<%BANNER%>

Growth and Characterization of Novel Gate Dielectrics for Gallium Nitride

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 E20110114_AAAACC INGEST_TIME 2011-01-14T12:07:25Z PACKAGE UFE0006644_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 12401 DFID F20110114_AABGOB ORIGIN DEPOSITOR PATH onstine_a_Page_093.pro GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
101ea0f49b6feb5b6d2f016f8388f3fb
SHA-1
c3e14549604299103d51f2d81f7d91ec5e8fa395
4450 F20110114_AABGNN onstine_a_Page_078.pro
7d67dcbe6ab100990822b1c041062bdf
db74feb4ec65b829a3e9a35101e60e8c6866b37c
43991 F20110114_AABGMY onstine_a_Page_063.pro
3df26bbac54a3714f34e9fc75379bfa6
76fc2379408c31e7341b32a2f25e79416ec268c5
15107 F20110114_AABGOC onstine_a_Page_094.pro
3d0facdd61d5f954460f1811f36bb6cc
ea9ee2a3f06eb0af1af55d20f6be150621ac5995
7739 F20110114_AABGNO onstine_a_Page_079.pro
b699b7394fc8ccebf62d205c96bd328f
96a609d05a3300ef1b5cd971d8294ccb10940b51
49871 F20110114_AABGMZ onstine_a_Page_064.pro
1f0616ee9cade29ea88aa98925e0df16
9e32b02deafe13b21b575f2d922c790f2e0a53b1
9680 F20110114_AABGOD onstine_a_Page_095.pro
9de0dd96619e54884332d05247e439a9
c403edc07645a6a66c43a7bc92a4a038e6915c88
6512 F20110114_AABGNP onstine_a_Page_081.pro
100cc9d6cb2612386529dfa2c46a0753
291c11bfdbab51cb395402135e30e2174115cbe8
2488 F20110114_AABGOE onstine_a_Page_096.pro
769721fe3197dc425b4db0c26e31f409
0e2d10256201fd86f0fb49e5d8a60c74e579c108
7232 F20110114_AABGNQ onstine_a_Page_082.pro
88eced9c15b1598bf4f0e165961957b3
82fc7b94c1b7dd111626c34d31b90cfc291cf9a1
4455 F20110114_AABGOF onstine_a_Page_097.pro
7f203967bb2b8308cf196112ecbfde39
5174502a69a502996fe4ff66b8221162b7e8e79d
1421 F20110114_AABGOG onstine_a_Page_099.pro
2c30dad976d919fbc498868455ba4290
070560c23f89368a1010ced41d93998912a779f1
1356 F20110114_AABGNR onstine_a_Page_083.pro
f08774bf6cb4fdb8b7b1d68f9969f7e3
b1ddd68fe2c6a38affaac5d3106d1c44968b75f2
3910 F20110114_AABGOH onstine_a_Page_101.pro
85c1a5c0b53a5b4e086bfcb0b44d0998
4032638b5af92e2270fc7a53842fdb093ebc7697
1877 F20110114_AABGNS onstine_a_Page_084.pro
932f09f6ac2013f2850ae309bec8f98d
08dda5693f2e08a57a893fe140cc2d4c4fce1528
10975 F20110114_AABGOI onstine_a_Page_102.pro
6a674f300b185666e8974205ec569573
1c0245c934c399dff8d543b349854cbb1b452938
12298 F20110114_AABGNT onstine_a_Page_085.pro
ebb5f9bc7097125ede6d6fe789d78a66
83feb831f11b9efa3f4412c9f9dc8157ce97d159
7054 F20110114_AABGOJ onstine_a_Page_103.pro
e617a6ceaef402f068159a391c20a30a
88302e51dea4814fbc316622e1c7618fd7a7db8d
8629 F20110114_AABGNU onstine_a_Page_086.pro
c49ebf6b70e0ed0ffadd7689cde2ef95
7927d9f5e2b369fc8174520bee3eb28ed54fa4d6
9382 F20110114_AABGOK onstine_a_Page_104.pro
714c426c0d7a2cb4bde731f8e31dca14
b20f02c1d3f298753d2aa88de038d0aff42e9517
42904 F20110114_AABGNV onstine_a_Page_087.pro
b94e0ef80d968a0d90954d22e1f2bfe7
fb720c9b3218d792a12b7b32f547946a3df26978
9954 F20110114_AABGOL onstine_a_Page_105.pro
b0a830565558db9ad076d1f7ea62a7f4
d39fb9785b3429827d594ec4388387fe2fef7ff3
52938 F20110114_AABGNW onstine_a_Page_088.pro
435a20e7c0d486d7352d3f9442b375cd
0ec4b7b4229b3d2f46f882af1f7769bb0e2be1da
9228 F20110114_AABGOM onstine_a_Page_106.pro
a68634149409cc61ba26339a24a003fb
d28b4f1c51e9e7d5d03d3d6bdaa9c5ba2aca3d11
46545 F20110114_AABGNX onstine_a_Page_089.pro
7c569666e217ab8fbcd0aec506131567
46767c1c8b440a7888081671c605eaf601f37ec4
10509 F20110114_AABGPA onstine_a_Page_120.pro
11cf9baa2ede57d823e4e783ecc2ffab
92f62142074556d11a1edddbd84d17bdb6514d06
12490 F20110114_AABGON onstine_a_Page_107.pro
d60f537e896c77d5348b59f060fcd99b
c83f08eeb63408bc675d94471171649f4a1e39ef
50437 F20110114_AABGNY onstine_a_Page_090.pro
60dab19bd6560fc5d7bbcb5b53426512
e46d6427edc464cb98c687e0cb00543c18bd52b5
3018 F20110114_AABGPB onstine_a_Page_121.pro
5e2f474174ec784ed2229a1839d8d0c3
0188d007a8ed1f1bc53b8dae5d040f29c41c190e
11070 F20110114_AABGOO onstine_a_Page_108.pro
49072d21c5c804f4608a9608b6c481f4
6930a7216f6ad3354fd2bada09ee4b68234be459
47680 F20110114_AABGNZ onstine_a_Page_091.pro
48f4aafe824f0b92103fd649efc974b6
8cbc7ddd9e883c4d6605759496e331cf14b0f49f
7064 F20110114_AABGPC onstine_a_Page_122.pro
63e2cfd214056b40446a19f892b9c03b
8df1d53a60f48d23175db5487787c18467b7eecd
9148 F20110114_AABGPD onstine_a_Page_123.pro
493ce0da9d0feb33d4408164ff887839
bf8d7cf294d4ad9e898bf83a04a355622cff8487
5644 F20110114_AABGOP onstine_a_Page_109.pro
69e7979737adb8dcfd5bf46b98c14c60
ed6d3868a55e2a5b34ddc08361c08cfb101c1041
21271 F20110114_AABGPE onstine_a_Page_124.pro
ac6500058ad0e915107a704dc7f46567
54bc4aaf5872a08679e626c0f633e72ec2ab6e00
6223 F20110114_AABGOQ onstine_a_Page_110.pro
0a2c1ba644e3e5194d6b38ad41f7d2ee
7e65e25c05d55631750f9dd0bec6c133e422524a
11197 F20110114_AABGPF onstine_a_Page_125.pro
c2ec039e340c166ecddbff64fae5c22a
4c9398181f1a480a107a4d5ccd18440b977b2194
8937 F20110114_AABGOR onstine_a_Page_111.pro
457ca1a5ed03d667fcf5a7ed3b3e6178
05502cd0e3471d6dd4210d8882403990e6aabb7b
12013 F20110114_AABGPG onstine_a_Page_126.pro
223a277a2c30a33fa1ae7f8662d8dbac
6f9ac982284ce0cf178bcaa35b9c178d9570645f
5962 F20110114_AABGPH onstine_a_Page_127.pro
3281f2067b3e86ca5e528fabe746a32d
dea2f8363a7585ddfd61fe066499a49ea97854ba
12475 F20110114_AABGOS onstine_a_Page_112.pro
86e5e90f300cd24d90545622e14bf1d9
24d49a2b941715c2e8762cf8aeb990e91f019a91
3336 F20110114_AABGPI onstine_a_Page_128.pro
36e2a6ee6efe16574b368c6c48e6d6c4
3d2e00ea56039f53197ef61c228d57420f6856ed
7980 F20110114_AABGOT onstine_a_Page_113.pro
d6617a55756a200b1cfe14cd2ae3a5bc
942c31d15d8a757ff9d26cf5892153f4e48d84f6
2525 F20110114_AABGPJ onstine_a_Page_129.pro
6d9999cf1800df4fa3b93673a23c1816
95d13aeac78d8d59f32ea06cdddfdee882858a2e
6054 F20110114_AABGOU onstine_a_Page_114.pro
eae6181f09eeab51d68ca3e1efb81855
57928b229e8d322018d56849efbceab0de1368db
5634 F20110114_AABGPK onstine_a_Page_130.pro
3d7b66c3abb5ee71c55cfe1690e188d6
b1caa99d98f18638e6f355ad81cf15d986017b34
9592 F20110114_AABGOV onstine_a_Page_115.pro
5dd6ed981672520ae2ad6cb899ed8930
95341e09cc2af4fcc24230b406bf992c4713e301
43271 F20110114_AABGPL onstine_a_Page_131.pro
16d9973b444979044dde686dfc5f45ff
38913e5a56138eae2652ffcda5ae9938719aef4d
40942 F20110114_AABGOW onstine_a_Page_116.pro
54f35bc09542b071443f1424e83ccab9
6f3faa901f6a5504a7d0f9dd05a52d711414229a
4329 F20110114_AABGQA onstine_a_Page_146.pro
cb7739a47a01d0596e37db3e7fc46057
ba7e3be45d3f2bfc342fdbdf66b812944b3e4094
44993 F20110114_AABGPM onstine_a_Page_132.pro
c5d8065286ca73978e53b094b8b55ff6
ae7b9c046651db192c4b263db392cea733faf041
47945 F20110114_AABGOX onstine_a_Page_117.pro
f1d3f34dfdf46f89ef2256c4724ea26a
a0a65e5c09db04e19e8bcc30eb7c50363c391ba0
39218 F20110114_AABGQB onstine_a_Page_147.pro
7b367907151d3550cba904ff82ba2204
87ffe584bc7a2dacb8b8c2f5b3ae378d5ed5df12
48235 F20110114_AABGPN onstine_a_Page_133.pro
0873f21509ec34d7aa0f105f9f1b67c3
98755fd397190f86631fae9e990dc4c1fc2a9d54
47241 F20110114_AABGOY onstine_a_Page_118.pro
358c741df43949763ec766b03e6556a4
4ae23767aff4ad83904536e1cc61ef80c3a5fb8f
47538 F20110114_AABGQC onstine_a_Page_148.pro
ad219ca6711638c612c5de5be621973a
0cbeb951d1b6cb5a619f2f90054a51b4254383e6
24664 F20110114_AABGPO onstine_a_Page_134.pro
19a69a0e1b0ca7b9368015d65d53c41f
bfa89b135d25034055113caa797b2690127bc1a9
49944 F20110114_AABGOZ onstine_a_Page_119.pro
f1cc8fa514bbe5bdc5c981fead9f75a1
d25ae243c921a3da2aa46ffb24c3e1cbd314307f
43332 F20110114_AABGQD onstine_a_Page_149.pro
11a11fb13f144bbc175b70b948b22e60
fb941d1890cc0ad0a49277ee73b7682b13ff62f2
11946 F20110114_AABGPP onstine_a_Page_135.pro
e012990472ac5d62a6ea82e0044668ce
947ae7cdf76119e9854627acf5008fd4dabd214c
32733 F20110114_AABGQE onstine_a_Page_150.pro
611ca153ecc613081282abeba7b83c32
69d03c725c13163ccebe68a3bb505143ff2fb4b7
21482 F20110114_AABGPQ onstine_a_Page_136.pro
39bbe329a009c523544b55ab18862b0b
d0cf2299d0baf568df1c9fc4dfa222f58d5c8759
41678 F20110114_AABGQF onstine_a_Page_151.pro
75ee1e54fa1dafbffd83beca156f7c42
68d8dfe89b3623af8bc85b113a3500daa6c0bd37
14600 F20110114_AABGPR onstine_a_Page_137.pro
b9b19fcc05a35f6927700d76437224b7
6bd60cbe1d85534edd56455c68974c84655588bd
24189 F20110114_AABFNE onstine_a_Page_133thm.jpg
c5f34c073855af0134203650b1f48567
6d8b34bb990858fd129c4ccb1c09efc644cf050f
46734 F20110114_AABGQG onstine_a_Page_152.pro
962d6ce8131ca2232ce38fa48344e42e
9a038f79a262d9f2db9986c50774a625ff7621e6
9183 F20110114_AABGPS onstine_a_Page_138.pro
20bc3760ca6d29d6ef4d001c25af408f
e11e4f9c4b98f9bf64eea23b8d5042af0222573a
28697 F20110114_AABFNF onstine_a_Page_103.QC.jpg
bf612678567d68bc42f9aa1338456994
a594bcfe947488bd1396456487744bd67c90168b
31173 F20110114_AABGQH onstine_a_Page_153.pro
90c0a09d3b3993c61a56c0f27a0a40eb
48c2884e4fd22fea6a5f0c98304acc0c4f69ec7a
208528 F20110114_AABFNG onstine_a_Page_045.jpg
177625c51e417792284cd61da0af6ca2
99c177582f6288cc19c1449ee5e20f21424790ff
46527 F20110114_AABGQI onstine_a_Page_154.pro
d0b8f46826d6c49ce34de85b72e831ce
30a85ae4ff76646da5a9b785c728a5956431dfb6
5898 F20110114_AABGPT onstine_a_Page_139.pro
1e7220d0a79b27158858c146532a684d
97c87ceeb1d6a15050db21027fd1a4b19c117258
22278 F20110114_AABFNH onstine_a_Page_152thm.jpg
78cdfb51d85926d4b8718c5a25e3e142
3b02abd159a43fd1781ec19f3764812427f5fcfd
43135 F20110114_AABGQJ onstine_a_Page_155.pro
84ba3d9df04db1b914b58ba2f3f0cf7c
50f424f0fd7eb11f3ac3523e80c46579689d7bfb
8288 F20110114_AABGPU onstine_a_Page_140.pro
3ae4359f64744a36e26812ecc66f5ea1
eae3ebbd2a940c9e2e8e3ddbe64c165d5db95d99
53153 F20110114_AABFNI onstine_a_Page_175.pro
a1dd31eda5002702ff7207fe2a9a85ee
33fbe7d6ed973aea91db5a4bf4eb4cc1748840f4
41605 F20110114_AABGQK onstine_a_Page_156.pro
eee67635ad7083bffee14abb1465e99b
0b2c9c4ec81c94f3351cda3dff5179863f1dbc04
7218 F20110114_AABGPV onstine_a_Page_141.pro
43ff589ed0042f1d76c2520c836c866f
0c324e3582d3e85a14bfdb5736994f7702a5f68b
47477 F20110114_AABFNJ onstine_a_Page_037.pro
803ef5f8c6fad06d0e3291078fe3ee71
17429580d9bb0dfc2f5adc687e6ecbae4c7958d3
38846 F20110114_AABGQL onstine_a_Page_157.pro
d320dc696f9f285aa9be296422515e1b
be4e19ffa8aea2ff4c9a60ab6b8770d6d53dd361
5602 F20110114_AABGPW onstine_a_Page_142.pro
238118db9338eb2f03641f062ae36623
b68ef164e112cd4bacfd145b2475459108dfac6d
8423998 F20110114_AABFNK onstine_a_Page_145.tif
d14801fc54014eca328f927714c7a36f
1b3b07d483e265a09cfc1aae616705448c370c13
48361 F20110114_AABGQM onstine_a_Page_158.pro
d05dd70472e1530496300ac7d580b21e
725e1d212d25b5cfa7924189dbe17fc793d1f422
6355 F20110114_AABGPX onstine_a_Page_143.pro
180d52449c8f046a3ae6bf3a90082545
d90f917fa552f5f70f80b38fd1576041519ea7c7
59141 F20110114_AABGRA onstine_a_Page_176.pro
c0f1653e88469bbfd65f71f7bee5d917
db6b85956fc0bf77d300366b9de4afc1668cd47a
47625 F20110114_AABFNL onstine_a_Page_036.pro
247784917d73ad96cee25daeb7870037
dc150829cf09efcfe53528c77bfd547587bd579f
41485 F20110114_AABGQN onstine_a_Page_159.pro
2259ea573f7be661039508ac7201c326
614f42851c624a7a748d736099acd2104b8cc981
4370 F20110114_AABGPY onstine_a_Page_144.pro
20ac43a0930a686059f9929eb722bb19
dc00d80184b32781e55e0910bee6740e4ae81f04
55319 F20110114_AABGRB onstine_a_Page_177.pro
8cad1eb89b05d5224bef5f76fa8cf16f
0aa8c6fe5ce6a8e1ffd78ba7580c7899f34ca903
2677 F20110114_AABFNM onstine_a_Page_100.pro
7e4414f1ec41dcb2b54b104e028ba31b
f732cfa7023190671dd4293d478f245cb116909c
34690 F20110114_AABGQO onstine_a_Page_160.pro
fa4ce72e8e35f193cd6243496853b564
e4386c70b4ac621db1b49befff5d622793a7e5ff
6031 F20110114_AABGPZ onstine_a_Page_145.pro
fecdd9f963178c5e31b6740886979edd
7fb2adca138e5096efaaff7cc6c31909eb633550
25271604 F20110114_AABFOA onstine_a_Page_011.tif
823127801581f9287e36143433f1d0c3
cee2ea40e378a088bf8ee7ca2e71232d3ede7007
48014 F20110114_AABGRC onstine_a_Page_178.pro
f058c74a5d7e0306749b8ff85b02ef72
e9b9ed165de616f6ec90de147b640309cf91f1f9
1053954 F20110114_AABFNN onstine_a_Page_004.tif
fa0f53fd0c6e873a51b2ef00135d607e
0daa6b1f56b8cb52816d8c1403697f49a3d176fc
18712 F20110114_AABGQP onstine_a_Page_161.pro
741c3daf49a20758f8f5f0ded80822e4
3e42a6a4040f9c00c5b96a149f01b3c6d9ac866e
6244 F20110114_AABFOB onstine_a_Page_097thm.jpg
728ce0d4f799026f90a357d3cd73b1bf
4173779a442e346f0cf44afd8d906cb3692e2e16
5396 F20110114_AABGRD onstine_a_Page_179.pro
eca1d89093bcadce420a037b69ad10f9
250a0563015e3e8c8c9587b7a45352f91852c912
1752 F20110114_AABFNO onstine_a_Page_149.txt
b06c63b79e24e3b6480f9c64e429f1ee
31ce368aa9362b8b70252a26745007cce8c78e18
2732 F20110114_AABGQQ onstine_a_Page_162.pro
80349da87fd7ab8105c3bca0141f25cb
17909053da64941587680cee1bf970768acca776
59870 F20110114_AABFOC onstine_a_Page_092.jp2
26def1c4dc703f61d3369f8be7c57777
e0b35da30267377bd4b2d426d178feef1bf1d0d5
16892 F20110114_AABGRE onstine_a_Page_180.pro
cd1e8155ae9f2d0dffb630e8f943205c
3be2f33c67ca2a54da0b749bac8735f2278d1f14
24386 F20110114_AABFNP onstine_a_Page_166thm.jpg
9f3ac589b38b64c4b49658199b2549b6
a3848c63057aa8953002bf0d53a7976d4c3748c5
5288 F20110114_AABGQR onstine_a_Page_164.pro
ae5b1f27a089c75fa537faede1755ebf
be70f57ad3a9defda021d7f350179fd3e92d94f5
F20110114_AABFOD onstine_a_Page_068.tif
47d52285264ea7f6d57cc3f77f021b0e
a4dac47a2a44bef0a868ea0b4f03b6d1760dc135
457 F20110114_AABGRF onstine_a_Page_001.txt
f64b27ac9a82da60b92a9e1d560d6244
961f4f549a3aa85db6082109a02e0cd4a1965644
80735 F20110114_AABFNQ onstine_a_Page_161.jpg
4860dcef15d03dc1d0f9e415c44a51de
ad173c870ac634be252d8d611a843a396f86ed4e
42569 F20110114_AABGQS onstine_a_Page_165.pro
3b135b8c27b33740400bc6f909f1144d
cb04aa00ad732f6fe360b935c3d37559f465251e
423 F20110114_AABFOE onstine_a_Page_138.txt
ae47fd40c481d5b8e2c34eb91b95d1f1
a7797629f6aa39881e4b9e92ffd0a3c364eb0eb6
113 F20110114_AABGRG onstine_a_Page_002.txt
1dd6a6296598c78da9c36eadf33f13bf
ceb9b4c0a4ceb0117f464d01f324bf838380a171
36553 F20110114_AABFNR onstine_a_Page_113thm.jpg
e6fc0d5851c7ddee10e7a19d9283d437
0c2935def7a2a45613f9ac88865a7e44607f5c23
48312 F20110114_AABGQT onstine_a_Page_166.pro
6ac1557bbe96e910f5569d421dd97343
625f811f173ae16781737a80fdf89c77a239de36
85870 F20110114_AABFOF onstine_a_Page_024.QC.jpg
786609e959cfdb723e69e9180147a628
e3224c97195fdff8c8f0ecf4f1942ee0ce6f7fd4
448 F20110114_AABGRH onstine_a_Page_003.txt
1d9ed3ed3b07d056e530dbf121949176
faa126708dbc5d792ea72f0d5f4baabf27119df2
65655 F20110114_AABFOG onstine_a_Page_116.QC.jpg
7f9b7bc8647e754f706b693e5494c12e
7a69cc1108104fe2950170eaeac1351af1a0f0ad
1790 F20110114_AABGRI onstine_a_Page_004.txt
c995c9d5d53969fdb2e17341068e9e02
100bc22bd09c87de62b3dde1f680b006eba3a55a
96574 F20110114_AABFNS onstine_a_Page_033.jpg
4c19c8cc0f8938556e8d834239094934
509413038a5ac72afa5a020106c92369599cc657
5160 F20110114_AABGQU onstine_a_Page_168.pro
abed502905a9ff59218d99529f040e2c
f39d560e08ab625d08cbf81f8b847be5c2ea79a7
5411 F20110114_AABFOH onstine_a_Page_098.pro
d686a8f0e3fabf6dd7ccb851ea58a00d
cb883115f764183f7b161a8350532a25a3007d09
3332 F20110114_AABGRJ onstine_a_Page_006.txt
c61b7c8bd54882137fc914c047aee300
62ec621862ed9d8b59b1858a4f75739ad119fc72
3848 F20110114_AABGRK onstine_a_Page_007.txt
7f42c0561325af0e59381a29ac6162f7
67c35111584540616a41954dc5c3dd02adc3de61
1984 F20110114_AABFNT onstine_a_Page_119.txt
4a456f7bd22c60b7c81950fa755d882d
14379d5d97850602b677dee289f664ecb1f04d12
2272 F20110114_AABGQV onstine_a_Page_169.pro
563adebd2aa98533241acaaf49a990f8
fd2a21177aeb0bc8ce39b2d2c6969baba64f9370
264646 F20110114_AABFOI onstine_a_Page_006.jpg
f4d5ff7d89a18e04eecd428845d71c43
0d1be9813a8c2c073bc10b52143c4bdd6285d849
1235 F20110114_AABGRL onstine_a_Page_008.txt
55ac3637d8a52cae2c5d415c2c8a2c43
cdc16d8af86ef1522368e5029c3cf48afc76f7d9
14497 F20110114_AABFNU onstine_a_Page_058.pro
1ed9af99c8379516f5c4ecd6eb8fd481
937e115c4c098069c7274ba33969a0dc4bde56b9
6163 F20110114_AABGQW onstine_a_Page_170.pro
1b2ccb44a11b8e771e9bc6c52e178598
773f86fee6882e3ba3d3e558a77f726b5bffc31b
10949 F20110114_AABFOJ onstine_a_Page_161thm.jpg
d801ad0c47c9e9c2bba11f9159de36ef
e5f60d263c2bb7e42ab23048ca7a86a159c56cd2
1917 F20110114_AABGSA onstine_a_Page_025.txt
72dd071720a78f32645dd4b5d9053247
2c62cc61d6034dec72c3aa7a7b3f5a380925d763
955 F20110114_AABGRM onstine_a_Page_009.txt
88073e8aae53e54cbb67f912bc955553
e91835616ff461135b0f1e7baa570747e71a1170
53972 F20110114_AABFNV onstine_a_Page_011thm.jpg
31a94657d9e6cf10006848b47487a403
be409a6d9291bbe1d414d0fe725653f1cf4763f1
1753 F20110114_AABGQX onstine_a_Page_171.pro
0fd3cc47cb0400a7905560289ef22e13
0e5eedb6c7ca191b097cb5ef815bf6afabcaac74
54617 F20110114_AABFOK onstine_a_Page_123.QC.jpg
428bd96da4d27f353f29946afaf7141c
29751d2c9f006a176ac8a16302471ab3a04b71bb
1876 F20110114_AABGSB onstine_a_Page_026.txt
76070978ea2d649bfe1b31171f2f0af2
44505cbd30b0fd5941430055908241a9f6df69aa
2642 F20110114_AABGRN onstine_a_Page_011.txt
bf7e1d5e6d25b1b884cb79c5c15ef099
4bbc7f03ca2cf252528b680feb6ae717c97ce443
79171 F20110114_AABFNW onstine_a_Page_042.QC.jpg
b87aee3293fcda6732ffe577e5aa1d8c
6b7c1e53c5c95357f0670a18f7c990d6c0302b57
2108 F20110114_AABGQY onstine_a_Page_172.pro
705f0c87d3af651662ac2641ebc6d3f2
cb850171170ce350b43f6623d26c5d205b224e40
123825 F20110114_AABFOL onstine_a_Page_011.QC.jpg
2fc9845139635289bcdb22f4f54448d3
ec3e0dd1e7e4b6fc7146cd70aaa206138ca38796
1935 F20110114_AABGSC onstine_a_Page_027.txt
ac02f49b9cc8abed1103bf785657d010
9ee4c62badd7d18c28efe06bafeff374f1cf7c2e
2437 F20110114_AABGRO onstine_a_Page_012.txt
67883bb3325cce4cd30b2acf1fde7556
d7c30eaee4980086b8c2369a84c74dd6f35b0605
61352 F20110114_AABFNX onstine_a_Page_093.jpg
cd64c2160e7629224073aea71f18cd6b
6835de5496acdfa5393a89d66b7167c5e967de32
1796 F20110114_AABGQZ onstine_a_Page_173.pro
917d3ac5f9f17f7fe49820fef16fb41f
a8e77cd3838dbd4ae404cb7462e3840c75487a7e
324649 F20110114_AABFPA onstine_a_Page_012.jpg
5928c8d96e74e589209c176456d6f365
f83d00e7786e59083fec12bb3f245ad18d2c0c66
80891 F20110114_AABFOM onstine_a_Page_067.QC.jpg
62d399435ce5fae8c56b996b6b784569
98d4b369ed1bb52bc997d4fa8b23fcab8cdccd0c
1990 F20110114_AABGSD onstine_a_Page_028.txt
ab446a92c0f7f2d3da834fd649c61953
d5f656d1d4c71f3b02595a3e74c0c964627607ca
1268 F20110114_AABGRP onstine_a_Page_013.txt
5ad9bd0e4b54c4e1ade7fcd00785904f
a02c673a00a7dea8645beccfb2fa07f70deab931
93687 F20110114_AABFNY onstine_a_Page_100.jpg
788fb05fc7b4f37632d56f172d82ba5a
a5b580dfd0aa3a1d1c672b89fec763f6b151167b
8865 F20110114_AABFPB onstine_a_Page_060.pro
6c938712df088948e3abd068816777ff
92f93ede91842df6e264f35fd2f2c03bcc334597
44902 F20110114_AABFON onstine_a_Page_153.QC.jpg
7ce35939b0399210858cda4894c5bbe0
0c901d08772107d8232bb91a62fa5572fbb356e6
1793 F20110114_AABGSE onstine_a_Page_029.txt
80f6d9bf56dc26d2de776df5957c2039
2071cf9ec5f932699b565d39acf28a057f476b8d
1717 F20110114_AABGRQ onstine_a_Page_014.txt
4dead0e5accd46119239556f27a21f07
49c0d456ac394264dc7e4a4184a94325b5d421af
209121 F20110114_AABFNZ onstine_a_Page_027.jpg
fdb7f456e9ae26f8efdffbff55603f90
17bc6f903d93a615ed39199cd7ac426964020516
F20110114_AABFPC onstine_a_Page_081.tif
64ed986a5a3078dc564c13cefe4a0d7e
63c6c1743416e48a263f6c63f4afba7d65abc851
1944 F20110114_AABFOO onstine_a_Page_042.txt
5af52e68aaa4e6438076fceffabff45f
adeb6ef1f470c418448d25e175c8068a419fcf1b
564 F20110114_AABGSF onstine_a_Page_030.txt
bbc760982eb127a23465592c27a85edf
db3a4735f9c65f9cbd433a6ec745d1c58ac0f0aa
1632 F20110114_AABGRR onstine_a_Page_015.txt
7496d11c3bda2e82cf8afaba025b4cd6
5e0a84ee91b073bf5fdd10018e8eae429003e74b
F20110114_AABFPD onstine_a_Page_075.tif
71e150b30c06e8bd962078f8da530dfb
94ed0647f814831ca918c59be12a95090be198cf
377 F20110114_AABFOP onstine_a_Page_032.txt
124395f82ce68570175631a8a376a1bf
43490599ab2ac6d144f56ddd960b7b4a03fe47ff
663 F20110114_AABGSG onstine_a_Page_031.txt
a9a14450819a2e674ba099756457994d
6e69f0efade3d0441a94584364216683754f98c6
1823 F20110114_AABGRS onstine_a_Page_016.txt
fb317ff49fdcb210daf567c92db3934e
6747c01b662e08651e17695ea4627c3bf017e314
2277 F20110114_AABFPE onstine_a_Page_010.txt
1905fda6b30394d33b64d4ab8510e584
cf07e9218e34fed65fad214f4f5011caec544580
32268 F20110114_AABFOQ onstine_a_Page_050.QC.jpg
d4c5a26dbf3542b0e38cf0224a3c27bb
5670c615318a48e49f352e65cd80ed8826a19e60
606 F20110114_AABGSH onstine_a_Page_033.txt
7753e0cd305a1390a7a410aa7b8b117b
8d885699e3fbc515a8c81f24e7b7763df5d4b96e
1999 F20110114_AABGRT onstine_a_Page_017.txt
2aae843ec186d7a0c44c415bd522529e
5c2b2ede3e09d2711f8d1d2eacff4fefe0e39cac
1671 F20110114_AABFPF onstine_a_Page_160.txt
4a8339752482e4300a72ef00e4ec3d6d
d743c05eb3cfce3a2c9bbbdb131eb83efcfaac22
24316 F20110114_AABFOR onstine_a_Page_001.jp2
c712a506d2bca1f215a8fbeb7baf7f88
61b432c24cd0bd9b34d09bc5f75b5886c68e0174
1727 F20110114_AABGSI onstine_a_Page_034.txt
13d04420da29585de2915225d59237a2
d105bfff94137c376237908c23ddeb469d7f309e
2026 F20110114_AABGRU onstine_a_Page_018.txt
bc3b367cf5d0a46e7d064ecec6e2080f
76125fdb915453b1375d3ae04cd231e22119bb2a
75495 F20110114_AABFPG onstine_a_Page_038.QC.jpg
704f647795c5a587d6e0ee4a5d573f60
a6d37399b76582bdb82474ef793157fd38991918
46520 F20110114_AABFOS onstine_a_Page_173.QC.jpg
1dea6d9257dade3d5713da9136cf09d8
ea371166582787853642a2081d7933294c780062
2066 F20110114_AABGSJ onstine_a_Page_035.txt
71eba1a629ff37167ac7bad29372fc51
597f7fce8021ada9f4da94e8d1ed777ad4d7057d
21117 F20110114_AABFPH onstine_a_Page_155thm.jpg
13d071850cf0baf8f21be6d47917504d
38527ca3c65d48c8829aae3ba3e1a9e10d0eb10b
1882 F20110114_AABGSK onstine_a_Page_036.txt
fbe6fa7db0fe4fc521736ed46be4855b
6106dfd5afe8de93bcfe5b3c9422a3ef7ec24f65
207 F20110114_AABGRV onstine_a_Page_019.txt
43105a64115e158b27241fcbb3b28efc
7364e4832ef185be6e2685752812915e934103f8
37553 F20110114_AABFPI onstine_a_Page_077.jpg
147c7fc851889be25e6f82b3a3684d5d
246efef88dc10c29fe4cbcf64abd4aa3ffff44fc
35416 F20110114_AABFOT onstine_a_Page_072thm.jpg
e638140f4020da78790423835fdebc80
fead93cdf2810aec20c05947c7e75bfb5ee30585
F20110114_AABGSL onstine_a_Page_037.txt
4f29c13f525abc39d45523e23795d6b2
72ce1e28a453a94560083a4e246531a5c677c755
2010 F20110114_AABGRW onstine_a_Page_021.txt
4fc49d8776a4fd723ff6efe96372a5fc
9fcee89ffd23f2e2aecfac73d5acd83cad93d83e
89767 F20110114_AABFPJ onstine_a_Page_176.QC.jpg
e58d1f2ac70f33c8058bb3438c462726
9782252bd6744865a4fa801ed4007cae99f022bb
F20110114_AABFOU onstine_a_Page_164.tif
ff50f76021f22984f3f760819d9f151e
06470e18e340718b4f293038be409fc0684b9a9f
237 F20110114_AABGTA onstine_a_Page_053.txt
d1b3976562c76671ace2d277dbd7d351
8e94d6d6e5f129e5a3ab769ac1a29f2e179cb480
F20110114_AABGSM onstine_a_Page_038.txt
6b4b3a6aa7b4b5facdb4f47318137299
cb2de38b0a514b943badfd6454dd822935dcb574
1956 F20110114_AABGRX onstine_a_Page_022.txt
99925f37d8315bf4dc769c63e347d9e8
b1ccaa915c4074ce54a2727b61f462fffc74d526
108024 F20110114_AABFPK onstine_a_Page_064.jp2
dba5d6638c712324d2d075eab9e5e601
a6aae5b07e9bc1045d1747fcd2bf3eb44c939b3b
61818 F20110114_AABFOV onstine_a_Page_153.jp2
4c0c57540bb054dda96ca878dbb0702c
384af8b0025c1f541aae17fcb66618680938963c
279 F20110114_AABGTB onstine_a_Page_054.txt
1c6d22a86eb54cd013fea0b9e8eec04f
dee1d12a04ff3ec39980ac38310a9754f301f602
1779 F20110114_AABGSN onstine_a_Page_039.txt
8a9cbb25dbf3349c5a93170c8748d9dc
2a8d7e959a488b072da34519931f2197fb1a5545
1986 F20110114_AABGRY onstine_a_Page_023.txt
56dc5f8f00adeb5a65071b0dff84da72
97656717bd2f50a0c59e99c1a7e7f7580e3a389b
32133 F20110114_AABFPL onstine_a_Page_111.jp2
d8f3adea96f31136664c9fcd50b0a74c
654f2b98a59a011f9d397f65558a5f04a0cb617e
F20110114_AABFOW onstine_a_Page_003.tif
9154734b8daaf69a2ce0986dc005f683
4ba5f367ba3c4de2bfc9c847c13cded11294d878
F20110114_AABGTC onstine_a_Page_055.txt
1dc14747978e6fe05799aa143150a8dc
2df9c35666611ea10c9574d1ae33755cc6fc7ec2
1862 F20110114_AABGSO onstine_a_Page_040.txt
cec7d66cca50106173f2cb6dc9c25bec
18264862617559d197e633ec1d4ab5e6a7ab0e6d
2150 F20110114_AABGRZ onstine_a_Page_024.txt
217c1fad60c13eb8584fd6d389ea1e04
46688993027bceca3c5dde19c72b6d13ea15796a
75632 F20110114_AABFQA onstine_a_Page_117.QC.jpg
ba92a859f55bde0db21f1c4e7e172d17
0cc2537207177401001c48f7b31b367b08df6a4e
353 F20110114_AABFPM onstine_a_Page_143.txt
c9739114d1ec40ec34c1edbda241b6d9
b77718e1f1201ab5696e11fe24bf4f4d44b98036
215189 F20110114_AABFOX onstine_a_Page_018.jpg
3d6f512f037e2b8e06baf497b9de1b85
0b4406ea185b50e29778acf03ffc3b0c2a341024
314 F20110114_AABGTD onstine_a_Page_056.txt
844a130c94910996ec0d097327ba868a
12a067dec43c1f05a323e24a43536c2597b4a2fc
2086 F20110114_AABGSP onstine_a_Page_041.txt
f3ece7807c3ca3eaed57806bf6c71fd1
b78fa6846a1c2706c52804677abfa6cc5733a7d7
14061 F20110114_AABFPN onstine_a_Page_092thm.jpg
ac4fdbfa8d4865f286f1b4cf0a7c8d11
8de1a85b5c0de992a90f8d96ff35fff3d0201e8d
10438 F20110114_AABFOY onstine_a_Page_180thm.jpg
9c0b53b0432ac554bb0a9612b09f810b
e3602eb535e403958a7b9f5c18846e41859522a3
275346 F20110114_AABFQB onstine_a_Page_141.jp2
5349fdaab05626c829b4aa469dbdbe92
0215686fe11384b24b4ae491325252f408734006
404 F20110114_AABGTE onstine_a_Page_057.txt
2eed926b1aa32257f384cc64e58c1c6b
fa4804baca3a10f78521b4f4f6ab22f446458ac2
2068 F20110114_AABGSQ onstine_a_Page_043.txt
fc844711e2ad5bd3571fc069867c1127
2be410a5e59f0bf214839f7b7844a598bff4cbf1
369 F20110114_AABFPO onstine_a_Page_122.txt
b315e3584fa3975bbb31f826b0d8748b
cd002a8a31e259d37dbc2de0002a066239c50f3d
74900 F20110114_AABFOZ onstine_a_Page_138.jpg
b33c20bba4c3a7dd05721f104f63d2a5
85ecc8d2de9a9379430e9fa00973e77769c4186a
F20110114_AABFQC onstine_a_Page_161.tif
4349bdff377ced48e24bd5744d1bd0c7
ffbb26fdf0d559446a022c475da04249f79f1d70
661 F20110114_AABGTF onstine_a_Page_058.txt
5cd900f5a1afa3106ba590ff46f0473f
484de258e076dad7aed4bbbd88b7679fb1cc7cb2
2037 F20110114_AABGSR onstine_a_Page_044.txt
e215b6391d708d5506e16c4b7353f774
941908584642a9b36497cea6f5f85f750ed3f28c
2266 F20110114_AABFPP onstine_a_Page_177.txt
349a30c1109c335315dc548eb0141625
5a886466139e68bff42e455e97568d2c3925ec11
15862 F20110114_AABFQD onstine_a_Page_153thm.jpg
334d22302436a79bc68bbd6fd86e6263
0392f61a9fc96c9427233909cb470ed635c2ced8
309 F20110114_AABGTG onstine_a_Page_059.txt
1b8ec9cc7b78feeca32d0a93f0af7975
a34841d17de5c4ff255d81315d3096ab70f1f4e4
1963 F20110114_AABGSS onstine_a_Page_045.txt
2ce4fd4bdbbdb19eaaa0b96d63be690a
0074b425e83d9f41916afc68b8e6f16d14c25b56
502 F20110114_AABFPQ onstine_a_Page_081.txt
ae937c14621c1fb22b42c97d6ef6b4f9
55a0aedbc01908a6063cf21303194cfcda36f4cd
13986 F20110114_AABFQE onstine_a_Page_047.QC.jpg
62739b141005c974c98dcac57331c8f5
d1c47a1cb9ce61d461edd2109b63454517384a08
656 F20110114_AABGTH onstine_a_Page_060.txt
869764e4c03b04b52048bbc6d9a312c1
11ce72d57efb846da85fa716f3942217705aa3e3
1969 F20110114_AABGST onstine_a_Page_046.txt
095280ee129dfd961ae714e344a3c552
467c00746af608300ae0fc298fcfffa741a2fee9
1051933 F20110114_AABFPR onstine_a_Page_006.jp2
bb1ea00f885f892f3ab0742472034d99
543e3ab16a59a08fdcf9aaf25ee6854336d6e50c
361 F20110114_AABFQF onstine_a_Page_163.txt
92d7c4dd7299c7ecda88026b1301aeba
ba5341cfd3f3c4d8b69a4a96be988c8b99c7e7e7
335 F20110114_AABGTI onstine_a_Page_061.txt
b04a9def23cb8d002281cf5237e6326e
6b5800f7cab52739cf8fddf9816d6d5c73fac9d1
302 F20110114_AABGSU onstine_a_Page_047.txt
8332bedb11e8689991cb8b48bd97af1d
923fdc9e1625006a7d89a66dd3fd862b06372a82
22805 F20110114_AABFPS onstine_a_Page_154thm.jpg
98c444697db3df5d7a7e0b646258ea0f
0240a6cbb3ef537eeb08e40e682b5f6f8e883bb4
17265 F20110114_AABFQG onstine_a_Page_160thm.jpg
4d5e359d3645058f60aeae8747ba0649
094db5461e4c9ed607016a68cdc32fa42bccea93
280 F20110114_AABGTJ onstine_a_Page_062.txt
33d616a02d16d764a2c0156567baf5ce
e234a3b457eb5f132d7bb6a307c1b4cc11f27578
221 F20110114_AABGSV onstine_a_Page_048.txt
646017159f62c789e223c68a7a0becdb
79f74bafb62454da7d57eec98a3e3bb21fc1a2e8
105304 F20110114_AABFPT onstine_a_Page_025.jp2
5e0fc2ec12f08c5f8f159df36c3c57ee
20944e2761ba29215042970c102e9d5cf9a8c39d
215070 F20110114_AABFQH onstine_a_Page_164.jp2
c629e3426777d6b4bc5b3147bbdd78d4
1f0e868856171b3a5078c90b7b7e3f4d6a7d7acc
F20110114_AABGTK onstine_a_Page_063.txt
4b791b9ab7f5436e592c9a87a14968b2
a7bc893f9501c95838c0e05d14ae254f2684a581
46216 F20110114_AABFQI onstine_a_Page_174.pro
bba4c1580cb40fd62ec9a857cc019d4b
d5e05f3d1c23d4d1d613a7fe967afa0e27b69008
F20110114_AABGTL onstine_a_Page_064.txt
c0fd7c227f9d23eb5a000bfb7f8dcc4d
0c50323cb74c57fb5b7fd7a44d9e1f9ccc1d756b
190 F20110114_AABGSW onstine_a_Page_049.txt
ee2946e13b2dff2581464066297c3fa7
1afe3df6618d47928b0a505ec66b2fd7736915b2
14829 F20110114_AABFPU onstine_a_Page_142thm.jpg
e3de275b9f936427042df8632500803b
39d7bbf8beda92ff7be004e4f157503dde63cdf0
8020 F20110114_AABFQJ onstine_a_Page_093thm.jpg
e8c0871948f8708992e57264c6ce4f49
22eb0324ca5b98224a1de07314effbc36c45fd80
337 F20110114_AABGUA onstine_a_Page_080.txt
f9e40bbd9db49e910b0d8fc1f6d1c7fe
7da2e08f59ff5d23cb2e0ce16ac24dd0f9f22a90
1966 F20110114_AABGTM onstine_a_Page_065.txt
de610c2175792721cb5ec54d86353f12
60594860a226eb1cfe09319cfb4bb2db91bc5e03
197 F20110114_AABGSX onstine_a_Page_050.txt
ed62971f33dbb2934493b29c52c38ce6
e5b1db8ed7e7594deeeaf0ea3e2d68cd7a4042da
35619 F20110114_AABFPV onstine_a_Page_074thm.jpg
5f5d7b268290794b12e031b569e9b82a
541c891d143c605463db9bc83bae1ff286499c00
12388 F20110114_AABFQK onstine_a_Page_102thm.jpg
29e83c453939848bcdc33c15eb9a9360
e7e6b1c0275232afa0ed87ca68a532b1c23e0810
760 F20110114_AABGUB onstine_a_Page_082.txt
0757181324599e4747d5c2b9277b6d27
b03ca9833974c804db5ef4de98f1a18739625b5e
2006 F20110114_AABGTN onstine_a_Page_066.txt
1f834421fc0300323eed7eb1acc5e233
851ca0751edb7a351ee70e0c00f6823956565136
1295 F20110114_AABGSY onstine_a_Page_051.txt
9fab77bcdb0bc7369e0a0ebdf2652e74
50314174aee9573ad64180beae64918e4af8abb5
10880 F20110114_AABFPW onstine_a_Page_019.QC.jpg
09ed427092b95727db5f6aac8e32257a
d9b2000ea4ffb4e0983397c6caa0b7990b4ebe94
37343 F20110114_AABFQL onstine_a_Page_071thm.jpg
a923bd630f4cfb376ffc65dc3b77b04d
5b1060c654bf99f64294fd449bafc5c7c2789bfd
112 F20110114_AABGUC onstine_a_Page_083.txt
f7772f25209e9734c20064d44fdd8535
4b49e16519fbcfea1bd3a3e30864089f5359d7ca
1989 F20110114_AABGTO onstine_a_Page_067.txt
ef1bd44e28c1ae3f76ed60edece77351
761821ad47cfa83927246132c4731fe09f3b330c
617 F20110114_AABGSZ onstine_a_Page_052.txt
3e9cecd8b60bbcd6d05d0eb369c00d04
5d08b3cff9698adf9b49c639f5b78c42538c9161
15853 F20110114_AABFPX onstine_a_Page_057thm.jpg
14795e5f7fe449bccb0b43882deace9c
09c3208e90f4222a35d7bc318b541d89f1cc0820
111543 F20110114_AABFRA onstine_a_Page_044.jp2
5dc022f1aa561f74972418f6a89dfcfc
a736274b4ac068438163b91a7f74ebbbda7c7931
F20110114_AABFQM onstine_a_Page_002.tif
0fa789098945f308d069941bb98d8e85
c0aba9af5d92903cf36594bd1c88540ab7c9552c
91 F20110114_AABGUD onstine_a_Page_084.txt
c069f7b0df7d85ff91f38442c2f28042
e7259d737029c7d5fd19337cb9261f99324310f4
1474 F20110114_AABGTP onstine_a_Page_068.txt
6de9073a72873480c7add5dafb930de4
6dae29f45ddabd9fa9194ea826ca80531c3739ef
62801 F20110114_AABFPY onstine_a_Page_115.QC.jpg
8c5de429e2ac371aacd87311275313b1
