<%BANNER%>

Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2010-08-31.

Permanent Link: http://ufdc.ufl.edu/UFE0021726/00001

Material Information

Title: Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2010-08-31.
Physical Description: Book
Language: english
Creator: Wu, Zhuangchun
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

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

Notes

Statement of Responsibility: by Zhuangchun Wu.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Rinzler, Andrew G.
Electronic Access: INACCESSIBLE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021726:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021726/00001

Material Information

Title: Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2010-08-31.
Physical Description: Book
Language: english
Creator: Wu, Zhuangchun
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

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

Notes

Statement of Responsibility: by Zhuangchun Wu.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Rinzler, Andrew G.
Electronic Access: INACCESSIBLE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021726:00001


This item has the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101222_AAAACC INGEST_TIME 2010-12-22T20:02:01Z PACKAGE UFE0021726_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 17093 DFID F20101222_AABKEQ ORIGIN DEPOSITOR PATH wu_z_Page_019.QC.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
093dec407efd8f18a161611ef789bdc4
SHA-1
87ce1285161c226c7d8c0620f184ee6d4f387515
10927 F20101222_AABJZK wu_z_Page_088.pro
23d9187704f7e90b32335b7fa24d4834
0d012a574bc47c3df238f73b44226f73ea8db4b2
5010 F20101222_AABJYV wu_z_Page_071.pro
f8ac4609624c0d1ccc7dbbec18cf5a54
7eff0463a3760b05a75e094e5c4d365f6c937bd3
33889 F20101222_AABKFE wu_z_Page_028.QC.jpg
9dd1839997665293cb4e05c6231da5ba
5e35c71b15a222b2917cb8d7ae75ccc0c2d9990d
4248 F20101222_AABKER wu_z_Page_019thm.jpg
d2d701ccf4da8dc2ce1963ca67bef880
58f3cbd26b5b90ec28297fb9ddafa8cb62273f82
6960 F20101222_AABJZL wu_z_Page_091.pro
c019e30063fa5061b501133cd45df753
80ee50dbbded2c8df1cb1a8ebf0b32f693b9b0d1
4876 F20101222_AABJYW wu_z_Page_073.pro
0ab7346bb34693cb129f4600447527f6
7ddfa1008a8800e6bb571336a7a758942e533edb
8393 F20101222_AABKFF wu_z_Page_028thm.jpg
2e2a913f3bf8f4924400310a4b0bafe9
ad4ea3f3371a6eaa21f255c768c10b2712250385
14563 F20101222_AABKES wu_z_Page_020.QC.jpg
d558a671ff5146506051c097b2dc7086
108b8402a818cdbfcb6f7424c098d0a9e8961305
6601 F20101222_AABJZM wu_z_Page_092.pro
63bd547a4e5891782392f24929ecb73a
2a93cc3731f4e0c72853e3e6036c4c31014899e9
54305 F20101222_AABJYX wu_z_Page_074.pro
f69b80ca2e5b7fb2f3469576906b5be4
dddc76e25ecb5261e79f6006243ccbe12f1e6fa0
35645 F20101222_AABKFG wu_z_Page_029.QC.jpg
3135cda14039eed0cf859e1797fa79a4
5b44217a27f8cd6a7caff530b050ef3e5e260435
4420 F20101222_AABKET wu_z_Page_020thm.jpg
01b8829a029e065b72ebee95253a3b43
1b1efae6c039065db834e8a25bbad5a2e785a222
7700 F20101222_AABJZN wu_z_Page_093.pro
114eec90a2f6f41150a8a4f3150cc03e
8c01e4934eec8e9f83f7ee5b560b188afb9847e7
57764 F20101222_AABJYY wu_z_Page_075.pro
5db90e17b5c8aefad991f9d33eef67ff
7dabfd457a2bff5c83df66f11a01f3938107d232
8841 F20101222_AABKFH wu_z_Page_029thm.jpg
f3ac03857fa2891c83ada5140624b553
2977307cc620951d44b6e96810b4882e782ae21e
12501 F20101222_AABKEU wu_z_Page_022.QC.jpg
352d9a52452d6d7c74513faf44469176
9814089e0561b8470f4e7be9228ef4cf1b8d79bc
2948 F20101222_AABJZO wu_z_Page_095.pro
766d068f860e31f40927730476d11dcb
a13b5b1ebbe18d7b17c415c90a20cab7d604ff52
53350 F20101222_AABJYZ wu_z_Page_076.pro
2e68fef7a2201ad97c93ab7dc0e7e74d
cc60d7cb2b426e65c7e6c585b663d112caf225cb
28565 F20101222_AABKFI wu_z_Page_030.QC.jpg
79860d5ec6075a72c7bc3f8292627c59
569031b8f07a1cec4e2709095b9aed25429f5a13
4159 F20101222_AABKEV wu_z_Page_022thm.jpg
9eda796af5e3143dfe725bf4cf7d5958
319dcdb381c90dc9a2b3d162b63ee5e04c76d080
6302 F20101222_AABJZP wu_z_Page_096.pro
d481f7e3dfb5663f3e2b916a186d97f5
3d17abc79a4a6d9b72a644ac3a7c697229176e4c
6929 F20101222_AABKFJ wu_z_Page_030thm.jpg
0c9fa35cf3cbe4a26d2af4deb8959d33
0a1d7d47dbc0480a7ef74f4600be0b7168dd0a6a
35707 F20101222_AABKEW wu_z_Page_023.QC.jpg
91831794015b47fa72e440185a2ad153
b9c478126fb60172951a090cbdda5c32f1448c70
5542 F20101222_AABJZQ wu_z_Page_097.pro
16d3620b58e46006dbc42debf612cc4c
cf7c7afb703834ae8c0bffe6b538efaba01bea5d
14687 F20101222_AABKFK wu_z_Page_031.QC.jpg
8c2636ea96b23712653c8335e8bae381
805e6a35381cf16041123b35412b3095b142ffd2
36382 F20101222_AABKEX wu_z_Page_024.QC.jpg
74b3bc88c3e67f907e7c447fc71468ef
94576896a826a0c3f1a9e38c00a7cc8726906081
8650 F20101222_AABJZR wu_z_Page_098.pro
5094f9d74c33ff87bd996566ee292f7a
47689fa71b2268dd4ac7f1b813def25ecae5ced2
35754 F20101222_AABKGA wu_z_Page_041.QC.jpg
0328e4d56d59cc275a20cc31b5aaeace
8bb2e089068e3999b34a7e4c425b78288c9557a9
4354 F20101222_AABKFL wu_z_Page_031thm.jpg
9a9d2968ecc8f1ce804d86269759fab1
de57283d4828a7deb772aba543d7762b6b1efbb4
35737 F20101222_AABKEY wu_z_Page_025.QC.jpg
c9082982676b5f72693b92fd6a8079c1
c4a2fd496e8634136f4dd3a60f4278d56a33de92
5236 F20101222_AABJZS wu_z_Page_099.pro
660845712f944f6d4bfe226b8497a848
df74d1eed8d623496acba34e670914f569afa66b
33701 F20101222_AABKGB wu_z_Page_042.QC.jpg
5bf1b59c464775c8dbeb04622f60853b
d2e58b5bd9eab6a8467ab956dc15862e27ef6a2f
13842 F20101222_AABKFM wu_z_Page_032.QC.jpg
52bbfad1cc9f40b6cee353727a57b306
ccbbe0f98a0558f7440ba7a954abfe9a5451c03c
9015 F20101222_AABKEZ wu_z_Page_025thm.jpg
2312704eec68073c6def126e5b1ba70c
827b6c68028a331b87b0292f33910bc86eae6df1
8473 F20101222_AABJZT wu_z_Page_100.pro
ef5c4acd581ce5c180cc9698881b7bf8
9e7491d5889540164ed8501077961385014799a5
4482 F20101222_AABKFN wu_z_Page_032thm.jpg
aa089418a0af82bf739cf3165fd85cbd
935649ee5b9b6044c42b90dfd35cf170cb49d646
10740 F20101222_AABJZU wu_z_Page_101.pro
19746d4318d6c709f2921775b0d8b7eb
9949ab14c7961de44942abc2f020a2b22a186ca0
8636 F20101222_AABKGC wu_z_Page_042thm.jpg
09e3dc74bfc45cd53ed9a2ed16166f50
832038eac658114efc3fa907483039954760b185
18828 F20101222_AABKFO wu_z_Page_033.QC.jpg
a4b04e791cae9ba9e302234c323e3fcb
52e63ffe1f304e4e382a2f1f8ff615668d59fab1
54199 F20101222_AABJZV wu_z_Page_103.pro
45c5ca91f4cec0a097e66e65a697def0
a45621555e2a3214fccabe56cf0c004d9298607e
15313 F20101222_AABKGD wu_z_Page_043.QC.jpg
c76d5219d094262a2f43ba4aaeb4276a
25b7f78df1029ca74999fb904dd45e0fb4180d2c
5361 F20101222_AABKFP wu_z_Page_033thm.jpg
23579681ff2f55bfb5b52ba7ffe52c16
11f0978b0d3b3083a5f99d68d5830596728bf8d9
3877 F20101222_AABKGE wu_z_Page_043thm.jpg
6fcd199565b79523f495538bf5259848
357d5cad45b7f50d465ded277a58695a64df23f8
8911 F20101222_AABKFQ wu_z_Page_034thm.jpg
cf62a41da7c1b908c2fb0de58e7cbeba
a29536dbbf163543c6731488622277f6048490d2
50506 F20101222_AABJZW wu_z_Page_104.pro
dd6b06c046274932f5105ee737c9313a
533f0422320bfc06c05e508c9a6959950f8ed9c2
9734 F20101222_AABKGF wu_z_Page_044.QC.jpg
a32e3624f3e2add1585c27e82733a0e2
4170a548387dc2333a3b4c07516b8532fb365afd
34820 F20101222_AABKFR wu_z_Page_035.QC.jpg
62f0eec8d252dd049bee8e3a1efe21a9
0af3b864d23601e7eb2bacc28baf0b0f36fbf747
3892 F20101222_AABJZX wu_z_Page_105.pro
3e5bde279c5f2aa57cb787bf302342d2
9cd445ec03466ffb4f48779e213cf5443274741c
11766 F20101222_AABKGG wu_z_Page_045.QC.jpg
2bb1546f1667303727d8ddd58ce78962
95c2279c931dc0e211fa1fb4cbfc7e3d4404018b
8481 F20101222_AABKFS wu_z_Page_035thm.jpg
40a5c5acc66906c60f3a3f8b4e50a050
f982030bc9320d69a86c95e9158891c118e85f7c
44231 F20101222_AABJZY wu_z_Page_106.pro
faa94375542cde50e1e299bcdf82e6a6
316b7f11c10ac9f6f5b6f8f3cc98e3630b2d4745
3899 F20101222_AABKGH wu_z_Page_045thm.jpg
d7226ca482ce2568e139e5b6b060f1a1
4c3c2c8a4de7112678579ddbc7e26c1f6cdd3359
8799 F20101222_AABKFT wu_z_Page_036thm.jpg
f536c367936c20869a3bb7224fbad553
41062a72a6a486122de3c426e251f5d4ac445ba0
63044 F20101222_AABJZZ wu_z_Page_107.pro
8b7822b71ac237fabb44a6e6f4a6ca8c
74d52b6caec7fcf8e877ce9faa145794b81a4b1a
14817 F20101222_AABKGI wu_z_Page_046.QC.jpg
b04725ddcb81faa2e64a4322795db9d1
4773460f18b9c961da7fc1948de2f52b1fe3229a
37109 F20101222_AABKFU wu_z_Page_037.QC.jpg
c2240fca09eee37a87f55275131821c8
9c98001c99013bfd7cd3355645aa0c83ac5c10ff
4641 F20101222_AABKGJ wu_z_Page_046thm.jpg
b777f1b53e0546f3bb66c92c45b1f100
75fb79b165f2b1d838ba4ffead7a6232a6b85bc2
9203 F20101222_AABKFV wu_z_Page_038thm.jpg
b3ee4bc031df69e876121c81bed74d3c
9cdf4239e1bb0690d411c21f220bfe6880c4a3d7
13975 F20101222_AABKGK wu_z_Page_047.QC.jpg
6441cf1607fe70000f6e7d2d1d585858
c6ac794002a08f06d97a82d202b16bfc27f1bd80
35223 F20101222_AABKFW wu_z_Page_039.QC.jpg
90d48a74dbd49423004e329485c9d5fc
345609932a0604bb7e14c136a6d4c956191be246
33583 F20101222_AABKHA wu_z_Page_057.QC.jpg
a7f432a85de4b991966921cd5f5021cc
fe62c9e8560b13eeecc325852e12776b608bcad2
4479 F20101222_AABKGL wu_z_Page_047thm.jpg
c41b740def697973f2d3d5a2f3161907
44ec1571ebf0dc3a967c0839d876ce0d33f98f79
8488 F20101222_AABKFX wu_z_Page_039thm.jpg
b341e6b26f18e55a7f45b5d7cd0293a3
bb58d8694db694f3b875c3bb4af10520989f9410
1201 F20101222_AABKHB wu_z_Page_058thm.jpg
d030c5acd379a5909cc54e5d84510abb
863329b0bc6e5c864f52a5e378fabd943bf413f8
17723 F20101222_AABKGM wu_z_Page_048.QC.jpg
b1e595adcb1cae78785388cb20ba5485
b1907a593ffcc9cfabc02ee69e0efa910af0ab17
35464 F20101222_AABKFY wu_z_Page_040.QC.jpg
8ec10a60eed70de15b5d07d3ee41cd5b
9ef13cf7a3da3ac5ba5cb6c496e39cf9bed63ec3
11985 F20101222_AABKHC wu_z_Page_059.QC.jpg
83f3bf2992e134c06231c9caf67cee38
ae517c01c4cc9da84350b5a7f5081016a87b9d5c
8757 F20101222_AABKGN wu_z_Page_049.QC.jpg
6e81c1ec0ad8fb90f3ead142af73dac4
2ddd4677cfc3753f5b9f4c3a961896dea5f4d899
8892 F20101222_AABKFZ wu_z_Page_040thm.jpg
b8fdc4a3084fec9b2f063401cf35ee7a
b0fcfa1c7be383dc7e0196160152d081a8780473
2939 F20101222_AABKGO wu_z_Page_049thm.jpg
d3f56d7ee46aa4df48da413db22e5337
8f86e70bfce6d32f31b191e21d354e8aca80563f
3436 F20101222_AABKHD wu_z_Page_059thm.jpg
bfadae7595e3d7cf521cc90668df1fc5
79f5a36cce8eb49327a6d455d35556d8aa9e26d8
8904 F20101222_AABKGP wu_z_Page_050.QC.jpg
2c6f77f948725da70815e64cabb19b9c
5879bff7f87924fc6f0a471271ecd241e3239378
13360 F20101222_AABKHE wu_z_Page_060.QC.jpg
008d50f16e0634a5e210303f79d3088b
64b6cb19103a058afe8c72af1d2f75027de4f2a2
2688 F20101222_AABKGQ wu_z_Page_050thm.jpg
60fe70b5b8cd5186b12344babcf626eb
d7718282357f18724476b4f93404c479bdea5aa9
4931 F20101222_AABKHF wu_z_Page_060thm.jpg
f3ae1fc5d40d9a147a8f1d76270b7b5f
e425aa7031af7058204fbe1d7c5d82e2b3ae477f
16195 F20101222_AABKGR wu_z_Page_051.QC.jpg
86ca26d0f03a1b368bddc29abae0f806
590b96b874c70ea653d1a76c03a3ac1a09d3795f
8673 F20101222_AABKHG wu_z_Page_061.QC.jpg
d764ca37c82035eea9ac851de1f6a453
5b6598ecd6797b31e9b4aca696e510863fba0db6
4953 F20101222_AABKGS wu_z_Page_051thm.jpg
7faa0b30c4d61e10b45ffb37daf07de9
3331e3153b80613788e4e255744f718093babc5b
6174 F20101222_AABKHH wu_z_Page_062.QC.jpg
5f7d2981e54e5ca6c63d2eada39cf864
5f830f932eaf83b69607333a6acbe37ea22cf02a
37972 F20101222_AABKGT wu_z_Page_052.QC.jpg
30b0de0c5d751199adda55a599b3806e
6615c6d8c74b7c62843eaf9f6eae78f9f59c8122
1925 F20101222_AABKHI wu_z_Page_062thm.jpg
7efd79c37e7e44b9b43f92f9adbb10f3
0a51fce7dfa5ec45ddc8e20cca89286acd292736
8836 F20101222_AABKGU wu_z_Page_052thm.jpg
cc91115d2f0279c62240ebaa0a11b2f2
8ed9f180969cdc8740124ed19cd51ea2fae252e1
3412 F20101222_AABKHJ wu_z_Page_063thm.jpg
418db4663608c16b468d8ce01379ed59
445afa5b6269a98976431e927ffc183d2d87fa5e
36660 F20101222_AABKGV wu_z_Page_053.QC.jpg
5e9bb41a1ff4477679ab4003f045c0df
8965f154f259b57185d4af35cb048b065a6e7998
32479 F20101222_AABKHK wu_z_Page_064.QC.jpg
02af58ad18fd317688c9a19e1bd476e4
8e479f3c8e791704e4a63a9cff7d6f58323f2a0c
8779 F20101222_AABKGW wu_z_Page_053thm.jpg
d3fefa4efdbb8945af7cbe4ad030d41a
273d32175541566af88035b36f096e33268eb33a
7989 F20101222_AABKHL wu_z_Page_064thm.jpg
68323c088e2bc8568f5170eeb6a83d5e
6120c1536d1f63997ba78c188fda38ab69f55d69
34421 F20101222_AABKGX wu_z_Page_055.QC.jpg
2a54cdff0f7797f245733769798d2152
430b59c76f75b6cb959338989d796fb2c6f07f02
2815 F20101222_AABKIA wu_z_Page_073thm.jpg
07f0e0783be612cdafaf3f6e33c30bc3
daeceec1cc5c0eba91bf638b21419fe0539d3fb1
9103 F20101222_AABKHM wu_z_Page_065thm.jpg
3b3a3a934dc85e6d9dc2ca8ab4a1baf7
ff8ff3a069401729e2e37ced357f07eeed90a25e
8585 F20101222_AABKGY wu_z_Page_055thm.jpg
3646d7846df5adbcfe1fea792f32782c
03ab1a8fb829ff5e6b914a9d9ca8b9bf18bb43ef
35975 F20101222_AABKIB wu_z_Page_074.QC.jpg
5f07155f67b7facb99fe398a11308151
3c18fab5dd7e15e938b0b41883002e7da8176d0c
35412 F20101222_AABKHN wu_z_Page_066.QC.jpg
1e8e0769edc828458ef00cbb0ea6fef1
f97fe1ccdee365400fbb1a4f84d1b51a45f892c9
33017 F20101222_AABKGZ wu_z_Page_056.QC.jpg
04ff0c92113d6c1a184d2356e6063162
96123b341d72347b8c6364e17416c98ec0ad5937
8680 F20101222_AABKIC wu_z_Page_074thm.jpg
0b74aaf60fa4731ffc5aefa607eb142f
868c8ad774c39e2227ce4eb1b32b8861926b6a14
8672 F20101222_AABKHO wu_z_Page_066thm.jpg
c932305fde44689c9dd9689d82d45ba0
d0d28e5900ed8c7d96348963781603c1a471a1a0
38551 F20101222_AABKID wu_z_Page_075.QC.jpg
41be038e3b9a3568514a8eb4060f1cfc
2d69163b9ed53bea0c42fda316cb99cc48cf272d
36619 F20101222_AABKHP wu_z_Page_067.QC.jpg
6f4bed38b179d05dedbfa5a54c401478
2d5576212b3d1c50456ef88ddc8b289deec0f8eb
9176 F20101222_AABKHQ wu_z_Page_067thm.jpg
14761e7a07875023d5df01fcd0af3b41
b559b60f746122d85dd3a612930eb15fe785b234
F20101222_AABKIE wu_z_Page_075thm.jpg
b876029982d1451e93ff6aef4369d307
b30d524c45eb9ecc2b91e1823519a1d1d309351e
36811 F20101222_AABKHR wu_z_Page_068.QC.jpg
c4b1a69df5c6e1a55d3738a51c9b7807
6e504b353b69327378d7c4b165998baa0ed311f2
35084 F20101222_AABKIF wu_z_Page_076.QC.jpg
34be15a6f5897222f3373a4165eee3da
59af1288c944595c7b56ce8e3cfe3a1369647e7d
8964 F20101222_AABKHS wu_z_Page_068thm.jpg
2923d579f3488a66c3805c146a7e55c4
f4d311be0372dbc4d11c818cca61eb016ac8a130
8866 F20101222_AABKIG wu_z_Page_076thm.jpg
ea8ebf912b4dafa263f4897931e8476b
e39b9038a6dc0fe10c86ba16ae4ae82ff6da2206
8956 F20101222_AABKHT wu_z_Page_069.QC.jpg
95d533807ebc0d37d171e489f1f19fb0
8a24b7a1fb660583db97206c9486629c03b161bd
34821 F20101222_AABKIH wu_z_Page_077.QC.jpg
69a899a71944d6dafea5889a1b0c68a2
90fcfcb4e19bf0c58dda12da7ed912207965ee97
2178 F20101222_AABKHU wu_z_Page_069thm.jpg
7f4bddf81f736c9ac98eab7b847e4d90
370622150091590aa573a1201c3827504038cb7d
8476 F20101222_AABKII wu_z_Page_077thm.jpg
0ed6495ee3dec58a52f475dc0f98e273
37d5bb9a3116a5ce5e48878e0437a2359fca1463
16683 F20101222_AABKHV wu_z_Page_070.QC.jpg
c4455b5e5a6c433348d766c0c41106e3
d58909743beb1eb408a39200e4db53fb050781a0
9250 F20101222_AABKIJ wu_z_Page_078thm.jpg
f981a354c9ba2e43f3acf89554269c87
1f135f0904ba0b759a656ab146115b5e6e01c888
5659 F20101222_AABKHW wu_z_Page_070thm.jpg
ceaeeb8287701eb036eb14dd35bcf4a3
085acdede95146802f1f1fc05886c198279d98e3
33992 F20101222_AABKIK wu_z_Page_079.QC.jpg
0537a2a2ba93a34b28ab4c039093f3dc
18ea0d88307b2c84366938e7fdb5e7812d401d68
10146 F20101222_AABKHX wu_z_Page_071.QC.jpg
03ae1089ab78d5d08d655557e4131545
db5de2a03de5783f53678069d6d4c7c7d0de8a31
12968 F20101222_AABKJA wu_z_Page_088.QC.jpg
a363cc6289e066019d7a295ceb18386b
bb449a0af39caf91c14f53d6e6c6522f4536cebf
8536 F20101222_AABKIL wu_z_Page_079thm.jpg
14e876b6a2163e3dfb4b44fc5d856ec0
7a77a5415a044eef5a4d67f28624c647f511bc47
9585 F20101222_AABKHY wu_z_Page_072.QC.jpg
9cd0b59223059ebc71df7a92364be348
2856a070832f2c3fd68df33e66b1f2ca0a73e183
4271 F20101222_AABKJB wu_z_Page_088thm.jpg
6c6e86a91637c6b06adeb7d7eeda8645
515ff958cf491f7ff56c518c299560cd8253a5e2
34705 F20101222_AABKIM wu_z_Page_080.QC.jpg
c0c24b7dce134b80e4e01bcc950619ba
c5b63d245396f5a16b4f353ba79a4b33a87c7022
3313 F20101222_AABKHZ wu_z_Page_072thm.jpg
ff3644e1f78d348daaa423e7bc6e0ea8
c085773a657d6ee98494408661dd9aa9814a3dd6
16350 F20101222_AABKJC wu_z_Page_089.QC.jpg
454ae671260357eb28f701bc6ef20525
57dbe47f071e2debdb65e48b0c861f5c30dceb06
8592 F20101222_AABKIN wu_z_Page_080thm.jpg
4cb15acd1d25e305c2caf563e86c846c
400bb34c4b414d91782f6ddbdd6aa66b3c6c8e09
5560 F20101222_AABKJD wu_z_Page_089thm.jpg
9b9b55bdb40928c468604f44a7ab9842
8dfe479cf18367172dcfe4fbde8e8f8ee38f94e2
33722 F20101222_AABKIO wu_z_Page_081.QC.jpg
1fee901e79d01f6e97a89a53d3ed7c59
b473d02d8c3799f89404a5a0de83e6b51f8684dc
7437 F20101222_AABKJE wu_z_Page_090.QC.jpg
19562a0613b1c68aeb5570d5f7979405
db7be78067443106ee461e6cdb6bed58532ff284
8818 F20101222_AABKIP wu_z_Page_081thm.jpg
f4430793319edacfc0b1ff59bad0fef3
cbc24cabf0dca2ce47797072d73bc00c33cc3f2f
34565 F20101222_AABKIQ wu_z_Page_082.QC.jpg
fb45b0b0a09bc5603aa1198bd6da2077
fe7b3e0f8d1d4b91f1c15fa0f07639f3d1fa3366
2671 F20101222_AABKJF wu_z_Page_090thm.jpg
30309403d55ed706c5b415ea829a2aa0
90e3435bf54339be255f1d38d38fff1e1f741528
8912 F20101222_AABKIR wu_z_Page_082thm.jpg
bb6d7357b3efd52e546a6cb416387b77
9aa09530aac0aab26361696ccd4ba264dbd0fd50
9501 F20101222_AABKJG wu_z_Page_091.QC.jpg
996d763317184065f3b37ca8a543cbb2
d14b63e20553cfc81102a0157214c53df7dca4b7
9793 F20101222_AABKIS wu_z_Page_084.QC.jpg
8866463ced54d7ac56e775f0377559d2
1ce4b39abc46008a5ec6d0b1ac1ea15a05cdf8b8
3556 F20101222_AABKJH wu_z_Page_092thm.jpg
4e88cafe1bee8bac88bfcb2227401f0e
f1d08e775630685dbb1f62b488b954ca8f534f89
3400 F20101222_AABKIT wu_z_Page_084thm.jpg
15a27f32c622fd01bba401d072f66a7f
ca83e48a1b12e37110701ad1f612695b1a5db60c
8647 F20101222_AABKJI wu_z_Page_093.QC.jpg
36b949600934d5caa9eb1375df91200c
e52bdb7ff31c33b53fdf67ab1e8d15f7d8b02327
11799 F20101222_AABKIU wu_z_Page_085.QC.jpg
7aeb2a8a727ca504ff9ab7e5aacd02d1
258008d0b3968c35361767900705eb87cea57598
24676 F20101222_AABJGH wu_z_Page_093.jpg
3dcc8da17b9b1f670ecb70136f844313
0c030fc35c676653c5049468c0511a0b0a831439
14756 F20101222_AABKJJ wu_z_Page_095.QC.jpg
04654cc57adb4d1d2ac693e4a8021123
dd02f9e8414506b16c537d8f45816a50620b74f6
4622 F20101222_AABKIV wu_z_Page_085thm.jpg
13fc1f4b52a54d4e90e11712dee2cfb9
4b9dccd0651305b9e6bd668e69707cc6e651edec
27339 F20101222_AABJGI wu_z_Page_091.jpg
1d226f8476d7901e235a33ada519efa0
422095ff7c91cdb794e1ffef1e0974076ecb0be2
5123 F20101222_AABKJK wu_z_Page_095thm.jpg
a7b719941b204c093a186be2aca6a63e
3260164be5183c8fc4da9476b3b9c0f054d66f7e
16142 F20101222_AABKIW wu_z_Page_086.QC.jpg
2b4c87a1ca7ec9bfd8714764eb73dab6
43b2be70334d4645dc953cb3ad56ca4d9a733d95
3971 F20101222_AABKKA wu_z_Page_105thm.jpg
c0f86b20b44abf0e14660b585289e36d
69e92c97980d8dfcb8ccbaa4840fa36b62b33996
106105 F20101222_AABJGJ wu_z_Page_039.jpg
99114e3d54db409c028f793ab51cc4b1
b5be28d2ee927dcff78fed100086d7cf80f15e54
11037 F20101222_AABKJL wu_z_Page_096.QC.jpg
05dcfd389156ec2b0fc1916b02975151
9c2dd5720be0583644d3fb0767732723cc7436fe
5954 F20101222_AABKIX wu_z_Page_086thm.jpg
7020725890b5007b07e171569fe11539
9c42d5faab8b34158c52c99f1dd3ac8128443f9c
30806 F20101222_AABKKB wu_z_Page_106.QC.jpg
52f9f4ef17020694b385578d5fbb7058
030d09c83151b176c84403823dd754f63f31bc84
3360 F20101222_AABJGK wu_z_Page_044thm.jpg
a8c9677a179f7b5c9e7778ecebf85235
4e8cdaf1567c09d8fc9f3732e9c96c1b22add467
3695 F20101222_AABKJM wu_z_Page_096thm.jpg
b3201124222734befa6c3c2673928003
07c3a4a578ff2fc8a60b5b03e8fa1ea563ec53c0
15360 F20101222_AABKIY wu_z_Page_087.QC.jpg
96d3ec6e73bb4a144f48b274cf945ef5
1b4f7710faafb1ec0fb3bb5fa5d6c6f0b875cf0a
530248 F20101222_AABJGL wu_z_Page_046.jp2
4c3a34259be1418f7ad7a4c26929eaba
507059aa1beebdad481bfc42daa578674f49705e
10742 F20101222_AABKJN wu_z_Page_097.QC.jpg
0b260dce17833642d7c49deff224d534
08c9b7643278a640529ae4567ffaf3114551a743
4868 F20101222_AABKIZ wu_z_Page_087thm.jpg
85533faf2ee699069c7a3e610ef4c0a9
fea3c868194da4f2423e07c061b94479b138f4de
5260 F20101222_AABJHA wu_z_Page_048thm.jpg
7b0fc939cf2ffe6cbca5d003b4d7c6f7
d2c693f13ccba59a5a33a07311f165aa268510b7
7690 F20101222_AABKKC wu_z_Page_106thm.jpg
a1eb046a0da60afdf7ed2ef147ccbdf7
fa8fceae7d2411c709291b121411f85c747f949f
1755 F20101222_AABJGM wu_z_Page_012.txt
4b37d672cb2299c11decc1013687f78d
700760672ba76b67825c5d2c9660a53ba89ebee6
14217 F20101222_AABKJO wu_z_Page_098.QC.jpg
98efe37e8232a7c166cc27b221d1617d
ec64c94c082df776fd5a368218fa8b40e59de9b6
25271604 F20101222_AABJHB wu_z_Page_079.tif
62b78850294c57999dc8bf156ac09339
78c9ff4ccbd3bf087b49156dc98e492be7fd8ae1
36608 F20101222_AABKKD wu_z_Page_108.QC.jpg
dc173051b084f9c029a729aa7d6ff128
7eb7ae28be8671172b4f46d09a9f02ece170a284
3022 F20101222_AABJGN wu_z_Page_091thm.jpg
d8230800791c49680a8bf37fc8ab7bc5
a56af89fd448bc6124d0047a4c27fed4a097a459
4368 F20101222_AABKJP wu_z_Page_098thm.jpg
852d64b4110e6baacbff8d4747eb5fcb
17fb26f2d9c6ad767b0c828744a5333528057e4d
1051940 F20101222_AABJHC wu_z_Page_037.jp2
0d57c4219706ecafc189a86e8c5363e0
fcd65ec5c9c45568968bf22a5c4361fa7a727d60
9281 F20101222_AABKKE wu_z_Page_108thm.jpg
115b870d831f5798a6d63699c0444b7c
0a2141a7b3bfd606eb8234b0fea755be88de2749
33413 F20101222_AABJGO wu_z_Page_104.QC.jpg
ac54371dce4471aa6a07afaf626370d8
b61a14bc68d9edeb6985be0bc0d2b4746cf45e35
13353 F20101222_AABKJQ wu_z_Page_099.QC.jpg
6bd4d450cfa1a87cab8103f2dea1b9bf
9302089d93410738da767f7c1fdce8eb667cdfa9
523449 F20101222_AABJHD wu_z_Page_051.jp2
cf0c32d48cc47c2e1190ad3df43cba4a
63f25d6f6e1e0e1038cbc21b2569983b32ef4b0c
36068 F20101222_AABKKF wu_z_Page_109.QC.jpg
c9e349198f8e7aa0d27c28f7115c76e1
7efa5b940cdee760a3c5b72a07a50fe67e39ac81
7939 F20101222_AABJGP wu_z_Page_005thm.jpg
81f43351bf8fae907dafd7051af0a505
e6da047e0363009ef0b256ac81a2c317d3c72b72
4456 F20101222_AABKJR wu_z_Page_099thm.jpg
44f703d12310f936a8a26ff12d255e09
e2f5905abcec2adc836ced1a45ed41922f1ca921
24975 F20101222_AABJGQ wu_z_Page_004.QC.jpg
df229069c3633551a08b34a09cefc45b
4cea895249c219c1fac271d2f172ec1e2ed961d7
14407 F20101222_AABKJS wu_z_Page_101.QC.jpg
cd07e20e73317698e6266d684e57ea11
d4d5f5c85eb89045c8f443163c5a2c1611fa4d6d
1175 F20101222_AABJHE wu_z_Page_002.QC.jpg
316cb380dd1f7c91714f440c382226d6
1405ea7810efc5101157c30fba5fc34a7ad26c32
6832 F20101222_AABKKG wu_z_Page_110.QC.jpg
a38700c9b156628dbbd57dcbcfe5d2c8
3167208edd790c68666af8ae7e2829d49c606e24
8457 F20101222_AABJGR wu_z_Page_017thm.jpg
5cb014461b2923f3718469e750adc0c1
b3051f159cc0b37e02fb3f788511f259ae90fb0a
4340 F20101222_AABKJT wu_z_Page_101thm.jpg
9c3d9fa47aa330c9d39cfe2181171753
54cd5b0ab76c3799ae4fc7f1ae5f24d302fdb56c
213 F20101222_AABJHF wu_z_Page_007.txt
af9490eebd839930d374f0697db5c1ef
d0fc8616832c90789944c2cf4e7becea8ac2e256
24793 F20101222_AABKKH wu_z_Page_111.QC.jpg
84114b738166b6661b261934d2309c9c
bc40542a12c005f28373c0cf4da1575d82be380d
75930 F20101222_AABJGS wu_z_Page_016.jpg
82169282b85ff79f60c892ac5e9cc858
fe1453f90cc805ab03abc14aaf0ad8c81d2ddfde
35282 F20101222_AABKJU wu_z_Page_102.QC.jpg
8ebcf241ebe99e016c6defb7818ef42a
bf72ed815d1891430a9650048b548df604dc80ee
2147 F20101222_AABJHG wu_z_Page_102.txt
48aa7204614e2c28129484867ca7116a
df29ca91ef371a0db9efea05243e823a3c665ed6
6375 F20101222_AABKKI wu_z_Page_111thm.jpg
e0207dc93d658e380e5ef863e4882980
2e282188f64425b0571cfb5ea54036e54a87f083
1466 F20101222_AABJGT wu_z_Page_111.txt
85578cea9f23243fb8ded3409b688b83
77c7709f9bbe09d1ca328200b9367496ca30bf3c
8545 F20101222_AABKJV wu_z_Page_102thm.jpg
e0566df43bdfc6cf71b3df91f7d6290a
b4b135eff0ce8d36d70c44850788cf9fb2ceb4b3
36310 F20101222_AABJHH wu_z_Page_038.QC.jpg
10254fa27f69402942d8fd969457e595
dd3297392a9322b5dc5d3d71b728017978a98ce3
27046 F20101222_AABJGU wu_z_Page_083.QC.jpg
6e6e1b16d26e37e709d4e632cbe728e9
a4f0376d9581b01c8c9ac44a6b57a2f6b87134ec
35335 F20101222_AABKJW wu_z_Page_103.QC.jpg
25bcdc9b514868df13cc105f44983b60
deb5d94677f9ec0121e2f2caa09f15df9986b626
2193 F20101222_AABJHI wu_z_Page_078.txt
859a2ef2b7c9d203ac4b8b55968f5e96
d43205a46f43ec31876d87d18ca8794dd96c6827
2519 F20101222_AABJGV wu_z_Page_108.txt
b7ad38b13fe4e5a22f4c5465663bedaf
744a894c1a56d1e005697ace4ceecf7dc12003d9
8807 F20101222_AABKJX wu_z_Page_103thm.jpg
0329960737d8ee7985ab459c4bc3bd17
4585dca191a3200200c233d0269d632b6d0d6cb8
9110 F20101222_AABJHJ wu_z_Page_107thm.jpg
3d06448280df5ef35f43b416573d3573
67a3fc3855811220853aa97818964aa64b52d312
2198 F20101222_AABJGW wu_z_Page_038.txt
92b396b8f50986d0ee4cee3a5757321c
57626131a74cb91e19a9a9a676d31db357fd6f48
8421 F20101222_AABKJY wu_z_Page_104thm.jpg
61b43fa6d60cee454b9e6e3060103cb1
47eda5dbc1fd1c42beb2df5af26908c2fa95bbb0
F20101222_AABJHK wu_z_Page_058.tif
7f2d73d74c405b34141f2410f0ac5ce5
de7e31cd6bb103e742b5db901453f58ad28524ab
1771 F20101222_AABJGX wu_z_Page_110thm.jpg
5e423e3cd6bba54b056e5ac968d9c923
5073bb84d865edb9ee02118a3d938abc9ae3438f
10809 F20101222_AABKJZ wu_z_Page_105.QC.jpg
2551cd02a9a9c9febf6dcd3e4fdebf51
97c30a8aef23dd472daeadbc3a051814fdd43a46
2289 F20101222_AABJIA wu_z_Page_026.txt
03457336dcb3496d8afa852b27f33478
3ba208ca6d7e6f80ca88b585d325c16cd59e1d2e
2863 F20101222_AABJHL wu_z_Page_061thm.jpg
9d67fd4b5636028d10e2a79cb65a7576
605dbdbbd0a75fbb91739303da345cf0e7b0ec96
107165 F20101222_AABJGY wu_z_Page_103.jpg
a4ab8dd2476ff352025019e5bb772f27
99c6fcec643ec74bbb9675dbf1cf7d5df5ce78fc
4446 F20101222_AABJIB wu_z_Page_058.QC.jpg
cee8c971fc6373bb31314ee4c35f98f1
15f84c9755985de54d2204b885500b21e517e414
110318 F20101222_AABJHM wu_z_Page_074.jpg
630c3398b1c284405ff5b8d68491b146
74bce98a3bebaed5df70d996b4f44f68faaf02dc
22001 F20101222_AABJGZ wu_z_Page_021.QC.jpg
37ff3c24ceb5d1002f4b568d07f9dd19
06ffbe48cb4bcf06de8724eee4101e16b92b0e92
F20101222_AABJIC wu_z_Page_054thm.jpg
8fe35a16e29e857f5dbfaf4d51a7118d
248e1229faaaa9a6dd827d9b794ea37501643b84
F20101222_AABJHN wu_z_Page_061.tif
5ebb65c2dbdf4b37e15e55e6654bf4e1
a0162f589bc7e4945ebfb7e16a85e0f7ff692321
802548 F20101222_AABJID wu_z_Page_004.jp2
02598aa54f661ba545847ee73f3c006c
fa0ad31ef435c68c01404d71456937631bc64113
25824 F20101222_AABJHO wu_z_Page_072.jpg
9c7214748839875ae697c0f293ed1685
52d41cd180fce29044e9ac4029929d2dcd0e58cb
105108 F20101222_AABJIE wu_z_Page_058.jp2
c7d06953c2b284bdd0cefbc39a3a4b28
a2b2fc9fded4b00b9a875cbb9116b6fd90708756
38656 F20101222_AABJHP wu_z_Page_065.QC.jpg
42dbeabbddc31f39c0b64115ab9fc215
3f9671e2b3fd708416f173480cae534b38bd8415
272829 F20101222_AABJHQ wu_z_Page_071.jp2
32d0a50d9b73e4915603f450cc893293
37b851e127e86985c2fc9e208f81c01645d32e66
277361 F20101222_AABJIF wu_z_Page_001.jp2
2ea7a20ef5ae0f0af609408724aa7ab3
215460a5351bb98affcb19ed7d5205c42ca3bded
52124 F20101222_AABJHR wu_z_Page_054.pro
5979c230035b5725d155be6a002388b7
649cd9fe00ba5fbc7d9d312c124b0a502a333ca1
35894 F20101222_AABJIG wu_z_Page_078.QC.jpg
8eee3802dd172c0db4cbe1b447a6a593
c0f3b9297aeca03d5270d5bb695be67ce53f529e
475251 F20101222_AABJHS wu_z_Page_099.jp2
2d3c4aa7a5d72e9af36f6275a9754f1b
d9384339be1b03eb00c9cba2c2713fb1711c61b7
250756 F20101222_AABJIH wu_z_Page_096.jp2
3e905a4c81a5e711d5859969d63ed46e
1b251ab0cdc2c982cc4cfd59007bc0665368630d
10255 F20101222_AABJHT wu_z_Page_063.QC.jpg
f01ad33f741867ec8721bc557fe09d3b
1dd6720b82366cefe7543c6ac68863c25ede7f84
21174 F20101222_AABJII wu_z_Page_043.pro
00f15fb12cf13077753362e0096ff156
61635f1eb9465cbab9a678dec8aec8c3707ece3f
F20101222_AABJHU wu_z_Page_064.tif
d17113157580a6bc94139ac6ac491ac3
499a6fe17291884a36bfc3a94903fec7dd5790bd
229768 F20101222_AABJIJ wu_z_Page_084.jp2
deeca51151b6b28c3cca6a27d935db16
b275b845f8ee1675a54ff6197f99ca4ed293abb9
619 F20101222_AABJHV wu_z_Page_096.txt
c9f5fa4a51f57f0d688e6490e3174ea4
e4a9b0a51b2c56182d74493b82a3468a08ad71ba
11756 F20101222_AABJIK wu_z_Page_092.QC.jpg
cb99736e6ad36be2784b210fb7ec27b4
a54edc6308966d6f09944f6bea1c2f02da686126
7621 F20101222_AABJHW wu_z_Page_049.pro
d64175cbc7fe1ae6831e8a70ed1a84eb
e2e0dcb9f34c78fecb99522d46a0bc7ba9b99b86
38726 F20101222_AABJJA wu_z_Page_094.jpg
b93adda0cd6f4cc05706eeb7f534378c
657709bab63759a09e26173848768def02a04571
462 F20101222_AABJIL wu_z_Page_032.txt
e1aa7792f2c845fb8b7839282887e9ed
bd9287d3d0a83d1f9181d9c8241d8d5ed88d008a
35902 F20101222_AABJHX wu_z_Page_054.QC.jpg
b9bdbeba5168536086256f6147d758e2
dae13c6cb705f4a58c9e006052110e3b6a1ef519
176 F20101222_AABJJB wu_z_Page_058.txt
5685c8470f562954b500af3a79090d10
256e0dd0eedef75c340eddf4299e135920da35b2
33918 F20101222_AABJIM wu_z_Page_036.QC.jpg
6291152fbf83ba4be7751fba5f7a03a6
4f71bbce29599b5b290100b8aba22a9b0071a939
258 F20101222_AABJHY wu_z_Page_094.txt
0631bdbf758435c2b6b5816ec2c71437
2ffc5472463638f9e13e7c1c62759553ff77a5f4
414 F20101222_AABJJC wu_z_Page_097.txt
dbef97c650582ee50047e1408972d700
ccb206fdf1a3939e30a6e4c50a66845646c17cf2
F20101222_AABJIN wu_z_Page_106.tif
6b37c16fe1c1106d4e3f228f7431986e
8a44f4ce3db83e7721d7d4f1edb4c3ab8e9f46ac
F20101222_AABJHZ wu_z_Page_102.tif
5104cbe39ee2de6e15cc0f40d77bbffa
49f978278f60ee7a304e509326da7535caece1ba
6697 F20101222_AABJJD wu_z_Page_083thm.jpg
15e677057c2a97a7ca7cab78c470b2df
2b13bf853c117e39f74f0b18de699a3c344449d3
369563 F20101222_AABJIO wu_z_Page_094.jp2
a5ad4e83014788ebd67d9c28b6487891
f445f4e395ab8fc1e0b95a678d82a85286a46eb9
108370 F20101222_AABJJE wu_z_Page_023.jpg
0441241456a3e191a25b09b33d0166db
1153c2099880d2d2fc1806c34ca446dca1659d2c
8089 F20101222_AABJIP wu_z_Page_056thm.jpg
171832dc37dfa76b85d7a25a2d7ff856
f79949e55a28e60db790b0b9d32f329e0d20fd1b
104286 F20101222_AABJJF wu_z_Page_055.jpg
c57943bc22e5cc3df72613728846d75d
62a3c32dbd7b055d0ddc6250e4ed7c429bd6a694
105193 F20101222_AABJIQ wu_z_Page_028.jpg
5ae8865208d87ac2d8208e2783287d52
6d94d4996b9d8d593c4ec5f816b5d5ae6cdadb42
1051977 F20101222_AABJIR wu_z_Page_053.jp2
38f2ae30b141744d7ac489937229d0a4
9819edf9907fdef88cc152b84c741b08775d5e80
F20101222_AABJJG wu_z_Page_084.tif
04fc29c691fe13e501f1ff0a06b50bdb
5ba5d76b72e58df82f5ae66c6df991c3698f8399
41580 F20101222_AABJIS wu_z_Page_087.jpg
16c7a823ec87f384afdd57e321517e01
93cfc4ed14a3b0c8190d78776acb6a0aa6bc0063
2036 F20101222_AABJJH wu_z_Page_036.txt
f981c33cbb8518e48a89226810bc8ccc
c67ef2806c5e544ab81509d659d58302b2733a3c
F20101222_AABJIT wu_z_Page_053.tif
72e20075116fc1f717a6b384cf8daf4d
bc5a81aa9fc43c1c9e1d4dcf13ff3043f359b5d9
F20101222_AABJJI wu_z_Page_096.tif
5552d0cb68248fd8c76774c4883f39dd
9fd784ca6867d25fac91bff4992b878b514b5354
3079 F20101222_AABJIU wu_z_Page_093thm.jpg
72c9617c6274dcc4536f222bd4f42914
b43eb1053ce23ae2a63aaaaeb4fddf5c1dd851ec
4098 F20101222_AABJJJ wu_z_Page_094thm.jpg
1e4f5500de66e58e3891e7bebb65126a
d61dcf6e6cb894bc27b76dd53e25e7a07bf73d31
271349 F20101222_AABJIV wu_z_Page_069.jp2
4bca9a8f93db0fce80cd632932ad8775
42e2b280aeb956a8fde51173955de05d90fd6a43
F20101222_AABJJK wu_z_Page_052.tif
9f51ca2ec26649e1867de12eb7677ab1
9ce1acbaf96578f1e3349e22667893f7737eef92
2148 F20101222_AABJIW wu_z_Page_103.txt
f6cc1c868a4763ba10a02a180cebd159
4c34951d5462478e927795344f4612a1ce017f49
6673 F20101222_AABJJL wu_z_Page_089.pro
1513001fd33cd569031740dc26bf853d
b807c504967eedb130e40e7bb1ecab544d4cabbc
F20101222_AABJIX wu_z_Page_055.tif
26c11621ab12ce734b4d9e5d5a5f38bf
eb8f2593e311cdb6442f5cb6bd897b6ef313591f
4452 F20101222_AABJKA wu_z_Page_072.pro
706c04c61dbcdb4a32fcac7911cda6ef
