<%BANNER%>

Process Development and Simulation of Hybrid Photovoltaic Cells

Permanent Link: http://ufdc.ufl.edu/UFE0022458/00001

Material Information

Title: Process Development and Simulation of Hybrid Photovoltaic Cells
Physical Description: 1 online resource (206 p.)
Language: english
Creator: Monroe, Matthew
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: Chemical Engineering -- Dissertations, Academic -- UF
Genre: Chemical Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The fabrication and simulation of hybrid bulk heterojunction photovoltaic cells was studied to develop an understanding of these devices and facilitate improvements in device fabrication techniques. The device simulation software package Medici was applied to this type of device for the first time, and novel device fabrication techniques were developed to improve device performance. Medici was adapted for a hybrid system consisting of an ordered array of inorganic nanorods interspersed with an absorbing, semiconducting polymer. A cell taken from the literature was simulated using a two-stage simulation technique which independently calculated the light absorption profiles and the cell performance. This technique applied a line source of carriers directly at the interface between the inorganic and organic regions of the cell, realistically imitating the exciton dissociation physics of real hybrid cells. The simulations showed low current densities as compared to the real cell, due to strong recombination at the material interface. Nitrogen plasma treatment of the transparent anode in bi-layer and hybrid photovoltaic cells was found to reduce surface roughness, increase the hydrophilic nature of the film, and improve cell performance. A range of solvents were tested for hybrid bulk heterojunction film deposition, with chloroform, chlorobenzene, and o-dichlorobenzene found to provide films with low surface roughness and strong uniformity. Hybrid bulk heterojunction solar cells were fabricated, and these cells showed low performance of less than 1% efficiency. The active layers degraded upon exposure to air, resulting in a drop in short-circuit current density that was more pronounced for hybrid films than for pure polymer single-layer cells. These findings highlight the environmental sensitivity of these devices and the need for an inert environment for cell fabrication and testing.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Matthew Monroe.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Anderson, Timothy J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022458:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022458/00001

Material Information

Title: Process Development and Simulation of Hybrid Photovoltaic Cells
Physical Description: 1 online resource (206 p.)
Language: english
Creator: Monroe, Matthew
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: Chemical Engineering -- Dissertations, Academic -- UF
Genre: Chemical Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The fabrication and simulation of hybrid bulk heterojunction photovoltaic cells was studied to develop an understanding of these devices and facilitate improvements in device fabrication techniques. The device simulation software package Medici was applied to this type of device for the first time, and novel device fabrication techniques were developed to improve device performance. Medici was adapted for a hybrid system consisting of an ordered array of inorganic nanorods interspersed with an absorbing, semiconducting polymer. A cell taken from the literature was simulated using a two-stage simulation technique which independently calculated the light absorption profiles and the cell performance. This technique applied a line source of carriers directly at the interface between the inorganic and organic regions of the cell, realistically imitating the exciton dissociation physics of real hybrid cells. The simulations showed low current densities as compared to the real cell, due to strong recombination at the material interface. Nitrogen plasma treatment of the transparent anode in bi-layer and hybrid photovoltaic cells was found to reduce surface roughness, increase the hydrophilic nature of the film, and improve cell performance. A range of solvents were tested for hybrid bulk heterojunction film deposition, with chloroform, chlorobenzene, and o-dichlorobenzene found to provide films with low surface roughness and strong uniformity. Hybrid bulk heterojunction solar cells were fabricated, and these cells showed low performance of less than 1% efficiency. The active layers degraded upon exposure to air, resulting in a drop in short-circuit current density that was more pronounced for hybrid films than for pure polymer single-layer cells. These findings highlight the environmental sensitivity of these devices and the need for an inert environment for cell fabrication and testing.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Matthew Monroe.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Anderson, Timothy J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022458:00001


