<%BANNER%>

Polarons and Impurities in Nickel Cobalt Oxide

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 E20110112_AAAADI INGEST_TIME 2011-01-13T04:24:52Z PACKAGE UFE0002242_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 51539 DFID F20110112_AABCQL ORIGIN DEPOSITOR PATH owings_r_Page_051.pro GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
0f8828a5ac471f4b322db504a3f04ce5
SHA-1
e38021c185fc69dcc8da6c2c2dacb1997e4f547e
37406 F20110112_AABCPW owings_r_Page_028.pro
0dad284ad93ef73fa285a55c6500d0ff
bd5dcd59878e961a44f3ba582b9c8c2d2c35e2bf
84937 F20110112_AABBNJ owings_r_Page_159.QC.jpg
4aac80e27d64d292c785fc6c1f5b7f84
6a0711039e5a0bef82ffb91e469f78e550e72cfc
238334 F20110112_AABBMU owings_r_Page_204.jpg
129c010491f43b6fec5fb3dee9396817
0d54523278b208e3a61b670cc42c4269e070cfd2
36621 F20110112_AABCRA owings_r_Page_072.pro
866556c44ad2068b4da47cffe189f5e3
2362c43c0dc2f803609f1a1f10d9f34df6a9e4d8
31425 F20110112_AABCQM owings_r_Page_052.pro
49879dfe3274f606e83cdf1d4768eca0
17471114f099343df7cacac3c5174701d27b5da4
46984 F20110112_AABCPX owings_r_Page_031.pro
f53efa2dd2da58f33744020fa553d2fa
dcb2c4a0be61e7b9240e2e8a97a50960bc470bb8
6520 F20110112_AABBNK owings_r_Page_001thm.jpg
51c1bd0db502ed4c6ea1594256a67d72
6b2473239a6b3c9702457f1e0d4a0af2f22f3656
1053954 F20110112_AABBMV owings_r_Page_096.tif
aa5c68762bc0d31e1700126f3bf6d2eb
aeb86882ba8cecdef0bc7de60edd385a0684ef0f
48066 F20110112_AABCRB owings_r_Page_073.pro
750a7e33aa9a3f08eac3eeb64b349812
161e6cf08c7e01e1f82c0e7c3ef07cdc1194ab22
38966 F20110112_AABCQN owings_r_Page_053.pro
017da79889fe7890d37d5e1e5e80dd5f
fc4d5a74a5c14a776158732fc85e077886d17081
51695 F20110112_AABCPY owings_r_Page_034.pro
ad9883e1eb4f46c5bd376ab5411e5397
31dcef52141237e3a05d44f2b8139fd42dbaea94
78118 F20110112_AABBNL owings_r_Page_166.QC.jpg
472965b9293e24160b89ddd1ccc338cd
df63acc591024659baba05fe55645eb8ce2adebe
99205 F20110112_AABBMW owings_r_Page_168.jp2
9a29a40962353684376e732a5893db93
4da9f9170600b0909211fb49e1b019b96078ce65
43411 F20110112_AABCRC owings_r_Page_074.pro
8eb3abbd6ba1859efe9f533460191bc6
ee214a5a12d0c29f592e4914743cccee039f826f
46056 F20110112_AABCQO owings_r_Page_054.pro
a8e97334421a911caa87eb083f1868e7
818b4c6046075f85e5b7cda9b2693598253f3eed
53926 F20110112_AABCPZ owings_r_Page_035.pro
eda869ccd40611d61c45b50d6d5b8cb8
1308f152263b9c71a4c4419682a6d15e7b375d3c
25271604 F20110112_AABBOA owings_r_Page_062.tif
5e13ff749ef467b42b29428f4aafd8df
f2dccd280cd1e8b2cf83b5a634b3d0bb5b096245
28521 F20110112_AABBNM owings_r_Page_196.pro
44f7e4e7dcf9dd9e389f0af883c2e52f
6701ae52b89701e209afc6bf4e52814f600826c8
F20110112_AABBMX owings_r_Page_206.tif
053e4b85d9a003b8d291996cbb89f651
219c524b38e1a9116a9c1e0ffc31b188c6acb510
41093 F20110112_AABCRD owings_r_Page_075.pro
0ba4a0d22d16b56102c19d35d8d1a158
97b79ff8aedb26dea1f3d5eee51687296bda0761
17420 F20110112_AABCQP owings_r_Page_056.pro
bc74053b1ff4064444c2e641f05d6e77
08231675f7617979bea4f0e2f0ed5aff1c84cbcd
F20110112_AABBOB owings_r_Page_179.tif
fa3ce474e003967ffa1776e9d3387cf8
e7979e1a149d1db32c448dc723d7ae8d5299066b
195326 F20110112_AABBNN owings_r_Page_167.jpg
8a3d0ec6f9e584f835918f1944f0f374
aec7bf2a6bcfe40f62819293a447afca6c1e85ac
108846 F20110112_AABBMY owings_r_Page_051.jp2
25130ee60e3772c6b84c8e0d4d6bbd9d
8c24df459bfede41482147d39c497ed68cce09b8
41271 F20110112_AABCRE owings_r_Page_076.pro
373a643c38ba355e3101f8191bee8c7f
5b3d571ff5ae8d1f3617c98ce382e3af3be1cf83
51713 F20110112_AABCQQ owings_r_Page_057.pro
0e7b99a23c23c3615b5bc1b5feec3f11
3a8e775b3bb625f6df492b01b2cc7e0d6771da65
1785 F20110112_AABBOC owings_r_Page_198.txt
d3b5a2d84296cc8b91343fb1301823a3
8904b9293eee21d33b66080af7ea75426afe6e84
49765 F20110112_AABBNO owings_r_Page_088thm.jpg
d81f25da769ab7d93acdcab0f0c5ab99
71e517177c419423ec31ca46e5ef03731ebc7118
111919 F20110112_AABBMZ owings_r_Page_124.jpg
866234f70cca6b6580d78feee0dc23b6
e7a471bb8b9e5a347d8c54fac8d063312fa0d5fc
44363 F20110112_AABCRF owings_r_Page_077.pro
dd6a5e2ada30a286981fcce7fa92d379
ec42e9a9606ec1c47759c3024b24cee5365e513a
42146 F20110112_AABCQR owings_r_Page_060.pro
c5b23b28e3bc749d800c7ee5504d570c
bc9d20442f312ff47dbaecd999fa9539d8c1ee29
F20110112_AABBOD owings_r_Page_080.tif
a3d6af3f8f492704bd5d1ec703177ba3
44c0e380cfb61975e7fee6ab96c3bbb18427a1bb
37805 F20110112_AABBNP owings_r_Page_198.pro
4c86d8a52899f372f880d81b751d6275
a512e4868bdcc2c810d7fdcdbe29aec0d3dc3aaf
35004 F20110112_AABCRG owings_r_Page_078.pro
a003e15c48c55b39bfe4b810b359ae40
7d5e95e9b8cbff631a6b92653c4343c99a1cdb89
54114 F20110112_AABCQS owings_r_Page_061.pro
ee37eb5e600fb3b1d41d91e60f28db35
43e4742bb1a9b56e68d3e1dce6a33e029229dfa9
146769 F20110112_AABBOE owings_r_Page_179.jpg
9d3cf3755806295d35eb9f346ce194a6
b79665c56a9b2254ec9a9899da7e27d1c109cc58
F20110112_AABBNQ owings_r_Page_043.tif
2b7f910ad69edfa1702d5b4f06024acc
4c0984316e1cadfdb7cbd41b3429781451109d30
47824 F20110112_AABCRH owings_r_Page_082.pro
30ed692cb6cf41bba34d4f978ae5fcab
89fab608c32fc671815425af6dd0bacc9246bce1
35887 F20110112_AABCQT owings_r_Page_062.pro
4bf1ca97c6d0c21bb1a22548573f0639
fec474b32bbd5e69c5907822a1981d872189d0b0
F20110112_AABBOF owings_r_Page_099.tif
7da6874b4cb0b4f13dd4b488be230515
8d4b2c84aa18f1aaddecaabc38cdd0d6ce831577
1896 F20110112_AABBNR owings_r_Page_105.txt
e3017b0b63745583f4360dd72216c1e9
4595b6d3654101ce04f22b4805599af16b5a8283
33700 F20110112_AABCRI owings_r_Page_083.pro
abb13d010269b7aa1bb17da0aebd32a0
34e25876231e879d14719b73addd00cd791ae632
46620 F20110112_AABCQU owings_r_Page_063.pro
5e0cb90810d4a2dab990ffec45d4c0eb
d7a52f71112c6b2bd1462266c952ba07121d47fa
2190 F20110112_AABBOG owings_r_Page_084.txt
9115b13b512031749c5352cca5b02957
7a6838b06c5e6e59c0da6015578cc474829c78f3
90440 F20110112_AABBNS owings_r_Page_133.QC.jpg
b36b8b58c6e5b0a94a76a18a2fe98c01
282f95e2edf7a4aa7c98fd5bbcb41633f19ee9eb
47288 F20110112_AABCRJ owings_r_Page_084.pro
b06e3b535673a73d4f146a1e70a0114b
bce142870b92582c302870078b14a68891b0a5e0
43247 F20110112_AABCQV owings_r_Page_064.pro
b71f5683a6d3123f240cee9f408facc0
3198a0084da6e114511c3315753f642a0282271c
79798 F20110112_AABBOH owings_r_Page_020.QC.jpg
94c6b293ab2306ae1bcffda19c6eb315
eb5cc3f2a36036d35fe9309118482d4f08387358
55698 F20110112_AABBNT owings_r_Page_058.pro
8412893fa1e23b68d7147aa80e14741d
e6030aa7cc7ef3bb4b8aa76714714e2f9d49f548
39468 F20110112_AABCRK owings_r_Page_085.pro
a219375757f6b78ccc090498f3b7db25
0ac0caa80c0b94bd259bb3e7f48c4e3aeee52063
F20110112_AABBOI owings_r_Page_054.tif
c1669b571dde1a6d53c031ca63cc4292
8783d945720f2f006511829c14963ab3644bc669
50751 F20110112_AABCRL owings_r_Page_087.pro
7721c613eda81460c1afb85b191bc3bc
5b112ee9d2bda2d5812654e452d8b9a4ab223463
48182 F20110112_AABCQW owings_r_Page_067.pro
98d19b16c74e47f82197b4969c457be2
c9071a20f3e43089d3098dcaaac0f0fede4ae15b
190549 F20110112_AABBOJ owings_r_Page_133.jpg
1b8a22e7be1392450c68ccd71c6df04c
48ccfd1e23305b9031920d5edd683c0a2d8c800e
88412 F20110112_AABBNU owings_r_Page_125.jp2
de0e6a0a0327da3707be662b963ae960
260c070811c2794d4d7356f69c69b0f096a046c7
37082 F20110112_AABCSA owings_r_Page_110.pro
b0728a28161ebcf2f0f5e933e3bb564d
dca70895028c54f963a1a7e66cf057d5e04b02c4
51981 F20110112_AABCRM owings_r_Page_090.pro
f16cb7b3c94d9d6531055822bbb69313
a1ea66dbc854c3311bfc6a7f91c6e7699f57ec74
36640 F20110112_AABCQX owings_r_Page_068.pro
9ee58c0810db3986168d6cc07f674b0d
06d62eac0311faee874b5e26998cc1836f817293
214838 F20110112_AABBOK owings_r_Page_093.jpg
aeb0d491bc7a2b0d74493daaba301921
956999bd1a7ac5e9f041e97d8c4c03b271b981cb
F20110112_AABBNV owings_r_Page_075.tif
16e0534105894723c89bae292c443d5e
8100d1b0ec83f93bd630df4a65dcb54ba1fa3871
43061 F20110112_AABCSB owings_r_Page_111.pro
0fb727eb1cac37f0572de2f4f1db2baa
e278a387bc0ea404b47f05c3d8a545165e021973
40394 F20110112_AABCRN owings_r_Page_091.pro
5919c900d0abb36264c38fc96d794852
e434319d65bafc9ecedacb809099fba1b2c55a01
42306 F20110112_AABCQY owings_r_Page_069.pro
2579e896f7a83cc529eadba650212c95
c2b85be94fdc1e8847486f47b3be2fbafa71c91b
12024 F20110112_AABBOL owings_r_Page_209thm.jpg
42b20038f7cec196c9ddb14ca67378f3
6b2ffac9027842f2d858d2af97984f63a630426e
33371 F20110112_AABBNW owings_r_Page_080.pro
8b677a5a7994b10cb6aa1d43c04b3306
d28522aab6e61a8cc77094e4abe280918194bb4e
42321 F20110112_AABCSC owings_r_Page_112.pro
972b31ffc18858e5f5fa61c4478a4fb0
57cebbbc257f310132cb8c9df722869b1eb4ed53
44276 F20110112_AABCRO owings_r_Page_092.pro
8911f0624d9c3d79d0e65b1c516599b6
a1e52bb2ba6ab039e6c5787c9978b0b34d8ede33
42317 F20110112_AABCQZ owings_r_Page_070.pro
11801591144c60003d67ef9f509fb211
c81ae0282e08326cc330ab254d256b1ce29a2384
55616 F20110112_AABBOM owings_r_Page_011.QC.jpg
6491070d6c855abe4ae1b8f795d10941
1a2782fe3e8f0e0b90a82609cc9268811f444fd6
23875 F20110112_AABBNX owings_r_Page_089.pro
98b23746ea3d77200f13426c723a92a5
2d0b3ddb2155854740f9dc775480110b32c2ccbe
47351 F20110112_AABBPA owings_r_Page_052thm.jpg
3e9e05765a7fd653befd30dae8313837
b080dc45af1bd29d6366526a1a4ad432761a3887
29793 F20110112_AABCSD owings_r_Page_113.pro
8787566362fda228db3d98d2632d6dec
884dbcc39834ab849804531b58ce93cafc55e6ec
30606 F20110112_AABCRP owings_r_Page_093.pro
9a6f242730cf4faeef59cdb3b924ecb0
3ce923834f4e4578213f4d0a79019b0462617604
100133 F20110112_AABBON owings_r_Page_171.jp2
57896425d68c9447c41a33987ee4e007
4cc7164cf2778450e19230f2b2b0a6dc37e3ec3f
1051952 F20110112_AABBNY owings_r_Page_198.jp2
4d3e29d1d3350a9cfad1e43d1ce3bb1b
4d74da16eedca31a4a66d91aa584556e79850b2b
207200 F20110112_AABBPB owings_r_Page_098.jpg
a30a31ae24b7ce29934e1bf210eae95b
ba1d796ce89c6ba00029f09f5529f4520a771982
38804 F20110112_AABCSE owings_r_Page_114.pro
4edecee21eaa8944e4b2b83db52f9339
616f8120a036ed615dd5c27d5d74a272524246d8
52024 F20110112_AABCRQ owings_r_Page_094.pro
c7f06d0b729e0a53f1aac64ea1908190
d786ea805a52867b15ab312ab431cb648bcc95df
F20110112_AABBOO owings_r_Page_002.tif
8e2fa7900ccbd17ad19f32e09263d76d
1404c070cfc8aab1e9384a6a59d65e4d3adda17c
23692 F20110112_AABBNZ owings_r_Page_051thm.jpg
a355ca480765ecb02841f064ceea8a96
dc132b7bbacfba7282f4dabbaa5625d18b16fbb1
84698 F20110112_AABBPC owings_r_Page_203.QC.jpg
8d32f91b0d3261c833df62983523e1c9
c40c4e4d17906817f87c19080b273ffdb45375dc
45897 F20110112_AABCSF owings_r_Page_116.pro
4e37df74ec8d3571a67bf2027cd8aab5
37ae62ec856eb82238768a013e73520a8ec48079
50274 F20110112_AABCRR owings_r_Page_095.pro
82e8b5f8ab0834c932a9f9bfe1fb1321
779edf655bdfed3f58df1717728b5a8e3dfd783c
990114 F20110112_AABBOP owings_r_Page_066.jp2
329fcb9484af26cb60e6fe7b6665f955
088820cb8a4e7c11f058068886b23c401a43ce95
95682 F20110112_AABBPD owings_r_Page_085.QC.jpg
d195d02d588f73820f6ad41a200e3f72
c6ecbab8b1fae2b5550987a59a6aa30b4125dffa
19109 F20110112_AABCSG owings_r_Page_117.pro
df6d7f4b191623c968596bab64bc8ab0
3c1d32e146466143b4611c6b15f2a281538aa5b1
25568 F20110112_AABCRS owings_r_Page_096.pro
8cb961e58ddafedf3dad392f564bf30f
ea238175ea23bb3b47147c9a2c82a5b9865eb1a7
F20110112_AABBOQ owings_r_Page_108.tif
e3663ba80cadf291eacba8ab40a8d5ab
6c509c349a4ff61d40d2b655603a671b0824615c
246055 F20110112_AABBPE owings_r_Page_149.jpg
bb84026908246841e43200f978be0071
e93875f6c8b1b8e6590dabc06817647a80fe7f0f
51807 F20110112_AABCSH owings_r_Page_118.pro
431e921059693a9cd09c7dab43d6ea20
f136761ebbb02b26ba1519b0c85701ee37ad11a6
42208 F20110112_AABCRT owings_r_Page_097.pro
74f40d4a280e66957ce2eb3fb7ac429b
683720e5e85d2df0b97c168e2a368fe370f1149f
1615 F20110112_AABBOR owings_r_Page_141.txt
862a8e66c2908937b333b37c686d0222
bfdfd7188222ee5ea8d637cd8c55b9fb60d67835
1779 F20110112_AABBPF owings_r_Page_025.txt
93ebd31a5e0ac21f7a3c72c8fedbef48
f2a5fca7fd0bb00d91af4c03e74e6811a02e6ccc
48101 F20110112_AABCSI owings_r_Page_120.pro
6c91964ccf5477b9042bdd3c8949e825
acce2a18d9031c74efd52a5211fe678bd8b40a73
50845 F20110112_AABCRU owings_r_Page_098.pro
ffb3cea92741e27cf841885df41c7844
a892b22cc52c4d0739c379f519a57b55ecc29f9f
235419 F20110112_AABBOS owings_r_Page_086.jpg
d1a5329134c2e86a5d5f819215231c6f
9b2a0cb7405e31d9dccba693c90e49005d895a54
48096 F20110112_AABBPG owings_r_Page_185thm.jpg
f943905e29d108ac5b4ddb130cdd683c
6326a30a77742f8e10751496a16f60959735deb3
50153 F20110112_AABCSJ owings_r_Page_121.pro
4f2afb0741fa5d0b8452595a05eab23f
9dc466e2e82ff2cc9bd995e27ae8e48a3cafec41
44038 F20110112_AABCRV owings_r_Page_100.pro
b4402cf1b98ac6c0ee870da5c7533899
c4cb0f63c05e60de94258b9ced925333ff95e8fd
18311 F20110112_AABBOT owings_r_Page_191thm.jpg
4b4b17de7ceb7c8f1dd43f6e6d5626e6
76319a0633052a1a305b9cb3f5d52054f8e7edc4
231954 F20110112_AABBPH owings_r_Page_116.jpg
9501b08f59a6073164861ffefda046e6
1f565a5cc6d15d1f4677386538168697209ffcfe
53562 F20110112_AABCSK owings_r_Page_122.pro
fa718d296c90f0503c9e22609987940b
2d571c58be267d3650bf4e069f9fc4d44eb50a95
44840 F20110112_AABCRW owings_r_Page_102.pro
610667db0c1af05d5d3d029be7bc667a
59a0812c757475b724699dd5142b1674dbe6c8d2
63592 F20110112_AABBOU owings_r_Page_031.QC.jpg
e5791719f8aa7929f6378d0e13b82cd3
33bc68543d42998bd178611c0e286699d07f60c9
1843 F20110112_AABBPI owings_r_Page_139.txt
d59f29ba291024cd2d89e38dcfa1747e
04504ce44e13e2ff8ff236834632289eec01347b
26037 F20110112_AABCSL owings_r_Page_124.pro
19ead38a724001a75e3b886de20247d9
1743efec967ceb6879e4324fa201073aeb558e02
41755 F20110112_AABBPJ owings_r_Page_086.pro
5af71960ffe1fefc152e634a382064e9
e21bbec324d099f6fc967c645e4598a5cbdbf27d
45606 F20110112_AABCTA owings_r_Page_148.pro
135080373c572218ac33381a8aa885f6
565c695f4fee74af08d00a1a3212f1e3bf4fde48
39976 F20110112_AABCSM owings_r_Page_125.pro
744e0c8e89311154e076857ddfde9955
bdba203a31bf8b4b41ab77be49270bf1f28b8300
47706 F20110112_AABCRX owings_r_Page_103.pro
a1772be9351e6a63fa01bd751e42140b
d745769dca7fb1d0d937ce8fb1309780b42d5cba
2565 F20110112_AABBOV owings_r_Page_202.txt
46b2fe264fb164b92939055ead9ca091
f34ebe1ba77b1f5170aaca2819357221b967678b
1302 F20110112_AABBPK owings_r_Page_182.txt
402e61b09de426ceb77402580ca3f696
3b1be1f0cf6bd982676fec66e782c1b04dc4b58c
43345 F20110112_AABCTB owings_r_Page_149.pro
fe31bcb6305ca53ed9510cf042373d29
cc41fb2897fbd309a99d32f428fb3954e940f8c5
51579 F20110112_AABCSN owings_r_Page_126.pro
3f043f712e8bfaae04d3d834066cc136
96b7406f0a687eeb6358a51154f9e09133063f50
45260 F20110112_AABCRY owings_r_Page_105.pro
fdfa77c995ed2e0143ef5a7c5eaf7b5f
0b8c4ae6a5bb1d2211dc70a28abfc6f28d91909d
78995 F20110112_AABBOW owings_r_Page_108.QC.jpg
07e5f47da9b954a1137e2c5531ec5193
24506d61a4ca31425fffc41de15ab38a89aa4f47
23949 F20110112_AABBPL owings_r_Page_200thm.jpg
8501bd62dc124be0620f9743e6e88d9a
4e0dab95f2345d6318dec1eb86a4d82e1215fc21
44936 F20110112_AABCTC owings_r_Page_152.pro
9b1b8cc4b1492478013016814a691963
7bdf0039ed30f141939fab4c93fb52e586fa59d0
28718 F20110112_AABCSO owings_r_Page_127.pro
a669d603307f4e73371496a6db4e505e
26a98332f7a10247faa4d2a8a4451094b2380227
47197 F20110112_AABCRZ owings_r_Page_107.pro
42276ed7d224c9892bd9c824954107b4
af3efd801c15c7b4a768e5f1f6e142770b66aa13
47538 F20110112_AABBOX owings_r_Page_078thm.jpg
3f235e2ed7d4bde959be5396192a52e5
87b8cd71a45e33c8ebf1f0c96e2bc5d596c3ff6a
112245 F20110112_AABBQA owings_r_Page_012.QC.jpg
1861baf59fd3ec39b34fba928af28916
73bdc1873309b265ac88d7b024fbc48e0616bbf4
194104 F20110112_AABBPM owings_r_Page_004.jpg
c998e2a06783f95faf774cbfe209bc47
5759905973f9433a1e4410890d76204bcd20504b
47500 F20110112_AABCTD owings_r_Page_153.pro
c9e4375eacf97833128956d3a30a3ca0
7cf88aa6ef3b6acb1ebc3c638d860060cf7b7883
47761 F20110112_AABCSP owings_r_Page_128.pro
abab2b3ea5982bc986f2987ef8611750
58c617f89822aad5d75f1ad702861f60a988c248
1186 F20110112_AABBOY owings_r_Page_079.txt
6e9227bb28563691ddb1685e3c122ed5
85ab8cb26983b5051813fe07ee7aad0263920e18
88021 F20110112_AABBQB owings_r_Page_117.QC.jpg
e43e96da0fcd54a930549db166307327
24b52bd915b0b77e4d73c264aacbb196f85ca65b
49449 F20110112_AABBPN owings_r_Page_059.pro
fdcaec2db18edf1a88a276f28f1eb4a3
c758570e0e20358665662a217d6e37e239ec1d34
34687 F20110112_AABCTE owings_r_Page_157.pro
0eda21206e65acf69845a9bde2f72a12
46b1526b207cfe21c5acd2fcbcbd537cf4850f8d
53758 F20110112_AABCSQ owings_r_Page_129.pro
77847f2665664a3b1c20520e566619a6
68f6224caa5a4c1f631b5abd4948894b6b90cc59
78725 F20110112_AABBOZ owings_r_Page_174.QC.jpg
2e468a6bc816c0188aff26366253b613
58d7ecfaca05bfe88fe5ec6f35a6eb53717c4082
226090 F20110112_AABBQC owings_r_Page_058.jpg
298cc3dbc7430af91c70048dbc54ac1a
0d72671642b97043e452452e5b01d21f1ff63891
83003 F20110112_AABBPO owings_r_Page_163.QC.jpg
a7943f125f44b21ed4841d0546a527cb
57c893a7aa47e483a13041e41c969adf084fcca6
31164 F20110112_AABCTF owings_r_Page_158.pro
8b8618e6947adf50c5a8e64cf2ce645b
594118c0c048c2933dbff33199b0c1f2ba84b6f7
34157 F20110112_AABCSR owings_r_Page_133.pro
c6cfb76eba1ba25bba4f47fe34947415
8b09c4ef8dcd4497ea0fac02f44075dae5c8bf05
61482 F20110112_AABBQD owings_r_Page_208.pro
6d674c02bcd5e3e92014b62f99a3fb0a
26f0a34e69fb6c0f49c936cbed45316877d06bb8
97505 F20110112_AABBPP owings_r_Page_066.QC.jpg
63fb0b8314ef955cb9495384c1807791
f1e208ca7b02d464a42d8b6df5a8c509a1d5a370
30399 F20110112_AABCTG owings_r_Page_159.pro
60ea8d40e7df2e640f68edbf550a93e3
1580af248fba81d19f60ddccb367fb52eb500b75
25259 F20110112_AABCSS owings_r_Page_134.pro
8ec28b56044fd776616ded23eb9a1890
1b815181facaf68b52e68cfe48688e6f46cc817e
711323 F20110112_AABBQE owings_r_Page_011.jp2
fc2859e688427c8c18a3a7837be778c2
a92596ef9911f7360b0f7882bb7bb284c39e8b11
1743 F20110112_AABBPQ owings_r_Page_155.txt
5ac1f8ee03f8461cf78035c255388617
28d96132fbb7e026b165342ccc094291024ac91a
49741 F20110112_AABCTH owings_r_Page_161.pro
3aea0ee28003fb1b730408d896a93691
16db64803d56ce11e954b26cc01277d569d9d3b4
37824 F20110112_AABCST owings_r_Page_135.pro
f81a2165fa1eb3f356270306d7d681e6
d68f690df2901338ff2b1de92aea7437eacfe79f
79408 F20110112_AABBQF owings_r_Page_079.QC.jpg
75879f20d8b7c7880f7c46c5aece2168
bd83ddbf350f78e7a91ec5b4bf2444d198a63fae
2805 F20110112_AABBPR owings_r_Page_050.txt
82056e41c1c62099db723cdd299b70e1
8fb1abfed81743c4ee5a4e6c449cbd28d8a294f4
51137 F20110112_AABCTI owings_r_Page_162.pro
f456dc14ed5db73291b8b1d3720be5b2
4f88ade3fe799423223bd76067a6ef93b303b0b3
40895 F20110112_AABCSU owings_r_Page_139.pro
42456e2009eb716748d2f6a76beb7751
101f7c8508084aecc0f95c11391fe0dbce981f2b
136381 F20110112_AABBQG owings_r_Page_014.QC.jpg
70f906442643ff28bd259e85a96d144b
fd7f315d4594a65e358cf671d013b0753d08d731
1528 F20110112_AABBPS owings_r_Page_157.txt
9f53f20425282d88c4b8af9a00cebba6
6f06874730a6915aa96de04cdb8bccf8b0799ee4
53004 F20110112_AABCTJ owings_r_Page_163.pro
c3ff6d533703724f1ef8674fa84d6982
d50c9754fc216a762bf8aa9d45ce22b032046e1f
28214 F20110112_AABCSV owings_r_Page_140.pro
d627bbc162baa66143ae230bace9aafb
583389a8821b1ec8c5751431c819cfbe4ec03bc4
112064 F20110112_AABBQH owings_r_Page_036.jp2
9da7afa1ed3efbeee7b971f9e27411aa
05aadfe291e54bcbb807d7b11d82fc2d234cd6d1
3210 F20110112_AABBPT owings_r_Page_013.txt
11d62872e9b484c5efa3db23554f0916
64ee92bbe8abaeb55fc7c713d1207e737c88a148
24971 F20110112_AABCTK owings_r_Page_164.pro
49a66a8c54160482317ee1a1b25ab790
aa98e57ac2f5542f0d8c57986104f7d743e4ef81
52648 F20110112_AABCSW owings_r_Page_143.pro
82b26d7784a04841931a38bd39c26e23
5d08c8eb43f480868443332199e75a2791e6b25d
F20110112_AABBQI owings_r_Page_139.tif
deb46ff558f5d6d8e90f5555915f38c0
d941641aef256b65c10581881696bd37f6fac4cc
25911 F20110112_AABBPU owings_r_Page_055.pro
f56b0492939b423859a6daa2f37d97ad
017257bf7c4f9cc2cb6ee79f7c1efe90405e675e
44696 F20110112_AABCTL owings_r_Page_165.pro
3d0d7705b227965321dbbb208ca847fa
cd72832dd0d2a72c3d114bf8d868914dbbbd66b3
49766 F20110112_AABCSX owings_r_Page_144.pro
6bfc2a288279669e20be2be5a03c49df
2d13c0a6fbb841226a6dc8f2b92787e738f19d89
25698 F20110112_AABBQJ owings_r_Page_033thm.jpg
58bd670da42235c52921c0453df59596
9d7ecfde368cb8a89af68a83ad5dfad8695853af
47799 F20110112_AABBPV owings_r_Page_055thm.jpg
5add456a9a8e63afb697f251f0547fe4
760283a53295e62c2ee4f58198be205986e9e351
39350 F20110112_AABCUA owings_r_Page_188.pro
625b3fd2a032260e904aa465987e64e4
68131cea27fbc407e504f33f416a965512859a99
45253 F20110112_AABCTM owings_r_Page_168.pro
b3482d8469a67d72dd427844ab61a792
0d8b10bd9ef32dea139a1769e8f0f312c075d4c6
1620 F20110112_AABBQK owings_r_Page_040.txt
c37f6da4cce84d38bf3a193f92307bfc
6611e39631a574a6f3c3ae388ba7bc97746fe3b4
43466 F20110112_AABCUB owings_r_Page_189.pro
6e47d07d525838ae53400a954999200e
1af4292e10b84e853ac1edf5a1df5c409ff7e56f
47262 F20110112_AABCTN owings_r_Page_170.pro
bc4b61cd19f693fd9a5009f13b106fdc
4b6db2b30152a29a969293905f72fa70340859ee
49909 F20110112_AABCSY owings_r_Page_145.pro
f041d337c66af1ef757cf8a5a934ec11
9865c6e39b2fb2459b6c22fbb3a0d80184a8e5fc
1051911 F20110112_AABBQL owings_r_Page_136.jp2
ad1736c92c39ef24d6b64c884af48dde
f8ac7112ff41211088a014d137bba8e04796c744
1253 F20110112_AABBPW owings_r_Page_002.pro
33c1ea7239d154baf88ad2baad98b1e1
da93a2a206c55891dd3589e787cd886c53fc3b8f
31646 F20110112_AABCUC owings_r_Page_190.pro
0e1162c90d71d8adf8c0dc7c54763381
28973f6894594cbfbca03270bb18a3fc091d423a
45782 F20110112_AABCTO owings_r_Page_171.pro
12d3764617c584dab6eddcbb725f8713
24d8f57dc2f161c9a1f8bc6d3227fe229a9a2a2a
48670 F20110112_AABCSZ owings_r_Page_146.pro
750d24406ead1785998cce0a973bc0ed
d1aca67bbad865f3edf98217f3d43da80137ef71
F20110112_AABBRA owings_r_Page_184.tif
7d2243039e4c8c615e5860fa5a58cff5
57001967943ab539fe0e5826685f1ebeefeb7999
25874 F20110112_AABBQM owings_r_Page_143thm.jpg
6566f4aff94b97ae1deec149b9426fb1
67750296f9161ee9f6f45dfc2820f974ef5adaec
1950 F20110112_AABBPX owings_r_Page_099.txt
a59de87a30fa80c57b2440a2a92516c1
0f6a6c0ae0383504ba3e4149b725ac5519610b5d
39901 F20110112_AABCUD owings_r_Page_191.pro
4a31998cf49bbaccf235b3a95095f051
b0ab32aafe9587ba72e987707dc2de572ee701b0
39984 F20110112_AABCTP owings_r_Page_172.pro
8a60c262147c51c2990d16d9047820c4
051dd8394bf7572f069855a2357669f0f3976476
106284 F20110112_AABBRB owings_r_Page_050.jp2
042c814b8e71dc628d10edca64cab148
1b8c561a025a2436388f73281d6e439b71ef2dbf
F20110112_AABBQN owings_r_Page_109.tif
1ead7460524143ddaa5dfa9403b44385
433943139dda28ccb062fe99b315048d1e9a1665
201257 F20110112_AABBPY owings_r_Page_049.jpg
fb1381e11aab62e57653eafddd589e28
e1a37afc27ba08250156e20e96ddb81d9fa300b7
20433 F20110112_AABCUE owings_r_Page_192.pro
2fbf45e706537951b5af91c62b98421d
618cfc7485ee9f608421b7776f01f3b6be6966fd
42408 F20110112_AABCTQ owings_r_Page_173.pro
a1a31198606c0a932c197cd480b2f96e
37170b88847d300ce0523db4e69b7fa6bb560653
207536 F20110112_AABBRC owings_r_Page_026.jpg
ece416bbed181ae4a5538995b5e92dd4
b94a1b9be9b04605cd5397550febc2c81943cdd5
F20110112_AABBQO owings_r_Page_202.tif
17cee1b9b7ab2fe3e1d2770a10ad9526
6ef0eb9b3cda193f6f68d8ad32f7620223bdeccc
185467 F20110112_AABBPZ owings_r_Page_025.jpg
c2c13b1ca2fd95a6606d40c933c99fa6
00da7b0dde11c16083dfd91fa8654a180870dd9d
49143 F20110112_AABDAA owings_r_Page_107thm.jpg
76fd4a70a6782b963d253e3f3287078c
4a825050e190ba509592b8994dd0f04133893759
36704 F20110112_AABCUF owings_r_Page_197.pro
63efdc16618b809f82ef84e2906f1027
c491ca656eace82786909b04e3142e0a536304f4
33993 F20110112_AABCTR owings_r_Page_175.pro
e8b7cd35d64dd3873952092ee9a43eec
885fda2cf3c39ab4ce6cf3e353334b325a48e5eb
F20110112_AABBRD owings_r_Page_187.tif
6c97c7b0072e4a44e2f741b684fc99a2
b27faa2bead461c3303250796e69b75db875ee77
2138 F20110112_AABBQP owings_r_Page_061.txt
f7dd2735e04afb1b0eb59f5f162a7d29
50c7009e9ad71796c49a2969d8bd8b29d6a3b4cf
74115 F20110112_AABDAB owings_r_Page_051.QC.jpg
fc4b49629e42d0afc63e8a576ec24d99
a55992f9d955db199c9358ccade578282b5e27d0
48570 F20110112_AABCUG owings_r_Page_199.pro
416c634ad030ff96a48cb942962bc64a
ea69e36fa92cc02b53f818c18b55d0fc7e70f819
27232 F20110112_AABCTS owings_r_Page_176.pro
8f483ae3874b680c5b87c9a8242ccf5a
f1f33e76420c2607df2364cdb4ff8ef60152c352
225090 F20110112_AABBRE owings_r_Page_074.jpg
39f2c336326cca1402b9813b9e2e3b6d
dcd774effbf2d04793dee4b880e24af0d079607a
1993 F20110112_AABBQQ owings_r_Page_087.txt
6ce0c6fecea6e2ca487525ecda94d279
d1aeaa90a419e1ea1e29d424a42bdad852909d2d
47660 F20110112_AABDAC owings_r_Page_085thm.jpg
255d3f9ff016231b1c5a80fe3fec43c8
39f80616480fb91b2b0fbb6ac61117abf0c48f8a
58770 F20110112_AABCUH owings_r_Page_200.pro
e7b2009ad7a311b329e8994293008401
beabe967db76ef0b1a9aa6aa337fa4f4dae48c41
26738 F20110112_AABCTT owings_r_Page_180.pro
567629c322f699773d3e86db3c57d68d
5466dceb3ffc90ade601e3566b55f6303f582c7b
10580 F20110112_AABBRF owings_r_Page_024thm.jpg
5dcde4468f142883b71f08a2ac5a0835
4977441831d4357ef8ad50c2f4747e0bd5314f1b
38886 F20110112_AABBQR owings_r_Page_193.pro
c9b121d2d85d97f0368bb587fe73d0d1
c88260178fe8903e044897fc88c8c516aa61d7a0
23467 F20110112_AABDAD owings_r_Page_082thm.jpg
e6a91c7061b5a7e44f5d04e106a0adb0
26a6e6e99b9ddef2e2adf899a94a4a3d72998b53
61242 F20110112_AABCUI owings_r_Page_201.pro
c43a95aab2fd4d0d9bda9edbc65199c4
c247ac9c96289eb3ec9b05706b614a0034f81e85
20115 F20110112_AABCTU owings_r_Page_181.pro
b01454a3140b6fd2a8adceb6ebe9273c
daa61286b56f2c8ddde587a97e1a3cb88271bb38
39812 F20110112_AABBRG owings_r_Page_115.pro
2b365d3cec3101bcd8c62efca670f3a5
0a7dbba0efbe5298053c3b4178b8ace1ff859c66
F20110112_AABBQS owings_r_Page_045.tif
a5244dd0a54ae82206f409db6cbcc8bc
afbcbd63f841448e2f37c36a06eedd7f713aceb5
80583 F20110112_AABDAE owings_r_Page_177.QC.jpg
b91d307d7f17f17c27f4fdd984a7a151
0064134d84614c63e9d740a3ad7741358af69958
61525 F20110112_AABCUJ owings_r_Page_203.pro
76c26e84f19db20373db228a5b0637aa
ac1ffe8c43c4f911fe66dc157600219396af34d2
21378 F20110112_AABCTV owings_r_Page_183.pro
74d160a10f56709623811ccc18fae5c4
d6265c330f80217a7e63d5309c40e2e453670085
2006 F20110112_AABBRH owings_r_Page_119.txt
73aafd2c080540d125be2fb5e80344b0
ff5757e4d60f259b24dd89bc98169974f7bcf227
98719 F20110112_AABBQT owings_r_Page_063.jp2
1c64cde5e8d8a516a33c977bde1a5ef2
06e15306e676dea31b94d21c6dbe7938512a326b
62799 F20110112_AABCUK owings_r_Page_205.pro
70a297c096b00bacc17a5d60483aaa04
c37d1f2e2c3e094e04b489ab27cbccc5ebd5e4f7
32664 F20110112_AABCTW owings_r_Page_184.pro
88dbed1c1d300cf1573bd120ae42f0a3
ce25859e22279316d9abac77f97a05f3c028f381
31081 F20110112_AABBRI owings_r_Page_151.pro
148f677158dbcc4b65ce5c6246aa01ac
0fba004a8dcfd14e51bd9a5883b744a626d5993a
25322 F20110112_AABBQU owings_r_Page_090thm.jpg
6da832d938426537274fb4a787814fe1
91ad2def8d4568543d6b84b7dec3c31882d0e950
46569 F20110112_AABDAF owings_r_Page_111thm.jpg
4dfb4ad9a7a4fb9875fbab2fa085780d
5d4e9c939fde2770a2dd87afe120284d57e54466
63159 F20110112_AABCUL owings_r_Page_206.pro
e5abe10a1694bac6082d1545e88c703a
268b8441d0db29516fb6845b8c8a067036e1a504
12720 F20110112_AABCTX owings_r_Page_185.pro
0e4848ef3b01e58b27cb4d71fc4c87c4
eac2acf336e3d4f5e027c780ecddef1d03f95ecb
554860 F20110112_AABBRJ owings_r_Page_178.jp2
75098f9d68f562f2c5aa90ce436a8d1b
2b612e1c6b972bd6b4cf58165f6e1230fdcd5e63
135046 F20110112_AABBQV owings_r_Page_205.jp2
5789858e89cd8ad3b2264a686537139f
b97672cd9b417530fba959a23cc7a7cf3e0adc3b
24826 F20110112_AABDAG owings_r_Page_118thm.jpg
1bafacf8ce4cd095277be40829bcee1d
1e6d04eab5e736528823f0ee8a1214bcb10d131c
2004 F20110112_AABCVA owings_r_Page_022.txt
5ea3bbad0daa3bf8823b4bd65675a77c
ada65d3c3b5bc878aa4b1e9e123ed8ad160c7493
58102 F20110112_AABCUM owings_r_Page_207.pro
8beee7c3931790edc05bbd528d12bfbb
3b7b40830badc6825160d4cd1b2948963f182cfc
54021 F20110112_AABCTY owings_r_Page_186.pro
12696de27b741aa39c1a8ad86aed46ed
6a22055c6b2c8793ead7a1eddbfb583f796d9899
22820 F20110112_AABBRK owings_r_Page_100thm.jpg
7fb113d0305647d0608964831e91f1c8
19762bcdb83e2f26bc3379c5a3face995af4ebdb
204238 F20110112_AABBQW owings_r_Page_131.jpg
453593b3aee36f2b4c6c3d6c3d23dcbd
77467d7f9bd748a7402c301b6a1507c4b31cb6e4
46548 F20110112_AABDAH owings_r_Page_184thm.jpg
084a4ae1cf32c46d7ff723248f8a3211
70550248f5fbd84f8d62b23d7a6f7571c11c62c6
731 F20110112_AABCVB owings_r_Page_024.txt
30549fdec9fca8d84ff5beeac30d1f13
f16c4aa13fc77db6b42d136fa65c53200a477390
111 F20110112_AABCUN owings_r_Page_003.txt
3bba4be8ba43ab73392160ba57475235
06cd4152a077ee0df4f4b5025c156d7597608298
F20110112_AABBRL owings_r_Page_193.tif
a5e692947706940401a9c3acebb35b75
6346b1ed71fd9e8d87457924968214fb7a860fb0
47217 F20110112_AABDAI owings_r_Page_137thm.jpg
edc38a65a53d24ecee2709cd9f56a1b1
c01b0658eac475bfd25d660ccc9bdb00ad350362
1925 F20110112_AABCVC owings_r_Page_026.txt
d530b8eed3f8bb73e1aecee5fa74aba0
4b6a04a6f12a7fdcad000ab0f924dfc2f6bbbb10
1825 F20110112_AABCUO owings_r_Page_004.txt
d5d032d8b9cbd195c516b774a20df77a
ce5ab4f954bb233884e675b5a033b4dba1a0f3a7
36195 F20110112_AABCTZ owings_r_Page_187.pro
5fe1aa6b16da3725b9b91c0b753470af
1dfd9403707c0b4ffe5f2b715fd5093f74830f68
F20110112_AABBRM owings_r_Page_128.txt
41d6da3462cebbf1b751abf1554fcbed
bd215dba18d27d05afbae72487493e65405dec8e
F20110112_AABBQX owings_r_Page_116.tif
9edb9833e8895e1160153fd2dac193ac
832dd1fa7a1a6be276a020fdd23819e344236b7a
192005 F20110112_AABBSA owings_r_Page_171.jpg
d8a0f7bbcdc755b07e26ea1d2510ba63
197fa503bcf29f3edc4365e65dffc7f5ab933a25
128468 F20110112_AABDAJ owings_r_Page_017.QC.jpg
5f04692420ab05b6ec2cf143d4ecf1d1
fcb5a6848ffd35d8629d23784ae27cf4d5d4acfe
1577 F20110112_AABCVD owings_r_Page_028.txt
642022b097f1c7499cdd457433983118
4028bdc79979c590e6ea1d02af9ed7f1f5c38963
1944 F20110112_AABCUP owings_r_Page_005.txt
b81039ada9633d5e3d6bbe79cee2b5ef
9600d0bacb35b0a99adc8eb594bb98553f46909b
70448 F20110112_AABBRN owings_r_Page_132.QC.jpg
db712570143bcc318984246dfafeecd8
0a8c5ceda4931cf35a7ac659fa2589ecbbbd9829
24685 F20110112_AABBQY owings_r_Page_142thm.jpg
858ffc9a4648b5e898f52bbebaa4673f
a61250b8048729221ae521a10d02fcf3ccc66838
46324 F20110112_AABBSB owings_r_Page_053thm.jpg
e281f3173b21577384b6db62023491c8
069ca099e6cd956fd81cf996c6dde1e19f30bab1
80581 F20110112_AABDAK owings_r_Page_202.QC.jpg
54083a5f7af7282e5e31e8f36a19b0d5
5329852a646111751c5c54c4ca719166080a0d1e
2324 F20110112_AABCVE owings_r_Page_029.txt
37b8b9aa433b921d0f21072a8f10f630
4c45e55ecedd0137a478beb07e9f34848f0afaa3
3218 F20110112_AABCUQ owings_r_Page_006.txt
7ed6e3e27a65d98ebf0be242a290f87c
141182bc2c1d43dd9b25ac69a760496c1cc76c73
50431 F20110112_AABBRO owings_r_Page_108.pro
ac458453538d157ed264423d8cc0cdca
5bf54460de77aa15a60e72ec5a5dba3e789f7618
F20110112_AABBQZ owings_r_Page_160.tif
0eed6296f03bfeec3334a4541495253e
f7eb84e5b59c3728ec2d93acb015fd86c6207af8
41579 F20110112_AABBSC owings_r_Page_106.pro
fbed561a26ffba391851dbc4eff1543e
86d2f8d4fb801a5a657dcd2622ef1a008a9f6113
24387 F20110112_AABDBA owings_r_Page_203thm.jpg
e887ea202ac7e6aa77268ab20f97a754
46b37e7ca61aa645a8441e78de5e7e9df304c343
44313 F20110112_AABDAL owings_r_Page_157thm.jpg
7ee7fb5fb923b7833eab51806369fa80
1ebf0db2f28b38d92237b46dcf47947f9e9838fb
2797 F20110112_AABCVF owings_r_Page_030.txt
58aa40f60d00db037bc4d7f0b9e17937
84c567b6e4549a682386cc55f655137fdc38d9af
4329 F20110112_AABCUR owings_r_Page_007.txt
b63912d9e4aa933a88ec5ef49ae78b03
559e2632020eadc0e513734c13029543c4316576
F20110112_AABBRP owings_r_Page_015.tif
e71cf4ef6aa883a011c2686eb7c3589d
9bbb9a3b5358a1a51116f96020a02917633564fb
1051983 F20110112_AABBSD owings_r_Page_149.jp2
7e2ec99a944da85ae6c1cbd84a56ceba
3baf8c2e745c540064f098359c904ba863b0fc06
103956 F20110112_AABDBB owings_r_Page_028.QC.jpg
08c8a0e3f4936d5dee27fdebf4a058b3
9bf231691fd636e2d2509048d66e18ada1c9b507
82991 F20110112_AABDAM owings_r_Page_196.QC.jpg
d0014fd83e82c6307a8ae0592f5caee0
564c2705bdb57c810748e4f6ec25552ba147a377
2045 F20110112_AABCVG owings_r_Page_031.txt
3efffa924eb69055eddc94049317d8e2
ddbec30d91b75d2fd4fb0065d412af078ed1e2cc
541 F20110112_AABCUS owings_r_Page_009.txt
027783c44ab408f06ce26673a1157b33
116686967bb619a4d60fe6646097b5503f308303
1118 F20110112_AABBRQ owings_r_Page_209.txt
9b9d5395da9bb4345dc868b09eac9ff9
6f1228d53828f5cd8285a88b7fef96c10afdc83d
1106 F20110112_AABBSE owings_r_Page_180.txt
f9e66e07f03e68cd4b3f3a768b3efe38
13fb87bbe4ed1b6839ccc4363e59631a758a4301
104225 F20110112_AABDBC owings_r_Page_092.QC.jpg
b188e041f3274249cc712fc78177b206
4c698aef691d6b007a5f36accd88ae37c9c30c2e
71543 F20110112_AABDAN owings_r_Page_152.QC.jpg
3600e78b09460ccc7a62f38edd12e5c0
fddd5f130d3e396d165d9bfaac1c0af1f4f44aec
2722 F20110112_AABCVH owings_r_Page_036.txt
c2a645420616c8faa7803f4861b40884
56c00df0472d2a9fbc16cb3f2a3a073c74c81293
709 F20110112_AABCUT owings_r_Page_011.txt
f65c4bccf1766affe034d48c1c512bac
6d409312df49cf6dc1ff69a16b1ef445c77aa482
103099 F20110112_AABBSF owings_r_Page_073.jp2
9b86c2c87b23aaa00d97cbfc80eacdd2
073e18dc9bb56c36bc7bc06876c35aa035a97f55
20951 F20110112_AABBRR owings_r_Page_132.pro
c75a7c092c0d796215e42b6413ac9c35
c074ee7570e9f8d1fbf06959afb29a4b424243dd
55951 F20110112_AABDBD owings_r_Page_014thm.jpg
b0373c999e8751fa3539ff8ca49eca2f
aad8395da537b596dc7649bce072fec7e227791d
25741 F20110112_AABDAO owings_r_Page_122thm.jpg
b76db78176f801ed536e0d28fbd8a46f
517599133ce8cb66ba8089dc3871db509c47995c
F20110112_AABCVI owings_r_Page_037.txt
6bd320b561a8695994f93570aa06bb7b
4da346e99a05be1fd27da5b44c8dff262b17fc3f
2486 F20110112_AABCUU owings_r_Page_012.txt
07027918f6e2273f9770ad7675f9cef2
2a9ff3cc841c217491f84e5404bf9330656c378e
2013 F20110112_AABBSG owings_r_Page_098.txt
cd372ff1ab17ab783d500fb0093fbf5b
e9347ae84cf4fb09f6661f53cbc95d055c9667d8
865726 F20110112_AABBRS owings_r_Page_020.jp2
c43b107ec1a25b767d21527a21d9e9f0
562f2640fa34845e076097cbe50eb5647f288896
96762 F20110112_AABDBE owings_r_Page_041.QC.jpg
1ba93b0144c1d464b2d7e7a72a653868
eee90fde966d7deb569c70ffd8a034770b895f58
97101 F20110112_AABDAP owings_r_Page_136.QC.jpg
36168c634eb76fee7a01bb2f5795af43
c0f3ad8c3d2bafb20e0ffc272b330b592470e2e3
2088 F20110112_AABCVJ owings_r_Page_038.txt
fb24d584cf49a8c30c3e30a83bef7ab5
daf782aa9ba80f738396a348c66161a62634ec8c
3229 F20110112_AABCUV owings_r_Page_015.txt
75da00fd86dee6d81b7703bf294ee3a2
88c14a12e5472ef63fcd2748017910ec246d5864
1768 F20110112_AABBSH owings_r_Page_086.txt
e48e91dcff3b170fb7d3c149e6cf4240
9066b8f850958de27626574aa5c2088604f4507f
975 F20110112_AABBRT owings_r_Page_183.txt
78d8d4daa3fedd95e80c59858fcef285
ef2e8cbfd60226505b45d0d0414ea29db82acc0e
95869 F20110112_AABDBF owings_r_Page_070.QC.jpg
9a3f4331c1055fd3ff6d17fef68bbafd
e4593f40e70f390ae1418c1142d14cc4d1b9e834
24768 F20110112_AABDAQ owings_r_Page_098thm.jpg
80835565dbbc2486e46061fc20697450
7a9851dc0cdd8b152267a05bf61aaead2df11882
1942 F20110112_AABCVK owings_r_Page_039.txt
8f305e7c8b6cfd8fa441ad39ec72213b
a8ea2b0e3360a073dd0b15cd8a2b15f87777206d
2973 F20110112_AABCUW owings_r_Page_016.txt
b988af9653931d9cdd2e00c6e18bf5bf
e655f567b9163c5c0151e25f7bf202a32f854d5a
97049 F20110112_AABBSI owings_r_Page_152.jp2
cf40f8f4c853b44ab4519626b66ea7c0
08c74bc832d20c601e32c258dd710bd1789dc4cd
20280 F20110112_AABBRU owings_r_Page_060thm.jpg
6895599cdaf57558e18e94dc41f08ad0
e9353ac66ab1dde9be2b92d8b27a86b22f1ed7c9
25658 F20110112_AABDAR owings_r_Page_163thm.jpg
ad949bec34b5c0764c202f8287a49a4a
1d0fdaef073145af41979ef421d90c36357bccfb
2048 F20110112_AABCVL owings_r_Page_042.txt
eeae7c0f019a17ff055cb9cce2cd2af5
d10ae02cd741b0883c2d9009d38b0ad1828391c5
2800 F20110112_AABCUX owings_r_Page_017.txt
35d58e102e1f08aa20b3f2d9298d4ba3
24b6c39e4846c1df9b3a5e80807e249315dc63ea
48401 F20110112_AABBSJ owings_r_Page_049.pro
6f5015bd1d26c7deeb151261d2e2eaa3
ef0169a3b3cab715e0085b2fbbf6af4f30b4973a
80630 F20110112_AABBRV owings_r_Page_187.jp2
5cb446324fec8127cc481b0b4719278d
8fcd4e2408307cf1b5e841905243aa7ef8302986
75616 F20110112_AABDBG owings_r_Page_183.QC.jpg
0b1ec74acb782a007d04223dfbc9f976
9885dce34ee50421b6e56a4547fed4e5d17f6e39
81753 F20110112_AABDAS owings_r_Page_129.QC.jpg
bddeef4af5a59bffd5deb9c984a90004
f13c400b0fa60d1aa8b6e490d93451bf90c7a105
1681 F20110112_AABCWA owings_r_Page_065.txt
be8511d91f1db77be81e761dc47165f7
a43cf70c9e7dfd6e01dc56a15a97df4eea7c564c
2005 F20110112_AABCVM owings_r_Page_043.txt
928b7471c03ec71b37dec33b4f82c550
bacef468a482590d7d480274579386d55c9783d3
1379 F20110112_AABCUY owings_r_Page_019.txt
376d6480f9cead5a501c3020a4224333
fdec4c834279b6a0f46f2d8cca8ec642f8621779
196887 F20110112_AABBSK owings_r_Page_170.jpg
38274dbf7c41dba22978c879d6f9de8f
10d3f9b3cb8948e9b9b21a78f7300b5cd7b563eb
167281 F20110112_AABBRW owings_r_Page_079.jpg
91056c8776d426e6b281adfcf1d6216f
6c7397871d245b60519ebc5dbab1d74801c54252
105441 F20110112_AABDBH owings_r_Page_149.QC.jpg
d7762c81c6ff62a6e30decb11bb3c0d4
8fd2d717da67169956e51af2e1bf5b07e59ae17e
23824 F20110112_AABDAT owings_r_Page_123thm.jpg
e2593690d6ff897a18732d3948c2c73c
c1170377b2d04d4d2d0da6688be30089dcba9d8c
1769 F20110112_AABCWB owings_r_Page_066.txt
fc59830efa5abaeab1c8508328531729
b1b30726e13e1fb4fe80a02d15e560e8fc09fffd
1722 F20110112_AABCVN owings_r_Page_044.txt
7ff271f6d598bb628b2569878e4dbb83
5c79ade05766566799d082c05dbd8a7ea13eefa5
1764 F20110112_AABCUZ owings_r_Page_021.txt
2656711bc7b118ab0db2209693a42e3f
d8934de6783fe35e2bb92408502b7199727b1d20
40471 F20110112_AABBSL owings_r_Page_037.pro
20d2f22d7677adc61e9879842a1f88db
79bdf298d77620c045459b0f3b6c44e2de461a86
24330 F20110112_AABBRX owings_r_Page_202thm.jpg
049b4d00d649523fe3cb6bdd20d1c965
7d96520ab86812a1884f860bbec3f3573ab890cd
74519 F20110112_AABDBI owings_r_Page_049.QC.jpg
3e8d67ce5610c3ea293243a7e7edbe26
2137ae0dc07a5a4cc55fba90edb914f956ce9639
74195 F20110112_AABDAU owings_r_Page_167.QC.jpg
4ccc54f9ed4a2abef2b99f78363beb84
97a3ccab4bac7c3e59228f121876092585833dc2
2127 F20110112_AABCWC owings_r_Page_067.txt
105617089284006c0d263373c87bf3eb
e13bda23f20e62467576ab49d7add3b9162bff44
2165 F20110112_AABCVO owings_r_Page_045.txt
9543ef459553230fd1ca147dafff1ac4
acb0557ef779b018073e40c7dcb813759db0136d
46744 F20110112_AABBTA owings_r_Page_027thm.jpg
06d4cb53ffecf085af75b690b1058afa
4c09c4b37827d5a79e773ad12fd889c637d6751c
F20110112_AABBSM owings_r_Page_075.txt
f37579b6227db374307b2d1c5b43b303
c9fef07104fe4419501eea530530ab469ad2430c
43683 F20110112_AABDBJ owings_r_Page_096.QC.jpg
6f72d9304700a9e81a8ba8e8cc13e7b7
eb73241ca916ddcab81c0d4bdd914c9992bfddc5
24309 F20110112_AABDAV owings_r_Page_043thm.jpg
b17f4c18ab12f6f5090466adb9af6637
92df047f9dd36ed33b55d7a3bcd7d025bccc1b1b
1651 F20110112_AABCWD owings_r_Page_068.txt
8fdec1f2fc5e958d8fd23b1bd9c1d5eb
3ac5bd2491db97f4c2214017dc00b1996ec716c0
2118 F20110112_AABCVP owings_r_Page_047.txt
ed98799dc74ae775e9308103d215d010
6fd5783e5522226444fa7af38981cb49794f5212
1051977 F20110112_AABBTB owings_r_Page_101.jp2
aee6676e66aec67c2e984ff8d954e614
8c3c7b3e872ff18a8b32d5710f05de44c5cf05ce
24436 F20110112_AABBSN owings_r_Page_054thm.jpg
ce57d10f4f4d2f57558d5aaa72232ab4
178073abf3a129d03e7de7390962902cb0231da3
42724 F20110112_AABBRY owings_r_Page_164.QC.jpg
68900e6745f8003796c6005f1708133a
7f58c2b3c254a2692b7e9cad3c4f2aa4104e41e7
81521 F20110112_AABDBK owings_r_Page_151.QC.jpg
3e15cab00e9dba6ecc937649b271dc36
b45c081f8687038b6ac1cd23940310fadc2dd594
49973 F20110112_AABDAW owings_r_Page_086thm.jpg
2d3d3e92ad9920e053d3eb3a2883fed7
a1b0d52c83193f58e113ae910d6ee3c0ccffb237
1795 F20110112_AABCWE owings_r_Page_069.txt
045fce96c799d0fca6279f8feedce020
b6e65ac3de2aa6201cdd887fd2553398cf0f631e
1754 F20110112_AABCVQ owings_r_Page_048.txt
6d3a646fcfcfe1914e038f3683bdd835
ed498e0a1d752fac03f8dcaed101b68fb8c3e223
85344 F20110112_AABBTC owings_r_Page_140.QC.jpg
a2f9b3176108a69703709e99deb23702
8fb97ee964a51850da830ffe1826f01f804f0fc7
1050147 F20110112_AABBSO owings_r_Page_182.jp2
965e2f875285dd9e36daea5c75f80fbc
d780597af400c817e68cffb13a9228ea70e4abde
113439 F20110112_AABBRZ owings_r_Page_035.jp2
f3e04675a63048452aa69aecfea00180
d3253bd6635b239dd4b9a36ace0228c3cf07e51a
74106 F20110112_AABDCA owings_r_Page_082.QC.jpg
9c96080a3e39b313d507569167641d85
6e0dc6159f32f10c1299574a81c792c72e7ae226
76781 F20110112_AABDBL owings_r_Page_146.QC.jpg
39118252ed196e7dcb555c1d2a4b7be1
00705b9c38e5367700baba919ecda4a9188a9296
25402 F20110112_AABDAX owings_r_Page_145thm.jpg
784e5d02b201b31840aacd78ca317307
2d32f587481b4603bc6ca7df1d9c6baed15f2d18
1836 F20110112_AABCWF owings_r_Page_070.txt
ccc0b6295e03ab98d5f9dc3c608a5daa
ec20b3c29784e92ef0edb8fedaebbb534dc313c1
1904 F20110112_AABCVR owings_r_Page_049.txt
ab2ce70bf156cdf22639988c23941544
073241ca9de3ee387797c980818f06739a404043
F20110112_AABBTD owings_r_Page_030.tif
ab8dd18a65f1df2b7031d3348a6f6faf
e0b088142b9a1425697923caa3a35a5ce4290f09
20398 F20110112_AABBSP owings_r_Page_097thm.jpg
707ee87315faf6761f4965e8dd1e1ce6
71f3f14843e2e5b1f09e76aa4b2c19a13b927db3
95200 F20110112_AABDCB owings_r_Page_131.QC.jpg
66b971732154470356d40efbcef938fe
6b7f7acd8d8f078b7adbe8873850895ab19e9dd7
52685 F20110112_AABDBM owings_r_Page_194.QC.jpg
e18f60ed9481bb31f3604f323cde4bf7
9528418fafa04a49bd9839917988d4677663e39e
80258 F20110112_AABDAY owings_r_Page_035.QC.jpg
f8c0d86a5d6ce65f9d9c48f0bfd69ae7
2bc56f9a5981067044f666e03f42409967a113b8
1733 F20110112_AABCWG owings_r_Page_071.txt
ce94d47d29f41038b5d7d83113dd49de
dd4fdfc40eb371a666ff824379a389b71e3edff9
2148 F20110112_AABCVS owings_r_Page_051.txt
1bb75668c7c1915f68992d326d007361
06733aa1fc200e666b57e7d04cef8155df5b9623
42168 F20110112_AABBTE owings_r_Page_066.pro
86f6b01e9a903ea9a77c6e4a9ada53fd
7597332b3c727be59a424cfd01c8d7aa37254b1a
F20110112_AABBSQ owings_r_Page_026.tif
57b5e2e0c6718a7e17ecf45c51a0e698
f4979236ddfb48c01b6b0d4aed764870666859f9
47050 F20110112_AABDCC owings_r_Page_068thm.jpg
288b2108531fb735ac97104f1ce4e42f
cc2f20e83d2708dc6fd6bc1f32d3e3e43f88365e
104789 F20110112_AABDBN owings_r_Page_102.QC.jpg
22c4d484cd48127996cafb73a38b092d
3e8af414f2cf21a18832bb2aa33668cace5ce9aa
13212 F20110112_AABDAZ owings_r_Page_124thm.jpg
7fd9f204ea2faf59340d4818dfb6e9f6
97fa2187f118121874eb4f295e06f88b53a70ec8
1930 F20110112_AABCWH owings_r_Page_073.txt
93958322618b573412f23d1c3dbbc32e
129b563fe0a19b37ca867cedde06425acf605771
1968 F20110112_AABCVT owings_r_Page_053.txt
94776b33a2f402aa313f9e154e2f8358
e3e538c0519e1722330f0f72801871e334a8c274
262420 F20110112_AABBTF owings_r_Page_203.jpg
9838fdd9668b0bdf9e5fad905fd7b051
40d24f770e034d6ce22c563f3e16e35c0f79782c
101076 F20110112_AABBSR owings_r_Page_197.QC.jpg
89ae9767df348c2356485bfae095bdd9
a99227d88e993089619e475a36e6cb0d41aa6571
42990 F20110112_AABDCD owings_r_Page_124.QC.jpg
e145dc440554d3dc6eb085b2c8ea6ef2
5d02faa114558c2d41ae322f53ac761ceefa9d95
102680 F20110112_AABDBO owings_r_Page_114.QC.jpg
4da5c382952f1519f0216a47eb2f741a
1268125bb174b3868f9fa0784d2c321f2f52a7ae
1990 F20110112_AABCWI owings_r_Page_076.txt
bdb42d6baed2a4957dff2cafa41bc002
7a9510d1b0be32c81d2d8dc01e61de7858811193
1834 F20110112_AABCVU owings_r_Page_054.txt
5091e6d0c2a7fc6e1d686a8882b51559
1a75a7a68a054e2ff5f858dc59bdfb9dfac5a782
F20110112_AABBTG owings_r_Page_129.tif
c1fa39e6970db3ececc61b0a5faa7b46
a28575e8ab584c9337ae42c1fe7c0c9a3a632790
3050 F20110112_AABBSS owings_r_Page_003thm.jpg
fd3e6ad5db84080c820592811f3a64a3
11538ee2b52981ac56515b1f696dc08c04d296a3
77841 F20110112_AABDCE owings_r_Page_207.QC.jpg
14ba8f4422b583e3191bb44c54004bad
b1545f5119f1dd8bec29973fcbe82d90e9e7e811
82389 F20110112_AABDBP owings_r_Page_058.QC.jpg
2db0a071e14f9971d30b0fb80051b786
de0a98d53ed711820f65072c6c71056b1559f99e
1593 F20110112_AABCWJ owings_r_Page_078.txt
e6c187b855d91b7786b72d8303df6123
160ed1dbe2e442c6665b903f1f93d51506a54dc8
2257 F20110112_AABCVV owings_r_Page_058.txt
393d91aa23ce968423d4ce7f5ec18093
760373d33f3ee48b742a7316634dc06be8cab3b4
F20110112_AABBTH owings_r_Page_147.tif
239498da3a8b22bfb2a676321bd548b2
a542cac24026ff8e44080e4c79eb5028efb2a1b7
23705 F20110112_AABBST owings_r_Page_058thm.jpg
958eb8bd0fd306150cbe5ad160211c38
97d1c961d1d5d0a282ee1d761a304685bdd91abc
99238 F20110112_AABDCF owings_r_Page_069.QC.jpg
ced3c0cc15bd78ae626a692e68ca6cf9
132c4329b837f09c5c3c43c3f1f7191cd3c56e10
5523 F20110112_AABDBQ owings_r_Page_003.QC.jpg
4f2c42c1511289447e6af62036dc97cb
aaca2f3595884a8008b409a800767efffbbea0ac
2187 F20110112_AABCWK owings_r_Page_081.txt
91f9cd4996dd0300ddb7f94808766472
108afb5e0ec4298498e361ade852d005b5f55505
2000 F20110112_AABCVW owings_r_Page_059.txt
107451d61c1aeeb582c003866dcae494
12abf464ba5ea5ce610498eaa10d7d76bab11202
80150 F20110112_AABBTI owings_r_Page_034.QC.jpg
48f06cba3a10b6717eb4c7ed420d45cf
8bd8afa91cc461675a57224e302dc11b47dcf71e
72173 F20110112_AABBSU owings_r_Page_190.jp2
8b28ee18bcb1cad115b2bf3260cec032
6b51f0d218e0d04a9e0379888272c20336c14ee2
102535 F20110112_AABDCG owings_r_Page_105.QC.jpg
1ba50ab671ccc403423a58ee64852f8d
0fe6d52e566b133165578f1d47ea10e44fb256bc
21376 F20110112_AABDBR owings_r_Page_130thm.jpg
3e9f7e43376e140c8852bf642dc5841c
ce76e0d2f8ed73a6cae57d9a73a1662a9c1f9ad1
1895 F20110112_AABCWL owings_r_Page_082.txt
5de4d3c453767e642beee4c61d758dcd
20fa0b749abf46a4e875217d3b55dee247f11bca
1771 F20110112_AABCVX owings_r_Page_060.txt
2ac4b52cccf368cdd876da054f9d466d
583465f730cd6ead0f7e611a3197eb6f304c2e7c
323526 F20110112_AABBTJ owings_r_Page_008.jpg
79cb3074dc44bad845ea886c3d4171f7
50f0f248e3d9e8314cec782df2e382b20e3cb0fe
F20110112_AABBSV owings_r_Page_149.tif
300b8bd6f9cacda20908870390383fcc
2baf83b24ab597b80073979e18a4d5336aaf2186
16487 F20110112_AABDBS owings_r_Page_001.QC.jpg
842c072375cec558d548deaa1b8e0570
a70a1f6a6cb3a21c71aaa295df4d8f2a689693a7
2051 F20110112_AABCXA owings_r_Page_107.txt
181de7ed56bea57b68bdaa74b63a290d
7f93a518ce9cc4bbc51a1e1075cf811f7d68cba3
1603 F20110112_AABCWM owings_r_Page_083.txt
5e8e45d63b23a5572ebe7714a5ba90ea
b5332ae8128782889f40c58307690765cab4b4cd
1645 F20110112_AABCVY owings_r_Page_062.txt
6ae6b805854994ea2d6878896ea4a3b4
0cd640f8278f20f628d5e51caf65115b3d600abd
198969 F20110112_AABBTK owings_r_Page_073.jpg
fe1d7bffc6286b2225faa8a1c6ea2849
7af1aca982ea94c33da4dcc0ae365848e2e7d158
113754 F20110112_AABBSW owings_r_Page_178.jpg
c4fb5b4f520dece6d6efdfbb916091f7
845dce5ee407277bf97ecb6a1a75265d5b0b5ad5
25107 F20110112_AABDCH owings_r_Page_208thm.jpg
0e949609df4f21519b16ec0b38094a16
41987c0ebddaa811872542d102d0c653ddb708ce
25030 F20110112_AABDBT owings_r_Page_129thm.jpg
be198316829449b10cc0e7956d23d461
89d647446d972554f8113e8a49eedc73fa8bcc6d
1775 F20110112_AABCXB owings_r_Page_110.txt
d9fc06b3a18a00d09aafcf56451e0fbd
b634df73883f03d353682ac02746055c28799ba4
1749 F20110112_AABCWN owings_r_Page_085.txt
15a3b786424956e8becea6c2df2b8ea4
266d183a19bbfeecf336462de50659470bf49220
F20110112_AABCVZ owings_r_Page_063.txt
0cbab03ef82fe59b20688614624e87ca
c7aa9339891961f2fb4bebd804e6a5b6fbdce767
22531 F20110112_AABBTL owings_r_Page_165thm.jpg
f15a8da2a248e843ecc372eddc7bae0c
8a4486b5f83c5eff7ccda7f151d4e740b7a5e3a7
45735 F20110112_AABBSX owings_r_Page_083thm.jpg
29437bcc9658a8a1cf026ff4d0136953
4015a9a9983e8d5be1ff8e7a2e92c77ef7ca2dda
48926 F20110112_AABDCI owings_r_Page_109thm.jpg
c47fc8621dda6895f7e75ba85a3e4956
14284885c6c56950285289abf747cfd251d57223
51868 F20110112_AABDBU owings_r_Page_044thm.jpg
f740408f9111e0417c5100b0a746b613
4c2aad76c2e01d6ae3cd8ac0e4feac3018e106f5
1965 F20110112_AABCXC owings_r_Page_111.txt
634cb8ca5330e2cf3a5ffdd2ae3dabf7
6d66e5bf4b05aa35057cfdec9f99a1b62b979b5e
2067 F20110112_AABCWO owings_r_Page_088.txt
438bddb19720389698718cd8ab7c4c1f
8fe6f4632be5baaf5311389459e96da1dd11e6cd
214762 F20110112_AABBTM owings_r_Page_091.jpg
dd63723a8f713927440b46a2d6818e09
ba5835d808125bb58dec1924653db4f3de00dbcb
6317 F20110112_AABBSY owings_r_Page_003.jp2
814490d888612506ec8a7384c7f5952c
2552fb23ea62cc181d1fcf1462dd942e40034afc
17871 F20110112_AABBUA owings_r_Page_019thm.jpg
b3f616f9d91bc37616024d6351a47eef
61abef42d9ab8a7693ea5390918d7cc32c65d810
79334 F20110112_AABDCJ owings_r_Page_043.QC.jpg
c3714391fc4842df18bd74f58530f601
620b21bffec5756d41c11a55f71134dc51b41fc5
25163 F20110112_AABDBV owings_r_Page_022thm.jpg
69b069ef3f6a84a4346305f163d821e5
c3ffc845f052132e153d44310a5d62a7a3cf03ca
1880 F20110112_AABCXD owings_r_Page_112.txt
7aa2e0d351d976797c81f7fdf50ae8e0
ea166728b86e07da55ba28ed2b69a639fd049364
F20110112_AABCWP owings_r_Page_090.txt
4caa20e8cf6926e2eab2f7c4b8f88337
e3d1f8b5d724e05e679496a76ca50a4933b33c78
46853 F20110112_AABBTN owings_r_Page_075thm.jpg
b46cfe255e16beabffabe7d539382e7c
b78611ab8596fe1deca90b45e199ce6f3c5be1dc
228833 F20110112_AABBUB owings_r_Page_077.jpg
e32729e4783f105cd80ad01149e0a3f4
4f4e8ddbb52fbdbf17ebe8a0f5a5d8ca05eb06ca
24188 F20110112_AABDCK owings_r_Page_023thm.jpg
28f3eff5696da2e95dbf5ed484573050
dcfe78ba4ec4fc40dd4b093c2434de6b1e1d9514
45658 F20110112_AABDBW owings_r_Page_133thm.jpg
0b3bcda279d2463c5eb36cf6d38a6e53
7bef5495f5551f5181947860fe851a620fac3e2c
F20110112_AABCXE owings_r_Page_113.txt
9ecc92dd7a01fcb0870a2ddad22b0c41
f414a17b7c68ccce44b96d00c96686d93dc506cc
1876 F20110112_AABCWQ owings_r_Page_091.txt
df5278eda6ac23b0aa98f18d1f4cd199
e09cb65e32d08e331e48f8fa487457b5b61cbef2
F20110112_AABBTO owings_r_Page_070.tif
6b9b7a8be7ece1efc180a5cf9531bbff
2a3e06e3ff1c6d6137f55a05e6383920a3b7613e
189477 F20110112_AABBSZ owings_r_Page_152.jpg
c47bb11c117328e0b00138cbd8cc0af8
e6790e4a278a44ebe0e1f46fa200dd5d48ccad4b
207881 F20110112_AABBUC owings_r_Page_062.jpg
fb73d6a07cbdb698b2171c6e46b5f186
f2e5dd953a04cdcc02ecaa2e03cbd5987fa4d7e4
24719 F20110112_AABDDA owings_r_Page_087thm.jpg
f98cea2a14fa7aa414fc8712a5f7f5ba
0f69df8dd3659ab4fb1781755589d01e5a4abb1c
13859 F20110112_AABDCL owings_r_Page_096thm.jpg
a1e98d84e55a94ef6a7b89054b384e27
a02330f434751757ee905fc21a291247839327f6
47674 F20110112_AABDBX owings_r_Page_039thm.jpg
2555572231ff0a575df3d53f6b872317
db503ad49f2f6cc0cbb1d049e5ac649e4c879fd1
1909 F20110112_AABCXF owings_r_Page_114.txt
6cfc8096ebe2fc0f815d66e08420249c
d48fab75466211df125018d604fbff0e48cef172
1838 F20110112_AABCWR owings_r_Page_092.txt
5f6dfd3079b6a6dadae8061946c0e4a6
58bc6f6fc50bbace2193ddf4fafb04b94939a000
41012 F20110112_AABBTP owings_r_Page_021.pro
17e6326c5f34ad45234306c67ef3a615
177a01d5809000562a554eab5190414958aa1321
39489 F20110112_AABBUD owings_r_Page_071.pro
8fe5f1969d0a045abbe3839cb02c18fa
f16bcbb892c38d2586987b2d3d9beb0fcd8ee497
96195 F20110112_AABDDB owings_r_Page_048.QC.jpg
bc3135a392752dfffe4d630c50f39f8f
5e7b6e47473df2e1579fd35aa712d28321e1816a
60671 F20110112_AABDCM owings_r_Page_018.QC.jpg
7aa69230d0fa460f4de6c638302bc96f
93981d1caba8c6bd8f47581ff407eef47cfb6f84
49903 F20110112_AABDBY owings_r_Page_084thm.jpg
61d61eef56772b8b56d35d9498b47f32
d1b8d4bea8195b1ab469b6f0abf3f40408c81280
1789 F20110112_AABCXG owings_r_Page_115.txt
030ed3e0cd33ef6204246ccffb40a8b8
fb86ee2d149a815a20a4143e5a8621ef1df18aa9
1520 F20110112_AABCWS owings_r_Page_093.txt
b1ee08cce552b4bc45fc42545c5c9a69
0f266387980c244985a22539d156069f851e63bd
50763 F20110112_AABBTQ owings_r_Page_149thm.jpg
6a52bb7a8e33a597c5ebc7df2b86d72e
3458f9b4528c80c31d4b87214a860c26c77397d8
44401 F20110112_AABBUE owings_r_Page_154.pro
9656eb36332f98e71c82fb44374fd144
ee53f9a6521b856dde9c22c8e339405f720e5a3a
49727 F20110112_AABDDC owings_r_Page_102thm.jpg
fe6664dccdb144d7364a69e7c1bb07d7
51358e637912eecc7d9b39ec1101fa3ad915720c
25351 F20110112_AABDCN owings_r_Page_205thm.jpg
9fc7e97af5cad89be97f61615373e463
4106799fe940223a6e1233f7b9ab01eae568d3be
67863 F20110112_AABDBZ owings_r_Page_148.QC.jpg
b9ec287c83a57974fc1be9d919cb65b9
be78dde3f8293f5e0affa8dbd945e38a978fa5cd
967 F20110112_AABCXH owings_r_Page_117.txt
19f9d0cd209116494c4fe95194b2d897
4b7559864af122d0338cc225ab3bdf7382d2cac3
2041 F20110112_AABCWT owings_r_Page_094.txt
0348d1335b929789d02508137815e386
8f68df7b596e45e97d53201a6640152055072db2
42782 F20110112_AABBTR owings_r_Page_024.jp2
e854a8608616fa0fca31bc2ab20d99f8
3acaf3ee56ac263ca582f21d2167ae52aff91ea8
195129 F20110112_AABCAA owings_r_Page_031.jpg
4a15dba4d01d2b25a16b29c514c1a2eb
be175cef7748bea6c7583911cd35b5b2ed4d4687
F20110112_AABBUF owings_r_Page_140.tif
bf0af380b136b55d146e1cf135130762
9578b14e2f56dc993bbcf89b621525280a7ec750
48819 F20110112_AABDDD owings_r_Page_116thm.jpg
1e16d002a28dbd08d24fc5c8ebbe9664
0b95390707f555a86daddfdb79d66ed8319ca9ee
101421 F20110112_AABDCO owings_r_Page_088.QC.jpg
7ae4457455a78ae9ef3a18d02647fb8c
dcb4e51d730cb5340827e73ac29652e71d1d170a
2073 F20110112_AABCXI owings_r_Page_118.txt
1817220b97ae91c75a5856624bae15bc
5097ddc62bd689f7fe10ab4458ba9ec3159fe6c6
F20110112_AABCWU owings_r_Page_095.txt
4ba77d70242f8ed03cd52d06e8e2885c
11d0e966cb85eff2e12f32b6b932d0ee4a1c255a
F20110112_AABBTS owings_r_Page_069.tif
b88ab708e8f168775beff27f18538059
63239d3faaabd0c94f9320aa380d5fe25dd9cda6
197497 F20110112_AABCAB owings_r_Page_032.jpg
823d63b86b06fbb3a925e22c43b8dc16
777682832fe6a2568395ea3739663ddc4480b663
237153 F20110112_AABBUG owings_r_Page_106.jpg
41e51a04419aad9aafb07749dc05c0e8
e7a53586367e47cda5a4ca9fe8645bdf38c1c00c
24922 F20110112_AABDDE owings_r_Page_035thm.jpg
0ba0e05fc12244f5a05e455554d437a3
e0862914bd2c8ff4ed062abf3846b6e98b977dee
32005 F20110112_AABDCP owings_r_Page_009thm.jpg
9dd20e4d5f8c7db1e2052bab91b3a587
50eb1ef862967abe088753bd883106a8a20851aa
1898 F20110112_AABCXJ owings_r_Page_120.txt
1d00824f4a2a6704c89daef2dc4f2d69
44743409e7708980d654a9158aecbf36d705302c
1770 F20110112_AABCWV owings_r_Page_097.txt
54caf5bfbdfe265e02ca11ae5acc0109
fbb2a4f7588227c52f1d2184745f17a707285777
1911 F20110112_AABBTT owings_r_Page_153.txt
b339367cea60b3bf8ee5d18f27b634b1
1cd7631e149ca9548a7db553ac670dc6c40f6a2b
212720 F20110112_AABCAC owings_r_Page_034.jpg
24e3514842bb208c51f18419daa09a7c
29d6b8ced85fafb85d3501851205c5a71f2c7dd5
46738 F20110112_AABBUH owings_r_Page_021thm.jpg
241332220cde4e36670f122ed353dbe8
d7091b0fba1f4c1a1b8e4dddcd443ae80b126771
82549 F20110112_AABDDF owings_r_Page_157.QC.jpg
3c857e6e17f5946da79b9d339a508c3a
7764142cbe443c75dc9c2e4f2b2225a3465b7afe
69229 F20110112_AABDCQ owings_r_Page_189.QC.jpg
25a1c57e322916c8708604d01e43be61
cfee6b3c4129e5cdf20a15507d21d43905a27bfc
2099 F20110112_AABCXK owings_r_Page_122.txt
5db58aabaf46b54332715df13ce68df5
92fa009c03479f1dbb8c7f21b1dc7bb622a08487
F20110112_AABCWW owings_r_Page_100.txt
ffe244b2cec191325965ca2a5a4dbb18
963df08061cc117ef52a8d8e538fb16cca08a1fd
51370 F20110112_AABBTU owings_r_Page_081.pro
7912836bc4bab651e0faae9ebede9e52
73a3ef1c4afae82445ca80b9059f2f8630b1925b
221267 F20110112_AABCAD owings_r_Page_035.jpg
1d6fd1022474cd39e99917cc69ae2175
b3075bcf3926c6841153c0221394ff85678cf85d
180031 F20110112_AABBUI owings_r_Page_060.jpg
d3f6debb1b5d81e0ca84632d65889710
73a0bdea528295ecdcbf20b52a2d69827139b010
78436 F20110112_AABDDG owings_r_Page_095.QC.jpg
fd74894415a72418c7d8ee5c295dd0dc
713bde4bae3a6b32c74656dafb6fd4546f89bab0
100041 F20110112_AABDCR owings_r_Page_077.QC.jpg
305e4b75f13f1188fa53f38952938661
c61930708e45dbc36b1bbc34ad11292f56e3b604
F20110112_AABCXL owings_r_Page_123.txt
0399b962c5d137d1fecc89396bea078d
faa7903597738f321b8d7cfd9e9776fbe6b731fc
1805 F20110112_AABCWX owings_r_Page_101.txt
6d4dcd7962f4b083e9a0f9a3c0d1c0ab
77c1074658e8c7999b264412c7b16e31e3c0752a
F20110112_AABBTV owings_r_Page_144.tif
73f0e4cec6def63567b1eb1df8ec413b
8f87ccbf23a2e13084ce7294777c0856651324ae
222158 F20110112_AABCAE owings_r_Page_036.jpg
5b32a635634cb53c44110394198e0224
d335051050b07604f2e947345fbd8f9e46485a50
1742 F20110112_AABBUJ owings_r_Page_064.txt
2196fda2df1cfbb5f7a8f10417150718
dd95a63ff76ab27e4e731a79bce01610171b6114
47025 F20110112_AABDDH owings_r_Page_117thm.jpg
fbbdbe88b313811fb99ffe075866eb6a
08b10e2e2a277d9e8aae621f31edf9d74d93f100
77458 F20110112_AABDCS owings_r_Page_180.QC.jpg
10a9d3d60d81944a345ce6045a356ffd
977257fcdfb43ed2ae0573ba1326e48235774676
254 F20110112_AABCYA owings_r_Page_147.txt
1792987b39c9a59059aa9e8613bb1ef6
a50848aebf8301180842e5295e36c27545a0e14e
1687 F20110112_AABCXM owings_r_Page_125.txt
839f534d465d0c718a54f364d343af9c
8243c2338bde4cb084397260affc233480372a9a
1928 F20110112_AABCWY owings_r_Page_102.txt
4f977cc512ac3e7b923defdc48452f86
9b6ff4751d6a1d8e8dfb0fb64c0898e3934e2354
72967 F20110112_AABBTW owings_r_Page_004.QC.jpg
02b765dcc618b6aa90bea9a71b9e67f9
578777af7b6e3f6a9ebafc905ed8252f9e1bb950
222308 F20110112_AABCAF owings_r_Page_042.jpg
10476ba9128cc22c2a970f6725d6bcbf
80672afea7fab4d2edb2aa8d3c1627a98b0e1c45
2151 F20110112_AABBUK owings_r_Page_161.txt
b8f1ff3449642231298aad3e464daffc
68d62ce5a851ea1fa80a573a67c4e9aaae8ee8c9
25013 F20110112_AABDCT owings_r_Page_207thm.jpg
0ac5869d6ca07326209de5efe10d0466
7bc0440fb94881c3851d0ea45faca974494904d9
1918 F20110112_AABCYB owings_r_Page_149.txt
eb0f7836e2928dceddfea3200cae69da
f1571ced03e207a6c67f10001aefa0327887ea10
2030 F20110112_AABCXN owings_r_Page_126.txt
9f08395ff1e875d820f2234735a0b44a
17c27fe7e4e02142e6a3ba70b3b2225deafd52a6
F20110112_AABCWZ owings_r_Page_106.txt
e01dc44623a734ee32ca06ffe5ca7e2c
9515d353390cca34bb9b10ee90991e1c54c39a65
76656 F20110112_AABBTX owings_r_Page_123.QC.jpg
f1aa93605dad0edc093baff5e0a7deb5
0c653e3a4a62a848e6a4145cf5f451b3576647ce
23729 F20110112_AABBUL owings_r_Page_030thm.jpg
791b5ae32af6f0a25b9d6394bab6f3ba
d5485cefefeb7173e24b45ad8a923325f547a092
47852 F20110112_AABDDI owings_r_Page_069thm.jpg
10fcfafe55785dd76ba6df174aa7cc39
07da52faf113de945f38c9042d88d905bb463cd0
99716 F20110112_AABDCU owings_r_Page_104.QC.jpg
8faa18eab9d4ee73087592264a20df2e
757b25694b563c13be191161d5f67b193913b028
2173 F20110112_AABCYC owings_r_Page_150.txt
79b4027bdd2bb2fc9ca1f0070ddb5a55
4f5091e878b122ff0c7534031c5387a5968895a0
1461 F20110112_AABCXO owings_r_Page_127.txt
737486ef50d3623a52b6ea6c6e5a97b7
70d255bf41330e2e0c4fdcf3210a11c2356d4d75
45933 F20110112_AABBTY owings_r_Page_006thm.jpg
13e7bb4ab1feb097966e8ee4a3d3eb45
4f48465c731e2940d21b7a76a9589c937fb0bc63
209162 F20110112_AABCAG owings_r_Page_043.jpg
a0069b03bb7b3e70b9971ad658b2f8b3
df6d4c47ec8bc8abdaeb92831f6d887810dd41e6
97626 F20110112_AABBVA owings_r_Page_139.QC.jpg
b5ad3194a6e577fd3dc364fdc7533ad2
bcc7a38fd4264483af73d438a7787f01ffb87ab8
92129 F20110112_AABBUM owings_r_Page_027.QC.jpg
3436b9d5e856e2e5b75fb9bc03daf450
ca082e9b944133b8e18039b6d408840fb5903680
43432 F20110112_AABDDJ owings_r_Page_180thm.jpg
9b5f1a7e3052e69982d60830297b8876
3c0e5a6527b2d2701e2f64d3747d1bc8b58dd90d
52185 F20110112_AABDCV owings_r_Page_201thm.jpg
cc1a068ee424e261f888f7968aeec418
25932b57bb96145627f9cfc6338e1fbcfb576044
F20110112_AABCYD owings_r_Page_152.txt
77823fee0875e5d89d8ec2c9475d8fdc
3fa729debd41833a040e757f76a547332083c018
2312 F20110112_AABCXP owings_r_Page_129.txt
841a1e29b3ffc5e715dee6f9522cdb4a
4aeae02d192b01851d19ba11702d7230d2f46991
41250 F20110112_AABBTZ owings_r_Page_056thm.jpg
10a06e913d077ac22675f66b714e8cf3
9d5cb912ca133460047731c02dd4b37f38d0b6a3
251082 F20110112_AABCAH owings_r_Page_044.jpg
6d74b4f89d19139ffbbe7db57423a659
f05dc4d34f95789aceac8454dd37a54496457630
F20110112_AABBVB owings_r_Page_074.tif
f641eddf25770d58f8eae81d454c63b4
5f4e2881f9fa3daf7f6be5392989d1d68448e908
78587 F20110112_AABBUN owings_r_Page_142.QC.jpg
b530bf2375e0d8845c22d7258bcf6707
ae291154b113611ce718afa49d9287c1fea50d39
65134 F20110112_AABDDK owings_r_Page_199.QC.jpg
48496315d646248b3b648d33a9ff7044
9f9d8f4f0cc306304880bda6377829471110aefa
49044 F20110112_AABDCW owings_r_Page_077thm.jpg
66a2eee2c04d82a6bc3618b804c7255e
f029dc12623a79919e9b26088ce1468f83a137d9
1940 F20110112_AABCYE owings_r_Page_154.txt
409f499a319236c79a6fd4f1c4696cc4
a4cecd842062ce79bc384a67fff7b8264b6c1a42
1737 F20110112_AABCXQ owings_r_Page_131.txt
82cd601c92461261a948310eb54834f6
1157ef5252bd48a48835d883558c8eeca9a08c3f
197924 F20110112_AABCAI owings_r_Page_045.jpg
fd6a3003994ca26da6279ed4cd174a2f
0dbb077f7d1d37b2a04997b59277dfc4fd9fd6c1
48628 F20110112_AABBVC owings_r_Page_099.pro
38c18ba8191682c105b272871a3d6c57
08e1f487edf35dffa445943773ddb690aff3ca20
1394 F20110112_AABBUO owings_r_Page_080.txt
7b0e4d5a99eb367348af784b72079cca
560759833fbc604e48e57ac560e65b18237d18d4
94764 F20110112_AABDEA owings_r_Page_135.QC.jpg
1922df2f610717d104c5fcdb2b69bd46
f02d7f1ae41cfaedbab1337453f58818d6a58b32
130731 F20110112_AABDDL owings_r_Page_016.QC.jpg
d39eb9070eb9e997482c3ed362c80588
82f7d1327679af834718e3d694da2da269902cb8
57847 F20110112_AABDCX owings_r_Page_015thm.jpg
b1173a2ff0e972fae365ee2e4fb6405c
560819a4b531dfca04ca8a1461783e9b94d58557
1985 F20110112_AABCYF owings_r_Page_160.txt
6ae981f8526a54b265d7e5ed3c3ebc9f
8eaf2df9be5913894496f2c40aa0ad01ae7a5be5
1091 F20110112_AABCXR owings_r_Page_132.txt
608607dce4c08749ea32b454608b86ec
bcb0fb9342da073d7b0cf2d8dd85fd728b97ead1
193838 F20110112_AABCAJ owings_r_Page_046.jpg
de6f5b854c4e393fde1859956a6decae
6ad3108cbd25a918ac69af04caae7c0fbae7c845
2002 F20110112_AABBVD owings_r_Page_116.txt
0c98eec02babb1ba3641b63d00ff5dce
9c4fb8cf81e12ce22c250025cd966fd0bac16c39
83735 F20110112_AABBUP owings_r_Page_009.jpg
d11de1d3b255db13f4db5bcdb80db6ea
4318cd5d63ba87ba7b654d49e1689eeb5b4b9656
46803 F20110112_AABDEB owings_r_Page_038thm.jpg
748108d0981ed75ff47223d3fcacf34f
932f0b32ead0e71dccace2fef9ea0008c9527d97
49576 F20110112_AABDDM owings_r_Page_067thm.jpg
2f8b1a8bd24f719c589cf179497a3a43
a68d3e20e6ddc91e896731865452cf762b0b4d7f
43655 F20110112_AABDCY owings_r_Page_177thm.jpg
ed642e0ce437515d26b63a40a3e11fa3
6d36c06f8a7e208eb2c7b44bc46a58c9fdd0d741
2429 F20110112_AABCYG owings_r_Page_162.txt
d44b6e559ac776ca262300006b4cd966
18c3819632c3d4d071d7a9e4c77fd8f988fc46dc
1626 F20110112_AABCXS owings_r_Page_133.txt
eaa24b6443724a2b7c03da526795bdd4
4f2b788552f3d8ba7527405b6f3e9055aa2a540c
220551 F20110112_AABCAK owings_r_Page_047.jpg
9ceb2086fc242aa865d4f19e74851fd0
31a27b720cb873c3f20000279daeaa2efae700bf
1051951 F20110112_AABBVE owings_r_Page_105.jp2
02927e984430290a4c2b8bb3ec723998
df637d8894d285a86feaf1071b2d863c904cac60
212244 F20110112_AABBUQ owings_r_Page_119.jpg
aa1901a9de79ba6a1991a31cfe37a530
5e6aa224498ebff0843e604defb82d1e4c5cd147
88961 F20110112_AABDEC owings_r_Page_138.QC.jpg
64e25057edf95558d62bfe1480e9dc74
b125a9c208e6d4fa532753c2662388085222650a
23140 F20110112_AABDDN owings_r_Page_128thm.jpg
793bcb3a67e1e8491fc1a9142d542ac8
c3594b64363c6a7a9a01f5b085cab13014c9ee56
95029 F20110112_AABDCZ owings_r_Page_075.QC.jpg
30cc1c6e681328c91ec2580e12278824
b7ee9f417d9f0241786ddd34a64822c374a1a89f
2084 F20110112_AABCYH owings_r_Page_163.txt
58bb79397de2443d367705d35a411284
0a9ad6ad79d8310df1c387f1766ecd180c31b274
1309 F20110112_AABCXT owings_r_Page_134.txt
849868c10c5da4446237093ac2416d7a
31bdb805e4a84f28dc4591da391990c203eb0cc3
207790 F20110112_AABCAL owings_r_Page_050.jpg
9e43b6deb809a2a9a519ce9f78f71ac8
e0cadc5bf42f493adc2079f2399dc363e638c770
50820 F20110112_AABBVF owings_r_Page_166.pro
f543c79d14f5ce8c5f751d3d0b8c7bd1
ac8224931d70453e36f31f396a102b6fdf49fe23
1252 F20110112_AABBUR owings_r_Page_055.txt
ad07a617aa7fc20287038e7c4aef7c94
16d7e8ff4bb435f36a21f5759a946022237a622a
218248 F20110112_AABCBA owings_r_Page_075.jpg
ea4ee81e24b110bc63125645632da94b
24966f9de850019daf9adcf3e5aef811dc2aea30
92986 F20110112_AABDED owings_r_Page_192.QC.jpg
ff8b14f0300598c8e20c45c51b359980
9552c6bdd126653f0f7307a902739acb322be600
23085 F20110112_AABDDO owings_r_Page_170thm.jpg
2e85ebac22fd32c4becb488d0d2fa268
b1f7157e8b67f7e3115ba985738f602bcc7ad0bd
1033 F20110112_AABCYI owings_r_Page_164.txt
a2435043d7fa7d7ecc374385017341bc
ef06ce8809bb814525f62645aaa45e58ad813892
1319 F20110112_AABCXU owings_r_Page_137.txt
bd13359f5c5ff5fda9526271a3800610
1ee6f2a990d9049ad4d6a113b8866383fdbc476e
214572 F20110112_AABCAM owings_r_Page_051.jpg
3e53ddd6ca2ee8abbbc5e604f3b0897d
2303a976c8b0fef3eb33137fbecaddb17eeae760
38602 F20110112_AABBVG owings_r_Page_178thm.jpg
43c96c3a3f144c3d9f67837ec481eac2
f18daa93fd590e61ecdacfc67fc142570f81e938
1849 F20110112_AABBUS owings_r_Page_167.txt
f62877905738118a067dcecc6eeb565f
fd01e7633aacd02e5862ffd385fbbad7117bc024
195271 F20110112_AABCBB owings_r_Page_078.jpg
dc5cb6863b25a4a3c34568d072ce01ce
2c23908421aad0e2f26c9fb46efd5f1d49fd3dac
51263 F20110112_AABDEE owings_r_Page_010thm.jpg
68c1dc1028ddf25a7776c223de8a4aa0
7a86687077566cc2a62c17d77346d3e1f5c84be6
24521 F20110112_AABDDP owings_r_Page_166thm.jpg
8802e27e653a4c6a7cdbd09f48cf2ece
bc37b717b2a1c3e4af027218c6c683c3508a7e59
1855 F20110112_AABCYJ owings_r_Page_165.txt
62c5e4795828969c505497542c9c997e
874fdfba59a78e645dff67890a7b971c7503ab1f
1596 F20110112_AABCXV owings_r_Page_138.txt
6936b4b85836a70e48f571eb99c72cce
4abadd686652ffa06fe54b9204cdee6963ab4197
207416 F20110112_AABCAN owings_r_Page_052.jpg
79c2c1922a9d6ec258b6745d11f7dbeb
7752d80f1e6962c0d8e323c453a6aa4e65a60f16
23326 F20110112_AABBVH owings_r_Page_036thm.jpg
c53c21c13d5acd0a551b2c38530f7400
54eac128ec16ac53374bf045570fa48658657fee
51400 F20110112_AABBUT owings_r_Page_008thm.jpg
6ec85e6cc59d45850c40f3d5c882365f
d3927b815407fcdde1bf6ecaa0b0965c4a394751
216479 F20110112_AABCBC owings_r_Page_080.jpg
e69a3d175a169519e3387fa557d3d08d
51fbc366f9e0ee8672e098d42abdb679036650fb
93053 F20110112_AABDEF owings_r_Page_182.QC.jpg
66b0b55db75739d3db0d676f0f6799ad
a8a72863b51dd623ad8c25bba5e00f84e8949da2
68833 F20110112_AABDDQ owings_r_Page_168.QC.jpg
eb5ca9c9d0640d947a5ecc110eb50921
150d5290a5216e1b9cd92d421d18c408db345644
F20110112_AABCYK owings_r_Page_166.txt
6844701d11f88419fb4a6902fd52203f
1ed7a40d7e87d965b5c9f982b0e10039b144a066
1421 F20110112_AABCXW owings_r_Page_140.txt
26ef089815a1a4360dc9bc588c4723aa
bdd378043104a6bcb5c99f4133012d4da35e832f
195900 F20110112_AABCAO owings_r_Page_053.jpg
052026046ab59b01b5cc404b70339e53
ba8df5c6ad525132c3eac0ed20286947ab03ce86
F20110112_AABBVI owings_r_Page_058.tif
fef7bf8a731da0e64d4fc05f7fdc3f58
e41c95ee00d5b8dc0d390556f45635a5a591d3cb
222392 F20110112_AABBUU owings_r_Page_085.jpg
8436e59e916e0ac5fc06933a7225dcd2
8237a55235fc9bd82c67622e604e28b8385c2d5f
299647 F20110112_AABCBD owings_r_Page_081.jpg
652697dcf17c1f0fdeb1d605a5e54e86
8cdda5b9f843aaf02ed2751240933542f6299f22
101412 F20110112_AABDEG owings_r_Page_186.QC.jpg
876ef37599453ac32f6b8d1b6b77775d
0f6e81c9c4d99578f183c69b99df6e64b8219d88
53914 F20110112_AABDDR owings_r_Page_193.QC.jpg
8850538dc1e75528ba633cf455f4e6c9
b01f1d30e9026f911a4cf27f9bf7cd4d0f8739bb
1798 F20110112_AABCYL owings_r_Page_168.txt
516c083bca4176eb198ce9f487356155
10bf705aedeff26cb78debf23d10a561a8727a67
2019 F20110112_AABCXX owings_r_Page_142.txt
aa26d1f12f364f84050656a767d7cca8
0272c49cc7f71b8bc46503ae396e4f3e11f843cf
192383 F20110112_AABCAP owings_r_Page_054.jpg
1070d5831c1bdad447f1ce84acec22d9
7551d771da06728fd638268e4f20a80a07408c0b
113306 F20110112_AABBVJ owings_r_Page_129.jp2
80ade2d17bd2b4397c57430b18801a08
124dcdc39290b58b9e6ab5c4656fb94e5f2435a9
104184 F20110112_AABBUV owings_r_Page_071.QC.jpg
c6a6a930a5043b258b0e1e83e11c7ccb
5b93082bc4e0889b70ba236e8b396b04c38899f8
201563 F20110112_AABCBE owings_r_Page_082.jpg
0259568f869d20b033360d492b412106
91364a6e70983a063fd2de47eab6066c6d4ad148
102093 F20110112_AABDEH owings_r_Page_141.QC.jpg
2abb84331824caf67f533eb99072859e
e8806d53e476ac958e997dfa65f336f7344e34d0
72172 F20110112_AABDDS owings_r_Page_063.QC.jpg
ef3a7c1bade4cd0902b51925348391fd
33f2a8b1039353b9553592bbd6a07ccd33a1959f
1584 F20110112_AABCZA owings_r_Page_188.txt
c08c4244433e06863c3ec9cae9c5115b
3c62bec29b98e643dc18076a77d200afacbe50da
1958 F20110112_AABCYM owings_r_Page_169.txt
cf0b1bca0b0c5a7c996b474520d03e95
9d9ddd09aeb4808e2cdcff3ec36dc9561c9d5993
F20110112_AABCXY owings_r_Page_144.txt
e76b3180ac6cd4ee2458aacc724e64b0
a0a6fea213bb24c27e5255e71679f945e82c94a5
136989 F20110112_AABCAQ owings_r_Page_056.jpg
fd9fa142ecd25a375fab78b9fc253f37
a54d1161b0d8e6502362ddcdb7cbaa08b7e78d28
1732 F20110112_AABBVK owings_r_Page_189.txt
0b3059b33861239af56825b960a763bc
6b0deb1cad30b7eb49cf8c71af04aa5ef2bbac2a
1051975 F20110112_AABBUW owings_r_Page_028.jp2
61ebcd895d123dfe159f76a7149b0fe4
b027838eec71da7a46a77c793e323a9af922bcd0
191317 F20110112_AABCBF owings_r_Page_083.jpg
30f14a0f547b267816e309929f768192
4448dad86cce6532458545ecd5919bb6e18838cf
19808 F20110112_AABDEI owings_r_Page_125thm.jpg
beacb95b3a9310c860c80c038dc54b78
de4a570d63e1bf5e604dc3a0dea0c779c0e82fd0
47216 F20110112_AABDDT owings_r_Page_110thm.jpg
8b2d27b755db47baaea5967913182c0a
61b5c84985ca33fa1f2249573da6c6d48a43e7f1
1362 F20110112_AABCZB owings_r_Page_190.txt
f1b54c0871e9592c3e6766bee0e83426
d55c3fa696457abdfdfff06cf2e2febff0320725
1869 F20110112_AABCYN owings_r_Page_170.txt
3963f1db244626ab23ecb8073fc5fa23
b1ad674e52e155ab268161db96c31e860925de1a
2001 F20110112_AABCXZ owings_r_Page_145.txt
8aed97df4cca6fb5875a3a2ee1b302c1
13e6f535f8907af589b4815a34de6d02b008e8d2
214244 F20110112_AABCAR owings_r_Page_057.jpg
ede5e7f5a8bde686bfe58aa8c7475b78
9fea37c5ffe5be13e241260bc71eb3f79bb5a8cc
937 F20110112_AABBVL owings_r_Page_089.txt
e8c0e7cfb4e11cee55a91eb8c58da065
e657476d0f1649ec58469fccde9c4f1339ce3417
1780 F20110112_AABBUX owings_r_Page_046.txt
330141b819d56a2fc832396f229d4be4
986af523028bc7fdb2159110e779730564405b89
213292 F20110112_AABCBG owings_r_Page_087.jpg
10d717c5b7088107f01c0ded25e542cd
2a5323eb7c835b3340c4e2ed8157c3365b88245b
81845 F20110112_AABDDU owings_r_Page_047.QC.jpg
57b6c2ab32658c75e319363207d62cce
917ae134e765577f48f19a825cf101bd32490693
2103 F20110112_AABCZC owings_r_Page_191.txt
d8678fe419b4747324e5101e69c2f2ce
88fbb3f7d5bfb19cfb7c389f5b7d23010567091f
1847 F20110112_AABCYO owings_r_Page_171.txt
7f3cba1d344a8044e607045eadf1c99e
da5cc7b9e106601e3255c5bf9703a9b8dddbfbb3
80828 F20110112_AABBWA owings_r_Page_134.QC.jpg
7add26c0b1261188e8b89a772063b32f
2db38d2f0c4cc09b993d95f0733f6e8e186dc66d
192981 F20110112_AABCAS owings_r_Page_063.jpg
f8c306cd9625f68e28d5e5209963141e
c50b71243ade3cf0f87a6ba88ebee9ad7bb55b62
1286 F20110112_AABBVM owings_r_Page_052.txt
4cef29e6b4a47589a1c7e0c6e4e7f9be
4386ba681b1851edd3a8525fd6fc21838b9cf490
22111 F20110112_AABBUY owings_r_Page_032thm.jpg
8d443f726f97850d9704ad3eb85a2908
a75b7cb1f8582717b8ccb5529ef55152b621f0d8
21741 F20110112_AABDEJ owings_r_Page_188thm.jpg
7916529368d5852cd2f37c6888e05eac
1ad113566cd65c1b23f1acf7d44da6fd0b3ac518
108992 F20110112_AABDDV owings_r_Page_007.QC.jpg
5bcfe2345051033f5554e4e1d63ee471
37a0bc56636fd188a5d17124a080f0f71b77bb4a
959 F20110112_AABCZD owings_r_Page_192.txt
6fd11159ac13884c9d0e877f9d5af6bf
eb215cf97cfb8b373611b0997f7c4ad4dfee11fa
1641 F20110112_AABCYP owings_r_Page_172.txt
761419604888bc84710df64285ca9b1a
4a82919c4d33f80be6756fa9bfde41c4c555ce4c
F20110112_AABBWB owings_r_Page_173.tif
116cdbe421d47a3365823b3b6bb19467
18ebdbb0755e106dcefe16976358d69bfd3671ae
182825 F20110112_AABCAT owings_r_Page_064.jpg
464230267c99bc60a9c725cd99bc9dec
a99bc554621c57a999f90b777e69ffc31746f5b7
1980 F20110112_AABBVN owings_r_Page_121.txt
aa2f6181e640071a32498254dda47412
3dae9670757d6b3068ee0a65057a44190e09d47d
1051949 F20110112_AABBUZ owings_r_Page_107.jp2
b44c0ac5447faac3b8f115f8b4ad9224
28628efdfaae688b777c3869a20dc788d204be1a
230231 F20110112_AABCBH owings_r_Page_088.jpg
001f118105b141f198e9400a66c59a31
9772797083334c8f90a26df24b6c3dde860ae524
79552 F20110112_AABDEK owings_r_Page_090.QC.jpg
0899afda31a4afc624024c5226243853
1f664addce7b89ac6dd65a610906f22409dbdfc9
96643 F20110112_AABDDW owings_r_Page_109.QC.jpg
8ae1afcdbf9f7190fd6f1e1115703983
e7f828b1a77cb5cf3dc30c4256de2b5c33b6a913
F20110112_AABCZE owings_r_Page_193.txt
e915f1b92f1fcf2b46205366971308a6
32f81887326d1780c9400b195567afedbf2e01f3
F20110112_AABCYQ owings_r_Page_173.txt
54795aa71bb1f7cd813b4f0a1e48e276
7555cd4a2b9c43119b35c1696ce832024b618303
F20110112_AABBWC owings_r_Page_061.tif
9eda33f447e935d77d75c2ad5a023467
cfb21ef113053925e3da45952423701dbb93be83
223401 F20110112_AABCAU owings_r_Page_065.jpg
348cf962d7c63b017d719418fcc28063
dbcbae22469552328b8573dd736e27d88314c43d
F20110112_AABBVO owings_r_Page_008.tif
e608b84095ffa8134f727a5e7e141177
e3e8d11c8cbe12d2fab06370875f68b8b5e82ccd
242394 F20110112_AABCBI owings_r_Page_089.jpg
76f73cf00ee62b04afc12554bd812977
5c9b435594aa3b0c89c13ec9429d0b1425bebe9e
99376 F20110112_AABDFA owings_r_Page_106.QC.jpg
2594299a354ef3134cc5a3dfe839f8b3
65d4cb45d334e67ff88c25a5b5d7c76568102aff
22406 F20110112_AABDEL owings_r_Page_029thm.jpg
180bdd4b2606876a88b92fd9ab99556e
5ae9add50673d959cf7dff4a67df6dd91049a50c
46606 F20110112_AABDDX owings_r_Page_182thm.jpg
2bd26970a7e4f81a3c0dd9fd34db8133
ab48a323597761514d43f4d2112e353121bbd4e5
1575 F20110112_AABCZF owings_r_Page_195.txt
4e164e5bc4ea5f3ec45f4edd0a2d7648
bd8176a16f52db5a1cdb0ac60adb6dcd6caf5304
F20110112_AABCYR owings_r_Page_176.txt
8ed46e6458fc95052f0cbb255f3a2917
534dde7041fd85af44d0c6d181989b58546c9d0c
21851 F20110112_AABBWD owings_r_Page_152thm.jpg
2a77abaebdcc44295e1ec4f7d4a256f3
6d366765ca1f1a28125888438cc731079391a4a4
243595 F20110112_AABCAV owings_r_Page_067.jpg
2170ec29086eae7d11e428767a228d63
663066d466f5ba3a9f876a242d5a325f5cd0429b
F20110112_AABBVP owings_r_Page_125.tif
41015d11def425917cfb7d4ac981abd8
2a41a1b308f29345c54a7f67a0e6298dbce0df3e
212576 F20110112_AABCBJ owings_r_Page_090.jpg
1410d533cdf059e19997995ceda9cdec
9eecfc6b9dfaa29ad4178c916277aac64cb871a6
79892 F20110112_AABDFB owings_r_Page_005.QC.jpg
352b64a61c7342017b90702e84cf540b
180447b76406972c12a47e98e8b9aab6d2334a0f
44834 F20110112_AABDEM owings_r_Page_159thm.jpg
66445f67b9271e3e43115633550c3296
56f5d5bfd14d1782fff31d8fa7fcfdb9bf4b49f5
81177 F20110112_AABDDY owings_r_Page_030.QC.jpg
278d4f178887b92c883c661d588849b6
c60dbbb517fbc50831ee2c5084e6f22f9a624a48
1504 F20110112_AABCZG owings_r_Page_196.txt
15d914c2b326f92640153021766ec650
32bca66f9d9d5f5094bb18508cb39562c393cb4f
1433 F20110112_AABCYS owings_r_Page_177.txt
b46ba2fa90ca15d4b3c07f153c0c2e18
d63dfd363e1e1bbdca27cf70ab5a8a68b7087097
1891 F20110112_AABBWE owings_r_Page_077.txt
ec16fca7e4d47e7638ab6562c94f4654
b98a01b8e6177ff715c593a1a45b08c6acf76efe
217555 F20110112_AABCAW owings_r_Page_068.jpg
21c912e4b06b4a871dda1e45ebc36e10
7459b6da15c7240cd967a37e2347d44d369ddb46
25266 F20110112_AABBVQ owings_r_Page_034thm.jpg
e21e9bcb35974a2c7542e8772698a604
5ec9d892fd9c8c854bfead6e6636803416372479
207851 F20110112_AABCBK owings_r_Page_095.jpg
7ff75d365a75639a524a0c21f729c2b7
ddb0075484bfb452ff4b3df528b0c2a38bd8f203
70158 F20110112_AABDFC owings_r_Page_054.QC.jpg
6c416cf77be860c814feb57845dc0021
04c85da6453809edfd34e8b502d0f4897f3983b0
25134 F20110112_AABDEN owings_r_Page_121thm.jpg
e36aeca595b85aefb17a1aeebe69f08e
c1e872c5b7b9569a4484f4fef4baf43e25c7b852
136542 F20110112_AABDDZ owings_r_Page_013.QC.jpg
da6ce249a26949b63736b02160f0b05b
b7861dcedc6498fab8576855cb5f049031114dba
1727 F20110112_AABCZH owings_r_Page_197.txt
f7b0f972f42ab9978fb8250df7aa92bf
9b3195d8186ea681d05ad5356146e588499e5730
528 F20110112_AABCYT owings_r_Page_178.txt
6540ec53f8b0e44170c95683dee6a932
3402ad0df6f11b2457dfd02772fb8dd1eaaa5650
F20110112_AABBWF owings_r_Page_151.tif
51ab8c62d83a44f2a738941f49fd9e11
89d679b2611ce46197ec8c65225eaa379b4ed3b1
218077 F20110112_AABCAX owings_r_Page_069.jpg
d918db09537efb1d0d6bc8ad58260549
489830dcc0c8d81d5cd98f85f69b20499eb7bba0
43558 F20110112_AABBVR owings_r_Page_079thm.jpg
063819fa3c42c019f95fd697f0f94c6c
810161eb973eeeee0b1c79b553248a5f0ac0095d
193450 F20110112_AABCCA owings_r_Page_117.jpg
24264536adf403d4c3e2adf0226f84f6
7b977606d5351a05ab4dc77377a7de90c92798fe
112872 F20110112_AABCBL owings_r_Page_096.jpg
ffa3ef935902e0821fd21168ce1783c0
25f512e7906f058ce6309b5ce54567f053bad4f1
77437 F20110112_AABDFD owings_r_Page_022.QC.jpg
8ae9163df00cd4d5d576908a491a493e
90f43f77ba1f460cb140e3db9720acb88d2a234a
92506 F20110112_AABDEO owings_r_Page_062.QC.jpg
53618ee9381cc5d932019fdd76965ce4
dc98c45d8b995d74337816d402e13f91596b2b9c
2008 F20110112_AABCZI owings_r_Page_199.txt
f71088e3730918b0752c149d3451c812
4b3cd9a086231427f9c604fe692e36c366fa42a5
913 F20110112_AABCYU owings_r_Page_179.txt
5e5de199fc1ac537ae5a0ed5f0bbb760
bfda1c9b34c1059b3242ec02f85a623d21383a24
99526 F20110112_AABBWG owings_r_Page_065.QC.jpg
88ad8b8033947299e2f3688104f4b45c
11fa449a455fabf07b733e94d1e3f046b693d63d
216741 F20110112_AABCAY owings_r_Page_070.jpg
111b4c677f03a58e578fa2042b73b1ea
fca1155ce2fb52b52550088a6609bcfb2024ed69
47532 F20110112_AABBVS owings_r_Page_032.pro
57e5233803ed1517d12c286cf3f0e153
182df2efaeb9aaeda1ba8896f77682dc31942a8d
213846 F20110112_AABCCB owings_r_Page_118.jpg
b541b6f99b68fb72b2e94ea7129ffcb3
23d61e7a9fe099a68d2db8dc3a6c71b56c175d5e
179557 F20110112_AABCBM owings_r_Page_097.jpg
0f667dc7ebc9143ec24b57cda6021618
cdf6baec1fa59f228aa0de1153fb57748864f6c8
48252 F20110112_AABDFE owings_r_Page_136thm.jpg
36f54cd65e8d7531d1f2d5d602291a7e
9c0c63adf46da553a7f09ab500a1e7a8e922a83f
46323 F20110112_AABDEP owings_r_Page_140thm.jpg
ae5c1e7df63fed165bab0b0709510a80
b9c1fd9d05529d73d84e0754535a5e3b83e90b2a
2516 F20110112_AABCZJ owings_r_Page_201.txt
fc9a2b1767f7fce9c0d8de7b47e7ec25
3a2f2aaea3b56466e244ecf35f1c50c7f605803c
799 F20110112_AABCYV owings_r_Page_181.txt
dd6f7022cf796636012a7f9569998c27
125adfdf8896c374fa681ad50c0a5c633e1c83d1
175801 F20110112_AABBWH owings_r_Page_196.jpg
fd935583c345d02a36c54c76cee0de73
89d6885c45849856069ec89b9b67bcf0efbf210b
206642 F20110112_AABCAZ owings_r_Page_072.jpg
7a7efe6dce9a5b8427e585fb5d14d03c
4a51dc0041665979f2f0a83e1747314102be9765
F20110112_AABBVT owings_r_Page_052.tif
ca48b4d8f9f7d4f404e95ace0d98f856
814b8657cb8e7eca511b5d7ceea83fac36c21092
209327 F20110112_AABCCC owings_r_Page_121.jpg
491cf1ebdd53d5c42bbe8e87e622ee0a
cd5f4716d224db9157bef92f46473af5c2209ce1
207075 F20110112_AABCBN owings_r_Page_099.jpg
30e754c6770ee0eb77f1c1c4df47f880
63fc8b8aa33edfb5c5625f7c142a2989473ef55b
23623 F20110112_AABDFF owings_r_Page_168thm.jpg
687ce0bd5e73ca15e763b0855165595a
0a8be8c28e72ddc2b6861a2f8fa51cb1ee870c4d
95911 F20110112_AABDEQ owings_r_Page_074.QC.jpg
c4a39d8a11fd7cefe2067d03ac331970
8854d8b193ddf904da8505f3ad90c5ad485c276c
2528 F20110112_AABCZK owings_r_Page_203.txt
63ec0ce9f77dc9a79d127b6cd060b21b
638a8b9499605124f99af157f103abd8e9b8b064
1351 F20110112_AABCYW owings_r_Page_184.txt
4c66792d9a33ac946b23f2da8717264e
72c6b6f0d8004cb5945ac5c30c15e604242382ed
47410 F20110112_AABBWI owings_r_Page_041thm.jpg
d2716156ffa3a116189c5f53d06e198d
e6f07e29569dcac239efc11a9214aeeee35ddf5d
189921 F20110112_AABBVU owings_r_Page_165.jpg
03dfb8d0467d3ef85fc74b25dbba909e
62cc5fe8438d837f01096d9e6b0e5c026551770b
219807 F20110112_AABCCD owings_r_Page_122.jpg
60eaaaa162556527ac24c893a3aa3ec5
16ffbafec5878249cc10318cd23c6d1d8a676bf6
185002 F20110112_AABCBO owings_r_Page_100.jpg
860942483760bf9bdbf96aa03c9d1155
9fe0f77369ce5aa365c9c7c908fab09406fdac33
48054 F20110112_AABDFG owings_r_Page_186thm.jpg
d9be0fb817f819310c17a7bdd1f083e2
0ea41f29c1adc74ce5c562ecceff120ec1e040c8
81667 F20110112_AABDER owings_r_Page_200.QC.jpg
3500b3f4c2aa0601e8f24e8d3c962e7d
4988a6632c7a6c66a3b16e93787e669cfbfccce7
2605 F20110112_AABCZL owings_r_Page_206.txt
87f0f5f10be1392b1d444b1e312541f2
80084a1f417e49b2800c4817cb7dab61b2948206
500 F20110112_AABCYX owings_r_Page_185.txt
952447e86f94f638e18d2f82cea898a6
330761f5cfdc45b80fc93d89054d778ec98eeba3
F20110112_AABBWJ owings_r_Page_068.tif
bca1d97df06391aa078d18db9dd07615
8b6f7ad6642e1e2991344d58e1ee5e8c040e37f3
1051953 F20110112_AABBVV owings_r_Page_077.jp2
92d5c62d2454e0d1172f992210a99df4
c9b28483fc0eadc3fa1c53693494445a85529788
203996 F20110112_AABCCE owings_r_Page_123.jpg
3ad5e279e1835bbf9df59812ab7009e6
d92f39038e03972c650f964b2d283e44cbaabc5f
238313 F20110112_AABCBP owings_r_Page_102.jpg
48237cc5dec24f52693c9685f5d2e9b6
899d311bf7bfabcc6af0d7ea4cdc7b24d1b8bce5
97924 F20110112_AABDFH owings_r_Page_101.QC.jpg
3f7a326f7f5c4d63294fac6c787da7e7
ea1134c3d52a31e6bd7d21d500a8a189d9c98a91
48669 F20110112_AABDES owings_r_Page_104thm.jpg
76d1f8741a7a53a8c9a9ace55c3a8dc6
f1d80931973fb8dc805dc858ce5b15350e316f86
2399 F20110112_AABCZM owings_r_Page_207.txt
156daf473e4b96bf2298c119c1380a69
172e1b8364b3eacbeba5a0728bad260a4be1f7c1
2564 F20110112_AABCYY owings_r_Page_186.txt
308ffcf9800a44eadfecdcc233c84949
435b8e77946d8f69b00a7b0de7c7182b4cdbf009
49982 F20110112_AABBWK owings_r_Page_071thm.jpg
9168315781aa242385d95c2c89696037
ec95d35c2d144ab5675b7209966869a58897884d
43388 F20110112_AABBVW owings_r_Page_009.QC.jpg
71e2e0cde623f86d7ae1a7f3c06ccd2c
7453256f28f173e89124644a0e4305ac8df69499
174652 F20110112_AABCCF owings_r_Page_125.jpg
3187bd28efd33930370faa951dece8a1
7dba65a2b3ad2a07130cbcd71afc4c9d0faee61a
252612 F20110112_AABCBQ owings_r_Page_103.jpg
d1e6ede562a216bcfa53f45b1d3cb332
2911c405a21f6393e130102e7888df82226803dd
83514 F20110112_AABDFI owings_r_Page_195.QC.jpg
faabd080a55b5370a9607fd8f0af76f2
9f16c57707efd8caa0399002391548a1c6cd19e3
17285 F20110112_AABDET owings_r_Page_190thm.jpg
9c866e23e514b791a61a9dbc461c3324
86c7577dac487c7e4aa1e7b6326a619c151bb416
821 F20110112_AABCZN owings_r_Page_210.txt
dca6b3f26a0d34c91dd101d4c5227e41
56ef7f7bbb2a71c9f343b4277756944d5c74995d
1585 F20110112_AABCYZ owings_r_Page_187.txt
b41b63187a926726a61fe8b56d86e264
eab21a7fb4c565709b2dc4d99c35892eb7bdb912
79138 F20110112_AABBWL owings_r_Page_059.QC.jpg
d6e1e3b792616f424064ccc7f73eed5e
6aeb04397a0d63146c90ee28e6ae6eedf853f511
74557 F20110112_AABBVX owings_r_Page_153.QC.jpg
4ea28e8753559285be83bc337ee460a1
6f48d2ec287b4f49052d3f69598d5de4cfb48b35
214775 F20110112_AABCCG owings_r_Page_126.jpg
d6092998e6aedd23b5ae555b3e6f576a
fa3f1254874f695eaafbd4af5b72c4f7227bcb97
225863 F20110112_AABCBR owings_r_Page_104.jpg
d14f83f3a27908ef333186efcc5f9942
158ace7229f6d60af087b4963ef475cc05d16819
68105 F20110112_AABDFJ owings_r_Page_162.QC.jpg
b865efdf4831a8fd47905421b54cbdac
fd3a433d0f51cb111914a5083208d79b4b248917
94424 F20110112_AABDEU owings_r_Page_111.QC.jpg
3736d1bb1726fc051991b82a5ce15e7e
f40663fc94323d5f24dc5896684b4d003be6c20b
7545169 F20110112_AABCZO owings_r.pdf
31478b753496976d158790df3003903c
9de78cd196804005837a7dcf04e2999e2a59ee9a
49599 F20110112_AABBWM owings_r_Page_169.pro
b55f9434218bbf733563f069408625db
0651b6cc0f802a253d755904e42b5f0c8c68d63b
723705 F20110112_AABBVY owings_r_Page_134.jp2
c12cbb055dbe0eb2c2b48af088f2dd93
0308e58e36351d5f6914f6b88268ad4dc81ce69c
197000 F20110112_AABCCH owings_r_Page_128.jpg
d33621c09948efbd19caa2f4f5484ae5
303d6bcd3cacbcae10e0aa902f2340ffe1979f74
24840 F20110112_AABBXA owings_r_Page_059thm.jpg
ce56443906b7c6b242bf790cae1f8513
e89e9bfaee4fe8a51081c1d093459880ce2ac115
238720 F20110112_AABCBS owings_r_Page_105.jpg
f45ae33644e32c452b6812d73a7c67de
6985c73b15f62c07bf833a3587c7e95c022fc964
97949 F20110112_AABDEV owings_r_Page_006.QC.jpg
f43d7ed1f634099bd91bf955e7b8ee32
1b51c5cd47d660efd31d151a9af78420af98f4d9
643880 F20110112_AABCZP polaronmovie1.avi
9cf98dc54dd9ac15f2539286199f8539
a6359d6e596d5d0e5945ba708ffb068accf575d0
1051928 F20110112_AABBWN owings_r_Page_112.jp2
1db978379e6b3bfa19df3de1f3395ccc
04cef06e7dcfb75640fb6935c8625364df4b7e7f
F20110112_AABBVZ owings_r_Page_102.tif
9c0ff93d1624c0096c94b7fac0223dcd
d6f89333a1c9bbd4f1da54697719c78c1ca13d2e
F20110112_AABBXB owings_r_Page_077.tif
767d3520b93d2ad9d699a21535c0b295
468e1524b520cc2eab696ae097be635ec0654d22
208381 F20110112_AABCBT owings_r_Page_108.jpg
9175cde2b384b5b17e16f31320382e47
58ff703cdf74639472aeaa54c50d9d678bd94661
48983 F20110112_AABDFK owings_r_Page_074thm.jpg
a0e3b9aa3d9eacab2505102a85c1ae90
b628007ef10a1ab19f9111d34e91060582369cec
94222 F20110112_AABDEW owings_r_Page_080.QC.jpg
0e290d2fba8d58902fda00925bed1ec2
14ce30a2f0cc47c610ec47a802eb696ac38d40c1
50750 F20110112_AABCZQ owings_r_Page_190.QC.jpg
0b215c11416bbe39640024c2f24f08e8
cc51ccd8621afba579859eac0e71d7f975fdd910
223646 F20110112_AABCCI owings_r_Page_129.jpg
8c784e8ae1ddd3d9118fe421fef3f73c
06da3068231df06f5f0c37518a832e3afb3c1e91
38661 F20110112_AABBXC owings_r_Page_136.pro
548667ad4b64c8a2b382f2c54adec48c
f72d1095ae89677e0447132732a855a302718769
223036 F20110112_AABCBU owings_r_Page_109.jpg
4884c1230a719027473c96ac0ab98782
864827ac5ae8eda90c0e83786190f0b52dabe586
26914 F20110112_AABBWO owings_r_Page_079.pro
acdbed51025c649f7d52cdca97397105
0e4fb701a01633da9dda85fd617f279b8c84cba3
3263 F20110112_AABDGA owings_r_Page_002thm.jpg
20a216d1ba25848b1dd31608201d571d
1667f0f531861bd126355400fc9156782e538b22
25418 F20110112_AABDFL owings_r_Page_144thm.jpg
20acccaad314d13b27f2a6831a99a4f8
26ae713cd4854eac792064fb2821c594747b6d67
19821 F20110112_AABDEX owings_r_Page_018thm.jpg
5edac73565f882653dfb8fb38cdbe099
83b4206bffd4549f600b0dc2af17bc9d8116545c
25437 F20110112_AABCZR owings_r_Page_061thm.jpg
9ac7ac85c794983517b337e9b0aa7b90
ec8758229e1091d8f309b90f3a686912df51987b
179283 F20110112_AABCCJ owings_r_Page_130.jpg
6c1a8cd9afbd43a79fd04292acd41e6f
949df4c3b3dc57528b5057a4c14fe1d77d0ac97d
111584 F20110112_AABBXD owings_r_Page_090.jp2
6c2f0319a6c0c88f9dbb0200ab2bf4bc
242c7d20ef50861ca715979623ccc810c70011d5
204029 F20110112_AABCBV owings_r_Page_110.jpg
705746bc3fcf719c5e05d1a46c46f74e
c648a56d3f8285ea5b2db13e1274de50a30b1cf1
76828 F20110112_AABBWP owings_r_Page_026.QC.jpg
17f748cdf60ab482e606e6e66a4f680e
928e9dc9371cb74178d0c05d884e7f24912bb108
24858 F20110112_AABDGB owings_r_Page_005thm.jpg
bbad6bfb7540db1035da1f71fead3b5c
d711c57b4867f816b80680a63f2cb25e38b0f62e
67575 F20110112_AABDFM owings_r_Page_056.QC.jpg
de8374a37e7f5c3a3dffccfe3637fc0d
b91741d1ae91b827f84d4ca21d8e99ce622a1bac
90795 F20110112_AABDEY owings_r_Page_176.QC.jpg
7fb8990e61ffb1b7d0d1b20c16a1f23a
8f5c22cdb2c2d852d1153d353ddd85f7bba1499f
12831 F20110112_AABCZS owings_r_Page_147.QC.jpg
ed73eb72bc9ee83bddcb2f3824578732
d7ad9bd2816cdcbcb0351482d5fb780f13bffb93
168868 F20110112_AABCCK owings_r_Page_134.jpg
3d53ab41a480a85537fe85f40928c97e
c36eff294ef5875dbba738332a81372b61824702
85961 F20110112_AABBXE owings_r_Page_137.QC.jpg
1655b10d3aba3f9be3cb22d79ff27a44
6a40559b61b9059dc2f8e138976304f708fcb875
225713 F20110112_AABCBW owings_r_Page_111.jpg
109a0d7c3e894d5b0f6fc7d00d7212ec
7a6e02e4d81f9f8e123e02985a27a7b5e08fb6a2
212223 F20110112_AABBWQ owings_r_Page_033.jpg
344899e64c5a37a3e1b997684a0c6bda
ee3d37108504add7f29006329702fc153602d2b5
49644 F20110112_AABDGC owings_r_Page_007thm.jpg
ed11e00d1adb95519dbb5ca710d42688
1b4d85bd404f7dcb509d684803a1c709a53f0677
48192 F20110112_AABDFN owings_r_Page_072thm.jpg
7e1be83094b00c825847d8c0169a3aeb
2808309b53145b9c93bea263a6e71128ac658ef1
31434 F20110112_AABDEZ owings_r_Page_024.QC.jpg
2059865142771833879966d9d683b54a
6939fd050c9f45597e4e11bb1842c5df2f24099d
96906 F20110112_AABCZT owings_r_Page_091.QC.jpg
b700abc71236169507cffe7a29097762
06cd65c495a07c0d6d1612499219750e794482c4
178446 F20110112_AABCDA owings_r_Page_157.jpg
e4933516e53f4802c70965d5e7caae65
8c82d905c82a679f004cd3aa9c67ee070ce4f5d4
190741 F20110112_AABCCL owings_r_Page_137.jpg
a12cd93434ef73c7c3208617459c339b
a6544d04f87ea85904db44ea9423e8e6fe9e12bb
45716 F20110112_AABBXF owings_r_Page_150.pro
29b49a80e66fd9b338c54d15e19b2879
5e51c792531131519883efc5d941671520e6b1e2
228780 F20110112_AABCBX owings_r_Page_112.jpg
4bb9c085c1d2775b227e02b0b9f07e90
9395f5ecc3d201f79eae78288bab3af2bf2feb46
46594 F20110112_AABBWR owings_r_Page_167.pro
8e0337e2199b591170fcd9314997bfe7
ab64ef2db5b399b4e4175725ea7437ff2d37ca76
116215 F20110112_AABDGD owings_r_Page_008.QC.jpg
205d2400bd6d4845c0559edaa4f0d278
0ea1323efd302d0f8db914fca7a8680a3275cc3d
23816 F20110112_AABDFO owings_r_Page_146thm.jpg
3d89aec658d07a95a851603f7cd68667
5e2892038a470511e05b6637c406f220ea62a1ec
78414 F20110112_AABCZU owings_r_Page_121.QC.jpg
491bd4cffa9a74831783d8e4a3273f99
5a482939ccafb37899b396938e432abcc87ac80d
194136 F20110112_AABCDB owings_r_Page_159.jpg
d99e818962d7dcbeb662b916d1a0e48e
73139f1fb7547c6d904411e6dcaa62bae6531248
218063 F20110112_AABCCM owings_r_Page_139.jpg
382f80576b737b239fe45b973eb91ee8
58047a301ca194ecfc5a7bf925297d089190367b
27149 F20110112_AABBXG owings_r_Page_209.pro
d1ceb2eef2d348f2516fd843dd146fb6
51888fc5cff905cf5236c2f530beb891b81e8a06
241457 F20110112_AABCBY owings_r_Page_113.jpg
15f00510175c51f77d9d229cf6be9793
2f34c395ce52e49fe23cee2459fbd558d63d6c64
1051980 F20110112_AABBWS owings_r_Page_175.jp2
09a6409c440469f3cb7774c09179e74c
dbc40de22262da0fef12956ce8f2f5b7523bb933
57301 F20110112_AABDGE owings_r_Page_013thm.jpg
5dc53eaedb32bf967783b4cdeb2fd4fa
cde793d43b91fd8b7d0fd1f646e3206acbb3ec3f
64345 F20110112_AABDFP owings_r_Page_097.QC.jpg
d5f73e19a0aa617e3e5ff6a108700a91
90bff0f7944d78ab7624668f6d899079d936d44d
79988 F20110112_AABCZV owings_r_Page_126.QC.jpg
756d1b5df2dc6d484ae91d89d56c2c39
51c4ff78986f76f12bb03d82aaad071e408755d6
201228 F20110112_AABCDC owings_r_Page_160.jpg
c619aa3de9d789d0fbe01b7f98208e1d
cab72d8352dc7f6b8bcbc3c2dfa86e781ec74b2f
256397 F20110112_AABCCN owings_r_Page_141.jpg
8aa28836e6b29a8e13456cb98bfb33cb
46b8a48b22120c45947b621485c9ca889ea536eb
49858 F20110112_AABBXH owings_r_Page_192thm.jpg
9782e65ed3ad3fef3c5980b3ad40e6ae
4747a3f03f171fe89d739c375d03c21aba13ef1a
218485 F20110112_AABCBZ owings_r_Page_115.jpg
9ad05a8e1070896fe05197fab7586b07
4608b8e7a2125b4b366dd1c84e3be22e65c61804
233639 F20110112_AABBWT owings_r_Page_092.jpg
470340085f63f182f592e6661d048aa3
592a812f2b7e8be92f079b0890f066b8cde4df93
96021 F20110112_AABDGF owings_r_Page_021.QC.jpg
de9c21e7a4e99a8b7430630162cfdc1e
812dcf16e6affbf5b84427cd284d4201ed35836c
35403 F20110112_AABDFQ owings_r_Page_011thm.jpg
65cd0cfb6dfbf7ff885188943a733ffb
669af7cfca247adaff55553cc9b4b9f1bd09ce67
79022 F20110112_AABCZW owings_r_Page_098.QC.jpg
5e0950fb3febcdad83b8d077244dcc98
c839e24505c31f1f903c87375b4a9f4d694862a5
204792 F20110112_AABCDD owings_r_Page_161.jpg
6888ed16a3543261089422bee856ae29
ddbf645f6fbac1635595fddedb285eb040917cf1
208442 F20110112_AABCCO owings_r_Page_142.jpg