57990aae5e3a8e4eb571a4bb874cee3cbffd1514
F20101222_AABJJM wu_z_Page_027.tif
df7fb513dde3d8771788ffed944cd27d
7f43e3283faf307b1c6051ef15ec3f4df5e19ad1
840 F20101222_AABJIY wu_z_Page_043.txt
d2d07a84a6910e280604b5de644a77a6
51a45402f9832979673c773c09d36e703dbdcb1d
F20101222_AABJKB wu_z_Page_066.tif
0b9b7bb6f50bf9ff5a978fceee72326a
a52f1e191b330540b27853aa862bf535aefcd344
2417 F20101222_AABJJN wu_z_Page_006.txt
e20d06311232cbaa48c9ac71955f8dd1
7f867e7d4ce412a43792874f2f27e1cee635d8fc
25322 F20101222_AABJIZ wu_z_Page_002.jp2
d3593de58d2f29cc362f4cb504080e13
70e66d2644af143861549b92727ba6d8a8e3b59f
34228 F20101222_AABJKC wu_z_Page_045.jpg
744a12106e758c85ac971747bdd9e7c4
a4c70409296d6a0b77cb97ef89b211a84ee40bec
8980 F20101222_AABJJO wu_z_Page_109thm.jpg
6139f2ffd0c1c377baf744b0c17ce456
f7a772ccbd001b4ff82ecd96301aa872a4f9d927
460289 F20101222_AABJKD wu_z_Page_101.jp2
a6ef6693c8a4dbb42cec46b925669c5e
3236a40325bc1af953875a5678542786cb32e130
1955 F20101222_AABJKE wu_z_Page_015.txt
1a705582644c8a94d5751160ffafb82a
cdfbbe90a715b91029790b67dcde008b765bef2d
11882 F20101222_AABJJP wu_z_Page_058.jpg
75158a177bc8cb5d2b1a8ab280a22ec2
110a196d6b78208769bc46cd41b8ec14c7664302
15640 F20101222_AABJKF wu_z_Page_100.QC.jpg
110065f1e79b2055b78a443343d57103
1c125a4445a2bc4455ab18e25cb83b80a2977261
8689 F20101222_AABJJQ wu_z_Page_023thm.jpg
d85a821f8721851d30d3b515c295fbde
8fdfd5c598678ef90dae7f79417ff227cd18564f
8677 F20101222_AABJKG wu_z_Page_001.QC.jpg
6bc02c3dbcab79300092804b19587013
b189ab7defad5275e0e01747fd326a21779698e8
35879 F20101222_AABJJR wu_z_Page_107.QC.jpg
ce4e728a3c9fac76bfdaccc3743f1bf8
b9f0b006f40ef3db0db32b334d2ae03e6d34eb0a
27187 F20101222_AABJJS wu_z_Page_044.jpg
85965c0012596c12603aa246e8c54c75
62c4489ac600ae01b565aef370c8b93eb190fab1
354483 F20101222_AABJKH wu_z_Page_087.jp2
1ff55403440e62a086e057179e3ddfa9
1781e1f738d72b8dd376bc193e7eb93963142d59
490861 F20101222_AABJJT wu_z_Page_070.jp2
0e0f4b5ef7497947622b52cbd25b71a1
501fad0140bb0d9225199ad0a1597b3f4c71cac0
13783 F20101222_AABJKI wu_z_Page_094.QC.jpg
982fa521fd04887540e3d5dc7b972533
7f6cf12d21d0ab8ab5c506578babe04777c5f518
95 F20101222_AABJJU wu_z_Page_003.txt
617a629823caa9a53b898b685ea23e6f
cf34c59d3223afc565806105ca634647534aad65
31474 F20101222_AABJKJ wu_z_Page_013.QC.jpg
5fbacc09450df557451ee3f440d4f6ae
4c4afa9072e2cbd0b278d57ade630aac50748fea
7124 F20101222_AABJJV wu_z_Page_010thm.jpg
ee86362d2c828dda78e19832e7da3a89
ea7a8d061d96aee3614a96be0cbe8a93647ecf1e
52440 F20101222_AABJKK wu_z_Page_102.pro
d7e1f71ca8f8bd5c2aa5d48881eabeae
4b6b0df4112fa1efa657a42111f606c08aaede33
1051972 F20101222_AABJJW wu_z_Page_066.jp2
a969365d92029c4214a83886b397ba8a
f79994a42e4915fe46cc47ca30f275da3c8bb5aa
F20101222_AABJLA wu_z_Page_067.tif
705249ebfd7fc4ecbd3f4850393ed5bd
e180200863b1aefcb5520d36187589b8f6f0f4d1
F20101222_AABJKL wu_z_Page_022.tif
9793fb5f8c9936386216bfec4f991032
5400a3be722b62d601467571dae609a24a0e8220
F20101222_AABJJX wu_z_Page_016.tif
b5d65c4dbc13b046c58ecf4368220199
39827337a52cc0bbd667849bfbe431e674e414c9
54625 F20101222_AABJLB wu_z_Page_053.pro
5d5c914e7265ae9404e9801e0a3e1d75
b191514dd66f694539a48301a316c492a6e48165
9071 F20101222_AABJKM wu_z_Page_037thm.jpg
7f758d1cefae663aa4bd0a4ba54da6a2
ef6367aa6a4fb76df55061422aa268b873079406
28245 F20101222_AABJJY wu_z_Page_061.jpg
24aaeeef674a24c7e1799f1354241e25
4ec281c997d5edfe2a5cd269cd986b4f99ae4852
419197 F20101222_AABJLC wu_z_Page_086.jp2
d2dc213f96b96b5b01fde9be9475b61f
a07085b19d1e788969ee06bcb754c6259b7a48a8
153138 F20101222_AABJKN wu_z_Page_009.jpg
6643a90e6646847bd5bc8493f6d89ed9
77e843810f016d1a1f77df780ef1a604febf1b7e
37386 F20101222_AABJJZ wu_z_Page_034.QC.jpg
93764c5f1be0df00dbd0566b25caa8a4
24326097802ff0ab953270732bfa6f9f561ed021
110808 F20101222_AABJLD wu_z_Page_053.jpg
e5b5475ef2479dbda964a32f7f6c90a8
fd8f5a78ec8fc5d3dddba9f8c91799e41a16a1a1
24491 F20101222_AABJKO wu_z_Page_110.jpg
1dccd7163e895d357df778a03b4437e4
4191b9de65e9ca783a66589fcc26cd3ef25a8b89
106131 F20101222_AABJLE wu_z_Page_036.jpg
bb95a4357f0686b34bfb4c97a67bd5a1
7902c34df6314c7ca85c0aacefa71c955c294740
678 F20101222_AABJKP wu_z_Page_059.txt
c3636ebf61b4e0077a8969b88bd094cc
bb99e3e8940aabc91db841ab574c4c532ab6ae74
7504 F20101222_AABJLF wu_z_Page_073.QC.jpg
04e5d88f447b34772a3e4407625d059d
30c7ac6682b7d4a9a86db84a5398a7a82d440859
8968 F20101222_AABJKQ wu_z_Page_086.pro
4c8dd3df0859363ad6de3671dbfe5575
78ae3a11a398f25767cc028c03315504a5c986d3
6354 F20101222_AABJLG wu_z_Page_021thm.jpg
cb8b530674833ee0410f1324533ed241
de5f0e097a64f997978278ca0e40d415942eaefb
1974 F20101222_AABJKR wu_z_Page_056.txt
1e9555faecc8fa38a4a4380e313f7cfc
df2670fdff4fcdadf0c769a2749904a755540766
34509 F20101222_AABJLH wu_z_Page_003.jp2
c7c1cc8cc72f20afb5bd487d47397550
f6aa6f1f1f5c625930da13e65138f568472a2c87
9156 F20101222_AABJKS wu_z_Page_041thm.jpg
1cdedf554755a311594f72ddda1800dd
92529d30b35f11c3a6c1ca23683b5012db61fc8a
8630 F20101222_AABJKT wu_z_Page_057thm.jpg
6c726b0adf49fe88e1b9ed63138ba3c2
2a15ea7dcc52f331047111baf8d57a5791eaf948
28629 F20101222_AABJLI wu_z_Page_010.QC.jpg
d6920d07569f747d888c41523e8e5d28
cdd2016c261d49df0d2a2848208c2cfcde6aeade
1998 F20101222_AABJKU wu_z_Page_104.txt
34d8fa859d685d935406730b75dd4f37
28b7ee5573bb486bbe69b5f469ebf1ee45676075
5040 F20101222_AABJLJ wu_z_Page_007.pro
0dc8ca6a23682cbf459c786565916c7b
88fe9cec70dc417369df1d4923e1561535375817
5274 F20101222_AABJKV wu_z_Page_031.pro
e957738ffc11e7923d0a3c374781d9c6
9713c1ac4a2fb6e32a480a3d90b96360dac4ebca
5943 F20101222_AABJLK wu_z_Page_006thm.jpg
11299af1c8a45ecbd23d2ad5455d29f5
e8cf1005b4175516f2d00e615e6acda04f217151
2071 F20101222_AABJKW wu_z_Page_028.txt
88eeb193684c24b99c572cfe2042d933
60665a6777cf8d46ac6723e19ed786e1a88030b6
F20101222_AABJLL wu_z_Page_042.tif
fe98d30cba85096b26f350e610ba6957
39d59d9136e6caf7a8caddc17282b7c08af8ee6e
3730 F20101222_AABJKX wu_z_Page_097thm.jpg
c79aee6a02bcd83d503bdaa74652230b
8573c5dd9ee3424f0e5f1eaaa50cf54f040d7002
75738 F20101222_AABJMA wu_z_Page_004.jpg
62d005dc3e510b578bc27ecfd2881e36
6b631e0b0da00f802e4804409f69cfb20e2f8e89
5041 F20101222_AABJLM wu_z_Page_090.pro
e56d3e5803a970bee44cd5826d4bd552
0cc263263d3393a24cbef93487ab42a11dbdb5a2
F20101222_AABJKY wu_z_Page_002.tif
d9484cce80e12a4d37adad33654d2188
1c666159455bc024b1471c783473ed2d398a6cb6
134923 F20101222_AABJMB wu_z_Page_005.jpg
3ea0c564afdf9993028aa17a03d677b1
874a84130780a9bb170c7c8695482500929dc8ce
3478 F20101222_AABJLN wu_z_Page_094.pro
6dcd0d038c3306253473638dc6c4433a
f2eaba921bc3e87a8fc31a4b49eaf3ee28a0e3b5
580128 F20101222_AABJKZ wu_z_Page_089.jp2
1b05aabf438efabaa847d73585476bde
de51a83f9fc0888d53ab2d15edbb4515f6f9d4b7
104539 F20101222_AABJMC wu_z_Page_006.jpg
0e3f211efe8ea8f75c7b0f3b9db8e7b5
7e6c6ecf4ee437dc279a590cf8a59df289f54851
521 F20101222_AABJLO wu_z_Page_048.txt
09394df2163bb4e2fb4a9d22e3c7c811
1d80ded1e67477d9e211e74083a88099846acf6c
16291 F20101222_AABJMD wu_z_Page_007.jpg
b3e04b940ee4d74574d1e05d3900c638
0ff46d278d74c16e691ec0c0032e85c212641146
4774 F20101222_AABJLP wu_z_Page_100thm.jpg
f7696a7882ed6a8c356b3327d615faee
ca76abb7930171d615742f60fe954f96c8fbcad1
155694 F20101222_AABJME wu_z_Page_008.jpg
1296753ed620ad49e02e5f3e04b4c9b3
78a4fc89d534dc6abda282758a7bb6d66b7c3175
19157 F20101222_AABJLQ wu_z_Page_062.jpg
dd0a53f425296f16cb048045b39df9ed
2675ece279e565bd4b27728fdca405341a72f21e
106710 F20101222_AABJMF wu_z_Page_010.jpg
113a13602e397aff399df7167b4465ee
2ec695d02c708ed1ca3576b59cc3b477a56c3368
3469 F20101222_AABJLR wu_z_Page_071thm.jpg
ef57b73f1cebb35649dbd23e6b2d7062
157af8ca0c8353a309e24916629bf2aecf0cc92a
22175 F20101222_AABJMG wu_z_Page_011.jpg
f466f7604fd7710a9ad1206b31101a39
524d612a3e26fc6a0c1637c5c70979491a3cd65c
1051956 F20101222_AABJLS wu_z_Page_014.jp2
4c2474aa35419371dc83fcafc976d85e
59a3f438b9afbae262a8bfd4b1fc6361b7954393
86067 F20101222_AABJMH wu_z_Page_012.jpg
52fdcc6256019a87cee14a7308bb4f41
c3a0443644e3bbfb94950d176e90191b71f30162
9031 F20101222_AABJLT wu_z_Page_024thm.jpg
75b4918228ae6854197250fab77cbca1
8516bff888ea7d8fb4f10c518c35e5548159df33
97415 F20101222_AABJMI wu_z_Page_013.jpg
105319bbb1336e8ef65dfe4c413fd5ad
a4f019ac222d8cfc6ffff03d7f7103b25b0bf3a7
163950 F20101222_AABJLU UFE0021726_00001.xml FULL
add84b3cdb03c21e13ef8c054bb4ed5a
7f745cc3de877a3853f47f562d6cb35edf1ce223
109124 F20101222_AABJMJ wu_z_Page_014.jpg
d65ce269367a339038f2699c068a7253
8be0b8740cf0960720e5a9b54bc77316fa746c5f
97798 F20101222_AABJMK wu_z_Page_015.jpg
bd105ddd607c34e55e0de195834f489c
55368291befbf0cd7b848c6a886944f05ff782e0
28395 F20101222_AABJLX wu_z_Page_001.jpg
96ee9ae08357cfbe62fa4901b9fa4809
866e9c18c3f6244687b75a2d3596d98d0d86c61f
112739 F20101222_AABJNA wu_z_Page_034.jpg
42a2bcb5f98cb90761bc98f4106b5414
db6744e107ad82654ef6d51b7d4400fc7a4daaa6
104638 F20101222_AABJML wu_z_Page_017.jpg
dae91a1288314791e8f3cbdc99f8ba2f
4045067919d84cb86ce4743868653403be3c67e6
3952 F20101222_AABJLY wu_z_Page_002.jpg
80be2f03d70ba32cb7a9b2c3920eeeec
03800fc2ad9b42caeccc2f7e6c26a08036048f05
107127 F20101222_AABJNB wu_z_Page_035.jpg
13babd8b1c10d48e888b8709eef4e122
9f969788c09814fa9e75153738e6d2494258409c
28812 F20101222_AABJMM wu_z_Page_018.jpg
2704ba099017b626501f672166e8292c
0353916242b8ddd4c89a232b9243abb62f2db4d7
4736 F20101222_AABJLZ wu_z_Page_003.jpg
8ab3e3447a2b754382924f7280849cc2
d4250de94ea0c45aedabe4363bd2a73c38296d85
111680 F20101222_AABJNC wu_z_Page_037.jpg
e145f37577430a4d28729e6307fe5d0c
ca328a7a613ba70aaa9b9e644ec8ba61183d929d
52068 F20101222_AABJMN wu_z_Page_019.jpg
6f80185f5e10983db531ae7f244db205
3d2623851371c21886064a7edf8f788aeeb4e575
109889 F20101222_AABJND wu_z_Page_038.jpg
741832e020efac123f989d6ff2a57262
be0edc3d34df583249e75c565221a92f74c029af
47190 F20101222_AABJMO wu_z_Page_020.jpg
7bef73907ee7b61402c7e46d575cc454
8729d647055d401f7f61441b6ffb570d85f13944
111298 F20101222_AABJNE wu_z_Page_040.jpg
8062a061de68430f443836dac3816684
67f9055493c731b3a37ec1c2ade9f56c1cd5cf69
71355 F20101222_AABJMP wu_z_Page_021.jpg
13d03c0b23840bd3510a10432385f928
e082d78d0201ca90b1841a82431487a0f5e9c666
110276 F20101222_AABJNF wu_z_Page_041.jpg
b14b36146f480d8cb72c9d9da42e9ef6
0b0c0348b82eb220e3e7b35dd17d400720f495b7
38444 F20101222_AABJMQ wu_z_Page_022.jpg
d1688d5f75b96b878716535f3fc42a19
c3d94fbc04cda039fb93bc95da32ff632dbb4d54
100996 F20101222_AABJNG wu_z_Page_042.jpg
491bd84e40912194d557a2cdda37f787
68e32983ac237611ccc17b1be7c2c876a945acea
111245 F20101222_AABJMR wu_z_Page_024.jpg
f63715bc366cc19baffa1776513c8ad0
8447305947ac6c7f76ea280fac06eb187856ca16
46236 F20101222_AABJNH wu_z_Page_043.jpg
027974878b310952740bf554d423dd5a
3cca5f32c26382ce0e1d9fb04aca96b3f60a17fd
109040 F20101222_AABJMS wu_z_Page_025.jpg
403661e48b98cc1b68005bf775d1ca4a
01063484997eaaf6c85e3905d7ef8fa97a4835dd
43570 F20101222_AABJNI wu_z_Page_046.jpg
0fb44acd5678b2eeb740d2e7233ad66a
c766fbbb8c495567cade9e146f8b0ca23236136b
115154 F20101222_AABJMT wu_z_Page_026.jpg
f61bc9947be9d8596ba05040606abed5
bc0101426b2bcca61d80c920861d81423a3abb68
36613 F20101222_AABJNJ wu_z_Page_047.jpg
58bb3b8843c05412e746b0bab532f9f3
d90df78ca5bb83d9408e241fe92954a7362acf29
110408 F20101222_AABJMU wu_z_Page_027.jpg
2beb74df32b6d037d55910fe164049e3
648ffd1f2dfe4f83bfad52a6c94c9350bffe15ed
108374 F20101222_AABJMV wu_z_Page_029.jpg
a89cba162184ae685b2792619d0cf919
81854cf7f415fd845dff5ece110eb568fb341745
55968 F20101222_AABJNK wu_z_Page_048.jpg
f2cc7d0238556a434bb15d9dbb18d64b
194cef767fe2294a54ff0cc51cf4802544862daa
86527 F20101222_AABJMW wu_z_Page_030.jpg
966b19f2e0705e3ed5340d282d56850b
4e82b9bc3e49fe2430d9d268425de47cd9dbf942
27770 F20101222_AABJOA wu_z_Page_069.jpg
905d6aec912e9a1baa6b3dbd27b778f5
ffcf560eae862970a2fbb1ecdc7ade1dede73c27
25434 F20101222_AABJNL wu_z_Page_049.jpg
b5cf381d3d58957246a73b71310b8b81
394ff26ef956f2ca33c8bb71d91cdce17a513543
41858 F20101222_AABJMX wu_z_Page_031.jpg
5600b8242e68d681da38eb0232f36b94
3cbc6ebc2074537f477478be0f81c3ab0dcd0737
50354 F20101222_AABJOB wu_z_Page_070.jpg
9646d275a251327f45717025b53ca367
357cb6e3ef11d917857a73f3a532f64869fba87c
24612 F20101222_AABJNM wu_z_Page_050.jpg
33e93bb6008064162e0e016a2dffeae8
d97c6a2f7c0992859c1ff92660879105ff491540
36555 F20101222_AABJMY wu_z_Page_032.jpg
68abdcab84bb53cd7691b8c8c46b7e79
245601dbd8b2524d596ca519184c270461997509
26368 F20101222_AABJOC wu_z_Page_071.jpg
07e5eb93c6d120e5235cdaeaee8addbe
7ec0d5970c30c7a44497332b5ecedd476fbac44a
47103 F20101222_AABJNN wu_z_Page_051.jpg
6ebd67d80054e411c4b246aa10d6c990
af7cef9846ac99857761b2c1d6d715391557ef1f
65913 F20101222_AABJMZ wu_z_Page_033.jpg
240ec90fd3374556bc22ff33914dd533
cce9e2a8ee613132d91942ccd2cf3693c3e9a90d
20062 F20101222_AABJOD wu_z_Page_073.jpg
f07904e2a76d1e2a683e49a241867b0d
ef0d3b6caeb8a3bc6af7e329e93e255e83e3efe9
116155 F20101222_AABJNO wu_z_Page_052.jpg
f9596ffd1583759bca68db3bb98271ac
23b42caba0a3b0409455e3883400a5f9ffd58b05
117850 F20101222_AABJOE wu_z_Page_075.jpg
af6304cb17786c09a52f04f338faa72d
a419356758aa4220df9f2f51e81114c1269e079f
107832 F20101222_AABJNP wu_z_Page_054.jpg
607740dae871e60cbb30051bfdf5e1d5
df2823c55a7bb65fbd76a1a51068bacfb6e215d0
106837 F20101222_AABJOF wu_z_Page_076.jpg
c4d68310d26018cb426f10d261e7de6f
66795eff2ad0200f1530ad58c77d420b5aa82843
99121 F20101222_AABJNQ wu_z_Page_056.jpg
f007da586029917f550846a68b4edf5f
4954ced736655a0c998eecc312bff714ffc6d850
106501 F20101222_AABJOG wu_z_Page_077.jpg
6238a64bd98b4574bfddf00150aafe35
3ec2a6563c30c585a19b33f4dab80f63ce729ad5
105522 F20101222_AABJNR wu_z_Page_057.jpg
3ec93935ed9b7f825618f2c816c1258e
e739f37a42ef0397453aef95577c6018f6188f4a
112153 F20101222_AABJOH wu_z_Page_078.jpg
1368c0f1488abf5e136d3d6b32d8d1d1
af32ba8a0e93deff383e42f7169eab434a4bdc84
34090 F20101222_AABJNS wu_z_Page_059.jpg
841f4bdc6da12b2e0d837d77e6617fff
86f28ed8639c5676fb166983f30ca0f9d90fe26c
102858 F20101222_AABJOI wu_z_Page_079.jpg
db338451fe8fa0c6ca5a8e859d4689bd
ebb7c94e5c50b1d0de5f6372f538f1f38876e80d
35726 F20101222_AABJNT wu_z_Page_060.jpg
959dc3422efe3d5728842a11b982be37
9fb41d9f1904983a4fc934a1d416b77c4fedc3a3
104041 F20101222_AABJOJ wu_z_Page_080.jpg
fd00cec2b7a138be88e44cc50207ead1
374a461ac22c9d0752876b85fa7dec8ca3c89cbe
30675 F20101222_AABJNU wu_z_Page_063.jpg
5141c1e975beebfaceb3950b3f2d609b
25ecfcc2ff16423004b8c3a0f4c04597eb9bbe30
104436 F20101222_AABJOK wu_z_Page_081.jpg
f1254258932d8dc2e874940bcf8c806e
577c904af4e6df908fd307876cecf6846fb1821e
99524 F20101222_AABJNV wu_z_Page_064.jpg
feb2aa1e2085f6862ff34cbea65806bb
46bd68056b8af88567e858e3dd79968ab8fdcbd9
116459 F20101222_AABJNW wu_z_Page_065.jpg
b70394f0174a24ccec4b45e25e1e8717
78b806d8b56b7ca644dc96b98c4b256ecd0aa7be
104769 F20101222_AABJOL wu_z_Page_082.jpg
69f5a4266200b20766d6e7213612cebe
cc58d5a0a0f2842e1962682787d421b5c38fec03
105585 F20101222_AABJNX wu_z_Page_066.jpg
4be4467f49925e37e64f6129c78d590b
ded82ad63cf02c9fca6bbb534150c6fd9421efff
44056 F20101222_AABJPA wu_z_Page_101.jpg
e6828a07ebcade6da62b51fe7f733a24
a5f0191161ea76e14a295f335d5ff87861fc1445
111730 F20101222_AABJNY wu_z_Page_067.jpg
f21ef3dfaea952efdc039a9f76732fc4
645195b379b5bd95af57874477644f2489fc3c00
107077 F20101222_AABJPB wu_z_Page_102.jpg
b4a101c3a392d13fc23c2fde716398f4
5c74e5c133919dd4cfd2fb758af719f204a732ed
82703 F20101222_AABJOM wu_z_Page_083.jpg
f7a671fc5ff162dbd49ae89027b970d9
9c00088da9f8e0da6bc205c95a353b36b9eaf982
111214 F20101222_AABJNZ wu_z_Page_068.jpg
366f3fb1ed0dddb4f8aa1a74cb015ae8
f2c33b8e90a2e52e42504bd8682744c4f7e3c9cd
101930 F20101222_AABJPC wu_z_Page_104.jpg
df6917a07c9d792fafd3a0f2807aa95c
aa5d08295fe4914036477f4ca7ed1f9aef707dc8
26985 F20101222_AABJON wu_z_Page_084.jpg
8204c277fdbce51dadef05b0a00a8a26
a5a363eacd0822553f268223f71e38067f218fc9
26398 F20101222_AABJPD wu_z_Page_105.jpg
b4d36f7b992105fd9e5cc966cf46a045
9bc67426ac08decc3b9796c387e29821891f0d8e
30696 F20101222_AABJOO wu_z_Page_085.jpg
227a5a35cc3488bcf894725ba9264295
0024deb6c6a95ef35b70d488aec033deda2853cd
93788 F20101222_AABJPE wu_z_Page_106.jpg
bf42da41a0d2f6584cdb3f29589e1e80
7adfa4fcc9091c5a020e13aac1835006e748ffa3
43491 F20101222_AABJOP wu_z_Page_086.jpg
3a78651ca60aa432fd9a0135c7908dff
25ac47586eae68a5cd5a07ffaeb1d58326bcd3b3
119387 F20101222_AABJPF wu_z_Page_107.jpg
5ea96a3b20473b1b43153c2b8ac01c96
34d6f769d63886e8467dda23884d9d4c5fb2518a
35233 F20101222_AABJOQ wu_z_Page_088.jpg
386b7e0966fdf94b4220fb0bdefdc8c5
638f33cf64ed3a6328da423bdee4d3f0096604a7
128311 F20101222_AABJPG wu_z_Page_108.jpg
e5c21f1cd8cefca87810195185a1eecc
c1ecfcec3cac949f9c2889c7c7c7e9faa87d826e
49594 F20101222_AABJOR wu_z_Page_089.jpg
f711c15e68b29b4345e9dca005d97b46
cee3c766ac66819fe951f507e0b7502fd954490c
118763 F20101222_AABJPH wu_z_Page_109.jpg
6b8d037c6283f35bbcdb3a9ba43fab78
b5ca486ffd928db01b0d7175f1eb25b05ba0edad
20903 F20101222_AABJOS wu_z_Page_090.jpg
7b3b36834ea234756d123de46f7bae08
c442bdd44ba8d321e4fe5274c7bb20b8f1f6505b
77051 F20101222_AABJPI wu_z_Page_111.jpg
56e3f16765105a603be131c699267e0c
ba0e239f347c48e39da30e6a445974948058642c
31229 F20101222_AABJOT wu_z_Page_092.jpg
2c41c19cf2fb540b891d51451419a5c6
1d527d533d0b41d012f1fba794c5112f87c52bbf
F20101222_AABJPJ wu_z_Page_005.jp2
2142f0897baa191323081988b991f753
13b4a3ba2e6d5d3d907d33018ba111b39eefda40
40562 F20101222_AABJOU wu_z_Page_095.jpg
e133255a4218a74163b2dd918007352b
e16ced05579acaa78bae149d2d47188367de13ac
1051964 F20101222_AABJPK wu_z_Page_006.jp2
b9bfdf4dea19ae00d9a38ce8e8938bea
5a29c761f46c605588e30b2ca5fe5c658c8a70c1
29302 F20101222_AABJOV wu_z_Page_096.jpg
3300261d1d844970378fd894780d08cd
cee6bd5d5b13c8ee026648084d1dfb3743f6e35d
223187 F20101222_AABJPL wu_z_Page_007.jp2
feffb52bdc5b0fcffd54b6f34d0b2068
f3a4c1bfd37a0bdd1d46025186508a1e247b45e3
28092 F20101222_AABJOW wu_z_Page_097.jpg
9e6d72811a2137071077abd97f48f09a
e7efcc63c244f9d689c2167f4fbb332fb487b1c6
1051980 F20101222_AABJQA wu_z_Page_023.jp2
b3b0dd3f9e5bfc686862083d4c2b83e2
a8caf4b68faf05bb307f975b4ef15c40ec956ab2
39976 F20101222_AABJOX wu_z_Page_098.jpg
99a41de2e77d800398578bffb6910a9e
86f1b318dee0322e45e07157572311a7273eb2c6
1051927 F20101222_AABJQB wu_z_Page_024.jp2
0dc5c176a5c3cbb9e362a37d7c6293db
1ed05157dc8a9e998812f13b7e444e94e81ead10
1051981 F20101222_AABJPM wu_z_Page_008.jp2
1ee53b36898a065b0eabc22b3fd1e8b9
a904dd2b02297ee68bd308ab325873c85fe88cfe
37158 F20101222_AABJOY wu_z_Page_099.jpg
75c8e832620a6c73ed7ab58389aee93f
ab5b9ed1956a8afc99da877643cfc50933eb47d4
1051925 F20101222_AABJQC wu_z_Page_025.jp2
327bc9d3747d3c2f9207de6324007145
f9aab514b39fd049853490955c5dd0f070c1df21
1051958 F20101222_AABJPN wu_z_Page_009.jp2
ea99b26a31a2adba9d7bc0a7f09bd012
0cc51b2abe2dafc840030328010f6ef1d9759d8f
45750 F20101222_AABJOZ wu_z_Page_100.jpg
84f23476ac0773bf00c87a0bcb14037e
a6cef0bd5eee158a9df9ae75406518783d5f4160
1051954 F20101222_AABJQD wu_z_Page_026.jp2
51f73921207adc741b837e8433810f8e
02143fc6896a51008684e5ce2a8dc860339dcc9b
1051985 F20101222_AABJPO wu_z_Page_010.jp2
b772529c0d3b7b3c0e1c7df3b1809154
14e39fd7bcd089b0e00ac5df26dbb3d885e1dfce
F20101222_AABJQE wu_z_Page_027.jp2
6f308df2f0a73045cac551f225e3dc33
256cd0aa199d21260d2bbae48562dc31e220b184
212926 F20101222_AABJPP wu_z_Page_011.jp2
e7bf5ff64e59da38bb611a57c71f2e63
576e6e6b9ab1804ae47ba93f5eaf9c98301078e7
F20101222_AABJQF wu_z_Page_028.jp2
3fe5f843f7bae146b8b58d51cd9c92d1
c9fe31c256a1ac7c4beed6341981834aa7e9b2c2
929970 F20101222_AABJPQ wu_z_Page_012.jp2
cbd033e8332f42ddff8864c374459a8a
c1296fee5d8879181ba9d72fa4d4213cda0869aa
F20101222_AABJQG wu_z_Page_029.jp2
957734811b4a33a0f39695b30888696b
39d7a24b829f811b45178dc53c3b52a687899bea
1051983 F20101222_AABJPR wu_z_Page_013.jp2
aec4965a6eaf8388a4cd7e40002d5d11
19ee2159fa76d21389c72cdc6b7347ba98846380
928514 F20101222_AABJQH wu_z_Page_030.jp2
fb06bf95302bfe85a5275e4a328f6360
7ced4c33d64edcc72973aed2649a9488659d3356
F20101222_AABJPS wu_z_Page_015.jp2
6d54ba5cf8833da8fdeafc6924c077af
1c122d46a6e8b07062f49f1ba95993e0bc6aa649
632886 F20101222_AABJQI wu_z_Page_031.jp2
89aa6b448da401e399f247e0480a9a68
440d19bbf57b948ff95ab149cafa0151ac6de2a9
820994 F20101222_AABJPT wu_z_Page_016.jp2
7d55f40fa1e7fe7c7c397d08ba6b6625
dca161e89399f5346720204713623fe798342e00
392516 F20101222_AABJQJ wu_z_Page_032.jp2
3d0a9a140c8179628fa846b599ae5341
370707bf2b7bb84e6b89e0aea613c21a3155db43
F20101222_AABJPU wu_z_Page_017.jp2
6acf1212a6a450efd7418dc3eddcfeee
003665574c6d48a02735ba4927e0f9ad188d7f6f
718601 F20101222_AABJQK wu_z_Page_033.jp2
47ccaa7f416fd367e109940bed578508
dea10b367ac099060dc65b6cbc7dc37d184d5b84
283928 F20101222_AABJPV wu_z_Page_018.jp2
55a68e0517882b2fffd626da6f043be6
d68a6da7fec96b39d425f628babf981dfba77707
1051941 F20101222_AABJQL wu_z_Page_034.jp2
25c80146e625b44efc137e891812960b
9b6318afdd66793a9c43e5b3c8cfde21e9cf295f
632349 F20101222_AABJPW wu_z_Page_019.jp2
ece0492e896a7285dddfa5b0c1bdd056
ed54f678d71cc1c1fd20e0545d58911ab9a71057
F20101222_AABJQM wu_z_Page_035.jp2
2bc11d81c95ffce6dc3ab9b09624703d
09f7309f3d2f7fa92cdb9d8a9f00bc094d0b8a2d
630872 F20101222_AABJPX wu_z_Page_020.jp2
99498f2b65fb23389de25ad448d1c597
e57e37b0c56d2c87552d5addedda9661d5314bf3
F20101222_AABJRA wu_z_Page_052.jp2
727289b9e643a0191beed1254fec1351
bc27707806a8f12e7271919783dc7d60af292c5d
786042 F20101222_AABJPY wu_z_Page_021.jp2
25d1dbeed3c5a0a6124f4ee88868691c
07073e1c06222ccd63db0baad862410586d72ecc
1051974 F20101222_AABJRB wu_z_Page_054.jp2
b29bdfec7d5e10e65a78fa30b7e78384
5eb17a885cf6ede3a861ac7c2e99ceb1f61c5843
F20101222_AABJQN wu_z_Page_036.jp2
6768f04ea2fb15143e0a100d39ca43e1
dcee7fe111afea66d83c26c33e4a9c05b57bbad2
367982 F20101222_AABJPZ wu_z_Page_022.jp2
b578c5bb94b4e7c9f2a2ab9d7c6eb1ac
c52c67d610fc63f604a4668d669d116d540d6550
1051973 F20101222_AABJRC wu_z_Page_055.jp2
4dbd68d9f07a60231542a856cfb506cb
7b991f822cebc00b0efdce982692ae1be8cb7001
F20101222_AABJQO wu_z_Page_038.jp2
8713ac5c7c11f883a30e78bb3441e540
13b0aa35cd34f98bb27fcc726e6458ae2cce7fa8
1051984 F20101222_AABJRD wu_z_Page_056.jp2
8717ff3578a6f3e79690360aec4548d0
cb4e17e2ac130bbdbb2d280c8ff1d66e0289cc01
F20101222_AABJQP wu_z_Page_039.jp2
e8890da903303c50c455f43ecad69399
5a6bdfed432da6942140c810f48d1a80853cc4ff
F20101222_AABJRE wu_z_Page_057.jp2
3b6538840eafff73233e9dee4ddeaa00
c8075e112e341828c3629895cc04fbeefdfcb553
1051978 F20101222_AABJQQ wu_z_Page_040.jp2
9259778dda6e4646610121a28ca262bd
27094c95fba1ea466de51a1db1bcfd0b510b9cb8
330073 F20101222_AABJRF wu_z_Page_059.jp2
e109ee4e754695a9cceece162451bc7f
fa43999fd91de6c8ace099df0162b3cdfe1f8627
1051982 F20101222_AABJQR wu_z_Page_041.jp2
eb30e07349ff1ad37c9bf3e3c3524f89
b7f574197f7366a5d3aa2274a3484183c0acbcb9
372131 F20101222_AABJRG wu_z_Page_060.jp2
343da9a04d7a5100d8c1b578deb5db89
18f35227872b6b08e7d7d53d1b883eab36f1cb3d
1051955 F20101222_AABJQS wu_z_Page_042.jp2
881a677f681f87a3683b4cae7abae75b
49375aeda41f6cd7bca939ff100c43418628953a
267740 F20101222_AABJRH wu_z_Page_061.jp2
1b44cb188e6f8a2d29fd4761b4cb3f73
81b6535abbbc69682ca018ccdd631308db136a73
484427 F20101222_AABJQT wu_z_Page_043.jp2
18f49c6eefceab4fe104cdeabbb92b74
1584a26b47fb84626fb1f3f11078f6cd4336d9e8
178442 F20101222_AABJRI wu_z_Page_062.jp2
97c03d1c2e1edb0f685d538c5a978daf
d3f986dae0ce3bd5e3ad7e046fd8610dee98321a
287100 F20101222_AABJQU wu_z_Page_044.jp2
fce62d237781c0e32a883f3cbfa69a63
ef918c433cf5d44c0040edd76157d2f08b2ca3d4
333262 F20101222_AABJRJ wu_z_Page_063.jp2
51526e58e28e38f1b0d2ffacbb6d5ea1
51907a94cbc883d17362d33e02087330afd89c23
372489 F20101222_AABJQV wu_z_Page_045.jp2
56805977f0d9a377ad7077cbc0869d86
904ff6149fd68903f9791bbf54dc39cd3317d803
1051948 F20101222_AABJRK wu_z_Page_064.jp2
3bb01344016a62b74acc527e36fdee82
7b0c9f926bd8ef13e3eab16317be0a244e5e2c47
403611 F20101222_AABJQW wu_z_Page_047.jp2
0e826b3c0dd69154fea12ba5400effbb
df03f5df117d21611a4f779c8f5bf2c77cb8949d
F20101222_AABJRL wu_z_Page_065.jp2
6e06e4f1d03b0d3a04149bbad249f34a
bc7fe3f4ba472ed1f502c6dfa3b1e4d8bb7f1d3e
741904 F20101222_AABJQX wu_z_Page_048.jp2
503fadb90f7e7e0973f3c005319c7dc4
9f9c9d0484479c260bacd3f103216865c5b48800
293771 F20101222_AABJSA wu_z_Page_085.jp2
662ec35ee500be6e63975e3625015786
7fe5ef29ea3cfda5220ccb7d640a78244c8fac73
1051971 F20101222_AABJRM wu_z_Page_067.jp2
1dc902bb5a6ca27b0fd019ae5201c6fb
498b6199f217b23eca59c33c0e1a6039d24c4ed8
275301 F20101222_AABJQY wu_z_Page_049.jp2
919d5ee7161b365f59c140561c18c01f
8f2d3cb89cadf95fc49ceefc8d1819165904932c
364571 F20101222_AABJSB wu_z_Page_088.jp2
8e40def8c16cc7680f2c0e6b69af8605
bef7cc6899eb7cc187f25f3789984d13af39fb2c
1051965 F20101222_AABJRN wu_z_Page_068.jp2
286ffb3950d61d69bb82d9e2e5a5385b
fa3fbea335fdc6b3f8a81a95b2b657e3800acb29
251936 F20101222_AABJQZ wu_z_Page_050.jp2
22481235bdbb874e8dc324953aca55df
b196ac43c0129040c9bd74f428691ffd3012196d
159933 F20101222_AABJSC wu_z_Page_090.jp2
092dfd339eee433109b6200428f95384
bbb5cca2187e503947812163abb38a4ecaf497a3
247099 F20101222_AABJSD wu_z_Page_091.jp2
8eb02a29412754a7604ff0a9ba883b46
3dbed966a36dd24160b697ce315da724016a97c2
271276 F20101222_AABJRO wu_z_Page_072.jp2
59420e0f9ebe9819a52403be7af76aeb
494d7f3208024a38c330af53d57aa3460a5a9adf
328469 F20101222_AABJSE wu_z_Page_092.jp2
e20c4eb83dee91d57db830ab2ed1c157
f3fed0ed253e7ca69227794cae11a594fccc9616
174242 F20101222_AABJRP wu_z_Page_073.jp2
139acce6071022f2d4e002cb8670d16f
ea7c4fbc50f993cafaba9bc578f7bf7dc795f0c7
238052 F20101222_AABJSF wu_z_Page_093.jp2
343a43432b0f89a380f248802cd27104
11f87adae2a3e1d7782d4bfb75c4d5785801bbef
F20101222_AABJRQ wu_z_Page_074.jp2
021f8e78e942dfbd2dee60297d117ee7
6726a7cf4dbf35df30cc292f2f00bd3a46199d36
389539 F20101222_AABJSG wu_z_Page_095.jp2
9d2806c03d79be1f35b869cd297598ba
3fb32f1e896c23b3b6c9f22dea96011da86c09cd
F20101222_AABJRR wu_z_Page_075.jp2
dd5d0799eed1eb3d3464b4718f78e09a
85a3387fbf9c27a70714fb438afcc8622550019d
297315 F20101222_AABJSH wu_z_Page_097.jp2
20f8b0596a1016af68eea2f3926c1083
5e82448f14996b223cfc60561c4c54f98ab3b3a1
F20101222_AABJRS wu_z_Page_076.jp2
e1ac66309e5d7440c2c261d1f1f9a6d9
2dbe3b533af2ad9a2eb4b4fd465cf4528477f2bd
514568 F20101222_AABJSI wu_z_Page_098.jp2
9702dbb5f54b8e12241b4ac48649d2a0
712056bab244a9a573c35be32805756b0d67447f
F20101222_AABJRT wu_z_Page_077.jp2
b12bae69c52ffa06ab8acf924ad46b11
b050aa094365a95bf13c258dd2583d0f1c9d9c9e
446623 F20101222_AABJSJ wu_z_Page_100.jp2
8ddae5fe54988fad4ad1c4d8799c2b17
827603f003389395bee53f9dc2e866258bf5ed02
1051966 F20101222_AABJRU wu_z_Page_078.jp2
65d43d936746c4543faa2cb77d42bdf9
10e95de1b8fb2956ebe3bd83c2b6e7e99dc60746
1051947 F20101222_AABJSK wu_z_Page_102.jp2
d11bbcf735962f3ec94a015a1fb77008
25b075f4813f5c4f35601ecf855d23b1b8574407
1051967 F20101222_AABJRV wu_z_Page_079.jp2
88fece4fe1e8cf71263c9d336606e4b8
31ff438395ee22d8340304c4ed44007b98622154
1051951 F20101222_AABJSL wu_z_Page_103.jp2
55e06be8a5bdc76a7455f8fce2673c17
7b70df192d1bfcc14466e552a2bafb5778728ad0
F20101222_AABJRW wu_z_Page_080.jp2
8af7dd761283ba24daa69487c08325d2
c5dee3fc0db6c842b6916edeb0fb787b777b9eb5
F20101222_AABJTA wu_z_Page_008.tif
d0bfde1609f716f46f615b0ea7fe6ebe
aec9c8d16f2a86e3d252ac677e6fb8447c93aae9
1051986 F20101222_AABJSM wu_z_Page_104.jp2
9f109d61a75ff94fa847d7652cf77eb7
9eb0dc24e3705d3200d72df6b446b5088369e143
F20101222_AABJRX wu_z_Page_081.jp2
5003196fbb60323304ac776ac14952cf
ba1ee436dd867273a1985ece180b3a0605a1c471
F20101222_AABJTB wu_z_Page_009.tif
a1b279a674fec1e700bff4f1744f0396
70009469f02a971a4d13160a73985cbf2e68e3e7
202669 F20101222_AABJSN wu_z_Page_105.jp2
f733b1650172dadb2a8e5f2d7282f762
94a7934ad81a452ea996499cd206a0d2427d7339
F20101222_AABJRY wu_z_Page_082.jp2
5bc096a5b2da953bd758352a9e2d58e5
4e929df4df27767da34ad5081406ded9c102007f
F20101222_AABJTC wu_z_Page_010.tif
6068b66b2e5f8c3c08c1e91196626afa
36f3abf048f14d5fbd7d5862d036ddc3c0d93000
1011984 F20101222_AABJSO wu_z_Page_106.jp2
5cb1e81a7e2638e52262a15534965a8d
4045864f04e629ae0ca5fa577984b2256fec1081
907025 F20101222_AABJRZ wu_z_Page_083.jp2
16cad7e96558fdf5e6263f6ebd3ea0fa
33cf96ece9f9098ec5144dbaac7e1fe45b5c5a32
F20101222_AABJTD wu_z_Page_011.tif
8eab1034235d861e921d689ee3784795
93d45761fb6bd3573a21cd7f9eda862348bd2fe4
F20101222_AABJTE wu_z_Page_012.tif
bfb9d365b2c3b9bac34d4ef4e644eacf
3c44e478c2ede507334b881d4672810c5e670fcd
1051979 F20101222_AABJSP wu_z_Page_107.jp2
0970cf6b1c1a73c337ef82a8a9fb0e72
b97823ada6b9b3733d731637356beb6c554dbd22
F20101222_AABJTF wu_z_Page_013.tif
6be6d0bbbf2c9dd617ff6cbf449ff34b
28b9bc86cec8e3d3515ff2e198eb9d79bd8f3781
F20101222_AABJSQ wu_z_Page_108.jp2
b917c46143099ff552f76c19f62b8814
3bf298cbfba07992358830a855995bd31f9ac5b0
F20101222_AABJTG wu_z_Page_014.tif
235c84d979944bbb467a3f4dc6abca2a
306daa49fb17afaa79de0ba9e19c61d6d57d0507
F20101222_AABJSR wu_z_Page_109.jp2
cb0b256e24c182de8f9c0c7da4496729
dd9139a2eac184ee13fa1c900b3903500999c1df
F20101222_AABJTH wu_z_Page_015.tif
69641bab37b82022fcdc7c1b925fc2fb
f6535124184b6c1c4ce4b14ee16e94672db191d3
229474 F20101222_AABJSS wu_z_Page_110.jp2
d0e0b7468ee8de249b11ec7edab2938f
8e66492442fee2196657c4fc369c48c05953ca3e
F20101222_AABJTI wu_z_Page_017.tif
b2f85f90fbc5ec6944881b4482d994d4
16186f7fe4d3224b8ba15ab76364aa03357409e1
818189 F20101222_AABJST wu_z_Page_111.jp2
37dc72258f4840f25f9c310bec618738
224438c347285a848f41c28b146d08a2a3f9da7c
F20101222_AABJTJ wu_z_Page_018.tif
b4da74144527abeb986bddb8dbeeaaef
ba255f0b633a405d6525e8995203ff707d907571
F20101222_AABJSU wu_z_Page_001.tif
ca20410d7d5a8725f7a06f9b6e19310a
7d162f2690da853395c1c9b996eaaf71c88ba8d0
F20101222_AABJSV wu_z_Page_003.tif
8049d3ee5f60ae8535846bbbb646486a
91e6a7c92c9d5c3b7d9c28fc18f5776b404f9398
F20101222_AABJTK wu_z_Page_019.tif
7858bf19c5d9b18fb10f8ca582fb0eeb
2c270485d8cd240bb083105581f25bc4df3e8361
F20101222_AABJSW wu_z_Page_004.tif
7063b14c6e23966760d43b054e1f2784
ca4f2bf03a0b172e45c481dec5807dfa2985851f
F20101222_AABJTL wu_z_Page_020.tif
b252407dad269aa3dff93b10592d8d3f
e9a297fc317340c1e7337e2a12a89c3803a860f8
F20101222_AABJSX wu_z_Page_005.tif
b0432e3148fcc07f3984bafeff34500f
121c6fe90da9a65f3582dd5381adf82caf0de3fd
F20101222_AABJUA wu_z_Page_037.tif
cc22df388b1c2cd36e80dcf47d30fda4
ddab395067493db3c5608800c268f985d903f78b
F20101222_AABJTM wu_z_Page_021.tif
825ed63a6ddc769286ffdf735f3e26e2
0461c6130af32ecfc9e4eaebc800b8d67fd8dac5
F20101222_AABJSY wu_z_Page_006.tif
ba8a71e0e26bfd3f689b874e0df449bb
06390ac915e58e8fed58af36787759528f3e7fdf
F20101222_AABJUB wu_z_Page_038.tif
3eaecf25645ce8d17b2c22ebc220be2e
52d57c7abfa0cc79d93f6072a38998644295349e
F20101222_AABJTN wu_z_Page_023.tif
47c82f8d87d7d6202373f110c049b4c7
460566fe0a2ff122afa6a49cbcac2793d864aafd
F20101222_AABJSZ wu_z_Page_007.tif
7d62e0a0b6171889855cee5035252fc4
26122a45e9f6b75cd236daa2477fbe50cf125c19
F20101222_AABJUC wu_z_Page_039.tif
ec379c19e3f354f209409c8b94353afa
2d615a847148ec659f14be5f322c254c59e7189a