This item has the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101106_AAAAFO INGEST_TIME 2010-11-06T23:32:06Z PACKAGE UFE0022458_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 12210 DFID F20101106_AABQFD ORIGIN DEPOSITOR PATH monroe_m_Page_169.QC.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
ed196cdd9ed3f8453b7b1121ad2e5107
SHA-1
471bf7926eef9b8f7012a75229ee660807b85672
13392 F20101106_AABQEP monroe_m_Page_160.QC.jpg
d856b54eaab93d4049d956c7aadbb029
aa8855baa67b887c943b35408ddf26a219d4ee3e
15577 F20101106_AABPZJ monroe_m_Page_076.QC.jpg
61fb2dddf5c3ecee891c2f090271f7ad
cc980781122214fb2d3488ce3f2c68ddd0d0f348
7226 F20101106_AABPYU monroe_m_Page_066thm.jpg
27bdac69152e44b4ab2e8403233b8215
89f5ea7be668ab6af19549bc880b1c952c2651f1
1038717 F20101106_AABOWH monroe_m_Page_004.jp2
1198e6f82cd23d4b7d3dd906e64624a4
a2fa4b019e22d0d52ae9fa0ed8015343dfc4fff8
679211 F20101106_AABPAZ monroe_m_Page_154.jp2
01e15c9f3db0ceaf6c3c4af79792be4f
5549c03911af8f0ce713337d349456de7a6dd733
45446 F20101106_AABOVS monroe_m_Page_193.jpg
8c26310e74bef606fe99e4d0b797df66
c36ec5fe75f3a55835b0ef46a7ea0dcbd5cc5123
1051983 F20101106_AABPCB monroe_m_Page_192.jp2
2ed65055384257c375412b8f8f767569
517c062d08fefed5b81dfd5b69ed1dc5cd3849ef
1051976 F20101106_AABPBN monroe_m_Page_172.jp2
f69d86f081aa6f802524c38e82f09192
06815382450b5f0d76ad207d77f1431a1d016e8b
21406 F20101106_AABQFE monroe_m_Page_170.QC.jpg
6d5c9216b0cc9e7438361cbd44c8df9a
b569151b348ff6f882ee9a6a0933c8fdddf595aa
14893 F20101106_AABQEQ monroe_m_Page_161.QC.jpg
7af60b7cb3fc95809e2df3ec14826809
a09d78a7b13f9a4818fc69ef5adccdce4f0c94a8
14195 F20101106_AABPZK monroe_m_Page_077.QC.jpg
9c037b1889e200b612cbb4f04f024d23
028023a2e8afd33f8190def3cf424fdb0aea41cd
24266 F20101106_AABPYV monroe_m_Page_067.QC.jpg
1cbcd3d1d14db5aad51c06be84f44f3a
d0e8b68b6d0fbef6c5c21bb87eb6d16060c3e7f6
F20101106_AABOWI monroe_m_Page_005.jp2
d230e7f7ac19f27fa7a9df1e5de4e59a
777b6cf7c9d329ead06373cb7fd31bc7861a7651
27885 F20101106_AABOVT monroe_m_Page_194.jpg
3674c49b86d54c4fdeaf7da5514c98f4
ec7a214c0b0ef41c80d5b55f498b442f22b91dae
1051779 F20101106_AABPCC monroe_m_Page_193.jp2
9bd23360e6e6315b21ee487babc82490
6ab40e61785d83dc62b9fd48fa5979a5693fc24b
766415 F20101106_AABPBO monroe_m_Page_173.jp2
f1341f3d916cff019d4b342710044532
ba3cc774883fcd89a26c2d7de27d8ffb394ce5df
6228 F20101106_AABQFF monroe_m_Page_170thm.jpg
2699239f9d6282a205c398fa25777ea6
fad56cd69073deae3590109c343e059e882c453e
4724 F20101106_AABQER monroe_m_Page_161thm.jpg
5f9c41713c4b8c61f3310865569ffba5
1e4dddc695c6f4f4e2454be18a20b02bd6afc7d8
4822 F20101106_AABPZL monroe_m_Page_077thm.jpg
b2fc6c8920c9bae8b6d3bfa0b63bfda9
4c47d1f5612cb56dbaf8b96df972a202ea9089cb
6651 F20101106_AABPYW monroe_m_Page_067thm.jpg
0d8b900f1d892db161b68ea345854c88
d822f2d9a0eed4113a80212565827d137c0e03b1
1051984 F20101106_AABOWJ monroe_m_Page_006.jp2
b333d179975d18ff9f09a081d6bc1b0b
35b19f123ac910fbef9be17e420e81a21d3e8432
75642 F20101106_AABOVU monroe_m_Page_195.jpg
11b90b719d8fea6a9c43add4517726a0
c4eb5e46c8be3b42516ecd98f8a4e7304582b892
1051986 F20101106_AABPCD monroe_m_Page_195.jp2
fe548eed07a69d4763f616a9ed47b92f
8466c57e88d80976a48dfbded69cf347b2297bec
1051981 F20101106_AABPBP monroe_m_Page_176.jp2
f3e723b78c9f87fcfd5feeeed73fc39f
996b24f88408d5a124403644e9f168140d162bc9
20267 F20101106_AABQFG monroe_m_Page_171.QC.jpg
40c9ea3c9107b2b9e11a797e26c643af
2fb49602a5c945c899f3442cc12835b899ae832a
5718 F20101106_AABQES monroe_m_Page_162thm.jpg
d745d7bdc08e49c7bf2f5866e8cd99c1
a82aa4bf1c558ff6fac4172f9e67fb809a52693d
10192 F20101106_AABPZM monroe_m_Page_078.QC.jpg
1f958eec608b696d4a29d03015bcc6b0
4cd5a3ecabab71723416a63fe215b312625a7644
26913 F20101106_AABPYX monroe_m_Page_068.QC.jpg
3764678b552ffc65a3d28081b26fc8ef
4b7056b9720702329de2e59de5309b38961a1fe8
1051977 F20101106_AABOWK monroe_m_Page_007.jp2
9e6c44d0cb4009292104e8d9997970ce
1cc7af4bd9ea30c438e54d682716516d619e4d46
80953 F20101106_AABOVV monroe_m_Page_196.jpg
d96d27c2d2e5adc1f31c9ffdcf8de567
c5bfeba60f9daafa288448a7dcdb61e2a8485bfe
1051954 F20101106_AABPCE monroe_m_Page_196.jp2
1116561aff7abdf290d4b90a863c5892
813c0832890bfe95ba18507c85809ae099ef0ba9
865541 F20101106_AABPBQ monroe_m_Page_177.jp2
8c9a5e7a7b0afad32a02c7794fac654d
490b0b9ab374e88214c26aa35b28f07a86881944
6198 F20101106_AABQFH monroe_m_Page_171thm.jpg
602e383c8d0288952b856954f3d82bf4
24cf13d3bbcc323edb05daaa96f2d445e2eca7f6
13424 F20101106_AABQET monroe_m_Page_163.QC.jpg
bd60424eb2706c7aff4fab1cceaa15d4
5594e034d3c0c39d58390805636c57d8f52ad471
3424 F20101106_AABPZN monroe_m_Page_078thm.jpg
06aec70ab0902df42247ed0eb381463b
71ecdb1618f1d8c790a243d2c394858adc909bdc
7520 F20101106_AABPYY monroe_m_Page_069thm.jpg
ceeb63481840f885a93e193268021cc3
099180954a3e823335555fa0c3fe98c44939024d
945497 F20101106_AABOWL monroe_m_Page_008.jp2
950738178ca249bd74b890db6caa0e40
4839848549422e3c2d72914db51534db2802f6cd
72973 F20101106_AABOVW monroe_m_Page_197.jpg
a2725d00fa34f844e832f4da1c0ff79c
ec9fc5e4896a78f7270851823fbc443a30a6b8b9
1051969 F20101106_AABPCF monroe_m_Page_198.jp2
6ffe7e643e6d68f843e5dcc8a940cb9c
6de38c0f2d5d2da9231c658638c49afbd216333c
544157 F20101106_AABPBR monroe_m_Page_178.jp2
042cd4ad373d8236170a9ea3415b81f4
455d0c5764c195deec2fa1e26eea77aab40f29c5
22844 F20101106_AABQFI monroe_m_Page_172.QC.jpg
60c2f4b7210cfba9f062d9d9eee8c2c7
4baa99379d889d3963f5749fb3d056d1a252087d
3972 F20101106_AABQEU monroe_m_Page_163thm.jpg
f6eaf07d28f17bc4ce86ceaeb9374529
ebd8df4c0a6b8fac7fc56ad463f14ad4e1c64f2a
22471 F20101106_AABPZO monroe_m_Page_079.QC.jpg
f11b25f4cbf5a7856ecba74e25421f42
fb98105e16f9b8942ac1d98eaa85f34acb732d44
26460 F20101106_AABPYZ monroe_m_Page_070.QC.jpg
9fa332baed81653a587e20f94fc7c7b7
06c2f4df4aa049f7c29fbddb155a57701430c890
F20101106_AABOWM monroe_m_Page_009.jp2
590be717159dd01ca02386ee1138ea5f
0b1e7dde6b29331759c6c75bec225be4deb45aeb
81656 F20101106_AABOVX monroe_m_Page_198.jpg
3cd51e38b65d8f21de4b780303c1d824
386c62a6ebbe121579cfdc56a0e2a9b17301ffa4
1051943 F20101106_AABPCG monroe_m_Page_200.jp2
1b54b301e6485f1a86ba0afe10b28cc9
7724e1630fe4e7d22676cb14dafb2b5ee8022a34
1051913 F20101106_AABOXA monroe_m_Page_027.jp2
a0f4962049110d4a2d234c286cf75b1d
88572d6eb133c9ece7f590b20e078e3c7130ef86
459371 F20101106_AABPBS monroe_m_Page_179.jp2
0a503cc55f894a0f1a98ca642ab390a0
726c85130a6544dc9ae519ac57455a7ff55664b4
3727 F20101106_AABQFJ monroe_m_Page_173thm.jpg
ff81604343333eeb70f0e1a435b2e447
9af5de41c54f9a7529e02b3466d7bfab2a181bf0
11662 F20101106_AABQEV monroe_m_Page_164.QC.jpg
42871df321a0261081aea6871f282c45
734117ebd904264b1989e825d2c7ede3e3eaceea
6067 F20101106_AABPZP monroe_m_Page_079thm.jpg
89d44ee5074f963928cc692d15d3e4e3
55a102f1b7ef62771659b15f7241e8f52999b401
1051973 F20101106_AABOWN monroe_m_Page_010.jp2
c4f3537fdc5080364ac6251c1099502a
de2370bf910eb36d0dfecd7cdd6fbe7be7353f7b
84083 F20101106_AABOVY monroe_m_Page_199.jpg
80a19ce4ad909a210162d38f531e1693
ad5dee45c187ca51ea3d5af77e6ec331afdc7b7e
1051915 F20101106_AABPCH monroe_m_Page_201.jp2
d2fa622f89352c3e83a43e9cec8d9bc7
5857994942df68f6d43f6eb5a5c52b18398af03b
1051978 F20101106_AABOXB monroe_m_Page_028.jp2
0592051f28fe46fab43db65712681aab
50ff294f0e6ffc11206fc183fb706e9629c62ea0
724643 F20101106_AABPBT monroe_m_Page_181.jp2
7d6e516322c0953eaae046de7188a61c
a0fb7b9b47380cef9307260f0fc186d2527ed757
8316 F20101106_AABQFK monroe_m_Page_174.QC.jpg
bcb8bde087ce0334fd3e2761ac3c33b6
790458125a4deec1e097ce05dbe5073677f9d0f4
3902 F20101106_AABQEW monroe_m_Page_164thm.jpg
bf278c56db902b0b1faa70e2ea6c82df
950816edfe2f4634c7c48feaa3244b10f62f3362
4738 F20101106_AABPZQ monroe_m_Page_080thm.jpg
2f1b407ef55aa3ee6fb762f1e439af62
4af54173d1999885c3bdccc488dde346e3f8242e
1051966 F20101106_AABOWO monroe_m_Page_011.jp2
10a395f68248fc811339e42994dd0cda
262de1fdb2d2f0f3ec371218e883b90a70fac2a9
84485 F20101106_AABOVZ monroe_m_Page_201.jpg
65abad692920cb64939791e16d9e09f6
ef2e9ed45d7de8a3d59df2d5015e59da51be4c04
1013291 F20101106_AABPCI monroe_m_Page_203.jp2
f0c7210d907586f5df01a01cabd9a73f
df385e4286944d9ec25300b73332613bce24cf2d
1051974 F20101106_AABOXC monroe_m_Page_029.jp2
4d2947bc8d1fd491921e9723db721cac
780b83a64040281f20271c0188576d0f5a2ef4fe
646466 F20101106_AABPBU monroe_m_Page_182.jp2
0f904411a1166cc10523e61112948ba3
ea4562ce3445a90606d883cf6fa055fa0153a9f4
12556 F20101106_AABQGA monroe_m_Page_186.QC.jpg
e91804404796a2b58b36804f71882437
d36dd61ebcab2ea9911e63321c1f54aab1007b8c
2705 F20101106_AABQFL monroe_m_Page_174thm.jpg
bd786b9fdc2bc6844ca1bcf64cce8a39
f3ac8c8ea6353225337c0649d666d5f0c1028b12
13283 F20101106_AABQEX monroe_m_Page_165.QC.jpg
d4aa46721187b39f9dc0b84176cd1f16
590279b5e80b4dc00708446b90d7945b74135595
17393 F20101106_AABPZR monroe_m_Page_081.QC.jpg
0df3349eb6c696190fe880b0a7720911
04c0d4d91fc87ed74f5ae682da865b196e03fe2e
F20101106_AABOWP monroe_m_Page_012.jp2
4aa150d41610d4f3519edcdb35feaa15
4f2d9ef633b052fb89111a61d4fa4409737aeaba
1051968 F20101106_AABPCJ monroe_m_Page_205.jp2
77845bf3081bbe8e03568c526033f66b
82dd56a09737b3ef75593285139c92ff82eb647b
F20101106_AABOXD monroe_m_Page_030.jp2
4a55c4db3e7f26a37ac1f650d27c3eb7
74c534e7bc2d730652c5aafd183d4086bd7d1b7f
563408 F20101106_AABPBV monroe_m_Page_183.jp2
aa5e65921085a389c972167d2605050f
09479a10ae96b7945e7f0e2a21ff6e3ca2a798b5
10909 F20101106_AABQFM monroe_m_Page_175.QC.jpg
77845c480586fcadd733f17962c61962
e68471d8fbd8cee0e176f3818b72d87f766724de
4399 F20101106_AABQEY monroe_m_Page_165thm.jpg
4ce98f26a563331b3b73981fd245c54d
4dc51bce9a5862725c8030f6166c18d7d0615fea
4849 F20101106_AABPZS monroe_m_Page_081thm.jpg
a08cf376c2258832253f53cccb95e27f
be417f55bafbe711b58d69e1ae4bfa5e3054889e
F20101106_AABOWQ monroe_m_Page_013.jp2
f57e1d714e652bb057fe1d61f2b889cc
d888a353ad18035234bf8b386611f289caccb2e3
293616 F20101106_AABPCK monroe_m_Page_206.jp2
70c62207ade462c059db4f719fba4c82
6a549e1745747507ca6e95c509675906f4ca514f
1014357 F20101106_AABOXE monroe_m_Page_031.jp2
c12da3aa0bcf563b859fdc0d49770a9a
3ba876afcd2fa8b872db59ad82bd8932d7d6ba03
648221 F20101106_AABPBW monroe_m_Page_184.jp2
5451ab8d5676c1619b0431992998e8df
f941e1a9696cfd6d990142e4d1f85d49a40b76a5
9816 F20101106_AABQGB monroe_m_Page_187.QC.jpg
1e323f7c359e34e04290c4ac70ff1537
6d8a8fb3e39b2ee08da42acd16a6b78a811bc94b
3742 F20101106_AABQFN monroe_m_Page_175thm.jpg
36dfd861c2df218e4a6405695397bf07
ea7370784ecbdd15c4b005f1ce69f6321297ea18
4242 F20101106_AABQEZ monroe_m_Page_166thm.jpg
b68f96ab91bfc5f2a1b3c0d44c67fd4d
4583a1000f13cd6ab9abf7a5e94dcb9040f904c3
17164 F20101106_AABPZT monroe_m_Page_082.QC.jpg
f5b55adbc236d1221ddbdf7e554e21a3
770948882c8f699c91da5d979983a67ca8322df5
327369 F20101106_AABOWR monroe_m_Page_014.jp2
10d7649055a1365e2c4df3a5ced0694b
ddc0cc0eda8ce4503a1ae77028326fd2ef017c9a
25271604 F20101106_AABPCL monroe_m_Page_001.tif
eb62159a010ea1f6262df9ad040aa638
4f8c7d8c4c5c160600f9754f367c334b948fc3ae
1051931 F20101106_AABOXF monroe_m_Page_032.jp2
85c7cd123ccb0a849e76809b902d1e9b
93a8263d8b507a9210b829ffb9689aa35a59acf8
321768 F20101106_AABPBX monroe_m_Page_185.jp2
c18c990bce60bcf09535268d8d10bc98
d082eae6821330862910a5c8bc4c24092901f305
10250 F20101106_AABQGC monroe_m_Page_188.QC.jpg
5f99bfd56788d4f4f4bf9c705b9e5a92
4cee6a8f0d77bb769e457c8b8094128d70df241d
16951 F20101106_AABQFO monroe_m_Page_176.QC.jpg
35fb3cab6b1ebe0ddd53590e31d54834
8ca0adc020466da3d74e4c63659bcc563771b257
5229 F20101106_AABPZU monroe_m_Page_082thm.jpg
cbcdabc57339b2cf3edb7da2f78b8d5a
a2b21cf1d1776c03fe5c3fcb7ec71069c2cbac3d
1005113 F20101106_AABOWS monroe_m_Page_015.jp2
cfa70e1e766a62cb3e5b96299e88a245
ca31a89071d3ac8874853a45fba70130856380a4
F20101106_AABPDA monroe_m_Page_017.tif
0da64d03acadc2e67fa00ca39eb930bc
4a7345c3f7d0f51013b2e1198de5134eb9ac5f7b
F20101106_AABPCM monroe_m_Page_002.tif
2223c0b3ef4eeb3a9ce8f3df8ca5fc39
a6dc57ae2349bb665cb29ac7ca018b11a01de851
F20101106_AABOXG monroe_m_Page_034.jp2
e77ce2c9988250a36dd5aa1afa184efb
7cd19c55f5ce5961109af36705078e76824f3f48
393715 F20101106_AABPBY monroe_m_Page_187.jp2
201a5d612ed658bc724d1f96375e0485
833ec1f446cb6a80aae14ce7a32214a951bfd2f0
12530 F20101106_AABQGD monroe_m_Page_191.QC.jpg
640fb0ba7c8a9bf0731f47d1d7e25031
392334e4941628798359947e4f4c3f9ca6ccba50
12009 F20101106_AABQFP monroe_m_Page_177.QC.jpg
8a214deb2fc43ec998e175f8fbaafb84
caff80b4c7001baee7d46c950ca73bcf96435a3e
F20101106_AABPDB monroe_m_Page_018.tif
ba565d6082c7209b1f0f2d22b106d9f4
855cdf5ac5ba0f24aa68efde11e2263f0890c18f
F20101106_AABPCN monroe_m_Page_003.tif
4ad2dcda470bd600d1be13b1e8179b7e
9d3314c295f043b3207a5e1afb85aef057ac84cf
1051985 F20101106_AABOXH monroe_m_Page_035.jp2
e8779aad6bf57bf0e0e34f6ce928a14a
4abe3aad29b74303fca05104a7003a6d5d0b5ca1
830978 F20101106_AABPBZ monroe_m_Page_190.jp2
510857dee27566c23ecf3c560a1a2526
0fd9be20b995ca53fc0c5dfb4ff244535ca26a0b
4200 F20101106_AABQGE monroe_m_Page_191thm.jpg
a3bea3f97ce19f6d1c9c436b5c0a6377
b3f9820931c6842456f347df9af5bb049a5af9a5
4216 F20101106_AABQFQ monroe_m_Page_177thm.jpg
38fbc44aff6e5edb8127a081b25c6864
db4fa7d947dc6818acf82a58b469a225238220f1
17923 F20101106_AABPZV monroe_m_Page_083.QC.jpg
d0b8ddcf3341d4962efe4f92435ea289
375cb860cb1e8d00fd77fb9a16fcf7ccee212fbe
F20101106_AABPDC monroe_m_Page_020.tif
16df6c5a377bfe5b914cd59c5ed67713
0ed19d5fdb58e571cffa02fac12b1faab2bb7c2d
F20101106_AABPCO monroe_m_Page_004.tif
d8ac7eca7031e3aafe7881e1d0a56490
90e154353a7a4a9e1306d5ff3bf373d3310c41c8
1051967 F20101106_AABOXI monroe_m_Page_036.jp2
2a0ec86e6e953cf12a0b136248a8be7b
b5e665babcb4322bd9cb0377aa88f778d3f312c3
358342 F20101106_AABOWT monroe_m_Page_016.jp2
2dda7d9ae7685bd2292cdca5e564b7c7
1ae01b3e9a9037edc03544ae8c0efb107af09f08
13616 F20101106_AABQGF monroe_m_Page_193.QC.jpg
c96ead01ae65b9e0521a3e0d5955d079
1eae3a0afbffd6af042a83969003f6cf3e67e620
9717 F20101106_AABQFR monroe_m_Page_178.QC.jpg
0ba69a0301ba600fd8f421ba090df485
015637acc021e15b959831089da7493167daa5b5
12496 F20101106_AABPZW monroe_m_Page_084.QC.jpg
31788e5a964f4a5eb96ebd620fa6b78b
61ecc3eafb89fe14a0e49b43862485b674921007
F20101106_AABPDD monroe_m_Page_021.tif
0c0273d3b358d73ad2cf0b0030690aa7
8e4a5962ff88536187dbe9c5b1789bda11936c21
F20101106_AABPCP monroe_m_Page_005.tif
120d0658d6a2f60e25f4dbb8990d9df9
a2cd81f7e242e8c588b9a18c3541d86d9f3330e5
1051979 F20101106_AABOXJ monroe_m_Page_040.jp2
b51b26a92938db2ecf033af9f2e888a0
746600d8a4e41546e76034e731633dc0ccd1f428
F20101106_AABOWU monroe_m_Page_018.jp2
98f599b32578158830326ab8ae0e98ff
7158299389861de67db0dc855270551a70af8fab
4396 F20101106_AABQGG monroe_m_Page_193thm.jpg
9bb3ff15b126cba24f27b987d88d709a
523fc7eca9c9c7dcd4d03df0a46524e0414837ce
2898 F20101106_AABQFS monroe_m_Page_178thm.jpg
a0aef777cb26e22c849494a16ca2fce1
661acf30217867d7600842d63dcab5e40b8aad59
5301 F20101106_AABPZX monroe_m_Page_085thm.jpg
579d6cf39fdcb1c820be05ade4699c34
86cb84eb9a7c8edfe1541d265c24b1c3dcda97f9
F20101106_AABPDE monroe_m_Page_022.tif
aebe0ac14ef3b1f83fdca1b4e2223c35
0a9bfea5935336232a6238e1808efca64d233577
F20101106_AABPCQ monroe_m_Page_006.tif
d253a77571df363a8efef18fd8486ddc
2433b89720869fdc82ac5d28cd6f9d9f47c497c9
1051952 F20101106_AABOXK monroe_m_Page_041.jp2
4d0d01f65eade24b841b4f9cc8b55f96
85ace7d8d92903c88ff86edcf10fc98a254630ed
1051944 F20101106_AABOWV monroe_m_Page_019.jp2
76bff859ff45410b971c69218055496e
974f78f57ed000984166b5207e2c3714635acf08
9476 F20101106_AABQGH monroe_m_Page_194.QC.jpg
1b29041f298b051e4e0637ae7ae3f235
980d835491db6248a35bff4796ac548f03f3836a
11719 F20101106_AABQFT monroe_m_Page_179.QC.jpg
76eb78108379bd01c628e5b1b6c6f8d5
28ea73a358b57819d0e1222390bb171987382c27
10484 F20101106_AABPZY monroe_m_Page_086.QC.jpg
906dab4effa0b2ca149b3ba67ed27de3
ca80443326f3418839543f2926a2dcea3f8c390d
F20101106_AABPDF monroe_m_Page_023.tif
13a52a614f994edce71c1caf152145f0
ccff824b48dd61b874948e370a2be60090296423
F20101106_AABPCR monroe_m_Page_007.tif
6209966833e0f6a231c9a209848a206e
c908076dc64b0d1743e3c844c384fd99fa6d3978
1051953 F20101106_AABOXL monroe_m_Page_042.jp2
2d0e1474daa1ead8b53745040198787c
5b7d77ddae71743b12001604b6a0863924c93a22
1051959 F20101106_AABOWW monroe_m_Page_020.jp2
a91550417592fb15ff21e034fca904e4
1294442c68a3cb7f9c3b0a5674b0268b312e139d
3242 F20101106_AABQGI monroe_m_Page_194thm.jpg
88e0e4dd10dcdc55c8b60a4595a440b7
6a921117459de159fcbe5c13207afa94171ca5ca
3555 F20101106_AABQFU monroe_m_Page_179thm.jpg
01dc2b41fe535c5b7563937dc354f09a
a022036e5650aa325eeac9c7e6d6f50272427d45
10961 F20101106_AABPZZ monroe_m_Page_087.QC.jpg
0f15adb8ce4333896262bf1794029180
de06e53f005d2998de83e2a2e0bb0593347afd0c
F20101106_AABPDG monroe_m_Page_024.tif
3a989824e7f5c53076e7f06956e45b55
b3894431b19958c8cf2fe14163496c637329bd20
F20101106_AABOYA monroe_m_Page_061.jp2
2aa511bab4dae2eddabb58569510d617
5d5e2389c699dd30be015fbfb13270781909808b
F20101106_AABPCS monroe_m_Page_008.tif
9da2636e83ccdcc7fc2f1816a83b58f4
8b6eb4cd117c19960fbc3055e7a9aab306df0d05
F20101106_AABOXM monroe_m_Page_043.jp2
a22bc13dca5f66991ce6bac6c96fbd7e
5425c51b7534d4734ff26c4158ec279cbce08d04
1051970 F20101106_AABOWX monroe_m_Page_021.jp2
9c0e567db4ff18416ac261a712ca1c71
5f1927f07068d85d3ec08be45b0270bcffbefdae
23789 F20101106_AABQGJ monroe_m_Page_195.QC.jpg
1d1f72ced7a85806562de3153da1c28c
925ce2ea11e10f5010abeccdbcf9b35dd7f1e63d
14809 F20101106_AABQFV monroe_m_Page_180.QC.jpg
02f00edd037353f9ed45f89604254a96
bdf3eaa0921f1673a898cb48f860e1a58bc1001d
F20101106_AABPDH monroe_m_Page_025.tif
19d3e2240acf971721d59b8538928fc2
28ab51dde1ea49b2aa94dc93189853ab1fbc2e00
F20101106_AABOYB monroe_m_Page_062.jp2
b2216b63f3b00b2a65986619901ba149
5976aacc2dd95f29d44b9a1dc5cf63f42588a178
F20101106_AABPCT monroe_m_Page_009.tif
2f712d8da0ca55baa61878c553fda985
a0fd3a06682a52fa84552198899fe5e942bb28da
F20101106_AABOXN monroe_m_Page_044.jp2
4efc7cf65cbe6ad5c09e3b34ac842675
6e722f543adbec07286b2412c9601fe0f95f68c1
1051950 F20101106_AABOWY monroe_m_Page_023.jp2
68b4e4794693d4a817872dfff653853f
c907ebd6d46f46ca28d4940bf970145678fadeef
6659 F20101106_AABQGK monroe_m_Page_195thm.jpg
7142af7d644afb19466e935ae2a3094a
6e5c566a28d244fedc3d7fbdeff4a0d4a2383728
14463 F20101106_AABQFW monroe_m_Page_181.QC.jpg
d11fdaf2334b6d87be4010fd49e7edee
b3d86d956227579ca16f430f13acb82fd954c02c
F20101106_AABPDI monroe_m_Page_026.tif
3468c8f1d1db72e515cecac55065f795
ec6a45b909ac5869bbef04b99cf1a697e0696d89
F20101106_AABOYC monroe_m_Page_063.jp2
6013a267fe6219776cd6084f85aaec75
12747f227f4cd23dfa352bfcdab34587170939e9
F20101106_AABPCU monroe_m_Page_010.tif
ab50b0aa87f55ace2375710f7a564133
36b967731077bf8d9cf92654106bed6839a0a56c
1051930 F20101106_AABOXO monroe_m_Page_045.jp2
3fead08d1506b642d780b12ce62a5026
804c0e88d7828c073803230268ae91ac53eaf408
519951 F20101106_AABOWZ monroe_m_Page_026.jp2
df2aac91cff980f4e28ad015f6350f78
82517be793df988aa5ce09ee75eed7ecabee5784
6057 F20101106_AABQHA monroe_m_Page_205thm.jpg
5dc5b813522787e0dad608c08ff3a725
639768151f2632af1c8152678820e5968deef23d
25883 F20101106_AABQGL monroe_m_Page_196.QC.jpg
fabf1de26617e24eb2a77e291e714507
70d08d4fa0c3523d57246d28cb8ac6a760ab38c8
12052 F20101106_AABQFX monroe_m_Page_184.QC.jpg
f7122878c4fb13ed8a35243c72b0f138
bb2176f9bc14cc7e647281f20082df4f4256f93c
F20101106_AABPDJ monroe_m_Page_027.tif
b3e699bc8ef9331df255d3ac25bff6ad
3bc69dd1e1da03c7c67d1d635ab2ed284570f73a
1051956 F20101106_AABOYD monroe_m_Page_064.jp2
6d72250923682b15510eca37b0ec45f8
4c7751534e98732215fa1eb83ab82ac57391c063
F20101106_AABPCV monroe_m_Page_011.tif
22861561dab7cb89f82d9049b9ba89d4
049712e4948a0e2876e19d170048f0a059d1f309
F20101106_AABOXP monroe_m_Page_046.jp2
4c1f079fbc0b9f105560aa27f12cba6f
5e7f8e5edded9e3d3311cfe2055472c0ba3d5d1a
8478 F20101106_AABQHB monroe_m_Page_206.QC.jpg
bfae287b45cbb4b9678e240305e269fa
af43ef1e08635261203776a7ad190fa30bd93aa2
23575 F20101106_AABQGM monroe_m_Page_197.QC.jpg
f0ec2a04813da2e4e9fba4cdd03bef2a
894c5b266e05ff816a6c7491e592e03cab2b5b99
8968 F20101106_AABQFY monroe_m_Page_185.QC.jpg
9b1d0e4a7e1dca754cbda06e8e95ea31
fb260c3baab940c7ad8a7dcec06e4259d9e890da
F20101106_AABPDK monroe_m_Page_028.tif
6df38e60e6afa980c943df1bf3380e91
c75923b32e5dbec37997e849a698c4af04dae02c
F20101106_AABOYE monroe_m_Page_065.jp2
301ece290b96faa0030ec7aa864afdab
4a5afdb98d99b4ea90401c2746ccd529d5da7b72
F20101106_AABPCW monroe_m_Page_012.tif
7251c6be75f89a1dd3c5b6a51ba3e354
f3fd37b6a49c9f7fe410900bee2ddd158602abb7
1051927 F20101106_AABOXQ monroe_m_Page_047.jp2
43798f7be342185288ce6d21658591a9
18ad02b8aa6850be65f9d309e6a0fa30dc5e4b7b
6453 F20101106_AABQGN monroe_m_Page_197thm.jpg
d8dbc8133527e6f171745ce0521ec87f
7ba1c25ce8086fa83f8d1dd5aaa95326e6896468
3067 F20101106_AABQFZ monroe_m_Page_185thm.jpg
d23dd079826ff7a9686a04096ac9f576
bec78058c2c2c3b4f2be73a14dbc3ee4e781abe1
F20101106_AABPDL monroe_m_Page_029.tif
94ee6563faeb1d9725a048cb0e262b40
947e353dabc00b2d7b64d05531329296a61647f0
1051965 F20101106_AABOYF monroe_m_Page_067.jp2
e7f86f0d94f384157afa328b7aa7c912
6e1eec88a215e9fac0949a51bc25059fdce96150
F20101106_AABPCX monroe_m_Page_013.tif
a0bc82ac8090f59f209bd5a27b0e8826
71c2aa662ba3b3377f7cf8f8318030d40f0437e0
F20101106_AABOXR monroe_m_Page_048.jp2
4c2c1e65b52fff53747d4d6f2530b6ae
ddff003aed9f433168cfe8df798fe53a5b8d1772
236744 F20101106_AABQHC UFE0022458_00001.mets FULL
b64761bc37f3677f0ac2603c4a24d90d
4094e9256ce92f8b0daf1baf05b28ba8317c3448
25854 F20101106_AABQGO monroe_m_Page_198.QC.jpg
aa331e95b654ebdb0514b1096108c6bd
c37a4d3aabfd72ce33a31689ee16ff667539ef64
F20101106_AABPDM monroe_m_Page_030.tif
d0c8eb35e3d15518aa209c377ec4f619
7647733fef29a067f91899b020c06aa9dd133a18
1051980 F20101106_AABOYG monroe_m_Page_068.jp2
2d1fea33906a8192d146c6647bdbcd0e
fc8fb43dfae0d3efe7dba2f26d6f307c9ef075e1
F20101106_AABPCY monroe_m_Page_014.tif
2781094282aec925d6668ea0f704d5c1
ad1f8a28b66dd8d005f8902831ff44ffbcbb3900
F20101106_AABOXS monroe_m_Page_050.jp2
0c694c70ac6e774b57c3030ad7ff9f7d
62b9860e1bf31c5a3bebc02edc3786e3b9d9e4e4
F20101106_AABPEA monroe_m_Page_050.tif
2b673a5145aa75cd54d0f3cda7dc9064
5a3c71495ce5b9f55b30528465b28bcf4809538e
7243 F20101106_AABQGP monroe_m_Page_198thm.jpg
00bc4b5ae520c0d2fa89a709ab2fc5f2
5d3d4943d1cd233c97346ceb616ff7c80da54f8b
F20101106_AABPDN monroe_m_Page_031.tif
c6bd72f105bae88a45e90d74c2b4b5d8
801a2c718db8efd7e6471b867072bddb3edd35cf
1051865 F20101106_AABOYH monroe_m_Page_069.jp2
7e0bb7acfb262d737ff0b37a0fd126dc
631c2b434c545346bc7af2f9d50f200ea5afc692
F20101106_AABPCZ monroe_m_Page_015.tif
c1b22ad4933f43449475eb2a80aec54f
033814ace4be067b3a70faca4d455c15437f4444
1051926 F20101106_AABOXT monroe_m_Page_051.jp2
60a736a298eb71eaad79d66894bfbf20
34c9fc68be132357e479ddfba7f71bd2eb54519a
F20101106_AABPEB monroe_m_Page_052.tif
3db5d277d101d7d3f0d0f4ab7f798b06
9913b9a19869710795cd1fabd84e7c0df4fd10f4
26538 F20101106_AABQGQ monroe_m_Page_199.QC.jpg
48d06cff5bb8f80789c3a74fad566696
a921c60d79a45017bbab7aa98a896d56bf2dded2
F20101106_AABPDO monroe_m_Page_032.tif
1369fdce1725982cbe321962e2198f4d
5f4423e191b67d2da9b0d1f2e0fe6a6021660d5c
1051975 F20101106_AABOYI monroe_m_Page_071.jp2
69ef13e769024d8c5285868b7f24e10e
4ec726236b1dec64e7cc299c43b6ef8f4f7f64fd
F20101106_AABPEC monroe_m_Page_053.tif
cb991c1e44a0b74f766b95835ec7957b
b187159520262278e3788497885c11273494cb9a
7296 F20101106_AABQGR monroe_m_Page_199thm.jpg
1d839b230208b2faa1d5e6f3c7d2a4f6
2848678e2462c8407863d9520cba6f130459bf07
F20101106_AABPDP monroe_m_Page_037.tif
1ed0a12bf6d42d89293da56aea298b3c
3387a2e206d08473e3b38c10b4d5a3e138515b50
F20101106_AABOYJ monroe_m_Page_072.jp2
dca6dd98af81fe32aef7ce4196f02c0f
03b22c0fbd7ef920de8810d0f9fd65e94a1be560
F20101106_AABOXU monroe_m_Page_052.jp2
74834775ad8133854726737cf18f0886
598a9bdeb348a3f1dfdb09532951109eec461c44
F20101106_AABPED monroe_m_Page_055.tif
63b4820fd31dca19fe3dc718ff1a6124
0ce97b2eaa52c263755aaec1bf4119739bc9428e
28250 F20101106_AABQGS monroe_m_Page_200.QC.jpg
ebd8216689388904729a58903e6daa8f
da32a9bc062973b527f2efb358ee7c9cf090f48f
1051949 F20101106_AABOYK monroe_m_Page_073.jp2
441633195cfad41181e9de6af3c704ff
560294846f2e0979b0ee51f9ff925f93ce85dcbc
F20101106_AABOXV monroe_m_Page_055.jp2
805010503ea3c6e6aec61dad95a79361
659d3f411408857b06c8c2689f0bb18457397af7
F20101106_AABPEE monroe_m_Page_056.tif
682ac8ecc94af620c273dcb6121c7540
4bd3eb3fd90985350dcbc872560210c2064d853d
F20101106_AABPDQ monroe_m_Page_038.tif
f77dcfc977ed3a4687a3cc0cae74bf87
68f1d800f9573c4f49d89f09e80e39e9237b68d5
7538 F20101106_AABQGT monroe_m_Page_200thm.jpg
90488187cd48ab2379af1b44ce1c4a3a
707c4766efaa61e9de27ffdab9becb59840c4528
1051961 F20101106_AABOYL monroe_m_Page_074.jp2
61f7013c053b0aa7779cc97d935594e3
15476dc3b21e1f602b8bb3238e89c1e3bc22779a
F20101106_AABOXW monroe_m_Page_057.jp2
a54895318bd81c22dc3d3b054f76ade8
72d66b49b9865f941ff806706baf9ac6c91c8463
F20101106_AABPEF monroe_m_Page_057.tif
dcc2e0bd042ae39267692694468f03bc
caf1ad48bd373deb14e7f403047bd8fbbd11c77f
F20101106_AABPDR monroe_m_Page_039.tif
43e988f49552d9e336ca1cd323e336f1
21c2ffb05905821a5d0890f3be2dc3826c515387
7126 F20101106_AABQGU monroe_m_Page_201thm.jpg
a572d5e7dd08c42f6476b3b78af0c92f
37db57cb1b21b74dddaa266390f0c1acc2e58650
621338 F20101106_AABOYM monroe_m_Page_076.jp2
3f8aa98b1bd7af3b37a508f75f916ce1
ca5d56cec7c2535fc7ca9e5f7c5efad961cebf85
1051948 F20101106_AABOXX monroe_m_Page_058.jp2
67a3de61a73645803c162fb2a8bece44
0a8fde39490c4fb730a765b672451b883d9ca483
F20101106_AABPEG monroe_m_Page_058.tif
892599ea726ca4eacc233949861416ef
50334ff8ed4955fcff38ef551c8a95b87591cdae
692245 F20101106_AABOZA monroe_m_Page_093.jp2
f795f460f95b04e9778bec4a950b5721
7a3533654fabbd8e02ee522f3da35633e24c6b4a
F20101106_AABPDS monroe_m_Page_041.tif
393aeb0e15bd397cc4ff50ab3326c240
5c2dc987c04b7c97f83838f0f25acb0a5c44940c
25113 F20101106_AABQGV monroe_m_Page_202.QC.jpg
7cdadaa0cb3b4e2a213e0713274d9690
105db6d110a16a55177676f5b1ee2949a37c4dd4
1000280 F20101106_AABOYN monroe_m_Page_079.jp2
570e5475c99dc0fb4b5e543e03105c9e
553daf0cb9e807670ad282aa78b650540c4d218b
1051971 F20101106_AABOXY monroe_m_Page_059.jp2
3017899dcdece0b6f876e8e1bf2a38ea
86052e368525bf8f86527420a4130cbe18c477d9
F20101106_AABPEH monroe_m_Page_059.tif
40c3134fed57498040a50a2bc011bbd9
d85a2321372b481a251ce5b2e31c18fd68d1c3f0
685538 F20101106_AABOZB monroe_m_Page_094.jp2
205fe0edde115bb17c4d115126f920f2
f8f87b17791f88db442d025f3f58f52b46ab6de7
F20101106_AABPDT monroe_m_Page_042.tif
cd4ffd8cac6d6697460af0f7f486bd81
104209746cd0f3d8b78c95f6b4545ac6995bbfd5
7189 F20101106_AABQGW monroe_m_Page_202thm.jpg
4d4cf5dfae97c32dc0431e7df198b255
edc57e041ccc20f7d932c7b45e1e5de875baecb1
824261 F20101106_AABOYO monroe_m_Page_080.jp2
019f8ff9ab5df00c2bcfb90ab802313e
eb64e9bbff6086c5a0467cba62d1f0e46f660bef
1051928 F20101106_AABOXZ monroe_m_Page_060.jp2
30e8312b8971d76712201e88558a0e9d
a4831b6d6b7cc90d84c83ec736eb7cd78fb58183
F20101106_AABPEI monroe_m_Page_060.tif
3a8db042c73e552299919626f4462928
9a09d302db767efaffe37362a58c3538c60aec2c
501807 F20101106_AABOZC monroe_m_Page_095.jp2
8256aeebdb4ae77e28697637bb269aaf
e4a2d646cb210a067b9f49010300cbd4c1f5c72f
F20101106_AABPDU monroe_m_Page_044.tif
c0c5f5bcf1b30b6d9de173d42fc78fda
66e6cf30368f05190590d9dc5cc1b9097e6827e1
6624 F20101106_AABQGX monroe_m_Page_203thm.jpg
2e863b3d1bf9d1432e2e5b44f3dfda11
e441cf243c5ccf9fcb8f77293d67959cee6f3ef3
973225 F20101106_AABOYP monroe_m_Page_081.jp2
fb556be77fd4ab0854be3548298f9f02
65173388be4912cfc51c67aea55253b63e5126d2
F20101106_AABPEJ monroe_m_Page_061.tif
75a00954429a92695e8accb2d8eea65a
2174f48c77eab79c83aa751da57e37cec1e39960
608963 F20101106_AABOZD monroe_m_Page_096.jp2
8c963818545af48a1155c6e21277c812
2ed4d977c12430c3331dc28ad2b736ea2f3167cd
F20101106_AABPDV monroe_m_Page_045.tif
7bed528fa2d69a98e5b283d232cb56bc
ba1a102bf2e5d8b1dee022730f1fab7ad4eaaf25
7074 F20101106_AABQGY monroe_m_Page_204thm.jpg
3dabcafc5eecd4b1ef47633c84581b35
730badd61b4b32c4ae927dee962b533a01a86066
954317 F20101106_AABOYQ monroe_m_Page_082.jp2
438dbb18c8cf3c61d9b905a257eac39a
713601074ab4cc000a46c984a0f167cb5c02740c
F20101106_AABPEK monroe_m_Page_062.tif
03e0148fd62bb17f5209a0ee3e57196e
59f056c0f8214450e84f27d6ca6ae2ed82f6ca1a
596324 F20101106_AABOZE monroe_m_Page_097.jp2
531711c98d32be05f27907c02c057c2e
4ade1f32c887e53934e7e4c235018ba5cabe3442
F20101106_AABPDW monroe_m_Page_046.tif
66e07e8cee7dc61776960b5d1ed9437e
fa3e463e535fb1cdfda622fa210b9414f666ccb6
21712 F20101106_AABQGZ monroe_m_Page_205.QC.jpg
4bc2462d178dcddd3245f39521bd3d96
d95d5228567b3ca12235b337b7414d178c489954
F20101106_AABOYR monroe_m_Page_083.jp2
0faf9a2e74d3a998161a211e73b720ae
2cbb3a2ba0e53cf56b394e2f38681ef7a7ecb7c0
F20101106_AABPFA monroe_m_Page_083.tif
a12f1a1d81befac0166be6735946927c
6f0da3aaa55fdf9bf09db5fab86dd62534351688
F20101106_AABPEL monroe_m_Page_063.tif
1b9329518d2e5c5ef886742390149fb2
8db75bb80a6e2e02ff30b5717748bc1daf4775d4
933220 F20101106_AABOZF monroe_m_Page_099.jp2
b23ff41e3f277f0475e7a66d3bd62c47
99c51ec4dffb91ca48a4be1f1db6e257b06033a6
F20101106_AABPDX monroe_m_Page_047.tif
f6ad6e414f653cae158ab32676f3fd01
c69d15e3ab25db806a556e7c36ae9c39ff46392d
624362 F20101106_AABOYS monroe_m_Page_084.jp2
5fd24526b4a5e5c555222854da045aca
877f2ef056568920af92e7e6c6411460a11e47b1
F20101106_AABPEM monroe_m_Page_064.tif
d1dcaac85758b1964fd0620b30ad18bf
d3ec06f7f7bd0cac46576950d5bb961a3e1696ec
560406 F20101106_AABOZG monroe_m_Page_100.jp2
a7355f183c04a92f019c17c576fc77d4
9b251aa4f1a1fdf03f0ac52711f50c3a057abe9b
F20101106_AABPDY monroe_m_Page_048.tif
57bdd90237df727a0a96a748e9c5f3bb
423836a0b68d5406bedcf01d24e2598f92024d91
699860 F20101106_AABOYT monroe_m_Page_085.jp2
5708b2634265e04fc69bfd5de4a45db9
e81a0b572551d2aab7e1efad267406e0cd5003eb
F20101106_AABPFB monroe_m_Page_086.tif
ac9ee81a7485212842275181e99ced58
d181f630753b2e8a5ca12d6e251c4cb243601fb3
F20101106_AABPEN monroe_m_Page_065.tif
a7d983e275469d3119e7f2fa5e946ad8
18f88ff209916ce386a6af63f209d9ddf4e0f7cf
746790 F20101106_AABOZH monroe_m_Page_102.jp2
71f533d75b8867147587d578a0084290
8c1dfe242eeb93d2c91023c00d7ae27f126f6bc5
F20101106_AABPDZ monroe_m_Page_049.tif
37176fe0093e4cd217a90090ed53c6be
7e8e19f51bab70860ee45e49449b12a62c3b0b6e
373454 F20101106_AABOYU monroe_m_Page_086.jp2
1f4137e9d072f5ab188dbf2409c9ca48
e8a63898e334856f1b93447737fefac5673f0c39
F20101106_AABPFC monroe_m_Page_087.tif
abd2b9193646cb5ac21ae757516597ed
903775c8a4c389bde2f47b3b5b3d2fb6c931dea9
F20101106_AABPEO monroe_m_Page_068.tif
20c049441d5da976f80114aff3098fca
b0ca1cfe0c7db6fbc5fe3476fd57a2a2a26ed386
740013 F20101106_AABOZI monroe_m_Page_103.jp2
e228b9386e03c5c6bd4b452f7029b27b
d5556af280cab8c1366789b24773b42368e14f58
F20101106_AABPFD monroe_m_Page_088.tif
f6d2166db6df214654ad1db9ff0dbc63
0d1d926e5a9ca309dfdf758a3482956b4b7314e5
F20101106_AABPEP monroe_m_Page_070.tif
62f3b72881edca5f3a839f60962d2a1b
d7359b9932d4c9b3ef8605ede954c19a739809a9
701777 F20101106_AABOZJ monroe_m_Page_104.jp2
d108fa389914586528d4baa75ef790be
84d5c7747abcb6d10f1d2333359ab9fbcd12cc8e
381503 F20101106_AABOYV monroe_m_Page_087.jp2
13050a459edead6b9113ac81cd73e5c8
142165a8105dd7bd0b57ef8c4c09b9941dc87da4
F20101106_AABPFE monroe_m_Page_089.tif
fe776edf960a7b36884d0dfe9cc13f06
88dc33af241d46d482209357311ba5de2a776fee
F20101106_AABPEQ monroe_m_Page_071.tif
423f0e1b57b00dc6c130645d060e0207
2b5805e07f839e7ed3475c0c1f16e9af739b8fb9
723808 F20101106_AABOZK monroe_m_Page_105.jp2
91a4918643ed437a4e7d12c0c48553d5
ce74d421a91644b6ca8e5180ec2f4f1119b3a7a0
543616 F20101106_AABOYW monroe_m_Page_088.jp2
c1eedb30120d18668453df405d64c85f
4009232a79b7cca9780ef82f858698dc344184eb
F20101106_AABPFF monroe_m_Page_090.tif
2f5fa4c256cd35b8b014bdc46a4b2296
0e6b935ec97e99460adf36f9e37c2622dba2180b
F20101106_AABPER monroe_m_Page_073.tif
9b320f830fbd346c6d102370ff621819
548a610abc321b2352aa512a2fe78801ce4d433e
930473 F20101106_AABOZL monroe_m_Page_106.jp2
0ef17b794b86fb5d4c09ad5d71b1ad16
1146cdc16b7a9e74fa9b6ddd91c0a94ea88b37af
304967 F20101106_AABOYX monroe_m_Page_089.jp2
db8c2c1fdffa7c2cfaab84bab3d03dc1
76e7e646301d42050637c5ae28edc454f119b4cd
F20101106_AABPFG monroe_m_Page_092.tif
ffd4c0f38d773b35ed7a45cc840233b0
f097fca38b7dea98dfb6772dbdc653a439e612bd
F20101106_AABPES monroe_m_Page_074.tif
78bc74186657e7a2bf89de45f8a71148
f1ee2642b2ea6881b9185277eeb032e6a46f3e1f
581853 F20101106_AABOZM monroe_m_Page_107.jp2
a5310430728109e52b59843eb9710103
afc221bc8dc9aa350fb1b56d55ce05f1446ad50b
653548 F20101106_AABOYY monroe_m_Page_090.jp2
cae6eca8a7231b313a183caa9f3debc0
895a1183c3ca9e1e63ad8b80c04ae134896ba992
F20101106_AABPFH monroe_m_Page_093.tif
df926ddfe108d6bcfea5c2c0260c9bf7
5ff2df5d8a0063500bebd3e3f3a936535dea126f
F20101106_AABPET monroe_m_Page_075.tif
86c28c3a36544fc286edd6dd03e94b46
1b566369e229f74df23febf5aa1b6dcbb7d6c936
669375 F20101106_AABOZN monroe_m_Page_109.jp2
78a4472c310c42d3bc30a7d75f5ef68a
49acdea98c270808435b29df4649e4382fc4ae27
495037 F20101106_AABOYZ monroe_m_Page_092.jp2
644dfee95f48928bd943c3e1c735d8af
917be1381e3b3d8093882f4e0a20ffad8e9753df
F20101106_AABPFI monroe_m_Page_094.tif
b7e05226160d249c9c2a0f5fd465b9fa
8df8c3efe7db42fe1b2398abd8ec4f1be28cd027
F20101106_AABPEU monroe_m_Page_076.tif
56bf4b684e70450fbcf424d5e3ccd4be
0f44a64f2fc0bce8ddc9225440a240c3db73ac2f
970534 F20101106_AABOZO monroe_m_Page_110.jp2
1a592c5cfbf2f8b572a895d11de7794d
bf7dff62300b9bce81f6518fcc7aeaa8dc5ab67c
F20101106_AABPFJ monroe_m_Page_095.tif
00cbd68406f917cd7ed0b9ffa51693b2
362c4844899299c659424faf86905283d4ba540c
F20101106_AABPEV monroe_m_Page_077.tif
377aefeadaadcd0aba58d0d989066f41
9083157cd26059d5458b6f96ea5f91c583cd8c54
1051934 F20101106_AABOZP monroe_m_Page_111.jp2
3cfc8c46b42f704ffa239ca4a8a167c7
e60ffea1cad2c5dd51c26d01686dbfb30a5db2bc
F20101106_AABPFK monroe_m_Page_096.tif
391d66bd8f7126aebbb350f8076f51ee
57caf9417b410df8cff94dba26492430d01fdfa8
F20101106_AABPEW monroe_m_Page_078.tif
8b2fb3b446acc7199449a9fa828eca1e
7b91a9c3e5fb8215bd163160a5fb94cdae66e852
280973 F20101106_AABOZQ monroe_m_Page_112.jp2
d8a18ad2a25bb2eaedf78925fe914d78
b5578bd38a6b6ca8f00131bf904121bc78639593
F20101106_AABPFL monroe_m_Page_097.tif
7ec4d9b0ecfffaa93b34edb8ca1a897e
2ad5e4e3cbc24d742b56502c7ffdef650bf74eab
F20101106_AABPEX monroe_m_Page_079.tif
310c9441235d3f69a01992ce0e07b8e7
b4c34985f563bca351966dc1487c5487bdcbfdf2
377394 F20101106_AABOZR monroe_m_Page_113.jp2
e4f815615d0cd0104eaab69b28715fb6
819b8d8eba84017b0af83450b01219f265ec8fa7
F20101106_AABPGA monroe_m_Page_116.tif
a3abc33937109ce7a3879ff0c5938fd6
cb9e624a0e1e38740185451236932f5845b57dae
F20101106_AABPFM monroe_m_Page_098.tif
d8c91f9919eb1a3aa2bfbece5aade010
364a9452436aac2f1b9ffd7ffcc5c3161bcaf238
F20101106_AABPEY monroe_m_Page_080.tif
b74af2c4fd71aede7d1d6f5368edad6c
19c6c78081fb1f449836544a509b60360a487fd5
264349 F20101106_AABOZS monroe_m_Page_114.jp2
7e267ddeec4e5894a8bbe911020305b3
94f962c9044ed862c07564567e4107637033ac62
F20101106_AABPGB monroe_m_Page_117.tif
2d8db76e1b061ab98ad4d8fe6526b8d7
bb546df6db477a3c62694c1ecdbe42ccc2022d04
F20101106_AABPFN monroe_m_Page_100.tif
b6145c722dbe16aa5eed47f5356bc7ba
7eb96097810c1dd4169bde68078784d29d32157e
F20101106_AABPEZ monroe_m_Page_081.tif
59289a76264be34fb4f7a67ac3677e9a
e0808645ac9e0739302cabc6c60a2b1c8240d120
765432 F20101106_AABOZT monroe_m_Page_115.jp2
ba2b9dc18c2fe995725d27aa8453b64d
b365efe650f7f44e869c870f10baed4b55c957e1
F20101106_AABPFO monroe_m_Page_101.tif
1a112d0fbb618a1c49d28b422df7fb11
b2d06e564e8c041381a4342024ac9eb2d6476133
996850 F20101106_AABOZU monroe_m_Page_116.jp2
3b77c1a4991b6bb97d7108701bc32622
abdd31e9321a4e3cb4efab8643ccd286cec863e7
F20101106_AABPGC monroe_m_Page_118.tif
a941014229a963d3cad4f3ead4dd7f65
88835de31238ef70a6f08bed364e7fedf0690a27
F20101106_AABPFP monroe_m_Page_102.tif
cf4e28d549b2291f9e350dc85deca242
12c4b4e352b2df0e63e9a1520e79552560596a65
643984 F20101106_AABOZV monroe_m_Page_117.jp2
c5f7e0873223af9d09b1ba6a4222b70d
0b360aeaa169cda7f8440d176814f8d0bc2aeab0
F20101106_AABPGD monroe_m_Page_119.tif
8f19b3f44dd54658d3baabac31f3a73f
66e730f1c5a716a3f152856ccff201fcf42ba7eb
F20101106_AABPFQ monroe_m_Page_103.tif
067bd0864b251b65dd8c5fdb8d6cded7
5446b77038aa90935b7654838b04acef4ac4020e
F20101106_AABPGE monroe_m_Page_120.tif
64b40c928acce7678e9d3600ffe2e700
b78905ad0a1fe78bbea5a8d149e746f210af87f2
F20101106_AABPFR monroe_m_Page_104.tif
48980ca1ec7b4b0d01234467fd04e6d1
afbc809832c3fe06f2aa4b1b22766dd2744496c0
F20101106_AABOZW monroe_m_Page_119.jp2
58cf5df67db319bc08e34ae118f59a87
87d368250ba7e2531ab8c0c5490efd732b2f3f5f
F20101106_AABPGF monroe_m_Page_121.tif
7a9acd0b681600abd673697d2a7d6cb6
11291cdf11fdce8b1a3ffdf7e16dfddd44e010e3
F20101106_AABPFS monroe_m_Page_106.tif
0309578f3453b9e59a77ded9960ef6b4
128bf91ca19ba7ef0b7be757c0a6e45f0646eb74
519800 F20101106_AABOZX monroe_m_Page_121.jp2
842845ec99fe7166d99fb0a877ba6727
82923f169829c306bbc97063d72fbe0deb0b509f
F20101106_AABPGG monroe_m_Page_122.tif
8aef6ce8f74e86b45c84ad1c6425134c
8ba4dc9bb714471dcc3781d8a5a30c2b63d8af1e
F20101106_AABPFT monroe_m_Page_107.tif
8f55cc4d974d2336aa1bc858a3d6bcbe
ff9a3a8bcd640c6e3755a6d6acda8b14641ddee2
F20101106_AABOZY monroe_m_Page_122.jp2
9c7fb17caf4978852919264147be4fe7
e1a355e9788ed7a181b4cab890705468d0be6d15
F20101106_AABPGH monroe_m_Page_123.tif
0a32328912bfdd3836706b28dad5a639
0637fedf06bdf41a1397e27fb3c468b38373646f
F20101106_AABPFU monroe_m_Page_109.tif
9e991652a2c65e2a02d3b7b048c5713a
b8dd610a6d675536623f8dbbb6e2e18f924bc9ae
F20101106_AABOZZ monroe_m_Page_123.jp2
dd766f8d2a676c72641df430e4f10f69
72cbd6b6847116f62e38febc9dee158cd1f25edb
F20101106_AABPGI monroe_m_Page_124.tif
5fd91aa44863274fbb20ef0da67b7388
902a5170f3bfcf375a05500c8f0d21d5e7c913e2
F20101106_AABPFV monroe_m_Page_110.tif
181d7498ca272cb3195724d25b62b166
7e941093832f19da02ab6680e7c92b5dcf3ba86b
F20101106_AABPGJ monroe_m_Page_126.tif
6cf96d1fad59943367b88bdda0e1516d
b8bb41a4f7e93558a19a5f6c98771972d1a4f8ed
F20101106_AABPFW monroe_m_Page_111.tif
0c4ccab220cedec287dfa0669e23f42d
cedcbc26ad46af0cf6a389e025af9a01f34596c7
F20101106_AABPGK monroe_m_Page_128.tif
eaac7ffbb01fc36c12817c8c317bf4a2
18d55495856ec231d15494a911ccb4630fa30304
F20101106_AABPFX monroe_m_Page_113.tif
f71f058c52cd92c4568c3b2b885d4503
d1e5c06ac6ecf5d02b5170a0b8e722c816b8510e
F20101106_AABPHA monroe_m_Page_149.tif
d476a819b823ec58dfaa8f711f45972d
377328dde117dfd0456d0b8f062f16c13ddc7fec
F20101106_AABPGL monroe_m_Page_129.tif
b4d9c19e5fae5bee9892215fcc31580f
ebbf7246efb0bd7236786117346cff49c0b6a26d
F20101106_AABPFY monroe_m_Page_114.tif
92a31bdd48d5b13e3f3b40d3ff2a8653
584be4a09342150265d2f6b751a18362dee4962e
F20101106_AABPHB monroe_m_Page_150.tif
985dbf10f1c6f6dcbe641d57c052f50c
afd0a7d4d4fbb1225afbba985ec3ac9712e3f0cc
F20101106_AABPGM monroe_m_Page_130.tif
5aa0aaa5d5189ed9e094d923bdf0fe9b
7652fceb8b9239ee6f38e6ede48799c430af80a4
F20101106_AABPFZ monroe_m_Page_115.tif
01cb36f17cacd5e825f63d3d55c73aeb
78b5926db8333f4f3dfe2baed1030ff27ec13993
38905 F20101106_AABOEA monroe_m_Page_095.jpg
0b12d39a77b3a2acf580567e79eb0b3f
b1a642e995c8e862d946fa8f6d6ad9e5d19148db
F20101106_AABPHC monroe_m_Page_151.tif
e830f1641c6886f096f69b5abb35575d
b495723444d4f6e2d7e2f28891439a823063e9ef
F20101106_AABPGN monroe_m_Page_133.tif
a98ec9b4dbdfd46b32288cc2fbc976a9
e34b3fccaf5a84e33eda685491d07ef0f8b14167
27352 F20101106_AABODM monroe_m_Page_055.QC.jpg
d6aac1b06d656f1a1a4ec1358fdec87f
8c4d9b87b357243cbce1fdb4e2e0aff416459f67
F20101106_AABPGO monroe_m_Page_134.tif
bb233051281f9146405a6aec25dcd45c
33abc66f472f52eb72fb33ce08e812060bb7b871
15195 F20101106_AABOEB monroe_m_Page_104.QC.jpg
20c56361f9018ee637c6eb495f7097c1
15f013b9ca886a1bf6546230a186b51844d55ad7
F20101106_AABPHD monroe_m_Page_152.tif
f72334fc184d47dfdd75e02098cb3073
e4bd3454b38a52b5979920974327bb0cefaaf4ca
7498 F20101106_AABODN monroe_m_Page_046thm.jpg
e6c8a786618d5f5aa4451e3811f5b509
6984262bd290a9ccfb21f2a9d666b83af39eb4dc
F20101106_AABPGP monroe_m_Page_135.tif
fdcb72518b0f430b2a8ff05b294ac433
389483210fb1b9684c842ea2d6123d8a31f78df6
F20101106_AABOEC monroe_m_Page_189.tif
d82331e31431ade6def146a57b708d78
4f4abe54668939656b563b83c56195d653e835ae
F20101106_AABPHE monroe_m_Page_153.tif
7330a2bbd777aab9a0f50019bc4f559d
8ac496ff6b484cdb6c7db3459149c056aedf41ce
F20101106_AABODO monroe_m_Page_139.jp2
1a04e4bb5988014ae9dbb6a8e3f50825
625c08de5004b1d57ab7a555e74dadf9bf890b4f
F20101106_AABPGQ monroe_m_Page_137.tif
c41a12f5c2ee10aef7fe6cfddfdd6ef2
317019f27b030a3d3a360ff80ce2f3ccd85f2304
27756 F20101106_AABOED monroe_m_Page_020.QC.jpg
4705a6c3bf9bf4e27ec5abc3f7d9560c
6f57a3cd5f330063da4022e0abcdc66e3a08656e
F20101106_AABPHF monroe_m_Page_154.tif
357737500339ccb80138f6d34facaccd
73184a76de2f0ae4f1449e0e8e87104b90030464
7459 F20101106_AABODP monroe_m_Page_132thm.jpg
a896c982b3694a0a350ba579c786ea5d
9640a0c4adbc5f8b86012d19f8db8d5e40434948
F20101106_AABPGR monroe_m_Page_138.tif
b8e6565c4341e09cc534369ee5b63400
08655a586c881e7b4974daabcb67f6eae520a4ec
675 F20101106_AABOEE monroe_m_Page_173.txt
662c99594dcca92e2a946f6d07a854e4
22842f7a1ebe8b85fa443bf94f890108f096c46c
F20101106_AABPHG monroe_m_Page_156.tif
b992804e473e651b90072c5008472313
1d60faff92243a5d7670725219765bb74776a6d8
11471 F20101106_AABODQ monroe_m_Page_177.pro
a787f4f5ca2cc71e31363434c7913b70
7a417e5446eb1b6bc58b3d83ca3af53c1a58a099
F20101106_AABPGS monroe_m_Page_139.tif
9a9c809f0f0914a2b7f976e905369c42
4f7a4ff0c14d46a6193f0f221eaaff201df461ff
2122 F20101106_AABOEF monroe_m_Page_068.txt
b055ee229dab2a12c88926c3449bab03
2a60e0e7e08e746abb9dba0281a8b93ca165f3e1
F20101106_AABPHH monroe_m_Page_157.tif
1319f698fc45cbfbb0655e28d1f5fa76
6a170fe1aa064c5d44cfe27a9532e6fb9ef1c150
4788 F20101106_AABODR monroe_m_Page_180thm.jpg
7d680c1bc77b2afe1140a0cf5d9ba0be
1b71f8b90c070bef769e900a2b6e2685ebb63861
F20101106_AABPGT monroe_m_Page_140.tif
150cd466b7b7761ce312412949339f3d
2870959291a365144a244687b0e75c7a1708c722
2113 F20101106_AABOEG monroe_m_Page_074.txt
990f07f3e42243ad27a2bd9d3b1608bb
4ee1f8ff210f8f06481eccc6a279b7102e4f6a97
F20101106_AABPHI monroe_m_Page_158.tif
cfa03528e6f7cb0a7595346e328a5b4d
3e0a480c53db9d15c4e21804bf9c1c34d161b71f
1051941 F20101106_AABODS monroe_m_Page_022.jp2
022d3c496c29a7dc408ce36a38a86166
169078ee01cb778c998cd25b851329bdc4387d97
F20101106_AABPGU monroe_m_Page_141.tif
6a8ee4de5db0afea595502ec8ee34afd
d50a7e121b5310425bba7d6e9d87412bd85ff41a
38361 F20101106_AABOEH monroe_m_Page_177.jpg
af4d54a10d5b7a072a63f20eac690ae8
1cd8c93d99b4b823b9e18b7795151f11b4e8d3eb
F20101106_AABPHJ monroe_m_Page_160.tif
d3b59dcfe465313475613e93e5c2c208
8db92957f7a444c1830c084ff8da364105ac6284
1314 F20101106_AABODT monroe_m_Page_080.txt
f503eee5c8d3d23eecb6bdc4654df862
bae536378e1709fd2e5c76d8cce2ca7d04b49150
F20101106_AABPGV monroe_m_Page_143.tif
b0bcae75bee63746ec268ea4badb887e
15ff61d7143ff1f5c7e9a2e6d403f184c48de7f2
7477 F20101106_AABOEI monroe_m_Page_047thm.jpg
866c59ae361e73921809bdd30d032a7d
d56d07acf256c99cc74e36d0392ff9744a619677
F20101106_AABPHK monroe_m_Page_161.tif
2db4e95e3e288332e6c32555c195cab7
fd9916bfa38cc04f95b3562ddd7dd41aa57dc040
F20101106_AABODU monroe_m_Page_190thm.jpg
3acf9f72ffdb300cbbee316d5de8d3fe
4594edf0b170a20de99054e55a41a83d3095d6d9
F20101106_AABPGW monroe_m_Page_144.tif
6085d462f4f4149b93c409870476f9c0
982c245790c15978abc41384ec66d19f66a9afb3
F20101106_AABPIA monroe_m_Page_179.tif
07d682f6a499a05861bd8099614f97a9
36d0d7b76ef3eb0a8639bfccfb16690a869985a6
594459 F20101106_AABOEJ monroe_m_Page_098.jp2
777a40ad670fcdd8bb3541cef6b28240
11fa80d99fc7bf4cb60f748812e3780dc9463cad
F20101106_AABPHL monroe_m_Page_162.tif
686896838e7f55e4535e10ba3a4f6aae
38b795ee7dc5b3d3ef6eb99cb4953ec1e5699613
1626 F20101106_AABODV monroe_m_Page_190.txt
2d4c8e222fdf01c386252be0501c9b96
e39352e4349c1b721ab0d44df0f02491a98847e7
F20101106_AABPGX monroe_m_Page_145.tif
d59afe545df847c33d5dc32b6baf4d1d
9a869d0a26d0fe30fc354420219d34077befed2d
F20101106_AABPIB monroe_m_Page_180.tif
8de677cd91ce508a83384a6d0d3fccaf
b8dc5a18f7a756a197c186cdd0c84f4ae7ee5fdc
F20101106_AABOEK monroe_m_Page_082.tif
303cbac1160714cc284e2dde41f32d65
3ec72d74abcf8e861e4f99a8cddba8cfa83898ae
F20101106_AABPHM monroe_m_Page_163.tif
af1c3a9e1fedc825a7e266d552c43d94
1afd2a58cff9f505e5cd6dad9b5ea72862864e1b
33485 F20101106_AABODW monroe_m_Page_190.jpg
66a420b9a6e4bb3a4f5c2c868857b34d
8fd04a0188cf5e4dc558fffceb8ddb2303b2323a
F20101106_AABPGY monroe_m_Page_147.tif
ae8409cb71e7a9108b3132e446ebde39
f9a9355c701c5f7d5b56d3c91ed7a3d75beb7fcd
F20101106_AABPIC monroe_m_Page_181.tif
2bf8d041a0a13a82504f1c89bb2667bc
86bfd14016eb5628b414ad86baaab56c4029f4af
81434 F20101106_AABOFA monroe_m_Page_126.jpg
04ced88a6e1591642948e78d987dfd66
720eb2c9dcc62f5965f72f193a16d09743bd79cd
12155 F20101106_AABOEL monroe_m_Page_173.QC.jpg
1931acac28f895b07c2791fc98062909
333623058d55d743cb72f22115a472bbc50cc3ea
8423998 F20101106_AABPHN monroe_m_Page_164.tif
db76a92d57c45357dd7ffbd69aa290d1
2d8706641303d0cebed0d81247c24c7ee6a50159
39131 F20101106_AABODX monroe_m_Page_173.jpg
b97f75eb38093df7ebfb6af158fed6b1
f153781ffef431dc24dca9aac9260eff0300642a
F20101106_AABPGZ monroe_m_Page_148.tif
0ad9b0da6f3f0360342286a2ef8c59c5
b4c1e08cb54c1e2d84a9a7ef05b3c17aca133127
F20101106_AABPID monroe_m_Page_182.tif
3eddea35ac06fd3485a16b28971f9e47
4c0ff055a83b36a438e6e0f93c45025b739cb679
6966 F20101106_AABOFB monroe_m_Page_152thm.jpg
11d254f83811d96c1a5276d1780f3ddc
dcae660c58334619e8a1b59ed2ec207a5200380d
3302 F20101106_AABOEM monroe_m_Page_187thm.jpg
e0d44f2178ef1ab60e6f6160c5fac54b
7d21a296fddb5f43570235e73fd30eb5a9f8be87
F20101106_AABPHO monroe_m_Page_165.tif
3c02818a9fbe392073f027cda32240f6
fe4a01a37494d7f86e8e35613733db25baeb360c
2093 F20101106_AABODY monroe_m_Page_058.txt
8362eef9219ff371f075e2281acaa9f2
06c0313c52d320d12bd44b6578b71d8ca5ba6cb5
7106 F20101106_AABOEN monroe_m_Page_018thm.jpg
d16d4f01702eefe1fc62dda659723a36
0bcdb90ef9f523e4651f778e3a9390e623655bf6
F20101106_AABPHP monroe_m_Page_166.tif
4295365722626657cb42f0e633c7e0e3
53eaf85aaf61789386fb8c4d36a3b75c0530e902
F20101106_AABODZ monroe_m_Page_127.tif
19005c73762defe25cfe6dd1dea987cc
4f1e45ccbcbda0974a0ffc918934ab690f97656f
F20101106_AABPIE monroe_m_Page_183.tif
40627eff7ad29b06406c21cd4959e301
880a407710aa06ab572277e55b6d068faa83a87a
F20101106_AABOEO monroe_m_Page_197.tif
183c90ce235384fd0385fe626e1a82b3
d9df71511879efa3dd5d9878b0d057d704dbb8ca
F20101106_AABPHQ monroe_m_Page_167.tif
46673c4889c06eedc70eeac7a9f93aaf
ee876e73ea29f628a13e15112ce63057b2bebd26
56637 F20101106_AABOFC monroe_m_Page_069.pro
af3b055daa7bbf978fa3fccb93684a9a
a35416ebc9dc40033ccdabe19010371bc70125c1
F20101106_AABPIF monroe_m_Page_185.tif
4560fe3339f1ca2ab75d686210053d30
83029a712a679ea96f0fd4d67b718214629ad0fb
2106 F20101106_AABOEP monroe_m_Page_029.txt
4ce64415be084af59aaa8c26e6677a37
c4eeaa2484ebe0a53787619cffc66896d2d3c07d
F20101106_AABPHR monroe_m_Page_168.tif
76d95ff6977518fd9a080681e6ffb736
00a831115c2d3f024886853fa88faa738f89d079
73176 F20101106_AABOFD monroe_m_Page_205.jpg
c7b82b386f6630a5a9974b6d939ffac6
c4faff2cd1c8cc4c6bf4897c1ee1def0330cfc03
F20101106_AABPIG monroe_m_Page_186.tif
9318c93a2c7ed4851cb3df0f1e4dc3ba
5d07672a19126dac4a3a22f829f7b79d04d58d71
F20101106_AABPHS monroe_m_Page_169.tif
c2f6ba50c93236fbf77db8601d8dd438
81b3ce4efb8572b4787f137859248b89ab031250
11396 F20101106_AABOEQ monroe_m_Page_092.QC.jpg
75161dd8343b3194398ba00ddd8186ed
e562ad1e34f9c8b8bf10d391a38308a2fa243963
789375 F20101106_AABOFE monroe_m_Page_108.jp2
a71d49a09898cdd3907d075f92d73c63
5ad6f8e2ec4e41369faf172e9f064baeb86e826a
F20101106_AABPIH monroe_m_Page_187.tif
bc74eae8f0a5c9210f3ede3a372be6a1
ffab208de40cdbd20697936d0707d7c0db9659e4
F20101106_AABPHT monroe_m_Page_172.tif
43de97855691fc873f78f7ae53fe99ad
8d8ad005fe9832f5d2cde8915b0dc42da38cea11
1051951 F20101106_AABOER monroe_m_Page_066.jp2
bf685482029b1093091a02fd41cfedcc
39fafb26e98b5dbce42168c37d33bda67c88eabd
26528 F20101106_AABOFF monroe_m_Page_146.QC.jpg
1e0d4009e8f2f33fe5967f2f567afa84
866b8c61ed3ed87f25c45a8cde8aeff2ac624ecd
F20101106_AABPII monroe_m_Page_188.tif
0ba7965aabc324a7186a79e8efbb94ca
f7e2ce435999c054f542838e74aaa0e8b8fe1662
F20101106_AABPHU monroe_m_Page_173.tif
f6fe6490deee8cf966a6b48dab54c573
ba8a422e6743d967c9fc9adb18086b7c8ad47e7a
F20101106_AABOES monroe_m_Page_024.jp2
d74ae1444020efe7f827d5c97780c27a
141f947b925165ac10cae5fc73da39edccd682e1
84507 F20101106_AABOFG monroe_m_Page_150.jpg
1080a1122cd0655e39363138a5013d9f
08bcb990d192a7602124e40b62d765c81171fd91
25265604 F20101106_AABPIJ monroe_m_Page_190.tif
0df863fd068bf03a6fcb82150328febd
2df4f6f330940f10a98ba03fec60ec0cdf1b6bcd
F20101106_AABPHV monroe_m_Page_174.tif
d6835f77267996bfb3e96ec4b3bee2b6
f9d9a45164aea18f6849cd259dd6f75b21226d5f
44821 F20101106_AABOET monroe_m_Page_090.jpg
4257d27e2e7dfe60ac68694d92bd425c
3036ae8998d3d6a69e5ede194448f5081883b148
F20101106_AABOFH monroe_m_Page_204.tif
fd04fa9dccf17c0e33fe89d177b195ee
eb1ace554fac5b82b07fdafdd589ff9458aad320
F20101106_AABPIK monroe_m_Page_191.tif
379968f8414af58158964b80a317f933
efcf915c8013bed2a31454eb6e819dde8995252f
F20101106_AABPHW monroe_m_Page_175.tif
1e0abfffb765dd86547e353a17294311
4bf4d4e037632bf97522f2f73442da30796d7541
F20101106_AABOEU monroe_m_Page_051.tif
da905663b21edfe0c3b2a1676181d2ae
d6dbdcce713348acc041e21e5e6dadc0e141f158
45569 F20101106_AABOFI monroe_m_Page_031.pro
0b5238a7e08fde0074c198a67a0f9033
0568695bf765a25d3834a8a12bd956f115d3ffa9
25947 F20101106_AABPJA monroe_m_Page_008.pro
892881d9b08252b34b79e6bf638ddb87
fa0cc7e14d2228ecf68fde61bfa2ca793a5ab64c
F20101106_AABPIL monroe_m_Page_192.tif
22be93e483e3b8ec033a7f0e83ef3f29
f2b50b50390eef2d25098d3fbcb169d3fff10de3
F20101106_AABPHX monroe_m_Page_176.tif
69378deb3fde35f52db4e860262217ad
6c6d65d011944645f5b76f343abffaf9c341e50f
15068 F20101106_AABOEV monroe_m_Page_155.QC.jpg
07805e5394fb86bde9e62815d9828e8d
24f3355193d3fd35c80bdd5bee8698d77e780f64
84199 F20101106_AABOFJ monroe_m_Page_036.jpg
b67ba7ebcb6f1148416ffa182e2d0c84
c18cd92e8f361cd07d8339960964ecc9b1146887
67150 F20101106_AABPJB monroe_m_Page_009.pro
9f924d1ac907a0a1ceb7500365cb9f18
8b118ae71f6aa458cce6b5dadbd8a7e057fb6e5c
F20101106_AABPIM monroe_m_Page_193.tif
5be5b0fa994942f341d26781008f3f0e
e506f861c3d72cb36a6b1062a974f8886632e090
F20101106_AABPHY monroe_m_Page_177.tif
9e6fc11ecb1d12490f59631ecdeda834
e21701591240d13fb3f38a2d528dd2b5b8914241
F20101106_AABOEW monroe_m_Page_099.tif
499bf9729d7da3ae7aa3c3bcb7f484be
321ccbd4d66a5bc4d308859818a7e66060ac470b
25293 F20101106_AABOFK monroe_m_Page_044.QC.jpg
f798b286c5856ee976bf3b49bf943430
507d8aa89ad5e63ad828bcd112cbe6396f3678ef
77613 F20101106_AABPJC monroe_m_Page_010.pro
3b5f2f8ae5b9ff1d51af059d2c6e7201
2376e54d00143293d25ff716a6f9d3ed1c41b2da
F20101106_AABPIN monroe_m_Page_194.tif
d2382b0cf515dcbbf7a8a516cfc413e4
f0aacbb56261b11b76345819167bdba1083f7f0a
F20101106_AABPHZ monroe_m_Page_178.tif
4075b895f121e5e81b8eedf7d45c1019
2db2d308d6e719a3d9f1ab3e59f34c6f0e7fe81f
347046 F20101106_AABOEX monroe_m_Page_120.jp2
ca564951d7a6738c9a35cd60c6b8d84b
6b29fdf84bfb254ab08608eaba4631ef7019ea89
14872 F20101106_AABOGA monroe_m_Page_166.QC.jpg
8d6403ecb9caae8a4f43cf804daf933d
cf3184d953cd33bfef179fd7637dde45fe1f6a06
1326 F20101106_AABOFL monroe_m_Page_102.txt
0af6209dc5c2ba2d5d2930f77e508210
4d0a4ef73a00797221443a80b6da137f94711d0e
79118 F20101106_AABPJD monroe_m_Page_012.pro
a5f89a888ad8cf7d9a1191c1b180c5fe
58b11ffdcb899ca1d481c0d1fb3dfed90689d6e9
F20101106_AABPIO monroe_m_Page_196.tif
ca55c239bf99d2038467c99c5aff26f5
c3901a8d97501f56668378733e2356e075cbaf10
4143 F20101106_AABOEY monroe_m_Page_084thm.jpg
9ad848caf8534389493c6bf4c6458583
2409d609e7fe5eb57c9cf38cb4d56072640b56d0
342457 F20101106_AABOGB monroe_m_Page_175.jp2
9aa66b0fcfbe0c6199b8ab0b647f00c4
e0d986994df34e70ee36f7d810d370e52c1378f8
6630 F20101106_AABOFM monroe_m_Page_136thm.jpg
d4e283cc148d1e39664e15de4364b474
fa81fa8209d43dc96f3335d5bd7c4ff2f2ebab7b
80595 F20101106_AABPJE monroe_m_Page_013.pro
1421449ce0a0804e5a2b899bb42fb633
f300dd43402f60ab9ca606fc4adfda7107d42b90
F20101106_AABPIP monroe_m_Page_198.tif
5d2a32a521d9f698ee4f9733bedd1908
04d3c686fd567be71ad4620a78e19b3c4ba955d6
49980 F20101106_AABOEZ monroe_m_Page_103.jpg
a1684ac785df3ae464ffd4c832f52318
fcee7feb80155794b69c0c1fb965368c9c9aeb84
6888 F20101106_AABOGC monroe_m_Page_172thm.jpg
634a12d60b4215a407609f12c9b9505b
c301fc913e8d596f275aa7e284e78c0c93b1ca22
6981 F20101106_AABOFN monroe_m_Page_196thm.jpg
cd831a658ed455f87ac4d309dbe74fec
23b1c5f357ab9be1a4989dc28f657ec83c5f5f19
F20101106_AABPIQ monroe_m_Page_200.tif
f5ee22473400c0625b7c9b3745916598
ba0a3c5b666b33e4eb2b81a7368eeb9d1b7348b0
50182 F20101106_AABOFO monroe_m_Page_202.pro
5fac47509198d16e182afd15e5735e49
158a0e860c79dcf2e3c6630763131f1ff6a8556c
8787 F20101106_AABPJF monroe_m_Page_014.pro
ae6513bbfbde16220b42ee913de0aa4c
cf6793ecd39fc33c4dfc7728d5d17cb8ef7ac3ae
F20101106_AABPIR monroe_m_Page_201.tif
0fe12b313ae5a97cb3abbb6d34659385
84e4f25a1fb6505dc5cf0e7f5a5ce056c479e5a5
F20101106_AABOGD monroe_m_Page_142.tif
f68fc85df8a882bbfd31edc2b5a06bed
8b5521845299b56a2757171d399e4bd0d5629137
49642 F20101106_AABOFP monroe_m_Page_072.pro
00180c5ec8e7f1bf39f4fb9f05e6c37d
901b205750f718af3d485681cbe2107ec8c0b8d2
43957 F20101106_AABPJG monroe_m_Page_015.pro
01827272e43590c57ad4ae4b2192b693
fb3edb8d724d5b57a2ba0dd2990a65781dbba61c
F20101106_AABPIS monroe_m_Page_202.tif
ae9c491ddd81ffce0e1d0fe48fa841fa
2be6e39758f7265e12bdce23addf0a2df64a992c
2117 F20101106_AABOGE monroe_m_Page_146.txt
d66ee501e21503511cbf93bf88264c9d
9b7ed5dd2a8733ca288ac1b4d969cef156259566
F20101106_AABOFQ monroe_m_Page_105.tif
d19e32351aa52d20381b63093e6a4e7d
6dbfa635cfe43c2683851771bc9a6e17ae78749d
47953 F20101106_AABPJH monroe_m_Page_017.pro
7d471a66b9619fdfda35f310c58ddd1e
5620210bbf92bf0986b660bc64e0e6a38da6315a
F20101106_AABPIT monroe_m_Page_205.tif
4ac974316fc3ce8ebb2c3c3b72af200f
8fdb51e3674aa13acea351ff4a7c2c65213372e2
2081 F20101106_AABOGF monroe_m_Page_063.txt
ca8bbced7006997c9d6f75d2208dec42
e0f8cf0708357a5d24400f30fec85abb18f558ba
15772 F20101106_AABOFR monroe_m_Page_182.QC.jpg
1319d975c7a6ab26b51389cbb356fee2
fe5eaf0586f11d1af3dbfd4e4a9f07049c100a08
57164 F20101106_AABPJI monroe_m_Page_018.pro
61d69aa224585c856484fa69114f6327
776a11e30a32a5b2a65a680b3f806b7a156be876
F20101106_AABPIU monroe_m_Page_206.tif
b7a4ca2e8b102e2046ab5369e9cd7416
c30884d4a8902d21139902dd2a644620b834dbd3
994137 F20101106_AABOGG monroe_m_Page_162.jp2
ddc3d4b4bcc42333050d1469868fce96
88c9188f7dd930fcf684c4a25981bd19c936996d
F20101106_AABOFS monroe_m_Page_136.tif
a9a8889a849eb328c428532890a89e97
fac25f680f9851b86ce31db297a598cfbc376b2d
56065 F20101106_AABPJJ monroe_m_Page_019.pro
0c3b92827870e76446c52e3499cae4d6
f873402100ccd67b713c8b3544b502e92d823a75
7816 F20101106_AABPIV monroe_m_Page_001.pro
0ac397c3156a631f392fcf5100bb8a3e
db3bea24daa179d668fdf738a4f7de75a484a61a
4136 F20101106_AABOGH monroe_m_Page_156thm.jpg
53f85775e88c3112cc8600218b910842
7a2a4b938ffd9154adae91ac89a148e786dcd1fa
7155 F20101106_AABOFT monroe_m_Page_070thm.jpg
7ff82c55a173a49c358613dc8cef516a
fbaf995a55121db54b79650a87b4100f97ddeeea
56724 F20101106_AABPJK monroe_m_Page_020.pro
f3afb4b57968191d883a2b1ea9e295e0
32617d7be326eb8f6ae7da1aacce05e2eb337fa2
913 F20101106_AABPIW monroe_m_Page_002.pro
e35f46cc68d998ce89f2ac721eede5ad
450397cb419ae4a491156f67b6eedcfb479ed791
F20101106_AABOGI monroe_m_Page_112.tif
a080e6dd6cf71dd566f0f2b0ef2d1ce9
f154f84ae2a691db1b3b8dd5095cbeb6eac7606a
107251 F20101106_AABOFU monroe_m_Page_005.pro
b72a319eda6072b47029bafd2621a378
e3305e9383d22728af6e6aa746c4a31d56e345cd
54307 F20101106_AABPKA monroe_m_Page_043.pro
4e6d87376fe21b709fdb2e2e62e0daa0
1639687a043aeefba65dbbee16b0212f72d2bbbf
52052 F20101106_AABPJL monroe_m_Page_022.pro
03b2eb358973efaf934f3ffa945a3883
ba0cd6905164287720f459ea3556d6f27922c88d
677 F20101106_AABPIX monroe_m_Page_003.pro
c157a5c12ab348a4cbf95217ac77df1a
f865f1765c38f8890e7824c0f42dbd205e8f8d48
42941 F20101106_AABOGJ monroe_m_Page_160.jpg
4c71358d6d5dbff76433f23f511c19cb
1e55b29641ec0c2b8a52d09990cbef1ac4eedb55
575853 F20101106_AABOFV monroe_m_Page_077.jp2
097f91f533033c6b0155c832e9e8cff7
d2fd8ecd3f672c6be2076fd02f46dc453021c51d
52764 F20101106_AABPKB monroe_m_Page_045.pro
a131f7fedfc4e45437ea4029ebd44578
e00179b7a17c7c13a87f9cb5a448a9146d67bd28
54242 F20101106_AABPJM monroe_m_Page_024.pro
f3e6018418050780799a21d5732b9869
31d6e3f767b589e9d5f69cd39cbe068dc0c59f11
46001 F20101106_AABPIY monroe_m_Page_004.pro
d4994b5d3069b6c87e7584a5ac093733
3c78565b206d9d84848c3606428c1713c1c5283f
89528 F20101106_AABOGK monroe_m_Page_021.jpg
83357d0478c0dd59a7ac1983d8a020d3
365fe7b6dcd43f6aedc3154f4c93e01b44ff2edd
27854 F20101106_AABOFW monroe_m_Page_135.QC.jpg
aed8e4d8f47a5136638222ed15746135
318b3728b11c92e0ece4d71d888669a09b23007b
55284 F20101106_AABPKC monroe_m_Page_046.pro
dd777eb76219c993fcead72d27014bbf
e7cda4afbec92ba16a6b6558b76373fccb8afdef
4306 F20101106_AABPJN monroe_m_Page_025.pro
956bdc7a59c1ff41e1bc1c670b46cb9d
e2a474b0c37c5e4858b6aa5f0a7a4911465c940f
70714 F20101106_AABPIZ monroe_m_Page_007.pro
5f13ef990fdf1cafda04e6e35c14f2b8
02eba1a9672803ce1ec66cfbbb769bfe11cc8a24
26426 F20101106_AABOHA monroe_m_Page_043.QC.jpg
9b36b20e7b078a27573b35d56fbade71
041e9414d6ad575d552688088a3161032f6b8c72
30037 F20101106_AABOGL monroe_m_Page_190.pro
be2dbdd7a311581d7c1d971122cc9f3d
796e59d6f48af9c7f13172ca84520667060a46c3
6254 F20101106_AABOFX monroe_m_Page_128thm.jpg
30fac1cc85d4b68298e0ff1d5601868b
46653d69460987062a8672f401413799c543e0d6
55195 F20101106_AABPKD monroe_m_Page_047.pro
d6a7a88cfd51eb013d96ec3429df018e
f8568498f9945c40d2d550453e9ab4bf078542aa
11788 F20101106_AABPJO monroe_m_Page_026.pro
ae8083c9e766a3119fdccb209be86bc8
dbc8cfb24d1fde3084502ae2b37ac109888d2d1b
50324 F20101106_AABOHB monroe_m_Page_204.pro
bef53c8cb761bd8442deb66732337c45
886d4373363d2740d67bb06a181c217d6386eb13
2101 F20101106_AABOGM monroe_m_Page_125.txt
8b5722afb9a7a0acfc66d810ade88521
c8d4c33703c2ccc9e3344683e98e23db73219de3
1051932 F20101106_AABOFY monroe_m_Page_017.jp2
cd0e7adefbfc05b3fca497e3d1b0fc92
9f41d48da108424e1de001d1dd3f51eff7f0cf7d
53085 F20101106_AABPKE monroe_m_Page_049.pro
01dc156c257f7870159b9f4a8c220694
4049a57d7ce2c816108236caaf78421be73ec8ac
50392 F20101106_AABPJP monroe_m_Page_027.pro
0de9fc246f12710ca340f26dfc2edb58
e42f2d6ac5bc1e21e63ee2a0457485723ff6d5ad
F20101106_AABOHC monroe_m_Page_043.tif
78c1373e5d7550255bc63ef53e67377a
8ded5f03158c2fcb697c689aea368e6275694fdb
53712 F20101106_AABOGN monroe_m_Page_150.pro
a3b598d8f135fa8037624e2592985e51
e90c763599db3f2f2198d1a391fc5725a76f22ba
7332 F20101106_AABOFZ monroe_m_Page_142thm.jpg
8e6ecad8f9011c2e989757c76dc6c441
2c40aa3a84ff4593af20035dc55ae808651b3809
54764 F20101106_AABPKF monroe_m_Page_052.pro
e2569dbd49b1a85541f4653b85ea147f
47ae33d48818302d3ba85d8a252d2bf5d38a61ab
53017 F20101106_AABPJQ monroe_m_Page_028.pro
334aa3dd48578539a337c55593718492
70f9d224b3dde8e636cde1b7a371fe9c3bdf9672
668 F20101106_AABOHD monroe_m_Page_188.txt
fc5696b619cda0761f71d737caba31ce
4fec60bd74d6af2085eb668d007c536c440b1726
6285 F20101106_AABOGO monroe_m_Page_165.pro
1d3a70dc1af766925ca5c0a3ccaf97c9
cae4edbe639640eb4065ad3e03138e06adc14337
52641 F20101106_AABPJR monroe_m_Page_029.pro
f71d926f612792040439bf31113b54be
39269028314241a0f1b047dc841af9519a4bb7ec
9626 F20101106_AABOGP monroe_m_Page_190.QC.jpg
eac2850bc42502fd5cdb810048b0e925
21899ce41c3b59bfccb0ddec144ed4c6bd5fe134
46400 F20101106_AABPKG monroe_m_Page_053.pro
9a4552c4833f085eafbee40bd38ca728
db4cb8185080309938390751dd4a30579e392a6f
52451 F20101106_AABPJS monroe_m_Page_030.pro
c76bfe285e9b49a96c322f22cdfa006f
b6cfefd88455514f0ad513c4e8134f92a8753582
2650 F20101106_AABOHE monroe_m_Page_206thm.jpg
ecac4355e6600cba4b6c65b9fb1fe272
3a1939145700c00c75c3c4f00d673b174c8df91d
48819 F20101106_AABOGQ monroe_m_Page_105.jpg
b79c2d11ef2713fa33da1234fea3993a
29501175b6e445df21eaf9f467fc3d440bc43d72
54042 F20101106_AABPKH monroe_m_Page_054.pro
7236539ee72a9ffb454b0bcca689ade9
c3c1659a64832a4b1f135ca0b7258d72a99df08a
52165 F20101106_AABPJT monroe_m_Page_032.pro
1c502c174057ead4079a2110ae030a27
ff6e5c3fa910a6ac152a717b5ed21a402dc9509d
1835 F20101106_AABOHF monroe_m_Page_166.txt
d138eab617ef9d076f5dd7ef9f5cc203
ec14122b35a8a888bd9c4e64bce9ad22639d900a
F20101106_AABOGR monroe_m_Page_019.tif
22fd3dab94b2dbfa17f94e85b2439d75
31d246eea4252f663f76f40a9e6e8b23907a6290
53797 F20101106_AABPKI monroe_m_Page_055.pro
0ba044297bad03c86d3b0ae085863fe8
25374b1b9504b3e6974ae95d9ec6e904d7642772
52452 F20101106_AABPJU monroe_m_Page_033.pro
8bb366d844d8087d459d5d5242997d56
fce1baefadac7d7d290baa2e6c6668ce0c90d139
15743 F20101106_AABOHG monroe_m_Page_016.pro
b3d693d1591dfae0d76d16f00358b928
096ae777b284a38eba18141feb35d415888779d4
14980 F20101106_AABOGS monroe_m_Page_086.pro
f4936069fb78db67e8bd7e97c452f03b
ed018b5d173f50b60dcfa839b496a0bbdb73b449
56505 F20101106_AABPKJ monroe_m_Page_056.pro
9909a273737a45417d72ba6e0de14505
909ba48ba0aa0393fd815ed116b764118368d560
51637 F20101106_AABPJV monroe_m_Page_034.pro
881b61ebb685d797e0a3508af9da9a3e
b6b919658392666335dfe50cc5eebf5cf85d0229
6557 F20101106_AABOHH monroe_m_Page_112.pro
fef568663915b30eabd9ce7f994bb399
c71da18c8a02714978536c7748e5be7239b0aa46
25459 F20101106_AABOGT monroe_m_Page_032.QC.jpg
8734dff9323dc5bb89b3816c8118c7e3
32a75e59ad2ce7be988ea582d85e3cadd1f92f3f
55292 F20101106_AABPKK monroe_m_Page_057.pro
d19d687e90310b1d97f278ce6c759e43
43eff98efba187888f67d13e3b76f073a61111cb
54864 F20101106_AABPJW monroe_m_Page_035.pro
a4bed2a549a5033396f71c37c6ff711b
9be3c43bd864614702d14f993dc2342386bd9b07
30854 F20101106_AABOHI monroe_m_Page_083.pro
66daa924dbd9931aa1fab3b0710ef357
5a7b70a474742c36ba6b48eece3027059a20e6dc
F20101106_AABOGU monroe_m_Page_199.tif
d81684d680132b345ed695303d92c21d
12f535c533be9ac8fc0e9cb2ae6b0459337c6674
35549 F20101106_AABPLA monroe_m_Page_079.pro
3a6f941bba34ac62995f7fcfe5ed447c
600669b17f14691cc6290d2d357fd7c99ad0ccf8
53261 F20101106_AABPKL monroe_m_Page_058.pro
2bad18370580f39a8765f4b19c324312
f8483c0733e4ce3b39c1dd1d9d6f87d58fdc903e
49311 F20101106_AABPJX monroe_m_Page_039.pro
8091e653036bc9969005d319dc2f3583
f3fbf287ced6d06c9ef91a51d11b80b634672cd4
53926 F20101106_AABOHJ monroe_m_Page_038.pro
d529f05767481910777a09b7c7a10b55
a247f4c254efa9be6cc66d75ee908f522c0293a4
42529 F20101106_AABOGV monroe_m_Page_163.jpg
cc78cf9ea09e8464911556c02c5e81e2
f95a16cda15ab12f209c494cec9f06625486c301
21672 F20101106_AABPLB monroe_m_Page_081.pro
ea25c248995ea0e3863f936b61f044ff
53e957c541159c9af2d86f2581c420bd9475c51a
51583 F20101106_AABPKM monroe_m_Page_059.pro
4f5ab2b4e293129e9263c87ff55d7854
35949a6155830eb39bb990f633c50267dc1be19b
55127 F20101106_AABPJY monroe_m_Page_041.pro
2de1b86dc77e70cf1fa74695873a27d0
dbb2c735ad040e7e3537fa5483b0e9c448811c39
F20101106_AABOHK monroe_m_Page_195.tif
852e61c2af9cc1b0eb45ac6bd90fc83a
af280e101d07a6b8b2d8908c84169ad2762c069b
437 F20101106_AABOGW monroe_m_Page_001.txt
8f1d0155a793a5e8ff037fd22edb69d8
c9c720b8a7c81c98e48fa7c63e6711edeb7a3839
19936 F20101106_AABPLC monroe_m_Page_082.pro
e62e6155b071de94b18277024abe7700
bfecd3b5f4dea7d2db1fc736f81b036e8e510c1b
53653 F20101106_AABPKN monroe_m_Page_060.pro
d3f9dccf1994f2d6e20220a6a5f72863
ddf6b699decf18b8b157f90d1a508ac4d1cafa59
55901 F20101106_AABPJZ monroe_m_Page_042.pro
9663930e4b51a3dcd6f5a1bff72c5216
caa860e7228b833dd855f6a290612d9184cf7e3e
50051 F20101106_AABOHL monroe_m_Page_123.pro
94269234ed4f298bc4c36976648bbb98
e842d86a61712b94e6ff68503a4f63e26d8d5fab
5707 F20101106_AABOGX monroe_m_Page_192thm.jpg
1964c4fb88624ab9faa69e509855ab09
2fce88c6de2327fd82ca5d71f83d18aca16bc46c
84079 F20101106_AABOIA monroe_m_Page_142.jpg
223cbdfab56ed1bc23dfbc4b2b5760b4
ebfcb75f407abcbd1f0d49196129c95c96ec4f40
11464 F20101106_AABPLD monroe_m_Page_084.pro
37d515dac9b53482c2fe80aa9b0a611d
a3d4efee5077d1ac0e1a87b66fa830c153b8f183
55061 F20101106_AABPKO monroe_m_Page_061.pro
4c8d6f3ecbfc8176267cfff813ae24dd
a34c01a19570151c0258369aab0461abaaa5488a
450023 F20101106_AABOHM monroe_m_Page_174.jp2
fa696cdc77ae71739b4c968be1e279d1
4ba15bbb8512d68ac7c632b42c343ad8d7e8bb29
2109 F20101106_AABOGY monroe_m_Page_144.txt
19dc372adf23888ee7758c8ff37dd40d
f5a3c27ef4f7483fce039c2468c9d16ad80d57cb
479050 F20101106_AABOIB monroe_m_Page_180.jp2
45a4e3ae004dc46e0d5be4ef657d28b0
7f9f17d4d3c18452fcadd30c9355e54eb712c38e
10389 F20101106_AABPLE monroe_m_Page_087.pro
67356d9a0ea337759e9e72f666b0a714
e45e470e5b73d3726412d865100fc3d345b27951
45301 F20101106_AABPKP monroe_m_Page_062.pro
362f58e87fcf043568e0f8c444f3142a
5b9b980e8bb8f8a354be67d9b96cc53d5c023329
84028 F20101106_AABOHN monroe_m_Page_052.jpg
8b312d1bf995cb8bb2506abf258953d2
0fb12b50207ce59a8ca9e499c723d306b6b6e768
F20101106_AABOGZ monroe_m_Page_016.tif
a2e23ed1c7fd0b7d45cad7dda8c78a52
ea7ca91537c34ddea69e7ca9b3ef11528ea4d66f
947 F20101106_AABOIC monroe_m_Page_088.txt
dd103661be671565c8e096555359553b
3bde7753e567c0c77339bfd3617f933c99a43067
17700 F20101106_AABPLF monroe_m_Page_088.pro
2f23bca6adf599861d570c8126930a66
df4dcaaae8a7735441f9754a837975812d5e27ff
56936 F20101106_AABPKQ monroe_m_Page_064.pro
946dea70252e8b25286435840d615174
fa554ec5e1136d117b26732dfab25cb8e9a25856
32314 F20101106_AABOHO monroe_m_Page_187.jpg
7cd54bdd74f8712ddf455a2026db519a
abda99b074b28bd1ec7c0adce646dd701e1a811f
3532 F20101106_AABOID monroe_m_Page_188thm.jpg
6bb14d8c19bcdaa30acdd782985e45d7
78a1e021fc6b2aa71e1a48eeb1c3f82c6f521232
9839 F20101106_AABPLG monroe_m_Page_089.pro
45271ee8df38808c9d5fa6f4832957c6
62efd4916f7ef5db1de8d596defc9d5fa3bccbd5
52429 F20101106_AABPKR monroe_m_Page_065.pro
624ff03555fe412e90ca93eb91820b93
9370c209b07718bfa674d9812040af1b8015c112
108896 F20101106_AABOHP monroe_m_Page_006.pro
58679098557061cf8d338223d6694b41
aa5f6387c385845e3255445ed72f2bd8068c7e95
65058 F20101106_AABOIE monroe_m_Page_101.jpg
dc589e267c8231d49d3d6db017599cd2
135c4c1448c9a7c1061b38febcc4b230358260dd
53321 F20101106_AABPKS monroe_m_Page_066.pro
f39e63343a9cf35d579db4ab694e9a8a
ec8bea61bc55c8759393f84129ea8f94663f02a4
31081 F20101106_AABOHQ monroe_m_Page_154.pro
14283676044f49433ef98bc268019467
67d3fd94769ab3350255e63c8bd533c5be41650d
15760 F20101106_AABPLH monroe_m_Page_091.pro
ff1ba49885b30a2e22e913d74a59a2f4
82474ec9649dc5dc43b3b54c26112fff8fb4fd66
48986 F20101106_AABPKT monroe_m_Page_067.pro
bbc3dcd71be0275515355bec60abd7e5
33128df2b0acba0b5fa408b14a1e24c56a9c1c61
83443 F20101106_AABOHR monroe_m_Page_011.pro
988bd1f3d09a23b2c1163fc8f7187333
a447386e9138a3baed8529509f0e726d820dddf4
34934 F20101106_AABOIF monroe_m_Page_169.jpg
735de4e9d295830dd06f4065681e4b67
1154aca21f2ef53434e4a86dbbac20055c26df48
22002 F20101106_AABPLI monroe_m_Page_092.pro
8385e5fb5818b6113ebaef6c9a752f63
2f97eb5257b759ef23c4484f25965e9a959d2353
53906 F20101106_AABPKU monroe_m_Page_068.pro
26864f37bf6c057eb29e10c33145b371
26a8c93e0f8c23f93894f482172c94cefc0cb53f
725 F20101106_AABOHS monroe_m_Page_100.txt
9aa941fc485a2f08e94b6363cd244957
8c79281f66d172a2f7e2ad2ef0a06f4839e74a10
F20101106_AABOIG monroe_m_Page_125.jp2
80de4231ffa4208d26acedb09649a029
fbcedd5ecfb8d8dfa2b75f0663159c36780f1368
25237 F20101106_AABPLJ monroe_m_Page_093.pro
d2d112381cb395f7d1d1101236eec296
7add9c27654ec4dc449cccea4fe1bf0b12cbc5ab
56183 F20101106_AABPKV monroe_m_Page_073.pro
cc9ff4e10a6d0d65dca86fe35eede591
5777a48a5e107d227a08e41fa075d1e60490c1c4
22198 F20101106_AABOHT monroe_m_Page_117.pro
deb6069ef4815c37cd2328fa0d40a5d2
a95bdde3d4a487bd799f0f4dae2e69a3851dcea8
402228 F20101106_AABOIH monroe_m_Page_188.jp2
4b1c81c1b395babcf77db8c5750813a0
18d24d3ff0b0151337d6693f557ee1dd7cc8236c
16864 F20101106_AABPLK monroe_m_Page_094.pro
e5f50cb5be3beed0a5c530b175aeadbc
ee9c087b4d5bb3702988238693ba4309da1751c5
53412 F20101106_AABPKW monroe_m_Page_074.pro
45d5362fb41a3e4300348af2b4bb94ee
850cb14451207980a9474fe47686609f10e6a9f7
12369 F20101106_AABOHU monroe_m_Page_156.QC.jpg
dbe95a8c82f89c4dc7114d426d168eea
0e2f9027c4bb898b165f4b9cdc8fb312030d9a77
F20101106_AABOII monroe_m_Page_198.txt
8c5c5044eb245ee755da3af6ed53648f
75a1398fb73f8632d2e5ce1fe2be7aa98dc4a098
7164 F20101106_AABPMA monroe_m_Page_114.pro
2a7ca8a9b0f55ad381b9646900c4be7b
0088e22e784aa6d77d775329fc25fd28ad1b3c98
15179 F20101106_AABPLL monroe_m_Page_095.pro
d59623cf1d168d5d323379b804f469f0
01826498a7869d5e6bfb3b660649f2751be85914
F20101106_AABPKX monroe_m_Page_075.pro
240ddfab77f27a6dba0b5dc7c807bc9f
525855f94e848125b27a0b18634667f76ff34e9e
616 F20101106_AABOHV monroe_m_Page_111.txt
e5ab249c83d9fcb38994e92e8d866fd6
e634bf622457f35b1bf84cb5e043670b0ddb7fbe
88426 F20101106_AABOIJ monroe_m_Page_050.jpg
76f299e871ac4ff824cccc60c022d488
67893fee187c50fc6ce1d74bef89a703307dffab
15218 F20101106_AABPMB monroe_m_Page_115.pro
c90114f832bd0204d7548b2b0b7056dc
7ea3a5e75da279592f0c197e477a6440c96c2168
19397 F20101106_AABPLM monroe_m_Page_096.pro
fcb8c9cd99cd429c670fe7e86abda2d0
a95512bda924b2da96ac864d2c0dd837bb58f3f2
16749 F20101106_AABPKY monroe_m_Page_077.pro
29369edde84c92ff7ef867222f5295e7
9f57bdf16a7825c9ea885e6e1c67ca53a3f9cd7d
11338 F20101106_AABOHW monroe_m_Page_183.QC.jpg
2d7ef4f983b8a024faab33c1455326ea
5d03588d921cb4cd89526bed38b391b2d9f15de4
53348 F20101106_AABOIK monroe_m_Page_070.pro
7f9a2a8fca5581030808ba2159d1da4f
898c07f21927f6bed4f9c50e4359a7c6b4a39d18
9981 F20101106_AABPMC monroe_m_Page_116.pro
e22a52322fd2fe6f3d6f9b823764c447
f7068608873a14d1fd56642fc341fbfeb26df78c
13790 F20101106_AABPLN monroe_m_Page_097.pro
d321e6bd2a96cfb60959a68f14a66473
8a432d24ccc7461536657f6fc2e447a45a1c9bf0
7696 F20101106_AABPKZ monroe_m_Page_078.pro
6dd678c67f5f1bd028587b73e757302a
3646553eb8f3790d79534aa889ec469e4159f6f4
F20101106_AABOHX monroe_m_Page_038.jp2
a73eb9c02617f0817a7aa3141bd2b162
85f48cb58027e313307b13b8a006da85b66cdced
89026 F20101106_AABOJA monroe_m_Page_200.jpg
b3421fec18e328d094ec12b50ed45f7f
4b65fc2a679df881561acf14cc2a8d5af40579c2
9935 F20101106_AABOIL monroe_m_Page_160.pro
099171bc0defc2c1f91db270007a68bb
18d8a15e2b7a5570af46abfcca6d0e3431665275
22764 F20101106_AABPMD monroe_m_Page_118.pro
05d4631152a8bd07868fdce665a5bf55
1baaaf4600c069c5976573ea26ed2bed47f8ac00
17965 F20101106_AABPLO monroe_m_Page_098.pro
4469685c1f0ad81aa8a3719a2692cea9
9fbececb16c6f9eeb3ad8d0c5544b850c6bf7a14
53603 F20101106_AABOHY monroe_m_Page_144.pro
11c46bd04fe4eec07fe35884e8666e03
4e2c069ce14532a396ab37d2d4d633e53ccb985e
19669 F20101106_AABOJB monroe_m_Page_080.pro
5d5926ae0215043e423c5fafe8027e81
a1ab473cf7d4c6730f4e5c3f1c6c0ce50acdfbf2
2132 F20101106_AABOIM monroe_m_Page_024.txt
fcba56fae2fe9fabbee5f431bfa53b55
571d8c544de5a49479ae2b3eb435f85e5f2493b6
22809 F20101106_AABPME monroe_m_Page_119.pro
0d1f38db482fe800b571c8fa81179f5b
7e53f5b7f06049ba3dc8d9d509b6b776906469b3
30837 F20101106_AABPLP monroe_m_Page_099.pro
cdd8ca59327da414839b4a9051780420
a1507d3e62a007752d0d31bdf5e2367c23004250
36389 F20101106_AABOHZ monroe_m_Page_183.jpg
23c355d226c967cab1984597c0a2111d
17baf231e922e4b41b4dd7793e56e18a2898e3f4
49127 F20101106_AABOJC monroe_m_Page_151.pro
fdb063a00543e6322a52e7487823d28f
856ace7ee44c4e8e6840ae008003ed8425cc6c1a
60243 F20101106_AABOIN monroe_m_Page_192.jpg
4d1af66b26ad9c7da87305eddc9d03ff
42841d8e08066194325ac022acc8f714097d8cd8
10185 F20101106_AABPMF monroe_m_Page_120.pro
dfa3edbdaf6bf1525525087fe3ee38e2
4300c63770c0f5204ab0cd3498e4f38489c72bfe
13200 F20101106_AABPLQ monroe_m_Page_100.pro
58e4c3f686b8b48eb885171046469f0e
a629990ce6d20550774bd0243d2d43adc9b132e4
15494 F20101106_AABOJD monroe_m_Page_080.QC.jpg
5bf7b7d036efc73bb360c6cd24f2079f
1e0eeb4d37302a13521e500fd9bca42f5d5d97df
23391 F20101106_AABOIO monroe_m_Page_090.pro
120ac4bd64538e75da05d7cb42019361
cb1e1834667cd457781b8c546f94687b84c7358c
17963 F20101106_AABPMG monroe_m_Page_121.pro
6bbfc0a908746d1b415cfd0be974ee61
f91d4947684709f0fedc46d12beaa5cce57dad77
29563 F20101106_AABPLR monroe_m_Page_101.pro
6286f229283f21604184e93e814921d7
99b8ef4872bd26d9cf0bc593dc3c7c1b7565ebeb
1977 F20101106_AABOJE monroe_m_Page_062.txt
bda1124a9b370477502552e2ee68c8fd
1b0c68098711c72e008e3d8d64717650104c4869
7225 F20101106_AABOIP monroe_m_Page_147thm.jpg
22cb522437a836571bad431626207736
cc4bc1afad9d4ebc0734256c8a54238310979139
54211 F20101106_AABPMH monroe_m_Page_124.pro
bd664019572e96cd51cb0f364c0a16c1
8ab5d44282a22d5d273289f9845e8483ebc63397
24643 F20101106_AABPLS monroe_m_Page_103.pro
6c487100ee1b066f602b14b085ffb0d4
e41b1282d131c9af8e681c473412a73ac4f4680e
27664 F20101106_AABOJF monroe_m_Page_073.QC.jpg
b80d353b6041f4acde94cd561da81796
831eca068ef6063e8707ab07d02997085b44f5d1
2044 F20101106_AABOIQ monroe_m_Page_202.txt
5a0b5c5f739b6f95f70cb21593377b9f
91cb32df365153837cc7af2ef9ed7f124de0b179
23491 F20101106_AABPLT monroe_m_Page_104.pro
2e49952da101be7c90e2c4b2d63f1e45
b3b2e9101a688543c481299d4ae44db7df7caa2d
10147 F20101106_AABOIR monroe_m_Page_113.QC.jpg
935e6745b8ce3ef015f881f804fcb247
ee2354ba2da953072a9daec07b10bd61d96d1f67
53327 F20101106_AABPMI monroe_m_Page_125.pro
a4a9e86203ef348ff698e9a12cde0db0
2e33a582aea4c71bca973837e7ef48c15bbe7fa4
10146 F20101106_AABPLU monroe_m_Page_105.pro
5f4c53023bbcff509a3ace8f6662cb9c
882e716774a2996e369eca252a596dbf06c8c16b
F20101106_AABOJG monroe_m_Page_039.jp2
f7b169c69f7321c3a69ce94194131af5
bc8e55a3c170f8b9e9ab9e0896df5dd42b6dc74d
F20101106_AABOIS monroe_m_Page_035.tif
c70128e14c11addb7232ec62cb782cd3
b51a70a93743e7c910067865bbdcfa1e67a9fdcb
50959 F20101106_AABPMJ monroe_m_Page_126.pro
c16a921d5b2cb39ce24d0af6a7de9fd9
8251d04670af05db66e04d1b13ef98069c15c9c3
14021 F20101106_AABPLV monroe_m_Page_106.pro
c325a51424030fc25172f76aa6a6e8fc
af7e368e1a3dd1fef68f576d6b2c8c8e34a4fc4e
7304 F20101106_AABOJH monroe_m_Page_143thm.jpg
d7d8f6d79c25b4c7ddd16bdc5d589ce9
2f36c4f03712e550b3af722df746545087356200
74551 F20101106_AABOIT monroe_m_Page_172.jpg
862fd4582fdc67c848cbccef5cda5130
b56eaa525e710e5aea8136122f3b6d80d0d042b3
47986 F20101106_AABPMK monroe_m_Page_127.pro
9845e783da57b93f12927288cc96dc34
97ff4149131a22c348bfd3e4b6f4448971290114
13865 F20101106_AABPLW monroe_m_Page_107.pro
a894def23dfb8525ce8dc8844c5da855
29b947b96509c19e9bd2d814c190e5d74d3ae998
27518 F20101106_AABOJI monroe_m_Page_042.QC.jpg
78e8f85bd4fb4a362fa5dd3305ecc809
324eb0a66f9849100fafc0dc14db59bce50b2215
1067 F20101106_AABOIU monroe_m_Page_184.txt
1d96e386bc24c504e5224f5147925fa7
e0c00c141ec9d9186f2650e3eb4d8b3689002bc6
54189 F20101106_AABPNA monroe_m_Page_147.pro
930d4d7896ffe3453a642afe73776614
af8d9b330d27e60ae6c5d987063e09f2d2d2b674
43462 F20101106_AABPML monroe_m_Page_128.pro
68e6e2132e3b01e4b048dddb309a7234
894a3af1f8a882166a81ee2d82baf3fab2604608
13834 F20101106_AABPLX monroe_m_Page_109.pro
d3dd437fbcdd6167cf33c4234ca47fbe
6cafd9b4566365abdfa99f9191c241d03a1e6e7b
2226 F20101106_AABOJJ monroe_m_Page_051.txt
05a0f3f1e380f77a9b52a714497d77e5
91e42195bd922bb7d974c18a670dccc9f659f26c
7750 F20101106_AABOIV monroe_m_Page_011thm.jpg
f0981c73b1bb01981bb33aee6c99f823
be5e31ea8b0918b7f07d857201e19627c3b12669
52008 F20101106_AABPNB monroe_m_Page_149.pro
79f76d4e86259dcfb0a6b5862962799b
d75ab41e3a2702e490f14f63b9c6e84e94b20b26
57033 F20101106_AABPMM monroe_m_Page_129.pro
2ec48873083b4b7bc537d51dbb3638ae
8dfda9ee015081f3c2d28d15c79a8ab52fbceef4
12446 F20101106_AABPLY monroe_m_Page_111.pro
98f0abcd97b64c2f165b50d77339f4a2
65981da2db4d2e3cbd2bd6cf7456ef71977d459a
3939 F20101106_AABOJK monroe_m_Page_186thm.jpg
21f93ff419727107c1b426bd579e33b4
c5d1a0be151bbb30dd28e7571c8f0102063d333f
16763 F20101106_AABOIW monroe_m_Page_162.QC.jpg
6f5f83a9266e32709f13971b08481b6a
c82ebae49790fc00d91f2da30fa712db10fa0788
47669 F20101106_AABPNC monroe_m_Page_152.pro
cae65700408da0bcb95429180d900287
aa5fefc551ff29ff6e0ba3b7c182dbb92e371014
53885 F20101106_AABPMN monroe_m_Page_130.pro
fe396d9facd1bebff47d1f642c212da0
3fd3a98eaf19efdc2403928f60865718c307c917
8281 F20101106_AABPLZ monroe_m_Page_113.pro
70e20de0c097bdd73123321c0b065b44
0ec3b398ee576cc2bc3844b4e17b7b314226645e
603962 F20101106_AABOJL monroe_m_Page_078.jp2
e2c2f4ed4b5164d0119a9b5c6fd8a00e
78f4eb41f28c488b42778f2d916fea869d30dd76
7403 F20101106_AABOIX monroe_m_Page_068thm.jpg
74cc491caea36c2c8c7eff7cdd78621c
cddd54205857acc2a6cfda3c1498cb62d01c2795
53468 F20101106_AABOKA monroe_m_Page_199.pro
6087458668d4136984230cd39e9ad474
4ebea41728bed46a3de4aa2a1a7110817ae229f4
12330 F20101106_AABPND monroe_m_Page_156.pro
4a4d8b89f7a0eff89b2e9a7c765030d0
2d24dfdc0e5043644054a9318140d607f2ba3caf
53674 F20101106_AABPMO monroe_m_Page_131.pro
72e5515414fe33202421fbe1fe2b5c6f
39d21b46f3ca994219e57ecfa5b646899e80827a
54699 F20101106_AABOJM monroe_m_Page_036.pro
eab8d6d92f7b616da60e48cbbdc03f37
de8a8788a4f9985a08f10419f317a357cdc045e4
5573 F20101106_AABOIY monroe_m_Page_101thm.jpg
22c012b99494d106eb8a906ab36c4de5
5aeeaf7fcf77c33cd85d67dd559ec3671007a878
56106 F20101106_AABOKB monroe_m_Page_148.pro
12b0063304a1436c21d0fce5a92ab762
884443f491f48724c5c04e1f9ebcd1e446afa811
8780 F20101106_AABPNE monroe_m_Page_157.pro
a072a25319ce8f9ac8d2eda6a72a45b4
2db7637f3a104da33c3e89e019d2b37b659e270a
49937 F20101106_AABPMP monroe_m_Page_134.pro
251d24b8eeb1218a928ce19ce0d4a8ed
a507f461d61616e01ffc8c5d13749456ea79a3d8
86584 F20101106_AABOJN monroe_m_Page_137.jpg
129cf997f199fd0f447c3ccccfe57af7
31a8d6c0334252166d285270e8b8c051cc1e7873
294443 F20101106_AABOIZ monroe_m_Page_194.jp2
c7c44653748e617298134dc45d73388a
a5803f6b3a2316fac91095ac7661d63b53c98bfb
15498 F20101106_AABOKC monroe_m_Page_115.QC.jpg
98e8a0cb60fec4339fdc3c939b9d59a3
697bb72f605cbee2f451d4fb1873f87ab4cce85e
9109 F20101106_AABPNF monroe_m_Page_158.pro
3c24c34b46072e6d9770ca5d57915802
eb996d806876eb534f4815aff5c729bee07afb67
56056 F20101106_AABPMQ monroe_m_Page_135.pro
23b1f7c19a8f6141c9e826a6568558a0
0b76c3efbf27925b62c50d8dcd4b4cd476d070d8
57351 F20101106_AABOJO monroe_m_Page_200.pro
f6e51e88c26ad22b049622c10fc72cdf
1ac13f33b9fae9eaf30d28b29e38190ac97ebbb2
1882 F20101106_AABOKD monroe_m_Page_203.txt
c8b49d693994d53e4c3e442f5b31eed8
d32f75dcf77895c825779ea2778430f7555dfb7b
8638 F20101106_AABPNG monroe_m_Page_161.pro
01fab47f884a4760dc34daeae88af521
8650190f5cf12808da1507a394b7e6d85e2a4fbb
46925 F20101106_AABPMR monroe_m_Page_136.pro
15108d608d7f09bac82d515fb1f284c4
d6832e68285a5478742c9e82a8ddee3b71ba757e
27665 F20101106_AABOJP monroe_m_Page_069.QC.jpg
c04bf9b9846ecc1734939d0be3e4146c
9523c9757c7bd635d77dd6c66abebd230f40c253
82798 F20101106_AABOKE monroe_m_Page_023.jpg
c5b08d62ab8c81a25d113551f445e5db
285e20819b0cf5e228dcf1ed169ebda7508474ac
23716 F20101106_AABPNH monroe_m_Page_162.pro
8f130a13cb13dc70e8749910bb4bfe98
9084fcc84dbc8cdabc399ede1e120f567415c6ba
55067 F20101106_AABPMS monroe_m_Page_137.pro
fd8506db18a518b409ed583b7e745ab9
8efc3c5a8a3dfa8daa6db3e706685403afa673b3
F20101106_AABOJQ monroe_m_Page_171.tif
87e70035064b317009d1344daaec9463
d08b24ce02d46a1815525fe5018efe1a7fa5fc80
4723 F20101106_AABOKF monroe_m_Page_189thm.jpg
e6f5d98520190a4414fd5d141fba7b81
271921f50c64706cd348f8766ee791549f480335
17641 F20101106_AABPNI monroe_m_Page_163.pro
6972a6c0faab6bc0226e106c11a60a79
40176fa8bc49131c55c6e235b27dc8a02076ebec
53453 F20101106_AABPMT monroe_m_Page_138.pro
ca633e9b4180075446bfd0e5b4409edd
835dc0ca445bcbc0a56d7583ed36348fa230a98e
643695 F20101106_AABOJR monroe_m_Page_189.jp2
7216b050ca6eb4308dc0853643263311
20df2db88d889b17489be20914d2451db7e26da0
27934 F20101106_AABOKG monroe_m_Page_129.QC.jpg
a87622b924ba7ef347a93a8c1f4eaf6b
41ad097ef7f9f3de49cdb10fafc0c92416b67a6e
53105 F20101106_AABPMU monroe_m_Page_139.pro
a115581ebad09453176199bf21408830
b32548169d9047efdb3037c9ccf4d166e60ef332
583 F20101106_AABOJS monroe_m_Page_171.txt
a21ad22b733fc69329009cd1de642269
27868b7961233ae64ecb034bf3011a11e6eec630
6906 F20101106_AABPNJ monroe_m_Page_164.pro
7473237aa13494fe305a5d05a3b0176c
e6c1ef6b6b02dfb4506f827ddc9fa91cf65ab73d
49690 F20101106_AABPMV monroe_m_Page_140.pro
0a72fcaca732ff71f69289335cb5aea8
22131520517a7eb07fa35d0f34de4dff41c85cce
52573 F20101106_AABOJT monroe_m_Page_063.pro
9774943b27e826af1dec12f11b9040bb
67b1f095e595c0015c2342762f8fb5f75bac3ba6
94811 F20101106_AABOKH monroe_m_Page_075.jp2
fa90d79d6672c87b6fca54d2e4212e02
dea7989f5ee1e85c7c286f3baf7ff488b866a264
29130 F20101106_AABPNK monroe_m_Page_166.pro
1e6679f2bc5c1c2e0f3186f2fff06d0c
d98d1925a46d99f0fd2fc2c7693b8906495832dc
54560 F20101106_AABPMW monroe_m_Page_141.pro
bebd34e8ac44fcf1b1ce08e78c7b5e71
0f294afc125e268c5cf942116a337c454616f148
24065 F20101106_AABOJU monroe_m_Page_017.QC.jpg
20f645b26ce034392ff634a1dc5e004c
2f937adc1ca76132b13efc4a6d17d3e53c0c2802
2033 F20101106_AABOKI monroe_m_Page_023.txt
21b35d9c44761ae602b3b513a7a67ed2
a67288840069e0aba61bc54b32f92e2e69dad917
17058 F20101106_AABPOA monroe_m_Page_184.pro
46b804ca323b80a76c2a5bcd6b60af95
f462883a0b5f66209d3b0cb85244173aa1925f85
15741 F20101106_AABPNL monroe_m_Page_167.pro
fc8d4d346af9ef86064e90f1bb298208
818a958db56685331e9e55184b4c228d29935473
54894 F20101106_AABPMX monroe_m_Page_143.pro
0d8c386038e7fbf18ff7ac0cc5fa2363
fa92bd1e88bfab0b03de45b13d8e09b1edb2d7c1
F20101106_AABOJV monroe_m_Page_170.jp2
fd61700d01f04c53af4ad564a89433ce
d59bb71f7895b13201873b64dfdc87db1c5ec74b
80055 F20101106_AABOKJ monroe_m_Page_063.jpg
97d29b4829411c638eca71c85687a126
72c14a2f2d5dd4c7ce4d2309fad2bc79e41e98a3
18090 F20101106_AABPOB monroe_m_Page_185.pro
713b6260e277640030518622f76ab095
8ffbc235c24395fb805e6895b280e03439333cf1
12366 F20101106_AABPNM monroe_m_Page_168.pro
37105fd71d3b74eae3d9efdf5fdb32a2
de707c02277e2c1b947be98541528383366dcbb4
54227 F20101106_AABPMY monroe_m_Page_145.pro
96e7071053b4b7c1ee069955d0af1861
cd14d1a1ab0c6c422e0f8f14c7ceccc07789f107
41451 F20101106_AABOJW monroe_m_Page_165.jpg
e05d5640948c3cc6f4e9e04b9315dffe
81a6fa746c54f8c8f04c35ad0f5d8f688c6ea8aa
18773 F20101106_AABOKK monroe_m_Page_192.QC.jpg
8a4b52b679129da1468e98c530381035
bb81d0e6ffc425471efd6373aced55c11ad9a83a
16893 F20101106_AABPOC monroe_m_Page_186.pro
08544727a13647ce0d70718560157529
b4d7e01951fa70c45439f13c44560cb0beb5b311
12492 F20101106_AABPNN monroe_m_Page_169.pro
3ccdb5963034423d33dbb4effac72efe
25db8f789d900484f562a6b4a2f9a5154afb758b
53545 F20101106_AABPMZ monroe_m_Page_146.pro
10162c3dc04cd4b06436060e1c16d54e
854a4924078a923cd97086be0561045fd3893929
816 F20101106_AABOJX monroe_m_Page_175.txt
4bd282f46f5447293161ccd98aeeab0a
087e4366f827bd75b021cf78c0c0f35c796284d8
13553 F20101106_AABOLA monroe_m_Page_159.pro
a983361d1d45a41bcc9b42ee21787ecf
83359547f80002d0bdbbbe44fde329127b74a2d0
30803 F20101106_AABOKL monroe_m_Page_113.jpg
29734678b8a97678bdfd8753687802e5
c520728097dcfff06971c1e0f406836f618a4fc6
10228 F20101106_AABPOD monroe_m_Page_188.pro
d5b571eadd4bdff396b0d1f28f9247e2
dcdd098c2c11e5c67c0a4020bbb994820afaebee
20002 F20101106_AABPNO monroe_m_Page_170.pro
40c02d5d823b764089ad7850203c5e77
380f5dfef3637e8ab034af2989705cc0bb6b00b3
1208 F20101106_AABOJY monroe_m_Page_162.txt
15f764932d5632661e99eb206747699a
44ebe641f0a29a7094c070ec13958a08ccccbb43
1189 F20101106_AABOLB monroe_m_Page_082.txt
f7e0e7fe18e9ed1d2afb1b61d877b373
326a99339f3bb4fd942a6bb2bccfe6e31477671b
87033 F20101106_AABOKM monroe_m_Page_020.jpg
effe46d40d145a4f7867fecf1da3b88a
53f09be853ba641a9d460b1a19dc117a999e275e
9661 F20101106_AABPOE monroe_m_Page_189.pro
0df8d61f7177429eae7e7b2fe9ff3f77
02b76406b403261e1fa2579e3f9b0494a875093c
9986 F20101106_AABPNP monroe_m_Page_171.pro
a677c5377fb10cd08f6cc3788da9ee8a
bfa5fbf02e1ce1f3507902e513b3fc9e74afdbdd
6384 F20101106_AABOJZ monroe_m_Page_062thm.jpg
24472cbbce77be557e5949c5f2f9a3e6
62012b78734fa8dca8b2eacf02440484cf3b1258
F20101106_AABOLC monroe_m_Page_054.tif
b5b69699541f8a3328d61c5a18efba6e
a4580ea3785570e4419268b7eca56c7695d8a98c
4007 F20101106_AABOKN monroe_m_Page_169thm.jpg
b800943350f074f2fce79ee074df9d7f
7649dc9086c1c10e0a07e869611a1f626adfddad
7958 F20101106_AABPOF monroe_m_Page_191.pro
dcd7cfbb8ad34d4fa6f2adfcd9a1c388
8bbe6b52d75aca9aade278afb9f2053cf90085fb
5123 F20101106_AABPNQ monroe_m_Page_172.pro
a2637239a9b7b18f70b53350b6ff62db
338071cf515b5f2e959a4fa1fb7574ab20b3b156
2159 F20101106_AABOLD monroe_m_Page_035.txt
d7ecd53c749ab8b97616bb35f611fc59
6a6f31ba7f08511f63645c334465924bb1a12147
25848 F20101106_AABOKO monroe_m_Page_045.QC.jpg
58769b58d9158365cfe26bebfd65beeb
6f6c6e01d1d42b792461f2a57698e733cede06b7
17880 F20101106_AABPOG monroe_m_Page_192.pro
31f114942e9d805550ca8ae7961813ea
03bee9cd8633dfd3f97d220adcafd8dc25943078
9743 F20101106_AABPNR monroe_m_Page_173.pro
74138335a9d5c6575a7496211c5847c6
4ad511631e5a7396d7f4bc51a8321fd4b99b5d60
17053 F20101106_AABOLE monroe_m_Page_108.pro
9084de3a3e9448a34834ff5e7358f526
bca82c9f7877dc668abbee179446e57c84a6e855
F20101106_AABOKP monroe_m_Page_142.jp2
e81cd2ca2aa057b62f896b12f995a33a
619a02049eaa9e590929e3aca159894dfedde2d7
4126 F20101106_AABPOH monroe_m_Page_193.pro
03d4d3e93f0d07162a8bc19716dd0967
d7530599a2b48d3805981f4407c25fcb422eddcc
5376 F20101106_AABPNS monroe_m_Page_174.pro
bad5c7a0408029a1df78ca2be24e2a26
3ac9af1bf623016efdee9bbe4bac2789b3863901
53785 F20101106_AABOLF monroe_m_Page_142.pro
665bb212ac886049158222f79d346fd2
358e2d754dbcdd6ed5e94cdf77d59b5e7794ad1c
22950 F20101106_AABOKQ monroe_m_Page_102.pro
0fd05b0a9f9e88e805e177f0899d7995
239f7abdb6a0ac08e45c07f09a89d3cb5b5c2763
8150 F20101106_AABPOI monroe_m_Page_194.pro
d84fad97effd22cd47919f63e08115ae
63476a243edd1486a089ce1894ab55e8a918d60c
11329 F20101106_AABPNT monroe_m_Page_175.pro
da84f9abf7df23bc991def1f4050b2e8
a35be3a820795bcf2b877cb9023556beb6211578
7212 F20101106_AABOLG monroe_m_Page_187.pro
fc1f045545504936de26c264ea827d45
9c17d21607578840219927a964c261c8a5fdc858
F20101106_AABOKR monroe_m_Page_056.jp2
f2b2a6e355d397f8f00b7653fe1c4bcb
2ea0affb82a790ecef37e87184e92fe73f322106
47034 F20101106_AABPOJ monroe_m_Page_195.pro
0686879673cf05f7eea4e70b323c6920
dc8bde935c6c1c18cd4132f4cc3a893a0e653178
22464 F20101106_AABPNU monroe_m_Page_176.pro
8b864cea19fb010bee12e566a9cfc608
624ee82d3d92d610cd82a26873db9d09e915a5d4
F20101106_AABOLH monroe_m_Page_049.jp2
bb15ab83d69749196ab129befc7e7d32
6b4265896adb8afab8f8e94cc3cf1a26c83fe80b
752424 F20101106_AABOKS monroe_m_Page_091.jp2
ac969409fac7b5968b3ed9a1379827fd
99a4701d21efe49dab0bf8907aee4cee941667e4
6523 F20101106_AABPNV monroe_m_Page_178.pro
181f727eecaab4e07d3f5a2a4123f023
73b79c6e56aeed836f2c9ecf31b3835b73c458eb
26081 F20101106_AABOKT monroe_m_Page_023.QC.jpg
70a86737beb02c8298fed9d9f52517f5
f2de91625a7041cd2d0fb035312cd8c77e19a3a9
50442 F20101106_AABPOK monroe_m_Page_196.pro
3a84ec382535288540ad550f950edc25
5f5cb9c6d4440e95d23c7fc0408f9333a0b03e10
17627 F20101106_AABPNW monroe_m_Page_179.pro
4ab9017514f1a7307aec8edcd36d7fc5
809b8ce08b73147242014abdfb9dc7144b307c54
7617 F20101106_AABOLI monroe_m_Page_001.QC.jpg
96ffebf78caff3c201868a337f0dc922
5191c7854a2b53b19e11249db207c7ed56acd4a8
1051923 F20101106_AABOKU monroe_m_Page_070.jp2
ecea03398d2be66d83fdae5e826a387a
b373930b8aca6de77b5cf1420445ade6f87a21ab
346 F20101106_AABPPA monroe_m_Page_014.txt
30a093e78449648196b8e1f6f358705f
eee39690a4f7c99ac3a04203b7d1ec8d025491e1
44208 F20101106_AABPOL monroe_m_Page_197.pro
b7784b135221fb6211e4e4bf58ae6009
cb6f3f2c28c9125d096c28ed1a31a7663561b2ac
F20101106_AABPNX monroe_m_Page_180.pro
af0f0a43593060f01d48b2150af4f183
52f1301057c4b96c8891f34f6691ad132806360d
5013 F20101106_AABOLJ monroe_m_Page_083thm.jpg
b2bc372e560757ad3a948c92a2e7f8f3
4a833ddca47568c88002f017c9002029f66971f2
26266 F20101106_AABOKV monroe_m_Page_058.QC.jpg
16b5d1235fcc3e2aa7a4634b4bb5d0a4
b3bf8e906cc4fb16146bca0661a7c0f494ba4d4c
1914 F20101106_AABPPB monroe_m_Page_015.txt
9531d5611118beb6a49e9816138738f4
8968d356b6825567c551418ef67d03a30c98d8b6
52111 F20101106_AABPOM monroe_m_Page_198.pro
7fe79a9d75171661ca9be75d0ec585b0
e81f9cbbc4b16d40f36fc7ca041dfc34763009a4
16147 F20101106_AABPNY monroe_m_Page_181.pro
f6036eaf71b00307b7e8b400c60ea853
afe0272e1fee66a5065773a8a135d5b8efde8cd4
22983 F20101106_AABOLK monroe_m_Page_085.pro
e6206d7f340e40d79fc84f8f82772959
9b4948ebd8407cbb08632538c591e5e703178097
24026 F20101106_AABOKW monroe_m_Page_076.pro
34327cecdf07a6938af70394ac042f9f
173816a5e915adf4ccb3fb0ba3038dba618e237b
626 F20101106_AABPPC monroe_m_Page_016.txt
fc184b83fd55590541612ce43b976336
fc4f2596ebbfcf7dd38091c5a4bba96229eb7ba6
51735 F20101106_AABPON monroe_m_Page_201.pro
7cba4a6602a8d7a5657f6f5c571736f0
bc54dbfaa93ad45a17f25e8d749d9475bcb26f79
14862 F20101106_AABPNZ monroe_m_Page_182.pro
3ed68047989434cb4120ae794633ed22
87f76033e57006ccfdbdd303381d7cd10cc85ecf
76691 F20101106_AABOMA monroe_m_Page_017.jpg
cb88f67502b0fd6745ef74deefc83b94
fda89b8bcd7c93455c82929ace37fa0e5d906b8f
F20101106_AABOLL monroe_m_Page_146.tif
7920f83ea98e25c294d49afe3e567620
df1006156b20d5d6947c5fefcac04fd5eb5e0d56
3449 F20101106_AABOKX monroe_m_Page_086thm.jpg
a1903b27c6189579bdce3951b6d2a035
e0bb31e662097579b14f7cc683fc725276103a3b
2051 F20101106_AABPPD monroe_m_Page_017.txt
ce43768886c779149d6f57987ada276e
0b90143414f8a675836b5dbb20f574f9d1641fb4
46485 F20101106_AABPOO monroe_m_Page_203.pro
de5451e1492f4464ea9419cb33389c90
e8f29d4d060dfe9ab52994c7b5a22e1aef9a0fdc
1045884 F20101106_AABOMB monroe_m_Page_101.jp2
7328fa5e2e1a5e1fe23387b84c70df56
ff43ccc68baa01f1d7695c13171db9a81c0ea13e
26483 F20101106_AABOLM monroe_m_Page_147.QC.jpg
d6b2da0d0d973ca0d31822dc548d52d2
1d30e1f18422fab1529607295b6f93edc52f83f0
52493 F20101106_AABOKY monroe_m_Page_094.jpg
01d8c9bf97bdf751c2f41fe2703f7a48
226ab6bec8fe93ea8d3f3df63e2e75390d6e57c0
2293 F20101106_AABPPE monroe_m_Page_018.txt
555404f980846d6f6e21830713b3cbaa
c69fa8b12d57189d948ab948180422a443e4110d
44109 F20101106_AABPOP monroe_m_Page_205.pro
0442d4ac8b9b12ea9b4f24f562ed4f29
24a4d987c5b59f5da3f1c63fcc2059d742a7942f
1051972 F20101106_AABOMC monroe_m_Page_054.jp2
3c13e565507db549dd8688751501be2c
40483bfa67a882e86a341cccf13ae5d736328913
F20101106_AABOLN monroe_m_Page_131.tif
d91183ca5dde67560506b7385db1b45d
b27a6b0457994507b01473a7f64571d57148848c
83831 F20101106_AABOKZ monroe_m_Page_028.jpg
e8f451475e7545091b017be125b9c024
fc37b50148a6fbfda0ffbc3c19d74c9ce5890889
2196 F20101106_AABPPF monroe_m_Page_019.txt
1a141ed2d114cd20f646c67fc7041ccb
28989641e4be25f591882408cc3c3c6229e2bd51
85 F20101106_AABPOQ monroe_m_Page_002.txt
f6971f7e74aed5cdd92416bb5977fbc6
1139179ccf555e57bb075b48826a603a6a8c42c1
1051920 F20101106_AABOMD monroe_m_Page_147.jp2
916f9bee7a90f6756c960b2d9d962bcf
857e36a845f6c5b6f38e8e70b0791392a30362a4
23224 F20101106_AABOLO monroe_m_Page_015.QC.jpg
8547a993c56bb71e131267ff5c44e690
189bb576dd88dc45cee0b4516234834e54a8ad5f
2230 F20101106_AABPPG monroe_m_Page_020.txt
698c0f7ea41b1baba96883e26837009f
26bb83d27dec3cf35e8eaaa27d3f945c22249377
1866 F20101106_AABPOR monroe_m_Page_004.txt
c666bf5617b5e9aec0a43652f4acf084
625e6a9414771b5e88213c2356ceeb64463fb794
F20101106_AABOME monroe_m_Page_033.tif
94e10510191a7ce3d965121965608ceb
57d5ca87c5ca3e7008c1805c2c3e38e2fade9b24
38893 F20101106_AABOLP monroe_m_Page_158.jpg
fe6cf335f0d9cb8453b76a1cd1487504
0f1306525237a0c6ed322d6885bfcf0f8d75a284
2183 F20101106_AABPPH monroe_m_Page_021.txt
26618a6a387e017c0e52464690b9bd22
0f704ecd7843cda93a24ec083b3e5206a55da180
4445 F20101106_AABPOS monroe_m_Page_005.txt
278fc906f15bc1ce462f2c60c8f6106a
24873b3f417ac69537441aece15d10807b5693c8
F20101106_AABOMF monroe_m_Page_060.txt
ae79ad48730d8e220bcc24b21bdbadc0
c0afb415ea16757061d89860ebdac2b22bcfdcdf
F20101106_AABOLQ monroe_m_Page_034.tif
84eda491084448a1356c58d928da34f5
1e93f7db68149a9de14cc9d7555ca28058d64ae1
174 F20101106_AABPPI monroe_m_Page_025.txt
876ae4ed20f31df9683f7e31ae20ad78
bed3f1766a9cf3686d0f42887e30210ba25474ab
4469 F20101106_AABPOT monroe_m_Page_006.txt
5f32c57a5d2a786030bbc4d9fd06a224
9914d7d5368aa508e7cc8952dad086d82b515e3a
F20101106_AABOMG monroe_m_Page_037.jp2
b6d4af296e01a4c3144b6b737c015e4a
6aba02c140f976a8b73a3ca668791615e82bbf8b
F20101106_AABOLR monroe_m_Page_202.jp2
918c1b56bff1b5edcb77bfd211538443
8760480993c66578a616ac0f03befce8fda17563
F20101106_AABPPJ monroe_m_Page_026.txt
a6e21e56f663f4630d2806f8759515d5
6e00f15dcea36a7b8fe725edad2a7fc6ffd4cbf5
2808 F20101106_AABPOU monroe_m_Page_007.txt
eb325abade239a2ec60af0f516e407c1
51961da79485b74e59398d815a80abc510d4a0f0
2210 F20101106_AABOMH monroe_m_Page_073.txt
bc3ce6c9248ac5be532099ad0e97b373
95b2ac47657165d9ec0b0136d11dba68ab7567c8
51150 F20101106_AABOLS monroe_m_Page_044.pro
de911b5e715b16a4ae38d63e8e437e1a
0099ae12db7265becc7ee2e2defa073b2a22f2d8
2090 F20101106_AABPPK monroe_m_Page_027.txt
bc804bb90cd02ad3de649428758af47e
57b008e8113b9e5accc2f324455bda29272f57cb
1026 F20101106_AABPOV monroe_m_Page_008.txt
2b021f16cd6ce18a466ff4e9fb338b65
c74050bcb888f59f73136cef1f6d4c7a3d8c5156
F20101106_AABOMI monroe_m_Page_108.tif
8faa12025d7eb1d018a70074773fc38b
ce3558d557b066f882689077234ddfa6f32fc885
52825 F20101106_AABOLT monroe_m_Page_153.pro
ebd12f2620712ce56081716a2a1ca6c9
fea5c066d10af0b619d6d39befd5474811e3f9a4
3058 F20101106_AABPOW monroe_m_Page_010.txt
2211b9cf7f9e1e4934bb62090238ed4b
af8754506eecac62dd94889a8278922be63748b7
55586 F20101106_AABOLU monroe_m_Page_021.pro
3ddebf026113bd68a04ccf1fc915e3c7
5246bd17ec4025e47a56d485ef5101c1982761b7
2075 F20101106_AABPQA monroe_m_Page_045.txt
a08ff5e35ff53e9f2619a7cefef76796
eb8acd8cd441d2a79066fb7f3aa0e053270a6776
2089 F20101106_AABPPL monroe_m_Page_028.txt
8ae50f68e4d80fda39c06a3dc2d00f8c
56fe0cd26017145333ac3bc3f6ed1b332efe8aa0
3310 F20101106_AABPOX monroe_m_Page_011.txt
f373062bd72305de061d15fea3b11b19
d2559f185b4a97ebf224ace4f3204ad1f7e609c2
27763 F20101106_AABOMJ monroe_m_Page_057.QC.jpg
13937ce59de0008c141d148095938052
32b20ce1d5c18fea677a12fb3fe0fea0a7d809cd
25005 F20101106_AABOLV monroe_m_Page_151.QC.jpg
b5b458049ca09f98ab634a500afd5bb6
91fdbf0b5ca8a43f9202e42b4e2122c29b8cdfe5
2172 F20101106_AABPQB monroe_m_Page_046.txt
45a4c5c70f2ffd73add7ef7c9c0c1b18
7f2aae2e828ab7510422759284d222e1150f283b
2070 F20101106_AABPPM monroe_m_Page_030.txt
150e9904cd1df903c1f97ce298383d8e
ffb3d54ba2262952330c79809537e693fc25aceb
3104 F20101106_AABPOY monroe_m_Page_012.txt
50768e067e284b1a7f07a9ef6e8dde96
34406ab628e68e16714a2db7527de9daf028bb41
F20101106_AABOMK monroe_m_Page_072.tif
9d21953b3abdc1167cadcb9631c92a81
52e3ec2e19ac894fc873fb3a836c550a869b6441
82815 F20101106_AABOLW monroe_m_Page_149.jpg
cb3a99c8e60ba536aa390aa231445d6b
b9cfbfb0fd5ca83412013d724eddf4edd9525326
2173 F20101106_AABPQC monroe_m_Page_047.txt
668c12ac2556c4fc1af17a42a1118379
1b16fb2d7b199259aec6a67c01eb3a9838172644
1840 F20101106_AABPPN monroe_m_Page_031.txt
e7546095c1e2e0c45f062db5d52e3c2a
cde81c69f8a704f87a51c36c6fd24a0a73bab123
3204 F20101106_AABPOZ monroe_m_Page_013.txt
821439b5de5bf280c7faa5899dbc9b58
734e98a46a0df5f5a42ef13448ece53529db72f8
51590 F20101106_AABOML monroe_m_Page_023.pro
8b63273182bde25fceebb16a4b5bd56d
22173d3676a3b72e36f366656f924e0257e5b6e7
1721 F20101106_AABOLX monroe_m_Page_079.txt
c9efca003a18ceef20372917527867ca
eb00c43aa64352778555eaf5ed0532425c86afbf
25267 F20101106_AABONA monroe_m_Page_204.QC.jpg
2cd7bd49c157906c151793cbc1d87d06
8c7dd65dcc98d019f962ead824a2c9136fb90299
2034 F20101106_AABPQD monroe_m_Page_048.txt
10a8de443404d8d407032e2f0c73f309
e10ea0d67ff3bba5a74ca357673e368811614dcb
2058 F20101106_AABPPO monroe_m_Page_032.txt
6d4d3f6a094cfec61acddf4e0a7fd8de
c87b6523d1c2f0bbe034371cb053f90a74397071
9117 F20101106_AABOMM monroe_m_Page_183.pro
e4d5aa839bc01ba02da82a731b71fb94
b8213060251eda99d490dfa504a83ba2135edac3
7085 F20101106_AABOLY monroe_m_Page_052thm.jpg
c3ac9cf308901070fb4a4f8bf83ad564
4921e55e39079de188735acc1b67e4b99669d4ca
F20101106_AABONB monroe_m_Page_022.txt
88748cee076ea5d318eda9c1880a174b
165e4b097258f5ceb5ebc1fa25250d15fd14cbfd
2100 F20101106_AABPQE monroe_m_Page_049.txt
abadea454fe06549e9211f30ce3a4812
34c3583c566f6f6b564eb5a855493639cf6ea890
2072 F20101106_AABPPP monroe_m_Page_033.txt
32cb485186e7b9d0140a25a25e340ace
8c85fafb375652aff587efd1ffe0bbe9bd5fdc47
1545 F20101106_AABOMN monroe_m_Page_076.txt
82737e4ff446da761a0e41465e0c9f03
2fff4bf0a3923f6c355738bb99ab2b0a23b9de41
2142 F20101106_AABOLZ monroe_m_Page_141.txt
2c70f92e104bad5573e51fa1e51ea16f
2d6a75ebc9c61780c500dbdb4af6864dd15b1d19
36768 F20101106_AABONC monroe_m_Page_121.jpg
cc3198edd27dc6f8ee31dcf64883fd24
5d585dd317583a2ff3dbe77fb9374c178cfc0642
2251 F20101106_AABPQF monroe_m_Page_050.txt
c04e7b5cb234a763315b3cbcfb1834af
674159e873502fabb754355eb0341665ee8e7d64
2042 F20101106_AABPPQ monroe_m_Page_034.txt
931b19dd5be64dcf04b239cc164b509f
dacc7fd69ee6490df081d56013942a9ce9b53b74
57230 F20101106_AABOMO monroe_m_Page_050.pro
1493d7788738623644a28ed021d8dde5
f0457fd8cbc00c16aa71919eeb9b603eed47b9fa
F20101106_AABOND monroe_m_Page_199.jp2
79cd1555c218d610c4c2cc70e5311f9c
a7c612d6f5fa2f4369777f2ba8059765a908093e
2149 F20101106_AABPQG monroe_m_Page_052.txt
d31a8ec89e11cd41066f120ed93ab1b7
36575045117c10172629de31ccaeda2fb7aeb702
2152 F20101106_AABPPR monroe_m_Page_036.txt
fea1036f2a9f415ebe550178c89030d9
8f1356e2bf4af68b3eeedb81c0d02b5372cca82c
5115 F20101106_AABOMP monroe_m_Page_182thm.jpg
abf739883aac31fa78e526c4b4b85168
a9342d98c7c85d763ada70ea80bb28d2bcad33a8
50528 F20101106_AABONE monroe_m_Page_133.pro
07410a608f98d674e8cbc52b4a61c3c9
b8e1f8b2f9be670462fca6e4594196f7ad8c6cb1
2125 F20101106_AABPQH monroe_m_Page_054.txt
321a791f194f300fd495c705d6ce253a
58ac7040bd92422b1051bcfac414ed429df9567a
F20101106_AABPPS monroe_m_Page_037.txt
2da4a1bf7e3ae69b936654c0efdd6dd3
4660c36a3d03de1d95a869f9a436e51827246f70
24019 F20101106_AABOMQ monroe_m_Page_001.jpg
e441ac9a918904949dee94844a3f8eed
4b73b81846f12129ce6de07a6932b750299989a4
999779 F20101106_AABONF monroe_m_Page_053.jp2
6b8678ae7730eb4460d94587f8a1a671
a023597c72f3fdbf292f20f0560d31d51c0cc5fc
F20101106_AABPQI monroe_m_Page_055.txt
296be0369f1b577afa11e8cc056cf6e2
c4a95bd4bec51af7a211014afe24f8bdbc48aafa
2118 F20101106_AABPPT monroe_m_Page_038.txt
c8eab656d05b6db43c86582dd94e5547
44ff23bd70a78ac4ece1f6395feef46095bacec6
49404 F20101106_AABOMR monroe_m_Page_071.pro
1622f310a17b4d59a10a6547799e3676
e27448c73efd277fccc994a93791cb6e996fa80b
56801 F20101106_AABONG monroe_m_Page_085.jpg
388c1f3fac776133710c41dcc5691abb
e6316705c4b21fe1d2a2e104d09474d8335d0e19
2211 F20101106_AABPQJ monroe_m_Page_056.txt
76a68a12598d7b2198e3363a739ad0db
da4c27fe1a790798b7041edef2c08411edfd1bf0
1976 F20101106_AABPPU monroe_m_Page_039.txt
af0cfa181e499d7ddef77fc8bfbec20a
5c6cb89609ce79cdc2965e5317d4a09bc297c1d7
F20101106_AABOMS monroe_m_Page_069.tif
46a6f03dd027adc22b7fddf3ecb1a288
b96149dd8da8eae35d235736fd29c5defa9b07fa
F20101106_AABONH monroe_m_Page_066.tif
2d33764f2634b0716b933961c82aeedf
623c7f64af6caa7231223cad9d4c181a3a54d75c
2181 F20101106_AABPQK monroe_m_Page_057.txt
31d991d4a5bfdbe30c213953fe1d250f
2a8b75c12b8aa4c44ca229bc6b515328fcfed7de
2154 F20101106_AABPPV monroe_m_Page_040.txt
981cf70f310e9700518b6667a2987e6b
c804e29da6d4ea816dd02c7a6bbada76cd0c534a
F20101106_AABOMT monroe_m_Page_203.tif
488e6e9fcf8147f39de6d54ab5a7195c
7c2fe00790aee81d6b6101730652b386f36a86ee
84466 F20101106_AABONI monroe_m_Page_131.jpg
1ec89d8eed5bc1ad5ca79ab7210ef3f1
f49dcf048482820f5e08f39e99e66bea985cfe77
2073 F20101106_AABPQL monroe_m_Page_059.txt
4aba4554164a34720f7db1e94d64bb0f
d830d9c56c7a743382220a33056674d0d390263f
2171 F20101106_AABPPW monroe_m_Page_041.txt
282ebfc6f2bf06cb0fc1b834f084f21e
68f4021c8ec9932589fd713ab65ef8840a0ceddd
F20101106_AABOMU monroe_m_Page_170.tif
59b3257ed80ed7599384b79260b7b9eb
3eaea0e7b41ac2bfeb4e9e48b9414591db535e47
12166 F20101106_AABONJ monroe_m_Page_206.pro
c0ede3656fec71786c46a93c3adf7a49
f9f935d5c25a5047cd08f4f5e4e0ecf07b4b4be6
573 F20101106_AABPRA monroe_m_Page_084.txt
b3540d6cee208d8e14fdc983cf12f0b8
1dd38b047e9989cea4c39fc8b595f7c1cbec1913
2195 F20101106_AABPPX monroe_m_Page_042.txt
27e91f3aabd6b2b725775937ff5b5e87
e13652947eb4865e91cee9d74155b7b5199f1884
84686 F20101106_AABOMV monroe_m_Page_037.jpg
26e582a5d958e787bea1355a2afae6a2
6cb726e7a34c1543dc912ca2a893c2e94865628a
1286 F20101106_AABPRB monroe_m_Page_085.txt
96f114c2818819928bd2f48f4bd3a353
90c7cd253f906a45926f68a91ab66256cd3b94f2
2162 F20101106_AABPQM monroe_m_Page_061.txt
61e6ed525e805cbbe418c0b529dd2703
62c9582a144fc1d346de7c4b8e51f6a8dceececd
2133 F20101106_AABPPY monroe_m_Page_043.txt
faac2f0f8d345fa701e1e2ff09192d44
4cb6cff6026ea8d1c9b63520a83e3b463b5b958b
46479 F20101106_AABOMW monroe_m_Page_097.jpg
a7c0e5634742ee1f88afcf20e958da73
e7c731ab417c89d02d204e01c828b2d89292dba1
2877 F20101106_AABONK monroe_m_Page_114thm.jpg
c608f36eec0907422644805ec04bd1d5
629e93cb5b86dfd20c9cba0561ea5c9687c2e1bc
928 F20101106_AABPRC monroe_m_Page_086.txt
476708d08cdd39ba7c9e102336d3812c
f9ef797d3887a1c5859b16f2fcd1fb84f6592af5
2235 F20101106_AABPQN monroe_m_Page_064.txt
e179711b83fbf472e4b57e4cad3e5c11
5ce4b26bc68c568036e7f67e56c70625e95f4d7c
2080 F20101106_AABPPZ monroe_m_Page_044.txt
a232f7364997b0f37a953e358993ef18
935776ab14c5fad54d4a9b4b52a58c1c18e2add1
53579 F20101106_AABOMX monroe_m_Page_037.pro
761d52013c635a36e0d37c517c6f33cb
d81b427b17fd33d58f8b9bf25a887f442e9c028f
3724 F20101106_AABOOA monroe_m_Page_184thm.jpg
708f4cb03635b6488624560e45c80c33
aeaf1ce6f84e308da06c5305efa9f5001f4e3034
4894 F20101106_AABONL monroe_m_Page_176thm.jpg
42c5175ec253f03305d8e42d3f3ccd11
021d663855bdc6757aa595dc55cf8e2f2d76fae9
574 F20101106_AABPRD monroe_m_Page_087.txt
af1c7378c23bd04b9810764474a4adaa
6f534ab44c382da9ea06dd406162407c3655ee6f
F20101106_AABPQO monroe_m_Page_065.txt
6201f3ac2d8eb7465e61502f763fea2d
02d6810bb8bce48ebd4b3bd358980bb9a9088b87
78612 F20101106_AABOOB monroe_m_Page_044.jpg
4e16a79726b2b9f2868b526d5f6ccb79
03a19e41a539b98b634cefb637bbc1211d0f7b7f
20217 F20101106_AABONM monroe_m_Page_155.pro
c6ce03f30188b12a7cfdbefe53ae7f22
fff6057306543945e78706dc0d249a0a90e6dcf5
72 F20101106_AABOMY monroe_m_Page_003.txt
fa495724fc64a4578f656d4d77d6aa08
a38c14cd831466df719be66a8d5c36280a9350a5
671 F20101106_AABPRE monroe_m_Page_089.txt
f44615c386004690c1c2d3e3a0b5a575
6afcce3f657cb18600619a4c4b9197f9e7326518
2098 F20101106_AABPQP monroe_m_Page_066.txt
2a558c18ff6723edaba700a7d1fe5a8b
90f209e643c0d7f822caf414ed132e129311edc6
F20101106_AABOOC monroe_m_Page_125.tif
0d41ea89ec4a86a32baa76abafa5dc90
152f8782df2cdb1bfd04b1c5123dc2a01a56e4b1
F20101106_AABONN monroe_m_Page_067.tif
19c4e882a3b3abea0a38cdde3f3392d3
4c33fc9502d8bb5907bc54192860ae86311e2545
1345 F20101106_AABOMZ monroe_m_Page_003thm.jpg
7fba3f83c8fe60ffb76114f6847fb44f
d463b2322a87e3f16399efaac780ce4800647d05
1375 F20101106_AABPRF monroe_m_Page_090.txt
e15fc445e12b392110f09bb5a30d72ff
72f6b34ae137f9815a82110d22328a4a113d6013
1943 F20101106_AABPQQ monroe_m_Page_067.txt
1532b22b90b7438d8a0689019708b0ac
093f30a9e75c222f9dfb55e4a51bb093bc0960cb
4908 F20101106_AABOOD monroe_m_Page_076thm.jpg
df6b44d4b7e7b5edc6b368f8bfe87649
d36cf3675b3236bf171299d63552a88e4b526fb1
23101 F20101106_AABONO monroe_m_Page_004.QC.jpg
66262ee23cd6eaf4b96034c08f1d2f15
8a8552520f98e22dbda395f2f5cff04ed092f9ae
781 F20101106_AABPRG monroe_m_Page_091.txt
95054c4c637bb399709affdc62a62639
f5f31e0cb27ac7f11e38c8e2cf97bd47769422f5
2221 F20101106_AABPQR monroe_m_Page_069.txt
f0b3b7c3f56e71d57f05d740aa8ea404
9628daa7e139db7d9f09b11e9dbccf37ec6f7b12
1010377 F20101106_AABOOE monroe_m_Page_197.jp2
e775bd58a9ae78ca5ae575a41e3072b9
24e9b198524fca7f73b1e039117f8ef65324a543
51636 F20101106_AABONP monroe_m_Page_122.pro
2f223e9a2e9817c3a71d63f6c820acc1
e54714f58a38c88880e0bd89eec1e72eba733ce1
1561 F20101106_AABPRH monroe_m_Page_092.txt
bfce8156bd830a57add2686d32380305
9fe37d2dc4d939e299ea65822326d75a3a8b4911
F20101106_AABPQS monroe_m_Page_070.txt
b84046b6f4e9b7d52ca56564526a41dd
446216dcf9f818aaebfd0766ab1b426e53f085e1
1051982 F20101106_AABOOF monroe_m_Page_033.jp2
dafad4fec649d6aa3e0e3acbcfe1e57b
e0ceae719c7f471114d63a4ebf9ac14c8ac68b9b
6485 F20101106_AABONQ monroe_m_Page_053thm.jpg
c1c6e4a187720e8785f15729879a0b35
71df1ab3b0d29ab6a69c7fcfcb02750ece7eb540
1362 F20101106_AABPRI monroe_m_Page_093.txt
8522299eb5727fd480a761fd5e7d889a
7f4e12fc05e1a9ba43d956a15fd34d181f891f26
1960 F20101106_AABPQT monroe_m_Page_071.txt
dec28c765da682c0d75e29fbbb44fac4
035d812628c6949cc9b4b620379d15e05476dde0
4716 F20101106_AABOOG monroe_m_Page_115thm.jpg
b23d5884e4292cd944842fd83c0b02fd
2d9dc33cb999689ebeee57dae71491cb271ecc66
24431 F20101106_AABONR monroe_m_Page_203.QC.jpg
a3dc040244d342f87e18c7b7929655d1
d87867d532b88f39982c7bcd9f39975d20fa99f1
924 F20101106_AABPRJ monroe_m_Page_094.txt
f085b3383da899a90f1fd14b089236b5
11303f7a5755b4a42be0470a027e0351b1e4f05b
1969 F20101106_AABPQU monroe_m_Page_072.txt
db47be7a98da1651e21855649804eefb
a4574b1caaeacea8735930c798034e2350374170
40672 F20101106_AABOOH monroe_m_Page_191.jpg
e08525050327865ee7ef831e188c3730
913b0bc1f4f085a1297103d8880a738d1ed33f8e
6408 F20101106_AABONS monroe_m_Page_015thm.jpg
ff31234da4e7423b241324aed73d5908
091f45f0442c2018aaa86e7c393d5cce3ac72f3d
788 F20101106_AABPRK monroe_m_Page_095.txt
16cc12aee9d6d991621c9f351fead7e6
94e115e7fdeec65baf8c5128351f3aea15f18aa0
F20101106_AABPQV monroe_m_Page_075.txt
a3c5247b84cc93f604998bb125b4cada
490656104aa3f9ad661e44860e30b9114fb53480
4643 F20101106_AABOOI monroe_m_Page_117thm.jpg
d183f598ae1e0eb1132d13302cfa1baf
c26e25ac9d8021ddba3b83bce7c18f31aee6e65a
7295 F20101106_AABONT monroe_m_Page_049thm.jpg
a40364ba45867321aa31a4174ba81c15
5647699d0abc2f1edf630c4b8c23b1758c52c8db
944 F20101106_AABPRL monroe_m_Page_096.txt
c94ae8b908e29e1ba195bf3fb98a9437
4c239565488371b96dd44452ce059400859a32d9
1046 F20101106_AABPQW monroe_m_Page_077.txt
c0fdf2268c03283a8ed22f49fb7cdc75
fd801627528783504b4801d3a0898743d0cf94b9
F20101106_AABOOJ monroe_m_Page_184.tif
a560c98767f340e370c6a43bca729276
a69cf8cdd3e8810b1f8325f766d7bdcfb137a0e1
F20101106_AABONU monroe_m_Page_159.tif
373b80b803dc7a4e62fdc0a98d51687f
658ebefb69549b5245cc5bcd6f35b29a9f021ca5
798 F20101106_AABPSA monroe_m_Page_115.txt
1244182a730aa91cee3ae993ea35d999
b89d306620c1c5e1927a6522896a4c06c53f7dfb
922 F20101106_AABPRM monroe_m_Page_097.txt
144a688510533b1777dd931ceed60e97
01ec6448aaeb78f236fddfc6ad5207b99a858809
516 F20101106_AABPQX monroe_m_Page_078.txt
28340daa867c299387cd2fb814541313
205ad5767e100ed9d3b0745d2e64ece67d04b9b4
4150 F20101106_AABOOK monroe_m_Page_168thm.jpg
6f88d8e5fdf05218a4f32033ab0bf4ff
18767a3104f64691ce9ee70479e9161d16050cc3
F20101106_AABONV monroe_m_Page_040.tif
f3cb98b258f58002dd197865fcd2e400
3b39c1b97bb4c74a9f132e9f98c4e6bd66bab30b
504 F20101106_AABPSB monroe_m_Page_116.txt
4ee879da5a27dcad143db57ab29558f7
11b5e25c7db9c765a374a7cc1aa365f1d3f728e2
1080 F20101106_AABPQY monroe_m_Page_081.txt
3956e2299664e73f5366a35867037e64
30a20b430ff780a0de59dd737d6de4c259d8d7a2
78138 F20101106_AABONW monroe_m_Page_134.jpg
185779eb5c5484194909c885d3759958
278516bd889764c5f2284b2a096ba503c65bdf0e
1227 F20101106_AABPSC monroe_m_Page_117.txt
d5278552a301b683154322c51c338339
2253d56d47bd354ae5eb53a10b7f58f7e60b98f3
895 F20101106_AABPRN monroe_m_Page_098.txt
dea8cd2ec85f314ee7adaac6d4f668f2
fa8b5055f8731d6c564eeced6e0d52382495a948
F20101106_AABPQZ monroe_m_Page_083.txt
e5658fb67e252ceb2b3691d62e9cedd8
4a73191f812b87c7bf808618dfb9d4137f160d66
56871 F20101106_AABOOL monroe_m_Page_051.pro
4663fe01954350e7e8d3f038f86c65f6
7cc8c65fcfb3e7dfde11f39e820abb8fa6efe1a9
25654 F20101106_AABONX monroe_m_Page_201.QC.jpg
f213d8ed46bc44a273bb6580ffd768fb
87b0024412a126be4a804d62b746fadc9bf84e8c
1057 F20101106_AABOPA monroe_m_Page_155.txt
7fa07c28c8a92864c5b55127a1ce073e
a3be192e2d386493aa486e418937f01676348953
1347 F20101106_AABPSD monroe_m_Page_118.txt
9a916be319baecd1c4aa5dd3a2e6146f
0092d65f8a271e48b44e0f9f7fa03dbef4f6857c
1564 F20101106_AABPRO monroe_m_Page_099.txt
fa619d5044b65f277bcf5d26d9fcf588
4914469618bb326aa9368570291955ecee070118
55261 F20101106_AABOOM monroe_m_Page_132.pro
eaa369650eb51a5b6ab7669ee82e0247
ec8c2eac961f5afcc8e9a04db3ed87bd7cb3305c
F20101106_AABONY monroe_m_Page_085.tif
29500994bbabb2bf06bc3026901889ce
c2f7de74f59071f4f13f675fee137ac540b1db14
F20101106_AABOPB monroe_m_Page_134.jp2
0eeefe068f0ab351fdcb5a03313fd02a
a46ee7030dfea12416c3284baf18e62caa0fb689
2356 F20101106_AABPSE monroe_m_Page_119.txt
23672fbc48f4ba0636d0c44d10e5e297
4e20665162004720cfa2bb90a1562a5f3afe1b0e
1628 F20101106_AABPRP monroe_m_Page_101.txt
b193c123c7c9c05c5fbd7eb5e0c3a764
40f9d35b46074437c5ba0099e7678022286f5ccc
26743 F20101106_AABOON monroe_m_Page_052.QC.jpg
2245ff9af53dd73f691606dcabdebcdf
70539f94e8af13cdb31c61454182211c37ee8baa
F20101106_AABONZ monroe_m_Page_053.txt
c03394bcaec755aea77ac3ae2371d3fb
e6c0082e707df26a6d4e97cf50aa66513419812d
90089 F20101106_AABOPC monroe_m_Page_007.jpg
4e510745bed1f6f939e2e8e0af8fc3e1
ce892a043478b6bd26df32b2958ca74fbb57fa6d
953 F20101106_AABPSF monroe_m_Page_120.txt
3184d6caf9b304492b0a5e8d2a853342
87c1faeadae89892a92de8e2e6dbd61487fa12dd
1469 F20101106_AABPRQ monroe_m_Page_103.txt
75eae3b5525d2210c36a044de4ab423a
6a842164a6da53d4a4748ea1e90979a98113b57b
51572 F20101106_AABOOO monroe_m_Page_048.pro
2cbcd20911bb1169c2c6cb7e26080474
5419437cb51d6ab7b12a3dc5d0b2917d1b484a23
4462 F20101106_AABOPD monroe_m_Page_181thm.jpg
1fda85c0d645cdbbdcd1c1804ffc7d85
201df3265692bb5c68c917bedd113cabd3beee2f
904 F20101106_AABPSG monroe_m_Page_121.txt
db45caaad9fe1893ecb904db16baacf9
0d186fe7716aec90f732bb7f9e3dbc68cf266b01
1338 F20101106_AABPRR monroe_m_Page_104.txt
3d2f2a2546b1827103e4fdba8fb0d4b2
654ba7bc3281e9a7b9216c1cbc953451428b6105
372384 F20101106_AABOOP monroe_m_Page_164.jp2
81c4c2fdbf20a5c2af5dd52c6bd144ca
b9b5c199b2dcef8d6dfe76802933f5bd29c4f50b
54570 F20101106_AABOPE monroe_m_Page_040.pro
b75170566982f2a5e1a1bb3df87aa5d7
6cd4f06090f8ace2ec1f1581532bea51a2a48ce4
2153 F20101106_AABPSH monroe_m_Page_122.txt
792c8af7d3d6baa0337782e8c93ab464
f0867c6d2057a2a45e51bf34412c1b854aacdafe
527 F20101106_AABPRS monroe_m_Page_105.txt
bcd1141eb7f169e5c7c51b60f216295a
58242d53a7e8b5f1bd5fb1ee76355f9a791bf772
85593 F20101106_AABOOQ monroe_m_Page_141.jpg
3efb50ec10ad7942231e92d46a6c1b17
d1cb8d29b48e4a8ff1667a0a4ed68d1ea3647fe4
4387 F20101106_AABOPF monroe_m_Page_160thm.jpg
6ef42ea404ac50f11fcaebcda4b4c537
7d5d9810455febef8281c8a86fa2a077a92571ff
1974 F20101106_AABPSI monroe_m_Page_123.txt
2694d4cb7d7cacc807d1659ae6b40c68
75251a0a13080e00b457bd62a12d217f0d02d2ec
971 F20101106_AABPRT monroe_m_Page_106.txt
e022b4bc5bd61dd44d131dfb8314b84d
a62700e19e26beaa3a79452151373725a4dba7d4
6627 F20101106_AABOOR monroe_m_Page_017thm.jpg
92cda8073bd7dedc0b0df2ca04b60903
6782ebebe31d9f9dfb36cbf4f70f278970922fda
681 F20101106_AABOPG monroe_m_Page_159.txt
22efc198e2c7a4ebc4f23683a8d51944
800b4b466095c36aee4fa1f7b8bf182e587039ca
2131 F20101106_AABPSJ monroe_m_Page_124.txt
dc65745ffa69deec2ade29380eed561b
19226e0bf59f021d984e32d2116eb4e7fb4dd8f6
1187 F20101106_AABPRU monroe_m_Page_108.txt
951196bc5413458145bba15608ae8c84
afc1dbcf68e739dd612f8e1b16356c7ab85b4ed6
F20101106_AABOOS monroe_m_Page_084.tif
5c8bd95c6a5ba8d4f714fd6308c3430c
a6aa15424a9ed38590884a8feb346c79d3a521ee
100461 F20101106_AABOPH monroe_m_Page_025.jp2
242215657dc85cf64b23aae56f695db5
bbdc0910ae1946d13e66b11ec9c51087c4ccd8a9
2043 F20101106_AABPSK monroe_m_Page_126.txt
7fafb8b979ef99272372c402c612a3ae
5d90c18b7ffc79c6650cc99df2a9515eaafacdb7
812 F20101106_AABPRV monroe_m_Page_109.txt
e24970f6c854c3fd422d000cbbda45d8
0aa44d9b64c425a377188730544ee0829b06419f
1051933 F20101106_AABOOT monroe_m_Page_204.jp2
cfdd566cd5b5686aa004a3b4b64f6752
a49fd1d4f60548eebb77478360c6ccea645e1495
786 F20101106_AABOPI monroe_m_Page_107.txt
e7e91432079af73c9abd38fa0ccaa6a2
72c0bac14d784e61f1588590c3d331b2b08a717b
1900 F20101106_AABPSL monroe_m_Page_127.txt
8ddd4a5136be2c551fa36c8dde013cc6
6a816e310e96eb6ea28b18c14d038dd1fbfbe88d
F20101106_AABPRW monroe_m_Page_110.txt
32c1c088ad2bdb65a1ca7e8a7ad05004
669a4d28c6fcbbf79c90c2e5f41a14c6f8c7edeb
21400 F20101106_AABOOU monroe_m_Page_128.QC.jpg
0fc4ed855f1b44f907f31bb66d5760be
e86047c90cff67057b4c664fd00562744e24712f
1053954 F20101106_AABOPJ monroe_m_Page_155.tif
60e6ca1cb7d9454d70127e05be43517d
216d7733618d24f09fd0f3b0fbeea556fdbf0ddf
F20101106_AABPTA monroe_m_Page_143.txt
6d15f1c2535255999747e5530f699467
16139c7eb1b44f5398324f895fb3bc060b9f2162
1803 F20101106_AABPSM monroe_m_Page_128.txt
562acb32ee0cc750b1956ad774e9f8cb
f980163e287542f0db0dd4d7e3164d5200cd0fd9
319 F20101106_AABPRX monroe_m_Page_112.txt
bf66b3d12f83d5195b6106dde0d204a6
fcdc4062fc055ca2111feeb5f0cca326bc579cd9
2668 F20101106_AABOOV monroe_m_Page_009.txt
3f5a53a1e1bfc34cdcc1830813d2c42e
f7ebba23044de3c5ee153858081347b1d9b7d2d5
449 F20101106_AABOPK monroe_m_Page_180.txt
43429b1acc60878b8c22b71bd5aa76a3
2d5894607cd803301246b22e96d868bfa64c6a7f
2127 F20101106_AABPTB monroe_m_Page_145.txt
afa778146be96991e9763f312b17eac4
bc4c42dfadcc07dac7fb3812c76499da6dabfca2
2233 F20101106_AABPSN monroe_m_Page_129.txt
211c9553423e26fddb84f5865cbdda8f
c69f549c14e95f92e4346ed5b6332723d92a1fc3
702 F20101106_AABPRY monroe_m_Page_113.txt
54ee339ec7ebecdd0a24a4295e3a0ba6
2ac8b194dd8da119a71a0abe0a47c95e23b728c8
1195 F20101106_AABOOW monroe_m_Page_176.txt
6a06d344b9cf11a9f3f807a4a2651517
aba7d12ecad3d991980da34a3fb8bdd800483c6c
15223 F20101106_AABOPL monroe_m_Page_189.QC.jpg
51d056b161c8e5ade6a9d050ba01b461
93ba7eecbff9be90fca66f1499bb352445057942
F20101106_AABPTC monroe_m_Page_147.txt
0cbe7aa9d40e553f80e6b0bcd269fa32
95c3a11d0f4badb7f156dedb076ab7a73626eee3
389 F20101106_AABPRZ monroe_m_Page_114.txt
a2edb953f4c72389bfd419632841a54c
ff3c740c09e2a3d7d5b12bd5f84f06d12f89016d
F20101106_AABOOX monroe_m_Page_091.tif
34a0bacb138b27b10cea21681c01189c
ead22867329a1be9b26fd3e532f3d68de7ff27b8
6413 F20101106_AABOQA monroe_m_Page_007thm.jpg
0c0d472873a6e8886d199909b856fa00
cbcd939fedf5849b62cc48e90022e0e6b3a8279c
2201 F20101106_AABPTD monroe_m_Page_148.txt
165308bc96ae2eeeb8bf18cf1d1334c2
2cb8ff03bc3e72845ed3ba3aabf8d9f54d9f7f93
F20101106_AABPSO monroe_m_Page_130.txt
e013874493fcfdad3ddfa69552618c5b
00e57501c441bc5db5697602b151e8d56e3dac8b
6437 F20101106_AABOOY monroe_m_Page_031thm.jpg
ab305bf7bb6e7dbeec4951b07bd1589b
becea4e88b62002487a1e62d7fdefcbfe29a2cc8
11740 F20101106_AABOQB monroe_m_Page_110.pro
661530a090b79c4a33154aec38b0ab1b
4acd23c50b5e0952e6643204c3c5e322ebe56c7c
106441 F20101106_AABOPM monroe_m_Page_011.jpg
9489024bb502eef86f2f3ad08c0f2a9c
2fdaec2862cd588b23acb90056b525be43015b5e
2050 F20101106_AABPTE monroe_m_Page_149.txt
8c2316fe7c214a2452749a89fc2d4d98
0d1bf45e886307fdfb549fb7dae05b60700095be
F20101106_AABPSP monroe_m_Page_131.txt
74895e371f5088e5b2175d870fc70f5d
80c84f9555f6caa78e78c7721173e51519401fd4
F20101106_AABOOZ monroe_m_Page_132.tif
f17925744bb2cb842f70576ba542a2e1
3b4279884e880e2bc268fc516a23103a629e7008
86372 F20101106_AABOQC monroe_m_Page_041.jpg
4ce7de9dfafffa32df6255f11f016cee
dc494ebd33dde01f95bbaa2cc4dfd757f018634d
665 F20101106_AABOPN monroe_m_Page_191.txt
0544336253560c634b5f715805eb78f0
3cfcea34269c3090997f4cb23e726b27baf22bb4
2114 F20101106_AABPTF monroe_m_Page_150.txt
d3bf58867ead396bc01fe8e39124e41a
103b034e90cb4771a2a877d7c775dfc618a1fa87
2167 F20101106_AABPSQ monroe_m_Page_132.txt
1eddb05b877a35a860ccf602171fdbce
153ebe6b3c294acaa62e26a71a5168ad9982449d
17960 F20101106_AABOQD monroe_m_Page_085.QC.jpg
e2206737211596547cb1d959f6399af1
a88e701b9e06f352b2e2720ac2ebefc28cc53551
9910 F20101106_AABOPO monroe_m_Page_157.QC.jpg
a45cf43d0ff78cf2a3c5afa1bf31ac17
3a1af1a2a166678b0930d7130f060ed05845e590
2040 F20101106_AABPTG monroe_m_Page_151.txt
e7f241dfc9537cb7516003ab1af6a437
b7b24e7144768421c61f38ac9e4c6007e8bca93f
1994 F20101106_AABPSR monroe_m_Page_133.txt
1f13490ff8219b9ee3e5ac323329d5d9
bb3ef1a133ecd617c536a8c3accab55dff7ed57a
3871 F20101106_AABOQE monroe_m_Page_183thm.jpg
9d504bc9e09a4d84fdd5be103b7c628a
49947b3af40bcf324972dec12484250d641d4088
67574 F20101106_AABOPP monroe_m_Page_171.jpg
01dfe29e8fff8a24da0b95b5b33f6463
447f5eef444c50c520a7d3e737f3a3658114728c
1917 F20101106_AABPTH monroe_m_Page_152.txt
5861a86b8ccb8f1321944953f7ba8ad4
a176cedbc66553fd27b7b715567b734c46bf7806
2053 F20101106_AABPSS monroe_m_Page_134.txt
19d73bcdd5e47e890e04a7cfdf66102f
26ed49f06a1fba951fa89d32181635251ca67d35
84005 F20101106_AABOQF monroe_m_Page_060.jpg
afc93d3a42fde777111a4a3feb8d9bec
6774ceb661c268b02b609e7630dd91c8d3bbe025
48583 F20101106_AABOPQ monroe_m_Page_080.jpg
c486b803b58e1b41218b3c1d5abce702
613bf8891b6c0fed2361a120006eda235994076f
2116 F20101106_AABPTI monroe_m_Page_153.txt
ff15b211c61f705622d7899c46537f41
024b60c741ca4cdbfca166cb500d96342d978242
2229 F20101106_AABPST monroe_m_Page_135.txt
2b38f27bce5f464dee3c3a885f9863a7
4020228112112f2eeaddbc83f8d7cccfcc0a916c
306681 F20101106_AABOQG UFE0022458_00001.xml
9f0041c2c5f5c8d8b397c003e53f8402
7d3b6c911fae2d7a7af1fa07d10a16bb16760c05
657822 F20101106_AABOPR monroe_m_Page_165.jp2
32379fd3411e54955856b0a9e6771d48
5e02c17733c36bfc71e6e785d59890c937ab339d
1306 F20101106_AABPTJ monroe_m_Page_154.txt
812a5dfb43814cc76d9ba420842d39e5
0a515297aacf15b5ce94299804666a2f9e555b2c
1867 F20101106_AABPSU monroe_m_Page_136.txt
2c17f167e33765896aad1f3a6cf95f9f
2170a48faddf590f7d715111b6bc2d439aa580ea
50412 F20101106_AABOPS monroe_m_Page_116.jpg
9993626d7bbf88327a5cd3a5cd2319b7
3f4c90dfaf4a0111b4b34832c7782dfa0169d20c
549 F20101106_AABPTK monroe_m_Page_156.txt
34f2a5503fdd3ea8a02da3bb9629c2e6
74a3d4a93f64c1d3c94c5ebfef5f8c3292cdac0c
2161 F20101106_AABPSV monroe_m_Page_137.txt
1c0abcc31e7b3b739c2aedb644b0fc13
adf0d32f60a90b1b5e7ab6d10a9fb546c009b75c
44843 F20101106_AABOPT monroe_m_Page_180.jpg
f975b8309df648818a60f8a76fbdc1d5
ed93ffab10054995161982828be65d5a869edf4f
606 F20101106_AABPTL monroe_m_Page_157.txt
8250fcc9ec77387500d0b6eb979f5933
64e6c9e1f5573b208dcb049531f2d947485bf9ad
2124 F20101106_AABPSW monroe_m_Page_138.txt
cbf8b4041736dc15c2ea1d854a87e614
ee677d205160456c29ee1684558d950c5cb85215
10088 F20101106_AABOQJ monroe_m_Page_002.jpg
21527ebab292310d6c7e0d5e8c9bdb5e
c8f9ae7a243e99266173d5e2a91c207f863816c7
3535 F20101106_AABOPU monroe_m_Page_157thm.jpg
d2d4c7b364152b4f9def3e0b6ea6ebc8
005054d5ba60071f149048f0eb92796fa57bd1d7
972 F20101106_AABPUA monroe_m_Page_179.txt
24a21d5aa528e780828c22fd31e79f17
8c4cd8d37d025e9ef12f7d986a2f001966f7b222
472 F20101106_AABPTM monroe_m_Page_158.txt
e6d9b13869ea72b837f91b6f1e23b4c5
50d432ae2992201557c38e0e17edac5254b6c7c7
2082 F20101106_AABPSX monroe_m_Page_139.txt
3bee17b55072900dbb7628abbd665ad4
335ae538dc1a194a4d0b8b6f221ac013923e171d
9550 F20101106_AABOQK monroe_m_Page_003.jpg
88fceac350d8a7a4bdb568e47bef224a
307d6ef8de88cb3777431edaff779d3bd19d3d08
729107 F20101106_AABOPV monroe_m_Page_186.jp2
4b26a6aeb509b131b4d45ddc088122ec
72f0b24ca0048245e2f6937940a0fe0535f1b729
749 F20101106_AABPUB monroe_m_Page_181.txt
706ab907dfe0a2618c8174c7a205450c
2534775f2e1228cfcec8bfac57e1f26d9b11b29e
590 F20101106_AABPTN monroe_m_Page_160.txt
0b71ecbce42f51b1ad873badc3114c9a
fb04ec208a7033a55051cb6cbe7560376dc11c18
1965 F20101106_AABPSY monroe_m_Page_140.txt
59a3c976807bc00150f0569924ee947c
1be2fea0b2e04c55dde0b2b26492f013c10e82a4
74977 F20101106_AABOQL monroe_m_Page_004.jpg
cea3e2052353bd1bd65cdd32ccf5c436
ac2100411ec2844b7316f6d14a12a854344af582
77772 F20101106_AABOPW monroe_m_Page_072.jpg
d54c811f380e61743df7d3d8d922ed81
55c8c8ba824db0ec6af6b7bf574ac6065c2b568a
750 F20101106_AABPUC monroe_m_Page_182.txt
1b1a4c64c7c77c8c9691a7b3f07be997
ab586cb15799a561950ed70b95711643cc87de21
459 F20101106_AABPTO monroe_m_Page_161.txt
2a5a24ace0cc7460f8dbfce17db69ca5
ddb5d8122788f5bbf23ede02cfe0a5260bdf5048
F20101106_AABPSZ monroe_m_Page_142.txt
da512767dd6a523c4dc20f165f376959
f6ac6122f6e31494ae79c1abfff3286204ad0d8e
14363 F20101106_AABORA monroe_m_Page_025.jpg
ace1cec47938e27c31862c2b3c9d2ed7
55def60c5a79a9d9aa7e171d9dd9093e1a6ff4e9
83290 F20101106_AABOQM monroe_m_Page_005.jpg
53e3bcb48460837a2211852d9ef301c4
49b67411212741fc4c4d0a47444a95d3476206e8
F20101106_AABOPX monroe_m_Page_036.tif
78b2ab7f794487b6379e5f51b23a689d
4fe75b41b7c44f109ad4665a3c6ab51637aa3584
537 F20101106_AABPUD monroe_m_Page_183.txt
4c333eb7034073db7384090eef46b2cf
21817067a42650620cf7405332db31480b78bea5
49010 F20101106_AABORB monroe_m_Page_026.jpg
2ef09d11c2aaec31b2524f0503e49068
d3a09362f6e143200ce0787a36e235c9e43b720a
83303 F20101106_AABOPY monroe_m_Page_043.jpg
a613274fb9bc75552f4597bf1cc296a7
2837b9a0111391e40983f48637acdaa6ebc6a211
1125 F20101106_AABPUE monroe_m_Page_185.txt
9c07136ce1285370cbcd0ff7b1e86f94
8287827bb54c6357398c7e4349fbf8b3dc7ca2a2
1083 F20101106_AABPTP monroe_m_Page_163.txt
82a16cab7d9c566ebc56971bbffeaacd
2e8dcac0229d5b927da11e8a9ae2c8c700dfa9af
81019 F20101106_AABORC monroe_m_Page_027.jpg
3c2682df1c7b2203e32e0e9e157a3d41
28f2c1232a15cffca2186295a214a3f0b43a7921
92458 F20101106_AABOQN monroe_m_Page_006.jpg
431064554778a926940cacc43bb60de6
ea4b9aff94eabe157f8b00e5cded90e4b21d1d9b
805766 F20101106_AABOPZ monroe_m_Page_118.jp2
fbb03c70436e39e38f8ac2cc551aa496
a9c8cb70cfa935ebcf5fecf0cacca3d72d85a31d
1133 F20101106_AABPUF monroe_m_Page_186.txt
26a1f849535c109d544fcdaa7e2ddba4
640d069333f9ab8d0f63f1e449b1a4dad61837dc
388 F20101106_AABPTQ monroe_m_Page_164.txt
f1287a9cc5a450a512f2af71c86a5dc4
56e862a54c2da4eb4e08c16076dc9d3633aaa9d9
82149 F20101106_AABORD monroe_m_Page_029.jpg
b2496bdc266646554437c6ef0454ffa9
cfb86aeb121a46a8c8e56f08984f08b0ef5d01d9
37254 F20101106_AABOQO monroe_m_Page_008.jpg
c97a063fd7da1f191266d74f34692bdb
96a3552834236b88c8807ee941da9c9cdfb34829
3925 F20101106_AABQAA monroe_m_Page_087thm.jpg
60197085198661d8287a5333a33e6a0d
7d80d1f134699efa109b9983745f603ac9645206
400 F20101106_AABPUG monroe_m_Page_187.txt
07afe4195943337705acab7992b992a1
8cfce72311fe13886aedd6b5877a117dfdd6e3a2
348 F20101106_AABPTR monroe_m_Page_165.txt
d65e007c6049435b681bbca05f41439d
2bbbbebccd67554611b1ae1faeddfa4f067420ba
82172 F20101106_AABORE monroe_m_Page_030.jpg
ffdc7fcaba2231c4db52e92d72ed5ec7
972628b9afc8b2f033759c5f1219ca6a0f553e4b
86639 F20101106_AABOQP monroe_m_Page_009.jpg
8f22927ccad2bf113a51ed0fa185f332
cc04e2afaa9dcb390611dd6ed603e6f0b0652f21
13104 F20101106_AABQAB monroe_m_Page_088.QC.jpg
e506ccef25f00e0635f63c2f2bcdd0ea
17abec617b829963ec1aed820484094c1a1b5406
565 F20101106_AABPUH monroe_m_Page_189.txt
2ab5b217237420a12d97fc2c35079a47
62fb791879126bb165cb7993f9f4f0c860054d19
860 F20101106_AABPTS monroe_m_Page_167.txt
1d7b9769ccc24d5a7a494effb1affb98
ed9ccf8ae87d504a72d83a92977c67e4cbe59613
74271 F20101106_AABORF monroe_m_Page_031.jpg
fd3f5cc4fd28b82f9872f0185b8378c1
2ef9270c3c87815987983a8a8c0dd1d29efa96dc
96893 F20101106_AABOQQ monroe_m_Page_010.jpg
c0d1b2c7b7a67bed72043b9ace3ede6b
ca004c466fe0e56ea6e8c9b730c424661ff72a26
4141 F20101106_AABQAC monroe_m_Page_088thm.jpg
09ecdbe53c576d96efad7cb001fc40b0
71cc810eee6fcfab8dd01b046ab753e62c585080
817 F20101106_AABPUI monroe_m_Page_192.txt
97bdc8c84e72cc139bc380c3d4df9d13
71318bc4a57baefdd3034d70a13feda3b2880b40
621 F20101106_AABPTT monroe_m_Page_168.txt
c99fa4686c6ef795531565a2e01e2992
2d0a8f654e08b19d241974df51972096c2f6d7bc
80681 F20101106_AABORG monroe_m_Page_032.jpg
b2ebc7dd1c4720fae04c9c160233ad9d
20e1b23923664d22a5f0445cc717656960a7a98a
96843 F20101106_AABOQR monroe_m_Page_012.jpg
76b005006dd5e733a230513fa2a83a63
3ee924a39236843fa00d42de1238e63664cf8164
9196 F20101106_AABQAD monroe_m_Page_089.QC.jpg
98fac98186e64c7fab5d6f453ca2a76f
f474fdf0a618218642c21188e88e670708be91b6
223 F20101106_AABPUJ monroe_m_Page_193.txt
2c1b67323f4e68347d3a410c1f6b11cb
2c108b49b541f973ec9ae33ff88fd68d8e04911e
651 F20101106_AABPTU monroe_m_Page_169.txt
d88b64c815cb858fe60b4893de7384b2
ce1501fe6850e57b0d3c4a7fbf05af05a1548719
83480 F20101106_AABORH monroe_m_Page_033.jpg
0bcd31c1368827104447fe068e68cd67
4a0bef9b5e7a0a6d4db51fade7792566df0b0659
104213 F20101106_AABOQS monroe_m_Page_013.jpg
33a9303d905ed446571e51fade5553ca
296529b55766fcbf5dfedd862242b9a803683dcd
3191 F20101106_AABQAE monroe_m_Page_089thm.jpg
c3679d15d4073bc2320f99981ea2fb77
d39e425267bf1127ce4350d6ff63dd6059d83d6b
603 F20101106_AABPUK monroe_m_Page_194.txt
5126c954f7da18ee681c3b9ef2a8b5da
ecba248815a7e2be15d55cae9b528c8838e31a2f
1028 F20101106_AABPTV monroe_m_Page_170.txt
bbbbd42247d8cf31b0219009fe84b5db
76d204d9fabd92f1cfc5c38e28bd7af9cd2339e4
81994 F20101106_AABORI monroe_m_Page_034.jpg
ebd2b7b867047ec5cc66eb4835787887
c817b8b86a0e2068fdb9a65c4be4bf1a468ad61f
18390 F20101106_AABOQT monroe_m_Page_014.jpg
438dd735a2058678f726d97786a941f2
29b4da046bb74061fd480aeab47a2bf0a0c52fb2
13863 F20101106_AABQAF monroe_m_Page_090.QC.jpg
684d16a661cf9987f6dc0cf5de6e4115
2d667b55535b7d2327ff9269e47b26bd531a4a4a
F20101106_AABPUL monroe_m_Page_195.txt
8c67b9689218eff4631b10c454b5317f
bd8a96f75b8bef2637c440995bcad4513e7dada7
308 F20101106_AABPTW monroe_m_Page_172.txt
4b7d2faa00c4933c781a7f7d33a6e70b
5059494ffa5558fc37eeaa016003de1573ea99c2
83995 F20101106_AABORJ monroe_m_Page_035.jpg
cb22015b13a09c6629a024f8ce9b451a
7ead7a1b6399b884d86bdaab407f3546bdf033b0
74184 F20101106_AABOQU monroe_m_Page_015.jpg
837e2633c7796e54fbe0a71b325750d7
7600a5069e5902d506a446f6736ea1fcb5fce621
4412 F20101106_AABQAG monroe_m_Page_090thm.jpg
66fdc0c56c43e5b6da3f2f49213c5b32
e20d83ac220d4574f3ae51272555a97ffb66dc38
22612 F20101106_AABPVA monroe_m_Page_005.QC.jpg
955e70ad13d01cd7d0195be35425dbe6
da7d37441e232dc7052c71564cd00f51508e6319
1987 F20101106_AABPUM monroe_m_Page_196.txt
18794871d7efc2113f4557ef12b0bdde
b92de1467bdf129b2e7df3cbfb09127b3edeadb4
381 F20101106_AABPTX monroe_m_Page_174.txt
605787d65b16741d9fa22369e7b3a001
3cef4875879a3f4aa039dcfb5b24c9bf5a3da8d9
83361 F20101106_AABORK monroe_m_Page_038.jpg
1fcb48561f7ea01441ab4175d473f13d
15d623611c79566bc4cd6426e6176ac0875f8ea4
31166 F20101106_AABOQV monroe_m_Page_016.jpg
104f27d3c5ea078fdeadc876e813602a
b5838bfc5730164c061981583cc0d63d647a5e26
15121 F20101106_AABQAH monroe_m_Page_091.QC.jpg
7231e4c5a7ba49f50e400ed3c244daea
97d21377b5f0aae56e2f2d121bdb472ba8ae70d8
5554 F20101106_AABPVB monroe_m_Page_005thm.jpg
8273d90970b76ab79492151b0cafa839
00bb9d99b0921c037223bed12b8f13e4bc581d0d
1773 F20101106_AABPUN monroe_m_Page_197.txt
98af0dd0cc24a1abbc9ef3a3e34bb360
d023ef27bd304f3b89774ec75fe7d42ada32b323
F20101106_AABPTY monroe_m_Page_177.txt
bd89f7a7ac02a2189ee765a58aa5faf2
18c63ca50643a7275ace15d2f6cc7758aff48611
80281 F20101106_AABORL monroe_m_Page_039.jpg
15f03a73cc6170383657752fec94f2b1
2cd306ea93e4fcfc5134bc754788d419ccffd8fe
90404 F20101106_AABOQW monroe_m_Page_018.jpg
31ce6674004b603c7983b742e7944fc9
4589fa1db9a1517030805a794121bef786a99dcb
4695 F20101106_AABQAI monroe_m_Page_091thm.jpg
63a16c05502b240823baad0bf4b5a109
9fe47b4c9f9123458653901722caaca8208b8356
22773 F20101106_AABPVC monroe_m_Page_006.QC.jpg
07f27d62ae3d5d4ed3ddb66700b7057a
fcb559d9685d9f74fc11a2621649d014f847bb1a
2102 F20101106_AABPUO monroe_m_Page_199.txt
75422a0b7c5a429167d926442b7c29fa
99e865cae821bc90da8adbbda0eb47f690a66694
396 F20101106_AABPTZ monroe_m_Page_178.txt
1716b4365b29e5a5d6275c390e4aabe5
acaf0aa9678db9ca189cae1f0f7096b7d0f373f7
86240 F20101106_AABORM monroe_m_Page_040.jpg
341f8b0dee82f9d8dae8dcb235fee1cc
1a7db338883a1d1afd97eec7c501c546f6ba4a56
87766 F20101106_AABOQX monroe_m_Page_019.jpg
8dc74381685155b9ee39b41930da621d
1ecaa43488db25e53982aa7eea82cb6b5078b187
78966 F20101106_AABOSA monroe_m_Page_059.jpg
e59e2dbf0690cdd9e5d0594c4d7f6715
8a2cb9d50bae14cc9f70fc15b556148b218f0482
3344 F20101106_AABQAJ monroe_m_Page_092thm.jpg
9c509b739a8c4d255b03ca51177c57c8
c0d9b2a12bde79c3753e461a328b9f59b3b517ab
5756 F20101106_AABPVD monroe_m_Page_006thm.jpg
72e28a3045b7e9a9f68f424337888088
4e0ada790be12edc06d8c1269dc8c3424cc5f773
2250 F20101106_AABPUP monroe_m_Page_200.txt
bfc0ed3ec6abd1b81c5928ae1c6b0ae5
dcd5ea6528d4a72017cff203d9a617e92d4a9104
85959 F20101106_AABORN monroe_m_Page_042.jpg
ad1267178d62e452cd06b488600c6dfc
c9303758eb1cc62bd62ba4b5b80c0a2619d6f132
82347 F20101106_AABOQY monroe_m_Page_022.jpg
924b973886c38f468cb9c407728b7ae6
d316692197601cca9dc54c4339009a1ed675d51a
83746 F20101106_AABOSB monroe_m_Page_061.jpg
159ae74f02430f502751d90523af8cf7
b830cb15cc83903aec3811a7a27db0f3c0bf00bf
15433 F20101106_AABQAK monroe_m_Page_093.QC.jpg
4806b1a6b3084ea8fbaca82f4347b2a7
c95b69409b66984211ba0c13fa9f1475867d1b24
25580 F20101106_AABPVE monroe_m_Page_007.QC.jpg
0511044a5fe55f957490ef0325cbff23
b99e4e9ecafd81c0f2ab489a0ac9542d8e1b96f4
84703 F20101106_AABOQZ monroe_m_Page_024.jpg
0aa8d05b7d1c57d2d3f054172a252cf5
8fd06875fd2b0ac6063bbbd61bccf64f735b15bd
68885 F20101106_AABOSC monroe_m_Page_062.jpg
abcfce8c84f2121dc50ffed3d537a5f3
d063e96c7d6534700bd90610fb9bb3697f663a66
4512 F20101106_AABQAL monroe_m_Page_093thm.jpg
2bb3b6dd3e002089fcb59d2ccf78685a
7a92bc968af4e7d5fa07b7d8f856f36f878b6f9e
11007 F20101106_AABPVF monroe_m_Page_008.QC.jpg
1dc7923d3d4e0f1b42ee264a98dcd8a9
a99e47685e54e7c793a788b736f99a1f2548510c
2141 F20101106_AABPUQ monroe_m_Page_201.txt
602f604eb291f576ec304ae05f7a986d
de8908dfe36ae6f13f7e516a88f37c7308adb614
81919 F20101106_AABORO monroe_m_Page_045.jpg
668d93c0914dade88d58f69a15da869c
23237f03896d74e6b0d5836f90f2876ecfa0016d
88391 F20101106_AABOSD monroe_m_Page_064.jpg
f8206d949cf53e4b2b86de94c081e7d4
6a4cbd749a6258247fb3da5edefc72d4d04dc5c4
19233 F20101106_AABQBA monroe_m_Page_101.QC.jpg
6fe5bb125231bee0ad0cf4015381317d
d6eeb6390cf92e3e420d9355f479ebe930855a39
15857 F20101106_AABQAM monroe_m_Page_094.QC.jpg
f146ed9831f4d5dc65411ddac27e55d4
1f079964287361434907a9c3845710a2ac8b1899
3349 F20101106_AABPVG monroe_m_Page_008thm.jpg
8fbc9ba13b034614d08243b99022088f
6b1308928f13c5dad02971139285dfced045e33d
2039 F20101106_AABPUR monroe_m_Page_204.txt
7ab8393c439061640321517317ea1093
ba7bbfaa00a44b58c7cdfbc770b6c92aa9d86df4
86845 F20101106_AABORP monroe_m_Page_046.jpg
672fd3f342ebca0add76b5a07ef70142
be22b0cfa19244cb80b470dc38b5df4f047f7e1e
79303 F20101106_AABOSE monroe_m_Page_065.jpg
1c15a3d22f5947515481672615c53688
fc67b41aa2156e00b0bc311d2203df26941dc9bd
16089 F20101106_AABQBB monroe_m_Page_102.QC.jpg
31f3bf834b4f08032b951e3ff328bef7
a36ecbcefe5b1b52dc1cf1047742e9e8334dc3ff
4818 F20101106_AABQAN monroe_m_Page_094thm.jpg
0916f4b18ef74c819122496eacd0f268
24e84e6cab454e8814b858dc1c4cc340367a3a77
25729 F20101106_AABPVH monroe_m_Page_009.QC.jpg
1b1c4286f4075dab01a0f5682fa3021b
7f3cb84692898892cc1c1b8c0a68e9bf47d3fb06
1797 F20101106_AABPUS monroe_m_Page_205.txt
379fc014e066e3378fc78108828ccbb5
8380385370f7e026674131ae42f206296f2ff52f
87486 F20101106_AABORQ monroe_m_Page_047.jpg
b6851de99d609ea25df70c1cf22c1cc2
4980e4ff5fc2e2319034516aaf87025185cad3fe
82791 F20101106_AABOSF monroe_m_Page_066.jpg
bbc6037f515509e17633d48e43d3b07d
bba463c79d70c36b07843b629161fd8fff528ba2
4833 F20101106_AABQBC monroe_m_Page_102thm.jpg
e0446dbedf6021f6d4b4a62a31dfbdc7
4bdc6461624c3b72b568a075065e1b3bc9826d5c
12480 F20101106_AABQAO monroe_m_Page_095.QC.jpg
b67e0fe13b0d43d25397339b900d9007
a655a01e09149471df5f0c7118bc0c05cf4b128d
6670 F20101106_AABPVI monroe_m_Page_009thm.jpg
fa360cd4b3fcbf8077b74ef5a68b9ccd
f4d7dec17cbb91617f99c7c355c2fa88e39a92b5
520 F20101106_AABPUT monroe_m_Page_206.txt
7075b4548d0c40c1d719f1e1e3d01470
e54f6847974dc8dfa9a10de221ea7bc46f5f34ca
80030 F20101106_AABORR monroe_m_Page_048.jpg
cec763f52036414f62511375b36cc85c
d44d39a087abf4f210cac269ff56e11ed77c43f0
77792 F20101106_AABOSG monroe_m_Page_067.jpg
66da8bd87cb9403c7dff1b7cb876a865
d35cca2a632d831484ffe107374a9156d415c0fe
15890 F20101106_AABQBD monroe_m_Page_103.QC.jpg
69bea3e3603900dcfd7aa4f92adb935c
a791a3da4bacf9010d1b572d53c60b2aa9725b0d
4240 F20101106_AABQAP monroe_m_Page_095thm.jpg
23bf2e5fa0580fa178bbdfd3bbd1fc26
426fcde902fdb697c2ec38a013988effb4d8149c
27439 F20101106_AABPVJ monroe_m_Page_010.QC.jpg
2bc924e9fec65d7fdb15e19ff498a680
b0ed04bcd3e6cb46027d8ef4c4fd158061cd3f75
2368 F20101106_AABPUU monroe_m_Page_001thm.jpg
84f1960c59e1ea2531a58d7ae34fadd9
1e4a2d4ea9c5a9c7d8960785fffe7350fc747cd3
83939 F20101106_AABORS monroe_m_Page_049.jpg
5efdb99a01f265d636852177b280a002
f996cbb61b41c18ae1516eb4bf3cd1086f3c53c8
84599 F20101106_AABOSH monroe_m_Page_068.jpg
b59b27f06b4ec8798d2651a1cf90b510
26da19b71059f1c576fcf5710a009765e4adcba4
4673 F20101106_AABQBE monroe_m_Page_103thm.jpg
fac9f22c51c59eccad7e2f300e9c7620
cda668c1a9cb59772122523dcc80ebc3075021a1
13019 F20101106_AABQAQ monroe_m_Page_096.QC.jpg
c377cb8519559ab1c0a9f843dd0f2ed7
f73c6e8b76ae4d661eecffa9ef17c1ef696891da
7184 F20101106_AABPVK monroe_m_Page_010thm.jpg
c5c806470598a88da7bac35488978bde
360f075b029d16c5dbd743783da0cadedea0f569
11789456 F20101106_AABPUV monroe_m.pdf
4e10eb340021d67f68e0132cbd639332
a08c3befe02a1b36ad84b8db02414ab94cd6e782
88153 F20101106_AABORT monroe_m_Page_051.jpg
f13cdb992bb654d19478984c18d108ab
2fec7188129a3a2fdc804720e91462ae3630a159
87409 F20101106_AABOSI monroe_m_Page_069.jpg
8ec868f97ce5696202256830733bd523
3bc19805fe430a9ce1374df1417795bcd2c5d288
4639 F20101106_AABQBF monroe_m_Page_104thm.jpg
0e1dd0efddc9361df90b87839911b4ea
4d729213f6dc016b24d85126bf0b305818a20fcd
4184 F20101106_AABQAR monroe_m_Page_096thm.jpg
d4591eef01bf9084081e4b0a43fedd89
1e6350a58f252344662ce61d28bd218f4550174b
30284 F20101106_AABPVL monroe_m_Page_011.QC.jpg
36dd5bb2e8f20f5071e159814e2d4ed7
99221c39ce745c7f53894cdd7c5346930ff1d1eb
3266 F20101106_AABPUW monroe_m_Page_002.QC.jpg
6e43f8d6473a153f1380102ae066e7b0
3017cd38b6c70a987e022f4e61c61f05c4c2fb45
72270 F20101106_AABORU monroe_m_Page_053.jpg
ce3782d1da78009bbf78c7c3633d0766
6df80ed598fd7cca33e85ff6c46a0ae65907f979
82804 F20101106_AABOSJ monroe_m_Page_070.jpg
3cb70d403574b7b566945dfac8e1a108
b4e164fd34afe03a232ec457b6f2ac1a4c29875e
14877 F20101106_AABQBG monroe_m_Page_105.QC.jpg
3ad6f24121bd330d0b33a246760f0c0b
be09d06543b7744d79e9ecaaf4f99b0a92e79309
14605 F20101106_AABQAS monroe_m_Page_097.QC.jpg
2865d3347965c6d199b447cb59d331fb
6d6e4333324f6225c9a052bcc7e3795807af5ecf
26117 F20101106_AABPWA monroe_m_Page_022.QC.jpg
98a4f49437f798387e28409ca0184db7
3e738dcda10ff1c78d4e26871c3b184537d94a9b
27281 F20101106_AABPVM monroe_m_Page_012.QC.jpg
eeee2041ad8daf378fa9122474ddea1d
d0cd5bf411463f4d8a44be2d010e80f2a2a9a757
1413 F20101106_AABPUX monroe_m_Page_002thm.jpg
e0585de6f0b1881f8b45ea09f7db3175
8103e6b89135ca4ad196febd2c0d5d572e31be29
77557 F20101106_AABOSK monroe_m_Page_071.jpg
0db292f8587b8d13483430d309a3e00d
b9873f493d8d9ac9b5956f5a44c36ab18cf0b2e5
84293 F20101106_AABORV monroe_m_Page_054.jpg
1a57e79e8e857b9ad14dd6a55742d1f3
904baf555382ed6ede12a39e34232c1ab246ea0d
4650 F20101106_AABQBH monroe_m_Page_105thm.jpg
975f9f33e4ab9602de4e2efb3a0cc70e
239459ed32a63439401c27034b11b55262bc95fe
4794 F20101106_AABQAT monroe_m_Page_097thm.jpg
3221ab883a9103f51c031a5f4ad9c698
6dfc1132cfbefed6d153b3535028665946c804a5
7269 F20101106_AABPWB monroe_m_Page_022thm.jpg
8a2f5549b9003aa942c5b5bf79112199
df13549d05d7f72c790c1ad5cbf4aa8e970e195f
F20101106_AABPVN monroe_m_Page_012thm.jpg
539a2dc330d3eadee089cdf0c449e1a9
3b158ac1c7ed6d58575dc5a7da738b236e4c528a
3085 F20101106_AABPUY monroe_m_Page_003.QC.jpg
080be75efb1c044b3ef836f774674605
2c7816bae74a0b76845ffc4c28114fa0a6b88c92
87155 F20101106_AABOSL monroe_m_Page_073.jpg
deede90169c7b7f888e7782ffa9a6211
813753f169209837e69171974ea7419540741e0a
85267 F20101106_AABORW monroe_m_Page_055.jpg
2dd3477bf540fb045b352a0ce6ba7dcd
98aa77c4a45a49c28084749700dd74f53521113b
12858 F20101106_AABQBI monroe_m_Page_106.QC.jpg
87a08507cbfb68a95ec1ea7d9739cd62
ec4c25ed6e5e4547532899788e8a169aed0734b1
13504 F20101106_AABQAU monroe_m_Page_098.QC.jpg
25d28ac699c41c9a5edee367f46a1f40
6c75bca76621a7cb82bfb3ad3f0b654e1dab4383
7030 F20101106_AABPWC monroe_m_Page_023thm.jpg
0b28a44a4c15ee78a650a10e08a7d256
ea99d3cafefe25e4eb0cb9f4531a47e9852ab492
29425 F20101106_AABPVO monroe_m_Page_013.QC.jpg
eec67ab9d0432cc4aa7399f7ef9bd079
90f56c452b19a6a8ffd8f2d7a9ba2912b206d9e3
6793 F20101106_AABPUZ monroe_m_Page_004thm.jpg
1b673d88b4b478237513f18dbd909a42
949b49a4a8faf05cefb5577f7da45bb111aa32a7
48448 F20101106_AABOTA monroe_m_Page_091.jpg
f744c1f537088a74272afc8ce2bdae93
264ba59bcce46cdea9d043eb46310d2191de5b19
84200 F20101106_AABOSM monroe_m_Page_074.jpg
0722586fda5c999eabe48eac04166ce8
2523c68fcb91077f56bfc5e472e4fba4b6fab07b
86608 F20101106_AABORX monroe_m_Page_056.jpg
6a88ecb981f8dc0af456b8c2c0123fe8
daad741c72a2efadec397d926e7f64da592d7225
3954 F20101106_AABQBJ monroe_m_Page_106thm.jpg
a5772739c544415892a67f0613923c9e
278e3969910722c3bb49b51d9d7d6188ed6d7c17
4135 F20101106_AABQAV monroe_m_Page_098thm.jpg
759d6154b486ced3b394b6d8268f6be9
31958e53e6ecb43959bba4459bafb0663b018010
26321 F20101106_AABPWD monroe_m_Page_024.QC.jpg
1cecd127820ee735bafd2e65d17aecb1
f12f86930d33249f390e4d63a6759fc669f7b26f
7476 F20101106_AABPVP monroe_m_Page_013thm.jpg
a5f97bfa2754c41e92e4d435ccb89d5b
5a26ec29cce2343d9570688356037706207ca221
36272 F20101106_AABOTB monroe_m_Page_092.jpg
0c1cf9738e28c3c3146149aafcd2ff41
7939b72fa0f05d58961f7405329abbb8200b1306
14153 F20101106_AABOSN monroe_m_Page_075.jpg
709b25e6d108db20a25774e33780a98f
1921dda04460cfb29c6c7bc4f3adfb7a80e90ff1
86919 F20101106_AABORY monroe_m_Page_057.jpg
5dc1b387291cfe536c5ecdcfc45e84df
be21352d41e8926ba1dc7d28a271f173adc76127
11822 F20101106_AABQBK monroe_m_Page_107.QC.jpg
57eabebecb1e5d184599069b2a425b85
c2ca9b08bceb8c9a05d8f702ef03822933157c8a
18555 F20101106_AABQAW monroe_m_Page_099.QC.jpg
365ac30e7a0af8bbc616e914c718a1df
a8169a7edc8f0a01919c558d7fcf1b6f8f44dd33
7305 F20101106_AABPWE monroe_m_Page_024thm.jpg
0dbb8c2ed0fab96b84183e98d36dc8c9
8a2401b07dabbc478d524d0080123aede9adea56
5605 F20101106_AABPVQ monroe_m_Page_014.QC.jpg
9955ee3ec4967bf2d3f6d1851d981a3f
9ca52129216875b7dde43d78a088d32615ce0e7a
51668 F20101106_AABOTC monroe_m_Page_093.jpg
4c3f8046beb8279076ab9983aab9c20c
8ed0cc4118cafedbd91924c8a41262678bf5485c
49058 F20101106_AABOSO monroe_m_Page_076.jpg
0207164444ae6bec60beb6d0879dfc90
b58c42a08e181ed2548c20f06c49ec47e96f3e18
82057 F20101106_AABORZ monroe_m_Page_058.jpg
e919089a4be90b5ea209f928e79d9c57
d0aaf38e3797b9d7f50eedd25454d4ba1ba1fc13
3485 F20101106_AABQBL monroe_m_Page_107thm.jpg
52574b34b5f11b3463a96a2ccf1cb62d
8bb78941342f879ec9db47d7fa16cba0269a7c33
5383 F20101106_AABQAX monroe_m_Page_099thm.jpg
16f108109e6adddb0cf3ef4555240400
88fa0e22b1a16d77adc76b1dc8dfb0f8ca4c9a2b
4813 F20101106_AABPWF monroe_m_Page_025.QC.jpg
d338aa67d77b4d3fff6c4b5def19fa9e
ab6ecdfba86539a7ecbd32816f476c9f0f129c64
43202 F20101106_AABOTD monroe_m_Page_096.jpg
735c91749aaf45c792206c9d2427dff3
8ee92f264211ab6fd6d651273ef81805e7892d85
16580 F20101106_AABQCA monroe_m_Page_117.QC.jpg
ff570b2b82bcdb2ebc2bf54b15e01338
58fddce79ae1c11dd61bc1e680717b3f653fbbd4
12674 F20101106_AABQBM monroe_m_Page_108.QC.jpg
56666e0aa04b41defd3c237036c24a7b
a058df13a4649cac7214c492efd0e3301c489947
13701 F20101106_AABQAY monroe_m_Page_100.QC.jpg
50631fb4d0d98191324b1f12a0f55fe0
2a568bc3356f745baeea1b21c5d5dcd39eec5e01
1776 F20101106_AABPWG monroe_m_Page_025thm.jpg
f8837b60ef0255e856453258206408ea
2260979d3de95be1e51874683b3f926bec0108a1
1979 F20101106_AABPVR monroe_m_Page_014thm.jpg
d7b0613b53f2b6d520dc660fae51d620
c861fd166611ab8de46caa307ae1e0a5c23555fe
44558 F20101106_AABOTE monroe_m_Page_098.jpg
a6972e426cb03936bf510fa59a4f6ab7
9f958691fa67a9f7a5372776ebbde6347aa2329c
42222 F20101106_AABOSP monroe_m_Page_077.jpg
f9df33ef95727fa12e1ae6d2a0768555
0c44b6f805fe24630e16ea6314e7e14cfb6b2847
14857 F20101106_AABQCB monroe_m_Page_118.QC.jpg
14f35400e2e824ed836b0aa50bb5c282
ca0a22b3fe3ce1d470666a4d0bc769a933d27670
3914 F20101106_AABQBN monroe_m_Page_108thm.jpg
f3896fe7f67565339ccfd2a0650eca78
fe42932153eb63e63cbe75c7ea83890bc4018653
4186 F20101106_AABQAZ monroe_m_Page_100thm.jpg
c0f3b22e5d838e3edc5470838329681c
7fc50a5eb56dfbaaf63f9614dc9def0485a1687c
15828 F20101106_AABPWH monroe_m_Page_026.QC.jpg
c4b5c440687d2440cfee97522ae3d708
dfa087a45e44cf0804edc27eb091f126323875ad
10131 F20101106_AABPVS monroe_m_Page_016.QC.jpg
24b2cfe357168bccf504c55783ae07a2
b578cabbd88009dd0c855bdc1c38ec37c24b7f00
62538 F20101106_AABOTF monroe_m_Page_099.jpg
50ff2dba2c3d3dfa45f83b51c6780549
e2907b78e669afdae4f27fdcceb0c8fd0af4d697
33727 F20101106_AABOSQ monroe_m_Page_078.jpg
455f9fe612d4a075866aee3b0c9d66a1
a364e54b83539a1e81f552103aed745308e65f4d
4178 F20101106_AABQCC monroe_m_Page_118thm.jpg
1156b37cfb9c125de1067220fe54ee1d
6a9ca91217d0a2f6ba243340f03f85e1497df9b8
14288 F20101106_AABQBO monroe_m_Page_109.QC.jpg
d055a9d2f5aaeadc5d6a15ebcf4306e3
1a9584176b925edfb50c0a952739bd2fe474458f
4785 F20101106_AABPWI monroe_m_Page_026thm.jpg
03a17c749c20c6ff6797b015936e263f
3bd8f2b4fff4ab8af8d3b664bded5398ed525ad4
2992 F20101106_AABPVT monroe_m_Page_016thm.jpg
dca62272be24aeec906e8d6ad1c37349
51b8e941054109d606773fb0164d60f716154772
44908 F20101106_AABOTG monroe_m_Page_100.jpg
5519a251c183b8a55fd5f5245d0335ef
4738ad6cfb8e12cf45546c17c99a7b4f2fa9dddc
78747 F20101106_AABOSR monroe_m_Page_079.jpg
60be6a205a652817c3005d05c6a9418a
1f14bf4fdafd3c1113a9a66aa389f9f2b65f3d5c
18150 F20101106_AABQCD monroe_m_Page_119.QC.jpg
cb6a6bb95b54b9c642530190d23636f1
0ee0647ce8cd810351d6908123f62b1b2cb9e35e
4236 F20101106_AABQBP monroe_m_Page_109thm.jpg
1e41ef7e3f2eac3de765920b2a2c331d
def687f720603281198ea13badbf6c813c0ea61b
25618 F20101106_AABPWJ monroe_m_Page_027.QC.jpg
bf10f10c12ba0a4d5b62c7bfbc5fd526
92f7a1af36c8badb94258e6be6d5544e8e05ebcf
27624 F20101106_AABPVU monroe_m_Page_018.QC.jpg
c836f423c6e948e2ebcb8caa42af5493
bae7a90e292e752efc606c35037161904f23e698
50093 F20101106_AABOTH monroe_m_Page_102.jpg
48acb47edc7ef46cb5383f781649af47
669896f68928de395f1f8fd07b992feb671f27bd
58839 F20101106_AABOSS monroe_m_Page_081.jpg
32bebd1eb4df2c504c0f29418ac26caa
d4a49f923062515e9ed0d65ddb79b02a687357e5
5532 F20101106_AABQCE monroe_m_Page_119thm.jpg
771cec7b154da8fc9ce6b85fafcc65c6
6da210665df098353075fff93518789e2c33685c
16586 F20101106_AABQBQ monroe_m_Page_110.QC.jpg
5532c8cf3d5bc9b986af92a546ee8dde
4bb1731c853a361aaa50c4acc74dc5eafb8da6c1
6879 F20101106_AABPWK monroe_m_Page_027thm.jpg
fe528a019e254ca13de257a4104a6bc7
4eb2a34c3913fdb63e8a58ae93866c52bc9c8879
27925 F20101106_AABPVV monroe_m_Page_019.QC.jpg
fa875de8ba6811609516bdacf5ce777d
d4d766ed4569e10cb9022d8c4f6d2b0fcaaad294
49171 F20101106_AABOTI monroe_m_Page_104.jpg
34104c5fa1968cb88bc49154c4f5544d
a7cde914f5f132e5d93ba3169240a06fb2567ecd
54248 F20101106_AABOST monroe_m_Page_082.jpg
e5d7298bcb1c08e6c6c579884489e954
405b8cf6316927b56dec18bf39e25412906b006e
9917 F20101106_AABQCF monroe_m_Page_120.QC.jpg
5755a7fe0e2d138627485d741e92246b
6a89e492a34c54346d479b4c1ea3a50bc2f71896
5047 F20101106_AABQBR monroe_m_Page_110thm.jpg
73093c80c28bd66049fcf80fece566f1
bb57cc0ec27be5fa48818adbd6c387247b729e50
26030 F20101106_AABPWL monroe_m_Page_028.QC.jpg
c0b03e5077027adeafd7340c2b01c75b
66442cc81ebbcd6904b7d1000e59fde4917d12e5
7380 F20101106_AABPVW monroe_m_Page_019thm.jpg
e4c1518ddaad99d31bf8d571fc802a64
e12d49e498df8f8de7a49bdf761ec02e0cbbf8e7
45183 F20101106_AABOTJ monroe_m_Page_106.jpg
87193d171b753471d9ad357f21ceeadb
bc66f6800331ffffddf57d9c8f8843c85f05e8e1
57363 F20101106_AABOSU monroe_m_Page_083.jpg
3d6e6e5b1b27aa76972d6a04ed9cf644
abd46fd8659e465dde52935f3e559c12dc6e9c8d
3770 F20101106_AABQCG monroe_m_Page_120thm.jpg
ccc0e900f3df4fbbeada34a0e4cc2ec5
654c9a880b1c825febccc6bc6e433f2ad48eb806
16375 F20101106_AABQBS monroe_m_Page_111.QC.jpg
7160724c6048b6409c5b6474e09efcca
f766a438e9d9e7166e5af052ac33a3e1d42d8619
7094 F20101106_AABPXA monroe_m_Page_036thm.jpg
43ceae0056e9213621be9355df7e5a71
c6a4ea824ce380c68156d878f733a3248b3f40cb
7242 F20101106_AABPWM monroe_m_Page_028thm.jpg
ab66fba3c8b50cbb6665601f557809a6
18bce2a74c11345947db0a8f0026c89b4b3ca362
7321 F20101106_AABPVX monroe_m_Page_020thm.jpg
d45839c5582e6d6e7359fce4f880d4e7
9e13eb1c9e79352cc95096d4b6c0a0a3f133ebbf
40176 F20101106_AABOTK monroe_m_Page_107.jpg
c261b5edc751e4c85c5bf635e94b479e
f96917d21689f0ab65afc72e8c229f1f36dab730
38688 F20101106_AABOSV monroe_m_Page_084.jpg
2bf74fcf511327f3164938ec4dce80b1
ae52070e101d17c56ba16b05affa1a96d26ec0c7
11314 F20101106_AABQCH monroe_m_Page_121.QC.jpg
d44d43b235c8dff8cce5d7683f8b9154
52bbbfcd35e1b37ed858f1a6c53de9fec6854933
5588 F20101106_AABQBT monroe_m_Page_111thm.jpg
039d4e9de32d7196faf17b138e5b9e7b
f085d0a024c15c351649965beffbd730d438b9cb
26646 F20101106_AABPXB monroe_m_Page_037.QC.jpg
c23a9035e49617747ba400bece49efe3
de3beed100fbba1366e36c08b6c9237ec3545c28
26360 F20101106_AABPWN monroe_m_Page_029.QC.jpg
2a105a3a985668c117f36e17db1b89bd
8bbd167f32f128bf33e851bb69bfea0b8d291872
28043 F20101106_AABPVY monroe_m_Page_021.QC.jpg
3383f9b3dbe6e679c552732468aefd82
0958c37eb656974480a039b60f223e31fd5dda95
43727 F20101106_AABOTL monroe_m_Page_108.jpg
b2cfd9958093e03a4b7a48a69c0d4427
7cc8fa7a96ebc1839d9b8c7f635fde434696c760
33971 F20101106_AABOSW monroe_m_Page_086.jpg
cd23d4802bf9ee1fba4df01b15c6d29f
54c03b9de9a668f673fed4740a9ea377fc87f4db
3755 F20101106_AABQCI monroe_m_Page_121thm.jpg
65bff761dfebfe0b1b480b8e50fb0533
d0db195ea4f546692e341e9aef419696a48dd03f
6880 F20101106_AABQBU monroe_m_Page_112.QC.jpg
26f2f8a0bd429444cc7e2a5163cd43e6
bf851b4b9dede11d23b75633640710b2618dd814
7227 F20101106_AABPXC monroe_m_Page_037thm.jpg
6e82f4734f2478b08fc97730c25844f6
00517bfa9d014f6b665f9a09422af9924ec17b89
F20101106_AABPWO monroe_m_Page_029thm.jpg
29f3d663e9760b24fea443edc6db6f4c
179e71ea0773186e5deaf97d8c8818b18b77e22f
F20101106_AABPVZ monroe_m_Page_021thm.jpg
a7bc1592d641385c80ca75d212132485
99d985b996e769a4917b89b33435c0474d8aa4fb
46895 F20101106_AABOTM monroe_m_Page_109.jpg
6a08f57e67300d104f96a719c234b88f
ca70757d5f88d7f201bd1e775285f3118a7702ba
32187 F20101106_AABOSX monroe_m_Page_087.jpg
5f7032e6f71d6f6a228d62d6935bd6e1
710e4dc0322b926f2f8ec96f84e8a409d615be33
77818 F20101106_AABOUA monroe_m_Page_127.jpg
f2a207806aa82192b74e57af57185844
08256517b08de915b3b603ef8ebe90747e3c26a4
26691 F20101106_AABQCJ monroe_m_Page_122.QC.jpg
b64d6b79931e4dd23b7dfa045697f235
f23d8d3ff47761498ab6a7c64074df18286f8c09
2430 F20101106_AABQBV monroe_m_Page_112thm.jpg
69648e93a21cd97667f7256390b217a4
c5c043956a6bb4e9ceafbdcb0ca83c06a6fb16c2
26281 F20101106_AABPXD monroe_m_Page_038.QC.jpg
c3db69dc2483d4c18b663dd14d6e07c3
6aa4018f9fba36e85910f3dd23cf51d0e62b409c
26651 F20101106_AABPWP monroe_m_Page_030.QC.jpg
60f7e395857bfa8da1e670f541f5ee3b
cb62f66d921717affb24f4e631eddbf4c35a6a19
57371 F20101106_AABOTN monroe_m_Page_110.jpg
cc14196aa3faaa4df4c17501649c1a85
fb4f09f327bb6e3e3aa3bab1bb03b94095c5a873
43539 F20101106_AABOSY monroe_m_Page_088.jpg
2274102eb16a4c13bc7733c25d541ccd
1cdfb8dd29ed2cc1f731b21922606b93a4824628
69607 F20101106_AABOUB monroe_m_Page_128.jpg
9df6328beba13803f3b41f380e906c0a
943b61a62c27e1f93d8ec8de24e410304efe125c
7167 F20101106_AABQCK monroe_m_Page_122thm.jpg
684f726f191ef47e2d0ac64d4263a6f5
f22b8918fb9fc5bc82245096afd2ff17f6d4c804
3490 F20101106_AABQBW monroe_m_Page_113thm.jpg
efe9d043420b392287d7f43b8e90257c
6898058b4aa1ac262110cfea4ff61eda7493129d
7317 F20101106_AABPXE monroe_m_Page_038thm.jpg
92f587aeace6e55a063e2093c413d5e8
4e397069f28f3478685844fbf98c5dfdbad691f4
7111 F20101106_AABPWQ monroe_m_Page_030thm.jpg
bafcc79580c0780b853adc71118582b5
331386d1328cbf73da686465982bbb69aa6dc663
53525 F20101106_AABOTO monroe_m_Page_111.jpg
15d40b6d42c2d83bd8ea729e23566ee0
8318178de39013eb1a9b0064048f16c59780c3f2
29907 F20101106_AABOSZ monroe_m_Page_089.jpg
1a78cb28f3a0d4d0971cf61caac8a309
18c90ed0ec529f9cbf7f74daa8dc5a10a8d9d872
88283 F20101106_AABOUC monroe_m_Page_129.jpg
f790c26e60eaf0b6e39a7df7f3b8e2b2
5825149c44768257307f4cac7d0f08dca2a1e7a1
25572 F20101106_AABQCL monroe_m_Page_123.QC.jpg
4dda807be61cda83efee13945c1598b5
e5607d38ab97bc429b21964f8ed0839a51cb88fa
9149 F20101106_AABQBX monroe_m_Page_114.QC.jpg
4456ad207f1e2db53869c15166df12b2
6eb4f8fc983ce9efc297536401e9f847170d4020
25420 F20101106_AABPXF monroe_m_Page_039.QC.jpg
65dc1b9cf150bdcc6543f5b575ba5955
153597e0b392d13bc6e5b38ceb296ebdb84d29aa
23402 F20101106_AABPWR monroe_m_Page_031.QC.jpg
16648321054b2b8d565c8e164b254903
008897f65ec42877dda65e752d56b526b078a344
22337 F20101106_AABOTP monroe_m_Page_112.jpg
ed582e45ae6acb303c1b68d09d0ed4ab
d8d0bc055235aa5c5defb31bbb2eeb0b781a2d75
84145 F20101106_AABOUD monroe_m_Page_130.jpg
bacec03a96cf1aa0dc6fe591ad69a59b
5e046c344def78032c977449faeabfa0b28bfbcc
27176 F20101106_AABQDA monroe_m_Page_132.QC.jpg
adc4d43837f923b917e13b23330f47d0
26464145e467a73b6095c5f268d6970210aef39d
7375 F20101106_AABQCM monroe_m_Page_123thm.jpg
a429b8d32f386dc26d167a6ca9eca6e1
5abed19834900014f276c3d202a848153bad408e
16468 F20101106_AABQBY monroe_m_Page_116.QC.jpg
e1b08b8a774f100e05bdfbdc15e411ee
8b09e789e83c50910895e49c3c02e916cdc85957
F20101106_AABPXG monroe_m_Page_039thm.jpg
810c305cdc3d1862cd0b6c06644bb139
4638f5d5f1bd288fdc3ddea205429d3b5ee3d7fb
86698 F20101106_AABOUE monroe_m_Page_132.jpg
e342b7a39b45325d129bcde9fc1bc90e
705d4c089d3e50a76bbf284d067323243a0b9c2b
25826 F20101106_AABQDB monroe_m_Page_133.QC.jpg
11bc0bb674e609986d4c95eddb5196a0
2f277b428c236baa2e4081997ca9acafadc82f2d
26557 F20101106_AABQCN monroe_m_Page_124.QC.jpg
9b146a98461e86d25828554fb80a651b
34e04dab823d4842a1f1e8b797c724169c60dd9d
F20101106_AABQBZ monroe_m_Page_116thm.jpg
1ecbddd4e6c9772b8313451b825b7fb8
43f86f3795c59b05e54b477713eca943b33ef058
27295 F20101106_AABPXH monroe_m_Page_040.QC.jpg
d9969ad31fd18084171086d43ffd88c5
8564fbea53192cdeddf6a0360f0e4a01215f577f
6837 F20101106_AABPWS monroe_m_Page_032thm.jpg
dc98c1271be237828643e48f1e298f77
8cc92e435ff11e7ab8b80176d14b6c6dbf915146
26412 F20101106_AABOTQ monroe_m_Page_114.jpg
2f1ca555f7cdfce8cde99aa1a10c3567
3c71eb087d91f784e3f034b538d1707ffcbab9cb
81657 F20101106_AABOUF monroe_m_Page_133.jpg
c4a2a0462397d31d680e990c22fb0cb0
a7ec30d0d6a85db2953d52286b32fa2b74a993d7
7120 F20101106_AABQDC monroe_m_Page_133thm.jpg
ce9584f5529a6b7094d274912585708f
47c457efe088711a4c778835baf034b0c0166366
7288 F20101106_AABQCO monroe_m_Page_124thm.jpg
e13e5a9bc14068cf6a04cfcb64c57cc2
40305a3cb249cdfbc2547a2784b3435612e47d9c
7360 F20101106_AABPXI monroe_m_Page_040thm.jpg
223585c29d42ad41220bab3784699b2b
77488e4b9c22a0f0e2ce3ebf917abadc6214dac7
25946 F20101106_AABPWT monroe_m_Page_033.QC.jpg
c980e25204a7903137ff9f67fa69210e
4f04203aac7a31b557c492f819c9ada3ec2a9087
47911 F20101106_AABOTR monroe_m_Page_115.jpg
729cf601df987df400639f7b9f533611
8a7cce3204b415cab625dc428ae7cc1dc2a31be9
F20101106_AABPAA monroe_m_Page_124.jp2
e21b7b7490d18597cc1567ecec5215d7
4da2bec9d07cddaf82e171c6312da602023c921e
87171 F20101106_AABOUG monroe_m_Page_135.jpg
4ab8949f711e8a0c068b7640df73fcd8
ca286c9f572727531e86457ce394f81f598131a7
24683 F20101106_AABQDD monroe_m_Page_134.QC.jpg
c1e25f652385f7d7d179d1854e9b9d00
feb76fe97140ea923b335c94e0245f4b44e2366c
26037 F20101106_AABQCP monroe_m_Page_125.QC.jpg
83bb3609dfd87e7d39640ddf444c632e
f55718157590a4bfe10463e74fafba1b2e25f48d
27434 F20101106_AABPXJ monroe_m_Page_041.QC.jpg
52420cebb06fe5910a25b67578e405ac
dbeb9cbf23ede08483ad59d5ec43607c8b0f9b0f
7278 F20101106_AABPWU monroe_m_Page_033thm.jpg
4bdd3527d6d35956578fa8e59a07fd7a
fc8f1112fc483b2ce2df0dd5bc3438f30ea3c1f2
52080 F20101106_AABOTS monroe_m_Page_117.jpg
92cd37b02266a35967c39df330c75ef4
759189739aaa50af23a328360733de59c058bfc4
F20101106_AABPAB monroe_m_Page_126.jp2
54a477b0bce8ac82a117f077375043a9
aeed9d37be6ea08996977996adbac5decd9c41c7
75489 F20101106_AABOUH monroe_m_Page_136.jpg
9ceabf7e6a33307e75b6da22fec62f65
2a8502ae308a42845f5483857594b47e2d7ba53a
6980 F20101106_AABQDE monroe_m_Page_134thm.jpg
aab467f96387f15b8c8c0cf863036a0e
5fd3db5d6b00ca35adb0fe20f2de775d2b5b5c52
7309 F20101106_AABQCQ monroe_m_Page_125thm.jpg
2c68323f236da74b57800eb6c780107b
bc2d150b0883986ce1274572c887b283c69ad560
7579 F20101106_AABPXK monroe_m_Page_041thm.jpg
b3add413d7f456219748efb82ea9baae
340be4b1f6e1ad372a2aa85d43d8825266ecac17
26159 F20101106_AABPWV monroe_m_Page_034.QC.jpg
4d0484abbd487f1b3492b60c15dbf5fb
4efc13cbbf8500ce731a0c1a6482cac6156d54df
49096 F20101106_AABOTT monroe_m_Page_118.jpg
0736cbef0455171a5b65feecf1f2c950
f16a134ec716521e9106f15829eb8b208af572c4
F20101106_AABPAC monroe_m_Page_127.jp2
fcaab0082dd233451d3bf0a5411fc45c
f22865bbf57f1db004370fb36ee79aa9a0fd92e0
85126 F20101106_AABOUI monroe_m_Page_138.jpg
447a8e69d4593ef430fc964c3ab2b028
b7d2a2e1f152e2ad624fda32a2817f08100557cb
7387 F20101106_AABQDF monroe_m_Page_135thm.jpg
4b9182b07393ba2721c3fa1be5d763ef
ec89859cb04334bdaa491925b5f3305c2d2bb3d8
25719 F20101106_AABQCR monroe_m_Page_126.QC.jpg
dc5f0b66050b2e3a977941ae91466ca9
cda1f6a5427b1a285003e8bd9b2f70fbb9f24fae
7600 F20101106_AABPXL monroe_m_Page_042thm.jpg
e7bd2f649e1593c69a5a04da745223d0
22d33b1890be6620aa021dbceb0e75444e225a04
F20101106_AABPWW monroe_m_Page_034thm.jpg
6d20cfb97fde50347c53179e545c15d4
4b047dc613901cdb050a39aee4f12eb4bae6b974
68325 F20101106_AABOTU monroe_m_Page_119.jpg
a1a58bd2f9e753cd324a83ac9ead5311
856f60559c8f8d27b8295a907453f3d369268968
939587 F20101106_AABPAD monroe_m_Page_128.jp2
b89ba3f3891a573a50799402cafb8ea0
fbaf2bbac033232c837c1e1e25cf057fde5b5dad
85521 F20101106_AABOUJ monroe_m_Page_139.jpg
400553d7602972c3298777f1ac77593d
76dbedbfda4099ef996f5a0d39cc5de87dbf2c8a
24268 F20101106_AABQDG monroe_m_Page_136.QC.jpg
04bfe941afa816643ce0bc4412fd4fa3
9ae1d2278d78eb30215c458e4e8595889e4817b0
7291 F20101106_AABQCS monroe_m_Page_126thm.jpg
0f64c371fc550341b1103072a7fbe95f
4530c9af64eb60a2cf51aa038c90c52fe1c1b2c9
7170 F20101106_AABPYA monroe_m_Page_054thm.jpg
0325149dea46dcb3f99274724e3520e3
54f3dc881410743c1750f79bcbbb35ab37780582
7275 F20101106_AABPXM monroe_m_Page_043thm.jpg
9d4b8ebba977ab0c9da5c300791dacd6
1dbfeda137a3e6e0b0dd8d6e826ace029890bd74
27190 F20101106_AABPWX monroe_m_Page_035.QC.jpg
45ea509a85d27606a33eb767e7a0a214
e42b50b77edbf2e62bf8e82dbd35594a7f88650c
30823 F20101106_AABOTV monroe_m_Page_120.jpg
445a65aa54a9291ccdced824725247f3
b3befe87768fa1d7edcb7d455c1508e152c1d60b
F20101106_AABPAE monroe_m_Page_129.jp2
9f88ab9041f4a41224786e448a77203b
52011d29ef5eb106a01c1bdb1e898f2891f09fd3
79751 F20101106_AABOUK monroe_m_Page_140.jpg
12119981addaba3151392eec280fe72f
eb3dbe7738a1a5456310b99c7c1a28c7c63ec041
27745 F20101106_AABQDH monroe_m_Page_137.QC.jpg
34d58858935712c5b5123043b3e410aa
97e52af437278bd62f6040047ea0938b8fc728d7
24474 F20101106_AABQCT monroe_m_Page_127.QC.jpg
666cc1adc7239e4ec9b1221e17c3b031
e94bd579b785edc63c8680e0b194eb2372a7c27b
7144 F20101106_AABPYB monroe_m_Page_055thm.jpg
fb7e9021bcb784f5988288efd6dd6c82
71802cf3dff422c8bc7646fda1df34d727ec2ddb
7102 F20101106_AABPXN monroe_m_Page_044thm.jpg
3132114d56d765955619a54f6a2b678a
1d8a2ea9c3ee111bb1a848adbfaaded4d9e4a49e
7415 F20101106_AABPWY monroe_m_Page_035thm.jpg
9b882068e8735cf327f0b25f009147a8
9ef83d997887ff675e9df4e277e1a7b81e6d12bc
83954 F20101106_AABOTW monroe_m_Page_122.jpg
359697c748d2af0dc005d89688ca85f4
ccf26e71c2216ce0c1df02e0e7c579df5c8a000a
F20101106_AABPAF monroe_m_Page_130.jp2
d993ea7a7aa3ec35f2cfc6af4e696b48
33e5361c1a466933c1baaa5768e3129cc23c6c69
85861 F20101106_AABOUL monroe_m_Page_143.jpg
98943eb17ba4ea9edd808053b2bc0206
91d44f7c901e3fb456cdf7443478276c423c7c1a
7532 F20101106_AABQDI monroe_m_Page_137thm.jpg
c78e9e986ac74c32c5ef3f423b90ee44
8e0545e00be24824dee671cad1758a3de2db557b
6882 F20101106_AABQCU monroe_m_Page_127thm.jpg
ba385a1ec9866d5848c660317b76fcb0
522072dadcbef8c7cd9ef3b7157bda4afa306842
27842 F20101106_AABPYC monroe_m_Page_056.QC.jpg
4796117e1950000cbd43986f02c818e2
b49469944acf9c5bd1c887012ca6900c5219531b
7124 F20101106_AABPXO monroe_m_Page_045thm.jpg
c12060554db9962acbb445b85af5e8b3
5a7fdf0fb7bc6ee6519d78a9084a9a52c43fa839
26586 F20101106_AABPWZ monroe_m_Page_036.QC.jpg
72fa6ef0093841127589014dd8741b6e
0b7d742ee9ff3497222c6593b205f88ddb05ec08
81067 F20101106_AABOTX monroe_m_Page_123.jpg
3cfbb6d8f621756f2adfbb683f97da75
deb17d738d36a36af8c5abc085a5aa7d52fbafff
F20101106_AABPAG monroe_m_Page_131.jp2
e3fcbd31414eacf4b9225b7b368e7eea
f396b9398930fa8cc531393be57972364b39fac7
53262 F20101106_AABOVA monroe_m_Page_162.jpg
6845d2edfa9d93545bf20b7879f445c7
12806ad9e75d7a9a437880c96c042800c796d3b4
84185 F20101106_AABOUM monroe_m_Page_144.jpg
af0e621e1a007119c36c12b3216b165b
6e4f322961bbaea2ffed8d965f9ecbfaaef1dbb0
27209 F20101106_AABQDJ monroe_m_Page_138.QC.jpg
79c53c85e0170a39e6d6ef99cbf90cab
937a75fddf7a216f6277169961902af566ae5e32
7829 F20101106_AABQCV monroe_m_Page_129thm.jpg
239efb78e4ef7224b47027d63f2ff177
8e248cb14eac62f5f04c9075cb4e2c44afd2bba5
7460 F20101106_AABPYD monroe_m_Page_056thm.jpg
b74c3201dbf37ed5d1996e3adcd11c92
55245aac4e5f9de59aff22eeaf9d66bae462e87c
27974 F20101106_AABPXP monroe_m_Page_046.QC.jpg
a642f808ec06f32cfcbc93918a70d2fd
d7ae5637e1c5468bc1e170a8d11b7ecab5f023bc
83936 F20101106_AABOTY monroe_m_Page_124.jpg
bea9cbd6d2465a63e38cb857e0c8d941
e181461343ab52fd42ca8aee8c78fad582d09599
F20101106_AABPAH monroe_m_Page_132.jp2
43937f6210e10aadb2306d54ab0c54cd
4e14e1369add00965ef858121c92eb0c0babda3a
35637 F20101106_AABOVB monroe_m_Page_164.jpg
f8d939f3d14b4741cc06cf67dce04087
71ba421c48b8f2dc4929f93e66a1ce54a27e425a
85496 F20101106_AABOUN monroe_m_Page_145.jpg
51b829c494930276d52c653aa216e2b6
dd747be3cc4f6b128e7a912b8720b866d9ef9781
7481 F20101106_AABQDK monroe_m_Page_138thm.jpg
395bb4ae2c8eb1586afcf2aee42e77a6
c4504f42d0fd8a82ecb21ac1938cbd1a47e5fa42
26341 F20101106_AABQCW monroe_m_Page_130.QC.jpg
37e69419fd4d0bf81aff4688cfbcb1f5
9ed6b65accac5eeae2db72d73c0c230101ecd74d
7491 F20101106_AABPYE monroe_m_Page_057thm.jpg
e4436bb428d6a198d33c35f47679421a
5f874a4daa8c9d0537bb6a119f92572d800a9b73
27990 F20101106_AABPXQ monroe_m_Page_047.QC.jpg
5fb7902af76e0492c9a5a753e985527a
6e735a8cf4eeacb53a4a0cdabc84991fe67356e8
82955 F20101106_AABOTZ monroe_m_Page_125.jpg
daf223dec9f0d10e9a0dd2a35a8a600d
2ffd22c62b9b5029c53a6b7e2caaf72ade20408d
1051937 F20101106_AABPAI monroe_m_Page_133.jp2
4283abe624abf76374f390d770ae5060
4ee79b5bb23cd017e08cbdb53ed24b3ba67b80d2
50161 F20101106_AABOVC monroe_m_Page_166.jpg
20b950b8dc1b140f80ed4d6449668ba7
d0efc6b8761d8abc3595602247d7e39d5a63ff5c
84509 F20101106_AABOUO monroe_m_Page_146.jpg
c090d8dd599cf762adbfb52a4cc4b996
2b0384f6b8a2bf7bcaa09ebbcc925bd555f80dbb
27276 F20101106_AABQDL monroe_m_Page_139.QC.jpg
e63940bbf4dab9bfd0a17a0873dfe0ab
743baca0c3afd21a05f4cd6d1f472dc3a844102c
7352 F20101106_AABQCX monroe_m_Page_130thm.jpg
662fc291f1b1d60efd234dd094f774b6
50b5b3ac95f37a4a88f2b1a392360dd5c4338d21
6957 F20101106_AABPYF monroe_m_Page_058thm.jpg
d95b58722749ea14f99a31bbac1662e8
b2d6290316d16b6afd6fb787a44710a9a1966fc3
26424 F20101106_AABPXR monroe_m_Page_048.QC.jpg
97ada0f726242453beff7d7c0af61c52
2ea86e90e2a7a845d66ffa1a8b788074dd49d02a
1051947 F20101106_AABPAJ monroe_m_Page_135.jp2
78f4fe31ea43b3ede987d35e3b8152c3
d9235841c6396cec391f5150daf689179e4fc9f3
38803 F20101106_AABOVD monroe_m_Page_167.jpg
92c468eaf3c509702d7ec2fe5159b781
45bb574a1f8c13ca76565768b957aefe6b7b666e
85360 F20101106_AABOUP monroe_m_Page_147.jpg
5163e3738d6aed97ed1946f561df8e49
d6d61656addb68355025e11eff1756cc8d9c49c2
26563 F20101106_AABQEA monroe_m_Page_149.QC.jpg
737d8671782f0a1afb194d0849fe9e25
cb58473538e4c2bedb421289eae24cc8b8bcf0d6
7486 F20101106_AABQDM monroe_m_Page_139thm.jpg
af938801bd9b1eb051fb2ecb8ea60cb8
b5c3b431811fe39afa697ca44cd9808a87aaec7f
26364 F20101106_AABQCY monroe_m_Page_131.QC.jpg
7c5f171018c6dd79f6301b30c19a6b3f
876388b4980a3772123ed97547a89197badb0d4e
25021 F20101106_AABPYG monroe_m_Page_059.QC.jpg
c99259c8e8acfe7e98e4d8b991caca18
6ef184afb3514cdcd763a0ee1a972c7d4084cc90
7108 F20101106_AABPXS monroe_m_Page_048thm.jpg
9b2e12bc65240453cc96a333036cbd41
3b97260729e418515cd5dde1da14ec1da05ce938
1050694 F20101106_AABPAK monroe_m_Page_136.jp2
f64e5fc0e3e08bf3008936eebeddf08e
e936232b6407481d5483b2675489a1c42a1702de
34744 F20101106_AABOVE monroe_m_Page_168.jpg
6e1280fb73005557493c7e5db8195afd
daf825aa005cc7a4260f22a0a101ff8cdba556f9
86513 F20101106_AABOUQ monroe_m_Page_148.jpg
0d79d6e9513cf28b5eaa2702fc3b50aa
b0c6091afaa1ba7bb9636adecc1c6d8d084e7929
7196 F20101106_AABQEB monroe_m_Page_149thm.jpg
e80c9f164766352ed5c94ba13b247451
6e86a2d340fcc02d35471dd55201c9bba2e53a4c
25739 F20101106_AABQDN monroe_m_Page_140.QC.jpg
a6b1ad02257319818f8f5d41e9080a05
eeb05a03098deecb3fcf8328f09a34aff41b079e
7322 F20101106_AABQCZ monroe_m_Page_131thm.jpg
040b4c57af6142caccbf34e8855ea052
445f973a3a8ef2767557aee3115c3ae904361ca5
F20101106_AABPYH monroe_m_Page_059thm.jpg
f22e1b3a81390e2bcc0209a43c345e0d
f51de217fa312391e8ea1882fbc11b696418323e
1051960 F20101106_AABPAL monroe_m_Page_137.jp2
5c328bcb82015420e44040542607a7a0
e7e44ecfc415166f9604feb72f6d3a59883d06bb
74064 F20101106_AABOVF monroe_m_Page_170.jpg
3d4f3387e60d6b0b271a5a03582d5202
169cf25a97f2517aef3943d904dfd78e8e7b6e4b
27225 F20101106_AABQEC monroe_m_Page_150.QC.jpg
f6b94e88830ddde50e449e974e9eb744
fe49446b1b525f040a6e3e34ce0a57c9732ce8e6
6808 F20101106_AABQDO monroe_m_Page_140thm.jpg
9e77625f3a199d63c7c64d6aabb99ed0
963e11f8f60985f43a18d1aae2551225a288ea9a
26590 F20101106_AABPYI monroe_m_Page_060.QC.jpg
756009bed09dd92c7155674615c73379
2d1cc6c9d11f4ef93dd581900e18befb44b709b0
26812 F20101106_AABPXT monroe_m_Page_049.QC.jpg
27fddff30f19f46fce04c9baa6bab336
31054b5e6e4673f98f1634e152fb252cc6a669ea
F20101106_AABPAM monroe_m_Page_138.jp2
ce16a59abd33c6ac0d6383636da51e9b
5b344cb14ae8fa982d2232f3378a3bc1c8a10bf8
26882 F20101106_AABOVG monroe_m_Page_174.jpg
6fc726da780c89b70a8c1f1313da65ce
086df994294479bc9218e51a955f6df3cd63a022
76076 F20101106_AABOUR monroe_m_Page_151.jpg
645c05371db68b04313ca16e32a27703
8e8aa6f3d7e7b3920f357d373e4a44c8590b367a
58350 F20101106_AABPBA monroe_m_Page_155.jp2
43440d78bc2e0f30aa363b711af2baa7
8cf353f3f8cc7487bcda57058ab1adec8e1a746f
7262 F20101106_AABQED monroe_m_Page_150thm.jpg
5df67bf67bc81e143f55bca55be963ac
6352739f8a7049b7a3da3cb414b78c7f3ef77b3b
27033 F20101106_AABQDP monroe_m_Page_141.QC.jpg
78028ea3966c54e841dda9b8483237fa
fa54804d9f0d653027a1df0fc823aa6e19779abc
7293 F20101106_AABPYJ monroe_m_Page_060thm.jpg
72b63dae75f264e15506613029f43184
6f733e3dd204160f1fca28e2b8bae5a8557cce80
27987 F20101106_AABPXU monroe_m_Page_050.QC.jpg
28b347def0956b7c8ccbf8796af9fe73
483886f7ea0ca1e6a93f3b2e4ee3e2902a689bd2
F20101106_AABPAN monroe_m_Page_140.jp2
c5ea9a8e728d814e424a1a205ce2c71d
7e6bad6a322263cfc30f6115ec53f278a2ba8ad9
30642 F20101106_AABOVH monroe_m_Page_175.jpg
7470c7ccc9f5bcbc1a58efda4f67fe2c
15999f903db138dc274adaf4b9da60478b9e8a40
75949 F20101106_AABOUS monroe_m_Page_152.jpg
6e702565b45ff094d573b10bceff8b90
76279a9d07f046967242796eb9040fe238734418
387104 F20101106_AABPBB monroe_m_Page_156.jp2
e472e5d985921b732699d262b1134674
d5f0a33ce8b8ed2ba057434fffe89706d251f42c
6876 F20101106_AABQEE monroe_m_Page_151thm.jpg
fc05439407a97131d753b6d232d50b35
727b3ad041fced313aabc9cc8ffac04890047b98
7509 F20101106_AABQDQ monroe_m_Page_141thm.jpg
1cb38a677617120d51ea0f6ce9516549
b40c00a2a560db482cf4f1bb1756b50a75ebe493
27510 F20101106_AABPYK monroe_m_Page_061.QC.jpg
1963c49da13ba7e93d52c8ac33ddbc83
3e876864b352f221fa6b893c2a3ca155a8779820
7597 F20101106_AABPXV monroe_m_Page_050thm.jpg
48d50765c050c59e75956f3e9f01c412
87cec443580b763e868c1407c8b63c47c5a593a9
1051945 F20101106_AABPAO monroe_m_Page_141.jp2
fc3e0285a342eae56e6df24a099a6e0b
2753d4166620d2f13c329cce2c3ebc09ab0df731
54672 F20101106_AABOVI monroe_m_Page_176.jpg
7aa56f257ebf416197a549c021a5e5b1
1fd2fcde9267b31cb860d95cf3192e0629ada253
81765 F20101106_AABOUT monroe_m_Page_153.jpg
6cf8a4d795fc9bff7b6585f7ee24ddfd
2a9cdcc8206b10f4ea25a7a125756d06bd10cf81
341805 F20101106_AABPBC monroe_m_Page_157.jp2
f420285c7c518a9dd9514c5bedfa68f9
5fb7862d962e51f35fb4e283e68b1d1f55ed8b23
24231 F20101106_AABQEF monroe_m_Page_152.QC.jpg
51a4ce1bdb6709a18af0f2e22240f40d
da33df06b4738f4cce60a9af4850e00cd1af9bd6
26595 F20101106_AABQDR monroe_m_Page_142.QC.jpg
ac331fc715e7f1fa619e82020b55773a
8338e27de413c26c41235f1fed4a8221f1672c82
7392 F20101106_AABPYL monroe_m_Page_061thm.jpg
d3f2ed2fcd4afb3659cf9559ccf7c3ce
bf716ff5d28ceec6a7fb2072348022e1dbf63d34
27774 F20101106_AABPXW monroe_m_Page_051.QC.jpg
03716efcd581161be7a6214ae2a90fc4
7e047536576a6cc19e64d1e0eafc7668e192ae5a
F20101106_AABPAP monroe_m_Page_143.jp2
7f84768000fa7146881854dcebd8d274
551e6d3fcf8819cdea451c6a954246293413c478
31606 F20101106_AABOVJ monroe_m_Page_178.jpg
1f1feae6cfe66f78c716b94fd59467fd
4a344b56a6f0a0dd811e338cb890606969d98e86
51156 F20101106_AABOUU monroe_m_Page_154.jpg
53818a7c40cf2034e974480969c16727
d40103022155b928a0a174a5da529f0bac390cc5
469794 F20101106_AABPBD monroe_m_Page_158.jp2
bff175d6707325efc65f23ac47883367
beb2d480aa62b46a1613333f72be132d6e3c37fe
26474 F20101106_AABQEG monroe_m_Page_153.QC.jpg
26d48a1c8e14aa6df41c12198342363f
625ece95b77c9a25a5ccc0853688e8107b17aa66
27162 F20101106_AABQDS monroe_m_Page_143.QC.jpg
ef16939f357b228cd6fc14ebc6cb7aad
3b1ccc1cbaf207ac54ba613f4d92186798bbc5f0
25180 F20101106_AABPZA monroe_m_Page_071.QC.jpg
8e71cc321a46cc277c7f55821f0ee5b7
f982b0788bf8b804f564c69e52d1491cd3d7c033
22127 F20101106_AABPYM monroe_m_Page_062.QC.jpg
f41f16c2339055c080fde045e341f987
fb6dbabf191440b947b020bf56474c95571af91c
7514 F20101106_AABPXX monroe_m_Page_051thm.jpg
6f68808e05a5fdcc839633adaf624c3d
93ea7e5322325f8cf08cfa0e4e6770b3f1cac6e0
F20101106_AABPAQ monroe_m_Page_144.jp2
46301460bfbedc55ee66de0902714745
fb3d68d6b8331c39652b754d534e96f8b28d5d46
39431 F20101106_AABOVK monroe_m_Page_179.jpg
1cb864071cc5f0c06127e42e6413fd4c
9e5f5b499f78eb96eab0f976a316ff2592cb6a74
48743 F20101106_AABOUV monroe_m_Page_155.jpg
4bc5db07c26a04e26604e788783fc59e
6a0399a112b97efaeae94440e4d616705699041c
575680 F20101106_AABPBE monroe_m_Page_159.jp2
94be3357686dd1206c40c0bb45709bd9
559bb9688841cb882c0c59045e8f9bff53b10f1a
7191 F20101106_AABQEH monroe_m_Page_153thm.jpg
984808f5a9d409d2b073ea8f03898108
43f940e9eaddfb4bfaacf52c324fecd6841eaaf8
27107 F20101106_AABQDT monroe_m_Page_144.QC.jpg
1a042956dbb64482665c6de740faaa10
4726daf54a3a2aa6dc96ff6daa07a6b9e0d426ca
6940 F20101106_AABPZB monroe_m_Page_071thm.jpg
c33b1eeacb04c135c10031e0308e6a39
fb832c63f54aa0efcf2da60ff6bb6d50492a2b47
25783 F20101106_AABPYN monroe_m_Page_063.QC.jpg
20e1bcdfedbe523d329956aec4af2f72
868de7e5d6c9a2dea51d9f05861b4b6e41718842
23171 F20101106_AABPXY monroe_m_Page_053.QC.jpg
8865c3bb1925774c1bd3291c190fdc02
d06ce046b7c68b62b4ac0c5a0ccffee9df094ed0
F20101106_AABPAR monroe_m_Page_145.jp2
642b238c82e84b43a6cc089fb19000bb
753791478d3b6dd3168b6172ae5e4a39b93a7471
47142 F20101106_AABOVL monroe_m_Page_181.jpg
b4422c77f6711e84d573af1432284915
06a231a44e2541e295226e3d9144822ac382433a
37089 F20101106_AABOUW monroe_m_Page_156.jpg
83e951a219206d1b90ae72a0c681ca1f
3d6ec14e8fd8c219fe507d247036048d323add75
632306 F20101106_AABPBF monroe_m_Page_160.jp2
4debae55f7c0da0882b8efb1aeb02155
3a4bc912b2c380b9067254c4103427c911addb92
16814 F20101106_AABQEI monroe_m_Page_154.QC.jpg
7f96a97c8df0db514884821b0775a694
a7fa3acb1adbc53c78c96fa51dbd114bda5d53ea
7462 F20101106_AABQDU monroe_m_Page_144thm.jpg
755aef674f2ab8735eb56f2666c80bc5
c0ad4e10634898155dc13d82e91393ba5653b539
24837 F20101106_AABPZC monroe_m_Page_072.QC.jpg
e7596a6261d75771c28023d0e1136305
da9b9bee8b4ee0b0d2b777f23e283dea18907fa5
7116 F20101106_AABPYO monroe_m_Page_063thm.jpg
c4d77d91e0b92592839221824872f6f2
a1d2dadc6a462021449fde7b4f2a36b9cd3cc244
26980 F20101106_AABPXZ monroe_m_Page_054.QC.jpg
b8f5ce81c56e7eda05f23981050ffdfe
12e6492bd9bd483b470fa6f322dded8a438bfe2f
79410 F20101106_AABOWA monroe_m_Page_202.jpg
9b89aa2b2eab616a07c42f6dc5971ac9
da58dc47c197a30336c6dd7abe8c1859a1317ab8
1051918 F20101106_AABPAS monroe_m_Page_146.jp2
eb645280f87a437a18355aa8ea73ba16
e18d9aef2f45b35178eb4bbe25b93573c35cf244
50861 F20101106_AABOVM monroe_m_Page_182.jpg
848573a71eabec274082f1c0f42a597f
c8e2d777c2c99e2e18960b41571f148af0c4212a
30587 F20101106_AABOUX monroe_m_Page_157.jpg
b6e0e28c327130e6ed975de7211e240c
6ae5aaeac751244cb8eea7ebd5201affd640a249
734171 F20101106_AABPBG monroe_m_Page_161.jp2
aa1b77fbd43de40af912382c0f833c9e
2974452812676157ae480f6a4e96499c3ab3d0f3
4623 F20101106_AABQEJ monroe_m_Page_154thm.jpg
bfc8dc511fd59414ecea370a65fb8eb3
920e8bffb7c8489d70703909173a67156a9c7e37
27606 F20101106_AABQDV monroe_m_Page_145.QC.jpg
7ac7bce778e41142cf91b7bdab65d6a4
927409eeb3c623f89a9a3bfb4f837051ae95f03a
F20101106_AABPZD monroe_m_Page_072thm.jpg
3e1132185121de386eca5911724275a6
10f62b4ee7bfed6310416f72d5b555064ca288df
28427 F20101106_AABPYP monroe_m_Page_064.QC.jpg
7ac0549262842d98ff92eaaff5f2ca30
b6abd8b3d1b815087d69fbf14616ecf6801b9874
76120 F20101106_AABOWB monroe_m_Page_203.jpg
19acad293580779502a2895e329cdb20
1cc338ad65a488d679fe34373cd984b731fdc060
F20101106_AABPAT monroe_m_Page_148.jp2
c4f610d11820d20a0c530536763de010
887837ba092838905b5b7e92ea5e347ad0db45ee
41839 F20101106_AABOVN monroe_m_Page_184.jpg
8bb438a8750f3edb579023921d2ed8f9
52ec5904ac5664c8dcb98c440202376f9c7c7011
49922 F20101106_AABOUY monroe_m_Page_159.jpg
d0f61dc09f099ae185f6cfc242d9f2b2
cde60639a32fca615c6398a1239a3a1d667b2e6d
509172 F20101106_AABPBH monroe_m_Page_163.jp2
2ab3554c575d163c306e9b9a8af5bc4c
cad1345ce2403ecc47a97e0a807a6b2c8e665e8d
F20101106_AABQEK monroe_m_Page_155thm.jpg
49523b3434d25a994b80813051dd70ec
0b7772765537edbdf548cc9bf2ceebbeac3eff2f
7311 F20101106_AABQDW monroe_m_Page_145thm.jpg
1f21cc64d8c2d90086b68fd7a791adbc
c4273becb2058937ea2f02acb58d103f8bd1d278
7267 F20101106_AABPZE monroe_m_Page_073thm.jpg
fc266ec715873d8eb1fd3519178877c5
b675be23576c91431d8034bd65f319f394199020
7611 F20101106_AABPYQ monroe_m_Page_064thm.jpg
8ed774d3f961e3bda93f996934b079fb
773d180a6c3427aab1c542258c45e957b354c420
82749 F20101106_AABOWC monroe_m_Page_204.jpg
83c8219a5880decb64a50bd6db1154a5
a6377b5f429c8f096bf7349c78d4475f35c60479
1051906 F20101106_AABPAU monroe_m_Page_149.jp2
3c121e9973780119fd262189311937a5
273ea5e926ff3b96080523a66b117baab9d8dac9
28897 F20101106_AABOVO monroe_m_Page_185.jpg
1282fab02793219a00751f33831e3b21
07d7f200cf800bdeb925613c8a55d9ae27abe584
47961 F20101106_AABOUZ monroe_m_Page_161.jpg
14ded6a78666e110e2fcfb5bea92bcca
e2819fc57565990b6cff0a6c4df921ae2b6bb8f7
540149 F20101106_AABPBI monroe_m_Page_166.jp2
47408c3a9752c22a07d1d85668b2e205
1565009af8019b07da41e67e2f1dee757a41a738
12167 F20101106_AABQEL monroe_m_Page_158.QC.jpg
e2909b1f2e7906a793e5c66398f1acd6
367c596fc8730985e191b5ce53efe799e013df8c
7394 F20101106_AABQDX monroe_m_Page_146thm.jpg
636050bc6e0d5e4011a212b3fa0a49ed
9d87a6abf1229dcf5a2e55298de83d8855c96bbf
26682 F20101106_AABPZF monroe_m_Page_074.QC.jpg
62daebf6736ce4c9a0f70c0f3b34ac65
4cf847200beecb6a42eb658a58a3e94b20b1ae4f
26138 F20101106_AABPYR monroe_m_Page_065.QC.jpg
7de86e562afe6a8b5421c583d1986807
6b397b4d62fde866645621585bfe1f4904cd02c4
27008 F20101106_AABOWD monroe_m_Page_206.jpg
382bfb339b3028560903f300d599f6dd
c1be5bb4aa26f9aeb0940add080d5e72671ce0dc
F20101106_AABPAV monroe_m_Page_150.jp2
8d757be2f4cd06eb19a948bc60b224b3
5218fbe2f7eb975df5c8f6246a8329c8f348177e
40456 F20101106_AABOVP monroe_m_Page_186.jpg
ea08067eb5e490e0e3d013f14770cef4
769e9402be0b43b84157f700bacb79abccd630e5
428743 F20101106_AABPBJ monroe_m_Page_167.jp2
f9eae5403d33625128a9f3fa9473c92d
74539ce6e3e078bb2bae264ac4b692084cc54449
12030 F20101106_AABQFA monroe_m_Page_167.QC.jpg
ca280a47340b1a10c347629580d1f744
e722e1116a56c9c5f3a2df348eb5c02970fb57dc
3847 F20101106_AABQEM monroe_m_Page_158thm.jpg
7c8709a4d1d622e656e57de762b34727
dcbc18de29aac10ebd5a7617d83b1f48954b1f06
28091 F20101106_AABQDY monroe_m_Page_148.QC.jpg
84c8bd47f949fcd7e897c3ff88ef5823
033b97cef5f627574c53689c2308c42b5afc09a8
7118 F20101106_AABPZG monroe_m_Page_074thm.jpg
f1d62a430186d43b9167c96a19368a25
00a0b1274e67db48a9404e4e8ee5d8828a6246fd
6979 F20101106_AABPYS monroe_m_Page_065thm.jpg
82c834ba5da45a1d670bfc9339d6f441
d95699607f8bb611b61d3c5dbb521734a7b75425
237385 F20101106_AABOWE monroe_m_Page_001.jp2
f20e1cce7cb7b2f2a907430e7bb1ffb2
4e559624078e6ded65381e41c16f81265be3a4a2
1037634 F20101106_AABPAW monroe_m_Page_151.jp2
c2ac520a050eba65b1d085c03b4e179c
3ce89a3270bd4b2fddb2aa5292903f6145868fd1
32831 F20101106_AABOVQ monroe_m_Page_188.jpg
a82372a60c131de68d1ce8283806142d
6bf25c2a6403d9fbb50fd5a77a7fb3d2878a051b
394268 F20101106_AABPBK monroe_m_Page_168.jp2
87f37cf9934c9e3a0caab94c660406cd
479f3fe44c2277b8ae343cd9052c2ff1000e7b0e
3947 F20101106_AABQFB monroe_m_Page_167thm.jpg
526c3f2895688b8bc48bb5315c4028dd
f774d2570a609539593d3c317500f211bf057bba
15234 F20101106_AABQEN monroe_m_Page_159.QC.jpg
160f806b3b47cd0fa6558068c73aeb1d
24c5159fbc14a9bdabe29bdc65f5932ec1a911e1
7631 F20101106_AABQDZ monroe_m_Page_148thm.jpg
3edc9ec2cadce4b34a3306f3e7e37179
82a1e491834f38639bbc90fda93946ec96d7668f
4803 F20101106_AABPZH monroe_m_Page_075.QC.jpg
4d00710034c2334f36114138e65fbda9
582fa9d9004fcd59d4296528e19ca412f90722c5
26399 F20101106_AABPYT monroe_m_Page_066.QC.jpg
bdfcd4b2df1b5275338a167d24667d0e
8cf44a98f3f3e5a909242be57847ba886b903775
27647 F20101106_AABOWF monroe_m_Page_002.jp2
9e6eb44c92bb3303a5c0bff3166527f1
00bde55bad6c1c1b541cfbbaffd8ebee003fac1c
F20101106_AABPAX monroe_m_Page_152.jp2
f4d0612abaa5aee05fd892fb9a6c2eed
8443c346350412f51dea883bf6b3e382e724ee29
47886 F20101106_AABOVR monroe_m_Page_189.jpg
3d1b5046dcb7005842ed573b3bd2653b
c38aa391f865c97b3175b53e7f4cdab40045a171
397898 F20101106_AABPBL monroe_m_Page_169.jp2
cf6148b5c2522524d547482a85437d9e
523dc32ea6a57554f503795d274ce6496b2822f4
11805 F20101106_AABQFC monroe_m_Page_168.QC.jpg
831e47112565e9d657a938512a2f4c17
c669f61cf7b8b9794bea94842cd6924a499cf0c6
4456 F20101106_AABQEO monroe_m_Page_159thm.jpg
45a764f2134c79c6755db677bb19ceb9
4d17f3ad414caa3340d675c1278be8223ddf0e6c
1765 F20101106_AABPZI monroe_m_Page_075thm.jpg
5f2581c914e2aa71c59bde6c1ba9fb26
b8277c9b98d7f559894cd9728fecddb5f9937c94
19733 F20101106_AABOWG monroe_m_Page_003.jp2
0e1e1a807b13011a7db435ce96193bbd
0bec0e4d46a98c0feaab086cfd557e77cbd16f70
1051914 F20101106_AABPAY monroe_m_Page_153.jp2
d122cb6eca05f3e1e2e021e48e8156da
7393ffa9a9f302806f5bd95aeee8e01bcd6d1eba
683859 F20101106_AABPCA monroe_m_Page_191.jp2
de7534188115713625a287c17eb8f44e
7f5108d8fbea7c6b810a1182316d70612c119512
F20101106_AABPBM monroe_m_Page_171.jp2
cc11a8ea49958888534603a3c71c7ebd
67babd2f2ac344936d2cf69eb85b98d253074924