be97766f4330ec5eac492bd7d552c2c128af6f46
719 F20110114_AABFRB onstine_a_Page_005.txt
e6f93e6b81eed7c2cce4afce940bbbb7
5a1af26cf365a41f97e3ae20f51997c547091173
66623 F20110114_AABFQN onstine_a_Page_031.jpg
9d09c95fece86186674e3a9e437a5139
e8e2b4b76f5f616d2631337bc46455638cdcd487
640 F20110114_AABGUE onstine_a_Page_085.txt
b6e9b777104079f05a1ce1c38c285de8
479fae95255f962a2035523fd2841694c6a1e121
152 F20110114_AABGTQ onstine_a_Page_070.txt
2bc139c8072a231180241b5b064fb740
3b565d3254a88e439a77de77132b13cc8945f8f2
7867410 F20110114_AABFPZ onstine_a.pdf
404e9df81fe3dda4af8522355aa7756a
45d1bef58f43671161b5d57049b496f939da8f90
31101 F20110114_AABFRC onstine_a_Page_167.pro
07496ac002ab59fefd897c369d545d01
7eb1fae367dd42b1dfaeeae08b633cd9b8790422
1031035 F20110114_AABFQO onstine_a_Page_128.jp2
07af030ddcf835cbb31a75dd84f8c92e
b56b32a74f31b50a26a670723d3af5ddb5ebaaf8
78137 F20110114_AABHAA onstine_a_Page_090.QC.jpg
d45fc28f1f21cdf4b5c9dc9906e884bb
081fb06d2add72c0ec9a705e798638546258dc7f
851 F20110114_AABGUF onstine_a_Page_086.txt
377af8085818d77e97ee3d7aec7ba515
30cccd97300dc24628325f05781b5d1dd364b2b8
327 F20110114_AABGTR onstine_a_Page_071.txt
1b0ec63195a8636f295f02c3b9289ef2
ee2870ed89232bf7e2011e11c7c9d2eef14f2873
641834 F20110114_AABFRD onstine_a_Page_061.jp2
ba2a09f3d2f3c44324356e6d66265a5d
6c5523f9c605febababab514f83029fdde8befa6
5908 F20110114_AABFQP onstine_a_Page_163.pro
30e3c5d7250485b034eebd3c18b9745c
f9f78dd96afbe9dc40da649ff6583c34067beb9a
34260 F20110114_AABHAB onstine_a_Page_172thm.jpg
861fac7d7d8b9d3f57130d0096b8ea11
9bd88ac73557a6ea122646a8b33146ac3558b6a2
1820 F20110114_AABGUG onstine_a_Page_087.txt
bec7e4b88153e123a1ab7669491be9a5
958018c0576defd9ede8f02c716d17aca63d88e7
276 F20110114_AABGTS onstine_a_Page_072.txt
467786cd7558e3c99fcba46edb1cbd02
3cff8c1c9de92c6ffab72ea878d2373b54b5ae28
17578 F20110114_AABFRE onstine_a_Page_051thm.jpg
fcbc2c8b651a950f64265871a43998f9
03eef16979bc8157810a043d7eb244470eee09b1
272 F20110114_AABFQQ onstine_a_Page_168.txt
7db44211bb520255e6539c93986189b9
a7ed0e5510db616f0bb1d944f0718164ef483d13
16484 F20110114_AABHAC onstine_a_Page_114thm.jpg
38efe9b7e7aa63fb92e77120df308b5a
822879b8eb16d98bb30fb146f85268ea7b256db1
2111 F20110114_AABGUH onstine_a_Page_088.txt
7c2cf97ea4d7abd063db3470cdbac0fe
7fb595643a703b26d4fc98a26225e0ee9e058947
269 F20110114_AABGTT onstine_a_Page_073.txt
86c01cfd11ad262d60526748c16150c0
cca05d760d43e8c2a723a4aea9e33571fdffa972
60997 F20110114_AABFRF onstine_a_Page_129.QC.jpg
c7816e428132d02468971f6980b56328
01fe04106c168ff5819dd61ea05ab3fe119912ac
1695 F20110114_AABFQR onstine_a_Page_020.txt
3ab212f9a0095600271bc6b01823e1c9
0b3c5280af0d24dae93d8344287e15b1a5502d73
1847 F20110114_AABGUI onstine_a_Page_089.txt
e9c6d3de96ffe0494119368ea8e28826
d0414c2a2e50a38e8a11a68c0b83d866790cb442
297 F20110114_AABGTU onstine_a_Page_074.txt
1f7e411fca32efbe86b02921befbcfdc
a397b072172be09c44662ab70d489ed66ed014a8
418 F20110114_AABFRG onstine_a_Page_113.txt
477b819e92077ac5be80e8a2edd38c1c
996b5a31adc7510cc47779f1ebe3b9b83a14d57c
996 F20110114_AABFQS onstine_a_Page_134.txt
4fbdc8d688a3912845597ef4b978a8a5
9ce4623deff3c3bfa9607de2b7daab8565161b79
21374 F20110114_AABHAD onstine_a_Page_139.QC.jpg
c293a2126df669dd4afe828c09962c96
0628293a752bb0ef0c9fbd79319247467ba18cc6
1988 F20110114_AABGUJ onstine_a_Page_090.txt
a7d96b0c499721c9ba0768043a5843f6
e8d39ed17d92c071eaa9f45513ded9214b7dc9c7
230 F20110114_AABGTV onstine_a_Page_075.txt
72726a9cd661d23dd152d8fc2be79cc4
efc5b9d67496271bdc4d9852cd485b8b31c0f159
42267 F20110114_AABFRH onstine_a_Page_016.pro
47e552b83f7c77d482e3514d8533a338
8fb00bb844af069c36fe8f4b80181cf5f6a2bba7
293250 F20110114_AABFQT onstine_a_Page_050.jp2
a3f8c696db424776c0bcb119722c95be
171c8ac82c1fd3dc3c3bb0e11d55ee242027d95a
23769 F20110114_AABHAE onstine_a_Page_025thm.jpg
9476ca51015264161a6d83f8bfb64ae9
be845b01abb4e1d8b3be403cc7fc3366583cbb4f
1883 F20110114_AABGUK onstine_a_Page_091.txt
d350687d80885e2681a5cedc37540b52
6b7c978ef27f99e0e3408aaeb887bf7e15c6a46c
360 F20110114_AABGTW onstine_a_Page_076.txt
a34168ae5a0f0d6f03a7122144c08b10
42ab392fb0b86101460115e848d112f6eb0605ee
120593 F20110114_AABFRI onstine_a_Page_052.jpg
86c26ebb92088706f88e20248e166dd3
b8700afac76fe67db36a2f144181f86780a809f7
251452 F20110114_AABFQU onstine_a_Page_143.jp2
42427e116f8930a19d7fc2f59fe64efb
cb8eede941956f9f4f6aefafb94763cacc857bb1
25490 F20110114_AABHAF onstine_a_Page_090thm.jpg
66f018d6e4c38fcedf4add3dc2df8c54
13fc583d726fabb48d826d05979284ebef84f132
1033 F20110114_AABGUL onstine_a_Page_092.txt
5546050da8c858bd23157c286362e39f
a3351488b5c2530903210c43eedaa827d8bfedb7
207761 F20110114_AABFRJ UFE0006644_00001.mets FULL
95cadc279f14d6f6153eac3975e981db
bd50a88618615fba526dbacddb385ba56024bd43
42380 F20110114_AABHAG onstine_a_Page_013thm.jpg
1110b1a1391629fba13a76f9b2b9a6e4
7deaa1f84759d63fd2c4871a8811d79e9711d5e0
600 F20110114_AABGVA onstine_a_Page_107.txt
c2f06401d0c6b023f48a8470c4170ad6
fd9d591c6414d0c4d430a3644e35db96f0c95d67
693 F20110114_AABGUM onstine_a_Page_093.txt
2835f7ccdee5308450fd8b78ad6eef13
6a665a06eacc7dee8204b87d16459c2cdb748c8f
552 F20110114_AABGTX onstine_a_Page_077.txt
ef1414bd43653bf9aa2ac056cdde3b1e
f7cec0f6fea6fafbe6db698046d2b47dfbd67846
6947 F20110114_AABFQV onstine_a_Page_080.pro
0f78b7ba1671418f1c7278a6cd93151c
0ee9192c635afb09e0db9d8690bb9e599e502d84
16863 F20110114_AABHAH onstine_a_Page_164thm.jpg
5c2db1d12e7a0d20e330e32521370a4c
3a5d9e3bf42ab9838d8d2461036cff1dcad2ac1b
464 F20110114_AABGVB onstine_a_Page_108.txt
4f08dc4c0c3eeb5dfd49e82c994cf389
2d1a1f8612f9be4c6e107c91171f29ce4cc5f06f
751 F20110114_AABGUN onstine_a_Page_094.txt
f9c2e15078712ae1742d07875c60d442
01320f70d77dbb21a8b7d654337a09a9b295b6a8
286 F20110114_AABGTY onstine_a_Page_078.txt
7cb0bb43a7a0c300433725850ca1736e
32dc2b86acc91e9608b9417d1668841043df8424
F20110114_AABFQW onstine_a_Page_102.tif
98ac1635a120369b518202df021c13fd
e3ea35b2193acda8d58f69b533f0781c5a46f331
23926 F20110114_AABHAI onstine_a_Page_046thm.jpg
b974202fe1665d26d85e363d1c4b09e2
85e55f6e019cec5ab6283894d5d441bec3a7ce36
452 F20110114_AABGVC onstine_a_Page_109.txt
ed9e60c0127cbe396abbae80718b0258
79be2e367ff4160b027977dc45ff52d870080c0a
559 F20110114_AABGUO onstine_a_Page_095.txt
67a9fba65c9ae706bb22fe2a3dc663f2
1c408394c3c477ce17378b8da9928e4df34c41d2
1020 F20110114_AABGTZ onstine_a_Page_079.txt
03064d270367a5b354d34a5b62572333
4c094c4fd681a60411179316c2bfc728aa06a735
212122 F20110114_AABFSA onstine_a_Page_017.jpg
f72691f40e5dfb3175cce20656e1fd94
be095973013d4b0369d9ac780213c48e9a91180d
54161 F20110114_AABFRM onstine_a_Page_001.jpg
8acd2a276dc5a7a7799150846f32a16e
7dc3f70df23c2a4d5a78682d9ea97cdac19a0cff
652 F20110114_AABFQX onstine_a_Page_069.txt
651a16f92b841251df83aaaf88ff1c87
50d08c512ef843cc0a8e5f67562c0b70de3d085a
35071 F20110114_AABHAJ onstine_a_Page_061thm.jpg
d8508c4302c353bb72d156974c9b1538
37f505070dec3c3eb93f44b31b3e25b68dab29b4
459 F20110114_AABGVD onstine_a_Page_110.txt
d6da34d6d3107d342cfbe0d9b9e4edd7
514f9188d3c579caa47388468fad07c64df81644
167 F20110114_AABGUP onstine_a_Page_096.txt
1e085dd70e72020b59f8d21899a5ac6f
92f1ae000ee17d9eb87cb5c23748283f4f883c5b
25046 F20110114_AABFSB onstine_a_Page_019.jpg
47d2b0e8043df462b56180ceb93d038a
438953a45bcf39619f6a0628bce6981a3fd820eb
14871 F20110114_AABFRN onstine_a_Page_002.jpg
74afc2400de8b0ba6924c317070ce29a
2e12f21102e2c227998e018dc58063694ba53178
114599 F20110114_AABFQY onstine_a_Page_134.jpg
5cf6becd2c2a0495ee67ce0c9c1440bd
c3f37c5bc91f4e9d5f60b1318030a541841c1a04
38541 F20110114_AABHAK onstine_a_Page_085thm.jpg
301798b719d90f8f184d9fbde9d35f14
e3d8caacf292f38c3326d37c546715d7139532d8
478 F20110114_AABGVE onstine_a_Page_111.txt
12bf558c7d2faef88a28ca79782440c8
c7b4e69e8c673a6b9ecfba35bbf3c13bae573e24
F20110114_AABGUQ onstine_a_Page_097.txt
73736b7ab4ddbd8a6fcedb80567cd3a1
e1803f2bffcf6563e82d3c4625bf30beb4b98640
158046 F20110114_AABFSC onstine_a_Page_020.jpg
cd6c800d101d58b852200a671e86e4dc
4d036d2e2946359ca5e5b15183951e4b20e344ab
50581 F20110114_AABFRO onstine_a_Page_003.jpg
350e785169c3768c26c6a32c82d0ff80
7b4c202410aa66f0efd56caee701505ec6188cf1
17011 F20110114_AABFQZ onstine_a_Page_150thm.jpg
49092373513be7954ebcadf744961cf5
81b736395e7f481ca2577417d0f68220b19da683
59911 F20110114_AABHBA onstine_a_Page_113.QC.jpg
871186842f8898ed55f827b3d691fa82
48352f5eb3f0c2d425cf030c81a41ac7519bae34
21542 F20110114_AABHAL onstine_a_Page_116thm.jpg
8f5f14cf39cb093433d1a0184411ec95
214635a6864209601948f9a4886e9fe94f025a98
665 F20110114_AABGVF onstine_a_Page_112.txt
bb93369e00365c0e9d2a5fa311051d8c
b5e990694c26ca7abcc6fc8e3e562b176ce7c885
278 F20110114_AABGUR onstine_a_Page_098.txt
79d6438cfef8faa3aeb6a13667308a5d
cb4919aa26b866a231b74b6f2856c79fdca5b892
194874 F20110114_AABFSD onstine_a_Page_021.jpg
3efd0216f87bcf7c13893eb43e9d78e6
bcf1c562c74c40f86bc605f96ea2673c43db5ddf
183975 F20110114_AABFRP onstine_a_Page_004.jpg
9243a094d03a7fdabb65e3b336b0e3dd
58addf67b358d082d4e80cc3a61ee1b98063bf1f
23242 F20110114_AABHBB onstine_a_Page_022thm.jpg
fad2e51f4b6fd0f0deda610a31a1b6d7
7f9cb710d322409df15379033e68c3e8fe0546b4
55292 F20110114_AABHAM onstine_a_Page_074.QC.jpg
71f7615842168599afb10c07460d55ec
c498841d4d1e5458789250294f487e44939b40ba
390 F20110114_AABGVG onstine_a_Page_114.txt
b42542c9b98c3aeccc4b4a5b5dc56dfc
1f95581f78b7b23626503995ef69b71aad20020b
74 F20110114_AABGUS onstine_a_Page_099.txt
7f37eb23794fd8cfaa27fe29707345c4
579de287a6bb3b6f7e0000b2797190a4b39a0d81
199064 F20110114_AABFSE onstine_a_Page_022.jpg
5ad9c47e792ba761ece54182471144e4
f83b467cd2676d80deb0d80b117e1b90a277478d
84117 F20110114_AABFRQ onstine_a_Page_005.jpg
025e797e8035691dc22d1c14df155e37
a1405e5076af6fb6ca438ae6b206c794ec3ee8a5
29922 F20110114_AABHBC onstine_a_Page_051.QC.jpg
1dd482097553ad3bb82dffdc135c7ecf
a4fc67ef30241c49064ca6c28b9b6b4365102853
74398 F20110114_AABHAN onstine_a_Page_091.QC.jpg
ad163bb7c627f987d17af62e7eb47b4f
4142249430e9cb945de13499546ee6816040c5a8
703 F20110114_AABGVH onstine_a_Page_115.txt
43ea59d5935096c622a84bedada91245
01baee04ac719d4c1baaaa77f68c875c6b074143
204 F20110114_AABGUT onstine_a_Page_100.txt
a6a5784a32c2660aa7a4d8367f952eda
a6c8168f0f21f0ab110493a03709536d15481fbd
207278 F20110114_AABFSF onstine_a_Page_023.jpg
e9309bbbf4ecdc36f44aa56d6faba7f8
dc730502392231a3086db87e1c5969423bf9a282
361627 F20110114_AABFRR onstine_a_Page_007.jpg
9d1196ea443f29924c84dde3f638b498
4533ce3bcd5c6e9c4c09d60754ed6b6f39e981fa
23539 F20110114_AABHBD onstine_a_Page_104.QC.jpg
1837829fdee52febe0a70d0737cb4ab5
7732d6ed0520e701ce62780f744ff5995246e5ec
22909 F20110114_AABHAO onstine_a_Page_140.QC.jpg
794ce857afd719e775871db651a62709
4bc45a3a61d8984f090a4182188639375826d8bf
F20110114_AABGVI onstine_a_Page_116.txt
fda80549412ac8ad3674867c3f217254
85a0cbc708de9f4d6b6e27e14162a803e7095997
287 F20110114_AABGUU onstine_a_Page_101.txt
27b04d5a102ab554a17b596f1e0a2b38
41e5718a1ad8f13e9cca461c4455b82c3cfc0c57
225956 F20110114_AABFSG onstine_a_Page_024.jpg
96a1d6fe12cc84803cfd72351d604f33
85f7dc8912ec166fb2bc22bbfde3711904fe8ae7
124780 F20110114_AABFRS onstine_a_Page_008.jpg
c70d87485d0f896b579625aeb1e1d8e0
e715f06aa5b19afa23030b47706d0d0fd8db5986
7459 F20110114_AABHAP onstine_a_Page_095thm.jpg
1e0e0bc5be4c9b991c4bddfe989a91fd
4e00507cb0975b51d36b5c0fea8716b079fd8a0c
1913 F20110114_AABGVJ onstine_a_Page_117.txt
d7bca68aac38fc737fe096cee624a179
e30ece6c94cbb1be30305c9350de3849acda7348
834 F20110114_AABGUV onstine_a_Page_102.txt
b4d6fa0ef7394ae5b32719d7b0d32c3a
70b66842d8e75b33fd37c2203158f8b261e08e23
203001 F20110114_AABFSH onstine_a_Page_025.jpg
763fe0a470035cd4e75c4c3eb613531a
3caa402cb1c598339906e905c259206a8defb2dc
125755 F20110114_AABFRT onstine_a_Page_009.jpg
7e2ab3a6e735adbfe26db3d3da890d84
d8011112749497859548976574b5cd28ad64801c
55503 F20110114_AABHBE onstine_a_Page_150.QC.jpg
a2a46312f30b35ce07cdde9fa7b70981
da49c795e3fec75b66cf6ca5b5b9d70e01a4bb0b
60591 F20110114_AABHAQ onstine_a_Page_080.QC.jpg
53bd7fcd0e0aaf5e5594da59990a095e
2f1b37a7cd2c2759162443b8dcf6096b6c7ce7c0
1892 F20110114_AABGVK onstine_a_Page_118.txt
1298c6d978ba4a88fc94456d23f41c2c
c7a8de75201d62d983aae5901cf6ad7e43520259
453 F20110114_AABGUW onstine_a_Page_103.txt
0d296072103d797ccd3f7d5bda9e4f26
9023d4cba474e895ecb093732933416e9dd28c16
206017 F20110114_AABFSI onstine_a_Page_026.jpg
c489aff2114e2a4fa5caee585282bc21
90a629056cd9f0431b7a100b528e1448b19cb872
252255 F20110114_AABFRU onstine_a_Page_010.jpg
f960175d85c62873c051ad627ba059d5
b2d4f49c8eb985f3651effa51e51727eea83ff87
36639 F20110114_AABHBF onstine_a_Page_130thm.jpg
e10634cc8b2dcd8b20de9af233ed9252
5320afd1aac6e464bc73f0ba8b51754084cb6803
23578 F20110114_AABHAR onstine_a_Page_042thm.jpg
2fbf2691b7e6029d4be2f111787ad03e
43089e731459876322ff4a26986122ff405d6293
527 F20110114_AABGVL onstine_a_Page_120.txt
34fc07a9ae225c30492e8e440f28a035
31124d7eb7bd3c4ef37e840c47d45c8fe7cc8d57
545 F20110114_AABGUX onstine_a_Page_104.txt
7a5576a9f1b728645cc77f7f2592a019
ec3abd434d08d696cebbb5c82da9d8dbcfceb8be
217738 F20110114_AABFSJ onstine_a_Page_028.jpg
0095454c00050b01b578c535de8bb165
f0a22f77307b693a3f400d098553dcb04df97e87
305020 F20110114_AABFRV onstine_a_Page_011.jpg
78a1ccbdff55cdb582dc436f4b324f48
125bd32016a3ba2301c6d84a460304db5b4ecff4
60674 F20110114_AABHBG onstine_a_Page_020.QC.jpg
88adbff5db37dc5f5feeffbfdd2a4d8d
9bb1a2cb23e5632582462599b09b37baba56d953
39162 F20110114_AABHAS onstine_a_Page_110thm.jpg
b266c33324ccd3b7b9cb257201258df6
e33bdbb9f1c9a5c4182f4a56c9e9a0d02d703f57
603 F20110114_AABGWA onstine_a_Page_137.txt
266ad03f1730147dca49b46c5de5e8d3
169d8824b8d96a3004c3edd09d85d2577c54c39a
205 F20110114_AABGVM onstine_a_Page_121.txt
de233724a9e70d89218e1173b22d8e24
238539e5546311a118507a1acaf4c6f12691161e
193381 F20110114_AABFSK onstine_a_Page_029.jpg
0fb5964c5a5e1a2983cab2310672cbcd
83a1da2906ae75a5b845f4b67040d83d261dbdef
9424 F20110114_AABHBH onstine_a_Page_111thm.jpg
3ecb128ad9808fe13ab7a0cf38fef721
6d83ef8bb2e2787a2f42678b633ffffe9813117b
62161 F20110114_AABHAT onstine_a_Page_071.QC.jpg
c0d90409b0807761b23e2b9a928f0922
76ad573e79bb7e4e3e50685acd67ea8cbd64e870
320 F20110114_AABGWB onstine_a_Page_139.txt
22b456ae9c499208efda81860ec03151
a3ac158cc6f78d444c4b9fff4965c75986299b4b
680 F20110114_AABGVN onstine_a_Page_123.txt
c2ac09682ccf02fb9daea5cd8263fe03
20644292efe8bd64626b4339e2efd14061a70c19
651 F20110114_AABGUY onstine_a_Page_105.txt
e00bb7d45ce79164d548f6d0fed87d56
26955fb244081144e6b2a24c6fc2c2602a227e73
60013 F20110114_AABFSL onstine_a_Page_030.jpg
9b130f7ac38661806eb013542b863be6
3304c062259bfb6b516e4969bd669058603e2b98
185781 F20110114_AABFRW onstine_a_Page_013.jpg
4fa88c4da6775e9d54b51c2f8c55d361
dac890ac8943d848592618fe8c798d295d0937c7
25252 F20110114_AABHBI onstine_a_Page_065thm.jpg
b62736fc45723581e9a4afb66b03b018
5d380c8e816a12c4af1577b83f7ac481cfc7a090
17869 F20110114_AABHAU onstine_a_Page_141thm.jpg
4e15f7266fbe2f8505623bf1e4d6fc13
1b6b7828b6e6a20974b21b6deb49471f493829dc
409 F20110114_AABGWC onstine_a_Page_140.txt
e59e2e93894092bd93fe03a05b2d9122
2d65dafb5994e8a47e80b8492d8bc80f62e8edde
1337 F20110114_AABGVO onstine_a_Page_124.txt
80f16d3b54f8eedb8cf643e23c4d6cb0
e9a316ce58884942f0cdcdf45552bf35e0083410
565 F20110114_AABGUZ onstine_a_Page_106.txt
225d615d0c61d6d4a013fc6158db9a07
028f88836df913a2c34cc42adaf97bed1d8b7609
77786 F20110114_AABFSM onstine_a_Page_032.jpg
d8a1788cc7a9e05e5e6190816ac01027
12e30db0fcfe9c08876826f40af922363635c633
171948 F20110114_AABFRX onstine_a_Page_014.jpg
2250b4ce997d95dec606b2e9eadc6bfe
2ba232b710e8a7a8583cf2df00030c7f87647197
72339 F20110114_AABFTA onstine_a_Page_048.jpg
4af12bba7ea8137a5681008fd303dba4
4bcae5f0288c3d4a91cdfcbb208a3a3b1f36085b
78564 F20110114_AABHBJ onstine_a_Page_046.QC.jpg
d89113cddd7e9562b26e45c91dae6f08
5d6fdf9004df1fad7eb69d39140fb838ce8a17a6
53702 F20110114_AABHAV onstine_a_Page_007thm.jpg
46d22917dbc6bc103049b2d9b735619d
6c764845d38816a68668c459f9060e4a260c97e3
516 F20110114_AABGWD onstine_a_Page_141.txt
46077bb0b10b2c75e3d54b9562f25883
b085ea2bc80cf1d94fc04baf5323abaf7609f8d2
802 F20110114_AABGVP onstine_a_Page_125.txt
663155fed50db41a42a6fcf1e3cd41c8
6a78e5558c93051eea4a89375d7c73b6c38e3764
174790 F20110114_AABFSN onstine_a_Page_034.jpg
395eafe53a9dda926d545f6fd0c8c657
35a4f8e432abf82defafa09ac0338e23b5ea94f6
179468 F20110114_AABFRY onstine_a_Page_015.jpg
1015f5f60fd104aaf3ca9dd22e032029
eeb019eb8ddad4b7c057906fd6523cc6ae279f18
78057 F20110114_AABFTB onstine_a_Page_049.jpg
1a92adef309c385129c4584dfc37068a
30e4955ba5c6da658d92f9add46a0be81d5e2f7b
60771 F20110114_AABHBK onstine_a_Page_159.QC.jpg
f14c7d7faaa4c23753344f6f00112531
e99b3db87e6117642757285ac1fc36e83cdca63c
18498 F20110114_AABHAW onstine_a_Page_120.QC.jpg
de70d5b2bdd781ad0d4bee0ea2ca6300
d90c55943f1aeca96010b363c455ea6ed23cac65
443 F20110114_AABGWE onstine_a_Page_142.txt
9984fb1b12ac9c7a96b3bdd8a4ad0ab4
222e486adc16efd00a832f955c2837efe84caa5c
881 F20110114_AABGVQ onstine_a_Page_126.txt
de60a4abc46e132f0544d5aa30ddf5b7
f09811201decaf3403e269de8693bb8724b0c4ad
217761 F20110114_AABFSO onstine_a_Page_035.jpg
20243b7a39fbef80b70a81f44eb59b1f
f86bf2c7c7b6c10fd738091ee8edd87fff006afd
182107 F20110114_AABFRZ onstine_a_Page_016.jpg
acac26984477626c8e9d96231b9fe175
434ca5cd1c320b6d74fda9c89bc2f7387f057966
72851 F20110114_AABFTC onstine_a_Page_050.jpg
325f513b56bde90da36f94b7b26743af
b8ea878487944357b175e59e30401b9521b25b65
24825 F20110114_AABHCA onstine_a_Page_017thm.jpg
7f2428b8b4565767a3e04db6c9b1a7f2
a6ac1903723cda679011ae0157ba927ae89c0368
66996 F20110114_AABHBL onstine_a_Page_015.QC.jpg
f22884ef1c0b197ea0688ecb1edb7dbd
e18641e60ed771a20475aea311308920e8348466
36062 F20110114_AABHAX onstine_a_Page_102.QC.jpg
4d123ff8e97a358e368029541e49b8fc
7b463d13d8578933cf78618bf8028dab086ba88d
F20110114_AABGWF onstine_a_Page_144.txt
317e8aecfb06a54b900839231dd6f0a0
1e35f0a89f2a34502c0d140a6c74d9c2553470bf
750 F20110114_AABGVR onstine_a_Page_127.txt
e653e07192ad06555a4a049994f82486
b4261e236efb4fa8327f31bb780b791ebf929737
201081 F20110114_AABFSP onstine_a_Page_036.jpg
7725cbfbfa4732e77fc208553e12c90f
ca3d80be939bce414b45487f435de37dfbba0a1f
64130 F20110114_AABFTD onstine_a_Page_051.jpg
a49deea87f76326b8afef092fdf7b94f
e7e5a0ea049135b7924f729a54cc94d41523a7f6
52023 F20110114_AABHCB onstine_a_Page_160.QC.jpg
4cfdb18a6e0eb91b65636f60cea96c20
0f68787f7e5cbc963973437a1e708fe6e17de17f
37406 F20110114_AABHBM onstine_a_Page_099thm.jpg
c764711157773bd4e49ceb76139c38fa
63ec0f0367e7c78cec9e38e9d4fd43c5a510f7bd
38058 F20110114_AABHAY onstine_a_Page_098thm.jpg
fdd53685c53b5fd00f45640761a7f7c2
9c35edfe2043c3af75f575387b3d254274ef7b0d
F20110114_AABGWG onstine_a_Page_145.txt
726f2e6339bd0e2a0c0f1818f3f2e615
c2c74305e11459f64f6ae275f238e0223ee41dce
220 F20110114_AABGVS onstine_a_Page_128.txt
73ce340ea5ad90861e10d58648fc8178
377255e7516828fa64da38a7d606105b4e36c189
204052 F20110114_AABFSQ onstine_a_Page_037.jpg
ff1fcd3bc7e7104e0affd3ac89b1921c
cc97150497f08052c58f166c2b59ff962f94df8e
107299 F20110114_AABFTE onstine_a_Page_053.jpg
bff3766c7c38b7bcaf257d968409621f
8338b8942da0ec6edd506d6e5f5e4d785d015cf3
31252 F20110114_AABHCC onstine_a_Page_005.QC.jpg
d83bd26f5211cc093cb9e2f13086ecab
847f61d87436b8307f6637d3f59d02eb61e152f1
32621 F20110114_AABHBN onstine_a_Page_100thm.jpg
2d2b5560f28555b6d9becfb80d48ef05
856e8876b4196a9b8bda9c29d9e87f135f15a91e
80690 F20110114_AABHAZ onstine_a_Page_177.QC.jpg
5da8f75f95df4104be874c363f95a838
168a3407d45cd156c70a8719ef984eefa8fe9431
262 F20110114_AABGWH onstine_a_Page_146.txt
62beae0cdabe69d2fbfe0f1d333c91f9
6c126c67acc921804ab9115aa0c2981c6272e422
F20110114_AABGVT onstine_a_Page_129.txt
364b744920d02e0a479cd652e26d5bb4
6a4fccfb7835f37811a8951760527974b609f49f
201799 F20110114_AABFSR onstine_a_Page_038.jpg
403cff8db227fb35490b4057733d11b7
e9b3e4d126ec1b299ec339ae32cec7eaecaea721
111035 F20110114_AABFTF onstine_a_Page_054.jpg
9f0d19966d45d234c2b751bd3e7f1d08
4c30ed5c83bc0ccb1f621d58fe16bd8c8ed06500
8324 F20110114_AABHCD onstine_a_Page_030thm.jpg
3a446d84d93167ce385c100dfb6bdc0f
1cbb77463d5af8a96178d913f8453230ec4f32f5
36435 F20110114_AABHBO onstine_a_Page_080thm.jpg
6f7ef0ded60006e183325821244a2acb
6904cecf238592ec742725d0a991c81e3bf75431
1649 F20110114_AABGWI onstine_a_Page_147.txt
00f186258b9a0301087d782cf2c4fa17
78f721ba424c04b6d0c1241a7332cce62b82bcb7
397 F20110114_AABGVU onstine_a_Page_130.txt
11c3a5adeff4343379572dc5a83ac8cb
1825b3777b6a76add13ea2d3ae9e236bfc6cad45
193515 F20110114_AABFSS onstine_a_Page_039.jpg
32f00d67edcf2dda71f71fb2647d7a58
3532a07d2fea9f56cb38c2a03866e758002453bf
114816 F20110114_AABFTG onstine_a_Page_055.jpg
dc10f02f15b29e9ec87b76086ebb9480
f68ca5881ed8c307752cb2c1437b1b080360cf15
59004 F20110114_AABHCE onstine_a_Page_052.QC.jpg
19eb2586765a77e030c872920a581956
a08852c69cb0f577528cb8f0ef72fb579546b3a0
59302 F20110114_AABHBP onstine_a_Page_101.QC.jpg
54b2274e4f5a1bcd17449c1da719e387
0f41044cac35f34a8a724762d7686ad0ac498cba
1906 F20110114_AABGWJ onstine_a_Page_148.txt
27b51a2138ca1bda3d45854cb69289fe
2e0b85d38f5e2f8c32516b165bcd89365c1f90e2
1833 F20110114_AABGVV onstine_a_Page_131.txt
c4ec0ba29a0f687a43b93e0007d4ccf5
894ab45e1daf6a302bf962749f8b2913d36d8369
193846 F20110114_AABFST onstine_a_Page_040.jpg
c2979bb6ad2ba9a65ad1931797602a10
ea3b0ab6b52ce8df473a104890375fd1dbb2847e
80600 F20110114_AABFTH onstine_a_Page_056.jpg
6befd40fd4d3da6c91058179363ff5f4
29708f13c77f2f30dc6b6c2a0cb91eb51d0aef57
68785 F20110114_AABHBQ onstine_a_Page_127.QC.jpg
1d07e583131e025ab80637dcd456a026
43586ca892467725d7f5b16c8f748efec334cb03
1323 F20110114_AABGWK onstine_a_Page_150.txt
e76a31c5664f73c1ab967c383bccb86e
8677ba57b06a9a3530e906de1e9eb6726b1cff98
1824 F20110114_AABGVW onstine_a_Page_132.txt
8f536605b04202203c854983ea72d80e
66600cb80a5348845265841d6fb08056722799d3
222620 F20110114_AABFSU onstine_a_Page_041.jpg
3644e55b0d45064df8cfe2a3aa066b30
cee2877aa4c88bda525aa9f975f0a8cc16ae6944
59173 F20110114_AABFTI onstine_a_Page_057.jpg
251a2dd5cf20d7c43b8670c8de562305
418974b2cfcc75921f5bd49cbf9e104ff66f50fc
20545 F20110114_AABHCF onstine_a_Page_157thm.jpg
f5f242a85327a91c936bf6693439128d
2ffa1e42d3fa34157107423f9de6d30ffc3ba027
77036 F20110114_AABHBR onstine_a_Page_178.QC.jpg
c0eaaed3906b2b29c5a0957925905232
98243845c5e89be10f3d158bb8dfec85530aac29
1724 F20110114_AABGWL onstine_a_Page_151.txt
8e6c0ac6a6b9f495767094897eda8ac6
a3daa07436ceddafd0dd3782e448893eb14c74b0
1905 F20110114_AABGVX onstine_a_Page_133.txt
8b09ea7b3f3e847c0e0e6aca2a206e9a
2b80b9aa226ffb43bbfa1adeb8c658271c3f5f72
203899 F20110114_AABFSV onstine_a_Page_042.jpg
2f5033552f39482b5c76f6ecfed059a5
bfe73e27e636bec40be866af7e7f435cd4daabf2
101080 F20110114_AABFTJ onstine_a_Page_058.jpg
bbac7c8eceb82b9132326db7c722857e
e8576782156c7e8dfc0b26c1e1d6e3450ab03fdf
25639 F20110114_AABHCG onstine_a_Page_168.QC.jpg
2b0516221606ad9b2d77814b15bceab6
f93e5312fe42fce097b47cc0b27174ccf65ab0c7
3149 F20110114_AABHBS onstine_a_Page_002thm.jpg
4cb412ea93ab921cf748f9a4d5dd62ba
87a07b6a421eff0d97bdffd52cd69a7222746c7b
162 F20110114_AABGXA onstine_a_Page_169.txt
a5e05e78cd4df7dc26eb87af94c20e16
39d0b00a4e62cf85946c2e955b94545054507b3f
1903 F20110114_AABGWM onstine_a_Page_152.txt
d571a0bb8c8a691c21885bd70d9928b1
e046d1bd5e7b891ea10920da478f9e5612f86782
592 F20110114_AABGVY onstine_a_Page_135.txt
562849c126fb6825dc87c8bcd7d105a3
4f3d1abb89bb23992abac5c52a465129eb7ee860
224736 F20110114_AABFSW onstine_a_Page_043.jpg
6c181f1558388a5da0e2c29a6d448a89
f8334324d353ccaa5e93ab23b63150ead961d130
79274 F20110114_AABFTK onstine_a_Page_059.jpg
a4d1b21ed868b37854625b491b18494f
da77b20cf0a95ac451dc83786591b9589a8f0453
18264 F20110114_AABHCH onstine_a_Page_059thm.jpg
1eb2c528e00eadd47fbcb99552b55e7a
4b127836e15f177a71162d0035e3bc7cd4919ebc
57575 F20110114_AABHBT onstine_a_Page_085.QC.jpg
83867b0751fb6572fec5eae6e00f1e2a
cf75706148086ef8c5fefcf1abc240f2fd867923
304 F20110114_AABGXB onstine_a_Page_170.txt
716676d9eaa64191e2f3af1683090ebe
9c95c4cdf4e2d87bece620f8a8612a13e9c6cda2
1581 F20110114_AABGWN onstine_a_Page_153.txt
2ac5caf33d25263781a1acb52074a10f
b92f76a001c41a1e92a6e5be885f0cf8674a63fe
97737 F20110114_AABFTL onstine_a_Page_060.jpg
fa57f29c9185996e1a04af603847de27
8d2ad3a5cae77b776732bf253a1e4ed5a9ef0a9a
53942 F20110114_AABHCI onstine_a_Page_054.QC.jpg
db9c2e45c71345b614a426ecf4a6656d
2415b9f1320de29e7a4551732043061427b20863
34994 F20110114_AABHBU onstine_a_Page_123thm.jpg
13111632356ea60d62a9acdff7cd7d61
a2f2ae922fb1fc31dc1aec12a01f827c4f17bd85
122 F20110114_AABGXC onstine_a_Page_171.txt
52ceeef4ab1d70469ea449f9d138464d
047836476ecf67f2f9e453ffc8abe5cc6e5492ec
1921 F20110114_AABGWO onstine_a_Page_154.txt
f1ccc2d0c9a299821919bb465481bf2b
7a42f12e4c2e3ad0c9feb3a2789a33792efe5c22
1051 F20110114_AABGVZ onstine_a_Page_136.txt
7035ca2d67fad20ace199b124764102e
d709d9289984d2760c2d2d05f635a4d169ff2058
214955 F20110114_AABFSX onstine_a_Page_044.jpg
527faa54d8ee63b36822aaf9b5b31aac
92ce926a796f8b978f13f76f39ba29e1fbc620b4
78143 F20110114_AABFUA onstine_a_Page_075.jpg
1e28a51be9cc14b31b9ecf32acac133d
8e9ac9db352f52d4411efe74947549592b7cdac2
117886 F20110114_AABFTM onstine_a_Page_061.jpg
391f2948d03c7971164789985da1fe99
7459a27eef8d7739f33bee0c1f576f3455208cbc
24814 F20110114_AABHCJ onstine_a_Page_028thm.jpg
091877dbdab255815a02324287b399ab
6de1b70f1eb3e4afb86209046eebae0b6b889d79
15066 F20110114_AABHBV onstine_a_Page_097.QC.jpg
7a8a2941f57dbce61ac07bba3c290eb3
90ec7b28838c42c9b55632d7ab4daaf8c63eaa48
141 F20110114_AABGXD onstine_a_Page_172.txt
223afde330653f6bc7fc6ca7b58c4b0f
94b5e56ca380c76be67c1e29d395cfb8b040e7b7
1830 F20110114_AABGWP onstine_a_Page_155.txt
38d0124433cf37e4cf009ec1bdc8b6b0
8c90fa224216709851c0c47006b8bd320923a190
209060 F20110114_AABFSY onstine_a_Page_046.jpg
74002f4d46a5520e26ed999ffc732df1
375da22b5d6641da1a80c0e9c9263aa780a5fc3d
125375 F20110114_AABFUB onstine_a_Page_076.jpg
5159a0ca52c8f05f94d8ca5f4153b5d7
a59dfa3349a48ff73c3acf96ac4f072a26e0591c
54604 F20110114_AABFTN onstine_a_Page_062.jpg
cb0ef3a265ad07c8fc6e775a98e9d53c
0411b69ec688554bc64808f87c9ae937619f05b2
25674 F20110114_AABHCK onstine_a_Page_144.QC.jpg
cebe87907a498c55378f42ef09efb0de
f5b5ca26ffd7865e949b438253e48b09858c66ab
62547 F20110114_AABHBW onstine_a_Page_156.QC.jpg
ee3c48b8c0194eddb6d3f75c34ae2d43
562264e9cf98963cbaba237a6c0e80bee6f859d9
121 F20110114_AABGXE onstine_a_Page_173.txt
06fba8d15f87d863942c4b239ba91383
e7f64701fdfc4b1590edd62391cf5529b1e834a7
1775 F20110114_AABGWQ onstine_a_Page_156.txt
d3f12f3d79627bf7418daa1e488a02b5
6e21001cb61724f08fc23f73a11c2c7511975706
70352 F20110114_AABFUC onstine_a_Page_078.jpg
e7dedc002691e07e47e3e24c542472d4
f88d87c824ab02f3446a2726d273820a6b6b65a1
191720 F20110114_AABFTO onstine_a_Page_063.jpg
44bb1e79764f59e01ddff3f850806a7b
099c905844617a3fdf7afec62953b614873a8aa5
33286 F20110114_AABFSZ onstine_a_Page_047.jpg
ace9a9cf7205a467181ae655a3978763
a90d117382167c8862f5f528a221b884d5ee6137
15541 F20110114_AABHDA onstine_a_Page_081thm.jpg
c459ca971081f91196c79fb80c15542e
debb61f3288c73608e1987d74ceb85cf10a8e0dc
15840 F20110114_AABHCL onstine_a_Page_162thm.jpg
f817c6d973ebbe8c986288815f4ddb8c
406054d4c996c44343bba4b63742f7a8c9926f4b
63384 F20110114_AABHBX onstine_a_Page_147.QC.jpg
75fc3504b00ead2f11a1b71911aab9f1
302c9a1fb8a9047dd44d8eb4cae9799301d96a25
1912 F20110114_AABGXF onstine_a_Page_174.txt
fc2a62c95f3d9c3f40e78288313c471d
d222bd8eda80d15f8ade98d5c924288147844d23
1663 F20110114_AABGWR onstine_a_Page_157.txt
34a11c6331f9d7c94c29e6f9df23427f
d71c06a23f1cbca627513d71fc9693448283ff81
104643 F20110114_AABFUD onstine_a_Page_079.jpg
5fe03f3c01ca2400401fc8f65a7b7dd4
75dac7f0b6ab2018dd48ea9290a0f9e81ccd1da2
210232 F20110114_AABFTP onstine_a_Page_064.jpg
358b26745bc8ea90d863465d2b17b9fa
c42d47b3cf5f715968313bb207ce069b6e48bcbd
17246 F20110114_AABHDB onstine_a_Page_138thm.jpg
912f6e6ff58eb871faca45d277b6b692
d1f55b479f452a7147c7b50775a24d72c22b521b
31849 F20110114_AABHCM onstine_a_Page_161.QC.jpg
a927ff69a0c131caf73fb82f285631bb
df857f87385eb67dbd3877306a4c63d5b2cbd297
24191 F20110114_AABHBY onstine_a_Page_175thm.jpg
da3c38471bf5a6c2801ac4d7bdaaa112
5ee3ec6e8ce429c85945497a9ab6c2cd87d26754
2185 F20110114_AABGXG onstine_a_Page_175.txt
be8802ca37ef4373c4a2cdfe82401ace
8a29056fd252459561036c0ba40d15a6f580585b
1915 F20110114_AABGWS onstine_a_Page_158.txt
89c71a4d7750eb919c14ad18b3af8eb3
f8b240676d47e2b9c7d8e51527aa318c37fa18f7
139797 F20110114_AABFUE onstine_a_Page_080.jpg
3cb924f291d72d9e1165fe7f63e96bc7
d1a4f93b7d73be8604b97fb12f605a7c66e36f2a
206868 F20110114_AABFTQ onstine_a_Page_065.jpg
d2ada067122d10082e29c89d03c3cde9
23e758acd8380b90de9b774cf4949da8b371f980
36917 F20110114_AABHDC onstine_a_Page_055thm.jpg
b7dbf8339e2fdc58ea5fb8ce8cf82774
270ec7fda44e07fed17530f1fea57b88c0dfe263
22164 F20110114_AABHCN onstine_a_Page_131thm.jpg
12a557ff54ffef6dedc11c75d0c1d784
fdef32ed980edc51b5c3526c32c7b54b1eefe442
10553 F20110114_AABHBZ onstine_a_Page_103thm.jpg
154fed0519e77466c57be54284d77886
3a9a3674924f666ac6ae49c6bbbd15a34ed8c572
2406 F20110114_AABGXH onstine_a_Page_176.txt
1e68781dfab63a78c8821751148f7b5f
97c0a3afb30357dd2918094b46fb4b4b40909957
1831 F20110114_AABGWT onstine_a_Page_159.txt
b6e03dda3c9fbe7b76e75b84e0180441
67a1e42301d3a43f03789eed93039caf5eaf4c3c
57341 F20110114_AABFUF onstine_a_Page_081.jpg
cab88f69000c0c83199f7b9fad84b03e
9a46967af576196e6383477b8f7f26fdaf016007
213389 F20110114_AABFTR onstine_a_Page_066.jpg
23680231b73413ea2fe97d565623e318
1fc48e04425f133626c28873de7a79c5f6956dbb
97974 F20110114_AABGAA onstine_a_Page_063.jp2
859c24d2a8d1512fcb863f6f4ed66c00
76e78ce08b12c547954eda7880717f0654146ae4
77142 F20110114_AABHDD onstine_a_Page_133.QC.jpg
42c4d087f057cc0914f35108c973eb01
6e6c1871fd2a8bba7fa2b621935b02297c7cabf2
33750 F20110114_AABHCO onstine_a_Page_122thm.jpg
918183a4141f9e066ea71137857c7547
e33263a92e9c42eeecd482cdf2de6f1b32f8577b
F20110114_AABGXI onstine_a_Page_178.txt
abd1d82c6d7f230e73fd66f150df80db
07cd318725716b72e903e107fb1f0d76008b725b
804 F20110114_AABGWU onstine_a_Page_161.txt
4a347ce4ca98ae37e1082df4f1d4f1b8
6c8e10b71575b999d3eee4ca3ce76f28a37c6ed8
146096 F20110114_AABFUG onstine_a_Page_082.jpg
91645d7f477bc1feb2f9d6affe2cbc2a
b5de4a41327b215d48654fa851004d3ac648060a
207577 F20110114_AABFTS onstine_a_Page_067.jpg
2562888aabcfdcd75d26e29397e99e4a
85878ba250e42608412d245e756d6a3e1d2dc652
107997 F20110114_AABGAB onstine_a_Page_065.jp2
f52886ef3e3a890a1ad853461dd73e3e
8dcfeccc5a22a9b2308dd84e1a62d5787a4596ca
19299 F20110114_AABHDE onstine_a_Page_020thm.jpg
13a52c4e86c39d6f897f62e9cac733b4
c87d5f430c925908739b1b42a82f970a0f60e909
18529 F20110114_AABHCP onstine_a_Page_135.QC.jpg
5f582c6cba60cc43a369d84a436c4a8f
4420012b683a938951fee14100188a509af90698