83a6eb053f49d9f46a94d6eca23805df
1281d82485b3ed7e655cdd53e77c7e13b672039a
95272 F20110112_AABBXI owings_r_Page_185.QC.jpg
285d114d929eeec0c6e3fcfce901c556
864ccf1777e848691c8a5f8d85a2979ad9cf5918
2178 F20110112_AABBWU owings_r_Page_035.txt
526204e4e60d1b3aadee7603541eec3c
d2e416d5587131bbe103cf675d38bddf5ad21844
79669 F20110112_AABDGG owings_r_Page_023.QC.jpg
d9c8f797b8cd8d72a4fb88d416d68771
250ac91db31e707953e94e455cb4c64e13f80f84
48846 F20110112_AABDFR owings_r_Page_101thm.jpg
3329fc6972aea377edb1f5ae927cb43a
8549bd58763b2eb33b29c7c2a830867341a35e53
72330 F20110112_AABCZX owings_r_Page_128.QC.jpg
5debe87c30d41ae492c608dfbef52a7a
ea701a998de61b6f4959305d99bdacb29b5b2fbd
190140 F20110112_AABCDE owings_r_Page_162.jpg
ea032ad44fda95b4880853a455765546
3c88781b4d3da06d14191fb8a7d680481a03fa5e
218323 F20110112_AABCCP owings_r_Page_143.jpg
15ee6cf7e8a8c8e07f88ccc497a49e61
1bf4cacdce62757408bfe0eb7fe244e6c1e85c03
216556 F20110112_AABBXJ owings_r_Page_038.jpg
fa69cf388a6157effe9f18ef5a103c18
ebfd3a174d1a9c37d467f1edd2a50ec5dd037f4d
245856 F20110112_AABBWV owings_r_Page_107.jpg
f68aa60ece98126906e5ff7b54ad6a7d
6a0ca0d97bdcabaa73caae61796adb678219b16b
68872 F20110112_AABDGH owings_r_Page_025.QC.jpg
8b95025d50c28afa79c365d4cd2ddab5
98a5fb05450c108a901e1c4a0dba2ce6745f419d
44201 F20110112_AABDFS owings_r_Page_046thm.jpg
2f7c8b5d9d0a311ba6b82276495b1dee
947a8ed8756142d3c63a0928400873bdd07b0366
55962 F20110112_AABCZY owings_r_Page_017thm.jpg
9f18df1d29b8af56a9af155180a63a49
4e620c85bee72579a51c5414c047fd6d88c6f90a
110216 F20110112_AABCDF owings_r_Page_164.jpg
b34075fefe4486c03e0e20369f15e97a
4119e3e355b201f633f0db5e1a663967af45d928
206073 F20110112_AABCCQ owings_r_Page_144.jpg
28786a07a8d42a000d2c6ae59dfafc1d
1e88dfcc0755295c01edb484f8352a16e72ea0ee
216481 F20110112_AABBXK owings_r_Page_039.jpg
175f2bcb51af87cdc93aa7da0bbb7d25
0c8fd4a0f5d5d1448d4af69311516e16603ac231
1894 F20110112_AABBWW owings_r_Page_148.txt
f84be4af050ac2ed2fdafb3b288b4b33
9c10f85708d4414e1bd389048f324923f5ac0ec6
24229 F20110112_AABDGI owings_r_Page_026thm.jpg
b4c9de21bd734663b7705f06af0197d4
13d65c6fab6b793ad78030fbfccc17440d8fc335
73315 F20110112_AABDFT owings_r_Page_073.QC.jpg
2e479952821e8b31b75adbef9f56fef4
762bfa8eb3a24fe58bb65f624ce671ff60544273
22196 F20110112_AABCZZ owings_r_Page_148thm.jpg
91fca94ab4c0f9cdaa4efe74d383da67
0e8c5efa92122cf616ba27f67190fc2f38d105bb
209325 F20110112_AABCDG owings_r_Page_166.jpg
ac2857441c558841c1a02489798ece49
4790aa215b6a0c1d65a5bd6b201a4459ba7657ca
209672 F20110112_AABCCR owings_r_Page_145.jpg
af96804ef3b19585deeca7865ebe2706
2ffc0eb8047d54610dc452010e2feda30e85f695
49487 F20110112_AABBXL owings_r_Page_113thm.jpg
5efbd36eeee053feea6811223d23e400
6f48c9c375417da6147d9844eebde3e1d3055e97
60083 F20110112_AABBWX owings_r_Page_209.jp2
2fe7d22a1847e5f27629f6c312a420b3
6400dfce5e745f29379af8e1a0bb50e668258d66
F20110112_AABDGJ owings_r_Page_028thm.jpg
6ee1e47b5ae584b2941c086c43ce9f8b
38160e75a97e35d0e42e42fb135ea5fd13b7123c
107908 F20110112_AABDFU owings_r_Page_044.QC.jpg
0707d3c95c88575033aa0380cfc1b0cd
4dd8e45531c4c1ea55b7c458f71143fa609007a5
185625 F20110112_AABCDH owings_r_Page_168.jpg
edeede6a149a103eab26e78a313ddd79
e5849648311a39ea73fdb8fed163e85744e1bec9
80554 F20110112_AABBYA owings_r_Page_119.QC.jpg
21cda7171faf55b7f8554c9d0f3468e6
2c0d2dcc506ad5a72888d7b76d25f204a925b556
201913 F20110112_AABCCS owings_r_Page_146.jpg
c62ea5b1ad4ac10d1493271035a818b3
5341c1adc94af08da378cece4d5620908506dcee
51145 F20110112_AABBXM owings_r_Page_142.pro
2738c10b1e56e813c1ad83a8e8a55bdb
99c90188bb47d2b5a649e20cbfa7a53f8adb956d
2420 F20110112_AABBWY owings_r_Page_200.txt
30254d3a70a4c487bf7d44f26b1f455c
d2d9f20c46e751dbf961477d9a6ce7e589a1cb0e
73792 F20110112_AABDGK owings_r_Page_029.QC.jpg
16c40cdf7a6aeca5f57e209f3e0686ea
b390197e030a58fe812676563c60a7638a61530a
50767 F20110112_AABDFV owings_r_Page_012thm.jpg
01137ac3dc64488b1ac7a0cf1a3caacd
eda8cb12e32e86efc4b2302d03d180cd1c9e1b2f
166488 F20110112_AABCDI owings_r_Page_172.jpg
b0027ee6082d6a3a71db7fe79ae7671b
f8e260c03cdde0a5ae44f7b5b247530abd3b4762
F20110112_AABBYB owings_r_Page_196.tif
02c466e06bb4fe1d9236196f2448b4df
8e3415b7e5bcbd69c0bb162a135cf5ca828428b9
192369 F20110112_AABCCT owings_r_Page_148.jpg
37656d98786b66c3176dec7eb05042c9
c9a252d5a0480d1796264503d7677ef0a771eafd
182389 F20110112_AABBXN owings_r_Page_140.jpg
77a3a866155c7dd83207620430232000
16b2f1479e78e2971eb5c6fa1d73a880365c6210
50090 F20110112_AABBWZ owings_r_Page_103thm.jpg
b498c49907b0a9209bcd5b67613a64f9
3b182a37ca051db7ac64bedc9395ae77d39b3733
F20110112_AABDFW owings_r_Page_048thm.jpg
997639e53f5dd8cccc74515488c843d8
ed1c85dea0a7cb9cc71c5ac888e4eff803d71ae7
2342 F20110112_AABBYC owings_r_Page_010.txt
564bbc858f93cdab5729feca79e9cecb
0dec2d7055828e7a9ec59c1b80082ed39ff2d9ac
221687 F20110112_AABCCU owings_r_Page_150.jpg
aaa9d0f1827c1b5c1ff4861f9e411f99
2d420ffd3b7a81f33bc22c50f4a42cbd9b85b45e
77599 F20110112_AABBXO owings_r_Page_169.QC.jpg
c4ccffda3256f6dd2b86b1126908f634
62f03dae1632546f281f1279e6666b7e97579939
66093 F20110112_AABDHA owings_r_Page_060.QC.jpg
5e0385119656aa82edd9777295d0c29f
0ce9aaa39f3e086ab0f02b926bf55a59b72a7b01
80635 F20110112_AABDGL owings_r_Page_033.QC.jpg
402dd3d78db514d06f06e4e2c182ebe1
d9ceead70d707d9a2b26fa8f3ce3a496e7f1452d
76032 F20110112_AABDFX owings_r_Page_204.QC.jpg
e64280adda961c025c0782cc54766475
651009c3ea9c7957d71a1808a3ee4605f73cc251
179734 F20110112_AABCDJ owings_r_Page_173.jpg
0d26df5d91a5c33e0034728394bd2a09
b565702cad06facd9ecd4116572cd989cf3ae13c
24611 F20110112_AABBYD owings_r_Page_169thm.jpg
6c14230f67325950d8f5ed32d967cbc9
5a11bdeeb24a06e5b3097d108139913c7a3c23ff
183528 F20110112_AABCCV owings_r_Page_151.jpg
3d060d29b3a1e23467c960e66f3dcbfd
bafe7a9ead3ebdba5ce1e79cc814e30930f2b9ab
6562 F20110112_AABBXP owings_r_Page_002.QC.jpg
5979af4772c4361495a69899b83f6646
a56e6b5bc34a3de4234ea863f4a750e00f172225
83427 F20110112_AABDHB owings_r_Page_061.QC.jpg
4d6b323675279fa1ddf6e4c936bc437d
6d1596d029b77ff4a6bfcb3fb8642b66e118e7c3
79092 F20110112_AABDGM owings_r_Page_036.QC.jpg
b1e12b6711afecd25903390a6a64ed66
46ec6f2a9da0b62a5a5fbf7f1cbc806be94af78b
F20110112_AABDFY owings_r_Page_197thm.jpg
c14bb2f08d6a508aff43ce9ba4e5c0bd
233682fc1e7679291923f971f086341d844da024
164808 F20110112_AABCDK owings_r_Page_174.jpg
ce6bccca936af9a1b64678c6e9326928
263163b78b758235fbb64ecf9f3dc3c2682e9c2c
2175 F20110112_AABBYE owings_r_Page_041.txt
2d7398284110add8185548babcaf1e5a
5e313169319aa5ff563a9d31e2e63368939d59c8
199320 F20110112_AABCCW owings_r_Page_153.jpg
8179532560f976f40bc01fc6d724e49a
784cace15f8ef279c3453e5e792ac007a7cadf6f
45399 F20110112_AABBXQ owings_r_Page_127thm.jpg
e7d1b9c4640fb4f9e3933f4df6bb8159
53af60fd223341c8497ec417318067404804bb15
48529 F20110112_AABDHC owings_r_Page_065thm.jpg
d0787bcba1a6c88db9c58aa2ad7fa917
1085748c548b6b548c1a8df8f326a2cb988a30da
47290 F20110112_AABDGN owings_r_Page_037thm.jpg
5718e4d866f3e061033c5cb3625697d8
555bbc263c0599c3837489ba248cb8e0b9c6f2ed
313397 F20110112_AABDFZ UFE0002242_00001.xml FULL
b3b2e2a618e10ca10662d9922018a1cd
a153b3a4584487428097347720c2d76b94b2fba5
90616 F20110112_AABBYF owings_r_Page_097.jp2
b8520448a01c1923b5f8969c97607ae7
dc0015b5aa03f71d56d48c951874a01f8014e24b
219384 F20110112_AABCCX owings_r_Page_154.jpg
3b8502ccceabdeebf03742a033cca8cd
18e65a2282869b9970826d8d8306ac28646f3aa5
30332 F20110112_AABBXR owings_r_Page_147.jpg
2165e3fd2a67bb9f357dcd558a368eb4
50f597b558868188e96bd9cf3145b6dd5b952525
171735 F20110112_AABCEA owings_r_Page_195.jpg
4a30ca740142356909e7add4e5c7429f
c7728e6f4018e623a6e2f1cfbc2ba9022cc6bcad
222543 F20110112_AABCDL owings_r_Page_175.jpg
2e37296711263072664b05d3be9e275c
0dfd7a3d72ef7c068e653c97b407b4097fba39a2
101414 F20110112_AABDHD owings_r_Page_067.QC.jpg
a70478e4b25e839de9393d099107e487
0ad4a91c614e5c05002ac2afe1a911771bdd0fcb
94220 F20110112_AABDGO owings_r_Page_037.QC.jpg
4820b71e087ad41d32a29722c8418e3a
ea31b078b0f0523c4c4564b1274fe70b7e37b670
33346 F20110112_AABBYG owings_r_Page_195.pro
d5260de2e7e9cd3fab358233eb3cbb37
3320eaa7a582dc47cd693571d9569dcab7b906ea
162490 F20110112_AABCCY owings_r_Page_155.jpg
47ea0b1c10659d9258cb69611eddf890
3532fa7eb6c74a514a4394ca7a4f5f0b282b7103
F20110112_AABBXS owings_r_Page_098.tif
73061a436f017e14b120fe7361db196b
8db1bee0ce2534923a217130d8722223b19e3fb6
215986 F20110112_AABCEB owings_r_Page_198.jpg
4023f2e8c6766d917eb54c057b8f9ef4
69130f514a950cfa9df3377ccbae40112e3d481a
212056 F20110112_AABCDM owings_r_Page_176.jpg
37febdc1184b094b6640598002e06c73
08bd779da1bd2c1f0e4040cb7109eb798180e1f2
47487 F20110112_AABDHE owings_r_Page_070thm.jpg
2a2c29be482181e8fd36a0f1a898c50c
a9a7c477f57b13c6396054a37669b20f57dad8f4
98705 F20110112_AABDGP owings_r_Page_038.QC.jpg
a991f79ec87bd4ece61134162e0ebe93
b0438a86b43c0002d1f2d4b733a6cdb3e85e8c89
1509 F20110112_AABBYH owings_r_Page_109.txt
25a53c8bbb035e2cab8a6a8c3b6d18a7
17e69648924b90ca23ac8c663b72950ee0d32277
205372 F20110112_AABCCZ owings_r_Page_156.jpg
d7ce198fc9adfe76b9968d8f9ceb5ab2
67d68181fc94e17bd9ad830a4b5750ea8930993e
115371 F20110112_AABBXT owings_r_Page_081.QC.jpg
5f1d56811dbb952f806fac70207dfe6a
4d223c2cf3544e096adb850cd522228d1dfb4cc3
203114 F20110112_AABCEC owings_r_Page_199.jpg
645e4b2e634f5f92def91553af4d7b4e
36f187bc8e657e6fb1fe0116cbd93fd7b1045838
170209 F20110112_AABCDN owings_r_Page_177.jpg
cad434809f71fa44db6ad62a5bdbb265
727e47e4a881d65c394c8b762aa918b1cdaa0ead
18687 F20110112_AABDHF owings_r_Page_076thm.jpg
d23f5f442e11b4456c10bb20091fd39e
9b9f49fde91ec5571ead7aa8b21013e8e27cd8e7
95363 F20110112_AABDGQ owings_r_Page_039.QC.jpg
e83ba89843eb90c3d44c852f5d31b3fa
0b624466902d82208ecb671e7e1c3fef26cd9194
77244 F20110112_AABBYI owings_r_Page_057.QC.jpg
b085b8d9a5c2ecfcc325429655d023e3
6dddf160fa22807030ab875a61584118410d2b8e
22212 F20110112_AABBXU owings_r_Page_031thm.jpg
16ce0ee62f3e0442ad3cd22cd82863bd
860bfe31f2b433653b032a397ab39dac44af6a4a
283218 F20110112_AABCED owings_r_Page_201.jpg
2b2c93a60e992db5839d918ed94effa4
5fc975b1b317d19851d9e1e3fd52e386757e395f
174915 F20110112_AABCDO owings_r_Page_180.jpg
c0ff97fa39c3151aae18f5d110fcc3c2
0ebe6a292c11abbaef162d353a2904a4b2640f7f
59976 F20110112_AABDHG owings_r_Page_076.QC.jpg
37706098891be18cbff2e6e83fb40b90
e16f7c5b70dc2ef8f380eefb986b63884eae53aa
91369 F20110112_AABDGR owings_r_Page_040.QC.jpg
f7a9cf909a28d2a543db0c24d97ca4f0
f285ff363c81a85160b994732c22acde7cfee373
116248 F20110112_AABBYJ owings_r_Page_010.QC.jpg
7592d7621c3b35b44ee752e147402e7e
a1cac8e450d38b0e3ef4c550bb0f58eb40f28950
49788 F20110112_AABBXV owings_r_Page_123.pro
c48c1a8d8c137056378d0526c3e244a4
85f03b66e42ba9b3e46bc8faef7712efeedced7f
251244 F20110112_AABCEE owings_r_Page_202.jpg
3071e502be49ce439157ffec3e0309cc
992de991183d6e4db57c1db48aa332e88513f4f3
204240 F20110112_AABCDP owings_r_Page_182.jpg
9430501f032cbed1ad4303a3117a94bb
3de96972b0a77e24a78c657be72ea23873fce10c
47627 F20110112_AABDHH owings_r_Page_080thm.jpg
222af349abf3438dd951299f6989b815
790ade8638513766a0a69b820138c39ddafe0d04
48233 F20110112_AABDGS owings_r_Page_042thm.jpg
a2131f215156f3ce2e6075f2959ebcee
03383b8e12c2f296d3dc5ebc9eeb708c42d912d0
2294 F20110112_AABBYK owings_r_Page_204.txt
bf411153bed07a33cb8646fbd5c77952
f8022678176098d26b04eedf8cc4a235bb9da660
214793 F20110112_AABBXW owings_r_Page_094.jpg
1824a82568411516e1767d92aa963497
a063b9c21692e1388c19f4a627426cea5e20316e
267884 F20110112_AABCEF owings_r_Page_205.jpg
61781a446e207a29a53c87c955875d16
1bd4af68cfaa40071b5dee5de86567c30edb3ff9
174252 F20110112_AABCDQ owings_r_Page_183.jpg
7b7a51e12246ae500b5b5a082cd3a713
7d5ccc87aca9b6d6e258791dc74a9604eb772ba5
50659 F20110112_AABDHI owings_r_Page_081thm.jpg
94152dae0ef37baa50cdc0c5f6532b49
cc68a01bbb43454a1820cef4709e1d8b58cc3733
97713 F20110112_AABDGT owings_r_Page_042.QC.jpg
f02c5af23e25ed14d3700d5258889b3c
a05a914fe1c750c1e061046c15adf0b61c8b030c
34695 F20110112_AABBYL owings_r_Page_019.pro
54b175f5423e50e15b2f6bae934a22da
903718cc496712e543f8562f46b67ebaf0c72231
22411 F20110112_AABBXX owings_r_Page_025thm.jpg
4818ca3ef91cf83044607052210c62b4
b212256cee0cba7f886c19e8615c581eb6c72aa4
272970 F20110112_AABCEG owings_r_Page_206.jpg
412fbc5709e267a5f3276188aacb73c0
0fb84264e037660ef3eecf985a80166e4f1705a6
206533 F20110112_AABCDR owings_r_Page_184.jpg
2b903953dc88938ff7cd287b9c33650a
22b6e9043098bac9ce46bb9662aa1ad26aa67bea
101387 F20110112_AABDHJ owings_r_Page_086.QC.jpg
01b903239ed42c409eaeef0c3bdf60ee
73c0b067ea9d7e3878c0cb932119dd86c0a94886
25784 F20110112_AABDGU owings_r_Page_047thm.jpg
adebd2c5d865ca89dd2125be8e7c6e2e
2a22c911574255d80dfb0c90ef5dac8db81fea92
1729 F20110112_AABBYM owings_r_Page_130.txt
75795053a0be7742145b2e7f5497c8e3
e4a682037a88a2cce07233cab1570ac208bb086a
178939 F20110112_AABBXY owings_r_Page_188.jpg
41c874f8e65c8776d20512e4a2d480b8
1038260877c5977dbbdc221ec4b72a651332600c
247113 F20110112_AABCEH owings_r_Page_207.jpg
4848cf3800745d7859a5dfdae513ba53
69ced9b91baa5e57b8ada2339230c2473a2ec6bb
1051889 F20110112_AABBZA owings_r_Page_117.jp2
102c22c4b0f1c7f2d578704dab57db3e
41a6e3b853e093ff0f471fee2fb50b8544fd76c1
258244 F20110112_AABCDS owings_r_Page_186.jpg
a8be33506a4d364be156098c60cff8a1
7aada85dc8bb630d62ef25450d3ae1806514fadf
79424 F20110112_AABDHK owings_r_Page_087.QC.jpg
c3166c8728ea2e7dbf6a5731c66fd811
fd550f54e6a15c8cfd1175dd6f5b30c888685731
23630 F20110112_AABDGV owings_r_Page_049thm.jpg
5b6604351f3681ac7619d2da1f15b362
6bf6b30a5892d8c6ee01a579d6ecd11d322e54e4
F20110112_AABBYN owings_r_Page_067.jp2
2ea44b96eb13bd6fa545a0c63fd5ca3f
c80283ccb7657fbf6831f1f7e636a54f5537b435
2065 F20110112_AABBXZ owings_r_Page_143.txt
6696a4d6c6547e855a3ff867ec9e4d83
3c32a8d8d7c6ea835f07484c0d49121912e154df
112670 F20110112_AABCEI owings_r_Page_209.jpg
cd2d85167f295afac4ae8ad9b0706e18
f54d82e09fc15724d70d7cde2b542bd56fe8fe90
50995 F20110112_AABBZB owings_r_Page_119.pro
9420a493a2b512aa1987f74d8db36b3a
c9e794e03e14f84404bc882499aa8edd7de76467
165097 F20110112_AABCDT owings_r_Page_187.jpg
cb920501a1f3b940df9dc7f303e850b5
a32c813ef255c1723ebc2dce7c8bc9b89073f9d2
96040 F20110112_AABDHL owings_r_Page_089.QC.jpg
cd95d956fee50db110a67f0c2355cf85
124316dcf3477f6276094429b284b26a62aacd6e
70008 F20110112_AABDGW owings_r_Page_050.QC.jpg
70c6b7236d65715dca3a38073e43680d
42b88614219a91cb11b3b835ea296330c6bfc753
1051974 F20110112_AABBYO owings_r_Page_093.jp2
2d8b34835f68f9f58acdbada2238e7e2
fc8d0c4cf12a54e32aa59e7bfa5355c52fb50ac4
90777 F20110112_AABCEJ owings_r_Page_210.jpg
2a94c64843f08976552f655b60f13704
b91cf54dea3392413c9f17aebec1868908b89936
61838 F20110112_AABBZC owings_r_Page_202.pro
178562cfd68bc2e10cdb9dad49f7e36a
fb59c15cdb1e4f025d41e2968bb2a3f624055f9b
190444 F20110112_AABCDU owings_r_Page_189.jpg
84fe7affcffd698d533a48d89195779d
f19429e92269ed9036a3bea77fefa00a69e9df36
101645 F20110112_AABDIA owings_r_Page_116.QC.jpg
74d0e9b3fda150cea599600289ae3d6b
ee910762c640315382f06104d0afb3e5dcabbd0d
87637 F20110112_AABDGX owings_r_Page_053.QC.jpg
8b61df7d10332e1e3346d8b4d568fc35
bd6f261c2d1eb1b55ee8a4a7a7af3ce75f3fe1c7
35374 F20110112_AABBYP owings_r_Page_138.pro
82c081d27e35fa8c3224985c83ab8517
d0bb1307953fe580d5ce3566bd03107097901b0e
241696 F20110112_AABBZD UFE0002242_00001.mets
bc13d2c42832fc63a9b3ae116e7bc50f
9a0224ea407a736743bbf6625c5fc2d1cae17f8d
138112 F20110112_AABCDV owings_r_Page_190.jpg
a0c52ddaf9fac5fb227e727c1921e056
24bdfe63b57e8779285a7c24306a37c9f406b5e0
79259 F20110112_AABDIB owings_r_Page_118.QC.jpg
e89c5837aed06a07eab09e8771e9491a
60893d5fc625bbe2e98644d5d8d53bd6cc494550
49332 F20110112_AABDHM owings_r_Page_092thm.jpg
e61b37760ecd89d63fc28115f2a6c591
02409654cee5d1753141259da7775c5307f8cf12
98081 F20110112_AABDGY owings_r_Page_055.QC.jpg
a4116b3ad80489dc6427ed2753896ece
611177602abf96444e6443a02802bc1dcb0f36a0
30988 F20110112_AABBYQ owings_r_Page_177.pro
076ed61c5a3698f36be6ec5eb5d2485b
dc9c4ea98d3b226f165b7dbac5157f03111d59b5
22333 F20110112_AABCEK owings_r_Page_001.jp2
c9d361b89195e34d27537b3bcaf010a8
8caa6fd25a7590a8543d0b559a8be0f89feb79df
152778 F20110112_AABCDW owings_r_Page_191.jpg
bacebcf58422c0a80e82d42f58e39641
140276b0be879d611a0933f3bbd9bb45d286f1ce
23238 F20110112_AABDIC owings_r_Page_120thm.jpg
15544d9824fdb044a306eff472d3a7f0
efc4a06447cca00cbb0e82015d12e66d983fed03
96706 F20110112_AABDHN owings_r_Page_093.QC.jpg
53e410a2c56f307a86a530b4412d807d
d1d4218c2a23f737fd4744bf83d7693d4918172c
24198 F20110112_AABDGZ owings_r_Page_057thm.jpg
89d0e47af996091c7b0ad9f9829a66ae
029ade11af7d28c44cc987b17be00acbd6005a2d
43030 F20110112_AABBYR owings_r_Page_183thm.jpg
2d6746434ecb0188365fd57b219bc336
b4274cc2e1ddc8b366fcf0054ae28f6335700739
109648 F20110112_AABCFA owings_r_Page_023.jp2
e7b84177248d78fb16cbc34ae132aa25
306cee0f5c17d375881da1a49a225d90b9089bf1
6295 F20110112_AABCEL owings_r_Page_002.jp2
445443adf4efe0ff3c74e4f1e07aaf41
c2ff928d9c51b6362711f54f23d27cbabb071cbf
202879 F20110112_AABCDX owings_r_Page_192.jpg
7cfb1fabfbe2bfc7a9d03d51b594e6a0
883cfb0521de084a01dc2a8272f3918849039afa
F20110112_AABDID owings_r_Page_122.QC.jpg
8712d70d73dcf5e8ed09cae4969ae814
7c00cf5492448636d854e3b0bf225caeb7fa1ffa
80955 F20110112_AABDHO owings_r_Page_094.QC.jpg
9027ba213ecca518ff785edf0bec8a4c
24d2b3777cbec10e8286724dac834306bc607f5f
F20110112_AABBYS owings_r_Page_047.tif
522a25541cffeb2c3c0a48edf5c263b1
e40c34675f08cdbbe483f1f3e517c9e4f58a0804
95494 F20110112_AABCFB owings_r_Page_025.jp2
40427e25efe134f5f6e6a81eefe6a8fe
4e7214ef06aced3a9a5ebe5df6da0e26f28743aa
99751 F20110112_AABCEM owings_r_Page_004.jp2
ba0afa5307a540e9b97c80f6172fa653
7b8c04e7d8f1787fd3e10756d5c28ba686edeba5
49052 F20110112_AABBZG owings_r_Page_001.jpg
f975032378fd6d0d56a769a864535e0e
1834690f404358de49e5d0774699e9666b123d31
155504 F20110112_AABCDY owings_r_Page_193.jpg
10824aa01f46d1b77692f45a8ce7d1c0
e97b7ce5f4810c5eb8ec004791e557a6f724fd36
63926 F20110112_AABDIE owings_r_Page_125.QC.jpg
bf6c56cc110d84c158a8b6f82c743632
86bcbd0e2a7e9453af5d061eeebdd877d5b5f7ce
23712 F20110112_AABDHP owings_r_Page_099thm.jpg
abca0c044ef217cc1c997d81eae92d84
a88868075bc52baca4784843509b0fa7e58f881a
1028 F20110112_AABBYT owings_r_Page_096.txt
3dc0cb0e9f31c610a4ae6a8b7b95af2c
0c62a75309a6b6fc61b8a2b67cf0df344f59f704
106369 F20110112_AABCFC owings_r_Page_026.jp2
27a6ed2dc6fae075d32a8043c8135814
52709c2bfa2a7562bf664555d5f34d2552f17477
F20110112_AABCEN owings_r_Page_006.jp2
b599dc820db01f2369e8aa4070672b10
a7708e17298e39e5bcbd8c043eb5a2e3e595b6fb
15982 F20110112_AABBZH owings_r_Page_003.jpg
b52b23a69220c62ebe3df7492ce43ca4
9482aedbf444ee23ebf7d9f27bba571a99c07ada
144534 F20110112_AABCDZ owings_r_Page_194.jpg
b1c43c946afec8cd6cd8bca897579f04
88c73a339ae378fb057369d92766f37e1f6a45ff
24665 F20110112_AABDIF owings_r_Page_126thm.jpg
52f0bf4833764b757d679f96485411bc
443adca4a21e0699d977c8f4652b86c31a7575fa
49964 F20110112_AABDHQ owings_r_Page_105thm.jpg
2a0c86e91a895bf5b84922de20961c3a
98a6d19b20317ad29c10c7673856acaffe207c54
F20110112_AABBYU owings_r_Page_040.tif
dfd1be5d2778e580e413b2430f9218d4
8bfdf9febf917856ad0172e11539f234aae51192
1031835 F20110112_AABCFD owings_r_Page_027.jp2
176eb2784a63c5f7e3f5d2965649a6af
6c9bde10947f26812847b2cc4c63aeeace5d8fb7
1051984 F20110112_AABCEO owings_r_Page_007.jp2
3dc54286dbec5795a35928ef9c89e5a6
1f94cb9e47a6688f09e0fdbdc45864f149ca7a6f
215046 F20110112_AABBZI owings_r_Page_005.jpg
4ccfd3d6818500a0aa0c3a681e65a719
292248ce7cfa2e8912e84039fc3f22d7e905ede8
71575 F20110112_AABDIG owings_r_Page_130.QC.jpg
a34100275626529f010611dcdc2de3b5
d6dccf1a4726a7019aa1a9d632d6b98bfa97fbcf
50527 F20110112_AABDHR owings_r_Page_106thm.jpg
900f21c27008e6f8c8ec89d902479774
a6f93722df9f31875e0a17b5b1ff5b75621a1524
42247 F20110112_AABBYV owings_r_Page_130.pro
2beee15b3316cfbb5f939aa8464332c1
d0023cef71f731eb2872e81726ae373df3b53e24
108976 F20110112_AABCFE owings_r_Page_029.jp2
26c5dd44f1a2457d1dab7c209b68ef49
91d64afbb02e8b6b2fd702932c1c6467ac0c84c2
413062 F20110112_AABCEP owings_r_Page_009.jp2
982450315166ec3dd085a457e746ff34
4c71bb012a64c47d15f954e529bb0058904f15cf
272096 F20110112_AABBZJ owings_r_Page_006.jpg
8b1ee8979014668b61583eaf9389e329
7eb085922e6a689d9c41b8343e6a2094bcebb613
47913 F20110112_AABDIH owings_r_Page_131thm.jpg
2f330209d5e592cd01ecb1dcc8077c70
4111b8ca00f6369f9f2650fdcef63d8ed296b273
103548 F20110112_AABDHS owings_r_Page_107.QC.jpg
62dd74c1af7fe1eb08c76af7dd2af6f7
edda2dff7df5bc47c62f88bfaa8de5921cf66035
61399 F20110112_AABBYW owings_r_Page_178.QC.jpg
b3b69b96917d9df30a8819810ca0047b
daff457113af9dd50a130edee6834acbeb345554
119971 F20110112_AABCFF owings_r_Page_030.jp2
8b4cc196c30b8323ec540e685bbc67d1
d9ca10c24ae35dfb1c7726d8d9f08e934348134b
1051979 F20110112_AABCEQ owings_r_Page_010.jp2
8908c27628ac496dcd19919b4a438b7c
75408afd45da5e08af279d8ffeeec101f4a45fe7
334382 F20110112_AABBZK owings_r_Page_007.jpg
1a8f4d636f023cf2c3d66525a498ca41
7fedaec137ed2a51a998677264c8db13fe8b798e
40844 F20110112_AABDII owings_r_Page_132thm.jpg
c1ee46c35a6454a424dec6c926468fef
f1f0411c1aefa11a3a7af1391b6d1667fc4a4a0d
24994 F20110112_AABDHT owings_r_Page_108thm.jpg
c5f1792d079e86c68b309f3c4de15c4e
f79b8d31ee9ec5b8553f00c849a4941ef564818d
109104 F20110112_AABBYX owings_r_Page_119.jp2
009e3d1bf3c04997e1f049a23a5483f3
e5c1c89bc8da315129fa608fac413e4e9c467006
98495 F20110112_AABCFG owings_r_Page_031.jp2
bf7f5b0d110ad36a71640689c8ffd172
f28a125d82fa7735208361acd4fead4c33a59480
F20110112_AABCER owings_r_Page_012.jp2
7fc68cface3b5ba8c51d32ff465510d6
28b921f8f6b1d66cd6650421fac1d758f6492158
299511 F20110112_AABBZL owings_r_Page_010.jpg
7ba1e4f2e7a8a9c7d2eaff4a1baa2c98
910c40eabfe1b08f4f52fe19c895f6d0102fa2ca
44454 F20110112_AABDIJ owings_r_Page_134thm.jpg
b4ae9ff3a31b9aa6479cc841d024afc6
bddbaf974b4eddb216a529b8233b4cadc6869512
91808 F20110112_AABDHU owings_r_Page_110.QC.jpg
0ff0c6120cb8cd3b501a255b34ce8971
3867ecd1edb0d8dcede40b53dc8cc4f9bf86d739
1992 F20110112_AABBYY owings_r_Page_108.txt
61187e05ebda6cf712669c22a59bfe44
034349dc0905ed915a96c22e299c4c48a4631832
100142 F20110112_AABCFH owings_r_Page_032.jp2
0f6ffc1466889c4d75f38e01a0deddad
1fb2b2ccf7a66a33353de21f39ca101e8ed4d330
1051981 F20110112_AABCES owings_r_Page_013.jp2
2d6515a0c241b8d342576eb2ec15d6a6
ae29d191f0c5caa6c2a761a6f553dc0dfbbf2dd4
111874 F20110112_AABBZM owings_r_Page_011.jpg
e884c945d107b007fc5a154f9941f26b
52c0278c946b864912851bfb3467412516e63d97
46162 F20110112_AABDIK owings_r_Page_138thm.jpg
5d8c6fc49d971df72facbc23085c6abe
3f5c5fb62d7120aaf0d5fbc9798b6822f1e6de83
48347 F20110112_AABDHV owings_r_Page_112thm.jpg
332c09ae6afe3ce799070abca5224483
29ff62c038545853b49fa4f8d47d4cd71b23ae75
19429 F20110112_AABBYZ owings_r_Page_210.pro
d80003ad0f6eb153efa3f9ea875e9ba3
ece93ed9e9c2cecc362e6c5c9c1b3fe43c098019
111545 F20110112_AABCFI owings_r_Page_034.jp2
2485082f3a8a2643e396ee101d0023d6
d3963dfa39bce515120b659c54bf12d696c2e6bb
F20110112_AABCET owings_r_Page_014.jp2
3d0eeb2d17b4dba53a437642d543b7dd
9d386604af285931e80a5fd1f55f023344eb1e87
280879 F20110112_AABBZN owings_r_Page_012.jpg
8fa448c513910a9c94afd3eeecffeb17
8817604d5b3add941edff7c3da445d773f430707
81882 F20110112_AABDIL owings_r_Page_143.QC.jpg
441638fc8c33116c00281bc46a531023
f2de1524187cc02fb89e80be4502194c210dee51
97421 F20110112_AABDHW owings_r_Page_112.QC.jpg
d01074014c92fa59f2ba8cde9994c0bc
c8b8d7bb85ad0a29900958f94232061eb21744d2
1008176 F20110112_AABCFJ owings_r_Page_037.jp2
428b6db718b9c7a2b935aa42a8f09868
3d4dfeba7bc27c462a0920e8fafadd689ed5b5e2
1051985 F20110112_AABCEU owings_r_Page_015.jp2
b63b7701b05384b158dc21a0df9b22c4
a85be4e0c71477340f0a1f4019b877c86c092fa7
367834 F20110112_AABBZO owings_r_Page_014.jpg
09f269c899bbb9ee1e9aa6db00733337
59d19b9c49a5f5509b6fa83b650aa4bb855d2da5
68300 F20110112_AABDJA owings_r_Page_165.QC.jpg
41b526a45e5f5e0563ed09e9072e9fff
bb6a69f31442e80aadb5ad424ca398ae7eaa2b23
79492 F20110112_AABDIM owings_r_Page_144.QC.jpg
d9288895a43f39e9438dfddd909f6068
4df75f3cca0ef3c4378d3d66db341a0f8b2f2270
100112 F20110112_AABDHX owings_r_Page_113.QC.jpg
fb75f81acbba07bc070408cac5de648f
aacc116c96507e8612db38771275a511c84b64aa
1041354 F20110112_AABCFK owings_r_Page_038.jp2
c9d890e5da498a5f41816e68ad0dcd5e
2502366adef1084f7705770dab9127b0af09cca6
1051982 F20110112_AABCEV owings_r_Page_016.jp2
6d3cb17ada589d511db516c172e448d2
22e030027e38796e0738a384977e9a6a8f314044
374955 F20110112_AABBZP owings_r_Page_015.jpg
0e8809bd714b412ea5989d535efd3c51
7e7c42bc65912274dac4f0672749950853720a09
70597 F20110112_AABDJB owings_r_Page_171.QC.jpg
3afecc04a3187fceefb179767b131c03
c4c881d75ca0983200e4e3d663874c9d7a256c4f
46825 F20110112_AABDHY owings_r_Page_115thm.jpg
0be92f46b2cfe1e67565b1b8e3d40995
e954c6638c2eb661ed44c30f3300ec04274922f4
1051973 F20110112_AABCEW owings_r_Page_017.jp2
1bbefa0335f75de5bb838a5f58592f8e
2f4851d9df6e7e97bbff321be73583c33034017f
335315 F20110112_AABBZQ owings_r_Page_016.jpg
5376689c98a38210e354507144da586f
89967e9a385010538e8865d54441b51b7c270299
20358 F20110112_AABDJC owings_r_Page_172thm.jpg
c2b1120630d55356c894222f7ff05c23
3982b2ce9d8d94ac2d951db813b2e3a94928904f
5005 F20110112_AABDIN owings_r_Page_147thm.jpg
eb28ea357cf6382df87dac929dc453d6
4eb1dadb198516f3a3d2572e009a63744f4ae129
95337 F20110112_AABDHZ owings_r_Page_115.QC.jpg
3a137df3dc5f6e31cd3284925021a460
fa77a8ecd07c183ca7340139ca090740d0caa92f
986949 F20110112_AABCFL owings_r_Page_039.jp2
5b1a680f6ff96bc614d1e83b794c22ee
8c5e8ada61fc99dc25666d0727fc249c5e8c8695
85428 F20110112_AABCEX owings_r_Page_018.jp2
534f11c12e176e346d21392f598c18b2
b3c9c1a64f03c3e34108a6e4614932792af72847
342423 F20110112_AABBZR owings_r_Page_017.jpg
9957fdb4fdfe3d9b00fbe67449c76a1f
1ba4e851ae2efcbfa32a984037562d145feba398
116790 F20110112_AABCGA owings_r_Page_058.jp2
fbc6c51904ed37c9271b96e51134c399
c2c8522bb0af30f48b895ef6aa6ea769abbcc116
63756 F20110112_AABDJD owings_r_Page_172.QC.jpg
e41779d5926d53af19afca62244647a3
ad65bdce8596a758b4e9ebf138e27ef1743c1b0d
48898 F20110112_AABDIO owings_r_Page_150thm.jpg
d304ff996d49495f63757e23a4981f55
dd2f87edf625c20d5e45f5e586cc16dd951c117f
997660 F20110112_AABCFM owings_r_Page_041.jp2
26c43ed4ee5aea92c4aed86ac6953ea8
0066e7f79e94c9764f23508fed25cd25fb1c1730
1021773 F20110112_AABCEY owings_r_Page_021.jp2
8bcb79f51f9749bb473d3e69c9f8c382
7450dbf17b58c8a5e140c8177f37139e00bf9785
149620 F20110112_AABBZS owings_r_Page_019.jpg
0150c03bc9b7f0ca1cbcc363f6eac4b7
983368f284cc926bbe0ca48b44a413b99cd8d9ba
107758 F20110112_AABCGB owings_r_Page_059.jp2
c00714bc438ecfdc06cf5020ffaac827
0b10c69200e566a58a8a3aad7cb06037f6b95058
21168 F20110112_AABDJE owings_r_Page_173thm.jpg
6f3c96b808dc3310e2d784b7013a65ff
0361ea9735d106bff10a0fc80a45618453ef3ee7
99901 F20110112_AABDIP owings_r_Page_150.QC.jpg
84de6080c2a82352712de4fdcd8624ef
566ee21fc7d016c7df262a47b8025936c3004898
1041082 F20110112_AABCFN owings_r_Page_042.jp2
acc3173b064f7976d095ef5d7d2497a7
b7fa88f6ec7f06c79b9961cf3f69cd18c120cfe6
109986 F20110112_AABCEZ owings_r_Page_022.jp2
5ad661158b52f8f2db7699b267d0a2db
48744b1a40a65cf92028a3cbe73bb6c3ead94d29
209500 F20110112_AABBZT owings_r_Page_022.jpg
708cdfd71de6312c76a8cd695a7f0842
29abd7245bab906a8dd3de936b309abd8a203353
91062 F20110112_AABCGC owings_r_Page_060.jp2
216eab2533cba319a150c91c59448435
b5bd8a58ef52d745c0cd323793b60c8d430696f4
95883 F20110112_AABDJF owings_r_Page_175.QC.jpg
781492f8e4492c0aaf2a4d0895e03821
ab676bb1949d7d0d5a7c5634c4c6bb4a5ebdabfc
42937 F20110112_AABDIQ owings_r_Page_151thm.jpg
e5b403c8715fd42b36ab3a192f101650
299f899e4c344e364b24e2218c6326260861c442
1051946 F20110112_AABCFO owings_r_Page_044.jp2
2be6419b36ef2a17bb62f2e936ef914e
ce5c0fbbfc1dec2898b1eb6c617307356c026d0b
214629 F20110112_AABBZU owings_r_Page_023.jpg
e174eb15941704e078a48e3a9ba40f61
1c5a200f5300370c5b3611ba3a7dd323be9fd97e
973853 F20110112_AABCGD owings_r_Page_062.jp2
f6f138f1c4c6c8ab40c055ae2b095d65
63e187dd1a524c849dedfb6778f86a4754b18d8e
42888 F20110112_AABDJG owings_r_Page_179thm.jpg
5ad8d73a39ba95a98c4f268d366a3ee1
f8996ed3d1ecb4a4dcac107ae647647bf8299b4c
23212 F20110112_AABDIR owings_r_Page_153thm.jpg
0a69a5700d59413c64967acbbe990641
3ba7e7de2a1111e4ac5dd5b2c15be494accf5c6a
100320 F20110112_AABCFP owings_r_Page_045.jp2
d02b476107dc66e0181ae294b036ab38
6a1936d265a4f324412dadcdfa706e6ac2285ae6
82268 F20110112_AABBZV owings_r_Page_024.jpg
58c5487186ca244a960bcfd084f812a7
a0eaa6a3934d29b1df39e27cba4e85e89e62d77f
95644 F20110112_AABCGE owings_r_Page_064.jp2
bee2fcc0c65b60c352eca7ca2093395a
5dcdb89e5e5cbe06d932d887463ea0d80505ebe7
72479 F20110112_AABDJH owings_r_Page_179.QC.jpg
9b78460e0302b5fc4c261b1135dc1e19
8109c3640e7cb94bad489e092bb011ba2809349c
F20110112_AABDIS owings_r_Page_154thm.jpg
2cd074c127ddf0b633b29ff85ea94c8a
b756ac8a9dd9bda45837164a41ad2f2aa1a0899f
863904 F20110112_AABCFQ owings_r_Page_046.jp2
0fd3befeda9a4db587073b68bcb0bb3e
7ba34ea777fa205fa299fa0c148c9df7fd6ba627
219159 F20110112_AABBZW owings_r_Page_027.jpg
ffa107fb9def3dc0136c6f786710a5f8
2890fe5c980da5f68c5fb686648fa2d0a6a1779d
F20110112_AABCGF owings_r_Page_065.jp2
6ee8f750f349b3e783f174ad6b8b6169
8a6a3f3a256565e7cb7ceabbf67cdd520f092b4f
18967 F20110112_AABDJI owings_r_Page_187thm.jpg
0d9f9d08d5199f6fb57b48fe98688704
41aff37db98be0a48d9789b737cf18c2dd8909be
97323 F20110112_AABDIT owings_r_Page_154.QC.jpg
a837e4fd02f59c8f3b3cb6bed4300e1b
e74a78df1b840e3aeffa82e11eac2823c8b16e4c
113752 F20110112_AABCFR owings_r_Page_047.jp2
363512ed25ad6f9a0b327c2ab38b1d34
8d2d7240e1bc3f97cb7e5725dd19497b3d256bf8
244749 F20110112_AABBZX owings_r_Page_028.jpg
e89d41b2b1f949d776f068c8f5f4f6f9
3eb3ef5b1694871e5e0c70fc769d8b895b93953c
1048008 F20110112_AABCGG owings_r_Page_069.jp2
31d6321c8717ab91da269f3bb2d7c60a
dc21806c39682b8828bafd802ef5a7cabaf99d36
57170 F20110112_AABDJJ owings_r_Page_187.QC.jpg
55372aa394db8651dc614bd5c30dd7c1
7a06237546340c620795f34012ec437513aeabfe
62685 F20110112_AABDIU owings_r_Page_155.QC.jpg
51b16cbcf59f4937b03f72cbda949389
dffad5135648e913c7e5422a38fd2b630989ed05
1051958 F20110112_AABCFS owings_r_Page_048.jp2
685c1911cba350f955ec44170bc53519
6e8f0d6469c8758f3d0691a526288d965ed67689
210898 F20110112_AABBZY owings_r_Page_029.jpg
ea4dd5436ce12d663a394eb86b1d26c6
6c025ff53d459c072207474c723c81dfcfa1a7fb
981034 F20110112_AABCGH owings_r_Page_070.jp2
8c8950383eea9e3c10d65eabbfac8592
f957b8e37de1741bbcfbbb3a3fbb5577e8f64b3c
65981 F20110112_AABDJK owings_r_Page_188.QC.jpg
940b902dbe074ed8d01b9b1b75480513
8a16f67843d88b8cc94c71cc1c511c37a5057ecc
42327 F20110112_AABDIV owings_r_Page_158thm.jpg
1681b0876ddf01ef62458fd1470b8a89
351c288910eaea17f8d57836e5780000332cd092
104219 F20110112_AABCFT owings_r_Page_049.jp2
56e54def290dc059fc2d3c0303d499e5
02d8e077af6b513d0f4553a9d7c5366376cf804c
239452 F20110112_AABBZZ owings_r_Page_030.jpg
e860ad7077db5d4fb0b50d6f8dd35183
e195c864f7f63345f114003d96aa7801aab165e0
1041016 F20110112_AABCGI owings_r_Page_072.jp2
fac9f516f2127e06ba20639209f1f2ba
99c5b67019ce25a1260986d743bc25d7db394cf8
24136 F20110112_AABDJL owings_r_Page_189thm.jpg
52ac298391327f59829f371aca0993a5
92b7267027e30f86fefda6259faa714948e0ee66
79586 F20110112_AABDIW owings_r_Page_158.QC.jpg
b0616691c303153659a36ac023ae0ad7
a40c6ebfe4d7c96d2987502f6b5dddbea3dba1ba
1051905 F20110112_AABCFU owings_r_Page_052.jp2
8b7b9b4672acc18c84e7fdd46b9d97fb
110aa1024737e0152c00d2c600982ca4f1df3066
F20110112_AABCGJ owings_r_Page_074.jp2
48bf34cc364033b4074e222df63073a8
6a9ca6dcf8e67472d4c8218ee60130694e77b39f
54335 F20110112_AABDJM owings_r_Page_191.QC.jpg
980db36b96115abcaaf3fbb3fe77a924
31eeffe5a9057e32baf28b0b3342ff79b1caeab3
24165 F20110112_AABDIX owings_r_Page_160thm.jpg
1497f1d086ef78e0633f0fbb0def1cb8
211d99dad9889a4a4f6f48b5101d1447b48c0c52
952737 F20110112_AABCFV owings_r_Page_053.jp2
74fb0e0020d2ef0b3eea21e4bf049416
469dc6031114faa572f66365386c226fb52ce404
1051971 F20110112_AABCGK owings_r_Page_075.jp2
ad59ab9495b5e39904cd5a5f1257ada5
b50eab09b71e1f46c8f9b0ac3a3b460294a815ef
96445 F20110112_AABDJN owings_r_Page_198.QC.jpg
f9db2f83a532b2930e6ce216e0535212
09c8dc7bf0946f37644ed8d5707ef75bcd311445
75027 F20110112_AABDIY owings_r_Page_160.QC.jpg
47eee80eebf105ce1bdf3c93803b0eec
86950b35d56c55c5727263789584367b323d44f4
100913 F20110112_AABCFW owings_r_Page_054.jp2
334c75c67b665df848338169931dcc0f
3fde476be5247af5f7a9e97738e334da11a48f24
82573 F20110112_AABCGL owings_r_Page_076.jp2
9d32af4a21156fce73431b4d4e4b2fef
6f079d0be4a421034612f2e9b7c9049c7b1344aa
20929 F20110112_AABDIZ owings_r_Page_162thm.jpg
43dce7e357d8266cfc3bcc59a625b3b6
7e41ad62269b437e0641d3af6e7040fdaaf1e698
F20110112_AABCFX owings_r_Page_055.jp2
f51d67b2d7483f0e9043acda5cbc9335
1fa1c8b47946762c2e3b3c1dc27a93b2810fabe2
58719 F20110112_AABCHA owings_r_Page_096.jp2
03fc64b6ac533d000755ad15e0ec5c4b
2f94975afe3f873f2e82f3af6bde109e002a1651
109900 F20110112_AABDJO owings_r_Page_201.QC.jpg
156f5f83bff5206437a694c41db0290c
e45094a3190baa5e3a678acb5596cf8fa4cd9e06
592782 F20110112_AABCFY owings_r_Page_056.jp2
64dd450a9e78c9fc42422a7db8f0c9d3
6b6e07dc240e6456b390adf162d616fbb8927be0
106583 F20110112_AABCHB owings_r_Page_098.jp2
3526dc088d696850750cfb4a5b869b8c
c8f25be7eac34118a2fbe4a872142166613553a8
950492 F20110112_AABCGM owings_r_Page_078.jp2
41809592c12e9807e5d1bf4bb9f89b60
4c76188605f784c9a65a1634ae2d85b4b463c655
22537 F20110112_AABDJP owings_r_Page_204thm.jpg
2334ad66d0e7ae3a7c08a39293e86ec6
f8d0cbd5c160f28edc252945fb199ad03938147f
109836 F20110112_AABCFZ owings_r_Page_057.jp2
90a2fe31bafb5051fe3136a806c70279
a4887c80568038f5346b25a95b06e88c02b5175d
104565 F20110112_AABCHC owings_r_Page_099.jp2
cf0f2dc9b7efc289a632e4e7d87d02ef
99fb80786e6417e17be9278a206386372ae0edd3
710849 F20110112_AABCGN owings_r_Page_079.jp2
dbc9d055c28430d690ddfbbd17accb45
8d307d3955bcfbb0e073e23b85723c1081e09497
83488 F20110112_AABDJQ owings_r_Page_206.QC.jpg
291b43b795126fa680f4b111cf062a80
f7be545099a4d3219f42fc85201c66fa57fa202d
97430 F20110112_AABCHD owings_r_Page_100.jp2
68afc3333075aa78d1397562991d775f
1906da723c8f08a13702e7f08e7fe4adeb55ddd0
F20110112_AABCGO owings_r_Page_080.jp2
89fe96f4d4c7881acabb798f81aba863
b0bdf0cad604aa6820f2a0e8ad9c1a8d522cc17a
78888 F20110112_AABDJR owings_r_Page_208.QC.jpg
16058b1b61fcfb33d0ad88dfb9d323b4
81a9d3a94326b0c4a336c1cae3fd6ffccd91a764
F20110112_AABCHE owings_r_Page_106.jp2
c0317c2f2a08274debe717525d325474
aa8abc7072be4f604550af5279b228b18fa4a3c3
103571 F20110112_AABCGP owings_r_Page_082.jp2
dbc0e60f6c92a87523bfe622b4dcd0e8
5aa21cbc578a3208dbb8535734042bcfce529503
37626 F20110112_AABDJS owings_r_Page_209.QC.jpg
246cc693985aafcdad998b0a25158412
77a3a637a4f154740a5f96049e2e41bf47e9aa82
F20110112_AABCHF owings_r_Page_109.jp2
6294215c5711999f6e275aedd9fde364
8c33318f1c2bbb2ebd54ea9c69936ba2487555fc
F20110112_AABCGQ owings_r_Page_084.jp2
2b2b12dd030172da3e4e5bc10d7be3b2
d3babc02adf706f54c211fc50a1ac48b3410b384
11148 F20110112_AABDJT owings_r_Page_210thm.jpg
abfba3918be8293c8af7213d080b77b8
7002e06da62302ada7bff6f8aa62548fcb5518f1
1047325 F20110112_AABCHG owings_r_Page_110.jp2
a8d687f395d46226cf3be46eb016845f
266b031d4e38dbb6fae85522dd08a9dc8e780497
F20110112_AABCGR owings_r_Page_085.jp2
619a289de5fa377d2144b91d5dadd811
05e897506a8b1d9127531ff53cbbd15b1dddeb39
1045175 F20110112_AABCHH owings_r_Page_111.jp2
496dd934524c0cdb0157b675c28fba7d
e20a8ee211e6b1b56918842b586b55fc82016d11
1051887 F20110112_AABCGS owings_r_Page_086.jp2
23b3f3445a2641e79eb7c247f8f1b097
9a29ca4ae73b9328a82938bb0e5d24569405bc4d
F20110112_AABCHI owings_r_Page_113.jp2
ee6dafc5ba270039ad0b6a3636b9bd4f
a097b6f39c1ce36ee96091e25e611bcaa6890eb6
107736 F20110112_AABCGT owings_r_Page_087.jp2
5b0d07361cb690daa872d2b76d511e55
a79486ea2c7cb3477435487739030433b48a7432
1043349 F20110112_AABCHJ owings_r_Page_115.jp2
57dadb3694e5a0788ead250dc596ff94
72c865ed1ffab1df1cf243cb8b395e17563bbf27
F20110112_AABCGU owings_r_Page_088.jp2
4d2339de93067dbecce99fac0b62d8ae
a79d5ee46aeab11f2c4bccc14a1e95156ccc394a
1051970 F20110112_AABCHK owings_r_Page_116.jp2
efc90242665f698e6d10e4d67274f78c
82c7c5bef9816ac87ada338262cfa68085c2c438
1051927 F20110112_AABCGV owings_r_Page_089.jp2
b4895f427508f8d35df709f8c465de9d
6ecadf9264f5161186446c8b9bc8fadfd24ea8c4
108687 F20110112_AABCHL owings_r_Page_118.jp2
ef2b1aa3f21ddd9a2538faf32b7725b0
e9db9750557ca928e87d688ba73eaf6c84e262e9
1021785 F20110112_AABCGW owings_r_Page_091.jp2
aff7929cf2c7604fd245b60694be57b8
8b8e7952fa8e7e92d8c306126645e8c828b05dfa
113170 F20110112_AABCIA owings_r_Page_143.jp2
0c20fb8f0baf567e147422eef7e4fdac
154c9554d597b3a07099984b2a7f2692b4ef626d
107694 F20110112_AABCHM owings_r_Page_121.jp2
2c322bf63af6bbfd188d1b80037fb36d
bba802b7bf325acaffd4667b217da2437f1b9732
F20110112_AABCGX owings_r_Page_092.jp2
76cd43bfa7320d2dfcba7d5c374c2c74
44ba691ea5a6859097397cd32865937563cb433c
106363 F20110112_AABCIB owings_r_Page_144.jp2
5d64618194f19421321b840e4baa0f37
f4a4cf22a3034d6d8fe7af81bbf273d633c42384
112491 F20110112_AABCGY owings_r_Page_094.jp2
f368148c6ecdc3d3335f220a6dc1d295
5996232a8f68085f66371a1083f65d656c73144c
108075 F20110112_AABCIC owings_r_Page_145.jp2
edc1222f6a51e0a051f332092bed9838
6324da4bb77a14334e1728893aeba231e4d2affa
112439 F20110112_AABCHN owings_r_Page_122.jp2
fca3145873baeb2f1a2eb68fb7d0f8b1
7dffa563eb36bbdb1949d91e2974ac40a0f60867
109499 F20110112_AABCGZ owings_r_Page_095.jp2
3469782ea6afad7d00e213adca62a29c
a7422891b1bb3e9081a6945d11650713220ea28a
109570 F20110112_AABBFA owings_r_Page_033.jp2
d4004f25cf55acf8a5f28994650bc69a
cab7b4e3f0ec03e0501b0039e003b5bfc78e2100
104808 F20110112_AABCID owings_r_Page_146.jp2
288995276c6c3db351a538cfeb4919cd
520da8537b981d1855f287800b93672316f7982a
56973 F20110112_AABCHO owings_r_Page_124.jp2
51b01a4c375e50af333ec31995c0b6ce
68f5fb26f19b65091b7e85bf1d79ff52c8991255
13807 F20110112_AABBFB owings_r_Page_164thm.jpg
0f8b1f052e4366a87c9632b3bf082fd9
49743f6cb743077cc29b8fea5cf07b10c12feb45
96849 F20110112_AABCIE owings_r_Page_148.jp2
2e6c6fe30015b3513b45a11eccc37d7e
d33c04754b1be9374cf6d8eec1097763d5fd51cf
109527 F20110112_AABCHP owings_r_Page_126.jp2
40ac017c578c6715a22326135b4a4236
d3588a1ff97bd49919e38034c11893b535988b29
22783 F20110112_AABBFC owings_r_Page_050thm.jpg
7f661753eaf35501db6882aa4664e54d
5c6d283c8d28cc8049f3249334190c6cabb0def8
813301 F20110112_AABCIF owings_r_Page_151.jp2
7dfb7167ecfd51c5570dcaad19f8c945
012481d333779cb0b3dd6dea690e15ee6f4a0b1b
100666 F20110112_AABCHQ owings_r_Page_128.jp2
d08cfc633fcde6996a7f273e9bf06d15
46da95ac0ae9620b51c1c017de5d4e4b395c1ac0
45679 F20110112_AABBFD owings_r_Page_195thm.jpg
c70571e7a2c6301776510d9a43ab526e
67d63e3c0c195029b1d71410391dae26f28860fa
104048 F20110112_AABCIG owings_r_Page_153.jp2
b61b93ceb3cceb3353a7977f98be34dc
a3e9e099a6156de1bdc47ebad5df50ed0adee679
94037 F20110112_AABCHR owings_r_Page_130.jp2
93699b4e38b73d511f52d4f3298210f6
637beb0a0940e18c5cf87aa9af65aedd31be6b1a
40508 F20110112_AABBFE owings_r_Page_131.pro
177ec178f1cdc2b8a70ee00f610e3c2a
bb78f456c01cdedff7e5be6f2cc3ed5ce5c7891d
1021790 F20110112_AABCIH owings_r_Page_154.jp2
a3fa488cac7e7329529ab9754ceecb40
50a21373a80a02d327e068b095f83e82a350f1fb
945517 F20110112_AABCHS owings_r_Page_131.jp2
4a08fad34ec4f6a8b75478ac7cd2b373
98e8df9da39ee942ab30eb334e0ea42c1e833604
93095 F20110112_AABBFF owings_r_Page_184.QC.jpg
5b05e3d84b5527ef9b3f4f426df8ba0c
1973f5cb7bd6b165e5bfcbb2b1691846da413454
83300 F20110112_AABCII owings_r_Page_155.jp2
59d36679004c4b106daa37f342b969d0
4741380066c76e4cadbd4c893312305f73228b3a
567218 F20110112_AABCHT owings_r_Page_132.jp2
df53a978f5ee148590cad6a8ae866c58
210c7e3ef004224f9af9ed7e2a9a6034e7550273
106762 F20110112_AABBFG owings_r_Page_108.jp2
6db32e08178c48f87f7175e0fa4a0219
304cfbaf7ee8ad7f0d9050e9954861dea0f1faec
1022267 F20110112_AABCHU owings_r_Page_135.jp2
6a7028385321e56c9963f880dc63ac1d
c4064658a44a385f87371255eb175840a44fd90a
50511 F20110112_AABBFH owings_r_Page_114thm.jpg
c774068ff9d03e1bfb28712e571bd41c
5f74f62815b76a9d9cafff2eb159a429029e7d76
106015 F20110112_AABCIJ owings_r_Page_156.jp2
e4654cd99ec560a7ac2e1066d09abf19
4ec59e86326bf0a2e63139f1ea845eafdb456a49
1044487 F20110112_AABCHV owings_r_Page_137.jp2
54d189635a0f21b596c6c49dcf00fa0c
6619a7ed20de24b9edc98be9ce46090a1201c2a2
59848 F20110112_AABBFI owings_r_Page_036.pro
1f72efd1211fc88b65fd4e5afd0c7bee
a62aa442fc5c8e31d6b08efefdc7da9ec4f75590
824418 F20110112_AABCIK owings_r_Page_157.jp2
b4778f0ef238250efef614a1be788102
97b97761d16668b830b35367cd9faab2c257f110
875416 F20110112_AABCHW owings_r_Page_138.jp2
a9efc4529e050c9bf599e5dd1807c253
0ded9f06cfdbcc1ce5683f8f23e8464989e6172e
1030181 F20110112_AABBFJ owings_r_Page_104.jp2
48733a99639002e10827baa9d41cafd5
a6126e36abe1dd9f69cf12477ab70046300a7554
753372 F20110112_AABCIL owings_r_Page_158.jp2
31605f695a03c9fa5b99c89402f1c217
bf5483b436684c533bb9a0047a36f796a35ff4a6
1051947 F20110112_AABCHX owings_r_Page_139.jp2
a4bb41ab2ecfb5fe4f25b86ee1ff253f
b6e84e03689b951dcd2e49f710de070965df18e1
775384 F20110112_AABCJA owings_r_Page_180.jp2
370fd8f79bd61519e127777c98592610
7299e7a65a0a921fc1d3512ec60e96903af70096
F20110112_AABBFK owings_r_Page_092.tif
784b8f3f5c719c225e6a19a59f49380a
21b5c8386d95cb9d254ea408105ed5e131347d3f
888565 F20110112_AABCIM owings_r_Page_159.jp2
66649546b51cce6f20513d2d41cbfd82
4ce406986afa8fc2b5ac298fd8dc1a91836c9e58
926800 F20110112_AABCHY owings_r_Page_140.jp2
d73aaf00655b85adf845f3d1ec0035a9
20e52f93c330a54c92efd73d657da7e58e938c90
46300 F20110112_AABCJB owings_r_Page_181.jp2
cedc3a34b26f9b4295505809bee27e53
84df047bf66cd41a7f09ca34c81060a306185e5c
5315 F20110112_AABBFL owings_r_Page_147.pro
a926d6bb7d9630792ff5bab84667db87
9922058c165218df095159ebb4002f6c0b18a543
105093 F20110112_AABCIN owings_r_Page_160.jp2
2e08ebf362031652616f298aa0fc0f0c
f0a234f40098cab3fba26d194fadee40f9a47cbd
F20110112_AABCHZ owings_r_Page_141.jp2
84c24e0b89c74f28be3f335c0f4a31d0
8f0893cc83f8154e25b01b805fecb61160ceb96f
103927 F20110112_AABBGA owings_r_Page_084.QC.jpg
944af678adcd07dd498ded22ae550481
593a8d3f529cbe97c6de38537c7cd737d973c345
765486 F20110112_AABCJC owings_r_Page_183.jp2
e1ead31b164e843597adb33cfb0dd066
ae0e7f03cbec8f7b36cd1a79dec12b1c85213a32
63460 F20110112_AABBGB owings_r_Page_030.pro
0a2ca0ad12f90aa7363348d0a9dc0646
d2fb30289ba1ddd1a03cca1cb5952413797a390d
1008771 F20110112_AABCJD owings_r_Page_184.jp2
9ba363b0ff86be548c993296ad793dac
64c6a0ebadce13200a89dc97709413e76d739586
43396 F20110112_AABBFM owings_r_Page_174thm.jpg
5c3b2c0410bb2fc100a40fbbb3891234
d56da993e05bcb1768c36da7f9e37fd72cbbc372
101830 F20110112_AABCIO owings_r_Page_161.jp2
914f5aefc73adbc0a3f45a48a044365b
c120b99ba73fec8a919c331f752069b0f62b9645
29853 F20110112_AABBGC owings_r_Page_182.pro
42982fb874469e3c8d505ca24ebad33d
3944f7e2f588fd0a00084351ddbc16132a5ca6d4
F20110112_AABCJE owings_r_Page_185.jp2
6fc087a83be7cdf25813f6a32ca28d31
e39add5da1f44b8e86a7d67223c57299b6f883aa
48128 F20110112_AABBFN owings_r_Page_091thm.jpg
133c748bb05633fef7383f33ff2291d4
96993dc2b38a79759e57840cf80009f0e5299781
97176 F20110112_AABCIP owings_r_Page_162.jp2
acac1b9d41f0a59e498e32be63e8be88
abbc529d78941b1155974b251e85a9a7620d6f46
1623 F20110112_AABBGD owings_r_Page_027.txt
c0c01465f4b57bd0dd101c92f4b6cb73
ff4285780deb4a6eddd94e3343cdaaa2e21a6e74
1051986 F20110112_AABCJF owings_r_Page_186.jp2
0aeccf0bde7e25ee45cec41574509a1f
88fc8feeafb7149f7df4559e817e7df4f1920589
2062 F20110112_AABBFO owings_r_Page_034.txt
b374ea2221f00dcca24922e62e4a84b2
98df83da7c16f2b7bdb8e987c62da75504df41b3
111645 F20110112_AABCIQ owings_r_Page_163.jp2
62bc630b3fe82c9f3de98897a94cde83
c388827b6028b8e546bf0c6eb30405f0a9e257fe
88993 F20110112_AABCJG owings_r_Page_188.jp2
fa9ee075657754c3378c748a47825f44
6044d56e40a154e295c83682df092e54ba9c9737
F20110112_AABBGE owings_r_Page_004.tif
879ad0755e8bc99a4a82804b4a1b946e
d97dba1ec7b3a3f57b975ed6d78608e5a43c082f
48379 F20110112_AABBFP owings_r_Page_135thm.jpg
f357778ba68c9c6ad38befdfdbf402aa
224d7abe95a2fa1e79a30ee4cdd7d25de4b1b493
57190 F20110112_AABCIR owings_r_Page_164.jp2
0d9ca4b2ac56eca03b79e60a11d63615
6819a96dba2b83ec41637272b9e37618823484c2
98268 F20110112_AABCJH owings_r_Page_189.jp2
9d8594cbf7776d1468ced4daf29934a4
57ef86ff928aebdc593a3670daac1fcf5b5e204e
2063 F20110112_AABBGF owings_r_Page_074.txt
ce490e7162ad21cc3116d360884b978c
d84da6cdc566342ab7fbe81bc871829f36ae69d4
108557 F20110112_AABBFQ owings_r_Page_043.jp2
00267cc3713b383e6d2363340dc4c6ae
97a087928ebc061fe47dd7c67b0e36c96ee55ae7
110556 F20110112_AABCIS owings_r_Page_166.jp2
705d03c8e0bb1270bc190e0874363e5c
2f4beac2bb75812022baf747768465503168fad3
75357 F20110112_AABCJI owings_r_Page_191.jp2
b50e10cf2f7b53952c1c197ea664e726
613e3b5cf5e737cca9ab09e98d73b280028e60d7
208884 F20110112_AABBGG owings_r_Page_059.jpg
4d7e0e40062800f5856fe7bf1039bea3
bf4780095e1033030368e093a3165c7156670b68
F20110112_AABBFR owings_r_Page_170.tif
bb70a7b168b02aac422c7986163dcaae
fa1e7c781387c9117b2d520a5cc30d786a03f41d
107642 F20110112_AABCIT owings_r_Page_169.jp2
df1e3c81a636e1efbdbcccef6f8d5d8a
970ea983c8a8e4b468620191c4231bd3b36f9d6a
F20110112_AABCJJ owings_r_Page_192.jp2
618eeb8799b68e6ed75650c82970e1b7
60db946720706e1e7bafb87bd7d6c979249fdfe5
2072 F20110112_AABBGH owings_r_Page_057.txt
33cead7ae6d3337a6cb71c042b68d319
6b847b466d214676f53f0900b39faf9eeff05b07
1506 F20110112_AABBFS owings_r_Page_020.txt
304cdd9a9568bbe3104d463cb948eb09
b3d4e503cce07601f0514f101a148e2663a8a354
102433 F20110112_AABCIU owings_r_Page_170.jp2
ceff014e4d1c84cfc7fc14ae483332ae
371781e14514159c6707284a48667bcb649df140
80831 F20110112_AABCJK owings_r_Page_193.jp2
3938c9547369d4044f5f668ffe7f699c
d4b12c12c9207cfd6f53a1c1a2ae2baeb4d7b6f8
33933 F20110112_AABBGI owings_r_Page_109.pro
da078662533fee7cde7ed42a49e68eff
f4bc5cc0ef70df3fb47e79e81f0bbf7469243950
37106 F20110112_AABBFT owings_r_Page_040.pro
a12ff6829a911dce2f0813048b28948c
0697ae274ab0446ac5b35b931c2ed4fdd870bc9f
85655 F20110112_AABCIV owings_r_Page_172.jp2
a16824f1551b569fd342a71af43bda27
05fb3f9f8d4ccc1cb47150813de71841dc88844e
75658 F20110112_AABCJL owings_r_Page_194.jp2
1161fcdebc8f96e87b71c94111da2fba
9a078a80aeb1a1db0a1a1ce63aa0624320e35f2f
74928 F20110112_AABBGJ owings_r_Page_016.pro
5308145d52f2fb1df4411f291d765994
22586e2f5116db8b76221ab4dcfe3c541a12914b
46757 F20110112_AABBFU owings_r_Page_141thm.jpg
be1bb40753e7e851aaf848a5f9de479c
a54969fa1b4bd0b76142655ea8db851aab7dfab4
91842 F20110112_AABCIW owings_r_Page_173.jp2
23c4cacfaea280c649e266882dcd6b59
c6f6ec0582bc91e387c80b62bccc7eea8646e76b
F20110112_AABCKA owings_r_Page_006.tif
b8913621ed376e26ab7fb66c4cc3e25b
67b3fdf3f0f6fd03e3f14f4d280d15683749fcd4
755715 F20110112_AABCJM owings_r_Page_195.jp2
9de7d05dbd0b8ddffb63802346768679
d6bf9ced14c3aa9b3be6dd7075b893ea0d199a32
72020 F20110112_AABBGK owings_r_Page_170.QC.jpg
314620e489bdc1253e332c95fe6f9ec6
8639b348ce7316ff082480f9f7e0a847b6d1137f
F20110112_AABBFV owings_r_Page_085.tif
97575c29f6c925b62333fa4148b45265
788315de025c46477a04179727fbdeb5d020c17d
838237 F20110112_AABCIX owings_r_Page_174.jp2
65af475cf976e79ccfab733615f3c38a
cfa9c95f70a964534c0ac6d3159868c5add35a59
F20110112_AABCKB owings_r_Page_007.tif
23a5904f67a6f452791c4baf27578c44
b7303381f900d5ec35628af91e9e024a4bd64281
738987 F20110112_AABCJN owings_r_Page_196.jp2
b521e7e3138ec1ec1b926b6a0338f7e3
02f85d131087de234db0094d589b8d419bf298d5
209736 F20110112_AABBGL owings_r_Page_037.jpg
4326c1745d00b1f9c84b3b3f200dd9c1
4216d9ae3a3b2e28904580e7295dfe77cb931103
F20110112_AABBFW owings_r_Page_071.tif
16a32f0bdce26013534b3cc0df2f4793
bc84a36c1d3bb09539e6808e6a423d28d8fa20ec
F20110112_AABCIY owings_r_Page_176.jp2
9d84e7f1dc5a98ce29b23053b0f1b1b3
6f6daab12636c5b3f2c1b9f51e933cbed4266ea2
F20110112_AABCKC owings_r_Page_009.tif
311e2ad93c1cc86f13f2d69d0e404d72
9b4d548546719a5e61baec22f3620400cd3e7ede
F20110112_AABCJO owings_r_Page_197.jp2
2e00682aa33b1a2ed6fde3b6f39fbcb6
b796559d0a49c63daa06d86a99807a81260eb8e3
23443 F20110112_AABBHA owings_r_Page_156thm.jpg
d5eed0fec69a39db96e3be5d2e2ac436
78e34a9e728a00bf03fdd268be60708818b4d89d
2046 F20110112_AABBGM owings_r_Page_156.txt
5e04e07360e4ac3ca722b0275766e170
0f91b722940ff0b59c07b3ef31ba074f6b267757
225359 F20110112_AABBFX owings_r_Page_061.jpg
f48bcd972d5de99238ce5c78731085fc
073c8e0182b3ab7f71a1d71315f7c4bc25ea3554
800118 F20110112_AABCIZ owings_r_Page_177.jp2
3659a26fb13faa466aae2fedb20566d0
cd5cb5fd19ba36d282627097602dbd1bcc19f6b9
F20110112_AABCKD owings_r_Page_010.tif
a63d4bcdebf59cb5195b1de9765638d1
b033541f52a6a906e810b09d92fc5c0c1eafebf5
22533 F20110112_AABBHB owings_r_Page_063thm.jpg
5dd4ba946d46219a0a36521509fb23f0
e087de5275b2afaa1569122d688f76cec1b08b6b
727068 F20110112_AABBFY owings_r_Page_179.jp2
e7674871ec45e1499e049065ebd65289
a5e19dbb96c686155f16e9b9ac64a117f9a3a5a3
F20110112_AABCKE owings_r_Page_011.tif
315bf97a24e734ee57662568659b5192
9a06040043514967956cb21477c6b8f6fe5c5636
102240 F20110112_AABCJP owings_r_Page_199.jp2
95964d16ca18d2c5d88bcddf8d59115c
e6eb89867b2b836ef150ed83796220767dd3e5a3
198549 F20110112_AABBHC owings_r_Page_120.jpg
28f4f911196eb97ce5c2711ada38b0b3
f1f58c35e6aa39cacfa845c472e0c1dfc907a239
108627 F20110112_AABBGN owings_r_Page_103.QC.jpg
a7fdedf8ec27b6293285811c0e7c7a31
4538b4bd846861b3c9e34a0481d06f0d854a77ba
87241 F20110112_AABBFZ owings_r_Page_083.QC.jpg
0dc5a4dd2d6d3086f160a891995870c7
e1a5116635b0772a5632781c53f34ccd0960149b
F20110112_AABCKF owings_r_Page_012.tif
89e69a6d93cdacf46d67859d5266aa50
dcb16b0b6b58e10308814f1427bcabdc8995d09f
127770 F20110112_AABCJQ owings_r_Page_200.jp2
3ebdbad1a32c74577b2fad61dc143ba4
e66c0960b997c846218a5966951be77b55a52443
1765 F20110112_AABBHD owings_r_Page_136.txt
606a8ce3615a6817204bc5f3fec21da0
97f6deba0bb723f3ad26cb14eb383546e0076f30
22261 F20110112_AABBGO owings_r_Page_171thm.jpg
ee8ea0aae347ddc694ccd3e948b6541f
4ed47e8a1d3b694e326ed3f13cff0388ad48120e
F20110112_AABCKG owings_r_Page_013.tif
9cd8fbc881455dae0772e902f710cc68
586db5eaddd9442ddb1e6b1b54ff0cc23ed3f43c
F20110112_AABCJR owings_r_Page_201.jp2
bc05e5bcd1b23abfc5e0609d9a026353
70a3f64ce3650207d3d3bc9a4d5271cf9964fb2d
72202 F20110112_AABBHE owings_r_Page_032.QC.jpg
b3d25730d534c7d3433c95e6655999b1
5ea35888ac5a243092169c22f2bdfa173f4fb0d6
12251 F20110112_AABBGP owings_r_Page_178.pro
ac18619630ffd93becca321ce8926ecf
441ac31826e8f29a79d6aef78c2384e1780f8c67
F20110112_AABCKH owings_r_Page_014.tif
f55980848b5d391ce60cc0bbabc2082f
8485e627f40e451bb18721f364b57b99a31c06f1
132256 F20110112_AABCJS owings_r_Page_202.jp2
dfa4e70d842b7c7c21638fe19603fa3a
4329b6f14c2b97b621e4738f8d981bf8cd6a5a5e
161624 F20110112_AABBHF owings_r_Page_076.jpg
04cefa7475636ffb4122851e052d277b
4fdbabb091bb80fc0298600fcc01a0353d67c939
221297 F20110112_AABBGQ owings_r_Page_163.jpg
48c727a89f16b9c76f93f814a6e7576b
59dfee70a9a2cc6be43ee41bffc26c9a70916f6a
F20110112_AABCKI owings_r_Page_016.tif
930b005533549b117929ca2c333a4ad8
54c7dda882d25dd5e61459e8f0acfb60801e1b67
131313 F20110112_AABCJT owings_r_Page_203.jp2
0bd858119977b584fb4b1a4af1982553
a54892cedc9ac82836e685962c16ae5e1dbee411
F20110112_AABBHG owings_r_Page_131.tif
be4b0b76ac509f8d5277c21ba62a709d
8e70dd7f1b56116f54b64352375846336a56e897
47612 F20110112_AABBGR owings_r_Page_062thm.jpg
ea4a5ce29cbbc8ad77c2d85750b2120f
7a44627e128d840d2921b4e2c565dfd4dcc2691e
F20110112_AABCKJ owings_r_Page_017.tif
d77cff9a0fcccd3239d2a285632f1c1f
56422c00f85bc45515e5d71483e5ec327fdbf171
119552 F20110112_AABCJU owings_r_Page_204.jp2
66abd5605c831ff63f08b6ce67ffe7f4
382091ac78f0639e0038acb36d6c09876cc59d73
40003 F20110112_AABBHH owings_r_Page_101.pro
d530165a583798cd92fa310fd8a49f2f
e248916f3fa5b2417669d800e7685ddc3ecaf9bc
22871 F20110112_AABBGS owings_r_Page_073thm.jpg
bac53006650c6905085dd5b82012ef32
64c76a1d17d97321957ce31f868a1f743aa79c2d
F20110112_AABCKK owings_r_Page_018.tif
48fc7a0fb2542f552b34235cd8734c3a
3b65aded7a63013757e30ff8aa537f67f52ae4a4
132919 F20110112_AABCJV owings_r_Page_206.jp2
a111da41cb233bc93811f2f0dd29b515
281a37d4a84f271932d2b2a44b552709bce5f14c
F20110112_AABBHI owings_r_Page_072.txt
05eff9a6ff9252e1c880b68ae8f266e0
33e35014c363c12dc183f0cfa9140c44b72edcbd
93529 F20110112_AABBGT owings_r_Page_072.QC.jpg
ee5e2d1d92404203e58a7a1f0613dc8a
7f80d0f5c5e3c5b509e07882656dbbf1c4c5dfd6
F20110112_AABCKL owings_r_Page_019.tif
1d4855d0cdac0fb5cbf9c3cb30b2085f
579da3c1d7c1439c3536871349310e8cc69fedd3
125628 F20110112_AABCJW owings_r_Page_207.jp2
28099207287851e39f8010c9ff81a9b3
3a3ff166f939d895148f7d8530e75581ab247f5a
76253 F20110112_AABBHJ owings_r_Page_156.QC.jpg
780702eb84b5828882ca372a4f24e48e
fa7de1bd91c70c2f153ac54ac22345fc877df7eb
25628 F20110112_AABBGU owings_r_Page_094thm.jpg
65d825a7731d26f359ade37711c78f29
15d3e21ca73c5186270bda3119f8e56555779b30
F20110112_AABCLA owings_r_Page_042.tif
75d15339ffb519755ddcbbd15603fef5
7146878bc3fe329c57c6cfb10e0d1c60b1cecef3
F20110112_AABCKM owings_r_Page_021.tif
7c0a535e9518e254e79d8cbc4c2a9540
47c7b13935dafc4534f14bcf69b1e7ece034905d
128412 F20110112_AABCJX owings_r_Page_208.jp2
abb7add65780014d948e94c5824ce9f5
d9a9fd80e1f983a370800964096cce865d1a20b2
1051940 F20110112_AABBHK owings_r_Page_103.jp2
388b70ba36de1f0d3375c23b7ab9aa7c
fbbf1dd96a5f35fbc3c00648afb27119ce4b3533
3061 F20110112_AABBGV owings_r_Page_014.txt
7893a5243a846ac46bd0aa46a5aa293a
c88158986b3fbfed65ca1198c863377f57caab9c
F20110112_AABCLB owings_r_Page_046.tif
31e8a436f2182b2e4b32da8d843986cb
dd7fb74c6b62d2a045866b7be195fb860e5c185b
F20110112_AABCKN owings_r_Page_022.tif
4e67178136b26605352728d811afc078
f74424624cc1906166e5e2aaabf1922f5d29800d
46149 F20110112_AABCJY owings_r_Page_210.jp2
344135b45c666864e11e78abdb7399a3
70a878c0242b14f7393be8444fa38053c048d9ac
33682 F20110112_AABBHL owings_r_Page_210.QC.jpg
7374d13d36cf97fa6e96f1a5eeb40d8a
7cfd436a1abe77f9b71d9eda355bc1a5dc99df84
19501 F20110112_AABBGW owings_r_Page_179.pro
2890d211dec28c28568371938bc1da86
8c71e0327f188322c6ead7fcb9a160da347186fc
F20110112_AABCLC owings_r_Page_048.tif
9256e3444feeb304b808aa65e9de6886
cbc3abde7234e15b912131b0195145ea796c3603
F20110112_AABCKO owings_r_Page_023.tif
2414a8fa0c2010329f6c378fdfc8948c
2196b633827581144f2d9c75983af2a9de964321
F20110112_AABCJZ owings_r_Page_001.tif
57873adf30a8e78388151351fac94424
7492a8fb0994eaed911afd82dcadb107245963a7
1675 F20110112_AABBHM owings_r_Page_159.txt
8ee66b5bd2f249a1d70b3673044989a7
7e827ae763c0f4d150ca29aaa131031ed5f1539d
22296 F20110112_AABBGX owings_r_Page_064thm.jpg
1ef5f7264bbef1753015842d7f0e5ec1
aba9275525dc57128aec45d2c038d1cf877313cf
404 F20110112_AABBIA owings_r_Page_001.txt
1e476c8819a99966cb3b0b611af90846
2af9dc3f7f7d82e528f136e367b3833ac0371f6c
F20110112_AABCLD owings_r_Page_050.tif
4267a9a120a6b0688e2a35be84b184bf
1d0f472f8fae87022fadb4a588f661952b44f72f
F20110112_AABCKP owings_r_Page_024.tif
545f0d4c68dca249cbf3b0a3401d4caa
de8c665c10c034d2607586e952a3285b3c4b1385
1032920 F20110112_AABBHN owings_r_Page_068.jp2
d3b6ed3859f72b311b7220f5b8c13828
f56eb72e6bb3308afc4f463da71e4c5fb8ef6fab
1684 F20110112_AABBGY owings_r_Page_018.txt
0529ced235ed340b4137e0be14702a9b
4e2e0c967f7eb5e3b971ef252909bef29ebfbfc1
141046 F20110112_AABBIB owings_r_Page_132.jpg
72afac6ea54e06a7f54e50262448d6f5
310e891068b203ebb8e9e04f581346c3a2332702
F20110112_AABCLE owings_r_Page_051.tif
0d15972d295b57e9d702888b6a25b66f
c9a7139d49fdd0f41cbaabb32a3559917afd5539
F20110112_AABBGZ owings_r_Page_020.tif
c9d78afd069a0124255beb6911c0cf42
14285eda75a2a5b00124ca1c255ef77a4b27d0df
48007 F20110112_AABBIC owings_r_Page_139thm.jpg
e01187d37e26b465994e322356644952
2a85d892145dbe71c1bbcb6d0d6b59113dc458e2
F20110112_AABCLF owings_r_Page_053.tif
a13b2e3847f2b081a8cb7977ffadf180
f900012b299f815dd0aea29535d5c039703b51b4
F20110112_AABCKQ owings_r_Page_025.tif
070f9c82937de0e11f57469ab7014eb4
0502d8cf7035f4d7f5b0697edaf8b894d22099eb
F20110112_AABBHO owings_r_Page_057.tif
2a80b69e23e95c89bed158ca5d552b72
51b384e1e7d6555c2e22a14afa17e56ca21740b8
912549 F20110112_AABBID owings_r_Page_133.jp2
7bbe98c82c6a5500e1299ce7086a3666
c98f46d42f525373dd5524a10fbba3749e35209e
F20110112_AABCLG owings_r_Page_055.tif
9b5ad9ed2c4a77f28091d61602fed358
4fa077582db83d9f4b236b3e83aa4323fe57b699
F20110112_AABCKR owings_r_Page_028.tif
83082c912b172385858acc8e30ac658e
898732e63af4d09d08072835476a8535218411a5
1931 F20110112_AABBHP owings_r_Page_146.txt
051462e00e838facc5e66c7b7b076e0c
b13012c7cc75b5a495fed31bb799c981b8f7dee9
168371 F20110112_AABBIE owings_r_Page_018.jpg
d5cef8cbc436ac6b658db5736643c3f5
caef5cc25185d28e485e5ebb2d16ee1511586d92
F20110112_AABCLH owings_r_Page_056.tif
401ec0a1e7d9df0cdb66ac575f74fd4f
7dd7a84501dd01c0cd6e65bc590fd3c89ec4111c
F20110112_AABCKS owings_r_Page_029.tif
4b962f0889f2fe9eeb89b3c0dd6a45b8
9e5a9f3db026b17942bccb2b53d4569779ec324c
F20110112_AABBHQ owings_r_Page_191.tif
0d2d14152187e81af1094d33af3b7aca
d22a301d190b53c4d039b35042f74f4f203d515e
71745 F20110112_AABBIF owings_r_Page_045.QC.jpg
06f8d572b4d25e418f9c9c4d9d9a90b2
0fbf6004a42dd8863f3be3da5d0f2a554e68b2df
F20110112_AABCLI owings_r_Page_059.tif
7e7848487c8be3cf875173928e28227e
532638c64f426664886fecaac3ee736b8ceb5218
F20110112_AABCKT owings_r_Page_032.tif
d91c24693b70ecbc397924ff731632a6
3d649adb58209866d6539e19ee50dd1456f8416a
F20110112_AABBHR owings_r_Page_106.tif
fd2e0a673b144b26ccf54d033fa5cea1
cfc08f3b256dd2c7a0d7c29499ecad1db0a12cbb
F20110112_AABBIG owings_r_Page_185.tif
844caa8683f444d96060f3ea279687b3
9356ea2db106a766da4e704ef931ba2a9dc4a87d
F20110112_AABCLJ owings_r_Page_060.tif
03de47023902ddd461b5b511f7790abb
e7705eaf87457a63fe9f420c1186280b361dbd2a
F20110112_AABCKU owings_r_Page_033.tif
5284544ba37ba49dfefae82b10987a11
cfea7ffeb585a0e4eff2e22d90d36c7e9ee21fe7
240870 F20110112_AABBHS owings_r_Page_048.jpg
7aeb50110bf386f33b8ab9b77bac35ad
7633b3084264f9d95b5f7640c94f97c1f6a3c69a
17907 F20110112_AABBIH owings_r_Page_194thm.jpg
505b2e3eebc573e2dc60a200262ef6cd
0748a530211b5647ae2fe56023bd2567137174a3
F20110112_AABCLK owings_r_Page_063.tif
3cccb7cf5b7efd8c9ea0b5e1eb23911d
4de6bf9d75a777ab92425e9ff49c8fb157b30ec7
F20110112_AABCKV owings_r_Page_035.tif
ef091e9ac5fa1fc4513fa5538eba3735
49436bd531888108b9e295d7a6880f851e35828d
77548 F20110112_AABBHT owings_r_Page_099.QC.jpg
f03f3f6ec2f247d6d001dcc26fe940e8
2106a5aac88719908f0cbf03cc34e7add9b687b9
1040 F20110112_AABBII owings_r_Page_124.txt
b498533a03459e1dc0ef7fd7d47877f4
00247d1ada4046459e6123a1abe65a8a90e6601f
F20110112_AABCLL owings_r_Page_064.tif
52b40b2e9889cd0a3942531b2e1aa707
429f48a927e01253e421411c58f55f2a1c5527f4
F20110112_AABCKW owings_r_Page_036.tif
d30e2f204d2b222067c4fa20c2c053b6
37bc3b98ff5939702a6a7025ffa0de39bb8f2df8
214052 F20110112_AABBHU owings_r_Page_135.jpg
750cdbc10959e73c9ef097fd547a6a8a
44eac702a84f532ec5421fae12243094b605b886
1026328 F20110112_AABBIJ owings_r_Page_040.jp2
565c68d14d38afe744a0c19c78d15b5e
23aaac3ef32943516751b1771c976a4d18276a2a
F20110112_AABCMA owings_r_Page_089.tif
9af4824f5efb1d30d898aa3f39725ff1
2bb57e0d571771df2b27423316f690ca2c1683f6
F20110112_AABCLM owings_r_Page_065.tif
65afb68c2bbcd36e2e83ad03435609d1
ac750d7f29a43c0bfa52735920e90cbeaa168042
F20110112_AABCKX owings_r_Page_037.tif
558d9614c3bd7ea3a663dec2b4771b2b
81365185a134ac02a5bffb6b992490d254e756f3
192789 F20110112_AABBHV owings_r_Page_138.jpg
d90a67a920d5edbbf69735fb51e3c232
48fce921c3a4d27483e9f80e620ff8b86e8d98ca
48085 F20110112_AABBIK owings_r_Page_198thm.jpg
dab91a41f954307435c060b4ae47e3fb
124f04b7b13f79b06f544c6962594c7013bb2941
F20110112_AABCMB owings_r_Page_090.tif
fc5ef7a9811fc8752194c74adba50c4d
b8b11208dae92f8e274afe91abe83264b6b8aeea
F20110112_AABCLN owings_r_Page_066.tif
09c72aebb10629d71253d8283299df24
ff59f1c3bb1557212c445d829b8cd51ef40f43e0
F20110112_AABCKY owings_r_Page_038.tif
d4a79b037ac1bd9a4cf97a349d4dcdfd
b4330c502710a528f199d04dd1166164b8edfba2
36990 F20110112_AABBIL owings_r_Page_194.pro
47803d797ee2ff842b112f965bc5b666
9bb818b767047ba874c0166f8deea79ab0a18490
1051966 F20110112_AABBHW owings_r_Page_114.jp2
fece54146f5a24859345b9f78fc1a360
913f7bd8345e1e94dc313e24631b0b0bd7c7aee8
F20110112_AABCMC owings_r_Page_091.tif
220399789bae7ed3fb39f225dcb36d3b
f829e06b4fbc5f4c574b28434381532c5a700572
F20110112_AABCLO owings_r_Page_067.tif
9f9f759824abf9d79480bbaf2a3da0bd
2cd0df354711415f1509465bceba99a39e56ac90
F20110112_AABCKZ owings_r_Page_041.tif
7c1a0c32d6337b4c42dd19ce370839f6
e4f9301d8902eefebe9b061480e5c366152775ca
38610 F20110112_AABBJA owings_r_Page_155.pro
1e7cba1d6ca86cd26e433d6598767a52
da56f7b09716d73e276045127f9c6e1f78de8913
1367 F20110112_AABBIM owings_r_Page_151.txt
fe472bc1aa4917000a3b7d3ad0fc3c5c
b138c8c42506404b3a62c0e04defa2a765967ce1
F20110112_AABBHX owings_r_Page_049.tif
7aa252139bf0c674b976b19dc45b2462
01e959b14d2d1c53b3c401dd46127dc17b0390e4
F20110112_AABCMD owings_r_Page_093.tif
db52e9d9ce1dc25cb552c7a19141829b
429043c231e38ada3caa89d611febc2eec790a81
F20110112_AABCLP owings_r_Page_072.tif
5908c7d03b6a481c2d7e7d14951f57d0
05fe1cc31644f21379ad9f36b58c715cb18ed9ed
23163 F20110112_AABBJB owings_r_Page_167thm.jpg
b4109a3898116208ad54f5fb7d6a9583
7df8ce0a50769cf9fea38917dd986bca895d03ac
94033 F20110112_AABBIN owings_r_Page_052.QC.jpg
00b326a446aae5a663adc71f7a5a9b1c
c86e9ccc50187811d0996c89ac7c5bdbce015bde
F20110112_AABBHY owings_r_Page_003.tif
a4439d6a18122603b0fae3c40a6ab6b4
039c76b3f0aee748229c95c3d6ce3170ebaaa5c7
F20110112_AABCME owings_r_Page_094.tif
f2caea1d4479dd975e26becf80274ba6
f648f7888552c378a3ddf1ebccae171fed7d7102
F20110112_AABCLQ owings_r_Page_073.tif
2f2ddd66cdade73d021ae10a178c8850
bf716334c04463327de096a28a1ea8563ebd209f
19731 F20110112_AABBJC owings_r_Page_155thm.jpg
0f4d2a4743515f2d0def2c1e5a9ed5d0
b96cf9319d3f039e39eab78612b1f01dfefa3437
107889 F20110112_AABBIO owings_r_Page_007.pro
234ecd06bf1e57d007413735194ee0f7
52f8425e7410d7910c72169b09507b69234d2b62
247934 F20110112_AABBHZ owings_r_Page_208.jpg
1ab760322083fc36213e04fa6c4ca28d
8e82d2fc761b5c907a0197d6543098beb2a11e3c
F20110112_AABCMF owings_r_Page_095.tif
c1777350d9dc162a5d248be6b222eea9
411bd2b066a2dd5b0eae57f76501669f9aa57f88
70615 F20110112_AABBJD owings_r_Page_100.QC.jpg
44f3513e6234ed76a4fe411f75dfc567
d0bc3e5afbca08eca61ea1da04ebf93cda6fcb7f
F20110112_AABCMG owings_r_Page_101.tif
68adb62652c8ffb8777fb12bab5c354c
848cd5c943015875cfb39a3fbca8a1b71cd99943
F20110112_AABCLR owings_r_Page_076.tif
b5aefac1b3e5d28ce9ae146823f969c1
cbb863f50c055f906ff5bb9ca65b4bf25847a088
1051959 F20110112_AABBJE owings_r_Page_081.jp2
29eea39ac0e825cf39318e9befaccc8c
f89a0a035523982bba9b5dd320cf3bebb1037528
46627 F20110112_AABBIP owings_r_Page_040thm.jpg
009001a6c1768792d3ac1c9f3dfea21d
179a3a7c12f3326c37242230dc41d521a01af5ee
F20110112_AABCMH owings_r_Page_103.tif
e64b77ba9280db447c008abf095ebbbd
1483b4cee45fa672712c2e76b05254edf960c681
F20110112_AABCLS owings_r_Page_078.tif
9721980e5602a3d6d7af1121d50ed68e
ecb2763c64d09c188ea06c453ad61a76d8871ebb
F20110112_AABBJF owings_r_Page_005.tif
0ba916d0b49bde3f5820e7b50ef21ce7
daa811e85a9d214037e26fd4e24ff6e384e20469
236271 F20110112_AABBIQ owings_r_Page_071.jpg
496a7a58719819c3a80b137aff0d1a50
61c9011027f5790c5ac78277bdcb1ed4fb92b1dc
F20110112_AABCMI owings_r_Page_105.tif
212b2613c7182562ed4fb1ae1194a45a
c40fcd6ddb32e3f4f2119bb2d491deddb709ac23
F20110112_AABCLT owings_r_Page_079.tif
dea049aa9f25c4e419af31b3ff53a40a
5bd610514997e6b2bd1f1d7938442d898ec26c84
106328 F20110112_AABBJG owings_r_Page_123.jp2
6084a3c4ecbc9aa63272a39b1e8f4fd1
8f7a3dad5f40ad0233f4a86591e8bece89b36bbb
1689 F20110112_AABBIR owings_r_Page_135.txt
75865cb9b02f5c9df7390446771d08c3
8c1b9e1d1ba630f02cc523fea8780601797ae96b
F20110112_AABCMJ owings_r_Page_107.tif
8ee5d8418ebdace02d34ba21621ea9a2
acf2954d82cf3e8dd796a8d012fe23c4f7ccef2b
F20110112_AABCLU owings_r_Page_081.tif
39be17192acc4dade964d878bc83fb92
b2eb1c842f37947b3d29609e9072b1d250602c42
122 F20110112_AABBJH owings_r_Page_002.txt
bf91942e9cd53a96a1d2100cdb6b51ab
052a637ba55829754a671cb480ccafc6eea4e14b
47066 F20110112_AABBIS owings_r_Page_089thm.jpg
f85694b902865681f5d4a5efe68a50f4
d6e14676c9e095467bb4f31c8e81686c80b42cbd
F20110112_AABCMK owings_r_Page_110.tif
54d32cef3261b416a8c024d5bda3b8b5
ed2485c5b324fcc515e48ee8e1b2d53be9ea97e6
F20110112_AABCLV owings_r_Page_082.tif
e729010224700d52aab58e2f456cd7d0
18391ed7c7b537eea763bafa7649cbe0b3f9ac34
2522 F20110112_AABBJI owings_r_Page_208.txt
9e6bed0df6fe418b63d95b9790bab35d
2c725b8fa9a00824da5775d90e2b2ff947e22e7f
F20110112_AABBIT owings_r_Page_027.tif
9a10ae854e7d949d08d8759eff59e4cd
dac5066e75547c48bfec5651a569ed8512eac632
F20110112_AABCML owings_r_Page_111.tif
d87c4d6b679caf886a69bf80b9f71e49
4d0c085cae2e2dcdde9387fa5ab3c00b8ace273e
F20110112_AABCLW owings_r_Page_083.tif
acc27812f28f817574c7c25b249beaa6
e0bc9d80b69885700fabfc299d1b6d0d98025706
11631 F20110112_AABBJJ owings_r_Page_181thm.jpg
87743aa3052e8d4e5307c3d6811e80cb
3b4c539480969a8ef14eb0f6cfee7ba3e6e61de8
218786 F20110112_AABBIU owings_r_Page_066.jpg
7042e63d363edc3e6a20bd1feaf23411
d017e1f3871444d1febc4e63af07a7f512301be8
F20110112_AABCNA owings_r_Page_130.tif
3b5e3d9f36b2065b39de30bc92623cc2
0ef88e7661648f84d381130f900dfe4dbda92d7f
F20110112_AABCMM owings_r_Page_112.tif
48fbe3cb7b424bec2afe7aeaad9bf768
79453c4026d29ed1b29c3fe35e7376361f8b7b0f
F20110112_AABCLX owings_r_Page_084.tif
70403d741a4cc7674f41de83141df45f
e2f9d78c314cce721df758e0a0eee7848e9297fc
110241 F20110112_AABBJK owings_r_Page_005.jp2
7c95f0f854eef2866774f649bc824647
882ca41350d35eacc40074aa2813686ce18e67c7
F20110112_AABBIV owings_r_Page_163.tif
821d6b9308bfeaf14b057756c5a4665e
35eda0f0051b88d4efaa6f8b8519cf77189aca92
F20110112_AABCNB owings_r_Page_132.tif
ecc4c9ba98f69491711d88b46c5f1fe2
2381dedf15810876feb889af4be07e1eaaee85fb
F20110112_AABCMN owings_r_Page_113.tif
fa0f52f45ac982112df55196088e2a3e
ec5d5472faa98e069f234b45b52bacce796bded5
F20110112_AABCLY owings_r_Page_086.tif
c437114bd13000db7bbf53b7ba8d273a
41d82144e44ea7ef7d324ed76416bb1fe75243af
F20110112_AABBJL owings_r_Page_205.tif
cbea0459c31333b68a6bd90715e59184
f7daba929a469b289148ade08c1bd6e589eae5f2
1851 F20110112_AABBIW owings_r_Page_104.txt
b75c9fa52c051f695262e7bb256dc0be
38af9dac04ccbaf3dbfb37bed0500cec84c2a334
F20110112_AABCNC owings_r_Page_133.tif
ffc5c1abc30217bef6c886de27ae17c5
9a6c073c71c41d32e981d38a8c7a40f575138a8a
F20110112_AABCMO owings_r_Page_114.tif
e2464bf1213abbb4daa618255e254169
c0baf53175500a34ec5942ff8f06e60772f53164
F20110112_AABCLZ owings_r_Page_087.tif
df5f81a1eda60df3edae4d6aab04e660
737388b3c79452ea58aa345b4069a81b845fde31
218347 F20110112_AABBJM owings_r_Page_101.jpg
499da9df32d9b2728da706e1ed82a1b3
57aab2df041b679f83a34339fea76b30b8783b52
238986 F20110112_AABBIX owings_r_Page_114.jpg
dd9c055289bc4f5e9ec312ae9c9b773e
0326c8a4d466e27afd598f3474b53ca5a4bc0046
34807 F20110112_AABBKA owings_r_Page_141.pro
638ffba960ccc6448f1693d9a7518081
98168f1f02879ef88315d7b10b6213693706f457
F20110112_AABCND owings_r_Page_134.tif
dea2e4e7455c714a479bbe862711a09b
c7f874741312b8b712cdb37160717dd9fc8a4ec0
F20110112_AABCMP owings_r_Page_115.tif
9e3ff529fbfb079fae198677f8c9b631
3871b1248ef4893c93a567c85a066911d4b99bb3
F20110112_AABBJN owings_r_Page_210.tif
358b822a3628f7b3338c5ae5c1267c50
cfe8207da903c7613033cc5e9f38e5ac7d036f4c
2575 F20110112_AABBIY owings_r_Page_205.txt
118db8769150f5ce87c4cf712953e380
f3e0eee8969aaa670ccf2abe14fa9329bf40e7ba
45295 F20110112_AABBKB owings_r_Page_088.pro
9ac4d7d1cedd7a572592a536e3ccefec
e874ac5f3c99df701cec932511aa2429adbacfe5
F20110112_AABCNE owings_r_Page_135.tif
214d7d6cc4e9cb17cee8a67b74803cc1
95155e810862e5fd19eedc9dc2052ddc2a2a034f
F20110112_AABCMQ owings_r_Page_117.tif
c2e97a82c8be5b994f2a29ff2fb517ed
30ea58c23484b511766749a28f9f15d6ad7a7345
237543 F20110112_AABBJO owings_r_Page_197.jpg
a2a1f6567eafab2c8d32702e2754715e
153641323735ea51ed51acd931ab5ef682a7af18
103802 F20110112_AABBIZ owings_r_Page_120.jp2
db98a9a866ce55ffdc066fc81f8a90de
cfe76b2d25220d895382999ea8c68b641f5a207d
15075 F20110112_AABBKC owings_r_Page_147.jp2
89c818533159f3742e3ee7663c64641b
54750c0c9ccebecaf128e321db0d470ca2fa0fd3
F20110112_AABCNF owings_r_Page_136.tif
28368b5142c034a8554c8cf2ce001c24
bccce0dbad34e92426588a5d04482b5e565e3057
F20110112_AABCMR owings_r_Page_118.tif
20fc863958223d3b87ff6e8d2ca2b98d
b175dbfbfbd1b9472543e3df7f339bc82b4a7024
47613 F20110112_AABBJP owings_r_Page_066thm.jpg
f4ebbe7a873ba77362166df6bbb23825
61f2e606482e825866b52db8f2f18d6dd925b133
1750 F20110112_AABBKD owings_r_Page_194.txt
c02a2535f809c4c03fa51942d1ceb77e
51b4e305259397adc064d644e80d4cd9e480c4aa
F20110112_AABCNG owings_r_Page_137.tif
fef66543dfa23abe3b5c398a251684ef
7675b8bcbd765899ae752473f1490071472c05d5
20580 F20110112_AABBKE owings_r_Page_174.pro
5f10c49b0be0defdb85f6c0d07dac878
239d77e05955c9403300ac0d2dc5b69f697b1318
F20110112_AABCNH owings_r_Page_138.tif
c9dcdf6af81b24eb3f85b9360f0ae8f5
d766faf0fe88ef9e11d6da1a1679d2a2848bd8e0
F20110112_AABCMS owings_r_Page_119.tif
ca3da1ad7e601de6b02f0c85a18aa0fa
07bd52b1212b1a0c6ab72dd668590f1b61510ccd
42676 F20110112_AABBJQ owings_r_Page_104.pro
61305505d7d8a403ecf0e19886a7c37f
11388d8828619a0f1c27d46d9aa9e72378855cc7
140649 F20110112_AABBKF owings_r_Page_015.QC.jpg
bdcb08981973e8805e04f12bfab0f923
4e030e06da593445dec511a65e7d7b39765f8ddf
F20110112_AABCNI owings_r_Page_141.tif
9106c0751b099514e51a8fbcc3cf51ae
ff5dc15775807f788cba75043d6350b05e7ce910
F20110112_AABCMT owings_r_Page_121.tif
017cadf71061468c62f2a2c9e78f9b38
a39042e736cace1bb70728b3279ed1dd37504dbb
241520 F20110112_AABBJR owings_r_Page_084.jpg
d25e71ac84b8011b840eb4a94afb939d
31f4efeca780c8620179e8052165e933b6c8aa9e
99378 F20110112_AABBKG owings_r_Page_008.pro
7c5c1eb39b48cb91327a31e19bd05274
1e4fc7578f79073117be3c4de47d3ed8a7fb9e97
F20110112_AABCNJ owings_r_Page_142.tif
02c0950a80c33600356287fa4bd26e69
dfde4ac8a9f52605a9abdc3fcd4353132ab94df2
F20110112_AABCMU owings_r_Page_122.tif
050e5955ad4b4e6cc22e53dc7f39ecdb
904b998c94ce3b23b716bc0a2a439c7baedaa2df
64832 F20110112_AABBJS owings_r_Page_173.QC.jpg
6295824f31a9664d2387e7e614d4848e
4bd87ea1aba4277f6219795d12397fa040bf7a47
49521 F20110112_AABBKH owings_r_Page_160.pro
1bd7060d22e6d84fbf27a61be5e084bb
dbda2e7ae495b203a19e737698ba66ec2129ea87
F20110112_AABCNK owings_r_Page_145.tif
915374e76ede6575e7d2a4101d0da30e
f8f6144c5cc04b0d4e510834fb4bab26f1851dd8
F20110112_AABCMV owings_r_Page_123.tif
3cf2c2f02c03cb32603b8c883d1dd98f
3af93e0474836008d3019c8ff898970bcecca4d7
55215 F20110112_AABBJT owings_r_Page_016thm.jpg
12bb5e3cd0b644a47be066c836165e7a
40a3d169bbae3f9844f0a7ad5dbc270465fe0641
46184 F20110112_AABBKI owings_r_Page_176thm.jpg
56512ab6d8a2f33758ac8c3fcef978bc
31b8144014c1a6be78952e90467e85d337463416
F20110112_AABCNL owings_r_Page_146.tif
d291c63599eb5cd45a356305b2b98949
8462ab6c81a17fb3f2ec654c5b33f98bae227b58
F20110112_AABCMW owings_r_Page_124.tif
8d0cea75998f28152e227c59ea3b2b9f
330e447c3b6d2c9cefef38a64f84b964e9839851
2261 F20110112_AABBJU owings_r_Page_103.txt
490fd5137e94768b30ea3ed523c91530
b2259c15f72b64586b860d3e7143591a4bd9bb1c
42734 F20110112_AABBKJ owings_r_Page_020thm.jpg
0a619c235c5fe735e234306acc2760fe
21c606f4c8fbcb437c3215fe21dc45ca1639691b
F20110112_AABCOA owings_r_Page_167.tif
984f5e3b0b22b2223bfb0d727796d643
ec08979f57da5805574434a43f53ce90a8abf894
F20110112_AABCNM owings_r_Page_148.tif
2a9a70383d393fffcb1d9ba67a4fda31
0454629ac82a1f226e5b43b123e3f0868dd1a289
F20110112_AABCMX owings_r_Page_126.tif
02aa81a8f152cf82f30e4c62ca44bc1f
6d41b539c0a9cfdaa77186edbf4b8c3c4c4e62de
16117 F20110112_AABBJV owings_r_Page_002.jpg
0ff4387531ec64cf35c3d32c2397819f
17df6504221a07c79e63001dd1e4524db62c0910
89582 F20110112_AABBKK owings_r_Page_181.jpg
4083ee38463ab758c08974b879de597c
ab1de991232059a1c99e8f86bf8428e139eef690
F20110112_AABCOB owings_r_Page_168.tif
b204e9cf25548d14fa64636f73c85930
992e81bba8811b67a348dd23d97ba2d020d40a99
F20110112_AABCNN owings_r_Page_150.tif
070ed7406f9f88b47fb506494074cc4e
878055759d5dc6452a37530e8e12e0f335e51017
F20110112_AABCMY owings_r_Page_127.tif
d0dc392aa89586085466deda87ebfab5
f7e9fc75473613dc0e6a77c0d608c2558ce3cff8
109439 F20110112_AABBJW owings_r_Page_142.jp2
d1947b2bcfc93c0848c657ba9d46e7d5
1521d83718fee5f4811035e5d9909b289781324e
48430 F20110112_AABBKL owings_r_Page_093thm.jpg
2ff303364bd2b2b7175a39f2dbd02a5c
af82c3b7c9f55d746090f33b7584f65541dfa2b3
F20110112_AABCOC owings_r_Page_169.tif
4f95ff4dddf7b2a8d490f9f17edbd8c1
aa7d16bad7812620c9858b5ac09314aede88b356
F20110112_AABCNO owings_r_Page_152.tif
a30368aca1030fc72a483fc8c3f462b0
375ab7936888c1d570f8eed94a9d1b9dac3f7127
F20110112_AABCMZ owings_r_Page_128.tif
b3b044ffed29a88c8aee96550d74d263
7c4ce1264bd4efb33e282f58b8a54ae73ecaf2f7
54017 F20110112_AABBJX owings_r_Page_029.pro
f0da5c976f3d3b30610f32cfe8596435
1c8b86df3b48f0ffbee7800b5daacae0c646ac4c
F20110112_AABBLA owings_r_Page_039.tif
bd05125643af1272843e2c1b60f796a7
b97ffe951d611c6bce515cb7e6832e12ca99b303
979158 F20110112_AABBKM owings_r_Page_083.jp2
693a89eea4847b305dc6f13e9e00e3f6
930c7dbe60bc894474f4e0ee0c784fbf148f947a
F20110112_AABCOD owings_r_Page_171.tif
d148e4e51bbc4e562441d0affc4c25f9
ddf3ccb66854fb0528bd55bfaa6d4034bb06ae7b
F20110112_AABCNP owings_r_Page_153.tif
cadf3fb0c5c5d25d699d7fe85943a8db
76c795c5c0e2f11030ac76c535dbedcbff84d6ff
F20110112_AABBJY owings_r_Page_097.tif
9abf3015f0fe4ce3a3aaa6c89112f502
1564f113c8406102e8c1982a1bc3e541b1f06572
218993 F20110112_AABBLB owings_r_Page_021.jpg
c434ec8029431ded4340c19cb37d61de
b6e69d0411f2f59c8ef19b96616467d757f2d70d
206960 F20110112_AABBKN owings_r_Page_169.jpg
154087e64a36213d782b619544487064
10b128eb21a064bcb8f2df52262d8593841c5188
F20110112_AABCOE owings_r_Page_172.tif
91070081f6b61e794e8fd5354c6abb7d
a99dd27b5531b26f7418f298cc37c4dd59bd92de
F20110112_AABCNQ owings_r_Page_155.tif
d1bf3ed8a190435cc351aa8c3362ba1a
7508d02cdc92555677dcf7a530f44585ff1ed0da
1357 F20110112_AABBJZ owings_r_Page_158.txt
b4d1e10699a264b296497e16aa7e8cce
7e0f49aa948466a3ada32505e96eee4035e9f609
207593 F20110112_AABBLC owings_r_Page_040.jpg
fa3e2ddb101727ac106c4366307989c0
5b2a626f777d8e500862c64ea447d3d59dc03298
1042936 F20110112_AABBKO owings_r_Page_150.jp2
2fec0735f0511231a1be54abbd010155
06c1a42cd7c93fd80ed1d41bf205f2d74ab6b258
F20110112_AABCOF owings_r_Page_174.tif
5dc412634ee5299c158e99d7659262ae
13b186d4047529168f1f414172dd37fd516846d9
F20110112_AABCNR owings_r_Page_156.tif
1295b6ab488da5c1a42aa6dc14990601
d46d0eb8991a8874875b8dea6b7bab4f25994abc
81134 F20110112_AABBLD owings_r_Page_015.pro
73fecbd82573cf701514ad6df1b0eb58
66e4dc7b690e715c137aa4b533db2e657b226fff
F20110112_AABBKP owings_r_Page_088.tif
5fed483e69e47af187ff6d56a9da969f
fd360587cb93122fa0c75585f57e9d54f37315f3
F20110112_AABCOG owings_r_Page_175.tif
0999310c999be2eced34ac7fe732ed75
4828d731766162f4eae283155d96f66e57448972
F20110112_AABCNS owings_r_Page_157.tif
1512b1ba641fec06552b1eca64967290
ad06ec5ffb2fcbac7f5a1f06117572d28693a369
2007 F20110112_AABBLE owings_r_Page_033.txt
435d501c08e6912e648d4340783e7a43
ace7ba86853fa2e0adc880f691a072374d995aa5
24221 F20110112_AABBKQ owings_r_Page_095thm.jpg
55592bb14d0b1f5dde612dacadd033a8
f5344535dcd27041b7d9c8ffaf3e1ae05e3cc6f2
F20110112_AABCOH owings_r_Page_176.tif
bfe2b68e01b09657a7585c4227666c4c
011d739b314be6fa5e6dab96df7708680ec2b401
35961 F20110112_AABBLF owings_r_Page_065.pro
02d9c0a18dc3d0975da4dff91e6890ed
7a726bae0b4fcb52ff32c9d634fb5d32af519a3e
F20110112_AABCOI owings_r_Page_177.tif
663dfea4424bf0f49303679ebdb55b2c
e5de68c6bf34e38a9ad482aabdf39b28b57135f6
F20110112_AABCNT owings_r_Page_158.tif
794f6f6e2b397b1451c7c2189c16c04b
c83e28ebffa1c0f0cfc95e3614259af75045c2d4
F20110112_AABBLG owings_r_Page_031.tif
4ae4e076f2c2bdcd9635e582df237eb4
ce506ff5fee2eebfa8f3f3f2fbae53527463e97b
24898 F20110112_AABBKR owings_r_Page_137.pro
e79110dca9eff6207de6676dc8e08b61
a05c1e2ff36692e83a301fec1d9644862c4c4f57
F20110112_AABCOJ owings_r_Page_178.tif
1a4e07e28611ab8ae4691b8012a21b0f
bd0a9ce1049cc0cee5e9b6137ce5201b37323f8c
F20110112_AABCNU owings_r_Page_159.tif
b24240a7d08bf944a2d89fe1289e4761
939b1e509395e418fca501afaf3f269caff7e2a0
87550 F20110112_AABBLH owings_r_Page_046.QC.jpg
3a03c01aca18a0c856a1ac4bd95c6aa9
ca266ef2e8be373f2bba51244a76be4c6bf0f338
259989 F20110112_AABBKS owings_r_Page_200.jpg
f248b3cfccc94cca5bc7f4bb6730031c
46372c2f8af5874571558b0bfe663d5473575f5b
F20110112_AABCOK owings_r_Page_180.tif
32fa52cb3d7103b91e28ead1829e0a43
de1757f5b554d8234b4cc4202de91abb7755043c
F20110112_AABCNV owings_r_Page_161.tif
adf3971e32319cbc2e5b9948be4b2b85
f7ba35a30cb7a3571640b7d0128af2e918521c6e
86690 F20110112_AABBLI owings_r_Page_078.QC.jpg
6e96fb12bce3add6e8498a95aa040d72
59f7105813f308be268d985c9b4551894e3f98b8
219879 F20110112_AABBKT owings_r_Page_185.jpg
09dd0ad9c11308387597284d80f9f4a2
669bd8f641619f2b51bcf53529d55fea61632875
F20110112_AABCOL owings_r_Page_181.tif
6ae454c4cca87a308b2033e2014cf001
cf936d7caba389af3c684a6e153b9c5f8f14500a
F20110112_AABCNW owings_r_Page_162.tif
da23a47e8b903cc817dc729d34607ac4
b65082cfade122d31c2cb2e56eb5b936c74ca043
96396 F20110112_AABBLJ owings_r_Page_165.jp2
0fc06accb9c2f9c205005501e635c978
e9ab0adf09d4ba78dada97c79d434a903bc7bb92
75576 F20110112_AABBKU owings_r_Page_161.QC.jpg
a5419acaec8f2919c339fb7c678469b7
6ac2c55970c4e9158cce773baedf39836de95e2e
F20110112_AABCPA owings_r_Page_207.tif
e2e9feaedde900dcfe4299480b5d1820
9a1ca957e46225b932569fd4379f6adc0b3162c5
F20110112_AABCOM owings_r_Page_182.tif
73f1e7c8f18ebde7f20493a2cfd10b89
f41c54c7d8e91a554de3925a76f71752a3a9ed9c
F20110112_AABCNX owings_r_Page_164.tif
b1fb1e2a5a9d65831123e0fe21782560
541f19a466b9574c1137025a525053214bca8d82
75476 F20110112_AABBLK owings_r_Page_120.QC.jpg
da19fe0c9145c2a19771b32645f064c7
2c73c252055644312b4b70ebf1057bafc7e79512
44237 F20110112_AABBKV owings_r_Page_196thm.jpg
a26f561126692837b31dddc8735d6f6b
678a07d729c2078e81e4b02ad78788005d84e2ff
F20110112_AABCPB owings_r_Page_208.tif
f164fd46b768ebf1e3ee9f3ace632375
85e1397de58cc712958dd22c367a3b38cf49cf4c
F20110112_AABCON owings_r_Page_183.tif
beb06310c1b1195c9e41e7dda6185db4
408693bc95623e4ddbb4cf4147124ea8eacb474e
F20110112_AABCNY owings_r_Page_165.tif
4353009f5dfa1c139b865f715bb03be9
60a02e1332c4196ed2b57f8061152ae657ee6755
F20110112_AABBLL owings_r_Page_100.tif
ae0277f0ca164b898d996b9f634a4ffe
fbc6bdf024c72c6a26ae5187d9eae423376a52f9
23408 F20110112_AABBKW owings_r_Page_161thm.jpg
c1f8d2b69751914ba2492d162be43fc1
33689df8dd891c3f58fdaa15487ec6bfb3227d40
F20110112_AABCPC owings_r_Page_209.tif
613ce5e4417ca7bf2a54c06990ce6ae4
012c5d6966a6b4fe659b086202aba72dee622d4c
F20110112_AABCOO owings_r_Page_186.tif
723f6fb85224ee6774119efc5f307129
0a52069741ceabc24c3e706b68929577b1dc337f
F20110112_AABCNZ owings_r_Page_166.tif
45c329283797d93bf484d491fc2afe8f
7bb8ce6623af73c33ddf35545137b611b0a120c5
22944 F20110112_AABBMA owings_r_Page_004thm.jpg
cf6a172400ea807385718d42cc5f80b9
d09203b177fa21d0a589f8b2baba6007e84e6d34
51013 F20110112_AABBLM owings_r_Page_033.pro
a093adca910659a99307089b847b9a1e
2059e69242b4b68d0c3c07bee9ca05eca1ccd67b
19529 F20110112_AABBKX owings_r_Page_199thm.jpg
5f2a3289189c974a4494799c8cd27454
24e162ad163945395602a75e335971cecd221207
7465 F20110112_AABCPD owings_r_Page_001.pro
57aef663e747614d47309c18ddc1464e
06636d88d9d8707e91faad032677e442166345f1
F20110112_AABCOP owings_r_Page_189.tif
41d91ddcaaa4caa4f28c66a524d9fe6d
8ad051f666684396ffacdefa1f46de93f245376f
24223 F20110112_AABBMB owings_r_Page_045thm.jpg
f22af83aa0f2180f9d09dd30127fe691
7a9b237f3a6e86aed4b032f76464f2115d859065
77946 F20110112_AABBLN owings_r_Page_145.QC.jpg
bca99271b1e1bd55f772b4a3f50d077b
78af3db6afa7843565594eddf36e660ce5d908b6
1029057 F20110112_AABBKY owings_r_Page_127.jp2
1e121eca08b61a808ea65294d270a737
74f9089257d6d715bbf14e4a5174f9d019e9baa5
1483 F20110112_AABCPE owings_r_Page_003.pro
641658d6ece602de308355f1ba7f90b7
8ba0d3563d70c78fec1f85e270a3a4e2c4f2e922
F20110112_AABCOQ owings_r_Page_190.tif
6e00bb09d64db30399725e1c55ad3343
2e30b2673adb56fd31acc02adbccdb9ce8df876d
88095 F20110112_AABBMC owings_r_Page_127.QC.jpg
f68d175e61b712d6b5a536dc29e4d1a6
7d09c98bafc21d300a597d79d13e9a82d178bd28
182388 F20110112_AABBLO owings_r_Page_020.jpg
49a89099c08bd046c5f22e23ca9ef328
dec89aaffad47030244297cd11bdc7c4b110f8ae
F20110112_AABBKZ owings_r_Page_102.jp2
17b0ac9186a1d14a72b60405e6dc04cc
47747da35c27cf9139e92b235d3f5d5d39a289a7
45596 F20110112_AABCPF owings_r_Page_004.pro
bdf3b7032ad89152f5c8c15005c97dc5
2b36a429da345de66cf908f96dd1ba6b6490bc9a
F20110112_AABCOR owings_r_Page_192.tif
476e70cc9cbb2fde2147141eb126af94
70f4c37824bbc598d75c64a03c6dcef2c8f08d06
25602 F20110112_AABBMD owings_r_Page_119thm.jpg
1e39c9f75e8fb73f82633c93ecc37fae
520dfbc38afa9ed69b20372d949ee43018e7913c
86493 F20110112_AABBLP owings_r_Page_205.QC.jpg
c726a37e58bdd2081a068c718e84c791
a9275b00bcda8421ad8ee7e5195f9547e8426e41
77614 F20110112_AABCPG owings_r_Page_006.pro
77249ea40a8b1d2481bab85563e3cb5d
973809ac6ab16ae7f360c3ea1f779f4a436a350e
F20110112_AABCOS owings_r_Page_194.tif
f8baea825d2572d629e3ea36864a4552
73838d55ed62244f37ccb182613611099cf3f499
2040 F20110112_AABBME owings_r_Page_023.txt
c0a54c722d0524c7c310719468ad9135
ede5e4fddaf4cddb5e281a1315e8a01cd940f2f0
77898 F20110112_AABBLQ owings_r_Page_019.jp2
7af20f6d6e82eda9c65ce44ebc944503
36bd8f867a280fc162d6e430742b9179b54eadbd
13495 F20110112_AABCPH owings_r_Page_009.pro
7d825c53e3d49bf562672941949d6cb0
e1be5306c5908992f649c62f505a6124155e3f4c
F20110112_AABCOT owings_r_Page_195.tif
6e04cf0e4f4148a0606755e5608bb363
0240a17ac5ca08afa65cf4e6771e73fdf075e808
974 F20110112_AABBMF owings_r_Page_174.txt
7da5537f8816c28bba2cc7a7bf5bbc4c
b9b0a2d14db280745800c1f462fff33712394612
99806 F20110112_AABBLR owings_r_Page_167.jp2
dbbba6cc5f4fc9a967e14f4eadca2568
236778efcbb82d166a9675339bb7dbbb400e9cc3
58429 F20110112_AABCPI owings_r_Page_010.pro
b58df005ccfff795c6776f16538f63fe
98c1f38f5448c14c9d495b8a8df7d12a7a9d5326
219090 F20110112_AABBMG owings_r_Page_136.jpg
dbb3bce66fbb06e9f5aa6368c0eea1b1
ea5bbb5c2800774e167e07a890226198e77a9ba7
17792 F20110112_AABCPJ owings_r_Page_011.pro
c6b8ad498106156752c5e00f73cd77ad
64c17ad00b778b627fbb238a4d0a1d8e9ffc1359
F20110112_AABCOU owings_r_Page_197.tif
6cdc3506ac6e19b4039dc264106db315
e6b4f7b66230ca85a26986e73b752655eecad3b9
F20110112_AABBMH owings_r_Page_199.tif
74e3f2a578b8c72afef667f153a7c0fa
1e42b086e2b7b9dcdbdb826f65fe6ca1444c573c
362717 F20110112_AABBLS owings_r_Page_013.jpg
bdb90312fdd832d0f913413609507369
85023f840313b5ce4f3a5821c3fb2e1ca289dbff
61519 F20110112_AABCPK owings_r_Page_012.pro
04ef70e8fb92b503d7f68e4a30950b98
6384ad35ceefeb985b396f2cb2471b2cd509e534
F20110112_AABCOV owings_r_Page_198.tif
007c8497d209530ee7d63e6c802997f3
9d1a7b5cc8ed55c98b4a1b6340f826541815956e
4088 F20110112_AABBMI owings_r_Page_008.txt
64ca5d3b4f2bcf3b0c95dddcf681d0ea
06206d54e39f76b2e2181e5a1afef0b871798ded
33483 F20110112_AABBLT owings_r_Page_181.QC.jpg
fe20f0f828563a732c23b0887e24015e
946ddd3faadb933144ce356939237bd8e7a26079
80986 F20110112_AABCPL owings_r_Page_013.pro
fd5bcc67ffb9b78b7057e4abd518634d
86fa6c2c9903d7fe555b00cfc74f62cfcb069169
F20110112_AABCOW owings_r_Page_200.tif
4164a028749932c102c5e113c9ad8b94
0dedcb037c9d5c9aa5d09611f3625e2a91ef724c
166803 F20110112_AABBMJ owings_r_Page_158.jpg
1fe60ca39b84e75f3a89fda4beb38a05
08670810acce8c5e0f7fad28a696e9a484e484b3
2026 F20110112_AABBLU owings_r_Page_032.txt
992cff36ab0c39dc69be44568aa644c7
1dfc905030b47e0e22fe9fc1e63573b8a2430aa0
44868 F20110112_AABCQA owings_r_Page_038.pro
e4ffa7377969b96fbbf835a3af39ae9e
a287599487add0f57e5a55ed3fd848fce739c690
77143 F20110112_AABCPM owings_r_Page_014.pro
9d3a043083294d7838c9ac3977cd52e0
82f2a416289a20b166a906b229a8e052584bc85e
F20110112_AABCOX owings_r_Page_201.tif
1001cf66f3b9b71c42e71c10bbc5e831
9d400be78982ee32b5827acaaac00b23781a48bc
767 F20110112_AABBMK owings_r_Page_056.txt
59e0ae530b59162b06ac43d738e7e73e
6fd8a959cbb4d6b485b6cc7ab72628229e9757e5
57711 F20110112_AABBLV owings_r_Page_019.QC.jpg
5df05c22df251714dd61b8e77c702ad0
f7a07e9897adfdd90adaab75da87187c55b96b2c
42353 F20110112_AABCQB owings_r_Page_039.pro
f25decee841815036baa3a6adcc2f595
73509884f9b1e5e80ac55255a85735c0039c7dbe
71098 F20110112_AABCPN owings_r_Page_017.pro
9f55215e011feb409e14bd7e188441ba
85137986b57f147cd8a3cec5ddd46e214bbf41d7
F20110112_AABCOY owings_r_Page_203.tif
e0e9eb263d70e5827b34bd55387d6494
cc800947641551c75c0119e25efe6a8acfe3c557
94527 F20110112_AABBML owings_r_Page_068.QC.jpg
33ac8fad8e0b9335829eeb66f299c8dc
907865787a40cef8e86b9c35824d3ffe71c80f38
F20110112_AABBLW owings_r_Page_154.tif
d0cef181b70a47e91c75897ed1424719
b17ec6003afb75628c85b9b02758e490dd84cf76
47540 F20110112_AABCQC owings_r_Page_041.pro
427e80784ed34eb6d42e3cc82882fb85
954180144ffafe4fea2e60f000e04d24ea951e94
38536 F20110112_AABCPO owings_r_Page_018.pro
b8b09054216c491369a1432c64bfb61d
ae179c9c489344970e05d9e44cf2c09ceb2f47e6
F20110112_AABCOZ owings_r_Page_204.tif
60ae5a53d59667979dfdb2bdf30348bb
dc00d9516fa2a2ca4b48038c717d37ce57ec293c
215607 F20110112_AABBMM owings_r_Page_055.jpg
3604fedc87ea1f0921825708e56b146c
b6a194e2ec5ec50711dd839b5cdf5d3bd8ccab8b
1422 F20110112_AABBLX owings_r_Page_175.txt
aa2789ce469f493ae8a8a5217646781e
51ed91bbd736f05ff0e13df295cb57981ecc87be
67772 F20110112_AABBNA owings_r_Page_064.QC.jpg
032bd559cf031340a6e6eef9f6999a82
e51de636b18aec39092e8d3c1facff1ec5cab61b
44143 F20110112_AABCQD owings_r_Page_042.pro
66f7c010beafd89eeffd4d0eedf190be
f3a39a565e9b36fe9d24f0a6185c7c62f9feacb3
30870 F20110112_AABCPP owings_r_Page_020.pro
433b34657b145b4bf4ce2ff6f1fb1cb6
c2c3546c4fddb44799ba01147eaa63f4cb7e53a1
F20110112_AABBMN owings_r_Page_044.tif
0fa196941fcf0b6efb4b1776504a8e5d
1c061c96de4908cfbd05afe0e16a835a4505b1e2
49308 F20110112_AABBLY owings_r_Page_005.pro
0e8a661075cd811c65d0a78ea401a81c
60c27f20e799aa018bb5689b1eff49302cb9b9dc
48317 F20110112_AABBNB owings_r_Page_175thm.jpg
03d1b2da41f1e0bd25fd9d4e2d048cf7
4af298ce40f30e59045565b5f3b6935e6d631974
50311 F20110112_AABCQE owings_r_Page_043.pro
e9b3d8efdeaf9365aea476da1cf39b57
def6d937762fee4a6438a6986cb57dc0c2ed6eb4
50949 F20110112_AABCPQ owings_r_Page_022.pro
23ebc90a5714c30dc7861aa2f9a055fb
f33ec0336ddba1214aa08a85f4c84dee1d5a9641
1051978 F20110112_AABBMO owings_r_Page_071.jp2
c4a723ac4ea21a13f18554d67567b01d
4711fe44a9cc8c775a34ec167114e8dc05a71ab6
24385 F20110112_AABBLZ owings_r_Page_206thm.jpg
03307bfedb6dbb39cd91951d4049a059
7e5bd8e11a58f3c5cd8e4a1e6e0f334e6200cc69
F20110112_AABBNC owings_r_Page_143.tif
5d2b776f4f72ad9d60aaabfacfc216f1
0eac7d05875e0b9fe631b304fa6e5c04a5489ae4
41352 F20110112_AABCQF owings_r_Page_044.pro
1454b516d07774c050aabdea351b57b3
a1e5bfb51cf0c18454421ed513bb6a5dff435f2c
51004 F20110112_AABCPR owings_r_Page_023.pro
8e948abb2b9f4e2ad2786187a94b0335
079bc21682831f7f28eaf72562a657b5c66e4724
F20110112_AABBMP owings_r_Page_120.tif
3fe7bc7c708b8df7bdef1a2896bfad09
92043e088329ffb557746a4dc45dd171f40dc904
F20110112_AABBND owings_r_Page_008.jp2
941510205fa6eeecc0602bd856cff1a9
39aa325038679329b014ff4d331605ff7347c651
50400 F20110112_AABCQG owings_r_Page_045.pro
eff1223ca658e1b43b0d0234fbb549d3
a77d60ca168c93f90aa0d2e1c4af208d92da437d
18410 F20110112_AABCPS owings_r_Page_024.pro
1cae5734a452977a87f0f79a634413f6
f8da6ffbddcfa2b1ab083acbd0d96ac762ca0831
55073 F20110112_AABBMQ owings_r_Page_204.pro
6ba53e99f3b6b7c4162aadb9f60415f8
052d04a2df1716d2c4f740943d8bb3ef9fe1ca50
F20110112_AABBNE owings_r_Page_104.tif
ce6efe0835299d6361899a24170c3127
b9c87790f03705e5576eec6c194ea2828855dcbc
39471 F20110112_AABCQH owings_r_Page_046.pro
8631f3025896f15bf9ade4ae9da1bfb5
446fe5ef0759dfbb6652f5e86523a3e4a1394c39
42774 F20110112_AABCPT owings_r_Page_025.pro
92c14a2320945be3ee24e4242bddda63
d873d6c2358d2704a39fb9aa235b1dfc801d04c2
196223 F20110112_AABBMR owings_r_Page_127.jpg
0515426644c65d77e7347206f2c54c2a
17f4c5197f42b36c71dd65738f65774c1a27e97a
50510 F20110112_AABBNF owings_r_Page_156.pro
58d02c24f43b307267d41dc07e7eb6e6
92a8a72355859c55041ca9147504812399d455a0
54148 F20110112_AABCQI owings_r_Page_047.pro
439135a243f5ab7b9a9030be15b36a8d
fb48e3dd18beac9a6e35b05279667db2c9b93159
48871 F20110112_AABCPU owings_r_Page_026.pro
c0acdc30908646ce7750c0100224f209
c9979bebc79e31c4fa1bfbfc220f9da9e465c908
116604 F20110112_AABBMS owings_r_Page_061.jp2
71b32a5e5b16cbbe65f0b510694f03e6
c79c1e205fc5d9a6ff757db969a4390ffa5dd088
221399 F20110112_AABBNG owings_r_Page_041.jpg
defe0d20a049bd7e01e0693700a0b308
f1a99bfdeac2e11541f6a3f4837bcc0859b62cb4
42811 F20110112_AABCQJ owings_r_Page_048.pro
6fbaa96bc5fca1a00f4ecdf94c057ff4
bd22d0ca8629711d962e3557914880fd687e3685
F20110112_AABBNH owings_r_Page_188.tif
1a8c37ee4c0d77eb00e2b683363691a7
2fa2bf4fdbcc7ce4e3124f4510c3eb443f42aac0
58546 F20110112_AABCQK owings_r_Page_050.pro
ecfe0e0d230fd83b4634dee189dc4893
7d23cba1453ed656be9a389af037204b7196bf91
35951 F20110112_AABCPV owings_r_Page_027.pro
a21f1b70fef73e0ad08ec5a1f044c273
912c19d9c5917a8578a5dbcbfdbe537c9b61e84a
F20110112_AABBNI owings_r_Page_034.tif
b83d7d8da117a51e8ed8d7e13d52012f
d9821346daa6f9901a0f6386bb9f2fdd2eab40a7
19200 F20110112_AABBMT owings_r_Page_193thm.jpg
3af2680434010f807ee97fdc23d7952c
886dafbd593e303b566315ec6452960552626e37



