UFDC Home myUFDC Home  |   Help
<%BANNER%>

# Fabrication of Asymmetric Pores for Biosensors and Transport Studies

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 E20101115_AAAACA INGEST_TIME 2010-11-15T10:59:32Z PACKAGE UFE0019751_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 192 DFID F20101115_AAAVKQ ORIGIN DEPOSITOR PATH wharton_j_Page_066.txt GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
ca3ff78b008fa66136379d12dfda45ef
SHA-1
6501fb98fc3417f30b16a565aa0262ed74a9e455
1882 F20101115_AAAVLE wharton_j_Page_081.txt
ee456308594bdaa09b147d4812252ef8
90f9ca0e859139fc36a7a5dae22b8da4e800aa3a
595 F20101115_AAAVKR wharton_j_Page_067.txt
3cbc1ab78f2d89027e5fe0105f36e350
515f3f53fcc0fc40b5951d4d9ab2e5851b8bbfa3
1677 F20101115_AAAVLF wharton_j_Page_082.txt
97e056a91bf8fba583b5a111fe7ab615
296 F20101115_AAAVKS wharton_j_Page_068.txt
4f1f3b322f3176182d25afa1f762aca5086df2c9
1470 F20101115_AAAVLG wharton_j_Page_083.txt
e0ce18083d40a6823c8d3000d165b603
2130 F20101115_AAAVKT wharton_j_Page_069.txt
ec26d1a21eb75fa405324ba121277907
2012 F20101115_AAAVLH wharton_j_Page_084.txt
7d8ed27a0879766790cbb3a0feca35e0
2135 F20101115_AAAVKU wharton_j_Page_071.txt
5f35ce6cda968d6ca3388350bccd6eaf
80f92faabd522b262150cce22d39b818b7975595
2158 F20101115_AAAVLI wharton_j_Page_085.txt
67eaf4d988fa35d56701b082e0b6c300
d8251ca495106f86ec06866d6c24759b539d527c
1950 F20101115_AAAVKV wharton_j_Page_072.txt
9c534101c286fb79a11f95e7774d885f
bc17253c458014e8b3bb9f0682d4fab9a2426c37
1919 F20101115_AAAVLJ wharton_j_Page_086.txt
62853b3ab0a7b51a6d14f5103890ed78
c096b962211a069fb8a511e5d82953c286a25270
701 F20101115_AAAVKW wharton_j_Page_073.txt
6047600dd4d482c251dd71292b5aeda9
4cb1bca3c3bf067635d278a5a82edd2a8d07aacf
1993 F20101115_AAAVLK wharton_j_Page_087.txt
a8a36644d8527ffd58bf5cb0e5f535ff
55e0a5857419ffa7c8f8c14f82e3c1ba28a806fc
282 F20101115_AAAVKX wharton_j_Page_074.txt
80893dd707dfbfc356992d6bf7c1ff77
9a57664025e1b3d5df2486013b199c8a033b8314
1935 F20101115_AAAVLL wharton_j_Page_088.txt
a3333b09d18572df712b58e3ed3b2452
c0ce7934efe0ee200c2bc265a0178a426de90099
109 F20101115_AAAVKY wharton_j_Page_075.txt
da3261d064a060f600448b9b34111b967e8fc140
1570 F20101115_AAAVMA wharton_j_Page_103.txt
a23bd523d8a40f0a1ef47d8fcbf6302b
2318 F20101115_AAAVLM wharton_j_Page_089.txt
fe33c874f803b46422733aa3c436c66a
5eedec0fdfee8bda31c62a9f284892ea4032e497
377 F20101115_AAAVKZ wharton_j_Page_076.txt
55b7a78ddc7cc52a03eda896fd5903ac
f4b2d3cee602b81ae8a060bf8b9329200d3d0f92
1877 F20101115_AAAVMB wharton_j_Page_104.txt
e55a2d3c003d226184585b9539e89839
d8d7dbe249ebeb31e865ba9b45b629846ea18ebf
1727 F20101115_AAAVLN wharton_j_Page_090.txt
d44633d887380cfb84c624693a9481e6
dc33a8f34e38947d3798c93ab524f64a5dc2c04d
2367 F20101115_AAAVMC wharton_j_Page_105.txt
c2c0e1e40f17fc507588dbbc0ac25283
b4ceb105c3d72d5cc50a39e981e7466cb25d098e
544 F20101115_AAAVLO wharton_j_Page_091.txt
1df1a601b7f4efa1d96ba0975d41a59a
becc6c2c7f93e798a6fbec9a84306c2dd1e284fc
2246 F20101115_AAAVMD wharton_j_Page_106.txt
302f5da5cb708a92e3a3605c26e25155
e2efc1846bb3e30e7dd74f3cbb58fe9b0640872b
587 F20101115_AAAVLP wharton_j_Page_092.txt
e3da29bc747ee353c250b4227f8a2154
c4d3ae76f27ea0c63f583070bfa449746b213b85
2178 F20101115_AAAVME wharton_j_Page_107.txt
8aa0335c5b320198e1fe619437070dc4
f13d8259a43267e06c00a98ca631ce2807b220d8
483 F20101115_AAAVLQ wharton_j_Page_093.txt
0f06f89dbe3aeddc49255a3994da69e9
4592c2ba19810d48f8e2df3defe65d535cb7481d
2350 F20101115_AAAVMF wharton_j_Page_108.txt
401b615f7e2e7a1614bcbeba62098cde
623a2d5972e4115bdab0fcab7699863b01e39087
355 F20101115_AAAVLR wharton_j_Page_094.txt
b3cc6f446fca2537965d7a866212090c
da2e4ef73269304dbc5025b84febd5659603430a
2239 F20101115_AAAVMG wharton_j_Page_109.txt
4ba6b675b22ee9458441a5cb50c6d5a6
4eee2788976312cae061a398473ec69d9d25b326
264 F20101115_AAAVLS wharton_j_Page_095.txt
0d433b35dbdd0ae7bca51e066dfaf37a
099e1f8af6eefd9cac330bf0609deae76f9c155c
2126 F20101115_AAAVMH wharton_j_Page_110.txt
8efdbf38378bb185fa6a1644ba18129f
8f158298ea4f5cf371a8506511c6f2379c100d1e
711 F20101115_AAAVLT wharton_j_Page_096.txt
b536e9aaf37442817a2637b4968905ab
ae8d894159d2e9fdf8fd01c45436b5aed9917000
961 F20101115_AAAVMI wharton_j_Page_111.txt
c60b69b729f3bae2f589e6a5b94836ff
cb0e3451fd67f426f12ce781c21af9747dd26eae
407 F20101115_AAAVLU wharton_j_Page_097.txt
52a90a70674cb7b96516cec380bc8464
295fd3f1c484a46629df3885fa5d3f7dc2c5f997
4482692 F20101115_AAAVMJ wharton_j.pdf
0f2f38ba956fbae4e5e98c50623eba02
360 F20101115_AAAVLV wharton_j_Page_098.txt
d91cf22320f644dbdf710dc574eca889
fa4b4c002845305957f20f5fb02cb1b3dbf0d4a6
2177 F20101115_AAAVMK wharton_j_Page_001thm.jpg
0d61d6ede428b861166c67ed838103ac
a2711ae0f74834ddc7ff8fc7846fc24189541d8e
683 F20101115_AAAVLW wharton_j_Page_099.txt
576bf4b3ce8bbcb8f1eecc6f6ca2cf4aa3a862b5
8153 F20101115_AAAVML wharton_j_Page_001.QC.jpg
2f3bf5aba1fa5da418649f204aa78d9e7a2a202c
870 F20101115_AAAVLX wharton_j_Page_100.txt
d0bf89063ac8aa949ba944373c885845
99aea28f92dba0c15eb2a48a5a95d2b0982b23ac
28337 F20101115_AAAVNA wharton_j_Page_010.QC.jpg
b35217863bf1eb7f2fe9d4d9a90dd908
7128057eb7537134875124c06833e526980b4049
561 F20101115_AAAVMM wharton_j_Page_002thm.jpg
2577ac7b5c0830eb10ab5e363ccf6174
15c8ca26c5ffd5fd202e36fb51c56391d67ae8cf
634 F20101115_AAAVLY wharton_j_Page_101.txt
1a71b58e832b5f28f49119f7514133d1
a549a922d11bd51e1dc57822fbe54343da92ea5c
7878 F20101115_AAAVNB wharton_j_Page_011thm.jpg
8ea7f7e65fb55a3f0cfca7b838b45118
f2b054d8aca05b6cee1c534a9588fb2362e8ffcc
1508 F20101115_AAAVMN wharton_j_Page_003.QC.jpg
96f3801c59bb0a2a8edc1319feb03267
657f08fb4f31728e9081a08ed344dd05938afb3a
2134 F20101115_AAAVLZ wharton_j_Page_102.txt
64c497bb91e4aa6e64d61d363c692448
67edeb3f9635b58954373b6ba634d32eb05ed033
32048 F20101115_AAAVNC wharton_j_Page_011.QC.jpg
e64c0ee5ba5404b95584ea2287d59110
3645f86037a7b8e67f517130c15387911c4bcbca
5307 F20101115_AAAVMO wharton_j_Page_004thm.jpg
c2839b327868404e6504e9985b362a0f
b39c806e53d71996705ebf32c677c27d1b1e3ec3
35449 F20101115_AAAVND wharton_j_Page_012.QC.jpg
78b0b22fc55ebe3484fd8164f515d414
14b79fa9f3403289c7f1d686ede1c82e6b2a84ec
21476 F20101115_AAAVMP wharton_j_Page_004.QC.jpg
64c5e91435623d128fcc1bc7977569063ddb39c7
7902 F20101115_AAAVNE wharton_j_Page_013thm.jpg
da4bf4dd12b706b7722bf74f311b8b49
678ec38f2866609c1106edd58e3dd8def10a5238
6990 F20101115_AAAVMQ wharton_j_Page_005thm.jpg
7e98dae13df43177ae42ef78a57fa003
9c0993904ff72555a53e8da8c8ed1bef6af3f5be
32327 F20101115_AAAVNF wharton_j_Page_013.QC.jpg
7fcaeb44f9e6194aa4d3e12e259ee085
ceb5eb7f60c230bbdb2e4b3d34aaa458927189ac
29267 F20101115_AAAVMR wharton_j_Page_005.QC.jpg
9d2b848cc754ebdb50e35bcb1e04f463
c23977764183af47fbf854b51241f268bbef411d
8245 F20101115_AAAVNG wharton_j_Page_014thm.jpg
d937e47d4ee5cfdc6f04f798006d20abb26d2146
7821 F20101115_AAAVMS wharton_j_Page_006thm.jpg
33f38d5d3235ac779578915a2258bfd8
b0747568128cc7cbd21e0f6d01b6aabd2645b474
33260 F20101115_AAAVNH wharton_j_Page_014.QC.jpg
4d44c1ba01d29061c602d29e220bece2
03172df514ce44f7c654b758ae1296df4489c1c1
33522 F20101115_AAAVMT wharton_j_Page_006.QC.jpg
35e29d5c54d145cddf967b38e53ff151
c7fbed152cab13664a24d56397c5784fff7a5636
8959 F20101115_AAAVNI wharton_j_Page_015thm.jpg
3db75bed18e4cce00f5fbb00a2e88de2
353df7e142351b59c7c163259e37dafbe99d369c
3147 F20101115_AAAVMU wharton_j_Page_007thm.jpg
94f5ee20c9c5a092a32e120a3af78d67
c89b27f344bb1a7aece02e15701a9cf5debe49eb
37600 F20101115_AAAVNJ wharton_j_Page_015.QC.jpg
650d75621198f9bd504f1cea5b1e4f05
b087f50cc1946289beb4c1860d077f7813854b04
12135 F20101115_AAAVMV wharton_j_Page_007.QC.jpg
dbdcf427da092fb5c7832f010f853c60
8406 F20101115_AAAVNK wharton_j_Page_016thm.jpg
67279b6018b098e204b73e240a1f5491
62e2a3d6ce48354ddfcc7ce3343c4d09321fa0e3
9010 F20101115_AAAVMW wharton_j_Page_008thm.jpg
f521bf5d3ca96dff992b9a06f0c4c0c2
32699 F20101115_AAAVNL wharton_j_Page_016.QC.jpg
b4c95077078d435bec5efb553d55ae42
94a66bee697b343b3305015d0d9c1a91641688f3
37189 F20101115_AAAVMX wharton_j_Page_008.QC.jpg
21466fa974564fa406a754bdd55b7dfbc7527bf3
32584 F20101115_AAAVOA wharton_j_Page_024.QC.jpg
7744a5cebbb44d59f97a8d03c08d75f9
7928 F20101115_AAAVNM wharton_j_Page_017thm.jpg
0f16a21f75eb147b537e9b8d76d5ce48
5fd5787c7f23b533c52ecf2f3745d1c8b6185497
9765 F20101115_AAAVMY wharton_j_Page_009thm.jpg
31810 F20101115_AAAVNN wharton_j_Page_017.QC.jpg
4de845da501b8bf07e7fa2fd4002e4c5
cd9952e78a81c21a5e1c6ab014a49a6706ebc832
39965 F20101115_AAAVMZ wharton_j_Page_009.QC.jpg
ce869395371400bf4069e0317c683484
2ac7a08c13ac162a6dacca0a09a2db1c96014e0c
8355 F20101115_AAAVOB wharton_j_Page_025thm.jpg
d4d976f81ba873fa6c9457d9cb71a5e5
2f265fc5be6dd7957f89a3c8fe014069535407f4
8313 F20101115_AAAVNO wharton_j_Page_018thm.jpg
338e73d87e038e444a1c3edcf96d0f1d
6576af0aaf396505c239555415abe8cb673a9f35
33868 F20101115_AAAVOC wharton_j_Page_025.QC.jpg
ba140f127793d664307953ba95e705f6
fcfdecd08f3ec08c8d272ed0da85733f266f84a1
33681 F20101115_AAAVNP wharton_j_Page_018.QC.jpg
b7b207ae79bae0b048f5964fab9fd2b7
a567b688fed69e2ba66b8a3e0d0ab47141587d7a
7858 F20101115_AAAVOD wharton_j_Page_026thm.jpg
5c2e52bd2fafbcfc9a3cbf96c984bf9b63dfdc60
8732 F20101115_AAAVNQ wharton_j_Page_019thm.jpg
f7cef2c17d3e2725597771b844b77f65
e725b74fce5c792ef1c0bb7ef96281dd9451fe60
32520 F20101115_AAAVOE wharton_j_Page_026.QC.jpg
ecd2ebd146bcc8b887443cbafce45022
5313cdc28c78cff38b75cfa934e65587b7e2cd56
36310 F20101115_AAAVNR wharton_j_Page_019.QC.jpg
684c80bd23fa8da13911b20d136956f3
4658 F20101115_AAAVOF wharton_j_Page_027thm.jpg
0ef9b1a9c0e89c9c1346fe9481af3633
a342f029d06b631f797fd58fc18cf2ff0f4c1e9a
37119 F20101115_AAAVNS wharton_j_Page_020.QC.jpg
9cb788a32c7e9ac7dd47f81eee08712d
37779a21f086320744cfa631e4389a1b0458ecbc
13698 F20101115_AAAVOG wharton_j_Page_027.QC.jpg
8334de9ec850c693f24faa7c8e3c15fc
202cd23910d4a29623c42366a86253bee398348a
8917 F20101115_AAAVNT wharton_j_Page_021thm.jpg
61eefa8780df73a1f235e6f565e94458
538c6078fe8739f59c39e6db2383af662c19701b
5657 F20101115_AAAVOH wharton_j_Page_028thm.jpg
a37d56a1029694255a25213b7912727c
37499 F20101115_AAAVNU wharton_j_Page_021.QC.jpg
6d5b6b54f2eeb802c08b2364e3e0f58b
d5b406d401974087886e3177c6e9e0472e309657
19459 F20101115_AAAVOI wharton_j_Page_028.QC.jpg
ee60717172b4fecf6850d6c2155b72e1
8815 F20101115_AAAVNV wharton_j_Page_022thm.jpg
44ed62e3aca5acf81b07f3660668a77d
7900be4d56db97baa8b9e00b47b4e2c82e46497a
2422 F20101115_AAAVOJ wharton_j_Page_029thm.jpg
5b96110f6919845315e7e76303c844a0
a3c03d85585d8a1411717965597954c476b8fc2b
35686 F20101115_AAAVNW wharton_j_Page_022.QC.jpg
3be8f39b76d0499eea2ae5fc3a50aa0d
9f95b53f32d21a3bd08df377a70022344a3b148f
6968 F20101115_AAAVOK wharton_j_Page_029.QC.jpg
219062aa5bf0b6609b9355c14b6071c7
64cfdc7441b8e8ec03347fc2f90f484ea4ae3cd2
8865 F20101115_AAAVNX wharton_j_Page_023thm.jpg
9046affd69448c38fe632baed210f6e6
209c56ae9281e21d81fc26214b2eb5fe7cf87947
33460 F20101115_AAAVPA wharton_j_Page_037.QC.jpg
e3abbd4aab4f8753cbf53b8ced91fa12
b298ed3b1511f120a3d57f0762e6d8f78383605d
4575 F20101115_AAAVOL wharton_j_Page_030thm.jpg
f0b84a42f91deabd6147df177e908f1c
861cda99a1acc5e873b572b1ff73a44f60ed0cb4
37065 F20101115_AAAVNY wharton_j_Page_023.QC.jpg
a8425fa30653b66439ecd2a742a8018d
c0b6ca7280858241d12f09e599001d806bcd8696
8572 F20101115_AAAVPB wharton_j_Page_038thm.jpg
dd96a109dfae7ddc18acc07b6b526980
15293 F20101115_AAAVOM wharton_j_Page_030.QC.jpg
04cc9cd870916b3fbbfa06ae151600c6dd090f92
7931 F20101115_AAAVNZ wharton_j_Page_024thm.jpg
6e318df5bb5f697b48b1a4e596f11184
1baa729fe6417ba4e2c2accb97ab76e636041682
3663 F20101115_AAAVON wharton_j_Page_031thm.jpg
69ea1ddd5b1c6383ca14a6fa76b05ba3
e02cf8835ff811ebf977a2d798f7361bae676cd3
35421 F20101115_AAAVPC wharton_j_Page_038.QC.jpg
d31f169f03b9efe3853ab6af8319902e
e1072655990dab39c4ee78073c3f565fb30a15e1
10815 F20101115_AAAVOO wharton_j_Page_031.QC.jpg
2126d544e8e73e5e7307d63852bf11bc05e0b045
8812 F20101115_AAAVPD wharton_j_Page_039thm.jpg
1f17c00cd3e72b76098913c50a8605b6
5e9ffa133950784e4dc10d7cc8a5dacac1bb8404
3821 F20101115_AAAVOP wharton_j_Page_032thm.jpg
6640f5f88c24a563545f0b87a0e52863
3757ba2a3d9991ccb55814cf2a3e9bab10ee45df
36344 F20101115_AAAVPE wharton_j_Page_039.QC.jpg
71f47618ab77b24fef864ae094cd7c350d8b7df6
11440 F20101115_AAAVOQ wharton_j_Page_032.QC.jpg
c34dce953e050fb75b0276a0b798581d
b98e149322dcc1c9f572645256222fde845e6362
8677 F20101115_AAAVPF wharton_j_Page_040thm.jpg
fa6180d21fffba54e80d324b5689e056
9f845a3c6ce650e03c55d6c929cc088a20900b73
4394 F20101115_AAAVOR wharton_j_Page_033thm.jpg
e912e124f367865ef950a7cf9161d8d8
00bb9718bd57c529ccdb8009a7e085b0576d4a39
36337 F20101115_AAAVPG wharton_j_Page_040.QC.jpg
3abe3bc5d2500615ebeabe606d46e4a1
1cf8d9b60eb0f4b6a6c63a80ed02651b79029570
12085 F20101115_AAAVOS wharton_j_Page_033.QC.jpg
4e5f967195cf31024f6914d696dd3133
eff236b7d48f3f7f9da6c4b0ab27447440abbdb6
8897 F20101115_AAAVPH wharton_j_Page_041thm.jpg
2e7831fb19217dd267ccf6f31683f101
e4e52cb8af0482e22c8ba06de35dc959071ecb33
4410 F20101115_AAAVOT wharton_j_Page_034thm.jpg
353bca841be231562124bdf492334398
c96143ec11b1b19a86397da5e7e3888d43986a8c
35706 F20101115_AAAVPI wharton_j_Page_041.QC.jpg
5dfc26aa23d6856e86fcef86b24a01d6
13462 F20101115_AAAVOU wharton_j_Page_034.QC.jpg
c041566b3e9954c8afe3ccbf0f8b74af
cafd9436196e4d3040ae5b4bc1fb1bf50fd58e83
8369 F20101115_AAAVPJ wharton_j_Page_042thm.jpg
1d7d4ffbbb2dfd4a6e830e70753f7eac
83bac51ee6f483ba5217ac2170e9f32414545e46
2273 F20101115_AAAVOV wharton_j_Page_035thm.jpg
415cbb623297e6df2ca9c77bc647fbea
dc1e37a26b51550b848c529604c4ab8ac181cfed
32297 F20101115_AAAVPK wharton_j_Page_042.QC.jpg
04c59dc64c69525479a4a814d643b264
20b27c361c2f20ffc82cd92cb8d86704760958f3
7358 F20101115_AAAVOW wharton_j_Page_035.QC.jpg
b3cd125cb8721e26659447198db68cc8
38e5fd384864e8cb5543bde6143aa38999c0f862
7333 F20101115_AAAVPL wharton_j_Page_043thm.jpg
bff7f2c5e98b53a7ca73a2704931a865
eb85f40228a146b0bdfa896d5acbfd737a9f0456
8779 F20101115_AAAVOX wharton_j_Page_036thm.jpg
b252d30ebdf3b55a0b14ab7655322469
05ee225bc37ae425fa6040ab30561d4f3f2e0fcc
10238 F20101115_AAAVQA wharton_j_Page_050.QC.jpg
779bf21bc8d5feb887f6d0790010128d
0605da7a684720ca195e9c4061340f7a606151bb
28278 F20101115_AAAVPM wharton_j_Page_043.QC.jpg
3ea6b2999a5be419740744de56cdcb6b
2d0e4caa264d7c3d15133645920e1625c8f1b151
37054 F20101115_AAAVOY wharton_j_Page_036.QC.jpg
76f7dcac0e3c8ce86e5046587f54be60
80eaa21f14557e5d4ee7cb651d977c9c13d3a988
3263 F20101115_AAAVQB wharton_j_Page_051thm.jpg
4826c77b65c7572ddd7af8042cec574b
a301182b99a848e93ba1f37ddd6a5ec1dff89998
8252 F20101115_AAAVPN wharton_j_Page_044thm.jpg
baaae7c3123f5e671a38c069d92682ae
8278 F20101115_AAAVOZ wharton_j_Page_037thm.jpg
a492fb63b2224ee74a384f6f296784b1
42bda66d00348a1af686973e2b3dc941d48d4109
12215 F20101115_AAAVQC wharton_j_Page_051.QC.jpg
f7a37772dbc6d66e25a2ffe1b9acd383
a352d36088cdc740d9c8c2f5c212c0d8112e1efd
32895 F20101115_AAAVPO wharton_j_Page_044.QC.jpg
70b3d058275711c37563c4427d406185
450e0b1abb22429c4afbf28e28aa81947d94df54
7493 F20101115_AAAVPP wharton_j_Page_045thm.jpg
e4c4578c8c5421cc814530388f033a8d
930389b9a8bbc3da6c99face31817bafc7aa19d3
4069 F20101115_AAAVQD wharton_j_Page_052thm.jpg
3c55c5dc5279f055b18a208294bae724
1e62a5e48e95fe42051a0dd7e79ee5c6e8c792d8
29046 F20101115_AAAVPQ wharton_j_Page_045.QC.jpg
e04e22852796fcebfdea8a817fd199c9
63039fec61822815cf2a9d530f32df1176166316
19329 F20101115_AAAVQE wharton_j_Page_052.QC.jpg
80bbace18eb46ea389a3452344ae335a
d21a334d1af61c7d57c551f8d8842cd8bebc5c73
8593 F20101115_AAAVPR wharton_j_Page_046thm.jpg
1e27f843562ec42d066eaaf1494184a2
8a72b20b9c06b1b69be1f567df8d97fa8189e4e2
4158 F20101115_AAAVQF wharton_j_Page_053thm.jpg
43028d2669ae00a1c5baf5914b93b2c3c40ca570
35106 F20101115_AAAVPS wharton_j_Page_046.QC.jpg
af4fc04e7ebd2edeac22bdbcaa3edf72
a2ff242701bf974790e653c4991aed745d22a535
13524 F20101115_AAAVQG wharton_j_Page_053.QC.jpg
71b3da882674712a0c8a941077315056
3779864c9a51ee5b620d485a92cda3da6239df51
7249 F20101115_AAAVPT wharton_j_Page_047thm.jpg
72f63a6868dcaa83a87676e832f8a86c
620ec2ae860b5fba731ca573d499c8f355abdf92
4601 F20101115_AAAVQH wharton_j_Page_054thm.jpg
8b0f275ee1d1c10008d1e45897146556
bf0b7467a43e93c6c07f265014fba7e4734c4a26
28816 F20101115_AAAVPU wharton_j_Page_047.QC.jpg
392fc841bd0995c88fce8c7117a97710
8049f5bc232afd6b6e170a9d547ffd711452d60c
16915 F20101115_AAAVQI wharton_j_Page_054.QC.jpg
a1bddcd723a1f37759246d0eb2664476
def6d9ea52990a847060e473f6d6b0b1f94ff4f5
5254 F20101115_AAAVPV wharton_j_Page_048thm.jpg
4e801a8cbd9121308887d900c85710e8
949d15b623e112ec761c0faf98a06e53ccf93ce9
8810 F20101115_AAAVQJ wharton_j_Page_055thm.jpg
d2fccdc9ae90e0a1319ea9d3b35341e1
452f07e449d66c46770d927bd579ed60e0f1c468
17026 F20101115_AAAVPW wharton_j_Page_048.QC.jpg
9e8fac7eca7766a5bacc15a1d9fb2e22
4aebe94d21fa52a59e4871da84743bd0441a61d4
36876 F20101115_AAAVQK wharton_j_Page_055.QC.jpg
2aeff125b5b3e93ff14b4397776f244b
eb7578a84bc5d369087df95d43e1bec2f1c79781
2189 F20101115_AAAVPX wharton_j_Page_049thm.jpg
900588bb75c768d2834e215a4088e0060fcf3e9a
31248 F20101115_AAAVRA wharton_j_Page_063.QC.jpg
19bdca62e25a1a50f00ee4888a6ccf44
efb910b24a08d749ba66b9e2302e79c51b9b9b77
8559 F20101115_AAAVQL wharton_j_Page_056thm.jpg
22f7ceaab0f0b1051dae30d720e7dc26
d1f29cf43716702cb43b60d80b9d61a62e5dcac0
6423 F20101115_AAAVPY wharton_j_Page_049.QC.jpg
97bbed342b38f8b7f0b17b96121aaed7
a56cc644d6ab690ec04d9c5a40b88c5b9af41871
5132 F20101115_AAAVRB wharton_j_Page_064thm.jpg
7cff4613500d2a1896713f43be1cb473
ab11017e4e1b252b86ae324b6077b38f39f477cc
34762 F20101115_AAAVQM wharton_j_Page_056.QC.jpg
dc32118658f9673c4d954cd8467816ec
3177 F20101115_AAAVPZ wharton_j_Page_050thm.jpg
382f5c6f6cd2c6c75198fab99b743dd1
89c087f2b81877ced568ea1f7f7457b4001aef32
17564 F20101115_AAAVRC wharton_j_Page_064.QC.jpg
f84e19db52cd914f3143a623d7d5eec3b3293b19
8859 F20101115_AAAVQN wharton_j_Page_057thm.jpg
958ee5a70ca2eb0172f7dfba8d09ba64
322ef71411189b5c55f5bc5e01f37bfa931d870b
3321 F20101115_AAAVRD wharton_j_Page_065thm.jpg
726cf8abb798b30e36ec6899a9c115e201dc233e
36569 F20101115_AAAVQO wharton_j_Page_057.QC.jpg
c1a23eedc52ca5b4f40dec9155b8ce54
ce0b4b5e5258596132950a3c4c32eb351c40bd7f
8440 F20101115_AAAVQP wharton_j_Page_058thm.jpg
0bac264f4836f74859e5928b42195b12
b2df635f54a83fd98534a69ce1d62234e26e93b6
11552 F20101115_AAAVRE wharton_j_Page_065.QC.jpg
0aa9670cf7027aa758ee2cd99f9568ea
ac65cb1758eb5ab147156a6e7a1454401b013d69
32199 F20101115_AAAVQQ wharton_j_Page_058.QC.jpg
26e18753ab316eda673739b63b671496
e470337ff4a8496376c82d44810015397811bc6b
2461 F20101115_AAAVRF wharton_j_Page_066thm.jpg
e72805f9dcc611c8c91d9dd1cfbfe974
951c9baa7324208cc839b531cf9856162b131b21
8101 F20101115_AAAVQR wharton_j_Page_059thm.jpg
fe55465f920d3a27dd81c72c6ce2992c
4d87432ae9a5b0761b40e59ffd20afbfc2535ec7
6633 F20101115_AAAVRG wharton_j_Page_066.QC.jpg
b1e05c4b1c3b2666df505e6ab1998a5c
0aa8860be1c44da15995609f98c44e1f5797f0f7
31991 F20101115_AAAVQS wharton_j_Page_059.QC.jpg
1222d33e2d8f73f8dbef1a0af35f790a
5788 F20101115_AAAVRH wharton_j_Page_067thm.jpg
e663e5cda9202734c9035337315a333c
0db209a0212943f6bf6371e2ab3e2369f5cc97d1
8761 F20101115_AAAVQT wharton_j_Page_060thm.jpg
67b4160db7571b8dfbb1ea46a741b4c6
c6e1f90291b8379fbf0333eda0c7d153b1438a34
19198 F20101115_AAAVRI wharton_j_Page_067.QC.jpg
a6ded15376ba2ffd607d150cdc9b32d6
2419e9b973bf4cc93ec2b26eac67af95955d1583
36345 F20101115_AAAVQU wharton_j_Page_060.QC.jpg
2e56fe10f83214afa6009c643bedfe1b
cdc3e3c4946375a2dcae029ecdee09740e37e0d3
4665 F20101115_AAAVRJ wharton_j_Page_068thm.jpg
20af58254d85f974a056db5bc3754d7f
a4e5789077437f72b060dcdbf935192f32b1d51c
8942 F20101115_AAAVQV wharton_j_Page_061thm.jpg
163922bde9ec24bc8d0e98c0a65e3700
835746a64a9aa9bc2c8a14302b698782112ff74d
12322 F20101115_AAAVRK wharton_j_Page_068.QC.jpg
c310ac576234a7c50e9f4dae7863cfeecbeb6c2e
35292 F20101115_AAAVQW wharton_j_Page_061.QC.jpg
f95d584a5d71f1690af305cdc68c19c9
b71cdfc4b3670c92cba933be71645c7bf6ecf771
7568 F20101115_AAAVSA wharton_j_Page_076.QC.jpg
02a3e41f46176976a6774c423f01eafd
630609589621e52b397fca9dd22e20354a26d488
8603 F20101115_AAAVRL wharton_j_Page_069thm.jpg
fea6b849f2af3097021343e24137bb59
8aaf2f85a07b4c82ecbf317865ea42bf0cdbd31f
9239 F20101115_AAAVQX wharton_j_Page_062thm.jpg
19fabd48347b363aa023ab4f8c391aab
6087bd2621780cb87e5b7544ef03d8b1d6bc6862
3948 F20101115_AAAVSB wharton_j_Page_077thm.jpg
e5d2ce2b14e8be47a105757e3d251967
863dd973867dc5dbde1ba8e626b67dfe22120c64
34934 F20101115_AAAVRM wharton_j_Page_069.QC.jpg
1270322390515ec3c506b3124ecb52a4
3439f5fb7a1d2bfc3d4735d92780a6eb156e5037
35970 F20101115_AAAVQY wharton_j_Page_062.QC.jpg
b6dfe2e992a29c9e720725aa6127ef5c
003e0a3517572a3d8a0b858cda4e7acbfe308771
8877 F20101115_AAAVSC wharton_j_Page_078thm.jpg
e757e507d04fe66b5f743fc14ec1f6a4
0eb54bef3b99ff7514904d9e89ecd5e4827a4175
8388 F20101115_AAAVRN wharton_j_Page_070thm.jpg
e36483d4f18c8d8db301842a77061451
bd7193082b3299a5832f2c9f129f89650b96e9fe
8256 F20101115_AAAVQZ wharton_j_Page_063thm.jpg
6ab1f039bdb61dc1584fd45fdcd754ce
b178b8c9f9637dcb682336acebcd3d5b518807d9
36921 F20101115_AAAVSD wharton_j_Page_078.QC.jpg
78ef93a79e74c993aa0308577dd620f8
cbdf0641b25050836022523b6cee0dd74bb6b86d
34348 F20101115_AAAVRO wharton_j_Page_070.QC.jpg
4c8981692704c953c27ae5b689a09a6b
b47ab0b213576ddf38c725cbe5c16de11ec7fc93
32069 F20101115_AAAVSE wharton_j_Page_079.QC.jpg
0a94e26ddbcf014b73399036ab6d7f03
71615c3dcfcc8105e60d007fc7ecab4527cf30e1
9017 F20101115_AAAVRP wharton_j_Page_071thm.jpg
68d2628fb9d30b38d97dce08a0c311425574a764
37059 F20101115_AAAVRQ wharton_j_Page_071.QC.jpg
60465f08e404bd1dd933ac2de21b87e7
b622a409e64ff0d8cc4b87e6c2996cdc88b67022
7266 F20101115_AAAVSF wharton_j_Page_080thm.jpg
924bce04cfcff7190d60d9ceeba94a8a
2792fbd48bfe72fe9fd40e1eabe7cbbbfd3d4e31
7921 F20101115_AAAVRR wharton_j_Page_072thm.jpg
385f017f506aab10d9586a6f04dbe2dc
1e8e09a71153e6e4634f245b4e1f597ee027252b
28772 F20101115_AAAVSG wharton_j_Page_080.QC.jpg
88a301672d5a90e2f70125892405194e
e98c8fb43a72f7d9105352fbca86852b1f21c858
32461 F20101115_AAAVRS wharton_j_Page_072.QC.jpg
9a1f9a0d1b64de03738b67d78657770435d77016
7913 F20101115_AAAVSH wharton_j_Page_081thm.jpg
88114055511b489cfbe0cc37b9463aa971770442
4629 F20101115_AAAVRT wharton_j_Page_073thm.jpg
448816a4b5540667429f86076d3d94cd
ea2a5ceef7512d1d7c63d4190b51f1e2da4dce22
31048 F20101115_AAAVSI wharton_j_Page_081.QC.jpg
b02ebd87f0f2633986345b78c89104b8e796eb09
16983 F20101115_AAAVRU wharton_j_Page_073.QC.jpg
8e5d663434f1f21b29b152cca049b642
7183 F20101115_AAAVSJ wharton_j_Page_082thm.jpg
c2b213ecdbce509f1f7a0d78979c66ea
4040 F20101115_AAAVRV wharton_j_Page_074thm.jpg
27440 F20101115_AAAVSK wharton_j_Page_082.QC.jpg
de96782b42f39e10f9947a6a4d747724
efe721e436c1a3b359453b6f966203352c8efae1
10559 F20101115_AAAVRW wharton_j_Page_074.QC.jpg
19562844ebbe54a8a43cdb68965d062b
8409029c62cb7c4b9cc4da917da4d9c912515e66
6407 F20101115_AAAVSL wharton_j_Page_083thm.jpg
474959222a07de52d476d6e2957c8e64
3537 F20101115_AAAVRX wharton_j_Page_075thm.jpg
ab8243a65d2963182c0e0531b006115f
2a762fe62a9c10fa4968f44b18e44bb3f3c5a361
28544 F20101115_AAAVTA wharton_j_Page_090.QC.jpg
d7cbc7552fbf1352bd921cfce4e98f6e
024e75261342619d78fdab1e2930de28b011f8b5
23276 F20101115_AAAVSM wharton_j_Page_083.QC.jpg
6b61ce58088b683c8fa6c669a0b285ac
f4d6a1e99bdb094ae4e15ed2338594cd6f9dd91d
12518 F20101115_AAAVRY wharton_j_Page_075.QC.jpg
0daf8401f6ce78ff340b00e28553b1c3be3a8aac
13601 F20101115_AAAVTB wharton_j_Page_091.QC.jpg
a14d18de219372edf956c2cfa5082150
4d1d383c3199dc902c51ee9f9a706870a14cee53
8413 F20101115_AAAVSN wharton_j_Page_084thm.jpg
4472062b3f59489b2d9970b6edb1c4bb
f520dc0e5db1d1d445b6248cdcfcda180d4bf4a2
2436 F20101115_AAAVRZ wharton_j_Page_076thm.jpg
6d65cbf53fcee5f2b76d0aab8ab34901ee314ccc
17037 F20101115_AAAVTC wharton_j_Page_092.QC.jpg
b6399807a29d782abdfa48ee77fb06d8
ebd663bbe58e45f9c3a8b0c48b9a243dbe9422fd
35228 F20101115_AAAVSO wharton_j_Page_084.QC.jpg
eaea2d3ef7434c01cf7acc2462fde73e
8a067228d528a0d2fbf364a90cff6acc7ea9a4ae
2119 F20101115_AAAVTD wharton_j_Page_093thm.jpg
3496b13705e68f9f52bbc0bf655cf39e
beebd570804d7a19071b2b86497370af31a67d4b
8958 F20101115_AAAVSP wharton_j_Page_085thm.jpg
dab2b3673ab6b38bb0e3108584106c01
F20101115_AAAVTE wharton_j_Page_093.QC.jpg
35966 F20101115_AAAVSQ wharton_j_Page_085.QC.jpg
f3f3807fd9f8bcc3b14e05695ba4f087
d2fede7d50276c7c1dea5af88251a33693e4344b
4899 F20101115_AAAVTF wharton_j_Page_094thm.jpg
192042c024e8bbe007bde32ce8f60b34
7841 F20101115_AAAVSR wharton_j_Page_086thm.jpg
c98bcfb78a32b9c097848c85021e2b4f
c75895f584c27c22170629cf62ae4a07081dc7d9
30257 F20101115_AAAVSS wharton_j_Page_086.QC.jpg
2a1023e7bf21590c23b2fc8939eca6bc
e4581a1bd91d0a6f9ab9223813b1712f2574a78f
15369 F20101115_AAAVTG wharton_j_Page_094.QC.jpg
d9a50700192bfe08b82708fac385731a
5a4f0f9cf91f7bbb1c8b9ddd753787e99cbdb81e
8543 F20101115_AAAVST wharton_j_Page_087thm.jpg
5429bb85809d29d37dbeb725f520266c
a00c5c2bddc92974f83a1da88a7f4df9b6cc4094
4537 F20101115_AAAVTH wharton_j_Page_095thm.jpg
4f2214728fde787af17566ff9b96b00a
74eed0e205fcae48695b48b52739fe1f7632cbaa
32227 F20101115_AAAVSU wharton_j_Page_087.QC.jpg
769f393d2ac4ce6da9483e1676011f5c
16012 F20101115_AAAVTI wharton_j_Page_095.QC.jpg
de88f436439234fff17bcfb8d3ebd1ff
5ebfcc142b7139b1b17055e9239dd7191abd2b12
8173 F20101115_AAAVSV wharton_j_Page_088thm.jpg
e60580d2f60e8db4cabf7947d3d68577
f1a6b1e586a2cf5c4e52245c88154730ce97483a
5006 F20101115_AAAVTJ wharton_j_Page_096thm.jpg
33726 F20101115_AAAVSW wharton_j_Page_088.QC.jpg
896d7f0ae421136df10f2645fcc6a42f
7be8ac6a10cc8b84c5137be3d708d92238cc5892
15270 F20101115_AAAVTK wharton_j_Page_096.QC.jpg
8a2287b6fe05842f3f79c0da9aaba712
37ba0b007c5a3f2c9a26be1093c9bfdb63f8a4ab
30169 F20101115_AAAVUA wharton_j_Page_104.QC.jpg
50de6ee5a65aff5103d79679c2925e02
164f464664ca3735ec796d65c0090532f3ec8ce6
4710 F20101115_AAAVTL wharton_j_Page_097thm.jpg
688aedbe8ccf21b3186542193acfa132
c10edfab10a5217ed1edc9d3805067ee3c84cd70
9133 F20101115_AAAVSX wharton_j_Page_089thm.jpg
78172ddfa64b37ce4130e1ee6f0c823a
0f70a063c9ce848a95716b9f5dc9611bea3e1bb7
9027 F20101115_AAAVUB wharton_j_Page_105thm.jpg
0074102a581a770b488669db43403af17200a78b
13757 F20101115_AAAVTM wharton_j_Page_097.QC.jpg
9cecde896c16ae15f97a3262457972c15b299bcb
38738 F20101115_AAAVSY wharton_j_Page_089.QC.jpg
0540b296770d150f84c533d892d393ec
fde86fe38124c506d3cca8ae270a1a69b7c4dc89
2062 F20101115_AAAURA wharton_j_Page_070.txt
665a62a071b20f1bbb6f55c739b4023f
64092217c69689fe85a213fbc4f02b948f4b9b5e
37275 F20101115_AAAVUC wharton_j_Page_105.QC.jpg
b1b90a6de6f82be0db209bf4938e3d9f
b2b5e86327f6bd951df5a71256c68ed288666f9a
4072 F20101115_AAAVTN wharton_j_Page_098thm.jpg
e12a89631da4e76e7c9952a9d135bf2e
bdde062542839a10aedf2f53249335f22baaed4a
7318 F20101115_AAAVSZ wharton_j_Page_090thm.jpg
1b457ba59e1fccbb32f95df346702520
ea32273a093c851ce2151125aa816b0fc174a0e8
1895 F20101115_AAAURB wharton_j_Page_044.txt
e169e2cbf3d13a028a7bae346d0d073f
3bd4f820c69e8474a739f44a0a7969950ceb314d
9054 F20101115_AAAVUD wharton_j_Page_106thm.jpg
6e574df793f130772f1d3d41baebf603
39e894958b995b4640f65327f5710b2ff84d2498
11289 F20101115_AAAVTO wharton_j_Page_098.QC.jpg
a8b74db813e61628793b1e2a8452372c
b7e4fd6466f6e8dd57f467da8c0cbf2ac011ddff
6018 F20101115_AAAURC wharton_j_Page_092thm.jpg
b72ab42f231444a1220158c8e448d0d9
67740b729839ca0df49f09078ff1e8ca07405afa
36442 F20101115_AAAVUE wharton_j_Page_106.QC.jpg
870c03f291c82f38d9b6f565dacdf176
68f7c9cb37dfa0f253f490a5e846e568e797381c
4589 F20101115_AAAVTP wharton_j_Page_099thm.jpg
eeb25cae374606f76e9c4a41ef4f1d3b
f7b35e555cf8b699dd5c603d54fcaf50585829b5
8589 F20101115_AAAURD wharton_j_Page_012thm.jpg
b3b67f7db8a9a8f43518f5d59f0a7b17
6f6d9b6417be2cbe420e11749e18706ac825e79d
8727 F20101115_AAAVUF wharton_j_Page_107thm.jpg
a43320c0c80fc847eb27abeff70ff4a5019b5430
1053954 F20101115_AAAUQO wharton_j_Page_055.tif
2bdebdda6ef7ce565ed916fdf4e5bbac
ca3381068a6a8b8f75289c9d4ee759257d3be8a4
13437 F20101115_AAAVTQ wharton_j_Page_099.QC.jpg
5b9db05d8b187014860b3a0a6c10c274
b07530f04b70ccb52cf17fa9c1187de17ce48b06
505678 F20101115_AAAURE wharton_j_Page_100.jp2
d7dfd9def1c01e85d0604f0b7031c4e4c63c3321
35120 F20101115_AAAVUG wharton_j_Page_107.QC.jpg
ed55ba7496cb4f46538c91441c87090748a46f9b
33562 F20101115_AAAUQP wharton_j_Page_032.jpg
b9d00df832595af356bbeca62faa1d34
a03f5db50cb9bfa421be7fe4931afd0c23850c27
4825 F20101115_AAAVTR wharton_j_Page_100thm.jpg
18249f0aac22c021e778c36792b84943
6e1505782d7a32b97656145a3abd42e109aef533
104993 F20101115_AAAUQQ wharton_j_Page_026.jp2
019a56c393a2ff96c4461a62c06736cb
15441 F20101115_AAAVTS wharton_j_Page_100.QC.jpg
e3cb0cf6b9373d4337b018656746f667
669 F20101115_AAAURF wharton_j_Page_003thm.jpg
08537890989ceeac29b43fe77a20ebb1
9041 F20101115_AAAVUH wharton_j_Page_108thm.jpg
d8e65fa818c119b481cd29d615ccfee17344363e
114390 F20101115_AAAUQR wharton_j_Page_041.jp2
5077 F20101115_AAAVTT wharton_j_Page_101thm.jpg
49ae5474a121bca201e70453cb9b970f
fe80010211f56e82142bf9a1c13b63559d199687
1051983 F20101115_AAAURG wharton_j_Page_005.jp2
a43c958bc29ff01e638ccca4e84f089a
616256bbe9a9fa2435951cd3f18e2edd2d9ecaf1
37987 F20101115_AAAVUI wharton_j_Page_108.QC.jpg
c94ccbf67e624fab1468d3d416c76f22
42f87e367b11a5df516754b4d0c454a572f0fc37
10937 F20101115_AAAUQS wharton_j_Page_077.QC.jpg
be508806feb5201645406db2874f25a1
55164011352346940a3b32cbdb00ff19a872a24b
14827 F20101115_AAAVTU wharton_j_Page_101.QC.jpg
0483a391e0a88d051a5e6841068f78ca
ed1b434c28e3b1cd8f919ae3c8c8cc28ac1d2926
6927 F20101115_AAAURH wharton_j_Page_010thm.jpg
32eab4f57f583ab7b4325302e4dc9382
2c4aea5bae19a1cb67c801bbbd8a4a49f1ee1355
8838 F20101115_AAAVUJ wharton_j_Page_109thm.jpg
db83e7d21c5ee7c41c1b96be2baa74e9
290a4d3a7a7eb1ce403d482a36f7f6eaf112b7a1
104000 F20101115_AAAUQT wharton_j_Page_088.jp2
28152495ee73ee3b63ed132952b0a350
cc37bbfd055e97321afe8164da4792db2c5d2c27
8392 F20101115_AAAVTV wharton_j_Page_102thm.jpg
1828df0f570cd02a1341aae72476985b
5186625a3dd58d0da0108a8ae053d1af230089a1
4091 F20101115_AAAURI wharton_j_Page_091thm.jpg
0b3791eb050c72ef3573b70b5bfd6641
b29529f6d0997f1385d1d62e972d729562b112f6
35731 F20101115_AAAVUK wharton_j_Page_109.QC.jpg
e218fb5bee49568638df03fc80892402
007f09275942990979c1b41e856b4a2becd11091
1946 F20101115_AAAUQU wharton_j_Page_059.txt
5efbe429cfa0da42ee0027dc57ebe2c7
33769 F20101115_AAAVTW wharton_j_Page_102.QC.jpg
995e75d62698cdbca75806eba4d0d475
de0da92cf530519e3a664de4906bc469bc221f51
25271604 F20101115_AAAURJ wharton_j_Page_048.tif
8e98758626774f6b187cf75ae90a7d72
bf824d7025fe48f386d7bf4d34585946abc521da
8329 F20101115_AAAVUL wharton_j_Page_110thm.jpg
caed0be57f73648350d74dc3b5480ae87cab88c6
115111 F20101115_AAAUQV wharton_j_Page_039.jp2
4d744ba71fb00c62c11dff793b5131db
392ec677a2c28dbae14cf471955a7ebcc07093d5
6677 F20101115_AAAVTX wharton_j_Page_103thm.jpg
232c3a86bdf34f32a0bf22df9299360e
f891dc4b94c794a549b47179fa3d141baa6bb2ff
8258 F20101115_AAAURK wharton_j_Page_079thm.jpg
b906b580b84fd369fea92334113380f7
34216 F20101115_AAAVUM wharton_j_Page_110.QC.jpg
d50f6dd744a614945570daa3f8e61acc
26f6775f31da63fd45b211959331e0148b5f7448
325 F20101115_AAAUQW wharton_j_Page_035.txt
cb308ee77cc1c7513a7e715592a78cd8
79b09626d83c5be22ff8b3dd5637228e4bdc5b1a
27062 F20101115_AAAVTY wharton_j_Page_103.QC.jpg
e6ec7ef9df50c6577e531bb5fa8ffcdb10a4a955
167630 F20101115_AAAURL UFE0019751_00001.xml FULL
8b8eaa339d9417e30ba1bff6004a5568
88319439d8919021882926f70e28322cd982dc8a
4284 F20101115_AAAVUN wharton_j_Page_111thm.jpg
df2d17febe0ac4cfa2a39d169a703d21
d2de175392802600b52528508f9b0429f58df5fd
1275 F20101115_AAAUQX wharton_j_Page_002.QC.jpg
143ee9a68974680a34dfa94ee888f899
9f822309f0ca6556d843751d511d25c456a13160
7868 F20101115_AAAVTZ wharton_j_Page_104thm.jpg
513cb98b7bf75c1c2b2878f8dbfd5271
58a99eb2a335759c1f3d280b14d45178bec02da7
101441 F20101115_AAAUSA wharton_j_Page_013.jpg
cdbb01f04549d5e560e63934a2ab6ac5
1e27245400a12e41ce7a2e25efc8853134b3f156
16666 F20101115_AAAVUO wharton_j_Page_111.QC.jpg
7efef664c218d4eb902496b9569beca5
8588a2b5e178f764eab61853881aeeff86697508
47762 F20101115_AAAUQY wharton_j_Page_073.jpg
fc53ecde6428093196785a35075a4159
aabb4fba598ac390a467f5b30e8ce6623e6f3db4
102599 F20101115_AAAUSB wharton_j_Page_014.jpg
5ede2527c70cc480b54b27342b2257ab
bc4efe02f3c06f14d8b43cec80b2d067b1a2edf8
129311 F20101115_AAAVUP UFE0019751_00001.mets
3b5c59a953e56a7845ffedc395ca0d1f
e3f4e0ce6680e2079086cd6eda5f7fe8d001725c
9345 F20101115_AAAUQZ wharton_j_Page_020thm.jpg
0980c1ab1513a21bb94194da24b3bb90
00a84a6bd0d68950f3ec0736958e8236db60e0b0
111144 F20101115_AAAUSC wharton_j_Page_015.jpg
21e719776eda103b63daf6af5b2beb2a
70dc1b7ef30a0bd192e11927912b4ec50ab7f48d
25533 F20101115_AAAURO wharton_j_Page_001.jpg
62338f536a4bb9c08f6bba1fbfb7a12b
4a041d17376d9a4af81affef45dcb80c17791351
101466 F20101115_AAAUSD wharton_j_Page_016.jpg
7455021efddf6a6f05fdfa7a75e5375e
ecf2d72bbfded8ecdb88c3ce537f4bfaccb4bc72
4519 F20101115_AAAURP wharton_j_Page_002.jpg
7188f1f6f4d89eabd5ba39ba1bd5f0eb
0cc5029cd7d9a327861b6ff25142882a749d4d3f
99720 F20101115_AAAUSE wharton_j_Page_017.jpg
007031aac833279eb6a354cc366053b6
74dd722a1378d7d479bc1f84ff91d80956dfd3a8
5589 F20101115_AAAURQ wharton_j_Page_003.jpg
980498c2ed477ff26feee9b0b80fd082
dd442aff18b75fe30909812eb1f5dcc2b29e564d
100450 F20101115_AAAUSF wharton_j_Page_018.jpg
176cdf62973729054141e1f7cf89ab38
d0a5814a4e181a03c1dbe7968ecbd8a4fe48d6a8
63707 F20101115_AAAURR wharton_j_Page_004.jpg
133376 F20101115_AAAURS wharton_j_Page_005.jpg
7dc9b50c5235b59554433dbe29fe3f02
bf7d5d65a25c117c5f79c10eb2339c6f9756b219
106762 F20101115_AAAUSG wharton_j_Page_019.jpg
10b536c804d76230bfa3a9a030fed471
7fd9c758a319a2a5e3e1fabedcd47e8cdc835752
149414 F20101115_AAAURT wharton_j_Page_006.jpg
b28f350b4db4acbb92eb8a1ce28a5e26
669c039793af3fc84363a7913b6a1411f2ddf5e3
114045 F20101115_AAAUSH wharton_j_Page_020.jpg
6d56728c7a66bf0ba115b8af82ecf908
9d2345a9915dae7f18434c52ecb968156cdfa849
54377 F20101115_AAAURU wharton_j_Page_007.jpg
4cc2aac85296ea9ef39aed01cb25be14
111343 F20101115_AAAUSI wharton_j_Page_021.jpg
640f522f9a66f86daa81e6b5fbbbee2c
2b34b4c36abea241e93eba3f93b3ffd72e68e99e
134201 F20101115_AAAURV wharton_j_Page_008.jpg
d198e8b304b207dc27e886ca6a50a7b7c751c55f
110854 F20101115_AAAUSJ wharton_j_Page_022.jpg
6c88766145b13d61887687301d5b2102
7a8f7f277ee3d594977d8e2560f01e59756d7529
142941 F20101115_AAAURW wharton_j_Page_009.jpg
fecce7997fb8a2cbaf05121110e50657
b1a32b388134312a5a9c9d8c7e0cca35d2fc5d76
112254 F20101115_AAAUSK wharton_j_Page_023.jpg
eaab285f4ddfd82db70e58d2bfb3e8ea
1684efc16ac44e8eaa9842b288aa095c8d012abb
100864 F20101115_AAAURX wharton_j_Page_010.jpg
29533820336d30a4047e4a14730b9119
b7247d4c7969c1d678a70818754d35f26871b87a
106463 F20101115_AAAUTA wharton_j_Page_040.jpg
2d162eec776b4c1a7628bce4427d8dbc
a5657f1ddd7c1172e468e7a197cc7b1eecce66ea
99034 F20101115_AAAUSL wharton_j_Page_024.jpg
a088b7fa0a9967264956bb868096e1f0
f8247f7528f639f5a6b0d0ca4394553b1273b9e5
99666 F20101115_AAAURY wharton_j_Page_011.jpg
3aaa203663abac4a00ed21e6596c8636
107123 F20101115_AAAUTB wharton_j_Page_041.jpg
9f0f933880b57cfd296cded11c390a57
435ec66c3bca6cc07e5c718a5308dd5a5dae2710
103489 F20101115_AAAUSM wharton_j_Page_025.jpg
109195 F20101115_AAAURZ wharton_j_Page_012.jpg
8826772495a89b8ef284d78cd7877f3b
3ccc537f8378eee47edce6785904ff34e1b8261c
98107 F20101115_AAAUTC wharton_j_Page_042.jpg
1a752991f5eb4f4e388980c1a741515d
1f92ec1734f12f3a8dc8044217d0fb3fde27a1cf
97852 F20101115_AAAUSN wharton_j_Page_026.jpg
2783f21762c83f1a132ccf6ba60332e1
82340 F20101115_AAAUTD wharton_j_Page_043.jpg
dd84505c678b059de8203c1a47e27d3d
df15fd9db63cc3378703960cc6430f3cbee74443
36763 F20101115_AAAUSO wharton_j_Page_027.jpg
a8158ee7559ac4d12c2012514f905405
ef202da08e21acc032ee4e85e58f3f2c8afa7291
96099 F20101115_AAAUTE wharton_j_Page_044.jpg
3e8045af0b4708e470d53a70c69c3f6a
c17d1eba2c0c9dcb11032041d99c8b284c4dc2e4
67157 F20101115_AAAUSP wharton_j_Page_028.jpg
9e2a0d427e923403075da816ee578ff8
86492 F20101115_AAAUTF wharton_j_Page_045.jpg
301a5ab4f57e12036bf35a940724a0db
64cb14c8628b6d687bd044219598f0ffbc1cfa52
23448 F20101115_AAAUSQ wharton_j_Page_029.jpg
a27ef9e03cc7b5f420fee8a8cb076821e7a32ec5
106916 F20101115_AAAUTG wharton_j_Page_046.jpg
c864c9f1bcfdbc7ac767b95786157d6f
90bd10924aab7dff4be380856a8304baecd18aeb
50923 F20101115_AAAUSR wharton_j_Page_030.jpg
4061c67d5eba42e3a0c1b5de9d9b28a5
28938 F20101115_AAAUSS wharton_j_Page_031.jpg
823cfa49531180f838dcb1f021d061ac
e372181b4278be6bfb8e8d496be2fc3a41abe08c
86046 F20101115_AAAUTH wharton_j_Page_047.jpg
e193c903bb40af65dbddbaeb2b515955
bb589a2ee6372c6a82491f61103824be39327b03
39607 F20101115_AAAUST wharton_j_Page_033.jpg
612a881818d1352bc40ab7df5063e615
47475 F20101115_AAAUTI wharton_j_Page_048.jpg
37dd0fc79e553c4d21d2ac909f0f786e
05f279569f84fa33b55a45f7141bd3658b755e89
40242 F20101115_AAAUSU wharton_j_Page_034.jpg
4a093933551d5cff88b6fd6c0a013186
ed0128b2fd743a2fa11c8a7c7ff789b39fe9f609
18229 F20101115_AAAUTJ wharton_j_Page_049.jpg
c97f3ca3fa551c7ab40539e4276fbdd7
c44c39041d4ac78e472db2aefc4793f1a75b4d47
22447 F20101115_AAAUSV wharton_j_Page_035.jpg
bddb1cdaae689f2bbba3df9003d65083
4dbd36d15d813af92fe705b2902082e989b91e4f
32770 F20101115_AAAUTK wharton_j_Page_050.jpg
4b54f28a503bd9399e0124e5fc6ba94f
1dd0e0b2a5c185b46b7c52fb8f033b099bf13794
110476 F20101115_AAAUSW wharton_j_Page_036.jpg
cb2c390c8b7fa484c050641e393526cd
eac97833ab3dacc1ec67d0b7d8dee1eef57abe24
17342 F20101115_AAAUUA wharton_j_Page_066.jpg
ce736962e25ea50336f94e638e047036
5d182a0fa0af015ba2651a6c98fca470e3947abf
39087 F20101115_AAAUTL wharton_j_Page_051.jpg
dea3d10a27784407e2846ba4928b360c
13362605a98d08317e9afac6e5df071322bd8809
101011 F20101115_AAAUSX wharton_j_Page_037.jpg
0c95d3d9961545ac4ea0acd6340b0ecc
da2213725a4e17b3040bcf45a0e08274d1104bf1
61744 F20101115_AAAUUB wharton_j_Page_067.jpg
4109c9b76abcc6641c33ceaaa24f54a5
511a03f3bbd2e1a3df5d6677ef1137b1dc62e992
73004 F20101115_AAAUTM wharton_j_Page_052.jpg
19e10e28f86787bcb1c6c7c7d32c343b
df2c16ba28e612e29345f0549da4a95da320d2c2
107815 F20101115_AAAUSY wharton_j_Page_038.jpg
2fce2e67013c34655187c90951b24c1c
ae67ae2a8b9541e88947e556c5676f545920e6be
36760 F20101115_AAAUUC wharton_j_Page_068.jpg
8c9124b26d6aee4563369865bdd0118e
79c70d278a02737fd912f6445fa3d047f33ef5b0
42974 F20101115_AAAUTN wharton_j_Page_053.jpg
bef58d87e7cba4182de89724327efcab
110096 F20101115_AAAUSZ wharton_j_Page_039.jpg
42c996dc72a0a39a53c35e63b67c890d
bb546468d64ab13b5796af5b991b9ee9d4b862d6
105847 F20101115_AAAUUD wharton_j_Page_069.jpg
f5b6b8d0fc2cfb922077f3c67ba82caa
8618a87c4be96e5a1a2f2dbd5e3cca1301ff2e19
52382 F20101115_AAAUTO wharton_j_Page_054.jpg
3e6bc577a3d5054a89a083dfc355bfd7
986efc5d453cbb9b31be7185e606004673e42b18
103470 F20101115_AAAUUE wharton_j_Page_070.jpg
c030a24b40867e9db3f96bfccaf99f77
aa9cb0c5f9d815dcf81174010c5a5b6ea9de65c3
110122 F20101115_AAAUTP wharton_j_Page_055.jpg
6774be6419d6b89815a986f45555ee7d
d0709cf9bfea4d03b243f88aef0a7a13e314ee22
110944 F20101115_AAAUUF wharton_j_Page_071.jpg
47c1ee073647ea21d440b0068480c0db
107520 F20101115_AAAUTQ wharton_j_Page_056.jpg
d585786f1a1f9538ca497f950df383cf
a8b0b85a5db1d248e13e8377c7d50228b4037db5
96348 F20101115_AAAUUG wharton_j_Page_072.jpg
156117ffd5355365fc4e1ab411f77b2b1fa27ff1
112344 F20101115_AAAUTR wharton_j_Page_057.jpg
a35b46019dc070ceae50c330738e1829
0ce2d5ab44f5e886763c9dec6de6323662a6f867
F20101115_AAAVAA wharton_j_Page_007.tif
af5a0ed6df9e5d869103a70530e664e7
454ef66fd190a4169e55215b9e530ee121a84cbd
28691 F20101115_AAAUUH wharton_j_Page_074.jpg
42e10dbec0bed3069f785cca2f07df3c
d9afce2788d385c7b7d2e403b22bf02cdd8dd154
101305 F20101115_AAAUTS wharton_j_Page_058.jpg
fbdd654d31b96f10d18b1c3dfc0ce5e5
58a0871b51176bb16ebd2ddab0723fe101b39f0e
F20101115_AAAVAB wharton_j_Page_008.tif
eb94dd9df6226074780573ce2fc066ae8be67f8e
100466 F20101115_AAAUTT wharton_j_Page_059.jpg
73f461924e27bf8fcc9987eff7be0ce7
f22667a500322ce598a03b70961b8714e45e2c75
F20101115_AAAVAC wharton_j_Page_009.tif
6fcaa285031317fcfc61a7c42ea921af
05ea665babb9fc2e7d2817cd8c46fe2230e97f87
40536 F20101115_AAAUUI wharton_j_Page_075.jpg
d2c7815d8b0c04004ca6750aee69deef
4fd2ac625a8cd5f472f21ab319daf2ec2b84c2bf
110723 F20101115_AAAUTU wharton_j_Page_060.jpg
9a76bff935545d7edd21e4fa5999ee74
28c2770a875f13d85f1eaba9cfe0a60604624516
eee4fcc0db2002c29e4b3caf72ee76d7
110616 F20101115_AAAUTV wharton_j_Page_061.jpg
23abdc47bb4930e7b68e348bbf3b5966
8ee8d7d48a5b547a36d631759de1cd80ac45cc4d
F20101115_AAAVAE wharton_j_Page_011.tif
d4f25d29ec67cc6ed414b635de78ef56
d21666b4fb72363005e626d9bf822c9397592970
21906 F20101115_AAAUUJ wharton_j_Page_076.jpg
c082cb6683ccc567ba7eaa93b4a10aca
a16ab096e7b82da7fd4cc7b6c3999c53bd8cc8c0
111865 F20101115_AAAUTW wharton_j_Page_062.jpg
0aed1d61e3607082b4383686199bb8eb
d69e40d2174226b1c04616f64f686f6481ab2c6c
F20101115_AAAVAF wharton_j_Page_012.tif
30ab01bd7f4f653f6c317606b8dbbbb7
80019e627e179fa392a2433688e2de5c9af92f11
33237 F20101115_AAAUUK wharton_j_Page_077.jpg
cb07ccbba61b89050c8b7c73e6d3989356dac476
92715 F20101115_AAAUTX wharton_j_Page_063.jpg
0785b900e5dbacb8aa1b9977cf30df9a
778e3b59949195decaa9071eb4e72113434a1353
F20101115_AAAVAG wharton_j_Page_013.tif
5f854e88d1280fc405e18ed7fe77f188
239aaee69d6191aea6f38691d6a4ac16f481f1a2
25586 F20101115_AAAUVA wharton_j_Page_093.jpg
24a9781acaec361f67a9dd4015f47bca
e9270707c9a738c6a3f7a3fb2dbf94cf295a15cd
111417 F20101115_AAAUUL wharton_j_Page_078.jpg
5610120d536f70da8ba979fe4fd6721c
d6515d717c64c1c352940607fd51d4b5f84b6ec3
49405 F20101115_AAAUTY wharton_j_Page_064.jpg
ebe6b9af29903f5bb7943c3dfcb425a2
e655e5f578d21f2108db745a4b9eaef8c5ec51cf
F20101115_AAAVAH wharton_j_Page_014.tif
84f550c79f63c3f71512e4dfc1636061
5bccdec9d36bab8542a80a62d7bcfc667e1acec6
42348 F20101115_AAAUVB wharton_j_Page_094.jpg
710b864631029366931e63d49bd92c3a
0e9e2804aeb8c037c9734fb81e95d53f53723307
100915 F20101115_AAAUUM wharton_j_Page_079.jpg
9d53efeeea76609184dd652412ce318da97ced3c
35544 F20101115_AAAUTZ wharton_j_Page_065.jpg
a521ef5d7eb9bece17448fc81c15e792
1101796cfb0f7a8f4e47802175f18f27f2fb1192
F20101115_AAAVAI wharton_j_Page_015.tif
58b1d3bee6c6315716f3fcb937178b8c
51042 F20101115_AAAUVC wharton_j_Page_095.jpg
6e162e415f2003e42bbeb8ced44ea58b
e1cb32ebc4958b432d4b7ddb55696e2c19211d0f
87820 F20101115_AAAUUN wharton_j_Page_080.jpg
5bfb0c5a9b2004df174e67d487677a1c
0025b0919f944a5bf7717f52b1a3f29838ebf66c
F20101115_AAAVAJ wharton_j_Page_016.tif
a9d84eb9d1e917aa8a618d7b54020fc1
a1d3a22a4171f9ee257cd05b310a13e8bf163bab
40213 F20101115_AAAUVD wharton_j_Page_096.jpg
ccefd3af5cfb2b1317b56604790bf722
824b3494969d199468d2e4bfd673a93244302f49
93794 F20101115_AAAUUO wharton_j_Page_081.jpg
327b5d47d2648086c95715c5bf04434c
F20101115_AAAVAK wharton_j_Page_017.tif
662faa258ccd2f64eaf9897c5ba34e38
96e49f50b4a823de0b185eb14a01299d7a65e0e9
35532 F20101115_AAAUVE wharton_j_Page_097.jpg
4dd51b846190e00b568f67f250f0c6d0
5601c762877f5de5261dd54658e63a4414033510
81026 F20101115_AAAUUP wharton_j_Page_082.jpg
faef05fbe54f7e54c2198f3e2493af66
012bfec0460f229d00c8720a0dd379ed05f4789b
F20101115_AAAVAL wharton_j_Page_018.tif
5968341d55ae95c943a31e47ecf82a39
9d750fdd6a4f4bcd20a2252d19236432bcc0119d
31724 F20101115_AAAUVF wharton_j_Page_098.jpg
588b973de4b5631aaf7f439ae76a790a
d0a175b8ba3d08c5e1a149f10328547328a76804
69942 F20101115_AAAUUQ wharton_j_Page_083.jpg
25e8038e7fc0d07f5592fbe442a2990a
5f76ac651873e21a727260aedf6f2d33dfe83fbf
F20101115_AAAVBA wharton_j_Page_033.tif
92ac17587b7c662c7353b1bc4198d91a
f72fd0b4a653752e36341e4f98718b71e0f02955
F20101115_AAAVAM wharton_j_Page_019.tif
8387c0d11fa98d3a3ba8c77a10a8c48f
412cd5e45521cf187e50f9c59899df1d9b485d10
