<%BANNER%>

Nanoscale Energy Storage Electrodes by Template-Synthesis

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 E20101119_AAAADA INGEST_TIME 2010-11-19T19:54:51Z PACKAGE UFE0012360_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 105361 DFID F20101119_AACCLK ORIGIN DEPOSITOR PATH sides_c_Page_021.jp2 GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
646d6a390d153fd047d8e262ca3dc408
SHA-1
552c10b0e94d4dbfbc80357baa235449c7934736
110927 F20101119_AACCLL sides_c_Page_022.jp2
15a6d0d72e91cd40449d74a0c9499220
31c7ca80640eb9f3e3dc784436a9065ca430be7c
82339 F20101119_AACCKW sides_c_Page_004.jp2
7bf5f5b5d1798c12982d57a21b8575fe
b7de76ad92aa84fa68393638ab66f11bb831e0bb
103141 F20101119_AACCLM sides_c_Page_023.jp2
dfca63f41099a0cecb6eb6e09738bc8e
16dace4f86fdda668c8530eff1de2b928c943297
1051983 F20101119_AACCKX sides_c_Page_005.jp2
971891a2a6bea9a05b2dcc65e86556f5
cb0e5dd1a91660e489d4d01a8b898f6eb681e238
956898 F20101119_AACCMA sides_c_Page_045.jp2
79946008f5d246f4f52503603983c63f
767a9152f872a0d198bbe3141c3064b1078a1f88
98113 F20101119_AACCLN sides_c_Page_024.jp2
2295a740f00504cd4f715c90c86067ff
d2832f729921e10be486abcdeff96720e1700344
1051978 F20101119_AACCKY sides_c_Page_006.jp2
5622ed36c1492d7dbc72e65a141a844b
21a87f629af2ac4ae5d2e73200f1692922d8c7bd
545782 F20101119_AACCMB sides_c_Page_046.jp2
31165e16900d9a6acfc6e31d1ad54e08
f32deb650d805f2846c734c6718dcfa6edd73c37
104090 F20101119_AACCLO sides_c_Page_025.jp2
c6b9372fa41c7fbf8fea7d72100b35eb
2ab4c5c4b25f8d982c30e6f2b277121b8f2cf03b
906738 F20101119_AACCKZ sides_c_Page_007.jp2
7dcdf83e0bcaae881551ad604e0a7f03
934b383e86f06097091587d156e4d02338d5367c
403383 F20101119_AACCMC sides_c_Page_048.jp2
0341a03741355f279ae3b9cf4a38bcfc
f444520593c7441fdab46f43136e7e4303a276f8
100353 F20101119_AACCLP sides_c_Page_027.jp2
fbd40622a6d0a842a35f32665461ea44
1e602a1cef6f5ac2d07f8ba8cd83a7fc4c14f691
112426 F20101119_AACCMD sides_c_Page_049.jp2
bda47857aeed38cae73971dcf03f7c5f
448434060e15956a192f0cad58449bf50d9b344f
106366 F20101119_AACCLQ sides_c_Page_028.jp2
c5b2704888d041c04837b75f40a6e9f6
0acc6b1c005dc3dfcd7f674ce5221c845052e91b
110718 F20101119_AACCME sides_c_Page_050.jp2
d37d1f66d34dab6939f33dbf29ec0d75
7f3e8f2b799845b3dab62af4642439ea43178ceb
98312 F20101119_AACCLR sides_c_Page_029.jp2
5b0418b60c0c19f6c3b594daf8bfb492
5fa7ed548b350532bd9e7e8d640badb0f48f851c
105659 F20101119_AACCMF sides_c_Page_053.jp2
66994cda0c0f47e44fcaea930be467ca
fe5814a810ee9c4d814b055af7ed30f4d9e27fb9
112155 F20101119_AACCLS sides_c_Page_030.jp2
be187dcad6e036a2dfaae28dd5e90d21
40bdd59d16d94033cc5752239a47ba941c2613d6
61764 F20101119_AACCMG sides_c_Page_054.jp2
b3daf633d017cdd9b3d42e5279c44cdb
a5d18de9f384a5a2017d5a764b4c602d10bb6be9
93907 F20101119_AACCMH sides_c_Page_055.jp2
c1db21632cb65e0165790e65d9a3347b
80242ee1060d568da024dbf19420dcc30ea7dc2b
102150 F20101119_AACCLT sides_c_Page_033.jp2
7e63b6cee8e11cafb848c3d55ed3f4b8
a12ef2cf100d4c60862b16dc9683bb8deba00c10
103940 F20101119_AACCMI sides_c_Page_056.jp2
f84d5e73f00befb01a8c5a0a22510364
87a93c6275a0442ead4a7ab53e9566d1f6869eb6
793231 F20101119_AACCLU sides_c_Page_034.jp2
e3db1d12f1fea9a39f4f56326b931fea
e7f1f91a0a86f77024365b711fcaf52f2e65a63e
106463 F20101119_AACCMJ sides_c_Page_057.jp2
4e1c54411a914ae360c5cb64a441f1bd
31ac4a404f9bc1704b85b76cac798f132fa170aa
768254 F20101119_AACCLV sides_c_Page_037.jp2
4138959c3b5d5a49488202f8bee94378
37e9b79788e444c65826855dd0d1b8444b3ce336
94900 F20101119_AACCMK sides_c_Page_058.jp2
693733d8655b971e7fb1ac0f8b526cc7
908009472cfa207a5748185376742ca1f8228f03
95688 F20101119_AACCLW sides_c_Page_038.jp2
4941b0ad26dd58c34da6c934672b4dcd
8f6d0bd80d193a6b2523206cb083349711575049
1051930 F20101119_AACCML sides_c_Page_059.jp2
a495974cda64dbc4bc5b5f1162f28eae
e5f0fd9bb4a673a8d883d76087938bfdfdeb972a
761874 F20101119_AACCNA sides_c_Page_080.jp2
1bf004fb6d770181e8384e16d7879ab5
f3e9a71319cff7270f791b4c4193b8c40df940ae
106054 F20101119_AACCMM sides_c_Page_060.jp2
4ef8ea0fcd01eac73e5b85b06ce08382
077c79506ef3d449bcd35eab894c7600433e1313
1005315 F20101119_AACCLX sides_c_Page_040.jp2
762f8f747e5ed29be60cbc56996c8694
a64bdad06071017be8675fabc6765d70c1c0c0d3
106867 F20101119_AACCNB sides_c_Page_081.jp2
5be01d9f1226e8d8c872bd144a702454
3932996835caf6fbfc81d93caafc14bc6dec8d21
347195 F20101119_AACCMN sides_c_Page_061.jp2
d3a09c13ebaf556030dca0a87a42648a
8fd93ecf31d0ca2f2b322faf42a3fba44999f5ac
104851 F20101119_AACCLY sides_c_Page_041.jp2
c745bab083b4e1a6bf49c1832ec6d8c7
d841c721980aa5d667c3f66107c9803005ab449e
571636 F20101119_AACCNC sides_c_Page_083.jp2
d299147a1263609a57503f96e12ced43
0a75b50bf13f6de82a2729c68485d139b238fb13
700568 F20101119_AACCMO sides_c_Page_065.jp2
6b8894c0b6fdd072bf99219861725c60
72aa3fb15879d10ad9d1febdb7edf064100b1889
709301 F20101119_AACCLZ sides_c_Page_042.jp2
bff2efc26b3898f00520a9ee5c44bc27
eb0d553bc7cebb0cf057d656602815cfa9ebb533
420977 F20101119_AACCND sides_c_Page_084.jp2
d3b14f955f2ffff25f81ed7e0985a5b1
a7d7735b92af6c2c810abd3476ca190319c0565b
700460 F20101119_AACCMP sides_c_Page_066.jp2
293481a6f2ad9be3dd43c10523c1aea3
2519a3fe97630d466867f0c3cb559433e0e9adea
76381 F20101119_AACCNE sides_c_Page_085.jp2
ad65fe8248876fed3ff2fe27f8d7ce8c
e6a6ab458926d3aef50a5487134be1262b469dae
700682 F20101119_AACCMQ sides_c_Page_067.jp2
5938766d22dd4ab096bfc14a1c690b1d
2aea77050cabb4e7aa74315d9b192b97e1b8dea7
94571 F20101119_AACCNF sides_c_Page_086.jp2
61f106d266c877a179276aa71e857443
9a4bb6c6e947c62ef81744df91b13400f156fb16
62357 F20101119_AACCMR sides_c_Page_068.jp2
62c85a45797b04d774b2539afcf13236
9da64bff979e7200bffede6353db06bd14a698ac
993715 F20101119_AACCNG sides_c_Page_087.jp2
fcbba8d793688507fb7ce3464ac949ba
e8a3ae3fbb4ef1a86f75b512c41d3cd82cb21603
542736 F20101119_AACCMS sides_c_Page_069.jp2
51bf36bc5e0288ffee95f545eaa3c248
646edfda90f5bf3f77781d87b4db4ebcd195295b
102747 F20101119_AACCNH sides_c_Page_088.jp2
6231404f4f73b1e70eb3949924a07da1
a5cc2e773c0cb1f44e4bc75cb1685af5f9ee5c2e
102041 F20101119_AACCMT sides_c_Page_070.jp2
3f4186c9535b9e6d92d3aafe0ae6a3c4
dd51e2097e9f2a46b907a4e7cad005c28c9b8f22
722600 F20101119_AACCNI sides_c_Page_089.jp2
23e586f8981f709e7480139fda3f4138
0b2f64f1aec3a98cf99f6eabf8811eedc1eb7fd3
92800 F20101119_AACCMU sides_c_Page_072.jp2
d69aa521d74512bd50efb335bd8396ff
92fa665fb13f8f0dacc93ba44dafec4c2c60ef69
90786 F20101119_AACCNJ sides_c_Page_090.jp2
87ac32554a06d68d40fc7184e5f676b7
65a8f118a2bebc81dbf89d87a2678278edb46e1c
935614 F20101119_AACCMV sides_c_Page_073.jp2
1bd9b14f0feb4536febe3edf9b25ea11
961998dbfd5fbb048f522ab2c48528cc09e18e1f
780814 F20101119_AACCNK sides_c_Page_091.jp2
a709340af476382f9b06a868ad4e9d67
718393a0eaada24888de15e64c28a8c71d650059
106338 F20101119_AACCMW sides_c_Page_074.jp2
6675bcf4a580343e56fae8550b48fa72
f6e73abb1187c7191aac23f3701c71b0e682cee6
100922 F20101119_AACCNL sides_c_Page_092.jp2
8afccebf3e89a3490c329edabfc43962
18b8832e5a4fe9dd6c371e3b8c1b82848554c432
810015 F20101119_AACCMX sides_c_Page_075.jp2
bcc37aa16cc11c1376d61e7e7300d131
d16d2b54dc07e5c38e7688d3069bffe28a51a7d6
742018 F20101119_AACCNM sides_c_Page_093.jp2
e0f9aa5c76517fff48906c79ae44528b
2c1eefc3d8e666d0a5b0341c630f0d3e1848f10b
101570 F20101119_AACCOA sides_c_Page_109.jp2
0445925c41514847f011aeb2a2684d44
65e020dd24d3bcf77bc28835ec436dd7ca4ebbc2
761222 F20101119_AACCNN sides_c_Page_094.jp2
e0f1887c501bdc8737c48362700bd82d
96f9de6feb616955378b6603e0faddb7eeaeed03
91073 F20101119_AACCMY sides_c_Page_076.jp2
7617b20590fe4ffe9c3d0fe37f941fd8
42cd9e13a3bce82c3a13361ba8dfd1bd6240ae3f
105186 F20101119_AACCOB sides_c_Page_110.jp2
d1fdf36d951476c22a8e13e16865efac
24164d3eb4b3285ac3e5178924d86b84a0a300c6
93944 F20101119_AACCNO sides_c_Page_095.jp2
4b1091793bb073b5f23a01658107801d
6a26432d0f01055d4506a62c1da57b064515633c
769404 F20101119_AACCMZ sides_c_Page_078.jp2
dd7f94e3cfd12529004dde699d67c21e
c0675eb73809b8bbc0442ca64758cec618870912
83922 F20101119_AACCOC sides_c_Page_111.jp2
980f8bbfd4fc5aabf2e1f40e52a577c0
083764ceaeebb0d11301e5a98688ad670aafa5ff
765330 F20101119_AACCNP sides_c_Page_097.jp2
2fc842f26cbad4ae2004cdf0910dc826
7e8642f78d72fbd5e9c4f3eb82728605f41106d6
1053954 F20101119_AACCOD sides_c_Page_002.tif
40c4e8085648fd6ef5988616bf19e89c
666d4c1e8bf1bd8d0635eae0d85e098d3cccfcfb
603349 F20101119_AACCNQ sides_c_Page_098.jp2
86e4cbc18ab271277505323277239290
53dcc018c5cf2f99c8ae0bab119c911048899a80
F20101119_AACCOE sides_c_Page_003.tif
51ae196b8814308724a65ec5f2652d4a
4434de07305b0896cb8f18fe3945a1d2484220ac
84719 F20101119_AACCNR sides_c_Page_100.jp2
257a20fc6306f0cf0ab269e5787b8edc
dc2c7bf5d58a347a19ea042a4b9bbe2dc2b25ab7
25271604 F20101119_AACCOF sides_c_Page_005.tif
2fc8e6fef7affc9b08bd679b852a14ee
f89343fe2acc3addfa2f3da89f9dd1cb9caa06f4
110543 F20101119_AACCNS sides_c_Page_101.jp2
a0f9deba96e2b143067d398b74dee725
c2e03612a8c19420bae9a6d070bfad3f34065628
F20101119_AACCOG sides_c_Page_006.tif
3044ce954b1333e9b4ce5a14888f318b
77c6df3419cc53a4b3421244c65d3d77892645d8
109708 F20101119_AACCNT sides_c_Page_102.jp2
d873047f8ff3f6e9de41d3cd8e9304aa
b3ff91787102f39c87105dc3b11077dd4c36eeab
F20101119_AACCOH sides_c_Page_007.tif
f938329f0d51602fff83fa4e6ad4b122
b1e7f105e760aca897a9ab3cd831be9d6fe06835
111781 F20101119_AACCNU sides_c_Page_103.jp2
688e3729eafe408b033e9ee703041fbb
9d847d785248898b49cba855514203b70fd8b0a6
F20101119_AACCOI sides_c_Page_009.tif
ef4284fd2c81b89fe2f7b079caa064e8
277db7e94431c1ac849c99e3640a949f0c2d2686
106274 F20101119_AACCNV sides_c_Page_104.jp2
803d5fb897f255bd318c9eacdcad910d
6cf51cba7139df9722ab44ba046e3b9a6285b5f6
F20101119_AACCOJ sides_c_Page_010.tif
90f586019a5edfd04172d40f7204f810
0a3d76577421726cfba29751a113f7dae01f2f2d
101584 F20101119_AACCNW sides_c_Page_105.jp2
a6d07ff790d3eea77c94488c104096cf
9cd82b84aa3133618afd2336e7635ac7cb4eb845
F20101119_AACCOK sides_c_Page_011.tif
a49ffc56d0c38144bd66f96adfee27d9
e9e06b2a2ffbdf915dcb895687a06a1dab76f61c
85156 F20101119_AACCNX sides_c_Page_106.jp2
dc83e8b80f2ae80cb87b0ac2e9dc023d
0f16f7e7439950934324468cf8d4e1fa6acd22e4
F20101119_AACCOL sides_c_Page_012.tif
7d25f33a9a896e4998ffb7531e2218ea
d19bbc8a4c39b5d067c7687401c1709627953b6c
96054 F20101119_AACCNY sides_c_Page_107.jp2
d96fe9dfda3f1716de68472a704f1764
7c2ddd598141211f074acac98ec855bc106b0c06
F20101119_AACCPA sides_c_Page_032.tif
ca71298a9ebc069a4dda909cef83b98c
6c7f6793dd54510cf6cd5ad63b88ca6c9eb32f1b
F20101119_AACCOM sides_c_Page_014.tif
0d9000040340a0c48c7059fa9e609d64
bad8e24949d395cd122c65ad9ba0173e774e6abe
F20101119_AACCPB sides_c_Page_033.tif
c650e014afbe5061ce27c6bdf4a563ef
161552f4a8b7530aaeabb2abeafda17900fcf874
F20101119_AACCON sides_c_Page_015.tif
c15a2766293e8a4ddd4c8007ee762058
c8200694b7b68b6c940d2fc2352a5406d620d28e
105925 F20101119_AACCNZ sides_c_Page_108.jp2
8d9c22afb33d57dd96f566ade01f625a
1b9b0ae93b2523e27a2221b6f7ebbb41ce606a72
8423998 F20101119_AACCPC sides_c_Page_034.tif
e1d3d7251d8a0fb819eb128f8d57e639
9887a46cc24ab89c254a245926999e450be7b4b2
F20101119_AACCOO sides_c_Page_016.tif
68726ac7b905282f2d8a632bbab65b6f
481fb0d0f5b3ba6715f142b58f3ca2691fc23233
F20101119_AACCPD sides_c_Page_035.tif
5cd783598ded8fe7a9d6c28bab1b8d0d
c46996767d99cbf446138ada2bbd9769d2ad2fb5
F20101119_AACCOP sides_c_Page_017.tif
a2414ad07ffcbf8617a6f02e0ae625d5
2150d973e5460c2272ae8d3c82c954d030ce03a2
F20101119_AACCPE sides_c_Page_036.tif
9cc81f38add315ffb9ad36e2b4bea23c
cc9dd0ba23a241e9a4cf6095bb7561c47d8bb334
F20101119_AACCOQ sides_c_Page_019.tif
4cc0f9b968f3e9f1e1c52bb2fc3a1d4d
2fcd6adf200c6eadec6adb796e5fd6c57c94f2d2
F20101119_AACCPF sides_c_Page_039.tif
408cfd946ad32be2d9fa135fc0385c98
7e3668d4731b56f7430104a55e0f5117a36bd896
F20101119_AACCOR sides_c_Page_020.tif
8545dab58d25b7a657ebd842b5d0ca41
ad605f774a1a3f3c3630277deaf23a3e10e29b50
F20101119_AACCPG sides_c_Page_040.tif
4bd867cd03f1e238e48859956c510a56
c0261aad231a3193164804ff20c03e817405148f
F20101119_AACCOS sides_c_Page_022.tif
8ca8033ffc81074449aa71569a42a91b
f4bb8065af4d87494de2e8b2cfbaf365aeb4b111
F20101119_AACCPH sides_c_Page_041.tif
eb75d3238985d63f357edd41e7f2c44a
7af294ea3a57c27067a767124948452227237ff2
F20101119_AACCOT sides_c_Page_023.tif
25dc54b1f77b5a89379bb5a51db4c841
9763e6b4d7ed8b64fc0b8899e22f6cb7a50c45a7
F20101119_AACCPI sides_c_Page_042.tif
792899d6af58a20e393c2c9a22b10c92
a4f9040cc36624f6839195e7707031f2d0ef2c27
F20101119_AACCOU sides_c_Page_025.tif
753a9acd1613c106d1e7afa75fb90929
1127f3854e77a7d607e112b1c1628302112ab098
F20101119_AACCPJ sides_c_Page_043.tif
0b2296cc051dbe4996ea4c5da45e7b18
8c0a1a83e02c03badc22c20b014bc61c04010925
F20101119_AACCOV sides_c_Page_026.tif
b2cdb2bbcf0ec2b2db0a5ceed299e3e8
c72e15d72c177cdb2e382237b4f5b288ec7559a4
F20101119_AACCPK sides_c_Page_044.tif
75440af152d54211531fdf1c1f6bc202
5366b52b192f08ad09aa1cad3ef924144d94d90c
F20101119_AACCOW sides_c_Page_027.tif
72d3f01c9a9135155cfb2ca99c4b0e85
73a00f6b6e58d550303885d2eb3b129d725decc2
F20101119_AACCPL sides_c_Page_047.tif
48140d14f346ecba72bd66de9caff5bd
3a26d65fa63d92cc9bf4bf310770bede49f409a7
F20101119_AACCOX sides_c_Page_028.tif
d7a310b80d0eeb0de64d089f00c8d575
be6ccfbaa8d4fbcbdd27f1639cff3d0e661bd13e
F20101119_AACCQA sides_c_Page_070.tif
37cd6cfecf3de54d4a6c92325df57f83
d240353949c12a9b2d46b2bc8db64448eb92e362
F20101119_AACCPM sides_c_Page_049.tif
6da0ea28acb4418a8a515a4fc2ad4e90
934e6da09282634592401567cd3d40e4168c72c5
F20101119_AACCOY sides_c_Page_029.tif
c1111700adcfd89e2f1bbbf0752499da
1958b352056fc961b40f8aa5f65a27d787d1e5fe
F20101119_AACCQB sides_c_Page_071.tif
145087eed3b8f74c5eace97b1ec9505c
81a7aa416d2e9a4ddb27ee4669590307aa530e5b
F20101119_AACCPN sides_c_Page_050.tif
3313c48266f73a08ed1627c9a6098c5e
3b0836d9161c5c5efe92f0355e06351db979861c
F20101119_AACCOZ sides_c_Page_030.tif
89af19ce1484e63637139067a819bebf
0746f8b0627ec2d492e18ed7484c01007967d7d8
F20101119_AACCQC sides_c_Page_072.tif
27d6ea7758bdb22961e333d67d089aa3
3dd691a2a724c060919061d914991ac28d2ce265
F20101119_AACCPO sides_c_Page_051.tif
5c16dd28b9165e61e56af06d91f7d02a
731d3a813c47dbde96c6e42f9582ecbff0cdddbb
F20101119_AACCQD sides_c_Page_073.tif
7ca50f43d8e9580e70fa4fbc63cb2b27
1d8048b795282d736ea57bd509a324e2c73c8618
F20101119_AACCPP sides_c_Page_052.tif
47459d3b5aa2cf43c6af4f4bd2abed1f
6d0378058565f67ecfec4bb8a44762389a400eb0
F20101119_AACCQE sides_c_Page_074.tif
dfc02218dae9b1594ed4c18f599c6ea5
ff9ec3491c9cb78153e6c0bc8e5bb5711134a031
F20101119_AACCPQ sides_c_Page_054.tif
d0bb2150602178aafcdd297618687645
60f9de7d5eda18a962f19c58f28a349de0df6b0c
F20101119_AACCQF sides_c_Page_075.tif
64e1f4bcba2986b9a20547d62c2ae709
67123207b1615bf3a13ddb79712e7e9db42142e0
F20101119_AACCPR sides_c_Page_057.tif
33dc26f54531fdd2bfedb3bb3bb1c1a8
0320d93980829e29e1b87842d8f9223f1578b264
F20101119_AACCQG sides_c_Page_076.tif
a80e3f5125a1460466edda7a26f536e4
f6e48c120147076ae9263b4f86af1f474f870d7b
F20101119_AACCPS sides_c_Page_060.tif
8dac766af4d5499f9885ebb6af5856f3
a541c9a16000a2ca95c9e30067ba5ff2822d0fdd
F20101119_AACCQH sides_c_Page_077.tif
98120ac2133e2c12539abcadd9458340
074e2eca8f7677c16422edec1ddbeb03bde7cb1c
F20101119_AACCPT sides_c_Page_062.tif
8040828b90bc78f1363c306decdac37a
5b69aa69df95373c577a760768138add71e2dd6d
F20101119_AACCQI sides_c_Page_078.tif
97b67494397f5dec742d6a5a7b9c488d
686cfa5eceb5111bab44d790e3643b8bb2cf6a27
F20101119_AACCPU sides_c_Page_063.tif
b0bac5ca8f867d0e6c3393b3d69efe80
3c5ccf622860511d81a38f92f725e19bb6d11218
F20101119_AACCQJ sides_c_Page_079.tif
4048739ac65e04982f9166ba8f46d48b
5d66c7df41c8ba70ea1ab0f0cfb55d94fda884ff
F20101119_AACCPV sides_c_Page_064.tif
20d1d89f65ef26fb25e827f30d52fe78
faea35e497b0a10ead75961ad2769718c06e8d04
F20101119_AACCQK sides_c_Page_081.tif
24a1c1a8f4ade6247c8f4b37a4acb432
260d57607e9b3de4f54a259fbc7eec83558b0cf8
F20101119_AACCPW sides_c_Page_065.tif
0299e84fe231845bac48d23960aa9851
6eed049bfdeb4c8036c8f29c1c2dbf9f735185a8
F20101119_AACCQL sides_c_Page_082.tif
413af501616edc4b4b49bd44ea46e70c
c3fe9bc23147028ecadd80149cb5a60d2d7a3678
F20101119_AACCPX sides_c_Page_066.tif
ea3a75b461e44860b46a0343273a806a
375025bd22a01aa53d1858ad01dbe26c9d825f15
F20101119_AACCQM sides_c_Page_083.tif
12631173d9174345bae7d3dd72928901
a94984cb0003877df6f8e0c8487975ed6b89069d
F20101119_AACCPY sides_c_Page_067.tif
08b541c3f1896437715f1d049786a9e0
ffb4c45915e1d8df1c37b86434d667f05d9f4a18
F20101119_AACCRA sides_c_Page_102.tif
6bee3dc253926cb14393fd1482b18456
429f9bb25a887979414aad9e4f04b3f76526e82d
F20101119_AACCQN sides_c_Page_084.tif
1b200c9bc7b39cd859238883a626ba0f
5cf54405a0cf9e9eaf016725b2ca1fe9bb0996cf
F20101119_AACCPZ sides_c_Page_068.tif
17556f14291487642212a21bd39c71b7
ce910b29f0f6d2291ef9158e85bd4e81809703ef
F20101119_AACCRB sides_c_Page_103.tif
fd72672bc79a9a503ab6b98fc583e854
cdcd2ff963987124f5434de1de4225cc9f92ed9e
F20101119_AACCQO sides_c_Page_086.tif
64253dedab354be55c43871319d36e51
a59bf9c106870a9232f5aee43cac7c9f45f6013f
F20101119_AACCRC sides_c_Page_104.tif
d0669e2fd37c95d8a969aa6e0b385753
b1b6d3ea3dae4c0d48ab37fe2e46317f8f26aff4
F20101119_AACCQP sides_c_Page_089.tif
478bdbc7c105adcd30e3a4d593372f06
3281e2a8bbb4fe5a83eac2fb74a428005916d983
F20101119_AACCRD sides_c_Page_105.tif
db87a3476ca12be2eed008d1f1f2c302
6c2338dae02e23e33d37deb7bb04841050bd9dea
F20101119_AACCRE sides_c_Page_107.tif
b90243d918f5b4b302c1308a164b7b47
cc03791c0acd5c3450df23ef3d5db68474e0a460
F20101119_AACCQQ sides_c_Page_090.tif
5dcd9c63ec040fbe42fa3ccff921489c
e9aa438069c5ee91966129e7328fb6236fd9b845
F20101119_AACCRF sides_c_Page_108.tif
66e27d623b8e2928ee054ba676f0a744
2961a419d5e4e4f3abfcf087c8e736957779edb6
F20101119_AACCQR sides_c_Page_091.tif
2bc6ffeb8f514f6ecffd70fc26904c4a
6f898b8360103e69ec2112b84536285711c56537
F20101119_AACCRG sides_c_Page_109.tif
dbbb22e4edf260d8d2ee2f45ce3cacbf
a97f499fce7c9a2db6d11b5e3951034bf9cf554c
F20101119_AACCQS sides_c_Page_092.tif
ff23edcd67b2beea5bf6f78ee8da6e78
3c4d3e41a8b2f926ba2996733f4482c2497f4be3
F20101119_AACCRH sides_c_Page_110.tif
8656c826ae87a7b332176f985341f3c7
a04a93940a6a55157229e9e7b13e8312f4654968
F20101119_AACCQT sides_c_Page_094.tif
93dd3d5182aa5966218a74783c781f05
56cf988afa6fb8db876eb3f21ef31cab391c0a13
F20101119_AACCRI sides_c_Page_111.tif
16d61093a2123d51e727eb4e6d57f6ed
484ae8bfa4f6c4192df82ee296611c894eceae87
F20101119_AACCQU sides_c_Page_095.tif
1b95b57337ae77dd77b7f1eaf6c8437d
5bc821d359997a5f6d6b3f53e94b1ac99278539f
F20101119_AACCRJ sides_c_Page_112.tif
f9c9ba550a547e51652de330df02224b
f829e0af5adf77f6837b32ab52507c29a16307dd
F20101119_AACCQV sides_c_Page_096.tif
17b85d11762cb13585ff2a387addd5d6
f06ad7f6bb2c4e3bd1482b0a4b5246968f724026
7538 F20101119_AACCRK sides_c_Page_001.pro
617e709f28afb0a8af0144457c158d17
a60adddd181f2f833ef72f870a712239072662f9
F20101119_AACCQW sides_c_Page_097.tif
a2f1730b8292d4bca12cc905549a178e
234fd8464421ecb48bc111b8aa1f9f0b078af275
1303 F20101119_AACCRL sides_c_Page_002.pro
720afd0f2a3d12023abd256d651e0dd0
da047d4ebd405217b0284c65170d39ae953b17f3
F20101119_AACCQX sides_c_Page_098.tif
aeb0578829bd3095c7d82575aca0b715
115d85e2b48643d5409656d51287f48aabdaa80c
51357 F20101119_AACCSA sides_c_Page_020.pro
d82e2b62b59cf4b7ef5b84c0fe2f2efa
1218c2634e7ae5b831a95454d8f7b83961b97f66
1767 F20101119_AACCRM sides_c_Page_003.pro
76b9f53098277430dd40c3210cc7d06b
6ddec0aec41805ed6221dc127a20a8e5a23e7ac4
F20101119_AACCQY sides_c_Page_099.tif
3894d215efbd883db13c16b69bc16c0a
f154cfd4ef0dd07b749efbac0d3a45a402c056d3
49001 F20101119_AACCSB sides_c_Page_021.pro
70b61951025a54254a898365268c17d3
604b5675059aac02d867507dce5f2d7d41ede3d0
37174 F20101119_AACCRN sides_c_Page_004.pro
fd5fdb7cd116c17e8ca51ba9a46a451f
2b6abb2ed4517ac7b1a96705306fab975b1e45b9
F20101119_AACCQZ sides_c_Page_100.tif
95ae2efc23aac9d7e0af239e54794180
dc4fbccf4b275a202fa0794134cb35883518a5e7
47200 F20101119_AACCSC sides_c_Page_023.pro
4fe3ba980aaa3541d1da671a414f782e
f76738229f6f1c2f05355f1d88124801e0b2e22c
80489 F20101119_AACCRO sides_c_Page_005.pro
47bbdbfc954bebdf496909b6c6f1fe4b
bbf5a209248f6a1d79b731c09e31f21245948d41
45929 F20101119_AACCSD sides_c_Page_024.pro
60e0eb77dadefe3a4c9e9fc2e54a5961
c591fb023ed852bb8adcf6f2e5bb8eb80b2aa925
106690 F20101119_AACCRP sides_c_Page_006.pro
0d0f3a5bc9375762f6990a0276741ee7
bb441d9601bcb8f784d45f1343b5a8ff0687f648
47966 F20101119_AACCSE sides_c_Page_025.pro
5763006862f69582cfc4a62980e5d09e
0821e9145794a52eea786d5a8256983400455013
42085 F20101119_AACCRQ sides_c_Page_007.pro
b4aba0505df02d854f957ef051dcc41c
692f30dc83f950e4752b8378abeeed672f1d37ee
47338 F20101119_AACCSF sides_c_Page_026.pro
7277d4034af41fca6423fcd93ed3e2e2
18ef8b120b62b2d4e7d7758d2b2e4014234b39b9
12213 F20101119_AACCRR sides_c_Page_008.pro
b714b21100834cd16e11a2b17a87597f
4bfc5ab1a6145af7045a45ce4373d5114d3ecb05
46171 F20101119_AACCSG sides_c_Page_027.pro
1a81a2c57a46f469638859aff58f5a3b
b648b3f80195b66fd14b2837554b0c9998a712e4
65019 F20101119_AACCRS sides_c_Page_009.pro
dec9d5edde2a6f01b61d13c27c59219b
88ed6ebcbd5558545fc5c24014c5084d2f474616
52348 F20101119_AACCSH sides_c_Page_030.pro
a94be02ce6f79b45006806197e79d9c2
1a17e24328164e673df3bd89ad0fbf21be3a0414
62485 F20101119_AACCRT sides_c_Page_011.pro
87cd807d48512e542046d023d0611b48
39c1d0c6c0df36a4813a198da53447e8eee44626
48533 F20101119_AACCSI sides_c_Page_031.pro
806ab2e091d166f05dd6ce97bafc1f6e
2482a64f4b16f19b81fbda937faefdb0f43eb103
38106 F20101119_AACCRU sides_c_Page_012.pro
1335be81689825c52b0c5c038266a03b
c9d06fe8069eb5020087c9fd1f410363fd87ccf5
50315 F20101119_AACCSJ sides_c_Page_032.pro
d519a7889403a0772ecddf7d8ab7c1f6
b9ce4e1a26b017e0035e0f78170a0254363e9b57
38630 F20101119_AACCRV sides_c_Page_013.pro
1b3c0769a36357aefdb434e80b4366bd
064679602daf29e51a91d87067899f25247250ed
34893 F20101119_AACCSK sides_c_Page_034.pro
50c58e855e720222c96febbcb89469ed
19f62639419a0b3ee852d033d982372718e95805
40345 F20101119_AACCRW sides_c_Page_014.pro
bf9267081a2950b3e71f8d23bfdb6533
3fbb776b1463292995105fb3e2d5d2902f2cb297
53710 F20101119_AACCSL sides_c_Page_035.pro
85de21e419c3398ab12514f6194a15a9
ff877a614cbe2427448d0257811f71e81950902f
37303 F20101119_AACCRX sides_c_Page_015.pro
f6b5fc07f92fb966d2a27f2853e472ef
079054dc85e087465c56ce5c216fdbfaff9ba56d
46677 F20101119_AACCSM sides_c_Page_036.pro
327d5bd2b0d0eebe4a5459792da76d5c
1ee385870a8f635e53f313714e422d57f1dd1428
49432 F20101119_AACCRY sides_c_Page_018.pro
284ca67775e863de161e235bc26a989a
86e0bfcf7c8c8145e28142ac69a74a3bf40295c6
43676 F20101119_AACCTA sides_c_Page_055.pro
2539e7ef2ba9fff907c95a1aca90af8c
823e53ce058f3e7a7c436243a0d4d3fd93b0b7ee
28769 F20101119_AACCSN sides_c_Page_037.pro
7494e3feb59ca2e33fbce5dc3b605543
7f2566c40aa10667dd154e956234698e29d4d1e8
31029 F20101119_AACCRZ sides_c_Page_019.pro
8cd6c0ef7b778a2e200c053393f32b08
f92797f614da4ac6d6920cefab553d2a5a838580
48268 F20101119_AACCTB sides_c_Page_056.pro
eb69b055a0263e73a76178b91185e1eb
c834ccb88f16e930f9cc3e86603ac8e8ac756b20
43601 F20101119_AACCSO sides_c_Page_038.pro
7f5c7606dec5616d3432c6ceecf779ca
9cd1b8d48bd34e5053e3975d5a6a6a4a9f020676
49836 F20101119_AACCTC sides_c_Page_057.pro
c5842183756df657497ca978f9cf7d82
91f42b5e57b83fdd2f3b9956ac35e1b340797c26
26548 F20101119_AACCSP sides_c_Page_039.pro
cf3f9d0f2564e6eb55f4fc0ae3cad994
4838a5791f98b6bfeceeca742b5366e414dea002
8258 F20101119_AACCTD sides_c_Page_059.pro
2291e09761295081c6e87175fedc0515
5d4eaca383e6e4faaf2a97aea5a857b4be0738b5
45654 F20101119_AACCSQ sides_c_Page_040.pro
a3a42205e1fa3f8d63b97f49d347e955
035f77007f2bef28ef095e9b38b3865fce2ff74a
50202 F20101119_AACCTE sides_c_Page_060.pro
c03d4000ef4e4b2f2c23ddd050999732
113066b9657f5a5e16bce10c84011fed35def137
31552 F20101119_AACCSR sides_c_Page_042.pro
a5f92c0ac59b3addef0492f6174da085
f0bda34192dddf004e69cc07acc315b9363c6248
42412 F20101119_AACCTF sides_c_Page_062.pro
84c2c61c078dc4c9e34accc3106302b5
67652b9b503f2614824ca8e43e87f7a98b2357c1
33009 F20101119_AACCSS sides_c_Page_043.pro
1e149257234851f37e7bceed6b13a700
1fba34753c62d98bdb0bfedb922054e1c7f92ed7
26782 F20101119_AACCTG sides_c_Page_063.pro
a52c5af43131c48886a7fdf8a230e637
ccf46d154c9cc52588493474d5fe8f35f0cce328
40315 F20101119_AACCST sides_c_Page_045.pro
69e479f258d90c88b0d9f08d40acfdc4
e95423c63ab313d4ff3de1e70c19ebb32606aa74
29178 F20101119_AACCTH sides_c_Page_064.pro
d8cd93fdd4b51bb9b13dbf6c6f5dfdc9
1aa5e085c84c5527122596392927d1969ee0af48
24942 F20101119_AACCSU sides_c_Page_046.pro
5e50e81a0648e4aeb91d5e4ca661981e
f2e1f87b0046ab96788cf112babbe07881f0b827
30780 F20101119_AACCTI sides_c_Page_065.pro
cc727e1cc53f79611ccd2445611ae4d6
e5c747a92e0cdd54a4c1771993145cfd221a83a3
44517 F20101119_AACCSV sides_c_Page_047.pro
aa10596ca2adfdfc9ca2fa5576829940
21286127db0c1b3d39ca3814ff5787dda4acbc33
33342 F20101119_AACCTJ sides_c_Page_066.pro
c8af12478b416a71464293baf9587d66
484499b138bf647a02420c321ae11e9cc31d9041
10568 F20101119_AACCSW sides_c_Page_048.pro
ffd2c501c869df15307274de605deb9c
f426ed6e7858b3c705d68f9cd726463ffa205853
33421 F20101119_AACCTK sides_c_Page_067.pro
be9e7bfe49bef1f1495935382ab1b9cf
9685c35501f19c751f1c075ceb4c383ecee42a31
52069 F20101119_AACCSX sides_c_Page_050.pro
b1250701deda98804773cfe9fbe87846
e4c92ca521d18d74d300e9cd0aedc0063b1166dc
27805 F20101119_AACCTL sides_c_Page_068.pro
81247d5b0c95b0eb28559afcab30ed1b
38fdd8bff68cde84538aaa02c1ae32361e107bb2
49118 F20101119_AACCSY sides_c_Page_053.pro
a018cf96e7da42e6928041a107e009e7
198abb48759035a37df00e12a5865fa6f6692301
47454 F20101119_AACCUA sides_c_Page_088.pro
c291c869aff1f1c30895d837af4aaa21
a53bac93d679bfb2edc1c52cb9a160c664d4cf6d
14541 F20101119_AACCTM sides_c_Page_069.pro
e894bd3eeac7d5d65c9f386a9b4320e1
a796c9719928f849e311931b2f68771f6c243ab3
27480 F20101119_AACCSZ sides_c_Page_054.pro
15bd539e3a15150c45556714f927f037
370ea75114fc5f9e735c92153ae17991fc048572
43504 F20101119_AACCUB sides_c_Page_090.pro
f8c1f978f0c1e9a20d658d37d6871f2b
dc784d85ebdf42952d4732373279bce2d5579b22
47495 F20101119_AACCTN sides_c_Page_070.pro
5983bf1ee338775b1665fceaaa83915e
d27cca5d6b8d9047376ebdb954c1269a1f9a02f5
29439 F20101119_AACCUC sides_c_Page_091.pro
6b295d5b61a89aea3daf24d65e160922
48b8ab45f9ed05e09875b689cd38e95bf490581e
24200 F20101119_AACCTO sides_c_Page_071.pro
d5545378c04482c74d350f68e6978836
1c845063e428ea7963669e4b01eb5f0ef0d70fec
30693 F20101119_AACCUD sides_c_Page_093.pro
bd8e51cc6ac1f26fb749b5c0710c9816
b0759f4f563877379b6588c0ea1192a4b2932851
49738 F20101119_AACCTP sides_c_Page_074.pro
2f0f8244704a8ac4ff6b4755505ac192
8de153c9a5d24263469afe93078a3b64c6ace04d
42164 F20101119_AACCUE sides_c_Page_095.pro
42358d24d0fa7b37617ccc88091212d5
ba6005822b14c649e21f74776faaea1752a3f254
25305 F20101119_AACCTQ sides_c_Page_075.pro
91336d9d9824ff37628526b69f67379d
980973b2c6ee4c6e52ff0ccf07bddd947e5cc156
22361 F20101119_AACCUF sides_c_Page_098.pro
f2927a382940d0fc0a6eb0644396e1b0
315a302cf705be39325f7665a738fd30d6cff6d4
40960 F20101119_AACCTR sides_c_Page_076.pro
f726c761e083c4183c7b687c83400682
e3ac48025538c2e4e4a6ee2d92bf7c2f3ff45374
22568 F20101119_AACDAA sides_c_Page_041.QC.jpg
cf81d06caa9b2367db89c3f9fad45a80
3e93cc0b3f255d0d9fdc6a9c61600ca058e55661
39062 F20101119_AACCUG sides_c_Page_099.pro
6bac8694d7e3bb0c8f503ed8909d8422
e79de0a50cc85b9a45f23f646b4e37099e354dd4
10373 F20101119_AACCTS sides_c_Page_077.pro
97035f57fdeb340dd279a699c4be7688
f596b5e03901cde683b140a2fb52d46c0b16ba4b
6409 F20101119_AACDAB sides_c_Page_041thm.jpg
53dbe9743fabe03119e631c6c6886aa8
5571e386c755c98295854276d35e15ceb8c7c2f8
38755 F20101119_AACCUH sides_c_Page_100.pro
5e1cb16d993d217b47cb22f177618f2a
c77ec6ff98f3d59e23fb90f03c751e9f4b09fa12
28435 F20101119_AACCTT sides_c_Page_079.pro
4b96cc5170ae9aaf4781af12ceaab710
605a0897f02411886e151b110353d7e8a263fab7
5022 F20101119_AACDAC sides_c_Page_042thm.jpg
eff3573eb12bb3c0e99529a6f1286c2e
ab7dcb347edd7cc3204376d7006b1f007cf32f7d
52145 F20101119_AACCUI sides_c_Page_101.pro
0e84dfdffe5108d94abbaa93709932df
e83b52c3fe77a71ceb01fdd4acbc5d78e9036e44
40174 F20101119_AACCTU sides_c_Page_082.pro
910bab594adda18aa1984f4b240f3ae0
0fcb17bf503fa01cbd38134f19f71d44eaa4d365
20800 F20101119_AACDAD sides_c_Page_043.QC.jpg
187873dbf5994160084118776ae49081
65434a696760d206310dbbc9afb6cb852d1b0ab5
50466 F20101119_AACCUJ sides_c_Page_102.pro
b1603dd66cbd44d2c2edbf090de55e95
d0224ad38bc96a0ac897c603d83cb272a945da30
13344 F20101119_AACCTV sides_c_Page_083.pro
5b894693a044005a5874e874fa4827ef
5d89439267ba7046195795a47562b84892d83954
6109 F20101119_AACDAE sides_c_Page_043thm.jpg
a890ffe8cf05c368a5309381a21cee9e
f18a386d010aebf78c26bb3fab8ec1904467297e
52617 F20101119_AACCUK sides_c_Page_103.pro
4bcfd47a359701982a1c9a59dd80885a
fcd39cd8eb410bdd4eda4e0e372fb10657a6ed9c
13183 F20101119_AACCTW sides_c_Page_084.pro
435df03083d8a20d9fede7f5099d7df0
71db8d94cf0c14d60e16f824411b41d0a311da8d
18906 F20101119_AACDAF sides_c_Page_044.QC.jpg
c74514e58f2e1042deb1affb7aef0191
ad7e09aac58a59948667f73600af6b8fffdf2ea4
1729 F20101119_AACCVA sides_c_Page_015.txt
0de24c6d7322badff8caa90803552568
f834b0ad78cdfa63de1e1fd5084b899deacb752e
47446 F20101119_AACCUL sides_c_Page_105.pro
9ebb307f90f5a30f28a0c240d562b0cf
628b3983d9b7183007d8b158c4196106fb0d626d
33271 F20101119_AACCTX sides_c_Page_085.pro
fc130829caa74466ca049d8a6bf05acf
a1f9d72239ec458d9877ff727d987582fd248067
6021 F20101119_AACDAG sides_c_Page_044thm.jpg
0d8bbba720ddddad32a1222c478cf3b1
eb801a393c9951265b0c120f196cd251259f49a5
47996 F20101119_AACCUM sides_c_Page_108.pro
ebcb9cb16f9466797a1539bebb47ba50
c40963f90d58830cce0ee49e457fdb53a1ba2ecb
42670 F20101119_AACCTY sides_c_Page_086.pro
69688f830cbe1ab881ca3da6fa78fe31
94855910160f674a132d0d5f6699873afb3d11e3
22256 F20101119_AACDAH sides_c_Page_045.QC.jpg
699bed91304986478d1f7542fc777f92
a3e27ffacbe6589cf87e03e01ef36f605b5a1d23
1972 F20101119_AACCVB sides_c_Page_018.txt
726eb85109b91ffe0075c5b2dc354e85
57e1c4ac269a0528c9c5542b30cecefb42c8d308
46528 F20101119_AACCUN sides_c_Page_109.pro
8f03fb60ca3e7e08ed115ab015218c13
d0697d76251282664ae721af3fbf27f5486d37c1
45698 F20101119_AACCTZ sides_c_Page_087.pro
890e615df9cc60bc3d0352ef6a9f21f2
83fe372b5f04b69e7f96f00688d76fe43a59d21e
6398 F20101119_AACDAI sides_c_Page_045thm.jpg
bce6f13f43e7ad941246dbe70edfab40
4d1797f3828affaa6eeb01822edc9f5f14096f40
2021 F20101119_AACCVC sides_c_Page_020.txt
7d12ebde3062f382026f9412173a4e59
c8f20eb542b32910bcace4e052f3fa544f1be132
36756 F20101119_AACCUO sides_c_Page_111.pro
82962c134afe4eb0efd7c09705293569
2aff0b7838a795b7d3666767d18c882e6ab4a1bf
4508 F20101119_AACDAJ sides_c_Page_046thm.jpg
aaac24fd9a262783471e97d7cd34886b
c5ab9f8015e2f32eb2e09a12f57d3e2af9db8e1b
1931 F20101119_AACCVD sides_c_Page_021.txt
60884ff4fa8654dc41e0dbcbfcd188a7
fc534ed6363333bf891b1ee61eeb27bb7fcaa46b
16276 F20101119_AACCUP sides_c_Page_112.pro
fb56d14b8a5f948f28a720f5ce02f9cd
f260db5bae7f0d9e4a4b0c8464f8949847540cdb
5798 F20101119_AACDAK sides_c_Page_047thm.jpg
671b17518d8ea151eaeaae3cf0923c18
791fbc15b5f30090f2f09631f30a8cb0196e6dca
2025 F20101119_AACCVE sides_c_Page_022.txt
69c2ab9cb974cf4667dd32e26a95ab42
3c12164f05ad64a0d4f282647cbaac0b59d40280
412 F20101119_AACCUQ sides_c_Page_001.txt
f5e19ba2843fffce5aac4a8d60100e6d
3fc89d7b5499f94f779a5cba9bd30f1ab0d5eaa6
1865 F20101119_AACCVF sides_c_Page_023.txt
ce7ce05a79ba4a0d62fd0a1cabb48147
bb279247bb4ceecfc61d581dd12384ce2c962c1c
1527 F20101119_AACCUR sides_c_Page_004.txt
0801c743ee720cb7cfaadfe325ec9aa8
446672e000fe6dce7a98458b857b0e320829d134
6245 F20101119_AACDBA sides_c_Page_057thm.jpg