PAGE 1

POLARONS AND IMPURITIES IN NICKEL COBALT OXIDE By ROBERT REED OWINGS 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 2003

PAGE 2

Copyright 2003 By ROBERT REED OWINGS

PAGE 3

To my lovely sweetheart, companion, and friend

PAGE 4

iv ACKNOWLEDGMENTS Many have contributed to this work in content and by means of support and encouragement. Perhaps the greatest insp iration and motivation for the completion of this work was my wife, Sharee, whose enc ouragement and support was continuous and at times intense as needed. I acknowledge the patience of our two children, Parker and Londyn, who have gone at times without the atten tion they were entitled so that this work could be completed. Paul Holloway, my co mmittee chair, wholeheartedly supported me in my desires to complete this program and offered timely advice and critical analysis as needed. The committee, David Norton, Rolf Hummel, Susan Sinnott, and Arthur Hebard have contributed of their time and for th at I acknowledge them. Greg Exarhos was instrumental in providing experimental guidance for conducting and compiling the research. I would like to acknowledge him as my mentor, for providing perspective, guidance and insight. Chuck Wi ndisch aided in the structural organization and proofread select sections in addition to sharing key results and data from previous studies. I acknowledge him for the significant amount of personal time he spent on reading and discussing section layouts. His patience in training me in the operation of the many lab instruments and fielding my une nding questions was very much appreciated and for that I acknowledge him. Kim Ferris developed the framework for the polaron disorder model, ran the calculation, and participated in many insightful discussions on polarons, conductivity, and life as we know it. His assistance with editing in the final hours of preparation was most appreciated as were his encouraging word s of perspective.