38658 F20101115_AAAUVG wharton_j_Page_099.jpg
71ae249391d97e1c44e81c486c03a9c5
105934 F20101115_AAAUUR wharton_j_Page_084.jpg
9540e86848121016bc9318e2c16c94df
ba881ac8f31d59ac28868f518c3db3d415232edc
F20101115_AAAVBB wharton_j_Page_034.tif
4b2389f5b1792555d9af2964f5a1a731
85670bc3f2fa7d5cf9371275f0f77e898106c03a
F20101115_AAAVAN wharton_j_Page_020.tif
c357178769488ee91189941ca2c5e031
6e6c2672bf6b8e1af7bafa66721dafd8449e7db4
50424 F20101115_AAAUVH wharton_j_Page_100.jpg
19edb459a99b70378fe9ca2a1f42235d
9ce2892e6ef2b07f24ff927e7befde0b691feede
109656 F20101115_AAAUUS wharton_j_Page_085.jpg
e970096ffcea801cb9abf0ba5ac57450
01bd978ddd323c776084cc7d76d04888192ab507
F20101115_AAAVBC wharton_j_Page_035.tif
e6a9d447178517cbdcf098b21231978d
517de7e9707e6fb1cc007d18c841b1f6829d207a
F20101115_AAAVAO wharton_j_Page_021.tif
d0b8f8a200a900cb439fde80cf04ac86
54317fed44b187860d59a4fcc5cd0e767881a6e2
40061 F20101115_AAAUVI wharton_j_Page_101.jpg
4cbdf2560bfba11d0b460c17d2e7cbf1
b049ba37fbc0fe9cb2bdd197952798fde6e62e1a
94764 F20101115_AAAUUT wharton_j_Page_086.jpg
770f7d1337e088535782dfa6780741ca299e66bc
F20101115_AAAVBD wharton_j_Page_036.tif
e3b72c80f52254b4cb7bdf79af46016e
94600fb2c1f8ab8714634ba513b75e57dc4a3d4f
F20101115_AAAVAP wharton_j_Page_022.tif
ea8a55dc288d3ed1af46e8e366233db6
100999 F20101115_AAAUUU wharton_j_Page_087.jpg
a7781f9213539fe867177924ee27be44
80e92c0732bb64c1c22cb624916d1a33d0291b84
F20101115_AAAVBE wharton_j_Page_037.tif
675e7173760deb2f03066636a86d7ecc
6fefcd0ac9e4e5464dd7f03dbe48fa6c0073161f
F20101115_AAAVAQ wharton_j_Page_023.tif
b8738788680101f271bfb7739a83f025
9c40d4b71eff5ddc18e5a03a25871406011e15bc
105581 F20101115_AAAUVJ wharton_j_Page_102.jpg
7178549ddb74c5d11bd762b1a551923f
98544 F20101115_AAAUUV wharton_j_Page_088.jpg
8525135372f74ea1692aff11b08d490e
83e319eac7fcf93476c7ff5be2396ff2f06ba0d7
F20101115_AAAVBF wharton_j_Page_038.tif
5e353d0ac53c7dbb9595fea559eb45174f9a862b
F20101115_AAAVAR wharton_j_Page_024.tif
af8a9d91574c4c2c4f899737630ef7be
2939658c3686f255abaaa3a2b17a805b076064dd
80882 F20101115_AAAUVK wharton_j_Page_103.jpg
50cdf76b000110aba8877659a85cdded
3812aaeb7eaf7a81452388e1b1fcc220ea024403
118810 F20101115_AAAUUW wharton_j_Page_089.jpg
2172cafca392252612af9b1cd10d5650
9db0691eb23fdb1dd7c50386e6d85dd89d484dc8
F20101115_AAAVBG wharton_j_Page_039.tif
97027697932728fe9d60459e10d1b51d
1051980 F20101115_AAAUWA wharton_j_Page_009.jp2
4e3a7e2c317816d394947b7f480eb011
2474f447f5fd00614a2ab0495790087644250900
F20101115_AAAVAS wharton_j_Page_025.tif
5bac10aa5cde64c7aa602b9f4250685be150a469
97197 F20101115_AAAUVL wharton_j_Page_104.jpg
ba839039a700b74ef96a8a2f74aedd27
87241 F20101115_AAAUUX wharton_j_Page_090.jpg
7c2923d8b52a4bd109cb388328ceb9bd
f7428100912fce0bb03f86e9eabeda578ed1fc78
1051964 F20101115_AAAUWB wharton_j_Page_010.jp2
093a1bd3300b1a5d9b611a8e479e4a50
4ae059127b262a3a083bf119c9eef91e83c62100
119188 F20101115_AAAUVM wharton_j_Page_105.jpg
d4645de48f3dd523d018b7260f7a9732
c93dd237a7c6615c8d59ed3c7904a10c8645ca67
41154 F20101115_AAAUUY wharton_j_Page_091.jpg
42aba9bbd01d4234e82193c42a9f3cc8
f7fb23335fe36b298b196c02fcaf6e62aeac130a
F20101115_AAAVBH wharton_j_Page_040.tif
fc5d65215790d07fa505d2a547f1d2f9
104377 F20101115_AAAUWC wharton_j_Page_011.jp2
df0eb51735b5b03dded7e22038fa95e1
dc1f2e9814544ccc1274d26e582e60029de25049
F20101115_AAAVAT wharton_j_Page_026.tif
2a183976c942734a177b4c06c23619c2
116461 F20101115_AAAUVN wharton_j_Page_106.jpg
055c5a7f9585eb8ea5156f804005ddab
45cfbb1349987a723f2045e4a6c0eebf5d4bdef9
46478 F20101115_AAAUUZ wharton_j_Page_092.jpg
6921a5ba0e262f07501408820d51a050
8d709d1347793f6ec2cd651c39eb86c020b8a557
F20101115_AAAVBI wharton_j_Page_041.tif
62d9a47278e15963280ce307de52f510
08ae7896f90ebf8020844c1946bcc118ffe0b5f0
116457 F20101115_AAAUWD wharton_j_Page_012.jp2
63622056ea0cd345c5ac6fddd245421d
04a41dac71e5310359acf531c1903e12b333e872
F20101115_AAAVAU wharton_j_Page_027.tif
cf64e893b88b431851ee3bfb82ba8f4e
113854 F20101115_AAAUVO wharton_j_Page_107.jpg
99a64fe40568fda7d826a2edfeef5a8e
c87af0db4695f1db81a4219a5a49c288aee67ed5
F20101115_AAAVBJ wharton_j_Page_042.tif
c925f933d961eec4a32db0c25465b671
58681067b0496b758b2b85256600290de1e5c6e2
108722 F20101115_AAAUWE wharton_j_Page_013.jp2
0bf2372c0d3dd3df654d79913a2bb2d8
F20101115_AAAVAV wharton_j_Page_028.tif
30e5e7a5b955661907c3f532e15ff431
303af9c828bf06ba0414915e06469ea791ed5e6b
121334 F20101115_AAAUVP wharton_j_Page_108.jpg
25303a86c58ed79bb943afbc4b9f6fea
F20101115_AAAVBK wharton_j_Page_043.tif
abd99e7370bdd4ee26b91642927811d4
384ac61d7b2d6953c7e4f0eba349ba563930d51d
108637 F20101115_AAAUWF wharton_j_Page_014.jp2
5eb4c6410e617bfe80f32de7b951f310
a98dd80e4dc0ef71e97bf0f35e98fab156b27de8
F20101115_AAAVAW wharton_j_Page_029.tif
1a248f6b899ed7fa9ea65f73aefc3dc2
4f522b466924c45ef74dd90381d4b92a2f380a6b
116009 F20101115_AAAUVQ wharton_j_Page_109.jpg
618d640ddd3686d7de663c800ed2ee5442d76efa
F20101115_AAAVBL wharton_j_Page_044.tif
f8b42e3a3f6fd5ace8db23d645da7c20
7c87ca1b534bcaefb084398345a957eef9c885d5
116964 F20101115_AAAUWG wharton_j_Page_015.jp2
5ec02265baf2799007d9bd82acb55f26
27d76325cdb2d8ee8fa4630c76e621d3d152c74c
F20101115_AAAVAX wharton_j_Page_030.tif
42cc14f8ff54e2ef9da4e758693c95c8
bb700b29939d96c826a6bf1d0e5634e685a3dce9
108368 F20101115_AAAUVR wharton_j_Page_110.jpg
07b5c45bca971576fe4fbe4282c3ec49
086697194fb2f18da153ed082d288228bd7c3830
F20101115_AAAVCA wharton_j_Page_061.tif
a8d737665b84b73ac9a9f3e4a2059f3e417347d2
F20101115_AAAVBM wharton_j_Page_045.tif
423ffec40b4a4d2cda077f4f3abf004645547b0c
106736 F20101115_AAAUWH wharton_j_Page_016.jp2
3591e9c2d9bdec12527c53b0a2ddc40e
4be2fdf59887ebc23117ccf412639546a95ea1c1
F20101115_AAAVAY wharton_j_Page_031.tif
dcde368011dd35b60712fa4a369d3bbb
51306 F20101115_AAAUVS wharton_j_Page_111.jpg
e7917c0e9a292c4e8bb95cebbe4caba94c029cc1
F20101115_AAAVCB wharton_j_Page_062.tif
1e632f1f6fb1bd0f2e6969c22e21226c
F20101115_AAAVBN wharton_j_Page_046.tif
e5d3ceb6ebf58001035baf3829b52c131a5a17cd
103328 F20101115_AAAUWI wharton_j_Page_017.jp2
198db298e6c398f71ef386fe8cd905f2
ef12bb96d1e01606d8c17793c91fbb65faa36cd8
F20101115_AAAVAZ wharton_j_Page_032.tif
edf261532bcbfe94e7684d7a56fdd669
481dcde3bf00a7130757e193f2cc2bc341e8da0a
25170 F20101115_AAAUVT wharton_j_Page_001.jp2
3303eb8007fb755f517e8dda769bcf16
7cfc52c3c1c07d3a6024aeee967d22481343a14d
F20101115_AAAVCC wharton_j_Page_063.tif
954fe3d81299570e1a5d3cdfafe82b3e
F20101115_AAAVBO wharton_j_Page_047.tif
80ecc321603afbc5649c17ea5495c4e39368bd0b
105691 F20101115_AAAUWJ wharton_j_Page_018.jp2
835dd2990f5a2577a09de41862208f27
27c0874157018241bf4383fa1ed56afd5d83df27
5894 F20101115_AAAUVU wharton_j_Page_002.jp2
1969f2a26b885a3dd2f1994efdfa575d
d9397816ab732215af131817b433db73984265a9
F20101115_AAAVCD wharton_j_Page_064.tif
3d7126fd32e881775dfd331a0b0712fc
9da03d92cecc2cd3bbbe4a45c8ca405765bbcd16
F20101115_AAAVBP wharton_j_Page_049.tif
609c38ce70d70aaa168f2014a91aaacd
e98a74d40d4ea0d8e2c89265993eedb24328c04d
6572 F20101115_AAAUVV wharton_j_Page_003.jp2
c427f10d397881c1c6e95c79f9f7604a
9ce4001ded0d3406ba7a20e6a833e080e90fd057
F20101115_AAAVCE wharton_j_Page_065.tif
1ecf6f388a87882264c8f5c13a2bcef926c2d439
F20101115_AAAVBQ wharton_j_Page_050.tif
8b128d5233292cb36c3b25d7a7057a3d3388ba71
115190 F20101115_AAAUWK wharton_j_Page_019.jp2
16021abc522780c2b4bc900134fff2ee
8b5728e466738ec7d87ae5d0a358aea7856a2481
66502 F20101115_AAAUVW wharton_j_Page_004.jp2
e56c2155b491ebd2612a8061a723352e
70ec91ebd017d9660c44a0cc0d1368d0061442b1
F20101115_AAAVCF wharton_j_Page_066.tif
d6338a63f72449e084d2c3c7fdfb2578
4e61b9b263cf46a6ee2ca98dc4f7e52a3ae20444
F20101115_AAAVBR wharton_j_Page_051.tif
1bcf46c4f63fa712d1a3e5568bc2ce00
9f833b6b7b081a1a29a10e1c61b708ae580069b9
119057 F20101115_AAAUWL wharton_j_Page_020.jp2
593fbdc14d0a63b8037e76ab08cdf472
1051947 F20101115_AAAUVX wharton_j_Page_006.jp2
ba7ca787253d53851363dc8784b3f7be
d63b844b25b36069fe6e57d4f7853e1db2d46f92
F20101115_AAAVCG wharton_j_Page_067.tif
ef14dcbd5226ca736725eb57a6f7f1b7
42cbec4ded84b232ac2eddd60f40db7ec49328eb
116753 F20101115_AAAUXA wharton_j_Page_036.jp2
e090fc1594a1e736918b8427dc81c338
d4e1dd68ab060f366460038348a0519fe1d0e6e3
F20101115_AAAVBS wharton_j_Page_052.tif
59ee7a2e00c61b135842d71f3a242bdedd36c971
116600 F20101115_AAAUWM wharton_j_Page_021.jp2
f0668627149c9a1d5bb158aaa31374eb
e01581086d44a7db046290889d9401f398935096
982170 F20101115_AAAUVY wharton_j_Page_007.jp2
b4a300dac257568ec16a25b12f798aeb
c0c4a2931ebffb39c83fe375a9ee7d5468ac2e37
F20101115_AAAVCH wharton_j_Page_068.tif
93d91de88737179d608bab012ba5ea6a
108744 F20101115_AAAUXB wharton_j_Page_037.jp2
fc2f8361b53598073d690da4b0e750b8
b986c6dedf08770df56968d9e5476cd93d76aeea
F20101115_AAAVBT wharton_j_Page_053.tif
ea9b37039b50e681c23d5c38eb93a8eb
118758 F20101115_AAAUWN wharton_j_Page_022.jp2
1353283490745d83c15bc644d223ccd9
72ea98fdd93810cd836b8a5b33edce3fb94d9ea2
F20101115_AAAUVZ wharton_j_Page_008.jp2
57d50fc270ba2db91214e5ba8b626cec
599bed2d447faf21a7993bd5562e67453357b5ef
F20101115_AAAVCI wharton_j_Page_069.tif
d2ab592622530599f6db1c9fb3bc296c
1c24988a73b829ff7e5617c1d954820b021d7591
113841 F20101115_AAAUXC wharton_j_Page_038.jp2
9d6f9cde15a4ac42a15fac39e39886f2
bfecb2d4deaeb0e1c69fc542c7f38213c6f0e731
120709 F20101115_AAAUWO wharton_j_Page_023.jp2
de5edb923e6fe0ddb5f257d69a4b255f
7c6b9f94c18a9171ddf546728f4bca0e296b1635
F20101115_AAAVCJ wharton_j_Page_070.tif
7dd5c97ba39a373e14df7024fea0c389
8de8e3ae26e60f10faa32e80a2fdee01fd643123
115265 F20101115_AAAUXD wharton_j_Page_040.jp2
6ee288bc06bd3e5c941bbf20912b572f
7817c5956f6b649c9b741e15243913965a82e8c9
F20101115_AAAVBU wharton_j_Page_054.tif
bbe20ca29641951580b55639055ff4cc0b9e534d
106388 F20101115_AAAUWP wharton_j_Page_024.jp2
44d5ce181029fde94aa4d2f8f117cf59
8fbe72735b7ef9f19a61a00ff794f440566cae7d
F20101115_AAAVCK wharton_j_Page_071.tif
ef18a9ca49407e552ae2d11140b6f2b4
d36e057ecee34f7e81ecacd08894a57d0e57af49
104633 F20101115_AAAUXE wharton_j_Page_042.jp2
3bde780f2579b9f122368a62a79ef637
968ab06193cb67df3ee4edc34f2caf9e1b1df7e9
F20101115_AAAVBV wharton_j_Page_056.tif
6e0f477289cab8621aa971cbe6f59799
aa9ff1891eb813ea180129573d6712bf2fb2645e
112242 F20101115_AAAUWQ wharton_j_Page_025.jp2
de0a1c81d7ef4360f7091307fd97c37d
F20101115_AAAVCL wharton_j_Page_072.tif
eb723d7c209ed8043c9b9daeca3aa8f9
0c8b24f466ebdcca9ca6ae4d4cd59a6097aba26d
89159 F20101115_AAAUXF wharton_j_Page_043.jp2
f881dc45361becc85f4931f839981450
0149a0c8eecb51e4b4b926ba3937176bbd89861e
F20101115_AAAVBW wharton_j_Page_057.tif
a75e3bb52690b67f099c0dd0df02a1f6
12b016fe0136b3fb6f56d2bf5a968a0bbbffbda2
410162 F20101115_AAAUWR wharton_j_Page_027.jp2
02e50e883188c5d5029f42f7e861e9439f267ff5
F20101115_AAAVDA wharton_j_Page_087.tif
e8dcb8401bae342e508a1bd4cd91ea28fb2e1407
F20101115_AAAVCM wharton_j_Page_073.tif
bec7a6c6b94bcd71462dafca9eea5d66
955fd1a005505dab7a7b81cf1a7bf1e66a8c6b02
100784 F20101115_AAAUXG wharton_j_Page_044.jp2
74e34ff4d27374f7e925b97bfc3d5e18
0c6759c54d7065c1633117471ac553529f748c4f
F20101115_AAAVBX wharton_j_Page_058.tif
98edc5f87a86479700542f96743a8c19
905526 F20101115_AAAUWS wharton_j_Page_028.jp2
2f52d18fd77911570c605384122ba6df
966e5cec5e84e27bcd4b1c84d9d5e19346794d91
F20101115_AAAVDB wharton_j_Page_088.tif
317801b083c57d6434dac7d55a0d9c32
82dc9321dc8332064f4ac16d08d2b2269249e0db
F20101115_AAAVCN wharton_j_Page_074.tif
e8e90251c9b4cf10491159550a02cec1
f664a9326ce65e95d29cb2996e1c722ddb496b87
90225 F20101115_AAAUXH wharton_j_Page_045.jp2
01976b3ae9cbb47749dbdf52e354f194
656b97e180a340158520c2f672856c34227db470
F20101115_AAAVBY wharton_j_Page_059.tif
6ae13e41b13e385145d5b2b272251219
805de952d3b2f177d5a222b12d675a660849c5e8
625576 F20101115_AAAUWT wharton_j_Page_029.jp2
9a8d05b694a75cbfcf95089114a12e99
62eeb0885b4ce9ac32d581cf1e2388df8314d82e
F20101115_AAAVDC wharton_j_Page_089.tif
1df58507308233f28eb3aae1f9c631a4
2ce8de10860c7f84e846835db27ce75a0cedbd94
F20101115_AAAVCO wharton_j_Page_075.tif
f597066a8a678e4a20c5d989fd0a201f
bbce1f309c2d98d4a33662f08d5989ab93b8f662
113713 F20101115_AAAUXI wharton_j_Page_046.jp2
8416f0ec77bc78ddb28d8457dc3cc6b9
79357bc065fbda287330e23a8048638fdf864e3a
F20101115_AAAVBZ wharton_j_Page_060.tif
bc4316a30cfe4a2314268a904b1017a7
9ca7bcf7eb62f2511aee92ca5d59427983dff57f
597624 F20101115_AAAUWU wharton_j_Page_030.jp2
21872265fa5b9b46bfafa3df058bfe32
ba6ac196a94f86fa2187acebdfec746717a40545
F20101115_AAAVDD wharton_j_Page_090.tif
7b5e49f2d7fa88ed5ec177a7c965b1838664e0d5
F20101115_AAAVCP wharton_j_Page_076.tif
1a557d74159cc3671592fc0d84dcbaa0
8b77753c83c8a6d4012ec419ae1b89e896239df1
90840 F20101115_AAAUXJ wharton_j_Page_047.jp2
86ee6c5ccde8143cdbcceaf83efbdb30
226964 F20101115_AAAUWV wharton_j_Page_031.jp2
c12a1b82bbb2227bffca7c972ef8c728
ba3611d1e7db38195ffc8761cb97f14240be7d1d
F20101115_AAAVDE wharton_j_Page_091.tif
F20101115_AAAVCQ wharton_j_Page_077.tif
373a1b4d51c53205bbcac9ef04278209
1f97fb3b2889dff15c2c92b73d74867b12ed75c4
533235 F20101115_AAAUXK wharton_j_Page_048.jp2
76dbc4a7b771928a4c1d8b5e1bc123f889992bb4
303564 F20101115_AAAUWW wharton_j_Page_032.jp2
344550d77c284ebd490997159015836d
cddbd9b13a87f1f9780112d19728ee718e5b65f9
F20101115_AAAVDF wharton_j_Page_092.tif
c5308f537ed46e028bfec23f232acfb2a2190514
F20101115_AAAVCR wharton_j_Page_078.tif
b12d1c251f08a37740c0c2c7aa0ec4aa
ff7ab1da70ce153e18606e50dae6b4e37d0753ac
401404 F20101115_AAAUWX wharton_j_Page_033.jp2
3d3b3f44e9b7e326d9db1a7bae6129ff
72ab0b28184c9ee332b844b0e355f028f500e58c
F20101115_AAAVDG wharton_j_Page_093.tif
13b1c46d0e355447a305eb0b0b7b6ddc
0c6fca28c27793f26fa976eae028fe1b6e85bc3f
617165 F20101115_AAAUYA wharton_j_Page_064.jp2
1e2681c22978234a4fe4ffddc7475915
cebcf742885a2ef3de66a236c30be95f1bff93b1
F20101115_AAAVCS wharton_j_Page_079.tif
37d34e209670c4cb4eeffc05e0c9e404
e173df2be19a12225033efa7db2640cac9a22a02
172329 F20101115_AAAUXL wharton_j_Page_049.jp2
5cf141fe3864f3734f7ed0ff18608496
c158f299f40603557f4fe7760421e7901ab4a80b
407547 F20101115_AAAUWY wharton_j_Page_034.jp2
a1728c16e6353f243a6751a59618067b
b42be92c77361fec55fdd40c26a0d44369db8e6b
F20101115_AAAVDH wharton_j_Page_094.tif
0b16223ea23b385a9ecda3828caac6cb
cc59f83843998dd0410f3187ac2ec31103fa5173
359214 F20101115_AAAUYB wharton_j_Page_065.jp2
3689cb9e810aa8c9a5e303ff884de343
22b90fe25b8a5d89acd8d8ef1f396208d80355ab
F20101115_AAAVCT wharton_j_Page_080.tif
f3f86d615bc9776171fb2470ff0d5027
90495d8b4a4b7a8d6674c5f0b611cc30c31d31df
355327 F20101115_AAAUXM wharton_j_Page_050.jp2
2950cd768bc44a097b6a3ae3cd34d8f8
b16a32e69e0fc92d6982f6da429029bdbb835c67
184045 F20101115_AAAUWZ wharton_j_Page_035.jp2
29f932a3d223b05b03e4f357acb14f54
420d7585b65debb915f85e5ec89f4308f625d6b9
F20101115_AAAVDI wharton_j_Page_095.tif
7c015bb119975fbb7c74befa8be910a3
9cf448b22f78a30685338edbce3046f005957e2d
130741 F20101115_AAAUYC wharton_j_Page_066.jp2
ffaf3b7327596c46a10afd972408e351
953cdd4659469297d72414244173a7e08fb5cdcd
F20101115_AAAVCU wharton_j_Page_081.tif
89215554246b13171724ec402c87e57f
eebde2537c785fe0945b31c564f7b8cdd4c10d91
566127 F20101115_AAAUXN wharton_j_Page_051.jp2
294ea13efce0eaf5e8257758e64641c9
f9ede2e92afe520e989e7cb1d7af129eba7daf6f
F20101115_AAAVDJ wharton_j_Page_096.tif
d5a74261e25cbb05b96083e018c1e89e
aef2b42d1cfabd2d24fa4df58c34e6f0ff34ca5e
1051972 F20101115_AAAUYD wharton_j_Page_067.jp2
68ea8baa83410f982a44269ffc550550
1051892 F20101115_AAAUXO wharton_j_Page_052.jp2
c7fbc5750b5315fc926d52c3de2e0d46
13207c127331eb15265a95762bd840eb8dcf00ca
F20101115_AAAVDK wharton_j_Page_097.tif
d242c699a641694d9056afc66b1081cf
8f3cc7aa81b6601cc1abba5a539a7d6454d676ba
724650 F20101115_AAAUYE wharton_j_Page_068.jp2
69c969b018ca2cd8daef2c225fcf6492
3eab66bda8a34940e085a952450000b8e82155e3
F20101115_AAAVCV wharton_j_Page_082.tif
1b1a88ebfd9ae33edd76cb37c790e6e0
38f5fcd2d3ab9a88d8c1f72a19a7f55968e83e41
454892 F20101115_AAAUXP wharton_j_Page_053.jp2
00e051911d8457d2133421c3509b93ae
0ee8bee414b6ed1582b2b508fc82bc567dcc595b
F20101115_AAAVDL wharton_j_Page_098.tif
12384d646b10d19cb4415b4dbd88f465
111127 F20101115_AAAUYF wharton_j_Page_069.jp2
f6da7cd1409d740755c98b4bce2847d6
04b68bcba649b66ae9a4caf45e1854a7e1fe7f95
F20101115_AAAVCW wharton_j_Page_083.tif
e1b08ae03537cd34d9157ed90c90e805296eddf0
414209 F20101115_AAAUXQ wharton_j_Page_054.jp2
d8c25c82ec79a48306da276fa6a4ef75
a0f005c3589bf2ff4176686f3af59bd3ba1200fb
F20101115_AAAVDM wharton_j_Page_099.tif
590128c835a45b3d55dee8a4a3b847d5
95a2901d18be15b75bce54c7f8b6d381e983a9ed
109960 F20101115_AAAUYG wharton_j_Page_070.jp2
a529a29e01ea895340be612c3527b6ae
d1bf2149a5537f32562b70ccf590b0a677cb391c
F20101115_AAAVCX wharton_j_Page_084.tif
4f5fd979aa1fc3ae0d670d8e66aac665
65aeb37a6531b83f652584f38d0e1e91468fe6b2
117323 F20101115_AAAUXR wharton_j_Page_055.jp2
e0a2991de42be5daedd8a8bd33af257f
1d490688c9d88d1e94815571faa7af3e17c0fd92
1039 F20101115_AAAVEA wharton_j_Page_002.pro
2d7b5f2577fe1e055cb5bcdd435609a930536745
F20101115_AAAVDN wharton_j_Page_100.tif
feb2b449bbfe7b371c4d40f4b059024c
717e77e0a473284a6241e88fd8d234896788f40d
115299 F20101115_AAAUYH wharton_j_Page_071.jp2
b90cc8de809c59fcf45980933bc8a7c1
1a300ca14a4bb484f0501ab21ebcff6d42314e43
F20101115_AAAVCY wharton_j_Page_085.tif
ef561954ab472989a93429a122efb790
113637 F20101115_AAAUXS wharton_j_Page_056.jp2
2b815d51992d606e7d709ea824a68aec
b6cb4f6d5e81cc99acb9128095c61ecebe815cb3
1585 F20101115_AAAVEB wharton_j_Page_003.pro
c2eb42f398b8c0dd97265d3188b051e82731a8ed
F20101115_AAAVDO wharton_j_Page_101.tif
05459ed6d2b3a729f502766239dd56de5f17f799
103147 F20101115_AAAUYI wharton_j_Page_072.jp2
2b02a47b6a2c6246e42324ce1265aa39
F20101115_AAAVCZ wharton_j_Page_086.tif
32b0352128457024969123d6a1441394
afeb79f62deef94458febc9c00d5c243fe70b707
118800 F20101115_AAAUXT wharton_j_Page_057.jp2
0b4d58e5efb3564bb05ab38e69e80afb
5d29c71808e3ae6008b9ec142daa028b9c22f64b
29562 F20101115_AAAVEC wharton_j_Page_004.pro
d9037550b07601e83101d7c45724b6a1
e63466c7aab02345f12547e567ecf47704387f0e
F20101115_AAAVDP wharton_j_Page_102.tif
c554db6d8ee6b29faf20f610fc80124fd0a03527
484895 F20101115_AAAUYJ wharton_j_Page_073.jp2
dc4615f021fd51d54511e068f84e4ce7
864e69b2e41ea5780501f5fd8683e1fd9805b025
104792 F20101115_AAAUXU wharton_j_Page_058.jp2
13746b1c36ce1815ed36a40ab98d5f25
f74b72498c8f95ffdce3d12d4a39b2ef15884b44
82014 F20101115_AAAVED wharton_j_Page_005.pro
9c9f1b6b05cf3afba460e3e338f59235
0ceefa87e9c6827affdaf3f3218264353c9354b3
F20101115_AAAVDQ wharton_j_Page_103.tif
672ea8086eb95da4ef69a59011034272
3c809bc737b6ae69130d27eff7ff2bff8ceeae86
243962 F20101115_AAAUYK wharton_j_Page_074.jp2
1648b2f3be3df63f2ba6c182536b60d2
899572c9b72957279d5be3e2f05ddcd254cd5501
103749 F20101115_AAAUXV wharton_j_Page_059.jp2
4589381fc53b726b1a80a90082ed5926
bbb27d8a5037d144ca0f240a2a6bec003c9d2763
91797 F20101115_AAAVEE wharton_j_Page_006.pro
c1855a43a42d601244c1310d086683b6
a93c3830c71cc66857e64461b816398d0f4345ed
F20101115_AAAVDR wharton_j_Page_104.tif
9694b2001a3d784128aa64eed9e869ff
6419553669492b89d7eb91cc02e04c27a4d222cb
544228 F20101115_AAAUYL wharton_j_Page_075.jp2
0a59cb44fabbcfc09c29657d467058c2
6704a030cd00c06bfc5ce8d5c43bb210edb29570
117014 F20101115_AAAUXW wharton_j_Page_060.jp2
ee70e7bca586a52c7924f603feaacd43
ab04fbd71c42a4ff8e7660d152fd9b05d9c2af32
29309 F20101115_AAAVEF wharton_j_Page_007.pro
d8df8b65dfca8b4f02aab5b09fddf588
01f32c8a3f5181eb4e7c1012c8f5496716e22cbe
463672 F20101115_AAAUZA wharton_j_Page_091.jp2
3c4226f6ea1540556dc5d49bdfa39f55
63eaf530b4673553428a930600db8f4cde096e10
F20101115_AAAVDS wharton_j_Page_105.tif
9665ba4319ee9835281a14c6cc02b8c0
115321 F20101115_AAAUXX wharton_j_Page_061.jp2
e7d59ef9ba3dd9b75a8faaf0fb37cab8
37154fd4f61f666c2ddb2ca0ca6039f7bf6d0b2d
65318 F20101115_AAAVEG wharton_j_Page_008.pro
4a5f795326cc43b3f12b95028c683e86
08d63a5930c3f6806c5220f092603bebf816b9b9
415815 F20101115_AAAUZB wharton_j_Page_092.jp2
40c3f32ed24e7df3e92151c702eb266e
72d64687e93750a55121544b31262998ba6f8ddb
F20101115_AAAVDT wharton_j_Page_106.tif
20d44f0d78542d84c74ca1db4ee9a4ed
c474367dfa3e574783bb76ee3535db3ac3b869d5
193444 F20101115_AAAUYM wharton_j_Page_076.jp2
6cb046d65e4419df75111673152fc59585e86009
117345 F20101115_AAAUXY wharton_j_Page_062.jp2
a3232d34a67849437d54cf8a9c5227e3
f0d00fd8d642cbce783480a275819849f02118c4
69370 F20101115_AAAVEH wharton_j_Page_009.pro
bf4f14d5bded36ac6da6cec627536425
9b2590cb27d24de69c76f4b3bfd089f9839457e4
249952 F20101115_AAAUZC wharton_j_Page_093.jp2
26e67b2860202bde59e039971e0dbf7b
36d459093ceec4a63f6cd9a3da145a84ec28f5b9
F20101115_AAAVDU wharton_j_Page_107.tif
c59408d09dcf79de02936b04069a60ea
623993abeae85800efd5c77947ed3361b256b9ac
468329 F20101115_AAAUYN wharton_j_Page_077.jp2
ed53743116431b37b0f713d5ea7e7425
37ed1ef22eb23f3be7a573dda19610f0a487b938
100271 F20101115_AAAUXZ wharton_j_Page_063.jp2
a3aa91cba290d7e002d74a80b3420caa
85587ddd50c392cc1bded0ec7510172499d78e2f
49493 F20101115_AAAVEI wharton_j_Page_010.pro
89c15253913d995de2df60cb0bb10ef0
466418 F20101115_AAAUZD wharton_j_Page_094.jp2
828f4b9550926224ef9104afb98e0b7a
4b7854a5affb6b6b3b115d01fc6451953f008260
F20101115_AAAVDV wharton_j_Page_108.tif
288e47b704c1637d0a073733bbf93a47
6312181aa74c5d5fe36494873c3a68f36073251b
117623 F20101115_AAAUYO wharton_j_Page_078.jp2
6cc855baa541c5f646ff2182677f23f3
8ea4e3fc6aabccd73764760ba15c041c356dfa3e
48021 F20101115_AAAVEJ wharton_j_Page_011.pro
fdc7f4b4048aa6a8be6ec2dd530c3c76
936622 F20101115_AAAUZE wharton_j_Page_095.jp2
86013369f0f03f3b1d46219f08edfced
d5ac51b328a660f409ea091d56163cd2bd2cafdc
107113 F20101115_AAAUYP wharton_j_Page_079.jp2
ce6f2ba5e8fa881c8010a88e4db800f2
f3a2747a3e5d6b4bfe9cbdb8a3c8dba376a60d38
52994 F20101115_AAAVEK wharton_j_Page_012.pro
2a01749d11c574446dc0decdeb63af8e
e5598d37c4655ec1dc7933b39fc00876a7995281
381977 F20101115_AAAUZF wharton_j_Page_096.jp2
5bd570bf97fdef0d9b3a3d5ee7466617
191c8fe9fd9e26ab6c93941b7360a33df772de8e
F20101115_AAAVDW wharton_j_Page_109.tif
116b4a027f20c86c7dd3d48e76cb1be9f9181604
92954 F20101115_AAAUYQ wharton_j_Page_080.jp2
57b64fa96a7e0053aa72636f1b171dc1
014ea623f894a11f9706c1709e7d74d579b946fa
49965 F20101115_AAAVEL wharton_j_Page_013.pro
1b0f0270ce78a199b2bca9e592baa2ac
66e4f33f82d06b15b8157bc2d6edc7141f6394c3
351434 F20101115_AAAUZG wharton_j_Page_097.jp2
490c5642df3b60689e9ced9ba2265d6657841c9e
F20101115_AAAVDX wharton_j_Page_110.tif
ff8fdc7d8a4d6101971286ca845654cb
99238 F20101115_AAAUYR wharton_j_Page_081.jp2
6fde2e1bd4b42562c1e243f520d87a18
12117 F20101115_AAAVFA wharton_j_Page_028.pro
a6afe8ba07f3d5a95fd333156a2dc65225836aa1
50355 F20101115_AAAVEM wharton_j_Page_014.pro
af4fbd1c63a3d5a63359feedf43abaf3
95d8b7ce45566b0cdd9b29ae4ee869b32fd3eef9
F20101115_AAAVDY wharton_j_Page_111.tif
3f2c76a9977eb33cc270edc887ba43158abd2e02
86347 F20101115_AAAUYS wharton_j_Page_082.jp2
dcee6e364a0e57a4de46ae4128b4ebf3
54ec8337c3266ab78e1197a149d07d1244305fe1
2625 F20101115_AAAVFB wharton_j_Page_029.pro
ca76bded9b949e9eb85fb5fa90405889
c33a0e4c549e83a444189630df085987315db9cd
54605 F20101115_AAAVEN wharton_j_Page_015.pro
332b6e5255ff6968e8c0c4ccf6ec8dd3
e59a1982b3d6b43e6770211ba7d596d38b3155f3
314464 F20101115_AAAUZH wharton_j_Page_098.jp2
58bd5cd1210cc81aa0fe096e606ccba2
05ff7fb2c1bda0fdb4b414dc5ece6597e57aa344
7923 F20101115_AAAVDZ wharton_j_Page_001.pro
06ce1da6854942372fe9afbcfe0d2117
74106 F20101115_AAAUYT wharton_j_Page_083.jp2
a461884e1c2732570aa12e4d1322b0a8
4044 F20101115_AAAVFC wharton_j_Page_030.pro
7609f401a2a3a94d4520a9491437317e
49596 F20101115_AAAVEO wharton_j_Page_016.pro
38b78d10031de72205d6a0cf1dbc8525
4858f31e2544fff5858d23a98803de13c67cf178
386568 F20101115_AAAUZI wharton_j_Page_099.jp2
325ae4aa4a1dc2f244bf1531b695b77247c8c3f6
110323 F20101115_AAAUYU wharton_j_Page_084.jp2
2631da3a43921220c005109f38527d5c
1b66a8ff01824497fe3cbb063e2a2d9ae3a9b894
5162 F20101115_AAAVFD wharton_j_Page_031.pro
f709fe76b72b77d3e891c6db7f52aff8
2360136a77200f413c335fb681501bb8ea1df36c
46929 F20101115_AAAVEP wharton_j_Page_017.pro
a9f83e5419ff7e74ab4cbc741d26a02c
b9a495fe6fcf66955d939e031444c41ac2c7e1cb
356018 F20101115_AAAUZJ wharton_j_Page_101.jp2
c4ee316b46fbc2fa7aec083f210be06d
684d981df08c99cbeac39dba29e93c1f3d19036d
116367 F20101115_AAAUYV wharton_j_Page_085.jp2
14b95886d1ec90e6c6b871d42013ed6209e55632
7091 F20101115_AAAVFE wharton_j_Page_032.pro
0bae6d9f12c26ef5f5ea4c688b57be01
bb9b50aa7c804b093d5458f5c3e375382cb3ece3
48841 F20101115_AAAVEQ wharton_j_Page_018.pro
98a4aa95e7f734b9374a46ec3801c338
c028842e4d420347160f28daf9a55d4375fccd60
113072 F20101115_AAAUZK wharton_j_Page_102.jp2
9b5656eb223b53d62de2f24435d67cd293e741a5
999347 F20101115_AAAUYW wharton_j_Page_086.jp2
0c46df1d3cf4b8c165ac71dc65c18312
8584d28906585ab84bc3691b78cabda3ea899c98
2099 F20101115_AAAVFF wharton_j_Page_033.pro
915103c2fef5da99fbac57beb9c12769
950974ebe118070d4bae0653145b2992d73c0ea4
53611 F20101115_AAAVER wharton_j_Page_019.pro
979dca72cb5ae30db62fcffd054c1851d6db6f4e
88199 F20101115_AAAUZL wharton_j_Page_103.jp2
028068215d1c96b6cf83860d15888326
3dfd0452df826b3d725424bed224138b6634bfe3
106913 F20101115_AAAUYX wharton_j_Page_087.jp2
d1d212fddddeb48a98d2e91808dc8719
1f3eb41541173d1d80f341a377dacd7ccf0de84d
9367 F20101115_AAAVFG wharton_j_Page_034.pro
9e172055abf367d6faac7c4d3c4e2420
915dc6c1afa655cb691a051ffced1110298ecf5d
56055 F20101115_AAAVES wharton_j_Page_020.pro
2766c361151de2b43f6f3ace9d14710e
94ab3ff1b8e708f7773ac7cd620b9cbc56b37046
102086 F20101115_AAAUZM wharton_j_Page_104.jp2
ea2d7b770e8730853300058acd5245e4
16510c08ed8e0d2c2f430ec03d500b424b4260bf
125649 F20101115_AAAUYY wharton_j_Page_089.jp2
a9b0efb07d6f9c5d9d1af4af7400bb63
d1cbb7103fcdec37627f021c3b1b93ed34f6da49
6893 F20101115_AAAVFH wharton_j_Page_035.pro
58dcf5e76d6a972c03b6d8d1a1de5837b03a9679
53683 F20101115_AAAVET wharton_j_Page_021.pro
2e3614c835b135fde1d9a1c06911590d
423ceda8018bae2f1d4159a54ae949cc888c5b7f
951635 F20101115_AAAUYZ wharton_j_Page_090.jp2
53654 F20101115_AAAVFI wharton_j_Page_036.pro
be6ab320774fc21201cfe645174a8c49
bd41735354c0a2a5c006f8a1b4491bfac21e5c0e
55028 F20101115_AAAVEU wharton_j_Page_022.pro
38d29e19f476c45dd4d27217c33bdba01e4e1d49
126308 F20101115_AAAUZN wharton_j_Page_105.jp2
12680d513e484c180da30093c1d1758f
16f51c6da205c6330b81d6df53728c3f9c349be2
49956 F20101115_AAAVFJ wharton_j_Page_037.pro
b4ab23c210a433ae03484ab1468046b3
fc7670295d3a781ca59b122e63621653174a1331
55604 F20101115_AAAVEV wharton_j_Page_023.pro
79ac478d0eb948a0084ef1b984d9dc8b
120570 F20101115_AAAUZO wharton_j_Page_106.jp2
024d9c841e5cc5da9702b310326e45c2
78f1b984b4be7e3be9c2d2854dd9ac4efb1eabab
52641 F20101115_AAAVFK wharton_j_Page_038.pro
ed5e98a444efb61e3b2e256bdf0318a6
f9bf33fb1bab256687f76218b4c2edf89f1b210f
48594 F20101115_AAAVEW wharton_j_Page_024.pro
6b36cf5f8dcf300b7e22c76d52698d57
8208f097b4b9a745aa06cd1b7aa80e00b94a2bb5
118175 F20101115_AAAUZP wharton_j_Page_107.jp2
5fbea34dc95faf937515db48991c24c8
e1b4e8722a1ec0296ae521f919d3612f7a3beed6
53019 F20101115_AAAVFL wharton_j_Page_039.pro
6e558fa931ebdeb78059f5f8e2f9df83
94bfcaf228f31c319e6c7f1e8032165f553de94a
129087 F20101115_AAAUZQ wharton_j_Page_108.jp2
9ec6ac1a48ec7c1fbcb020e95f15bfa8
10313 F20101115_AAAVGA wharton_j_Page_054.pro
b1e0c0b9ab6f0b6eb2f84e6e2a5a4381
19b4e3371360e791180b0f13d6707633bfee959c
53450 F20101115_AAAVFM wharton_j_Page_040.pro
9074db8c8f761333f02140a47914a5d8
28f57a16dc6f5587f5a3a460ccde4488d36d227c
51477 F20101115_AAAVEX wharton_j_Page_025.pro
f5cd659279f3ab09da75f428739fa8c3
39d590f3d51a682a19e34fe98ec332256615f564
120544 F20101115_AAAUZR wharton_j_Page_109.jp2
2e87079583781b9e7beb068899552b6d
54235 F20101115_AAAVGB wharton_j_Page_055.pro
61da675a1b5caf29bffa47cd0bd72c94
080385dd2c947b3bf29c7218888c34bf391061c4
53972 F20101115_AAAVFN wharton_j_Page_041.pro
76f4dffbe2ba3f34d16d6c727321b2f2
aaee6b2dc8ca8c18e6f354e8488cd82a2a075de3
47050 F20101115_AAAVEY wharton_j_Page_026.pro
31886c0017d983b1416c3a95d8f3812f
b0642f5669866d0c6361edd98c86e9fde2e8a297
115737 F20101115_AAAUZS wharton_j_Page_110.jp2
9c9239fea7a2f3326b14d141f7c27653e2b436de
53592 F20101115_AAAVGC wharton_j_Page_056.pro
b61acf9b7897572c4c3e00f17be47c9c
075da1df875c73b9c54982e63c47eb4d925190d2
48796 F20101115_AAAVFO wharton_j_Page_042.pro
d80e91c068f1bcf1ddb1cb7a7559b046
11213 F20101115_AAAVEZ wharton_j_Page_027.pro
3bcd613723c453a6b9ac561f037fa741
fa5161316768ea905cf3afe9861f7621f9812052
52430 F20101115_AAAUZT wharton_j_Page_111.jp2
11437b75e1a43d2d944bb8cab9ef43c0
0848e770466668d3bbbddbf4265e1532ce9f081a
54885 F20101115_AAAVGD wharton_j_Page_057.pro
d0551887bf4b77aee310d71125d96608
fcdd2948161f8137d559daa033cc77ee5daf90e0
40407 F20101115_AAAVFP wharton_j_Page_043.pro
e77896846c992bd66b505e3cd7c09aae
383eaa30d56e601d16b4c1123fb62dd2827bf7d9
F20101115_AAAUZU wharton_j_Page_001.tif
2b39439b32a750da70cebc9551807a264a843cea
47513 F20101115_AAAVFQ wharton_j_Page_044.pro
d4189cfa1bdec2d9dbbb21335c63c3d2458b8f61
F20101115_AAAUZV wharton_j_Page_002.tif
e49c9a532a614b3299995f18dce85148
72e2270fb788fb0feeaec5a6558db2357d7de852
48466 F20101115_AAAVGE wharton_j_Page_058.pro
ca83b80bc571e68e1f9bee1cbd94d3b4
235d98b6f78001364b6f7c5f1ce2fb8bde8ac271
43096 F20101115_AAAVFR wharton_j_Page_045.pro
cca13bae5c390f6c6c1c48e019de9389
b361dc2531d4f8abc817108f14b610cb09ae255c
F20101115_AAAUZW wharton_j_Page_003.tif
9565608e32136ea75cd6689d5b020b69
315404dfe1e8d5e6d0b0ff239a82f757eb6d61cd
49225 F20101115_AAAVGF wharton_j_Page_059.pro
80dee1764e68dda856495931693cf1ca
54138 F20101115_AAAVFS wharton_j_Page_046.pro
3438db4bd24b803d064da768fef2e462
2eaa6876fded8ac3b3f71f36c117b6cc9620207a
F20101115_AAAUZX wharton_j_Page_004.tif
5a35d22f867069503c7114888db884ff
811186699e93b815fa0075043e5bfd5e836313b3
53579 F20101115_AAAVGG wharton_j_Page_060.pro
b0b181891075449fb1b85a514af3c230
e2f410d948e07cd44994c4f6de489f6e2f317c0f
41741 F20101115_AAAVFT wharton_j_Page_047.pro
3cbffacea661064a884f06cc52958929
d5f9d52ec9d66714071cbab97710789e63891d40
F20101115_AAAUZY wharton_j_Page_005.tif
b5c87242b259e2daa874954ae439b0d7
5a32451dd2e36776e93f55b9627e8b886194d9da
53837 F20101115_AAAVGH wharton_j_Page_061.pro
12965 F20101115_AAAVFU wharton_j_Page_048.pro
0186d6092d1051edb5284031e6ba47ca
F20101115_AAAUZZ wharton_j_Page_006.tif
67ca9c3e2b2b790b61b704dab969b96c
55458 F20101115_AAAVGI wharton_j_Page_062.pro
483d5b636dd9c257f0840d4ab6df6a1a
f2e55c11884db1edbafd693ce1120b05a11de4ce
4844 F20101115_AAAVFV wharton_j_Page_049.pro
5fd595bbac11d545fc3994d6d251990c
2cda0f2ca19dc04c17c86d6053fe188a12d964f4
45904 F20101115_AAAVGJ wharton_j_Page_063.pro
5ffac7c7415d38098676533dc947bdcc
2df090914c3dcb6193a48e7ca2ee10ac9004ee07
8652 F20101115_AAAVFW wharton_j_Page_050.pro
4de836b99371f6ca20cd75e3dd230c4b
19397 F20101115_AAAVGK wharton_j_Page_064.pro
bce8052bc77be443e138fa212acfa488
0670d5af6fec0262cf394a86b8a14c7871b550dd
5853 F20101115_AAAVFX wharton_j_Page_051.pro
0661ed63d8ff4208ba8b17d9f9464dd2
7b0cf5fa00c1f2ce86b37b9c6efecc8fbddf50c9
6789 F20101115_AAAVGL wharton_j_Page_065.pro
44f1b92fbac05ce66ee53ac3a6974d19
43736 F20101115_AAAVHA wharton_j_Page_080.pro
72d5dda05c5581d14bd6d1267f216c1b
bea56e9bf0838d2464bba33d940b3a1a93b22f8b
4124 F20101115_AAAVGM wharton_j_Page_066.pro
ff06fd57767ce7cfbf48814729b3c3d6
4fbd5cc53eed9ab93ae99b4ce1aa25c3c02fb35d
3108 F20101115_AAAVFY wharton_j_Page_052.pro
d9fecb94409119371de04e35e3f22ebc
0d6cba9df6804e5acfb476a6807d9e36af66d8d6
47100 F20101115_AAAVHB wharton_j_Page_081.pro
0b245a9787a9f300ce8cd702e85381b2
5797d76cc6f38eeb3a16254419c70b6249e266c4
12531 F20101115_AAAVGN wharton_j_Page_067.pro
83fc012b61f07b25e356a69f0e438497
c1ce6139bfc2a78bc679e3b3dfdfdba4fd477e68
14772 F20101115_AAAVFZ wharton_j_Page_053.pro
eb0a32e9d1f3b74e058ba8f301e1e092
20e487872034ea73457197d7ffd59532f4d28c83
40282 F20101115_AAAVHC wharton_j_Page_082.pro
71a1c1197e275d82b3be37b0242d5125
26f6a6ef699b825bd08b324e6ac83316ae4b5759
5965 F20101115_AAAVGO wharton_j_Page_068.pro
2d7962ee0d5b06a7c7c704d95191cae4
78b9f48e73e2a336dd189d6eeaab6bee86caa865
33386 F20101115_AAAVHD wharton_j_Page_083.pro
89f7b0a0ba4e35e704e9015b014bdc27
51239 F20101115_AAAVGP wharton_j_Page_069.pro
f5b306a2b7437deba033c443cf929f9a
2365742b9a2879a4eca13d4e6273232325292c32
50936 F20101115_AAAVHE wharton_j_Page_084.pro
1fea1f18fee0dc20f57d855a2bcdc294
55cf8f8ca33940ec97ab299ab469f456e78d4080
51360 F20101115_AAAVGQ wharton_j_Page_070.pro
4a37e1f80d69f046289fbbeef717623dd44162a9
54898 F20101115_AAAVHF wharton_j_Page_085.pro
09ca649f170502b432285b928b4d5d53
917ce5b3e2a4ae82c42cf88b6d5ba11928306f46
54351 F20101115_AAAVGR wharton_j_Page_071.pro
1e37c3f730aa6c8cd0d78d9890d4a674
13139a5cd35e679114d78a8b75dd2fc3efe999d5
46325 F20101115_AAAVHG wharton_j_Page_086.pro
68d4371ac62b8749f964bcd1c8c94e49
47345 F20101115_AAAVGS wharton_j_Page_072.pro
e504a2655d492d2d277cbd3a7008035b
1f1ed75b3c384cd5d35a75effe77be1415d88320
49682 F20101115_AAAVHH wharton_j_Page_087.pro
9748f275a1dfdac45ff9fcdfcac68093
6b17cd1ea7006c01474404e9e702a94f0ab9b91d
15417 F20101115_AAAVGT wharton_j_Page_073.pro
0ef76b8b65a579b0c7ec269185eae7ca
427a9bcd634e915c7548938bb66e4f34c99e7965
48939 F20101115_AAAVHI wharton_j_Page_088.pro
1a21d8808660af42d85daa07d52d0fac
9a85be91fa7eb6c9664b3cc6e00fa1f3d764401d
6685 F20101115_AAAVGU wharton_j_Page_074.pro
69ca05935a979a58a9708d4d5c1a71aa
bc06bb42907145e79277a8bcb150b0585640c0ae
59379 F20101115_AAAVHJ wharton_j_Page_089.pro
3af3e34cc0f1d4fc416058feb5f02070
3535f9b333f3645049e4959cc07c270cd6b06b42
2569 F20101115_AAAVGV wharton_j_Page_075.pro
459e079cc2a32b729eda1a19189bcf72
c2135f9a34140d0f04371a57e269209644669858
41642 F20101115_AAAVHK wharton_j_Page_090.pro
54e8a0f1f091ec644ed18aec5a4c4b16
ab2aa9443df405bcae245574b1a173fa2d49e47d
6933 F20101115_AAAVGW wharton_j_Page_076.pro
5a6f07a8751792f4e6c1de2d7ea72300
32e30e4d28e73d97576eb6049614e403200296e3
11067 F20101115_AAAVHL wharton_j_Page_091.pro
1a24602dcef7089c767ac76afdb783cd
0de65c3aaa444a343b87a67eabfd4d228e867355
7766 F20101115_AAAVGX wharton_j_Page_077.pro
87eded3b2eb547340032196a97040c71
1c3f97fc171ac2173b513c50d16bc2bd2d9bc1a5
55708 F20101115_AAAVIA wharton_j_Page_106.pro
45a88274bc78cc21435a26a8eee53511
09e386e9d3c2df985e328f7e047eee856a25dffe
10140 F20101115_AAAVHM wharton_j_Page_092.pro
1775bc6fe927dee9edef414b54ba957a
02b60655e628334f583d08e4dacf5e4166efd32b
55268 F20101115_AAAVGY wharton_j_Page_078.pro
84bb404ec8e93ea7c3825de8da74d149
472ce6203ab65147ace7205df11a9f8b7ec30665
54174 F20101115_AAAVIB wharton_j_Page_107.pro
b5d716430ab4a9163fc022d181b2af81
9618 F20101115_AAAVHN wharton_j_Page_093.pro
424d74a0e3925eb28d637159ef192465
173e3708496f98e576403aff7e4b61bcffaa7f3f
58544 F20101115_AAAVIC wharton_j_Page_108.pro
90fc73c5c143eccc9e90c704d8c6aa6426d51aa1
7365 F20101115_AAAVHO wharton_j_Page_094.pro
d474c26e7fb5f6c000c276ebae7474d93cac9b76
49662 F20101115_AAAVGZ wharton_j_Page_079.pro
40eff8d1084d1183842d20a6fb2dfeec
d9c4b44d0438279dbdec62945d3d67046462c33e
55193 F20101115_AAAVID wharton_j_Page_109.pro
e1677268454ba087b1848d37d7c85248
3397a09bac2561172be52020f894657539544090
5560 F20101115_AAAVHP wharton_j_Page_095.pro
714e89d7d40aa250bd08f6c48ce09a26
52494 F20101115_AAAVIE wharton_j_Page_110.pro
88c7dd4d2e3bc8eea1ecb701d006cdce
f4bf46d86778de3b73c4ab2c33c24cb0bf05b5d2
11484 F20101115_AAAVHQ wharton_j_Page_096.pro
d20c3ddc714f678076b784ab917af62e
ab44398a838952fcd12aa0c8315561f01e2b2c2b
23030 F20101115_AAAVIF wharton_j_Page_111.pro
f100002ca888ee5f2f622fd31f6e99ba
b83a9d11dceec8679a8f64f5c4e4b7f163d9fa1e
7345 F20101115_AAAVHR wharton_j_Page_097.pro
b063d9457ef4d7e6f2cf312965cff633
2d494dc6c2aa6618087f82e565763bec44be2463
481 F20101115_AAAVIG wharton_j_Page_001.txt
38696b70d92958f5153c5f46e80964a8
6233 F20101115_AAAVHS wharton_j_Page_098.pro
a578308f6962eeeb9ec66910b3cf4723
2216b6064bd8a04457f0f40565762f8caae20f38
100 F20101115_AAAVIH wharton_j_Page_002.txt
d66c9ee46719b21ffcfaba88de2d6c32
8d06ce9eef64dac70da09cb1b6baf4f4aa69b44d
10895 F20101115_AAAVHT wharton_j_Page_099.pro
7dfd2428897bdcf7ec422af8e899d12d
64c286db1fe22fbe5f4a9a13808dde714e510cc8
117 F20101115_AAAVII wharton_j_Page_003.txt
dd3c14cf28517c5ca5b23876521f89a1
5a69b7eb3a229aee9bd9a9412b430c2cc317e0aa
17865 F20101115_AAAVHU wharton_j_Page_100.pro
4e3a5a10c62c01aa47f524ae899cb6c1
1232 F20101115_AAAVIJ wharton_j_Page_004.txt
931c528327d42d3aa7af9a56400f3660
10720 F20101115_AAAVHV wharton_j_Page_101.pro
35e08b30148d1332dd64bd0256973ace
fe9a51f19916d30fb2ec4afbb20a065f7c97f9f5
3600 F20101115_AAAVIK wharton_j_Page_005.txt
f8e9953ff92598db27a2b2095de5735c
f3fb915bae3c64147838927264575a2a5028da1d
52040 F20101115_AAAVHW wharton_j_Page_102.pro
eaf75ce64cf0d5d69c7af9c9ca2d5fb8
010a6851330081abbe79369f95d6ddf66d0841b8
3901 F20101115_AAAVIL wharton_j_Page_006.txt
ef3f509f819925d44b39be16985d3335
3ac505ceab9f282126ee3db696760120343d66d4
39343 F20101115_AAAVHX wharton_j_Page_103.pro
ae1b61a5394ba7b7dfe847b4c6c521fc
1281 F20101115_AAAVIM wharton_j_Page_007.txt
bea87c480c044cf0fbd85d699ae81a68
6f8e4070b048cbf8322c0621d55cc25ccd0ea744
46079 F20101115_AAAVHY wharton_j_Page_104.pro
1c4ab9ed71082350886d3d1786ac6381
056ee052208db9559b35c495572a41c9b78cc012
2145 F20101115_AAAVJA wharton_j_Page_021.txt
db7b9852f9c14faef7286c3e127ce57d
8d1040fb82c84565b3ca10565700f417a65886f2
2680 F20101115_AAAVIN wharton_j_Page_008.txt
7109c24c3e3ce7bcc278f4e06414b782
5691e51b19140452eeee18c94abb5ded7211e976
58806 F20101115_AAAVHZ wharton_j_Page_105.pro
bb2dc5d4df08f4b23bdeabd9066f26ec
2168 F20101115_AAAVJB wharton_j_Page_022.txt
68214d65cbe3a2943bc99f387d200cf0
2817 F20101115_AAAVIO wharton_j_Page_009.txt
2b5f39b4790fec6056e73c103bb71633
e084d5d1dc8f8fda2f3cb0bf5105f8945c1c4288
2220 F20101115_AAAVJC wharton_j_Page_023.txt
491074b5b5ce9c4162599220e4633a6e
fb05d96bed54237379d2f32b08d00479a6425447
2112 F20101115_AAAVIP wharton_j_Page_010.txt
05295bfc88de458e7bc5efeca93ccc188b3b5176
1961 F20101115_AAAVJD wharton_j_Page_024.txt
4f278d133b7e1af8de0c0a39b18c70d9
95b9df5f8ff04eee640e5351b0a807107fd6611e
2096 F20101115_AAAVIQ wharton_j_Page_011.txt
83c25a956657941c2edb62be534423bf
58cb86a12931f635b2ae5010e925289b7df15aa2
2067 F20101115_AAAVJE wharton_j_Page_025.txt
880cb332400c6ef4844062d6ea2063a0
8ccfafbb1dfd9d41b6ea400aa5e2538cec709f93
2088 F20101115_AAAVIR wharton_j_Page_012.txt
38c5be7863458b38e7ea372b7658f442bb376113
1871 F20101115_AAAVJF wharton_j_Page_026.txt
b717a7dfeaf46ede0dbfcf59bf11f28e
fd1c64f1160166d544fd66fac1258b9180c5b90f
2041 F20101115_AAAVIS wharton_j_Page_013.txt
66e9edf41e5bf23154a4a0a65dd59081
dd58363cb4c71003169ec0f121e7c20f991a22f2
486 F20101115_AAAVJG wharton_j_Page_027.txt
2042d276028bf57a2dd713c63834a828
04e4a3a00f481f56c328b1731e0a4192fe8e5a75
2016 F20101115_AAAVIT wharton_j_Page_014.txt
f9ec323ef20f7163ac88110d33ee4cbe
b290ccee38580de320dfd32fca2042f597d73bf3
533 F20101115_AAAVJH wharton_j_Page_028.txt
5d26635a218350e6c0172cf2b0286a48e854394f
2141 F20101115_AAAVIU wharton_j_Page_015.txt
19fcca281ff95c47eb382f0dfceb8f81
7af4f77865b4bb16427eb8d255de45688d49656f
155 F20101115_AAAVJI wharton_j_Page_029.txt
7a8b9062f87374135be5de4bfea20363
d513fdc59417fd8a76980595fd02794dc9ac8014
1964 F20101115_AAAVIV wharton_j_Page_016.txt
e9ef1e7149f3729137eaf91e7b233c652a435559
218 F20101115_AAAVJJ wharton_j_Page_030.txt
fe88ec854baa2812b86091da36fba90e
fd4fdc5c90f9876db50a2661466c42d8c957e75c
1865 F20101115_AAAVIW wharton_j_Page_017.txt
136c692dc9555f333f317b6dd5a69b59
9f7f02a9d6658bd1b87c84295e0a7893887aeb23
283 F20101115_AAAVJK wharton_j_Page_031.txt
2ca6639d21164a07ab756c5c554a70d7
4262bbb291ddf191c4eaf6d183c4907a85bbd230
1973 F20101115_AAAVIX wharton_j_Page_018.txt
7401a985e0a2e3fa3ccfe19bfd0d0e69
4141e4e70cb7ca8118de8ef25b98562bca127c58
637 F20101115_AAAVJL wharton_j_Page_032.txt
9c1ba6c0b3f8f5cf88b6c411713c919f
2a94d3e9950b1af965f1d491f2d74fc35f07316d
2104 F20101115_AAAVIY wharton_j_Page_019.txt
e29720b53e0ca7aaf612c5c314f2d2de
263 F20101115_AAAVKA wharton_j_Page_049.txt
ee882169ce4a7369e0ecb5364a624e2f
83c10783d2af21c3a162e43494f2eb990025ac88
186 F20101115_AAAVJM wharton_j_Page_033.txt
9690c84d652285231921853cf99b0c9d
3c30e2d0906ef24c5d940d533fbce8f18e806564
2200 F20101115_AAAVIZ wharton_j_Page_020.txt
cf417ce0c1a1100109b7e39dc6879361
102c6cf1b259bfffae88bacdb2ffc201f29a119f
456 F20101115_AAAVKB wharton_j_Page_050.txt
c44cde5aed6302849f7589a39befef08
467d51642e2f0c68515c14c81af413329c5426c9
408 F20101115_AAAVJN wharton_j_Page_034.txt
bd1d261d85803c76001af2e611e6306d
014eb101788a29822ddd5099a08ffbb25e313029
F20101115_AAAVKC wharton_j_Page_051.txt
5d12cea44bc9091e94df44e72c28d9c2
7a40c3526c5f07d7b0c7f2d21d8be05b42879751
2222 F20101115_AAAVJO wharton_j_Page_036.txt
495735f0f024634e2297c99c379ea898
64f6235fb1de10bb3db00bf9f578919a2329cfb2
190 F20101115_AAAVKD wharton_j_Page_052.txt
32e3607a558136849b33f7d38c75c5ab
66f2a4db5e50f2022162d70130689799bd4f8d07
2017 F20101115_AAAVJP wharton_j_Page_037.txt
c787142c709d05c438d206f8ce9d8af0e94fcdbc
723 F20101115_AAAVKE wharton_j_Page_053.txt
eda8f95e13cb711abb7c3fce5788e107
d48ba8fc494441684eef8341f12582afcaf220bd
2069 F20101115_AAAVJQ wharton_j_Page_038.txt
3f1c61374840cd8e64fc626fc48cd414
469 F20101115_AAAVKF wharton_j_Page_054.txt
e2015e19fca789bc154d1a0e47cb3850
F20101115_AAAVJR wharton_j_Page_039.txt
19a08fa10d6988f6426f5fe12f9d0433
526bef8cd0425784b5edd198dd698d4b43322480
2242 F20101115_AAAVKG wharton_j_Page_055.txt
98d8a9cd23caa78e8101ec2368486c24
529a6eea83bff96a13c3101af0d3e192d723ed1d
F20101115_AAAVJS wharton_j_Page_040.txt
8a43f1c9af35833e3aaa387cb5108ea1
f266c62fc97ff45077c96f8781e36300c405f13b
F20101115_AAAVKH wharton_j_Page_056.txt
93280d241ece349d934f9d249b783aaa
a27db85cd75f9f4e89d3b2cfee73dec5d026e90b
2127 F20101115_AAAVJT wharton_j_Page_041.txt
1355f1dcfe9e8a3caa70c67973953264
2192 F20101115_AAAVKI wharton_j_Page_057.txt
f7bcd9111f87cefde211c6b2f90d7227
6dd2832eaed229d83d9aa9a25d1f6a0c18cf628d
1934 F20101115_AAAVJU wharton_j_Page_042.txt
0a76c1f597c9481a49168f9434818176
fcd91093886c61230891487a9fe7c6532bb2685e
1923 F20101115_AAAVKJ wharton_j_Page_058.txt
dea85933ec50353ea2366692bc48ce2c
5902acf75d98e00ca578ed25fd4c6e6f3b59d31f
1658 F20101115_AAAVJV wharton_j_Page_043.txt
52bb2ee2c5fff115eee4b50eabb69bd1
a212de2aac14c8e6569511a205a6f1af7144e04e
2102 F20101115_AAAVKK wharton_j_Page_060.txt
d3239a3eb6a6af13c996a7ab563845a9
174076af3c637d85cc2fb9d2f485412464357200
1798 F20101115_AAAVJW wharton_j_Page_045.txt
daa644c9ce7355bbea6f2247f9f69a29
364889fe16128b250ab66b35c2629b24dc037cfe
F20101115_AAAVKL wharton_j_Page_061.txt
b7c0b500bed6196de0691e772c6d5428
a824af2b5515f2c1a4c638a09610d869693304ea
2133 F20101115_AAAVJX wharton_j_Page_046.txt
29f4444d67e36da5fd5e09185925d845
9070b097888954129c196fb9fbe2269a18909760
378 F20101115_AAAVLA wharton_j_Page_077.txt
b0e3dbba64ff7b6a35fc5b7a67164ec9
7b188c50340795a7eddc0e63645beabf299e549a
2171 F20101115_AAAVKM wharton_j_Page_062.txt
57d46218b3e79cefc356aef28efa1a5f
fcb8d073f035291470256d1d856631e44d6be7f4
1713 F20101115_AAAVJY wharton_j_Page_047.txt
b05f8bf59b447c3cc0715c082a5e4f2a
105b7987546f032dd63536a1cb68d31539f9e1e2
F20101115_AAAVKN wharton_j_Page_063.txt
5504f4ccc9da5b2997624799b8565570
564 F20101115_AAAVJZ wharton_j_Page_048.txt
2ea578033c648b411a70097ded26a28c
da225d1f6ab17430b776855b8f18884a25a452e4
2277 F20101115_AAAVLB wharton_j_Page_078.txt
835 F20101115_AAAVKO wharton_j_Page_064.txt
ce5dd87231520375520c38fabd84c39e
40b2432192b3cb31de93fa1fd22a3b9621fa4e64
1962 F20101115_AAAVLC wharton_j_Page_079.txt
e71be18fcc41fcfa0f8c387fa954d31c
45f7d4a6ae9095806dce4ca12b861db63fb71563
375 F20101115_AAAVKP wharton_j_Page_065.txt
d68c451215042698d165b37c87bf3ee7
8454878f2d0239c68c87cb014136f82e1cee8f68
1804 F20101115_AAAVLD wharton_j_Page_080.txt
8a44c07b5a4c364a90a825bf2a60f604