4e5f12ffd2614c9e4fa24ca06d82ed91
6e810c612e06a77c730f69af5acea4c6d8c76d65
3371 F20101119_AACDAL sides_c_Page_048thm.jpg
f9dde985259eeb11adc87d57770500df
24cccc85890c9c6bf22c393a4ef015d22ca596a9
1821 F20101119_AACCVG sides_c_Page_024.txt
3963c30d5d9572c536ccc8784a472e31
0f8341313886d6024890d533633ef37930bbc48a
4300 F20101119_AACCUS sides_c_Page_006.txt
3625fff12a9fa397800ac9804d176461
78669bd559dbd8110bb6076c368d209c5cca3667
5984 F20101119_AACDBB sides_c_Page_058thm.jpg
6bce6adfb4489130f86888a1da453c53
e12a0adf63cc5da4b64c690ec6eaeaea5bfc32c6
23783 F20101119_AACDAM sides_c_Page_049.QC.jpg
92e5efd576fcfe4c125e5359bc0e6ce5
a9fd19bc95dc245ef2562564ee79c54cb8e1f5c7
1899 F20101119_AACCVH sides_c_Page_025.txt
2ce5128cf23a0fd3c4de7c7ed3557948
91182e22fd277b7bb2a8c318e071ab20b91e5111
1694 F20101119_AACCUT sides_c_Page_007.txt
6c21c62ad55ede5380fe337b02327c13
9574099f016a55d6b8e2e39dcbfabe8d6e0538ff
22420 F20101119_AACDBC sides_c_Page_059.QC.jpg
bf0e7e1b448e657933fd31d1ec712e71
6ccf1b95b78b5571019bc1c91af7f38f062591a0
6460 F20101119_AACDAN sides_c_Page_049thm.jpg
86050926f2b57cb4e41a58322be4e276
3b6f9a436abc03125adaf7c692931d9cb83667a9
1882 F20101119_AACCVI sides_c_Page_026.txt
647c0afb05bb253610764341eb410cbe
8ffccce7d15801a4c30aadd6cf99fa4993af396e
537 F20101119_AACCUU sides_c_Page_008.txt
70d83bd0401f8d54fd51c99e066727d2
d3557a3b9c59dac88e8a8b6683edfb3bcc7214c7
6683 F20101119_AACDBD sides_c_Page_059thm.jpg
d068506ffae1aa36024405dd76e247bc
88e0fee7ae0abb8da53d6de7d2b1ea4b92e5cbed
6537 F20101119_AACDAO sides_c_Page_050thm.jpg
052e4078b7d78fe6e751443c1e3be52b
a84292bb84ca86f848af3cb92c97e1bc6dcd3443
1827 F20101119_AACCVJ sides_c_Page_027.txt
a84a32a333c251f48aa14d9e87593183
66fe56545e08422811aa31f8ead1b692ac954af3
2980 F20101119_AACCUV sides_c_Page_010.txt
ee7e47c2c33a3fe454d453a4ebbb648f
bd2fd2653a52f69f04a470460185143300e2c6ce
6437 F20101119_AACDBE sides_c_Page_060thm.jpg
944b5664e904bf07c9edd5ec7831dd45
1c9908df7f392bda84288169ad9a2e66f70c0982
20184 F20101119_AACDAP sides_c_Page_051.QC.jpg
3904a32c4510bf2eb68b6d576f50e7d9
dbfd2fb33c1517ce4c8b74f90c8ebb6fb63c2686
1960 F20101119_AACCVK sides_c_Page_028.txt
70c04ee07311d098bdfdc9157b45b191
30f75f75f8d39986868f440b755a16d7a9fd1f20
2538 F20101119_AACCUW sides_c_Page_011.txt
24002cb7894d1d0dcb5eefae1e63c2b5
f85ca2e32000503f0ecd6665219d46724386c211
10721 F20101119_AACDBF sides_c_Page_061.QC.jpg
cd99aabacbe70597beda2465bfeeceac
3061b298f80296e27dec8f3c5a235cd5b4a4c061
5541 F20101119_AACDAQ sides_c_Page_051thm.jpg
f9ca54b80b03eded3c6ec9e68800355b
82129690f91796a2b1dc374f53d3811955e72fb1
1831 F20101119_AACCVL sides_c_Page_029.txt
6ad677cdbde383b3ab1cd174a78226bf
f97742ad29cee1c50a2cdc5c83e4be1eba3a2945
1669 F20101119_AACCUX sides_c_Page_012.txt
46727395f012e9b5e3188fa866e07dbf
e3c54c6a3eaebc3c9882246a8cf4108c90793393
18949 F20101119_AACDBG sides_c_Page_062.QC.jpg
6b2f5f06d3d9f1867c31e1964058e639
efbacbb198ff8acfdb3ad5a5693faaa1494e38a9
1568 F20101119_AACCWA sides_c_Page_046.txt
1ac22bfd7e6267a213f4b407b88f84c9
3eea14b4df145ba45906fdccaa0c8c8e904fceba
4876 F20101119_AACDAR sides_c_Page_052thm.jpg
2341902eeeac3cfee82d08e728302293
045fd0d68008ff693ee11cb637e278a02e8167e3
2061 F20101119_AACCVM sides_c_Page_030.txt
7d92d90e0f2fbc8c7d8073bd746e6144
375e63c992aedd2bc1de1600878a9a7eccdd87fd
1545 F20101119_AACCUY sides_c_Page_013.txt
ac32706c014600242e420d93996e6b00
08a2f37329b396f4d68a107da1227777fb3966aa
5525 F20101119_AACDBH sides_c_Page_062thm.jpg
a95bcac31da8d8708b8e7cf007169260
a08889f889746b7fd0351997f7870a502da11a04
638 F20101119_AACCWB sides_c_Page_048.txt
89093a090d5ba5d6510c2377beaed514
d33b15c2dca14b3132c365a5c6fec53ff122efa7
22944 F20101119_AACDAS sides_c_Page_053.QC.jpg
d90c1fd832608d1bdd554a4eebaf0b11
a40446ab91998ec98c86ea2f54c41ee8802899b6
1697 F20101119_AACCUZ sides_c_Page_014.txt
78e2af58dd58e0743a189383c972dae3
ec03f1c3171397c4cb1fbab38ff8b30d56030c59
16191 F20101119_AACDBI sides_c_Page_063.QC.jpg
5ce80c42c441088423c6706e61929bf7
345ef433ed7f0cf15c4bdf9bcb4fbc4197b1c09f
6481 F20101119_AACDAT sides_c_Page_053thm.jpg
50aa30d928aafd815e1b1bcea3c178f3
c4111399c57a93423f77952747001ba5af78f2de
1936 F20101119_AACCVN sides_c_Page_031.txt
bda9afee7c88814e7000871c71a61ddf
2fcb567d8ccbf072f289cbf2085b188f229fab3e
4711 F20101119_AACDBJ sides_c_Page_063thm.jpg
55132c0562a164cca7aea1d6c49e76a2
4d718dd2803ccb164d1c58e11fc47a0b65f71c61
2051 F20101119_AACCWC sides_c_Page_049.txt
acf24edd97d991f6b7f7cce0e3865432
d3ddc07b2c047fa19717174f41e2415e2be249ee
13704 F20101119_AACDAU sides_c_Page_054.QC.jpg
af0f684f89d40b9be02ad0822138a3c7
2d517829f4af5cee91dda6f72dae92c93491f1a1
2008 F20101119_AACCVO sides_c_Page_032.txt
14896cc3e76594a78ba32cedd5767d4b
0e6e92dbe63b8c534ae8dc532da4b155276a6a86
5052 F20101119_AACDBK sides_c_Page_064thm.jpg
2f4c4395242025f7e91503bb66ffa2de
d2cb2e8d3be3f74fcfa6f02d9c9f2e79c5a0ed60
1798 F20101119_AACCWD sides_c_Page_051.txt
923280fbed787107aa89c1dd309f42d6
62c672ae8312417b2ad34da6f6e19bd9a63c91f9
4058 F20101119_AACDAV sides_c_Page_054thm.jpg
3f224842d3dfd3698d1d9eb5ab56186d
3c53e4dc9aa6a556a2fb102766e07a83136b742a
1909 F20101119_AACCVP sides_c_Page_033.txt
b3e690e088e9511f554fc65939ecb1a0
2f6515cfef3cbe03577a67ccd2417b8834974e26
17227 F20101119_AACDBL sides_c_Page_065.QC.jpg
b1b50c2e7fd1f1d82c1e1ffdd958f98d
437df98491bf077e32b77842926d3cf317bc7798
1973 F20101119_AACCWE sides_c_Page_053.txt
f02597ddc4db54e08dd5160efc73c634
0a5636d3d1518a11d40af6de2bceae0f910f245a
20796 F20101119_AACDAW sides_c_Page_055.QC.jpg
a86b3170a9bbd7879b996f0a5f98b601
8d24394991a60a249e71baafab260ac406e1d4f0
1577 F20101119_AACCVQ sides_c_Page_034.txt
2b8e9da4bf369bd41a2e0124cee6e9ba
d1c1267c6ff77ade1123bb163fab3a410867513d
17697 F20101119_AACDCA sides_c_Page_075.QC.jpg
94515fec5c4e1ae1b981e29421a224ab
3a145a660ee2a354390c33f5909adcc77e11f617
1098 F20101119_AACCWF sides_c_Page_054.txt
0b29bfcec5c225e9dfa349b288749ca1
69e456ef8bccc65599e495e276614da68ec21913
22405 F20101119_AACDAX sides_c_Page_056.QC.jpg
53fe3ee686e16ebb3b5899e7655a480a
6d33ac2529b62bef214158baf05c6c1b59498a11
1905 F20101119_AACCVR sides_c_Page_036.txt
14775b448dede1755c77ed3ba5cc7558
818643b260d619ce99f27683f12bb1f744c59247
5477 F20101119_AACDCB sides_c_Page_075thm.jpg
96e1fd1eed316e38c56503ace115e89f
aabaf145dd50607fa9752d784467ed02efb6d3ca
17326 F20101119_AACDBM sides_c_Page_066.QC.jpg
f6cb3f838665e9c3a2df3da7e3badd38
23104955c855a585a5188c232a7749a6eec79c39
1833 F20101119_AACCWG sides_c_Page_055.txt
68bf87f85fc3d11004a975f16420d24d
e551a99d10cfaf1d4ff4b5efda97ad295e5f636f
6367 F20101119_AACDAY sides_c_Page_056thm.jpg
a1f7d772f9857664f26cf934939a37a5
f3cc945a85b8707d5dff7040768a4b1c02690224
1327 F20101119_AACCVS sides_c_Page_037.txt
fbaea6ee30be7b320c7800623df55891
bf7b0becf6a6bf6f30c32bacfe3b5917f4a633e9
5639 F20101119_AACDCC sides_c_Page_076thm.jpg
c98dc1baffbffd990e51009a4d111072
fd4296653d30f501ceb39b022c94174226c1edf3
5461 F20101119_AACDBN sides_c_Page_066thm.jpg
e808c907afe9841a48a8dbc7aec07afe
7a7e6d6054d869f6572f12746bb8e3547f002af9
1935 F20101119_AACCWH sides_c_Page_056.txt
1edc4077655cc46ecf7dcfc8533586e2
ba181939cf61b5e807d506e7df1ddff03ad0036f
22948 F20101119_AACDAZ sides_c_Page_057.QC.jpg
7a60ec01d03f6b1bbbd03c4737187887
0f717439c824a5c49e79e5236acbbf3d1edd03d1
1824 F20101119_AACCVT sides_c_Page_038.txt
5dfe6253d1cb699aa504fe19e94935d7
ba2253572c462b033648fcaa86f0c23206fcd4ce
16297 F20101119_AACDCD sides_c_Page_077.QC.jpg
dd7d5dad4d0f53984c8f9efc61dc4325
1d6fa5f9d13f0290a6e93d542547db9f2c1ea294
14409 F20101119_AACDBO sides_c_Page_068.QC.jpg
eac06a1cce6604e76b199c82436371df
4176bdc304081d65bc9b45f003402b21d8797bc7
1996 F20101119_AACCWI sides_c_Page_057.txt
ce615082de356c4941e88a7373d49268
fca96d70fcfc8fc682dc72c0dd5df31124c3ad93
1406 F20101119_AACCVU sides_c_Page_039.txt
e7137d4f92758647debd142f03726ca5
e585514e744be78d1d616091d2f4a47ba61b271b
4978 F20101119_AACDCE sides_c_Page_077thm.jpg
79c543c9b1a65e07550deb40848d12d2
da47527f4abd98098c51f2ed66a22a72a905f022
4086 F20101119_AACDBP sides_c_Page_068thm.jpg
45675912c921b3585848581d6dbf6236
7b151326d4207ca4a083724333dfb2372b048331
1776 F20101119_AACCWJ sides_c_Page_058.txt
1497b5cf80ef1945db48da9e0a1ddf28
753dd31c5509aeccc625d2cfdc6c4780de861afd
1934 F20101119_AACCVV sides_c_Page_040.txt
5266a281d4ed531e6bd1a2aa1672ceb5
10cef94a1700d655280e2b0c4b997738d315dea8
17191 F20101119_AACDCF sides_c_Page_078.QC.jpg
7058f6dc9ac3ed67b92bfcf24cf65eee
607848056dee851aab31f71338454cde45facf89
11915 F20101119_AACDBQ sides_c_Page_069.QC.jpg
31c02655f66413814938c4fb9f77e6c3
e46de46ef3bcfeffeb99d607408e7e6f0995c378
1148 F20101119_AACCWK sides_c_Page_061.txt
0a2cbd94201feb14d265353bec87fb73
83798f753c4160571b0422886ea7449cd065e649
1900 F20101119_AACCVW sides_c_Page_041.txt
52e8759c28aa8106021cdef6fb384b8d
309297e1fd4ed848f475348c63bfdd79d0c36dfc
5472 F20101119_AACDCG sides_c_Page_078thm.jpg
f4921567250144463a9401a0bc2cefb2
c12401b011b66b5555cf7f7a56b3337cde4041c4
667 F20101119_AACCXA sides_c_Page_083.txt
3ae79bffb1026d392b3a3c9f6b500086
6bf227ca437e7789d02e52a059315bf7e21ea944
3836 F20101119_AACDBR sides_c_Page_069thm.jpg
79572058ed0be3d61130e5a247e46d21
9bd73aead0fdf9f930c05949efaab8428c598b8a
1440 F20101119_AACCWL sides_c_Page_064.txt
c8c360588ac94b94e7cde320679dd022
3165fd328cc39c3591110ba71a8565e55f7afba6
1394 F20101119_AACCVX sides_c_Page_043.txt
9453fdb68f1ad4c275e1efaf67bb9a58
bbb021ffd7268d062856cce16c0ddbc8af9c9e8d
14798 F20101119_AACDCH sides_c_Page_079.QC.jpg
128779a3531fd45335c5f5e65d753c21
f360bb6716e29deeaac7b70c47f31f3e120c6fe3
636 F20101119_AACCXB sides_c_Page_084.txt
2a0d77552a82175c8191d3a646d5bf24
d27a22ba9bf3734345b5b4da2d953ed17b4f78bc
21899 F20101119_AACDBS sides_c_Page_070.QC.jpg
16ff92a0db703a586442c283035cf4d3
9f7da309cf01e4d9fe84ded089edd967001076ea
1474 F20101119_AACCWM sides_c_Page_065.txt
c735dc1bf953c04751c7046ebcee2c66
092a09ebb359f7253e942f9328d86d8b2593bf40
771 F20101119_AACCVY sides_c_Page_044.txt
9a1e85b06b9f0c9a16c8df74c4dd61bf
9021a418ee4b031b6e49d3baa9f0bbabbebfd784
18463 F20101119_AACDCI sides_c_Page_080.QC.jpg
41f55183f51de05429f7d9188be1f3ff
d37c3916730948f6b58e3423c79b3467eaa03ebe
1778 F20101119_AACCXC sides_c_Page_086.txt
29c8b2223b4e299a8647e985a52cc9fb
996256fd377daa8042bfe999c6a7d09826e2835a
6115 F20101119_AACDBT sides_c_Page_070thm.jpg
99e0bcee7109bb08dbddb0a6847707f7
1ba0fc4e8cb06e2f86555caacbd99c513ce72d7a
1626 F20101119_AACCWN sides_c_Page_066.txt
14cdc600519ec8a038cceeca28ea992a
8433431a8f2b8ae89236363558f81689e003b4f8
1923 F20101119_AACCVZ sides_c_Page_045.txt
893a52dd022094d32cc19c4411278c19
7bf273941ccceee300a2ff770dfe47f2c9d2e5a1
5493 F20101119_AACDCJ sides_c_Page_080thm.jpg
640e8dfb6ff38316a23de89b40c3852a
51685254b4bcd4571c05214c181f00e084fc7e42
3709 F20101119_AACDBU sides_c_Page_071thm.jpg
0c8293218d18bead5f31ab40a74aac3d
ec53deeade48c5756c2093a47462b302bfa2adb6
724 F20101119_AACCWO sides_c_Page_069.txt
642d97fef60ed11a7a560bc754974771
1e009b6e0dc94e0086390a57fd7ccd67a41af40c
22724 F20101119_AACDCK sides_c_Page_081.QC.jpg
2fa07824131cd119b6be3ad246f6d1bf
0f15a9dda2ead0d9914e42e80ec650322d1c9cf1
1915 F20101119_AACCXD sides_c_Page_088.txt
a75d568ff3dd60143ba0fd4efdc06b04
b7e296ba8ffd7096d198e09f47450229d728fcc8
20516 F20101119_AACDBV sides_c_Page_072.QC.jpg
e3367c7783a2dc77458f59ec25a66efb
ed3fa35e11fe6babe0dc4e4276d43d2556e03fb2
1921 F20101119_AACCWP sides_c_Page_070.txt
2b700686fd7222d54ec3bf37d036fb2f
5c80704b34e6aa6b00f23a6c9730ed69bb03e2b0
1335 F20101119_AACCXE sides_c_Page_089.txt
8394c07bd7093423a8f6e81c6bc0ee94
6b31354490656807b709d300f5f0ccf89262360f
21433 F20101119_AACDBW sides_c_Page_073.QC.jpg
d2ba0542014d31bd74acd38c4d6d8823
5a25fd5bd7ae71bd2bd57d3e2910437de6e1b486
1769 F20101119_AACCWQ sides_c_Page_072.txt
dbc9e61e48f1d9f060d222aec3e73f67
c4bb555b07106cc50ce777ceb968db5d48a2d21e
6215 F20101119_AACDCL sides_c_Page_081thm.jpg
e32d7ee298fe6053fd971f062a0ee67f
563e971159818823881afb494eabac38e677bc17
1853 F20101119_AACCXF sides_c_Page_090.txt
bd3ad1f702112198c50e190b1ae01466
058312d9c85f2d54eb57bd2ece034dc6d6a0b030
6516 F20101119_AACDBX sides_c_Page_073thm.jpg
65a2969ef113e99ce9baf85a23472182
9ca32852deba6e8068a16c2f9f3ed9e38a40c55f
1786 F20101119_AACCWR sides_c_Page_073.txt
8da69b61dc655e813c0ba3ca7bd1280e
a320c255e7c0c77a7721e52a7f038a38c9e497c2
17343 F20101119_AACDDA sides_c_Page_093.QC.jpg
0d0721b09f7d930952b7a6e6fb0e0187
5ba1cb1ee49dde522674264881ad3c5369952547
19119 F20101119_AACDCM sides_c_Page_082.QC.jpg
dd0ca06f9932118a498315173e06645f
1674b1e5f8e73b12ab21a794f04ad28e24092c0a
1496 F20101119_AACCXG sides_c_Page_093.txt
5fbc901e58d26b54607f6059e543f5a4
0bf8fa89c342745221f1f91433f3c7d15b5b18d4
23439 F20101119_AACDBY sides_c_Page_074.QC.jpg
22c4d87e7f927633e0b20fc735e80322
51fd86dbd5bdaa315af105a4a18409d016b295f8
1965 F20101119_AACCWS sides_c_Page_074.txt
ee47c765d5454c84bd6714ee49c4af9c
b0743f7da3fadc7e78b2a32ba3b24d50241169a0
17258 F20101119_AACDDB sides_c_Page_094.QC.jpg
d16748eaa111c967e2e7d895f9e58ffc
60de28a655e652b133fb8a47b081285367563ca8
1280 F20101119_AACCXH sides_c_Page_094.txt
17f7563f5638bb82abdfd28f4a28bf9e
68ca46c818b02ee3aae808a036bd9db53536bcbc
6568 F20101119_AACDBZ sides_c_Page_074thm.jpg
85a39f88fd99ad575f92bcbd95c14142
1c2719873f9cd83d368efe9fe6c9bb9ebbaa66a0
1300 F20101119_AACCWT sides_c_Page_075.txt
147fc94b239ab5cb2b383b8f0b368c35
21a7b274f3d85e76e6ddd43d44f2b2550ec5a48d
69536 F20101119_AACCAA sides_c_Page_081.jpg
2a25ef416d989203e82727ed9a573330
a80865dfbd4f84a127a1c1b65edafb1e0953a2b1
5274 F20101119_AACDDC sides_c_Page_094thm.jpg
a990ff39f07f8550d6836b2cab850324
f70726956334751d638d4ab78568ea99df806e68
5534 F20101119_AACDCN sides_c_Page_082thm.jpg
3a26751a387d36feea69cd59a53628d3
d98ed475823fc9b4e6cf5c5f647d305174abd2b1
1087 F20101119_AACCXI sides_c_Page_096.txt
46d7c4c1e9a0cb589f06643880a06db5
c7cdd2544a8e626d8c4f1423de0f0e540a700d37
1637 F20101119_AACCWU sides_c_Page_076.txt
0b67887483314aa00ea8b1825797472a
7d51dff107fcf872b6cbcb4892c8799210d236e6
44124 F20101119_AACCAB sides_c_Page_058.pro
1624afdce4162570a5f625e650ea14db
c55755435a25320ce5710b4f61e2f24835617903
5812 F20101119_AACDDD sides_c_Page_095thm.jpg
94daa69f879a06d5a476e0b9eaa13330
cc5081688094f7865c633765f63f9b5881212026
11952 F20101119_AACDCO sides_c_Page_083.QC.jpg
2868ffe755f90ddaa281b86e38799764
1c9848298473e2f30d02e647807665871a61cf9b
1651 F20101119_AACCXJ sides_c_Page_100.txt
8dc43345cebf31b8109149c0a0e74043
03d7e26ea631f90e65fde53ffd52d75dcd5a6f2b
1649 F20101119_AACCWV sides_c_Page_078.txt
d3fa8f23032178a4a8f734914d8b9e7d
50b6b53b32f9dec7d9188f5f556fb762a966f0dd
17481 F20101119_AACCAC sides_c_Page_042.QC.jpg
3c16d478a1c557bee7728ef9939f6ad1
5ec3624e596f94a01207fba91d724fac3980bc67
4471 F20101119_AACDDE sides_c_Page_096thm.jpg
8134007fe50730cb5e2d000f809893bd
73f873268af2ac89fa37aa624e5eb064ad40b02d
16695 F20101119_AACDCP sides_c_Page_085.QC.jpg
4d5f898539df3696c15e744e1400e6e6
1f80784910f4ec962c147411d748fae57ed6f5a3
2121 F20101119_AACCXK sides_c_Page_101.txt
5ebbec0bf47477fa49dda23892de5562
eec8c783268ad7cfa55d19cb501a16e27b16d481
1141 F20101119_AACCWW sides_c_Page_079.txt
deab260b01c1e4bc3d19464f16bf453a
7416b5e94e8bea505054becd97891fb26b8e87d6
13610 F20101119_AACCAD sides_c_Page_098.QC.jpg
67e39af4922bf30a6e233b0cb826a0e2
4397efabefa7b85ec08ea8cc92218fe1bbe6d3f7
5299 F20101119_AACDDF sides_c_Page_097thm.jpg
c32601186ba0e916571dfdf81da23db4
036d8a0a50ae32fc56b1e5aa859a2cd27b499817
4849 F20101119_AACDCQ sides_c_Page_085thm.jpg
78f548a09595f76248855a16a450190c
d0ab3717f21abe80266aae7bff947ad3aee56f56
1995 F20101119_AACCXL sides_c_Page_102.txt
9a948078aceeb99e08c573c948ce60dc
8888869125905dead9b5bc8a731eb26b4f5564df
1349 F20101119_AACCWX sides_c_Page_080.txt
b85b3d1eb31173857cf0cf6d3b6afbb5
7586ad100386b29db643130fc3f91f24dd067b3c
96626 F20101119_AACCAE sides_c_Page_047.jp2
d21461c8518bdd31a9eedc22c154d736
b07104c19c22648d6eed1a100890d8cbafd6adc1
5411 F20101119_AACDDG sides_c_Page_099thm.jpg
91a7a9c7935216b60c83ac3c00656861
a4ebcba82da958a8591a2e9eb323c01a6c9ad22b
24067 F20101119_AACCYA sides_c_Page_009.QC.jpg
b95b59d2665303aa83f0cde39d647c51
7fb0065924f44a868b168f086aa61d9e4ba108a3
23275 F20101119_AACDCR sides_c_Page_087.QC.jpg
e5fd67d2c3ba55691cfdb6022d3fc126
557a37da80b4927c476cd066e3eae77142d331f8
1879 F20101119_AACCXM sides_c_Page_105.txt
cd16b3b2a81475432a0acd91636d2181
bb7654027bfa1e2e1fa1ba6d98f03619f663e403
1939 F20101119_AACCWY sides_c_Page_081.txt
d96621beff8b245f44287024d1658c3e
28410651fec5d0b9d6121ac7d6944a9eebc03f00
23856 F20101119_AACCAF sides_c_Page_050.QC.jpg
0d25232613f19b3c8c934230167a6d9b
29b1eea94c0b5b16600381629b438f83a49a4478
19379 F20101119_AACDDH sides_c_Page_100.QC.jpg
6cefb4231b6963e12eca551cf16afa88
ccbb439f6ed73deda40e06b5f561d295347e6278
6248 F20101119_AACCYB sides_c_Page_009thm.jpg
1f5d4c250cf1f92030c46534bdac6da6
b548d392d0eb74fe33c8a46ac86e8719950c8b40
21980 F20101119_AACDCS sides_c_Page_088.QC.jpg
0b62aa5c4059840ec7fa95d45487d252
d215fa76374a51a80213dee464f9c19c42a4b30a
1771 F20101119_AACCXN sides_c_Page_107.txt
95bad64d38274cf05cd2c633aed14f1a
d13c93e2d705bf188096510acf794c168e781bb5
1600 F20101119_AACCWZ sides_c_Page_082.txt
82d82477ee32797e8ed1748cfa3ab11f
b3d30d7b884923a670b0844673c53a7512756fc9
58543 F20101119_AACCAG sides_c_Page_015.jpg
4b65520c4e291b5a04213e9485584ec1
584584d6baacdbf279e660a89d407230e581edfc
5370 F20101119_AACDDI sides_c_Page_100thm.jpg
62f8c9b00a5b07ad79dbe617becbc3fd
ae756b753806453ede91988906dd308840de1382
28713 F20101119_AACCYC sides_c_Page_010.QC.jpg
095276c366face1b642fe85217ebd105
96c142b3c82f67e33f67027e9c04e0e5397d3bc4
17792 F20101119_AACDCT sides_c_Page_089.QC.jpg
27ee8a418e0b7c1bd629c7960ae04f82
4a0df4a334ac2cdfcbb069a2ce19e92442c186e2
F20101119_AACCXO sides_c_Page_108.txt
7c84e9f30b5b17d7df047b7616bf8af3
2b365c9e077b22dc47879edfbe50fe973b6f6b27
29314 F20101119_AACCAH sides_c_Page_052.pro
5300e7f10a0eea92ed3646d23c86a69e
c0e0326cc895dc97542635d927f3c925ef70411c
6598 F20101119_AACDDJ sides_c_Page_101thm.jpg
a51f43e308f21edf402b7ee3c8604f18
3ef108c81e69b656dd31436cf27344c801780f56
7118 F20101119_AACCYD sides_c_Page_010thm.jpg
e7aed39857b6fc32dc25dd1a9eaf31fa
d2216980b7b2a73aa52d026ec35aef4fd7adc378
5551 F20101119_AACDCU sides_c_Page_089thm.jpg
3f2def087c01d44fabfdbc0456d684c6
169a534a4edac9f9c9921848a4a6e492c3a4226a
1964 F20101119_AACCXP sides_c_Page_110.txt
334ec095b06240359413f4d832a8256c
7399a3ba1fb311d37b583d4f909099b2bef2ecd4
46184 F20101119_AACCAI sides_c_Page_029.pro
3f516c45169f6e476d8c346b30364181
da44bfcdc2cb37ef49e517200f832d7002d05d1c
23839 F20101119_AACDDK sides_c_Page_102.QC.jpg
44db7557a03dcd45a4eb37a06d66f0fd
adf7ef718cb9e39cd9698f7bf8ee214c366e982d
20002 F20101119_AACDCV sides_c_Page_090.QC.jpg
5d467dc8b277089fec367f12a327d2b6
f69e7a6e28930a8d34d28e91d620bcac9fca9314
1519 F20101119_AACCXQ sides_c_Page_111.txt
c6559595edae80e9dbef30d06f1531a7
6e0d37187c6546f23328ce2601a08ebb6d1ad52c
17406 F20101119_AACCAJ sides_c_Page_097.QC.jpg
6e199f9f381c3fecba650c9b2e3b33b5
48d3d0e2e53cdfa7d0f746cae39f281d3c2aa53e
6587 F20101119_AACDDL sides_c_Page_102thm.jpg
065a21f063afe0731a8ad38b2462aeb5
683c545dc63fd5b05b8533498a064f0fa9e950bc
6188 F20101119_AACCYE sides_c_Page_011thm.jpg
20d93d961e9c5772dd274f80b5dc0ebd
78996982b3ad9b5dbc82eeaf41bbf0924e82eba5
5710 F20101119_AACDCW sides_c_Page_090thm.jpg
2cbad44c1418da51346d594cf1a34eff
228b77dbcd4e023a99463375645f7319b3625bd6
694 F20101119_AACCXR sides_c_Page_112.txt
443c977fbbd591c7faa0f3897af7c5d2
f57eca60164eda3519249db699114ba62335df0e
6321 F20101119_AACDEA sides_c_Page_110thm.jpg
e3490db7a811d9c370bb7c4c6b71698a
39077f876188a88558eb38f0c2e8a8aed543676b
67815 F20101119_AACCAK sides_c_Page_105.jpg
5cddb84ef1490475e19024754dbb5c0f
fa3f489a733a47e8b2d0117706abad76d087406e
24327 F20101119_AACDDM sides_c_Page_103.QC.jpg
45becfb2fcb4cb987d0ffafce3991aa8
41c332f3af648e6197e42e2c8ee79be47f953426
18171 F20101119_AACCYF sides_c_Page_012.QC.jpg
90771e872d0ebbcd549b1549f56a53c8
74e9d89130b64fbe46f257bf89d4464f1d668a19
17535 F20101119_AACDCX sides_c_Page_091.QC.jpg
423f945e69bdff87e5d333f0c46ecfd8
558eb5984e4228a4377fde64f2235e01e5f761c2
3096904 F20101119_AACCXS sides_c.pdf
7fda38ff65a9d6af84a9e2c3a31d8f05
4ff03ac214463512777acb11af96e961167c88cf
17629 F20101119_AACDEB sides_c_Page_111.QC.jpg
11243bae196a96ad09584393dea27d8b
5ef1a0c73b49a40e7a1d0fbd3fc996978ee11bb6
49442 F20101119_AACCAL sides_c_Page_104.pro
9480913918df510e9bd9b20e0ead9d7c
f0617065f07c1b2721c79f8108cd20b72cfbe7c6
6555 F20101119_AACDDN sides_c_Page_103thm.jpg
0d59206d2e5f104c29dac37e210b42f2
faaba5870b98b22c1b832a9b939c41682d872178
5166 F20101119_AACCYG sides_c_Page_012thm.jpg
75e41b23095c6c8e338bf446f48e4c1e
427670ce42c8394338b5066688288482040c6ad1
5339 F20101119_AACDCY sides_c_Page_091thm.jpg
65e1e6bcce86d31e3fca5b8d62f84b9a
d90146ff22b2ea4e9df2520a2e9d5a1330f8d8bf
10058 F20101119_AACDEC sides_c_Page_112.QC.jpg
cd2458fadb6a9fe8bdedf4aaa64007db
8fae6666b9a3e096f580c4a61269e2603960e555
3332 F20101119_AACCXT sides_c_Page_003.QC.jpg
29f1f44c7f5c1321caf25298b6749f64
48fbe4b842af4966a9d196381e0adb14e98ca232
68782 F20101119_AACCBA sides_c_Page_026.jpg
27d9e51ec298d8d5c7daa7c49b4330a3
a7f8da7e083955ab881c8ac17ec3617b659dff85
18378 F20101119_AACCYH sides_c_Page_013.QC.jpg
48e3508b85bc5152f11d8bc22be77a08
43c80f3e862c65d11e269d40e400e19220157b6c
6364 F20101119_AACDCZ sides_c_Page_092thm.jpg
030c5ea2721d30fd47b786322290f8c9
4f254f608c4d905e6d22ba3ac68c23443d946878
3069 F20101119_AACDED sides_c_Page_112thm.jpg
1ab829a5ed386a2be2f1d98ce2eaa951
3c8c07e3ddac15a97885f6a699ea956dc955def3
1428 F20101119_AACCXU sides_c_Page_003thm.jpg
71d0d7e573223de7acb9cc0d025579b9
4ef53bce72fa543c271ed4d58a56ef3d60943e9c
961 F20101119_AACCBB sides_c_Page_071.txt
6ae6b396e75cca7fce5ffdc8384e522c
825762aae93dc8f37211c3fdd6f14c79dc957db3
67397 F20101119_AACCAM sides_c_Page_088.jpg
09ca8e17609a867bd046c403fe82ceb5
e013a3587b9092e3f769b78d93f469c7afab1109
23266 F20101119_AACDDO sides_c_Page_104.QC.jpg
93c19dec6191522aa91b5d5847d3e0ba
15171b15bd03ea49162f8dc6b02da48c613e5fa8
5252 F20101119_AACCYI sides_c_Page_013thm.jpg
0c15c8e587442fb7b17cb79086742d67
a9e9a8dd40e7a1ca0787172cc86a6f9f2e0624fc
129434 F20101119_AACDEE UFE0012360_00001.mets FULL
c5d33c56ab16d01818e810218465f9e5
b938c1ffbc67bd72bb6c884e7eab9ae0bb256bc1
18890 F20101119_AACCXV sides_c_Page_004.QC.jpg
fe5521cf8daeca7834a3d89d3c22eba4
e9c91fdf21c420550a95ac455009ec65c98f6ffb
5397 F20101119_AACCBC sides_c_Page_015thm.jpg
a6b5088232b0c7bf062011f55634614d
da08ce6c76286dcfa571a1c8ca701448b61afa24
1051986 F20101119_AACCAN sides_c_Page_077.jp2
255119d440d9c0ba225bce41472c5f24
a647fe82b983453cf31cde797dff78628f5cba19
6410 F20101119_AACDDP sides_c_Page_104thm.jpg
06d1c1de885f54575b1b6492596748a7
24aa7a42d9121aa60c63f1414357c69912542b5d
5308 F20101119_AACCYJ sides_c_Page_014thm.jpg
c4fe58cec8f5aedb7e09d4eab71ce5c5
809dec940a849e4d8e9f8bcb62a0fe5022b279a3
18589 F20101119_AACCXW sides_c_Page_005.QC.jpg
acdad899584f2cfef6e2ab86e3e09111
52b6762472535b7c11442cc248c0b629a9c064f4
1338 F20101119_AACCBD sides_c_Page_019.txt
a24db9ac5034b7d5b589c0e189d90e4a
5ead0f7391ee79dc4f1295047200b6fd215c5842
7034 F20101119_AACCAO sides_c_Page_001.QC.jpg
a8dd238271501178e35b3f25d5bb0365
81253868eb3bcf45ae1f65ee4ba067657f373663
22022 F20101119_AACDDQ sides_c_Page_105.QC.jpg
4f87db545f5647262524941921c7f6df
ae7f8dadd5a9cca98caae0b9f287d2d3efc3331b
19698 F20101119_AACCYK sides_c_Page_015.QC.jpg
210be2ca39bbc683c1607dff1ff81ff8
1d37faa60f1e6c217dcb2f7c2c726760d9f00cfa
4946 F20101119_AACCXX sides_c_Page_005thm.jpg
a34f321f85e5bf62a16d482875a77810
b52b807caeb08c3813172d2e823b020e79b09fa0
9949 F20101119_AACCBE sides_c_Page_048.QC.jpg
e128fcc191d2ebf8d0d3a4585ced356f
f24148177d1ae4597afb35826a52a450a5afea70
39403 F20101119_AACCAP sides_c_Page_016.pro
6d3a7581c61fd70e6986c5127e76889f
6fd2fc173230eaba28ce12559f45161730f842bd
6253 F20101119_AACCZA sides_c_Page_024thm.jpg
cfd9e24dcc0b9570977ebbd87130c3c3
9462ee9f4c5d10613dfca67902772ad873197a80
6241 F20101119_AACDDR sides_c_Page_105thm.jpg
c072db2a8505b3eef593f394c6f2b71a
dcf9e6a9cfe48edf9d74660df1a399b4cbe7ddd7
21873 F20101119_AACCYL sides_c_Page_016.QC.jpg
d05c96fe5b4fe2418c6fc6ce60a4b52c
3423407db03e40e2eb879cc0a0bb640344a07e9c
24477 F20101119_AACCXY sides_c_Page_006.QC.jpg
9f2bf4ec8bc40cf07e16190fb0c83289
1e2b143978dfbc04d6963610af65aedfcde88d15
F20101119_AACCBF sides_c_Page_087.tif
0a422bfaa208f7fbe518112736284157
2b02b340763c3d86b0753c8daca5e60992203334
6260 F20101119_AACCAQ sides_c_Page_006thm.jpg
d544a7f6535cc2232b5f34cc8b9930df
e8a2cfcdcb39bc6059c6ba4d432a4b0cf340c9d9
22698 F20101119_AACCZB sides_c_Page_025.QC.jpg
5a9d18d434af17624d5e7c71e94a3997
210a9c6ce52f4f377788c313451c76b3cbb26f3e
19239 F20101119_AACDDS sides_c_Page_106.QC.jpg
210bc695c456e8ea68318f01f213d861
7a02e563f6235b2c835479724298121211fdcf39
6641 F20101119_AACCYM sides_c_Page_016thm.jpg
ca024b94f1961240b1a218877c4a2ee2
d2bf2ae5f7dee907c4d69e6b25bb2a5fd3eedf17
6247 F20101119_AACCXZ sides_c_Page_008.QC.jpg
e2dfc1e264f08dbb1402b702892c2dac
daf2ac4a011db5d869bd1ca27ebced5eae01681c
1448 F20101119_AACCBG sides_c_Page_052.txt
dc2005dfd2070a0d992bfd4bf0123a93
198406a573a1d410911265387a376e3fb6490fdf
19084 F20101119_AACCAR sides_c_Page_099.QC.jpg
4c511bf06d9047194940c50ef3fe56cc
228f9c6e797c4a9c4b5ad7c7dc72f5ef26800c3f
22526 F20101119_AACCZC sides_c_Page_026.QC.jpg
c522b557c7b2e185382b5e510d507a98
c7499e7dceeb85169148ee04100527f3f257f2ec
5646 F20101119_AACDDT sides_c_Page_106thm.jpg
f00b8a9220fb777a7e3da9e1c475413a
86d71173d32d1a5532f57a7727c114ccc521446e
20324 F20101119_AACCYN sides_c_Page_017.QC.jpg
4279993f68ea3acd2ce64960cc87f2f9
4e40e8f60a564869a9945c8c3ea75409b238e870
123 F20101119_AACCBH sides_c_Page_002.txt
685839f5577a0d28faa7886fc1861a9e
67d94e90f6850015b794e5a4ceda279dafd7dd0b
505 F20101119_AACCAS sides_c_Page_077.txt
396247970886263b47f229804e7f1209
13c9aed586a301ff873e21d0cfe440371a264123
6267 F20101119_AACCZD sides_c_Page_026thm.jpg
164dc54ca156ba71b2e7a34a3c438b83
45c7045b07657dcada0ccb66d38ab60b46e5b1a9
19627 F20101119_AACDDU sides_c_Page_107.QC.jpg
8af8ab09ea57ed2f5272db4a0e8a1520
15de0d997b4bd343c3e16fd218cd89f24541e862
6172 F20101119_AACCYO sides_c_Page_017thm.jpg
d5b6c701935f446d89bda8190192f434
ae8b00fdea47c21287eebfa7d1f9dddbfbe64a1c
1854 F20101119_AACCBI sides_c_Page_016.txt
09fe637df2da4f65539e53c65b1811f4
03d7dd6ad9638848542eaffdb9d3ae3ba42949f8
F20101119_AACCAT sides_c_Page_055.tif
606942aa218ebec72231004931b85aad
d03185f665b474af7d53b9e66e732cc50624cf29
22594 F20101119_AACCZE sides_c_Page_028.QC.jpg
0cd59e2e5f0dcc2dfb4c8009ab1dd8f0
18710ca1a41546a7e35002ce60b8723c36de4560
6219 F20101119_AACDDV sides_c_Page_107thm.jpg
3ed2befc6ad0c1f7089bd984d548c0b4
c64743bfddac19fe9db52ec6e261d51fea2e0437
22445 F20101119_AACCYP sides_c_Page_018.QC.jpg
1f06ece03f75c2dc795cd56f831f62ed
d60808b7a78af8ec519951c76f668258966715ed
686258 F20101119_AACCBJ sides_c_Page_064.jp2
75c150f01cab8322b608ad7db7a63ee4
8ea0ffb64c6162ce55b4b18f76014b5b20d529e7
3592 F20101119_AACCAU sides_c_Page_061thm.jpg
1439825e9fe1d68a0d24e2cf89d77e1c
e65617bf2ade9f1df14ef0cead6977b73e7aeaed
6656 F20101119_AACDDW sides_c_Page_108thm.jpg
ad545d8099d5bf4aa6ea9d922fc4e3a9
c4acb34ab616100ec3e208e6e58860484f5c6691
6307 F20101119_AACCYQ sides_c_Page_018thm.jpg
11f123f22719aedad11de96288c3b3ac
d98e2d7b40899131e9f936a8db1c2d64d59218a6
1119 F20101119_AACCBK sides_c_Page_068.txt
4b90632e5312531f45f5f53f03369959
b6859955d63a0aa50c4fdce82771b26a4da6d07d
47876 F20101119_AACCAV sides_c_Page_110.pro
49b919e00245e46788dfad47beab05eb
97ad0323fc73db6634308601190686f04ea107e4
6354 F20101119_AACCZF sides_c_Page_028thm.jpg
81347882b1214080a3986e5afbb93385
870c3e81c308e0c61ecaf14700cd3e6879e4784a
21315 F20101119_AACDDX sides_c_Page_109.QC.jpg
1179a8bbbf910f95cd6b69b7f9654bc2
a4a1cdad6f16b339cc03a62b3448dd1fa4ca3db3
F20101119_AACCYR sides_c_Page_019thm.jpg
277ef88c474e2e1bd19b6742d290998d
d631470396a68ddc9e7210fc424c25732d75c02d
48381 F20101119_AACCBL sides_c_Page_033.pro
d363e10606eeebd4a50c2acad1064adb
3ad37c7f63dcc22331d8a72689e4790337b97f5c
30732 F20101119_AACCAW sides_c_Page_097.pro
5a23fec53f33082d3056b8a72875d422
990afda83f4748a94342b858abbd13d3500564e8
21247 F20101119_AACCZG sides_c_Page_029.QC.jpg
4134df11557da78fc22f7026705633ef
c3e56dcfe082863966a3bf8238b9390a17c862ba
6167 F20101119_AACDDY sides_c_Page_109thm.jpg
3946a43cdf5db9fe61f55cb6d9c658ab
93f84fcbe713d60b5cb6b3e9fa8628bcec571eff
23752 F20101119_AACCYS sides_c_Page_020.QC.jpg
53b2da4da3bd0c1381a695dab5821ed5
1c7c3d13d53194d406d8da85a0e43faab2f4536a
F20101119_AACCBM sides_c_Page_045.tif
bab150b4fd91f178bf74bbee313f4550
b2ec5749282e911985ca046cb9e5b815dcf8da4a
69222 F20101119_AACCAX sides_c_Page_021.jpg
e25d1809e4bbd900846f4a976461b2cc
3365247aaf9c50bc0b2fe653b3624a08977a5cad
6031 F20101119_AACCZH sides_c_Page_029thm.jpg
07eda6d69e7a83bfa8a153e6dc0ccb6b
4a2a6a9d2adc07a4ba22f39d83b8144103a1b45e
21496 F20101119_AACDDZ sides_c_Page_110.QC.jpg
49b1beee375501c8acc4fb04e736094a
1cd4f5721261fca3cb62862a3716e540c1aaea51
6636 F20101119_AACCYT sides_c_Page_020thm.jpg
d3a19879ecd3c2932856b63d84cc0d43
4477ede906d82f4c509f89f795aa22bd1757abc1
1787 F20101119_AACCCA sides_c_Page_067.txt
fd59900b069886fd97277eb625d6d153
cc228afe392e95c14dc15676019f33c81a926c7b
4468 F20101119_AACCAY sides_c_Page_079thm.jpg
97434943353136d109b9748a78f255cb
d51f41b06d8e68101dfbba3d03232464f83e175a
24124 F20101119_AACCZI sides_c_Page_030.QC.jpg
a85d86a3b64fc4e2933def505137e1be
66918e5728c6cb2ac1606733a0a1258ac9e144ed
22323 F20101119_AACCYU sides_c_Page_021.QC.jpg
ca9cb187270955d019a3a4b9e456a9a7
ef19aad30f09c94cd562df39797e39bc029c2183
41528 F20101119_AACCCB sides_c_Page_051.pro
a6419406f51d56beca281a76da156d9f
4f89073c975a0b88174606e624bb4a89663d986f
F20101119_AACCBN sides_c_Page_018.tif
b5b5ac291c2ea08028ecee7a7a394f5f
60a85db1d590b1cd662d0ee35e267ffb096b551b
F20101119_AACCAZ sides_c_Page_101.tif
09d9d5d97efc2931bf65dec78fd990d8
c5f763e70f3fecec5488c1dcd0398f0b7d70bfc1
6618 F20101119_AACCZJ sides_c_Page_030thm.jpg
24050a29e0b3e58077508803205a5e15
0153a6053f5a5142c4d02e7da8836cdb9567f3c9
6235 F20101119_AACCYV sides_c_Page_021thm.jpg
e812c40aac17e19068489e11c9884ff4
ebb856d7ec3e91584ba2faffd165291eaf420883
F20101119_AACCCC sides_c_Page_037.tif
7a46d30ecac86303a1fa642e6377d014
eea1567f7c6e692eceff356ac93772a3efb2cd32
F20101119_AACCBO sides_c_Page_031.tif
e643c0e9b40140cd5b6f93a7dfeb68ee
f45d11dd6384267f8629d86859ddf714c87a6254
22752 F20101119_AACCZK sides_c_Page_031.QC.jpg
6e80461bb26ccd032fb0c98b212c31c8
dbe704112d64db8dbe771a14cf8a11694578109f
23909 F20101119_AACCYW sides_c_Page_022.QC.jpg
b71e380c68860197194db35fd3bdd455
0624732b24e9dceac53a60b4df7ef4c36a426b22
F20101119_AACCCD sides_c_Page_108.QC.jpg
3533ee8e6e2fea4103b8fa96541dcbde
132d4b02049a87978a708edef9c9f82abcdbe070
5408 F20101119_AACCBP sides_c_Page_111thm.jpg
a7ca21f5b775a71ff70bd50c4e2f15ae
112bae37cb2c4477eb5f35b3a88c96f18e0d6b84
23420 F20101119_AACCZL sides_c_Page_032.QC.jpg
8e3677dddf7c3a0e84f2346d9500a18d
7bed2f3f1135f4e7a7d57f26ad7630b604ca91ae
6592 F20101119_AACCYX sides_c_Page_022thm.jpg
9d94149d858ea6694872417f68e732a9
a08dbe983864c13fcbb3832a403d0d13a01e248b
86025 F20101119_AACCCE sides_c_Page_062.jp2
ec9a36d5455e1960acee7f60b7246594
1bb9cb1d5d734433a32f5b665f282b3d521d6bef
16145 F20101119_AACCBQ sides_c_Page_052.QC.jpg