261 F20110114_AABGXJ onstine_a_Page_179.txt
65c679d97851ef8bff1a4d0736cc7ef1
a37382a84e89273613f80c85e456628fab16e1a9
131 F20110114_AABGWV onstine_a_Page_162.txt
353db054e19dfbe5b78f58845d44dec8
c68673662127cc02245997f8120454c44de0971c
118140 F20110114_AABFUH onstine_a_Page_083.jpg
d4d03c01a3a4806d7dfde9274f83fd18
4669be88b0869e41d24d06d870b2ff51bfc4cb02
160015 F20110114_AABFTT onstine_a_Page_068.jpg
8ebb4bbd18f354262350b71a1b06aaad
9e16daaea362d23e2a0394b38b413b9b727a8e94
110104 F20110114_AABGAC onstine_a_Page_066.jp2
12d5ca7771115f25d113ef91ac3f8567
f80da36ba1bb679afe542b0b934f2afe4d6c2039
27118 F20110114_AABHDF onstine_a_Page_162.QC.jpg
c4f700e00a2cace4b280b932ea521c6d
1f59c6f3beff3f62b018e09d973ac953db4c9d9b
75514 F20110114_AABHCQ onstine_a_Page_040.QC.jpg
58f8ccfee708ef048a78009a75e2a70f
86e87eeaa8f7bc6bcd2d2714bcbf60d2ae7d9f75
F20110114_AABGXK onstine_a_Page_180.txt
6c4cbdec993497a2e886f4aa4bc5da87
ad96a9a9a6137071eb6d7cac5f2a7f2bebcc5196
293 F20110114_AABGWW onstine_a_Page_164.txt
50af480a5cf7603d156f0509a5cfdda5
8eecf29375fc02d9abd1dece4c0dd89e6270bb1a
147367 F20110114_AABFUI onstine_a_Page_084.jpg
b682eb366ca8d0c36676302cfbc9a693
8b99fce1bb2316923b1920f12adacf07115873e1
61901 F20110114_AABFTU onstine_a_Page_069.jpg
69998e92f45bacad01526449087315c2
242ca2f4938449eb80d3006efff4922b201963fe
108874 F20110114_AABGAD onstine_a_Page_067.jp2
f641eb9ad86b626949fa75f0ad36fdf9
3b3773b871634b029f8f37456277a1183f033b01
31770 F20110114_AABHCR onstine_a_Page_143.QC.jpg
cf5b674d3433f7a4c5c17037786f5889
668954443fdbb9e41f519e3687490716e18ff05d
81950 F20110114_AABGXL onstine_a_Page_044.QC.jpg
13bd1ba17ef66ca7e1a9010ec1d12ba8
e2cd67e0cf7cc2bec990d4afc2bb9422781c1ab9
1784 F20110114_AABGWX onstine_a_Page_165.txt
c0dacd2a69c6b90e6c1314054f017f49
f4e0c429eec04ad2f57e62ebc7646937423e4f3d
109796 F20110114_AABFUJ onstine_a_Page_085.jpg
4762507941e07c36b37b457dc4626d0a
cf55a81cf8fcd4f09138bcc6f29ff3318cee42a1
91229 F20110114_AABFTV onstine_a_Page_070.jpg
3e614ac63cc36d2f79004415b299541f
ca8be1c1dc746266c92075e5ba057d320ee9148b
24886 F20110114_AABHDG onstine_a_Page_031.QC.jpg
44636acbd326a10d9345e37c9362cee9
303529db39bbb63b1e75d2ae2b3a37f99187d6c4
18955 F20110114_AABHCS onstine_a_Page_001.QC.jpg
c1e136e3e47f44adeb2d553844e5a017
042a92a7650fe45bbb8bbbce2ea78f2805f8e2a1
36404 F20110114_AABGYA onstine_a_Page_059.QC.jpg
2fc152714863e81c0605ef8ee27de385
97cf9266e33034609aff6e8eacdd47316f9b8645
47251 F20110114_AABGXM onstine_a_Page_049.QC.jpg
7908a6d70c12ad9d9be27f1dc6921b64
32713cfe7138e6a3a0023f704043b809cb06fca6
1936 F20110114_AABGWY onstine_a_Page_166.txt
59aa14969fe7fb2680e0b8c980c5c6cb
d3c3b844ff3a93ae65d9383a52314af8054bd987
162087 F20110114_AABFUK onstine_a_Page_086.jpg
677b8d7828650359d435441b52538216
d046275aae6547e5ffd6781fda92e1d505538ce3
132279 F20110114_AABFTW onstine_a_Page_071.jpg
3cfbbe2c908abce9c52100532d49bb02
0c670d6d22939ba4eaa2543ff454e391ba7bd10d
83287 F20110114_AABGAE onstine_a_Page_068.jp2
3af38a50b9714c4d5d284a17bfd50503
2c9b360eeed7943fc35d2db27b414996df5d2291
4324 F20110114_AABHDH onstine_a_Page_019thm.jpg
30e4083ee3b4266d418ca746327f0508
64df588dda9922c125154de8bc994f73be9d42f6
38775 F20110114_AABHCT onstine_a_Page_084thm.jpg
bf660ce8649866ccf255ce9cd9da0f21
b4a78899560dd0a186e9f3dfcc3d162755978ab6
39523 F20110114_AABGYB onstine_a_Page_106thm.jpg
ce602bf52d8511d970b35fea04992a41
03c39a03dd81b529c3212b53f098bd5083849d32
76212 F20110114_AABGXN onstine_a_Page_166.QC.jpg
4a6e255a4c1bedd93a35537283a668f8
dbda4a5a3b1ec71ea29320981e939caf68d919c0
1260 F20110114_AABGWZ onstine_a_Page_167.txt
4a70cc1b985ed467098b0b551dddc2be
5cff2e302a67a7b9dc2302ac11eeaa8af4756a45
188294 F20110114_AABFUL onstine_a_Page_087.jpg
9f4fcfe044cd7c424629283c014828d0
545c40f53188d834aae91a6aa2dffbdc9e4e1a39
108871 F20110114_AABFTX onstine_a_Page_072.jpg
5e19080859491c7a04f2397c681973ee
cd29cbfce2f3385d2aee46dfb695cda7491a18e1
27568 F20110114_AABGAF onstine_a_Page_069.jp2
2fd588f2f3f2ff25bf06b6577599e870
19c0cb2ca6b24c7a2a48bcba12d2f132be83dc69
47647 F20110114_AABHDI onstine_a_Page_100.QC.jpg
c8663d3c6bacd58d46e8c2a4c25fc439
fd2e059cff491bf0ec6f85afb6011e37c6d6ec62
33616 F20110114_AABHCU onstine_a_Page_073thm.jpg
5551953c4df6a6937a51ea3cd4b34137
2464ea4aff4ac3fe16f7407fec8964ecab385e4b
60000 F20110114_AABGYC onstine_a_Page_068.QC.jpg
b87176ce863453666395f4519376085c
4f8cd3314493c09e5366704c2e615ac031f505b6
49636 F20110114_AABGXO onstine_a_Page_122.QC.jpg
da1092b3ce470c167c0c8aa79779436e
1649b377f3346a384fcb830c49380f3602b79b2b
225877 F20110114_AABFUM onstine_a_Page_088.jpg
ad32be5b1f170c4d2e83def957823346
8c73f87cccc151668581122170411866b35df216
350047 F20110114_AABGAG onstine_a_Page_070.jp2
4e2729cb7be983985fadea67c1edad82
af799e58cf092e25d391a5ac2cabb99c20d0f1b5
58848 F20110114_AABFVA onstine_a_Page_104.jpg
62857122a876bb891c47a57ecb053be1
d7433c23fb029606cdf7ef9e27feb40559befd5a
6453 F20110114_AABHDJ onstine_a_Page_003thm.jpg
f5ff425b34f3ec30fde6bd70eb733517
0f371cc51eab1c018e4c57976aac522b79cadf28
75140 F20110114_AABHCV onstine_a_Page_128.QC.jpg
f17b3a1084320f01010ed8f8bd9d596b
e4c7746bb57a4280c24646a0195e6716791a67f5
18053 F20110114_AABGYD onstine_a_Page_056thm.jpg
d2555b7fefb72043d4dfe3452581ffb2
6650c3b19d77bfe2055c0280ad34cbf910c12646
23975 F20110114_AABGXP onstine_a_Page_148thm.jpg
6e5f44fa4b7955c641348354d4b653cc
8a1e8fddf6c9f3acf8720de881d752dad02ede3c
197152 F20110114_AABFUN onstine_a_Page_089.jpg
60b1840acf7dd2db185b1d8cdd48ddae
23cde18386e928c12bd88754966e3f8048ab1eae
91272 F20110114_AABFTY onstine_a_Page_073.jpg
51581ec287e4f413f7732b6d21fca6e0
72757c6a987566c6aac21755b879dfcedd6c9a7a
579240 F20110114_AABGAH onstine_a_Page_071.jp2
5612a538800a7dfc6aea47fba60391df
d7a7c100e5b8b8e60f5a0668c148ec36410fcf2e
80517 F20110114_AABFVB onstine_a_Page_105.jpg
5195a1cd08039f1fa1429c7169a28505
c89edee086f3a2d7b0ebe82c3bc65f4847a00360
63647 F20110114_AABHDK onstine_a_Page_110.QC.jpg
d7504fa046b48298389643310433359e
2daa7be0fa2cb5ff61309a0038128ebb35175850
70190 F20110114_AABHCW onstine_a_Page_131.QC.jpg
c78ce9e0c9e78f86482481cf0f2748f1
098fbfdefa8eb477e200dedb21cdae4d45eb6457
39096 F20110114_AABGYE onstine_a_Page_127thm.jpg
6ff15e44df9d594365bcf320b0a72809
ca870c449b9592e50088e43013190b19b5336ccc
70240 F20110114_AABGXQ onstine_a_Page_152.QC.jpg
cd38c4a136064db14b6781bebb9beee1
09db0f9a7953ba5654db4f655874f5c6ec10ea87
211014 F20110114_AABFUO onstine_a_Page_090.jpg
9b5b41c847ffc17685069a910ceea05b
5bdfcf7c995f3387aa4f75999903387763c7e456
125852 F20110114_AABFTZ onstine_a_Page_074.jpg
bb1624b6afde148547fa79e5bfdbfa55
527656726497abd398e4180f4159b1d73ed39549
525902 F20110114_AABGAI onstine_a_Page_072.jp2
20587ddddb22986dd8cf6f5f9da26f5a
aa4836a56009a421e8db91779cc15f821bba8a50
190632 F20110114_AABFVC onstine_a_Page_106.jpg
9e447b826507dd8a8b693d938710718a
23fb2331fc9c598eb22254c9d6a75f2b3e0ee8b3
47276 F20110114_AABHEA onstine_a_Page_010thm.jpg
d35d845dcb7090cc481e2e4d8b0a5a63
668be0cd12809f3636b634d7a61e3d6edc96845b
15296 F20110114_AABHDL onstine_a_Page_170thm.jpg
45b178f6877cd714ab8d4db8ae5b991c
1eb730083f274f54b263ccef81896081ea46243e
34863 F20110114_AABHCX onstine_a_Page_053thm.jpg
4279cc90e4c8fc62f1539620662ad249
4b1cdc5bd22b0e098d22aa20b16d64740877f8de
21623 F20110114_AABGYF onstine_a_Page_174thm.jpg
93ac8710535d5eca866d4ed0de1eb0f8
f54697b2459f8fb5a7e31ad56d34153626fd2afe
14622 F20110114_AABGXR onstine_a_Page_134thm.jpg
264dc778c88297468174879c0c8a4dda
67662f8c6ed5810b9eeacf3f1e7cb20cacbf5a24
201053 F20110114_AABFUP onstine_a_Page_091.jpg
e37b6e0cea2e86a23c176199dacc5dbc
0d9bcdea51a5ae4cfa1b9b06d1963802e1e7060a
505956 F20110114_AABGAJ onstine_a_Page_073.jp2
eaac04d702f0af247a583348f08a2006
6d98cf585c1d447d06e0fa45e255cb625560a808
61816 F20110114_AABFVD onstine_a_Page_107.jpg
1f5450397ecb8ca74dcbdec2298a6969
4c1b037a72697e8c08a69c2119a67d182e5b2ea2
104427 F20110114_AABHEB onstine_a_Page_010.QC.jpg
36316b096147b583f1561801f85f2885
b577c95c196fae6a81e8779307250204c436d874
35285 F20110114_AABHDM onstine_a_Page_054thm.jpg
e326254560c1964ab6f29637b8ec901b
746707fe9212595d650b6ac3dedaf555b26e82d0
31601 F20110114_AABHCY onstine_a_Page_141.QC.jpg
5b6f4021c6885c4bfcc71fcb955a65ec
1a58aebd7d5e8501c28e88b56a3f1ae4990a00d0
78982 F20110114_AABGYG onstine_a_Page_035.QC.jpg
b36af6b9b2253e722ab1bbe8c658d405
027615bdc76f50a43a76a5df6451172dc64cbef5
21470 F20110114_AABGXS onstine_a_Page_063thm.jpg
aceb849383e09b2dbd6a248c5b2e147e
c282c590ace1230a8c2880ff61c9854b361fca0e
116529 F20110114_AABFUQ onstine_a_Page_092.jpg
afe9e0975f103d809f17c876c78b9898
bd709216469a521f6b049347e6503a2d18271c7d
697501 F20110114_AABGAK onstine_a_Page_074.jp2
d40fe4415fdaf407b31c22f39997e697
6197151bc9931f6555f2784119db93ac59a50b55
52145 F20110114_AABFVE onstine_a_Page_108.jpg
dee4006d419c25e89fef18ed6e95dae0
f8fae7a78c1649b4ac215e3534c99e97c1fce5e2
55688 F20110114_AABHEC onstine_a_Page_012thm.jpg
55d7309037ef98e743f76e3e2a765163
669ae2e153a1a32ade7684c3828abf600b343755
269641 F20110114_AABHDN UFE0006644_00001.xml
018146dfea14b5fd2d629abff5a3ad1f
eaa72c32277f106baf62821f86799c6744612ba3
28185 F20110114_AABHCZ onstine_a_Page_057.QC.jpg
7d990a432bc078d08496069b935a1289
5ee2bbb315d10e5d7afa37eb4a72ed92eb7dce30
67147 F20110114_AABGYH onstine_a_Page_154.QC.jpg
d0e448086301b03e83b4a42c3472dc3f
94903b88b741fba44bf3a915cd6b52189e68a9ad
8529 F20110114_AABGXT onstine_a_Page_108thm.jpg
aa6025d9fdf25b696b119831b91dc5c9
7f6a8fc6995f4906ce564c773fe35c4d54f534fb
68994 F20110114_AABFUR onstine_a_Page_094.jpg
62e316e9b08084ab0d3fba65a402d30e
4abc8bc1510c1a09195a776362b7e7d20c6ea1b3
110943 F20110114_AABGBA onstine_a_Page_090.jp2
25e6759b229c83aaf1c659d4f90237e0
b940378cf5d20805761a3bf509061cf351810b54
245916 F20110114_AABGAL onstine_a_Page_075.jp2
1419707995a9ee64516bf844a880a03d
76fc3c2712531f6bce25bfb64001886976ecfbe5
60114 F20110114_AABFVF onstine_a_Page_109.jpg
6ce95e8c793efeb0563063b6e9c3d1d2
37f9b15c6a9a575c5f83012a962fc85bfdab9a38
129629 F20110114_AABHED onstine_a_Page_012.QC.jpg
6403d4058a134242df6043771a573c4c
4ba2c7c930efa9727e609d24f0f9e2a5b7c02df8
7131 F20110114_AABHDO onstine_a_Page_001thm.jpg
a3c865230de42c879875ad56225145ec
c6dc22ad1bf53980a1a45c81222fc3a1bb30396d
23823 F20110114_AABGYI onstine_a_Page_091thm.jpg
1a4eafa548b9ca270a0bcf1459bd9306
20fb6016b73511a2041985c0b7525de16607a53a
64974 F20110114_AABGXU onstine_a_Page_084.QC.jpg
3d510ba0dcacdc0dcc950cbfa5419669
f1a0ae2174017999340a157483efa854d9d0026f
56692 F20110114_AABFUS onstine_a_Page_095.jpg
69320d1aabeadfef17205a76a470eca2
54f57bd93a80ee5413563a74ced9f1bcaed08998
103908 F20110114_AABGBB onstine_a_Page_091.jp2
8dc04f06a594b8eb1820ab58a3d19eef
c0c598fc5075aeea3d05663fbdeece834bf5c0d9
796010 F20110114_AABGAM onstine_a_Page_076.jp2
e62af487b635e4bf16a9fb12463b17da
d7fe162e9625a075f25908fe6f9769cd0cee9d6a
131303 F20110114_AABFVG onstine_a_Page_110.jpg
694db56dfe06a7c51e1a46c5269350c1
4aab22089a4b9b9cce48b58013a6c8a0c659a04e
81049 F20110114_AABHEE onstine_a_Page_013.QC.jpg
be6dd0a2118b87b4094b48677663f1d8
249c80eade9483a90bce961d7481e4410ede24b9
5811 F20110114_AABHDP onstine_a_Page_002.QC.jpg
39a41368a058bb96807abe3465e4926f
7430fca3ce1b84a61d0c99ee2962996877637bb4
98009 F20110114_AABGYJ onstine_a_Page_006.QC.jpg
d8998818f9828ff81a2f25bc9cabde25
8b7d009edc8ad785ad293b73c3ea526ed8bb3a28
75746 F20110114_AABGXV onstine_a_Page_118.QC.jpg
01c0b9a251fd67b48a873689ace4a862
601bfe1fe0891b5d2e69a303479bb1e50279934e
91467 F20110114_AABFUT onstine_a_Page_096.jpg
5f622c8f6f8ffe8b72826d8a176f3d2c
f3ea244905d632113e493f0bc65c193236cef82f
26672 F20110114_AABGBC onstine_a_Page_093.jp2
0bcf73206b05bd0808f94a938afcbe76
a57c9442b33ef2805e436463bd42fe689f4f5255
17833 F20110114_AABGAN onstine_a_Page_077.jp2
ad5d724eb38510372415a339d10f245c
f585510800e6f3e356bb26a30a1ce3ee0c3e4367
66370 F20110114_AABFVH onstine_a_Page_111.jpg
aaba199098c41d8e52e1e2747a474b54
6effdb03da3f4ca11133b4db477e6a6c21eeb3a3
60299 F20110114_AABHEF onstine_a_Page_014.QC.jpg
39cb382d8a84cd234397a4093b2141d5
7f7f7060098bb2ae7f7d933976097744c811077d
16740 F20110114_AABHDQ onstine_a_Page_003.QC.jpg
b9f8192446f94315e638f0910514639c
b060cebdc34e5f9c6ff04aaa73bb75cdc447c565
21466 F20110114_AABGYK onstine_a_Page_165thm.jpg
267f9c19006d2248340f2cca8a9eb812
737d056d838c448ff33e49d72a11cccb5b1ded43
74276 F20110114_AABGXW onstine_a_Page_029.QC.jpg
29586b7926e364bfa547aea04ad0d7e6
defd19517c10da9a5301802c89c0a1460c6145cb
38071 F20110114_AABFUU onstine_a_Page_097.jpg
33da5a623adade9f4828602d7d92f118
0605d4d3b299f2c3634e94132303fa00a5a1d117
31268 F20110114_AABGBD onstine_a_Page_094.jp2
bc256e0fb26a3cb61a4156b7e5afb0b4
ba46041061f28df2aaa0503d4c55767608a631af
181403 F20110114_AABGAO onstine_a_Page_078.jp2
5e3fcaf5d98f4e156058e00e55b06695
f09e4781133fd1aa46076c4810531d5004c370ff
73429 F20110114_AABFVI onstine_a_Page_112.jpg
c5e62afeea0f153264eaf78bb861e55f
5ba0b849e03059f0d5d92bbcfda49592b2436858
20450 F20110114_AABHEG onstine_a_Page_015thm.jpg
405a826954036bf4d485f7f2800e9a79
5dc9eb2dd45fc8411e2fb6aec9e8376b4c51f6c5
21617 F20110114_AABHDR onstine_a_Page_004thm.jpg
6ea4c74207ca06dec01ebd1f56a222dc
3d6cb984cd40b4cf690b7bb745da453eaa6fb930
23797 F20110114_AABGYL onstine_a_Page_149thm.jpg
9e0a87ad89ae69f1d5aa2d27bd522d7c
7c817fd3c53d2960343991fbd750a92ac91e43ff
23306 F20110114_AABGXX onstine_a_Page_040thm.jpg
2cc0fe253c78f16e08120661144f3a86
09d5727cfa5d30c8a2e9ac7678c2f4109ff71c1d
111611 F20110114_AABFUV onstine_a_Page_098.jpg
0cfd56aabf4ad978eaaf5adf233362b5
e471b5d1657d47fe6d02fb28ff3e307f3d6103a9
23612 F20110114_AABGBE onstine_a_Page_095.jp2
9c07cbf5fbef09dd0991a37e9cb288ee
5ddace043ea1518356f16a5fc0e2fe5558c69a86
583814 F20110114_AABGAP onstine_a_Page_079.jp2
6ca455f67f0d947159e19eebc8897f2b
7cc4cf76dd822660ea111b8807324adcddb83fbb
137118 F20110114_AABFVJ onstine_a_Page_113.jpg
cd9de75372f1be50204af6ed9f63487a
53b8ca136cb9473cdba52104b795b98b5a4f3c3c
67604 F20110114_AABHDS onstine_a_Page_004.QC.jpg
2e61d6b069b446c43e9e94ae1e70a737
681b38521531bc27fb9a7d3f937cfdff755dfc28
80085 F20110114_AABGZA onstine_a_Page_043.QC.jpg
e1c33fbc89d0aa2f8ec8d7377b028f7f
7e72a7ee77a79db94ad8b1ef8b255cf5b273d8b6
10379 F20110114_AABGYM onstine_a_Page_136thm.jpg
8858967e8375ceca5fbf4a9d5042d7e9
65220a645a75cd65b40e37290146d634f6241d8f
19898 F20110114_AABGXY onstine_a_Page_014thm.jpg
028ba327055c53e5221b24b391ac592a
3b1068d0f448a14421590754606aeff644c62e9d
157223 F20110114_AABFUW onstine_a_Page_099.jpg
27bfe4fec1a9b22e9258bec1b0e9fb9b
57bba111768ab5c8811f2c3fccecb4cbc575abe8
817334 F20110114_AABGAQ onstine_a_Page_080.jp2
155574bfb5aef5c9d41eb7a396ca9914
3c54fdc531c6ead3cba2db13e0101e324c07c0ec
65384 F20110114_AABFVK onstine_a_Page_114.jpg
3ef77de3c774204bf9326ca461dff501
f74b98be951989201b8f36d4f5c31d3e0d42231b
21086 F20110114_AABHEH onstine_a_Page_016thm.jpg
e04ea1ca22be6e44446ab727e2c0db08
a2e77f2eb1d2d6cb9785efcd5af01dfde93c9cb5
11491 F20110114_AABHDT onstine_a_Page_005thm.jpg
d80f92aa3bf86c2a2ae269452794ae02
32938298ab029177c15d1c00e4e20735e0a3f75c
32739 F20110114_AABGZB onstine_a_Page_078thm.jpg
4e2cfc562780335e5e00775a1c51d785
fbc39ae6bb8f40760a3cf959d1cc9252c21ba86f
24409 F20110114_AABGYN onstine_a_Page_026thm.jpg
625ff32b73fccf2bf3488f75a323aa89
894a0f5f7a8af102b475e88f9d2a5f2f7354f6a4
74308 F20110114_AABGXZ onstine_a_Page_037.QC.jpg
a4506e4eced3ff798726807c99dc7f09
f3c9f400b79eb32301a30e8c23222849854b5211
128206 F20110114_AABFUX onstine_a_Page_101.jpg
42914b719ba6f9e7aab0fc33d63cffa8
033143e21dd184dfb1da96edc31ffea0aa078f33
356272 F20110114_AABGBF onstine_a_Page_096.jp2
ce16b7416e9801c98d91638d4c266fcb
1164cc87f349c5d7f0a1d720b5c0fbf6f0de16b8
216224 F20110114_AABGAR onstine_a_Page_081.jp2
7cad9434fae5f7850ed04e636294cfac
6351670988134caa845035bed0ff1f70332ceabb
146557 F20110114_AABFVL onstine_a_Page_115.jpg
35708a0d44ff70ad4f96081886d7b0f3
46871fc3913672f3a765c02c0ddadb390be50144
63857 F20110114_AABHEI onstine_a_Page_016.QC.jpg
72e3196c5c98aaff454bee8663cb7cb5
d7daa0938c1ecc91da5b228b81d84c0c4f4d6df9
46692 F20110114_AABHDU onstine_a_Page_006thm.jpg
2c2ad9483f0148c01dc579f0d079b9b1
d0d9eb7762f982c25b5eee3785768bc154784e8d
70717 F20110114_AABGZC onstine_a_Page_087.QC.jpg
ea809fffcc569f01716b203a12579145
21dccb6d9b58319ea4da0f0931805d16a1947ade
23598 F20110114_AABGYO onstine_a_Page_089thm.jpg
42c124ca1d071944bc3a2798e3a599eb
9c4e9b391d077c5c118ea39b4831ef3173c290f4
91780 F20110114_AABFUY onstine_a_Page_102.jpg
971f99eac2c60ecfa309fb3e5a47aed6
a29786f1a4a5291948dc345c7c8f6bdb334234f4
17448 F20110114_AABGBG onstine_a_Page_097.jp2
cc7963c5ca7d8e6a2f2d2745df3110d0
b240cdce56a7c7a60a76743dc9f5746c862fcb40
114668 F20110114_AABFWA onstine_a_Page_130.jpg
043423e6eca03dc5b323a03709d51d3d
dccace88263f87852100ac6990a38e74afea30be
690493 F20110114_AABGAS onstine_a_Page_082.jp2
839d43deac113716dd37a3311df7c665
55ebb778989ee8f29829b1d2036df34d4e06a013
184474 F20110114_AABFVM onstine_a_Page_116.jpg
64e89f5ccac302876c1b7777264f3a83
e2373172f6be8382df4e5fb2df7f0101b9f0b464
81466 F20110114_AABHEJ onstine_a_Page_017.QC.jpg
3cdda85ac75c4fe6fee67e820512d8a2
27831ad071ecd105560e5cd51de3ad01cd2d09c0
126012 F20110114_AABHDV onstine_a_Page_007.QC.jpg
43c43adb8d95e22f3543a325ff7322b4
7a7be08d69778c7e2171f5adf8c942cc1a178d7f
11799 F20110114_AABGZD onstine_a_Page_179.QC.jpg
79e879680e5c7fde95cc9d9299265f02
1b14403025962b2f3c3e98f6873e627c86432dfd
53664 F20110114_AABGYP onstine_a_Page_079.QC.jpg
000efa5f55cb66859f080ab8552bb1b3
c104aac48fc7d0e62e6a656db073a78d72e4f584
403271 F20110114_AABGBH onstine_a_Page_098.jp2
f652c575e6cf451c20ff543d6d9531a8
b34f87739c3b80da82496119861e211baf15c1e5
187425 F20110114_AABFWB onstine_a_Page_131.jpg
c47316a1d05573046cb85cd0a2771937
d58ce99521745c829c044c4a5ef79dfc52f50d2f
698282 F20110114_AABGAT onstine_a_Page_083.jp2
bf61ff289800e68e412cc09deeaaff3a
0c991ccdfc3b1eaed35dfdf239e84842d2b55059
206996 F20110114_AABFVN onstine_a_Page_117.jpg
4c24d3b713e7412129bc50e5545d28a2
95ba4b2fe75b935f4ec53a8b3d12c2b1041e5a6c
24899 F20110114_AABHEK onstine_a_Page_018thm.jpg
f07f7f0c87627de3adf4dd4564c73e84
6adcf9fdf2ea2e412f3ae6a8038737be6a8cc1f2
35298 F20110114_AABHDW onstine_a_Page_008thm.jpg
def264394429bf0e5dcec0bcb81d092b
5ad62292aa36e34f29e2c375fb6edfb12a47bc18
25318 F20110114_AABGZE onstine_a_Page_117thm.jpg
c0290c33b5943096a8dcd9e259c40eec
e7d7efd6171dee891056f4a97579e6e8889bc49f
6647 F20110114_AABGYQ onstine_a_Page_135thm.jpg
31eae1cccab6ff805c84b97546c79936
f2b742cde6e6f4042ebca5bd0bf6a740ef0fb2ac
76297 F20110114_AABFUZ onstine_a_Page_103.jpg
c9b4511f19aed8c8264fae3174145954
d22a21074a474e032e6632e82c85812c83c6f3f3
1042323 F20110114_AABGBI onstine_a_Page_099.jp2
96134719c6eee208e0d8a33f1ebd4b5a
0708dd9d6989ef21c110c3b8bd8cf087aa43fa5a
200224 F20110114_AABFWC onstine_a_Page_132.jpg
a3aae72c230fb06147bf890da4bdcb19
73f12a5a31bdc5bb09d53b0bcaebeeb3dee5cb2c
925378 F20110114_AABGAU onstine_a_Page_084.jp2
3bdf4472fd269c9c0646227cc1330b58
fd942453ed8af204a81b4f0ea4bd9c3d4b4b7459
204226 F20110114_AABFVO onstine_a_Page_118.jpg
e79c7943e8b140228b5f04b0aeaabf54
46fe099ec700e3e100283ff600d564e6e44b296f
51754 F20110114_AABHFA onstine_a_Page_033.QC.jpg
ae57f34bbf3cc226cff63bb7850443a8
298dff10f8abfeb1fffe93aa3024ba6262976569
23597 F20110114_AABHEL onstine_a_Page_021thm.jpg
448d195981364e0817dedc36a8d0fb72
6c2d2adfe4dc2f19770b49d3a9dc7a144585409a
56663 F20110114_AABHDX onstine_a_Page_008.QC.jpg
b4a6d93383e376f27acdb10bb73cfdb1
72a9c87e18da7b5e3a3d084420d013e8cf658a09
33096 F20110114_AABGZF onstine_a_Page_173thm.jpg
22aa21be97fac45c45c9ef07e3b87049
f60127cc2896b57d81dc72a6741e4a7410465172
36304 F20110114_AABGYR onstine_a_Page_137thm.jpg
30f292f66d6223b61089a7f67d7e61c3
9a9fc12e15839f50bae45ebf0c79e04432b7e5f5
581849 F20110114_AABGBJ onstine_a_Page_100.jp2
5374f26efecc091bd55a35aadb997190
205e8e85911d53f8632deff91864aaf23833c79a
205528 F20110114_AABFWD onstine_a_Page_133.jpg
d16a890838d5331c525d54e9abcbdcc5
69d92fa809ccbbdbcfe78cd412297e3c89d1f1d4
470289 F20110114_AABGAV onstine_a_Page_085.jp2
64bafeb58baeb23993e394be7e435211
295e14cbdeeb12b8bc8177cc6483ce0259ab35b8
215958 F20110114_AABFVP onstine_a_Page_119.jpg
41c83cc2956ce57888c4c006a8493552
31b6f3ef12a5c42aed104f879dcb08831bb3a6c0
21090 F20110114_AABHFB onstine_a_Page_034thm.jpg
5b8d85fdf2eda7097b72998b90333962
1cddf47c5cf0f4aae17d814d3d08218204e6c175
73308 F20110114_AABHEM onstine_a_Page_021.QC.jpg
2041597197d1d5200b7b3ab426cafdd1
497f8e860a371ca035525ca99e0237c87d718919
36980 F20110114_AABHDY onstine_a_Page_009thm.jpg
f61cd3123b5b04253ffb1846a96ad726
b6389376a0ca0ae94ab7e17e7ace491287d201de
21712 F20110114_AABGZG onstine_a_Page_156thm.jpg
214e9d3cf39f9cff40ec91d44f699f13
62f35487e8143b575f7860b981f008d49a0649f2
55413 F20110114_AABGYS onstine_a_Page_124.QC.jpg
d56c5c17adf2d2502520005c13b9ff32
7e2a50e6efabc22211c7614cddaed3cdeaec2b4c
770111 F20110114_AABGBK onstine_a_Page_101.jp2
28261fbcef6d963fc3194984f0042e65
43749c2d0d4fb0e031fc0dea0ac737afa4eb2d93
50360 F20110114_AABFWE onstine_a_Page_135.jpg
c174f47e873e8179858eeec990f6f387
e3b95db133c577b718b0260b086535b2d4a25ef5
1051944 F20110114_AABGAW onstine_a_Page_086.jp2
cfe0bea95a9db998cb03458e557e8a55
ae656231e4b5c4088941df4ae2f7ed8f513f0d20
54657 F20110114_AABFVQ onstine_a_Page_120.jpg
9d64c9bd3d34a9000cfb046ca8ea18f6
68d0e7e884ccff0707f3647d54d447fc60662974
63670 F20110114_AABHFC onstine_a_Page_034.QC.jpg
ec6316392281e79d605510d46ad68ac8
863781c282dad1251172a50b608bf0366bc54ecb
73355 F20110114_AABHEN onstine_a_Page_022.QC.jpg
c303e2ffd8f3a16b75ce5f846b6aa055
fe438f8f34b546d73ea30445668a80c9c0225bc4
60561 F20110114_AABHDZ onstine_a_Page_009.QC.jpg
82a028fc8e77ca0ed4060b03dfcf7728
09716d53111bb7fc4c363344d8798c8b47bfb4e9
20723 F20110114_AABGZH onstine_a_Page_147thm.jpg
0d6e601d990543bc879f322762aa91b6
550e5e13ba96b88f30a4b1002021cb97489b6d82
35060 F20110114_AABGYT onstine_a_Page_058thm.jpg
6b2e87819e033e89be1e78561a1aa93c
8a499da1f522b64392ee53ffd92c76b5eafce9c3
41689 F20110114_AABGBL onstine_a_Page_102.jp2
3f0fad222fff4b42974fb825336586fa
50ac96c3292bb5886721e51850f97041b26ea8b5
86329 F20110114_AABFWF onstine_a_Page_136.jpg
c0f51f1783d4b4237f98284db281e842
af2f471801798e0c5b0cdaabbadfbad55c31c7e6
95269 F20110114_AABGAX onstine_a_Page_087.jp2
1d024993e4d8fa3d03d1f3f833195924
7d4818e548108e3c9fa952d37990ee26ac40d860
92145 F20110114_AABFVR onstine_a_Page_121.jpg
bb56d99baacde2c85fb24c429c4b2b7f
ecef5f36ccfccdf0893daf4c83cfdea305c69284
105617 F20110114_AABGCA onstine_a_Page_118.jp2
34a81bf45d34e3b2e14c52fc9578fa1a
8d9870854b8547f46765ac8859f4b3af7a171aef
25115 F20110114_AABHFD onstine_a_Page_035thm.jpg
64ea3952988487bf9b9d9465c6f6442b
6a9d73f2d3867b35096ebfab2872bc22a06ae3ce
24200 F20110114_AABHEO onstine_a_Page_023thm.jpg
b8c9cebe79560e14616363ce8c09e2f7
fa8aade3a6abd13a14c92450fc4f46c3547fedc1
57681 F20110114_AABGZI onstine_a_Page_137.QC.jpg
f29b4c91beeddafd9fba9e2ac1b136ea
45d1eb0541b3092e5f1b4d4a1573527cc11baf73
22758 F20110114_AABGYU onstine_a_Page_029thm.jpg
c425348c3eb0b18b31f37505fa07990a
4c48f4e72f9887e3b9d0fd6c81b449e963089deb
32721 F20110114_AABGBM onstine_a_Page_103.jp2
eb428fe8792132e93de61fb5d72e60a8
599137e2ed90afda0e6d79bdd30fb255571a09a2
118182 F20110114_AABFWG onstine_a_Page_137.jpg
f1639b776fe41d3718ee39cbab7ef895
a05388f328204e9a8306892cfd095822efc487c7
116795 F20110114_AABGAY onstine_a_Page_088.jp2
04bb87fa3f0a4af8d8c2c61407c844c2
f4db36c6ae9f17153d262d81256656349fc14314
93696 F20110114_AABFVS onstine_a_Page_122.jpg
45deffb5b27ca74dc14cb43571f1871d
49342e4379acd53aea7f5d7feef7dbc427a1b5dc
110623 F20110114_AABGCB onstine_a_Page_119.jp2
ebfe848bbe0c26d2c26aea92f0fd6e6b
c81d5ee84cd0d3e1dce534c319691f9fbfb472a0
23786 F20110114_AABHFE onstine_a_Page_036thm.jpg
db728ed24c184317020df8aa416343b2
3ff718bc36d1078f8a7790e6144876de0145b320
77412 F20110114_AABHEP onstine_a_Page_023.QC.jpg
308e40c4733d5c2a7d84e03dea92e43c
04458911d674879f992e377b70050e766f2dc07b
22946 F20110114_AABGZJ onstine_a_Page_093.QC.jpg
9a403ed1390350fe7c07ac2a0dc718d8
6eab12bb5cef085bdefe5d1296657faa2cad1725
58332 F20110114_AABGYV onstine_a_Page_076.QC.jpg
78f535778ef72fc5c5a4b1c8f4913668
796000e48ca97621b657e819dc40e1c2a1e9be47
30733 F20110114_AABGBN onstine_a_Page_104.jp2
4b8f7274b3df3243c93ce2c0801c213d
55e0e09aac10ab29365be8e55d58cc2535d41066
51661 F20110114_AABFWH onstine_a_Page_139.jpg
99deea2700ea1a9f5091c7836bd1be5e
d78e92f15887a7fb68586789cfc8361f00bd43fd
102149 F20110114_AABGAZ onstine_a_Page_089.jp2
c77889549ad2560724c182a9dc17f5b0
26bff2d93eec043ac94e2b538e458ef7a972cdc2
118933 F20110114_AABFVT onstine_a_Page_123.jpg
85416e818b61b2b29d6132ab5d5c4122
a56fde9a51b4b7595bda8d5d98764cf7ee5403b1
26458 F20110114_AABGCC onstine_a_Page_120.jp2
9e7b5f79c618a09608d8bc492fe3c820
c1746e18fc1cb8322958729b1acef875c21db7b2
74036 F20110114_AABHFF onstine_a_Page_036.QC.jpg
7e717cf158a3578cdd1357562356cf34
30bc98923d27adfcf840312c1fac7ba2f22f984d
25431 F20110114_AABHEQ onstine_a_Page_024thm.jpg
9ba5235025a750a2fefd7c0c29135584
1f3d13a9075b2202cbc49b10ee18fe638efe14e4
70756 F20110114_AABGZK onstine_a_Page_063.QC.jpg
609d4996dfe82c023b6f798b8f462f4d
aabd70bf681b2d2f6995b4b036c9beb875693072
33339 F20110114_AABGYW onstine_a_Page_049thm.jpg
bb951aaa2ec215cdf54d49bc5c324734
c0efb878208705a2545d90390b47c887ccd132eb
318398 F20110114_AABGBO onstine_a_Page_105.jp2
0119f08bba6d42b94e16a37b06796a08
f5ad407cb99175cc754702dc5fd2bf15db0d36af
56449 F20110114_AABFWI onstine_a_Page_140.jpg
3d66f47f269c6ccb80b4b6d4280add49
69346308522ac40e197fe0f27835924a72518010
109599 F20110114_AABFVU onstine_a_Page_124.jpg
a0e3e494b8986f2e4a8f102c1a553dc1
3fecb4ac8659ccd76a38fb189f9d5edb0ba2c08c
315970 F20110114_AABGCD onstine_a_Page_121.jp2
b4d6385690f98d0c66a344c5f928c7a3
85a7e810a5a4e1593154dd62d91e38b800a29b31
23541 F20110114_AABHFG onstine_a_Page_037thm.jpg
17ca02d0793e72ad8cc366818b83cc4c
71104ab87b3d5ebef789a08ebca946405dd73493
77417 F20110114_AABHER onstine_a_Page_025.QC.jpg
e49a9dad86cd2cb708958b95ae55d157
c139716ff94dea67f358088a205a4ea9ac173bfe
80568 F20110114_AABGZL onstine_a_Page_018.QC.jpg
3b1f0f99d5f1fa49bc84dcea963e21e1
bb6b6c6a537e92008e637e0145931b45dd3de46f
51368 F20110114_AABGYX onstine_a_Page_070.QC.jpg
01dc0803dbfc73706742a58e36a6b31f
2d9abd2b3bfd80205f76aee741e122dfc22c1a68
1051925 F20110114_AABGBP onstine_a_Page_106.jp2
dbec5a3143c78a9927502f6a55cad366
47527e0c8d91e206d839fd2927a9c0aa022d596e
73273 F20110114_AABFWJ onstine_a_Page_141.jpg
dbc665c798baf783d8d25c6e508d1e55
84e396a8f78ae16d1d06c5af094ab1a6a55cc971
60421 F20110114_AABFVV onstine_a_Page_125.jpg
a14c7f4cc921ddbb15164595c8e9f84f
2a9f57c0a209522c754bceca3823b8e32eb02746
302346 F20110114_AABGCE onstine_a_Page_122.jp2
d77ed14e07b144dc898dcde3b0904000
eb6f86c7448765ac608a7c5513537f0bbafe9732
24024 F20110114_AABHFH onstine_a_Page_038thm.jpg
50964fbdb0a1c9064ec23c921d5763b8
9938e69921757983fe0927ef0238ff996a400bf6
75515 F20110114_AABHES onstine_a_Page_026.QC.jpg
36c6e14666af7cfc7a6dc3edf74fbe64
d30cf4e1b6a4f0296fd32269a039f63e4ff46ab0
20472 F20110114_AABGZM onstine_a_Page_095.QC.jpg
2305b8b96c41d9451c2590999f6cb51a
0d78b654338bb6e67bfb00cad9d156ec053f1d91
44770 F20110114_AABGYY onstine_a_Page_032.QC.jpg
87234b336fd3e83ff2e69a5eaab6401d
7ea7cc269c13c76081ca44f34242c31108fca5b9
27155 F20110114_AABGBQ onstine_a_Page_107.jp2
97041ecace76232c73da36e1b0e6c1aa
749132aa3fc58e0ef5fcfcb7ccaf3c81ee77494a
56082 F20110114_AABFWK onstine_a_Page_142.jpg
3f763b0cd356325129c91dfc6a3e543e
16a1c3a021cdeb7f566deea139a2aa3f9515c4af
70990 F20110114_AABFVW onstine_a_Page_126.jpg
90393afbe339d4c3211fa746a4f179b4
611823233f4bc6e8f5d8e3c7ecb9ee9a7302bacb
762952 F20110114_AABGCF onstine_a_Page_123.jp2
08141b9a80c8e895a3492825598607a0
99f3c4827fdd21754c1e4b458f105fcae01264b0
24966 F20110114_AABHET onstine_a_Page_027thm.jpg
4b5561c134a26a718e3b7d342dee021b
588e6b589e34b0dd0c9eef7dfe1309e6b1edb20d
33068 F20110114_AABGZN onstine_a_Page_126.QC.jpg
2d64cf3c5b256b00c938829f2d65afbe
e11a281941113a5978cbfcbd40ac8a067e216a25
39574 F20110114_AABGYZ onstine_a_Page_128thm.jpg
4c1f94545684d2d3d37a877fa0e6af4a
e059fce42f7eeb661e9a7a2f4ff726063eb83821
24338 F20110114_AABGBR onstine_a_Page_108.jp2
3a8b8e9057457ef5870b18a87b45fed3
6ba4bfa979e3ec780deab02192c1ebf1d6636465
67806 F20110114_AABFWL onstine_a_Page_143.jpg
229f166f349f4b858195557cb04706cb
432d23a9ece9ba993f0702aff78c290eedb0cd68
148747 F20110114_AABFVX onstine_a_Page_127.jpg
8b9b8bc74800464c2b9e697194773392
4a124ad61297056f3030f918bd0cebb267814322
24129 F20110114_AABHFI onstine_a_Page_039thm.jpg
b8a8be015e6a813a07e4033c0ee18d1d
57cd3c963d62d752458b656bc995477d36d88474
77836 F20110114_AABHEU onstine_a_Page_027.QC.jpg
44866119e68922d4b3b1a7e27664fc69
f8401678b6d6a44434d18a58d081a72ae41a82cd
48904 F20110114_AABGZO onstine_a_Page_163.QC.jpg
3fb5ba893a0094611329ab2ce3cb82ee
ff6398fd2b5c3d93967b0e8857efdb0aaf855d7d
200331 F20110114_AABFXA onstine_a_Page_158.jpg
1739e671b79c37606f9f4e4ff8cd9399
b07cfee5b0e6196de539ad5fa21a096ad6e13f9d
222993 F20110114_AABGBS onstine_a_Page_109.jp2
b9c02c9137574a8d089491bbd3c42bbe
7c890febfe194c68e65d2e556ce3534e9bf9c816
57803 F20110114_AABFWM onstine_a_Page_144.jpg
ca1b8d86c19439898a6d8a9eeac91aa3
ef05d4d80d63c7982eba89c1fbb47e98150f835a
175253 F20110114_AABFVY onstine_a_Page_128.jpg
779717b3ab5f2d222ad2a11e6daf2097
7f287ece8dd531d835a4e826df6eeacf332eee84
462063 F20110114_AABGCG onstine_a_Page_124.jp2
98f7681c6ba3190a4533159baec29f96
54744747b5b5ce4732410eb57fe7913e20eef47b
75679 F20110114_AABHFJ onstine_a_Page_039.QC.jpg
916be810b74382256db0b11fbe7dcdfb
cd83fa7d9e21ce6e28f27653990cad701df074e4
81529 F20110114_AABHEV onstine_a_Page_028.QC.jpg
470f70e013723fff9b4db8abfab4b7d0
4878338d624cb39f61ccb41b09a4f42bfc97bdec
30912 F20110114_AABGZP onstine_a_Page_136.QC.jpg
0ca77cbd32616e10004a82658e685b05
ff55b30285273037ccd3827546512a09ec1984a4
164362 F20110114_AABFXB onstine_a_Page_159.jpg
28fe70f29d5c408c44c0bf1cebd75173
bd46580dcce6f9d6f9783173bc4ddfa708dba619
660691 F20110114_AABGBT onstine_a_Page_110.jp2
2ce2775aed62ca17c099971e055c0e5e
cd04d2b26f902cc624751d43b90a2047e2c181bb
63489 F20110114_AABFWN onstine_a_Page_145.jpg
5c8dc95f90c28dafc2896df75114b1e4
f4707a4eec50f62f69917c67134891aee183a2ff
147834 F20110114_AABFVZ onstine_a_Page_129.jpg
ef725b9c5e70278c2e5dff3a91a88eb4
1454aeba87d55e17edd6711d2aa91437f60d66f6
27743 F20110114_AABGCH onstine_a_Page_125.jp2
4c9383a136f9e7bdfba12284da2a2ac9
907a2be9b925eaa2210a2fd5dd1dbc9acee82548
25445 F20110114_AABHFK onstine_a_Page_041thm.jpg
0509502482fbf4c5a1a65cebb46c9bfd
b7ade979cbc03e023a3ea45e89d38b99faf4cfbf
23182 F20110114_AABHEW onstine_a_Page_030.QC.jpg
6c093057d716e53c3b69093c87564fbd
1023d62553dd38b02160420a234822b950e90706
84971 F20110114_AABGZQ onstine_a_Page_088.QC.jpg