PAGE 1

1 FABRICATION OF ASYMMETRIC PORES FOR BIOSENSORS AND TRANSPORT STUDIES By JOHN EDWARDSON WHARTON 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 2007

PAGE 2

2 2007 John Edwardson Wharton

PAGE 3

3 To my family, for their continued support and love

PAGE 4

4 ACKNOWLEDGMENTS I wish to acknowledge the many people who have given me advice, help, and encouragement during my years in graduate school. I would like to thank Prof. Dr. Charles R. Martin and the entire Martin group for the opportunity to work with them over the years. Prof. Martin was continuously supportive and always willing and ready to discus s some of our new ideas for resear ch. I appreciate the level of independence that Prof. Martin allowed all of his students. Undoubted ly, this created an atmosphere of confidence among the st udents where creativity flourished. I would like to thank Drs. P unit Kohli, Zuzanna Siwy, and Lane Baker for their patience and expert advice along the way. I wish to thank Fan Xu, Lindsay Sexton, Pu Jin, Lloyd Horne, Stefanie Sherrill, and Warren Mino for very impor tant contributions to some of my projects; these great students have been very patient with me and are always willing to help with very high interest and good spirits. Finally, I would like to thank my family and friends for showing confidence in me and for giving their love and sup port throughout the years.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF FIGURES................................................................................................................ .........8 ABSTRACT....................................................................................................................... ............11 CHAPTER 1 INTRODUCTION AND BACKGROUND...........................................................................13 Ion Track-Etch Method.......................................................................................................... .14 Track Formation..............................................................................................................14 Acceleration and Irrad iation Facilities............................................................................15 Chemical Etching of Ion Tracks......................................................................................15 The Effect of Storage of Tracked Material......................................................................16 Effect of Etch Promoters.................................................................................................16 Etch Properties of Selected Polymers..............................................................................17 Effect of Thermal Annealing...........................................................................................17 The Effect of Temperature during Etching.....................................................................17 Effect of Detergents.........................................................................................................18 Etch Properties in Muscovite Mica.................................................................................18 Template Synthesis............................................................................................................. ....18 Template Synthesis Strategies.........................................................................................19 Electroless Deposition.....................................................................................................19 Chemical Vapor Deposition (CVD)................................................................................21 Resistive-Pulse Sensing........................................................................................................ ..21 Plasma-Based Etching........................................................................................................... .23 Asymmetric Diffusion........................................................................................................... .24 Dissertation Overview.......................................................................................................... ..25 2 A METHOD FOR REPRODUCIBLY PREPARING SYNTHETIC NANOPORES FOR RESISTIVE-PULSE BIOSENSORS.............................................................................36 Introduction................................................................................................................... ..........36 Experimental................................................................................................................... ........37 First Etch Step................................................................................................................ .37 Determination of the Diameter of the Base Opening......................................................38 Electrochemical Measurement of the Tip Diameter........................................................38 Second Etch Step.............................................................................................................38 Bovine Serum Albumin (BSA) Resistive-Pulse Sensing................................................39 Results and Discussion......................................................................................................... ..39 Conical Shaped Nanopores are Ideal Resistive-Pulse Sensor Elements.........................39 The Core Technology: The Track-Etch Method.............................................................40 The First Etch Step..........................................................................................................40

PAGE 6

6 Measuring the Diameter of the Ti p Opening after the First Etch....................................41 The Second Etch Step......................................................................................................42 Measuring the Diameter of the Tip Opening after the Second Etch...............................43 Reproducibly Varying the Tip Diameter.........................................................................44 The Mathematical Model................................................................................................45 Electrochemical Details...................................................................................................47 Conclusions.................................................................................................................... .........47 3 ETCH-FILL-ETCH METHOD FOR PREP ARING TAPERED PORES IN ION TRACKED MICA FILMS.....................................................................................................55 Introduction................................................................................................................... ..........55 Experimental................................................................................................................... ........57 Materials...................................................................................................................... ....57 Initial Etching of Mica Tracks to Prepare Very Small Pores..........................................58 Preparation of Tin Sensitizing Solution..........................................................................58 Preparation of Silver Plating Solution.............................................................................58 Filling the Pores with Silver Wires.................................................................................59 Etching Silver Filled Mica to form Tapered Pores..........................................................59 Making Replicas of the Tapered Pores............................................................................60 Preparation of the Carbon Tube Replicas for SEM Imaging..........................................60 Results and Discussion......................................................................................................... ..61 Conclusion..................................................................................................................... .........63 4 ELECTROLESS AU PLATI NG OF TRACK-ETCHED KAPTON POLYIMIDE NANOPOROUS MEMBRANES...........................................................................................69 Introduction................................................................................................................... ..........69 Experimental................................................................................................................... ........70 Materials...................................................................................................................... ....70 Chemical Etching............................................................................................................70 Electroless Plating of Kapton..........................................................................................71 Pore Diameter Measurement...........................................................................................71 Results and Discussion......................................................................................................... ..72 SEM and Ion Current Measurements..............................................................................72 Atomic Force Microscope Images...................................................................................72 Conclusion..................................................................................................................... .........72 5 ASYMMETRY IN DIFFUSIONAL TRAN SPORT OF MOLECULES THROUGH KAPTON CONICAL NANOPORES....................................................................................78 Introduction................................................................................................................... ..........78 Diffusional Transport Desc ribed by Ficks Laws...........................................................79 Hindered Diffusion in Cylindrical Pores.........................................................................80 Hindered Diffusion in Conical Pores..............................................................................81 Experimental................................................................................................................... ........83 Materials...................................................................................................................... ....83

PAGE 7

7 Kapton Polyimide Membrane..........................................................................................84 Irradiation-Track Formation............................................................................................84 Chemical Etching of Membrane Tracks..........................................................................84 Pore Diameter Measurement...........................................................................................86 Transport Measurement..................................................................................................86 Viscosity Measurements..................................................................................................87 Results and Discussion......................................................................................................... ..87 Membrane Characterization............................................................................................87 Transport Measurements of Phthalazine.........................................................................87 The Influence of Cosolute Concentration on Asymmetry...............................................89 Conclusions.................................................................................................................... .........90 6 CONCLUSION.....................................................................................................................102 LIST OF REFERENCES.............................................................................................................104 BIOGRAPHICAL SKETCH.......................................................................................................111