PAGE 5

v Others at the Pacific Northwest Nationa l Laboratory encouraged my progression and contributed by answering my questions and providing data and insight. Special thanks go to Mark Engelhard for XPS data and analysis, Scott Lea for AFM training, Dan Gaspar for SIMS data and analysis, Da ve McCready for XRR and GIXRD data and analysis, and Tim Droubay for UPS data a nd VSM data all conducted at the EMSL facility. I would also like to acknowledge Ma rk Gross, John Johnston, Pete Martin, Dean Matson, Pete Reike, Bill Samuels, Don Stewar t, Rick Williford, and others at PNNL and for their assistance with tec hnical equipment, details, or operations. Maggie PugaLambers collected SIMS data at the Universi ty of Florida Microfabritech facility. Valentin Cracium performed X-ray modeling an d data acquisition at the University of Florida MAIC facility. For their efforts and data provided, I thank them. I wish to extend a special thanks to Ludie Harmon, the administrative professional who, in addition to securing logistical arrangements, provi ded encouragement, plenty of sugar and emotional energy at both high and low tim es making the process more enjoyable. Funding was provided by a fellowship from the University of Florida Alumni Association. This research was suppor ted through the Materials Sciences and Engineering Division of the Office of Basic Energy Sciences, U.S. DOE, and the ARO through DARPA contract AO J209/00. A portion of the research in this work was performed at the W. R. Wiley Environmenta l Molecular Sciences Laboratory, a national scientific user facility spons ored by the U.S. Department of Energy's Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory. Pacific Northwest National Laboratory is operated by Battelle Memorial Institute for DOE under Contract DE-AC06-76RLO 1830.

PAGE 6

vi TABLE OF CONTENTS Page ACKNOWLEDGMENTS................................................................................................IV LIST OF TABLES..............................................................................................................X LIST OF FIGURES.........................................................................................................XII ABSTRACT.................................................................................................................XVIII CHAPTER 1 INTRODUCTION........................................................................................................1 1.1 Introduction.............................................................................................................1 1.2 Organization of the Dissertation.............................................................................4 2 BACKGROUND OF NICKEL COBALT OXIDE......................................................6 2.1 Background of Transparent Oxide Conductors......................................................6 2.1.1 The Basic p-n Junction Diode......................................................................7 2.1.2 N -Type Transparent Conducting Oxides......................................................9 2.1.3 P -Type Transparent Conducting Oxides....................................................12 2.2 Nickel-Cobalt Oxide.............................................................................................15 2.3 Nickel-Cobalt Oxide Conduction Mechanism......................................................24 2.3.1 Fundamentals of Polaron Conduction........................................................24 2.3.2 Fundamentals of Free Carrier Conduction.................................................26 2.3.3 Observed Properties of Free Carriers and Polarons...................................28 2.4 Nickel-Cobalt Oxide Spinel Structure..................................................................32 2.4.1 Spinel Sites.................................................................................................35 2.4.2 Variations of the Spinel Structure..............................................................38 2.4.3 Spinel and Conductivity.............................................................................39 2.5 Summary of Literature Review............................................................................39 3 SPUTTERED NICKEL-COBALT OXIDE...............................................................41 3.1 Introduction...........................................................................................................41 3.2 Film Preparation and Characterization Procedures..............................................42 3.3 Characterization Results and Discussion..............................................................45 3.3.1 Electrical Properties....................................................................................45 3.3.2 Optical Properties.......................................................................................54

PAGE 7

vii 3.3.3 Film Structural Properties...........................................................................59 3.3.4 Post Deposition Heat Treatment.................................................................68 3.4 Summary and Conclusions...................................................................................76 4 THE ROLE OF LITHIUM.........................................................................................78 4.1 Introduction...........................................................................................................78 4.2 Experimental Procedure........................................................................................80 4.2.1 Deposition...................................................................................................80 4.2.2 Characterization..........................................................................................81 4.2.3 Sample Heat Treatment..............................................................................82 4.3 Characterization Results.......................................................................................82 4.3.1 Electrical Properties....................................................................................82 4.3.2 Optical Properties.......................................................................................87 4.3.3 Structural Properties and Composition.......................................................90 4.3.4 Chemical Properties....................................................................................94 4.4 Discussion.............................................................................................................99 4.4.1 Lithium Effects...........................................................................................99 4.4.2 Effects of Heat Treatment........................................................................101 4.5 Conclusions.........................................................................................................104 5 RHODIUM SUBSTITUTION FOR COBALT........................................................106 5.1 Introduction.........................................................................................................106 5.2 Experimental Procedure......................................................................................109 5.2.1 Film Deposition........................................................................................109 5.2.2 Post Deposition Heat Treatment...............................................................111 5.2.3 Characterization........................................................................................111 5.3 Results.................................................................................................................111 5.3.1 Electrical Properties..................................................................................111 5.3.2 Optical Properties.....................................................................................114 5.3.3 Structural Properties.................................................................................116 5.3.4 Chemical Properties..................................................................................119 5.4 Discussion...........................................................................................................122 5.5 Conclusions.........................................................................................................126 6 DISORDER AND POLARON CONDUCTIVITY..................................................129 6.1 Introduction.........................................................................................................129 6.2 Structural Disorder..............................................................................................130 6.3 Cation Charge Disorder......................................................................................134 6.3.1 Introduction..............................................................................................134 6.3.2 Disorder and the Polaron H opping Activation Energy Model.................135 6.4 Resistivity Limit.................................................................................................141 6.5 Conclusions.........................................................................................................145

PAGE 8

viii 7 CONCLUSIONS, FUTURE WORK, AND APPLICATIONS...............................146 7.1 Conclusions.........................................................................................................146 7.1.1 Sputtered Nickel-Cobalt Oxide................................................................147 7.1.2 The Role of Lithium.................................................................................148 7.1.3 Rhodium Substitution for Cobalt..............................................................149 7.1.4 Heat Treatment.........................................................................................151 7.1.5 Disorder and Polaron Conductivity..........................................................152 7.2 Future Work........................................................................................................152 7.3 Applications........................................................................................................154 APPENDIX A DEPOSITION DETAILS.........................................................................................155 A.1 Sputtering...........................................................................................................155 A.1.1 Definition.................................................................................................155 A.1.2 Equipment................................................................................................156 A.1.3 Configuration/Setup.................................................................................157 A.1.4 Target Production....................................................................................160 A.2 Solution Deposition...........................................................................................161 B CHARACTERIZATION METHODS AND DETAILS..........................................163 B.1 Electrical Characterization.................................................................................163 B.2.1 Hall Effect................................................................................................163 B.2.2 van der Pauw............................................................................................164 B.2 Optical Characterization....................................................................................164 B.3.1 FTIR.........................................................................................................164 B.3.2 UVVIS.....................................................................................................165 B.4 Structural Characterization................................................................................165 B.4.1 Profilometry.............................................................................................165 B.4.2 XRD.........................................................................................................167 B.4.3 TEM / STEM...........................................................................................169 B.5 Chemical Characterization.................................................................................169 B.5.1 SIMS........................................................................................................169 B.5.2 UPS..........................................................................................................170 B.5.3 XPS..........................................................................................................170 C CALCULATIONS, EQUATIONS, AND EFFECTS..............................................172 C.1 Unit Cell Specifics............................................................................................172 C.2 Theoretical Film Density...................................................................................173 C.3 Heike’s Rule.......................................................................................................174 C.4 Absorption Coefficient Calculation...................................................................174 C.5 Jahn-Teller Effect...............................................................................................175 C.6 Verway Transition..............................................................................................176

PAGE 9

ix C.7 Nickel-Cobalt Oxide Activation Energy Curves...............................................176 C.8 Lithium in Nickel-Cobalt Oxide Activation Energy Curves.............................178 C.9 Heat Treatment Quench Model..........................................................................179 LIST OF REFERENCES.................................................................................................180 BIOGRAPHICAL SKETCH...........................................................................................191

PAGE 10

x LIST OF TABLES Table page 2-1. Choice of available n -type transparent conductors....................................................10 2-2. Figure-of-merit, absorption coeffici ent and sheet resistance comparison..................11 2-3. Approximate minimum resistivities and plasma wavelengths for some transparent conductors................................................................................................................12 2-4. Electrical properties of CuAlO2, CuGaO2, SrCu2O2, and NiCo2O4 thin films...........13 2-5. Comparison of nickel-cobalt oxid e to nickel oxide and cobalt oxide.......................17 2-6. Observed properties of polaron s compared with free carriers...................................31 2-7. Key characteristics of small polaron hopping for nickel-cobalt oxide......................32 3-1. Film deposition parameter a nd sputtering setup matrix for NixCo3-xO4....................44 3-2. Values of index of refraction for film s and substrates are s hown individually with calculated transmission losses due to in terfaces alone assuming no absorption......57 3-3. Extrapolated band gap values lines in Figure 3-13....................................................59 3-4. Increase in activ ation energies of NixCo3-xO4 from the temperature regions in Figure 3-23 graphed in Figure 3-27.....................................................................................74 4-1. Numerical data from Figur e 4-10 are listed in columns............................................92 4-2. Lattice parameter of Ni0.75Co2.25-zLizO4 as a function of lithium fraction (z) before and after rapid cooling from a 10 minute 375C heat treatment..............................94 5-1. Composition recipe of NiCo2-vRhvO4 used for solution deposited samples............110 6-1. Spinel sites and occupation for polaron density calculation....................................142 6-2. Theoretical value of conductivity a ssuming a polaron density given by the number of hopping polarons in a unit cell with a constant mobility...................................143 B-1. Powder diffraction file number 73-1702 us ed as the standard of comparison for nickel cobalt oxide th in films and powders............................................................167

PAGE 11

xi B-2. X-ray reflectivity (XRR) experi mental system setup parameters...........................168 B-3. Grazing-incidence x-ray diffraction (GIX RD) experimental summary system setup parameters..............................................................................................................168 C-1. Unit cell sites and wei ghts for density calculation..................................................172 C-2. Activation energies of NixCo3-xO4 films from Figure C-4......................................177 C-3. Activation energies of NiCo2-zLizO4 films from solution deposition.....................178 C-4. Activation energies of Ni0.75Co2.25-zLizO4 films from solution deposition.............178

PAGE 12

xii LIST OF FIGURES Figure page 1-1. A basic light emitting thin film stack consists of p-n junction and transparent electrode. ....................................................................................................................1 1-2. A basic light detecting thin film stack device consists of a p-n junction and a transparent electrode..................................................................................................2 2-1. A simple p-n junction diode and accompanying ener gy band structure illustrates the locations of charge carriers a nd direction of charge flow..........................................8 2-2. Basic optoelectronic devices.......................................................................................9 2-3. Visible transmission spectra of nick el-cobalt oxide from solution deposited and sputtered samples.....................................................................................................18 2-4. FTIR transmission spectrum of solution deposited nickel-cobalt oxide thin film. .19 2-5. Seebeck data from nickel-cobalt oxide......................................................................20 2-6. The oxygen 1s binding energy region fr om XPS shows a peak at 531.2 eV that scales with conductivity...........................................................................................21 2-7. Comparison of TCO transparency regions with conductivity disp layed as a function of transmission wavelength......................................................................................22 2-8. Effects of deposition techniques on the resistivity of nickel-cobalt oxide show that sputtered films of similar compositions to solution deposited films have lower resistivities................................................................................................................23 2-9. A bound carrier in a two-dimensional lattice is the enla rged blue atom...................25 2-10. The particle in a box plot of a particle in a one dimensional lattice bound by an infinite energy barrier...............................................................................................27 2-11. A polaron consists of the molecule s involved and the local charge region.............29 2-12. green tetrahedral cations (A), yellow octahedral cations (B), and black oxygen anions (O) form the spinel unit cel l with a chemical formula of AB2O4.................33

PAGE 13

xiii 2-13. A calculated diffraction pattern for NiCo2O4 annotated with peak positions, planes, and relative intensities..............................................................................................34 2-14. Spinel unit cell layers stack w ith interlocking tetrahedral sites...............................36 2-15. Tetrahedral site atom (green) bonded to four surrounding oxygen (black).............37 2-16. The octahedral site atom (yello w) bonds with six surrounding oxygen (black) anions to form a six fold orientation........................................................................37 3-1. Combinatorial sputtering uses a dual cathod e setup with targets of different material composition..............................................................................................................43 3-2. Conductivity versus position for combinat orial runs are displayed as a function of oxygen and argon gas composition..........................................................................46 3-3. Position 7 from Figure 3-2 is th e position with the highest conductivity..................47 3-4. Fraction of Ni and Co detected by XPS as a function of position on a combinatorial sputtered NixCo3-xO4 film.........................................................................................48 3-5. Combinatorial sputtered NixCo3-xO4 film combining conductivity and position with composition with position data................................................................................49 3-6. A combinatorial substrate thickness prof ile measured by a stylus profilometer shows that the film thickness is also graded from cobalt rich on the left to nickel rich on the right....................................................................................................................49 3-7. The effect of sputtering gas pressure on NiCo alloy sputtered Ni1.5Co1.5O4 thin film resistivity before and after 10 minutes at 375C heat treatment..............................50 3-8. Target-substrate distance affects the resis tivity of sputtered nickel-cobalt oxide thin films from a NiCo alloy target.................................................................................51 3-9. Conductivity of NixCo3-xO4 films on PET substrate is al ways less than films on other substrates by a factor of 2-4x for as-d eposited samples for all pressures and compositions shown.................................................................................................52 3-10. UPS work function measurement of NiCo2O4 and Ni1.5Co1.5O4 films....................53 3-11. FTIR traces from of reactively sputtered NixCo3-xO4 thin films.............................55 3-12. Optical transmission through a thin film.................................................................56 3-13. Target-substrate distance effects are oppo site for the optical absorption coefficient and resistivity from sputtered nickel-cobalt oxide thin films from a NiCo alloy target......................................................................................................................... 58 3-14. Tauc’s plot of NixCo1-xO4, films show a band gap between 3 and 3.75 eV............59

PAGE 14

xiv 3-15. Increasing target to substrate spu ttering distance decreases film density...............60 3-16. An oxygen plasma shows a confined plas ma region (white plume) near the target and a more dispersed plasma region (y ellow-green) at longer distances.................60 3-17. The three zone model described by Campbell for film growth in a vacuum..........61 3-18. A spinel unit cell at different angles of rotation containing octahedral atoms, minimized tetrahedral atoms, and select oxygen atoms for points of reference such as the enlarged black oxygen atom..........................................................................62 3-19. Grazing incident x-ray di ffraction from sputtered films 50 nm thick appear to be structured as spinel-type, but do not match exactly.................................................64 3-20. Sims profile of nickel-cobalt oxide th in film shows surface composition appears to differ from the bulk composition.............................................................................65 3-21. Cross-sectional TEM image of sputtered nickel cobalt oxide f ilms on (100) silicon. Nickel-cobalt oxide thin film gr ows in multi-grained columns...............................66 3-22. Nickel-cobalt oxide-silicon wafer interface............................................................67 3-23. An Arrhenius plot of conductivity and th e reciprocal of temper ature (K) shows that at high temperatures, the f ilm conductivity is high..................................................69 3-24. TEM image and diffraction patterns at 300 K and 600 K showing no detectable structural changes for the two temperatures.............................................................70 3-25. Effects of heat treatment cooling rate after heat treatment on optical properties of NixCo3-xO4 samples from Figure 3-23......................................................................72 3-26. Resistivity of NixCo3-xO4.........................................................................................73 3-27. Rapidly quenched sample activation energies of NixCo3-xO4 change after heating when they are slowly cooled....................................................................................74 4-1. Thin films with added of lithium deposited from solution precursors or sputtering.82 4-2. Conductivity changes based on the cooling rate follow ing heat treatment for all compositions of lithium in nickel-cobalt oxide films...............................................83 4-3. Conductivity as a function of temperature for 50 nm th ick solution deposited thin films.......................................................................................................................... 84 4-4. Rate of sample cooling after heat tr eatment had a dramatic effect on conductivity.85 4-5. Activation energy dependence for Ni0.75Co2.25-zLizO4 and NiCo2-zLizO4 from solution on lithium content.......................................................................................86

PAGE 15

xv 4-6. FTIR mid IR transmission spectra of alloy-target reactive sputter-deposited Ni0.95Co1.95Li0.15O4...................................................................................................87 4-7. FTIR mid IR transmission spectra of oxide-target sputter-deposited Ni1.2Co1.2Li0.6O4.......................................................................................................88 4-8. Normalized FTIR spectra at 1200 cm-1 show the absorption region near 1400 cm-1 assigned to carbonate on the surface when compared with a carbonate reference spectrum shown in grey............................................................................................90 4-9. Grazing incidence XRD from combinator ial sputter deposited films deposited from oxide targets. Note the weak crysta lline diffraction peaks from the film...............91 4-10. Films sputter deposited from an alloy target have a smaller lattice expansion than films deposited from oxide targets...........................................................................92 4-11. XRD of sputtered films with incremental amounts of lithium (not normalized for intensity)...................................................................................................................93 4-12. XRD spectra show no obvious changes for fast versus slow cooling for Ni0.75Co2.25zLizO4. Fitting the curves reveals changes shown in Table 4-4...............................94 4-13. XPS of the Carbon 1s region from the co mbinatorial sputtered f ilm (a) at positions identified by the color code, and (b) curv e fit to show compos ition of position 9. 95 4-14. XPS of combinatorial deposited nickel -cobalt oxide film with variable lithium concentration............................................................................................................96 4-15. SIMS depth profile of a thin film sputtered from a Ni1.5Co1.5Li0.6O4 target............97 4-16. XPS of (a) Ni 2p, (b) Co 2p, (c) O 1s, and (d) C 1s binding energy regions from solution deposited films of nickel-coba lt oxide containing lithium following slow and fast cooling after heat treatment........................................................................98 5-1. First principles d-band density of st ates calculation of rhodium substituted for cobalt. Energy of 0 eV is the Fermi energy...........................................................108 5-2. Effect of rhodium fraction (h) on conductivity for four-layer films on silicon substrates deposited from solution.........................................................................112 5-3. Sputtered nickel-rhodium oxide and nick el-cobalt oxide films deposited at 7.5 cm in 10 mTorr of 100% oxygen.....................................................................................113 5-4. Resistivity of NiRh4Ox as a function of target-substrat e distance. Increased distance increases resistivity................................................................................................113 5-5. Solution deposited NiCo2-hRhhO4 FTIR transmission spectra were all corrected for the silicon substrate absorption peaks....................................................................114

PAGE 16

xvi 5-6. Resistivity and optical absorption coeffi cient at 3 m as a function of sputtering target-substrate de position distance.......................................................................115 5-7. NiRh2Ox FTIR data referenced to air (lower trace) and corrected for the silicon substrate (upper trace)............................................................................................115 5-8. FTIR spectra of two NiRh4Ox samples of different thicknesses deposited from NiRh4 alloy target by DC sputtering......................................................................116 5-9. FTIR of solution deposited NiCo2-hRhhO4 and sputtered NiRhsO4 thin films corrected for silicon substrate absorption...............................................................117 5-10. Far IR spectra of solution deposited NiCo2-hRhhO4 from FTIR show the structure of the films.............................................................................................................118 5-11. XRD of NiRh4Ox film from DC all oy sputter deposition......................................118 5-12. XRD of amorphous NiRh2Ox thin film from DC alloy sputter deposition............119 5-13. XPS depth profile showing th e surface of sputter deposited NiRh2Ox is different than the bulk concentration....................................................................................120 5-14. XPS depth profile of a film sputtered from a NiRh4 target shows an oxygen depleted surface......................................................................................................121 5-15. Multiplex binding energy da ta from XPS depth profiles......................................122 6-1. Schematic of disorder..............................................................................................130 6-2. The effect of composition on temper ature dependant conduc tivity and activation energy.....................................................................................................................131 6-3. NaPO3 glass phase (amorphous) with a broad band covering the bond vibrational modes.....................................................................................................................132 6-4. A double well potential with the associated energies th at influence carrier motion between the wells...................................................................................................135 6-5. Schematic illustration of one (a) and tw o-dimensional (b) spin models for polarons defined by adjacent site interactions......................................................................138 6-6. Illustration of dipole-dipo le interactions between adjacent sites for polaronic spin model variables in Equation 6.2.............................................................................139 6-7. Energy distribution function for one -dimensional polaron disorder model............140 6-8. Energy distribution function for tw o-dimensional polaron disorder model............140

PAGE 17

xvii A-1. A sputtering target (red) mounted on a cathode (silver) with impinging gas ions (purple) and target partic les being removed (red)..................................................155 A-2. Vacuum chambers used for sputtering deposition...................................................156 A-3. Three-inch sputter cathode used to deposit thin films............................................157 A-4. Combinatorial sputtering se tup inside of chamber with mounted substrates prior to pump-down for sputtering......................................................................................158 A-5. Single rotation setup...............................................................................................159 A-6. Offset rotation setup...............................................................................................159 A-7. Double planetary rotation setup..............................................................................160 A-8. Pressed target produced from ni trate solutions used for sputtering........................161 A-9. The solution deposition method invol ves dropping a precursor solution on a spinning substrate prior to baking to form a film...................................................161 B-1. Hall effect apparatus made by MMR technologies.................................................163 B-2. A van der Pauw resistivity prob e setup and measurement schematic....................164 B-3. Nexus 570 FTIR with 3 detectors was used for high sensitivity mid IR (4000 cm-1 to 650 cm-1, mid IR (4000 cm-1 to 400 cm-1), and far IR (500 cm-1 to 100 cm-1) measurements.........................................................................................................164 B-4. Varian CAREY 5 dual beam ultr a violet-visible spectrophotometer.....................165 B-5. Stylus profilometer used to measur e step height of the deposited film..................166 B-6. Optical profilometer software screen capture. Instrument made by Zygo..............166 C-1. Hexagonally packed cylinders in 3-D on film and top view..................................173 C-2. Cubic packed cylinders top view and in 3-D as would be seen in the film............173 C-3. Degenerate energy level orbitals in the normal spinel state and schematic of orbital energy shift due to the Jahn-Teller effect...............................................................176 C-4. Activation energies are the slope taken from the curve fits to the heating traces...177 C-5. Measured conductivity as a function of temperature for NiCo2-zLizO4 films from solution...................................................................................................................178 C-6. Conductivity is temperature dependent for Ni0.75Co2.25-zLizO4 from solution........179

PAGE 18

xviii Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy POLARONS AND IMPURITIES IN NICKEL COBALT OXIDE By ROBERT REED OWINGS DECEMBER, 2003 Chair: Paul Holloway Major Department: Materials Science and Engineering The optical transparency from 0.3 to >10 m and the electrical conductivity of NixCo3-x-yLiyO4 and NixCo3-x-yRhyO4 films deposited by either planar magnetron sputter deposition or from nitrate solution were investigated. The films had low optical transparencies ( 15%) over the visible region (~ 300-800nm), but the transparency increased ( 50-80%) at infrared wavelengths from 2 to 25 m. The DC electrical conductivity ranged from 10 to 500 Scm-1. These unique properties result in conduction by p -type polarons. Despite low mobility of polarons ( 0.1 cm2V-1s-1), good conductivity results from high concentra tions of small polarons ( 1022 cm-3). For sputtered films, the conductivity was larger with a 50% O2-50% Ar composition, at lower sputter gas pressures (2 versus 10 mTorr), and with smalle r target-substrate distances (best at 5 cm). The effects of composition are due to the re sulting cation oxidati on states, while the effects of pressure and substr ate distance were attributed to more energetic sputtered particles causing increased mobility of su rface adatoms on the depositing films.

PAGE 19

xix Increased adatom mobility led to smooth films which were 20% more dense, and crystalline with a spinel struct ure. Li additions to sputter and solution deposited films of NixCo3-x-yLiyO4 were found to decrease and increas e the electrical c onductivity by less than a factor of two. The decreased el ectrical conductivity fo r sputter deposited NixCo3– x–yLiyO4 films resulted from Li occupying interstitia l sites, rather than substitutional sites in the lattice. Surface segregation and Li2CO3 formation were found. Quenching ( 150 C/min) produced higher conductivities (2x), while slow cooling (10 C/min) resulted in lower conductivities after heat treatment at 375 C for 10 minutes in air. This was attributed to a larger concentration of polarons upon quenching. Solution and sputter deposition of NixCo3-x-yRhyO4 films were crystalline (spinel) and amorphous, respectively, and remained so with heat treatments up to 375 C. Electrical conductivities were 350 Scm-1. Transparency from 2 to 10 m was ~10% higher than for nickel-cobalt oxide films. A model was developed for th e relationship between cation disorder and polaron formation affecting the activation en ergy of electrical c onductivity. Also, the physical limit of polaron concentration wa s used to project a maximum conductivity between 360 and 2400 Scm-1.

PAGE 20

1 CHAPTER 1 INTRODUCTION 1.1 Introduction Materials possessing optical transparency and electrical conductivity make up a small class of specialized materials. Tr ansparent conducting oxide (TCO) materials are basically doped* semiconductors categorized most often by their carrier type ( n -type or p type) or by the mechanism of charge transpor t (free carrier or bound). Each classification of carrier type or conduction mechanism exhi bits complementary optical and electrical properties. Figure 1-1. A basic light emitting thin film stack consists of p-n junction and transparent electrode. Injected electrons and holes combine at the p-n junction interface and relax to emit a photon. Note that th e diagram is merely for illustrative purposes and actual thickness of films may not be proportional in scale. Modern optoelectronics are made possibl e by transparent con ducting oxide (TCO) electrodes. Simple devices that use TCOs ar e made of nanometer-thick material layers Doping is adding trace amount s of impurities to a material.

PAGE 21

2 stacked in various geometries to function as a light emitter or detector (see Figure 1-1 and Figure 1-2). Figure 1-2. A basic light detecting th in film stack device consists of a p-n junction and a transparent electrode. Impinging photons promote electrons in the valence band into the conduction band and out of the device to an ammeter or external circuit device. Note that the diagram is merely for illustrative purposes and actual thickness of films may not be proportional in scale. A very common implementation of both devices is the remote control. The hand held unit contains the infrared emitting diode and the electronic unit to be controlled contains either a silicon diode or a photoconductor to detect the remote control’s optical signal and interpret it to produce the desire d result, such as changing the channel. Novel light emitting diodes (LEDs) su ch as organic light emitting devices (OLEDs), polymer light emitting devices (PLEDs), and longer wavelength infrared emitting diodes (IREDs) are examples of emittin g devices that require an electrode that will transmit light and conduct electricity. This demand will increase for improved solidstate lighting or flexible devices made on plas tic (Bergh et al., 2001; Fo rrest et al., 2000). TCOs are of interest for improving optical de vice performance and efficiency (for items such as a solar cells and LEDs). A thorough understanding of the basic fundamentals of

PAGE 22

3 how an electronic charge carrier moves in this class of materials will provide avenues for future exploitation in device development. Electrical conduction in a solid occurs when a charged particle (referred to as a charge carrier) such as an electron moves under the influence of an applied electric field. Highly conductive materials are most of ten metals with large populations (~1028 cm-3) of unbound or free electrons residing in the conduction band. Electron conductors are n -type materials, while the p -type materials conduct by the moveme nt of electron vacancies or holes Conductivity of both types of TCOs can be controlled by doping levels to allow insulating, semiconducting, or metallic be havior (Kawazoe et al., 2000). Charge transport (electrical conduction) mechanisms in TCOs are similar to metallic free-carrier conduction or bound carrier hopping. Oxides are transparent in the visible region when the band gap exceeds 3.1 eV (Kawazoe et al., 1997; Lewis & Paine, 2000). Most often these oxides are insulators rather than conductors due to relatively few carriers contributi ng to conductivity. Researchers have enhanced transparent oxide s by introducing donor st ates just below the conduction band to produce more carriers in the band and result in condu ctivity increases. One trick to adding donor states requires slightly reducing the film (such as annealing in a hydrogen atmosphere) to create oxygen defect sites that act as donors (Lewis & Paine, 2000). Attempts to make transparent oxide s conducting have yielded higher carrier concentrations but have also affected tran smission properties. One problem with free carriers is a limited transmission region. While transparent in the optical spectrum, the onset of absorption by free carriers at the plas ma frequency abruptly halts transmission. Still, other researchers have attempted to pus h this absorption region further out into the

PAGE 23

4 IR by shifting the region of fr ee carrier absorption to a l onger wavelength. Conducting a hole instead of an electron shif ts the absorption region to a lo nger wavelength because the hole has a larger effective ma ss than the electron. This p -type system is however in the free carrier regime and still subject to free carri er absorption, only at a longer wavelength. Larger masses also move slower than smaller masses, so conductivity of the p -type material is reduced. A material system conducting by a bound carrier would escape this absorption peril caused by free carriers. A polaron is a localized carrier with an accompanying lattice strain. A polaron conducting material inherently has a low mob ility as well as the absence of a plasma frequency cutoff in the infrared region. De spite a low rate of polaron hopping at room temperature, a high polaron concentration will promote a high conductivity. Recent studies have focused on increasi ng the density of polarons in such materials to obtain a net increase in conductivity. Polaron density can be increased by a number of ways. Disordered cation arrangement, increased oxidation state due to selective doping, or structural disorder introduced by cation size mismatch in the lattice are all ways of increasing the number of polarons in a given volume of material. Nickel-cobalt oxide is a recently studied polaron conducting material that shows promise as an infrared-transparent conducti ng oxide (ITCO). NO other ITCO has been studied or reported to date. This work discusses the polaron c onducting nickel-cobalt oxide system, its IR optical and electrical properties and how they are affected by key deposition parameters and the addition of impurities. 1.2 Organization of the Dissertation. Chapter 2 discusses common TCOs and de tails the current kn owledge of their properties. Base properties of the nickel -cobalt oxide system will include the polaron

PAGE 24

5 conduction mechanism and its spinel structure as reported in the l iterature. Chapter 3 describes efforts to enhance the nickel-cob alt oxide properties for use as an ITCO by altering the thin film deposition parameters. Chapters 4 and 5 report the effects of doping the nickel-cobalt system with lithium and rhodium, respectivel y. Chapter 6 discusses the conduction mechanism in the nickel-cobalt oxide system and outlines a model for understanding the effects of diso rder on conductivity to explain the experimental results. A projected maximum range for conductivity is es timated as well. Conclusions from this research and recommendations for future work and applications comprise Chapter 7.

PAGE 25

6 CHAPTER 2 BACKGROUND OF NICKEL COBALT OXIDE This chapter provides the necessary b ackground information to understand the significance of the infrared transparent nickel -cobalt oxide system and how its properties are unique among transparen t conducting oxides. An explanation of two basic p-n junction diodes is given below to illustrate the difference between p -type and n -type materials and shows how a transparent conducting layer functions as part of a light emitting or de tecting device. Limitations of each type of TCO include a particular transmission window and conductivity, both of which depend on the conduction mechanism and carrier type. N-type TCOs are wi dely used and well developed. The p -type TCOs are limited primarily by poor conductivity and are less developed. Research has take n two different approaches to improve the conductivity of p -type TCOs. Explanations of conduction mechanisms such as free carrier movement and polaron hopping provide the background knowle dge for this discussion. Nickelcobalt oxide is the p -type material system of interest to function as an infrared transparent conducting oxide (ITCO) because of its infrared transparency, stability in oxygen, ease of preparation, phase purity, and high conductivit y. Nickel-cobalt ox ide exhibits these characteristics due in part to the carrier type, the conduction mechanism, and the crystal structure. 2.1 Background of Transparent Oxide Conductors Transparent conducting oxides are basica lly doped metal-oxide semiconductors classified as either n -type or p -type. Majority electron conductors are n -type materials,

PAGE 26

7 while the p -type materials conduct by the moveme nt of electron vacancies called holes Conductivity of both types is controlled by dop ing to allow insulative, semiconducting, or metallic behavior (Hummel, 2001; Kawazoe et al., 2000). Currently, n -type TCOs dominate with respect to commercial use and have wide spread application because they exhibit superior conductivity and minimal photon absorption over the visible range as compared with p -type materials. However, p -type oxides exhibit some unique properties such as a majority hole carrier concentration resulting in increased infrared transparency not possessed by n -type oxides. This p -type conductivity and infrared transparency make them of interest for future development (Kawazoe et al., 2000; Kawazoe et al., 1997; Ohya et al., 1998; Windisch et al., 2001a; Wi ndisch et al., 2001b; Windisch et al., 2002b) 2.1.1 The Basic p-n Junction Diode A fundamental optoelectronic device illustrates the differences between n -type and p -type materials that make up a p-n junction. A diode (or rectifie r) is a device that allows current to pass in one direction, but not in the reverse direction. The most basic diode consists of a p -type region next to an n -type region referred to as a p-n junction. Junctions can be homojuctions, made of the sa me material (e.g., s ilicon doped with boron for p -type behavior or with phosphorus for n -type behavior (Hummel, 2001)), or heterojunctions, made from two different materi als. In either case, the behavior of the diode results from the interface between the two different types of materials. A depletion region forms at the interface with an accomp anying electric field (Figure 2-1), which effectively removes generated carriers or allows carrier recombination (also referred to as the space charge region or the depletion layer (Hummel, 2001)).

PAGE 27

8 Figure 2-1. A simple p-n junction diode and accompanying energy band structure illustrates the locations of charge carriers and direction of charge flow. The n type material has electrons in the conduction band while the p -type material has holes in the valence band. When a bias is applied to the junction, the bands shift and electrons (or holes) move from left to right (or opposite) and through the rest of the circuit. Light emission occurs as an electron and hole recombine when injected into the depletion region. The energy of the photon em itted is the difference between the energy of the electron in the conduc tion band and the hole in the valence band (band gap energy). Photo current is generated as photons impinge on the depletion region and excite electrons to the conduction band (over the energy gap) leavi ng behind holes in the valence band. Carriers are c onducted away from the depletio n region similar to the photo diode current diagram in Figure 2-2. Di fferent carrier types are imperative to p-n junction device behavior. Without the two di fferent types, neithe r light emission nor photo-generation of carriers would o ccur. Figure 2-2 shows simple p-n junction diodes, one as a light emitter and the other as a light detector.

PAGE 28

9 Figure 2-2. Basic optoelectronic devices. The light emitting diode (LED, on the left) incorporates n -type and p -type materials to essent ially “convert” electrical current to light by combining electro ns and holes and produce a photon. The photo diode (solar cell -on the right) us es light (photons from the LED) to produce electrical current by generating electron-hole pairs. Materials in these devices conduct a majority of holes or electrons and are classified as p type or n -type respectively. TCOs are used in these devices to allow current and light to pass through to the p-n junction. Key to the operation of a diode is the ability to get current in or out of the device. Transparent conducting oxides serve a unique function in optoelectronic p-n junction stacks because of thei r conductivity and tran sparency. TCOs allow current to pass through to the device depletion region without significantly blocking the light that is being emitted or detected. 2.1.2 N -Type Transparent Conducting Oxides Prevalent TCOs typically possess extrinsic n -type conduction (Chopra et al., 1983) and are more researched due to their superior conductivities. Increased research has sped development and applications of common n -type TCO systems such as (1) the zinc oxide system (e.g., zinc oxide doped with fluorine, boron, gallium, indium, or aluminum), (2)

PAGE 29

10 cadmium stannate (Cd2SnO4), (3) the tin oxide system (e .g., tin oxide and tin oxide doped with antimony [SnO2:Sb referred to as ATO]), and (4) indium oxide doped with tin (In2O3:Sn referred to as indium tin oxide or ITO) (Gordon, 2000). No single TCO is suited for all applications. Table 2-1 shows that if high conductivity is important, then ITO is the best material, however if highest transparency is most important, cadmium stannate or zinc oxide may be more appropriate. Table 2-1. Choice of available n -type transpar ent conductors. Property Material Highest transparency ZnO:F, Cd2SnO4 Highest conductivity In2O3:Sn Lowest plasma frequency SnO2:F, ZnO:F Highest plasma frequency In2O3:Sn Highest work function, best contact to p-Si SnO2:Sb, ZnSnO3 Lowest work function, best contact to n-Si ZnO:F Best thermal stability SnO2:F, Cd2SnO4 Best mechanical durability SnO2:F Best chemical durability SnO2:F Easiest to etch ZnO:F Best resistance to hydrogen Plasmas ZnO:F Lowest deposition temperature SnO2:F, ZnO:B Least toxic ZnO:F SnO2:F Lowest cost SnO2:F Reproduced from Gordon, R.G. (2000). Crit eria for choosing transparent conductors. MRS Bulletin, 25 (8), 52-57, (Table 8, p.55) by permission of MRS Bulletin Currently the most common transparent condu cting material used in electronics is indium tin oxide (ITO). It tran smits up to ninety percent of vi sible light with a resistivity near 1x(10)-4 cm. High transparency, high conductivity, and chemical stability contribute to the wide use of ITO. An oxide conductor like indium tin oxide has a carrier density on the order of 1x1020 cm-3 to 1x1021 cm-3 (Minami, 2000; Minami et al., 1995; Minami et al., 2000) and a mobility typically greater than 10 cm2V-1s-1 (Minami, 2000). Other material systems are better suited fo r specific applications, because each system has specific strengths and weaknesses.

PAGE 30

11 Table 2-2. Figure-of-merit, absorption co efficient and sheet resistance comparison. Material Sheet Resistance Rs ( /square) Visible Absorption Coefficient Figure of Merit a / ZnO:F 5 0.03 7 Cd2SnO4 7.2 0.02 7 ZnO:Al 3.8 0.05 5 In2O3:Sn 6 0.04 4 SnO2:F 8 0.04 3 ZnO:Ga 3 0.12 3 ZnO:B 8 0.06 2 SnO2:Sb 20 0.12 0.4 ZnO:In 20 0.20 0.2 Reproduced from Gordon, R.G. (2000). Crit eria for choosing transparent conductors. MRS Bulletin 25 (8), 52-57, (Table 2, p.53) by permission of MRS Bulletin a Figure-of-merit ( / ), the calculated ratio of a ma terial’s electrical conductivity ( ) to its total optical absorption coefficient ( ) is one method of comparing different TCOs. / = -{RsLn(T+R)}-1. Rs is the sheet resistance, T is the total visible transmittance, R is the total visible reflectance, all of which ar e required to calculate the figure of merit (Chopra et al., 1983). Larger fi gure of merit values indicat e higher performing materials. Rs = (t)/ t = 1/( (t) t). Resistivity, a function of thickness, when divided by film thickness is equal to sheet resistance. Conductiv ity is the reciprocal of resistivity (Chopra et al., 1983). One method of comparing di fferent TCOs on the same criteria is to devise a figureof-merit number as found in Table 2-2. Sheet resistance divided by the total optical absorption coefficient gives a number to comp are both electrical a nd optical properties simultaneously. Zinc oxide doped with fluorin e and cadmium stannate have the highest figure-of-merit at 7, but due to lower conductivit y and higher cost, they are not utilized as much as ITO or tin oxide. It is important to note that the figure of merit is an extremely sensitive to the wavelength of the light being transmitted. ITO is limited in that it is essentially a free electron conductor. Nearly-free electrons absorb photons with energies typi cally less than 1.1 eV corresponding roughly to photon wavelengths longer than 1 m. This onset of absorption by free electrons is called the plasma frequency (Hummel, 2001). Typical values are included in Table 2-3. While this absorption is what makes tin oxi de extremely valuable to the structural

PAGE 31

12 building community when used as an insulati ng thin film on windows, it limits the use of ITO as a transparent electrode at longer wavelengths. Table 2-3. Approximate minimum resistiv ities and plasma wavelengths for some transparent conductors. Material Resistivity (m cm) Plasma ( m) In2O3:Sn 0.100 >1.0 Cd2SnO4 0.130 >1.3 ZnO:Al 0.150 >1.3 SnO2:F 0.200 >1.6 ZnO:F 0.400 >2.0 Reproduced from, Gordon, R.G. (2000). Crit eria for choosing transparent conductors. MRS Bulletin 25 (8), (Table 3, p.54) by permission of MRS Bulletin Many studies have been conducted on the n -type oxides to develop applications, however transparent p -type materials are not well deve loped despite the need for them (Giesbers et al., 1997; He et al., 1999; Park et al., 2002 ). 2.1.3 P -Type Transparent Conducting Oxides Investigations into p -type TCOs include two major systems (1) the nickel oxide systems, nickel oxide (Giesbers et al., 1997; Sato et al., 1993) with additions of cobalt oxide (He et al., 1999; Windisch et al., 2001a ; Windisch et al., 2001b), and (2) the copper oxide systems, strontium copper oxide (Kaw azoe et al., 2000), copper scandium oxide (Duan et al., 2000), copper gallium oxide (Ueda et al., 2001), lanthanum copper oxide (Ueda et al., 2000), and copper aluminum oxide (Kawazoe et al., 1997). Research conducted on p -type TCO material syst ems has yielded a few p-n devices, but beyond prototypes, they have not been further de veloped (Giesbers et al., 1997; Ohya et al., 1998). Infrared spectra of all the listed p -type materials have a plas ma cutoff deeper in the IR than the n -type TCOs except the nickel systems that do not have a published plasma

PAGE 32

13 cutoff. Conducting via holes they may be used as a p -type electrode next to the p -type layer of a p-n junction diode or as the p -type layer in the junction. Table 2-4. Electrical properties of CuAlO2, CuGaO2, SrCu2O2, and NiCo2O4 thin films. CuAlO2 a CuGaO2 a SrCu2O2 a NiCo2O4 b Electrical conductivity (S cm-1) 9.5x10-1 6.3x10-2 4.83x10-2 333 Carrier density (cm-3) 1.3x1017 1.7x1018 6.1x1017 ~1x1021 Hall mobility (cm2/V s) 10.4 0.23 0.46 < 0.1 Seebeck coefficients ( V K-1) +183 +560 +260 ~+20 a(Kawazoe et al., 2000) b(Windisch et al., 2002b) Limitations of p -type TCOs include low conductivity (near 0.10.01 Scm-1) and 70%-90% transparency in the visible regi on. With conductivity about two orders of magnitude lower (at best) than a good n -type TCO such as ITO at 10000 Scm-1, p -type TCO electrical properties shown in Table 2-4 do not yet compare with n -type TCOs. Plasma absorption of free or quasi-free carrier s limits transparency at longer wavelengths regardless of the conducting ca rrier type. Interest in p -type TCOs has recently renewed with new ideas on how to overcome thes e limitations (Kawazoe et al., 2000). Conductivity, the major limitation for p -type TCOs, is a function of the number of carriers and the speed at which those carri ers move through the material. Shown in equation 2.1, conductivity is a f unction of carrier density ( n ) multiplied by mobility () and the charge of each carrier ( q ). 1 nq (2.1) Assuming this equation is suitable for al l types of materials, two choices are available to increase conductivity because th e charge of each carrier is constant. One choice is to increase mobility by controlling the valence band; the other option is to boost the carrier density.

PAGE 33

14 Chemical modulation of the valence band. Kawazoe et al. suggest the limited conductivity of the p -type TCOs result from valence cations bonding with oxygen forming localized carriers. In contrast, high conductivity observed in n -type TCOs is often due to high free carrier densities ev en with carrier mobilities at a low 10 cm2V-1s-1. Locally bound p -type carriers present a problem becau se they are less mobile than free carriers. The solution is referred to as ch emical modulation of the valence band (CMVB) where cation selection in the lattice controls the valence band while maintaining quasifree carrier movement. CMVB rids the valen ce band of positive holes that become deep acceptor levels by introducing covalency in the metal-oxygen bonding to form an extended valance-band structure. Transition metal oxides with unfilled d10 shells are not recommended for TCO applications because d-d transitions exhibit strong coloration in the visible. Copper oxide with select cations added was the example system (Kawazoe et al., 2000). The best conductivity repo rted in Table 2-4 is 0.95 Scm-1 for CuAlO2 indicating that this regime worked to some degree, but conductivity remains low. If mobility is at the maximum for the system, and e is constant at 1.606x10-19 C (per electron), then the only othe r parameter to influence in the conductivity equation (Equation 2.1) is the number of carriers. Increase carrier density. Another approach to impr oving conductivity is to leave the carriers localized and attempt to add more of them (Windisch et al., 2002b). Localized carriers do not exhibi t the infrared absorption edge allowing transparency at longer wavelengths and are called polarons when accompanied by lattice strain. They have a very low mobility due to a hopping tr ansport mechanism unlike free carriers. In addition, cations that have open d shell orbitals such as nickel and cobalt in nickel-cobalt

PAGE 34

15 oxide reported by Windisch et al. ha ve a conductivity of up to 333.3 Scm-1 and have exhibited transparency up to 60% at 600 nm and higher at higher wavelengths (Windisch et al., 2001a). The nickel-cobalt system is not as transparent in the visible region as the copper system, though very transparent in the infrared. 2.2 Nickel-Cobalt Oxide Interest in nickel-cobalt oxi de stemmed from early work on either nickel oxide or cobalt oxide. Nickel oxide studies explored electrical properties with interest in its superior transmission (near 80%) in the visibl e region. Attempts to exploit this in electrochromic window applica tions (Kitao et al., 1994), as p -type transparent conducting films (Sato et al., 1993), an antiferromagnetic material (Fuj ii et al., 1996), a functional sensor layer for chemical sensors (Kumagai et al., 1996), or as a photocathode for a solar cell (He et al., 1999), fueled interest in this system. Cubic nickel oxide has also been investigated with respect to electrical properties (Morin, 1953 ) and its conduction mechanism (Biju & Khadar, 2001; Parravano, 1954; Sanz et al., 1998; Snowden, 1965). Studies of nickel oxide showed experimental results, but as alluded to previously, poor conductivity of the p -type oxides has not encouraged interest in developing p -type TCOs in general for applications beyond lab experimentation. Spinel cobalt oxide has also been studied previously as a result of its magnetic spin states (Belova et al., 1983), optical nonlinea rity (Yamamoto et al ., 2003), gas sensing capabilities, and solar ener gy reflecting properties (Cheng et al., 1998). Doping single crystals of cobalt ox ide with nickel produced a signi ficant increase in electrical conductivity (up to 105x) while maintaining the spinel structure. Nickel cations were found to reside in octahedral sites with a va lence of 2+ and 3+ (R oginskaya et al., 1997; Tareen et al., 1984) repres ented by the equation below.

PAGE 35

16 Co1-y 2+Coy 3+Co2-x 3+Niy 2+Nix-y 3+O4 (2.2) Nickel doped cobalt oxide shows p -type semiconducting behavior similar to intrinsic spinel cobalt oxide (Tareen et al.). Investigators have work ed to develop nickelcobalt oxide for applications intended origin ally for nickel oxide or cobalt oxide (Monk & Ayub, 1997). Other interests in nickel-coba lt oxide include uses as electrodes in batteries (Liu et al., 1999; Polo Da F onseca et al., 1999; Yoshimura et al., 1998), electrodes in solar cells (Park et al., 2002 ), electrodes in molten carbonate fuel cells (Kuk et al., 2001), or as a heterogeneous optic al recording media, (Iida & Nishikawa, 1994). These intended uses of nickel-cobalt oxide are not paramount to this study but show other research interests in the material have been pursued in the past to develop specific applications such as using NiCo2O4 as an electrocatalyst for anodic oxygen evolution (Carey et al., 1991; Haenen et al., 1986), in or ganic or inorganic electrosynthesis (Roginskaya et al., 1997), as a supercapacitor (Hu & Cheng, 2002), or as an infrared-transparent conducting electrode fo r flat panel displays, sensors, or optical limiters and switches (Goodwin-J ohansson et al., 2000; Windisch et al., 2001a; Windisch et al., 2001b; Windisch et al., 2002b). Studies of nickel-cobalt spinel have in cluded bulk crystals (Haenen et al., 1986; Roginskaya et al., 1997; Tareen et al., 1984), powders, (Windisch, 2003) and thin films (Carey et al., 1991; Galtayri es & Grimblot, 1999; Hu & Ch eng, 2002; Kim et al., 2000; Marco et al., 2000; Monk & Ayub, 1997). Ni ckel-cobalt oxide studies have correlated the crystal structure and activity of NiCo2O4 and related oxides (King & Tseung, 1974), explained the NiCo2O4 spinel surface by Auger and X PS (Kim et al., 2000), and have expounded on the deposition parameters and substo ichiometric structures (Carey et al., 1991). Work on the nickel-cobalt oxide syst em and stoichiometric deviations has

PAGE 36

17 included infrared spectroscopy (Windisch et al., 2001b), laser Raman spectroscopy (Windisch et al., 2002a), Seebeck and conduc tivity measurements (Windisch et al., 2002b), and conductivity measurement variat ions correlated to film composition (Windisch et al., 2001b). Post deposition he at treatment and resulting increases in conductivity were correlated with oxygen binding energy changes (Windisch et al., 2001b; Windisch et al., 2002b; Windisch, 2003b). Table 2-5. Comparison of nickel-cobalt oxide to nickel oxid e and cobalt oxide. NiO NiCo2O4 Co3O4 Breakdown temp. ( C) 1100+ 400 a 895 a Conductivity (S/cm) 2x10-2,b 333 c 10-6 d, 0.05 e, 0.07 f Carrier type p –type small polaron, p g p -type small polaron h p hopping e,f,i Carrier density (cm-3) at room temp. 9.8x1020 j, 1020,k, 1019 k 1.3x1019 l b 5x1021 m 2.4x1019 e Lattice constant () 4.195 n, 4.176 o 8.114 p 8.084 q, 8.11 o Structure Cubic n Spinel r Spinel d,r Seebeck coef.(V/ C) 350-500 j 20 m 0 d Band gap (eV) 4.0 s, 3.8 l Not reported 1.5 t, 1.65 u, 2.2 v, 2.02 w a (Haenen et al., 1986) b (Morin, 1953) c (Windisch et al., 2001a) d (Tareen et al., 1984) e (Patil et al., 1996) f (Cheng et al., 1998) g (Sato et al., 1993) h (Windisch et al., 2002b) i (Ohya et al., 1998) j (Parravano, 1954) k (Snowden, 1965) l (Sato et al., 1993) m (Windisch et al., 2002a) n (Hotovy et al., 1998b) o (Kennedy, 1996) p Powder Diffraction File #73-1702 (see Appendix B for specifics) q (Taylor & Kagle, 1963) r (King & Tseung, 1974) s (Lunkenheimer et al., 15) t (Schumacher et al., 1990) u (Varkey & Fort, 1993) v (Pejova et al., 2001) w (Yamamoto et al., 2003) In Table 2-5 a comparison of nickel-cobalt oxide to nickel oxide and cobalt oxide outlines key differences in the base oxides and the mixed material. When mixed together, the breakdown temperature is 400 C (far below either bi nary oxide), while at the same time the carrier con centration increases by up to 102x and the conductivity

PAGE 37

18 increases by 105x. The lattice parameter is largest for the mixture of nickel-cobalt oxide and the material exhibits the same p -type nature as both nickel oxide and cobalt oxide. Work function measurements were not avai lable for these materials including nickelcobalt oxide. Band gap information was al so not available nor was the index of refraction Figure 2-3. Visible transmi ssion spectra of nickel-cobalt ox ide from solution deposited and sputtered samples. Reprinted from Windisch, C.F., Exarhos, G.J., Ferris, K.F., Engelhard, M.H., & Stewart, D.C. (2001). Infrared transparent spinel films with p-t ype conductivity. Thin Solid Films 398, 45-52 (Figure 3, p.47) with permission from Elsevier, color added. Using the available data one might reasona bly interpolate the existing reports and surmise a calculated bandgap in the neighbor hood of 2.5-3.0 eV, if Vegard's law were valid for this system. However, the band ga p for NiO may not be a valid number to use in this interpolation method because nickel, in NiCo2O4, is in the same chemical state as it would be found in Ni3O4. Since Ni3O4 is not a stable phase existing at standard temperature and pressure conditi ons, no electrical information is available for it. Nickel

PAGE 38

19 oxide and cobalt oxide individua lly differ in structure and el ectrical properties, but they both exhibit electrical pr operties far inferior to mixed nickel-cobalt oxide. According to Figure 2-3, optical abso rption begins near 600 nm where the transmission is nominally less than 30%. As stated previously, this is most likely due to d-d transitions absorbing the visible light. By TCO standards, this transparency is unacceptable for use in any practical device Nickel-cobalt oxide transparency is acceptable at longer wavelengths as seen in Figure 2-4 below. Figure 2-4. FTIR transmission spectrum of solution deposited nickel-cobalt oxide thin film. Reprinted from Windisch, C.F., Exarhos, G.J., Ferris, K.F., Engelhard, M.H., & Stewart, D.C. (2001). Infrared transparent spinel films with p-type conductivity. Thin Solid Films 398, 45-52, (Figure 4(b), p.48) with permission from Elsevier, color added. Separately neither nickel oxide nor cobalt oxide has the electrical properties to be of industrial interest though they are both p -type with high carri er counts and low mobility. Windisch et al. determined the carrier to be p -type using the Seebeck coefficient (Windisch et al., 2002a; Windisc h et al., 2002b). Figure 2-5 shows Seebeck data taken from a nickel-cobalt oxide thin film. They report that Ni3+ also has a role in conductivity such that the mechanism of conducti on is a charge transf er between resident divalent and trivalent cations suggesting it is possibly assisted by the magnetic nature of

PAGE 39

20 the oxide film (Windisch et al., 2002b). The ma gnetic nature of the film has been blamed for difficulty in obtaining reli able results from the Hall eff ect measurement. Anomalous Hall effect measurements made an absolute determination of carri er concentration and carrier type difficult using the Hall effect measurement. A primary assumption of the Hall effect is that the measurement occurs in a material with free carriers. Figure 2-5. Seebeck data from nickel-coba lt oxide. Reprinted from Windisch, C.F., Ferris, K.F., Exarhos, G.J., & Sharma S.K. (2002). Conducting spinel oxide films with infrared transparency. Thin Solid Films 420, 89-99, (Figure 6(a) p.93) with permission from Elsevier, color added. Emin addressed an anomalous Hall sign in oxide semiconductors. The sign of the Hall effect measurement depends on the nature and relative orientations of the local orbitals between which the carrier moves and on the local geometry. The sign of the Hall angle is not unambiguously dete rmined by the sign of the ch arge carrier in the case of small-polaron hopping motion. The Hall effect depends on the local geometry and on the nature of the local electronic states. The observed anomalies of the sign of the Hall angle may be explained as simply being a manifest ation of the hopping natu re of the transport in phonon-assisted hopping motion (Emin, 1977). Polaron conductors and the Hall effect

PAGE 40

21 have been discussed in the literature for th e mentioned reasons. The Hall effect does not produce reliable reproducible values of carrier type, density, or mobility. This Hall effect anomaly is likely due to the localized natu re of the carriers an d the hopping conduction mechanism of nickel-cobalt oxide. Figure 2-6. The oxygen 1s binding energy regi on from XPS shows a peak at 531.2 eV that scales with conductivity. Reprin ted from Windisch, C.F., Exarhos, G.J., Ferris, K.F., Engelhard, M.H., & Stewar t, D.C. (2001a). Infrared transparent spinel films with p -type conductivity. Thin Solid Films 398, 45-52, (Figure 6. p.49), with permission from Elsevier, color added. Nickel-cobalt oxide is probably a defect conductor due to adsorption of oxygen similar to nickel oxide where high oxygen partial pressure increases conductivity (Hotovy et al., 1998a). Cation vacancies produ ced from oxygen adsorption create holes in the valence band making the material p -type and it is therefore classified as an electron-defect semiconductor. (Azaroff, 1960) Additionally, the defect states of lattice oxygen monitored by XPS in the oxygen 1s region (see Figure 2-6) at a binding energy of 531.2 eV are believed to scale with conductiv ity and may be an indicator of polarons (Windisch et al., 2001a).

PAGE 41

22 Figure 2-7. Comparison of TC O transparency regions with conductivity displayed as a function of transmission wavelength. Th e long line in center is nickel-cobalt oxide transparency range. It extends off the chart into the far IR with much better conductivity than the other p -type TCOs.* Windisch et al. have demonstrated the nick el-cobalt oxide system as a prospect for development with respect to conductivity a nd infrared transparency (shown in Figure 24) (Windisch et al., 2001a; Windisch et al ., 2002a; Windisch et al., 2001b; Windisch et al., 2002b; Windisch, 2003b). Their work has suggested that the conductivity of the nickel-cobalt oxide system could be improved by the addition or subs titution of selected cations, such as lithium, rhodium, or palla dium. (Windisch et al., 2001a; Windisch et al., 2001b; Windisch et al., 2002b) Figure 2-7 shows that nickel cobalt oxide has a transparency window that extends more into the IR than typical n -type or other recently studied p -type TCOs. No other reported transp arent conductor has a similar transparent region. UV cutoff values of the TCOs are approximate for a qualitative comparison of the transparent regions in the IR. Plasma cutoff values for each are taken from the lite rature (Gordon, 2000; Wi ndisch et al., 2001b; Windisch et al., 2002b). Plasma values for p -type TCOs estimated.

PAGE 42

23 The nickel cobalt oxide system is unique in that it exhibits p -type conductivity with highly localized carriers bound to the lattice with an accompanying lattice strain, i.e. with polaron conduction. This bound car rier and lattice strain toge ther are known as a small polaron. Carrier mobility is on the order of 0.1 cm2V-1s-1 due to the lattice bound localized carriers (Windisch et al., 2002b). Conduction occurs in these materials in spite of the low mobility because of a high polarons density, ~1021 cm-3 (Windisch et al., 2002b). Conductivity up to 333 Scm-1 has been measured (Windisch et al., 2001a). Figure 2-8. Effects of de position techniques on the resist ivity of nickel-cobalt oxide show that sputtered films of similar compositions to solution deposited films have lower resistivities. Reprinted fro m Windisch, C.F., Exarhos, G.J., Ferris, K.F., Engelhard, M.H., & Stewart, D.C. (2001a). Infrared transparent spinel films with p-t ype conductivity. Thin Solid Films 398, 45-52. (Figure 2, p.47) with permission of El sevier, color added. Windisch et al. report that the best co nductivity of sputtered films was found at nickel to cobalt ratio of 1:1 for sputtered films and 1:2 for solution films (Figure 2-8). Both the 1:1 sputtered and 1:2 solution deposit ed nickel-cobalt oxide s were cubic spinel structured. Also included in Figure 2-8 was a compar ison of solution and sputter deposited films showing the difference at th e same composition for films deposited by

PAGE 43

24 two different methods. It is believed that the difference in the film conductivity comes from lattice cation disorder accommodating po laron charge carriers (Emin & Bussac, 1994). Sputtering allowed a hi gher conductivity while at the same time preventing less transmission in the visible re gion due to strong absorption by d-d transitions from either the nickel or the cobalt ions. The spin state of Co3+ in the octahedral site may act as an acceptor and Ni2+ may exist in the octahedral site. Both of these cation states may enhance the conductivity (W indisch et al., 2002b). 2.3 Nickel-Cobalt Oxide Conduction Mechanism Nickel-cobalt oxide conducts electrically by a mechanism of small polaron hopping (Windisch et al., 2002b). An understandi ng of why polarons and free carriers are different sheds light on why the material properties are also different. Polaron conductivity gives nickel-cobalt oxide its unique properties of infrared transparency and high conductivity despite low values of mobility Large numbers of polarons form with increased structural disorder that can be introduced by processing, selective cation addition, or cation substitution. Understanding polarons may give insight to discovering polaron-hopping conduction in ot her materials as well. 2.3.1 Fundamentals of Polaron Conduction When a charge carrier such as an electr on or hole distorts its neighboring lattice structure and traps itself, it is called a “pol aron.” A polaron is described as an entity including both the displaced neighboring atoms (localized st rain) and the trapped carrier (Cox, 1987; Emin, 1982; Austin, 1969). Mistakenly called a polaron due to early investigations of self-trapping in polar and ionic solids, self-trapping is not restricted just to polar a nd ionic solids with long-range dipolar electron-lattice interactions. Polarons consist of two main types: large and small.

PAGE 44

25 The two are distinguished by the se verity of the localized strain area. “A small polaron is an extra electron or hole severely localized within a potential well that it creates by displacing the atoms that surround it.” (Emin, 1982, p.34). When the electronic carrier and the lattice distortion together have a linear dimension less than the lattice parameter, it is referred to as a small polaron (Kingery et al., 1975, p.870). Short-range electronlattice interaction plays the major role allowing small polarons to occur in polar, covalent, and ionic materials and generally is comprised of interactions of the carrier with acoustic and optic vibrational modes of the lat tice. Polaron motion can be described as a succession of phonon-assisted hopping steps (Emin, 1982). Figure 2-9. A bound carrier in a two-dimensional lattice is th e enlarged blue atom. The polaron is the carrier and the accompanying strain. Schematic (a) is before the hop, (b) is the move, and (c) is th e strain movement after the hop. A hop consists of the following three steps shown in Figure 2-9 above: 1. Atoms arrange to allow multiple positions for a charge carrier. 2. The charge moves between degene rate electronic energy levels 3. Local deformation follows. Polarons typically have a low mobility b ecause the process of hopping as described previously takes more time than does movi ng a free carrier. Fr ee carriers travel unbound with a given drift velocity ( vf) under the influence-of-an applie d electric field (E) as seen

PAGE 45

26 in Equation 2.3. In a periodi c lattice, nearly-fr ee electrons travel with a random drift velocity. With no external electric field applied, the drift velocity nets out to zero. Drift velocity is a function of the averag e time between scattering events ( ) and the mass of the carrier (m) shown in Equation 2.3. vf Ee m (2.3) Free carriers have small masses lending to highe r velocities and therefore higher mobility given the same applied field assuming no increase in collision scattering. v E (2.4) When combined with a high carrier density, th e nearly free carrier provides for a high conductivity. Nfe2m (2.5) Assuming that a periodic lattice moves in ti me with some given displacement at a frequency determined by the temperature, at an y give time there is a probability that the lattice will arrange itself to allow a charge carrier to move between sites. Lattice distortion will follow. This occurrence is statis tical in nature and has a finite probability at a given temperature. A bound carrier must be activated by a discrete amount of energy to hop from one site to another followed by the strain. Low mobility is a result of the probability of hopping combined with the en ergy required to hop. So, for a polaron conductor, the time between scattering events ( ) may well be converted to a frequency within a given time for a hop to occur. This jump rate relationship ties the motion of the carrier to the phonon motion or natu ral frequency of the lattice. 2.3.2 Fundamentals of F ree Carrier Conduction Nearly the opposite of polarons, free carrier s are not locked into a specific position within the lattice. Free ca rriers typically have a high mobility until the carrier concentration increases to the point that they begin to interact by colliding and scattering.

PAGE 46

27 Figure 2-10. The particle in a box plot of a particle in a one dimensional lattice bound by an infinite energy barrier. The particle is confined to the region inside the 1-D well. The particle resides on a line and moves in a linea r position (left and right) direction bounded by the energy barriers on either end. Modeling free carriers mathematically is si mple and most often done with a simple particle in a box model (the box term may be a misnom er when the system is one or two dimensional). The assumption for the particle in a box is a single molecule or particle in a one-dimensional position that can move linearly in two directions bounded by a potential well on each side. The graph of this energy versus pos ition looks like a box seen in Figure 2-10, hence the name particle in a box The energy for a particle in the box is described by Equation 2.3, where En is the allowed energy, n is the principle quantum number, h is Planck’s constant, m is the mass of the particle and L is the length of the one-dimensional box. En n2h28 mL2 (2.6) The associated normalized wave function for the charge carrier in the well is given by Equation 2.7, where x is the position of the charge carrier in the box nx 2 L sin nx L (2.7) The solution to the wave function yields discrete nodes of allowed frequencies for the particle (charge carrier). These nodes correspond to discre te energies that are allowed

PAGE 47

28 within the confines of the box The Pauli exclusion principle* dictates the energy levels of the quantum states of the confined electr ons (or charge carriers). Each will have a unique set of quantum numbers though the energy value may be the same (degenerate). When the density of electrons is large, ener gy levels corresponding to quantum states are so close that the distribution is nearly continuous. This e ssentially continuous band of energies is referred to as a conduction or valence band for electronic conduction and is used to describe p -type and n -type materials. With free carriers, the box size is much larger than the crystal unit cell dimensions. This particular model matches classical physics based on billiard ball collisions when th e dimensions of the box approach the centimeter scale. The picture fails when the deBroglie wavelength of a particle is reduced to atomic dimensions. (Emin, 1980). Modeling free carriers using the particle in a box is well accepted. A model of similar simplicity for polarons does not exist. 2.3.3 Observed Properties of Free Carriers and Polarons While high conductivity is a ch aracteristic of free elec tron carriers, for instance degenerately doped indium tin oxide (ITO) c onducts electricity by free electrons. Free electrons absorb photons at wavelengths grea ter than ~1-2 m, but do not prevent high transparency in the visible spectrum from 400-80 0 nm. It is because of this free carrier absorption that n -type free carrier materials do not function as IR transmitting electrodes but serve as low emission “low e” glass co atings for insulatin g windows (Svensson & Granqvist, 1986). Light of energy less than ~1 eV is easily absorbed and/or reflected as the free carriers are promoted by the incoming photons to higher ener gy states that relax and give off phonons or photons. The energy ons et of this absorp tion activity due to The Pauli exclusion principle requires that each quant um state can be filled with at most two electrons each with opposite spin s (Hummel, 2001).

PAGE 48

29 carrier excitation is referred to as the plasma frequency*. Polarons do not exhibit a plasmon resonance near the same region becau se they are locally bound to the lattice. (a) (b) Figure 2-11. A polaron consists of the molecu les involved and the local charge region: (a) Liquid ammonia (color less) forms polarons (dark blue) when sodium metal is added (b) Molecular orientation of polaron in liquid ammonia. Shaded areas are regions of high electr on density (donated by sodium). ** As a dramatic demonstration of the effect s of polarons on color in white light, the property changes upon formation of polarons in liquid ammoni a from electrons injected by sodium metal pellets is illustrated in Figur e 2-11. Polarons form as regions of high electron density are created around the sodi um ions. The polar ammonia molecules preferentially arrange around the charged regions This localized charge distorts bonding and coupled with vibrational modes shifts th e characteristic optic al spectrum of the molecule. Temperature is a factor to consider when dealing with polarons and free carriers. Free carriers and polarons in semiconductors beha ve differently as the lattice is heated. Free electrons in metals will decrease in m obility with increasing temperature due to Plasma frequency is a characteristic frequency that separates the optically reflective region from the transparent region. The dielectric constant goes to zero and conditions are right for plasma (fluid-like) oscillation for the entire electron gas. ** Video contained in a separate file ( PolaronClip1.avi ).

PAGE 49

30 carrier collisions and sca ttering with a net result of decreased conductivity. Semiconductors behave in two different wa ys depending on their doping regime. The intrinsic semiconductor will decrease in mobility and consequently conductivity until at a high enough temperature thermal energy excite s electrons from the valence band to the conduction band. Carrier generation will a llow conductivity to increase up to a certain threshold value where carriers begin to collid e and scatter causing an overall decrease in conductivity due to the reduction in mobility even as the carrier density increases. Thermal generation of carriers from extrinsi c dopants will also affect the conductivity due to dopant ionization and increased carrier density until the point of saturation. The increased temperature will no longer increa se carrier density beyond the saturation concentration, but enhance carrier collision and scattering will reduce mobility and decrease conductivity with increased temper ature until the material begins intrinsic carrier generation. Materials in whic h conduction is limited by hopping experience the opposite effect due to the hi gh population of bound carriers ( polarons). As temperature increases, the lattice vibrati ons increase and the number c onfigurations per second that allow hopping. The time for a carrier to hop is believed to remain constant with increased temperature, but the opportunities for hopping increase. Polarons ideally experience a mobility increase with temper ature due to increased hopping. Carrier density remains constant so the net effect is an increased conductivity with increased temperature. When a polaron-conducting ma terial is compared to a free carrier conducting material at higher temperature, the two are distinguished on the basis of mobility and carrier density.

PAGE 50

31 Table 2-6. Observed properties of polarons compared with free carriers. Observed property Polaron Observed in transition-metal oxides with ions in multiple oxidation states Free carrier Metals; n -type or p -type semiconductors Seebeck coefficient Value Small < 20VK-1.a Large 0.1 – 1 mVK-1 Positive for p -type and negative for n -type. a Seebeck coefficient temperature dependence Nearly independent a Dependent, decreases with increased temperature. Conductivity value Good ~ 10-1000 Scm-1.a Better 1000+ Scm-1.e Conductivity temp. dependence Related to exp (-Ea )/(kT) where Ea is about 0.2 eV. b Degrades; Plateaus with temperature then decreases. c Carrier density value High ~1021 cm-3.b Medium to high 1018-21+ cm-3 Increases with temperature Hall mobility Small << 1 cm2V-1s-1.a > 10 cm2V-1s-1 Hall mobility and temperature Increases with temperature. c Falls as temperature increases. c Carrier density Promoted by structural disorder. b Increases with temperature. Hall sign Anomalous. a, c p -type positive, n -type negative. a(Windisch et al., 2002b) b(Windisch et al., 2002a) c(Emin, 1982) The Seebeck coefficient* changes significantly with temperature for free-carrier conductors but only a negligible amount or not at all for a pola ron conductor. This nearly temperature-independent behavi or is one of the key indica tors of polaron hopping. Table 2-6 summarizes some properties of pol aron hopping versus free carrier conducting materials. Hall measurements show an increa se in carrier density with temperature and a decrease in mobility. Seebeck measurements give a higher value for carrier density that is not influenced with temperature. In the Hall measurement, carrier count is measured and mobility is calculated. A measure of the voltage change with temperature in a material from heat driven diffusion gradient of carriers in the material. If the change is positive it indi cates holes are the majority carrier, if negative, then electrons are dominant.

PAGE 51

32 Nickel-cobalt oxide is reported to conduc t via a small polaron hopping mechanism (Windisch et al., 2002b). The key consider ations for small polaron hopping, shown in Table 2-7, verify that nickel-c obalt oxide is a polaron conductor. Table 2-7. Key characteristics of sm all polaron hopping for nickel-cobalt oxide Characteristic Observed? Observed in oxides with ions in more than one oxidation state Yes (Ni(+2,+3), Co(+2,+3)) a Small Seebeck coefficient 20 VK-1 b Small charge carrier mobility < (0.1 cm2V-1s-1)b Conductivity related to temperature S=exp (-Ea )/(kT) where Ea is about 0.2 eV Yes b High carrier density Ye s, approaching 1x(10)22 b Seebeck coefficient is nearly temperature independent Yes b Correlates with phonon behavior Yes: XRD & Raman c Promoted by structural disorder Yes c a(Roginskaya et al., 1997; Tareen et al., 1984) b (Windisch et al., 2002b) c (Windisch et al., 2002a) Temperature dependent measurements of the Hall effect show mobility increasing with temperature while the carrier con centration remains high for polaron conducting films. Conductivity is due to the high density of carriers in spite of low mobility making conduction possible because each carrier hops to contribute to the net current. Activation energy of electrical conductivity is influenced by temperature. Mobility increases as activation energy decreases w ith increased temperature. Conductivity increases with temperature because mobility increases with temperature. Based on observed properties, a polaron model will be introduced that will relate charge disorder to activation energy for electrical conduction in the nickel-cobalt oxide system. 2.4 Nickel-Cobalt Oxide Spinel Structure Transition metal oxides take on a variety of structures including the delafossite, perovskite, pyrochlore, smectite, sodalite, stib iconite, and spinel structures (Fleischer,

PAGE 52

33 1995). Cations stack in a solid based on a preferred nearest neighbor arrangement. These arrangements created from the stacki ng orientation determine the nature of the structure and may influence the conducti on mechanism. Ternary compounds often arrange in two common struct ures: perovskite and spinel. (Smyth, 2000) A third more recently studied ternary arrangement is the dela fossite structure. (Kawazoe et al., 2000) This section provides a brief background of th e spinel structure and its characteristic arrangements in the nickel-cobalt oxide system. Figure 2-12. Green tetrahedral cations (A), yellow octahedral cations (B), and black oxygen anions (O) form the spinel unit cell with a chemical formula of AB2O4 (based on atomic layout (K ingery et al., 1975, p.65). Spinel, a gem found in nature often mi staken for Ruby, contains magnesium, aluminum and oxygen ( Gem by gem, 2003; Hughes, 1999; Sickafus, 1999)). The arrangement of cations and oxygen in the ge m spinel, shown in Figure 2-12, is the common crystallographic arrangement for a number of other minerals and oxides. Crystallographers refer to other similarly structured systems as having the spinel

PAGE 53

34 structure. The spinel group of natural cr ystalline materials includes mineral such as spinel (MgAl2O4), chromite (FeCr2O4), coulsonite (FeV2O4), gahnite (ZnAl2O4), hercynite (FeAl2O4), magnesiochromite (MgCr2O4), magnesioferrite (MgFe2O4), magnetite (Fe3O4), trevorite (NiFe2O4), zincochromite (ZnCr2O4). Another isostructural mineral is the linnaeite group with sulfur instead of oxyge n occupying the anion sites. (Fleischer, 1995) This sulfide containing lin naeite structure however, is often also referred to as spinel (Balla l & Mande, 1977; Benco et al., 1999; Kishimoto et al., 2000; Ohmuro et al., 1995; Schellenschlager & Lutz, 2000). Figure 2-13. A calculated diffraction pattern for NiCo2O4 annotated with peak positions, planes, and relative intensities from the powder diffraction file database number 73-1702. As nickel is added to cobalt oxide to make the NixCo3-xO4 mixture, the spinel structure from the cobalt oxide persists up unt il cubic nickel oxide precipitates out. The most conductive mixture of nickel-cobalt oxid e is spinel structured (Windisch et al., 2002b). Figure 2-13 shows a calculated XRD pattern of the spinel structure.