PROCESS DEVELOPMENT AND SIMULATION OF HYBRID PHOTOVOLTAIC CELLS


By

MATTHEW L. MONROE





















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 Matthew L. Monroe





























To my wife, Kim









ACKNOWLEDGMENTS

I acknowledge my loving wife, Kim, for her patience, perseverance, and support. She was

a source of inspiration, and without her this could not have been possible. I thank my parents for

their continued love and support that has guided me through my personal and academic life.

I thank my advisor, Dr. Tim Anderson, for his advice and guidance in my research. I

would like to thank all of my committee members for their guidance and direction. Dr. Chinho

Park granted me the use of his laboratory and much assistance and advice in organic

photovoltaics. Dr. Kirk Ziegler provided laboratory space and assistance with nanocrystal

chemistry. Dr. Oscar Crisalle and Dr. Sheng Li provided valuable guidance through their roles

in the interdisciplinary photovoltaics team.

I thank Dr. Chinho Park's students at Yeungnam University (particularly Jiyoun Seol,

Young Wook Kim, Trong Nguyen Tam Nguyen, and Md. Azizul Hasnain) for extensive

experimental support and for making a foreign country feel like home. I thank Dr. Woo Kyoung

Kim for sharing his expertise in Medici and photovoltaics. I thank Dr. Ziegler's students,

particularly Justin Hill and Randy Wang, for assistance with their laboratory equipment and for

many helpful discussions.

I thank all of the support staff at the Chemical Engineering department, particularly Sean

Poole for helping me set up my simulation work. Special thanks go to Sherrie Jenkins, for

performing scheduling miracles on a regular basis.

Finally, I would like to thank all of my friends and family that have stood behind me and

helped me through my doctoral research and my whole life up to this point. They have helped

mold me into the person I am today, and I am grateful for it.









TABLE OF CONTENTS

page

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

L IST O F T A B L E S ..................................................................................................... . 7

LIST OF FIGURES .................................. .. ..... ..... ................. .9

A B S T R A C T ................................ ............................................................ 15

1 IN TR OD U CTION ......................................................... ................. .. ........ 17

Intro du action ................... .......................................................... ................ 17
M o tiv atio n ...............................................................................1 7
P hotovoltaic T technologies ...................................................................... ..................18
Organic Photovoltaics ................................................. ......19
Simulation of Hybrid Photovoltaic D evices................................................................ 23
T targeted R research ................. ........ ............................ .. .......... .. .............24

2 MEDICI SIMULATIONS OF HYBRID SOLAR CELLS .......................................... 27

In tro d u ctio n ............. .... ........... ....................................................................................... 2 7
Initial Modeling Efforts ...................... .......................... ......... 29
Prelim inary M odel D description ............................................... ............................ 29
D definition of m odel in M edici.............. .......................................... .... ........... 30
Im purity profile .......................................................................32
Cell dim ension adjustm ents .................................. .....................................34
Illu m in ation sou rce ............. .............................................................. ....... ............... 3 7
S u m m ary ................................................................3 8
Simulation of a Real Cell ........ ..................................... .. .... ........ ................. 39
Model Parameters ......... ..... .... ............. ...... ..............39
Initial Sim ulations ......................... .................... .. .. .................... 42
Reflectance of A g electrode .............................................................................. 43
M o b ility ............ .. ................. ................. ..............................................4 5
Exciton diffusion length ........... ........... .............. ................... ............... 46
D o p in g d en sity ................................................................................................... 4 7
D ensity of states ................................................................................................. 48
O pen-circuit voltage exam nation ........................................ ........................ 48
A bsorption coefficient adjustm ent ........................................ ....................... 50
M u lti-stag e ab sorption .......... ............................................................ .. .... .. .. .... 5 1
P3HT replacement with CIS .......................................................................54
Shortcomings of the initial model ................................. .................61
Tw o-Step Sim ulation Technique ......... ................. ...................................... ...................... 62
Photogenerated Carrier Distribution.......................................... 63
Line Source G generation .......................................................... .. .......... 64









J-V C u rv es ...................................................70..........
S u m m ary o f R esu lts .............................................................................. ................ .. 7 4

3 ORGANIC AND HYBRID SOLAR CELL PROCESS DEVELOPMENT..................... 122

In tro du ctio n ...................................................22..........
ITO A node Treatm ent ............. ...... .... ........ .. ..................... ........ .. 122
Bi-layer Organic Solar Cell Fabrication........................... .. ..................................126
C ell F fabrication Procedure ........... ................. ................. .................... ............... 127
Sub state P reparation ......... .............................................................. .......... ...... 127
Spin -C eating .................................................................................................... 12 8
E v ap o ratio n ........................................................................................12 8
E n c a p su la tio n .......................................................................................................... 12 8
F ilm D ry in g .............................................................................12 9
PEDOT:PSS .................................................130
P3HT ........................................................ ...............133
Bi-layer Cell Fabrication ................................. .......................... ..........133
Solvent C om prisons ..................................................................................................... 135
Hybrid Bulk H eterojunction Cell Fabrication ........................................ ............... 138
N anocrystal Synthesis and Surfactant .......................................................... ........ 139
H y b rid F ilm s ............................................................................................................ 14 0
TOPO-coated CdSe ....................... .................................. ... ............... 140
P3HT solubility in chloroform and pyridine.............................. ............... 142
C ell F ab rication ...................................................................................................... 144
Cells deposited from chloroform solutions ........................................ .....144
Thicker-film cells in chloroform solution .............................................................145
Chlorobenzene solvent ............... ......... ........ ........149
Com m ercial CdSe nanocrystals ................................ .. ........................ 150
Hybrid cell performance summary .................................... 150
Particle Induced Nanostructuring ................................. ........................... ...... 152

4 CONCLUSIONS AND FUTURE WORK ................................................................. 195

C o n c lu sio n s ............... ...... ........................................................................... .............. 1 9 5
Hybrid Photovoltaic Simulation ...................................................... 195
Anode Surface Treatment............................................. 196
Solvent Selection ................................... ............................196
Hybrid Bulk Heterojunction Photovoltaic Development .............................................197
F u tu re W o rk ................................................................................................................... 19 8
Organic Photovoltaic Simulations ................................. ......................... .... 198
Development of Hybrid Photovoltaic Cells .........................................................199

L IST O F R EFER EN CE S ................................................................................... ............... 201

BIO GRAPH ICAL SK ETCH .................................206.............................





6









LIST OF TABLES


Table page

2-1. P3HT Properties for Device Simulations. ........................................................................76

2-2. CdSe Properties for Device Simulations. ........................................ ......................... 76

2-3. Electrode Properties for Device Simulations..................................... .............76

2-4. Impurity profile inputs for simulations .................................................... ...................79

2-5. Solar cell performance measures for unit cells of 12 nm width and varying thickness........81

2-6. Performance measures for simulated hybrid cells with varying CdSe half-width ...............83

2-7. Cell performance measures from published and digitally converted J-V curves .................85

2-8. ZnO properties used for hybrid solar cell simulation ................ .................. ............... 86

2-9. Performance measures for simulated cells with varying carrier mobility...........................90

2-10. Performance measures for simulated cells with varying exciton diffusion lengths. ...........92

2-11. Absorption data and short-circuit current for graded absorption simulations ...................99

2-12. Materials properties for P3HT and CIS used in cell simulations..................................... 101

2-13. Performance measures for ZnO:CIS cells with an individual material property set at
the P3H T value. ............................................................................101

2-14. Performance measures for simulated ZnO:CIS solar cells with the P3HT absorption
sp e ctru m ................... ............................ .......................... ................ 10 2

2-15. Performance measures for simulated ZnO:CIS solar cells with P3HT values for
absorption coefficient and electron affinity. ........................................ ...............103

2-16. Performance measures for simulated ZnO:CIS solar cells with P3HT values for
absorption coefficient, electron affinity, and energy band gap................................... 104

2-17. Cell performance measures for real and simulated solar cells. .....................................118

2-18. Performance measures for simulated cells with 40 nm LD and varying mobility
values ............. ......... ......................................... .............. ................ 118

2-19. Generated carriers and performance measures for simulated solar cells ........................121

3-1. Chem ical com position of ITO film s................................ ...................... ............... 155









3-2. Performance of organic solar cells on treated ITO substrates............................................155

3-3. RMS surface roughness of P3HT films shown in Figure 3-11. ........................................163

3-4 J-V data for bi-layer solar cells............................................. ......................................... 166

3-5. Solvents considered for hybrid bulk heterojunction film deposition. ...............................166

3-6. Mean rms surface roughness in nm for hybrid films deposited from selected solvents......170

3-7. Hybrid solutions of P3HT and TOPO-coated CdSe nanocrystals in a mixed solvent of
chloroform and pyridine. ........................................................................ ................... 176

3-8. Film properties for P3HT films deposited from various solvents............. ................181

3-9. Hybrid solar cell fabrication information and performance data. .......................................190









LIST OF FIGURES


Figure page

1-1. Worldwide cumulative installed PV Power in Megawatts from 1992 to 2006...................26

1-2. Common organic materials used in solar cell development ............................................26

2-1. Hybrid solar cell and corresponding unit cell used for device simulation ..........................76

2-2. Wavelength-dependent absorption coefficient data used in simulations............. ...............77

2-3. J-V curves for simulated hybrid solar cells with different methods of specifying doping
density ................ ............... .........................77

2-4. Simulated J-V curves showing the effect of doping density in the CdSe nanorods..............78

2-5. Variation of open circuit voltage with doping density of CdSe nanorods...........................79

2-6. Simulated J-V curves with varying unit cell thickness............................... ............... 80

2-7. Solar cell parameters for unit 12 nm wide unit cells with varying cell thickness ................80

2-8. J-V curves for hybrid cells with varying nanorod width ............................................. 81

2-9. J-V curves for simulated hybrid cells with CdSe nanorod half-thickness between 10
an d 2 5 nm ................................................................................82

2-10. Solar cell performance measures for simulated hybrid cells with varying CdSe half-
w idth .......................................................... ................................... 82

2-11. J-V curves for hybrid solar cells with different light source specifications ......................83

2-12. Illustration of the PHOTOGEN command in Medici .......................................................84

2-13. SEM images of ZnO nanofibers and nanofiber and P3HT composite films........................ 84

2-14. J-V curve for a real ZnO:P3HT solar cell to be used for verification of Medici
sim ulations. ............................................................................... 85

2-15. J-V and P-V curves for the real solar cell fabricated by Olson et al. .................................85

2-16. Unit cell used for simulations of ZnO/P3HT hybrid cells.. .............................................86

2-17. M edici unit cell used for device simulation.......................................................... ......... 87

2-18. Simulated J-V curves for ZnO:P3HT solar cell using two P3HT regions..........................88









2-19. J-V curves for simulated cells with zero absorption in the P3HT2 region and varying
reflectance from the A g electrode ......... ................. ................................ ............... 89

2-20. Simulation results showing the effect of changing charge mobilities in the P3HT
region s ............... ......... .......... ......... ........ ................................. 90

2-21. J-V curves for simulated cells with varying exciton diffusion length.............................91

2-22. Extrapolations to estimate Voc for simulated cells with varying exciton diffusion
le n g th .............. .... ........ ..... ............................................... ................ 9 1

2-23. Solar cell performance measures for simulated cells with varying exciton diffusion
lengths..................... .......................................92

2-24. Real and simulated J-V curves for cells with varying P3HT doping density. ...................93

2-25. Real and simulated J-V curves for hybrid cells with varying ZnO doping density.............93

2-26. Real and simulated J-V curves for hybrid solar cells with varying P3HT density of
state s................. .. .......... .................. ................................................ 9 4

2-27. Simulated J-V curves for hybrid solar cells with varying P3HT mobility........................94

2-28. Simulated J-V curves for hybrid cells with varying P3HT doping concentrations.............95

2-29. Energy band diagram for P3HT ZnO hybrid solar cells. .............................................95

2-30. J-V curves for simulated cells with varying energy band gap in the active layers..............96

2-31. Absorption coefficient vs. wavelength as tabulated in Medici....................... ......... 96

2-32. AM 1.5 solar spectrum. ..................................... .. .. ......... .. ............97

2-33. Carrier generation in simulated solar cells plotted with absorption coefficients for
P 3H T and Z nO ............................................................................97

2-34. J-V curves for simulated cells showing the original P3HT absorption profile and an
edited absorption profile limiting absorption between 0.2 and 0.3 m ............................. 98

2-35. J-V curves for simulated cells with exciton diffusion length of 10 nm and multi-stage
ab so rp tio n reg io n s..............................................................................................................9 8

2-36. Examples of cumulative distribution function with mean of 10 nm and a range of
standard deviation ....................................................... ................. 99

2-37. Simulated J-V curves for cells with graded absorption profiles.............. ... ................100

2-38. Simulated J-V curves with CIS replacing P3HT..................................... ...... ............... 100









2-39. Simulated J-V curves for ZnO:CIS solar cells with an individual material property
changed to the P3H T value ......... ................. ................. ...................... ............... 101

2-40. Simulated J-V curves for ZnO:CIS solar cells with the P3HT absorption spectrum
applied ............................................ ..... ...................... ................. 102

2-41. Simulated ZnO:CIS solar cells with P3HT values for absorption coefficient and
electron affinity ............... ................................................ ............ ...... 103

2-42. Simulated ZnO:CIS solar cells with P3HT values for absorption coefficient, electron
affinity, and energy band gap................................................. .............................. 104

2-43. Simulated J-V curves for ZnO:P3HT solar cells with varying carrier mobility. .............105

2-44. Calculation method for estimated second derivatives of J-V curves............................. 105

2-45. Estimated second derivative J"est for simulated ZnO:P3HT solar cells with varying
carrier mobilities and carrier generation in the full P3HT region. ..................................106

2-46. Simulated J-V curves for ZnO:P3HT solar cells with varying carrier mobilities and
carrier generation in the 10-nm exciton diffusion length region. ................................... 107

2-47. Estimated second derivative J"est for simulated ZnO:P3HT solar cells with varying
carrier mobilities and carrier generation in the 10-nm exciton diffusion length region. .108

2-48. Simulated J-V curves for ZnO:P3HT solar cells with varying absorption coefficients
in P3HT ................................................... ............... ................ 109

2-49. Simulated J-V curves for ZnO:P3HT solar cells with varying energy band gap.............09

2-50. Simulated J-V curves for ZnO:P3HT solar cells with varying band gap and absorption
in the P 3H T region. .................................................. ................................ 110

2-51. Simulated J-V curves for ZnO:P3HT solar cells with P3HT energy band gap of 1.0 eV
and hole m ability of 500 cm 2/V -s. ....................................................... ............... 110

2-52. Photogenerated carrier distribution in pairs/cm3 for the full unit cell and the region of
13 nm < x < 27 nm along the edge of the ZnO nanorod.......... ........................111

2-53. Carrier mapping scheme for two-stage simulations. ..........................112

2-55. Cumulative number of photogenerated carriers in simulated cells with varying exciton
diffusion length. ................................................................ .. ..... ........ 114

2-56. Photogenerated carrier distribution along x- and y- coordinates for models with
varying exciton diffusion lengths........................................................................ ....... 115

2-57. Contour plots of the photogenerated carrier difference between the full absorption
simulation and the 40 nm LD simulation........ .. ............................. .. ............... 116









2-58. Photogenerated carriers at the tip and base corer points of the nanorod in simulated
h y b rid c e lls ...................................... ................................................... 1 1 7

2-59. J-V curves for simulated solar cells using line-source carrier generation.........................117

2-60. J-V curves for simulated cells with a 40 nm LD and varying carrier mobility ................118

2-61. Photogenerated carrier distribution for a simulated unit cell with LD = 40 nm and LD
= 40 with the P3HT absorption coefficient increased by 50% ............... .................119

2-62. Difference in photogenerated carriers between 150% and 100% P3HT absorption
coefficients in simulated cells with 40 nm LD. .................................... .................120

2-63. J-V curves for simulated cells with varying absorption coefficient in P3HT ...................121

3-1. J-V curves for organic solar cells on treated ITO substrates ................................ .......... 155

3-2. Energy band diagram for bi-layer organic solar cells ........................................................ 156

3-3. Steps for bi-layer solar cell fabrication .................................................... ..............156

3-4. PEDOT:PSS film thickness vs. spin-coater speed................................... ...............157

3-5. P3HT film thickness vs. spin-coating speed. ............................................. ............... 157

3-6. FTIR spectra for P3HT film and solution. ........................................ ....................... 158

3-7. FTIR spectra for PEDOT:PSS film and solution...................................... ...............159

3-8. J-V curves for bi-layer organic solar cells fabricated with a single 80 nm thick layer of
P E D O T :P S S ...................................... .................................................. 16 0

3-9. J-V curves for bi-layer organic solar cells fabricated with two 40 nm thick layers of
P E D O T :P S S ...................................... .................................................. 16 1

3-10. Calibration curves for slow- and fast-filtered PEDOT:PSS...........................................162

3-11. A FM im ages ofP3H T film s....................................................................... ..................162

3-12. J-V curves for bi-layer solar cells fabricated on untreated ITO substrates. ....................163

3-13. J-V curves in the dark and under 100 mW/cm2 illumination. .........................................164

3-14. Surface roughness measurements by profilometry for hybrid films deposited from
v ariou s solv ents.............................. ........................................................ ............... 16 7

3-15. RMS surface roughness for 5 x 5 ptm surface area samples measured with AFM............168

3-16. RMS surface roughness for 1 x 1 pm surface area samples measured with AFM...........169









3-17. Optical microscope images of selected films ............................................ ..................170

3-18. SEM im ages of selected hybrid film s ...................................................... ............. 173

3-19. Film surface roughness vs. pyridine concentration in the chloroform solvent for
TOPO-coated and pyridine-coated CdSe nanocrystals..............................175

3-20. Optical microscope images of hybrid films deposited from 20.4 mg/ml, 19 mg/ml and
5 m g/m l solutions.............. ............. ..... .. ........... ......... ... .......... 176

3-21. Optical microscope image of 19 mg/ml hybrid film deposited at 3000 rpm and
subjected to a pure solvent spin-coating step at 8000 rpm. ...........................................177

3-22. Optical microscope images of P3HT films deposited from 5 mg/ml solutions in
chloroform, 1:1 chloroform pyridinee, and pyridine.................................... ............... 177

3-24. Surface profiles of P3HT films deposited from chloroform, 1:1 chloroform:pyridine,
and pyridine solvents. ................................... ..... .......... ........ ..... 179

3-25. Dark and illuminated J-V curves for cells generated from 5 mg/ml composite
solutions in chloroform mixed with 2% pyridine and pure chloroform .........................181

3-26. Dark and illuminated J-V curves for 10 mg/ml hybrid solution deposited with low-
speed spin-coating ........... .. .................. ............. ........ ..... ............ .. .. 82

3-27. Surface profiles of film deposited from 12 mg/ml P3HT and 25 mg/ml hybrid
solutions in 2% pyridine in chloroform ............................................... ............... 183

3-28. Dark and illuminated J-V curves for hybrid bulk heterojunction solar cell deposited
from 25 mg/ml solution and P3HT polymer cell deposited from 12 mg/ml solution......184

3-29. Short-circuit current decay for hybrid and P3HT.................................... ............... 185

3-30. Dark and illuminated J-V curves for hybrid bulk heterojunction and pure P3HT solar
cells with limited air exposure during processing................................. ............... 186

3-31. Dark and illuminated J-V curves for hybrid solar cell fabricated from chlorobenzene
w ith 2% pyridine solution .......................................................................................... 187

3-32. Dark and illuminated J-V curves for a hybrid solar cell fabricated with commercial
CdSe nanopowder .................... .. ......................... ............ 188

3-33. Illuminated J-V curves for hybrid bulk heterojunction solar cells with various
fab rication con edition s............................................................................. .................... 189

3-34. J-V curves for the best bi-layer and hybrid cells shown in this dissertation ...................189

3-35. Film thickness for multi-layer hybrid films............................................................ 191









3-36. SEM and AFM surface images of multi-layer hybrid films................................191

3-37. Optical microscope images for multi-layer hybrid films............................................193

3-38. RM S surface roughness for multi-layer hybrid films....................................................... 194









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

PROCESS DEVELOPMENT AND SIMULATION OF HYBRID PHOTOVOLTAIC CELLS

By

Matthew L. Monroe

December 2008

Chair: Timothy J. Anderson
Major: Chemical Engineering

The fabrication and simulation of hybrid bulk heterojunction photovoltaic cells was studied

to develop an understanding of these devices and facilitate improvements in device fabrication

techniques. The device simulation software package Medici was applied to this type of device

for the first time, and novel device fabrication techniques were developed to improve device

performance.