PAGE 8

8 LIST OF FIGURES Figure page 1-1 Swift heavy ions impinge on a dielectr ic material creating damaged ion tracks...............27 1-2 After irradiation, the material s are subject to chemical etchin g which preferentially removes the latent ion track.............................................................................................27 1-3 Etched pore geometry in a homogeneous is otropic medium to a first approximation, showing track etch rate, VT, and bulk at rate, VB................................................................28 1-4 Scanning electron micrographs of Au na nocones replicas of PET conical pores. ...........28 1-5 Scanning electron micrographs of etched particle tracks in single-crystal mica...............29 1-6 Scanning electron micrographs of a porous polycarbonate, alumina and mica membranes used for template s ynthesis (A-C, respectively).............................................30 1-7 Schematic diagram of Au el ectroless plating procedure....................................................31 1-8 Schematic illustration of Au nanotubes obtained from electroless gold deposition..........32 1-9 Schematic illustration of resistive-pulse sensing...............................................................33 1-10 Essential features of the staphylococcal hemolysin pore shown in a crosssection based on the crystal structure.............................................................................................34 1-11 Schematic of a conical nanopore sensor element showing the base diameter and range of tip diameters used in these studies (drawing to scale).....................................35 2-1 Schematic of a conical nanopore se nsor element and etching cell. ................................48 2-2 A typical current-voltage curve used to measure the tip diameter of the conical nanopore....................................................................................................................... ......49 2-3 Current-time transients obtained during the second etch step for three membranes that were subjected to the same first etch. ......................................................................50 2-4 Scanning electron micrograph of the base openings of two conical nanopores in a multi-track PET membrane................................................................................................51 2-5 Scanning electron micrograph of conical gold nanotubes deposited in a conical nanopore membrane...........................................................................................................52 2-6 Plot of tip diameter measured after the second etch step vs. the final nanopore ion current (If) at which the second etch was stopped. ..........................................................53

PAGE 9

9 2-7 Current-pulse data obtained for a protot ype protein analyte, bovine serum albumin (BSA) using PEG-modified conical nanotube sensors. ...................................................54 3-1 After irradiation, the materials are subject to chemical etching which preferentially removes the damaged ion track..........................................................................................64 3-2 Schematic of cell used to do the et ching and to make all electrochemical measurements................................................................................................................... ..65 3-3 Schematic diagram of etch-fill-etch method......................................................................66 3-4 Scanning electron micrographs of mica me mbrane that was exposed on one face to 20% HF and10% nitric acid solution at 25 C for 3 hrs. ..................................................67 3-5 Scanning electron micrographs of mica me mbrane that was exposed on one face to 40% HF and10% nitric acid solution at 25 C for 3 hrs. ..................................................67 3-6 Scanning electron micrograph of carbon tape red nanotube replica of the mica tapered pore........................................................................................................................... .........68 3-7 Low magnification SEM images of car bon tapered nanotube replicas of the mica tapered pore................................................................................................................... .....68 4-1 Definition of bulk etch rate Vb and track etch rate Vt.........................................................73 4-2 Schematic of cell used for electrochemical measurements................................................74 4-3 Pore diameters for diffe rent Au plating times....................................................................75 4-4 Pore diameter as a function of platin g time with measurements taken from SEM image (0-7.5 hrs) and ion current resistance measurements (8-12 hrs).............................76 4-5 Atomic Force micrographs of Kapton and PC membranes before and after electroless Au plating..................................................................................................................... ......77 5-1 The partitioning of a spherical molecule of radius, a in cylindrical pores of radius, r .....90 5-2 Spherical molecule of radius, a, moving within a cylindrical pore of radius, r.................91 5-3 Swift heavy ions impinge on a dielectric solid leading to damaged ion tracks witch can be chemically etched to form pores.............................................................................91 5-4 Definition of bulk etch rate Vb and track etch rate Vt.........................................................92 5-5 Conductivity cell used to prepare conical pores................................................................92 5-6 Etching curve showing moment of breakth rough with sharp increas e in ion current.......93 5-7 Kapton membran showing large op ening (base) of conical pores.....................................94

PAGE 10

10 5-8 Experimental set-up for transport measurements..............................................................94 5-9 Scanning electron micrograph of base side showing pore density (107 pores/cm2) of Kapton membrane..............................................................................................................95 5-10 Scanning electron micrograph of the base showing diameter of 1.68 m.........................95 5-11 A typical current-voltage curve used to measure the tip diameter of the conical nanopores...................................................................................................................... .....96 5-12 Base and tip fluxes for 1 mM phthalazine feed solution...................................................96 5-13 Base and tip fluxes for 3 mM phthalazine feed solution...................................................97 5-14 Base and tip fluxes for 10 mM phthalazine feed solution.................................................97 5-15 Base and tip fluxes for 50 mM phthalazine feed solution.................................................98 5-16 Asymmetric behavior of flux versus concentration...........................................................98 5-17 Theoretical and experimental tip flux vs concentration.....................................................99 5-18 Increase of flux through base and tip, resp ectively, with respect to the flux measured at 0.1 mM...................................................................................................................... .....99 5-19 Experimental set-up for transport meas urements of phthalazine with cosolute dextrose....................................................................................................................... .....100 5-20 Calibration curve of phthalazine, and phtha lazine with the highest concentration of dextrose cosolute used.....................................................................................................100 5-21 Viscosity of dextrose and phthalazi ne over the concentration range used......................101 5-22 Influence of dextrose cosolute concentr ations on the base and tip diffusion from 5 mM phthalazine...............................................................................................................101

PAGE 11

11 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 FABRICATION OF ASYMMETRIC PORES FOR BIOSENSORS AND TRANSPORT STUDIES By John Edwardson Wharton May 2007 Chair: Charles R. Martin Major: Chemistry The goals of this research are to develop asymmetric nanopores and nanotubes (single and multipore) in polymers and mica membranes by the track-etch method, and an extension of this technology, for sensing, nanostructure fabrication, and to investigat e the transport properties of conical pores. The first part of this work de scribes an extension of the track-etch method to make conical pores in a reproducib le fashion. We have demonstrat ed here that we can, not only reproducibly prepare trac k-etched based conical nanopore sens or elements, but that we can predict from the experimental parameters used during the second etch, what the diameter of the all-important nanopore tip will be. For these reas ons, we believe that the track-etch method will prove to be the technology of c hoice for taking artificial-nanopor e resistive-pulse sensors from the bench top to the practical protot ype-device stage of the R&D effort. The second part of this work describes a me thod to make asymmetric pores in tracked muscovite mica films using an etch-refill-etch a pproach. Tracks in the films were initially etched away with hydrofluoric acid to form na noporous membranes. These nanopores were then refilled with silver nanowires or metal tracks using an electroless plating method. One face of the membrane was then exposed to a solution of hydrofluoric acid and nitric acid, which etched the bulk material and the nanowires respectively, at two different rates. By controlling the

PAGE 12

12 concentration ratio of hydrofluoric acid to nitric acid, tapered pores with diamond shaped crosssection were obtained. Replicas of the asym metric pores were accomplished by carbon vapor deposition, and scanning electron microscopy was used to give evidence of the resulting nanotubes. In this study, excel lent control over tip size and cone angle was demonstrated. In the third section, electrol ess gold plating properties on the surface and pore walls of track-etched Kapton polyimide nanoporous memb ranes were studied. Scanning electron microscopy (SEM), atomic force microscopy (AFM), and ion current measurements were used to characterize the surfaces and di mensions of the pores in th e membrane. Nanoporous Kapton polyimide membranes were elect roless gold plated over different times and the pore diameter characterized using SEM. Ion current measurements were used to measure the diameter of very small pores. AFM images show that after electroless gold plating, the gold surface layers are smooth compared to a similarly structured poly carbonate membrane. Electroless plating the membranes for 12 hours produced Au wires in the pores. Removing the membrane by oxygen plasma revealed that the plating in the walls is also relatively smooth. The final part of this work describes asymmetric diffusional transport of neutral molecular species through nanoporous polymer membranes. Membranes containing conical nanopores from polyimide (Kapton HN Dupont) foils were prepar ed by the track-etching technique, based on irradiation of the polymer with swift heavy ions and subsequent etching of the latent tracks. The transport properties (e.g., flux) of these membranes were inves tigated using UV-Vis spectroscopy. Transport experiments were perf ormed with bare polymer membranes without any modification. We report the preferential direc tion of the flow of mo lecules through conical nanopores and spatial or concentration dependence of diffusion coefficient for a particular flow direction.

PAGE 13

13 CHAPTER 1 INTRODUCTION AND BACKGROUND Nanopore fabrication technology has produced nanoporous materials that are potential candidates for applications in vari ous fiels, such as bionanotechnology,1, 2 gas separation,3 catalysts4 and micro-electronics2, 5 In particular, the track-etch method has become an indispensable technology for the producti on of nanoand microstructured materials.6 This technology can be applied directly to most polymers and via th e replication techniqueto a wide variety of materi als, including metals.6 Track-etch membranes were first marketed three decades ago and remain the best product for a number of biological, medical, analytical and scientific applications.6 Recently, there has been a surge of interest in developing abiotic analogues of biological nanopores as sensing elements for chemicals and biological sensors.7 We have explored the fabrication of such synthetic nanopores using the tr ack-etch method. Particular interest is in the fabrication of synthetic conical pores for resistive-pul se sensing. Resistive-pulse sensing using conical nanopores is in its infa ncy. Studies in this area coul d prove beneficial for future applications of conical po res for science and technology. Membranes and porous materials have f ound various applications in filtration and separation processes.8, 9, 10 Modern biotechnology has posed ne w challenges in the application of such membranes, and requires pores with diamet ers similar to those of molecules under study,11 (e.g., as small as several nanometers). The nanomet er scale of such pores is necessary in both achieving optimal control of the flow of biomolecu les, as well as in developing sensors for their detection. The transport properties of such na nometer scale pores are not well understood yet. The hint that nanopores behave differently from micropores, comes from Mother Nature.12

PAGE 14

14 Biological channels and pores have diameter of ~ 1 nm and are cr itical for functioning of living organisms. This chapter is divided into six sections wh ich provide the background information for this research. Section 1 reviews th e track-etch method, first discovered by Price and Walker, for preparation of pores in dielectric materials. The track-etch method is used in the preparation of conical pores described in Chapters 2, 3, 4 and 5. Section 2 reviews membrane-based template synthesis. This method, pioneered by the Martin group, is used in preparation of nanostructured materials described in Chapters 2 and 3. Section 3 review resistive-pulse sensing; which is the focus of the application of Chapter 2 and 3. Section 4 describes plasma based etching; a procedure necessary for the libera tion of carbon tube in mica templates in Chapter 2. In Section 5, an introduction of Asymmetric Diffusion is given. Asymmetric diffu sion is the topic of Chapter 5. Finally, Section 6 is the dissertation overview. Ion Track-Etch Method Track Formation When dielectric materials, such as poly mers, ceramics and minerals, are bombarded by swift ions, latent tracks are formed al ong the path of the ions (Figure 1-1).13, 14 Ion track materials can be divided into two categories: (a) single-tracked and (b) multiple-tracked materials. Single-tracked materials can be pr oduced by controlling the beam optics and fluence of the heavy ion beam.15 Commonly tracked materials ar e Makrofol-KG, Kapton-H, PVDF, mica films, cellulose nitrate, CR-39 and Lexan polycarbonate. Swift io n beams are produced by cyclotrons or linear accelerator and the radiation is characteri zed by extremely high linear energy transfer.6 The conventional ionization radiation sources such as radioactive isotopes or electron accelerators,16 are less sophisticated and less expensive and are mostly used in the industrial

PAGE 15

15 processes.6 Present-day heavy ion accel erators provide beams with i on energies in the order of 10MeV/u and even 100MeV/u which expands the treatment dept up to millimeters.6 Acceleration and Irradiation Facilities About a half dozen heavy ion accelerators em ployed for irradiation of materials on the industrial scale exist. The Tanden Van de Graaff accelerators at Brookheaven National Laboratory are used to bombard materials with ions for manufacturing and testing purposes.6 At the Grand Accelerator National dIons Lours, Fran ce, ions are produced in an electron cyclotron resonance(ECR) source.6 The cyclotron of Louvain la Neuve is a multiparticle variable energy, cyclotron capable of accelerating protons, alpha particles and hea vy ions. The RFQ + cyclotron combination at the Hahn-Meitner Institute in Be rlin, delivers intense beams of ion species such as Kr or Xe with energies from approximately 1.5-6 MeV/u17 At the Flerov Laboratory of Nuclear Reactions(Dubna) a beam line connected at the U-400 cyclotron is equipped with scanning systems which allows one to obtain a homogeneous distributi on of ion tracks on the target up to 60 cm in width and 6 cm in height.6 Additionally, a line r accelerator at GSI (Darmstadt),6 the AVF cyclotron at TRCRE JAERI (Takasaki)6 and some others are used in experiments on polymer modification.6 Chemical Etching of Ion Tracks After irradiation, the materials are subject to chemical etching which preferentially removes the latent ion track (Figure 1-2)16 This etching process results in pore formation in the material.16 Etching is the pore-size -determining and pore-shapedetermining stage of the technology. In a homogeneous isot ropic medium, mainly two influe ntial parameter describe the etch processthe bulk etch rate VB and the track etch rate VT (Figure 1-3).6 The ratio of track etch rate to the bulk etch rate is called the track-etch-ratio. When VT is >> VB, pores turn out to be cylindrical as opposed to conical. In other word s, high track-etch-ratio yields cylindrical pores,

PAGE 16

PAGE 17

17 Etch Properties of Selected Polymers In PET (OOC-C6H4-COO-CH2CH2), the main points of etch attack are the partially charged COOester groups, which are hydrolyz ed by alkalis. During alkaline etching the ordinary bond between carbon and oxyge n is broken which produces COOand HOat the ends of the formed fragments. For polycarbonate (OOC-O-C6H4-C(CH3)2-C6H4), the main point of etch attack is the carbonate group O-COO-. During the alkaline etching, chemical bonds are ruptured on both sides of the carbonate group, leading to the formation of carbonate ions, CO3 2-. The other product is diphenylol HO-C6H4-C(CH3)2_C6H4-OH. In Polyimide (C6H4-OC6H4), the preferential point of et ch attack is the oxygen. At high pH the imide group will be hydrolyzed. The et ching mechanism is complex because of the simultaneous factors of oxidation and alkalinity. The chemical reaction responsible for etching different polyimide can be different becau se they are made up of monomer units. 19 Effect of Thermal Annealing By increasing the temperature of the tracked material, ion tracks can be thermally annealed. For polymers, heating above the gla ss transition temperature mobilizes the polymer fragments formed along the ion path. These fr agments are sucked in to the voids of the neighboring pristine material, wi ping out the latent ion track.14 The Effect of Temperature during Etching Etch rates usually increase with temperature. Therefore, to obtain high throughput makes it necessary to work at high temperatures. It is found that by alternati ng the temperature during etching, high aspect ratio pores with large diameters can be obtained. At room temperature diffusion processes can be faster than chemical reactions of the etchant within the polymer. 14

PAGE 18

18 From this basis, the technique to make pores wi th large cone angles we re develop by soaking at low temperature and etching at high temperature.20 Effect of Detergents Amphiphilic detergents may increase the track-etch-ratio. These detergents have been shown to produce nanopores with cigar-like shapes and very sma ll entrance openings. It is thought that amphiphilic detergents attach to the hydrophilic surface layer, rendering it less permeable to the etchant.21 Etch Properties in Muscovite Mica Damage tracks are form in single crystal muscovite mica (KAl2(AlSiO3O10)(OH)2 from ion irradiation along the (001) directi on. Because the etching rate along the tracks is much faster than both the lateral and bulb etch rate, when the tracks are etch ed through, nanopores with small cross-section and sm all taper angle (0.02 ) are created.22 In muscovite mica, all the pores are diamond-shaped with inner angles very close to 60 and 120 (Figure 1-5). All the pores in a given sample have the same size and orientation.22 By correlating th e results of scanning electron microscopy and X-ray diffraction on et ched mica crystals, it was found that the orientation of the diamonds is exactly the same as that of the mica unit cell. The four sides of the diamonds are parallel to the four oxygen-terminated planes within the unit cell. These facts points out that the diamondshaped pores have their origin in the mica crystal structure. Also, this shows that the uniform diamond shape aris es because the oxygen-terminated planes are those with the slowest etch rate, and the pores are aligned because th e template are single crystals.22 Template Synthesis The Martin group has pioneered a general me thod for the preparation of nanostructured materials called template synthesis23-25 Template synthesis method entails depositing a desired

PAGE 19

19 material of interest into a porous solid. The size and shape of the nanomaterial depend on the dimentions of the nanocavities within the porous template material. Depending on the membrane and synthetic material used, nanostructu res such as solid nanofibers or nanotubes can be obtained. The method is termed general because nearly any chemical synthesis method used to prepare bulk materials can be adapted to synthesize materials. There are reports of metals,10, 26-32 polymers,33-36 carbons37-39 and semiconductors40, 41 prepared by the template synthesis method. More advance material pr eparation include composite nanostructured material, both concentric and tubular42, 43 and segmented composites nanowires.44 Template Synthesis Strategies Three commonly used templates are; porous polymers, alumina and mica membranes (Figure 1-6). Some of the more common synthetic strategies us ed to prepare nanomaterials include chemical vapor deposition,38, 45, 46 electrochemical47-49 and electroless deposition,50, 51 chemical52 and electrochemical polymerization,53 and sol-gel chemistry.40, 54 Special attentions will be devoted to the electroless deposition in side polymers and mica, and chemical vapor deposition in mica since these are the methods used for preparing templates in chapters 2, 3, and 4 respectively. Electroless Deposition The electroless deposition met hod involves the use of a chemi cal reducing agent to plate a metal from a solution onto a surface. Unlike el ectrochemical deposition, a conductive surface is not necessary. The key requirement of electrole ss deposition is arrangement of the chemistry such that the kinetics of the homogenous electr on transfer from the reducing agent to the metal ions is slow. This is essential because the metal ions would simply be reduced in the bulk solution for fast electron transfer In electroless depos ition, a catalyst needs to be coated onto the pore walls so that reduction of th e metal ion only occurs at the pore surfaces. Figure 1-7 shows

PAGE 20

20 the schematic representation that was used to prepare silver and gold nanowires and nanotubes within PET, Kapton and Mica track-etched membrane s. The membranes were first exposed to a sensitizer (Sn2+). This is accomplished by simply immersing the membrane for 45 minutes in a solution that is 0.026 M in SnCl2 and 0.07 M in trifluoroacitic acid in 50/50 methanol/water. The tin sensitizer binds to the pore walls and memb rane surfaces via complexation with the amine, carbonyl and hydroxyl groups. 63 After sensitization, the membra ne is rinsed thoroughly with methanol and immersed into an aqueous solu tion of ammoniac silver nitrate (0.029 M Ag (NH3)2 +) for 5 minutes. A redox reation occ ours in which the surface bound tin ( ) is oxidized to tin ( V) and the Ag+ is reduced to elemental Ag. As a result, the pore walls and the membrane surface become coated with nanoscopic silver particles. The membrane is again thoroughly rinsed with methanol. The silver coated membrane is then immersed into a gold plating bath that is 7.9 x 10-3 M in Na3Au (SO3)2, 0.127 M Na2SO3, and 0.625 M in formaldehyde at 4 C. The Au galvanically displaces the Ag particles because the reduction potential of Au is more positive than that of Ag. As a result, the pore walls and surfaces become coated with Au particles. These particles are excellent catalyti c sites for the oxidation of fo rmaldehyde and the concurrent reduction of Au (I) to Au (0).63 Without a catalyst, the kinetics of the electron transfer from the reducing agent (formaldehyde) to Au (I) is slow ; therefore, the gold plating continues on gold particles instead of in the bulk solution. The reaction can be represented as follows: 2Au (I) + HCHO + 3OH HCOO+ 2H2O + 2Au (0) (1-1) This method yields the Au nanowires or nanot ubes within pores plus Au surface layers on both face of the membrane. These structures r un through the entire thic kness of the template membrane (Figure 1-8). By controlling the platin g time, the inside diameter of the tubes can be varied because the thickness of both the Au surface films and nanotube wall increase with

PAGE 21

21 plating time. By controlling the plating time, the inside diameter of the nanotubes can be varied, even as low as 1 nm in diameter.28 As a result, these membranes can be used in a simple membrane permeation experiment to cleanly separa te small molecules on the basis of molecular size.28 Also, by chemisorbing appropriate thiols to the Au nanotube wall based on well known gold-thiol chemistry, the Au nanotube membrane can be made to preferentially transport cations vs. anions and hydrophobic vs. hydrophilic molecules.10, 27, 31, 55 In addition, Au nanotube membranes are electronically conductive and can be charged electrostatically in an electrolyte solution.27 This introduces ion transport selectivity, allowing the Au nanotube membranes to be electromodulated between ideal-catio n and idea-anion transport states.27 Thus these Au nanotube memebranes are ideal model systems for studying how pore size, chemistry, and charge affect the transport selectivity at the nanometer scale. Chemical Vapor Deposition (CVD) CVD is commonly used to prep are carbon nanomaterials. We38, 56 and others57, 58 have synthesized carbon nanotubes with in the porous alumina membranes using CVD. This involves placing an alumina membrane in a high-temperature furnace (ca. 700 C) and passing a gas such as ethane, propene or ethylene through the memb rane. Thermal decomposition of the gas occurs on the pore walls, resulting in th e deposition of carbon nanotubes w ithin the pores. High surface area microporous carbon with long-range order has been synthesized by using zeolite Y as a template with propylene CVD.59 Besides carbon nanostructures, other nanomaterials have been obtained by CVD. For example, the martin group has used a CVD method to coat an ensemble of gold nanotubes with concentric TiS2 outer nanotubes.43 Resistive-Pulse Sensing Resistive-pulse60 sensors for molecular and macromolecules analytes60-77 use a nanopore in a synthetic or biological membra ne as the sensor element. This method, which when applied to

PAGE 22

22 such analytes is sometimes called stochastic sensing,60, 69 entails mounting the membrane containing the nanopore between two electrolyte solutions, applying a transmembrane potential difference, and measuring the resulting ion current flowing through the electrolyte-filled nanopore. In simplest terms, when the analyte en ters and translocates th e nanopore, it transiently blocks the ion current, resulting in a downward current pulse (Fi gure 1-9). The frequency of such analyte-induced current puls es is proportional to the concen tration of the analyte, and the identity of the analyte is encoded in the magnitude and duration of the current pulse.60-77 The majority of such resistiv e-pulse biosensing data has been obtained using a biological nanopore, -hemolysin ( -HL), embedded in a supported lipid-bilayer membrane as the sensor element (Figure 1-10).60-69 This biological nanopor e sensor has two key adva ntages. First, it can be made analyte selective by using chemical or genetic-engineering methods to attach molecularrecognition agents to the nanopore. As a result numerous different analyte types including metal ions,64 DNA,65, 66 proteins,67 and small molecules68 have been selectively detected with the -HL nanopore. Second, the biological nanopore can be reproducibly prepared from the commercially available -HL protein, which is obviously of great im portance if practical, real-world, sensing devices are ultimately to be derived from this technology.78 There is however a key impediment to developing practical sensors based on the biological nanopore. This problem concerns the fragility of the supported bilayer membrane that houses the nanopore. Such membranes typically survive for periods of only hours before rupture, much too short of a time to make a practical sensing device.69 One approach for solving this problem is to replace the biological nanopore, and bilayer membrane, with an artificial nanopore embedde d in a mechanically and chemically robust synthetic membrane.69-77, 79, 80 Such artificial nanopores are often prepared by modern

PAGE 23

23 microlithographic methods, using for example a focused ion71 or electron72 beam to bore the nanopore into a silicon or Si3N3 membrane. We and others are exploring an alternative technology, called the track-etch method,16, 19, 81, 82 for preparing nanopores for resistive-pulse sensors.70, 75-77, 79, 80 Analytes detected with prototype track-etched nanopore sensors include small molecules,70 DNA,75, 76 proteins77 and nanoparticles.79 Furthermore, there are older reports of developing virus sensors based on track-etched nanopores.80 The sensor elements we evaluated were conically shaped nanopores70, 75-77, 79, 81 prepared by the track-etch method in polyethylene terepht halate (PET) membranes. Such conical nanopores have two openings the large-diameter (base) opening at one face of the membrane and the small-diameter (tip) opening at the opposi te face (Figure 1-11). Fabrication methods to prepare these conically shaped nanopores for resis tive pulse sensing, is addressed in chapters 1. Plasma-Based Etching Plasma etching, a dry etching process, has b ecome a very useful means of removing small quantities of material from a variety of substrates quickly and efficiently.83 Plasma processes have been used in many highly sensitive integrat ed applications to precisely remove specific materials from sample surfaces. To generate pl asma, a pair of electrode is needed; one is connected to a radio frequency (R F) voltage and the other is gr ounded. RF energy is applied to the electrodes which accelerates el ectrons to increase their kinetic energy. The electrons collide with a neutral gas to form a collection of gase ous species including ions, free radicals, electrons, photons and neutrals.83 The gaseous species can react with the surface to be etched such that reaction product is volatile and can be pumped away. There are several types of plasma etchingphysical etching (anisotropic), chemical etchi ng (isotropic) an d reactive ion (combination of physical etching and chemical etching). Usually in plasma based etching, both chemical etching and physical etching occur.83

PAGE 24

24 There are many applications for plasma etch ing such as photoresist removal, glass-like compound etching (e.g., SiO2) and polymer etching to produce microstructures and nanostructures. For example, large-area, well-o rdered, periodic nanopillar arrays with lateral dimensions as small as 40 nm have been developed based on a combination of colloidal lithography and plasma etching techniques.84, 85 The etching mask on silicon substrates were prepared using the close-packed structures formed by monodisperse polystyrene beads.84, 85 Polymer surfaces can also be modified by plasma treatment, for example, to improve wetting properties and to enhance the adhesi on of plasma-deposited coatings.86 Plasma etching technique is described in Chapter 2 to remove carbon surface layers from mica tracked etched membrane that was previous ly exposed to CVD. This was necessary to expose the underlying mica surface for dissolution with HF so that the carbon nanotube can be revealed. Asymmetric Diffusion In some systems, diffusive transport thr ough membranes may in some circumstances pass more readily in one direction than the othe r. This phenomenon is known as asymmetric diffusion, and is known in the contex t of transport across membranes,87-89 and in the context of osmosis.90, 91 The above asymmetric phenomena is explai ned either by the binding of particles at intra-pore sites or other electrosta tic interaction with the pore surf ace. In this research, we show that asymmetric diffusion can occur with no binding or electrostatic interaction of particle with the pore surface. It is demonstrated in Chap ter 5 that asymmetric diffusion can take place by purely geometric constraints. He re, conical nanopores, with tip di ameter comparable to that of the diffusion molecule, are the used for asym metric diffusion of neutral molecules.

PAGE 25

25 Dissertation Overview The goal of this research is to develo p conical nanopores and nanotube (single and multipore) in polymers and mica membranes by th e track-etch method and extension of this technology for sensing and other app lications, and to inve stigate the tr ansport properties of these conical pores. The previous part of Chapter 1 has reviewed background information for this dissertation including the ion tr ack-etch method, membrane based te mplate synthesis, electroless metal deposition, chemical vapor deposition, resi stive-pulse sensing, plasma based etching and asymmetric diffusion. In Chapter 2, an extension of the track-etch method to make conical pores in a reproducible fashion is demonstrated. We have shown here th at we can not only repr oducibly prepare tracketched based conical nanopore sens or elements, but that we can predict from the experimental parameters used during the second etch, what the diameter of the all-important nanopore tip will be. For these reasons, we believe that the track -etch method will prove to be the technology of choice for taking artificial-nanopore resistive-pulse sensors from the bench top to the practical prototype-device stage of the R&D effort. Chapter 3 describes a method to make asymme tric pores in tracked muscovite mica films using an etch-refill-etch approach. Tracks in the films were initially etched away with hydrofluoric acid to form nanoporous membranes. These nanopores were then refilled with silver nanowires or metal tr acks using an electroless pl ating method. One face of the membrane was then exposed to a solution of hydr ofluoric acid and nitric acid, which etched the bulk material and the nanowires respectively, at two different rates. By controlling the concentration ratio of hydrofluoric acid to nitric acid, tapered pores with diamond shaped crosssection were obtained. Replicas of the asym metric pores were accomplished by carbon vapor

PAGE 26

26 deposition, and scanning electron microscopy was used to give evidence of the resulting nanotubes. In this study, excel lent control over tip size and cone angle was demonstrated. In Chapter 4, electroless gol d plating properties on the su rface and pore walls of tracketched Kapton polyimide nanoporous membranes we re studied. Scanning electron microscopy (SEM), atomic force microscopy (AFM), and ion current measurements were used to characterize the surfaces and por e dimensions of the membrane. Nanoporous Kapton polyimide membranes were electroless gold plated over different times and the pore diameter characterized using SEM. Ion current measurements were used to measure the diameter of very small pores. AFM images show that after electroless gold plating, the gold surface layers are smooth compared to a similarly structured polycarbonat e membrane. Electroless plating the membranes for 12 hours produced Au wires in the pores Removing the membrane by oxygen plasma revealed that the plating in the walls is also relatively smooth. Finally, Chapter 5 describes asymmetric diffusi onal transport of neutral molecular species through nanoporous polymer membranes. Me mbranes containing conical nanopores from polyimide (Kapton HN Dupont) foils were prepar ed by the track-etching technique, based on irradiation of the polymer with sw ift heavy ions and subsequent etch ing of the latent tracks. The transport properties (e.g, flux) of these membranes were investigated using UV-Vis spectroscopy. Transport experiments were perf ormed with bare polymer membranes without any modification. We report the preferential direc tion of the flow of mo lecules through conical nanopores and spatial or concentration dependence of diffusion coefficient for a particular flow direction.

PAGE 27

27 Figure 1-1. Swift heavy ions impinge on a diel ectric material creating damaged ion tracks. Figure 1-2. After irradiation, the ma terials are subject to chemical et ching which preferentially removes the latent ion track. Etch-stop solution Etchant Etch-stop solution Etchant Etch-stop solution Etchant Etch-stop solution Etchant

PAGE 28

28 Figure 1-3. Etched pore geometry in a homogeneous isotropic medium to a first approximation, showing track etch rate, VT, and bulk at rate, VB. A B C A B C Figure 1-4. Scanning electron mi crographs of Au nanocones repl icas of PET conical pores. Increasing fraction of ethanol in the etch solution results in greater cone angle (AC).[Adapted from Scopece, P.; Baker, L. A.; Ugo, P.; Martin, C. R. Nanotechnology 2006 17 3951-3956.]

PAGE 29

29 Figure 1-5. Scanning electron micrographs of et ched particle tracks in single-crystal mica

PAGE 30

30 A B C A B C Figure 1-6. Scanning electr on micrographs of a porous polycarbonate, alumina and mica membranes used for template synthesis (A-C, respectively).

PAGE 31

31 Figure 1-7. Schematic diagram of Au electroless plating procedure

PAGE 32

32 Side View Electroless Gold Plating Membrane Pores Gold tubes lining pores Side View Electroless Plating Membrane Pores Au nanotubes lining the pores Electroless Gold Plating Membrane face Pore Gold surface layer Top View Top View Membrane Face Pore Electroless Plating Gold surface film Side View Electroless Gold Plating Membrane Pores Gold tubes lining pores Side View Electroless Plating Membrane Pores Au nanotubes lining the pores Electroless Gold Plating Membrane face Pore Gold surface layer Top View Top View Membrane Face Pore Electroless Plating Gold surface film Figure 1-8. Schematic illustration of Au na notubes obtained from electroless gold deposition

PAGE 33

33 Figure 1-9. Schematic illustrati on of resistive-pulse sensing

PAGE 34

34 Figure 1-10. Essential features of the staphylococcal hemolysin pore shown in a crosssection based on the crystal structure. [Adapted from Bayley, H.; Martin, C. R. Chemical Reviews (Washington, D. C.) 2000 100 2575-2594.]

PAGE 35

35 Figure 1-11. Schematic of a conical nanopore se nsor element showing the base diameter and range of tip diameters used in these studies (drawing to scale)

PAGE 36

36 CHAPTER 2 A METHOD FOR REPRODUCIBLY PREPARING SYNTHETIC NANOPORES FOR RESISTIVE-PULSE BIOSENSORS Introduction Resistive-pulse60 sensors for molecular and macromolecules analytes60-77 use a nanopore in a synthetic or biological membra ne as the sensor element. This method, which when applied to such analytes is sometimes called stochastic sensing,60-69 entails mounting the membrane containing the nanopore between two electrolyte solutions, applying a transmembrane potential difference, and measuring the resulting ion current flowing through the electrolyte-filled nanopore. In simplest terms, when the analyte en ters and translocates th e nanopore, it transiently blocks the ion current, resulting in a downward current pulse. The frequency of such analyteinduced current pulses is proportional to the concen tration of the analyte, and the identity of the analyte is encoded in the magnitude and duration of the current pulse.60-77 The majority of such resistiv e-pulse biosensing data has been obtained using a biological nanopore, -hemolysin ( -HL), embedded in a supported lipid-bilayer membrane as the sensor element.60-69 A key advantage of this biological-na nopore sensor element is that it can be reproducibly prepared from the commercially available -HL protein. This is of great importance if practical, real-world, sensing devi ces are ultimately to be derived from this technology.78 There is, however, a key impediment to developing practical sensors based on the biological nanopore. This problem concerns the fragility of the supported bilayer membrane that houses the nanopore. Such membranes typically survive for periods of only hours before rupture, much too short of a time to make a practical sensing device.69 One approach for solving this problem is to replace the biological nanopore, and bilayer membrane, with an artificial nanopore embedde d in a mechanically and chemically robust synthetic membrane.69-77, 79, 80 Such artificial nanopores are often prepared by microlithographic

PAGE 37

37 methods, using for example a focused ion71 or electron72 beam to bore the na nopore into a silicon or Si3N3 membrane. We and others are exploring an alternative technology, called the track-etch method,16, 19, 81, 82 for preparing nanopores fo r resistive-pulse sensors.70, 75-77, 79, 80 Analytes detected with prototype track-etched nanopore sensors include small molecules,70 DNA,75, 76 proteins77 and nanoparticles.79 Furthermore, there are older re ports of developing virus sensors based on track-etched nanopores.80 This field of artificial-nanopor e resistive-pulse sensing is cu rrently in its infancy. A key question that must be addressed before practical sensors can be developed is can the nanopore sensor element be prepared reproducibly, as the biological nanopore can? We address this critically important issue here. The sensor elements we evaluated were conically shaped nanopores70, 75-77, 79, 81 prepared by the track-etch method in polyethylene terepht halate (PET) membranes. Such conical nanopores have two openings the large-diameter (base) opening at one face of the membrane and the small-diameter (tip) opening at the opposite face (Figure 2-1a). We have found that the diameters of both of these opening can be contro lled with good reproducib ility using a new twostep pore-etching procedure. Furthermore, we ha ve developed a simple mathematical model that allows us to predict the diameter of the tip opening from the parameters used during pore etching. Good agreement was obtained between th e predicted and experimentally measured tip diameters. Experimental First Etch Step The tracked PET membrane (from GSI) was mo unted in cell shown in Figure 2-1b, and the etch solution (9 M NaOH) was placed in one half cell and the stop solution (1M formic acid plus 1 M KCl) in the other. A platinum wire el ectrode was placed in each solution and a Keithley

PAGE 38

38 6487 was used to apply a transmembrane potential difference of 1V during etching, with polarity such that the anode was in the etch solution. The electrochemical reac tions occurring at the anode and cathode are discussed in the Supplementa ry Materials. Etchin g was terminated after two hours by replacing the contents of the etch half-cell with stop-etch solution. The membrane was then rinsed with purified water (Barnstead D4641, E-pure filters). Determination of the Diamet er of the Base Opening Multi-track membranes (106 cm-2) were subjected to the same first etch step as used for the single-track membranes, and th e base openings were imaged via FESEM (JEOL JSM-6335F). The average base diameter obtained (520+ 45 nm) is associated with measurements of 50 pores in five different multi-track membrane samples. Electrochemical Measurement of the Tip Diameter The membrane was mounted in the cell, and the half cells were filled with 100 mM phosphate-buffered saline, pH = 7.0 that was also 1 M in KCl. The specific conductivity of this solution was measured using a conductometer (YSI 3200) at 25 C; a conductivity of 0.107 S cm-1 was obtained. A Ag/AgCl electrode immersed into each solution was used in conjunction with the Kiethly 6487 to obtain the current-volta ge curve for the nanopore (Figure 2-2). We have validated this electrochemical method by comparing diameters obtained via this method with diameters for the same pores measured by electron microscopy.92, 93 Second Etch Step The etch solution in this case was 1 M NaOH, and its conductivity was measured at 0.160 S cm-1 (25 C). The membrane was mounted in the cel l, and the half cells were filled with this etch solution. A platinum electro de was immersed into each ha lf-cell solution, and the Keithley 6487 was used to apply a transmembrane potenti al difference of 1V and measure the nanopore ion current as a function of time. Etching was te rminated at specified current values (Figure 2-3,

PAGE 39

39 Figure 2-4) by replacing the etch solution in both half cells with the stop-etch solution. The membrane remained in the stop etch for at least 30 min and was then rinsed with purified water. Bovine Serum Albumin (BSA) Resistive-Pulse Sensing Conical nanopores sensors ha ving two different tip diameters, 58 nm and 44 nm, were prepared; the base diameter fo r both sensors was 520 nm. After etching the pore walls were coated with gold nanotubes using the electr oless plating method de scribed previously.28 After electroless plating the tip diameters of the re sulting gold nanotubes were measured using the electrochemical method discussed above; tip diameters of 32 nm and 23 nm were obtained. The Au-coated nanopore walls were th en functionalized with a thio lated PEG (MW 5000 Da, Nektar Therapeutics) to prevent non-specific protein adsorption.94 This was accomplished by immersing the nanotube-containing membrane in a 0.1 mM solution of the PEG at 4 oC for ~15 hours. The membrane was then immersed in purified water for 1.5 hours to remove any unbound PEG. The tip diameters were then re-measur ed; values of 27 nm and 17 nm were obtained. The BSA (Sigma) was dissolved in 10 mM phosphate-buffered saline that was also 100 mM in KCl (pH 7.4). The concentration of BS A was 100 nM, and the BSA solution was placed on the tip side of the membrane. Buffer was placed on the base side and a transmembrane potential of 1000 mV was used to drive the protein through the nanopor e (tip to base) by electrophoresis. Results and Discussion Conical Shaped Nanopores are Ideal Resistive-Pulse Sensor Elements Resistive-pulse sensing entails mounting the nanopore membrane between the two halves of an electrolyte-filled cell (Figure 2-1 (b)) and passing an io nic current through the electrolyteflooded nanopore. In a conically shaped nanopore the voltage drop caused by this ion current is focused to the electrolyte solution in the tip opening of the nanopore.79 Indeed, the field strength

PAGE 40

40 in the solution within the nanopor e tip can be greater than 106 V m-1, when the total voltage drop across the nanopore membrane is only 1 V.79 A consequence of this fo cusing effect is that the nanopore ion current is extremely sensitive to anal yte species present in the nanopore tip. That is, there is an analyte sensing zone just insi de the tip, which makes conically shaped nanopores ideally suited for the resistive-pulse sensing ap plication. This has b een demonstrated with prototype conical-nanopore sensors for analyte speci es ranging in size from small molecules, to proteins, to nanoparticles.70, 75-77, 79 The Core Technology: The Track-Etch Method The track-etch method has been practiced comm ercially for decades to make polymeric nanopore membranes for filtration applications.16, 19, 81, 82 It entails passing high energy particles through the membrane, to create damage tracks, followed by chemical etching to convert these damage tracks into pores. While the commercia l process yields membranes that contain high pore densities, a method for preparing single-damage-track membranes was developed at the Gellsellschaft fur Schw erionenforschung (GSI).19 We purchased such single-track PET membranes from GSI for these studies. This is an important point with regard to the overall sensor-fabrication technology the key precu rsor material, the tracked membrane, can be obtained commercially. The First Etch Step Both etch steps used the cell shown in Figure 2-1b. Step 1 entails placing a solution that etches the damage track on one si de of the membrane and a soluti on that neutralizes this etchant on the other side.81 For PET the etchant is NaOH, and the etch-stop is formic acid. This yields a conically shaped pore (Figure 21a) with the base opening facing the etch solution and the tip opening facing the etch-stop solution. To determ ine when the etchant ha s broken through to the etch-stop solution, and a contiguous pore has been obtained, an el ectrode is placed in each

PAGE 41

41 solution, and a potential difference is applied across the membrane. Before breakthrough, the transmembrane ion current is zer o, and breakthrough is signaled by a sudden rise in the current.81 We previously showed that the diameter of the base opening could be controlled by varying the potential applied across the membrane during this first etch step.92 An applied transmembrane potential of 1.0 V was used in th e first etch step for all of the nanopores investigated here. To obtain a measure of the re producibility of the base diameter obtained after the first etch step, we subjected multi-track membranes (106 tracks cm-2) to the same first etch as used for the single-track membranes, and imag ed the base openings using field emission scanning electron microscopy (FESEM, Figure 2-4) We used multi-track membranes for this study because it is difficult to locate the base opening in electron micrographs of a singlenanopore membrane. We have previously shown that the pore diameter obtained for trackedetch membranes is independent of track density.95 A base diameter of 520+ 45 nm was obtained, indicating good reproducibility in base di ameter after the first etch step. However, the tip diameter varied between 1 and 7 nm and could not be reproduced from etch to etch. We reasoned that th is is an inherent feature of this anisotropic etch process. This is because the etch and etch-stop solutions ar e mixing (and neutralizing each other) in the nascent tip, which makes it difficult to control the etch rate in this critically important region of the nanopore. Measuring the Diameter of the Tip Opening after the First Etch Because the tips after the first etch are so small, they are very difficult to find and image via electron microscopy. Therefore, an el ectrochemical method described previously70, 76, 81, 92 was used to measure the diameter of the tip. Th is entailed mounting the membrane sample in the cell (Figure 2-1b), filling both half cells with an electrolyte so lution of known ionic conductivity,

PAGE 42

42 and obtaining a current-voltage (I-V) curve a ssociated with ion-tran sport through the nanopore (Figure 2-2). The experimental slope of this linear I-V curve is the ionic conductance, G1, (in Siemens, S) of the nanopore, which is given by81 G1 = ( KCl db dt) / 4 L (2-1) where KCl is the experimentally measured conductivity of the KCl-based electrolyte used (S cm1), L is the length of the nanopore (membrane th ickness), db is the experimentally measured diameter of the base opening, and dt is the diameter of the tip ope ning. Because all of the other parameters in Equation 1 are known, dt can be calculated. The Second Etch Step In the second step NaOH etch solution is pl aced on both sides of the membrane. Again, a transmembrane potential is ap plied, and the ion current fl owing through the nanopore is measured as a function of time during this etch. Our key innovation is that the second etch is stopped at a prescribed value of this nanopore ion current rather than at some prescribed time after starting the second etch (e.g., Figure 2-3). We adopted this approach because of the variability in tip diameter obtained after the first etch step. The c onsequence of this variability is that if we stopped the second etch at a prescribed time, we would obtain a corresponding variability in the tip diameters obtained after the second etch step. In contrast, as we will see below, there is an exact mathematically relati onship between the ion cu rrent flowing through the nanopore when the second etch is stopped (If) and the diameter of tip opening. To prove that the pores obtained after the s econd etch truly are conically shaped, we used an electroless plating method28 to deposit corres pondingly conically shaped gold nanotubes within the pores of the multipore membranes de scribed above. The PET membrane was then

PAGE 43

43 dissolved and the conical nanopores colle cted by filtration and imaged by FESEM.18 These images show that a nearly ideal conically shaped pore is obtained (Figure 2-5). Measuring the Diameter of the Tip Opening after the Second Etch The same electrochemical method was used, but the mathematics is slightly different. We define the diameters of the base and tip openings after the first etch step as db1 and dt1 and the diameters after the second etch as dbf and dtf. These diameters are related via dbf = (db1 + x) (2-2) dtf = (dt1 + x) (2-3) where x is the change in diameter during the se cond etch. Substituting Equations 2-2 and 3-3 for db and dt in Equation 2-1 yields G2 = ( KCl (db1 + x)(dt1 + x)) / 4 L (2-4) where G2 is the slope of the current-voltage curve us ed to determine the tip diameter after the second etch, db1 and dt1 are the experimentally determined base and tip diameters, respectively, after the first etch, KCl is the experimentally measured cond uctivity of the electrolyte used (100 mM phosphate-buffered saline, pH 7, that was also 1 M in KCl; = 0.107 S cm-1), L is the membrane thickness, and x is the change in diameter between the first and second etch steps. We define the parameter M as L MKCl4 (2-5) which allows us to write, after some simple algebraic manipulation, 2 1 1 1 1 2) ( x x d d d d M Gt b t b (2-6) Substituting (dtf dt1) for x (where dtf is the diameter of the tip opening after the second etch), and again applying some simple algebraic manipulations, yield

PAGE 44

44 0 ) (2 1 1 2 M G d d d dtf t b tf (2-7) This is a quadratic equation in dtf for which the solution is 2 / 4 ) ( ) (2 2 1 1 1 1M G d d d d dt b t b tf (2-8) Note that the quadratic formula has two roots; i.e., there should be a + instead of a + between the two terms in the numerator of Equation 2-8. Howe ver, the root that results when subtraction is used yields a negative valu e for the tip diameter. Because all of the parameters on th e RHS of Equation 2-8 are known, dtf can be calculated. Furthermore, because the base diameter at the start of the second etch is large (520 nm), the change in base diameter during the second etch ( x in Equation 2-2) is negligibly small for all but the very largest tip investigat ed here (60 nm, Figure 2-6). Reproducibly Varying the Tip Diameter Figure 2-6 shows a plot of nanopor e tip diameter, measured after the second etch step (Eq. 2-5), vs. the nanopore ion current at which this etch was stopped (If). We see that the tip diameter can be reproducibly vari ed over the range from 10 to 60 nm (data points in Figure 2-6). This is important because this is exactly the ra nge in tip diameters we used in our prototype protein,77 DNA,76 and nanoparticles79 sensors. The lower limit (10 nm) is determined by the tip diameter obtained after the first etch which, again, was in the range of 1 to 7 nm. However, we have shown that the walls of such na nopores can be lined with gold nanotubes,77 and that the diameter of these tubes can be c ontrolled at will down to 1 nm.96 Hence, if tip diameters smaller than 10 nm are needed, a pore with a 10 nm tip can be gold plated to reduce the tip to any desired value. Furthermore, tips larger than the 60 nm maximum shown in Figure 2-6 can be easily prepared by simply stopping the sec ond etch at larger values of If.