PAGE 54

35 The atomistic arrangement in the spinel structure accommodates multiple cations in multiple oxidation states and allows hopping conduction (Windisch et al., 2002b). While the literature agrees that the bonding nature between atom s in the spinel structure is mostly ionic, a complete understanding of the bonding is subject to controversy. (Azaroff, 1960). Grazing incident x-ray diffraction (GIX RD) from both sputtered and solution deposited films exhibit a spinel pattern, similar to the powder diffraction file (PDF) number 73-1702 for nickel-cobalt spinel show n in Figure 2-13. This structure is characterized by a primary peak from the (100) plane at 36.70 (2-theta) and secondary peaks from (220), (400), (511), and (440) at 31.15 44.63 59.11 and 64.96 (2-theta) respectively. Nickel addition to cobalt oxide maintains the spinel structure up to its solubility limit when it separates out to form the NiCo2O4 spinel and a separate cubic NiO phase detectable by XRD. According to Windisch et al. this occurs when nickel concentrations exceed 33% in solution deposited samples a nd 50% in sputter deposited films. All samples experience nickel oxide precipitati on when heated above four hundred degrees (Tareen et al., 1984; Windisch et al., 2001a), however Petrov and Will demonstrated that heating to 1000C in oxygen and KClO3 at a pressure of 60kbar reversed the phase separation to form spinel structured Ni1.71Co1.29O4 (Petrov & Will, 1987). 2.4.1 Spinel Sites The unit cell from Figure 2-12 is made up of a series of repeating layers illustrated in Figure 2-14. Using the nomenclature AB2O4 to represent cations and oxygen in the normal spinel structure, the A represents the doubly ionized ca tion found in the

PAGE 55

36 tetrahedral site, the B represents the triply io nized cation found in the octahedral site, and “O” represents the oxygen occupied sites according the space group F d 3 m #227. (Azaroff, 1960) Figure 2-14. Spinel unit cell la yers stack with interlocking tetrahedral sites. Oversized green tetrahedral atoms (A) in the number ed layers stick up and connect in the vacancies of the level above. This layered schematic is based on a similar spinel layout detailed by Kingery (Kingery et al., 1975) p.65. Tetrahedral site. The tetrahedral site (Figure. 215), or four-fold coordination site accommodates a cation which bonds with four n earest neighboring oxygen anions. Eight out of 64 possible tetrahedral sites are occupied by a doubly ionized cation. Cobalt is believed to occupy the tetr ahedral sites in the nick el-cobalt oxide lattice.

PAGE 56

37 Figure 2-15. Tetrahedral site atom (green ) bonded to four surrounding oxygen (black). Figure 2-16. The octahedral site atom (yellow) bonds with six surrounding oxygen (black) anions to form a six fold orientation. Octahedral site. The octahedral site (F igure 2-16), or six-fold coordination site, typically is occupied by a cat ion with a 3+ valence to bond with its 6 nearest oxygen anions. The spinel unit cell contains 32 octa hedral sites, of which 16 are occupied. Nickel prefers the octahedral site in the nickel-cobalt ox ide (Tareen et al., 1984), and cobalt fills the remaining octahedral sites and then the tetrahedral sites.

PAGE 57

38 The octahedral site plays a role in conductiv ity. As nickel is added to cobalt oxide, the conductivity increases by up to five orders of magnitude and nickel occupies the octahedral site. However after heat tr eatment, nickel oxide precipitation causes conductivity to decrease, sugges ting that nickel in the octa hedral site must have a significant effect. Oxygen site. Oxygen occupies the remaining thirty -two anion sites. Oxygen sites are nearly a close packed cubic lattice (Kingery et al., 1975). 2.4.2 Variations of the Spinel Structure Not all spinel structures arrange as noted by the AB2O4 with the A2+ in the tetrahedral site and the B3+ atoms in the octahedral site. The inverse spinel sometimes occurs, where half of the B3+ species occupy the tetrahedral si tes. The remaining half of the B3+ cations along with all of the A2+ cations occupy the octahe dral sites. Inverse spinel structures are often iden tified structurally as B[AB]O4 (Azaroff, 1960). Other distributions of cations between the various lattice sites are possibl e (Deer, 1962; Kingery et al., 1975). “The partially inverse spinels can be view ed as partially disordered versions of either end member, the ideal normal spinel or the ideal inverse spinel. The disordered structures contain lattice defect s relative to either end member in that cations are located on lattice sites where th ey do not appear in the ideal reference structure.” (Smyth, 2000, p.20) Nickel and cobalt compose an oxide sp inel where both cations are found in multiple valence states and in multiple sites. This condition of a mixture of multiple valence states in multiple sites is referred to as cation disorder. Verwey et al. postulated a cation charge ordering within iron oxide at a low temperature (Verwey 1939; Verwey & Haayman, 1941; Verwey et al. 1947). Be low the Verwey temp erature, conductivity decreases dramatically (Monk & Ayub, 1997). Much like Verwey’s postulate, cation

PAGE 58

39 arrangement must have an effect on polaron interaction. Hopping between sites will depend on whether the sites are dissimilar or bot h tetrahedral or octahedral. Arranging as primarily an inverse spinel (Windisch et al ., 2001b), the nickel and cobalt cations may substitute for each other at both the tetrah edral and octahedral lattice sites. If arrangement has an effect, then the opposite must be true as we ll. Defects and disorder of the cation charge distribution within the spinel structure also has an effect on the conductivity (Windisc h et al., 2002a). 2.4.3 Spinel and Conductivity The electrical conductivity for these spinels is generally very limited. In general their bandgaps are large (>2.5 eV) and impurities form deep traps for charge carriers. To understand the limited observed conductivity, th e electronic bonding must be considered. As reported by Smyth et al., for the iron oxide system the following is observed: “In an octahedral environment, the crys tal field splitting gives three equivalent levels of lower energy, the t2g levels whose orbitals are directed in between the six closest anions, and two equivalent levels of higher energy the eg levels whose orbitals point directly at nearest neighbors. The energy separation in this case is sufficient to promote a violation of Hund’ s rule (that “equivalent” electron states are singly filled before any of them ar e doubly occupied), and the six d electrons fill the three lower levels.” (Smyth, 2000, p.18). Smyth concludes that electrical conduc tivity is achieved by carrier movement through the bonding or bitals or t2g orbitals along the octahedral sites. Conduction may be due to movement of holes in anion p -orbitals. (Ballal & Mande, 1977) 2.5 Summary of Literature Review TCOs are classified by their majority carri er type or conduction mechanism. Both n -type and p -type TCOs are needed, but n -type are more developed and available. The primary reason p -type TCOs are not as developed is a result of the conductivity being orders of magnitude less than n -type TCOs. Nickel-cobalt ox ide conductivity is order of

PAGE 59

40 magnitudes better than other p -type TCOs, yet is still poor in comparison to other well used n -type TCOs. N-type TCOs block light in the IR while nickel-cobalt oxide transmits it. N-type TCOs are free electr on conductors with high mobility and nickelcobalt oxide is a polaron hopping conductor with low mobility. Highly localized charge carriers and the accompanying la ttice strain known as small polarons are found in nickelcobalt oxide and behave distin ctly different than the free carrier analog in other oxide semiconductors. Most often polaron hopping is “discovered” as the conduction mechanism by observing material properties. Key differences between free carrier conductors and small polaron conductors ar e evident from temperature dependent properties such as conductivity and the S eebeck coefficient. Conductivity becomes significant in small polaron conducting material when the carrier c oncentration becomes extremely large. Polarons conduct thr ough a charge hopping mechanism. Polaron formation does not create a plasma absorpti on region therefore IR transmission remains high. Properties exhibited by nickel cobalt ox ide such as high conductivity and infrared transparency result from the spinel st ructural arrangement and the high polaron concentrations. Hopping is a low mobility pr ocess that requires energy for activation. Achieving conductivity improvements is approached in three ways: Increase the cation disorder and lower activation energy by adjusting film deposition conditions and post depos ition heat treatment processing. Add polarons by doping the system with a monovalent impurity atom such as lithium to further oxidize metallic ions and create more polarons Substitute rhodium for cobalt to increase disorder by atomic size distortion. These methods of enhancing conductivity by are reported in the Chapters 3-5.

PAGE 60

41 CHAPTER 3 SPUTTERED NICKEL-COBALT OXIDE 3.1 Introduction This chapter discusses the electrical a nd optical properties of nickel-cobalt oxide and how sputtering conditions and heat treat ment conditions affect them. Changing the growth rate or nucleation m ode (growth mechanism) by ad justing sputtering conditions may influence the film qualities such as cr ystallinity, morphology or density (Mattox, 1998). The effects of introducing a third cation such as lithium or rhodium on thin film electrical and optical properties are disc ussed in Chapters 4 and 5, respectively. Two primary methods used to deposit nickel -cobalt oxide as infrared transparent conducting oxide (ITCO) thin films incl ude solution deposi tion and sputtering.* Conductivity and transparency both vary de pending on the method of film deposition. The solubility limit of nickel in cobalt oxide appears to be a func tion of the deposition method. Nickel solubility in cobalt oxide is enhanced by sputteri ng and superior film conductivity is observed at increased nickel concentrations. Solution deposition is limited to a maximum concentration of 33% nickel to produce the stoichiometric NiCo2O4 spinel composition. Sputtered samples allow a nickel con centration of up to 50% to give Ni1.5Co1.5O4. Windisch et al. have explai ned the superior conductivity (order of magnitude improvement) of sputte r deposited films in comparison to solution deposited films (Windisch et al., 2001a; Wi ndisch et al., 2001b; Windisch et al., 2002b). Refer to Appendix A for additional details of sputtering and solution deposition

PAGE 61

42 A technique of sputter deposition that produces a compositionally varying film (combinatorial sputtering) used in this study produced a fine r range of film compositions to find the highest conductivity composition. This combinatorial sputtering deposition technique was repeated to study the effect of select gas compositions on deposited film properties. Traditi onal sputtering of Ni1.5Co1.5O4 and NiCo2O4 from alloy targets show how process conditions of film deposition such as to gas pressure and target-substrate distance affect electrical, optic al and structural properties. Heat treatment methods are shown to have a dramatic effect on film properties as well. 3.2 Film Preparation and Characterization Procedures Radio frequency (RF) sputter deposition at 13.56 MHz with a magnetron cathode allows tight control of process variables su ch as cathode power, s ubstrate temperature, gas flow rate, gas composition, gas pressure and target-substrate distance inside the vacuum chamber to produce consistent films ove r large areas. This study analyzes films sputtered with process adjustments such as sputtering gas composition, target-to-substrate distance, and gas pressure. Resulting optical transmission and electrical conductivity will be reported as a function of th ese sputtering process conditions. Sputtering for this study occurred in one of two configurations: (1) the combinatorial film deposition method used two cathodes simultaneously to generate many compositions in a single run, and (2) the traditional sputtering method used a single cathode and a rotating substrate holder. The first sputter deposition configura tion uses two cathodes simultaneously to produce a film with varying composition. Co mbinatorial sputtering in Figure 3-1 (also referred to as combinatoric sputtering (Freeman et al., 2000)) produced films containing a graded composition between nickel oxide a nd cobalt oxide in an attempt to find the composition of a nickel-cobalt ox ide film with the highest conductivity. The basic setup

PAGE 62

43 included three microscope slides placed endto-end, divided into ni ne one-inch sections, and labeled 1-9, as shown in Figure 3-1. A ll sputtered films were deposited with a base vacuum pressure near 1x10-6 Torr. The process gas pressu re was held constant at 2 mTorr for the combinatorial experiments. Figure 3-1. Combinatorial sput tering uses a dual cathode setu p with targets of different material composition. The resulting co mpositionally graded film is divided into subsections and numbered 1-9. Sapphire, microscope slides and/or silicon wafer slices served as substrat es for combinatorial deposited films. Factors such as background contaminati on within the chamber (e.g. pump oil, previously sputtered material, or gas line impurities) may cause slight film variation from one process run to an identical one. By us ing the combinatorial technique, an array of compositions can be fabricated simultaneously with identical deposition conditions to minimize incidental error from one run to the next. Combinatorial sputter deposition does have its own complexity concerning uni form film thickness attributed to the geometry of the setup and the different materials used as ta rgets. Uneven film thickness

PAGE 63

44 is minimized by adjusting the sputtering rate of the individual cathodes by depositing a film and then changing the cathode power and verifying the effect. During calibration, all other process parameters are held constant. The second sputtering configuration is ca lled traditional sputtering and includes a single two or three inch cathode powered at 100 or 200 watts, respectively. Process gas pressure and target-substrate distance are al so varied. Sputtering time is adjusted to produce films of similar thickness from setups with the varying target -substrate distance. The substrate holder setup for single cathode de position rotated primarily in an offset rotation setup where the axis of rotation is offset from the center axis of the cathode (see appendix A for clarification on sputtering geometry). Samples of NixCo3-xO4 were produced on fused silica, p -type silicon, sapphire, poly( ethylene-teraphthalate) (PET), and microscope slides with x being equal to 1 or 1.5. Characterization results consist of data from both combinatorial and traditional sputtered films. The matrix for the experiments is shown in Table 3-1. Table 3-1. Film deposition paramete r and sputtering setup matrix for NixCo3-xO4 Parameter Setup Variable range Film Composition Combinatorial 0.8 < x < 1.75 Gas Composition Combinatorial 0, 20, 50, 75, 100% O2 Gas Pressure Traditional 2, 5, 10 mTorr; x=1, x=1.5 Target-Substrate Distance Traditional 7.5, 10, 15, 30 cm; x=1.5 Characterization included x-ray photoelectron spectroscopy (XPS) *(Brundle et al., 1992), secondary ion mass spectroscopy (SIMS)* (Brundle et al., 1992) transmission See Appendix B for instrumental setup parameters and operating conditions.

PAGE 64

45 electron microscopy (TEM) *(Brundle et al., 1992), x-ray diffraction (XRD) (Brundle et al., 1992), Fourier transform in frared spectroscopy (FTIR)* (Brundle et al., 1992), ultraviolet and visible spectroscopy*, optical and stylus profilometry*, and room-temperature and temperature-dependent van der Pauw* measurements. Heat treatment consisted of a ten minute post-deposition heat treatmen t at 375C followed by rapid cooling. XPS was used to determine the chemical composition of select films and then correlate the other characterization informa tion to a specific composition. Electrical van der Pauw measurements were performed w ith films on all substrates, while FTIR transmission measurements were performed with films on silicon and sapphire substrates. Visible and near-infrared optical spectra were measured from films on fused silica and/or sapphire substrates. 3.3 Characterization Results and Discussion 3.3.1 Electrical Properties Variable composition films from combinat orial sputtering show the effect of process gas composition while traditionally sp uttered films from alloy targets show the effects of total chamber sputter gas pressure a nd target to substrate distance on electrical properties. Substrate material has an effect on the produced film regardless of the film composition and the work function does not appear to vary a significantly with composition. Gas composition. The effects of gas composition, with the gas mixtures ranging from 100% oxygen to 100% argon, at a constant pressure of 2 mTorr are shown in Figure See Appendix B for instrumental setup parameters and operating conditions.

PAGE 65

46 3-2. Films deposited at constant gas pressure with different gas concentrations of argon and oxygen show that position 7 ha d the highest conductivity of ~350 cm. Figure 3-2. Conductivity versus position for comb inatorial runs is displayed as a function of oxygen and argon gas composition. The film deposited with the gas concentration of 50% oxygen and 50% argon is the most conductive at position 7. Combinatorial sputtered films from nickel oxide and cobalt oxid e targets deposited with variable gas composition showed marked changes in resistivity as the amount of argon increased up to 50%. The gas with 50% oxygen and 50% argon yielded the highest conductivity across all positions. Resistivit y degraded when the argon concentration exceeded 50%. Gas composition appeared to affect the sputter rate and possibly the oxidation state of the cations in the produced fi lm. Argon, with its la rger mass, inflicts a greater amount of damage on the target, in creasing the sputtering rate compared to oxygen, likely producing a slightly reduced f ilm. Oxygen however promotes complete oxidation of the film. Combining the eff ects of oxygen and argon produced the highest

PAGE 66

47 conductivity films. Figure 3-3 shows positi on 7, the highest conduc tivity position of all the films (~350 cm), plotted as a function of gas pressure. Figure 3-3. Position 7 from Figure 3-2 is the position with the highest conductivity. Conductivity is enhanced at a sputte r deposition process gas composition containing 50% oxygen and 50% argon. This combinatorial study was conducted to fill the gaps between the data points previously reported (Figure 2-8) where the composition with highest conductivity film contains equal parts of nickel and cobalt. Combinatorial sputtering with 5 cm nickel oxide and cobalt oxide targets produced a film with a graded composition as determined by XPS (Figure 3-4). Possibly due to different magnetic species in the plasma, more cobalt is measured in the sample in Figure 3-4. An ideal deposit ion would show an equal ratio of nickel to cobalt at position 5. It is inte resting that while proportionally mo re cobalt is in this film, shifting the equal ratio composition Ni1.5Co1.5O4 to the nickel-rich side, the data show that the nickel-rich side of the film is actually thicker. This suggests that the nickel oxide target and the cobalt oxide ta rget sputtered at different ra tes. Nickel oxide, being a nonmagnetic target, likely had a higher de position rate, but being influenced by the magnetic field of the cathode, deposited mate rial in a more confined region. The magnetic nature of the Co3O4 target may have interacted w ith the magnetic field from the

PAGE 67

48 cathode to alter its effect on the desired ion trajectory and produce a more diffuse deposition resulting in a lower sputte r rate and a higher area of coverage. Figure 3-4. Fraction of Ni and Co dete cted by XPS as a function of position on a combinatorial sputtered NixCo3-xO4 film. Equal portions of nickel and cobalt were found at position 7 (x value of 1.5). The blue arrow indicates the solubility limit of nickel from soluti on deposition. The light blue shaded region shows the increased nickel ava ilable from sputtering. Embedded along the abscissa of the graph is a picture of a combinat orial sputtered film on a silicon substrate from this study. The co lor change is due in part to thickness and in part to compositional variations. XPS of the film combined with resistivity and thickness data, each as a function of numbered position, yields a relationship between composition and conductivity. XPS of the most conductive combinatorial film is s hown in Figure 3-5. Film conductivity is found to be highest at a composition of Ni1.5Co1.5O4 at ~350 Scm-1, agreeing with previously reported results (Windisch et al., 2001a). The composition limit imposed by solution deposition is included for contrast. All compositions to the right of the red vertical line are not possible from solution deposition. As an indication of the different sputteri ng rates of nickel an d cobalt, a thickness calibration profile shows that the nickel-rich side is thicker than the cobalt-rich side (Figure 3-6). Interference due to the magne tic nature of the cobalt oxide target may contribute to the non-uniform de position thickness. Sputtering RF power and tilt angle of

PAGE 68

49 the individual cathodes would re quire fine tuning to flatten the thickness curve. This iterative process of finite adjustment should be addressed in a future study. Figure 3-5. Combin atorial sputtered NixCo3-xO4 film combining conductivity and position with composition with position data to yield c onductivity as a function of composition. The highe st conductivity film appears at a composition near Ni1.5Co1.5O4. The vertical red line at x=1 indicates the solubility limit of nickel when deposited from solution methods. Figure 3-6. A combinator ial substrate thickness prof ile measured by a stylus profilometer shows that the film thickne ss is also graded from cobalt rich on the left to nickel rich on the right. Followings an initial calibration run, cathode power was adjusted as was depos ition time to give a more uniform film. Subsequent combinatorial films were processed using this calibrated power regime. Figure 3-6 shows the initial calibration run (run 1) with a non-uniform thickness of about 30 nm on the cobalt side and 100 nm on the nickel side. The power was adjusted

PAGE 69

50 and the second deposition ran with a 25% increase in the cobalt cathode power and the nickel cathode power reduced by 75%. Sputte r time was doubled to ensure a film of similar thickness. No further increase in the cobalt cathode power was practical because a general rule of thumb fo r sputtering (specified by one cathode manufacturer) discouraged exceeding a power rating of 4 Wcm-3 on a 5 cm cathode. Since the thickness difference was now a factor of 2x instead of nearly a factor of 4x, it was deemed acceptable. Chamber pressure. Using the composition from the combinatorial run with the lowest resistivity, an alloy target with equa l parts of nickel and cobalt was reactively RF sputtered in 100% oxygen at 10, 5, and 2 mTorr in the traditional setup with an offset rotating substrate holder at a distance of 10 cm. A mixed gas composition was not used for reactive sputtering. Film s in Figure 3-7 deposited at 2 mTorr maintained lower resistivity before and after heat treatment. No te that the variance is less than a factor of two and that the best resistivity is near 2 m cm (a conductivity of 500 Scm-1). Figure 3-7. The effect of sputtering gas pressure on NiCo alloy sputtered Ni1.5Co1.5O4 thin film resistivity before and afte r 10 minutes at 375C heat treatment. While the results for the as-deposited samples are not in complete agreement, the general trend shows that the lower pressure in the range studied is better increase

PAGE 70

51 conductivity. The result of decr eased resistivity at a lower pr essure this system exhibits is not surprising, but was previously unknown. Molecules in a vacuum travel an average distance between molecular collisions which changes with pressure. This molecular mean free path ( ) in Equation 3.1 has units of centimeters when pressure (P) is in units of Pascals (O'hanlon, 1989) and is determined by the number of molecules o ccupying the swept out volume. Higher pressure means more molecules per unit volume and a lower mean free path. 6.6P (3.1) In the vacuum chamber this becomes important as the molecules being sputtered from the target surface collide with and lose energy to other molecules before reaching the substrate for film formation. Pressure ch anges directly influence this mean free path. The experimental parameter of distance between target and substrate may also affect the energy of depositing atoms in a similar fashion. Figure 3-8. Target-substrate distance affect s the resistivity of sputtered nickel-cobalt oxide thin films from a NiCo alloy target. Resistivity increases with distance. Target-substrate distance. Increasing target-substra te distance (Figure 3-8) allows a larger area of deposition and enables uniform coating of multiple substrates with the proper substrate motion (see Appendix A for details on planetary rotation). Increased target-substrate sputtering distance results in increased resistivity. Discussion of the

PAGE 71

52 effect evident by these data follows in the Optical Properties Section (3.3.2) and in the Film Structural Propert ies Section (3.3.3). Substrate material. Substrate material in some cases had an effect on conductivity. Poly(ethelyne-ter apthalate), a polymer substrate used for OLEDs, had a detrimental effect on film conduc tivity. It is believed that carbon from the PET substrate reduced the nickel-cobalt oxide film somewhat which resulted in the observed decrease in conductivity. Figure 3-9. Conductivity of NixCo3-xO4 films on PET substrate is always less than films on other substrates by a factor of 2-4x for as-deposited samples for all pressures and compositions shown. Work function. Sputter condition changes to nick el-cobalt oxide may also affect the work function, a property not reported in the literature. From UPS data, a work function was calculated by with Equation 3.2. An ultraviolet He I source with an energy of 21.218 eV (+ or 0.001 eV) excites electrons across the band gap. A bias voltage (VB) of two different values was applied and the energy (EVBcutoff) is extrapolated from Figure 3-10 (energy on the left side of the peak where inte nsity is zero). The work function is calculated by ta king the known photon source ener gy, subtracting the bias

PAGE 72

53 voltage, and then subtracting the measured va lence band cutoff value. The detector work function is important to know al so, but is calibrated out of this data and therefore not included in Equation 3.2. hv VB E VBcutoff (3.2) Figure 3-10. UPS work function measurement of NiCo2O4 and Ni1.5Co1.5O4 films. Higher nickel content shows a slightly lower work function. Work function values may track inversely with conduc tivity or increased nickel content. Compositional differences of oxide films from sputtered NiCo and NiCo2 alloy targets does not appear to have a dramatic effect on the work function, These results suggest that the work function may vary sl ightly with composition or conductivity. Assuming the variance is with conductivity, the work f unction would change with processing conditions. Additional work would be required to quantif y this assertion. However, given the small change due to co mposition in the work function it is doubtful that processing conditions such as sputtering pressure or sputtering distance would have a significant impact on the work function value.

PAGE 73

54 3.3.2 Optical Properties Changes in processing of TCO films affects optical properties though with the opposite general trends of the electrical propert ies. Variation of op tical transmission as a function of film composition has been re ported previously (Windisch et al., 2001b). Incident light impinging on a sample will become a sum of several interactions, including transmission, absorption, reflec tion, and scattering (Equation 3.3). Often absorption and scattering are assumed to be zero and the calculation simplifies to three terms. I 0 I trans I abs I refl I scatter I 0 I trans I refl (3.3) Gas composition. Due to the thickness variatio n in the combinatorial sputtered films shown in Section 3.2.1, variations in transmission data will not necessarily represent changes with film composition or gas composition, so such data have been excluded from the optical property analysis The optical transmission measurements would likely yield useful information if the films are of uniform thickness. Chamber pressure. For films of the same com position deposited at different pressures, optical transparency was seen to vary from 5-10%. Transparency decreases with pressure (Figure 3-11) and increas ed conductivity, agre eing with reported conductivity and transparency relationships (Windisch et al., 2002b). Lower pressures from both compositions (x=1.5 and x=1) exhi bit lower transmission. The absence of correction for silicon substrate limits the ma ximum transparency to 50%, but even with correction, the effect would be similar. F ilms on silicon substrates, shown in Figure 3-11, show that the transparency of the nickel-cob alt oxide films in the nearto mid-infrared regions (corresponding to light wavelengths of 2.5 m to 25 m) if normalized to 50% would be well above that value.

PAGE 74

55 Figure 3-11. FTIR traces from of reactively sputtered NixCo3-xO4 thin films. (a) x = 1.5 (from NiCo alloy target) and (b) x = 1 (from NiCo2 alloy target) show the effect of pressure on transmission. Films nominally 50 nm thick are shown as a function of sputte ring gas pressure. The decrease in transparency seen with decreased sputtering pressure may be due to a change in the index of refraction of the deposited film likely caused by a chemical or structural change induced by the processi ng conditions. Changes in the index of refraction may result in a 3% change in transmission due to an increased absorption/damping region likely attribut able to defects in the lattice. The index of refraction shown in Equation 3.4 (n), is the ratio of the speed of light in a vacuum (c) divided by the speed of light in the material (v). n c v (3.4) A difference in index of refraction of tw o different materials across an interface results in a loss of transmitted light due to reflection at the interface. The reflected intensity is expressed in Equation 3.5 where n1 and n2 are the indices of refraction for the two materials. Percent reflection is the intensity of reflected light (Irefl) divided by the incident light (I0) multiplied by 100. This expression is valid for transparent materials and neglects damping. Complex values of n could be used in the expression to take damping into account.

PAGE 75

56 Irefl I0n1 n2n1 n2 2 (3.5) Figure 3-12. Optical transmission through a thin film. On a substrat e, it requires a low index of refraction mismatch at all in terfaces (to avoid complete reflection) and a low damping coefficient (to av oid extinction) for the particular wavelength of transmitted light (this pi cture does not show the effects of damping). The index of refraction is important for TCO films because the light must pass through three interfaces as illustrated in Figure 3-12. A significant difference in index of refraction at any of these inte rfaces will result in increased optical reflection or damping with an associated loss in transmission. Lo sses due to reflection from index mismatches of nickel-cobalt oxide (n = 2.6-2.8) on a silicon substrate (n = 3.5) exceed 50% when damping is assumed to be zero. Absorp tion or damping mechanisms may include electron excitation with phonon emi ssion (activation of lattice vibrations) and in reality are not zero, but are assumed to be so in Table 3-2. Assuming that the process pressure causes a shift in the index of refraction from 2.6-2.8, the shift in transparency would be less than 3 percent (see calculation in Table 32). A shift greater than 3% could be attri buted to a difference in thickness or possibly scattering from process-induced morphology ch anges, such as increased grain size or grain boundary area.

PAGE 76

57 Table 3-2. Values of index of refraction fo r films and substrates are shown individually with calculated transmission losses due to interface index mismatch assuming no absorption or extinction. n of Film Substrate n of Substrate % Loss in Transmission NiCo2O4 2.6 Si 3.5 (1370 nm) 52.8 NiCo2O4 2.7 Si 3.5 53.6 NiCo2O4 2.8 Si 3.5 54.5 NiCo2O4 2.6 SiO2 1.46 (300-800 nm) 31.13 NiCo2O4 2.6 Al2O3 1.77 (300-5000 nm) 26.85 Target-substrate distance. The distance of sputtering between the target and the substrate also had an effect on the optical transmission from which an optical absorption coefficient is calculated. The absorption coefficient (), given in Equations 3.6 and 3.7 multiplied by the film thickness (d) is proportional to th e intensity of light (Itrans) detected after the beam passes through the film: Itrans=I0ed (3.6) Percent transmission is the ratio of measured intensity (Itrans) normalized to the background intensity (I0) and then multiplied by 100. When transmission has been measured at a specific wavelength and the film thickness is known, an absorption coefficient can be calculated (Equation 3.7). 1d ln ItransIreflI0 (3.7) Care must be taken to account for reflection (Irefl) at the air-film, film-substrate, and substrate-atmosphere interfaces in order to obtain an accurate absorption coefficient (interfaces and reflectivity are shown in Figure 3-12).

PAGE 77

58 The optical absorption decreases in Figure 3-13 with increased target to substrate distance. The resistivity trace from Figure 3-8 is included for contrast to show that the electrical properties are affected opposite to the optical properties. Figure 3-13. Target-substra te distance effects are opposit e for the optical absorption coefficient and resistivity from sputtere d nickel-cobalt oxide thin films from a NiCo alloy target. As distance increases, decreases. The optical absorption coefficient was calculated from a transmission measurement at 3 m (see Appendix C for absorption coefficient calculation details). Electrical band gap. Sputtering condition changes will likely affect the transmission window of the produced films. Optical tr ansmission spectra are often described by the electrical band gap, or energy of the forbidden region between the valence band and the conduction band. This band gap energy allows photons of a lower energy to pass through the material without ex citing electrons from the valence band to the conduction band causing film coloration. Band gap values extrapolated from the Tauc’s plot in Figure 3-14 consist of the square of the absorption coefficient multiplied by the photon energy plotted as a function of th e photon energy. The plot is extrapolated down to an energy value at the x-axis, which is equal to the optical band gap. Band gap energies are shown in Table 33 as a function of nickel conc entration. The effect on the

PAGE 78

59 band gap from processing may be affected by process conditions, but that has not yet been determined nor is it availa ble elsewhere in the literature. Figure 3-14. Tauc’s plot of NixCo1-xO4, films show a band gap between 3 and 3.75 eV. Table 3-3. Extrapolated band ga p values lines in Figure 3-13. Ni1.5Co1.5O4 NiCo2O4 Ni0.75Co2.25O4 Band Gap (eV) 3.2 3.4 3.5 3.3.3 Film Structural Properties The electrical and optical pr operties are dictated by the atomic structure and grain structure of the films. Structural changes of different target-substrat e distances affect the film properties. Basic film properties such as surface and bulk composition are also briefly discussed in this section. Target-substrate distance. Increasing target-to-s ubstrate sputtering distance decreased absorption and increased resistivity for films of similar thickness. Modeling of x-ray diffraction data of NiCo2O4 films deposited 5 and 15 cm from the target show a decrease in density as sputtering distance increased (see Figure 3-15). The distance a sputtered metallic ion must travel through the oxygen plasma to get from the target to the substrate will determ ine its oxidation state and kinetic energy when

PAGE 79

60 it reaches the substrate. Th e sputter deposition rate, or number of energetic particles incident on the surface, decreases with the squa re of the difference in distance. The ion flux from the target to the substrate will decrease at longer distances as seen by the fading of the plasma color from the cathode in Figure 3-16. Figure 3-15. Increasing target to substrate sputtering distance decreases film density. Increased porosity is believed to de crease electrical conduc tivity and increase optical transparency. Figure 3-16. An oxygen plasma shows a confin ed plasma region (white plume) near the target and a more dispersed plasma re gion (yellow-green) at longer distances. Two factors may influence the film density with distance: (1) temperature of the substrate and (2) the mean free path of the ga s. Substrate heating is affected by distance in two ways: (a) a longer dist ance would allow more time for oxidation reaction in the

PAGE 80

61 plasma with less heating of the substrate fr om the exothermic oxidation reaction, and (b) a longer distance reduces the ener gy transfer attributed to th e plasma-substrate interaction (the plasma is most energetic near the cat hode surface). The two effects combined yield a lower density film at longer distances because a lower substrate temperature will not support surface diffusion required to increase the film density. Decreased target-substrate distance shows an increase in conduc tivity likely due to the opposite effect. Oxidation of cobalt metal to cobalt oxide or the nickel metal to the nickel oxide on the surface along with increa sed surface-plasma interaction could provide heat to the surface allowing gr eater surface ion mobility to obtain an energetically favorable atomic arrangement. Increased substrate temperature allows diffusion for creating a denser matrix. Figure 3-17. The three zone model describe d by Campbell for film growth in a vacuum (Campbell, 1996). Campbell elaborates on a thre e zone model that explains the deposited films. Figure 3-17 shows a schematic of the regions of film that are affected by the substrate temperature and the incident ion energy described by Campbell.

PAGE 81

62 “At the lowest temperature and ion energy, the film will be an amorphous, highly porous solid with a low mass density. This is the first zone of the diagram. It is caused by the low adatom mobility of the growing film. Metal films deposited in this region can readily oxidize when expos ed to air and so may also have high resistivities. If the chamber pressure is lowered or the substrate temperature is raised, the deposition process enters the “T” zone. Films depos ited in this region are highly specular and have very sm all grains. For may microelectronic applications, this is the most desira ble region of operation. Increasing the temperature and/or impinging energy furthe r cause the grain size to increase. The second zone has tall narrow columnar grains that grow vertically from the surface. The grains end in facets. Finally, in Zone 3, the film has large 3-D grains. The surfaces of the films in the second and third zones are moderately rough and the films appear milky or hazy ,” (Campbell, 1996, p.299) Films deposited at closer distances are lik ely shifting to the T region from region 1 with an increase in temperature from the heat transfer of the plasma, and from the ions being more energetic in the plasma. As the pressure decreases, the effect is similar. Figure 3-18. A spinel unit cell at different angles of rotati on containing octahedral atoms, minimized tetrahedral atoms, and sel ect oxygen atoms for points of reference such as the enlarged black oxygen atom. Lines along octahedral sites are to simply illustrate a crisscrossing three-dimensional network with possible routes for conduction. Conductivity in the spinel structure may be relatively high due to the multiple pathways for carrier movement to occur. Th e cation arrangement of the spinel structure allows isotropic electrical c onduction that is not restrict ed to a single dimension,

PAGE 82

63 direction, or a preferred plan e. Conductivity likely occu rs in all three-dimensions through the octahedral cation network, whic h forms crisscrossing lines as shown in Figure 3-18. This simplified illustration shows only select reference oxygen ions. The actual network pathways along the octahedral-oxy gen ions would be similar, but the path would include an alternating oxygen anions along the path between each cation. Closer target-substrate di stance may produce more of the “cross-linked” conduction lines connections as the film is denser. Greater distance would e quate to more broken conduction lines from pores or voids. Vary ing the distance and angle of incidence may also change the stoichiometry in some cases, which will result in a change in transparency and conductivity. Slight stoi chiometry deviations such as increasing oxygen content with longer targ et-substrate distances may be a possible explanation to the observed property changes. Film structure. The grazing incident x-ray diffr action traces shown in Figure 3-19 appear to be spinel, but do not match the di ffraction pattern exactly. X-ray scans of the sputtered nickel-cobalt oxide films are simila r to the reference powder diffraction file 731702. As such, these films are classified as spinel-type. Deviations occur in peaks near 60. The peak at 55.5 does not exist in the green trace and both of the 59.5 and the 65 peaks appear to be slightly shifted to lower angles. As suggested earlier, disorder aids in polar on formation with an associated increase in conductivity. An understanding of thes e deviations and abnormalities may provide insight on carrier generation to further impr ove film properties such as conductivity. Distortions in the lattice may be due to one or both of the following effects: the Jahn–

PAGE 83

64 Teller effect (Dionne, 1990), or the Verwey transition (charg e ordering) effect (Verway & Haayman, 1941; Verwey et al., 1947; Verwey, 1939). Figure 3-19. Grazing incident x-ray diffraction from sputtere d films 50 nm thick appear to be structured as spinel-t ype, but do not match exactly. The Jahn-Teller effect is a distortion in bond length that occu rs allowing the t2g or eg orbitals to shift energy a nd lose degeneracy to obtain lo wer system energy. It is possible that the cubic spinel lattice may slightly distort to contain some tetragonal unit cells and still be spinel-like (Blasse, 1963). Ca re must be taken in attributing this shift only to the Jahn-Teller effect because it ma y also be due in part to a Verwey-type transition effect. Verwey postulated that the cation arrangeme nt and charge distribution with respect to the tetrahedral and octahedral sites being occupied may change at a given temperature. The spinel may shift from normal to inverse. Verwey’s system was the Fe3O4 spinel. Since nickel-cobalt oxide is isostructural with Fe3O4, it is possible that the Verwey See Appendix C for more information

PAGE 84

65 transition may occur during heated processing of these thin films which will be discussed in Section 3.3.4 (Verway & Haayman, 1941; Verwey et al., 1947; Verwey, 1939). Distortions can affect the film properties although the qualitative knowledge at this point suggests that distortion may have a role, but exactly what that role is has not been determined. Literature data on these nickel-c obalt oxide thin films has concentrated on surface sensitive techniques such as XPS for st oichiometric determination. Bulk film composition may aide in determining th e role of distortion or its origin. Film surface and bulk composition. SIMS data in Figure 3-20 shows that the surface composition may be different from the bulk, probably due to the well understood SIMS shift from preferentia l sputtering making the film surface and interface appear compositionally incorrect. Bulk studies may reveal more information regarding the film composition effects on the optical and electr ical properties. Th e film surface shows nearly equal amounts of ni ckel and cobalt, while the bulk shows more cobalt. Figure 3-20. Sims profile of nickel-coba lt oxide thin film shows surface composition appears to differ from the bulk compos ition. XPS analysis for chemical composition on the surface agrees with SIMS data. Calibration standards were not available for this material. Effects such as preferential sputtering have not been cons idered, but may take into account phenomena

PAGE 85

66 that occur on the surface during sputter-etching. Sensitivity factors for the nickel-cobalt oxide matrix have been calculated using X PS as a SIMS calibration, however the results may be skewed due to the nature of sputter etching that occurs in the individual systems to remove material and acquire the depth profiles. XPS de pth profiling was done with an argon sputtering source to remove surface laye rs. SIMS was done with an oxygen source and therefore the oxygen is not included with the depth profile in Figure 3-20. The results may not accurately represent the materi al due to the possible preferential removal of either metal species from the exposed su rface during the depth profiling. SIMS did show that in the bulk the film composition is constant. Figure 3-21. Cross-sectional TEM image of sputtered nickel coba lt oxide films on (100) silicon. Nickel-cobalt oxide thin film grows in multi-grained columns. Micrographs from a transmission electron microscope (TEM) show that the films were homogeneous from the surface down to the interface. The interf ace is interesting. Notice the dark band that runs diagonally from top center to lower right side at the filmsubstrate interface in Figure 3-21. The white line is the native silicon dioxide layer found on silicon wafers and is typically 2 nm thick. The dark region at the interface may be a metallic growth layer and the cause of poor transmission and high conductivity. A highly

PAGE 86

67 conductive metal alloy film, thin enough to be transparent and undetectable by XRD could be part of the dark band at the inte rface surface. It may be a substoichiometric phase that acts as an interface layer on the silicon substrate. Another possibility is that it could include a metal-silicide layer of reacted metal with the surface of the substrate. Nickel silicide is used in silicon wafer fabr ication processing as an interconnect because of its high conductivity, but nick el or cobalt silicide optical properties have not been reported. Both are thermodynamically stable and could form as cations migrate through the silicon dioxide barrier laye r into the bulk wafer. The SIMS profile in Figure 3-20 does not agree with the data in Figure 3-22 wh ere increased concentr ations of cobalt is detected at the interface and substrate regions which would substantiate this claim. Figure 3-22. Nickel-cobalt oxide-silicon wafer interface. (a) Energy dispersion spectroscopy (EDS) of area the shown in (b), the S TEM micrograph of filmsubstrate interface scan region used for EDS analysis in (a). The interface layer is cobalt rich according to the EDS data from the STEM shown in Figure 3-22. The cobalt rich region may be part of a substoichiometric spinel film at the interface between the nickel-cobalt oxide and the silicon dioxide native layer on the surface of the silicon wafer. Closer inspecti on of the SIMS data in Figure 3-20 shows the

PAGE 87

68 cobalt content falls off and the nickel content remains steady before tailing off. It is possible that this is merely evidence of the preferential sputtering ar tifact. EDS does not rely on film layer removal for characterization as SIMS and is therefore not subject to preferential sputtering effect s. The cobalt rich region de tected by EDS is probably not indicative of silicide formation, otherwise the silicon trace in th e SIMS depth profile would have a spike in it repr esenting the metal-silicide layer was present. A closer examination reveals a change in slope of the silicon trace about mid way through the interface region. This slope change could indi cate a metal-silicide fo rmation, but it could also result from the native silicon dioxide la yer, which would have a different sputtering rate than unoxidized silicon. The data are best interpreted to mean that the film-substrate interface consists of a cobalt rich nickel-cobalt oxide la yer on the native silicon dioxide surface of the silicon wafer. 3.3.4 Post Deposition Heat Treatment Heat treatment of the nickel-cobalt oxide films result in increases in conductivity with concomitant decreases in transparen cy depending on the method used for heat treatment and the temperature involved As mentioned in Chapter 2, temperatures above 400C cause phase separation of nickel oxide w ithin the spinel film. Film properties degradation occurs over periods of days fo r as deposited and annealed samples left exposed to air at room temperature. Typical degradation of conductiv ity is a factor of two after one week. Successive heat treatments can return the film conductivity to nearly the same value as before the degradation if the cooling rate is on the order of ~150C/min. When the cooling rate is less than 15C/min, the film conductivity can be artificially aged to the degradation value had the samp le had been left exposed to air for

PAGE 88

69 several days. Transparency responds the oppos ite to conductivity after heat treatment activity. Heat and conductivity. Figure 3-23 shows a plot of the natural log of conductivity plotted against the reciprocal of temperature multiplied by 1000. As temperature increases from 300 K to 625 K, shown by th e blue traces for the two different composition films, the conductivity measured at temperature increases. These particular films were first heat treated at 375C (a 1000/T value of 1.54) for ten minutes and then rapidly quenched to room temperature be fore commencing the heated conductivity measurement. The conductivity incr eases with increasing temperature. Figure 3-23. An Arrhenius plot of conductiv ity and the reciprocal of temperature (K) shows that at high temperatures, the film conductivity is high. A temperature of 525 K graphed near 1.5 on the absci ssa divides the low temperature region on the right from the high temperature region on the left. The two regions have different activation energies for electrical conduction. The heating and cooling rate is limited to a maximum of 15/min to avoid instrument damage. Notice the difference in c onductivity at the same temp erature upon cooling from the highest temperature shown by the red cu rves. The room temperature conductivity upon cooling is always below its original value. There is a temperat ure where the heating and cooling data are nearly superimposed above, but depart belo w. Windisch et al.

PAGE 89

70 reported this anomaly of conduc tivity being different for two regions of slope associated with each of the heating and cooling curves, and have different slopes versus temperature at high versus low values (Windisch et al., 2002b). Figure 3-24. TEM image and diffracti on patterns at 300 K and 600 K showing no detectable structural change s for the two temperatures. Their speculation was that the transition c ould be the result of a Verwey transition near 525K that accounts for the change in sl ope of both the heating and cooling curves. Verwey et al. described the a temperature transition in Fe3O4 in terms of an order/disorder transition resu lting from the preferential or dering of the cations which changed the activation energy for electri cal conduction (Verway & Haayman, 1941; Verwey et al., 1947; Verwey, 1939). Rather than being a Verwey transition, it could be a

PAGE 90

71 structural transition, a magneti c transition or possibly a mobile species within the lattice that has a different hopping activation ener gy triggered at a higher temperature. The explanation for the effects of heat tr eatment is open to speculation. One such speculation invalidated by Figure 3-24 is that of a structural transformation. If a structural transformation occurred, the TEM diffraction pattern gene rated at 300 K would be different from the patter n at 600 K. No change is indicated, suggesting that no structural changes occurred and therefore rule s out the Jahn-Teller effect as a significant contributor. It is, however, possible that the metastable state quenched in relaxed out before the STEM was performed or was pro cessed out as the sample was prepared or during the analysis from localized electron beam heating. If the 600 K temperature experiment was conducted first then the stru cture is expected to be similar at 300K. Magnetic measurements would be require d to invalidate the suggestion of a magnetic transition. Samples were run on a vibrating sample magne tometer (VSM), but no significant hysteresis could be detected ei ther due to lack of a transition or more probably due to a lack of sensitivity because the film was too thin. A magnetic transition such as a spin state realignment triggered at a specific temperatur e may correlate with a Verwey transition in that the arrangement of the cations and ordering/disordering of their charges may contribute or detr act from a magnetic domain arrangement of the crystal. This is still unknown and could be the subject of a future study. The last suggestion was that a mobile speci es could be the cause of the change in electrical conductivity slope upon heating. The only mobile species in the lattice is oxygen. Oxygen is typically present as O2anions occupying the 32 available anion sites. Oxidizing conditions during s puttering and heat treatmen t may allow excess atomic

PAGE 91

72 oxygen to adsorb on the surface from dissociat ed carbon dioxide, water molecules, or molecular oxygen. Oxygen in th e spinel cell would not necessa rily be restricted to one site only and at elevated temperatures ma y have enough energy to site hop or form Frenkel defects involving oxygen ions (Callister 1997). In addition, a superstoichiometric film could result in the lattice in the presence of O1-. These oxygen defects could add additional el ectrical carriers to increase conductivity and will be further discussed below. Effect of heat treatment cooling parameters. Film cooling rate also has an effect on the observed properties. Figure 3-25 shows th e effect of cooling rate on transmission in the visible and near infrared regions. Optical transmission measurements were collected after quenching from the heat tr eatment described earlier, and after the temperature dependent conductivity meas urements shown in Figure 3-23. NixCo3-xO4 films where x = 1 are lower in conductivity overall than the films where x = 1.5. Figure 3-25. Effects of heat treatment cooling rate after heat treatment on optical properties of NixCo3-xO4 samples from Figure 3-23. Heating alone does not ensure high c onductivity. When cooled slowly, the conductivity degrades rather than improves. Conductivity can be in creased by up to an

PAGE 92

73 order of magnitude by quenching rapidly fr om 375C to room temperature by using a large heat sink versus slowly cooling the sample at a controlled 10-15C per minute. Figure 3-26. Resistivity of NixCo3-xO4. (1) quenched to room temperature after heat treatment at 375C for 10 minutes in air, (2) at 300C, the first point of the temperature dependent conductivity measur ement on the heating cycle, (3) at 300C, the last point of the temperatur e dependent conductivity measurement on the cooling cycle, and (4) ~16 week s after the cooling heat treatment. Effect of time after heat treatment. Samples exposed to ambient atmosphere for days to months after quenching exhibited resistivities that sl owly increase by a factor of two or more. Figure 3-26 shows a Ni1.5Co1.5O4 sample after a heat treatment cycle. Discussion of heat treatment effects. Heat treatment of nickel-cobalt oxide thin films has an appreciable effect on both optical transmission and electrical conductivity. Electrical conductivity im provements are a factor of c ooling rate with the optical properties degrading with increased conductivity Heat treatment may either repair or induce defects depending on the cooling proce dure. Since the phase transition does not appear to be probable, it is po ssible that localized defects fo rm from atomic disorder or cation disorder, or Frenkel defect oxygen and carriers get trapped in these locally deformed regions becoming polarons. Disorder increases the number of polarons in the lattice and therefore increases the conductivity. Quenched sa mples have lower activation

PAGE 93

74 energies for conductivity in bot h low and high temperature re gions compared to slowly cooled films (summarized in Figure 3-27). Figure 3-27. Rapidly quenched sa mple activation energies of NixCo3-xO4 change after heating when they are slowly cooled. Activation energies calculated from Figur e 3-23 and included in Table 3-4 show that the values calculated fr om heating a quenched sample are all lower than those activation energies calculated from slowly cooling the same samples. Rapid quenching must therefore lock in some degree of disorder or defect structure. Table 3-4. Increase in activation energies of NixCo3-xO4 from the temperature regions in Figure 3-23 graphed in Figure 3-27. Heating to cooling low temp Heating to cooling high temp x=1 0.020 0.035 x=1.5 0.018 0.025 The possibility that a magnetic or physical phase transition could be the cause of the two different transition re gions can be ruled out by the same arguments advanced in

PAGE 94

75 the heat and conductivity subsection above. Tr ansport of surface adsorbed species such as oxygen into the lattice to increase the oxidation state of nickel or cobalt to form more polarons and increase conductivity would be a thermally activated process. STEM data does not show a significant struct ural change with heat treatm ent, which indicates that the bulk of the grains do not experience the in creased oxygen effect, but the grain boundaries may accommodate extra oxygen. These high ly defective regions with excess oxygen would be frozen in the film when it is rapidly cooled. The steady state preferred structure would not be achieved due to kinetic limitations upon cooling. The th in film would relax to this condition perhaps by the excess oxygen to escaping. These extra defects may distort the unit cell via the JahnTeller effect or a Verwey ch arge ordering effect just enough to allow a lower ensemble average of activation energy by increasing the distribution of activation energi es for individual sites. A model that approximates this disorder idea is presented in Chapter 6. Another possible explanation for the increased conductivity at higher temperatures would be densification due to heat treatment Heat would result in increased density from the increased solid state diffusion and the increase in film conductivity could result from elimination of voids. However, it is unlikely that porosity would increase at low sample cooling rates making this explanation less likely. XPS data showed high surface concentrations of carbonyl on the nickel-c obalt oxide films (Windisch et al., 2001a; Windisch et al., 2001b; Windisch et al., 2002b). If carbon was to permeate the lattice and take oxygen to form carbonates upon cooling, th is could increase the disorder during heating. If this mechanism is operati ve, the data suggest that carbonates would decompose and reoxygenate the lattice or even form bubbles of carbon dioxide molecules

PAGE 95

76 thereby increasing porosity. While this idea cannot be entirely discounted, it is considered unlikely that the car bonate will decompose upon cooling. 3.4 Summary and Conclusions The combinatorial approach is effective for depositing a film with a continuously variable composition over a large area in one run. The effects of gas composition during combinatorial sputtering and the effects of to tal chamber gas pressure and different target to substrate distances using si ngle target sputtering in the offset rotation setup made up this study. Optimum conductivity is achieved from comb inatorial films containing equal parts of nickel and cobalt when sputtered in a gas mixture of 50% argon and 50% oxygen. The best film conductivity of 375 Scm-1 may be a result of ideal growth conditions with the gas mixture promoting complete oxi dation and a higher deposition rate. The gas pressure study shows th at a chamber pressure of 2 mTorr yields the highest reported film conductivity to date at 500 Scm-1 compared to the 5 mTorr (400 Scm-1) and 10 mTorr (333 Scm-1) following rapidly quenched after he at treatment. This enhanced conductivity is attributed to a higher growth rate and in creased adatom mobility during growth due to the lower molecular mean free path. Increased distance of target to substrate decreased the film density increased film porosity and decreased conductivity, but incr eased transparency. Target-substrate distance effects on the film density are believ ed to be controlled by phenomena similar to those from lower pressures i.e. higher surface mobility from depositing species due to less gas phase scattering. Less plasma-film in teraction at longer distances may also be a factor. Closer target -substrate distances re sult in denser more electrically conductive films with lower optical transmission, opposite of the films deposited at longer distances.

PAGE 96

77 Heat treatment showed improvement in conductivity when films were rapidly quenched and degradation when slowly c ooled. Transparency once again behaved opposite to the electrical conduc tivity. Temperature dependent conductivity showed that after a rapid quenching heat treatment, the activation energy of conduction is lower than from a film that was slowly cooled. This heat treatment cooling rate effect is a new discovery for this system and is believed to be due to excess oxygen in the lattice. Rapid cooling from heat treatment at 375C to r oom temperature may lock in defects that increase the concentration of polarons Slow cooling allows them to anneal out, resulting in improved transparency across the in frared region and lower conductivity. Band gaps of 3.2 and 3.5 eV and work f unctions of 4.39 and 4.27 eV are reported for NixCo3-xO4 at x = 1 and x = 1.5, respectively. These band gap and work function values have not previously been reported to date.

PAGE 97

78 CHAPTER 4 THE ROLE OF LITHIUM 4.1 Introduction Adding lithium to the nickel oxide, cobalt oxide, and nickel-cobalt oxide systems is a currently acceptable practice in lithium ba ttery electrodes (Appanda irajan et al., 1981; Banov et al., 1995; Benqlilou-Moudden et al., 19 98; Carewska et al., 1997; El-Farh et al., 1999; Fransson et al., 2002; Ganguly et al ., 1997; Gendron et al., 2003; Gover et al., 1999; Han et al., 1999; Julien, 2000; Julien et al., 1999; Kim et al., 2002; Koumoto & Yanagida, 1981; Moshtev et al., 2002; Mo shtev et al., 1996; Moshtev et al., 1999; Robertson et al., 1999; Seguin et al., 1999; Shirakami et al ., 1998; Stoyanova et al., 1997; Yoshimura et al., 1998; Zhecheva et al., 1996). The same is true for molten carbonate fuel cell electrodes (Fukui et al., 2000; Kuk et al., 1999; Kuk et al., 2001). Improvement in some properties resulted as lithium was added in specific quantities, however the material requirements for battery electr ode and fuel cell electrode technology is fundamentally different than those for optoe lectronics. Optoelectronics is primarily concerned with photon-solid in teractions and electrical condu ction. Batteries and fuel cells deal with electrochemical properties such as ionic conduc tion and intercalation (Koumoto & Yanagida, 1981; Wolverton & Zunger, 1999; Zh echeva et al., 1996). Many studies have been done with lithium in ni ckel-cobalt oxide (Puspharajah et al., 1997; Urbano et al., 2001), but none have attempted to probe nickel-cobalt oxide with added lithium for use as an infrared-transparent thin-film electrode.

PAGE 98

79 Initial studies suggested that the additi on of lithium to the nickel-cobalt oxide system would favorably improve conductivity and transparency based the fact that lithium could be incorporated into the spinel unit cell at a tetrahedral site (Appandairajan et al., 1981; Windisch et al., 2001a; Windi sch et al., 2002a; Windisch et al., 2001b; Windisch et al., 2002b). Lithium, a monovale nt ion (for general spinel structure information see (Azaroff, 1960; Kingery et al ., 1975; Smyth, 2000)), could substitute for divalent or trivalent cobalt resulting in a net negative charge in the lattice. An adjacent cation in a tetrahedral site s hould then increase its oxidati on state to maintain charge neutrality in the crystal. It is also possible, though not as likely, that a neighboring octahedral cation could increase its oxidation state by donating an elec tron to balance the charge offset from the lithium substitution on the tetrahedral site (Appandairajan et al., 1981). Such an increase in oxi dation state will change the bon ding nature at that site (create a polaron) resulting in increased conductivity. This prediction assumes that the lithium w ill substitute for a cation on a tetrahedral site, however it is possible that lithium, being a small atom, may prefer to enter the lattice as an interstitial and produce an opposite effect. Inters titial lithium woul d add a positive charge, likely requiring a nei ghboring cation to reduce its oxi dation state. The result would be a net decrease in conductivity from polaron annihilation. Assuming the first case, XPS binding energies might be expected to scale or shift with lithium addition, as would the lattice parameter. Changes in tr ansparency and electrical conductivity may occur as well. Preliminary proof of principle data ar e shown in Figure 4-1 for contrast and comparison to experimental results of this st udy. A favorable improvement in resistivity

PAGE 99

80 for solution deposited nickel cobalt oxide with 10% addition of lithium was recorded. The improvement from lithium in the solution nickel-cobalt oxide films combined with an order of magnitude incr eased conductivity re ported in the literature for sputter deposited versus solution depos ited films (Figure 2-8) provided motivation to determine the effects of lithium in sputtered samples. Experimental results will show that lithium behaved in a manner unexpected and yielded ne w and interesting details of the nickelcobalt oxide system with and without lithium doping. This chapter focuses on the electrical, op tical, structural, and chemical changes produced by the presence of lithium in the nickel-cobalt oxide system. Electrical properties were largely of interest with respect to heat treatment parameters and activation energies of conductivity to explore disorder in the system. Optical properties provided insight to understand the electrical results that were confirmed by both the structural and chemical analysis. 4.2 Experimental Procedure 4.2.1 Deposition A solution deposition method produced some of the films for this study*. Solutions of metal nitrates and a combustion agent ar e mixed in aqueous solution and eye-dropped onto the substrate just before spinning at 3500 rpm for ~30 seconds. Placing the wet film substrate on a hot plate at ~ 375C ignites the glycine com bustion agent included in the solution to decompose the nitrates and produce a uniform film ~50 nm thick. Both traditional and combinatorial films were sputter deposited* in a vacuum chamber evacuated to near 1x10-6 Torr and backfilled to 10 mT orr for all depositions. Reactive RF sputtering was performed in 100% oxygen using alloy target while Additional details of solu tion and sputter deposition are included in Appendix A.

PAGE 100

81 sputtering of oxide targets was performe d with 50% oxygen and 50% argon. Oxide targets of Ni1.5Co1.5O4 and Ni1.35Co1.35Li0.3O4 were used in the combinatorial setup to deposit a film with a graded lithium c oncentration (schematic in Figure 3-1). 4.2.2 Characterization Film thickness was measured with an op tical or stylus profilometer. X-ray reflectivity (XRR) measurements of sele ct films provided confirmation of the profilometry data (See Appendix B for more information on characterization techniques and data). Electrical measurements conducte d in a van der Pauw apparatus provided resistivity values when combined with measured film thickness. Temperature dependent van der Pauw data were collected from room temperature up to 675 K. Fourier transform infrared (FTIR)* (Brundle et al., 1992) was used to measure film transparency with respect to air and with resp ect to the silicon or sapphire substrate over the range of 4000 cm-1 to 400 cm-1after 2 minutes of a nitrogen purge. A dual beam ultraviolet-visible spectrometer was used to measure transparency from the ultra violet region to the near infrared region (200 nm-3300 nm). The deposited film composition was determined using x-ray photoelectron spectroscopy (XPS) (Brundle et al., 1992) and secondary ion mass spectrometry (SIMS) (Brundle et al., 1992) Depth profiled dynamic SIMS data was calibra ted with high resolution XPS depth profile scans. Grazing incident x-ray diffraction (GIXRD) (Brundle et al., 1992) provided information on the crystal structure, crystallite size and orientation, and lattice parameter. For details on characterization techni ques, please refer to Appendix B.

PAGE 101

82 4.2.3 Sample Heat Treatment Films were heat treated to ~375C for 10 minutes in air. Films on fused silica, sapphire, or silicon substrates were cooled by removal to a heat sink for rapid quenching to room temperature followed by characteri zation. Cooling rate s were monitored and controlled at rates <15C/min on the van der Pauw stage. Rapid cooling rates on the aluminum heat sink were calculated to be well over 150C/min, but did not exceeding 15C/s according to a model where ideal conduction conditions were assumed. 4.3 Characterization Results 4.3.1 Electrical Properties Figure 4-1. Thin films with added lith ium deposited from solution precursors or sputtering. Thin films from several deposition methods including solution deposition, reactive RF sputtering from alloy targets, and RF sputtering from metal-oxide targets, were compared with respect to the concentration of lithium added to the starting material target or solution (Figure 4-1). Cobalt oxide films deposited from solution with 1, 5, 10, and 15% lithium showed a favorable decrease in resistivity with lithium addition, (blue squares Figure 4-1). The optimum amount of nickel for conductivity was added to the

PAGE 102

83 cobalt oxide solution, along with 10% lithium, and these films also showed a favorable decrease in resistivity. The data from th e cobalt oxide system (Figure 4-1), show little change in resistivity for lithium content be tween 10 and 20% suggesting the effect has saturated. Figure 4-2. Conductivity changes based on the cooling rate following heat treatment for all compositions of lithium in nickel-cobalt oxide films. As mentioned earlier, the design of this experiment was to in crease the conductivity of the best nickel-cobalt film by adding lithiu m. Sputtering had shown lower resistivity values for nickel-cobalt oxide thin films. Samples of NiCo2O4 were reactively RF sputtered in 100% oxygen using the composition of the solution samples as a guide. A small decrease in resistivity with the NiCo2O4 spinel composition was seen (Figure 4-1 yellow squares). This effect was very sm all, and the conductivity from sputtering deposited films could be increased much more by adding more nickel. The Ni1.5Co1.5O4 films sputtered from oxide target s fabricated in the laboratory* and found (Chapter 3) to exhibit the lowest resistivity, increased in resistivity upon addition of lithium (green circles). The effects of lithium on electrical resistivity of sputter deposited films Recipes and procedure of oxide target production included in Appendix A.

PAGE 103

84 prompted a reinvestigation of the solution deposition method. High lithium concentrations in the presence of high ni ckel concentrations appear to degrade conductivity, so very small amounts of lithium we re added to the solutions used for film deposition. Figure 4-2 shows the result of starting with the ideal composition of NiCo2O4 and adding lithium and decreasing coba lt by equal concentrations, i.e. NiCo2–zLizO4. Figure 4-3. Conductivity as a function of te mperature for 50 nm thick solution deposited thin films. Films of (a) NiCo2-zLizO4 and (b) Ni0.75Co2.25-zLizO4 show that small amounts of lithium increase conductiv ity. All films exhibited a decrease in room temperature conductivity after th e heating cycle. Each film was heat treated and rapidly cooled prior to the temperature de pendent conductivity measurement. Small amounts of lithium increase conductivity in the substituted solution deposited samples. In fact a z value of 0.01 in NiCo2-zLizO4, (NiCo1.99Li0.01O4,) yields the highest conductivity for that series. The same concentration of lithium does not have the same effect when the nickel content in the film is reduced (cobalt to nickel ratio of 3to-1 or Ni0.75Co2.25-zLizO4). Substituting lithium in this system shows that a z value of 0.1 (Ni0.75Ni2.15Li0.1O4) was required for optimization. When small doses of lithium substitute for cobalt, increased conductivity is observed. Th e changes are ~2x 3x rather

PAGE 104

85 than near10x as expected. Conductivity afte r rapid and slow cooling is shown in Figure 4-2 and 4-3. The faster cooling rate always yields a better conductivity. Figure 4-3 shows temperature dependent c onductivity values for solution deposited films. Four samples from Figure 4-2 (the two samples without any lithium and the two highest conductivity samples) are shown. For clarity, lower conductivity compositions are omitted from the plot, but are tabulated in Appendix C. Conductivity as a function of temperature for high conductivity films increases with temperature, but upon cooling the samples exhibit lower room temperature conduc tivities. The conductiv ity after sequential heat treatments with different cooling rates (10-15C/min versus quenching on an aluminum, brass or water chilled brass plate (~150C/min)) for one sample are shown in Figure 4-4. Figure 4-4. Rate of sample cooling afte r heat treatment had a dramatic effect on conductivity. Aluminum and Brass heat sinks have effective cooling rates greater than at leas t 150C per minute, possibly as high as 15C per second. The faster the cooling rate, the higher the conductivity. A maximum cooling rate of 15/min was allowed by the van der Pauw measurement instrument. The quenching medium for most samples in this study was an aluminum block. While the exact cooling rate for the aluminum and brass heat sinks are unknown, a calculation was done using the therma l mass of the fused si lica substrate with

PAGE 105

86 a heat flux through the surface in contact with the cooling block. The cooling block heat capacity is assumed to be infinite and the c onduction contact is assumed to be perfect to yield a maximum theoretical c ooling rate of ~15C/s. Th e heat sinks used were of sufficient thermal mass that no signifi cant temperature rise was expected. The break between a larger slope and a sm aller slope for the conductivity data in Figure 4-3 near 450K (1000/T = 2.22) has prev iously been reported (Windisch et al., 2002b). Note that above this temperatur e, the conductivity is independent of temperature, while below it conductivity depe nds upon quenching rate. The origin of this break will be discussed further below. Figure 4-5. Activation energy dependence for Ni0.75Co2.25-zLizO4 and NiCo2-zLizO4 from solution on lithium content. Lithiu m substitutes in increasing amounts for cobalt and as conductivity decreases, the activation energy increases. All samples were heated and cooled at 15C/min. Hopping of polarons in the nickel-cobalt ox ide spinel is the primary mechanism of electrical conductivity. The slopes of the Arrhenius plot s in Figure 4-3 (additional plots found in Appendix C) are proporti onal to the activation energy required to induce charge carrier motion. Of interest are the activati on energies for the low temperature slopes of the quenched state and the slowly cooled state. These differences in activation energy for the two series (NiCo2-zLizO4 and Ni0.75Co2.25-zLizO4) are shown in Figure 4-5. The most

PAGE 106

87 conductive film has the lowest activation en ergy and the most resistive film has the highest activation energy, consistent with expect ation. Note that more lithium in the film results in a larger difference between the ac tivation energies of th e quenched film state and the slowly cooled state. Also, the larger variance is observed in samples with lower overall conductivity. Conductivity degrades when too much lithiu m is added to the target or precursor solution. At low concentrations, and dependi ng on the nickel content in the film, lithium had either a positive or negative effect on conduc tivity. The next section will discuss the role of lithium in the nickel-cobalt oxide system and its effect on optical properties. Figure 4-6. FTIR mid IR transmission spectra of alloy-target reactive sputter-deposited Ni0.95Co1.95Li0.15O4. Transmission spectra measured before and after heat treatment with rapid cooling, referenced to air and to a bare silicon substrate. Absorption regions at 633 cm-1 and 546 cm-1 are characteristic of spinel lattice vibrations. 4.3.2 Optical Properties Optical properties are also affected by th e different lithium film concentrations. Thin film transparency from FTIR between 400 cm-1 and 4000 cm-1 of a reactive sputter deposited 50 nm thick Ni0.95Co1.95Li0.15O4 film is shown in Figure 4-6 (Post means after

PAGE 107

88 quench). The transmission spectrum of a film (as deposited) is compared to the transmission spectrum after heat treatm ent followed by rapid quenching (with an aluminum block). The IR transparency decreases after the quenching treatment, consistent with an increased conductivity. Similar results are observed for all samples after quenching. When the transparency of th e films is corrected to remove the silicon absorption spectra, the characteristic spinel absorption peaks at 633 cm-1 and 546 cm-1 are observed. Figure 4-7. FTIR mid IR transmission spectra of oxide-target sputter-deposited Ni1.2Co1.2Li0.6O4. Transmission spectra measured before and after heat treatment with rapid cooling. The sp ectra marked post is after quenching on an Al block. Those marked air are sp ectra taken of the film and substrate uncorrected for the silicon substrate refere nced to air. Those marked Si have been corrected for absorption of the sili con substrate. A region of interest near 1400 cm-1 was found (blue arrow) in samples prior to heat treatment. Thin film transparency from FTIR of a sputtered 250 nm thick Ni1.3Co1.3Li0.6O4 film is shown in Figure 4-7. The transmi ssion spectrum of the film as deposited is compared to its transmission spectrum after h eat treatment and rapid quenching just as in Figure 4-6. A decrease in optic al transparency is seen after the quenching treatment. When the film spectra are corrected for si licon absorption spectra, an absorption band near 1400 cm-1 (blue arrow) is obvious and the charac teristic spinel absorptions were not

PAGE 108

89 observed. Traces labeled Post are the transmission data ta ken after quenching the sample from heat treatment of ten minutes at 375C. The absorption feature at 1400 cm-1 is significantly reduced immediately afte r quenching. The surface of the Ni1.3Co1.3Li0.6O4 sample exhibits a milky white haze af ter prolonged exposure to atmosphere. The effect of lithium on films sputter deposited using the combinatorial technique (discussed earlier) from a Ni1.5Co1.5O4 target and a Ni1.35Co1.35Li0.3O4 target revealed a chemical reaction at the film surface. Fi gure 4-8 shows the FTIR spectra from various substrate positions. Positions 7 and 9 (close r to the lithium targ et) showed the same substantial absorption near 1400 cm-1 compared to film positions 1 and 3 (closer to the target with no lithium). More lithium app ears to lead to this absorption peak. The spectra were not all uniform films with the same thickness, therefore they were normalized to a common value of 45% at 1200 cm-1 to the intensity of the feature near 1400 cm-1. Based on the interpretation below, the feature near 1400 cm-1 is not a function of thickness of the films. The absorption at 1400 cm-1 is attributed to carbonate formation, as seen from the grey carbonate reference spectrum shown in Figure 4-8 from lithium carbonate in potassium bromide that also show s a strong absorption near 1400 cm-1. The broad absorption band near 1600 cm-1 is from water and accounts for the remaining deviation of the reference spectra. Formation of the surf ace carbonate is discussed in Section 4.3.4. The combinatorial film provided interest ing information with respect to the absorption region near 1400 cm-1 and its correlation with lithium, but it was not structurally identical with other lithium-cont aining nickel-cobalt oxide films. The film conductivity across all positions was an order of magnitude less than optimized sputter

PAGE 109

90 deposited nickel-cobalt oxide thin films, pr obably due to the large target-to-substrate distances required to get adequate co verage for the combinatorial film. Figure 4-8. Normalized FTIR spectra at 1200 cm-1 show the absorption region near 1400 cm-1 assigned to carbonate on the surface when compared with a carbonate reference spectrum shown in grey. F ilm position in relation to the lithium target appears to determine the strength of the absorption. 4.3.3 Structural Properties and Composition For corrected FTIR data, near 610 cm-1 and 540 cm-1 indicate that the films are not amorphous. The spinel structure is confir med by grazing incide nt x-ray diffraction (GIXRD referred to hereafter as XRD). Th e combinatorial film shown in Figure 4-9 appears to be weakly crystalline with only low intensity spinel peaks. The film is most probably highly disordered with a low density, similar to those reported in Chapter 3 with a large target-substrate distance. A peak near 43 appears to shift to a lower angle from areas of the film positioned closer to the lithium-containing target

PAGE 110

91 Figure 4-9. Grazing incidence XRD from combinatorial sputter deposited films deposited from oxide targets. Note the weak crystalline diffraction peaks from the film. Other films deposited by s puttering and solution were analyzed with XRD. The lattice parameters calculated from fitted XRD patterns are is shown in Figure 4-10. The method of deposition determines the lattice pa rameter. The oxide target sputtered films all had a much larger lattice parameter that increases with the addition of lithium (blue). Films deposited from reactive sputtering with alloy targets all had a smaller lattice parameter. It is reported that nickel a ddition to cobalt oxide increases the lattice parameter (Windisch et al., 2001a; Windisc h et al., 2001b; Windisc h et al., 2002b). The same trend is evident from films deposited from an alloy target (magenta and green). The addition of lithium to nickel cobalt oxide (green data) also increased the lattice parameter. The theoretical lattice parameter calculated fr om the powder diffraction data is included as a reference point. All measured lattice pa rameters were larger than the theoretical value. An explanation of this observation is discussed below.

PAGE 111

92 Figure 4-10. Films sputter deposited from an alloy target have a smaller lattice expansion than films deposited from oxide targets. The theoretical lattice parameter from powder diffraction files is also shown. The increase of the lattice parameters appears to be pr oportional to the amount of lithium. Data from Windisch et al. show that as nickel is added to into cobalt oxide, the lattice parameter expands (Windisch et al ., 2001b; Windisch et al., 2002b). As more impurities are added in the spinel system, the unit cell expands to accommodate the different cations. Table 4-3 shows the data plotted in Figure 4-10 w ith the error margin determined from XRD data. Table 4-1. Numerical data fr om Figure 4-10 are listed in columns. The calculated lattice parameter from the powder diffraction file is less than the experimentally measured values. The experimental va lues increase with lithium addition and depend on the method of production of the film. Source Formula Lattice Parameter () Error Theoretical NiCo2O4 8.114 Alloy Target NiCo2O4 8.123 0.001 Alloy Target Ni0.95Co1.85Li0.15O4 8.127 0.001 Alloy Target Ni1.5Co1.5O4 8.137 0.001 Oxide Target Ni1.5Co1.5O4 8.237 0.021 Oxide Target Ni1.35Co1.35Li0.3O4 8.242 0.007 Oxide Target Ni1.2Co1.2Li0.6O4 8.304 0.004

PAGE 112

93 A lattice parameter increase from 8.23 to 8.30 in films sputter deposited from oxide targets indicates that lithium may not have substituted for cobalt, but occupy interstitial sites. XRD scans of the same film s showed the spinel structure as is evident in Figure 4-11. Changes in the texture of the film with increased lithium are reasonable. Figure 4-11. XRD of sputtered films with incremental amounts of lithium (not normalized for intensity). The peak near 65 (below the black arrow) shifts to a lower angle and the peak near 18 d ecreases in intensity with increased lithium concentration. XRD data collected after heat treatment of solution deposited samples to observe any structural changes that may occur from changes in cooling rates as a function of lithium concentration. Thee XRD scans from solution deposited samples are shown in Figure 4-12, collected after heat treatmen t followed by rapid quenching and again after heat treatment followed by slow cooling. Th e spectra are identical to the unaided eye, but some peak narrowing after rapid quenchi ng can be demonstrated. This narrowing of peaks reflects a larger calculated crystallite size in these samples. A change in the lattice parameter between the two cooling rates could be detected by fitting the XRD pattern, with a smaller lattic e parameter at a faster cooling rate as shown in Table 4-2. While crystallite size va ried, it did not appear to scale with cooling rate.

PAGE 113

94 Figure 4-12. XRD spectra show no obvious ch anges for fast versus slow cooling for Ni0.75Co2.25-zLizO4. Fitting the curves reveals changes shown in Table 4-4. Table 4-2. Lattice parameter of Ni0.75Co2.25-zLizO4 as a function of lithium fraction (z) before and after rapid cooling from a 10 minute 375C heat treatment. Ni0.75Co2.25-zLizO4 Z=0.00Z=0.01 Z=0.03Z=0.05 Z=0.1 Z=0.3 Z=0.5 Lattice Parameter 8.1313 8.144978.1414 8.1548 8.1419 8.1332 8.1279 After Quench 8.1327 8.1360 8.1327 Crystallite Size 80 50 55 50 90 63 After Quench 55 55 50 4.3.4 Chemical Properties Film composition was measured by XPS and SIMS. XPS analysis. of the combinatorial film from the oxide targets (discussed in Section 4.3.2) with the absorption band at 1400 cm-1, as well as the variable lattice parameter films from the previous structural section were of particul ar interest. SIMS was collected from films

PAGE 114

95 sputter deposited from the oxide target with varying lithium concentrations including the highest concentration. From both XPS and SIMS analyses, depth profiling data were collected, but the sensitivity to Li was low and high, respectively. Figure 4-13. XPS of the Carbon 1s region from the combinatorial sputtered film (a) at positions identified by the color code, and (b) curve fit to show composition of position 9. Carbonate at a binding en ergy of 290 eV appears with lithium only at position 9. All curves show the presence of carbonyl near 288 eV. The combinatorial film sputtered from oxide targets of Ni1.5Co1.5O4 and Ni1.35Co1.35Li0.3O4 was analyzed with XPS to observe surface compositional changes with position and lithium concentration. Changes in the carbon 1s peak are shown in Figure 4–13. Carbon bound as carbonyl, whose 1s peak shifted from 284.6 to 288.2 eV, was detected all across the film. The ar ea of the film closest to the Ni1.35Co1.35Li0.3O4 target (green trace, position 9) shows a large peak at 290 eV, a binding energy associated with carbon bonded with three oxygen atoms in a carbonate. This carbonate binding energy region is not prominent at ot her positions of the film.

PAGE 115

96 Figure 4-14. XPS of combinat orial deposited nickel-cobalt oxide film with variable lithium concentration. Lithium is onl y detected at position 9, the position closest to the Ni1.35Co1.35Li0.3O4 target. See Figure 4-13 for trace numbering position key. Figure 4-14 shows XPS data from the co mbinatorial film over an energy range which includes the Co2p and the Li 1s photoele ctron peaks. Lithium is only detected at position 9, which indicates a high lithium concentration based on the low sensitivity of XPS for lithium. Position 7 may show trace amounts of lithium, while any lithium signal form the remainder of the films is below th e noise level. Carbona te and lithium are both in the film at position 9. SIMS depth profile data were collected from the film showing the FTIR carbonate absorption peak near 1400 cm-1. The SIMS data shows predominantly lithium at the surface, which decreases after sputtering for a few hundred seconds to a low, constant level at which time the nickel and cobalt signal increased and remained constant throughout the film. As shown in Figure 4-15, th e film is composed of nearly equal parts of nickel and cobalt and a nearly-constant concen tration of lithium in the bulk of the film. Based on quantitative XPS calibration, the solu bility limit of lithium in nickel-cobalt oxide is ~15%. Another interesting featur e shown by the SIMS is the interface layer

PAGE 116

97 where the film meets the substrate. The con centration of lithium drops before the other cations with nickel concentration s lightly increasing at the interface. Figure 4-15. SIMS depth profile of a thin film sputtered from a Ni1.5Co1.5Li0.6O4 target. XPS analysis is correlated with heat treatment of solution deposited lithiumcontaining films is shown in Figure 4-16. The Li 1s peak was undetectable and is therefore omitted. Rapid quenching of Ni0.75Co2.24Li0.01O4, Ni0.75Co2.22Li0.03O4, and Ni0.75Co2.2Li0.05O4 produced no significant changes in the Ni 2p and Co 2p binding energy regions. Minor changes may appear to be visible but may be in intensity only. A case might be made to differentiate the peak shapes at 858 eV in the Ni 2p region, but a calibrated quantitative measure will be required to make such an assertion. Significant changes are seen comparing the slow cooling ve rsus rapid cooling O 1s and C 1s peaks at 531.2 eV and 285 eV, respectively. After quenching, the O 1s shoulder at 531.2 eV appears to decrease. Also the carbon content, likely due to adventitious carbon on the surface, decreases after the rapid cooling heat treatment. The lithium film concentrations do not appear to influence the order or arra ngements in any of the plots suggesting that the effects seen here with heat treatment are not a function of the lithium concentration. For example, the O 1s peak at 531.2 eV show s the purple curve at th e top with blue and then red beneath. The C 1s region shows the order as red, purple, and then blue. If

PAGE 117

98 lithium concentration is a factor ordering of re d, purple, and then bl ue or blue, purple, and then red is expected in bot h cases, but they are different. Figure 4-16. XPS of (a) Ni 2p, (b) Co 2p, (c) O 1s, and (d) C 1s binding energy regions from solution deposited films of nickel-cobalt oxide containing lithium following slow and fast cooling after heat treatment.

PAGE 118

99 4.4 Discussion 4.4.1 Lithium Effects The resistivity of nickel-cobalt oxide did not decrease with additions of lithium as was expected based in the effects of addi ng more nickel to cobalt oxide. Lithium substitution did not decrease the resistivity of cobalt oxide to the same extent as just adding nickel. This may be a result of lithium being in a different oxidation state, or energetically preferring a different site than ni ckel. Only specific lithium concentrations in specific nickel-cobalt oxide samples were beneficial i.e. led to increased conductivity. For example, electrical conductiv ity of solution deposited NiCo1.99Li0.01O4, and Ni0.75Co2.15Li0.1O4, both increased, however the sample with more nickel requires less lithium to increase the conductivity. The samples with less nickel require more lithium to produce a positive effect, yet the absolute conduc tivity is not as high as the sample with a higher nickel concentration. Compositions with higher and lower concentrations of lithium for these amounts of nickel and coba lt do not enhance electrical conductivity. This leads to the conclusion that there may be an optimum number of carriers within the lattice that can be activated. Once this num ber is exceeded, reduced conductivity will be observed. Lithium may perturb the localized stated required for polarons formation and reduce the carrier concentration and the c onductivity. Note that lithium-enhanced conductivity exhibited improveme nts less than 2x, not the order-of-magnitude changes between the conductivities of sputter deposit ed (high conductivities) versus solution deposited nickel-cobalt oxide films. In all cases, the resistivity values for nickel-cobalt oxide films with or without lithiu m are much higher than those from n-type TCOs. The improved conductivity by lithium additi on to solution deposited films is not realized in the sputter deposited films. One nickel-cobalt oxid e film deposited by

PAGE 119

100 reactive RF sputtering from an alloy target (making Ni0.95Co1.9Li0.15O4) shows a small decrease in resistivity. The reason for this single improvement is unknown and is considered to be insignificant due to the small change. Films sputter deposited from oxide targ ets with added lithium showed lower conductivities upon lithium addition. Accord ing to XRD data, the unit cell expanded even though a similar or somewhat smalle r lattice parameter wa s predicted upon lithium substitutional addition. Latti ce expansion suggests that l ithium incorporated as an interstitial rather than occupying a tetrahedral site occupied by cobalt. Inte rstitial lithium would add a positive charge balanced in the matrix by the reduction of a neighboring cation. The net result would be a decreas e in conductivity from th e reduction in the number of polarons charge carriers, just as was observed. Reduced conductivity may also result fr om insulating layers of carbon and oxygen species found on the surface. XPS and SI MS analysis indicate that lithium is concentrated on the surface and XPS data show it is bound as a carbonate. A logical sequence leading to surface carbonate would be for lithium to segregate to the surface where it forms a hydroxide. Adsorbed car bon dioxide reacts with the lithium hydroxide to form surface layers of lithium carbonate. The SIMS data suggest that lithium has a solubility limit of ~15% in the sputter depos ited nickel-cobalt oxide. Observation of a concentrated lithium surface layer and lower con centration in the film may be due to two possible scenarios: (1) lithium diffuses to th e surface at room temperature, or (2) the SIMS sputtering process greatly enhances the lithium sputte r yield due to lithium’s low mass and high ionization probability. Both are likely to be true, but the XPS data support the presence of a true lithium surface laye r bound as carbonate. In addition to artifacts

PAGE 120

101 from SIMS analysis, preferential sputtering of lithium during deposition may create an initial lithium-rich layer on the substrate which remains on the surface as a “floating layer” while the film grows underneath. This lithium “floating or surfactant layer” may then react with wet air forming lithium hydroxi de when the chamber is vented. Lithium hydroxide on the surface then reacts with adsorbed carbon dioxide on the surface to form lithium carbonate. No matter the formation mechanism, this hazy carbonate layer could block electrical contact without significantly reducing optical transparency. Lithium is therefore detrimental to film conductivity from sputtering due to its high surface concentration and subsequent carbonate formation on the surface. It is also possible that conductivity re duces when too much lithium is added because metastable nickel forms nickel oxide similar to the effect observed when nickelcobalt is overheated and nickel oxide precipitates. Heat tr eatment procedures exceeding 400C cause decreased conductivity due to this phase segrega tion. Lithium may change the decomposition temperature or act as a nuc leation agent or inhibitor at sufficiently high concentrations. If th e heat treatment and quenching lead to nickel oxide precipitation, the change in conductivity will not be recoverable, contra ry to the data. It is therefore concluded that nickel oxide precipitation did not cause the changes in conductivity. 4.4.2 Effects of Heat Treatment Activation energies show that quenching ha s a greater effect when the conductivity is lower, as seen in Figure 4-3. Quen ching may freeze in cation distribution disorder among the available spinel octahedral and tetr ahedral sites (discussed more in chapter 6) with increased activation seen as resistivity increases.

PAGE 121

102 The rate of cooling after heat treatment has an eff ect on conductivity that is dependent upon the rate of cooling. Fa ster cooling leads to higher conductivity, presumably by creating more polarons from the freezing of a disordered structure. This conductivity improvement relaxes with time at room temperature because it is a metastable state as seen in Figure 4-4. Quenching of the films decreased optical transparency. More polarons, believed to result from the nickel and oxygen electron exch ange in the octahedral site, presumably introduce more d-d transitions and increase the optical absorption as seen in Figure 4-6. Heat treatment also reduced the carbonate f eature in Figure 4-7. This is most easily explained by carbonate evaporating during the heat treatment. Heat treatment affects nickel-cobalt-lithium oxide films by increasing the conductivity and reducing the ca rbon and an oxygen species when the film is cooled quickly following heat treatment. XPS data shows that heat treatment has no significant effect on the nickel or cobalt 2p regions. The effect of heat treatment is shown in the C 1s and the O 1s regions with a decrease in the adventitious carbon, carbonyl and a decrease in the higher binding energy oxygen region, consistent w ith volatization of carbonate. A decrease in the higher binding en ergy O 1s peak, shown in Figure 4-16c, does not scale proportionally wi th conductivity that increased after the quenching heat treatment. Improved conductiv ity with the decreased O 1s peak may contradict the reported data shown in Figure 2-6 that sugge sts the peak intensity scales proportionally with conductivity. The oxygen peak shown in the data for nickel-cobalt-lithium oxide may originate from both lattice defect oxyge n as well as the surface carbonate layer which is not a factor in films with no lithium. For films with lithium, is not an absolute

PAGE 122

103 indicator of defect oxygen. It may be a combination of seve ral effects. The interference of the higher binding energy O 1s peak may be explained as (1) th e increased oxidation of the surface when heat treated in the presence of air, or (2) surface adsorbed species in pores of the film may be thermally desorbed during the heat treatment According to the FTIR spectra, the original spectra (Figure 47 and Figure 4-8) show that the absorption regions at 1400 cm-1 will disappear after the heat treatment. A combination of both effects may be possible and resu lt in a shift in the electron density of the nickel atoms near the surface to provide a conducting path on the surface and into the bulk film below, which would also explain the decrease in tr ansparency. A third less probable explanation may be possible, but would be more a function the experimental setup. If the carbonate layers simply desorbed during heat treatment, then the surface would be in better contact with the probe. Contact probe pressure could greatly influence the conductivity measurement in this instan ce and would introduce additiona l uncertainty. Data as a function of probe pressure may answer this question, but was not pursued. Table 4-4 shows that the lattice parameter of the quenched nickel-cobalt oxide with small amounts of lithium is reduced. It seems counter-intuitive to suggest that the lattice shrinks as a result of more cation disorder a nd lattice distortion, but it would be consistent with lithium changing from an interstitial to substitutional ion in a tetrahedral or an octahedral site, and upon quenc hing, the disordered arrangeme nt of these cations in the lattice could produce more polarons. By de finition, the polaron induces a localized lattice strain and therefore may exhibit latt ice shrinkage when the majority carriers are holes. In addition, cations in a higher oxidation state will be smaller and therefore the lattice parameter will decrease. This shift to a smaller lattice parameter is therefore

PAGE 123

104 plausible and indicative of an increased polar on concentration from cation disorder in the crystalline spinel solid. A maximum amount of disorder in the lat tice may be induced by nickel when it is present. The addition of lithium may have merely ordered the lattice in a different way and therefore had the eff ect of increasing the activation energy of conduction, resulting in a lower conductivity. 4.5 Conclusions Lithium improves electrical properties only in three sets of conditions: two compositions deposited from solution and one deposited by sputtering. The improvements in electrical conductivity from lithium in these three compositions is by less than a factor of two a nd were less than the improve ment made in conductivity in films containing no lithium and deposited by spu ttering rather than from solution. Other compositions of nickel-cobalt oxide with lithium, whether sputtered or solution deposited, produces films with increased resist ivity and expanded lattice parameters. The behavior of films containing lithium is si milar to those not containing lithium with respect to optical and electrical properties following heat treatment. Heat treatment followed by quenching increases conductivity and decreases transparency. In contrast to high concentrations of added lithium, which causes a lattice expansi on, heat treating and quenching films with small lithium doses causes a lattice shrinkage and an increase in conductivity. Increased lithium causes a rapid increase in the activation energy for film conductivity for both quenched and slowly cool ed films, but the magnitude depends upon whether the temperature is cooled fast (quenched-high conductivit y) or slow (slow cooling-low conductivity). Heat treatmen t effects upon conductivity and transparency are reversible for films with a nd without lithium. It is conc luded that the film is in a

PAGE 124

105 metastable state following a rapid quench fr om after heat treatment, and that cation disorder is responsible for this change. Lithium in films deposited from solution te nds to increase con ductivity when added in the correct amount, which is a function of the nickel-to-cobalt rati o. Higher nickel-tocobalt ratios require less lithium for increas ed conductivity. The improved conductivity value is still small in comparison to the high co nductivity sputtered films without lithium. Up to 15% of lithium incorporates in the la ttice. Lattice expansion suggests this high lithium concentration results in occupation of interstitial sites rather than substitutional site replacement for cobalt or nickel. Po laron annihilation inst ead of polaron creation results from the inters titial lithium. Excess lithium be yond the solubility limit segregates to the surface, forms carbonate and results in reduced conductivity. The formation of the carbonate surface layer was de tected FTIR and XPS.

