241Am 1.5907E05 1.4680E05 to 1.5982E05 243Am 2.9191E05 2.7797E05 to 3.1620E05 242Cm 2.5536E06 2.3253E06 to 2.6204E06 243Cm 7.2443E08 5.8709E08 to 7.0660E08 244Cm 1.7168E05 9.2778E06 to 1.3050E05 245Cm 1.7035E06 8.1985E07 to 1.2348E06 95Mo 2.9454E05 2.8330E05 to 3.4297E05 99Tc 4.1886E05 4.1811E05 to 4.4400E05 101Ru 4.3865E05 4.2320E05 to 4.5186E05 103Rh 4.0075E05 3.7469E05 to 4.0914E05 109Ag 9.0959E06 8.0989E06 to 1.0640E05 133Cs 4.6303E05 4.4403E05 to 4.6602E05 143Nd 3.0328E05 2.9970E05 to 3.1809E05 145Nd 2.1678E05 2.1322E05 to 2.2191E05 147Sm 2.1430E06 2.4130E06 to 2.4742E06 149Sm 4.0373E07 3.5930E07 to 4.1990E07 150Sm 1.0296E05 1.0197E05 to 1.1350E05 151Sm 1.5541E06 1.5306E06 to 1.8870E06 152Sm 4.9712E06 5.0468E06 to 5.9850E06 153Eu 4.9325E06 5.0480E06 to 5.4745E06 155Gd 1.6016E08 7.9700E09 to 2.8935E08 PAGE 159 159 Table 529. EOC 3 results for Case A of the Phase IVB Benchmark. Nuclide MCNPX (atoms/bcm) Range of Values from Other Depletion Codes (atoms/bcm) 234U 9.4561E07 1.4169E07 to 1.0509E06 235U 2.6715E05 2.6596E05 to 2.7129E05 236U 6.2239E06 5.8461E06 to 6.1140E06 238U 2.0648E02 2.0657E02 to 2.0678E02 238Pu 4.0397E05 2.9577E05 to 3.9688E05 239Pu 5.0933E04 5.0368E04 to 5.3435E04 240Pu 4.1054E04 4.0836E04 to 4.2870E04 241Pu 2.3785E04 2.3107E04 to 2.4058E04 242Pu 1.8086E04 1.5231E04 to 1.6558E04 237Np 4.3966E06 6.9348E07 to 4.5652E06 241Am 1.8127E05 1.7960E05 to 2.0066E05 243Am 3.6401E05 3.6793E05 to 4.2000E05 242Cm 4.3963E06 3.5650E06 to 4.0436E06 243Cm 1.5834E07 1.2551E07 to 1.4820E07 244Cm 3.1800E05 1.8435E05 to 2.4950E05 245Cm 4.1506E06 2.1928E06 to 3.2335E06 95Mo 4.3988E05 4.3990E05 to 4.9929E05 99Tc 5.9641E05 6.0068E05 to 6.3850E05 101Ru 6.4487E05 6.1990E05 to 6.6364E05 103Rh 5.3043E05 5.1041E05 to 5.5357E05 109Ag 1.2395E05 1.0524E05 to 1.4480E05 133Cs 6.4706E05 6.2829E05 to 6.6172E05 143Nd 4.2327E05 4.2830E05 to 4.5385E05 145Nd 3.1270E05 3.0386E05 to 3.2104E05 147Sm 3.1679E06 4.0280E06 to 4.1579E06 149Sm 3.2126E07 3.2302E07 to 3.8520E07 150Sm 1.5873E05 1.5410E05 to 1.7660E05 151Sm 1.5666E06 1.5340E06 to 2.0110E06 152Sm 6.3075E06 6.3335E06 to 7.9580E06 153Eu 7.5412E06 8.1310E06 to 8.6655E06 155Gd 1.9499E08 1.3920E08 to 4.2299E08 PAGE 160 160 Table 530. Five year cooling time results for Case A of the Phase IVB Benchmark. Nuclide MCNPX (atoms/bcm) Range of Values from Other Depletion Codes (atoms/bcm) 234U 2.6607E06 1.4169E07 to 2.7220E06 235U 2.6785E05 2.6668E05 to 2.7206E05 236U 6.4390E06 5.8461E06 to 6.2030E06 238U 2.0648E02 2.0657E02 to 2.0678E02 238Pu 4.3111E05 2.8432E05 to 4.1989E05 239Pu 5.1081E04 5.0542E04 to 5.3613E04 240Pu 4.1585E04 4.1217E04 to 4.3260E04 241Pu 1.8680E04 1.8167E04 to 1.8891E04 242Pu 1.8086E04 1.5231E04 to 1.6558E04 237Np 4.8116E06 7.0645E07 to 4.9848E06 241Am 6.8783E05 6.6320E05 to 7.0845E05 243Am 3.6406E05 3.6793E05 to 3.9880E05 242Cm 2.5594E09 1.6887E09 to 5.0942E09 243Cm 1.4022E07 1.1143E07 to 1.2852E07 244Cm 2.6279E05 1.5224E05 to 1.9671E05 245Cm 4.1491E06 2.1928E06 to 3.2321E06 95Mo 4.9799E05 4.9929E05 to 5.0765E05 99Tc 5.9856E05 6.0280E05 to 6.4090E05 101Ru 6.4487E05 6.3251E05 to 6.6366E05 103Rh 5.6463E05 5.1041E05 to 5.8208E05 109Ag 1.2416E05 1.0524E05 to 1.4490E05 133Cs 6.5257E05 6.2829E05 to 6.6172E05 143Nd 4.3169E05 4.4116E05 to 4.5385E05 145Nd 3.1278E05 3.0386E05 to 3.2104E05 147Sm 7.8905E06 9.9452E06 to 1.0734E05 149Sm 3.6852E07 3.7494E07 to 4.2796E07 150Sm 1.5873E05 1.5410E05 to 1.6679E05 151Sm 1.5196E06 1.4779E06 to 1.8916E06 152Sm 6.3090E06 6.3335E06 to 7.9580E06 153Eu 7.5868E06 8.1650E06 to 8.6716E06 155Gd 3.6011E07 3.3400E07 to 1.0081E06 PAGE 161 161 Table 531. EOC 1 results for Case