PAGE 45

45 The Mathematical Model We begin by defining a new conductance, Getch, which is the ion current at which the second etch is stopped (If) divided by the transmembrane potential applied during the second etch (Eap). With this definition, Eq uation 4 can be rewritten as If = Eap ( etch (db1 + x)(dt1 + x)) / 4 L (2-9) where etch is the experimentally meas ured conductivity of the NaOH solution used in the second etch. Equation 2-9 is agai n a quadratic Equation in dtf for which the solution is 2 / 4 ) ( ) (2 1 1 1 1K I d d d d df t b t b tf (2-10) where K = Eap etch /4L (see Derivation of Equation 2-5, be low). Equation 2-10 allows us to calculate the value of the tip diam eter after the second etch step (dtf) for any value of If at which the second etch was stopped. We noted above that the base diameter before and after the second et ch is essentially the same for all but the largest tip in Figure 2-6. This allows us to simplify Equation 2-9 to If = Eap etch db1 dtf / 4 L (2-11) which can be rearranged to dtf = If 4 L/ (Eap etch db1 ) (2-12) This obviously provides a much simpler relationship between dtf and If. Plots of dtf vs. If calculated using the simplified e quation (Equation 2-12) and the exact equation (Equation 2-10) are shown as the two solid lines in Fi gure 2-6. The tip diameters calculated by these two equations are identical for tips below ~20 nm. Furthermore, the agreement between the experimentally measured (Equation 2-8) and theoretically calculated (Equation 2-10) tip diameters is good, especially considering that there are no adjustable

PAGE 46

46 parameters in the calculations. For example, at If = 20 nA the experimental and calculated tip diameters differ by less than 10%, and at 40 nA they are identical. The calculated tip-diameter values are, in general, slightly smaller than the experimental values. This results from an interesting featur e of the transport properties of conical nanopores if there is charge on th e pore wall and if the tip opening is small, such nanopores act as ion current rectifiers.97 The consequence of this rectifi cation phenomenon is that the ionic conductivity of an electroly te solution within the tip of the nanopore can be lower than the value measured for a bulk sample of the same elect rolyte. Since we used the bulk-solution conductivity in our calculations, the calculated values are in genera l low. Excellent agreement is obtained between the experimental and calculate d diameters for the largest tip (Figure 2-6) because large-tip pores do not rectify the ion current.98 We believe that procedures to obviate the small disagreement between the experimental and calculated tip diameters can be developed, and we are currently pursuing this issue. To illustrate the importance of controlling the tip diameter in resistive-pulse sensing, we obtained current-pulse data for a prototype protein analyte, bovi ne serum albumin (BSA), with nanopore sensors having two different tip diameters. The sensors in this case were conical PET nanopores that had been lined with gold nanotubes28, 77 and then coated with a poly(ethylene glycol thiol) (PEG) to preven t nonspecific protein adsorption.94 The tip diameters, 17 nm and 27 nm, were measured after PEG functionalization. Figure 2-7 shows current-pulse data obtained for BSA with th ese two different sensors. The current-pulse signature can be define d by the average duration and magnitude ( I) of the current pulses. The magnitude of the current pulse is important because if I is not larger than the peak-to-peak noise in the background current, the current pulse will be undetectable. We

PAGE 47

47 found that I is larger for the nanopore sensor with the smaller tip opening ( I = 80+ 20 pA) than for the sensor with the larger tip opening ( I = 35+ 9 pA). This is because the roughly 4 nm x 4 nm x 14 nm94 BSA molecule more effectively blocks the ion current as it translocates the smaller, 17 nm, tip. Electrochemical Details Pt electrodes were used to a pply the transmembrane potential di fference in both of the etch steps, and the applied potential was 1.0 V in bot h cases. The half reactio n occurring at the Pt cathode was the reduction of the dissolved O2 in the solution. O2 + 4H+ + 4 e2H20 (2-13) The low (nA-level) currents, a nd the fact that the solutions were exposed to air during etching, insured that the O2 was not depleted. The half reacti on occurring at the Pt anode was the reverse of Equation 2-15. Conclusions In his review of nanowire-based chemical a nd biosensors, Lieber stresses the importance of being able to reproducibly pr epare the nanowire sensing element.78 The same is true for artificial nanopores to be used as resistive-pulse sensor elements. We have shown here that we can not only reproducibly prepare track-etched ba sed conical nanopore sensor elements, but that we can predict from the experimental parameters used during the second et ch, what the diameter of the all-important nanopore tip will be. For these reasons, we believe that the track-etch method will prove to be the technology of choice for taking artificial-nanopore resistive-pulse sensors from the bench top to the practical prototype-device stage of the R&D effort.

PAGE 48

48 PET membrane 12 m Conical nanopore Base opening(520 nm) Tip opening(10 to 60 nm) a A PET membrane 12 m Conical nanopore Base opening(520 nm) Tip opening(10 to 60 nm) a PET membrane 12 m Conical nanopore Base opening(520 nm) Tip opening(10 to 60 nm) a A Poly(chlorotrifluoroethylene) cell Pressure plate 3.5 cm 1.0 cm 3.5 cm 1.0 cm Electrodes Membrane 3.5 cm 1.0 cm b Aluminum frame Clamping screw Solution chamber B Poly(chlorotrifluoroethylene) cell Pressure plate 3.5 cm 1.0 cm 3.5 cm 1.0 cm Electrodes Membrane 3.5 cm 1.0 cm b Aluminum frame Clamping screw Solution chamber Poly(chlorotrifluoroethylene) cell Pressure plate 3.5 cm 1.0 cm 3.5 cm 1.0 cm Electrodes Membrane 3.5 cm 1.0 cm b Aluminum frame Clamping screw Solution chamber B Figure 2-1. Schematic of a conical nanopore sensor element a nd etching cell. (A) The base diameter and range of tip diameters used in these studies (drawing not to scale). (B) Cell used to do the etching and to ma ke all electrochemical measurements.

PAGE 49

49 -2 -1 0 1 2 -0.2-0.100.10.2Applied transmembrane potential (V)Nanopore ion current (nA)a -2 -1 0 1 2 -0.2-0.100.10.2Applied transmembrane potential (V)Nanopore ion current (nA)a -2 -1 0 1 2 -0.2-0.100.10.2Applied transmembrane potential (V)Nanopore ion current (nA)a -2 -1 0 1 2 -0.2-0.100.10.2Applied transmembrane potential (V)Nanopore ion current (nA)a Figure 2-2. A typical current-voltage curve used to measure the tip diameter of the conical nanopore.

PAGE 50

50 020406080100120140 0 5 10 15 20 25 30 35 40 45 Nanopore ion current (nA)Etch time (minutes) Figure 2-3. Current-time transients obtained during th e second etch step for three membranes that were subjected to the same first et ch. The second etch was stopped in each case when a final nanopore ion current (If) of 40 nA was obtained.

PAGE 51

51 Figure 2-4. Scanning electron mi crograph of the base openings of two conical nanopores in a multi-track PET membrane that had been etch ed (first etch step) as per the singletrack membranes used in these studies.

PAGE 52

52 Figure 2-5. Scanning electron micrograph of conical gold na notubes deposited in a conical nanopore membrane.

PAGE 53

53 0 10 20 30 40 50 60 70 80 010203040Final nanopore ion current during second etch, If(nA)Tip diameter (nm) Figure 2-6. Plot of tip diameter measured after the second etch step vs. the final nanopore ion current (If) at which the second etch was stopped. The points are the experimentally measured tip diameters. The error bars are measurements on three different membrane samples prepared identically. The solid curves were calculated using the simplified equation (Equation 2-12, red curv e) and the exact e quation (Equation 2-10, blue curve).

PAGE 54

54 100 s 20 pA 100 s 20 pAa b A B 100 s 20 pA 100 s 20 pAa b 100 s 20 pA 100 s 20 pAa b A B Figure 2-7. Current-pulse data obtained for a prototype protein analyte, bovine serum albumin (BSA) using PEG-modified conical nanotube sensors. (A) Tip diameters of 17 nm and (B) Tip diameters of 27 nm. BSA concentration = 100 nM. Applied transmembrane = 1000 mV.

PAGE 55

55 CHAPTER 3 ETCH-FILL-ETCH METHOD FOR PREPARING TAPERED PORES IN ION TRACKED MICA FILMS Introduction The Martin research group and others have been investigating a general method for preparing nanomaterials know n as template synthesis.23, 96 This method entails the synthesis or deposition of a desired material within the po res of a nanopore membrane that serves as a template. These template membranes contain mo nodisperse pores that ar e typically cylindrical in geometry, and the pore diameter can be varied at will from tens of nanometers to tens of microns. Since typical pore geometries are cylindrical, corr espondingly cylindrical nanostructures are usually synthesized via the template method; depending on the membrane and synthetic method used, these may be solid nanowires or hollow nanotubes. 99 Recently, our research team and others have become interested in nanopores that have a conical pore shape and the correspondingly conical nanostructures synthesized via the template method within these pores. A number of applic ations can potentially benefit from conical pore geometry. For instance, it has been shown that su ch conically shaped nanopores can be used as the sensing element for new types of small molecule, 70 DNA, 75, 98, 100, 101 protein 77 and particle79 sensors. Conically shaped gold nanotubes deposited within such pores can also act as mimics of voltage gated ion channels. 102 Membranes used for separations might also benefit from a highly asymmetric pore structure. Fi nally, in addition to sensing and separations platforms, conical nanostructures prepared by more conventional methods have been proposed for use as cathodes in field-emission displays. 103 To date, the vast ma jority of conical pores have been fabricated using tracked polymers. 81, 89, 93, 95, 97, 104, 105 However, we are interested in exploring materials other than polymers to use as conical pore templates because their different properties may prove potentially superior for cert ain applications, and other applications might

PAGE 56

56 be realized that are not possibl e with polymers. One such material of interest is tracked muscovite mica. Because mica is an inorganic crystalline mate rial, it possesses some properties not present in polymers that might make it superior in some aspects for certain applications. For example, the surface of mica is molecularly flat, 106 and is a good candidate for platform for AFM imaging of DNA, 106 and support layer for lipid bilayers. 107 Also, mica is very chemically resistant and has high thermal and mechanical stability. 22, 108 Mica conical pores might prove much more stable for resistive pulse sensor s than polymers. Furthermore, these properties make it possible for template synthesis of materials that requir e high temperatures. Additionally, nanostructures that demand special geometry that is diffi cult to obtain by conven tional methods may be realized. For instance, tapered-shape carbon structure can provide mechanical stability yet provide very sharp tips, which may be usef ul for enhanced electron field emission. 109 We and other have shown that the well-known track-etch method 16 can be used as a starting place for prepari ng such conical nanopores. 81, 89, 93, 95, 97, 104, 105 This method entails bombarding a thin film (5-20 m) film of the material with a collimated beam of high-energy particles to create parallel damaged tracks throug h the film. To make cylindrical pores, the tracked membrane is simply immersed into a ch emical etch bath, wher e preferential etching along the damaged track converts each track into a cylindrical pore. To make conical pores, the tracked membrane is mounted in an etching cell with an etch solu tion on one side of the tracked membrane and a stop solution on the other side.81 This is shown schematically in Figure 3-1. Since the damaged track is etched at a longe r duration time and faster at the face of the membrane exposed to the etch solution than at the face of the membrane exposed to the stop solution, conically shaped nanopores are obtained.

PAGE 57

57 While this stop-etch approach has been successf ul in making conical na nopores in a variety of polymer materials, the method fails for the pr eparation of conical pores in mica membranes. The key impediment is due to the ratio of the trac k to bulk etching rate. The track etch rate is much faster than the bulk etch ra te in mica (about 3000 time faster), 108 so that the etch solution traverses the entire membrane before any signifi cant bulk etching takes place. This means that all parts of the membrane start etching almost at the same time in an isotropic fashion giving rise to uniform pores instead of asymmetrical pore shap e. One approach to solve this problem is to replace the tracks of the mica films with a material that is more controllably etched, for instance metal nanowires. In this study, we developed a method to independently control the solution etch rate traversing the membrane, and also the etch rate of the surrounding bulk material, to give asymmetric pores. We prove this by using car bon vapor deposition to replicate the pores, dissolving the membrane to expose the tapered tubes, and imaging the nanotubes using scanning electron microscopy. The results of th ese investigations are reported here. Experimental Materials Muscovite mica wafers (1.181X 0.0004 inches ) were purchased from Spruce Pine Co. USA. Then these bare wafers were irradiated by swift heavy U25+ ions of 2.2GeV kinetic energy with fluence of 104 to107 cm-2 (GSI Darmstadt, Germany), which produced damage tracks through the mica membranes. H ydrofluoric acid (HF, 48~51% from ACROS), for etching and dissolving mica membranes, was used as received. Anhydrous tin (II) chloride 98% (Aldrich) and hydrochloric acid (A CROS) were used as received to sensitize the mica membranes for electroless plating. Ammonium hydroxide (Fis her), silver nitrate (M allinckrodt), potassium sodium tartrate tetrahydrade (Aldrich), and magn esium sulfate (Fisher) were used as received to prepare silver plating so lution. Ethylene (30% balanced with Helium, from Praxair) was used as

PAGE 58

58 the CVD carbon precursor gas. Purified water wa s prepared by passing house-distilled water through a Millipore Milli-Q water purification system. Initial Etching of Mica Tracks to Prepare Very Small Pores Mica wafers, containing damaged tracks, were exposed to low concentration HF solution on both faces to create pores of about 10nm in diameter. The wafers were sandwiched between two half cells of a conductivity ce ll (Figure 3-2.) and 2% HF at 25 C were placed in each half cell for a period of 10 minutes. The etch proce ss was terminated by quickly removing the etch solution and replacing it with wa ter for 2 minutes. Fresh water was replaced several times for two minute intervals. Finally water was allo wed to sit in the cell for another two hours. Preparation of Tin Sensitizing Solution Tin (II) chloride was used to sensitize the wafe rs so that electroless plating of silver can take place on the surfaces. Tin (II) chloride crys tals (0.5 g) were placed in 100 mL water and stirred to give a cloudy appearance. 2 mL of 10 % hydrochloric acid was added using a pipette causing the mixture to become a clear ti n (II) chloride sensitizing solution. Preparation of Silver Plating Solution Filling the pores in the wafer with silver wire was done to make a more controllably etched material. A two part electroless silver plating solution was made, where solution A contains the silver ions and solution B contains a reducing agent. Solution A was made by dissolving 45.4 g of silver nitrate to 450 mL of water and th en adding ammonium hydroxide drop wise using a pipette until the solution goes from clear to da rk brown and returns to clear. Solution B was made by dissolving 159 g of potassium sodium tartrate tetrahydrate and 11.4 g magnesium sulfate to 364 mL of water. Both so lutions were stored away from light.

PAGE 59

59 Filling the Pores with Silver Wires The purpose of silver plating the nanopores in the mica membrane was to control the rate at which etch solution traverses the membrane relative to the lateral or bulk etch rate. First, the intended stop side of the membrane was exposed to tin by filling one half cell with the solution. The other half cell was left empty so that one face of the membrane was exposed to air. Tin solution remains in the cell for 45 minutes giving it time to properly wet and sensitize the inner walls of the pores. The membrane was then rinsed with water several times then was left to sit in water for at least one hour. Aft er removal of water from the both half cells, the membrane was now ready for silver electroless plating. A dilute solution A (0.5 mL Solution A in 45 mL water) was cooled to 4 C and then 0.5 mL of solution B was added. This mixture was then placed in the half cell that was not exposed to tin solution (sid e to be etched). The c onductivity ce ll containing the membrane was placed in a refrigerator, which was set at a temperature of 4 C, for one hour. Following electroless plating of silver, the remaining solution in the cell was removed and the cell was thoroughly rinsed with water. Figure 33 shows the schematic outline for this etch-filletch method. Etching Silver Filled Mica to form Tapered Pores The silver wire-containing mica membrane was exposed to HF/HNO3 etch solutions on one face. This etch solution of HF (variable concentration) and HNO3 (10%) etched mica and silver wire respectively at two different rates. After 3 hours th e etch solution was removed and rinsed with water several times. The me mbrane was then left to sit in 10% HNO3 for at least 3 hours to get rid of any residual s ilver. Finally, the membrane was rinsed in water several times followed by soaking for at least 3 hours.

PAGE 60

60 Making Replicas of the Tapered Pores Tapered carbon tubes were obtained using a chemical vapor de position (CVD) method described in detail previously. 38, 110, 111 A piece of a porous mica membrane (preparation methods were described in the previous secti on) was placed vertically into a quartz tube (diameter: 4.5 cm, length: 48 cm). This tube wa s then inserted into a high-temperature tube furnace (Thermolyne 21100) and the furnace was heated to 670oC under Ar flow. Once the temperature stabilized, the Ar gas was replaced with an ethylene gas (20 sccm), which thermally decomposed into carbon on the inner-wall and bot h faces of the mica template. After a desired deposition time, the heating was terminated, th e ethylene gas was replaced by Ar flow, and the furnace was cooled down to room temperature. Unlike our previous CVD procedure with alumina templates, 111 heat pretreatment of the template in this experiment was skipped since the mica membranes we used can withstand temperatures above 900oC without any physical deformation. The yielded carbon thickness can be controlled by vary ing the duration of deposition. Preparation of the Carbon T ube Replicas for SEM Imaging First, CVD carbon/mica membrane was put into 48~51% HF solution for 16 hr to dissolve away the mica template. Next, HF was then removed by pipette, leaving the liberated carbon nanoboxes (connected together by the carbon surface f ilm) which were rinsed with methanol and suspended in methanol. The next preparati on procedure entailed the removal of the carbon surface layer on one face of the CVD-treated mi ca membrane to expose the carbon tubes. This was accomplished by using an oxygen plasma etch procedure. A 1 cm x 1 cm hole was premade in one piece of aluminum foil. This hole de fines the area of the membrane that is exposed to the oxygen plasma. The entire assembly, with the hole-containing Al foil facing up, was then placed in the center of vacuum chamber of a plasma reactive-i on etching system (Samco, model

PAGE 61

61 RIE-1C). The following etch cond itions were used: power = 100 W, O2 pressure = 300 Pa, O2 flow rate = 30 sccm. After etching away the carbon surface film (determined by measuring the conductivity of the membrane su rface), the membrane was immersed into ~ 48-51 wt % HF solution to dissolve the mica template. This st ep does not result in free carbon tubes because they are held together by carbon surface film that was not exposed to oxygen plasma. Finally, the sample was rinsed with distilled water and air dried overnight. Sample imaging was conducted using JEOL 6335F field emission scanni ng electron microscope (FESEM). Prior to FESEM imaging, all samples were sputtered with Au/Pd using the Desk II Cold Sputter instrument (Denton Vacuum, LLC). The sputter current = 45 mA, Ar pressure = 75 mTorr, sputtering time = 60sec. The resu lting Au/Pd film was ~ 16 nm. Results and Discussion During chemical etching of ion-tracked membrane s, the damaged zone of the latent track is transformed into nanopores. 16 The simplest description of the etching pro cess defines two parameters: the bulk etch rate ( VB) and the track etch rate ( VT). VB depends on the material, etchant composition and temperature. VT depends on additional parame ters, such as sensitivity of the material to trackin g, post-irradiation conditi ons and etching conditions. 14 When the tracked film is exposed to an etchant on one f ace as described above, the results give conical pores. In most ion-tracked materials, without exposure to extreme conditions (like exposure to high illumination) the VB/ VT ratio defines the cone angle that is formed. 14 However, in mica VT is 3000 times faster than VB, 108 thus producing c one angles (0.02 )108 that is almost zero, giving essentially pores with almost identical cross-sect ions along the membrane thickness. To improve the cone angle in polymeric materials, the VB/ VT ratio is increased by a number of methods including increasing the etchant concentration, 108 applying a high transmembrane potential, 92

PAGE 62

62 and modifying the etch solution composition. 18 However, none of the above methods can work with mica since VT is orders of magnitude greater than VB. 108 To get around this problem, the etched ion tracks in mica was re placed with silver metal that can be independently etched with nitric acid to a wide variety of etch rates. Figu re 3-4a is an SEM image of a membrane that was exposed on one face to 20% HF a nd10% nitric acid solution at 25 C for 3 hrs. Here, HF and HNO3 solutions etch the bulk mica membrane and th e metal track respectively. The base side clearly shows the tapered cone shape that resulte d from this etch. The decreasing pores size is made quite evident from the progressive mica la yers going down into the cone. The opposite face of the membrane (Figure 3-4b) shows the tip to be on the order of a magnitude smaller than the base. The Martin group has been using conical pores for resistive pulse sens ing of molecules. One important feature that make conical pores idea l for sensing is, that most of the resistance is focus in a short distant of the tip. As shown by Lee et al,79 the electric field in conical pores is focused at the tip. The greater the half cone opening angle, the smaller the focus and hence a sensing zone for molecular translocation of the pore for resistive pulse sensing. Half cone opening angle depends on the VL/ VT ratio, and we changed the concentration ratio of the etchants to achieve this. Because the concentration ratio of HF (the lateral etchant) to HNO3 (the track etchant) is greater than used previously, the lateral to track etch rate increases VL/ VT resulting in greater half cone opening angle (~ 21 ). Figure 3-5 is an SEM image of a tapered mica pore that was etched with a higher percen tage of HF solution than previously in Figure 3-4. We demonstrate here that the cone angle can be c ontrolled at will, because we can independently control the solution etch rate traversing the memb rane, and also the lateral etch rate, to give asymmetric pores. This implies that the eff ective pore length (the part of the pore where

PAGE 63

63 resistance is focused), and hence the focus of el ectric field can be varied in these tapered mica pores. This is an important feature for resistivepulse sensing that is like ly responsible for pulse duration. To indirectly capture the entire geometry of the tapered mica pores in the membrane, a replica was done using CVD met hod. Figure 3-6 clearly shows SEM images of carbon tapered nanotubes replicas of the mica pores. The angl es are well defined and surface of the tube appears rather smooth at th e magnification shown. The half cone opening angle was calculated to be ~ 6 This is a relatively large cone angle when compared to those of polymers that are etched without any promoters (Lane) or applied high potential (chad). The half cone opening angle can be calculated as ) 2 ) (( arctan L d dt b (3-1) where L is the length of the pore, db and dt are the large and sma ll openings of the pore, respectively. For db >> dt the equation simplifies to: ) 2 ( arctan L db (3-2) Figure 3-7 is a low magnificati on image of the carbon replicas of mica tapered pores indicating that we can indeed reproduce the tapered geometry mica pore uniformly. Conclusion This study described a method to make asymmetric pores in tracked muscovite mica films using an etch-fill-etch approach. Tracks in the films were initially etched away with hydrofluoric acid to form nanoporous membranes. We demonstrated th at by controlling the concentration ratio of hydrofluoric acid to nitric acid during etch ing, tapered pores with diamond shaped cross-section can be obtai ned. Additionally, we have s hown that the cone angle of the pores can be controlled be changing the concentra tion ratio of the bulk and metal etch solutions.

PAGE 64

64 Replicas of the asymmetric pores were accomplished by carbon vapor deposition, and scanning electron microscopy was used to give evidence of the resulting nanotubes. These conical mica pores make prove more stable for resistive pulse sens ing. Because it is so easy to tailor the cone angle, this might make these mica pores more su itable sensing devises. One potential capability that might be realized in resis tive pulse sensing, is the tuning of the cone angle to control pulse duration. Etch-stop solution Etchant Etch-stop solution Etchant Damaged ion track Etch-stop solution Etchant Etch-stop solution Etchant Etch-stop solution Etchant Etch-stop solution Etchant Damaged ion track Figure 3-1. After irradiation, th e materials are subject to chemi cal etching which preferentially removes the damaged ion track

PAGE 65

65 Figure 3-2. Schematic of cell used to do th e etching and to make all electrochemical measurements. Poly(chlorotrifluoroethylene) cell Pressure plate 3.5 cm 1.0 cm 3.5 cm 1.0 cm Electrodes Membrane 3.5 cm 1.0 cm b Aluminum frame Clamping screw Solution chamber

PAGE 66

66 HF Electroless plating of Silver Mica membrane Tracks Silver Pore HF/ HNO3 Figure 3-3. Schematic diag ram of etch-fill-etch method

PAGE 67

67 AB AB Figure 3-4. Scanning electron micrographs of mica membrane that was exposed on one face to 20% HF and10% nitric acid solution at 25 C for 3 hrs. (A) Side exposed to the etchant (the base). (B) Side exposed to water (the tip). A B Figure 3-5. Scanning electron micrographs of mi ca membrane that was exposed on one face to 40% HF and10% nitric acid solution at 25 C for 3 hrs. (A) Side exposed to the etchant (the base). (B) Side exposed to water (the tip).

PAGE 68

68 Figure 3-6. Scanning electr on micrograph of carbon tapered nanotube replica of the mica tapered pore. Figure 3-7. Low magnification SEM images of carbon tapered nanotube replicas of the mica tapered pore.

PAGE 69

69 CHAPTER 4 ELECTROLESS AU PLATI NG OF TRACK-ETCHED KAPTON POLYIMIDE NANOPOROUS MEMBRANES Introduction Track-etched polymer membranes have found ma ny applications in industry and research as filtration and separation materials. 6 Chemical modification of the inner walls of these membranes has made them more selective and sophisticated se paration structures. 27, 28, 31 These chemical modifications have been made possible by first plating th e inner walls of the pores of these membranes with gold, 23 then attaching a desired thiol terminated functional group. 31 For reproducible results of any analytical measurem ents done using the pores of these nanoporous membranes, it is crucial that the pores remain stable and have a well define internal diameter, particularly where the pore approaches the size of the analyte molecule. One of the problems faced when using some polymers such as poly carbonate (PC) and poly (e thylene terephthalate) (PET) membranes, is that very small pores tend to temporarily bloc k during ion transport mesurement. 112 Another disadvantage discovered wh en using these membranes to measure ion current, is that possi ble dangling alkyl groups 113 render the pores from been well define and thus gives inconsistent results and lots of noise in the current measurement. 112 The polyimide Kapton on the other hand does not exhibit these disadvantages. 112 Due to its abilities of maintaining excellent physical, elect rical and mechanical properties at both low and high temperature extremes, Kapton is a very attractive polymer for use as a particle track-etch membrane for application in separation and filtration in industry and research. 16, 114, 115 It will be advantageous therefore, if lik e polycarbonate and PET, Kapton can similarly be electrolessly plated with gold to tailor pore diameter and, furt her, be modified with desired thiol terminated functional gr oups for selective separation.

PAGE 70

70 The objective of this study was to determine if the polyimide Kapton, a very chemically resistant and stable polymer, can be electrole ssly gold plated with similar or better quality compared to PC porous membranes. Additionally, investigate whether pla ting can be controlledthat is, tailor pore diameter of the membrane with pl ating time. It is interesting to note that to date; there is no known report of elect roless plating of Kapton membranes. Experimental Materials Kapton 50 HN foils (12.5 m thick, 107 tracks per cm2) were obtained from the linear accelerator laboratory UNI LAC at the GSI (Darmstadt, Germany). Boric Acid was obtained from Fisher and used as receiv ed. NaOCl (13% active Cl), KI, SnCl2, AgNO3, NaHCO3, were obtained from Aldrich and used as received. Trifluoroacetic acid, Na2SO3, NH4OH, formaldehyde and methanol were obtained from Mall inckrodt and used as received. Commercial gold-plating solution (Oromerse SO Part B) was obtained from T echnic Inc. Milli Q water was used to prepare all solutions and to rinse the membranes. Chemical Etching Cylindrical pores were etched in the tracks of the Kapton foils using sodium hypochlorite solution as describes in detail elsewhere. 116 The shape and size of th e pores can be tailored by the choice of etchant an d the etching conditions. 116 To obtain cylindrical pores the etching rate along the track, the-so called track etch rate Vt has to be much faster than the non-specific etching of the polymer ca lled the bulk etched rate Vb 14, 81 The relation between Vt and Vb is explained in figure 4-1. 14, 81 Studies have shown that an e fficient etching can be performed in sodium hypochlorite containing 13% active chlorine content. 116 Furthermore, it has been demonstrated that the shape of the pore can be regulated with an a ppropriate choice of pH. 116 When sodium hypochlorite is not buffered its pH ~ 12.6. At these conditions, and elevated

PAGE 71

71 temperature of 50C, Vb of Kapton is high (~ 0.21 ( m / h) and the pores become strongly conical. Buffering the etchant with boric acid to pH ~9 enables one to obtain cylindrical pores. It is important to note that th e etching works only at basic pH when the hydrolysis of imide bonds by OHis possible. The membrane was immersed in a Teflon container containing 400 ml of NaOCl solution at pH 9.8 and temperature of 50C. First, the container with the etchant solution is brought to 50C by plac ing it in a water bath controled by a water heater. The size of the pores increases with etching time. After etching, the membrane was rinsed with D. I. water and left to soak for two hours. The memb rane was allowed to dry in air overnight. Electroless Plating of Kapton In order to better control th e size and surface chemistry of Kapton, the membranes can be plated electrolessly with gold. 10 Tailoring of the pore size in polymer membranes by the time of performing the electroless plating with gold has been demonstrated previously. 28, 51 Since Kapton membranes possess carboxylate groups ma de available via imide hydrolysis by the etchant, it is expected that these would act as active sites for the bonding of tin ( ). Tin can then reduce silver ion which later acts as a nucleation site for the re duction of gold. The procedure followed the recipe for the electroless plating of polycarbonate. 28, 51 Pore Diameter Measurement Cylindrical pores were ch aracterized by taking scanni ng electron microscopy (SEM) images of the membrane surfaces. Also, we have chosen to use an electrochemical technique based on measuring ion current to measure pore diameter. This entailed mounting the membrane sample in the cell (Figure 4-2), filling both half cells with an electrolyt e solution of known ionic conductivity, and obtaining a curren t-voltage (I-V) curv e associated with ion-transport through the nanopore. The experimental slope of this li near I-V curve is the i onic conductance, G, (in Siemens, S) of the nanopore, which is given by

PAGE 72

72 G = ( N A KCl d2) / 4 L (4-1) where N is the pore density, A is the membrane area, KCl is the experimentally measured conductivity of the KCl-base d electrolyte used (S cm-1), L is the length of the nanopore (membrane thickness), d is the pore diameter. Because all of the other parameters in Equation 41 are known, d can be calculated. Results and Discussion SEM and Ion Current Measurements Figure 4-3 shows SEM images of Kapton porous membranes before and after electroless gold plating. Here we see that th e pore diameter decreases with plating time. Figure 4-4 shows a plot of pore diameter as a func tion of Au plating time from 0 to 12 hours. The two methods of measuring pore diameter gave very similar result s. This is important because only resistance measurements can be used for small pore diameter that cannot be resolved by SEM. This data indicate that one can tailor th e pore diameter with plating time down to the nanometer scale. This capability is important for different transpor t studies and in sensor research where the size of the pore with respect to the analyte is important. 10, 28, 77, 96 Atomic Force Microscope Images Figure 4-5 shows the atomic force microscope images of the Kapton membranes before and after electroless plating. The membrane s remain relatively sm ooth after gold plating compared to a similarly porous structured poly carbonate membrane after plating with gold. Conclusion Electroless gold plating propert ies on the surface and pore walls of track-etched Kapton polyimide nanoporous membranes we re studied. SEM, AFM, and ion current measurements were used to characterize the surfaces and pore dimensions of the membrane. Nanoporous Kapton polyimide membranes were electroless gold plated over different times and the pore

PAGE 73

73 diameter characterized using SEM. Ion curr ent measurements were used to measure the diameter of very small pores. AFM images show that after electroless gold plating, the gold surface layers are smooth compared to a sim ilarly structured polycarbonate membrane. Electroless plating the membranes for 12 hours pr oduced Au wires in the pores. Etching away the membrane with sodium hypochlorite to expose gol d tube replicas revealed that the plating in the walls is also smooth. Figure 4-1. Definiti on of bulk etch rate Vb and track etch rate Vt.[Adapted from Apel, P. Radiation Measurements 2001, 34 559-566.]

PAGE 74

74 Figure 4-2. Schematic of cell used for electrochemical measurements.

PAGE 75

75 Before AU1.5 hrs plating3 hrs plating 4.5 hrs plating7.5 hrs plating Figure 4-3. Pore diameters fo r different Au plating times.

PAGE 76

76 0 20 40 60 80 100 120 140 160 180 200 220 012345678910111213Time (hrs)Pore diameter (nm) Figure 4-4. Pore diameter as a function of plating time with measurements taken from SEM image (0-7.5 hrs) and ion current resistance measurements (8-12 hrs).

PAGE 77

77 PC membraneBefore Au Before Au Roughness ~ 13.6 Roughness ~ 5.1 Roughness ~ 6.3nm Roughness ~ 3.6nmKapton membrane After 2.5 hrs Au plating After 2.5 hrs Au plating Figure 4-5. Atomic Force micrographs of Kapton and PC membranes before and after electroless Au plating

PAGE 78

78 CHAPTER 5 ASYMMETRY IN DIFFUSIONAL TRANSPO RT OF MOLECULES THROUGH KAPTON CONICAL NANOPORES Introduction Membranes and porous materials have found various applications in filtration and separation processes. 8-10 Modern biotechnology has posed new challenges in th e application of such membranes, and requires pores with diamet ers similar to those of molecules under study 11 (e.g., as small as several nanometers). The nanom eter scale of such pores is necessary in both achieving optimal control of the flow of biomolecu les, as well as in developing sensors for their detection. The transport properties of such na nometer scale pores are not well understood yet. The hint that nanopores behave differently from micropores, comes from Mother Nature. 12 Biological channels and pores have diameter of ~ 1 nm and are cr itical for functioning of living organisms. Ion channels and pores exhibit tran sport properties not obser ved with larger pores, for example (i) selectivity fo r ions or molecules, (ii) rectification of ion current117 (iii) ion current for constant voltages applied across the membrane118 (iv) facilitated transport of molecules (v) transport of ions and molecules against th eir electro-chemical potential gradient. 119 It has recently been demonstrated that asym metric conical nanopores in polymer films can exhibit transport and r ectification properties similar to biochannels. For example, conical nanopores in PET and Kapton membranes are cati on selective and rectif y ion current with a preferential direction of cation flow from the tip to the base of the cone. 81, 88, 105, 120, 121 However, unlike the above cases, in this inve stigation there is no applied transmembrane potential. Also, there are no electrostatic or bi nding interactions of the molecule with the pore surface. Since the molecule under study is neutra l, transport through the membrane is by purely diffusional and geometrical constraints. Yet we found in our studies that the diffusion of molecules across these membranes, exhibit a rectif ication behaviour. It means that there was a

PAGE 79

79 preferential direction of diffusion flow. This effect cannot be predicted by considering classical diffusion with constant diffusion coefficients. The rectificatio n of conical nanopores is of tremendous significance for industr ial filtration processes in wh ich asymmetric membranes are often applied. One type of asymmetric membranes used in industry consists of a thin skin of nanoporous material placed on a low resistance support, which assures high fluxes and good size separation. 122 Finding an optimal direction of concentr ation gradient will improve the filtration process. Asymmetric diffusion has been observed before with multi-membrane systems and membranes with two skins on two memb rane faces of a membrane support. 123 Diffusional Transport Described by Ficks Laws Diffusional transport is desc ribed by Ficks two diffusion laws. For diffusion in one dimension, Ficks first law of diffusion describes the flux (the net number of moles of particles crossing per unit time, t, through a unit area perpendicular to the x-axis and located at x ), stating that the flux of a molecule with diffusion coefficient, D is directly proportional to the concentration gradient. x t x c D t x J , (5-1) D normally is assumed constant, however there are known examples when D depends on position x concentration c or even time t 124, 125 The diffusion coefficient determines the time it takes a solute to diffuse a given distance in a medium. The diffusion coefficient depends on the physical characteristics of the solute as well as those of the medium. An approximate bulk diffusion coefficient (like that of molecules of similar masses) was used here, because, we are more concerned about emphasizing th e preferential flux from one side of the membrane than the absolute flux values, themselves. The second of Ficks law expresses how the concentration of

PAGE 80

80 the species depends on time. Here again, the gene ral form of the law in one dimension, taking into account the possible forms of D is as follows: x c D x t c (5-2) Hindered Diffusion in Cylindrical Pores The effective diffusion coefficient of a solute with in a pore of comparable size is lower than that of the bulk solution value. This phenomenon is ca lled hindered diffusion and results from steric exclusion of the solute at pore opening and hydrodynamic or wall drag resistance due to the presence of the pore wall. For steady-state di ffusion through a membrane with cylindrical pores that are comparable to the size of an uncharged solute, the effective diffusion coefficient can be expressed in terms of the solute-to-pore size ratio ) ( r a by the Renking equation: 126, 127 ) 948 0 089 2 1044 2 1 ( ) 1 ( /5 3 2 D D (5-3) Here, restriction to diffusion due to steric hindrance at the entr ance to the pores is given by the partition coefficient, =2) 1 ( (5-4) As established by Ferry,128 a molecule must pass through the opening without striking the edge. Therefore the center of the solu te particle cannot be locate d at a radius that exceeds ( r a ) (Figure 5-1). The partition coefficient ther efore, is equivalent to the frac tion of the cross-sectional area of the pore that is accessible to the center of the molecule. The second factor in the Renking equation, called the inverse enhanced drag,127 ) 948 0 089 2 1044 2 1 (5 3 1 (5-5) corrects for hydrodynamic or wall drag resistance, which is th e friction between a molecule moving within a pore and its walls (Figure 5-2).

PAGE 81

81 The total flux, J through a membrane with cylindrical pores of length, L and pore density, can then be given as: L C D r J 1 2 (5-6) where Dis the free diffusion coefficient of the solute in the bulk solution, C is the concentration difference across the membrane, and N r2 is the membrane porosity, Hindered Diffusion in Conical Pores Equation 5-6 is for a membrane with cylindrical pores where r is constant. The pores used for this study are conical, and therefore r increases or decreases from one end of the membrane to the other, thus varying linearly with position. The porosity of conical pores is different from that of cylindrical pores, and is obtained by th e product of the pore density and the geometric mean cross-sectional pore area: 4t bd d (5-7) where bd and td are the diameters of the large and small opening of the pores respectively. There are varying hindrance effects (steric and hydrodynamic) on diffusion over the length of the membrane. The average inverse enhanced drag, K-1, is ~ 1 for conical pores because most of the wall drag resistance occurs in the region of the tip, which is a very small faction of the pore. For cylindrical pores, the partition coefficient determin es the rate of entry of the molecules into the pores, and is equal to the ratio of the solute concentration in the pore to that of the bulk solution at equilibrium. However, unlike cylindrical por es, the partition coefficient that determines molecular entry into conical pores at steady stat e diffusion is not the same as the equilibrium partition coefficient. The partition coefficient, ( r ( x )) varies with r ; which is a function of position, x and therefore changes along the length of the pores. Hence it is expected that the

PAGE 82

82 equilibrium concentration along a pore length is different. However, the partition coefficient and also the enhanced inverse drag within the pore, av erages to be of neglig ible contribution to the flux since only a very small fraction of the pore approaches the size of the solute molecule. The value of these coefficients therefore approximate s to 1. The asymmetry in diffusion, then, may be caused by the difference in steric hindrance at the entrance of either side of the pore. This will determine the amount of molecules that ente r and leave the pore per unit time. The above arguments suggest that molecules entering from the large opening of a conical pore will have greater flux than those that enter from the sma ller opening, provided that they can leave at the same rate. The limit to diffusion will be the pa rtition coefficient at the tip. With these assumptions, the fluxes can be written for dilute solute concentrations as L C D Jb b 1 (5-8) L C D Jt t 1 (5-9) where the subscripts b and t represents the directions of th e net flux entering the large and small opening of the pores respectively, and K-1 is equal to 1. We can also solve the diffusion problem with a constant D through a conical nanopore with opening angle openings dt and db, with solute concentraton c0 at the tip, and zero concentration of solute at the base: 1 ) (0x L d d c x cb t, L x dt cot (5-10) valid for dt<
PAGE 83

83 x L d d c x cb t1 ) (0, L x dt cot (5-11) When we calculate the fluxes (moles/s) in the tw o above mentioned conditions they are equal to each other: L d d c D jb t0 (5-12) These calculations indicated that classical diffusion with constant D cannot describe our experiments, unless one takes into account th e difference in boundary conditions due to the difference in the partition coefficient. It was the purpose of this study to investig ate the diffusion of neut ral molecules from both ends of conical pores of the polyimide Kapton. Here the dimension of the tip of the conical pores approaches that of the molecules. To th is end the diffusion of th e water soluble neutral molecule phthalazine was studied. Experimental Materials Kapton 50 HN foils (12.5 m thick, 107 tracks cm-2) were obtained from the linear accelerator laboratory UNILAC at the GSI (D armstadt, Germany). Sodium hypochlorite (NaOCl, 13% active Cl) and potassium iodide (KI), were obtained from Aldrich and used as received. Phthalazine, sodium chloride, sodium phosphate dibasic, sodium phosphate monobasic, sodium azide was obtaine d from Fisher and used as received. Milli Q water was used to prepare all solutions and to rinse the membranes.

PAGE 84

84 Kapton Polyimide Membrane We used polyimide foils to prepare nanoporous membranes. Polyimide possesses a unique combination of properties that are ideal for a vari ety of applications in many different fields. The film maintains excellent physical, electrical, and mechanical prope rties over a wide temperature range. Polyimide also has very good chemical resistance and does not dissolve in organic solvents. We used 12.5 m thick commercially available Kapton 50 HN, produced by DuPont. Irradiation-Track Formation For the preparation of membranes we used th e-so called track etchi ng technique. It is based on irradiating a dielectr ic film with swift heavy i ons and subsequent chemical development (etching) of the damaged ion tracks. A unique feature of heavy ion irradiation is single-particle recording. That is to say, one swift heavy ion wh ich penetrates the foil produces one damaged track. Therefore, counting the number of ions used for irradiation enables one to prepare membranes with tailored number of pores from the range 1 up to 1010 ions/cm2. Kapton foils, which we used for these experiments, were irradiated with uranium ions of energy of 11.4 MeV/u, at the heavy ion accelerat or UNILAC at the Institute fo r Heavy Ions Research, (GSI) Darmstadt, Germany. We used foils irradiated with the fluencies 107 and 108 ions/cm2. The range of the ions is in all cases larger than th e thickness of the polyimide membranes. The ions penetrate the membranes at normal incidence, cr eating damaged tracks, which is then followed by chemical etching to form pores (Figure 5-3). Chemical Etching of Membrane Tracks After irradiation of the Kapton foil with heavy ions, the latent tracks have to be chemically etched. The shape and size of the pores can be tailored by the choice of etchant and the etching conditions. To obtain cylindrical pores the etching rate along the track, the-so called track etch

PAGE 85

85 rate Vt has to be much faster than the non-specific etching of the polymer called the bulk etched rate Vb The relation between Vt and Vb is explained in Figure 5-4. To obtain conical pores, one needs to choose etching c onditions which assure high vb. Preparation of conical pores is normally performed in a cond uctivity cell with etchant placed only on one side of the membrane. The other side of the membrane is in contact with a stopping medium, which neutralizes the etchant as soon as the pore is etched throug h. For example, if NaOH is used as an etchant, we use an aci dic stopping medium. The chemical stopping is further supported by an electric stopping. The etching is performed under voltage with electrodes arranged in such a fashi on that the anode is on the side of etchant, which retracts ions active in the etching process (e.g. OH-) out from the pore ( Figure5-5). Etching with an applied voltage and measuring electric curren t affords a method to monitor the process. At the beginning of etching the cu rrent is zero, because the two chambers of the conductivity cell are no t connected with one another. When the pore is etched through, the current value is finite and increases in time, i ndicating an increase of the pore diameter (Figure 5-6). Kapton is very resistant chemically, therefore, development of latent tracks has to be performed by a very aggressive etchant and at elevated temperatures. Previous studies have shown that an efficient etching can be perfor med in sodium hypochlorite containing 13% active chlorine content. Furthermore, it has been de monstrated that the shape of the pore can be regulated with an appropriate c hoice of pH. When sodium hypochl orite is not buffered its pH ~ 12.6. At these conditions, and elevated temperature of 50C, Vb of Kapton is high (~ 0.42 ( m / h) and the pores become strongly conical. Bufferi ng the etchant with boric acid to pH ~9 enables one to obtain cylindrical pores. It is important to note that the etching works only at basic pH when the hydrolysis of imide bonds by OHis possible.

PAGE 86

86 To obtain conical pores in Kapton, the ir radiated samples were placed between two chambers of a conductivity cell and etched from one side in sodium hypochlo rite. The other half of the cell is filled with 1 M potassium iodide (KI) solution as a stopping medium for the OClions of the etchant. As soon as the etchant co mpletely penetrates the membrane, iodide ions reduce OClto Clions: 112, 130 OCl+ 2H+ + 2II2 + Cl+ H2O 5-13 Via this reaction, the etching process stops immediately afte r the breakthrough, allowing the preparation of extremely narrow pores. Pore Diameter Measurement SEM was used to characterize the opening diam eters of conical pores, especially the big opening, which we call the base (Figure 5-7). The small opening of conical pores, called the tip is below resolution of SEM, therefore, we have to use another technique for its size estimation. We have chosen to use an elec trochemical technique ba sed on measuring ion current. The ionic conductance of a conical pore is related to its diameter by the following equation: 5-14 where n is the number of pores, is the conductivity of electrolyte, L is the length of the pore (or the equivalent membrane thickness), and db and dt are the diameters at the base and tip of the cone respectively. Transport Measurement Measurements of neutral molecular trans port through conical nanopor es were performed using UV-Vis spectrometry (Agi lent 8453). A U-cell set-up cont ained the permeate molecule solution on one side and a buffer, in our case PBS (in which the permeate was prepared) on the other side of the cell (Figure 5-8). The me mbrane was sandwiched between two transparent L d d n Gt b4

PAGE 87

87 tapes that have holes in the center that defined the transport area. This tape-membrane composite was clamped between the U-cell. Th e permeate solution was removed at certain time intervals and the permeate concentration was meas ured be UV-Vis spectrometry. Feed solution concentrations were varied in 4 ml PBS (pH 7.2) and the permeate side was only 4 ml PBS. The solution on both sides of the U-tube was stirre d using stir bars and stir-plate set-up. Viscosity Measurements The viscosities of phthalazine and dext rose were measured using Cannon-Fenske viscometer No. 75 (model P200; Cannon Instruments) Results and Discussion Membrane Characterization Figure 5-9 is an SEM image of the base side of a typical porous Kapton membrane used for the transport study. The average pore density taken from approximately five hundred pores in five different locations of th e image under low magnification, was 107/cm2. Figure 5-10 is an SEM image of the base s howing a diameter of 1.68 m. The size of the tip, which is below the resolution of SEM, was calculated from ionic co nductance measurements (Figure 5-11) using equation 5-14. The average tip size used in this study was ~ 2 nm. The average diameter of the molecule for transport was ~ 0.7 nm. Transport Measurements of Phthalazine From equation 5-4, the partiti on coefficients at the base a nd tip approximate to 1 and 0.42 respectively. These values put th e partition coefficient at the base about 2 times that at the tip. From the proposed equations, Figures 5-8 and 5-9 fo r base and tip fluxes, it is expected that at low concentrations the flux from base to tip should be about twice that in the opposite direction. Indeed, Figures 5-12 and 5-13 show that base fluxes are about twice that of tip fluxes for 1 and 3 mM solute concentrations. However, base fluxes approach those of tip fluxes as the

PAGE 88

88 concentration of the solute increases, and bot h base and tip fluxes are similar at higher concentrations (Figures 5-14 a nd 5-15). Figure 5-16 shows this asymmetric behavior over a wider rage of concentrations. The higher flux from base to tip direction can be explained by a lower ac cess resistance of the solute to the pore. Figure 5-17 shows a comp arison between the theoretical flux obtain from equation 5-9, and the experimental flux from tip to base. We see that these are in good agreement over the wide range of concentration used The experimental flux from base to tip, on the other hand, deviates from theoretical values obtained from equation 5-8, particularly at higher concentrations. To determ ine whether the base flux or tip flux behaves in a classical way, a plot showing increase of flux through base and tip, respectively, w ith respect to the flux measured at 0.1 mM was generated (Figure 5-18 ). The straight line: Flux(c)/Flux(0.1mM) = c/0.1 obtained for the tip flux indicates that tran sport through tip behaves according to the Fick's law with a constant D. Transport through base is hindered for higher co ncentrations indicating that diffusion coefficient is concentration dependent. The flux (mol/s) through an aperture of diameter dt with boundary conditions c0 is given by: .0 limDc d jt ited This value is smaller than the flux as given by eq. 5-12 The flux through the tip is limited by the value of.limitedj, while the flux through the base is not. The value of flux through a conical nanopore as gi ven by eq. 5-12 can be easily obt ained when the solute passes from base to tip. The open question however remains, why the ratio of fluxes is concentration dependent. In order to answer this question, we examined behavior of the flux from base and tip side, respectively for different con centrations of the solute.