524425dfab6aca972dfedc090e072ae8
224e50356c1c8a960900b46f410746220648846a
133769 F20110114_AABFXC onstine_a_Page_160.jpg
54c87825545dfd6c4beb83e771ad16cf
2500062df614b559dc27799330377665a886246d
38689 F20110114_AABGBU onstine_a_Page_112.jp2
6a5249d97e9b9efe1584a1806a3475fb
ca89b61fa94e7d66cd2978111ecd59df51231212
58844 F20110114_AABFWO onstine_a_Page_146.jpg
71c72025a870c2ee7d76da82dd2e8408
f2bdac0066e9661bc79fce470c9f43cd87f0bfc3
270073 F20110114_AABGCI onstine_a_Page_126.jp2
4ced07aa451f64fb8f2f3ada65334d46
9cb558c793c74fee6f79f199b1fe3a0883420b6b
52964 F20110114_AABHGA onstine_a_Page_060.QC.jpg
a0feb3075658ad4720088f690c8f8d66
c52e23f8422c83a9efe770b085a610a7f78e22d6
82918 F20110114_AABHFL onstine_a_Page_041.QC.jpg
911e8a615c01da8a1c1ce27b59b9718e
76354692ed12363f8a5e822c3f41add97396987e
9684 F20110114_AABHEX onstine_a_Page_031thm.jpg
f8f690799da05c87f9c6af63b494101d
3ad3be13435d195f87ccd7c2eb8c2417078844a6
6980 F20110114_AABGZR onstine_a_Page_120thm.jpg
ff7b9cdfd4c9f9f009f1f3368eaf72dd
78bb7f8386ac70aa49afad003b2316bd0809dd14
55008 F20110114_AABFXD onstine_a_Page_162.jpg
a47b41fe0a6ce1a983ce5e33872ff05f
df9a5cad61355ca84b4efd45c94048cda2585e65
838538 F20110114_AABGBV onstine_a_Page_113.jp2
2b602a073376dc215b3f22016b8ac9ec
b14794687721082cf9cf875a792928267a996df5
174226 F20110114_AABFWP onstine_a_Page_147.jpg
1c9d75c8fa41cb1e86e890bde813461c
b773a5fbd74294571d9701cf6f44e3e9d21a2e0f
908647 F20110114_AABGCJ onstine_a_Page_127.jp2
5846703678652d66b2e96341197508e2
59b9adeaedaa70e3294abbe0ff894283541b1704
56339 F20110114_AABHGB onstine_a_Page_061.QC.jpg
3154579ed7be5bd24c3212221434e990
a334c0b977e76910e9fc6db806abaf4077f49dc4
25647 F20110114_AABHFM onstine_a_Page_043thm.jpg
922d8625d8bf37992fbb524f1db27ab8
0ae65689f0f7a852661c9bc73cbcbddd71f37b4f
32316 F20110114_AABHEY onstine_a_Page_032thm.jpg
b53637590ccf377546386efd64b48617
6aca56e24007f2c2802d6943052066cfbedfde87
66675 F20110114_AABGZS onstine_a_Page_082.QC.jpg
bc962e0a1d22c7ec36b1db093878f647
414331847d6ae50a8ebe7e9a8bee99bb78f90407
82268 F20110114_AABFXE onstine_a_Page_163.jpg
de335301b39b09f398c852745aac2d54
3f9db6291c9ba60d9d6e7fc49543fb4e24129abe
275783 F20110114_AABGBW onstine_a_Page_114.jp2
6cbaa85a2e431e7fac5970dfcc43fc40
81023f4c5bab51281fb3efca750b99c59af41f04
205664 F20110114_AABFWQ onstine_a_Page_148.jpg
bbda1234076802af8b7b197453a92679
0d3031b76d4e4aeaea162f5cb6c22ef4af8a809e
915269 F20110114_AABGCK onstine_a_Page_129.jp2
3dfca9579a7334502c3f0faa78129635
b85bbac9b0e77305d1eba7448659edefcbb7ad04
16257 F20110114_AABHGC onstine_a_Page_062thm.jpg
1845f01a2eec66387410f316a19d8c38
37394d84ff70363444dfb84580d727e518d2ffc4
25298 F20110114_AABHFN onstine_a_Page_044thm.jpg
f0cf248bbe71e22cb4a85918b4ef0537
30d7297425a7db431bc7f70871f25cc946d967df
35381 F20110114_AABHEZ onstine_a_Page_033thm.jpg
76a03f11496999176f9be1efd5525638
899aeea697fcebe95ce429214024b4c2142ae0ee
25208 F20110114_AABGZT onstine_a_Page_066thm.jpg
070c55a7ffdd41fe24a41575c798df51
066a2b5b5301df64dd8328077cb6db4c4fb6e8ae
61772 F20110114_AABFXF onstine_a_Page_164.jpg
3aad604e7505442e78972384bfcd67be
77bf3f6f5f65c0ca6ac97f5ced8cd6a0d69ff3d4
979555 F20110114_AABGBX onstine_a_Page_115.jp2
b1ce610c7c831b057b2f62bfb53163aa
4ba4058ad545c7b8d8bd7043706c6e1994a3f324
193953 F20110114_AABFWR onstine_a_Page_149.jpg
0e584788521c5108e685515a08eda37f
585cd90a9367bc6c299d098a9fd5673532087a65
90388 F20110114_AABGDA onstine_a_Page_147.jp2
fc2c3740a7f83aff473fcc0637fbb1a9
bd1476f3e791b243f60817e0af8c1f71215493e0
574268 F20110114_AABGCL onstine_a_Page_130.jp2
bc38c1b0d10b7b04870434e8316f690f
dde6e46e8b157f99a649a60c6101e34328b703b2
27394 F20110114_AABHGD onstine_a_Page_062.QC.jpg
1633ad172dfe67832e2f9fcfbf45fe1a
15976e067301f357bb49068e9d585c82a67daf84
24538 F20110114_AABHFO onstine_a_Page_045thm.jpg
7dc879c81d693078e0d4e8ac67906cf5
08e36449d4f10aa03cb663ec88b31eef7fc0a944
22836 F20110114_AABGZU onstine_a_Page_094.QC.jpg
e5c2c49bbfa829242533df0074c5f926
5a464a7642b809ca79005e02311a4f2fc33626a2
182886 F20110114_AABFXG onstine_a_Page_165.jpg
f75a404113d9ac13ba764c219273d7ef
58e73ea4c4a4d8ec840cf69c053b1a9f901886c2
93061 F20110114_AABGBY onstine_a_Page_116.jp2
e19adb67cf864578061b631be15c063f
a85c61b18fd522c74a0dfd294f1dabc82d5019dc
144417 F20110114_AABFWS onstine_a_Page_150.jpg
f5f47d65a6813acce86e2bd8e32ccf18
8cbecf7c3b984aa317e6cb7ecceb0e5177360cf8
105922 F20110114_AABGDB onstine_a_Page_148.jp2
52acf978278cede908e4d4e43db3d8c1
8ac3077c9b0573b15df5b33e896575aaba2c8ef7
95047 F20110114_AABGCM onstine_a_Page_131.jp2
a54a134a4a5f9fc4da97f48dfef4b47e
d01254c92bb0c375b60017441c6013b24cd14f75
24778 F20110114_AABHGE onstine_a_Page_064thm.jpg
f56dbb8a3d2033b03ee94b7d8acba057
1e579f0201ad22005cb8c13661ddcfb3547d1d1f
78960 F20110114_AABHFP onstine_a_Page_045.QC.jpg
1c629dea39118f1042cfc3dad95790f8
287b31bb9621bac7a10972036c7ca9a56ef5d764
31063 F20110114_AABGZV onstine_a_Page_164.QC.jpg
96548a370f3c02c4415895741a2fb344
62cc1b504a0f7b91f5804877f11f8acb4fde1cab
201803 F20110114_AABFXH onstine_a_Page_166.jpg
15c1c712d39243ac6d3200f6c72df1e7
83a147f25f85b4dbfd710f7bdbca1bb942d8bdd3
106779 F20110114_AABGBZ onstine_a_Page_117.jp2
8914da44c9a0989c6864e5df1ed35e78
5e01e999bbf58b5213ae59243468b91db7c66387
175572 F20110114_AABFWT onstine_a_Page_151.jpg
b145b046f4b928c51c74caf4479dd906
c48db7454a54c815e99952eb4040fad50a549832
99366 F20110114_AABGDC onstine_a_Page_149.jp2
b37740af3d63efa1851a994f200c0581
72171f34ae93c3e9790a789360bf9fe62afd0ef5
102519 F20110114_AABGCN onstine_a_Page_132.jp2
76bb3ea8b6f22c9d82d948ff899072bf
208aaf2b91b9ed400d67e9bb7b8755b7125314d0
77351 F20110114_AABHGF onstine_a_Page_064.QC.jpg
424a319bd91940762ed525ab845d34b1
319370316c92b1bd86ea9116dbaa6cbeb22d4790
5632 F20110114_AABHFQ onstine_a_Page_047thm.jpg
3d4d08132b3e37ed6431bda303469d19
037d1f0af0f20d9d8c1e9f2bad428ae5812a8e74
52685 F20110114_AABGZW onstine_a_Page_072.QC.jpg
84f2c68d4065ec64655eaa45dd467805
f05e897c5ceaa280fb33687d5c25e787ee28ec6d
134540 F20110114_AABFXI onstine_a_Page_167.jpg
e28b661d45fb11091a1b435999476c12
ab0a31f4f623c28df3edc957afc7312621f11452
193486 F20110114_AABFWU onstine_a_Page_152.jpg
a5b53d506a656b17a39a3d570ccfd50f
137b635e162f8da9d31113d8256fd43126fbae5d
76901 F20110114_AABGDD onstine_a_Page_150.jp2
83babb22f37c015b2b22037d732b59f9
156ad588ba31efa5f95f21c9e16971a8bead24a8
106935 F20110114_AABGCO onstine_a_Page_133.jp2
b7813537ccf13924798e030506a76f6f
317140261e0ad47345f412df5934e2ed1fc78cd0
77861 F20110114_AABHGG onstine_a_Page_065.QC.jpg
1f4026fe55c5d75f27bfc8d5ca20ad53
2baf8f8d6c006700ecff9aa85ef008096d0b1ff5
16538 F20110114_AABHFR onstine_a_Page_048thm.jpg
a6454d5044902116d338d24b060ba6fe
106b6e84ae51fae80ebfed45a38c99d397b01b2f
9479 F20110114_AABGZX onstine_a_Page_125thm.jpg
d059c9370e11f9f754b533901e3d2b33
8b7d808716042bacc36485c1342dc9ff36ce9b97
50772 F20110114_AABFXJ onstine_a_Page_168.jpg
a1b5ae4e3fe3df445799bc9ec9a70592
3d09de316724d4c07acda080ea1abb85d8fb1b6e
119832 F20110114_AABFWV onstine_a_Page_153.jpg
e36062cc62de75a3be960809a196e627
dbca53e61912483382089bfd4275fc91a4f9fe84
91022 F20110114_AABGDE onstine_a_Page_151.jp2
b90e1f9488f1881f9e67c313de0a79d7
076ebbdaee94b7b85c092498e39af3fee74d004b
60698 F20110114_AABGCP onstine_a_Page_134.jp2
4fceaf75671e9e7a652c9386682729de
0236c25e53c7e63f3eabcff21270166b4366823d
80172 F20110114_AABHGH onstine_a_Page_066.QC.jpg
ec1afd3016d037e69b54c07a05eace63
12346df6a91e3b2e908e51db038bf774d7141ae2
31638 F20110114_AABHFS onstine_a_Page_048.QC.jpg
d8a94453e4dfa903fbfb1b902f2d0e20
7dffeba79a206ac5ba9df1b35d50be1db23b3577
38108 F20110114_AABGZY onstine_a_Page_086thm.jpg
eb0dd33908f7fb6d3743e1bee443cf75
634376c2e9b94dd2912c82dc600aa91f0b835029
36084 F20110114_AABFXK onstine_a_Page_169.jpg
cf07e381c1402f18f5bf9f4560880d0e
7ffc2ba3861c6561ec1b6fffb170cd24a8740453
182295 F20110114_AABFWW onstine_a_Page_154.jpg
20fae722f403a8ec3ea953cb638e67e0
c88d638ea1ae3b05802d2de30f90160633985e03
98497 F20110114_AABGDF onstine_a_Page_152.jp2
2b6a59f931b3c0db315acb6c3c07571f
0abfbb08781bca40cc9aa86fc4aabc8933d4f311
F20110114_AABGCQ onstine_a_Page_135.jp2
2b29de98d57c08bca24d1c8ebb0121db
9e6d3b2a819c738213cb17d411c3d06410138298
25026 F20110114_AABHGI onstine_a_Page_067thm.jpg
3ba246402e45813e35995d467dfaf8b6
fa3552e4bd3cb6975f8c208ff53745f7ac1fa44d
16948 F20110114_AABHFT onstine_a_Page_050thm.jpg
99889d1725c174970e782a5ecbcb1640
c7eddb974ca566eaa2df8a76acbcd7df55683b24
41835 F20110114_AABGZZ onstine_a_Page_078.QC.jpg
20ec76f920f44609f5b23c70a9608c23
67c823bbffcc6dcda65d08b5025f4edeca588617
59776 F20110114_AABFXL onstine_a_Page_170.jpg
eaf30254370fbe2f447461ef4ae05bad
331984260268ca1a5ea3c295c140888b763fc04e
175412 F20110114_AABFWX onstine_a_Page_155.jpg
68a281cdb3cef5493e27813fde8215f4
fa5426ef99c62b8c14e1c42a1dea2b50626a5d26
95960 F20110114_AABGDG onstine_a_Page_154.jp2
e76bf5c5ff0d80bb784d13d295090989
5d2275fc3d13779fbe8f335c1bd1a5fffcab648b
40496 F20110114_AABGCR onstine_a_Page_136.jp2
3bee4f776c455724c3c2ec71eb8f71fa
27db23570f28f6ffc7e896290f5c17d86f0aa4e0
36742 F20110114_AABHFU onstine_a_Page_052thm.jpg
19357d3916c333227c46b08b7a17c9d1
abbbabda4f24f4390f4a4bdebd0b058b80c4686c
122504 F20110114_AABFXM onstine_a_Page_171.jpg
eb1e3a5dd97077cec0525dbd7f2ad462
66cc59e3091bbab31f80d52a6d206158b1538d48
167526 F20110114_AABFWY onstine_a_Page_156.jpg
8489dc0e7fc283ed90da70c7bbfb22f7
4716be53c0393e13b333d93a66de2c7efa03f903
1051986 F20110114_AABFYA onstine_a_Page_007.jp2
cd6587877347536f2ac6ae956f6e9431
746e28864f5225ae10de85de42235881e60c6191
475845 F20110114_AABGCS onstine_a_Page_137.jp2
a5df8aa9e4a106da261155eadf365371
e42ba5f97a72ea2104f2cefe49070ec39d66366d
18670 F20110114_AABHGJ onstine_a_Page_068thm.jpg
c2fbbf882cfdcd60eb46fdf225e26ae7
ae65e156938924bb90b2c7ffedb0bd7699bc52e3
51235 F20110114_AABHFV onstine_a_Page_053.QC.jpg
597bbce59fff065a47967dd62c09b271
b50b0d5439f54a8eed1182070fcab8d730a97cdc
104728 F20110114_AABFXN onstine_a_Page_172.jpg
715dcb0d9d68bb8b504fd1022078e5ad
d67404f45b30e4ebe567afbe021925fa74e19c12
162182 F20110114_AABFWZ onstine_a_Page_157.jpg
25062ea5058a7526ce996131be0ed1bb
46eca51e54df4e7a5fde2093fa2a4727d3e857ab
90955 F20110114_AABGDH onstine_a_Page_155.jp2
0783af546c492929cc0ec40154f6e389
a6a2f4f6cbb044491444351cef68e5dc34011a45
731659 F20110114_AABFYB onstine_a_Page_008.jp2
34788414d7df9a7e93f7095352d33038
cc04370db6eb68956e1ee27a3886f28dd5751c1b
269629 F20110114_AABGCT onstine_a_Page_138.jp2
176da28bf53e523618ff6954d4360a92
624c2f37289aca9eec975e39d267c4a8035c39cb
6996 F20110114_AABHGK onstine_a_Page_069thm.jpg
43d6e2353b82ad5046c591628c4fa597
568cac01fcb68219ce101430653fcc3aac6411cb
56440 F20110114_AABHFW onstine_a_Page_055.QC.jpg
d0692fc3b77bc65d3ad6c5dcc77eaf68
60358c52599b9327128d830b091699615928b5fb
92629 F20110114_AABFXO onstine_a_Page_173.jpg
2aeca12ff1ff655cba3a28442aeaa3e2
bdb0a39bece2f4b6684d23eb7672f96701b5f0df
87238 F20110114_AABGDI onstine_a_Page_156.jp2
651e58d7b20bdbb9eb60eb787f2ed66a
bc678a78b172ac9f4304c55f90b40e0fd63247bd
791202 F20110114_AABFYC onstine_a_Page_009.jp2
01a386a4328d8b3e081830f3b2208165
94ffe0d8793b6c1eebabf5264ca71c1d57d4b341
22647 F20110114_AABGCU onstine_a_Page_139.jp2
71f8d56b2f32d6f9b07b5becaddb5ea1
95eea2603fbd89c19f1ae7820c0449df4b328054
25622 F20110114_AABHHA onstine_a_Page_088thm.jpg
5d290b10a1072b27e4e92d5953cc8c09
7a4228c10c2c22c54d10d040dcef46cd4023d974
19610 F20110114_AABHGL onstine_a_Page_069.QC.jpg
2bab4755153f8113f9b655bce9a5787e
8ede1a09e62acbaf67f40fb424cbc58b78b613dc
36030 F20110114_AABHFX onstine_a_Page_056.QC.jpg
14edf8d61524af61dbdea77aba0ea334
bb14ba2b9c623bad91aa76431d9db92266e6ca4a
202424 F20110114_AABFXP onstine_a_Page_174.jpg
45347ad610d07a9a725e97540bb7ade0
71cc8958e0894d6e826da55aa72403f1a8da2f0f
82927 F20110114_AABGDJ onstine_a_Page_157.jp2
4ec97914ee71af0d5c74e16f02e729dd
6b73b2c415ccfb347420ba68b2b512afb5eb03df
F20110114_AABFYD onstine_a_Page_010.jp2
8712e8d4a8b8ab3000eb8ab18ae95af5
91bc9f76dc842a35a55a73a67b4c0a2b740690e7
24253 F20110114_AABGCV onstine_a_Page_140.jp2
ee598c28dc8715d54788db8ce4b6793e
1629beec15bafebda9aa11c87f3bf0369025e431
72886 F20110114_AABHHB onstine_a_Page_089.QC.jpg
021f33f19cc55d541f73b9c2d609838b
5c5762dbaf36b274388487afe47bfc1fe37a8ec4
34864 F20110114_AABHGM onstine_a_Page_070thm.jpg
26d3d8ebbb41534cd470e5de3bb2f685
1fb30c93f56f6a4dee89c9495e534467adbefc36
51362 F20110114_AABHFY onstine_a_Page_058.QC.jpg
2b100e3cb9325266c07cd774a70a074a
5f27b62545c9d5f2ce831a8ecef8bd08f0734103
239758 F20110114_AABFXQ onstine_a_Page_175.jpg
cd8dca828f33a6cb7cda667df8288623
084bc0770fc4fb9a4be89f06b7030b49d8d0382c
101342 F20110114_AABGDK onstine_a_Page_158.jp2
eedbaf3cdb4762313d458e41dd9b0af5
c99e1956c90169099c3a891263dc486527386cb7
1051927 F20110114_AABFYE onstine_a_Page_011.jp2
5f1c1695bfd64b38348636e04654ba7e
b712b3423b374d21257e55e384044b2422c74bbe
210586 F20110114_AABGCW onstine_a_Page_142.jp2
cc863018a3faf48d9f247833db5e0220
02a16a21eab6b96d8a1c4358e7a04fbf46fbf459
45010 F20110114_AABHHC onstine_a_Page_092.QC.jpg
ab3d3ca3cac0464cc25ffc6b48865258
ec9fe57b5d41890d225f34f725361e2b8ca42f25
47200 F20110114_AABHGN onstine_a_Page_073.QC.jpg
cc8859df88974453d4d1519c7dacc8d0
2d292b7381519bd3fa4742179c4d93398bb768ea
37004 F20110114_AABHFZ onstine_a_Page_060thm.jpg
76813fc6f10ad55a493acebd8e941b0e
d4adf66320584e435d5b3e48e71ae15f0bcfc275
253809 F20110114_AABFXR onstine_a_Page_176.jpg
7d4e2aac745353451e34dce6d076f304
4788d9e54374e2125c12cd04df515b6cfca57d53
117001 F20110114_AABGEA onstine_a_Page_175.jp2
491bdd4913ee0f4162535b373fe2e66f
6e58c381233da12711b5ff672718cca4f0a16215
84835 F20110114_AABGDL onstine_a_Page_159.jp2
630948effc68d97acea83cfd710c1a9d
66599fc0a95052c7a0a70fcd6584eb94da1363cf
1051983 F20110114_AABFYF onstine_a_Page_012.jp2
cab601fb6a63a60685c047f3c16c9c2f
dfc45f0b4814cf33c502c9bab01491430734beac
222708 F20110114_AABGCX onstine_a_Page_144.jp2
4943eae6ebc29f40864ea7a06db1df47
fda5ed025c6f67c6090569afefa8ed925eb1dbd9
7683 F20110114_AABHHD onstine_a_Page_094thm.jpg
6812dd0daceeb1f934d1e93204135889
725007af2499d388ba73eb44cbc37c12546e207a
32807 F20110114_AABHGO onstine_a_Page_075thm.jpg
f268859aa5ed7444bd95b8262ecb3800
063c42977316a9269949deae51d77df7b2116d45
236544 F20110114_AABFXS onstine_a_Page_177.jpg
1810d9ed1ebd479fa2e07d28e04d27f1
da151041ba9f3156daa8b68849fea9728461d251
125518 F20110114_AABGEB onstine_a_Page_176.jp2
3e597e99e8b9d65231db7b5f462bbc5d
54cb94d3d56e78876c29840b1b1e4baa8a423ffc
68395 F20110114_AABGDM onstine_a_Page_160.jp2
205140bc2aaded369acf0bd631b1b744
5b1b98a08b1a7b23f904b741021f985822bf97cf
1051979 F20110114_AABFYG onstine_a_Page_013.jp2
ca919a95d3dc0cfae250fdfa3a6ac81b
355292ef8e1173d4476f70e4056aa70e8e38743b
235896 F20110114_AABGCY onstine_a_Page_145.jp2
332943e73e6044d0de1cd88f2bb0a098
e2fc51eaa16d513f2cd450a70eaced1840eccd92
34801 F20110114_AABHHE onstine_a_Page_096thm.jpg
d27874ac8b8bde664701289d6f6411c1
fb3993fa296764177353cbbb302d9433c265c839
45059 F20110114_AABHGP onstine_a_Page_075.QC.jpg
a3b4df3d48e551937ab5b366cb16d30d
c562908a547283a03d1f1b558dd664c594f29da4
212142 F20110114_AABFXT onstine_a_Page_178.jpg
ece648f55abe2386eb1045c09bad30bb
6ab3170a3e5bc523323474c75da1f0c04d50a48e
118840 F20110114_AABGEC onstine_a_Page_177.jp2
62313459583b43345bee7fbd091501b1
9a09f464728f906c6b99ed5eda86290c1d83b262
F20110114_AABGDN onstine_a_Page_161.jp2
7ed6ea7b700473459a786e91031acaa1
11219b21dbc8d350d1e8f840e57a05d31235cf20
86171 F20110114_AABFYH onstine_a_Page_014.jp2
025e2e339ec565fe8926739717935a2d
f9b1cace57f98782f493d39f4d138b746c9c7e6b
231286 F20110114_AABGCZ onstine_a_Page_146.jp2
38e14c25e34caa996651ca7a41b5c289
b0d06fed9e8e62e97b6bb2b6fc2ffd509a5b1488
52212 F20110114_AABHHF onstine_a_Page_096.QC.jpg
ab011b549f47c830a641d6d55667d115
54b2d7d9cf411b24056a5696959fe12f0a01e49c
36730 F20110114_AABHGQ onstine_a_Page_076thm.jpg
ff11a4ec13466d716f1bbfd2024979b2
ee8d757b8d6d90efce264a56c76169b34cb2d39d
31864 F20110114_AABFXU onstine_a_Page_179.jpg
a2df67060b7a63182d7001504369e5a2
200534b8a37bca6d308e856ad711d2136dca306f
107084 F20110114_AABGED onstine_a_Page_178.jp2
84202137b3204126d41209c09a132918
0027c72d440bf34d1be0b020dcc3fd4dc92d570e
186304 F20110114_AABGDO onstine_a_Page_162.jp2
cfb8517423a890f85ca71049ae265680
c4602a5f4b3d184aaa4e4efd95811f4dba387bdd
93323 F20110114_AABFYI onstine_a_Page_015.jp2
f0e518c0be6a6a442fe314334cf292e0
796b78c8f1157a70021d2d5f30e70298ae10c269
57498 F20110114_AABHHG onstine_a_Page_098.QC.jpg
2bb9213c968c57d3906918f747216bcd
a68701ba0066642ac72672fe75210ee72f931b6e
6175 F20110114_AABHGR onstine_a_Page_077thm.jpg
a6d8ccc57c7863bc888fc18345c8f231
7cfab70bbb4474337465d0c667ee13b94507c283
81316 F20110114_AABFXV onstine_a_Page_180.jpg
46d657bea8e05daed920f07f901e885d
63ecbe13fd17b6e4d4ad4377524427a1133197f9
13712 F20110114_AABGEE onstine_a_Page_179.jp2
30438b811ddff54c30b545d05263f98a
7debf38ca59e564e4ab3985ea9ad05c617ce8522
245987 F20110114_AABGDP onstine_a_Page_163.jp2
29545d1be3c299d92337046390397937
80bbb813000b3d8c454a0be6869c02107bea3880
92940 F20110114_AABFYJ onstine_a_Page_016.jp2
31db2264b4ab003429a80531a206fe7a
a9aa2e10006a23acec81df534509508781dbd00e
64174 F20110114_AABHHH onstine_a_Page_099.QC.jpg
62a36d4d155e7d7ead1fb4ac60efa2f4
f6ea46c50ea54985ca1b6ebb9f01c8854f2cdf43
13854 F20110114_AABHGS onstine_a_Page_077.QC.jpg
85b18f822fac44abc76ef93fd97076c0
981e2a1849ca87319dd16341029e36c040362e16
5657 F20110114_AABFXW onstine_a_Page_002.jp2
d59f7baa42469f61de7a263d95cb8029
38f8a3d97a6496a637915c85b717e6a511a8a243
40627 F20110114_AABGEF onstine_a_Page_180.jp2
6ece7bbc0f2cc2dd096339bfcb257d66
f9c0a3baafcc50e28bd5c765e0ad33db370bacbd
91817 F20110114_AABGDQ onstine_a_Page_165.jp2
08c3984a1fdf180701dac7ca7a92d7b6
ffd1f4156205d84ba1835881a4b162e1142a5d8d
110429 F20110114_AABFYK onstine_a_Page_017.jp2
fe0d751337902ee44c80db3eaaa3230f
a16129351aeaad85f0c18da1b6627008fb34f256
36334 F20110114_AABHHI onstine_a_Page_101thm.jpg
49549f9b564cd41744166f290ee1b47c
12d4214c8e55760ff4bee85d73ee85323da26ea5
35270 F20110114_AABHGT onstine_a_Page_079thm.jpg
2d861c0caa6b7cb31802eded309926ab
f39ef6ee9f31d7fe1d4f1b5363bf133b15ce32b5
F20110114_AABGEG onstine_a_Page_001.tif
f3b99b15370f4c019beeaa1035db71b5
9aab442272f7a4c9cfd21cabc0d6df9125d0f0cb
102514 F20110114_AABGDR onstine_a_Page_166.jp2
80086d815dc4b502795ba536b27f0e77
25bd1e702c02b11cffa1d7a70929e272948b3a64
110509 F20110114_AABFYL onstine_a_Page_018.jp2
6419fe661c6284eec945a38a81c9b325
9157798f4123fdccc8d3edebcc51d83fe03cff74
25430 F20110114_AABFXX onstine_a_Page_003.jp2
374ad9ba923f8eba31260be79be9679d
d8e6a43258dea5c2b430863b27b634542425d420
9154 F20110114_AABHHJ onstine_a_Page_104thm.jpg
6a89faea46ab603247cc377e6514dfb8
25c13f848b40ba7abe74d4e78d2801ea4b399e18
25792 F20110114_AABHGU onstine_a_Page_081.QC.jpg
38df73c735b4dfd447f7922ed20f9209
75bbf286fd2f44c4bfe2c93a95ec2f8e0caf9cbc
F20110114_AABGEH onstine_a_Page_005.tif
6298707573907591f5078757c9caeae6
6ee8f02832f16d2908cbbfd8bcce34bb898c6f57
88300 F20110114_AABFZA onstine_a_Page_034.jp2
19988cc9d6b4c0425eb50514db0fef3f
7ea4c48fcd0fe4d9ef801f6b5dfafb0b96a0a167
68211 F20110114_AABGDS onstine_a_Page_167.jp2
cf347f46c56345fb6fc08f4f73c49944
51c887e02b8b71fade6a6e259faede5150ef4cfc
11831 F20110114_AABFYM onstine_a_Page_019.jp2
5656836534430801f8d02943a86b5552
674efc37277e385d7b9ab803eb65cdeb70be2a2a
93370 F20110114_AABFXY onstine_a_Page_004.jp2
4114972a0bdfe0180a9d634dd356e953
8b029d1f57aa61fb13d7089747d794286d89e90a
39510 F20110114_AABHGV onstine_a_Page_082thm.jpg
4710521305daab5f277b70422b648262
1ca6255f2a347c6154abc6df5ce6dcab0f03180e
111337 F20110114_AABFZB onstine_a_Page_035.jp2
bbcb1619e31a2b57613d972a0eccf079
a91c3a8d0b9f3f4c98b19e304d3059f5bcfc8832
173810 F20110114_AABGDT onstine_a_Page_168.jp2
8fc09f321028fc945847a0851716448d
3344a531770081a84c4894179397331e128ded96
79667 F20110114_AABFYN onstine_a_Page_020.jp2
5b2f26b6e74d52f702951bdcc0adf4c9
05568ec63df2cb91c0acfa2121f75de91cbade64
41934 F20110114_AABFXZ onstine_a_Page_005.jp2
f1b887d7a412d133b3bf66345c99e223
70677fe7e1828f9292be319cefc2f085e8fea5be
18679 F20110114_AABHHK onstine_a_Page_105thm.jpg
db211925136591b36585e0612095d298
b83a80af585ca1beefae45f58517bc4df57aa502
34022 F20110114_AABHGW onstine_a_Page_083thm.jpg
e4fb0c7a2dff2671e5d9f96c11b37d9f
ac67c9596a1b73f7980e83b9bc73c84f6386d99b
F20110114_AABGEI onstine_a_Page_006.tif
222953f98239c6caec1552d59a12ee23
d368b60d4f8ffb873938bd74381aff3555e15ca3
104046 F20110114_AABFZC onstine_a_Page_036.jp2
98b0f0ee88627e7a72d6bc4ba6eb6949
b3c6830cd756402472e1bf93aeb656d3a17e4f3e
92113 F20110114_AABGDU onstine_a_Page_169.jp2
60d43408898d6edfa1c649f3cd50e2f3
c5ade65753d5702e24f947c7c7e8f779cdf5a43a
100112 F20110114_AABFYO onstine_a_Page_021.jp2
ea7c413cee6ebc8108543b0b874d71ee
1e8cb4a73ec8be34065bcd9e02af2b4ccca0b690
34278 F20110114_AABHIA onstine_a_Page_121thm.jpg
507d07db3a361e6febf4dc4240695dcd
2b40b8359fa0ca0786a5120624fa369c6708c640
36861 F20110114_AABHHL onstine_a_Page_105.QC.jpg
0605a80fe0308a3231d427da4c2ecc8d
cb78636926ae75f68f8d6f911df63ba71a9a9b1c
54248 F20110114_AABHGX onstine_a_Page_083.QC.jpg
10917af19e1b87cd0df3980472b332dc
d68e30a0d7e2b60c97c2f5646e0d75396466136b
F20110114_AABGEJ onstine_a_Page_007.tif
92bcf27301fb7c6a81835bb415ecd879
d6dafd3c985039feae1b6610d625334aad65b327
103778 F20110114_AABFZD onstine_a_Page_037.jp2
328a36d8fa247d3162bc6600937f46a0
22c671be3308d540420f10ff8f3b4e6e13b0da37
210820 F20110114_AABGDV onstine_a_Page_170.jp2
bc3bf11bb080f89dd992fe15d0f77721
dcdd0318ee2d0c4ffcba50b69900939f2e4ac24b
103925 F20110114_AABFYP onstine_a_Page_022.jp2
b7b8e731ae6f6ec577dec68f4ad81b17
5b386ae2645316a9d7b7622ae79d239dd145e2b3
49270 F20110114_AABHIB onstine_a_Page_121.QC.jpg
58e82049a635989aed6d2c36f0995e40
c0597c231b32a99fbbe19c9beba6e364e20c8bb3
73268 F20110114_AABHHM onstine_a_Page_106.QC.jpg
02b661707f55dbaedfc3e8ef2699040e
25891361bb4cf7116d989f105df5288e0e37212b
69000 F20110114_AABHGY onstine_a_Page_086.QC.jpg
00d62ec4e6d9e19d25b8ff956fc4b133
39f921813ad55a1315f59e810344cf3a59362753
F20110114_AABGEK onstine_a_Page_008.tif
846bbe88576e05e94ababf21100a373f
634e2841b562bbbec08a2e1a8962e6e2b8df472b
104836 F20110114_AABFZE onstine_a_Page_038.jp2
38a6267a4b00cd5fb4b8eae868fb4229
34eb3cf4f4022ea8fc536a08ec621d991a063a6e
876844 F20110114_AABGDW onstine_a_Page_171.jp2
2ded848575109b26f86b1eccb15fc988
22916d0a5272bea903713138866e9f31c5646cd0
108218 F20110114_AABFYQ onstine_a_Page_023.jp2
534dcbe1e6d5833f47c436cbed8f2512
ac6eddad056110e7f8fe47b991499a3cc24600ec
35556 F20110114_AABHIC onstine_a_Page_124thm.jpg
d0cb6cff30700cc49112b8630fbeb784
9b98d888ca7ddc7cce892ba3d74a0996c161a75e
8538 F20110114_AABHHN onstine_a_Page_107thm.jpg
9e0f7878a1beae72087217dbfee62f8d
f0cf238cca24acdee1672eed861ef28a00efe3fe
22612 F20110114_AABHGZ onstine_a_Page_087thm.jpg
5015bb083867b17716c4e3f5f251e2da
dd090075af8002612b6fc19c9f77fa0bfa67b1ec
F20110114_AABGEL onstine_a_Page_009.tif
6bd6a6ab67fe6bc997c6432f824889df
5f144ed1c96cce96fabb2beddc7a40d1c54c3211
101941 F20110114_AABFZF onstine_a_Page_039.jp2
aea61bf8b3a46431255528e4e596dd40
ec7b60946b9a6ae64665ea1bb44c27b42befc587
846059 F20110114_AABGDX onstine_a_Page_172.jp2
c6179d0da919f1208dfd745a2388b3b0
e09fe2d0f2b72f15abb34dd8fe75e54622867241
114725 F20110114_AABFYR onstine_a_Page_024.jp2
6ee0e015ad5811a79fbe4495558d5405
14e3726cfd3754531c0b81e8ac8f3adf31e1f168
F20110114_AABGFA onstine_a_Page_025.tif
37a8effa400c804637fb6648124fcfdd
af39d8341a4ca3be520f55626a4d0c25b77137bf
23439 F20110114_AABHID onstine_a_Page_125.QC.jpg
2353965a20cc1c1712117ee6d2de9460
02f640f60fe97c88e0970223e9e402a0a8fbd5e6
23783 F20110114_AABHHO onstine_a_Page_107.QC.jpg
b0d4ba8949a0e98455d27e18d8a8920a
c578b85c2e7434ac50afde450ac21cf15e0669c9
F20110114_AABGEM onstine_a_Page_010.tif
5685255907b177bd5b75324d6c49a46c
8e0dac8ed008045a7e578b0bd02a5ed66b783433
101785 F20110114_AABFZG onstine_a_Page_040.jp2
3efb2c0206ed4c01ea4961044caebd6a
daf06a117e0b3470175ddb3def734f473e9266df
328578 F20110114_AABGDY onstine_a_Page_173.jp2
7944b84a43fe476881699b6593fcc6ee
4d107c0c780fb11aadc81dc648dfe6f0f3e5fa37
105263 F20110114_AABFYS onstine_a_Page_026.jp2
70cf15fed32145ff2c9a0e49ba74fefb
2cca62c173d54f175e8a33d9d4a1ab29f74c9b28
F20110114_AABGFB onstine_a_Page_026.tif
dfb50bbe3175ba0866993bee101f1ad2
aef9d46d90ef7ccfd97fd656dbe0c43b4169da10
17951 F20110114_AABHIE onstine_a_Page_126thm.jpg
dade22db0c45e551a0e63d5de33e3fb2
acfcfb79bdb7b56683f2cfa4147c23b7822619fa
20535 F20110114_AABHHP onstine_a_Page_108.QC.jpg
f3fa4195256d5109ab6b641888681263
25d2210da6db2f33b88707ffb58ded6c63e84140
F20110114_AABGEN onstine_a_Page_012.tif
6905b3cdcb586ce7ac851923b9f31546
ac56998cc105623da37cfcb6d2ef2ab8749fd806
112541 F20110114_AABFZH onstine_a_Page_041.jp2
e229698893b8065d78783b0bf885a8b8
5f1e8ef1db7d0204250a48fedced8eb720cd9216
98624 F20110114_AABGDZ onstine_a_Page_174.jp2
436162b4ef353cc2971865c242664ce0
e07cce6ac3176daa657371b13dbbad6456f24eb7
108048 F20110114_AABFYT onstine_a_Page_027.jp2
c3b9f416d964e193962d459d2d96e89d
702715e1aef45c6d93537606ddd0df5c7bea2d36
F20110114_AABGFC onstine_a_Page_027.tif
0ea069b8330d6a8161b1fe57b974e1c9
dbbad4c3d0fd9a846e175fc4d8931edc5f46ebca
36053 F20110114_AABHIF onstine_a_Page_129thm.jpg
91ac0196bf2500b65dee7503392853b2
03939f9b437e75457ff55f78ea48fc82210ff794
15958 F20110114_AABHHQ onstine_a_Page_109thm.jpg
9f927bc2f5ff83167bed3317547a8501
a3ef4f300d833183bc9a97f693f730e5c9b8787b
F20110114_AABGEO onstine_a_Page_013.tif
582ffb8936559adf80a865bd73553dec
e167d945a5860e96d41dc6ffbd9cd5def6c57afb
105097 F20110114_AABFZI onstine_a_Page_042.jp2
23c2a2a56baef41e9f03ef6f9dec54c9
591bcc51852b4c7a297af2bbf8f110709e525f08
111362 F20110114_AABFYU onstine_a_Page_028.jp2
3f3872e73397c36bfaf8efa7c19eb4cc
31cd80841f4b4677ebfbbae100d66de9bcb9f681
F20110114_AABGFD onstine_a_Page_028.tif
0112a8db8b206c522b4cef3be107d995
8538efe16759862c2af0fdf1a72da2b850b4d30f
58088 F20110114_AABHIG onstine_a_Page_130.QC.jpg
9a0bfce8243feb381c1092891728e01e
890fd9a90be5fed09ea60500e7ffc4e13ca23792
25961 F20110114_AABHHR onstine_a_Page_109.QC.jpg
222e71e6c063fe6eb967965f8a998e9d
7d5306ddf94185317a688dd9363d7e1c65027144
F20110114_AABGEP onstine_a_Page_014.tif
6c876ac3183f5a1ffaedd1104973a6af
7b53d11a8736fd4c73a458989afa7715efd66377
113944 F20110114_AABFZJ onstine_a_Page_043.jp2
92350b7ba5949433022b6dcc42cb0676
c6dea2c37b37608fad500c9aa1054ee5adc8e5b6
102499 F20110114_AABFYV onstine_a_Page_029.jp2
6c2a03ca66d3babbd46a917365395d2f
ac214cf40ea879a8ee582b2e7324d211c143237f
F20110114_AABGFE onstine_a_Page_029.tif
7bc9a88379af39da2c5549b0a124a4b5
990f480da8f5aee6c893ecdb11d45ca75fe648b5
24302 F20110114_AABHIH onstine_a_Page_132thm.jpg
205f6dbe30f7bd4890c500b4301ad814
aa2b1261b43b1e762f938ad99eb879c13e22439e
25069 F20110114_AABHHS onstine_a_Page_111.QC.jpg
5db6dc4fbb56886990034c19207e1d3e
7b8018be1cc4482383b15582667b40403d92654d
F20110114_AABGEQ onstine_a_Page_015.tif
7584f0282fb6442852f4e9ef94e2c0b4
20b68a447a10171e4e5031655304cd5dacee2107
107252 F20110114_AABFZK onstine_a_Page_045.jp2
f6bb5f21c30cbf5145a97bfece6aa9e8
d22bf2a65032c37439c4288f76a2072cd2f6c61e
31449 F20110114_AABFYW onstine_a_Page_030.jp2
fa150941720daf1cea62c0860ab97636
d27978ece4c8b5e0b03f84fbe837053bbafe9e39
F20110114_AABGFF onstine_a_Page_030.tif
7a54b27e46f5657bf8370b56722354f5
b00e84c6c31afd1761c5e098e85d9c0b47b841e4
76248 F20110114_AABHII onstine_a_Page_132.QC.jpg
798e02dba9be182ab69c8ffe7acd3dc1
e4a32a2d4e24c110bae03bcdae3d7e7eef5d6efe
9516 F20110114_AABHHT onstine_a_Page_112thm.jpg
ab9944c93e87feeb8f969ea5081827ad
b55ecc1bcc60f370b2bc0c17f0edd82df275d603
F20110114_AABGER onstine_a_Page_016.tif
bbbb853b0d4e43d166c9af15eabb95c8
6a1587f4f4d134838c9c012fcfa11e14fd841f21
108599 F20110114_AABFZL onstine_a_Page_046.jp2
1d1cd193f0ba2be022c0bbd7a88bf18e
fba3d01b0dbf90ab3c22bb375144380c45a5c9f2
32308 F20110114_AABFYX onstine_a_Page_031.jp2
a04bea318cb1a16c10e3c77c94f0aeca
b0048e35880e91e670a7c77ca6e7c1a9b0e47570
F20110114_AABGFG onstine_a_Page_031.tif
847b095368e9560a61f09249d2396a83
c12cf3f5fb5adfc93f38df23ce2f6cc87146d7aa
45215 F20110114_AABHIJ onstine_a_Page_134.QC.jpg
9f08d2e86bd3f7a18235d7ddcb104ee6
40017d3402333302585d860b976326f72376ca74
27048 F20110114_AABHHU onstine_a_Page_112.QC.jpg
5cef906c6a688ab6456d16311789bf72
67dccd9b72f3c6dac3a9745a03f15a2f289fdeca
F20110114_AABGES onstine_a_Page_017.tif
626cadc4c700b5a047b9af78feecb4cc
6a0ec3f39a47718e410756ccf14d5b9254e0afe9
17288 F20110114_AABFZM onstine_a_Page_047.jp2
46071c94e43597df2fe80644b282a363
4aa7f2248456afd7155cfeee658fb7810422f096
223097 F20110114_AABFYY onstine_a_Page_032.jp2
4052f12232b3b11f561e42be03287fee
97b35d2f9cb6d56e53e1261814daa8c9d34e33ea
F20110114_AABGFH onstine_a_Page_032.tif
7a6db971628e97f7ab91a63a06272f46
8748574285702e662b8c5b2e30bb09ce4651548d
35101 F20110114_AABHIK onstine_a_Page_138.QC.jpg
6fd9ff580b7bbc36109a34630cfbf814
86265aaf8397c0440c016809eabc53b9af01cdcc
30632 F20110114_AABHHV onstine_a_Page_114.QC.jpg
3e9eb3943e249f2ca01cc37369d01340
6850576e5d7a715fd2e30e51f43c11594319174c
F20110114_AABGET onstine_a_Page_018.tif
44fe2830ab7bc15ad286d8a00e1315c5
7a533167ab500ff27cda462738b73e8495b88cc1
266795 F20110114_AABFZN onstine_a_Page_048.jp2
ee57ecb4d9a7d03cc565cb64ec508875
dde94667258a2f0ba6227b13f308b7b34a597992
340494 F20110114_AABFYZ onstine_a_Page_033.jp2
d4bd4fb7471f60bd6c9c5f7333b7654e
0336441145c2f9950d197d9c75187315090f9eb6
F20110114_AABGFI onstine_a_Page_033.tif
3451b530b4dda4bc251c953792770d01
7d74d06094118145caf6b50c0dfd0f3db30ba1e2
37878 F20110114_AABHHW onstine_a_Page_115thm.jpg
b5ea8d9ea41b12e1df1d55ddd6b56cd2
b6b6500976d2686d024f6e7d673815e728267323
218760 F20110114_AABFZO onstine_a_Page_049.jp2
ffbb56a5f5dbb1d99b2e8396de82d8b6
ae766dad6ccd234450aa000006615f7bcd17f1f2
F20110114_AABGEU onstine_a_Page_019.tif
6cab189cef648722da2a5594d30557d7
9488b1557224db0637c1e3429c1104e337f1ffac
23626 F20110114_AABHJA onstine_a_Page_158thm.jpg
3b7fa22a4c36ab0df303b9e07aa69095
6838fa49b54a36d4f0e4d47fd65c3937148c46ae
8467 F20110114_AABHIL onstine_a_Page_139thm.jpg
60d27e35b5fff68930d79b05189ac282
c75ea7024dd44ce2a4f364cbb09bd171f790bfb3
24167 F20110114_AABHHX onstine_a_Page_118thm.jpg
06078fba749d535569cc0fc46948efa9
4d88f643a5aad598043daabcb2c635772b944c90
356107 F20110114_AABFZP onstine_a_Page_051.jp2
ae41949ed6a764507b1f838edafa6fa7
18c2f4fec1e52179f85add3f3e9ceedc75fb5df5
F20110114_AABGFJ onstine_a_Page_034.tif
4be19c2d0a27e3972b4ff67377a25751
63459b732e0702e1957533f3d350c8d0b9d7fb5c
F20110114_AABGEV onstine_a_Page_020.tif
7a97293183d597b43bd6feaf6198e1db
632ccd3b06809792546953a7ebba6ae3bd17a5c4
77124 F20110114_AABHJB onstine_a_Page_158.QC.jpg
0e1c76e84ae4df01abcf5b725f2fe8c2