B of the Phase IVB Benchmark. Nuclide MCNPX (atoms/bcm) Range of Values from Other Depletion Codes (atoms/bcm) 234U 2.1859E07 2.1683E07 to 2.2681E07 235U 4.2074E05 4.1730E05 to 4.2085E05 236U 3.1619E06 3.0287E06 to 3.1900E06 238U 2.2041E02 2.2000E02 to 2.2049E02 238Pu 1.1757E06 3.7913E07 to 7.1826E07 239Pu 5.6960E04 5.6280E04 to 5.7659E04 240Pu 1.5241E04 1.4600E04 to 1.5008E04 241Pu 6.0895E05 5.6781E05 to 6.3800E05 242Pu 7.1572E06 5.3572E06 to 6.1170E06 237Np 1.5688E06 1.3364E07 to 1.6488E06 241Am 1.6417E06 1.3107E06 to 1.4360E06 243Am 9.6882E07 6.2689E07 to 8.0100E07 242Cm 1.8012E07 1.5269E07 to 1.7950E07 243Cm 2.0951E09 1.6814E09 to 2.0300E09 244Cm 2.4339E07 9.3251E08 to 1.3100E07 245Cm 1.1257E08 3.4090E09 to 4.6024E09 95Mo 1.3153E05 1.2181E05 to 1.8204E05 99Tc 2.1838E05 2.1900E05 to 2.2950E05 101Ru 2.1979E05 2.1769E05 to 2.3058E05 103Rh 2.0336E05 1.8759E05 to 2.1765E05 109Ag 4.3321E06 3.6778E06 to 5.5840E06 133Cs 2.4560E05 2.3439E05 to 2.4574E05 143Nd 1.5453E05 1.5100E05 to 1.6346E05 145Nd 1.0969E05 1.1117E05 to 1.1454E05 147Sm 7.6879E07 7.6500E07 to 7.8701E07 149Sm 2.3529E07 1.9930E07 to 2.2432E07 150Sm 4.7705E06 4.9051E06 to 5.2700E06 151Sm 8.6623E07 8.8308E07 to 1.0200E06 152Sm 2.8949E06 2.9600E06 to 3.3370E06 153Eu 2.0836E06 2.0570E06 to 2.2659E06 155Gd 4.5104E09 1.8000E09 to 7.6886E09 PAGE 162 162 Table 532. EOC 2 results for Case B of the Phase IVB Benchmark. Nuclide MCNPX (atoms/bcm) Range of Values from Other Depletion Codes (atoms/bcm) 234U 1.8049E07 1.7098E07 to 1.9333E07 235U 2.9106E05 2.8760E05 to 2.9257E05 236U 5.5884E06 5.3234E06 to 5.5600E06 238U 2.1773E02 2.1767E02 to 2.1800E02 238Pu 2.6720E06 2.9817E07 to 2.0587E06 239Pu 3.7031E04 3.6608E04 to 3.8300E04 240Pu 1.8694E04 1.8070E04 to 1.8624E04 241Pu 9.7981E05 9.4093E05 to 1.0080E04 242Pu 2.2218E05 1.9550E05 to 2.1840E05 237Np 3.0365E06 4.5945E07 to 3.1897E06 241Am 4.1152E06 3.6692E06 to 3.9280E06 243Am 4.3621E06 3.6612E06 to 4.7300E06 242Cm 8.9769E07 7.7886E07 to 8.9591E07 243Cm 4.3621E06 1.6565E08 to 2.0900E08 244Cm 1.8948E06 1.0277E06 to 1.4100E06 245Cm 1.4495E07 6.0866E08 to 8.6913E08 95Mo 2.9951E05 2.9300E05 to 3.5096E05 99Tc 4.1181E05 4.1730E05 to 4.3780E05 101Ru 4.3022E05 4.2699E05 to 4.5057E05 103Rh 3.5805E05 3.3166E05 to 3.7099E05 109Ag 7.7718E06 6.5223E06 to 9.7410E06 133Cs 4.5646E05 4.3917E05 to 4.6063E05 143Nd 2.8262E05 2.8500E05 to 3.0200E05 145Nd 2.0952E05 2.1048E05 to 2.1882E05 147Sm 1.9461E06 2.2520E06 to 2.2924E06 149Sm 2.0460E07 1.7336E07 to 1.9700E07 150Sm 1.0170E05 1.0500E05 to 1.1600E05 151Sm 8.4275E07 8.4937E07 to 1.0900E06 152Sm 4.7605E06 4.8514E06 to 5.9030E06 153Eu 4.8992E06 5.1650E06 to 5.5763E06 155Gd 9.7191E09 3.8100E09 to 1.1850E08 PAGE 163 163 Table 533. EOC 3 results for Case A of the Phase IVB Benchmark. Nuclide MCNPX (atoms/bcm) Range of Values from Other Depletion Codes (atoms/bcm) 234U 1.1520E13 1.3174E07 to 1.7954E07 235U 1.8445E05 1.8119E05 to 1.8791E05 236U 7.1794E06 6.8091E06 to 7.0700E06 238U 2.1481E02 2.1467E02 to 2.1500E02 238Pu 5.5067E06 2.2123E07 to 4.7927E06 239Pu 2.5259E04 2.4767E04 to 2.7200E04 240Pu 1.8124E04 1.7470E04 to 1.8437E04 241Pu 1.0690E04 1.0323E04 to 1.0850E04 242Pu 4.3597E05 3.9815E05 to 4.3080E05 237Np 4.1982E06 8.5608E07 to 4.4424E06 241Am 4.9089E06 4.9773E06 to 5.4746E06 243Am 1.0121E05 9.3994E06 to 1.1700E05 242Cm 1.9925E06 1.5821E06 to 1.8030E06 243Cm 6.2628E08 4.8232E08 to 5.9200E08 244Cm 6.7113E06 4.1608E06 to 5.4200E06 245Cm 6.6887E07 3.2603E07 to 4.7115E07 95Mo 4.4224E05 4.5000E05 to 5.0570E05 99Tc 5.7742E05 5.9218E05 to 6.2220E05 101Ru 6.2895E05 6.2300E05 to 6.5908E05 103Rh 4.3988E05 4.2315E05 to 