1a4d97d759e4b0d25201cf48b4845d58
397cde681ce7f332b01f3c69d11a9e18a8d570e5
F20101119_AACCZM sides_c_Page_032thm.jpg
d3c9380ee8423c16d4486ec8b51eff3c
a3bef25d309ae461cf0daf7e65195b63d69463e4
22470 F20101119_AACCYY sides_c_Page_023.QC.jpg
8ea51e60c8eddd36b3d44e6a6eecf06d
f8986a876cc1e52413bf9f9c06c370fd70297c0f
73541 F20101119_AACCCF sides_c_Page_103.jpg
82c05d23934909eaa415d0704ecc261d
7ecfe9f93dafa7d9ad6830ecde7a543e3c6dfe34
55026 F20101119_AACCBR sides_c_Page_077.jpg
c863c8538a50cd967dab1d01d9aacf9e
6c880535c56ffe67d2427bda19699086c7eb3115
22089 F20101119_AACCZN sides_c_Page_033.QC.jpg
e066be8dd49cdf1cfbddaca6fb5fd5ac
5c5f736c968eb3148e6844770b3005fc4b88d297
6270 F20101119_AACCYZ sides_c_Page_023thm.jpg
06206fbaa0951519fe3f933a3e5fc106
8319290150e8d0dd1213642303a11334588d0719
1893 F20101119_AACCCG sides_c_Page_109.txt
b193da6c416680adfce8ba55f0a78c00
9cc91d6a852b7d9735cd21a9c8d34dde52507966
10011 F20101119_AACCBS sides_c_Page_084.QC.jpg
da8557e4875cc60acc38846b247263e7
ed0fa3580233219d5a1dbe55902222853d4d20df
19207 F20101119_AACCZO sides_c_Page_034.QC.jpg
92b45d61335231993ef5b86b24877766
e6a8be5e71f07d886b1893805d5b1857d547dac1
20405 F20101119_AACCCH sides_c_Page_058.QC.jpg
072c074afa67b1dd60175f5a7fa5d6ee
b15964a0ce38278698743feef7663bc90eca3184
21602 F20101119_AACCBT sides_c_Page_096.pro
5ab9f1ca6c1ad533ee3b0ef6c61406b6
1c70b1627144437f188c710d2d395babd3b88a44
5695 F20101119_AACCZP sides_c_Page_034thm.jpg
3288ad4d2de87aeba1b87bb15986d5b9
f89959e3ad666f8bd7591c88a18191e6fb72d08f
1366 F20101119_AACCCI sides_c_Page_085.txt
02fd2ae1cec7a8143c00339ed13d2199
9f5ffea479954457716960436a0e83df798d3a2d
105750 F20101119_AACCBU sides_c_Page_031.jp2
701bf4809c54a529147b0bbf2f585dc4
facc21950c1bf18d6d893063e1fbbd60306ef421
24483 F20101119_AACCZQ sides_c_Page_035.QC.jpg
6437379190721cf8ad88e806e6a670cd
6af6b49b0143dd43ac9f2a97ebc9667271d3e317
21473 F20101119_AACCCJ sides_c_Page_024.QC.jpg
e756ae73b48f03b32dbf243c4c464e56
86883faf0f1122d910195f57b5edfcdb9bec758e
2097 F20101119_AACCBV sides_c_Page_103.txt
7c91267435e743663174afa8c171c8a6
9a1ead90ceccaf6b00a8089b20479d62f0260d98
6797 F20101119_AACCZR sides_c_Page_035thm.jpg
ef3b5a81470727cd3aba0249795ec7da
58e8928c295978132bb36a3482e14e10eae973e9
14922 F20101119_AACCCK sides_c_Page_096.QC.jpg
9589ee691d264898acdc357abd475ce9
27c24aec43f5bc7ec59ff1e0e685384cca809470
970263 F20101119_AACCBW sides_c_Page_016.jp2
9f63e49abea4ad465e5dfdc9405ec175
7063d2068f43beee7ea760b159baff78162cd173
21220 F20101119_AACCZS sides_c_Page_036.QC.jpg
57baa32d08bbd5240342cd6971137e66
b478a45643fcbd0e095c2737e128eed9cfff586d
6553 F20101119_AACCCL sides_c_Page_040thm.jpg
fcf0aebd0ac6dbfa7b2704ddb840fb45
430ea9c7d1d27e2d1d24ec2c3d0a9fe0c4ebbd7b
48392 F20101119_AACCBX sides_c_Page_081.pro
b16cd86e08f6ea38b9cc023852d38695
79d958dc09cac84e0ad0aa6581b2b192222c169d
5890 F20101119_AACCZT sides_c_Page_036thm.jpg
3addc1669cdff3255451dd4e83f71f9b
0418e1b3a523b33e62aa016fca35b5c94cbb199c
49825 F20101119_AACCDA sides_c_Page_039.jpg
503572c0ae2b57e9443222cdbdc62dd9
f44c7ea17f2f77e5e73ee62947d43ea12ffc4833
108014 F20101119_AACCCM sides_c_Page_032.jp2
f5f6ca3814760fde7ba29078cfa07ae1
4c354a31135599dde074cde107d90f7a56e2ce33
17498 F20101119_AACCZU sides_c_Page_037.QC.jpg
7390f7fe0d7ce043fda34951e8d97339
f08e9e652bb26e83c42ee7c9f7b226e6f1ddbb39
F20101119_AACCDB sides_c_Page_088.tif
f85214d43d4661b4b4ba1cf2b6f42665
6baf245020d1f8e4be80103a8cd1f9e82438b52e
1289 F20101119_AACCCN sides_c_Page_091.txt
e0867576f1a4db0173fe925eead13050
9d8c9ffa4c0e5ff21d699be154d481354ccd6b01
F20101119_AACCBY sides_c_Page_093.tif
1ac87d5f00d27c82a91226755c6021e4
e11e2384fc1dbc9ba001b1a6c32e5dfb1c9646b8
5261 F20101119_AACCZV sides_c_Page_037thm.jpg
a05edbe2982fb43070b7e4b79ac256aa
01f7c568f46bc46c48a8470459e93ac09845c0aa
63862 F20101119_AACCDC sides_c_Page_055.jpg
6189e3fad7596ffb60c4cd8e3fb75b97
015697d74cd8bf9397698e9f31e18150db3d89bf
12749 F20101119_AACCBZ sides_c_Page_071.QC.jpg
dcc177023f627b58025cb66c5c678fc5
501ecd501ce174fe4ce758cbc36935d9c69f9a26
20695 F20101119_AACCZW sides_c_Page_038.QC.jpg
a97835575a3fb087006f047f0a53ad7d
67f895be8cb88bfb1e29d3c5c3ecb39727d00290
49615 F20101119_AACCDD sides_c_Page_028.pro
0db013cb847159933555c8f8f67481c1
4ef64dc90b31ea58693dd315e59bd4081b65a84d
103884 F20101119_AACCCO sides_c_Page_026.jp2
fc568322e4c04f810678a68593539823
7fddd8e8835204c681b2e05e2ff7c9a406fae547
5938 F20101119_AACCZX sides_c_Page_038thm.jpg
009a2fd60432d2d8b258fcaec39fc14d
2d5a175e4693f4848ff6028d47d7e31479a8e435
18191 F20101119_AACCDE sides_c_Page_019.QC.jpg
929c3a58b806ffdf36d79f34ad3fc3b5
c042a8d6ca185f0525f35492da2c6985fb006bc0
16983 F20101119_AACCCP sides_c_Page_044.pro
8625450b92b3adbf1b585a3aeac0f54d
6b49974fe860d6561bef7eb49b855eb8257f79b4
4885 F20101119_AACCZY sides_c_Page_039thm.jpg
c2b98137cdba32d42212eddda18c5120
46ef3190d177b0570f8cebbecfbb5331658314c9
20706 F20101119_AACCDF sides_c_Page_086.QC.jpg
9f70d99908de8a07a5ca169a733dbbec
67c06c0dc168ec889c32ff2e0ac6b5a4899635af
37896 F20101119_AACCCQ sides_c_Page_106.pro
d052108b8a846cba01ff860e8097499e
606c9e263ec3294d1eec40b921a5fda75bf19866
22847 F20101119_AACCZZ sides_c_Page_040.QC.jpg
aae05877a4bd7e1467e32da09c456551
0f49a45f98d3bafab376aa130657de33d015a426
41925 F20101119_AACCDG sides_c_Page_073.pro
045b839a3c6f58ce9a241a50e1bc143c
ca4a2f9fbcd95caa2574398c96c9e8ac7a7bccc6
F20101119_AACCCR sides_c_Page_053.tif
2927623c4109228eb0fdb7aa896b3464
c2928d3efbc03074663458e7e9184e813c6868fd
3245 F20101119_AACCDH sides_c_Page_084thm.jpg
6e24d1a8aaa04f1932d64c862e9a0717
c23a5be269d66392d62dcc1e5c776b8effebbb8b
2647 F20101119_AACCCS sides_c_Page_009.txt
4f8b3da78a03c6c0a7d8229271510e94
db857c01619c9d78ae500a97ff81206a3126c633
1850 F20101119_AACCDI sides_c_Page_092.txt
1e0245562f924aeec285d56f1921631a
ee99cd9132d265bc9e775cf0f11d7ddaab1fccaa
122 F20101119_AACCCT sides_c_Page_003.txt
6ce79f099fbe41b84c54f6cfdd1d7afc
199220b401fd35727d676728af74c4723e803e55
24186 F20101119_AACCDJ sides_c_Page_101.QC.jpg
f5cd254c8852b53ecae1b3b26dce0fb1
007401ba4a9b2d040c91ca2df8309407c5268903
F20101119_AACCCU sides_c_Page_059.tif
b497336f48f85a02647b43cb87361683
a82abac12b3a12354ed161e558957295f0ce3474
2176 F20101119_AACCDK sides_c_Page_008thm.jpg
abfeddff5397f396bc0780aef3db1a14
6d186ff7f3dfe8da7bd6f1f84029b6854052cfbf
65955 F20101119_AACCCV sides_c_Page_079.jp2
e8dcd7d16a5d549c7dc371fcfc6508d2
1baf2d282a4f049671f962a0e72d52308212cceb
2328 F20101119_AACCDL sides_c_Page_001thm.jpg
2645bc2891dd5edcd5ef0e08d7fa4714
e2e687d2db365d3e29b891ed5b92ca6c6da5cf49
F20101119_AACCCW sides_c_Page_004.tif
e5b48f0cf2e908b4d3ffbbdb2625b1ec
7ac036c7780c1a4c14a825c4be57fbfc57237655
1597 F20101119_AACCDM sides_c_Page_099.txt
4098becdebccab501cdc42690b9ff9df
6c7a42d39ec58f5bcb405640a6fd9440b22d383e
6932 F20101119_AACCCX sides_c_Page_003.jp2
acec96bcad2f154bb164cdee45c04947
75d5e0da2b007708003e6586cb00f8df53802730
2946 F20101119_AACCEA sides_c_Page_007thm.jpg
586833ed61ddcd3fd619f1b8e87c1d61
65f239b165edb2f84693ec0da3ba1fc656a6c25d
67859 F20101119_AACCDN sides_c_Page_059.jpg
7a5dbf6ef5de70bc3a225c7549cc7a43
731249838d28ebe2256ac7062df9c4768f35cfe8
F20101119_AACCCY sides_c_Page_046.tif
8c328ebe5211ac49302a1e01f2bd998d
0ea6f2fe8fb8ff7a99f73636479cfe245583d0ed
55383 F20101119_AACCEB sides_c_Page_093.jpg
e7cbf79d14007e6dfd41ecf0b5612593
89e7ad6d6d2d67def60ef67bcad96df28a4dae46
1340 F20101119_AACCDO sides_c_Page_042.txt
a85a26ffa759907296849958276a6f79
6826418017cd3e3de27ade7f56b57594768ba772
15473 F20101119_AACCCZ sides_c_Page_039.QC.jpg
7d92c363607a97df592e3bac6afbd363
238a56a1c23e34b1b5b8318c42d0cead00e3f20a
1500 F20101119_AACCEC sides_c_Page_097.txt
08663cb9dbfb677691d21aefc043b3fe
81817560b385e20df5558d2905c0584ae61d3ea7
5298 F20101119_AACCED sides_c_Page_004thm.jpg
ea9ffb7996f2f4e79f3880618ec13860
4bacc7bdc3f23cd72d37249a8261c9e0d204e855
656589 F20101119_AACCDP sides_c_Page_063.jp2
5774421a21aae5272442c33280a85596
b4e72e91da87f575d42790973cdfdca8d81ef7b9
44078 F20101119_AACCEE sides_c_Page_068.jpg
27f5e05b2c2b1fb7974b74fd4819a4c5
6d6691af6118d0e9559456915603738277977fa0
113474 F20101119_AACCDQ sides_c_Page_035.jp2
dd987d83e5b47ff1c2764189ed275bf2
4a1549e31119bd00ec47ba4827312c6f18e6dbde
68809 F20101119_AACCEF sides_c_Page_016.jpg
47714b7eed78ec49fa48923d02d60fe1
6c7781132b0198cb85bfe97547731aaa7e55c558
72922 F20101119_AACCDR sides_c_Page_049.jpg
650c1572b6c638be3405ddf81a0fbf7a
bd340b462de191df7b9b2d9f24caa2e7397b959d
23584 F20101119_AACCEG sides_c_Page_011.QC.jpg
9f1d901d36f72dadf4d1d87763bf4676
3a447a1a33d750c822be8ffa71713c4126658020
1812 F20101119_AACCDS sides_c_Page_017.txt
f86bbf8fa75612bb4e9dc6b71e25b014
783248b82dc4cd691789315ec4052bec1b5801fb
30306 F20101119_AACCEH sides_c_Page_089.pro
9f34f911ac287471205d7584063c138f
f29e9471c5db0125e22eaef4f8d011c6594a6279
F20101119_AACCDT sides_c_Page_008.tif
0439af6b249ed315dea9bba6b11a9c69
7ea3afe216ccbe8a6074c9811965e46ca0efd01e
713977 F20101119_AACCEI sides_c_Page_039.jp2
ae3fddbc3e1c608ed7ef56c82afac309
f1aa1e8e05d6c894bc3118cfd86dc02caa750069
1945 F20101119_AACCDU sides_c_Page_087.txt
bc321ae5595661c449b4ce6df36136b6
fe584dbc406c83e70c5b7b77cb8ba9104a6b33f0
F20101119_AACCEJ sides_c_Page_038.tif
0d35e8e0684a9e8fc31ffdd3cb234697
a337447a8ba1cfc6cf6e3784a7729791211f8c37
42932 F20101119_AACCDV sides_c_Page_107.pro
33c9407120405d5531dc65965ca3b532
1cf0b0fa1f8e87eefa20046424bc22bc3e825945
29790 F20101119_AACCEK sides_c_Page_094.pro
60a032cc0dfef7665228a7b5eb075ce7
f730bbc1b7a371f92c86fdcf5ef9e844143f14d0
1576 F20101119_AACCDW sides_c_Page_106.txt
01b65f440d1db486d0459d41bfa6ba89
220be83342184ef8bdf8f115561d3647f94ac413
1478 F20101119_AACCEL sides_c_Page_063.txt
3b59cb9a601f30f6d3eeb03c5c5b9157
1cfc1c63fe5e1c9a89d83bfd25af09d19e237567
5246 F20101119_AACCDX sides_c_Page_065thm.jpg
046bcdc12629fa8d266b77d60d5d918d
97b5f3d8a7f8641d7412387a2b84cbe1c9f57a81
16605 F20101119_AACCFA sides_c_Page_061.pro
3485fe9052118db6fa976426278d299c
ef7f1d220b1f714ff6e55e0c989a1f262ecff8ce
59220 F20101119_AACCEM sides_c_Page_099.jpg
065a7c0f2b750b9d1db76cf10f7065b0
2b3697bcbdbabeb324f0e0a466b0d28167a7431b
6059 F20101119_AACCDY sides_c_Page_027thm.jpg
5e5b4af2326beafc614447c7dbdc327f
55c8709e8d745cd38fe95d4185425314f1e75798
F20101119_AACCFB sides_c_Page_013.tif
cfde3f30c54e94c7feb302665f83878f
f37d09dc4c267af66013d57585b1e7eae0327adf
6297 F20101119_AACCEN sides_c_Page_033thm.jpg
e421cd35c2971783f5d74bb6fba25091
2147000d107e23f81f6d62f7f1045fa4e11036da
F20101119_AACCDZ sides_c_Page_056.tif
58386f7fb4030d43f9a35868719cd595
9b6fdd9a05ef512b7229e86c58a52a3d903bac1b
F20101119_AACCFC sides_c_Page_080.tif
b7b00c34227750dd05f7ef4f742325f5
54574709d1daf857c27c89cc242897c85dc4b4a5
53397 F20101119_AACCEO sides_c_Page_052.jpg
7ebf689eeb5ebe073c8c472cf8a82161
5e9d2f0bab3a6f89e72f0c6d84ba130458ca8fbb
9964 F20101119_AACCFD sides_c_Page_007.QC.jpg
277cce2ed61cc2df334a22900d2e80b5
2109ae8b75365190683c3af074f67a1971472e04
96173 F20101119_AACCEP sides_c_Page_036.jp2
aefebd0f8cd6ef9095b56fb8a7848c88
d0645d5ef3ec51bf0bf5d9dd46cb144178e83019
60753 F20101119_AACCFE sides_c_Page_051.jpg
0c910a44afe1b6d32025d9b9941a1783
3f9378364fcf76308fb7f2bc580bf1c9c1c6dc91
F20101119_AACCFF sides_c_Page_050.txt
7a1a82c140afbb81b33695a1b74af873
63bcdd8ff2f0743a45cf2425250c1eb8a5cb3b9d
21843 F20101119_AACCEQ sides_c_Page_027.QC.jpg
4a295d8e4c0a7de3da395af6f4714d59
577028a71e2ccc50581f70aec0c32b3dd8c95671
1051941 F20101119_AACCFG sides_c_Page_044.jp2
d7ecc099896586d24296e41bc0c04daf
aac94c378fd44c7725fd9b392ea8eceef1505038
20200 F20101119_AACCER sides_c_Page_095.QC.jpg
cbcb23d27bd42c71cceb1728bb6d2eb6
3949324c7662ddbbfaae157f2626253da7648d94
40723 F20101119_AACCFH sides_c_Page_017.pro
55df1452d67063095b1a7c3fb844df3c
580c1e0c9730c2bac2aa2b7394bc4a4fcd1329a5
38959 F20101119_AACCES sides_c_Page_112.jp2
c5b0fe6468162fb5ef0af8c4b57b8025
a97c48589e9a8307ce0834e527aec530cb1307da
3560 F20101119_AACCFI sides_c_Page_002.QC.jpg
958808f79f5f6d131d1ed70e515d01f2
1b8b49c98e8a60200248aa81192f417e35293d9f
6502 F20101119_AACCET sides_c_Page_031thm.jpg
1936d1adf1e14abc6639450af0d9b9f0
c3588a17b7f0c6c00904bfcf369a931177a63c3f
6281 F20101119_AACCFJ sides_c_Page_025thm.jpg
c1bbf890e03d3ff2e62761685fa74682
0caaa8197e3f398637e97eb39013773a490d6227
1947 F20101119_AACCEU sides_c_Page_104.txt
d30b6ce92161ec764b0e004542d8d944
c5bda0569573a815ac5537039826fc7c80964cb1
F20101119_AACCFK sides_c_Page_061.tif
6453621da559289484d78456b3b13780
4cd44feab7ef3ae56b76b180366114041fff5ab9
4941 F20101119_AACCEV sides_c_Page_067thm.jpg
34eb1e0ceb96c081c16b9b48d2fc888a
e8ea9b1cf56be2692a01ae7d82381cdfef76e555
32395 F20101119_AACCFL sides_c_Page_078.pro
4b1bb404ece0fb3003d92ff9a3e0e436
caa1035d6bc386797cd434cb17bbe30d48a31af6
5823 F20101119_AACCEW sides_c_Page_086thm.jpg
dbb1e05328a15285187deb3e3d0b4cb4
9517d951615c7994729bb31224a737e09af07e5d
795377 F20101119_AACCGA sides_c_Page_052.jp2
1a4630b4a91d0f66d5cf96ed9d6c3569
af80dbb4e8459fd48df8753f331a2bef142254f2
6035 F20101119_AACCFM sides_c_Page_088thm.jpg
9a03b679a52ba9fbc81bfc525055ee6a
15d70f636d02aceef6b2fb541604e5b09db44142
55411 F20101119_AACCEX sides_c_Page_097.jpg
299e15be59ccb3e48700dd73f0a19626
bd351d86c1696eeb1a616299e68429ef6a6a9c13
21095 F20101119_AACCGB sides_c_Page_092.QC.jpg
0a6ff50817ca80d9c56aa523aa2fa21c
0b75b0d21eee0d5bda4742d4dad11f25bac779e8
15074 F20101119_AACCFN sides_c_Page_046.QC.jpg
2148f2aa41a4ed804f5983ff7b7cf722
edc6d5de68508cc8bb152c95847450c761d2cdb4
20898 F20101119_AACCEY sides_c_Page_047.QC.jpg
b0f68545f4ebbf2413b51eadc04fe507
0c19fcb9a4cc02601808aa5b78361404f957dc02
42562 F20101119_AACCGC sides_c_Page_072.pro
764aa4328f807cc4aca5609f34ebf6e8
c93775e8050bd269b6ec5c6ee4b5eaa7fa72c3a0
428 F20101119_AACCFO sides_c_Page_059.txt
8f0d27111fff3db0db4422f44d93b6f0
e57638ff4c71e7fd0099b5a0b7a7cd14842ed419
F20101119_AACCEZ sides_c_Page_106.tif
8a84fbb411c67ddd6f559835ca3bcb2e
e23f3f89050e8137cd1ef3bad8cf8640999f5723
735963 F20101119_AACCGD sides_c_Page_096.jp2
2bcdce929e7ccc751b8d2210a073fe4b
9163979fa89c23ad58702ce67dfa35210b32bb4e
73367 F20101119_AACCFP sides_c_Page_010.pro
d5ae671795a86ef398936083158b8d13
ac82e454593c34a00ca15a372cfda39d5e2e3f7e
1000907 F20101119_AACCGE sides_c_Page_019.jp2
cf945ff486c10366ed15246d7997f4cc
2ee0b518da0c7427a5797a8809066a3d0631d845
1782 F20101119_AACCFQ sides_c_Page_047.txt
a2ba3d6dd5df05a3768b29ecda8457f2
4f3e76d2edb6f020c2fb94f86b04d0de41f33ce5
46743 F20101119_AACBZW sides_c_Page_092.pro
bbe35f63ea158b563f34c9af680e933e
03ebf58efbb5c901f07ae42b6c7850a6a14b34da
F20101119_AACCGF sides_c_Page_024.tif
454d9e8876d98fefb16e6c005ded06ef
224de18c8c292d298e0a9c59a8e76e282bab3b00
977 F20101119_AACBZX sides_c_Page_098.txt
d7629e4e21c20b0644fd59d5f93169d0
451f12e30860e5d48ca71ece1094b9693e3aa58d
F20101119_AACCGG sides_c_Page_087thm.jpg
d5e2e44abb66356fce2ec290a5d53cc8
0496ffa429d9569db05dbb81615a25390bf5e110
51682 F20101119_AACCFR sides_c_Page_049.pro
15acbe7a0c8ad289b7c7ac47564fb3ad
e51b0ba6f849072cb87bff6cbdbeedb1d112eba0
1685 F20101119_AACCGH sides_c_Page_095.txt
35ca4257f6743cacd81f7dd58d2c489a
ffe2901f77e9bf7d374c0b9bc99aad918eeae98b
48082 F20101119_AACCFS sides_c_Page_041.pro
0d5435becfc83a4d3acf8c9f0a673647
75b13b5225fb068bea66dc20770cc72025da3e9b
1043494 F20101119_AACBZY sides_c_Page_043.jp2
5e93d9c696bb2f098166b8af41593114
57d4ab869828bdcd4e4f7cb66a4add5719c436fc
86898 F20101119_AACCGI sides_c_Page_099.jp2
a3fe88db154af05246c3541984e1ef12
9168d285c6a8eb054867f9fd9ae58aac01a94600
70827 F20101119_AACCFT sides_c_Page_108.jpg
15377f8ebae2c8d31249a90fa054941e
ca8f84385135ebb0c3d864e852c7c0d9a69f4452
19807 F20101119_AACBZZ sides_c_Page_076.QC.jpg
937b8149e977b6d5d54f412cb25f475e
a0c2f280d92a898049ec8dde41b911ecd12abd3a
71165 F20101119_AACCGJ sides_c_Page_087.jpg
bb674a4415f5a416abd575119ffd0f3e
220c317dc66a6a1904661930d14a2589550667c2
F20101119_AACCFU sides_c_Page_001.tif
cb0601c2618f9c506897ac68fed79251
864b4ece3630f31238b9e973eed98a3d182628be
57062 F20101119_AACCGK sides_c_Page_013.jpg
b062a454abb0f751b637d86cb2515eaf
b848238f1e474c1a6f6c5f1319b8bf4d3a75546a
2006 F20101119_AACCFV sides_c_Page_060.txt
6c191c8c1c54bc806ae09b083cfd0bbb
926b10c0b9869c9296ace1df959869fa6972efc5
3944 F20101119_AACCGL sides_c_Page_083thm.jpg
5d143e0e7a34dde20de01f579c11ccb5
2139c66dea36bddab2d86564e5edfc2c23b66577
F20101119_AACCFW sides_c_Page_069.tif
35f8282734b8b326f73a13982de76154
e622dbc296eca50a41196bce7fadf2e2826c106e
23318 F20101119_AACCGM sides_c_Page_060.QC.jpg
50ab42b10f5775957e013b1357d9591a
afbe4f51ee22a3ffc31b0e15b7fa9d167011a2dd
F20101119_AACCFX sides_c_Page_058.tif
e79310bc01c923cda566622f1b3965ba
b20179592994f0bf2d1baa402e2501b783c41c2a
2108 F20101119_AACCHA sides_c_Page_035.txt
68d8373e697d8a102517939d192b12fd
0ab38f8347f0b4f8179cab90c813a207a5cd14c6
1829 F20101119_AACCGN sides_c_Page_062.txt
5a08639491f6a34970b2892ffe14a8fc
7187d96d90ae8995986c844833c139cb9555ee87
F20101119_AACCFY sides_c_Page_048.tif
b3280abe83e0f8dfee29cbb9edcb3ba1
f3220b4758512947f5f048b90a931d97ae11b65b
17025 F20101119_AACCHB sides_c_Page_067.QC.jpg
ed03837032279843e1df5ea4d5723c6a
9184c1df6257b6962a4409fabdb577fbafc6efb5
29984 F20101119_AACCGO sides_c_Page_080.pro
93d153ac427a56ab2625cabf7d12c420
fd32a831d96b5536f18ab463c0f7e93ce2565b5b
60957 F20101119_AACCFZ sides_c_Page_034.jpg
fc90095346a27678995378c4e8c13b21
6809f9cf3ead7809cc99e6d393cc348719576713
4239 F20101119_AACCHC sides_c_Page_098thm.jpg
fdb769bc760b556a0df5de66eaa4147d
d72c0e4dde3fba3fb0115fd34bee97d1124deddd
53742 F20101119_AACCGP sides_c_Page_071.jp2
86b346b944830b76c4f9c0b2405b6e9b
22199799e2f3d59241030f7d52ef0019fce8ad83
873118 F20101119_AACCHD sides_c_Page_017.jp2
4e357e06bb3fe5bbd32a3f35af096c2b
f81f07648949dbbd5c5f0d7fea9cf1ffc4766a3e
3428 F20101119_AACCGQ sides_c_Page_005.txt
ba6292bd0b126c2c6ab7240e79781fc7
722cd3bb010dc200ccceb6e412d7ffd837eedcc0
5605 F20101119_AACCHE sides_c_Page_093thm.jpg
6cb52ce323eb8685f7d3a5d8c9cfd8fa
6c9287c659312b942d38d1464c233390e616cca2
70471 F20101119_AACCGR sides_c_Page_057.jpg
76d848c8cbe02797acb771742417e510
992d3de4a50044893e32dcbf7e3e87a0e4a8cfd7
22512 F20101119_AACCHF sides_c_Page_001.jpg
8c50841f59b435eecf04a596f3171612
7f289aa338d6d8672194b84cb37fb73440d6534b
51585 F20101119_AACCHG sides_c_Page_022.pro
6c785ef890d4691b6d22a5dbba6d37bb
9457188259c69c5b26663f2221d6ce981be78033
F20101119_AACCGS sides_c_Page_002thm.jpg
8a05a4faba122da76c7457f654aff056
1cf518d78c7b92c1ca2406c2a4ea3248e4a7f6eb
F20101119_AACCHH sides_c_Page_021.tif
a9780e1a389d32c48ffe3f25fc5dace9
51315793e296f0c6ca47b4141982dac16de160e0
F20101119_AACCGT sides_c_Page_085.tif
6f9b4fd08811ec0435e5f585d8d40bfe
e69aedb0fd821482ca69cc8d90b7fd6b2c0ca9d6
86374 F20101119_AACCHI sides_c_Page_051.jp2
52b9fd764f326cac0d135817f47912f4
36ed6feaab6f0791a9b743d4d53325d3e42776b5
5921 F20101119_AACCGU sides_c_Page_055thm.jpg
9a34b98f2890cf903532aff36face72e
e7c1ca1c00b4c3ed4ab4e96bc7e78a54b943242e
90577 F20101119_AACCHJ sides_c_Page_082.jp2
febe1acc76fb8ec113a89cf9c20c3b09
1a0cb3b86a337c010562c51456cce968797cfb28
68363 F20101119_AACCHK sides_c_Page_033.jpg
94ef5de74685a3e1d775d3403ee936b6
99efced4657f8952a3661365d858e4e7f546154d
63737 F20101119_AACCGV sides_c_Page_086.jpg
d8f13de8cdd7c4cebd8429022cfaaf64
23b509af9e7c3d82e1689e3fb47bbdcf4bd0115f
74827 F20101119_AACCHL sides_c_Page_035.jpg
2fbcfb423ca88654ab4b4c1612ad63a1
daa6fea130a5c8d98214afa30450d615c99eef2d
16358 F20101119_AACCGW sides_c_Page_064.QC.jpg
c612f68c4d2aae7f2765b25473107161
3ce858453cf4fbef43bc49aabb57acea576bc5cf
60865 F20101119_AACCIA sides_c_Page_014.jpg
2c9669dcee989bea6399ca4f5347c132
9911a59ff70ffc5d07560d2c88f5b1f5cbbe8d5b
167482 F20101119_AACCHM UFE0012360_00001.xml
319dd4cf1ddf32fba5e623858e1ea3cd
ad6ed32019fc4a3b009005a0019067dc8735b445
5623 F20101119_AACCGX sides_c_Page_072thm.jpg
84df2c001e4ba46cd31ec3567d3484e1
44e978229fa303f6c2d6644a1b6c907cff1ee0a4
64552 F20101119_AACCIB sides_c_Page_017.jpg
3a857d62182f195b20755bfe0b0ed576
30c9c322002851b913171699cffefb6c4b3c16e7
19653 F20101119_AACCGY sides_c_Page_014.QC.jpg
c1beb6b0798bdbd57a844750f7cd091a
32d5bd6d9546a281234e21d76caead37562abe0f
68602 F20101119_AACCIC sides_c_Page_018.jpg
bd619639602d873446e80cb440189dfa
9ede1cc4946c801d5090335cc1d137e1fed8a008
42745 F20101119_AACCGZ sides_c_Page_098.jpg
d03056d823f214f6dc05d96218e0a5d7
3ac0d2915ba25b5de515ed63f8dd1b23a936669f
60679 F20101119_AACCID sides_c_Page_019.jpg
f33ac846d70c0d9225e184c409ad40e4
94d2f015c5a79edd5b08a034750734bae5955561
10876 F20101119_AACCHP sides_c_Page_002.jpg
3ec1ef12b1f5016712b3980249fdc76d
8ea3aae9b0834a538f74ab64a9bbd22685add9f4
72356 F20101119_AACCIE sides_c_Page_020.jpg
a2748093e593aef2987f2df429916a3f
6f828b7ebb45f9dc87d944921497d1257ad8fce5
11113 F20101119_AACCHQ sides_c_Page_003.jpg
4e914ad9b033167f0cfc5d7f5dd83bde
12bfa1ba2f8b9de44f662b44e287124b1542db48
72126 F20101119_AACCIF sides_c_Page_022.jpg
b98d6ae7ff8cc18d2d667c47449fb4da
25f62b9872aa7392e7c5bb0e4edd5aaa9d550194
57291 F20101119_AACCHR sides_c_Page_004.jpg
18c238763e6f6762a76c2d31b544fd55
260844dc8e6073f4dd8cfcfe8f869cf90403b747
67532 F20101119_AACCIG sides_c_Page_023.jpg
10cb722aee6e4ef178868c71f12ba2e5
42b6104ea08f1ae3017a5dcfe1a31cf49d202128
75369 F20101119_AACCHS sides_c_Page_005.jpg
cb51ba2a30b1b82f78f7da449ecb8a84
4f584117386973c9e745db9ad98794f869c10280
65788 F20101119_AACCIH sides_c_Page_024.jpg
99849a493cd814399141d572608828eb
6379d9149cfbda8f3b67c275aafb59da5c763dc2
69518 F20101119_AACCII sides_c_Page_025.jpg
c17ddcaf6c21473830c895f64baa6d6f
1f907a780d7620eabf10202a660dba500a1038ba
98817 F20101119_AACCHT sides_c_Page_006.jpg
e345f4249e011a2ff900b512d63e6979
495e1f98f7b2953c8d47462e1cd7818e794422a2
67024 F20101119_AACCIJ sides_c_Page_027.jpg
454d354d8697c8ead97178a35a674ba3
4a034f1f975fa667e7d767a0c3f599e7518acd6f
37836 F20101119_AACCHU sides_c_Page_007.jpg
a9e1616026467917efc4693368c6dd17
fc8101c844763f54338f45f004048562825f4ba8
70022 F20101119_AACCIK sides_c_Page_028.jpg
22ddb128c92b218f631fc2e75a1ad915
9c3e25a05c25607669cc801f41c45bbb3ec10caf
19798 F20101119_AACCHV sides_c_Page_008.jpg
6ef14266cdd233df3f106c9daed2ee0f
46d474861157dfea7ce3b79c7e944d8451905b12
65386 F20101119_AACCIL sides_c_Page_029.jpg
4aaf0f9605efd73d7ceece26ce1732cb
fec4a372d5488b439ff968fbcfa0d8f8628f635c
88674 F20101119_AACCHW sides_c_Page_009.jpg
f92d4d6b01528063b9e78afc63faef07
d0d721d67215d817e12c813b381254b4c6ba99aa
73723 F20101119_AACCIM sides_c_Page_030.jpg
937e13fa0207ab02e6aa661ebc5bfc98
0af7e06061901721e117764de2b9388658c9a572
109574 F20101119_AACCHX sides_c_Page_010.jpg
6253cb433f0e5308a8b0c1af64336277
8f3ef2eab1143596ec2402e5fa0adcfe1cd518ba
30634 F20101119_AACCJA sides_c_Page_048.jpg
1b1aec4aca6f55f5aefadf8e7cecaf3a
076454f621b0b55c0c6a8d576b3de04413a7c716
70574 F20101119_AACCIN sides_c_Page_031.jpg
aaea8a38c17e63994d31c982b46695e1
2bad1886651ea81b3e7c9419fd09fe5be46ff1ab
85417 F20101119_AACCHY sides_c_Page_011.jpg
6683d11c64feff3466ae3490854bf96a
ddc5c96b1253f74b85412a54f2c16727d132ac9f
72851 F20101119_AACCJB sides_c_Page_050.jpg
e136b4bdb84b6a0caac6a47991596033
ba3331750da4ea11572b94ae932064234d7bb3b1
71102 F20101119_AACCIO sides_c_Page_032.jpg
8a5e50720598ca285e42d09952286aba
e267482ed2c72c028155556b88996678662bbcef
57893 F20101119_AACCHZ sides_c_Page_012.jpg
63a7c232f6da5c608b8cf3502f081904
16dc792431fd32a1d6dd6944d8f07569c7d55af3
69607 F20101119_AACCJC sides_c_Page_053.jpg
3b0dcd008538e6460c946a6a5db3757a
fe98b6bfa1480ad2fa9eaa6e5ca218532455cde4
64686 F20101119_AACCIP sides_c_Page_036.jpg
cd9b2b4c1b6bf46a2aefebf6af0fdc2e
05ac6f134c618488ae147d5e77e9dda23f028470
42630 F20101119_AACCJD sides_c_Page_054.jpg
c10a4641f294e3591e71001481e05322
98dd8cdf08046d069722e951552f44afd78ccc71
55451 F20101119_AACCIQ sides_c_Page_037.jpg
1284991109373248a4e116c88676cad7
2e1f992d196f7e67e7f3753e3f3cac3b8f82f69a
67922 F20101119_AACCJE sides_c_Page_056.jpg
0b21c78755f15d2d915a683079daffcb
bb851cb5ccad20d9c666a3d4f69ce505b0cbf04a
64257 F20101119_AACCIR sides_c_Page_038.jpg
04ebb4c41c83aaac9c24b7ba6fe17420
1314c075700eccb6693b0fd3f732de82e8042818
63941 F20101119_AACCJF sides_c_Page_058.jpg
07537fd0fdcb7c767ecdd901a91e4b02
32ae28fd36c1e98a0342eff77b9b6c7b50d97406
72054 F20101119_AACCIS sides_c_Page_040.jpg
638035f81e1d1dff8ccc68d831091d8a
83a8b98e990979a737aa093476997ca3498603ff
71196 F20101119_AACCJG sides_c_Page_060.jpg
878e5a9d53d97792625dac1f7f1e151e
60e982e85a761098fe4bbf3d614f5f30aaed95d2
69433 F20101119_AACCIT sides_c_Page_041.jpg
cc1187711710602e8997a3d34c802c98
44e35383a826965ef94334a70c6ba472bc9c97a3
33567 F20101119_AACCJH sides_c_Page_061.jpg
dac5639892242ea291eec57655393f3b
1c9a344b1d1ab1d3645d28f998e4b64ccb20b8f1
58454 F20101119_AACCJI sides_c_Page_062.jpg
625d693700bfc5467d0b7feaa535d975
b2e97b44562dde9a30de4b2ebb61caea0509451f
54742 F20101119_AACCIU sides_c_Page_042.jpg
ac608d41db241b5322fa9c8f9d8235e7
17b470fdf0ba1425c0432f3a8705e81b6f92526c
51253 F20101119_AACCJJ sides_c_Page_063.jpg
8ef28dfa3347972e52a614b527637bf4
458c7bd96118395291d2b9b4d193a88fb6a61089
68534 F20101119_AACCIV sides_c_Page_043.jpg
ef509c9c2a3c58adf58d2be2e3d4ac30
4c28401afb7f0f270609c02da9c27a325d8e0dad
51256 F20101119_AACCJK sides_c_Page_064.jpg
e654c29bd4721d6b50a4fb5c8cbd871a
eb10b140337c404e2760f1057c6c30a63c04b629
58156 F20101119_AACCIW sides_c_Page_044.jpg
325d8b66923b852c8b6e517b40e28de6
0989432fad2c048b24bc513625379489e75d3462
53313 F20101119_AACCJL sides_c_Page_065.jpg
82fd0bd4a4429de0af03eec1138567ff
1dd7283ac1b3c4e6a4b3963f30dc6e3c686ae9a7
70566 F20101119_AACCIX sides_c_Page_045.jpg
3becec200e36e9fb915213835b01d73e
d3cf6cc7345549c7c3c3b9140c42236902165265
38114 F20101119_AACCKA sides_c_Page_083.jpg
07a85311ccd0676c25693ff636bfacdb
73fa49a22403638c25f897f92a34babdb7bb9b5f
53553 F20101119_AACCJM sides_c_Page_066.jpg
58308f0b075989bcfc07e1c6dab277dc
f0606967b066dd7ea9afd2dbdf9e4ca956bf4768
45969 F20101119_AACCIY sides_c_Page_046.jpg
65b028e677a547219879f77d4ba673eb
bc036f6ec73aae2b69ac89cdb303ff1ad350b180
31040 F20101119_AACCKB sides_c_Page_084.jpg
a2c9faa23072971cf3fea87801cf710c
7c34ddb7545188a5291f204e9d295d5daee10e10
53500 F20101119_AACCJN sides_c_Page_067.jpg
819ad2f04453e0b98893ff19496e71d3
1e7af2292b46df6511a85a347cc510a9e7bc99de
63741 F20101119_AACCIZ sides_c_Page_047.jpg
83138960c613973eca8b67a186ba5b6b
fdd45df7ed7fb5b9b69c948d488f3090d02d9765
52193 F20101119_AACCKC sides_c_Page_085.jpg
a6ba82ee69988b5d5f4881e0101b4ac4
44c08b3d3daaa16d7a91847952660ab0bed8d9db
35825 F20101119_AACCJO sides_c_Page_069.jpg
f482af2a63834225aa1de0f96c70144b
4c7f48dc2491a94de74b22d505edeec01d52fde5
56429 F20101119_AACCKD sides_c_Page_089.jpg
416d63e500d285d9ac0c35aab8dc1b35
8907ca226a0d9a530be81b1271f19b495d885116
67626 F20101119_AACCJP sides_c_Page_070.jpg
7f083de73cbb4c9d1a079748cef239c6
03de08e27e733c4cc91ae3f63e450a55db51ab1f
60812 F20101119_AACCKE sides_c_Page_090.jpg
64d5db0801cf5bbff3acf97b1e75cee5
b8fd413cdd1f7f70f9f979d97b07085ecb5b51f9
38742 F20101119_AACCJQ sides_c_Page_071.jpg
229bfd51abe2856232f3b73a985210be
852223e1711eb536d2128e02fbaee2c2b07a02a5
56142 F20101119_AACCKF sides_c_Page_091.jpg
d053d48b22849b85920a7f4bba6eae27
b1c4ea3b5e38032cffc5d6bb304bd38c86661a92
63094 F20101119_AACCJR sides_c_Page_072.jpg
270154a431eb5e8258f5c89683daf608
a4ac3bfa246032df2224ee029fc956e859687878
66398 F20101119_AACCKG sides_c_Page_092.jpg
6afa66a6841c29d04acc70a847c8d5c3
aecc3047cfac53060d51c7b28b21f609a6de3cbc
67475 F20101119_AACCJS sides_c_Page_073.jpg
b58692b0f9af2eee9f0570fe56c9bc45
7440e9de3d96be90479baec0d7a519f8978287d1
53349 F20101119_AACCKH sides_c_Page_094.jpg
3012cfa582973718c5f605b4c2015171
b01f5dfd1eb27bae3daff031eda4db0e8d63a9dc
69946 F20101119_AACCJT sides_c_Page_074.jpg
035f4b6568bc7902e1aa6282706be5bb
7f954857ca06ac91d8ced27c06335b9321bcaa5c
62795 F20101119_AACCKI sides_c_Page_095.jpg
747fe45ffb437f045e828031b46172bc
00d1302f8059f192d8f69118ce0303da75e0036e
56548 F20101119_AACCJU sides_c_Page_075.jpg
14dd48b42c1a42beb98f381c23ea5527
48311055e5e3f1b36bc4dc09daca7dcd6a91a182
47830 F20101119_AACCKJ sides_c_Page_096.jpg
e82f4ab0f5cb75e66bed6d359e8205b6
f528587db3a39d346a9815f56725d4f0058da251
58212 F20101119_AACCKK sides_c_Page_100.jpg
185c2cad78e7ae3c35997ed4adad00bf
cfe93543210890c3561040ff4ba49a7d2a33f394
61057 F20101119_AACCJV sides_c_Page_076.jpg
8532b7bb1bcf7b2fbe52014a5109e3a8
c0c1483a2dcea992880c4feaa7bd83411eac0c8b
74301 F20101119_AACCKL sides_c_Page_101.jpg
88955aa319c1435ebb939d86def077ae
eea7883c45c4d210b318c7bee1e4412dfe0561fe
57398 F20101119_AACCJW sides_c_Page_078.jpg
fabba2f873187b06e5351dfc4e9047e1
f78d5ac572bfe1d82a6911fef27a96e06a95f7eb
376762 F20101119_AACCLA sides_c_Page_008.jp2
3782574ad3f4df6596a334be9360ec60
1647a3d3f305b93cbc7db9f5f277b8fa5a361a7f
73309 F20101119_AACCKM sides_c_Page_102.jpg
c3d2710815777b2cc1ba84cb268d63c8
987bcd481de4022f7e74a609530e21314bc43ad5
45390 F20101119_AACCJX sides_c_Page_079.jpg
872d0dff24fb40bcad69aa228fa6f9b4
c4c5266f7eddc7eb469675ae7835830e68797d33
1051942 F20101119_AACCLB sides_c_Page_009.jp2
6e1c190e761c41e7512aa5db3814defc
586095726d806254301aebbf2adb2523fef0ef08
70158 F20101119_AACCKN sides_c_Page_104.jpg
65476ac9d9e9963bfa9446463c1bca47
3a962f96ff915e0ab66bce3bb8bad0e39f6069e8
58833 F20101119_AACCJY sides_c_Page_080.jpg
ceed8a673a4cc6d44d6e71f1433c7a3a
4729187a1936a4110fc12d6faeb2d53f8fdd8ef9
1051973 F20101119_AACCLC sides_c_Page_010.jp2
c632397928325f3e198aea3ee9d9a8be
b3c7d9636dbd217579f1168362696bff8edc3628
58382 F20101119_AACCKO sides_c_Page_106.jpg
bf1e7a3630a9ccc5e99a31005eaee293
c72ff75ff79ebc552bff60226ab878dff6ed4ba0
58870 F20101119_AACCJZ sides_c_Page_082.jpg
e31de64aa3307c7215806342971a0cf4
cdab6b34035f82268bfe18afa93f4898fb4f516e
1051984 F20101119_AACCLD sides_c_Page_011.jp2
2ad06330a83b0ed11c4743b5063705a9
233bd5932742489d159332e1929215f3057265e2
62505 F20101119_AACCKP sides_c_Page_107.jpg
81f858145124187cc27f2fff42e7e702
43133af486d240bca5decc5660c46e8f66016d76
85168 F20101119_AACCLE sides_c_Page_012.jp2
4797102b5577ef106866df2059702787
719fc65b861428082ee1e1d286a563419eadaa65
68948 F20101119_AACCKQ sides_c_Page_109.jpg
e1752195f994f17ddf24999667953f66
869a09125a2c51d3ea164364ecd35e0cca2d7aaa
85360 F20101119_AACCLF sides_c_Page_013.jp2
8ea04770e14eb46827c6e6a7fa51a0ff
6e1f95a90ecd27d81ebe089da10e22df8144e7f7
69668 F20101119_AACCKR sides_c_Page_110.jpg
2ec0f0bb95184ba8a6a99b8cdb013180
16bce8f5d13b9c3d2e2002aa9151d9b2818f9533
88485 F20101119_AACCLG sides_c_Page_014.jp2
8b6849f36d601f596eab931efbf52600
08670026d699bb3e3fd962c83390f15911439ab0
55562 F20101119_AACCKS sides_c_Page_111.jpg
750b75885ca1ca4ca95dd483caab61da
cb945689783ed2f92e884c5205bf0eb375ea75dc
79964 F20101119_AACCLH sides_c_Page_015.jp2
86370d681f1fd2e7e99f33b1b9436034
95563fc2ef339541eb36794a8e74c602f0ba66a0
30536 F20101119_AACCKT sides_c_Page_112.jpg
3eca0595c2d6b06bba3659b04520cf48
3d78cac41d1cf05ae97e84023f82314287d758b1
105385 F20101119_AACCLI sides_c_Page_018.jp2
cef27af1b8e6e99ff889808aeca2a454
6dbbe62915897b0305b550f4cff67a6b28fc240a
23553 F20101119_AACCKU sides_c_Page_001.jp2
a4895cba28cf5f7c1e3e2214417eb3e5
94f97b4aff96232d51f200aa46e2ac9c7dea01df
111355 F20101119_AACCLJ sides_c_Page_020.jp2
9751fcbd476da3bb2f15c26b8f4c765e
6ac3d8b6b3fe5d06b7b8a0477ad8de429a950594
6316 F20101119_AACCKV sides_c_Page_002.jp2
8198e9b1f98dea31e81c8a15b1cd5dc0
e63524697447d98f95cc58ad885a1da62dc28941