F20101222_AABJTO wu_z_Page_024.tif
7c50fce265c57a2984c2c050e1968ee2
483f703fccaad2ec69e96f1cabe7cbdcc460fa94
F20101222_AABJUD wu_z_Page_040.tif
871d855c0daaa8e8f1ff8c853a43ee30
34c85f7a9e0e44d20646efb5f4a652734f97e588
F20101222_AABJTP wu_z_Page_025.tif
30c39c213cb072533fc1a454143eeca4
0b30c2139905b6deecf69f09e7a99b21aff3e21b
F20101222_AABJUE wu_z_Page_041.tif
26309dc63e239e51f24df8bd35961344
f5987f17bebe9c9ed80d0c27cac9f167235a9bb9
F20101222_AABJUF wu_z_Page_043.tif
d928401e547f918a58c4bd7588b3173e
c4df402d56d7969606ba628602bf38caf8feff20
F20101222_AABJTQ wu_z_Page_026.tif
955e4058d1c68b2c425bc5cda756a85e
d2982ffe42645a079c74c7e39abe8a7f40cb4a86
62112 F20101222_AABKAA wu_z_Page_108.pro
0bc1339f4f26cb0f4611b460ce4f997f
68e4ca6b78080bb72d8ae6ecb9e9790d48193362
F20101222_AABJUG wu_z_Page_044.tif
7175b270f217cb3ce25bf707276b26a2
0be44634bc9b89990f3406a855aae80fd384315d
F20101222_AABJTR wu_z_Page_028.tif
2067ee82559df754572920dcb310fe3a
b8768f3a3ae6d04e644a4f2fa2468321a53910e7
56470 F20101222_AABKAB wu_z_Page_109.pro
539ac4e3df1a749659b94658636b4fa1
db9447a5c1228309cff91bc212f7692441d1c18b
F20101222_AABJUH wu_z_Page_045.tif
6d363f5e2e328c24a403f4211d442d38
82ea673c882ce06ea3f917877641088339088e61
F20101222_AABJTS wu_z_Page_029.tif
c5026847cab412c5110bb29a9ba37aa9
710de4174c6540ee98a30c1edf00c6f949f1d1ab
9975 F20101222_AABKAC wu_z_Page_110.pro
e9f86b399cf83a0079090673d286b3e9
18d936edfdf3949184c5ea26442395e06daaf0ac
F20101222_AABJUI wu_z_Page_046.tif
ed8c9a3359481d127de3add35853e0f4
c43bd2ab51f37be09bc489a57ac2f8aaaee5225c
F20101222_AABJTT wu_z_Page_030.tif
fb7376edd7b984e8044f26ea02f2a936
b8bf882afad3d0a959ce78d913d1e120de76c235
35765 F20101222_AABKAD wu_z_Page_111.pro
3941fe34af56a29d7d974e257c9ae354
269b3b70d5c7f1f7d59a62e696dc4ea76e1dd951
F20101222_AABJUJ wu_z_Page_047.tif
c0628ef47456cd6d3da78cb5b916e5f1
3adf47afb2ec135832f668474b900a487073c15e
F20101222_AABJTU wu_z_Page_031.tif
2b5e60051fcfbe625c5ae0ee18b1edf8
c40e5878628521b0f3dae19b6a8e8824a6d9b38c
509 F20101222_AABKAE wu_z_Page_001.txt
0ac9715c500b8df4718cf0e3cef3541a
41eb8747c71881f5e907330afa33abc99eb6d260
F20101222_AABJUK wu_z_Page_048.tif
30249009fb051d6b4f45401f7058d2a1
3b65b6d5a9bdad705189ffc7b3a63046e1c87a8c
F20101222_AABJTV wu_z_Page_032.tif
56f5c87a5606d667ca366e9d675c7219
f9029ae86e2a1331f9df20cff190278d34d60732
85 F20101222_AABKAF wu_z_Page_002.txt
6c92e993b7cd5932dd7a02eabc4afc30
d6171152b97cd9e0aeb456d8aab391a0e64853ed
F20101222_AABJUL wu_z_Page_049.tif
23ba10a9ec286355f80e8961c5952c4d
81f4e45d71a997ea6c28c4fa09340bea2ce3e459
F20101222_AABJTW wu_z_Page_033.tif
4b476cbe9f41bab51a344f51ce796a6b
ec72152c554641db4b57c5236c5f9c90d8f1f639
1423 F20101222_AABKAG wu_z_Page_004.txt
e8cc5b82c6916068c213c855219dff1c
4830bd55d5370302a5536dbd193908923d5c7e01
F20101222_AABJVA wu_z_Page_072.tif
8dd7bc3ca7b5cc6960c175e8c5ea49a7
76f746b2cd4c3a7cfdf8e1d21fc600759828cbda
F20101222_AABJUM wu_z_Page_050.tif
7014571d2366f85053813e5a2bb7ff50
29db521ff24cf7bbc04b20c0af05e1975939a350
F20101222_AABJTX wu_z_Page_034.tif
5017e6345222493955c5433f8f494e48
9e609d6ca698ddfd1aee03ce3dc45e20dddadf9d
F20101222_AABJVB wu_z_Page_073.tif
32ece19265c18e097ee65a8303a661ea
623857e48cc561aa1d5d5ff4a1a2f99a25a0628b
F20101222_AABJUN wu_z_Page_051.tif
d5f983c020dd477f96d297bdbee2f1db
b97f92b73a5adcecfcdd1256992db202a95c705e
F20101222_AABJTY wu_z_Page_035.tif
cb8aefc731199f89a761f55358d6224c
15125a6979cc1e903ccfa8661314f63927b03f29
3353 F20101222_AABKAH wu_z_Page_005.txt
c9a4afd674cfe7e8cf41f2c15e977248
90732a5bf18b6359932937ef4b312b9e11e36129
F20101222_AABJVC wu_z_Page_074.tif
9a56eb96bd411605eee051701a930e09
166a4f6b0125de8f45d28f4309ca03f4688296d4
F20101222_AABJUO wu_z_Page_054.tif
e0cd4384f0eae70c3ff7f54b1d3a6ffe
9139af2891a8f38e7730ec8e6b706e41d7f91887
F20101222_AABJTZ wu_z_Page_036.tif
d7abfcb1850114d84beb01fb6ebc515d
8fbcb0014346f0871e50d79537124ac922743142
3151 F20101222_AABKAI wu_z_Page_008.txt
1788c6c6c674e825a0763fe2899ae9e4
5369edfe88ea976a0905664d8702648526b368f9
F20101222_AABJVD wu_z_Page_075.tif
08c3a5e7cda798128c4202f5ae9e3185
fa94b302b32dccd9afa43f7c9700af53f409cc3a
F20101222_AABJUP wu_z_Page_056.tif
f5db7a884397780de71eda66f1097f7c
94e621fc51af114433a5f4b21a9e1b13130f8123
2914 F20101222_AABKAJ wu_z_Page_009.txt
c1aaf4220327d844b7dbe93cceaf2f58
6afdd538a1f0c4b4cc2d3442fdec5ac3d6c04f5e
F20101222_AABJVE wu_z_Page_076.tif
2a01ef0b1218aa5f7aacff6dc6ef8365
e1852103a03d39b14c4591d60818e66eb620ceba
F20101222_AABJUQ wu_z_Page_057.tif
d40c18aa1f780ffeab5af7176d3d5c2b
8de236c0eac3ca66df3c6d9cbac13b2df0f2ee08
2168 F20101222_AABKAK wu_z_Page_010.txt
b470c2c20c6110908303b2c3ba1ecf2b
a52129b4490421ea0b100220833a76de4cfbc47e
F20101222_AABJVF wu_z_Page_077.tif
223ae4b8f4512c5dd0b464b038f9b1ea
c747fb85d85ab46f417bf10bad88e8d98d549de1
381 F20101222_AABKAL wu_z_Page_011.txt
d9982ffda1a3aee58974a35ebf3281cc
0aaf4b0c696c8aa3038e05abd3dacaa55754868c
F20101222_AABJVG wu_z_Page_078.tif
11c6c1f7cf59abda387a7d7aca7673c0
49a21a31aa950a84da745c8f12e8700aedd5438a
F20101222_AABJUR wu_z_Page_059.tif
904f2ba1bd7e0f3e1df2abff3b7966b2
17312dd5f4f90fe1bf475c988a37bdd2db6d0adf
1668 F20101222_AABKBA wu_z_Page_030.txt
aca9b7ddefbb452a1e923766c2982dd5
3a99946fa9e6c47d871b276fa1f55687e8e92d94
1886 F20101222_AABKAM wu_z_Page_013.txt
6caf13b52b1b3f266c0d21327253c11d
87af843acc6a8a38bb15aec0950748c0ad3b5f75
F20101222_AABJVH wu_z_Page_080.tif
137993b856d0d584dc98a29580e281d6
f247dcb1375985b207b6e47e887da4ab0ee1fb16
F20101222_AABJUS wu_z_Page_060.tif
d8ca9632f3e830bff7caa415419d9d4a
2e060ff9a4f8c0165953a09cb34ba93975301b21
292 F20101222_AABKBB wu_z_Page_031.txt
f6ef75a83172a0e54fa977d0d2d84bf1
c3ef4c28699f23fd23df0f2a10233006460fa144
2195 F20101222_AABKAN wu_z_Page_014.txt
d9048777be66d190b0a8d5e945cc57e6
fba0b7fdbc3d209725dc43a183b20eea477e1cf2
F20101222_AABJVI wu_z_Page_081.tif
e29b1d84d3fb2d57da80419853c7330b
9c76a15bf52b89b7685758946c67e31e736e17ee
F20101222_AABJUT wu_z_Page_062.tif
f43a76756e9f46dbfcb4676947da2a9a
c8b2d846da5a442fc822c3a6dbb3045eb29f29f1
453 F20101222_AABKBC wu_z_Page_033.txt
46ab2a2ac249fb59cf3fa7595b15367b
6fb8bd42f79f33c69f91694471ca2c29f8399edd
1479 F20101222_AABKAO wu_z_Page_016.txt
c4c516c0a69c7a6db03c2b53b11d8fa9
40a825f6ac887003acb5b80baa2d0409834e15b4
F20101222_AABJVJ wu_z_Page_082.tif
9765b879583ba208dd211b0088a0d07b
715384667f7689ccefc0d42b990892ec5989bde7
F20101222_AABJUU wu_z_Page_063.tif
ae90529524083992aedb80dc803e1107
bb8b0b871c532f65ffd0920bf99f932558a6d986
2270 F20101222_AABKBD wu_z_Page_034.txt
d27d01f4f28077faa974eb1b43936c04
b073229fa728d07d7f3a509ca450182bffa9a5f3
2085 F20101222_AABKAP wu_z_Page_017.txt
b12c072b4cd4e0fef890584fb5622dae
7fa8ab50db8cc70976d4b33c844d661b6531d86b
F20101222_AABJVK wu_z_Page_083.tif
e453260d8c3a0b34b9af79ce0257f392
fa15ff5aafcdc56ec284e23236b6f9a2b9add9f7
F20101222_AABJUV wu_z_Page_065.tif
785647576be80f8079589389d62be3a2
c3dbe1bd6afbdaf32a5f757e05fc3fd5ad742ce1
2084 F20101222_AABKBE wu_z_Page_035.txt
790bbed67cd11f5c663952f72726796a
7680c4282d286302a8f556b781622271c2a63aba
501 F20101222_AABKAQ wu_z_Page_018.txt
44560a3eec726b1d58c407f5063a3720
f98b367c30818a3232e1180b2d7cb4fb8ed44522
F20101222_AABJVL wu_z_Page_085.tif
020a87ef1c320acdbcc3c36c78001e32
45b08891cc40b5c2930e1db62f7bd54eec8176e3
F20101222_AABJUW wu_z_Page_068.tif
8db130b104719e9abe80afa7464aa1dc
b84561ff6c9002e41866af8a978ec1b6ecfada33
F20101222_AABKBF wu_z_Page_037.txt
3c7d71fc25c5668acd2920dd63ba9c26
a4bd8dc80e919ad8424294495f4b76c1ef454535
408 F20101222_AABKAR wu_z_Page_019.txt
0d15f087ce5bcd487a96b8e9d27587ce
ae5de0acb367cea1b50a713a49447b7516988da1
F20101222_AABJVM wu_z_Page_086.tif
e10da0f079373bed6d402a8ab0c7cd52
9f2310b46ac9fa86bc2537b06d7189bfc4833d7e
F20101222_AABJUX wu_z_Page_069.tif
bb3ae43e28d1e2a9dd780992a6e26dbe
ec6a7957ddae1b48b9d915744cc38c1c1656c67d
2158 F20101222_AABKBG wu_z_Page_039.txt
39dcbcf6f53987ed5ac3da4454c76136
e517d9380ea9ebd04ea613d7866da3803f64b398
F20101222_AABJWA wu_z_Page_101.tif
53a00bd3efd973cd9b8677fe7f591d04
de80625176591e09ee9383b8764eade0aa20ad0b
579 F20101222_AABKAS wu_z_Page_020.txt
647afa287e6c15488dc79558143477b3
e11480778ad76b463a7b5f1bd9ce518a489b0dc2
F20101222_AABJVN wu_z_Page_087.tif
c4ce63cc2cc67c1cb0803d9b05ede010
8262fc40f832435ceb7b7e974f1c9e89680f9371
F20101222_AABJUY wu_z_Page_070.tif
be9474923aee382b21b06db0dce8fd05
3a4b521afaa5e5b505e2754f56050a3eca26a80e
2189 F20101222_AABKBH wu_z_Page_040.txt
17bd5b89f1fe55445a7ee997ff26ffa4
eb4b8ca3f8333f2bdad8f26fbbb90d6dcace0695
F20101222_AABJWB wu_z_Page_103.tif
dfb3d4f35cff5355288d8674e23e83ad
12cea7b136d23684add8d7a9eb7b68b60b9c9f3e
F20101222_AABKAT wu_z_Page_021.txt
40c05ec3277ab9b8edb8c905ed572e15
f0203e4222cf2ff6e296881ab7b79559b548822c
F20101222_AABJVO wu_z_Page_088.tif
90b2510c51b10157ca411198797f1658
11ab82856f73fecb5dfda5dc9c652db4d47ba936
F20101222_AABJUZ wu_z_Page_071.tif
45c9bb04ada326478152a3cc98f2cdf1
1171f180632f1d960d55022c5cda3665ac5a2cc5
2177 F20101222_AABKBI wu_z_Page_041.txt
69e6464547622260eb03eba06643ad46
e93eba4809a441899ff498cc9506e506669d3a48
F20101222_AABJWC wu_z_Page_104.tif
ff47fc7196663b430400efd7bb1f4737
b73aededbc9f2bc690300c08e3eeef159d25d774
556 F20101222_AABKAU wu_z_Page_022.txt
99f59da1bffbdef9cba3e628192312b8
a1148b05431c1ad54076beb2b270497db85353c0
F20101222_AABJVP wu_z_Page_089.tif
4912bb6e7912d8bff39b6b4df843e269
de52b116eb4ecda68b79a6990fa65256b2dd3e88
2056 F20101222_AABKBJ wu_z_Page_042.txt
cbfe807440b6b4711e28b0e220725966
8dd5c282e6d0942268f4733dbf2662bff6ef0fd8
F20101222_AABJWD wu_z_Page_105.tif
7da3b29702a6fbed816272d22d65159b
625f4cefb5acd4a84a4f9902d8db700268fa93c9
2167 F20101222_AABKAV wu_z_Page_023.txt
9d1b5d520c19a7cbeb85685bb40d515d
d863d8e34f27493a633606552ac610b48f7b56b9
F20101222_AABJVQ wu_z_Page_090.tif
31cc8101f4287442fbc5a3b8ac7c69eb
93db65c50d78c50a92e29cf34dc76c7a30be6f68
279 F20101222_AABKBK wu_z_Page_044.txt
f5c4234c9481e77185a0f709d0996773
8768ec35cdae2651c4e560d76df74d5b4c86135e
F20101222_AABJWE wu_z_Page_107.tif
d035087224cac59133a1e4728c418a92
c3db77db8b2ff401107eddfce27e07ff71c5f609
F20101222_AABKAW wu_z_Page_024.txt
7076f67369d36ebb5db159e2526a1f81
3d685ec7e652a44e975676e2855e3b55b8a8a0e1
F20101222_AABJVR wu_z_Page_091.tif
9c78baa599addfe6dacdd5727ae4bd34
5adc52e8c9d31b52eb9bb3508b12488b9d182922
481 F20101222_AABKBL wu_z_Page_045.txt
bb441209015165924ea3161645273d1c
06c0ee2762414097adfa8b36771665045702290a
F20101222_AABJWF wu_z_Page_108.tif
911d58877b1daa73dcb4821121df9cb6
e458091eec55bb23e39e9c0a6e10a7ff09d7e8f9
2145 F20101222_AABKAX wu_z_Page_025.txt
58df78b0bda41e4115805b686e48e161
3f817fd4b4ba3c4394450e32a703dd0fc52cfa04
1981 F20101222_AABKCA wu_z_Page_064.txt
fe71332cc76a6dc92621f00a58afbefc
c292cb5bdca2c8be759ae8af9232fccbd98856f0
591 F20101222_AABKBM wu_z_Page_046.txt
f5f51f23df9fbd4a2ccb9950fa9250d5
41155e4baadcd7c488cb14d78cab4fd254bcb50c
F20101222_AABJWG wu_z_Page_109.tif
f1f19b97fed9031cfa9516cdd2991e2e
b714e6222445530352b138f909d545f75d9f5d6a
2218 F20101222_AABKAY wu_z_Page_027.txt
f3eae0c2ccbdf39522965491ea5f50bf
2af53460fb156fd1c583157aaea114391c498da4
F20101222_AABJVS wu_z_Page_092.tif
7880952350997879e8190b27dc479504
cd3ccc4143a4b4bba1fc59e80c21f7038f765950
2220 F20101222_AABKCB wu_z_Page_065.txt
4a7e094bc85e8dacb91771ea82e8bca8
8a603c51dbaa2723152919ed08cfe238c11b0dea
574 F20101222_AABKBN wu_z_Page_047.txt
29ab5535d13323a5d17bbb95e11705aa
525011646a804476e40de0f99cdbda62fbc52b8b
F20101222_AABJWH wu_z_Page_110.tif
e4aade32cca437744c0c713efcb8c81d
5793c1360e93df3d3949ec947f1f904ad0e39478
2143 F20101222_AABKAZ wu_z_Page_029.txt
22f398be24720203dfb8d2ceec117de7
e2c3e3c09f778bcb91e15e1a0baa69c03d607b34
F20101222_AABJVT wu_z_Page_093.tif
4dc3ccc0f44d8404acec6c2c0fb7d51a
a2edadde0cf524d1e4304bb9ab1f0a60b2416e59
2082 F20101222_AABKCC wu_z_Page_066.txt
e92c777ee905595f960ee4e4844722e7
21dc6e02481a3cb47191d48e8ef296e0b58c17b0
383 F20101222_AABKBO wu_z_Page_049.txt
77096531e6f12be62629cf8123ab3495
37b1e37b320cf50029db2412bb71c84bbaf19c7c
F20101222_AABJWI wu_z_Page_111.tif
df8c154eb2806173ab5f6f0156eab378
e4d56a43eced97b98096f842739914150ddd52f1
F20101222_AABJVU wu_z_Page_094.tif
a9d8313881ce5d07ee274c81ad5181c0
fb3bd8bc961abe802349cd827df4673b844961e0
2122 F20101222_AABKCD wu_z_Page_067.txt
b760b1f202aa4447e5471315a3cb9882
182332e3d47170f583464b40c6b59fcbe786c80c
F20101222_AABKBP wu_z_Page_050.txt
5da36c0c4fa6d04913e317fa25c0da49
9c47aa45e9fadad527f3e802406c27dd1e4192be
8970 F20101222_AABJWJ wu_z_Page_001.pro
ce8218a4e2c04f9a434984173fd716f8
a3fc33d8f087dc76820f8e59095eb514114ac083
F20101222_AABJVV wu_z_Page_095.tif
5fffab39ff0894c5a46b2557fdbaed9d
029d73d1720c5b7374dc473cb855c655c27295f2
2131 F20101222_AABKCE wu_z_Page_068.txt
b384ef2650416ec3ca3b93826e7781a4
39f751729fe8e95ddf4851403f3be7580e61321e
254 F20101222_AABKBQ wu_z_Page_051.txt
7168a4ebc82b36031853084fad109780
77f3711fa6ce842603e53ef2ed4f2af204fae8bb
800 F20101222_AABJWK wu_z_Page_002.pro
b2986837745cbb96dbe9494e641bd860
a52b6eb3c7213c91521b04cd954a5be8867bd7c4
F20101222_AABJVW wu_z_Page_097.tif
4eb2cee2c15b71b8aa6e3df09819a15d
888737c98fc6283398c42bcc625e256c06b0331e
F20101222_AABKCF wu_z_Page_069.txt
d9298fc9ab04033d04349f8cd05df291
ab924f39e349be17b5407881feadbf90452d183a
F20101222_AABKBR wu_z_Page_052.txt
8c6b3bcf82f02eec6da6d13dc595c7bf
a4e2b0e070bc4a8edb086e7246080a71dd8380e1
1268 F20101222_AABJWL wu_z_Page_003.pro
7ec5371447f7c2079c2ef33cc094f457
d7cf3818de9b8f1e565ea53c85156dac7bd2e55d
F20101222_AABJVX wu_z_Page_098.tif
7d159de519b8ce89ca44a6403bf79f3a
7218503216e20af83ff722ebbbed352d782e8c26
503 F20101222_AABKCG wu_z_Page_070.txt
cb17670fce49b959b8b7bf5144e19f82
043ac94727d0dc01fa66feb46a7466f3a33777c6
7830 F20101222_AABJXA wu_z_Page_019.pro
c99254af3293c1d6984e51aa368b093c
b84ccb4d416952775cbcf97c7b9b04ac739745b5
F20101222_AABKBS wu_z_Page_053.txt
74d1f1914f3abe22c2c66e1cf1682955
e4b9a495a9b55e32e48449255d9b1359736bb9ea
34695 F20101222_AABJWM wu_z_Page_004.pro
816d5d81ba10e46969820218607d0178
02b95c53d906e228a1b059c8c569be3f0f932b6a
F20101222_AABJVY wu_z_Page_099.tif
fb9c34b5456ed7b65dc848cb7b30920f
b0b57081ec5895c8abb306a704f7c9f9dd29e38c
378 F20101222_AABKCH wu_z_Page_071.txt
35d21f2ae4b02205201491e52c78b6d7
465710b2c7cdfd2884ded6188a441114ecd12b02
11636 F20101222_AABJXB wu_z_Page_020.pro
b816a3144a0076271a5522c08030e4ac
a0f40dfa772703701045de6f2b9195ff353882d8
2074 F20101222_AABKBT wu_z_Page_054.txt
c8d15186c334c89942ac9ac5eabd26e3
b45db2f072f52b0b91c2e7d7dbc88c15008e1406
78620 F20101222_AABJWN wu_z_Page_005.pro
fb5b9250db53fcfec60f83f703f6f1e9
fa1d74b316650a7dc422729c920d6def1a5998cc
F20101222_AABJVZ wu_z_Page_100.tif
0cd74fbf0bf61acfaf08aa024e9c9c27
630238274eb7905b434c3822e95cc9b053336bb6
339 F20101222_AABKCI wu_z_Page_072.txt
ea270e5b9ce74228f3fc645b265cbf55
7d6e68609e2e9273f89f06cac86f19b347886620
9795 F20101222_AABJXC wu_z_Page_021.pro
5184a8b21a8bc7c74b00c698f76a5571
3466fe371ec445ad5fd325a36eff00b1ba0e9753
2057 F20101222_AABKBU wu_z_Page_055.txt
d4a89a64091dcf3e653c2c2f444574b8
f8fb70b2b7544477b52348cf580e55a144dabcec
58671 F20101222_AABJWO wu_z_Page_006.pro
3b9cb51a9af4e4dee069c36de023b72c
f158ec1dc062cda5d07faee957bfb6237c273b17
317 F20101222_AABKCJ wu_z_Page_073.txt
e123123a80fd602419e85c0f052f888c
3379bbfdc7bd96f94b3a7dc749a5a5b708f49ec1
11158 F20101222_AABJXD wu_z_Page_022.pro
14e3a5d34e8836a51ee27c9e68ef2d7f
65fc2b8b4a4bfb4721c1d5a0ec6d4dccebe0f5d7
2060 F20101222_AABKBV wu_z_Page_057.txt
018e7145a4f98b93713fc9c3eb1a32c3
15bd757315f9b41d47dd2012b68c9efb343adb76
75153 F20101222_AABJWP wu_z_Page_008.pro
7b7e31f8e6e54f7ffdf73cb55732b97a
450be1ccb19471de50c2bd6bea32d08bb0af1f68
2210 F20101222_AABKCK wu_z_Page_074.txt
59ac0bfed297502cfee2b113ace7eec6
5c94303cfb1fb6f0415891e3a0176c46ff658579
52540 F20101222_AABJXE wu_z_Page_023.pro
6c8258bb1b27924413e382182e86621f
a8978c293fa1f12b840977e2cae009ec408ca4e4
439 F20101222_AABKBW wu_z_Page_060.txt
b97420fbe3374c704d4a41140a99ed9d
06d8c062b581cd51a2f029931d09fe46b524ae64
71215 F20101222_AABJWQ wu_z_Page_009.pro
2d2b243a0d99d8cb42bc347cea07e149
fe5825bb7b00589dba54fee5371b8df920073758
2287 F20101222_AABKCL wu_z_Page_075.txt
ad30b7db44c69ffcd6f5c59d9be4dd98
32785d9d4a1886b656d5e431ef96ac02df6dda5e
55848 F20101222_AABJXF wu_z_Page_024.pro
72f3d0b7be14f0c965699cfff94cc2d8
3657875b73782b93bd91c50c0c7eb07358f58e21
631 F20101222_AABKBX wu_z_Page_061.txt
c72d589ae937172e154b66e548b92ecc
3298e87d185266b6b2392fcbf14ef09fe58bd2d4
51652 F20101222_AABJWR wu_z_Page_010.pro
fa022f42b72bf0a8c06687bf7d310b18
adae8763592dfb168d9bda5d2efd2702349a3bef
2115 F20101222_AABKCM wu_z_Page_076.txt
eac33578734a03c670a241491789864b
8d1af301a854aba4c9e5963162d13794639f96b8
54632 F20101222_AABJXG wu_z_Page_025.pro
3ae2e1e201c33657af7e838bf0560c94
52c7901f051d24bc83efcbc6557128f75776148f
426 F20101222_AABKBY wu_z_Page_062.txt
4a031fd79a1fbb4dfe074d289b8cb855
76cb7e9f4fd2ce279d2ba65c819dd51af3ad52ab
8459 F20101222_AABJWS wu_z_Page_011.pro
ea3323f00a8f27e3985cd5c65f28f599
dc61abe46c82c018991ca84dfa09d7795af30333
312 F20101222_AABKDA wu_z_Page_091.txt
011f279f9b70b94c6feba3b2d75f3300
c808cc0df2e788689801578c36cf95c3da405359
2108 F20101222_AABKCN wu_z_Page_077.txt
603154210656841825f5d3c4f5e34ee8
c10834999bb87dbf0c1ea6852c4e5ab54a7cf9c1
57498 F20101222_AABJXH wu_z_Page_026.pro
9f92423ac1cd356bf91cd21a05d63824
b45b9cb2c34dd4c26ad6ce39634ead9205fc797b
F20101222_AABKBZ wu_z_Page_063.txt
5bc801011ecad21944fb71b1e0611746
a987b3182cd4d77bfb51794d5e181aacd1298058
411 F20101222_AABKDB wu_z_Page_092.txt
bb04e057c103ae80279e632d2d00fdb9
7664d6d5a65316dce2f6bf28fab83becabb5c443
2032 F20101222_AABKCO wu_z_Page_079.txt
6a57c2fa488fb7d1af10a334afce2a62
8f3ea0068210ceca73b7b82bd7a9cba2d9c68829
55388 F20101222_AABJXI wu_z_Page_027.pro
0e3590fbcb85608421097fcf6792f360
b1471b791faea3e668106bfc2833a9ab43d804d2
39684 F20101222_AABJWT wu_z_Page_012.pro
49e4fd873ce1689f03a6a041f2205e6e
f36604a2c29d312264039f2267eb6872ce04d760
437 F20101222_AABKDC wu_z_Page_093.txt
a84813e5551e683b18dc45de4741b84c
3fe6cc4c68060cc92dc2c2872f234366bf262aea
2073 F20101222_AABKCP wu_z_Page_080.txt
78c30a92b9f8674c8bcb29b36e955c41
cb45ea9b3a7538fdc393485267d7d0e2858e1887
52525 F20101222_AABJXJ wu_z_Page_028.pro
01876a5ae6ddd0087edcaa3a795901c9
e485573e79668e3db33c55523e659bfe1d12aa12
47243 F20101222_AABJWU wu_z_Page_013.pro
fed9ce2b1872a9b61982f5e3ef238148
3ab886469cee6802004fedbeee7830ef8e99a7f7
133 F20101222_AABKDD wu_z_Page_095.txt
f5418b6127a965f5c2b279ec9550591a
9117d1fd7b7c77de5ac283af95986b9d82d46136
2066 F20101222_AABKCQ wu_z_Page_081.txt
e9befc88a2aef88db87b0eb9720dea65
d56c8da053dc6f9d5259d8a97d1a750d84c11029
54566 F20101222_AABJXK wu_z_Page_029.pro
4aa94ea37790c1672169de1eb2c450c8
e57236ff624ac1fbee7666f2798fd797e88bdd7c
53313 F20101222_AABJWV wu_z_Page_014.pro
0d97405245c36e6da337776356d15c83
dff12380fd3fff4da5a6772a98087691dfd8bd3c
689 F20101222_AABKDE wu_z_Page_098.txt
9f1af8a5585805846dbcb717173c8945
9d6484769ddd82f33d657d77e5b6bb5d97791f73
F20101222_AABKCR wu_z_Page_082.txt
827b1484a34ce94bec475e3a50ef5bce
09b6ccaa12a1706dc91c567b0b69562c8aea3749
41663 F20101222_AABJXL wu_z_Page_030.pro
5095f5007a2d86ec117002ba14621ed3
dc43c5e845acf061dcec787340319a3e1c56dd88
48419 F20101222_AABJWW wu_z_Page_015.pro
5e39cdd2f05023530782d10f0a7e2a34
18ad26197181e449759b060431d3f272290e635b
370 F20101222_AABKDF wu_z_Page_099.txt
9871aa0aa01da179063a5d6b715b6331
521ebdd698ab94aac17cd8f7e4f5a583d023fa75
9997 F20101222_AABJYA wu_z_Page_047.pro
5df0a26e8888ca4bcf8018e436b8c84b
178c9543e1910dd267b399b95e18aaa771f50168
1608 F20101222_AABKCS wu_z_Page_083.txt
9d373ccb062fbcfff03ba146a7f24201
b90b13156393609f0dea7cea9c244187e9673567
6125 F20101222_AABJXM wu_z_Page_032.pro
15df0b45240169d8c864488b7b3454d6
d59d6cf39b870af8fff1b1ddd1b1f234e5df4e28
35328 F20101222_AABJWX wu_z_Page_016.pro
4555f30f48b92ac3c7e088c8b58bc7d2
f30c3400fb2a98b6c7e396d7ac7c8639a82b0026
F20101222_AABKDG wu_z_Page_100.txt
d6bae4fb960fba8bd57fb88465190c7a
8b11d0d866b30e88b9041466937fd722cf535f16
9682 F20101222_AABJYB wu_z_Page_048.pro
7e1878651e3c49a0a8e44c44e1bd17b8
b0bea9832941f245dcb0ee87d40ace98afd22008
226 F20101222_AABKCT wu_z_Page_084.txt
62ac8afa0c65b304e8e9be93811d0b6b
23849837281860c4162b7e2146587c41e898e7a2
9524 F20101222_AABJXN wu_z_Page_033.pro
0dfea9301424cd406b5e1b21a9122435
6751771d6409bc25ca1e19159ed7631a96bb03d7
51933 F20101222_AABJWY wu_z_Page_017.pro
bf04c6ffdc8e53b8894f465de6f7e042
38205fb32835a6e613e4758fb4f6297b6ed442e2
F20101222_AABKDH wu_z_Page_101.txt
9d6bde46f15bd2ac85b3227df60a47ec
aaefa452bcdf7d55b69dabd2cc7381821e200d4f
6590 F20101222_AABJYC wu_z_Page_050.pro
6af030e71bcfb8b250a9a5fd70e88922
c635248d86eb8071c99d57b62ad320bd3c9bae52
303 F20101222_AABKCU wu_z_Page_085.txt
3f5f78ce4c0201a29c32e56f5c92d8b0
38175d1e8232bf5acb38c15866547267b35a112b
55773 F20101222_AABJXO wu_z_Page_034.pro
8bc5529665e4f6c2bc9263b6ed4597fb
76e1cc83841b4780830a67c1fb31d4ece82b7ca2
12132 F20101222_AABJWZ wu_z_Page_018.pro
7a5cfa9c3b4d832307c23acce9f7e57d
56b125a8db9133167e64d470c2f32cd0110127ee
188 F20101222_AABKDI wu_z_Page_105.txt
1dd0003e1ceceb2c95bdeb9ee2cee86d
b8f60f6dcff30af6a2e4f01a879a140b77320be2
3909 F20101222_AABJYD wu_z_Page_051.pro
c0adf9c8d430396e5001f2726a187b0a
88d17cb1090b117cfbfef5dc9d0e28edb4f5902b
434 F20101222_AABKCV wu_z_Page_086.txt
e3614d034611f4011a86598a690224f0
b9f1415082287c8af4bc720e15d20d76eb4ee491
52561 F20101222_AABJXP wu_z_Page_035.pro
16bee7efee9b377258c3c7bf5ce7e2a7
8795b80a7e34973f9ab477a31be47cdda09336ff
1903 F20101222_AABKDJ wu_z_Page_106.txt
f15a578771f5b07b375c8860b55aec59
543e562724e25024fee225cab9c371e91c5adff6
56853 F20101222_AABJYE wu_z_Page_052.pro
4b7b1ea595b0181644482ae85021f6d5
1559a5a57d7556e80394ef8fef06e09f8fe42f0e
706 F20101222_AABKCW wu_z_Page_087.txt
16a20ae265f3836728f3f0b18544af17
515d0b9cc68eb9fa339723292abf467b540a248e
51605 F20101222_AABJXQ wu_z_Page_036.pro
a152321c9bd301b9f708f7ecf396b83d
a1c7b5464a64d17cb15b67340fe0eab085da1216
2534 F20101222_AABKDK wu_z_Page_107.txt
61ac764f96fec5f0ec5061db73ce6273
681f0931e1523365593a93aa85e2d07e950acc96
51113 F20101222_AABJYF wu_z_Page_055.pro
a9b13f57502bf8fa4011bfd527547364
81d59d4fa5ae5290afeb77c4ee3e7be31d4d43dc
729 F20101222_AABKCX wu_z_Page_088.txt
13322ad6c64ed6c7a68f90d056cc0cfb
3f8839c579616328cafc3e228272ad0343ef656f
54394 F20101222_AABJXR wu_z_Page_037.pro
b4343a0e277f5db173df4394fa38c261
f676785ce3524c08df7ffae57f1039544b0e3fc7
2283 F20101222_AABKDL wu_z_Page_109.txt
c306dafbf2cb7b74907a0edc30ee2754
40886db275ea00752575f5c227c48ba5093d636c
48903 F20101222_AABJYG wu_z_Page_056.pro
c1b2b637e1ecc49ff95b0b249b0db2b2
c5478a18a4d17c38581f48db2f795ff5622f27b1
418 F20101222_AABKCY wu_z_Page_089.txt
664b2c1ef7ced2bd00cb95c4c447944d
2044191630b2e1ccba9b312c20d9517a1bdf9970
55967 F20101222_AABJXS wu_z_Page_038.pro
e2bb012ebb158f6cbd543c4ee1c43d4d
4e476c0460f23cf42d150db6e1b22c71f8fe2d07
41596 F20101222_AABKEA wu_z_Page_009.QC.jpg
dd844fb07111466bd66dcbf7bb4147f2
262f05dd6c0a69c4c5700c630510ba4bc8bb2eb7
416 F20101222_AABKDM wu_z_Page_110.txt
41b7820e22e04316758f41739c3a0818
7fd7420fbae8fa6cce18b51818b9ace63b6ee290
236 F20101222_AABKCZ wu_z_Page_090.txt
789a42ead4bbc016a9a8a00487f4c015
6666547ebc8817012f837088bd412e4dad092357
53652 F20101222_AABJXT wu_z_Page_039.pro
1bf34dff9514c09a02771d799e6b1e6a
c16f705a12c18b7dba3233aebdb7ea660ecf35fd
9697 F20101222_AABKEB wu_z_Page_009thm.jpg
253aa5847be46c64d9e1be40d35e772c
96bc54b7e95fa5dbe1fbf9b93a0af1c5e3f5a241
2208 F20101222_AABKDN wu_z_Page_001thm.jpg
7bc48674b51b40a4439074bc38643133
d65eaa2f2e438ed38d21cf37f665a66c9e58602a
51259 F20101222_AABJYH wu_z_Page_057.pro
4cc626e78a002d72c0b7df9a8b02ad06
54b6123bdab4cf1a0653729b8e53d58b91225fca
7634 F20101222_AABKEC wu_z_Page_011.QC.jpg
eef7d1a91ae64f89328ed11e6f7362df
5ee0b500f4fadc78688b55ffa9de674d820b8385
2191040 F20101222_AABKDO wu_z.pdf
74607788633fdb6cc42f81d15b34ee58
a4f084acbe90f37c39df0d2670b541cb9dbe3cac
4351 F20101222_AABJYI wu_z_Page_058.pro
00ff628528faeb30993f4e41aa693c62
f6b8903a44a4da23e5ce961ebd2494f153a43c69
55541 F20101222_AABJXU wu_z_Page_040.pro
8844a3dd51b2f4454c56fa77a93aae29
a10e09076a7a2a61a109afda651bd2be82839146
2029 F20101222_AABKED wu_z_Page_011thm.jpg
7c0d1a7b459f508d39e0c32c9c886d09
6b350303d35688af4c3404261467153be8a0f3bf
126915 F20101222_AABKDP UFE0021726_00001.mets
64a5dfb2f919470bed2ed62ac6409bec
38b3fa1c72ddc143f0b577852d12bbc05321404e
13731 F20101222_AABJYJ wu_z_Page_059.pro
0d911d2013265a65eb3ecf8594587465
5eada21a9dd1938d358f6bab984cc80895ba48eb
55488 F20101222_AABJXV wu_z_Page_041.pro
a2a622dfa562637e6ab9fe355d9a2502
d3fda74400c052dfc8d20bb0201405d3f36cc6c7
27213 F20101222_AABKEE wu_z_Page_012.QC.jpg
db15dbf52c620448da8f4b55c85732c6
64b6c08e94a9739afe93dc07527c84921b77d4bd
529 F20101222_AABKDQ wu_z_Page_002thm.jpg
16cf6196a427f0c18660150a03541120
6a536d656c69044de91ad2fcd54f691901655e3e
9068 F20101222_AABJYK wu_z_Page_060.pro
cd0cd5307fda959ca3e15b48d087a087
f25f4780a664e5b6fd8b80725b0767df35308751
49909 F20101222_AABJXW wu_z_Page_042.pro
6ad4afeb00465ce785af36e0304abf17
4bc33e5b6233aa16d2c36b112d5381525a558303
6668 F20101222_AABKEF wu_z_Page_012thm.jpg
c01862f75b69db40950559702708fa5c
932dfeeb740c10e34f5a2b37ca0fec35698ec9f9
1480 F20101222_AABKDR wu_z_Page_003.QC.jpg
3bcfe5093bacae456cd7a27bbbbaf4a7
33c2ed0ebc2a96e6b616c15e1a9f8d05b8b28bc7
10942 F20101222_AABJYL wu_z_Page_061.pro
752f942bddae3d12198c5489ab0442ca
d2f5fb3a0e52cf32782cb0f0f116ccacc0b7dcce
6417 F20101222_AABJXX wu_z_Page_044.pro
84bb1761aba0ca9107f8f7f5ad5fd824
39dbfdd73abe59a4b0a9043482624c152923629f
8236 F20101222_AABKEG wu_z_Page_013thm.jpg
69308c22183cc256088ec4c1c4919094
86e72322379b63eb972ada93cbbe5b08effa868e
53514 F20101222_AABJZA wu_z_Page_077.pro
4f418e893f7f1cccc02458ae557722df
4967abcc15ecb7efabea645b6f622123e313c2fa
628 F20101222_AABKDS wu_z_Page_003thm.jpg
ff8cac7baa2a377fda732ef98458cafc
0c2c45d1375fa36473f5b8d0dfc325c7db566ac8
7736 F20101222_AABJYM wu_z_Page_062.pro
83d6fa06e9230622c9fdfc44c5b4d5af
1536943dbf125c91638eb120d9b8cf3484557516
8467 F20101222_AABJXY wu_z_Page_045.pro
fcd0c6b60c9f9bb8af1f166531d5666a
f65308e86e378551f81ea2267e3b96a2c7ad4004
36171 F20101222_AABKEH wu_z_Page_014.QC.jpg
e1e9ac655e472d1ee22a3e86889e9605
49f5edba21d4f99e9018306d80acef4396c6ff70
55682 F20101222_AABJZB wu_z_Page_078.pro
11090bb977e2cd5b067cf91dd8b22c2e
5a704c38c7cbced193ae29a41e16dfdfbba8568b
6023 F20101222_AABKDT wu_z_Page_004thm.jpg
28eccb60a33c522ec8843214a52858bc
cc28781d16b010450f223c54adabf2da7ac18c79
7049 F20101222_AABJYN wu_z_Page_063.pro
56e0b9fb9e54e21ae84ca6d2bd4e8a71
9df8f53cffefa66598b4052fa772a62cdd814285
6229 F20101222_AABJXZ wu_z_Page_046.pro
dbb76a1458fab9217f6bee8feb72e63d
ea2f2d1941a37b5bd48a9add5bd114688b891a99
8631 F20101222_AABKEI wu_z_Page_014thm.jpg
bc866f544e0113e9a8fa9c948ebfa8f3
1a738bba095ab1e51a0c14cf2e74e6f61a561436
50703 F20101222_AABJZC wu_z_Page_079.pro
c62adbcf134350a5444aa7a94fe9392c
cd8b968c8cd955caf6bf34857961f0b81832142d
33356 F20101222_AABKDU wu_z_Page_005.QC.jpg
0f98fd4d037b0ff45090587b9e336b06
edc5abd8f714bacccfc1c6c81fbd3fa049d15adc
46718 F20101222_AABJYO wu_z_Page_064.pro
7be8f2cf7cc9b70cbe682941a515eb25
34474bec52f0a89a46e490881fa82fbd111ea7a9
31483 F20101222_AABKEJ wu_z_Page_015.QC.jpg
a37a6b202c80e22e6f4d1509146caac2
da2174fe39f177288d93614c61dd34c463a11e40
51670 F20101222_AABJZD wu_z_Page_080.pro
bd8d32afe197c34bffc514acda283b75
915330d088a92f59906a94636fdaa7b970254908
24780 F20101222_AABKDV wu_z_Page_006.QC.jpg
251257e36cd8d8c1fca7c29b9e1ab81b
f9c85652f695e1ff324ef4c3ad208b84549f1e49
55703 F20101222_AABJYP wu_z_Page_065.pro
2b0a04626f526ab1403551ce69ba9631
5974c71c3afea9deeb97cf62c7efb65befb41c77
8109 F20101222_AABKEK wu_z_Page_015thm.jpg
d8513d3d3dff897946fde6d543c9a458
ccb48c1607591416bc769e75dbe9fb3b6fd92810
52330 F20101222_AABJZE wu_z_Page_081.pro
533a59027c5077089489169ceeea4e44
6793e58f0d79ce4dd3d83ff37761a4d96d4dae65
4501 F20101222_AABKDW wu_z_Page_007.QC.jpg
b9df3a17a57ca3a6167b3b3fa0f90ae1
57eea7821f7fdd4fe66728f964f1943223a99ce7
52027 F20101222_AABJYQ wu_z_Page_066.pro
e1b1bc1958681ce3e7462a37d902ff0a
e3ada3db1bdd6b3a0f57e29d469b096f82562b55
38495 F20101222_AABKFA wu_z_Page_026.QC.jpg
236dae34ebea27b85558feb0246b2ce0
1a07f48ab1220511f22acd60a33c862d7ee31872
25329 F20101222_AABKEL wu_z_Page_016.QC.jpg
f50a8525b0e89f5a66b5e1e1cc611323
c9fb3fd405bdb94f4ace85df24ba4d4872f78f27
51988 F20101222_AABJZF wu_z_Page_082.pro
6432b84cdd0fad1d8f65207d8f8b036e
aad6dea5fb19242673bc07349a06903bc0b8ff37
1597 F20101222_AABKDX wu_z_Page_007thm.jpg
7c7c2fbdb60ee69b856ac55febfaa891
108d93deb4494378a68fc1ce3d07f148b76afd4b
54162 F20101222_AABJYR wu_z_Page_067.pro
7348d6efdad9b124e19efe68e0e24e27
a6709f7185c74e6659ae0760c77072e4ff5de889
7254 F20101222_AABKEM wu_z_Page_016thm.jpg
999a34ecdc621fdb5218c13f74c5f042
ce2087e87515484ffae1075f1b91a04e461d9691
40106 F20101222_AABJZG wu_z_Page_083.pro
7251e2097cefbe87fa00010b072dfc0a
87504fbd5fee81ae87433c63ea97414b9064ebc2
39958 F20101222_AABKDY wu_z_Page_008.QC.jpg
aaea31cd02f56c8436a7f97390b2d9d9
f1bca1e789bc666aa514249f9df4595e1fad5960
F20101222_AABJYS wu_z_Page_068.pro
37824e09f86877f6cb372610e4065caf
1ff2ff14f7376c72fb3ec40f267ab9848e8c05af
9270 F20101222_AABKFB wu_z_Page_026thm.jpg
6761c300accfb41d23d984fb848c346d
61073148342e31f0c6bda8ca32e18408d1fce3e9
34205 F20101222_AABKEN wu_z_Page_017.QC.jpg
ee163366ed205e322fe995b9ac058201
88b8f8115d97c7f06169ce0963c0a063b7c9b522
4323 F20101222_AABJZH wu_z_Page_084.pro
035f767ad536db2035ae8d48cd96bd18
7a609db95e1f642ea219b918d29b579264fe1289
9546 F20101222_AABKDZ wu_z_Page_008thm.jpg
74c16b3332ca0fcfe09a5f2f93e36576
6081ec6992459f041156efde3c47e3e2abbd74dc
10989 F20101222_AABJYT wu_z_Page_069.pro
6c7491ecbb2748df34693e15e1c907c7
d61e0f7b0540260cf539ba49df322cb7242e68e2
36194 F20101222_AABKFC wu_z_Page_027.QC.jpg
3eb2aeb8a1ea9ad5a7daa233b71b4dff
2a1c5de079f4fc9441c111118f49b672f2fde479
9988 F20101222_AABKEO wu_z_Page_018.QC.jpg
4aae27871f52a3cfeb2c627ea407add4
edd0d9e2f59355517f4081e86b914990ef755071
6969 F20101222_AABJZI wu_z_Page_085.pro
9a5f6cf969d7febe666284bdd6110fec
02f92412abf987d7f9e2453f294ec2aa680e4533
6397 F20101222_AABJYU wu_z_Page_070.pro
3c068660e6711b933b26a17bbe997b8f
15793eedcc07b569bcbbc3a7f2e2384f8d327405
8843 F20101222_AABKFD wu_z_Page_027thm.jpg
9e8c2b735b2082ba21d2a1c3b0f2af76
20249007c8a09b858a408368dde524e94820720c
2430 F20101222_AABKEP wu_z_Page_018thm.jpg
dc08f0a293777a1bbce3f8eb75e816c3
4ceec063313027d7bfe844f3d3b50f761b18afb9
6135 F20101222_AABJZJ wu_z_Page_087.pro
657a990e2ba1dd88ae68dcd46a69f2b2
305ba3a4144442a6c5ce4ce7e11465db220520e7