4.7011E05 109Ag 1.0536E05 8.5551E06 to 1.2920E05 133Cs 6.2697E05 6.1230E05 to 6.4442E05 143Nd 3.7203E05 3.8700E05 to 4.0746E05 145Nd 2.9770E05 2.9641E05 to 3.1166E05 147Sm 2.7210E06 3.5820E06 to 3.6688E06 149Sm 1.2934E07 1.4821E07 to 1.6900E07 150Sm 1.5432E05 1.5880E05 to 1.7900E05 151Sm 7.8954E07 8.0820E07 to 1.1000E06 152Sm 5.6451E06 5.7314E06 to 7.4850E06 153Eu 7.1295E06 7.9849E06 to 8.6528E06 155Gd 8.0878E09 6.3000E09 to 1.5641E08 PAGE 164 164 Table 534. Five year coo ling time results for Case B of the Phase IVB Benchmark. Nuclide MCNPX (atoms/bcm) Range of Values from Other Depletion Codes (atoms/bcm) 234U 4.4485E07 1.3174E07 to 4.1870E07 235U 1.8480E05 1.8119E05 to 1.8830E05 236U 7.2745E06 6.9037E06 to 7.0634E06 238U 2.1481E02 2.1467E02 to 2.1493E02 238Pu 7.2430E06 2.1267E07 to 6.3483E06 239Pu 2.5476E04 2.4983E04 to 2.7284E04 240Pu 1.8232E04 1.7540E04 to 1.8510E04 241Pu 8.3970E05 8.1013E05 to 8.5710E05 242Pu 4.3597E05 3.9817E05 to 4.3090E05 237Np 4.3965E06 8.7398E07 to 4.6418E06 241Am 2.7709E05 2.6942E05 to 2.8303E05 243Am 1.0125E05 9.3994E06 to 1.0616E05 242Cm 9.8698E10 7.5522E10 to 1.4676E09 243Cm 5.5486E08 4.2821E08 to 5.1277E08 244Cm 5.5500E06 3.4362E06 to 3.8927E06 245Cm 6.6887E07 3.2603E07 to 4.7096E07 95Mo 5.0010E05 5.0570E05 to 5.1528E05 99Tc 5.7945E05 5.9428E05 to 6.2460E05 101Ru 6.2895E05 6.2654E05 to 6.5909E05 103Rh 4.7293E05 4.2315E05 to 4.9970E05 109Ag 1.0552E05 8.5551E06 to 1.2940E05 133Cs 6.3248E05 6.1230E05 to 6.4442E05 143Nd 3.8029E05 3.9608E05 to 4.0746E05 145Nd 2.9778E05 2.9642E05 to 3.1176E05 147Sm 6.8077E06 8.7798E06 to 9.5399E06 149Sm 1.7692E07 2.0516E07 to 2.2260E07 150Sm 1.5432E05 1.5880E05 to 1.7344E05 151Sm 7.7139E07 7.7864E07 to 9.9180E07 152Sm 5.6451E06 5.7314E06 to 7.4860E06 153Eu 7.1797E06 7.9849E06 to 8.7110E06 155Gd 3.3990E07 3.2182E07 to 9.2307E07 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 odecode 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 39 and 310. As a result of the he ritage of the data release, MCNPX depletion expects Am242m 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 63group 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 Am243 which decays to Cm243. The nuclide Am242 decays quickly to Cm242 which then decays to Pu238. 242Am has a higher capturetofission 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 Cm243 Cm244 and Cm245. This hypothesis seems to be validated by the fact that Cm243, Cm 244 and Cm245 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 timedependent 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 easytouse depletion calculations in a single, well established, radia tion transport Monte Carlo code. The capability links the wellestablished MCNPX steadystat 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 dystate 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 CELL2 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 fueltomoderator 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, uraniumdioxide 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 codetocode 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 capturetofission 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 2.433 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 odetocode 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 adjustedfueltomoderator 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 1020% 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 nonrare events leading to an inadequate sampling of nonrare 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 Carlolinked 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 6.87812e2 1 u=2 imp:n=1 vol=192.287 $ fuel 3 2 4.5854e2 1 2 u=2 imp:n=1 vol=66.43 $ clad 4 3 7.1594e2 2 u=2 imp:n=1 vol=370.82 $ water 6 4 6.87812e2 1 u=3 imp:n=1 vol=192.287 $ fuel 7 5 4.5854e2 1 2 u=3 imp:n=1 vol=66.43 $ clad 8 6 7.1594e2 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= 1e32 MATVOL= 384.57 384.57 MATMOD= 2 1 1 4 1 94238 1e6 2 2 1 2 94238 1e6 94241 1e6 4 1 94238 1e6 ... 