PAGE 125

106 CHAPTER 5 RHODIUM SUBSTITUTION FOR COBALT 5.1 Introduction Nickel-cobalt oxide conducts via a pola ron hopping mechanism. Polaron hopping is a relatively slow process b ecause a polaron is made up of a charge carrier and a lattice strain, both of which move during electrical conduction. Conductivity is the product of polaron density, mobility, and an electron’s charge. In a polaron hopping system, conductivity is regulated by the polaron density because of the hopping mobility is low. Improvement to conductivity in a hopping c onduction system is achieved by increasing the number of carriers. Substitution of rhodi um for cobalt in the nickel-cobalt oxide system is expected to increase conductivity and transparency. C onductivity will increase with increased polaron density from structural distortion introduced as the larger rhodium ion occupies the octahedral site (Windisc h et al., 2002b). Substitution of rhodium for cobalt should improve optical transparency be cause moving deeper in the periodic table to a larger mass will shift the absorption fr equencies to lower values and increase the forbidden band gap of transition metal oxides. Rhodium may also play a role in cation char ge disorder as discussed in Chapters 3 and 4 where Verwey-type transitions affect the cation charge disorder following quenching heat treatment that does not signifi cantly reduce crystallinity. Disorder of cation charges may change the charge distri bution from the normal to inverse spinel cation arrangement.

PAGE 126

107 Nickel-cobalt-rhodium oxide spinel has been reported (Blasse, 1963). Rhodium may cause a structural shift in stability from the spinel structure to forming separate phases and reducing the crystallin ity of the films as structural disorder in the lattice increases (Da Silva Pereira et al., 1994; Tavares et al., 1996). Structural disorder would increase the polaron density, but may present complications if the conduction pathways are disrupted to the extent that carrier hopping is obstructed by the di sorganization. An increase in the carrier density amounts to li ttle or no increase in conductivity if their formation creates potential barriers or di scontinuities in the hopping paths. Emin, Stoneham, and Tauc have all commented on th e role of structural disorder and the increased polaron formation in such an a rrangement (Emin, 1994; Emin & Bussac, 1994,; Stoneham, 1980; Tauc, 1976). In these cases, structural disorder increases the polaron density and the conductivity increases. A calculation done by Windisch, Ferris, et al. shows that rhodium substitution in the nickel-cobalt oxide system may shift the density of states of the d orbitals and result in a decreased optical absorption due to an in crease in the disorder in the system, shown in Figure 5-1 (Windisch et al., 2002a; Windi sch et al., 2002b). The calculation assumes that rhodium will be substituted completely for cobalt. The density of electron quantum states in the quantum levels at energies between 0 and 2 eV shifts down with rhodium and appear to combine and shift left at en ergies of 0 to –4 eV a to a more negative energy. The energy spectrum between 0 and –2 eV showed a shift to the left with substituted rhodium. Shifts to a decreased density of stat es in the 0 to 2 eV region indicate that the addition of rhodium may reduce optical ab sorption in that region from decreased d-d transitions and result in improved optical and infrared transparency.

PAGE 127

108 Other studies on bulk powder samples show that rhodium spinel forms as rhodium is added to the nickel-cobalt oxide system and the degree of crystallinity of the spinel structure decreases with rhodium (Da Silv a Pereira et al., 1994). When the powder samples are heated to 500C, NiO precipita tes, however processing at 350C produces a spinel phase for NiCo2-xRhxO4 at a value of x up to 0.5 (D a Silva Pereira et al., 1994; Tavares et al., 1996). Figure 5-1. First principles d-band density of states calculation of rhodium substituted for cobalt. Energy of 0 eV is the Fermi energy. Blasse produced a set of powder samp les of nickel-cobalt-rhodium oxide by heating at 1100C for several days With complete substitution of Rh for Co in the spinel crystal cell, the normally cubic spinel system is reported as a tetragonal variant, being elongated with c/a > 1. The cubic to tetragonal structural shift is attributed to the Jahn-

PAGE 128

109 Teller* effect. Spin states of the tetrahedrally located ions are not quenched if the lattice is cubic. In other words, the magnetic moments of the divalent metal ions are considerably larger in the cubic phase than in the tetragonal phase because they are threefold degenerate in the tetrahedral site. With distortion from the Jahn-Teller effect, the degeneracy is removed and the angular or bital momentum is for the greater part quenched. The structure returned to a c ubic spinel above 400 K, reversing the JahnTeller effect (Blasse, 1963). Nickel rhodium conductivity resu lts were not given but due to the change in structure from cubic to te tragonal, a more disordered structure allowing more polarons should be be neficial to conductivity. This chapter discusses results from elect rical, optical, structural, and chemical characterization of solution and sputter de posited nickel-cobalt-rhodium thin films. 5.2 Experimental Procedure 5.2.1 Film Deposition Solution deposition. Solution precursors containing specific molar quantities of nickel nitrate, cobalt nitrate, aqueous rhodium nitrate, and ma lonic acid were dissolved in ethanol to total 20 mL followed by filteri ng with a 5 m filter. Films produced by solution deposition were composed of the amounts shown in Table 5-1. Precursor solutions were dropped on silicon substrates and spun at ~3500 rpm and for 5-10 seconds until the substr ate came to full rotation speed and a constant green color showed. Spinning was stopped and the substrat e removed from the spin coater and baked at ~375C for 10 minutes. Film thickness was on the order of 25-50 nm per layer The Jahn-Teller effect is where degeneracy of orbitals is lost as they shift and distort to obtain a lower symmetry and lower more en ergetically favorable state.

PAGE 129

110 deposited. Multiple spin cast film layers were successively deposited on individual substrates to obtain films of greater thickness for greater ease in characterization. Table 5-1. Composition recipe of NiCo2-vRhvO4 used for solution deposited samples. h value Ni(NO3)2 Co(NO3)2 RhNO3 (aq) Malonic Acid Ethanol (mL) moles 0 0.0022 0.0044 0.009798 20 0.25 0.0022 0.00385 0.00055 0.009798 20 0.5 0.0022 0.0033 0.0011 0.009798 20 1 0.0022 0.0022 0.0022 0.009798 20 grams 0 0.63879 1.280576 1.02 20 0.25 0.63879 1.120504 0.4051397 1.02 19 0.5 0.63879 0.960432 0.8102795 1.02 18 1 0.63879 0.640288 1.620559 1.02 18 Sputtering. Sputtering conditions similar to those found in Chapter 3 and 4 were used to deposit various film compositions with the exception of a DC power supply was used instead of an RF power supply. Chamber base pressure was ~1x10-6 Torr backfilled to a pressure of 10 mTorr with 100% oxygen (See Appendix A for sputtering systems and other deposition details). Twoinch alloy targets of Rh, NiRh2, and NiRh4 were reactively sputtered with the DC power supply operating in controlled amperage mode set at 0.25 A resulting in 408 V and a power of about 90 W for 5-10 minutes. Target to substrate distances varied from 10 cm to 30 cm with substrate fixtur es in double planetary and single rotation setups. Two-inch all oy targets were powered by a 1 kW MDX DC power supply. Deposition distances of 12.5 cm or less were used in the offset rotation setup. Samples were placed on the substrate holder platen along the radius of rotation directly over the cathode ax is. Rotation speeds were ~60 rpm. Films deposited at distances between 15 and 30 cm were done by either the offset rotation sample setup or the double planetary rotation. Double planetar y rotation provides uniform films and is most often used for commercial producti on where film homogeneity is of utmost

PAGE 130

111 importance. Single rotation requires less targ et material and less time to produce a film. Substrates were cleaned in an ionized air stream prior to introduction in the vacuum chamber to remove any particles that ma y cause pinholes in the produced films. 5.2.2 Post Deposition Heat Treatment Annealing heat treatment occurred at temp eratures at 375C in air for 10 minutes. Films of fused silica, sapphire, or silicon s ubstrates were cooled on a heat sink followed by immediate electrical and optical measurements. 5.2.3 Characterization Film thickness was measured with an op tical or stylus profilometer. X-ray reflectivity (XRR) measurements of sele ct films provided confirmation of the profilometry measurements. Van der Pauw m easurements and Hall effect measurements were conducted at room temperature and in crementally up to 675 K with measurements taken both before and after heat treatment. Fourier transform infrared (FTIR) (Brundle et al., 1992) measured film transparency with respect to air and with resp ect to the silicon or sapphire substrate. A dual beam ultra violet visible (UVVIS) spect rometer measured transparency from the ultra violet region to the near infrared region (~200 nm-3300 nm). Deposited film composition was determined using x-ray phot oelectron spectroscopy (XPS) (Brundle et al., 1992) and secondary ion mass spectrome try (SIMS) (Brundle et al., 1992). 5.3 Results 5.3.1 Electrical Properties Solution deposited films. Samples with rhodium substituted for cobalt have reasonably high conductivity with values between 110 Scm-1 and 290 Scm-1 (Figure 5-2)

PAGE 131

112 compared to other solution deposited samp les with conductivity values up to 70 Scm-1 from aqueous solution deposition methods reported in Chapter 4. Figure 5-2. Effect of rhodium fraction (h) on conductivity for four-layer films on silicon substrates deposited from solution. The addition of rhodium to nickel-c obalt oxide was predicted to increase conductivity. As seen in Figure 5-2, the rhodium concentration doubles between each point on the plot and the c onductivity staying nearly the same, except where h = 0. Nickel-cobalt oxide films were originally reported to have conductivity values of ~10-90 Scm-1. It is possible that using a combustion ag ent such as malonic acid in an ethanol solvent, instead of glycine as a combustion agent in a water solvent, may have changed the film so the bake-off procedure creates more voids. Malonic acid as a combustion agent, if added in too high of a concentration, causes the film to appear to boil as microscopic bubbles emerging from the bulk of the film shatte r the surface. The visible pock marks could easily be a source of reduced conductivity from the discontinuous conduction lines through the craters. Sputtered films. As in the nickel cobalt oxide system, the sputtered films show increased conductivity over solu tion deposited films by only a factor of 2x or 3x. It appears that there is a slight decrease in conductivity when replaci ng cobalt with rhodium

PAGE 132

113 (Figure 5-3). It reduces conductivity, but only s lightly for the alloy sputtered NiRh4Ox film in Figure 5–3. Figure 5-3. Sputtered nickel -rhodium oxide and nickel-coba lt oxide films deposited at 7.5 cm in 10 mTorr of 100% oxygen. Sputtered films of NiRh4Ox show the e ffect of changing target to substrate distance on resistivity in Figure 5-4. Similar to the nickel cobalt oxide system (Figure 38), the electrical resistivity increases w ith increased target-substrate distance. Figure 5-4. Resistivity of NiRh4Ox as a function of target-sub strate distance. Increased distance increases resistivity

PAGE 133

114 It is probable that the growth mechanis m is governed by the molecular mean free path and plasma-substrate interaction whic h will produce a lower density film at the longer distance. 5.3.2 Optical Properties Solution deposited films. Transmission spectra of NiCo2-hRhhO4 films 4 layers thick are shown in Figure 5-5 after correction for the silicon substrate absorption peaks. The most conductive film is th e least transparent in the IR spectrum and contains no rhodium. Spinel absorption lines app ear in all spectra near 550 and 650 cm-1. The spinel absorption peaks for NiCo2-hRhhO4 are sharpest in the films where h = 0 most probably due to the highly crystalline spinel film stru cture (Windisch et al., 2002a). As discussed in Chapter 3, increased conductivity accompanies a concomitant decrease in transparency. Figure 5-5. Solution deposited NiCo2-hRhhO4 FTIR transmission spectra were all corrected for the silicon substrate absorption peaks. Sputter deposited films. A comparison of calculated absorption coefficients described in Equation 3.6 is shown in Figure 5-6. As s ubstrate distance increases, the

PAGE 134

115 calculated absorption coefficient decreases indicating that increased target-substrate distance leads to a lower density. Decrease d density is responsib le for the increased electrical resistivity for films deposited at different distances described in Chapter 3. Figure 5-6. Resistivity and optical absorp tion coefficient at 3 m as a function of sputtering target-substrat e deposition distance. Figure 5-7. NiRh2Ox FTIR data referenced to air (low er blue trace) and corrected for the silicon substrate (upper red trace). FTIR data of a NiRh2Ox film with a transmission near 70% at 700 cm-1 is shown in Figure 5-7. Optical correction removes th e absorption bands normally seen from the

PAGE 135

116 silicon substrate (blue) in addi tion to the film features. The red data plot is essentially featureless suggesting that th e sputtered films are amorphous because the typical spinel absorption peaks are not observed. Figure 5-8. FTIR spectra of two NiRh4Ox samples of different thicknesses deposited from NiRh4 alloy target by DC sputtering. Both films are shown with and without substrate correction. Notice th at there are no sharp absorption bands at either 540-1 cm-1 or 630 cm-1 from spine absorption. Figure 5-8 shows FTIR data of a NiRh4Ox film with a transmission near 80% at 700 cm-1. Thin films (90 nm) of nickel-rhodium oxide reactively DC sputtered on silicon substrates show transparency of ~75% for f ilms nearly twice as thick as nickel-cobalt oxide (50 nm). Nickel-rhodium oxide optical transmission compared to nickel cobalt oxide optical transmission shows improved transparency at similar thickness with similar electrical conductivity. 5.3.3 Structural Properties Structurally the solution and sputtered f ilms are not identical. Sputtered films do not exhibit the spinel absorption peaks at 546 cm-1 and ~640 cm-1, while the solution films all exhibit the spinel absorption peaks. NiCo2-hRhhO4 films in Figure 5-9 show that

PAGE 136

117 as h increases, the spinel peak intensities a ppear to decrease, and the structure is more amorphous. Apparently the addition of rhodium does introduce disorder into the lattice as seen in the IR spectra for NiRhsO4 where s = 2 and s = 4. Both are nearly flat lines in Figure 5-9. Figure 5-9. FTIR of solution deposited NiCo2-hRhhO4 and sputtered NiRhsO4 thin films corrected for silicon substrate absorpti on. All of the solution compositions show the spinel absorption bands near 640 cm-1 and 540 cm-1. None of the sputtered films show the spinel absorption lines. The higher frequency mode at ~640 cm-1 is correlated with the octahedral site while the mode at ~540 cm-1 is a function of both the octahedral and the tetrahedral site vibrations. The differences between sputtered and solution deposited films, which are amorphous and crystalline spinel, respectiv ely, probably arise from the growth mechanism being limited by the growth speed. Far infrared spectra of the 4-layer so lution deposited films show the spinel absorption at 546 cm-1, which appears to shift to lower frequency as rhodium concentration increases. The film with a co mposition of h = 1 does not appear to have a significant spinel peak despite the stoich iometry of equal parts nickel, cobalt and

PAGE 137

118 rhodium. This stoichiometry should allow r hodium and nickel to occupy all of the octahedral sites and cobalt should occupy the te trahedral sites. Also evident from these data is the broad peak near 400 cm-1 attributed to the Ni3+ cation state when no rhodium is present. Figure 5-10. Far IR spectr a of solution deposited NiCo2-hRhhO4 from FTIR show the structure of the films. Figure 5-11. XRD of NiRh4Ox film from DC all oy sputter deposition. Sputtered films did not exhibit XRD peak s (Figure 5-11 and Figure 5-12). As suggested above, the addition of rhodium adds distortions to the lattice. Beyond what was predicted, the rhodium actua lly distorts the la ttice beyond the Jahn-Te ller effect to an

PAGE 138

119 amorphous state. Both NiRh4Ox and NiRh2Ox compositions of spu ttered nickel-rhodium were amorphous. Film heat treatment up to 600C followed by rapid quenching did not improve the film crystallinity from the amorphous state. Figure 5-12. XRD of amorphous NiRh2Ox thin film from DC alloy sputter deposition. 5.3.4 Chemical Properties Nickel rhodium appears to be stoichiome tric on the surface as expected, however depth profiling shows, the films lack oxygen. It has not been determined whether this lack of oxygen in the bulk of the film is intr insic to the material or an artifact of its preferential removal. Interaction of oxyge n with the argon sputte ring beam used to perform the depth profile (Smentkowski, 2000). If this oxygen depletion is intrinsic to the film, dangling bonds would be a source of blocked polaron hopping and may cause a reduction in conductivity. The oxygen depletion may be a result of the argon bombardment that takes place during the XPS depth profiling. Surface stoichio metry, while not always indicative of the bulk, may actually be more representative of the bulk of the film XPS depth profiles show rhodium in the silicon substrate d eeper than would be expected. Another

PAGE 139

120 possibility is that the rhodium-oxide octahedr al bonding, being less covalent because it is a 5d bond with a 2p orbital as opposed to a cobalt 3d bond with a 2p orbital may more easily releases the oxygen in the lattice. The result w ould be a diffusion of oxygen through the native silicon dioxide layer to form a deeper SiO2 layer. As rhodium is reduced, which is thermodynamically favor able, the formerly bound oxygen is free to diffuse through the lattice and possibly into the substrate as seen in Figure 5-13 with the extended tail of the red trace. Figure 5-13. XPS depth profile showing the surface of sputter deposited NiRh2Ox is different than the bulk concentration. Concentrations of Ni and Rh at 35 nm indicate diffusion into the substrate. Depth was calibrated by sputtering a standard and confirmed with opt ical and stylus profilometry. Data in Figure 5-14 are similar to Figure 5-13 except that the ratio of nickel to rhodium is 1:4 instead of 1:2. Films in Figure 5-14 were deposited under the same conditions as the film in Figure 5-13, but fo r a longer period of time giving a thickness of ~200 nm compared with ~35 nm. The film surface in Figure 514 also appears oxygen depleted prior to sputtering. In both cases, the oxygen always decrea ses with the onset of sputtering indicating that the surface is oxidized, but the sput tered bulk of the film is

PAGE 140

121 reduced. The concentrations of rhodium, nickel, and oxygen appear to be constant through the film to the substrate interface. Figure 5-14. XPS depth profile of a film sputtered from a NiRh4 target shows an oxygen depleted surface. The multiplexed data from the XPS depth prof iles, shown in Figures 5-15 indicate that the chemical state of th e nickel and rhodium species ch anges with depth. At the interface, the O 1s peaks shift to higher energy indicating a region of silicon dioxide (black data plot in Figure 5-15c and Figure 5-15d). The Ni 2p3/2 doublet peak located at ~855 eV and ~861 eV in Figure 5-15a and Fi gure 5-15b appears to change immediately upon sputtering with the peak near 861 eV decreasing in intensity with increasing sputtering time. Rhodium also a ppears to shift chemical state at the onset of sputtering as the Rh 3d line (shown in Figure 5-16e and Figur e 5-16f) at 309 eV shifts to lower binding energy near 308 eV.

PAGE 141

122 Figure 5-15. Multiplex binding energy data from XPS depth profiles. Nickel 2p binding energy region of films deposited from (a) NiRh4 and (b) NiRh2 alloy targets. Oxygen 1s binding energy region of films deposited from (c) NiRh4 and (d) NiRh2 alloy targets. Rhodium 3d bind ing energy region of films deposited from (e) NiRh4 and (f) NiRh2 alloy targets. Data plotted near the base of each plot is the surface. Successive stacked data plots are deeper in the sample towards the film-s ubstrate interface. 5.4 Discussion More polarons may form with the increas ed lattice distortion brought about by the substitution of rhodium for cobalt, however, an overall increase in conductivity is not observed. An increase in polaron density may not make up for discontinuous conduction paths from the increased structural disorder that allows the formation of the extra carriers.

PAGE 142

123 This results in a decreased mobility due to obstructed hopping. More polarons cannot lead to higher conductivity if they cannot hop to c ontribute to the current. Evident from solution deposition parameters the films may have micro craters as a result of the burning of the malonic acid a nd ethanol that occurs during the deposition process heat treatment. Cr aters could also obstruct hoppi ng along the octahedral sites (illustrated in Figure 3-18 as crisscrossing lines of conduction), but were not considered critical because all films were deposited under identical conditions and multiple film coatings were essentially continuous. Similar to the process study in Chapter 3, increasing the distance of the target to the substrate for sputtered samples increases resistivity while the absorption coefficient decreases. This effect is attributed to th e combination of lost energy due to increased collisions as distance increases and a reduced plasma interaction at the longer distance so less energy is transferred to the substrate. Lower energy transfer at the substrate equates to a lower substrate temperature and lower surface mobility. An energetically preferred structural arrangement may not occur with lower surface mobility from the lower energy at the surface and may result in discontinuous conductivity path s along octahedral sites. Longer sputter distance results in lower film density, increased resi stivity, and increased transparency, similar to the nickel-cobalt oxide system. Optical transmission data show some de gree of spinel structure in solution deposited films exhibited by charac teristic spinel peaks at ~540 cm-1 and ~640 cm-1. Far infrared spectra show a shift to a lower frequency of the 546 cm-1 spinel peak with increasing rhodium concentration accompanied by a broadening of the band. This shift is expected because the higher mass of rhodium will cause a shift to a lower vibrational

PAGE 143

124 frequency in the octahedral mode. Far infr ared data also indicate a characteristic Ni3+ band near 380 cm-1 for the NiCo2O4 sample that appears to weaken with increased rhodium concentration. These results agree with the proposed incorporation of rhodium in the octahedral site. Weakening of the specified absorption p eaks agrees with the addition of structural disorder to the system with increase d rhodium concentrations. The amorphous sputter deposited films of NiRh2Ox and NiRh4Ox are surprising, but not unexpected. The flat-lined infrared data at 640 cm-1 and 540 cm-1 were confirmed with featureless XRD scans indicati ng an amorphous structure. The amorphous structure may occur because both nickel and rhodium are octa hedral site occupants. When rhodium is partially substituted for cobalt, it is probable that the cobalt remains in the tetrahedral sites affording a measure of spinel phase stabil ity. With no tetrahedral site resident, the spinel structure is not supported. As seen in the IR spectra in Figure 5-9, the decrease in spinel vibrational modes with the decrease in co balt concentration supports this assertion. Previous data indicate th at powder spinel nickel-rhodi um oxide was produced via several days of high temperature proce ssing exceeding 1100C (B lasse, 1963). XRD data was not shown to be pure spinel. Da Silv a Perieira et al. indica te that nickel oxide separates from electrodes produced at te mperatures of 500C, a finding that is commensurate with the results shown by Windisch et al. show ing nickel oxide precipitation at 400C in the ni ckel-cobalt oxide thin films (D a Silva Pereira et al., 1994; Windisch et al., 2001b). The amor phous sputtered films of NiRh2Ox and NiRh4O in this study indicate possible film-substrate reactions. Blasse’s claims a tetragonal to cubic Jahn-Teller shift may occur and refers to XRD in his text and lets the reader assume all the materials were crystalline spinel. The temperatures used to process those nickel-

PAGE 144

125 rhodium powders would have likely contaminat ed the samples with traces of precipitated nickel oxide. Further high temperature treat ment tests of nickel-rhodium oxide films on infrared transparent, other than silicon, is required to address this disparity of information. It is probable that this disordered struct ural arrangement gives the nickel rhodium its high conductivity, though its conductivity is not be tter than crystalline NiCo2O4 suggesting that there may be a maximum de nsity of polarons in a given volume that limits the maximum conductivity for a pol aron conducting film. A discussion of maximum conductivity will follow in Chapter 6. XPS depth profiles of sputtered nickel r hodium films indicate that the film surface may be more oxidized than the bulk film seen by shifts of metallic peaks to lower binding energies at the onset of sputtering. This does not necessarily indi cate that the film is substoichiometric in the bulk. The disparity between bulk and surface may, however, be due a preferential sputtering artifact articulated by Smentkowski, who explored the sputtering of various metal oxide systems a ddressing the problem of the preferential removal of oxygen over the metal cation duri ng sputtering. He addresses many systems that show a depletion of oxygen on the surface after the process of sputtering used in depth profiling (Smentkowski, 2000, Holloway & Nelson, 1979). The results from this study are inconclusive as to whether the bulk of the film is partiall y reduced in the bulk or completely the result of an artifact from s puttering, a point to be further investigated. Depth profiles also indicate that rhodium and nickel diffuse into the silicon wafer beyond the native oxide layer. Formation of metal silicides has not been confirmed. Nickel silicide is highly c onductive and used extensively as an interconnect material in

PAGE 145

126 the microelectronics industry (Xu et al ., 1998). Thermodynamically the formation of rhodium silicide or rhodium silicate is more favorable than nickel silicide based on a free energy calculation, but kinetics of formation of one phase versus th e other has not been studied. Nickel oxide has been shown to form nickel silicide under the influence of argon bombardment (Gonzalez-Elipe et al., 1991). It is possible that a ba se layer of nickel silicide forms during the initial nickel-rhodium oxide film deposition or during the sputter depth profile of the XPS anal ysis; however, a silicide phase formation as shown by Gonzales-Elipe et al. exhibits a single strong bind ing energy peak near 853 eV from XPS, which is not observed in this analysis. Binding energy peaks at 854.5 eV and 862 eV merely broadened and appear to co nverge, indicative of reduction. It may be that the heat from the dept h profiling argon beam induces localized diffusion of nickel and rhodium into the subs trate during sputtering. The idea of rhodium diffusion into the substrate may be plausible, but once again, the resu lts are not definitive to support this assertion. Future studies w ill need to further ela borate on the interaction of the argon plasma with the ni ckel-cobalt-rhodium oxide system. 5.5 Conclusions Conductivity decreases with increased rhodium concentration dropping from 290 Scm-1 to about 110 Scm-1 for solution deposited samples. Sputtered oxide films from alloy targets of NiRh4 and NiRh2 with conductivities of 260 Scm-1 and 225 Scm-1, respectively, show comparable conductiv ity with nickel-cobalt oxide at 310 Scm-1. Sputtered film resistiv ity increased from 3.8 mcm to 48 mcm with increased targetsubstrate distance similar to observations in ot her nickel cobalt oxide films that increase in resistivity with increased target-substrate sputtering distance (Chapter 3).

PAGE 146

127 Infrared transparency for solution deposited NiCo2-hRhhO4 films compositions with h = 0, 0.25, 0.5 and 1 show improvements of 10-15% and a decrease in conductivity of nearly 50%. Sputtered films exhibit an incr ease in the optical absorption coefficient at 3 m as target-substrate distance decreases from 30 cm to 7 cm. This change in absorption typifies increased density with shorter target-substrate di stance, common in sputtered films. FTIR data of sputtered films s hows that oxide films deposited from the NiRh4 alloy target were similar in transparency and more conductive than the oxide films of similar thickness deposited from the NiRh2 target. Structurally, the sputtered films are am orphous. The solution deposited films exhibit spinel absorption peaks at 546 cm-1 and ~640 cm-1 in the infrared spectrum that broaden and shift to lower frequency with increased rhodium concentration. Low frequency shifts are expected from the larger mass of the rhodium ion compared to the cobalt ion being replaced. XPS results show that the film compos ition was homogeneous through the bulk of the film. The disparity of oxygen concentrati on at the surface and th e bulk of the film may be real or an indication of preferentia l removal of oxygen that occurs during the XPS sputter depth profiling. Silicide formation at the surface of the silicon wafer and the nickel-rhodium film most probably does not occur; however the presence of a silicide film is not completely ruled out. Amorphous nickel-rhodium oxide structures from sputtering are unexpected, but not surprising because the addition of rhodium, by design, is to induce structural disorder and allow increased polaron formation. Structur al disorder is believe d to be due to the octahedral occupation of both nickel and rhodium with neit her occupying the tetrahedral

PAGE 147

128 position. While more polarons may form, a competing degradation of the conduction pathways due to the disordered struct ure impedes the hopping motion for increased electrical conductivity.

PAGE 148

129 CHAPTER 6 DISORDER AND POLARON CONDUCTIVITY 6.1 Introduction While it is customary to think of electrons and holes as conducting elements in a solid state lattice, when the lattice is distor ted by these charges, sp ecial properties result and the charge carrier under these conditions is called a polaron. Since polarons are electrons or holes trapped in bandgap quantum states associated with lattice distortions, their motion is thermally activated resulting in low mobility at room temperature (< 0.1 cm2V-1s-1). Polarons form as the lattice distor ts to accommodate the extra charge state, creating the potential well, and effectively trapping the carrier in the self-induced strain region. Similar to electrons and holes, conduc tivity is proportional to the product of the polaron density times its mobility. When the lat tice is disordered, an electronic carrier is more easily trapped. In the case of polarons disorder increases th e polaron density and increases the conductivity of the film. Two types of disorder are discussed in th is chapter (see Figure 6-1). First is a structural disorder of the cations and anions that constitute the crystal lattice (Section 6.2). The films produced were expected to be a pure phase of crysta lline spinel rather than amorphous. Typical amorphous structures exhibit little if any long-range structural order. The second disorder type deals with the disordered distribution of the cation charges in spinel film (Section 6.3). Cation charge disorder, attributed to a Verwey-like order/disorder transition (Chapt ers 3 and 4), affects the activ ation energy of nickel-cobalt oxide films as a function of the rate of coo ling that follows heat treatment. A model for

PAGE 149

130 the cation charge disorder based on localized carrier interactions that detail a similar effect on activation energy is described in this chapter. Section 6.4 discusses an estimate for maximum conductivity in the spinel nickel -cobalt oxide system c onsidering realistic boundaries of disorder. Figure 6-1. Schematic of disorder. (a) Am orphous structural disorder, (b) an ordered structure, and (c) a crystalline structure with disordered charge distribution. Red corresponds to one oxidation state of a cation, while blue is a second oxidation state. 6.2 Structural Disorder Structural disorder is the random arrangeme nt of ions in a solid. In many cases, lattice disorder can be correlated with a d ecrease in crystallinity Due to the random atomistic arrangement, structural disorder produ ces a variety of trapped states (polarons) with a distribution of activation energies that enable conduction. Work by Emin and Tauc has shown that disorder in amorphous SiO2 can produce polarons by trapping holes in the oxygen p orbitals (Emi n & Bussac, 1994; Tauc, 1976). Figure 6-2 shows the effects of nickel c oncentration on activa tion energy. With lower nickel concentration, the film conductivity decreases and activation energy increases. The addition of nickel was show n to induce disorder. The activation energy shifts in nickel-cobalt oxide thin films as a function of composition as shown from temperature dependent conductivity data (Windi sch et al., 2002b). Data in Figure 6-2a

PAGE 150

131 were gathered during the heat ing cycle from samples prev iously quenched after being subjected to heat treatment of ~375C for 10 minutes in air. Figure 6-2. The effect of composition on temperature dependant conductivity and activation energy. Plots of (a) Arrhen ius conductivity plot and (b) activation energies calculated from the two different energy regions in plot (a). (Based on data presented by Windisch et al., 2002b, Figure 5, p. 93, slightly modified and color added). Optical techniques such as laser Rama n and Fourier transform far infrared absorption spectroscopy were also used to show the effects of nickel in the cobalt oxide lattice. Raman peaks for cobalt oxide we re seen to rapidly broaden to amorphous features with small additions of nickel (W indisch et al., 2002a; Windisch et al., 2002b). An additional study by Windisch et al. showed that a mode near 373 cm-1 increased in intensity with an increased concentration of Ni3+ as activation energy decreased indicating polaron formation (Windisch, 2003a). These films structures were reported as spinel despite the change in nickel concentr ation suggesting this system may have been affected by disorder. A shift of NaPO3 from a crystalline to amorphou s structure induced by processing is shown in Figure 6-3. The amorphous NaPO3 structure is made by rapid cooling the melt on a metal plate. It exhibits a broad, re latively featureless Raman band in the far IR,

PAGE 151

132 whereas the crystalline state (made by heat treating the amorphous material at 300C for 2 hours) shows distinct peaks correlating with specific molecular vibrational modes. Figure 6-3. NaPO3 glass phase (amorphous) with a broad band covering the bond vibrational modes. Crystalline NaPO3 phase shows distinct individual active vibrational modes easily dis tinguished due to the ordered structure of the film. Reprinted with permission from Exarhos, G.J. (1974) Spectroscopic investigations of glasses and polymers. Ph.D., Brown University. XRD data for the nickel-cobalt oxide data in Figure 6-2 did not show a decrease in crystallinity with the increase in nickel concentration. Windi sch et al. also reported that the lattice parameter increased with increased nickel concentration in cobalt oxide (Roginskaya et al., 1997; Windisc h et al., 2002b). This shift in lattice parameter suggests that spinel remains as the major phase, nickel incorporates into the lattice (Windisch et al., 2002b), and that structural la ttice disorder may be part of but is not the entire cause of increased conductivity.

PAGE 152

133 Data from the nickel-cobalt-rhodium oxide system prepared by solution deposition (presented in Figure 5-9) show that with increased rhodium, the structure shifts to an amorphous state. More structural disord er in that case did not cause improved conductivity, an unexpected result based on Em in’s predictions. The shift to an amorphous state does not necessarily increase the conductivity (presented in Chapter 5 and illustrated in Figures 5-2 and 5-3). St oneham describes transition metal oxides from a modeling standpoint and addresses the eff ects of impurity additions. His modeling results indicate that the addition of lithium impurities to nickel oxide cause a trapped hole to localize on a cation adjacent to the substi tutional impurity (Stoneham, 1980). Trapped holes localize on the rhodium impurity in the NiCo2-hRhhO4 system and require more energy to hop because of the rhodium inter action with the cobalt in the lattice (as h increases from 0 to 1). When the rhodium completely replaces th e cobalt, conductivity values are nearly recovered, suggesting that the carrier gets trapped at the im purity site. Stru ctural disorder introduced by the substituted cation does not produce the desired result of increased conductivity. This may be caused by the coup ling of the disorder with some type of bonding anomaly such as a change in tr apped-carrier activation energy from the interaction of rhodium with its neighbors. Tauc reasons that carriers are much less mobile in glasses than in crystals (Tauc, 1976) It is probable that mobility decreases due to disrupted carrier paths within the i rregular lattice as th e increased rhodium concentration shifts the spin el to an amorphous state.

PAGE 153

134 Raman data of nickel-cobalt oxide show some degree of disorder may occur as indicated by the broadening of the spinel band s. This disorder is probably a decoupling of the active modes from the JahnTeller effect rather than a sh ift to an amorphous state. 6.3 Cation Charge Disorder 6.3.1 Introduction Verwey et al. postulated that the charge distribution with in a structured solid may affect the activation energy of el ectrical conductivity in spinel Fe3O4 (Verway & Haayman, 1941; Verwey et al., 1947; Verwey 1939). Cation charge order/disorder is suggested to decrease/increas e conductivity, respectively. A transition that occurred during conductivity experiments was attributed to a disorder/order charge distribution within the crystalline lattice. This reporte d transition for the iron oxide spinel system occurs below room temperature near the Neel temperature of 200K for Fe3O4 (Verway & Haayman, 1941; Verwey et al., 1947; Verwey, 1939). Nickel-cobalt oxide films quenched after he at treatment are more conductive than the slowly cooled films (shown in Figure 3-23) and exhibit a decrea sed slope (activation energy) in the Arrhenius conductiv ity curve. This change in slope is attributed to a Verwey-type transition caused by an increase cation charge disorder induced by the kinetics of cooling. Film conductivity after quenching decreases with exposure to atmosphere. Equilibration requires time for kinetics to reduce the film to a more favorable energy state of an or dered charge distribution. Sl ow cooling of the films after heat treatment shows effects similar to pr olonged exposure to atmosphere. Quenched-in cation charge disorder and its effects on th e activation energy fo r carrier conduction are modeled in the next section by a site-to-site interaction model.

PAGE 154

135 Charge disorder may be affected by a ddition of impurities. The addition of impurities may affect the populations of defect s (polarons form at defect sites), in addition to the types of defects and the m echanism allowing more to form (Stoneham, 1980). Cation charge disorder affects carri ers hopping between two neighboring ions. This interchange requires two ions of differe nt oxidation states. If the cation order is random, then the probability of an adjacent site where the carrier can hop will be high. Bringing the concept of a bound carrier together with th e influence of disorder, there are several classical models for a bound ca rrier in a potential well. In the following, a model for polaron conductivity of the nickel-cobalt oxide spinel system is outlined based on the particle in a box model for lattice dynamics and a spin model for carrier disorder effects. 6.3.2 Disorder and the Polaron Hopping Activation Energy Model Figure 6-4. A double well potential with the as sociated energies th at influence carrier motion between the wells. EBW labels a positional distance whose value depends upon the distance the carrier must hop between the two wells (r). The carrier is found in position 1 (blue) or position 2 (red). The particle in a box model approximates a carrier in a one dimensional well with infinite potential energy surface barriers (descr ibed in Chapter 2). The particle is bound to the regions inside the 1-D box (energy well). Two wells placed side by side share an adjoining potential barrier with a finite valu e. Two mechanisms are possible that allow a

PAGE 155

136 trapped carrier in state 1 to m ove to state 2 (Figure 6-4). Th e first is to tunnel through the barrier, and the second is to hop over it. A ssuming tunneling is negligible at room temperature, there are three energies that co mbine to equal the activation energy required for the barrier hopping process: 1. Barrier height, EBH. Energy of the barrier height is influenced by the polaron nature i.e. the atomic species involved with the self-trapped state or the cation-anion interacti on at the occupied site. 2. Barrier width, EBW. Energy of the barrier widt h is directly proportional to the distance between two hopping sites. Hopping distance is also described as the polaron size. The extent of the local distortion in the lattice is what classifies a polaron as large or small. A large polaron displaces atoms beyond its near neighbors, while a small polaron is highly localized to the site and its nearest neighbors. As more disorder is introduced, barrier widths may become distributed in sizes. 3. Well depth, EWD. Energy of the well depth on either side of the barrier is not necessarily the same in a disordered system. In a perfectly ordered system, the well depth would be the same. This base energy value of organizational energy is affected by th e interaction between polarons based on the polaron distribution throughout the lattice. Disordered cation charge arrangement, grain structure defects, or an amorphous arrangement affect this energy.

PAGE 156

137 The well width is not considered because th e lattice is assumed to be a distribution of well sizes within the range of being classifi ed as small polarons. The extent of a small polaron wave function is not greate r than the lattice parameter. As temperature increases, the probability of a hop occurring increases from well depth (EWD) or barrier height (EBH) reduction. Physically this correlates to increased lattice vibrational motion that will increase the probability of the polaron being in a position energetically favorable to move. Availability of neighboring sites and neighboring carrier interactions also affect the activati on energy required for hopping. The interactions of the carri ers and their effect on the ac tivation energy have important implications on the conductivity Conductivity has been rela ted to cation disorder in previous sections. The relati onships between disorder and activation energy needs closer examination. Conductivity is the product of the carrier density (n) multiplied by the mobility of the carrier () and the charge of an electron (q) as shown in Equation 6.1: q n (6.1) Since the carrier charge is constant at 1.602x10-19 C, the only ways to affect conductivity is to create mo re carriers by producing distortions in the lattice where carriers will be trapped, or arranging the char ges in the lattice such that the activation energy for movement is minimal. Rapid cool ing from heat treatment must affect the mobility in order to change the conductivity because it changes the activation energy. Activation energy decreases will increase carrier mobility. As will be shown by the disorder model, a disordered stat e decreases the activation energy. While the particle in the box model for pol aronic materials is illustrative for the hopping nature of conduction, the heterogeneity of the trap energies and the effect of

PAGE 157

138 disorder need to be examined in closer detail. For the particle in the box model, polaronic sites are represented without in teractions with the remaining material. If we were to picture electronic cond uction between adjacent sites, the question remains: how do the neighboring sites affect the conduction process? Using a spin model for site-site interactions, this question is examined. Both one and two dimensional spin lattices are illustrated in Figur e 6-5a and b, respectively, with the charge carrier (electron or hole) residing at the center (darker colored) site. To represent the interaction between neighboring sites, the dipole-di pole interaction ener gy is used as a first approximation of the interaction between sites for both a 1-D and 2-D solid (Figure 6-5, a and b). Figure 6-5. Schematic illustration of one (a ) and two-dimensional (b) spin models for polarons defined by adjacent site inter actions. Darker colored site is coordinate origin. Equation 6.2 shows the relation of the dipole stabilization energy between two dipoles at neighboring si tes. The dipoles (1 and 2) are located on adjacent sites (1 and 2) spaced at a distance (r). The dipoles are oriented at 1 and 2. 3 0 2 1 2 1 2 14 sin sin cos cos 2 ) ( r u u r E (6.2)

PAGE 158

139 Figure 6-6. Illustration of dipole-dipole inte ractions between adj acent sites for polaronic spin model variables in Equation 6.2. Figure 6-6 schematically illustrates the dipol e magnitude and orientation. Position 1 in pink is canted slightly towards dipole 2 that is nearly normal to the distance vector r. For this model, the orientation angle () of each dipoles is constrained between 0 and 90 degrees. Imposing a unit dipole moment ( = 1) on all sites, and replacing the numerical constants with ‘k’, Equation (6.2) for the one-dimensional case reduc es to Equation 6.3. 2 1 2 1sin sin cos cos 2 ) ( k r E (6.3) The interaction energy for an indivi dual site becomes Equation 6.4. E ( site ) E ( dipole dipoleabove) E ( dipole dipolebelow) (6.4) When the spins of the adjacent sites are al l perfectly aligned (maximum alignment), the largest dipole-dipole stabi lization energy is found (highest activation energy). Using this relationship and an analogous form for the two dimensional case, heterogeneity is introduced into this system by generation of a random distribution of orientations of the dipole vectors with respect to each other.

PAGE 159

140 Figure 6-7. Energy distributi on functions for one-dimensional polaron disorder model. The distribution is taken as an ensemb le average containing 10,000 sites with interaction energies normalized to an ideal ordering case where equals 0. Using a 10,000 element lattice, the follo wing histograms (Figure 6-7 and Figure 68) of interaction energies were generated fo r one and two-dimensiona l lattices for random distributions of dipole orientations constr ained within a specified angular amplitude. Interaction energies were normalized to the maximum value as determined by the ideal alignment ( = 0) case. Figure 6-8. Energy distributi on functions for two-dimensional polaron disorder model. Distributions were taken from an ense mble of 10,000 sites, with interaction energies normalized to ideal ordering case where equals 0.

PAGE 160

141 By introducing this heteroge neity, the effect of disord er is shown to create distribution of site energies, with the maximum value varying as a function of the extent of angular variation. As the experimental system will exhibit a degree of disorder, the ensemble-average value is constituted from an energetically defined distribution of sites, with site energies both higher and lower than th e measured value. In terms of the particle in the box model described earlier, the ener gy level of the polaronic site is better described as a distribution of energy states with an accompanying range of activation energies for site hopping. 6.4 Resistivity Limit When nickel is added to cobalt oxide, improvements in conductivity were on the order of 102 104 from a lower value of 0.001 Scm-1 up to ~333 Scm-1 (Windisch et al., 2001a; Windisch et al., 2001b; Windisch et al., 2002b). Conductivity reported in this work is up to 500 Scm-1 with the enhancement from processing. Both lithium and rhodium dopants in addition to several di fferent process optimizations designed to increase conductivity* showed some small measure of enhancement. The conductivity in these materials remains high due to the hi gh carrier density despite the low mobility characteristic of the hopping mechanism of polaron motion. The rational behind dopant additions or cation charge di sorder was to create more carriers for conductivity. More carriers, created by dopant addi tion, cation charge disorder or induced by processing were designed to increase conductivity. There is a realistic saturati on limit of polarons in the lattice. It is possible that a saturation lim it of polarons exists in the lattice. Once the saturation limit is reached, additional carriers will only increase the lattice distortion and See Figures 3-7, 3-9, 3-13, 4-1, 42, 4-3, 4-4, 5-2, 5-3, 5-4, 5-6)

PAGE 161

142 impede the hopping motion of available carrier s. Polaron saturation will impose a lower limit to resistivity. Hopping distance (barrier width) could change, or the localization of the carrier (well depth) could increase due to the increased strain. Additional interaction among carriers may also degrade the hopping process. Based on a unit cell with a theoretical la ttice parameter of 8.114 and assuming the crystal is 100% dense, an estimation of the number of allowed pol arons in one spinel unit cell is made. This estimation of the sa turation limit of polarons in the material will allow a projection of the maximum conductivity in these materials. It should be noted that these estimates are based on a number of idealized conditions and that this limit is a maximum achievable goal for these c onditions. A number of processing and compositional factors will affect the validity of this model and its agreement with experimental measurements. A maximum of one polaron per lattice cation (each polaron is a bound carrier) would yield a total of 24 cation sites in a spinel cell made up of 16 octahedral and 8 tetrahedral coordinated locations. It is unrealistic to assume that every site will be distorted to create a polaron, so the number of polaron site s per unit cell is an arbitrary value chosen that is less than the total num ber of sites available. Table 6-1 shows the available sites and which cations occupy them. Table 6-1. Spinel sites and occupa tion for polaron density calculation. Octahedral Tetrahedral Oxygen (anion) Total Sites Available 16 8 32 Ni 8 or 12* Co 8 or 4* 8 Li** YES Rh** YES *Depends on the metallic ratio of nickel and cobalt —if equal parts, nickel will occupy the octahedral site. If the c oncentration of nickel is grea ter than the concentration of

PAGE 162

143 cobalt then nickel oxide precipitation may reduce the accuracy of the site density calculation. **Li and Rh are included to indicate which sites are affected as they are added. Since the increase in conduc tivity is attributed to the increase in nickel concentration, polarons must form around the nickel ions, providing a starting point for calculating the theoretical conductivity. The spinel composition of NiCo2O4 has 8 nickel ions per unit cell. The theoreti cal polaron density, in units of cm-3 (Equation 6.5), is the product of the number of cells per cubi c centimeter (Equation C.2, Appendix C) multiplied by the number of sites per cell. 8 Ni cell 1 ( Polaron ) Ni 1.87195 1021cell cm3 1.49756 1022( Polaron ) cm3 (6.5) If the number of polarons per cell increases to 12, then the polaron density will increase to 2.246x1022 cm-3. Table 6-2 shows the calcu lated conductivity with the mobility at 0.1 and 1 cm2V-1s-1 when equation 6.1 is used with q as the constant charge of an electron equal to 1.602x10-19 C. Values of mobility shown were chosen based on Emin’s statement that small polaron hopping was indicated by a mobility less than 1 cm2V-1s-1 (Emin, 1982). Table 6-2. Theoretical value of conduc tivity assuming a polaron density given by the number of hopping polarons in a un it cell with a constant mobility. Polarons / Cell Polaron Density, cm-3 Mobility, cm2V-1s-1 Conductivity, Scm-1 0.8 1.498x1021 0.10 24 1.2 2.246x1021 0.10 36 2.4 4.492x1021 0.10 72 8.0 1.498x1022 0.10 240 12.0 2.246x1022 0.10 360 24.0 4.492x1022 0.10 720 0.8 1.498x1021 1.0 240 1.2 2.246x1021 1.0 360 8.0 1.498x1022 1.0 2400 12.0 2.246x1022 1.0 3600 24.0 4.492x1022 1.0 7200

PAGE 163

144 Windisch et al. stated that 10% of all nickel is in th e 3+ oxidation state. The Ni3+ is the polaron entity (Windisch et al., 2002b). The maximum would be 0.8 per unit cell for a theoretical density of 1.498x1021 cm-3. If mobility was 1cm2V-1s-1 then the conductivity would increase by an order of magnitude to 2400 Scm-1. High conductivity ITO films are reported with conductivities near 4,000 Scm-1 (Minami et al., 2000). Nickel cobalt oxide films will likely not reach those values unless the carrier mobility increases. Increased mobility will, however cause scattering collisi ons due to such high carrier concentrations and may not necessarily result in a higher conductiv ity. It is unrealistic to assume that all octahedral sites will host polarons, so the projected ma ximum conductivity of 7200 Scm-1 should be considered too high. This high concentration of pol arons would also distort the spinel lattice resulting in the lower mobili ty corresponding to the lower conductivity that was reported earlier. Thus the maximum realis tic conductivity of the nickel-cobalt oxide system should be much less than 2400 Scm-1 in agreement with the experimental data where the highest reported value is ~500 Scm-1. The Hall effect data mentioned in Chapter 2 is anomalous with carrier densities as high as 5x1023 cm-3. This value suggests the physical ly unrealistic interpretation that every cation in the lattice is a polaron host simultaneously. A potentia l explanation of the carrier density exceeding the theoretical maximum is the thermal generation of carriers, resulting in a change of c onductivity mechanism, or a mi xed carrier population of both quasi-free electrons and p-type polarons. Thermal energy translated into lattice distortions would reduce the well depth and shift the conduction regime to a more quasifree type of carrier population. This argument is not a verifiable option as the Hall measurement is anomalous for this material. Seebeck data on the ma terial should support

PAGE 164

145 this mechanism. Further information on the anomalous Hall effect can be explored by reading the works of Emin (Emi n, 1977, 1980; Emin & Emin, 1977) 6.5 Conclusions This chapter discussed the effects of disord er in the nickel-cobalt oxide system with respect to structural disorder and cation char ge disorder with a model that illustrates the change in activation energy of electrical conduc tivity with a disordered arrangement of cations. The attempts to increase the polaron density in Chapters 4 and 5, suggest that a polaron saturation effect occurs wh en the polaron density reaches 2.25x1022 cm-3 limiting the conductivity to a range between ~360 Scm-1 and ~2400 Scm-1 from polaron hopping when the mobility is between ~0.1 cm2V-1s-1 and 1.0 cm2V-1s-1. Attempts to increase the polaron concentration to achie ve a higher conductivity may pe rturb the lattice and reduce mobility or carrier population resulti ng in a net decrease in conductivity.

PAGE 165

146 CHAPTER 7 CONCLUSIONS, FUTURE WORK, AND APPLICATIONS 7.1 Conclusions Nickel-cobalt oxide is the p-type material system of in terest to function as an infrared transparent conducting oxide (ITCO) because of its infrared transparency (up to 90% at 16 m), and relatively high conductivity (~500 Scm-1). Nickel-cobalt oxide exhibits these characteristics due in part to the intrinsic carrier type and the conduction mechanism attributed to the spinel crystal structure. This study investigated three methods of increasing conductivity. First was the optimization of sputter deposition proce ss conditions. Adjusting sputter process parameters was designed to adjust the growth conditions affecting film properties such as density, morphology or crystallinity and incr ease the film conductivity. The second method for increased conductivity was substituting lithium for cobalt. Replacing a divalent cobalt ion on a tetrahed ral site with a monovalent li thium ion was anticipated to shift a neighboring cobalt cat ion to a higher oxidation stat e and form a polaron. The increased polaron (carrier) de nsity would result in a net in crease in conductivity. The third method for increased conductivity in the nickel-cobalt oxide system was to substitute rhodium for cobalt. Rhodium is similar in electronic structure to cobalt, and is located in the row beneath on the periodic ta ble. Rhodium size was expected to induce structural distortions and provi de sites for polaron formation, resulting in an increased conductivity. Due to the shifted energy of th e d orbital electrons in the rhodium ion, transparency at wavelengths closer to the optical spectrum was expected to increase.

PAGE 166

147 Films from each study were subjected to heat treatment to ensure complete oxidation and to see the effects of thermal treatment. Heat treatment cooling studies suggest that cation disorder is a dominant f actor in polaron formation and a significant factor affecting the activa tion energy of electrical c onduction. A model relating activation energy to cation site disorder s howed that activation energy decreased as disorder increased. 7.1.1 Summary of Sputter Processing of Nickel-Cobalt Oxide For sputtered films, the conductiv ity was larger with a 50% O2-50% Ar composition, a lower sputter gas pressure (2 mTorr vs. 10 mTorr), and a smaller targetto-substrate distance (best at 5 cm). The e ffects of composition are due to the resulting cation oxidation states, while the effects of pressure and substrate distance were attributed to more energetic sputtered pa rticles causing increased mobility of surface adatoms on the substrate and film surfaces. Th e 5 cm target-to-substrate distance results in a 20% increase in film de nsity, compared to a 15 cm distance with both producing a crystalline spinel structure. The best film conductivity of the gas mixture series of ~350 Scm-1 may be a result of ideal growth conditions promoted by complete oxidation and a higher deposition rate. The ga s pressure study showed that a chamber pressure of 2 mTorr yielded a higher c onductivity film (~500 Scm-1) compared to the 5 mTorr (~400 Scm-1) and 10 mTorr (~333 Scm-1) for films rapidly quenched after heat treatment. This enhanced conductivity is attribut ed to a higher growth rate an d increased adatom mobility during growth due to the shorter molecular tr ansport path and more energy transfer from the plasma heating the substrate. Increased target to substrate distance decr eased the film density from 5.7 g/cm to 5 g/cm, increased film porosity, decreased conductivity from ~250 Scm-1 to ~35 Scm-1, and

PAGE 167

148 decreased the optical absorption coefficient at 3 m from ~105 to ~2x104. Targetsubstrate distance effects on the film density are believed to be controlled by phenomena similar to those from lower pressures of 2 mTorr instead of 10 mTorr i.e. higher surface mobility from depositing species due to less gas phase scattering. Less plasma-film interaction at longer distances may also be a factor. Band gaps of 3.5 eV, 3.4 eV, and 3.2 eV were reported from x equal to 0.75, 1, and 1.5 in NixCo3-xO4, respectively. Work functions of 4.39 and 4.27 were also reported for composition (x) values of 1 and 1.5, respectively. In summary, lower gas pressures of 2 mTo rr and short target-substrate distances of 5 cm resulted in conductivities of up to 500 Scm-1 and transmission of up to 40% (if corrected for the silicon substrate absorpti on and index mismatch, the transmission would be ~80%) at 7.15 m (1400cm-1). 7.1.2 Summary of the Role of Lithium Lithium additions to sputter and solution deposited films of NixCo3-x-yLiyO4 were found to decrease and increase th e electrical conductiv ity by less than a f actor of two. Lithium in solution films tends to increase conduc tivity when added in specific amounts, depending on the nickel concentration. L ithium improved electrical properties only in three sets of conditions: two compositions de posited from solution and one deposited by sputtering. The two solution de posited conditions were in NixCo3-x-yLiyO4 at y = 0.01 when x = 1 and y = 0.1 when x = 0.75. Hi gher nickel-to-cobalt ratios required less lithium for increased conductivity. The improved conductivity value of 72 Scm-1, up from 37 Scm-1 was still small in comparison to the high conductivity sputtered films without lithium (500 Scm-1).

PAGE 168

149 When sputtered, lithium concentrations up to 15% incorporate in the lattice according to SIMS and XPS analysis. Lattice expansion suggests this high lithium concentration resulted in occupation of lithium interstitials causing polaron annihilation. Excess lithium beyond the solubility limit segr egated to the surf ace, formed carbonate and resulted in reduced conductivity. The fo rmation of the carbonate surface layer was detected FTIR and XPS. In summary, lithium substitution in solution deposited NiCo2-zLizO4 films with a z value of up to 0.1 results in conductivity in creases of 35% and up to 84% transmission at 7.15 m. 7.1.3 Rhodium Substitution for Cobalt Sputtered oxide films from alloy targets of NiRh4 and NiRh2 with conductivities of 260 Scm-1 and 225 Scm-1, respectively, showed compar able conductivity with nickelcobalt oxide at 310 Scm-1. Sputtered film resistivity increased from 3.8 mcm to 48 mcm with increased target-substrate distance similar to observations in other nickel cobalt oxide films that increased in resistivity with increased target-substrate sputtering distance. An increase in conductivity was not observed with increased rhodium concentration in solution deposited sa mples as conductivity dropped from 290 Scm-1 to about 110 Scm-1. Sputtered nickel-rhodium oxide films are more transparent into th e visible and near infrared regions with a transparency of 60% at 600 nm with similar conductivity compared to nickel-cobalt oxide films. Transparency from 2 m to 10 m was improved by ~10% with the rhodium present. Data s how that nickel-rhodium oxide is just as conductive at 200 S/cm-1with nickel-cobalt oxide at similar thicknesses, but optically

PAGE 169

150 nickel-rhodium oxide is more transparent; however, nickel-r hodium oxide film properties change after heat treatment and do not show the reversible heat treatment properties possibly due to rhodium reduction in oxidation state. Sputtere d films exhibit an increase in the optical absorption coefficient from 1.5x104 to 5x105 at 3 m as the target-substrate distance decreased from 25 cm to 7.5 cm. FTIR data of sputtered f ilms shows that oxide films deposited from the NiRh4 alloy target were similar in transparency and more conductive than the oxide films of similar thickness deposited from the NiRh2 target. Infrared transparency for solution deposited NiCo2-hRhhO4 films of h equal 0, 0.25, 0.5 and 1 showed improvements of 10-15% and a decrease in conductivity of up to ~50%. The solution deposited films exhibited spinel infrared absorption peaks at ~546 cm-1 and ~640 cm-1 that broadened and shifted to lower frequency with increased rhodium concentration. Low frequency shifts were expected from the larger mass of the rhodium ion compared to the cobalt ion being replaced. Solution and sputter deposition of NixCo3-x-yRhyO4 films were crystalline (spinel) and amorphous, respectively, and remained so with heat treatments up to 550C. Attempts to crystallize the sputtered f ilms were unsuccessful. Amorphous nickelrhodium oxide structures from sputtering were unexpected, but not surprising because the addition of rhodium, by design, was to induce structural distortion and allow additional polaron formation. Structural disorder is due to the octahedral s ite occupation of both nickel and rhodium. With no tetrahedral site occupants, the spinel structure collapses unless cobalt remains to occupy it. While more polarons may have formed, a competing degradation of the conduction pathways due to the disordered stru cture impeded their hopping motion.

PAGE 170

151 In summary, sputter deposited NiRhsO4 films with an s value of up to 4 results in conductivity decreases of up to 22% and up to 70% transmission at 7.15 m. Solution deposited NiCo2-hRhhO4 films with an h value of up to 0.5 results in conductivity decreases of up to 60% and up to 70% transmission at 7.15 m because the rhodium created an amorphous structure impeding carrier motion, but did not greatly affect the optical transmission. 7.1.4 Heat Treatment Heat treatment showed improvement in conductivity by up to 2x when films were rapidly quenched and of up to 2x degradati on when slowly cooled. Transparency behaved opposite to the electri cal conductivity. After a rapi d quenching heat treatment, the activation energy (~ 0. 01 eV 0.08 eV) of conduction is about 0.01 eV lower compared to a slow cooled sample. The cha nge was attributed to a larger concentration of polarons upon quenching due to a greater de gree of disorder. This heat treatment cooling rate effect is a ne w discovery for this system. The behavior of films containing lithiu m was similar to those not containing lithium with respect to optical and electrical properties following heat treatment. Quenching (150C/min) produced higher conductiviti es (2x), while slow cooling (10C/min) results in lower conductivi ties after heat treatment at 375C for 10 minutes in air. Heat treatment increases conductivity by up to 2x and decreases transparency by up to 15%. In contrast to spu ttered films with large lithium concentrations (> 5% added) which caused a lattice expansion, heat treatment and quenching of lithium films from solution with small lithium concentrations caused slight lattice shrinkage according to XRD data fits. This anomaly of lattice shrinkage is not fully explained with an

PAGE 171

152 accompanying increase in conductivity. Increas ed lithium caused a rapid increased in the activation energy for film conductivity for bot h quenched and slowly cooled films, but the magnitude depends on the rate of cooli ng. Heat treatment effects upon conductivity and transparency were reversib le for films with and without lithium. It was concluded that the film was in a metastable state followi ng a rapid quench from after heat treatment, and that cation disorder was responsib le the 100% increas e in conductivity. 7.1.5 Disorder and Polaron Conductivity Having attempted with dopants to increas e the polaron density through processing and impurity addition, it is concluded that a polaron saturation e ffect probably occurs when the polaron density reaches 2.25x(10)22 cm-3 limiting the conductivity to a projected range between ~360 Scm-1 and ~2400 Scm-1 from polaron hopping when the mobility is between ~0.1 and 1.0 cm2V-1s-1. Attempts to increase the conductivity beyond this value will perturb the lattice such that either mobility decreases or the carrier population decreases resulting in lower c onductivity. The effects of cation charge disorder and structural disorder in the nickel-cobalt oxide system were modeled and mathematically related to activation energy of electrical conductivity. The model shows that as disorder increases, activation energy decreases, consistent with the experimental data. 7.2 Future Work Due to the non-existence of IR transmitting conductors, there is a pressing need to further develop the nickel-cob alt oxide system and expound on its limitations and provide a baseline of comparison for other material sy stems that may be developed in the future. Future work may include:

PAGE 172

153 Additional processing studies of nick el-cobalt oxide to find a deposition technique or geometry to produce a further optimized ITCO film. Maximizing both conductivity and transp arency while prov iding a figure of merit at various wavelengths in the infrared would be useful. Investigation on the effects of doping such as lithium and rhodium on the carrier population and whethe r they only affect the num ber of carriers or the mobility. This will require improved me thods of characterizing mobility. Increases in conductivity we re not conclusively due to one or the other. A method to determine experimentally the disorder in the system and whether or not super ion exchange or the neighboring environment surrounding the polaron influences its mobility. A detailed investigation into the model used to approximate the effect of disorder on activation energy of electr ical conductivity might yield a more empirical fit to the system that would take into consideration the two activation energy regions. Beam line studies of films could be used to determine the site occupied by the substituting lithium ions and whet her it is substituting for nickel or cobalt preferentially. Lithium is predicted to occupy the tetrahedral site, but further evidence could subs tantiate this claim. Additional studies to determine the stru ctural differences between solution and sputter deposited films may provi de further insight as to why the sputtered films are more conduc tive, but less transparent.

PAGE 173

154 Magnetic property studies of nickel-coba lt oxide would reveal the effects of heat treatment and doping on the magnetic nature of the thin films. Quantitative calibrated XPS depth profiles could be used to explain the bulk and surface film compositional differences and if there is indeed an effect from preferential sputtering of oxyge n from the argon sputtering beam. Transparency studies of combinator ial sputtered films at various gas compositions would produce data to better relate composition and growth conditions with optical properties. High temperature studies of the st ability of nickel-rhodium oxide on substrates other than silicon would show if a reaction at the film-substrate interface is affecting film conductivity and oxygen concentration. Learning more about the film structural properties and how th ey relate to the electrical and optical properti es will allow incremental improvements in film conductivity and transparency enabling it for use in future applications in IR devices. 7.3 Applications A wealth of applications including IR se nsors and actuators such as night vision displays, friend-foe identification, and local area networking, are in production, but would be greatly enhanced by the development of an IR transparent conducting material such as nickel-cobalt oxide. Additional devices such as organic IR emitting devices (OLEDs), longer wavelength fiber optics, higher power IR lasers, or solid-state optical temperature sensors could also be developed. The future is bright and with further development the world of the IR could be put to use.

PAGE 174

155 APPENDIX A DEPOSITION DETAILS A.1 Sputtering Figure A-1. A sputtering ta rget (red) mounted on a cathode (silver) with impinging gas ions (purple) and target pa rticles being removed (red). A.1.1 Definition Sputtering is a process of using high voltage to create a plasma (ionized gas) and bombard a surface (target) and cause it to “sputtering” off pa rticles (Mahajan & Harsha, 1999). This flux of target partic les coming off is collected as a thin film on a substrate of the user’s choosing such as a microscope slid e, fused-silica, sapphire, silicon wafers, or poly(ethelene terephthalate). Sputter coati ng is a well developed method used for thin film deposition and is widely used in the micr oelectronics and data storage manufacturing

PAGE 175

156 processes as well as food pro cessing for metallization of su ch item as food storage bags deposition although the process conditions vary gr eatly from one application to the next. For a more technical description, see (textbook on sputter deposition. A.1.2 Equipment Sputtering is done in a vacuum chamber. These chambers can be made of glass, aluminum or stainless steel as long as they can be evacuated to a sufficient background pressure to preclude the sputtering process from being contaminated by water vapor or carbon species normally found in the air. The eq uipment used in this study consisted of two chambers (Figure A-2) each of stai nless steel construction pumped by Varian diffusion pumps backed by oil rotary vane pumps. Figure A-2. Vacuum chambers used for sput tering deposition. (a ) E-beam deposition chamber located in room 1317 at the EMSL at PNNL (b) the M chamber located in RTL room 120 at PNNL. Both systems were used for sputter deposition for the films in this study. Cathode. A cathode is a fixture used to f eed voltage and a cooling medium (usually water) into the chamber and it holds th e target material. While they vary in size from one inch to several inches, the ones us ed for this study were two inches and three inches.

PAGE 176

157 Figure A-3. Three-inch sputte r cathode used to deposit thin films. Reprinted from US Inc. (2003) Mak™ Owner’s operation and main tenance manual of the mak sputtering sources. San Jose: US Inc. (Figure on p.7) with permission. Target. A target is a piece of material fabr icated for sputtering that is placed on the cathode. Targets used in this study were both alloy and oxide targets of two different sizes: 5 cm (Rh, Pd, NiRh4, NiRh2, NiO, Co3O4, Ni1.5Co1.5O4, Ni1.35Co1.35Li0.3O4, Ni1.2Co1.2Li0.6O4) 7.5 cm (Ni, Co, NiCo, NiCo2, NiCo3, Ni0.95Co1.95Li0.15O4) A.1.3 Configuration/Setup Combinatorial sputtering. Benefits of the combinatorial sputtering technique include adequate process control, fewer necessary targets, and generation of multiple compositions in a single run. Combinatoria l films provide ample data to compare and contrast compositionally variant properties such as conduc tivity and transparency from a

PAGE 177

158 single deposition run. Numbered Positions (9 divisions-one per inch on a 3 inch microscope slide) Cathodes may be tilted for more uniform coverage Less time to produce a range of films More data is produced per run Compositional variation requires fewer targets Easier to relate proper ties only to composition Fixed substrate geometry may limit film uniformity Mismatched cathodes require calibration Figure A-4. Combinatorial sputtering setup inside of chamber with mounted substrates prior to pump-down for sputtering. Single rotation. Single rotation is where the substrate holder spins on axis with the longitudinal axis of the cathode (Figure A-5). This setup was used for a few select films. Offset rotation. Offset rotation (Figure A-6) is like single rotation, but the axis of substrate holder is parallel, but not in line with th e longitudinal axis of the cathode. This setup was used for a majority of film depositions. Double planetary rotation. Double planetary rotation is like the earth orbiting the sun shown in Figure A-7. Double planetary rotation is used in many cases to obtain

PAGE 178

159 uniform thickness and morphology over larger ar eas not possible with single or offset rotation. Often the planetary setup requires a longer target to substrate distance to work effectively. Double planetary rotation was available in both chambers and was used in some cases. Figure A-5. Single rotation setup. Figure A-6. Offset rotation setup.

PAGE 179

160 Figure A-7. Double planetary ro tation setup. The target is blue, the substrate is greenblue, and the cathode is red. A.1.4 Target Production Custom target manufacturing incorporat ed techniques of solution preparation where metal nitrates in aqueous solution were boiled down and then heated to facilitate nitrate decomposition w ithout the organic additive. Th e produced cake was then ground and again heated to 350C to ensure complete dissociation of the nitrates resulting in finely divided powder. Pressi ng a green disk required a 6 cm die and a uniaxial press at

PAGE 180

161 15 ksi. Using a heating and cooling rate of 5C/min the green target was sintering at 900C for two hours yielding a ~5 cm sputtering target. Figure A-8. Pressed target produced from nitrate solu tions used for sputtering. A.2 Solution Deposition Figure A-9. The solution deposition method involves dropping a precursor solution on a spinning substrate prior to baking to form a film. Nitrate solutions require a combustion agent such as glycine or malonic acid, in aqueous based or alcohol based solutions, respectively, to ensure complete decomposition of the nitrates leaving an oxide film. Films are baked at 400C for 10-15 minutes after spinning. Solution deposition comprises of droppi ng an aqueous or alcohol based metal nitrate solution containing an organic com bustion agent on a spinning substrate. After spinning, the substrate is immediately transferred to a hotplate and cooked at approximately 400C for 10 minutes. The organic addi tive ignites and the nitrates

PAGE 181

162 dissociate leaving behind a thin metal oxi de film on the order of 50 nm thick. Combustion agents such as be glycine is us ed in aqueous soluti on for deposition on SiO2. Malonic acid is the combustion agent used in ethanol based soluti ons for deposition on Si. Solution deposition is simple because the solutions are simple to prepare, one solution can make many samples, and solutions are easily mixed in various compositions. Substrate preparation includes baking substr ates at >400C for up to 1 hour prior to deposition. The ability to reproduce high quality films proves very difficult however when attempting to maintain film consistency. Also, creating thick films requires multiple coating spins and uniformity degrades with each successive deposition.

PAGE 182

163 APPENDIX B CHARACTERIZATION METHODS AND DETAILS Electrical characterization included Hall and van der Pauw measurements, and temperature dependent van der Pauw measur ements. Optical spectral characterization included infrared transmission measuremen ts with Fourier transform infrared spectroscopy (FTIR) and ultra-violet and visible spectroscopy ( UVVIS). Structural analysis included optical and stylus profilome try to measure film thickness, transmission electron microscopy (TEM), and x-ray di ffraction (XRD). Secondary ion mass spectrometry (SIMS) and x-ray photoelect ron spectroscopy (XPS) were used to determine the chemical composition and bonding states in the film. B.1 Electrical Characterization B.2.1 Hall Effect The Hall effect system used had a 1 T magne t with a four-pin probe in the van der Pauw setup. The gold contacting pins were space d at 0.5 cm to the sample with a spring. The apparatus is enclosed in aluminum, is olated from light during the measurement. Figure B-1. Hall effect apparatus made by MMR technologies. It consists of sample dewer, magnet, measurement equipment and control computer.

PAGE 183

164 B.2.2 van der Pauw The van der Pauw setup uses the same e quipment as the Hall measurement, but does not require the magnet. Care was taken to ensure proper seating of the contacts with the film (see Figure B-1 and Figure B-2). Figure B-2. A van der Pauw resistivity pr obe setup and measurement schematic. Four probes are used to apply a current and measure an applied voltage. Eight measurements are collected using bot h a positive and negative current between each pin e.g. source: 3-4, meas ure: 1-2; source 41, measure 2-3; source 1-2, measure 3-4; etc. B.2 Optical Characterization B.3.1 FTIR Figure B-3. Nexus 570 FTIR with 3 detector s was used for high sensitivity mid IR (4000 cm-1 to 650 cm-1, mid IR (4000 cm-1 to 400 cm-1), and far IR (500 cm-1 to 100 cm-1) measurements.

PAGE 184

165 A Nicollet 210, a NEXUS 570 (Figure B-3), and a Nicollet 560 were all used to measure transmission of samples in the wave number regions between 4000 cm-1400 cm-1. The NEXUS 570 also has the capabil ity to measure the far IR from 500 cm-1 down to 150 cm-1. In each case, the chamber was allowed to equilibrate with a flowing nitrogen purge after admitting the sample into the holder. Purge times were typically at least one minute, but most often two minutes before both the background scan and the sample measurement. B.3.2 UVVIS. Visible and near IR was performed using the Carey 5 UVVis, shown in Figure B-3. The UVVis has a wavelength range of 190 nm-33 00 nm with the capability of measuring reflectance, and absorption/transmission in these ranges. The dual beam spectrometer offers the option of a zero and 100% transmi ssion spectra corrections to ensure an absolute zero reference and a ma ximum that will not exceed 100%. Figure B-4. Varian CAREY 5 dual beam ultra violet-visible spectrophotometer. B.4 Structural Characterization B.4.1 Profilometry A Tencor Alphastep 200 (Figur e B-5) was used to dete rmine film thickness of sputtered samples from a taped witness substr ate. The tape was removed and the stylus

PAGE 185

166 dragged across the film-substrate interface to gi ve a film thickness. Optical profilometry using a white light interferometer was used to verify results of the stylus profilometer. Results from both of these methods were co mpared with the average values provided by x-ray reflectivity measurements. Figure B-5. Stylus profilometer used to measure step height of the deposited film. Figure B-6. Optical profilometer software sc reen capture. Instrument made by Zygo.

PAGE 186

167 B.4.2 XRD The powder diffraction file is a database resource of both measured and calculated x-ray diffraction patterns for known substances used to compare with generated XRD. Table B-1 shows the relevant information included in the database entry. Table B-1. Powder diffracti on file number 73-1702 used as the standard of comparison for nickel cobalt oxide thin films a nd powders. Taken from the JCPDSInternational Center for Diffrac tion data. PCPDFWIN v. 2.2. 73-1702 Wavelength = 1.54060 d(A) Int h k l NiCo2O4 Nickel Cobalt Oxide 4.6846 132 1 1 1 2.867 298 2 2 0 2.4464 999* 3 1 1 2.3423 87 2 2 2 2.0285 200 4 0 0 Rad: CuK1 : 1.54060 d-sp: Calculated Cut off: 17.7 Int.: Calculated I/Icor.: 4.51 Ref: Calculated fro ICSD using POWD-12++, (1997) Ref: Knop, O., Reid, KIG., Su tarno, Nakagawa, Y., Can. J. Chem., 46, 3463 (1968) 1.8614 7 3 3 1 1.6562 71 4 2 2 1.5615 244 5 1 1 Sys.: Cubic S.G.: Fd3m (227) a: 8.114(14) Z: 8 Ref: Ibid. Dx: 5.982 ICSD#: 024211 1.4343 295 4 4 0 1.3715 11 5 3 1 1.3523 1 4 4 2 1.2829 19 6 2 0 1.2373 52 5 3 3 1.2232 28 6 2 2 1.1711 18 4 4 4 Peak height intensity. R -factor: 0.054. PSC: cF56. Structural reference: Knop, O., Reid, K.I.G., Sutarno, Nakagawa, Y., Can. J. Chem., 46, 3463 (1968). Mwt: 240.56. Volume[CD]: 534.20 1.1361 5 55 1 The following tables display the configuration data for the XRD system used in collecting the data for this st udy. The two setups used included x-ray reflectivity (XRR) (Table B-2), grazing incident x-ray diffraction (GIXRD) (Table B-3). XRR was used to gather information on the thickness of the produced thin films. GIXRD was used instead of the powder diffraction setup because the thickness of the films did not provide a suitable signal to noise ratio for peak analys is to provide lattice analysis such as the crystallite size or the lattice parameter.

PAGE 187

168 Table B-2. X-ray reflectivity (XRR) e xperimental system setup parameters. Apparatus Diffractometer Philips X’Pert MPD System (PW3040/00 type) X-ray Source Sealed Ceramic Tube Anode Fixed, Long-Fine-Focus, Cu Wavelength, K1, 1.5406 Voltage 45 kV Current 40 mA Power 1.8 kW Goniometer (Vertical, 220 mm radius)Sample Platform Philips Multi-Purpose Sample Stage (PW1821 type) Sample Support Glass Optical Flat Scan Geometry Symmetric Incident Optics Beam Conditioner Gbel Mirror (Paralle l Beam, Equatorial Divergence <0.04) Slit 1/32 (Fixed) Soller Slit None Axial Beam Mask 10 mm Receiving Optics Anti-scatter Slit 0.1 mm (Fixed) Receiving Slit 0.1 mm (Fixed) Soller Slit 0.04 radians Monochromator Curved Graphite Detector Xe-filled Proportional Counter Table B-3. Grazing-inciden ce x-ray diffraction (GIXRD) ex perimental summary system setup parameters. Apparatus Diffractometer Philips X’Pert MPD System (PW3040/00 type) X-ray Source Sealed Ceramic Tube Anode Fixed, Long-Fine-Focus, Cu Wavelength, K1, 1.5406 Voltage 45 kV Current 40 mA Power 1.8 kW Goniometer (Vertical, 220 mm radius)Sample Platform Philips Multi-Purpose Sample Stage (PW1821 type) Sample Support Glass Optical Flat Scan Geometry Asymmetric

PAGE 188

169 continued continued Incident Optics Beam Conditioner Gbel Mirror (Paralle l Beam, Equatorial Divergence <0.04) Slit 1/2 (Fixed) Soller Slit None Axial Beam Mask 20 mm Receiving Optics Collimator Parallel Plate, 0.27 radians Receiving Slit None Soller Slit None Monochromator Flat Graphite Detector Xe-filled Proportional Counter B.4.3 TEM / STEM Cross sectional high resolu tion scanning transmission electron microscopy was done on selected ~0.5 um thick films deposited from alloy target reactive RF sputtering. Cross-sectional specimens were prepared by gluing two samples film-to-film and then cutting vertical sections which were first m echanically thinned from the both side to a thickness of ~ 20 m. Final thinning to el ectron transparency was accomplished by ion milling using a 5 keV Ar+ ion beam incident at 12 XTEM analyses was carried our in a JEOL 2010F instrument operated at 200 keV using selected area electron diffraction (SAED) and bright-field (BF), dark-field (DF), high-angle annular dark field (HAADF), EDS analysis, and lattice re solution imaging techniques. B.5 Chemical Characterization B.5.1 SIMS SIMS mass surveys and depth profiles we re acquired with a Perkin-Elmer PHI 660 SIMS system using a 6keV oxygen primary ion beam with positive secondary acquisition. The current intensity was set at 108nA. During mass survey and depth profile acquisition, the raster size was varied between 200x200 m2, 300x300 m2,

PAGE 189

170 350x350 m2 and 400x400 m2 with 65% gating. The neut ralizer was used during the analysis for charge compensation. Quantitative XPS was use to calculate a SIMS sensitivity factor assuming ionization probabi lities for the elements being analyzed. Once the sensitivity values were determined, the raw data was adjusted accordingly and reported. B.5.2 UPS Using the He I line of a helium lamp with a photon energy of 21.218 eV, an XPS analysis system was used to acquire th e UPS spectrum at a vacuum level of ~1x10-9 Torr B.5.3 XPS A k x-ray beam was used with a pass energy of 45 eV in multiplex mode with a 5 KeV argon sputtering source energy. Wide and narrow scan data collection. The XPS measurements were made using a Physical Electronics Quantum 2000 Scanni ng ESCA Microprobe. This system uses a focused monochromatic Al Ka x-rays (1486. 7 eV) source for excitation and a spherical section analyzer. The instrument has a 16 element multichannel detection system. The X-ray beam used was a 100 W, 107 um beam ra stered over a 1.4 mm x 0.2 mm area. The x-ray beam is incident normal to the sample and the x-ray detector is at 45 away from the normal. The survey scans were collected using a pass energy of 117.4 eV. For the Ag 3d5/2 these conditions produce a FWHM of better than 1.6 eV. The high energy resolution data was collect ed using a pass energy of 23.5 eV. For the Ag 3d5/2 these conditions produce a FWHM of better than 0.74 eV. The collected data were referenced to an energy scale with binding energies fo r Cu 2p 3/2 at 932.62 0.05 eV and Au 4f at 83.96.0 0.05 eV. Analyzer type: Spherical Section Analyzer

PAGE 190

171 Analyzer mode: Constant Pass Energy or FAT Detector: Multichannel resistive plate Emission angle: 45 degrees Incident angle: 0 degrees Source to analyzer angle: 45 degrees Analyzer angular acceptance width: 20 degrees Depth profile data collection. The XPS measurements were made using a Physical Electronics Quantum 2000 Scanning ESCA Microprobe. This system uses a focused monochromatic Al Ka x-rays (1486. 7 eV) source for excitation and a spherical section analyzer. The instrument has a 16 element multichannel detection system. The X-ray beam used was a 40, 200 um beam spot. The x-ray beam is incident normal to the sample and the x-ray detector is at 45 aw ay from the normal. The profile data was collected using a pass energy of 58.7 eV. For the Ag 3d5/2 these conditions produce a FWHM of better than 0.87 eV. The collected data were referenced to an energy scale with binding energies for Cu 2p 3/2 at 932.62 0.05 eV and Au 4f at 83.96.0 0.05 eV. Ion gun: 2kV Ar+ ions Ion beam raster area 5 mm x 3 mm Spot size: 150 um minimum beam diameter Sputter source incident angle: 45 degrees Sputter source polar angle 60 degrees Sputter source azimuthal angle: 90 degrees

PAGE 191

172 APPENDIX C CALCULATIONS, EQUATIONS, AND EFFECTS C.1 Unit Cell Specifics Taking 8.114 as the theoretical lattice parameter, the unit cell consists of 24 cations (8 tetrahedral, 16 octahedral) and 32 oxygen anions for a total of 56. Table C-1. Unit cell sites and weights for density calculation. Ni Co O Octahedral site 8 8 Tetrahedral site 8 Oxygen site 32 Atomic wt. 59.8 58.9 15.9997 Wt. per cell 478.4 3534 7680 Valence state 3+, 2+ 3+, 2+ 2-, 1Ionic rad (2+, tet) 69 pm 73 pm 124 pm Ionic rad (2+, oct) 83 pm 79 pm Ionic rad (2+, oct, h.spin) 88.5 pm Ionic rad (3+) 70 pm 68.5 pm Ionic rad (3+, h.spin) 74 pm 75 pm Using the total number of sites in the cell and multiplying the weight of each atom give the grams per mol per cell. Dividing by Avogadro’s number give s the weight of one cell (Equation C.1). 59.8gm Ni 8Ni Cel l 58.9gm Co 16Co Cel l 16gm O 32O cel l 3.195748910 21gm cel l (C.1) Taking the lattice parameter of the un it cell, cubing it and converting from angstroms to centimeters gives the volume occ upied per cell, the reciprocal of which is the number of cells per cubic centi meter, shown in Equation C.2. 8.114 cell 10 8cm 1 3 5.34201 10 22cm3cell 1.87195 1021cell cm3 (C.2)

PAGE 192

173 C.2 Theoretical Film Density Multiplying the number of cells per volume and the grams per cell gives a theoretical density in grams pe r cubic centimeter (Equation C.3). 1.871951021 3.1961021 5.98 gm cm3 (C.3) If cells were stacked in cylindrical colu mns and packed hexagonally (Figure C-1), then the maximum percent density would be ~90.6%. Figure C-1. Hexagonally packed cyli nders in 3-D on film and top view. If the packing of the cylindrical column s changed to a cubic arrangement (Figure C-2) then the density drops to 78.5% of the theoretical 100% dense matrix. Figure C-2. Cubic packed cylinders top view and in 3-D as would be seen in the film.

PAGE 193

174 C.3 Heike’s Rule The formula used for determining the car rier concentration from the Seebeck measurement is called Heike’s Rule. Alpha is the Seebeck coefficient () is the slope of the change in voltage (V) measured with the change in temperature (T in Equation C.4). V T (C.4) Heike’s rule is the proportional relationship of the Seebeck coefficient to the density of carriers in the material (R ao & Raveau, 1998, Tareen et al., 1984). c c e k 1 ln (C.5) The letter c represents the carrier concentra tion ratio of the number of active carriers over the number of carriers available. C.4 Absorption Coefficient Calculation. The procedure for calculating absorption coe fficient involves using a transparency measurement system similar to the UVVis system mentioned in Appendix B. The first step is to measure the percent transmission as a function of wavelength. Using Equation C.6, and assuming that absorption is 0 and s cattering is zero, the only other effect that needs to be considered is the reflection. Th e index of refraction of each material present that creates and interf ace the light must pass through is required to accurately calculate the percent of light reflected at the in terfaces and added to the transmission. I transmission I 0 I absorption I reflection (C.6) Percent transmission is the ratio of transm itted light and incident light so divide equation C.6 by I0.

PAGE 194

175 0 0 01 I I I I I Ireflection absorption on transmissi (C.7) Assume Reflection can be calculated from the refractive indice s of both materials 2 2 1 2 1 0100 % 100 n n n n R I Ireflection (C.8) The index of refraction for air is 1 and the spin el is 2.6, ~25% is lost at the film-air interface. The index of refraction of the fuse d silica is 1.46 resulting in ~ 4% loss from the film-substrate interface, and so add 29% to all measured values and the %T lost to absorption Convert the reflection corrected transmi ssion value to absorbance by subtracting sum of %R value and %T value from 100. Dividing by 100 you get absorbance which is found in Equation C.10. d absorptione I I 0 (C.9) Solving for the absorption coefficient, yields get Equation C.10. Plug in the absorption figure calculated above for Iabsorption/I0 and the film thickness to yield the absorption coefficient. 1 d Ln IabsorptionI0 (C.10) C.5 Jahn-Teller Effect The Jahn-Teller effect is a term used to describe a system going to a lower energy state from a state of degener acy by shifting the degenerate or bitals slightly due to an atomistic distortion. For example a cubic syst em may become slightly tetragonal to allow a stepped energy level of the t2g orbitals rather than a triple degeneracy (Dionne, 1990).