Medici was adapted for a hybrid system consisting of an ordered array of inorganic

nanorods interspersed with an absorbing, semiconducting polymer. A cell taken from the

literature was simulated using a two-stage simulation technique which independently calculated

the light absorption profiles and the cell performance. This technique applied a line source of

carriers directly at the interface between the inorganic and organic regions of the cell,

realistically imitating the exciton dissociation physics of real hybrid cells. The simulations

showed low current densities as compared to the real cell, due to strong recombination at the

material interface.

Nitrogen plasma treatment of the transparent anode in bi-layer and hybrid photovoltaic

cells was found to reduce surface roughness, increase the hydrophilic nature of the film, and

improve cell performance. A range of solvents were tested for hybrid bulk heterojunction film









deposition, with chloroform, chlorobenzene, and o-dichlorobenzene found to provide films with

low surface roughness and strong uniformity. Hybrid bulk heterojunction solar cells were

fabricated, and these cells showed low performance of less than 1% efficiency. The active layers

degraded upon exposure to air, resulting in a drop in short-circuit current density that was more

pronounced for hybrid films than for pure polymer single-layer cells. These findings highlight

the environmental sensitivity of these devices and the need for an inert environment for cell

fabrication and testing.









CHAPTER 1
INTRODUCTION

Introduction

Since the discovery of the photovoltaic effect and the design of the first functional solar

cell, photovoltaic technology has consistently developed to become an increasingly viable energy

source. Photovoltaics has developed into many classes of devices and materials systems.

Photovoltaic technology can provide clean, efficient, and portable energy.

Motivation

There is currently a strong push toward alternative energy sources as the price of oil

increases and nations worldwide work to slow the emission of greenhouse gases such as COx and

NOx. Many countries have established conservation and alternative energy programs in attempts

to control the output of these gases. Recent increases in oil and gas prices and controversy

surrounding global warming have driven public recognition of the need for alternative renewable

energy sources. While small-scale steps such as hybrid cars stand to relieve a small amount of

the world's fossil fuel consumption, new technologies such as photovoltaic energy must be

developed to fulfill the world's large-scale energy needs.

Of the available candidates for alternative large-scale energy production, photovoltaic

energy conversion has many qualities that make it a leading technology, including:

Photovoltaic technology has been studied intensely for many years, initially driven by its
use in the space program to provide energy for satellites and space vehicles, to prepare it
for commercialization.

Energy generated from solar cells can be generated locally, resulting in reduced energy
distribution costs and a more reactive system.

Because of its local power generation, photovoltaics are optimal for power generation in
remote locations where it is difficult or impossible to connect to a local grid.









Solar energy is plentiful. America's energy needs could be supplied by a single solar
array covering an area of 100 x 100 miles in the uninhabited deserts of the western
United States.

Photovoltaic sources provide peak energy during peak consumption times. Energy
consumption is at its maximum during the middle of the day, requiring power plants to
work harder to provide the necessary energy to their consumers. Because the sun's rays
on earth are also at a maximum during that time, solar energy provides maximum energy
at the most critical portion of the day.

Worldwide photovoltaic installations have begun to grow rapidly, as shown in Figure 1-1

[1]. The U.S., who at one time was the world leader in photovoltaics, has fallen behind countries

such as Japan and Germany who have made strong efforts to boost their solar energy programs.

However, recent initiatives such as the Million Solar Roofs Initiative and Solar America

Initiative, along with federal and state sponsored cost-sharing programs for residential and

commercial solar installations, have created a new boom in American solar installations [2].

Photovoltaic Technologies

Many photovoltaic technologies are currently in the developmental or production stages,

including crystalline and polycrystalline silicon, thin film, concentrator arrays, space

photovoltaics, and organic cells. Polycrystalline silicon is currently the dominant technology in

the photovoltaics market, drawing on years of silicon processing technologies to provide a low

production cost. Polycrystalline silicon modules area commercially available for as low as

$4.29/Watt in April 2008, approximately $0.50/Watt cheaper than the U.S. average cost of

$4.81/Watt [3]. These panels are easily identified by their deep blue color and are currently in

production worldwide. The next wave of photovoltaic production appears to be thin-film

photovoltaics. These cells employ direct-bandgap materials with strong absorption coefficients

such as amorphous silicon (a-Si), cadmium telluride (CdTe), and copper indium-gallium

diselenide (CuInxGai-xSe2). Many of these cells have demonstrated extremely promising









laboratory efficiencies, and the thin layers reduce material costs and provide the ability to create

flexible cells for a wider range of applications.

Concentrator photovoltaic arrays generate a large amount of power from a single cell by

using an array of mirrors to focus the sun's energy on the cell. This technology has been used to

reduce large-scale production costs because the cost of a mirror is lower than the cost of a

photovoltaic panel. Space solar cells are the most advanced cells, using expensive growth

methods to produce high quality photovoltaic material. These cells provide extremely high

performance and durability through their multi-junction stack technology, but this comes at a

high cost that makes the cells too expensive for standard terrestrial applications. Under

concentration, however, their high efficiency (approaching 40%) mitigates their higher cost.

Organic and hybrid solar cells provide extremely lightweight devices and low-cost

manufacturing, but currently their low performance and life time limits their applications.

Organic Photovoltaics

Compared to their inorganic counterparts, the development of organic PV is in its

infancy. Drawing on advances in organic light emitting diodes, however, considerable progress

has been achieved that encourages continued exploration of organic solar cells. Organic

photovoltaic technology uses organic molecules or polymers to absorb sunlight and generate

photocurrent. Many of these materials have extremely high absorption coefficients with maxima

in the visible region of the spectrum (a z 105 cm-1), so an extremely thin layer (hundreds of

nanometers) is sufficient for absorbing incident light. Because the films are extremely thin and

flexibile as compared to inorganic crystals, organic photovoltaics show promise for flexible and

lightweight portable devices. Many organic materials can be rapidly deposited through

inexpensive techniques such as thermal evaporation at moderate vacuum, or by spin-coating, dip-

coating, screen printing, ink jet printing, and spray coating at room temperature and pressure.









The approaches to using organic for light conversion can be categorized into three general

classes of cell structures: bi-layer, dye-sensitized, and bulk heterojunction. Similar to inorganic

designs, bi-layer cells use a flat junction created by stacking p- and n-type organic layers, with

additional layers incorporated for charge transport enhancement [4-6]. Dye-sensitized solar cells

use an organic dye adsorbed on inorganic transport materials, typically TiO2, so that the dye

absorbs photons and the inorganic phase allows for efficient charge transport [7-10]. Bulk

heterojunction cells consist of donor and acceptor materials mixed together to form a blended

junction throughout the device active layer.

Organic cells are distinguished from their inorganic counterparts by exciton creation upon

photon absorption. Due to their extremely low exciton binding energy, inorganic p-n junction

cells generate free carriers upon photon absorption, and the carriers are primarily collected by the

field across the depleted junction. Organic materials, on the other hand, primarily generate

excitons that have significant binding energy and transport by diffusion until they recombine or

dissociate at an energetic interface to produce free carriers for eventual collection.

Excitons are efficiently dissociated at a p-n junction in organic devices, although exciton

dissociation occurs to a lesser extent at interfaces with electrodes, polymer chain defects,

absorbed oxygen sites, or active-layer impurities [11]. Because of this dissociation requirement,

only excitons generated within a diffusion length of the junction will contribute to the collected

current. Exciton diffusion lengths are typically on the order of 5 to 20 nm for organic

semiconductors, placing a limit on the thickness of active layers and therefore the photon

absorption extent [12-16]. The blended junction of a bulk heterojunction cell attempts to provide

an interface within an exciton diffusion length throughout the entire active layer, thus allowing

for thicker active layers with better adsorption and more efficient exciton dissociation. Once the









free carriers are generated they must be collected at their respective electrodes before they

recombine, which occurs to some extent in the bulk or more prevalently at interfaces.

Bulk heterojunction devices have been fabricated by blending several classes of materials,

including multiple organic small molecules, polymer and organic molecules, polymer and carbon

nanotubes, polymer and inorganic nanoparticles, and polymers deposited in a prefabricated

inorganic nanostructure. Heterojunctions based on small organic molecules such as copper

phthalocyanine (CuPc) receive little attention compared to polymer-based heterojunctions, but

have demonstrated reasonable efficiency using a variety of deposition techniques [17-19].

Carbon nanotubes have very recently received consideration as a solar cell material, both as an n-

type conductor in a heterojunction [20] and as a structured electrode [21].

Bulk heterojunctions fabricated from conjugated polymers and C60 represent the most

widely-studied class of bulk heterojunction solar cell. Since the discovery of ultra-fast

photoinduced charge transfer from conducting polymers to C60 [22], polymer-C60 bi-layer and

bulk heterojunctions have been studied extensively. The first cells using semiconducting

polymers and C60 were fabricated by the same group [23]. Early bulk heterojunction cells using

C60 employed poly (pheneylene vinylene) (PPV) derivatives as the polymer material, and issues

centered on co-dispersion of the two materials to create a well-blended structure [24]. This issue

was solved with the synthesis and application of [6, 6]-phenyl C61 butyric acid methyl ester

(PCBM), a highly-soluble C60 derivative, [25, 26], and led to a record efficiency of2.5% in 2000

[27]. Since this development, other fullerene derivatives have been synthesized and evaluated,

but PCBM remains the most widely used [28, 29]. The most popular PPV-based polymers are

poly [2-methoxy,5-( 2'-ethylhexyloxy) -1,4-phenylene-vinylene] (MEH-PPV) and poly [2-

methyl,5-(3*,7** dimethyloctyloxy)]-p-phenylene vinylene (MDMO-PPV), although other









derivatives have been demonstrated [30]. These cells frequently feature a hole transporting layer

of poly (3, 4-ethylenedioxythiophene):poly (styrenesulfonate) (PEDOT:PSS) to improve hole

collection. The structures of PCBM and some polymers commonly used for bulk heterojunction

solar cells are shown in Figure 1-2.

In particular, polythiophene derivatives are now popular alternatives to PPV, with poly (3-

hexylthiophene) (P3HT) being the most commonly used [29, 31]. P3HT is synthesized with a

regio-regular configuration where the polymer side chains alternate on opposite sides of the

backbone chain. This arrangement aids in aligning the polymer chains for efficient charge

transfer along the backbone, with additional chain straightening attributed to steric hindrance

from the fullerene molecules [32]. Bulk heterojunction cells fabricated from P3HT and PCBM

have reached efficiencies of 3.5% in 2003 [33] and 4.4% in 2005 [34].

The use of inorganic nanocrystals as a replacement for C60 is a relatively recent trend. In

addition to their high electron mobilities, semiconductor nanocrystals can contribute to

absorption and photocurrent generation in the heterojunction active layer. These nanocrystals

offer the possibility of bandgap engineering by material selection and tightly controlling the size

distribution, as well as the growth of different crystal shapes such as rods and tetrapods [35-37]

to create more efficient charge transport pathways. As with early work involving C60, dispersion

of the nanocrystals remains an issue in process development. One approach to enhance

nanocrystal dispersion is the use of sol-gel processing to grow nanocrystals in the polymer film

[18, 38-40]. The nanocrystals are typically grown in solution, and are formed with a surfactant

capping layer to relieve the high surface energy [41, 42]. Exchange of this surfactant has been

demonstrated to obtain improved solubility and electrical properties [43-45].









Ordered heterojunctions have been fabricated from a variety of materials including

mesoporous TiO2 [46, 47], but ZnO nanostructures have been the most popular. Regular arrays

of well-aligned ZnO nanowires have been grown using a variety of techniques including

MOCVD [48, 49], evaporation [50, 51], and solution-based thermal growth [52]. Tak et al.

demonstrated selective MOCVD growth of ZnO nanoneedles on a patterned buffer layer [53],

allowing for substrate patterning with nanomaterial coverage. Low-temperature growth has been

achieved using a sol-gel precursor method [54]. Device performance of these ordered

heterojunction devices has been limited by rod spacings several times larger than the exciton

diffusion length of the active polymer.

Simulation of Hybrid Photovoltaic Devices

The incorporation of excitons in device physics models has been done both inside and

outside the organic photovoltaics world. Models of silicon solar cells show that the inclusion of

excitons causes a decrease in dark current but an increase in photocurrent [55], while further

studies showed that this effect is only substantial when the exciton diffusion length is

significantly greater than the electron diffusion length [56]. This work was further expanded by

Burgelman and Minnaert to include exciton dissociation at surfaces, showing that even purely

excitonic devices such as organic solar cells can be effective as long as the rate of exciton

dissociation is high [57].

Simulations of polymer-inorganic hybrid cells have been performed with various

assumptions. Studies have proven that effective charge transport only takes place in these

devices when the dimensions of the polymer region are less than or equal to the exciton diffusion

length [58]. Additional models have validated this effect in polymer-C60 bulk heterojunction

cells [59], highlighting the importance of excitons in these types of cells. The traditional circuit









model of a solar cell has been modified to include an additional rectification diode to include the

effect of exciton recombination [60].

While work exists in development of models to describe organic photovoltaics, little work

has been done in applying existing semiconductor modeling programs to simulate these cells.

Recently, Takshi et. al. applied Medici to simulate an organic transistor using P3HT as the

semiconductor [61]. Medici is a 2-dimensional device simulation program developed by Avant!

Corporation. It is designed for the simulation of MOS and bipolar transistors, and models

potential and carrier concentrations in a device by solving Poisson's equation and the electron

and hole continuity equations.

Targeted Research

The work presented in this dissertation provides a study of hybrid photovoltaic devices

from an experimental and theoretical perspective. The device simulation program Medici is used

for the first time to provide simulations of an ordered bulk heterojunction photovoltaic device

with an array of inorganic nanorods interspersed with a semiconducting organic polymer. This

design was chosen due to its semi-regular structure that allows the device to be broken into a

representative unit cell, providing greater detail in the simulations. By modeling the cell in this

way, effective values of key parameters such as the carrier mobility, exciton diffusion length,

and energy gap of the interface can be estimated for the device during operation.

This simulation supplements experimental work focusing on process development for a

polymer-nanocrystal bulk heterojunction solar cell. The issues of charge transport and exciton

dissociation are targeted. Charge transport is improved through surface treatment of the ITO

electrode prior to active layer deposition. This generates a smoother surface and promotes

adhesion of subsequent layers. Exciton dissociation is addressed through control of the

morphology of the bulk heterojunction active layer. This is achieved through surface exchange









of the nanocrystal surfactant, selection of an appropriate solvent for film deposition, and the

introduction of a novel layer-by-layer deposition process.






































Figure 1-1. Worldwide cumulative installed PV Power in Megawatts from 1992 to 2006 [1].


Figure 1-2. Common organic materials used in solar cell development regioregular P3HT (a),
PCBM (b), MDMO-PPV (c), MEH-PPV (d), and PEDOT:PSS (e).


6000

5000



-a
4000 -

S3000

2000

1000
E
S 0

rsj m o t Ll 0 0- CD -i rj rn It Ll WD
M MJCD CD CD CD CD CD CD
I m i i -i i rj rj rj r j rj









CHAPTER 2
MEDICI SIMULATIONS OF HYBRID SOLAR CELLS

Introduction

Device modeling has lagged behind cell fabrication for organic and hybrid photovoltaics.

Cell performance has been characterized using equivalent circuit theories [60, 62], and some

work has been performed to model certain parameters such as cell lifetimes [63], charge

recombination [64], and short-circuit current [59]. The use of existing simulation programs to

provide a fully-encompassed view of the device performance has not been attempted to this

point.

Organic photoabsorbing materials differ fundamentally from most inorganic crystals in

that the absorption of a photon generates an exciton, or bound electron-hole pair, with high

binding energy. Excitons can be generated in inorganic materials as well, but this occurs only

over a very limited wavelength and temperature range, and the resulting excitons exist with a low

binding energy on the order ofkT at room temperature [57]. This low binding energy causes the

exciton to be highly unstable in the presence of elevated temperatures or strong fields, such as in

the space-charge region of a p-n junction. In organic materials, excitons are generated nearly

exclusively and exist with a binding energy or approximately 10x that of the inorganic excitons,

or 300 meV. These highly stable excitons dissociate at an interface with another material,

where one of the carriers is transferred into the neighboring material. In organic photovoltaic

cells, the electron is typically transferred into an n-type material due to the relatively low

electron mobility in most organic materials.

Because the organic exciton exists as an essentially uncharged particle unaffected by

fields, it travels through the solar cell through diffusion, with the exciton diffusion length (LD)

representing the maximum distance it can travel before self-annihilation. For many organic









photoabsorbers, including P3HT, LD has been measured to be on the order of 10 nm [15]. This

restricts the current in organic photovoltaics because only excitons generated within LD of an

electron acceptor can generate free carriers, and any photons absorbed by the organic absorber

outside of this distance are lost to recombination. This inherent limitation is the motivation

behind the bulk heterojunction cell design, in which the organic absorber and electron acceptor

are blended on a scale that approximates LD.

The hybrid bulk heterojunction design considered for simulations in this dissertation is an

ordered array of inorganic nanorods with an organic absorbing polymer interspersed between and

on top of the rods. This inorganic array is assumed to be uniform in terms of rod length, rod

width, and rod spacing, and the polymer is assumed to fully occupy vacant areas between the

rods. The simulation program chosen for this effort was Medici [65].

Medici is a device modeling software package from Synopsis designed to simulate diodes,

MOS transistors, and bipolar transistors, as well as emissive devices. The program is provides a

two-dimensional simulation area and solves Poisson's equation and the electron and hole

continuity equations to generate two-dimensional distributions of potential and carrier

concentrations [65]. Medici is designed for the simulation of crystalline inorganic materials, but

has been used in certain rare instances for the modeling of organic semiconductors. Recently,

Takshi et al. used Medici to simulate a dual gate organic transistor using P3HT as the organic

semiconductor [61]. The simulation applied literature values for P3HT properties and measured

the performance of single- and dual-gate OFET designs with varying film thickness. No effort

was made by the authors to account for exciton dynamics in the device because the polymer was

functioning as a transistor rather than a solar absorber, so the influence of excitons would be

minimal.









The semiconductor properties that are necessary for a proper simulation using Medici

include the electron and hole mobility, band gap energy, permittivity, electron affinity, density of

states in the conduction and valence band, and doping density. These properties are well-known

for most common semiconductor materials, but are less well-characterized for many organic

materials. Despite this, measurements and assumptions regarding the necessary properties for

P3HT can be found in the literature, and these values were used as starting points for device

simulation.

Initial Modeling Efforts

A dispersion of P3HT in an ordered array of CdSe nanorods was chosen as a first attempt at

hybrid solar cell modeling using Medici. The choice of materials and dimensions were

somewhat arbitrary, but represent a pseudo-realistic scenario with which to expore the

capabilities of Medici for hybrid cell modeling. This initial modeling effort serves as a precursor

to attempts to simulate a real cell from literature.

Preliminary Model Description

Hybrid solar cells consisting of an ordered array of CdSe nanorods dispersed in a P3HT

matrix were simulated using the device modeling software package Medici. The nanorods were

assumed to be vertically aligned, be in contact with the aluminum back electrode, have uniform

dimensions, and have uniform spacing. An illustration of this design is shown in Figure 2-1.

Due to the symmetry imposed in this structure, the cell can be divided into a basic repeating unit

cell, as illustrated in the figure. The active layer is fixed at 100 nm thick, while the CdSe

nanorod has dimensions of 90 nm x 10 nm. The rod spacing is fixed at 20 nm. These

dimensions define a P3HT "capping" layer of 10 nm over the tip of each of the CdSe nanorods.

Additionally, the unit cell has a polymer thickness of 10 nm to the side of the nanorod, due to the

20 nm rod spacing imposed in the model. This is chosen because it is equal to the approximate









exciton diffusion length in P3HT [15]. The unit cell used for simulation has dimensions of 100

nm thick (the thickness of the device, not including electrodes) by 15 nm wide (half of the 10 nm

rod width plus half of the 20 nm rod spacing). The top contact is indium tin oxide (ITO), and the

bottom contact is aluminum.

The unit cell occupies 0.1 [tm in the y direction and 0.015 [tm in the x direction. The CdSe

nanorod occupies an area from 0.01 < y < 0.1 [um and 0 < x < 0.005 [tm. P3HT occupies the

other regions, for y < 0.01 [tm and x > 0.005 [tm.

The materials properties assigned for the simulation were taken from the literature and are

displayed in Tables 2-1 2-3. Material properties for CdSe were taken from previous

simulations from Dr. Woo Kyoung Kim [66]. Figure 2-2 shows the absorption coefficient data

used for several different materials in simulations, with the sources of the data noted in the

caption. Note that the electrical band gap and optical band gap can be individually specified in

MEDICI, but they are assumed to be equal in all simulations unless otherwise noted. The doping

density was calculated to depend on the amount of time that P3HT is exposed to air, but falls in

the narrow range of 1 x 1016 to 2 x 1016 cm-2 for air exposure times on the order of 100 hr [67].

This value was further extended to 5 x 1016 cm-2 by Takshi et al. by assuming extended exposure

times [61], and this value was used as a starting point for simulations. The P3HT hole mobility

of 0.1 cm2/V-s is the highest reported mobility for this material [68].

Definition of model in Medici

This section gives a brief description of the steps required to perform a simulation in

Medici. Medici runs based on user-defined input files that are easily edited in text format with

any text editing program, typically Notepad or Wordpad on Windows computers. For

photovoltaic simulations, these input files contain the following information, typically in this

order:









1) Mesh definition
2) Region definitions
3) Electrode definitions
4) Material property definitions
5) Illumination definition
6) J-V calculation
7) Output definition.

The simulation mesh is the collection of points where calculations are performed

throughout the simulation area. The spacing of the mesh points can be specified by the user and

is somewhat arbitrary, with the exception that mesh points should exist at boundaries between

regions to facilitate convergence of the calculations. Medici allows a maximum of 3,200 mesh

points, with more points providing a more well-defined calculation at the expense of

computational time.

Regions of the mesh can be defined to correspond to different materials in the simulation.

These are specified through ranges of (x, y) coordinates in units of [m, with y = 0 corresponding

to the top of the simulation area. The electrodes are specified in terms of their optical properties

and are not defined by (x, y) coordinates. Calculations are performed by Medici at the interface

points with the electrodes, but not within the actual electrode.

Material properties are specified to correspond to the material ranges given in the region

definition. An example would be

MATERIAL REGION=P3HT PERMITTI=3 NC300=2E18 NV300=2E19 EG300=1.7

+ EGO300=1.7 AFFINITY=3.15 ABS.FILE="abs-p3ht.txt" PR.TAB

where the properties for the area defined as "P3HT" are specified. Carrier mobilities and

impurity profiles are defined in separate statements. Further details on impurity profile

definitions are given in the following section. The absorption coefficient is defined through an

external text file, where the value can be specified by the user for a range of wavelengths.









The illumination source can be defined in multiple ways. For absorption models such as

photovoltaic cells, radiation can be defined to follow any specified profile and generated from a

point source or line source at any location outside of the simulation area. Further details are

given in a later section describing the effect of altering the origin of the illumination source. In

addition to general illumination, sources of photogenerated carriers can be specified along any

line within the simulation area.

With the defined illumination source, Medici calculates the photogenerated carrier

distribution throughout the simulation area based on the absorption coefficient data given for

each material region. The program then begins iterating over a range of applied biases specified

by the user until the current and potential has been mapped and converged throughout the

simulation area. Once a solution has been achieved, Medici provides a vast array of output

mechanisms for the user to extract data from the simulation, including plot generation and

numerical data extraction.

Impurity profile

The way in which Medici assigns impurity profiles in simulations is a bit of a mystery. A

few simulations were performed to determine the proper way to call this function. The profile

can be defined in terms of (x, y) coordinates or by the "REGION" statement which applies

values to an area defined by a specific name. The impurity profile input statements used are

listed in Table 2-4.

The file "impurity_l" specifies the doping areas by the "REGION" statement, which

generates doping in an area which has previously been defined in coordinate space and assigned

a label (P3HT or CdSe). The file "impurity_2" first applies p-type doping at a concentration of 5

x 1016 cm-2 to the entire cell area and then applies an n-type doping of 6 x 1016 cm-2 to the area

occupied by CdSe. This depends on Medici overwriting the impurity values when multiple









definitions are made in the same region, in this case the CdSe region. In "impurity_3" the cell is

sectioned into three rectangles, two which encompass the P3HT region and one which defines the

CdSe region. The P3HT region consists of the regions y < 0.01 [m and (0.01 < y < 0.1 am,

0.005 < x < 0.015 am). The CdSe region is defined of the area from (0.01 < y < 0.1 am, 0 < x <

0.005 [tm). The file "impurity_4" considers the possibility that MEDICI sums doping

specifications rather than overwriting them. It takes the same format as "impurity_3" but assigns

the n-type doping concentration as 1 x 1016 cm-2 to account for the possibility that 5 x 1016 cm-2

is cancelled out by the p-type impurities when the statements are superimposed. In "impurity_5"

this concept is again tested by breaking the cell into the same three sections as in "impurity_3",

but assigning each section a p-type doping of 5 x 1016 cm-2. Then, the CdSe section is re-

specified as n-type with a concentration of 6 x 1016 cm-2

The files impurity_4 and impurity_5 failed to converge. It is interesting to note that

"impurity_5" failed, but "impurity_2" did not. Both appear to perform the same actions of

filling the entire cell with p-type doping at a concentration of 5 x 1016 cm-2, and then applying an

n-type doping of 6 x 1016 cm-2 to the CdSe region. The file "impurity_4" directly applied an

impurity profile of 1 x 1016 cm-2 in the CdSe region.

J-V curves from the three successful simulations are shown in Figure 2-3. From the graph,

it is obvious that "impurity_2" was significantly different than "impurity_l" or "impurity_3".

Both "impurity_l" and "impurity_3" show nearly identical J-V curves, suggesting that the

calculation is very similar. It is interesting, however, that the results were not identical.

The curve for file "impurity_2" shows an open circuit voltage of 0.185 V, approximately

30% lower than the values of 0.246 and 0.253 V for the other curves. This would seem to be









indicative of a lower effective doping density. The purpose of input files "impurity_4" and

"impurity_5" was to determine this difference, but these simulations failed to converge.

Further simulations demonstrated the effect of doping density in the CdSe nanorods, with

the results shown in Figure 2-4. The CdSe doping density was varied from 3 x 1016 cm-2 7 x

1016 cm-2. The graph shows a linear dependence of Voc on the CdSe doping density. This

dependence is further illustrated in Figure 2-5, along with the linear trendline fit through the

data.

The fit in Figure 2-5 predicts a Voc of 0.181 V for an undoped semiconductor nanorod and

a value of 0.193 V for a doping level of 1 x 1016 cm 2. The curve generated from the input file

"impurity_2" displayed an open-circuit voltage of 0.185 V, which corresponds to a doping

density of 3.3 x 1015 cm-2 according to the trendline in Figure 2-5. However, this linear trend

should not hold for doping levels approaching zero, because zero doping in the CdSe region

should coincide with a Voc of zero, due to the lack of a p-n junction in the device. Any error

from this value could arise from a space-charge region forming between P3HT and one of the

contacts.

From these simulations, the optimal method for defining impurity profiles was determined

to be the use of the "REGION" statement. This is simpler than defining the impurities for

multiple regions as in the "impurity_3" simulation. The "REGION" statement ensures that each

region is properly assigned the appropriate doping density.

Cell dimension adjustments

After definition of the basic model parameters, simulations were performed to determine

the effect of varying cell dimensions. The first parameter adjusted was the cell thickness, which

was accomplished by adjusting the length of the nanorod while maintaining a 10 nm polymer

capping layer. These simulations were performed with the CdSe nanorod width set to 4 nm









rather than 10 nm. This sets the nanorod half-width in the unit cell simulation area as 2 nm. The

resulting simulated J-V curves are shown in Figure 2-6. The length listed in the legend is the

unit cell thickness, not including electrodes, so the CdSe rod length is 10 nm shorter than that

distance due to the constant 10 nm capping layer.

The cells showed an increase in Voc, Jsc, FF, and efficiency as the cell thickness increased

due to increased absorption in the devices. However, due to increasing series resistance, there

was a diminishing return as the film thickness increased. Interestingly, the short-circuit current

density continued increasing as the film thickness was increased to 1 [tm, despite P3HT's low

hole mobility of 0.01 cm2/V-s.

Figure 2-7 displays solar cell performance measures for the simulated J-V curves shown in

Figure 2-6, with additional data points that were not displayed in Figure 2-6 for clarity. The

short-circuit current density of the unit cells continues to increase with cell thickness, although it

begins to level off. The fill factor and Voc of the cells reach a maximum in this simulation, with

the value of the 1 [tm cell showing slightly lower values than the 500 nm cell. The peak values

are approximately Voc = 60 mV and FF = 0.36. The efficiency continues increasing up to the 1

[tm cell thickness, but as with the short-circuit current, the rate of increase slows dramatically. A

summary of the results from these simulations is shown in Table 2-5.

With the effects of varying cell thickness characterized, simulations were then performed

to determine the impact of nanorod width on cell performance. Similar to the thickness

variations displayed previously, these simulations maintain a constant P3HT thickness of 10 nm

on the top and side of the CdSe nanorod. The resulting J-V curves are shown in Figure 2-8. The

values displayed in the legend represent the nanorod half-width rather than the unit cell width.

The unit cell width is 10 nm higher than the listed value due to the constant 10 nm of P3HT on









the side of the nanorod. The results show that performance increases with increasing nanorod

width, which is counterintuitive. The absorption coefficient for CdSe is significantly lower than

that of P3HT over most wavelengths, so the photocurrent generation increase with the wider cell

is expected to be minimal. Additionally, the current density calculation involves dividing the

total current by the cross-sectional area of the cell, which increases as the unit cell width

increases.

One interesting feature of the simulation is that nearly all the cell widths showed a short-

circuit current density of approximately 9 mA/cm2, with the exception of two. The curves for 15

nm and 20 nm CdSe nanorod half-widths showed a short-circuit current density of approximately

6 mA/cm2. To further explore this strange behavior at 15 and 20 nm nanorod half-widths, a full

range of half-widths was explored between 10 and 25 nm, with the resulting J-V curves shown in

Figure 2-9. The results show that the drop in short-circuit current density is a collection of

seemingly arbitrary thickness values that result in a reduced current density rather than a trend in

the range from 15 20 nm of CdSe half-thickness..

The cells with half-thicknesses of 15, 20, and 23 nm showed a short-circuit current density

in the range 5.4 to 6.4 mA/cm2, the cell with a half-thickness of 10 nm resulted in a short-circuit

current density of 8.5 mA/cm2 while all other cells were slightly above 9.0 mA/cm2.

Figure 2-10 shows the collection of cell performance data generated through the

simulations involving variations in the width of the CdSe nanowires, and this data is tabulated in

Table 2-6. The odd behavior of the short-circuit current density is a prominent feature in the

graph, with the cells at 15, 20, and 23 nm half-thickness showing a significant drop in Jsc. This

results in a drop in cell efficiency at these points, from over 3% to approximately 2%. With the

exception of these outlying data points, the short-circuit current density shows a minor but steady









decrease as the CdSe width grows from 5 to 30 nm. This is to be expected as series resistance

and the voltage across the junction increase and the cell current is being divided by increasing

cell areas to calculate current density. The cells' fill factors remain steady for CdSe half-widths

between 10 and 25 nm. At 30 nm, the Voc become very large (> 0.9 V), which causes a

significant drop in the fill factor (< 0.5). These values were unable to calculated directly, so they

are not included in Table 2-6.

Illumination source

Simulations were performed to determine the effect of altering the illumination source.

Medici generates photons from an illumination source at a specified location that illuminates the

sample at a specified angle and beam width. The illumination used for the cell dimension study

used the form

PHOTOGEN RAYTRACE BB.RADI BB.TEMP=5800 WAVE.START=0.2
+ WAVE.END=1.0 WAVE.NUM=@WL X.ORG=0.006 Y.ORG=-2.5 ANGLE=90
+ INT.RATIO=1E-2 N.INTEG=10 RAY.N=1 RAY.W=1.0

This code generates radiation from a blackbody source at 5800K, which simulates the

AMO radiation spectrum. The statements X.ORG= and Y.ORG= define the origin of the light

source. In the example above, the source is 6 nm to the right of the origin (centered on a 12 nm

unit cell) and 2.5 microns off the surface of the cell. The Y.ORG value is negative because the

surface of the device is at y = 0, so the area above the top of the device is in the negative y range.

The light is incident at 900 on the surface, with a ray width (RAY.W) of 1 am.

The results of varying the coordinates of the light origin are shown in Figure 2-11. Three

of the four data sets are identical, with only one showing a difference. The perfect match

between the two simulations with a centered light source shows that the distance away from the

surface is not a factor. Additionally, placing the source at the left edge of the cell shows no

change. However, placing the light source at the right edge of the cell results in a total loss of









photocurrent in the cell. The J-V curve shows the same exponential growth as the other

simulations and matches the other curves at large voltages where the current is primarily driven

by applied voltage rather than photogenerated carriers.

The lack of photocurrent generation when the sample is illuminated from over the right

edge is puzzling. According to the MEDICI manual, the ray will generate on each side of the

specified origin [65]. In this simulation, that should mean that the beam will generate from a

range of 0.5 [am on each side of the origin. Because the cell is 12 nm wide, this should easily

illuminate the entire cell. This is illustrated in Figure 2-12. Because the angle of incidence

specified is 900, generated rays will be perpendicular to the top surface of the cell. As shown in

the illustration, the beam width of 1 [tm would be more than sufficient to illuminate the 12 nm

cell width.

The simulations shown in Figure 2-11 were performed a second time with the input file

modified to remove the RAY.W statement. This allows MEDICI to set the beam width, which

by default is 2x the greatest dimension of the substrate [65]. The results of this simulation are

not shown because they were identical to the curves shown in Figure 2-11. There was no

photogeneration when the illumination source was placed at the right edge of the unit cell. The

cause of this phenomenon is unclear, but it is obvious that placing the beam origin at the right

edge of the simulation area results in a total lack of photocurrent.

Summary

These preliminary simulations serve as a guide for how to appropriately define commands

in Medici, as well as to create a basic understanding of the effect of certain model properties on

the cell performance. It was found that impurity profiles should be defined using the REGION

command to ensure the desired doping density is applied in each region. It was also found that

the illumination source should be placed in the center of the cell, but that the distance from the









surface of the cell and the definition of the ray width are unimportant, provided that the ray width

is high enough to illuminate the entire cell. Further studies will focus on the simulation of a real-

world hybrid solar cell from literature.

Simulation of a Real Cell

Model Parameters

The cell chosen for Medici simulation was reported in 2006 by Olson et al. [75]. The cell

was a hybrid cell fabricated from an array of ZnO nanofibers and P3HT, and has a cell structure

of ITO/ZnO/ZnO:P3HT/Ag. The ZnO layer was spin-coated onto the ITO substrate and the

nanofibers were grown through hydrothermal growth. P3HT was then spin-coated at a reported

thickness of 200 nm. SEM images of the ZnO nanofiber array before and after P3HT spin-

coating are shown in Figure 2-13. The performance measures reported for the device were as

follows: Voc = 440 mV, Jsc = 2.2 mA/cm2, FF = 0.56, and rI = 0.53%. The J-V curve for this

device is shown by the solid line in Figure 2-14.

From the SEM images shown in Figure 2-13 (a), the average height of the ZnO nanofiber

array is approximately 260 nm. From Figure 2-13 (b), the thickness of the full ZnO:P3HT layer

is approximately 430 nm. These values leave a P3HT capping layer of approximately 170 nm on

top of the nanofiber array. The image in Figure 2-13 (a) shows an average rod diameter of

approximately 30 nm. The authors note the rod spacing is approximately 100 nm. It is

extremely difficult to accurately estimate the spacing from the cross-section SEM image shown

in Figure 2-13 (a), but this figure appears to be reasonable and was likely verified with other

unpublished data, so it will be assumed to be accurate. From the SEM image, a thin base coat of

ZnO is visible on the ITO film. This film was estimated to be 25 nm thick from the SEM

images, and this height was included in the 260 nm rod length.









The authors note that the thickness of their P3HT film was 200 nm, but are somewhat

unclear whether this 200 nm thickness is measured from the top of the nanofiber structure or if it

represents the full film thickness. If the scale bars for the SEM images are reliable, it appears

that the 200 nm figure refers to the excess P3HT film on top of the fiber array. This seems to be

an excessive amount of capping considering that the exciton diffusion length of P3HT is on the

order of 10 nm in P3HT [15], a fact which is referenced by the authors. However, light is

incident from the ITO/ZnO side of the cell in the reported device, so photons must pass through

more than 200 nm ofZnO:P3HT film before reaching this capping layer. Additionally, the

inconsistent length of the ZnO nanofibers must be considered, as the relatively high mobility of

both electrons and holes in ZnO compared to P3HT would create a short-circuit in the device if

the wire tips were exposed. With this consideration, the additional buffer layer thickness may be

important experimentally to ensure that the device will function as a diode.

The J-V curve shown in Figure 2-14 was converted to numerical data using a graph

digitizer program [76] so it could be compared to simulated curves. The converted J-V and P-V

curves are shown in Figure 2-15. To verify the effectiveness of this conversion, the published

cell performance measures were compared to the performance measures calculated from the

digitized curve. The results are shown in Table 2-7. The results were Voc = 0.44 V, Jsc = 2.2

mA/cm2, FF = 0.57, and rI = 0.55%. This shows exact matches for Voc and Jsc and values for

FF and rI that are 0.01 and 0.02 higher than the published results. This is less than a 5% error in

both cases, and demonstrates that the curve was re-produced with a high accuracy. This new

curve can now be plotted with the results of simulated cells to find a best fit.

From previous work performed by Dr. Woo Kyoung Kim, the material properties for ZnO

were defined as shown in Table 2-8 [66].









The unit cell used for these simulation efforts is shown in Figure 2-16, consisting of half

the width of a nanorod and half of the space between the nanorod in the x direction and the full

cell thickness in the y direction. The unit cell dimensions were set based on the published cell

dimensions observed in Figure 2-13. In the unit cell, the transparent conducting oxide of ITO is

shown as the top contact. There is a thin 25 nm layer of ZnO coating the ITO electrode, and the

nanorods extend out from this structure. The P3HT region exists beside and above the nanorod.

The back electrode of Ag completes the structure. From the SEM images in Figure 2-13, the

nanorod width was approximated as 30 nm and the average rod spacing was approximated as

100 nm. Because the unit cell consists of half of a nanorod and half of a rod spacing distance,

the rod width shown in the unit cell is 15 nm and the P3HT width is 50 nm, resulting in a unit cell

width of 65 nm.

The model contains two regions of P3HT, a photocurrent generating region and a non-

generating region. For convenience, the region shown in Figure 2-16 c) will be listed as P3HT1,

and the region is Figure 2-16 d) will be listed as P3HT2. The P3HT1 region is the region within

one exciton diffusion length (LD) that generates photocurrent. Any references to the P3HT1

region, the LD region, or the P3HT absorbing region refer to this area. The P3HT2 region is

outside of the diffusion length region, and conducts holes to the Ag elecrode but does not

contribute to photocurrent in the cell. This division is included to account for the effects of the

exciton diffusion length in the polymer. The P3HT2 region is interchangeably referred to as the

P3HT bulk region. The effective diffusion length (LD) is a parameter that can be adjusted in the

simulations, but has been shown experimentally to be on the order of 10 nm [15].

Materials properties will be adjusted in the non-generating region to prevent photocurrent

from originating in this region. Several potential adjustments will be considered, including









reduction of the electron mobility to zero, setting the electron free carrier lifetime to zero, or

setting the absorption coefficient in the region to zero for all wavelengths. For simulation

purposes, it is assumed that the ITO electrode is perfectly transparent and that the Ag electrode

has a reflectance of 90%.

The unit cell structure as seen in Medici is shown in Figure 2-17. In this image, the blue

region is ZnO, the red region is the P3HT1 region and encompasses all areas within 10 nm of the

ZnO, and the green region is the P3HT2 region which includes all areas outside of 10 nm from

the ZnO. Note the difference in scale between the x- and y- axes.

Initial Simulations

Simulations were performed using the materials properties found in Table 2-1 for P3HT

and Table 2-8 for ZnO. The J-V curves for these initial simulations are shown in Figure 2-18,

along with the curve for the literature cell. These simulations demonstrate the effect of some

variation in the properties for the P3HT2 region. In simulation tsf496_1, the absorption

coefficient for the P3HT2 region is set to zero for all wavelengths. In simulation tsf496_2, no

changes are made to the film properties, so all parts of the P3HT region are allowed to generate

and transport free carriers. In simulation tsf496_4, the electron mobility is set to 0.0001 cm2/V-s

in the P3HT2 region, representing a reduction by one order of magnitude from the original value.

The simulation tsf496_2 is a case where the entire P3HT region is treated as a photocurrent

generating layer. This represents a case where the exciton diffusion length is infinite. This

should represent the maximum possible performance of a cell constricted by the materials

properties shown in Table 2-1 for P3HT. The efficiency of that simulated cell was 5.55% with a

short-circuit current of 11.66 mA/cm2. The simulation data range was stopped at 0.9 V to limit

the size of output files and the simulation time, so a direct measurement of Voc was unable to be

obtained. From the shape of the curve, it appears that the Voc would be approximately 1.0 V.









Those numbers give an estimated fill factor of approximately 0.5, which is relatively low

considering that this represents a best case scenario.

In simulation tsf496_4, the electron mobility is reduced by an order of magnitude in the

P3HT2 region. This was an attempt to allow realistic absorption in the region but strictly limit

the ability of the electrons generated in this area to reach the ZnO and contribute to the overall

photocurrent. The resulting J-V curve showed a strange double-curve behavior, with the

maximum power point occurring in the second elbow at approximately 0.75 V. The double-

curve behavior is not physically realistic, but it is interesting to note that the performance of this

simulation was higher than the performance of tsf496_1, showing that even with the low electron

mobility, current generated in the P3HT2 region was able to make a contribution to cell

performance.

Simulation tsf496_1 shows the case where the absorption coefficient is set to zero in the

non-generating P3HT2 region. Of the three curves shown in Figure 2-18, this is the most similar

to the real curve. Although the Voc and Jsc were significantly higher than that of the real cell,

the shape of the curve seems to be very similar. The current density increases very slowly over

the low-voltage regime until approximately 0.6 V, where there is a distinctive elbow in the curve

and the current begins increasing more rapidly. The slope of the curve as it approaches Voc

appears to approximate that of the real curve.

Reflectance of Ag electrode

Because the incident light angle is set at 900 in simulation tsf496_1 and the boundaries

between layers are set at right angles, one physical inconsistency in this model is the amount of

reflected light that returns to the P3HT1 and ZnO regions. On the left-hand side of the cell where

the nanorod resides, reflected light in the real cell would need to pass through 320 nm of the non-

absorbing P3HT2 region between its first and second passes through the absorbing P3HT1 region.









On the right-side of the cell, light would have to pass through 790 nm of the non-absorbing

P3HT2 region before returning to the thin absorbing layer of polymer and ZnO at the surface of

the cell. Based on this, it is obvious that this simulation allows more light to be absorbed in the

generating regions of the model than would exist in reality. This was corrected by studying the

effect of reducing, reflection at the back electrode.

The absorption coefficient of P3HT ranges from 2 x 104 cm-1 at X < 350 nm and X > 680

nm to a maximum value of- 2 x 105 for X = 540 nm. At the low-end absorption value,

approximately 20% of incident light is absorbed over the 10 nm LD region, according to

Equation 2-1.


I0
_e (2-1)


Ii represents the amount of light transmitted through the film, where Io is the intensity of incident

light, a is the absorption coefficient, and x is the film thickness. Nearly 99.8% of the remaining

photons would be absorbed over the 320 nm path length consisting of the forward and backward

pass through the non-absorbing region in the simulation. Even in this case, with a low

absorption coefficient and the shortest considered path length, virtually no photons would remain

in a real cell to be absorbed after reflection from the back electrode.

With this consideration, the next line of simulations was performed with reflectance at the

Ag electrode either reduced or completely removed. The resulting J-V curves are shown in

Figure 20. Despite the removal of all reflectance from the Ag electrode, simulated cell

performance was still substantially higher than that of the real cell for all simulations. All of

these simulations prevent absorption in the P3HT2 region.

By comparing simulations with varying electrode reflectance and constant mobility and

absorption properties (mobilities as defined in Table 2-1, absorption set to zero in the P3HT2









region) we can see the effect of reflectance at the Ag electrode with no absorption in the P3HT2

region. These curves are shown in Figure 2-19. As expected, the short-circuit current density

decreases as the Ag reflectance is decreased. The simulation with the Ag reflectance set to zero

results in a Jsc = 3.6 mA/cm2, which is still approximately 1.5x higher than the real cell value of

Jsc = 2.2 mA/cm2.

Mobility

In addition to the removal of reflectance from the back electrode, the charge mobilities in

P3HT were reduced to restrict current flow in the device. The physical justification for this is

that the initial value of hole mobility was taken from the literature [71], and was the highest

reported value for hole mobility in P3HT. Additionally, this value is a field-effect mobility

which relies on a strong applied bias to drive current flow. In photovoltaic devices, biases are

significantly lower, and this field-effect mobility may be a serious over-estimation of the true

film properties. Also, charge mobility has been shown to be anisotropic in P3HT [77], so values

of mobility can vary depending on the alignment of the polymer chains in the film.

Based on these reasons, simulations were performed to evaluate the effect of reducing

mobility values in the polymer regions of the model, with the resulting J-V curves shown in

Figure 2-20. As predicted, simulated cells with higher mobilities resulted in stronger device

performance. The black and green curves, corresponding to no change in mobility values and a

50% reduction of mobility values in the P3HT2 region only, are nearly identical. A change from

the initial mobility values to a 50% reduction in both P3HT regions, shown by the black and red

curves, results in an efficiency drop from ir = 2.12% to ir = 2.00%, or approximately 6%.

Reducing the P3HT mobility b 90% in both regions results in an efficiency drop of

approximately 23%, from i = 2.12% to i = 1.64%. Attempts to reduce the P3HT2 mobility









values by 90% while holding the P3HT1 values at their original values failed to converge in

Medici.

Although the J-V curves generated on this graph stop at a value of 0.9 V before reaching

the open-circuit voltage, extrapolations estimate that the Voc is virtually unchanged in these

simulations, with a value of approximately 1.04 V. Using this estimation, cell performance

measures were calculated and are displayed in Table 2-9. The short-circuit current density

decreases with each drop in the mobility. For the original mobility values, Jsc = 3.61 mA/cm2.

For 50% and 90% reductions in mobility, the Jsc drops to 3.58 mA/cm2 and 3.44 mA/cm2,

respectively. Using the extrapolated value of Voc = 1.04 V, fill factors for each cell can be

calculated as 0.57, 0.54, and 0.46 for mobility values of 100%, 50%, and 10%, respectively.

This reduced fill factor is obvious from Figure 2-19, as the curve corresponding to a 90%

reduction in mobility shows an obvious reduction in the sharpness of the "elbow" shape of the

curve. Based on these values, a 50% decrease in mobility resulted in a 6% drop in efficiency, a <

1% drop in Jsc, and an estimated 5% drop in fill factor. The 90% decrease in mobility resulted in

a 23% drop in efficiency, a 5% drop in Jsc, and an estimated 19% drop in fill factor.

Exciton diffusion length

Even with a reduction in charge mobilities by 90% in the P3HT regions, complete removal

of absorption in P3HT outside of a diffusion length, and elimination of all reflection from the

silver electrode, simulated cell performance greatly exceeds that of the real cell. The simulation

with the lowest performance showed a Jsc of 3.44 mA/cm2 and i of 1.64%, along with estimated

values for a Voc of 1.04 V and a fill factor of 0.49. To match the published experimental data,

the Jsc must be reduced by another 1.2 mA/cm2, the Voc by 0.6 V, and the efficiency by 1.1%.

The exciton diffusion length was adjusted downward from the literature value of 10 nm

[15] to compensate for this inconsistency. As noted previously, the exciton does not exist in









Medici, so the definition of an exciton diffusion length in this study is taken to be the width of

the P3HT1 region. Since this is the only P3HT region generating photocurrent in the model, it is

taken to be an accurate approximation of the exciton diffusion length in a real cell. For this

study, with results shown in Figure 2-21, the absorption coefficient was again set to zero in the

P3HT2 region and the carrier mobilities in both P3HT regions were set to their original values of

Ln = 0.001 cm2/V-s and ip = 0.01 cm2/V-s. As predicted, the simulated cell performance

decreased as the exciton diffusion length decreased. This was a result of decreasing short-circuit

current density, which decreased linearly with exciton diffusion length, as shown in Figure 2-20.

Estimates for open-circuit voltage actually increased slightly as the diffusion length decreased, as

shown in Figure 2-22. This trend was very slight, however, so it showed virtually no impact on

cell performance. Based on these simulations, an exciton diffusion length of 6 nm was chosen

for future simulations.

Doping density

From previous simulations of CdSe:P3HT hybrid cells, changing the doping level of CdSe

showed a strong impact on the Voc of the cells. Based on these results, the doping density of

P3HT was varied from the initial value of 5 x 1016 cm-3 to 1 x 1014 cm-3 in an attempt to reduce

the Voc of the simulated cells. The simulated J-V curves are shown in Figure 2-24. Contrary to

expectations, the Voc of the cells did not appear to change with the reduction in P3HT doping.

A key difference between this study and the simulations performed on the test cell of CdSe

and P3HT is that CdSe and P3HT showed very similar doping densities: 6x1016 cm-3 for CdSe

and 5x1016 cm-3 for P3HT. The doping density of ZnO is 5x1017 cm-3, so it is possible that the

Voc is dominated by this wide discrepancy.

To test this hypothesis, simulations were performed with reduced ZnO doping density.

These tests fixed the P3HT doping density at lxl016 cm-3, based on the results shown in Figure 2-









24. The results from this test are shown in Figure 2-25. Oddly, this variation fails to show a

decrease in Voc. The Voc of a cell should depend on factors such as doping density and

minority carrier lifetime [78], but the doping density variations here doesn't show any impact.

Density of states

Simulations were performed to determine the effect of the density of states in the

conduction and valence band of P3HT, and the resulting J-V curves are shown in Figure 2-26.

The curves show a decrease in performance as the density of valence band states increases,

although extrapolated values for open-circuit voltage remain relatively constant. The curves are

independent of the density of conduction band states, as shown by the overlap of the black and

red data points and the green and yellow data points.

Open-circuit voltage examination

Simulations focused on the effect of carrier mobility were again performed, but with an

expanded voltage range so that the open-circuit voltage could be observed rather than estimated.

These resulting J-V curves from these simulations are shown in Figure 2-27. As predicted

through previous extrapolations, the open-circuit voltage is independent of the carrier mobility

and the short-circuit current density is strongly impacted by it. Extrapolated values for Voc in

previous simulations predicted values near 1.04 V, but these simulations show that Voc = 1.27 V

for these cells. As noted earlier, the cells show a double-elbow effect that was unable to be

explained.

The effect of P3HT doping density was also reexamined with the expanded voltage range,

as shown in Figure 2-28. As predicted through previous extrapolations, the Voc is independent

of the doping density of P3HT. Again, the Voc = 1.27 V for these simulations, and the short-

circuit current density decreases as the P3HT doping density decreases. This mimics the

behavior seen in previous studies on this material system.









Although the short-circuit current density and fill factor can be altered by variations in

carrier mobility, doping density, and valence band density of states in P3HT, these tools offer no

control over the open circuit voltage of the simulated cells. From a theoretical perspective, the

open circuit voltage of organic and hybrid cells is controlled by the energy difference between

the HOMO level of the electron donor and the LUMO level of the electron acceptor, or the

conduction band level if the acceptor is an inorganic [79]. In the case of a hybrid cell, the

LUMO level of the organic electron acceptor is replaced by the conduction band energy of the

inorganic material. In other words, carriers do not exist with energy equal to the band gap of the

material where they are generated; instead they exist with energy equal to the spacing between

the bands of the junction materials. The energy band diagram for the P3HT ZnO system

displayed in Figure 2-29 shows this band offset to be 0.85 eV. Based on this, free carriers

resulting from photoabsorption in P3HT should exist at an energy of 0.85 eV rather than 1.7 eV.

Medici allows independent specification of the energy band gap and optical band gap in

each material. The optical band gap should remain fixed at the appropriate level to dictate the

location of the absorption cutoff for each material. The energy band gap, however, presents an

additional tuning control to describe the energy of photogenerated carriers in the device.

Figure 2-30 shows the results of simulations with variations in the energy band gap of

P3HT. In the curve tsf496_45, the energy band gap is set to 0.85 eV in the P3HT regions of the

cell. In curve tsf496_53, both the energy band gap of both the P3HT and ZnO regions is fixed at

0.85 eV. From these curves, fixing the energy band gap ofP3HT accurately simulates the Voc of

the literature cell. Altering the ZnO band gap in addition drops the Voc to below 0.4 V, beyond

the published cell performance data.









Absorption coefficient adjustment

Under close inspection of a Medici output file, it was noticed that the program

automatically assigns absorption coefficient values for wavelengths outside of the range

specified in the input file if this range is smaller than the full spectrum used for calculation.

These values are not set to zero, but assigned a default value based dependent on the wavelength.

Figure 2-31 shows the absorption coefficient curves as assigned by Medici for three absorption

input files. The first two files, absnone and abs_p3ht, have an input range specified from 0.3 to

0.7 pm and correspond to zero absorption and the absorption profile ofP3HT, respectively. The

range of 0.3 to 0.7 pm was chosen because this is the range of data shown in the absorption

curve from literature [74]. The file absizno was borrowed from Dr. Woo Kyoung Kim in his

simulations of CIS solar cells [65]. This input file ranges from approximately 0.2 to 1.0 rm.

From Figure 2-31, it is obvious that Medici assigns identical values to all absorption files

in the range from 0.12 to 0.2 rm. Similarly, it assigns an identical data point to all files at 1.24

rm. Although absorption in the high wavelength region is set to zero by the program, absorption

in the short wavelength region is set to a very high value in the range of 106 cm-1. Although the

photon flux is low in this region, as shown in the AM1.5 spectrum in Figure 2-32, this extremely

high absorption coefficient leads to a significant amount of carrier generation in the cell. This

leads to an over-estimation of the number of free carriers in the cell and artificially enhances cell

performance in the simulations. This effect has no impact on the ZnO absorption, because in the

simulation photons are generated from 0.2 to 1.0 tm. Figure 2-33 shows the number of

photogenerated carriers as a function of wavelength.

Figure 2-33 shows the carriers generated by this absorption region between 0.2 and 0.4

rlm, but does not show the exact effect on the J-V curves. In Figure 2-34, J-V curves are

compared for cells simulated with the P3HT absorption profiles shown in Figure 2-31 and an









absorption profile with all absorption between 0.2 and 0.3 im set to 2.24x104 cm-1. The curve

with the corrected absorption coefficient, shown in blue, shows a drop of 0.14 mA/cm2 in the

short-circuit current when compared to the curve with Medici-assigned values for the absorption

coefficient, shown in purple. This difference is minimal, but should be included in simulations

for improved accuracy in the simulations.

Multi-stage absorption

The technique used for simulations up to this point involved a 2-layer P3HT region with a

thin absorbing region near the ZnO region and a large non-absorbing region. The width of the

absorbing region was set equal to the assumed diffusion length of an exciton. Because the

diffusion length of an exciton is not explicitly included in the program, Medici assumes that all

photogenerated carriers within this absorbing region are free carriers, which are driven across the

space-charge region by the built-in electric field. This is only a first-order approximation of the

true device physics of a hybrid solar cell. The exciton diffusion length is an average distance

that an exciton can travel before it recombines, not a firm line beyond which all excitons are

doomed to recombination. The excitons do not selectively move toward the inorganic organic

interface, they simply travel by diffusion until they recombine or dissociate at the interface. To

approximate this distinction, the model was modified to include multi-stage absorption in the

P3HT region where the probability of excitons reaching the interface is included.

A framework for a simulation involving multiple layers in the absorbing region was

developed with a multi-layer region extending to 20 nm from the ZnO region. J-V curves shown

in Figure 2-35 correspond to simulations in which the first 10 nm of this region is set as an

absorbing region using the P3HT absorption profile. The second 10 nm is set as a non-absorbing

region, identical to the "P3HT2" region. These cells all have identical structures, with the only

difference being the number of stages within this 20 nm region. For example, the simulation









with 10 stages has 10 layers that are each 2 nm thick. The first 5 stages are set to absorb based

on the P3HT absorption profile, and the final 5 stages are set as non-absorbing. Similarly, the

simulation with 4 stages contains 4 layers with 5 nm thickness in each layer, with the first two

set to absorb and the final two set as non-absorbing. In the case of the single stage cell, the

absorption region was set from 0 to 10 nm away from the ZnO interface, and the "P3HT2"

region began at 10 nm.

Because the cell structures are identical in terms of exciton diffusion length, simulations

were expected to produce identical J-V curves. Although the curves are all similar in shape and

performance data, they are not identical, due to slight differences in the iterative calculation in

Medici. In all the simulations shown in Figure 2-35, absorption was set to 100% of the P3HT

value in all stages between 0 and 10 nm into the P3HT region, and it was set to 0 in all stages

between 10 and 20 nm.

This result is interesting, but is not the goal for this multi-stage model design. The purpose

is to create a graded carrier generation profile that will average to the specified LD, but allow for

some carriers inside of LD to fail to reach the interface and allow some carriers outside of LD to

succeed in reaching the interface. The structures used to generate the J-V curves in Figure 2-35

can be thought of as normal curves with standard deviations of zero. Absorption is 100% of the

P3HT value up to the specified diffusion length of 10 nm, and then drops as a step function to 0%

of the P3HT value for the next 10 nm. Using this multi-stage framework, distribution curves

with nonzero values for standard deviation can be imposed on the simulation. In this case the

property being distributed is the absorption coefficient, as this is the property chosen to simulate

the exciton diffusion length. It should be noted that these curves are not exactly equal to

cumulative distribution functions for a true normal distribution because in this case the area









considered was only between 0 and 20 nm. In a true normal distribution with a mean of 10 nm

and a standard deviation of 4 nm or greater, an appreciable area exists under the curve beyond 0

and 20 nm. To remove this area from consideration, normalization was performed by dividing

by the sum of the area between 0 and 20 to establish these boundaries. The distribution was

calculated as shown in Equation 2-2.


F(x)= ex 2 (2-2)


In Equation 2-2, x is the distance into the P3HT region and x is the mean of the distribution, set

to 10 nm in this case. By dividing this by the sum of all distribution values between x = 0 and x

= 20 nm, the normalized distribution was obtained. The cumulative distribution was calculated

using Equation 2-3.


F(x)
G(x)= 1 =0 (2-3)
x=20

x=0

Again, this cumulative distribution is normalized by the area under the distribution function

between 0 and 20 nm rather than between negative and positive infinity. This cumulative

distribution function is shown in Figure 2-36 for a range of standard deviations and a mean of 10

nm. By coupling these distribution functions with the multi-stage cell design demonstrated in

Figure 2-35, graded absorption can be generated in the simulation by applying a fractional

absorption coefficient in each stage of the absorption region.

Simulations were performed to consider the effect of this absorption distribution in the

cells. These simulations used a 10-stage absorption region with a graded absorption coefficient

in an attempt to more accurately define the absorption in the cell. Absorption profiles

corresponding to standard deviations of 0 to 10 nm were generated by using a fractional value for









the absorption coefficient in each region, rounded to the nearest 10% value. The input data used

for the simulations are shown in Table 2-11, as well as the resulting short-circuit current density

for that simulation. The J-V curves for these simulations are shown in Figure 2-37.

Table 2-11 shows that absorption in the cell is distributed over a wider range as the

standard deviation increases. The average multiplier to the absorption coefficient is calculated

for each situation and as targeted it is equal to 50% or 51% for all cases, with any error coming

from rounding to the nearest 10% in each region. The J-V curves in Figure 2-36 show that this

grading has virtually no effect on the final performance of the cells. The short-circuit current

density shows extremely minor variations in the curves, and this value is tabulated in Table 2-11.

The short-circuit current density increases slightly as the absorption distributions become wider.

This shows that extending the generated carrier distribution toward the back electrode improves

device performance, while limiting the carrier distribution to a narrower region, even with the

same number of carriers being generated, hinders performance.

P3HT replacement with CIS

The J-V curves generated in previous simulations shows a double-elbow shape that is not

characteristic of photovoltaic cells. To determine the root cause of this phenomenon, P3HT was

replaced with copper indium diselenide (CIS), a popular p-type thin-film photovoltaic material.

The same cell structure is used, with ZnO nanorods as the n-type material. Within this

framework, material properties of the CIS layer were adjusted to values corresponding to P3HT.

Preliminary simulations in this form were performed, and the resulting J-V curves are shown in

Figure 2-38. This set of simulations compares ZnO:P3HT cells with 10 nm and full cell

absorption regions to ZnO:CIS cells with full cell absorption, a 10 nm absorption range, and a 10

nm absorption range with an energy band gap set at 0.85 eV.









Figure 2-38 shows that a limited absorption range in the simulations does not cause the

double-elbow shape of the J-V curve, as the ZnO:CIS cell with a 10 nm absorption region shows

a similar shape to the one with absorption in the full CIS region. The same holds true for

comparisons of the two ZnO:P3HT cells, which both show the double-elbow shape. The

ZnO:CIS cells produce J-V curves with the anticipated shape and high fill factor as compared to

the ZnO:P3HT cells. This also held true for the cell in which the CIS electrical bandgap was

reduced to 0.85 eV, although this cell resulted in a low Voc of 0.25 V.

Parameters used for P3HT and CIS in the simulations are displayed in Table 2-12. Data for

CIS was obtained from previous simulations from Woo Kyoung Kim [66]. The doping density is

the only parameter that is the same in both materials. By individually adjusting parameters

between the values for CIS and P3HT, an attempt will be made to determine the origin of the

double-elbow J-V curve shape. Although not shown in the table, the absorption profile for each

material was also adjusted.

Using the same 10 nm absorption region as the ZnO:CIS cell shown in red in Figure 2-38,

the properties shown in Table 2-12 were individually changed from the CIS values to the P3HT

values. The resulting J-V curves are shown in Figure 2-39. Performance measures for these

cells are displayed in Table 2-13. These results show that the permittivity and density of states

have virtually no impact on the cell performance. The three parameters that strongly impacted

the J-V curves are electron affinity, carrier mobility, and absorption profiles.

The change in electron affinity causes a shift in Voc from 0.44 V to 0.49 V due to the

higher conduction band level for P3HT. The short circuit current is minimally affected by this

change, but there is a notable drop in the fill factor of this curve, down nearly 30% from the

values of the original ZnO:CIS cell. The variation in carrier mobilities showed a 1.5 mA/cm2









reduction in short circuit current paired with a 1 V increase in open-circuit voltage. The reduced

mobility results in fewer carriers escaping the cell, causing a lower short-circuit current density.

The strong increase in Voc is more difficult to explain, as adjustments to carrier mobility in

previous simulations did not show this type of response. There is also a slight decrease in the fill

factor for the cell with P3HT mobility values, but this effect was minor. When the absorption

profile is changed from CIS values to P3HT values, the short-circuit current density of the

simulated cell drops by nearly 50%, down to 4.1 mA/cm2. The Voc of that cell is slightly

reduced, but the fill factor remains nearly unchanged.

Based on the J-V curves shown in Figure 2-38 and detailed in Table 2-13, simulation 72

was chosen as a basis for further efforts. In this simulation, the absorption coefficient values

were assigned values corresponding to P3HT rather than CIS, while maintaining a 10 nm

absorption range in the simulation. Variations around this base case were performed by

individually varying the other material properties from their CIS value to their P3HT value. The

results are shown in Figure 2-40, with performance measure data shown in Table 2-14. As in the

previous set of simulations, the addition of P3HT values for permittivity or density of states

showed virtually no change from the control cell. The values for open-circuit voltage, short-

circuit current density, and fill factor of these cells were within 2% of the values for the control

cell. As seen in the previous set of simulations, when the P3HT value for electron affinity is

applied the J-V curve shifts to a higher Voc with a 30% drop in fill factor and a virtually

unnoticeable drop in Jsc. The application of P3HT values for charge mobility results in a 0.8

mA/cm2 decrease in Jsc and a 0.1 V increase in Voc with an approximately 10% drop in fill

factor. Despite reductions in fill factor for simulated cells with P3HT values for electron affinity

and mobility, none of the J-V curves in Figure 2-40 displayed the double-elbow shape.









The electron affinity was fixed at its P3HT value of 3.15 eV, setting the conduction band

level in the simulations to the appropriate value for P3HT. The electronic band gap of the

simulations is still set at the CIS value of 1.04 eV rather than the P3HT value of 1.7 eV or the

effective band gap of 0.85 eV at the ZnO-P3HT junction. Again, the absorption coefficient is set

at the P3HT value and absorption is allowed over a 10 nm range near the material interface.

Variations in the other materials properties resulted in the J-V curves shown in Figure 2-41 and

detailed in Table 2-15. As seen in previous simulations, there is virtually no change in the J-V

curve or device properties with the application of the P3HT values for permittivity and density of

states. The application of P3HT values for carrier mobility, however, resulted in a drastic shift in

the nature of the curve. This change resulted in a 0.5 mA/cm2 reduction in the short-circuit

current density and an increase of -0.6 V in the open-circuit voltage. Additionally, the J-V curve

takes on the double-curve shape, which drops the fill factor to approximately 0.20. This shape

will be discussed in more detail after the next set of data.

As discussed previously, the J-V curves shown in Figure 2-41 resulted from simulations

where the electron affinity was set at the P3HT value of 3.15 eV, but the electronic band gap

remained at the CIS value of 1.04 eV. Although fixing the electron affinity sets the LUMO level

of P3HT, the band gap value sets an inappropriate HOMO level and results in the generation of

carriers with higher energy than appropriate. The simulations of Figure 2-41 were modified to

include the appropriate ZnO-P3HT interfacial band gap of 0.85 eV, corresponding to the energy

gap between the conduction band of ZnO and the HOMO level of P3HT. The results, shown in

Figure 2-42, mimic those in Figure 2-41.

The curves resulting from adjustments in the permittivity and density of states show the

expected shape for a solar cell J-V curve despite having a significantly lower Voc than their









counterparts in Figure 2-41. This variation is expected, due to the shift in energy band levels

causing the generated carriers to exist with more energy. The Jsc of the cells remains virtually

unchanged as compared to their counterparts in Figure 2-41, but the shift in Voc results in a

slight reduction of the fill factor as the curves cross the voltage axis at a slightly lower slope.

Curve 82, corresponding to a change in carrier mobility, also shows a reduction in Voc when

compared the curve shown in Figure 2-41. This curve shows a very small fill factor of

approximately 0.15 due to the double-elbow shape which eliminates most of the active area of

the curve.

The origin of the change in shape accompanying lower mobility values is unclear, but it

obviously occurs only when the absorption, electrical energy levels, and mobilities take on their

P3HT values. This shape was observed for all ZnO:P3HT cells simulated up to this point, but it

was not seen in other simulations using a ZnO:CIS cell as a basis. From the graphs in Figures 2-

39 2-41, it can be concluded can state that the permittivity and density of states have no effect

in causing this shape. Additionally, the application of P3HT levels of absorption and energy

levels to a ZnO:CIS cell did not cause this shape without the addition of P3HT levels for carrier

mobility. In the simulations, the carrier mobilities were set as [n = 0.001 cm2/V-s, ip = 0.01

cm2/V-s for P3HT, and Ln = 30 cm2/V-s, Lp = 300 cm2/V-s for CIS. Changing from CIS values to

P3HT values represents a severe drop of 4 orders of magnitude with no further details for

intermediate values.

Using simulations of a ZnO:P3HT cell, a wide range of mobility values were examined,

with the resulting J-V curves shown in Figure 2-43. Note that in all cases, Cp = 10*pn.

Additionally, these simulations allow absorption throughout the full P3HT region of the cell to

boost the level of current flow and more clearly show the effect of the mobility variations.









These simulations demonstrate that at higher mobility values, the J-V curve shows the

expected shape for a solar cell. At lower mobility values, the curve inverts to the double-elbow

shape. An ideal solar cell J-V curve should have a positive second derivative over the entire

active range, from V = 0 to V = Voc. It is difficult to calculate a second derivative for these

curves because they are not easily fit to empirical equations. However, the sign and approximate

magnitude of the second derivative over a short range of the data can be determined by using

Equation 2-4.

J"est = (y3 y) / 2 y2 (2-4)

A visual representation of this calculation is shown in Figure 2-44. The values yi, y2, and

y3 represent current density values for equally spaced applied voltages. For these J-V curves, y

is current density and x is voltage. The first term of Equation 2-4 is noted in Figure 2-44 as y2,

and is the midpoint of a line drawn through the points (xi, yi) and (x3, y3). If this value is greater

than the actual value of y2, J"est calculated from Equation 2-4 will be positive and will

correspond to a curve that is concave up, as shown in the figures. Although this is not an exact

measure of the second derivative, it will result in a positive value (the value for y20 is greater

than the value y2) for curves that are concave up and a negative value for curves that are concave

down. Additionally, the magnitude of J"est will give a relative idea of how close to linear the

curve is over the range from xi to x3.

This calculation is applied to the simulated J-V curves for ZnO:P3HT cells shown in Figure

2-43, and the calculated J"est values are plotted in Figure 2-45. Note that these calculations

correspond to cells with absorption in the full P3HT region. Of all the curves shown, only those

corresponding to I, = 5,000 and I, = 10,000 cm2/V-s showed second derivative values over the

entire active region.









When these same simulations are performed on ZnO:P3HT cells featuring carrier

generation only within a 10 nm exciton diffusion length, the curves shift significantly, as shown

in Figure 2-46. The most obvious and expected feature is that the current density drops

significantly due to the lower number of carriers being generated in the limited diffusion length

region. There is also a slight drop in the open-circuit voltage for these cells, but an increase in

the fill factor as the curves show less of an inverted shape. This is shown more clearly in Figure

2-47, which displays the J"est values for these simulated cells. Cells with mobility values as low

as p = 500 cm2/V-s show positive J"est values for the entire active region in this case, a full order

of magnitude smaller than the threshold mobility in the full cell absorption simulations.

Lower values of carrier mobility generate J-V curves with consistently positive second

derivatives in the 10 nm absorption region simulations as compared to the full cell absorption

region simulations. This refutes the theory that a second p-n junction region arises at the

interface of the absorbing and non-absorbing P3HT regions. Instead, the charges generated at a

greater distance from the ZnO-P3HT interface drive this inversion of the J-V curves. This is due

to the low mobility of electrons in P3HT and their difficulty in traveling through large regions of

the polymer to reach the ZnO regions.

The comparisons of ZnO:CIS cells demonstrated that three parameters are key to

controlling the shape of the J-V curve in these simulations. Carrier mobility dictates the shape of

the curve, as demonstrated in Figures 2-43 and 2-46. Altering absorption in the cell, such as by

changing the exciton diffusion length or the absorption coefficient, has a direct impact on the

number of charge carriers and therefore on the current density of the cell. This is demonstrated

in Figure 2-48 for cells with the P3HT hole mobility set to 500 cm2/V-s and a 10 nm absorption









region. The energy band gap of the P3HT region in the cell alters the open-circuit voltage of the

cell. This is shown in Figure 2-49 for cells with I, = 500 cm2/V-s and LD = 10 nm.

The curves shown in the two figures above demonstrate the level of independent control of

JsC and Vo that is available from variations in absorption and energy band gap, respectively. To

accurately simulate the experimental data, however, an array of combinations must be examined.

Figure 2-50 shows a set of curves spanning four values for each of these parameters. In the

graph, colors are constant for constant band gap, while line style is constant for constant

absorption coefficient values.

The experimental data falls somewhere in the range between 20% and 50% absorption

with a band gap of 1.0 eV. This set of curves was expanded to show a wider range of absorption,

all with the band gap held constant at 1.0 eV and the hole mobility set at 500 cm2/V-s. The

results are shown in Figure 2-51. Although the curve for 30% absorption produces a near-

perfect match for Jsc and a close match for Voc, the simulation produces a curve with a lower fill

factor than the real data. Based on this, the search must be re-expanded to three parameters:

carrier mobility to control fill factor, band gap to control Voc, and absorption coefficients to

control Jsc. However, it must be considered that each of these controls impacts all three target

parameters, not just the intended one.

Shortcomings of the initial model

This simulation strategy was abandoned as the physics of the model were more closely

analyzed. All simulations up to this point suffered from one distinct inconsistency the

existence of electrons in the P3HT region. This is demonstrated in the strong impact of electron

mobility on the shape of the J-V curve for ZnO:P3HT cells. The polymer absorbs photons to

generate excitons rather than free carriers, but the model is not aware of this distinction. As a

result, the electron mobility of P3HT is an important factor in these simulations, but it is of minor









importance in real cells. Because of this distinction, a new strategy has been employed to more

accurately represent exciton physics in Medici.

Two-Step Simulation Technique

The important feature of excitonic solar cells, as has been discussed, is the generation of

excitons rather than electron-hole pairs. One important consideration, addressed in simulation

efforts described previously, is the limited carrier generation due to the exciton diffusion length.

However, another important effect, not directly considered in previous simulations, is that the

exciton dissociates at the material interface. This means electrons should not be generated in the

P3HT region; instead, free carriers should be generated directly at the material interface. A new

modeling strategy is needed to incorporate this effect into the simulations.

This new strategy involves a two-stage simulation. In the first stage, the photogenerated

carrier distribution is measured over the entire device area under simulated solar illumination. In

the second stage, this distribution is compressed to a line source of carriers generated directly at

the organic-inorganic material interface by mapping the carriers generated at all points to their

nearest interfacial point.

Specifying a line source of carriers in Medici requires specifying the origin and endpoints

of the line with X and Y coordinates, as well as parameters to describe the carrier distribution

along that line, as shown in Equations 2-5 2-8.

G.(Crivdct^ ) = 1L" )' (t) (2-5)

L( )= 41 4 A2. -+ A axA4-- I (2-6)


r-r 'tLp-kl (2-7)

1() = 1 (2-8)









Gn and Gp are generated electrons and holes, respectively. This electron-hole pair generation

equation consists of three components: lateral, radial, and time-dependent. The physical

dimensions I and r represent lateral distance along the line and radial distance from the line,

respectively.

The lateral dependence can be specified as an exponential-linear function that requires the

definition of parameters Al, A2, A3, and A4. These parameters are determined through

regression for each scenario studied, and are discussed in further detail in the following sections.

The radial dependence follows an exponential decay using the decay constant R.CHAR,

which is set to 0.0001 [tm for these simulations. This decays the radial component of the

generation by 44 orders of magnitude within 1 nm of the line, so that the generation occurs only

at the material interface and not in the individual materials.

Carrier generation is assumed to be uniform in time. The time dependence is set to the

default value of T(t) = 1.

Photogenerated Carrier Distribution

The distribution of photogenerated carriers in the unit cell was determined through Medici.

The unit cell was exposed to simulated solar illumination as described in previous sections.

Medici calculates photoabsorption and carrier generation at each point in the simulation mesh,

and this data can be displayed as a contour plot. Unfortunately, it seems to be impossible to

extract the numerical source data from contour plots in the program. This is possible, however,

for line plots, as this is the technique used in to extract J-V data from the simulations.

Line plots were generated at each mesh point along the x-axis of the unit cell, and these

lines extended through the thickness of the device so that the carrier generation data for every

mesh point in the unit cell was collected. Using this (x, y, z) data set, contour plots were created

in Sigmaplot to show the distribution across the device area.









A contour plot of the photogenerated carrier distribution in the full cell and a surface plot

focused on the area within LD from the side of the nanorod are shown in Figure 2-52. The

exciton diffusion length for the cells depicted is 10 nm. Note that in the plots, y = 0 represents

the top surface of the cell, bordering on the ITO electrode. There is a 0.025 [m surface layer of

ZnO, with the ZnO nanorod extending to y = 0.26 [m and being 0.015 nm wide. This region

appears as dark blue to purple in Figure 2-52A, as ZnO is nearly transparent to most

wavelengths. The 10 nm absorbing region of P3HT appears as the brightly-colored region in

Figure 2-52A, and is the focus of Figure 2-52B. Although absorption is strong in the strips

above the nanorod tip and just beyond the ZnO base layer, a large portion of the carrier

generation in the unit cell occurs in the area on the side of the nanorod. It is interesting to note

from Figure 2-52B that the strong absorption from the region above the nanorod tip extends

slightly into the area to the side of the nanorod. It is unclear from simulations if this occurs due

to refraction at the interface or if it simply a numerical anomaly to facility convergence. The

P3HT bulk region is seen in blue in Figure 2-52A. Although the absorption coefficient is set to

zero in this region, a low level of carriers still exist.

Line Source Generation

Line sources of carriers were imposed along the ZnO-P3HT interface, consisting of three

lines due to the shape of the interface. The first line is at the tip of the ZnO nanorod. For

simplicity, this is noted as Line 1. The line extending along the edge of the nanorod is referred

to as Line 2, and the line running along the top of the ZnO base layer is Line 3.

To determine the density of photogenerated carriers to be imposed along these line sources,

all carriers generated by the simulated solar radiation are mapped to their nearest ZnO-P3HT

interface point. This mapping scheme is illustrated in Figure 2-53. Gray areas represent ZnO,

while blue areas represent P3HT. Carriers in a particular region are directed to the nearest line as









illustrated with the orange arrows. Dotted white lines show the edges of regions that map to a

specific line. As one of these lines is crossed, the carriers begin mapping to a different line.

There are two blocks, in the top-left and bottom-right corners, that map to a single point at the

intersection of two lines.

Once the carrier densities from each mesh point in the unit cell are mapped to their

corresponding point along one of the three lines, these carrier distributions are fit to exponential-

linear curves to determine the carrier vs. line distance profile along each line.

The mapping process is identical for all simulations using the same unit cell. In other

words, the exciton diffusion length does not impact the mapping process. Because carriers are

mapped to the nearest line, the (x, y) coordinates of the point are the only factor not how many

carriers are generated there. However, changes in the exciton diffusion length do impact the

number of carriers mapped to each point on the lines, and therefore impact the shape of the

carrier distribution along those lines.

To clarify the mapping process, Figure 2-54 displays the photogenerated carrier

distribution for a simulated cell with an exciton diffusion length of 20 nm. The dotted lines

shown in the illustration of Figure 2-53 are overlayed on this contour plot, seen as dashed white

lines. Line 1, Line 2, and Line 3 are shown as solid white lines and represent the borders

between the ZnO and P3HT regions. Note that the x- and y-axes are not on the same scale. The

diagonal dashed lines on the plot are at 450 angles through the unit cell, although scaling makes

them appear to be less sloped.

The behavior of the three carrier generation lines depends on the exciton diffusion length

of the simulated cell and the lateral distance along the line. For Line 1 (x < 0.015 [am, y = 0.26

am), the carrier distribution is nearly constant. As seen in Figure 2-54, this line collects carriers









from the P3HT region beyond the tip of the nanorod, as well as a small area within nanorod.

This area inside the nanorod decreases as the x-coordinate increases because portions of this

region begin mapping to Line 2. This decrease results in a negligible decrease in the number of

carriers, however, because the ZnO absorbs poorly in comparison to the active P3HT region

included in the calculation. This effect holds true for any exciton diffusion length that is

considered in this study, because the P3HT region will always absorb much more strongly than

the ZnO region. Due to this, the carrier distribution for Line 1 is approximated as a constant for

all simulations.

Along Line 2 (x = 0.015 am, 0.025 am < y < 0.26 am), photogeneration increases steadily

at low values of y because a larger section of the absorbing region is included at each point.

Seen in Figure 2-54, this occurs for approximately one LD, from 0.025 < y < 0.045 am. Beyond

that point, the new regions being added are not absorbing regions, so even though the size of the

area being summed is larger, this larger area is not contributing a significant number of carriers.

This is compounded by the fact that the LD region at this greater depth produces a lower number

of carriers due to a drop in the number of photons remaining in the cell. This exponential decay

of absorption becomes the dominant contribution to the number of carriers along the line, and

continues to the tip of the nanorod. The point where the distribution along Line 2 turns from

linear growth to exponential decay depends on the exciton diffusion length of the simulated cell.

For Line 3 (x > 0.015 a, y = 0.025 am), the number of photogenerated carriers initially

shows a linear increase with x for the same reasons as Line 2. As the line is traversed, additional

area is being mapped to this line, which increases the number of carriers contributed. After a

distance of LD along the line, the distribution becomes nearly constant. This occurs because the









new area being included in each summation contains relatively few carriers due to lack of

absorption in these regions.

The total amount of photogeneration in the simulated cell is dependent on the exciton

diffusion length specified in the simulation. This parameter defines the size of the strongly-

absorbing area in the P3HT polymer. Figure 2-55 displays the total count of photogenerated

carriers for simulated cells with varying exciton diffusion lengths defined from 10 nm up to the

full polymer region of the cell.

In addition to dictating the total number of carriers in the cell, variations in the exciton

diffusion length also change the distribution of these carriers. Figure 2-56 displays the line

sources used in Medici to simulate photoabsorption based on the model described previously.

Note that the curves seen in Figure 2-56 are not the actual summations of carriers calculated

from the contour plots, they are the fit lines applied to Medici with the form shown in Equation

2-6. As evidence of the quality of the fit, Figure 2-56C displays the true carrier summations for

Line 2 of a cell with LD = 20 nm, along with the fit curve applied in Medici.

As described previously, photogeneration in Line 1, for 0 < x < 0.015 atm, is set as

constant in all cases. In Line 2, shown in Figue 2-56B, carrier generation increases nearly

linearly until a certain point where the exponential decay becomes the dominant effect. For Line

3, carrier generation again increases nearly linearly for x > 0.015 [tm before reaching a point

where it becomes approximately constant. Unlike Line 1, the nearly-constant region for Line 3

was not assumed to be exactly constant, and all parameters for the exponential-linear equation

were calculated. For the 10 nm and 20 nm cases, this region shows a lower slope than in the 40

nm and full cell cases.









Interestingly, the curves in Figure 2-56 show regions where the carrier generation is

higher for the 40 nm scenario than for the full cell scenario. This effect is observed to a very

small degree near x = 0.035 [tm in Figure 2-56A and more noticeably near y = 0.06 [tm in Figure

2-56B. This is not simply an inconsistency due to curve fitting, this shift occurs in the

absorption profiles that were summed in the cells.

To clearly see the regions where the 40 nm and full cell simulations produced inconsistent

carrier generation, the generation profile for the 40 nm absorption case was subtracted from the

profile for the full cell case, and contour plots of this difference are shown in Figure 2-57. In the

figure, positive values represent areas where the full absorption simulation produced more

carriers than the 40 nm simulation. Figure 2-57A shows that the positive-valued regions have

extremely large values compared to the negative-valued regions, accounting for the increase in

total absorption in the cell. Figure 2-57B is shown with a smaller scale so that positive and

negative-valued areas are clearly displayed. The green areas show no equal carrier generation

for the two simulations, reds and yellows show areas where the full absorption case produced

more carriers, and blues and purples show areas where the 40 nm simulation produced more

carriers.

In general, the ZnO phase shows no difference between the two simulations, the P3HT

region beyond the 40 nm LD shows greater carrier generation in the full absorption simulation,

and portions of the diffusion length region show greater generation in the 40 nm case. In the

region beyond the tip of the nanorod, the plot shows a zero value for 4 nm, then a negative area

with a value of approximately -2 x 1020 pairs/cm3 stretches for 16 nm before jumping to a

positive value of nearly 6 x 1021 pairs/cm3. To the side of the nanorod, the contour plot shows a

zero value for nearly 10 nm into the P3HT, but then turns negative for 26 nm. In the P3HT bulk









region, there is no difference between the simulations up to x = 0.022 tm. Beyond the interface

for approximately 10 nm, values are all negative, indicating the 40 nm simulation generated

more carriers. Above the interface in the ZnO phase, there is a thin strip of 2 4 nm where there

is an inconsistency. From x = 0.05 to 0.06 [tm, this strip shows a negative value. For x = 0.06 to

0.065 am, the strip shows a positive value.

The full cell absorption simulation showed stronger carrier generation in the P3HT region

beyond 40 nm from the ZnO interface. This is to be expected, as the full cell simulation allows

this region to absorb any remaining carriers, while the 40 nm simulation does not. However, the

increased carrier generation for the 40 nm simulation within the LD region is puzzling. Photons

entering this region have the exact same absorption history in both simulations. They have

passed through a weakly-absorbing ZnO layer and entered a strongly-absorbing P3HT region.

For both simulations, this P3HT region has identical absorption properties for a path length of at

least 40 nm. In the area within 40 nm of the side of the nanorod, the two simulations have

identical absorption properties for 300 nm into the P3HT region. There is never a point in the

simulation area where the 40 nm simulation displays a stronger absorption coefficient than the

full cell simulation. With that in mind, this inconsistency is considered to be a numerical

anomaly in Medici during the convergence process. It is a strange occurance, but due to the

difference in value between the negative regions and positive regions, as seen in Figure 2-57A,

the total number of carriers in the cells is still increasing as the absorption area increases.

There are two data points ignored in the graphs in Figure 2-56 the points at the base and

tip corners of the nanorod. The carrier concentration at these points is plotted against the exciton

diffusion length in Figure 2-58. The point designated as the nanorod tip has coordinates of

(0.015, 0.260) and shows a growth as the exciton diffusion length increases. This point collects









all generated carriers in the P3HT region for x > 0.015 [tm and y > 0.260 am. As the exciton

diffusion length increases, this region produces a larger number of carriers because a larger

portion of the region contains a strong absorption coefficient. The point designated as the

nanorod base is the point at (0.015, 0.025) where the edge of the nanorod meets the base ZnO

layer. This point collects all photogenerated carriers for the region of x < 0.015 am and y <

0.025 [tm, which is composed entirely of ZnO. Because the absorption coefficient of ZnO is

held constant in all simulations, the number of carriers collected at this point remains constant

regardless of the exciton diffusion length.

With this procedure established for measuring carrier generation across the simulated unit

cell and compressing this distribution to line sources in Medici, the J-V response of cells can be

determined.

J-V Curves

The carrier generation line sources shown in Figure 2-56 and 2-58 were input to Medici to

generate J-V curves for the simulated cells. As described previously, these line sources of

carriers restrict the existence of free electrons and holes to the interface between ZnO and P3HT,

which matches the physics of excitonic solar cells. The resulting J-V curves were compared with

the published data and are shown in Figure 2-59. As anticipated, the short-circuit current density

increases with increases in the exciton diffusion length. Interestingly, this trend does not

continue to the fully absorbing unit cell, where the short-circuit current density drops by 0.2

mA/cm2 from the level of the LD = 40 nm cell.

All simulations failed to achieve the short-circuit current density of 2.2 mA/cm2 reported

for the published cell. This is interesting considering that the simulated cell was allowed to

generate photocurrent for specified diffusion lengths ranging to the full cell area. The open-

circuit voltage of the real cell was well approximated by the experiments due to the 0.85 eV









energy band gap specified in P3HT. Fill factors for the simulated cells were higher than for the

real cell. Table 2-17 displays a summary of the performance characteristics of the real and

simulated cells shown in Figure 2-59.

The simulated cell using a 40 nm exciton diffusion length showed the best match to the

real cell from literature. The Voc was nearly a perfect match, and the efficiency was

approximately 5% higher than the literature value. However, the fill factor and Jsc for the two

cells did not match. The simulated cell showed a fill factor that was 50% higher and a Jsc that

was 33% lower than the real cell.

Further simulations were performed using the 40 nm LD simulation as a basis. The effect

of changing the carrier mobilities by up to two orders of magnitude is shown in Figure 2-60, and

the resulting performance measures are shown in Table 2-18. While four of the curves appeared

as nearly identical, the curve corresponding to a 100x reduction in mobility began to show a

noticeable drop in fill factor. Attempts to reduce the mobility by an extra order of magnitude

failed due to an inability of Medici to converge around such small values. Because Medici is

designed for the simulation of inorganic semiconductors, mobility values of p = 1 x 10-5 and [n

= 1 x 10-6 Cm2/V-s are more than 6 orders of magnitude lower than the program was designed to

handle. This leads to convergence difficulties due to the low current values in the cell.

The simulation using a 100x reduction in mobility values shows a better match to the real

cell than any other simulation. The Voc and efficiency differ by only about 2% between the two

cells. The Jsc is again off by 33% from the real cell, as that value did not change with the

variation in mobility. With the lower mobility, however, the fill factor dropped to within 40% of

the published value.









Because these simulations fail to produce a short-circuit current on par with the real cell,

the P3HT absorption coefficient was adjusted to a higher value. This generates more carriers in

the cell, which is expected to generate more photocurrent at all applied biases. The absorption

coefficient is a parameter that is easily measured experimentally, but there is some physical of

justification for this adjustment.

The unit cell in Medici is defined by perfectly vertical nanowires with flat tips and perfect

spacing between them. From the image in Figure 2-14, this is far from reality. Although care

was taken to properly estimate the nanowire length, width, and inter-wire spacing, these values

are purely averages taken from the visible portion of the cross-sectional view. The real wire

array contains overlapping non-vertical wires that create non-uniform spaces between them

where P3HT exists. In addition, these randomly angled structures would create light reflection

and refraction patterns that are not considered in Medici. It is not unreasonable to expect that

this disordered array of nanowires could create light-trapping effects in the film, where incident

rays are refracted in such a way to increase their residence time in the film and increase the

degree to which they are absorbed. From this, it is not unreasonable to expect that this increased

absorption occurs to some degree in the polymer regions within an exciton diffusion length from

the material interfaces.

For these reasons, the absorption coefficient of P3HT was increased by 50% in a

simulation. The photogenerated carrier distribution for this simulation is compared to the

previous 40 nm LD simulation, with an absorption coefficient of 100%, with the resulting

surface plots shown in Figure 2-61. As expected, this resulted in a larger number of

photogenerated carriers in the film.









Although the total number of photogenerated carriers increased, there was not a uniform

increase at all points in the film. Specifically, in the column of polymer within one exciton

diffusion length of the side of the nanorod, the number of carriers decreases quickly for the

simulation with an increased absorption coefficient. This is because the number of photons

penetrating deeper into the film is reduced due to the stronger absorption. This is clearly seen in

Figure 2-62, which shows a comparison of carrier generation for the increased absorption case

and the standard absorption case. The carrier distributions were subtracted, and green regions

show no change in absorption between the two simulations. Red, orange, and yellow regions

show areas where the 150% absorption coefficient resulted in stronger absorption, and this is

primarily contained within a range of 40 nm (LD) from the upper surface of the ZnO-P3HT

interface. Blue regions represent areas where the 150% absorption coefficient resulted in less

absorption, and this is limited to the exciton diffusion length region along the nanorod edge.