692f7fd2808190bbee0f79265355a6fcab65618d
8420 F20110114_AABHIM onstine_a_Page_140thm.jpg
4decdab7980f3d6fa70bbd40fc726edc
573cb1a166dddc627d5da271c33fe7e62818da1c
24875 F20110114_AABHHY onstine_a_Page_119thm.jpg
9020f67929dc46efb62acb0fc360d548
b99b3c47e27f7eb95f50405ed320d5807c707c2f
760149 F20110114_AABFZQ onstine_a_Page_052.jp2
7d71be9bdf447fafab4f1bd9ba8d36e0
7c37a07cce8549ebdd40635d9444acc69c5d0dbb
F20110114_AABGFK onstine_a_Page_035.tif
398ce485fc62b2f0034d4c89c6cbceed
1a4f172342606c0a1989d4955a72d1b6934be5bf
F20110114_AABGEW onstine_a_Page_021.tif
e04563c267b8693fffffb21ab796a98d
02e94d859a6035f6de5d6c38914d0288e1dcef29
20178 F20110114_AABHJC onstine_a_Page_159thm.jpg
7760a81e300f8ba59370fed3aa43da0d
3b652a8b1ba489fd0d475ef812ba57c8215a531f
26392 F20110114_AABHIN onstine_a_Page_142.QC.jpg
664509976d3df494309855531cf78490
695c388f767305d464fa3c7da1e4189d1bbb308b
79129 F20110114_AABHHZ onstine_a_Page_119.QC.jpg
c41cbe26b806acdb3300a875e0ac31d3
e81ce7e21635b2e1d100ae8b8c926d6a10f23e3f
780731 F20110114_AABFZR onstine_a_Page_053.jp2
a4e0630f6c59c8f84b6efda485526051
7e096a03e49eeb894ab5879f5c02418bf3bd719f
8425398 F20110114_AABGGA onstine_a_Page_051.tif
f783842cc29c1eb0d8e5f73fe2ad1dd3
884b54f7ad0d124d3d9c050ef906c1efc9020a9b
F20110114_AABGFL onstine_a_Page_036.tif
7e019b14575310917766264e9be14573
158b1e88bd358b17d463d2fa1b32a97fcc860069
F20110114_AABGEX onstine_a_Page_022.tif
6fa3cbf42f8f27d10517327e2250fda6
fbe73b664e3add6f8475215ecac5f681b96df334
34180 F20110114_AABHJD onstine_a_Page_163thm.jpg
34a1cf2f464ff91d046459a696f5f76d
cf124006b5b626cfaed81462e1c3ca4522b8a102
15482 F20110114_AABHIO onstine_a_Page_143thm.jpg
bfa108b4855be4285e6027582ef37125
58fa0c31a39e400f73f47a23c3740294e6fee489
474829 F20110114_AABFZS onstine_a_Page_054.jp2
7100866b5c48be8cdba70a9555478cd9
89030ae97ee217b41715d5a5ddf3828dde906663
F20110114_AABGGB onstine_a_Page_052.tif
ad9ef90108962cc854de9dc74b9fdcd1
3fa8c4acd735398714478f94476c56881e66c4ee
F20110114_AABGFM onstine_a_Page_037.tif
6407ca0a3d1b3420cb5751c662303d70
d394446c157224f236d8bbf719ff7af9d1877147
F20110114_AABGEY onstine_a_Page_023.tif
e9acf4d88f8cb3af178f98dff3548857
3a412cc45a61995b81287c69a3fc38ee18c35e42
68966 F20110114_AABHJE onstine_a_Page_165.QC.jpg
ac274f9c9e4a4ff467f73cfbf75531be
7dd43bb8dcf18caf4e0b52a92cf47f6d426f02d4
15449 F20110114_AABHIP onstine_a_Page_144thm.jpg
d0605c783f88d89ca28075e4e06a6799
e24680910b1259bb433b3d55022b0e5e3a62b9dd
541290 F20110114_AABFZT onstine_a_Page_055.jp2
8a00e50135faf1eee5fc2f90e17c5a00
8371f9a3d14081b4a67676da2632a7e77dbb03aa
F20110114_AABGGC onstine_a_Page_053.tif
bba88d291b5c5607205398467980e5bd
adfa0b77ec3c90b5fd09b922f4143e365a125409
F20110114_AABGFN onstine_a_Page_038.tif
6d5695c2c960deaa44bdba2c0b9e3095
dcb8e8f9e61167114c53b583d913986162f2b82e
F20110114_AABGEZ onstine_a_Page_024.tif
ec4dc1eab86beab28631a2237d2dcc90
e0609e4c114b972c1e42873c8aa9c6a9c7939947
16826 F20110114_AABHJF onstine_a_Page_167thm.jpg
34a961b44a77c287caefc80e520780fc
1393a4e5cca4836fcc5170ae8037a2c8c05b794b
15603 F20110114_AABHIQ onstine_a_Page_145thm.jpg
4ca03c3b0059a99e31c8c5606a214af1
02c40328f0fdfcf816e4024d73bcb88f5d412cbf
325739 F20110114_AABFZU onstine_a_Page_056.jp2
1333b5b125e8dd7f8f50f6966ac53e9a
19b6e24b4967d0c65af65f6dc7bd3f1b0e755d28
F20110114_AABGGD onstine_a_Page_054.tif
6de3a91fc6f525491335e105baf4066f
1319477fc23d174b65f4d51e702d6866b962b98e
F20110114_AABGFO onstine_a_Page_039.tif
f1975938edcab3f26b3bf8a8b5f87502
18bfe03715400b68f1b4a79a3f50b76d1d5f294a
49680 F20110114_AABHJG onstine_a_Page_167.QC.jpg
4e83e008d201f6a531387601d33a716e
f618413043ddf7ac03992518a6bbd2ff35824b88
29102 F20110114_AABHIR onstine_a_Page_145.QC.jpg
63448e67db0a713f3b91e5ba50aa8093
90a446fd9aa7d4122ec867032a92bcd6ec7d9db6
205129 F20110114_AABFZV onstine_a_Page_057.jp2
e06d1be55ae5a368a09b738a52648479
c5b5b60a283fb4ded51237be99b3fe48368acd52
F20110114_AABGGE onstine_a_Page_055.tif
f2645b709e29cdd9a582728c5ff32432
55c29b35c3935f64c88f1fd86fc321cbcc2ebd4d
F20110114_AABGFP onstine_a_Page_040.tif
b4a10a0f7da041ce91ea25253126ff07
0b7cf3dbfa6bf2c4224e68996b15a7859b395b44
14680 F20110114_AABHJH onstine_a_Page_168thm.jpg
0f8c17c8e16b1463a8ff044ed25fe701
845f279767277c6942eb757ff8f161471ef32a5d
15395 F20110114_AABHIS onstine_a_Page_146thm.jpg
a424841fa3b89dd5bd4bf5196a221dab
fcb24e9764cad1948749fc36ed0aef88e2669241
364530 F20110114_AABFZW onstine_a_Page_058.jp2
6ac25c850ce1a594083c26c7310108ad
c1088c774c30d8a8c8f964386c4926e907390954
F20110114_AABGGF onstine_a_Page_056.tif
9ed8b16715cd2762b6fa32ea11492b37
7730b79596c342aea7f01aba9323e47d40c21f3e
F20110114_AABGFQ onstine_a_Page_041.tif
b1822a94d8212e16eae55e8b0a69ccf3
a7acb9ac32edb49a29964de8487463a5b2732b37
12590 F20110114_AABHJI onstine_a_Page_169thm.jpg
1a11d70e2a6c5090037aa9ecee88bde5
a4013785e56bbb402fe3153bcff1e2782924b399
27623 F20110114_AABHIT onstine_a_Page_146.QC.jpg
1d5e064499f2545ece6137751816f97c
c6249f52d1680ca20894ad83ee6130d3c1c70ac0
309805 F20110114_AABFZX onstine_a_Page_059.jp2
7899eae8c6ec7152ad24cba9f7a9a040
aadb3ee3b845468a9b6215a25eb2af66f71f2580
F20110114_AABGGG onstine_a_Page_057.tif
f886865055d3a2979b45f7b8959d91f0
3f1af38f30d3bd8131bf134d2cb9cee6496f807d
F20110114_AABGFR onstine_a_Page_042.tif
fefce52c9323c30d6c4ec172d367242b
770c117a0ee5ec8bbca8d1432111195b5d575433
19380 F20110114_AABHJJ onstine_a_Page_169.QC.jpg
da2dd035fd7f7a196e253fddfab716fe
0fd8c92f69d4fdd3fa575adb1b5173c3da4a0001
75149 F20110114_AABHIU onstine_a_Page_148.QC.jpg
f18c8d8b6aafee508c0a67b35fbb89d2
65291a42929f5f655ad5a9e02e77d5f9f6d4ad06
302468 F20110114_AABFZY onstine_a_Page_060.jp2
c5d9f20079c588b1ef58fa2771c21a0b
4bf4db33ef10bb054b7f85c9d0c63df0b84bac91
F20110114_AABGGH onstine_a_Page_058.tif
1d0ebbc6e04efe104ef302908ee2c132
ca04a9e871b52ebc9fece3697f9f404f9b8a0a91
F20110114_AABGFS onstine_a_Page_043.tif
d6d9cf2c0d04a57a4de822358c241420
942d7969fd1c550a7ed85f371a6f9508a0d1b37b
27501 F20110114_AABHJK onstine_a_Page_170.QC.jpg
7df47d330da99ffc1998e807dbbdbaad
69a7ae8ad01e2626dc67c558970383e59530f449
71486 F20110114_AABHIV onstine_a_Page_149.QC.jpg
322e79f771b280261c4b6a03b3f34dd5
ca560ac659101c68cb55d3e196dfb169d839ee58
271014 F20110114_AABFZZ onstine_a_Page_062.jp2
b3fcc0fa8bc0db71df0dee84929a9517
7d7e76c2ce0efaded06b2b215fa1339ff1c09e2e
F20110114_AABGGI onstine_a_Page_059.tif
5515bc4c6ca96b43ad2d3645d0058c0a
de8780651dd5d02bbb26b58e2b5756f9508a09d5
F20110114_AABGFT onstine_a_Page_044.tif
515ca905d63b43609bbfb64b888ab5a3
0be63bd074a1aeb2aaf080ed1e6eb81561f258c6
35151 F20110114_AABHJL onstine_a_Page_171thm.jpg
7f866f952e40cd9b46fabd68f1763f1b
e283597a0b82c9b1b410af68ea6507dbaa8065b1
21199 F20110114_AABHIW onstine_a_Page_151thm.jpg
99f6d00bc091d7c36bea10b45bfaef63
e98499cf5bccffda9c3db859d9f829894c0dae6a
F20110114_AABGGJ onstine_a_Page_060.tif
799dfdddd4aab9e9978b401763994d3c
590f932df80de77fd71956c3dbd61acaa2ac78cf
F20110114_AABGFU onstine_a_Page_045.tif
535b0f7602a34ef5db9b7e64abb5d1fd
e1d2264868bd4a0edb389a175b177965e151b465
63003 F20110114_AABHIX onstine_a_Page_151.QC.jpg
b35ce0c68c0c46d529e495a1f28547d9
1e9de5215ace7c7a833cebebb10c079242444503
F20110114_AABGFV onstine_a_Page_046.tif
3e632fcfb456bc91f034894b7c8fb9f1
ff17ffba81a10a1f21c4f7159bc51916555d8146
56210 F20110114_AABHJM onstine_a_Page_171.QC.jpg
2329178a8b4c3db2b73209b6f55f4163
4673b05ef6eb7742edd2e4c0f2b3c516e67196f0
65315 F20110114_AABHIY onstine_a_Page_155.QC.jpg
b4069005a8f5a754ecbf4d6436cb1099
717cc1fcddcba58fa4d7da2fa04800d38d73982b
F20110114_AABGGK onstine_a_Page_061.tif
8882853173270a3cbe3a568445ec6fa3
be3a0294af9df00318bd47500e3f49169093aae2
F20110114_AABGFW onstine_a_Page_047.tif
7fa3177bd9b5d3a3f14424c6ce205f7a
3dd9c66d93b110cc4ae3fe3ecf9853e6f7504fe0
51162 F20110114_AABHJN onstine_a_Page_172.QC.jpg
4be5eb0f3efff0f484ed9e70e8ecd969
4e3dba5cfa9cf07cc5af1e3468df4baa41a3cd71
62587 F20110114_AABHIZ onstine_a_Page_157.QC.jpg
a254350db4775c22dc19bfcd32355c15
19dc103a07e226a51ab11a0b02b22fd4086262bf
F20110114_AABGGL onstine_a_Page_062.tif
4acdfaa1384b33dab0809d9f9711c476
31adfb4e84a4a9d67255612c2347c605fa13ebb2
F20110114_AABGFX onstine_a_Page_048.tif
8de828333c9e3a5397f1f8a88d2ab5d8
933947c5cdc5320896c81d16189d07047daf4a2c
F20110114_AABGHA onstine_a_Page_079.tif
3682b1a906bf8f3e511b16327fb36ad6
f09f52c07c7e2d47c2fe338b635be62010f1c92c
70703 F20110114_AABHJO onstine_a_Page_174.QC.jpg
c49fbe46cd5f6ab4f3eb16bdf7c8e2b4
27da3ae5677beb2a7e0c969f27a235c051ff5650
F20110114_AABGGM onstine_a_Page_063.tif
18cefa9dc0db75618e9d55be1461a16c
65389ba151824f66e7c36088045f991c8aa03751
F20110114_AABGFY onstine_a_Page_049.tif
10cc95b17318493af36625fa89699b22
4b27f4ba15deaabaf182eced80a32c497460c41b
F20110114_AABGHB onstine_a_Page_080.tif
c22b33a7c19caa9c1f2029df773dbe38
44dce6fabcf77557e95db8b5684e337bf936edeb
80379 F20110114_AABHJP onstine_a_Page_175.QC.jpg
9ad8da5f995175bb33c0a2a71bee5bb6
cea2870ba335fa8144fa10398c2222db0f5f5029
F20110114_AABGGN onstine_a_Page_064.tif
cdf290b3f14a58e199030499429d68d6
5ce8260a93af97cb7c8b9e6ca712088312cb8fc4
F20110114_AABGFZ onstine_a_Page_050.tif
fbbf5851797742ef82fbe10bc719f789
46d6643ee0ac5f57bfd75a5e56b8e094eb323d16
F20110114_AABGHC onstine_a_Page_082.tif
7989d9932f43db4172f43aa13d68df88
ff6589510cb89180af509fd3b3564279a76b8c72
24148 F20110114_AABHJQ onstine_a_Page_176thm.jpg
a453d590a752257fa02c6293f94104a8
09aaa1549105fe0d3ac5b3d9ef2c829b73862c1c
F20110114_AABGGO onstine_a_Page_065.tif
68a9b9ef222651da56b45846076622ba
a3b51ac3814d6ce0207b176f59486e39c12c7bde
F20110114_AABGHD onstine_a_Page_083.tif
a978c85fbb4203b1825fa9dcd1045649
802ad2a3d979fcc139b2bef7307b14998fa6b623
24703 F20110114_AABHJR onstine_a_Page_177thm.jpg
e4265e94e3c961729641f7128e880a5e
2c66a007c9b3f021f5edc2adc14ef38fc532e4a6
F20110114_AABGGP onstine_a_Page_066.tif
9db5f2b1bdd0c69263db148c9866deb3
8ad0c15240b1376bd750ef373dab0357d9019f1f
F20110114_AABGHE onstine_a_Page_084.tif
4b6db5947fcf112155e440bd7cca702d
fb1bd6795e68e4ab1e104910ec1efc29eabb6510
23034 F20110114_AABHJS onstine_a_Page_178thm.jpg
29923c8f6629bb98e045e340bb2a0e0e
3a5dd50f7ca1270f13c5525115ef39285ec3bdc3
F20110114_AABGGQ onstine_a_Page_067.tif
b6f410d59057ffa6bf60ed9f024116ed
a037fe1ec5ec4a8edc472e3dfcda33c5ccd6e4ee
F20110114_AABGHF onstine_a_Page_085.tif
54cf7868450a3bf48917d3d735f5e2cf
5ed6d4ed20c34c8dc07b4aa152b23ab3d69d168c
4177 F20110114_AABHJT onstine_a_Page_179thm.jpg
fd22ce6ca1a3d4fb9c86168376573229
262f57b1210f9f94cae5a81dfe8069876d418976
F20110114_AABGGR onstine_a_Page_069.tif
ecd6784ab15598c81cf6cccb37fcf07f
8e8d543ce3b4fb89ec5d20cf318200d5d8b806d3
F20110114_AABGHG onstine_a_Page_086.tif
e63b391855f8f98fa568ee8e8c69b027
956426bce26c6b6c5a6fb2a02abcb381b92c022c
30369 F20110114_AABHJU onstine_a_Page_180.QC.jpg
d53dfc69adb8e3b3c8d932605b1ef348
38468bafff5cd4cb237583bbccd9572a59021999
F20110114_AABGGS onstine_a_Page_070.tif
18cd729db1b4997f30f2b1ba02f896ce
462a1267cd07bb3a20e28adc8bec4b21fb812772
F20110114_AABGHH onstine_a_Page_087.tif
9831ca18296ada69dc732d7f8fa4c1e6
d989230b7c2545860601761f547eb51749585f65
F20110114_AABGGT onstine_a_Page_071.tif
4391963da5022273dfe1f87bb4cf8629
ed93b690ba8e642a9b537841e9a61e0c22e8ea99
F20110114_AABGHI onstine_a_Page_088.tif
304246758f3acc0e7e6f967cb5a24202
78f347ce7cf61790b8f1c5cad4ed226cd96a7e55
F20110114_AABGGU onstine_a_Page_072.tif
163c87910872744bfbd467b1f0da9392
f8f8cd1f45ebc210b3f89124e38834132f526cf0
F20110114_AABGHJ onstine_a_Page_089.tif
f8e62cf586dedee0b0a108bd2aeec312
93b072ac6a5e42d6a63d0a9cc9d77ff72a011637
F20110114_AABGGV onstine_a_Page_073.tif
e42bed7186692863a81c1016c0719750
55530fe9deaaa66310d379a6a4a5a5083c8d2f1c
F20110114_AABGHK onstine_a_Page_090.tif
80437b2342f82de68d56d76f11d08b2c
21b78ac298b53e98ade3094dc5cc593c44c14704
F20110114_AABGGW onstine_a_Page_074.tif
45ebda36a8b0a241fcdf1fbe8388ef8c
a19d052e20bf99c4c3dbdb69cb3db0229b7dafa7
F20110114_AABGGX onstine_a_Page_076.tif
0a6e76acbdecd84383e46009bdb98d4a
f448007f2f6bc97134587ac97feb66882e9d318e
F20110114_AABGIA onstine_a_Page_107.tif
ac7d9746527954e65dfb8c07010021a4
14cfd42cdeb334f31099e67b76daa622ed5d88e4
F20110114_AABGHL onstine_a_Page_091.tif
5d7a82d5a58b4752c8cf1da18efc243e
6ebfdfd5d2dce0c7f3d53c9f8eef87ed499d41ab
F20110114_AABGGY onstine_a_Page_077.tif
02a58fce7eb6ea528cf4106b6ab07247
83cbd548c1438bb98325da4a7486dee9343bf47c
F20110114_AABGIB onstine_a_Page_108.tif
8f48f4768ea53e8a7fdf6680b4665730
d41716b67f4260dd7f227d61919e204b5e23da04
F20110114_AABGHM onstine_a_Page_092.tif
8f60723cc79a7e72ff82f70455ba432f
3c3c0e6f11e6929c15611f715727fa2fd487b586
F20110114_AABGGZ onstine_a_Page_078.tif
7e7cd733fb866493e547a652b47acd75
f8b83360d35a06d73777dbdddf86f9675b897ad3
F20110114_AABGIC onstine_a_Page_109.tif
13d5b52e678178bf851be19b71b0306c
b9a325644e6fbeb5f0f9ff3769eedf69b1b833fa
F20110114_AABGHN onstine_a_Page_093.tif
e17e34ec00ba90950adeb64ed26486c1
46ea3fc9a4be6279fa113d2c6a5daf60ab450cae
F20110114_AABGID onstine_a_Page_110.tif
5c152af3094b3d6e946f356becb63cdc
c5a7c41d31d3bfbbaa56344e05de8216787bea2e
F20110114_AABGHO onstine_a_Page_094.tif
93c2ccc3f70162dbb745b7214e866ec3
ca7aa70c530c52ad48661302816d8f29275704ea
F20110114_AABGIE onstine_a_Page_111.tif
b15df52c2d5086fe1d31f6a061ce5ab1
0734a63af84228e22d8ba36586cd7b1f3da30b66
F20110114_AABGHP onstine_a_Page_095.tif
aec85259a60159b89df0f6462cec4d0c
a076f552942b6b48fb63cc49c527ba379ffdd1a7
F20110114_AABGIF onstine_a_Page_112.tif
cc5d748bff76e467ef47b6260b1dbcc4
87e0c4f8bcb6f69f2e4cb647d04be647af8df6e3
F20110114_AABGHQ onstine_a_Page_096.tif
322dc81831fea2aafc28da5938175cc1
09a27c1fa91d5885ac81c6ea320896271bfd340a
F20110114_AABGIG onstine_a_Page_113.tif
b617626bb646e847bddd176957bd0a83
c5e6040890dc558e82263a7ae4f0f0c161086593
F20110114_AABGHR onstine_a_Page_097.tif
090a8867754ad6ebdbfbba7d8e3410c2
40e04554580854513486846783558cb85456ac2b
F20110114_AABGIH onstine_a_Page_114.tif
d29bc44789cfdd1891dd8dd92ad12884
e1f42b6bbf98a9412f23ab7e5ce0d9531419735e
F20110114_AABGHS onstine_a_Page_098.tif
cefa6d5181411590714dbb3294ca2aa3
61732c5db6faba91d59241134b1b26f50d789957
F20110114_AABGII onstine_a_Page_115.tif
4b427d29b0bfeba32c7b40cedc050a39
fe34e50ccd72067b6d7f8b6ed7f39d451bd4b0b8
F20110114_AABGHT onstine_a_Page_099.tif
ab4ae34e78932e21587d203e4bbaf4b3
6e037daf0b4c7f8e78d000be7191bab0c1fac66a
F20110114_AABGIJ onstine_a_Page_116.tif
8cfa4a373707835415f424afa9a66ddd
deef0ec1aae09e46439d1f68deff9341534fb3a7
F20110114_AABGHU onstine_a_Page_100.tif
665593b11d00c3076db0b3bfc720d01a
7bb04c66e47b3ed8fc95a75dc8920a27bd854351
F20110114_AABGIK onstine_a_Page_117.tif
eec2da3e7bebfbb30f0692000077de5e
c7c661456f2d139b6473da95bcb101342bd884ef
F20110114_AABGHV onstine_a_Page_101.tif
26dc83386712f780f7a0b496aaa6e319
04bd6d1064282b691cfb79527f65cceb835a3473
F20110114_AABGIL onstine_a_Page_118.tif
ff05bdee346dc9e6368e8234b9cc4a71
663464a6fca82682f172a1643f656547445fe388
F20110114_AABGHW onstine_a_Page_103.tif
34074128f6519722f42430647bebe753
606c68ad34a779b55bdb35abd76fac133c2db326
F20110114_AABGJA onstine_a_Page_133.tif
b0efce4bf2968c8d133bc4ec5e8ff313
daa72986d351fcf3f52989a24cc660fc08c3c935
F20110114_AABGHX onstine_a_Page_104.tif
703745c8ac6960d19dbc252a82bbb4d4
f9461d66f37ca5b9258ea865bd2fdb3e4d85af23
F20110114_AABGJB onstine_a_Page_134.tif
213634c91531e1896747e4bfe472b857
08c9a1430ed5c13d4a866fa5306681c287688786
F20110114_AABGIM onstine_a_Page_119.tif
eeb971e0b52ebf2f9c5f6a3821359121
56eb8f4b4ca365c44bd2ddda02b5d402e16648b2
F20110114_AABGHY onstine_a_Page_105.tif
6cbecd00ad16dfd257605ef56dcc61af
3cef2b02a14a5302637f2fa2cdad5df1fd5863e1
F20110114_AABGJC onstine_a_Page_135.tif
6e50cdda3622b296a207d973a11d596c
40df94869accd01d1935d70f7ca717c070df67c8
F20110114_AABGIN onstine_a_Page_120.tif
f217372eaf1b49cdfc46773768ce5c24
bd18e6e59408edea8fc3069f45f1d441adb2e056
F20110114_AABGHZ onstine_a_Page_106.tif
ab2ee8b508225f9632dbc0e87e2f12ef
aeafd1a006daf4600f5aba4250f71e10044ba6e8
F20110114_AABGJD onstine_a_Page_136.tif
212a803eff1b0caf343780a99bb5bf7c
0683432d30186fbadd2ca7eafe2e2482916e0397
F20110114_AABGIO onstine_a_Page_121.tif
0fb38bdee150d1241805f68b2d5cef2d
b77231a98c15ddcf182b20010eae9679d4b283d9
F20110114_AABGJE onstine_a_Page_137.tif
a7a0fcc88fb35ed63b2468ccb28cff9e
98bf3a2291695f1b3c9b838f323ce4b0f366a4cd
F20110114_AABGIP onstine_a_Page_122.tif
87dae2aa30ecc6712885a2d46624ac1d
c1f58310717e9d38b8cb1a55f3154fa0b2840984
F20110114_AABGJF onstine_a_Page_138.tif
c72350187397f0d33ff1829d14106384
e201dfb7d7517e115ca663dac42ae910393bfc6c
F20110114_AABGIQ onstine_a_Page_123.tif
84c1075232e6ff51f8d7e9ff27fb2585
ca23ebed34930fce12864e5f397f2e2eb3f25f7e
F20110114_AABGJG onstine_a_Page_139.tif
c19d410e8db7b2c0c16c67eb69d3f432
0d2aa6afddd9b64543a7aefd68bd21f96b4fe507
F20110114_AABGIR onstine_a_Page_124.tif
617717f8ea87a1c40fba67aaae45bd6f
c00438ca4e9ccff78778953595e9f345276c397c
F20110114_AABGJH onstine_a_Page_140.tif
ed0c7ec48a6000706c6f76629ddcc829
5e31252cafde1499dfff7ee92f2dc5452330eede
F20110114_AABGIS onstine_a_Page_125.tif
4c6adcd14b6a6dc4a7cc4c0d4bbf378c
a82b879f7f0e3c32469c984f8f29e0f73b741533
F20110114_AABGJI onstine_a_Page_141.tif
605a8023042bb3e02fcefa8dd09a4dc1
487211772e431945fa4af230a59a9a8f3f2d9d42
F20110114_AABGIT onstine_a_Page_126.tif
4fa55c4b2d23a0aedf9158a78dcc2e03
4799595210772bbbf0f40d476e2df86c87f6aef7
F20110114_AABGJJ onstine_a_Page_142.tif
0a229e8a06b600fedc55d0e5f5955da8
ebae0ca32be5473006d24c7c6975e7ec02e88013
F20110114_AABGIU onstine_a_Page_127.tif
acdc8324ab4fe8b335fd07e82a89594b
4cc20edd1ea9267db8937a27d88618b61a3b8ea1
F20110114_AABGJK onstine_a_Page_143.tif
415c64e30ad7cb395a096072247e2760
df22f39103f045ee0a5beeea931624441a3938df
F20110114_AABGIV onstine_a_Page_128.tif
7c00568593b17c1d7557b992cb2784cd
25205981e105fac5d3be0cb33fd351c2c3d8f021
F20110114_AABGJL onstine_a_Page_144.tif
c38cc9849af1bbdd4afc6feaf1bbd5f9
054713c7ffd11c76264d8a242198f85f86775812
F20110114_AABGIW onstine_a_Page_129.tif
d4befaa026f2297a28ed79fcc88a1952
9b3ad3ffbddf9221ba3ad1f851a1399c3337da34
F20110114_AABGJM onstine_a_Page_146.tif
87dee5cd6a1088137f20b508366d6876
d62c8fbc8c3d3ba3b8482cb12eb5333cacfb8ef0
F20110114_AABGIX onstine_a_Page_130.tif
f08c5dde5ea65b435f0aa4e7a90f9ac6
cc06411e1514cc1d6e6a023c9b27edfb36b579c8
F20110114_AABGKA onstine_a_Page_160.tif
e122324f76eb1569f1767371c34032c3
a97ed640b44e810b0dcea3fc33ed700a9abb9e8a
F20110114_AABGIY onstine_a_Page_131.tif
830d0c46117d556b9bbb8af5728588a6
07284f7697f4064c48a1455eea42595449ead9c2
F20110114_AABGKB onstine_a_Page_162.tif
2f7290aa096c83b07a6153254942b99a
49d0ec2b8cb0a3b81c01197de203d8392778d71f
F20110114_AABGJN onstine_a_Page_147.tif
58ffc1beee35470c37dbad58f08558e7
82d04621773f5d34f35082125c71d7def9f3634d
F20110114_AABGIZ onstine_a_Page_132.tif
34bbe1f333e7b54a49af5bc2557b5863
34452b59cdb318d3e6cad2ce6a94d45002feaa8d
F20110114_AABGKC onstine_a_Page_163.tif
754deff1ebe881771d11ee144982bae0
a38cfa788a3676f81300aa6300a488e72bf9a533
F20110114_AABGJO onstine_a_Page_148.tif
33ac2684af84e390febe410b3d28f6d4
2dbeb9d41ea90c186cf6796fce9d138ef922ce8d
F20110114_AABGKD onstine_a_Page_165.tif
61da9a5f6b6c27166e706a09fac116f2
116d60151fdfd27b06b0fa32c46dcb4572bce816
F20110114_AABGJP onstine_a_Page_149.tif
baf4ce3572477dae8d27da2cd00cad86
602ca67b8da4dac0c72129ff20abb52a611592ba
F20110114_AABGKE onstine_a_Page_166.tif
8779e0492f9ffa11b3fe2eb96c1d1341
e1433fb1cf5d51d29605d329215ff0b71a2b0238
F20110114_AABGJQ onstine_a_Page_150.tif
b77b7476d4e34b4af482a7cf76f1b667
31181864d16096f83744d5d5c764241c0b3f2b0f
F20110114_AABGKF onstine_a_Page_167.tif
c4140bcf72c30581185b4da5a0dcb1b8
1657b9bd239ce166206c3d3fd04eb377f78ba3d1
F20110114_AABGKG onstine_a_Page_168.tif
da3e266a2864818ac2b657e72d9d39a3
fc6fd0a577d65053a8d39126d1c697384d8c0f78
F20110114_AABGJR onstine_a_Page_151.tif
29f5e4e585aaad984e44f9e473d066f7
53a8a97031c1badfa8a04f4e713f9d78da22bf30
F20110114_AABGKH onstine_a_Page_169.tif
05a9fa22157a6ab712670b453fbaa065
a5cfd4c15fe5bd042dd8a997dcac5232d8b7105c
F20110114_AABGJS onstine_a_Page_152.tif
7c10a33515be0956fffd632474d550be
ffb99b49ef7a07e4d5c7d4d93950c84e83410c73
F20110114_AABGKI onstine_a_Page_170.tif
85a5473a2a1eb8debce1e4062b08079c
e3a8315188092588b2bd463139aa43cc4dd2782c
F20110114_AABGJT onstine_a_Page_153.tif
276726f241aa9d25062be3f1cf5e57a9
1f281451ceca816e785482e81ea6a73f08805247
F20110114_AABGKJ onstine_a_Page_171.tif
463f316123e8b0936a4f06dd0d806910
47a058459b5b09a7b99e76a7d68dfa12edf938a6
F20110114_AABGJU onstine_a_Page_154.tif
b422fea78b9700297fb8f6114f86787c
0b5102ac53e44dc722471475e83981247f85a5e1
F20110114_AABGKK onstine_a_Page_172.tif
70b192dac8397996054434adf8398fc1
2a5b33e0c6c6ce7943bc5d6dff1572c0c6f810d4
F20110114_AABGJV onstine_a_Page_155.tif
ff9622dc9ab4d308d64098bc4f55ec68
088b5bd76048376b572bc11560f95fc8f23f6102
F20110114_AABGKL onstine_a_Page_173.tif
0a2aede6e9ee8746714885a91bb0d918
15e9bccf849911e9b368d2972e35e7301f5cd20b
F20110114_AABGJW onstine_a_Page_156.tif
48f603bd91a9427b7ba4ae980b61e14b
8739bfe1f98b52fa23c77ac2a5b0a22bbc85b990
F20110114_AABGLA onstine_a_Page_008.pro
b9a5f93dde744da81caa5a7dd4fc0b49
d776cd3a86e5b55d926f696ddb115fa0827149e6
F20110114_AABGKM onstine_a_Page_174.tif
49a523d6dade959c3f82b528459ab350
a9b00aa358fc921373f754e8f46639937d59bb7d
F20110114_AABGJX onstine_a_Page_157.tif
a3c6c6c6dd535223dd6c80f39b8f23be
6da78778373888cac12836b26ef511dcf5d75623
22936 F20110114_AABGLB onstine_a_Page_009.pro
4fdd14a64e07f78a84f55c19bcba5cbb
65f666ebf0e436f84041a15231ae0c28ac73a60e
F20110114_AABGKN onstine_a_Page_175.tif
a4d14446dfec98b84260db58dc078b95
1c8e0f8322c82d32a54c94a1f2f7d06f1705ad82
F20110114_AABGJY onstine_a_Page_158.tif
306bc340a82137d33049b5695f537332
ed547fb803bed3265a3632c7f89ea9a956ec5da0
57861 F20110114_AABGLC onstine_a_Page_010.pro
8028f463856dac5c1fe2a3f6d1c07225
ea65dedaa05dc2b76c53a39f8cdd7b29563575a1
F20110114_AABGJZ onstine_a_Page_159.tif
18926ed4684cd639ac1386ea5e5ea3bb
c9edb55bd5e82b456a4c3a9c83724c586792ebf6
67576 F20110114_AABGLD onstine_a_Page_011.pro
4ab0f83ed08072fa3e6963e40304a853
ac1c033bf8eb9ac46a0ca197893009930b29c9b4
F20110114_AABGKO onstine_a_Page_176.tif
e91f55abac73ce818c2e99dd4ab623a3
5e9a8841bcade0a7394fedcf56c6c978ff4fb6c1
61915 F20110114_AABGLE onstine_a_Page_012.pro
eeff319a83a2ee723fe306f0b6d43028
0eb24077d8774f9f30a946307f282f1724c1701f
F20110114_AABGKP onstine_a_Page_177.tif
3cb64ab26119c094aada6b83bef45e73
d53c13c29e51df80670c138fd332bd92a2ab749c
31525 F20110114_AABGLF onstine_a_Page_013.pro
5ec31485b9ad0cac910d2430d0c64b1e
af4613b315f6f9f4c34454a3b2f17be26b7453e4
F20110114_AABGKQ onstine_a_Page_178.tif
82dd7ab5be912f811202f44fc6bbed83
ab2cac0c07f9c80367a1215042bfb11a859cd2f3
38618 F20110114_AABGLG onstine_a_Page_014.pro
5a2bdb99068b50e648c80f05ff4bb2f9
5cabb46d04563f11d6d13d7c490a53c7af281042
F20110114_AABGKR onstine_a_Page_179.tif
2c16cf9fe056f97ca42a0dfc039310a8
e479867e1f9408738f16f01ad3a85fcb407914ba
40657 F20110114_AABGLH onstine_a_Page_015.pro
5e715706ea99482e78808cd4d108f0ea
3bef8cf566e88edf24fbbfd37ebacbe3006e4d7a
F20110114_AABGKS onstine_a_Page_180.tif
beda42ec65513288535ff908c498b466
74179d9ffb9d99b538ae3526e79781a2c4e91c3d
50801 F20110114_AABGLI onstine_a_Page_017.pro
a9a0010d6a45363f2d264c2e696e3e14
9aa6d25a2bf48fe54fdeb09657fd59f3776866b4
8213 F20110114_AABGKT onstine_a_Page_001.pro
af6a820c483b5ca32872746b8b1adeac
86f0972f6b0c3452b23c416a7f9f464e7515c6fb
50808 F20110114_AABGLJ onstine_a_Page_018.pro
e37c2c343e4f8c05078f3efffba1aa40
943b55438999e86c32f5652c8bf70e8c0e96b3b4
1228 F20110114_AABGKU onstine_a_Page_002.pro
d3983ae8fe4724efa5d37ad49f2255e0
7b2d99db43ea0dd495656f6b65916f6a705fa814
4109 F20110114_AABGLK onstine_a_Page_019.pro
28848bb1d52122adf60d06b42121bd50
60b4ce23cbe3ea4ce1e15ed26a784dfeeaf9ec8e
10232 F20110114_AABGKV onstine_a_Page_003.pro
04ddae232e1e17340ed260cb814aa66e
9d998ef9520cc0cb1b1bd8adcfbf1e6cad00c907
38765 F20110114_AABGLL onstine_a_Page_020.pro
feb399d91fd2dd82647d2e3845ec38e5
0cf4e9aa48bfe4cc17ad76f56167012473803090
43162 F20110114_AABGKW onstine_a_Page_004.pro
d62815d37d47a5b6a5acb126de4a372f
272da00f703851d35de78566b389a8cac57ec6f6
48644 F20110114_AABGLM onstine_a_Page_021.pro
c0902b1c3f1ddfe3268b195fa1be8520
9f503e202b45a9348e6dd7f971d49ba737281f80
17906 F20110114_AABGKX onstine_a_Page_005.pro
0d2d6465d76099938da949705e14d142
803056efa0855f81fe53ee1041917b313bfd844e
51596 F20110114_AABGMA onstine_a_Page_035.pro
acff98988d3417f5a2017dfb25659bd9
79b584d8f112d597ad816784ecb07aba2ccccb49
48639 F20110114_AABGLN onstine_a_Page_022.pro
cd96c3ec3a53a95f9a17caf774fa092d
cf71d46727f8191b184285ccc8951a1b802a6339
79655 F20110114_AABGKY onstine_a_Page_006.pro
bfc1815c5a083fc9e0ab1d2c5c68464a
2c3eb0c27f9dec7fb06e18057d273f68f3cc9018
47365 F20110114_AABGMB onstine_a_Page_038.pro
7e59feeac98f89acd3e5c76c754b6b77
7f4e6c489d29ee77eaaf3412ffd955902ff192e2
50300 F20110114_AABGLO onstine_a_Page_023.pro
7334775311a8a6b4183d3a8d97317ebf
8918df5eb0aa73aa228d325fb5678ef23753c432
94697 F20110114_AABGKZ onstine_a_Page_007.pro
09485d510fb98962fb52ab1f402cb138
b01b5a6991c92a4e3d1c81003c72eec342ff8c7c
45121 F20110114_AABGMC onstine_a_Page_039.pro
94e9935dd0db86d3bf717e4e5da72934
5a3c3fe1fb63bd9bc6a7d2ea555f22854b8aa7a5
46586 F20110114_AABGMD onstine_a_Page_040.pro
683bf8aad771a41e9b64cfbc5da36c63
57f8f86a3c9af5d2e02f9163c3655fe7bb0a72b6
54061 F20110114_AABGLP onstine_a_Page_024.pro
127d049bff64585a11fea446c9a6a46b
7ac711435127b7bc3816d7b5fc3b4da112205a5f
53118 F20110114_AABGME onstine_a_Page_041.pro
91862d7e582e457970f61d8d75e8b4ae
7933fa92d492cda910cf1e70f7e199dc24b301e1
48291 F20110114_AABGLQ onstine_a_Page_025.pro
7ae31ffac99ddde7f5871cb2d4acae72
6381ed1845967d8ed1bbdcaa269f6504831fc225
48155 F20110114_AABGMF onstine_a_Page_042.pro
4e377e43ba3fbf01738faf5b63097f5b
24a399b074813e8c35ca21005d3b0cef4df6bb6f
47635 F20110114_AABGLR onstine_a_Page_026.pro
7b8cff8c3282146785b1506cda6652b3
63f383fcc13db469f25ab554c1cbe0b462543286
52606 F20110114_AABGMG onstine_a_Page_043.pro
7f0dc000c415ac3efe1c92fbd594e58c
0e68085f2d6aab4b8059c74bbc58a1e365b49200
49239 F20110114_AABGLS onstine_a_Page_027.pro
ea4ea8ca48c5e1305253507e769d8ca3
9e81c2821d7adc68b1db5593ca3631f8f3daed95
51911 F20110114_AABGMH onstine_a_Page_044.pro
0378ac064e8710a037360991c8bd096b
485d054357620aefe12879466ca630c2b98150e7
50537 F20110114_AABGLT onstine_a_Page_028.pro
2a9001cc3a1b285083ae8c939b264fd5
225452d5a97fb98c464faf2feae1a2d7f3ff6e3a
49984 F20110114_AABGMI onstine_a_Page_045.pro
fcb37a10163a15d881df61abbf898b1a
768906006994963ccfbc704b7a3f3fe1829c48db
45360 F20110114_AABGLU onstine_a_Page_029.pro
0dc0e1926a7d4b1bd5135ace47d9b262
cd04724cdbe349ca138e8e179c0a5bf1d24b82d4
49783 F20110114_AABGMJ onstine_a_Page_046.pro
03e5cbc4a845a783b230047e9202cb09
0fe250c4da1261d2346cfc0e6d9ef3eeb96056f3
12983 F20110114_AABGLV onstine_a_Page_030.pro
3ebdf9d73c8e1afe4fc64aff4f7b0410
c2187beaa39d63f70fb66c3db7a9c9de9407f252
6514 F20110114_AABGMK onstine_a_Page_047.pro
e21e6bc731f4492c756b8b5e4ff52206
d625698861d9c1632bd1a89290820df27e237655
11524 F20110114_AABGLW onstine_a_Page_031.pro
8e3646ec3a4cac9c81425e02cd714976
168bbc8edbdfc6ce718b2b087e619cb8c40e1037
4397 F20110114_AABGML onstine_a_Page_048.pro
4f711c0411da48753ebb52d8626b8d8b
f68106e406bd2a479be73ae86233b15e0eef1f21
7654 F20110114_AABGLX onstine_a_Page_032.pro
671381dcf3ed20902bfaefb41a7d579b
c3639bed3a7b262eb2812701f9e4ed9a114b6374
49233 F20110114_AABGNA onstine_a_Page_065.pro
d0bd6b19f08aadcf4f8fc8d528166167
141d65253bdcb060825f4855b64871d91eac5bba
3401 F20110114_AABGMM onstine_a_Page_049.pro
47f483d4f0f7c10b15b5f266b48bca47
cf2cb5930c2e40043b744e2fa8285d4e0f588013
10872 F20110114_AABGLY onstine_a_Page_033.pro
9e777622881b969a4c301f88356cbec2
3ad149b9d81818d42ff2823132078f6e765972bd
51152 F20110114_AABGNB onstine_a_Page_066.pro
009db3c4added0ab8fca635ba6729d91
e7cc28a650ab2b256360750f0ac718d968ef4956
4601 F20110114_AABGMN onstine_a_Page_050.pro
aac20df6ffc2b154effd85b190861b43
9c2f2909b7c8631dbb67fd15bb4cee2bc9059b95
41275 F20110114_AABGLZ onstine_a_Page_034.pro
72b23d181d91632b0d9fe5f4dcb1bede
aec1191d425402caaf1e0cb0f13a3d62374c615a
49603 F20110114_AABGNC onstine_a_Page_067.pro
41abe11f17cac6a43ac282247537046e
fdc8ae3266d73ea46c85ce2bce9be452f29aa6bf
12205 F20110114_AABGMO onstine_a_Page_051.pro
86ba8c8bb99d1a27ea1f64910fc712ff
b87fb3ca4e495f914d6c1b6e10ec4cd29612bdfc
35972 F20110114_AABGND onstine_a_Page_068.pro
6ac1557ba486d9a0df94da2aa7b1d756
d4e5b4677f59443c3d16cba70b2d11ef01129bb6
7526 F20110114_AABGMP onstine_a_Page_052.pro
3ac57e4351a14d376318c340616c9759
ff612d52bb04eb32e21e00b9b42756d2cee4e431
12719 F20110114_AABGNE onstine_a_Page_069.pro
b5a55cd0921f3ea8b2dadfe94d378fd2
37772e68a9d44d19c8d53657cd341b34af229fdf
2114 F20110114_AABGNF onstine_a_Page_070.pro
32dfe83a983a8670d4ea4d3eeadf025b
386ae6eb9c73318d0939b3ae8e3ad41a66573af0
3969 F20110114_AABGMQ onstine_a_Page_053.pro
1c6547a6ac239b6ecd13599bb6ae165c
c3def1915843d14e4fdd90b337648f15f4c5f064
7279 F20110114_AABGNG onstine_a_Page_071.pro
71e0939e0ea52243c738db4e0a6e50f9
1483ad12d7c6d792943375d93a61af87fbf20e72
4634 F20110114_AABGMR onstine_a_Page_054.pro
9ffa49851e198b577f649a6a11276bca
20ae200e0bff84010890186525ac69922de46053
4648 F20110114_AABGNH onstine_a_Page_072.pro
87a18fe1c433f8749d06a82b0ab76759
205bdeea822135a00917ee14fcd581967accf1ec
4965 F20110114_AABGMS onstine_a_Page_055.pro
0613a26f2f473fccba30c361c551094a
f3ccbf8cb3485103c44d622ec6258925ac918759
5861 F20110114_AABGNI onstine_a_Page_073.pro
17d8190c9d7791d8c33870106b2353c5
dfc580d57e6ccdfa8020dfa05b9f017e9383ccf5
5375 F20110114_AABGMT onstine_a_Page_056.pro
a90c9fa140b4885ec05ba235343e4102
b1a48fecd63efd97c630a1941cee6127f3282fd2
5757 F20110114_AABGNJ onstine_a_Page_074.pro
e4207439a432f6d0106ad8932a8e21ce
89f08a8b3e9ff126e40005815e38eef7beb701f7
6417 F20110114_AABGMU onstine_a_Page_057.pro
2518d016f7f4466a159cb6068fdbeab4
c1215ec86069ec9028fed416d59588c8c0a63ebf
3403 F20110114_AABGNK onstine_a_Page_075.pro
1594ae40c1ef9194e37df161c9a5bd96
4cdccfea65515550183151cfa4767871b0d07cdb
5028 F20110114_AABGMV onstine_a_Page_059.pro
e7e49dfc9e9521d2278dd59ec1002195
463e77ab2fcc1936a08afd6a7a99c40995be9754
6184 F20110114_AABGNL onstine_a_Page_076.pro
ca8c2345da016186a488a80cd2f6fefb
5c63b937e7e49a2baac9ef22c65fd183d96fec77
7188 F20110114_AABGMW onstine_a_Page_061.pro
bb547f2fed9cd17fb4e521e45aecf017
fb27d4c4bef25de206f0ff79f3c3a67a56ebbcf6
25689 F20110114_AABGOA onstine_a_Page_092.pro
148939b3b32e75e54b63fe76dce9340b
96651493f141a5675983aeb0c37692cde143f023
7287 F20110114_AABGNM onstine_a_Page_077.pro
ebecaa6c6135bb2eeef89ca2851c9fe7
c0b8b3b1c2bc2827b0d1337c0294ffea8fe4d88a
6279 F20110114_AABGMX onstine_a_Page_062.pro
04093679f533785f51037348b9e8f94c
a116809f453f4409484bc2e0324c591facfa1bfc