PAGE 195

176 Figure C-3. Degenerate energy level orbitals in the normal spinel state and schematic of orbital energy shift due to the Jahn-Teller effect. C.6 Verway Transition Charge ordering of cations was postulated by C.J. Verway in 1939 and has been used as an argument to describe the behavior of Fe3O4 and its multiple cation states. At a temperature of X (400K for Fe3O4) the Verway or order-disorder transition occurs and is believed to be an ordering of the cation charge s. This charge orderi ng alters the activation energy of electrical conductivity. C.7 Nickel-Cobalt Oxide Activation Energy Curves Using Equation C.11, when rearranged as C.12 will match the form given in Equation C.13 that is easily graphed. Th e ordinate axis (y) corresponds with ln( ), the abscissa (x) corresponds with 1/T, m is the sl ope and is the value from the graph (Ea/k). When multiplied by the Boltzman constant of 8.6175 eVK-1 gives Ea. The remaining value is the y intercept (b) corresponding to the preexponentia l term of the conductivity equation. kT E kT Ea norm norm act aexp or exp0 (C.11) ) ( ) (0 Ln kT E Lna (C.12) y mx b (C.13)

PAGE 196

177 The slopes from Figure C-4 were used as m in this case to find the value of the activation energy (Ea) using Equation C.14 rearranged to give Equation C.15. The Boltzman constant in eVK-1 is shown with 1/T being the variable x in Equation C.14 and Equation C.15. ) ( 1 10 6175 8 ) (5K T K eV eV E kT Ea a (C.14) 1000 10 6175 85 K eV m Ea (C.15) Figure C-4. Activation energies are the slope taken from th e curve fits to the heating traces Table C-2. Activation energies of NixCo3-xO4 films from Figure C-4. All values are in millivolts taken from slopes of curves in Figure C-4. NixCo3-xO4 Low Up High Up High Down Low Down x=1 14.75 41.27 76.02 35.03 x=1.5 8.07 41.2 76.02 35.03

PAGE 197

178 C.8 Lithium in Nickel-Cobalt Oxide Activation Energy Curves Table C-3. Activati on energies of NiCo2-zLizO4 films from solution deposition. All values are in millivolts taken from slopes of curves in Figure C-5. NiCo2-zLizO4 z=0.00 z=0.01 z=0.03z=0.05z=0.1 z=0.3 z=0.5 Ea Low Up 11.7 19.6 19.4 19.4 13.2 15.2 6.17 Ea Low Down 35.2 32.2 35.3 35.3 30.7 41.4 62.5 Ea High Up 47.5 58.5 59.3 66.9 56.2 80.6 130 Ea High Down 58.9 58.9 59.3 66.9 63.9 88.2 140 Figure C-5. Measured conductivity as a function of temperature for NiCo2-zLizO4 films from solution. The highes t conductivity is from the sample with the lowest concentration of lithium. Table C-4. Activation energies of Ni0.75Co2.25-zLizO4 films from solution deposition. All values are in millivolts taken from slopes of curves in Figure C-6. Z=0.00 Z=0.05 Z=0.1 Z=0.25 Z=0. 5 Ea Low Up 11.71 9.91 7.505 35.20 13.08 Ea High Up 47.50 52.11 53.67 70.41 115.65 Ea High Down 68.59 56.88 56.76 76.12 114.22 Ea Low Down 12.38 28.43 27.52 36.87 54.20

PAGE 198

179 Figure C-6. Conductivity is temperature dependent for Ni0.75Co2.25-zLizO4 from solution. Lithium in increasing amounts substitutes for cobalt. All samples were heated and cooled at 15/minute. C.9 Heat Treatment Quench Model If the sample substrate of fused silica is 3 mm thick and 2.54 cm in diameter, the effective thermal mass is low. Thermal contact quality between the substrate surface and the cooling block surface were subject to uncertainty and beli eved to be the reason for the differences in cooling between the aluminum a nd brass. Brass in cool water was a faster quench because the brass sink was below room temperature when the quench began increasing the cooling rate. Using the th ermal conductivity of fu sed silica, the time required to remove the amount of thermal en ergy stored in the substrate by its heat capacity times its thermal mass wa s ~20 s. Cooling from ~400 C to room temperature at 25 C was 375 C in 20 s. The effective maximum cooling rate is ~15 C/s or ~1200 C per minute.

PAGE 199

180 LIST OF REFERENCES Appandairajan, N.K., Viswanathan, B., & G opalakrishnan, J. (1981). Lithium-substituted cobalt oxide spinels LixM1-xCo2O4 (M = Co2+, Zn2+, 0 < or = x < or = 0.4). Journal of Solid State Chemistry, 40 (1), 117-121. Austin, I.G., Mott, N.F. (1969). Polarons in crystalline and non-crystalline materials. Advances in Physics 18 41-102. Azaroff, L.V. (1960). Introduction to solids New York: McGraw-Hill. Ballal, M.M., & Mande, C. (1977). Chemical shifts of copper and cobalt k absorption discontinuities in the spinels CuCr2X4 (X = 0, S, Se, Te), CoCr2X4 (X = O, S) and Cu0.5Co0.5Cr2-xRhxS4 (x=0, 1, 2). Journal of the Physics and Chemistry of Solids, 38 (8), 843-848. Banov, B., Bourilkov, J., & Ml adenov, M. (1995). Cobalt st abilized layered lithiumnickel oxides, cathodes in lithium rechargeable cells. Journal of Power Sources, 54 (2), 268-270. Belova, I.D., Roginskaya, Y.E., Shifrina, R.R., Gagarin, S.G., Plekhanov, Y.V., & Venevtsev, Y.N. (1983). Co (III) ions high-spin configuration in nonstoichiometric Co3O4 films. Solid State Communications, 47 (8), 577-584. Benco, L., Barras, J.-L., Atanasov, M., Daul C., & Deiss, E. (1999). First principles calculation of electrode material for lithium intercalation batteries: TiS2 and LiTi2S4 cubic spinel structures. Journal of Solid State Chemistry, 145 (2), 503-510. Benqlilou-Moudden, H., Blondiaux, G., Vinatier, P., & Levasseur, A. (1998). Amorphous lithium cobalt and nickel oxides thin films: Preparation and characterization by RBS and PIGE. Thin Solid Films, 333 (1-2), 16-19. Bergh, A., Craford, G., Duggal, A., & Haitz R. (2001). The promise and challenge of solid-state lighting. Physics Today, 54 (12), 42-47. Biju, V., & Khadar, M.A. (2001) DC conductivity of consolidat ed nanoparticles of NiO. Materials Research Bulletin, 36 (1-2), 21-33.

PAGE 200

181 Blasse, G. (1963). New type of superexchange in the spinel stru cture some magnetic properties of oxides Me2+Co2O4 and Me2+Rh2O4 with spinel structure. Philips Research Reports, 18 383-392. Brundle, C.R., Evans, C.A.J., & Wilson, S. (1992). Encyclopedia of materials characterization: Surfaces, interfaces, thin films Boston: ButterworthHeinemann. Callister, W.D. (1997). Materials science and engi neering: An introduction (Fourth ed.). New York: Wiley. Campbell, S.A. (1996). The science and engineering of microelectronic fabrication New York: Oxford University Press. Carewska, M., Scaccia, S., Croce, F., Ar umugam, S., Wang, Y., & Greenbaum, S. (1997). Electrical conductivity and 6,7Li NMR studies of Li1+yCoO2. Solid State Ionics, Diffusion & Reactions, 93 (3-4), 227-237. Carey, M.J., Spada, F.E., Berkowitz, A.E., Cao, W., & Thomas, G. (1991). Preparation and structural characterization of sputtered CoO, NiO, and Ni0.5Co0.5O thin epitaxial-films. Journal of Materials Research, 6 (12), 2680-2687. Cheng, C.-S., Serizawa, M., Sakata, H., & Hi rayama, T. (1998). El ectrical conductivity of Co3O4 films prepared by chemical vapour deposition. Materials Chemistry and Physics, 53 (3), 225-230. Chopra, K.L., Major, S., & Pandya, K. (1983). Transparent conductors a status review. Thin Solid Films, 102 1-46. Cox, P.A. (1987). The electronic structure and chemistry of solids Oxford [Oxfordshire]; New York: Oxford University Press. da Silva Pereira, M.I., da Costa, F.M.A ., & Tavares, A.C. (1994). Electrochemical behaviour of NiCo2-xRhxO4 spinel system. Electrochimica Acta, 39 (11-12), 15711578. Deer, W.A. (1962). Rock-forming minerals London: Longmans. Dionne, G.F. (1990). Spin states and electronic conduction in Ni oxides. Journal of Applied Physics Duan, N., Sleight, A.W., Jayaraj, M.K., & Ta te, J. (2000). Transpar ent p-type conducting CuSCo2+x films. Applied Physics Letters, 77 (9), 1325-1326. El-Farh, L., Massot, M., Lemal, M., Julien, C., Chitra, S., Kalyani, P., Mohan, T., & Gangadharan, R. (1999). Physical propert ies and electrochemical features of lithium nickel-cobalt oxide cathode materials prepared at moderate temperature. Journal of Electroceramics, 3 (4), 425-432.

PAGE 201

182 Emin, D. (1977). The sign of the ha ll effect in hopping conduction. Philosophical Magazine, 35, 1189-1198. Emin, D. (1980). The hall effect in hopping c onduction. In C.L. Chien & C.R. Westgate (Eds.), Hall effect and its applications. Proceedings of the commemorative symposium, 13 Nov. 1979 (pp. 281-298). Baltimore, MD, USA: Plenum. Emin, D. (1982). Small polarons. Physics Today, 35 (6), 34-40. Emin, D. (1994). Disorder effects on small-polaron formation and hopping. International Journal of Modern Physics B, 8 (7), 819-827. Emin, D., & Bussac, M.-N. (1994). Disorder-induced small-polaron formation. Physical Review B (Condensed Matter), 49 (20), 14290-14300. Fleischer, M. (1995). Glossary of mineral species 1995 (7th ed.). Tucson, AZ: Mineralogical Record. Forrest, S., Burrows, P., & Thompson, M. (2000). The dawn of organic electronics. IEEE Spectrum, 37 (8), 29-34. Fransson, L., Nordstrom, E., Edstrom, K., Haggstrom, L., Vaughey, J.T., & Thackeray, M.M. (2002). Structural tr ansformations in lithiated '-Cu6Sn5 electrodes probed by in situ Mossbauer spectr oscopy and x-ray diffraction. Journal of the Electrochemical Society, 149 (6), 736-742. Freeman, A.J., Poeppelmeier, K.R., Mason, T. O., Chang, R.P.H., & Marks, T.J. (2000). Chemical and thin-film strategies fo r new transparent conducting oxides. MRS Bulletin, 25 (8), 45-51. Fujii, E., Tomozawa, A., Torii, H., & Takayama R. (1996). Preferred orientations of NiO films prepared by plasma-enhanced meta lorganic chemical vapor deposition. Japanese Journal of App lied Physics Part 2, 35 (3A), L328-L330. Fukui, T., Ohara, S., Okawa, H., Hotta, T ., & Naito, M. (2000). Properties of NiO cathode coated with lithiated Co a nd Ni solid solution oxide for MCFS. Journal of Power Sources, 86 (1-2), 340-346. Galtayries, A., & Grimblot, J. (1999). Forma tion and electronic properties of oxide and sulphide films of Co, Ni and Mo studied by XPS. Journal of Electron Spectroscopy and Related Phenomena, 99 267-275. Ganguly, P., Venkatraman, T.N., Rajamohanan, P.R., & Ganapathy, S. (1997). Evidence for multiple M sites in AMO2 compounds: 59Co solid state NMR studies on LiCoO2. Journal of Physical Chemistry B, 101 (51), 11099-11105. Gem by gem (2003). Retrieved June 10, 2003, from http://www.gemstone.org/gem-bygem/english/spinel.html

PAGE 202

183 Gendron, F., Castro-Garcia, S., Popova, E., Zi olkiewicz, S., Soulette, F., & Julien, C. (2003). Magnetic and electron ic properties of lithium cobalt oxide substituted by nickel. Solid State Ionics, Diffusion & Reactions, 157 125-132. Giesbers, J.B., Prins, M.W.J., Cillessen, J.F.M., & vanEsch, H.A. (1997). Dry etching of all-oxide transparent thin film memory transistors. Microelectronic Engineering, 35 (1-4), 71-74. Gonzalez-Elipe, A.R., Alvarez, R., Holgado, J.P., Espinos, J.P., Munuera, G., & Sanz, J.M. (1991). XPS study of the Ar+-induced reduction of Ni2+ in NiO and Ni-Si oxide systems. Applied Surface Science (1985), 51 (1-2), 19-26. Goodwin-Johansson, S., Holloway, P.H., McGuire, G., Buckley, L., Cozzens, R., Schwartz, R., & Exarhos, G. (2000). Artific ial eyelid for protection of optical sensors. Smart Structures and Materials 2000: Electroactive Polymer Actuators and Devices (EAPAD), (3987), 225-231. Gordon, R.G. (2000). Criteria for choosing transparent conductors. MRS Bulletin, 25 (8), 52-57. Gover, R.K.B., Yonemura, M., Hirano, A ., Kanno, R., Kawamoto, Y., Murphy, C., Mitchell, B.J., & Richardson, J.W., Jr. ( 1999). The control of nonstoichiometry in lithium nickel-cobalt oxides. Journal of Power Sources, 81-82 535-541. Haenen, J., Visscher, W., & Barendrecht, E. (1986). Characterization of NiCo2O4 electrodes for O-2 evolution .2. Nonelectrochem ical characterization of NiCo2O4 electrodes. Journal of Electroanalytical Chemistry, 208 (2), 297-321. Han, K.-S., Tsurimoto, S., & Yoshimura, M. (1999). Fabrication te mperature and applied current density effects on the direct fabrication of lithium nickel oxide thin-film electrodes in LiOH solution by the el ectrochemical-hydrothermal method. Solid State Ionics, Diffusion & Reactions, 121 (1-4), 229-233. He, J., Lindstrom, H., Hagfeldt, A., & Lindquist, S.-E. (1999). Dye-sensitized nanostructured p-type nickel oxide f ilm as a photocathode for a solar cell. Journal of Physical Chemistry B, 103 (42), 8940-8943. Holloway, P.H., & Nelson, G.C. (1979) Preferential sputtering of Ta2O5 by argon ions. Journal of Vacuum Science & Techno logy A-Vacuum Surfaces and Films, 16 (2), 793-796. Hosono, H., Yasukawa, M., & Kawazoe, H. (1996). Novel oxide amorphous semiconductors: Transparent conducting amorphous oxides. Journal of NonCrystalline Solids, 203 334-344.

PAGE 203

184 Hotovy, I., Buc, D., Hascik, S., & Nennewitz O. (1998a). Character ization of NiO thin films deposited by reactive sputtering. Vacuum Electronic Properties of Metal/Non-Metal Microsystems., 50 (1-2), 41-44. Hotovy, I., Huran, J., Janik, J., & Kob zev, A.P. (1998b). Deposition and properties of nickel oxide films produced by DC reactive magnetron sputtering. Vacuum, 51 (2), 157-160. Hu, C.-C., & Cheng, C.-Y. (2002). Ideally pseudocapacitive behavior of amorphous hydrous cobalt-nickel oxide pr epared by anodic deposition. Electrochemical and Solid-State Letters, 5 (3), 43-46. Hughes, R.A.S., K.E. (1999). Spinel com pounds: structure and property relations. Journal of the American Ceramic Society, 82 (12), 3277–3278. Hummel, R.E. (2001). Electronic properties of materials (3rd ed.). New York: Springer. Iida, A., & Nishikawa, R. (1994). A thin-film of an Ni-NiO heterogeneous system for an optical-recording medium. Japanese Journal of App lied Physics Part 1, 33 (7A), 3952-3959. Julien, C. (2000). Local cationic environment in lithium nickel-cobalt oxides used as cathode materials for lithium batteries. Solid State Ionics, 136-137 887-896. Julien, C., El-Farh, L., Rangan, S., & Massot, M. (1999). Studies of LiNi0.6Co0.4O2 cathode material prepared by the citric acid-assisted sol-gel method for lithium batteries. Journal of Sol-Gel Sc ience and Technology, 15 (1), 63-72. Kawazoe, H., Yanagi, H., Ueda, K., & Hosono, H. (2000). Transparent p-type conducting oxides: design and fabricati on of p-n heterojunctions. MRS Bulletin, 25 (8), 28-36. Kawazoe, H., Yasukawa, M., Hyodo, H., Kur ita, M., Yanagi, H., & Hosono, H. (1997). P-type electrical conduc tion in transparent thin films of CuAlO2. Nature, 389 (6654), 939-942. Kennedy, R.J. (1996). The growth of iron oxide, nickel oxide and cobalt oxide thin films by laser ablation from metal targets. IEEE transactions on magnetics 31 (6), 38293831. Kim, D., Kim, M.-K., Son, J.-T., & Kim, H. -G. (2002). Effect of target properties on deposition of lithium nickel cobalt oxide thin-films using RF magnetron sputtering. Journal of Power Sources, 108 (1-2), 239-244. Kim, J.-G., Pugmire, D.L., Battaglia, D., & Langell, M.A. (2000). Analysis of the NiCo2O4 spinel surface with auger and x-ray photoelectron spectroscopy. Applied Surface Science, 165 (1), 70-84.

PAGE 204

185 King, W.J., & Tseung, A.C.C. (1974). Reduction of oxygen on nickel-cobalt oxides .2. Correlation between crystalstructure and activity of Co2NiO4 and related oxides. Electrochimica Acta, 19 (8), 493-498. Kingery, W.D., Bowin, H.K ., & Uhlmann, D.R. (1975). Introduction to ceramics (2nd Ed.). New York: John Wiley & Sons. Kishimoto, Y., Ohno, T., Nagata, S., Matsum oto, N., Kanashiro, T., Michihiro, Y., & Nakamura, K. (2000). NMR study of Cu(Ir0.95Rh0.05)2S4. Physica B, 281-282 634635. Kitao, M., Izawa, K., Urabe, K., Komats u, T., Kuwano, S., & Yamada, S. (1994). Preparation and electrochromic properties of RF-sputtered NiOx films prepared in Ar/O2/H2 atmosphere. Japanese Journal of Applied Physics Part 1, 33 (12A), 6656-6662. Koumoto, K., & Yanagida, H. (1981). Elec trical conduction in pure and Li-substituted Co3O4. Communications of the American Society C-156 C-157. Kuk, S.T., Song, Y.S., & Kim, K. (1999). Pr operties of a new type of cathode for molten carbonate fuel cells. Journal of Power Sources, 83 (1-2), 50-56. Kuk, S.T., Song, Y.S., Suh, S., Kim, J.Y ., & Kim, K. (2001). The formation of LiCoO2 on a NiO cathode for a molten carbonat e fuel cell using electroplating. Journal of Materials Chemistry, 11 (2), 630-635. Kumagai, H., Matsumoto, M., Toyoda, K., & Obara, M. (1996). Preparation and characteristics of nickel oxide thin f ilm by controlled growth with sequential surface chemical reactions. Journal of Materials Science Letters, 15 (12), 10811083. Lewis, B.G., & Paine, D.C. (2000). Applic ations and processing of transparent conducting oxides. MRS Bulletin, 25 (8), 22-27. Liu, Z.L., Yu, A.S., & Lee, J.Y. (1999). Synthesis and characterization of LiNi 1xyCoxMnyO2 as the cathode materials of secondary lithium batteries. Journal of Power Sources, 81-82, 416-419. Lunkenheimer, P., Loidl, A., Ottermann, C. R., & Bange, K. (15). Correlated barrier hopping in NiO films. Physical Review B (Condensed Matter), 44 (11), 59275930. Mahajan, S., & Harsha, K.S.S. (1998). Principles of growth and processing of semiconductors New York: WCB/McGraw-Hill.

PAGE 205

186 Marco, J.F., Gancedo, J.R., Gracia, M., Gau tier, J.L., Rios, E., & Berry, F.J. (2000). Characterization of the nickel cobaltite, NiCo2O4, prepared by several methods: an XRD, XANES, EXAFS, and XPS study. Journal of Solid State Chemistry, 153 (1), 74-81. Mattox, D.M. (1998). Handbook of physical vapor depos ition (PVD) processing : film formation, adhesion, surface pre paration and contamination control Westwood, N.J.: Noyes Publications. Minami, T. (2000). New n-type transparent conducting oxides. MRS Bulletin, 25 (8), 3844. Minami, T., Sonohara, H., Kakumu, T., & Taka ta, S. (1995). Physics of very thin ITO conducting films with high transparency prepared by DC magnetron sputtering. Thin Solid Films, 270 (1-2), 37-42. Minami, T., Yamamoto, T., Toda, Y., & Miya ta, T. (2000). Transparent conducting zincco-doped ITO films prepared by magnetron sputtering. Thin Solid Films, 373 (12), 189-194. Monk, P.M.S., & Ayub, S. (1997). Solid-state properties of thin film electrochromic cobalt-nickel oxide. Solid State Ionics, Diffusion & Reactions, 99 (1-2), 115-124. Morin, F.J. (1954). Electrical properties of NiO. Physical Review, 93 (6), 1199-1204. Moshtev, R., Zhatilova, P., Bakalova, I ., & Vassilev, S. (2002). Synthesis, XRD characterization, and cycling performa nce of cobalt doped lithium nickelates. Journal of Power Sources, 112 (1), 30-35. Moshtev, R., Zlatilova, P., Manev, V., & Tagawa, K. (1996). Synthesis of LiNiO2 in air atmosphere: x-ray diffraction characterizat ion and electrochemical investigation. Journal of Power Sources, 62 (1), 59-66. Moshtev, R., Zlatilova, P., Vasilev, S., Ba kalova, I., & Kozawa, A. (1999). Synthesis, XRD characterization and electrochemical performance of overlithiated LiNiO2. Journal of Power Sources, 81-82 434-441. O'Hanlon, J.F. (1989). A user's guide to vacuum technology (2nd ed.). New York: Wiley. Ohmuro, Y., Ohachi, T., & Taniguchi I. (1995). Ionic conduction of Ag+ and Cu+ ions in vacancy defect spinel compounds (MIn5S8)1-x(In2S3)x (M=Cu or Ag, x<1) in microwave regime. Solid State Ionics, Diffusion & Reactions, 79 53-59. Ohya, Y., Koyama, H., Ban, T., & Takahashi, Y. (1998). Electrical properties of p-n contact with oxide semiconductor thin f ilms fabricated by liquid phase method. Materials Science and Engineering B-So lid State Materials For Advanced Technology, 54 (1-2), 55-59.

PAGE 206

187Park, S., Keszler, D.A., Valencia, M.M., Hoffman, R.L., Bender, J.P., & Wager, J.F.(2002). Transparent p-type conducting bacu/sub 2/s/sub 2/ films. Applied PhysicsLetters, 80(23), 4393-4394.Parravano, G. (1955). Thermoelectric behavior of nickel oxide. The Journal of ChemicalPhysics, 23(1), 5-10.Patil, P.S., Kadam, L.D., & Lokhande, C.D. (1996). Preparation and characterization ofspray pyrolysed cobalt oxide thin films. Thin Solid Films, 272(1), 29-32.Pejova, B., Isahi, A., Najdoski, M., & Grozdanov, I. (2001). Fabrication andcharacterization of nanocrystalline cobalt oxide thin films. Materials ResearchBulletin, 36(1-2), 161-170.Petrov, K., & Will, G. (1987). A new cobalt-nickel oxide spinel prepared under highpressure in an oxygen atmosphere. Journal of Materials Science Letters, 6(10),1153-1155.Polo da Fonseca, C.N., Davalos, J., Kleinke, M., Fantini, M.C.A., & Gorenstein, A.(1999). Studies of LiCoOx thin film cathodes produced by rf Sputtering. Journalof Power Sources, 81-82, 575-580.Puspharajah, P., Radhakrishna, S., & Arof, A.K. (1997). Transparent conducting lithium-doped nickel oxide thin films by spray pyrolysis technique. Journal of MaterialsScience, 32(11), 3001-3006.Rao, C.N.R., & Raveau, B. (1998). Transition metal oxides : structure, properties, andsynthesis of ceramic oxides (2nd ed.). New York: Wiley-VCH.Robertson, A.D., Trevino, L., Tukamoto, H., & Irvine, J.T.S. (1999). New inorganicspinel oxides for use as negative electrode materials in future lithium-ionbatteries. Journal of Power Sources, 81-82, 352-357.Roginskaya, Y.E., Morozova, O.V., Lubnin, E.N., Ulitina, Y.E., Lopukhova, G.V., &Trasatti, S. (1997). Characterization of bulk and surface composition ofCoxNi1xOy mixed oxides for electrocatalysis. Langmuir, 13(17), 4621-4627.Sanz, J.M., Soriano, L., Prieto, P., Tyuliev, G., Morant, C., & Elizalde, E. (1998).Electronic structure and chemical characterization of ultrathin insulating films.Thin Solid Films, 332(1-2), 209-214.Sato, H., Minami, T., Takata, S., & Yamada, T. (1993). Transparent conducting p-typeNiO thin-films prepared by magnetron sputtering. Thin Solid Films, 236(1-2), 27-31.Schellenschlager, V., & Lutz, H.D. (2000). Raman spectroscopic studies ondecomposition of spinel-type cobalt chromium indium sulfide solid solutions.Materials Research Bulletin, 35(4), 567-574.

PAGE 207

188Schumacher, L.C., Holzhueter, I.B., Hill, I.R., & Dignam, M.J. (1990). Semiconductingand electrocatalytic properties of sputtered cobalt oxide films. ElectrochimicaActa, 35(6), 975-984.Seguin, L., Amatucci, G., Anne, M., Chabre, Y., Strobel, P., Tarascon, J.M., & Vaughan,G. (1999). Structural study of NiO2 and CoO2 as end members of the lithiatedcompounds by in situ high resolution X-ray powder diffraction. Journal of PowerSources, 81-82, 604-606.Shirakami, T., Takematsu, M., Hirano, A., Kanno, R., Yamaura, K., Takano, M., &Atake, T. (1998). Spin glass-like magnetic properties of LiNiO2. MaterialsScience & Engineering B 54(1-2), 70-72.Sickafus, K.E., Hughes, Richard. (1999, March 6, 2000). Spinel compounds: Background& historical perspective. Retrieved June 10, 2003, from http://www.ruby-sapphire.com/spinel.htmSmentkowski, V.S. (2000). Trends in sputtering. Progress in Surface Science, 64(1-2), 1-58.Smyth, D.M. (2000). The defect chemistry of metal oxides. New York: Oxford UniversityPress.Snowden, D.P.S., H. (1965). Hopping conduction in NiO. Physical Review Letters,14(13), 497-499.Stoneham, A.M. (1980). Theory of defect processes. Physics Today, 33(1), 34-42.Stoyanova, R., Zhecheva, E., Alcantara, R., Lavela, P., & Tirado, J.L. (1997). EPRstudies of Li1-x(NiyCo1-y)1+xO2 solid solutions. Solid State Communications,102(6), 457-462.Svensson, J., & Granqvist, C.G. (1986). Electrochromic hydrated nickel-oxide coatingsfor energy efficient windows: optical-properties and coloration mechanism.Applied Physics Letters, 49(23), 1566-1568.Tareen, J.A.K., Malecki, A., Doumerc, J.P., Launay, J.C., Dordor, P., Pouchard, M., &Hagenmuller, P. (1984). Growth and electrical-properties of pure and Ni-dopedCo3O4 single-crystals. Materials Research Bulletin, 19(8), 989-997.Tauc, J. (1976). Amorphous semiconductors. Physics Today, 29(10), 23-24.Tavares, A.C., Bochatay, L., Da Silva Pereira, M.I., & Da Costa, F.M.A. (1996). Oxygenevolution on NiCo2-xRhxO4 spinel system. Electrochimica Acta, 41(13), 1953-1959.Taylor, A.K., & Kagle, J.B. (1963). Crystallographic data on metal and alloy structures.New York: Dover Publications, Inc.

PAGE 208

189Tipler, P.A. (1999). Physics for scientists and engineers (4 ed. Vol. I, II, III). New York:W. H. Freeman.Ueda, K., Hase, T., Yanagi, H., Kawazoe, H., Hosono, H., Ohta, H., Orita, M., & Hirano,M. (2001). Epitaxial growth of transparent p-type conducting CuGaO2 thin filmson sapphire (001) substrates by pulsed laser deposition. Journal of AppliedPhysics, 89(3), 1790-1793.Ueda, K., Inoue, S., Hirose, S., Kawazoe, H., & Hosono, H. (2000). Transparent p-typesemiconductor: Lacuos layered oxysulfide. Applied Physics Letters, 77(17), 2701-2703.Urbano, A., Ferreira, F.F., DeCastro, S.C., Landers, R., Fantini, M.C.A., & Gorenstein,A. (2001). Electrochromism in lithiated nickel oxide films deposited by rfsputtering. Electrochimica Acta, 46(13-14), 2269-2273.Varkey, A.J., & Fort, A.F. (1993). Solution growth technique for deposition of nickel-oxide thinfilms. Thin Solid Films, 235(1-2), 47-50.Verwey, E.J., Haayman, P.W., & Romeijn, F.C. (1947). Physical properties and cationarrangement of oxides with spinel structures .2. Electronic conductivity. Journalof Chemical Physics, 15(4), 181-187.Verwey, E.J.W. (1939). Electronic conduction of magnetite (Fe3O4) and its transitionpoint at low temperatures. Nature, 144(3642), 327-328.Verway, E.J.W., & Haayman, P.W. (1941). Electronic conductivity and transition pointof magnetite ("fe3o4"). Physica, 8(9), 979-987.Windisch, C.F., Exarhos, G.J., Ferris, K.F., Engelhard, M.H., & Stewart, D.C. (2001a).Infrared transparent spinel films with p-type conductivity. Thin Solid Films, 398,45-52.Windisch, C.F., Exarhos, G.J., & Sharma, S.K. (2002a). Influence of temperature andelectronic disorder on the Raman spectra of nickel cobalt oxides. Journal ofApplied Physics, 92(9), 5572-5574.Windisch, C.F., Exarhos, G.J., Owings, R.O. (2003a). Site occupancy distribution ofcations in nickel cobalt oxides. Journal of Applied Physics, Submitted.Windisch, C.F., Ferris, K.F., & Exarhos, G.J. (2001b). Synthesis and characterization oftransparent conducting oxide cobalt-nickel spinel films. Journal of VacuumScience & Technology A-Vacuum Surfaces and Films, 19(4), 1647-1651.Windisch, C.F., Ferris, K.F., Exarhos, G.J., & Sharma, S.K. (2002b). Conducting spineloxide films with infrared transparency. Thin Solid Films, 420, 89-99.

PAGE 209

190Windisch, C.F.J., Exarhos, G. J., Engelhard, M. H., Stewart, D. C. (2003b). Influence ofheat treatment on properties of spinel ITCO films. Unpublished Manuscript.Wolverton, C., & Zunger, A. (1999). First-principles theory of cation and intercalationordering in LixCoO2. Journal of Power Sources, 81-82, 680-684.Xu, D.-X., Das, S.R., Peters, C.J., & Erickson, L.E. (1998). Material aspects of nickelsilicide for ULSI applications. Thin Solid Films, 326(1-2), 143-150.Yamamoto, H., Tanaka, S., & Hirao, K. (2003). Effects of substrate temperature onnanostructure and band structure of sputtered Co3O4 thin films. Journal of AppliedPhysics, 93(7), 4158-4162.Yoshimura, M., Han, K.S., & Tsurimoto, S. (1998). Direct fabrication of thin-filmLiNiO2 electrodes in LiOH solution by electrochemical-hydrothermal method.Solid State Ionics, 106(1-2), 39-44.Zhecheva, E., Stoyanova, R., & Zarkova, L. (1996). Lithium doping of cobalt-nickelspinel oxides at low temperatures. Materials Research Bulletin, 31(6), 593-602.

PAGE 210

191 BIOGRAPHICAL SKETCH Robert Reed Owings graduated from Gr eeley West High School in Greeley, Colorado. After one semester at Ri cks College (now renamed Brigham Young University – Idaho) he served two years of volunteer missionary service for The Church of Jesus Christ of Latter Da y Saints. He returned to st udy at Ricks College, where he married and graduated with an associate’ s degree in mechanical engineering. Transferring to the University of Idaho, he graduated cum laude in metallurgical engineering and received the outstanding senior award from the college of mines and earth resources. His master’s work was comp leted in materials science and engineering at the University of Florida. Robert is the father of two children.


Permanent Link: http://ufdc.ufl.edu/UFE0002242/00001

Material Information

Title: Polarons and Impurities in Nickel Cobalt Oxide
Physical Description: Mixed Material
Language: English
Creator: Owings, Robert Reed
Publication Date: 2003
Copyright Date: 2003

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: UFE0002242:00001

Permanent Link: http://ufdc.ufl.edu/UFE0002242/00001

Material Information

Title: Polarons and Impurities in Nickel Cobalt Oxide
Physical Description: Mixed Material
Language: English
Creator: Owings, Robert Reed
Publication Date: 2003
Copyright Date: 2003

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: UFE0002242:00001


This item has the following downloads:


Full Text











POLARONS AND IMPURITIES IN NICKEL COBALT OXIDE


By

ROBERT REED OWINGS
















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


2003


































Copyright 2003

By

ROBERT REED OWINGS

































To my lovely sweetheart, companion, and friend















ACKNOWLEDGMENTS

Many have contributed to this work in content and by means of support and

encouragement. Perhaps the greatest inspiration and motivation for the completion of

this work was my wife, Sharee, whose encouragement and support was continuous and at

times intense as needed. I acknowledge the patience of our two children, Parker and

Londyn, who have gone at times without the attention they were entitled so that this work

could be completed. Paul Holloway, my committee chair, whole-heartedly supported me

in my desires to complete this program and offered timely advice and critical analysis as

needed. The committee, David Norton, Rolf Hummel, Susan Sinnott, and Arthur Hebard

have contributed of their time and for that I acknowledge them. Greg Exarhos was

instrumental in providing experimental guidance for conducting and compiling the

research. I would like to acknowledge him as my mentor, for providing perspective,

guidance and insight. Chuck Windisch aided in the structural organization and proofread

select sections in addition to sharing key results and data from previous studies. I

acknowledge him for the significant amount of personal time he spent on reading and

discussing section layouts. His patience in training me in the operation of the many lab

instruments and fielding my unending questions was very much appreciated and for that I

acknowledge him. Kim Ferris developed the framework for the polaron disorder model,

ran the calculation, and participated in many insightful discussions on polarons,

conductivity, and life as we know it. His assistance with editing in the final hours of

preparation was most appreciated as were his encouraging words of perspective.









Others at the Pacific Northwest National Laboratory encouraged my progression

and contributed by answering my questions and providing data and insight. Special

thanks go to Mark Engelhard for XPS data and analysis, Scott Lea for AFM training, Dan

Gaspar for SIMS data and analysis, Dave McCready for XRR and GIXRD data and

analysis, and Tim Droubay for UPS data and VSM data all conducted at the EMSL

facility. I would also like to acknowledge Mark Gross, John Johnston, Pete Martin, Dean

Matson, Pete Reike, Bill Samuels, Don Stewart, Rick Williford, and others at PNNL and

for their assistance with technical equipment, details, or operations. Maggie Puga-

Lambers collected SIMS data at the University of Florida Microfabritech facility.

Valentin Cracium performed X-ray modeling and data acquisition at the University of

Florida MAIC facility. For their efforts and data provided, I thank them. I wish to

extend a special thanks to Ludie Harmon, the administrative professional who, in addition

to securing logistical arrangements, provided encouragement, plenty of sugar and

emotional energy at both high and low times making the process more enjoyable.

Funding was provided by a fellowship from the University of Florida Alumni

Association. This research was supported through the Materials Sciences and

Engineering Division of the Office of Basic Energy Sciences, U.S. DOE, and the ARO

through DARPA contract AO J209/00. A portion of the research in this work was

performed at the W. R. Wiley Environmental Molecular Sciences Laboratory, a national

scientific user facility sponsored by the U.S. Department of Energy's Office of Biological

and Environmental Research and located at Pacific Northwest National Laboratory.

Pacific Northwest National Laboratory is operated by Battelle Memorial Institute for

DOE under Contract DE-AC06-76RLO 1830.
















TABLE OF CONTENTS
Page

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

LIST OF TABLES .............. ........................................................X

LIST OF FIGU RE S ......... ............................. ...... ................ .............. X II

A B ST R A C T ........... ..................................................... ................. .. ........ .. .. X V III

CHAPTER

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

1.1 Introduction .................................................................................................. ....... 1
1.2 Organization of the D issertation ................................ ........................ ......... 4

2 BACKGROUND OF NICKEL COBALT OXIDE.....................................................6

2.1 Background of Transparent Oxide Conductors................................................ 6
2.1.1 The B asicp-n Junction D iode ........................................... ............... 7
2.1.2 N-Type Transparent Conducting Oxides.....................................................9
2.1.3 P-Type Transparent Conducting Oxides............. .......... ............... 12
2 .2 N ick el-C ob alt O x ide ................................................................ .................. .... 15
2.3 Nickel-Cobalt Oxide Conduction Mechanism..........................................24
2.3.1 Fundamentals of Polaron Conduction ................................................24
2.3.2 Fundamentals of Free Carrier Conduction ...............................................26
2.3.3 Observed Properties of Free Carriers and Polarons .................................28
2.4 Nickel-Cobalt Oxide Spinel Structure........................... .................... 32
2.4.1 Spinel Sites ............................... ..... ...... ............. ............ 35
2.4.2 V ariations of the Spinel Structure ................................... ............... ..38
2.4.3 Spinel and C onductivity ........................................ ......... ............... 39
2.5 Sum m ary of Literature Review ........................................ ........................ 39

3 SPUTTERED NICKEL-COBALT OXIDE .................................... ............... 41

3 .1 In tro d u ctio n ...................................... ............ ........... ................ 4 1
3.2 Film Preparation and Characterization Procedures ...........................................42
3.3 Characterization Results and Discussion ................... ..................... ......... 45
3.3.1 E electrical Properties......................................................... ............... 45
3.3.2 Optical Properties ........... ..... .. ......... ... ....................... 54









3.3.3 Film Structural Properties...................................... ......................... 59
3.3.4 Post D position H eat Treatm ent...................................... ............... 68
3.4 Sum m ary and Conclusions ...........................................................................76

4 TH E R O LE O F L ITH IU M ........................................................................... .... ... 78

4 .1 In tro d u ctio n ..................................................................................................... 7 8
4.2 Experim ental Procedure............................................... ............................. 80
4.2.1 Deposition.......................... .......... ........ ..........80
4.2.2 Characterization..................... ......... ......... 81
4.2.3 Sam ple H eat Treatm ent ................................... ............................ ......... 82
4.3 Characterization R results ......................................................... .............. 82
4.3.1 E electrical Properties......................................................... ............... 82
4.3.2 O ptical Properties ......................................................... .............. 87
4.3.3 Structural Properties and Composition............................. ..............90
4.3.4 C hem ical Properties......................................................... ............... 94
4 .4 D iscu ssion .................................................................................................. ......99
4.4.1 Lithium Effects .......................................... .... .... ................. 99
4.4.2 Effects of H eat Treatm ent ................................. ............. .................. 101
4 .5 C on clu sion s................................................. ................ 104

5 RHODIUM SUBSTITUTION FOR COBALT ............................... 106

5 .1 In tro d u ctio n ................................................................................................... 1 0 6
5.2 Experim ental Procedure.............................................. ............................ 109
5.2.1 Film D position ................................................. .... .. .... .. .......... 109
5.2.2 Post Deposition Heat Treatment.........................................................111
5.2.3 C haracterization...... ...................................... ...... ............ ............. .. 111
5.3 R esults............................................................................... 111
5.3.1 E electrical Properties............................................................. .. ............. 111
5.3.2 O ptical P properties .............................................................................. .... 114
5.3.3 Structural Properties ................................ ....................... .. ......... 116
5.3.4 C hem ical Properties .......................................................... ............... 119
5.4 D discussion ................................. ......................... .... ... ........ 122
5 .5 C o n clu sio n s................................................. ................ 12 6

6 DISORDER AND POLARON CONDUCTIVITY ............................................129

6.1 Introduction .............. ..... ........... ....... .................. ......... 129
6 .2 Structural D disorder ................................................................. .................. 130
6.3 Cation Charge D disorder ......................................................... .............. 134
6.3.1 Introduction ................. .......... ...................... .... ........... ............. 134
6.3.2 Disorder and the Polaron Hopping Activation Energy Model ................135
6.4 Resistivity Lim it ............. ........ .......... ...... .. ........ .. .. ......... 141
6 .5 C o n clu sio n s................................................. ................ 14 5









7 CONCLUSIONS, FUTURE WORK, AND APPLICATIONS .............................146

7.1 Conclusions........................................................ ............. 146
7.1.1 Sputtered N ickel-Cobalt Oxide ........................................ ............... 147
7.1.2 The R ole of L ithium .............................................. ............................. 148
7.1.3 Rhodium Substitution for Cobalt.................................. ............... 149
7.1.4 Heat Treatment ........... ...... ............. ..... ...... ...... ................. 151
7.1.5 Disorder and Polaron Conductivity ...................... .............152
7.2 Future Work ............. .. ..................... ...... ...............152
7.3 Applications ............. ........ .................................154

APPENDIX

A DEPOSITION DETAILS .................................................. ...... ............... 155

A 1 Sputtering .................................................. ..... ........................ 155
A. 1.1 Definition............... .............................155
A. 1.2 Equipment................................................................... 156
A 1.3 Configuration/Setup................. ........... .......................... ............... 157
A 1.4 Target Production ............................................. .... ............... 160
A .2 S solution D ep o sition ............................................. ........................................ 16 1

B CHARACTERIZATION METHODS AND DETAILS .......................................163

B. 1 Electrical Characterization.................. ....... .. ......................... ........ ........... 163
B .2 .1 H all E ffect............................................................................... .... 163
B .2.2 van der Pauw .................. .......................................... 164
B .2 Optical Characterization ............................................................................. 164
B .3 .1 F T IR ......... ..................................................................................1 6 4
B .3.2 U V V IS. ..............................................................165
B .4 Structural Characterization ........................................ .......................... 165
B .4.1 Profilom etry .............................................. .. ........ .......... ..... 165
B .4.2 X R D ................................................................... ......... 167
B .4.3 TEM / STEM ................................. .. ... ...... ...............169
B .5 Chem ical Characterization........................................... .......................... 169
B .5.1 SIM S ....................................................................... .. ......... 169
B .5 .2 U P S ..................................................................... 1 7 0
B .5 .3 X P S ..................................................................... 1 7 0

C CALCULATIONS, EQUATIONS, AND EFFECTS ...........................................172

C 1 U n it C ell S p ecifics ................................................................. .......... .. .. 172
C.2 Theoretical Film Density ........................................................173
C .3 H eike's R ule................................................174
C.4 Absorption Coefficient Calculation. ...................................... ............... 174
C.5 Jahn-Teller Effect........ .. ................................... ............................... 175
C .6 V erw ay T ransition............................................. ....................................... 176


viii









C.7 Nickel-Cobalt Oxide Activation Energy Curves ............................................176
C.8 Lithium in Nickel-Cobalt Oxide Activation Energy Curves ...........................178
C.9 Heat Treatment Quench Model..................................................179

LIST OF REFEREN CES .................................................................. ............... 180

B IO G R A PH ICA L SK ETCH .................................... ........... ................. .....................191
















LIST OF TABLES


Table page

2-1. Choice of available n-type transparent conductors............... .................. ............10

2-2. Figure-of-merit, absorption coefficient and sheet resistance comparison ................11

2-3. Approximate minimum resistivities and plasma wavelengths for some transparent
con du actors. ........................................................ ................ 12

2-4. Electrical properties of CuA102, CuGaO2, SrCu202, and NiCo204 thin films...........13

2-5. Comparison of nickel-cobalt oxide to nickel oxide and cobalt oxide ....................17

2-6. Observed properties of polarons compared with free carriers...............................31

2-7. Key characteristics of small polaron hopping for nickel-cobalt oxide....................32

3-1. Film deposition parameter and sputtering setup matrix for NixCo3-xO4....................44

3-2. Values of index of refraction for films and substrates are shown individually with
calculated transmission losses due to interfaces alone assuming no absorption......57

3-3. Extrapolated band gap values lines in Figure 3-13..............................................59

3-4. Increase in activation energies of NixCo3-xO4 from the temperature regions in Figure
3-23 graphed in Figure 3-27 .......... .... ..................... ............... 74

4-1. Numerical data from Figure 4-10 are listed in columns........................................92

4-2. Lattice parameter of Nio.75Co2.25-zLiO4 as a function of lithium fraction (z) before
and after rapid cooling from a 10 minute 3750C heat treatment............................94

5-1. Composition recipe of NiCo2-vRh04 used for solution deposited samples.......... 110

6-1. Spinel sites and occupation for polaron density calculation.............. .......... 142

6-2. Theoretical value of conductivity assuming a polaron density given by the number
of hopping polarons in a unit cell with a constant mobility............................. 143

B-1. Powder diffraction file number 73-1702 used as the standard of comparison for
nickel cobalt oxide thin films and powders.................... ..........167









B-2. X-ray reflectivity (XRR) experimental system setup parameters......................168

B-3. Grazing-incidence x-ray diffraction (GIXRD) experimental summary system setup
param eters ....................................................................... ....... ....... 168

C-1. Unit cell sites and weights for density calculation......................................172

C-2. Activation energies of NixCo3-xO4 films from Figure C-4...................................177

C-3. Activation energies of NiCo2-zLizO4 films from solution deposition. ....................178

C-4. Activation energies of Nio.75Co2.25-zLizO4 films from solution deposition. ............178
















LIST OF FIGURES


Figure page

1-1. A basic light emitting thin film stack consists ofp-n junction and transparent
electro d e ........................................................... ................ 1

1-2. A basic light detecting thin film stack device consists of ap-n junction and a
transparent electrode. ..................... .... .......... ........................... .2

2-1. A simple p-n junction diode and accompanying energy band structure illustrates the
locations of charge carriers and direction of charge flow. ........................................8

2-2. B asic optoelectronic devices. ........................................................... .....................9

2-3. Visible transmission spectra of nickel-cobalt oxide from solution deposited and
sputtered sam ples. .....................................................................18

2-4. FTIR transmission spectrum of solution deposited nickel-cobalt oxide thin film. .19

2-5. Seebeck data from nickel-cobalt oxide................... ........... ..... .. ............. 20

2-6. The oxygen is binding energy region from XPS shows a peak at 531.2 eV that
scales w ith conductivity .......................... ...................... .... ........ ...... ............2 1

2-7. Comparison of TCO transparency regions with conductivity displayed as a function
of transmission wavelength .............. .. ................................. 22

2-8. Effects of deposition techniques on the resistivity of nickel-cobalt oxide show that
sputtered films of similar compositions to solution deposited films have lower
re sistiv itie s........................... .......................................................... ............... 2 3

2-9. A bound carrier in a two-dimensional lattice is the enlarged blue atom .................25

2-10. The particle in a box plot of a particle in a one dimensional lattice bound by an
infinite energy barrier ................................................................ 27

2-11. A polaron consists of the molecules involved and the local charge region.............29

2-12. green tetrahedral cations (A), yellow octahedral cations (B), and black oxygen
anions (0) form the spinel unit cell with a chemical formula of AB204 ...............33









2-13. A calculated diffraction pattern for NiCo204 annotated with peak positions, planes,
and relative intensities ........................... ........................ .... .... ........... 34

2-14. Spinel unit cell layers stack with interlocking tetrahedral sites............................36

2-15. Tetrahedral site atom (green) bonded to four surrounding oxygen (black).............37

2-16. The octahedral site atom (yellow) bonds with six surrounding oxygen (black)
anions to form a six fold orientation. ............................................ ............... 37

3-1. Combinatorial sputtering uses a dual cathode setup with targets of different material
com p o sition ....................................................... ................. 4 3

3-2. Conductivity versus position for combinatorial runs are displayed as a function of
oxygen and argon gas com position. ........................................................................46

3-3. Position 7 from Figure 3-2 is the position with the highest conductivity..................47

3-4. Fraction of Ni and Co detected by XPS as a function of position on a combinatorial
sputtered N ixCo3-xO 4 film ..................................... ......................... ............... 48

3-5. Combinatorial sputtered NixCo3-xO4 film combining conductivity and position with
com position w ith position data ........................................... ......................... 49

3-6. A combinatorial substrate thickness profile measured by a stylus profilometer shows
that the film thickness is also graded from cobalt rich on the left to nickel rich on
th e rig h t ............................................................................. 4 9

3-7. The effect of sputtering gas pressure on NiCo alloy sputtered Nil.5Co1.504 thin film
resistivity before and after 10 minutes at 3750C heat treatment. ..........................50

3-8. Target-substrate distance affects the resistivity of sputtered nickel-cobalt oxide thin
film s from a N iCo alloy target. ........................................ .......................... 51

3-9. Conductivity of NixCo3-xO4 films on PET substrate is always less than films on other
substrates by a factor of 2-4x for as-deposited samples for all pressures and
com positions show n. ........................ ....... .... .. ..... ............... 52

3-10. UPS work function measurement of NiCo204 and Nil.5Col.504 films ..................53

3-11. FTIR traces from of reactively sputtered NixCo3-xO4 thin films. ............................55

3-12. Optical transmission through a thin film. ..................................... ............... 56

3-13. Target-substrate distance effects are opposite for the optical absorption coefficient
and resistivity from sputtered nickel-cobalt oxide thin films from a NiCo alloy
ta rg et ...................................... .................................................... 5 8

3-14. Tauc's plot ofNixCol-xO4, films show a band gap between 3 and 3.75 eV...........59









3-15. Increasing target to substrate sputtering distance decreases film density. ..............60

3-16. An oxygen plasma shows a confined plasma region (white plume) near the target
and a more dispersed plasma region (yellow-green) at longer distances ...............60

3-17. The three zone model described by Campbell for film growth in a vacuum. .........61

3-18. A spinel unit cell at different angles of rotation containing octahedral atoms,
minimized tetrahedral atoms, and select oxygen atoms for points of reference such
as the enlarged black oxygen atom .............................................. ............... 62

3-19. Grazing incident x-ray diffraction from sputtered films 50 nm thick appear to be
structured as spinel-type, but do not match exactly. ..............................................64

3-20. Sims profile of nickel-cobalt oxide thin film shows surface composition appears to
differ from the bulk com position. ........................................ ........................ 65

3-21. Cross-sectional TEM image of sputtered nickel cobalt oxide films on (100) silicon.
Nickel-cobalt oxide thin film grows in multi-grained columns. ...........................66

3-22. Nickel-cobalt oxide-silicon wafer interface. ................................... ..................... 67

3-23. An Arrhenius plot of conductivity and the reciprocal of temperature (K) shows that
at high temperatures, the film conductivity is high ............................. ............... 69

3-24. TEM image and diffraction patterns at 300 K and 600 K showing no detectable
structural changes for the two temperatures .................. ...............................70

3-25. Effects of heat treatment cooling rate after heat treatment on optical properties of
NixCo3-xO4 samples from Figure 3-23 .......................................... ................... 72

3-26. R esistivity of N iCo3-O 4................................................ .............................. 73

3-27. Rapidly quenched sample activation energies of NixCo3-xO4 change after heating
when they are slowly cooled. ..... .......................................................................74

4-1. Thin films with added of lithium deposited from solution precursors or sputtering. 82

4-2. Conductivity changes based on the cooling rate following heat treatment for all
compositions of lithium in nickel-cobalt oxide films.................. ................83

4-3. Conductivity as a function of temperature for 50 nm thick solution deposited thin
film s ...................................... ..................................... ................ 8 4

4-4. Rate of sample cooling after heat treatment had a dramatic effect on conductivity. 85

4-5. Activation energy dependence for Nio.75Co2.25-zLizO4 and NiCo2-zLizO4 from
solution on lithium content ............................ .................... 86









4-6. FTIR mid IR transmission spectra of alloy-target reactive sputter-deposited
N i0.95Co1.95Lio.150 4. ............. .......................................................... ......... 87

4-7. FTIR mid IR transmission spectra of oxide-target sputter-deposited
N il.2C o 1.2L i0.60 4. ......................... .............................................. 88

4-8. Normalized FTIR spectra at 1200 cm-1 show the absorption region near 1400 cm-1
assigned to carbonate on the surface when compared with a carbonate reference
sp ectrum sh ow n in grey ......................................... .............................................90

4-9. Grazing incidence XRD from combinatorial sputter deposited films deposited from
oxide targets. Note the weak crystalline diffraction peaks from the film ..............91

4-10. Films sputter deposited from an alloy target have a smaller lattice expansion than
film s deposited from oxide targets. ......................................................................... 92

4-11. XRD of sputtered films with incremental amounts of lithium (not normalized for
intensity) .................................. ........................... ....... ........... 93

4-12. XRD spectra show no obvious changes for fast versus slow cooling for Nio.75Co2.25-
zLizO4. Fitting the curves reveals changes shown in Table 4-4...............................94

4-13. XPS of the Carbon Is region from the combinatorial sputtered film (a) at positions
identified by the color code, and (b) curve fit to show composition of position 9. 95

4-14. XPS of combinatorial deposited nickel-cobalt oxide film with variable lithium
concentration ...................................................... ................. 96

4-15. SIMS depth profile of a thin film sputtered from a Nil.5Co1.5Li0.604 target............97

4-16. XPS of(a) Ni 2p, (b) Co 2p, (c) O Is, and (d) C Is binding energy regions from
solution deposited films of nickel-cobalt oxide containing lithium following slow
and fast cooling after heat treatment. ............................................ ............... 98

5-1. First principles d-band density of states calculation of rhodium substituted for
cobalt. Energy of 0 eV is the Fermi energy................................................. 108

5-2. Effect of rhodium fraction (h) on conductivity for four-layer films on silicon
substrates deposited from solution. .............................................. ................... 112

5-3. Sputtered nickel-rhodium oxide and nickel-cobalt oxide films deposited at 7.5 cm in
10 m Torr of 100% oxygen. ............. ......... ..............................................113

5-4. Resistivity of NiRh40x as a function of target-substrate distance. Increased distance
increases resistivity .................. .............................. ......... .. .......... ..113

5-5. Solution deposited NiCo2-hRhhO4 FTIR transmission spectra were all corrected for
the silicon substrate absorption peaks. ............ ........... ...... ............... 114









5-6. Resistivity and optical absorption coefficient at 3 .im as a function of sputtering
target-substrate deposition distance. ............ .............................. ...............115

5-7. NiRh2Ox FTIR data referenced to air (lower trace) and corrected for the silicon
substrate (upper trace). ...................................... .......... .... .. ........ .... 115

5-8. FTIR spectra of two NiRh40x samples of different thicknesses deposited from
NiRh4 alloy target by DC sputtering. ..................... .......... ............... 116

5-9. FTIR of solution deposited NiCo2-hRhhO4 and sputtered NiRhsO4 thin films
corrected for silicon substrate absorption.............. ............................................ 117

5-10. Far IR spectra of solution deposited NiCo2-hRhhO4 from FTIR show the structure
of the film s. ...................................................................... .........118

5-11. XRD of NiRh40x film from DC alloy sputter deposition.................................... 118

5-12. XRD of amorphous NiRh2Ox thin film from DC alloy sputter deposition ..........19

5-13. XPS depth profile showing the surface of sputter deposited NiRh2Ox is different
than the bulk concentration. ........................................... ............................ 120

5-14. XPS depth profile of a film sputtered from a NiRh4 target shows an oxygen
depleted surface.......................................................................................... 121

5-15. Multiplex binding energy data from XPS depth profiles. .....................................122

6-1. Schem atic of disorder. .......................................... .............................................. 130

6-2. The effect of composition on temperature dependant conductivity and activation
energy ..............................................................................13 1

6-3. NaPO3 glass phase (amorphous) with a broad band covering the bond vibrational
m o d es. .......................................................................... 13 2

6-4. A double well potential with the associated energies that influence carrier motion
betw een the w ells. ............................ .......... .. ......... .... .......... 135

6-5. Schematic illustration of one (a) and two-dimensional (b) spin models for polarons
defined by adjacent site interactions. ........................................ ............... 138

6-6. Illustration of dipole-dipole interactions between adjacent sites for polaronic spin
m odel variables in Equation 6.2 ................................................................ ....... 139

6-7. Energy distribution function for one-dimensional polaron disorder model ..........140

6-8. Energy distribution function for two-dimensional polaron disorder model ..........140









A-1. A sputtering target (red) mounted on a cathode (silver) with impinging gas ions
(purple) and target particles being removed (red). ............................................155

A-2. Vacuum chambers used for sputtering deposition ......... ..................................156

A-3. Three-inch sputter cathode used to deposit thin films ................. ...............157

A-4. Combinatorial sputtering setup inside of chamber with mounted substrates prior to
pum p-dow n for sputtering ...................... .... ............................... ............... 158

A -5. Single rotation setup. ............................... ..... ......... .......... .. ...... 159

A -6. O offset rotation setup. ..................................................................... ...................159

A -7. D double planetary rotation setup .......... ......................................... ....... ........ 160

A-8. Pressed target produced from nitrate solutions used for sputtering.....................161

A-9. The solution deposition method involves dropping a precursor solution on a
spinning substrate prior to baking to form a film ............ ...... .................161

B-1. Hall effect apparatus made by MMR technologies ..............................................163

B-2. A van der Pauw resistivity probe setup and measurement schematic. .................164

B-3. Nexus 570 FTIR with 3 detectors was used for high sensitivity mid IR (4000 cm-1
to 650 cm-1, mid IR (4000 cm-1 to 400 cm-1), and far IR (500 cm-1 to 100 cm-1)
m easurem ents .................................................................... ........ 164

B-4. Varian CAREY 5 dual beam ultra violet-visible spectrophotometer ...................165

B-5. Stylus profilometer used to measure step height of the deposited film. .................166

B-6. Optical profilometer software screen capture. Instrument made by Zygo............166

C-1. Hexagonally packed cylinders in 3-D on film and top view. ...............................173

C-2. Cubic packed cylinders top view and in 3-D as would be seen in the film ..........173

C-3. Degenerate energy level orbitals in the normal spinel state and schematic of orbital
energy shift due to the Jahn-Teller effect.................................. .................. .....176

C-4. Activation energies are the slope taken from the curve fits to the heating traces... 177

C-5. Measured conductivity as a function of temperature for NiCo2-zLizO4 films from
solution .............................................................................178

C-6. Conductivity is temperature dependent for Nio.75Co2.25-zLizO4 from solution....... 179















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

POLARONS AND IMPURITIES IN NICKEL COBALT OXIDE

By

ROBERT REED OWINGS

DECEMBER, 2003

Chair: Paul Holloway
Major Department: Materials Science and Engineering

The optical transparency from 0.3 to >10 |tm and the electrical conductivity of

NixCo3-x-yLiyO4 and NixCo3--yRhyO4 films deposited by either planar magnetron sputter

deposition or from nitrate solution were investigated. The films had low optical

transparencies (-15%) over the visible region (-300-800nm), but the transparency

increased (-50-80%) at infrared wavelengths from 2 to 25 |tm. The DC electrical

conductivity ranged from 10 to 500 Scm-1. These unique properties result in conduction

by p-type polarons. Despite low mobility of polarons (-0.1 cm2V- s-), good conductivity

results from high concentrations of small polarons (~1022 cm-3). For sputtered films, the

conductivity was larger with a 50% 02-50% Ar composition, at lower sputter gas

pressures (2 versus 10 mTorr), and with smaller target-substrate distances (best at 5 cm).

The effects of composition are due to the resulting cation oxidation states, while the

effects of pressure and substrate distance were attributed to more energetic sputtered

particles causing increased mobility of surface adatoms on the depositing films.


xviii









Increased adatom mobility led to smooth films which were 20% more dense, and

crystalline with a spinel structure. Li additions to sputter and solution deposited films of

NixCo3-x-yLiyO4 were found to decrease and increase the electrical conductivity by less

than a factor of two. The decreased electrical conductivity for sputter deposited NixCo3

x-yLiyO4 films resulted from Li occupying interstitial sites, rather than substitutional sites

in the lattice. Surface segregation and Li2CO3 formation were found. Quenching

(-150C/min) produced higher conductivities (2x), while slow cooling (100C/min)

resulted in lower conductivities after heat treatment at 3750C for 10 minutes in air. This

was attributed to a larger concentration of polarons upon quenching. Solution and sputter

deposition of NixCo3--yRhyO4 films were crystalline (spinel) and amorphous,

respectively, and remained so with heat treatments up to 3750C. Electrical conductivities

were -350 Scm-1. Transparency from 2 to 10 |tm was -10% higher than for nickel-cobalt

oxide films. A model was developed for the relationship between cation disorder and

polaron formation affecting the activation energy of electrical conductivity. Also, the

physical limit of polaron concentration was used to project a maximum conductivity

between 360 and 2400 Scm1.














CHAPTER 1
INTRODUCTION

1.1 Introduction

Materials possessing optical transparency and electrical conductivity make up a

small class of specialized materials. Transparent conducting oxide (TCO) materials are

basically doped* semiconductors categorized most often by their carrier type (n-type orp-

type) or by the mechanism of charge transport (free carrier or bound). Each classification

of carrier type or conduction mechanism exhibits complementary optical and electrical

properties.







Electrons
Transparent electrode -
n-type active layer Holes

p-type active layer

Figure 1-1. A basic light emitting thin film stack consists ofp-n junction and transparent
electrode. Injected electrons and holes combine at thep-n junction interface
and relax to emit a photon. Note that the diagram is merely for illustrative
purposes and actual thickness of films may not be proportional in scale.

Modem optoelectronics are made possible by transparent conducting oxide (TCO)

electrodes. Simple devices that use TCOs are made of nanometer-thick material layers



SDoping is adding trace amounts of impurities to a material.









stacked in various geometries to function as a light emitter or detector (see Figure 1-1 and

Figure 1-2).






av am



Transparent electrode
n-type active layer o
p-type active layer -

Figure 1-2. A basic light detecting thin film stack device consists of ap-n junction and a
transparent electrode. Impinging photons promote electrons in the valence
band into the conduction band and out of the device to an ammeter or external
circuit device. Note that the diagram is merely for illustrative purposes and
actual thickness of films may not be proportional in scale.

A very common implementation of both devices is the remote control. The hand

held unit contains the infrared emitting diode and the electronic unit to be controlled

contains either a silicon diode or a photoconductor to detect the remote control's optical

signal and interpret it to produce the desired result, such as changing the channel.

Novel light emitting diodes (LEDs) such as organic light emitting devices

(OLEDs), polymer light emitting devices (PLEDs), and longer wavelength infrared

emitting diodes (IREDs) are examples of emitting devices that require an electrode that

will transmit light and conduct electricity. This demand will increase for improved solid-

state lighting or flexible devices made on plastic (Bergh et al., 2001; Forrest et al., 2000).

TCOs are of interest for improving optical device performance and efficiency (for items

such as a solar cells and LEDs). A thorough understanding of the basic fundamentals of









how an electronic charge carrier moves in this class of materials will provide avenues for

future exploitation in device development.

Electrical conduction in a solid occurs when a charged particle (referred to as a

charge carrier) such as an electron moves under the influence of an applied electric field.

Highly conductive materials are most often metals with large populations (-1028 cm-3) of

unbound or free electrons residing in the conduction band. Electron conductors are n-type

materials, while the p-type materials conduct by the movement of electron vacancies or

holes. Conductivity of both types of TCOs can be controlled by doping levels to allow

insulating, semiconducting, or metallic behavior (Kawazoe et al., 2000). Charge

transport (electrical conduction) mechanisms in TCOs are similar to metallic free-carrier

conduction or bound carrier hopping.

Oxides are transparent in the visible region when the band gap exceeds 3.1 eV

(Kawazoe et al., 1997; Lewis & Paine, 2000). Most often these oxides are insulators

rather than conductors due to relatively few carriers contributing to conductivity.

Researchers have enhanced transparent oxides by introducing donor states just below the

conduction band to produce more carriers in the band and result in conductivity increases.

One trick to adding donor states requires slightly reducing the film (such as annealing in

a hydrogen atmosphere) to create oxygen defect sites that act as donors (Lewis & Paine,

2000). Attempts to make transparent oxides conducting have yielded higher carrier

concentrations but have also affected transmission properties. One problem with free

carriers is a limited transmission region. While transparent in the optical spectrum, the

onset of absorption by free carriers at the plasma frequency abruptly halts transmission.

Still, other researchers have attempted to push this absorption region further out into the









IR by shifting the region of free carrier absorption to a longer wavelength. Conducting a

hole instead of an electron shifts the absorption region to a longer wavelength because the

hole has a larger effective mass than the electron. This p-type system is however in the

free carrier regime and still subject to free carrier absorption, only at a longer wavelength.

Larger masses also move slower than smaller masses, so conductivity of the p-type

material is reduced. A material system conducting by a bound carrier would escape this

absorption peril caused by free carriers.

A polaron is a localized carrier with an accompanying lattice strain. A polaron

conducting material inherently has a low mobility as well as the absence of a plasma

frequency cutoff in the infrared region. Despite a low rate of polaron hopping at room

temperature, a high polaron concentration will promote a high conductivity. Recent

studies have focused on increasing the density of polarons in such materials to obtain a

net increase in conductivity. Polaron density can be increased by a number of ways.

Disordered cation arrangement, increased oxidation state due to selective doping, or

structural disorder introduced by cation size mismatch in the lattice are all ways of

increasing the number of polarons in a given volume of material.

Nickel-cobalt oxide is a recently studied polaron conducting material that shows

promise as an infrared-transparent conducting oxide (ITCO). NO other ITCO has been

studied or reported to date. This work discusses the polaron conducting nickel-cobalt

oxide system, its IR optical and electrical properties and how they are affected by key

deposition parameters and the addition of impurities.

1.2 Organization of the Dissertation.

Chapter 2 discusses common TCOs and details the current knowledge of their

properties. Base properties of the nickel-cobalt oxide system will include the polaron









conduction mechanism and its spinel structure as reported in the literature. Chapter 3

describes efforts to enhance the nickel-cobalt oxide properties for use as an ITCO by

altering the thin film deposition parameters. Chapters 4 and 5 report the effects of doping

the nickel-cobalt system with lithium and rhodium, respectively. Chapter 6 discusses the

conduction mechanism in the nickel-cobalt oxide system and outlines a model for

understanding the effects of disorder on conductivity to explain the experimental results.

A projected maximum range for conductivity is estimated as well. Conclusions from this

research and recommendations for future work and applications comprise Chapter 7.














CHAPTER 2
BACKGROUND OF NICKEL COBALT OXIDE

This chapter provides the necessary background information to understand the

significance of the infrared transparent nickel-cobalt oxide system and how its properties

are unique among transparent conducting oxides.

An explanation of two basic p-n junction diodes is given below to illustrate the

difference betweenp-type and n-type materials and shows how a transparent conducting

layer functions as part of a light emitting or detecting device. Limitations of each type of

TCO include a particular transmission window and conductivity, both of which depend

on the conduction mechanism and carrier type. N-type TCOs are widely used and well

developed. Thep-type TCOs are limited primarily by poor conductivity and are less

developed. Research has taken two different approaches to improve the conductivity of

p-type TCOs. Explanations of conduction mechanisms such as free carrier movement

and polaron hopping provide the background knowledge for this discussion. Nickel-

cobalt oxide is the p-type material system of interest to function as an infrared transparent

conducting oxide (ITCO) because of its infrared transparency, stability in oxygen, ease of

preparation, phase purity, and high conductivity. Nickel-cobalt oxide exhibits these

characteristics due in part to the carrier type, the conduction mechanism, and the crystal

structure.

2.1 Background of Transparent Oxide Conductors

Transparent conducting oxides are basically doped metal-oxide semiconductors

classified as either n-type orp-type. Majority electron conductors are n-type materials,









while the p-type materials conduct by the movement of electron vacancies called holes.

Conductivity of both types is controlled by doping to allow insulative, semiconducting, or

metallic behavior (Hummel, 2001; Kawazoe et al., 2000). Currently, n-type TCOs

dominate with respect to commercial use and have wide spread application because they

exhibit superior conductivity and minimal photon absorption over the visible range as

compared with p-type materials. However, p-type oxides exhibit some unique properties

such as a majority hole carrier concentration resulting in increased infrared transparency

not possessed by n-type oxides. This p-type conductivity and infrared transparency make

them of interest for future development (Kawazoe et al., 2000; Kawazoe et al., 1997;

Ohya et al., 1998; Windisch et al., 2001a; Windisch et al., 2001b; Windisch et al., 2002b)

2.1.1 The Basic p-n Junction Diode

A fundamental optoelectronic device illustrates the differences between n-type and

p-type materials that make up ap-n junction. A diode (or rectifier) is a device that allows

current to pass in one direction, but not in the reverse direction. The most basic diode

consists of a p-type region next to an n-type region referred to as ap-n junction.

Junctions can be homojuctions, made of the same material (e.g., silicon doped with boron

forp-type behavior or with phosphorus for n-type behavior (Hummel, 2001)), or

heterojunctions, made from two different materials. In either case, the behavior of the

diode results from the interface between the two different types of materials. A depletion

region forms at the interface with an accompanying electric field (Figure 2-1), which

effectively removes generated carriers or allows carrier recombination (also referred to as

the space charge region or the depletion layer (Hummel, 2001)).













------------------------


SValence Band








n-type material p-type material
Depletion Region
Figure 2-1. A simple p-n junction diode and accompanying energy band structure
illustrates the locations of charge carriers and direction of charge flow. The n-
type material has electrons in the conduction band while the p-type material
has holes in the valence band. When a bias is applied to the junction, the
bands shift and electrons (or holes) move from left to right (or opposite) and
through the rest of the circuit.

Light emission occurs as an electron and hole recombine when injected into the

depletion region. The energy of the photon emitted is the difference between the energy

of the electron in the conduction band and the hole in the valence band (band gap

energy). Photo current is generated as photons impinge on the depletion region and

excite electrons to the conduction band (over the energy gap) leaving behind holes in the

valence band. Carriers are conducted away from the depletion region similar to the photo

diode current diagram in Figure 2-2. Different carrier types are imperative to p-n

junction device behavior. Without the two different types, neither light emission nor

photo-generation of carriers would occur. Figure 2-2 shows simple p-n junction diodes,

one as a light emitter and the other as a light detector.





























Figure 2-2. Basic optoelectronic devices. The light emitting diode (LED, on the left)
incorporates n-type and p-type materials to essentially "convert" electrical
current to light by combining electrons and holes and produce a photon. The
photo diode (solar cell -on the right) uses light (photons from the LED) to
produce electrical current by generating electron-hole pairs. Materials in
these devices conduct a majority of holes or electrons and are classified asp-
type or n-type respectively. TCOs are used in these devices to allow current
and light to pass through to the p-n junction.

Key to the operation of a diode is the ability to get current in or out of the device.

Transparent conducting oxides serve a unique function in optoelectronic p-n junction

stacks because of their conductivity and transparency. TCOs allow current to pass

through to the device depletion region without significantly blocking the light that is

being emitted or detected.

2.1.2 N-Type Transparent Conducting Oxides

Prevalent TCOs typically possess extrinsic n-type conduction (Chopra et al., 1983)

and are more researched due to their superior conductivities. Increased research has sped

development and applications of common n-type TCO systems such as (1) the zinc oxide

system (e.g., zinc oxide doped with fluorine, boron, gallium, indium, or aluminum), (2)


N-type TCO
N-type layer
P-type layer









cadmium stannate (Cd2SnO4), (3) the tin oxide system (e.g., tin oxide and tin oxide doped

with antimony [SnO2:Sb referred to as ATO]), and (4) indium oxide doped with tin

(In203:Sn referred to as indium tin oxide or ITO) (Gordon, 2000). No single TCO is

suited for all applications. Table 2-1 shows that if high conductivity is important, then

ITO is the best material, however if highest transparency is most important, cadmium

stannate or zinc oxide may be more appropriate.

Table 2-1. Choice of available n-type transparent conductors.
Property Material
Highest transparency ZnO:F, Cd2SnO4
Highest conductivity In203:Sn
Lowest plasma frequency Sn02:F, ZnO:F
Highest plasma frequency In203:Sn
Highest work function, best contact to p-Si SnO2:Sb, ZnSnO3
Lowest work function, best contact to n-Si ZnO:F
Best thermal stability SnO2:F, Cd2SnO4
Best mechanical durability Sn02:F
Best chemical durability Sn02:F
Easiest to etch ZnO:F
Best resistance to hydrogen Plasmas ZnO:F
Lowest deposition temperature SnO2:F, ZnO:B
Least toxic ZnO:F SnO2:F
Lowest cost Sn02:F
Reproduced from Gordon, R.G. (2000). Criteria for choosing transparent conductors.
MRS Bulletin, 25(8), 52-57, (Table 8, p.55) by permission of MRSBulletin.

Currently the most common transparent conducting material used in electronics is

indium tin oxide (ITO). It transmits up to ninety percent of visible light with a resistivity

near lx(10)-4 Qcm. High transparency, high conductivity, and chemical stability

contribute to the wide use of ITO. An oxide conductor like indium tin oxide has a carrier

density on the order of lxl120 cm-3 to lxl021 cm-3 (Minami, 2000; Minami et al., 1995;

Minami et al., 2000) and a mobility typically greater than 10 cm2V-ls- (Minami, 2000).

Other material systems are better suited for specific applications, because each system

has specific strengths and weaknesses.









Table 2-2. Figure-of-merit, absorption coefficient and sheet resistance comparison.
Sheet Resistance Rs Visible Absorption Figure of Merit a
Material (Q/square) Coefficient a cY/a
ZnO:F 5 0.03 7
Cd2SnO4 7.2 0.02 7
ZnO:Al 3.8 0.05 5
In203:Sn 6 0.04 4
SnO2:F 8 0.04 3
ZnO:Ga 3 0.12 3
ZnO:B 8 0.06 2
SnO2:Sb 20 0.12 0.4
ZnO:In 20 0.20 0.2
Reproduced from Gordon, R.G. (2000). Criteria for choosing transparent conductors.
MRS Bulletin, 25(8), 52-57, (Table 2, p.53) by permission of MRS Bulletin.
aFigure-of-merit (l/a), the calculated ratio of a material's electrical conductivity (o) to
its total optical absorption coefficient (a) is one method of comparing different TCOs.
o/a = -{RsLn(T+R)}1. Rs is the sheet resistance, T is the total visible transmittance, R is
the total visible reflectance, all of which are required to calculate the figure of merit
(Chopra et al., 1983). Larger figure of merit values indicate higher performing materials.
Rs = p(t)/ t = 1/(o(t) t). Resistivity, a function of thickness, when divided by film
thickness is equal to sheet resistance. Conductivity is the reciprocal of resistivity (Chopra
et al., 1983).

One method of comparing different TCOs on the same criteria is to devise afigure-

of-merit number as found in Table 2-2. Sheet resistance divided by the total optical

absorption coefficient gives a number to compare both electrical and optical properties

simultaneously. Zinc oxide doped with fluorine and cadmium stannate have the highest

figure-of-merit at 7, but due to lower conductivity and higher cost, they are not utilized as

much as ITO or tin oxide. It is important to note that the figure of merit is an extremely

sensitive to the wavelength of the light being transmitted.

ITO is limited in that it is essentially afree electron conductor. Nearly-free

electrons absorb photons with energies typically less than 1.1 eV corresponding roughly

to photon wavelengths longer than 1 im. This onset of absorption by free electrons is

called the plasma frequency (Hummel, 2001). Typical values are included in Table 2-3.

While this absorption is what makes tin oxide extremely valuable to the structural









building community when used as an insulating thin film on windows, it limits the use of

ITO as a transparent electrode at longer wavelengths.

Table 2-3. Approximate minimum resistivities and plasma wavelengths for some
transparent conductors.
Material Resistivity p (mQcm) Plasma X ([tm)
In203:Sn 0.100 >1.0
Cd2SnO4 0.130 >1.3
ZnO:Al 0.150 >1.3
SnO2:F 0.200 >1.6
ZnO:F 0.400 >2.0
Reproduced from, Gordon, R.G. (2000). Criteria for choosing transparent conductors.
MRS Bulletin, 25(8), (Table 3, p.54) by permission of MRS Bulletin.
Many studies have been conducted on the n-type oxides to develop applications,

however transparent p-type materials are not well developed despite the need for them

(Giesbers et al., 1997; He et al., 1999; Park et al., 2002).

2.1.3 P-Type Transparent Conducting Oxides

Investigations into p-type TCOs include two major systems (1) the nickel oxide

systems, nickel oxide (Giesbers et al., 1997; Sato et al., 1993) with additions of cobalt

oxide (He et al., 1999; Windisch et al., 2001a; Windisch et al., 2001b), and (2) the copper

oxide systems, strontium copper oxide (Kawazoe et al., 2000), copper scandium oxide

(Duan et al., 2000), copper gallium oxide (Ueda et al., 2001), lanthanum copper oxide

(Ueda et al., 2000), and copper aluminum oxide (Kawazoe et al., 1997). Research

conducted on p-type TCO material systems has yielded a fewp-n devices, but beyond

prototypes, they have not been further developed (Giesbers et al., 1997; Ohya et al.,

1998).

Infrared spectra of all the listedp-type materials have a plasma cutoff deeper in the

IR than the n-type TCOs except the nickel systems that do not have a published plasma









cutoff. Conducting via holes, they may be used as ap-type electrode next to thep-type

layer of a p-n junction diode or as the p-type layer in the junction.

Table 2-4. Electrical properties of CuA102, CuGaO2, SrCu202, and NiCo204 thin films.
CuA1O2 a CuGaO2 a SrCu202 a NiCo204b
Electrical conductivity (S cm ) 9.5x10- 6.3x10-2 4.83x10-2 333
Carrier density (cm-3) 1.3x1017 1.7x1018 6.1x1017 lx1021
Hall mobility (cm2/V s) 10.4 0.23 0.46 < 0.1
Seebeck coefficients ([V K-1) +183 +560 +260 -+20
a(Kawazoe et al., 2000)
b(Windisch et al., 2002b)

Limitations ofp-type TCOs include low conductivity (near 0.1- 0.01 Scm-1) and

70%-90% transparency in the visible region. With conductivity about two orders of

magnitude lower (at best) than a good n-type TCO such as ITO at 10000 Scm-, p-type

TCO electrical properties shown in Table 2-4 do not yet compare with n-type TCOs.

Plasma absorption of free or quasi-free carriers limits transparency at longer wavelengths

regardless of the conducting carrier type. Interest inp-type TCOs has recently renewed

with new ideas on how to overcome these limitations (Kawazoe et al., 2000).

Conductivity, the major limitation forp-type TCOs, is a function of the number of

carriers and the speed at which those carriers move through the material. Shown in

equation 2.1, conductivity is a function of carrier density (n) multiplied by mobility (/)

and the charge of each carrier (q).

cr = = n/q (2.1)
P
Assuming this equation is suitable for all types of materials, two choices are

available to increase conductivity because the charge of each carrier is constant. One

choice is to increase mobility by controlling the valence band; the other option is to boost

the carrier density.









Chemical modulation of the valence band. Kawazoe et al. suggest the limited

conductivity of the p-type TCOs result from valence cations bonding with oxygen

forming localized carriers. In contrast, high conductivity observed in n-type TCOs is

often due to high free carrier densities even with carrier mobilities at a low 10 cm2V s-1.

Locally bound p-type carriers present a problem because they are less mobile than free

carriers. The solution is referred to as chemical modulation of the valence band (CMVB)

where cation selection in the lattice controls the valence band while maintaining quasi-

free carrier movement. CMVB rids the valence band of positive holes that become deep

acceptor levels by introducing covalency in the metal-oxygen bonding to form an

extended valance-band structure. Transition metal oxides with unfilled dio shells are not

recommended for TCO applications because d-d transitions exhibit strong coloration in

the visible. Copper oxide with select cations added was the example system (Kawazoe et

al., 2000). The best conductivity reported in Table 2-4 is 0.95 Scm-1 for CuA102

indicating that this regime worked to some degree, but conductivity remains low. If

mobility is at the maximum for the system, and e is constant at 1.606x10-19 C (per

electron), then the only other parameter to influence in the conductivity equation

(Equation 2.1) is the number of carriers.

Increase carrier density. Another approach to improving conductivity is to leave

the carriers localized and attempt to add more of them (Windisch et al., 2002b).

Localized carriers do not exhibit the infrared absorption edge allowing transparency at

longer wavelengths and are called polarons when accompanied by lattice strain. They

have a very low mobility due to a hopping transport mechanism unlike free carriers. In

addition, cations that have open d shell orbitals such as nickel and cobalt in nickel-cobalt









oxide reported by Windisch et al. have a conductivity of up to 333.3 Scm-1 and have

exhibited transparency up to 60% at 600 nm and higher at higher wavelengths (Windisch

et al., 2001a). The nickel-cobalt system is not as transparent in the visible region as the

copper system, though very transparent in the infrared.

2.2 Nickel-Cobalt Oxide

Interest in nickel-cobalt oxide stemmed from early work on either nickel oxide or

cobalt oxide. Nickel oxide studies explored electrical properties with interest in its

superior transmission (near 80%) in the visible region. Attempts to exploit this in

electrochromic window applications (Kitao et al., 1994), as p-type transparent conducting

films (Sato et al., 1993), an antiferromagnetic material (Fujii et al., 1996), a functional

sensor layer for chemical sensors (Kumagai et al., 1996), or as a photocathode for a solar

cell (He et al., 1999), fueled interest in this system. Cubic nickel oxide has also been

investigated with respect to electrical properties (Morin, 1953) and its conduction

mechanism (Biju & Khadar, 2001; Parravano, 1954; Sanz et al., 1998; Snowden, 1965).

Studies of nickel oxide showed experimental results, but as alluded to previously, poor

conductivity of the p-type oxides has not encouraged interest in developing p-type TCOs

in general for applications beyond lab experimentation.

Spinel cobalt oxide has also been studied previously as a result of its magnetic spin

states (Belova et al., 1983), optical nonlinearity (Yamamoto et al., 2003), gas sensing

capabilities, and solar energy reflecting properties (Cheng et al., 1998). Doping single

crystals of cobalt oxide with nickel produced a significant increase in electrical

conductivity (up to 105x) while maintaining the spinel structure. Nickel cations were

found to reside in octahedral sites with a valence of 2+ and 3+ (Roginskaya et al., 1997;

Tareen et al., 1984) represented by the equation below.









Co2Co+ [Co3NiNi 3 (2.2)
O-OY Y 0 Y X-Y ]4 (2.2)r4
Nickel doped cobalt oxide showsp-type semiconducting behavior similar to

intrinsic spinel cobalt oxide (Tareen et al.). Investigators have worked to develop nickel-

cobalt oxide for applications intended originally for nickel oxide or cobalt oxide (Monk

& Ayub, 1997). Other interests in nickel-cobalt oxide include uses as electrodes in

batteries (Liu et al., 1999; Polo Da Fonseca et al., 1999; Yoshimura et al., 1998),

electrodes in solar cells (Park et al., 2002), electrodes in molten carbonate fuel cells

(Kuk et al., 2001), or as a heterogeneous optical recording media, (lida & Nishikawa,

1994). These intended uses of nickel-cobalt oxide are not paramount to this study but

show other research interests in the material have been pursued in the past to develop

specific applications such as using NiCo204 as an electrocatalyst for anodic oxygen

evolution (Carey et al., 1991; Haenen et al., 1986), in organic or inorganic

electrosynthesis (Roginskaya et al., 1997), as a supercapacitor (Hu & Cheng, 2002), or as

an infrared-transparent conducting electrode for flat panel displays, sensors, or optical

limiters and switches (Goodwin-Johansson et al., 2000; Windisch et al., 2001a; Windisch

et al., 2001b; Windisch et al., 2002b).

Studies of nickel-cobalt spinel have included bulk crystals (Haenen et al., 1986;

Roginskaya et al., 1997; Tareen et al., 1984), powders, (Windisch, 2003) and thin films

(Carey et al., 1991; Galtayries & Grimblot, 1999; Hu & Cheng, 2002; Kim et al., 2000;

Marco et al., 2000; Monk & Ayub, 1997). Nickel-cobalt oxide studies have correlated

the crystal structure and activity of NiCo204 and related oxides (King & Tseung, 1974),

explained the NiCo204 spinel surface by Auger and XPS (Kim et al., 2000), and have

expounded on the deposition parameters and substoichiometric structures (Carey et al.,

1991). Work on the nickel-cobalt oxide system and stoichiometric deviations has









included infrared spectroscopy (Windisch et al., 2001b), laser Raman spectroscopy

(Windisch et al., 2002a), Seebeck and conductivity measurements (Windisch et al.,

2002b), and conductivity measurement variations correlated to film composition

(Windisch et al., 2001b). Post deposition heat treatment and resulting increases in

conductivity were correlated with oxygen binding energy changes (Windisch et al.,

2001b; Windisch et al., 2002b; Windisch, 2003b).

Table 2-5. Comparison of nickel-cobalt oxide to nickel oxide and cobalt oxide.
NiO NiCo204 Co304
Breakdown temp. (C) 1100+ 400 a 895 a
Conductivity (S/cm) 2x10-2'b 333 c 10-6 d, 0.05 e, 0.07 f
Carrier type p-type small polaron, p-type small hp efi
Carrier type g l p- hopping
p polaron h
Carrier density (cm3) at 9.8x1020 1020,k, 1019 km
room temp. 1.3x109 b 5x102 2.4x0l9 e
Lattice constant (A) 4.195 n, 4.176 o 8.114 P 8.084 q, 8.11 o
Structure Cubic n Spinelr Spinel d,r
Seebeck coef.(iV/C) 350-500j 20 m 0 d
Band gap (eV) 4.0 s, 3.8' Not reported 15 1.65 2.0 2
1_ 1_ _2.02wJ
a (Haenen et al., 1986) m (Windisch et al., 2002a)
b (Morin, 1953) n (Hotovy et al., 1998b)
c (Windisch et al., 2001a) o (Kennedy, 1996)
d (Tareen et al., 1984) P Powder Diffraction File #73-1702 (see
e(Patil et al., 1996) Appendix B for specifics)
f(Cheng et al., 1998) q (Taylor & Kagle, 1963)
S(Sato et al., 1993) r(King & Tseung, 1974)
h (Windisch et al., 2002b) s(Lunkenheimer et al., 15)
(Ohya et al., 1998) t (Schumacher et al., 1990)
J (Parravano, 1954) u (Varkey & Fort, 1993)
k (Snowden, 1965) (Pejova et al., 2001)
1(Sato et al., 1993) w (Yamamoto et al., 2003)

In Table 2-5 a comparison of nickel-cobalt oxide to nickel oxide and cobalt oxide

outlines key differences in the base oxides and the mixed material. When mixed

together, the breakdown temperature is 4000C (far below either binary oxide), while at

the same time the carrier concentration increases by up to 102x and the conductivity









increases by 105x. The lattice parameter is largest for the mixture of nickel-cobalt oxide

and the material exhibits the samep-type nature as both nickel oxide and cobalt oxide.