PAGE 1

NANOSCALE ENERGY STORAGE ELECTR ODES BY TEMPLATE-SYNTHESIS By CHARLES ROBERT SIDES 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 2005

PAGE 2

Copyright 2005 by CHARLES ROBERT SIDES

PAGE 3

This document is dedicated to my beloved wife and family.

PAGE 4

ACKNOWLEDGMENTS I would like to use this platform to expression sincere appreciation to my research advisor, Charles R. Martin, for his guidance through my graduate career. I believe that I leave with us equally pleased with our accomplishments. I would like to thank my parents, Charles R. Sides and Martha C. Sides; my sister Anna Dobbins, her husband Clint, and their two new children Bella and Chase; my younger brother Matthew and his wife Kelly; my youngest brother William, unmarried as of yet; my grandparents, Evelyn and Raymond Sides and Evelyn Campbell; and the most recent addition, my new wife, the former Alison S. Knefely. Each of these family members has motivated and encouraged me during each turn in this winding road. I had the distinct pleasure of working with Stephen E. Creager (undergraduate research advisor), Bruno Scrosati, and Fausto Croce (hosts during collaboration visit to Rome and Chieti, Italy). I enjoyed both the professional and personal interactions. I also thank the following members of the Martin group who advised me on experimental design, presentation style, and my golf swing: Dr. Naichao Li, Dr. Lane Baker, Dr. Punit Kohli, and Dr. C. Chad Harrell. There are a number of other individuals that I want to acknowledge for helping me to enjoy my nine years in college Wist, Riles, Grant, Wil, Baker, Buck, Elizabeth, Scott, Heather, Bruce and Butler. iv

PAGE 5

TABLE OF CONTENTS Page ACKNOWLEDGMENTS .................................................................................................iv LIST OF TABLES ...........................................................................................................viii LIST OF FIGURES ...........................................................................................................ix ABSTRACT ......................................................................................................................xii CHAPTER 1 INTRODUCTION AND TEMPLATE-SYNTHESIS..................................................1 Introduction...................................................................................................................1 Background on Li-ion Batteries....................................................................................3 Nanomaterials by Template-Synthesis.........................................................................5 Templates......................................................................................................................6 Track-etch Membranes..........................................................................................6 Anodized alumina..........................................................................................8 Colloidal-crystals...........................................................................................9 Material Deposition Strategies............................................................................10 Sol-gel..........................................................................................................10 Other deposition strategies of template-synthesis........................................12 Impact of Template-Synthesis to Nanoelectrochemsitry............................................13 Fundamental Electrochemical Investigations......................................................13 Gold nanoelectrode ensemble......................................................................13 Carbon nanotube membrane.........................................................................14 Li-ion Battery Nanoelectrodes............................................................................14 Polymeric track-etch templates....................................................................14 Nanosphere-templated structures.................................................................17 Li-ion Battery Electrode Fabrication Strategies.........................................................18 Commercial Process Method...............................................................................18 Template-Synthesis Method................................................................................18 An Electrochemical Primer.........................................................................................19 Comparing Electrochemical Methods.................................................................19 Comparison of Diffusion Regimes......................................................................21 Electrochemical Methods for Li-ion Battery Electrodes.....................................22 Chapter Summary.......................................................................................................24 v

PAGE 6

2 NANOSTRUCTURED ELECTRODES AND THE LOW-TEMPERATURE PERFORMANCE OF LI-ION BATTERIES.............................................................25 Introduction.................................................................................................................25 Electrode Synthesis.....................................................................................................27 Scanning Electron Microscopy...................................................................................30 Electrode Electrochemical Investigations...................................................................31 Constant Current Discharge Experiments...........................................................31 Comparison of Rate Capabilities.........................................................................34 Evaluation of S c ...................................................................................................38 Effect of Cycle Life.............................................................................................38 Effect of Electronic Conductivity........................................................................40 Chapter Summary.......................................................................................................40 3 EVALUATION OF SOLID-STATE DIFFUSION COEFFICIENT OF LI-IONS AT LOW TEMPERATURE.......................................................................................42 Introduction.................................................................................................................42 Electrode Synthesis.....................................................................................................43 Electrolyte Synthesis..................................................................................................44 Measurement of Diffusion Distance...........................................................................45 Electrochemical Investigations...................................................................................47 Potentiostatic Intermittent Titration Technique (PITT)......................................47 Determining D Li+ (Data analysis)........................................................................51 Activation Energy of Diffusion Process..............................................................53 Cyclic Voltammetry............................................................................................55 Chapter Summary.......................................................................................................57 4 A HIGH-RATE, NANOCOMPOSITE LIFEPO 4 /CARBON CATHODE.................59 Introduction.................................................................................................................59 Electrode Synthesis.....................................................................................................60 Structural Investigations.............................................................................................62 Scanning Electron Microscopy............................................................................62 X-ray Diffraction Studies (XRD)........................................................................65 X-ray Photoelectron Spectroscopy......................................................................65 Analysis of Carbon Content................................................................................67 Electrochemical Investigations...................................................................................68 Cyclic Voltammetry............................................................................................69 Rate Capabilities..................................................................................................69 Chapter Summary.......................................................................................................72 5 MAGNESIUM-ION INTERCALATION INTO TEMPLATE-SYNTHESIZED NANOSCALE ELECTRODES IN THE ABSENCE OF CARBON.........................73 Introduction.................................................................................................................73 Electrode Synthesis.....................................................................................................74 vi

PAGE 7

XRD Studies...............................................................................................................75 Electrochemical Investigations...................................................................................77 Ferrocene Pseduo-Reference...............................................................................78 Cyclic Voltammetry............................................................................................79 ITO-glass current collector...........................................................................79 Pt foil current collector.................................................................................81 Rate Capability....................................................................................................82 Investigations of a Possible Mg-Sn Alloy..................................................................84 Chapter Summary.......................................................................................................86 6 CONCLUSIONS AND FUTURE DIRECTIONS.....................................................87 Conclusions.................................................................................................................87 Future Directions........................................................................................................90 LIST OF REFERENCES...................................................................................................93 BIOGRAPHICAL SKETCH.............................................................................................99 vii

PAGE 8

LIST OF TABLES Table page 1-1 Comparison of commercial rechargeable batteries......................................................2 2-1 The effect of discharge rate and temperature on the specific capacity of the 70-nm electrode...................................................................................................................33 2-2 Parameters needed to calculate S c ..............................................................................38 viii

PAGE 9

LIST OF FIGURES Figure page 1-1 Schematic of discharge process of Li-ion battery. Charge moves from anode to cathode during discharge. Adapted from van Schalkwijk. 5 .....................................3 1-2 Scanning electron micrograph of the surface of a commercial nanoscale polycarbonate track-etch template membrane............................................................6 1-3 Fundamentals of electrochemical reactions comparing oxidation and reduction reactions. E F is the Fermi level, the average energy of the electrons of the current collector........................................................................................................21 1-4 Concentration polarization of intercalating Li-ions. Ions are non-uniformly distributed and are concentrated at the electrode surface. Gray represents LiV 2 O 5 ; whereas, yellow represents V 2 O 5 ...............................................................24 2-1 Discharge proprieties of a commercial Li-ion battery (AT&T Cylindrical 17500) at low temperature. This cell has a rated capacity of 800 mAh..............................26 2-2 Schematic of the template-synthesis of a V 2 O 5 electrode..........................................30 2-3 Scanning electron micrographs. A) 70-nm electrode. B) 0.45-m electrode. C) 0.8-m electrode. D) Polycarbonate template used to prepare the 70-nm electrode...................................................................................................................31 2-4 Electrochemical cell used for the characterization of V 2 O 5 electrodes......................32 2-5 Slow-rate (C/3) constant current discharge of 70-nm V 2 O 5 electrode using the cell pictured above..........................................................................................................33 2-6 Capacity ratio (see text) versus discharge rate for experiments conducted at 25 o C (Blue), 0 o C (Orange), and -20 o C (Green). A) R 70/0.8 B) R 70/0.45 ...........................35 2-7 Cycle life of 70-nm V 2 O 5 electrode charged and discharged at 10 C for 400 cycles. Columbic efficiency (Q out / Q in ) over this range is greatly than 95%.........39 3-1 Scanning electron micrograph. A) This low magnification image shows the uniformity over the large number of wires that constitute the electrode. B) This high magnification image is used to identify the solid-state diffusion path-length (radius of a nanowire) to be 50 nm..........................................................................46 ix

PAGE 10

3-2 Electrochemical response of nanostructured V 2 O 5 electrode to a E = -15 mV potential step. A) Current-time transient recorded for 45 minutes as the current decayed to a steady-state value. B) Current-time transient shown in Cottrell-plot form. The plateau represents the Cottrellian value It 1/2 ...................................48 3-3 Cottrell-plots as a function of potential and thus level of intercalation. At the potentials corresponding to the voltammetric minima, the Cottrellian value It 1/2 is shifted to more negative values and occurs at longer times.................................50 3-4 Charge intercalated per potential step. The majority of charge is intercalated at the potentials corresponding to voltammetric minima and galvanostatic plateaus..51 3-5 Solid-state diffusion coefficient as a function of potential. Y1-axis is the solid-state diffusion coefficient and the Y2-axis corresponds to characteristic diffusion time constant () at room temperature. Note that both y-axes are plotted on a logarithmic scale......................................................................................................52 3-6 Dependence of D Li+ conducted at 25 o C (Blue squares), 0 o C (Orange circles), and -20 o C (Green triangles)...........................................................................................53 3-7 Arrhenius plot to solve for activation energy of diffusive process. The slope of this line represents E A /R. All values of D were taken at E = 3.24 V. At this potential the intercalation is between phases of intercalation (see text)..................54 3-8 Cyclic voltammetric responses. A) Typical cyclic voltammetric response of 0.8-m electrode at room temperature. Scan rate = 0.1 mV s -1 B) Comparison of E pk 1 from cyclic voltammograms of nanoand micro-structured electrodes at various temperatures................................................................................................56 4-1 Schematic of template-synthesis of LiFePO 4 /carbon nanocomposite electrode........62 4-2 Scanning electron micrographs. A) Lower magnification image of the nanocomposite LiFePO 4 /carbon electrode. B) Higher magnification image of the nanocomposite LiFePO 4 /carbon electrode. Composite fiber diameter is 350 nm. C) Image of LiFePO 4 electrode synthesized by template dissolution method (absent of carbon). LiFePO 4 fiber diameter is 170 nm..............................64 4-3 XRD pattern from nanocomposite, microcomposite, compared to accepted literature values for LiFePO 4 The substrate is quartz. Acquisition time is 10 s...65 4-4 X-ray photoelectron spectroscopy data for C1s peak for the carbon in the nanocomposite electrode. Relative peak areas show amphorous form is dominant (~80%), but graphitic carbon is also present (~20%)...............................67 x

PAGE 11

4-5 Cyclic voltammograms. A) Cyclic voltammogram for the nanocomposite LiFePO 4 /carbon electrode prepared using a template with 50 nm-diameter pores. Scan rate = 0.1 mV s -1 B) Comparison of voltammetric response to theoretical response exhibiting diffusional tailing. The diffusional wave was generated from the normalized function. 79 ...........................................................................70 4-6 Constant current experiments. A) Constant-current (3 C) discharge of LiFePO 4 /carbon nanocomposite electrode. B) Specific capacity versus C-rate for the nanocomposite LiFePO 4 /carbon electrode prepared using a template with 50 nm-diameter pores...............................................................................................71 5-1 X-ray diffraction experiment. The sample is nanostructured V 2 O 5 on Pt. Acquisition time =10 s. The discrete lines represent the accepted values for orthorhombic phase V 2 O 5 (JCPDS 41-1426)...........................................................76 5-2 Cyclic voltammogram of ferrocence in Li-based and Mg-based systems (see text). Scan rate = 50 mV s -1 ...............................................................................................78 5-3 Cyclic voltammogram of ITO-glass/Mg x V 2 O 5 nanowire electrode. Scan rate = 0.1 mV s -1 X2-axis is calculated from data shown in Figure 5-1...........................80 5-4 Cyclic voltammogram of Pt/Mg x V 2 O 5 nanowire electrode. Scan rate = 0.1 mV s -1 The potential reference is Mg/Mg 2+ .....................................................................81 5-5 Comparison of rate capabilities of Li x V 2 O 5 and Mg x V 2 O 5 The mass for these calculations is the electroactive mass determined by integration of the respective cyclic voltammogram...............................................................................................83 5-6 Comparison of rate capabilities of nanoand micro-structured Mg x V 2 O 5 The mass for these calculations is the mass determined by ICP-AES analysis of the V-ion from post-electrochemical dissolution...........................................................83 5-7 Cyclic voltammogram for Sn-based electrode in Li-system prepared using a template with 50 nm-diameter pores. Scan rate = 0.1 mV s -1 .................................84 5-8 Cyclic voltammogram for Sn-based electrode in Mg-system prepared using a template with 50 nm-diameter pores. Scan rate = 0.1 mV s -1 .................................85 xi

PAGE 12

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 NANOSCALE ENERGY STORAGE ELECTRODES BY TEMPLATE-SYNTHESIS By Charles Robert Sides December 2005 Chair: Charles R. Martin Major Department: Chemistry Lithium-ion batteries powered the recent boom of personal electronic devices, such as cell phones, laptops and digital cameras. This success spawned a global research initiative to adapt this technology to more demanding applications, such as low-temperature systems or those relying on pulse-power. The electrodes of these batteries store charge by reversibly intercalating Li-ions. The facile insertion flux of Li-ions into the electrode and sluggish solid-state diffusion from surface to non-surface intercalation sites causes a polarization of charge. Therefore, under demanding conditions, the electrode discharges without fully accessing all charge-storage sites. The electrode used in these studies is created by template-synthesis. Template-synthesis is a general nanofabrication method capable of creating structures of known geometry. Nanomaterial-based electrodes mitigate the rate-limiting effects of sluggish electron-kinetics and mass-transport. The large surface-area of this design serves to distribute the current density, improving electron-kinetics, while the small size ensures xii

PAGE 13

that intercalation sites reside close to the surface, minimizing the distance Li-ions must diffuse in the solid-state. The intent of this dissertation is to highlight the success of nanomaterials in the study and design of energy-storage systems. It begins with a discussion of the low-temperature performance of Li-ion battery electrodes. Charge-storage characteristics and reproducible electrode geometries identify the fundamental breakdown of Li-ion batteries at low-temperature as the decrease in solid-state diffusion coefficient of the Li-ion (D Li+ ). This study then quantifies the value of D Li+ as a function of both intercalation-level and temperature. Next, it describes a variation of the previous template-synthesis method, in which the polymeric template is pyrolyzed to create a LiFePO 4 /carbon composite electrode. This composite improves upon the poor electron-conductivity of the otherwise attractive LiFePO 4 cathode. Finally, preliminary results from polyvalent-ion (Mg 2+ ) intercalation into the V 2 O 5 electrode are presented. Polyvalent-ions allow for more charge to be stored than is stoichiometrically possible by singly charged ions. The advantage of incorporating nanomaterials into the design of energy-storage devices is the recurring theme of this document. While the studies are approached from a fundamental view, conclusions such as those reported here will undoubtedly have profound commercial impact on the increasingly portable world. xiii

PAGE 14

CHAPTER 1 INTRODUCTION AND TEMPLATE-SYNTHESIS Introduction Batteries provide power to an incredible number of devices that we rely on daily such as automotives, electronics, and hearing aids and pacemakers. In addition to being portable, batteries operate more efficiently than solar power, cleaner than fossil fuels, and safer than nuclear power. Andrew Volta is credited for the first battery device in 1799 and sixty years passed until Gaston Plante discovered the first practical rechargeable battery (the lead-acid battery still used in cars today.) Now there are several classes of rechargeable batteries, which include Plantes lead acid battery, nickel-based systems (nickel-metal hydride and nickel cadmium) and the Li-ion battery. The introduction of Li-ion batteries by T. Nagaura and K. Tozawa of SonyTec Inc. in 1990 marked a major advance in battery technology. 1 Li-ion batteries have generated great interest as lightweight, portable, rechargeable power sources over the last decade. Li-ion batteries are now the power source of choice for laptops, cell phones, digital cameras, and camcorders. Table 1-1 compares several of the rechargeable batteries currently on the market. 2 Note that the Li-ion battery has unparalleled energy density, but is comparatively expensive. Li-ion batteries are popular because of their high cell potential, large cycle life, high energy (Wh L -1 Wh kg -1 ,) and power density (W L -1 W kg -1 .) For many applications these advantages outweigh the expense of these batteries. 1

PAGE 15

2 Table 1-1: Comparison of commercial rechargeable batteries. Energy Density Battery Type Nominal Voltage (V) By Weight (Wh kg -1 ) By Volume (Wh L -1 ) Pb-acid 2 25 80 Ni-Cd 1.2 50 150 Ni-MH 1.2 75 200 Li-ion 3.6 150 325 The public has quickly embraced this technology, which accounts for an approximately 3 billion dollar annual market. 3 Despite the commercial success of these batteries, a global research initiative exists to improve the existing design. The goal of this research is to apply this technology to more demanding applications, such as those relying on pulse-power or improved low-temperature performance. A specific example is for use as the electric component of hybrid vehicles, which is currently a NiMH based system. However, the current Li-ion battery design cannot adequately satisfy the power requirements of such systems due to the inability to deliver a sufficient quantity of charge at high discharge currents 4 combined with concerns of safety and capital expenses. The emergence of the burgeoning field of nanotechology has already tremendously impacted the field of scientific research. The fields of biotechnology and electronics have revolutionized. This dissertation details my efforts to incorporate the field of nanomaterials to improve the design of Li-ion batteries.

PAGE 16

3 Background on Li-ion Batteries The three primary components of any battery are the cathode, the electrolyte, and the anode. Li-ion batteries operate by reversibly intercalating charge in each of the two electrodes. Intercalation is the process by which a specific quantity of guest species (Li + ) is able to reversibly enter/exit a host structure (e.g., V 2 O 5 carbon), causing little or no difference to the host. These electrodes are separated by an ion-conductive electrolyte. Upon discharge, the Li-ions deintercalate from the low-potential electrode, migrate through the electrolyte, and insert into the high-potential electrode. The ions then must rely on solid-state diffusion to fill the non-surface intercalation sites. A corresponding quantity of charge travels the circuit and provides power to the load. This process is detailed in Figure 1-1. Figure 1-1: Schematic of discharge process of Li-ion battery. Charge moves from anode to cathode during discharge. Adapted from van Schalkwijk. 5 If current flow is reversed (from cathode to anode), Li-ions insert into the low-potential electrode and the system is charged. The low-potential electrode is the anode

PAGE 17

4 and the high-potential electrode is the cathode. This convention (adopted from the discharge process) is obeyed regardless of the direction of current flow. As in a traditional electrochemical process, the reaction is limited by either mass-transfer (diffusion of ionic species) or electron-kinetics. It does differ though, because it is the host, not the diffusing species, which oxides/reduces. General forms of the charge/discharge reactions for a Li-ion battery cathode and anode are shown in Equation 1-1 and 1-2, respectively. 1-1 1-2 For the cathode material shown in Equation 1-1 M = transition metal or a mixture of transition metals. The most popular choices are Co, Mn, Ni, and V. These are the constituents of the complex that undergo the redox reaction. The anode material is shown in Equation 1-2 as carbon. Development of the carbon-based anode was an engineering improvement to mitigate the early safety issues of lithium-metal anodes. Replacing the Li metal anode with the carbonaceous material obviated the short-circuit issues caused by dendritic growth during the Li-deposition process. Graphite has a layered structure and is a popular alternative to lithium because it can store a large amount of charge (~370 mAh g -1 ) and has a reduction potential just positive of lithium metal (~ 100 mV). The cell voltage delivered by a battery is the

PAGE 18

5 difference in the operating potential of the two electrodes, so a significantly negative anodic potential is necessary to maintain the large cell voltage. Nanomaterials by Template-Synthesis The Martin research group has pioneered the nanofabrication strategy of template synthesis. 6 This general method has been used to synthesize nanostructures of a variety of materials such as gold, 7-10 carbon, 11-13 semiconductors, 14,15 polymers, 6,16 and Li-ion battery electrodes, 12,14,17-26 our focus here. This method involves deposition of a material precursor into a microor nanoporous template. This template can be a variety of porous materials, such as commercially available track-etch polymer filters, anodized alumina, or even colloidal crystals. Depending on both the pore-diameter and the specific chemical interactions between the pore wall and the precursor, the resulting structures may be tubes (hollow) or wires (solid). The template is functional by restricting particle-growth. It may remain intact to impart directionality or increased mechanical strength during the experiments. For battery materials, the template is a commercially-available polycarbonate filter. The pores of this filter are monodisperse, nanoscopic in diameter, nominally cylindrical in shape, and traverse the entire length of the membrane. Electrode precursor is deposited into the pores of the organic template. The template is then preferentially etched by oxygen plasma, leaving structures of identical geometry as the pores. A sintering process imparts crystallinity to the structures to ensure that the host-structure is maintained during the intercalation/deintercalation process. These structures are referred to as nano if one or more of their dimensions are on the nanoscale (< 100 nm). However, the aspect ratio (length / width) is often on the order of 10.

PAGE 19

6 Templates Track-etch Membranes Microand nanoporous polymeric filtration membranes prepared via the track-etch method are available from commercial sources (e.g., GE Osmonics) in a variety of materials and pore geometries. 27 Polycarbonate is perhaps the most common example of a track-etch filter material. Other options of materials include polyester, Teflon and polyethersulfone. An electron micrograph of the surface of a polycarbonate template membrane is shown in Figure 1-2. Figure 1-2: Scanning electron micrograph of the surface of a commercial nanoscale polycarbonate track-etch template membrane. The term track-etch refers to the pore-production process. 28,29 Pores of the filters are created by exposing the solid material film to nuclear fission fragments, which leave randomly-dispersed damage-tracks in the film. The high energy (on the order of 2 GeV) of the fragments ensures that the tracks span the entire length of the membrane (typically from 6 to 10 m). These reactive chains end at the damage-tracks in the polycarbonate film and are then etched with a basic chemical solution, and they become pores. One ion

PAGE 20

7 creates one track, which in turn becomes one pore. During production the pore density is controlled by the duration of time that the polycarbonate film is exposed to the charged particles. Typical track-etch membrane pore-densities are 10 4 to 10 8 pores cm -2 30 Varying parameters of this etching solution such as temperature, strength and exposure time dictate the pore diameter. Commercial membranes are available with pore diameters ranging from 10 nm to 20 m. The microporous material etches to form uniform cylindrical pores, but as the pore diameter is reduced to the smaller nanoscopic dimensions the shape of the pore becomes like a cigar, slightly tapered at the ends. Microscopic investigations of template-synthesized nanostructures prepared within the pores of such nanoporous membranes have shown that the diameter of the pore in the center of the membrane is larger than the diameter at the membrane surface. 22,31 This pore geometry may arise from the fission fragment that creates the damage track and also generates secondary electrons, which contribute to the damage along the track. The number of secondary electrons generated at the faces of the membrane is less than in the central region of the membrane. An alternate theory is that the surfactant protective layer adsorbed to the surface of the membrane retards the local etching process. 29 Either suggested mechanism leads to bottleneck pores. In the production of template-synthesized nanostructured battery materials, polymeric (specifically polycarbonate) track-etch membranes are the current template of choice. These templates may be easily removed in conditions that do not adversely affect the nanostructures themselves. The wide variety of commercially available pore diameters and densities can generate comparative structures of differing geometries that are key tools for fundamental investigations. The disadvantage that is associated with

PAGE 21

8 these types of membranes and their application to electrode materials is their low porosity. These values are typically between 2 10%. This decreases the ratio of active material to a given footprint area or volume region on the current collector surface. It is important to remember, however, that these electrodes will serve as tools for fundamental studies, not a viable approach for commercial synthesis. Anodized alumina Another template membrane used in our laboratory is anodized alumina. Anodization of aluminum metal in an acidic environment causes the metal to etch in a fashion that leaves a porous structure. 32 These pores are extremely regular, having monodisperse diameters and cylindrical shapes in a hexagonal array. Unlike the track-etch process, this process is systematic and generates an isolated, non-connected pore structure. The pore densities of these alumina filters can be on the order of 10 11 pores cm -2 which is about 1000 times the density available in the track-etch polycarbonate membranes. The porosities can be as high as 50%. Also, alumina filters have much greater mechanical stability and chemical resistivity than polycarbonate. However, there is an extremely limited selection of commercially available pore sizes, and the smaller pores are branched. These membranes can be very thick (10 -100 m). It is also notable that this alumina structure is electronically insulating. There are advantages of alumina templates, but, to date, the application to Li-ion battery research has been limited. The high porosity (template pore volume becomes material volume) would dramatically increase the volumetric capacity (mAh L -1 ) available for a template-synthesized electrode. However, the chemical resistivity of the membrane dictates that more harsh dissolution methods be employed. There are few Li

PAGE 22

9 ion battery intercalation electrodes that survive such exposure. Recently though another research group has shown reversible Li-ion intercalation into MnO 2 wires electrochemically deposited in alumina template 33 Colloidal-crystals In the previous examples of templates, the pore-structure is cylindrical in general shape. An alternative to this configuration is a nanosphere template (see Steins recent review). 34 In this method, spherical particles with diameters of nanoto microscale dimensions (typically ~ 100s of nm) are deposited in a close-packed array. This is commonly accomplished by a solvent evaporation technique. If a solvent evaporates at a slow, controlled rate, it imparts an order to the particles. These particles are typically made of polymers (latex, polysterene) or silica. Since they are spherical in shape, void-volume exists in the interstitial sites, even when close-packed. These interstitial sites serve as the porous network and the nanosphere array act as the template. These close-packed spheres have a theoretical packing efficiency (defined as [volume of space occupied by the spheres/total volume]) depending on specific arrangements of between 68 74%. Therefore, the theoretical void volume is approximately 30% in a close-packed array and increases to 48% in a monolayer of spheres. Again by physically restricting particle growth (as opposed to a chemical technique), the deposition strategy becomes general. The requirements of size, shape, and uniformity are transferred from the material to the template. This is demonstrated with examples from the literature of synthesis of a variety of materials such as metal salts, 34,35 metal particles, 36 polymeric materials, 37,38 and evaporated or electroplated metal films. 39 Though the spherical shape is the most common, it is possible that the sphere can be distorted to other shapes, such as an ellipsoid or even a doughnut-shape. 38

PAGE 23

10 Material Deposition Strategies There are a number of differing structures of templates used for template-synthesis. There are just as many material-deposition strategies. It is the combination of templates and deposition-methods that provide a multiplicative number of solutions to a given challenge. In our laboratory alone we have shown examples of sol-gel deposition, electrochemical deposition (metals and polymers), electroless deposition, and chemical vapor deposition. Sol-gel Sol-gel chemistry has recently evolved as a powerful approach for preparing inorganic materials such as glasses and ceramics. This method for the synthesis of inorganic materials has a number of advantages over more conventional synthetic procedures. For example, high-purity materials can be synthesized at a lower temperature. In addition, homogeneous multi-component systems can be obtained by mixing precursor solutions; this allows for easy chemical doping of the materials prepared. 40 Such a versatile deposition technique partners well with the template-synthesis nanofabrication method. The sol-gel process typically involves hydrolysis of a solution of the precursor molecule to obtain first a suspension of colloidal particles (the sol) and then a gel composed of aggregated sol particles. The amorphous gel may then be thermally treated to yield a more crystalline product. We have recently conducted various sol-gel syntheses within the pores of the alumina and polycarbonate membranes to create both tubes and wires of a variety of inorganic oxide materials, including semiconductors 14,15 and Li-ion battery intercalation materials. 20,22,24,25 First, the template membrane is immersed into a sol for a given period of time, and the sol deposits on the pore walls.

PAGE 24

11 After hydrolysis, either a tube or wire of the gelled material is formed within the pores. As with other template synthesis techniques, longer immersion times yield wires, while brief immersion times produce tubes. The formation of tubes after short immersion times indicates that the sol particles adsorb to the template membranes pore walls. It has also been found that the rate of gelation is faster within the pore than in bulk solution. This is most likely due to the enhancement in the local concentration of the sol particles owing to adsorption on the pore walls. An electrostatic interaction may also pull charged sol particles to the walls of the template membrane. 14 Previously, our group has demonstrated the formation of TiO 2 semiconductor nanostructures via template-synthesis. 15 The mechanism of formation of TiO 2 from acidified titanium alkoxide solutions is well documented. In the early stages of the synthesis, sol particles are held together by a network of -Ti-Obonds. These particles ultimately coalesce to form a three-dimensional infinite network, the gel. The fact that tubes are initially obtained when this process is done in the alumina membrane indicates that the sol particles adsorb to the pore walls. It is well-known that at the acidic pH values, the sol particles are weakly positively charged. Tubes are formed because these positively charged particles interact with anionic sites on the alumina pore wall. 14 Several other inorganic oxides, specifically MnO 2 Co 3 O 4 ZnO, WO 3 and SiO 2 have been synthesized in a similar fashion. Since transition-metal oxides are typical choices for Li-ion battery cathodes, it can easily be seen how this method partners well with research of this field.

PAGE 25

12 Other deposition strategies of template-synthesis Electrochemical deposition is a based on electroplating metal from a solution into the pores of the template. The structures plate axially from a current collector. By controlling the charge (current and time) passed, the length of the structure can be controlled. This method typically produces wires, but tubes are possible when a molecular anchor is employed. We have made electrochemical plated structures of gold, silver, nickel, platinum and even polymers. Conversely, the electroless plating method relies on the electrochemical series for metals to replace ones more noble than themselves. A polymer (nonconductive) template is sensitized by Sn-ions, which are replaced by Ag particles, which are replaced by Au particles. These gold particles coalesce to form a continuous structure. This method plates metals from the walls of the template (radially); therefore, they plate as tubes with a controllable inner diameter down to when they become solid-wires. The plating occurs indiscriminately onto all faces of the membrane exposed to the solution. The most common example is the electroless deposition of gold into a track-etch polymeric template. Vapor deposition is useful because of the conformal coating that it can create by depositing materials in the vapor phase. It is commonly applied for making carbon nanotubes from ethylene gas; however, there are examples of conducting, semi-conducting and insulating materials. The difficulty with this method is the selection of suitable precursors. Anodized alumina templates are ideal selections to accompany this method, because of the ruggedness of that template. We deposit carbon by this method at greater than 600 o C. Another advantage to this method is that tubes of thin wall

PAGE 26

13 diameters are possible. The Martin laboratory has shown that vapor-deposited carbon nanotubes are ideal tools for fundamental studies of electroosmotic flow investigations. 11 Impact of Template-Synthesis to Nanoelectrochemsitry Fundamental Electrochemical Investigations Gold nanoelectrode ensemble One of the first fields where these template-synthesized nanomaterials demonstrated superior functional capabilities is as an array of gold nanoelectrodes elements. 10 Our group has created an array of gold nanoelectrodes by electroless deposition of Au to form solid wires in a track-etch polycarbonate template. Nanoelectrodes offer opportunities to perform electrochemical experiments to investigate the kinetics of redox processes that are too fast to measure at conventional macroscopic electrodes. 41,42 Also, they have the ability to serve as useful electrodes even in highly resistive media. 43 Depending on the spacing (pore density) and timescale of the experiment (voltammetric scan rate), we were able to demonstrate by radial and total overlap diffusional and voltammetric responses. The radial case refers to when the diffusion layer of each element is independent, whereas the total overlap refers to when the layers merge to act as a single diffusion layer. Template-synthesized gold nanowires operating in the total overlap regime are able to measure electrochemical response of redox molecules at trace concentrations (less than 2 nM for TMAFc + ). 10,44 This is a result of the faradaic response being a function of geometric gold surface area, and the background double-layer charging current being a function of active Au area. Gold nanoelectrodes were shown to provide advantageous electrochemical results, due to their unique template-synthesized geometry.

PAGE 27

14 Carbon nanotube membrane Carbon nanotubes are of great interest in both fundamental and applied science. Our group has created template-synthesized carbon nanotubes by chemical vapor deposition of ethylene gas onto a home-grown alumina membrane. This carbon nanotube membrane (CNM) is used for fundamental electrochemical investigations, specifically as tools to monitor and control electroosmotic flow (EOF). 11,45 EOF refers to the movement of solution past a stationary surface due to an externally applied electric field and is a consequence of the way ions are distributed near surfaces. 46 The Martin laboratory has shown the ability to measure EOF through carbon nanotube membranes. This work was further developed into the ability to control the rate and direction of EOF inside a CNM by depositing a thin-layer of PVFc. By applying a potential, the sign and magnitude of surface-charge, and thus EOF, can be effectively modulated. Li-ion Battery Nanoelectrodes Polymeric track-etch templates Nanomaterials are advantageous in regards to both ionic and electronic conductivity. Decreasing particle-size decreases the solid-state diffusion distance for Li + and increases the specific surface area, thus distributing the current and decreasing the effective current-density. This serves to increase the electron-transfer kinetics of the system. Solid-state diffusion coefficients are dependent on material and state-of-charge, but are extremely small, in the range of 10 -11 to 10 -15 cm 2 s -1 This is an intrinsic property. Therefore, our strategy is to use nanoparticles and minimize the distance the Li-ion must diffuse. This creates a situation in which the intercalation sites are as close to the electrode surface as possible.

PAGE 28

15 The Martin laboratory published its preliminary data of Li-ion batteries in 1997. 12,14 This was not long after these batteries were introduced to the marketplace. Patrissi 22 and Li 19 were the first of the Martin laboratory to demonstrate the structural and electrochemical characterization of template-synthesized Li-ion batteries electrodes, V 2 O 5 cathodes and Sn-based anodes, respectively. The work detailed here builds upon the preliminary findings of these scientists. They built the foundations for a number of the experiments discussed below. A tremendous number of laboratories across the world are involved in this research. Several premier journals devote sections of every issue to featuring advances in energy storage (specifically, Li-ion based technology). Current major goals in this research are to 1) create high-rate pulse-power compatible structures, 2) identify alternative electrode materials (specifically more energetic, less expensive, and less hazardous cathodes), 3) fabricate solid-electrolytes with viable ionic conductivity at room temperature. Energy storage and production are topics of conversation at every corner of the industrialized world. The focus of research in the field of Li-ions batteries is to create systems of the highest energy and power densities, both on a gravimetric and volumetric basis. Essentially the goal is to get more power from a smaller, lighter device. Template-synthesis as applied to the field of Li-ion batteries creates ideal tools for the fundamental studies of factors that limit both rate and power. Previously, we have focused on using constant-current to discharge template-synthesized electrodes and comparing that to the response of thin-film control electrodes. During the discharge process, Li-ions deintercalate from the anode, migrate through a Li-ion conducting electrolyte, and then

PAGE 29

16 intercalate into the cathode. When these Li-ions intercalate into an electrode, they are not able to solid-state diffuse rapidly enough to compensate for the facile nature of the insertion-flux (D ~ 10 -8 to 10 -11 cm 2 s -1 ). This results in concentration-polarization of Li-ions at the surface during intercalation; conversely, the polarization is in the core during the deintercalation process. 47 This is a practical limitation to the stoichiometric quantity of charge-storage (mAh g -1 ) that is theoretically possible. The discharge-rate (current) determines this rate of insertion-flux; therefore at the high currents of a demanding application, this concentration-polarization problem is exacerbated. Since the diffusion coefficient is an intrinsic property of the electrode material, we work with the strategy of decreasing the size of the electrode-particles. This allows the intercalation sites to be closer to the surface, shortening the distance over which the sluggish solid-state diffusion process must propagate. 20,22,24,25 This plan of using smaller particles has another attribute. The increased surface-area per total-volume fraction serves to decrease the effective current density at any given rate. This works to offset any sluggish electron-kinetics of the system. Since the electrodes must exhibit both good electronic and ionic conductivity, a structure comprised of electronically addressable small particles is extremely favorable. This has been demonstrated by template-synthesis of both cathodic 20,22,24,25 and anodic materials. 20,48 In every case, the nanostructured material was able to deliver higher specific capacity (mAh g -1 ) at any given discharge rate (C, C = h -1 ) than the microstructured control electrode.

PAGE 30

17 For an electrode to function in a rechargeable system, it must be able to maintain its ability to be charged and discharged for many cycles. This cycle-life parameter is measured as the charge-discharge cycle is repeated at the same rate. Our nanostructured electrodes has been cycled for greater than 1400 times at the high-rate of 58 C, without diminishing the quantity of charge delivered per cycle. 20 As demonstrated in the above discussion, electrodes comprised of small particles are advantageous. Void-volume is used to keep these particles small. While this has no effect on the gravimetric capacity (mAh g -1 ), it does on volumetric capacity (mAh L -1 ). Volumetric restrictions of an application may be just as stringent as gravimetric restrictions. Our lab has shown the ability to refill a portion of this void-volume with more active material to increase the charge stored per liter. 23 In that investigation, the volumetric charge delivered by the template-synthesized nanostructured electrode at high-rates of discharge actually surpassed that delivered by the thin-film control. Nanosphere-templated structures Laboratories throughout the world are employing template-synthesis methods. Other groups (most notably from the University of Minnesota) have worked with a template method that uses nanospheres to create similar short-diffusion-distance materials. 49 An array of monodisperse PMMA spheres of diameter of 300 nm serves as the template. Approximately 10 layers of these spheres construct an array. The void-volume of these close-packed spheres hosts a Sn-based precursor solution. After processing the precursor into a state capable of reversible Li-ion intercalation, the PMMA-template spheres were removed via calcination. The electrode structure shrinks as a result of this heating. Average solid-state diffusion distance for Li-ions is about 170 nm. This electrode showed good order to an area of 100 m 2 The resulting electrode

PAGE 31

18 was characterized electrochemically and was able to reversibly store and deliver charge under conditions similar to those previously described. 49 Li-ion Battery Electrode Fabrication Strategies Commercial Process Method This synthetic method can be contrasted to a general method for commercial Li-ion battery systems. 47 There are many types of commercial batteries, such as prismatic cells, coin cells, and jelly-roll cylindrical cells. The assembly and application of each are different; however, the synthesis and components are general. The commercial battery cathode is created in a high-temperature procedure that gives particles of the active transition-metal-oxide. This process generates particles that range in diameter from approximately 2 to 20 microns. These discrete particles are then mixed into slurry with polymeric binder (PVDf) and a conductive element (carbon) and pressed into a metal mesh current collector. The conductive element is necessary to improve the electronic conductivity of the system, which is very low in this system because the particles are only point-connected. The polymer is necessary to simply ensure that the slurry physically contacts the current collector. In our template-synthesis approach, these inactive (not capable of Li-ion intercalation) components are not necessary and are excluded so as not to decrease volumetric and gravimetric energy densities or complicate analysis. 50 Notice that competing technologies use the mass of only the active material when determining charge-per-gram. Template-Synthesis Method In this embodiment of Li-ion electrodes a precursor-impregnated polycarbonate template membrane is attached to a section of metal foil. The foil has dual-functionality as it serves as a substrate during synthesis and as a common current collector during

PAGE 32

19 electrochemical characterization. Template-synthesis methods yield a fiber-based electrode that consists of structures that mirror the geometry (length, diameter, and number density) of the pores of the template. Typical values are 100-nm in diameter, 6 microns in length, and 10 8 elements per cm 2 These structures extend from the surface of the current collector like the bristles of a brush. There are only two components of the electrode active material and current collector. Each electrode is then electrochemically characterized as a half-cell. Because of the parallel electronic-conduction mechanism of our template-synthesized nanostructrued electrodes, carbon is not required. 25 No binder is necessary because the structures are directly attached to the metal foil that serves as both a physical substrate and a common current collector. The absence of non-active components in our electrodes also assists in increasing volumetric energy densities. An Electrochemical Primer Comparing Electrochemical Methods A standard electrochemical experiment could be envisioned as a planar metal electrode (working electrode) and a facile redox couple such as ferrocene/ferrocenium. The ferrocene would be dissolved in water, along with a supporting electrolytic salt. This other salt is composed of nonelectroactive ions and serves to decrease the contribution to migration to the mass transfer, as well as decreasing the cell resistance. A reference electrode is in the system as well. The purpose of this electrode is to provide a comparative value for the applied potential. If the ionic conductivity (S cm -1 ) of the electrolyte is comparatively low, a Luggin capillary may be employed to minimize the effective distance between the working and reference electrodes. This decreases the cell resistance, minimizes iR drop, and ensures that the potential measured is accurate. A

PAGE 33

20 counter electrode is inert in the system conditions and serves to pass current through the system. A potentiostat is connected to the electrodes. Envision a potentiostat as the instrument that sets the working electrode to the appropriate potential compared to the reference electrode and measures the flow of current between the working and counter electrodes. For a reduction-oxidation (redox) reaction to occur, an electron and ion must unite. Figure 1-3 illustrates this from thermodynamics. Electron-kinetics describes the ability for electrons to be transferred at a certain rate constant. Ion transport (mass transfer) is accomplished by one of three mechanisms: diffusion, migration, and convection. Commonly systems eliminate convection by not stirring the system. The effect of migration, movement of ions in an electric field, is minimized by the supporting electrolyte. This leaves diffusion as the predominant mass-transfer mechanism. Diffusion is a result of a concentration gradient. This concentration gradient is related to flux of the species by a proportionality constant, its diffusion coefficient (D). This parameter has units of area per time. A small molecule in water has D ~ 1 x 10 -5 cm 2 s -1 While the diffusive movements of species are random in nature, the bulk movement serves to smooth concentration gradients. Electrochemical measurements on Li-ion batteries are similar, but not identical. We test our electrodes in a half-cell setup, meaning that the cathode and anode are individually characterized. Li-ions are solvated in a non-aqueous electrolyte. This setup is inside of an argon glovebox, to protect the dry electrolyte and the lithium metal. The diffusive movement is associated with the Li-ions, but Li-ions are stored and the host material is subject to the redox reactions.

PAGE 34

21 Figure1-3: Fundamentals of electrochemical reactions comparing oxidation and reduction reactions. E F is the Fermi level, the average energy of the electrons of the current collector. The host material is electronically conductive and injects electrons by contact with a metal current collector. This intermediary between the ions and the current collector serves to slow electron transfer. In addition, though Li-ions are small, they must diffuse through solid-material as opposed to only liquid in a traditional system. The diffusion coefficient for this process is about 6 orders of magnitude lower than for a liquid-phase system. Therefore, the rates of the experiments must be several orders of magnitude less than those used to typical experiments. Again, the rate of the reaction is either limited by mass-transfer or electron-transfer. Comparison of Diffusion Regimes This leads us to a discussion of two regimes of diffusion that are discussed in this dissertation semi-infinite and finite. Semi-infinite diffusion refers to the situation when Species loses eOxidation Species gains eReduction Electrode Solution Vacant MO Occupied MO E F E G G = nFE

PAGE 35

22 movement of ions is unrestricted. Ions may freely diffuse without boundary constraints, so during the timescale of the experiment effects of the cell walls are not felt on the system. Therefore, at some value away from the electrode there is no concentration gradient. The direct opposite is the case for finite diffusion when the movement of ions is constrained. This is often the case in thin-layer experiments where propagation of the diffusion layer is truncated at some distance, the thickness of the thin-layer. A colleague once compared these diffusion regimes to me as an analogy of two people each holding one end of a rope. The first causes the rope to oscillate by moving his hand up and down. This movement represents the diffusion process. If the rope is sufficiently long enough, the second person cannot feel the response (semi-infinite diffusion). However, if the rope is short enough, the perturbation is felt by the second person and it is stopped by his hand (finite diffusion). There is no infinite diffusion regime, because the diffusing species is confined by the walls of the cell, and as t approaches infinity it will eventually encounter this system constraint. Electrochemical Methods for Li-ion Battery Electrodes We will use several standard electrochemical methods to characterize the Li-ion battery electrodes cyclic voltammetry, chronopotentiometry, and small potential-steps. Cyclic voltammetry will be used to confirm the activity of the electrode, identify the potential region of electroactivity, and comment on the electron-kinetics and diffusion regimes of the process. This is accomplished by sweeping an applied potential, while monitoring the current response. Chronopotentiometry uses constant applied currents to charge and discharge the electrode, while monitoring the change in potential. By applying a known current to a known electrode mass, the time is easily converted to charge stored and thus quantity of Li + Larger currents (faster rates) decrease the time

PAGE 36

23 regime for the intercalation process. Therefore, the solid-state diffusion process can access less of the intercalation sites at larger currents. Measurement of the rate-capability is a useful way to compare electrodes. The final method uses small applied potential steps and monitors the current-response over time. Data from these small potential steps are interpreted via the Cottrell relationship to identify the solid-state diffusion coefficient of the Li-ion and identify the activation energy of the diffusion process via an Arrenhenius relationship. Since the quantity of charge stored in a Li-ion battery electrode is dependent on the mass of the electrode, the molecular weight of the host species, as well as the stiochiometric theoretical maximum, a normalized parameter is necessary when discussing charge-discharge experiments. This parameter is C-Rate, 1 C = 1 h -1 described in Equation 1-3. mCiRateCs 1-3 Here, i represents the applied current, C s represents the theoretical specific capacity (mAh g -1 ), and m represents the mass of the electrode. The rate of 1 C is the current necessary to incorporate the theoretical maximum charge in 1 h. For reference, if a laptop battery lasts for 2 h, it is discharged at an average rate of C/2 (though, the rate is actually a little higher, because this calculation assumes theoretical charge-storage.) Figure 1-4 is a schematic of the result of the concentration polarization effect. This effect depends greatly on C-Rate. At small rates, there is enough time for Li-ions to occupy/vacate nearly all of the intercalation sites, even at the core of the electrode particle. At high rates, there is much less time for these Li-ions to diffuse in the solid

PAGE 37

24 state, creating this large number of inaccessible intercalation sites, resulting in poor charge-storage characteristics. Figure 1-4: Concentration polarization of intercalating Li-ions. Ions are non-uniformly distributed and are concentrated at the electrode surface. Gray represents LiV 2 O 5 ; whereas, yellow represents V 2 O 5 Chapter Summary The intent of this dissertation is to demonstrate how nanoscale Li-ion battery design succeeds where conventional technology fails. It will begin with a discussion of the improved low-temperature performance of these electrodes. This will then be extended to quantify the solid-state diffusion coefficient of the Li-ion as a function of intercalation-level and temperature. Next, it describes a variation of the general synthetic method, in which the polymer template is pyrolzed to create a LiFePO 4 /carbon composite electrode. Finally, this is followed by a demonstration of polyvalent-ion (Mg 2+ ) intercalation into V 2 O 5

PAGE 38

CHAPTER 2 NANOSTRUCTURED ELECTRODES AND THE LOW-TEMPERATURE PERFORMANCE OF LI-ION BATTERIES Introduction Li-ion batteries have become the power source of choice for consumer electronic devices such as cell phones and laptop computers. 2,51-58 This is because these batteries have good rechargeability (1000+ cycles) and offer higher energy density (stored charge per unit volume or mass of the battery) than competing battery technologies. 2,51 However, it is well documented that Li-ion batteries show poor performance at low temperature or under the extreme currents necessary for pulse-power. 59-63 Specifically, the amount of charge delivered from the battery at temperatures below 0 o C is substantially lower than the amount of charge delivered at room temperature. 59,60,62 This precludes the utilization of these batteries in a number of defense, space and even terrestrial applications. 64 Figure 2-1 is generated from data published by Nagasubramainan 60 on the effect of temperature on the performance of a commercial Li-ion battery cell. Note that the C-Rates are very low; however, still there is a dramatic decrease in the charge-storage ability of the cell as temperature is decreased. We have been investigating the application of nanotechnology to Li-ion battery electrode design. 17,20-22,24 Previously, the Martin laboratory has shown that template-synthesized Li-ion battery electrodes are capable of delivering more charge-per-gram at the same normalized discharge-rate (C-Rate) than control film electrodes. Based on these studies it seemed likely that Li-ion battery electrodes composed of nanoscopic particles 25

PAGE 39

26 of the electrode material could mitigate this low-temperature performance problem. We prove this case here by showing that nanofibers (diameter = 70 nm) of the electrode material V 2 O 5 deliver dramatically higher specific discharge capacities at low temperatures than V 2 O 5 fibers with micro-sized diameters. Figure 2-1: Discharge proprieties of a commercial Li-ion battery (AT&T Cylindrical 17500) at low temperature. This cell has a rated capacity of 800 mAh. Li-ion batteries store charge by intercalating Li + from a contacting solution phase, along with an equivalent number of electrons, into the batterys anode material. 2 During discharge the Li + must diffuse out of the anode material, through the electrolyte, and intercalate into the cathode material. This process can be seen in Figure 1-1.

PAGE 40

27 For the cathode material V 2 O 5 investigated here, the reversible charging and discharging reactions can be written as Equation 2-1. 2-1 Other laboratories have concluded that the cathode, anode, and electrolyte were the source of the fundamental breakdown that occurs at low temperature. While there is controversy in the scientific literature, 59,60,62,63 the most likely causes of the poor low-temperature performance are either 1.) Diminution in the rates of these electrochemical charge/discharge reactions at low temperature. 2.) Diminution in the rate at which Li + diffuses within the particles which constitute the electrode at low temperature. We hypothesized that in either case, an electrode composed of nanoscopic particles (diameter less than 100 nm) would provide better low temperature performance than the ~10-m -sized particles 65 used in commercial battery electrodes. This is because an electrode composed of nanoscopic particles would, in general, have higher surface area than an electrode composed of large particles, and this would mitigate the slow electrochemical kinetics problem. Furthermore, the distance that Li + must diffuse within the particle would be decreased for nanoscopic particles, and this would mitigate the slow solid-state diffusion problem. Electrode Synthesis To prove this point we have used the template-synthesis method 6 to prepare cathodes composed of monodisperse V 2 O 5 nanofibers (diameter = 70 nm) that protrude from a current collector surface like the bristles of a brush (Figure 2-2A). We compare the low-temperature charge/discharge performance of these nanofiber cathodes with

PAGE 41

28 cathodes composed of V 2 O 5 fibers with diameters of 0.8 m (Figure 2-2C), as well as with cathodes composed of 0.45 m-diameter fibers (Figure 2-2B). If our hypothesis is correct, the low-temperature performance of the cathode composed of the 70 nm-diameter nanofibers (designated the 70-nm electrode) should be dramatically better than that of the cathode composed of the 0.8 m-diameter fibers (0.8-m electrode), and the low-temperature performance of the cathode composed of the 0.45 m-diameter fibers (0.45-m electrode) should be between these two extremes. This experimental design exploits the power of the template-synthesis method for preparing monodisperse fibers having any desired diameter, 6 and builds on our prior work on the electrochemistry of nanostructured Li-ion battery electrodes. 17,20-22,24 The template-synthesis method entails using the pores in a synthetic membrane or other solid to form nanofibers, nanowires or nanotubes of a desired material. 6 Commercially-available (Poretics, Inc) polycarbonate filtration membranes (figure 2-2D) were used as the templates for these studies. These filters have monodisperse cylindrical pores that run through the complete thickness (6 10 m) of the membrane and are available with a wide range of pore diameters. The pore diameter of the template determines the diameter of the fibers synthesized within the pores. 6 This method was described in detail previously. 22 Briefly, a 1.0 cm x 0.7 cm piece of the template membrane was placed on the surface of a Pt foil electrode, and 0.6 L of the liquid V 2 O 5 precursor triisopropoxyvanadium (TIVO) was applied to the membrane surface. TIVO fills the pores in the template, and hydrolysis in air converts this material to V 2 O 5 This chemistry is straight-forward and it shown in Equation 2-2.

PAGE 42

29 2-2 In fact, this reaction is so facile that the deposition must occur inside an argon-filled glovebox, so that the precursor may impregnate the pores in this low-viscosity state. It is then transferred to a low O 2 environment (antechamber of the glovebox), so that they hydrolysis can proceed slowly enough as to form continuous structures inside the pores. This was allowed to proceed at room temperature for 12 hours, followed by hydrolysis for 2 hours at 80 o C in air. The film of V 2 O 5 that covered the upper face of the membrane was then removed by wiping with a damp cotton swab, and this procedure was repeated to ensure that the pores were completely filled with V 2 O 5 Oxygen plasma (25 W, 15 Pa, 2 hours) was then used to remove the polycarbonate template. To ensure complete conversion to crystalline V 2 O 5 the fibers were than heated at 400 o C for 10 hours in flowing O 2 gas. X-ray diffraction studies confirmed that the material obtained is orthorhombic V 2 O 5 (experimental details discussed later). 66 A schematic of this process is shown in Figure 2-2.

PAGE 43

30 Poly carbonate T e mplate Membrane 50-nm no min a l d i am eter por es 6 m in len g th 6 x 10 8 pore s cm -2 Te m p late with Sol-Gel Precurso r Gel TIVO precur sor Air; Heat 80 o C; 2 h Etch template Ox ygen plas ma; 20 W, 2 h, 15 Pa Heat 40 0 o C in O 2 ; 10 h V 2 O 5 Nanos t ruc t ures on Pt Current Coll ecto r 100 % activ e Figure 2-2: Schem a tic of th e tem p late-synthesis of a V 2 O 5 electrod e Scanning Electron Microscopy Scanning electron m i crographs of electrode s of created in various tem p lates are shown Figure 2-3. These sam p les are attach ed to stubs by conductiv e copper tape prior to im aging. The electrodes were not spu ttered with any conductive m a terial. This dem onstrates the sufficient electron conductiv ity of these samples. The polycarbonate tem p late was sputte red with a th in layer of Au/Pd to prevent ex cessive charging during im age acquisition. These tem p late-syn thes ized V 2 O 5 nano and m i crofibers elect rodes serve as useful tools in fundam ental investigations, though the void-volum e, that is responsible for the short diffusion distances, sacrifices v o lum e tric ca pacity (m Ah L -1 ). Previously, we dem onstrated a m e thod that creates a com p ro m i s e of these tw o interrelated param e ters. 23

PAGE 44

31 Figure 2-3: Scanning electron micrographs. A) 70-nm electrode. B) 0.45-m electrode. C) 0.8-m electrode. D) Polycarbonate template used to prepare the 70-nm electrode. Electrode Electrochemical Investigations Constant Current Discharge Experiments The charge/discharge reactions for these nano and microfiber electrodes were investigated in an electrolyte solution that was 1 M in LiClO 4 dissolved in a 1:1:1 (by volume) mixture of ethylene carbonate, diethyl carbonate, and dimethyl carbonate. 67 Li metal ribbons served as the counter and reference electrodes. A schematic of this electrochemical cell can be seen in Figure 2-4.