This is due to strong absorption near the surface of the P3HT region, leaving fewer photons

available for absorption in the deeper region.

Although there are regions of increased and decreased carrier generation in the cell, the

overall number of generated carriers increased by approximately 3 x 1024 pairs/cm3 over the unit

cell. This additional charge generation did not translate into additional photocurrent as expected,

however. Figure 2-63 shows J-V curves for the two cells, calculated with [p = 1 x 10-4 cm2/V-s.

Despite the increased carriers density for the 150% absorption coefficient simulation, the

short-circuit current density decreased by 0.23 mA/cm2. This is very similar to the effect seen

when the full P3HT area was allowed to contribute to absorption. In fact, a comparison of those

cells shows extremely similar properties. Note that the full cell simulation was performed with

the standard values for mobility in the P3HT regions, while the two 40 nm simulations were









performed with a these mobility values reduced by two orders of magnitude. This explains the

difference in fill factor shown in Table 2-19, but Figure 2-60 demonstrated that this mobility

reduction does not impact the short-circuit current density in these simulations.

It is believed that the reason for the reduced performance associated with this increase in

carrier generation is an effect of increased annihilation near the material interface. This large

number of carriers is being produced along a very narrow region directly at the material

interface. This results in many free electrons and holes in a small area, which could cause

attractive forces between them to promote annihilation immediately following their generation.

There seems to be a breaking point between 3.14 x 1025 and 3.23 x 1025 pairs/cm3 of total

carrier density, as this is the limit where the reduced charge transport seems to take hold. The

short-circuit current density drops by about 15%, the Voc remains constant, and the fill factor

drops by less than 10%.

Summary of Results

The device modeling program Medici was used for simulation of hybrid solar cells.

Several nuances of the program were probed with a test model before attempts were made to

simulate an existing cell from the literature.

After probing the effect of various parameters, a two-step simulation strategy was adopted

that separated photon absorption and charge transport. This model more accurately

approximates exciton dynamics by eliminating free electrons in the polymer regions of the cell

and applying free carrier generation directly at the material interface lines.

Simulations of this type initially showed low performance, but increases in the exciton

diffusion length up to 40 nm provided increased charge generation and current flow. Further

increases in the number of carriers generated reduced the current in the simulation, presumably

due to increased charge attraction and annihilation. It was found that decreasing the carrier









mobilities in P3HT from the field effect mobility values found in literature resulted in a decrease

in the fill factor to levels similar to the real cell.













ITO I


BjZr~


Figure 2-1. Hybrid solar cell (A) and corresponding unit cell (B) used for device simulation.


Table 2-1. P3HT Properties for Device Simulations.
Property Value Reference
Doping Density p-type 5x1016 cm-2 [67]
Permittivity 3 [61]
Nc 2x10 [69]
Nv 2x1019 [69]
Eg 1.7 eV [70]
Electron Affinity 3.15 eV [71]
[ie 0.01 cm2/V-s 10% of [h
[th 0.1 cm2/Vs [68]


Table 2-2. CdSe Properties for Device Simulations.
Property Value
Doping Density n-type 6x1016 cm-2
Permittivity 10.2
Nc 2x1018
Nv 2x1019
Eg 1.74 eV
Electron Affinity 3.75 eV
[e ~650 cm2/V-s
[Ph 30 cm2/V-s


Table 2-3. Electrode Properties for Device Simulations.
Electrode Optical Properties Reference
ITO Transparent [72]
Al 90% Reflectance [73]














2.5e+5


2.0e+5 -




1.5e+5 -




1.0e+5 -




5.0e+4 -


A A A A
0.0 A
0.2


A A A A A 4AAA


A0
0.4


-- P3HT
A ZnO
-4- CdSe


















A A A A A A
0.8 1.0


Wavelength (pm)


Figure 2-2. Wavelength-dependent absorption coefficient data used in simulations [66, 74].


-2.0 -- iipui rity_l"
"E 'i it_2 / /
4.0 i ity3
E

S-6.0
C
o -j
8.0


-10.0--


-12.0
0.00 0.05 0.10 0.15 0.20 0.25 0.30

Voltage(V)



Figure 2-3. J-V curves for simulated hybrid solar cells with different methods of specifying
doping density. See Table 2-4 for a description of the differences.


















-2 5E16
E 05 6E16
,o A 7E16

E -4



-6


0
-8



-10
0.00 0.05 0.10 0.15 0.20 0.25 0.30

Voltage (V)


Figure 2-4. Simulated J-V curves showing the effect of doping density in the CdSe nanorods.
The legend shows the n-type doping density measured in cm-2









Table 2-4. Impurity profile inputs for simulations.
Filename Impurity Profile Input Statement
impurityl PROFILE P-TYPE REGION=P3HT UNIFORM N.PEAK=5E16
PROFILE N-TYPE REGION=CdSe UNIFORM N.PEAK=6E16
impurity_2 PROFILE P-TYPE Y.MIN=0 Y.MAX=0.100 UNIFORM N.PEAK=5E16
PROFILE N-TYPE Y.MIN=0.01 Y.MAX=0.100 X.MIN=0 X.MAX=0.005
+ UNIFORM N.PEAK=6E16
impurity_3 PROFILE P-TYPE Y.MIN=0 Y.MAX=0.01 UNIFORM N.PEAK=5E16
PROFILE P-TYPE Y.MIN=0.01 Y.MAX=0.100 X.MIN=0.005 X.MAX=0.015
+ UNIFORM N.PEAK=5E16
PROFILE N-TYPE Y.MIN=0.01 Y.MAX=0.100 X.MIN=0 X.MAX=0.005
+ UNIFORM N.PEAK=6E16
impurity_4 PROFILE P-TYPE Y.MIN=0 Y.MAX=0.01 UNIFORM N.PEAK=5E16
PROFILE P-TYPE Y.MIN=0.01 Y.MAX=0.100 X.MIN=0.005 X.MAX=0.015
+ UNIFORM N.PEAK=5E16
PROFILE N-TYPE Y.MIN=0.01 Y.MAX=0.100 X.MIN=0 X.MAX=0.005
+ UNIFORM N.PEAK=1E16
impurity_5 PROFILE P-TYPE Y.MIN=0 Y.MAX=0.01 UNIFORM N.PEAK=5E16
PROFILE P-TYPE Y.MIN=0.01 Y.MAX=0.100 X.MIN=0.005 X.MAX=0.015
+ UNIFORM N.PEAK=5E16
PROFILE P-TYPE Y.MIN=0.01 Y.MAX=0.100 X.MIN=0 X.MAX=0.005
+ UNIFORM N.PEAK=5E16
PROFILE N-TYPE Y.MIN=0.01 Y.MAX=0.100 X.MIN=0 X.MAX=0.005
+ UNIFORM N.PEAK=6E16



0.3

S0.25-

0.2

S0.15


S00Vn = 0.012N, + 0.181
c- 0.05
0I I

0 2 4 6 8

CdSe Doping Density (x1016)


Figure 2-5. Variation of open circuit voltage with doping density of CdSe nanorods.































-15


-20 -
-0.10


-0.05 0.00 0.05


Voltage (V)


Figure 2-6.


Simulated J-V curves with varying unit cell thickness.


0.4 16

0.35 ,- 14
LL
> 0.3 12

0 0.25 -- Efficiency () -- 10 E
> U
O Voc(V)
0.2 8
FE
S0.15 _sc(mA/cm2) 6
'0 0.1 4

0.05 -. 0 rl 0 l 2

0 0
0 200 400 600 800 1000

Cell Thickness (nm)


Figure 2-7. Solar cell parameters for unit 12 nm wide unit cells with varying cell thickness.










Table 2-5. Solar cell performance measures for unit cells of 12 nm width and varying thickness.


Thickness (nm)
11
20
30
40
50
60
70
80
90
100
120
150
200
300
500
1000


Figure 2-8. J-V


q (%)
0.0000
0.0000
0.0052
0.0172
0.0335
0.0507
0.0669
0.0817
0.0954
0.1082
0.1310
0.1598
0.1966
0.2456
0.2975
0.3275


Voc (V)
0.0000
0.0009
0.0067
0.0163
0.0252
0.0317
0.0361
0.0393
0.0418
0.0437
0.0466
0.0497
0.0531
0.0568
0.0597
0.0588


Jsc (mA/cm2)
1.251
2.257
3.106
3.899
4.627
5.297
5.916
6.491
7.023
7.513
8.368
9.394
10.625
12.150
13.791
15.594


FF
0.0003
0.0000
0.2533
0.2699
0.2873
0.3019
0.3131
0.3203
0.3250
0.3295
0.3360
0.3422
0.3484
0.3559
0.3614
0.3572


0.0 0.2 0.4 0.6
Voltage (V)
curves for hybrid cells with varying nanorod width.































-8 z4 nm


-10
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Voltage (V)
Figure 2-9. J-V curves for simulated hybrid cells with CdSe nanorod half-thickness between 10
and 25 nm.


Figure 2-10. Solar cell
width.


performance measures for simulated hybrid cells with varying CdSe half-


1.0 10
0.9 9
0.8 8
0.7 7
E






0.2 OFF 2
0. Efficiency Q- ,..
0.1 -- 1
012 A' Jsc (mA/cm2)
0.0 0
0 5 10 15 20 25 30 35

CdSe Nanorod Half-Width (nm)










Table 2-6. Performance measures for simulated hybrid cells with varying CdSe half-width.
CdSe Width (nm) Voc (V) Jsc (mA/cm2) FF q (%)
2 0.044 7.513 0.330 0.11
5 0.253 9.658 0.623 1.52
10 0.410 8.575 0.661 2.32
11 0.435 9.488 0.670 2.77
12 0.454 9.453 0.669 2.87
13 0.470 9.421 0.668 2.96
14 0.485 9.391 0.666 3.03
15 0.487 6.461 0.675 2.12
16 0.512 9.338 0.662 3.16
17 0.523 9.315 0.659 3.21
18 0.533 9.295 0.658 3.26
19 0.542 9.276 0.656 3.30
20 0.535 5.745 0.677 2.08
21 0.558 9.243 0.654 3.37
22 0.564 9.229 0.653 3.40
23 0.553 5.414 0.679 2.03
24 0.576 9.204 0.652 3.45
25 0.580 9.194 0.652 3.48
30 9.071 4.93


50

0 Center, Y=-2.5
40 v Center, Y=-5.0
Left Edge=-
Right Edcge ,:-

30


20


10 -


0


-10 '. --
0.00 0.05 0.10 0.15


','
LAF


0.20 0.25 0.30 0.35
0.20 0.25 0.30 0.35


Voltage (V)
Figure 2-11. J-V curves for hybrid solar cells with different light source specifications.











RAY.W Origin


I E |
! I Is
Si i ir i Is
i r i.


Cell Width


Figure 2-12. Illustration of the PHOTOGEN command in Medici.


200Uam I(NONENI 200rnm 10 0X

Figure 2-13. SEM images of(a) ZnO nanofibers and (b) nanofiber and P3HT composite films.
Reprinted with permission from D.C. Olson, J. Piris, R.T. Collins, S.E. Shaheen, D.S.
Ginley, Thin Solid Films 496 (2006) 26 Figure 2 (a), (b) p. 28.












E


ro I I I I





a--1
I I I I I I I I


-0.2 -0.1 0.0 0.1 0.2
Voltage (V)


0.3 0.4 0.5


Figure 2-14. J-V curve for a real ZnO:P3HT solar cell (solid line) to be used for verification of
Medici simulations. Reprinted with permission from D.C. Olson, J. Piris, R.T.
Collins, S.E. Shaheen, D.S. Ginley, Thin Solid Films 496 (2006) 26 Figure 3 p. 28.


0.6

Current Density
Power Density
-0.3



m"m


-0.3







-0.9
-0.1 0.0 0.1 0.2 0.3 0.4 0.5


Voltage (V)


Figure 2-15. J-V and P-V curves for the real solar cell fabricated by Olson et al. [75]


Table 2-7. Cell performance measures from published and digitally
Curve Voc (V) Jsc (mA/cm2) FF


Published
Converted


0.44
0.44


0.56
0.57


converted J-V curves.
1q (%)
0.53
0.55


-3 -
-0.2











Table 2-8. ZnO properties used for hybrid solar cell simulation.
Property Value
Doping Density n-type 5 x 1017 cm-2
Permittivity 9.0
No 2 x 1018
Nv 2 x 1019
Eg 3.3 eV
Affinity 4.0 eV
_pe 50 cm2N-s
Pth 5 cm N-s


15 nm
- 50 nm
p------p*


260nm


430nm


25nm
ILD






I LD
170 nm


--* 65 nm
LD


Figure 2-16. Unit cell used for simulations of ZnO/P3HT hybrid cells. Device areas area a) ITO
electrode, b) ZnO, c) P3HT photocurrent generating region, d) P3HT non-generating
region, and e) Ag electrode.












Sl-J II 'I -A I i I- '. I .. Ii e HM .' II i F II'
lr''*: i ui F
















,--,



,--








L I. .00 00 -I. L1L1 ,. .. 0
[i0 1 i hr.l: -* i I .:r.:.r. ; IL 1 -


Figure 2-17. Medici unit cell used for device simulation. The blue area is ZnO, the red area is
the photogenerating region of P3HT, and the green area is the non-generating region
of P3HT.

















0 -



E



o -6 Keal Cell
"E W I .f'7,.", 1
L -8- t .
0

-10 -


-12
-0.2 0.0 0.2 0.4 0.6 0.8

Voltage (V)


Figure 2-18. Simulated J-V curves for ZnO:P3HT solar cell using two P3HT regions. The first
region is the exciton diffusion length region, where all properties are set as shown in
Table 2-1. The second region is the non-generating region outside of the diffusion
length, with varying properties. In tsf496_1, the absorption coefficient is set to zero.
In tsf496_2, the properties remain the same as the exciton diffusion length region. In
tsf496_4 the electron mobility is reduced by an order of magnitude (0.0001 cm2/Vs).
















-- 90% Reflectance
-1.550% Reflectance
0% Reflectance
o" -2.0
E

< -2.5


cn -3.0
C
(.
C)
S-3.5 /


0 -4.0 .





-5.0
-0.2 0.0 0.2 0.4 0.6 0.8 1.0

Voltage (V)


Figure 2-19. J-V curves for simulated cells with zero absorption in the P3HT2 region and
varying reflectance from the Ag electrode.

















-1.0


-1.5


-2.0


-2.5


-3.0


-3.5 -


-4.0
-0.2


I I I I I
0.0 0.2 0.4 0.6 0.8 1


Voltage (V)


Figure 2-20. Simulation results showing the effect of changing charge mobilities in the P3HT
regions.

Table 2-9. Performance measures for simulated cells with varying carrier mobility.
Mobility (cm2/V-s) Jsc (mA/cm2) Voc (V) FF q (%)
Cln = 0.01
=0.01 3.61 1.04 0.57 2.12
pp = 0.001
pn = 0.005
S0.005 3.58 1.04 0.54 2.00
pp = 0.0005
Cln = 0.01
p.=. 3.58 1.04 0.57 2.13
pp = 0.0005
pn = 0.001
S0.001 3.44 1.04 0.46 1.64
pp = 0.0001


-- tsf496_05
n = 0.001 cm2/V-s, Lp = 0.01 cm2N-s
- tsf496 07
i's reduced by 50% in both P3HT regions
tsf496 08
i's reduced by 50% in P3HT2 region
tsf496 09
i's reduced by 90% in both P3HT regions









































0.0 0.2 0.4 0.6 0.8


Voltage (V)


Figure 2-21.


J-V curves for simulated


cells with varying exciton diffusion length.


Figure 2-22. Extrapolations to estimate Voc for simulated cells with varying exciton diffusion
length.


-3



-4 -
-0.2


-0.2 C

-0.4

-0.6

-0.8

-1

-1.2

-1.4


0.9 0.95


Voltage (V)


v LD= 10 rni
E LD= 9 nm
LD= 8 nm
X LD= 7 nm
SLD= 6 nm
O LD= 5 nm
- Linear(LD = 10 nm)
-- Linear(LD= 9 nm)
SLinear (LD = 8 nm)
------------ Llnear(LD= 8nm)
- Linear (LD = 7 nm)
SLinear (LD = 6 nm)
-- Linear (LD = 5 nm)


).


1 1.05










1.4 3.5

1.2 3

1 2.5

S0.8 -2 < V
E DVoc
0.6 1.5 XFF
> .ft
0.4 1 Eff ( .
..... Jsc
0.2 0.5 s

0 0
4 5 6 7 8 9 10 11

Exciton Diffusion Length (nm)


Figure 2-23. Solar cell performance measures for simulated cells with varying exciton diffusion
lengths.

Table 2-10. Performance measures for simulated cells with varying exciton diffusion lengths.
LD (nm) Voc (V) Jsc (ma/cm2) FF q (%)
10 1.04 3.44 0.459 1.64
9 1.04 3.17 0.468 1.54
8 1.05 2.90 0.473 1.44
7 1.05 2.61 0.482 1.32
6 1.06 2.32 0.487 1.20
5 1.07 2.02 0.492 1.06





























-3-
-0.2


0.0 0.2 0.4 0.6 0.8


Voltage (V)

Figure 2-24. Real and simulated J-V curves for cells with varying P3HT doping density.



nn -


-0.5


-1.0


-1.5


-2.0


0.0 0.2 0.4


0.6 0.8


Voltage (V)

Figure 2-25. Real and simulated J-V curves for hybrid cells with varying ZnO doping density.


Real Cell
- P3HT Doping 5x1016 cm-3
-- P3HT Doping 1x1016 cm-3
P3HT Doping 5x1015 cm-3
P3HT Doping 1x1015 cm-3
- P3HT Doping 5x1014 cm-3
- P3HT Doping 1x1014 cm-3


S
0
S






/


Real Cell 0
-- ZnO Doping Density 5x1017
-- nO Doping Deniit, 1.1017 S
n-.-- Dogpin g EE-nijr-, '5 U-
nnO D'oping Densit, '5- 10"
-nO 'Doping Dlenslt, 2. 10"
n-- 'onC ping Clenslt, 1.10"
n- nO hoping DCenirl, 10 T ,/

-n0' 'Doping D'ensit, 5. 10'
n-nC' Doping ',ensi 1 10 ///
-- -'nO CLoPing CLenilt, 1.10"' /,i/,


----~--T ^//
_--_^ ,: I


-2.5 -
-0.2


00
























S











-0.2 0.0 0.2 0.4


0.6 0.8


Voltage (V)


Figure 2-26. Real and simulated J-V curves for hybrid solar cells with varying P3HT density of
states.


1.0


0.5


0.0


-0.5


-1.0


-1.5


-0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

Voltage (V)
Figure 2-27. Simulated J-V curves for hybrid solar cells with varying P3HT mobility.


* Real Cell
0 P3HT Nc=2x1018 Nv=2x1019
V P3HT Nc=2x1019 Nv=2x1019
P3HT Nc=2x1019 Nv=2x1020
P3HT Nc=2x1020 Nv=2x1020
A P3HT Nc=2x1020 Nv=2x1021


i
0


-- =0.001 cm2/V-s, p=0.01 cm2/V-s
-- =0.0001 cm2/V-s, p=0.001 cm2/V-s
=0.00001 cm2/V-s, P=0.0001 cm2/V-s


















0.0 -


-0.5 -


-1.0


-1.5


-2.0 -

-2.5 ,,,,,,
-0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
Voltage (V)

Figure 2-28. Simulated J-V curves for hybrid cells with varying P3HT doping concentrations.

iaculum


Figure 2-29. Energy band diagram for P3HT ZnO hybrid solar cells.


- P3HT Doping: 5x1016
- P3HT Doping: 1x1016
P3HT Doping: 5x1015
P3HT Doping: 1x1015
- P3HT Doping: 5x1014
- P3HT Doping: 1x1014


-- P3HT Doping: 5x101
-- P3HT Doping: lx101
P3HT Doping: 5x10
P3HT Doping: lx10
-- P3HT Doping: 5x10 4
-- P3HT Doping: lx10 4







































-4 4-


-0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6

Voltage (V)


Figure 2-30. J-V curves for simulated cells with varying energy band gap in the active layers.


-e- absnone
abs_p3ht
abs i zno


0.0 0.2 0.4 0.6


0.8 1.0 1.2 1.4


Wavelength (lim)
Figure 2-31. Absorption coefficient vs. wavelength as tabulated in Medici for input files
corresponding to zero absorption (red), P3HT (green), and ZnO (teal).


* Real ,Cell
A t..if4,ii,_ 4
I tl.f4 ':, _, '=, "


U,,



o~-


2.5e+6




, 2.0e+6
E
CE
*A 1.5e+6
C-

c,


c 1.0e+6




S5.0e+5


I I I I I I I









































0.2 0.4 0.6 0.8 1.0 1.2

Wavelength (u)m)


1.4 1.6 1.8


Figure 2-32. AM1.5 solar spectrum.





3.5e-13


3.0e-13 -

E "
2.5e-13 -


(D
2.0e-13 -
Co
C-
S1.5e-13

a(
0

o
5.0e-14 -


0.0


2.25e+6

2.00e+6

1.75e+6
E
1.50e+6

'1.25e+6
0
0
0

o
2.00e+5

J3


0.0 0.2 0.4 0.6 0.8 1.0

Wavelength (jlm)

Figure 2-33. Carrier generation (left axis) in simulated solar cells plotted with absorption
coefficients for P3HT and ZnO (right axis).





































0.0 0.1


0.2 0.3 0.4


Voltage (V)
Figure 2-34. J-V curves for simulated cells showing the original P3HT absorption profile and an
edited absorption profile limiting absorption between 0.2 and 0.3 rnm.


-3 I
-0.1 0.0 0.1


0.2 0.3 0.4 0.5


Voltage (V)
Figure 2-35. J-V curves for simulated cells with exciton diffusion length
stage absorption regions.


of 10 nm and multi-


- u- lMedici-assigned a values ....
Fully specified a values


-4-
-0.1


- tsf496_45 1 stage
- tsf496_47 2 stages
tsf496_48 4 stages
tsf496_49 6 stages
- tsf496_50 8 stages
- tsf496_51 -10 stages











1.0

o_
I 0.8
O

_
0 0.6

E
-3 a = 0.001
E 0.4

O
0
c 0.2 0 =8
.2 --~=10
Cu
L 0.0.
00 5 10 15 20
P3HT Thickness


Figure 2-36. Examples of cumulative distribution function with mean of 10 nm and a range of
standard deviation.

Table 2-11. Absorption data and short-circuit current for graded absorption simulations.
C5 0 nm 1 nm 2 nm 3 nm 4 nm 5nm 6nm 1 nm
0-2 nm 100% 100% 100% 100% 100% 100% 100% 100%
a, 2-4nm 100% 100% 100% 100% 100% 90% 90% 90%
c, 4100% 100% 10010000% 9 909% 0% 80% 90%
a, 6-8 nm 100% 100% 90% 80% 80% 70% 70% 70%

ca, 8-10 nm 100% 80% 70% 60% 60% 60% 60% 60%
c, 10-12 nm 0% 20% 30% 40% 40% 40% 40% 40%
12-14nm 0% 0% 10%0 20% 20% 30% 30% 30%
14-16nm 0% 0% 0% 10% 10% 10% 20% 20
,16-18 nm 0% 0% 0% 0% 0% 10% 10% 10%
c, 18-20 nm 0% 0% 0% 0% 0% 0% 0% 0%
Avera e a 50% 50% 50%0 51% 50% 50% 50% 51%
Jsc (mA/LcM2 1.62 1.62 1.66 1.67 1.68 1.69 1.69 1.71

































-2


-3
-0.1


0.0 0.1 0.2 0.3 0.4


Voltage (V)


Figure 2-37. Simulated J-V curves for cells with graded absorption profiles.


0


-0.1 0.0 0.1 0.2 0.3 0.4

Voltage (V)


Figure 2-38. Simulated J-V curves with CIS replacing P3HT.


- ZnO:P3HT with 10 nm absorption region
- ZnO:CIS with 10 nm absorption region
ZnO:P3HT with full absorption region
ZnO:CIS with full absorption region
- ZnO:CIS with 10 nm absorption region and Eg = 0.85 eV









Table 2-12. Materials properties for P3HT and CIS used in cell simulations.
Property P3HT CIS
Doping Density p-type 5x1016 cm-2 p-type 5x1016 cm-2
Permittivity 3 13.6
Nc 2x1018 3x1018
Nv 2x1019 1.5x1019
Eg 1.7 eV 1.04 eV
Electron Affinity 3.15 eV 3.93 eV
[te 0.1 cm2/V-s 300 cm2/V-s
[lh 0.01 cm2/V-s 30 cm2/V-s


-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6
Voltage (V)

Figure 2-39. Simulated J-V curves for ZnO:CIS solar cells with an individual material property
changed to the P3HT value.

Table 2-13. Performance measures for ZnO:CIS cells with an individual material property set at
the P3HT value.
Simulation Parameter Adjusted Voc (V) Jsc (mA/cm2) FF q (%)
64 None 0.439 7.774 0.772 2.633
68 Permittivity 0.443 7.747 0.778 2.669
69 Density of States 0.441 7.773 0.774 2.655
70 Affinity 0.487 7.654 0.485 1.808
71 Mobility 0.545 6.309 0.662 2.277
72 Absorption 0.421 4.105 0.762 1.317


































0.0 0.1 0.2 0.3 0.4 0.5 0.6

Voltage (V)


Figure 2-40. Simulated J-V
applied.


curves for ZnO:CIS solar cells with the P3HT absorption spectrum


Table 2-14. Performance measures for simulated ZnO:CIS solar cells with the P3HT absorption
spectrum.
Simulation Properties Varied Voc (V) Jsc (mA/cm2) FF (%)
72 Absorption 0.421 4.105 0.762 1.317

73 Absorption 0.427 4.065 0.765 1.328
Permittivity

74 Absorption 0.423 4.105 0.765 1.328
Density of States

75 Absorption 0.467 3.980 0.493 0.916
Affinity

76 Absorption 0.517 3.293 0.657 1.118
Mobility


- tsf496_72 P3HT Absorption
- tsf496_73 P3HT Absorption, Permittivit
tsf496_74 P3HT Absorption, Density of
tsf496_75 P3HT Absorption, Affinity
- tsf496_76 P3HT Absorption, Mobilities


-5


-10
-0.1


y
States














tsf496_75 P HT Absorption, Affinity
15 tsf496_77 P3HT Absorption, Affinity, Permittivity
tsf496_78 P3HT Absorption, Affinity, Density of States
tsf496_79 P3HT Absorption, Affinity, Mobilities
10 -


5 -


0 -


-5


-10
-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Voltage (V)


Figure 2-41. Simulated ZnO:CIS
electron affinity.


solar cells with P3HT values for absorption coefficient and


Table 2-15. Performance measures for simulated ZnO:CIS solar cells with P3HT values for
absorption coefficient and electron affinity.
Simulation Properties Varied Voc (V) Jsc (mA/cm2) FF i (%)
Absorption
75 Absorption 0.467 3.980 0.493 0.916
Affinity
Absorption
77 Affinity 0.455 3.979 0.502 0.909
Permittivity
Absorption
78 Affinity 0.471 3.975 0.479 0.896
Density of States
Absorption
79 Affinity 0.613 3.439 0.199 0.419
Mobility
































1 0.0 0.1 0.2 0.3 0.4 0.5 0.6
1 0.0 0.1 0.2 0.3 0.4 0.5 0.6


Voltage (V)

Figure 2-42. Simulated ZnO:CIS solar cells with P3HT values for absorption coefficient,
electron affinity, and energy band gap.

Table 2-16. Performance measures for simulated ZnO:CIS solar cells with P3HT values for
absorption coefficient, electron affinity, and energy band gap.
Simulation Properties Varied Voc (V) Jsc (mA/cm2) FF (%)
Absorption
80 Affinity, E, 0.265 3.633 0.418 0.402
Permittivity
Absorption
81 Affinity, Eg 0.281 3.608 0.395 0.401
Density of States
Absorption
82 Affinity, E, 0.423 2.164 0.159 0.145
Mobility


-- tsf496_80 P3HT Absorption, Affinity, Eg, Permittivity
-- tsf496_81 P3HT Absorption, Affinity, Eg, Density of States
tsf496_82 P3HT Absorption, Affinity, Eg, Mobilities






























-10 -



-0.1


0.0 0.1 0.2 0.3 0.4


Voltage (V)


Figure 2-43. Simulated J-V curves for ZnO:P3HT solar cells with varying carrier mobility. In all
cases, carrier generation is allowed in the full P3HT region and the electron mobility
is set to 10% of the hole mobility.


Figure 2-44. Calculation method for estimated second derivatives of J-V curves.



















'e-4 -
2e-4 -4 p = 10 cm2/V-s
F Lp = 100 cm2N-s

0 D 1ip = 200 cm2N-s


Lpp = 1000 cm2N-s
-2e-4 0 0L2 p = 5000 cm2N-s

a p = 10,000 cm2/V-s

-4e-4 -


0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Voltage (V)
Figure 2-45. Estimated second derivative J"est for simulated ZnO:P3HT solar cells with varying
carrier mobilities and carrier generation in the full P3HT region. In all cases, the
electron mobility is set to 10% of the hole mobility.
















8 = 1 cm2/V-s
= 10 cm2N-s
E 6- lp = 100 cm2N-s

S --- p = 200 cm2N-s
E 4 p = 500 cm2N-s
Sip = 800 cm2N-s
S- p = 1000 cm2N-s
S -- = 5000 cm2N-s
S0- L = 10,000 cm2N-s


-2 -

-4 -

-0.1 0.0 0.1 0.2 0.3 0.4 0.5

Voltage (V)


Figure 2-46. Simulated J-V curves for ZnO:P3HT solar cells with varying carrier mobilities and
carrier generation in the 10-nm exciton diffusion length region. In all cases, the
electron mobility is set to 10% of the hole mobility.
















4e-4 -
4e- u a O p =0.1 cm2/V-s
i- O = 1 cm2N-s
_l 0 up = 10 cm2N-s

Sp = 100 cm2N-s
2e-4 D P = 200 cm2N-s

S' p = 500 cmN-s
le-4 -p = 800 cm2N-s
I = 1000 cm2N-s
0 A Ip = 5000 cm2-s
A lIp = 10,000 cm2N-s
-1 e-4 --


-2e-4
0.0 0.2 0.4 0.6

Voltage (V)
Figure 2-47. Estimated second derivative J"est for simulated ZnO:P3HT solar cells with varying
carrier mobilities and carrier generation in the 10-nm exciton diffusion length region.
In all cases, the electron mobility is set to 10% of the hole mobility.






































-0.1 0.0 0.1


0.2 0.3 0.4


Voltage (V)
Figure 2-48. Simulated J-V curves for ZnO:P3HT solar cells with varying absorption
coefficients in P3HT.


-0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

Voltage (V)


Figure 2-49. Simulated J-V curves for ZnO:P3HT solar cells with varying energy band gap.


tsf496_98 100% absorption
tsf496_103 80% absorption
-- tsf496_104 50% absorption
- tsf496_105 20% absorption




t. 7
-

































-4 --
a = 100%

-5 i i
0.0 0.2 0.4 0.6 0.8 1.0

Voltage (V)
Figure 2-50. Simulated J-V curves for ZnO:P3HT solar cells with varying band gap and
absorption in the P3HT region.


0.0 0.1


0.2 0.3 0.4 0.5


Voltage (V)
Figure 2-51. Simulated J-V curves for ZnO:P3HT solar cells with P3HT energy band gap of 1.0
eV and hole mobility of 500 cm2/Vs.


-4 -
-40.1
-0.1
















0.05


0.10


E 0.15


S0.20

0
o 0.25
0
>-


0.00 0.01 0.02 0.03 0.04 0.05 0.06


X-Coordinate (pim)


le+23


E

0
*. 1


0-

Co



a(
a

Q-


e+22




le+21



le+20(



le+19


0.10
0.15
Cy o 0.20
rinate (p,)


Figure 2-52. Photogenerated carrier distribution in pairs/cm3 for A) the full unit cell and B) the
region of 13 nm < x < 27 nm along the edge of the ZnO nanorod. Note that the x and
y axes do not follow the same scale.




























Figure 2-53. Carrier mapping scheme for two-stage simulations. Gray regions are ZnO and blue
regions are P3HT. Orange arrows represent the nearest interface points for
photogenerated carriers in the region defined by the dotted white lines.


















0.05


0.10


0.15
E

0.20

0
0 0.25

M 2e+21
0.30 M 4e+21
M 6e+21
8e+21
0.35 M le+22


0.40


0.00 0.01 0.02 0.03 0.04 0.05 0.06

X-Coordinate (tm)


Figure 2-54. Photogenerated carrier distribution for a simulated cell with LD = 20 nm.











S1.E+25
E
1.6E+25
1.4E+25
1.2E+25
L-


X 8E+24
6E+24
S41E+24
2 2E+24
-
0
10 nm 20 nm 40 nm Full Cell

Exciton Diffusion Length


Figure 2-55. Cumulative number of photogenerated carriers in simulated cells with varying
exciton diffusion length.

















------- ---- _



1



F',
IFq n


0 0.01 0.02 0.03 0.04

X Coorindate


0 0.05 0.1 0.15

Y Coordinate


0 0.05


0.1 0.15

Y-Coordinate (unm)


3E+23

2.5E+23

2E+23

1.5E+23

1E+23

5E+22

0


0.2 0.25 0.3


0.2 0.25 0.3


Figure 2-56. Photogenerated carrier distribution along x- and y- coordinates for models with
varying exciton diffusion lengths.


0.05 0.06 0.07


3.00E+23

2.50E+23

2.00E+23

1.50E+23

1.00E+23

5.00E+22

0.00E+D0


1.8E+23

1.6E+23
1.4E+23

1.2E+23

1E+23
8E+22

6E+22
4E+22

2E+22











6e+21 --


5e+21-


4l+21 -


3e+21


2e+21


le+21


0


-le+21


Ar

I/ id


I2
LJ I ESi


Y coordinate (i' 0


Y Coordinate (,m)


0.00


0.01 0.02 0.03 0.04 0.05 0.06


X Coordinate (inm)
Figure 2-57. Contour plots of the photogenerated carrier difference between the full absorption
simulation and the 40 nm LD simulation.






























Figure 2-58. Photogenerated carriers at the tip (left axis) and base (right axis) corner points of
the nanorod in simulated hybrid cells.


-3 -
-0.1


0.0 0.1 0.2 0.3 0.4

Voltage (V)


Figure 2-59. J-V curves for simulated solar cells using line-source carrier generation.


1.60E+24 3.249E+22
C. 'a
'F
a 1.40E+24 -- --NanorodTip 3.248E+22 0
o C
0 1.20E+24 -- -Nanorod Base 3.247E+22 z

" 1.00E+24 3.246E+22 E
E 8.00E+23 3.245E+22
--

m 6.00E+23 3.244E+22
1 4.00E+23 3.243E+22 C
c 2.00E+23 3.242E+22
O.OOE+00 3.241E+22
. 10nm 20nm 40nm Full

Exciton Diffusion Length










Table 2-17. Cell performance measures for real and simulated solar cells.
Cell Jsc (mA/cm) Voc (V) FF i (%)
Real 2.20 0.44 0.56 0.53
10 nm LD 0.59 0.44 0.84 0.22
20 nm LD 1.01 0.45 0.84 0.38
40 nm LD 1.47 0.45 0.84 0.56
Full Cell 1.27 0.45 0.85 0.48


0.0 -


-0.2 -


E -0.4 -
o

E -0.6 -


c -0.8 -


o -1.0 -

1.2
-1.2 -


-1.4 -


Voltage (V)
Figure 2-60. J-V curves for simulated cells with a 40 nm LD and varying carrier mobility.

Table 2-18. Performance measures for simulated cells with 40 nm LD and varying mobility
values.
Mobility Jsc (mA/cm2) Voc (V) FF i (%)
0.01 x 1.47 0.45 0.78 0.52
0.1 x 1.47 0.45 0.84 0.56
1 x 1.47 0.45 0.84 0.56
10 x 1.47 0.45 0.84 0.56
100 x 1.47 0.45 0.84 0.56
Real Cell 2.20 0.44 0.56 0.53

















1.8e+22

E 1.6e+22 0.0
S2.0e+21
1.4e+22 .. .
1.2e+22
1.0e+22
0 8.0e+21
( 6.0e+21
"D 4.0e+21
C
0 2.0e+21
0
S 0.0
0 0.00
0.05
0.10
0.15



0.35 0.02
0.40 .0 0.01 0062
0.00 ,,o

B)




1.8e+22

E 1.6e+22

-1.4e+22 O
0-. 0
1.2e+22
S 1
S1.0e+22 0
O 8.0e+21
a 6.0e+21 I l
a 4.0e+21 I
o 2.0e+21
0.0
.0.00
0.05
S 0.10
CO 0.15 0.06
S0.2 0.05
0.05
0.04
)0.30 0.02 0.03
0.40 0.01 G -,,a,
0.00 ,,oo'

Figure 2-61. Photogenerated carrier distribution for a simulated unit cell with (A) LD = 40 nm
and (B) LD = 40 and the P3HT absorption coefficient increased by 50%.
















0.05


0.10


0.15


0.20


o 0.25
>- -4e+20

0.30-3e+20
-2e+20

0.35
le+20
2e+20
M 3e+20
0.40 M 4e+20


0.00 0.01 0.02 0.03 0.04 0.05 0.06

X Coordinate (jim)


Figure 2-62. Difference in photogenerated carriers (in pairs/cm3) between 150% and 100%
P3HT absorption coefficients in simulated cells with 40 nm LD.





















0.0 -


-0.5 -


-1.0


-1.5


-Z.U


I I I I


Voltage (V)


Figure 2-63. J-V curves for simulated cells with varying absorption coefficient in P3HT.

Table 2-19. Generated carriers and performance measures for simulated solar cells.

Cell Total Carriers Jsc) FF
Cell (pairs/cm3) (mA/cm2) V (V)
40 nm 3.14 x 1025 1.47 0.45 0.78 0.52
Full Cell 3.23 x 1025 1.27 0.45 0.85 0.48
40 nm, 150% a 3.42 x 1025 1.24 0.45 0.79 0.44


-- Standard Absorption Coefficient
-- + 50% Absorption Coefficient


j/









CHAPTER 3
ORGANIC AND HYBRID SOLAR CELL PROCESS DEVELOPMENT

Introduction

The experimental details for process development of organic and hybrid solar cells are

presented in this chapter. In the first section, pretreatment of the indium tin oxide anode is

studied. This work was done in collaboration with the research group of Dr. Chinho Park at

Yeungnam University in South Korea, particularly Jiyoun Seol. This work was previously

presented at the 2006 World Conference on Photovoltaic Energy Conversion sponsored by IEEE

and was published in their Conference Record [80]. The second section focuses on the

development of bilayer photovoltaic cells using absorbing polymers. The next section describes

efforts to characterize solvents appropriate for use in hybrid bulk heterojunction films. The

fourth section details work characterizing hybrid films and the fabrication of photovoltaic cells

from these films. The final section introduces Particle Induced Nanostructuring, a process for

hybrid film deposition intended to control the distribution of nanocrystals in the polymer matrix.

ITO Anode Treatment

Organic solar cells incorporate transparent conducting substrates as an anode, with indium-

tin-oxide (ITO) coated glass most often used. ITO films offer several positive characteristics as

substrates for optical devices, including a high luminous transparency, good electrical

conductivity, and good infrared reflectivity. For these reasons, ITO is widely adopted as

transparent anodes in light-emitting diodes, liquid crystal displays, and solar cells [4, 81-82].

ITO coated glass substrates are commercially produced by sputter deposition followed by

processes to improve surface roughness and microstructure. As-received substrate surfaces,

however, have to be further processed prior to application to current flowing devices such as

OLEDs and solar cells because a sputter-deposited surface microstructure and chemical









composition can be degraded during extended device operation. Surface treatment of an ITO

surface by several techniques can alter the chemical and physical properties of the surface such

as work function, surface roughness, and oxidation property, and thus it could affect the device

performance.

In this study, nitrogen and oxygen plasma treatment along with electron bombardment of

commercially available ITO-coated glass substrates were revisited as approaches to improving

the efficiency and stability of organic solar cells. The effect of these surface treatments on the

surface morphology and chemical composition was characterized, and organic solar cells with

the structure ITO/PEDOT:PSS(50 nm)/CuPc(25 nm)/C60(15 nm)/Al(100 nm) were fabricated

and the device performance measured.

The substrate used in this study was commercially available ITO-coated glass with an ITO

film thickness of- 1800 A and sheet resistance of 7 Q/sq. The as-received substrate was

chemically cleaned by sequential ultrasonification in trichloroethylene (TCE), acetone, and

methanol, followed by nitrogen blow-drying. The cleaned substrate was then exposed to either a

nitrogen or oxygen plasma or an electron beam. Plasma treatment was carried out in a barrel-

type plasma chamber for 10 min at a power input in the range 50 to 300 W and pressure in the

range 50 mTorr to 1 Torr with N2 gas flow. Electron beam irradiation was performed for 15 sec

in a nitrogen environment with beam power varied from 0.5 to 2 kGy (kJ/kg). After treatment a

PEDOT:PSS layer was spin-coated and dried in a vacuum oven. Organic films (CuPc and C60)

and aluminum were deposited at room temperature in a thermal evaporator with a base pressure

of 2.0x10-6 Torr. The chemical composition and surface morphology of the ITO surface were

measured using XPS, AFM, and a video contact angle system (VCAS). Power conversion









efficiencies were measured under illumination from a solar simulator set to produce an AM 1.5

100 mW/cm2 spectrum.

The measured contact angle of a water droplet on an ITO substrate changed significantly

after surface treatment. The contact angle for untreated ITO substrates was 650, while films

subjected to electron beam treatment showed a reduced contact angle of- 500. After exposure to

the N2 plasma, the contact angle decreased dramatically to a value < 100. In both cases, the

reduction is due to two effects. First, the exposure to reactive radicals removes contamination

from the film surface that may have remained after wet chemical cleaning. Secondly, these

treatments increase the activity of the film surface by incorporating nitrogen radicals into the

film. This effect is significantly stronger in the case of N2 plasma than in the case of electron

beam treatment because the plasma treatment supplied a greater flux of highly reactive nitrogen

radicals to the film surface. The change in surface polarity in the case of electron beam

treatment was not as significant as that in the case of plasma treatment, and as electron beam

energy is significantly increased (larger than 2 kGy), the ITO film started to change its color to

light gray, which degraded the luminous transparency of the substrate.

AFM measurements were performed to quantify the surface roughness of the ITO films.

As-received ITO films showed a RMS surface roughness of 1.1 nm. This value was reduced to

0.8 nm by electron beam treatment and < 0.6 nm by N2 plasma treatment at optimized

conditions. Sputtered ITO film generally contains irregular surface features, even though the

subsequent polishing and annealing of the film improves its roughness. The surface treatment

procedures used in this study are expected to attack the higher surface features first, resulting in a

decrease in surface roughness [83, 84]. The effect, however, was not very significant, because

the surface roughness of the as-received substrate was already low enough.









XPS spectra from treated films showed a change in the chemical composition of the film

surface. Films treated with N2 plasma and electron beam bombardment showed an increase in

nitrogen content accompanied by a decrease in oxygen composition, confirming that near-surface

oxygen is replaced by nitrogen in films subjected to these treatments as shown in Table 3-1. In

the case of N2 plasma exposure, there was a decrease in the In/Sn ratio of the films as well,

which can cause a slight decrease in the work function of the films [85]. It is believed this

change is a result of indium reacting with nitrogen radicals to form InN, which is subsequently

removed from the film. The chemical shift of In-Sn-O bonding was also investigated by XPS,

but no noticeable change in chemical shift was observed under the treatment conditions

investigated in this study.

The incorporation of nitrogen into the near surface region of films treated in N2 plasma was

confirmed by glow discharge spectroscopy (GDS). Films exposed to the plasma at a constant

power but increasing pressure showed an increase in nitrogen levels in the film up to 700 mTorr.

At higher pressures the nitrogen level decreased slightly, due to a decrease in nitrogen radical

activity in the plasma.

Organic solar cell devices were fabricated from the surface-treated ITO films to determine

the impact on device performance as shown in Figure 3-1 and Table 3-2. Electron beam

treatment produced no change in device efficiency; however, there was a decrease in Voc and an

increase in Jsc in these devices. 02 plasma treatment gave a slight increase in Jsc, but a

significant drop in Voc led to a reduction in overall device efficiency. N2 plasma treatment

resulted in cells with efficiency nearly double that of the untreated ITO cells. All treated ITO

devices showed an improvement in the shape of the I-V curve compared to that of untreated ITO

devices. The changes in performance can be attributed to several effects due to the treatment,









including surface roughness, chemical stability, and reduced work function. Electron beam and

N2 plasma treatment replace oxygen with nitrogen, reducing the electron affinity of the ITO film

and resulting in a lowering of the work function that improves the charge collection efficiency.

The effect of several ITO surface treatments on the performance of organic solar cells was

determined. It was shown that exposure to a N2 plasma was more beneficial than either the

oxygen plasma or e-beam treatment. The N2 plasma was most successful in improving the

surface characteristics, as evidenced by the extent of lowering the contact angle and decreasing

the surface roughness (AFM). This treatment incorporates nitrogen into the near surface region

and produces a slight change in the In/Sn ratio, which reduces the ITO work function. These

changes optimize the energy band diagram and improve charge collection at the ITO anode.

Bi-layer Organic Solar Cell Fabrication

A process for fabricating bi-layer organic solar cells with the cell structure

ITO/PEDOT:PSS/P3HT/C60/Al was developed. Figure 3-2 shows the energy band diagram for a

cell with this structure [45, 86]. This material system has received much attention for

applications in bulk heterojunction solar cells due after the discovery of ultrafast charge transfer

at interfaces between conjugated polymers and C60 molecules.

In these bi-layer organic cells, P3HT serves as the primary absorber layer in the cells, with

a bandgap of approximately 1.7 eV and a very strong absorption coefficient. Excitons generated

in the polymer are separated at the C60 interface, with electrons dropping to the lower energy

level of the C60 and holes returning to the P3HT. Electrons are transported to the backside

aluminum contact, while holes are transported through the P3HT layer and the hole transport

layer (HTL) of PEDOT:PSS to the transparent ITO frontside contact.









Cell Fabrication Procedure

Bi-layer organic solar cells were fabricated on ITO-coated glass substrates. The substrates

were patterned, cleaned, and treated in-house under N2 plasma. A hole transport layer of

PEDOT:PSS and the absorber layer of P3HT were spin-coated onto the substrate and dried under

vacuum. The electron transport layer of C60 and the aluminum back contact were deposited by

evaporation. For cells to be transported, glass encapsulation was performed to extend the cell

lifetime. The encapsulation process is described in greater detail below. A graphic synopsis of

the fabrication process is shown in Figure 3-3.

Substrate Preparation

The cells were fabricated on ITO-coated glass substrates obtained from Samsung-Corning.

The ITO films had a resistance of < 7 Q/sq. and a thickness of approximately 0.18 nm. As-

received ITO substrates were cut into 2.5 x 2.5 cm squares and patterned by HC1 vapor etching.

A 2 mm wide strip of electrical tape was fixed to the ITO substrate to cover a strip that would

become the ITO anode. The substrates were suspended from the inside lid of a glass dish with a

small amount of HC1 in the bottom of the dish to generate vapor. Vapor etching occurred over

25 min, after which the substrates were thoroughly rinsed with de-ionized water and blown dry

with nitrogen. They were then chemically cleaned by successive 10 min sonication steps in

trichloroethylene, acetone, and methanol, followed by blow-drying under nitrogen. Cleaned

substrates were subjected to N2 plasma for 10-min in a barrel-type plasma chamber wrapped in

induction coils. The chamber offered a continuous flow of nitrogen at a constant flow rate

during the treatment, and the conditions used for treatment were optimized to a supplied power

of 50 W and a chamber pressure of 200 mTorr.









Spin-Coating

A solution of PEDOT:PSS in water obtained from Bayer was filtered with a 0.45 gm

syringe filter and spin-coated onto the treated substrate for 30 sec. The film was then dried in a

vacuum oven at 90 OC for 30 min. A calibration curve was generated for dried film thickness vs.

spin-coating speed and is shown in Figure 3-4. The curve was fit to power law as has been noted

in the literature (87), with the curve fitting Equation 3-1.

t =(e14.7)* -0.99 (3-1)

The absorber layer of P3HT was then spin-coated from a solution of 5 mg/ml in

chlorobenzene. The films were spin-coated for 30 sec and then dried under vacuum. The film

thickness versus spin-coating speed was fit to Equation 3-2, and the results are shown in Figure

3-5.

t= (e12.5)*o 0.93 (3-2)

Both spin-coating steps occurred in a clean room environment. Film thicknesses for these

measurements were performed with a profilometer. After the active layer was dried, selected

areas of the substrates were wiped clean of the polymer films to provide clean surfaces for

external contacts.

Evaporation

The cells were placed in a glove box under nitrogen and loaded into a thermal evaporator.

C60 was evaporated through a shadow mask at a rate of approximately 2 A/s under a pressure of

10-6 Torr. Finally, a thick layer of aluminum was rapidly evaporated through a second shadow

mask to form the back contact of the cells.

Encapsulation

Encapsulation was performed to extend the lifetime of the fabricated cells by shielding

them from moisture in the atmosphere. The two components of the encapsulation procedure are









the encapsulant and the dessicant. The encapsulant consisted of a small glass slab to serve as the

backing and a rubber spacer to prevent contact between the substrate and encapsulant glass. The

assembly of substrate / spacer / encapsulant was sealed with epoxy.

Prior to fixing the encapsulant onto the substrate, a dessicant was added to protect the cells

from moisture exposure. The dessicant used was barium oxide. A small pouch was created from

part of a piece of weighing paper. Inside the nitrogen glove box, the pouch was filled with BaO

powder and sealed with double-sided tape. It was then affixed to the inside of the encapsulant

glass. Several tiny holes were punctured in the pouch to allow moisture to reach the dessicant,

and the final assembly was sealed to the substrate with epoxy.

Film Drying

Due to the potential impact of residual solvent in these films, care must be taken to ensure

drying is complete after each film deposition step. Residual organic solvent from the active layer

film serves as an insulator to cripple electrical performance, while excess water remaining from

the HTL film can oxidize and degrade the active layer polymer. To confirm the effectiveness of

the drying step, FTIR spectra were compared between the dry films, the solutions used for spin-

coating, and spectra obtained from Sigma-Aldrich for the pure solvent. Figure 3-6 shows the

spectra for a P3HT film spin-coated from chlorobenzene, as well as spectra for the solution and

pure solvent. The large peak at approximately 3050 cm-1 in the chlorobenzene spectra is clearly

visible in the solution spectra, but is noticeably missing in the P3HT film spectra. Also, several

sharp, narrow peaks between 1600 and 500 cm-1 match in the solution and solvent spectra, but

are missing from the film spectra. In Figure 3-7, the broad O-H stretching peak from

approximately 3700 to 3000 cm-1 is an obvious feature in the FTIR spectrum for pure water.

This wide peak is also obvious in the solution spectrum but is missing from the PEDOT:PSS film

spectrum. Similarly, the strong peak at approximately 1640 cm-1 in the water spectrum is present









in the solution spectrum but missing from the film spectra. These FTIR spectra provide

confirmation that film drying is complete under the conditions specified, and in that respect,

these films are suitable for use in organic solar cells. Both solvents analyzed, chlorobenzene and

water, produce strong peaks that are easily distinguished from the polymer spectrum, making this

an effective technique for identifying residual solvent in the films after deposition.

PEDOT:PSS

The PEDOT:PSS layer in bi-layer solar cells was deposited with a thickness in the range of

80 to 100 nm. One potential issue in this charge transport layer is that it is susceptible to

pinholes, which can lead to shorting of the devices. Pinholes can be generated during the drying

process as the solvent evaporates and rises through the drying polymer. To see the if pinholes

were problematic in this deposition process, cells were fabricated using a single-layer and

double-layers of PEDOT:PSS. The double-layer devices should eliminate the presence of layer-

spanning pinholes by providing two separate films so that any pinholes would only span half of

the final film.

Experiments were performed by fabricating bi-layer solar cells using single- or double-

layers of PEDOT:PSS. A consistent final film thickness of 80 nm was used for the PEDOT:PSS

layer: one 80 nm layer for the single-layer devices, and two 40 nm layers for the double-layer

devices. All other cleaning, preparation, deposition, and encapsulation steps were held constant.

After fabrication, the cell performance was characterized under 100 mW/cm2 light from a solar

simulator. The resulting J-V curves are shown for single-layer devices in Figure 3-8 and for

double-layer devices in Figure 3-9.

From the results, it is clear that the performance of cells using the double-layer

PEDOT:PSS film suffer dramatically. Of the four cells using the double-layer structure, two fail

to show any diode characteristics, while the other two show only a minimal photovoltaic effect.









While the performance of the cells using the single-layer PEDOT:PSS were not phenomenal,

they all showed a significant photocurrent and diode characteristics.

Cells fabricated with a single 80 nm layer of PEDOT:PSS showed efficiencies as high as

0.168% (cell 8-2), with all cells showing efficiencies of at least 0.047%. In contrast, cells

fabricated with the double-layer structure failed to show a measurable efficiency, although

estimates put them in the range of 0.015% for the best cell (cell 9-1). Open circuit voltage values

for the single-layer cells were in the range of 0.15 V, while values for the double-layer cells are

less than 0.05 V. Short circuit current density values ranged from 1.5 to 2.5 mA/cm2 for the

single-layer cells and from 0.7 to 1.4 mA/cm2 for the double-layer cells.

The data show that pinholes are not a concern for the PEDOT:PSS films under these

deposition and drying conditions. In fact, there is a different effect causing the double-layer

films to perform more poorly than the single-layer films. The total thickness of the layers was

held constant to keep series resistance constant. However, the series resistance is impacted by

the film resistivity in addition to the path length. It seems that the double-layer films showed a

higher resistance to current flow due to the lower cell performance. For positive biases, the

single-layer films showed current densities 7 10 % higher than the double-layer films. This can

be attributed to two possible causes interfacial resistance and film resistance. Because of the

double-layer structure, there is an extra film interface which could cause an increase in the

overall resistance. Additionally, the inherent film resistivity could be increased due to the

deposition conditions. The films were deposited via spin-coating, with the 40 nm films

deposited at a much higher rpm than the 80 nm films. This causes faster solvent evaporation and

can inhibit the ability of the polymers to self-align in a configuration that could minimize









resistance. This effect has been observed in the past, particularly in active layer films where

extremely slow solvent evaporation has resulted in strong increases in the film's conductivity.

Another effect observed in PEDOT:PSS deposition is that of different particle filtering

speeds that produce different solution properties. In polymer electronics processing,

PEDOT:PSS is typically filtered prior to deposition to remove particles in the solution and

provide a smoother film upon spin-coating. In our labs, the PEDOT:PSS solution was filtered by

hand using a 0.45 tm disc filter attached to a 100 ml syringe. It was found that there were two

different methods of filtering depending on the person doing the work, but because the work was

done by hand rather than with a mechanical system. In one method, the solution was filtered

very slowly, with the worker applying just enough pressure to force the solution through the

membrane in a dropwise fashion. The filter was periodically replaced as the filtration became

more difficult due to clogging. This slow filtration method resulted in a relatively low viscosity

solution. In the other method, the solution was forced through the filter in one steady motion.

This process was much faster, with several ml of the solution passed through in a matter of

seconds, and resulted in a more viscous solution. The difference was noticed while developing

calibration curves for film thickness depending on spin-coating speed. The slow filtration results

in a much thinner film than the fast filtration step because a lower percentage of the polymer is

forced across the membrane at the lower pressure difference. Calibration curves using the two

solutions are shown in Figure 3-10. The data set labeled "051116" represents the slow filtration

method, while the other three data sets show the faster filtration method. For cell fabrication, the

fast filtration method was used to provide the ability to deposit approximately 100 nm thick films

while still using a reasonably fast spin-coating speed to produce smooth, uniform films.









P3HT

Prior to cell fabrication, P3HT films were analyzed under AFM to determine the surface

quality of the films. Measurements were made under different solution concentrations and spin-

coating speeds. The results, shown in Figure 3-11, show high-quality films with low surface

roughness, indicating their suitability for bi-layer cell construction. All films spin-cast from the

5 mg/ml solution showed an RMS surface roughness of around 1 nm. For the films cast from the

10 mg/ml solution, the RMS surface roughness was between 2.5 nm and 3.6 nm, with the higher

roughness occurring at the slowest spin speed. The data from the images shown in Figure 3-11 is

tabulated in Table 3-3.

Bi-layer Cell Fabrication

Following these film and process characterization investigations, bi-layer organic solar

cells were fabricated with the device structure ITO/PEDOT:PSS/P3HT/C60/Al. The bi-layer

device structure is commonly used in OLED devices [88] and has been explored in molecular

organic photovoltaics [89], but less work has been performed regarding bi-layer devices using

polymer active layers [90]. The devices were fabricated on ITO-coated glass substrates prepared

as described previously. A PEDOT:PSS film was deposited by spin-coating for 30 sec at 2500

rpm and drying for 30 min. The active layer of P3HT was deposited by spin-coating from a 5

mg/ml solution in 1,2-dichlorobenzene for 30 sec at 3500 rpm and dried under vacuum. The

samples were loaded into an evaporator where a 150 A film of C60 was deposited, followed by a

800 A Al electrode.

Bi-layer cell J-V measurements were performed at Busan National University in Busan,

South Korea using a Keithley I-V measurement system under illumination from a solar

simulator. Current measurements were converted to current density by dividing by the active

cell area (0.04 cm2). Illuminated measurements were performed under 100 mW/cm2









illumination. Power conversion efficiency (rI) was calculated using Equation 3-3, with Vmax,

Imax, and Jmax representing the voltage, current, and current density at the maximum power point.

The fill factor (FF) is calculated using Equation 3-4 with Voc and Jsc being the open circuit

voltage and short circuit current density.



P,, J1mW/cm2


FF = Vmaxmax (3-4)
V J
OC SC

The rectification ratio (RR) of a diode is the ratio of forward to reverse current at some

applied bias. Higher rectification ratios show stronger diode characteristics in the I-V curve for

devices. Rectification ratios displayed in this section were calculated from dark current

measurements at 0.5 V of forward and reverse bias unless otherwise noted.

Bi-layer solar cells were fabricated using the procedure detailed previously, but with no

treatment performed on the ITO electrode. J-V curves for these cells are shown in Figure 3-12.

The cells fabricated in this set all showed a measurable photocurrent, demonstrating

functioning solar cell behavior. The best-performing cell in the set, Set I 3, showed a power

conversion efficiency of 0.04%. Despite having the lowest Voc of all cells in the set, this

champion cell showed a short-circuit current density of 1.55 mA/cm2, which was significantly

higher than any other cell in the set. Performance was low, but measurable, in all cells. With the

exception of cell I 3 with a Voc of 0.11 V, all cells showed a Voc of almost exactly 0.15 V.

The fill factor for the cells ranged from 0.15 for cell I 4 to 0.24 for cell I-1. Cells I-1 and I-4

showed a Jsc of approximately 0.5 mA/cm2. The Jsc for cell I-2 was 1.03 mA/cm2.

Another set of bi-layer cells were fabricated with the same cell structure, but applying

plasma treatment to the ITO substrate. The performance of these cells was poor, with the lack of









performance attributed to possible poor encapsulation resulting in cell decay before

measurement, or non-optimized plasma treatment inhibiting the cell performance. These cells

showed a minimal photovoltaic effect, with open-circuit voltages below 0.1 V and short circuit

current densities no higher than 1.10 mA/cm2. Although fill factors were in the range of 0.20-

0.25, efficiencies were no higher than 0.02% due to the low photovoltage and current in the cells.

J-V curves for the cells are shown in Figure 3-13 and the cell performance is detailed in Table 3-

4.

Solvent Comparisons

Various solvents were compared to determine their applicability for hybrid bulk

heterojunction films. The solvents were all commonly available chemicals, so no high-cost

specialty materials were used that would add substantial cost to the fabrication process.

The goal was to identify a solvent or a family of solvents that would provide the highest-

quality hybrid films for cell fabrication. The solvents used for the tests are shown in Table 24.

Only those which showed an appropriate level of solubility were used in further testing. Polymer

solubility was qualitatively assessed by mixing a small amount of the polymer with a few

milliliters of the chosen solvent into a small vial and subjecting the mixture to ultrasonication for

at least one hour. Solvents with solubility labeled as "No" showed solid flakes of the polymer in

the clear solvent after the mixing. Those labeled as "Poor" resulted in a color change of the

liquid to indicate some degree of dissolution, but still contained significant solid particles of

polymer. The label of "Yes" indicates that the polymer was fully dissolved to give a red colored

solution that is characteristic of P3HT. The solvent's normal boiling temperature and polarity

values are listed in Table 3-5 as these parameters strongly impact the film-forming properties of

a solution. The values of polarity in the chart were taken from a solvent miscibility and polarity

chart from Phenomenex [91] where higher values correspond to more polar solvents, with water









having a value of 9.0. In the table, DMF is dimethyl formamide, MED is methyl ethyl ketone,

THF is tetrahydrofuran, and TCE is trichloroethylene.

Hybrid films deposited from the solvents that showed solubility for P3HT were analyzed

with profilometry, AFM, and imaged with optical microscopy and SEM. Film roughness

measurements taken from profilometry are shown in Figure 3-14. The results are plotted against

both boiling temperature and polarity, although no trend is evident in either case. These

measurements were taken over 5 line scans of 1.0 to 1.5 mm on the surface of two separate

substrates.

AFM surface scans were taken for films deposited using 6 of the selected solvents, and rms

surface roughness was measured for 5 x 5 [tm and 1 x 1 tm surface areas. The results for each

test are shown in Figures 3-15 and 3-16, with separate graphs shown to plot rms surface

roughness vs. solvent boiling temperature and solvent polarity.

On the large-scale profilometer measurements, chlorobenzene was the best performing

solvent, showing a mean surface roughness of 64.9 nm on a line scan. The worst performers

were THF and o-xylene, with mean rms values of 136 and 198 nm, respectively. For the 5 x 5

[tm AFM measurements, chloroform showed the lowest roughness, with the rms roughness being

measured at 16.2 nm. THF displayed the highest roughness, with a value of 36.8 nm. For the

small-area 1 x 1 tm AFM measurement, Toluene was the superior solvent with an average rms

value of 2.93 nm. Again, THF showed the highest value, at 15.0 nm. The results are

summarized in Table 3-6.

SEM and optical microscope images of film surfaces were taken to visually assess the film

quality in conjunction with the surface roughness data. Figure 3-17 shows optical images of









selected films that were representative of the samples taken for each solvent. Figure 3-18 shows

SEM images from films, also representative of the data taken for each film solvent.

In the optical microscope images, dark spots on the image represent surface features on the

films. Some of the larger dark spots on the image, such as in the top left corer and near the

right hand side of the 100x THF image, appear fuzzy because they are beyond the depth of field

of the image, and different areas of this spot could be visualized by re-focusing the lens. In the

1Ox magnification images, it is noted that all solvents produce at least a few of these large black

features. Chlorobenzene, o-dichlorobenzene, and toluene show these spots in smaller sizes and

regularities than the other materials. Benzene shows large wispy collections of the dark spots

that were not seen from other solvents, which seem to suggest precipitation or clustering

occurring on a much larger scale than the other solvents show. These features are also visible in

the 10Ox image for benzene, and again, it does not appear for other solvents. The 10Ox image for

THF shows a high frequency of very large features. Dichlorobenzene, on the other hand, shows

a very low frequency and size of features in both the 1Ox and 100x images. Chlorobenzene

shows a few large features, but the image overall shows a high-quality film with a low frequency

of features.

The SEM images shown in Figure 3-18 display surface images of the hybrid films at

5,000x and 15,000x magnification. These magnification levels are relatively low for SEM

images, but attempts to focus the electron beam to produce images in the 30,000x to 50,000x

range resulted in beam damage of the sample. In fact, an example of this is visible in the

15,000x image for o-dichlorobenzene. The dark box near the center of the image, just to the left

of the bright surface feature, represents an area where a tighter focus was attempted and the

sample was burned. In these images, surface features of the films are clearly visible as bright









spots in the secondary electron image. Images for chlorobenzene and o-dichlorobenzene show

high-quality films with relatively few surface features. The images for TCE also shows a

minimal amount of surface features, particularly in the 15,000x image. Chloroform displays a

moderate amount of features, but less so than THF, which similarly to the optical images shows

the poorest film quality.

Attempts to identify the composition of these surface features failed to yield results.

Backscattered electron images of the films showed images of the features similar to the

secondary electron images, but with no additional bright spots to demonstrate possible clustering

of the higher density nanocrystals. Compositional scans were performed using energy-dispersive

x-ray analysis (EDX), and these scans detected cadmium and selenium at a constant

concentration in both the features and in smooth areas of the film surface.

From this study, it was determined that chloroform, chlorobenzene, and o-dichlorobenzene

are good candidates for hybrid bulk heterojunction film deposition. These solvents all produce

films that were among the lowest in surface roughness for all of the measurement techniques

used. Additionally, the film quality could be visually confirmed from optical microscopy and

SEM surface images.

Hybrid Bulk Heterojunction Cell Fabrication

Bulk heterojunction photovoltaic cells were fabricated using P3HT as the absorbing

semiconductor and nano-CdSe as the electron transporter. The cell design was the modeled after

the bi-layer organic cell design described previously. The bulk heterojunction cell structure was

ITO/PEDOT:PSS/P3HT:CdSe/Al, which is very similar to the bi-layer structure with the

exception of CdSe replacing C60 as the electron acceptor, and that acceptor is now blended into

the active layer film rather than deposited on top. Performance measurements for these cells

were performed in-house rather than remotely, so the encapsulation process was not performed.









Measurements were performed with a Keithley I-V-L measurement system under illumination

from a JEOL solar simulator.

Nanocrystal Synthesis and Surfactant

CdSe nanocrystals were synthesized using the solution-phase growth mechanism

demonstrated by Peng et al. [92]. In this method cadmium oxide is added to a flask containing

tri-n-octylphosphine oxide (TOPO) and hexyl-phosphonic acid. The solution is then heated

before the addition of selenium powder dissolved in liquid tri-n-octylphosphine. The reaction

progresses and is halted by removal from the heating source. The nanocrystals, coated in a

surfactant layer of TOPO, are precipitated from the solution by the addition of methanol and

centrifugation. The TOPO-coated nanocrystals can further be modified by dissolution in

pyridine, precipitation using hexane, and centrifugation to isolate the crystals. This ligand

exchange process replaces the TOPO molecules with pyridine molecules on the surface. After

-5 steps of this process, the TOPO coating is fully replaced and the pyridine-coated nanocrystals

can be used for processing. CdSe nanocrystals used in this dissertation were synthesized by Md.

Azizul Hasnain and Trong Nguyen Tam Nguyen at Yeungnam University unless otherwise

noted.

Solutions were generated using CdSe nanocrystals with both types of surfactant coatings,

and films deposited from these solutions were tested to determine the appropriate deposition

parameters. The solutions consisted of a 60:40 mixture of CdSe nanocrystals and P3HT polymer

dissolved in chloroform with a variable volume fraction of pyridine added to enhance the CdSe

solubility. The surface roughness of spin-coated films is shown in Figure 3-19, plotted against

the pyridine content of the solution. Note the scale difference in the two graphs. These

measurements were performed by Md. Azizul Hasnain at Yeungnam University and are

displayed to clearly represent the effect of nanocrystal surfactant on film quality for hybrid films.









For TOPO-coated nanocrystals, the surface roughness of the films decreased as the pyridine

content increased, up to 50%. This is due to the low solubility of TOPO-coated nanocrystals in

chloroform. Pyridine concentrations of 45-50% were required to generate films with less than 10

nm rms roughness. After surface exchange with pyridine is performed, the pyridine-terminated

nanocrystals become significantly more soluble in chloroform. In this case, pyridine

concentrations of less than 10% yield rms surface roughness values of less than 10 nm.

Hybrid Films

To fabricate bulk heterojunction solar cells with organic polymers and inorganic

nanocrystals, great care must be taken to ensure proper mixing between the two phases. The

exciton diffusion length of most semiconducting polymers is in the range 5 to 20 nm. Because of

this limitation, the organic and inorganic phases in the hybrid active layer must be well-mixed so

that excitons generated in the organic phase can reach an inorganic phase that is within one

diffusion length.

In this investigation, the properties of hybrid films is studied through optical microscopy,

electron microscopy, atomic force microscopy, and surface profilometry. Solutions are

generated by dissolving blends of P3HT polymer and CdSe nanopowder into solvent mixtures of

chloroform and pyridine. Hybrid films are spin-cast from these solutions onto ITO-coated glass

substrates that were subjected to N2 plasma treatment as described previously, and the films were

dried under vacuum.

TOPO-coated CdSe

Initially, hybrid solutions were created with P3HT and TOPO-coated CdSe nanocrystals

with a radius of approximately 5 nm. The solvent used in these solutions was a 50-50 mixture of

chloroform and pyridine, based on the results shown in Figure 3-19. Three hybrid solutions were

prepared with composition shown in Table 3-7.









Films were spin-cast from these solutions, and their visual quality was studied using

optical microscopy. The resulting images are shown in Figure 3-20. The images were taken at

100x magnification, and the films imaged were spin-coated at 3000 rpm from the solutions

described in Table 3-7. In these images, darker regions correspond to features on the surface of

the films. From the images, it is observed that films from the higher-concentrated solutions

show significantly more numerous and larger features than the film from the weakly-

concentrated solution. The weakly-concentrated Solution 3 shows only small and well-dispersed

dark regions. Solutions 1 and 2 show a darker overall image, as is to be expected as these

higher-concentrated solutions produce thicker films. However, they also show large surface

features, some of which appear out of focus due to the depth of field of the microscope. These

observed differences are due to the low solubility of both the polymer and nanocrystals in these

composite solutions when TOPO-coated nanocrystals are used. The polymer is poorly soluble in

pyridine, and the TOPO-coated CdSe nanocrystals are poorly soluble in chloroform, so the only

way to control the film morphology in this solvent system is to reduce the total solute load in the

solution.

The 20.4 mg/ml and 19 mg/ml solutions produced a significantly rougher surface than that

of the 5 mg/ml solution, seen in the images in Figure 3-20. In an attempt to decrease the

roughness of these films, a second spin-coating step was performed using pure chloroform. The

hypothesis was that the solvent would selectively attack the more pronounced surface features,

similar to the effect of nitrogen plasma smoothing the surface of ITO. The resulting film did

show a decrease in roughness, similar to that of the 5 mg/ml solution. An optical microscope

image of the film is shown in Figure 3-21. The surface roughness decreased, but the film was

nearly completely etched during this process. Even under spin speeds up to 8000 rpm, designed









to significantly reduce the contact time between the chloroform and the film, the original film

was nearly completely removed.

P3HT solubility in chloroform and pyridine

To further investigate the effect of the solvent composition on film qualities, studies were

performed to compare the quality of pure P3HT films cast from pure chloroform, a 1:1

chloroform:pyridine mixture, and pure pyridine. Films were deposited from 5 mg/ml solutions

of P3HT in each of the three solvent types and then examined under optical and secondary

electron microscopes and the roughness measured with a surface profilometer.

Under an optical microscope at 100x magnification, clear differences are observed in film

quality depending on the type of solvent used, as shown in Figure 3-22. The chloroform solution

produces a film that appears smooth with a number of small surface features. The film deposited

from the mixed solvent shows a rougher base film with larger and more numerous surface

features. The pyridine solution results in a smooth base film, but with several very large surface

features.

SEM analysis of the chloroform and mixed solvent films show results similar to those

obtained under optical microscopy. Figure 3-23 displays a comparison of these films at 2,000

and 10,000 x magnification. For the films deposited from chloroform solvent, there is a

significant reduction in the number of surface features of the films. In the 2,000x image, there is

one large area feature that is visible at the bottom of the image, but this was not observed to be

common in the film. On the other hand, the films deposited from the mixed solvent show a

distribution of surface features with varying shape and sizes ranging from a few microns to tens

of microns. This offers further evidence of phase separation occurring in the films deposited

from the mixed solvent. Because these films are pure P3HT rather than hybrid films, the

observed features must be regions of P3HT that formed a non-uniform surface. This could occur









through non-uniform precipitation of the polymer during deposition. In the chloroform solvent,

due to the high polymer solubility, the smooth film is a result of a uniform deposition and film

drying process. In the mixed solvent, however, the solubility is poorer and the P3HT will

precipitate more quickly as solvent evaporates during the spin-coating and drying processes,

resulting in the polymer freezing in its current state rather than being allowed to relax to a

preferred alignment that results in a smoother film.

Surface profiles of the films measured through profilometry revealed films with significant

increases in surface roughness as the amount of pyridine in the solvent mixture increased, as

shown in Figure 3-24. When the film thickness was measured using this technique, it was found

that the film thickness decreased as the pyridine content of the solvent increased, due to

decreasing solubility of the P3HT polymer. The profiles shown in Figure 3-24 are the result of a

line scan across the surface of a sample that had the film stripped away on the left-hand side.

The tall, wide peak at approximately 1 mm is the edge where the film was wiped clean and the

film surface was distorted. The polymer film is shown on the right-hand side, which is where the

mean and median film thickness was measured. On the profiles, the solid teal line represents the

median film thickness, the red dash-dot line represents the mean film thickness, and the blue

dashed lines represent + one standard deviation from the mean. Note that not all lines appear in

all graphs due to the scaling.

These scans show an increase in the mean film thickness as the pyridine content of the

solvent increases, resulting from an increase in surface roughess. They show, however, a

decrease in median film thickness, due to the base film thickness being smaller as the polymer is

poorly dispersed by the pyridine solvent. These properties are summarized in Table 3-8, and

represent averages for measurements across different regions of each sample. It is interesting to









note that for the pyridine solvent, the error range for the mean film thickness and the standard

deviation of film thickness is larger than the actual measurement. This is due to the extremely

nonuniform film which shows extremely large features over a very thin base film.

Cell Fabrication

Replacing TOPO-coated nanocrystals with pyridine-coated nanocrystals is the key to

reducing the amount of pyridine needed in the solvent mixture. Because of this, pyridine-coated

nanocrystals were used for attempts at hybrid bulk heterojunction cell fabrication. The

processing steps for these cells were similar to those used for the organic bi-layer solar cells

described earlier. ITO-coated glass substrates were etched with HC1, then cleaned under

ultrasonication in TCE, acetone, and methanol. The clean surface was exposed to a N2 plasma

under specific conditions, followed by deposition of PEDOT:PSS, then the hybrid solution,

followed by evaporation of the Al electrode. Cell performance was tested in the dark and under

simulated solar illumination.

Cells deposited from chloroform solutions

Bulk heterojunction cells were fabricated from pure chloroform solution and a 2% pyridine

in chloroform solution. Nitrogen plasma treatment was performed at 50 W and 200 mTorr for 10

minutes after etching and cleaning of the substrates. A thin film of PEDOT:PSS was deposited

via spin-coating and dried under vacuum. The hybrid solution concentration was 5 mg/ml,

consisting of 60% CdSe by weight. The aluminum electrode was between 125 and 150 nm in

thickness. The resulting dark and light J-V curves for these cells are shown in Figure 3-25.

The active cell performance is low for both cells, particularly the cell deposited from pure

chloroform solvent. The first cell, deposited from the chloroform-pyridine mixed solvent, shows

Jsc = 8.99 x 10-3 mA/cm2 with a maximum efficiency = 7 x 10-4 %. The second cell, in pure

chloroform solution, shows Jsc = 2.62 x 10-5 mA/cm2 and maximum efficiency = 1.3 x 10-6 %.









Although these films show some photovoltaic effect, it is extremely low, presumably due to the

very thin nature of the films. From visual inspection, the films are faint pink in color and highly

transparent. These thin films fail to absorb a significant number of photons to yield high-

performance solar cells.

To enhance film thickness, a new hybrid solution was created with the total solute

concentration doubled to 10 mg/ml while maintaining the nanocrystal weight percentage at 60%.

The solvent was 2% pyridine in chloroform. The resulting J-V curves are shown in Figure 3-26.

This cell showed higher values of Jsc than the cell deposited from the 5 mg/ml solution. The

measured cell performance characteristics were Jsc = 3.39 x 10-2 mA/cm2, Voc = 0.256 V, FF =

0.275, and efficiency = 2.4 x 10-3 %. This is a 2.5x increase in efficiency compared to the 5

mg/ml hybrid solution. Although performance is still low, the film produces photocurrent that is

half an order of magnitude higher.

Thicker-film cells in chloroform solution

By further increasing the concentration of the active layer solution, light absorption in the

cells can be further enhanced. In these cells the weight ratio of the hybrid film was altered from

60% CdSe to 50% CdSe. Because the CdSe serves primarily as an electron transporter while

P3HT is the absorber, the cell performance can be improved by decreasing the CdSe volume in

the hybrid films as long as electron transport pathways exist to allow current flow to the

electrode. A 25 mg/ml hybrid solution with equal weights of P3HT and pyridine-coated CdSe

was dissolved in a solvent of chloroform with 2% pyridine. To determine the effects of the

nanocrystals in the solution, a solution consisting of 12 mg/ml P3HT in the 2% pyridine mixed

solvent was also generated. This solution has essentially the same P3HT concentration in

solution (12.5 mg/ml in the hybrid solution, 12 mg/ml in the pure polymer solution), so the

presence of nanocrystals in the hybrid solution differentiates the two.









Surface profiles of the films generated with a profilometer are shown in Figure 3-27. For

these scans, the film was wiped clean at the left-hand side of the image to allow measurement of

the film thickness. Note that the scale for both profiles is the same. The P3HT film shows a

thickness of 95.7 + 17.8 nm with an rms roughness of 15 nm. The hybrid solution produces a

slightly thicker film at 130 16.8 nm, but with an rms roughness of 239 nm. The inclusion of

nanocrystals increases the surface roughness of the film by more than an order of magnitude.

However, the nanocrystals have only a minimal effect on the film thickness. Although the

hybrid solution had a total solute concentration that was twice as high as the P3HT solution, the

final film thickness only increased by 35%. This demonstrates that the polymer is a much

stronger factor in film thickness than the nanocrystals.

Dark and illuminated J-V curves for cells fabricated from the 25 mg/ml hybrid solution

and the 12 mg/ml P3HT solution are shown in Figure 3-28. The hybrid cell showed performance

characteristics of Jsc = 4.99 x 10-2 mA/cm2, Voc = 0.701 V, FF = 0.232, and efficiency =

0.0175%. These measurements show a short-circuit current density 150% higher than that of the

10 mg/ml hybrid cell shown in Figure 3-26. As expected, J-V curves for the pure P3HT cell

showed poorer performance than that of the hybrid cell. Performance characteristics for the

P3HT cell showed Jsc = 2.03 x 10-2 mA/cm2, Voc = 0.523 V, FF = 0.310, and efficiency = 1.25 x

10-3 %, which are similar to that of the 10 mg/ml hybrid cell. Despite the poor measurements for

the P3HT cell, the dark J-V curve shows a strong rectification ratio, signifying a strong diode.

J-V measurements in the dark for each of these cells showed a strong rectification ratio,

something that was not observed previously for thinner cells. This shows that the thicker films

limit reverse leakage current that was easily driven through the thin active films in the previous

cells.









Cell lifetime measurements were also performed for these two cells. The cells were periodically

measured under dark and illuminated conditions under exposure to room air. The cell

performance decreased quickly with exposure time, as seen in Figure 3-29. The first

measurements for each cell occurred after approximately 40 minutes after removal from a

nitrogen glove box. The short-circuit current density for the hybrid cell decays approximately

70% in 12 min between the first and second measurements and by more than 95% in the 30 min

between the first and third sets of measurements. The short-circuit current density for the P3HT

cell decays more slowly by approximately 40% in 11 min between the first and second sets of

measurements, 50% in 26 min between the first and third sets, and 80% in approximately 3.5 hr

between the first and final measurements. The data was fit to an exponential decay function of

the form shown in Equation 3-5 with J in units of mA/cm2 and t in units of min. This fit was

performed using Sigmaplot, and the resulting equations are displayed over the graph in Figure 3-

29. Rsq values for the curves were 0.939 for the hybrid cell and 0.950 for the P3HT cell, showing

a good fit for the data range.

J = Jo + a exp(-b t) (3-5)

The hybrid cell shows a half-life of 7.42 min, while the P3HT cell shows a half-life of

16.05 min. This short half-life was confirmed for other hybrid cells fabricated in the same data

set. It is interesting to note the difference in the initial and final values extrapolated from the

exponential decay curves. The hybrid cell curve is extrapolated to an initial value of Jsc = 1.53

mA/cm2, while the P3HT cell curve shows an initial value of 8.09 x 10-2 mA/cm2. The P3HT

cell, however, shows a Jsc approximately twice as high as that of the hybrid cell as time

approaches infinity: 5.80 x 103 mA/cm2 vs. 2.89 x 10-3 mA/cm2. These extrapolated values

cannot be taken as absolute truths due to the uncertain nature of curve-fitting, but the









experimental data in Figure 3-29 confirm that the short-circuit current density for the hybrid cell

is initially higher than that of the P3HT cell and drops to a lower value after approximately 60

min of exposure time. The cause of the accelerated decay for the hybrid cells as compared to the

pure P3HT cells is unknown.

The exposure times shown in Figure 3-29 were measured from the time the cells were

removed from a glove box after the back contact deposition, but this may not be an appropriate

measurement of the exposure time, as exposure occurred throughout the fabrication process.

After weighing out appropriate amounts of the polymer and nanocrystals, both of which were

stored in a glove box under nitrogen, these materials were removed from the glove box in a vial

where the solvent mixture was added under exposure to room air. Solvents were not stored in

the glove box to prevent contamination of that environment. The solutions were sealed in their

vials to limit further exposure during mixing, but the vial contained room air rather than nitrogen

at this point. Spin-coating occurred in a cleanroom environment to limit particle contamination,

so the air was filtered and humidity-controlled to some degree, but not inert. After the films

were dried under vacuum, they were again opened to the cleanroom atmosphere where the films

were wiped clean in the electrode contact areas. At this point, the cells entered the glove box for

electrode deposition. From the time that the solid P3HT was removed from the nitrogen

environment to the time that the J-V curves of finished cells were measured, the total amount of

air exposure could vary from 75 min to over 2 hr during the fabrication process. A large part of

this exposure occurs while the P3HT is in solution, as the mixing time was typically at least 1 hr.

A second set of cells fabricated from these solutions yielded improved performance for the

hybrid cell. For these cells, care was taken to minimize the amount of environmental exposure

of the cells and films prior to testing. Approximately 15 min elapsed from when the finished









cells were removed from the glove box and when J-V testing was performed. The best hybrid

cell in this set showed performance measures of Jsc = 0.205 mA/cm2, Voc = 0.705 V, FF =

0.288, and efficiency = 0.0416 %. The top performing P3HT cell resulted in Jsc = 5.26 x 10-2

mA/cm2, Voc = 0.237 V, FF = 0.242, and efficiency = 3 x 103 %. The J-V curves for these cells

are shown in Figure 3-30.

The hybrid cell shown in Figure 3-30 displays a 300% improvement in short-circuit current

density and a 24% improvement in fill factor when compared to the hybrid cell from Figure 91.

The open-circuit voltage remained nearly constant, resulting in a nearly 140% improvement in

power conversion efficiency. The P3HT cell shown in Figure 3-30 shows a diode with a lower

rectification ratio as compared to the one shown in Figure 3-28. This cell showed a nearly 160%

increase in short-circuit current density, but a 55% reduction in the open-circuit voltage resulted

in an approximately 20% reduction in fill factor. Despite these changes, the maximum efficiency

improved by 140%.

These cells saw approximately 15 min of air exposure between removal from the glove box

and J-V measurement. Using the short-circuit current density vs. exposure time curves shown in

Figure 3-29, the predicted exposure times for these cells would have been 21 min for the hybrid

and 10 min for the polymer. This is within about 5 min of the actual exposure time, which is a

reasonable prediction considering this curve does not take into account exposure time accrued

during the fabrication process.

Chlorobenzene solvent

Based on comparisons of hybrid films deposited from various solvents, chlorobenzene was

found to be a solvent that granted good morphology with low surface roughness. To test the

performance of hybrid bulk heterojunction cells deposited from chlorobenzne, a 30 mg/ml

solution was generated with a 1:1 mixture of P3HT and pyridine-coated CdSe dissolved in a









mixture of 98% chlorobenzene with 2% pyridine. The best cell generated from this solution

showed maximum performance of Jsc = 0.138 mA/cm2, Voc = 0.391 V, FF = 0.292, and

efficiency = 0.0158%. Dark and illuminated J-V curves for this cell are shown in Figure 3-31.

This cell showed a lower short-circuit current current and open-circuit voltage than the similar

cell deposited from a 25 mg/ml hybrid solution in a chloroform:pyridine solution.

Commercial CdSe nanocrystals

To test the quality of the synthesized CdSe nanocrystals used for cell fabrication,

commercial CdSe nanopowder was acquired from Meliorum Technologies, Inc. [93] and used

for the fabrication of solar cells. The particles were 5 nm in diameter, just as the particles

synthesized in-house. The hybrid solution using these particles was 25 mg/ml using a 1:1 by

weight mixture of the commercial CdSe and P3HT in a solvent of chloroform with 2% pyridine

by volume. The best cell fabricated from this solution showed the following performance

characteristics: Jsc = 9.44 x 10-2 mA/cm2, Voc = 0.329 V, FF = 0.247, and efficiency = 7.65 x 10

3 %. Dark and illuminated J-V curves for this cell are shown in Figure 3-32. These curves are

inferior to those fabricated using in-house CdSe nanocrystals. The dark J-V curve shows a

rectification ratio of only 0.98 at + 1 V. Although the short-circuit current is lags only the best-

performing hybrid cell, the open-circuit voltage was less than 0.4 V and led to the low efficiency.

The terminating group on these nanocrystals is unknown, as the company refused to give up this

information. No attempts were made to alter the crystals; all cells were fabricated with these

particles as they were received.

Hybrid cell performance summary

Selected J-V curves for hybrid bulk heterojunction solar cells are shown in Figure 3-33.

These curves are all shown in previous figures, but are presented here on one graph for

comparison. A compilation of the J-V data for hybrid cells shown in this section is shown in









Table 3-9. The cell fabricated from a 25 mg/ml solution of equal weights P3HT and CdSe

dissolved in 2% pyridine in chloroform with minimal air exposure showed the highest

performance by simultaneously demonstrating the highest short-circuit current density and open-

circuit voltage of all cells measured.

The top-performing bi-layer and hybrid cell J-V curves are plotted together in Figure 3-34.

The current flow through the bi-layer cell is an order of magnitude higher than that of the hybrid

cell, demonstrating the need for improved morphology control in the hybrid active layers. The

bi-layer cell shows a low Voc compared to that of the hybrid cell.

The series and shunt resistances of these cells were estimated from the J-V curves using

Equation 3-6 and Equation 3-7, respectively [62]. This calculation is good for shunt resistance,

but series resistance is more accurately calculated as the applied voltage approaches infinity. For

the bi-layer cell, J-V data was not sufficiently collected to make this calculation. From these

equations, resistances for the bi-layer cell were calculated as Rs = 1.59 x 103 Q and Rsh = 2.25 x

103 Q. For the hybrid cell, resistances were calculated as Rs = 4.5 x 104 Q and Rsh = 8.33 x 104

Q. The series resistance for the hybrid cell calculated at the maximum measured voltage point

was 1.5 x 10-2 Q. For another bi-layer cell with a more extensive set of J-V data, the series

resistance is calculated as 1.8 x 102 Q.


R (3-6)
s dl )I=)


Rh(d- (3-7)


These resistance calculations show shunt resistances approximately 5 orders of magnitude

higher than the series resistance, which should result in high-quality solar cells. Further

improvements in cell design to maximize absorption and charge separation in the hybrid cells









should result in further reduction of the series resistance and drastic improvements in current

flow.

Particle Induced Nanostructuring

A key challenge in hybrid bulk heterojunction solar cells is control of the nanocrystal

distribution throughout the active layer. The film must be well-mixed to allow efficient exciton

dissociation, but also provide percolation pathways to provide charge collection pathways. A

new concept developed to resolve this issue is particle induced nanostructuring (PIN). The PIN

concept involves depositing multiple ultra-thin layers to control nanocrystal distribution

throughout the thickness of the film.

In this study, ITO-coated glass substrates were chemically cleaned and treated with

nitrogen plasma before film deposition. Multiple layers were deposited from a weakly

concentrated 5 mg/ml solution of 60 wt. % CdSe and 40 wt. % P3HT in chloroform with 2%

pyridine by volume. Deposition occurred at a spin-coating speed of 5000 rpm to produce very

thin films. The thicknesses of the films were measured with a profilometer after removing a

portion of the film to create a step.

Film thickness measurements on samples with 1 to 6 film layers showed an interesting

phenomenon. As expected, the film thickness grew with the addition of multiple layers. The

film thickness increased, however, only after two deposition steps were performed, as shown in

Figure 3-35. On the first deposition, a film of approximately 10 nm was deposited on the

substrate. After the second, the film thickness remained constant. On the third, another film of

approximately 10 nm was deposited, followed by another spin-coating resulting in no film

growth. After the next deposition step, the film grows by another 10 nm, with the 6th resulting in

a minimal amount of growth.









The dotted line drawn through the data in Figure 3-35 is the best fit line for film thickness

vs. number of layers, which follows Equation 3-8. With this trend, to reach a hybrid film

thickness of 100 nm, 21 layers would be required. In the fabrication setup used for these

experiments where film deposition equipment is open to atmosphere, this would result in a

prohibitive amount of atmospheric exposure time for the P3HT. For deposition in a nitrogen or

argon environment like a glove box, however, this process is feasible.

FilmThickness = 4.903 (# Layers) (3-8)

The film surface roughness was measured with AFM, and the surface was visualized with

SEM and optical microscopes. Figure 3-36 displays images of the results of these measurements

at 1, 3, 5, and 7 layers. The films analyzed by SEM and AFM were different from the films

analyzed with profilometry due to a difference in the sample size required for these techniques.

All SEM images are shown at the same scale (at 30,000x magnification), and all AFM images

show a 3 x 3 [tm scan area with a 100 nm scale on the z-axis.

Optical microscope images of the films are shown in Figure 3-37. These images are taken

at 100x magnification and are representative of the multiple images taken of these surfaces.

The rms surface roughness was measured from the AFM images for 3 x 3 [tm and 1 x 1 tm

areas on the surface. The results are shown in Figure 3-38. Additionally, lines are fit to

determine the trends for roughness as more layers are deposited.

The trendlines follow Equation 3-9 for the large area measurement and Equation 3-10 for the

small area measurement. If these trendlines are extrapolated to the required 21 layers for a 100

nm active layer film, the rms surface roughness is predicted to be 39.9 nm for a 5 x 5 tm area

and 15.0 nm for a 1 x 1 [tm area. From previous measurements of surface roughness by

profilometry and AFM, it was found that profilometer roughness measurements were









approximately 3.4x higher than 5 x 5 [tm AFM measurements and approximately 14x higher than

1 x 1 am AFM measurements. By this extrapolation, the rms surface roughness of a 21 layer

PIN film will be in the range 136 210 nm. The hybrid film surface profile shown in Figure 3-

27 displayed an rms roughness of 239 nm when measured by profilometry. This comparison

relies on a large amount of extrapolation, but implies that this technique is capable of producing

films with surface roughness as low or lower than similarly thick films deposited from

concentrated solutions in a single deposition step.

Rws = 1.20 (# Layers) + 14.65 (3-9)

Rws = 0.26 (#Layers) + 9.60 (3-10)

Cells were not fabricated using this technique, as lifetime measurements on other cells

indicated that the required air exposure time would cripple the cells before their performance

could be evaluated. Once the cell fabrication process line is housed under an inert environment,

this limitation no longer applies and cells consisting of multiple PIN layers can be compared to

single-layer cells to quantitatively determine the applicability of this technique.












Table 3-1. Chemical composition of ITO films
Con n Atomic Percent Atomic Ratio
Condition
O Sn In N O/In In/Sn
No Treatment 56.75 2.65 40.6 1.397 15.32
N2 Plasmaa 55.49 2.67 40.49 1.35 1.37 15.16
e-beamb 56.67 2.38 38.76 2.19 1.462 16.28
a After N2 plasma treatment at 50 W, 1 Torr, 10 min
b After e-beam treatment at 2kGy


-0.4 -0.2 0.0 0.2 0.4 0.6 0.8


Voltage (V)

Figure 3-1. J-V curves for organic solar cells on treated ITO substrates

Table 3-2. Performance of organic solar cells on treated ITO substrates
Treatment Voc (V) Jsc (mA/cm2) rl (%)
None 0.52 1.61 0.36
N2 Plasma 0.46 2.73 0.62
E-Beam 0.38 2.08 0.36
02 Plasma 0.32 1.82 0.27









ITO
4.8
5.3
PEDOT:PSS


4.0


P3HT

6.0


4.5


6.2
6.2


Al


Figure 3-2. Energy band diagram for bi-layer organic solar cells. All energy levels are listed in
units of eV.


Figure 3-3. Steps for bi-layer solar cell fabrication. A) ITO-coated glass substrate. B) Patterned
ITO anode. C) After PEDOT:PSS spin-coating. D) After P3HT spin-coating. E) After
polishing. F) After C60 evaporation. G) After Al evaporation. H) Encapsulated solar
cell.


ni~L














4000





3000-


(,
rS
S2000

1--



1000





0
1000 2000 3000 4000 5000

Spin Speed (rpm)


Figure 3-4. PEDOT:PSS film thickness vs. spin-coater speed.


600



500



i 400

(,

S300
-c
I-
E
'- 200
LL


1000 2000 3000

Spin Speed (rpm)


Figure 3-5. P3HT film thickness vs. spin-coating speed.


4000







100


5D


1 D-
41i


P3HT




PHTF i P3HT in Chlorobenzene
P T So I j io


W a ve n u n b e r ( m -1 )
Figure 3-6. Upper Spectrum: FTIR spectrum for chlorobenzene from Sigma Aldrich. Lower
Spectrum: P3HT film (purple) spin-coated from P3HT in chlorobenzene solution
(red).


V~Chlorobenzene n _


4 0 O 0


20


P F..





2 5 0 0


P E D 0 T P S S F im
P E D 0 T P S S o luto n

2 0 00 15 00 1 0 0 0


3



E


W a e n u m b er (cm )
Figure 3-7. Upper spectrum: FTIR spectrum of water from Sigma Aldrich. Lower spectrum:
PEDOT:PSS film (gray) spin-coated from solution of PEDOT:PSS in water (blue).


55


WAPEDOT:PSS
PEDOT:PSS


PEDOT:PSS in Water


3 0 AI-
4 000


3 M 3 2 M M = i 1 1M 12 I a
-- - --- -.----- ."
S11f 1
:: t/

,,-
i /"-


ii
0 -




































-0.2 -0.1 0.0 0.1 0.2 0.3


Voltage (V)


1000


100


0.001 ----- 8-3 Illu
-- 8-4 Illu
0.0001
-1.0 -0.5 0.0 0.5


Voltage (V)
Figure 3-8. Linear (A) and base 10 log-scale (B) J-V curves for bi-layer organic solar cells
fabricated with a single 80 nm thick layer of PEDOT:PSS.



































-0.2 -0.1 0.0 0.1 0.2 0.3


Voltage (V)


1000


100


0.01


0.001


-1.0 -0.5 0.0 0.5


Voltage (V)
Figure 3-9. Linear (a) and base 10 log-scale (b) J-V curves for bi-layer organic solar cells
fabricated with two 40 nm thick layers of PEDOT:PSS.