TUNABLE CONTACT BARRIER OF SINGLE WALL CARBON NANOTUBE FILMS FOR
ELECTRICAL CONTACT TO SEMICONDUCTORS AND POLYMERS






















By

ZHUANGCHIUN WU


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

UNIVERSITY OF FLORIDA

2008

































O 2008 Zhuangchun Wu





























To my parents, my wife and my daughter









ACKNOWLEDGMENTS

I thank my academic advisor, Dr. Andrew G. Rinzler, for his thoughtful knowledge and

emotional support throughout my PhD study. I also thank my committee members for their

advice. The cooperative research proj ects with Dr. Reynold' s group gave me insights about

polymer science. I have used various instruments and equipment from Dr. Herbard's lab which

made my research much easier, and I was benefited from his "Solid State Physics" lecture.

Dr. Jeremiah Mwaura worked closely with me for the OLED proj ect. I also thanks to Dr.

Kyu-pil Lee, who did collaboration with me on the GaN LED proj ect.

Dr. Zhihong Chen showed me the ropes of the lab and I benefited from discussions and

hand on hand training from her. Dr. Jeniffer Sippel-Oakley, Mr. Bo Liu, Mr. Mitch McCarthy

helped me in various ways throughout my Ph.D study. Mr. McCarthy helped me to build a

chamber for nanotube/silicon heterojunction sensitivity to environmental gas study. Guneeta

Singh deserves my special thanks for all the enlightening conversations about life, culture and

everything beyond physics. I would like to thank Ms. Debra Anderson from UFIC and Ms.

Barbara Bostin for their continuous help for my family. I would like to thank Physics machine

shop personnel for helping me machining various experimental setups.











TABLE OF CONTENTS


page

ACKNOWLEDGMENT S .............. ...............4.....


LI ST OF T ABLE S ................. ...............7................

LI ST OF FIGURE S .............. ...............8.....


AB S TRAC T ............._. .......... ..............._ 12...


1. INTRODUCTION .............. .....................14


1.1 Single Walled Carbon Nanotube .................. ...............14.......... ...
1.2 Chiral Vectors, Unit Cell of Graphene Lattice ........._......... ....... ..._.. .. ........._.......1
1.3 Energy Dispersion of the Graphene Lattice and Single Wall Carbon Nanotube
(SW N T) ............... .. ............ ... ...............16.......
1.4 Density of States of SW NT ................... .... ........... ..... .........1
1.5 Motivation for the Single Wall Nanotube Film Studies in this thesis .............. ..... ..........18

2. SINGLE WALLED CARBON NANOTUBE THIN FILM-FABRICATION AND
PHYSICAL PROPERTIES .............. ...............23....


2. 1 Fabrication of a Carbon Nanotube Film ................ ...............23...
2.2 Transfer of a Carbon Nanotube Film to a Substrate ................. ................. ..........26
2.3 Surface Morphology of a Carbon Nanotube Film .............. ...............26....
2.4 Resistivity of a Carbon Nanotube Film .............. ...............27....
2.5 Optical Spectroscopy of a Carbon Nanotube Film ................. ...............27.............

3. SHIFTING THE FERMI LEVEL OF CARBON NANOTUBE FILM ................ ................34


3 .1 Introducti on ............... .... ................... ..... .... ........ ...... ........ ........ ...3
3.2 Optical Analog to the Electrolyte-Gated Nanotube Based FET (O-NFET) ................... ..35
3.2.1 Experim ental Details ........................ ...... ... .. ... .. ........3
3.2.2 Transmittance Spectrum of O-NFET as a Function of the Applied Gate
V oltage ................... .. .......... .. .... ....... ... .. .. ........ .......3
3.2.3 The Film Resistivity as a Function of the Applied Gate Voltage. ................... .......39
3.3 Discussion and Explanation............... ..............3
3.4 Time Drive of the O-NFET ................. ...............42.......... ...
3.5 Conclusion ................ ...............42........... ....


4. OHMIC CONTACT COUPLING CARBON NANOTUBE FILM TO P-GALLIUM
NITRIDE .............. ...............52....


4. 1 GaN Light Emitting Diode Background .................. .. ............ ......... .......... ..... 5
4.2 Experiments Details for Fabricating SWNT Film Based GaN LED ............... .... ...........54
4.3 Re sults ............ ..... .._ ...............55..












4.3.1 Contact Resistance............... .. ......... ...................5
4.3.2 I-V Characteristic of Carbon Nanotube Film/p-GaN ................ ............. .......56
4.4 Discussion ................. ...............56........... ...
4.5 Conclusion ................ ...............57........... ....


5. ORGANIC LIGHT EMITTING DIODE WITH SWNT FILM AS ANODE ................... ........64


5 .1 Introducti on ................. ...............64........... ...
5.2 M EH-PPV ............. ...... ._ ...............64...
5.3 Experimental Details .............. ...............65....
5.4 Discussion ............... ....__ ............... ...........6
5.4.1 MEH PPV does not wet with SWNT .............. ...............68....
5.5 Conclusion ................ ...............69................


6. CARBON NANOTUBE/SILICON HETEROJUNCTION ................. .......... ...............74


6. 1 Introduction about Schottky Barrier ................. .. ...............74......... ...
6.2 Experimental Details of SWNT/p-Si Heterojunction ................. ......... ................75
6.3 Results of SWNT/p-Si Heterojunction and Discussion............... ...............7
6.4 SWNT/n-Si Photovoltaic Device............... .....................7

6.5 Experimental details for SWNT/n-Photovoltaic device .............. ....................8
6.6 SWNT/n-Si Photovoltaic cell results and discussion .............. .... ...............80.

6.7 Conclusion on SWNT/p-Si heterojunction and SWNT/n-Si solar cell ............................82


7. PATTERNING OF SWNT FILM .............. ...............102....


7. 1 Patterning of SWNT Film to Sub-Millimeter ................. ...............102.............


8. SUMMARY AND FUTURE WORK .............. ...............106....


8. 1 Summary ................. ...............106................
8.2 Future Work............... ...............106.


LIST OF REFERENCES ................. ...............107................


BIOGRAPHICAL SKETCH ................. ...............111......... ......










LIST OF TABLES


Table page

4-1 Contact resistances of Ti/Al/Pt/Au on carbon nanotubes, carbon nanotube film on p-
GaN, and standard Ni/Au on p-GaN............... ...............59.










LIST OF FIGURES


FiMr page

1-1. Chiral vector Ch, translational vector T and the unit cell (OACB) of a (4,2) SWNT
illustrated on a graphene sheet. ............. ...............19.....

1-2. Band structure of graphene lattice (top) and the Brillouin zone of graphene lattice
(bottom). The quantization of the wave vectors perpendicular to the tube axis of the
nanotube lead to a set of discrete set of energy sub-bands (red parallel lines in the
bottom ). ............. ...............20.....

1-3. Energy dispersion relationship of (10, 10) and (10, 0) SWNT. Notice the cross of the
valence band and conduction band of the (10, 10) nanotube at the Fermi level, which
implies it is a metallic nanotube, whereas the energy gap at the Fermi level (0 in
these plots) show the (10, 0) nanotube to be semiconducting. ............ .....................21

1-4. Density of states (DOS) of (10, 10) and (10, 0) nanotube. Notice the finite states at the
Fermi level of (10, 10) nanotube, resulting it' s characterization as a metal. While the
(10, 0) has a gap, indicating it's a semiconductor. ................ .............. ........ .....22

2-1. AFM image of a 50nm SWNT film. ............. ...............3......1

2-2. The transmittance of a 50Onm before and after baking at 600oC in Argon. .........................3 2

2-3. Transmittance of a 240nm free standing SWNT fi1m before (grey curve) and after
(black solid curve) baking, with wavelength up to 120 micron meter. Also shown is
the transmittance vs. much shorter wavelength range of a 50nm SWNT fi1m on
quartz substrate. The inset shows the electronic transaction between valence band
and conduction band of the DOS from (12,8) semiconducting nanotube and (10,10)
m etallic nanotube. .............. ...............33....

3-1 The schematic drawing of a solid state optical modulator ............. ........ .............4

3-2. A sketch of the optical analog to the electrolyte-gated NFET .............. ....................4

3-3. Three transmittance spectrum as a function of applied counter electrode voltage, i.e.
OV, +0.4V and -0.5V. ............. ...............46.....

3 -4. Density of states (DO S) of (12, 8) single walled carbon nanotube ................ ................. .47

3-5. The transmittance of the nanotube fi1m as a function of the applied gate voltage, from
+1.8V to -1.8V. The transmittance at S1 peak (1676nm) increased from 44% to 92%,
while the S2 peak (932nm) varied from 51% to 68%. The IR at 3080nm decreased
from 97% to 75%, at opposite direction to the S1 peak modulation. ............ .................48

3-6. The resistivity of the nanotube fi1m as a function of the applied gate voltage......................49










3-7. The transmittance at S1, S2 and Ml depends on the gate voltage. ............ ....................50

3-8. Time drive of the S1 transmittance peak with +/- 1V gate voltage. .........__............_._.....51

4-1. Schematic view of the GaN based light-emitting diode using SWNT fi1m as the p-
Ohmic contact. ............. ...............60.....

4-2. IV characteristics of GaN LED with different metal, i.e. Ti/Al/Pt/Au, Pd only and
Pd/Au contact on carbon nanotube Eim. All the devices are using carbon nanotube
fi1m as the p-GaN contact. ............ ...............61.....

4-3. Emission spectrum of the GaN LED with inj section current 0.1ImA ................. ................ .62

4-4. Picture of the visible emission from the GaN LED with carbon nanotube fi1m as the p-
GaN contact el ectrodes. ............. ...............52.....

5-1. A) Structure of 1VEH-PPV (top) and B) Illustration of the OLED device with
ITO/PEDOT PSS as anode (middle) C) OLED device with SWNT fi1m (bottom) ..........70

5-2. ITO/PEDOT PSS based OLED radiance and current vs. voltage............ ... .........___...71

5-3. SWNT fi1m based OLED radiance and current vs. voltage .............. ....................7

5-4. Illustration of the energy band alignment of the OLED devices............. .._........._ ....73

6-1. Flat band model of contact barrier and Schottky barrier height ................. .....................84

6-2. Illustration of single gate and double gate of carbon nanotube/Si heterojunction ................85

6-3. Three steps explaining the SWNT/p-Si single gate heterojunction device fabrication
process............... ...............86

6-4. Circuit diagram of the single gate device (top) and double gate device (bottom) .................87

6-5. Source drain current as a function of gate voltage for SWNT/Si heterojunction with
heavily and lightly doped p-Silicon. Over a course of +/-0.7V gate voltage, Isd
modified 10 and 36 times, for heavily and lightly doped silicon, respectively. ................88

6-6. IV characteristics of the SWNT/p-Silicon heterojunction at different gate voltage A)
Lightly doped p-silicon and B) heavily doped p-Si ................ .............................89

6-7. Flat band model of modulating the contact barrier of SWNT/Si heterojunction by
shifting the Fermi level of SWNTs. .............. ...............90....

6-8. The current decreased significantly after 5 hour of continuously application of gate
voltage for the lightly doped silicon device. Same phenomena observed for heavily
doped silicon device as well. ............. ...............91.....










6-9. Double gating effect, with identical source gate and drain gate. Isd can be modulated
more than 300 times by applying gate voltage +0.7V. ............. ...... ............... 9

6-10. Carbon nanotube silicon junction current in argon environment will change upon
introduce oxygen or air. .............. ...............93....

6-11. Carbon nanotube silicon junction transport vs different ambient condition. ................... .....94

6-12. Circuit diagram of the boot-strap PV device ................ ...............95.............

6-13. I-V characteristic of the SWNT/n-Si solar cell, dark (top) and under illumination
(b ottom) ................. ...............96.................

6-14. The series resistance was determined to be 283 ohm for low doping n-silicon solar
cell. ............ ...............97.....

6-15. IV characteristic for the solar cell under different bias .............. ...............98....

6-16. The close up of Figure 6-14, to show the open circuit voltage change under different
gate bias. ............. ...............99.....

6-17. Flat band model of gating effect for SWNT/n-Si junction (top) and illustration of
photovoltaic effect (bottom). The blue line illustrates the over-gated situation, where
there is a barrier generated for the hole to transfer from the silicon to the SWNT. ........100

6-18. Energy band alignment at the SWNT/n-Si interface before (red) and after (black and
blue) gate voltage is applied. For the black and blue curve, we can see the built in
potential is increased. However, the further increase the gate voltage will push the
Fermi level of the SWNT too low, to generate a barrier for hole to transfer from
Silicon to SWNT, reduce the charge harvest efficient ................. .........................101

7-1. Illustration of coarse patterning of interdigitated finger with line width 0.1Imm ........._......105









LIST OF ABBREVIATIONS

SWNT: Single walled carbon nanotube

OLED: Organic light emitting diode

PV: Photovoltaic

TEM: Transmission electron microscopy

AFM: Atomic force microscopy

SEM: Scanning electron microscopy

ITO: Indium tin oxide

EMI-B TI: 1 -ethyl-3 -methylimidazolium bi s-(trifluoromethyl sulfonyl)imi de










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

TUNABLE CONTACT BARRIER OF SINGLE WALL CARBON NANOTUBE FILMS FOR
ELECTRICAL CONTACT TO SEMICONDUCTORS AND POLYMERS

By

Zhuangchun Wu

August 2008

Chair: Andrew G. Rinzler
Major: Physics

Single walled carbon nanotubes (SWNTs) are quasi one dimentional molecule formed with

pure sp2 carbon-carbon bonding. They are chemically stable, near ballistic transporters, and have

high surface area. According to their chiral vector and diameter, they can be either metallic or

semiconducting. They are a new class of material in the nanometer range which has unique

physical and chemical properties. Their unique one dimensional character gives them unusual

electrical properties.

A method of manufacturing thin films made of 100% SWNTs was explored. The carbon

nanotube films made by this method are thin enough to be transparent, yet still conducting. The

electrical and optical properties of those films have been studied, as well as application examples

for integrating the film into opto-electrical devices. The SWNT films made by this method have

following advantages: they are thickness controllable, uniform, transparent, conducting, and

formed at room temperature. We found the SWNT films were comparable in terms of the

conductivity and transparency to the standard industry transparent electrode, indium tin oxide

(ITO).









An optical analog of SWN\T based FET was constructed which demonstrated the ability to

shift the Fermi level of the SWNT film by about 0.9eV. We can shift the Fermi level by such

large amount is due to the one dimensional characteristic of SWNT such that the density of states

(DOS) of the nanotubes are finite and easy to be filled or emptied upon injecting or depleting

electrons by applying a gate field. This unique feature makes the SWNT film favorable in

making ohmic coupling to usually difficult materials, like GaN.

The SWNT film has been demonstrated to form ohmic contact with p-GaN light emitting

diode (LED). Studies on using SWNT as anode for constructing organic LED (OLED) have

shown great hopes. Several challenges of this system have been identified and further research is

still on the way.

The heterojunction between SWNT film and silicon has been investigated. The carrier

inj section across the p-silicon and nanotube film junction has been modulated by a factor of 300

by application of a gate voltage of less than 2V. The contact barrier was found sensitive to the

environment. Both oxygen and other component of the atmosphere (most likely water vapor) will

contribute to the junction resistance change. The n-silicon and nanotube film will form Schottky

barrier and showed photovoltaic effect.

Patterning the SWNT films to sub-millimeter range was demonstrated, which can be

beneficial to all the applications.

In summary, the SWN\T film provides a new type of film for electrical coupling to opto-

electronic devices as well as various inorganic and organic materials. The ease of Fermi level

tunability for this nanotube film provides the new opportunity for working with a wide range of

materials, including previously proven to be difficult ones.










CHAPTER 1
INTTRODUCTION

1.1 Single Walled Carbon Nanotube

A single walled carbon nanotube (SWNT) is essentially a single atomic layer of a graphene

sheet rolled to form a seamless tube, with a diameter on the order of one nanometer, its length in

the range of micrometers up to one centimeter. Multi walled carbon nanotubes (MWNTs) are

nested coaxial layers of SWNTs. Carbon nanotubes have remained a highly active research area

since the MWNT structure was elucidated in 1991 by lijima (1). SWNTs are quasi one

dimensional objects. They have rich physical properties as well as many potential applications.

According to their chirality and diameter, SWNT can be either metallic or semiconducting,

making them very attractive as the next generation of bottom up building blocks for electronic

devices (2-4). All the carbon atoms in the SWNT lattice are bonded with sp2 o bonds, which are

among the strongest chemical bonds, making SWNT very stable and inert to covalent chemistry

except under extreme conditions. Various sensors (5-8) were manufactured based on the SWNTs

by detecting their resistance change upon exposure to various chemicals. Functionalization of the

SWNT was another research topic to make SWNTs either binding to different agents for the

purpose of biosensor applications or for separating metallic nanotubes from semiconducting ones

(9-11). Nanotube based field effect transistors (NFET) also have been demonstrated (12-14), as

well as integrated logical circuits based on semiconducting SWNTs (15,16).

A central topic in this thesis concerns the ease of shifting the Fermi level of a conducting

film made of SWNTs. I will first introduce a method of making thin, transparent and conducting

SWNT films, followed by a description of the physical properties of the film. Then, two

examples of using this thin film as the electrode for optoelectronic devices, in which the Fermi

level shift of the nanotubes plays an important role will be discussed. Finally, by shifting the









Fermi level of the nanotubes in the film, the transport across the heterojunction between the

SWNT film and silicon will be modulated.

This first Chapter will give the basic fundamentals of the SWNTs as well as the motivation

of this study. In Chapter 2, the fabrication as well as electrical and optical properties of the

SWNT film will be explored. In Chapter 3, a SWNT film is utilized as the electrode to make

Ohmic contact with the p-GaN side of a GaN LED. Chapter 4 demonstrates an optical analog of

nanotube based FET and successfully shifts the Fermi level of a nanotube film by about 0.9

electron Volts. Chapter 5 discusses using the nanotube film as an anode to construct an organic

light emitting diode (OLED). Chapter 6 investigates the nanotube silicon heterojunction and

modulating the transport across the junction by applying a gate field. Chapter 7 explains a

method of patterning the nanotube films down to sub millimeter scale. Chapter 8 gives a

summary to this thesis and points out the future research needed to be done.

The remainder of this chapter will introduce the fundamental physical properties of

SWNT, which are closely followed according to reference 17 and 19. Chiral vectors, followed

with energy dispersion relationship and density of states of the SWNT will be explained and will

lay the theoretical foundation for the experimental phenomena in this thesis.

1.2 Chiral Vectors, Unit Cell of Graphene Lattice

Starting from the unit vectors of the two dimentional graphene lattice at and a- 2, we define

a chiral vector Ch.


Ch, =nt 1~ Ei 2 H,m) With Inl
As illustrated in Figure 1-1, start from point O, we have the chiral vector (4 aE 1+2 aE 2) to

point A, such that OA is the chiral vector Ch Later we will roll this OA to form a seamless tube,

which will called (4,2) nanotube.










Next we define a translational vector T (also starting with point O, OB), which is

orthogonal to chiral vector Ch, i.e.


Ch F= 0

BC can be obtained through translating OA by amount of T This OACB will make up a

unit cell of the (n,m) nanotube, in this case, a (4,2) nanotube.

The nanotube (n,m) will be formed if we cut the OACB out of a graphene sheet and roll

along the OA direction to form the circumference of the nanotube. So the diameter dh, of the

nanotube is


dh=2L/, = Chh h _"u 2 2+


This OACB is the unit cell in real space.

1.3 Energy Dispersion of the Graphene Lattice and Single Wall Carbon Nanotube (SWNT)

By using tight binding approximation, only considering the nearest neighbor atoms, one

can get the energy dispersion of the graphene sheet, as illustrated in Figure 1-2.

In order to calculate the energy dispersion of a SWNT, we need to impose the boundary

condition to the 2D graphene lattice, since the nanotubes are rolled up into a tube.

Ch K = 2@q, where q=1, 2, ....N, N is the number of hexagons in the real space unit cell

OAB'B.

This will result in a quantized wave vector


k = (k, I T, the nanotube axis, q=1,2,...N)


By applying this boundary condition to the 2D graphene lattice, the SWNT dispersion

becomes simple cut lines through the 2D dispersion relation of graphene in the direction









specified by the boundary conditions (in turn dictated by the real space direction of the nanotube

axi s).

Examples of the energy dispersion relationship of (10,10) and (10, 0) nanotube are shown

in Figure 1-3. It is clear that the valence band and conduction band of (10, 10) nanotube has a

cross point right at the Fermi level, which makes it intrinsically metallic. While for (10, 0)

nanotube, there is a band gap at the Fermi level. So, it is a semiconducting nanotube.

1.4 Density of States of SWNT

In order to calculate the density of states (DOS) of a SWNT, one can take the energy

dispersion relationship of the SWNT, divide the energy into small intervals and sum the k

vectors appearing in each energy segment. This procedure leads to density of states diagrams

specific to each nanotube of distinct (n,m) index such as shown in Figure 1-4.

As shown in Figure 1-4, (10, 10) and (10, 0) nanotube density of states again confirms that

the (10, 10) nanotubes are metallic, with finite density of states at the Fermi level, while the

semiconducting (10, 0) nanotube has a band gap right at the Fermi level. The sharp features in

the DOS of the nanotubes are called von Hove singularities. There can be electronic transitions

between symmetric valence band and conduction band von Hove singularities, resulting in

absorption bands in the optical spectrum of the SWNTs. The band gap for the semiconducting

nanotubes is inversely proportional to the diameter of the nanotube (17).

The density of states of SWNT is finite, but the mobility along the nanotube is near

ballistic (18,20). The effective scattering length is on the order of micrometers for metallic tubes

and a few hundred nanometers for semiconducting nanotubes (18,20). Because the carbon

nanotubes are only quasi one dimensional, they have some tolerance for defects. Although the

density of states for the nanotubes is low (compared to typical metals) the charge mobility is very









high making the nanotubes good conductors. The Fermi level of the nanotube is easily tuned

because of the low density of states.

1.5 Motivation for the Single Wall Nanotube Film Studies in this thesis

Techniques developed to produce thin homogeneous films of SWNT that are

simultaneously transparent and electrically conducting are exploited to study the potential of the

SWNTs as Fermi level tunable electrodes in electronic and optoelectronic devices.













































Figure 1-1. Chiral vector Ch, translational vector T and the unit cell (OACB) of a (4,2) SWNT
illustrated on a graphene sheet.














i Reprinted with permission from Ph. Avouris, Z. Chen and V. Perebeinos, Nature Nanotechnology 2, 605 (2007)
Copyright C 2007, Nature Publishing Group































Figure 1-2. Band structure of graphene lattice (top) and the Brillouin zone of graphene lattice
(bottom). The quantization of the wave vectors perpendicular to the tube axis of the
nanotube lead to a set of discrete set of energy sub-bands (red parallel lines in the
bottom). 2


























2 Reprinted with permission from Ph. Avouris, Z. Chen and V. Perebeinos, Nature Nanotechnology 2, 605 (2007)
Copyright C 2007, Nature Publishing Group













8 (10,10) 8












-8 --~~ -----e-S" -8-












-8~- -8




Fiur 1-3 Enrydseso eainhpo 1,1)ad(0 )S NT. oietecoso h

valence~~~~~~~~~( badadcnuto bn fte(0 0) nntb tteFrilvl hc
imle iti ealcnntbweesteenrygpa h em ee 0i hs

plts showC-- th (0,0 nnoub o e emcndctng









Fi Reprn with perm y ispsion frmZ..Celains heisUiest of Flori0 ada(, 2003W T otc h roso






























-3 -2 -1 0 1 2 3
Energy (ekV)


(1 0,0)


-3 -2 -1 0 1 2 3
Ene~rgy (eV)


Figure 1-4. Density of states (DOS) of (10, 10) and (10, 0) nanotube. Notice the finite states at
the Fermi level of (10, 10) nanotube, resulting it' s characterization as a metal. While
the (10, 0) has a gap, indicating it' s a semiconductor. 4



4 Reprint with permission from Shigeo Mamuyama www.photon.t.u-tokyo.ac~jp/









CHAPTER 2
SINGLE WALLED CARBON NANOTUBE THIN FILM-FABRICATION AND PHYSICAL
PROPERTIES

Prior to 2002, much of the effort in the community of SWNT research was focused on the

individual nanotubes (2,3,5). In this thesis, a different approach was explored by using the

SWNT effectively in bulk as thin films possessing thickness less than 200 nm over macroscopic

areas (up to 5.5 inches in diameter). The film fabrication and properties are described here.

2.1 Fabrication of a Carbon Nanotube Film

All the single walled carbon nanotubes used in this study were synthesized from laser

vaporization growth and later purified by Dr. Andrew Rinzler (4,24). It will be referred to as

"carbon nanotubes" or "nanotubes". The growth and purification process is briefly described as

follows.

The laser vaporization growth apparatus involves a dual pulsed laser beam, precisely timed

to fire one after another to hit a rotating target. The target was composed of a mixture of catalyst

cobalt and nickel particles, 1 atomic percent each, together with carbon, sitting in the hot zone of

the 1200oC furnace, with flowing argon (4,24), all in a two inch diameter closed quartz tube. The

SWNT deposit on the down stream of the quartz tube wall, together with a mixture of the

amorphous carbon and metal catalyst particles. All the amorphous carbon and catalyst particles

need to be removed in order to get the pure SWNT.

The carbon nanotubes were purified by a cross flow filtration method to remove the

amorphous carbon and catalyst particles (24). In order to separate into small bundles and

individual nanotubes, typically they will be suspended in an aqueous solution (25), otherwise the

nanotubes entangle with each other to form an intractable mass. Since the surface volume ratio of

the nanotubes is tremendous, the Van der Waals force is strong enough to hold them together. In

order to suspend in water, some kind of surfactant is needed. It is found 1 weight percent of










Triton-X 100 surfactant solution will stably suspend the nanotubes for a long time (24, 25). The

surfactant has a hydrophobic head and a long hydrophilic tail. The surfactant will form hydration

shells around the nanotubes. Only once the surfactant concentration reaches the so called critical

micelle concentration (CMC), does it starts to coat the nanotubes. The process of surfactant

coating of the nanotube is a dynamic process. Some surfactant will always come off the nanotube

and some other surfactant molecules newly bind onto the nanotube surface. If the surfactant

concentration is lower than the CMC, the surfactant polymers are rather associated with each

other to form the micelles. Once the concentration is above CMC, there will be enough free

surfactant for the nanotube surfaces to begin to be coated.

Even with the use of a surfactant energy must be put into the material to break apart the

large agglomerates of nanotubes. This is accomplished using ultrasonication. The ultra

sonication energy will break the aggregates apart. The surfactant can then quickly coat the

nanotube bundles preventing reaggregation. Too much sonication will start to cleave the

nanotubes and result in shorter tube length. So, it' s a matter of balancing the degree of bundle

disaggregation to individual nanotubes versus the tube length. To make a uniform SWNT fi1m, it

is desired to have more individual and small bundle nanotubes in the solution. It will make the

film more homogeneous and decrease surface roughness. This becomes more important when

making organic optoelectronic devices where the working polymer layer is in the order of 100

nm thick and the surface roughness of the SWNT fi1m becomes critical. Apart from the bundle

size, the particles in the nanotube fi1m are another critical issue, just like the silicon industry

needs to lower the particle concentration in a silicon wafer. Much effort including extra filtration

process and the environment where those fi1ms were made have been taken into account to

address the particle issue. More details will be discussed in chapter 5.









The nanotube solution concentration (mass/volume) in a stock solution was determined

simply by making a thick nanotube film by the filtration method (described below) from a

known volume of solution on a Teflon filter membrane so that the film could be peeled off the

membrane (so called buckypaper) allowing determination of its mass with a microbalance. Once

the concentration was determined a thin film was made from a known volume of the stock

solution and its thickness measured by atomic force microscopy (AFM). Once the film thickness

(for a particular area) resulting from one particular volume of the stock concentration was

known, a simple calculation will tell us how much solution is needed for any desired thickness

film of a given area.

The film is made by the filtration process trapping the nanotubes on filtration membranes

with pores too small for the nanotubes to permeate. The filtration process sketch is shown in

Figure 2-1. A membrane (cellulous ester, Millipore, VCWP) with pore size 100nm was typically

used. This material was carefully chosen so that it can be easily dissolved away later by

dissolution with acetone during the transfer process, resulting in a pure nanotube film, either

attached to a substrate or free standing. This fabrication process is self-regulated to form uniform

thickness films. If one region of the membrane were to accumulate more nanotubes, the

nanotubes themselves begin to impede the flow of the solution through that portion of the

membrane hence bringing fewer nanotubes to that region. The pore size of the nanotube film

made from a well sonicated solution usually is in the order of few to tens of nanometer, as seen

on AFM images. This means the pore sizes are much smaller than the average pore sizes of the

filter membrane used. After a few layers of the nanotube landed on the filter membrane, the

regulation of the flow was dominated by the smaller pore size of the nanotube film. Once the

solution was all gone through, the film was then rinsed with sufficient amount of DI water to









remove the remaining surfactant in the film. The nanotube film is dried on the filtration

membrane, which brings the nanotubes into intimate contact with each other.

2.2 Transfer of a Carbon Nanotube Film to a Substrate

In order to use the film, a transfer process is needed, either being transferred to a substrate

or held up by a frame for free standing purpose. The transfer process starts with a nanotube film

sitting on the membrane, wetted with DI water, then bringing the nanotube film in contact with a

clean surface. It can be virtually any surface of the material for which nanotube film to contact

with or the surface of a supporting substrate, as long as that substrate can tolerate the membrane

dissolving and the rinsing solvent, in this case, acetone and methyl alcohol. Once the two

surfaces were in contact, the two obj ects will be sandwiched by two metal plates and a clamp

will be used to apply pressure to the assembly. While sitting in an 800 C oven for half an hour,

the maj ority of the DI water initially for wetting the nanotube film will evaporate and the surface

tension of the water will bring the nanotube film into intimate contact with the substrate surface.

Then the membrane will be dissolved away, using sequential acetone baths (typically 4 baths) to

ensure removal of residual membrane material. Finally methyl alcohol was used as the final bath

to then dry the nanotube film sample out of a low residue solvent.

2.3 Surface Morphology of a Carbon Nanotube Film

The surface morphology of the nanotube film was characterized by the scanning electron

microscope (SEM) and the atomic force microscope (AFM), as shown in Figure 2-1. The root

mean square surface roughness is about 7 nm. For a thinner film, for example, a 7nm film, there

are appreciable areas of open spaces and other areas with nanotube bundles. In order to lower the

surface roughness, smaller bundles and more individual nanotubes will help. From AFM images,

we will see particles embedded in the nanotube films and also on the surface of the film. These

particles are typically on the order of micrometers, which is much bigger than the nanotube film









thickness. For using the nanotube film as electrodes in organic optoelectronic devices, those

particles will cause serious shorting issues, because those polymer photoactive layers are

typically on the order of 100nm thick. Increasing the polymer layer will decrease the device

efficiency because the carrier mobility is limited for polymers. It is an ongoing research project

being carried out in our group for improving the surface smoothness of the nanotube film.

2.4 Resistivity of a Carbon Nanotube Film

The resistivity of the nanotube film was measured by the Van der Pauw method. For as

prepared 50nm film, the resistivity is typically 3x10-4 G-cm (60 ohm/square) and the value is

quite stable, for as long as a month. The resistivity will increase 5-10 times when the film get de-

doped by baking at 600oC in Argon for half an hour. Once the film has been de-doped, it can be

re-doped by exposure to various dopants at room temperature, such as bromine vapor or nitric

acid vapor. The resistivity will recover or even be slightly lower than the as-prepared sample.

However, this doping is not stable, once the film is exposed to air for a couple of hours, it will

become de-doped.

2.5 Optical Spectroscopy of a Carbon Nanotube Film

Optical spectroscopy of the nanotube film has been characterized by a Perkin-Elmer dual

beam spectrometer (Lambda 900). The dual beam refers to sample beam and reference beam. By

placing the sample on a substrate in the sample beam and an identical substrate in the reference

beam the instrument normalizes the sample transmittance at each wavelength to the substrate

transmittance in the reference beam. Hence the absorbance of the sample is determined, without

substrate absorbance. The absorption of the light was come from the sample, in this case, a

nanotube film. The spectrometer generated a monochromatic light, through a slit, incident on the

sample (in our case, perpendicular to the sample surface), and the detector integrated the

radiation that passed through the sample at the same wavelength for a given period of time










(integration time). At the Ultra violet/Visible range, the slit width was kept as small as possible

to maintain the accuracy of the wavelength, but large enough to give a good signal to noise ratio.

In the near infrared region (in our case, above 850nm) where the detector has poor dynamic

range, the intensity striking the detector was kept constant by modifying the slit width..

Therefore, the system adjusted the slit width to achieve the required constant intensity and the

width of the slit would be recorded and converted into the transmittance of the film.

A typical transmittance spectrum of the nanotube film is shown in Figure 2-2, before and

after baking at 600oC in argon environment. The dips in the transmittance around 1670nm,

930nm and 650nm are labeled S1, S2 and Ml, respectively. S1 and S2 peaks are the absorption

bands which correspond to the electronic transitions between the first and second pair of the Van

Hove singularities in the conduction band and valence band of the density of states of the

semiconducting SWNTs, as illustrated in chapter 1. Ml peak corresponds to the absorption from

the first pair of such transitions contributed by the metallic nanotubes. The absorptions observed

are broad, despite the very sharp structure of the van Hove singularities, because of the

dependence of the energy separation between the singularities on the nanotube diameter (recall

the inverse proportionality of the band gap to the nanotube diameter) and the diameter

distribution of the nanotubes present in the sample (17). The bundling of nanotubes in the sample

also acts to perturb the singularities smearing out the contribution from the discrete tubes to the

absorption.

It is worth while to emphasize that the absorption peak intensity for a given nanotube film

provides information about the occupancy of the nanotube electronic density of states. As the

Fermi level shifts inside the nanotube film, the S1, S2 and Ml absorption peak intensity will









change accordingly. The transmittance spectra thus provide information about the location of the

the Fermi level in the SWNT films.

Figure 2-2 shows the transmittance of a 50nm thick SWN\T film, before and after baking at

600oC inside an argon environment. The difference in the S1 and S2 intensity is obvious for pre-

and post baking of the film. As described below this change is due to a shift in the Fermi level of

the nanotubes upon baking. This Fermi level shifting is the result of chemical dopants removal.

We will show shifting the Fermi level of the nanotube film through electric field in chapter 3 and



In order to understand the change in the intensity of the S1 and S2 peaks before and after

baking for the nanotube film, we have to understand from the point of view of the DOS of the

nanotube. The inset of Figure 2-3 shows the DOS of the (10, 10) and (12,8) nanotubes. The

arrows in the inset correspond to electronic transition from the valence band singularities to the

conduction band singularities for S1 and S2 transition of the (12,8) nanotube and Ml transition

of the (10,10) nanotube, which correspond to the absorption peak in the transmittance spectrum.

If the Fermi level sits in the middle of the gap, the available electrons in the fully occupied

valence band and the available states in the empty conduction band are maximized, hence the

transition probability is maximized,. Once we p-dope the SWN\T film, the Fermi level will shift

towards the left. When it moves into the valence band, it will empty some states in the valence

band, making fewer electrons available for participating in the electronic transition, which will

result in the decrease of the absorption S1 peak. On the other hand, if we n-dope the nanotube,

the conduction band states will be partially occupied. Then the transition from valence band to

the conduction band will also get smaller, resulting in a smaller absorption S1 peak. If the Fermi

level moves deep into the valence band, further into the second van Hove singularity, it will start









to decrease the S2 absorption peak. The same analysis is applied to the Ml peak as well. This is

why we can use the spectrum as the indicator of where the Fermi level of the nanotube film sits

but cannot tell whether it is p doped or n doped. It is easy to tell whether it' s p or n doping by

shifting the Fermi level corresponding to the known external field applied to the nanotube film

while monitoring the spectrum, which will be explained further in the next chapter.

Figure 2-3 shows the transmittance of two films, one 50nm film sitting on quartz,

measured with a narrow wavelength window, the other a 240nm free standing SWNT film, with

a much broader wavelength window, up to 300Cpm, before and after baking at 600oC with argon.

Notice in Figure 2-3, there is a big transparency window between the wavelengths of 2 to

10 Cpm, at which range the industry standard transparent conducting film indium tin oxide (ITO)

is totally opaque. This unique feature makes SWNT film an outstanding candidate for using in

microwave electronic applications.

In summary, we described a simple method for manufacturing pure single walled carbon

nanotube films and characterized the electrical and optical properties of the film. The SWNT

films are comparable to ITO in terms of the conductivity and the transparency in the visible

range but far better transparent in the infrared region (ITO is totally cut off after 2Cpm). The

Fermi level of the SWNT film is readily shiftable upon depleting or adding electrons to the

nanotube film, either by chemical or as will be shown electrical means.












1.50


1.00





0.50


0 0.50 1.00 1.50


Figure 2-1. AFM image of a 50nm SWNT film.

































SReproduced in part with permission from [Z.Wu et.al, Science 305, 1273-1276 ] Copyright
[2004] AAAS











100

95-





E so st



70- M S2
-as-preparec
65-1 -600oC bakl
500 1000 1500 2000 2

Wavelength (nm)

Figure 2-2. The transmittance of a 50nm before and after baking at 600oC in Argon.


d


500



















1 10rr 100
Wavlegt (pm)

Fiue23 rnmtac fa20mfe tndn W Tfl eoe(rycre n fe
(bac soi ure ainwt waelnth pt 2 irnmtr losoni
th trnmtac vs. muc shre aeeghrneo a5n W Tfl nqat
subtrte Th ne hw h letoi rnato etenvlnebn n
codcto bado h O rm(28 eiodcignntb n 1,0
rmtali nanotube.6~'









6 eroue in ar w itpemsinfo [ZW eta Sine3512317]Cprgh[04]A S









CHAPTER 3
SHIFTING THE FERMI LEVEL OF CARBON NANOTUBE FILM

3.1 Introduction

In previous research, Dr. Zhihong Chen in our research group has designed a solid state

optical modulator (Figure 3-1) and demonstrated that upon applying a gate voltage, the Fermi

level of the carbon nanotube film can be shifted (as shown by a 0. 1% modulation of the S1

transmittance peak at 6V of gating voltage) (19). The solid state optical modulator is essentially

a transparent capacitor, with two electrodes sandwiched with AlOx as the dielectric layer. The

two electrodes are a thin layer of carbon nanotube film and indium tin oxide (ITO), respectively.

The whole structure sits on a glass substrate. By applying a voltage between the two electrodes

of the capacitor, the optical transmittance of the whole device was monitored by a dual beam

spectrometer, having the sample beam going through the entire device. Since S1 peak is

corresponding to the electronic transitions between the first pair of von Hover singularities of

semiconductor single wall carbon nanotubes, we reach the conclusion that the Fermi level of the

SWNT has been changed upon the application of the gate field. The reasons the total modulation

effect is rather small are two fold: the first and most important one is electronic screening. The

gate field will be mostly screened by the first several layers of the carbon nanotubes next to the

dielectric layer, leaving the rest of the nanotubes not experiencing the gate electric field. Second,

the gating field is rather small, since the dielectric layer is about 81nm thick. In order to increase

the modulation percentage, more charges are needed to inj ect into the electrodes. By simply

increasing the capacity, one can inj ect more charges. The capacitance per unit area is important

in this modulation effect. One way to increase the capacitance is to reduce the distance between

the two electrodes of the capacitor, which is reducing the dielectric layer thickness. But reducing

the dielectric layer will cause breaking down of the dielectric layer and cause leakage current.









For their operable voltage range electrolytic capacitors have much greater charge storage

capacity than dielectric capacitors. This is due to the fact that in the electrolyte, there are ions in

the solution that can move freely, in response to the electric field generated by the electrodes

immersed into the electrolyte solution. The ions will accumulate very close to the electrode,

which has the equivalent effect of reducing the spacing between the capacitor electrodes, helping

to inj ect more charges onto the electrodes. The distance for those ions that accumulate next to the

electrode is within the so called Helmholtz layer or double layer, which has a thickness typically

on the order of a few nanometers. Here the phenomenon is exploited to greatly enhance the

electric field gated modulation of the absorption bands in the nanotube films.

3.2 Optical Analog to the Electrolyte-Gated Nanotube Based FET (O-NFET)

We designed a device which is an optical analog to the electrolyte-gated nanotube based

field effect transistor (O-NFET). In this device we have two electrodes, one sample and one

counter (gate) electrode. They are identical SWNT films. The two films were both immersed in

the electrolyte, i.e., an ionic liquid EMI-BTI (1 -ethyl-3-methylimidazolium bis-

(trifluoromethyl sulfonyl)imide) (23). By applying a voltage between the two films, we could

inj ect electrons or holes onto each of the nanotube electrodes, respectively. With the help of the

ionic liquid, the charge we can inj ect onto the nanotube film electrodes will be much increased

compared to having no ionic liquid (as we explained above). The transmittance through the

sample SWNT electrode was monitored while applying a different potential between the sample

and counter electrodes. As discussed in the previous chapter, monitoring the transmittance of the

SWNT film at wavelengths corresponding to particular electronic transitions provides

information about the Fermi level within the film.









3.2.1 Experimental Details

As illustrated in Figure 3-2, Two identical, 150nm thick single wall carbon nanotube films

(15mm by 8.5mm) were transferred onto a sapphire substrate (25mm by 25 mm), with 2mm gap

between them. Two thin (50nm) strips of palladium film were thermally evaporated across the

top edges of the nanotube films to make electrical contact with each of them. At the bottom of

the sapphire substrate, a U shape thin rubber gasket was placed along the edge of the substrate on

the film side, and covered with a thin glass plate (25mm by 8mm, hence not in the path of the

light) to form a reservoir for the ionic liquid. The gasket, glass plate and substrate were held

together by a clamp.

Once the device was constructed it was held horizontally and the nanotube films were

wetted and saturated by ionic liquid EMI-BTI. This ionic liquid is viscous, with a very low vapor

pressure and stable at room temperature in ambient environment (23). Once the films were

wetted the device was tilted to be in a vertical position, the excess amount of ionic liquid drained

to the reservoir but by capillary forces a thin layer remained on top of the two nanotube films.

The two nanotube film strips were also in electrolytic contact through the ionic liquid drained

into the reservoir.

The O-NFET device was placed inside a Perkin-Elmer dual beam spectrometer (Lambda

900). By measurement it was previously confirmed that EMI-BTI is transparent from 300nm to 2

Cpm of wavelength. The monochromatic sample beam passes through the sample nanotube film

and the other nanotube film serves as the gate electrode. Initially we used a platinum wire

(1.5mm diameter) as the gate electrode. But since the surface area of the Pt wire (although we

increased the length of the wire by making several circles inside the ionic liquid reservoir) is

much smaller compared to that of the sample nanotube film, it would be a limiting effect for the










capacitance of this device. This led to the decision to use an identical nanotube film as the gating

electrode, thus solving this problem.