PAGE 89

89 The Influence of Cosolute Concentration on Asymmetry We investigated whether molecular crowding / jamming had any influence in the behavior of the base flux at higher concentrations. To do th is, we used a constant concentration (5 mM) of phthalazine plus varying concentra tions (0 to 100 mM) of the coso lute dextrose on the feed side of the cell. We then placed identical total c oncentrations on the permeat e side of the transport cell using dextrose (Figure 5-19). This procedure was done to ensu re that there was no osmotic influence on transport. Figure 5-20 is a calib ration curve for phthalazine, and phthalazine with dextrose cosolute, generated from UV-visible absorption measurements. The two plots were placed together to show that the addition of dextrose as a cosolute to phthalazine does not affect the absorbance to any significant degree. Also to address the issue of viscosity, kinematic viscosity measurements were done on phthalazi ne and dextrose over the concentration range used in this study. Figure 5-21 in dicates that there is negligible change in viscosity over the range of concentrations (0 to 100 mM) used for this study. Figure 5-22 shows the influence of varying concentrations of the co solute dextrose on the flux of pht halazine, which is kept at a constant concentration of 5 mM. The tip flux is not affected over the entire range of concentrations. However, the base flux is sharply decreased to that of the tip flux when dextrose is added to the feed solution. Even though phtha lazine concentration wa s kept constant, base flux approached the tip flux in a very simila r way when phthalazine concentrations were increased. The increase in total concentration on the base side of the membrane causes base flux to approach the limit of tip flux. Proper mathem atical modeling of this effect is needed, taking into account dependence of diffusion coefficient on concentration of the solute and position in the channel. We would like to me ntion that theoretical studies were reported in which two sizes gas molecules were placed on the base side of co nical pores, and one size molecule was too large to pass through the tip. In this situation a total jamming was found, causing the diffusion of the

PAGE 90

90 small molecules to be zero over time. As exp ected, the rate of such jamming increased with concentration. On the other hand, when the mo lecules were placed on the tip side, no jamming occurred from that entrance. 131 We think that in our case, an increase in total concentration of the solutes might lead to similar partial molecular jamming of diffusion from base to tip. Decrease of the base flux provides ev idence that it is the exit rate of the solute from the pore that limits the diffusion transport in the direction from base to tip. Conclusions Diffusion rates through a membrane can be asymmetric due to molecular binding, electrostatic interaction, and difference in osmotic potential. In this study we have demonstrated that asymmetric diffusion can also occur by purel y geometric constrains of conical pores on the diffusing particles, where the tip of the cone is comparable to the dimension of the molecules. We show that asymmetric behavi or of diffusion is concentration dependent and there appear to be some sort of partial jamming effect related to the increase in concentration of molecules from the large opening of the pores. To further shed more light on this interesting phenomenon, proper mathematical modeling of this effect is needed, taking into account dependence of diffusion coefficient on concentration of the solute and position in the channel. Figure 5-1. The partitioning of a spherical molecule of radius, a, in cylindrical pores of radius, r.[Adapted from Davidson, M. G.; Deen, W. M. Macromolecules 1988, 21, 34743481.]

PAGE 91

91 = a/r Figure 5-2. Spherical molecule of radius, a, moving within a cylindrical pore of radius, r.[Adapted from Davidson, M. G.; Deen, W. M. Macromolecules 1988, 21, 34743481.] A B A B Figure 5-3. Swift heavy ions impinge on a dielec tric solid leading to damaged ion tracks witch can be chemically etched to form pores(A and B, respectively)

PAGE 92

92 Figure 5-4. Definiti on of bulk etch rate Vb and track etch rate Vt.[Adapted from Apel, P. Radiation Measurements 2001, 34, 559-566.] Figure 5-5. Conductivity cell used to prepare conical pores.

PAGE 93

93 Figure 5-6. Etching curve s howing moment of breakthrough with sharp increase in ion current.[Adapted from Siwy, Z.; Apel, P.; Dobrev, D.; Neumann, R.; Spohr, R.; Trautmann, C.; Voss, K. Nuclear Instruments & Met hods in Physics Research, Section B: Beam Interacti ons with Materials and Atoms 2003, 208, 143-148.]

PAGE 94

94 Figure 5-7. Kapton membran showing la rge opening (base) of conical pores Figure 5-8. Experimental setup for transport measurements

PAGE 95

95 Figure 5-9. Scanning electron micrograph of base side showing pore density (107 pores/cm2) of Kapton membrane. Figure 5-10. Scanning electron micrograph of the base showing diameter of 1.68 m

PAGE 96

96 I = 0.0077E R2 = 0.9918-0.00020 -0.00015 -0.00010 -0.00005 0.00000 0.00005 0.00010 0.00015 0.00020 -0.02-0.015-0.01-0.00500.0050.010.0150.02E (V)I (A) Figure 5-11. A typical current-vol tage curve used to measure th e tip diameter of the conical nanopores. 0 10 20 30 40 50 60 70 80 90 100 01002003004005006007008009001000Time (min)Nanomoles transported/cm2 Base Tip Figure 5-12. Base and tip fluxes fo r 1 mM phthalazine feed solution.

PAGE 97

97 Figure 5-13. Base and tip fluxes fo r 3 mM phthalazine feed solution. 0 20 40 60 80 100 120 050100150200250300350400450Time (min)Nanomoles transported/cm2 Base to tip Tip to base Figure 5-14. Base and tip fluxes fo r 10 mM phthalazine feed solution. 0 5 10 15 20 25 30 35 40 45 50 0100200300400500Time (min)Nanomoles transported/cm2 Base Tip

PAGE 98

98 0 50 100 150 200 250 050100150200250 Time (min)Nanomoles transported/cm2 Tip Base Figure 5-15. Base and tip fluxes fo r 50 mM phthalazine feed solution. 0 0.5 1 1.5 2 2.5 051015202530Concentration (mM)Base Flux/Tip Flux Figure 5-16. Asymmetric behavior of flux versus concentration.

PAGE 99

99 0 0.1 0.2 0.3 0.4 0.5 0.6 051015202530Concentration (mM)Flux (nmoles/cm2 min)Theoretical Experimental Figure 5-17. Theoretical and experi mental tip flux vs concentration 0 50 100 150 200 250 300 350 0100200300 c [mM]/0.1 [mM]Flux (c)/Flux (0.1 mM) from tip from base Figure 5-18. Increase of flux through base a nd tip, respectively, with respect to the flux measured at 0.1 mM.

PAGE 100

100 Stir bar Conical nanopore membrane Phthalazine+ Dextrose+ PBS (pH 7.2) PBS (pH 7.2) + Dextrose U-cell Stir bar Conical nanopore membrane Phthalazine+ Dextrose+ PBS (pH 7.2) PBS (pH 7.2) + Dextrose U-cell Figure 5-19. Experimental set-up for transport measurements of phthalazine with cosolute dextrose y = 0.003021x R2 = 0.999811 y = 0.003058x R2 = 0.998450 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1 05101520253035CONCENTRATION (micrmolar)ABSORBANCE (a.u.) Phthalazine Phthalazine with 490 mM Dextrose y = 0.003021x R2 = 0.998406 0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035 00.20.40.60.811.2 CONCENTRATION (micromolar)ABSORBANCE (a.u. y = 0.003021x R2 = 0.999811 y = 0.003058x R2 = 0.998450 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1 05101520253035CONCENTRATION (micrmolar)ABSORBANCE (a.u.) Phthalazine Phthalazine with 490 mM Dextrose y = 0.003021x R2 = 0.998406 0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035 00.20.40.60.811.2 CONCENTRATION (micromolar)ABSORBANCE (a.u. Figure 5-20. Calibration curve of phthalazine, and phthalazine with the highest concentration of dextrose cosolute used

PAGE 101

101 0 0.2 0.4 0.6 0.8 1 1.2 1.4 050100150200250300350Concentration (mM)Viscosity ( mm2/ sec) Dextrose Phthalazine Figure 5-21. Viscosity of dextrose and phtha lazine over the concentr ation range used 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 020406080100Dextrose Concentration ( mM)Flux (nmoles/(cm2 min) Base Flux Tip Flux Figure 5-22. Influence of dextrose cosolute conc entrations on the base and tip diffusion from 5 mM phthalazine

PAGE 102

102 CHAPTER 6 CONCLUSION The goals of this research were to deve lop conical nanopores and nanotube (single and multipore) in polymers and mica membranes by th e track-etch method and extension of this technology for sensing and other app lications, and to inve stigate the tr ansport properties of these conical pores. Chapter 1 reviewed background info rmation for this dissert ation including the ion track-etch method, membrane based template synt hesis, electroless metal deposition, chemical vapor deposition, resistive-pulse sensing, plasma based etch ing and asymmetric diffusion. In Chapter 2, an extension of the track-etch method to make conical pores in a reproducible fashion was demonstrated. We have shown he re that we can not only reproducibly prepare track-etched based conical nanopor e sensor elements, but that we can predict from the experimental parameters used during the second etch, what the diameter of the all-important nanopore tip will be. For these reasons, we belie ve that the track-etch method will prove to be the technology of choice for taki ng artificial-nanopore resistive-pul se sensors from the bench top to the practical prototype-device stage of the R&D effort. A method to make asymmetric pores in tracked muscovite mica films using an etch-refilletch approach was the topic of Chapter 3. By controlling the concentra tion ratio of hydrofluoric acid to nitric acid, tapered pores with diamond shap ed cross-section were obtained. Here, we see that a two dimensional control of the etch process was developed. Replicas of the asymmetric pores were accomplished by carbon vapor depos ition, and scanning electron microscopy was used to give evidence of the resulting nanotubes. In this study, excellent control over cone angle was demonstrated. This control over cone angle has great potential to make the sensing zone of the conical pores limited to a very short distance from the tip. This is ideal for single molecule detection.

PAGE 103

103 In Chapter 4, electroless gol d plating properties on the su rface and pore walls of tracketched Kapton polyimide nanoporous membranes we re studied. SEM, AFM, and ion current measurements were used to characterize the su rfaces and dimensions of the pores in the membrane. AFM images showed that after electr oless gold plating, the gold surface layers are smooth compared to a similarly structured polycarbonate membra ne. It is important that membranes remain relatively smooth after gold plat ing in order to have a well define pore area for transport and sensing. These properties sugg est that Kapton might be a better candidate than PC for use in transport studi es and molecular sensing. Finally, asymmetric diffusiona l transport of neutral mol ecular species through Kapton conical pores was described in Chapter 5. Membranes containing conical nanopores from polyimide (Kapton HN Dupont) foils were prepared by the track-etching tec hnique. We report the preferential direction of th e flow of molecules through conical nanopores from the base side at low concentration. However, there is con centration dependence from the base side of the pores which causes the flux to appr oach that of the tip as conc entration increases. Study of transport with a cosolute suggested that there may also be some sort of partial jamming effect with concentration from the base entry. It is hoped that further mathematical modeling of this system would shed some light on th is interesting transport phenomenon

PAGE 104

104 LIST OF REFERENCES (1) Sotiropoulou, S.; Vamvakaki, V.; Chaniotakis, N. A. Biosensors & Bioelectronics 2007, 22, 1566. (2) Heo, K.; Yoon, J.; Jin, K. S.; Jin, S.; Ree, M. IEE Proceedings: Nanobiotechnology 2006, 153, 121-128. (3) Hinds, B. J.; Chopra, N.; Rantell, T.; Andrews, R.; Gavalas, V.; Bachas, L. G. Science (Washington, DC, United States) 2004, 303, 62-65. (4) Lev, O.; Tsionsky, M.; Rabinovich, L.; Glezer V.; Sampath, S.; Pankratov, I.; Gun, J. Analytical Chemistry 1995, 67, 22A-30A. (5) Ree, M.; Yoon, J.; Heo, K. Journal of Materials Chemistry 2006, 16, 685-697. (6) Apel, P. Nuclear Instruments & Methods in Ph ysics Research, Section B: Beam Interactions with Materials and Atoms 2003, 208, 11-20. (7) Choi, Y.; Baker, L. A.; H illebrenner, H.; Martin, C. R. Physical Chemistry Chemical Physics 2006, 8, 4976-4988. (8) Van der Bruggen, B.; Everaert, K.; Wilms, D.; Vandecasteele, C. Journal of Membrane Science 2001, 193, 239-248. (9) Gilron, J.; Gara, N.; Kedem, O. Journal of Membrane Science 2001, 185, 223-236. (10) Jirage, K. B.; Hulteen, J. C.; Martin, C. R. Analytical Chemistry 1999, 71, 4913-4918. (11) Desai, T. A.; West, T.; Cohen, M.; Boiarski, T.; Rampersaud, A. Advanced Drug Delivery Reviews 2004, 56, 1661-1673. (12) Cooper, G. M. In The CellA Molecular Approach, 2 ed.; Sinauer Associates, i., Ed., 2000, pp 81-84; 476-491. (13) Fischer, B. E.; Spohr, R. Interdisciplinary Science Reviews 1984, 9, 329-335. (14) Apel, P. Radiation Measurements 2001, 34, 559-566. (15) Virk, H. S.; Kaur, S. A.; Randhawa, G. S. Indian Journal of Environmental Protection 2001, 21, 529-533. (16) Fleischer, R. L.; Price, P. B.; Walker, R. M. Nuclear Tracks in Solids: Principles and Applications, 1975.

PAGE 105

105 (17) Engels, O.; Schempp, A.; Pelzer, W.; Homeyer, H. Proceedings of the Particle Accelerator Conference, 17th, Vancouver, B. C., May 12-16, 1997 1998, 1, 1078-1080. (18) Scopece, P.; Baker, L. A.; Ugo, P.; Martin, C. R. Nanotechnology 2006, 17, 3951-3956. (19) Spohr, R. Radiation Measurements 2005, 40, 191-202. (20) Toerber, G.; Enge, W.; Beaujean, R.; Siegmon, G. Solid State Nucl. Track Detect., Proc. Int. Conf., 11th 1982, 307-310. (21) Apel, P. Y.; Blonskaya, I. V.; Didyk, A. Y.; Dmitriev, S. N.; Orelovitch, O. L.; Root, D.; Samoilova, L. I.; Vutsadakis, V. A. Nuclear Instruments & Met hods in Physics Research, Section B: Beam Interacti ons with Materials and Atoms 2001, 179, 55-62. (22) Chien, C. L.; Sun, L.; Tanase, M.; Bauer, L. A.; Hultgren, A.; Silevitch, D. M.; Meyer, G. J.; Searson, P. C.; Reich, D. H. Journal of Magnetism and Magnetic Materials 2002, 249, 146-155. (23) Martin, C. R. Science (Washington, D. C.) 1994, 266, 1961-1966. (24) Martin, C. R.; Mitchell, D. T. Analytical Chemistry 1998, 70, 322A-327A. (25) Hulteen, J. C.; Martin, C. R. Nanoparticles and Nanos tructured Films 1998, 235-262. (26) Possin, G. E. Review of Scientific Instruments 1970, 41, 772-774. (27) Nishizawa, M.; Menon, V. P.; Martin, C. R. Science (Washington, D. C.) 1995, 268, 700702. (28) Jirage, K. B.; Hulteen, J. C.; Martin, C. R. Science (Washington, D. C.) 1997, 278, 655658. (29) Hornyak, G. L.; Patrissi, C. J.; Martin, C. R. Journal of Physical Chemistry B 1997, 101, 1548-1555. (30) Williams, W. D.; Giordano, N. Review of Scientific Instruments 1984, 55, 410-412. (31) Hulteen, J. C.; Jirage, K. B.; Martin, C. R. Journal of the American Chemical Society 1998, 120, 6603-6604. (32) Hou, Z.; Abbott, N. L.; Stroeve, P. Langmuir 2000, 16, 2401-2404. (33) In Handbook of Conducting Polymers, 2 ed.; Skotheim, T. A., Elsenbaumer, R. L., Reynols, J. R., Eds., 1997; Vol. p, pp 1112. (34) Duchet, J.; Legras, R.; Demoustier-Champagne, S. Synthetic Metals 1998, 98, 113-122. (35) Demoustier-Champagne, S.; Stavaux, P.-Y. Chemistry of Materials 1999, 11, 829-834. (36) Sukeerthi, S.; Contractor, A. Q. Analytical Chemistry 1999, 71, 2231-2236.

PAGE 106

106 (37) Che, G.; Lakshmi, B. B.; Martin, C. R.; Fisher, E. R. Langmuir 1999, 15, 750-758. (38) Che, G.; Lakshmi, B. B.; Fisher, E. R.; Martin, C. R. Nature (London) 1998, 393, 346349. (39) Kyotani, T.; Tsai, L.-f.; Tomita, A. Chemical Communications (Cambridge) 1997, 701702. (40) Lakshmi, B. B.; Dorhout, P. K.; Martin, C. R. Chemistry of Materials 1997, 9, 857-862. (41) Lakshmi, B. B.; Patrissi, C. J.; Martin, C. R. Chemistry of Materials 1997, 9, 2544-2550. (42) Cepak, V. M.; Hulteen, J. C.; Che, G.; Jirage, K. B.; Lakshmi, B. B.; Fisher, E. R.; Martin, C. R. Journal of Materials Research 1998, 13, 3070-3080. (43) Cepak, V. M.; Hulteen, J. C.; Che, G.; Jirage, K. B.; Lakshmi, B. B.; Fisher, E. R.; Martin, C. R.; Yoneyama, H. Chemistry of Materials 1997, 9, 1065-1067. (44) Martin, B. R.; Dermody, D. J.; Reiss, B. D.; Fang, M.; Lyon, L. A.; Natan, M. J.; Mallouk, T. E. Advanced Materials (Weinheim, Germany) 1999, 11, 1021-1025. (45) Bowes, C. L.; Malek, A.; Ozin, G. A. Chemical Vapor Deposition 1996, 2, 97-103. (46) Shelimov, K. B.; Moskovits, M. Chemistry of Materials 2000, 12, 250-254. (47) Foss, C. A., Jr.; Hornyak, G. L.; Stockert, J. A.; Martin, C. R. Journal of Physical Chemistry 1994, 98, 2963-2971. (48) Blondel, A.; Meier, J. P.; Doudin, B.; Ansermet, J. P. Applied Physics Letters 1994, 65, 3019-3021. (49) Schwarzacher, W. Electrochemical Society Interface 1999, 8, 20-24. (50) Ang, L.-M.; Hor, T. S. A.; Xu, G.-Q.; Tung, C.-h.; Zhao, S.; Wang, J. L. S. Chemistry of Materials 1999, 11, 2115-2118. (51) Menon, V. P.; Martin, C. R. Analytical Chemistry 1995, 67, 1920-1928. (52) Parthasarathy, R. V.; Phani, K. L. N.; Martin, C. R. Advanced Materials (Weinheim, Germany) 1995, 7, 896-897. (53) Cepak, V. M.; Martin, C. R. Chemistry of Materials 1999, 11, 1363-1367. (54) Patrissi, C. J.; Martin, C. R. Journal of the Electrochemical Society 1999, 146, 31763180. (55) Lee, S. B.; Martin, C. R. Anal Chem FIELD Full Journal Title:Analytical chemistry 2001, 73, 768-775. (56) Miller, S. A.; Martin, C. R. Journal of Electroanalytical Chemistry 2002, 522, 66-69.

PAGE 107

107 (57) Kyotani, T.; Tsai, L.-f.; Tomita, A. Chemistry of Materials 1996, 8, 2109-2113. (58) Yuan, Z. h.; Huang, H.; Liu, L.; Fan, S. s. Chemical Physics Letters 2001, 345, 39-43. (59) Kyotani, T.; Ma, Z.; Tomita, A. Carbon 2003, 41, 1451-1459. (60) Bayley, H.; Martin, C. R. Chemical Reviews (W ashington, D. C.) 2000, 100, 2575-2594. (61) Gu, L. Q.; Braha, O.; Conlan, S.; Cheley, S.; Bayley, H. Nature 1999, 398, 686-690. (62) Movileanu, L.; Cheley, S.; Bayley, H. Biophysical journal 2003, 85, 897-910. (63) Gu, L. Q.; Bayley, H. Biophysical journal 2000, 79, 1967-1975. (64) Kasianowicz, J. J.; Burden, D. L.; Han, L. C.; Cheley, S.; Bayley, H. Biophysical journal 1999, 76, 837-845. (65) Astier, Y.; Braha, O.; Bayley, H. Journal of the American Chemical Society 2006, 128, 1705-1710. (66) Mathe, J.; Aksimentiev, A.; Nelson, D. R.; Schulten, K.; Meller, A. Proceedings of the National Academy of Sciences of the United States of America 2005, 102, 12377-12382. (67) Howorka, S.; Nam, J.; Bayley, H.; Kahne, D. Angewandte Chemie, International Edition 2004, 43, 842-846. (68) Gu, L. Q.; Cheley, S.; Bayley, H. Science 2001, 291, 636-640. (69) Schmidt, J. Journal of Materials Chemistry 2005, 15, 831-840. (70) Heins Elizabeth, A.; Siwy Zuzanna, S.; Baker Lane, A.; Martin Charles, R. Nano letters 2005, 5, 1824-1829. (71) Li, J.; Stein, D.; McMullan, C.; Bran ton, D.; Aziz, M. J.; Golovchenko, J. A. Nature (London, United Kingdom) 2001, 412, 166-169. (72) Storm, A. J.; Chen, J. H.; Ling, X. S.; Zandbergen, H. W.; Dekker, C. Nature Materials 2003, 2, 537-540. (73) Li, J.; Gershow, M.; Stein, D. ; Brandin, E.; Golovchenko, J. A. Nature Materials 2003, 2, 611-615. (74) Han, A.; Schurmann, G.; Mondin, G.; Bitterli, R. A.; Hegelbach, N. G.; de Rooij, N. F.; Staufer, U. Applied Physics Letters 2006, 88, 093901/093901-093901/093903. (75) Mara, A.; Siwy, Z.; Trautmann, C.; Wan, J.; Kamme, F. Nano Letters 2004, 4, 497-501. (76) Harrell, C. C.; Choi, Y.; Horne, L. P.; Baker, L. A.; Siwy, Z. S.; Martin, C. R. Langmuir, ACS ASAP.

PAGE 108

108 (77) Siwy, Z.; Trofin, L.; Kohli, P.; Baker, L. A.; Trautmann, C.; Martin, C. R. Journal of the American Chemical Society 2005, 127, 5000-5001. (78) Patolsky, F.; Zheng, G.; Lieber, C. M. Analytical Chemistry 2006, 78, 4260-4269. (79) Lee, S.; Zhang, Y.; White, H. S.; Harrell, C. C.; Martin, C. R. Analytical Chemistry 2004, 76, 6108-6115. (80) DeBlois, R. W.; Wesley, R. K. Journal of virology 1977, 23, 227-233. (81) Apel, P. Y.; Korchev, Y. E.; Siwy, Z.; Spohr, R.; Yoshida, M. Nuclear Instruments & Methods in Physics Research, Section B: Beam Interactio ns with Materials and Atoms 2001, 184, 337-346. (82) Fleischer, R. L.; Viertl, J. R. M.; Price, P. B. Review of Scientific Instruments 1972, 43, 1708-1709. (83) Shul, R. J.; Pearton, S. J. Handbook of Advanced Plasma Processing Techniques; Sringer: Berlin Heidelberg: New York, 2000. (84) Kuo, C.-W.; Shiu, J.-Y.; Cho, Y.-H.; Chen, P. Advanced Materials (Weinheim, Germany) 2003, 15, 1065-1068. (85) Kuo, C.-W.; Shiu, J.-Y.; Chen, P. Chemistry of Materials 2003, 15, 2917-2920. (86) Hegemann, D.; Brunner, H.; Oehr, C. Nuclear Instruments & Methods in Physics Research, Section B: Beam Intera ctions with Materials and Atoms 2003, 208, 281-286. (87) Hille, B. Ion Channels of Excitable Membranes; Sinauer Associates, 2001. (88) Siwy, Z.; Fulinski, A. Physical Review Letters 2002, 89, 198103/198101-198103/198104. (89) Siwy, Z.; Kosinska, I. D.; Fulinski, A.; Martin, C. R. Physical Review Letters 2005, 94, 048102/048101-048102/048104. (90) Chou, T. Journal of Chemical Physics 1999, 110, 606-615. (91) Kim, K. S.; Davis, I. S.; Macpherson, P. A.; Pedley, T. J.; Hill, A. E. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences 2005, 461, 273-296. (92) Harrell, C. C.; Siwy, Z. S.; Martin, C. R. Small 2006, 2, 194-198. (93) Li, N.; Yu, S.; Harrell, C. C.; Martin, C. R. Analytical Chemistry 2004, 76, 2025-2030. (94) Yu, S.; Lee, S. B.; Kang, M.; Martin, C. R. Nano Letters 2001, 1, 495-498. (95) Harrell, C. C.; Lee, S. B.; Martin, C. R. Analytical Chemistry 2003, 75, 6861-6867. (96) Martin, C. R.; Nishizawa, M.; Jirage, K.; Kang, M. Journal of Physical Chemistry B 2001, 105, 1925-1934.

PAGE 109

109 (97) Siwy, Z.; Heins, E.; Harrell, C. C.; Kohli, P.; Martin, C. R. Journal of the American Chemical Society 2004, 126, 10850-10851. (98) Harrell, C. C.; Kohli, P.; Siwy, Z.; Martin, C. R. Journal of the American Chemical Society 2004, 126, 15646-15647. (99) Kovtyukhova, N. I.; Mallouk, T. E. Chemistry--A European Journal 2002, 8, 4354-4363. (100) Harrell, C. C.; Choi, Y.; Horne, L. P.; Baker, L. A.; Siwy, Z. S.; Martin, C. R. Langmuir 2006, 22, 10837-10843. (101) Schiedt, B.; Healy, K.; Morrison, A. P.; Neumann, R.; Siwy, Z. Nuclear Instruments & Methods in Physics Research, Section B: Beam Interactio ns with Materials and Atoms 2005, 236, 109-116. (102) Siwy, Z. S. Advanced Functional Materials 2006, 16, 735-746. (103) Merkulov, V. I.; Guillorn, M. A.; Lowndes, D. H.; Simpson, M. L.; Voelkl, E. Applied Physics Letters 2001, 79, 1178-1180. (104) Heins, E. A.; Baker, L. A.; Siwy, Z. S.; Mota, M. O.; Martin, C. R. Journal of Physical Chemistry B 2005, 109, 18400-18407. (105) Siwy, Z.; Apel, P.; Baur, D.; Dobrev, D. D.; Korchev, Y. E.; Neumann, R.; Spohr, R.; Trautmann, C.; Voss, K.-O. Surface Science 2003, 532-535, 1061-1066. (106) Sasou, M.; Sugiyama, S.; Yoshino, T.; Ohtani, T. Langmuir 2003, 19, 9845-9849. (107) Heim, M.; Cevc, G.; Guckenberger, R.; Knapp, H. F.; Wiegrabe, W. Biophys J FIELD Full Journal Title:Biophysical journal 1995, 69, 489-497. (108) Sun, L.; Searson, P. C.; Chien, C. L. Applied Physics Letters 1999, 74, 2803-2805. (109) Chen, L. H.; AuBuchon, J. F.; Gapin, A.; Daraio, C.; Bandaru, P.; Jin, S.; Kim, D. W.; Yoo, I. K.; Wang, C. M. Applied Physics Letters 2004, 85, 5373-5375. (110) Che, G.; Lakshmi, B. B.; Martin, C. R.; Fisher, E. R.; Ruoff, R. S. Chemistry of Materials 1998, 10, 260-267. (111) Miller, S. A.; Young, V. Y.; Martin, C. R. Journal of the American Chemical Society 2001, 123, 12335-12342. (112) Siwy, Z.; Apel, P.; Dobrev, D.; Neumann, R.; Spohr, R.; Trautmann, C.; Voss, K. Nuclear Instruments & Methods in Physics Research, Section B: Beam Interactions with Materials and Atoms 2003, 208, 143-148.

PAGE 110

110 (113) Schmidt-Rohr, K.; Hu, W.; Zumbulyadis, N. Science (Washington, D. C.) 1998, 280, 714-717. (114) Ganz, M.; Feige, G.; Edelstein, N.; Bucher, J. ; Zhu, T. C.; Vater, P.; Brandt, R.; Vetter, J. Nuclear Tracks and Radiation Measurements 1988, 15, 799-802. (115) Vater, P. Nuclear Tracks and Radiation Measurements 1988, 15, 743-749. (116) Trautmann, C.; Bruechle, W.; Spohr, R.; Vetter, J.; Angert, N. Nuclear Instruments & Methods in Physics Research, Section B: Beam Interactio ns with Materials and Atoms 1996, 111, 70-74. (117) Hille, B. Ionic Channels of Excitable Membranes, 2nd ed.; Sinauer Associates, Inc: Sunderland, MA, 1992. (118) Sakmann, B.; Neher, E. Single-Channel Recording, 2nd ed.; Plenum Press: New York, 1995. (119) Sperelakis, N. Cell Physiology; Academic Press: San Diego, Califonia, 1998. (120) Siwy, Z.; Gu, Y.; Spohr, H. A.; Baur, D.; Wolf-Reber, A.; Spohr, R.; Apel, P.; Korchev, Y. E. Europhysics Letters 2002, 60, 349-355. (121) Siwy, Z.; Fulinski, A. American Journal of Physics 2004, 72, 567-574. (122) Soltys, P. J.; Ofsthun, N. J.; Zydney, A. L. Journal of Membrane Science 1996, 118, 199212. (123) Patlak, C. S.; Goldstein, D. A.; Hoffmann, J. F. J. Theoret. Biol. 1963, 5, 426-442. (124) Crank, J. The Mathematics of Diffusion. 2d Ed, 1975. (125) Crank, J.; Park, G. S.; Editors Diffusion in Polymers, 1968. (126) Renkin, E. M. J Gen Physiol FIELD Full Journal Title:The Journal of general physiology 1954, 38, 225-243. (127) Davidson, M. G.; Deen, W. M. Macromolecules 1988, 21, 3474-3481. (128) Ferry, J. D. Journal of General Physiology 1936, 20, 95-104. (129) Vlassiouk, I.; Siwy, Z. S. Nano Letters 2007, 7, 552-556. (130) Siwy, Z.; Dobrev, D.; Neumann, R.; Trautmann, C.; Voss, K. Applied Physics A: Materials Science & Processing 2003, 76, 781-785. (131) Packard, N.; Shaw, R. Los Alamos National Laboratory, Preprint Archive, Condensed Matter 2004, 1-17, arXiv:cond-mat/0412626.

PAGE 111

111 BIOGRAPHICAL SKETCH John Edwardson Wharton was born in St. Kitts, West I ndies. He spent two years at the St. Kitts-Nevis Teachers College, earning a diploma in education in 1985. After graduation, he taught for two and a half years, before taki ng up employment at the St. Kitts Biomedical Research Foundation, where he worked for a few y ears. He later went on to pursue a B.S. in chemistry with physics at the University of th e Virgin Island from 1992 to 1996. In August, 1999, John began his graduate work in analyti cal chemistry under the guidance of Prof. Dr. Charles R. Martin. He earned his M.S. in 2003 and continued in the Martin group in pursuit of his Ph.D. He completed his research in the sp ring of 2007, obtaining a Doctor of Philosophy in analytical chemistry in the area of nanostructu red asymmetric pore design and fabrication for biosensor applications.

## Material Information

Title: Fabrication of Asymmetric Pores for Biosensors and Transport Studies
Physical Description: Mixed Material

## Record Information

Source Institution: University of Florida
Holding Location: University of Florida
System ID: UFE0019751:00001

## Material Information

Title: Fabrication of Asymmetric Pores for Biosensors and Transport Studies
Physical Description: Mixed Material

## Record Information

Source Institution: University of Florida
Holding Location: University of Florida
System ID: UFE0019751:00001

Full Text

FABRICATION OF ASYMMETRIC PORES FOR BIOSENSORS AND TRANSPORT
STUDIES

By

JOHN EDWARD SON WHARTON

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

2007

O 2007 John Edwardson Wharton

To my family, for their continued support and love

ACKNOWLEDGMENTS

I wish to acknowledge the many people who have given me advice, help, and

encouragement during my years in graduate school.

I would like to thank Prof. Dr. Charles R. Martin and the entire Martin group for the

opportunity to work with them over the years. Prof. Martin was continuously supportive and

always willing and ready to discuss some of our new ideas for research. I appreciate the level of

independence that Prof. Martin allowed all of his students. Undoubtedly, this created an

atmosphere of confidence among the students where creativity flourished.

I would like to thank Drs. Punit Kohli, Zuzanna Siwy, and Lane Baker for their patience

and expert advice along the way. I wish to thank Fan Xu, Lindsay Sexton, Pu Jin, Lloyd Horne,

Stefanie Sherrill, and Warren Mino for very important contributions to some of my proj ects;

these great students have been very patient with me and are always willing to help with very high

interest and good spirits.

Finally, I would like to thank my family and friends for showing confidence in me and for

giving their love and support throughout the years.

page

ACKNOWLEDGMENTS .............. ...............4.....

LIST OF FIGURES .............. ...............8.....

AB S TRAC T ............._. .......... ..............._ 1 1..

CHAPTER

1 INTRODUCTION AND BACKGROUND .............. ...............13....

Ion Track-Etch Method............... ...............14.
Track Form ation ................. ... ........... ...............14.......
Acceleration and Irradiation Facilities .............. ...............15....
Chemical Etching of lon Tracks ................. ...............15........... ...
The Effect of Storage of Tracked Material............... ...............16
Effect of Etch Promoters ................. ...............16.......... ....
Etch Properties of Selected Polymers ................. ...............17........... ...
Effect of Thermal Annealing .................. ...............17................
The Effect of Temperature during Etching .............. .....................17
Effect of Detergents ................... ...............18................
Etch Properties in Muscovite Mica ................. ......... ...............18.....
Template Synthesis................ ..............1
Template Synthesis Strategies ................. ...............19........... ....
Electroless Deposition ................... .......... ...............19.......
Chemical Vapor Deposition (CVD) .............. ...............21....
Resistive-Pulse Sensing ................. ...............21.......... ......
Plasma-Based Etching .............. ...............23....
Asymmetric Diffusion .............. ...............24....
Dissertation Overview .............. ...............25....

2 A METHOD FOR REPRODUCIBLY PREPARING SYNTHETIC NANOPORES
FOR RESISTIVE-PULSE BIOSENSORS ................. ...............36........... ....

Introducti on ................. ...............36.................

Experimental ................. ...............37.................
First Etch Step .............. ..... ... .. .......3
Determination of the Diameter of the Base Opening ................. .......... ...............3 8
Electrochemical Measurement of the Tip Diameter ................. .......... ................3 8
Second Etch Step ................... ..... ....... ..... ........ ...........3
Bovine Serum Albumin (BSA) Resistive-Pulse Sensing .............. .....................3
Results and Discussion .............. ........... .... .. .. .. ............3
Conical Shaped Nanopores are Ideal Resistive-Pulse Sensor Elements .........................39
The Core Technology: The Track-Etch Method .............. ...............40....
The First Etch Step .............. ...............40....

Measuring the Diameter of the Tip Opening after the First Etch ................. ................41
The Second Etch Step.................. .... .... .. .... ..... .......4
Measuring the Diameter of the Tip Opening after the Second Etch .............. .... ...........43
Reproducibly Varying the Tip Diameter ................. ...............44........... ...
The Mathematical Model .............. ...............45....
Electro chemical Detail s............... ...............47
Conclusions............... ..............4

3 ETCH-FILL-ETCH METHOD FOR PREPARING TAPERED PORES IN ION
TRACKED MICA FILMS .............. ...............55....

Introducti on ................. ...............55.................

Experimental ................. ...............57.................
M materials ................. .. .... ..... ... .. .......... .. ... ..........5
Initial Etching of Mica Tracks to Prepare Very Small Pores ................ ............... ....58
Preparation of Tin Sensitizing Solution .............. ...............58....
Preparation of Silver Plating Solution ................. ...............58........... ...
Filling the Pores with Silver Wires .................. ...............59..
Etching Silver Filled Mica to form Tapered Pores ................. ................ ......... .59
Making Replicas of the Tapered Pores ................... .......... ............... 60 ...
Preparation of the Carbon Tube Replicas for SEM Imaging ................. ............... ....60
Results and Discussion .............. ...............61....
Conclusion ................ ...............63.................

4 ELECTROLESS AU PLATING OF TRACK-ETCHED KAPTON POLYIMIDE
NANOPOROU S MEMBRANE S ................. ...............69.......... ......

Introducti on ................. ...............69.................

Experim ental ................. ...............70.......... ......
M material s ................. ...............70.......... .....
Chemical Etching .............. ...............70....
Electroless Plating of Kapton ................. ...............71.......... ....
Pore Diameter Measurement ................. ...............71................
Results and Discussion .............. .... ...............72
SEM and lon Current Measurements .............. ...............72....
Atomic Force Microscope Images............... ...............72.
Conclusion ................ ...............72.................

5 ASYMMETRY IN DIFFUSIONAL TRANSPORT OF MOLECULES THROUGH
KAPTON CONICAL NANOPORES .............. ...............78....

Introducti on .............. .. .... .... .. ....... ..._ ... ...........7
Diffusional Transport Described by Fick's Laws ................. .............................79
Hindered Diffusion in Cylindrical Pores ................. ...............80........... ...
Hindered Diffusion in Conical Pores .............. ...............81....

Experim ental ................. ...............83.......... ......
M material s ................. ...............83.......... .....

Kapton Polyimide Membrane............... ...............84

Chemical Etching of Membrane Tracks .....__.....___ ..........._ ...........8
Pore Diameter Measurement ............. ...... ...............86...

Transport Measurement ............_...... ...............86....
Viscosity Measurements ............_...... ...............87....
Results and Discussion .............. ...............87....
Membrane Characterization .............. ...............87....

Transport Measurements of Phthalazine .............. ......___.....__ ............8
The Influence of Cosolute Concentration on Asymmetry ....._____ ..... .. ....__..........89
Conclusions............... ..............9

6 CONCLUSION................ ..............10

LIST OF REFERENCES ............_ ..... ..__ ...............104..

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

LIST OF FIGURES

Figure page

1-1 Swift heavy ions impinge on a dielectric material creating damaged ion tracks. ..............27

1-2 After irradiation, the materials are subj ect to chemical etching which preferentially
removes the latent ion track. .............. ...............27....

1-3 Etched pore geometry in a homogeneous isotropic medium to a first approximation,
showing track etch rate, VT, and bulk at rate, Va ...._.................... ................ ..2

1-4 Scanning electron micrographs of Au nanocones replicas of PET conical pores. ...........28

1-5 Scanning electron micrographs of etched particle tracks in single-crystal mica.............. .29

1-6 Scanning electron micrographs of a porous polycarbonate, alumina and mica
membranes used for template synthesi s (A-C, respectively) ................. ........_._. .......30O

1-7 Schematic diagram of Au electroless plating procedure............... ...............3

1-8 Schematic illustration of Au nanotubes obtained from electroless gold deposition ..........32

1-9 Schematic illustration of resistive-pulse sensing ................. ...............33...............

1-10 Essential features of the staphylococcal a -hemolysin pore shown in a cross- section
based on the crystal structure. .............. ...............34....

1-11 Schematic of a conical nanopore sensor element showing the base diameter and
range of tip diameters used in these studies (drawing to scale) .........._.. ..........._......3 5

2-1 Schematic of a conical nanopore sensor element and etching cell. ........... ..................48

2-2 A typical current-voltage curve used to measure the tip diameter of the conical
nanopore. .........._._.. ...._ ... ...............49.....

2-3 Current-time transients obtained during the second etch step for three membranes
that were subj ected to the same first etch. .......... ...............50......

2-4 Scanning electron micrograph of the base openings of two conical nanopores in a
multi-track PET membrane ................. ...............51........... ....

2-5 Scanning electron micrograph of conical gold nanotubes deposited in a conical
nanopore membrane ................. ...............52.................

2-6 Plot of tip diameter measured after the second etch step vs. the final nanopore ion
current (If) at which the second etch was stopped. .......... ...............53......

2-7 Current-pulse data obtained for a prototype protein analyte, bovine serum albumin
(BSA) using PEG-modified conical nanotube sensors. ........... ....... ............... 5

3-1 After irradiation, the materials are subj ect to chemical etching which preferentially
removes the damaged ion track. ........... ......__ ...............64...

3-2 Schematic of cell used to do the etching and to make all electrochemical
measurements ................. ...............65.................

3-3 Schematic diagram of etch-fill-etch method ................. ...............66...............

3-4 Scanning electron micrographs of mica membrane that was exposed on one face to
20% HF andl0% nitric acid solution at 250C for 3 hrs. ........... ......................6

3-5 Scanning electron micrographs of mica membrane that was exposed on one face to
40% HF andl0% nitric acid solution at 250C for 3 hrs. .......... .......................6

3-6 Scanning electron micrograph of carbon tapered nanotube replica of the mica tapered
pore. ............. ...............68.....

3-7 Low magnification SEM images of carbon tapered nanotube replicas of the mica
tapered pore............... ...............68..

4-1 Definition of bulk etch rate Vb and track etch rate Vt ................. ................ ........ .73

4-2 Schematic of cell used for electrochemical measurements............... ..............7

4-3 Pore diameters for different Au plating times............... ...............75.

4-4 Pore diameter as a function of plating time with measurements taken from SEM
image (0-7.5 hrs) and ion current resistance measurements (8-12 hrs). ............................76

4-5 Atomic Force micrographs of Kapton and PC membranes before and after electroless
Au plating............... ...............77

5-1 The partitioning of a spherical molecule of radius, a, in cylindrical pores of radius, r.....90

5-2 Spherical molecule of radius, a, moving within a cylindrical pore of radius, r. ................91

5-3 Swift heavy ions impinge on a dielectric solid leading to damaged ion tracks witch
can be chemically etched to form pores............... ...............91.

5-4 Definition of bulk etch rate Vb and track etch rate Vt ........._..... ...._... ........_.......92

5-5 Conductivity cell used to prepare conical pores. ............. ...............92.....

5-6 Etching curve showing moment of breakthrough with sharp increase in ion current. ......93

5-7 Kapton membran showing large opening (base) of conical pores ................. ................. 94

5-8 Experimental set-up for transport measurements .............. ...............94....

5-9 Scanning electron micrograph of base side showing pore density (10' pores/cm2) Of
Kapton membrane. ........._.._.. ...._... ...............95....

5-10 Scanning electron micrograph of the base showing diameter of 1.68 pm ................... ......95

5-11 A typical current-voltage curve used to measure the tip diameter of the conical
nanopores. ............. ...............96.....

5-12 Base and tip fluxes for 1 mM phthalazine feed solution. ............. .....................9

5-13 Base and tip fluxes for 3 mM phthalazine feed solution. ............. .....................9

5-14 Base and tip fluxes for 10 mM phthalazine feed solution. ............. .....................9

5-15 Base and tip fluxes for 50 mM phthalazine feed solution. ............. .....................9

5-16 Asymmetric behavior of flux versus concentration ......._._............_ ........._..__....98

5-17 Theoretical and experimental tip flux vs concentration. ........._..... ....._... ............99

5-18 Increase of flux through base and tip, respectively, with respect to the flux measured
at 0.1 m M .............. ...............99....

5-19 Experimental set-up for transport measurements of phthalazine with cosolute
dextrose. .............. ...............100....

5-20 Calibration curve of phthalazine, and phthalazine with the highest concentration of
dextrose cosolute used. ............. ...............100....

5-21 Viscosity of dextrose and phthalazine over the concentration range used ...................... 101

5-22 Influence of dextrose cosolute concentrations on the base and tip diffusion from 5
mM phthalazine .............. ...............101....

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

FABRICATION OF ASYMMETRIC PORES FOR BIOSENSORS AND TRANSPORT
STUDIES

By

John Edwardson Wharton

May 2007

Chair: Charles R. Martin
Major: Chemistry

The goals of this research are to develop asymmetric nanopores and nanotubes (single and

multipore) in polymers and mica membranes by the track-etch method, and an extension of this

technology, for sensing, nanostructure fabrication, and to investigate the transport properties of

conical pores. The first part of this work describes an extension of the track-etch method to

make conical pores in a reproducible fashion. We have demonstrated here that we can, not only

reproducibly prepare track-etched based conical nanopore sensor elements, but that we can

predict from the experimental parameters used during the second etch, what the diameter of the

all-important nanopore tip will be. For these reasons, we believe that the track-etch method will

prove to be the technology of choice for taking artificial-nanopore resistive-pulse sensors from

the bench top to the practical prototype-device stage of the R&D effort.

The second part of this work describes a method to make asymmetric pores in tracked

muscovite mica films using an etch-refill-etch approach. Tracks in the films were initially

etched away with hydrofluoric acid to form nanoporous membranes. These nanopores were then

refilled with silver nanowires or "metal tracks" using an electroless plating method. One face of

the membrane was then exposed to a solution of hydrofluoric acid and nitric acid, which etched

the bulk material and the nanowires respectively, at two different rates. By controlling the

concentration ratio of hydrofluoric acid to nitric acid, tapered pores with diamond shaped cross-

section were obtained. Replicas of the asymmetric pores were accomplished by carbon vapor

deposition, and scanning electron microscopy was used to give evidence of the resulting

nanotubes. In this study, excellent control over tip size and cone angle was demonstrated.

In the third section, electroless gold plating properties on the surface and pore walls of

track-etched Kapton polyimide nanoporous membranes were studied. Scanning electron

microscopy (SEM), atomic force microscopy (AFM), and ion current measurements were used to

characterize the surfaces and dimensions of the pores in the membrane. Nanoporous Kapton

polyimide membranes were electroless gold plated over different times and the pore diameter

characterized using SEM. Ion current measurements were used to measure the diameter of very

small pores. AFM images show that after electroless gold plating, the gold surface layers are

smooth compared to a similarly structured polycarbonate membrane. Electroless plating the

membranes for 12 hours produced Au wires in the pores. Removing the membrane by oxygen

plasma revealed that the plating in the walls is also relatively smooth.

The Einal part of this work describes asymmetric diffusional transport of neutral molecular

species through nanoporous polymer membranes. Membranes containing conical nanopores

from polyimide (Kapton HN Dupont) foils were prepared by the track-etching technique, based

on irradiation of the polymer with swift heavy ions and subsequent etching of the latent tracks.

The transport properties (e.g., flux) of these membranes were investigated using UV-Vis

spectroscopy. Transport experiments were performed with bare polymer membranes without

any modification. We report the preferential direction of the flow of molecules through conical

nanopores and spatial or concentration dependence of diffusion coefficient for a particular flow

direction.