3000


2500 -


" 2000 -

(,
C,
- 1500
--
I--
E
1000
-L


2000


3000


4000


5000


Spin Speed (rpm)


Figure 3-10. Calibration curves for slow- and fast-filtered PEDOT:PSS. Slow-filtered
PEDOT:PSS is shown in the "051116" data set. Fast-filtered PEDOT:PSS is shown
in data sets "051122", "051123", and "051130".



A B C











D LE F D.2







I0 110.


Figure 3-11. AFM images of P3HT films. Images A C were spin-cast from 5 mg/ml P3HT in
chlorobenzene solutions, while images D F were spin-cast from 10 mg/ml P3HT in
chlorobenzene solutions. Images A and D were spin-cast at 2000 rpm, images B and
E at 3000 rpm, and images C and F at 4000 rpm. For all images, the scale bar for the
film height axis is 50 nm, and the scan area is 1 [tm x 1 [tm.


051116
V 051122
051123
051130









* V
*










Table 3-3. RMS surface roughness of P3HT films shown in Figure 3-11.
Label Solution Spin Speed RMS Roughness
A 5 mg/ml 2000 rpm 0.89 nm
B 5 mg/ml 3000 rpm 1.27 nm
C 5 mg/ml 4000 rpm 0.94 nm
D 10 mg/ml 2000 rpm 3.66 nm
E 10 mg/ml 3000 rpm 2.59 nm
F 10 mg/ml 4000 rpm 2.54 nm


/
/
/


0 -- -"










-4 I I I
-0.2 -0.1 0.0 0.1 0.2 0.3

Voltage (V)

Figure 3-12. J-V curves for bi-layer solar cells fabricated on untreated ITO substrates.


0--- Set I dark
........ v ........ S et I 1
Set I 1
--- --- Set 1-2
et I 3
Set I 4



































-0.1 0.0 0.1 0.2 0.3


Voltage (V)


-0.4 -0.2 0.0 0.2 0.4


Figure 3-13. J-V


Voltage (V)
curves in the dark (A and B) and under 100 mW/cm2 illumination (C and D).


10



1



0.1



0.01


0.001



































-0.2 -0.1 0.0 0.1 0.2 0.3


Voltage (V)


0.001


-0.4 -0.2 0.0 0.2 0.4


Voltage (V)
Figure 3-13 continued. J-V curves under 100 mW/cm2 illumination (C and D).









Table 3-4. J-V data for bi-layer solar cells.
Sample RR Jsc (mA/cm2) Voc (V) FF 1 (%)
I-1 2.4 0.55 0.16 0.23 0.02
I-2 1.03 0.15 0.19 0.03
1-3 1.55 0.11 0.23 0.04
I-4 0.41 0.16 0.15 0.01
II-1 8.53 1.10 0.08 0.24 0.02
II-2 3.96 0.68 0.07 0.22 0.01
II-3 1.81 0.70 0.06 0.19 0.01
II-4 0.88 0.07 0.21 0.01


Table 3-5. Solvents considered for hybrid bulk heterojunction film deposition.
Solvent B.T. (C) Polarity P3HT Solubility
Acetone 56.2 5.1 No
2-butanol 79.6 4.0 No
DMF 153 6.4 No
Methanol 64.6 5.1 No
2-propanol 82.4 3.9 No
MEK 80.0 4.7 Poor
Pyridine 115.3 5.3 Poor
Benzene 80.1 2.7 Yes
Chlorobenzene 131.7 2.7 Yes
Chloroform 61.2 4.1 Yes
o-dichlorobenzene 180 2.7 Yes
THF 66.0 4.0 Yes
Toluene 110.6 2.4 Yes
TCE 87.2 1.0 Yes
o-xylene 144.4 2.5 Yes















* THF
* Chlorobenzene
o-dichlorobenzene
Chloroform
* Toluene
* o-xylene
TCE
Benzene


40 60 80 100 120 140 160 180 200 220

Solvent Boiling Temperature C


.5 1.0 1.5 2.0


2.5 3.0 3.5 4.0


Solvent Polarity
Figure 3-14. Surface roughness measurements by profilometry for hybrid films deposited from
various solvents.


300
E
C,
C,
c,

- 200
o
0
aI

100




0


I


* THF
* Chlorobenzene
o-dichlorobenzene
Chloroform
* Toluene
* o-xylene
TCE
Benzene


400





300
E
(D
c
v,



0
o
o

(1)
c
r
100 -





0
0


6I





























40 60 80 100


120 140


160 180


Solvent Boiling Temperature (OC)


'*1


0.5 1.0 1.5 2.0 2.5 3.0 3.5


4.0 4.5


Solvent Polarity
Figure 3-15. RMS surface roughness for 5 x 5 pm surface area samples measured with AFM.


* THF
* Chlorobenzene
o-dichlorobenzene
Chloroform
* Toluene
* o-xylene


* THF
* Chlorobenzene
o-dichlorobenzene
Chloroform
* Toluene
* o-xylene


4' I































40 60 80 100
40 60 80 100


140 16 1
140 160 180


Solvent Boiling Temperature (oC)


1.0 1.5


3.5 4.0


Solvent Polarity
Figure 3-16. RMS surface roughness for 1 x 1 pm surface area samples measured with AFM.


* THF
* Chlorobenzene
o-dichlorobenzene
Chloroform
* Toluene
* o-xylene


* THF
* Chlorobenzene
o-dichlorobenzene
Chloroform
* Toluene
* o-xylene


t


El
I


4

2

0 -
0.5








Table 3-6. Mean rms surface roughness in nm for hybrid films deposited from selected solvents
Solvent Profilometer AFM (5 x 5 m) AFM (1 x 1 pm)
THF 136 36.8 15.0
Chlorobenzene 64.9 25.0 8.42
o-DCB 81.2 21.1 9.58
Chloroform 71.8 16.2 3.22
Toluene 66.2 27.4 2.93
Xylene 198 16.8 7.76
TCE 72.0 Not measured Not measured
Benzene 73.9 Not measured Not measured


I ._-"_ ___ "__ "__ ___ ._'_ __ _____ .________. ..,
Figure 3-17. Optical microscope images of selected films. The first column displays images
taken at 10x magnification, while the second column displays images taken at 100x
magnification.



































* .I" .1 q ^ tfk
IL ...., -W




.:: "' .*
.4 .4


00 "o S"
:o. '. -,." -; > .'
*. :'
'- .u.& "" ." ""
'-" ,'-" ,. ,*'*,: ".'
.. ...


v: "
t.
*


S
4.


'.4'


a'


a
lb


9i


I.
S. **


'.,


Figure 3-17 continued. Optical microscope images of selected films.


10xI

































lox I


7M -


p-r ,* I


I S


F 3 -1s se t





Figure 3-17 continued. Optical microscope images of selected films.


al e


.l .


r




b.. L*P































































Figure 3-18. SEM images of selected hybrid films. The first column image was taken at 5,000x
magnification and the second column image was taken at 15,000 x magnification.


N HD2 1 5 0 k V X 5 0 6 k' C; 6'0'A" ri'l
















N IA D 2 1 5 0 k V X 5 0 C;
















N-HD2 1 5 0 k V X 5 0 6 k' E;


N 1-1 D 2 1 5 0 k V X 1 5
















N-IAD2 15.OkV X15.OK 2.00pro








C







N HD2 1 5 Ok V X 1 5 2: 6'0'r









































Figure 3-18 continued. SEM images of selected hybrid films.


tl











N PID2 1 5 0 k V X 5 0 C; i3'0'p' M'














N IAD2 1 5 0 k V X 5 o 6 W 'c; We'


N-HD2 15.OkV X15.OK 200prn














N-IAD2 15.OkV X15.OK 2.eOpm










A)

50



S40



30

0

20

Cn

S10 -



0
0 10 20 30 40 50 60
Pyridin e Percentage in Solvent (%)
B)

12



S10



,r-
-)
r 2 8


Co
6U 6








2
O 2 4 6 8 10 12

Pyridinre Percentage in Solvent(%J
Figure 3-19. Film surface roughness vs. pyridine concentration in the chloroform solvent for
TOPO-coated CdSe nanocrystals (A) and pyridine-coated CdSe nanocrystals (B).



175








Table 3-7. Hybrid solutions of P3HT and TOPO-coated CdSe nanocrystals in a mixed solvent of
chloroform and pyridine.
Solution P3HT Wt. CdSe Wt. Chloroform Vol. Pyridine Vol. Solution Concentration
1 26.8 mg 35.5 mg 1.5 ml 1.5 ml 20.4 mg/ml
2 24.8 mg 23.2 mg 1.5ml 1.5 ml 19 mg/ml
3 15.0 mg 25.0 mg 1.5 ml 1.5 ml 5 mg/ml

S. 'i
















A)






Figure 3-20. Optical microscope images of hybrid films deposited from 20.4 mg/ml (A), 19
mg/ml (B) and 5 mg/ml (C) solutions. Images were taken at 10Ox magnification and
the films were deposited at 3000 rpm and dried under vacuum at 120 C.
r i"



~ ~ "1::" :.C !,: :'! ":..






Figure~ ~ ~ ~~ ~~~~~" 3-20 Optic. .irscp imge :" hyri .:"m .. oste frm 04 gml(
mg/ml! .......g/l(C sltin Imgswr akna 0xmgnfcto n























Figure 3-21. Optical microscope image of 19 mg/ml hybrid film deposited at 3000 rpm and
subjected to a pure solvent spin-coating step at 8000 rpm. The image was taken at
100x magnification.














A) B













C)
Figure 3-22. Optical microscope images of P3HT films deposited from 5 mg/ml solutions in
chloroform (A), 1:1 chloroform:pyridine (B), and pyridine (C). All images are taken
at 100x magnification.

















O
























Figure 3-23 continued. SEM images of P3HT films deposited from pure chloroform and an equal
mixture of chloroform and pyridine solvents. Images are shown at 2,000x and 10,000x
for each film.
-5

rC























for each film.


2,000 x


10,000 x









A)
100


80

P3HT Film
C 60
Co

S40 ITO
.- 40 -

I--c t- L
20------------ - -


0

0.0 0.5 1.0 1.5 2.0 2.5
Lateral Distance (mm)
Figure 3-24. Surface profiles of P3HT films deposited from chloroform (A), 1:1
chloroform:pyridine (B), and pyridine (C) solvents. In the graphs, the bare substrate
appears on the left hand side with the film on the right hand side beyond the wide
peak. The solid teal line represents the median film thickness, the red dash-dot line
represents the mean film thickness, and the blue dashed lines represent + one standard
deviation from the mean.































0.5 1.0 1.5 2.0


Lateral Distance mm)


0.5 1.0 1.5 2.0


Lateral Distance (mm)
Figure 3-24 continued. Surface profiles of P3HT films deposited from chloroform (A), 1:1
chloroform:pyridine (B), and pyridine (C) solvents.






180










Table 3-8. Film properties for P3HT films deposited from various solvents.
Solvent Mean Thickness (nm) Median Thickness (nm) Standard Deviation (nm)
Chloroform 21.2 + 8.18 23.1 + 7.95 9.58 + 5.41
Mixed 56.2 34.3 16.2 4.42 227 157
Pyridine 78.5 85.4 10.9 8.61 338 457


-0.4 4-
-1.5


-1.0 -0.5 0.0 0.5 1.0


Voltage (V)


Figure 3-25. Dark and illuminated J-V curves for cells generated from 5 mg/ml composite
solutions in (A) chloroform mixed with 2% pyridine and (B) pure chloroform. Note
the scale difference of the graphs.













4.0e-4

2.0e-4 -

0.0 -


-2.0e-4

-4.0e-4 -

-6.0e-4 -

-8.0e-4 -


-*- Dark
IlluImininaedj


-1.0e-3

_1 I


-02 -0.1 00 01 02 03 04 05


-1.5 -1.0 -0.5 0.0 0.5 1.0


Voltage (V)
Figure 3-25 continued. Dark and illuminated J-V curves for cells generated from 5 mg/ml
composite solution in pure chloroform.


-1.5 -1.0 -0.5 0.0 0.5 1.0


Voltage (V)
Figure 3-26. Dark and illuminated J-V curves for 10 mg/ml hybrid solution deposited with low-
speed spin-coating.


~----~"El


- I. -














500


400 -


300 -


200 -


100 -


0


-100
0.


0


1.0 1.5

Lateral Distance (mm)


0.5 1.0 1.5 2.0 2.5


Lateral Distance (mm)

Figure 3-27. Surface profiles of film deposited from (A) 12 mg/ml P3HT and (B) 25 mg/ml
hybrid solutions in 2% pyridine in chloroform.


P3HT Film










ITO HT Film

















-0.8 -
-0.05

0.00 u
E -0.6 0
o 0.05 .

E -0.4 0.10 O

0.15
c -0.2 020
a, 0 ,5,5 .0 .0 I .0 6 .,io

0.0

E0 H-brid Ciark
0.2 O I H bnri Illuin ated
-B- P. HT Dark
0.4 I F'-.HT IIIiminn ated


1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5

Voltage (V)
Figure 3-28. Dark and illuminated J-V curves for hybrid bulk heterojunction solar cell deposited
from 25 mg/ml solution and P3HT polymer cell deposited from 12 mg/ml solution.
The insert shows a zoom-in on the active area of the cells.












0.06

(E Hybrid Cell
5 0.05 P3HT Cell

E Decay- Hybrid
4 Decay- P3HT
0.04 \ -------

S\ Hybrid Cell
c 0.03 J= 2.894E-3 + 1.525 exp(-9.369E-2* t)

S\ \ P3HT Cell
0.02 \ J = 5.804E-3 + 7.506E-2 exp(-4.820E-2* t)
0

o 0.01

U) t -----f-I
S-- -- -- -- -

0.00 ---
0 50 100 150 200 250 300

Exposure Time (min)
Figure 3-29. Short-circuit current decay for hybrid (gold circles) and P3HT (green squares).

















-1.5 0o.o00 7 0 1

o o0.05 0 -
E /
S-1.0 010

E 0.15- o 0
.t: 0.20 o
S-05 0
( 0.25 1 0
Q 0.2 0.0 -0.2 -0.4 -0.6 -0.. O



CO -- Hybrid Dark
0.5 O-.^ 0 Hybrid Illuminated
-8- P3HT Dark
P3HT- Illuminated
1.0
1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5

Voltage (V)
Figure 3-30. Dark and illuminated J-V curves for hybrid bulk heterojunction and pure P3HT
solar cells with limited air exposure during processing.































1.5 1.0 0.5 0.0 -0.5 -1.0


Voltage (V)
Figure 3-31. Dark and illuminated J-V curves for hybrid solar cell fabricated from
chlorobenzene with 2% pyridine solution. Red curves represent illuminated J-V and
black curves represent dark J-V.













1.5

0.4

1.0 0.2

E O0 ....*****

-.-2 02
E 0.5 -
>. -0.4

-0.2 0.0 0.2 0.4 106 0.8 1.0
0.0



-0.5




-1.0
-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5

Voltage (V)


Figure 3-32. Dark and illuminated J-V curves for a hybrid solar cell fabricated with commercial
CdSe nanopowder. Green curves represent illuminated J-V and black curves
represent dark J-V.


































-1.0 -0.5 0.0 0.5


Figure 3-33.


Voltage (V)
Illuminated J-V curves for hybrid bulk heterojunction solar cells with various


fabrication conditions.


-3 +L
-0.2


0.0 0.2 0.4 0.6 0.8


Voltage (V)


Figure 3-34. J-V curves for the best bi-layer and hybrid cells shown in this dissertation.


189









Table 3-9. Hybrid solar cell fabrication information and performance data.
SSolution CdSe wt.% %Pyridine in (1) Vo() (2) Notes
Figure S C 0 0 Jsc Voc (V) FF 1 Notes
SConcentration in P3HT Chloroform
3-25 5 mg/ml 60% 2% 8.99 0.338 0.230 0.7
3-25 5 mg/ml 60% 0% 0.026 0.194 0.252 0.001 Pure Chloroform Solvent
3-26 10 mg/ml 60% 2% 33.9 0.256 0.275 2.4
3-28 25 mg/ml 50% 2% 49.9 0.701 0.232 17.5
3-28 12 mg/ml 0% 2% 20.3 0.523 0.310 12.5 Pure P3HT film
3-30 25 mg/ml 50% 2% 205 0.705 0.288 41.6 Minimized air exposure
Pure P3HT film
3-30 12 mg/ml 0% 2% 52.6 0.237 0.242 3.0 ureHTlm
Minimized air exposure
3-31 30 mg/ml 50% 2%(3) 138 0.391 0.292 15.8 Pyridine in Chlorobenzene Solvent
3-32 25 mg/ml 50% 2% 94.4 0.329 0.247 7.65 Commercial CdSe
(1) Jsc displayed in units of aA/cm2 (mA/cm2 x 10-3)
(2) Efficiency displayed in units of % x 10-3
(3) This solvent is 2% pyridine in chlorobenzene












40
/



30 -
E /




10



10





0 1 2 3 4 5 6 7

Number of Layers

Figure 3-35. Film thickness for multi-layer hybrid films.

# SEM Image AFM Image














Figure 3-36. SEM and AFM surface images of multi-layer hybrid films.









191




















AFM Image

lli l n ntr i ONplS 2O6

I.." Dt a Hl ivht
la PD* .1. a* Il 1If upi.


blt tnjrlt
light tmlt


2X I.00 iI diV
2 100.000 -wAi.


O des


Disital Intrt>nts rl:uo :'
Datn si. U 4w0 -
Scm rtel o 40 I z
HI a,,or P uplt 24S,
loo.e Dit. H11i0nl
DI u tso. srjI i-,








1 ulest ansl*
^^ "1 -* light ansle


X 1.l II NWdi
2 100.00 nt/tiW


0 d"s


Disitol Intrzotrrts Hnwo[os
Set. rale 0.4002 Hz
HuIsr Por nupl 25IS
Duta sarls 100,0 .


beit atVlu
light Wantg


2 1.000i MM/iua 0 d
2 100.50 /CIt^V


Figure 3-36 continued. SEM and AFM surface images of multi-layer hybrid films.


32.000


F,


43%.000
































1 ey


Figure 3-37. Optical microscope images for multi-layer hybrid films. The number of film layers
were a) 1, b) 2, c) 3, d) 4, e) 5, and f) 6.














30
E
V)



20-


15

10 -
10 ------------------





0 1 2 3 4 5 6 7

Number of Layers

Figure 3-38. RMS surface roughness for multi-layer hybrid films. Red squares represent
measurements over a 3 x 3 tm area and blue circles represent measurements on a 1 x
1 tm area.


























194









CHAPTER 4
CONCLUSIONS AND FUTURE WORK

Conclusions

This dissertation presented the results of exploratory research on the processing and

performance of hybrid PV. It provided encouragement for a more complete study of hybrid

photovoltaic devices. With collaboration and assistance from several teams and individuals, the

groundwork was laid for future studies to continue this project and achieve high-performance

hybrid photovoltaic devices.

The innovations of this study include the first simulations of hybrid photovoltaics using

existing semiconductor modeling software, development of anode surface treatment processes,

solvent selection for hybrid films, and hybrid bulk heterojunction photovoltaic process

development, including an interesting multiple spjn coating process sequence to better disperse

the inorganic phase.

Hybrid Photovoltaic Simulation

Simulations of an ordered heterojunction photovoltaic cell provided interesting results. It

was found that reported values of certain parameters reported in the literature from organic field-

effect transistor fabrication were poor estimates for organic photovoltaic simulation.

The hole mobility values given in the literature [71] proved to be higher than the

simulations estimated. This value of 0.01 cm2/V-s for the hole mobility produced J-V curves

with fill factors around 0.85, which is considerably larger than published values which typically

range from 0.4 to 0.6 [27, 33-34, 44, 75]. By reducing the mobility to 1 x 10-4 cm2/V-s the fill

factor was reduced to 0.78, which is closer to the published range.

The short-circuit current density of the real cell could not be matched by the simulations.

All attempts using the two-step model gave values lower than the observed ones, including









simulations that allowed the full P3HT region to generate photocurrent and a 50% increase in the

absorption coefficient over a 40 nm exciton diffusion length. The highest short-circuit current

density achieved with this simulation technique was 1.47 mA/cm2 from a cell featuring a 40 nm

LD region, but this value is approximately 0.8 mA/cm2 lower than the real cell. It is believed

that the two-step simulation process creates a strong attractive force between charges when very

high levels of charges are generated along a narrow line, resulting in increased annihilation

between the free carriers.

Anode Surface Treatment

Treatment of the ITO substrate with a N2 plasma was optimized to grant a smooth, stable

surface, which served as the basis of cell fabrication. This work was done in close collaboration

with Jiyoun Seol at Yeungnam University. The chemically cleaned substrate was subjected to

nitrogen plasma at 50 W and 200 mTorr for 30 min. This process resulted in a smoother, more

hydrophilic surface which aided deposition of further layers, and an increase in nitrogen content

which lowered the film work function and eased the pathway for hole current. This treatment

procedure was used for all hybrid film and hybrid photovoltaic studies that followed to ensure a

stable surface and allow direct comparison of the results.

Solvent Selection

The solvents TCE, THF, benzene, toluene, o-xylene, chlorobenzene, o-dichlorobenzene,

and chloroform were compared for their applicability to hybrid film deposition. Although no

correlations were found between film properties and solvent polarity or solvent boiling

temperature, certain solvents were identified as strong candidates for these films. These included

chloroform, and chlorobenzene, which were each used in further studies. These solvents

produced films with low surface roughness at all measurement scales, from microns to









millimeters. These solvents provided simultaneous dissolution both the organic and inorganic

phases which limited early precipitation of the solutes, resulting in a more uniform film surface.

Hybrid Bulk Heterojunction Photovoltaic Development

Hybrid photovoltaic cells were fabricated with varying film properties. It was found that

the best cells featured thick absorber films spin-coated from solutions of at least 25 mg/ml with a

50:50 weight ratio between P3HT and the nano-CdSe. Cells consisting of pure P3HT sandwiched

between the PEDOT:PSS and Al layers showed maximum performance approximately an order

of magnitude lower than the best hybrid cells.

Cells fabricated from commercial CdSe nanopowder performed at less than 20% of the

best cells using in-house CdSe nanopowder coated with pyridine, although a direct comparison

between the two is difficult because the surface passivation material remains unknown for the

commercial nanopowder.

Hybrid cells fabricated with chlorobenzene solution rather than chloroform saw a 60%

reduction is efficiency despite an increase in solution concentration from 25 to 30 mg/ml.

Despite Jsc and fill factor values that were nearly the same as that of the chloroform cell, the

chlorobenzene cell showed a Voc of 0.3 V lower than the chloroform cell.

The best cell was fabricated from a 25 mg/ml solution of 50% CdSe in P3HT dissolved in

2% pyridine in chloroform. The process was tailored to minimize the amount of time the cell

was exposed to atmosphere during fabrication and between fabrication and testing. It showed a

maximum efficiency of 4.16 x 10-2 %, Jsc of 0.205 mA/cm2, Voc of 0.705 V, and a fill factor of

0.288.









Future Work

Although the field of organic photovoltaics is rapidly growing and advancing, several

research directions are suggested by this work for its continuation. This section describes some

potential research directions stemming from the work presented in this dissertation.

Organic Photovoltaic Simulations

The models presented in this dissertation used the powerful modeling software package

Medici. Although the software has powerful optical and electrical simulation abilities, this

research seems to have pushed its limits by demanding nano-scale material specifications and

low levels of free carriers, carrier mobilities, and current flow in the devices. Simulation

attempts were frequently cut short due to convergence issues in the software under the specified

conditions. Additionally, a key component of the modeling work focused on the correct way to

simulate the effect of excitons in Medici, which are not explicit in the package.

In light of these difficulties, a new software package designed to simulate organic

electronic materials would be a great boost to this work. A product such as Fluxim [94] would

be able to more accurately simulate the effect of excitons in these hybrid devices.

Another direction that should be pursued is the variation of the geometry and materials of

the simulated cells. While ZnO is a well-researched material for the growth of aligned and

ordered nanowires, these structures are now being grown for other materials with stronger

absorption spectra such as CdSe and InP [66, 95]. These materials could see more use in

nanowire hybrid cells in the future, driving the need for effective simulations to study device

properties.

The most common organic cell design is currently the bulk heterojunction cell using

semiconductor nanoparticles or soluble C60 derivatives. This design presents a challenge for

simulations because little work has been done to characterize the particle distributions in these









films. Because the physical dimensions of the model are vital for accurate simulation, this

provides both a limitation and an opportunity. On the one hand, this lack of information limits

the results that can be generated through simulations. On the other hand, if an accurate, robust

model can be developed for a hybrid material system, it could be used to back-calculate

unknown physical dimensions of the system.

Development of Hybrid Photovoltaic Cells

A central theme of this dissertation is the morphology control of hybrid films for

photovoltaic applications. This study only focused on nearly spherical nanocrystals, but future

studies should expand the field to include nanoparticles with dimensionality. Shaped

nanoparticles, nanowires, tetrapods, and more exotic branched structures are being grown using

techniques similar to the one used for spherical nanocrystals in this study [35-37]. These

dimensional crystals offer the promise of directed charge transport without the multiple electron

"jumping" processes required for small spherical nanocrystals.

Alternative semiconductor nanoparticles should be considered as well. Although CdSe is

one of the easiest particles to be synthesized, other compound semiconductors such as CdS,

CdTe, PbSe, and CIS can be grown on the nano scale. Some semiconductors such as PbSe have

demonstrated multi-carrier generation on short timescales that offer the possibility of

constructing photovoltaic devices with quantum efficiencies greater than 1 if these charges can

be harvested [96-97].

Regioregular P3HT, as used in this study, is currently the most promising candidate for

polymer in electronic devices such as organic thin-film transistors and solar cells [33-34].

However, this polymer shows some limitations for solar applications. The absorption spectrum

shows a cut-off above 650 nm, limiting absorption of near-IR photons which are plentiful in the

solar spectrum. Although the hole mobility of P3HT is high compared to many conductive









polymers, it is several orders of magnitude lower than most inorganic semiconductors and limits

charge collection in devices. Additionally, it degrades quickly under water and oxygen

atmospheres, particularly in the presence of radiation. As new polymers are developed with a

broader absorption spectrum, improved carrier mobility, improved environmental resistance, and

the ability to deposit high-quality films with strong adhesion and low surface roughness they will

quickly find relevance in organic photovoltaics research.

Regardless of the targeted future research direction, some processing equipment changes

should be considered. The recent installation of a photolithographic patterning system will allow

for the fabrication of multiple cells on single substrates, and this will greatly improve the speed

of research and also the quality of the devices fabricated. As long as P3HT is used for the active

layer film, all processing equipment should be set up under an inert atmosphere of argon or

nitrogen in a glove box. Because of the sensitivity of this polymer to water and oxygen, high-

quality device fabrication at this scale requires that it be protected from exposure at all phases of

the process. This is not true for the HTL layer of PEDOT:PSS, which is in fact deposited from a

water solution and shows no negative effects of short-term atmospheric exposure.

All processing steps beyond the deposition of the HTL should be contained in this glove

box, including mixing for the hybrid solution, deposition and drying of the hybrid films, possible

inclusion of exciton blocking layers, back electrode deposition, and cell characterization. This is

a difficult challenge due to the size of some processing equipment, but this is the design used by

groups fabricating world-record organic photovoltaic cells. The processing equipment line

currently in use already involves an evaporation chamber opening directly into the glove box,

which is the largest piece of equipment used in the process. In order to improve these cells to the

highest performance level, this environmental protection is a step that absolutely must be taken.


200









LIST OF REFERENCES

1. International Energy Agency, Trends in Photovoltaic Applications, Report IEA-PVPS
T1-16:2007, August 2007.

2. U.S. Department of Energy, Energy Efficiency and Renewable Energy,
http://wwwl.eere.energy.gov/solar/deployment.html (2006) July 2008.

3. Solarbuzz, LLC., www.solarbuzz.com/moduleprices.htm (2008) July 2008.

4. C.W. Tang, Appl. Phys. Lett. 48 (1986) 183.

5. Md.K.H. Bhuiyan, T. Mieno, Thin Solid Films 441 (2003) 187.

6. N.S. Sariciftci, L. Smilowitz, A.J. Heeger, F. Wudl, Synth. Met. 59 (1993) 333.

7. B. O'Regan, M. Gratzel, Nature 335 (1991) 737.

8. M.K. Nazeeruddin, P. Pechy, T. Renouard, S.M. Zakeeruddin, R. Humphry-Baker, P.
Comte, P. Liska, L. Cevey, E. Costa, V. Shklover, L. Spiccia, G.B. Deacon, C.A.
Bignozzi, M. Gratzel, J. Am. Chem. Soc. 123 (2001) 1613.

9. K. Hara, M. Kurashigo, Y. Dan-oh, C. Kasasa, Y. Ohga, A. Shinpo, S. Suga, K. Sayama,
H. Arakawa, New J. Chem. 27 (2003) 783.

10. Q.B. Meng, K. Takahashi, X.T. Zhang, I. Sutanto, T.N. Rao, O. Sato, A. Fujishima, H.
Watanabe, T. Nakamori, Uragami Langmuir 19 (2003) 3572.

11. L.J. Rothberg, M. Yan, F. Papadimitrakopoulos, M.E. Galvin, E.W. Kwock, T.M. Miller,
Synth. Met. 80 (1996) 41.

12. J.J.M. Halls, K. Pichler, R.H. Friend, S.C. Moratti, A.B. Holmes, Appl. Phys. Lett. 68
(1996)3120.

13. L.A. Pettersson, L.S. Roman, O. Inganas, J. Appl. Phys. 86 (1999) 487.

14. J.J.M. Halls, C.A. Walsh, N.C. Greenham, E.A. Marseglia, R.H. Friend, S.C. Moratti,
A.B. Holmes, Nature 376 (1995) 498.

15. T.J. Savenije, J.M. Warman, A. Goossens, Chem. Phys. Lett. 287 (1998) 148.

16. M. Theander, A. Yartsev, D. Zigmantas, V. Sundstrom, W. Mammo, M.R. Andersson, O.
Inganas, Phys. Rev., B 61 (2000) 12957.

17. H.L. Wong, L.S.M. Lam, K.W. Cheng, K.Y.K. Man, W.K.Chan, C.Y. Kwong, A.B.
Djurisic, App. Phys. Lett. 84 (2004) 2557.

18. G.D. Sharma, R. Kumar, S.K. Sharma, M.S. Roy, S. En. Mater. & Solar Cells 90 (2006)
933.









19. P. Peumans, S. Uchida, S.R. Forrest, Nature 425 (2003) 158.

20. R.P. Raffaelle, B.J. Landi, J.D. Harris, S.G. Bailey, A.F. Hepp, Mat. Sci. & Eng., B 116
(2005) 233.

21. A.D. Pasquier, H.E. Unalan, A. Kanwal, S. Miller, M. Chhowalla, App. Phys. Lett. 87
(2005) 203511.

22. N.S. Sariciftci, L. Smilowitz, A.J. Heeger, F. Wudl, Science 258 (1992) 1474.

23. N.S. Sariciftci, L. Smilowitz, A.J. Heeger, F. Wudl, Synth. Met. 59 (1993) 333.

24. G. Yu, K. Pakbaz, A.J. Heeger, Appl. Phys. Lett. 64 (1994) 3422.

25. J.C. Hummelen, B.W. Knight, F. LePeq, F. Wudl, J. Yao, C.L. Wilkins, J. Org. Chem. 60
(1995) 532.

26. G. Yu, J. Gao, J.C. Hummelen, F. Wudl, A.J. Heeger, Science 270 (1995) 1789.

27. S.E. Shaheen, C.J. Brabec, N.S. Sariciftci, F. Padinger, T. Fromherz, J.C. Hummelen,
Appl. Phys. Lett. 78 (2001) 841.

28. N. Camaioni, G. Ridolfi, G. Casalbore-Miceli, G. Possamai, L. Garlaschelli, M. Maggini,
Solar Energy Materials & Solar Cells 76 (2003) 107.

29. M. Al-Ibrahim, H.K. Roth, M. Schroedner, A. Konkin, U. Zhokhavets, G. Gobsch, P.
Scharff, S. Sensfuss, Organic Electronics 6 (2005) 65.

30. W. Geens, T. Aernouts, J. Poortmans, G. Hadziioannou, Thin Solid Films, 403 (2002)
438.

31. D. Gebeyehu, F. Padinger, C.J. Brabec, T. Fromherz, J.C. Hummelen, N.S. Sariciftci, Int.
J. ofPhotoenergy 1 (1999) 1.

32. A.J. Mozer, N.S. Sariciftci, C.R. Chimie 9 (2006) 568.

33. F. Padinger, R.S. Rittberger, N.S. Sariciftci, Adv. Funct. Mat. 13 (2003) 85.

34. G. Li, V. Shrotriya, J. Huang, Y. Yao, T. Moriarty, K. Emery, Y. Yang, Nat. Mat. 4
(2005) 864.

35. M. Chen, Yi Xie, J. Lu, Y. Xiong, S. Zhang, Y. Qian, X. Liu, J. Mat. Chem. 12 (2002)
748.

36. L. Manna, D.J. Million, A. Meisel, E.C. Scher, A.P. Alivisatos, Nat. Mat. 2 (2003) 382.

37. K.A. Dick, K. Deppert, M.W. Larsson, T. Martensson, W. Seifert, L.R. Wallenberg, L.
Samuelson, Nat. Mat. 3 (2004) 380.









38. W.J.E. Beek, L.H. Slooff, M.M. Wienk, J.M. Kroon, R.A.J. Janssen, Adv. Funct. Mater.
15 (2005) 1703.

39. P.A.C. Quist, L.H. Slooff, H. Donker, J.M. Kroon, T.J. Savenije, L.D.A. Siebbeles,
Superlattices and Microstructures 38 (2005) 308.

40. A. Watt, H. Rubinsztein-Dunlop, P. Meredith, Mat. Lett. 59 (2005) 3033.

41. X. Peng, J. Wickham, A.P. Alivisatos, J. Am. Chem. Soc. 120 (1998) 5343.

42. Z.A. Peng, X. Peng, J. Am. Chem. Soc. 123 (2001) 183.

43. M. Kuno, J.K. Lee, B.O. Dabbousi, F.V. Mikulec, M.G. Bawendi, J. Chem. Phys. 106
(1997) 9869.

44. W.U. Huynh, J.J. Dittmer, W.C. Libby, G.L. Whiting, A.P. Alivisatos, Adv. Funct.
Mater. 13 (2003) 73.

45. H.J. Choi, J.K. Yang, H.H. Park, Thin Solid Films 494 (2006) 207.

46. K.M. Coakley, M.D. McGehee, App. Phys. Lett. 83 (2003) 3380.

47. K.M. Coakley, Y. Lui, M.D. McGeehee, K.M. Frindell, G.D. Stucky, Adv. Funct. Mater.
13 (2003) 301.

48. J.J. Wu, S.C. Liu, Adv. Mater., Vol. 14 (2002) 215.

49. W. Lee, M.C. Jeong, J.M. Myoung, Acta Mater. 52 (2004) 3949.

50. C. Geng, Y. Jiang, Y. Yao, X. Meng, J.A. Zapien, C.S. Lee, Y. Lifshitz, S.T. Lee, Adv.
Funct. Mater. 14 (2004) 589.

51. S.C. Lyu, Y. Zhang, C.J. Lee, H. Ruh, H.J. Lee, Chem Mater. 15 (2003) 3294.

52. L. Vayssieres, K. Keis, A. Hagfeldt, S. Lindquist, Chem. Mater. 13 (2001) 4395.

53. Y. Tak, K. Yong, C. Park, J. Crystal Growth 285 (2005) 549.

54. M. Wang, E.J. Kim, S.H. Hahn, C. Park, Physica Status Solidi (a) 203 (2006) 2418.

55. R. Corkish, D.S.P. Chan, M.A. Green, J. Appl. Phys. 79 (1996) 195.

56. Y. Zhang, A. Mascarenhas, S. Deb, J. Appl. Phys. 84 (1998) 3966.

57. M. Burgelman, B. Minnaert, Thin Solid Films 511 (2006) 214.

58. B. Kannan, K. Castelino, A. Majumdar, Nano Lett. 3 (2003) 1729.









59. W. Geens, T. Martens, J. Poortmans, T. Aernouts, J. Manca, L. Lutsen, P. Heremans, S.
Borghs, R. Mertens, D. Vanderzande, Thin Solid Films 451 (2004) 498.

60. B. Mazhari, S. En. Mater. & Solar Cells 90 (2006) 1021.

61. A. Takshi, A. Dimopoulos, J.D. Madden, Solid-State Electronics 52 (2008) 107.

62. W.U. Huynh, J.J. Dittmer, N. Teclemariam, D.J. Million, A.P. Alvisatos, K.W.J.
Barnham, Phys. Rev., B 67 (2003) 115326.

63. R.D. Bettignies, J. Leroy, M. Firon, C. Sentein, Syn. Met. 156 (2006) 510.

64. V. Lemaur, M. Steel, D. Beljonne, J.L. Bredas, J. Cornil, J. Am. Chem. Soc. 127 (2005)
6077.

65. Medici User's Manual, Avant Corporation, Beaverton, OR, Release 2001.4, (2000).

66. W.K. Kim, Ph.D. Dissertation, University of Florida, Gainesville, FL, (2006).

67. E.J. Meijer, C. Detcheverry, P.J. Baesjou, E.V. Veenendaal, D.M. de Leeuw, T.M.
Klapwijk, J. Appl. Phys. 93 (2003) 4831.

68. H. Sirringhaus, N. Tessler, R.H. Friend, Science 280 (1998) 1741.

69. C. Tanase, E.J. Meijer, P.W.M. Blom, D.M. de Leeuw, Org. Elec. 4 (2003) 33.

70. T.A. Chen, X. Wu, R.D. Reike, J. Am. Chem. Soc. 117 (1995) 233.

71. A.J. Casico, J.E. Lyon, M.M. Beerbom, R. Schlaf, Y. Zhu, S.A. Jenekhe, App. Phys. Lett.
88 (2006) 62103.

72. A.J.C. Wen, K.L. Chen, M.H. Yang, W.T. Hsiao, L.G. Chao, M.S. Leu, Surf. Coat. Tech.
198 (2005) 362.

73. Melles Griot Inc., http://mellesgriot.com/products/optics/oc_5_1.htm (2002) July 2008.

74. Kim, Y., S.A. Choulis, J. Nelson, D.D.C. Bradley, S. Cook, and J.R. Durrant, J. Mat. Sci.
40 (2005) 1371.

75. D.C. Olson, J. Piris, R.T. Collins, S.E. Shaheen, D.S. Ginley, Thin Solid Films 496
(2006) 26.

76. GetData Graph Digitizer, http://getdata-graph-digitizer.com (2008) July 2008.

77. K.R. Amundson, B.J. Sapjeta, A.J. Lovinger, Z. Bao, Thin Solid Films 414 (2002) 143.

78. S. Li, "Semiconductor Physical Electronics," 2nd Ed., Springer, 2006.









79. C.J. Brabec, A. Cravino, D. Meissner, N.S. Sariciftci, T. Fromherz, M.T. Rispens, L
Sanchez, J.C. Hummelen, Adv. Funct. Mater. 11 (2001) 374.

80. J. Seol, M.L. Monroe, T.J. Anderson, M.A. Hasnain, C.H. Park, Conference Record of
the 2006 IEEE 4th World Conference on Photovoltaic Energy Conversion 1 (2006) 236.

81. C.W. Tang, S.A. VanSlyke, C.H. Chem, J. Appl. Phys. 65 (1989) 3610.

82. Kobayashi, H. Hori, and Y. Tanaka, in Handbook of Liquid Crystal Research, edited by
P.J. Collings and J.S. Patel (Oxford University Press, New York, 1997), Chap. 10, 415.

83. H.S. Kim, J.H. Lee, C.H. Park, J. Kor. Phys. Soc. 41 (2002) 395.

84. J.S. Kim, R.H. Friend, F. Cacialli, J. Appl. Phys. 86 (1999) 2774.

85. K. Sugiyama, H. Ishii, Y. Ouchi, J. Appl. Phys. 87 (2000) 295.

86. W.W. Flack, D.S. Soong, A.T. Bell, D.W. Hess, J. Appl. Phys. 56 (1984) 1199.

87. U. Mitschke, P. Bauerle, J. Mat. Chem. 10 (2000) 1471.

88. P. Peumans, A. Yakimov, S.R. Forrest, J. App. Phys. 93 (2003) 3693.

89. E. Kymakis, E. Kourdoumas, I. Franghiadakis, S. En. Mater. & Solar Cells 90 (2006)
1705.

90. Phenomenex, Solvent Miscibility Table, http://designer-
drugs.com/pte/12.162.180.114/dcd/pdf/solvent.miscibility.pdf July 2008.

91. Z.A. Peng, X. Peng, J. Am. Chem. Soc. 123 (2001) 183.

92. Meliorum Technologies, Inc., www.meliorum.com (2008) July 2008.

93. Fluxim, www.fluxim.com (2008) July 2008.

94. H. Burkhard, H.W. Dinges, E. Kuphal, J. App. Phys. 53 (1982) 655.

95. R.J. Ellingson, M.C. Beard, J.C. Johnson, P. Yu, O.I. Micic, A.J. Nozik, A. Shabaev,
A.L. Enfros, Nano Lett. 5 (2005) 865.

96. R.D. Schaller, V.I. Klimov, Phys. Rev. Lett. 92 (2004) 186601.


205









BIOGRAPHICAL SKETCH

Matthew Lowell Monroe was born in 1978, in Marietta, Georgia, to Ronald L. and Debbie

B. Monroe. He earned a Bachelor of Science degree from the Chemical Engineering Department

at the Georgia Institute of Technology in Atlanta in 2002. He joined the Chemical Engineering

Department at the University of Florida in 2002 and joined Dr. Anderson's research group in

2003. He earned a Doctor of Philosophy in chemical engineering in 2008.





PAGE 1

PROCESS DEVELOPMENT AND SIMULATI ON OF HYBRID PHOTOVOLTAIC CELLS By MATTHEW L. MONROE 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 1

PAGE 2

2008 Matthew L. Monroe 2

PAGE 3

To my wife, Kim 3

PAGE 4

ACKNOWLEDGMENTS I acknowledge my loving wife, Kim, for her pa tience, perseverance, and support. She was a source of inspiration, and without her this could not have been possible. I thank my parents for their continued love and support that has guide d me through my personal and academic life. I thank my advisor, Dr. Tim Anderson, for hi s advice and guidance in my research. I would like to thank all of my committee member s for their guidance and direction. Dr. Chinho Park granted me the use of his laboratory and much assistance and advice in organic photovoltaics. Dr. Kirk Ziegler provided labo ratory space and assistance with nanocrystal chemistry. Dr. Oscar Crisalle and Dr. Sheng Li provided valuable guidance through their roles in the interdisciplinary photovoltaics team. I thank Dr. Chinho Parks students at Ye ungnam University (par ticularly Jiyoun Seol, Young Wook Kim, Trong Nguyen Tam Nguyen, a nd Md. Azizul Hasnain) for extensive experimental support and for making a foreign c ountry feel like home. I thank Dr. Woo Kyoung Kim for sharing his expertise in Medici and phot ovoltaics. I thank Dr Zieglers students, particularly Justin Hill and Randy Wang, for assi stance with their laboratory equipment and for many helpful discussions. I thank all of the support staff at the Chemical Engineering department, particularly Sean Poole for helping me set up my simulation work. Special thanks go to Sherrie Jenkins, for performing scheduling miracles on a regular basis. Finally, I would like to thank all of my friends and family that have stood behind me and helped me through my doctoral research and my w hole life up to this point. They have helped mold me into the person I am today, and I am grateful for it. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4LIST OF TABLES ...........................................................................................................................7LIST OF FIGURES .........................................................................................................................9ABSTRACT ...................................................................................................................... .............151 INTRODUCTION ................................................................................................................ ..17Introduction .................................................................................................................. ...........17Motivation .......................................................................................................................17Photovoltaic Technologies ..............................................................................................18Organic Photovoltaics .....................................................................................................19Simulation of Hybrid Photovoltaic Devices ....................................................................23Targeted Research ...........................................................................................................242 MEDICI SIMULATIONS OF HYBRID SOLAR CELLS ....................................................27Introduction .................................................................................................................. ...........27Initial Modeling Efforts ..........................................................................................................29Preliminary Model Description .......................................................................................29Definition of model in Medici ..................................................................................30Impurity profile ........................................................................................................32Cell dimension adjustments .....................................................................................34Illumination source ...................................................................................................37Summary ..................................................................................................................38Simulation of a Real Cell ..................................................................................................... ...39Model Parameters ............................................................................................................39Initial Simulations ........................................................................................................... 42Reflectance of Ag electrode .....................................................................................43Mobility ....................................................................................................................45Exciton diffusion length ...........................................................................................46Doping density .........................................................................................................47Density of states .......................................................................................................48Open-circuit voltage examination ............................................................................48Absorption coefficient adjustment ...........................................................................50Multi-stage absorption ..............................................................................................51P3HT replacement with CIS .....................................................................................54Shortcomings of the initial model ............................................................................61Two-Step Simulation Technique ............................................................................................62Photogenerated Carrier Distribution ................................................................................63Line Source Generation ...................................................................................................64 5

PAGE 6

J-V Curves .......................................................................................................................70Summary of Results ........................................................................................................743 ORGANIC AND HYBRID SOLAR CELL PROCESS DEVELOPMENT ........................122Introduction .................................................................................................................. .........122ITO Anode Treatment ...........................................................................................................122Bi-layer Organic Solar Cell Fabrication ...............................................................................126Cell Fabrication Procedure ............................................................................................127Substrate Preparation .....................................................................................................127Spin-Coating .................................................................................................................. 128Evaporation ................................................................................................................... .128Encapsulation ................................................................................................................1 28Film Drying ................................................................................................................... 129PEDOT:PSS ..................................................................................................................130P3HT ............................................................................................................................ ..133Bi-layer Cell Fabrication ...............................................................................................133Solvent Comparisons ............................................................................................................135Hybrid Bulk Heterojunction Cell Fabrication ......................................................................138Nanocrystal Synthesis and Surfactant ...........................................................................139Hybrid Films .................................................................................................................. 140TOPO-coated CdSe ................................................................................................140P3HT solubility in chloroform and pyridine ...........................................................142Cell Fabrication .............................................................................................................144Cells deposited from chloroform solutions ............................................................144Thicker-film cells in chloroform solution ..............................................................145Chlorobenzene solvent ...........................................................................................149Commercial CdSe nanocrystals .............................................................................150Hybrid cell performance summary .........................................................................150Particle Induced Nanostructuring .........................................................................................1524 CONCLUSIONS AND FUTURE WORK ...........................................................................195Conclusions ...........................................................................................................................195Hybrid Photovoltaic Simulation ....................................................................................195Anode Surface Treatment ..............................................................................................196Solvent Selection ...........................................................................................................196Hybrid Bulk Heterojunction Photovoltaic Development ..............................................197Future Work ..........................................................................................................................198Organic Photovoltaic Simulations .................................................................................198Development of Hybrid Photovoltaic Cells ..................................................................199LIST OF REFERENCES .............................................................................................................201BIOGRAPHICAL SKETCH .......................................................................................................206 6

PAGE 7

LIST OF TABLES Table page 2-1. P3HT Properties for Device Simulations. ..............................................................................762-2. CdSe Properties for Device Simulations. ..............................................................................762-3. Electrode Properties for Device Simulations. ........................................................................762-4. Impurity profile inputs for simulations. ................................................................................ .792-5. Solar cell performance measures for unit ce lls of 12 nm width and varying thickness. .......812-6. Performance measures for simulated hybrid cells with varying CdSe half-width. ................832-7. Cell performance measures from publis hed and digitally converted J-V curves. .................852-8. ZnO properties used for hybrid solar cell simulation. ...........................................................862-9. Performance measures for simulated cells with varying carrier mobility. ............................902-10. Performance measures for simulated cells with varying exciton diffusion lengths. ...........922-11. Absorption data and short-circuit cu rrent for graded absorption simulations. ....................992-12. Materials properties for P3HT and CIS used in cell simulations. ......................................1012-13. Performance measures for ZnO:CIS cells with an individual material property set at the P3HT value. ................................................................................................................1012-14. Performance measures for simula ted ZnO:CIS solar cells with the P3HT absorption spectrum. ..................................................................................................................... .....1022-15. Performance measures for simu lated ZnO:CIS solar cells with P3HT values for absorption coefficient and electron affinity. ....................................................................1032-16. Performance measures for simu lated ZnO:CIS solar cells with P3HT values for absorption coefficient, electron affinity, and energy band gap. .......................................1042-17. Cell performance measures for real and simulated solar cells. .........................................1182-18. Performance measures for simulated cells with 40 nm LD and varying mobility values. ..............................................................................................................................1182-19. Generated carriers and performan ce measures for simulated solar cells. ..........................1213-1. Chemical composition of ITO films ....................................................................................155 7

PAGE 8

3-2. Performance of organic solar cells on treated ITO substrates .............................................1553-3. RMS surface roughness of P3HT films shown in Figure 3-11. ...........................................1633-4. J-V data for bi-layer solar cells. ...........................................................................................1663-5. Solvents considered for hybrid bulk heterojunction film deposition. .................................1663-6. Mean rms surface roughness in nm for hybrid films deposited from selected solvents ......1703-7. Hybrid solutions of P3HT and TOPO-coated CdSe nanocrystals in a mixed solvent of chloroform and pyridine. .................................................................................................1763-8. Film properties for P3HT films deposited from various solvents. .......................................1813-9. Hybrid solar cell fabrication information and performance data. .......................................190 8

PAGE 9

LIST OF FIGURES Figure page 1-1. Worldwide cumulative installed PV Power in Megawatts from 1992 to 2006. ....................261-2. Common organic materials used in solar cell development. .................................................262-1. Hybrid solar cell and correspondi ng unit cell used for device simulation. ...........................762-2. Wavelength-dependent absorption coe fficient data used in simulations. ..............................772-3. J-V curves for simulated hybrid solar cell s with different methods of specifying doping density.. ..................................................................................................................... .........772-4. Simulated J-V curves showing the effect of doping density in the CdSe nanorods.. ............782-5. Variation of open circuit voltage with doping density of CdSe nanorods. ............................792-6. Simulated J-V curves with varying unit cel l thickness. .........................................................802-7. Solar cell parameters for unit 12 nm wi de unit cells with varying cell thickness. ................802-8. J-V curves for hybrid cells with varying nanorod width. ......................................................812-9. J-V curves for simulated hybrid cells w ith CdSe nanorod half-thickness between 10 and 25 nm. .................................................................................................................... ......822-10. Solar cell performance measures for simu lated hybrid cells with varying CdSe halfwidth. ........................................................................................................................ .........822-11. J-V curves for hybrid solar cells with different light sour ce specifications. .......................832-12. Illustration of the P HOTOGEN command in Medici. .........................................................842-13. SEM images of ZnO nanofibers and nanofiber and P3HT composite films. .......................842-14. J-V curve for a real ZnO:P3HT solar cell to be used for verification of Medici simulations. .................................................................................................................. ......852-15. J-V and P-V curves for the real so lar cell fabricated by Olson et al. ..................................852-16. Unit cell used for simulations of ZnO/P3HT hybrid cells.. .................................................862-17. Medici unit cell used for device simulation. ........................................................................872-18. Simulated J-V curves for ZnO:P3HT solar cell using two P3HT regions. ...........................88 9

PAGE 10

2-19. J-V curves for simulated cells with zer o absorption in the P3HT2 region and varying reflectance from the Ag electrode. .....................................................................................892-20. Simulation results showing the effect of changing charge mobilities in the P3HT regions. ...................................................................................................................... .........902-21. J-V curves for simulated cells with varying exciton diffusion length. ................................912-22. Extrapolations to estimate VOC for simulated cells with varying exciton diffusion length..................................................................................................................................912-23. Solar cell performance measures for si mulated cells with varying exciton diffusion lengths. ...................................................................................................................... .........922-24. Real and simulated J-V curves for cells with varying P3HT doping density. .....................932-25. Real and simulated J-V curves for hybr id cells with varying ZnO doping density. ............932-26. Real and simulated J-V curves for hybrid solar cells with varying P3HT density of states. ....................................................................................................................... ...........942-27. Simulated J-V curves for hybr id solar cells with varying P3HT mobility. ..........................942-28. Simulated J-V curves fo r hybrid cells with varying P3HT doping concentrations. .............952-29. Energy band diagram for P3HT ZnO hybrid solar cells. ..................................................952-30. J-V curves for simulated cells with varying energy band gap in the active layers. .............962-31. Absorption coefficient vs. wavelength as tabulated in Medici. ...........................................962-32. AM1.5 solar spectrum. .................................................................................................. ......972-33. Carrier generation in simulated solar ce lls plotted with absorption coefficients for P3HT and ZnO. ...................................................................................................................972-34. J-V curves for simulated cells showing the original P3HT absorption profile and an edited absorption profile limiting absorption between 0.2 and 0.3 m. ............................982-35. J-V curves for simulated cells with ex citon diffusion length of 10 nm and multi-stage absorption regions. .............................................................................................................982-36. Examples of cumulative distribution f unction with mean of 10 nm and a range of standard deviation. .............................................................................................................992-37. Simulated J-V curves for cells with graded absorption profiles. .......................................1002-38. Simulated J-V curves with CIS replacing P3HT. ...............................................................100 10

PAGE 11

2-39. Simulated J-V curves for ZnO:CIS solar cells with an indivi dual material property changed to the P3HT value. ..............................................................................................1012-40. Simulated J-V curves for ZnO:CIS solar cells with the P3HT absorption spectrum applied. ...................................................................................................................... .......1022-41. Simulated ZnO:CIS solar cells with P3HT values for absorption coefficient and electron affinity. ...............................................................................................................1032-42. Simulated ZnO:CIS solar cells with P3HT values for absorption coefficient, electron affinity, and energy band gap. ..........................................................................................1042-43. Simulated J-V curves for ZnO:P3HT solar cells with varying carrier mobility. ...............1052-44. Calculation method for estimated second derivatives of J-V curves. ................................1052-45. Estimated second derivative Jest for simulated ZnO:P3HT solar cells with varying carrier mobilities and carrier generation in the full P3HT region. ...................................1062-46. Simulated J-V curves for ZnO:P3HT solar cells with varying carrier mobilities and carrier generation in the 10-nm exciton diffusion length region. ....................................1072-47. Estimated second derivative Jest for simulated ZnO:P3HT solar cells with varying carrier mobilities and carrier generation in the 10-nm exciton diffusion length region. .1082-48. Simulated J-V curves for ZnO:P3HT solar cells with varying absorption coefficients in P3HT.............................................................................................................................1092-49. Simulated J-V curves for ZnO:P3HT solar cells with varying energy band gap. ..............1092-50. Simulated J-V curves for ZnO:P3HT solar cells with varying band gap and absorption in the P3HT region. ..........................................................................................................1102-51. Simulated J-V curves for ZnO:P3HT solar cells with P3HT energy band gap of 1.0 eV and hole mobility of 500 cm2/Vs. ...................................................................................1102-52. Photogenerated carrier distribution in pairs/cm3 for the full unit ce ll and the region of 13 nm x 27 nm along the edge of the ZnO nanorod. .................................................1112-53. Carrier mapping scheme for two-stage simulations. .........................................................1122-55. Cumulative number of photogenerated carrie rs in simulated cells with varying exciton diffusion length. ...............................................................................................................1142-56. Photogenerated carrier distribution along xand ycoordinates for models with varying exciton diffusion lengths.....................................................................................1152-57. Contour plots of the photogenerated car rier difference between the full absorption simulation and the 40 nm LD simulation. ........................................................................116 11

PAGE 12

2-58. Photogenerated carriers at the tip and base corner poin ts of the nanorod in simulated hybrid cells. ......................................................................................................................1172-59. J-V curves for simulated solar cells using line-source car rier generation. ........................1172-60. J-V curves for simulated cells with a 40 nm LD and varying carrier mobility. ................1182-61. Photogenerated carrier distribution for a simulated unit cell with LD = 40 nm and LD = 40 with the P3HT absorption coefficient increased by 50%. ........................................1192-62. Difference in photogenerated carriers between 150% and 100% P3HT absorption coefficients in simulated cells with 40 nm LD. ...............................................................1202-63. J-V curves for simulated cells with varying absorption coefficient in P3HT. ...................1213-1. J-V curves for organic solar cells on treated ITO substrates ...............................................1553-2. Energy band diagram for bi -layer organic solar cells. .........................................................1563-3. Steps for bi-layer solar cell fabrication. ............................................................................... .1563-4. PEDOT:PSS film thickness vs. spin-coater speed. ...............................................................1573-5. P3HT film thickness vs. spin-coating speed. ........................................................................1573-6. FTIR spectra for P3HT film and solution. ...........................................................................1583-7. FTIR spectra for PEDOT:PSS film and solution. ................................................................1593-8. J-V curves for bi-layer or ganic solar cells fabricated with a single 80 nm thick layer of PEDOT:PSS. .................................................................................................................... 1603-9. J-V curves for bi-layer organic solar cells fabricated with two 40 nm thick layers of PEDOT:PSS. .................................................................................................................... 1613-10. Calibration curves for slow and fast-filtered PEDOT:PSS. .............................................1623-11. AFM images of P3HT films. ..............................................................................................1623-12. J-V curves for bi-layer solar cells fabricated on untreated ITO substrates. ......................1633-13. J-V curves in the dark and under 100 mW/cm2 illumination. ...........................................1643-14. Surface roughness measurements by profilometry for hybrid films deposited from various solvents. ............................................................................................................. ..1673-15. RMS surface roughness for 5 x 5 m surface area samples measured with AFM. ...........1683-16. RMS surface roughness for 1 x 1 m surface area samples measured with AFM. ...........169 12

PAGE 13

3-17. Optical microscope imag es of selected films. ...................................................................1703-18. SEM images of selected hybrid films. ...............................................................................1733-19. Film surface roughness vs. pyridine con centration in the chloroform solvent for TOPO-coated and pyridine-coated CdSe nanocrystals. ...................................................1753-20. Optical microscope images of hybrid films deposited from 20.4 mg/ml, 19 mg/ml and 5 mg/ml solutions. ............................................................................................................ 1763-21. Optical microscope im age of 19 mg/ml hybrid film deposited at 3000 rpm and subjected to a pure solvent spin -coating step at 8000 rpm. .............................................1773-22. Optical microscope images of P3HT films deposited from 5 mg/ml solutions in chloroform, 1:1 chloroform: pyridine, and pyridine. ........................................................1773-24. Surface profiles of P3HT films deposited from chloro form, 1:1 chloroform:pyridine, and pyridine solvents. ......................................................................................................17 93-25. Dark and illuminated J-V curves fo r cells generated from 5 mg/ml composite solutions in chloroform mixed with 2% pyridine and pure chloroform. .........................1813-26. Dark and illuminated J-V curves for 10 mg/ml hybrid solution deposited with lowspeed spin-coating. ...........................................................................................................1823-27. Surface profiles of film deposited from 12 mg/ml P3HT and 25 mg/ml hybrid solutions in 2% pyridine in chloroform. ..........................................................................1833-28. Dark and illuminated J-V curves for hybr id bulk heterojunction solar cell deposited from 25 mg/ml solution and P3HT polymer cell deposited from 12 mg/ml solution. .....1843-29. Short-circuit current decay for hybrid and P3HT. ..............................................................1853-30. Dark and illuminated J-V curves for hybrid bulk heterojunction and pure P3HT solar cells with limited air exposure during processing. ...........................................................1863-31. Dark and illuminated J-V curves for hybrid solar cell fabricated from chlorobenzene with 2% pyridine solution. ...............................................................................................1873-32. Dark and illuminated J-V curves for a hybrid solar cell fabricat ed with commercial CdSe nanopowder. ...........................................................................................................1883-33. Illuminated J-V curves for hybrid bul k heterojunction solar cells with various fabrication conditions.......................................................................................................1893-34. J-V curves for the best bi-layer and hybrid cells shown in this dissertation. ....................1893-35. Film thickness for multi-layer hybrid films. ......................................................................191 13

PAGE 14

3-36. SEM and AFM surface images of multi-layer hybrid films. .............................................1913-37. Optical microscope images for multi-layer hybrid films. ..................................................1933-38. RMS surface roughness for multi-layer hybrid films. .......................................................194 14

PAGE 15

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 PROCESS DEVELOPMENT AND SIMULATI ON OF HYBRID PHOTOVOLTAIC CELLS By Matthew L. Monroe December 2008 Chair: Timothy J. Anderson Major: Chemical Engineering The fabrication and simulation of hybrid bulk heterojunction photovoltaic cells was studied to develop an understanding of these devices an d facilitate improvements in device fabrication techniques. The device simulation software package Medici was app lied to this type of device for the first time, and novel device fabricati on techniques were developed to improve device performance. Medici was adapted for a hybrid system cons isting of an ordered array of inorganic nanorods interspersed with an absorbing, se miconducting polymer. A cell taken from the literature was simulated using a two-stage simu lation technique which in dependently calculated the light absorption profiles and the cell performa nce. This technique ap plied a line source of carriers directly at the inte rface between the inorganic and organic regions of the cell, realistically imitating the exciton dissociation phy sics of real hybrid cells. The simulations showed low current densities as compared to th e real cell, due to str ong recombination at the material interface. Nitrogen plasma treatment of the transparen t anode in bi-layer and hybrid photovoltaic cells was found to reduce surface roughness, increas e the hydrophilic nature of the film, and improve cell performance. A range of solvents were tested for hybrid bulk heterojunction film 15

PAGE 16

16 deposition, with chloroform, chlo robenzene, and o-dichlorobenzen e found to provide films with low surface roughness and strong uniformity. Hybr id bulk heterojunction solar cells were fabricated, and these cells showed low performance of less than 1% efficien cy. The active layers degraded upon exposure to air, re sulting in a drop in short-circu it current density that was more pronounced for hybrid films than for pure polymer single-layer cells. These findings highlight the environmental sensitivity of these devices and the need for an inert environment for cell fabrication and testing.

PAGE 17

CHAPTER 1 INTRODUCTION Introduction Since the discovery of the photovoltaic effect and the design of the first functional solar cell, photovoltaic technology has consistently developed to become an increasingly viable energy source. Photovoltaics has developed into many classes of devices and materials systems. Photovoltaic technology can provide clea n, efficient, and portable energy. Motivation There is currently a strong push toward alte rnative energy sources as the price of oil increases and nations worldwide work to slow the emission of greenhouse gases such as COx and NOx. Many countries have established conservation and alternative energy programs in attempts to control the output of these gases. Recent in creases in oil and gas prices and controversy surrounding global warming have driven public rec ognition of the need for alternative renewable energy sources. While small-scale steps such as hybrid cars stand to relieve a small amount of the worlds fossil fuel consumption, new tec hnologies such as photovoltaic energy must be developed to fulfill the worlds large-scale energy needs. Of the available candidates for alterna tive large-scale ener gy production, photovoltaic energy conversion has many qualities that make it a leading technology, including: Photovoltaic technology has been studied intensely for many year s, initially driven by its use in the space program to provide energy fo r satellites and space vehicles, to prepare it for commercialization. Energy generated from solar cells can be ge nerated locally, resulting in reduced energy distribution costs and a more reactive system. Because of its local power generation, photovol taics are optimal for power generation in remote locations where it is difficult or impossible to connect to a local grid. 17

PAGE 18

Solar energy is plentiful. Americas energy needs could be supplied by a single solar array covering an area of 100 x 100 miles in the uninhabited deserts of the western United States. Photovoltaic sources provide peak energy during peak consumption times. Energy consumption is at its maximum during the mi ddle of the day, requiring power plants to work harder to provide the necessary energy to their consumers. Because the suns rays on earth are also at a maximu m during that time, solar en ergy provides maximum energy at the most critical portion of the day. Worldwide photovoltaic installations have begun to grow rapidly, as shown in Figure 1-1 [1]. The U.S., who at one time was the world le ader in photovoltaics, ha s fallen behind countries such as Japan and Germany who have made str ong efforts to boost their solar energy programs. However, recent initiatives such as the Mil lion Solar Roofs Initiative and Solar America Initiative, along with federal and state sponsor ed cost-sharing program s for residential and commercial solar installations, have created a ne w boom in American solar installations [2]. Photovoltaic Technologies Many photovoltaic technologies are currently in the develo pmental or production stages, including crystalline and polycrystalline si licon, thin film, concen trator arrays, space photovoltaics, and organic cells. Polycrystalline silicon is currently the dominant technology in the photovoltaics market, drawing on years of silicon processing technologies to provide a low production cost. Polycrystalline silicon modules area commercia lly available for as low as $4.29/Watt in April 2008, approximately $0.50/Watt ch eaper than the U.S. average cost of $4.81/Watt [3]. These panels are ea sily identified by thei r deep blue color and are currently in production worldwide. The next wave of phot ovoltaic production app ears to be thin-film photovoltaics. These cells employ direct-bandgap materials with strong absorption coefficients such as amorphous silicon (a-Si), cadmium te lluride (CdTe), and copper indium-gallium diselenide (CuInxGa1-xSe2). Many of these cells have demonstrated extremely promising 18

PAGE 19

laboratory efficiencies, and the thin layers reduce material costs and provide the ability to create flexible cells for a wider range of applications. Concentrator photovoltaic arrays generate a large amount of power from a single cell by using an array of mirrors to focus the suns ener gy on the cell. This tech nology has been used to reduce large-scale production costs because the cost of a mirror is lower than the cost of a photovoltaic panel. Space solar cells are the most advanced cells, using expensive growth methods to produce high quality photovoltaic material. These cells provide extremely high performance and durability through their multi-ju nction stack technology, but this comes at a high cost that makes the cells too expensive fo r standard terrestrial applications. Under concentration, however, their high efficiency (a pproaching 40%) mitigates their higher cost. Organic and hybrid solar cells provide extr emely lightweight devices and low-cost manufacturing, but currently th eir low performance and life time limits their applications. Organic Photovoltaics Compared to their inorganic counterparts, the development of organic PV is in its infancy. Drawing on advances in organic light emitting diodes, however, considerable progress has been achieved that encourag es continued exploration of organic solar cells. Organic photovoltaic technology uses organi c molecules or polymers to absorb sunlight and generate photocurrent. Many of these materials have extr emely high absorption coefficients with maxima in the visible region of the spectrum ( 105 cm-1), so an extremely th in layer (hundreds of nanometers) is sufficient for absorbing incident light. Because the films are extremely thin and flexibile as compared to inorga nic crystals, organic photovoltaics show promise for flexible and lightweight portable devices. Many organic materials can be rapidly deposited through inexpensive techniques such as thermal evapora tion at moderate vacuum, or by spin-coating, dipcoating, screen printing, ink jet printing, and spray coating at room temperature and pressure. 19

PAGE 20

The approaches to using organi cs for light conversion can be categorized into three general classes of cell structures: bi-lay er, dye-sensitized, and bulk hetero junction. Similar to inorganic designs, bi-layer cells use a flat junction create d by stacking pand n-type organic layers, with additional layers incorporated for charge transport enhancement [4-6 ]. Dye-sensitized solar cells use an organic dye adsorbed on inorgani c transport materials, typically TiO2, so that the dye absorbs photons and the inorganic phase allows fo r efficient charge transport [7-10]. Bulk heterojunction cells consist of donor and acceptor materials mixed together to form a blended junction throughout the device active layer. Organic cells are distinguishe d from their inorganic count erparts by exciton creation upon photon absorption. Due to their extremely low exciton binding energy, inorganic p-n junction cells generate free carriers upon photon absorption, and the carriers are primarily collected by the field across the depleted junction. Organic ma terials, on the other hand, primarily generate excitons that have significant binding energy and transport by diffusion until they recombine or dissociate at an energetic interface to produ ce free carriers for eventual collection. Excitons are efficiently dissoci ated at a p-n junction in or ganic devices, although exciton dissociation occurs to a lesser extent at inte rfaces with electrodes, polymer chain defects, absorbed oxygen sites, or active-layer impurities [ 11]. Because of this dissociation requirement, only excitons generated within a diffusion length of the junction will contribute to the collected current. Exciton diffusion lengt hs are typically on the order of 5 to 20 nm for organic semiconductors, placing a limit on the thickness of active la yers and therefore the photon absorption extent [12-16]. The blended junction of a bulk heterojunction ce ll attempts to provide an interface within an exciton diffusion length throughout the entire active layer, thus allowing for thicker active layers with better adsorption an d more efficient exciton dissociation. Once the 20

PAGE 21

free carriers are generated they must be collected at their re spective electrodes before they recombine, which occurs to some extent in th e bulk or more prevalently at interfaces. Bulk heterojunction devices have been fabricat ed by blending several cl asses of materials, including multiple organic small molecules, polymer and organic molecules, polymer and carbon nanotubes, polymer and inorganic nanoparticles, and polymers deposited in a prefabricated inorganic nanostructure. Heterojunctions base d on small organic molecules such as copper phthalocyanine (CuPc) receive li ttle attention compared to polymer-based heterojunctions, but have demonstrated reasonable efficiency using a variety of depositi on techniques [17-19]. Carbon nanotubes have very recently received consid eration as a solar cell material, both as an ntype conductor in a heterojunction [20] and as a structured electrode [21]. Bulk heterojunctions fabricated from conjugated polymers and C60 represent the most widely-studied class of bulk heterojunction solar cell. Sin ce the discovery of ultra-fast photoinduced charge transfer from conducting polymers to C60 [22], polymer-C60 bi-layer and bulk heterojunctions have been studied exte nsively. The first cells using semiconducting polymers and C60 were fabricated by the same group [23] Early bulk heterojunction cells using C60 employed poly (pheneylene vinylene) (PPV) deri vatives as the polymer material, and issues centered on co-dispersion of the two materials to create a wellblended structure [24]. This issue was solved with the synthesis a nd application of [6, 6]-phenyl C61 butyric acid methyl ester (PCBM), a highly-soluble C60 derivative, [25, 26], and led to a record efficiency of 2.5% in 2000 [27]. Since this development, other fullerene derivatives have been s ynthesized and evaluated, but PCBM remains the most widely used [28, 29]. The most popular PPV-based polymers are poly [2-methoxy,5-( 2-ethylhexyloxy) -1,4-phe nylene-vinylene] (MEH-PPV) and poly [2methyl,5-(3*,7** dimethyloctyl oxy)]-p-phenylene vinylene (M DMO-PPV), although other 21

PAGE 22

derivatives have been demonstrated [30]. These cells frequently feature a hole transporting layer of poly (3, 4-ethylenedioxythi ophene):poly (styrenesulfonate) (PEDOT:PSS) to improve hole collection. The structures of PCBM and some polymers commonly used for bulk heterojunction solar cells are shown in Figure 1-2. In particular, polythiophene derivatives are now popular alternatives to PPV, with poly (3hexylthiophene) (P3HT) being the most commonly used [29, 31]. P3HT is synthesized with a regio-regular configuration wher e the polymer side chains alte rnate on opposite sides of the backbone chain. This arrangement aids in ali gning the polymer chains for efficient charge transfer along the backbone, with additional chai n straightening attributed to steric hindrance from the fullerene molecules [32]. Bulk heterojunction cells fabricated from P3HT and PCBM have reached efficiencies of 3.5% in 2003 [33] and 4.4% in 2005 [34]. The use of inorganic nanocrystals as a replacement for C60 is a relatively recent trend. In addition to their high electron mobilities, se miconductor nanocrystals can contribute to absorption and photocurrent generation in the hete rojunction active layer. These nanocrystals offer the possibility of bandgap engineering by mate rial selection and tightly controlling the size distribution, as well as the growth of different crystal shapes such as rods and tetrapods [35-37] to create more efficient charge transport pathways. As with early work involving C60, dispersion of the nanocrystals remains an issue in pro cess development. One approach to enhance nanocrystal dispersion is the use of sol-gel proc essing to grow nanocrystals in the polymer film [18, 38-40]. The nanocrystals are typically grown in solution, a nd are formed with a surfactant capping layer to relieve the high surface energy [41, 42]. Exchange of this surfactant has been demonstrated to obtain improved solubility and electrical properties [43-45]. 22

PAGE 23

Ordered heterojunctions have been fabricat ed from a variety of materials including mesoporous TiO2 [46, 47], but ZnO nanostructures have b een the most popular. Regular arrays of well-aligned ZnO nanowires have been gr own using a variety of techniques including MOCVD [48, 49], evaporation [50, 51], and soluti on-based thermal growth [52]. Tak et al. demonstrated selective MOCVD growth of Zn O nanoneedles on a patterned buffer layer [53], allowing for substrate patterning with nanomateria l coverage. Low-temperature growth has been achieved using a sol-gel precursor method [ 54]. Device performance of these ordered heterojunction devices has been limited by rod sp acings several times larger than the exciton diffusion length of the active polymer. Simulation of Hybrid Photovoltaic Devices The incorporation of excitons in device phys ics models has been done both inside and outside the organic photovoltaics world. Models of silicon solar cells show that the inclusion of excitons causes a decrease in dark current but an increase in photocurre nt [55], while further studies showed that this eff ect is only substantial when the exciton diffusion length is significantly greater than the electron diffusion le ngth [56]. This work was further expanded by Burgelman and Minnaert to include exciton diss ociation at surfaces, showing that even purely excitonic devices such as organic solar cells ca n be effective as long as the rate of exciton dissociation is high [57]. Simulations of polymer-inorganic hybrid ce lls have been performed with various assumptions. Studies have proven that effectiv e charge transport only takes place in these devices when the dimensions of the polymer regi on are less than or equal to the exciton diffusion length [58]. Additional m odels have validated this effect in polymer-C60 bulk heterojunction cells [59], highlighting the importance of excitons in these types of cells. The traditional circuit 23

PAGE 24

model of a solar cell has been modified to includ e an additional rectificat ion diode to include the effect of exciton recombination [60]. While work exists in development of models to describe organic phot ovoltaics, little work has been done in applying existing semiconductor modeling programs to simulate these cells. Recently, Takshi et. al. applied Medici to simulate an organic transistor using P3HT as the semiconductor [61]. Medici is a 2-dimensiona l device simulation program developed by Avant! Corporation. It is designed for the simulati on of MOS and bipolar tr ansistors, and models potential and carrier concentrations in a devi ce by solving Poissons equation and the electron and hole continuity equations. Targeted Research The work presented in this dissertation provides a study of hybrid photovoltaic devices from an experimental and theoretic al perspective. The device si mulation program Medici is used for the first time to provide simulations of an ordered bulk heteroj unction photovoltaic device with an array of inorganic nanorods interspersed with a semi conducting organic polymer. This design was chosen due to its semi-regular structur e that allows the device to be broken into a representative unit cell, providing greater detail in the simulations. By modeling the cell in this way, effective values of key parameters such as the carrier mobility, exciton diffusion length, and energy gap of the interface can be esti mated for the device during operation. This simulation supplements experimental work focusing on process development for a polymer-nanocrystal bulk heterojunction solar cell. The issues of charge transport and exciton dissociation are targeted. Charge transport is improved through surface treatment of the ITO electrode prior to active laye r deposition. This generates a smoother surface and promotes adhesion of subsequent layers Exciton dissociation is a ddressed through control of the morphology of the bulk heterojunction active layer. This is achieved through surface exchange 24

PAGE 25

of the nanocrystal surfactant, se lection of an appropriate solv ent for film deposition, and the introduction of a novel layer-by-layer deposition process. 25

PAGE 26

26 Figure 1-1. Worldwide cumulative installed PV Power in Megawatts from 1992 to 2006 [1]. Figure 1-2. Common organic mate rials used in solar cell de velopment regioregular P3HT (a), PCBM (b), MDMO-PPV (c), MEH-PPV (d), and PEDOT:PSS (e). S S O O OSO3H S O O * O O a) b) O c) e) d)

PAGE 27

CHAPTER 2 MEDICI SIMULATIONS OF HYBRID SOLAR CELLS Introduction Device modeling has lagged behind cell fabri cation for organic and hybrid photovoltaics. Cell performance has been characterized using e quivalent circuit theories [60, 62], and some work has been performed to model certain para meters such as cell lif etimes [63], charge recombination [64], and short-circuit current [59] The use of existing simulation programs to provide a fully-encompassed view of the device performance has not been attempted to this point. Organic photoabsorbing materials differ fundame ntally from most i norganic crystals in that the absorption of a photon generates an exciton, or bound electron-hole pair, with high binding energy. Excitons can be generated in inor ganic materials as well, but this occurs only over a very limited wavelength and temperature range and the resulting excitons exist with a low binding energy on the order of k T at room temperature [57]. Th is low binding energy causes the exciton to be highly unstable in th e presence of elevated temperatur es or strong fields, such as in the space-charge region of a p-n junction. In organic materials, excitons are generated nearly exclusively and exist with a bi nding energy or approximately 10x th at of the inorganic excitons, or ~ 300 meV. These highly stable excitons diss ociate at an interface with another material, where one of the carriers is transferred into th e neighboring material. In organic photovoltaic cells, the electron is typically transferred into an n-type material due to the relatively low electron mobility in most organic materials. Because the organic exciton exists as an e ssentially uncharged particle unaffected by fields, it travels through the so lar cell through diffusion, with the exciton diffusion length (LD) representing the maximum distan ce it can travel before self-a nnihilation. For many organic 27

PAGE 28

photoabsorbers, including P3HT, LD has been measured to be on the order of 10 nm [15]. This restricts the current in organic photovoltaics because only excitons generated within LD of an electron acceptor can generate free carriers, and any photons abso rbed by the organic absorber outside of this distance are lost to recombination. This inhere nt limitation is the motivation behind the bulk heterojunction cell design, in wh ich the organic absorber and electron acceptor are blended on a scale that approximates LD. The hybrid bulk heterojunction desi gn considered for simulations in this dissertation is an ordered array of inorganic nanorods with an organic absorbing pol ymer interspersed between and on top of the rods. This inorganic array is a ssumed to be uniform in terms of rod length, rod width, and rod spacing, and the polymer is assume d to fully occupy vacant areas between the rods. The simulation program chosen for this effort was Medici [65]. Medici is a device modeling software package from Synopsis designed to simulate diodes, MOS transistors, and bipolar transistors, as well as emissive devices. The program is provides a two-dimensional simulation area and solves Po issons equation and the electron and hole continuity equations to generate two-dimens ional distributions of potential and carrier concentrations [65]. Medici is designed for the simulation of crystalline inorganic materials, but has been used in certain rare instances for the modeling of organic semiconductors. Recently, Takshi et al. used Medici to simulate a dual gate organic transistor using P3HT as the organic semiconductor [61]. The simulation applied literature values for P3HT properties and measured the performance of singleand dual-gate OFET de signs with varying film thickness. No effort was made by the authors to account for exciton d ynamics in the device because the polymer was functioning as a transistor rather than a solar absorber, so the influence of excitons would be minimal. 28

PAGE 29

The semiconductor properties that are nece ssary for a proper simulation using Medici include the electron and hole mobility, band gap en ergy, permittivity, electron affinity, density of states in the conduction and valence band, and doping density. These properties are well-known for most common semiconductor materials, but are less well-characterized for many organic materials. Despite this, measurements and assu mptions regarding the necessary properties for P3HT can be found in the literature, and these va lues were used as starting points for device simulation. Initial Modeling Efforts A dispersion of P3HT in an ordered array of CdSe nanor ods was chosen as a first attempt at hybrid solar cell modeling using Medici. The choice of materials and dimensions were somewhat arbitrary, but represent a pseudo-re alistic scenario with which to expore the capabilities of Medici for hybrid cell modeling. This initial modeli ng effort serves as a precursor to attempts to simulate a real cell from literature. Preliminary Model Description Hybrid solar cells consisting of an ordered array of CdSe nanorods dispersed in a P3HT matrix were simulated using the device modeling software package Medici. The nanorods were assumed to be vertically aligned, be in contact with the aluminum back electrode, have uniform dimensions, and have uniform spacing. An illustra tion of this design is shown in Figure 2-1. Due to the symmetry imposed in this structure, the cell can be divided in to a basic repeating unit cell, as illustrated in the figure. The active layer is fixed at 100 nm thick, while the CdSe nanorod has dimensions of 90 nm x 10 nm. The rod spacing is fixed at 20 nm. These dimensions define a P3HT capping layer of 10 nm over the tip of each of the CdSe nanorods. Additionally, the unit cell ha s a polymer thickness of 10 nm to the side of the nanorod, due to the 20 nm rod spacing imposed in the model. This is chosen because it is equal to the approximate 29

PAGE 30

exciton diffusion length in P3HT [15]. The unit cell used for simulation has dimensions of 100 nm thick (the thickness of the device, not includ ing electrodes) by 15 nm wide (half of the 10 nm rod width plus half of the 20 nm rod spacing). The top contact is indium tin oxide (ITO), and the bottom contact is aluminum. The unit cell occupies 0.1 m in the y direction and 0.015 m in the x direction. The CdSe nanorod occupies an area from 0.01 y 0.1 m and 0 x 0.005 m. P3HT occupies the other regions, for y < 0.01 m and x > 0.005 m. The materials properties assigned for the simula tion were taken from the literature and are displayed in Tables 2-1 2-3. Material properties for CdSe were taken from previous simulations from Dr. Woo Kyoung Kim [66]. Figur e 2-2 shows the absorp tion coefficient data used for several different materials in simulatio ns, with the sources of the data noted in the caption. Note that the electrical band gap and op tical band gap can be i ndividually specified in MEDICI, but they are assumed to be equal in all simulations unless otherwise noted. The doping density was calculated to depe nd on the amount of time that P3HT is exposed to air, but falls in the narrow range of 1 x 1016 to 2 x 1016 cm-2 for air exposure times on the order of 100 hr [67]. This value was further extended to 5 x 1016 cm-2 by Takshi et al. by as suming extended exposure times [61], and this value was used as a starting point for simulations. The P3HT hole mobility of 0.1 cm2/V-s is the highest reported mobility for this material [68]. Definition of model in Medici This section gives a brief de scription of the steps require d to perform a simulation in Medici. Medici runs based on use r-defined input files that are easil y edited in text format with any text editing program, typically Notepa d or Wordpad on Windows computers. For photovoltaic simulations, these input files contain the following in formation, typically in this order: 30

PAGE 31

1) Mesh definition 2) Region definitions 3) Electrode definitions 4) Material property definitions 5) Illumination definition 6) J-V calculation 7) Output definition. The simulation mesh is the collection of points where calculations are performed throughout the simulation area. The spacing of the mesh points can be specified by the user and is somewhat arbitrary, with the exception that mesh points sh ould exist at boundaries between regions to facilitate convergen ce of the calculations. Medici allows a maximum of 3,200 mesh points, with more points pr oviding a more well-defined cal culation at the expense of computational time. Regions of the mesh can be defined to correspond to different materials in the simulation. These are specified through ranges of (x, y) coordinates in units of m, with y = 0 corresponding to the top of the simulation area. The electrodes are specified in terms of their optical properties and are not defined by (x, y) coordinates. Calcul ations are performed by Medici at the interface points with the electrodes, but not within the actual electrode. Material properties are specifi ed to correspond to the materi al ranges given in the region definition. An example would be MATERIAL REGION=P3HT PERMITTI=3 NC300=2E18 NV300=2E19 EG300=1.7 + EGO300=1.7 AFFINITY=3.15 AB S.FILE="abs-p3ht.txt" PR.TAB where the properties for the area defined as P3HT are specified. Carrier mobilities and impurity profiles are defined in separate statements. Further details on impurity profile definitions are given in the following section. The absorption coefficient is defined through an external text file, where the value can be sp ecified by the user for a range of wavelengths. 31

PAGE 32

The illumination source can be defined in multiple ways. For absorption models such as photovoltaic cells, radiation can be defined to follow any specified profile and generated from a point source or line source at a ny location outside of the simula tion area. Further details are given in a later section describing the effect of altering the origin of the illumination source. In addition to general illumination, sources of phot ogenerated carriers can be specified along any line within the simulation area. With the defined illumination source, Medici calculates the photogenerated carrier distribution throughout the simula tion area based on the absorption coefficient data given for each material region. The program then begins ite rating over a range of applied biases specified by the user until the current and potential has been mapped and converged throughout the simulation area. Once a solution has been achie ved, Medici provides a vast array of output mechanisms for the user to extract data fr om the simulation, including plot generation and numerical data extraction. Impurity profile The way in which Medici assigns impurity profile s in simulations is a bit of a mystery. A few simulations were performed to determine th e proper way to call this function. The profile can be defined in terms of (x, y) coordinate s or by the REGION statement which applies values to an area defined by a specific name. The impurity profile input statements used are listed in Table 2-4. The file impurity_1 specifies the doping areas by the REGION statement, which generates doping in an area which has previously been defined in coordinate space and assigned a label (P3HT or CdSe). The file impurity_2 fi rst applies p-type doping at a concentration of 5 x 1016 cm-2 to the entire cell area and then applies an n-type doping of 6 x 1016 cm-2 to the area occupied by CdSe. This depends on Medici overwriting the impurity values when multiple 32

PAGE 33

definitions are made in the same region, in this case the CdSe region. In impurity_3 the cell is sectioned into three rectangl es, two which encompass the P3HT region and one which defines the CdSe region. The P3HT region consists of the regions y 0.01 m and (0.01 y 0.1 m, 0.005 x 0.015 m). The CdSe region is defined of the area from (0.01 y 0.1 m, 0 x 0.005 m). The file impurity_4 considers the possibility that MEDICI sums doping specifications rather than overwriti ng them. It takes the same fo rmat as impurity_3 but assigns the n-type doping concentration as 1 x 1016 cm-2 to account for the pos sibility that 5 x 1016 cm-2 is cancelled out by the p-type impurities when th e statements are superimposed. In impurity_5 this concept is again tested by breaking the cell in to the same three sectio ns as in impurity_3, but assigning each section a p-type doping of 5 x 1016 cm-2. Then, the CdSe section is respecified as n-type with a concentration of 6 x 1016 cm-2. The files impurity_4 and impurity_5 failed to conv erge. It is intere sting to note that impurity_5 failed, but impurity_2 did not. Both appear to perform the same actions of filling the entire cell with p-type dop ing at a concentration of 5 x 1016 cm-2, and then applying an n-type doping of 6 x 1016 cm-2 to the CdSe region. The file impurity_4 directly applied an impurity profile of 1 x 1016 cm-2 in the CdSe region. J-V curves from the three successful simulati ons are shown in Figure 2-3. From the graph, it is obvious that impurity_2 was significantly different than impurity_1 or impurity_3. Both impurity_1 and impurity_3 show nearly identical J-V curves, suggesting that the calculation is very similar. It is interesting, however, that th e results were not identical. The curve for file impurity_2 shows an open circuit voltage of 0.185 V, approximately 30% lower than the values of 0.246 and 0.253 V for the other curves. This would seem to be 33

PAGE 34

indicative of a lower effective doping density. The purpose of input files impurity_4 and impurity_5 was to determine this difference, but these simulations failed to converge. Further simulations demonstrat ed the effect of doping density in the CdSe nanorods, with the results shown in Figure 2-4. The Cd Se doping density was varied from 3 x 1016 cm-2 7 x 1016 cm-2. The graph shows a linear dependence of VOC on the CdSe doping density. This dependence is further illustrated in Figure 2-5, along with the linear tre ndline fit through the data. The fit in Figure 2-5 predicts a VOC of 0.181 V for an undoped semiconductor nanorod and a value of 0.193 V for a doping level of 1 x 1016 cm-2. The curve generated from the input file impurity_2 displayed an open-circuit voltage of 0.185 V, which corresponds to a doping density of 3.3 x 1015 cm-2 according to the trendline in Figure 2-5. However, this linear trend should not hold for doping levels approaching zero, because zero doping in the CdSe region should coincide with a VOC of zero, due to the lack of a p-n junction in the device. Any error from this value could arise from a space-charge region forming between P3HT and one of the contacts. From these simulations, the optimal method for defining impurity profiles was determined to be the use of the REGION statement. This is simpler than defining the impurities for multiple regions as in the impurity_3 simulatio n. The REGION statement ensures that each region is properly assigned the appropriate doping density. Cell dimension adjustments After definition of the basic model parameters simulations were performed to determine the effect of varying cell dimensions. The first parameter adjusted was the cell thickness, which was accomplished by adjusting the length of th e nanorod while maintaining a 10 nm polymer capping layer. These simulations were perfor med with the CdSe nanorod width set to 4 nm 34

PAGE 35

rather than 10 nm. This sets the nanorod half-w idth in the unit cell simulation area as 2 nm. The resulting simulated J-V curves are shown in Fi gure 2-6. The length liste d in the legend is the unit cell thickness, not including electrodes, so the CdSe rod lengt h is 10 nm shorter than that distance due to the constant 10 nm capping layer. The cells showed an increase in VOC, JSC, FF, and efficiency as the cell thickness increased due to increased absorption in the devices. Howe ver, due to increasing se ries resistance, there was a diminishing return as the film thickness in creased. Interestingly, th e short-circuit current density continued increasing as the film thickness was increased to 1 m, despite P3HTs low hole mobility of 0.01 cm2/V-s. Figure 2-7 displays solar cell performance meas ures for the simulated J-V curves shown in Figure 2-6, with additional data points that were not displayed in Figure 2-6 for clarity. The short-circuit current density of the unit cells c ontinues to increase with cell thickness, although it begins to level off. The fill factor and VOC of the cells reach a maximu m in this simulation, with the value of the 1 m cell showing slightly lower values th an the 500 nm cell. The peak values are approximately VOC = 60 mV and FF = 0.36. The efficiency continues increasing up to the 1 m cell thickness, but as with the short-circuit curre nt, the rate of increase slows dramatically. A summary of the results from these s imulations is shown in Table 2-5. With the effects of varying cell thickness ch aracterized, simulations were then performed to determine the impact of nanorod width on cel l performance. Similar to the thickness variations displayed prev iously, these simulations maintain a constant P3HT thickness of 10 nm on the top and side of the CdSe nanorod. The resulting J-V curves are shown in Figure 2-8. The values displayed in the legend represent the nano rod half-width rather than the unit cell width. The unit cell width is 10 nm higher than the li sted value due to the constant 10 nm of P3HT on 35

PAGE 36

the side of the nanorod. The results show that performance increases with increasing nanorod width, which is counterintuitive. The absorption coefficient for CdSe is significantly lower than that of P3HT over most wavelengths, so the photocurrent generation increase with the wider cell is expected to be minimal. Additionally, the current density calculation involves dividing the total current by the cross-sectional area of th e cell, which increases as the unit cell width increases. One interesting feature of the simulation is th at nearly all the cell widths showed a shortcircuit current density of approximately 9 mA/cm2, with the exception of two. The curves for 15 nm and 20 nm CdSe nanorod half-widths showed a short-circuit current de nsity of approximately 6 mA/cm2. To further explore this strange behavior at 15 and 20 nm nanorod half-widths, a full range of half-widths was explored between 10 and 25 nm, with the resultin g J-V curves shown in Figure 2-9. The results show that the drop in s hort-circuit current dens ity is a collection of seemingly arbitrary thickness values that result in a reduced current density rather than a trend in the range from 15 20 nm of CdSe half-thickness.. The cells with half-thicknesses of 15, 20, and 23 nm showed a short-ci rcuit current density in the range 5.4 to 6.4 mA/cm2, the cell with a half-thickness of 10 nm resulted in a short-circuit current density of 8.5 mA/cm2 while all other cells were slightly above 9.0 mA/cm2. Figure 2-10 shows the collection of cell performance data generated through the simulations involving variations in the width of the CdSe nanowires, and this data is tabulated in Table 2-6. The odd behavior of th e short-circuit current density is a prominent feature in the graph, with the cells at 15, 20, and 23 nm ha lf-thickness showing a significant drop in JSC. This results in a drop in cell efficiency at these point s, from over 3% to approximately 2%. With the exception of these outlying data points, the shortcircuit current density shows a minor but steady 36

PAGE 37

decrease as the CdSe width grows from 5 to 30 nm. This is to be expected as series resistance and the voltage across the junction increase and the cell current is bei ng divided by increasing cell areas to calculate current density. The cells fill factors remain steady for CdSe half-widths between 10 and 25 nm. At 30 nm, the VOC become very large (> 0.9 V), which causes a significant drop in the fill factor (< 0.5). These values were unable to calculated directly, so they are not included in Table 2-6. Illumination source Simulations were performed to determine the ef fect of altering the illumination source. Medici generates photons from an illumination source at a specified location that illuminates the sample at a specified angle and beam width. The illumination used for the cell dimension study used the form PHOTOGEN RAYTRACE BB.RADI BB.TEMP= 5800 WAVE.START=0.2 + WAVE.END=1.0 WAVE.NUM=@WL X.ORG=0.006 Y.ORG=-2.5 ANGLE=90 + INT.RATIO=1E-2 N.INTEG=10 RAY.N=1 RAY.W=1.0 This code generates radiation from a bl ackbody source at 5800K, which simulates the AM0 radiation spectrum. The statements X.ORG= and Y.ORG= define the origin of the light source. In the example above, the source is 6 nm to the right of the origin (centered on a 12 nm unit cell) and 2.5 microns off the surface of the cell. The Y.ORG value is negative because the surface of the device is at y = 0, so the area above the top of the devi ce is in the negative y range. The light is incident at 90 on the su rface, with a ray width (RAY.W) of 1 m. The results of varying the coordinates of the light origin are shown in Figure 2-11. Three of the four data sets are identical, with onl y one showing a difference. The perfect match between the two simulations with a centered light source shows that the distance away from the surface is not a factor. Additionally, placing the source at the left edge of the cell shows no change. However, placing the light source at the right edge of the cell results in a total loss of 37

PAGE 38

photocurrent in the cell. The J-V curve show s the same exponential growth as the other simulations and matches the other curves at large voltages where the current is primarily driven by applied voltage rather than photogenerated carriers. The lack of photocurrent generation when the sample is illuminated from over the right edge is puzzling. According to the MEDICI ma nual, the ray will generate on each side of the specified origin [65]. In this simulation, that should mean that the beam will generate from a range of 0.5 m on each side of the origin. Because the cell is 12 nm wide, this should easily illuminate the entire cell. This is illustrated in Figure 2-12. Because the angle of incidence specified is 90, generated rays wi ll be perpendicular to the top su rface of the cell. As shown in the illustration, the beam width of 1 m would be more than sufficient to illuminate the 12 nm cell width. The simulations shown in Figure 2-11 were performed a second time with the input file modified to remove the RAY.W statement. This allows MEDICI to set the beam width, which by default is 2x the greatest dimension of the subs trate [65]. The results of this simulation are not shown because they were id entical to the curves shown in Figure 2-11. There was no photogeneration when the illumination source was placed at the right edge of the unit cell. The cause of this phenomenon is uncl ear, but it is obvious that placing the beam origin at the right edge of the simulation area results in a total lack of photocurrent. Summary These preliminary simulations serve as a guide for how to appropriately define commands in Medici, as well as to create a basic understanding of the effect of certain model properties on the cell performance. It was found that impurity profiles should be defined using the REGION command to ensure the desired dopin g density is applied in each region. It was also found that the illumination source should be pl aced in the center of the cell, but that the distance from the 38