PAGE 45

32 Figure 2-4: Electrochemical cell used for the characterization of V 2 O 5 electrodes. Temperature was set and maintained via a circulation bath of 80:20 ethylene glycol:water. All potentials are quoted here vs. the Li/Li + reference. A galvanostat was used to apply a constant current to sequentially discharge and charge the V 2 O 5 (Equation 2-1), and monitor the change in potential of the V 2 O 5 electrode as a function of time. The discharge reaction was continued until a potential of 2.8 V was achieved, and the charge reaction was terminated when the potential reached 3.8 V. Over this potential window 1 mol of V 2 O 5 is known to reversibly intercalate 1 mol of Li + (i.e., x = 1 in Equation 2-1). 68 This corresponds to a maximum specific (per g) charge-storage capacity of 148 mAh g -1 The mass of V 2 O 5 present in all of our electrodes was determined gravimetrically so that the specific capacities could be determined from the potential-vs.-discharge-time data. A typical potential-vs.-time curve associated with constant-current discharge of a 70-nm electrode at 25 o C is shown in Figure 2-5. A specific capacity of 145 mAh g -1 (equivalent to x = 0.98) was obtained from these data.

PAGE 46

33 Figure 2-5: Slow-rate (C/3) constant current discharge of 70-nm V 2 O 5 electrode using the cell pictured above. This shows that at this temperature and current density, nearly 100% of the maximum specific capacity of the cathode is being delivered. The effect of discharge current or C-Rate and temperature on the specific capacity of the 70-nm electrode is shown in Table 2-1. Table 2-1: The effect of discharge rate and temperature on the specific capacity of the 70-nm electrode. C-Rate / h -1 Specific Capacity / mAh g -1 25 o C 0 o C -20 o C 0.2 77.5 1 144.8 104.9 63.8 5 123.0 74.4 40.7 10 107.5 60.2 33.5 20 90.4 48.7 27.3 25 25.5 30 43.1 24.8 35 23.0 40 72.9 39.6 22.7

PAGE 47

34 Comparison of Rate Capabilities At all temperatures, we see the decrease in capacity with discharge rate, typical of Li-ion battery electrodes. 20,21,51,57 However, as expected, 60-62 this decrease in capacity becomes much more pronounced at low temperatures. Data of this type were obtained for all three of the V 2 O 5 fiber electrodes. Again, our objective is to demonstrate that the 70-nm electrode is able to deliver greater specific capacity at low temperatures than the electrodes composed of the larger-diameter fibers. We define the specific capacity of the 70-nm electrode at a given discharge current, I, and temperature, T (Table 2-1), as C s-70,I,T We define the identical parameter (i.e., same current and temperature) for the 0.8-m electrode as C s-0.8,I,T Finally, we define a capacity ratio R 70/0.8 = C s-70,I,T / C s-0.8,I,T If R 70/0.8 is greater than unity, then at these particular values of I and T, the 70-nm electrode shows a capacity advantage relative to the 0.8-m electrode. Figure 2-6A shows plots of R 70/0.8 vs. the discharge rate for experiments done at temperatures of 25 o C, 0 o C and -20 o C. Looking at the 25 o C curve first, we see that at low discharge rates, R 70/0.8 is nearly unity, which means that at room temperature and low discharge rate, both electrodes are delivering nearly 100% of their maximum specific capacities; hence, at room temperature and low discharge rates, there is no capacity advantage for the 70-nm electrode relative to the0.8-m electrode. At higher discharge rates R 70/0.8 increases above unity, indicating that there is now a capacity advantage for the nanofiber electrode; however, at room temperature this advantage is modest.

PAGE 48

35 Figure 2-6: Capacity ratio (see text) versus discharge rate for experiments conducted at 25 o C (Blue), 0 o C (Orange), and -20 o C (Green). A) R 70/0.8 B) R 70/0.45

PAGE 49

36 Turning now to the -20 o C data, we find that at the lowest discharge rate the 70-nm electrode delivers twice the specific capacity of the 0.8-m electrode. Furthermore, R 70/0.8 increases dramatically with discharge rate; indeed, at the highest rate studied, the specific capacity delivered by the 70-nm electrode is almost two-orders of magnitude greater than the specific capacity delivered by the 0.8-m electrode. These data prove our hypothesis that an electrode composed of nanoscopic particles shows better (indeed, dramatically better) low-temperature performance than an electrode composed of micron-sized particles. The 70-nm electrode also shows a capacity advantage at the intermediate temperature of 0 o C, but as would be expected, the advantage is not nearly as dramatic as at -20 o C. These data confirm the results obtained with commercial Li-ion batteries that performance at room temperature and 0 o C are not so different but that performance becomes markedly worse at -20 o C. 62,67 We obtained similar data for the 0.45-m electrode and calculated the analogous specific capacity ratio R 70/0.45 = C s-70,I,T / C s-0.45,I,T If our hypothesis is correct, the 70-nm electrode should show better low-temperature performance than the 0.45-m electrode (R 70/0.45 > 1), but at any value of discharge current, R 70/0.45 should be less than R 70/0.8 A comparison of the curves in Figure 2-6A and B shows that this is indeed the case. We have proven that a nanostructured Li-ion battery electrode shows better low-temperature performance than analogous electrodes composed of micron-sized particles. Two questions remain. 1.) How do these studies relate to practical battery-electrode design? 2.) From a fundamental viewpoint, what is the genesis of this improved low-T performance? With regard to question 1, our results suggest that to improve low-T performance, practical Li-ion battery electrodes should be prepared using smaller

PAGE 50

37 particles of the active material than the ~10 m-sized particles currently used. However, a practical Li-ion battery electrode contains other components in addition to the active material; these include a polymeric binder and graphitic particles to improve electronic conduction through the electrode. 50 The effect of amount and particle size of these ancillary components would also have to be evaluated. In this regard, our experimental design based on template-synthesized micro and nano particles has tremendous advantage for the fundamental studies performed here because these ancillary components are not needed and therefore do not complicate the analysis. Our data also provide an answer to question 2. We suggested above that the poor low-temperature performance of Li-ion batteries is due to either a decrease in the diffusion coefficient within the electrode particles, or a decrease in the rate of the electrode reactions (Equation 2-1), with decreasing temperature. If the solid-state diffusion coefficient is the culprit then low-temperature performance should improve with decreasing particle diameter, and this is exactly what is observed here. If electrode reaction rate is the culprit, then low-temperature performance should improve with a parameter we designate S c the electrode-particle surface area per cm 2 of current collector area. The beauty of the template method is that S c can be easily calculated from the fiber diameter, the pore density of the template membrane, and the fiber length. Table 2-2 shows these parameters. These calculations show that because of the interplay between pore diameter and density in the template membranes, S c is higher for the 0.45-m electrode (S c = 14.1) than for the 70-nm (S c = 7.9) or the 0.8-m (S c = 6.8) electrodes. This indicates that if electrode reaction rate is the culprit, then the 0.45-m electrode would show the best low-temperature performance, and this is not observed

PAGE 51

38 experimentally. Hence, our data show that the temperature dependence of the solid-state diffusion coefficient determines the low-temperature performance of the electrodes studied here. Evaluation of S c As discussed in the text of the paper, S c is the electrode-particle surface area per cm 2 of current collector area. The parameters needed to calculate S c are presented in Table 2-2. Table 2-2: Parameters needed to calculate S c Pore Diameter (nm) Pore Density (cm -2 ) Membrane thickness (m) 70 6x10 8 6 450 1x10 8 10 800 3x10 7 9 To calculate S c for the template-synthesized nano or microfiber electrode, one first calculates the circumference of an individual fiber. The surface area of the fiber is the product of the circumference and the length of the fiber. The length is given by the membrane thickness. The number of fibers per cm 2 of current collector surface area is equivalent to the pore density of the membrane. Therefore, the surface area per fiber multiplied by the pore density is S c Effect of Cycle Life The order of the series of experiments dictates that each test delivers less charge than the previous (e.g., low to high C-rate of discharge and high to low temperature.) Therefore, it is helpful to examine any adverse effects on the cycle life on the system. After the entire set of experiments was completed, if the system was returned to room temperature (25 o C), the electrodes maintained 98+% of the capacity shown in the

PAGE 52

39 original series of tests at that temperature. This is an important control experiment to help validate the data that are the basis for our conclusions. Also, the cycle-life can easily be characterized by charging and discharging the electrode with a constant current. Figure 2-7 shows the cycle life of a V 2 O 5 nanostructured electrode that is charged and discharged at the constant rate of 10 C. Since the same current is used to charge and discharge the electrode, the columbic efficiency (charge out / charge in) reduces to a ratio of discharge time / charge time. The columbic efficiency of this electrode was greater than 95%. The electrode stored and delivered virtually the same charge over nearly 400 cycles at a high rate of discharge. Figure 2-7: Cycle life of 70-nm V 2 O 5 electrode charged and discharged at 10 C for 400 cycles. Columbic efficiency (Q out / Q in ) over this range is greatly than 95%.

PAGE 53

40 Effect of Electronic Conductivity In commercial Li-ion battery cathodes, it is necessary to add an electronically conductive material, typically carbon particles, to improve electronic conductivity through the electrode. We have shown that this is not necessary with template synthesized electrodes. 20-22,24 This is because while the electronic resistance of a single fiber (R f ) making up the template-synthesized electrode might be high, the electrode is a parallel ensemble of such fibers. As such, the total resistance of the electrode (R t ) is given by R t = R f /N, where N is the number of fibers in the electrode. That electronic conductivity does not dominate the rate capabilities of the electrodes studied here is easy to prove. Assuming the same length, it is easy to show that the electronic resistance of a 70 nm diameter fiber is 33 times higher than that of a 0.8 m diameter fiber composed of the same material. If the rate capability was limited by electronic conductivity, the 70-nm electrode would show the worst rate capability and the 0.8-m electrode would show the best. This is exactly the opposite of what is observed experimentally. Chapter Summary This chapter clearly demonstrates the advantages of template-synthesis as a method to create ideal electrodes for fundamental studies of Li-ion batteries. The electrode precursor was deposited into the pores of a commercially available polycarbonate template membrane. The template serves to restrict particle growth as the precursor is hydrolysized to V 2 O 5 a known Li-battery cathode material. These V 2 O 5 wires extend from the surface of a current collector like bristles of a brush. The geometry of the pores of the template is imparted onto these wires a nanoporous template yields nanostructured wires and microporous template yields microstructured wires. We

PAGE 54

41 demonstrated the ability to create structures of reproducible diameters by this template-synthesis method. Three of these electrodes with different wire geometries were electrochemically charged and discharged by a constant-current. By creating particles of the shortest solid-state diffusion distance, we were able to demonstrate superior low-temperature rate-capabilities when compared to microstructured control electrodes. However, these superior discharge characteristics would be achieved with any competitive nanofabrication method. The ability of template-synthesis to create structures of reproducible solid-state diffusion distance and surface-area is what differentiates it from other nanofabrication methods. Analysis of these data identifies the decrease in solid-state diffusion coefficient of the Li-ion associated with the decrease in temperature to be the rate-limiting factor in Li-ion batteries. The conclusions presented in this chapter are a benchmark study in the field of low-temperature Li-ion battery research.

PAGE 55

CHAPTER 3 EVALUATION OF SOLID-STATE DIFFUSION COEFFICIENT OF LI-IONS AT LOW TEMPERATURE Introduction Since their introduction in the early 1990s by Sony 1 Li-ion batteries have been the focus of intense research to adapt their design to more demanding systems, such as those operating on pulse-power and at low-temperature. It is widely known that these batteries can only provide limited capacity (mAh) at low temperatures. 25,60,61 The previous chapter identified the fundamental breakdown that occurs in these batteries at low temperature to be the decrease in solid-state diffusion coefficient of the Li-ion (D Li+ ). 25 The critical step in this analysis was the ability to produce nanowire-based electrodes with controllable particle-diameter, length, and density. This was accomplished by template-synthesis. 69 Template-synthesis is a general nanofabrication method, pioneered in the Martin laboratory, capable of creating nanostructures of gold, 8,9 carbon, 12 semiconductors, 14 polymers, 6 and Li-ion battery electrodes. 20,22,24 Since we identified the decrease in D Li+ as the rate-limiting factor in low-temperature environment, 25 the next logical progression is to quantify that decrease. Aurbach et al. 70,71 have demonstrated the ability of the potentiostatic intermittent titration technique (PITT) to measure the value D Li+ in Li-ion battery electrodes. This experimental method is based on a series of small potential steps and data analysis via the Cottrell-relationship. This was first described in this embodiment by Huggins and coworkers. 72 This method results in the systematic discharge (intercalation) of a Li-ion 42

PAGE 56

43 battery cathode, such as V 2 O 5 Lithium is known to reversibly intercalate 1 mol of charge into an equivalent of V 2 O 5 host over the potential window 2.8 to 3.8 V vs Li/Li + as described by Delmas et al. 68 according to the reversible intercalation reactions of Equation 2-1. One of the complications identified in prior studies is the ability to create structures of reproducible diffusion distance. 70 Template-synthesized nanostructured V 2 O 5 electrodes are again an ideal tool for this analysis, due to the reproducibility of the nanowires solid-state diffusion distance (a factor closely related to the analysis) and their ability to equilibrate uniformly after an electrochemical perturbation. The electrochemical response of these template-synthesized electrodes is not as subject to the interferences of conventional systems, such as concentration-polarization of the Li-ion 25 and the effects of conductive carbon and polymeric binders 50 (absent in this system). This chapter reports our efforts to quantify this decrease in solid-state diffusion-coefficient of the Li-ion with decrease in system temperature. Electrode Synthesis Template-synthesis was used to fabricate V 2 O 5 electrodes. This process has been detailed elsewhere. 25 A 3 cm 2 section of commercially-available polycarbonate template (Poretics) was placed atop a section of Pt foil. A 1 L drop of triisopropoxide vanadium (TIVO) precursor impregnated the pores of the template from the top. This was performed in an argon environment to prevent hydrolysis of the precursor prior to complete filling the pores. This assembly is then transferred to a low humidity environment (antechamber of the glovebox) and hydrolysis proceeds slowly over 8 hours. The particle-growth is restricted by the pore wall of the template; therefore, a nanoporous

PAGE 57

44 template yields an array of nanoscale diameter wires and microporous template yields microscale diameter wires. The templates used here are polycarbonate. The nanoporous template has nominally 50-nm diameter pores, 6 microns in length with a density of 6 x 10 8 pores cm -2 The microporous template has nominally 0.8-m diameter pores, 9 microns in length with a density of 3 x 10 7 pores cm -2 In order to ensure the micropores are completely filled, the filling procedure is repeated. 25 The assembly is then heated to 80 o C for 2 hours in air to ensure complete hydrolysis. Oxygen plasma (20 Pa, 25 W, 2 hours) preferentially etches the organic template, leaving structures that have the same geometry of the pores. These structures extend from the Pt current collector like the bristles of a brush. This assembly is then heated in flowing O 2 for 10 hours to convert the V 2 O 5 into its crystalline form. Note there is no conductive carbon or polymeric binder used in this synthesis. Electrolyte Synthesis Li-ion battery electrolytes are based on non-aqueous solvents. The primary reason is that the available potential envelop is greatly extended in such solvents. Aqueous-based systems are limited to ~ 1.2 Volts as not to engage undesired reactions, i.e., the splitting of water. The solvent system is a mixture of aprotic carbonates. Ethylene carbonate (EC) has a high dielectric constant; therefore, it ensures that the LiClO 4 dissociates. However, EC is a solid at room temperature and must be gently heated to melt prior to use. Therefore, diethyl carbonate (DEC) is added to decrease the viscosity of the system; which greatly improves ionic conductivity. Dimethyl carbonate (DMC) is added in proportion with diethyl carbonate in low temperature experiments, because of its ability to maintain ionic conductivity at low temperatures. These solvent systems are

PAGE 58

45 relatively standard in Li-ion battery research. They are dry and it is imperative that they remain dry throughout the experiment. The salt used for all of these experiments is LiClO 4 There are several examples in the literature of alternative counter-ions. Perhaps the most common seen in commercial-systems is LiPF 6 Others include LiBF 4 and LiAsF 6 We selected LiClO 4 because it is much easier to handle than the competing salts. LiPF 6 for example, is much more hydroscopic and hazardous than LiClO 4 and has only slightly improved conductivity. The LiClO 4 is heated on a hot plate in the argon glovebox to aid the electrolyte to remain dry. Measurement of Diffusion Distance As stated previously, a primary advantage of using template-synthesized Li-ion battery electrodes for this analysis is the ability to create structures of reproducible solid-state diffusion distance. We have used scanning electron microscopy (SEM) to verify the geometry of our structures. Figure 3-1A is a low-magnification image that demonstrates the uniformity of the structures. Figure 3-1B is a high-magnification image of these structures to identify the solid-state diffusion distance (l), determined by the radius of the nanofiber. This is a valid assumption since these wires are equally accessible to radial Li-ion intercalation and their aspect ratio is high (i.e., the surface area available for radial diffusion vastly exceeds that available for axial diffusion). This value is 50 nm. The uniformity of diffusion distance of the particles should provide a single characteristic diffusion time-regime, where all are subject to semi-infinite diffusion at the same times.

PAGE 59

46 Figure 3-1: Scanning electron micrograph. A) This low magnification image shows the uniformity over the large number of wires that constitute the electrode. B) This high magnification image is used to identify the solid-state diffusion path-length (radius of a nanowire) to be 50 nm.

PAGE 60

47 Electrochemical Investigations Potentiostatic Intermittent Titration Technique (PITT) Each electrode is characterized in a three-electrode jacketed cell. Temperature is set and maintained by a cooling circulation bath. Lithium ribbon is used as both the reference and counter electrodes. The electrolyte is 1 M LiClO 4 in an equivolume solvent mixture of ethylene carbonate, diethyl carbonate, and dimethyl carbonate (EC:DEC:DMC.) The PITT electrochemical experiments begin with the electrode completely in the charged (as synthesized) state. A -15 mV potential step polarized the electrode for the period of 45 minutes. This process was repeated over the entire range of 3.8 to 2.8 Volts versus Li/Li + The temperature is then reduced to the next level. A slow linear potential sweep (0.1 mV s -1 ) is used to return the electrode to the charged-state after 3 hours has elapsed for temperature equilibration. The experimental potential-step series was repeated. The cyclic voltammetric experiments are performed in an identical system with a scan rate of 0.1 mV s -1 A typical electrochemical response (current peak and decay) after the negative potential step is shown in Figure 3-2A. Since the electrode material is synthesized in the delithiated (x = 0) charged state, the potential steps are negative (E = -15 mV), so as to intercalate Li + with every step. Therefore, the currents are negative as well. Note that a disproportionate quantity of charge is intercalated at the potentials that correspond to the voltammetric peaks, as expected. After the 45 minute duration of the polarization, the currents have reached a small, but non-zero, steady state value. Preliminary analysis of these data follows the work of Aurbach et al. 70 In that experiment, they identified distinct kinetic regions during the charge reaction of anodic graphite by examining the It 1/2 versus log t plot.

PAGE 61

48 Figure 3-2: Electrochemical response of nanostructured V 2 O 5 electrode to a E = -15 mV potential step. A) Current-time transient recorded for 45 minutes as the current decayed to a steady-state value. B) Current-time transient shown in Cottrell-plot form. The plateau represents the Cottrellian value It 1/2

PAGE 62

49 Figure 3-2B represents the current-time transient (Fig. 3-2A) as a similar It 1/2 versus log t plot. This method conveniently identifies the solution to the Cottrell relationship 73 (i.e., where semi-infinite linear diffusion plays a major role), as it is the plateau of the plot. This particular plateau (3.200 V) lasts approximately 20 s in duration and is identified on the figure. The kinetic regimes 70 are clearly seen and correlate to 1.) the time-constant associated with the double-layer capacitance and the charge-transfer resistance (the short-time linear region prior to the plateau) 2.) the solution to the Cottrell relationship, where the value of It 1/2 is independent of time (the plateau) and 3.) the transition into the finite diffusion regime, where enough time has elapsed that the diffusion layer has propagated to encompass the entire nanowire [ )2(Dt ] (The long-time linear region post to the plateau.) Figure 3-3 demonstrates the potential-dependency of the time-region associated with this plateau. As the potentials approach that of the voltammetric peaks the plateaus shift to longer times and more negative values. This data can be easily converted to the solid-state diffusion coefficient of the Li-ion by means of Equations 3-1 and 3-2. Equation 3-1 establishes tau () as the characteristic diffusional time constant. Equation 3-2 expresses the simple relationship of this time constant to the solid-state diffusion coefficient of the Li-ion (D Li+ ). 22/12ItQ (3-1) 2LiD (3-2)

PAGE 63

50 Figure 3-3: Cottrell-plots as a function of potential and thus level of intercalation. At the potentials corresponding to the voltammetric minima, the Cottrellian value It 1/2 is shifted to more negative values and occurs at longer times. Again, It 1/2 is determined from the Cottrell-relationship plot (the ordinate value at plateau) (see Figure 3-2B); Q represents the quantity of charge intercalated per step (this value is determined via a digital integration of the I-t response using the CorrView software package); l is the characteristic diffusion path length (nanowire radius). The value for Q is adjusted to account for the effect of background currents. Figure 3-4 shows this value of Q at each E for the room temperature experiment.

PAGE 64

51 Figure 3-4: Charge intercalated per potential step. The majority of charge is intercalated at the potentials corresponding to voltammetric minima and galvanostatic plateaus. Determining D Li+ (Data analysis) Figure 3-5 clearly shows the dependency of D Li+ on T and E. Note that the y-axis (D Li+ ) is plotted on a log scale. First, let us discuss the value of D Li+ at room temperature (blue curve.) This curve exhibits similar dependence of D versus E seen by several research groups. 71,74-77 Minima are observed at potentials of the voltammetric peaks and are attributed to the effects of ion-ion interaction on the energy of activation. 70,78 This is intuitive as the movement of ions between intercalation sites becomes retarded in the presence of more of these species. Also of note, the particular values are within the approximations made by other groups. 75,77

PAGE 65

52 Figure 3-5: Solid-state diffusion coefficient as a function of potential. Y1-axis is the solid-state diffusion coefficient and the Y2-axis corresponds to characteristic diffusion time constant () at room temperature. Note that both y-axes are plotted on a logarithmic scale. Previously, we have seen less charge delivered from template-synthesized nanostructured electrodes as the temperature is decreased. 25 We attributed the decrease to a decrease in D Li+ Comparing data collected at room temperature (blue) and those at 0 o C (orange), we see this is indeed the case. When comparing these curves, we notice the values of the orange curve are less than the blue curve in every instance. The minima are again observed and again correlate to the potentials of the voltammetric peaks. The minima of the low temperature curve are shifted slightly less positive of those observed with the room temperature curve. This is attributed to sluggish electrokinetics and is discussed in detail later.

PAGE 66

53 Figure 3-6: Dependence of D Li+ conducted at 25 o C (Blue squares), 0 o C (Orange circles), and -20 o C (Green triangles). Figure 3-6 shows the trend continues in a comparison of the 0 o C curve to the -20 o C (green) curve. Again the decrease in D Li+ is seen, particularly at the low-potential region. The low-potential D Li+ -minimum at -20 o C (2.2 x 10 -15 cm 2 s -1 ) is an order of magnitude lower than the minima at 25 o C (2.3 x 10 -14 cm 2 s -1 ). These data confirm our previous conclusions that the rate-limiting factor of these electrodes at low-temperature is the decrease in solid-state diffusion coefficient. Again, the minima potentials are more negative than those at 0 o C. Activation Energy of Diffusion Process Energy of activation represents the thermal energy barrier that must be overcome for a process to proceed. Equation 3-3 is the general form of the Arrhenius equation that

PAGE 67

54 describes the relationship between the rate constant k and the temperature T. Equation 3-3 is the form of the general Equation 3-4 and conveys the relationships of our data to E A RTEAkA/exp (3-3) TREADALi1/lnln (3-4) As stated earlier and clearly observed in the figures above, the value of D is dependent on potential. We selected the D at potential of 3.24 V for this activation energy analysis, because it is between the two minima at every temperature, so the electrodes are in comparable states. The slope of this graph (Figure 3-7) is -5700 K and represents the E A / R value shown in Equation 3-4. Therefore, we can assign E A a value of 690 J. This represents the thermal energy barrier associated with the solid-state diffusion of the Li-ion. Figure 3-7: Arrhenius plot to solve for activation energy of diffusive process. The slope of this line represents E A /R. All values of D were taken at E = 3.24 V. At this potential the intercalation is between phases of intercalation (see text).

PAGE 68

55 Cyclic Voltammetry To further bolster this conclusion, we performed cyclic voltammetry (CV) experiments at all three temperatures with nanowires and microwires. In this experiment, the parameter E pk is the difference in the potentials at which the peak currents occur during the charging and discharging reactions. This parameter provides information concerning the electron-kinetics of the reaction (Equation 2-1). 79 A typical CV of a 0.8-m electrode is shown in Figure 3-8A. The data for the peak separation at the low-potential (E pk 1) for the nanowire and microwire electrode at the three different temperatures are presented in Figure 3-8B. These data show a correlation between the decrease in temperature and an increase in E pk Also from these data, the E pk for the nanowire and microwire are very similar at each temperature. Therefore, we can state that the electron-kinetics of this process does decrease with temperature; however, it is not the dominant effect, as the decrease is virtually independent of these electrode geometries.

PAGE 69

56 Figure 3-8: Cyclic voltammetric responses. A) Typical cyclic voltammetric response of 0.8-m electrode at room temperature. Scan rate = 0.1 mV s -1 B) Comparison of E pk 1 from cyclic voltammograms of nanoand micro-structured electrodes at various temperatures.

PAGE 70

57 Chapter Summary In this chapter, we demonstrated the measurement of solid-state diffusion coefficient of Li-ions (D Li+ ) as a function of potential and temperature. This is accomplished using the potential intermittent titration technique (PITT). PITT acquires data from small potential steps and analyzes data via the Cottrell relationship. Template-synthesis created electrode particles of reproducible and determinable solid-state diffusion distance. This method mitigates a critical data analysis issue identified by other research groups. D Li+ was seen to vary greatly with potential. This potential represents various states of charge (Li-ion intercalation level) of the electrode. Minimas were observed in the potentials close to the voltammetric peaks. These peaks represent the potential at which the majority of charge is actively intercalating causing ion-ion interactions to decrease the diffusitivy of the ions. D Li+ decreases over the range of potentials with temperature. We performed these experiments at five different temperatures ranging from room temperature to -20 o C. This generates the data required to identify the activation energy for the diffusion process by an Arrenheius relationship. We also report the results of cyclic voltammetric experiments where E pk is seen to be dependent on temperature, but independent of electrode-geometry (solid-state diffusion distance and electrode-particle area per template-area. Since E pk is a kinetic parameter, kinetics is again identified as a secondary contributor to the decrease in rate capabilities discussed in detail in the previous chapter. In discussions on the research contained in this chapter, the validity of this PITT method has come under debate. The Cottrell equation is a globally accepted and

PAGE 71

58 implemented method to determine the diffusion coefficient of small particles in aqueous systems diffusing in the semi-infinite regime to a planar electrode. The debate arises due to the uniqueness of this solid-state intercalation process. This method is analogous to ion diffusion through a polymer-coated electrode. Research of this type (e.g., Nafion-coated electrodes) gained popularity with the last generation of scientists. Collectively they describe this parameter as an apparent diffusion coefficient. Debate ensues on the spurious nature of the relationship of diffusion coefficient to voltammetric peak potentials. Many researchers believe this to be an artifact of the application of the Cottrell method to the battery intercalation mechanism. 80 It would be remiss to not mention this debate in a discussion of this method. At the time of publication of this dissertation, the debate continues. 80,81

PAGE 72

CHAPTER 4 A HIGH-RATE, NANOCOMPOSITE LIFEPO 4 /CARBON CATHODE Introduction Lithium-ion batteries are the power source of choice for portable electronics, a multi-billion dollar market. 1 This outstanding commercial success has spawned great international interest in applying this technology to systems that demand higher power, such as the electric component of hybrid vehicles. 3 This would require new electrode materials that are less expensive, more energetic, and more environmentally friendly than the present ones. Of particular interest is the olivine-structured LiFePO 4 cathode developed by Goodenough and co-workers, 82 which offers several appealing features, such as a high, flat voltage profile and relatively high theoretical specific capacity (168 mAh g -1 ), combined with low cost and low toxicity. However, the current designs of cells based on LiFePO 4 technology have not shown the ability to deliver high specific capacity at high discharge rates. For this reason, LiFePO 4 is currently not a promising electrode material for high-rate and pulse-power applications. The discharge reaction for LiFePO 4 (Equation 4-1) entails intercalation of Li + (from the contacting electrolyte phase) along with an equivalent number of electrons into the electrode material. The rate capabilities of LiFePO 4 are limited primarily by its intrinsically poor electronic conductivity and by the low rate of Li + transport within the micron-sized particles used to prepare the battery electrode. A number of approaches have been proposed to improve this materials inherent poor electronic conductivity, 59

PAGE 73

60 including carbon coating, 83 nano-fibril textures, 84 optimized synthesis procedures 85 and foreign metal doping. 86 4-1 We describe here a new approach for preparing high rate-capability LiFePO 4 electrodes. This approach builds on the application of the template synthesis method for preparing nanofiber Li-ion battery electrodes. 17,19,20,22,24 However, the method was modified such that the template-prepared LiFePO 4 nanofibers are mixed with carbon particles, and coated by thin carbon films, to yield a nanocomposite LiFePO 4 /carbon matrix. As we have shown previously, 19,20,22 the nanofiber morphology mitigates the slow Li + -transport problem, because the distance Li + must diffuse within the electrode material is minimized. The carbon matrix obtained with this new template-based method obviates the poor electronic conductivity problem. These nanocomposite LiFePO 4 /carbon electrodes can deliver a capacity of 150 mAh g -1 at a rate of 5 C and maintains a substantial fraction of the theoretical capacity even at rates exceeding 50 C. To our knowledge, performance at this level has never been achieved by other types of LiFePO 4 Electrode Synthesis The sol-gel method developed by Croce et al. was employed for the synthesis of the electrode precursor solution. 87 Accordingly, LiOH monohydrate (Aldrich), ferric nitrate nonahydrate (Fisher), ascorbic acid (Fisher), phosphoric acid, and ammonium hydroxide were used to create the LiFePO 4 precursor solution. The template membranes were

PAGE 74

61 commercially available polycarbonate filters (Poretics). Pt foil (2.5 x 1.5 x 0.025 cm, 99.99% purity, Aldrich) was used as the current collector. LiClO 4 (Aldrich), ethylene carbonate (Aldrich), diethyl carbonate (Aldrich) were used as received for preparing the electrolyte. An approximately 1 cm 2 piece of the polycarbonate filter was immersed in a precursor solution of 1 M LiFePO 4 in water in 24 hours. This solution was synthesized with ferric nitrate, lithium hydroxide, and phosphoric acid in proportions for a 1:1:1 molar ratio. Ascorbic acid, in equimolar ratio to the total metal (Li + plus Fe 2+ ) content, aided the synthesis by forming a complex with the iron, and ammonium hydroxide was used to raise the pH to ~2. The impregnated template was then attached to a Pt current collector and dried in air at 80 o C for 10 minutes. A 10 L drop of precursor solution was placed on top of the dried filter to increase the amount of active material in the sample. It was dried again under the same conditions. This assembly, template intact, was heated in a reducing atmosphere of flowing Ar/H 2 gas (95/5 %). The temperature was slowly taken over the course of 4 hours from 250 o C to 650 o C and held there for 12 hours. This procedure yields the Fe(II) oxidation state necessary for LiFePO 4 and decomposes the template into the carbon necessary for improved conductivity. A schematic of this process is shown in Figure 4-1. In all of our previous examples of template-synthesized nanofiber electrodes 19,20,22 after synthesis of the nanofibers the template would be totally removed to yield the nanofibers protruding from an underlying current collector surface like the bristles of a brush. For the LiFePO 4 nanofibers prepared here we instead pyrolyzed the polycarbonate in a reducing Ar/H 2 environment at a temperature of 650 o C. This yields graphitic carbon

PAGE 75

62 particles intimately mixed with the LiFePO 4 nanofibers, and thin carbon films that coat these fibers. Figure 4-1: Schematic of template-synthesis of LiFePO 4 /carbon nanocomposite electrode. Two different templates were used. A template with a slightly larger pore-diameter of 100 nm is used for microscopy. A template of 50 nm pore-diameter is used for electrochemistry. The template being used for each experiment is clearly stated. Structural Investigations Scanning Electron Microscopy A FEG SEM JEOL JSM 6335F instrument was used to obtain SEM images of the electrode. The nanostructured electrodes on Pt foil were prepared for imaging by attaching them to a SEM stub by conductive copper tape. No conductive metal sputtering was required for the composite electrodes, but a thin Au/Pd sputtering was applied to the bare LiFePO 4 wires (Figure 4-2C) prior to imaging.

PAGE 76

63 To obtain detailed images of the morphology of the nanocomposite structure we found it prudent to use a template with nominally 100 nm-diameter pores. The larger LiFePO 4 nanofibers obtained from this template are more easily imaged with scanning electron microscopy. This template was also 6 m thick and had a pore density of 4x10 8 pores per cm 2 A lower magnification scanning electron microscopic (SEM) image of the resulting LiFePO 4 /carbon nanocomposite electrode is shown in Figure 4-2A. Because of the relatively low porosity of the template, there is substantial void volume, but in analogy to our prior nanofiber electrodes of this type, the LiFePO 4 nanofibers can be seen crossing through this void space. 19,20,22 Higher magnification images (Figure 4-2B) show that there are carbon particles dispersed through this matrix and that the LiFePO 4 nanofibers are coated with thin carbon films. To prove that these fibers are coated with carbon films, we prepared fibers in the same template, but instead of then pyrolyzing the template, we simply removed it quantitatively by burning it away in O 2 plasma. Hence, in this sample the fibers are not coated with carbon films. 19,20,22 An image of these fibers is shown in Figure 4-2C. The fibers from the pyrolyzed membrane (Figures 4-2A and B) have a textured surface morphology and have a larger diameter than the fibers from the plasma-removed membrane (Figure 4-2C). Both the larger diameter and the textured surface are due to the carbon coating surrounding the fibers from the pyrolyzed membrane.

PAGE 77

64 Figure 4-2: Scanning electron micrographs. A) Lower magnification image of the nanocomposite LiFePO 4 /carbon electrode. B) Higher magnification image of the nanocomposite LiFePO 4 /carbon electrode. Composite fiber diameter is 350 nm. C) Image of LiFePO 4 electrode synthesized by template dissolution method (absent of carbon). LiFePO 4 fiber diameter is 170 nm.

PAGE 78

65 X-ray Diffraction Studies (XRD) The X-ray diffraction data are presented in Figure 4-3. The pattern represents the diffraction of LiFePO 4 /carbon on quartz. The quartz substrate was used because the Pt foil would give a large signal in this region. This signal would dwarf the relative signal from the LiFePO 4 /nanocomposite. The lines below correspond to the accepted literature values for LiFePO 4 olivine group, triphylite subgroup, published as JCPDS 40-1499. The large amorphous wave ranging from 15 to 30 2 is characteristic of disordered carbon. Our group has seen similar diffraction patterns when working with disorderd carbons. Figure 4-3: XRD pattern from nanocomposite, microcomposite, compared to accepted literature values for LiFePO 4 The substrate is quartz. Acquisition time is 10 s. X-ray Photoelectron Spectroscopy The XPS studies were performed on a Kratos XSAM 800 spectrometer with Al-K exitation (180W). The sample, mounted onto a stainless steel sample stub, was inserted into the sample analyzer chamber by means of a quick insertion probe, and spectral

PAGE 79

66 acquisition commenced after the pressure decreased to 5x10 -9 Torr. High resolution C1s spectra were recorded at a take-off angle of 75 relative to the sample surface. Data analysis was done by using the DS 800 software package. Peak positions were all referenced to 70.9 eV for the Pt4f 7/2 peak (literature value for metallic platinum used as sample support). The presence of carbon was also confirmed by XPS analysis (Figure 4-4). The high resolution C1s spectrum may be fitted to three peaks with binding energies of 283.2 0.5, 284.7 0.3 and 286.0 0.3 eV. According to previous work by the Martin laboratory 11 the lowest binding-energy peaks may be assigned to graphitic (283.2 eV) and amorphous (284.7 eV) carbon. The predominance of the 284.7 eV peak indicates that most of the carbon present is amorphous; this is confirmed by the X-ray diffraction data. The peak at the highest binding-energy (286.0 eV) is due to oxygen-containing surface functional groups. Oxygen functional groups are nearly always observed on carbon surfaces that have been exposed to air. 11

PAGE 80

67 Figure 4-4: X-ray photoelectron spectroscopy data for C1s peak for the carbon in the nanocomposite electrode. Relative peak areas show amphorous form is dominant (~80%), but graphitic carbon is also present (~20%). Analysis of Carbon Content We estimated the weight percent of carbon in the composite electrode gravimetrically using an Ultra-Micro-Balance SC2 (Satorius). LiFePO 4 nanofibers were synthesized within the pores of the polycarbonate template on a Pt current collector, and the polymer was pyrolzyed as described previously. The mass of this composite, corresponding to the masses of the Pt current collector, the nanofibers and the carbon, was obtained. This composite was then heated in air at 600 o C for 30 minutes to burn off the carbon, and the mass was measured again. The difference between these two masses is the mass of carbon in the composite. Replicate analyses on four identically prepared samples gave a carbon content of 7 + 4 % in the nanofiber/carbon composite.

PAGE 81

68 The presence of carbon in this matrix was confirmed by X-ray diffraction analysis, Raman spectroscopy, and X-ray photoelectron spectroscopy (XPS). A Raman spectrum (data not shown) of the nanocomposite LiFePO 4 /carbon electrode was also taken. The most intense peak (centered around 1000 cm -1 ) and others at the lowest frequencies correspond to the PO 4 stretching modes of LiFePO 4 Bands at 1350 cm -1 and 1580 cm -1 are assigned to carbon. 88 Electrochemical Investigations For the electrochemical studies we wanted a template with small-diameter pores so that correspondingly small-diameter nanofibers of LiFePO 4 would be obtained. For this reason we used a template with nominally 50 nm-diameter pores for the electrochemical studies. This template was 6 m thick and had a pore density of 6x10 8 pores per cm 2 of surface area. Cyclic voltammetric and constant current charge/discharge experiments were performed in a three-electrode cell using a Solartron 1287 Potentiostat, driven by the CorrWare software package. The electrolyte for these experiments was 1 M LiClO 4 in ethylene carbonate:diethyl carbonate (3:7 v/v). The nanostructured LiFePO 4 was the working electrode and lithium ribbon was the reference electrode and counter electrode. Potentials are reported versus the Li/Li + reference. The experiments were performed in the inert atmosphere of a glovebox filled with argon gas. This unique LiFePO 4 /carbon nanocomposite electrode should be ideally suited for high rate applications because the distance that Li + must diffuse in the electrode material is limited to the radius of the nanofibers 19,20,22 and because the carbon matrix should provide for good electronic conductivity through the composite. This was confirmed experimentally via electrochemical characterization of these electrodes. Again, the

PAGE 82

69 electrochemical studies were done on electrodes prepared in templates with 50 nm-diameter pores. Cyclic Voltammetry Cyclic voltammograms (CVs) for the nanocomposite electrode show reversible waves centered at 3.5 V associated with the reduction and re-oxidation of the LiFePO 4 (Figure 4-5A). 87,89 The difference in peak potentials (E pk ) for this nanostructured electrode is 60 mV (Figure 4-5A). This may be contrast to CVs for conventional, non-nanostructured, LiFePO 4 electrodes, which at comparable scan rates and in comparable electrolyte solutions show E p > 200 mV. 87,89 This clearly shows that our nanostructured LiFePO 4 electrodes lack a resistive component that is present in the conventional electrodes. This verifies the major premise of this work that the conductive carbon matrix overcomes the inherently poor electronic conductivity of LiFePO 4 Rate Capabilities The constant-current discharge curve (lithium insertion, Equation 4-1) for the nanocomposite electrode shows the flat voltage plateau centered at 3.5 V, characteristic of LiFePO 4 (Figure 4-6A) 83,87,89 At the lowest discharge rate used (3 C), the specific capacity for the composite is 165 mAh g -1 essentially identical to the maximum theoretical capacity, 168 mAh g -1 While capacity falls off with increasing discharge rate (Figure 4-6B) the electrode retains 36% of its theoretical capacity at discharge rates as high as 65 C. There are no other examples in the literature of LiFePO 4 being discharged at such enormous rates.

PAGE 83

70 Figure 4-5: Cyclic voltammograms. A) Cyclic voltammogram for the nanocomposite LiFePO 4 /carbon electrode prepared using a template with 50 nm-diameter pores. Scan rate = 0.1 mV s -1 B) Comparison of voltammetric response to theoretical response exhibiting diffusional tailing. The diffusional wave was generated from the normalized function. 79

PAGE 84

71 Figure 4-6: Constant current experiments. A) Constant-current (3 C) discharge of LiFePO 4 /carbon nanocomposite electrode. B) Specific capacity versus C-rate for the nanocomposite LiFePO 4 /carbon electrode prepared using a template with 50 nm-diameter pores.

PAGE 85

72 Chapter Summary We have described here a new type of template-prepared nanostructured LiFePO 4 electrode. LiFePO4 was developed as an alternative cathode material to LiCoO2, the industry standard. LiFePO4 is an energetic material with advantages of both cost and safety. However, this material is inherently electronically insulating. This material is also susceptible to the concentration polarization with larger micron-diameter particles. Our nanocomposite electrode consists of nanofibers of the LiFePO 4 electrode material mixed with an electronically conductive carbon matrix. It is created by a modified template-synthesis procedure, where the polymeric template is pyrolzed. This carbon matrix coats the LiFePO 4 nanostructures and provides an improved electron pathway. Therefore, both rate-limiting effects of electron-transfer and ion-transport are delayed. This unique nanocomposite morphology allows these electrodes to deliver high capacity, even when discharged at extreme rates necessary for many pulse-power applications. This nanocomposite is a specific example of nanotechnology overcoming the limits of conventional technology. We are currently working toward developing a commercially viable route for preparing such nanocomposite electrodes.

PAGE 86

CHAPTER 5 MAGNESIUM-ION INTERCALATION INTO TEMPLATE-SYNTHESIZED NANOSCALE ELECTRODES IN THE ABSENCE OF CARBON Introduction Li-ion batteries have powered the mobile electronic industry. Though these energy-storage devices have enjoyed tremendous commercial success, they remain the subject of international research. This is primarily due to the consumer driven demands of an increasingly portable world, specifically, the desire for increased pulse-power and decreased size. The Li-ion intercalation reaction (mechanism by which the host-electrodes store charge) is shown previously. Lithium-ions were chosen because of their low reduction potential (-3.05 vs. NHE) and their small dimensions facilitate both the intercalation process and the solid-state diffusion process. We have had great success synthesizing and characterizing Li-intercalation into template-synthesized nanostructured electrodes. Template-synthesis is a general nanofabrication method pioneered in the Martin group. Employing this method, we have shown that with no addition of inactive ancillary components (carbon or polymeric binder), template-synthesized nanostructured electrodes are capable of reversibly storing and delivering a dramatically greater portion of the electrodes theoretical capacity (mAh g -1 ) than similarly constructed microstructured electrodes. These nanostructured electrodes are capable of this, due to their small solid-state diffusion distance and large surface-area per gram. These traits serve to delay the effects of concentration polarization and sluggish electron-kinetics, respectively. 73

PAGE 87

74 Dramatically fewer researchers have explored the field of polyvalent-ion intercalation. 90-92 The charge-storage mechanism for this device is similar to that of the Li-ion battery; however, whereas the Li-ion intercalation reaction has 1 electron equivalent per Li-ion each Mg-ion has 2 electrons. Simply stated, by doubling the quantity of electrons available, one can store an amount of specific energy that is not stoichiometrically possible via the Li-ion intercalation process. However, the diffusivity of these ions is further hindered by its polyvalent nature and larger dimensions. Therefore, a structure that has proven to be able to delay the limiting effects of such phenomena may be able to successfully intercalate such ions. Template-synthesized V 2 O 5 nanostructured electrodes have, in fact, this ability. 25 Here we detail these studies and demonstrate the ability to reversibly (de)intercalate Mg 2+ into/out of the host (template-synthesized V 2 O 5 nanostructures) by Equation 5-1. 5-1 Other advantages of Mg-based technology are that concerns still exist with the safety and stability of lithiated carbon as well as the expense of the constituents of the battery. Electrode Synthesis As stated, these electrodes were created by the template-synthesis method. This process is similar to those described previously; however, the notable exception is that the current collector is Indium Tin-Oxide (ITO)-glass. Another research group has demonstrated the usefulness of the substitution. 91 This is a piece of glass that has a thin layer of conductive Indium Tin-Oxide deposited onto it. Therefore, one of the surfaces of

PAGE 88

75 the glass is electronically conductive. Electrodes were synthesized on both ITO-glass substrates as well as platinum. A commercially available (Poretics) polycarbonate filter was used as the template. This membrane has nominally-cylindrical track-etched pores (~ 50 nm in diameter) that run its entire length of 6 microns. There are 6 x 10 8 of these pores per cm 2 of template area. In an inert environment (Ar-filled gloebox) a 3 cm 2 section of template is placed atop a section of electrically conducting ITO-glass. A 1 L volume of TIVO (triisopropoxide vanadium) was placed atop the template membrane. The precursors low viscosity allows it to flood the pores of the filter. This assembly is then moved to a low water environment (antechamber of glovebox). In the presence of the atmospheric water, the TIVO precursor hydrolyzes. At this point, the assembly is heat to 80 o C for 2 h in air to ensure complete hydrolysis and conversion to the gel-phase. Any top surface layer is then removed with a damp cotton swab. The assembly is then placed into oxygen plasma (20 W; 10 Pa; 2 hours) to etch away the organic template. The resulting nanowires extend from the ITO-glass (or platinum) communal current collector and mirror the geometry (length, diameter, and number density) of the pores of the membrane. The electrode is then heated to 400 o C in flowing O 2 for 10 hours to form crystalline V 2 O 5 nanowires. XRD Studies X-ray diffraction is a convenient experiment to identify the crystalline phase of the electrode material. Like SEM, it is non-destructive. In a powder x-ray diffraction (XRD) experiment, a quantity of powder-form sample is exposed to X-rays at a sweeping angle. These incident X-rays deflect off the sample onto a detector at a pattern relative to the

PAGE 89

76 crystalline lattice structure of the powder. The arrangement of the atoms in a relationship to each other can be determined from the pattern and Braggs law. It is displayed in terms of Miller indices (hkl), coordinates of parallel planes in a unit cell. Figure 5-1: X-ray diffraction experiment. The sample is nanostructured V 2 O 5 on Pt. Acquisition time =10 s. The discrete lines represent the accepted values for orthorhombic phase V 2 O 5 (JCPDS 41-1426). This continuous line is the response of a nanostructured V 2 O 5 electrode. The discrete lines represent the internationally accepted pattern for orthorhombic V 2 O 5 as per JCPDS card 41-1426, which replaced card 9-387. These two patterns are similar, with only the relative intensity of the peak at 226.1 (representing the (110) plane) being smaller than the standard. The 2 range was limited to 15 to 35, so as to not record the intense Pt peak at 39.8, which would dwarf the relative intensities of the V 2 O 5 sample. An acquisition time of 10 s per step was used to increase resolution of the pattern.

PAGE 90

77 The XRD pattern of a powder can also be used to estimate the crystallite size of the sample. This relationship is described in the following Equation 5-2 first described by Scherrer 93 and refined by Biscoe and Warren. 94 BBtcos**89.0 5-2 The term t is the size of the crystallite. The term is the wavelength of the incident X-rays (Copper = 1.54 Angstroms). The B is the Braggs angle being analyzed. The term B represents the full-width half-max of the crystallographic peak in radians. Here the data from the most intense 001 plane is compared. Using this equation, the nanostructured electrode has an average crystallite size of 20.1 nm; whereas, the microstructured electrode (data not shown) has an average crystallite size of 32.3 nm. So, qualitatively the particles of the micronstructured electrode are lager on a crystallographic plane than the nanostructure electrode. This also indicates that each wire may consist of a number of smaller crystallites. Electrochemical Investigations The electrode was characterized in a three-electrode half-cell configuration. The cell differs significantly from the ones discussed earlier. The V 2 O 5 nanostructures on ITO-glass (or platinum) were the working electrode. Lithium must be excluded from the cell to ensure that all charge-storage is due to Mg 2+ Polished Mg ribbons were used as the reference and counter electrodes. The electrolyte is 1 M Mg(ClO 4 ) 2 in dry acetonitrile. This system been used by other researchers in this field. 91 This cell was assembled and characterized in the Ar-filled glovebox.