One disadvantage of this device is the the response time is rather long, typically in the

order of minutes. The rearrangement of ions in the viscous ionic liquid is slow. Accordingly a

delay of more than 5 minutes was used between each change of voltage and the spectral

measurement. The delay time was determined by monitoring the non-Faradaic charging current

of the device when the voltage was changed. Once the gate current dropped to tens of nanoamps,

the spectrum was recorded (spectra recorded when the charging currents differed by a factor of 2

in this range showed no discernable difference).

3.2.2 Transmittance Spectrum of O-NFET as a Function of the Applied Gate Voltage

The transmittance of the SWNT film electrode as a function of the gate voltage is shown in

Figure 3-3. The measurement takes place from UV to near infrared (IR), i.e. from 350nm to

3080nm. The gate voltages are OV (no gate voltage), positive 0.5V and negative 0.4V. First, let's

focus on the absorption spectrum when the gate voltage is OV. There are three absorption peaks,

S1 (1676nm), S2 (932nm) and Ml (652nm). As discussed in chapter two, the peaks S1 and S2

correspond to the electronic transitions from the conduction band to the valance band of the

semiconducting nanotubes in the SWNT sample film. This electronic transition is illustrated in

Figure 3-4, which shows the density of states of the (12, 8) semiconducting SWNT. As we

know, the energy gap is inverse proportional to the nanotube diameter (17). The diameter

distribution of the nanotubes in our sample is determined by TEM images in the reference 24,

which is mainly between 1.1-1.6nm. That energy gap agreed well with the (12,8) nanotube,

which has the diameter of 1.365nm.

Compare to the DOS of (12,8) nanotube, the S1 and S2 peaks in the transmittance

spectrum of the nanotube film are much more broadened. The broadening of the peaks is mainly









due to the distribution of the nanotube diameter in the film as well as to the bundling effect, as

we discussed in chapter two. Typical bundles in our films are about 5-10nm in diameter. These

bundles will also contribute to one of the gating effects, which will be discussed later in this

chapter.

In other words, for a given sample with a Eixed electronic status, the transmittance or

absorbance spectrum will be constant. The electronic status is the availability of the valence band

carriers and the conduction band carriers, or where the Fermi level is sit for a given material. If

we apply a voltage to inj ect electrons or holes onto that sample, it will shift the Fermi level of the

sample, by changing the conduction band or valence band carrier distributions, hence changing

the corresponding absorption spectrum. This is the case for Figure 3-5, where we applied +0.4V,

OV and -0.5V to the nanotube fi1m counter electrode. Because there is ionic liquid next to the

carbon nanotubes and the Helmholtz layer is on the order of nanometers, the nearby counter ions

allowed us to inj ect much more charge onto the entire bulk of the nanotube fi1m with a lower

gate voltage, compared to the solid state device discussed at the beginning of this chapter. When

a negative voltage (-0.5V) is applied to the counter electrode, positive ions from the ionic liquid

will accumulate at the counter electrode, and negative ions will accumulate at the sample

nanotube film. This will allow more holes to be inj ected onto the sample nanotube film, shifting

the Fermi level to the p side, resulting in less absorption. As a result, the transmittance at S1 peak

increased from 68% (V,=0V) to 73.9% (V,=-0.5V), which is consistent with our prediction. The

opposite situation happens when the positive voltage is applied to the counter electrode, in which

case the transmittance of the film at S1 decreases from 68% to 5 8%.

The transmittance of the nanotube film as a function of the applied voltages in the range of

positive 1.8V to negative 1.8V on the counter electrode (gate voltage) is shown in Figure 3-5.









This gate voltage range was chosen to remain below the reduction/oxidation threshold for EMI

BT I, .

3.2.3 The Film Resistivity as a Function of the Applied Gate Voltage

In order to confirm that the carrier density at given point of the gate voltage on the

nanotube fi1m agreed well with the Fermi level shifting picture, the resistivity of the nanotube

fi1m was measured as a function of gate voltage in a similar device using a Linear Research LR

700 AC Resistance Bridge operated at 16Hz. To facilitate the four probe resistance measurement

the sample film in this case had four 1 mm wide Pd strips deposited across the thin direction of

the film separated by 3 mm. Figure 3-6 shows the fi1m resistivity corresponding to the gate

voltage. As the gate voltage starts from negative 1.8V increasing towards zero and then positive,

the resistivity continuously increases until the gate voltage reaches +1.4V, after which the

resistivity decreases a little as the gate voltage continues increasing to +1.8V.

3.3 Discussion

If the semiconducting nanotubes in the nanotube film are intrinsic or un-doped, the Fermi

level of those nanotubes will be sitting at the middle gap of the DOS of the nanotubes. The

nanotubes in the films are a mixture of semiconducting and metal nanotubes, and the Fermi level

of the films will be the equilibrium of all the nanotubes in the films. So, the Fermi level of the

film will be sitting at the middle of the DOS of the nanotubes if all the nanotubes are intrinsic.

The following direction of "left" or "right" refers to the horizontal axis of the Figure 3-4,

the electronic energy of the nanotube film.

For intrinsic nanotubes, the S1 peak of the absorption spectrum will be at a maximum

when the gate voltage is zero. By inj ecting electrons to the nanotube film, those electrons will

occupy the conduction band density of states, which will shift the Fermi level to the right. Those

newly occupied conduction band states will be unavailable to participate into the electronic









transitions between valence band and conduction band, which will decrease the absorption of the

S1 peak. On the other hand, if we inj ect holes into the nanotube film (or deplete electrons from

the nanotube film), shifting the Fermi level to the left side, that will decrease the valence band

carriers to be excited into conduction band, also resulting in decrease of the absorption S1 peak.

The experimental observation is that when applying negative gate voltage, the

transmittance of S1 increases, or in other words the absorption decreases, which is agreed well

with our previous analysis. But by applying positive gate voltage, the transmittance of S1

decreases, or equivalent to the increase absorption of S1 peak. This tells us that the Fermi level is

not sitting in the mid-gap, rather, in the p-side (left of the gap), under the S1 valence singularities

when the gate voltage is zero. This is somewhat surprising since the nanotube films were baked

to 6000 C in Ar to de-dope them prior to their saturation with the ionic liquid. This seemingly

intrinsic p-type behavior of the nanotube film is likely due to the equilibration of the chemical

potential of the mixed metallic and semiconducting nanotubes with the Pd contact electrodes in

the presence of the surrounding ionic liquid.

A second initially perplexing phenomenon was that far before completely saturating the

changes in the S1 peak, the S2 peak starts to change. This is better observed in Figure 3-7, which

plots the transmittance at the corresponding peak positions versus the applied gate voltage.

According to the simple picture of the gate voltage induced Fermi-level shift, the change of S1,

S2 and Ml should appear sequentially as the Fermi level progress sequentially through the

corresponding valence band singularities.

This seemingly paradoxical behavior is readily explained by electrostatic screening and the

fact that the nanotubes in the films are not individual nanotubes but rather bundles of nanotubes.

For bundles, the inside nanotubes are not directly exposed to ionic liquid. At the lower applied









voltages only the outer nanotubes of the bundles will form double layer with ionic liquid and

participate in the electronic doping process. The charges on those outer layer nanotubes will

partially screen the inner nanotubes from the ionic field. Once the equilibrium is established for a

given gate voltage, the Fermi level for the outer layer nanotubes can well lie below Ml while the

Fermi level for the inner nanotubes can still lie below the S1 singularity. Since however the

carrier density of nanotubes is low compared to conventional metals the outer layer of nanotubes

can only partially screen the inner nanotubes so that at the higher applied voltages the S1 peak

can is virtually gone, which means even the Fermi level has shifted below the S1 valence

singularity even for the innermost nanotubes in the bundles.

The transmittance of the film in the near IR region (above 2200nm) responds to gate field

in the opposite direction as that of the S1 peak. When the gate voltage is increased in the

negative direction, the transmittance of S1 increases while that in the near IR region decreases.

The transmittance in the near IR region is governed by free carrier absorption. As the gate

voltage is made more negative the concentration ofp type carriers is increased (making the film

more conducting) and increasing the free carrier absorption explaining the observed changes inj

the near IR Transmittance.

From the resistivity versus gate voltage relationship, we can figure out whether the

nanotube film is intrinsic or either n or p doped. If the Fermi level is initially sitting in the n-side,

the resistivity will decrease when the gate voltage moving toward the negative, since the

negative voltage will make the film less n doped, which is not the phenomena we see in our

resistivity data. What is really going on is that when a negative voltage is applied to the counter

electrode, the Fermi level of the nanotube film shifts away from the band gap, resulting in the

drop of the resistivity of the film. This implies that the nanotubes were previously p-doped under










the zero gate voltage. When the Fermi level shifting away from the gap, the nanotubes will be

more p-doped, in other words, more conducting. Indeed the resistivity of the nanotube film will

decrease as the gate voltage becomes more negative, which is also confirmed by the near IR

region's transparency drop.

As the gate voltage increases from zero to +1.4V, the resistivity (Figure 3-6) increases

accordingly, which corresponds to a decrease in the carrier (hole) concentration, as the Fermi

level moves into the gap. This change in the resistivity saturates after 1.4V gate voltage and

actually begins to turn around indicating that the nanotubes have just begun to be n doped at the

largest voltage of 1.8V. This behavior is consistent with saturation appeared in the S1 peak after

+1.4V gate voltage.

3.4 Time Drive of the O-NFET

In order to confirm that the gating effect of the transmittance spectrum comes from the

gating effect, we performed a time drive measurement at the S1 wavelength to confirm that the

whole process is reversible. This step helps to rule out any redox chemistry contribution.

A square wave of +/- IV gate voltage was applied, each lasting 50 seconds with a period of

100 seconds, while monitoring the transmittance of the S1 peak at 1676nm. The transmittance of

S1 peak will alternate from 71% to 57%, fully recoverable, as illustrated in Figure 3-8. Notice

that the transmittance gradually changing but does not reach a plateau during the 50 seconds

period. As discussed previously, this is due to the slow reorganization of the ions in the viscous

ionic liquid during charging and discharging. The transmittance will reach a plateau after a few

minutes.

3.5 Conclusion

The Fermi level of the single wall carbon nanotube can be shifted upon applying an

electrical field inside an optical analogy nanotube based FET has been confirmed. The Fermi









level shifting effect was confirmed through monitoring the transmittance intensity change of the

nanotube film at certain wavelength which corresponding to the electronic transition of the

SWNT while applying a gate voltage. Since spectral changes are observed from the Ml band

associated with a depletion of the first valence band of the metallic nanotubes in the sample (see

the DOS for the (10, 10) nanotube shown in the inset of Figure 2-3) the Fermi level of the

nanotube can be shifted by about 0.9eV by applying a gate voltage less than 2V. The near IR

transmittance decrease was corresponding to the free carrier absorption was also observed as the

carrier concentration increase in the SWNT film. The film resistivity as a function of the gate

voltage showed the SWNT was initially p-doped.








ITO (200oA)


Glas s







ITr115111ittedl light


SWcNT fihn


\Au


Figure 3-1 The schematic drawing of a solid state optical modulator


Reprint with permission from Z.H. Chen, Thesis, University of Florida, 2003


Al0, (810A1)











sapph1i re
subjstrate


sample
beamn


IL, reservoir .'/ | lass
.4- plate
rubber
gasket



Figure 3-2. A sketch of the optical analog to the electrolyte-gated NFETs















SReproduced in part with permission from [Z.Wu et.al, Science 305, 1273-1276 ] Copyright [2004] AAAS


palladiumn
Metal


fillms












100



Counter
electrode
votage
r~ -0.5V
? OV
Er i!+0.4V






500 1000 1500 2000 2500 3000

Wavelength (nm)


Figure 3-3. Three transmittance spectrum as a function of applied counter electrode voltage, i.e.
OV. +0.4V and -0.5V.










r 1 1 j I I


S 2


i,-, 2 t


( 12,8)


0.6


0.4


-


SI


0.2 -





0,0-


-0.5


1.0


-1.0


0.0


0 5


1.5


Energy (eV)



Figure 3-4. Density of states (DOS) of (12, 8) single walled carbon nanotube9
























9 Reproduced in part with permission from [Z.Wu et.al, Science 305, 1273-1276 ] Copyright [2004] AAAS


I
I
r
I
r
r
r
I
r
r
r
I


I


























aeI I I I I Id~
a 50 100 150 200 250 300
WIlrthnr

Figure 3-5. Th rnmtac ftenntb ma ucionofte plid ae otae
fro +.8 t -.8V Te rasmttaceatS1pek (67nm icrase fom44 t
92%l whl h 2pek(3n)vrie frm 1 t 8%.TeI t38n
decreased~~~~~~~~~~f frm9%t 5,atopst ietint h 1pakmdlto.1











"'Rpoucdi ar ihpemsin rm[.W ta, cec 305 123 127 ]CoY rih 20]A A





Ferrni igel grted


Forml level gules
timerrd gap


curwrelectrode Vahgee

Figure 3-6. The resistivity of the nanotube film as a function of the applied gate voltagell



































" Reproduced in part with permission from [Z.Wu et.al, Science 305, 1273-1276 ] Copyright [2004] AAAS













as S1






as-i M1i


4 -1r 0 1 2
Cournier-elechede8 UMana

Figure 3-7. The transmittance at S1, S2 and Ml depends on the gate voltage. 12


































'2Reproduced in part with permission from [Z.Wu et.al, Science 305, 1273-1276 ] Copyright [2004] AAAS
























I86 Iru e~r
0 100 200 300 400
time (s)

Figure 3-8. Time drive of the S1 transmittance peak with +/- 1V gate voltage.










CHAPTER 4
OHMIC CONTACT COUPLINTG CARBON NANOTUBE FILM TO P-GALLIUM NITRIDE

We have demonstrated that the SWNT fi1m is a new type of conducting, transparent, Fermi

level shiftable electrode material. These features of the SWN\T fi1m lead us to test their ability to

incorporate into optoelectronic devices as transparent electrodes through which the charges

transferred across the interface of the nanotube fi1m and the semiconductors. It is well known

that when two distinct materials come into intimate contact with each other, there will be energy

barriers across their interface. Such energy barriers possess a height and a width (depletion

width, into both side of the physical interface). To Birst order, in the Schottky-Mott model the

height will be determined by the energy band alignment of the two materials while the width of

the associated depletion region on each each side will be governed by the height of the barrier

and the carrier density of that material. For metal-metal contact, since the electron density is so

high, the barrier will be negligible because the barrier width is very thin at both sides to tunnel

through. In metal semiconductor contacts in contrast the carrier concentration of the

semiconductor is so much lower compare to that of the metal, that the depletion width on the

semiconductor side can be very wide (up to microns), resulting in an insurmountable barrier to

charge transport across the junction. Schottky barriers can be useful for example as in Schottky

diodes as well as Schottky FETs, But for most semiconductor devices, an Ohmic contact (with

negligible barrier height and/or width) is preferred since the efficiency of charge transfer across

the metal semiconductor interface is of priority. Ohmic contact causes less energy to be

dissipated at the junction in terms of charge inj section or extraction.

Wide band gap semiconductors, such as GaN are very useful in making visible LED as

well as high temperature operation applications. The band gap of the semiconductor is wide such

that the p-band of the semiconductor is far away (greater than -7eV) from the vacuum level. The









work functions of most metals are less than 5.5eV, which makes it difficult to make ohmic

contact with the p-band, resulting in a barrier for charge transfer into the p-band. There is a need

for a p-type transparent electrode which can ohmically couple to p-GaN. In this chapter, a GaN

LED device was demonstrated by using SWNT film to couple to the p-GaN side of the LED. The

contact resistance of the nanotube film to p-GaN versus the conventional metal contact to the p-

GaN was also compared.

4.1 GaN Light Emitting Diode Background

Gallium nitride (GaN) is a wide band semiconductor with a 3.4eV band gap and 4.1eV

electron affinity. This means the valence band lies at 7.5eV. Since the p-GaN is p doped it's

Fermi level lies near this valence band edge meaning that Ohmic contact requires a metal with a

work function approaching the valence band edge. Most metals however have a work function

less than 5.5eV. This remains a technical challenge for wide band gap semiconductors. It is well

known that if the contact barrier between metal and semiconductor is high, it requires higher

voltage to overcome the high contact barrier, resulting in overheating, low efficiency and

electromigration. An approach to make ohmic contact to semiconductors is by heavily doping a

thin buffer layer right at the interface to reduce the barrier width, which will enhance the

probability for charges tunneling through the barrier. This additional layer however adds cost to

the device fabrication process.

A wide variety of metals and alloys have been explored for making low contact barrier and

therefore low contact resistance contact to p-GaN. These include Ni/Au, Ni, Au, Pd, Pd/Au,

Pt/Au, Au/Mg/Au, Au/C/Ni, Ni/Cr/Au and Pd/Pt/Au (26-28). Typically Ni, Pt, or Pd is the metal

in direct contact to p-GaN and the structure will be annealed at 400-750oC. This will produce a

contact resistance in the range of 10- -10-3 CM 112. Higher annealing temperatures can degrade

metal contacts, usually because they will react with GaN to form metal gallides.









In order to observe the light produced in the LED structure, transparent conducting

contacts are needed. Indium tin oxide (ITO) has been a candidate however, ITO shows rectifying

behavior on p-GaN, and is more commonly used as the n-GaN contact.

One of the failure modes for the electronic device is the electro-migration of the metal

contacts into the semiconductor. Typically, it requires 1-3eV to remove one metal atom from the

bulk, compared to 7eV to remove one carbon atom from nanotubes tightly bound lattice.

Furthermore, the entire carbon nanotube is simply too large to migrate. Thus, with SWNT fi1m as

the electrode, there will be no electro-migration problem.

From chapter 2, we already learned that the thin nanotube film is transparent all the way

from UV to infrared. The GaN LED will emit light at 434nm, where the transmittance for a

100nm thick carbon nanotube fi1m is ~60% (thinner film will be more transmissive but at the

price of lower conductance).

4.2 Experiments Details for Fabricating SWNT Film Based GaN LED

This research was carried out with collaboration with Dr. Lee, Dr. Ren and Dr. Pearton

from material science and engineering department, University of Florida. First, metal organic

chemical vapor deposition (MOCVD) method was used to grow n-GaN followed with p-GaN on

c-plane sapphire substrate. A 100nm thick carbon nanotube fi1m was deposited on top of p-GaN,

covering the entire surface of the sample. Once the nanotube film was attached to the p-GaN, the

assembly was first dried and annealed at 600oC for 6 hours in argon to remove any residual

cellulous ester. Thermally deposited Ti/Al/Pt/Au, Pd/Au, or Pd was patterned by a standard e-

beam lift-off process and a mesa was formed by Cl2/Ar inductively coupled plasma ethching to

expose the n-GaN side of the device. The nanotube film was simultaneously patterned by this

standard e-beam lithography process. Annealing of the device was carried out under nitrogen









environment for one minute, at 700oC. The whole structure of the diode is illustrated in Figure 4-



4.3 Results

4.3.1 Contact Resistance

In order to investigate the contact resistance of the carbon nanotube fi1m/ p-GaN interface,

the carbon nanotube fi1m was deposited on single layer p-GaN on sapphire substrate. The

transmission line method (TLM) was used to determine the contact resistance of the nanotube

fi1m/p-GaN contact resistance (29). The control devices of the conventional Ni/Au metal

contacts on p-GaN were also fabricated. Upon fabrication, the conventional metal contact is non-

Ohmic, while the as deposited carbon nanotube fi1m contact was Ohmic with a resistance of 0. 12

Ohm-cm2. After 500oC annealing in nitrogen for one minute of rapid thermal anneal (RTA), the

resistance of the conventional metal became 0.033 Ohm cm2, while for the carbon nanotube film

structure after a 700oC RTA the contact resistance was 0.0110hm cm2, a factor of 3 smaller than

that of the conventional metal. This data is summarized in Table 4. 1. The annealing temperature

was optimized for least contact resistance for the different devices. It is worth while to note that

the carbon nanotube fi1m itself is very conducting. There is another layer of metal film at the

corner of the nanotube film, which makes the electrical connection to the device, as illustrated in

Figure 4-1. The contact resistance between the metal film and carbon nanotube is negligible, two

orders of magnitude smaller (after annealing) compare to the contact resistance between

nanotube film and p-GaN. The measurement of the contact resistance of as deposited metal

Ti/Al/Pt/Au on the carbon nanotube is 5.3X10-3 ~ZCM2 reducing to 1.3X10-4 ~ZCM2 after the

700oC RTA.

The lowest contact resistance of the conventional metal was achieved at the annealing

temperature of 500oC, while the lowest contact resistance of the nanotube film/p-GaN was










achieved at 700oC, The latter temperature was tried with the metal contact but at 700oC, the

conventional metal was severely degraded through the reaction of the metal to the GaN. Thus,

excellent thermal stability is another advantage of the nanotube film, an important feature in

order to achieve the lowest contact resistance which also adds simplicity to the whole device

fabrication since it can tolerate higher temperature required for n-GaN patterning and RTA

treatment.

4.3.2 I-V Characteristic of Carbon Nanotube Film/p-GaN

In order to investigate whether different metal contacts to the nanotube fi1m will change

the overall LED performance, we measured I-V characteristics of the junction with different

metal contact pads. In all cases, the junctions showed rectifying effects, as expected. They all

show similar forward and reverse current. However as shown in Figure 4-2 there were subtle

differences depending on the metal used, likely due to a Fermi level equilibration between the

nanotubes and the distinct metals used.

The GaN LED emitted blue light at 434nm, as shown in the spectrum of Figure 4-3 and a

picture of the actual device glowing light in Figure 4-4.

4.4 Discussion

First of all, why will the RTA process improve the contact in both cases? For the metal

semiconductor interface, contact resistance improvement upon rapid thermal annealing results

from the surface atoms will gaining enough energy at high temperature to rearrange themselves

and lower the overall surface energy. On the other hand, for nanotube fi1m/p-GaN, it is very

likely that upon rapid thermal annealing, the surface atoms of the GaN rearrange themselves and

more carbon nanotube come in intimate contact to the p-GaN, so that the actual contact area

between nanotube fi1m and p-GaN increases, hence lowering the contact resistance.









Secondly, the Fermi level of the carbon nanotube lies at about 5eV, which is 2.5eV away

from 7.5eV, where the p-GaN valence band sits which should result in a large barrier. Hence the

question becomes why do nanotube films provide a good ohmic contact p-GaN? The most likely

reason is that the carbon nanotubes posses a low density of states permitting large shifts in their

Fermi-level. As shown in Chapter 3 under relatively low voltages (1.8 V) the nanotube Fermi

level could be shifted 0.9eV. Upon applying an appropriate voltage, the Fermi-level shift lowers

the contact barrier for hole inj section into the p-GaN. Note that the nanotubes provide a very

unconventional metal in this respect. For conventional metals, which have three dimensional

density of states, and very high carrier density, the shift of the Fermi level at low biased voltages

is negligible.

4.5 Conclusion

Low contact resistance was achieved by using single walled carbon nanotube films coupled

to the p-GaN. The SWNT film's thermal stability simplifies the LED manufacturing process by

allowing the n-GaN RTA process with the SWNT film in place. The contact resistance is 3 times

smaller than the conventional metal after 700oC RTA in N2 for 1min. This SWNT film provides a

new class of p electrodes to all the high work function p type semiconductors. The tight bonding

of the carbon atoms in the nanotube implies that the principle cause of lifetime degradation in

these devices: electromigration, will simply not occur when nanotubes are used for electrical

contact to the devices.

The mechanism we propose for SWNTs making ohmic contact to high work function

semiconductors is due to the limited density of states of the one dimensional carbon nanotubes.

This enables one to shift the Fermi level up by several electron volts upon applying a small

voltage, to accommodate the large p-band of the semiconductor. The side wall of the nanotube









prevents any covalent bonding between nanotube to the semiconductor, which allows this Fermi

level shifting to happen without so called Fermi level pinning.









Table 4-1. Contact resistances of Ti/Al/Pt/Au on carbon nanotubes, carbon nanotube film on p-
GaN, and standard Ni/Au on p-GaN13
Specific contact resistance (0Z cm2)
Structure
700oC, N2, Imin annealing As prepared (no annealing)

Ti/Al/Pt/Au on carbon
1.31X10-4 5.4X10-3
nanotube film

Carbon nanotube film/p-GaN 0.011 0.12

Ni/Au/p-GaN 0.033 after 500oC annealing Non-Ohmic


13 Reproduced in part with permission from [Lee, K. et.al, Nano Lett. 4(5), 911-914] Copyright
[2004] American Chemical Society











































3.4


,, 7.5


14 Reproduced in part with permission from [Lee, K. et.al, Nano Lett. 4(5), 911-914] Copyright
[2004] American Chemical Society


p+ carbona
nanotube


Metal


GaN


Ti/Al/Pt/Au


SWNT


Figure 4-1. Schematic view of the GaN
Ohmic contact. 14


based light-emitting diode using SWNT film as the p-










0.15


-i 0.10 TAU,~
-ePdCAu



~ 0.05




-0.10
-8 -4 O 4
Bias (V}

Figure 4-2. IV characteristics of GaN LED with different metal, i.e. Ti/Al/Pt/Au, Pd only and
Pd/Au contact on carbon nanotube fim. All the devices are using carbon nanotube
film as the p-GaN contact. 1






































"5 Reproduced in part with permission from [Lee, K. et.al, Nano Lett. 4(5), 911-914] Copyright [2004]
American Chemical Society






















300 400 _500 600
\rc've Law16l lmil


1.2

~1.11
r3

~08
Oci
O.J


Figure 4-3. Emission spectrum of the GaN LED with inj section current 0.1lmAl6








































16 Reproduced in part with permission from [Lee, K. et.al, Nano Lett. 4(5), 911-914] Copyright [2004]
American Chemical Society






























Figure 4-4. Picture of the visible emission from the GaN LED with carbon nanotube film as the
p-GaN contact electrodes. 1




























"7 Reproduced in part with permission from [Lee, K. et.al, Nano Lett. 4(5), 911-914] Copyright [2004]
American Chemical Society









CHAPTER 5
ORGANIC LIGHT EMITTINTG DIODE WITH SWNT FILM AS ANODE

5.1 Introduction

We have demonstrated the ability for the SWNT film to ohmically couple with inorganic

semiconductor GaN. It is natural to test if SWN\T film will be a good candidate for coupling to

organic semiconductors, since organic optoelectronic devices have drawn considerable attention

due to their ease of process and low cost (30-33).

Indium Tin Oxide (ITO) is presently the transparent electrode used in most OLED devices.

However as a rigid oxide material ITO has limited flexibility before it breaks hence it excludes

the possibility of flexible/foldable displays. Carbon nanotube films in contrast (deposited on

flexible substrates, such as Polyethylene terephthalate, or PET) can be bent indefinitely without

damage making them excellent candidates for flexible display applications. Additionally the

SWNTs possess an extreme chemical tolerance making them compatible with virtually any

process step that might be required in OLED fabrication.

5.2 MEH-PPV

In order to test if SWN\Ts integrate well with organic light emitting devices, it is wise to

choose a widely used organic material to start with. After consulting with Dr. Reynolds from

Chemistry Department of University of Florida, Poly[2-methoxy-5 -(2'-ethyl-hexyloxy)- 1,4-

phenylene vinylene] (MEH-PPV) was chosen as the candidate polymer in our first trial OLED

devices. Because MEH-PPV has been fully studied, it is easy to process, commercially available,

air stable and its electroluminescent emission is in the visible range (32).

The structure of MEH-PPV is shown in Figure 5-1, along with the illustration of the layer

by layer OLED structure. The MEH PPV initially used for this study was synthesized by Dr.









Reynolds' group. Later we also used a batch purchased from Aldrich. This work was done in

close collaboration with Dr. Jeremiah Mwaura from Dr. Reynolds' group.

5.3 Experimental Details

MEH-PPV was dissolved with dichloroethene over night at room temperature, with a

concentration of 5mg/ml. The solution was constantly stirred with a stir bar to help fully dissolve

into the polymer. Before using the dissolved polymer solution, it was filtered with a syringe filter

of 5Cpm pore size to remove un-dissolved polymer and large impurity particles. After filtering,

the solution was spin casting onto nanotube film at 1000rpm for 45second, which resulted in a

film with thickness in the range of 60-100nm. The MEH PPV film thickness was estimated based

on the thickness of the same polymer concentration prepared under the same conditions on an

ITO substrate. The nanotube film surface is much rougher compare to that of ITO/PEDOT PSS,

which made the polymer film thickness measurement difficult. The MEH PPV film thickness on

SWNT will be discussed further when we get to the wetting issue of MEH PPV on SWNT.

A SWNT film with a 50nm thickness was deposited onto a one inch square microscope

glass slide substrate. All the substrates were cleaned with acetone and methanol and then rinsed

with DI water prior to deposition of SWNT films. At the same time, control devices with ITO as

the anode were fabricated side by side as the SWNT film based devices. ITO coated glass was

plasma cleaned and spin coated with 100nm of PEDOT PSS as the buffer layer, baked at 150oC

in vacuum oven for 2 hours to remove excess water from the PEDOT PSS layer.

After spin coating the MEH-PPV, a cathode consisting of 10nm of calcium and 150nm of

aluminum was thermally evaporated onto the MEH-PPV through a shadow mask. Before the

evaporation of the cathode, the sample was vacuum de gassed in the vacuum (2x 10 7torr)

deposition chamber over night to allow the solvent to fully evaporate. The area of the cathode

(through the shadow mask) defines the area of the working OLED. The illustration of the OLED









is shown in Figure 5-1. Once the device was taken out of the vacuum chamber it was

immediately coated using minute epoxy to prevent oxidation of the calcium electrode greatly

shortening the device life time.

5.4 Discussion



Since the working polymer is on the order of 100nm thick, the cleanliness of the substrate

and the polymer solution are crucial to the successful functioning of the devices. Dust particles

are usually on the order of a micrometer. We fabricated 8 pixels on one glass substrate, only one

out of 8 pixels worked reasonably well, but with a dimmer glow than the ITO control device.

The only working pixel was also nonuniformly illuminated, visible to the naked eye when it was

turned on. The control ITO devices were much better: bright with all devices working. The fact

that there is one pixel working meant that the energy band alignment of the carbon nanotube fi1m

with the MEHPPV was not the maj or problem. The nonuniformity in the light of the working

pixel indicates there may be intrinsic nonuniformity to start with in the nanotube fi1m. This was

confirmed by the fact that the current is higher than that of the ITO device under the same bias

voltage but with lower light emission.

We performed transmission electron microscopy (TEM, JOEL 2010F) on the SWNT fi1ms

finding large micron scale particles. Furthermore using the energy dispersive X-ray spectrometer

(EDS) to analyze the chemical component of the particles, it was found that these were

principally made of principally calcium and magnesium. There were also smalled particulates

made of cobalt and nickel, the catalysts used in the growth of the SWNTs. The calcium and

magnesium contamination was initially mysterious until it was recognized that these were most

likely dust particles from the laboratory cement walls and floor, since these SWNT fi1ms were

prepared in the ambient laboratory environment. Additional cross flow filtration was done to the









stock nanotube solution to remove more of the residual catalyst particles. To avoid the dust

particles, a class 100 clean room was constructed in the laboratory for manufacture of the SWNT

films.

Once the SWNT films became cleaner, the fraction of nanotube based OLEDs that emitted

any light was greatly increased (~5 out of 8 pixels vs. less than 1 out of 8 previously) though the

performance of most of the nanotube devices was poor. The performance of a typical

ITO/PEDOT PSS OLED device current and radiance as a function of the applied voltage is

shown in Figure 5-2 (these did not vary much from device to device). The performance of one of

the best performing SWNT film based OLEDs is shown in Figure 5-3. Note that under the same

applied voltage, the radiance of the SWNT film based device was roughly one order of

magnitude less than the ITO/PEDOT PSS based device. A possible explanation for this poor

performance of the SWNT based devices is that while the steps taken above to improve the

purity of the films helped, there were still particles observed in the SWN\T films. While not

generating dead shorts completely killing the maj or fraction of the devices these particles

generated high current regions that drew down the local potential across the other regions like a

voltage divider circuit. Alternatively or perhaps coinciding with this is the possibility that the

nanotubes were too efficient at hole inj section into the devices. The resulting electron hole

imbalance then responsible for the poorer performance. The energy band alignment of the OLED

system is shown in Figure 5-4. We can see that the alignment of the PEDOT PSS to the p-band

of the IVEH-PPV is better than that of the SWNT film. But since we have demonstrated the

ability of shifting the Fermi level of SWNT film, this mismatch in the energy band only requires

slightly higher applied voltage. On the other hand the most efficient OLEDs should have

balanced electrons and holes inj ected from both electrodes and recombine in the middle of the









electroluminescent polymer layer away from the luminescence quenching anode and cathode.

Most polymers, including MEH PPV, have much better hole mobility than electron mobility

(3 5). If we have indeed improved the hole inj section from SWNT, say, much better than the

electron inj section, that will result in luminescence quenching at the cathode.

5.4.1 MEH PPV does not Wet with SWNT

Further study revealed that the MEH PPV does not wet the SWNT surface. In order to take

advantage of the high surface area of the SWNT film, MEH-PPV was diluted to roughly 1/10 of

the original concentration. More over, the diluted MEH PPV solution was soaked for 30 seconds

to allow better penetration of the solution into the nanotube network, since the pore size of the

nanotube film was on the order of tens of nanometer. The spin casting procedure was repeated

for 10 times to compensate the diluted solution in order to obtain same polymer film thickness.

After spin casting, the MEH-PPV film looks nonuniform in color to the naked eye and the device

performance was poor.

In order to find out what was the cause of this nonuniformity, a regular concentration

solution was used and one drop of MEH-PPV solution was put on top of SWNT film. This

should gave a much thicker film since during the spin casting process, the maj ority of the

solution will be spin off the substrate. When the solvent began to evaporate, the wetted drop area

became smaller, leaving behind no sign of MEH-PPV coated on SWNT film from regions where

the drop had wetted the film. Since the MEH PPV has orange color, it can be easily identified if

there is a coating on the nanotube film. When the solvent had fully evaporated, the MEH-PPV

had all accumulated at the center of the drop with a much smaller area than the initial MEH-PPV

solution drop area. This leads to the conclusion that MEH PPV polymer itself has little affinity

for the carbon nanotube surface. Poor wet ability will result in less contact area between the two,

causing charge inj section problems.









5.5 Conclusion

SWNT film as anode for making MEH-PPV organic LED was successfully manufactured

and compared with ITO/PEDOT PSS based devices. The radiance of the nanotube based device

was one order of magnitude lower than the ITO/PEDOT PSS based device. The possible reason

of wet ability of SWNT film to MEH PPV has been suggested and further work is necessary to

determine the cause of the poor performance.



























Al
~ Ca

MEH PPV


Al
Ca

*** MEH PPV
SWN T

| +******. Glass

Figure 5-1. A) Structure of 1VEH-PPV (top) and B) Illustration of the OLED device with
ITO/PEDOT PSS as anode (middle) C) OLED device with SWNT film (bottom)
































Figure 5-2. ITO/PEDOT PSS based OLED radiance and current vs. voltage


_


I


600






S300-

E


20


-3mm .7d

/ /
/ n


-5000


Voltag~e (V)


5 6











1200

1000 -d -200

E.0 ,O 150 O


c ,"P-100
E 400

200 7 50



S1 2 3 4 5 6 7
Voltage (V]





Figure 5-3. SWNT film based OLED radiance and current vs. voltage










Vacuum level


GI~o Al(4.3)



5.3

5.2 e


ITO /
SWNT PETDOT
shiftable PSS MEH-PPV


Figure 5-4. Illustration of the energy band alignment of the OLED devices.









CHAPTER 6
CARBON NANOTUBE/SILICON HETEROJUNCTION

6.1 Introduction about Schottky Barrier

The metal semiconductor heterojunction plays an important role for all semiconductor

devices. The transport across the metal semiconductor junction is determined by the contact

barrier height. For ideal situation, without surface states, the barrier height is given by the

difference of the metal work function and the electron affinity of the semiconductor (for n-type

semiconductor), this is called the Schottky-Mott limit (37). The flat band model illustrates the

contact barrier height in Figure 6-1. When the metal and semiconductor come into intimate

contact, the Fermi level across the interface will be lined up. Since the semiconductor has higher

electron affinity in Figure 6-1, electrons will flow from semiconductor to the metal until thermo

dynamic equilibrium was achieved. The Fermi level in the semiconductor side will be lowered

by an amount equal to the difference of their work function. The negative charge transferred

from the semiconductor will build up in the metal side and the equal amount but opposite sign of

the charges will be built up in the semiconductor side. Since the density of electronic states of the

semiconductor usually is lower compare to that of the metal, this positive charge will be

distributed over a region called depletion region (since oppositely charged maj ority carriers are

neutralized there). The width of this depletion region will be determined by the doping level of

the semiconductor and the amount of charge transferred. A small barrier height and narrow

depletion region results in what is called Ohmic contact. A large barrier height and large

depletion region will give rectifying behavior. For simple electrical contact purposes, an Ohmic

contact is preferable for better transport across the junction. A contact barrier in contrast will

consume power and generate heat. As the device size shrinks and device density increases in

integrated circuits, heat dissipation becomes a problem.









According to experimental measurements, flat band model usually does not work. Most

metals will form covalent bonds with semiconductor surfaces. This covalent bonding will add

intermediate states in the band gap of the semiconductor. This is the reason why the contact

barrier height is not sensitive to the work function difference between metals, a phenomena

called Fermi level pinning (38-40), widely observed in semiconductor/metal contacts.

Given the demonstrated ability to modify the Fermi level in chapter 3, the idea described

here was to form a sort of transistor based on modulating the Schottky barrier contact between

SWNTs and a doped semiconductor (silicon). Figure 6-2 illustrates the relevant features needed

for such a device, in which there is a SWNT/Si interface (called the Source), an Al/Silicon ohmic

contact (serving as the Drain), and a strip of Pd thin film sitting on top of SiO2 that will act as the

gate contact to an electrolytic gate in electrolytic contact with the nanotube film/Si surface. Since

the Schottky barrier depends on the Fermi level offset, if one modifies the Fermi level, the

current passing through the junction will be modulated as well.

6.2 Experimental Details of SWNT/p-Si Heterojunction

Lightly (with resistivity 2-10 Ohm cm) and heavily (<0.005 Ohm cm) Boron-doped silicon

with thermally grown 250nm oxide was purchased from Silicon Quest International and diced

into 20 x30mm2 pieces. The surface of the silicon substrate was cleaned with Acetone, methanol

alcohol and DI water. To construct the device shown in Figure 6-2, a thick layer of photoresist

was painted with one window of 2x4mm2 and 20x 7mm2 Strip left unpainted, those area later

will be etched with hydrofluoric acid (HF), as explained in Figure 6-3 a). The fresh etched

silicon was rinsed with DI water and blow dry with nitrogen and immediately put into a high

vacuum chamber for thermally evaporating a 100nm aluminum (Al), to form an Ohmic contact

with the silicon, as explained in Figure 6-3b). The Al film was split into two in order to check the

contact resistance between Al and silicon. The resistance varies from several ohms to tens of









Ohms depending on the doping level of the silicon. It is considered Ohmic contact due to the

large contact area (~100mm2) between the Al and silicon. Also shown in Figure 6-3b) is the two

Pd film pads thermally evaporated through a shadow mask, with one pad next to the silicon

window, being careful to avoid any direct contact with the bare silicon. The other Pd bar is

parallel to the first one, and will be served as the gate electrode.

A 50nm thick nanotube film (3*10mm) was transferred across onto the exposed silicon

window, with 2mm overlap the Palladium bar very next to this Si window, as illustrated in

Figure 6-3c). A small well (not shown on the figure) made of poster putty was built around the

nanotube film and the gate electrode, and several drops of ionic liquid, EMI-BTI were put into

this well to make the source and gate electrode both immersed inside the ionic liquids. The

electrical connections to all three electrodes (Source, Drain and Gate) are connected through the

gold wire pressed with fresh cut indium dot onto the corresponding metal films. The circuit

diagram is illustrated in Figure 6-4, where a source drain voltage is applied. We keep the source

drain voltage as a constance vary the gate voltage to see the current change as a function as the

gate voltage.

6.3 Results of SWNT/p-Si Heterojunction and Discussion

The I-V characteristic as a function of gate voltage for both highly doped and lightly doped

silicon was measured. The source drain voltage held constantly at 0.1V, by changing the gate

voltage from -0.7V to +0.7V, the source drain current was recorded. For lightly doped silicon (2-

10 Ohm cm), the gating effect is obvious. When the gate voltage was varied from -0.7V to

+0.7V, the source drain current dropped from 128CLA to 3.5CLA, a factor of 36 times smaller, as

shown in Figure 6-5. While the heavily doped silicon device source drain current reduced from

256CLA to 25CLA, a factor of 10 times change, less than the lightly doped silicon device.









In Figure 6-6, the gate voltage was held at a constant value while measuring the source

drain current as a function of the source drain voltage, both for lightly doped silicon (top) and the

heavily doped silicon (bottom). When the gate voltage is zero, the IV curve showed rectifying

characteristic, indicating there is a barrier at the carbon nanotube silicon interface. When we

apply positive gate voltage, the transport gets reduced. When negative gate voltage is applied,

the current increased, both in the forward and reverse direction. The increase in the forward

direction is much more obvious. When the gate voltage increased to negative 0.7V, the reverse

direction current gets significantly increased.

The gating effect is explained as the following two reasons. First let' s explain from the

energy diagram, as illustrated in Figure 6-7. When the gate voltage is zero, there is a built in

potential barrier right at the SWNT/p-Si junction because of the difference in the Fermi level of

the SWNT and the silicon. Upon applying negative gate voltage, the source is relatively positive,

move the Fermi level of the SWNT downward on the Figure 6-7 energy diagram. This will make

the energy alignment between carbon nanotube and silicon valence band better, or in other

words, reduce the barrier at the interface. The second reason comes from the carrier

concentration change upon application of gate voltage. When the negative gate voltage was

applied, the source electrode was positive relative to the gate electrode. It draws negative ions in

the ionic liquid to the vicinity of the carbon nanotube. Since the carbon nanotube is in intimate

contact with the silicon surface, that excess amount of negative charge near the silicon surface

will increase the positive carrier concentration inside silicon locally, which will result in a

thinner barrier region for better inj ecting charge from the nanotube to silicon and vice versa. The

positive gate voltage will do the reverse to the negative gate voltage.










After continuously applying the gate voltage for 5 hours, the source drain current reduced

dramatically, as shown in Figure 6-8. This might because when the gate voltage is applied, the

silicon surface will attract ionic liquid and those ions will lift the nanotube from intimate contact

with the silicon surface, hence reducing the effective contact surface between nanotube and

silicon.

Double gating structure (also being illustrated in Figure 6-2) was also explored, in which

case another SWNT/p-Si junction replaced the Si/Al drain. The fabrication process is very

similar to that of the single gate device. This is a symmetric device, with the gate on source and

drain can be independently modulated. When two identical gate voltages were applied for source

gate and drain gate, the source drain voltage was kept at constant of 0. 1V. The source drain

current can be modulated more than 300 times when the gate voltage changed from positive 0.7V

to negative 0.7V, as shown in Figure 6-10. When the source junction and the drain junction

working simultaneously to be open or close so that the source drain current on/off ratio is much

bigger than the single gate device, which is 36 times modulation for the lightly doped device. In

contrast, the single junction device will have the drain (Al/Si ohmic contact) always turned on.

Finally we studied the ambient environment influence on the transport across the carbon

nanotube silicon heterojunction. According to our observation, the current across the nanotube

film/ silicon junction will stabilize when the sample was left in any of the following environment

more than half an hour. We have tested air, oxygen, argon or under Vacuum. But upon ambient

environment change, the current will have a sudden change, as illustrated in Figure 6-10 and

Figure 6-11. In Figure 6-10, it showed the current suddenly reduces when the device is stabilized

in argon upon exposure to oxygen. While in Figure 6-11, when the device was stabilized in argon

with a stabilized current, a bigger drop in current will occurred immediately upon exposure to










oxygen. This can be explained by the oxygen playing an important role in the charge transfer

process between the interface of carbon nanotube and silicon. It is very likely water vapor in air

also contribute to the charge inj section across the interface (42).

6.4 SWNT/n-Si Photovoltaic Device

We just demonstrated the ability to electrochemically shift the Fermi level of the SWNT

by applying a gate voltage relative to the SWNT with ionic liquid as the electrolyte. We can

modulate the current across the SWNT/p-Si interface. On the other hand, if we investigate the

SWNT/n-Si interface, there will be a bigger barrier compare to SWNT/p-Si, according to the flat

band model. Since the SWNT fi1ms are transparent, this barrier will be useful for photovoltaic

application. This is a Schottky barrier type photovoltaic device, since the SWNT fi1m can be

regarded as a conducting film.

Because the Fermi level in the n-Si is higher than that of the SWNT, silicon will donate

electrons to SWNT to reach an equilibrium state. As a result, there will be a depletion region in

silicon near the interface. As we know, when the light incident onto that interface, the photon

will be absorbed by the semiconductor and the energy will be transferred as the exciton--the

electron hole pair. This light absorption will mainly happen in the silicon depletion region. The

depletion region is extended from SWNT side to the silicon side. The width of the depletion

region at each side was determined by the carrier concentrations of the SWNT and the n-Si. The

product of the two (carrier concentration and the depletion width) at SWNT and n-Si side will be

equal to each other. The built in electric field inside the depletion region will separate the

electron and hole, driving the holes goes toward the SWNT side and electrons goes to the silicon,

then passing through the ohmic contact, going out to the external circuit.









6.5 Experimental Details for SWNT/n-Photovoltaic Device

The n-Si (Phosphorus) wafer with doping level 1014 (4-20 0Z cm) was purchased from

Silicon quest international. wafer is along <111> direction and has one micron meter thermal

grown oxide on top. The silicon wafer was diced into 1 by 1 and half inch substrate. A one by

two millimeter window was etched on the oxide to expose the silicon. The back side of the chip

was being etched and thermal evaporate with 70nm Al. The resistance between the Al and Si was

measured in the order of 1002. A 50nm thick SWNT film (1.5 by 3mm) was transferred onto the

silicon window, followed by thermal evaporate Pd film to make electrical contact with SWNTs.

A second Pd bar was also deposit parallel to the first bar, to be the gate electrode. The whole

structure of the device was shown in Figure 1 1. The SWNT was the cathode (also being called

source in this chapter) and the Al ohmic contact (being called drain) is the anode of this

photovoltaic cell. With extra gate electrode sitting next to the cathode, but was isolated from

silicon by the SiO2. We will explain the gate electrode later.