PAGE 39

surface of the cell and the definition of the ray wi dth are unimportant, provided that the ray width is high enough to illuminate the entire cell. Further studies will focus on the simulation of a realworld hybrid solar cell from literature. Simulation of a Real Cell Model Parameters The cell chosen for Medici simulation was repor ted in 2006 by Olson et al. [75]. The cell was a hybrid cell fabricated from an array of ZnO nanofibers and P3HT, and has a cell structure of ITO/ZnO/ZnO:P3HT/Ag. The ZnO layer was spin-coa ted onto the ITO substrate and the nanofibers were grown thr ough hydrothermal growth. P3HT was then spin-coated at a reported thickness of 200 nm. SEM images of the ZnO nanofiber array before and after P3HT spincoating are shown in Figure 2-13. The performa nce measures reported for the device were as follows: VOC = 440 mV, JSC = 2.2 mA/cm2, FF = 0.56, and = 0.53%. The J-V curve for this device is shown by the solid line in Figure 2-14. From the SEM images shown in Figure 2-13 (a), the average height of the ZnO nanofiber array is approximately 260 nm. From Figur e 2-13 (b), the thickness of the full ZnO:P3HT layer is approximately 430 nm. These values leave a P3HT capping layer of approximately 170 nm on top of the nanofiber array. Th e image in Figure 2-13 (a) shows an average rod diameter of approximately 30 nm. The authors note the rod spacing is approximately 100 nm. It is extremely difficult to accurately estimate the spacing from the cross-section SEM image shown in Figure 2-13 (a), but this figure appears to be reasonable and was likely verified with other unpublished data, so it will be assumed to be accurate. From the SEM image, a thin base coat of ZnO is visible on the ITO film. This film wa s estimated to be 25 nm thick from the SEM images, and this height was in cluded in the 260 nm rod length. 39

PAGE 40

The authors note that the thickness of their P3HT film was 200 nm, but are somewhat unclear whether this 200 nm thickness is measured from the top of the nanof iber structure or if it represents the full film thickness. If the scale bars for the SEM images are reliable, it appears that the 200 nm figure refers to the excess P3HT film on top of the fiber array. This seems to be an excessive amount of capping considering that the exciton diffusion length of P3HT is on the order of 10 nm in P3HT [15], a fact which is referenced by the authors. However, light is incident from the ITO/ZnO side of the cell in the reported device, so photons must pass through more than 200 nm of ZnO:P3HT film before reaching this capping layer. Additionally, the inconsistent length of the ZnO nanofibers must be considered, as the rela tively high mobility of both electrons and holes in ZnO compared to P3HT would create a short-ci rcuit in the device if the wire tips were exposed. With this consider ation, the additional buffer layer thickness may be important experimentally to ensure that the device will function as a diode. The J-V curve shown in Figure 2-14 was conve rted to numerical data using a graph digitizer program [76] so it coul d be compared to simulated curv es. The converted J-V and P-V curves are shown in Figure 2-15. To verify th e effectiveness of this conversion, the published cell performance measures were compared to th e performance measures calculated from the digitized curve. The results are show n in Table 2-7. The results were VOC = 0.44 V, JSC = 2.2 mA/cm2, FF = 0.57, and = 0.55%. This shows exact matches for VOC and JSC and values for FF and that are 0.01 and 0.02 higher than the published results. This is less than a 5% error in both cases, and demonstrates that the curve was re-produced with a high accuracy. This new curve can now be plotted with the results of simulated cells to find a best fit. From previous work performed by Dr. Woo K young Kim, the material properties for ZnO were defined as shown in Table 2-8 [66]. 40

PAGE 41

The unit cell used for these simulation effort s is shown in Figure 2-16, consisting of half the width of a nanorod and half of the space between the nanorod in the x direction and the full cell thickness in the y direction. The unit cell dimensions were set based on the published cell dimensions observed in Figure 2-13. In the unit cell, the transparent cond ucting oxide of ITO is shown as the top contact. There is a thin 25 nm layer of ZnO co ating the ITO electrode, and the nanorods extend out from this structure. The P3HT region exists beside and above the nanorod. The back electrode of Ag completes the structur e. From the SEM images in Figure 2-13, the nanorod width was approximated as 30 nm and the average rod spacing was approximated as 100 nm. Because the unit cell consis ts of half of a nanorod and ha lf of a rod spacing distance, the rod width shown in the unit cell is 15 nm and the P3HT width is 50 nm, resulting in a unit cell width of 65 nm. The model contains two regions of P3HT, a photocurrent generating region and a nongenerating region. For convenience, the region shown in Figure 2-16 c) will be listed as P3HT1, and the region is Figure 2-16 d) will be listed as P3HT2. The P3HT1 region is the region within one exciton diffusion length (LD) that genera tes photocurrent. Any references to the P3HT1 region, the LD region, or the P3HT absorbing region refer to this area. The P3HT2 region is outside of the diffusion length region, and c onducts holes to the Ag elecrode but does not contribute to photocurrent in the cel l. This division is included to account for the effects of the exciton diffusion length in the polymer. The P3HT2 region is interchangeably referred to as the P3HT bulk region. The effective diffusion length (LD) is a parameter that can be adjusted in the simulations, but has been shown experimenta lly to be on the order of 10 nm [15]. Materials properties will be adjusted in the non-generating region to prevent photocurrent from originating in this region. Several pote ntial adjustments will be considered, including 41

PAGE 42

reduction of the electron mobility to zero, setting the electron free carrier lifetime to zero, or setting the absorption coefficient in the regi on to zero for all wavelengths. For simulation purposes, it is assumed that the ITO electrode is perfectly transp arent and that the Ag electrode has a reflectance of 90%. The unit cell structure as seen in Medici is shown in Figure 2-17. In this image, the blue region is ZnO, the red region is the P3HT1 region and encompasses all areas within 10 nm of the ZnO, and the green region is the P3HT2 region which includes all areas outside of 10 nm from the ZnO. Note the difference in scale between the xand yaxes. Initial Simulations Simulations were performed using the mate rials properties found in Table 2-1 for P3HT and Table 2-8 for ZnO. The J-V curves for th ese initial simulations are shown in Figure 2-18, along with the curve for the literature cell. These simulations de monstrate the effect of some variation in the properties for the P3HT2 region. In simulation tsf496_1, the absorption coefficient for the P3HT2 region is set to zero for all wavelengths. In simulation tsf496_2, no changes are made to the film properties, so all parts of the P3HT region are allowed to generate and transport free carriers. In simulation ts f496_4, the electron mob ility is set to 0.0001 cm2/Vs in the P3HT2 region, representing a reduction by one order of magnitude from the original value. The simulation tsf496_2 is a case where the entire P3HT region is treated as a photocurrent generating layer. This represents a case wher e the exciton diffusion lengt h is infinite. This should represent the maximum possible performance of a cell constricted by the materials properties shown in Table 2-1 for P3HT. The efficiency of that simulated cell was 5.55% with a short-circuit current of 11.66 mA/cm2. The simulation data range was stopped at 0.9 V to limit the size of output files and the simulati on time, so a direct measurement of VOC was unable to be obtained. From the shape of th e curve, it appears that the VOC would be approximately 1.0 V. 42

PAGE 43

Those numbers give an estimated fill factor of approximately 0.5, which is relatively low considering that this repres ents a best case scenario. In simulation tsf496_4, the electr on mobility is reduced by an order of magnitude in the P3HT2 region. This was an attempt to allow real istic absorption in the region but strictly limit the ability of the electrons generated in this area to reach the ZnO and contribute to the overall photocurrent. The resulting J-V curve showed a strange double-curve behavior, with the maximum power point occurring in the sec ond elbow at approximately 0.75 V. The doublecurve behavior is not physically r ealistic, but it is interesting to note that the performance of this simulation was higher than the performance of ts f496_1, showing that even with the low electron mobility, current generated in the P3HT2 region was able to make a contribution to cell performance. Simulation tsf496_1 shows the case where the abso rption coefficient is set to zero in the non-generating P3HT2 region. Of the three curves shown in Figure 2-18, this is the most similar to the real curve. Although the VOC and JSC were significantly higher th an that of the real cell, the shape of the curve seems to be very similar. The current density in creases very slowly over the low-voltage regime until approximately 0.6 V, wh ere there is a distinctive elbow in the curve and the current begins increasing more rapidly. The slope of the curve as it approaches VOC appears to approximate th at of the real curve. Reflectance of Ag electrode Because the incident light angle is set at 90 in simulation tsf496_1 and the boundaries between layers are set at right angles, one physical inconsistency in this model is the amount of reflected light that returns to the P3HT1 and ZnO regions. On the le ft-hand side of the cell where the nanorod resides, reflected light in the real cell would need to pass through 320 nm of the nonabsorbing P3HT2 region between its first and s econd passes through the absorbing P3HT1 region. 43

PAGE 44

On the right-side of the cell, light would ha ve to pass through 790 nm of the non-absorbing P3HT2 region before returning to the thin absorbing layer of poly mer and ZnO at the surface of the cell. Based on this, it is obvious that this si mulation allows more light to be absorbed in the generating regions of the model th an would exist in reality. Th is was corrected by studying the effect of reducing, reflection at the back electrode. The absorption coefficient of P3HT ranges from ~ 2 x 104 cm-1 at < 350 nm and > 680 nm to a maximum value of ~ 2 x 105 for = 540 nm. At the low-end absorption value, approximately 20% of incident light is absorbed over the 10 nm LD region, according to Equation 2-1. xe I I0 1 (2-1) I1 represents the amount of light transmitted through the film, where I0 is the intensity of incident light, is the absorption coefficient, and x is the film thickness. Nearly 99.8% of the remaining photons would be absorbed over the 320 nm path length consisting of the forward and backward pass through the non-absorbing region in the si mulation. Even in this case, with a low absorption coefficient and the shortest considered path length, virtually no photons would remain in a real cell to be absorbed afte r reflection from the back electrode. With this consideration, the next line of simu lations was performed with reflectance at the Ag electrode either reduced or completely removed. The resulting J-V curves are shown in Figure 20. Despite the removal of all reflec tance from the Ag electrode, simulated cell performance was still substantiall y higher than that of the real cell for all simulations. All of these simulations prevent absorption in the P3HT2 region. By comparing simulations with varying electrode reflectance and constant mobility and absorption properties (mobilities as defined in Table 2-1, absorption set to zero in the P3HT2 44

PAGE 45

region) we can see the effect of reflectance at the Ag electrode with no absorption in the P3HT2 region. These curves are shown in Figure 2-19. As expected, the short-circuit current density decreases as the Ag reflectance is decreased. Th e simulation with the Ag reflectance set to zero results in a JSC = 3.6 mA/cm2, which is still approximately 1.5x hi gher than the real cell value of JSC = 2.2 mA/cm2. Mobility In addition to the removal of reflectance from the back electrode, the charge mobilities in P3HT were reduced to restrict curre nt flow in the device. The physical justification for this is that the initial value of hole mobility was take n from the literature [71], and was the highest reported value for hole mobility in P3HT. Additionally, this value is a field-effect mobility which relies on a strong applied bi as to drive current flow. In photovoltaic devices, biases are significantly lower, and this field-effect mobility may be a serious over-estimation of the true film properties. Also, charge mobility has been shown to be anisotropic in P3HT [77], so values of mobility can vary depending on the alignment of the polymer chains in the film. Based on these reasons, simulations were performed to evaluate th e effect of reducing mobility values in the polymer regions of the mo del, with the resulting J-V curves shown in Figure 2-20. As predicted, simulated cells with higher mobilities resulte d in stronger device performance. The black and green curves, corres ponding to no change in mobility values and a 50% reduction of mobility values in the P3HT2 region only, are nearly id entical. A change from the initial mobility values to a 50% reduction in both P3HT regions, shown by the black and red curves, results in an efficiency drop from = 2.12% to = 2.00%, or approximately 6%. Reducing the P3HT mobility b 90% in both regions results in an efficiency drop of approximately 23%, from = 2.12% to = 1.64%. Attempts to reduce the P3HT2 mobility 45

PAGE 46

values by 90% while holding the P3HT1 values at their original values failed to converge in Medici. Although the J-V curves generated on this gr aph stop at a value of 0.9 V before reaching the open-circuit voltage, extrapolations estimate that the VOC is virtually unchanged in these simulations, with a value of approximately 1.04 V. Using this estimation, cell performance measures were calculated and are displayed in Table 2-9. The short-circuit current density decreases with each drop in the mobility. For the original mobility values, JSC = 3.61 mA/cm2. For 50% and 90% reductions in mobility, the JSC drops to 3.58 mA/cm2 and 3.44 mA/cm2, respectively. Using the extrapolated value of VOC = 1.04 V, fill factors for each cell can be calculated as 0.57, 0.54, and 0.46 for mobility values of 100%, 50%, and 10%, respectively. This reduced fill factor is obvious from Figur e 2-19, as the curve corresponding to a 90% reduction in mobility shows an obvious reduction in the sharpness of the elbow shape of the curve. Based on these values, a 50% decrease in mobility resulted in a 6% drop in efficiency, a < 1% drop in JSC, and an estimated 5% drop in fill factor. The 90% decrease in mobility resulted in a 23% drop in efficiency, a 5% drop in JSC, and an estimated 19% drop in fill factor. Exciton diffusion length Even with a reduction in charge mobilities by 90% in the P3HT regions, complete removal of absorption in P3HT outside of a diffusion length, and elimination of all reflection from the silver electrode, simulated cell performance greatly exceeds that of the real cell. The simulation with the lowest performance showed a JSC of 3.44 mA/cm2 and of 1.64%, along with estimated values for a VOC of 1.04 V and a fill factor of 0.49. To match the published experimental data, the JSC must be reduced by another 1.2 mA/cm2, the VOC by 0.6 V, and the efficiency by 1.1%. The exciton diffusion length was adjusted dow nward from the literature value of 10 nm [15] to compensate for this inconsistency. As noted previously, the ex citon does not exist in 46

PAGE 47

Medici, so the definition of an exciton diffusion length in this study is ta ken to be the width of the P3HT1 region. Since this is the only P3HT region generating photocurrent in the model, it is taken to be an accurate approximation of the exci ton diffusion length in a real cell. For this study, with results shown in Figur e 2-21, the absorption coefficient was again set to zero in the P3HT2 region and the carrier mobilities in both P3HT regions were set to th eir original values of n = 0.001 cm2/Vs and p = 0.01 cm2/Vs. As predicted, the simulated cell performance decreased as the exciton diffusion length decreased. This was a re sult of decreasing short-circuit current density, which decreased linearly with exciton diffusion le ngth, as shown in Figure 2-20. Estimates for open-circuit voltage actually increase d slightly as the diffusion length decreased, as shown in Figure 2-22. This trend was very slight, however, so it showed virtually no impact on cell performance. Based on these simulations, an exciton diffusion length of 6 nm was chosen for future simulations. Doping density From previous simulations of CdSe:P3HT hybrid cells, changing the doping level of CdSe showed a strong impact on the VOC of the cells. Based on thes e results, the doping density of P3HT was varied from the initial value of 5 x 1016 cm-3 to 1 x 1014 cm-3 in an attempt to reduce the VOC of the simulated cells. The simulated J-V cu rves are shown in Figure 2-24. Contrary to expectations, the VOC of the cells did not appear to change with the reduction in P3HT doping. A key difference between this study and the simu lations performed on the test cell of CdSe and P3HT is that CdSe and P3HT showed very similar doping densities: 6x1016 cm-3 for CdSe and 5x1016 cm-3 for P3HT. The doping density of ZnO is 5x1017 cm-3, so it is possible that the VOC is dominated by this wide discrepancy. To test this hypothesis, simulations were performed with reduced ZnO doping density. These tests fixed the P3HT doping density at 1x1016 cm-3, based on the results shown in Figure 247

PAGE 48

24. The results from this test are shown in Fi gure 2-25. Oddly, this variation fails to show a decrease in VOC. The VOC of a cell should depend on fact ors such as doping density and minority carrier lifetime [78], but the doping density variations here doesnt show any impact. Density of states Simulations were performed to determine the effect of the density of states in the conduction and valence band of P3HT, and the resulting J-V curves are shown in Figure 2-26. The curves show a decrease in performance as the density of valence band states increases, although extrapolated values for open-circuit voltage remain relatively constant. The curves are independent of the density of conduction band stat es, as shown by the overlap of the black and red data points and the green and yellow data points. Open-circuit voltage examination Simulations focused on the effect of carrier mobility were again performed, but with an expanded voltage range so that th e open-circuit voltage could be observed rather than estimated. These resulting J-V curves from these simula tions are shown in Figure 2-27. As predicted through previous extrapolations, the open-circuit voltage is independent of the carrier mobility and the short-circuit current de nsity is strongly impacted by it. Extrapolated values for VOC in previous simulations predicted values near 1.04 V, but these simulations show that VOC = 1.27 V for these cells. As noted earlier, the cells s how a double-elbow effect that was unable to be explained. The effect of P3HT doping density was also reexamined with the expanded voltage range, as shown in Figure 2-28. As predicte d through previous ex trapolations, the VOC is independent of the doping density of P3HT. Again, the VOC = 1.27 V for these simulations, and the shortcircuit current density decreases as the P3HT doping density decreases. This mimics the behavior seen in previous studies on this material system. 48

PAGE 49

Although the short-circuit current density and fill factor can be altered by variations in carrier mobility, doping density, and vale nce band density of states in P3HT, these tools offer no control over the open circuit voltage of the simulated cells. From a theoretical perspective, the open circuit voltage of organic a nd hybrid cells is controlled by the energy difference between the HOMO level of the electron donor and the LU MO level of the electron acceptor, or the conduction band level if the acceptor is an inorga nic [79]. In the case of a hybrid cell, the LUMO level of the organic el ectron acceptor is replaced by th e conduction band energy of the inorganic material. In other words, carriers do no t exist with energy equal to the band gap of the material where they are generated; instead they exist with energy equal to the spacing between the bands of the junction materials. The energy band diagram for the P3HT ZnO system displayed in Figure 2-29 shows th is band offset to be 0.85 eV. Based on this, free carriers resulting from photoabsorption in P3HT should exist at an energy of 0.85 eV rather than 1.7 eV. Medici allows independent specification of the energy band gap and optical band gap in each material. The optical band gap should remain fixed at the appropriate level to dictate the location of the absorption cutoff for each materi al. The energy band gap, however, presents an additional tuning control to desc ribe the energy of photogenerated carriers in the device. Figure 2-30 shows the results of simulations with variations in the energy band gap of P3HT. In the curve tsf496_45, the energy band gap is set to 0.85 eV in the P3HT regions of the cell. In curve tsf496_53, both th e energy band gap of both the P3HT and ZnO regions is fixed at 0.85 eV. From these curves, fixing the energy band gap of P3HT accurately simulates the VOC of the literature cell. Altering the ZnO band gap in addition drops the VOC to below 0.4 V, beyond the published cell performance data. 49

PAGE 50

Absorption coefficient adjustment Under close inspection of a Medici output file, it was noticed that the program automatically assigns absorption coefficient va lues for wavelengths outside of the range specified in the input f ile if this range is smaller than th e full spectrum used for calculation. These values are not set to zero, but assigned a default value based dependent on the wavelength. Figure 2-31 shows the absorption coefficient curves as assigned by Medici for three absorption input files. The first two files, abs_none and abs_p3ht, have an input range specified from 0.3 to 0.7 m and correspond to zero absorption and the absorption profile of P3HT, respectively. The range of 0.3 to 0.7 m was chosen because this is the range of data shown in the absorption curve from literature [74]. The file abs_i_ zno was borrowed from Dr. Woo Kyoung Kim in his simulations of CIS solar cells [65]. This input file ranges from approxima tely 0.2 to 1.0 m. From Figure 2-31, it is obvious th at Medici assigns identical va lues to all absorption files in the range from 0.12 to 0.2 m. Similarly, it assi gns an identical data point to all files at 1.24 m. Although absorption in the high wavelength region is set to zero by the program, absorption in the short wavelength region is set to a very high value in the range of 106 cm-1. Although the photon flux is low in this region, as shown in the AM1.5 spectrum in Figure 2-32, this extremely high absorption coefficient leads to a significant amount of carrier generation in the cell. This leads to an over-estimation of th e number of free carriers in the ce ll and artificially enhances cell performance in the simulations. This effect ha s no impact on the ZnO absorption, because in the simulation photons are generated from 0.2 to 1.0 m. Figure 2-33 shows the number of photogenerated carriers as a function of wavelength. Figure 2-33 shows the carriers generated by this absorption region between 0.2 and 0.4 m, but does not show the exact effect on the J-V curves. In Figure 2-34, J-V curves are compared for cells simulated with the P3HT absorption profiles shown in Figure 2-31 and an 50

PAGE 51

absorption profile with all absorption between 0.2 and 0.3 m set to 2.24x104 cm-1. The curve with the corrected ab sorption coefficient, shown in blue, shows a drop of 0.14 mA/cm2 in the short-circuit current when compared to the curv e with Medici-assigned va lues for the absorption coefficient, shown in purple. This difference is minimal, but should be included in simulations for improved accuracy in the simulations. Multi-stage absorption The technique used for simulations up to this point involved a 2-layer P3HT region with a thin absorbing region near the ZnO region and a large non-absorbing region. The width of the absorbing region was set equal to the assumed diffusion length of an exciton. Because the diffusion length of an exciton is not explicitly included in the program, Me dici assumes that all photogenerated carriers within this absorbing region are free car riers, which are driven across the space-charge region by the built-in el ectric field. This is only a first-order approximation of the true device physics of a hybrid solar cell. The exciton diffusion length is an average distance that an exciton can travel before it recombin es, not a firm line beyond which all excitons are doomed to recombination. The excitons do not se lectively move toward the inorganic organic interface, they simply travel by diffusion until they recombine or dissociate at the interface. To approximate this distinction, the model was modified to include multi-stage absorption in the P3HT region where the probability of exc itons reaching the interface is included. A framework for a simulation involving multip le layers in the absorbing region was developed with a multi-layer region extending to 20 nm from the ZnO region. J-V curves shown in Figure 2-35 correspond to simulations in which the first 10 nm of this region is set as an absorbing region using the P3HT absorption profile. The second 10 nm is set as a non-absorbing region, identical to the P3HT2 region. These cells all have identical structures, with the only difference being the number of stages within this 20 nm region. For example, the simulation 51

PAGE 52

with 10 stages has 10 layers that are each 2 nm th ick. The first 5 stages are set to absorb based on the P3HT absorption profile, and the final 5 stages are set as non-absorbing. Similarly, the simulation with 4 stages contains 4 layers with 5 nm thickness in each layer, with the first two set to absorb and the final two set as non-absorbing. In the case of the single stage cell, the absorption region was set from 0 to 10 nm away from the ZnO interface, and the P3HT2 region began at 10 nm. Because the cell structures ar e identical in terms of exc iton diffusion length, simulations were expected to produce identical J-V curves. Although the curves are all similar in shape and performance data, they are not identical, due to slight differences in the iterative calculation in Medici. In all the simulations shown in Fi gure 2-35, absorption was set to 100% of the P3HT value in all stages betwee n 0 and 10 nm into the P3HT region, and it was se t to 0 in all stages between 10 and 20 nm. This result is interesting, but is not the goal for this multi-stage model design. The purpose is to create a graded carrier generation profile that will average to the sp ecified LD, but allow for some carriers inside of LD to fail to reach the interface and allow some carriers outside of LD to succeed in reaching the interface. The structures used to generate the J-V curves in Figure 2-35 can be thought of as normal curves with standard deviations of zero. Ab sorption is 100% of the P3HT value up to the specified diffusion length of 10 nm, and then drops as a step function to 0% of the P3HT value for the next 10 nm. Using this multi-stage framework, distribution curves with nonzero values for standard deviation can be imposed on the simulation. In this case the property being distributed is the ab sorption coefficient, as this is the property chosen to simulate the exciton diffusion length. It should be noted that these curves are not exactly equal to cumulative distribution functions for a true nor mal distribution because in this case the area 52

PAGE 53

considered was only between 0 and 20 nm. In a true normal distribution with a mean of 10 nm and a standard deviation of 4 nm or greater, an appreciable area exists under the curve beyond 0 and 20 nm. To remove this area from consid eration, normalization was performed by dividing by the sum of the area between 0 and 20 to esta blish these boundaries. The distribution was calculated as shown in Equation 2-2. 2 22 exp 2 1 xx xF (2-2) In Equation 2-2, x is th e distance into the P3HT region and x is the mean of the distribution, set to 10 nm in this case. By dividing this by the sum of all distribution values between x = 0 and x = 20 nm, the normalized distribution was obtained. The cumulative distribution was calculated using Equation 2-3. 20 0 0)( 1x x x xxF xF xG (2-3) Again, this cumulative distribu tion is normalized by the area under the distribution function between 0 and 20 nm rather than between nega tive and positive infinity. This cumulative distribution function is shown in Figure 2-36 for a range of standard deviations and a mean of 10 nm. By coupling these distribution functions with the multi-stage cell design demonstrated in Figure 2-35, graded absorption can be genera ted in the simulation by applying a fractional absorption coefficient in each stage of the absorption region. Simulations were performed to consider the effect of this absorption distribution in the cells. These simulations used a 10-stage absorp tion region with a graded absorption coefficient in an attempt to more accurately define the absorption in the cell. Absorption profiles corresponding to standard deviations of 0 to 10 nm were generated by using a fractional value for 53

PAGE 54

the absorption coefficient in each region, rounded to the nearest 10% value. The input data used for the simulations are shown in Table 2-11, as we ll as the resulting short-circuit current density for that simulation. The J-V curves for th ese simulations are shown in Figure 2-37. Table 2-11 shows that absorpti on in the cell is distribute d over a wider range as the standard deviation increases. The average multiplier to the absorption coefficient is calculated for each situation and as targeted it is equal to 50% or 51% for all cases, with any error coming from rounding to the nearest 10% in each region. The J-V curves in Figure 2-36 show that this grading has virtually no effect on the final performance of the cel ls. The short-circuit current density shows extremely minor variations in the cu rves, and this value is tabulated in Table 2-11. The short-circuit current density increases slightly as the absorption distributions become wider. This shows that extending the generated carrier distribution toward the back electrode improves device performance, while limiting the carrier di stribution to a narrower region, even with the same number of carriers being ge nerated, hinders performance. P3HT replacement with CIS The J-V curves generated in previous simula tions shows a double-elbow shape that is not characteristic of photovoltaic cells. To determin e the root cause of this phenomenon, P3HT was replaced with copper indium diselenide (CIS), a popular p-type thin-film photovoltaic material. The same cell structure is used, with ZnO nanor ods as the n-type material. Within this framework, material properties of the CIS laye r were adjusted to values corresponding to P3HT. Preliminary simulations in this form were perf ormed, and the resulting J-V curves are shown in Figure 2-38. This set of simulations compares ZnO:P3HT cells with 10 nm and full cell absorption regions to ZnO:CIS cells with full ce ll absorption, a 10 nm absorption range, and a 10 nm absorption range with an energy band gap set at 0.85 eV. 54

PAGE 55

Figure 2-38 shows that a limited absorption ra nge in the simulations does not cause the double-elbow shape of the J-V curve, as the Zn O:CIS cell with a 10 nm absorption region shows a similar shape to the one with absorption in the full CIS region. The same holds true for comparisons of the two ZnO:P3HT cells, which both show the double-elbow shape. The ZnO:CIS cells produce J-V curves with the anticipa ted shape and high fill factor as compared to the ZnO:P3HT cells. This also held true for the cell in which the CIS electrical bandgap was reduced to 0.85 eV, although this cell resulted in a low VOC of 0.25 V. Parameters used for P3HT and CIS in the simulations are displayed in Table 2-12. Data for CIS was obtained from previous simulations fr om Woo Kyoung Kim [66]. The doping density is the only parameter that is the same in both mate rials. By individually adjusting parameters between the values for CIS and P3HT, an attempt will be made to determine the origin of the double-elbow J-V curve shape. Although not shown in the table, the absorption profile for each material was also adjusted. Using the same 10 nm absorption region as th e ZnO:CIS cell shown in red in Figure 2-38, the properties shown in Table 2-12 were individually changed fr om the CIS values to the P3HT values. The resulting J-V curves are shown in Figure 2-39. Performance measures for these cells are displayed in Table 2-13. These results show that the permittivity and density of states have virtually no impact on the cell performance. The three parameters that strongly impacted the J-V curves are electron affinity, carri er mobility, and absorption profiles. The change in electron affinity causes a shift in VOC from 0.44 V to 0.49 V due to the higher conduction band level for P3HT. The short circuit current is minimally affected by this change, but there is a notable drop in the fill fa ctor of this curve, down nearly 30% from the values of the original ZnO:CIS cell. The va riation in carrier mob ilities showed a 1.5 mA/cm2 55

PAGE 56

reduction in short circuit current paired with a 1 V increase in open-circuit voltage. The reduced mobility results in fewer carriers escaping the cell, causing a lower short-circuit current density. The strong increase in VOC is more difficult to explain, as adjustments to carrier mobility in previous simulations did not show this type of response. There is also a s light decrease in the fill factor for the cell with P3HT mobility values, but this effect was minor. When the absorption profile is changed from CIS values to P3HT values, the short-circuit current density of the simulated cell drops by nearly 50%, down to 4.1 mA/cm2. The VOC of that cell is slightly reduced, but the fill factor remains nearly unchanged. Based on the J-V curves shown in Figure 2-38 and detailed in Table 2-13, simulation 72 was chosen as a basis for further efforts. In this simulation, the absorption coefficient values were assigned values corresponding to P3HT rather than CIS, while maintaining a 10 nm absorption range in the simulation. Variati ons around this base case were performed by individually varying the other material pr operties from their CIS value to their P3HT value. The results are shown in Figure 2-40, with performance measure data shown in Table 2-14. As in the previous set of simulations, the addition of P3HT values for permittivity or density of states showed virtually no change from the control cell. The values for open-circuit voltage, shortcircuit current density, and fill factor of these cells were within 2% of the values for the control cell. As seen in the previous set of simulations, when the P3HT value for electron affinity is applied the J-V curve shifts to a higher VOC with a 30% drop in fill factor and a virtually unnoticeable drop in JSC. The application of P3HT values for charge mobility results in a 0.8 mA/cm2 decrease in JSC and a 0.1 V increase in VOC with an approximately 10% drop in fill factor. Despite reductions in fill factor for simulated cells with P3HT values for electron affinity and mobility, none of the J-V curves in Figure 2-40 displayed the double-elbow shape. 56

PAGE 57

The electron affinity was fixed at its P3HT value of 3.15 eV, setting the conduction band level in the simulations to the appropriate value for P3HT. The electronic band gap of the simulations is still set at the CIS value of 1.04 eV rather than the P3HT value of 1.7 eV or the effective band gap of 0.85 eV at the ZnO-P3HT junction. Again, the abso rption coefficient is set at the P3HT value and absorption is a llowed over a 10 nm range near the material interface. Variations in the other materials properties resu lted in the J-V curves shown in Figure 2-41 and detailed in Table 2-15. As seen in previous si mulations, there is virtua lly no change in the J-V curve or device properties w ith the application of the P3HT values for permittivity and density of states. The application of P3HT values for carrier mobility, howev er, resulted in a drastic shift in the nature of the curve. This change resulted in a 0.5 mA/cm2 reduction in the short-circuit current density and an increase of ~0.6 V in the open-circuit voltage. Additionally, the J-V curve takes on the double-curve shape, which drops the fill factor to approximately 0.20. This shape will be discussed in more detail after the next set of data. As discussed previously, the J-V curves show n in Figure 2-41 resulted from simulations where the electron affinity was set at the P3HT value of 3.15 eV, but the electronic band gap remained at the CIS value of 1.04 eV. Although fi xing the electron affinity sets the LUMO level of P3HT, the band gap value sets an inappropriate HOMO level and results in the generation of carriers with higher energy than appropriate. The simulations of Figure 2-41 were modified to include the appropriate ZnO-P3HT interfacial band gap of 0.85 eV, corresponding to the energy gap between the conduction band of ZnO and the HOMO level of P3HT. The results, shown in Figure 2-42, mimic those in Figure 2-41. The curves resulting from adjustments in th e permittivity and density of states show the expected shape for a solar cell J-V curv e despite having a significantly lower VOC than their 57

PAGE 58

counterparts in Figure 2-41. This variation is ex pected, due to the shift in energy band levels causing the generated carriers to exist with more energy. The JSC of the cells remains virtually unchanged as compared to their counterparts in Figure 2-41, but the shift in VOC results in a slight reduction of the fill factor as the curves cross the voltage axis at a slightly lower slope. Curve 82, corresponding to a change in carri er mobility, also shows a reduction in VOC when compared the curve shown in Figure 2-41. This curve shows a very small fill factor of approximately 0.15 due to the double-elbow shape wh ich eliminates most of the active area of the curve. The origin of the change in shape accompanying lower mobility values is unclear, but it obviously occurs only when the ab sorption, electrical energy levels and mobilities take on their P3HT values. This shape was observed for all ZnO:P3HT cells simulated up to this point, but it was not seen in other simulations using a ZnO:CIS cell as a basis. From the graphs in Figures 239 2-41, it can be concluded can state that the permittivity and density of states have no effect in causing this shape. Additionally, the application of P3HT levels of absorption and energy levels to a ZnO:CIS cell did not cause this shape without the addition of P3HT levels for carrier mobility. In the simulations, the carrier mobilities were set as n = 0.001 cm2/Vs, p = 0.01 cm2/Vs for P3HT, and n = 30 cm2/Vs, p = 300 cm2/Vs for CIS. Changing from CIS values to P3HT values represents a severe drop of 4 orders of magnit ude with no further details for intermediate values. Using simulations of a ZnO:P3HT cell, a wide range of mobility values were examined, with the resulting J-V curves shown in Fi gure 2-43. Note that in all cases, p = 10*n. Additionally, these simulations a llow absorption throughout the full P3HT region of the cell to boost the level of current flow a nd more clearly show the effect of the mobility variations. 58

PAGE 59

These simulations demonstrate that at highe r mobility values, the J-V curve shows the expected shape for a solar cell. At lower mob ility values, the curve inverts to the double-elbow shape. An ideal solar cell J-V curve should ha ve a positive second deri vative over the entire active range, from V = 0 to V = VOC. It is difficult to calculate a second derivative for these curves because they are not easily fit to empirical equations. However, the sign and approximate magnitude of the second derivative over a short range of the data can be determined by using Equation 2-4. Jest = (y3 y1) / 2 y2 (2-4) A visual representation of this calculati on is shown in Figure 2-44. The values y1, y2, and y3 represent current density values for equally sp aced applied voltages. For these J-V curves, y is current density and x is voltage. The first te rm of Equation 2-4 is noted in Figure 2-44 as y2 0, and is the midpoint of a line drawn through the points (x1, y1) and (x3, y3). If this value is greater than the actual value of y2, Jest calculated from Equation 2-4 will be positive and will correspond to a curve that is con cave up, as shown in the figures. Although this is not an exact measure of the second derivativ e, it will result in a posit ive value (the value for y2 0 is greater than the value y2) for curves that are concave up and a negative value for curves that are concave down. Additionally, th e magnitude of Jest will give a relative idea of how close to linear the curve is over the range from x1 to x3. This calculation is applied to the simulated J-V curves for ZnO:P3HT cells shown in Figure 2-43, and the calculated Jest values are plotted in Figure 2-45. Note that these calculations correspond to cells with absorption in the full P3HT region. Of all the curves shown, only those corresponding to p = 5,000 and p = 10,000 cm2/V-s showed second derivative values over the entire active region. 59

PAGE 60

When these same simulations are performed on ZnO:P3HT cells featuring carrier generation only within a 10 nm exciton diffusion length, the curves shift significantly, as shown in Figure 2-46. The most obvious and expected feature is that the current density drops significantly due to the lower number of carriers being generated in the limited diffusion length region. There is also a slight drop in the open-ci rcuit voltage for these ce lls, but an increase in the fill factor as the curves show less of an inve rted shape. This is shown more clearly in Figure 2-47, which displays the Jest values for these simulated cells. Cells with mobility values as low as p = 500 cm2/Vs show positive Jest values for the entire active re gion in this case, a full order of magnitude smaller than the threshold mobility in the full cell absorption simulations. Lower values of carrier mobility generate JV curves with consistently positive second derivatives in the 10 nm absorption region simula tions as compared to the full cell absorption region simulations. This refutes the theory th at a second p-n junction region arises at the interface of the absorbing and non-absorbing P3HT regions. Instead, the charges generated at a greater distance from the ZnO-P3HT interface drive this inversion of the J-V curves. This is due to the low mobility of electrons in P3HT and their difficulty in traveling through larg e regions of the polymer to reach the ZnO regions. The comparisons of ZnO:CIS cells demonstr ated that three parameters are key to controlling the shape of the J-V curve in these simu lations. Carrier mobility dictates the shape of the curve, as demonstrated in Figures 2-43 and 246. Altering absorption in the cell, such as by changing the exciton diffusion length or the absorp tion coefficient, has a direct impact on the number of charge carriers and th erefore on the current density of th e cell. This is demonstrated in Figure 2-48 for cells with the P3HT hole mobility set to 500 cm2/V-s and a 10 nm absorption 60

PAGE 61

region. The energy band gap of the P3HT region in the cell alters th e open-circuit voltage of the cell. This is shown in Figure 2-49 for cells with p = 500 cm2/V-s and LD = 10 nm. The curves shown in the two figures above demo nstrate the level of independent control of Jsc and Voc that is available from variations in ab sorption and energy band gap, respectively. To accurately simulate the experimental data, however, an array of combinations must be examined. Figure 2-50 shows a set of curves spanning four values for each of these parameters. In the graph, colors are constant for constant band gap, while line style is constant for constant absorption coefficient values. The experimental data falls somewhere in the range between 20% and 50% absorption with a band gap of 1.0 eV. This set of curves was expanded to s how a wider range of absorption, all with the band gap held constant at 1.0 eV and the hole mobility set at 500 cm2/Vs. The results are shown in Figure 2-51. Although th e curve for 30% absorp tion produces a nearperfect match for JSC and a close match for VOC, the simulation produces a curve with a lower fill factor than the real data. Based on this, the search must be re-expande d to three parameters: carrier mobility to control fill factor, band gap to control VOC, and absorption coefficients to control JSC. However, it must be considered that each of these controls impacts all three target parameters, not just the intended one. Shortcomings of the initial model This simulation strategy was abandoned as the physics of the model were more closely analyzed. All simulations up to this point suffered from one distinct inconsistency the existence of electrons in the P3HT region. This is demonstrated in the strong impact of electron mobility on the shape of the J-V curve for ZnO:P3HT cells. The polymer absorbs photons to generate excitons rather than free carriers, but th e model is not aware of th is distinction. As a result, the electron mobility of P3HT is an important factor in th ese simulations, but it is of minor 61

PAGE 62

importance in real cells. Because of this dis tinction, a new strategy has been employed to more accurately represent exciton physics in Medici. Two-Step Simulation Technique The important feature of excitonic solar cells, as has been discussed, is the generation of excitons rather than electron-ho le pairs. One important consid eration, addressed in simulation efforts described previously, is the limited carrier generation due to the exciton diffusion length. However, another important effect, not directly c onsidered in previous s imulations, is that the exciton dissociates at the material interface. This means electrons should not be generated in the P3HT region; instead, free carriers should be generated directly at the material interface. A new modeling strategy is needed to incorporate this effect into the simulations. This new strategy involves a two-stage simula tion. In the first st age, the photogenerated carrier distribution is measured over the entire de vice area under simulated solar illumination. In the second stage, this distributi on is compressed to a line source of carriers generated directly at the organic-inorganic material interface by mapping the carriers generated at all points to their nearest interfacial point. Specifying a line source of carriers in Medici requires specifying th e origin and endpoints of the line with X and Y coordinates, as well as parameters to describe the carrier distribution along that line, as shown in Equations 2-5 2-8. (2-5) (2-6) (2-7) (2-8) 62

PAGE 63

Gn and Gp are generated electr ons and holes, respectiv ely. This electron-hole pair generation equation consists of th ree components: lateral, radial, and time-dependent. The physical dimensions l and r represent lateral distance along the line and radial distance from the line, respectively. The lateral dependence can be specified as an exponential-linear function that requires the definition of parameters A1, A2, A3, and A4. These parameters are determined through regression for each scenario studied, and are discu ssed in further detail in the following sections. The radial dependence follows an exponentia l decay using the decay constant R.CHAR, which is set to 0.0001 m for these simulations. This d ecays the radial component of the generation by 44 orders of magnitude within 1 nm of the line, so that the generation occurs only at the material interface and not in the individual materials. Carrier generation is assumed to be uniform in time. The time dependence is set to the default value of T(t) = 1. Photogenerated Carrier Distribution The distribution of photogenerate d carriers in the unit cell was determined through Medici. The unit cell was exposed to simulated solar illu mination as described in previous sections. Medici calculates photoabsorption and carrier generation at each point in the simulation mesh, and this data can be displayed as a contour pl ot. Unfortunately, it seems to be impossible to extract the numerical source data from contour plots in the program This is possible, however, for line plots, as this is the technique used in to extract J-V data from the simulations. Line plots were generated at each mesh point along the x-axis of the unit cell, and these lines extended through the thickness of the device so that the car rier generation data for every mesh point in the unit cell was collected. Using this (x, y, z) data set, co ntour plots were created in Sigmaplot to show the dist ribution across the device area. 63

PAGE 64

A contour plot of the photogene rated carrier distribution in the full cell and a surface plot focused on the area within LD from the side of the nanorod are shown in Figure 2-52. The exciton diffusion length for the cells depicted is 10 nm. Note that in the plots, y = 0 represents the top surface of the cell, bordering on the ITO electrode. There is a 0.025 m surface layer of ZnO, with the ZnO nanorod extending to y = 0.26 m and being 0.015 nm wide. This region appears as dark blue to purp le in Figure 2-52A, as ZnO is nearly transparent to most wavelengths. The 10 nm absorbing region of P3HT appears as the brightly-colored region in Figure 2-52A, and is the focus of Figure 2-52B Although absorption is strong in the strips above the nanorod tip and just beyond the ZnO base layer, a large portion of the carrier generation in the unit cell occurs in the area on the side of the nanorod. It is interesting to note from Figure 2-52B that the strong absorption fr om the region above the nanorod tip extends slightly into the area to the side of the nanorod. It is unclear from simulations if this occurs due to refraction at the interface or if it simply a numerical anomaly to facility convergence. The P3HT bulk region is seen in blue in Figure 2-52A. Although the ab sorption coefficient is set to zero in this region, a low leve l of carriers still exist. Line Source Generation Line sources of carriers we re imposed along the ZnO-P3HT interface, consisting of three lines due to the shape of the in terface. The first line is at the tip of the ZnO nanorod. For simplicity, this is noted as Line 1. The line ex tending along the edge of the nanorod is referred to as Line 2, and the line running along th e top of the ZnO base layer is Line 3. To determine the density of photogenerated carri ers to be imposed along these line sources, all carriers generated by the simulated solar radiation are mapped to their nearest ZnO-P3HT interface point. This mapping scheme is illustra ted in Figure 2-53. Gray areas represent ZnO, while blue areas represent P3HT. Carriers in a particular region are directed to the nearest line as 64

PAGE 65

illustrated with the orange arrows. Dotted white lines show the edges of regions that map to a specific line. As one of these lines is crosse d, the carriers begin mapping to a different line. There are two blocks, in the top-left and bottom-ri ght corners, that map to a single point at the intersection of two lines. Once the carrier densities from each mesh point in the unit cell are mapped to their corresponding point along one of th e three lines, these carrier dist ributions are fit to exponentiallinear curves to determine the carrier vs line distance profile along each line. The mapping process is identical for all simu lations using the same unit cell. In other words, the exciton diffusion length does not imp act the mapping process. Because carriers are mapped to the nearest line, the (x, y) coordinate s of the point are the onl y factor not how many carriers are generated there. However, change s in the exciton diffusion length do impact the number of carriers mapped to each point on the li nes, and therefore impact the shape of the carrier distribution along those lines. To clarify the mapping process, Figure 2-54 displays the photogenerated carrier distribution for a simulated cell with an exciton diffusion length of 20 nm. The dotted lines shown in the illustration of Figure 2-53 are overlayed on this contour plot, seen as dashed white lines. Line 1, Line 2, and Line 3 are shown as solid white lines and represent the borders between the ZnO and P3HT regions. Note that the xand y-ax es are not on the same scale. The diagonal dashed lines on the plot are at 45 a ngles through the unit cell, although scaling makes them appear to be less sloped. The behavior of the three carrier generation lines depends on the ex citon diffusion length of the simulated cell and the lateral di stance along the line. For Line 1 (x 0.015 m, y = 0.26 m), the carrier distribution is nearly constant. As seen in Figure 2-54, this line collects carriers 65

PAGE 66

from the P3HT region beyond the tip of the nanorod, as well as a small area within nanorod. This area inside the nanorod decreases as the x-coordinate increases because portions of this region begin mapping to Line 2. This decrease results in a negligible decrease in the number of carriers, however, because the ZnO abso rbs poorly in comparison to the active P3HT region included in the calculation. This effect holds true for any exciton diffusion length that is considered in this study, because the P3HT region will always absorb much more strongly than the ZnO region. Due to this, the carrier distributi on for Line 1 is approximated as a constant for all simulations. Along Line 2 (x = 0.015 m, 0.025 m y 0.26 m), photogeneration increases steadily at low values of y because a larger section of the absorbing region is included at each point. Seen in Figure 2-54, this occurs for approximately one LD, from 0.025 y 0.045 m. Beyond that point, the new regions being added are not absorbing regions, so even though the size of the area being summed is larger, this larger area is not contributing a significant number of carriers. This is compounded by the fact that the LD region at this greater depth produces a lower number of carriers due to a drop in the number of photons remaining in the cell. This exponential decay of absorption becomes the dominant contribution to the number of carrie rs along the line, and continues to the tip of the nanorod. The point where the distribution along Line 2 turns from linear growth to exponential decay depends on the exciton diffusion length of the simulated cell. For Line 3 (x 0.015 y = 0.025 m), the number of photogenerated carriers initially shows a linear increase with x for the same reasons as Line 2. As the lin e is traversed, additional area is being mapped to this line, which increases the number of carrier s contributed. After a distance of LD along the lin e, the distribution becomes nearly c onstant. This occurs because the 66

PAGE 67

new area being included in each summation contai ns relatively few carriers due to lack of absorption in these regions. The total amount of photogeneration in the simulated cell is dependent on the exciton diffusion length specified in the simulation. Th is parameter defines the size of the stronglyabsorbing area in the P3HT polymer. Figure 2-55 displays the total count of photogenerated carriers for simulated cells with varying excit on diffusion lengths defined from 10 nm up to the full polymer region of the cell. In addition to dictating the to tal number of carriers in the ce ll, variations in the exciton diffusion length also change the distribution of these carriers. Figure 2-56 displays the line sources used in Medici to simulate photoabsor ption based on the model described previously. Note that the curves seen in Figure 2-56 are not the actual summations of carriers calculated from the contour plots, they are the fit lines a pplied to Medici with the form shown in Equation 2-6. As evidence of the quality of the fit, Figur e 2-56C displays the true carrier summations for Line 2 of a cell with LD = 20 nm, along w ith the fit curve applied in Medici. As desccribed previously, photogeneration in Line 1, for 0 x < 0.015 m, is set as constant in all cases. In Line 2, shown in Figue 2-56B, carrier genera tion increases nearly linearly until a certain point wher e the exponential decay becomes th e dominant effect. For Line 3, carrier generation again incr eases nearly linearly for x 0.015 m before reaching a point where it becomes approximately constant. Unlike Line 1, the nearly-const ant region for Line 3 was not assumed to be exactly constant, and a ll parameters for the e xponential-linear equation were calculated. For the 10 nm and 20 nm cases, this region shows a lower slope than in the 40 nm and full cell cases. 67

PAGE 68

Interestingly, the curves in Figure 2-56 s how regions where the carrier generation is higher for the 40 nm scenario than for the full cell scenario. This effect is observed to a very small degree near x = 0.035 m in Figure 2-56A and more noticeably near y = 0.06 m in Figure 2-56B. This is not simply an inconsistency due to curve fitting, this shift occurs in the absorption profiles that were summed in the cells. To clearly see the regions where the 40 nm a nd full cell simulations produced inconsistent carrier generation, the generation pr ofile for the 40 nm absorption case was subtracted from the profile for the full cell case, and contour plots of this difference are shown in Figure 2-57. In the figure, positive values represent areas wher e the full absorption simulation produced more carriers than the 40 nm simulation. Figure 2-57A shows that the positive-valued regions have extremely large values compared to the negative-valued regions, accounting for the increase in total absorption in the cell. Fi gure 2-57B is shown with a sma ller scale so that positive and negative-valued areas are clearly displayed. Th e green areas show no equal carrier generation for the two simulations, reds and yellows show areas where the full absorption case produced more carriers, and blues and pur ples show areas where the 40 nm simulation produced more carriers. In general, the ZnO phase shows no difference between the two simulations, the P3HT region beyond the 40 nm LD shows greater carrier generation in the full absorption simulation, and portions of the diffusion length region show greater generation in the 40 nm case. In the region beyond the tip of the nanorod, the plot sh ows a zero value for 4 nm, then a negative area with a value of approximately -2 x 1020 pairs/cm3 stretches for 16 nm before jumping to a positive value of nearly 6 x 1021 pairs/cm3. To the side of the nanorod, the contour plot shows a zero value for nearly 10 nm into the P3HT, but then turns negative for 26 nm. In the P3HT bulk 68

PAGE 69

region, there is no difference betw een the simulations up to x = 0.022 m. Beyond the interface for approximately 10 nm, values are all negati ve, indicating the 40 nm simulation generated more carriers. Above the interface in the ZnO phase there is a thin strip of 2 4 nm where there is an inconsistency. From x = 0.05 to 0.06 m, this strip shows a negative value. For x = 0.06 to 0.065 m, the strip shows a positive value. The full cell absorption simulation showed stronger carrier generation in the P3HT region beyond 40 nm from the ZnO interface. This is to be expected, as the full cell simulation allows this region to absorb any remaining carriers, wh ile the 40 nm simulation does not. However, the increased carrier generation for the 40 nm simulati on within the LD region is puzzling. Photons entering this region have the ex act same absorption history in both simulations. They have passed through a weakly-absorbing ZnO layer and entere d a strongly-absorbing P3HT region. For both simulations, this P3HT region has identical absorption properties for a path length of at least 40 nm. In the area within 40 nm of the side of the nanorod, the two simulations have identical absorption properties for 300 nm into the P3HT region. There is never a point in the simulation area where the 40 nm simulation displays a stronger absorption coefficient than the full cell simulation. With that in mind, this in consistency is considered to be a numerical anomaly in Medici during the convergence process. It is a strange occurance, but due to the difference in value between the negative regions and positive regions, as seen in Figure 2-57A, the total number of carriers in the cells is st ill increasing as the absorption area increases. There are two data points ignored in the graphs in Figure 2-56 the points at the base and tip corners of the nanorod. The carrier concentra tion at these points is pl otted against the exciton diffusion length in Figure 2-58. The point designated as the nanorod tip has coordinates of (0.015, 0.260) and shows a growth as the exciton di ffusion length increases. This point collects 69

PAGE 70

all generated carriers in the P3HT region for x 0.015 m and y 0.260 m. As the exciton diffusion length increases, this region produces a larger number of carriers because a larger portion of the region contains a strong absorpti on coefficient. The point designated as the nanorod base is the point at (0.015, 0.025) where the edge of the nanorod meets the base ZnO layer. This point colle cts all photogenerated carri ers for the region of x 0.015 m and y 0.025 m, which is composed entirely of ZnO. Because the absorption coefficient of ZnO is held constant in all simulations, the number of carriers collected at this point remains constant regardless of the exciton diffusion length. With this procedure established for measuri ng carrier generation across the simulated unit cell and compressing this distributi on to line sources in Medici, th e J-V response of cells can be determined. J-V Curves The carrier generation line sour ces shown in Figure 2-56 and 258 were input to Medici to generate J-V curves for the simulated cells. As described previously, these line sources of carriers restrict the existence of free electrons and holes to the inte rface between ZnO and P3HT, which matches the physics of excitonic solar cells. The resulting J-V curves were compared with the published data and are shown in Figure 2-59. As anticipated, the shortcircuit current density increases with increases in the exciton diffu sion length. Interestingly, this trend does not continue to the fully absorbing unit cell, wher e the short-circuit curre nt density drops by 0.2 mA/cm2 from the level of th e LD = 40 nm cell. All simulations failed to achieve the s hort-circuit current density of 2.2 mA/cm2 reported for the published cell. This is interesting cons idering that the simula ted cell was allowed to generate photocurrent for speci fied diffusion lengths ranging to the full cell area. The opencircuit voltage of the real cell was well appr oximated by the experiments due to the 0.85 eV 70

PAGE 71

energy band gap specified in P3HT. Fill factors for the simulated cells were higher than for the real cell. Table 2-17 displays a summary of the performance characteristics of the real and simulated cells shown in Figure 2-59. The simulated cell using a 40 nm exciton diffu sion length showed the best match to the real cell from literature. The VOC was nearly a perfect match, and the efficiency was approximately 5% higher than the literature value. However, the fill factor and JSC for the two cells did not match. The simulated cell showed a fill factor that was 50% higher and a JSC that was 33% lower than the real cell. Further simulations were performed using the 40 nm LD simulation as a basis. The effect of changing the carrier mobilitie s by up to two orders of magnit ude is shown in Figure 2-60, and the resulting performance measures are shown in Table 2-18. While four of the curves appeared as nearly identical, the curve corresponding to a 100x reduction in mobility began to show a noticeable drop in fill factor. Attempts to reduce the mobility by an extra order of magnitude failed due to an inability of Medici to converg e around such small values. Because Medici is designed for the simulation of inorganic semiconductors, mobility values of p = 1 x 10-5 and n = 1 x 10-6 cm2/V-s are more than 6 orders of magnitude lower than the program was designed to handle. This leads to convergence difficulties due to the low current values in the cell. The simulation using a 100x reduction in mobilit y values shows a better match to the real cell than any other simulation. The VOC and efficiency differ by only about 2% between the two cells. The JSC is again off by 33% from the real cell, as that value did not change with the variation in mobility. With the lower mobility, however, the fill factor dropped to within 40% of the published value. 71

PAGE 72

Because these simulations fail to produce a short-circuit current on par with the real cell, the P3HT absorption coefficient was adjusted to a hi gher value. This gene rates more carriers in the cell, which is expected to generate more photocurrent at all applied biases. The absorption coefficient is a parameter that is easily measur ed experimentally, but th ere is some physical of justification for this adjustment. The unit cell in Medici is defined by perfectly vertical nanowires with flat tips and perfect spacing between them. From the image in Figure 2-14, this is far from reality. Although care was taken to properly estimate the nanowire le ngth, width, and inter-wir e spacing, these values are purely averages taken from th e visible portion of the cross-se ctional view. The real wire array contains overlapping non-vertical wires that create non-uniform spaces between them where P3HT exists. In addition, these randomly angled structures would cr eate light reflection and refraction patterns that are not considered in Medici. It is not unreas onable to expect that this disordered array of nanowires could create light-trapping effects in the film, where incident rays are refracted in such a way to increase th eir residence time in the film and increase the degree to which they are absorbed. From this, it is not unreasonable to exp ect that this increased absorption occurs to some degree in the polymer regions within an exciton diffusion length from the material interfaces. For these reasons, the absorption coefficient of P3HT was increased by 50% in a simulation. The photogenerated car rier distribution for this simulation is compared to the previous 40 nm LD simulation, with an absorp tion coefficient of 100%, with the resulting surface plots shown in Figure 2-61. As expected this resulted in a larger number of photogenerated carriers in the film. 72

PAGE 73

Although the total number of phot ogenerated carriers increase d, there was not a uniform increase at all points in the film. Specificall y, in the column of polym er within one exciton diffusion length of the side of the nanorod, the number of carriers decreases quickly for the simulation with an increased absorption coeffici ent. This is because the number of photons penetrating deeper into the film is reduced due to the stronger absorption. This is clearly seen in Figure 2-62, which shows a comparison of carrier generation for the increased absorption case and the standard absorption case. The carrier distributions were subtra cted, and green regions show no change in absorption between the two simulations. Red, ora nge, and yellow regions show areas where the 150% absorption coefficien t resulted in stronger ab sorption, and this is primarily contained within a range of 40 nm (LD) from the upper surface of the ZnO-P3HT interface. Blue regions represent areas where the 150% absorption coefficient resulted in less absorption, and this is limited to the exciton di ffusion length region along the nanorod edge. This is due to strong absorption near the surface of the P3HT region, leaving fewer photons available for absorption in the deeper region. Although there are regions of increased and d ecreased carrier genera tion in the cell, the overall number of generated carriers increased by approximately 3 x 1024 pairs/cm3 over the unit cell. This additional charge generation did not tr anslate into additional photocurrent as expected, however. Figure 2-63 shows J-V curves for the two cells, calculated with p = 1 x 10-4 cm2/V-s. Despite the increased carriers density for th e 150% absorption coefficient simulation, the short-circuit current density decreased by 0.23 mA/cm2. This is very similar to the effect seen when the full P3HT area was allowed to contribute to absorption. In fact, a comparison of those cells shows extremely similar properties. Note that the full cell simulation was performed with the standard values fo r mobility in the P3HT regions, while the two 40 nm simulations were 73

PAGE 74

performed with a these mobility values reduced by two orders of magnitude. This explains the difference in fill factor shown in Table 2-19, bu t Figure 2-60 demonstrated that this mobility reduction does not impact the short-circui t current density in these simulations. It is believed that the reason for the reduced performance associated with this increase in carrier generation is an effect of increased annihilation near the material interface. This large number of carriers is being produced along a very narrow regi on directly at the material interface. This results in many free electrons and holes in a small area, which could cause attractive forces between them to promote anni hilation immediately follo wing their generation. There seems to be a breaking point between 3.14 x 1025 and 3.23 x 1025 pairs/cm3 of total carrier density, as this is the limit where the re duced charge transport seems to take hold. The short-circuit current density drops by about 15%, the VOC remains constant, and the fill factor drops by less than 10%. Summary of Results The device modeling program Medici was used for simulation of hybrid solar cells. Several nuances of the program were probed with a test model before attempts were made to simulate an existing cell from the literature. After probing the effect of various parameters, a two-step simulation strategy was adopted that separated photon absorption and charge transport. This m odel more accurately approximates exciton dynamics by eliminating free electrons in the polymer regions of the cell and applying free carrier generation directly at the ma terial interface lines. Simulations of this type initially showed lo w performance, but increases in the exciton diffusion length up to 40 nm provided increased charge generation and cu rrent flow. Further increases in the number of carriers generated re duced the current in the simulation, presumably due to increased charge attract ion and annihilation. It was found that decreasing the carrier 74

PAGE 75

mobilities in P3HT from the field effect mobility values f ound in literature resulted in a decrease in the fill factor to levels similar to the real cell. 75

PAGE 76

A B Figure 2-1. Hybrid solar cell (A) and correspond ing unit cell (B) used fo r device simulation. Table 2-1. P3HT Properties for Device Simulations. Property Value Reference Doping Density p-type 5x1016 cm-2[67] Permittivity 3 [61] NC 2x1018 [69] NV 2x1019 [69] E g 1.7 eV [70] Electron Affinity 3.15 eV [71] e 0.01 cm2/Vs 10% of h h 0.1 cm2/Vs [68] Table 2-2. CdSe Properties for Device Simulations. Property Value Doping Density n-type 6x1016 cm-2 Permittivity 10.2 NC 2x1018 NV 2x1019 E g 1.74 eV Electron Affinity 3.75 eV e 650 cm2/Vs h 30 cm2/Vs Table 2-3. Electrode Propert ies for Device Simulations. Electrode Optical Properties Reference ITO Transparent [72] Al 90% Reflectance [73] 76

PAGE 77

Wavelength ( m) 0.2 0.4 0.6 0.8 1.0 Absorption Coefficient (cm-1) 0.0 5.0e+4 1.0e+5 1.5e+5 2.0e+5 2.5e+5 P3HT ZnO CdSe Figure 2-2. Wavelength-de pendent absorption coefficient data used in simulations [66, 74]. Figure 2-3. J-V curves for simulated hybrid sola r cells with different methods of specifying doping density. See Table 2-4 for a description of the differences. 77

PAGE 78

Voltage (V) 0.000.050.100.150.200.250.30 Current Density (mA/cm2) -10 -8 -6 -4 -2 0 3E16 4E16 5E16 6E16 7E16 Figure 2-4. Simulated J-V curves showing the e ffect of doping density in the CdSe nanorods. The legend shows the n-type doping density measured in cm-2. 78

PAGE 79

Table 2-4. Impurity profile inputs for simulations. Filename Impurity Profile Input Statement impurity_1 PROFILE P-TYPE REGI ON=P3HT UNIFORM N.PEAK=5E16 PROFILE N-TYPE REGION=Cd Se UNIFORM N.PEAK=6E16 impurity_2 PROFILE P-TYPE Y.MIN=0 Y.MAX=0.100 UNIFORM N.PEAK=5E16 PROFILE N-TYPE Y.MIN=0.01 Y. MAX=0.100 X.MIN=0 X.MAX=0.005 + UNIFORM N.PEAK=6E16 impurity_3 PROFILE P-TYPE Y.MIN=0 Y.MAX=0.01 UNIFORM N.PEAK=5E16 PROFILE P-TYPE Y.MIN=0.01 Y.MAX=0.100 X.MIN=0.005 X.MAX=0.015 + UNIFORM N.PEAK=5E16 PROFILE N-TYPE Y.MIN=0.01 Y. MAX=0.100 X.MIN=0 X.MAX=0.005 + UNIFORM N.PEAK=6E16 impurity_4 PROFILE P-TYPE Y.MIN=0 Y.MAX=0.01 UNIFORM N.PEAK=5E16 PROFILE P-TYPE Y.MIN=0.01 Y.MAX=0.100 X.MIN=0.005 X.MAX=0.015 + UNIFORM N.PEAK=5E16 PROFILE N-TYPE Y.MIN=0.01 Y. MAX=0.100 X.MIN=0 X.MAX=0.005 + UNIFORM N.PEAK=1E16 impurity_5 PROFILE P-TYPE Y.MIN=0 Y.MAX=0.01 UNIFORM N.PEAK=5E16 PROFILE P-TYPE Y.MIN=0.01 Y.MAX=0.100 X.MIN=0.005 X.MAX=0.015 + UNIFORM N.PEAK=5E16 PROFILE P-TYPE Y.MIN=0.01 Y.MAX=0.100 X.MIN=0 X.MAX=0.005 + UNIFORM N.PEAK=5E16 PROFILE N-TYPE Y.MIN=0.01 Y.MAX=0.100 X.MIN=0 X.MAX=0.005 + UNIFORM N.PEAK=6E16 Figure 2-5. Variation of open circuit volta ge with doping density of CdSe nanorods. 79

PAGE 80

Voltage (V) -0.10 -0.05 0.00 0.05 0.10 Current Density (mA/cm2) -20 -15 -10 -5 0 5 10 15 20 20 nm 30 nm 40 nm 50 nm 100 nm 200 nm 300 nm 500 nm 1 m Figure 2-6. Simulated J-V curves with varying unit cell thickness. Figure 2-7. Solar cell parameters for unit 12 nm wide unit cells with varying cell thickness. 80

PAGE 81

Table 2-5. Solar cell performance measures for un it cells of 12 nm width a nd varying thickness. Thickness (nm) (%) VOC (V) JSC (mA/cm2) FF 11 0.0000 0.0000 1.251 0.0003 20 0.0000 0.0009 2.257 0.0000 30 0.0052 0.0067 3.106 0.2533 40 0.0172 0.0163 3.899 0.2699 50 0.0335 0.0252 4.627 0.2873 60 0.0507 0.0317 5.297 0.3019 70 0.0669 0.0361 5.916 0.3131 80 0.0817 0.0393 6.491 0.3203 90 0.0954 0.0418 7.023 0.3250 100 0.1082 0.0437 7.513 0.3295 120 0.1310 0.0466 8.368 0.3360 150 0.1598 0.0497 9.394 0.3422 200 0.1966 0.0531 10.625 0.3484 300 0.2456 0.0568 12.150 0.3559 500 0.2975 0.0597 13.791 0.3614 1000 0.3275 0.0588 15.594 0.3572 Voltage (V) 0.0 0.2 0.4 0.6 0.8 Current Density (mA/cm2) -10 -8 -6 -4 -2 0 2 4 2 nm 5 nm 10 nm 15 nm 20 nm 25 nm 30 nm Figure 2-8. J-V curves for hybrid cells with varying nanorod width. 81

PAGE 82

Voltage (V) 0.00.10.20.30.40.50.60.70.8 Current Density (mA/cm2) -10 -8 -6 -4 -2 0 2 4 10 nm 11 nm 12 nm 13 nm 14 nm 15 nm 16 nm 17 nm 18 nm 19 nm 20 nm 21 nm 22 nm 23 nm 24 nm 25 nm Figure 2-9. J-V curves for simulated hybrid ce lls with CdSe nanorod half-thickness between 10 and 25 nm. Figure 2-10. Solar cell performance measures for simulated hybrid cells w ith varying CdSe halfwidth. 82

PAGE 83

Table 2-6. Performance measures for simulated hybrid cells with varying CdSe half-width. CdSe Width (nm) Voc (V) Jsc (mA/cm2) FF (%) 2 0.044 7.513 0.330 0.11 5 0.253 9.658 0.623 1.52 10 0.410 8.575 0.661 2.32 11 0.435 9.488 0.670 2.77 12 0.454 9.453 0.669 2.87 13 0.470 9.421 0.668 2.96 14 0.485 9.391 0.666 3.03 15 0.487 6.461 0.675 2.12 16 0.512 9.338 0.662 3.16 17 0.523 9.315 0.659 3.21 18 0.533 9.295 0.658 3.26 19 0.542 9.276 0.656 3.30 20 0.535 5.745 0.677 2.08 21 0.558 9.243 0.654 3.37 22 0.564 9.229 0.653 3.40 23 0.553 5.414 0.679 2.03 24 0.576 9.204 0.652 3.45 25 0.580 9.194 0.652 3.48 30 9.071 4.93 Voltage (V) 0.000.050.100.150.200.250.300.350.40 Current Density (mA/cm2) -10 0 10 20 30 40 50 Center, Y=-2.5 Center, Y=-5.0 Left Edge, Y=-2.5 Right Edge, Y=-2.5 Figure 2-11. J-V curves for hybrid solar cell s with different light source specifications. 83

PAGE 84

Figure 2-12. Illustration of the PHOTOGEN command in Medici. Figure 2-13. SEM images of (a) Zn O nanofibers and (b) nanofiber and P3HT composite films. Reprinted with permission from D.C. Olson, J. Piris, R.T. Collins, S.E. Shaheen, D.S. Ginley, Thin Solid Films 496 (2006) 26 Figure 2 (a), (b) p. 28. 84

PAGE 85

Figure 2-14. J-V curve for a real ZnO:P3HT solar cell (solid line) to be used for verification of Medici simulations. Reprinted with perm ission from D.C. Olson, J. Piris, R.T. Collins, S.E. Shaheen, D.S. Ginley, Thin Solid Films 496 (2006) 26 Figure 3 p. 28. Voltage (V) -0.2-0.10.00.10.20.30.40.5 Current Density (mA/cm2) -3 -2 -1 0 1 2 Power Density (mW/cm2) -0.9 -0.6 -0.3 0.0 0.3 0.6 Current Density Power Density Figure 2-15. J-V and P-V curves for the real solar cell fabricated by Olson et al. [75] Table 2-7. Cell performance measures from pu blished and digitally converted J-V curves. Curve VOC (V) JSC (mA/cm2) FF (%) Published 0.44 2.2 0.56 0.53 Converted 0.44 2.2 0.57 0.55 85

PAGE 86

Table 2-8. ZnO properties used for hybrid solar cell simulation. Property Value Doping Density n-type 5 x 1017 cm-2 Permittivity 9.0 Nc 2 x 1018 Nv 2 x 1019 E g 3.3 eV Affinity 4.0 eV e 50 cm2/V-s h 5 cm2/V-s Figure 2-16. Unit cell used for simulations of ZnO/P3HT hybrid cells. Device areas area a) ITO electrode, b) ZnO, c) P3HT photocurrent generating region, d) P3HT non-generating region, and e) Ag electrode. 86

PAGE 87

Figure 2-17. Medici unit cell used for device simu lation. The blue area is ZnO, the red area is the photogenerating region of P3HT, and the green area is the non-generating region of P3HT. 87

PAGE 88

Voltage (V) -0.20.0 0.2 0.4 0.6 0.8 Current Density (mA/cm2) -12 -10 -8 -6 -4 -2 0 2 Real Cell tsf496_1 tsf496_2 tsf496_4 Figure 2-18. Simulated J-V curves for ZnO:P3HT solar cell using two P3HT regions. The first region is the exciton diffusion length region, where all properties are set as shown in Table 2-1. The second region is the non-ge nerating region outside of the diffusion length, with varying properties In tsf496_1, the absorption co efficient is set to zero. In tsf496_2, the properties remain the same as the exciton diffusion length region. In tsf496_4 the electron mobility is reduced by an order of magnitude (0.0001 cm2/Vs). 88

PAGE 89

Voltage (V) -0.20.00.20.40.60.81.0 Current Density (mA/cm2) -5.0 -4.5 -4.0 -3.5 -3.0 -2.5 -2.0 -1.5 -1.0 90% Reflectance 50% Reflectance 0% Reflectance Figure 2-19. J-V curves for simulated cells with zero absorption in the P3HT2 region and varying reflectance from the Ag electrode. 89

PAGE 90

Voltage (V) -0.20.00.20.40.60.81.0 Current Density (mA/cm2) -4.0 -3.5 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 tsf496_05n = 0.001 cm2/V-s, p = 0.01 cm2/V-s tsf496_07's reduced by 50% in both P3HT regions tsf496_08's reduced by 50% in P3HT2 region tsf496_09's reduced by 90% in both P3HT regions Figure 2-20. Simulation results showing the effect of changing charge mobilities in the P3HT regions. Table 2-9. Performance measures for simulated cells with varying carrier mobility. Mobility (cm2/V-s) JSC (mA/cm2) VOC (V) FF (%) n = 0.01 p = 0.001 3.61 1.04 0.57 2.12 n = 0.005 p = 0.0005 3.58 1.04 0.54 2.00 n = 0.01 p = 0.0005 3.58 1.04 0.57 2.13 n = 0.001 p = 0.0001 3.44 1.04 0.46 1.64 90

PAGE 91

Voltage (V) -0.20.00.20.40.60.81.0 Current Density (mA/cm2) -4 -3 -2 -1 0 1 2 Real Cell tsf496_09, LD = 10 nm tsf496_11, LD = 9 nm tsf496_12, LD = 8 nm tsf496_13, LD = 7 nm tsf496_14, LD = 6 nm tsf496_15, LD = 5 nm Figure 2-21. J-V curves for simulated cel ls with varying exciton diffusion length. Figure 2-22. Extrapolations to estimate VOC for simulated cells with varying exciton diffusion length. 91

PAGE 92

Figure 2-23. Solar cell performance measures fo r simulated cells with varying exciton diffusion lengths. Table 2-10. Performance measures for simulate d cells with varying exciton diffusion lengths. LD (nm) Voc (V) Jsc (ma/cm2) FF (%) 10 1.04 3.44 0.459 1.64 9 1.04 3.17 0.468 1.54 8 1.05 2.90 0.473 1.44 7 1.05 2.61 0.482 1.32 6 1.06 2.32 0.487 1.20 5 1.07 2.02 0.492 1.06 92

PAGE 93

Voltage (V) -0.20.00.20.40.60.81.0 Current Density (mA/cm2) -3 -2 -1 0 1 2 Real Cell P3HT Doping 5x1016 cm-3 P3HT Doping 1x1016 cm-3 P3HT Doping 5x1015 cm-3 P3HT Doping 1x1015 cm-3 P3HT Doping 5x1014 cm-3 P3HT Doping 1x1014 cm-3 Figure 2-24. Real and simulated J-V curves for cells with varying P3HT doping density. Voltage (V) -0.20.00.20.40.60.81.0 Current Density (mA/cm2) -2.5 -2.0 -1.5 -1.0 -0.5 0.0 Real Cell ZnO Doping Density 5x1017 ZnO Doping Density 1x1017 ZnO Doping Density 5x1016 ZnO Doping Density 2x1016 ZnO Doping Density 1x1016 ZnO Doping Density 9x1015 ZnO Doping Density 7x1015 ZnO Doping Density 5x1015 ZnO Doping Density 1x1015 Figure 2-25. Real and simulated J-V curves fo r hybrid cells with varying ZnO doping density. 93

PAGE 94

Voltage (V) -0.20.00.20.40.60.81.0 Current Density (mA/cm2) -2.5 -2.0 -1.5 -1.0 -0.5 0.0 Real Cell P3HT NC=2x1018 NV=2x1019 P3HT NC=2x1019 NV=2x1019 P3HT NC=2x1019 NV=2x1020 P3HT NC=2x1020 NV=2x1020 P3HT NC=2x1020 NV=2x1021 Figure 2-26. Real and simulated J-V curves for hybrid solar cells with varying P3HT density of states. Voltage (V) -0.20.00.20.40.60.81.01.21.4 Current Density (mA/cm2) -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 n=0.001 cm2/V-s, p=0.01 cm2/V-s n=0.0001 cm2/V-s, p=0.001 cm2/V-s n=0.00001 cm2/V-s, p=0.0001 cm2/V-s Figure 2-27. Simulated J-V curves fo r hybrid solar cells with varying P3HT mobility. 94

PAGE 95

Voltage (V) -0.20.00.20.40.60.81.01.21.4 Current Density (mA/cm2) -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 P3HT Doping: 5x1016 P3HT Doping: 1x1016 P3HT Doping: 5x1015 P3HT Doping: 1x1015 P3HT Doping: 5x1014 P3HT Doping: 1x1014 Figure 2-28. Simulated J-V curves for hybrid cells with varying P3HT doping concentrations. Figure 2-29. Energy band diagram for P3HT ZnO hybrid solar cells. 95

PAGE 96

Voltage (V) -0.2-0.10.00.10.20.30.40.50.6 Current Density (mA/cm2) -4 -2 0 2 4 6 8 10 Real Cell tsf496_45 tsf496_53 Figure 2-30. J-V curves for simulated cells w ith varying energy band ga p in the active layers. Wavelength ( m) 0.00.20.40.60.81.01.21.4 Absorption Coefficient (cm-1) 0.0 5.0e+5 1.0e+6 1.5e+6 2.0e+6 2.5e+6 abs_none abs_p3ht abs_i_zno Figure 2-31. Absorption coeffici ent vs. wavelength as tabulated in Medici for input files corresponding to zero absorption (red), P3HT (green), and ZnO (teal). 96

PAGE 97

Wavelength ( m) 0.20.40.60.81.01.21.41.61.8 Normalized Flux 0.0 0.2 0.4 0.6 0.8 1.0 1.2 AM1.5 Solar Spectrum Figure 2-32. AM1.5 solar spectrum. Wavelength ( m) 0.0 0.2 0.4 0.6 0.8 1.0 Photogenerated Carriers (C m-1s-1) 0.0 5.0e-14 1.0e-13 1.5e-13 2.0e-13 2.5e-13 3.0e-13 3.5e-13 Absorption Coefficient (cm-1) 0.00 1.00e+5 2.00e+5 1.25e+6 1.50e+6 1.75e+6 2.00e+6 2.25e+6 Generated Carriers abs_p3ht abs_zno Figure 2-33. Carrier generation (left axis) in simulated solar cells plotted with absorption coefficients for P3HT and ZnO (right axis). 97

PAGE 98

Voltage (V) -0.10.00.10.20.30.40.5 Current Density (mA/cm2) -4 -2 0 2 4 Medici-assigned values Fully specified values Figure 2-34. J-V curves for simula ted cells showing the original P3HT absorption profile and an edited absorption profile limiting absorption between 0.2 and 0.3 m. Voltage (V) -0.10.00.10.20.30.40.5 Current Density (mA/cm2) -3 -2 -1 0 1 tsf496_45 1 stage tsf496_47 2 stages tsf496_48 4 stages tsf496_49 6 stages tsf496_50 8 stages tsf496_51 10 stages Figure 2-35. J-V curves for simulated cells w ith exciton diffusion length of 10 nm and multistage absorption regions. 98

PAGE 99

P3HT Thickness 0 5 10 15 20 Fraction of Maximum Absorption Coefficient 0.0 0.2 0.4 0.6 0.8 1.0 = 0.001 = 2 = 4 = 6 = 8 = 10 Figure 2-36. Examples of cumu lative distribution function with mean of 10 nm and a range of standard deviation. Table 2-11. Absorption data a nd short-circuit current for graded absorption simulations. 0 nm 1 nm 2 nm 3 nm 4 nm 5 nm 6 nm 10 nm 0-2 nm 100% 100% 100% 100% 100% 100% 100% 100% 2-4nm 100% 100% 100% 100% 100% 90% 90% 90% 4-6 nm 100% 100% 100% 100% 90% 90% 80% 90% 6-8 nm 100% 100% 90% 80% 80% 70% 70% 70% 8-10 nm 100% 80% 70% 60% 60% 60% 60% 60% 10-12 nm 0% 20% 30% 40% 40% 40% 40% 40% 12-14nm 0% 0% 10% 20% 20% 30% 30% 30% 14-16 nm 0% 0% 0% 10% 10% 10% 20% 20% 16-18 nm 0% 0% 0% 0% 0% 10% 10% 10% 18-20 nm 0% 0% 0% 0% 0% 0% 0% 0% Average 50% 50% 50% 51% 50% 50% 50% 51% JSC (mA/cm2) 1.62 1.62 1.66 1.67 1.68 1.69 1.69 1.71 99

PAGE 100

Voltage (V) -0.10.00.10.20.30.40.5 Current Density (mA/cm2) -3 -2 -1 0 1 2 3 = 0 nm = 1 nm = 2 nm = 3 nm = 4 nm = 5 nm = 6 nm = 10 nm Figure 2-37. Simulated J-V curves for cells with graded absorption profiles. Voltage (V) -0.10.00.10.20.30.40.5 Current Density (mA/cm2) -30 -20 -10 0 10 20 30 ZnO:P3HT with 10 nm absorption region ZnO:CIS with 10 nm absorption region ZnO:P3HT with full absorption region ZnO:CIS with full absorption region ZnO:CIS with 10 nm absorption region and Eg = 0.85 eV Figure 2-38. Simulated J-V curves with CIS replacing P3HT. 100

PAGE 101

Table 2-12. Materials properties for P3HT and CIS used in cell simulations. Property P3HT CIS Doping Density p-type 5x1016 cm-2 p-type 5x1016 cm-2 Permittivity 3 13.6 NC 2x1018 3x1018 NV 2x1019 1.5x1019 E g 1.7 eV 1.04 eV Electron Affinity 3.15 eV 3.93 eV e 0.1 cm2/Vs 300 cm2/Vs h 0.01 cm2/Vs 30 cm2/Vs Voltage (V) -0.10.00.10.20.30.40.50.6 Current Density (mA/cm2) -10 -5 0 5 10 tsf496_64 ZnO:CIS tsf496_68 P3HT Permittivity tsf496_69 P3HT Density of States tsf496_70 P3HT Affinity tsf496_71 P3HT Mobilities tsf496_72 P3HT Absorption Figure 2-39. Simulated J-V curves for ZnO:CIS solar cells with an individual material property changed to the P3HT value. Table 2-13. Performance measures for ZnO:CIS cells with an individual ma terial property set at the P3HT value. Simulation Parameter Adjusted Voc (V) JSC (mA/cm2) FF (%) 64 None 0.439 7.774 0.772 2.633 68 Permittivity 0.443 7.747 0.778 2.669 69 Density of States 0.441 7.773 0.774 2.655 70 Affinity 0.487 7.654 0.485 1.808 71 Mobility 0.545 6.309 0.662 2.277 72 Absorption 0.421 4.105 0.762 1.317 101

PAGE 102

Voltage (V) -0.10.00.10.20.30.40.50.6 Current Density (mA/cm2) -10 -5 0 5 10 15 20 tsf496_72 P3HT Absorption tsf496_73 P3HT Absorption, Permittivity tsf496_74 P3HT Absorption, Density of States tsf496_75 P3HT Absorption, Affinity tsf496_76 P3HT Absorption, Mobilities Figure 2-40. Simulated J-V curves for ZnO:CIS solar cells with the P3HT absorption spectrum applied. Table 2-14. Performance measures for si mulated ZnO:CIS solar cells with the P3HT absorption spectrum. Simulation Properties Varied VOC (V) JSC (mA/cm2) FF (%) 72 Absorption 0.421 4.105 0.762 1.317 73 Absorption Permittivity 0.427 4.065 0.765 1.328 74 Absorption Density of States 0.423 4.105 0.765 1.328 75 Absorption Affinity 0.467 3.980 0.493 0.916 76 Absorption Mobility 0.517 3.293 0.657 1.118 102

PAGE 103

Voltage (V) -0.10.00.10.20.30.40.50.60.7 Current Density (mA/cm2) -10 -5 0 5 10 15 20 tsf496_75 P3HT Absorption, Affinity tsf496_77 P3HT Absorption, Affinity, Permittivity tsf496_78 P3HT Absorption, Affinity, Density of States tsf496_79 P3HT Absorption, Affinity, Mobilities Figure 2-41. Simulated ZnO:CIS solar cells with P3HT values for absorption coefficient and electron affinity. Table 2-15. Performance measures for simulated ZnO:CIS solar cells with P3HT values for absorption coefficient and electron affinity. Simulation Properties Varied VOC (V) JSC (mA/cm2) FF (%) 75 Absorption Affinity 0.467 3.980 0.493 0.916 77 Absorption Affinity Permittivity 0.455 3.979 0.502 0.909 78 Absorption Affinity Density of States 0.471 3.975 0.479 0.896 79 Absorption Affinity Mobility 0.613 3.439 0.199 0.419 103

PAGE 104

Voltage (V) -0.10.00.10.20.30.40.50.6 Current Density (mA/cm2) -10 -5 0 5 10 15 20 tsf496_80 P3HT Absorption, Affinity, Eg, Permittivity tsf496_81 P3HT Absorption, Affinity, Eg, Density of States tsf496_82 P3HT Absorption, Affinity, Eg, Mobilities Figure 2-42. Simulated ZnO:CIS solar cells with P3HT values for absorption coefficient, electron affinity, and energy band gap. Table 2-16. Performance measures for simulated ZnO:CIS solar cells with P3HT values for absorption coefficient, electr on affinity, and energy band gap. Simulation Properties Varied VOC (V) JSC (mA/cm2) FF (%) 80 Absorption Affinity, Eg Permittivity 0.265 3.633 0.418 0.402 81 Absorption Affinity, Eg Density of States 0.281 3.608 0.395 0.401 82 Absorption Affinity, Eg Mobility 0.423 2.164 0.159 0.145 104

PAGE 105

Voltage (V) -0.10.00.10.20.30.40.5 Current Density (mA/cm2) -10 0 10 20 p = 0.1 cm2/V-s p = 1 cm2/V-s p = 10 cm2/V-s p = 100 cm2/V-s p = 200 cm2/V-s p = 500 cm2/V-s p = 800 cm2/V-s p = 1000 cm2/V-s p = 5000 cm2/V-s p = 10,000 cm2/V-s Figure 2-43. Simulated J-V curves for ZnO:P3HT solar cells with varying carrier mobility. In all cases, carrier generation is allowed in the full P3HT region and the electron mobility is set to 10% of the hole mobility. Figure 2-44. Calculation method for estim ated second derivatives of J-V curves. 105

PAGE 106

Voltage (V) 0.00.10.20.30.40.50.60.7 J"est -4e-4 -2e-4 0 2e-4 4e-4 p = 0.1 cm2/V-s p = 1 cm2/V-s p = 10 cm2/V-s p = 100 cm2/V-s p = 200 cm2/V-s p = 500 cm2/V-s p = 800 cm2/V-s p = 1000 cm2/V-s p = 5000 cm2/V-s p = 10,000 cm2/V-s Figure 2-45. Estimated second derivative Jest for simulated ZnO:P3HT solar cells with varying carrier mobilities and carrier generation in the full P3HT region. In all cases, the electron mobility is set to 10% of the hole mobility. 106

PAGE 107

Voltage (V) -0.10.00.10.20.30.40.5 Current Density (mA/cm2) -4 -2 0 2 4 6 8 10 p = 0.1 cm2/V-s p = 1 cm2/V-s p = 10 cm2/V-s p = 100 cm2/V-s p = 200 cm2/V-s p = 500 cm2/V-s p = 800 cm2/V-s p = 1000 cm2/V-s p = 5000 cm2/V-s p = 10,000 cm2/V-s Figure 2-46. Simulated J-V curves for ZnO:P3HT solar cells with varying carrier mobilities and carrier generation in the 10-nm exciton di ffusion length region. In all cases, the electron mobility is set to 10% of the hole mobility. 107

PAGE 108

Voltage (V) 0 00 20 40 6 J"est -2e-4 -1e-4 0 1e-4 2e-4 3e-4 4e-4 5e-4 p = 0.1 cm2/V-s p = 1 cm2/V-s p = 10 cm2/V-s p = 100 cm2/V-s p = 200 cm2/V-s p = 500 cm2/V-s p = 800 cm2/V-s p = 1000 cm2/V-s p = 5000 cm2/V-s p = 10,000 cm2/V-s Figure 2-47. Estimated second derivative Jest for simulated ZnO:P3HT solar cells with varying carrier mobilities and carrier generation in the 10-nm exciton diffusion length region. In all cases, the electr on mobility is set to 10% of the hole mobility. 108

PAGE 109

Voltage (V) -0.10.00.10.20.30.40.5 Current Density (mA/cm2) -4 -2 0 2 4 tsf496_98 100% absorption tsf496_103 80% absorption tsf496_104 50% absorption tsf496_105 20% absorption Figure 2-48. Simulated J-V curves for ZnO:P3HT solar cells with varying absorption coefficients in P3HT. Voltage (V) -0.20.00.20.40.60.81.01.21.4 Current Density (mA/cm2) -4 -2 0 2 4 tsf496_98 Eg = 0.85 eV tsf496_106 Eg = 1.0 eV tsf496_107 Eg = 1.2 eV tsf496_108 Eg = 1.5 eV tsf496_109 Eg = 1.7 eV Figure 2-49. Simulated J-V curves for ZnO:P3HT solar cells with varying energy band gap. 109

PAGE 110

Voltage (V) 0 00 20 40 60 81 0 Current Density (mA/cm2) -5 -4 -3 -2 -1 0 1 Eg = 0.85 eV Eg = 1.0 eV Eg = 1.2 eV Eg = 1.5 eV = 100% = 80% = 50% Figure 2-50. Simulated J-V curves for ZnO:P3HT solar cells with varying band gap and absorption in the P3HT region. Voltage (V) -0.10.00.10.20.30.40.5 Current Density (mA/cm2) -4 -3 -2 -1 0 1 2 Real Data 100% absorption 80% absorption 50% absorption 40% absorption 30% absorption 20% absorption Figure 2-51. Simulated J-V curves for ZnO:P3HT solar cells with P3HT energy band gap of 1.0 eV and hole mobility of 500 cm2/Vs. 110

PAGE 111

A) X-Coordinate ( m) 0.000.010.020. 030.040.050.06 Y-Coordinate ( m) 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 1e+20 1e+21 1e+22 B) 1e+19 1e+20 1e+21 1e+22 1e+23 0.014 0.016 0.018 0.020 0.022 0.024 0.026 0.05 0.10 0.15 0.20 0.25P h o t o g e n e r a t e d C a r r i e r s ( p a i r s / c m3)X C o o r d i n a t e ( m )Y C o o r d i n a t e ( m ) 1e+19 1e+20 1e+21 1e+22 1e+23 Figure 2-52. Photogenerated ca rrier distribution in pairs/cm3 for A) the full unit cell and B) the region of 13 nm x 27 nm along the edge of the ZnO nanorod. Note that the x and y axes do not follow the same scale. 111

PAGE 112

Figure 2-53. Carrier mapping scheme for two-stag e simulations. Gray regions are ZnO and blue regions are P3HT. Orange arrows represent the nearest interface points for photogenerated carriers in the region de fined by the dotted white lines. 112

PAGE 113

X-Coordinate ( m) 0.000.010.020.030.040.050.06 Y-Coordinate ( m) 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 2e+21 4e+21 6e+21 8e+21 1e+22 Line 1 Line 2 Line 3 Figure 2-54. Photogenerated ca rrier distribution for a simu lated cell with LD = 20 nm. 113

PAGE 114

Figure 2-55. Cumulative number of photogenerated carriers in si mulated cells with varying exciton diffusion length. 114

PAGE 115

A) B) C) Figure 2-56. Photogenerated carr ier distribution along xand ycoordinates for models with varying exciton diffusion lengths. 115

PAGE 116

A) -1e+21 0 1e+21 2e+21 3e+21 4e+21 5e+21 6e+21 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40D i f f e r e n c e i n P h o t o g e n e r a t e d C a r r i e r s ( p a i r s / c m3)X C o o r d i n a t e ( m )Y C o o r d i n a t e ( m ) -1e+21 0 1e+21 2e+21 3e+21 4e+21 5e+21 6e+21 B) X Coordinate ( m) 0.000.010.020.030.040.050.06 Y Coordinate ( m) 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 -4e+18 -2e+18 0 2e+18 4e+18 Figure 2-57. Contour plots of the photogenerate d carrier difference between the full absorption simulation and the 40 nm LD simulation. 116

PAGE 117

Figure 2-58. Photogenerated carriers at the tip (left axis) and base (right axis) corner points of the nanorod in simulated hybrid cells. Voltage (V) -0.10.00.10.20.30.40.5 Current Density (mA/cm2) -3 -2 -1 0 1 2 3 4 Real Data 10 nm 20 nm 40 nm Full Cell Figure 2-59. J-V curves for simulated sola r cells using line-sour ce carrier generation. 117

PAGE 118

Table 2-17. Cell performance measures for real and simulated solar cells. Cell JSC (mA/cm2) VOC (V) FF (%) Real 2.20 0.44 0.56 0.53 10 nm LD 0.59 0.44 0.84 0.22 20 nm LD 1.01 0.45 0.84 0.38 40 nm LD 1.47 0.45 0.84 0.56 Full Cell 1.27 0.45 0.85 0.48 Voltage (V) 0.0 0.1 0.2 0.3 0.4 0.5 Current Density (mA/cm2) -1.4 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 x 0.01 x 0.1 x 1 x 10 x 100 Figure 2-60. J-V curves for simulated cells with a 40 nm LD and varying carrier mobility. Table 2-18. Performance measures for simulated cells with 40 nm LD and varying mobility values. Mobility JSC (mA/cm2) VOC (V) FF (%) 0.01 x 1.47 0.45 0.78 0.52 0.1 x 1.47 0.45 0.84 0.56 1 x 1.47 0.45 0.84 0.56 10 x 1.47 0.45 0.84 0.56 100 x 1.47 0.45 0.84 0.56 Real Cell 2.20 0.44 0.56 0.53 118

PAGE 119

A) 0.0 2.0e+21 4.0e+21 6.0e+21 8.0e+21 1.0e+22 1.2e+22 1.4e+22 1.6e+22 1.8e+22 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40P h o to g e n e r a t e d C a r r i e r s ( p a i r s /c m3)X C o o r d i n a t e ( m )Y C o o r d i n a t e ( m ) 0.0 2.0e+21 4.0e+21 6.0e+21 8.0e+21 1.0e+22 1.2e+22 1.4e+22 1.6e+22 1.8e+22 B) 0.0 2.0e+21 4.0e+21 6.0e+21 8.0e+21 1.0e+22 1.2e+22 1.4e+22 1.6e+22 1.8e+22 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40P h o t o g e n e r a t e d C a r r i e r s ( p a i r s / c m3)X C o o r d i n a t e ( m )Y C o o r d i n a t e ( m ) 0.0 2.0e+21 4.0e+21 6.0e+21 8.0e+21 1.0e+22 1.2e+22 1.4e+22 1.6e+22 1.8e+22 Figure 2-61. Photogenerated carr ier distribution for a simulated unit cell with (A) LD = 40 nm and (B) LD = 40 and the P3HT absorption coefficient increased by 50%. 119

PAGE 120

X Coordinate ( m) 0.000.010.020.030.040.050.06 Y Coordinate ( m) 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 -4e+20 -3e+20 -2e+20 -1e+20 0 1e+20 2e+20 3e+20 4e+20 Figure 2-62. Difference in photogenerated carriers (in pairs/cm3) between 150% and 100% P3HT absorption coefficients in simulated cells with 40 nm LD. 120

PAGE 121

121 Voltage (V) 0 00 10 20 30 40 5 Current Density (mA/cm2) -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 Standard Absorption Coefficient + 50% Absorption Coefficient Figure 2-63. J-V curves for simulated cells with varying absorption coefficient in P3HT. Table 2-19. Generated carrier s and performance measures for simulated solar cells. Cell Total Carriers (pairs/cm3) JSC (mA/cm2) VOC (V) FF (%) 40 nm 3.14 x 1025 1.47 0.45 0.78 0.52 Full Cell 3.23 x 10251.27 0.45 0.85 0.48 40 nm, 150% 3.42 x 1025 1.24 0.45 0.79 0.44

PAGE 122

CHAPTER 3 ORGANIC AND HYBRID SOLAR CELL PROCESS DEVELOPMENT Introduction The experimental details for process devel opment of organic and hybrid solar cells are presented in this chapter. In the first section, pretreatment of the indium tin oxide anode is studied. This work was done in collaboration wi th the research group of Dr. Chinho Park at Yeungnam University in South Korea, particular ly Jiyoun Seol. This work was previously presented at the 2006 World Conference on Phot ovoltaic Energy Conversion sponsored by IEEE and was published in their Conference Record [80]. The second section focuses on the development of bilayer photovoltaic cells using ab sorbing polymers. The next section describes efforts to characterize solvents appropriate for use in hybrid bulk hete rojunction films. The fourth section details work ch aracterizing hybrid films and the fabrication of photovoltaic cells from these films. The final section introduces Particle Induced Nanos tructuring, a process for hybrid film deposition intended to control the distribution of nanoc rystals in the polymer matrix. ITO Anode Treatment Organic solar cells incorporate transparent conducting substrates as an anode, with indiumtin-oxide (ITO) coated glass most often used. ITO films offer several positive characteristics as substrates for optical devices including a high luminous tr ansparency, good electrical conductivity, and good infrared reflectivity. For these reasons, ITO is widely adopted as transparent anodes in light-emitting diodes, liquid crystal displays, and solar cells [4, 81-82]. ITO coated glass substrates are commercia lly produced by sputter deposition followed by processes to improve surface roughness and micr ostructure. As-received substrate surfaces, however, have to be further proc essed prior to application to cu rrent flowing devices such as OLEDs and solar cells because a sputter-de posited surface microstructure and chemical 122

PAGE 123

composition can be degraded during extended de vice operation. Surface treatment of an ITO surface by several techniques can alter the chemical and physical properties of the surface such as work function, surface roughness, and oxidation property, and thus it could affect the device performance. In this study, nitrogen and oxygen plasma treatment along with electron bombardment of commercially available ITO-coated glass substrat es were revisited as approaches to improving the efficiency and stability of organic solar cells. The effect of these surface treatments on the surface morphology and chemical composition was ch aracterized, and organic solar cells with the structure ITO/PEDOT:PSS(50 nm)/CuPc(25 nm)/C60(15 nm)/Al(100 nm) were fabricated and the device performance measured. The substrate used in this study was commercia lly available ITO-coated glass with an ITO film thickness of ~ 1800 and sheet resistance of ~7 /sq. The as-received substrate was chemically cleaned by sequential ultrasonificatio n in trichloroethylene (TCE), acetone, and methanol, followed by nitrogen blow-drying. The cl eaned substrate was then exposed to either a nitrogen or oxygen plasma or an electron beam. Plasma treatment was ca rried out in a barreltype plasma chamber for 10 min at a power input in the range 50 to 300 W and pressure in the range 50 mTorr to 1 Torr with N2 gas flow. Electron beam irradiation was performed for 15 sec in a nitrogen environment with beam power varied from 0.5 to 2 kGy (kJ/kg). After treatment a PEDOT:PSS layer was spin-coated and dried in a vacuum oven. Organic films (CuPc and C60) and aluminum were deposited at room temperature in a thermal evaporator with a base pressure of 2.0x10-6 Torr. The chemical composition and surface morphology of the ITO surface were measured using XPS, AFM, and a video cont act angle system (VCAS). Power conversion 123