PAGE 91

78 Ferrocene Pseduo-Reference As stated previously, Li-ions must be excluded from the electrochemical system to ensure that the response is strictly from the Mg-ion intercalation. Therefore, the reference electrode in this experiment is Mg/Mg 2+ This reference reaction is not as common, especially in this particular electrolyte system. We use the well established ferrocence / ferrocenium redox couple as a pseduo-reference. The first system has lithium ribbon as the counter and the reference electrode. Platinum metal is used as the working electrode. The electrolyte is 1 M LiClO 4 in EC:DEC (3:7). Ferrocence is also dissolved into the solution. A cyclic voltammogram is taken using this system. Figure 5-2 shows the potential of the reversible ferrocence redox reaction to be 3.25V vs. Li/Li + Figure 5-2: Cyclic voltammogram of ferrocence in Li-based and Mg-based systems (see text). Scan rate = 50 mV s -1

PAGE 92

79 The comparative system has magnesium ribbon as the counter and the reference electrode. Platinum metal is again used as the working electrode. The electrolyte is 1 M Mg(ClO 4 ) 2 in dry acetonitrile. Ferrocence is also dissolved into the solution. A cyclic voltammogram is taken using this system. The potential of the reversible ferrocence redox reaction is 1.5 V vs. Mg/Mg 2+ The same redox reaction occurs in the Mg-based system at a potential that is 1.75 V negative of the potential at which is occurs in the Li-based system. Therefore, the Mg/Mg 2+ reference is 1.75 V negative of the Li/Li + reference (i.e., 0 V versus Mg/Mg 2+ is +1.75 V versus. Li/Li + ). Cyclic Voltammetry ITO-glass current collector The CV experiment (Figure 5-3) was performed at the scan rate of 0.1 mV s -1 The electrolyte is the Mg-based system described above. There is a peak, indicating a charge-storage process, during both intercalation and deintercalation scans. The most notable characteristic is the large peak separation associated with these peaks. Interestingly, this low-potential peak does represent the deintercalation of Mg 2+ because if the experiment is terminated at 1.5 Volts, then the high-potential peak is not seen again in the next scan (i.e., it becomes an irreversible reaction). To date, we are uncertain as to the source of this phenomenon. It appears too large to be iR drop, an effect of arising from the uncompensated resistance between the reference and working electrodes. For such small currents, the resistance would need to be on the order of 100 k. The conductivity of the solution was measured to be 25 S cm -1 This translates to a R uncomp of > 25 in this system. (The distance between working and reference electrode would need to be 4 m for

PAGE 93

80 this to be the source of the peak separation.) Also note that the R uncomp was estimated by Electrochemical Impedance Spectroscopy to be approximately 10 The effect is most probably related to either the vulnerability of the ITO layer during the thermal processing step or a resistive passivation layer. The poor thermal stability of ITO is undoubtedly an effect, because the mass of the ITO-glass post-heat treatment is significantly lower mass prior to the heat treatment. Other groups 91 were able to use ITO-glass as a substrate, because they apply the active electrode material post-processing (i.e., it is not necessary for the current collector to be compatible with the fabrication process of the material synthesis, as it is in our case). Figure 5-3: Cyclic voltammogram of ITO-glass/Mg x V 2 O 5 nanowire electrode. Scan rate = 0.1 mV s -1 X2-axis is calculated from data shown in Figure 5-1.

PAGE 94

81 Pt foil current collector For comparison, we also used our standard Pt foil as a substrate. The results of this experiment are shown in Figure 5-4. The current response during the voltammetric sweep is much greater for approximately the same sample mass. This represents the inertness during the synthetic process seen with Pt as opposed to ITO-glass. However, the large peak separation still exists. This suggests that this peak-separation is not dominated by interactions between the current collector and the electrolyte system. It is possible that a Mg-insulating passivation layer forms on the surface of the electrode. Other groups have also proposed this theory 90 Our electrodes created by template-synthesis may, in fact, be particularly susceptible to such limiting factors, due to the high surface area. This theory would explain the resistive nature of the current-response. Figure 5-4: Cyclic voltammogram of Pt/Mg x V 2 O 5 nanowire electrode. Scan rate = 0.1 mV s -1 The potential reference is Mg/Mg 2+

PAGE 95

82 Rate Capability We also characterized this ITO-glass/V 2 O 5 electrode using the galvanostatic method to charge and discharge the electrode. The potential limits used were 2.7 and 0.8 versus the Mg/Mg 2+ reference. While the ITO-glass/V 2 O 5 electrode has advantages during the electrochemical characterization, this system has a distinct problem in template-synthesis. As stated earlier, the standard method of Li-ion battery construction processes the active material of the electrode separately from the current collector. However, in the template-based method the electrode material is processed from cradle to grave with the current collector. The chemical and thermal inertness of platinum makes it a good choice. During the final heating stage of the synthesis, the ITO-layer degrades. This is easily seen by noting that the mass of an ITO-glass is significantly less after a similar heating process than before. Therefore, in Figure 5-5 can analyze the rate capability data using the electroactive mass, as we did with the LiFePO 4 We have taken data comparing rate-capabilities of nanostructured (50-nm template) microstructured (0.8 micron) and film electrodes. These data are presented in Figure 5-6. Each of these was analyzed for mass post-electrochemical characterization by dissolving the V 2 O 5 electrode material in 2 M H 2 SO 4 and analyzing for the V-ion by ICP-AES at 310 nm. Again, we observe an advantage in rate capabilities of the nanostructured electrode compared to the controls. However, there is markedly less specific capacity, when using the mass determined by ICP-AES.

PAGE 96

83 Figure 5-5: Comparison of rate capabilities of Li x V 2 O 5 and Mg x V 2 O 5 The mass for these calculations is the electroactive mass determined by integration of the respective cyclic voltammogram. Figure 5-6: Comparison of rate capabilities of nanoand micro-structured Mg x V 2 O 5 The mass for these calculations is the mass determined by ICP-AES analysis of the V-ion from post-electrochemical dissolution.

PAGE 97

84 Investigations of a Possible Mg-Sn Alloy There is evidence in the literature that Mg and Sn form an alloy. 95 Previously, we and others have demonstrated the superior charge-storage properties associated with a reversible Sn-Li alloying process. 20 We propose that if the Mg-Sn could form a reversible alloy by electrochemistry, then it may be an alternative anode for a Mg-based system. This technology is based on two interesting reactions. First, a sol-gel route is used to synthesize SnO 2 nanowires. These wires are electrochemically converted to Sn, which reversibly alloys with Li, thus storing charge. Figure 5-7 is the typical electrochemical response in the Li-system, which serves as a control for the synthetic process. The first wave is the electrochemical conversion to Sn; whereas, the reversible waves are attributed to the Sn-Li alloying process. Figure 5-7: Cyclic voltammogram for Sn-based electrode in Li-system prepared using a template with 50 nm-diameter pores. Scan rate = 0.1 mV s -1

PAGE 98

85 However, when the SnO 2 electrode is inserted into the Mg-based system, there is only a single reduction wave. We can attribute this to the conversion of SnO 2 to Sn, analogous to what was seen in the Li-system. As the potential is swept more negative, we fail to see the reversible waves of charge-storage. Therefore, we can conclude that where SnO 2 may successfully become Sn, Sn has trouble forming Mg x Sn by this electrochemical means. The apparent inability of this reaction eliminates it from consideration as an alternative anode system. Figure 5-8: Cyclic voltammogram for Sn-based electrode in Mg-system prepared using a template with 50 nm-diameter pores. Scan rate = 0.1 mV s -1

PAGE 99

86 Chapter Summary We were able to form several conclusions from this preliminary series of experiments. Our primary goal of demonstrating reversible intercalation of Mg-ions into V 2 O 5 nanowires without carbon or binder was achieved. We also showed that at a low discharge rate, more charge can be stored per gram in the Mg-based system than the Li-based system. This is from the rate capability experiment ITO-glass, using the electroactive mass obtained via voltammetric integration. Another observation is that at large discharge rates, the charge-per-gram stored by the Li-based system comes to exceed that stored by the Mg-based system. This suggests a decrease in diffusivity of the Mg-ion compared to the Li-ion. When diffusion profiles are given less time to propagate, this is more of a concern. ITO-glass was determined to be a satisfactory current collector because of its available potential window, but it is incompatible with this synthetic process. Data suggest that this results in only islands of material that is electronically addressable. Also, a comparison between the ITO-glass and the Pt systems reveals some background specific to the platinum. This may be a result of excess capacitance associated with the exposed back of the metal. This would not be a factor with the ITO-glass, because the exposed back of that substrate is not conductive. It is notable though, that this effect appears to be unique to this system, as is not dominant in previous studies.

PAGE 100

CHAPTER 6 CONCLUSIONS AND FUTURE DIRECTIONS Conclusions Li-ion batteries are the current power source for portable electronics. We are part of an intense, global effort to improve upon this existing design, in order to expand the applications of these batteries. We are specifically investigating the effects of nanomaterials on the design of Li-ion batteries. This dissertation details our use of nanostructured electrodes created by template-synthesis as tools in fundamental studies of these batteries. Chapter 1 is an introduction to the field of Li-ion battery research and nanofabrication method of template-synthesis. This chapter lays a framework for the dissertation and explains the significance of our research. In Chapter 2, we demonstrate the advantages of template-synthesis as it pertains to Li-ion batteries. The electrode precursor is deposited into the pores of a commercially available polycarbonate template membrane. The template serves to restrict particle growth as the precursor is hydrolysized to V 2 O 5 a known Li-battery cathode material. These V 2 O 5 wires extend from the surface of a current collector like bristles of a brush. The geometry of the pores of the template is imparted onto these wires a nanoporous template yields nanostructured wires and microporous template yields microstructured wires. We demonstrated the ability to create structures of reproducible diameters by this template-synthesis method. 87

PAGE 101

88 Three of these electrodes with different wire geometries were electrochemically charged and discharged by a constant-current. By creating particles of the shortest solid-state diffusion distance, we were able to demonstrate superior low-temperature rate-capabilities when compared to microstructured control electrodes. However, these superior discharge characteristics would be achieved with any competitive nanofabrication method. The ability of template-synthesis to create structures of reproducible solid-state diffusion distance and surface-area is what differentiates it from other nanofabrication methods. Analysis of these data identifies the decrease in solid-state diffusion coefficient of the Li-ion associated with the decrease in temperature to be the rate-limiting factor in Li-ion batteries. The conclusions presented in this chapter are a benchmark study in the field of low-temperature Li-ion battery research. Chapter 3 is derivative of the previous study, as we demonstrate the measurement of solid-state diffusion coefficient of Li-ions (D Li+ ) as a function of potential and temperature. This is accomplished using the potential intermittent titration technique (PITT). D Li+ decreases over the range of potentials with temperature. Using this value recorded at several temperatures, we are able to calculate activation energy for the diffusive process via the Arrehenius relationship. We also report the results of cyclic voltammetric experiments where E pk is seen to be dependent on temperature, but independent of electrode-geometry (solid-state diffusion distance and electrode-particle area per template-area. The key to this analysis is the ability of template-synthesis to create structures of determinable diameter, length and number density. In Chapter 4, we have described a new type of template-prepared nanostructured LiFePO 4 electrode. LiFePO 4 is a promising alternative cathode material, but, to date, its

PAGE 102

89 use is limited by its inherently poor electron-conductivity. The electrode described in this chapter is a composite consisting of nanofibers of the LiFePO 4 material mixed with an electronically conductive carbon matrix. This composite is created by a modified template-synthesis procedure, where the polymer template membrane is pyrolyzed. This nanocomposite electrode provides an intimate-contact mechanism between carbon and LiFePO 4 nanostructures. This encompassing carbon matrix provides an improved electron-pathway, while the nanostructured geometry minimizes Li-ion diffusion distance. Therefore, the composite assists on both rate-limiting factors. This unique nanocomposite morphology allows these electrodes to deliver high capacity, even when discharged at extreme rates necessary for many pulse-power applications. In Chapter 5, we were able to demonstrate the reversible intercalation of a polyvalent-ion into a Li-ion battery electrode. From this preliminary series of experiments, we can form several conclusions. Our primary goal of demonstrating reversible intercalation of Mg-ions into V 2 O 5 nanowires without carbon or binder was achieved. We also showed that at a low discharge rate, more charge can be stored per gram in the Mg-based system than the Li-based system. This is from the rate capability experiment ITO-glass, using the electroactive mass obtained via voltammetric integration. Another observation is that at large discharge rates, the charge-per-gram stored by the Li-based system comes to exceed that stored by the Mg-based system. This suggests a decrease in diffusivity of the Mg-ion compared to the Li-ion. When diffusion profiles are given less time to propagate, this is more of a concern. ITO-glass was determined to be a satisfactory current collector because of its available potential window, but it is incompatible with this synthetic process. Data

PAGE 103

90 suggest that this results in only islands of material that is electronically addressable. Also, a comparison between the ITO-glass and the Pt systems reveals some background specific to the platinum. This may be a result of excess capacitance associated with the exposed back of the metal. This would not be a factor with the ITO-glass, because the exposed back of that substrate is not conductive. It is notable though, that this effect appears to be unique to this system, as is not dominant in previous studies. Future Directions When given the forum to layout the future directions of a project, I first consider the goals of academic research. I perceive the goal of academic research in the pure sciences as the exploration of fundamental principles and the fostering of creativity. Approaching experiments and challenges by a strategy well founded in the scientific method ensures an understanding of underlying processes. It is the responsibility of the private sector to adapt the creative findings of academia to the marketplace. Graduate school is as much about the contribution of the field to the student as it is about the contribution of the student to the field. My journey through graduate school has instilled a tremendous respect for original thought. With these goals in mind, I believe that the existence and interplay of the physical and chemical phenomena of this complex system can be resolved Electrochemical Impedance Spectroscopy (EIS). EIS imposes a small amplitude sinusoidal AC potential onto a DC potential. The response to this voltage perturbation is a current. The signal in (voltage) and signal out (current) are related by the impedance. This impedance has both a resistive and capacitive contribution. This response can be modeled as the response of an equivalent electric circuit constructed of different inductors, capacitors, resistors, and diffusive components in different connectivity and of different value. These electric

PAGE 104

91 components represent true physical-chemical processes, such as charge-transfer resistance and double-layer capacitance. By modulating the frequency of the voltage perturbation, processes on different timescales can be probed. For instance, a typical experiment would investigate frequencies from the 100s of kHz to 1s of mHz. The geometries available from the template-synthesis of nanoscale electrodes studied by EIS should help to separate the collective current response into the contributions of faradaic and non-faradaic processes. Here a complex system requires a complex method of investigation. Another topic where focused resources can be productive is expanding upon the preliminary studies outlined in Chapter 5. The results presented in that chapter lay a framework for a polyvalent-ion intercalation project. Since there are orders of magnitude difference in the number of researchers investigating these principles compared to Li-ion intercalation, it is a field ripe for progress. This could result in a new class of less expensive batteries. This new class of batteries could become a compromise of cost and performance, a niche of systems more powerful than those based on Ni (NiMH or Ni-Cd), while less expense than the Li-ion. On a broader view, the field of Li-ion batteries works to constantly improve on the interplay of size, weight and power. Improvements are sectioned into materials and design. Research in the field of materials has been a focus of engineers since the Li-ion battery introduction in the early 1990s. After fifteen years of research, the standard commercial battery still uses the same (or very similar) components. If there is to be another great advance in the field, it will come in the design of the battery. These materials have sufficient intrinsic properties to function as Li-ion battery components.

PAGE 105

92 The key to providing more power using a smaller, lighter device is to become more elegant in its design. The issues with the current materials are their expense and toxicity. Both of these issues are direct functions of quantity. Apparently, the consumer is perfectly content with the hidden cost of this type of battery. This cost is hidden, because it is integrated into the total price of a laptop or the contract from a cellular phone service provider. An explicit price such as in a replacement battery, it warrants a raise of the eyebrow (about $50 for a cellular phone and $200 for a high-performance laptop). Why are these batteries so expensive? A main contributor to this is cost of the raw materials. Conservation comes from better design. What is meant by better design? The use of smaller, better contacted particles. Another fundamental tool in elegant design is a polymer electrolyte. Such an electrolyte allows the engineers more design flexibility. This should reduce the capital cost of these batteries. An electrochemically deposited polymeric electrolyte that has room-temperature Li-ion conductivity to rival that of the liquid electrolyte would be a major advance. Without a doubt there is a pot of gold at the end of this rainbow. A battery with a polymeric electrolyte is termed a solid-state battery. It is a difficult proposition though as the deposition process for each component of a solid-state battery has to be successively compatible. The components can be constructed of nanoparticles or nano-machined from deposited macrostructures by laser or plasma ablation. Indeed, the ultimate goal of this document is to inspire the next generation of graduate students to take hold of the methods and findings presented here and to direct them as they see fit.

PAGE 106

LIST OF REFERENCES (1) Nagaura, T. Tozawa, K. Progress in Batteries and Solar Cells 1990, 9, 209-217. (2) Tarascon, J. M. Armand, M. Nature 2001, 414, 359-367. (3) Tullo, A. H. Chemical & Engineering News 2002, 80, 25-26. (4) Moshtev, R. Johnson, B. Journal of Power Sources 2000, 91, 86-91. (5) Van Schalkwijk, W. A. Scrosati, B. In Eds.; Kluwer Academic / Plenum Publishers: New York, 2002, (6) Martin, C. R.; Menon, V. P. Parthasarathy, R. V. Polymer Preprints (American Chemical Society, Division of Polymer Chemistry) 1994, 35, 229-30. (7) Jirage, K. B.; Hulteen, J. C. Martin, C. R. Science 1997, 278, 655-658. (8) Brumlik, C. J. Martin, C. R. Journal of the American Chemical Society 1991, 113, 3174-5. (9) Wirtz, M. Martin, C. R. Advanced Materials 2003, 15, 455-458. (10) Menon, V. P. Martin, C. R. Analytical Chemistry 1995, 67, 1920-8. (11) Miller, S. A.; Young, V. Y. Martin, C. R. Journal of the American Chemical Society 2001, 123, 12335-12342. (12) Che, G.; Lakshmi, B. B.; Fisher, E. R. Martin, C. R. Nature 1998, 393, 346-349. (13) Che, G.; Lakshmi, B. B.; Martin, C. R.; Fisher, E. R. Ruoff, R. S. Chemistry of Materials 1998, 10, 260-267. (14) Lakshmi, B. B.; Patrissi, C. J. Martin, C. R. Chemistry of Materials 1997, 9, 2544-2550. (15) Lakshmi, B. B.; Dorhout, P. K. Martin, C. R. Chemistry of Materials 1997, 9, 857-862. (16) Martin, C. R. Accounts of Chemical Research 1995, 28, 61-8. 93

PAGE 107

94 (17) Che, G.; Jirage, K. B.; Fisher, E. R. Martin, C. R. Journal of the Electrochemical Society 1997, 144, 4296-4302. (18) Kuwabata, S.; Idzu, T.; Martin, C. R. Yoneyama, H. Journal of the Electrochemical Society 1998, 145, 2707-2710. (19) Li, N.; Martin, C. R. Scrosati, B. Electrochemical and Solid-State Letters 2000, 3, 316-318. (20) Li, N. Martin, C. R. Journal of the Electrochemical Society 2001, 148, A164-A170. (21) Nishizawa, M.; Mukai, K.; Kuwabata, S.; Martin, C. R. Yoneyama, H. Journal of the Electrochemical Society 1997, 144, 1923-1927. (22) Patrissi, C. J. Martin, C. R. Journal of the Electrochemical Society 1999, 146, 3176-3180. (23) Patrissi, C. J. Martin, C. R. Journal of the Electrochemical Society 2001, 148, A1247-A1253. (24) Sides, C. R.; Li, N.; Patrissi, C. J.; Scrosati, B. Martin, C. R. MRS Bulletin 2002, 27, 604-607. (25) Sides, C. R. Martin, C. R. Advanced Materials 2005, 17, 125-128. (26) Sides, C. R.; Croce, F.; Young, V.; Martin, C. R. Scrosati, B. Electrochemical and Solid-State Letters 2005, 8, A484-A487. (27) GE Osmotics, I., 2003, 1220457, Revision B. (28) Fleisher, R. L.; Price, P. B. Walker, R. M. Nuclear Tracks in Solids; University of California Press: Berkeley, 1975. (29) Apel, P. Radiation Measurements 2001, 34, 559-566. (30) Poretics, Product Guide, 1995. (31) Martin, C. R.; Nishizawa, M.; Jirage, K. Kang, M. Journal of Physical Chemistry B 2001, 105, 1925-1934. (32) Despic, A. Parkhutik, V. P. In Modern Aspects in Electrochemistry; Bockris, J. O., White, R. E. Conway, B. E., Eds.; Plenum Press: New York, 1989, 20, Chap. 6, 401-503.

PAGE 108

95 (33) West, W. C.; Myung, N. V.; Whitacre, J. F. Ratnakumar, B. V. Journal of Power Sources 2004, 126, 203-206. (34) Stein, A. Schroden, R. C. Current Opinion in Solid State and Materials Science 2001, 5, 553-564. (35) Turner, M. E.; Trentler, T. J. Colvin, V. L. Advanced Materials 2001, 13, 180-183. (36) Velev, O. D.; Tessier, P. M.; Lenhoff, A. M. Kaler, E. W. Nature 1999, 401, 548. (37) Jiang, P.; Bertone, J. F. Colvin, V. L. Science 2001, 291, 453-457. (38) Velev, O. D.; Lenhoff, A. M. Kaler, E. W. Science 2000, 287, 2240-2243. (39) Hulteen, J. C. Van Duyne, R. P. Journal of Vacuum Science and Technology A 1995, 13, 1553-1558. (40) Cepak, V. M.; Hulteen, J. C.; Che, G.; Jirage, K. B.; Lakshmi, B. B.; Fisher, E. R.; Martin, C. R. Yoneyama, H. Chemistry of Materials 1997, 9, 1065-1067. (41) Russell, A.; Repka, K.; Dibble, T.; Ghoroghchian, J.; Smith, J. J.; Fleischmann, M. Pons, S. Analytical Chemistry 1986, 58, 2961-4. (42) Karpinski, Z. J. Osteryoung, R. A. Journal of Electroanalytical Chemistry 1993, 349, 285-97. (43) Drew, S. M.; Wightman, R. M. Amatore, C. A. Journal of Electroanalytical Chemistry and Interfacial Electrochemistry 1991, 317, 117-24. (44) Hulteen, J. C.; Menon, V. P. Martin, C. R. Journal of the Chemical Society, Faraday Transactions 1996, 92, 4029-4032. (45) Miller, S. A. Martin, C. R. Journal of the American Chemical Society 2004, 126, 6226-6227. (46) Giddings, J. C. Unified Separation Science; John Wiley & Sons, Inc.: New York, 1991. (47) Johnson, B. A. White, R. E. Journal of Power Sources 1998, 70, 48-54. (48) Li, N.; Mitchell, D. T.; Lee, K.-P. Martin, C. R. Journal of the Electrochemical Society 2003, 150, A979-A984. (49) Lytle, J. C.; Yan, H.; Ergang, N. S.; Smyrl, W. H. Stein, A. Journal of Materials Chemistry 2004, 14, 1616-1622.

PAGE 109

96 (50) Fransson, L.; Eriksson, T.; Edstrom, K.; T., G. Thomas, J. G. Journal of Power Sources 2001, 101, 1-9. (51) Linden, D. Reddy, T. B., Eds., Handbook of Batteries, 3rd ed; McGraw-Hill, New York, NY, 2002. (52) Scrosati, B. Nature 1995, 373, 557-8. (53) Thomas, J. Nature Materials 2003, 2, 705-706. (54) Morcrette, M.; Rozier, P.; Dupont, L.; Mugnier, E.; Sannier, L.; Galy, J. Tarascon, J. M. Nature Materials 2003, 2, 755-761. (55) Hammami, A.; Raymond, N. Armand, M. Nature 2003, 424, 635-636. (56) Thackeray, M. Nature Materials 2002, 1, 81-82. (57) Polzot, P.; Laruelle, S.; Grugeon, S.; Dupont, L. Tarascon, J. M. Nature 2000, 407, 496-499. (58) Croce, F.; Appetecchi, G. B.; Persi, L. Scrosati, B. Nature 1998, 394, 456-458. (59) Plichta, E. J.; Hendrickson, M.; Thompson, R.; Au, G.; Behl, W. K.; Smart, M. C.; Ratnakumar, B. V. Surampudi, S. Journal of Power Sources 2001, 94, 160-162. (60) Nagasubramanian, G. Journal of Applied Electrochemistry 2001, 31, 99-104. (61) Salomon, M.; Lin, H. P.; Plichta, E. Hendrickson, M. In Advances in Lithium-Ion Batteries; Van Klinken, W. Scrosati, B., Eds.; Kluwer Academic/Plenum Publishers: New York, 2002, Ch 11. (62) Lin, H. P.; Chua, D.; Salomon, M.; Shiao, H. C.; Hendrickson, M.; Plichta, E. Slane, S. Electrochemical and Solid-State Letters 2001, 4, A71-A73. (63) Huang, C. K.; Sakamoto, J. S.; Wolfenstine, J. Surampudi, S. Journal of the Electrochemical Society 2000, 147, 2893-2896. (64) Ratnakumar, B. V.; Smart, M. C.; Kindler, A.; Frank, H.; Ewell, R. Surampudi, S. Journal of Power Sources 2003, 119-121, 906-910. (65) Aurbach, D.; Markovsky, B.; Rodkin, A.; Cojocaru, M.; Levi, E. Kim, H.-J. Electrochimica Acta 2002, 47, 1899 1911. (66) X-Ray Powder Data File, 1962.

PAGE 110

97 (67) Smart, M. C.; Ratnakumar, B. V.; Surampudi, S.; Wang, Y.; Zhang, X.; Greenbaum, S. G.; Hightower, A.; Ahn, C. C. Fultz, B. Journal of the Electrochemical Society 1999, 146, 3963-3969. (68) Delmas, C.; Cognac-Auradou, H.; Cocciantelli, J. M.; Menetrier, M. Doumerc, J. P. Solid State Ionics 1994, 69, 257-64. (69) Martin, C. R. Science 1994, 266, 1961-6. (70) Levi, M. D.; Levi, E. A. Aurbach, D. Journal of Electroanalytical Chemistry 1997, 421, 89-97. (71) Lu, Z.; Levi, M. D.; Salitra, G.; Gofer, Y.; Levi, E. Aurbach, D. Journal of Electroanalytical Chemistry 2000, 491, 211-221. (72) Wen, C. J.; Boukamp, B. A.; Huggins, R. A. Weppner, W. Journal of the Electrochemical Society 1979, 126, 2258-66. (73) Bard, A. J. Faulkner, L. R. In Electrochemical Methods: Fundamentals and Applications; Bard, A. J. Faulkner, L. R., Eds.; John Wiley & Sons: New York, 1980, (74) Lantelme, F.; Mantoux, A.; Groult, H. Lincot, D. Journal of the Electrochemical Society 2003, 150, A1202-A1208. (75) McGraw, J. M.; Bahn, C. S.; Parilla, P. A.; Perkins, J. D.; Readey, D. W. Ginley, D. S. Electrochimica Acta 1999, 45, 187-196. (76) Miyazaki, H.; Sakamura, H.; Kamei, M. Yasui, I. Solid State Ionics 1999, 122, 223-229. (77) Vivier, V.; Farcy, J. Pereira-Ramos, J.-P. Electrochimica Acta 1998, 44, 831-839. (78) Levi, M. D. Aurbach, D. Electrochimica Acta 1999, 45, 167-185. (79) Bard, A. J. Faulkner, L. R. In Electrochemical Methods: Fundamentals and Applications; Bard, A. J. Faulkner, L. R., Eds.; John Wiley & Sons: New York, 1980, (80) Deiss, E. Electrochimica Acta 2002, 47, 4027-4034. (81) Eftekhari, A. Electrochimica Acta 2005, 50, 2541-2543. (82) Padhi, A. K.; Nanjundaswamy, K. S.; Masquelier, C.; Okada, S. Goodenough, J. B. Journal of the Electrochemical Society 1997, 144, 1609-1613.

PAGE 111

98 (83) Ravet, N.; Chouinard, Y.; Magnan, J. F.; Besner, S.; Gauthier, M. Armand, M. Journal of Power Sources 2001, 97-98, 503-507. (84) Huang, H.; Yin, S.-C. Nazar, L. F. Electrochemical and Solid-State Letters 2001, 4, A170-A172. (85) Arnold, G.; Garche, J.; Hemmer, R.; Strobele, S.; Vogler, C. Wohlfahrt-Mehrens, M. Journal of Power Sources 2003, 119-121, 247-251. (86) Chung, S.-Y.; Bloking, J. T. Chiang, Y.-M. Nature Materials 2002, 1, 123-128. (87) Croce, F.; Epifanio, A. D.; Hassoun, J.; Deptula, A.; Olczac, T. Scrosati, B. Electrochemical and Solid-State Letters 2002, 5, A47-A50. (88) Doeff, M. M.; Hu, Y.; McLarnon, F. Kostecki, R. Electrochemical and Solid-State Letters 2003, 6, A207-A209. (89) Takahashi, M.; Tobishima, S.-I.; Takei, K. Sakuri, Y. Solid State Ionics 2002, 148, 283-289. (90) Aurbach, D.; Lu, Z.; Schechter, A.; Gofer, Y.; Gizbar, H.; Turgeman, R.; Cohen, Y.; Moshkovich, M. Levi, E. Nature (London) 2000, 407, 724-727. (91) Imamura, D.; Miyayama, M.; Hibino, M. Kudo, T. Journal of the Electrochemical Society 2003, 150, A753-A758. (92) Jiao, L.; Yuan, H.; Wang, Y.; Cao, J. Wang, Y. Electrochemistry Communications 2005, 7, 431-436. (93) Scherrer, P. Nachr. Ges. Wiss. Gottingen 1918, 96-100. (94) Biscoe, J. Warren, B. E. Journal of the American Ceramic Society 1938, 21, 287-93. (95) Van der Planken, J. Journal of Materials Science 1969, 4, 927-9.

PAGE 112

BIOGRAPHICAL SKETCH Charles Rob Sides is the second of the four children of Charles Ray and Martha Campbell Sides. He was born in North Carolina and spent the vast majority of his life just across the state line in Fort Mill, SC. He graduated salutatorian of the class of 1996 from Fort Mill High School. Go Yellow Jackets! Rob received an undergraduate degree in chemistry at Clemson University in 2000 and enrolled immediately in graduate school. Go Tigers! Upon graduation from University of Florida, he hopes to spend several years doing an honest days work making other people a lot of money. Go Gators! 99


Permanent Link: http://ufdc.ufl.edu/UFE0012360/00001

Material Information

Title: Nanoscale Energy Storage Electrodes by Template-Synthesis
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0012360:00001

Permanent Link: http://ufdc.ufl.edu/UFE0012360/00001

Material Information

Title: Nanoscale Energy Storage Electrodes by Template-Synthesis
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0012360:00001


This item has the following downloads:


Full Text












NANOSCALE ENERGY STORAGE ELECTRODES BY TEMPLATE-SYNTHESIS


By

CHARLES ROBERT SIDES













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


2005

































Copyright 2005

by

CHARLES ROBERT SIDES

































This document is dedicated to my beloved wife and family.















ACKNOWLEDGMENTS

I would like to use this platform to expression sincere appreciation to my research

advisor, Charles R. Martin, for his guidance through my graduate career. I believe that I

leave with us equally pleased with our accomplishments.

I would like to thank my parents, Charles R. Sides and Martha C. Sides; my sister

Anna Dobbins, her husband Clint, and their two new children Bella and Chase; my

younger brother Matthew and his wife Kelly; my youngest brother William, unmarried as

of yet; my grandparents, Evelyn and Raymond Sides and Evelyn Campbell; and the most

recent addition, my new wife, the former Alison S. Knefely. Each of these family

members has motivated and encouraged me during each turn in this winding road.

I had the distinct pleasure of working with Stephen E. Creager (undergraduate

research advisor), Bruno Scrosati, and Fausto Croce (hosts during collaboration visit to

Rome and Chieti, Italy). I enjoyed both the professional and personal interactions. I also

thank the following members of the Martin group who advised me on experimental

design, presentation style, and my golf swing: Dr. Naichao Li, Dr. Lane Baker, Dr. Punit

Kohli, and Dr. C. Chad Harrell.

There are a number of other individuals that I want to acknowledge for helping me

to enjoy my nine years in college Wist, Riles, Grant, Wil, Baker, Buck, Elizabeth,

Scott, Heather, Bruce and Butler.
















TABLE OF CONTENTS

Page

A C K N O W L E D G M E N T S .................................................................... ......... .............. iv

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

LIST OF FIGURES ......... ......................... ...... ........ ............ ix

ABSTRACT .............. ..................... .......... .............. xii

CHAPTER

1 INTRODUCTION AND TEMPLATE-SYNTHESIS..................................................1

In tro d u ctio n .................................................................................. 1
B background on Li-ion B atteries......................................................... ............... 3
N anom materials by Tem plate-Synthesis ........................................ ...... ............... 5
T e m p late s ................... ...................6.............................
T rack-etch M em branes ............................................................... .....................6
Anodized alumina ....................... .... .. ...... ........... 8....8
C olloidal-cry stals ................................................................. 9
M material D position Strategies ........................................ ........................ 10
Sol-gel ................................ ........... .................. ......... 10
Other deposition strategies of template-synthesis......................................12
Impact of Template-Synthesis to Nanoelectrochemsitry................ ..................13
Fundamental Electrochemical Investigations.....................................................13
Gold nanoelectrode ensem ble ............................. ..... ....... ............... 13
Carbon nanotube m em brane..................................... ......... ............... 14
Li-ion B attery N anoelectrodes ........................................ ........................ 14
Polym eric track-etch tem plates ....................................... ............... 14
N anosphere-tem plated structures ............................................................... 17
Li-ion Battery Electrode Fabrication Strategies ............................... ................18
C om m ercial Process M ethod ...................................... ........................... ........ 18
T em plate-Synthesis M ethod ......................................................................... ... 18
A n Electrochem ical Prim er .................................................................... ............... 19
Comparing Electrochemical Methods .....................................................19
Comparison of Diffusion Regimes.................................... ............... 21
Electrochemical Methods for Li-ion Battery Electrodes................................22
Chapter Sum m ary ............................................ .. .. .... ........ ......... 24



v









2 NANOSTRUCTURED ELECTRODES AND THE LOW-TEMPERATURE
PERFORMANCE OF LI-ION BATTERIES ........ .. ............ .....................25

Introduction ............. .. ....... ......... ........................................................ 25
E lectrode Synthesis............ .............................................................. ........ .......... 27
Scanning Electron Microscopy ........... .................................... 30
Electrode Electrochemical Investigations............ ................................31
Constant Current Discharge Experim ents ................................. ............... 31
Comparison of Rate Capabilities............................ ............... 34
E valuation of Sc ............................ ........ .. ... ..... .... ...........38
Effect of Cycle Life ............................................. .... .... ... ........ .... 38
Effect of Electronic Conductivity............................................... .................. 40
C chapter Sum m ary ............................................ .. .. .... ........ .... .. ... 40

3 EVALUATION OF SOLID-STATE DIFFUSION COEFFICIENT OF LI-IONS
AT LOW TEM PERATURE......................................................... .............. 42

In tro d u ctio n .......................................................................................4 2
E lectrode Synthesis............ .............................................................. ........ .......... 43
E lectrolyte Synthesis ........................ .... .......................... .. ...... .. ................44
M easurem ent of D iffusion Distance...................................................... ............... 45
E lectrochem ical Investigations................................. ............ .. .........................47
Potentiostatic Intermittent Titration Technique (PITT) ....................................47
D eterm ining DL,+ (D ata analysis)............................................... .................. 51
A ctivation Energy of D iffusion Process ........................................ .................53
C yclic V oltam m etry ....................... .. ....................... ...... .. .... ...........55
Chapter Sum m ary ............................................ .. .. .... ........ ......... 57

4 A HIGH-RATE, NANOCOMPOSITE LIFEPO4/CARBON CATHODE................. 59

In tro d u ctio n ........................................................................................................... 5 9
E lectrode Synthesis........... .............................................................. .......... ...... 60
Structural Investigations .................................... ............... .............................. 62
Scanning Electron M icroscopy..................................................................... 62
X -ray D iffraction Studies (XRD ) ............................................. ............... 65
X-ray Photoelectron Spectroscopy .............................................. ..............65
A analysis of Carbon C content ........................................ .......................... 67
Electrochem ical Investigations.............................................................. ............... 68
C yclic V oltam m etry ....................... .. ....................... ...... .. .... ...........69
R ate C apabilities........................................................................ ............... 69
Chapter Sum m ary ............................................ .. .. .... ........ ......... 72

5 MAGNESIUM-ION INTERCALATION INTO TEMPLATE-SYNTHESIZED
NANOSCALE ELECTRODES IN THE ABSENCE OF CARBON.........................73

Introduction ..................................................... ................. ..... ...... 73
E le ctro d e S y n th e sis............................................................................................... 7 4









X R D S tu d ie s ............................................................................................................... 7 5
Electrochemical Investigations........................................................ ..........77
Ferrocene P seduo-R reference .................................... .... .....................................78
Cyclic Voltammetry .............................. .. .......... ................. 79
ITO-glass current collector........................... .................................... 79
Pt foil current collector........................................... .......................... 81
R ate C ap ab ility .............................................................................................. 8 2
Investigations of a Possible Mg-Sn Alloy ...................................... ............... 84
C h ap ter S u m m ary ............................................................................ .................... 8 6

6 CONCLUSIONS AND FUTURE DIRECTIONS ..................................................87

C o n c lu sio n s ........................................................................................................... 8 7
F utu re D direction s ................................................................90

L IST O F R E F E R E N C E S ....................................................................... ... ................... 93

B IO G R A PH IC A L SK E TCH ..................................................................... ..................99
















LIST OF TABLES


Table page

1-1 Comparison of commercial rechargeable batteries. ..............................................2

2-1 The effect of discharge rate and temperature on the specific capacity of the 70-nm
electrode. ............................................................................33

2-2 Param eters needed to calculate S ........................................ ........................ 38















LIST OF FIGURES


Figure page

1-1 Schematic of discharge process of Li-ion battery. Charge moves from anode to
cathode during discharge. Adapted from van Schalkwijk. .................................3

1-2 Scanning electron micrograph of the surface of a commercial nanoscale
polycarbonate track-etch template membrane ........................................................6

1-3 Fundamentals of electrochemical reactions comparing oxidation and reduction
reactions. EF is the Fermi level, the average energy of the electrons of the
cu rrent collector....................................................................................... ...... 2 1

1-4 Concentration polarization of intercalating Li-ions. Ions are non-uniformly
distributed and are concentrated at the electrode surface. Gray represents
LiV2O5; whereas, yellow represents V205. ................................... ............... 24

2-1 Discharge proprieties of a commercial Li-ion battery (AT&T Cylindrical 17500)
at low temperature. This cell has a rated capacity of 800 mAh. ..........................26

2-2 Schematic of the template-synthesis of a V205 electrode. .......................................30

2-3 Scanning electron micrographs. A) 70-nm electrode. B) 0.45-rim electrode. C)
0.8-tm electrode. D) Polycarbonate template used to prepare the 70-nm
electrode. ........................................................................... 3 1

2-4 Electrochemical cell used for the characterization of V205 electrodes .................32

2-5 Slow-rate (C/3) constant current discharge of 70-nm V205 electrode using the cell
pictured ab ov e. ..................................................... ................. 33

2-6 Capacity ratio (see text) versus discharge rate for experiments conducted at 25 C
(Blue), 0 C (Orange), and -20 C (Green). A) R70/o0. B) R70/0.45. .......................35

2-7 Cycle life of 70-nm V205 electrode charged and discharged at 10 C for 400
cycles. Columbic efficiency (Qout / Qin) over this range is greatly than 95%. ........39

3-1 Scanning electron micrograph. A) This low magnification image shows the
uniformity over the large number of wires that constitute the electrode. B) This
high magnification image is used to identify the solid-state diffusion path-length
(radius of a nanowire) to be 50 nm .............................................. ............... 46









3-2 Electrochemical response of nanostructured V205 electrode to a AE = -15 mV
potential step. A) Current-time transient recorded for 45 minutes as the current
decayed to a steady-state value. B) Current-time transient shown in Cottrell-
plot form. The plateau represents the Cottrellian value It. ................................48

3-3 Cottrell-plots as a function of potential and thus level of intercalation. At the
potentials corresponding to the voltammetric minima, the Cottrellian value It'2
is shifted to more negative values and occurs at longer times. .............................50

3-4 Charge intercalated per potential step. The majority of charge is intercalated at
the potentials corresponding to voltammetric minima and galvanostatic plateaus..51

3-5 Solid-state diffusion coefficient as a function of potential. Y1-axis is the solid-
state diffusion coefficient and the Y2-axis corresponds to characteristic diffusion
time constant (c) at room temperature. Note that both y-axes are plotted on a
logarithm ic scale. ................................................... ................. 52

3-6 Dependence of DL,+ conducted at 25 C (Blue squares), 0 C (Orange circles), and
-20 C (G reen triangles). .............................................. ................................ 53

3-7 Arrhenius plot to solve for activation energy of diffusive process. The slope of
this line represents -EA/R. All values of D were taken at E = 3.24 V. At this
potential the intercalation is between phases of intercalation (see text). .................54

3-8 Cyclic voltammetric responses. A) Typical cyclic voltammetric response of 0.8-
|tm electrode at room temperature. Scan rate = 0.1 mV s 1. B) Comparison of
AEpkl from cyclic voltammograms of nano- and micro-structured electrodes at
various tem peratures. ......................... ...... ............................ .. .... ..... .. 56

4-1 Schematic of template-synthesis of LiFePO4/carbon nanocomposite electrode. .......62

4-2 Scanning electron micrographs. A) Lower magnification image of the
nanocomposite LiFePO4/carbon electrode. B) Higher magnification image of
the nanocomposite LiFePO4/carbon electrode. Composite fiber diameter is 350
nm. C) Image of LiFePO4 electrode synthesized by template dissolution
method (absent of carbon). LiFePO4 fiber diameter is 170 nm. ........................... 64

4-3 XRD pattern from nanocomposite, microcomposite, compared to accepted
literature values for LiFePO4. The substrate is quartz. Acquisition time is 10 s. ..65

4-4 X-ray photoelectron spectroscopy data for C Is peak for the carbon in the
nanocomposite electrode. Relative peak areas show amphorous form is
dominant (-80%), but graphitic carbon is also present (-20%) ............................67









4-5 Cyclic voltammograms. A) Cyclic voltammogram for the nanocomposite
LiFePO4/carbon electrode prepared using a template with 50 nm-diameter pores.
Scan rate = 0.1 mV s-1. B) Comparison ofvoltammetric response to theoretical
response exhibiting diffusional tailing. The diffusional wave was generated
from the normalized X function.79 ....................................................................... 70

4-6 Constant current experiments. A) Constant-current (3 C) discharge of
LiFePO4/carbon nanocomposite electrode. B) Specific capacity versus C-rate
for the nanocomposite LiFePO4/carbon electrode prepared using a template with
50 nm -diam eter pores ........................................... .... ................ .. .. 71

5-1 X-ray diffraction experiment. The sample is nanostructured V205 on Pt.
Acquisition time =10 s. The discrete lines represent the accepted values for
orthorhombic phase V205 (JCPDS 41-1426). ................. ............................... 76

5-2 Cyclic voltammogram of ferrocence in Li-based and Mg-based systems (see text).
Scan rate = 50 m V s ....................... ...................... .. .... .... ............... 78

5-3 Cyclic voltammogram of ITO-glass/MgxV205 nanowire electrode. Scan rate =
0.1 mV s-1. X2-axis is calculated from data shown in Figure 5-1.........................80

5-4 Cyclic voltammogram of Pt/MgxV205 nanowire electrode. Scan rate = 0.1 mV s-
The potential reference is M g/M g2+......................... ............................ 81

5-5 Comparison of rate capabilities of Li xV205 and Mg xV20. The mass for these
calculations is the electroactive mass determined by integration of the respective
cyclic voltam m ogram ...................... .... .............. ................... .. ......83

5-6 Comparison of rate capabilities of nano- and micro-structured MgxV205. The
mass for these calculations is the mass determined by ICP-AES analysis of the
V-ion from post-electrochemical dissolution. ................. ............................... 83

5-7 Cyclic voltammogram for Sn-based electrode in Li-system prepared using a
template with 50 nm-diameter pores. Scan rate = 0.1 mV s1 ................................84

5-8 Cyclic voltammogram for Sn-based electrode in Mg-system prepared using a
template with 50 nm-diameter pores. Scan rate = 0.1 mV s. ................................85















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

NANOSCALE ENERGY STORAGE ELECTRODES BY TEMPLATE-SYNTHESIS

By

Charles Robert Sides

December 2005

Chair: Charles R. Martin
Major Department: Chemistry

Lithium-ion batteries powered the recent boom of personal electronic devices, such

as cell phones, laptops and digital cameras. This success spawned a global research

initiative to adapt this technology to more demanding applications, such as low-

temperature systems or those relying on pulse-power. The electrodes of these batteries

store charge by reversibly intercalating Li-ions. The facile insertion flux of Li-ions into

the electrode and sluggish solid-state diffusion from surface to non-surface intercalation

sites causes a polarization of charge. Therefore, under demanding conditions, the

electrode discharges without fully accessing all charge-storage sites.

The electrode used in these studies is created by template-\yiheiil\ Template-

synthesis is a general nanofabrication method capable of creating structures of known

geometry. Nanomaterial-based electrodes mitigate the rate-limiting effects of sluggish

electron-kinetics and mass-transport. The large surface-area of this design serves to

distribute the current density, improving electron-kinetics, while the small size ensures









that intercalation sites reside close to the surface, minimizing the distance Li-ions must

diffuse in the solid-state.

The intent of this dissertation is to highlight the success of nanomaterials in the

study and design of energy-storage systems. It begins with a discussion of the low-

temperature performance of Li-ion battery electrodes. Charge-storage characteristics and

reproducible electrode geometries identify the fundamental breakdown of Li-ion batteries

at low-temperature as the decrease in solid-state diffusion coefficient of the Li-ion (DL,+).

This study then quantifies the value of DL,+ as a function of both intercalation-level and

temperature. Next, it describes a variation of the previous template-synthesis method, in

which the polymeric template is pyrolyzed to create a LiFePO4/carbon composite

electrode. This composite improves upon the poor electron-conductivity of the otherwise

attractive LiFePO4 cathode. Finally, preliminary results from polyvalent-ion (Mg2+)

intercalation into the V205 electrode are presented. Polyvalent-ions allow for more

charge to be stored than is stoichiometrically possible by singly charged ions.

The advantage of incorporating nanomaterials into the design of energy-storage

devices is the recurring theme of this document. While the studies are approached from a

fundamental view, conclusions such as those reported here will undoubtedly have

profound commercial impact on the increasingly portable world.














CHAPTER 1
INTRODUCTION AND TEMPLATE-SYNTHESIS

Introduction

Batteries provide power to an incredible number of devices that we rely on daily

such as automotive, electronics, and hearing aids and pacemakers. In addition to being

portable, batteries operate more efficiently than solar power, cleaner than fossil fuels, and

safer than nuclear power. Andrew Volta is credited for the first battery device in 1799

and sixty years passed until Gaston Plante discovered the first practical rechargeable

battery (the lead-acid battery still used in cars today.) Now there are several classes of

rechargeable batteries, which include Plante's lead acid battery, nickel-based systems

(nickel-metal hydride and nickel cadmium) and the Li-ion battery.

The introduction of Li-ion batteries by T. Nagaura and K. Tozawa of SonyTec Inc.

in 1990 marked a major advance in battery technology.1 Li-ion batteries have generated

great interest as lightweight, portable, rechargeable power sources over the last decade.

Li-ion batteries are now the power source of choice for laptops, cell phones, digital

cameras, and camcorders. Table 1-1 compares several of the rechargeable batteries

currently on the market.2 Note that the Li-ion battery has unparalleled energy density,

but is comparatively expensive. Li-ion batteries are popular because of their high cell

potential, large cycle life, high energy (Wh L-1, Wh kg-1,) and power density (W L-1, W

kg-1.) For many applications these advantages outweigh the expense of these batteries.









Table 1-1: Comparison of commercial rechargeable batteries.


Energy Density

Nominal By Weight By Volume
Battery Type Voltage (V) (Wh kg-1) (Wh L1)

Pb-acid 2 25 80

Ni-Cd 1.2 50 150

Ni-MH 1.2 75 200

Li-ion 3.6 150 325


The public has quickly embraced this technology, which accounts for an

approximately 3 billion dollar annual market.3 Despite the commercial success of these

batteries, a global research initiative exists to improve the existing design. The goal of

this research is to apply this technology to more demanding applications, such as those

relying on pulse-power or improved low-temperature performance. A specific example is

for use as the electric component of hybrid vehicles, which is currently a NiMH based

system. However, the current Li-ion battery design cannot adequately satisfy the power

requirements of such systems due to the inability to deliver a sufficient quantity of charge

at high discharge currents4 combined with concerns of safety and capital expenses.

The emergence of the burgeoning field of nanotechology has already tremendously

impacted the field of scientific research. The fields of biotechnology and electronics

have revolutionized. This dissertation details my efforts to incorporate the field of

nanomaterials to improve the design of Li-ion batteries.









Background on Li-ion Batteries

The three primary components of any battery are the cathode, the electrolyte, and

the anode. Li-ion batteries operate by reversibly intercalating charge in each of the two

electrodes. Intercalation is the process by which a specific quantity of guest species (Li )

is able to reversibly enter/exit a host structure (e.g., V205, carbon), causing little or no

difference to the host. These electrodes are separated by an ion-conductive electrolyte.

Upon discharge, the Li-ions deintercalate from the low-potential electrode, migrate

through the electrolyte, and insert into the high-potential electrode. The ions then must

rely on solid-state diffusion to fill the non-surface intercalation sites. A corresponding

quantity of charge travels the circuit and provides power to the load. This process is

detailed in Figure 1-1.

e--


ANODE __/ CATHODE

Electro I yte/Separator
L



,- --Ch: ; +B
-.,' -- ..



Cu | +
Currart LC Li, MO2 Currna
Odlector Collector

Figure 1-1: Schematic of discharge process of Li-ion battery. Charge moves from anode
to cathode during discharge. Adapted from van Schalkwijk. 5

If current flow is reversed (from cathode to anode), Li-ions insert into the low-

potential electrode and the system is charged. The low-potential electrode is the anode









and the high-potential electrode is the cathode. This convention (adopted from the

discharge process) is obeyed regardless of the direction of current flow. As in a

traditional electrochemical process, the reaction is limited by either mass-transfer

(diffusion of ionic species) or electron-kinetics. It does differ though, because it is the

host, not the diffusing species, which oxides/reduces. General forms of the

charge/discharge reactions for a Li-ion battery cathode and anode are shown in Equation

1-1 and 1-2, respectively.

discharge
Cathode : MO2 + xLi+ + xe- -- LixMO2
charge 1-1



charge
Anode: 6 C + xLi+ + xe- -- LixC6
discharge 1-2

For the cathode material shown in Equation 1-1 M = transition metal or a mixture

of transition metals. The most popular choices are Co, Mn, Ni, and V. These are the

constituents of the complex that undergo the redox reaction.