CHAPTER 1
INTRODUCTION AND BACKGROUND

Nanopore fabrication technology has produced nanoporous materials that are potential

candidates for applications in various fiels, such as bionanotechnology, 1, 2 gas separation,3

catalysts4 and micro-electronics2, 5 In particular, the track-etch method has become an

indispensable technology for the production of nano- and microstructured materials.6 This

technology can be applied directly to most polymers and via the replication technique- to a

wide variety of materials, including metals.6 Track-etch membranes were first marketed three

decades ago and remain the best product for a number of biological, medical, analytical and

scientific applications.6

Recently, there has been a surge of interest in developing abiotic analogues of biological

nanopores as sensing elements for chemicals and biological sensors.' We have explored the

fabrication of such synthetic nanopores using the track-etch method. Particular interest is in the

fabrication of synthetic conical pores for resistive-pulse sensing. Resistive-pulse sensing using

conical nanopores is in its infancy. Studies in this area could prove beneficial for future

applications of conical pores for science and technology.

Membranes and porous materials have found various applications in filtration and

separation processes.8 9, 10 Modern biotechnology has posed new challenges in the application of

such membranes, and requires pores with diameters similar to those of molecules under study,ll

(e.g., as small as several nanometers). The nanometer scale of such pores is necessary in both

achieving optimal control of the flow of biomolecules, as well as in developing sensors for their

detection. The transport properties of such nanometer scale pores are not well understood yet.

The hint that nanopores behave differently from micropores, comes from Mother Nature.12

Biological channels and pores have diameter of ~ 1 nm and are critical for functioning of living

organism s.

This chapter is divided into six sections which provide the background information for this

research. Section 1 reviews the track-etch method, first discovered by Price and Walker, for

preparation of pores in dielectric materials. The track-etch method is used in the preparation of

conical pores described in Chapters 2, 3, 4 and 5. Section 2 reviews membrane-based template

synthesis. This method, pioneered by the Martin group, is used in preparation of nanostructured

materials described in Chapters 2 and 3. Section 3 review resistive-pulse sensing; which is the

focus of the application of Chapter 2 and 3. Section 4 describes plasma based etching; a

procedure necessary for the liberation of carbon tube in mica templates in Chapter 2. In Section

5, an introduction of Asymmetric Diffusion is given. Asymmetric diffusion is the topic of

Chapter 5. Finally, Section 6 is the dissertation overview.

Ion Track-Etch Method

Track Formation

When dielectric materials, such as polymers, ceramics and minerals, are bombarded by

swift ions, latent tracks are formed along the path of the ions (Figure 1-1).13, 14 I0n track

materials can be divided into two categories: (a) single-tracked and (b) multiple-tracked

materials. Single-tracked materials can be produced by controlling the beam optics and fluence

of the heavy ion beam." Commonly tracked materials are Makrofol-KG, Kapton-H, PVDF,

mica films, cellulose nitrate, CR-39 and Lexan polycarbonate. Swift ion beams are produced by

cyclotrons or linear accelerator and the radiation is characterized by extremely high linear energy

transfer.6 The conventional ionization radiation sources, such as radioactive isotopes or electron

accelerators,16 are less sophisticated and less expensive and are mostly used in the industrial

processes.6 Present-day heavy ion accelerators provide beams with ion energies in the order of

10MeV/u and even 100MeV/u which expands the treatment dept up to millimeters.6

About a half dozen heavy ion accelerators employed for irradiation of materials on the

industrial scale exist. The Tanden Van de Graaff accelerators at Brookheaven National

Laboratory are used to bombard materials with ions for manufacturing and testing purposes.6 At

the Grand Accelerator National d'lons Lours, France, ions are produced in an electron cyclotron

resonance(ECR) source.6 The cyclotron of Louvain la Neuve is a multiparticle variable energy,

cyclotron capable of accelerating protons, alpha particles and heavy ions. The RFQ + cyclotron

combination at the Hahn-Meitner Institute in Berlin, delivers intense beams of ion species such

as Kr or Xe with energies from approximately 1.5-6 MeV/ul7 At the Flerov Laboratory of

Nuclear Reactions(Dubna) a beam line connected at the U-400 cyclotron is equipped with

scanning systems which allows one to obtain a homogeneous distribution of ion tracks on the

target up to 60 cm in width and 6 cm in height.6 Additionally, a liner accelerator at GSI

(Darmstadt),6 the AVF cyclotron at TRCRE JAERI (Takasaki)6 and some others are used in

experiments on polymer modification.6

Chemical Etching of lon Tracks

After irradiation, the materials are subj ect to chemical etching which preferentially

removes the latent ion track (Figure 1-2)16 This etching process results in pore formation in the

material.16 Etching is the pore-size-determining and pore-shape-determining stage of the

technology. In a homogeneous isotropic medium, mainly two influential parameter describe the

etch process-the bulk etch rate VI3 and the track etch rate yr (Figure 1-3).6 The ratio of track etch

rate to the bulk etch rate is called the track-etch-ratio. When Vr is >> VI3, pores turn out to be

cylindrical as opposed to conical. In other words, high track-etch-ratio yields cylindrical pores,

where as low track-etch-ratio yields conical pores. The arctangent of the inverse track-etch-ratio

(VI3/VT) yields the half cone angle of the pore. The bulk etch rate depends on the material, on the

etchant composition and on the temperature.14 The track etch rate depends on the sensitivity of

The Effect of Storage of Tracked Material

The most important storage factors are the atmosphere in which the material is stored, the

temperature and the illumination conditions during storage. In the presence of oxygen the latent

ion tracks becomes susceptible to track etching. This is due to the oxidation of the radicals

formed during irradiation. When polymers are stored at temperatures close their glass transition

temperature, rearrangement on a molecular scale can take place which may lead to annealing of

the ion tracks. Storing under illumination may lead to photo oxidation and is able to increase the

track etch ratio by orders of magnitude. It is reported that tracks in poly(ethylene terephthalate)

(PET) may be sensitized by uv radiation of 310 to 400 nm. Soaking in weak solvents, such as

dimethyl formamide or water-soluble gass can sensitize ion tracks in PET.6

Effect of Etch Promoters

Etch promoters are organic solvents that accelerate the etch process when added to the etch

bath. It has been observed that track etch ratios in polycarbonate (PC) can be above ten

thousand. On the other hand, track etch ratio can be dramatically decreased down to 2 to 4 in

PET by the addition of solvents such as methanol, ethanol or propanol, leading to wide cone

angles. These organic solvents help to dissolve large fragments ready to move into the liquid

phase by disengaging them from their neighbors.6 Figure 1-4 shows an example of etch

promotion using ethanol in PET.l

Etch Properties of Selected Polymers

In PET (OOC-C6H4-COO-CH2CH2), the main points of etch attack are the partially

charged -COO- ester groups, which are hydrolyzed by alkalis. During alkaline etching the

ordinary bond between carbon and oxygen is broken which produces -COO- and HO- at the ends

of the formed fragments.

For polycarbonate (OOC-O-C6H4-C(CH3)2- 6H4), the main point of etch attack is the

carbonate group -O-COO-. During the alkaline etching, chemical bonds are ruptured on both

sides of the carbonate group, leading to the formation of carbonate ions, CO32-. The other

product is diphenylol HO-C6H4-C(CH3)2_ 6H4-OH.

In Polyimide (C6H4-O- C6H4), the preferential point of etch attack is the oxygen. At high

pH the imide group will be hydrolyzed. The etching mechanism is complex because of the

simultaneous factors of oxidation and alkalinity. The chemical reaction responsible for etching

different polyimide can be different because they are made up of monomer units. 19

Effect of Thermal Annealing

By increasing the temperature of the tracked material, ion tracks can be thermally

annealed. For polymers, heating above the glass transition temperature mobilizes the polymer

fragments formed along the ion path. These fragments are sucked into the voids of the

neighboring pristine material, wiping out the latent ion track.14

The Effect of Temperature during Etching

Etch rates usually increase with temperature. Therefore, to obtain high throughput makes

it necessary to work at high temperatures. It is found that by alternating the temperature during

etching, high aspect ratio pores with large diameters can be obtained. At room temperature

diffusion processes can be faster than chemical reactions of the etchant within the polymer. 14

From this basis, the technique to make pores with large cone angles were develop by soaking at

low temperature and etching at high temperature.20

Effect of Detergents

Amphiphilic detergents may increase the track-etch-ratio. These detergents have been

shown to produce nanopores with cigar-like shapes and very small entrance openings. It is

thought that amphiphilic detergents attach to the hydrophilic surface layer, rendering it less

permeable to the etchant.21

Etch Properties in Muscovite Mica

Damage tracks are form in single crystal muscovite mica (KAl2(AI SiO3010)(OH)2 fTOm ion

irradiation along the (001) direction. Because the etching rate along the tracks is much faster

than both the lateral and bulb etch rate, when the tracks are etched through, nanopores with small

cross-section and small taper angle (0.020) are created.22 In muscovite mica, all the pores are

diamond-shaped with inner angles very close to 600 and 1200 (Figure 1-5). All the pores in a

given sample have the same size and orientation.22 By correlating the results of scanning

electron microscopy and X-ray diffraction on etched mica crystals, it was found that the

orientation of the diamonds is exactly the same as that of the mica unit cell. The four sides of the

diamonds are parallel to the four oxygen-terminated planes within the unit cell. These facts

points out that the diamond-shaped pores have their origin in the mica crystal structure. Also,

this shows that the uniform diamond shape arises because the oxygen-terminated planes are

those with the slowest etch rate, and the pores are aligned because the template are single

crystals.22

Template Synthesis

The Martin group has pioneered a general method for the preparation of nanostructured

materials called template synthesis23-2 Template synthesis method entails depositing a desired

material of interest into a porous solid. The size and shape of the nanomaterial depend on the

dimentions of the nanocavities within the porous template material. Depending on the

membrane and synthetic material used, nanostructures such as solid nanofibers or nanotubes can

be obtained. The method is termed "general" because nearly any chemical synthesis method

used to prepare bulk materials can be adapted to synthesize materials. There are reports of

metals,10, 26-32 pOlymers,33-36 carbons37-39 and semiconductors40, 41 prepared by the template

synthesis method. More advance material preparation include composite nanostructured

material, both concentric and tubular42, 43 and segmented composites nanowires.44

Template Synthesis Strategies

Three commonly used templates are; porous polymers, alumina and mica membranes

(Figure 1-6). Some of the more common synthetic strategies used to prepare nanomaterials

include chemical vapor deposition,38, 45, 46 electrochemical47-49 and electroless deposition,50, 51

chemical and electrochemical polymerization,53 and sol-gel chemistry.40, 54 Special attentions

will be devoted to the electroless deposition inside polymers and mica, and chemical vapor

deposition in mica since these are the methods used for preparing templates in chapters 2, 3, and

4 respectively.

Electroless Deposition

The electroless deposition method involves the use of a chemical reducing agent to plate a

metal from a solution onto a surface. Unlike electrochemical deposition, a conductive surface is

not necessary. The key requirement of electroless deposition is arrangement of the chemistry

such that the kinetics of the homogenous electron transfer from the reducing agent to the metal

ions is slow. This is essential because the metal ions would simply be reduced in the bulk

solution for fast electron transfer. In electroless deposition, a catalyst needs to be coated onto the

pore walls so that reduction of the metal ion only occurs at the pore surfaces. Figure 1-7 shows

the schematic representation that was used to prepare silver and gold nanowires and nanotubes

within PET, Kapton and Mica track-etched membranes. The membranes were first exposed to a

sensitizer (Sn2+). This is accomplished by simply immersing the membrane for 45 minutes in a

solution that is 0.026 M in SnCl2 and 0.07 M in trifluoroacitic acid in 50/50 methanol/water. The

tin sensitizer binds to the pore walls and membrane surfaces via complexation with the amine,

carbonyl and hydroxyl groups. 63 After sensitization, the membrane is rinsed thoroughly with

methanol and immersed into an aqueous solution of ammoniac silver nitrate (0.029 M Ag

(NH3)2 ) for 5 minutes. A redox reaction occours in which the surface bound tin (II) is oxidized

to tin (IV) and the Ag' is reduced to elemental Ag. As a result, the pore walls and the membrane

surface become coated with nanoscopic silver particles. The membrane is again thoroughly

rinsed with methanol. The silver coated membrane is then immersed into a gold plating bath that

is 7.9 x 10-3 M in Na3Au (SO3)2, 0. 127 M Na2SO3, and 0.625 M in formaldehyde at 40C. The Au

galvanically displaces the Ag particles because the reduction potential of Au is more positive

than that of Ag. As a result, the pore walls and surfaces become coated with Au particles. These

particles are excellent catalytic sites for the oxidation of formaldehyde and the concurrent

reduction of Au (I) to Au (0).63 Without a catalyst, the kinetics of the electron transfer from the

reducing agent (formaldehyde) to Au (I) is slow; therefore, the gold plating continues on gold

particles instead of in the bulk solution. The reaction can be represented as follows:

2Au (I) + HCHO + 30H 4 HCOO- + 2H20 + 2Au (0) (1-1)

This method yields the Au nanowires or nanotubes within pores plus Au surface layers on

both face of the membrane. These structures run through the entire thickness of the template

membrane (Figure 1-8). By controlling the plating time, the inside diameter of the tubes can be

varied because the thickness of both the Au surface films and nanotube wall increase with

plating time. By controlling the plating time, the inside diameter of the nanotubes can be varied,

even as low as 1 nm in diameter.28 As a result, these membranes can be used in a simple

membrane permeation experiment to cleanly separate small molecules on the basis of molecular

size.28 Also, by chemisorbing appropriate thiols to the Au nanotube wall based on well known

gold-thiol chemistry, the Au nanotube membrane can be made to preferentially transport cations

vs. anions and hydrophobic vs. hydrophilic moleculeS.10, 27, 31, 55 In addition, Au nanotube

membranes are electronically conductive and can be charged electrostatically in an electrolyte

solution.27 This introduces ion transport selectivity, allowing the Au nanotube membranes to be

electromodulated between ideal-cation and idea-anion transport states.27 Thus these Au nanotube

memebranes are ideal model systems for studying how pore size, chemistry, and charge affect

the transport selectivity at the nanometer scale.

Chemical Vapor Deposition (CVD)

CVD is commonly used to prepare carbon nanomaterials. We38, 56 and others57 have

synthesized carbon nanotubes within the porous alumina membranes using CVD. This involves

placing an alumina membrane in a high-temperature furnace (ca. 7000C) and passing a gas such

as ethane, propene or ethylene through the membrane. Thermal decomposition of the gas occurs

on the pore walls, resulting in the deposition of carbon nanotubes within the pores. High surface

area microporous carbon with long-range order has been synthesized by using zeolite Y as a

template with propylene CVD.59 Besides carbon nanostructures, other nanomaterials have been

obtained by CVD. For example, the martin group has used a CVD method to coat an ensemble

of gold nanotubes with concentric TiS2 Outer nanotubes.43

Resistive-Pulse Sensing

Resistive-pulse60 SCHSOTS for molecular and macromolecules analytes60-77 USe a nanopore in

a synthetic or biological membrane as the sensor element. This method, which when applied to

such analytes is sometimes called stochastic sensing,60, 69 entails mounting the membrane

containing the nanopore between two electrolyte solutions, applying a transmembrane potential

difference, and measuring the resulting ion current flowing through the electrolyte-filled

nanopore. In simplest terms, when the analyte enters and translocates the nanopore, it transiently

blocks the ion current, resulting in a downward current pulse (Figure 1-9). The frequency of

such analyte-induced current pulses is proportional to the concentration of the analyte, and the

identity of the analyte is encoded in the magnitude and duration of the current pulse.60-77

The maj ority of such resistive-pulse biosensing data has been obtained using a biological

nanopore, a-hemolysin (a-HL), embedded in a supported lipid-bilayer membrane as the sensor

element (Figure 1-10).60-69 This biological nanopore sensor has two key advantages. First, it can

be made analyte selective by using chemical or genetic-engineering methods to attach molecular-

recognition agents to the nanopore. As a result numerous different analyte types including metal

ions,64 DNA,65, 66 prOteins,67 and small molecules68 have been selectively detected with the a-HL

nanopore. Second, the biological nanopore can be reproducibly prepared from the commercially

available a-HL protein, which is obviously of great importance if practical, real-world, sensing

devices are ultimately to be derived from this technology.' There is however a key impediment

to developing practical sensors based on the biological nanopore. This problem concerns the

fragility of the supported bilayer membrane that houses the nanopore. Such membranes typically

survive for periods of only hours before rupture, much too short of a time to make a practical

sensing device."

One approach for solving this problem is to replace the biological nanopore, and bilayer

membrane, with an artificial nanopore embedded in a mechanically and chemically robust

synthetic membrane.69-77, 79, 80 Such artificial nanopores are often prepared by modern

microlithographic methods, using for example a focused ion" or electron72 beam to bore the

nanopore into a silicon or Si3N3 membrane. We and others are exploring an alternative

technology, called the track-etch method,16, 19, 81, 82 for preparing nanopores for resistive-pulse

sensorS.70, 75-77, 79, 80 Analytes detected with prototype track-etched nanopore sensors include

small molecules,70 DNA,75, 76 prOteins7 and nanoparticles.79 Furthermore, there are older reports

of developing virus sensors based on track-etched nanopores.so

The sensor elements we evaluated were conically shaped nanoporeS70, 75-77, 79, 81 prepared

by the track-etch method in polyethylene terephthalate (PET) membranes. Such conical

nanopores have two openings the large-diameter (base) opening at one face of the membrane

and the small-diameter (tip) opening at the opposite face (Figure 1-11). Fabrication methods to

prepare these conically shaped nanopores for resistive pulse sensing, is addressed in chapters 1.

Plasma-Based Etching

Plasma etching, a dry etching process, has become a very useful means of removing small

quantities of material from a variety of substrates quickly and efficiently.83 Plasma processes

have been used in many highly sensitive integrated applications to precisely remove specific

materials from sample surfaces. To generate plasma, a pair of electrode is needed; one is

connected to a radio frequency (RF) voltage and the other is grounded. RF energy is applied to

the electrodes which accelerates electrons to increase their kinetic energy. The electrons collide

with a neutral gas to form a collection of gaseous species including ions, free radicals, electrons,

photons and neutrals.83 The gaseous species can react with the surface to be etched such that

reaction product is volatile and can be pumped away. There are several types of plasma etching-

physical etching (anisotropic), chemical etching isotropicc) and reactive ion (combination of

physical etching and chemical etching). Usually in plasma based etching, both chemical etching

and physical etching occur.83

There are many applications for plasma etching such as photoresist removal, glass-like

compound etching (e.g., SiO2) and polymer etching to produce microstructures and

nanostructures. For example, large-area, well-ordered, periodic nanopillar arrays with lateral

dimensions as small as 40 nm have been developed based on a combination of colloidal

lithography and plasma etching techniques.84, 85 The etching mask on silicon substrates were

prepared using the close-packed structures formed by monodisperse polystyrene beads.84, 85

Polymer surfaces can also be modified by plasma treatment, for example, to improve wetting

properties and to enhance the adhesion of plasma-deposited coatings.86

Plasma etching technique is described in Chapter 2 to remove carbon surface layers from

mica tracked etched membrane that was previously exposed to CVD. This was necessary to

expose the underlying mica surface for dissolution with HF so that the carbon nanotube can be

revealed.

Asymmetric Diffusion

In some systems, diffusive transport through membranes may in some circumstances pass

more readily in one direction than the other. This phenomenon is known as asymmetric

diffusion, and is known in the context of transport across membranes,87-8 and in the context of

osmosis.90, 91 The above asymmetric phenomena is explained either by the binding of particles at

intra-pore sites or other electrostatic interaction with the pore surface. In this research, we show

that asymmetric diffusion can occur with no binding or electrostatic interaction of particle with

the pore surface. It is demonstrated in Chapter 5 that asymmetric diffusion can take place by

purely geometric constraints. Here, conical nanopores, with tip diameter comparable to that of

the diffusion molecule, are the used for asymmetric diffusion of neutral molecules.

Dissertation Overview

The goal of this research is to develop conical nanopores and nanotube (single and

multipore) in polymers and mica membranes by the track-etch method and extension of this

technology for sensing and other applications, and to investigate the transport properties of these

conical pores. The previous part of Chapter 1 has reviewed background information for this

dissertation including the ion track-etch method, membrane based template synthesis, electroless

metal deposition, chemical vapor deposition, resistive-pulse sensing, plasma based etching and

asymmetric diffusion.

In Chapter 2, an extension of the track-etch method to make conical pores in a reproducible

fashion is demonstrated. We have shown here that we can not only reproducibly prepare track-

etched based conical nanopore sensor elements, but that we can predict from the experimental

parameters used during the second etch, what the diameter of the all-important nanopore tip will

be. For these reasons, we believe that the track-etch method will prove to be the technology of

choice for taking artifieial-nanopore resistive-pulse sensors from the bench top to the practical

prototype-device stage of the R&D effort.

Chapter 3 describes a method to make asymmetric pores in tracked muscovite mica films

using an etch-refi11-etch approach. Tracks in the fi1ms were initially etched away with

hydrofluoric acid to form nanoporous membranes. These nanopores were then refi11ed with

silver nanowires or "metal tracks" using an electroless plating method. One face of the

membrane was then exposed to a solution of hydrofluoric acid and nitric acid, which etched the

bulk material and the nanowires respectively, at two different rates. By controlling the

concentration ratio of hydrofluoric acid to nitric acid, tapered pores with diamond shaped cross-

section were obtained. Replicas of the asymmetric pores were accomplished by carbon vapor

deposition, and scanning electron microscopy was used to give evidence of the resulting

nanotubes. In this study, excellent control over tip size and cone angle was demonstrated.

In Chapter 4, electroless gold plating properties on the surface and pore walls of track-

etched Kapton polyimide nanoporous membranes were studied. Scanning electron microscopy

(SEM), atomic force microscopy (AFM), and ion current measurements were used to

characterize the surfaces and pore dimensions of the membrane. Nanoporous Kapton polyimide

membranes were electroless gold plated over different times and the pore diameter characterized

using SEM. Ion current measurements were used to measure the diameter of very small pores.

AFM images show that after electroless gold plating, the gold surface layers are smooth

compared to a similarly structured polycarbonate membrane. Electroless plating the membranes

for 12 hours produced Au wires in the pores. Removing the membrane by oxygen plasma

revealed that the plating in the walls is also relatively smooth.

Finally, Chapter 5 describes asymmetric diffusional transport of neutral molecular species

through nanoporous polymer membranes. Membranes containing conical nanopores from

polyimide (Kapton HN Dupont) foils were prepared by the track-etching technique, based on

irradiation of the polymer with swift heavy ions and subsequent etching of the latent tracks. The

transport properties (e.g, flux) of these membranes were investigated using UV-Vis

spectroscopy. Transport experiments were performed with bare polymer membranes without

any modification. We report the preferential direction of the flow of molecules through conical

nanopores and spatial or concentration dependence of diffusion coefficient for a particular flow

direction.

Swilt heavy tom

''L'
r:';$~5~ ;~ ,Latant lon tackg I, r ~I ~.'~'. .~ Figure 1-1. Swift heavy ions impinge on a dielectric material creating damaged ion tracks. \0bitfion Etlds:Int ; n.-arm r. Etchi-\to a solutionn Elds~ant Figure 1-2. After irradiation, the materials are subject to chemical etching which preferentially removes the latent ion track. DIctric material r T - lon track Figure 1-3. Etched pore geometry in a homogeneous isotropic medium to a first approximation, showing track etch rate, Fr, and bulk at rate, Va~ Figure 1-4. Scanning electron micrographs of Au nanocones replicas of PET conical pores. Increasing fraction of ethanol in the etch solution results in greater cone angle (A- C).[Adapted from Scopece, P.; Baker, L. A.; Ugo, P.; Martin, C. R. Nanotechnology 2006, 1 7, 3951-3956.] V Figure 1-5. Scanning electron micrographs of etched particle tracks in single-crystal mica Figure 1-6. Scanning electron micrographs of a porous polycarbonate, alumina and mica membranes used for template synthesis (A-C, respectively). Figure 1-7. Schematic diagram of Au electroless plating procedure ,>SnLI Ag ,Sn2 7,SXn4+ Ag SAg '>iSn 'Ag /) Ag SnCl Au `X4+ Au 'Sn *A Au + Ag* Au' Formaldhdeyd 4 OC Top View Membrane Face Pore Gold surface film Electroless Plating O Gold Plating Au nanotubes lining SSide View 1 the pores Electroless Gold Plating Membrane Pores Figure 1-8. Schematic illustration of Au nanotubes obtained from electroless gold deposition Tim Time$ nlyewt ngtvecag
Figre1-. chmaicilusra ionofrsitv-usse in

Vestibule/cavity\

soA

IlrliM

Constriction***
Lipid bilayer 11

Transmembrane
il barrel

~- 20A

Figure 1-10. Essential features of the staphylococcal a -hemolysin pore shown in a cross-
section based on the crystal structure.[Adapted from Bayley, H.; Martin, C. R.
Chemical Reviews (Washington, D. C.) 2000, 100, 2575-2594.]

100 A

PET membrane

12 prm

Base
opening
(520 nm)

TIp
Opening
(10O to 60 nm)

Conical nanopore

Figure 1-11. Schematic of a conical nanopore sensor element showing the base diameter and
range of tip diameters used in these studies (drawing to scale)

CHAPTER 2
A1VIETHOD FOR REPRODUCIBLY PREPARING SYNTHETIC NANOPORES FOR
RESISTIVE-PULSE BIOSENSORS

Introduction

Resistive-pulse60 SCHSOTS for molecular and macromolecules analytes60-77 USe a nanopore in

a synthetic or biological membrane as the sensor element. This method, which when applied to

such analytes is sometimes called stochastic sensing,60-69 entails mounting the membrane

containing the nanopore between two electrolyte solutions, applying a transmembrane potential

difference, and measuring the resulting ion current flowing through the electrolyte-filled

nanopore. In simplest terms, when the analyte enters and translocates the nanopore, it transiently

blocks the ion current, resulting in a downward current pulse. The frequency of such analyte-

induced current pulses is proportional to the concentration of the analyte, and the identity of the

analyte is encoded in the magnitude and duration of the current pulse.60-77

The maj ority of such resistive-pulse biosensing data has been obtained using a biological

nanopore, a-hemolysin (a-HL), embedded in a supported lipid-bilayer membrane as the sensor

element.60-6 A key advantage of this biological-nanopore sensor element is that it can be

reproducibly prepared from the commercially available a-HL protein. This is of great

importance if practical, real-world, sensing devices are ultimately to be derived from this

technology.' There is, however, a key impediment to developing practical sensors based on the

biological nanopore. This problem concerns the fragility of the supported bilayer membrane that

houses the nanopore. Such membranes typically survive for periods of only hours before

rupture, much too short of a time to make a practical sensing device.69

One approach for solving this problem is to replace the biological nanopore, and bilayer

membrane, with an artificial nanopore embedded in a mechanically and chemically robust

synthetic membrane.69-77, 79, 80 Such artificial nanopores are often prepared by microlithographic

methods, using for example a focused ion" or electron72 beam to bore the nanopore into a silicon

or Si3N3 membrane. We and others are exploring an alternative technology, called the track-etch

method,16, 19, 81, 82 for preparing nanopores for resistive-pulse sensorS.70, 75-77, 79, 80 Analytes

detected with prototype track-etched nanopore sensors include small molecules,70 DNA,75, 76

proteins" and nanoparticles.79 Furthermore, there are older reports of developing virus sensors

based on track-etched nanopores.so

This field of artificial-nanopore resistive-pulse sensing is currently in its infancy. A key

question that must be addressed before practical sensors can be developed is can the nanopore

sensor element be prepared reproducibly, as the biological nanopore can? We address this

critically important issue here.

The sensor elements we evaluated were conically shaped nanoporeS70, 75-77, 79, 81 prepared

by the track-etch method in polyethylene terephthalate (PET) membranes. Such conical

nanopores have two openings the large-diameter (base) opening at one face of the membrane

and the small-diameter (tip) opening at the opposite face (Figure 2-la). We have found that the

diameters of both of these opening can be controlled with good reproducibility using a new two-

step pore-etching procedure. Furthermore, we have developed a simple mathematical model that

allows us to predict the diameter of the tip opening from the parameters used during pore

etching. Good agreement was obtained between the predicted and experimentally measured tip

diameters.

Experimental

First Etch Step

The tracked PET membrane (from GSI) was mounted in cell shown in Figure 2-1b, and the

etch solution (9 M NaOH) was placed in one half cell and the stop solution (lM formic acid plus

1 M KC1) in the other. A platinum wire electrode was placed in each solution and a Keithley

6487 was used to apply a transmembrane potential difference of IV during etching, with polarity

such that the anode was in the etch solution. The electrochemical reactions occurring at the

anode and cathode are discussed in the Supplementary Materials. Etching was terminated after

two hours by replacing the contents of the etch half-cell with stop-etch solution. The membrane

was then rinsed with purified water (Barnstead D4641, E-pure filters).

Determination of the Diameter of the Base Opening

Multi-track membranes (106 CA-12) were subj ected to the same first etch step as used for the

single-track membranes, and the base openings were imaged via FESEM (JEOL JSM-633 5F).

The average base diameter obtained (520+45 nm) is associated with measurements of 50 pores in

five different multi-track membrane samples.

Electrochemical Measurement of the Tip Diameter

The membrane was mounted in the cell, and the half cells were filled with 100 mM

phosphate-buffered saline, pH = 7.0 that was also 1 M in KC1. The specific conductivity of this

solution was measured using a conductometer (YSI 3200) at 250 C; a conductivity of 0. 107 S

cm-l was obtained. A Ag/AgCl electrode immersed into each solution was used in conjunction

with the Kiethly 6487 to obtain the current-voltage curve for the nanopore (Figure 2-2). We

have validated this electrochemical method by comparing diameters obtained via this method

with diameters for the same pores measured by electron microscopy.92, 93

Second Etch Step

The etch solution in this case was 1 M NaOH, and its conductivity was measured at 0. 160

S cml (250 C). The membrane was mounted in the cell, and the half cells were filled with this

etch solution. A platinum electrode was immersed into each half-cell solution, and the Keithley

6487 was used to apply a transmembrane potential difference of IV and measure the nanopore

ion current as a function of time. Etching was terminated at specified current values (Figure 2-3,

Figure 2-4) by replacing the etch solution in both half cells with the stop-etch solution. The

membrane remained in the stop etch for at least 30 min and was then rinsed with purified water.

Bovine Serum Albumin (BSA) Resistive-Pulse Sensing

Conical nanopores sensors having two different tip diameters, 58 nm and 44 nm, were

prepared; the base diameter for both sensors was 520 nm. After etching the pore walls were

coated with gold nanotubes using the electroless plating method described previously.28 After

electroless plating the tip diameters of the resulting gold nanotubes were measured using the

electrochemical method discussed above; tip diameters of 32 nm and 23 nm were obtained. The

Au-coated nanopore walls were then functionalized with a thiolated PEG (MW 5000 Da, Nektar

Therapeutics) to prevent non-specific protein adsorption.94 This was accomplished by

immersing the nanotube-containing membrane in a 0. 1 mM solution of the PEG at 4 oC for ~15

hours. The membrane was then immersed in purified water for 1.5 hours to remove any unbound

PEG. The tip diameters were then re-measured; values of 27 nm and 17 nm were obtained.

The BSA (Sigma) was dissolved in 10 mM phosphate-buffered saline that was also 100

mM in KCl (pH 7.4). The concentration of BSA was 100 nM, and the BSA solution was placed

on the tip side of the membrane. Buffer was placed on the base side and a transmembrane

potential of 1000 mV was used to drive the protein through the nanopore (tip to base) by

electrophoresis.

Results and Discussion

Conical Shaped Nanopores are Ideal Resistive-Pulse Sensor Elements

Resistive-pulse sensing entails mounting the nanopore membrane between the two halves

of an electrolyte-filled cell (Figure 2-1 (b)) and passing an ionic current through the electrolyte-

flooded nanopore. In a conically shaped nanopore the voltage drop caused by this ion current is

focused to the electrolyte solution in the tip opening of the nanopore.79 Indeed, the field strength

in the solution within the nanopore tip can be greater than 106 V m- when the total voltage drop

across the nanopore membrane is only 1 V.79 A consequence of this focusing effect is that the

nanopore ion current is extremely sensitive to analyte species present in the nanopore tip. That

is, there is an analyte "sensing zone" just inside the tip, which makes conically shaped nanopores

ideally suited for the resistive-pulse sensing application. This has been demonstrated with

prototype conical-nanopore sensors for analyte species ranging in size from small molecules, to

proteins, to nanoparticleS.70, 75-77, 79

The Core Technology: The Track-Etch Method

The track-etch method has been practiced commercially for decades to make polymeric

nanopore membranes for filtration applications.16, 19, 81, 82 It entails passing high energy particles

through the membrane, to create damage tracks, followed by chemical etching to convert these

damage tracks into pores. While the commercial process yields membranes that contain high

pore densities, a method for preparing single-damage-track membranes was developed at the

Gellsellschaft fur Schwerionenforschung (GSI).19 We purchased such single-track PET

membranes from GSI for these studies. This is an important point 0I ithr regard to the overall

sensor-fabrication technology the key precursor material, the tracked membrane, can be

obtained commercially.

The First Etch Step

Both etch steps used the cell shown in Figure 2-1b. Step 1 entails placing a solution that

etches the damage track on one side of the membrane and a solution that neutralizes this etchant

on the other side.8 For PET the etchant is NaOH, and the etch-stop is formic acid. This yields a

conically shaped pore (Figure 2-la) with the base opening facing the etch solution and the tip

opening facing the etch-stop solution. To determine when the etchant has broken through to the

etch-stop solution, and a contiguous pore has been obtained, an electrode is placed in each

solution, and a potential difference is applied across the membrane. Before breakthrough, the

transmembrane ion current is zero, and breakthrough is signaled by a sudden rise in the current.8

We previously showed that the diameter of the base opening could be controlled by

varying the potential applied across the membrane during this first etch step.92 An applied

transmembrane potential of 1.0 V was used in the first etch step for all of the nanopores

investigated here. To obtain a measure of the reproducibility of the base diameter obtained after

the first etch step, we subj ected multi-track membranes (106 tracks cm-2) to the same first etch as

used for the single-track membranes, and imaged the base openings using field emission

scanning electron microscopy (FESEM, Figure 2-4). We used multi-track membranes for this

study because it is difficult to locate the base opening in electron micrographs of a single-

nanopore membrane. We have previously shown that the pore diameter obtained for tracked-

etch membranes is ind pendent of track densi y.95 A base diameter of 520+45 nm was obtained,

indicating good reproducibility in base diameter after the first etch step.

However, the tip diameter varied between 1 and 7 nm and could not be reproduced from

etch to etch. We reasoned that this is an inherent feature of this "anisotropic" etch process. This

is because the etch and etch-stop solutions are mixing (and neutralizing each other) in the

nascent tip, which makes it difficult to control the etch rate in this critically important region of

the nanopore.

Measuring the Diameter of the Tip Opening after the First Etch

Because the tips after the first etch are so small, they are very difficult to find and image

via electron microscopy. Therefore, an electrochemical method described previously70, 76, 81, 92

was used to measure the diameter of the tip. This entailed mounting the membrane sample in the

cell (Figure 2-1b), filling both half cells with an electrolyte solution of known ionic conductivity,

and obtaining a current-voltage (I-V) curve associated with ion-transport through the nanopore

(Figure 2-2).

The experimental slope of this linear I-V curve is the ionic conductance, G1, (in Siemens,

S) of the nanopore, which is given bys

G1 = (GKal x db dt) / 4 L (2-1)

where GKal is the experimentally measured conductivity of the KCl-based electrolyte used (S cm-

1), L is the length of the nanopore (membrane thickness), db is the experimentally measured

diameter of the base opening, and dt is the diameter of the tip opening. Because all of the other

parameters in Equation 1 are known, dt can be calculated.

The Second Etch Step

In the second step NaOH etch solution is placed on both sides of the membrane. Again, a

transmembrane potential is applied, and the ion current flowing through the nanopore is

measured as a function of time during this etch. Our key innovation is that the second etch is

stopped at a prescribed value of this nanopore ion current rather than at some prescribed time

after starting the second etch (e.g., Figure 2-3). We adopted this approach because of the

variability in tip diameter obtained after the first etch step. The consequence of this variability is

that if we stopped the second etch at a prescribed time, we would obtain a corresponding

variability in the tip diameters obtained after the second etch step. In contrast, as we will see

below, there is an exact mathematically relationship between the ion current flowing through the

nanopore when the second etch is stopped (If) and the diameter of tip opening.

To prove that the pores obtained after the second etch truly are conically shaped, we used

an electroless plating method28 to deposit correspondingly conically shaped gold nanotubes

within the pores of the multipore membranes described above. The PET membrane was then

dissolved and the conical nanopores collected by filtration and imaged by FESEM.l These

images show that a nearly ideal conically shaped pore is obtained (Figure 2-5).

Measuring the Diameter of the Tip Opening after the Second Etch

The same electrochemical method was used, but the mathematics is slightly different. We

define the diameters of the base and tip openings after the first etch step as db1 and dtl and the

diameters after the second etch as dbf and dtf. These diameters are related via

dbf = (db1 + Ax) (2-2)

dtf = (dr y + Ax) (2-3)

where Ax is the change in diameter during the second etch. Substituting Equations 2-2 and 3-3

for db and dt in Equation 2-1 yields

G2 (Kal n (db1 + Ax)(dr y + Ax)) / 4 L (2-4)

where G2 is the slope of the current-voltage curve used to determine the tip diameter after the

second etch, db1 and dol are the experimentally determined base and tip diameters, respectively,

after the first etch, GKal is the experimentally measured conductivity of the electrolyte used (100

mM phosphate-buffered saline, pH 7, that was also 1 M in KCl; a = 0.107 S cm- ), L is the

membrane thickness, and Ax is the change in diameter between the first and second etch steps.

We define the parameter M as

M = KCl (2-5)
4L

which allows us to write, after some simple algebraic manipulation,

u2= dbldtl + (db1 + dtl)Ax + Ax2 (2-6)

Substituting (dtf dtl) for Ax (where dtf is the diameter of the tip opening after the second etch),

and again applying some simple algebraic manipulations, yield

d- + (db1 + d,,)dv = 0 (2-7)

This is a quadratic equation in dtf for which the solution is

-(b d,,)+ (db1 ,-d,)z)+4G', /M
d,. = (2-8)

Note that the quadratic formula has two roots; i. e., there should be a + instead of a + between the

two terms in the numerator of Equation 2-8. However, the root that results when subtraction is

used yields a negative value for the tip diameter.

Because all of the parameters on the RHS of Equation 2-8 are known, dte can be calculated.

Furthermore, because the base diameter at the start of the second etch is large (520 nm), the

change in base diameter during the second etch (Ax in Equation 2-2) is negligibly small for all

but the very largest tip investigated here (60 nm, Figure 2-6).

Reproducibly Varying the Tip Diameter

Figure 2-6 shows a plot of nanopore tip diameter, measured after the second etch step (Eq.

2-5), vs. the nanopore ion current at which this etch was stopped (If). We see that the tip

diameter can be reproducibly varied over the range from 10 to 60 nm (data points in Figure 2-6).

This is important because this is exactly the range in tip diameters we used in our prototype

protein," DNA,76 and nanoparticles79 Sensors. The lower limit (10 nm) is determined by the tip

diameter obtained after the first etch which, again, was in the range of 1 to 7 nm. However, we

have shown that the walls of such nanopores can be lined with gold nanotubes," and that the

diameter of these tubes can be controlled at will down to 1 nm.96 Hence, if tip diameters smaller

than 10 nm are needed, a pore with a 10 nm tip can be gold plated to reduce the tip to any desired

value. Furthermore, tips larger than the 60 nm maximum shown in Figure 2-6 can be easily

prepared by simply stopping the second etch at larger values of If.

The Mathematical Model

We begin by defining a new conductance, Getch, which is the ion current at which the

second etch is stopped (If) divided by the transmembrane potential applied during the second

etch (Eap). With this definition, Equation 4 can be rewritten as

If = Eap (etch n: (db1 + Ax)(dtl + Ax)) / 4 L (2-9)

where oetch is the experimentally measured conductivity of the NaOH solution used in the second

etch. Equation 2-9 is again a quadratic Equation in dtf for which the solution is

-(db1 -d,,)f + j(db1- d,,) + 4I /
der (2-10)

where K = Eap oetch n/4L (see Derivation of Equation 2-5, below). Equation 2-10 allows us to

calculate the value of the tip diameter after the second etch step (dtf) for any value of If at which

the second etch was stopped.

We noted above that the base diameter before and after the second etch is essentially the

same for all but the largest tip in Figure 2-6. This allows us to simplify Equation 2-9 to

If = Eap oetch n: db1 dtf / 4 L (2-11)

which can be rearranged to

dtf = If 4 L/ (Eap Getch n: dbl ) (2-12)

This obviously provides a much simpler relationship between dtf and If.

Plots of dtf vs. If calculated using the simplified equation (Equation 2-12) and the exact

equation (Equation 2-10) are shown as the two solid lines in Figure 2-6. The tip diameters

calculated by these two equations are identical for tips below ~20 nm. Furthermore, the

agreement between the experimentally measured (Equation 2-8) and theoretically calculated

(Equation 2-10) tip diameters is good, especially considering that there are no adjustable

parameters in the calculations. For example, at If = 20 nA the experimental and calculated tip

diameters differ by less than 10%, and at 40 nA they are identical.

The calculated tip-diameter values are, in general, slightly smaller than the experimental

values. This results from an interesting feature of the transport properties of conical nanopores -

if there is charge on the pore wall and if the tip opening is small, such nanopores act as ion

current rectifiers.97 The consequence of this rectifieation phenomenon is that the ionic

conductivity of an electrolyte solution within the tip of the nanopore can be lower than the value

measured for a bulk sample of the same electrolyte. Since we used the bulk-solution

conductivity in our calculations, the calculated values are in general low. Excellent agreement is

obtained between the experimental and calculated diameters for the largest tip (Figure 2-6)

because large-tip pores do not rectify the ion current.98 We believe that procedures to obviate the

small disagreement between the experimental and calculated tip diameters can be developed, and

we are currently pursuing this issue.

To illustrate the importance of controlling the tip diameter in resistive-pulse sensing, we

obtained current-pulse data for a prototype protein analyte, bovine serum albumin (BSA), with

nanopore sensors having two different tip diameters. The sensors in this case were conical PET

nanopores that had been lined with gold nanotubeS28, 77 and then coated with a poly(ethylene

glycol thiol) (PEG) to prevent nonspecific protein adsorption.94 The tip diameters, 17 nm and 27

nm, were measured after PEG functionalization.

Figure 2-7 shows current-pulse data obtained for BSA with these two different sensors.

The current-pulse signature can be defined by the average duration and magnitude (AI) of the

current pulses. The magnitude of the current pulse is important because if AI is not larger than

the peak-to-peak noise in the background current, the current pulse will be undetectable. We

found that AI is larger for the nanopore sensor with the smaller tip opening (AI = 80+20 pA) than

for the sensor with the larger tip opening (AI = 35+9 pA). This is because the roughly 4 nm x 4

nm x 14 nm94 BSA molecule more effectively blocks the ion current as it translocates the

smaller, 17 nm, tip.

Electrochemical Details

Pt electrodes were used to apply the transmembrane potential difference in both of the etch

steps, and the applied potential was 1.0 V in both cases. The half reaction occurring at the Pt

cathode was the reduction of the dissolved 02 in the solution.

02 + 4H+ + 4 e- + 2H20 (2-13)

The low (nA-level) currents, and the fact that the solutions were exposed to air during

etching, insured that the Oz WAS not depleted. The half reaction occurring at the Pt anode was the

reverse of Equation 2-15.

Conclusions

In his review of nanowire-based chemical and biosensors, Lieber stresses the importance

of being able to reproducibly prepare the nanowire sensing element.'" The same is true for

artificial nanopores to be used as resistive-pulse sensor elements. We have shown here that we

can not only reproducibly prepare track-etched based conical nanopore sensor elements, but that

we can predict from the experimental parameters used during the second etch, what the diameter

of the all-important nanopore tip will be. For these reasons, we believe that the track-etch

method will prove to be the technology of choice for taking artificial-nanopore resistive-pulse

sensors from the bench top to the practical prototype-device stage of the R&D effort.

PET membrane

12 pm

Base

(520 nm)

Conical
nanopore

Electrodes Pressure Clamping
J plate screw

IC1.0 cm3.cm

Solution
chamber\

/ 1
Membrane Poly(chlorotrifluoroethylene) cell

Figure 2-1. Schematic of a conical nanopore sensor element and etching cell. (A) The base
diameter and range of tip diameters used in these studies (drawing not to scale). (B)
Cell used to do the etching and to make all electrochemical measurements.

Tip
opening
(10 to 60 nm)

Q

-.2-0.1 0D 0.1 0.2

o -1-

2 -2-
Applied transmembrane potential (V)

Figure 2-2. A typical current-voltage curve used to measure the tip diameter of the conical
nanopore.

40-

S35-

F 20-

E 15-

o 10-

z5,

0 20 40 60 80 100 120 140

Etch time (minutes)

Figure 2-3. Current-time transients obtained during the second etch step for three membranes
that were subjected to the same first etch. The second etch was stopped in each case
when a final nanopore ion current (If) of 40 nA was obtained.

Figure 2-4. Scanning electron micrograph of the base openings of two conical nanopores in a
multi-track PET membrane that had been etched (first etch step) as per the single-
track membranes used in these studies.

Figure 2-5. Scanning electron micrograph of conical gold nanotubes deposited in a conical
nanopore membrane.

80

70

50

f 40~

I I I

0 10 20 30 40
Final nanopore ion current during second etch, I,(nA)

Figure 2-6. Plot of tip diameter measured after the second etch step vs. the final nanopore ion
current (If) at which the second etch was stopped. The points are the experimentally
measured tip diameters. The error bars are measurements on three different
membrane samples prepared identically. The solid curves were calculated using the
simplified equation (Equation 2-12, red curve) and the exact equation (Equation 2-10,
blue curve).

.r"r" nl- nr rr71-~nrr~lrl-~r~n 'Imn~nr~rrmrr

.~LLL 1.

100 s

20 pA [

1 1 I

1 (

Tll"r~ n I

Figure 2-7. Current-pulse data obtained for a prototype protein analyte, bovine serum albumin
(BSA) using PEG-modified conical nanotube sensors. (A) Tip diameters of 17 nm
and (B) Tip diameters of 27 nm. BSA concentration = 100 nM. Applied
transmembrane = 1000 mV.