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: Qvalue 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 63group flux from MCNPX to be matched to a 63group 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 1e32 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 1e6 atoms/barncm. At time step 2, material 1 will have the atom densities of 2 isotopes changed: material 1 will have isotope 94238 changed to 1e6 atoms/barncm and isotope 94241 will be changed to 1e6 at oms/barncm, material 4 will have the atom density of 1 isotope changed: 94238 w ill be set to 1e6 atoms/barncm. PAGE 179 179 LIST OF REFERENCES 1. U.S. Department of Energy Office of Civilian Radioactive Waste Management, Oklo: Natural Nuclear Reactors, November 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. 8996 (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: LAUR041572, 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: GNEP167312 (2007). 19. M. D. LOWENTHAL, Transmutation in the Nucl ear Fuel Cycle: Approaches and Impacts, Journal of Nuclear Technology 138, pg. 284299 (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. 449456 (2002). 22. M. SALVATORES, Nuclear Fuel Cycle Strategies Including Partitioning and Transmutation, Nuclear Engineerin g and Design, 235, pg. 805816 (2005). 23. United States Department of Energy, Global Nuclear Energy Partnership: Improved Nuclear Safeguards, December 2006, 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/TM252, 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. 612620 (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. 97108. 30. M. EDENIUS, K. EKBERG, B. H. FORSSEN, and D. KNOTT, Casmo4, A Fuel Assembly Burnup Program, Users Manual, Studsvik Repor t: SOA95/15, 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 NextGeneration Neutronics Design System, Nuclear Science and Engineering 155, pp. 276289 (2007). 32. D. KNOTT, E. WEHLAGE, Description of the LANCER02 Lattice Physics Code for SingleAssembly and Multibundle Analysis, Nuclear Science and Engineering 155, pp. 331354 (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. 222233 (1998). 34. A. NUNEZCARRERA, J. L. FRANCOIS, G. ESPINOSAPAREDES, Comparison between HELIOS CriticalDepletion Calcul ations and a PWR Thorium Cell Burnup Benchmark, Annuls of Nuclear Energy 31, pg. 713722 (2004). 35. M. L. FENSIN and S. ANGHAIE, Current Me thods of Depletion Analysis, Los Alamos National Laboratory Report LAUR063933, American Nuclear Societys 2006 Winter Meeting, Albuquerque, NM (2006). 36. G. S. CHANG, J. M. RYSKAMP, Depletion An alysis of MixedOxide Fuel Pins in Light Water Reactors and the Advanced Test Reactor, Nuclear Technology 129, pp. 326337 (2000). 37. W. HAECK, An Optimum Approach to Monte Carlo BurnUp, PhD thesis, Ghent University, Belgium (2007). 38. E. J. PARMA, BURNCAL: A Nuclear Reactor Burnup Code Using MCNP Tallies, Sandia National Laboratory Report: SAND 20023868, 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, MCWOLinking 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 MonteCarlo 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. 127132 (2003). 