6.6 SWNT/n-Si Solar Cell Results and Discussion

The solar simulator was calibrated such that the light output was 100mW/cm2 at the

position where the sample sits. The open circuit voltage is 0.51V and the close circuit current is

1.31ImA for the 2mm2 Solar cell under the AM1.5 solar simulator (100mW/cm2). The efficiency

of the solar cell was 2.53% and the filling factor is 0. 15. The IV curve for the solar cell was

shown in Figure 12. As we can see from Figure 13, the series resistance was calculated to be 283

ohms for the low doping PV cell, which is the maj or contribution for the low filling factor.

Typical series resistance of a silicon p-n solar cell is less than 1 ohm (43). The series resistance is

mainly due to the poor contact between the Aluminum and the gold wire, we use indium dot to









make connection between the gold wire and the aluminum back contact. Improvement for

reducing series resistance is under way, including using higher doping n-Si.

As we know, the Fermi level of the nanotube can be readily shifted. Since the open circuit

voltage of the solar cell was depend on the built in potential and that was determined by the

Fermi level difference between the metal and the semiconductor. As we have demonstrated in

the SWNT/p-Si interface, we can modulate the transport across the interface by applying a gate

voltage. We use the same idea to test the gate Hield on the solar cell.

As we illustrated in the layout of the device, there is a gate electrode sitting parallel to the

source electrode (the SWNT fi1m). The gate electrode was isolated from the silicon by the oxide

layer underneath it. We put a drop of ionic liquid EMI-BTI on top of the SWNT and gate

el ectrode.

We apply a gate Hield between the gate electrode and the drain electrode. This gate Hield is

provided with a battery, with floating ground. When the gate is connected to the drain, without

additional battery, we call that OV. When the gate is applied with a negative potential, we call it

negative gate voltage. In Figure 6-14, we showed the IV characteristic under different gate bias.

As we can see, the positive gate voltage will decrease the solar cell efficiency, both decreasing

the open circuit voltage and the close circuit current. While with a negative gate bias of -0.3V,

the open circuit voltage will slightly increase from 0.52V (when OV gate bias was applied) to

0.548V. When the gate voltage increased to -0.45V, the open circuit voltage remains at 0.548V.

Further increase the gate voltage to -0.6V, the open circuit voltage starts to drop back to 0.528V.

We only tested up to this value because we need to keep the gate bias below the redox voltage. It

is clear that the overall efficiency for this solar cell was best when the gate voltage is -0.45V and









followed by -0.3V and -0.6V, all of which are better than that of the OV. The efficiency under OV

bias is better than that of open circuit.

The explanation is as follow. When the SWNT was in contact with n-Si, the Fermi level of

the SWNT was lie below that of the n-Si, as shown in Figure 6-16. The electron from the silicon

side will donate to the SWNT, forming a depletion region and a built in potential. That forms a

Schottky barrier type solar cell. When we apply a positive gate Hield, the gate electrode will be

positive compare to OV situation. That positive potential on the gate electrode will attract

negative ions from the ionic liquid EMI-BTI, as a result, the SWNT will be surrounded by net

positive ions. The relative potential of the SWNT will be negative, which shifts the Fermi level

of the SWNT up, making the built in potential for the solar cell smaller. On the opposite

situation, when the negative gate voltage was applied, the opposite situation will happen. This

explains why negative 0.3V gate voltage will increase the open circuit voltage of the solar cell.

However, as the Fermi level moves further down when the gate electric field gets more negative,

the Fermi level will move below the p-band of the silicon, which will form a barrier for the holes

to across that barrier. This was illustrated in Figure 6-17, in which the blue line shows this

situation.

6.7 Conclusion on SWNT/p-Si Heterojunction and SWNT/n-Si Solar Cell

This chapter explored the SWNT/Si heterojunction behavior and we investigated the gate

effect of that transportation.

The heterojunction between SWNT and p-Si was explored. The transport can be tuned by a

factor of more than 300 when applied a gate voltage of +0.6V. Such an obvious gating effect was

due to the following two reasons: (1) the limited available electronic states of the SWNT making

the Fermi level shift feasible upon applying a small gate potential; (2) the porosity of the SWNT









film allow ions get access to maj ority of the nanotube which reduces the distance between the

counter electrode, enhanced the gating effect.

The SWNT/n-Si solar cell showed an efficiency of 2.53% and the filling factor was 0. 15.

The IV characteristic was changed when the gate voltage was applied, which agrees well with

what happened in the SWNT/p-Si case. The gate field changed the Fermi level of the SWNT

hence changed the performance of the SWNT/n-Si solar cell.

The variable contact barrier was demonstrated by Lonergan et al in 1997 (41). Lonergan' s

experiment use a hybrid of inorganic-organic, n-Indium Phosphide | poly (pyrrole) | nonaqueous

electrolyte architecture. By applying a potential through the gate electrode, the permeate network

of conjugated polymer allows the electrolyte to access and enable electrochemical tuning of the

contact barrier between the n-Indium Phosphide and poly (pyrrole) for 0.6V. The disadvantage

of this system is that the polymer is air sensitive and thus prone corrosion. So it has to be carried

out in a glove box and by using nonaqueous electrolyte to eliminate the possible corrosion.

Our carbon nanotube films are an air stable, open, conductive network. More over, the

graphene wall of the carbon nanotube is very stable and known not to easily form covalent

bonding. Plus we already demonstrated that the Fermi level of the nanotube is readily shiftable in

previous chapters. All of the above make nanotube films very desirable for studying

heterojunction properties.

















Vacuum

S~*


9X


Ec
EFs




Ev


EFm


Xd Xd


Flat band model Schottky Barrier

Figure 6-1. Flat band model of contact barrier and Schottky barrier height










Drop of ionic liquid
(EMI:BTI)


SWNT
Gate film


Al
Ohmic
Contact


Pd pad


SiO2


Drop of ionic liquid
(EMI:BTI)


SWNT
film


Pd pad


SiO2


Figure 6-2. Illustration of single gate and double gate of carbon nanotube/Si heterojunction











a) Etch two
windows
down to bare
Si, the rest is
SiO2


Sb) Deposit
AlPd and Al
pads






c) Transfer
SWNT
Film and add
a drop of
ionic liquid
coving source
and gate
electrode


Pd






swN
Film



Ionic
Liquid
drop


Figure 6-3. Three steps explaining the SWNT/p-Si single gate heterojunction device fabrication

process


/



















VSG
VSD


Gate
A drop of
ionic liauld












VSG Su i
VSD

Gate
Gate
A drop of
ionic lianid 1




Figure 6-4. Circuit diagram of the single gate device (top) and double gate device (bottom).













250.1 I Heavily doped Si
*Lightly doped Si

200o -I "


'150 -1 .


-a100
e" .
r,

50-1




-0.6 -0.3 0.0 0.3 0.6

va (v)

Figure 6-5. Source drain current as a function of gate voltage for SWNT/Si heterojunction with
heavily and lightly doped p-Silicon. Over a course of +/-0.7V gate voltage, Isd
modified 10 and 36 times, for heavily and lightly doped silicon, respectively.




























-80
-0.30 -0.15 0.00 0.15 0.30

Vsd (V)



900. -- V = 0.7V
-*- V = 0.35V
V=0V
600-
-r- V = 0.35V
V = 0.7V B
h 300-





-03 01 .001 .4
-cVs (V)'


Fiure 6-.I hrceitc fteSN/-iio eeoucina ifrnaevlaeA
Lihl oe -slcnadB eaiydpdpS
















1.12eV


p-Si


4.05eV


Al 4.28eV


SWNT


SWN T


Figure 6-7. Flat band model of modulating the contact barrier of SWNT/Si heterojunction by
shifting the Fermi level of SWNTs.





0.42 0.70


-0.70 -0.42 -0.14 0.14
Vg(V)


Figure 6-8. The current decreased significantly after 5 hour of continuously application of gate
voltage for the lightly doped silicon device. Same phenomena observed for heavily
doped silicon device as well.











1100-



10~ \ Vsd=0.3V


V ,=0.1V





0.1-

-0.6 -0.3 0.0 0.3 0.6

VSG & VDG (V)
Figure 6-9. Double gating effect, with identical source gate and drain gate. Isd can be modulated
more than 300 times by applying gate voltage +0.7V.






































I I


-1000 -

-1100 -

-1200
-3 -


-1300 -

-1400 -

-1500 -
-7 -


-1600 -


back to air "


extra polate
point


Argon


10.5


11.0

time (min)


Figure 6-10. Carbon nanotube silicon junction current in argon environment will change upon
introduce oxygen or air.





















-~ Isd (pA)





-1500




time, condition

Figure 6-11. Carbon nanotube silicon junction transport vs different ambient condition.













A drop of
ionic liauid


SiO2


COntact


Figure 6-12. Circuit diagram of the boot-strap PV device













1.6

1.4

1.2-






0.4


0.2

0.0

-0.2
0.0 0.2 0.4 0.6
Voltage (V)









S-10-



-20-



-30-



-40-


0.0 0.2 0.4 0.6
Voltage (V)



Figure 6-13. I-V characteristic of the SWNT/n-Si solar cell, dark (top) and under illumination
(b ottom).











1.2!


S0.6-





0.3-- I
Linear fit for IV

0.8 0.9 '1.0

V(V)


Figure 6-14. The series resistance was determined to be 283 ohm for low doping n-silicon solar
cell .













0--3




S-6. ~I~+0.6V bias
c / I~+0.45V bias
*ar 1 /I- +0.3V bias

b OV bias
O -Yr -0.3V bias
-0.45V bias
-0.6V bias

0.0 0.2 0.4 0.6i

V(V)

Figure 6-15. IV characteristic for the solar cell under different bias














eY0.05. --ov


E -0.45V
S0.00- 0V




g -0.05-



-0.10
0.0 0.2 0.4 0.6j

V(V)


Figure 6-16. The close up of Figure 6-14, to show the open circuit voltage change under
different gate bias.



















1.12eV


n-Si


4.05eV


SWN


Figure 6-17. Flat band model of gating effect for SWNT/n-Si junction (top) and illustration of
photovoltaic effect (bottom). The blue line illustrates the over-gated situation, where
there is a barrier generated for the hole to transfer from the silicon to the SWNT.





































Figure 6-18. Energy band alignment at the SWNT/n-Si interface before (red) and after (black
and blue) gate voltage is applied. For the black and blue curve, we can see the built in
potential is increased. However, the further increase the gate voltage will push the
Fermi level of the SWNT too low, to generate a barrier for hole to transfer from
Silicon to SWNT, reduce the charge harvest efficient.









CHAPTER 7

PATTERNING OF SWNT FILM

In order to use the SWNT fi1ms in devices, they will typically need to be patterned into

some desired shape. For film pattern dimensions bigger than a millimeter, it is easy to cut the

nanotube fi1m on the membrane prior to transfer with scissors or a razor blade followed by

transfer of the shaped film. However, for sub millimeter dimension patterning, it is necessary to

develop some technique to form the desired pattern. In this chapter, we describe the techniques

for making such patterns with line width in the order of 100 pm. Standard lithography and e-

beam lithography method to produce Eine line width down to 50 nm has been demonstrated by

another group (43). The maj ority content of this chapter was published as a US patent in 2007

(44).

7.1 Patterning of SWNT Film to Sub-Millimeter

The idea is to pre-pattern the filtration membrane within a manner that blocks the pores of

the membrane in the inverse pattern desired for the nanotube film, such that the nanotubes, when

they are vacuum filtered to the surface of the membrane only deposit in the non-occuluded

regions. As one implementation of this we used a solid ink printer (Xerox Phaser 8400) to

produce the pattern in the membrane. An example of this is the interdigitated pattern shown in

Figure 7-1, with line width about 100 Cpm and comparable spacing between the interdigital

fingers. The printer' s ink is a thermoplastic, which is a combination of waxes, polyethylene and

proprietary compounds who's melting point was determined by experiment, to be about 90oC. To

form a patterned nanotube film the inverse pattern of that desired is first printed transparency

sheet (Phaser 840/850 standard transparency film). Transparency sheet is preferred to paper

because the wax spreads more on paper decreasing the resolution of the resulting pattern. The

transparency film is then placed (with printed pattern face down) into contact with the filtration









membrane, a cellulose ester membrane (Millipore, VCWP), on a hot plate set at a temperature

just above the melting point of the thermoplastic ink. The thermoplastic ink melts and is partly

transferred into the pores of cellulose ester membrane, filling the pores in the pattern that was

printed on the transparency sheet. The hotplate is turned off allowing the wax to solidify and the

membrane and transparency sheet are peeled apart. From here the film fabrication process and

the transfer to the desired substrate proceeds as normally. During the filtration the nanotubes are

not drawn to those part of the membrane where the pores have been plugged by the wax and in

fact the hydrodynamic force sweep the nanotubes away from those region to deposit only in

those regions where the pores have not been blocked. This can result in partial alignment of the

nanotube / bundles along the long direction of the edges between the blocked and unblocked

regions of the membrane. If the unblocked regions are narrow lines of order 100 Cpm the

nanotubes can end up preferentially aligned along the long direction of the line enhancing the

conductivity of the film along that direction. Since most application for the patterned nanotube

film will be to serve as electrodes for devices, this technique will result in the enhancement of

the conductivity along the direction where it is most needed. Note that the standard photo

lithography, e-beam lithography or plasma etching does not provide such advantage.

The limitation of this technique comes from the following two factors:

1. the resolution of the printer

2. the length of the nanotube and nanotube bundles

The first one is trivial. The second one is because once the line spacing gets too small,

there will be cross talk between adj acent lines. When one bundle or long nanotube was drawn

down onto the filtration membrane, it will lie parallel to the membrane surface, but will be more

or less random in plane. Once it lie perpendicular to the line direction, it will lie across the










spacing, even if that spacing was been plugged with thermoplastic and no flow rate. If the

numbers of such cross-talk nanotubes are small, we can pass through a high current between the

adj acent pattern to burn those nanotubes away. But once there are significant numbers of

nanotubes or nanotube bundles, as is the case when the pattern line and spacing are in the same

order as the average nanotube bundle lengths (tens of micron meters in our samples), the pattern

lines are no longer separated by the line spacing.

One issue of the above transfer method is that when the thermoplastic melts during the

transfer process, it tends to smear around, which will decrease the resolution of the pattern. We

solve this problem by pulling a vacuum while transferring the pattern. We clamped the

membrane and PET substrate, on which sits the thermoplastic pattern, to the filtration apparatus.

By applying a vacuum while adding hot water (slightly above the melting point of the

thermoplastic ink) to the top funnel. When the thermoplastic is melting, there is a suction force to

guide the melted thermoplastic to the corresponding position of the cellulous ester membrane.

The water was kept at the funnel, since the PET substrate will not allow water to follow through.

Once the thermoplastic pattern was transferred, we can replace the hot water with nanotube

solution, and remove the PET membrane, to allow the nanotube film to form onto the cellulous

ester membrane, with the desired pattern formed.

This is a simple and cost efficient technique to form coarse line width at the order of 100

micron meter, which is useful for many applications such as display purpose. However, nanotube

film can be patterned to submicron scale, which has been demonstrated by Dr. Ant Ural's group

(ref). They use standard photo lithography as well as E-beam lithography showed line width as

thin as 50nm can be achieved.
























Printed Pattern (reversed) Desired Pattern


Figure 7-1. Illustration of coarse patterning of interdigitated finger with line width 0.1Imm









CHAPTER 8
SUMMARY AND FUTURE WORK

8.1 Summary

This dissertation provides a method to fabricate single walled carbon nanotube films and

pattern them down to less than 100nm thick line width. The electronic and optical properties of

the film with different film thicknesses have been studied. This SWNT film provides a new class

of conducting and transparent electrodes for optoelectronic devices, such as LED, photovoltaic

(PV) device, organic LED and organic PV.

An optical analogy based on carbon nanotube FET device was manufactured and

successfully demonstrated by applying a small gate voltage to shift the Fermi level of the SWNT

film about 0.9eV. Baking at 600oC and doping with nitric acid will de-dope and redope the

SWNT film, changing its transparency as well as its conductivity.

Several examples of using SWNT film as electrode to build optoelectronic devices were

shown. First, it was shown that by utilizing the SWNT film to form ohmic contact to p-GaN, it is

possible to light up the GaN LED. Then it is demonstrated SWNT film can be used as the anode

to fabricate MEH-PPV organic LED. Finally, the SWNT film silicon heterojunction was studied

and the transport across the heterojunction can be modulated by a factor of 300.

8.2 Future Work

As discussed in chapter 5, the OLED device still needs further study to improve the

efficiency, as well as the solar cell efficiency improvement in chapter 6. Other possible

applications include using the high surface area of the SWNT film to fabricate capacitors.

Another possibility is by using the nano scale pore size of the SWN\T film to serve as a filter

membrane for exchanging the nano particles.










LIST OF REFERENCES


1. S. lijima, Helical Microtubules of graphitic carbon. Nature 354, 56 (1991).

2. Ph. Avouris, T. Hertel, R. Martel, T. Schmidt, H.R. Shea, and R.E. Walkup, Carbon
Nanotubes: Nanomechamics, Manipulation, and Electronic Devices. Applied Surface
Science 141, 201 (1999).

3. S.J. Tans, A.R.M. Verschueren and C. Dekker, Room-temperature transistor based on a
single carbon nanotube, Nature 393, 49 (1998)

4. A. Thess, R. Lee, P. Nikolaev, H. Dai, P. Petit, J. Robert, C. Xu, Y.H. Lee, S.G. Kim,
A.G. Rinzler, D.T. Colbert, G.E. Scuseria, D. Tomanek, J.E. Fischer, & R.E. Smalley,
Crystalline ropes of metallic carbon nanotubes. Science 273, 483 (1996).

5. J. Kong, N.R. Franklin, C.W. Zhou, M.G. Chapline, S. Peng, K. Cho, H.J. Dai,
Nanotube Molecular Wires as Chemical Sensors, Science 287, 622 (2000)

6. K.H. An, S.Y. Jeong, H.R. Hwang, Y.H. Lee, Enhanced Sensitivity of a Gas Sensor
Incorporating Single-Walled Carbon Nanotube-Polypyrrole Nanocomposites,
Advanced Ma'~terials 16, 1005 (2004)

7. S Chopra, K McGuire, N Gothard, AM Rao, A Pham, Selective gas detection using a
carbon nanotube sensor, Appl. Phys. Lett. 83, 2280 (2003)

8. J. Sippel-Oakley, H.T.Wang, B.S. Kang, Z.C. Wu, F. Ren, A.G. Rinzler and S.J.
Pearton, Carbon nanotube films for room temperature hydrogen sensing,
NaNNNNNNnotcnology~NNN~~~NN 16 2218 (2005)

9. M. S. Strano, C.A. Dyke, M.L. Usrey, P.W. Barone, M.J. Allen, H. Shan, C. Kittrell,
R.H. Hauge, J.M. Tour,R.E. Smalley, Electronic Structure Control of Single-Walled
Carbon Nanotube Functionalization, Science, 301, 1519 (2003)

10. M. Shim, N.W. Kam, R.J. Chen, Y. Li, and H. Dai, Functionalization of Carbon
Nanotubes for Biocompatibility and Biomolecular Recognition, Nano Letters, 2, 285
(2002)

11. A.Modi, N. Koratkar, E. Lass, B. Wei and P.M. Aj ayan, Miniaturized gas ionization
sensors using carbon nanotubes, Nature 424, 171 (2003)

12. S. J. Tans, A.R.M. Verscheuren, and C. Dekker, Room-temperature transistor based on
a single carbon nanotube. Nature 393, 49 (1998).

13. R. Martel, T. Schmidt, H.R. Shea, T. Hertel, and Ph. Avouris, Single and multi-wall
carbon nanotube field-effect transistors. Appl. Phys. Lett. 73, 2447 (1998)

14. J.A. Misewich, R.Martel, Ph. Avouris, J.C. Tsang, S. Heinze, J. Tersoff, Electrically
Induced Optical Emission from a Carbon Nanotube FET, Science 300, 783 (2003)










15. A. Bachtold, P. Hadley, T. Nakanishi, and C. Dekker, Logic circuits with carbon
nanotube transistors. Science 294, 1317 (2001)

16. Z. Chen, J. Appenzeller, Y.M. Lin, J. Sippel-Oakley, A.G. Rinzler, J. Tang, S.J. Wind,
P.M. Solomon, Ph. Avouris, An integrated logic circuit assembled on a single carbon
nanotube. Science 311, 1735 (2006)

17. R. Saito, Dresselhaus, G. & Dresselhaus, M. S. Physical properties of carbon
nanotubes. London: World Scientific Publishing Company, Imperial College Press,
(1998).

18. A. Javey, J. Guo, Q. Wang, M. Lundstrom and H.J. Dai, Ballistic carbon nanotube
field-effect transistors, Nature 424, 654 (2003)

19. Z.H. Chen, Electric field induced transparency modulation in single wall carbon
nanotube ultra-thin films and a method to separate metallic and semiconducting
nanotubes. Thesis, University of Florida, 2003

20. Ph. Avouris, Z. Chen and V. Perebeinos, Carbon-based electronics, Nature
Nanotechnology 2, 605 (2007)

21. Z.H. Chen, X. Du, M.H. Du, C.D. Rancken, H.-P. Cheng, A.G. Rinzler, Bulk
Separative Enrichment in Metallic or Semiconducting Single Wall Carbon Nanotubes,
Nano Letters 3, 1245 (2003)

22. M. S. Arnold, A. A. Green, J.F. Hulvat, S.I. Stupp, & M.C. Hersam, Sorting carbon
nanotubes by electronic structure using density differentiation, Nature Nanotech. 1, 60
(2006)

23. M.J. Earle, J.MSS Esperanca, M.A. Gilea, J.N.C. Lopes, L.P.N. Rebelo, J.W. Magee
K.R. Seddon and J.A. Widegren, The distillation and volatility of ionic liquids, Nature
439, 831 (2006)

24. A.G. Rinzler J. Liu, H. Dai, P. Nikolaev, C.B. Huffman, F.J. Rodriguez-Macias, P.J.
Boul, A.H. Lu, D. Heymann, D.T. Colbert, R.S. Lee, J.E. Fischer, A.M. Rao, P.C.
Eklund & R.E. Smalley, Large-scale purification of single-wall carbon nanotubes:
process, product, and characterization. Applied Physics A 67, 29 (1998)

25. V.C. Moore, M. S. Strano, E.H. Haroz, R.H. Hauge and R.E. Smalley, Individually
suspended single-walled carbon nanotubes in various surfactants, Nano Lett. 3 1379
(2003)

26. S. Nakamura, T. Mukai, M. Senoh, Candela-class high-brightness InGaN/AlGaN
double-heterostructure blue-light-emitting diodes, Appl. Phys. Lett. 64 1687 (1994)

27. D.J. King, L. Zhang, J.C. Ramer, S.D. Hersee, L.F. Lester, Temperature Behavior of
Pt/Au Ohmic Contacts to p-GaN, Mater. Res. Soc. Symp. 468, 421 (1997)









28. L.F. Lester, D.J. King, L. Zhang, J. C. Ramer, and S.D. Hersee, Ohmic Contacts to n-
and p-GaN, Proc. Electrochemical Society 971, 171 (1997)

29. G.K. Reeves, H.B. Harrison, Obtaining the specific contact resistance from
transmission line model measurements, Electron Device Letters, IEEE 3, 111 (1982)

30. Peter K. H. Ho, D. Stephen Thomas, Richard H. Friend, Nir Tessler, All-Polymer
Optoelectronic Devices, Science 285, 233 (1999)

31. Conjugated Polymeric Materials: Opportunities in Electronics, Optoelectronics, and
Molecular Electronics, J.L. Bredas (Editor), R.R. Chance (Editor), Springer-Verlag
New York, LLC; (May 1990)

32. Terj e A. Skotheim (Editor), John Reynolds (Editor), Conjugated Polymers: Processing,
Devices, and Applications (Handbook of Conducting Polymers), 3rd ed., Boca Raton,
FL: CRC Press, Taylor & Francis, Inc., December (2006)

33. Joseph Shinar, Organic Light-Emitting Devices, New York: Springer-Verlag New
York, LLC, October (2003)

34. see www.cheaptubes.com website, Brattleboro, VT 05301 USA, April, 2008

35. G. G. Malliaras, J. R. Salem, P. J. Brock, and C. Scott, Electrical characteristics and
efficiency of single-layer organic light-emitting diodes, Phys. Rev. B 58, R13411
(1998)

36. S.M. Sze, Physics of semiconductor devices, John Wiley & Sons, New York, (1981)

37. Raymond T. Tung, Recent advances in Schottky barrier concepts, Materials Science
and Engineering: R: Reports 35, 1 (2001)

38. Leonard J. Brillson (Editor), Contacts to Semiconductors: Fundamentals and
Technology (Materials Science and Process Technology), Noyes Publications,
Webster, New York (October 1, 1993)

39. Winfried Moinch, Semiconductor Surfaces and Interfaces, Springer, Berlin (May 11,
2001)

40. Mark C. Lonergan, A Tunable Diode Based on an Inorganic Semiconductor
|Conjugated Polymer Interface, Science 278, 2103 (1997)

41. L. Valentini, I. Armentano and J.M. Kenny, Electrically switchable carbon nanotubes
hydrophobic surfaces, Diamnonda~ndRelae 2~~tedr atrial 14, 121 (2005)

42. R.J. Handy, Theoretical analysis of the series resistance of a solar cell, Solid State
Electron. 10, 765 (1967)









43. A. Behnam, L. Noriega, Y. Choi, Z. Wu, A.Rinzler, A. Ural, Resistivity scaling in
single-walled carbon nanotube films patterned to submicron dimensions, Applied
Physics Letters 89, 093107 (2006)

44. A.G. Rinzler, Z. Wu, Low temperature methods for forming patterned electrically
conductive thin films, Publication No.:WO/2007/03 583 8, Publication date: 29, 03,
2007









BIOGRAPHICAL SKETCH

Zhuangchun Wu was born at 1971, in a small village called WuWei, YiXing

county, JiangSu province, China. His father is ZhiYing Wu and his mother HongFeng

Huang. Both his parents did not get a chance to get education and are peasants. He has

two elder sisters Guifang and Chaofang, whom he developed closed relationships with.

He is a student who came from a rural Chinese education. During his high school,

his physics teacher, Mr. Liangcai Lin, inspired his interest in physics and his English

teacher, Mr. MinShi Yang, showed him the door into learning English. Both of these

experiences played an important roles in his later career development. He managed to get

into Nanjing University, one of the best universities in China. Although while there he

only pursued a college diploma, he still had the dream of higher education. He received a

master' s degree from the physics while he was working in Nanjing University as a lab

technician in the Physics Department. He applied to study in the USA and was accepted

by the University of Vermont. While there he received a master' s degree in material

science. Later, he applied to the University of Florida and worked with Dr. Andrew G.

Rinzler on carbon nanotubes for 5 years.

He has a daughter, Emma, who is the source of all of his j oy over the years of

graduate study in Florida.





PAGE 1

TUNABLE CONTACT BARRIER OF SING LE WALL CARBON NANOTUBE FILMS FOR ELECTRICAL CONTACT TO SEMI CONDUCTORS AND POLYMERS By ZHUANGCHUN WU A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008 1

PAGE 2

2008 Zhuangchun Wu 2

PAGE 3

To my parents, my wife and my daughter 3

PAGE 4

ACKNOWLEDGMENTS I thank my academic advisor, Dr. Andrew G. Rinzler, for his thoughtful knowledge and emotional support throughout my PhD study. I also thank my committee members for their advice. The cooperative research projects with Dr. Reynolds group gave me insights about polymer science. I have used various instrume nts and equipment from Dr. Herbards lab which made my research much easier, and I was benef ited from his Solid State Physics lecture. Dr. Jeremiah Mwaura worked closely with me for the OLED project. I also thanks to Dr. Kyu-pil Lee, who did collaboration with me on the GaN LED project. Dr. Zhihong Chen showed me the ropes of the lab and I benefited from discussions and hand on hand training from her. Dr. Jeniffer Si ppel-Oakley, Mr. Bo Li u, Mr. Mitch McCarthy helped me in various ways throughout my Ph.D study. Mr. McCarthy helped me to build a chamber for nanotube/silicon heterojunction sensitivity to environmental gas study. Guneeta Singh deserves my special thanks for all the en lightening conversations about life, culture and everything beyond physics. I would like to th ank Ms. Debra Anderson from UFIC and Ms. Barbara Bostin for their continuous help for my family. I would like to thank Physics machine shop personnel for helping me machining various experimental setups. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES.........................................................................................................................8 ABSTRACT...................................................................................................................................12 1. INTRODUCTION................................................................................................................ .....14 1.1 Single Walled Carbon Nanotube......................................................................................14 1.2 Chiral Vectors, Unit Ce ll of Graphene Lattice.................................................................15 1.3 Energy Dispersion of the Graphene Lattice and Single Wall Carbon Nanotube (SWNT)...............................................................................................................................16 1.4 Density of States of SWNT..............................................................................................17 1.5 Motivation for the Single Wall Nanotube Film Studies in this thesis..............................18 2. SINGLE WALLED CARBON NANOTUB E THIN FILM-FABRICATION AND PHYSICAL PROPERTIES....................................................................................................23 2.1 Fabrication of a Carbon Nanotube Film...........................................................................23 2.2 Transfer of a Carbon Nanotube Film to a Substrate.........................................................26 2.3 Surface Morphology of a Carbon Nanotube Film............................................................26 2.4 Resistivity of a Carbon Nanotube Film............................................................................27 2.5 Optical Spectroscopy of a Carbon Nanotube Film...........................................................27 3. SHIFTING THE FERMI LEVEL OF CARBON NANOTUBE FILM....................................34 3.1 Introduction............................................................................................................... ........34 3.2 Optical Analog to the Electrolyte-Gated Nanotube Based FET (O-NFET).....................35 3.2.1 Experimental Details..............................................................................................36 3.2.2 Transmittance Spectrum of O-NFET as a Function of the Applied Gate Voltage.........................................................................................................................37 3.2.3 The Film Resistivity as a Func tion of the Applied Gate Voltage...........................39 3.3 Discussion and Explanation..............................................................................................39 3.4 Time Drive of the O-NFET..............................................................................................42 3.5 Conclusion........................................................................................................................42 4. OHMIC CONTACT COUPLING CARB ON NANOTUBE FILM TO P-GALLIUM NITRIDE................................................................................................................................52 4.1 GaN Light Emitting Diode Background...........................................................................53 4.2 Experiments Details for Fabric ating SWNT Film Based GaN LED................................54 4.3 Results...............................................................................................................................55 5

PAGE 6

6 4.3.1 Contact Resistance..................................................................................................55 4.3.2 I-V Characteristic of Carbon Nanotube Film/p-GaN.............................................56 4.4 Discussion.........................................................................................................................56 4.5 Conclusion........................................................................................................................57 5. ORGANIC LIGHT EMITTING DIODE WITH SWNT FILM AS ANODE...........................64 5.1 Introduction............................................................................................................... ........64 5.2 MEH-PPV.........................................................................................................................64 5.3 Experimental Details........................................................................................................65 5.4 Discussion.........................................................................................................................66 5.4.1 MEH PPV does not wet with SWNT.....................................................................68 5.5 Conclusion........................................................................................................................69 6. CARBON NANOTUBE/SILI CON HETEROJUNCTION.......................................................74 6.1 Introduction about Schottky Barrier.................................................................................74 6.2 Experimental Details of SWNT/p-Si Heterojunction.......................................................75 6.3 Results of SWNT/p-Si Hete rojunction and Discussion....................................................76 6.4 SWNT/n-Si Photovoltaic Device......................................................................................79 6.5 Experimental details for SWNT/n-Photovoltaic device...................................................80 6.6 SWNT/n-Si Photovoltaic cel l results and discussion.......................................................80 6.7 Conclusion on SWNT/p-Si heteroj unction and SWNT/n-Si solar cell............................82 7. PATTERNING OF SWNT FILM...........................................................................................102 7.1 Patterning of SWNT F ilm to Sub-Millimeter.................................................................102 8. SUMMARY AND FUTURE WORK.....................................................................................106 8.1 Summary.........................................................................................................................106 8.2 Future Work................................................................................................................ ....106 LIST OF REFERENCES.............................................................................................................107 BIOGRAPHICAL SKETCH.......................................................................................................111

PAGE 7

LIST OF TABLES Table page 4-1 Contact resistances of Ti/Al/Pt/Au on carbon nanotubes, carbon nanotube film on pGaN, and standard Ni/Au on p-GaN..................................................................................59 7

PAGE 8

LIST OF FIGURES Figure page 1-1. Chiral vector Ch, transl ational vector T and the unit cell (OACB) of a (4,2) SWNT illustrated on a graphene sheet. .........................................................................................19 1-2. Band structure of graphene lattice (top) and the Brill ouin zone of graphene lattice (bottom). The quantization of the wave vector s perpendicular to the tube axis of the nanotube lead to a set of discrete set of energy sub-bands (red parallel lines in the bottom). ...................................................................................................................... .......20 1-3. Energy dispersion relati onship of (10, 10) and (10, 0) SWNT. Notice the cross of the valence band and conduction band of the (10, 10) nanotube at the Fermi level, which implies it is a metallic nanotube, whereas the energy gap at the Fermi level (0 in these plots) show the (10, 0) nanotube to be semiconducting. .........................................21 1-4. Density of states (DOS) of (10, 10) and (1 0, 0) nanotube. Notice the finite states at the Fermi level of (10, 10) nanotube, resulting its characterization as a metal. While the (10, 0) has a gap, indicati ng its a semiconductor. ............................................................22 2-1. AFM image of a 50nm SWNT film. .....................................................................................31 2-2. The transmittance of a 50nm before and after baking at 600oC in Argon.............................32 2-3. Transmittance of a 240nm free standing SW NT film before (grey curve) and after (black solid curve) baking, with wavelengt h up to 120 micron meter. Also shown is the transmittance vs. much shorter wave length range of a 50nm SWNT film on quartz substrate. The inset shows the electronic transaction between valence band and conduction band of the DOS from ( 12,8) semiconducting nanotube and (10,10) metallic nanotube.............................................................................................................. .33 3-1 The schematic drawing of a solid state optical modulator .....................................................44 3-2. A sketch of the optical analog to the electrolyte-gated NFET..............................................45 3-3. Three transmittance spectrum as a function of applied counter electrode voltage, i.e. 0V, +0.4V and -0.5V.........................................................................................................46 3-4. Density of states (DOS) of (12, 8) single walled carbon nanotube.......................................47 3-5. The transmittance of the nanotube film as a function of the applied gate voltage, from +1.8V to -1.8V. The transmittance at S1 peak (1676nm) increased from 44% to 92%, while the S2 peak (932nm) varied from 51% to 68%. The IR at 3080nm decreased from 97% to 75%, at opposite direction to the S1 peak modulation. ...............................48 3-6. The resistivity of the nanotube film as a function of the applied gate voltage......................49 8

PAGE 9

3-7. The transmittance at S1, S2 and M1 depends on the gate voltage. ......................................50 3-8. Time drive of the S1 transmittance peak with +/1V gate voltage.......................................51 4-1. Schematic view of the GaN based light-emitting diode using SWNT film as the pOhmic contact.................................................................................................................. ..60 4-2. IV characteristics of GaN LED with diffe rent metal, i.e. Ti/Al/Pt/Au, Pd only and Pd/Au contact on carbon nanotube fim. All the devices are using carbon nanotube film as the p-GaN contact. ................................................................................................61 4-3. Emission spectrum of the GaN LED with injection current 0.1mA......................................62 4-4. Picture of the visible emission from th e GaN LED with carbon nanotube film as the pGaN contact electrodes......................................................................................................52 5-1. A) Structure of MEH-PPV (top) and B) Illustration of the OLED device with ITO/PEDOT PSS as anode (middle) C) O LED device with SWNT film (bottom)..........70 5-2. ITO/PEDOT PSS based OLED radiance and current vs. voltage.........................................71 5-3. SWNT film based OLED radiance and current vs. voltage..................................................72 5-4. Illustration of the energy band alignment of the OLED devices...........................................73 6-1. Flat band model of contact ba rrier and Schottky barrier height............................................84 6-2. Illustration of single gate and double gate of carbon nanotube/Si heterojunction................85 6-3. Three steps explaining the SWNT/p-Si sing le gate heterojuncti on device fabrication process................................................................................................................................86 6-4. Circuit diagram of the single gate de vice (top) and double gate device (bottom).................87 6-5. Source drain current as a function of ga te voltage for SWNT/Si heterojunction with heavily and lightly doped p-Silicon. Over a course of +/-0.7V gate voltage, Isd modified 10 and 36 times, for heavily and lightly doped silicon, respectively.................88 6-6. IV characteristics of the SWNT/p-Silicon heterojunction at differe nt gate voltage A) Lightly doped p-silicon and B) heavily doped p-Si...........................................................89 6-7. Flat band model of modulating the cont act barrier of SWNT/Si heterojunction by shifting the Fermi level of SWNTs....................................................................................90 6-8. The current decreased significantly after 5 hour of con tinuously application of gate voltage for the lightly doped silicon de vice. Same phenomena observed for heavily doped silicon device as well..............................................................................................91 9

PAGE 10

10 6-9. Double gating effect, with identical source gate and drain gate. Isd can be modulated more than 300 times by applying gate voltage .7V.......................................................92 6-10. Carbon nanotube silicon junction curre nt in argon environment will change upon introduce oxygen or air......................................................................................................93 6-11. Carbon nanotube silicon junction transpot vs different ambient condition.........................94 6-12. Circuit diagram of the boot-strap PV device.......................................................................95 6-13. I-V characteristic of the SWNT/n-Si solar cell, dark (top) and under illumination (bottom)....................................................................................................................... .......96 6-14. The series resistance was determined to be 283 ohm for low doping n-silicon solar cell. ........................................................................................................................ ...........97 6-15. IV characteristic for the solar cell under different bias.......................................................98 6-16. The close up of Figure 6-14, to show th e open circuit voltage change under different gate bias.............................................................................................................................99 6-17. Flat band model of gating effect for SWNT/n-Si junction (top) and illustration of photovoltaic effect (bottom). The blue line illustrates the over-gat ed situation, where there is a barrier generated for the hole to transfer from the silicon to the SWNT.........100 6-18. Energy band alignment at the SWNT/n-Si interface before (red) and after (black and blue) gate voltage is applied. For the black and blue curve, we can see the built in potential is increased. However, the furthe r increase the gate voltage will push the Fermi level of the SWNT too low, to genera te a barrier for hole to transfer from Silicon to SWNT, reduce the charge harvest efficient.....................................................101 7-1. Illustration of coarse patterning of interdigitated finger with line width 0.1mm................105

PAGE 11

LIST OF ABBREVIATIONS SWNT: Single walled carbon nanotube OLED: Organic light emitting diode PV: Photovoltaic TEM: Transmission electron microscopy AFM: Atomic force microscopy SEM: Scanning electron microscopy ITO: Indium tin oxide EMI-BTI: 1-ethyl-3-methylimidazolium bis-(trifluoromethylsulfonyl)imide 11

PAGE 12

Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy TUNABLE CONTACT BARRIER OF SING LE WALL CARBON NANOTUBE FILMS FOR ELECTRICAL CONTACT TO SEMI CONDUCTORS AND POLYMERS By Zhuangchun Wu August 2008 Chair: Andrew G. Rinzler Major: Physics Single walled carbon nanotubes (SWNTs) are quasi one dimentional molecule formed with pure sp2 carbon-carbon bonding. They are chemically stable, near balli stic transporters, and have high surface area. According to their chiral vector and diameter, they can be either metallic or semiconducting. They are a new class of material in the nanometer range which has unique physical and chemical properties. Their unique one dimensional character gives them unusual electrical properties. A method of manufacturing thin films made of 100% SWNTs was explored. The carbon nanotube films made by this method are thin eno ugh to be transparent, yet still conducting. The electrical and optical properties of those films ha ve been studied, as well as application examples for integrating the film into opto-electrical de vices. The SWNT films made by this method have following advantages: they are thickness controllable, uniform, transparent, conducting, and formed at room temperature. We found the SW NT films were comparable in terms of the conductivity and transparency to the standard industry transp arent electrode, indium tin oxide (ITO). 12

PAGE 13

An optical analog of SWNT based FET was cons tructed which demonstrated the ability to shift the Fermi level of the SWNT film by about 0.9eV. We can shift the Fermi level by such large amount is due to the one dime nsional characteristic of SWNT su ch that the density of states (DOS) of the nanotubes are finite and easy to be filled or emptied upon in jecting or depleting electrons by applying a gate fi eld. This unique feature makes the SWNT film favorable in making ohmic coupling to usually difficult materials, like GaN. The SWNT film has been demonstrated to fo rm ohmic contact with p-GaN light emitting diode (LED). Studies on using SWNT as anode for constructing organic LED (OLED) have shown great hopes. Several challenges of this system have been identified and further research is still on the way. The heterojunction between SWNT film and silicon has been investigated. The carrier injection across the p-silicon and nanotube film junction has been modulated by a factor of 300 by application of a gate voltage of less than 2V. The contact barrier was found sensitive to the environment. Both oxygen and other component of the atmosphere (most likely water vapor) will contribute to the junction resistance change. Th e n-silicon and nanotube film will form Schottky barrier and showed photovoltaic effect. Patterning the SWNT films to sub-millimeter range was demonstrated, which can be beneficial to all the applications. In summary, the SWNT film provides a new t ype of film for electrical coupling to optoelectronic devices as well as various inorganic an d organic materials. The ease of Fermi level tunability for this nanotube film provides the ne w opportunity for working with a wide range of materials, including previously proven to be difficult ones. 13

PAGE 14

14 CHAPTER 1 INTRODUCTION 1.1 Single Walled Carbon Nanotube A single walled carbon nanotube (SWNT) is essen tially a single atomic layer of a graphene sheet rolled to form a seamless tube, with a diam eter on the order of one nanometer, its length in the range of micrometers up to one centimeter. Multi walled carbon nanotubes (MWNTs) are nested coaxial layers of SWNTs. Carbon nanotub es have remained a highly active research area since the MWNT structure was elucidated in 1991 by Iijima (1). SWNTs are quasi one dimensional objects. They have rich physical prope rties as well as many potential applications. According to their chirality and diameter, SW NT can be either meta llic or semiconducting, making them very attractive as the next genera tion of bottom up building blocks for electronic devices (2-4). All the car bon atoms in the SWNT lat tice are bonded with sp2 bonds, which are among the strongest chemical bonds, making SWNT ve ry stable and inert to covalent chemistry except under extreme conditions. Various sensors (5-8) were manufactured based on the SWNTs by detecting their resistance change upon exposure to various chemicals. Functionalization of the SWNT was another research topi c to make SWNTs either bindi ng to different agents for the purpose of biosensor applications or for separa ting metallic nanotubes from semiconducting ones (9-11). Nanotube based field effect transistors (NFET) also have been demonstrated (12-14), as well as integrated logical circuits based on semiconducting SWNTs (15,16). A central topic in this thesis concerns the ease of shifting the Fermi level of a conducting film made of SWNTs. I will first introduce a method of ma king thin, transparent and conducting SWNT films, followed by a description of the physical properties of the film. Then, two examples of using this thin f ilm as the electrode for optoelect ronic devices, in which the Fermi level shift of the nanotubes plays an important role will be discussed. Finally, by shifting the

PAGE 15

Fermi level of the nanotubes in the film, the transport across the hete rojunction between the SWNT film and silicon will be modulated. This first Chapter will give the basic fundame ntals of the SWNTs as well as the motivation of this study. In Chapter 2, the fabrication as we ll as electrical and op tical properties of the SWNT film will be explored. In Chapter 3, a SWNT film is utilized as the electrode to make Ohmic contact with the p-GaN side of a GaN LED. Chapter 4 dem onstrates an optical analog of nanotube based FET and successfully shifts th e Fermi level of a nanotube film by about 0.9 electron Volts. Chapter 5 discusse s using the nanotube film as an anode to construct an organic light emitting diode (OLED). Chapter 6 investigat es the nanotube silicon heterojunction and modulating the transport across the junction by applying a gate field. Chapter 7 explains a method of patterning the nanotube films down to sub millimeter scale. Chapter 8 gives a summary to this thesis and points out the future research needed to be done. The remainder of this chapter will introdu ce the fundamental physical properties of SWNT, which are closely followed according to reference 17 and 19. Chiral vectors, followed with energy dispersion relationship and density of states of the SWNT will be explained and will lay the theoretical foundation for the expe rimental phenomena in this thesis. 1.2 Chiral Vectors, Unit Cell of Graphene Lattice Starting from the unit vectors of th e two dimentional graphene lattice 1 and 2, we define a chiral vector a a hC hC =n1+m2 (n,m) with a a m
PAGE 16

Next we define a translational vector T (also starting with point O, OB), which is orthogonal to chiral vector i.e. hC 0 TCh BC can be obtained through translating OA by amount of T This OACB will make up a unit cell of the (n,m) nanotube, in this case, a (4,2) nanotube. The nanotube (n,m) will be formed if we cut the OACB out of a graphene sheet and roll along the OA direction to form the circumference of the nanotube. So the diameter dh, of the nanotube is dh= 2L/ L= hC = nmmnaCChh22 This OACB is the unit cell in real space. 1.3 Energy Dispersion of the Graphene Lattice and Single Wall Carbon Nanotube (SWNT) By using tight binding approxi mation, only considering the n earest neighbor atoms, one can get the energy dispersion of the graphene sheet, as illustrated in Figure 1-2. In order to calculate the energy dispersion of a SWNT, we need to impose the boundary condition to the 2D graphene lattice, since the nanotubes ar e rolled up into a tube. qKCh2 where q=1, 2, .N, N is the number of hexagons in the real space unit cell OABB. This will result in a quantized wave vector 2 222nmmna q C q kh (Tk the nanotube axis, q=1,2,N) By applying this boundary condition to the 2D graphene lattice, the SWNT dispersion becomes simple cut lines through the 2D disper sion relation of graphene in the direction 16

PAGE 17

specified by the boundary conditions (in turn dictat ed by the real space di rection of the nanotube axis). Examples of the energy dispersion relationshi p of (10,10) and (10, 0) nanotube are shown in Figure 1-3. It is clear that the valence band and conduction band of (10, 10) nanotube has a cross point right at the Fermi level, which makes it intrinsically meta llic. While for (10, 0) nanotube, there is a band gap at the Fermi level. So, it is a semiconducting nanotube. 1.4 Density of States of SWNT In order to calculate the de nsity of states (DOS) of a SWNT, one can take the energy dispersion relationship of the SWNT, divide th e energy into small intervals and sum the k vectors appearing in each energy segment. This pr ocedure leads to density of states diagrams specific to each nanotube of distinct ( n,m) index such as shown in Figure 1-4. As shown in Figure 1-4, (10, 10) and (10, 0) nanotube density of states again confirms that the (10, 10) nanotubes are metallic, with finite dens ity of states at the Fermi level, while the semiconducting (10, 0) nanotube has a band gap right at the Fermi le vel. The sharp features in the DOS of the nanotubes are called von Hove singularities. There can be el ectronic transitions between symmetric valence band and conduction band von Hove singularities, resulting in absorption bands in the optical spectrum of the SWNTs. The band gap for the semiconducting nanotubes is inversely proportional to th e diameter of the nanotube (17). The density of states of SWNT is finite, but the mobility along the nanotube is near ballistic (18,20). The effective scattering length is on the order of micrometers for metallic tubes and a few hundred nanometers for semic onducting nanotubes (18,20). Because the carbon nanotubes are only quasi one dimens ional, they have some tolerance for defects. Although the density of states for the nanotubes is low (compared to typical metals) the charge mobility is very 17