CHAPTER 3
ETCH-FILL-ETCH METHOD FOR PREPARING TAPERED PORES IN ION TRACKED
MICA FILMS

Introduction

The Martin research group and others have been investigating a general method for

preparing nanomaterials known as template synthesis.23, 96 This method entails the synthesis or

deposition of a desired material within the pores of a nanopore membrane that serves as a

template. These template membranes contain monodisperse pores that are typically cylindrical

in geometry, and the pore diameter can be varied at will from tens of nanometers to tens of

microns. Since typical pore geometries are cylindrical, correspondingly cylindrical

nanostructures are usually synthesized via the template method; depending on the membrane and

synthetic method used, these may be solid nanowires or hollow nanotubes. 99

Recently, our research team and others have become interested in nanopores that have a

conical pore shape and the correspondingly conical nanostructures synthesized via the template

method within these pores. A number of applications can potentially benefit from conical pore

geometry. For instance, it has been shown that such conically shaped nanopores can be used as

the sensing element for new types of small molecule, 70 DNA, 75, 98, 100, 101 prOtein 77and

particle79 Sensors. Conically shaped gold nanotubes deposited within such pores can also act as

mimics of voltage gated ion channelS. 102 Membranes used for separations might also benefit

from a highly asymmetric pore structure. Finally, in addition to sensing and separations

platforms, conical nanostructures prepared by more conventional methods have been proposed

for use as cathodes in field-emission displayS. 103 To date, the vast majority of conical pores

have been fabricated using tracked polymers. 81, 89, 93, 95, 97, 104, 105 However, we are interested in

exploring materials other than polymers to use as conical pore templates because their different

properties may prove potentially superior for certain applications, and other applications might

be realized that are not possible with polymers. One such material of interest is tracked

muscovite mica.

Because mica is an inorganic crystalline material, it possesses some properties not present

in polymers that might make it superior in some aspects for certain applications. For example,

the surface of mica is molecularly flat, 106 and is a good candidate for platform for AFM imaging

of DNA, 106 and support layer for lipid bilayers. 107 Also, mica is very chemically resistant and

has high thermal and mechanical stability. 22, 108 Mica conical pores might prove much more

stable for resistive pulse sensors than polymers. Furthermore, these properties make it possible

for template synthesis of materials that require high temperatures. Additionally, nanostructures

that demand special geometry that is difficult to obtain by conventional methods may be

realized. For instance, tapered-shape carbon structure can provide mechanical stability yet

provide very sharp tips, which may be useful for enhanced electron field emission. 109

We and other have shown that the well-known track-etch method 16 can be used as a

starting place for preparing such conical nanopores. 81, 89, 93, 95, 97, 104, 105 This method entails

bombarding a thin film (5-20 Cpm) film of the material with a collimated beam of high-energy

particles to create parallel damaged tracks through the film. To make cylindrical pores, the

tracked membrane is simply immersed into a chemical etch bath, where preferential etching

along the damaged track converts each track into a cylindrical pore. To make conical pores, the

tracked membrane is mounted in an etching cell with an etch solution on one side of the tracked

membrane and a stop solution on the other side.8 This is shown schematically in Figure 3-1.

Since the damaged track is etched at a longer duration time and faster at the face of the

membrane exposed to the etch solution than at the face of the membrane exposed to the stop

solution, conically shaped nanopores are obtained.

While this stop-etch approach has been successful in making conical nanopores in a variety

of polymer materials, the method fails for the preparation of conical pores in mica membranes.

The key impediment is due to the ratio of the track to bulk etching rate. The track etch rate is

much faster than the bulk etch rate in mica (about 3000 time faster), los so that the etch solution

traverses the entire membrane before any significant bulk etching takes place. This means that

all parts of the membrane start etching almost at the same time in an isotropic fashion giving rise

to uniform pores instead of asymmetrical pore shape. One approach to solve this problem is to

replace the tracks of the mica films with a material that is more controllably etched, for instance

metal nanowires. In this study, we developed a method to independently control the solution

etch rate traversing the membrane, and also the etch rate of the surrounding bulk material, to give

asymmetric pores. We prove this by using carbon vapor deposition to replicate the pores,

dissolving the membrane to expose the tapered tubes, and imaging the nanotubes using scanning

electron microscopy. The results of these investigations are reported here.

Experimental

Materials

Muscovite mica wafers (1.181X 0.0004 inches) were purchased from Spruce Pine Co.

USA. Then these bare wafers were irradiated by swift heavy U25+ ions of 2.2GeV kinetic energy

with fluence of 104 tol07 cm-12 (GSI Darmstadt, Germany), which produced damage tracks

through the mica membranes. Hydrofluoric acid (HF, 48~51% from ACROS), for etching and

dissolving mica membranes, was used as received. Anhydrous tin (II) chloride 98% (Aldrich)

and hydrochloric acid (ACROS) were used as received to sensitize the mica membranes for

electroless plating. Ammonium hydroxide (Fisher), silver nitrate (Mallinckrodt), potassium

sodium tartrate tetrahydrade (Aldrich), and magnesium sulfate (Fisher) were used as received to

prepare silver plating solution. Ethylene (30% balanced with Helium, from Praxair) was used as

the CVD carbon precursor gas. Purified water was prepared by passing house-distilled water

through a Millipore Milli-Q water purification system.

Initial Etching of Mica Tracks to Prepare Very Small Pores

Mica wafers, containing damaged tracks, were exposed to low concentration HF solution

on both faces to create pores of about 10nm in diameter. The wafers were sandwiched between

two half cells of a conductivity cell (Figure 3-2.) and 2% HF at 250 C were placed in each half

cell for a period of 10 minutes. The etch process was terminated by quickly removing the etch

solution and replacing it with water for 2 minutes. Fresh water was replaced several times for

two minute intervals. Finally water was allowed to sit in the cell for another two hours.

Preparation of Tin Sensitizing Solution

Tin (II) chloride was used to sensitize the wafers so that electroless plating of silver can

take place on the surfaces. Tin (II) chloride crystals (0.5 g) were placed in 100 mL water and

stirred to give a cloudy appearance. 2 mL of 10 % hydrochloric acid was added using a pipette

causing the mixture to become a clear tin (II) chloride sensitizing solution.

Preparation of Silver Plating Solution

Filling the pores in the wafer with silver wire was done to make a more controllably etched

material. A two part electroless silver plating solution was made, where solution A contains the

silver ions and solution B contains a reducing agent. Solution A was made by dissolving 45.4 g

of silver nitrate to 450 mL of water and then adding ammonium hydroxide drop wise using a

pipette until the solution goes from clear to dark brown and returns to clear. Solution B was

made by dissolving 159 g of potassium sodium tartrate tetrahydrate and 1 1.4 g magnesium

sulfate to 364 mL of water. Both solutions were stored away from light.

Filling the Pores with Silver Wires

The purpose of silver plating the nanopores in the mica membrane was to control the rate

at which etch solution traverses the membrane relative to the lateral or bulk etch rate. First, the

intended stop side of the membrane was exposed to tin by filling one half cell with the solution.

The other half cell was left empty so that one face of the membrane was exposed to air. Tin

solution remains in the cell for 45 minutes giving it time to properly wet and sensitize the inner

walls of the pores. The membrane was then rinsed with water several times then was left to sit in

water for at least one hour. After removal of water from the both half cells, the membrane was

now ready for silver electroless plating. A dilute solution A (0.5 mL Solution A in 45 mL water)

was cooled to 40C and then 0.5 mL of solution B was added. This mixture was then placed in the

half cell that was not exposed to tin solution (side to be etched). The conductivity cell containing

the membrane was placed in a refrigerator, which was set at a temperature of 40C, for one hour.

Following electroless plating of silver, the remaining solution in the cell was removed and the

cell was thoroughly rinsed with water. Figure 3-3 shows the schematic outline for this etch-fill-

etch method.

Etching Silver Filled Mica to form Tapered Pores

The silver wire-containing mica membrane was exposed to HF/HNO3 etch solutions on

one face. This etch solution of HF (variable concentration) and HNO3 (10%) etched mica and

silver wire respectively at two different rates. After 3 hours the etch solution was removed and

rinsed with water several times. The membrane was then left to sit in 10% HNO3 for at least 3

hours to get rid of any residual silver. Finally, the membrane was rinsed in water several times

followed by soaking for at least 3 hours.

Making Replicas of the Tapered Pores

Tapered carbon tubes were obtained using a chemical vapor deposition (CVD) method

described in detail previously. 38, 110, 111 A piece of a porous mica membrane (preparation

methods were described in the previous section) was placed vertically into a quartz tube

(diameter: 4.5 cm, length: 48 cm). This tube was then inserted into a high-temperature tube

furnace (Thermolyne 21100) and the furnace was heated to 670oC under Ar flow. Once the

temperature stabilized, the Ar gas was replaced with an ethylene gas (20 sccm), which thermally

decomposed into carbon on the inner-wall and both faces of the mica template. After a desired

deposition time, the heating was terminated, the ethylene gas was replaced by Ar flow, and the

furnace was cooled down to room temperature. Unlike our previous CVD procedure with

alumina templates, 11 heat pretreatment of the template in this experiment was skipped since the

mica membranes we used can withstand temperatures above 900oC without any physical

deformation. The yielded carbon thickness can be controlled by varying the duration of

deposition.

Preparation of the Carbon Tube Replicas for SEM Imaging

First, CVD carbon/mica membrane was put into 48~51% HF solution for 16 hr to dissolve

away the mica template. Next, HF was then removed by pipette, leaving the liberated carbon

nanoboxes (connected together by the carbon surface film) which were rinsed with methanol and

suspended in methanol. The next preparation procedure entailed the removal of the carbon

surface layer on one face of the CVD-treated mica membrane to expose the carbon tubes. This

was accomplished by using an oxygen plasma etch procedure. A 1 cm x 1 cm hole was pre-

made in one piece of aluminum foil. This hole defines the area of the membrane that is exposed

to the oxygen plasma. The entire assembly, with the hole-containing Al foil facing up, was then

placed in the center of vacuum chamber of a plasma reactive-ion etching system (Samco, model

RIE-1C). The following etch conditions were used: power = 100 W, Oz preSsure = 300 Pa, Oz

flow rate = 30 sccm. After etching away the carbon surface film (determined by measuring the

conductivity of the membrane surface), the membrane was immersed into~- 48-5 1 wt % HF

solution to dissolve the mica template. This step does not result in free carbon tubes because

they are held together by carbon surface film that was not exposed to oxygen plasma. Finally,

the sample was rinsed with distilled water and air dried overnight. Sample imaging was

conducted using JEOL 6335F field emission scanning electron microscope (FESEM). Prior to

FESEM imaging, all samples were sputtered with Au/Pd using the Desk II Cold Sputter

instrument (Denton Vacuum, LLC). The sputter current = 45 mA, Ar pressure = 75 mTorr,

sputtering time = 60sec. The resulting Au/Pd film was ~ 16 nm.

Results and Discussion

During chemical etching of ion-tracked membranes, the damaged zone of the latent track is

transformed into nanopores. 16 The simplest description of the etching process defines two

parameters: the bulk etch rate (Va) and the track etch rate (VT). VB depends on the material,

etchant composition and temperature. VT depends on additional parameters, such as sensitivity

of the material to tracking, post-irradiation conditions and etching conditions. 14 When the

tracked film is exposed to an etchant on one face as described above, the results give conical

pores. In most ion-tracked materials, without exposure to extreme conditions (like exposure to

high illumination) the VB/VT ratio defines the cone angle that is formed. 14 However, in mica VT

is 3000 times faster than VB, 108 thus producing cone angles (0.020 los that is almost zero, giving

essentially pores with almost identical cross-sections along the membrane thickness. To improve

the cone angle in polymeric materials, the VB/VT ratio is increased by a number of methods

including increasing the etchant concentration, 1os applying a high transmembrane potential, 92

and modifying the etch solution composition. IsHowever, none of the above methods can work

with mica since VT is orders of magnitude greater than Va. 10s To get around this problem, the

etched ion tracks in mica was replaced with silver metal that can be independently etched with

nitric acid to a wide variety of etch rates. Figure 3-4a is an SEM image of a membrane that was

exposed on one face to 20% HF andl0% nitric acid solution at 250 C for 3 hrs. Here, HF and

HNO3 Solutions etch the bulk mica membrane and the metal track respectively. The base side

clearly shows the tapered "cone" shape that resulted from this etch. The decreasing pores size is

made quite evident from the progressive mica layers going down into the cone. The opposite

face of the membrane (Figure 3-4b) shows the tip to be on the order of a magnitude smaller than

the base.

The Martin group has been using conical pores for resistive pulse sensing of molecules.

One important feature that make conical pores ideal for sensing is, that most of the resistance is

focus in a short distant of the tip. As shown by Lee et al,79 the electric field in conical pores is

focused at the tip. The greater the half cone opening angle, the smaller the focus and hence a

sensing zone for molecular translocation of the pore for resistive pulse sensing. Half cone

opening angle depends on the VL/VT rtio, and we changed the concentration ratio of the etchants

to achieve this. Because the concentration ratio of HF (the lateral etchant) to HNO3 (the track

etchant) is greater than used previously, the lateral to track etch rate increases VL/VT TOSulting in

greater half cone opening angle (~ 210). Figure 3-5 is an SEM image of a tapered mica pore that

was etched with a higher percentage of HF solution than previously in Figure 3-4. We

demonstrate here that the cone angle can be controlled at will, because we can independently

control the solution etch rate traversing the membrane, and also the lateral etch rate, to give

asymmetric pores. This implies that the effective pore length (the part of the pore where

resistance is focused), and hence the focus of electric field can be varied in these tapered mica

pores. This is an important feature for resistive-pulse sensing that is likely responsible for pulse

duration.

To indirectly capture the entire geometry of the tapered mica pores in the membrane, a

replica was done using CVD method. Figure 3-6 clearly shows SEM images of carbon tapered

nanotubes replicas of the mica pores. The angles are well defined and surface of the tube

appears rather smooth at the magnifieation shown.

The half cone opening angle was calculated to be ~ 60. This is a relatively large cone

angle when compared to those of polymers that are etched without any promoters (Lane) or

applied high potential (chad). The half cone opening angle can be calculated as

P = arctan((db d,)/2L) (3-1)

where L is the length of the pore, db and dt are the large and small openings of the pore,

respectively. For db >> dt the equation simplifies to:

p = arctan (db /2L) (3 -2)

Figure 3-7 is a low magnifieation image of the carbon replicas of mica tapered pores indicating

that we can indeed reproduce the tapered geometry mica pore uniformly.

Conclusion

This study described a method to make asymmetric pores in tracked muscovite mica films

using an etch-fill-etch approach. Tracks in the films were initially etched away with

hydrofluoric acid to form nanoporous membranes. We demonstrated that by controlling the

concentration ratio of hydrofluoric acid to nitric acid during etching, tapered pores with diamond

shaped cross-section can be obtained. Additionally, we have shown that the cone angle of the

pores can be controlled be changing the concentration ratio of the bulk and metal etch solutions.

Replicas of the asymmetric pores were accomplished by carbon vapor deposition, and scanning

electron microscopy was used to give evidence of the resulting nanotubes. These conical mica

pores make prove more stable for resistive pulse sensing. Because it is so easy to tailor the cone

angle, this might make these mica pores more suitable sensing devises. One potential capability

that might be realized in resistive pulse sensing, is the tuning of the cone angle to control pulse

duration.

Damallged
ion tratck

Etchatnt a a .

Figure 3-1. After irradiation, the materials are subject to chemical etching which preferentially
removes the damaged ion track

Etchi-stop,
solution

Etcha nt Etch-stop,
solution

Electrodes

Solution
chamber

1ii~ ri ~ 1 I

\$1.0 cm 3.5 cm -

I 1
Membrane Poly(chlorotrifluoroethylene) cell

Figure 3-2. Schematic of cell used to do the etching and to make all electrochemical
measurements.

Pressure Clamping
plate screw

Mica membrane

Tracks

Electroless plating
of Silver

HF

Silver

Pore

Figure 3-3. Schematic diagram of etch-fill-etch method

HF/ HNO3

Figure 3-4. Scanning electron micrographs of mica membrane that was exposed on one face to
20% HF andl0% nitric acid solution at 250C for 3 hrs. (A) Side exposed to the
etchant (the base). (B) Side exposed to water (the tip).

Figure 3-5. Scanning electron micrographs of mica membrane that was exposed on one face to
40% HF andl0% nitric acid solution at 250C for 3 hrs. (A) Side exposed to the
etchant (the base). (B) Side exposed to water (the tip).

Figure 3-6. Scanning electron micrograph of carbon tapered nanotube replica of the mica
tapered pore.

Figure 3-7. Low magnification SEM images of carbon tapered nanotube replicas of the mica
tapered pore.

CHAPTER 4
ELECTROLESS AU PLANTING OF TRACK-ETCHED KAPTON POLYIMIDE
NANOPOROUS 1VE1VBRANES

Introduction

Track-etched polymer membranes have found many applications in industry and research

as filtration and separation materials. 6 Chemical modification of the inner walls of these

membranes has made them more selective and sophisticated separation structures. 27, 28, 31 These

chemical modifications have been made possible by first plating the inner walls of the pores of

these membranes with gold, 23 then attaching a desired thiol terminated functional group. 31 For

reproducible results of any analytical measurements done using the pores of these nanoporous

membranes, it is crucial that the pores remain stable and have a well define internal diameter,

particularly where the pore approaches the size of the analyte molecule. One of the problems

faced when using some polymers such as polycarbonate (PC) and poly (ethylene terephthalate)

(PET) membranes, is that very small pores tend to temporarily block during ion transport

measurement. 112 Another disadvantage discovered when using these membranes to measure ion

current, is that possible dangling alkyl groups 113 render the pores from been well define and

thus gives inconsistent results and lots of noise in the current measurement. 112 The polyimide

Kapton on the other hand does not exhibit these disadvantages. 112

Due to its abilities of maintaining excellent physical, electrical and mechanical properties

at both low and high temperature extremes, Kapton is a very attractive polymer for use as a

particle track-etch membrane for application in separation and filtration in industry and research.

16, 114, 115 It will be advantageous therefore, if like polycarbonate and PET, Kapton can similarly

be electrolessly plated with gold to tailor pore diameter and, further, be modified with desired

thiol terminated functional groups for selective separation.

The objective of this study was to determine if the polyimide Kapton, a very chemically

resistant and stable polymer, can be electrolessly gold plated with similar or better quality

compared to PC porous membranes. Additionally, investigate whether plating can be controlled-

that is, tailor pore diameter of the membrane with plating time. It is interesting to note that to

date; there is no known report of electroless plating of Kapton membranes.

Experimental

Materials

Kapton 50 HN foils (12.5 Cpm thick, 107 tracks per cm2) were obtained from the linear

accelerator laboratory UNILAC at the GSI (Darmstadt, Germany). Boric Acid was obtained

from Fisher and used as received. NaOCl (13% active Cl), KI, SnCl2, AgNO3, NaHCO3, were

obtained from Aldrich and used as received. Trifluoroacetic acid, Na2SO3, NH40H,

formaldehyde and methanol were obtained from Mallinckrodt and used as received. Commercial

gold-plating solution (Oromerse SO Part B) was obtained from Technic Inc. Milli Q water was

used to prepare all solutions and to rinse the membranes.

Chemical Etching

Cylindrical pores were etched in the tracks of the Kapton foils using sodium hypochlorite

solution as describes in detail elsewhere. 116 The shape and size of the pores can be tailored by

the choice of etchant and the etching conditions. 116 To obtain cylindrical pores the etching rate

along the track, the-so called track etch rate yt has to be much faster than the non-specific

etching of the polymer called the bulk etched rate yb. 14, 81 The relation between yt and yb is

explained in figure 4-1. 14, 81 Studies have shown that an efficient etching can be performed in

sodium hypochlorite containing 13% active chlorine content. 116 Furthermore, it has been

demonstrated that the shape of the pore can be regulated with an appropriate choice of pH. 116

When sodium hypochlorite is not buffered its pH ~ 12.6. At these conditions, and elevated

temperature of 50oC, yb of Kapton is high (~ 0.21 (Cpm / h) and the pores become strongly

conical. Buffering the etchant with boric acid to pH ~9 enables one to obtain cylindrical pores.

It is important to note that the etching works only at basic pH when the hydrolysis of imide

bonds by OH- is possible. The membrane was immersed in a Teflon container containing 400 ml

of NaOCl solution at pH 9.8 and temperature of 50oC. First, the container with the etchant

solution is brought to 50oC by placing it in a water bath controled by a water heater. The size of

the pores increases with etching time. After etching, the membrane was rinsed with D. I. water

and left to soak for two hours. The membrane was allowed to dry in air overnight.

Electroless Plating of Kapton

In order to better control the size and surface chemistry of Kapton, the membranes can be

plated electrolessly with gold. 10 Tailoring of the pore size in polymer membranes by the time of

performing the electroless plating with gold has been demonstrated previously. 28' 51 Since

Kapton membranes possess carboxylate groups made available via imide hydrolysis by the

etchant, it is expected that these would act as active sites for the bonding of tin (H). Tin can then

reduce silver ion which later acts as a nucleation site for the reduction of gold. The procedure

followed the recipe for the electroless plating of polycarbonate. 28' 51

Pore Diameter Measurement

Cylindrical pores were characterized by taking scanning electron microscopy (SEM)

images of the membrane surfaces. Also, we have chosen to use an electrochemical technique

based on measuring ion current to measure pore diameter. This entailed mounting the membrane

sample in the cell (Figure 4-2), filling both half cells with an electrolyte solution of known ionic

conductivity, and obtaining a current-voltage (I-V) curve associated with ion-transport through

the nanopore. The experimental slope of this linear I-V curve is the ionic conductance, G, (in

Siemens, S) of the nanopore, which is given by

G = (NA GKal x d2) / 4 L (4-1)

where N is the pore density, A is the membrane area, GKal is the experimentally measured

conductivity of the KCl-based electrolyte used (S cm )~, L is the length of the nanopore

(membrane thickness), d is the pore diameter. Because all of the other parameters in Equation 4-

1 are known, d can be calculated.

Results and Discussion

SEM and lon Current Measurements

Figure 4-3 shows SEM images of Kapton porous membranes before and after electroless

gold plating. Here we see that the pore diameter decreases with plating time. Figure 4-4 shows a

plot of pore diameter as a function of Au plating time from 0 to 12 hours. The two methods of

measuring pore diameter gave very similar results. This is important because only resistance

measurements can be used for small pore diameter that cannot be resolved by SEM. This data

indicate that one can tailor the pore diameter with plating time down to the nanometer scale.

This capability is important for different transport studies and in sensor research where the size

of the pore with respect to the analyte is important. 10, 28, 77, 96

Atomic Force Microscope Images

Figure 4-5 shows the atomic force microscope images of the Kapton membranes before

and after electroless plating. The membranes remain relatively smooth after gold plating

compared to a similarly porous structured polycarbonate membrane after plating with gold.

Conclusion

Electroless gold plating properties on the surface and pore walls of track-etched Kapton

polyimide nanoporous membranes were studied. SEM, AFM, and ion current measurements

were used to characterize the surfaces and pore dimensions of the membrane. Nanoporous

Kapton polyimide membranes were electroless gold plated over different times and the pore

diameter characterized using SEM. Ion current measurements were used to measure the

diameter of very small pores. AFM images show that after electroless gold plating, the gold

surface layers are smooth compared to a similarly structured polycarbonate membrane.

Electroless plating the membranes for 12 hours produced Au wires in the pores. Etching away

the membrane with sodium hypochlorite to expose gold tube replicas revealed that the plating in

the walls is also smooth.

Latn intrc

Fiue -. eintono bl ec rt V n tak th ae t[Aate ro peP

Electrodes

Pressure
plate

0.5 cm
1t 1

Clamping
screw

Solution
chamber

lon track membrane

Poly(chlorotrifluoroethylene) cell

Aluminum frame

Figure 4-2. Schematic of cell used for electrochemical measurements.

3.5 cm --+

Before AU

3 hrs plating

Figure 4-3. Pore diameters for different Au plating times.

220
200
180
160
S140

S120
100
80
o
60
40
20 *

0 1 2 3 4 5 6 7 8 9 10 11 12 13
Time (hrs)

Figure 4-4. Pore diameter as a function of plating time with measurements taken from SEM
image (0-7.5 hrs) and ion current resistance measurements (8-12 hrs).

Before Au

Kapton
\4 membrane

Roughness ~ 3.6nm

After 2.5 hrs Au plating

~ 6.3nm

PC membrane After 2.5 hrs Au plating

Before Au

IRoughness ~ 13.6

SRoughness ~ 5.1

Figure 4-5. Atomic Force micrographs of Kapton and PC membranes before and after
electroless Au plating

CHAPTER 5
ASYMMETRY INT DIFFUSIONAL TRANSPORT OF MOLECULES THROUGH KAPTON
CONICAL NANOPORES

Introduction

Membranes and porous materials have found various applications in filtration and

separation processes. s-10 Modern biotechnology has posed new challenges in the application of

such membranes, and requires pores with diameters similar to those of molecules under study 11

(e.g., as small as several nanometers). The nanometer scale of such pores is necessary in both

achieving optimal control of the flow of biomolecules, as well as in developing sensors for their

detection. The transport properties of such nanometer scale pores are not well understood yet.

The hint that nanopores behave differently from micropores, comes from Mother Nature. 12

Biological channels and pores have diameter of ~ 1 nm and are critical for functioning of living

organisms. Ion channels and pores exhibit transport properties not observed with larger pores,

for example (i) selectivity for ions or molecules, (ii) rectification of ion current" (iii) ion current

for constant voltages applied across the membranell (iv) facilitated transport of molecules (v)

transport of ions and molecules against their electro-chemical potential gradient. 119

It has recently been demonstrated that asymmetric conical nanopores in polymer films can

exhibit transport and rectification properties similar to biochannels. For example, conical

nanopores in PET and Kapton membranes are cation selective and rectify ion current with a

preferential direction of cation flow from the tip to the base of the cone. 81, 88, 105, 120, 121

However, unlike the above cases, in this investigation there is no applied transmembrane

potential. Also, there are no electrostatic or binding interactions of the molecule with the pore

surface. Since the molecule under study is neutral, transport through the membrane is by purely

diffusional and geometrical constraints. Yet we found in our studies that the diffusion of

molecules across these membranes, exhibit a rectification behaviour. It means that there was a

preferential direction of diffusion flow. This effect cannot be predicted by considering classical

diffusion with constant diffusion coefficients. The rectification of conical nanopores is of

tremendous significance for industrial filtration processes in which asymmetric membranes are

often applied. One type of asymmetric membranes used in industry consists of a thin 'skin' of

nanoporous material placed on a low resistance support, which assures high fluxes and good size

separation. 122 Finding an optimal direction of concentration gradient will improve the filtration

process. Asymmetric diffusion has been observed before with multi-membrane systems and

membranes with two skins on two membrane faces of a membrane support. 123

Diffusional Transport Described by Fick's Laws

Diffusional transport is described by Fick' s two diffusion laws. For diffusion in one

dimension, Fick' s first law of diffusion describes the flux (the net number of moles of particles

crossing per unit time, t, through a unit area perpendicular to the x-axis and located at x), stating

that the flux of a molecule with diffusion coefficient, D, is directly proportional to the

J(x, t) =-D Scx )(5-1)

D normally is assumed constant, however there are known examples when D depends on

position x, concentration c or even time t. 124, 125 The diffusion coefficient determines the time it

takes a solute to diffuse a given distance in a medium. The diffusion coefficient depends on the

physical characteristics of the solute as well as those of the medium. An approximate bulk

diffusion coefficient (like that of molecules of similar masses) was used here, because, we are

more concerned about emphasizing the preferential flux from one side of the membrane than the

absolute flux values, themselves. The second of Fick' s law expresses how the concentration of

the species depends on time. Here again, the general form of the law in one dimension, taking

into account the possible forms ofD is as follows:

Sc f Sc\
-o (5-2)
at ax \ xc

Hindered Diffusion in Cylindrical Pores

The effective diffusion coeffcient of a solute within a pore of comparable size is lower than that

of the bulk solution value. This phenomenon is called hindered diffusion and results from steric

exclusion of the solute at pore opening and hydrodynamic or wall drag resistance due to the

presence of the pore wall. For steady-state diffusion through a membrane with cylindrical pores

that are comparable to the size of an uncharged solute, the effective diffusion coefficient can be

expressed in terms of the solute-to-pore size ratio Ai (= a/r) by the Renking equation: 126, 127

D / D, = (1 A1) (1- 2. 1044Al + 2.089il + 0.948il) (5-3)

Here, restriction to diffusion due to steric hindrance at the entrance to the pores is given by the

partition coeffcient,

O = (1 A) .(5-4)

As established by Ferry,128 a molecule must pass through the opening without striking the edge.

Therefore the center of the solute particle cannot be located at a radius that exceeds (r-a) (Figure

5-1). The partition coeffcient therefore, is equivalent to the fraction of the cross-sectional area

of the pore that is accessible to the center of the molecule.

The second factor in the Renking equation, called the inverse enhanced drag,127

K = (1- 2. 1044Al+ 2.089il + 0.948il) (5-5)

corrects for hydrodynamic or wall drag resistance, which is the friction between a molecule

moving within a pore and its walls (Figure 5-2).

The total flux, J, through a membrane with cylindrical pores of length, L, and pore density, N ,

can then be given as:

Nz ~r2 OK D, AC
J = (5-6)

where D.,is the free diffusion coefficient of the solute in the bulk solution, AC is the

concentration difference across the membrane, and N x r2 is the membrane porosity, E.

Hindered Diffusion in Conical Pores

Equation 5-6 is for a membrane with cylindrical pores where r is constant. The pores used

for this study are conical, and therefore r increases or decreases from one end of the membrane

to the other, thus varying linearly with position. The porosity of conical pores is different from

that of cylindrical pores, and is obtained by the product of the pore density and the geometric

mean cross-sectional pore area:

Nx ~d, d
E = '(5-7)

where d, and d, are the diameters of the large and small opening of the pores respectively.

There are varying hindrance effects (steric and hydrodynamic) on diffusion over the length of the

membrane. The average inverse enhanced drag, K 1, is ~ 1 for conical pores because most of the

wall drag resistance occurs in the region of the tip, which is a very small faction of the pore. For

cylindrical pores, the partition coefficient determines the rate of entry of the molecules into the

pores, and is equal to the ratio of the solute concentration in the pore to that of the bulk solution

at equilibrium. However, unlike cylindrical pores, the partition coefficient that determines

molecular entry into conical pores at steady state diffusion is not the same as the equilibrium

partition coefficient. The partition coefficient,0(r(x)) varies with r; which is a function of

position, x, and therefore changes along the length of the pores. Hence it is expected that the

equilibrium concentration along a pore length is different. However, the partition coefficient and

also the enhanced inverse drag within the pore, averages to be of negligible contribution to the

flux since only a very small fraction of the pore approaches the size of the solute molecule. The

value of these coefficients therefore approximates to 1. The asymmetry in diffusion, then, may

be caused by the difference in steric hindrance at the entrance of either side of the pore. This

will determine the amount of molecules that enter and leave the pore per unit time. The above

arguments suggest that molecules entering from the large opening of a conical pore will have

greater flux than those that enter from the smaller opening, provided that they can leave at the

same rate. The limit to diffusion will be the partition coefficient at the tip. With these

assumptions, the fluxes can be written for dilute solute concentrations as

E Ob K-' D, AC
Jb =(5-8)

E O, K D, AC
J, =(5-9)

where the subscripts b and t represents the directions of the net flux entering the large and small

opening of the pores respectively, and K-1 is equal to 1.

We can also solve the diffusion problem with a constant D, through a conical nanopore

with opening angle ot, openings dt and db, with solute concentration co at the tip, and zero

concentration of solute at the base:

c(x) =C co d, -cot a <; < (5-10)

valid for dt<
The diffusion problem with solute placed at the base side has the following form:

c:lx)=i co 1 ,dt -cota< x

When we calculate the fluxes (moles/s) in the two above mentioned conditions they are equal to

each other:

j =Dod'x (5-12)

These calculations indicated that 'classical' diffusion with constant D cannot describe our

experiments, unless one takes into account the difference in boundary conditions due to the

difference in the partition coefficient.

It was the purpose of this study to investigate the diffusion of neutral molecules from both

ends of conical pores of the polyimide Kapton. Here the dimension of the tip of the conical

pores approaches that of the molecules. To this end the diffusion of the water soluble neutral

molecule phthalazine was studied.

Experimental

Materials

Kapton 50 HN foils (12.5 Cpm thick, 107 tracks cm-2) were obtained from the linear

accelerator laboratory UNILAC at the GSI (Darmstadt, Germany). Sodium hypochlorite

(NaOC1, 13% active Cl) and potassium iodide (KI), were obtained from Aldrich and used as

received. Phthalazine, sodium chloride, sodium phosphate dibasic, sodium phosphate

monobasic, sodium azide was obtained from Fisher and used as received. Milli Q water was used

to prepare all solutions and to rinse the membranes.

Kapton Polyimide Membrane

We used polyimide foils to prepare nanoporous membranes. Polyimide possesses a unique

combination of properties that are ideal for a variety of applications in many different fields.

The film maintains excellent physical, electrical, and mechanical properties over a wide

temperature range. Polyimide also has very good chemical resistance and does not dissolve in

organic solvents. We used 12.5 Cpm thick commercially available Kapton 50 HN, produced by

DuPont.

For the preparation of membranes we used the-so called track etching technique. It is

based on irradiating a dielectric film with swift heavy ions and subsequent chemical

development (etching) of the damaged ion tracks. A unique feature of heavy ion irradiation is

single-particle recording. That is to say, one swift heavy ion which penetrates the foil produces

one damaged track. Therefore, counting the number of ions used for irradiation enables one to

prepare membranes with tailored number of pores from the range 1 up to 1010 ions/cm2. Kapton

foils, which we used for these experiments, were irradiated with uranium ions of energy of 1 1.4

MeV/u, at the heavy ion accelerator UNILAC at the Institute for Heavy Ions Research, (GSI)

Darmstadt, Germany. We used foils irradiated with the fluencies 107 and 10s ions/cm2. The

range of the ions is in all cases larger than the thickness of the polyimide membranes. The ions

penetrate the membranes at normal incidence, creating damaged tracks, which is then followed

by chemical etching to form pores (Figure 5-3).

Chemical Etching of Membrane Tracks

After irradiation of the Kapton foil with heavy ions, the latent tracks have to be chemically

etched. The shape and size of the pores can be tailored by the choice of etchant and the etching

conditions. To obtain cylindrical pores the etching rate along the track, the-so called track etch

rate Yt has to be much faster than the non-specific etching of the polymer called the bulk etched

rate y,. The relation between yt and yb is explained in Figure 5-4.

To obtain conical pores, one needs to choose etching conditions which assure high vb.

Preparation of conical pores is normally performed in a conductivity cell with etchant placed

only on one side of the membrane. The other side of the membrane is in contact with a stopping

medium, which neutralizes the etchant as soon as the pore is etched through. For example, if

NaOH is used as an etchant, we use an acidic stopping medium. The chemical stopping is

further supported by an electric stopping. The etching is performed under voltage with

electrodes arranged in such a fashion that the anode is on the side of etchant, which retracts ions

active in the etching process (e.g. OH-) out from the pore ( Figure5-5).

Etching with an applied voltage and measuring electric current affords a method to monitor

the process. At the beginning of etching the current is zero, because the two chambers of the

conductivity cell are not connected with one another. When the pore is etched through, the

current value is finite and increases in time, indicating an increase of the pore diameter (Figure

5-6). Kapton is very resistant chemically, therefore, development of latent tracks has to be

performed by a very aggressive etchant and at elevated temperatures. Previous studies have

shown that an efficient etching can be performed in sodium hypochlorite containing 13% active

chlorine content. Furthermore, it has been demonstrated that the shape of the pore can be

regulated with an appropriate choice of pH. When sodium hypochlorite is not buffered its pH ~

12.6. At these conditions, and elevated temperature of 50oC, yb of Kapton is high (~ 0.42 (Cpm /

h) and the pores become strongly conical. Buffering the etchant with boric acid to pH ~9 enables

one to obtain cylindrical pores. It is important to note that the etching works only at basic pH

when the hydrolysis of imide bonds by OH- is possible.

To obtain conical pores in Kapton, the irradiated samples were placed between two

chambers of a conductivity cell and etched from one side in sodium hypochlorite. The other half

of the cell is filled with 1 M potassium iodide (KI) solution as a stopping medium for the OCl-

ions of the etchant. As soon as the etchant completely penetrates the membrane, iodide ions

reduce OCl- to Cl- ions: 112, 130

OCl- + 2H' + 21[- It + Cl- + H20 5-13

Via this reaction, the etching process stops immediately after the breakthrough, allowing the

preparation of extremely narrow pores.

Pore Diameter Measurement

SEM was used to characterize the opening diameters of conical pores, especially the big

opening, which we call the 'base' (Figure 5-7). The small opening of conical pores, called the

'tip' is below resolution of SEM, therefore, we have to use another technique for its size

estimation. We have chosen to use an electrochemical technique based on measuring ion

current. The ionic conductance of a conical pore is related to its diameter by the following

equation:

G = nc trdbd, 5-14
4L
where n is the number of pores, cris the conductivity of electrolyte, L is the length of the pore (or

the equivalent membrane thickness), and db and d, are the diameters at the base and tip of the

cone respectively.

Transport Measurement

Measurements of neutral molecular transport through conical nanopores were performed

using UV-Vis spectrometry (Agilent 8453). A U-cell set-up contained the permeate molecule

solution on one side and a buffer, in our case PBS (in which the permeate was prepared) on the

other side of the cell (Figure 5-8). The membrane was sandwiched between two transparent

tapes that have holes in the center that defined the transport area. This tape-membrane

composite was clamped between the U-cell. The permeate solution was removed at certain time

intervals and the permeate concentration was measured be UV-Vis spectrometry. Feed solution

concentrations were varied in 4 ml PBS (pH 7.2), and the permeate side was only 4 ml PBS. The

solution on both sides of the U-tube was stirred using stir bars and stir-plate set-up.

Viscosity Measurements

The viscosities of phthalazine and dextrose were measured using Cannon-Fenske

viscometer No. 75 (model P200; Cannon Instruments)

Results and Discussion

Membrane Characterization

Figure 5-9 is an SEM image of the base side of a typical porous Kapton membrane used

for the transport study. The average pore density, taken from approximately five hundred pores

in five different locations of the image under low magnification, was 10 /cm2. Figure 5-10 is an

SEM image of the base showing a diameter of 1.68 pm. The size of the tip, which is below the

resolution of SEM, was calculated from ionic conductance measurements (Figure 5-11) using

equation 5-14. The average tip size used in this study was ~ 2 nm. The average diameter of the

molecule for transport was ~ 0.7 nm.

Transport Measurements of Phthalazine

From equation 5-4, the partition coefficients at the base and tip approximate to 1 and 0.42

respectively. These values put the partition coefficient at the base about 2 times that at the tip.

From the proposed equations, Figures 5-8 and 5-9 for base and tip fluxes, it is expected that at

low concentrations the flux from base to tip should be about twice that in the opposite direction.

Indeed, Figures 5-12 and 5-13 show that base fluxes are about twice that of tip fluxes for 1 and 3

mM solute concentrations. However, base fluxes approach those of tip fluxes as the

concentration of the solute increases, and both base and tip fluxes are similar at higher

concentrations (Figures 5-14 and 5-15). Figure 5-16 shows this asymmetric behavior over a

wider rage of concentrations.

The higher flux from base to tip direction can be explained by a lower access resistance of the

solute to the pore. Figure 5-17 shows a comparison between the theoretical flux obtain from

equation 5-9, and the experimental flux from tip to base. We see that these are in good

agreement over the wide range of concentration used. The experimental flux from base to tip, on

the other hand, deviates from theoretical values obtained from equation 5-8, particularly at

higher concentrations. To determine whether the base flux or tip flux behaves in a classical way,

a plot showing increase of flux through base and tip, respectively, with respect to the flux

measured at 0.1 mM was generated (Figure 5-18). The straight line: Flux(c)/Flux(0.1mM) =

c/0.1 obtained for the tip flux indicates that transport through tip behaves according to the Fick's

law with a constant D. Transport through base is hindered for higher concentrations indicating

that diffusion coefficient is concentration dependent.

The flux (mol/s) through an aperture of diameter dt with boundary conditions co is given

by: jlim,,= ited Dco. This value is smaller than the flux as given by eq. 5-12 The flux through

the tip is limited by the value of jlimited, While the flux through the base is not. The value of flux

through a conical nanopore as given by eq. 5-12 can be easily obtained when the solute passes

from base to tip.

The open question however remains, why the ratio of fluxes is concentration dependent. In

order to answer this question, we examined behavior of the flux from base and tip side,

respectively for different concentrations of the solute.

The Influence of Cosolute Concentration on Asymmetry

We investigated whether molecular crowding / jamming had any influence in the behavior

of the base flux at higher concentrations. To do this, we used a constant concentration (5 mM) of

phthalazine plus varying concentrations (0 to 100 mM) of the cosolute dextrose on the feed side

of the cell. We then placed identical total concentrations on the permeate side of the transport

cell using dextrose (Figure 5-19). This procedure was done to ensure that there was no osmotic

influence on transport. Figure 5-20 is a calibration curve for phthalazine, and phthalazine with

dextrose cosolute, generated from UV-visible absorption measurements. The two plots were

placed together to show that the addition of dextrose as a cosolute to phthalazine does not affect

the absorbance to any significant degree. Also, to address the issue of viscosity, kinematic

viscosity measurements were done on phthalazine and dextrose over the concentration range

used in this study. Figure 5-21 indicates that there is negligible change in viscosity over the

range of concentrations (0 to 100 mM) used for this study. Figure 5-22 shows the influence of

varying concentrations of the cosolute dextrose on the flux of phthalazine, which is kept at a

constant concentration of 5 mM. The tip flux is not affected over the entire range of

concentrations. However, the base flux is sharply decreased to that of the tip flux when dextrose

is added to the feed solution. Even though phthalazine concentration was kept constant, base

flux approached the tip flux in a very similar way when phthalazine concentrations were

increased. The increase in total concentration on the base side of the membrane causes base flux

to approach the limit of tip flux. Proper mathematical modeling of this effect is needed, taking

into account dependence of diffusion coefficient on concentration of the solute and position in

the channel. We would like to mention that theoretical studies were reported in which two sizes

gas molecules were placed on the base side of conical pores, and one size molecule was too large

to pass through the tip. In this situation a total j amming was found, causing the diffusion of the

small molecules to be zero over time. As expected, the rate of such j amming increased with

concentration. On the other hand, when the molecules were placed on the tip side, no j amming

occurred from that entrance. 131 We think that in our case, an increase in total concentration of

the solutes might lead to similar partial molecular j amming of diffusion from base to tip.

Decrease of the base flux provides evidence that it is the exit rate of the solute from the pore that

limits the diffusion transport in the direction from base to tip.

Conclusions

Diffusion rates through a membrane can be asymmetric due to molecular binding,

electrostatic interaction, and difference in osmotic potential. In this study we have demonstrated

that asymmetric diffusion can also occur by purely geometric constrains of conical pores on the

diffusing particles, where the tip of the cone is comparable to the dimension of the molecules.

We show that asymmetric behavior of diffusion is concentration dependent and there appear to

be some sort of partial j amming effect related to the increase in concentration of molecules from

the large opening of the pores. To further shed more light on this interesting phenomenon,

proper mathematical modeling of this effect is needed, taking into account dependence of

diffusion coefficient on concentration of the solute and position in the channel.

r~ra

Figure 5-1. The partitioning of a spherical molecule of radius, a, in cylindrical pores of radius,
r.[Adapted from Davidson, M. G.; Deen, W. M. Macromolecules 1988, 21, 3474-
3481.]

h = a/r

Figure 5-2. Spherical molecule of radius, a, moving within a cylindrical pore of radius,
r.[Adapted from Davidson, M. G.; Deen, W. M. Macromolecules 1988, 21, 3474-
3481.]

Dielectric
mate rial
Swift heavy ions

a-Betenion

Figure 5-3. Swift heavy ions impinge on a dielectric solid leading to damaged ion tracks witch
can be chemically etched to form pores(A and B, respectively)

.a- t

' ~a

Figure 5-4. Definition of bulk etch rate Vb and track etch rate Vt.[Adapted from Apel, P.

Pressure
0. cmplate

Solution 10c
chamber I I cm 5 e~m

Clamping
screw

Figure 5-5. Conductivity cell used to prepare conical pores.

lon track membrane

Aluminum frame Poly(chlorotrifluoroethylene) cell

150-

242 244 246 4 250 252
time (min)

Figure 5-6. Etching curve showing moment of breakthrough with sharp increase in ion
current.[Adapted from Siwy, Z.; Apel, P.; Dobrev, D.; Neumann, R.; Spohr, R.;
Trautmann, C.; Voss, K. Nuclear Instruments & M~ethods in Physics Research,
Section B: Beamn Interactions nI itr Materals and Atoms 2003, 208, 143-148.]

Figure 5-7. Kapton membran showing large opening (base) of conical pores

Feed side

Phthalazine + *--
PBS (pH 7.2) '

U-cell

pH 7.2)

Stir bar

Conical nanoporous
membrane

Figure 5-8. Experimental set-up for transport measurements

Permeate side

Figure 5-9. Scanning electron micrograph of base side showing pore density (107 pores/cm2) Of
Kapton membrane.

Figure 5-10. Scanning electron micrograph of the base showing diameter of 1.68 Cpm

V.VVVY~

I = 0.0077E
R2 = 0.9918

0.00020
0.00015
0.00010
0.00005

-

0.02 -0.015 -0.01.005

-0.00010
-0.00015
-0.00020

E (V)

0.005 0.01 0.015 0.02

Figure 5-11. A typical current-voltage curve used to measure the tip diameter of the conical
nanopores.

100

90

HBase
A Tip

0 100 200 300 400 500 600 700 800 900 1000

Time (min)

Figure 5-12. Base and tip fluxes for 1 mM phthalazine feed solution.

mBase

0 100 200 300 400

Time (min)

Figure 5-13. Base and tip fluxes for 3 mM phthalazine feed solution.

120

HBase to tlp
| ATlp to base

0 50 100 150 200 250 300 350 400 450
Time (min)

Figure 5-14. Base and tip fluxes for 10 mM phthalazine feed solution.

E

n 150
Z,

A Tip
aBase

0 50 100 150 200 250
Time (min)

Figure 5-15. Base and tip fluxes for 50 mM phthalazine feed solution.

Concentration (mM)

Figure 5-16. Asymmetric behavior of flux versus concentration.

0.6

S0.5
(V_ Theoretical
S0.4

O 0.3

0. Experimental

0 5 10 15 20 25 30

Co ncentration (mM)

Figure 5-17. Theoretical and experimental tip flux vs concentration

350

300 -

250

200 + from ti p
S150 -I from base

100 +

50 m

0 100 200 300

c [mM]/0.1 [mMV]

Figure 5-18. Increase of flux through base and tip, respectively, with respect to the flux
measured at 0.1 mM.

U-cell

'BS (pH 7.2)
+ Dextrose

Phthalazine +

Dextrose+

PBS (pH 7.2)

Conical nanopore Si a
membrane

Figure 5-19. Experimental set-up for transport measurements of phthalazine with cosolute
dextrose

y = 0003058x
R2 = 0 998450

y = 0003021x
FF = 0 999811

* Phthalazine
SPhthalazine with 490 mM Dextrose

y = 0003021x
R2 = 0 998406

o 1

0 09

0 08

0 07

0d 05

001

0

-0 0035
S0 003
m 0 0025
z 0 002
S0 0015
S0001
m 0 0005
0

02 04 06 08
CONCENTRATION(m icrom olar)

5 10 15 20 25 30 35
CONCENTRATION(m icrm olar)

Figure 5-20. Calibration curve of phthalazine, and phthalazine with the highest concentration of
dextrose cosolute used