PAGE 124

efficiencies were measured under illumination fr om a solar simulator set to produce an AM 1.5 100 mW/cm2 spectrum. The measured contact angle of a water drople t on an ITO substrate changed significantly after surface treatment. The c ontact angle for untreated ITO s ubstrates was ~ 65, while films subjected to electron beam treatment showed a reduced contact angle of ~ 50. After exposure to the N2 plasma, the contact angle decreased dramatica lly to a value < 10. In both cases, the reduction is due to two effects. First, the e xposure to reactive radicals removes contamination from the film surface that may have remained after wet chemical cleaning. Secondly, these treatments increase the activity of the film surf ace by incorporating nitrogen radicals into the film. This effect is significan tly stronger in the case of N2 plasma than in the case of electron beam treatment because the plasma treatment su pplied a greater flux of highly reactive nitrogen radicals to the film surface. The change in surface polarity in th e case of electron beam treatment was not as significant as that in the ca se of plasma treatment, and as electron beam energy is significantly increased (larger than 2 kG y), the ITO film started to change its color to light gray, which degraded the luminous transparency of the substrate. AFM measurements were performed to quant ify the surface roughness of the ITO films. As-received ITO films showed a RMS surface roughness of 1.1 nm. This value was reduced to 0.8 nm by electron beam treatment and < 0.6 nm by N2 plasma treatment at optimized conditions. Sputtered ITO film generally contai ns irregular surface features, even though the subsequent polishing and annealing of the film improves its roughness. The surface treatment procedures used in this study are ex pected to attack the higher surf ace features first, resulting in a decrease in surface roughness [83, 84]. The effe ct, however, was not very significant, because the surface roughness of the as-receive d substrate was already low enough. 124

PAGE 125

XPS spectra from treated films showed a chan ge in the chemical composition of the film surface. Films treated with N2 plasma and electron beam bombardment showed an increase in nitrogen content accompanied by a decrease in oxygen composition, confirming that near-surface oxygen is replaced by nitrogen in films subjected to these treatments as shown in Table 3-1. In the case of N2 plasma exposure, there was a decrease in the In/Sn ratio of the films as well, which can cause a slight decrease in the work f unction of the films [85]. It is believed this change is a result of indium reacting with nitroge n radicals to form InN, which is subsequently removed from the film. The chemical shift of In-Sn-O bonding was also investigated by XPS, but no noticeable change in chemical shift was observed under th e treatment conditions investigated in this study. The incorporation of nitrogen into the n ear surface region of films treated in N2 plasma was confirmed by glow discharge spectroscopy (GDS). Films exposed to the plasma at a constant power but increasing pressure showed an increase in nitrogen levels in the film up to 700 mTorr. At higher pressures the nitrogen level decreased slightly, due to a decrease in nitrogen radical activity in the plasma. Organic solar cell devices were fabricated fr om the surface-treated IT O films to determine the impact on device performance as shown in Figure 3-1 and Table 3-2. Electron beam treatment produced no change in device effici ency; however, there was a decrease in VOC and an increase in JSC in these devices. O2 plasma treatment gave a slight increase in JSC, but a significant drop in VOC led to a reduction in overall device efficiency. N2 plasma treatment resulted in cells with efficiency nearly double th at of the untreated ITO cells. All treated ITO devices showed an improvement in the shape of th e I-V curve compared to that of untreated ITO devices. The changes in performance can be attri buted to several effects due to the treatment, 125

PAGE 126

including surface roughness, chemical stability, a nd reduced work function. Electron beam and N2 plasma treatment replace oxygen with nitrogen, re ducing the electron affin ity of the ITO film and resulting in a lowering of the work function that improves the charge collection efficiency. The effect of several ITO surface treatments on the performance of organic solar cells was determined. It was shown that exposure to a N2 plasma was more beneficial than either the oxygen plasma or e-beam treatment. The N2 plasma was most successful in improving the surface characteristics, as evidenced by the extent of lowering the contact angle and decreasing the surface roughness (AFM). This treatment incor porates nitrogen into the near surface region and produces a slight change in the In/Sn rati o, which reduces the ITO work function. These changes optimize the energy band diagram and im prove charge collection at the ITO anode. Bi-layer Organic Solar Cell Fabrication A process for fabricating bi-layer orga nic solar cells with the cell structure ITO/PEDOT:PSS/P3HT/C60/Al was developed. Figure 3-2 s hows the energy band diagram for a cell with this structure [45, 86]. This materi al system has received much attention for applications in bulk heterojunction solar cells due after the discovery of ultrafast charge transfer at interfaces between conjugated polymers and C60 molecules. In these bi-layer organic cells, P3HT serves as the primary absorber layer in the cells, with a bandgap of approximately 1.7 eV and a very str ong absorption coefficient. Excitons generated in the polymer are separated at the C60 interface, with electrons dropping to the lower energy level of the C60 and holes returning to the P3HT. Electrons are transported to the backside aluminum contact, while holes are transported through the P3HT layer and the hole transport layer (HTL) of PEDOT:PSS to the tran sparent ITO frontside contact. 126

PAGE 127

Cell Fabrication Procedure Bi-layer organic solar cells were fabricated on ITO-coated glass substrates. The substrates were patterned, cleaned, a nd treated in-house under N2 plasma. A hole transport layer of PEDOT:PSS and the ab sorber layer of P3HT were spin-coated onto th e substrate and dried under vacuum. The electron transport layer of C60 and the aluminum back contact were deposited by evaporation. For cells to be transported, gla ss encapsulation was perf ormed to extend the cell lifetime. The encapsulation process is described in greater detail below. A graphic synopsis of the fabrication process is shown in Figure 3-3. Substrate Preparation The cells were fabricated on ITO-coated gla ss substrates obtained from Samsung-Corning. The ITO films had a resistance of < 7 /sq. and a thickness of approximately 0.18 m. Asreceived ITO substrates were cut into 2.5 x 2.5 cm squares and patterned by HCl vapor etching. A 2 mm wide strip of elec trical tape was fixed to the ITO substrate to cover a strip that would become the ITO anode. The substrates were suspended from the inside lid of a glass dish with a small amount of HCl in the bottom of the dish to generate vapor. Vapor etching occurred over 25 min, after which the substrates were thoroughly rinsed with de-ionized water and blown dry with nitrogen. They were then chemically cleaned by successive 10 min sonication steps in trichloroethylene, acetone, and methanol, fo llowed by blow-drying under nitrogen. Cleaned substrates were subjected to N2 plasma for 10-min in a barrel-type plasma chamber wrapped in induction coils. The chamber offered a continuous flow of nitrogen at a constant flow rate during the treatment, and the conditions used for treatment were optimized to a supplied power of 50 W and a chamber pressure of 200 mTorr. 127

PAGE 128

Spin-Coating A solution of PEDOT:PSS in water obtaine d from Bayer was filtered with a 0.45 m syringe filter and spin-coated onto th e treated substrate for 30 sec. The film was then dried in a vacuum oven at 90 C for 30 min. A calibration cu rve was generated for dried film thickness vs. spin-coating speed and is shown in Figure 3-4. Th e curve was fit to power law as has been noted in the literature (87), with the curve fitting Equation 3-1. t = (e14.7)* -0.99 (3-1) The absorber layer of P3HT was then spin-coated from a solution of 5 mg/ml in chlorobenzene. The films were spin-coated for 30 sec and then dried under vacuum. The film thickness versus spin-coating speed was fit to Eq uation 3-2, and the results are shown in Figure 3-5. t = (e12.5)* .93 (3-2) Both spin-coating steps occurred in a clean room environment. Film thicknesses for these measurements were performed with a profilomete r. After the active layer was dried, selected areas of the substrates were wiped clean of th e polymer films to provide clean surfaces for external contacts. Evaporation The cells were placed in a glove box under nitrog en and loaded into a thermal evaporator. C60 was evaporated through a shadow mask at a ra te of approximately 2 /s under a pressure of 10-6 Torr. Finally, a thick layer of aluminum wa s rapidly evaporated through a second shadow mask to form the back contact of the cells. Encapsulation Encapsulation was performed to extend the lifetime of the fabricated cells by shielding them from moisture in the atmosphere. The two components of the encapsulation procedure are 128

PAGE 129

the encapsulant and the dessicant. The encapsulant consisted of a small gla ss slab to serve as the backing and a rubber spacer to prevent contact be tween the substrate and encapsulant glass. The assembly of substrate / spacer / en capsulant was sealed with epoxy. Prior to fixing the encapsulant onto the substrat e, a dessicant was adde d to protect the cells from moisture exposure. The dessicant used wa s barium oxide. A small pouch was created from part of a piece of weighing paper. Inside th e nitrogen glove box, the pouch was filled with BaO powder and sealed with double-sided tape. It was then affixed to the inside of the encapsulant glass. Several tiny holes were punctured in the pouch to allow moisture to reach the dessicant, and the final assembly was sealed to the substrate with epoxy. Film Drying Due to the potential impact of residual solvent in these films, care must be taken to ensure drying is complete after each film deposition step. Residual organic solvent from the active layer film serves as an insulator to cripple electrica l performance, while excess water remaining from the HTL film can oxidize and degrade the active layer polymer. To confirm the effectiveness of the drying step, FTIR spectra were compared betw een the dry films, the solutions used for spincoating, and spectra obtained from Sigma-Aldrich for the pure solvent. Figure 3-6 shows the spectra for a P3HT film spin-coated from chlorobenzene, as well as spectra for the solution and pure solvent. The large peak at approximately 3050 cm-1 in the chlorobenzene spectra is clearly visible in the solution spectra, but is noticeably missing in the P3HT film spectra. Also, several sharp, narrow peaks between 1600 and 500 cm-1 match in the solution and solvent spectra, but are missing from the film spectra. In Figure 3-7, the broad O-H stretching peak from approximately 3700 to 3000 cm-1 is an obvious feature in the FTIR spectrum for pure water. This wide peak is also obvious in the soluti on spectrum but is missing from the PEDOT:PSS film spectrum. Similarly, the str ong peak at approximately 1640 cm-1 in the water spectrum is present 129

PAGE 130

in the solution spectrum but missing from the film spectra. These FTIR spectra provide confirmation that film drying is complete under the conditions specified, and in that respect, these films are suitable for use in organic solar cells. Both solvents analyzed, chlorobenzene and water, produce strong peaks that are easily distin guished from the polymer spectrum, making this an effective technique for identifying residual solvent in the films after deposition. PEDOT:PSS The PEDOT:PSS layer in bi-layer solar cells wa s deposited with a thic kness in the range of 80 to 100 nm. One potential issue in this charge transport layer is that it is susceptible to pinholes, which can lead to shor ting of the devices. Pinholes can be generated during the drying process as the solvent evaporates and rises through the drying pol ymer. To see the if pinholes were problematic in this deposition process, cells were fabricated using a single-layer and double-layers of PEDOT:PSS. The double-layer devi ces should eliminate the presence of layerspanning pinholes by providing two separate films so that any pinholes would only span half of the final film. Experiments were performed by fabricating bi -layer solar cells using singleor doublelayers of PEDOT:PSS. A consistent final film thickness of 80 nm was used for the PEDOT:PSS layer: one 80 nm layer for the single-layer devi ces, and two 40 nm layers for the double-layer devices. All other cleaning, preparation, deposition, and encapsulation steps were held constant. After fabrication, the ce ll performance was characterized under 100 mW/cm2 light from a solar simulator. The resulting J-V curves are show n for single-layer devices in Figure 3-8 and for double-layer devices in Figure 3-9. From the results, it is clear that the performance of cells using the double-layer PEDOT:PSS film suffer dramatically. Of the fo ur cells using the double-la yer structure, two fail to show any diode characteristics, while the ot her two show only a minimal photovoltaic effect. 130

PAGE 131

While the performance of the cells using th e single-layer PEDOT:PSS were not phenomenal, they all showed a significant photocurre nt and diode characteristics. Cells fabricated with a single 80 nm layer of PEDOT:PSS showed efficiencies as high as 0.168% (cell 8-2), with all cells showing efficiencies of at l east 0.047%. In contrast, cells fabricated with the double-layer structure failed to show a measurable efficiency, although estimates put them in the range of 0.015% for the be st cell (cell 9-1). Open circuit voltage values for the single-layer cells were in the range of 0.15 V, while valu es for the double-layer cells are less than 0.05 V. Short circ uit current density values ranged from 1.5 to 2.5 mA/cm2 for the single-layer cells and from 0.7 to 1.4 mA/cm2 for the double-layer cells. The data show that pinholes are not a c oncern for the PEDOT:PSS films under these deposition and drying conditions. In fact, there is a different effect causing the double-layer films to perform more poorly than the single-layer films. The to tal thickness of the layers was held constant to keep series re sistance constant. However, the series resistance is impacted by the film resistivity in ad dition to the path length. It seems that the double-layer films showed a higher resistance to current flow due to the lower cell performance. For positive biases, the single-layer films showed current densities 7 10 % higher than the double-la yer films. This can be attributed to two possible cau ses interfacial resistance and f ilm resistance. Because of the double-layer structure, there is an extra film interface which could cause an increase in the overall resistance. Additionall y, the inherent film re sistivity could be increased due to the deposition conditions. The films were deposit ed via spin-coating, w ith the 40 nm films deposited at a much higher rpm than the 80 nm fi lms. This causes faster solvent evaporation and can inhibit the ability of the polymers to self -align in a configurati on that could minimize 131

PAGE 132

resistance. This effect has been observed in the past, particularly in acti ve layer films where extremely slow solvent evaporation has resulted in strong increases in the films conductivity. Another effect observed in PEDOT:PSS deposition is that of different particle filtering speeds that produce different solution properties. In polymer electronics processing, PEDOT:PSS is typically filtered prior to deposition to remove particles in the solution and provide a smoother film upon spin-coating. In our labs, the PEDOT:PSS solution was filtered by hand using a 0.45 m disc filter attached to a 100 ml syri nge. It was found that there were two different methods of filtering depending on the person doing the work, but because the work was done by hand rather than with a mechanical system. In one method, the solution was filtered very slowly, with the worker applying just e nough pressure to force the solution through the membrane in a dropwise fashion. The filter was pe riodically replaced as the filtration became more difficult due to clogging. This slow filtrat ion method resulted in a relatively low viscosity solution. In the other method, the solution was fo rced through the filter in one steady motion. This process was much faster, with several ml of the solution passed through in a matter of seconds, and resulted in a more viscous solution. The difference was noticed while developing calibration curves for film thickness depending on sp in-coating speed. The slow filtration results in a much thinner film than the fast filtration st ep because a lower percentage of the polymer is forced across the membrane at the lower pressure difference. Calibration curves using the two solutions are shown in Figure 3-10. The data set labeled 116 represents the slow filtration method, while the other three data sets show the faster filtration method. For cell fabrication, the fast filtration method was used to provide the ability to deposit approximately 100 nm thick films while still using a reasonably fast spin-coating speed to pr oduce smooth, uniform films. 132

PAGE 133

P3HT Prior to cell fabrication, P3HT films were analyzed under AFM to determine the surface quality of the films. Measurements were made under different solution concentrations and spincoating speeds. The results, s hown in Figure 3-11, show high-qua lity films with low surface roughness, indicating their suitability for bi-layer cell construction. All films spin-cast from the 5 mg/ml solution showed an RMS surface roughness of around 1 nm. For the films cast from the 10 mg/ml solution, the RMS surface roughness was between 2.5 nm and 3.6 nm, with the higher roughness occurring at the slowest spin speed. The data from the images shown in Figure 3-11 is tabulated in Table 3-3. Bi-layer Cell Fabrication Following these film and process characteriza tion investigations, bi-layer organic solar cells were fabricated with th e device structure ITO/PEDOT:PSS/P3HT/C60/Al. The bi-layer device structure is commonly used in OLED devi ces [88] and has been explored in molecular organic photovoltaics [89], but le ss work has been performed rega rding bi-layer devices using polymer active layers [90]. The devices were fabr icated on ITO-coated glass substrates prepared as described previously. A PEDOT:PSS film was deposited by spin-coating for 30 sec at 2500 rpm and drying for 30 min. The active layer of P3HT was deposited by spin-coating from a 5 mg/ml solution in 1,2-dichlorobenzene for 30 s ec at 3500 rpm and dried under vacuum. The samples were loaded into an ev aporator where a 150 film of C60 was deposited, followed by a 800 Al electrode. Bi-layer cell J-V measurements were perfor med at Busan National University in Busan, South Korea using a Keithley I-V measurement system under illumination from a solar simulator. Current measurements were convert ed to current density by dividing by the active cell area (0.04 cm2). Illuminated measurements were performed under 100 mW/cm2 133

PAGE 134

illumination. Power conversion efficiency ( ) was calculated using Equation 3-3, with Vmax, Imax, and Jmax representing the voltage, current, and cu rrent density at the maximum power point. The fill factor (FF) is calculated using Equation 3-4 with VOC and JSC being the open circuit voltage and short circu it current density. %100 /100 %1002 maxmax maxmax cmmW JV P IVin (3-3) SCOCJV JV FFmaxmax (3-4) The rectification ratio (RR) of a diode is the ratio of forward to reverse current at some applied bias. Higher rectificati on ratios show stronger diode characteristics in th e I-V curve for devices. Rectification ratios displayed in this section were calculated from dark current measurements at 0.5 V of forward a nd reverse bias unless otherwise noted. Bi-layer solar cells were fabricated using the procedure detailed pr eviously, but with no treatment performed on the ITO electrode. J-V cu rves for these cells are shown in Figure 3-12. The cells fabricated in this set all show ed a measurable photocurrent, demonstrating functioning solar cell behavior. The best-performing cell in the se t, Set I 3, showed a power conversion efficiency of 0.04%. Despite having the lowest VOC of all cells in the set, this champion cell showed a short-circu it current density of 1.55 mA/cm2, which was significantly higher than any other cell in the se t. Performance was low, but measu rable, in all cells. With the exception of cell I 3 with a VOC of 0.11 V, all cells showed a VOC of almost exactly 0.15 V. The fill factor for the cells ranged from 0.15 for ce ll I 4 to 0.24 for cell I-1. Cells I-1 and I-4 showed a JSC of approximately 0.5 mA/cm2. The JSC for cell I-2 was 1.03 mA/cm2. Another set of bi-layer cells were fabricated with the same cell structure, but applying plasma treatment to the ITO substrate. The perf ormance of these cells was poor, with the lack of 134

PAGE 135

performance attributed to possible poor encapsulation resulting in cell decay before measurement, or non-optimized plasma treatment inhibiting the cell performance. These cells showed a minimal photovoltaic effect, with open-circuit voltages below 0.1 V and short circuit current densities no high er than 1.10 mA/cm2. Although fill factors were in the range of 0.200.25, efficiencies were no higher than 0.02% due to the low photovolta ge and current in the cells. J-V curves for the cells are shown in Figure 3-13 and the cell perfor mance is detailed in Table 34. Solvent Comparisons Various solvents were compared to dete rmine their applicab ility for hybrid bulk heterojunction films. The solvents were all commonly available chemicals, so no high-cost specialty materials were used th at would add substantial cost to the fabrication process. The goal was to identify a solvent or a family of solvents that woul d provide the highestquality hybrid films for cell fabricat ion. The solvents used for th e tests are shown in Table 24. Only those which showed an approp riate level of solubility were u sed in further testing. Polymer solubility was qualitatively asse ssed by mixing a small amount of the polymer with a few milliliters of the chosen solvent in to a small vial and subjecting th e mixture to ultrasonication for at least one hour. Solvents with solubility labeled as No showed solid flak es of the polymer in the clear solvent after the mixing. Those labeled as Poor result ed in a color change of the liquid to indicate some degree of dissolution, bu t still contained significant solid particles of polymer. The label of Yes indicates that the po lymer was fully dissolved to give a red colored solution that is characteristic of P3HT. The solvents normal bo iling temperature and polarity values are listed in Table 3-5 as these parameters strongly impact the f ilm-forming properties of a solution. The values of polarity in the chart were taken from a solvent miscibility and polarity chart from Phenomenex [91] where higher values correspond to more polar solvents, with water 135

PAGE 136

having a value of 9.0. In the table, DMF is di methyl formamide, MED is methyl ethyl ketone, THF is tetrahydrofuran, and TCE is trichloroethylene. Hybrid films deposited from the solv ents that showed solubility for P3HT were analyzed with profilometry, AFM, and imaged with optical microscopy and SEM. Film roughness measurements taken from profilometry are shown in Figure 3-14. The results are plotted against both boiling temperature and polar ity, although no trend is evident in either case. These measurements were taken over 5 line scans of 1.0 to 1.5 mm on the surface of two separate substrates. AFM surface scans were taken for films deposited using 6 of the select ed solvents, and rms surface roughness was measured for 5 x 5 m and 1 x 1 m surface areas. The results for each test are shown in Figures 3-15 and 3-16, with separate graphs shown to plot rms surface roughness vs. solvent boiling temperature and solvent polarity. On the large-scale profilometer measurement s, chlorobenzene was the best performing solvent, showing a mean surface roughness of 64.9 nm on a line scan. The worst performers were THF and o-xylene, with mean rms values of 136 and 198 nm, respectively. For the 5 x 5 m AFM measurements, chloroform showed the lowest roughness, with the rms roughness being measured at 16.2 nm. THF displayed the highest roughness, with a value of 36.8 nm. For the small-area 1 x 1 m AFM measurement, Toluene was the supe rior solvent with an average rms value of 2.93 nm. Again, THF showed the hi ghest value, at 15.0 nm. The results are summarized in Table 3-6. SEM and optical microscope images of film su rfaces were taken to vi sually assess the film quality in conjunction with the surface roughness data. Figure 3-17 shows optical images of 136

PAGE 137

selected films that were represen tative of the samples taken for each solvent. Figure 3-18 shows SEM images from films, also representative of the data taken fo r each film solvent. In the optical microscope images, dark s pots on the image represen t surface features on the films. Some of the larger dark spots on the imag e, such as in the top left corner and near the right hand side of the 100x THF image, appear fuzzy because they are beyond the depth of field of the image, and different areas of this spot co uld be visualized by re-foc using the lens. In the 10x magnification images, it is noted that all solvents produce at least a few of these large black features. Chlorobenzene, o -dichlorobenzene, and toluene show these spots in smaller sizes and regularities than the other materials. Benzene shows large wispy collections of the dark spots that were not seen from other solvents, which seem to suggest precipitation or clustering occurring on a much larger scale than the other solvents show. These feat ures are also visible in the 100x image for benzene, and again, it does not appear for other solvents. The 100x image for THF shows a high frequency of very large features. Dichlorobenzene, on the other hand, shows a very low frequency and size of features in both the 10x and 100x images. Chlorobenzene shows a few large features, but the image overall shows a high-quality film with a low frequency of features. The SEM images shown in Figure 3-18 displa y surface images of the hybrid films at 5,000x and 15,000x magnification. These magnification levels are relatively low for SEM images, but attempts to focus the electron beam to produce images in the 30,000x to 50,000x range resulted in beam damage of the sample. In fact, an example of this is visible in the 15,000x image for o -dichlorobenzene. The dark box near the center of the image, just to the left of the bright surface feature, represents an area where a tighte r focus was attempted and the sample was burned. In these imag es, surface features of the films are clearly visible as bright 137

PAGE 138

spots in the secondary electron imag e. Images for chlorobenzene and o -dichlorobenzene show high-quality films with relatively few surface features. The images for TCE also shows a minimal amount of surface features, particularly in the 15,000x image. Chloroform displays a moderate amount of features, but less so than THF, which simila rly to the optical images shows the poorest film quality. Attempts to identify the composition of th ese surface features failed to yield results. Backscattered electron images of the films showed images of the features similar to the secondary electron images, but with no additional bright spots to demonstrate possible clustering of the higher density nanocrystals. Compositiona l scans were performed using energy-dispersive x-ray analysis (EDX), and these scans detect ed cadmium and selenium at a constant concentration in both the features and in smooth areas of the film surface. From this study, it was determined that chlo roform, chlorobenzene, and o-dichlorobenzene are good candidates for hybrid bulk heterojunction film deposition. These solvents all produce films that were among the low est in surface roughness for all of the measurement techniques used. Additionally, the film quality could be visually confirmed from optical microscopy and SEM surface images. Hybrid Bulk Heterojunction Cell Fabrication Bulk heterojunction pho tovoltaic cells were fabricated using P3HT as the absorbing semiconductor and nano-CdSe as the electron transpor ter. The cell design was the modeled after the bi-layer organic cell design described previously. The bulk heterojunction ce ll structure was ITO/PEDOT:PSS/P3HT:CdSe/Al, which is very similar to the bi-layer structure with the exception of CdSe replacing C60 as the electron acceptor, and that acceptor is now blended into the active layer film ra ther than deposited on top. Performance measurements for these cells were performed in-house rather than remotely, so the encapsulation process was not performed. 138

PAGE 139

Measurements were performed with a Keithley I-V-L measurement system under illumination from a JEOL solar simulator. Nanocrystal Synthesis and Surfactant CdSe nanocrystals were synthesized using the solution-phase growth mechanism demonstrated by Peng et al. [92]. In this met hod cadmium oxide is adde d to a flask containing trin -octylphosphine oxide (TOPO) and hexyl-phosphonic acid. The solution is then heated before the addition of selenium powder dissolved in liquid trin -octylphosphine. The reaction progresses and is halted by removal from the hea ting source. The nanocrystals, coated in a surfactant layer of TOPO, are pr ecipitated from the solution by the addition of methanol and centrifugation. The TOPO-coate d nanocrystals can further be modified by dissolution in pyridine, precipitation using hexane, and centrifugation to isolate the crystals. This ligand exchange process replaces the TO PO molecules with pyridine mo lecules on the surface. After ~5 steps of this process, the TO PO coating is fully replaced a nd the pyridine-coated nanocrystals can be used for processing. CdSe nanocrystals u sed in this dissertation were synthesized by Md. Azizul Hasnain and Trong Nguyen Tam Nguye n at Yeungnam Universi ty unless otherwise noted. Solutions were generated using CdSe nanocryst als with both types of surfactant coatings, and films deposited from these solutions were tested to determine the appropriate deposition parameters. The solutions consisted of a 60:40 mixture of CdSe nanocrystals and P3HT polymer dissolved in chloroform with a variable volume fraction of pyrid ine added to enhance the CdSe solubility. The surface roughness of spin-coated films is shown in Figure 3-19, plotted against the pyridine content of the solution. Note th e scale difference in the two graphs. These measurements were performed by Md. Azizul Hasnain at Yeungnam University and are displayed to clearly represent the effect of nanocrystal surfactant on film quality for hybrid films. 139

PAGE 140

For TOPO-coated nanocrystals, the surface rough ness of the films decreased as the pyridine content increased, up to 50%. This is due to the low solubility of TOPO-coated nanocrystals in chloroform. Pyridine concentrations of 45-50% we re required to generate films with less than 10 nm rms roughness. After surface exchange with pyr idine is performed, the pyridine-terminated nanocrystals become significantly more soluble in chloroform. In this case, pyridine concentrations of less than 10% yield rms su rface roughness values of less than 10 nm. Hybrid Films To fabricate bulk heterojunc tion solar cells with organic polymers and inorganic nanocrystals, great care must be taken to ensu re proper mixing between the two phases. The exciton diffusion length of most semiconducting polym ers is in the range 5 to 20 nm. Because of this limitation, the organic and inorganic phases in the hybrid active layer mu st be well-mixed so that excitons generated in the organic phase can reach an inorganic phase that is within one diffusion length. In this investigation, the properties of hybrid films is studied through optical microscopy, electron microscopy, atomic force microscopy, and surface profilometry. Solutions are generated by dissolving blends of P3HT polymer and CdSe nanopowde r into solvent mixtures of chloroform and pyridine. Hybrid films are sp in-cast from these solutions onto ITO-coated glass substrates that were subjected to N2 plasma treatment as described previously, and the films were dried under vacuum. TOPO-coated CdSe Initially, hybrid solutions were created with P3HT and TOPO-coated CdSe nanocrystals with a radius of approximately 5 nm. The solven t used in these solutions was a 50-50 mixture of chloroform and pyridine, based on the results show n in Figure 3-19. Three hybrid solutions were prepared with composition shown in Table 3-7. 140

PAGE 141

Films were spin-cast from these solutions, a nd their visual quality was studied using optical microscopy. The resulting images are s hown in Figure 3-20. The images were taken at 100x magnification, and the films imaged were spin-coated at 3000 rpm from the solutions described in Table 3-7. In these images, darker regions correspond to features on the surface of the films. From the images, it is observed that films from the higher-concentrated solutions show significantly more numerous and larger features than the f ilm from the weaklyconcentrated solution. The w eakly-concentrated Solution 3 show s only small and well-dispersed dark regions. Solutions 1 and 2 show a darker overall image, as is to be expected as these higher-concentrated solutions produce thicker film s. However, they al so show large surface features, some of which appear out of focus due to the depth of field of the microscope. These observed differences are due to the low solubility of both the polymer and nanocrystals in these composite solutions when TOPO-coated nanocrystal s are used. The polymer is poorly soluble in pyridine, and the TOPO-coated CdSe nanocrystals are poorly soluble in chloroform, so the only way to control the film morphology in this solvent sy stem is to reduce the to tal solute lo ad in the solution. The 20.4 mg/ml and 19 mg/ml solutions produced a significantly rougher surface than that of the 5 mg/ml solution, seen in the images in Figure 3-20. In an attempt to decrease the roughness of these films, a second spin-coating step was performed using pure chloroform. The hypothesis was that the solvent would selectively attack the mo re pronounced surface features, similar to the effect of nitrogen plasma smoothi ng the surface of ITO. The resulting film did show a decrease in roughness, simila r to that of the 5 mg/ml solu tion. An optical microscope image of the film is shown in Figure 3-21. The surface roughness decreas ed, but the film was nearly completely etched during this process. Even under spin speeds up to 8000 rpm, designed 141

PAGE 142

to significantly reduce the contac t time between the chloroform and the film, the original film was nearly completely removed. P3HT solubility in chloroform and pyridine To further investigate the effect of the so lvent composition on film qualities, studies were performed to compare the quality of pure P3HT films cast from pure chloroform, a 1:1 chloroform:pyridine mixture, and pure pyridine. Films were de posited from 5 mg/ml solutions of P3HT in each of the three solvent types a nd then examined under optical and secondary electron microscopes and the roughness measured with a surface profilometer. Under an optical microscope at 100x magnificat ion, clear differences are observed in film quality depending on the type of solvent used, as shown in Figure 3-22. The chloroform solution produces a film that appears smooth with a number of small surface features. The film deposited from the mixed solvent shows a rougher base f ilm with larger and more numerous surface features. The pyridine solution results in a smooth base film, but with sev eral very large surface features. SEM analysis of the chloroform and mixed so lvent films show results similar to those obtained under optical microscopy. Figure 3-23 displays a co mparison of these films at 2,000 and 10,000 x magnification. For the films deposited from chloroform solvent, there is a significant reduction in the number of surface features of the films. In the 2,000x image, there is one large area feature that is visible at the botto m of the image, but this was not observed to be common in the film. On the ot her hand, the films deposited fr om the mixed solvent show a distribution of surface features wi th varying shape and sizes rangi ng from a few microns to tens of microns. This offers further evidence of ph ase separation occurring in the films deposited from the mixed solvent. Because these films are pure P3HT rather than hybrid films, the observed features must be regions of P3HT that formed a non-uniform surface. This could occur 142

PAGE 143

through non-uniform precipitation of the polymer dur ing deposition. In the chloroform solvent, due to the high polymer solubility the smooth film is a result of a uniform deposition and film drying process. In the mixed solvent, however, the solubility is poorer and the P3HT will precipitate more quickly as solvent evaporates during the spin-coating and drying processes, resulting in the polymer freezing in its current state rather than being allowed to relax to a preferred alignment that r esults in a smoother film. Surface profiles of the films m easured through profilometry revealed films with significant increases in surface roughness as the amount of pyridine in the solvent mixture increased, as shown in Figure 3-24. When the film thickness was m easured using this technique, it was found that the film thickness decrea sed as the pyridine content of the solvent increased, due to decreasing solubility of the P3HT polymer. The profiles shown in Figure 3-24 are the result of a line scan across the surface of a sample that had the film stripped away on the left-hand side. The tall, wide peak at approximately 1 mm is the edge where the film was wiped clean and the film surface was dist orted. The polymer film is shown on the right-hand side, which is where the mean and median film thickness was measured. On the profiles, the solid teal line represents the median film thickness, the red dash-dot line repr esents the mean film thickness, and the blue dashed lines represent one standa rd deviation from the mean. Note that not all lines appear in all graphs due to the scaling. These scans show an increase in the mean film thickness as the pyridine content of the solvent increases, resulting from an increase in surface rough ess. They show, however, a decrease in median film thickness, due to the base film thickness being smaller as the polymer is poorly dispersed by the pyridine solvent. These properties are summarized in Table 3-8, and represent averages for measurements across different regions of each sample. It is interesting to 143

PAGE 144

note that for the pyridine solvent, the error range for the mean film thickness and the standard deviation of film thickness is larger than the ac tual measurement. This is due to the extremely nonuniform film which shows extremely larg e features over a very thin base film. Cell Fabrication Replacing TOPO-coated nanocrystals with pyrid ine-coated nanocrystals is the key to reducing the amount of pyridine needed in the solvent mixture. Because of this, pyridine-coated nanocrystals were used for attempts at hybrid bulk heterojunction ce ll fabrication. The processing steps for these cells were similar to those used for th e organic bi-layer solar cells described earlier. ITO-coated glass substrates were etched with HCl, then cleaned under ultrasonication in TCE, acetone, and methanol The clean surface was exposed to a N2 plasma under specific conditions, followed by deposition of PEDOT:PSS, then the hybrid solution, followed by evaporation of the Al electrode. Cell performance was tested in the dark and under simulated solar illumination. Cells deposited from chloroform solutions Bulk heterojunction cells were fabricated from pure chloroform solu tion and a 2% pyridine in chloroform solution. Nitrogen plasma treatment was pe rformed at 50 W and 200 mTorr for 10 minutes after etching and cleaning of the substrates. A thin film of PEDOT:PSS was deposited via spin-coating and dried under vacuum. The hybrid solution concentr ation was 5 mg/ml, consisting of 60% CdSe by wei ght. The aluminum electrode was between 125 and 150 nm in thickness. The resulting dark and light J-V curv es for these cells are shown in Figure 3-25. The active cell performance is low for both cells, particularly the cell deposited from pure chloroform solvent. The first cell, deposited fr om the chloroform-pyridine mixed solvent, shows JSC = 8.99 x 10-3 mA/cm2 with a maximum efficiency = 7 x 10-4 %. The second cell, in pure chloroform solution, shows JSC = 2.62 x 10-5 mA/cm2 and maximum efficiency = 1.3 x 10-6 %. 144

PAGE 145

Although these films show some photovoltaic effect it is extremely low, presumably due to the very thin nature of the films. From visual insp ection, the films are faint pink in color and highly transparent. These thin films fail to absorb a significant number of photons to yield highperformance solar cells. To enhance film thickness, a new hybrid so lution was created with the total solute concentration doubled to 10 mg/ml while maintaining the nanocrysta l weight percentage at 60%. The solvent was 2% pyridine in chloroform. The resulting J-V curves are shown in Figure 3-26. This cell showed higher values of JSC than the cell deposited from the 5 mg/ml solution. The measured cell performance characteristics were JSC = 3.39 x 10-2 mA/cm2, VOC = 0.256 V, FF = 0.275, and efficiency = 2.4 x 10-3 %. This is a 2.5x increase in efficiency compared to the 5 mg/ml hybrid solution. Although performance is s till low, the film produces photocurrent that is half an order of magnitude higher. Thicker-film cells in chloroform solution By further increasing the concentration of the active layer solution, light absorption in the cells can be further enhanced. In these cells the weight ratio of the hybrid film was altered from 60% CdSe to 50% CdSe. Because the CdSe serv es primarily as an elect ron transporter while P3HT is the absorber, the cell pe rformance can be improved by decreasing the CdSe volume in the hybrid films as long as electron transport path ways exist to allow current flow to the electrode. A 25 mg/ml hybrid solu tion with equal weights of P3HT and pyridine-coated CdSe was dissolved in a solvent of ch loroform with 2% pyridine. To determine the effects of the nanocrystals in the solution, a so lution consisting of 12 mg/ml P3HT in the 2% pyridine mixed solvent was also generated. This solution has essen tially the same P3HT concentration in solution (12.5 mg/ml in the hybr id solution, 12 mg/ml in the pure polymer solution), so the presence of nanocrystals in the hybri d solution differentiates the two. 145

PAGE 146

Surface profiles of the films generated with a profilometer are shown in Figure 3-27. For these scans, the film was wiped cl ean at the left-hand side of the image to allow measurement of the film thickness. Note that the scale for both prof iles is the same. The P3HT film shows a thickness of 95.7 17.8 nm with an rms roughne ss of 15 nm. The hybrid solution produces a slightly thicker film at 130 16.8 nm, but with an rms roughness of 239 nm. The inclusion of nanocrystals increases the surface roughness of the film by more than an order of magnitude. However, the nanocrystals have only a minimal effect on the film thickness. Although the hybrid solution had a total solute concentra tion that was twi ce as high as the P3HT solution, the final film thickness only increased by 35%. This demonstrates that the polymer is a much stronger factor in film thickn ess than the nanocrystals. Dark and illuminated J-V curves for cells fa bricated from the 25 mg/ml hybrid solution and the 12 mg/ml P3HT solution are shown in Figure 3-28. The hybrid cell showed performance characteristics of JSC = 4.99 x 10-2 mA/cm2, VOC = 0.701 V, FF = 0.232, and efficiency = 0.0175%. These measurements show a short-circuit current density 150% higher than that of the 10 mg/ml hybrid cell shown in Figure 3-26. As expected, J-V curves for the pure P3HT cell showed poorer performance than that of the hybrid cell. Perf ormance characteristics for the P3HT cell showed JSC = 2.03 x 10-2 mA/cm2, VOC = 0.523 V, FF = 0.310, and efficiency = 1.25 x 10-3 %, which are similar to that of the 10 mg/m l hybrid cell. Despite the poor measurements for the P3HT cell, the dark J-V curve shows a strong rectification ra tio, signifying a strong diode. J-V measurements in the dark for each of th ese cells showed a str ong rectification ratio, something that was not observed previously for thinner cells. This shows that the thicker films limit reverse leakage current that was easily driven through the thin active films in the previous cells. 146

PAGE 147

Cell lifetime measurements were also performed fo r these two cells. The cells were periodically measured under dark and illuminated conditions under exposure to room air. The cell performance decreased quickly with exposur e time, as seen in Figure 3-29. The first measurements for each cell occu rred after approximately 40 minutes after removal from a nitrogen glove box. The short-circuit current density for the hybrid cell decays approximately 70% in 12 min between the firs t and second measurements and by more than 95% in the 30 min between the first and third sets of measurements. The short-ci rcuit current density for the P3HT cell decays more slowly by approximately 40% in 11 min between the first and second sets of measurements, 50% in 26 min betw een the first and third sets, and 80% in approximately 3.5 hr between the first and final measurements. The da ta was fit to an exponential decay function of the form shown in Equation 35 with J in units of mA/cm2 and t in units of min. This fit was performed using Sigmaplot, and the resulting eq uations are displayed ove r the graph in Figure 329. Rsq values for the curves were 0.939 fo r the hybrid cell and 0.950 for the P3HT cell, showing a good fit for the data range. )exp(0tbaJJ (3-5) The hybrid cell shows a half-life of 7.42 min, while the P3HT cell shows a half-life of 16.05 min. This short half-life was confirmed for other hybrid cells fabricated in the same data set. It is interesting to note the difference in the initial and final values extrapolated from the exponential decay curves. The hybrid cell curve is extrapolated to an initial value of JSC = 1.53 mA/cm2, while the P3HT cell curve shows an in itial value of 8.09 x 10-2 mA/cm2. The P3HT cell, however, shows a JSC approximately twice as high as that of the hybrid cell as time approaches infinity: 5.80 x 10-3 mA/cm2 vs. 2.89 x 10-3 mA/cm2. These extrapolated values cannot be taken as absolute truths due to the uncertain nature of curve-fitting, but the 147

PAGE 148

experimental data in Figure 3-29 confirm that the short-circuit current density for the hybrid cell is initially higher than that of the P3HT cell and drops to a lower value after approximately 60 min of exposure time. The cause of the accelerate d decay for the hybrid cells as compared to the pure P3HT cells is unknown. The exposure times shown in Figure 3-29 were measured from the time the cells were removed from a glove box after the back contact de position, but this may not be an appropriate measurement of the exposure time, as exposure occurred throughout the fabrication process. After weighing out appropriate amounts of the polymer and nanocrystals, both of which were stored in a glove box under nitrogen, these material s were removed from the glove box in a vial where the solvent mixture was added under exposure to room air. Solvents were not stored in the glove box to prevent contamin ation of that environment. Th e solutions were sealed in their vials to limit further exposure dur ing mixing, but the vial contained room air rather than nitrogen at this point. Spin-coating occurred in a cleanro om environment to limit particle contamination, so the air was filtered and humidity-controlled to some degree, but not inert. After the films were dried under vacuum, they were again opened to the cleanroom atmosphere where the films were wiped clean in the electrode contact areas. At th is point, the cells en tered the glove box for electrode deposition. From the time that the solid P3HT was removed from the nitrogen environment to the time that the J-V curves of fi nished cells were measured, the total amount of air exposure could vary from 75 min to over 2 hr during the fabrication process. A large part of this exposure occurs while the P3HT is in solution, as the mixing tim e was typically at least 1 hr. A second set of cells fabricated from these so lutions yielded improved performance for the hybrid cell. For these cells, car e was taken to minimize the am ount of environmental exposure of the cells and films prior to testing. Approximately 15 min elapsed from when the finished 148

PAGE 149

cells were removed from the glove box and when J-V testing was performed. The best hybrid cell in this set showed performance measures of JSC = 0.205 mA/cm2, VOC = 0.705 V, FF = 0.288, and efficiency = 0.0416 %. The top performing P3HT cell resulted in JSC = 5.26 x 10-2 mA/cm2, VOC = 0.237 V, FF = 0.242, and efficiency = 3 x 10-3 %. The J-V curves for these cells are shown in Figure 3-30. The hybrid cell shown in Figure 3-30 displays a 300% improvement in short-circuit current density and a 24% improvement in fill factor when compared to the hybrid cell from Figure 91. The open-circuit voltage remained nearly constant, resulting in a nearly 140% improvement in power conversion efficiency. The P3HT cell shown in Figure 3-30 shows a diode with a lower rectification ratio as compared to the one shown in Figure 3-28. This cell showed a nearly 160% increase in short-circuit current density, but a 55% reduction in th e open-circuit voltage resulted in an approximately 20% reduction in fill factor. Despite these changes, the maximum efficiency improved by 140%. These cells saw approximately 15 min of air ex posure between removal from the glove box and J-V measurement. Using the short-circuit current density vs exposure time curves shown in Figure 3-29, the predicted exposure times for th ese cells would have been 21 min for the hybrid and 10 min for the polymer. This is within abou t 5 min of the actual exposure time, which is a reasonable prediction considering this curve doe s not take into account exposure time accrued during the fabrication process. Chlorobenzene solvent Based on comparisons of hybrid films deposited from various solvents, chlorobenzene was found to be a solvent that granted good mor phology with low surface roughness. To test the performance of hybrid bulk heterojunction cell s deposited from chlorobenzne, a 30 mg/ml solution was generated with a 1:1 mixture of P3HT and pyridine-coated CdSe dissolved in a 149

PAGE 150

mixture of 98% chlorobenzene with 2% pyridine. The best cell generated from this solution showed maximum performance of JSC = 0.138 mA/cm2, VOC = 0.391 V, FF = 0.292, and efficiency = 0.0158%. Dark and illuminated J-V curves for this cell are shown in Figure 3-31. This cell showed a lower short-circuit current cu rrent and open-circuit voltage than the similar cell deposited from a 25 mg/ml hybrid solutio n in a chloroform:pyridine solution. Commercial CdSe nanocrystals To test the quality of the synthesized CdSe nanocrystals used for cell fabrication, commercial CdSe nanopowder was acquired from Meliorum Technologies, Inc. [93] and used for the fabrication of solar cells. The particl es were 5 nm in diameter, just as the particles synthesized in-house. The hybrid solution using these particles was 25 mg/ml using a 1:1 by weight mixture of the commercial CdSe and P3HT in a solvent of chloroform with 2% pyridine by volume. The best cell fabricated from this solution showed the following performance characteristics: JSC = 9.44 x 10-2 mA/cm2, VOC = 0.329 V, FF = 0.247, and efficiency = 7.65 x 103 %. Dark and illuminated J-V curves for this cell are shown in Figure 3-32. These curves are inferior to those fabricated using in-house CdSe nanocrystals. The dark J-V curve shows a rectification ratio of only 0.98 at 1 V. Although the short-circu it current is lags only the bestperforming hybrid cell, the open-circ uit voltage was less than 0.4 V and led to the low efficiency. The terminating group on these nanocrystals is u nknown, as the company refused to give up this information. No attempts were made to alter the crystals; all cells we re fabricated with these particles as they were received. Hybrid cell performance summary Selected J-V curves for hybrid bulk heterojunc tion solar cells are s hown in Figure 3-33. These curves are all shown in previous figures, but are presented here on one graph for comparison. A compilation of the J-V data for hybrid cells shown in this section is shown in 150

PAGE 151

Table 3-9. The cell fabricated from a 25 mg/ml solution of equal weights P3HT and CdSe dissolved in 2% pyridine in chloroform with minimal air exposure showed the highest performance by simultaneously demonstrating the highest short-circuit cu rrent density and opencircuit voltage of all cells measured. The top-performing bi-layer and hybrid cell J-V cu rves are plotted together in Figure 3-34. The current flow through the bi-lay er cell is an order of magnitude higher than that of the hybrid cell, demonstrating the need for improved morpho logy control in the hybrid active layers. The bi-layer cell shows a low VOC compared to that of the hybrid cell. The series and shunt resistances of these cells were estimated from the J-V curves using Equation 3-6 and Equation 3-7, respectively [62]. This calculation is good for shunt resistance, but series resistance is more accura tely calculated as the applied vo ltage approaches infinity. For the bi-layer cell, J-V data was not sufficiently collected to make this calculation. From these equations, resistances for the bi-l ayer cell were calculated as Rs = 1.59 x 103 and Rsh = 2.25 x 103 For the hybrid cell, resist ances were calculated as Rs = 4.5 x 104 and Rsh = 8.33 x 104 The series resistance for th e hybrid cell calculated at the maximum measured voltage point was 1.5 x 10-2 For another bi-layer cell with a more extensive set of J-V data, the series resistance is calculated as 1.8 x 10-2 0 I sdI dV R (3-6) 0 V shdI dV R (3-7) These resistance calculations s how shunt resistances approximat ely 5 orders of magnitude higher than the series resistan ce, which should result in high-quality solar cells. Further improvements in cell design to maximize absorpti on and charge separation in the hybrid cells 151

PAGE 152

should result in further reduction of the series resistance and drastic improvements in current flow. Particle Induced Nanostructuring A key challenge in hybrid bulk heterojunction solar cells is control of the nanocrystal distribution throughout the act ive layer. The film must be well -mixed to allow efficient exciton dissociation, but also provide pe rcolation pathways to provide charge collection pathways. A new concept developed to resolve this issue is particle induced nanostructuring (PIN). The PIN concept involves depositing multiple ultra-thin layers to control nanocrystal distribution throughout the thickn ess of the film. In this study, ITO-coated glass substrates we re chemically cleaned and treated with nitrogen plasma before film deposition. Mu ltiple layers were deposited from a weakly concentrated 5 mg/ml solution of 60 wt. % CdSe and 40 wt. % P3HT in chloroform with 2% pyridine by volume. Deposition occurred at a sp in-coating speed of 5000 rpm to produce very thin films. The thicknesses of the films were measured with a profilo meter after removing a portion of the film to create a step. Film thickness measurements on samples with 1 to 6 film layers showed an interesting phenomenon. As expected, the film thickness grew with the addition of multiple layers. The film thickness increased, however, only after two deposition steps were performed, as shown in Figure 3-35. On the first deposition, a film of approximately 10 nm was deposited on the substrate. After the second, the film thickness remained constant. On the third, another film of approximately 10 nm was deposited, followed by another spin-coating resulting in no film growth. After the next deposition step, th e film grows by another 10 nm, with the 6th resulting in a minimal amount of growth. 152

PAGE 153

The dotted line drawn through the data in Figure 3-35 is the best fit line for film thickness vs. number of layers, which follows Equation 3-8. With this trend, to reach a hybrid film thickness of 100 nm, 21 layers would be required. In the fabricati on setup used for these experiments where film deposition equipment is open to atmosphere, this would result in a prohibitive amount of atmosphe ric exposure time for the P3HT. For deposition in a nitrogen or argon environment like a glove box, how ever, this process is feasible. )(#*903.4 Layers ess FilmThickn (3-8) The film surface roughness was measured with AFM, and the surface was visualized with SEM and optical microscopes. Figure 3-36 displays images of the results of these measurements at 1, 3, 5, and 7 layers. The films analyzed by SEM and AFM were different from the films analyzed with profilometry due to a difference in the sample size required for these techniques. All SEM images are shown at the same scale (at 30,000x magnification), and all AFM images show a 3 x 3 m scan area with a 100 nm scale on the z-axis. Optical microscope images of the films are s hown in Figure 3-37. These images are taken at 100x magnification and are representative of the multiple images taken of these surfaces. The rms surface roughness was measured from the AFM images for 3 x 3 m and 1 x 1 m areas on the surface. The results are shown in Figure 3-38. Additionally, lines are fit to determine the trends for roughness as more layers are deposited. The trendlines follow Equation 3-9 for the large area measurement and Equation 3-10 for the small area measurement. If these trendlines are ex trapolated to the required 21 layers for a 100 nm active layer film, the rms surface roughness is predicted to be 39.9 nm for a 5 x 5 m area and 15.0 nm for a 1 x 1 m area. From previous m easurements of surface roughness by profilometry and AFM, it was found that profilometer roughness measurements were 153

PAGE 154

approximately 3.4x higher than 5 x 5 m AFM measurements and approximately 14x higher than 1 x 1 m AFM measurements. By this extrapolation, the rms surface roughness of a 21 layer PIN film will be in the range 136 210 nm. Th e hybrid film surface profile shown in Figure 327 displayed an rms roughness of 239 nm when measured by profilometry. This comparison relies on a large amount of extrapolation, but imp lies that this technique is capable of producing films with surface roughness as low or lower than similarly thick films deposited from concentrated solutions in a single deposition step. (3-9) 65.14)(#20.1 Layers RRMS (3-10) 60.9)(#26.0 Layers RRMSCells were not fabricated using this tec hnique, as lifetime measurements on other cells indicated that the required air exposure time would cr ipple the cells before their performance could be evaluated. Once the cell fabrication process line is housed under an inert environment, this limitation no longer applies and cells consisti ng of multiple PIN layers can be compared to single-layer cells to quantitatively determ ine the applicability of this technique. 154

PAGE 155

Table 3-1. Chemical composition of ITO films Condition Atomic Percent Atomic Ratio O Sn In N O/In In/Sn No Treatment 56.75 2.65 40.6 1.397 15.32 N2 Plasmaa 55.49 2.67 40.49 1.35 1.37 15.16 e-beam b 56.67 2.38 38.76 2.19 1.462 16.28 a After N2 plasma treatment at 50 W, 1 Torr, 10 min b After e-beam treatment at 2kGy Voltage (V) -0.6-0.4-0.20.00.20.40.60.81.0 Current Density (mA/cm2) -4 -2 0 2 4 6 8 10 12 No treatment O2 Treatment Electron Beam N2 Treatment Figure 3-1. J-V curves for organic so lar cells on treated ITO substrates Table 3-2. Performance of organic solar cells on treated ITO substrates Treatment VOC (V) JSC (mA/cm2) (%) None 0.52 1.61 0.36 N2 Plasma 0.46 2.73 0.62 E-Beam 0.38 2.08 0.36 O2 Plasma 0.32 1.82 0.27 155

PAGE 156

Figure 3-2. Energy band diagram for bi-layer organic solar cells. All energy levels are listed in units of eV. Figure 3-3. Steps for bi-layer sola r cell fabrication. A) ITO-coat ed glass substrate. B) Patterned ITO anode. C) After PEDOT:PSS spin-coating. D) After P3HT spin-coating. E) After polishing. F) After C60 evaporation. G) After Al eva poration. H) Encapsulated solar cell. 156

PAGE 157

Spin Speed (rpm) 10002000300040005000 Film Thickness (A) 0 1000 2000 3000 4000 Figure 3-4. PEDOT:PSS film thickness vs. spin-coater speed. Spin Speed (rpm) 1000 2000 3000 4000 Film Thickness (A) 0 100 200 300 400 500 600 Figure 3-5. P3HT film thickness vs. spin-coating speed. 157

PAGE 158

Chlorobenzene P3HT P3HT in Chlorobenzene Figure 3-6. Upper Spectrum: FTIR spectrum for chlorobenzene from Sigma Aldrich. Lower Spectrum: P3HT film (purple) spin-coated from P3HT in chlorobenzene solution (red). 158

PAGE 159

Water PEDOT:PSS PEDOT:PSS in Wate r Figure 3-7. Upper spectrum: FT IR spectrum of water from Sigm a Aldrich. Lower spectrum: PEDOT:PSS film (gray) spin-coated from solution of PEDOT:PSS in water (blue). 159

PAGE 160

A) Voltage (V) -0.2-0.10.0 0.1 0.2 0.3 Current Density (mA/cm2) -10 0 10 20 30 8-2 Dark 8-1 Illuminated 8-2 Illuminated 8-3 Illuminated 8-4 Illuminated B) Voltage (V) -1.0 -0.5 0.0 0.5 1.0 Current Density (mA/cm2) 0.0001 0.001 0.01 0.1 1 10 100 1000 8-2 Dark 8-1 Illuminated 8-2 Illuminated 8-3 Illuminated 8-4 Illuminated Figure 3-8. Linear (A) and b ase 10 log-scale (B) J-V curves for bi-layer organic solar cells fabricated with a single 80 nm thick layer of PEDOT:PSS. 160

PAGE 161

A) Voltage (V) -0.2-0.10.0 0.1 0.2 0.3 Current Density (mA/cm2) -10 0 10 20 30 9-1 Dark 9-1 Illuminated 9-2 Illuminated 9-3 Illuminated 9-4 Illuminated B) Voltage (V) -1.0 -0.5 0.0 0.5 1.0 Current Density (mA/cm2) 0.001 0.01 0.1 1 10 100 1000 9-1 Dark 9-1 Illuminated 9-2 Illuminated 9-3 Illuminated 9-4 Illuminated Figure 3-9. Linear (a) and base 10 log-scale (b ) J-V curves for bi-layer organic solar cells fabricated with two 40 nm thick layers of PEDOT:PSS. 161

PAGE 162

Spin Speed (rpm) 10002000300040005000 Film Thickness () 0 500 1000 1500 2000 2500 3000 051116 051122 051123 051130 Figure 3-10. Calibration curves for slowa nd fast-filtered PEDOT :PSS. Slow-filtered PEDOT:PSS is shown in the data set. Fast-filtered PEDOT:PSS is shown in data sets , and A B D E C F F C E DB A Figure 3-11. AFM images of P3HT films. Images A C were spin-cast from 5 mg/ml P3HT in chlorobenzene solutions, while images D F were spin-cast from 10 mg/ml P3HT in chlorobenzene solutions. Images A and D were spin-cast at 2000 rpm, images B and E at 3000 rpm, and images C and F at 4000 rpm. For all images, the scale bar for the film height axis is 50 nm, and the scan area is 1 m x 1 m. 162

PAGE 163

Table 3-3. RMS surface roughness of P3HT films shown in Figure 3-11. Label Solution Spin Speed RMS Roughness A 5 mg/ml 2000 rpm 0.89 nm B 5 mg/ml 3000 rpm 1.27 nm C 5 mg/ml 4000 rpm 0.94 nm D 10 mg/ml 2000 rpm 3.66 nm E 10 mg/ml 3000 rpm 2.59 nm F 10 mg/ml 4000 rpm 2.54 nm Voltage (V) -0.2-0.10.0 0.1 0.2 0.3 Current density (mA/cm2) -4 -2 0 2 4 6 Set I dark Set I 1 Set I 2 Set I 3 Set I 4 Figure 3-12. J-V curves for bi-layer solar cells fabricated on untreated ITO substrates. 163

PAGE 164

A) Voltage (V) -0.2-0.10.0 0.1 0.2 0.3 Current Density (mA/cm2) -4 -2 0 2 4 6 8 10 II-1 dark II-2 dark II-3 dark B) Voltage (V) -0.4 -0.2 0.0 0.2 0.4 Current Density (mA/cm2) 0.001 0.01 0.1 1 10 100 II-1 dark II-2 dark II-3 dark Figure 3-13. J-V curves in the da rk (A and B) and under 100 mW/cm2 illumination (C and D). 164

PAGE 165

C) Voltage (V) -0.2-0.10.0 0.1 0.2 0.3 Current Density (mA/cm2) -4 -2 0 2 4 6 8 10 II-1 dark II-1 II-2 II-3 II-4 D) Voltage (V) -0.4 -0.2 0.0 0.2 0.4 Current Density (mA/cm2) 0.001 0.01 0.1 1 10 100 II-1 dark II-1 II-2 II-3 II-4 Figure 3-13 continued. J-V curves under 100 mW/cm2 illumination (C and D). 165

PAGE 166

Table 3-4. J-V data for bi-layer solar cells. Sample RR JSC (mA/cm2) VOC (V) FF (%) I-1 2.4 0.55 0.16 0.23 0.02 I-2 1.03 0.15 0.19 0.03 I-3 1.55 0.11 0.23 0.04 I-4 0.41 0.16 0.15 0.01 II-1 8.53 1.10 0.08 0.24 0.02 II-2 3.96 0.68 0.07 0.22 0.01 II-3 1.81 0.70 0.06 0.19 0.01 II-4 0.88 0.07 0.21 0.01 Table 3-5. Solvents considered for hybr id bulk heterojunction film deposition. Solvent B.T. (C) Polarity P3HT Solubility Acetone 56.2 5.1 No 2-butanol 79.6 4.0 No DMF 153 6.4 No Methanol 64.6 5.1 No 2-propanol 82.4 3.9 No MEK 80.0 4.7 Poor Pyridine 115.3 5.3 Poor Benzene 80.1 2.7 Yes Chlorobenzene 131.7 2.7 Yes Chloroform 61.2 4.1 Yes o-dichlorobenzene 180 2.7 Yes THF 66.0 4.0 Yes Toluene 110.6 2.4 Yes TCE 87.2 1.0 Yes o-xylene 144.4 2.5 Yes 166

PAGE 167

A) Solvent Boiling Temperature oC 406080100120140160180200220 RMS Roughness (nm) 0 100 200 300 400 THF Chlorobenzene o-dichlorobenzene Chloroform Toluene o-xylene TCE Benzene B) Solvent Polarity 0.51.01.52.02.53.03.54.04.5 RMS Roughness (nm) 0 100 200 300 400 THF Chlorobenzene o-dichlorobenzene Chloroform Toluene o-xylene TCE Benzene Figure 3-14. Surface roughness measurements by profilometry for hybrid films deposited from various solvents. 167

PAGE 168

A) Solvent Boiling Temperature (oC) 406080100120140160180200 RMS Roughness (nm) 0 10 20 30 40 50 60 THF Chlorobenzene o-dichlorobenzene Chloroform Toluene o-xylene B) Solvent Polarity 0.51.01.52.02.53.03.54.04.5 RMS Roughness (nm) 0 10 20 30 40 50 60 THF Chlorobenzene o-dichlorobenzene Chloroform Toluene o-xylene Figure 3-15. RMS surface roughness for 5 x 5 m surface area samples measured with AFM. 168

PAGE 169

A) Solvent Boiling Temperature (oC) 406080100120140160180200 RMS Roughness (nm) 0 2 4 6 8 10 12 14 16 18 20 THF Chlorobenzene o-dichlorobenzene Chloroform Toluene o-xylene B) Solvent Polarity 0.51.01.52.02.53.03.54.04.5 RMS Roughness (nm) 0 2 4 6 8 10 12 14 16 18 20 THF Chlorobenzene o-dichlorobenzene Chloroform Toluene o-xylene Figure 3-16. RMS surface roughness for 1 x 1 m surface area samples measured with AFM. 169

PAGE 170

Table 3-6. Mean rms surface roughness in nm fo r hybrid films deposited from selected solvents Solvent Profilometer AFM (5 x 5 m) AFM (1 x 1 m) THF 136 36.8 15.0 Chlorobenzene 64.9 25.0 8.42 o-DCB 81.2 21.1 9.58 Chloroform 71.8 16.2 3.22 Toluene 66.2 27.4 2.93 Xylene 198 16.8 7.76 TCE 72.0 Not measured Not measured Benzene 73.9 Not measured Not measured THF Chlorobenzene 10 x 100 x 100 x 10 x Figure 3-17. Optical microscope images of selec ted films. The first column displays images taken at 10x magnification, while the sec ond column displays images taken at 100x magnification. 170

PAGE 171

o-dichlorobenzene Chloroform Toluene 10 x 100 x 100 x 10 x 100 x 10 x Figure 3-17 continued. Optical microscope imag es of selected films. 171

PAGE 172

o-xylene Benzene TCE 100 x 10 x 100 x 10 x 100 x Figure 3-17 continued. Optical microscope imag es of selected films. 10 x 172

PAGE 173

THF Chlorobenzene o-dichlorobenzene Figure 3-18. SEM images of selected hybrid films. The first column image was taken at 5,000x magnification and the second column imag e was taken at 15,000 x magnification. 173

PAGE 174

Chloroform TCE Figure 3-18 continued. SEM imag es of selected hybrid films. 174

PAGE 175

A) B) Figure 3-19. Film surface roughn ess vs. pyridine concentration in the chloroform solvent for TOPO-coated CdSe nanocrystals (A) and pyr idine-coated CdSe nanocrystals (B). 175

PAGE 176

Table 3-7. Hybrid solutions of P3HT and TOPO-coated CdSe nanocry stals in a mixed solvent of chloroform and pyridine. Solution P3HT Wt. CdSe Wt. Chloroform Vol. Py ridine Vol. Solution Concentration 1 26.8 mg 35.5 mg 1.5 ml 1.5 ml 20.4 mg/ml 2 24.8 mg 23.2 mg 1.5 ml 1.5 ml 19 mg/ml 3 15.0 mg 25.0 mg 1.5 ml 1.5 ml 5 mg/ml A) B) C) Figure 3-20. Optical microscope images of hybrid films deposited from 20.4 mg/ml (A), 19 mg/ml (B) and 5 mg/ml (C) solutions. Images were taken at 100x magnification and the films were deposited at 3000 rpm and dried under vacuum at 120 C. 176

PAGE 177

Figure 3-21. Optical microscope image of 19 mg/ml hybrid film deposited at 3000 rpm and subjected to a pure solvent spin-coating step at 8000 rpm. The image was taken at 100x magnification. A) B) C) Figure 3-22. Optical microscope images of P3HT films deposited from 5 mg/ml solutions in chloroform (A), 1:1 chloroform:pyridine (B ), and pyridine (C). All images are taken at 100x magnification. 177

PAGE 178

2,000 x 10,000 x Chloroform Solvent Mixed Solvent Figure 3-23 continued. SEM images of P3HT films deposited from pure chloroform and an equal mixture of chloroform and pyridine solv ents. Images are shown at 2,000x and 10,000x for each film. 178

PAGE 179

A) Lateral Distance (mm) 0.0 0.5 1.0 1.5 2.0 2.5 Film Thickness (nm) 0 20 40 60 80 100 ITO P3HT Film Figure 3-24. Surface profiles of P3HT films deposited from chloroform (A), 1:1 chloroform:pyridine (B), and pyridine (C) solvents. In the graphs, the bare substrate appears on the left hand side with the film on the right hand side beyond the wide peak. The solid teal line represents the me dian film thickness, the red dash-dot line represents the mean film thic kness, and the blue dashed li nes represent one standard deviation from the mean. 179

PAGE 180

B) Lateral Distance mm) 0.0 0.5 1.0 1.5 2.0 2.5 Film Thickness (nm) 0 20 40 60 80 100 C) Lateral Distance (mm) 0.0 0.5 1.0 1.5 2.0 2.5 Film Thickness (nm) 0 20 40 60 80 100 ITO P3HT Film Figure 3-24 continued. Surface profiles of P3HT films deposited from chloroform (A), 1:1 chloroform:pyridine (B), and pyridine (C) solvents. 180

PAGE 181

Table 3-8. Film properties for P3HT films deposited from various solvents. Solvent Mean Thickness (nm) Median Thickn ess (nm) Standard Deviation (nm) Chloroform 21.2 8.18 23.1 7.95 9.58 5.41 Mixed 56.2 34.3 16.2 4.42 227 Pyridine 78.5 85.4 10.9 8.61 338 A) Voltage (V) -1.5-1.0-0.50.00.51.01.5 Current Density (mA/cm2) -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 Dark Illuminated -0.2-0.10.00.10.20.30.40.5 -0.04 -0.02 0.00 0.02 0.04 Figure 3-25. Dark and illuminated J-V curves for cells generated from 5 mg/ml composite solutions in (A) chloroform mixed with 2% pyridine and (B) pure chloroform. Note the scale difference of the graphs. 181

PAGE 182

B) Voltage (V) -1.5-1.0-0.50.00.51.01.5 Current Density (mA/cm2) -1.2e-3 -1.0e-3 -8.0e-4 -6.0e-4 -4.0e-4 -2.0e-4 0.0 2.0e-4 4.0e-4 Dark Illuminated -0.2-0.10.00.10.20.30.40.5 -1e-4 -5e-5 0 5e-5 1e-4 Figure 3-25 continued. Dark and illuminated J-V curves for cells generated from 5 mg/ml composite solution in pure chloroform. Voltage (V) -1.5-1.0-0.50.00.51.01.5 Current Density (mA/cm2) -0.4 -0.2 0.0 0.2 0.4 Figure 3-26. Dark and illuminated J-V curves for 10 mg/ml hybrid solutio n deposited with lowspeed spin-coating. 182

PAGE 183

A) Lateral Distance (mm) 0.0 0.5 1.0 1.5 2.0 2.5 Film Thickness (nm) -100 0 100 200 300 400 500 P3HT Film ITO P3HT FilmB) Lateral Distance (mm) 0.0 0.5 1.0 1.5 2.0 2.5 Film Thickness (nm) -100 0 100 200 300 400 500 Hybrid Film ITO Hybrid FilmFigure 3-27. Surface profiles of f ilm deposited from (A) 12 mg/ml P3HT and (B) 25 mg/ml hybrid solutions in 2% pyridine in chloroform. 183

PAGE 184

Voltage (V) -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 Current Density (mA/cm2) -1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 Hybrid Dark Hybrid Illuminated P3HT Dark P3HT Illuminated -0.8 -0.6 -0.4 -0.2 0.0 0.2 -0.10 -0.05 0.00 0.05 0.10 0.15 0.20 Figure 3-28. Dark and illuminated J-V curves for hybrid bulk heterojunction solar cell deposited from 25 mg/ml solution and P3HT polymer cell deposited from 12 mg/ml solution. The insert shows a zoom-in on the active area of the cells. 184

PAGE 185

Exposure Time (min) 0 50100150200250300 Short Circuit Current Density (mA/cm2) 0.00 0.01 0.02 0.03 0.04 0.05 0.06 Hybrid Cell P3HT Cell Decay Hybrid Decay P3HT Hybrid Cell J = 2.894E-3 + 1.525 exp(-9.369E-2 t) P3HT Cell J = 5.804E-3 + 7.506E-2 exp(-4.820E-2 t) Figure 3-29. Short-circuit current de cay for hybrid (gold circles) and P3HT (green squares). 185

PAGE 186

Voltage (V) -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 Current Density (mA/cm2) -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 Hybrid Dark Hybrid Illuminated P3HT Dark P3HT Illuminated -0.8 -0.6 -0.4 -0.2 0.0 0.2 -0.10 -0.05 0.00 0.05 0.10 0.15 0.20 0.25 Figure 3-30. Dark and illuminated J-V cu rves for hybrid bulk heterojunction and pure P3HT solar cells with limited ai r exposure during processing. 186

PAGE 187

Voltage (V) -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 Current Density (mA/cm2) -30 -20 -10 0 Dark Illuminated -1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 -0.4 -0.2 0.0 0.2 0.4 Figure 3-31. Dark and illuminated J-V cu rves for hybrid solar cell fabricated from chlorobenzene with 2% pyridine solution. Red curves represent illuminated J-V and black curves represent dark J-V. 187

PAGE 188

Voltage (V) -1.5-1.0-0.50.00.51.01.5 Current Density (mA/cm2) -1.0 -0.5 0.0 0.5 1.0 1.5 -0.20.00.20.40.60.81.0 -0.4 -0.2 0.0 0.2 0.4 Figure 3-32. Dark and illuminated J-V curves fo r a hybrid solar cell fabr icated with commercial CdSe nanopowder. Green curves represen t illuminated J-V and black curves represent dark J-V. 188

PAGE 189

189 Voltage (V) -1.0 -0.5 0.0 0.5 1.0 Current Density (mA/cm2) -0.4 -0.2 0.0 0.2 0.4 25 mg/ml in Chloroform-Pyridine 25 mg/ml in Chloroform-Pyridine short exposure time 25 mg/ml in Chloroform-Pyridine commercial CdSe 30 mg/ml in Chlorobenzene-Pyridine Figure 3-33. Illuminated J-V curves for hybrid bulk heterojunction sola r cells with various fabrication conditions. Voltage (V) -0.20.00.20.40.60.81.0 Current Density (mA/cm2) -3 -2 -1 0 1 2 Bi-layer Cell Hybrid Cell Figure 3-34. J-V curves for the best bi-layer and hybrid cells shown in this dissertation.

PAGE 190

Table 3-9. Hybrid solar cell fabricat ion information and performance data. Figure Solution Concentration CdSe wt. % in P3HT % Pyridine in Chloroform JSC (1) VOC (V) FF (2) Notes 3-25 5 mg/ml 60% 2% 8.99 0.338 0.230 0.7 3-25 5 mg/ml 60% 0% 0.026 0.194 0.252 0.001 Pure Chloroform Solvent 3-26 10 mg/ml 60% 2% 33.9 0.256 0.275 2.4 3-28 25 mg/ml 50% 2% 49.9 0.701 0.232 17.5 3-28 12 mg/ml 0% 2% 20.3 0.523 0.310 12.5 Pure P3HT film 3-30 25 mg/ml 50% 2% 205 0.705 0.288 41.6 Minimized air exposure 3-30 12 mg/ml 0% 2% 52.6 0.237 0.242 3.0 Pure P3HT film Minimized air exposure 3-31 30 mg/ml 50% 2% ( 3 ) 138 0.391 0.292 15.8 Pyridine in Chlorobenzene Solvent 3-32 25 mg/ml 50% 2% 94.4 0.329 0.247 7.65 Commercial CdSe (1) JSC displayed in units of A/cm2 (mA/cm2 x 10-3) (2) Efficiency displayed in units of % x 10-3 (3) This solvent is 2% pyridine in chlorobenzene 190

PAGE 191

Number of Layers 0123456 7 Film Thickness (nm) 0 10 20 30 40 Figure 3-35. Film thickness fo r multi-layer hybrid films. # SEM Image AFM Image 1 Figure 3-36. SEM and AFM surface im ages of multi-layer hybrid films. 191

PAGE 192

# SEM Image AFM Image 3 5 7 Figure 3-36 continued. SEM and AFM surface images of mu lti-layer hybrid films. 192

PAGE 193

a) b) c) d) e) f) Figure 3-37. Optical microscope images for multilayer hybrid films. The number of film layers were a) 1, b) 2, c) 3, d) 4, e) 5, and f) 6. 193

PAGE 194

194 Number of Layers 01234567 RMS Surface Roughness (nm) 5 10 15 20 25 30 35 Figure 3-38. RMS surface roughness for multi-la yer hybrid films. Red squares represent measurements over a 3 x 3 m area and blue circles represent measurements on a 1 x 1 m area.

PAGE 195

CHAPTER 4 CONCLUSIONS AND FUTURE WORK Conclusions This dissertation presented the results of exploratory research on the processing and performance of hybrid PV. It provided encour agement for a more complete study of hybrid photovoltaic devices. With collaboration and assi stance from several teams and individuals, the groundwork was laid for future st udies to continue this project and achieve high-performance hybrid photovoltaic devices. The innovations of this study include the firs t simulations of hybrid photovoltaics using existing semiconductor modeling software, devel opment of anode surface treatment processes, solvent selection for hybrid films, and hybr id bulk heterojuncti on photovoltaic process development, including an interesting multiple spjn coating process sequence to better disperse the inorganic phase. Hybrid Photovoltaic Simulation Simulations of an ordered hete rojunction photovoltaic cell provided interes ting results. It was found that reported values of certain parameters repo rted in the literatur e from organic fieldeffect transistor fabrication were poor estim ates for organic photovol taic simulation. The hole mobility values given in the liter ature [71] proved to be higher than the simulations estimated. This value of 0.01 cm2/V-s for the hole mobility produced J-V curves with fill factors around 0.85, which is considerab ly larger than published values which typically range from 0.4 to 0.6 [27, 33-34, 44, 75]. By reducing the mobility to 1 x 10-4 cm2/V-s the fill factor was reduced to 0.78, which is closer to the published range. The short-circuit current density of the real cell could not be matched by the simulations. All attempts using the two-step model gave values lower than the observed ones, including 195

PAGE 196

simulations that allowed the full P3HT region to generate photocur rent and a 50% increase in the absorption coefficient over a 40 nm exciton diffu sion length. The highest short-circuit current density achieved with this simu lation technique was 1.47 mA/cm2 from a cell featuring a 40 nm LD region, but this value is approximately 0.8 mA/cm2 lower than the real ce ll. It is believed that the two-step simulation process creates a strong attractive for ce between charges when very high levels of charges are generated along a na rrow line, resulting in increased annihilation between the free carriers. Anode Surface Treatment Treatment of the ITO substrate with a N2 plasma was optimized to grant a smooth, stable surface, which served as the basis of cell fabrication. This work was done in close collaboration with Jiyoun Seol at Yeungnam Univ ersity. The chemically cleaned substrate was subjected to nitrogen plasma at 50 W and 200 mTorr for 30 min. This process resulted in a smoother, more hydrophilic surface which aided deposition of furthe r layers, and an increase in nitrogen content which lowered the film work function and eased the pathway for hole current. This treatment procedure was used for all hybrid film and hybrid photovoltaic studies that followed to ensure a stable surface and allow direct comparison of the results. Solvent Selection The solvents TCE, THF, benzene, toluene, o-xylene, chlorobenzene, o-dichlorobenzene, and chloroform were compared for their app licability to hybrid film deposition. Although no correlations were found between film propert ies and solvent polarity or solvent boiling temperature, certain solvents were identified as strong candidates for these films. These included chloroform, and chlorobenzene, which were eac h used in further studies. These solvents produced films with low surface roughness at all measurement scales, from microns to 196

PAGE 197

millimeters. These solvents provided simultaneous dissolution both the organic and inorganic phases which limited early precipitati on of the solutes, resulting in a more uniform film surface. Hybrid Bulk Heterojunction Photovoltaic Development Hybrid photovoltaic cells were fa bricated with varying film pr operties. It was found that the best cells featured thick absorber films spin-coated from solutions of at least 25 mg/ml with a 50:50 weight ratio between P3HT and the nano-CdSe. Cells consisting of pure P3HT sandwiched between the PEDOT:PSS and Al layers showed maximum performance approximately an order of magnitude lower than the best hybrid cells. Cells fabricated from commercial CdSe na nopowder performed at less than 20% of the best cells using in-house CdSe nanopowder coated with pyridine, althoug h a direct comparison between the two is difficult because the surf ace passivation material remains unknown for the commercial nanopowder. Hybrid cells fabricated with chlorobenzene solution rath er than chloroform saw a 60% reduction is efficiency despite an increase in solution concentration from 25 to 30 mg/ml. Despite JSC and fill factor values that were nearly th e same as that of the chloroform cell, the chlorobenzene cell showed a VOC of 0.3 V lower than the chloroform cell. The best cell was fabricated from a 25 mg/ml solution of 50% CdSe in P3HT dissolved in 2% pyridine in chloroform. The process was ta ilored to minimize the amount of time the cell was exposed to atmosphere during fabrication and between fabrication and test ing. It showed a maximum efficiency of 4.16 x 10-2 %, JSC of 0.205 mA/cm2, VOC of 0.705 V, and a fill factor of 0.288. 197

PAGE 198

Future Work Although the field of organic photovoltaics is rapidly grow ing and advancing, several research directions are suggested by this work for its continua tion. This section describes some potential research directions stemming fr om the work presented in this dissertation. Organic Photovoltaic Simulations The models presented in this dissertation u sed the powerful modeling software package Medici. Although the software h as powerful optical and electric al simulation abilities, this research seems to have pushed its limits by dema nding nano-scale material specifications and low levels of free carriers, carrier mobilities, and current flow in the devices. Simulation attempts were frequently cut short due to conve rgence issues in the soft ware under the specified conditions. Additionally, a key component of th e modeling work focused on the correct way to simulate the effect of excitons in Medici which are not explicit in the package. In light of these difficulties, a new software package designed to simulate organic electronic materials would be a great boost to this work. A product such as Fluxim [94] would be able to more accurately simulate the effect of excitons in these hybrid devices. Another direction that should be pursued is the variation of the geometry and materials of the simulated cells. While ZnO is a well-researc hed material for the growth of aligned and ordered nanowires, these stru ctures are now being grown for other materials with stronger absorption spectra such as CdSe and InP [66, 95]. These materials could see more use in nanowire hybrid cells in the fu ture, driving the need for eff ective simulations to study device properties. The most common organic cell design is cu rrently the bulk heterojunction cell using semiconductor nanoparticles or soluble C60 derivatives. This design presents a challenge for simulations because little work has been done to characterize the pa rticle distribut ions in these 198

PAGE 199

films. Because the physical dime nsions of the model are vital for accurate simulation, this provides both a limitation and an opportunity. On the one hand, this lack of information limits the results that can be generated through simulati ons. On the other hand, if an accurate, robust model can be developed for a hybrid material system, it could be used to back-calculate unknown physical dimensions of the system. Development of Hybrid Photovoltaic Cells A central theme of this dissertation is the morphology control of hybrid films for photovoltaic applications. This study only focused on nearly sphe rical nanocrystals, but future studies should expand the field to include nanoparticles with dimensionality. Shaped nanoparticles, nanowires, tetrapods and more exotic branched st ructures are bei ng grown using techniques similar to the one used for spheri cal nanocrystals in th is study [35-37]. These dimensional crystals offer the promise of direct ed charge transport without the multiple electron jumping processes required for sm all spherical nanocrystals. Alternative semiconductor nanoparticles should be considered as well. Although CdSe is one of the easiest particles to be synthesized, other compound semiconductors such as CdS, CdTe, PbSe, and CIS can be grown on the nano scal e. Some semiconductors such as PbSe have demonstrated multi-carrier generation on s hort timescales that offer the possibility of constructing photovoltaic devices with quantum e fficiencies greater than 1 if these charges can be harvested [96-97]. Regioregular P3HT, as used in this study, is curren tly the most promising candidate for polymer in electronic devices such as organic thin-film transist ors and solar cells [33-34]. However, this polymer shows some limitations fo r solar applications. The absorption spectrum shows a cut-off above 650 nm, limiting absorption of near-IR photons which are plentiful in the solar spectrum. Although the hole mobility of P3HT is high compared to many conductive 199

PAGE 200

200 polymers, it is several orders of magnitude lower than most inorganic semiconductors and limits charge collection in devices. Additiona lly, it degrades quickly under water and oxygen atmospheres, particularly in the presence of ra diation. As new polymer s are developed with a broader absorption spectrum, improved carrier m obility, improved environmental resistance, and the ability to deposit high-quality films with strong adhesion and low surface roughness they will quickly find relevance in or ganic photovoltaics research. Regardless of the targeted fu ture research direction, some processing equipment changes should be considered. The recent installation of a photolithographic patterning system will allow for the fabrication of multiple cells on single substrates, and this will greatly improve the speed of research and also the quality of the devices fabricated. As long as P3HT is used for the active layer film, all processing equipment should be set up under an inert atmosphere of argon or nitrogen in a glove box. Because of the sensitiv ity of this polymer to water and oxygen, highquality device fabrication at this scale requires that it be protec ted from exposure at all phases of the process. This is not true fo r the HTL layer of PEDOT:PSS, which is in fact deposited from a water solution and shows no negative effects of short-term atmospheric exposure. All processing steps beyond the deposition of the HTL should be contained in this glove box, including mixing for the hybrid solution, depos ition and drying of the hybrid films, possible inclusion of exciton blocking layers back electrode deposition, and cell characterization. This is a difficult challenge due to the size of some pro cessing equipment, but this is the design used by groups fabricating world-record organic photovo ltaic cells. The processing equipment line currently in use already involv es an evaporation chamber opening directly into the glove box, which is the largest piece of equipment used in the process. In order to improve these cells to the highest performance level, this environmental protect ion is a step that absolutely must be taken.

PAGE 201

LIST OF REFERENCES 1. International Energy Agency, Trends in Photovoltaic Applicat ions, Report IEA-PVPS T1-16:2007, August 2007. 2. U.S. Department of Energy, Energy Efficiency and Renewable Energy, http://www1.eere.energy.gov/solar/deployment.html (2006) July 2008. 3. Solarbuzz, LLC., www.solarbuzz.com/moduleprices.htm (2008) July 2008. 4. C.W. Tang, Appl. Phys. Lett. 48 (1986) 183. 5. Md.K.H. Bhuiyan, T. Mieno, Th in Solid Films 441 (2003) 187. 6. N.S. Sariciftci, L. Smilowitz, A.J. Heeg er, F. Wudl, Synth. Met. 59 (1993) 333. 7. B. ORegan, M. Gratze l, Nature 335 (1991) 737. 8. M.K. Nazeeruddin, P. Pechy, T. Renouard, S. M. Zakeeruddin, R. Humphry-Baker, P. Comte, P. Liska, L. Cevey, E. Costa, V. Shklover, L. Spiccia, G.B. Deacon, C.A. Bignozzi, M. Grtzel, J. Am Chem. Soc. 123 (2001) 1613. 9. K. Hara, M. Kurashigo, Y. Dan-oh, C. Kasasa, Y. Ohga, A. Shinpo, S. Suga, K. Sayama, H. Arakawa, New J. Chem. 27 (2003) 783. 10. Q.B. Meng, K. Takahashi, X.T. Zhang, I. Su tanto, T.N. Rao, O. Sato, A. Fujishima, H. Watanabe, T. Nakamori, Uragami Langmuir 19 (2003) 3572. 11. L.J. Rothberg, M. Yan, F. Papadimitrakopoulos M.E. Galvin, E.W. Kwock, T.M. Miller, Synth. Met. 80 (1996) 41. 12. J.J.M. Halls, K. Pichler, R.H. Friend, S.C. Moratti, A.B. Holmes, Appl. Phys. Lett. 68 (1996) 3120. 13. L.A. Pettersson, L.S. Roman, O. I nganas, J. Appl. Phys. 86 (1999) 487. 14. J.J.M. Halls, C.A. Walsh, N.C. Greenham, E.A. Marseglia, R.H. Friend, S.C. Moratti, A.B. Holmes, Nature 376 (1995) 498. 15. T.J. Savenije, J.M. Warman, A. Goo ssens, Chem. Phys. Lett. 287 (1998) 148. 16. M. Theander, A. Yartsev, D. Zigmantas, V. Sundstrom, W. Mammo, M.R. Andersson, O. Inganas, Phys. Rev., B 61 (2000) 12957. 17. H.L. Wong, L.S.M. Lam, K. W. Cheng, K.Y.K. Man, W. K.Chan, C.Y. Kwong, A.B. Djurisic, App. Phys. Lett. 84 (2004) 2557. 18. G.D. Sharma, R. Kumar, S.K. Sharma, M.S. Roy, S. En. Mater. & Solar Cells 90 (2006) 933. 201

PAGE 202

19. P. Peumans, S. Uchida, S.R. Forrest, Nature 425 (2003) 158. 20. R.P. Raffaelle, B.J. Landi, J.D. Harris, S.G. Bailey, A.F. Hepp, Mat. Sci. & Eng., B 116 (2005) 233. 21. A.D. Pasquier, H.E. Unalan, A. Kanwal, S. Miller, M. Chhowalla, App. Phys. Lett. 87 (2005) 203511. 22. N.S. Sariciftci, L. Smilowitz, A.J. Heeger, F. Wudl, Science 258 (1992) 1474. 23. N.S. Sariciftci, L. Smilowitz, A.J. Heeg er, F. Wudl, Synth. Met. 59 (1993) 333. 24. G. Yu, K. Pakbaz, A.J. Heeger, Appl. Phys. Lett. 64 (1994) 3422. 25. J.C. Hummelen, B.W. Knight, F. LePeq, F. Wudl, J. Yao, C.L. Wilkins, J. Org. Chem. 60 (1995) 532. 26. G. Yu, J. Gao, J.C. Hummelen, F. W udl, A.J. Heeger, Science 270 (1995) 1789. 27. S.E. Shaheen, C.J. Brabec, N.S. Sariciftci, F. Padinger, T. Fromherz, J.C. Hummelen, Appl. Phys. Lett. 78 (2001) 841. 28. N. Camaioni, G. Ridolfi, G. Casalbore-Miceli, G. Possamai, L. Garlaschelli, M. Maggini, Solar Energy Materials & Solar Cells 76 (2003) 107. 29. M. Al-Ibrahim, H.K. Roth, M. Schroedner, A. Konkin, U. Zhokhavets, G. Gobsch, P. Scharff, S. Sensfuss, Orga nic Electronics 6 (2005) 65. 30. W. Geens, T. Aernouts, J. Poortmans, G. Hadziioannou, Thin Solid Films, 403 (2002) 438. 31. D. Gebeyehu, F. Padinger, C.J. Brabec, T. Fr omherz, J.C. Hummelen, N.S. Sariciftci, Int. J. of Photoenergy 1 (1999) 1. 32. A.J. Mozer, N.S. Sariciftci, C.R. Chimie 9 (2006) 568. 33. F. Padinger, R.S. Rittberger, N.S. Sari ciftci, Adv. Funct. Mat. 13 (2003) 85. 34. G. Li, V. Shrotriya, J. Huang, Y. Yao, T. Moriarty, K. Emery, Y. Yang, Nat. Mat. 4 (2005) 864. 35. M. Chen, Yi Xie, J. Lu, Y. Xiong, S. Zhang, Y. Qian, X. Liu, J. Mat. Chem. 12 (2002) 748. 36. L. Manna, D.J. Milliron, A. Me isel, E.C. Scher, A.P. Alivis atos, Nat. Mat. 2 (2003) 382. 37. K.A. Dick, K. Deppert, M.W. Larsson, T. Ma rtensson, W. Seifert, L.R. Wallenberg, L. Samuelson, Nat. Mat. 3 (2004) 380. 202

PAGE 203

38. W.J.E. Beek, L.H. Slooff, M.M. Wienk, J.M. Kroon, R.A.J. Janssen, Adv. Funct. Mater. 15 (2005) 1703. 39. P.A.C. Quist, L.H. Slooff, H. Donker, J.M. Kroon, T.J. Savenije, L.D.A. Siebbeles, Superlattices and Microstr uctures 38 (2005) 308. 40. A. Watt, H. Rubinsztei n-Dunlop, P. Meredith, Mat. Lett. 59 (2005) 3033. 41. X. Peng, J. Wickham, A.P. Alivisat os, J. Am. Chem. Soc. 120 (1998) 5343. 42. Z.A. Peng, X. Peng, J. Am Chem. Soc. 123 (2001) 183. 43. M. Kuno, J.K. Lee, B.O. Dabbousi, F.V. Mikulec, M.G. Bawendi, J. Chem. Phys. 106 (1997) 9869. 44. W.U. Huynh, J.J. Dittmer, W.C. Libby, G.L. Whiting, A.P. Alivisatos, Adv. Funct. Mater. 13 (2003) 73. 45. H.J. Choi, J.K. Yang, H.H. Park, Thin Solid Films 494 (2006) 207. 46. K.M. Coakley, M.D. McGehee, App. Phys. Lett. 83 (2003) 3380. 47. K.M. Coakley, Y. Lui, M.D. McGeehee, K.M. Frindell, G.D. Stucky, Adv. Funct. Mater. 13 (2003) 301. 48. J.J. Wu, S.C. Liu, Adv. Mater., Vol. 14 (2002) 215. 49. W. Lee, M.C. Jeong, J.M. Myoung, Acta Mater. 52 (2004) 3949. 50. C. Geng, Y. Jiang, Y. Yao, X. Meng, J.A. Za pien, C.S. Lee, Y. Lifshitz, S.T. Lee, Adv. Funct. Mater. 14 (2004) 589. 51. S.C. Lyu, Y. Zhang, C.J. Lee, H. Ruh, H.J. Lee, Chem Mater. 15 (2003) 3294. 52. L. Vayssieres, K. Keis, A. Hagfeldt, S. Lindquist, Chem. Mater. 13 (2001) 4395. 53. Y. Tak, K. Yong, C. Park, J. Crystal Growth 285 (2005) 549. 54. M. Wang, E.J. Kim, S.H. Hahn, C. Park, Physica Status Solidi (a) 203 (2006) 2418. 55. R. Corkish, D.S.P. Chan, M.A. Green, J. Appl. Phys. 79 (1996) 195. 56. Y. Zhang, A. Mascarenhas, S. Deb, J. Appl. Phys. 84 (1998) 3966. 57. M. Burgelman, B. Minnaert, Thin Solid Films 511 (2006) 214. 58. B. Kannan, K. Castelino, A. Ma jumdar, Nano Lett. 3 (2003) 1729. 203

PAGE 204

59. W. Geens, T. Martens, J. Poor tmans, T. Aernouts, J. Manca, L. Lutsen, P. Heremans, S. Borghs, R. Mertens, D. Vanderzan de, Thin Solid Films 451 (2004) 498. 60. B. Mazhari, S. En. Mater. & Solar Cells 90 (2006) 1021. 61. A. Takshi, A. Dimopoulos, J.D. Madden, Solid-State Electronics 52 (2008) 107. 62. W.U. Huynh, J.J. Dittmer, N. Teclemariam, D.J. Milliron, A.P. Alvisatos, K.W.J. Barnham, Phys. Rev., B 67 (2003) 115326. 63. R.D. Bettignies, J. Leroy, M. Firon, C. Se ntein, Syn. Met. 156 (2006) 510. 64. V. Lemaur, M. Steel, D. Belj onne, J.L. Bredas, J. Cornil, J. Am. Chem. Soc. 127 (2005) 6077. 65. Medici Users Manual, Avant Corporati on, Beaverton, OR, Release 2001.4, (2000). 66. W.K. Kim, Ph.D. Dissertation, University of Florida, Gainesville, FL, (2006). 67. E.J. Meijer, C. Detcheverry, P.J. Baesjou, E.V. Veenendaal, D.M. de Leeuw, T.M. Klapwijk, J. Appl. Phys. 93 (2003) 4831. 68. H. Sirringhaus, N. Tessler, R.H. Friend, Science 280 (1998) 1741. 69. C. Tanase, E.J. Meijer, P.W.M. Blom, D. M. de Leeuw, Org. Elec. 4 (2003) 33. 70. T.A. Chen, X. Wu, R.D. Reike, J. Am. Chem. Soc. 117 (1995) 233. 71. A.J. Casico, J.E. Lyon, M.M. Beerbom, R. Sc hlaf, Y. Zhu, S.A. Jenekhe, App. Phys. Lett. 88 (2006) 62103. 72. A.J.C. Wen, K.L. Chen, M.H. Yang, W.T. Hsiao, L.G. Chao, M.S. Leu, Surf. Coat. Tech. 198 (2005) 362. 73. Melles Griot Inc., http://mellesgriot.com/p roducts/optics/oc_5_1.htm (2002) July 2008. 74. Kim, Y., S.A. Choulis, J. Nelson, D.D.C. Bradle y, S. Cook, and J.R. Durrant, J. Mat. Sci. 40 (2005) 1371. 75. D.C. Olson, J. Piris, R.T. Collins, S.E. Shaheen, D.S. Ginley, Thin Solid Films 496 (2006) 26. 76. GetData Graph Digitizer, http://getdata-graph-digitizer.com (2008) July 2008. 77. K.R. Amundson, B.J. Sapjeta, A.J. Lovinger, Z. Bao, Thin Solid Films 414 (2002) 143. 78. S. Li, Semiconductor Physical Electronics, 2nd Ed., Springer, 2006. 204

PAGE 205

205 79. C.J. Brabec, A. Cravino, D. Meissner, N.S. Sariciftci, T. Fromherz, M.T. Rispens, L Sanchez, J.C. Hummelen, Adv. Funct. Mater. 11 (2001) 374. 80. J. Seol, M.L. Monroe, T.J. Anderson, M.A. Hasnain, C.H. Park, Conference Record of the 2006 IEEE 4th World Conference on Photovoltaic Energy Conversion 1 (2006) 236. 81. C.W. Tang, S.A. VanSlyke, C.H. Chem, J. Appl. Phys. 65 (1989) 3610. 82. Kobayashi, H. Hori, and Y. Tanaka, in Handbook of Liquid Crystal Research, edited by P.J. Collings and J.S. Patel (Oxford Univ ersity Press, New York, 1997), Chap. 10, 415. 83. H.S. Kim, J.H. Lee, C.H. Park, J. Kor. Phys. Soc. 41 (2002) 395. 84. J.S. Kim, R.H. Friend, F. Cacial li, J. Appl. Phys. 86 (1999) 2774. 85. K. Sugiyama, H. Ishii, Y. Ouchi, J. Appl. Phys. 87 (2000) 295. 86. W.W. Flack, D.S. Soong, A.T. Bell, D. W. Hess, J. Appl. Phys. 56 (1984) 1199. 87. U. Mitschke, P. Bauerle, J. Mat. Chem. 10 (2000) 1471. 88. P. Peumans, A. Yakimov, S.R. Fo rrest, J. App. Phys. 93 (2003) 3693. 89. E. Kymakis, E. Kourdoumas, I. Franghiadak is, S. En. Mater. & Solar Cells 90 (2006) 1705. 90. Phenomenex, Solvent Miscibility Table, http://designerdrugs.com/pte/12.162.180.114/dcd/ pdf/solvent.miscibility.pdf July 2008. 91. Z.A. Peng, X. Peng, J. Am Chem. Soc. 123 (2001) 183. 92. Meliorum Technologies, Inc., www.meliorum.com (2008) July 2008. 93. Fluxim, www.fluxim.com (2008) July 2008. 94. H. Burkhard, H.W. Dinges, E. K uphal, J. App. Phys. 53 (1982) 655. 95. R.J. Ellingson, M.C. Beard, J.C. Johnson, P. Yu, O.I. Micic, A.J. Nozik, A. Shabaev, A.L. Enfros, Nano Lett. 5 (2005) 865. 96. R.D. Schaller, V.I. Klimov, Phys. Rev. Lett. 92 (2004) 186601.

PAGE 206

BIOGRAPHICAL SKETCH Matthew Lowell Monroe was born in 1978, in Ma rietta, Georgia, to Ronald L. and Debbie B. Monroe. He earned a Bachelor of Science degree from the Ch emical Engineering Department at the Georgia Institute of Tec hnology in Atlanta in 2002. He jo ined the Chemical Engineering Department at the University of Florida in 2002 and joined Dr. Andersons research group in 2003. He earned a Doctor of Philosophy in chemical engineering in 2008.