PAGE 18

high making the nanotubes good conductors. The Fe rmi level of the nanot ube is easily tuned because of the low density of states. 1.5 Motivation for the Single Wall Nanotube Film Studies in this thesis Techniques developed to produce thin homogeneous films of SWNT that are simultaneously transparent and el ectrically conducting are exploited to study the potential of the SWNTs as Fermi level tunable electrodes in electronic and optoelectronic devices. 18

PAGE 19

a1 a2 (4,2) ChT (4,-5)O A B C Figure 1-1. Chiral vector Ch, tr anslational vector T and the un it cell (OACB) of a (4,2) SWNT illustrated on a graphene sheet. 1 1 Reprinted with permission from Ph. Avouris, Z. Chen and V. Perebeinos, Nature Nanotechnology 2 605 (2007) Copyright 2007, Nature Publishing Group 19

PAGE 20

Figure 1-2. Band structure of graphe ne lattice (top) and the Brill ouin zone of graphene lattice (bottom). The quantization of the wave vector s perpendicular to the tube axis of the nanotube lead to a set of discrete set of energy sub-bands (red parallel lines in the bottom). 2 2 Reprinted with permission from Ph. Avouris, Z. Chen and V. Perebeinos, Nature Nanotechnology 2 605 (2007) Copyright 2007, Nature Publishing Group 20

PAGE 21

Figure 1-3. Energy disper sion relationship of (10, 10) and (10, 0) SWNT. Notice the cross of the valence band and conduction band of the (10, 10) nanotube at the Fermi level, which implies it is a metallic nanotube, whereas the energy gap at the Fermi level (0 in these plots) show the (10, 0) na notube to be semiconducting. 3 3 Reprint with permission from Z.H. Chen Thesis, University of Florida, 2003 21

PAGE 22

22 Figure 1-4. Density of states ( DOS) of (10, 10) and (10, 0) nanotube Notice the finite states at the Fermi level of (10, 10) nanotube, resulti ng its characterization as a metal. While the (10, 0) has a gap, indicating its a semiconductor. 4 4 Reprint with permission from Shigeo Maruyama www.photon.t.u-tokyo.ac.jp/

PAGE 23

CHAPTER 2 SINGLE WALLED CARBON NANOTUBE THIN FILM-FABRICATION AND PHYSICAL PROPERTIES Prior to 2002, much of the effort in the co mmunity of SWNT research was focused on the individual nanotubes (2,3,5). In this thesis, a different approa ch was explored by using the SWNT effectively in bulk as thin films possessi ng thickness less than 20 0 nm over macroscopic areas (up to 5.5 inches in diamet er). The film fabrication a nd properties are described here. 2.1 Fabrication of a Carbon Nanotube Film All the single walled carbon nanotubes used in this study were synthesized from laser vaporization growth and later pur ified by Dr. Andrew Rinzler (4,24) It will be referred to as carbon nanotubes or nanotubes. The growth and purification process is briefly described as follows. The laser vaporization growth a pparatus involves a dual pulsed laser beam, precisely timed to fire one after another to hit a rotating target. The target was co mposed of a mixture of catalyst cobalt and nickel particles, 1 atomic percent each together with carbon, sitting in the hot zone of the 1200oC furnace, with flowing argon (4,24), all in a two inch diameter closed quartz tube. The SWNT deposit on the down stream of the quartz tube wall, together with a mixture of the amorphous carbon and metal catalyst particles. A ll the amorphous carbon and catalyst particles need to be removed in order to get the pure SWNT. The carbon nanotubes were purified by a cross flow filtration method to remove the amorphous carbon and catalyst particles (24). In order to separate into small bundles and individual nanotubes, typically they will be susp ended in an aqueous solution (25), otherwise the nanotubes entangle with each other to form an in tractable mass. Since the surface volume ratio of the nanotubes is tremendous, the Van der Waals force is strong enough to hold them together. In order to suspend in water, some kind of surfact ant is needed. It is found 1 weight percent of 23

PAGE 24

Triton-X 100 surfactant solution will stably su spend the nanotubes for a long time (24, 25). The surfactant has a hydrophobic head and a long hydrophilic tail. The surfactant will form hydration shells around the nanotubes. Only once the surfactant concentration reaches the so called critical micelle concentration (CMC), does it starts to coat the nanotubes. The process of surfactant coating of the nanotube is a dynamic process. So me surfactant will always come off the nanotube and some other surfactant molecules newly bind onto the nanotube surface. If the surfactant concentration is lower than the CMC, the surfac tant polymers are rather associated with each other to form the micelles. Once the concentr ation is above CMC, there will be enough free surfactant for the nanotube surfaces to begin to be coated. Even with the use of a surfact ant energy must be put into th e material to break apart the large agglomerates of nanotubes. This is accomplished using ultrasonication. The ultra sonication energy will break the aggregates apar t. The surfactant can th en quickly coat the nanotube bundles preventing reaggregation. Too much sonication will start to cleave the nanotubes and result in shorter tube length. So, its a matter of balancing the degree of bundle disaggregation to individual nanot ubes versus the tube length. To make a uniform SWNT film, it is desired to have more indi vidual and small bundle nanotubes in the solution. It will make the film more homogeneous and decrease surface roughness. This becomes more important when making organic optoelectronic devices where the working polymer layer is in the order of 100 nm thick and the surface roughne ss of the SWNT film becomes critical. Apart from the bundle size, the particles in the nanotube film are another critical issue, just like the silicon industry needs to lower the particle concentration in a silicon wafer. Much effort including extra filtration process and the environment where those films we re made have been taken into account to address the particle issue. More details will be discussed in chapter 5. 24

PAGE 25

The nanotube solution concentration (mass/vol ume) in a stock solution was determined simply by making a thick nanotube film by the filtration method (described below) from a known volume of solution on a Teflon filter membrane so that the film could be peeled off the membrane (so called buckypaper) allowing determin ation of its mass with a microbalance. Once the concentration was determined a thin film was made from a known volume of the stock solution and its thickness measured by atomic force microscopy (AFM). Once the film thickness (for a particular area) resulting from one part icular volume of the stock concentration was known, a simple calculation will te ll us how much solution is needed for any desired thickness film of a given area. The film is made by the filtration process trapping the nanotubes on filtration membranes with pores too small for the nanotubes to permeate. The filtration proces s sketch is shown in Figure 2-1. A membrane (cellulous ester, Mi llipore, VCWP) with pore size 100nm was typically used. This material was carefully chosen so that it can be easily dissolved away later by dissolution with acetone during the transfer process, resulting in a pure nanotube film, either attached to a substrate or free standing. This fabr ication process is self-re gulated to form uniform thickness films. If one region of the membrane were to accumulate more nanotubes, the nanotubes themselves begin to impede the flow of the solution through that portion of the membrane hence bringing fewer nanotubes to th at region. The pore size of the nanotube film made from a well sonicated solution usually is in the order of few to tens of nanometer, as seen on AFM images. This means the pore sizes are much smaller than the aver age pore sizes of the filter membrane used. After a few layers of the nanotube landed on the filter membrane, the regulation of the flow was dominated by the sma ller pore size of the nanotube film. Once the solution was all gone through, the film was then ri nsed with sufficient amount of DI water to 25

PAGE 26

remove the remaining surfactant in the film. The nanotube film is dried on the filtration membrane, which brings the nanotubes into intimate contact with each other. 2.2 Transfer of a Carbon Nanotube Film to a Substrate In order to use the film, a transfer process is needed, either being transferred to a substrate or held up by a frame for free standing purpose. The transfer process starts with a nanotube film sitting on the membrane, wetted with DI water, then bringing the nanotube film in contact with a clean surface. It can be virtually any surface of the material fo r which nanotube film to contact with or the surface of a supporting substrate, as lo ng as that substrate can tolerate the membrane dissolving and the rinsing solvent, in this ca se, acetone and methyl alcohol. Once the two surfaces were in contact, the two objects will be sandwiched by two metal plates and a clamp will be used to apply pressure to the assembly. While sitting in an 80o C oven for half an hour, the majority of the DI water initially for wett ing the nanotube film will evaporate and the surface tension of the water will bring th e nanotube film into intimate cont act with the substrate surface. Then the membrane will be dissolved away, using sequential acetone baths (typically 4 baths) to ensure removal of residual membrane material. Fina lly methyl alcohol was used as the final bath to then dry the nanotube film sample out of a low residue solvent. 2.3 Surface Morphology of a Carbon Nanotube Film The surface morphology of the nanotube film was characterized by the scanning electron microscope (SEM) and the atomic force microsc ope (AFM), as shown in Figure 2-1. The root mean square surface roughness is about 7 nm. For a thinner film, for exam ple, a 7nm film, there are appreciable areas of open spaces and other areas with nanotube bundles. In order to lower the surface roughness, smaller bundles and more indi vidual nanotubes will help. From AFM images, we will see particles embedded in the nanotube f ilms and also on the surface of the film. These particles are typically on the orde r of micrometers, which is much bigger than the nanotube film 26

PAGE 27

thickness. For using the nanotube film as electro des in organic optoelectronic devices, those particles will cause serious shorting issues, because those polymer photoactive layers are typically on the order of 100nm thick. Increasing the polymer layer will decrease the device efficiency because the carrier mobility is limite d for polymers. It is an ongoing research project being carried out in our group for improving the surface smoothness of the nanotube film. 2.4 Resistivity of a Carbon Nanotube Film The resistivity of the nanotube film was measured by the Van der Pauw method. For as prepared 50nm film, the re sistivity is typically 3x10-4 cm (60 ohm/square) and the value is quite stable, for as long as a month. The resistivit y will increase 5-10 times when the film get dedoped by baking at 600oC in Argon for half an hour. Once the film has been de-doped, it can be re-doped by exposure to various dopan ts at room temperature, such as bromine vapor or nitric acid vapor. The resistivity will recover or even be slightly lower than the as-prepared sample. However, this doping is not stable, once the film is exposed to air for a couple of hours, it will become de-doped. 2.5 Optical Spectroscopy of a Carbon Nanotube Film Optical spectroscopy of the nanotube film ha s been characterized by a Perkin-Elmer dual beam spectrometer (Lambda 900). The dual beam refe rs to sample beam and reference beam. By placing the sample on a substrate in the sample b eam and an identical substrate in the reference beam the instrument normalizes the sample transmittance at each wavelength to the substrate transmittance in the reference beam. Hence the ab sorbance of the sample is determined, without substrate absorbance. The absorption of the light was come from the sample, in this case, a nanotube film. The spectrometer generated a monochr omatic light, through a slit, incident on the sample (in our case, perpendicular to the samp le surface), and the detector integrated the radiation that passed through the sample at th e same wavelength for a given period of time 27

PAGE 28

(integration time). At the Ultra violet/Visible rang e, the slit width was kept as small as possible to maintain the accuracy of the wavelength, but large enough to give a good signal to noise ratio. In the near infrared region (in our case, above 850nm) where the detector has poor dynamic range, the intensity striking the detector was kept constant by modifying the slit width.. Therefore, the system adjusted the slit width to achieve the required constant intensity and the width of the slit would be recorded and conve rted into the transmittance of the film. A typical transmittance spectrum of the nanotube film is shown in Figure 2-2, before and after baking at 600oC in argon environment. The dips in the transmittance around 1670nm, 930nm and 650nm are labeled S1, S2 and M1, respectively. S1 and S2 peaks are the absorption bands which correspond to the electronic transitions between the first and second pair of the Van Hove singularities in the conduction band and vale nce band of the density of states of the semiconducting SWNTs, as illustrated in chapter 1. M1 peak corresponds to the absorption from the first pair of such transitions contributed by the metallic nanotubes. The absorptions observed are broad, despite the very sharp structure of the van Hove singular ities, because of the dependence of the energy separation between th e singularities on the nano tube diameter (recall the inverse proportionality of the band gap to the nanotube diameter) and the diameter distribution of the nanotubes present in the samp le (17). The bundling of nanotubes in the sample also acts to perturb the singular ities smearing out the contribution from the discrete tubes to the absorption. It is worth while to emphasize that the absorp tion peak intensity for a given nanotube film provides information about the occupancy of the nanotube electronic density of states. As the Fermi level shifts inside the nanotube film, the S1, S2 and M1 absorption peak intensity will 28

PAGE 29

change accordingly. The transmittance spectra thus provide information about the location of the the Fermi level in the SWNT films. Figure 2-2 shows the transmittance of a 50nm th ick SWNT film, before and after baking at 600oC inside an argon environment. The difference in the S1 and S2 intensity is obvious for preand post baking of the film. As described below this change is due to a shift in the Fermi level of the nanotubes upon baking. This Fermi level shifting is the result of chemical dopants removal. We will show shifting the Fermi level of the nanot ube film through electric field in chapter 3 and 6. In order to understand the change in the intensity of the S1 and S2 peaks before and after baking for the nanotube film, we have to understand from the point of view of the DOS of the nanotube. The inset of Figure 2-3 shows the DOS of the (10,10) and (12,8) nanotubes. The arrows in the inset correspond to electronic transition from the valence band singularities to the conduction band singularities for S1 and S2 trans ition of the (12,8) nanotube and M1 transition of the (10,10) nanotube, which correspond to the absorption peak in the transmittance spectrum. If the Fermi level sits in the middle of the ga p, the available electrons in the fully occupied valence band and the available states in th e empty conduction band are maximized, hence the transition probability is maximi zed,. Once we p-dope the SWNT film, the Fermi level will shift towards the left. When it moves into the valen ce band, it will empty some states in the valence band, making fewer electrons available for partic ipating in the electronic transition, which will result in the decrease of the absorption S1 peak. On the other hand, if we n-dope the nanotube, the conduction band states will be partially occupi ed. Then the transition from valence band to the conduction band will also get sm aller, resulting in a smaller ab sorption S1 peak. If the Fermi level moves deep into the valence band, further in to the second van Hove singularity, it will start 29

PAGE 30

to decrease the S2 absorption peak. The same analysis is applied to the M1 peak as well. This is why we can use the spectrum as the indicator of where the Fermi level of the nanotube film sits but cannot tell whether it is p dope d or n doped. It is easy to te ll whether its p or n doping by shifting the Fermi level co rresponding to the known external fi eld applied to the nanotube film while monitoring the spectrum, which will be explained further in the next chapter. Figure 2-3 shows the transmittance of two films, one 50nm film sitting on quartz, measured with a narrow wavelength window, the other a 240nm free stan ding SWNT film, with a much broader wavelength window, up to 300 m, before and after baking at 600oC with argon. Notice in Figure 2-3, there is a big transparen cy window between the wavelengths of 2 to 10 m, at which range the industry standard transp arent conducting film indium tin oxide (ITO) is totally opaque. This unique feature makes SWNT film an outstanding candidate for using in microwave electronic applications. In summary, we described a simple method for manufacturing pure single walled carbon nanotube films and characterized the electrical and optical prop erties of the film. The SWNT films are comparable to ITO in terms of the conductivity and the transp arency in the visible range but far better transparent in the infr ared region (ITO is totally cut off after 2 m). The Fermi level of the SWNT film is readily shif table upon depleting or a dding electrons to the nanotube film, either by chemical or as will be shown electrical means. 30

PAGE 31

Figure 2-1. AFM image of a 50nm SWNT film. 5 5 Reproduced in part with pe rmission from [Z.Wu et.al, Science 305, 1273-1276 ] Copyright [2004] AAAS 31

PAGE 32

Figure 2-2. The transmittance of a 50nm before and after baking at 600oC in Argon. 32

PAGE 33

33 Figure 2-3. Transmittance of a 240nm free standing SWNT film before (grey curve) and after (black solid curve) baking, with wavelengt h up to 120 micron meter. Also shown is the transmittance vs. much shorter waveleng th range of a 50nm SWNT film on quartz substrate. The inset shows the electroni c transaction between valence band and conduction band of the DOS from (12,8) semiconducting nanotube and (10,10) metallic nanotube.6 6 Reproduced in part with permission from [Z.Wu et.al, Science 305 1273-1276 ] Copyright [2004] AAAS

PAGE 34

CHAPTER 3 SHIFTING THE FERMI LEVEL OF CARBON NANOTUBE FILM 3.1 Introduction In previous research, Dr. Zhihong Chen in our research group has designed a solid state optical modulator (Figure 3-1) and demonstrat ed that upon applying a gate voltage, the Fermi level of the carbon nanotube film can be shif ted (as shown by a 0.1% modulation of the S1 transmittance peak at 6V of gating voltage) (19). The solid state optical modulator is essentially a transparent capacitor, with two electrodes sandwiched with AlOx as the dielectric layer. The two electrodes are a thin layer of carbon nanotube film and indi um tin oxide (ITO), respectively. The whole structure sits on a glass substrate. By applying a voltage be tween the two electrodes of the capacitor, the optical transmittance of the whole device was monitored by a dual beam spectrometer, having the sample beam going th rough the entire device. Since S1 peak is corresponding to the electronic transitions betwee n the first pair of von Hover singularities of semiconductor single wall carbon nanotubes, we reach the conclusion that the Fermi level of the SWNT has been changed upon the application of th e gate field. The reasons the total modulation effect is rather small are two fold: the first and most important one is electronic screening. The gate field will be mostly screen ed by the first several layers of the carbon nanotubes next to the dielectric layer, leaving the re st of the nanotubes not experienci ng the gate electric field. Second, the gating field is rather small, since the dielectric la yer is about 81nm thick. In order to increase the modulation percentage, more charges are needed to inject into the electrodes. By simply increasing the capacity, one can inject more char ges. The capacitance per unit area is important in this modulation effect. One way to increase the capacitance is to reduce the distance between the two electrodes of the capacitor, which is reducing the dielectric layer thickness. But reducing the dielectric layer will cause breaking down of the dielectric layer and cause leakage current. 34

PAGE 35

For their operable voltage range electrolytic capacitors have much greater charge storage capacity than dielectric capacitors. Th is is due to the fact that in th e electrolyte, there are ions in the solution that can move freely, in response to the electric field ge nerated by the electrodes immersed into the electrolyte solution. The ions will accumulate very close to the electrode, which has the equivalent effect of reducing the spacing between the capacitor electrodes, helping to inject more charges onto the electrodes. The distance for those ions that accumulate next to the electrode is within the so calle d Helmholtz layer or double layer, which has a thickness typically on the order of a few nanometers. Here the ph enomenon is exploited to greatly enhance the electric field gated modula tion of the absorption bands in the nanotube films. 3.2 Optical Analog to the Electrolyte-G ated Nanotube Based FET (O-NFET) We designed a device which is an optical an alog to the electrolyt e-gated nanotube based field effect transistor (O-NFET). In this device we have two electrodes, one sample and one counter (gate) electrode. They are identical SWNT films. The two films were both immersed in the electrolyte, i.e., an ionic liquid EMI-BTI (1-et hyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide) (23) By applying a voltage between the two films, we could inject electrons or holes onto each of the nanotube electrodes, respectively. With the help of the ionic liquid, the charge we can in ject onto the nanotube film electrodes will be much increased compared to having no ionic liquid (as we expl ained above). The transmittance through the sample SWNT electrode was monitored while appl ying a different potential between the sample and counter electrodes. As discussed in the previous chapter, monitoring the transmittance of the SWNT film at wavelengths co rresponding to particular elec tronic transitions provides information about the Fermi level within the film. 35

PAGE 36

3.2.1 Experimental Details As illustrated in Figure 3-2, Two identical, 150nm thick single wall carbon nanotube films (15mm by 8.5mm) were transferred onto a sapphire substrate (25mm by 25 mm), with 2mm gap between them. Two thin (50nm) strips of palladi um film were thermally evaporated across the top edges of the nanotube films to make electric al contact with each of them. At the bottom of the sapphire substrate, a U shape thin rubber gasket was placed al ong the edge of the substrate on the film side, and covered with a thin glass plate (2 5mm by 8mm, hence not in the path of the light) to form a reservoir for the ionic liquid. Th e gasket, glass plate and substrate were held together by a clamp. Once the device was constructed it was held horizontally and the nanotube films were wetted and saturated by ionic liquid EMI-BTI. This i onic liquid is viscous, with a very low vapor pressure and stable at room temperature in ambient environment (23). Once the films were wetted the device was tilted to be in a vertical position, the excess amount of ionic liquid drained to the reservoir but by ca pillary forces a thin layer remained on top of the two nanotube films. The two nanotube film strips were also in elec trolytic contact through the ionic liquid drained into the reservoir. The O-NFET device was placed inside a Perk in-Elmer dual beam spectrometer (Lambda 900). By measurement it was previously confirme d that EMI-BTI is transparent from 300nm to 2 m of wavelength. The monochromatic sample b eam passes through the sample nanotube film and the other nanotube film serves as the gate electrode. Initially we used a platinum wire (1.5mm diameter) as the gate electrode. But since the surface area of the Pt wire (although we increased the length of the wire by making several circles inside the ioni c liquid reservoir) is much smaller compared to that of the sample na notube film, it would be a limiting effect for the 36

PAGE 37

capacitance of this device. This le d to the decision to use an identical nanotube film as the gating electrode, thus solv ing this problem. One disadvantage of this device is the the response time is rather long, typically in the order of minutes. The rearrangement of ions in the viscous ionic liquid is slow. Accordingly a delay of more than 5 minutes was used between each change of volta ge and the spectral measurement. The delay time was determined by monitoring the non-Faradaic charging current of the device when the voltage was changed. Once the gate current dropped to tens of nanoamps, the spectrum was recorded (spectra recorded when the charging currents differed by a factor of 2 in this range showed no discernable difference). 3.2.2 Transmittance Spectrum of O-NFET as a Function of the Applied Gate Voltage The transmittance of the SWNT film electrode as a function of the gate voltage is shown in Figure 3-3. The measurement takes place from UV to near infrared (IR), i.e. from 350nm to 3080nm. The gate voltages are 0V (no gate voltage), positive 0.5V and negative 0.4V. First, lets focus on the absorption spectrum when the gate vo ltage is 0V. There are three absorption peaks, S1 (1676nm), S2 (932nm) and M1 (652nm). As disc ussed in chapter two, the peaks S1 and S2 correspond to the electronic tran sitions from the conduction band to the valance band of the semiconducting nanotubes in the SWNT sample film. This electronic transition is illustrated in Figure 3-4, which shows the dens ity of states of the (12, 8) semiconducting SWNT. As we know, the energy gap is inverse proportional to the nanotube diameter (17). The diameter distribution of the nanotubes in our sample is determined by TEM images in the reference 24, which is mainly between 1.1-1.6nm. That ener gy gap agreed well with the (12,8) nanotube, which has the diameter of 1.365nm. Compare to the DOS of (12,8) nanotube, th e S1 and S2 peaks in the transmittance spectrum of the nanotube film are much more broadened. The broadening of the peaks is mainly 37

PAGE 38

due to the distribution of the nanotube diameter in the film as well as to the bundling effect, as we discussed in chapter two. Typical bundles in our films are about 5-10nm in diameter. These bundles will also contribute to one of the gating effects, which will be di scussed later in this chapter. In other words, for a given sample with a fixed electronic status, the transmittance or absorbance spectrum will be constant. The electronic status is the availability of the valence band carriers and the conduction band carriers, or where th e Fermi level is sit for a given material. If we apply a voltage to inject elect rons or holes onto that sample, it will shift the Fermi level of the sample, by changing the conduction band or valen ce band carrier distributions, hence changing the corresponding absorption spectrum. This is the case for Figure 3-5, where we applied +0.4V, 0V and -0.5V to the nanotube film counter elec trode. Because there is ionic liquid next to the carbon nanotubes and the Helmholtz layer is on the order of nanometers, th e nearby counter ions allowed us to inject much more charge onto th e entire bulk of the nanotube film with a lower gate voltage, compared to the solid state device di scussed at the beginning of this chapter. When a negative voltage (-0.5V) is applied to the coun ter electrode, positive i ons from the ionic liquid will accumulate at the counter electrode, and negative ions will accumulate at the sample nanotube film. This will allow more holes to be injected onto the sample nanotube film, shifting the Fermi level to the p side, resulting in less ab sorption. As a result, the transmittance at S1 peak increased from 68% (Vg=0V) to 73.9% (Vg=-0.5V), which is consistent with our prediction. The opposite situation happens when the positive voltage is applied to the counter electrode, in which case the transmittance of the film at S1 decreases from 68% to 58%. The transmittance of the nanotube film as a func tion of the applied volta ges in the range of positive 1.8V to negative 1.8V on the counter elec trode (gate voltage) is shown in Figure 3-5. 38

PAGE 39

This gate voltage range was chosen to remain below the reduction/oxidation threshold for EMI BTI,. 3.2.3 The Film Resistivity as a Functi on of the Applied Gate Voltage In order to confirm that the carrier density at given point of the gate voltage on the nanotube film agreed well with the Fermi level sh ifting picture, the resistivity of the nanotube film was measured as a function of gate voltage in a similar de vice using a Linear Research LR 700 AC Resistance Bridge operated at 16Hz. To facilitate the four probe resistance measurement the sample film in this case had four 1 mm wide Pd strips deposited acro ss the thin direction of the film separated by 3 mm. Figure 3-6 shows th e film resistivity co rresponding to the gate voltage. As the gate voltage starts from nega tive 1.8V increasing towards zero and then positive, the resistivity continuously increases until the gate voltage reaches +1.4V, after which the resistivity decreases a little as the gate voltage continues in creasing to +1.8V. 3.3 Discussion If the semiconducting nanotubes in the nanotube film are intrinsic or un-doped, the Fermi level of those nanotubes will be sitting at the middle gap of the DOS of the nanotubes. The nanotubes in the films are a mixture of semic onducting and metal nanotubes, and the Fermi level of the films will be the equilibrium of all the na notubes in the films. So, the Fermi level of the film will be sitting at the middle of the DOS of the nanotubes if all the nanotubes are intrinsic. The following direction of left or right refers to the hori zontal axis of the Figure 3-4, the electronic energy of the nanotube film. For intrinsic nanotubes, the S1 peak of the absorption spectrum will be at a maximum when the gate voltage is zero. By injecting elec trons to the nanotube film, those electrons will occupy the conduction band density of states, which wi ll shift the Fermi level to the right. Those newly occupied conduction band states will be un available to participate into the electronic 39

PAGE 40

transitions between valence band and conduction band, which will decrease the absorption of the S1 peak. On the other hand, if we inject holes in to the nanotube film (or deplete electrons from the nanotube film), shifting the Fermi level to the left side, that will decrease the valence band carriers to be excited into conductio n band, also resulting in decrease of the absorption S1 peak. The experimental observation is that when applying negative gate voltage, the transmittance of S1 increases, or in other words the absorption decreases, which is agreed well with our previous analysis. But by applying positive gate voltage, the transmittance of S1 decreases, or equivalent to the increase absorption of S1 peak. This tells us that the Fermi level is not sitting in the mid-gap, rather, in the p-side (left of the gap) under the S1 valence singularities when the gate voltage is zero. This is somewhat surprising since the nanotube films were baked to 600o C in Ar to de-dope them prior to their saturation with the ionic liquid. This seemingly intrinsic p-type behavior of the nanotube film is likely due to the equilibration of the chemical potential of the mixed metallic and semiconducting nanotubes with the Pd contact electrodes in the presence of the surrounding ionic liquid. A second initially perplexing phenomenon was that far before completely saturating the changes in the S1 peak, the S2 peak starts to change. This is better observed in Figure 3-7, which plots the transmittance at the corresponding peak positions versus the applied gate voltage. According to the simple picture of the gate volta ge induced Fermi-level sh ift, the change of S1, S2 and M1 should appear sequentially as the Fermi level progress se quentially th rough the corresponding valence band singularities. This seemingly paradoxical behavior is readily explained by electrosta tic screening and the fact that the nanotubes in the f ilms are not individual nanotubes but rather bundles of nanotubes. For bundles, the inside nanotubes are not directly exposed to ionic liquid. At the lower applied 40

PAGE 41

voltages only the outer nanotubes of the bundles w ill form double layer with ionic liquid and participate in the electronic doping process. Th e charges on those outer layer nanotubes will partially screen the inner nanotubes from the ionic field. Once the equilibrium is established for a given gate voltage, the Fermi level for the outer layer nanotubes can well lie below M1 while the Fermi level for the inner nanotubes can still lie below the S1 singular ity. Since however the carrier density of nanotubes is low compared to conventional metals the outer layer of nanotubes can only partially screen the inne r nanotubes so that at the higher applied voltages the S1 peak can is virtually gone, which means even the Fe rmi level has shifted below the S1 valence singularity even for the innerm ost nanotubes in the bundles. The transmittance of the film in the near IR region (above 2200nm) responds to gate field in the opposite direction as that of the S1 pea k. When the gate voltage is increased in the negative direction, the transmittance of S1 increases while that in the near IR region decreases. The transmittance in the near IR region is governed by free carrier absorption. As the gate voltage is made more negative th e concentration of p t ype carriers is increa sed (making the film more conducting) and increasing the free carrier absorption explaining th e observed changes inj the near IR Transmittance. From the resistivity versus gate voltage relationship, we can figure out whether the nanotube film is intrinsic or either n or p doped. If th e Fermi level is initially sitting in the n-side, the resistivity will decrease when the gate voltage moving toward the negative, since the negative voltage will make the film less n doped, which is not the phenomena we see in our resistivity data. What is really going on is that when a negative voltage is applied to the counter electrode, the Fermi level of the nanotube film shifts away from the band gap, resulting in the drop of the resistivity of the film. This implies that the nanotubes were previously p-doped under 41

PAGE 42

the zero gate voltage. When the Fermi level shifting away from the gap, the nanotubes will be more p-doped, in other words, more conducting. Ind eed the resistivity of the nanotube film will decrease as the gate voltage becomes more negative, which is also confirmed by the near IR regions transparency drop. As the gate voltage increases from zero to +1.4V, the resistivity (Figure 3-6) increases accordingly, which corresponds to a decrease in the carrier (hole) concentration, as the Fermi level moves into the gap. This change in the resistivity saturates afte r 1.4V gate voltage and actually begins to turn around indicating that the nanotubes have just begun to be n doped at the largest voltage of 1.8V This behavior is consistent with sa turation appeared in the S1 peak after +1.4V gate voltage. 3.4 Time Drive of the O-NFET In order to confirm that the gating effect of the transmittance spectrum comes from the gating effect, we performed a time drive measurem ent at the S1 wavelength to confirm that the whole process is reversible. This step help s to rule out any redox chemistry contribution. A square wave of +/1V gate voltage was applied, each lasting 50 se conds with a period of 100 seconds, while monitoring the transmittance of the S1 peak at 1676nm. The transmittance of S1 peak will alternate from 71% to 57%, fully recoverable, as illustrated in Figure 3-8. Notice that the transmittance gradually changing but does not reach a plateau during the 50 seconds period. As discussed previously, this is due to the slow reorganization of the ions in the viscous ionic liquid during charging and discharging. The transmittance will reach a plateau after a few minutes. 3.5 Conclusion The Fermi level of the single wall car bon nanotube can be shifted upon applying an electrical field inside an op tical analogy nanotube based FET has been confirmed. The Fermi 42

PAGE 43

level shifting effect was confirmed through monito ring the transmittance inte nsity change of the nanotube film at certain wavele ngth which corresponding to the electronic transition of the SWNT while applying a gate voltage. Since spectral changes are observed from the M1 band associated with a depletion of the first valence band of the metallic nanotubes in the sample (see the DOS for the (10,10) nanotube shown in the in set of Figure 2-3) th e Fermi level of the nanotube can be shifted by about 0.9eV by applyi ng a gate voltage less than 2V. The near IR transmittance decrease was corresponding to the free carrier absorption was also observed as the carrier concentration increase in the SWNT film. The film resistiv ity as a function of the gate voltage showed the SWNT was initially p-doped. 43

PAGE 44

Figure 3-1 The schematic drawing of a solid state optical modulator 7 7 Reprint with permission from Z.H. Chen Thesis, University of Florida, 2003 44

PAGE 45

Figure 3-2. A sketch of the optical analog to the electrolyte-gated NFET8 8 Reproduced in part with permission from [Z.Wu et.al, Science 305 1273-1276 ] Copyright [2004] AAAS 45

PAGE 46

Figure 3-3. Three transmittance spectrum as a function of applied counter electrode voltage, i.e. 0V, +0.4V and -0.5V. 46

PAGE 47

S1 S2 Figure 3-4. Density of states (DOS) of (12, 8) single walled carbon nanotube9 9 Reproduced in part with permission from [Z.Wu et.al, Science 305 1273-1276 ] Copyright [2004] AAAS 47

PAGE 48

Figure 3-5. The transmittance of the nanotube fi lm as a function of the applied gate voltage, from +1.8V to -1.8V. The transmittance at S1 peak (1676nm) increased from 44% to 92%, while the S2 peak (932nm) varied from 51% to 68%. The IR at 3080nm decreased from 97% to 75%, at opposite direction to the S1 peak modulation. 10 10 Reproduced in part with permission from [Z.Wu et.al, Science 305 1273-1276 ] Copyright [2004] AAAS 48

PAGE 49

Figure 3-6. The resistivity of the nanotube f ilm as a function of th e applied gate voltage11 11 Reproduced in part with permission from [Z.Wu et.al, Science 305 1273-1276 ] Copyright [2004] AAAS 49

PAGE 50

Figure 3-7. The transmittance at S1, S2 and M1 depends on the gate voltage. 12 12 Reproduced in part with permission from [Z.Wu et.al, Science 305 1273-1276 ] Copyright [2004] AAAS 50

PAGE 51

Figure 3-8. Time drive of the S1 transm ittance peak with +/1V gate voltage. 51

PAGE 52

52 CHAPTER 4 OHMIC CONTACT COUPLING CARBON NANOTUBE FILM TO P-GALLIUM NITRIDE We have demonstrated that the SWNT film is a new type of conducting, transparent, Fermi level shiftable electrode material. Th ese features of the SWNT film lead us to test their ability to incorporate into optoelectronic devices as tr ansparent electrodes th rough which the charges transferred across the interface of the nanotube film and the semiconductors. It is well known that when two distinct materials come into intima te contact with each other, there will be energy barriers across their interface. Such energy barr iers possess a height and a width (depletion width, into both side of the physical interface). To first order, in the Schottky-Mott model the height will be determined by the energy band alig nment of the two materials while the width of the associated depletion region on each each side will be governed by the height of the barrier and the carrier density of that material. For meta l-metal contact, since the electron density is so high, the barrier will be negligible because the barrier width is very thin at both sides to tunnel through. In metal semiconductor contacts in co ntrast the carrier concentration of the semiconductor is so much lower compare to that of the metal, that the depletion width on the semiconductor side can be very wide (up to microns), resulting in an in surmountable barrier to charge transport across the juncti on. Schottky barriers can be usef ul for example as in Schottky diodes as well as Schottky FETs But for most semiconductor de vices, an Ohmic contact (with negligible barrier height and/or width) is preferred since the efficiency of charge transfer across the metal semiconductor interface is of priority. Ohmic cont act causes less energy to be dissipated at the junction in terms of charge injection or extraction. Wide band gap semiconductors, such as GaN ar e very useful in making visible LED as well as high temperature operation applications. Th e band gap of the semiconductor is wide such that the p-band of the semiconductor is far away (g reater than -7eV) from the vacuum level. The

PAGE 53

work functions of most metals are less than 5.5eV, which makes it difficult to make ohmic contact with the p-band, re sulting in a barrier for charge transf er into the p-band. There is a need for a p-type transparent electrode which can ohmically couple to p-GaN. In this chapter, a GaN LED device was demonstrated by using SWNT film to couple to the p-GaN side of the LED. The contact resistance of the nanotube film to p-GaN versus the conve ntional metal c ontact to the pGaN was also compared. 4.1 GaN Light Emitting Diode Background Gallium nitride (GaN) is a wide band semiconductor with a 3.4eV band gap and 4.1eV electron affinity. This means the valence band lies at 7.5eV. Since the p-GaN is p doped its Fermi level lies near this valence band edge m eaning that Ohmic contact requires a metal with a work function approaching the va lence band edge. Most metals however have a work function less than 5.5eV. This remains a technical challenge for wide band gap semiconductors. It is well known that if the contact barrier between me tal and semiconductor is high, it requires higher voltage to overcome the high contact barrier, resulting in overheating, low efficiency and electromigration. An approach to make ohmic contact to semiconductors is by heavily doping a thin buffer layer right at the interface to reduce the barrier width, which will enhance the probability for charges tunneling through the barrier. This additional layer however adds cost to the device fabrication process. A wide variety of metals and alloys have b een explored for making low contact barrier and therefore low contact resistance contact to p-GaN. These inclue Ni/Au, Ni, Au, Pd, Pd/Au, Pt/Au, Au/Mg/Au, Au/C/Ni, Ni/Cr/Au and Pd/Pt/Au (26-28). Typically Ni, Pt, or Pd is the metal in direct contact to p-GaN and the structure will be annealed at 400-750oC. This will produce a contact resistance in the range of 10-1-10-3 cm2. Higher annealing temperatures can degrade metal contacts, usually because they will react with GaN to form metal gallides. 53

PAGE 54

54 In order to observe the light produced in the LED structure, transparent conducting contacts are needed. Indium tin oxide (ITO) has been a candidate however, ITO shows rectifying behavior on p-GaN, and is more co mmonly used as the n-GaN contact. One of the failure modes for the electronic device is the electro-migration of the metal contacts into the semiconductor. Typically, it requires 1-3eV to re move one metal atom from the bulk, compared to 7eV to remove one carbon atom from nanotubes tightly bound lattice. Furthermore, the entire carbon nanotube is simply too large to migrate. T hus, with SWNT film as the electrode, there will be no electro-migration problem. From chapter 2, we already learned that the thin nanotube film is transparent all the way from UV to infrared. The GaN LED will emit light at 434nm, where the transmittance for a 100nm thick carbon nanotube film is ~60% (thinner film will be more transmissive but at the price of lower conductance). 4.2 Experiments Details for Fabricating SWNT Film Based GaN LED This research was carried out with collabora tion with Dr. Lee, Dr. Ren and Dr. Pearton from material science and engine ering department, University of Florida. First, metal organic chemical vapor deposition (MOCVD) method was used to grow n-GaN followed with p-GaN on c-plane sapphire substrate. A 100nm thick carbon nanotube film was deposited on top of p-GaN, covering the entire surface of th e sample. Once the nanotube film was attached to the p-GaN, the assembly was first dried and annealed at 600oC for 6 hours in argon to remove any residual cellulous ester. Thermally deposited Ti/Al/Pt/A u, Pd/Au, or Pd was patterned by a standard ebeam lift-off process and a mesa was formed by Cl2/Ar inductively coupled plasma ethching to expose the n-GaN side of the device. The nanot ube film was simultaneously patterned by this standard e-beam lithography proce ss. Annealing of the device was carried out under nitrogen

PAGE 55

environment for one minute, at 700oC. The whole structure of the diode is illustrated in Figure 41. 4.3 Results 4.3.1 Contact Resistance In order to investigat e the contact resistance of the carbon nanotube film/ p-GaN interface, the carbon nanotube film was deposited on si ngle layer p-GaN on sapphire substrate. The transmission line method (TLM) was used to de termine the contact resistance of the nanotube film/p-GaN contact resistance (29). The control devices of the conventional Ni/Au metal contacts on p-GaN were also fa bricated. Upon fabrication, the c onventional metal contact is nonOhmic, while the as deposited carbon nanotube f ilm contact was Ohmic with a resistance of 0.12 Ohmcm2. After 500oC annealing in nitrogen for one minute of rapid thermal anneal (RTA), the resistance of the conventional metal became 0.033 Ohm cm2, while for the carbon nanotube film structure after a 700oC RTA the contact resistance was 0.011Ohm cm2, a factor of 3 smaller than that of the conventional metal. This data is summarized in Table 4.1. The annealing temperature was optimized for least contact resistance for the di fferent devices. It is worth while to note that the carbon nanotube film itself is very conducting. There is another layer of metal film at the corner of the nanotube film, which makes the electr ical connection to the de vice, as illustrated in Figure 4-1. The contact resistance between the metal film and ca rbon nanotube is negligible, two orders of magnitude smaller (after annealing) compare to the contact resistance between nanotube film and p-GaN. The measurement of th e contact resistance of as deposited metal Ti/Al/Pt/Au on the carbon nanotube is 5.3X10-3 cm2 reducing to 1.3X10-4 cm2 after the 700oC RTA. The lowest contact resistance of the conven tional metal was achieved at the annealing temperature of 500oC, while the lowest contact resist ance of the nanotube film/p-GaN was 55

PAGE 56

achieved at 700oC, The latter temperature was tried with the metal contact but at 700oC, the conventional metal was severely degraded throug h the reaction of the me tal to the GaN. Thus, excellent thermal stability is another advantage of the nanotube film, an important feature in order to achieve the lowest contact resistance which also adds simplicity to the whole device fabrication since it can tolera te higher temperature required for n-GaN patterning and RTA treatment. 4.3.2 I-V Characteristic of Carbon Nanotube Film/p-GaN In order to investigate whether different me tal contacts to the nanotube film will change the overall LED performance, we measured I-V characteristics of the j unction with different metal contact pads. In all cases, the junctions sh owed rectifying effects, as expected. They all show similar forward and revers e current. However as shown in Figure 4-2 there were subtle differences depending on the metal used, likely due to a Fermi level equilibration between the nanotubes and the distinct metals used. The GaN LED emitted blue light at 434nm, as shown in the spectrum of Figure 4-3 and a picture of the actual device gl owing light in Figure 4-4. 4.4 Discussion First of all, why will the RTA process impr ove the contact in both cases? For the metal semiconductor interface, contact resistance impr ovement upon rapid thermal annealing results from the surface atoms will gain ing enough energy at high temperat ure to rearrange themselves and lower the overall surface energy. On the ot her hand, for nanotube f ilm/p-GaN, it is very likely that upon rapid thermal annealing, the surf ace atoms of the GaN rearrange themselves and more carbon nanotube come in intim ate contact to the p-GaN, so that the actual contact area between nanotube film and p-GaN increases hence lowering the contact resistance. 56

PAGE 57

Secondly, the Fermi level of the carbon nanotube lies at about 5eV, which is 2.5eV away from 7.5eV, where the p-GaN valence band sits whic h should result in a large barrier. Hence the question becomes why do nanotube films provide a good ohmic contact p-GaN? The most likely reason is that the carbon nanotubes posses a low density of states permitting large shifts in their Fermi-level. As shown in Chapter 3 under rela tively low voltages (1.8 V) the nanotube Fermi level could be shifted 0.9eV. U pon applying an appropriate voltage, the Fermi-level shift lowers the contact barrier for hole injec tion into the p-GaN. Note that the nanotubes provide a very unconventional metal in this respect. For conventional metals, which have three dimensional density of states, and very high carrier density, th e shift of the Fermi level at low biased voltages is negligible. 4.5 Conclusion Low contact resistance was achieved by using single walled carbon nanotube films coupled to the p-GaN. The SWNT films thermal stability simplifies the LED manufacturing process by allowing the n-GaN RTA process with the SWNT film in place. The contact resistance is 3 times smaller than the conventional metal after 700oC RTA in N2 for 1min. This SWNT film provides a new class of p electrodes to all the high work function p type semiconductors. The tight bonding of the carbon atoms in the nanotube implies that the principle cause of lifetime degradation in these devices: electromigration, will simply not occur when nanotubes are used for electrical contact to the devices. The mechanism we propose for SWNTs making ohmic contact to high work function semiconductors is due to the limited density of states of the one dimensional carbon nanotubes. This enables one to shift the Fermi level up by several electron volts upon applying a small voltage, to accommodate the larg e p-band of the semiconductor. The side wall of the nanotube 57

PAGE 58

prevents any covalent bonding be tween nanotube to the semiconductor, which allows this Fermi level shifting to happen without so called Fermi level pinning. 58

PAGE 59

Table 4-1. Contact resistances of Ti/Al/Pt/Au on carbon nanotubes, carbon nanotube film on pGaN, and standard Ni/Au on p-GaN13 Specific contact resistance ( cm2) Structure 700oC, N2, 1min annealing As prepared (no annealing) Ti/Al/Pt/Au on carbon nanotube film 1.31X10-45.4X10-3 Carbon nanotube film/p-GaN 0.011 0.12 Ni/Au/p-GaN 0.033 after 500oC annealing Non-Ohmic 13 Reproduced in part with permission from [Lee, K. et.al, Nano Lett. 4 (5), 911-914] Copyright [2004] American Chemical Society 59

PAGE 60

GaN Ti/Al/Pt/Au SWNT 3.4 7.5 4.1 Figure 4-1. Schematic view of the GaN based light-emitting diode using SWNT film as the pOhmic contact.14 14 Reproduced in part with permission from [Lee, K. et.al, Nano Lett. 4 (5), 911-914] Copyright [2004] American Chemical Society 60

PAGE 61

Figure 4-2. IV characteristics of GaN LED with different metal, i.e. Ti/Al/Pt/Au, Pd only and Pd/Au contact on carbon nanotube fim. All the devices are using carbon nanotube film as the p-GaN contact. 15 15 Reproduced in part with permission from [Lee, K. et.al, Nano Lett. 4 (5), 911-914] Copyright [2004] American Chemical Society 61

PAGE 62

Figure 4-3. Emission spectrum of the GaN LED with injection current 0.1mA16 16 Reproduced in part with permission from [Lee, K. et.al, Nano Lett. 4 (5), 911-914] Copyright [2004] American Chemical Society 62

PAGE 63

Figure 4-4. Picture of the visi ble emission from the GaN LED with carbon nanotube film as the p-GaN contact electrodes.17 17 Reproduced in part with permission from [Lee, K. et.al, Nano Lett. 4 (5), 911-914] Copyright [2004] American Chemical Society 63