The anode material is shown in Equation 1-2 as carbon. Development of the

carbon-based anode was an engineering improvement to mitigate the early safety issues

of lithium-metal anodes. Replacing the Li metal anode with the carbonaceous material

obviated the short-circuit issues caused by dendritic growth during the Li-deposition

process. Graphite has a layered structure and is a popular alternative to lithium because it

can store a large amount of charge (-370 mAh g-) and has a reduction potential just

positive of lithium metal (- 100 mV). The cell voltage delivered by a battery is the









difference in the operating potential of the two electrodes, so a significantly negative

anodic potential is necessary to maintain the large cell voltage.

Nanomaterials by Template-Synthesis

The Martin research group has pioneered the nanofabrication strategy of template

synthesis.6 This general method has been used to synthesize nanostructures of a variety

of materials such as gold,7-10 carbon, 1113semiconductors,14'15 polymers,6'16 and Li-ion

battery electrodes,12,14,17-26 our focus here. This method involves deposition of a material

precursor into a micro- or nanoporous template. This template can be a variety of porous

materials, such as commercially available track-etch polymer filters, anodized alumina,

or even colloidal crystals. Depending on both the pore-diameter and the specific

chemical interactions between the pore wall and the precursor, the resulting structures

may be tubes (hollow) or wires (solid). The template is functional by restricting particle-

growth. It may remain intact to impart directionality or increased mechanical strength

during the experiments.

For battery materials, the template is a commercially-available polycarbonate filter.

The pores of this filter are monodisperse, nanoscopic in diameter, nominally cylindrical

in shape, and traverse the entire length of the membrane. Electrode precursor is

deposited into the pores of the organic template. The template is then preferentially

etched by oxygen plasma, leaving structures of identical geometry as the pores. A

sintering process imparts crystallinity to the structures to ensure that the host-structure is

maintained during the intercalation/deintercalation process. These structures are referred

to as nanoo" if one or more of their dimensions are on the nanoscale (< 100 nm).

However, the aspect ratio (length / width) is often on the order of 10.









Templates

Track-etch Membranes

Micro- and nanoporous polymeric filtration membranes prepared via the "track-

etch" method are available from commercial sources (e.g., GE Osmonics) in a variety of

materials and pore geometries.27 Polycarbonate is perhaps the most common example of

a track-etch filter material. Other options of materials include polyester, Teflon, and

polyethersulfone. An electron micrograph of the surface of a polycarbonate template

membrane is shown in Figure 1-2.













500 nm



Figure 1-2: Scanning electron micrograph of the surface of a commercial nanoscale
polycarbonate track-etch template membrane.

The term track-etch refers to the pore-production process.28'29 Pores of the filters

are created by exposing the solid material film to nuclear fission fragments, which leave

randomly-dispersed damage-tracks in the film. The high energy (on the order of 2 GeV)

of the fragments ensures that the tracks span the entire length of the membrane (typically

from 6 to 10 irm). These reactive chains end at the damage-tracks in the polycarbonate

film and are then etched with a basic chemical solution, and they become pores. One ion









creates one track, which in turn becomes one pore. During production the pore density is

controlled by the duration of time that the polycarbonate film is exposed to the charged

particles. Typical track-etch membrane pore-densities are 104 to 108 pores cm-2.30

Varying parameters of this etching solution such as temperature, strength and

exposure time dictate the pore diameter. Commercial membranes are available with pore

diameters ranging from 10 nm to 20 im. The microporous material etches to form

uniform cylindrical pores, but as the pore diameter is reduced to the smaller nanoscopic

dimensions the shape of the pore becomes like a cigar, slightly tapered at the ends.

Microscopic investigations of template-synthesized nanostructures prepared within the

pores of such nanoporous membranes have shown that the diameter of the pore in the

center of the membrane is larger than the diameter at the membrane surface.22'31 This

pore geometry may arise from the fission fragment that creates the damage track and also

generates secondary electrons, which contribute to the damage along the track. The

number of secondary electrons generated at the faces of the membrane is less than in the

central region of the membrane. An alternate theory is that the surfactant protective layer

adsorbed to the surface of the membrane retards the local etching process.29 Either

suggested mechanism leads to "bottleneck" pores.

In the production of template-synthesized nanostructured battery materials,

polymeric (specifically polycarbonate) track-etch membranes are the current template of

choice. These templates may be easily removed in conditions that do not adversely affect

the nanostructures themselves. The wide variety of commercially available pore

diameters and densities can generate comparative structures of differing geometries that

are key tools for fundamental investigations. The disadvantage that is associated with









these types of membranes and their application to electrode materials is their low

porosity. These values are typically between 2 10%. This decreases the ratio of active

material to a given footprint area or volume region on the current collector surface. It is

important to remember, however, that these electrodes will serve as tools for fundamental

studies, not a viable approach for commercial synthesis.

Anodized alumina

Another template membrane used in our laboratory is anodized alumina.

Anodization of aluminum metal in an acidic environment causes the metal to etch in a

fashion that leaves a porous structure.32 These pores are extremely regular, having

monodisperse diameters and cylindrical shapes in a hexagonal array. Unlike the track-

etch process, this process is systematic and generates an isolated, non-connected pore

structure. The pore densities of these alumina filters can be on the order of 1011 pores

cm-2, which is about 1000 times the density available in the track-etch polycarbonate

membranes. The porosities can be as high as 50%. Also, alumina filters have much

greater mechanical stability and chemical resistivity than polycarbonate. However, there

is an extremely limited selection of commercially available pore sizes, and the smaller

pores are branched. These membranes can be very thick (10 -100 [tm). It is also notable

that this alumina structure is electronically insulating.

There are advantages of alumina templates, but, to date, the application to Li-ion

battery research has been limited. The high porosity (template pore volume becomes

material volume) would dramatically increase the volumetric capacity (mAh L1)

available for a template-synthesized electrode. However, the chemical resistivity of the

membrane dictates that more harsh dissolution methods be employed. There are few Li-









ion battery intercalation electrodes that survive such exposure. Recently though another

research group has shown reversible Li-ion intercalation into MnO2 wires

electrochemically deposited in alumina template33

Colloidal-crystals

In the previous examples of templates, the pore-structure is cylindrical in general

shape. An alternative to this configuration is a nanosphere template (see Stein's recent

review).34 In this method, spherical particles with diameters of nano- to microscale

dimensions (typically 100s of nm) are deposited in a close-packed array. This is

commonly accomplished by a solvent evaporation technique. If a solvent evaporates at a

slow, controlled rate, it imparts an order to the particles. These particles are typically

made of polymers (latex, polysterene) or silica. Since they are spherical in shape, void-

volume exists in the interstitial sites, even when close-packed. These interstitial sites

serve as the porous network and the nanosphere array act as the template. These close-

packed spheres have a theoretical packing efficiency (defined as [volume of space

occupied by the spheres/total volume]) depending on specific arrangements of between

68 74%. Therefore, the theoretical void volume is approximately 30% in a close-

packed array and increases to 48% in a monolayer of spheres.

Again by physically restricting particle growth (as opposed to a chemical

technique), the deposition strategy becomes general. The requirements of size, shape,

and uniformity are transferred from the material to the template. This is demonstrated

with examples from the literature of synthesis of a variety of materials such as metal

salts,34'35 metal particles,36 polymeric materials,37'38 and evaporated or electroplated metal

films.39 Though the spherical shape is the most common, it is possible that the sphere can

be distorted to other shapes, such as an ellipsoid or even a "doughnut"-shape.38









Material Deposition Strategies

There are a number of differing structures of templates used for template-synthesis.

There are just as many material-deposition strategies. It is the combination of templates

and deposition-methods that provide a multiplicative number of solutions to a given

challenge. In our laboratory alone we have shown examples of sol-gel deposition,

electrochemical deposition (metals and polymers), electroless deposition, and chemical

vapor deposition.

Sol-gel

Sol-gel chemistry has recently evolved as a powerful approach for preparing

inorganic materials such as glasses and ceramics. This method for the synthesis of

inorganic materials has a number of advantages over more conventional synthetic

procedures. For example, high-purity materials can be synthesized at a lower

temperature. In addition, homogeneous multi-component systems can be obtained by

mixing precursor solutions; this allows for easy chemical doping of the materials

prepared.40 Such a versatile deposition technique partners well with the template-

synthesis nanofabrication method.

The sol-gel process typically involves hydrolysis of a solution of the precursor

molecule to obtain first a suspension of colloidal particles (the sol) and then a gel

composed of aggregated sol particles. The amorphous gel may then be thermally treated

to yield a more crystalline product. We have recently conducted various sol-gel

syntheses within the pores of the alumina and polycarbonate membranes to create both

tubes and wires of a variety of inorganic oxide materials, including semiconductors14'15

and Li-ion battery intercalation materials.20'22'24'25 First, the template membrane is

immersed into a sol for a given period of time, and the sol deposits on the pore walls.









After hydrolysis, either a tube or wire of the gelled material is formed within the pores.

As with other template synthesis techniques, longer immersion times yield wires, while

brief immersion times produce tubes.

The formation of tubes after short immersion times indicates that the sol particles

adsorb to the template membrane's pore walls. It has also been found that the rate of

gelation is faster within the pore than in bulk solution. This is most likely due to the

enhancement in the local concentration of the sol particles owing to adsorption on the

pore walls. An electrostatic interaction may also pull charged sol particles to the walls of

the template membrane.14

Previously, our group has demonstrated the formation of TiO2 semiconductor

nanostructures via template-synthesis.15 The mechanism of formation of TiO2 from

acidified titanium alkoxide solutions is well documented. In the early stages of the

synthesis, sol particles are held together by a network of -Ti-O- bonds. These particles

ultimately coalesce to form a three-dimensional infinite network, the gel. The fact that

tubes are initially obtained when this process is done in the alumina membrane indicates

that the sol particles adsorb to the pore walls. It is well-known that at the acidic pH

values, the sol particles are weakly positively charged. Tubes are formed because these

positively charged particles interact with anionic sites on the alumina pore wall.14

Several other inorganic oxides, specifically MnO2, C0304, ZnO, WO3, and Si02, have

been synthesized in a similar fashion. Since transition-metal oxides are typical choices

for Li-ion battery cathodes, it can easily be seen how this method partners well with

research of this field.









Other deposition strategies of template-synthesis

Electrochemical deposition is a based on electroplating metal from a solution into

the pores of the template. The structures plate axially from a current collector. By

controlling the charge (current and time) passed, the length of the structure can be

controlled. This method typically produces wires, but tubes are possible when a

molecular anchor is employed. We have made electrochemical plated structures of gold,

silver, nickel, platinum and even polymers.

Conversely, the electroless plating method relies on the electrochemical series for

metals to replace ones more noble than themselves. A polymer (nonconductive) template

is "sensitized" by Sn-ions, which are replaced by Ag particles, which are replaced by Au

particles. These gold particles coalesce to form a continuous structure. This method

plates metals from the walls of the template (radially); therefore, they plate as tubes with

a controllable inner diameter down to when they become solid-wires. The plating occurs

indiscriminately onto all faces of the membrane exposed to the solution. The most

common example is the electroless deposition of gold into a track-etch polymeric

template.

Vapor deposition is useful because of the conformal coating that it can create by

depositing materials in the vapor phase. It is commonly applied for making carbon

nanotubes from ethylene gas; however, there are examples of conducting, semi-

conducting and insulating materials. The difficulty with this method is the selection of

suitable precursors. Anodized alumina templates are ideal selections to accompany this

method, because of the ruggedness of that template. We deposit carbon by this method at

greater than 600 C. Another advantage to this method is that tubes of thin wall









diameters are possible. The Martin laboratory has shown that vapor-deposited carbon

nanotubes are ideal tools for fundamental studies of electroosmotic flow investigations.11

Impact of Template-Synthesis to Nanoelectrochemsitry

Fundamental Electrochemical Investigations

Gold nanoelectrode ensemble

One of the first fields where these template-synthesized nanomaterials

demonstrated superior functional capabilities is as an array of gold nanoelectrodes

elements.10 Our group has created an array of gold nanoelectrodes by electroless

deposition of Au to form solid wires in a track-etch polycarbonate template.

Nanoelectrodes offer opportunities to perform electrochemical experiments to investigate

the kinetics of redox processes that are too fast to measure at conventional macroscopic

electrodes.41,42 Also, they have the ability to serve as useful electrodes even in highly

resistive media.43 Depending on the spacing (pore density) and timescale of the

experiment (voltammetric scan rate), we were able to demonstrate by "radial" and "total

overlap" diffusional and voltammetric responses. The radial case refers to when the

diffusion layer of each element is independent, whereas the total overlap refers to when

the layers merge to act as a single diffusion layer. Template-synthesized gold nanowires

operating in the "total overlap" regime are able to measure electrochemical response of

redox molecules at trace concentrations (less than 2 nM for TMAFc+).10,44 This is a result

of the faradaic response being a function of geometric gold surface area, and the

background double-layer charging current being a function of active Au area. Gold

nanoelectrodes were shown to provide advantageous electrochemical results, due to their

unique template-synthesized geometry.









Carbon nanotube membrane

Carbon nanotubes are of great interest in both fundamental and applied science.

Our group has created template-synthesized carbon nanotubes by chemical vapor

deposition of ethylene gas onto a home-grown alumina membrane. This carbon nanotube

membrane (CNM) is used for fundamental electrochemical investigations, specifically as

tools to monitor and control electroosmotic flow (EOF).11'45 EOF refers to the movement

of solution past a stationary surface due to an externally applied electric field and is a

consequence of the way ions are distributed near surfaces.46 The Martin laboratory has

shown the ability to measure EOF through carbon nanotube membranes. This work was

further developed into the ability to control the rate and direction of EOF inside a CNM

by depositing a thin-layer of PVFc. By applying a potential, the sign and magnitude of

surface-charge, and thus EOF, can be effectively modulated.

Li-ion Battery Nanoelectrodes

Polymeric track-etch templates

Nanomaterials are advantageous in regards to both ionic and electronic

conductivity. Decreasing particle-size decreases the solid-state diffusion distance for Li

and increases the specific surface area, thus distributing the current and decreasing the

effective current-density. This serves to increase the electron-transfer kinetics of the

system. Solid-state diffusion coefficients are dependent on material and state-of-charge,

but are extremely small, in the range of 1011 to 10-15 cm2 s-1. This is an intrinsic

property. Therefore, our strategy is to use nanoparticles and minimize the distance the

Li-ion must diffuse. This creates a situation in which the intercalation sites are as close

to the electrode surface as possible.









The Martin laboratory published its preliminary data of Li-ion batteries in 1997.12,14

This was not long after these batteries were introduced to the marketplace. Patrissi22 and

Li19 were the first of the Martin laboratory to demonstrate the structural and

electrochemical characterization of template-synthesized Li-ion batteries electrodes,

V205 cathodes and Sn-based anodes, respectively. The work detailed here builds upon

the preliminary findings of these scientists. They built the foundations for a number of

the experiments discussed below.

A tremendous number of laboratories across the world are involved in this research.

Several premier journals devote sections of every issue to featuring advances in energy

storage (specifically, Li-ion based technology). Current major goals in this research are

to 1) create high-rate pulse-power compatible structures, 2) identify alternative electrode

materials (specifically more energetic, less expensive, and less hazardous cathodes), 3)

fabricate solid-electrolytes with viable ionic conductivity at room temperature. Energy

storage and production are topics of conversation at every corer of the industrialized

world.

The focus of research in the field of Li-ions batteries is to create systems of the

highest energy and power densities, both on a gravimetric and volumetric basis.

Essentially the goal is to get more power from a smaller, lighter device. Template-

synthesis as applied to the field of Li-ion batteries creates ideal tools for the fundamental

studies of factors that limit both rate and power. Previously, we have focused on using

constant-current to discharge template-synthesized electrodes and comparing that to the

response of thin-film control electrodes. During the discharge process, Li-ions

deintercalate from the anode, migrate through a Li-ion conducting electrolyte, and then









intercalate into the cathode. When these Li-ions intercalate into an electrode, they are not

able to solid-state diffuse rapidly enough to compensate for the facile nature of the

insertion-flux (D 10-8 to 1011 cm2 s-1). This results in concentration-polarization of Li-

ions at the surface during intercalation; conversely, the polarization is in the core during

the deintercalation process.47

This is a practical limitation to the stoichiometric quantity of charge-storage (mAh

g-1) that is theoretically possible. The discharge-rate (current) determines this rate of

insertion-flux; therefore at the high currents of a demanding application, this

concentration-polarization problem is exacerbated. Since the diffusion coefficient is an

intrinsic property of the electrode material, we work with the strategy of decreasing the

size of the electrode-particles. This allows the intercalation sites to be closer to the

surface, shortening the distance over which the sluggish solid-state diffusion process

must propagate.20'22'24'25

This plan of using smaller particles has another attribute. The increased surface-

area per total-volume fraction serves to decrease the effective current density at any given

rate. This works to offset any sluggish electron-kinetics of the system. Since the

electrodes must exhibit both good electronic and ionic conductivity, a structure

comprised of electronically addressable small particles is extremely favorable. This has

been demonstrated by template-synthesis of both cathodic20,22,24,25 and anodic

materials.20'48 In every case, the nanostructured material was able to deliver higher

specific capacity (mAh g-l) at any given discharge rate (C, C = h-1) than the

microstructured control electrode.









For an electrode to function in a rechargeable system, it must be able to maintain its

ability to be charged and discharged for many cycles. This cycle-life parameter is

measured as the charge-discharge cycle is repeated at the same rate. Our nanostructured

electrodes has been cycled for greater than 1400 times at the high-rate of 58 C, without

diminishing the quantity of charge delivered per cycle.20

As demonstrated in the above discussion, electrodes comprised of small particles

are advantageous. Void-volume is used to keep these particles small. While this has no

effect on the gravimetric capacity (mAh g-1), it does on volumetric capacity (mAh L-1).

Volumetric restrictions of an application may be just as stringent as gravimetric

restrictions. Our lab has shown the ability to "refill" a portion of this void-volume with

more active material to increase the charge stored per liter.23 In that investigation, the

volumetric charge delivered by the template-synthesized nanostructured electrode at

high-rates of discharge actually surpassed that delivered by the thin-film control.

Nanosphere-templated structures

Laboratories throughout the world are employing template-synthesis methods.

Other groups (most notably from the University of Minnesota) have worked with a

template method that uses nanospheres to create similar short-diffusion-distance

materials.49 An array of monodisperse PMMA spheres of diameter of 300 nm serves as

the template. Approximately 10 layers of these spheres construct an array. The void-

volume of these close-packed spheres hosts a Sn-based precursor solution. After

processing the precursor into a state capable of reversible Li-ion intercalation, the

PMMA-template spheres were removed via calcination. The electrode structure shrinks

as a result of this heating. Average solid-state diffusion distance for Li-ions is about 170

nm. This electrode showed good order to an area of 100 trm2. The resulting electrode









was characterized electrochemically and was able to reversibly store and deliver charge

under conditions similar to those previously described.49

Li-ion Battery Electrode Fabrication Strategies

Commercial Process Method

This synthetic method can be contrasted to a general method for commercial Li-ion

battery systems.47 There are many types of commercial batteries, such as prismatic cells,

coin cells, and "jelly-roll" cylindrical cells. The assembly and application of each are

different; however, the synthesis and components are general. The commercial battery

cathode is created in a high-temperature procedure that gives particles of the active

transition-metal-oxide. This process generates particles that range in diameter from

approximately 2 to 20 microns. These discrete particles are then mixed into slurry with

polymeric binder (PVDf) and a conductive element (carbon) and pressed into a metal

mesh current collector. The conductive element is necessary to improve the electronic

conductivity of the system, which is very low in this system because the particles are only

point-connected. The polymer is necessary to simply ensure that the slurry physically

contacts the current collector. In our template-synthesis approach, these inactive (not

capable of Li-ion intercalation) components are not necessary and are excluded so as not

to decrease volumetric and gravimetric energy densities or complicate analysis.50 Notice

that competing technologies use the mass of only the active material when determining

charge-per-gram.

Template-Synthesis Method

In this embodiment of Li-ion electrodes a precursor-impregnated polycarbonate

template membrane is attached to a section of metal foil. The foil has dual-functionality

as it serves as a substrate during synthesis and as a common current collector during









electrochemical characterization. Template-synthesis methods yield a fiber-based

electrode that consists of structures that mirror the geometry (length, diameter, and

number density) of the pores of the template. Typical values are 100-nm in diameter, 6

microns in length, and 108 elements per cm2. These structures extend from the surface of

the current collector like the bristles of a brush. There are only two components of the

electrode active material and current collector. Each electrode is then

electrochemically characterized as a half-cell.

Because of the parallel electronic-conduction mechanism of our template-

synthesized nanostructrued electrodes, carbon is not required.25 No binder is necessary

because the structures are directly attached to the metal foil that serves as both a physical

substrate and a common current collector. The absence of non-active components in our

electrodes also assists in increasing volumetric energy densities.

An Electrochemical Primer

Comparing Electrochemical Methods

A standard electrochemical experiment could be envisioned as a planar metal

electrode (working electrode) and a facile redox couple such as ferrocene/ferrocenium.

The ferrocene would be dissolved in water, along with a supporting electrolytic salt. This

other salt is composed of nonelectroactive ions and serves to decrease the contribution to

migration to the mass transfer, as well as decreasing the cell resistance. A reference

electrode is in the system as well. The purpose of this electrode is to provide a

comparative value for the applied potential. If the ionic conductivity (S cm-) of the

electrolyte is comparatively low, a Luggin capillary may be employed to minimize the

effective distance between the working and reference electrodes. This decreases the cell

resistance, minimizes iR drop, and ensures that the potential measured is accurate. A









counter electrode is inert in the system conditions and serves to pass current through the

system. A potentiostat is connected to the electrodes. Envision a potentiostat as the

instrument that sets the working electrode to the appropriate potential compared to the

reference electrode and measures the flow of current between the working and counter

electrodes.

For a reduction-oxidation redoxx) reaction to occur, an electron and ion must unite.

Figure 1-3 illustrates this from thermodynamics. Electron-kinetics describes the ability

for electrons to be transferred at a certain rate constant. Ion transport (mass transfer) is

accomplished by one of three mechanisms: diffusion, migration, and convection.

Commonly systems eliminate convection by not stirring the system. The effect of

migration, movement of ions in an electric field, is minimized by the supporting

electrolyte. This leaves diffusion as the predominant mass-transfer mechanism.

Diffusion is a result of a concentration gradient. This concentration gradient is related to

flux of the species by a proportionality constant, its diffusion coefficient (D). This

parameter has units of area per time. A small molecule in water has D 1 x 105 cm2 s-1

While the diffusive movements of species are random in nature, the bulk movement

serves to smooth concentration gradients.

Electrochemical measurements on Li-ion batteries are similar, but not identical.

We test our electrodes in a half-cell setup, meaning that the cathode and anode are

individually characterized. Li-ions are solvated in a non-aqueous electrolyte. This setup

is inside of an argon glovebox, to protect the dry electrolyte and the lithium metal. The

diffusive movement is associated with the Li-ions, but Li-ions are stored and the host

material is subject to the redox reactions.











0 Electrode Solution


t


Species gains e-
Reduction


". Vacant MO

EF

4 Occupied MO


0 Species loses e-
AG E Oxidation
AG = nFE

Figure -3: Fundamentals of electrochemical reactions comparing oxidation and
reduction reactions. EF is the Fermi level, the average energy of the electrons
of the current collector.
The host material is electronically conductive and injects electrons by contact with

a metal current collector. This intermediary between the ions and the current collector

serves to slow electron transfer. In addition, though Li-ions are small, they must diffuse

through solid-material as opposed to only liquid in a traditional system. The diffusion

coefficient for this process is about 6 orders of magnitude lower than for a liquid-phase

system. Therefore, the rates of the experiments must be several orders of magnitude less

than those used to typical experiments. Again, the rate of the reaction is either limited by

mass-transfer or electron-transfer.

Comparison of Diffusion Regimes
This leads us to a discussion of two regimes of diffusion that are discussed in this

dissertation semi-infinite and finite. Semi-infinite diffusion refers to the situation when









movement of ions is unrestricted. Ions may freely diffuse without boundary constraints,

so during the timescale of the experiment effects of the cell walls are not felt on the

system. Therefore, at some value away from the electrode there is no concentration

gradient. The direct opposite is the case for finite diffusion when the movement of ions

is constrained. This is often the case in thin-layer experiments where propagation of the

diffusion layer is truncated at some distance, the thickness of the thin-layer.

A colleague once compared these diffusion regimes to me as an analogy of two

people each holding one end of a rope. The first causes the rope to oscillate by moving

his hand up and down. This movement represents the diffusion process. If the rope is

sufficiently long enough, the second person cannot feel the response (semi-infinite

diffusion). However, if the rope is short enough, the perturbation is felt by the second

person and it is stopped by his hand (finite diffusion). There is no infinite diffusion

regime, because the diffusing species is confined by the walls of the cell, and as t

approaches infinity it will eventually encounter this system constraint.

Electrochemical Methods for Li-ion Battery Electrodes

We will use several standard electrochemical methods to characterize the Li-ion

battery electrodes cyclic voltammetry, chronopotentiometry, and small potential-steps.

Cyclic voltammetry will be used to confirm the activity of the electrode, identify the

potential region of electroactivity, and comment on the electron-kinetics and diffusion

regimes of the process. This is accomplished by sweeping an applied potential, while

monitoring the current response. Chronopotentiometry uses constant applied currents to

charge and discharge the electrode, while monitoring the change in potential. By

applying a known current to a known electrode mass, the time is easily converted to

charge stored and thus quantity of Li Larger currents (faster rates) decrease the time









regime for the intercalation process. Therefore, the solid-state diffusion process can

access less of the intercalation sites at larger currents. Measurement of the rate-

capability is a useful way to compare electrodes. The final method uses small applied

potential steps and monitors the current-response over time. Data from these small

potential steps are interpreted via the Cottrell relationship to identify the solid-state

diffusion coefficient of the Li-ion and identify the activation energy of the diffusion

process via an Arrenhenius relationship.

Since the quantity of charge stored in a Li-ion battery electrode is dependent on the

mass of the electrode, the molecular weight of the host species, as well as the

stiochiometric theoretical maximum, a normalized parameter is necessary when

discussing charge-discharge experiments. This parameter is C-Rate, 1 C = 1 h-1,

described in Equation 1-3.


C -Rate = 1-3
C, m
Here, i represents the applied current, C, represents the theoretical specific capacity

(mAh g-1), and m represents the mass of the electrode. The rate of 1 C is the current

necessary to incorporate the theoretical maximum charge in 1 h. For reference, if a

laptop battery lasts for 2 h, it is discharged at an average rate of C/2 (though, the rate is

actually a little higher, because this calculation assumes theoretical charge-storage.)

Figure 1-4 is a schematic of the result of the concentration polarization effect. This

effect depends greatly on C-Rate. At small rates, there is enough time for Li-ions to

occupy/vacate nearly all of the intercalation sites, even at the core of the electrode

particle. At high rates, there is much less time for these Li-ions to diffuse in the solid-









state, creating this large number of inaccessible intercalation sites, resulting in poor

charge-storage characteristics.

Commercial particle Nanowire
Diffusion distance = 1 5 pm Diffusion distance = 35 nm







V20s





LiV205
Figure 1-4: Concentration polarization of intercalating Li-ions. Ions are non-uniformly
distributed and are concentrated at the electrode surface. Gray represents
LiV205; whereas, yellow represents V205.

Chapter Summary

The intent of this dissertation is to demonstrate how nanoscale Li-ion battery

design succeeds where conventional technology fails. It will begin with a discussion of

the improved low-temperature performance of these electrodes. This will then be

extended to quantify the solid-state diffusion coefficient of the Li-ion as a function of

intercalation-level and temperature. Next, it describes a variation of the general synthetic

method, in which the polymer template is pyrolzed to create a LiFePO4carbon composite

electrode. Finally, this is followed by a demonstration of polyvalent-ion (Mg2+)

intercalation into V20s.














CHAPTER 2
NANOSTRUCTURED ELECTRODES AND THE LOW-TEMPERATURE
PERFORMANCE OF LI-ION BATTERIES

Introduction

Li-ion batteries have become the power source of choice for consumer electronic

devices such as cell phones and laptop computers.2'51-58 This is because these batteries

have good rechargeability (1000+ cycles) and offer higher energy density (stored charge

per unit volume or mass of the battery) than competing battery technologies.2'51

However, it is well documented that Li-ion batteries show poor performance at low

temperature or under the extreme currents necessary for pulse-power.59-63 Specifically,

the amount of charge delivered from the battery at temperatures below 0 C is

substantially lower than the amount of charge delivered at room temperature.59'60'62 This

precludes the utilization of these batteries in a number of defense, space and even

terrestrial applications.64 Figure 2-1 is generated from data published by

Nagasubramainan60 on the effect of temperature on the performance of a commercial Li-

ion battery cell. Note that the C-Rates are very low; however, still there is a dramatic

decrease in the charge-storage ability of the cell as temperature is decreased.

We have been investigating the application of nanotechnology to Li-ion battery

electrode design.17,20-22,24 Previously, the Martin laboratory has shown that template-

synthesized Li-ion battery electrodes are capable of delivering more charge-per-gram at

the same normalized discharge-rate (C-Rate) than control film electrodes. Based on these

studies it seemed likely that Li-ion battery electrodes composed of nanoscopic particles









of the electrode material could mitigate this low-temperature performance problem. We

prove this case here by showing that nanofibers (diameter = 70 nm) of the electrode

material V205 deliver dramatically higher specific discharge capacities at low

temperatures than V205 fibers with micro-sized diameters.





--- 25 C
740
---- 10 "C
m------- -A- -10 oC

720 -

*
c 700-


b 680 O--

a A

660 -


640 A

0.0 0.2 0.4 0.6 0.8 1.0

C-Rate / C, h-'

Figure 2-1: Discharge proprieties of a commercial Li-ion battery (AT&T Cylindrical
17500) at low temperature. This cell has a rated capacity of 800 mAh.

Li-ion batteries store charge by intercalating Li+ from a contacting solution phase,

along with an equivalent number of electrons, into the battery's anode material.2 During

discharge the Li+ must diffuse out of the anode material, through the electrolyte, and

intercalate into the cathode material. This process can be seen in Figure 1-1.









For the cathode material V205, investigated here, the reversible charging and

discharging reactions can be written as Equation 2-1.

discharge
V205 + xLi+ + xe- LixV205
charge 2-1


Other laboratories have concluded that the cathode, anode, and electrolyte were the

source of the fundamental breakdown that occurs at low temperature. While there is

controversy in the scientific literature,59'60'6263 the most likely causes of the poor low-

temperature performance are either 1.) Diminution in the rates of these electrochemical

charge/discharge reactions at low temperature. 2.) Diminution in the rate at which Li

diffuses within the particles which constitute the electrode at low temperature. We

hypothesized that in either case, an electrode composed of nanoscopic particles (diameter

less than 100 nm) would provide better low temperature performance than the -10-lm -

sized particles65 used in commercial battery electrodes. This is because an electrode

composed of nanoscopic particles would, in general, have higher surface area than an

electrode composed of large particles, and this would mitigate the slow electrochemical

kinetics problem. Furthermore, the distance that Li must diffuse within the particle

would be decreased for nanoscopic particles, and this would mitigate the slow solid-state

diffusion problem.

Electrode Synthesis

To prove this point we have used the template-synthesis method6 to prepare

cathodes composed of monodisperse V205 nanofibers (diameter = 70 nm) that protrude

from a current collector surface like the bristles of a brush (Figure 2-2A). We compare

the low-temperature charge/discharge performance of these nanofiber cathodes with









cathodes composed of V205 fibers with diameters of 0.8 [im (Figure 2-2C), as well as

with cathodes composed of 0.45 pm-diameter fibers (Figure 2-2B). If our hypothesis is

correct, the low-temperature performance of the cathode composed of the 70 nm-

diameter nanofibers (designated the 70-nm electrode) should be dramatically better than

that of the cathode composed of the 0.8 .im-diameter fibers (0.8-rm electrode), and the

low-temperature performance of the cathode composed of the 0.45 pm-diameter fibers

(0.45-rm electrode) should be between these two extremes. This experimental design

exploits the power of the template-synthesis method for preparing monodisperse fibers

having any desired diameter,6 and builds on our prior work on the electrochemistry of

nanostructured Li-ion battery electrodes.17,20-22,24

The template-synthesis method entails using the pores in a synthetic membrane or

other solid to form nanofibers, nanowires or nanotubes of a desired material.6

Commercially-available (Poretics, Inc) polycarbonate filtration membranes (figure 2-2D)

were used as the templates for these studies. These filters have monodisperse cylindrical

pores that run through the complete thickness (6 10 im) of the membrane and are

available with a wide range of pore diameters. The pore diameter of the template

determines the diameter of the fibers synthesized within the pores.6

This method was described in detail previously.22 Briefly, a 1.0 cm x 0.7 cm piece

of the template membrane was placed on the surface of a Pt foil electrode, and 0.6 [iL of

the liquid V20s precursor triisopropoxyvanadium (TIVO) was applied to the membrane

surface. TIVO fills the pores in the template, and hydrolysis in air converts this material

to V205. This chemistry is straight-forward and it shown in Equation 2-2.









water vapor


V- O H20 V20 + HO )


Vanadium (V) triisopropoxide Vanadium pentoxide isopropanol vapor 2-2


In fact, this reaction is so facile that the deposition must occur inside an argon-

filled glovebox, so that the precursor may impregnate the pores in this low-viscosity

state. It is then transferred to a low 02 environment (antechamber of the glovebox), so

that they hydrolysis can proceed slowly enough as to form continuous structures inside

the pores. This was allowed to proceed at room temperature for 12 hours, followed by

hydrolysis for 2 hours at 80 C in air. The film of V205 that covered the upper face of the

membrane was then removed by wiping with a damp cotton swab, and this procedure was

repeated to ensure that the pores were completely filled with V205. Oxygen plasma (25

W, 15 Pa, 2 hours) was then used to remove the polycarbonate template. To ensure

complete conversion to crystalline V205, the fibers were than heated at 400 C for 10

hours in flowing 02 gas. X-ray diffraction studies confirmed that the material obtained is

orthorhombic V205 (experimental details discussed later).66 A schematic of this process

is shown in Figure 2-2.









Polycarbonate Template Membrane
50-nm nominal diameter pores
6 |tm in length
6 x 108 pores cm-2

I
Template with Sol-Gel Precursor
Gel TIVO precursor
Air; Heat 80 oC; 2 h
Etch template
Oxygen plasma; 20 W, 2 h, 15 Pa
Heat 400 C in 02; 10h



V205 Nanostructures on
Pt Current Collector

100% active


Figure 2-2: Schematic of the template-synthesis of a V205 electrode.

Scanning Electron Microscopy

Scanning electron micrographs of electrodes of created in various templates are

shown Figure 2-3. These samples are attached to stubs by conductive copper tape prior

to imaging. The electrodes were not sputtered with any conductive material. This

demonstrates the sufficient electron conductivity of these samples. The polycarbonate

template was sputtered with a thin layer of Au/Pd to prevent excessive charging during

image acquisition.

These template-synthesized V205 nano and microfibers electrodes serve as useful

tools in fundamental investigations, though the void-volume, that is responsible for the

short diffusion distances, sacrifices volumetric capacity (mAh L-1). Previously, we

demonstrated a method that creates a compromise of these two interrelated parameters.23


Sl~~-li




































Figure 2-3: Scanning electron micrographs. A) 70-nm electrode. B) 0.45-rm electrode.
C) 0.8-rm electrode. D) Polycarbonate template used to prepare the 70-nm
electrode.

Electrode Electrochemical Investigations

Constant Current Discharge Experiments

The charge/discharge reactions for these nano and microfiber electrodes were

investigated in an electrolyte solution that was 1 M in LiC104 dissolved in a 1:1:1 (by

volume) mixture of ethylene carbonate, diethyl carbonate, and dimethyl carbonate.67 Li

metal ribbons served as the counter and reference electrodes. A schematic of this

electrochemical cell can be seen in Figure 2-4.










*" Argon Atmosphere


Electrolyte Solution

S1 M LiCIO4 in

I Ethylene Carbonate (EC)
Diethyl Carbonate (DEC)
Dimethyl Carbonate (DMC)

S;1 vtvlv Ratio

Counter Blectmrode
Ci ~nbir Betrode :Working Electrode
Lithium Ribbon
Lithium Ribbon rence Electrode V2,O on Pt
Lithium Ribbon
Figure 2-4: Electrochemical cell used for the characterization of V205 electrodes.

Temperature was set and maintained via a circulation bath of 80:20 ethylene

glycol:water. All potentials are quoted here vs. the Li/Li reference. A galvanostat was

used to apply a constant current to sequentially discharge and charge the V205 (Equation

2-1), and monitor the change in potential of the V205 electrode as a function of time. The

discharge reaction was continued until a potential of 2.8 V was achieved, and the charge

reaction was terminated when the potential reached 3.8 V. Over this potential window 1

mol of V205 is known to reversibly intercalate 1 mol of Li+, (i.e., x = 1 in Equation 2-1).68

This corresponds to a maximum specific (per g) charge-storage capacity of 148 mAh g1.

The mass of V205 present in all of our electrodes was determined gravimetrically so that

the specific capacities could be determined from the potential-vs.-discharge-time data.

A typical potential-vs.-time curve associated with constant-current discharge of a

70-nm electrode at 25 C is shown in Figure 2-5. A specific capacity of 145 mAh g-1

(equivalent to x = 0.98) was obtained from these data.











3.8-


3.6
i|
> 3.4


(r 3.2

cu
S3.0-


2.8

0 1000 2000 3000
Time s

Figure 2-5: Slow-rate (C/3) constant current discharge of 70-nm V205 electrode using
the cell pictured above.

This shows that at this temperature and current density, nearly 100% of the

maximum specific capacity of the cathode is being delivered. The effect of discharge

current or C-Rate and temperature on the specific capacity of the 70-nm electrode is

shown in Table 2-1.

Table 2-1: The effect of discharge rate and temperature on the specific capacity of the
70-nm electrode.


C-Rate / h- Specific Capacity / mAh g1
25 C 0 C -20 C
0.2 77.5
1 144.8 104.9 63.8
5 123.0 74.4 40.7
10 107.5 60.2 33.5
20 90.4 48.7 27.3
25 25.5
30 43.1 24.8
35 23.0
40 72.9 39.6 22.7









Comparison of Rate Capabilities

At all temperatures, we see the decrease in capacity with discharge rate, typical of

Li-ion battery electrodes.20'21'51'57 However, as expected,60-62 this decrease in capacity

becomes much more pronounced at low temperatures. Data of this type were obtained

for all three of the V205 fiber electrodes. Again, our objective is to demonstrate that the

70-nm electrode is able to deliver greater specific capacity at low temperatures than the

electrodes composed of the larger-diameter fibers. We define the specific capacity of the

70-nm electrode at a given discharge current, I, and temperature, T (Table 2-1), as Cs-

70,I,T. We define the identical parameter (i.e., same current and temperature) for the 0.8-

lm electrode as Cs-0.8,I,T. Finally, we define a capacity ratio R70/o.8 = Cs-70,I,T / Cs-0.8,I,T. If

R70/o.8 is greater than unity, then at these particular values of and T, the 70-nm electrode

shows a capacity advantage relative to the 0.8-m electrode.

Figure 2-6A shows plots of R70/0.8 vs. the discharge rate for experiments done at

temperatures of 25 C, 0 C and -20 C. Looking at the 25 C curve first, we see that at

low discharge rates, R70/0.8 is nearly unity, which means that at room temperature and low

discharge rate, both electrodes are delivering nearly 100% of their maximum specific

capacities; hence, at room temperature and low discharge rates, there is no capacity

advantage for the 70-nm electrode relative to the0.8-gm electrode. At higher discharge

rates R70/o.8 increases above unity, indicating that there is now a capacity advantage for

the nanofiber electrode; however, at room temperature this advantage is modest.













80-



60-
0
O
0.
t 40-


o
err
rr0
0


A -A- -20 C
-o- 0 C
-o- + 25 C









AA

Asi

- ______ 9~-


C-Rate / C, h"'


o



5-



0
0-


--20 C
-- 0 C
-- 25 C


C-Rate / C, h-1


Figure 2-6: Capacity ratio (see text) versus discharge rate for experiments conducted at
25 C (Blue), 0 C (Orange), and -20 oC (Green). A) R70/o.8 B) R70/0.45-


I









Turning now to the -20 C data, we find that at the lowest discharge rate the 70-nm

electrode delivers twice the specific capacity of the 0.8-pm electrode. Furthermore,

R70/o.8 increases dramatically with discharge rate; indeed, at the highest rate studied, the

specific capacity delivered by the 70-nm electrode is almost two-orders of magnitude

greater than the specific capacity delivered by the 0.8-m electrode. These dataprove

our 1y)pithei% -that an electrode composed of nanoscopic particles shows better

(indeed, dramatically better) low-temperature performance than an electrode composed

of micron-sized particles. The 70-nm electrode also shows a capacity advantage at the

intermediate temperature of 0 C, but as would be expected, the advantage is not nearly

as dramatic as at -20 C. These data confirm the results obtained with commercial Li-ion

batteries that performance at room temperature and 0 C are not so different but that

performance becomes markedly worse at -20 OC.62,67

We obtained similar data for the 0.45-rm electrode and calculated the analogous

specific capacity ratio R70/0.45 = Cs-70,I,T / Cs-0.45,I,T. If our hypothesis is correct, the 70-nm

electrode should show better low-temperature performance than the 0.45-rm electrode

(R70/0.45 > 1), but at any value of discharge current, R70/0.45 should be less than R70/o.8. A

comparison of the curves in Figure 2-6A and B shows that this is indeed the case.

We have proven that a nanostructured Li-ion battery electrode shows better low-

temperature performance than analogous electrodes composed of micron-sized particles.

Two questions remain. 1.) How do these studies relate to practical battery-electrode

design? 2.) From a fundamental viewpoint, what is the genesis of this improved low-T

performance? With regard to question 1, our results suggest that to improve low-T

performance, practical Li-ion battery electrodes should be prepared using smaller









particles of the active material than the -10 tm-sized particles currently used. However,

a practical Li-ion battery electrode contains other components in addition to the active

material; these include a polymeric binder and graphitic particles to improve electronic

conduction through the electrode.50 The effect of amount and particle size of these

ancillary components would also have to be evaluated. In this regard, our experimental

design based on template-synthesized micro and nano particles has tremendous advantage

for the fundamental studies performed here because these ancillary components are not

needed and therefore do not complicate the analysis.

Our data also provide an answer to question 2. We suggested above that the poor

low-temperature performance of Li-ion batteries is due to either a decrease in the

diffusion coefficient within the electrode particles, or a decrease in the rate of the

electrode reactions (Equation 2-1), with decreasing temperature. If the solid-state

diffusion coefficient is the culprit then low-temperature performance should improve

with decreasing particle diameter, and this is exactly what is observed here.

If electrode reaction rate is the culprit, then low-temperature performance should

improve with a parameter we designate So, the electrode-particle surface area per cm2 of

current collector area. The beauty of the template method is that So can be easily

calculated from the fiber diameter, the pore density of the template membrane, and the

fiber length. Table 2-2 shows these parameters. These calculations show that because of

the interplay between pore diameter and density in the template membranes, So is higher

for the 0.45-rim electrode (Sc = 14.1) than for the 70-nm (Sc = 7.9) or the 0.8-pm (Sc =

6.8) electrodes. This indicates that if electrode reaction rate is the culprit, then the 0.45-

Lm electrode would show the best low-temperature performance, and this is not observed









experimentally. Hence, our data show that the temperature dependence of the solid-state

diffusion coefficient determines the low-temperature performance of the electrodes

studied here.

Evaluation of Sc

As discussed in the text of the paper, So is the electrode-particle surface area per

cm2 of current collector area. The parameters needed to calculate So are presented in

Table 2-2.

Table 2-2: Parameters needed to calculate Sc.


Pore Diameter Pore Density Membrane thickness
(nm) (cm-2) (Gm)
70 6x108 6
450 Ix108 10
800 3x107 9


To calculate So for the template-synthesized nano or microfiber electrode, one first

calculates the circumference of an individual fiber. The surface area of the fiber is the

product of the circumference and the length of the fiber. The length is given by the

membrane thickness. The number of fibers per cm2 of current collector surface area is

equivalent to the pore density of the membrane. Therefore, the surface area per fiber

multiplied by the pore density is Sc.

Effect of Cycle Life

The order of the series of experiments dictates that each test delivers less charge

than the previous (e.g., low to high C-rate of discharge and high to low temperature.)

Therefore, it is helpful to examine any adverse effects on the cycle life on the system.

After the entire set of experiments was completed, if the system was returned to room

temperature (25 C), the electrodes maintained 98+% of the capacity shown in the









original series of tests at that temperature. This is an important control experiment to

help validate the data that are the basis for our conclusions.

Also, the cycle-life can easily be characterized by charging and discharging the

electrode with a constant current. Figure 2-7 shows the cycle life of a V205

nanostructured electrode that is charged and discharged at the constant rate of 10 C.

Since the same current is used to charge and discharge the electrode, the columbic

efficiency (charge out / charge in) reduces to a ratio of discharge time / charge time. The

columbic efficiency of this electrode was greater than 95%. The electrode stored and

delivered virtually the same charge over nearly 400 cycles at a high rate of discharge.



120


100


E 80
o Charge
o Discharge
o 60
CU




(D
o 40


CL 20
0)
0- 1 --I -I I

0 100 200 300 400

Cycle Number

Figure 2-7: Cycle life of 70-nm V205 electrode charged and discharged at 10 C for 400
cycles. Columbic efficiency (Qout / Qin) over this range is greatly than 95%.









Effect of Electronic Conductivity

In commercial Li-ion battery cathodes, it is necessary to add an electronically

conductive material, typically carbon particles, to improve electronic conductivity

through the electrode. We have shown that this is not necessary with template

synthesized electrodes.2022'24 This is because while the electronic resistance of a single

fiber (Rf) making up the template-synthesized electrode might be high, the electrode is a

parallel ensemble of such fibers. As such, the total resistance of the electrode (Rt) is

given by Rt = Rf/N, where N is the number of fibers in the electrode.

That electronic conductivity does not dominate the rate capabilities of the

electrodes studied here is easy to prove. Assuming the same length, it is easy to show

that the electronic resistance of a 70 nm diameter fiber is 33 times higher than that of a

0.8 rm diameter fiber composed of the same material. If the rate capability was limited

by electronic conductivity, the 70-nm electrode would show the worst rate capability and

the 0.8-rm electrode would show the best. This is exactly the opposite of what is

observed experimentally.

Chapter Summary

This chapter clearly demonstrates the advantages of template-synthesis as a method

to create ideal electrodes for fundamental studies of Li-ion batteries. The electrode

precursor was deposited into the pores of a commercially available polycarbonate

template membrane. The template serves to restrict particle growth as the precursor is

hydrolysized to V205, a known Li-battery cathode material. These V205 wires extend

from the surface of a current collector like bristles of a brush. The geometry of the pores

of the template is imparted onto these wires a nanoporous template yields

nanostructured wires and microporous template yields microstructured wires. We









demonstrated the ability to create structures of reproducible diameters by this template-

synthesis method.

Three of these electrodes with different wire geometries were electrochemically

charged and discharged by a constant-current. By creating particles of the shortest solid-

state diffusion distance, we were able to demonstrate superior low-temperature rate-

capabilities when compared to microstructured control electrodes. However, these

superior discharge characteristics would be achieved with any competitive

nanofabrication method. The ability of template-synthesis to create structures of

reproducible solid-state diffusion distance and surface-area is what differentiates it from

other nanofabrication methods. Analysis of these data identifies the decrease in solid-

state diffusion coefficient of the Li-ion associated with the decrease in temperature to be

the rate-limiting factor in Li-ion batteries. The conclusions presented in this chapter are a

benchmark study in the field of low-temperature Li-ion battery research.














CHAPTER 3
EVALUATION OF SOLID-STATE DIFFUSION COEFFICIENT OF LI-IONS AT
LOW TEMPERATURE

Introduction

Since their introduction in the early 1990s by Sony1, Li-ion batteries have been the

focus of intense research to adapt their design to more demanding systems, such as those

operating on pulse-power and at low-temperature. It is widely known that these batteries

can only provide limited capacity (mAh) at low temperatures.25'60'61 The previous chapter

identified the fundamental breakdown that occurs in these batteries at low temperature to

be the decrease in solid-state diffusion coefficient of the Li-ion (DL,+).25 The critical step

in this analysis was the ability to produce nanowire-based electrodes with controllable

particle-diameter, length, and density. This was accomplished by template-synthesis.69

Template-synthesis is a general nanofabrication method, pioneered in the Martin

laboratory, capable of creating nanostructures of gold,8'9 carbon,12 semiconductors,14

polymers,6 and Li-ion battery electrodes.20'22'24

Since we identified the decrease in DL,+ as the rate-limiting factor in low-

temperature environment,25 the next logical progression is to quantify that decrease.

Aurbach et al. 70,71 have demonstrated the ability of the potentiostatic intermittent

titration technique (PITT) to measure the value DL,+ in Li-ion battery electrodes. This

experimental method is based on a series of small potential steps and data analysis via the

Cottrell-relationship. This was first described in this embodiment by Huggins and

coworkers.72 This method results in the systematic discharge intercalationn) of a Li-ion









battery cathode, such as V205. Lithium is known to reversibly intercalate 1 mol of

charge into an equivalent of V205 host over the potential window 2.8 to 3.8 V vs Li/Li

as described by Delmas et al.68 according to the reversible intercalation reactions of

Equation 2-1.