PAGE 1

GROWTH AND CHARACTERIZATION OF NOVEL GATE DIELECTRICS FOR GALLIUM NITRIDE By ANDREA H. ONSTINE 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 2004

PAGE 2

Copyright 2004 by Andrea H. Onstine

PAGE 3

“I don’t think there is an y concept that you can’t ma ke understandable to the educated lay public. I always tell my st udents and postdocs if you can’t explain to your grandmother what you are doing, pr obably you don’t understand it yourself properly.” —Nobelist Gunter Blobel “Between the stirrup and the ground I as ked for mercy and mercy there I found.” —English/Irish Proverb.

PAGE 4

iv ACKNOWLEDGMENTS This dissertation would not be possibl e without the help and support of many people in my life. I would first like to thank my advisor, Dr Cammy Abernathy, for accepting me into the research group, for al ways having an encouraging word, and for being an inspiration that you can have it all— career, life marriage, and family. Dr. Abernathy remembered that we were people, a nd encouraged us to ha ve a life outside of school. I thank my committee members (Dr. Steve Pearton, Dr. David Norton, Dr. Fan Ren, Dr. Rajiv Singh) for their support a nd knowledge in the field of GaN device processing and semiconductor device physics. I would also like to thank Dr. Amlan Biswas for substituting for Dr. Ren on such short notice. Many thanks go to Dr. Brent Gila for teac hing me everything he knows (or at least as much as I could learn) in the past 4 year s. From pumps and MBE, to characterization, to how to deal with people, Brent’s help and guidance have been invaluable. People will always enjoy working for and with him. Thanks go to my parents and family, who supported me completely in anything and everything that I wanted to try. Special thanks go to Aunt Ja neen and my father, for all of their help editing and proofreading my disse rtation even though they may not have understood everything. I regret that all of my grandparents, especially my paternal grandmother, did not live long enough to see me graduate with my Ph.D. I would also like to thank my group member s and co-workers (Jerry Thaler, Rachel Fraser, Jen Hite, Kimberly Allums, Daniel le Stodilka, Mark Hlad, Andrew Hererro,

PAGE 5

v Jiyhun Kim, Rishab Mehandru, and Wayne Johnson) for all of their help and support. I thank Nina Burbure, Carrie Ro ss, and Erik Kuryliw for all the time they spent with the Focused Ion Beam system. Thanks go to Kerry Siebein for all of her time spent working in the cold with the High Resolution-Tran smission Electron Micros cope as well as the conversation. I thank my long-time friends Aamir Qaiyumi, Amy Matthews, Jennifer Babin nee Adam, Bill Weisner, Bill O’Conner, Pete Me yer, Kathy Daly, Gloria Bergman, Jacques Palmer, Lori Kornberg, and everyone from the Swing Dancing Club, and from Tallwood, for all of their support. It’s hard to believe that it has been so long.

PAGE 6

vi TABLE OF CONTENTS Page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................ix LIST OF FIGURES.............................................................................................................x ABSTRACT.....................................................................................................................xi v CHAPTER 1 INTRODUCTION........................................................................................................1 1.1 Motivation...............................................................................................................1 1.2 Dissertation Outline................................................................................................3 2 BACKGROUND AND LI TERATURE REVIEW......................................................5 2.1 Introduction to Dielectric Films, Capacitors, and MOS(MIS)FETs.......................5 2.1.1 Dielectric Films............................................................................................5 2.1.2 Metal/Oxide/Semiconductor (MOS) Capacitor............................................7 2.1.3 Metal/Insulator/Semic onductor (MIS) Capacitor.........................................8 2.2 GaN Based Electronic Devices...............................................................................9 2.2.1 Silicon Oxide on GaN................................................................................10 2.2.2 Silicon Nitride on GaN...............................................................................10 2.2.3 Aluminum N itride on GaN.........................................................................11 2.2.4 Gallium Oxide on GaN...............................................................................12 2.2.5 Silicon Dioxide on Gallium Oxide on Gallium Nitride.............................13 2.2.6 Gallium Gadolinium Oxide on GaN...........................................................13 2.2.7 Gadolinium Oxide on GaN.........................................................................14 2.2.8 Scandium Oxide on GaN............................................................................14 3 EXPERIMENTAL APPROACH...............................................................................19 3.1 Molecular Beam Epitaxy......................................................................................19 3.1.1 Substrate Preparation..................................................................................21 3.1.1.1 Silicon...............................................................................................22 3.1.1.2 Gallium Nitride................................................................................22 3.1.2 Magnesium Oxide Growth.........................................................................23 3.1.3 Magnesium Calcium Oxide Growth...........................................................24

PAGE 7

vii 3.1.4 Magnesium Scandium Oxide Growth........................................................24 3.2 Materials Characterization....................................................................................25 3.2.1 Reflection High Energy El ectron Diffraction: RHEED.............................25 3.2.2 Transmission Electr on Microscopy: TEM.................................................26 3.2.3 X-Ray Diffraction: XRD............................................................................27 3.2.4 Atomic Force Microscopy: AFM...............................................................28 3.2.5 Scanning Electron Microscopy: SEM........................................................28 3.2.6 Auger Electron Spectroscopy: AES...........................................................29 3.2.7 Ellipsometry...............................................................................................30 3.2.8 Current-Voltage (I-V) measurements.........................................................31 3.2.9 Capacitance-Voltage (C-V) measurements ................................................31 4 MAGNESIUM OXIDE: RESULTS AND DISCUSSION.........................................48 4.1 Effect of Oxygen Plasma Source..........................................................................48 4.2 Effects of Oxygen Pressure..................................................................................50 4.3 Effect of Substrate Temperature...........................................................................52 4.4 Scandium Oxide Capping Layer...........................................................................53 5 MAGNESIUM CALCIUM OXIDE : RESULTS AND DISCUSSION.....................72 5.1 Growth of CaO.....................................................................................................72 5.2 Growth of Ternary MgCaO..................................................................................73 6 SCANDIUM MAGNESIUM OXIDE AN D MAGNESIUM SCANDIUM OXIDE: RESULTS AND DISCUSSION...............................................................................101 6.1 Scandium Magnesium Oxide..............................................................................102 6.2 Magnesium Scandium Oxide..............................................................................103 7 ENVIRONMENTAL AND THERMAL STABILIY..............................................116 7.1 Environmental Stability......................................................................................116 7.1.1 Magnesium Oxide....................................................................................116 7.1.2 Magnesium Calcium Oxide......................................................................117 7.1.3 Magnesium Scandium Oxide and Scandium Magnesium Oxide.............117 7.2 Thermal Stability................................................................................................117 7.2.1 Magnesium Oxide....................................................................................118 7.2.2 Magnesium Calcium Oxide......................................................................118 7.2.3 Magnesium Scandium Oxide and Scandium Magnesium Oxide.............119 8 SUMMARY AND FUTURE WORK......................................................................132 8.1 Magnesium Oxide...............................................................................................132 8.2 Magnesium Calcium Oxide................................................................................133 8.3 Scandium Magnesium Oxide and Magnesium Scandium Oxide.......................134 8.4 Environmental and Thermal Stability.................................................................134

PAGE 8

viii APPENDIX A ELLIPSOMETRY....................................................................................................136 A.1 Maxwell’s Equations and the EM Plan Wave...................................................138 A.2 Jones Vectors and Matrices...............................................................................140 A.3 Light Polarization States....................................................................................141 A.4 Single Films on Thick Substrates......................................................................142 B TEM SAMPLE PREPARATION.............................................................................150 B.1 Old-fashioned Hand Grinding-..........................................................................150 B.2 Focused Ion Beam Milling.................................................................................151 LIST OF REFERENCES.................................................................................................159 BIOGRAPHICAL SKETCH...........................................................................................165

PAGE 9

ix LIST OF TABLES Table Page 2.1 Properties of dielectric material s that have been used on GaN................................16 4-1 Electrical characteriza tion of MgO/GaN diodes......................................................54 5-1 Growth rate and AES data for MgCaO samples......................................................78 5-2 Growth rate, AFM and AES data for MgCaO grown at 300 C...............................79 5-3 Composition and mi smatch from XRD....................................................................80 6-1 Dependence of growth rate, RMS roughness, and AES ratio................................105 7-1 Ellipsometry data. Change in index of refraction.................................................120 7-2 XRR data of as-grown and annealed samples........................................................121

PAGE 10

x LIST OF FIGURES Figure Page 2-1 Capacitor diagrams...................................................................................................17 2-2 Cross-section illustration of a depletion mode n-MOSFET.....................................18 3-1 Typical Knudsen effusion oven (After B.P. Gila 2000)...........................................33 3-2 Riber MBE used for oxide growth...........................................................................34 3-3 WaveMat 610 ECR plasma head.............................................................................35 3.4 Schematic of the Oxford RF plasma source.............................................................36 3-5 AFM images of as received MOCVD GaN and MBE GaN....................................37 3-6 RHEED images showing treate d surface of MOCVD grown GaN.........................38 3-7 RHEED photos indicat ing a (1x3) pattern...............................................................39 3-8 RHEED photos showi ng different types of diffraction patterns..............................40 3-9 TEM column............................................................................................................41 3-10 The relation between the latti ce parameter and Bragg angle...................................42 3-11 Atomic force microscope (after K.K. Harris 2000).................................................43 3-12 Schematic of SEM column.......................................................................................44 3-13 SEM operation. .......................................................................................................45 3-14 The penetration depth and inte raction of an electron beam.....................................46 3-15 Ellipsometer schematic (After D.K Schroder74)......................................................47 4-1 The MgO structure (from Cullity 197875)................................................................55 4-2 Illustration showing the sy mmetry between (111) NaCl and (0002) wurtzite plane.56 4-3 RHEED images indicating clean GaN surface and MgO growth............................57

PAGE 11

xi 4-4 SEM image of the MgO surface grown at 350C with ECR plasma source............58 4-5 AFM images of MgO grown at 350C and MgO grown at 100C..........................59 4-6 RHEED images of MgO after 1 minute a nd after 90 minutes of growth at 100C.60 4-7 SEM and AFM of MgO grown at 100 C with RF plasma source............................61 4-8 Mg/O ratio as determined by AES as a function of oxygen pressure......................62 4-9 Dependance of growth rate on oxygen pressure......................................................63 4-10 AFM of MgO grown at 100 C with oxygen pressure of 1x10-5 or 1x10-4 Torr.......64 4-11 SEM and TEM of MgO............................................................................................65 4-12 XRD shows that the MgO film grown at 300 C vs 100 C.......................................66 4-13 AFM of MgO grown at 100C and grown at 300C................................................67 4-14 HRTEM of MgO grown at 300 C.............................................................................68 4-15 SEM of degraded MgO film.....................................................................................69 4-16 I-V and C-V of MgO/n-GaN diodes, aged15 weeks................................................70 4-17 AFM of MgO with and without a capping layer......................................................71 5-1 Illustration of the CaO structure (from Cullity 1978)..............................................81 5-2 AES shows only Ca and O after growth of CaO at 300 C.......................................82 5-3 MgO-CaO phase diagram.........................................................................................83 5-4 HR TEM of CaO......................................................................................................84 5-5 AFM of CaO grown at 100 C..................................................................................85 5-6 AFM of CaO grown at 300 C..................................................................................86 5-7 AES depth profiling of MgCaO...............................................................................87 5-8 XRD of MgCaO shows no signs of pha se separation or secondary phases.............88 5-9 Powder XRD showing increase in latti ce constant with th e addition of Ca.............89 5-10 HR-XRD of the relative positions of MgO (222) and MgCaO (222) peaks............90 5-11 AFM of MgCaO grown at 100C.............................................................................91

PAGE 12

xii 5-12 AES shows that no Mg or Ca is present after growth time......................................92 5-13 AES of MgCaO film grown at substrate temperature of 300 C...............................93 5-14 XRD shows the peak from the MgCaO film grow at 300 C vs 100 C....................94 5-15 AFM of MgCaO grown at 100 C and 300 C...........................................................95 5-16 AES scans of continuously grown and digitally grown samples.............................96 5-17 HR-XRD showing MgCaO and MgO te xturing in the (111) direction....................97 5-18 High resolution XTEM of MgO...............................................................................98 5-19 HR XRD showing change in peak po sition with change in shutter timing..............99 5-20 AFM of capped MgCaO at di fferent shutter sequences.........................................100 6-1 The Bixbyite crystal structure of scandium oxide. (B.P Gila 2000)......................106 6-2 AES depth profiles taken from layers grown under different growth conditions..107 6-3 AFM surface scans for films grow n at different gorwth conditions......................108 6-4 XRD of ScMgO grown at TMg =350 C and Tsub = 300 C.....................................109 6-5 Powder XRD showing the peak position of MgScO peak.....................................110 6-6 AES of MgScO with Sc2O3 capping layer.............................................................111 6-7 AFM of MgScO grown at 100 C, uncapped and with a Sc2O3 cap.......................112 6-8 AFM of MgScO grown at 300 C without capping layer.......................................113 6-9 TEM of MgScO shows epitaxi al growth of MgScO on GaN................................114 6-10 EDS of ScMgO.......................................................................................................115 7-1 Accelerated aging experimental set up...................................................................122 7-2 Ellipsometry of the degradation of MgO with and without a capping layer..........123 7-3 Ellipsometry of the degradation of MgCaO with and without capping layer........124 7-4 Ellipsometry of the degradation of MgScO with and without capping layer.........125 7-5 XRR of MgO uncapped, before and after annealing at 1000 C for 2 minutes......126 7-6 XRR of capped MgO before and after annealing at 1000 C for 2 minutes............127

PAGE 13

xiii 7-7 XRR of MgCaO uncapped before and after annealing at 1000 C at 2 minutes.....128 7-8 XRR of MgCaO capped before and after annealing at 1000 C for 2 minutes.......129 7-9 XRR of MgScO uncapped, before and after ann ealing at 1000 C for 2 minutes...130 7-10 XRR of MgScO capped, before and after annealing at 1000 C for 2 minutes......131 A-1 An electromagnetic plane wave.............................................................................147 A-2 Geometry of an ellipsometric expe riment, showing pand s-directions................148 A-3 Multiple reflected and transm itted beams for a single film...................................149 B-1 Strips of sample and support sapphire glued together into a sample stack............153 B-2 TEM sample polished to 25 m thick....................................................................154 B-3 The sample glued to a 3mm Cu suppor t ring and orientation of ion milling.........155 B-4 Top view of FIB thinned sample............................................................................156 B-5 Tilted picture of undercut step of FIB sample preparation....................................157 B-6 Sample is now cut free and ready for lift-out.........................................................158

PAGE 14

xiv 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 GROWTH AND CHARACTERIZATION OF NOVEL GATE DIELECTRICS FOR GALLIUM NITRIDE By Andrea H. Onstine December 2004 Chair: Cammy R. Abernathy Major Department: Material Science and Engineering Novel crystalline dielectric materials for gate application on gallium nitride were studied. These dielectric materials must operate at high temper atures and under high power loads. To meet these needs, the sele cted dielectric materials must be thermally stable to temperatures above 1000 C for device fabrication, must be chemically stable to prevent diffusion into the semiconductor, and must have a low defect density to reduce the charged trap sites in th e dielectric and the dielectric/semiconductor interface. The dielectric materials studied were magnes ium oxide (MgO), magnesium calcium oxide (MgCaO), and magnesium s candium oxide (MgScO). These materials were deposited using molecular beam epitaxy (MBE) where the individual elements are supplied independent of each other. This technique allows for the use of a wide range of growth conditions in order to obtain the highest quality material and precise control of the f ilm composition. The dielectr ics were deposited on gallium nitride for characterization and device fabrica tion. The samples were characterized using

PAGE 15

xv a variety of techniques to determine surf ace roughness, crystal structure, chemical composition, and electrical properties. MgO deposited using the MBE approach gr ew epitaxially on GaN for the first 40 monolayers, after which it became po lycrystalline. MgO grown at 300 C using a radio frequency (RF) oxygen plasma source showed the best properties. Optimization of the MgO did not eliminate the problems with MgO as a gate dielectric, namely poor environmental and thermal stability. On annealing, the interfa ce becomes rougher as evidenced by x-ray reflectometry. Crystalline, single-phase MgCaO was gr own by the continuous and digital MBE methods. The digital approach showed be tter uniformity and morphology. From XRD it was found that the addition of Ca could be us ed to vary the bond mismatch over a range of -6.5% to +0.96%. As with MgO, environm ental stability was still problematic because of reactivity with moisture. Bixbyite ScMgO exhibited a solid solubil ity limit of about 9% Mg, after which a magnesium-rich second phase was observed by XR D. This severely limits the usefulness of ScMgO. Rock salt MgScO was grown below this solubility limit and grew epitaxially on GaN. The environmental and thermal stab ility are not significantly affected by the addition of Sc. The environmental and thermal stability of the oxides were also investigated. A Sc2O3 capping layer was shown to improve both thermal and environmental stability of the oxides. The most stable dielec tric was found to be MgCaO with a Sc2O3 cap.

PAGE 16

1 CHAPTER 1 INTRODUCTION 1.1 Motivation The modern microelectronics industry is largely based on solid-state silicon technology. Compound semiconductors are becomi ng an increasingly important area of research and technology because of the limita tions of silicon-based technology. The bandgap of silicon limits the temperature a nd power operation range of the devices. Also, since silicon is an indirect bandgap ma terial, it will never be an efficient light emitter. Compound semiconductors are base d on elements from Groups III and Groups V of the periodic table, such as gallium arsenide (GaAs) and indium phosphide (InP). These compound semiconductor devices have higher carrier mobilities, resulting in faster devices and lower-saturation electric fields than silicon semiconductor devices. Also, some compounds have direct band gaps that lead to efficient light emission and light detection. Transistor research based on compou nd semiconductors has led to several breakthroughs in device performance. Recen tly, research in this field has produced a GaAs metal/oxide/semiconductor (MOS) capacitor that demonstrated properties useful for transistors.1 This discovery led to an operational GaAs-based MOS transistor incorporating gallium gadolinium oxide as a gate dielectric.2 Further research has shown that the gadolinium content of the oxide was responsible for surface passivation, and improved the electrical properties of the device.3-7 There are still limits to these exciting compound materials, such as the thermal ope rating limit and power-handling capabilities.

PAGE 17

2 These obstacles can be overcome by using a semiconductor material with a wider bandgap. The term “wide bandgap” refers to the forbidden energy gap of a material, a region in the energy diagram that is not occ upied by electrons, that is typically greater than 2 eV, (Figure 1.1). Wide bandgap se miconductors have been researched for decades, beginning with silicon carb ide (SiC) in the middle of the 20th century. In more recent years, research has turned to Group III-nitrides (such as gallium nitride), and Group II-VI materials. Because of the ease of defect propagat ion through the II-VI device structure during operation at room te mperature, these materials have been set aside for materials that show more potential The III-V nitride and SiC semiconductors are more thermally stable at device operating temperatures, and therefore do not have the defect-propagation problems associated with II-VI semiconductors. Table 1.1 lists some wide bandgap semiconductors and their properties, as well as those of silicon and gallium arsenide. Nitride-based semiconductors have become a focus for research into optoelectronic devices. By creating ternary alloys, the II I-nitride light emitting devices (LEDs) have covered the entire visible sp ectrum. Other photonic devi ces include UV detectors and laser diodes. Many of the lessons learne d (and processes developed) by photonics research have led to the grow th of high-quality material, im proved electrical contacts, and controllable materials processing. From this base, research has been initiated to create microwave, ultrahigh-power switches and de vices that operate at high temperatures.8 Advances in SiC and GaN have led to power switches based on diffe rent configurations like metal-semiconductor field effect transistor s (MESFETs), heterojunctions field effect transistors (HJFETs), and heterojunction bipola r transistors (HBTs). While these devices

PAGE 18

3 have shown promise for a number of app lications, the metal oxide (or insulator) semiconductor field effect transistor (MOSFET or MISFET) is also a desirable structure. Complementary devices are required for logic circuits. The MOSFET or MISFET structure can be made into complementary metal oxide semiconducto r, CMOS, logic. The MESFET, HJFET, and HBT structures canno t be made into complementary devices. A complementary circuit based on wide bandga p semiconductors will allow for an entire monolithic control circuit to be construc ted for high-temperature/high-power use. For a MOS(MIS)FET to be realized, a highquality dielectric material must be created for the gate insulator. This mate rial must have a bandgap wider than the semiconductor, a dielectric constant larger than the semiconductor, and high temperature stability similar to the semiconductor. Th e materials that have been previously researched for this roll are discussed in Sect ion 2.2. Because of deficiencies in various areas, each of these materials is not optimum for devices on GaN. Finding new materials to satisfy these requirements is the goal of our study. Materials selected for this study were MgO, MgCaO, ScMgO, and MgScO. 1.2 Dissertation Outline The objective of our study was to explore the feasibility of growing lattice-matched oxides for GaN devices by molecular beam epitaxy and characterization of these materials. The background and literature review is given in Chapter 2. In the background, definitions of dielectric materi als, capacitance, and MOSFET are given. The literature review contains descriptions and results of dielectric materials used on GaN. In Chapter 3, the growth methods of the oxides are explained along with the characterization methods. Chapters 4, 5 and 6 describe the results of the dielectric materials grown and discuss how the different di electric materials compare to each other.