44. R. L. MOORE, B. G. SCHNITZLER, C. A. WEMPLE, R. S. BABCOCK, D. E. WESSOL, MOCUP: MCNPORIGEN2 Coupled Utility Program, Idaho National Engineering Laboratory Report: INEL95/0523, 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 LAUR994999, 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. 1925 (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: ANLTDCP97492, 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, HighFidelity 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 LACP050369, 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 LAUR933080, 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 IB 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/TM12667, Oak Ridge, TN, USA (1995). 54. G. J. OCONNOR and P. H. LIEM Burnup Credit Criticality Benchmark PHASE IVB: 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. 407414 (1993). 56. M. YAMAMOTO, Coarse TimeStep Integrati on Method for Burnup Calculation of LWR Lattice Containing GadoliniumPoisoned Rods, Journal of Nuclear Science and Technology 22, pg.115 (1985). 57. G. ILAS, B. D. MURPHY, and I. C. GAULD, O verview of OrigenARP 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. 335352 (1983). 59. G. F. THOMAS and D. H. BARBER, Sti ffness in Radioactive Decay Chains, Annuls of Nuclear Energy 21, pg 309320 (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: ORNL5621, 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. 197210 (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. 6577 (1999). 70. H. HERMAN, ENDF6 FORMATS MANUAL, Brookhaven National Laboratory Report: BNLNCS4494505Rev., Long Island, NY (2005). 71. M. L. FENSIN and D. I. POSTON, Optimum Reflector Configurat ions for Minimizing Fission Power Peaking in a LithiumCooled, Li quidMetal Reactor with Sliding Reflectors, Los Alamos National Laboratory Report LAUR7154, 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 LAUR058018, 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, MCNPA General Monte Carlo NParticle Transport Code, Los Alamos National Laboratory Report: LA13709M, 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 LA13514T, Los Alamos, NM (1998). 76. S. M. BOWMAN, M. D. DEHART, and L. M. PETRIE, Integrated Keno Monte Carlo Transport for 3Depletion 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 Tokaimura, 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. 154157 (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: LAUR058225, 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/BVII.0: Next Generation Evaluated Nuclear Data Library for Nuclear Science and Technology", Nuclear Data Sheets 107, pg. 29313060 (2006). 81. T. R. ENGLAND, B. F. RIDER, ENDF349 evaluation and compilation of fission product yields, Los Alamos Nationa l Laboratory Report: LAUR943106, 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 LAUR057895, 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 TwoTiered System Comprising Light Water Reactors and AcceleratorDriven Systems,Los Alamos National Laboratory Report: LA14 052T, 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: LAUR067991, 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 LAUR060363, 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: LAUR076632, 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, 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: LAUR063248, 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: LAUR074137, 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: LAUR080305, 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 Burnup Credit Criticality Safety, September 2003, 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. 