One of the complications identified in prior studies is the ability to create structures

of reproducible diffusion distance.70 Template-synthesized nanostructured V205

electrodes are again an ideal tool for this analysis, due to the reproducibility of the

nanowires solid-state diffusion distance (a factor closely related to the analysis) and their

ability to equilibrate uniformly after an electrochemical perturbation. The

electrochemical response of these template-synthesized electrodes is not as subject to the

interference of conventional systems, such as concentration-polarization of the Li-ion 25

and the effects of conductive carbon and polymeric binders 5o (absent in this system).

This chapter reports our efforts to quantify this decrease in solid-state diffusion-

coefficient of the Li-ion with decrease in system temperature.

Electrode Synthesis

Template-synthesis was used to fabricate V205 electrodes. This process has been

detailed elsewhere.25 A 3 cm2 section of commercially-available polycarbonate template

(Poretics) was placed atop a section of Pt foil. A 1 ptL drop oftriisopropoxide vanadium

(TIVO) precursor impregnated the pores of the template from the top. This was

performed in an argon environment to prevent hydrolysis of the precursor prior to

complete filling the pores. This assembly is then transferred to a low humidity

environment (antechamber of the glovebox) and hydrolysis proceeds slowly over 8 hours.

The particle-growth is restricted by the pore wall of the template; therefore, a nanoporous









template yields an array of nanoscale diameter wires and microporous template yields

microscale diameter wires. The templates used here are polycarbonate. The nanoporous

template has nominally 50-nm diameter pores, 6 microns in length with a density of 6 x

108 pores cm-2. The microporous template has nominally 0.8-tlm diameter pores, 9

microns in length with a density of 3 x 107 pores cm-2. In order to ensure the micropores

are completely filled, the filling procedure is repeated.25

The assembly is then heated to 80 C for 2 hours in air to ensure complete

hydrolysis. Oxygen plasma (20 Pa, 25 W, 2 hours) preferentially etches the organic

template, leaving structures that have the same geometry of the pores. These structures

extend from the Pt current collector like the bristles of a brush. This assembly is then

heated in flowing 02 for 10 hours to convert the V205 into its crystalline form. Note

there is no conductive carbon or polymeric binder used in this synthesis.

Electrolyte Synthesis

Li-ion battery electrolytes are based on non-aqueous solvents. The primary reason

is that the available potential envelop is greatly extended in such solvents. Aqueous-

based systems are limited to 1.2 Volts as not to engage undesired reactions, i.e., the

splitting of water. The solvent system is a mixture of aprotic carbonates. Ethylene

carbonate (EC) has a high dielectric constant; therefore, it ensures that the LiC1O4

dissociates. However, EC is a solid at room temperature and must be gently heated to

melt prior to use. Therefore, diethyl carbonate (DEC) is added to decrease the viscosity

of the system; which greatly improves ionic conductivity. Dimethyl carbonate (DMC) is

added in proportion with diethyl carbonate in low temperature experiments, because of its

ability to maintain ionic conductivity at low temperatures. These solvent systems are









relatively standard in Li-ion battery research. They are dry and it is imperative that they

remain dry throughout the experiment.

The salt used for all of these experiments is LiC104. There are several examples in

the literature of alternative counter-ions. Perhaps the most common seen in commercial-

systems is LiPF6. Others include LiBF4 and LiAsF6. We selected LiC104, because it is

much easier to handle than the competing salts. LiPF6, for example, is much more

hydroscopic and hazardous than LiC1O4 and has only slightly improved conductivity.

The LiC1O4 is heated on a hot plate in the argon glovebox to aid the electrolyte to remain

dry.

Measurement of Diffusion Distance

As stated previously, a primary advantage of using template-synthesized Li-ion

battery electrodes for this analysis is the ability to create structures of reproducible solid-

state diffusion distance. We have used scanning electron microscopy (SEM) to verify the

geometry of our structures. Figure 3-1A is a low-magnification image that demonstrates

the uniformity of the structures. Figure 3-1B is a high-magnification image of these

structures to identify the solid-state diffusion distance (), determined by the radius of the

nanofiber. This is a valid assumption since these wires are equally accessible to radial

Li-ion intercalation and their aspect ratio is high (i.e., the surface area available for radial

diffusion vastly exceeds that available for axial diffusion). This value is 50 nm. The

uniformity of diffusion distance of the particles should provide a single characteristic

diffusion time-regime, where all are subject to semi-infinite diffusion at the same times.























































Figure 3-1: Scanning electron micrograph. A) This low magnification image shows the
uniformity over the large number of wires that constitute the electrode. B)
This high magnification image is used to identify the solid-state diffusion
path-length (radius of a nanowire) to be 50 nm.









Electrochemical Investigations

Potentiostatic Intermittent Titration Technique (PITT)

Each electrode is characterized in a three-electrode jacketed cell. Temperature is

set and maintained by a cooling circulation bath. Lithium ribbon is used as both the

reference and counter electrodes. The electrolyte is 1 M LiC104 in an equivolume

solvent mixture of ethylene carbonate, diethyl carbonate, and dimethyl carbonate

(EC:DEC:DMC.) The PITT electrochemical experiments begin with the electrode

completely in the charged (as synthesized) state. A -15 mV potential step polarized the

electrode for the period of 45 minutes. This process was repeated over the entire range of

3.8 to 2.8 Volts versus Li/Li The temperature is then reduced to the next level. A slow

linear potential sweep (0.1 mV s-1) is used to return the electrode to the charged-state

after 3 hours has elapsed for temperature equilibration. The experimental potential-step

series was repeated. The cyclic voltammetric experiments are performed in an identical

system with a scan rate of 0.1 mV s1.

A typical electrochemical response (current peak and decay) after the negative

potential step is shown in Figure 3-2A. Since the electrode material is synthesized in the

delithiated (x = 0) charged state, the potential steps are negative (AE = -15 mV), so as to

intercalate Li with every step. Therefore, the currents are negative as well. Note that a

disproportionate quantity of charge is intercalated at the potentials that correspond to the

voltammetric peaks, as expected. After the 45 minute duration of the polarization, the

currents have reached a small, but non-zero, steady state value. Preliminary analysis of

these data follows the work of Aurbach et al.70 In that experiment, they identified distinct

kinetic regions during the charge reaction of anodic graphite by examining the It1/2 versus

log t plot.






48


0.0

^ A

< -0.1
E
4-'

S-0.2
0


-0.3



0 1000 2000 3000
Time s


-1x104 B



-2x10-4



S-3x104



-4x10-4
Cottrellian Value

-5x 10-4 --------
-5I 0 I I I I t
-1 0 1 2 3 4
Log (t / s)


Figure 3-2: Electrochemical response of nanostructured V205 electrode to a AE = -15
mV potential step. A) Current-time transient recorded for 45 minutes as the
current decayed to a steady-state value. B) Current-time transient shown in
Cottrell-plot form. The plateau represents the Cottrellian value It'2.









Figure 3-2B represents the current-time transient (Fig. 3-2A) as a similar It1/2

versus log t plot. This method conveniently identifies the solution to the Cottrell

relationship73 (i.e., where semi-infinite linear diffusion plays a major role), as it is the

plateau of the plot. This particular plateau (3.200 V) lasts approximately 20 s in duration

and is identified on the figure. The kinetic regimes70 are clearly seen and correlate to 1.)

the time-constant associated with the double-layer capacitance and the charge-transfer

resistance (the short-time linear region prior to the plateau) 2.) the solution to the Cottrell

relationship, where the value of It1/2 is independent of time (the plateau) and 3.) the

transition into the finite diffusion regime, where enough time has elapsed that the

diffusion layer has propagated to encompass the entire nanowire [ < ](2Dt)] (The

long-time linear region post to the plateau.)

Figure 3-3 demonstrates the potential-dependency of the time-region associated

with this plateau. As the potentials approach that of the voltammetric peaks the plateaus

shift to longer times and more negative values. This data can be easily converted to the

solid-state diffusion coefficient of the Li-ion by means of Equations 3-1 and 3-2.

Equation 3-1 establishes tau (c) as the characteristic diffusional time constant. Equation

3-2 expresses the simple relationship of this time constant to the solid-state diffusion

coefficient of the Li-ion (DL,+).


S 2= (3-1)


2 (3-2)
DL,+ = (3-2)






50


0.0-
3.050
S -.........- ,.
-1.0x10-4 ...... 3.125



< -2.0x10-4 3.215 /
-4-

-3.0x1 0.4 "\ /

3.185 /

-4.0x10-4 "
3.200


-1 0 1 2 3

Log (t s)
Figure 3-3: Cottrell-plots as a function of potential and thus level of intercalation. At the
potentials corresponding to the voltammetric minima, the Cottrellian value
It/2 is shifted to more negative values and occurs at longer times.

Again, It1/2 is determined from the Cottrell-relationship plot (the ordinate value at

plateau) (see Figure 3-2B); AQ represents the quantity of charge intercalated per step

(this value is determined via a digital integration of the I-t response using the CorrView

software package); fis the characteristic diffusion path length (nanowire radius). The

value for AQ is adjusted to account for the effect of background currents. Figure 3-4

shows this value of AQ at each E for the room temperature experiment.











0.000- 1o D15
0o 000000 /



-0.010 -

2-0.015
CU


-0.020 -


-0.025
-0.025 0]
I I I I I
3.1 3.2 3.3 3.4 3.5
Potential/ E

Figure 3-4: Charge intercalated per potential step. The majority of charge is intercalated
at the potentials corresponding to voltammetric minima and galvanostatic
plateaus.

Determining DLi+ (Data analysis)

Figure 3-5 clearly shows the dependency of DL,+ on T and E. Note that the y-axis

(DL,+) is plotted on a log scale. First, let us discuss the value of DL,+ at room temperature

(blue curve.) This curve exhibits similar dependence of D versus E seen by several

research groups.71'74-77 Minima are observed at potentials of the voltammetric peaks and

are attributed to the effects of ion-ion interaction on the energy of activation.70'78 This is

intuitive as the movement of ions between intercalation sites becomes retarded in the

presence of more of these species. Also of note, the particular values are within the

approximations made by other groups.75'77






52


1011



101
1 7

102

S1010
10-
o 10. o







10- 103

1014
3.1 3.2 3.3 3.4 3.5

Potential I E

Figure 3-5: Solid-state diffusion coefficient as a function of potential. Y1-axis is the
solid-state diffusion coefficient and the Y2-axis corresponds to characteristic
diffusion time constant (c) at room temperature. Note that both y-axes are
plotted on a logarithmic scale.

Previously, we have seen less charge delivered from template-synthesized

nanostructured electrodes as the temperature is decreased.25 We attributed the decrease

to a decrease in DL,+. Comparing data collected at room temperature (blue) and those at 0

C (orange), we see this is indeed the case. When comparing these curves, we notice the

values of the orange curve are less than the blue curve in every instance. The minima are

again observed and again correlate to the potentials of the voltammetric peaks. The

minima of the low temperature curve are shifted slightly less positive of those observed

with the room temperature curve. This is attributed to sluggish electrokinetics and is

discussed in detail later.






53


1 E-11




C 1E-12
E

A


A A A A
S1E-13-

o : A;

Potential / V vs. Li








o (green) curve. Again the decrease in DL is seen, particularly at the low-potential
S1E-145 I A i


1E-15- --------------------
3.0 3.1 3.2 3.3 3.4 3.5

Potential / V vs. Li/Li'


Figure 3-6: Dependence ofDL,+ conducted at 25 C (Blue squares), 0 C (Orange
circles), and -20 C (Green triangles).

Figure 3-6 shows the trend continues in a comparison of the 0 C curve to the -20

C (green) curve. Again the decrease in DL,+ is seen, particularly at the low-potential

region. The low-potential DL,+-minimum at -20 C (2.2 x 10-15 cm2 s-1) is an order of

magnitude lower than the minima at 25 C (2.3 x 10-14 cm2 S-1). These data confirm our

previous conclusions that the rate-limiting factor of these electrodes at low-temperature is

the decrease in solid-state diffusion coefficient. Again, the minima potentials are more

negative than those at 0 oC.

Activation Energy of Diffusion Process

Energy of activation represents the thermal energy barrier that must be overcome

for a process to proceed. Equation 3-3 is the general form of the Arrhenius equation that










describes the relationship between the rate constant k and the temperature T. Equation 3-

3 is the form of the general Equation 3-4 and conveys the relationships of our data to EA.

k= Aexp(-E /RT) (3-3)


InDL =InA (EA/R) (3-4)


As stated earlier and clearly observed in the figures above, the value ofD is

dependent on potential. We selected the D at potential of 3.24 V for this activation

energy analysis, because it is between the two minima at every temperature, so the

electrodes are in comparable states. The slope of this graph (Figure 3-7) is -5700 K and

represents the EA/R value shown in Equation 3-4. Therefore, we can assign EA a value

of 690 J. This represents the thermal energy barrier associated with the solid-state

diffusion of the Li-ion.




-31 \
E = 3.24 V (between phases)

) y y=-5739.3x 11.093
E -32 R = 0.9422



-33-

\
\

-34 -
3.4 3.6 3.8 4.0
(1 /) *103/K1


Figure 3-7: Arrhenius plot to solve for activation energy of diffusive process. The slope
of this line represents -EA/R. All values of D were taken at E = 3.24 V. At
this potential the intercalation is between phases of intercalation (see text).









Cyclic Voltammetry

To further bolster this conclusion, we performed cyclic voltammetry (CV)

experiments at all three temperatures with nanowires and microwires. In this experiment,

the parameter AEpk is the difference in the potentials at which the peak currents occur

during the charging and discharging reactions. This parameter provides information

concerning the electron-kinetics of the reaction (Equation 2-1). 79

A typical CV of a 0.8-[tm electrode is shown in Figure 3-8A. The data for the peak

separation at the low-potential (AEpkl) for the nanowire and microwire electrode at the

three different temperatures are presented in Figure 3-8B. These data show a correlation

between the decrease in temperature and an increase in AEpk. Also from these data, the

AEpk for the nanowire and microwire are very similar at each temperature. Therefore, we

can state that the electron-kinetics of this process does decrease with temperature;

however, it is not the dominant effect, as the decrease is virtually independent of these

electrode geometries.











0.3-


0.2-


0.1 -


0.0-


-0.1


-0.2-


-0.3-


3.4


3.6


3.8


Potential / V


---- Nanostructured electrode I
-e- Microstructured electrode
II


-20 -10 0


10 20 30


Temperature / C


Figure 3-8: Cyclic voltammetric responses. A) Typical cyclic voltammetric response of
0.8-ptm electrode at room temperature. Scan rate =0.1 mV s-. B)
Comparison of AEpkl from cyclic voltammograms of nano- and micro-
structured electrodes at various temperatures.


3.2


3.0


100-





80-












Chapter Summary

In this chapter, we demonstrated the measurement of solid-state diffusion

coefficient of Li-ions (DL,+) as a function of potential and temperature. This is

accomplished using the potential intermittent titration technique (PITT). PITT acquires

data from small potential steps and analyzes data via the Cottrell relationship. Template-

synthesis created electrode particles of reproducible and determinable solid-state

diffusion distance. This method mitigates a critical data analysis issue identified by other

research groups.

DL,+ was seen to vary greatly with potential. This potential represents various

states of charge (Li-ion intercalation level) of the electrode. Minimas were observed in

the potentials close to the voltammetric peaks. These peaks represent the potential at

which the majority of charge is actively intercalating causing ion-ion interactions to

decrease the diffusitivy of the ions. DL,+ decreases over the range of potentials with

temperature. We performed these experiments at five different temperatures ranging

from room temperature to -20 C. This generates the data required to identify the

activation energy for the diffusion process by an Arrenheius relationship.

We also report the results of cyclic voltammetric experiments where AEpk is seen to

be dependent on temperature, but independent of electrode-geometry (solid-state

diffusion distance and electrode-particle area per template-area. Since AEpk is a kinetic

parameter, kinetics is again identified as a secondary contributor to the decrease in rate

capabilities discussed in detail in the previous chapter.

In discussions on the research contained in this chapter, the validity of this PITT

method has come under debate. The Cottrell equation is a globally accepted and









implemented method to determine the diffusion coefficient of small particles in aqueous

systems diffusing in the semi-infinite regime to a planar electrode. The debate arises due

to the uniqueness of this solid-state intercalation process. This method is analogous to

ion diffusion through a polymer-coated electrode. Research of this type (e.g., Nafion-

coated electrodes) gained popularity with the last generation of scientists. Collectively

they describe this parameter as an "apparent" diffusion coefficient. Debate ensues on the

spurious nature of the relationship of diffusion coefficient to voltammetric peak

potentials. Many researchers believe this to be an artifact of the application of the

Cottrell method to the battery intercalation mechanism.80 It would be remiss to not

mention this debate in a discussion of this method. At the time of publication of this

dissertation, the debate continues.80',s














CHAPTER 4
A HIGH-RATE, NANOCOMPOSITE LIFEPO4/CARBON CATHODE

Introduction

Lithium-ion batteries are the power source of choice for portable electronics, a

multi-billion dollar market.1 This outstanding commercial success has spawned great

international interest in applying this technology to systems that demand higher power,

such as the electric component of hybrid vehicles.3 This would require new electrode

materials that are less expensive, more energetic, and more environmentally friendly than

the present ones. Of particular interest is the olivine-structured LiFePO4 cathode

developed by Goodenough and co-workers,82 which offers several appealing features,

such as a high, flat voltage profile and relatively high theoretical specific capacity (168

mAh g-1), combined with low cost and low toxicity. However, the current designs of

cells based on LiFePO4 technology have not shown the ability to deliver high specific

capacity at high discharge rates. For this reason, LiFePO4 is currently not a promising

electrode material for high-rate and pulse-power applications.

The discharge reaction for LiFePO4 (Equation 4-1) entails intercalation of Li+

(from the contacting electrolyte phase) along with an equivalent number of electrons into

the electrode material. The rate capabilities of LiFePO4 are limited primarily by its

intrinsically poor electronic conductivity and by the low rate of Li+ transport within the

micron-sized particles used to prepare the battery electrode. A number of approaches

have been proposed to improve this material's inherent poor electronic conductivity,









including carbon coating,83 nano-fibril textures,84 optimized synthesis procedures85 and

foreign metal doping.86


discharge
FeP04 + x Li + x e- LixFeP04
charge 4-1


We describe here a new approach for preparing high rate-capability LiFePO4

electrodes. This approach builds on the application of the template synthesis method for

preparing nanofiber Li-ion battery electrodes.17'19'20'22'24 However, the method was

modified such that the template-prepared LiFePO4 nanofibers are mixed with carbon

particles, and coated by thin carbon films, to yield a nanocomposite LiFePO4/carbon

matrix. As we have shown previously,19'20,22 the nanofiber morphology mitigates the

slow Li+-transport problem, because the distance Li+ must diffuse within the electrode

material is minimized. The carbon matrix obtained with this new template-based method

obviates the poor electronic conductivity problem. These nanocomposite

LiFePO4/carbon electrodes can deliver a capacity of 150 mAh g-1 at a rate of 5 C and

maintains a substantial fraction of the theoretical capacity even at rates exceeding 50 C.

To our knowledge, performance at this level has never been achieved by other types of

LiFePO4.

Electrode Synthesis

The sol-gel method developed by Croce et al. was employed for the synthesis of the

electrode precursor solution."7 Accordingly, LiOH monohydrate (Aldrich), ferric nitrate

nonahydrate (Fisher), ascorbic acid (Fisher), phosphoric acid, and ammonium hydroxide

were used to create the LiFePO4 precursor solution. The template membranes were









commercially available polycarbonate filters (Poretics). Pt foil (2.5 x 1.5 x 0.025 cm,

99.99% purity, Aldrich) was used as the current collector. LiC104 (Aldrich), ethylene

carbonate (Aldrich), diethyl carbonate (Aldrich) were used as received for preparing the

electrolyte.

An approximately 1 cm2 piece of the polycarbonate filter was immersed in a

precursor solution of 1 M LiFePO4 in water in 24 hours. This solution was synthesized

with ferric nitrate, lithium hydroxide, and phosphoric acid in proportions for a 1:1:1

molar ratio. Ascorbic acid, in equimolar ratio to the total metal (Li plus Fe2+) content,

aided the synthesis by forming a complex with the iron, and ammonium hydroxide was

used to raise the pH to -2. The impregnated template was then attached to a Pt current

collector and dried in air at 80 C for 10 minutes. A 10 ptL drop of precursor solution

was placed on top of the dried filter to increase the amount of active material in the

sample. It was dried again under the same conditions. This assembly, template intact,

was heated in a reducing atmosphere of flowing Ar/H2 gas (95/5 %). The temperature

was slowly taken over the course of 4 hours from 250 C to 650 C and held there for 12

hours. This procedure yields the Fe(II) oxidation state necessary for LiFePO4 and

decomposes the template into the carbon necessary for improved conductivity. A

schematic of this process is shown in Figure 4-1.

In all of our previous examples of template-synthesized nanofiber electrodes19'20,22

after synthesis of the nanofibers the template would be totally removed to yield the

nanofibers protruding from an underlying current collector surface like the bristles of a

brush. For the LiFePO4 nanofibers prepared here we instead pyrolyzed the polycarbonate

in a reducing Ar/H2 environment at a temperature of 650 C. This yields graphitic carbon









particles intimately mixed with the LiFePO4 nanofibers, and thin carbon films that coat

these fibers.



Polycarbonate
template



LiFePO,-solutiOn
irnpreg a led














Figure 4-1: Schematic of template-synthesis of LiFeP04/carbon nanocomposite
electrode.

Two different templates were used. A template with a slightly larger pore-diameter

of 100 nm is used for microscopy. A template of 50 nm pore-diameter is used for

electrochemistry. The template being used for each experiment is clearly stated.

Structural Investigations

Scanning Electron Microscopy

A FEG SEM JEOL JSM 6335F instrument was used to obtain SEM images of the

electrode. The nanostructured electrodes on Pt foil were prepared for imaging by

attaching them to a SEM stub by conductive copper tape. No conductive metal sputtering

was required for the composite electrodes, but a thin Au/Pd sputtering was applied to the

bare LiFePO4 wires (Figure 4-2C) prior to imaging.









To obtain detailed images of the morphology of the nanocomposite structure we

found it prudent to use a template with nominally 100 nm-diameter pores. The larger

LiFePO4 nanofibers obtained from this template are more easily imaged with scanning

electron microscopy. This template was also 6 |tm thick and had a pore density of 4x108

pores per cm2. A lower magnification scanning electron microscopic (SEM) image of the

resulting LiFePO4/carbon nanocomposite electrode is shown in Figure 4-2A. Because of

the relatively low porosity of the template, there is substantial void volume, but in

analogy to our prior nanofiber electrodes of this type, the LiFePO4 nanofibers can be seen

crossing through this void space.19,20,22 Higher magnification images (Figure 4-2B) show

that there are carbon particles dispersed through this matrix and that the LiFePO4

nanofibers are coated with thin carbon films. To prove that these fibers are coated with

carbon films, we prepared fibers in the same template, but instead of then pyrolyzing the

template, we simply removed it quantitatively by burning it away in 02 plasma. Hence,

in this sample the fibers are not coated with carbon films.19,20,22 An image of these fibers

is shown in Figure 4-2C.

The fibers from the pyrolyzed membrane (Figures 4-2A and B) have a textured

surface morphology and have a larger diameter than the fibers from the plasma-removed

membrane (Figure 4-2C). Both the larger diameter and the textured surface are due to the

carbon coating surrounding the fibers from the pyrolyzed membrane.




















































Figure 4-2: Scanning electron micrographs. A) Lower magnification image of the
nanocomposite LiFePO4/carbon electrode. B) Higher magnification image of
the nanocomposite LiFePO4/carbon electrode. Composite fiber diameter is
350 nm. C) Image of LiFePO4 electrode synthesized by template dissolution
method (absent of carbon). LiFePO4 fiber diameter is 170 nm.










X-ray Diffraction Studies (XRD)

The X-ray diffraction data are presented in Figure 4-3. The pattern represents the

diffraction of LiFePO4/carbon on quartz. The quartz substrate was used because the Pt

foil would give a large signal in this region. This signal would dwarf the relative signal

from the LiFePO4/nanocomposite. The lines below correspond to the accepted literature

values for LiFePO4, olivine group, triphylite subgroup, published as JCPDS 40-1499.

The large amorphous wave ranging from 15 to 30 20 is characteristic of disordered

carbon. Our group has seen similar diffraction patterns when working with disorder

carbons.






Nanocomposite
50 nm template

Microcomposite
0.4 jIm template



Si JCPDS 40-1499

10 20 30 40 50 60 70
20


Figure 4-3: XRD pattern from nanocomposite, microcomposite, compared to accepted
literature values for LiFePO4. The substrate is quartz. Acquisition time is 10
s.

X-ray Photoelectron Spectroscopy

The XPS studies were performed on a Kratos XSAM 800 spectrometer with Al-Kc~

exitation (180W). The sample, mounted onto a stainless steel sample stub, was inserted

into the sample analyzer chamber by means of a quick insertion probe, and spectral









acquisition commenced after the pressure decreased to 5x 109 Torr. High resolution C1s

spectra were recorded at a take-off angle of 750 relative to the sample surface. Data

analysis was done by using the DS 800 software package. Peak positions were all

referenced to 70.9 eV for the Pt4f- : peak (literature value for metallic platinum used as

sample support).

The presence of carbon was also confirmed by XPS analysis (Figure 4-4). The

high resolution C1s spectrum may be fitted to three peaks with binding energies of 283.2

0.5, 284.7 0.3 and 286.0 + 0.3 eV. According to previous work by the Martin

laboratory11, the lowest binding-energy peaks may be assigned to graphitic (283.2 eV)

and amorphous (284.7 eV) carbon. The predominance of the 284.7 eV peak indicates

that most of the carbon present is amorphous; this is confirmed by the X-ray diffraction

data. The peak at the highest binding-energy (286.0 eV) is due to oxygen-containing

surface functional groups. Oxygen functional groups are nearly always observed on

carbon surfaces that have been exposed to air.1









100- Cls

80-

S60o

< 40

t 20
C)
( 0--

288 286 284 282
Binding energy /eV


Figure 4-4: X-ray photoelectron spectroscopy data for Cls peak for the carbon in the
nanocomposite electrode. Relative peak areas show amphorous form is
dominant (-80%), but graphitic carbon is also present (-20%).

Analysis of Carbon Content

We estimated the weight percent of carbon in the composite electrode

gravimetrically using an Ultra-Micro-Balance SC2 (Satorius). LiFePO4 nanofibers were

synthesized within the pores of the polycarbonate template on a Pt current collector, and

the polymer was pyrolzyed as described previously. The mass of this composite,

corresponding to the masses of the Pt current collector, the nanofibers and the carbon,

was obtained. This composite was then heated in air at 600 C for 30 minutes to bum off

the carbon, and the mass was measured again. The difference between these two masses

is the mass of carbon in the composite. Replicate analyses on four identically prepared

samples gave a carbon content of 7 + 4 % in the nanofiber/carbon composite.









The presence of carbon in this matrix was confirmed by X-ray diffraction analysis,

Raman spectroscopy, and X-ray photoelectron spectroscopy (XPS). A Raman spectrum

(data not shown) of the nanocomposite LiFePO4/carbon electrode was also taken. The

most intense peak (centered around 1000 cm-1) and others at the lowest frequencies

correspond to the P04 stretching modes ofLiFePO4. Bands at 1350 cm-1 and 1580 cm-1

are assigned to carbon.88

Electrochemical Investigations

For the electrochemical studies we wanted a template with small-diameter pores so

that correspondingly small-diameter nanofibers of LiFePO4 would be obtained. For this

reason we used a template with nominally 50 nm-diameter pores for the electrochemical

studies. This template was 6 |tm thick and had a pore density of 6x108 pores per cm2 of

surface area. Cyclic voltammetric and constant current charge/discharge experiments

were performed in a three-electrode cell using a Solartron 1287 Potentiostat, driven by

the CorrWare software package. The electrolyte for these experiments was 1 M LiC104

in ethylene carbonate:diethyl carbonate (3:7 v/v). The nanostructured LiFePO4 was the

working electrode and lithium ribbon was the reference electrode and counter electrode.

Potentials are reported versus the Li/Li reference. The experiments were performed in

the inert atmosphere of a glovebox filled with argon gas.

This unique LiFePO4/carbon nanocomposite electrode should be ideally suited for

high rate applications because the distance that Li+ must diffuse in the electrode material

is limited to the radius of the nanofibers19,20,22 and because the carbon matrix should

provide for good electronic conductivity through the composite. This was confirmed

experimentally via electrochemical characterization of these electrodes. Again, the









electrochemical studies were done on electrodes prepared in templates with 50 nm-

diameter pores.

Cyclic Voltammetry

Cyclic voltammograms (CVs) for the nanocomposite electrode show reversible

waves centered at 3.5 V associated with the reduction and re-oxidation of the LiFePO4

(Figure 4-5A).87,89 The difference in peak potentials (AEpk) for this nanostructured

electrode is 60 mV (Figure 4-5A). This may be contrast to CVs for conventional, non-

nanostructured, LiFePO4 electrodes, which at comparable scan rates and in comparable

electrolyte solutions show AEp > 200 mV. 87,89 This clearly shows that our

nanostructured LiFePO4 electrodes lack a resistive component that is present in the

conventional electrodes. This verifies the major premise of this work that the

conductive carbon matrix overcomes the inherently poor electronic conductivity of

LiFePO4.

Rate Capabilities

The constant-current discharge curve (lithium insertion, Equation 4-1) for the

nanocomposite electrode shows the flat voltage plateau centered at 3.5 V, characteristic

of LiFePO4 (Figure 4-6A)83'87,89. At the lowest discharge rate used (3 C), the specific

capacity for the composite is 165 mAh g-l, essentially identical to the maximum

theoretical capacity, 168 mAh g-1. While capacity falls off with increasing discharge rate

(Figure 4-6B) the electrode retains 36% of its theoretical capacity at discharge rates as

high as 65 C. There are no other examples in the literature of LiFePO4 being discharged

at such enormous rates.
































2.8





30


20

<
S10
4-

0 0

-lo
-10-


-20


-30 -


3.0 3.2 3.4

Potential / E


3.4
Potential / E


Figure 4-5: Cyclic voltammograms. A) Cyclic voltammogram for the nanocomposite
LiFePO4/carbon electrode prepared using a template with 50 nm-diameter
pores. Scan rate = 0.1 mV s-. B) Comparison of voltammetric response to
theoretical response exhibiting diffusional tailing. The diffusional wave was
generated from the normalized X function.79


I I ( I I I I













3.6-


3.4-



3.2-


3.0-


0 40 80 120

Specific Capacity / mAh g-1


B
0
[]
0]


-r I I 1 T I I I I r1 i


0 10 20 30 40 50
C-Rate / C, h1


Figure 4-6: Constant current experiments. A) Constant-current (3 C) discharge of
LiFePO4/carbon nanocomposite electrode. B) Specific capacity versus C-rate
for the nanocomposite LiFePO4/carbon electrode prepared using a template
with 50 nm-diameter pores.


160-



120-



80-



40-


60 70









Chapter Summary

We have described here a new type of template-prepared nanostructured LiFePO4

electrode. LiFePO4 was developed as an alternative cathode material to LiCoO2, the

industry standard. LiFePO4 is an energetic material with advantages of both cost and

safety. However, this material is inherently electronically insulating. This material is

also susceptible to the concentration polarization with larger micron-diameter particles.

Our nanocomposite electrode consists of nanofibers of the LiFePO4 electrode

material mixed with an electronically conductive carbon matrix. It is created by a

modified template-synthesis procedure, where the polymeric template is pyrolzed. This

carbon matrix coats the LiFePO4 nanostructures and provides an improved electron

pathway. Therefore, both rate-limiting effects of electron-transfer and ion-transport are

delayed. This unique nanocomposite morphology allows these electrodes to deliver high

capacity, even when discharged at extreme rates necessary for many pulse-power

applications. This nanocomposite is a specific example of nanotechnology overcoming

the limits of conventional technology. We are currently working toward developing a

commercially viable route for preparing such nanocomposite electrodes.














CHAPTER 5
MAGNESIUM-ION INTERCALATION INTO TEMPLATE-SYNTHESIZED
NANOSCALE ELECTRODES IN THE ABSENCE OF CARBON

Introduction

Li-ion batteries have powered the mobile electronic industry. Though these

energy-storage devices have enjoyed tremendous commercial success, they remain the

subject of international research. This is primarily due to the consumer driven demands

of an increasingly portable world, specifically, the desire for increased pulse-power and

decreased size. The Li-ion intercalation reaction (mechanism by which the host-

electrodes store charge) is shown previously. Lithium-ions were chosen because of their

low reduction potential (-3.05 vs. NHE) and their small dimensions facilitate both the

intercalation process and the solid-state diffusion process.

We have had great success synthesizing and characterizing Li-intercalation into

template-synthesized nanostructured electrodes. Template-\yii hei, is a general

nanofabrication method pioneered in the Martin group. Employing this method, we have

shown that with no addition of inactive ancillary components (carbon or polymeric

binder), template-synthesized nanostructured electrodes are capable of reversibly storing

and delivering a dramatically greater portion of the electrode's theoretical capacity (mAh

g-1) than similarly constructed microstructured electrodes. These nanostructured

electrodes are capable of this, due to their small solid-state diffusion distance and large

surface-area per gram. These traits serve to delay the effects of concentration

polarization and sluggish electron-kinetics, respectively.









Dramatically fewer researchers have explored the field of polyvalent-ion

intercalation.90-92 The charge-storage mechanism for this device is similar to that of the

Li-ion battery; however, whereas the Li-ion intercalation reaction has 1 electron

equivalent per Li-ion each Mg-ion has 2 electrons. Simply stated, by doubling the

quantity of electrons available, one can store an amount of specific energy that is not

stoichiometrically possible via the Li-ion intercalation process. However, the diffusivity

of these ions is further hindered by its polyvalent nature and larger dimensions.

Therefore, a structure that has proven to be able to delay the limiting effects of such

phenomena may be able to successfully intercalate such ions. Template-synthesized

V205 nanostructured electrodes have, in fact, this ability.25 Here we detail these studies

and demonstrate the ability to reversibly (de)intercalate Mg2+ into/out of the host

(template-synthesized V205 nanostructures) by Equation 5-1.

discharge
x Mg2+ + 2x e- + V205 MgxV205
charge 5-1

Other advantages of Mg-based technology are that concerns still exist with the

safety and stability of lithiated carbon as well as the expense of the constituents of the

battery.

Electrode Synthesis

As stated, these electrodes were created by the template-synthesis method. This

process is similar to those described previously; however, the notable exception is that

the current collector is Indium Tin-Oxide (ITO)-glass. Another research group has

demonstrated the usefulness of the substitution.91 This is a piece of glass that has a thin

layer of conductive Indium Tin-Oxide deposited onto it. Therefore, one of the surfaces of









the glass is electronically conductive. Electrodes were synthesized on both ITO-glass

substrates as well as platinum.

A commercially available (Poretics) polycarbonate filter was used as the template.

This membrane has nominally-cylindrical track-etched pores (- 50 nm in diameter) that

run its entire length of 6 microns. There are 6 x 108 of these pores per cm2 of template

area. In an inert environment (Ar-filled gloebox) a 3 cm2 section of template is placed

atop a section of electrically conducting ITO-glass. A 1 p.L volume of TIVO

(triisopropoxide vanadium) was placed atop the template membrane. The precursor's

low viscosity allows it to flood the pores of the filter. This assembly is then moved to a

low water environment (antechamber of glovebox). In the presence of the atmospheric

water, the TIVO precursor hydrolyzes. At this point, the assembly is heat to 80 C for 2 h

in air to ensure complete hydrolysis and conversion to the gel-phase. Any top surface

layer is then removed with a damp cotton swab.

The assembly is then placed into oxygen plasma (20 W; 10 Pa; 2 hours) to etch

away the organic template. The resulting nanowires extend from the ITO-glass (or

platinum) communal current collector and mirror the geometry (length, diameter, and

number density) of the pores of the membrane. The electrode is then heated to 400 C in

flowing 02 for 10 hours to form crystalline V205 nanowires.

XRD Studies

X-ray diffraction is a convenient experiment to identify the crystalline phase of the

electrode material. Like SEM, it is non-destructive. In a powder x-ray diffraction (XRD)

experiment, a quantity of powder-form sample is exposed to X-rays at a sweeping angle.

These incident X-rays deflect off the sample onto a detector at a pattern relative to the









crystalline lattice structure of the powder. The arrangement of the atoms in a relationship

to each other can be determined from the pattern and Bragg's law. It is displayed in

terms of Miller indices (hkl), coordinates of parallel planes in a unit cell.



c)i

'

.i-

I-


I-
-*1'





15 20 25 30 35
20


Figure 5-1: X-ray diffraction experiment. The sample is nanostructured V205 on Pt.
Acquisition time =10 s. The discrete lines represent the accepted values for
orthorhombic phase V205 (JCPDS 41-1426).

This continuous line is the response of a nanostructured V205 electrode. The

discrete lines represent the internationally accepted pattern for orthorhombic V205 as per

JCPDS card 41-1426, which replaced card 9-387. These two patterns are similar, with

only the relative intensity of the peak at 20 = 26.1 (representing the (110) plane) being

smaller than the standard. The 20 range was limited to 15 to 35, so as to not record the

intense Pt peak at 39.8, which would dwarf the relative intensities of the V205 sample.

An acquisition time of 10 s per step was used to increase resolution of the pattern.









The XRD pattern of a powder can also be used to estimate the crystallite size of the

sample. This relationship is described in the following Equation 5-2 first described by

Scherrer93 and refined by Biscoe and Warren.94

0.89* A 5-2
t- 5-2
B cos O

The term t is the size of the crystallite. The term X is the wavelength of the incident

X-rays (Copper c = 1.54 Angstroms). The OB is the Bragg's angle being analyzed. The

term B represents the full-width half-max of the crystallographic peak in radians.

Here the data from the most intense 001 plane is compared. Using this equation,

the nanostructured electrode has an average crystallite size of 20.1 nm; whereas, the

microstructured electrode (data not shown) has an average crystallite size of 32.3 nm.

So, qualitatively the particles of the micronstructured electrode are lager on a

crystallographic plane than the nanostructure electrode. This also indicates that each wire

may consist of a number of smaller crystallites.

Electrochemical Investigations

The electrode was characterized in a three-electrode half-cell configuration. The

cell differs significantly from the ones discussed earlier. The V205 nanostructures on

ITO-glass (or platinum) were the working electrode. Lithium must be excluded from the

cell to ensure that all charge-storage is due to Mg2+. Polished Mg ribbons were used as

the reference and counter electrodes. The electrolyte is 1 M Mg(C104)2 in dry

acetonitrile. This system been used by other researchers in this field.91 This cell was

assembled and characterized in the Ar-filled glovebox.










Ferrocene Pseduo-Reference

As stated previously, Li-ions must be excluded from the electrochemical system to

ensure that the response is strictly from the Mg-ion intercalation. Therefore, the

reference electrode in this experiment is Mg/Mg2 This reference reaction is not as

common, especially in this particular electrolyte system. We use the well established

ferrocence / ferrocenium redox couple as a pseduo-reference.

The first system has lithium ribbon as the counter and the reference electrode.

Platinum metal is used as the working electrode. The electrolyte is 1 M LiC104 in

EC:DEC (3:7). Ferrocence is also dissolved into the solution. A cyclic voltammogram is

taken using this system. Figure 5-2 shows the potential of the reversible ferrocence redox

reaction to be 3.25V vs. Li/Li+


x









N
E
0






z
O


1.0

0.8

0.6

0.4

0.2

0.0

-0.2

-0.4

-0.6

-0.8

-1.0
0.5


-- Mg-system
-- Li-system


I I I I
1.0 1.5 2.0 2.5

Potential / E


3.0 3.5 4.0


Figure 5-2: Cyclic voltammogram of ferrocence in Li-based and Mg-based
text). Scan rate =50 mV s1.


systems (see









The comparative system has magnesium ribbon as the counter and the reference

electrode. Platinum metal is again used as the working electrode. The electrolyte is 1 M

Mg(C104)2 in dry acetonitrile. Ferrocence is also dissolved into the solution. A cyclic

voltammogram is taken using this system. The potential of the reversible ferrocence

redox reaction is 1.5 V vs. Mg/Mg2+. The same redox reaction occurs in the Mg-based

system at a potential that is 1.75 V negative of the potential at which is occurs in the Li-

based system. Therefore, the Mg/Mg2+ reference is 1.75 V negative of the Li/Li

reference (i.e., 0 V versus Mg/Mg2+ is +1.75 V versus. Li/Li+).

Cyclic Voltammetry

ITO-glass current collector

The CV experiment (Figure 5-3) was performed at the scan rate of 0.1 mV s-1. The

electrolyte is the Mg-based system described above. There is a peak, indicating a charge-

storage process, during both intercalation and deintercalation scans. The most notable

characteristic is the large peak separation associated with these peaks. Interestingly, this

low-potential peak does represent the deintercalation of Mg2+, because if the experiment

is terminated at 1.5 Volts, then the high-potential peak is not seen again in the next scan

(i.e., it becomes an irreversible reaction). To date, we are uncertain as to the source of

this phenomenon. It appears too large to be iR drop, an effect of arising from the

uncompensated resistance between the reference and working electrodes. For such small

currents, the resistance would need to be on the order of 100 kM. The conductivity of the

solution was measured to be 25 S cm-1. This translates to a Runcomp of> 25 Q in this

system. (The distance between working and reference electrode would need to be 4 m for






80


this to be the source of the peak separation.) Also note that the Runcomp was estimated by

Electrochemical Impedance Spectroscopy to be approximately 10 Q.

The effect is most probably related to either the vulnerability of the ITO layer

during the thermal processing step or a resistive passivation layer. The poor thermal

stability of ITO is undoubtedly an effect, because the mass of the ITO-glass post-heat

treatment is significantly lower mass prior to the heat treatment. Other groups91 were

able to use ITO-glass as a substrate, because they apply the active electrode material

post-processing (i.e., it is not necessary for the current collector to be compatible with the

fabrication process of the material synthesis, as it is in our case).


Potential / E vs. Li/Li+
1 2.5 3.0 3.5 4.0 4.5

8

6

4

< 2-

0
-2
C
D -2 -
3 -4
0
-6

-8

-10

-12- i
0.5 1.0 1.5 2.0 2.5 3.0

Potential / E vs. Mg/Mg2+


Figure 5-3: Cyclic voltammogram of ITO-glass/MgxV205 nanowire electrode. Scan rate
=0.1 mV s-1. X2-axis is calculated from data shown in Figure 5-1.









Pt foil current collector

For comparison, we also used our standard Pt foil as a substrate. The results of this

experiment are shown in Figure 5-4. The current response during the voltammetric

sweep is much greater for approximately the same sample mass. This represents the

inertness during the synthetic process seen with Pt as opposed to ITO-glass. However,

the large peak separation still exists. This suggests that this peak-separation is not

dominated by interactions between the current collector and the electrolyte system. It is

possible that a Mg-insulating passivation layer forms on the surface of the electrode.

Other groups have also proposed this theory90. Our electrodes created by template-

synthesis may, in fact, be particularly susceptible to such limiting factors, due to the high

surface area. This theory would explain the resistive nature of the current-response.



100-



50-


-50 -



-100 -


0.5 1.0 1.5 2.0 2.5

Potential / V vs. Mg/Mg2+

Figure 5-4: Cyclic voltammogram of Pt/MgxV205 nanowire electrode.
mV s-1. The potential reference is Mg/Mg2+


I
3.0



Scan rate = 0.1









Rate Capability

We also characterized this ITO-glass/V205 electrode using the galvanostatic

method to charge and discharge the electrode. The potential limits used were 2.7 and 0.8

versus the Mg/Mg2+ reference.

While the ITO-glass/V205 electrode has advantages during the electrochemical

characterization, this system has a distinct problem in template-synthesis. As stated

earlier, the standard method of Li-ion battery construction processes the active material

of the electrode separately from the current collector. However, in the template-based

method the electrode material is processed from "cradle to grave" with the current

collector. The chemical and thermal inertness of platinum makes it a good choice.

During the final heating stage of the synthesis, the ITO-layer degrades. This is easily

seen by noting that the mass of an ITO-glass is significantly less after a similar heating

process than before. Therefore, in Figure 5-5 can analyze the rate capability data using

the electroactive mass, as we did with the LiFePO4.

We have taken data comparing rate-capabilities of nanostructured (50-nm template)

microstructured (0.8 micron) and film electrodes. These data are presented in Figure 5-6.

Each of these was analyzed for mass post-electrochemical characterization by dissolving

the V205 electrode material in 2 M H2SO4 and analyzing for the V-ion by ICP-AES at

310 nm. Again, we observe an advantage in rate capabilities of the nanostructured

electrode compared to the controls. However, there is markedly less specific capacity,

when using the mass determined by ICP-AES.










300-


250-


200-


150-


100-


50-


12
C-Rate / h-1


16 20


Figure 5-5: Comparison of rate capabilities of Li xV205 and MgxV20s. The mass for
these calculations is the electroactive mass determined by integration of the
respective cyclic voltammogram.


-a- Nanostructured electrode
-o- Microstructured electrode
--- Film electrode


0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

C-Rate / h-1

Figure 5-6: Comparison of rate capabilities of nano- and micro-structured MgxV205.
The mass for these calculations is the mass determined by ICP-AES analysis
of the V-ion from post-electrochemical dissolution.


O
-o- Mg system
-- Li system


Rate capabilities of Mg and Li based systems
Mass determined by integration of CV
Electroactive mass

0


^ ~ ~ ~ ~ ~ ~ ~O






84


Investigations of a Possible Mg-Sn Alloy

There is evidence in the literature that Mg and Sn form an alloy.95 Previously, we

and others have demonstrated the superior charge-storage properties associated with a

reversible Sn-Li alloying process.20 We propose that if the Mg-Sn could form a

reversible alloy by electrochemistry, then it may be an alternative anode for a Mg-based

system. This technology is based on two interesting reactions. First, a sol-gel route is

used to synthesize SnO2 nanowires. These wires are electrochemically converted to Sn,

which reversibly alloys with Li, thus storing charge. Figure 5-7 is the typical

electrochemical response in the Li-system, which serves as a control for the synthetic

process. The first wave is the electrochemical conversion to Sn; whereas, the reversible

waves are attributed to the Sn-Li alloying process.


1.0-

0.5

0.0
E
4-I
-0.5

5 -1.0

-1.5 /

-2.0

-2.5
I I I I
0.0 1.0 2.0 3.0
Potential/V vs. Li/Li

Figure 5-7: Cyclic voltammogram for Sn-based electrode in Li-system prepared using a
template with 50 nm-diameter pores. Scan rate = 0.1 mV s-1






85


However, when the SnO2 electrode is inserted into the Mg-based system, there is

only a single reduction wave. We can attribute this to the conversion of SnO2 to Sn,

analogous to what was seen in the Li-system. As the potential is swept more negative,

we fail to see the reversible waves of charge-storage. Therefore, we can conclude that

where Sn02 may successfully become Sn, Sn has trouble forming MgxSn by this

electrochemical means. The apparent inability of this reaction eliminates it from

consideration as an alternative anode system.


-1.0 -0.5
-1.0 -0.5


I I I
0.0 0.5 1.0

Potential / V vs. Mg/Mg2+


Figure 5-8: Cyclic voltammogram for Sn-based electrode in Mg-system prepared using a
template with 50 nm-diameter pores. Scan rate = 0.1 mV s-1


0-


-100-


-200-


-300-


-400-


-500 -
-1.5









Chapter Summary

We were able to form several conclusions from this preliminary series of

experiments. Our primary goal of demonstrating reversible intercalation of Mg-ions into

V205 nanowires without carbon or binder was achieved. We also showed that at a low

discharge rate, more charge can be stored per gram in the Mg-based system than the Li-

based system. This is from the rate capability experiment ITO-glass, using the

electroactive mass obtained via voltammetric integration. Another observation is that at

large discharge rates, the charge-per-gram stored by the Li-based system comes to exceed

that stored by the Mg-based system. This suggests a decrease in diffusivity of the Mg-ion

compared to the Li-ion. When diffusion profiles are given less time to propagate, this is

more of a concern.

ITO-glass was determined to be a satisfactory current collector because of its

available potential window, but it is incompatible with this synthetic process. Data

suggest that this results in only islands of material that is electronically addressable.

Also, a comparison between the ITO-glass and the Pt systems reveals some background

specific to the platinum. This may be a result of excess capacitance associated with the

exposed back of the metal. This would not be a factor with the ITO-glass, because the

exposed back of that substrate is not conductive. It is notable though, that this effect

appears to be unique to this system, as is not dominant in previous studies.















CHAPTER 6
CONCLUSIONS AND FUTURE DIRECTIONS

Conclusions

Li-ion batteries are the current power source for portable electronics. We are part

of an intense, global effort to improve upon this existing design, in order to expand the

applications of these batteries. We are specifically investigating the effects of

nanomaterials on the design of Li-ion batteries. This dissertation details our use of

nanostructured electrodes created by template-synthesis as tools in fundamental studies of

these batteries.

Chapter 1 is an introduction to the field of Li-ion battery research and

nanofabrication method of template-synthesis. This chapter lays a framework for the

dissertation and explains the significance of our research.

In Chapter 2, we demonstrate the advantages of template-synthesis as it pertains to

Li-ion batteries. The electrode precursor is deposited into the pores of a commercially

available polycarbonate template membrane. The template serves to restrict particle

growth as the precursor is hydrolysized to V 0 a known Li-battery cathode material.

These V O wires extend from the surface of a current collector like bristles of a brush.
2 5

The geometry of the pores of the template is imparted onto these wires a nanoporous

template yields nanostructured wires and microporous template yields microstructured

wires. We demonstrated the ability to create structures of reproducible diameters by this

template-synthesis method.