PAGE 19

4 Chapter 7 covers the environmental and ther mal stability of the dielectrics. Finally, conclusions and future experi ments are given in Chapter 8.

PAGE 20

5 CHAPTER 2 BACKGROUND AND LITERATURE REVIEW 2.1 Introduction to Dielectric Films, Capacitors, and MOS(MIS)FETs The following sections discuss the basics of dielectric films, their properties, and applications. The capacitor is an important tool for testing these dielectric materials. The metal/insulator/semiconductor transistor is a so lid-state switch that is controlled by the capacitor structure in the gate region. 2.1.1 Dielectric Films Insulators are characterized by the absence of charge transport. Insulators have positive and negative charges in the form of atom nucleus and electron cloud; but these charges are bound to the atom or molecule, a nd are not available for conduction. When materials are placed in an electric field, there is a shift,( or polarization) in the charge distribution, and it is this polar ization that leads to dielectr ic behavior in the material.9 The polarization induces dipoles within the atomic or molecular structures that are aligned with the applied field. Th e ability of a material to resi st the polarization of charge is described as the dielectric constant, which is the ratio of the permittivity of the material, i, to the permittivity of vacuum, v. = i/ v (2-1) The dielectric constant can also be relate d to the internal field created within the material and the external app lied field, through the equation 2-2.10 Einternal = Eapplied/ (2-2)

PAGE 21

6 The polarization, P, of the material is related to th e dielectric constant by the Equation 2-3, where is the strength of the electric fiel d (V/m). It can be assumed from this relation that the polarization increases as the electric field strength increases, until all the dipoles are aligned such that P=( -1) o (2-3) There are several applications for dielectric materials. Passivation of high-voltage junctions, isolation of devices and interconne cts, and gate insula tion of field-effect transistors are a few applications that are relevant to this di scussion. For a material to be a successful dielectric, it must meet certain criteria. Desirable characteristics include chemical stability over the lif etime of the device, immob ile charge traps (to avoid shorting and frequency limits), and a dielec tric constant higher that that of the semiconductor (to avoid generating a high electric field in the di electric). In the case of the wide-bandgap semiconductor devices, the dielectric materials must also have excellent thermal stability, since the high-pow er applications will result in elevated operating temperatures. Another criterion im portant to wide bandgap semiconductors is that the band gap of the dielectric must be greater than that of the semiconductor. The ideal dielectric would keep the electrons and holes in the semiconductor, and out of the dielectric. To achieve these properties, the ba ndgap of the dielectric must be larger than the semiconductor, and the electron affinity near er to the vacuum leve l, according to the electron affinity rule, Equation 2-4. Ec + Ev = Eg (2-4) The dielectric/semiconductor interface is also an important focus of research in the area of device processing. The inte rface state density of car rier traps must be

PAGE 22

7 <1011 eV-1cm-2 for a device to be considered succe ssful. Another important focus is the fixed dielectric charge densit y, or carrier trap density with in the dielectric. To date, several dielectric materials have been rese arched for use in wide-bandgap semiconductor switches, including AlN, Al2O3, Ga2O3, Gd2O3, Ga2O3(Gd2O3), SiOx and Si3N4, Table 2.1. 2.1.2 Metal/Oxide/Semiconductor (MOS) Capacitor Since the gate is the actual on/off switch in the transistor structure, defining the properties of the gate and its operation are extremely important. The gate structure is identical to the metal/oxide/semiconductor ca pacitor, (Figure 2-1) A capacitor is a device made of two parallel, conducting plates separated by an insulating material. When a direct current (DC) voltage is applied to one side of a capacitor, an equal and opposite charge forms on the other side of the capacitor. In most cas es, the DC voltage is applied to the metal side of the cap acitor, and the charge is formed in the semiconductor. The amount of capacitance that th e capacitor can hold is directly related to the dielectric constant of the dielectric material,11 (Equation 2-5) where C is the capacitance, 0 is the permittivity of vacuum ( 0=8.854x10-14 F/cm), i is the permittivity of the material, A is the area of the metal contact, and d is th e thickness of the dielectric material. C= 0 i A/d (2.5) The capacitance is independent of the applie d voltage to the gate, but is completely dependant on the geometry and the dielectr ic constant. This gives a theoretical capacitance for a given device geometry. If the applied electric field becomes too great, the charges are ripped from the ma terial and conducted to the charged plates. This leads to a short in the material, a nd is termed dielectric breakdown.

PAGE 23

8 However, as usually happens in practice, the theoretical value and the measured value rarely match. The measured capaci tance value is actually the sum of two capacitors in series. These are the dielec tric capacitance and the semiconductor spacecharge layer capacitance. The semiconductor capacitance, Cs, is responsible for deviation in the measured capacitance, and is voltage-dependant.12 As the bias is applied to the metal, the majority carriers in th e semiconductor are repelled from the oxide/semiconductor interface, resu lting in a space-charge laye r. This is assuming that the applied bias is the same charge as the majority carriers in the semiconductor. The space-charge layer is populated by minority carri ers, and given the name majority-carrier depletion layer.11 The frequency of the field applied to the cap acitor can also affect the trap layer in the semiconductor. Traps within the semic onductor material become filled with carriers that are attracted to the diel ectric/semiconductor interface. Wh en the bias is released, the traps are emptied. If the frequency become s too high, the traps do not have sufficient time to empty and form a charge layer. At low frequencies, this layer is almost nonexistent. As the frequency is increased, however, this trap laye r becomes thicker and adds to the total capacitance. Capacitance measurements must be made at extremely low frequencies, called quasi-static frequencies, to obtain the capac itance that is purely within the dielectric material. 2.1.3 Metal/Insulator/Semiconductor (MIS) Capacitor Modern silicon technology is based on complementary pairs of metal oxide semiconductor (CMOS) transistors. This is one of the most common devices found in logic and memory circuits. The gate region of the transistor determines the capabilities of the device. The two types of metal oxi de semiconductor field effect transistor

PAGE 24

9 (MOSFET) devices are depleti on mode and enhancement mode. In the depletion-mode device, the material type under the source, ga te and drain regions is the same. This device is in the “on” state when no gate vol tage is applied. In the enhancement-mode device, the material type under the gate is oppo site that under the source and drain. This device is in the “off” state when no applied ga te voltage is applied. Figure 2-2 shows a cross-section of a depletion-mode MOSFET. Because of the relatively low p-type carrier concentration available for p-GaN, only n-type depletion-mode devices were considered. The n-type MOSFET was used to describe the operation of the gate in the transistor. For a p-type MOSFET, the gate volta ge is reversed. When there is a zero gate voltage, carriers are free to flow from the sour ce to the drain in the MOSFET structure. The switch is “on”. As a negative voltage is applied to the gate contact, electrons under the gate dielectric are repelled (like charges repel) and a posi tive charge is induced in the semiconductor under the gate region. This positive charge hinders the ability for electrons to flow from the source region to the drain regi on. As the gate voltage is increased, more positive charges collect under the gate until the flow from source to drain is completely stopped. The voltage is called the pinch-off voltage, since it effectively pinches the channel shut. The transistor is now “off”. As the current through the sourcedrain is increased, it requires more gate volta ge to successfully pinchoff the carrier flow. Thus the maximum operating parameters of the device are determined by the amount of electric field that can be a pplied to the gate before the dielectric breakdown occurs. 2.2 GaN Based Electronic Devices Gallium nitride research has resulted in l ong-lifetime, room temperature operation of photonic devices. These include LEDs that cover the visible spectrum, laser diodes in the blue and blue-green regime, and UV det ectors. These devices are just recently

PAGE 25

10 reaching production levels with problems still to be solved in the fi elds of n-Ohmic and p-Ohmic contacts, p-type doping issues, Schottk y contacts, and dielectric materials. Also, with the lack of availability of highquality GaN substrates, research of epitaxial growth and substrate selecti on is still ongoing. From materi al and processing advances learned from the photonics research, high-pow er and high-temperature switches have been realized. The following summarizes so me of the dielectrics research to date. 2.2.1 Silicon Oxide on GaN Silicon oxide is a very attractive choice for a dielectric material since is has been well studied and the processing is well esta blished. Silicon oxide deposited by plasma enhanced chemical vapor deposition (PECVD)13-16 has been reported to give interface state densities on the order of low 1011 eV-1cm-2. Silicon oxide deposited by electron beam (EB) evaporation has show n interface state density of 5.3x1011 eV-1cm-2.15 After annealing the EB deposited SiOx at 650C the valance band offset was measured to be 2eV and the conduction band offset was measured to be 3.6 eV.17 The EB evaporated SiOx shows a silicon rich stoichiometr y when compared to the PECVD SiOx. There are several inhere nt problems with SiOx as a dielectric material for wide bandgap semiconductors. The high Dit of SiOx is due to uncontrolled oxidation of the surface18. The most significant limitation is that SiOx has a dielectric constant ( ) of 3.9, which is significantly lower than that of GaN ( = 8.9). This will create a very large electric field in the dielectric leading to further breakdown. 2.2.2 Silicon Nitride on GaN Silicon nitride deposition by PECVD15,16 reported an interf ace state density of 6.5x1011 eV-1cm-2. This value is reasonable for a first attempt. But when compared to

PAGE 26

11 GaN MESFETs, the Si3N4 MISFET was out performed. Electrical measurements showed the MISFET structure had a large flat band voltage shift (3.07 V) and a low breakdown voltage (1.5 MV/cm). There was no microstructu ral analysis performe d on the deposited Si3N4 films. SiN deposited by ECR plasma CVD showed a Dit of 1x1011cm-2eV-1 but has excess leakage current due to small conduction band offset.18 The ECR-CVD MIS diode showed a Dit of 4x1011 cm-2eV-1, fixed oxide charge of 1.1x1011 cm-2 and breakdown 5.7 MV/cm with a dielectric constant of 6.19 A unique dielectric structure of SiO2/Si3N4/SiO2 (ONO) was reported to have breakdown field strength of 12.5 MV/c m for temperatures as high as 300 C.20 The ONO structure was deposited by jet vapor deposition to a thicknes s of 10 nm /20 nm /10 nm. The multiplayer structure does allow for unique engineering of a dielectric, but multiple interfaces can lead to an extremely large number of inte rface state traps and increased processing. The Dit for the ONO structure was shown to be less than 5x1010 eV-1cm-2 with breakdown fields greater than 12 MV/cm.21 2.2.3 Aluminum Nitride on GaN Aluminum nitride deposited by MBE a nd MOCVD has been used to create MISFET devices and insulated gate heterostru cture field effect transistors (IG-HFET) devices.22,23 The AlN MISFET structure grown at 400 C was polycrystalline. From xray reflectivity measurements, the AlN/Ga N interface showed a roughness of 2.0 nm. This may be due to the polycrystalline nature of the film or from intermixing of the AlN and GaN at the interface. The dielectric breakdown fi eld was calculated to be 1.2 MV/cm. The AlN IG-HFET structure was grown at 990 C, forming a single crystal film of 4.0 nm. This device operated in enhancement mode and had a pinch-off voltage

PAGE 27

12 of 0 V. Hexagonal aluminum nitride has a 2.4% lattice mismatch with hexagonal GaN on the (0001) plane. The 4.0 nm film thickne ss is greater than th e critical thickness allowed for elastic deformation leading to th reading dislocations forming from plastic deformation. Single crystal AlN and polycry stalline AlN films suffer from defects and grain boundaries that cause shorting. 2.2.4 Gallium Oxide on GaN Gallium nitride forms a stable native oxide. This oxide has been considered as a dielectric material, like the native oxide of silicon. Thermal oxidation of the GaN surface has lead to some interesting resear ch. Oxidation was performed in dry24,25 and wet26 atmospheres. Dry oxidation of GaN epilayers at temperatures below 900 C showed minimal oxidation. Dry oxidation at 880 C for 5 hours showed 1110 nm of -Ga2O3 with a Dit of 1x1010 eV-1cm-2 and showed inversion.27 At temperatures above 900 C, a polycrystalline monoclinic Ga2O3 forms at a rate of 5.0 nm/hr. This oxidation rate is too slow to be viable as a processing step. Wet oxidation of GaN also forms polycrystalline monoclinic Ga2O3, but at a rate of 50.0 nm/hr at 900 C. From cross-sectional transmission electron microscopy, the interface between the oxide and the GaN is found to be non-uniform. Scanning electron micr oscopy shows that both films are rough and faceted. Electrical characterization of th e oxide shows the dry oxide dielectric field strength of 0.2 MV/cm and the wet oxide dielec tric field strength of 0.05 to 0.1 MV/cm. Some limits to the thermal oxidation of GaN are that only one microstructure has been formed from this process and GaN is consumed in the process. XRD shows this to be a high temperature hexagonal phase28. Ga2O3 passivates the surface,28-31 and has a Dit of 1011 eV-1cm-2 for GaN MOS. A negative oxide char ge as well as high capacitance and

PAGE 28

13 reduced reverse leakage where shown for thicker oxides grown by PEC.29 Using PEC and HeCd laser, a low reverse leakage curren t of 200 pA at 20 Vm has been achieved. For this oxide the forward breakdown, Efb, is 2.8 MV/cm and the reverse breakdown, Erb, is 5.7 MV/cm with a Dit of 2.53 x 1011 cm-2eV-1. The dielectric constant of Ga2O3 grown under these conditions is estimated at 10.6.33 With amorphous GaO deposited by PEC, low leakage currents of <10 x 10-8 /cm at –15 V have been measured. Using Ga2O3 as both gate dielectric and passi vation layer, a breakdown fiel d of 0.4 MV/cm was observed. The bandgap, Eg, of Ga2O3 was measured to be 4.4 eV.34 2.2.5 Silicon Dioxide on Gall ium Oxide on Gallium Nitride Depending on the growth technique, the in terface between the SiO and the GaN can vary drastically. When SiO is deposited by RPECVD, a parasitic subcutaneous layer of native gallium oxide is grown on the GaN surface. This layer as been shown to have a direct effect on the device pe rformance. When the thickness of the initial GaO layer is controlled by a pre-oxidation step the device characteristics improve markedly. Remote Plasma Assisted Oxidation first followed by RPECVD gave a lower Dit and a smaller flat band shift over the RPECVD of the SiOx alone.35 Real and ideal CV curves are nearly identical.36 After an anneal in an RTA for 1 min at 900 C in Ar the Dit is 2-3x1011 cm-2.37,38 Another group used a similar oxide growth technique and measured a Dit of 3.9x1010 eV-1cm-2 and a low leakage current.39,40 2.2.6 Gallium Gadolinium Oxide on GaN Due to the recent success of gallium gadolinium oxide (GGG) as a dielectric in GaAs MOSFETs41-46, attention has turned toward this as a dielectric material for GaN. The GGG dielectric was deposited on a Ga N epilayer by EB evaporation of a single crystal GGG source.22 The substrate temperature was 550 C. The interface roughness

PAGE 29

14 was calculated to be 0.3 nm from x-ray re flectivity. Metal oxide semiconductor (MOS) capacitors were formed and tested. A br eakdown field of greate r than 12 MV/cm was estimated. More recently, the thermal stability of the film and the interface has been proven to temperatur es as high as 950 C and operation of a depletion mode MOSFET has been performed at temperatures up to 400 C.47 The EB evaporated GGG stoichiometry is heavily dependant upon the substrate temper ature. Changes in temperature lead to changes in the stoichiometry.48 This limits the available microstructure obtainable within the stoichiometric limits of GGG. 2.2.7 Gadolinium Oxide on GaN GaN based MOSFETs have been made that used a stacked gate oxide consisting of single crystal gadolinium oxide and amorphous SiO2.49 The gadolinium oxide provides a good oxide /semiconductor interface and the SiO2 reduces the gate leakage current and enhances oxide breakdown voltages. The dislocations in the Gd2O3 film limit the breakdown field that can be su stained in the dielectric. 2.2.8 Scandium Oxide on GaN Scandium oxide grown by MBE has been us ed as a gate dielectric and passivation layer for GaN based devices. Scandium oxi de has the bixbyite crystal structure, a reasonable band gap of 6.3 eV and a lattice mismatch to GaN of 9.2%. The scandium oxide was grown by MBE using an RF oxyge n plasma, substrate temperature of 650 C and an effusion cell temperature of 1130 C. The surface state density is 8.2x1012eV-1cm-2 and showed inversion for gated diodes.50 This oxide was also grown under the same conditions except the substrate temperature was lower to 100 C, which resulted in an interface state density of 5x1011 eV-1cm-2.51

PAGE 30

15 Scandium oxide has also been used as a fi eld passivation layer fo r GaN devices. It has been shown to reduce the reverse leakage current an increase the fT and fMAX. The passivation films have been grown w ith a substrate temperature of 100 C and a cell temperature of 1130 C.52 Scandium oxide has better long-term stability than SiNx as a passivation film for GaN based HEMTs. It has been shown to dr amatically reduce the gate lag problems due to surface states on AlGaN/GaN HEMTs.53

PAGE 31

16 Table 2.1. Properties of dielectric ma terials that have been used on GaN Material Bandgap (eV) Dielectric ( ) Melting Point (K) References AlN Al2O3 Ga2O3 Gd2O3 Ga2O3(Gd2O3) SiOx Si3N4 MgO Sc2O3 6.2 5.75 4.4 8.5 4.7 8-9 5.0 7.3 6.3 8.5 12 10 11.4 14.2 3.9 7.5 9.8 11.4 3273 2319 2013 2668 2023 1993 2173 3073 2678 54,55,56 57,58,59 60,61,62 63,64,65 2,66 12,59,67 12,67,68 59,69,70,71 72

PAGE 32

17 Figure 2-1. Capacitor diagrams. A) Typi cal metal-insulator-semiconductor capacitor with backside ohmic contact. B) Typi cal planar capacitor produced in our study. Semiconductor VGMetal Insulator Ohmic contact d A B

PAGE 33

18 Figure 2-2. Cross-section illustration of a depletion mode n-MOSFET. In the top figure, the device is in the “ON” state with VG=0. The bottom figure is the device in the “OFF” state with VG<0, notice the conduction ch annel is pinched-off. n-GaN +n-GaN source gate drain VG=0 n-GaN +n-GaN source gate drain VG<0 + + + + + + + + + +

PAGE 34

19 CHAPTER 3 EXPERIMENTAL APPROACH 3.1 Molecular Beam Epitaxy Molecular beam epitaxy (MBE) was the grow th method employed in this work to produce the dielectric films. The dielectri c materials were deposited in an ultra-high vacuum environment from the purest attain able elements. In MBE, the individual elements of the compound are provided to th e growth surface independent of each other, allowing for a high degree of control over the stoi chiometry of the dielectric material that conventional sputtering and electron beam ev aporation do not allow. Also, MBE allows for precise control of the substrate temperat ure, which in turn helps to control the microstructure of these materials. The growth rate is dependent upon the substrate temperature, the ratio of the elements, and also the rate at which the elements are supplied to the substrate. In conventional MBE, beam s of atomic or diatomic species are produced from ovens called Knudsen cells (Figure 3-1). The purity of the atomic beam depends upon the vacuum level in the chamber and the purity of the source material. The number of atoms emitted from the Knudsen cell is relate d to the temperature of the cell and the relative atomic mass of the material in the ce ll. This relation is modeled by the Equation 3-1 where F is the flux of the Knudsen cell in atoms/cm2-s, p is the vapor pressure in the cell in Torr, a is the orifice area in cm2, d is the cell to substrate distance in cm, M is the atomic mass of the element in amu, and T is th e temperature of the ce ll in degrees Kelvin.

PAGE 35

20 s cm atoms MT d a p x F2 2 / 1 2 22) ( ) )( ( 10 18 1 (3-1) For these oxides, a Riber model 2300 MBE wa s modified to perform the growth. A sketch of this system is shown in Figure 3.2. The main growth chamber was pumped on with a cryopump allowing for a base pressure of 2x10-9 Torr. This modified system is equipped with a Reflection High-Energy Electron Diffraction (RHEED) system, described in Section 3.2.1. This allows for the arrangement of the top few monolayers of atom to be determined. This is extremely important in determining the atomic arrangement of the starting growth surface and in determining the structure of the film. The oxygen source for the oxide growth was a WaveMat model 610 electron cyclotron resonance (ECR) plasma source, (F igure 3-3) operating at a frequency of 2.54 GHz and powers ranging from 100 to 200 watts or an Oxford Applied Research radio frequency (RF) oxygen plasma source, Figur e 3-4, operating at 13.56 MHz with the RF power set at 300W. Oxygen is supplied to the plasma generator through a leak valve using a 99.995% oxygen source. The ECR plasma sources works by microwave energy that is guided into the sour ce chamber and coupled into the oxygen molecule electron cloud. A series of permanent magnets around the source chamber create a magnetic field, which accelerates the electron motion into helic al paths that collide and ionize the source gas molecules. This creates a dense plasma in the source chamber. The plasma contains the atomic species for growth, as well as ionic and molecular species. The RF plasma source operates by means of an electrical di scharge created from i nductively-coupled RF excitation.73 Atoms produced by dissociation in th e discharge tube can escape into the vacuum environment along with the undissociated molecules via an array of fine holes in the aperture plate. The electrical potentials are such that negligible currents of ions or

PAGE 36

21 electrons escape form the discharge during nor mal operation of the source. Dissociated atoms undergoing wall collisions in the discha rge chamber exhibit a low recombination coefficient and may also ultimately contribute to the radical beam flux. The pressure in the discharge chamber is sufficiently low that atom-gas collisions are minimal over the dimensions of the discharge chamber.73 The substrate temperature was determined by a backside thermocouple in close proximity with the substrate holder. The s ubstrate thermocouple was calibrated by using the melting points of galliu m antimonide (GaSb) at 707 C and indium antimonide (InSb) at 525 C. Pieces of GaSb and InSb were heated in the growth position with a nitrogen plasma impinging on the surface. This reduces the chance for loss of the Group V, Sb, species during the heating process, whic h would result in an incorrect melting temperature. With out the nitrogen over pr essure, the InSb and GaSb would degrade by loss of Sb, the more volatile Group V species, before melting. 3.1.1 Substrate Preparation Prior to any epitaxial growth, the substrates receive an ex-situ and in-situ surface treatment to remove any contamination and the native oxide. The surface of the semiconductor must be as clean and planar as possible to ensure high quality dielectric film deposition. Surface contaminat ion leads to impurities at the dielectric/semiconductor interface, which ultimat ely results in creating interface traps, as described in Section 2.1.2. The substrates we re visually inspected as well as scanned using atomic force microscopy (AFM), descri bed in Section 3.2.4. This gave a reference for the surface roughness to compare to the fi nal product. The substr ates used in this work are silicon (Si) an d gallium nitride (GaN).

PAGE 37

22 3.1.1.1 Silicon All of the initial experiments were carri ed out on Si single crystal substrates oriented in the <001> direction. This was due to the wide availability of Si substrates and their low cost. The data gathered from the Si substrates is used to calibrate the growth rate and composition of the diel ectric films. The microstruc ture of the dielectric films grown on Si will be different from that on GaN since the surface atomic spacing and the crystal structure of the substrates are different. Silicon wafers received as ex-situ treatment of a 30 second wet etch in a buffered oxide solution consisting of 6 parts ammoni um fluoride and 1 part hydrofluoric acid, rinsed in deionized water (DI), and dried under nitrogen gas. This tr eatment results in a smooth surface with a surface roughness root mean square (RMS) value of 0.08 nm, as seen in AFM. This surface is oxide free and stable for a period of up to one hour. The in-situ cleaning consists of h eating the silicon up to 200 C in order to drive off any moisture from exposure to the atmosphere. 3.1.1.2 Gallium Nitride Since gallium nitride wafers substrates are not currently available, gallium nitride grown on sapphire wafers oriented <0001> were used. These will be referred to throughout this work as GaN s ubstrates. Two different t ypes of growth of the GaN substrates were employed in this work, MBE and metal-organic chemical vapor deposition (MOCVD). The MBE GaN substrat es were provided by SVT Associates and the MOCVD substrates were provided by Epitronics, QinetiQ, and Uniroyal Optoelectronics (UOE). From AFM, ther e is a large differen ce in surface roughness between the two types of GaN. The MOCV D substrates are 13 nm RMS roughness and the MBE substrates are ~6 nm RMS roughness (Figure 3-5).

PAGE 38

23 The GaN substrates received an ex-situ treatment consisting of a 3-minute etch in (1:1) hydrochloric acid: water, followed by a DI rinse and blown dry by nitrogen. This was use to remove any organic residue from the surface. Then a 25-minute exposure to ozone produced by an ultraviolet lamp in a UVOCS UVO Cleaner model number 42-220 was used to oxidize the carbon on the surface and create a thin native oxide layer. Next, the substrates received another etch in buffe red oxide etch for 5 minutes, in the solution described for silicon, to remove the nativ e oxide. This is shown by observing the reflection high-energy electr on diffraction (RHEED) pattern produced from the surface, described in Section 3.2.1. The RHEED patte rn produced by the native oxide is more diffuse than the pattern produced by the buffered oxide etched surface (Figure 3-6). The GaN substrates were then mounted to molybdenum blocks using indium solder, and then loaded into the load-loc k of the MBE. Room temperature RHEED measurements showed a reasonably clean (1x1) surface (Figure 3-6). Two different crystal directions are observed in RHEED to create a more complete understanding of the surface. Here, the <1-100> and the < 11-20> directions are observed. An in-situ thermal treatment was employed to further remove any oxide or contamination left on the surface. The substrates were heated to 700 C in vacuum and no overpressure of nitrogen was used. The RHEED patterns reco rded at this temperature in dicate a sharp (1x3) pattern (Figure 3-7). 3.1.2 Magnesium Oxide Growth The magnesium oxide samples were grown from 99.99% pure magnesium and Knudsen cell temperatures ranging from 350 C to 400 C. Substrate temperatures between 100 C and 340 C were used. The oxygen was provided by an Oxford RF

PAGE 39

24 plasma source at 300 W forward power or an ECR Wavemat plasma source set to 200 W forward power. There was a significant differen ce in the properties of the films between the 2 plasma sources. The RF plasma sour ce produced the superior films. Oxygen pressure was varied from 8x10-6 up to 1x10-5 Torr. In all cases, the sample rotation was kept at 15 rpm. 3.1.3 Magnesium Calcium Oxide Growth The magnesium calcium oxide samples were grown from the same Mg sources as that used for the MgO with the addition of 99.999% pure calcium, Ca, from another Knudsen cell with temperatures ranging from 450 C to 500 C. Substrate growth temperatures were between 100 C and 300 C. The two growth methods that were used are continuous where all shu tters open at once and exposed to the substrate and digital alloying where alternating layers of MgO and CaO. Changing the flux of the metal sources during a continuous growth or ch anging the timing of the shutter sequences during digital growth varied the composition of the film. Oxygen pressure was held at 8x10-6 Torr and used only the RF plasma source. As in the MgO growth, the sample rotation was kept at 15 rpm. 3.1.4 Magnesium Scandium Oxide Growth The magnesium scandium oxide films were grown using the same Mg source with the addition of a Sc metal Knudsen ce ll with temperatures ranging from 1090 C to 1190 C due to the extremely low vapor pressure of scandium. Substrate temperatures were between 100 C and 300 C. Growth methods for this ternary oxide were continuous and digital as in the MgCaO films. It wa s not possible to take a flux reading of the scandium due to severe fluctuations in r eadings, the needle bounced around, when the

PAGE 40

25 cell at temperature due to out-gassing of He from the Sc source metal. The out-gassing of He was shown by the mass spectrometer. The composition of the film was varied by changing the temperature of the source meta ls. As a convention, films with a higher amount of scandium then magnesium are refe rred to as scandium magnesium oxides, ScMgO. Films with a greater amount of magnesium than scandium are referred to as magnesium scandium oxides, MgScO. Oxygen pressure was held at 8x10-6 Torr and used only the RF plasma source. As in the MgO growth, the sample rotation was kept at 15 rpm. 3.2 Materials Characterization The films were heavily characterized afte r growth. Emphasis of the research was placed on, but not limited to, the microstructu re and the stoichiometry of the epitaxial films and how these properties related to th e electrical properties, environmental and thermal stability of the dielectric materials. The dielectric films were annealed to temperatures as high as 1000 C by a rapid thermal anneal (RTA) process to determine the thermal stability of the films. 3.2.1 Reflection High Energy Electron Diffraction: RHEED In-situ structural characterizati on can be done in the grow th chamber via reflection high-energy electron diffraction (RHEED). A RHEED system consists of an electron gun, typically 5 to 30 kV, and a phosphorescent sc reen. The electrons from a filament are collimated, accelerated, and reflected off the surface of the sample. A diffraction pattern is seen on the phosphorous screen. Fr om this diffraction pa ttern, single crystal, polycrystalline and amorphous films can be di fferentiated. This technique was used to determine the starting substrate surface quality and the quality of the films grown while

PAGE 41

26 in the ultra-high vacuum syst em. Also, the method of growth initiation, which has an enormous impact on the overall film qualit y, can be determined from RHEED. RHEED reflections are created by diffraction from the surface of the substrate. The incoming electron beam has a grazing incide nt angle of 1 to 2 degrees. Diffraction occurs only along certain crystal directions in a singl e crystal material. From the type of pattern, intensity, and spacing between different diffraction events, a 2-dimensional map of the surface can be created. It is this ma p that will help determine the condition of the starting substrate as well as of the grown film. A surface growing layer-by-layer (2D) will produce a pattern with streaky lines, whereas a surface growing by islanding (3D) will produce a pattern that is spotty. Poly crystalline surfaces show a ringed pattern and amorphous surfaces show almost no pattern at all (Figure 3-8). 3.2.2 Transmission Electron Microscopy: TEM One of the most powerful microstructu ral analysis techni ques available is transmission electron microscopy (TEM). Fr om TEM, not only can the microstructure of an epitaxial film be determined, but also deta iled analysis of defect s in the film, atomic imaging of the interface, and an accurate cal culation of lattice spacing is determined. TEM uses a beam of electrons that pass through and interact with a very thin sample to form an image on the other side of the sample (Figure 3-9). The in teractions between the atoms in the sample and the electrons produce th e contrast seen in the image. One of the major drawbacks of TEM is sample prepara tion required to obtain the images. The sample must be cut, polished, and thinned to electron transparen cy (~100 nm) via hand polishing and ion beam milling or by using a focused ion beam (FIB) system. A complete description of the sample prepar ation is given in Appendix B. This is especially difficult for the nitride materials due to their hardness. The FIB used to make

PAGE 42

27 these sample was a FEI Strata DB (Dua l Beam) 235 FIB. A JOEL 200CX operating at 200 keV was used for film analysis and JO EL 2010FX operating at 400 keV was used for high-resolution analysis of the interface. 3.2.3 X-Ray Diffraction: XRD Another structural analysis technique is x-ray diffracti on (XRD). This technique had virtually no sample preparation when co mpared to TEM. X-rays are diffracted off the sample to produce characteristic peaks of the atomic planes in the sample. The full width at half maximum (FWHM) of these char acteristic peaks is used to determine the crystalline quality of the film s. Powder x-ray diffraction can be used if the samples are polycrystalline or polycrystalline/amorphous. This was used in preliminary sample analysis to determine if second phases were present. X-ray reflectivity (XRR) from the air/film interface and the film /substrate interface help determine the roughness of these interfaces and the thickness of the film. The powder system is a Phillips 3720. The highresolution system is a Phillips MPD 1880/HR w ith a 5-crystal analyzer. The x-ray source for both systems is a copper (Cu) K x-ray source. Figure 3-10 shows an illustration of the sample geometry for the x-ray diffrac tion. Samples were scanned measuring the 2 with the GaN peak optimized for the high -resolution system. For the powder system, peak intensity2 was measured. For the x-ray scan, the 2 peak positions are obtain and using Bragg’s Law, Equation 3-2, the d-spacing between the corresponding planes are calculated. sin 2 d n (3-2) Here, is the wavelength of the incident xray, d is the spacing between the planes, and is the measured peak position. The d-sp acings are compared to known values of the material to determ ine crystal orientation.

PAGE 43

28 3.2.4 Atomic Force Microscopy: AFM The surfaces of the grown f ilms were characterized using atomic force microscopy (AFM) to give a quantitative analysis of film morphology or surface roughness. Tapping mode AFM was used to obtain a root mean square (RMS) roughness of the surface. In tapping mode, the tip of a stylus, made from Si3N4, is brought into close proximity to the surface, close enough to be defected by van de r Waals forces of the surface atoms. A laser is reflected off the AFM tip and colle cted into a photodiode (Figure 3-11). The intensity of the reflected light is read as height. The tip is rastered across the surface and each point is read as a height, creating a 3-di mensional map of the surface. From this 3dimensional map, a surface roughness is calculat ed and from this a RMS roughness. This is very useful in characterizing the starting subs trate, the dielectric film, and the effects of the various growth and processing steps. The sensitivity of the AFM is largely dependent on the sharpness of the tip and the sensitivity of the deflection. The tapping mode tip used here has a tip radius of 5 nm and the deflection sensitivity is about 0.01 nm. This makes tapping mode extremely sensitive to su rface roughness. An alternative mode of operation is contact mode. However, the tip ra dius is about 20 nm, which greatly reduces the resolution. The AFM used in this st udy was a Digital Instruments Nanoscope III. 3.2.5 Scanning Electron Microscopy: SEM The surface morphology of the samples is characterized on the macro level by using the secondary electron microscope (SEM ). The SEM technique enables an image of the surface to be taken at very high magnifications, from between 5000X to 100,000X. In this technique, an electron beam of en ergy between 5 keV and 30 keV is rastered across the sample surface. A schematic of the SEM is shown in Figure 3-12. One requirement of an SEM sample is that th e surface be conducting in order to prevent

PAGE 44

29 surface charging due to interaction of the beam with the sample. In the case of insulating samples a thin carbon film is applied or b eam conditions are used that reduce surface charging. The interaction of th e beam with the sample produces several different species at differing depth within the sample incl uding secondary electr ons, Auger electrons, backscattered electrons, and characteristic x-rays. Sec ondary electrons have energies below 50 eV, and due to this low energy can only escape from the sample if they are produce within a few nm of the surface. As the electron beam is rastered, a detector picks up the secondary electron si gnal. This signal is fed in to a cathode ray tube that is scanned at the same rate as the be am, producing an image (Figure 3-13). The SEM used in this research is a JO EL 6400. The technique gives a qualitative analysis of the surface, indicating the overa ll surface morphology of the film. This is important for future processes in fabri cating capacitors and MOS(MIS)FET’s, since device processing requires annealing, etch ing, and metal deposition, all of which are sensitive to surface morphology. SEM can also be used to look at the surface topography, which should determine how well the surface is covered and see any obvious defects and pin holes. 3.2.6 Auger Electron Spectroscopy: AES Auger electron spectroscopy (AES) was used to determine qualitatively the elements present in the grown dielectrics. Auger electrons are also emitted from the sample during the electron/sample interaction (F igure 3-14). An inci dent electron strikes an inner shell electron of an atom and ejects that electron. An upper shell electron fills the void and energy is given off in the form of an Auger electron. The Auger electron is specific in energy to the element it came from and is thus a ch aracteristic electron to that element. The Auger electrons are collected a nd an elemental analysis of the surface is

PAGE 45

30 obtained. The AES technique can detect elem ents down to the all oy level (~1%) within the top 1.0 nm of the surface. Auger elec trons are produced throughout the interaction volume of the incident electrons, however, because of their low energies, only those produced near or at the surface can escape. From the ratio of the peak heights from each element and published sensitivity factors, an approximate ratio of elements can be determined. A Perkin Elmer 660 AES was used for these measurements. This system is also equipped with an ion gun for creating de pth profile Auger electron spectra. From this, changes in the element ratios perpendicu lar to the interface and the interface itself can be studied. This will help in determin ing if film composition is constant throughout the film and if there is segregation in the film. Also, approximate film thickness can be determined from known standards. 3.2.7 Ellipsometry Ellipsometry is used primarily to determine the thickness of thin dielectric films on highly absorbing substrates but can also be used to determ ine the optical constants of films or substrates. Ellipsometry is base d on measuring the state of polarization of polarized light. When light is reflected fr om a single surface it will generally be reduced in amplitude and shifted in phase. For multiple reflecting surfaces, the various reflecting beams will further interact and give maxima and minima as a function of wavelength or incident angle. Since ellipsometry depends on angle measurements, optical variables can be measured with great precision, being indepe ndent of light intensit y, total reflectance, and detector-amplitude sensitivity. A general sc hematic of the machine is in Figure 3-15. For further explanation of ellipsometry see Appendix A. The ellipsometer used for this experiment is a JA Woollam Variable Angle Spectroscopic Ellipsometer.

PAGE 46

31 3.2.8 Current-Voltage (I-V) measurements Current-voltage measurements were ma de using a Hewlett Packard Model 4145. In these measurements, the current is set to an upper and lower limit, typically 5A to10A and the voltage is swept from nega tive to positive. The voltage range is increased until the forward and reverse breakdowns are reached. The current limit is then set to 1mA and the voltage is measured again. The voltage at 1 mA is divided by the dielectric thickness and the brea kdown field strength is obtained. This is one parameter used in defining the quality of the dielectric film. This helps to determine the breakdown field of the dielectric at elev ated temperatures. There is a heated stage with the currentvoltage measurement equipment with a maximum temperature for testing of the 300C. This was used to study the dependence of th e temperature with th e breakdown of the dielectric. Ohmic contacts were made to the silicon substrates using a Pt/Au (300 /1000 ) bilayer structure using electr on beam evaporation. Ohmic contacts made to the gallium nitride were made using a multiplayer st ructure of Ti/Al/Pt/Au with the following thicknesses20 nm Ti /70 nm Al/40 nm Pt /100 nm Au. The contacts on the GaN were annealed for 30 seconds at a temperature of 450 C in a nitrogen ambient. Contacts on the dielectric were Pt/Au and were deposited th rough a shadow mask with varying contact sizes. The most commonly used co ntact sizes were 50 m and 80 m. 3.2.9 Capacitance-Voltage (C-V) measurements Capacitance-voltage measurements were made using a Hewlett Packard Model 4284. Here a bias of 2 to 20 volts is applied across the capacitor and cycled at a selected frequency, typically 100 Hz to 1 MHz, and the resulting capacitance is recorded. From capacitance-voltage plots, the flat band voltage shift, dielectric c onstant, and interface

PAGE 47

32 state density can be calculate d. Calculating the carrier co ncentration of the substrate from the data obtained can test the accuracy of these measurements. This carrier concentration should match the quote d value from the manufacturer.

PAGE 48

33 Figure 3-1. Typical Knudsen effu sion oven (After B.P. Gila 2000). PBN crucible Source material Ta Heater element Thermocouple Atoms or atom clusters

PAGE 49

34 Figure 3-2. Riber MBE used for oxide growth. PLASMA SOURCE SOLID SOURCES RHEED Gun Load/lock Buffer chamber

PAGE 50

35 Figure 3-3. WaveMat 610 ECR plasma head. Source gas inlet Active species Permanent magnets Tuning antenna Cooling gas inlet and microwave power lead

PAGE 51

36 Figure 3.4. Schematic of the Oxford RF plasma sour ce (After MDP21S Operati ng and Service Manual 198973). Tuning Knob Retainin g Knob Location Button Rf Tunin g Box O p tical Filte r Photodiode S a pp hire W indow RF Coupling Flange Aperture Retaining Ring PBN Apeture Plate Discharge Tube RF Coils Water Cooled

PAGE 52

37 Figure 3-5. AFM images of as received A) MOCVD GaN from Epitronics, and B) as received MBE GaN from SVT. A B

PAGE 53

38 A B Figure 3-6. RHEED images showing A) the UV-ozone treated surface of MOCVD grown GaN and B) the buffered oxide etched UV-ozone surface of GaN.

PAGE 54

39 A B Figure 3-7. RHEED photos indicat ing a (1x3) pattern. The to p photo is along the <11-20> crystal direction and the bo ttom photo is along the <1-100> crystal direction.

PAGE 55

40 A B C Figure 3-8. RHEED photos s howing A) an amorphous diffraction pattern, B) a polycrystalline diffraction pattern, and C) a single crystal diffraction pattern.

PAGE 56

41 Figure 3-9. TEM column. Electron Source Condenser Lens Specimen Objective Lens Back Focal Plane First Intermediate Image Intermediate Lens Second Intermediate Image Projector Lens Image

PAGE 57

42 Figure 3-10. The relation betw een the lattice parameter a nd Bragg angle for film and substrate. Film Substrate dsub < dfilm fil m fil m sub> fil m sub sub Incident x-rays Diffracted x-rays

PAGE 58

43 A piezoelectric scanner which moves the tip over the sample (or the sample under the tip) in a raster pattern A feedback system to control the vertical position of the tip A sharp tip A way of sensing the vertical position of the tip A computer system that drives the scanner, measures data and converts the data into an image. A course positioning system to bring to tip into the general vicinity of the sample. Sample Figure 3-11. Atomic force micros cope (after K.K. Harris 2000).

PAGE 59

44 Figure 3-12. Schematic of SEM column. Spot Size Objective Lens Objective Aperture Intermediate Beam Size Condenser Lens Initial Beam Size Electron Source Optic Axis Specimen

PAGE 60

45 Figure 3-13. SEM operation. Electron beam is rastered over the sample producing secondary electrons (a fter S.M. Donovan 1999). Cathode ray tube Screen Electron source Detector giving modulating signal Deflection coils DEFLECTION Secondary electron signal Sample

PAGE 61

46 Figure 3-14. The penetration depth and intera ction of an electron beam in a material. Notice that Auger only escape the top 1.0 nm (after Goldstein).