PAGE 64

CHAPTER 5 ORGANIC LIGHT EMITTI NG DIODE WITH SWNT FILM AS ANODE 5.1 Introduction We have demonstrated the ability for the SW NT film to ohmically couple with inorganic semiconductor GaN. It is natural to test if SWNT f ilm will be a good candidate for coupling to organic semiconductors, since orga nic optoelectronic devices have drawn considerable attention due to their ease of proce ss and low cost (30-33). Indium Tin Oxide (ITO) is presently the transp arent electrode used in most OLED devices. However as a rigid oxide material ITO has limite d flexibility before it breaks hence it excludes the possibility of flexible/foldable displays. Ca rbon nanotube films in contrast (deposited on flexible substrates, such as Poly ethylene terephthalate, or PET) can be bent indefinitely without damage making them excellent candidates for flexible display applications. Additionally the SWNTs possess an extreme chemical tolerance making them compatible with virtually any process step that might be re quired in OLED fabrication. 5.2 MEH-PPV In order to test if SWNTs integrate well with organic light emitting devices, it is wise to choose a widely used organic material to star t with. After consulting with Dr. Reynolds from Chemistry Department of University of Fl orida, Poly[2-methoxy-5-(2'-ethyl-hexyloxy)-1,4phenylene vinylene] (MEH-PPV) was chosen as th e candidate polymer in our first trial OLED devices. Because MEH-PPV has been fully studied, it is easy to process, commercially available, air stable and its electroluminescent em ission is in the visible range (32). The structure of MEH-PPV is shown in Figure 5-1, along with the illustration of the layer by layer OLED structure. The MEH PPV initiall y used for this study was synthesized by Dr. 64

PAGE 65

Reynolds group. Later we also used a batch purc hased from Aldrich. This work was done in close collaboration with Dr. Jeremiah Mwaura from Dr. Reynolds group. 5.3 Experimental Details MEH-PPV was dissolved with dichloroethene over night at room temperature, with a concentration of 5mg/ml. The solution was constantly stirred with a stir bar to help fully dissolve into the polymer. Before using the dissolved polymer solution, it was filtered with a syringe filter of 5 m pore size to remove un-dissolved polymer a nd large impurity partic les. After filtering, the solution was spin casting ont o nanotube film at 1000rpm for 45second, which resulted in a film with thickness in the range of 60-100nm. The MEH PPV film thickness was estimated based on the thickness of the same polymer concentr ation prepared under the same conditions on an ITO substrate. The nanotube film surface is much rougher compare to that of ITO/PEDOT PSS, which made the polymer film thickness measurem ent difficult. The MEH PPV film thickness on SWNT will be discussed further when we get to the wetting issue of MEH PPV on SWNT. A SWNT film with a 50nm thickness was depos ited onto a one inch square microscope glass slide substrate. All the substrates were cl eaned with acetone and methanol and then rinsed with DI water prior to deposition of SWNT films. At the same time, control devices with ITO as the anode were fabricated side by side as the SWNT film based devices. ITO coated glass was plasma cleaned and spin coated with 100nm of PEDOT PSS as the buffer layer, baked at 150oC in vacuum oven for 2 hours to remove excess water from the PEDOT PSS layer. After spin coating the MEH-PPV, a cathode co nsisting of 10nm of calcium and 150nm of aluminum was thermally evaporated onto the MEH-PPV through a shadow mask. Before the evaporation of the cathode, the sample was vacuum degassed in the vacuum (2 10-7torr) deposition chamber over night to allow the solvent to fully evaporate. The area of the cathode (through the shadow mask) defines the area of th e working OLED. The illustration of the OLED 65

PAGE 66

is shown in Figure 5-1. Once the device was taken out of the vacuum chamber it was immediately coated using 5minute epoxy to prev ent oxidation of the calcium electrode greatly shortening the device life time. 5.4 Discussion Since the working polymer is on the order of 100nm thick, the cleanli ness of the substrate and the polymer solution are crucial to the succes sful functioning of the devices. Dust particles are usually on the order of a micrometer. We fabr icated 8 pixels on one gl ass substrate, only one out of 8 pixels worked reasonably well, but with a dimmer glow than the ITO control device. The only working pixel was also nonuniformly illu minated, visible to the naked eye when it was turned on. The control ITO devices were much bette r: bright with all devices working. The fact that there is one pixel working meant that the energy band alignment of the carbon nanotube film with the MEHPPV was not the major problem. The nonuniformity in the light of the working pixel indicates there may be intrinsic nonuniformity to start with in the na notube film. This was confirmed by the fact that the cu rrent is higher than that of the ITO device under the same bias voltage but with lower light emission. We performed transmission electron micr oscopy (TEM, JOEL 2010F) on the SWNT films finding large micron scale particles. Furthermore using the energy dispersive X-ray spectrometer (EDS) to analyze the chemical component of the particles, it was found that these were principally made of principally calcium and magnesium. There we re also smalled particulates made of cobalt and nickel, the catalysts used in the growth of the SWNTs. The calcium and magnesium contamination was initially mysterious until it was recognized that these were most likely dust particles from the laboratory cement walls and floor, since th ese SWNT films were prepared in the ambient laboratory environment. Additional cross flow filtration was done to the 66

PAGE 67

stock nanotube solution to remove more of th e residual catalyst particles. To avoid the dust particles, a class 100 clean room was constructed in the laboratory for manufacture of the SWNT films. Once the SWNT films became cleaner, the fraction of nanotube based OLEDs that emitted any light was greatly increased (~5 out of 8 pixe ls vs. less than 1 out of 8 previously) though the performance of most of the nanotube devices was poor. The performance of a typical ITO/PEDOT PSS OLED device current and radiance as a function of the applied voltage is shown in Figure 5-2 (these did not vary much from device to device). The performance of one of the best performing SWNT film based OLEDs is shown in Figure 5-3. Note that under the same applied voltage, the radiance of the SWNT film based devi ce was roughly one order of magnitude less than the ITO/PEDOT PSS based de vice. A possible explanation for this poor performance of the SWNT based devices is that while the steps taken above to improve the purity of the films helped, there were still particles observed in the SWNT films. While not generating dead shorts completely killing the major fraction of the devices these particles generated high current regions that drew down th e local potential across the other regions like a voltage divider circuit. Alternatively or perhaps coinciding with this is the possibility that the nanotubes were too efficient at hole injection into the devices. The resulting electron hole imbalance then responsible for the poorer perfor mance. The energy band alignment of the OLED system is shown in Figure 5-4. We can see that the alignment of the PEDOT PSS to the p-band of the MEH-PPV is better than that of the SW NT film. But since we have demonstrated the ability of shifting the Fermi level of SWNT film this mismatch in the energy band only requires slightly higher applied voltage. On the other hand the most efficient OLEDs should have balanced electrons and holes injected from bot h electrodes and recombine in the middle of the 67

PAGE 68

electroluminescent polymer layer away from the luminescence quenching anode and cathode. Most polymers, including MEH PPV, have much better hole mobility th an electron mobility (35). If we have indeed improved the hole inj ection from SWNT, say, much better than the electron injection, that will result in lu minescence quenching at the cathode. 5.4.1 MEH PPV does not Wet with SWNT Further study revealed that the MEH PPV does not wet the SWNT surface. In order to take advantage of the high surface area of the SWNT film, MEH-PPV was diluted to roughly 1/10 of the original concentration. More over, the diluted MEH PPV solution was soaked for 30 seconds to allow better penetration of the solution into the nanotube network, since the pore size of the nanotube film was on the order of tens of nanome ter. The spin casting procedure was repeated for 10 times to compensate the diluted solution in order to obtain same polymer film thickness. After spin casting, the MEH-PPV film looks nonunifo rm in color to the na ked eye and the device performance was poor. In order to find out what was the cause of this nonuniformity, a regular concentration solution was used and one drop of MEH-PPV so lution was put on top of SWNT film. This should gave a much thicker film since during th e spin casting process, the majority of the solution will be spin off the substrate. When th e solvent began to evaporate, the wetted drop area became smaller, leaving behind no sign of MEH-PPV coated on SWNT film from regions where the drop had wetted the film. Since the MEH PPV ha s orange color, it can be easily identified if there is a coating on the nanotube film. When the solvent had fully evaporated, the MEH-PPV had all accumulated at the center of the drop with a much smaller area than the initial MEH-PPV solution drop area. This leads to the conclusion that MEH PPV polym er itself has little affinity for the carbon nanotube surface. Poor wet ability will result in less contact area between the two, causing charge injection problems. 68

PAGE 69

5.5 Conclusion SWNT film as anode for making MEH-PPV organic LED was successfully manufactured and compared with ITO/PEDOT PSS based devices. The radiance of the nanotube based device was one order of magnitude lower than the IT O/PEDOT PSS based device. The possible reason of wet ability of SWNT film to MEH PPV has been suggested and further work is necessary to determine the cause of the poor performance. 69

PAGE 70

PEDOT PSS ITO MEH PPV Glass Al Ca MEH PPV SWNT Glass Ca Al Figure 5-1. A) Structure of ME H-PPV (top) and B) Illustrati on of the OLED device with ITO/PEDOT PSS as anode (middle) C) OLED device with SWNT film (bottom) 70

PAGE 71

3mm Figure 5-2. ITO/PEDOT PSS based OLED radiance and current vs. voltage 71

PAGE 72

Figure 5-3. SWNT film based OLED radiance and current vs. voltage 72

PAGE 73

73 5.3 5.2 Ca (2.9) Vacuum level ITO / PETDOT PSS Al ITO Al (4.3) 3.2 Ee SWNT shiftable MEH-PPV Figure 5-4. Illustration of the energy band alignment of the OLED devices.

PAGE 74

CHAPTER 6 CARBON NANOTUBE/SILI CON HETEROJUNCTION 6.1 Introduction about Schottky Barrier The metal semiconductor heterojunction plays an important role for all semiconductor devices. The transport across the metal semiconduc tor junction is determined by the contact barrier height. For ideal situation, without surface states, th e barrier height is given by the difference of the metal work function and the electron affinity of the semiconductor (for n-type semiconductor), this is called the Schottky-Mott limit (37). The flat band model illustrates the contact barrier height in Figur e 6-1. When the metal and semiconductor come into intimate contact, the Fermi level across the interface will be lined up. Si nce the semiconductor has higher electron affinity in Figure 6-1, electrons will fl ow from semiconductor to the metal until thermo dynamic equilibrium was achieved. The Fermi level in the semiconductor side will be lowered by an amount equal to the difference of their work function. The negativ e charge transferred from the semiconductor will build up in the metal side and the equal amount but opposite sign of the charges will be built up in the semiconductor side Since the density of electronic states of the semiconductor usually is lower compare to that of the metal, this positive charge will be distributed over a region called depletion region (since oppositely ch arged majority carriers are neutralized there). The width of this depletion region will be determined by the doping level of the semiconductor and the amount of charge tr ansferred. A small barri er height and narrow depletion region results in what is called Ohmic contact. A la rge barrier height and large depletion region will give rectifying behavior. Fo r simple electrical contact purposes, an Ohmic contact is preferable for better transport across the junction. A cont act barrier in contrast will consume power and generate heat. As the devi ce size shrinks and device density increases in integrated circuits, heat di ssipation becomes a problem. 74

PAGE 75

According to experimental measurements, fl at band model usually does not work. Most metals will form covalent bonds with semic onductor surfaces. This covalent bonding will add intermediate states in the band gap of the semi conductor. This is the reason why the contact barrier height is not sensitive to the work function difference between metals, a phenomena called Fermi level pinning (38-40), widely observed in semiconductor/metal contacts. Given the demonstrated ability to modify the Fermi level in chapter 3, the idea described here was to form a sort of tr ansistor based on modulating the Schottky barrier contact between SWNTs and a doped semiconductor (silicon). Figure 62 illustrates th e relevant features needed for such a device, in which there is a SWNT/Si interface (called the Sour ce), an Al/Silicon ohmic contact (serving as the Drain) and a strip of Pd thin film sitting on top of SiO2 that will act as the gate contact to an electrolytic ga te in electrolytic contact with the nanotube film/Si surface. Since the Schottky barrier depends on the Fermi level o ffset, if one modifies the Fermi level, the current passing through the juncti on will be modulated as well. 6.2 Experimental Details of SWNT/p-Si Heterojunction Lightly (with resistivity 210 Ohm cm) and heavily (<0.005 Ohm cm) Boron-doped silicon with thermally grown 250nm oxide was purchased from Silicon Quest In ternational and diced into 20 30mm2 pieces. The surface of the silicon substr ate was cleaned with Acetone, methanol alcohol and DI water. To constr uct the device shown in Figure 62, a thick layer of photoresist was painted with one window of 2 4mm2 and 20 7mm2 strip left unpainted, those area later will be etched with hydrofluoric acid (HF), as explained in Figure 6-3 a). The fresh etched silicon was rinsed with DI water and blow dry with nitrogen and imme diately put into a high vacuum chamber for thermally evaporating a 100n m aluminum (Al), to form an Ohmic contact with the silicon, as explained in Figure 6-3b). The Al film was split into tw o in order to check the contact resistance between Al and silicon. The re sistance varies from several ohms to tens of 75

PAGE 76

ohms depending on the doping level of the silicon. It is considered Ohmic contact due to the large contact area (~100mm2) between the Al and silicon. Also shown in Figure 6-3b) is the two Pd film pads thermally evaporated through a sh adow mask, with one pad next to the silicon window, being careful to avoid any direct contact with the bare silicon. The other Pd bar is parallel to the first one, and will be served as the gate electrode. A 50nm thick nanotube film (3*10mm) was tr ansferred across onto the exposed silicon window, with 2mm overlap the Palladium bar very next to this Si window, as illustrated in Figure 6-3c). A small well (not shown on the fi gure) made of poster putty was built around the nanotube film and the gate electrode, and several drops of ionic liquid, EMI-BTI were put into this well to make the source and gate electrode both immersed inside the ionic liquids. The electrical connections to all three electrodes (Source, Drain and Gate) are connected through the gold wire pressed with fresh cut indium dot onto the corresponding metal films. The circuit diagram is illustrated in Figure 6-4, where a source drain voltage is applied. We keep the source drain voltage as a constance vary the gate voltage to see the curre nt change as a function as the gate voltage. 6.3 Results of SWNT/p-Si Het erojunction and Discussion The I-V characteristic as a function of gate voltage for both highly doped and lightly doped silicon was measured. The source drain voltage he ld constantly at 0.1V, by changing the gate voltage from -0.7V to +0.7V, the source drain current was recorde d. For lightly doped silicon (210 Ohm cm), the gating effect is obvious. When the gate voltage was varied from -0.7V to +0.7V, the source drain cu rrent dropped from 128 A to 3.5 A, a factor of 36 times smaller, as shown in Figure 6-5. While the heavily doped sili con device source drain current reduced from 256 A to 25 A, a factor of 10 times change, less th an the lightly doped silicon device. 76

PAGE 77

In Figure 6-6, the gate voltage was held at a constant value while measuring the source drain current as a function of the source drain volta ge, both for lightly dope d silicon (top) and the heavily doped silicon (bottom). When the gate vo ltage is zero, the IV curve showed rectifying characteristic, indicating there is a barrier at the carbon nanotube silicon interface. When we apply positive gate voltage, the transport gets reduced. When negative gate voltage is applied, the current increased, both in the forward and reverse directi on. The increase in the forward direction is much more obvious. When the gate voltage increased to ne gative 0.7V, the reverse direction current gets si gnificantly increased. The gating effect is explained as the followi ng two reasons. First lets explain from the energy diagram, as illustrated in Figure 6-7. When the gate voltage is zero, there is a built in potential barrier right at the SW NT/p-Si junction because of the di fference in the Fermi level of the SWNT and the silicon. Upon applying negative gate voltage, the source is relatively positive, move the Fermi level of the SWNT downward on the Figure 6-7 energy diagram. This will make the energy alignment between carbon nanotube an d silicon valence band better, or in other words, reduce the barrier at the interface. The second r eason comes from the carrier concentration change upon applic ation of gate voltage. When th e negative gate voltage was applied, the source electrode was positive relative to the gate electrode. It draws negative ions in the ionic liquid to the vicini ty of the carbon nanotube. Since the carbon nanotube is in intimate contact with the silicon surface, that excess amount of negative ch arge near the silicon surface will increase the positive carrier concentration inside silicon locally, which will result in a thinner barrier region for better injecting charge from the nanotube to silicon and vice versa. The positive gate voltage will do the reverse to the negative gate voltage. 77

PAGE 78

After continuously applying the gate voltage for 5 hours, the source drain current reduced dramatically, as shown in Figure 6-8. This might because when the gate voltage is applied, the silicon surface will attract ionic liquid and those ions will lift the nanotube from intimate contact with the silicon surface, hence reducing the e ffective contact surface between nanotube and silicon. Double gating structure (also bei ng illustrated in Figure 6-2) was also explored, in which case another SWNT/p-Si junction replaced the Si/Al drain. The fabrication process is very similar to that of the single gate device. This is a symmetric device, with the gate on source and drain can be independently modulated. When two id entical gate voltages were applied for source gate and drain gate, the source drain voltage was kept at cons tant of 0.1V. The source drain current can be modulated more than 300 times wh en the gate voltage changed from positive 0.7V to negative 0.7V, as shown in Figure 6-10. When the source junction and the drain junction working simultaneously to be open or close so that the source dr ain current on/off ratio is much bigger than the single gate device, which is 36 times modulation for the lightly doped device. In contrast, the single junction device will have the drain (Al/Si ohmic contac t) always turned on. Finally we studied the ambient environment influence on the transport across the carbon nanotube silicon heterojunction. According to our observation, the curr ent across the nanotube film/ silicon junction will stabilize when the samp le was left in any of the following environment more than half an hour. We have tested ai r, oxygen, argon or under Vacuum. But upon ambient environment change, the current will have a s udden change, as illustrated in Figure 6-10 and Figure 6-11. In Figure 6-10, it showed the current suddenly reduces when th e device is stabilized in argon upon exposure to oxygen. While in Figure 611, when the device was stabilized in argon with a stabilized current, a bigger drop in cu rrent will occurred immedi ately upon exposure to 78

PAGE 79

oxygen. This can be explained by the oxygen playing an important role in the charge transfer process between the interface of carbon nanotube and si licon. It is very likely water vapor in air also contribute to the charge in jection across the interface (42). 6.4 SWNT/n-Si Photovoltaic Device We just demonstrated the ability to electroch emically shift the Fermi level of the SWNT by applying a gate voltage relativ e to the SWNT with ionic liqui d as the electrolyte. We can modulate the current across the SWNT/p-Si interface. On the other hand, if we investigate the SWNT/n-Si interface, there will be a bigger barrie r compare to SWNT/p-Si, according to the flat band model. Since the SWNT films are transparen t, this barrier will be useful for photovoltaic application. This is a Schottky barrier type photovoltaic device, since the SWNT film can be regarded as a conducting film. Because the Fermi level in the n-Si is highe r than that of the SWNT, silicon will donate electrons to SWNT to reach an e quilibrium state. As a result, ther e will be a depletion region in silicon near the interface. As we know, when th e light incident onto that interface, the photon will be absorbed by the semiconductor and the en ergy will be transferred as the excitonthe electron hole pair. This light absorption will ma inly happen in the silicon depletion region. The depletion region is extended from SWNT side to the silicon side. The width of the depletion region at each side was determined by the carrier concentrations of the SWNT and the n-Si. The product of the two (carrier concentr ation and the depletion width) at SWNT and n-Si side will be equal to each other. The built in electric field inside the depletion region will separate the electron and hole, driving the holes goes toward the SWNT side and electrons goes to the silicon, then passing through the ohmic contact, going out to the external circuit. 79

PAGE 80

6.5 Experimental Details for SWNT/n-Photovoltaic Device The n-Si (Phosphorus) wafer with doping level 1014 (4-20 cm) was purchased from Silicon quest international. wafer is along <111> direction and has one micron meter thermal grown oxide on top. The silicon wa fer was diced into 1 by 1 and ha lf inch substrate. A one by two millimeter window was etched on the oxide to expose the silicon. The back side of the chip was being etched and thermal evaporate with 70n m Al. The resistance between the Al and Si was measured in the order of 10 A 50nm thick SWNT film (1.5 by 3mm) was transferred onto the silicon window, followed by thermal evaporate Pd f ilm to make electrical contact with SWNTs. A second Pd bar was also deposit parallel to the first bar, to be the ga te electrode. The whole structure of the device was shown in Figure 11. The SWNT was the cathode (also being called source in this chapter) and the Al ohmic contact (being called drain) is the anode of this photovoltaic cell. With extra gate electrode sit ting next to the cathode, but was isolated from silicon by the SiO2. We will explain the gate electrode later. 6.6 SWNT/n-Si Solar Cell Results and Discussion The solar simulator was calibrated such that the light ou tput was 100mW/cm2 at the position where the sample sits. The open circuit vo ltage is 0.51V and the cl ose circuit current is 1.31mA for the 2mm2 solar cell under the AM1.5 solar simulator (100mW/cm2). The efficiency of the solar cell was 2.53% and the filling factor is 0.15. The IV curve for the solar cell was shown in Figure 12. As we can see from Figure 13, th e series resistance was calculated to be 283 ohms for the low doping PV cell, which is the major contribution for the low filling factor. Typical series resistance of a silicon p-n solar cell is less than 1 ohm (43). The series resistance is mainly due to the poor contact between the Alum inum and the gold wire, we use indium dot to 80

PAGE 81

make connection between the gold wire and th e aluminum back contact. Improvement for reducing series resistance is under wa y, including using higher doping n-Si. As we know, the Fermi level of the nanotube can be readily shifted. Since the open circuit voltage of the solar cel l was depend on the built in potential and that was determined by the Fermi level difference between the metal and the semiconductor. As we have demonstrated in the SWNT/p-Si interface, we can modulate the tr ansport across the interface by applying a gate voltage. We use the same idea to test the gate field on the solar cell. As we illustrated in the layout of the device, ther e is a gate electrode sitting parallel to the source electrode (the SWNT film). The gate electr ode was isolated from the silicon by the oxide layer underneath it. We put a drop of ionic li quid EMI-BTI on top of the SWNT and gate electrode. We apply a gate field between the gate electrode and the drain electrode This gate field is provided with a battery, with floating ground. When the gate is connected to the drain, without additional battery, we call that 0V When the gate is applied with a negative potential, we call it negative gate voltage. In Figure 6-14, we showed th e IV characteristic unde r different gate bias. As we can see, the positive gate voltage will decrease the solar cell efficiency, both decreasing the open circuit voltage and the cl ose circuit current. While with a negative gate bias of -0.3V, the open circuit voltage will slightly increase fr om 0.52V (when 0V gate bias was applied) to 0.548V. When the gate voltage increased to -0.4 5V, the open circuit voltage remains at 0.548V. Further increase the gate voltage to -0.6V, the open circuit voltage starts to drop back to 0.528V. We only tested up to this value because we need to keep the gate bias below the redox voltage. It is clear that the overall efficiency for this solar cell was best when the gate voltage is -0.45V and 81

PAGE 82

followed by -0.3V and -0.6V, all of which are better than that of the 0V. The efficiency under 0V bias is better than that of open circuit. The explanation is as follow. When the SWNT was in contact with n-Si, the Fermi level of the SWNT was lie below that of the n-Si, as shown in Figure 6-16. The electron from the silicon side will donate to the SWNT, forming a depletion region and a built in potential. That forms a Schottky barrier type solar cell. When we apply a positive gate field, the gate electrode will be positive compare to 0V situation. That positive potential on the gate electrode will attract negative ions from the ionic liquid EMI-BTI, as a result, the SWNT will be surrounded by net positive ions. The relative potential of the SWNT will be negative, which shifts the Fermi level of the SWNT up, making the built in potential for the solar cell sma ller. On the opposite situation, when the negative gate voltage was ap plied, the opposite situation will happen. This explains why negative 0.3V gate voltage will incr ease the open circuit volta ge of the solar cell. However, as the Fermi level moves further down when the gate electric field gets more negative, the Fermi level will move below the p-band of the silicon, which will form a barrier for the holes to across that barrier. This was illustrated in Figure 6-17, in which the blue line shows this situation. 6.7 Conclusion on SWNT/p-Si Hetero junction and SWNT/n-Si Solar Cell This chapter explored the SWNT/Si heterojunc tion behavior and we i nvestigated the gate effect of that transportation. The heterojunction between SWNT and p-Si was explored. The transport can be tuned by a factor of more than 300 when applied a gate vo ltage of .6V. Such an obvious gating effect was due to the following two reasons: (1) the limited available electronic stat es of the SWNT making the Fermi level shift feasible upon applying a small gate potential; (2) the porosity of the SWNT 82

PAGE 83

film allow ions get access to majority of the nanotube which reduces the distance between the counter electrode, enhan ced the gating effect. The SWNT/n-Si solar cell showed an efficiency of 2.53% and the filling factor was 0.15. The IV characteristic was changed when the gate voltage was applied, which agrees well with what happened in the SWNT/p-Si case. The gate field changed the Fermi level of the SWNT hence changed the performance of the SWNT/n-Si solar cell. The variable contact barrier was demonstrated by Lonergan et al in 1997 (41). Lonergans experiment use a hybrid of inorganic-organic, n-Indium Phosphide | pol y (pyrrole) | nonaqueous electrolyte architecture. By applying a potential through the gate electrode, the permeate network of conjugated polymer allows the electrolyte to access and enable electr ochemical tuning of the contact barrier between the n-Indium Phosphide and poly (pyrrole) for 0.6V. The disadvantage of this system is that the polymer is air sensitiv e and thus prone corrosion. So it has to be carried out in a glove box and by using nonaqueous electr olyte to eliminate the possible corrosion. Our carbon nanotube films are an air stable open, conductive network. More over, the graphene wall of the carbon nanotube is very stable and known not to easily form covalent bonding. Plus we already demonstrated that the Ferm i level of the nanotube is readily shiftable in previous chapters. All of the above make nanotube films very desirable for studying heterojunction properties. 83

PAGE 84

X d q EFm q m q EC EFs Vacuum EV Schottky Barrier Flat band model Figure 6-1. Flat band model of contact barrier and Schott ky barrier height 84

PAGE 85

Pd pad Drop of ionic liquid (EMI:BTI) Al Ohmic Contact SWNT film SiO2 Gate p-Si Drop of ionic liquid (EMI:BTI) Gate SWNT film p-Si Pd pad SiO2 Figure 6-2. Illustration of si ngle gate and double gate of carbon nanotube/Si heterojunction 85

PAGE 86

SiO2 Si a) Et ch t wo windows down to bare Si, the rest is SiO2 Si b) Deposit Pd and Al pads Al Pd SW NT Film Ionic Liquid drop c) Tr ansf er SWNT Film and add a dr op of ioni c li quid covi ng s o urce and gate electrode Figure 6-3. Three steps explaini ng the SW NT/p-Si single gate he terojunction device fabrication process 86

PAGE 87

ISD Drain A drop of ionic liqui d VSD Gate Sourc VSG ISD Sourc Gate A drop of ionic liqui d VSG VSD Gate VDG Drain Figure 6-4. Circuit diagram of the single gate device (top) and double gate device (bottom). 87

PAGE 88

Figure 6-5. Source drain current as a function of gate voltage for SWNT/Si heterojunction with heavily and lightly doped p-Silicon. Over a course of +/-0.7V gate voltage, Isd modified 10 and 36 times, for heavily and lightly doped silicon, respectively. 88

PAGE 89

-0.30-0.150.000.150.30 -80 0 80 160 240 320 I sd (A)Vsd (V) Vg= 0.7V Vg= 0.35V Vg= 0 V Vg= 0.35V Vg= 0.7V -0.30-0.150.000.150.30 -600 -300 0 300 600 900 Vg= 0.7V Vg= 0.35V Vg= 0 V Vg= 0.35V Vg= 0.7VI sd (A)Vsd (V) A B Figure 6-6. IV characteristics of the SWNT/p-Silicon heterojunction at different gate voltage A) Lightly doped p-silicon and B) heavily doped p-Si 89

PAGE 90

p-Si 4.05eV Al 4.28eV 1.12eV SWNT SWNT Figure 6-7. Flat band model of modulating the contact barrier of SW NT/Si heterojunction by shifting the Fermi level of SWNTs. 90

PAGE 91

4 8 12 16 -0.70-0.42-0.140.140.420.70 V g( V )Isd(uA) Figure 6-8. The current decrease d significantly after 5 hour of c ontinuously application of gate voltage for the lightly doped silicon de vice. Same phenomena observed for heavily doped silicon device as well. 91

PAGE 92

Figure 6-9. Double gating effect, w ith identical source gate and dr ain gate. Isd can be modulated more than 300 times by applying gate voltage .7V. 92

PAGE 93

Figure 6-10. Carbon nanotube silicon junction current in argon environment will change upon introduce oxygen or air. 93

PAGE 94

-1950 -1900 -1850 -1800 -1750 -1700 -1650 -1600 -1550 -1500Ar2 h r A r 6 h r immed.E xpos et o a ir 8 mi n 13 mi n 18m i n 23 mi n 2 8 min backt o p umpi ng withA r w ith A r,aft e r 2 min time,conditionIsd( A) Isd( A) Figure 6-11. Carbon nanotube silicon junction transpot vs different ambient condition. 94

PAGE 95

holes R SWN n-Si junct / ion Gate on insulator Al/Si back contact A drop of ionic liqui d I + load electrons Figure 6-12. Circuit diagram of the boot-strap PV device Gate Voltage Pd pad Al Ohmic Contact SWNT fil m SiO2 Gate n-Si Am m e ter R Light 95

PAGE 96

Figure 6-13. I-V characteristic of the SWNT/n -Si solar cell, dark (t op) and under illumination (bottom). 96

PAGE 97

Figure 6-14. The series resistance was determ ined to be 283 ohm for low doping n-silicon solar cell. 97

PAGE 98

Figure 6-15. IV characteristic for the solar cell unde r different bias 98

PAGE 99

Figure 6-16. The close up of Figure 6-14, to show the open circuit voltage change under different gate bias. 99

PAGE 100

1.12eV SWN n-Si 4.05eV Figure 6-17. Flat band model of gating effect for SWNT/n-Si j unction (top) and illustration of photovoltaic effect (bottom). The blue line illustrates the over-gat ed situation, where there is a barrier generated for the hole to transfer from the silicon to the SWNT. Hole SWNT Light Barrier Electron 100

PAGE 101

Fermi level shift q B q B SWNT Figure 6-18. Energy band alignment at the SWNT /n-Si interface before (re d) and after (black and blue) gate voltage is applied. For the black and blue cu rve, we can see the built in potential is increased. However, the furthe r increase the gate voltage will push the Fermi level of the SWNT too low, to genera te a barrier for hole to transfer from Silicon to SWNT, reduce the charge harvest efficient. 101

PAGE 102

CHAPTER 7 PATTERNING OF SWNT FILM In order to use the SWNT films in devices, they will typically need to be patterned into some desired shape. For film pattern dimensions bigger than a millimeter, it is easy to cut the nanotube film on the membrane prior to transfer with scissors or a razor blade followed by transfer of the shaped film. However, for sub millimeter dimension patterning, it is necessary to develop some technique to form the desired patter n. In this chapter, we describe the techniques for making such patterns with line width in the order of 100 m. Standard lithography and ebeam lithography method to produce fine line width down to 50 nm has been demonstrated by another group (43). The majority content of this chapter was published as a US patent in 2007 (44). 7.1 Patterning of SWNT Film to Sub-Millimeter The idea is to pre-pattern the filtration membra ne within a manner that blocks the pores of the membrane in the inverse pattern desired for the nanotube film, such that the nanotubes, when they are vacuum filtered to the surface of the membrane only deposit in the non-occuluded regions. As one implementation of this we used a solid ink printer (X erox Phaser 8400) to produce the pattern in the membrane. An example of this is th e interdigitated pattern shown in Figure 7-1, with lin e width about 100 m and comparable spacing between the interdigital fingers. The printers ink is a th ermoplastic, which is a combination of waxes, polyethylene and proprietary compounds whos melting point was determined by experiment, to be about 90oC. To form a patterned nanotube film the inverse pattern of that desired is firs t printed transparency sheet (Phaser 840/850 standard transparency film). Transparency sheet is preferred to paper because the wax spreads more on paper decreasi ng the resolution of the resulting pattern. The transparency film is then placed (with printed pa ttern face down) into contact with the filtration 102

PAGE 103

membrane, a cellulose ester membrane (Millipore, VCWP), on a hot plate set at a temperature just above the melting point of th e thermoplastic ink. The thermopl astic ink melts and is partly transferred into the pores of cellulose ester membrane, filling the pores in the pattern that was printed on the transparency sheet. The hotplate is turned off allowing the wax to solidify and the membrane and transparency sheet are peeled apar t. From here the film fabrication process and the transfer to the desired substrate proceeds as normally. During the filtration the nanotubes are not drawn to those part of the membrane where the pores have been plugged by the wax and in fact the hydrodynamic force sweep the nanotubes away from those region to deposit only in those regions where the pores have not been blocke d. This can result in partial alignment of the nanotube / bundles along the long direction of the edges between the blocked and unblocked regions of the membrane. If the unblocke d regions are narrow lines of order 100 m the nanotubes can end up preferentially aligned along the long direct ion of the line enhancing the conductivity of the film along that direction. Since most app lication for the patterned nanotube film will be to serve as electr odes for devices, this technique will result in the enhancement of the conductivity along the direc tion where it is most needed. Note that the standard photo lithography, e-beam lithography or plasma etching does not provide such advantage. The limitation of this technique come s from the following two factors: 1. the resolution of the printer 2. the length of the nanotube and nanotube bundles The first one is trivial. The second one is because once the line spacing gets too small, there will be cross talk between adjacent line s. When one bundle or lo ng nanotube was drawn down onto the filtration membrane, it will lie paralle l to the membrane surface, but will be more or less random in plane. Once it lie perpendicu lar to the line directi on, it will lie across the 103

PAGE 104

spacing, even if that spacing was been plugged w ith thermoplastic and no flow rate. If the numbers of such cross-talk nanotubes are sma ll, we can pass through a high current between the adjacent pattern to burn those nanotubes awa y. But once there are significant numbers of nanotubes or nanotube bundles, as is the case when the pattern line and spacing are in the same order as the average nanotube bundle lengths (tens of micron meters in our samples), the pattern lines are no longer separate d by the line spacing. One issue of the above transfer method is that when the thermopl astic melts during the transfer process, it tends to smear around, whic h will decrease the resolution of the pattern. We solve this problem by pulling a vacuum while transferring the pattern. We clamped the membrane and PET substrate, on which sits the th ermoplastic pattern, to the filtration apparatus. By applying a vacuum while adding hot water (slightly above the melting point of the thermoplastic ink) to the top funne l. When the thermoplastic is me lting, there is a suction force to guide the melted thermoplastic to the correspond ing position of the cellulous ester membrane. The water was kept at the funnel, since the P ET substrate will not allow water to follow through. Once the thermoplastic pattern was transferred, we can replace the hot water with nanotube solution, and remove the PET membrane, to allow the nanotube film to form onto the cellulous ester membrane, with the desired pattern formed. This is a simple and cost effi cient technique to form coarse line width at the order of 100 micron meter, which is useful for many applicati ons such as display purpose. However, nanotube film can be patterned to submicron scale, whic h has been demonstrated by Dr. Ant Urals group (ref). They use standard photo lithography as we ll as E-beam lithography showed line width as thin as 50nm can be achieved. 104

PAGE 105

105 Printed Pattern (reversed) Desired Pattern Figure 7-1. Illustration of coar se patterning of interdigitated finger with line width 0.1mm

PAGE 106

CHAPTER 8 SUMMARY AND FUTURE WORK 8.1 Summary This dissertation provides a method to fabr icate single walled carbon nanotube films and pattern them down to less than 100nm thick line width. The electronic and optical properties of the film with different film thicknesses have b een studied. This SWNT f ilm provides a new class of conducting and transparent el ectrodes for optoelectronic devices, such as LED, photovoltaic (PV) device, organic LED and organic PV. An optical analogy based on carbon nanotube FET device was manufactured and successfully demonstrated by applying a small gate voltage to shift the Fermi level of the SWNT film about 0.9eV. Baking at 600oC and doping with nitric acid will de-dope and redope the SWNT film, changing its transparen cy as well as its conductivity. Several examples of using SWNT film as elect rode to build optoelectronic devices were shown. First, it was shown that by utilizing the SWNT film to form ohmic contact to p-GaN, it is possible to light up the GaN LED. Then it is demons trated SWNT film can be used as the anode to fabricate MEH-PPV organic LED. Finally, the SWNT film silicon heterojunction was studied and the transport across the heterojunction can be modulated by a factor of 300. 8.2 Future Work As discussed in chapter 5, the OLED devi ce still needs further study to improve the efficiency, as well as the solar cell effici ency improvement in chapter 6. Other possible applications include using the high surface area of the SWNT film to fabricate capacitors. Another possibility is by using the nano scale po re size of the SWNT film to serve as a filter membrane for exchanging the nano particles. 106

PAGE 107

LIST OF REFERENCES 1. S. Iijima, Helical Microtubules of graphitic carbon. Nature 354, 56 (1991). 2. Ph. Avouris, T. Hertel, R. Martel, T. Schm idt, H.R. Shea, and R.E. Walkup, Carbon Nanotubes: Nanomechamics, Mani pulation, and Electronic Devices. Applied Surface Science 141, 201 (1999). 3. S.J. Tans, A.R.M. Verschueren and C. Dekker, Room-temperature transistor based on a single carbon nanotube, Nature 393, 49 (1998) 4. A. Thess, R. Lee, P. Nikolaev, H. Dai, P. Petit, J. Robert, C. Xu, Y.H. Lee, S.G. Kim, A.G. Rinzler, D.T. Colbert, G.E. Scuseria D. Tomanek, J.E. Fischer, & R.E. Smalley, Crystalline ropes of metallic carbon nanotubes. Science 273, 483 (1996). 5. J. Kong, N.R. Franklin, C.W. Zhou, M.G. Ch apline, S. Peng, K. Cho, H.J. Dai, Nanotube Molecular Wires as Chemical Sensors, Science 287, 622 (2000) 6. K.H. An, S.Y. Jeong, H.R. Hwang, Y.H. Lee, Enhanced Sensitivity of a Gas Sensor Incorporating Single-Walled Carbon Na notube-Polypyrrole Nanocomposites, Advanced Materials 16, 1005 (2004) 7. S Chopra, K McGuire, N Gothard, AM Rao, A Pham, Selective gas detection using a carbon nanotube sensor, Appl. Phys. Lett. 83, 2280 (2003) 8. J. Sippel-Oakley, H.T.Wang, B.S. Kang, Z.C. Wu, F. Ren, A.G. Rinzler and S.J. Pearton, Carbon nanotube films for ro om temperature hydrogen sensing, Nanotechnology 16 2218 (2005) 9. M.S. Strano, C.A. Dyke, M.L. Usrey, P.W. Barone, M.J. Allen, H. Shan, C. Kittrell, R.H. Hauge, J.M. Tour,R.E. Smalley, Electronic Structure Control of Single-Walled Carbon Nanotube Functionalization, Science 301 1519 (2003) 10. M. Shim, N.W. Kam, R.J. Chen, Y. Li, and H. Dai, Functionalization of Carbon Nanotubes for Biocompatibility and Biomolecular Recognition, Nano Letters 2, 285 (2002) 11. A.Modi, N. Koratkar, E. Lass, B. Wei and P.M. Ajayan, Miniaturized gas ionization sensors using carbon nanotubes, Nature 424, 171 (2003) 12. S. J. Tans, A.R.M. Verscheuren, and C. Dekker, Room-temperature transistor based on a single carbon nanotube. Nature 393, 49 (1998). 13. R. Martel, T. Schmidt, H.R. Shea, T. He rtel, and Ph. Avouris, Single and multi-wall carbon nanotube field-e ffect transistors. Appl. Phys. Lett. 73 2447 (1998) 14. J.A. Misewich, R.Martel, Ph. Avouris, J.C. Tsang, S. Heinze, J. Tersoff, Electrically Induced Optical Emission from a Carbon Nanotube FET, Science 300, 783 (2003) 107

PAGE 108

15. A. Bachtold, P. Hadley, T. Nakanishi, and C. Dekker, Logic circuits with carbon nanotube transistors. Science 294, 1317 (2001) 16. Z. Chen, J. Appenzeller, Y.M. Lin, J. Sippel-Oakley, A.G. Rinzler, J. Tang, S.J. Wind, P.M. Solomon, Ph. Avouris, An integrated logic circuit assemble d on a single carbon nanotube. Science 311, 1735 (2006) 17. R. Saito, Dresselhaus, G. & Dresselhaus, M. S. Physical properties of carbon nanotubes. London: World Scientific Publis hing Company, Imperial College Press, (1998). 18. A. Javey, J. Guo, Q. Wang, M. Lundstrom and H.J. Dai, Ballistic carbon nanotube field-effect transistors, Nature 424, 654 (2003) 19. Z.H. Chen, Electric field induced transp arency modulation in single wall carbon nanotube ultra-thin films and a method to separate metallic and semiconducting nanotubes. Thesis, University of Florida, 2003 20. Ph. Avouris, Z. Chen and V. Perebeinos, Carbon-based electronics, Nature Nanotechnology 2, 605 (2007) 21. Z.H. Chen, X. Du, M.H. Du, C.D. Ranc ken, H.-P. Cheng, A.G. Rinzler, Bulk Separative Enrichment in Metallic or Se miconducting Single Wall Carbon Nanotubes, Nano Letters 3, 1245 (2003) 22. M. S. Arnold, A. A. Green, J.F. Hulvat, S.I. Stupp, & M.C. Hersam, Sorting carbon nanotubes by electronic structur e using density differentiation, Nature Nanotech 1, 60 (2006) 23. M.J. Earle, J.MSS Esperanca, M.A. Gilea, J.N.C. Lopes, L.P.N. Rebelo, J.W. Magee K.R. Seddon and J.A. Widegren, The distil lation and volatility of ionic liquids, Nature 439, 831 (2006) 24. A.G. Rinzler J. Liu, H. Dai, P. Nikolaev, C.B. Huffman, F.J. Rodriguez-Macias, P.J. Boul, A.H. Lu, D. Heymann, D.T. Colbert, R.S. Lee, J.E. Fischer, A.M. Rao, P.C. Eklund & R.E. Smalley, Large-scale purif ication of single-wall carbon nanotubes: process, product, and characterization. Applied Physics A 67, 29 (1998) 25. V.C. Moore, M.S. Strano, E.H. Haroz, R.H. Hauge and R.E. Smalley, Individually suspended single-walled carbon nano tubes in various surfactants, Nano Lett. 3 1379 (2003) 26. S. Nakamura, T. Mukai, M. Senoh, Ca ndela-class high-brightness InGaN/AlGaN double-heterostructure blue-light-emitting diodes, Appl. Phys. Lett. 64 1687 (1994) 27. D.J. King, L. Zhang, J.C. Ramer, S.D. Hersee, L.F. Lester, Temperature Behavior of Pt/Au Ohmic Contacts to p-GaN, Mater. Res. Soc. Symp 468, 421 (1997) 108

PAGE 109

28. L.F. Lester, D.J. King, L. Zhang, J. C. Rame r, and S.D. Hersee, Ohmic Contacts to nand p-GaN, Proc. Electrochemical Society 971, 171 (1997) 29. G.K. Reeves, H.B. Harrison, Obtaining the specific contact resistance from transmission line model measurements, Electron Device Letters, IEEE 3, 111 (1982) 30. Peter K. H. Ho, D. Stephen Thomas, Richard H. Friend, Nir Tessler, All-Polymer Optoelectronic Devices, Science 285 233 (1999) 31. Conjugated Polymeric Materials: Opportuni ties in Electronics, Optoelectronics, and Molecular Electronics, J.L. Brdas (Edito r), R.R. Chance (Editor), Springer-Verlag New York, LLC; (May 1990) 32. Terje A. Skotheim (Editor), John Reynolds (Editor), Conjugated Polymers: Processing, Devices, and Applications (Handbook of Conducting Polymers), 3rd ed., Boca Raton, FL: CRC Press, Taylor & Francis, Inc. December (2006) 33. Joseph Shinar, Organic Light-Emitting De vices, New York: Springer-Verlag New York, LLC, October (2003) 34. see www.cheaptubes.com website, Brattleboro, VT 05301 USA, April, 2008 35. G. G. Malliaras, J. R. Salem, P. J. Brock, and C. Scott, Electrical characteristics and efficiency of single-layer organic light-emitting diodes, Phys. Rev. B 58, R13411 (1998) 36. S.M. Sze, Physics of semiconductor devi ces, John Wiley & Sons, New York, (1981) 37. Raymond T. Tung, Recent advances in Schottky barrier concepts, Materials Science and Engineering: R: Reports 35, 1 (2001) 38. Leonard J. Brillson (Editor), Contacts to Semiconductors: Fundamentals and Technology (Materials Science and Pro cess Technology), Noyes Publications, Webster, New York (October 1, 1993) 39. Winfried Mnch, Semiconductor Surfaces and Interfaces, Springer, Berlin (May 11, 2001) 40. Mark C. Lonergan, A Tunable Diode Based on an Inorganic Semiconductor |Conjugated Polymer Interface, Science 278, 2103 (1997) 41. L. Valentini, I. Armentano and J.M. Kenny, Electrically switchable carbon nanotubes hydrophobic surfaces, Diamond and Related Materials 14, 121 (2005) 42. R.J. Handy, Theoretical analysis of th e series resistance of a solar cell, Solid State Electron. 10, 765 (1967) 109

PAGE 110

110 43. A. Behnam, L. Noriega, Y. Choi, Z. Wu, A. Rinzler, A. Ural, Resistivity scaling in single-walled carbon nanotube films patte rned to submicron dimensions, Applied Physics Letters 89, 093107 (2006) 44. A.G. Rinzler, Z. Wu, Low temperature me thods for forming patterned electrically conductive thin films, Publication No.:WO/2007/035838, Publication date: 29, 03, 2007

PAGE 111

BIOGRAPHICAL SKETCH Zhuangchun Wu was born at 1971, in a small village called WuWei, YiXing county, JiangSu province, China. His father is ZhiYing Wu and his mother HongFeng Huang. Both his parents did not get a chance to get education and are peasants. He has two elder sisters Guifang and Chaofang, whom he developed closed relationships with. He is a student who came from a rural Chinese education. During his high school, his physics teacher, Mr. Liangcai Lin, inspired his interest in physics and his English teacher, Mr. MinShi Yang, showed him the door into learning English. Both of these experiences played an important roles in his later career development. He managed to get into Nanjing University, one of the best uni versities in China. Although while there he only pursued a college diploma, he still had th e dream of higher education. He received a masters degree from the physics while he was working in Nanjing University as a lab technician in the Physics Department. He applied to study in the USA and was accepted by the University of Vermont. While there he received a masters degree in material science. Later, he applied to the University of Florida and worked with Dr. Andrew G. Rinzler on carbon nanotubes for 5 years. He has a daughter, Emma, who is the source of all of his jo y over the years of graduate study in Florida. 111