Work function measurements were not available for these materials including nickel-

cobalt oxide. Band gap information was also not available nor was the index of

refraction

40

35 -
30 -

2 -

20 -

S 15

10 '-- --

Solution Deposition -- Sputtered
0 I --- i ---- i ---- i ---- i --

300 400 500 600 700 800
Wavelength (nm)
Figure 2-3. Visible transmission spectra of nickel-cobalt oxide from solution deposited
and sputtered samples. Reprinted from Windisch, C.F., Exarhos, G.J., Ferris,
K.F., Engelhard, M.H., & Stewart, D.C. (2001). Infrared transparent spinel
films with p-type conductivity. Thin Solid Films, 398, 45-52 (Figure 3, p.47)
with permission from Elsevier, color added.

Using the available data one might reasonably interpolate the existing reports and

surmise a calculated bandgap in the neighborhood of 2.5-3.0 eV, if Vegard's law were

valid for this system. However, the band gap for NiO may not be a valid number to use

in this interpolation method because nickel, in NiCo204, is in the same chemical state as

it would be found in Ni304. Since Ni304 is not a stable phase existing at standard

temperature and pressure conditions, no electrical information is available for it. Nickel









oxide and cobalt oxide individually differ in structure and electrical properties, but they

both exhibit electrical properties far inferior to mixed nickel-cobalt oxide.

According to Figure 2-3, optical absorption begins near 600 nm where the

transmission is nominally less than 30%. As stated previously, this is most likely due to

d-dtransitions absorbing the visible light. By TCO standards, this transparency is

unacceptable for use in any practical device. Nickel-cobalt oxide transparency is

acceptable at longer wavelengths as seen in Figure 2-4 below.


90 -

I85-

80- 641
Sso

S75 -
546
70 -
1 I I I I I I
3500 2500 1500 500
Wave number (cm ')
Figure 2-4. FTIR transmission spectrum of solution deposited nickel-cobalt oxide thin
film. Reprinted from Windisch, C.F., Exarhos, G.J., Ferris, K.F., Engelhard,
M.H., & Stewart, D.C. (2001). Infrared transparent spinel films with p-type
conductivity. Thin Solid Films, 398, 45-52, (Figure 4(b), p.48) with
permission from Elsevier, color added.

Separately neither nickel oxide nor cobalt oxide has the electrical properties to be

of industrial interest, though they are both p-type with high carrier counts and low

mobility. Windisch et al. determined the carrier to bep-type using the Seebeck

coefficient (Windisch et al., 2002a; Windisch et al., 2002b). Figure 2-5 shows Seebeck

data taken from a nickel-cobalt oxide thin film. They report that Ni3+ also has a role in

conductivity such that the mechanism of conduction is a charge transfer between resident

divalent and trivalent cations suggesting it is possibly assisted by the magnetic nature of









the oxide film (Windisch et al., 2002b). The magnetic nature of the film has been blamed

for difficulty in obtaining reliable results from the Hall effect measurement. Anomalous

Hall effect measurements made an absolute determination of carrier concentration and

carrier type difficult using the Hall effect measurement. A primary assumption of the

Hall effect is that the measurement occurs in a material withfree carriers.

60
SNiCo204 Film
50 75 nm Thick

S40



20 o
29* *.

10 -
**
0
I I I
300 400 500 600
Temperature (K)
Figure 2-5. Seebeck data from nickel-cobalt oxide. Reprinted from Windisch, C.F.,
Ferris, K.F., Exarhos, G.J., & Sharma, S.K. (2002). Conducting spinel oxide
films with infrared transparency. Thin Solid Films, 420, 89-99, (Figure 6(a)
p.93) with permission from Elsevier, color added.

Emin addressed an anomalous Hall sign in oxide semiconductors. The sign of the

Hall effect measurement depends on the nature and relative orientations of the local

orbitals between which the carrier moves and on the local geometry. The sign of the Hall

angle is not unambiguously determined by the sign of the charge carrier in the case of

small-polaron hopping motion. The Hall effect depends on the local geometry and on the

nature of the local electronic states. The observed anomalies of the sign of the Hall angle

may be explained as simply being a manifestation of the hopping nature of the transport

in phonon-assisted hopping motion (Emin, 1977). Polaron conductors and the Hall effect









have been discussed in the literature for the mentioned reasons. The Hall effect does not

produce reliable reproducible values of carrier type, density, or mobility. This Hall effect

anomaly is likely due to the localized nature of the carriers and the hopping conduction

mechanism of nickel-cobalt oxide.






Ni,. Co,,04
NS il)75,Co02254








540 536 532 528 524 520
Binding Energy (eV)
Figure 2-6. The oxygen Is binding energy region from XPS shows a peak at 531.2 eV
that scales with conductivity. Reprinted from Windisch, C.F., Exarhos, G.J.,
Ferris, K.F., Engelhard, M.H., & Stewart, D.C. (2001a). Infrared transparent
spinel films withp-type conductivity. Thin Solid Films, 398, 45-52, (Figure 6.
p.49), with permission from Elsevier, color added.

Nickel-cobalt oxide is probably a defect conductor due to adsorption of oxygen

similar to nickel oxide where high oxygen partial pressure increases conductivity

(Hotovy et al., 1998a). Cation vacancies produced from oxygen adsorption create holes

in the valence band making the material p-type and it is therefore classified as an

electron-defect semiconductor. (Azaroff, 1960) Additionally, the defect states of lattice

oxygen monitored by XPS in the oxygen Is region (see Figure 2-6) at a binding energy of

531.2 eV are believed to scale with conductivity and may be an indicator of polarons

(Windisch et al., 2001a).









100000ITO
-- ITO
0000 ........ ..................-- ZnO:Al
S- CdSn42

S-- ZnO:F
.100 ---------- ---------- NiCo2O4
S--CuAlO,
S--Visible Spectrum CuGaO2
*e 10-------------------------------CuaO2-
S--SrCu202
S1,----- -........................................



0.01 ,
0 2 4 6 8 10
Wavelength (pm)

Figure 2-7. Comparison of TCO transparency regions with conductivity displayed as a
function of transmission wavelength. The long line in center is nickel-cobalt
oxide transparency range. It extends off the chart into the far IR with much
better conductivity than the otherp-type TCOs.*

Windisch et al. have demonstrated the nickel-cobalt oxide system as a prospect for

development with respect to conductivity and infrared transparency (shown in Figure 2-

4) (Windisch et al., 2001a; Windisch et al., 2002a; Windisch et al., 2001b; Windisch et

al., 2002b; Windisch, 2003b). Their work has suggested that the conductivity of the

nickel-cobalt oxide system could be improved by the addition or substitution of selected

cations, such as lithium, rhodium, or palladium. (Windisch et al., 2001a; Windisch et al.,

2001b; Windisch et al., 2002b) Figure 2-7 shows that nickel cobalt oxide has a

transparency window that extends more into the IR than typical n-type or other recently

studiedp-type TCOs. No other reported transparent conductor has a similar transparent

region.


* UV cutoff values of the TCOs are approximate for a qualitative comparison of the transparent regions in
the IR. Plasma cutoff values for each are taken from the literature (Gordon, 2000; Windisch et al., 2001b;
Windisch et al., 2002b). Plasma values forp-type TCOs estimated.









The nickel cobalt oxide system is unique in that it exhibits p-type conductivity with

highly localized carriers bound to the lattice with an accompanying lattice strain, i.e. with

polaron conduction. This bound carrier and lattice strain together are known as a small

polaron. Carrier mobility is on the order of 0.1 cm2V- s- due to the lattice bound

localized carriers (Windisch et al., 2002b). Conduction occurs in these materials in spite

of the low mobility because of a high polarons density, ~1021 cm-3 (Windisch et al.,

2002b). Conductivity up to 333 Scm-1 has been measured (Windisch et al., 2001a).

4.0
3.0^ Solution Deposited
0" \ o Sputtered
2.0.

1.0

0.0

3-1.0 *

-2.0

-3.0
3.00 2.25 1.50 0.75 0.00
Composition (x) in NixCo3,04
Figure 2-8. Effects of deposition techniques on the resistivity of nickel-cobalt oxide
show that sputtered films of similar compositions to solution deposited films
have lower resistivities. Reprinted from Windisch, C.F., Exarhos, G.J., Ferris,
K.F., Engelhard, M.H., & Stewart, D.C. (2001a). Infrared transparent spinel
films with p-type conductivity. Thin SolidFilms, 398, 45-52. (Figure 2, p.47)
with permission of Elsevier, color added.

Windisch et al. report that the best conductivity of sputtered films was found at

nickel to cobalt ratio of 1:1 for sputtered films and 1:2 for solution films (Figure 2-8).

Both the 1:1 sputtered and 1:2 solution deposited nickel-cobalt oxides were cubic spinel

structured. Also included in Figure 2-8 was a comparison of solution and sputter

deposited films showing the difference at the same composition for films deposited by









two different methods. It is believed that the difference in the film conductivity comes

from lattice cation disorder accommodating polaron charge carriers (Emin & Bussac,

1994). Sputtering allowed a higher conductivity while at the same time preventing less

transmission in the visible region due to strong absorption by d-d transitions from either

the nickel or the cobalt ions. The spin state of Co3+ in the octahedral site may act as an

acceptor and Ni2+ may exist in the octahedral site. Both of these cation states may

enhance the conductivity (Windisch et al., 2002b).

2.3 Nickel-Cobalt Oxide Conduction Mechanism

Nickel-cobalt oxide conducts electrically by a mechanism of small polaron hopping

(Windisch et al., 2002b). An understanding of why polarons and free carriers are

different sheds light on why the material properties are also different. Polaron

conductivity gives nickel-cobalt oxide its unique properties of infrared transparency and

high conductivity despite low values of mobility. Large numbers of polarons form with

increased structural disorder that can be introduced by processing, selective cation

addition, or cation substitution. Understanding polarons may give insight to discovering

polaron-hopping conduction in other materials as well.

2.3.1 Fundamentals of Polaron Conduction

When a charge carrier such as an electron or hole distorts its neighboring lattice

structure and traps itself, it is called a "polaron." A polaron is described as an entity

including both the displaced neighboring atoms (localized strain) and the trapped carrier

(Cox, 1987; Emin, 1982; Austin, 1969).

Mistakenly called a polaron due to early investigations of self-trapping in polar and

ionic solids, self-trapping is not restricted just to polar and ionic solids with long-range

dipolar electron-lattice interactions. Polarons consist of two main types: large and small.









The two are distinguished by the severity of the localized strain area. "A small polaron is

an extra electron or hole severely localized within a potential well that it creates by

displacing the atoms that surround it." (Emin, 1982, p.34). When the electronic carrier

and the lattice distortion together have a linear dimension less than the lattice parameter,

it is referred to as a small polaron (Kingery et al., 1975, p.870). Short-range electron-

lattice interaction plays the major role allowing small polarons to occur in polar,

covalent, and ionic materials and generally is comprised of interactions of the carrier with

acoustic and optic vibrational modes of the lattice. Polaron motion can be described as a

succession of phonon-assisted hopping steps (Emin, 1982).










(a) (b) (c)
Figure 2-9. A bound carrier in a two-dimensional lattice is the enlarged blue atom. The
polaron is the carrier and the accompanying strain. Schematic (a) is before the
hop, (b) is the move, and (c) is the strain movement after the hop.

A hop consists of the following three steps shown in Figure 2-9 above:

1. Atoms arrange to allow multiple positions for a charge carrier.

2. The charge moves between degenerate electronic energy levels

3. Local deformation follows.

Polarons typically have a low mobility because the process of hopping as described

previously takes more time than does moving a free carrier. Free carriers travel unbound

with a given drift velocity (vf) under the influence-of-an applied electric field (E) as seen









in Equation 2.3. In a periodic lattice, nearly-free electrons travel with a random drift

velocity. With no external electric field applied, the drift velocity nets out to zero. Drift

velocity is a function of the average time between scattering events (') and the mass of

the carrier (m) shown in Equation 2.3.

Ee
Vf = (2.3)
m
Free carriers have small masses lending to higher velocities and therefore higher mobility
given the same applied field assuming no increase in collision scattering.
v
P= (2.4)
E
When combined with a high carrier density, the nearly free carrier provides for a high
conductivity.
r= e- (2.5)
m
Assuming that a periodic lattice moves in time with some given displacement at a

frequency determined by the temperature, at any give time there is a probability that the

lattice will arrange itself to allow a charge carrier to move between sites. Lattice

distortion will follow. This occurrence is statistical in nature and has a finite probability

at a given temperature. A bound carrier must be activated by a discrete amount of energy

to hop from one site to another followed by the strain. Low mobility is a result of the

probability of hopping combined with the energy required to hop. So, for a polaron

conductor, the time between scattering events (c) may well be converted to a frequency

within a given time for a hop to occur. This jump rate relationship ties the motion of the

carrier to the phonon motion or natural frequency of the lattice.

2.3.2 Fundamentals of Free Carrier Conduction

Nearly the opposite of polarons, free carriers are not locked into a specific position

within the lattice. Free carriers typically have a high mobility until the carrier

concentration increases to the point that they begin to interact by colliding and scattering.


















Position
Figure 2-10. The particle in a box plot of a particle in a one dimensional lattice bound by
an infinite energy barrier. The particle is confined to the region inside the 1-D
well. The particle resides on a line and moves in a linear position (left and
right) direction bounded by the energy barriers on either end.

Modeling free carriers mathematically is simple and most often done with a simple

particle in a box model (the box term may be a misnomer when the system is one or two

dimensional). The assumption for the particle in a box is a single molecule or particle in

a one-dimensional position that can move linearly in two directions bounded by a

potential well on each side. The graph of this energy versus position looks like a box

seen in Figure 2-10, hence the name particle in a box. The energy for a particle in the

box is described by Equation 2.3, where En is the allowed energy, n is the principle

quantum number, h is Planck's constant, m is the mass of the particle and L is the length

of the one-dimensional box.

n2h2
En= (2.6)
8mL2
The associated normalized wave function for the charge carrier in the well is given

by Equation 2.7, where x is the position of the charge carrier in the box.


2/ (x)= 1 sin- (2.7)
L L
The solution to the wave function yields discrete nodes of allowed frequencies for

the particle (charge carrier). These nodes correspond to discrete energies that are allowed









within the confines of the box. The Pauli exclusion principle* dictates the energy levels

of the quantum states of the confined electrons (or charge carriers). Each will have a

unique set of quantum numbers though the energy value may be the same (degenerate).

When the density of electrons is large, energy levels corresponding to quantum states are

so close that the distribution is nearly continuous. This essentially continuous band of

energies is referred to as a conduction or valence band for electronic conduction and is

used to describep-type and n-type materials. With free carriers, the box size is much

larger than the crystal unit cell dimensions. This particular model matches classical

physics based on billiard ball collisions when the dimensions of the box approach the

centimeter scale. The picture fails when the deBroglie wavelength of particle is

reduced to atomic dimensions. (Emin, 1980). Modeling free carriers using theparticle in

a box is well accepted. A model of similar simplicity for polarons does not exist.

2.3.3 Observed Properties of Free Carriers and Polarons

While high conductivity is a characteristic of free electron carriers, for instance

degenerately doped indium tin oxide (ITO) conducts electricity by free electrons. Free

electrons absorb photons at wavelengths greater than -1-2 itm, but do not prevent high

transparency in the visible spectrum from 400-800 nm. It is because of this free carrier

absorption that n-type free carrier materials do not function as IR transmitting electrodes

but serve as low emission "low e" glass coatings for insulating windows (Svensson &

Granqvist, 1986). Light of energy less than -1 eV is easily absorbed and/or reflected as

the free carriers are promoted by the incoming photons to higher energy states that relax

and give off phonons or photons. The energy onset of this absorption activity due to

* The Pauli exclusion principle requires that each quantum state can be filled with at most two electrons
each with opposite spins (Hummel, 2001).









carrier excitation is referred to as the plasma frequency*. Polarons do not exhibit a

plasmon resonance near the same region because they are locally bound to the lattice.













(a) (b)
Figure 2-11. A polaron consists of the molecules involved and the local charge region:
(a) Liquid ammonia (colorless) forms polarons (dark blue) when sodium metal
is added (b) Molecular orientation of polaron in liquid ammonia. Shaded
areas are regions of high electron density (donated by sodium). **

As a dramatic demonstration of the effects of polarons on color in white light, the

property changes upon formation of polarons in liquid ammonia from electrons injected

by sodium metal pellets is illustrated in Figure 2-11. Polarons form as regions of high

electron density are created around the sodium ions. The polar ammonia molecules

preferentially arrange around the charged regions. This localized charge distorts bonding

and coupled with vibrational modes shifts the characteristic optical spectrum of the

molecule.

Temperature is a factor to consider when dealing with polarons and free carriers.

Free carriers and polarons in semiconductors behave differently as the lattice is heated.

Free electrons in metals will decrease in mobility with increasing temperature due to


* Plasma frequency is a characteristic frequency that separates the optically reflective region from the
transparent region. The dielectric constant goes to zero and conditions are right for plasma (fluid-like)
oscillation for the entire electron gas.

*Video contained in a separate file (PolaronClipl.avi).









carrier collisions and scattering with a net result of decreased conductivity.

Semiconductors behave in two different ways depending on their doping regime. The

intrinsic semiconductor will decrease in mobility and consequently conductivity until at a

high enough temperature thermal energy excites electrons from the valence band to the

conduction band. Carrier generation will allow conductivity to increase up to a certain

threshold value where carriers begin to collide and scatter causing an overall decrease in

conductivity due to the reduction in mobility even as the carrier density increases.

Thermal generation of carriers from extrinsic dopants will also affect the conductivity

due to dopant ionization and increased carrier density until the point of saturation. The

increased temperature will no longer increase carrier density beyond the saturation

concentration, but enhance carrier collision and scattering will reduce mobility and

decrease conductivity with increased temperature until the material begins intrinsic

carrier generation. Materials in which conduction is limited by hopping experience the

opposite effect due to the high population of bound carriers (polarons). As temperature

increases, the lattice vibrations increase and the number configurations per second that

allow hopping. The time for a carrier to hop is believed to remain constant with

increased temperature, but the opportunities for hopping increase. Polarons ideally

experience a mobility increase with temperature due to increased hopping. Carrier

density remains constant so the net effect is an increased conductivity with increased

temperature. When a polaron-conducting material is compared to a free carrier

conducting material at higher temperature, the two are distinguished on the basis of

mobility and carrier density.










Table 2-6. Observed properties of polarons compared with free carriers.
Observed property Polaron Free carrier
Observed in transition-metal Metals; n-type orp-type
oxides with ions in multiple semiconductors
oxidation states
Seebeck coefficient Sml < -a Large 0.1 1 mVK-' Positive for
Small < 20iVK a
Value p-type and negative for n-type.
Seebeck coefficient Nearly independent a Dependent, decreases with
Nearly independent. .
temperature dependence increased temperature.
Conductivity value Good 10-1000 Scm 1.a Better 1000+ Scm-'.e
Conductivity temp. Related to exp (-Ea )/(kT) Degrades; Plateaus with
dependence where Ea is about 0.2 eV. b temperature then decreases.
1 -3b Medium to high 101802 cm
Carrier density value High -102 cm3.b Medium to high 101821+ cm-3
Increases with temperature
Hall mobility Small << 1 cm2V s-'.a > 10 cm2V-'s-1
Hall mobility and
l m Increases with temperature. Falls as temperature increases.
temperature
Promoted by structural
Carrier density ordere, b Increases with temperature.
disorder.
Hall sign Anomalous. a' c p-type positive, n-type negative.
a(Windisch et al., 2002b)
b(Windisch et al., 2002a)
C(Emin, 1982)

The Seebeck coefficient* changes significantly with temperature for free-carrier

conductors but only a negligible amount or not at all for a polaron conductor. This nearly

temperature-independent behavior is one of the key indicators of polaron hopping. Table

2-6 summarizes some properties of polaron hopping versus free carrier conducting

materials. Hall measurements show an increase in carrier density with temperature and a

decrease in mobility. Seebeck measurements give a higher value for carrier density that

is not influenced with temperature. In the Hall measurement, carrier count is measured

and mobility is calculated.


SA measure of the voltage change with temperature in a material from heat driven diffusion gradient of
carriers in the material. If the change is positive it indicates holes are the majority carrier, if negative, then
electrons are dominant.









Nickel-cobalt oxide is reported to conduct via a small polaron hopping mechanism

(Windisch et al., 2002b). The key considerations for small polaron hopping, shown in

Table 2-7, verify that nickel-cobalt oxide is a polaron conductor.

Table 2-7 Key characteristics of small polaron hopping for nickel-cobalt oxide


Characteristic Observed?
Observed in oxides with ions in more than one Yes Ni(+2,+3), C(+2,+3) a
oxidation state
Small Seebeck coefficient 20 pVK1 b
Small charge carrier mobility < (0.1 cm2V s-)b
Conductivity related to temperature Yes b
S=exp (-Ea )/(kT) where Ea is about 0.2 eV
High carrier density Yes, approaching lx(10)22 b
Seebeck coefficient is nearly temperature Yes b
independent
Correlates with phonon behavior Yes: XRD & Raman c
Promoted by structural disorder Yes c


a(Roginskaya et al., 1997; Tareen et al., 1984)
b (Windisch et al., 2002b)
c (Windisch et al., 2002a)

Temperature dependent measurements of the Hall effect show mobility increasing

with temperature while the carrier concentration remains high for polaron conducting

films. Conductivity is due to the high density of carriers in spite of low mobility making

conduction possible because each carrier hops to contribute to the net current. Activation

energy of electrical conductivity is influenced by temperature. Mobility increases as

activation energy decreases with increased temperature. Conductivity increases with

temperature because mobility increases with temperature. Based on observed properties,

a polaron model will be introduced that will relate charge disorder to activation energy

for electrical conduction in the nickel-cobalt oxide system.

2.4 Nickel-Cobalt Oxide Spinel Structure

Transition metal oxides take on a variety of structures including the delafossite,

perovskite, pyrochlore, smectite, sodalite, stibiconite, and spinel structures (Fleischer,









1995). Cations stack in a solid based on a preferred nearest neighbor arrangement.

These arrangements created from the stacking orientation determine the nature of the

structure and may influence the conduction mechanism. Ternary compounds often

arrange in two common structures: perovskite and spinel. (Smyth, 2000) A third more

recently studied ternary arrangement is the delafossite structure. (Kawazoe et al., 2000)

This section provides a brief background of the spinel structure and its characteristic

arrangements in the nickel-cobalt oxide system.





















Figure 2-12. Green tetrahedral cations (A), yellow octahedral cations (B), and black
oxygen anions (0) form the spinel unit cell with a chemical formula of AB204
(based on atomic layout (Kingery et al., 1975, p.65).

Spinel, a gem found in nature often mistaken for Ruby, contains magnesium,

aluminum and oxygen (Gem by gem, 2003; Hughes, 1999; Sickafus, 1999)). The

arrangement of cations and oxygen in the gem spinel, shown in Figure 2-12, is the

common crystallographic arrangement for a number of other minerals and oxides.

Crystallographers refer to other similarly structured systems as having the spinel









structure. The spinel group of natural crystalline materials includes mineral such as

spinel (MgAl204), chromite (FeCr204), coulsonite (FeV204), gahnite (ZnAl204),

hercynite (FeAl204), magnesiochromite (MgCr204), magnesioferrite (MgFe204),

magnetite (Fe304), trevorite (NiFe204), zincochromite (ZnCr204). Another isostructural

mineral is the linnaeite group with sulfur instead of oxygen occupying the anion sites.

(Fleischer, 1995) This sulfide containing linnaeite structure however, is often also

referred to as spinel (Ballal & Mande, 1977; Benco et al., 1999; Kishimoto et al., 2000;

Ohmuro et al., 1995; Schellenschlager & Lutz, 2000).

36.700
(100)
100%




M 31.150 64.960
59.110
t (220) 44.63 (511) (40)
1 18.930 30% (400) 24% 30%
(111) 38.400 20% 55.430
13% (222) (422)
9% 7%

20 30 40 50 60
2-Theta (0)
Figure 2-13. A calculated diffraction pattern for NiCo204 annotated with peak positions,
planes, and relative intensities from the powder diffraction file database
number 73-1702.

As nickel is added to cobalt oxide to make the NixCo3-xO4 mixture, the spinel

structure from the cobalt oxide persists up until cubic nickel oxide precipitates out. The

most conductive mixture of nickel-cobalt oxide is spinel structured (Windisch et al.,

2002b). Figure 2-13 shows a calculated XRD pattern of the spinel structure.









The atomistic arrangement in the spinel structure accommodates multiple cations in

multiple oxidation states and allows hopping conduction (Windisch et al., 2002b). While

the literature agrees that the bonding nature between atoms in the spinel structure is

mostly ionic, a complete understanding of the bonding is subject to controversy.

(Azaroff, 1960).

Grazing incident x-ray diffraction (GIXRD) from both sputtered and solution

deposited films exhibit a spinel pattern, similar to the powder diffraction file (PDF)

number 73-1702 for nickel-cobalt spinel shown in Figure 2-13. This structure is

characterized by a primary peak from the (100) plane at 36.700 (2-theta) and secondary

peaks from (220), (400), (511), and (440) at 31.150, 44.630, 59.110, and 64.960 (2-theta)

respectively.

Nickel addition to cobalt oxide maintains the spinel structure up to its solubility

limit when it separates out to form the NiCo204 spinel and a separate cubic NiO phase

detectable by XRD. According to Windisch et al. this occurs when nickel concentrations

exceed 33% in solution deposited samples and 50% in sputter deposited films. All

samples experience nickel oxide precipitation when heated above four hundred degrees

(Tareen et al., 1984; Windisch et al., 2001a), however Petrov and Will demonstrated that

heating to 10000C in oxygen and KC103 at a pressure of 60kbar reversed the phase

separation to form spinel structured Nil.71Co1.2904 (Petrov & Will, 1987).

2.4.1 Spinel Sites

The unit cell from Figure 2-12 is made up of a series of repeating layers illustrated

in Figure 2-14. Using the nomenclature AB204 to represent cations and oxygen in the

normal spinel structure, the A represents the doubly ionized cation found in the








tetrahedral site, the B represents the triply ionized cation found in the octahedral site, and
"O" represents the oxygen occupied sites according the space group Fd3m #227.
(Azaroff, 1960)

1 T-O f 2




i0 0



Mb 0
3 4 4- 0

-4 o1f 0


Figure 2-14. Spinel unit cell layers stack with interlocking tetrahedral sites. Oversized
green tetrahedral atoms (A) in the numbered layers stick up and connect in the
vacancies of the level above. This layered schematic is based on a similar
spinel layout detailed by Kingery. (Kingery et al., 1975) p.65.
Tetrahedral site. The tetrahedral site (Figure. 2-15), or four-fold coordination site
accommodates a cation which bonds with four nearest neighboring oxygen anions. Eight
out of 64 possible tetrahedral sites are occupied by a doubly ionized cation. Cobalt is
believed to occupy the tetrahedral sites in the nickel-cobalt oxide lattice.




















Figure 2-15. Tetrahedral site atom (green) bonded to four surrounding oxygen (black).





4O







Figure 2-16. The octahedral site atom (yellow) bonds with six surrounding oxygen
(black) anions to form a six fold orientation.
Octahedral site. The octahedral site (Figure 2-16), or six-fold coordination site,
typically is occupied by a cation with a 3+ valence to bond with its 6 nearest oxygen
anions. The spinel unit cell contains 32 octahedral sites, of which 16 are occupied.
Nickel prefers the octahedral site in the nickel-cobalt oxide (Tareen et al., 1984), and
cobalt fills the remaining octahedral sites and then the tetrahedral sites.









The octahedral site plays a role in conductivity. As nickel is added to cobalt oxide,

the conductivity increases by up to five orders of magnitude and nickel occupies the

octahedral site. However after heat treatment, nickel oxide precipitation causes

conductivity to decrease, suggesting that nickel in the octahedral site must have a

significant effect.

Oxygen site. Oxygen occupies the remaining thirty-two anion sites. Oxygen sites

are nearly a close packed cubic lattice (Kingery et al., 1975).

2.4.2 Variations of the Spinel Structure

Not all spinel structures arrange as noted by the AB204 with the A2+ in the

tetrahedral site and the B3+ atoms in the octahedral site. The inverse spinel sometimes

occurs, where half of the B3+ species occupy the tetrahedral sites. The remaining half of

the B3+ cations along with all of the A2+ cations occupy the octahedral sites. Inverse

spinel structures are often identified structurally as B[AB]O4 (Azaroff, 1960). Other

distributions of cations between the various lattice sites are possible (Deer, 1962; Kingery

et al., 1975).

"The partially inverse spinels can be viewed as partially disordered versions of
either end member, the ideal normal spinel or the ideal inverse spinel. The
disordered structures contain lattice defects relative to either end member in that
cations are located on lattice sites where they do not appear in the ideal reference
structure." (Smyth, 2000, p.20)

Nickel and cobalt compose an oxide spinel where both cations are found in

multiple valence states and in multiple sites. This condition of a mixture of multiple

valence states in multiple sites is referred to as cation disorder. Verwey et al. postulated a

cation charge ordering within iron oxide at a low temperature (Verwey 1939; Verwey &

Haayman, 1941; Verwey et al. 1947). Below the Verwey temperature, conductivity

decreases dramatically (Monk & Ayub, 1997). Much like Verwey's postulate, cation









arrangement must have an effect on polaron interaction. Hopping between sites will

depend on whether the sites are dissimilar or both tetrahedral or octahedral. Arranging as

primarily an inverse spinel (Windisch et al., 2001b), the nickel and cobalt cations may

substitute for each other at both the tetrahedral and octahedral lattice sites. If

arrangement has an effect, then the opposite must be true as well. Defects and disorder of

the cation charge distribution within the spinel structure also has an effect on the

conductivity (Windisch et al., 2002a).

2.4.3 Spinel and Conductivity

The electrical conductivity for these spinels is generally very limited. In general

their bandgaps are large (>2.5 eV) and impurities form deep traps for charge carriers. To

understand the limited observed conductivity, the electronic bonding must be considered.

As reported by Smyth et al., for the iron oxide system the following is observed:

"In an octahedral environment, the crystal field splitting gives three equivalent
levels of lower energy, the t2g levels whose orbitals are directed in between the six
closest anions, and two equivalent levels of higher energy the eg levels whose
orbitals point directly at nearest neighbors. The energy separation in this case is
sufficient to promote a violation of Hund's rule (that "equivalent" electron states
are singly filled before any of them are doubly occupied), and the six d electrons
fill the three lower levels." (Smyth, 2000, p.18).

Smyth concludes that electrical conductivity is achieved by carrier movement

through the bonding orbitals or t2g orbitals along the octahedral sites. Conduction may be

due to movement of holes in anionp-orbitals. (Ballal & Mande, 1977)

2.5 Summary of Literature Review

TCOs are classified by their majority carrier type or conduction mechanism. Both

n-type andp-type TCOs are needed, but n-type are more developed and available. The

primary reasonp-type TCOs are not as developed is a result of the conductivity being

orders of magnitude less than n-type TCOs. Nickel-cobalt oxide conductivity is order of









magnitudes better than otherp-type TCOs, yet is still poor in comparison to other well

used n-type TCOs. N-type TCOs block light in the IR while nickel-cobalt oxide

transmits it. N-type TCOs are free electron conductors with high mobility and nickel-

cobalt oxide is a polaron hopping conductor with low mobility. Highly localized charge

carriers and the accompanying lattice strain known as small polarons are found in nickel-

cobalt oxide and behave distinctly different than the free carrier analog in other oxide

semiconductors. Most often polaron hopping is "discovered" as the conduction

mechanism by observing material properties. Key differences between free carrier

conductors and small polaron conductors are evident from temperature dependent

properties such as conductivity and the Seebeck coefficient. Conductivity becomes

significant in small polaron conducting material when the carrier concentration becomes

extremely large. Polarons conduct through a charge hopping mechanism. Polaron

formation does not create a plasma absorption region therefore IR transmission remains

high. Properties exhibited by nickel cobalt oxide such as high conductivity and infrared

transparency result from the spinel structural arrangement and the high polaron

concentrations. Hopping is a low mobility process that requires energy for activation.

Achieving conductivity improvements is approached in three ways:

Increase the cation disorder and lower activation energy by adjusting film

deposition conditions and post deposition heat treatment processing.

Add polarons by doping the system with a monovalent impurity atom such

as lithium to further oxidize metallic ions and create more polarons

Substitute rhodium for cobalt to increase disorder by atomic size distortion.

These methods of enhancing conductivity by are reported in the Chapters 3-5.














CHAPTER 3
SPUTTERED NICKEL-COBALT OXIDE

3.1 Introduction

This chapter discusses the electrical and optical properties of nickel-cobalt oxide

and how sputtering conditions and heat treatment conditions affect them. Changing the

growth rate or nucleation mode (growth mechanism) by adjusting sputtering conditions

may influence the film qualities such as crystallinity, morphology or density (Mattox,

1998). The effects of introducing a third cation such as lithium or rhodium on thin film

electrical and optical properties are discussed in Chapters 4 and 5, respectively.

Two primary methods used to deposit nickel-cobalt oxide as infrared transparent

conducting oxide (ITCO) thin films include solution deposition and sputtering.*

Conductivity and transparency both vary depending on the method of film deposition.

The solubility limit of nickel in cobalt oxide appears to be a function of the deposition

method. Nickel solubility in cobalt oxide is enhanced by sputtering and superior film

conductivity is observed at increased nickel concentrations. Solution deposition is

limited to a maximum concentration of 33% nickel to produce the stoichiometric

NiCo204 spinel composition. Sputtered samples allow a nickel concentration of up to

50% to give Nil.5Col.504. Windisch et al. have explained the superior conductivity

(order of magnitude improvement) of sputter deposited films in comparison to solution

deposited films (Windisch et al., 2001a; Windisch et al., 2001b; Windisch et al., 2002b).



*Refer to Appendix A for additional details of sputtering and solution deposition









A technique of sputter deposition that produces a compositionally varying film

(combinatorial sputtering) used in this study produced a finer range of film compositions

to find the highest conductivity composition. This combinatorial sputtering deposition

technique was repeated to study the effect of select gas compositions on deposited film

properties. Traditional sputtering of Nil.5Co1.504 and NiCo204 from alloy targets show

how process conditions of film deposition such as to gas pressure and target-substrate

distance affect electrical, optical and structural properties. Heat treatment methods are

shown to have a dramatic effect on film properties as well.

3.2 Film Preparation and Characterization Procedures

Radio frequency (RF) sputter deposition at 13.56 VMHz with a magnetron cathode

allows tight control of process variables such as cathode power, substrate temperature,

gas flow rate, gas composition, gas pressure, and target-substrate distance inside the

vacuum chamber to produce consistent films over large areas. This study analyzes films

sputtered with process adjustments such as sputtering gas composition, target-to-substrate

distance, and gas pressure. Resulting optical transmission and electrical conductivity will

be reported as a function of these sputtering process conditions. Sputtering for this study

occurred in one of two configurations: (1) the combinatorial film deposition method used

two cathodes simultaneously to generate many compositions in a single run, and (2) the

traditional sputtering method used a single cathode and a rotating substrate holder.

The first sputter deposition configuration uses two cathodes simultaneously to

produce a film with varying composition. Combinatorial sputtering in Figure 3-1 (also

referred to as combinatoric sputtering (Freeman et al., 2000)) produced films containing a

graded composition between nickel oxide and cobalt oxide in an attempt to find the

composition of a nickel-cobalt oxide film with the highest conductivity. The basic setup









included three microscope slides placed end-to-end, divided into nine one-inch sections,

and labeled 1-9, as shown in Figure 3-1. All sputtered films were deposited with a base

vacuum pressure near 1x10-6 Torr. The process gas pressure was held constant at 2

mTorr for the combinatorial experiments.

1 2 3 4 5 6 7 8 9


Substrates 7cm
1-7.5cm-*
22.5 30 cm Target to
Substrate Distance
0 c30m 10











Figure 3-1. Combinatorial sputtering uses a dual cathode setup with targets of different
material composition. The resulting compositionally graded film is divided
into subsections and numbered 1-9. Sapphire, microscope slides and/or
silicon wafer slices served as substrates for combinatorial deposited films.

Factors such as background contamination within the chamber (e.g. pump oil,

previously sputtered material, or gas line impurities) may cause slight film variation from

one process run to an identical one. By using the combinatorial technique, an array of

compositions can be fabricated simultaneously with identical deposition conditions to

minimize incidental error from one run to the next. Combinatorial sputter deposition

does have its own complexity concerning uniform film thickness attributed to the

geometry of the setup and the different materials used as targets. Uneven film thickness









is minimized by adjusting the sputtering rate of the individual cathodes by depositing a

film and then changing the cathode power and verifying the effect. During calibration,

all other process parameters are held constant.

The second sputtering configuration is called traditional sputtering and includes a

single two or three inch cathode powered at 100 or 200 watts, respectively. Process gas

pressure and target-substrate distance are also varied. Sputtering time is adjusted to

produce films of similar thickness from setups with the varying target-substrate distance.

The substrate holder setup for single cathode deposition rotated primarily in an offset

rotation setup where the axis of rotation is offset from the center axis of the cathode (see

appendix A for clarification on sputtering geometry). Samples of NixCo3-x04 were

produced on fused silica, p-type silicon, sapphire, poly(ethylene-teraphthalate) (PET),

and microscope slides with x being equal to 1 or 1.5.

Characterization results consist of data from both combinatorial and traditional

sputtered films. The matrix for the experiments is shown in Table 3-1.

Table 3-1. Film deposition parameter and sputtering setup matrix for NixCo3-x04
Parameter Setup Variable range

Film Composition Combinatorial 0.8 < x < 1.75
Gas Composition Combinatorial 0, 20, 50, 75, 100% 02

Gas Pressure Traditional 2, 5, 10 mTorr; x=l, x=1.5
Target-Substrate Distance Traditional 7.5, 10, 15, 30 cm; x=1.5


Characterization included x-ray photoelectron spectroscopy (XPS) *(Brundle et al.,

1992), secondary ion mass spectroscopy (SIMS)* (Brundle et al., 1992) transmission


See Appendix B for instrumental setup parameters and operating conditions.









electron microscopy (TEM) *(Brundle et al., 1992), x-ray diffraction (XRD)* (Brundle et

al., 1992), Fourier transform infrared spectroscopy (FTIR)* (Brundle et al., 1992), ultra-

violet and visible spectroscopy*, optical and stylus profilometry*, and room-temperature

and temperature-dependent van der Pauw measurements. Heat treatment consisted of a

ten minute post-deposition heat treatment at 3750C followed by rapid cooling.

XPS was used to determine the chemical composition of select films and then

correlate the other characterization information to a specific composition. Electrical van

der Pauw measurements were performed with films on all substrates, while FTIR

transmission measurements were performed with films on silicon and sapphire substrates.

Visible and near-infrared optical spectra were measured from films on fused silica and/or

sapphire substrates.

3.3 Characterization Results and Discussion

3.3.1 Electrical Properties

Variable composition films from combinatorial sputtering show the effect of

process gas composition while traditionally sputtered films from alloy targets show the

effects of total chamber sputter gas pressure and target to substrate distance on electrical

properties. Substrate material has an effect on the produced film regardless of the film

composition and the work function does not appear to vary a significantly with

composition.

Gas composition. The effects of gas composition, with the gas mixtures ranging

from 100% oxygen to 100% argon, at a constant pressure of 2 mTorr are shown in Figure


See Appendix B for instrumental setup parameters and operating conditions.









3-2. Films deposited at constant gas pressure with different gas concentrations of argon

and oxygen show that position 7 had the highest conductivity of -350 fcm.

1000




1
lOOO---------------







S0.1 100%Ar
o 080% Ar 20% 02
0.01 50% Ar 50% 0,
25% Ar 75% 02

0.001 100% 02
1 2 3 4 5 6 7 8 9
Position
Figure 3-2. Conductivity versus position for combinatorial runs is displayed as a function
of oxygen and argon gas composition. The film deposited with the gas
concentration of 50% oxygen and 50% argon is the most conductive at
position 7.

Combinatorial sputtered films from nickel oxide and cobalt oxide targets deposited

with variable gas composition showed marked changes in resistivity as the amount of

argon increased up to 50%. The gas with 50% oxygen and 50% argon yielded the highest

conductivity across all positions. Resistivity degraded when the argon concentration

exceeded 50%. Gas composition appeared to affect the sputter rate and possibly the

oxidation state of the cations in the produced film. Argon, with its larger mass, inflicts a

greater amount of damage on the target, increasing the sputtering rate compared to

oxygen, likely producing a slightly reduced film. Oxygen however promotes complete

oxidation of the film. Combining the effects of oxygen and argon produced the highest









conductivity films. Figure 3-3 shows position 7, the highest conductivity position of all

the films (-350 Qcm), plotted as a function of gas pressure.








200-
Q 0 20 40 60 80 100
Percent Argon in Oxygen Process Gas
Figure 3-3. Position 7 from Figure 3-2 is the position with the highest conductivity.
Conductivity is enhanced at a sputter deposition process gas composition
containing 50% oxygen and 50% argon.

This combinatorial study was conducted to fill the gaps between the data points

previously reported (Figure 2-8) where the composition with highest conductivity film

contains equal parts of nickel and cobalt. Combinatorial sputtering with 5 cm nickel

oxide and cobalt oxide targets produced a film with a graded composition as determined

by XPS (Figure 3-4).

Possibly due to different magnetic species in the plasma, more cobalt is measured

in the sample in Figure 3-4. An ideal deposition would show an equal ratio of nickel to

cobalt at position 5. It is interesting that while proportionally more cobalt is in this film,

shifting the equal ratio composition Nil.5Co1.504 to the nickel-rich side, the data show

that the nickel-rich side of the film is actually thicker. This suggests that the nickel oxide

target and the cobalt oxide target sputtered at different rates. Nickel oxide, being a

nonmagnetic target, likely had a higher deposition rate, but being influenced by the

magnetic field of the cathode, deposited material in a more confined region. The

magnetic nature of the Co304 target may have interacted with the magnetic field from the









cathode to alter its effect on the desired ion trajectory and produce a more diffuse

deposition resulting in a lower sputter rate and a higher area of coverage.


g NiCo204 Nil .sCO.504
= 1.5


Cobalt Increasing x -- Nickel
0.5

0
1 2 3 4 5 6 7 8 9
Substrate Position
Figure 3-4. Fraction of Ni and Co detected by XPS as a function of position on a
combinatorial sputtered NixCo3-x04 film. Equal portions of nickel and cobalt
were found at position 7 (x value of 1.5). The blue arrow indicates the
solubility limit of nickel from solution deposition. The light blue shaded
region shows the increased nickel available from sputtering. Embedded along
the abscissa of the graph is a picture of a combinatorial sputtered film on a
silicon substrate from this study. The color change is due in part to thickness
and in part to compositional variations.

XPS of the film combined with resistivity and thickness data, each as a function of

numbered position, yields a relationship between composition and conductivity. XPS of

the most conductive combinatorial film is shown in Figure 3-5. Film conductivity is

found to be highest at a composition of Nil.5Co.504 at -350 Scm-1, agreeing with

previously reported results (Windisch et al., 2001a). The composition limit imposed by

solution deposition is included for contrast. All compositions to the right of the red

vertical line are not possible from solution deposition.

As an indication of the different sputtering rates of nickel and cobalt, a thickness

calibration profile shows that the nickel-rich side is thicker than the cobalt-rich side

(Figure 3-6). Interference due to the magnetic nature of the cobalt oxide target may

contribute to the non-uniform deposition thickness. Sputtering RF power and tilt angle of









the individual cathodes would require fine tuning to flatten the thickness curve. This

iterative process of finite adjustment should be addressed in a future study.


500

g 400

300

| 200

100


.8 1 1.2 1.4 1.6
Composition (x)


Figure 3-5. Combinatorial sputtered NixCo3-xO4 film combining conductivity and
position with composition with position data to yield conductivity as a
function of composition. The highest conductivity film appears at a
composition near Nii.5Col.504. The vertical red line at x=l indicates the
solubility limit of nickel when deposited from solution methods.


Run 1 Co304-100W, NiO-100W, 2 Hours
--- Run 2 Co0O4-125W, NiO-25W, 4 Hours
1201


S100-
N 80
S60
S4O
H 20
0


Cobalt Nickel

1 2 3 4 5 6 7 8 9
Position


Figure 3-6. A combinatorial substrate thickness profile measured by a stylus
profilometer shows that the film thickness is also graded from cobalt rich on
the left to nickel rich on the right. Followings an initial calibration run,
cathode power was adjusted as was deposition time to give a more uniform
film. Subsequent combinatorial films were processed using this calibrated
power regime.

Figure 3-6 shows the initial calibration run (run 1) with a non-uniform thickness of

about 30 nm on the cobalt side and 100 nm on the nickel side. The power was adjusted









and the second deposition ran with a 25% increase in the cobalt cathode power and the

nickel cathode power reduced by 75%. Sputter time was doubled to ensure a film of

similar thickness. No further increase in the cobalt cathode power was practical because

a general rule of thumb for sputtering (specified by one cathode manufacturer)

discouraged exceeding a power rating of 4 Wcm-3 on a 5 cm cathode. Since the thickness

difference was now a factor of 2x instead of nearly a factor of 4x, it was deemed

acceptable.

Chamber pressure. Using the composition from the combinatorial run with the

lowest resistivity, an alloy target with equal parts of nickel and cobalt was reactively RF

sputtered in 100% oxygen at 10, 5, and 2 mTorr in the traditional setup with an offset

rotating substrate holder at a distance of 10 cm. A mixed gas composition was not used

for reactive sputtering. Films in Figure 3-7 deposited at 2 mTorr maintained lower

resistivity before and after heat treatment. Note that the variance is less than a factor of

two and that the best resistivity is near 2 mQcm (a conductivity of 500 Scm-1).


26
C As Deposited
5
"*- After Heat Treatment
14
3 ---



10 mtorr 5 mtorr 2 mtorr
Figure 3-7. The effect of sputtering gas pressure on NiCo alloy sputtered Nil.5Coi.504
thin film resistivity before and after 10 minutes at 3750C heat treatment.

While the results for the as-deposited samples are not in complete agreement, the

general trend shows that the lower pressure in the range studied is better increase









conductivity. The result of decreased resistivity at a lower pressure this system exhibits

is not surprising, but was previously unknown.

Molecules in a vacuum travel an average distance between molecular collisions

which changes with pressure. This molecular mean free path (k) in Equation 3.1 has

units of centimeters when pressure (P) is in units of Pascals (O'hanlon, 1989) and is

determined by the number of molecules occupying the swept out volume. Higher

pressure means more molecules per unit volume and a lower mean free path.

A=6 (3.1)
P
In the vacuum chamber this becomes important as the molecules being sputtered

from the target surface collide with and lose energy to other molecules before reaching

the substrate for film formation. Pressure changes directly influence this mean free path.

The experimental parameter of distance between target and substrate may also affect the

energy of depositing atoms in a similar fashion.

,35



15

S5 <----

5 10 15 20 25 30
Target-Substrate Distance (cm)
Figure 3-8. Target-substrate distance affects the resistivity of sputtered nickel-cobalt
oxide thin films from a NiCo alloy target. Resistivity increases with distance.

Target-substrate distance. Increasing target-substrate distance (Figure 3-8)

allows a larger area of deposition and enables uniform coating of multiple substrates with

the proper substrate motion (see Appendix A for details on planetary rotation). Increased

target-substrate sputtering distance results in increased resistivity. Discussion of the









effect evident by these data follows in the Optical Properties Section (3.3.2) and in the

Film Structural Properties Section (3.3.3).

Substrate material. Substrate material in some cases had an effect on

conductivity. Poly(ethelyne-terapthalate), a polymer substrate used for OLEDs, had a

detrimental effect on film conductivity. It is believed that carbon from the PET substrate

reduced the nickel-cobalt oxide film somewhat, which resulted in the observed decrease

in conductivity.

450

350

1 250-

150--

50'

M. Slide Fused Silica Sapphire Silicon PET
Substrate Material
x=1.5 10mtorr x1 x=1.5 5mtorr x=1.5 2mtorr
x=1 l0mtorr x=l 5mtorr x1 2mtorr

Figure 3-9. Conductivity ofNixCo3-xO4 films on PET substrate is always less than films
on other substrates by a factor of 2-4x for as-deposited samples for all
pressures and compositions shown.

Work function. Sputter condition changes to nickel-cobalt oxide may also affect

the work function, a property not reported in the literature. From UPS data, a work

function was calculated by with Equation 3.2. An ultraviolet He I source with an energy

of 21.218 eV (+ or 0.001 eV) excites electrons across the band gap. A bias voltage

(VB) of two different values was applied and the energy (EvBcutoff) is extrapolated from

Figure 3-10 (energy on the left side of the peak where intensity is zero). The work

function is calculated by taking the known photon source energy, subtracting the bias









voltage, and then subtracting the measured valence band cutoff value. The detector work

function is important to know also, but is calibrated out of this data and therefore not

included in Equation 3.2.

= hv V -Ev E ff (3.2)

NiCoO24 -15V Bias
S- -NiCoO24 -9.1V Bias
S\ Nil. Co.504 -15V Bias
S-Ni1.5Co.i504 -9.1V Bias

S\ Nil.sCoL504 Work Function 4.27 eV

NiCo2O4 Work Function -4.39 eV






10 5 0 -5 -10 -15
Voltage (eV)
Figure 3-10. UPS work function measurement of NiCo204 and Nil.5Co1.504 films.
Higher nickel content shows a slightly lower work function. Work function
values may track inversely with conductivity or increased nickel content.

Compositional differences of oxide films from sputtered NiCo and NiCo2 alloy

targets does not appear to have a dramatic effect on the work function, These results

suggest that the work function may vary slightly with composition or conductivity.

Assuming the variance is with conductivity, the work function would change with

processing conditions. Additional work would be required to quantify this assertion.

However, given the small change due to composition in the work function it is doubtful

that processing conditions such as sputtering pressure or sputtering distance would have a

significant impact on the work function value.









3.3.2 Optical Properties

Changes in processing of TCO films affects optical properties though with the

opposite general trends of the electrical properties. Variation of optical transmission as a

function of film composition has been reported previously (Windisch et al., 2001b).

Incident light impinging on a sample will become a sum of several interactions,

including transmission, absorption, reflection, and scattering (Equation 3.3). Often

absorption and scattering are assumed to be zero and the calculation simplifies to three

terms.

0 = Itran + Iabs + Irefl + scatter 0 = Itra,, refl (3.3)
Gas composition. Due to the thickness variation in the combinatorial sputtered

films shown in Section 3.2.1, variations in transmission data will not necessarily

represent changes with film composition or gas composition, so such data have been

excluded from the optical property analysis. The optical transmission measurements

would likely yield useful information if the films are of uniform thickness.

Chamber pressure. For films of the same composition deposited at different

pressures, optical transparency was seen to vary from 5-10%. Transparency decreases

with pressure (Figure 3-11) and increased conductivity, agreeing with reported

conductivity and transparency relationships (Windisch et al., 2002b). Lower pressures

from both compositions (x=1.5 and x=l) exhibit lower transmission. The absence of

correction for silicon substrate limits the maximum transparency to 50%, but even with

correction, the effect would be similar. Films on silicon substrates, shown in Figure 3-11,

show that the transparency of the nickel-cobalt oxide films in the near- to mid-infrared

regions (corresponding to light wavelengths of 2.5 [m to 25 [m) if normalized to 50%

would be well above that value.










e -- 10 mtorr 5 mtorr 2 mtorr


40
30


0
0x= 1.5 x= 1

S 3400 2400 1400 3400 2400 1400
Wavenumber (cm-') Waavenu nber (cm-1)

Figure 3-11. FTIR traces from of reactively sputtered NixCo3-xO4 thin films. (a) x = 1.5
(from NiCo alloy target) and (b) x = 1 (from NiCo2 alloy target) show the
effect of pressure on transmission. Films nominally 50 nm thick are shown as
a function of sputtering gas pressure.

The decrease in transparency seen with decreased sputtering pressure may be due to

a change in the index of refraction of the deposited film likely caused by a chemical or

structural change induced by the processing conditions. Changes in the index of

refraction may result in a 3% change in transmission due to an increased

absorption/damping region likely attributable to defects in the lattice.

The index of refraction shown in Equation 3.4 (n), is the ratio of the speed of light

in a vacuum (c) divided by the speed of light in the material (v).

c
n=
v (3.4)

A difference in index of refraction of two different materials across an interface

results in a loss of transmitted light due to reflection at the interface. The reflected

intensity is expressed in Equation 3.5 where nl and n2 are the indices of refraction for the

two materials. Percent reflection is the intensity of reflected light (Irenf) divided by the

incident light (Io) multiplied by 100. This expression is valid for transparent materials

and neglects damping. Complex values ofn could be used in the expression to take

damping into account.










i= I -n 12 (3.5)

Interface

Film Substrate
n, n,
Air n=l Air n=1




Photon Path


Film-Air Interface Substrate Air Interface
Figure 3-12. Optical transmission through a thin film. On a substrate, it requires a low
index of refraction mismatch at all interfaces (to avoid complete reflection)
and a low damping coefficient (to avoid extinction) for the particular
wavelength of transmitted light (this picture does not show the effects of
damping).

The index of refraction is important for TCO films because the light must pass

through three interfaces as illustrated in Figure 3-12. A significant difference in index of

refraction at any of these interfaces will result in increased optical reflection or damping

with an associated loss in transmission. Losses due to reflection from index mismatches

of nickel-cobalt oxide (n = 2.6-2.8) on a silicon substrate (n = 3.5) exceed 50% when

damping is assumed to be zero. Absorption or damping mechanisms may include

electron excitation with phonon emission (activation of lattice vibrations) and in reality

are not zero, but are assumed to be so in Table 3-2.

Assuming that the process pressure causes a shift in the index of refraction from

2.6-2.8, the shift in transparency would be less than 3 percent (see calculation in Table 3-

2). A shift greater than 3% could be attributed to a difference in thickness or possibly

scattering from process-induced morphology changes, such as increased grain size or

grain boundary area.









Table 3-2. Values of index of refraction for films and substrates are shown individually
with calculated transmission losses due to interface index mismatch assuming
no absorption or extinction.
n of Film Substrate n of Substrate % Loss in
Transmission
NiCo204 2.6 Si 3.5 (1370 nm) 52.8
NiCo204 2.7 Si 3.5 53.6
NiCo204 2.8 Si 3.5 54.5

NiCo204 2.6 Si02 1.46 (300-800 nm) 31.13
NiCo204 2.6 A1203 1.77 (300-5000 nm) 26.85


Target-substrate distance. The distance of sputtering between the target and the

substrate also had an effect on the optical transmission from which an optical absorption

coefficient is calculated. The absorption coefficient (ca), given in Equations 3.6 and 3.7

multiplied by the film thickness (d) is proportional to the intensity of light (Itrans) detected

after the beam passes through the film:

Itrans 0e( ad) (3.6)
Percent transmission is the ratio of measured intensity (Itrans) normalized to the

background intensity (Io) and then multiplied by 100. When transmission has been

measured at a specific wavelength and the film thickness is known, an absorption

coefficient can be calculated (Equation 3.7).


a=- n It rans+ Irefl (3.7)
d

Care must be taken to account for reflection (Irenf) at the air-film, film-substrate, and

substrate-atmosphere interfaces in order to obtain an accurate absorption coefficient

(interfaces and reflectivity are shown in Figure 3-12).









The optical absorption decreases in Figure 3-13 with increased target to substrate

distance. The resistivity trace from Figure 3-8 is included for contrast to show that the

electrical properties are affected opposite to the optical properties.

40
30 Resistivity (mQcm)
30 -
--0 Absorption coefficient at 3pm (oa*10"4)
20

10



0 5 10 15 20 25 30
Target to substrate distance (cm)

Figure 3-13. Target-substrate distance effects are opposite for the optical absorption
coefficient and resistivity from sputtered nickel-cobalt oxide thin films from a
NiCo alloy target. As distance increases, a decreases. The optical absorption
coefficient was calculated from a transmission measurement at 3 .im (see
Appendix C for absorption coefficient calculation details).

Electrical band gap. Sputtering condition changes will likely affect the

transmission window of the produced films. Optical transmission spectra are often

described by the electrical band gap, or energy of the forbidden region between the

valence band and the conduction band. This band gap energy allows photons of a lower

energy to pass through the material without exciting electrons from the valence band to

the conduction band causing film coloration. Band gap values extrapolated from the

Tauc's plot in Figure 3-14 consist of the square of the absorption coefficient multiplied

by the photon energy plotted as a function of the photon energy. The plot is extrapolated

down to an energy value at the x-axis, which is equal to the optical band gap. Band gap

energies are shown in Table 3-3 as a function of nickel concentration. The effect on the









band gap from processing may be affected by process conditions, but that has not yet

been determined nor is it available elsewhere in the literature.


x=.75
-- x=l
x=1.5








0 1 2 3 4 5 6 7
Energy (eV)
Figure 3-14. Tauc's plot of NixCol-xO4, films show a band gap between 3 and 3.75 eV.

Table 3-3. Extrapolated band gap values lines in Figure 3-13.
Nil.5Col.504 NiCo204 Nio.75Co2.2504
Band Gap (eV) 3.2 3.4 3.5


3.3.3 Film Structural Properties

The electrical and optical properties are dictated by the atomic structure and grain

structure of the films. Structural changes of different target-substrate distances affect the

film properties. Basic film properties such as surface and bulk composition are also

briefly discussed in this section.

Target-substrate distance. Increasing target-to-substrate sputtering distance

decreased absorption and increased resistivity for films of similar thickness. Modeling of

x-ray diffraction data of NiCo204 films deposited 5 and 15 cm from the target show a

decrease in density as sputtering distance increased (see Figure 3-15).

The distance a sputtered metallic ion must travel through the oxygen plasma to get

from the target to the substrate will determine its oxidation state and kinetic energy when









it reaches the substrate. The sputter deposition rate, or number of energetic particles

incident on the surface, decreases with the square of the difference in distance. The ion

flux from the target to the substrate will decrease at longer distances as seen by the fading

of the plasma color from the cathode in Figure 3-16.

6.5





Q 5 f

4.5
Theoretical 5 cm 15 cm
Figure 3-15. Increasing target to substrate sputtering distance decreases film density.
Increased porosity is believed to decrease electrical conductivity and increase
optical transparency.


















Figure 3-16. An oxygen plasma shows a confined plasma region (white plume) near the
target and a more dispersed plasma region (yellow-green) at longer distances.

Two factors may influence the film density with distance: (1) temperature of the

substrate and (2) the mean free path of the gas. Substrate heating is affected by distance

in two ways: (a) a longer distance would allow more time for oxidation reaction in the









plasma with less heating of the substrate from the exothermic oxidation reaction, and (b)

a longer distance reduces the energy transfer attributed to the plasma-substrate interaction

(the plasma is most energetic near the cathode surface). The two effects combined yield

a lower density film at longer distances because a lower substrate temperature will not

support surface diffusion required to increase the film density.

Decreased target-substrate distance shows an increase in conductivity likely due to

the opposite effect. Oxidation of cobalt metal to cobalt oxide or the nickel metal to the

nickel oxide on the surface along with increased surface-plasma interaction could provide

heat to the surface allowing greater surface ion mobility to obtain an energetically

favorable atomic arrangement. Increased substrate temperature allows diffusion for

creating a denser matrix.





1 3













Figure 3-17. The three zone model described by Campbell for film growth in a vacuum
(Campbell, 1996).

Campbell elaborates on a three zone model that explains the deposited films.

Figure 3-17 shows a schematic of the regions of film that are affected by the substrate

temperature and the incident ion energy described by Campbell.









"At the lowest temperature and ion energy, the film will be an amorphous, highly
porous solid with a low mass density. This is the first zone of the diagram. It is
caused by the low adatom mobility of the growing film. Metal films deposited in
this region can readily oxidize when exposed to air and so may also have high
resistivities. If the chamber pressure is lowered or the substrate temperature is
raised, the deposition process enters the "T" zone. Films deposited in this region
are highly specular and have very small grains. For may microelectronic
applications, this is the most desirable region of operation. Increasing the
temperature and/or impinging energy further cause the grain size to increase. The
second zone has tall narrow columnar grains that grow vertically from the surface.
The grains end in facets. Finally, in Zone 3, the film has large 3-D grains. The
surfaces of the films in the second and third zones are moderately rough and the
films appear milky or hazy," (Campbell, 1996, p.299)

Films deposited at closer distances are likely shifting to the T region from region 1

with an increase in temperature from the heat transfer of the plasma, and from the ions

being more energetic in the plasma. As the pressure decreases, the effect is similar.









L j. r- \ .
2 .








Figure 3-18. A spinel unit cell at different angles of rotation containing octahedral atoms,
minimized tetrahedral atoms, and select oxygen atoms for points of reference
such as the enlarged black oxygen atom. Lines along octahedral sites are to
simply illustrate a crisscrossing three-dimensional network with possible
routes for conduction.

Conductivity in the spinel structure may be relatively high due to the multiple

pathways for carrier movement to occur. The cation arrangement of the spinel structure

allows isotropic electrical conduction that is not restricted to a single dimension,









direction, or a preferred plane. Conductivity likely occurs in all three-dimensions

through the octahedral cation network, which forms crisscrossing lines as shown in

Figure 3-18. This simplified illustration shows only select reference oxygen ions. The

actual network pathways along the octahedral-oxygen ions would be similar, but the path

would include an alternating oxygen anions along the path between each cation.

Closer target-substrate distance may produce more of the "cross-linked" conduction

lines connections as the film is denser. Greater distance would equate to more broken

conduction lines from pores or voids. Varying the distance and angle of incidence may

also change the stoichiometry in some cases, which will result in a change in

transparency and conductivity. Slight stoichiometry deviations such as increasing

oxygen content with longer target-substrate distances may be a possible explanation to

the observed property changes.

Film structure. The grazing incident x-ray diffraction traces shown in Figure 3-19

appear to be spinel, but do not match the diffraction pattern exactly. X-ray scans of the

sputtered nickel-cobalt oxide films are similar to the reference powder diffraction file 73-

1702. As such, these films are classified as spinel-type. Deviations occur in peaks near

600. The peak at 55.50 does not exist in the green trace and both of the 59.50 and the 650

peaks appear to be slightly shifted to lower angles.

As suggested earlier, disorder aids in polaron formation with an associated increase

in conductivity. An understanding of these deviations and abnormalities may provide

insight on carrier generation to further improve film properties such as conductivity.

Distortions in the lattice may be due to one or both of the following effects: the Jahn-









Teller effect (Dionne, 1990), or the Verwey transition (charge ordering) effect (Verway

& Haayman, 1941; Verwey et al., 1947; Verwey, 1939). *





I \ Ni ,Co0,.504



NiCo2O4

Spinel Phase (Expt D-I List)

PDF 73-1702 NiCo2O4

20 30 40 50 60 70
2-Theta()
Figure 3-19. Grazing incident x-ray diffraction from sputtered films 50 nm thick appear
to be structured as spinel-type, but do not match exactly.

The Jahn-Teller effect is a distortion in bond length that occurs allowing the t2g or

eg orbitals to shift energy and lose degeneracy to obtain lower system energy. It is

possible that the cubic spinel lattice may slightly distort to contain some tetragonal unit

cells and still be spinel-like (Blasse, 1963). Care must be taken in attributing this shift

only to the Jahn-Teller effect because it may also be due in part to a Verwey-type

transition effect.

Verwey postulated that the cation arrangement and charge distribution with respect

to the tetrahedral and octahedral sites being occupied may change at a given temperature.

The spinel may shift from normal to inverse. Verwey's system was the Fe304 spinel.

Since nickel-cobalt oxide is isostructural with Fe304, it is possible that the Verwey


See Appendix C for more information









transition may occur during heated processing of these thin films which will be discussed

in Section 3.3.4 (Verway & Haayman, 1941; Verwey et al., 1947; Verwey, 1939).

Distortions can affect the film properties although the qualitative knowledge at this

point suggests that distortion may have a role, but exactly what that role is has not been

determined. Literature data on these nickel-cobalt oxide thin films has concentrated on

surface sensitive techniques such as XPS for stoichiometric determination. Bulk film

composition may aide in determining the role of distortion or its origin.

Film surface and bulk composition. SIMS data in Figure 3-20 shows that the

surface composition may be different from the bulk, probably due to the well understood

SIMS shift from preferential sputtering making the film surface and interface appear

compositionally incorrect. Bulk studies may reveal more information regarding the film

composition effects on the optical and electrical properties. The film surface shows

nearly equal amounts of nickel and cobalt, while the bulk shows more cobalt.


Sti Substrate --
S0.8
-- Bulk Film I
0.6 si
SI i Co59
0. i I [ Ni58
0 0.4
0.2 I |

Surface / Interface
0
0 500 1000 1500 2000 2500 3000 3500 4000
Sputter Time (seconds)

Figure 3-20. Sims profile of nickel-cobalt oxide thin film shows surface composition
appears to differ from the bulk composition. XPS analysis for chemical
composition on the surface agrees with SIMS data.

Calibration standards were not available for this material. Effects such as

preferential sputtering have not been considered, but may take into account phenomena









that occur on the surface during sputter-etching. Sensitivity factors for the nickel-cobalt

oxide matrix have been calculated using XPS as a SIMS calibration, however the results

may be skewed due to the nature of sputter etching that occurs in the individual systems

to remove material and acquire the depth profiles. XPS depth profiling was done with an

argon sputtering source to remove surface layers. SIMS was done with an oxygen source

and therefore the oxygen is not included with the depth profile in Figure 3-20. The

results may not accurately represent the material due to the possible preferential removal

of either metal species from the exposed surface during the depth profiling. SIMS did

show that in the bulk the film composition is constant.


Substrate






Interface

Film



Figure 3-21. Cross-sectional TEM image of sputtered nickel cobalt oxide films on (100)
silicon. Nickel-cobalt oxide thin film grows in multi-grained columns.

Micrographs from a transmission electron microscope (TEM) show that the films

were homogeneous from the surface down to the interface. The interface is interesting.

Notice the dark band that runs diagonally from top center to lower right side at the film-

substrate interface in Figure 3-21. The white line is the native silicon dioxide layer found

on silicon wafers and is typically 2 nm thick. The dark region at the interface may be a

metallic growth layer and the cause of poor transmission and high conductivity. A highly









conductive metal alloy film, thin enough to be transparent and undetectable by XRD

could be part of the dark band at the interface surface. It may be a substoichiometric

phase that acts as an interface layer on the silicon substrate. Another possibility is that it

could include a metal-silicide layer of reacted metal with the surface of the substrate.

Nickel silicide is used in silicon wafer fabrication processing as an interconnect because

of its high conductivity, but nickel or cobalt silicide optical properties have not been

reported. Both are thermodynamically stable and could form as cations migrate through

the silicon dioxide barrier layer into the bulk wafer. The SIMS profile in Figure 3-20

does not agree with the data in Figure 3-22 where increased concentrations of cobalt is

detected at the interface and substrate regions which would substantiate this claim.

Co Ka (cts)
15 aNi Kb (cts)


10


5-

0

0 50 100 150 200
Location (nm)
(a) (b)

Figure 3-22. Nickel-cobalt oxide-silicon wafer interface. (a) Energy dispersion
spectroscopy (EDS) of area the shown in (b), the STEM micrograph of film-
substrate interface scan region used for EDS analysis in (a).

The interface layer is cobalt rich according to the EDS data from the STEM shown

in Figure 3-22. The cobalt rich region may be part of a substoichiometric spinel film at

the interface between the nickel-cobalt oxide and the silicon dioxide native layer on the

surface of the silicon wafer. Closer inspection of the SIMS data in Figure 3-20 shows the









cobalt content falls off and the nickel content remains steady before tailing off. It is

possible that this is merely evidence of the preferential sputtering artifact. EDS does not

rely on film layer removal for characterization as SIMS and is therefore not subject to

preferential sputtering effects. The cobalt rich region detected by EDS is probably not

indicative of silicide formation, otherwise the silicon trace in the SIMS depth profile

would have a spike in it representing the metal-silicide layer was present. A closer

examination reveals a change in slope of the silicon trace about mid way through the

interface region. This slope change could indicate a metal-silicide formation, but it could

also result from the native silicon dioxide layer, which would have a different sputtering

rate than unoxidized silicon. The data are best interpreted to mean that the film-substrate

interface consists of a cobalt rich nickel-cobalt oxide layer on the native silicon dioxide

surface of the silicon wafer.

3.3.4 Post Deposition Heat Treatment

Heat treatment of the nickel-cobalt oxide films result in increases in conductivity

with concomitant decreases in transparency depending on the method used for heat

treatment and the temperature involved As mentioned in Chapter 2, temperatures above

4000C cause phase separation of nickel oxide within the spinel film. Film properties

degradation occurs over periods of days for as deposited and annealed samples left

exposed to air at room temperature. Typical degradation of conductivity is a factor of

two after one week. Successive heat treatments can return the film conductivity to nearly

the same value as before the degradation if the cooling rate is on the order of

-1500C/min. When the cooling rate is less than 15C/min, the film conductivity can be

artificially aged to the degradation value had the sample had been left exposed to air for









several days. Transparency responds the opposite to conductivity after heat treatment

activity.

Heat and conductivity. Figure 3-23 shows a plot of the natural log of conductivity

plotted against the reciprocal of temperature multiplied by 1000. As temperature

increases from 300 K to 625 K, shown by the blue traces for the two different

composition films, the conductivity measured at temperature increases. These particular

films were first heat treated at 3750C (a 1000/T value of 1.54) for ten minutes and then

rapidly quenched to room temperature before commencing the heated conductivity

measurement. The conductivity increases with increasing temperature.

T (K)
675 600 525 450 375 300
--6.5
-



U

4.5 --D- Ni,,Co, 504 Heating ---NiCo2O4 Heating
m- Nil CoI.sO4 Cooling -- NiCo2O4 Cooling
,4 4 1 ..
1 1.5 2 2.5 3 3.5
1000/T
Figure 3-23. An Arrhenius plot of conductivity and the reciprocal of temperature (K)
shows that at high temperatures, the film conductivity is high. A temperature
of 525 K graphed near 1.5 on the abscissa divides the low temperature region
on the right from the high temperature region on the left. The two regions
have different activation energies for electrical conduction. The heating and
cooling rate is limited to a maximum of 15/min to avoid instrument damage.

Notice the difference in conductivity at the same temperature upon cooling from

the highest temperature shown by the red curves. The room temperature conductivity

upon cooling is always below its original value. There is a temperature where the heating

and cooling data are nearly superimposed above, but depart below. Windisch et al.









reported this anomaly of conductivity being different for two regions of slope associated

with each of the heating and cooling curves, and have different slopes versus temperature

at high versus low values (Windisch et al., 2002b).


Figure 3-24. TEM image and diffraction patterns at 300 K and 600 K showing no
detectable structural changes for the two temperatures.

Their speculation was that the transition could be the result of a Verwey transition

near 525K that accounts for the change in slope of both the heating and cooling curves.

Verwey et al. described the a temperature transition in Fe304 in terms of an

order/disorder transition resulting from the preferential ordering of the cations which

changed the activation energy for electrical conduction (Verway & Haayman, 1941;

Verwey et al., 1947; Verwey, 1939). Rather than being a Verwey transition, it could be a









structural transition, a magnetic transition or possibly a mobile species within the lattice

that has a different hopping activation energy triggered at a higher temperature.

The explanation for the effects of heat treatment is open to speculation. One such

speculation invalidated by Figure 3-24 is that of a structural transformation. If a

structural transformation occurred, the TEM diffraction pattern generated at 300 K would

be different from the pattern at 600 K. No change is indicated, suggesting that no

structural changes occurred and therefore rules out the Jahn-Teller effect as a significant

contributor. It is, however, possible that the metastable state quenched in relaxed out

before the STEM was performed or was processed out as the sample was prepared or

during the analysis from localized electron beam heating. If the 600 K temperature

experiment was conducted first then the structure is expected to be similar at 300K.

Magnetic measurements would be required to invalidate the suggestion of a

magnetic transition. Samples were run on a vibrating sample magnetometer (VSM), but

no significant hysteresis could be detected either due to lack of a transition or more

probably due to a lack of sensitivity because the film was too thin. A magnetic transition

such as a spin state realignment triggered at a specific temperature may correlate with a

Verwey transition in that the arrangement of the cations and ordering/disordering of their

charges may contribute or detract from a magnetic domain arrangement of the crystal.

This is still unknown and could be the subject of a future study.

The last suggestion was that a mobile species could be the cause of the change in

electrical conductivity slope upon heating. The only mobile species in the lattice is

oxygen. Oxygen is typically present as 02- anions occupying the 32 available anion sites.

Oxidizing conditions during sputtering and heat treatment may allow excess atomic









oxygen to adsorb on the surface from dissociated carbon dioxide, water molecules, or

molecular oxygen. Oxygen in the spinel cell would not necessarily be restricted to one

site only and at elevated temperatures may have enough energy to site hop or form

Frenkel defects involving oxygen ions (Callister 1997). In addition, a

superstoichiometric film could result in the lattice in the presence of 01-. These oxygen

defects could add additional electrical carriers to increase conductivity and will be further

discussed below.

Effect of heat treatment cooling parameters. Film cooling rate also has an effect

on the observed properties. Figure 3-25 shows the effect of cooling rate on transmission

in the visible and near infrared regions. Optical transmission measurements were

collected after quenching from the heat treatment described earlier, and after the

temperature dependent conductivity measurements shown in Figure 3-23. NixCo3-xO4

films where x = 1 are lower in conductivity overall than the films where x = 1.5.

60

a 50

40

30
x=1SloS 150/min
| 20 x=l Rapid 1500/min
S~~0 x =1.5 Slow
10
x=1.5 Rapid
300 900 1500 2100 2700 3300

Wavelength (nm)
Figure 3-25. Effects of heat treatment cooling rate after heat treatment on optical
properties of NixCo3-xO4 samples from Figure 3-23.

Heating alone does not ensure high conductivity. When cooled slowly, the

conductivity degrades rather than improves. Conductivity can be increased by up to an









order of magnitude by quenching rapidly from 3750C to room temperature by using a

large heat sink versus slowly cooling the sample at a controlled 10-15C per minute.

20


S14 x=l




2-
Quenched 3000C Heating 3000C Cooling 2600 hours later
Figure 3-26. Resistivity of NixCo3-xO4. (1) quenched to room temperature after heat
treatment at 3750C for 10 minutes in air, (2) at 3000C, the first point of the
temperature dependent conductivity measurement on the heating cycle, (3) at
3000C, the last point of the temperature dependent conductivity measurement
on the cooling cycle, and (4) -16 weeks after the cooling heat treatment.

Effect of time after heat treatment. Samples exposed to ambient atmosphere for

days to months after quenching exhibited resistivities that slowly increase by a factor of

two or more. Figure 3-26 shows a Nil.5Co1.504 sample after a heat treatment cycle.

Discussion of heat treatment effects. Heat treatment of nickel-cobalt oxide thin

films has an appreciable effect on both optical transmission and electrical conductivity.

Electrical conductivity improvements are a factor of cooling rate with the optical

properties degrading with increased conductivity. Heat treatment may either repair or

induce defects depending on the cooling procedure. Since the phase transition does not

appear to be probable, it is possible that localized defects form from atomic disorder or

cation disorder, or Frenkel defect oxygen and carriers get trapped in these locally

deformed regions becoming polarons. Disorder increases the number of polarons in the

lattice and therefore increases the conductivity. Quenched samples have lower activation









energies for conductivity in both low and high temperature regions compared to slowly

cooled films (summarized in Figure 3-27).


0.08

0.07

S0.06


6 0.04

0.03


0.01

0 High Temp
x=1.5 x=l Low Temp
x=1l
Heating x=l'5 x=l


Figure 3-27. Rapidly quenched sample activation energies of NixCo3-xO4 change after
heating when they are slowly cooled.

Activation energies calculated from Figure 3-23 and included in Table 3-4 show

that the values calculated from heating a quenched sample are all lower than those

activation energies calculated from slowly cooling the same samples. Rapid quenching

must therefore lock in some degree of disorder or defect structure.

Table 3-4. Increase in activation energies of NixCo3-xO4 from the temperature regions in
Figure 3-23 graphed in Figure 3-27.
Heating to cooling low temp Heating to cooling high temp
x=l 0.020 0.035
x=1.5 0.018 0.025


The possibility that a magnetic or physical phase transition could be the cause of

the two different transition regions can be ruled out by the same arguments advanced in









the heat and conductivity subsection above. Transport of surface adsorbed species such

as oxygen into the lattice to increase the oxidation state of nickel or cobalt to form more

polarons and increase conductivity would be a thermally activated process. STEM data

does not show a significant structural change with heat treatment, which indicates that the

bulk of the grains do not experience the increased oxygen effect, but the grain boundaries

may accommodate extra oxygen. These highly defective regions with excess oxygen

would be frozen in the film when it is rapidly cooled. The steady state preferred structure

would not be achieved due to kinetic limitations upon cooling. The thin film would relax

to this condition perhaps by the excess oxygen to escaping. These extra defects may

distort the unit cell via the Jahn-Teller effect or a Verwey charge ordering effect just

enough to allow a lower ensemble average of activation energy by increasing the

distribution of activation energies for individual sites. A model that approximates this

disorder idea is presented in Chapter 6.

Another possible explanation for the increased conductivity at higher temperatures

would be densification due to heat treatment. Heat would result in increased density

from the increased solid state diffusion and the increase in film conductivity could result

from elimination of voids. However, it is unlikely that porosity would increase at low

sample cooling rates making this explanation less likely. XPS data showed high surface

concentrations of carbonyl on the nickel-cobalt oxide films (Windisch et al., 2001a;

Windisch et al., 2001b; Windisch et al., 2002b). If carbon was to permeate the lattice and

take oxygen to form carbonates upon cooling, this could increase the disorder during

heating. If this mechanism is operative, the data suggest that carbonates would

decompose and reoxygenate the lattice or even form bubbles of carbon dioxide molecules









thereby increasing porosity. While this idea cannot be entirely discounted, it is

considered unlikely that the carbonate will decompose upon cooling.

3.4 Summary and Conclusions

The combinatorial approach is effective for depositing a film with a continuously

variable composition over a large area in one run. The effects of gas composition during

combinatorial sputtering and the effects of total chamber gas pressure and different target

to substrate distances using single target sputtering in the offset rotation setup made up

this study.

Optimum conductivity is achieved from combinatorial films containing equal parts

of nickel and cobalt when sputtered in a gas mixture of 50% argon and 50% oxygen. The

best film conductivity of 375 Scm-1 may be a result of ideal growth conditions with the

gas mixture promoting complete oxidation and a higher deposition rate.

The gas pressure study shows that a chamber pressure of 2 mTorr yields the highest

reported film conductivity to date at 500 Scm-1 compared to the 5 mTorr (400 Scm-1) and

10 mTorr (333 Scm-1) following rapidly quenched after heat treatment. This enhanced

conductivity is attributed to a higher growth rate and increased adatom mobility during

growth due to the lower molecular mean free path.

Increased distance of target to substrate decreased the film density increased film

porosity and decreased conductivity, but increased transparency. Target-substrate

distance effects on the film density are believed to be controlled by phenomena similar to

those from lower pressures i.e. higher surface mobility from depositing species due to

less gas phase scattering. Less plasma-film interaction at longer distances may also be a

factor. Closer target-substrate distances result in denser more electrically conductive

films with lower optical transmission, opposite of the films deposited at longer distances.









Heat treatment showed improvement in conductivity when films were rapidly

quenched and degradation when slowly cooled. Transparency once again behaved

opposite to the electrical conductivity. Temperature dependent conductivity showed that

after a rapid quenching heat treatment, the activation energy of conduction is lower than

from a film that was slowly cooled. This heat treatment cooling rate effect is a new

discovery for this system and is believed to be due to excess oxygen in the lattice. Rapid

cooling from heat treatment at 3750C to room temperature may lock in defects that

increase the concentration of polarons. Slow cooling allows them to anneal out, resulting

in improved transparency across the infrared region and lower conductivity.

Band gaps of 3.2 and 3.5 eV and work functions of 4.39 and 4.27 eV are reported

for NixCo3-xO4 at x = 1 and x = 1.5, respectively. These band gap and work function

values have not previously been reported to date.














CHAPTER 4
THE ROLE OF LITHIUM

4.1 Introduction

Adding lithium to the nickel oxide, cobalt oxide, and nickel-cobalt oxide systems is

a currently acceptable practice in lithium battery electrodes (Appandairajan et al., 1981;

Banov et al., 1995; Benqlilou-Moudden et al., 1998; Carewska et al., 1997; El-Farh et al.,

1999; Fransson et al., 2002; Ganguly et al., 1997; Gendron et al., 2003; Gover et al.,

1999; Han et al., 1999; Julien, 2000; Julien et al., 1999; Kim et al., 2002; Koumoto &

Yanagida, 1981; Moshtev et al., 2002; Moshtev et al., 1996; Moshtev et al., 1999;

Robertson et al., 1999; Seguin et al., 1999; Shirakami et al., 1998; Stoyanova et al., 1997;

Yoshimura et al., 1998; Zhecheva et al., 1996). The same is true for molten carbonate

fuel cell electrodes (Fukui et al., 2000; Kuk et al., 1999; Kuk et al., 2001). Improvement

in some properties resulted as lithium was added in specific quantities, however the

material requirements for battery electrode and fuel cell electrode technology is

fundamentally different than those for optoelectronics. Optoelectronics is primarily

concerned with photon-solid interactions and electrical conduction. Batteries and fuel

cells deal with electrochemical properties such as ionic conduction and intercalation

(Koumoto & Yanagida, 1981; Wolverton & Zunger, 1999; Zhecheva et al., 1996). Many

studies have been done with lithium in nickel-cobalt oxide (Puspharajah et al., 1997;

Urbano et al., 2001), but none have attempted to probe nickel-cobalt oxide with added

lithium for use as an infrared-transparent thin-film electrode.









Initial studies suggested that the addition of lithium to the nickel-cobalt oxide

system would favorably improve conductivity and transparency based the fact that

lithium could be incorporated into the spinel unit cell at a tetrahedral site (Appandairajan

et al., 1981; Windisch et al., 2001a; Windisch et al., 2002a; Windisch et al., 2001b;

Windisch et al., 2002b). Lithium, a monovalent ion (for general spinel structure

information see (Azaroff, 1960; Kingery et al., 1975; Smyth, 2000)), could substitute for

divalent or trivalent cobalt resulting in a net negative charge in the lattice. An adjacent

cation in a tetrahedral site should then increase its oxidation state to maintain charge

neutrality in the crystal. It is also possible, though not as likely, that a neighboring

octahedral cation could increase its oxidation state by donating an electron to balance the

charge offset from the lithium substitution on the tetrahedral site (Appandairajan et al.,

1981). Such an increase in oxidation state will change the bonding nature at that site

(create a polaron) resulting in increased conductivity.

This prediction assumes that the lithium will substitute for a cation on a tetrahedral

site, however it is possible that lithium, being a small atom, may prefer to enter the lattice

as an interstitial and produce an opposite effect. Interstitial lithium would add a positive

charge, likely requiring a neighboring cation to reduce its oxidation state. The result

would be a net decrease in conductivity from polaron annihilation. Assuming the first

case, XPS binding energies might be expected to scale or shift with lithium addition, as

would the lattice parameter. Changes in transparency and electrical conductivity may

occur as well.

Preliminary proof of principle data are shown in Figure 4-1 for contrast and

comparison to experimental results of this study. A favorable improvement in resistivity









for solution deposited nickel cobalt oxide with 10% addition of lithium was recorded.

The improvement from lithium in the solution nickel-cobalt oxide films combined with

an order of magnitude increased conductivity reported in the literature for sputter

deposited versus solution deposited films (Figure 2-8) provided motivation to determine

the effects of lithium in sputtered samples. Experimental results will show that lithium

behaved in a manner unexpected and yielded new and interesting details of the nickel-

cobalt oxide system with and without lithium doping.

This chapter focuses on the electrical, optical, structural, and chemical changes

produced by the presence of lithium in the nickel-cobalt oxide system. Electrical

properties were largely of interest with respect to heat treatment parameters and

activation energies of conductivity to explore disorder in the system. Optical properties

provided insight to understand the electrical results that were confirmed by both the

structural and chemical analysis.

4.2 Experimental Procedure

4.2.1 Deposition

A solution deposition method produced some of the films for this study*. Solutions

of metal nitrates and a combustion agent are mixed in aqueous solution and eye-dropped

onto the substrate just before spinning at 3500 rpm for -30 seconds. Placing the wet film

substrate on a hot plate at -3750C ignites the glycine combustion agent included in the

solution to decompose the nitrates and produce a uniform film -50 nm thick.

Both traditional and combinatorial films were sputter deposited* in a vacuum

chamber evacuated to near 1x10-6 Torr and backfilled to 10 mTorr for all depositions.

Reactive RF sputtering was performed in 100% oxygen using alloy target while

* Additional details of solution and sputter deposition are included in Appendix A.









sputtering of oxide targets was performed with 50% oxygen and 50% argon. Oxide

targets ofNil.5Co1.504 and Ni1.35Co1.35Li0.304 were used in the combinatorial setup to

deposit a film with a graded lithium concentration (schematic in Figure 3-1).

4.2.2 Characterization

Film thickness was measured with an optical or stylus profilometer. X-ray

reflectivity (XRR) measurements of select films provided confirmation of the

profilometry data (See Appendix B for more information on characterization techniques

and data). Electrical measurements conducted in a van der Pauw apparatus provided

resistivity values when combined with measured film thickness. Temperature dependent

van der Pauw data were collected from room temperature up to 675 K.

Fourier transform infrared (FTIR)* (Brundle et al., 1992) was used to measure film

transparency with respect to air and with respect to the silicon or sapphire substrate over

the range of 4000 cm-1 to 400 cm-'after 2 minutes of a nitrogen purge. A dual beam

ultraviolet-visible spectrometer was used to measure transparency from the ultra violet

region to the near infrared region (200 nm-3300 nm). The deposited film composition

was determined using x-ray photoelectron spectroscopy (XPS)* (Brundle et al., 1992)

and secondary ion mass spectrometry (SIMS)* (Brundle et al., 1992) Depth profiled

dynamic SIMS data was calibrated with high resolution XPS depth profile scans.

Grazing incident x-ray diffraction (GIXRD)* (Brundle et al., 1992) provided information

on the crystal structure, crystallite size and orientation, and lattice parameter.


For details on characterization techniques, please refer to Appendix B.