PAGE 62

47 Figure 3-15. Ellipsometer schematic (After D.K Schroder74) Light Source Analyzer Compensator Polarizer Detector Unpolarized Linearly Polarized Elliptically Polarized Linearly Polarized Extinguished Sample

PAGE 63

48 CHAPTER 4 MAGNESIUM OXIDE: RESULTS AND DISCUSSION Magnesium oxide was chosen as a possible dielectric material for GaN because of its high melting point (2850C), large band gap (7.3 eV), large dielect ric constant (9.8) and good band offsets with GaN. The structur e of MgO is the NaCl crystal structure75 (Figure 4-1). This is a cubi c structure with a lattice constant of 4.20 . The symmetry alignment for MgO:GaN is the MgO (111) and the GaN (0001). The cation spacing for the MgO (111) plane is 2.97 and for the GaN (0001) plane is 3.19 which gives a mismatch of -6.9% (Figure 4.2). 4.1 Effect of Oxygen Plasma Source Many factors affect the growth of Mg O such as oxygen source, oxygen pressure, substrate temperature, and pl asma power. Initial work was carried out using an ECR plasma head for the oxygen source. Sample s grown using a substrate temperature of 350 C and a cell temperature, TMg, of 340 C, showed spotty RHEE D diffraction patterns (Figure 4-3). As seen in this series of im ages, the pattern changed from streaky pattern, indicative of the single crystal GaN surface, to a broken line pattern indicative of a roughened surface. The surface diffraction also showed 6-fold symmetry indicating that the MgO grew with the (111) plane parallel to the GaN (0001) basal plane. During growth the pattern changed slowly until at th e end of 120 minutes of growth, the pattern was highly three-dimensional and showed arcs associated with a polycrystalline pattern. This indicates that the initial layers were single crystal and that the microstructure gradually changed to polycry stalline as more layers were deposited. Using a

PAGE 64

49 profilometer, the film thickness was measured to be 150 nm, yielding a growth rate of ~1.25 nm/min. From XRD results, the MgO film was found to be cubic and oriented with the (111) direction normal to the surface. Needle-like structures were observed on the surface by SEM (Figure 4-4) and the sa mple surface had an AFM root-mean-square (RMS) roughness of 4.07 nm (Figure 4-5). Th ese needle-like feat ures were found to decrease with increasing film thickness. Samples grown at a substrate temperature of 100 C using the ECR plasma head were significantly different from those grown at 350 C. Initiating growth on the same GaN surface, the films remained single crys tal for 1-2 minutes (~2.5-5.0 nm), then a polycrystalline RHEED pattern began to form, Figure 4-6. This pattern remains for the duration of the growth. Th e polycrystalline RHEED pattern shows six-fold rotational symmetry, indicating that the film is textured towards the <111>. This polycrystalline pattern differs from the pattern obtained at 350 C, in that the arcs are more pronounced and the spots are not visible. This suggests a less textured film with a smaller grain size. From an etch step measurement, a growth rate of 2.5 nm/min was calculated. The sample surface has an AFM RMS roughness value of 1 .26 nm and a reduced number of needlelike features. These needle-like features ar e not visible on films w ith a thickness of 100 nm or more. From XRD measurements, the polycrystalline MgO films were confirmed to be highly oriented to ward the (111) direction. Diodes were fabricated from these materials to measure the electrical properties of the dielectric film and the dielectric/GaN interface. Ohmic windows were created by etching the MgO with dilute phosphoric aci d. Ohmic contacts were made by e-beam evaporating Ti/Al/Pt/Au. The gate metal wa s Pt/Au with contact sizes ranging from 100-

PAGE 65

50 50 m. Because of the leaky nature of the single crystal MgO grown at temperatures of 350 C, no C-V results could be measured. The leakage in these films could be due to the needle-like microstructure. For th e polycrystalline MgO grown at 100 C, a forward breakdown field of 2.3 MV/cm was measured at a current density of 5.1 mA/cm2. From C-V measurements a corresponding interface state density of 4 x 1011 cm-2eV-1 was calculated for the diode using the Terman me thod. It is clear that the different microstructure obtained at the 100 C is critical to improving the electrical behavior of the MgO dielectric. The combination the single crystalline interface a nd the polycrystalline material on top appears to produce a low inte rface trap density while at the same time eliminating the shorting paths obtained in th e films which are mostly single crystal but highly defective. When an RF plasma head was used as the oxygen source for MgO growth the materials properties improved further. For samples grown under the same conditions but with an RF plasma oxygen source, the need le-like microstructure was not observed. SEM and AFM also show a smoother surface, Figure 4-7. For these reasons, all further growths used the RF plasma head for th e oxygen source and the ECR plasma head was removed from the system. 4.2 Effects of Oxygen Pressure To study the effect of oxygen pressure on the properties of the growth of MgO, several different pressures were investigated1x10-5, 3x10-5, 7x10-5, and 1x10-4 Torr as measured by the beam flux monitor. The MgO flux was held constant. The higher oxygen pressures produced a decrease in the gr owth rate (Figure 4.9). From AES, the films were shown to contain only Mg a nd oxygen, with the Mg/O decreasing from 0.72 to 0.63 as the pressure was increased, (Figur e 4.8). By comparison, the ratio of peak

PAGE 66

51 heights of a standard single crystalline sample of MgO was measured and showed the ratio to be 0.60. The reduction in growth ra tes at higher pressures may be an indication of a reduction in the concentra tion of reactive oxygen species at the surface or of site blocking due to the higher c oncentration of oxygen incident on the surface. Both cases would result in a reduction in the Mg sticking coefficient. The lower Mg/O ratio at the higher pressures would seem to favor the site blocking explanation. AFM analysis indicated that as the pressure was increas ed the surface morphology became smoother, as evidenced by the decrease in RMS roughness from 0.998 nm at a pressure of 1x10-5 Torr to 0.247 nm at 1x10-4 Torr, (Figure 4-10). All of the films appeared smooth when examined by SEM. XTEM of the MgO grown at the lowest pressure showed that the initial 40 monolayers were epitaxial, with the remainder of the layer appearing to be finegrained polycrystalline, (Figur e 4-11). The precise microstr uctures of the films grown at higher pressure are not yet known. It is quite likely that given the superior morphology, these films retain their single crystal nature for a greater percentage of their thickness before becoming poly-crystalline. From structural and compositional analys is it would appear that the higher pressures are beneficial to the growth of MgO layers. However, electrical characterization of the MgO/GaN diodes s uggests the opposite, (Table 4-1). The breakdown field, Vbd, and the interfa ce state density, Dit, improve with decreasing pressure. Ironically, the reduction in the dielec tric strength may be due to the superior but not perfect microstructure in the films grow n at higher pressures. Previous work with other dielectrics such as Gd2O3 49 has shown that if the layer does not contain a substantial polycrystalline region, then the breakdown fi eld will be substantially lowered due to

PAGE 67

52 leakage through the defects, which propagate through the layer. The presence of a nanocrystalline layer on top of the single crys tal material at the interface appears to improve the breakdown strength of the layer in spite of the presence of numerous grain boundaries. The effect of the pressure on the interf ace and the bulk charge densities suggests that the electrical behavior of the layer is enhanced by the presence of higher ion energy species at the surface. Since the total power to the plasma remains constant, increasing the oxygen flow will decrease the average energy per ion, and possibly decrease the concentration of ionized oxygen species as well Studies with ECR plasmas suggest that the average ion energy is a critical paramete r. ECR plasmas typically exhibit very low ion energies. MgO films grown using an E CR plasma with similar oxygen pressures to those in the RF grown films exhibit breakdow n fields which are up to four times lower than those obtained with the RF plasmas. As the ion energy is increased, damage of the interface will eventually become a factor and begin to increase the density of interface states. Clearly, however, this does not occur at the standard pressures and powers used in this study, making the RF plasmas the opt imum choice for the deposition of MgO dielectric on GaN. 4.3 Effect of Substrate Temperature For MgO samples grown with the RF plasma at 300 C, XRD shows a sharper peak, which indicates that the film grow n at a substrate temperature of 300 C is more crystalline than that grown at a substrate temperature of 100 C, (Figure 4-12). AFM of films grown at 300 C are rougher than films grown at 100 C, (Figure 4-13). The RMS roughness of the 300 C film is 1.998 nm while the 100 C grown samples have RMS roughnesses of 1.334 nm. AES shows that the Mg /O ratio of the films deposited at 100 C

PAGE 68

53 is 0.75 while the 300 C grown samples have an Mg/O ratio of 0.66. Single crystal MgO has a Mg/O ratio of 0.60 when investigated in the Auger system. This shows more incorporation of oxygen at a substrate temperature of 300 C under the same oxygen flux and Mg flux. TEM of MgO grown at 300 C (Figure 4-14) shows that the films is still epitaxial and is single crystalline for a grea ter thickness before rotation as compared to MgO grown at 100 C, (Figure 4-11). 4.4 Scandium Oxide Capping Layer Despite the positive aspects of MgO, it has been found to be environmentally unstable as shown by SEM, (Figure 4-15). The MgO films degrade over time in atmosphere due to the presence of water vapor.76,77 From device processing, it does not appear that the MgO under the metal contac ts degrades, only the areas exposed to atmosphere, (Figure 4-16). Capping layers of scandium oxide of various thicknesses have been tried in order to prevent this degradation. AFM shows that the capping layer smoothes the surface for both MgO grown at 100 C and 300 C. The RMS for MgO grown at 300 C goes from 1.921 nm uncapped to 0.337 nm with a Sc2O3 cap of 20 nm, (Figure 4-17). For the MgO grown at 100 C the RMS goes from 1.539 nm uncapped to 0.869 nm with a 20 nm Sc2O3 cap. The effect of the ca pping layer on the environmental stability of MgO is discussed further in Chapter 7.

PAGE 69

54 Table 4-1. Electrical characteri zation of MgO/GaN diodes. VBD was the applied voltage, which produced a leakage current of 1 mA/cm2. Dit was the defect value at 0.4 eV below the conduction band calculated using the Terman method. The diode grown at 1x10-4 Torr was too leaky to be measured. Oxygen Pressure (Torr) VBD (MV/cm) Dit (eV-1cm-2) 1x10-5 4.4 3.4x1011 3x10-5 4 7.1x1011 7x10-5 1.2 1.8x1012

PAGE 70

55 Figure 4-1. The MgO structure (from Cullity 197875)

PAGE 71

56 (111) (0002) FCC HCP Figure 4-2. Illustration show ing the symmetry between the (111) NaCl plane and the (0002) wurtzite plane. The spacing between atoms marked A is 2.97 for NaCl MgO and 3.19 for wurtzite GaN. (from Cullity 1978)

PAGE 72

57 Figure 4-3. RHEED images i ndicating A) GaN at 350C befo re growth, B) MgO after 1 minute of growth, and C) MgO after 120 minutes of growth at 350C. A) B) C)

PAGE 73

58 Figure 4-4. SEM image of the MgO surface grown at 350C with ECR plasma source. Scale bar is 5m.

PAGE 74

59 A) B) Figure 4-5. AFM images of A) MgO gr own at 350C (4.07 nm RMS) and B) MgO grown at 100C (1.26 nm RMS). Both films are grown with the ECR oxygen plasma source. Images are 5 m per side.

PAGE 75

60 Figure 4-6. RHEED images indicating MgO af ter A) 1 minute of growth, and B) after 90 minutes of growth at 100C. A) B)

PAGE 76

61 Figure 4-7. SEM, above, and AFM, below, of MgO grown at 100 C with RF plasma source, RMS = 1.334 nm.

PAGE 77

62 Figure 4-8. Mg/O ratio as determined by AES as a function of oxygen pressure. 1x10-53x10-57x10-5 0.60 0.62 0.64 0.66 0.68 0.70 0.72 0.74 0.76 0.78 Mg/O RatioOxygen Pressure (torr)

PAGE 78

63 0 50 100 150 200 250 1x10-53x10-57x10-51x10-4 Ox yg en Pressure ( torr ) Growth Rate (nm/hr) Figure 4-9. Dependance of gr owth rate on oxygen pressure

PAGE 79

64 Figure 4-10. AFM scans of MgO grown at 100 C using an oxygen pressure of A) 1x10-5 Torr, or B) 1x10-4 Torr. The RMS roughnesses were 0.998 nm and 0.247 nm respectively. A) B)

PAGE 80

65 Figure 4-11. SEM and TEM of MgO. An SEM image (10,000x) of the MgO layer grown at 1x10-5 Torr is shown above and an XTEM image of the same layer is shown below. GaN MgO GaN Interface M g O 5 nm

PAGE 81

66 74767880 counts (arb)2(degrees) Top MgO grown at 300C Bottom MgO grown at 100C Figure 4-12. XRD shows that the MgO film grown at 300 C has a much sharper peak than the MgO film grown at 100 C.

PAGE 82

67 Figure 4-13. AFM of MgO A) grown at 100C with and RMS roughness of 1.334 nm, and B) grown at 300C with an RMS roughness of 1.998 nm. A) B)

PAGE 83

68 Figure 4-14. HRTEM of MgO grown at 300 C.

PAGE 84

69 Figure 4-15. SEM of degraded MgO film. 200 m 5 m

PAGE 85

70 Figure 4-16. I-V and C-V of MgO/n-GaN diodes, aged15 weeks with and without processing -40-2002040 -2.0x10-80.0 2.0x10-84.0x10-86.0x10-88.0x10-81.0x10-7 as-grown aged w/gate aged w/o gate current (A)voltage (V)-4-3-2-10123 1.00E-012 2.00E-012 3.00E-012 4.00E-012 5.00E-012 capacitance (F)voltage (V) as-grown aged w/gate aged w/o gate

PAGE 86

71 Figure 4-17. AFM of MgO with and without a capping layer. A) MgO without a cap, RMS = 1.921 nm, B) MgO with a 20 nm Sc2O3 cap, RMS = 0.337 nm. The MgO is grown at 300 C. The Sc2O3 cap is grown at 100 C. A) B)

PAGE 87

72 CHAPTER 5 MAGNESIUM CALCIUM OXIDE: RESULTS AND DISCUSSION It is expected that by latti ce matching the oxide to the GaN, the interfacial trap density can be reduced. One method of accompli shing this is to alloy the MgO with an oxide of larger lattice constant. Calcium oxi de was chosen for this purpose because CaO is also a rock salt dielectric, (Figure 5-1) like MgO, but has a larger lattice constant, 4.779 . The CaO dielectric constant (11.8) and bandgap (7.1eV) are similar to those of MgO. From Vegard’s Law, the optimum ratio of Mg :Ca should be 1:1 in order to get material lattice matched to GaN. Though the phase diagram78 shows poor miscibility between MgO and CaO, (Figure 5-2) single-phase th in films on GaAs whose compositions span the entire composition range have been reported using LT-MBE as the deposition method.79 5.1 Growth of CaO First, to prove that CaO can be grown by MBE, starting with conditions similar to those used for the growth of MgO, CaO films were grown on GaN. The solid metal Ca source washeld at temperature of 405 C to achieve a flux of 8x10-8 Torr. The same fluxwas used for the growth of MgO. The RF oxygen plasma was se t at a pressure of 2x10-5 Torr. The substrate temperature was 100 C or 300 C. AES shows that the film was comprised of Ca and O, (Figure 5-3) but XRD showed no evidence of CaO or Ca diffraction peaks. TEM confirmed that the fi lm was poly-crystallin e at the interface but amorphous after 2 nm, (Figure 5-4). The am orphous nature of the bulk of the film explains the absence of peaks in XRD. The CaO film is very sensitive to the electron

PAGE 88

73 beam in the TEM making it difficult to obtain good images. The growth rate of CaO at 100 C was 25.8 A/min, while at 300 C the growth rate was 2.2 nm/min. This is to be expected since an increase in temperature s hould decrease the sticking coefficient. AFM shows no change in RMS roughness of CaO grown at 100 C versus 300 C but stays constant at about 0.667 nm, Figure 5-5 and Fi gure 5-6. Unfortunately, CaO etches in deionized water, which makes processing difficult so no devices were made with CaO. 5.2 Growth of Ternary MgCaO The standard MgO growth conditions th at have produced oxide/GaN interfaces with low Dit consist of a Mg beam equivalent pressure, BEP, of 10x10-8 Torr. The addition of Ca to this beam at a comparable Ca BEP of 8x10-8 Torr, produced an increase of less than 50% in growth rate. This suggest s that the sticking coefficient of the Ca is significantly lower than that of the Mg. This is furthe r confirmed by AES analysis, which shows an Mg/Ca ratio less than that expected for a 50/50 composition film, (Table 5-1). In addition, AES depth pr ofiling analysis show s that the Ca has severely segregated to the surface, (Figure 5-7). This would al so indicate a low Ca st icking coefficient. In spite of the apparent segregation of the Ca, XRD analysis of the MgCaO layer shows no evidence of phase separation, (Figur e 5-8). MgO layers grown under similar conditions typically show primarily a (222) pe ak due to the texturi ng of the film. The MgCaO layer shows no evidence of either th e MgO or the CaO (222) peaks suggesting that phase separation into the tw o binaries has not occurred. Instead there appears to be a shoulder on the GaN (004) peak that is not observed in spectra taken from either GaN substrates or MgO layers grown on GaN. This peak is most likely the (222) peak from the ternary MgCaO. The peak position is shif ted to larger plane spacing relative to the MgO peak as would be expected from the addi tion of Ca, (Figure 5-9). The proximity of

PAGE 89

74 this peak to the GaN (004) peak is encourag ing and suggests that th e addition of Ca may be useful in reducing the lattice mismatch between the dielectric and the GaN. Unfortunately due to the severe segregation, the peak is broadened indicative of a range of compositions present in the film, (Figure 5-10). AFM analysis of the ternar y grown at the highest fl ux rates shows a very rough surface morphology, (Figure 5-11). MgO laye rs grown under similar conditions show smooth morphology with an RMS of ~1 nm suggesting that the Ca addition has dramatically altered th e microstructure. It was thought th at the higher growth rate used for the deposition of the ternary might have caused the rough morphology. To investigate a lower growth rate, th e fluxes were reduced to BEP ~ 5.7x10-8 Torr. This did produce a substantial drop in the growth ra te to ~1 nm/min and was successful in improving the surface morphology. Reducing the growth rate did not, however, suppress the Ca segregation. In fact there appears to be even greater disparity in the Ca surface and interface concentrations. This is not surprising since lo wer growth rates are usually found to enhance segregation. Substrate temperature also affects the growth of MgCaO. At a substrate temperature of 340 C, no film was grown even after 20 minutes of growth time as shown by AES, (Figure 5-12). A slightly lower substrate temperature of 300 C did result in deposition of a film as determined by AES, (Figure 5-13). This film has a reasonable growth rate of 51.6 A/min, and enhanced crysta llinity as shown by XR D, (Figure 5-14). Though for this substrate temperature, AFM does show a slightly rougher surface than that produced with a substrate temperature of 100 C, (Figure 5-15). Due to

PAGE 90

75 improvements in film quality with the increased substrate temperature of 300 C, all further growths were done at a substrate temperature of 300 C. In order to reduce the segregation, a digi tal growth technique was used. This showed superior compositional and structural properties as compared to the continuous growth. To study the effects of growth t echnique on the characteristics of MgCaO deposited on GaN, several films were gr own under different c onditions. Initially conditions were set so that the Mg and Ca fluxes were equal. The timing sequence was 10 seconds Mg followed by 10 seconds Ca ( 10/10) with continuous operation of the oxygen plasma. The intended thickness of each layers was 3 and the substrate temperature was 300 C. The resulting films were Mg-ri ch, which showed that Ca has a lower sticking coefficient than Mg. S ubsequent samples were grown using a progressively higher Ca flux in order to incorporate more Ca into the film and reduce the lattice mismatch. Digital samples were grow n at the same fluxes and oxygen pressures as the continuous samples. AFM shows that th e digital samples have a slightly smoother surface than the continuous samples. Also th e growth rate of the continuous samples is about twice that of the digita l samples, which was expected. Because of a combination of the growth rate and the growth sequence, th e digital samples all showed a much more uniform depth profile in AES especially near the surface, (Figure 5-16). In the continuous samples there is a dip in the oxygen concentration near the surface as well as in the Ca profile. The ratio of oxygen to total metal shows a great er incorporation of oxygen in the digital samples than the continuous samples, (Table 5-2). Powder XRD shows no oxide peaks other than those expected for the MgCaO. The oxide (222) peak is found to shift toward the GaN peak as the amount of Ca incorporated

PAGE 91

76 into the film is increased for both the di gital and the continuous growth sequences, (Figure 5-9). This peak pos ition is the same for either growth method grown under the same fluxes, (Figure 5-17). At the highest Ca concentration, the lattice mismatch has been reduced from –6.5% for MgO to –2.05% for the ternary, (Table 5-3). Highresolution XRD shows a full-width-at-h alf-maximum (FWHM) of 3542 arcseconds, (Figure 5-17). Though this is substan tially higher than the GaN FWHM of 507 arcseconds, it is a significant improvement over the value of 4327 arcseconds measured for MgO grown on GaN using similar growth conditions. Similar to the XRD data, XTEM shows im proved crystal quality in the ternary as compared to the binary, (Figure 5-18). In both cases, the oxide /GaN interface is epitaxial. For the binary, continued growth produces a change in microstructure indicative of a nanocrystalline film. For the ternary, this transition is not observed and the overall defect density appears to be si gnificantly lower. This improvement in structural quality is most likely due to the reduction in mismatch for the ternary relative to the binary MgO. In order to increase the am ount of Ca incorporated into the films and reduce the lattice mismatch further, the fluxes of the Mg and Ca were held constant and the length of time the shutters were open was varied. Th e standard procedure was 10 seconds of Mg followed by 10 seconds of Ca(10/10) with continuous operation of the oxygen plasma. This shutter sequence was continued for the en tire growth time. Keeping the cycle time constant at 20 seconds, shutter times of 8 seconds Mg followed by 12 seconds Ca (8/12), and 5 seconds Mg followed by 15 seconds of Ca (5/15) were tried. All samples began with an Mg layer as the first layer.

PAGE 92

77 HRXRD shows the10/10 sample to have the expected peak positi on to the right of the GaN (004) peak at 73.9 degrees. This corresponds to a composition of 60% Mg and – 1.5 lattice mismatch. For the 8/12 shutter seque nce more Ca is incorporated and the film is slightly Ca rich. The Mg content goe s down to 40.5% Mg and the mismatch is now +0.96%. The 8/12 sample peak shows up as a shoulder on the GaN (004) peak. The 5/15 shutter sequence incorp orates even more Ca and has a peak position of 70.175 degrees. This results in a Mg content of 24% and a lattice mismatch of +3.04%. Figure 5-19 shows the XRD plots of these 3 samples. AFM shows a slight difference in surface roughness for the different shutter sequences, (Figure 5-20). For 10/10 RM S is 0.790 nm, for 8/12 RMS is 1.081 nm, for 5/15 RMS is 0.979 nm. All of these samples have a 20 nm Sc2O3 cap. HRTEM shows the oxide to be crystalline a nd the capping layer to be polyc rystalline for the 10/10 shutter sequence.

PAGE 93

78 Table 5-1. Growth rate and AES data for MgCaO samples. AES data taken from an MgO single crystal is shown for comparison. Mg Beam Equivalent Pressure (Torr) Ca Beam Equivalent Pressure (Torr) Growth Rate (nm/min) Mg/Ca Ratio (Mg+Ca)/O Ratio 5.7x10-8 5.7x10-8 1.0 0.81 0.92 10x10-8 8x10-8 5.3 1.27 0.83 10x10-8 3.4 0.72 0.72 MgO Ref. 0.6

PAGE 94

79 Table 5-2. Growth rate, AFM and AES data for MgCaO grown at 300 C. Growth Method Mg Beam Equivalent Pressure (Torr) Ca Beam Equivalent Pressure (Torr) Growth Rate (nm/min) RMS roughness (nm) O/(Mg+Ca) from AES Continuous 10x10-8 9.0x10-8 5.6 1.931 0.52 Digital 9.6x10-8 8.6x10-8 3 0.712 1.40 Continuous 8.0x10-8 9.0x10-8 5.3 0.774 1.29 Digital 8.0x10-8 9.0x10-8 1.8 0.554 1.81 MgO ref. 12x10-8 --2.4 1.998 0.50

PAGE 95

80 Table 5-3. Composition and mismatch from XRD layer Ccontinuous Ddigital 2-theta% Mg % mismatch to GaN MgO78.59100-6.45 MgCaO-C75.4372.8-3.11 MgCaO-D75.4172.8-3.09 MgCaO-C74.7974.79-2.40 MgCaO-D74.4774.47-2.05 50/50~72.950-0.23 CaO67.37506.86

PAGE 96

81 Figure 5-1. Illustration of the CaO structure (from Cullity 1978) Ca

PAGE 97

82 Figure 5-2. AES shows only Ca a nd O after growth of CaO at 300 C 500100015002000 Ca Ocounts (arb)Kinetic Energy (eV)

PAGE 98

83 Figure 5-3. MgO-CaO phase diagram.78

PAGE 99

84 Figure 5-4. HR TEM of CaO. GaN Interface Polycrystalline CaO Amorphous CaO

PAGE 100

85 Figure 5-5. AFM of CaO grown at 100 C, RMS = 0.667 nm.

PAGE 101

86 Figure 5-6. AFM of CaO grown at 300 C, RMS roughness is 0.667 nm.

PAGE 102

87 Figure 5-7. AES depth prof iling of MgCaO grown using Ca and Mg BEP of 5.7x10-8 and 5.7x10-8 (top) and 8x10-8 and 10x10-8 (bottom). 010203040 0 10000 20000 30000 40000 50000 60000 70000 N Ga O Mg CaCounts (arb. units)Cycles 020406080100 0 5000 10000 15000 20000 25000 30000 35000 40000 45000 50000 N Ga Ca Mg OCounts (arb. units)Cycles

PAGE 103

88 Figure 5-8. XRD of MgCaO s hows no signs of phase separa tion or secondary phases. 3040607080102103104105106 CaO (222) MgO (222) Sapphire (006) GaN (002) MgCaO (222) GaN (004)2(degrees)

PAGE 104

89 Figure 5-9. Powder XRD showing increase in lattice constant with the addition of Ca. 747576777879808182 Sapphire Kb (0012) MgO (222) 2-theta = 78.59 2(degrees) 747576777879808182 Sapphire Kb (0012) MgCaO (222) 2-theta = 75.43 2(degrees) 747576777879808182 Sapphire Kb (0012) MgCaO (222) 2-theta 74.792(degrees)

PAGE 105

90 Figure 5-10. HR-XRD shows the relative po sitions of the MgO (222) and the MgCaO (222) peaks. 68707274767880102103104105106 FWHM = 4327 arcsec FWHM = 3452 arcsec FWHM = 507 arcsec MgO (222) MgCaO (222) GaN (004)counts2-

PAGE 106

91 A) B) Figure 5-11. AFM of MgCaO grown at 100 C using Mg and Ca fluxes of A) 5.7x10-8 or B) 8x10-8. RMS roughnesses for the two samples were 4 nm and 9 nm respectively. MgO grown under simila r conditions shows an RMS roughness of ~1 nm

PAGE 107

92 500100015002000 C N OGacounts (arb)Kinetic Energy (eV) Figure 5-12. AES shows that no Mg (1175 eV) or Ca (294 eV) is present after growth time. Oxygen is due to exposure of surface to oxygen plasma for duration of growth time.

PAGE 108

93 Figure 5-13. AES of MgCaO film gr own at substrate temperature of 300 C. 500100015002000 -800 -600 -400 -200 0 200 400 600 O Ca Mgcounts (arb)Kinetic Energy (eV)

PAGE 109

94 Figure 5-14. XRD shows that the peak from the MgCaO film grow at 300 C is much sharper than the MgCaO film grown at 100 C. 747678 counts (arb)2TopMgCaO at 300C BottomMgCaO at 100C

PAGE 110

95 Figure 5-15. AFM of MgCaO grown at A)100 C, RMS = 1.256 nm, B) 300 C, RMS = 1.931 nm. A) B)

PAGE 111

96 Figure 5-16. AES scans of continuously grow n sample (at left), and digitally grown sample (at right). Both films we re deposited at 300C under the same conditions. 0501001502000 10k 20k 30k 40k 50k 60k 70k 80k Ca Mg N Ga Ocycles 0204060801000 10k 20k 30k 40k 50k 60k Mg Ca O N Gacycle

PAGE 112

97 68707274767880102103104105106 FWHM = 4327 arcsec FWHM = 3452 arcsec FWHM = 507 arcsec MgO (222) MgCaO (222) GaN (004)counts-2697071727374757677102103104105106 GaN (004) MgCaO (222) Digital Bottom Continuous Topcounts-2 Figure 5-17. HR-XRD showing MgCaO and MgO texturing in the (111) direction. Peak positions were the same for the digital and continuous growth techniques under the same growth conditions. The ternary oxide showed a significantly lower FWHM than the binary MgO

PAGE 113

98 Figure 5-18. High resolution XTEM showing th e epitaxial nature of the initial oxide layers grown on GaN. The ternary laye r, shown at left, shows no evidence of crystallite formation while the MgO layer, shown at right, shows clear evidence of rotational relaxation. Ga Interface MgO 5nm GaMgCaO Interface

PAGE 114

99 Figure 5-19. HR XRD showing change in peak position with change in shutter timing. 6870737578 101102103104105106 5/15 70.175 8/12 71.79 10/10 73.9counts (arb)2(degrees)

PAGE 115

100 A) B) C) Figure 5-20. AFM of capped MgCaO at differe nt shutter sequences. A) 10/10 with RMS roughness of 0.970 nm, B) 8/12 with RMS roughness of 1.081 nm, andC) 5/15 with RMS roughness of 0.979 nm.

PAGE 116

101 CHAPTER 6 SCANDIUM MAGNESIUM OXIDE AN D MAGNESIUM SCANDIUM OXIDE: RESULTS AND DISCUSSION Scandium oxide, Sc2O3, has been explored for use as a dielectric for GaN and has many favorable properties. GaN gated diodes using Sc2O3 have shown inversion but the Dit of Sc2O3 is still higher than that seen for GaN diodes using MgO. Unlike MgO, Sc2O3 has the bixbyite structure. The bixbyite stru cture is a face-centered cubic, FCC, array of Sc atoms with 3/4 of tetrahedral sites fille d with oxygen atoms, (Figure 6-1). The bond length mismatch between the GaN(0001) and the Sc2O3(111) is 9%. Scandium oxide is very environmentally stable. Even though Sc2O3 has a much different crystal structure and lattice constant, the ternary MgScO and ScMgO films have much potential. The addition of Sc to MgO should increase the environmental stability of the film. It should also increase the lattice constant of the material which would improve the Dit and hence device performance. Films with a higher percentage of Mg than Sc are referred to as MgScO. Adding Mg to Sc2O3 may also improve the Dit of Sc2O3 by decreasing the lattice constant. The addition of Mg to Sc2O3 should improve confinement on the valence band, which is necessary for CMOS technology, and improve the lattice match. As stated before the reduction in la ttice mismatch reduced D it, increases EBD and will enable novel device structures such as Ga N single crystal grown over diel ectric and 3-D integration. Films with a higher percen tage of Sc than Mg are referred to as ScMgO.

PAGE 117

102 6.1 Scandium Magnesium Oxide To study the effects of growth conditions and composition on the characteristics of ScMgO deposited on GaN, several films were grown under different scandium rich conditions. The composition of the films was varied by using a constant Sc cell temperature and varying the Mg cell temperature from 330 C to 350 C. As the Mg cell temperature was increased, the growth rate increased (Table 6-1) in agreement with previous work with MgO.80 Although the growth rate increased, the RMS roughness remained about the same for a given substrate temperature. All of the samples grown at a substrate temperature of 100 C had an RMS roughness of about 0.6 nm independent of the Mg cell temperature. XRD showed a secondary phase peak at 76.7 degrees for samples with a Mg cell temperature over 330 C. No secondary phase peak was seen at a cell temperature of 330 C. AES shows an increase in the ratio of Mg to the total amount of metal in the samples with increasing cel l temperature and growth rate, which shows that more Mg is being incorporated into th e film. AES also shows that the ratio of oxygen incorporated into the f ilm decreased with increasing growth rate. Depth profile AES showed an increasingly large dip in the amount of O at the surface with increasing Mg, (Figure 6-2). The effect of the substrate temperature, Tsub, on the film was investigated by holding the fluxes constant at TSc = 1190 C and TMg = 350 C, and increasing the substrate temperature from 100 C to 500 C. As the substrate temperature was increased the growth rate remained unchanged, which has been seen in previous work with scandium oxide growth.81 But unlike Sc2O3, as the substrate temperature increased the surface roughness also increased si gnificantly from 0.6 nm at 100 C to 24.63 nm at

PAGE 118

103 500 C, (Figure 6-3). AES showed a decrease in the amount of Mg incorporated into the film, which is consistent with a reduction of the Mg sticking coefficient with increased temperature. Depth profile AES (Figure 6-2) showed a reduction in surface effects, such as the dip in oxygen near the su rface, with increased substrat e temperature. XRD showed an increase in the peak height but no change in position. This sugge sts that little or no Mg is being incorporated at the higher subs trate temperatures a nd thus no improvement in lattice mismatch between the oxide a nd the underlying GaN is being achieved. All of the multi-phase films showed a ma gnesium rich secondary phase peak at 76.7 degrees, which corresponds to a composition of 98%Mg, (Figure 6-4). This peak occurs in the same position regardless of the growth conditions used for deposition, suggesting the solid solub ility limit of MgO in Sc2O3 has been reached. The scandium oxide peak is masked by the GaN K peaks at the same position making an accurate determination of the lattice c onstant of the single phase ScMg O alloy difficult. If the lattice constant follows Vegard’s law then based upon the composition as determined by AES, the lattice mismatch between the ScMgO and the GaN should be reduced from the 9.5% obtained with pure Sc2O3 to ~1.5%. Because of the formation of second phases this lattice matching is not possible but a slight reduction in lattice mismatch is achieved below the composition at wh ich second phases appear. 6.2 Magnesium Scandium Oxide To study the effects of growth conditions and composition on the characteristics of MgScO deposited on GaN, several films were grown under different magnesium rich conditions. The composition of the films was varied by using a constant Mg cell

PAGE 119

104 temperature and varying the Sc cell temperature, TSc, from 1090 C to 1130 C. As the Sc cell temperature was increased, the growth rate increased only slightly Initial work was done on Si to see the e ffect of annealing on the oxide and to determine the composition at various Sc cell temperatures. The Mg/Sc ratio decreased as the TSc increased, from 8.5 to 3.3. Etch rates al so decreased as the Sc cell temperature increased, from 1.32 nm/sec at TSc = 1090 C to 0.3 nm/sec at TSc = 1130 C due to the increase in the amount of Sc in the f ilm. The scandium cell temperature of 1090 C produced a film composition most likely to produce single-phase material on GaN. From this initial work on Si, MgScO was grown on GaN at a Sc cell temperature of 1090 C. Powder XRD (Figure 6-5) showed a peak position of 78.09 degrees, which is closer to the GaN (004) peak than MgO showi ng that Sc has changed the lattice constant of the film. AES confirmed the composition of the capping layer and the oxide film (Figure 6-6). The MgScO film grown on GaN at 100 C was much rougher than that grown on Si, RMS = 26.075 nm, (Figure 6-7). With a protective Sc2O3 cap grown at 100 C, the RMS roughness was decreased to 0. 728 nm, (Figure 6-8). For MgScO grown at 300 C the RMS roughness increased to10.24 nm, (Figure 6.9). The amount of Mg incorporated in the film decreased with incr easing substrate temperature. The sticking coefficient of Sc is not as te mperature dependant as Mg. The Sc2O3 growth rate is relatively independent of growth temperat ure while almost no MgO film grows over a substrate temperature of 350 C. HR-TEM (Figure 6-10) of MgScO grown at 100 C with a Sc2O3 cap shows the oxide/GaN interface to be epitaxial and single crystalline. EDS shows this film to be the expected composition, (Figure 6-11). The Sc2O3 capping layer is polycrystalline, similar to that grown on GaN.

PAGE 120

105 Table 6-1. Dependence of growth rate, RMS roughness, and AES ratio on magnesium cell temperature and substrate temperature. TMg ( C) Tsub ( C) G rate (/min) RMS (nm) Mg/(Sc+M g) O/(Sc+Mg) 330 100 13.33 0.694 0.49 0.81 340 100 25.00 0.580 0.61 0.74 350 100 35.33 0.645 0.71 0.65 350 300 36.67 2.947 0.68 0.68 350 500 37.67 24.63 0.55 0.96

PAGE 121

106 Figure 6-1. The Bixbyite cr ystal structure of scandium oxide. (B.P Gila 2000) Sc

PAGE 122

107 Figure 6-2. AES depth prof iles taken from layers gr own under different growth conditions: TMg = 330 C, Tsub = 100 C (at left); TMg = 350 C, Tsub = 100 C (middle); and TMg = 350 C, Tsub = 300 C (at right). 020406080100120140 N Ga O Sc Mg cycles 020406080100120140 N Ga O Sc Mg cycles 020406080100 N Ga O Sc Mg cycles

PAGE 123

108 A) B) Figure 6-3. AFM surface scans for films grown at A) TMg = 350 C, Tsub = 100 C showing an RMS roughness of 0.645 nm and B) TMg =350 C, Tsub = 300 C showing an RMS of 2.947 nm.

PAGE 124

109 Figure 6-4. XRD of ScMgO grown at TMg =350 C and Tsub = 300 C (at left) and comparison of the secondary peak position of scandium rich ScMgO with magnesium rich MgScO and MgO (at right). All peak positions are normalized with respect to the GaN (004) peak. 20406080100 101102103104105106 ScMgO (222) Sapphire K (0012) Sapphire K (0012) GaN K (004) GaN K (004) Sapphire K (006) GaN K (002) GaN K (002)counts2(degrees) 757677787980 0 200 400 600 800 1000 ScMgO 76.67 MgScO 78.09 MgO (222) 78.47 2(degrees)

PAGE 125

110 Figure 6-5. Powder XRD showing the peak position of MgScO peak. 406080 101102103104105106 GaN Kb (002) GaN Kb (004) MgScO GaN (002) Ka1,2 Sapphire (0012) Ka1,2 GaN (004) Ka1,2 Sapphire K(0012)counts (arb)2(degrees)

PAGE 126

111 Figure 6-6. AES of MgScO with Sc2O3 capping layer. 500100015002000-3k -2k -1k 0 1k 2k 3k O Sc MgdN(E)Kinetic Energy (eV)5101520253035 0.0 20.0k 40.0k 60.0k 80.0k Sc O Mg N Gacounts (arb)cycles

PAGE 127

112 Figure 6-7. AFM of MgScO grown at 100 C, A) uncapped, RMS = 34.868 nm and B) with a Sc2O3 cap, RMS = 1.69 nm. A) B)

PAGE 128

113 Figure 6-8. AFM of MgScO grown at 300 C without capping layer, RMS = 10.345 nm.

PAGE 129

114 Figure 6-9. TEM of MgScO shows ep itaxial growth of MgScO on GaN. GaN MgScO

PAGE 130

115 Figure 6-10. EDS of ScMgO showing that th e composition of the film is MgScO and the cap layer is Sc2O3.

PAGE 131

116 CHAPTER 7 ENVIRONMENTAL AND THERMAL STABILIY 7.1 Environmental Stability Environmental stability is important for th e viability of a gate oxide in production and long-term use. If the oxide reacts ra pidly upon exposure to light, moisture, or atmospheric gases and its properties change, then it is not viable. The environmental stability of the oxides under study was tested under the accelerated aging conditions of 100% humidity and elevated temperature. A diagram of the accelerated aging experiment is shown in Figure 7-1. The stabili ty of the films was m easured as the change of the index of refraction, n, over time. Th e index of refraction was measured using a J.A. Woollam Spectroscopic Ellipsometer as described in Section 3.2.7 and Appendix A. 7.1.1 Magnesium Oxide Despite all the good qualities of MgO, it has been found to be environmentally unstable as shown by SEM, (Figure 4-17). The MgO films degrade over time in atmosphere due to the presence of water vapor. It is believed that the water vapor reacts with the MgO to form magnesium hydroxide, Mg(OH)2.76,77 From device characteristics, it does not appear that th e MgO under the metal contact s degrades, only the areas exposed to atmosphere. Capping layers of scandium oxide of various thicknesses were tried in order to prevent this degradation. Samples with scandium oxide caps of 5 nm, 10 nm, and 20 nm were compared to bare MgO under accelerated agi ng conditions, (Figure 7-1). For the bare, uncapped MgO sample, the index of refraction st arts at 1.71 and after 20 days is 1.54, (Table 7-1). A cappi ng layer of 5 nm greatly improves the

PAGE 132

117 environmental stability of the MgO. The 10 nm and 20 nm capped MgO shows no apparent degradation over the entire duration of the aging st udy. The index of refraction for the 5 nm, 10nm and 20 nm caps remains about 1.64, 1.41, and 1.15, respectively. 7.1.2 Magnesium Calcium Oxide Environmental stability of bare, unca pped MgCaO (60%Mg) was compared to MgCaO with a 20 nm scandium oxide cap. Uncapped MgCaO degraded more rapidly than the uncapped MgO, (Figure 7-3). The uncapped MgO showed rapid decrease in n over time, (Figure 7-2). For the bare, uncappe d sample, the index of refraction starts at 1.62 and after 20 days is 0.89, (Table 7-1). Th e index of refraction for the 20 nm capped sample remained about 1.07. Thus the ca pping layer of scandium oxide improved the stability of MgCaO. 7.1.3 Magnesium Scandium Oxide and Scandium Magnesium Oxide Environmental stability of bare, unca pped MgScO (2%Mg) was compared to MgScO with a 20 nm scandium oxide cap, (F igure 7-4). The uncapped MgScO did show some degradation but at a slower rate than the MgO. For the bare, uncapped sample, the index of refraction starts at 1.77 and after 20 days is 1.59, (T able 7-1). The index of refraction for the 20 nm capped sample remained about 1.31. The addition of scandium to the MgO did improveslightly the environm ental stability of the oxide. The capped MgScO showed no apparent degradation over the duration of the study. Due to the two-phase nature of the Sc MgO films, no aging study was preformed on these oxides. 7.2 Thermal Stability The thermal stability of the gate oxid e is important for both operation and processing. If the oxide degrades upon e xposure to moderate temperatures over an

PAGE 133

118 extended time then the oxide is not su itable due to degradation at operational temperatures. More importantly if the oxide degrades upon annealing at elevated temperatures of up to 1000 C then processing limitations beco me an issue as anneals for Ohmic contact formation and implant activ ation are typically done at elevated temperatures. The oxide films we re annealed in an RTA at 1000 C for 2 minutes to test their thermal stability. X-ra y reflectivity (XRR), Section 3.2.6, was used to measure the interface roughness before and after annealing to determine the thermal stability of the film. Philips Electronics WinGixa softwa re, Version V1.102, was used to model and interpret the results. 7.2.1 Magnesium Oxide Uncapped MgO showed severe degradati on or change in film properties after annealing, (Figure 7-5). The surface and in terface roughness increase d and the density of the oxide layer changed, (Table 7-2) The roughness of the GaN/MgO interface increased from 7.65 to 16.84 . The roughness of the MgO/Air interface increased from 35.4 to 56.33 . The density of the MgO film changed from 2.69 g/cm3 to 3.33 g/cm3. This is likely due to recrystallinzation of the polycrystalline MgO. The capping layer seems to reduce these effects (Figure 76), but it is still not stable under these conditions, (Table 7-2). The MgO/GaN inte rfacial roughness increases about the same amount for the capped sample as the uncapped sample. 7.2.2 Magnesium Calcium Oxide Uncapped MgCaO (60% Mg) also showed degradation after annealing (Figure 77), but less severely than the MgO. The capped MgCaO showed onl y slight differences after annealing, (Figure 7-8). The majority of this difference comes from a change in the capping layer possibly due to recrystallizat ion of the polycrystalline scandium oxide

PAGE 134

119 capping layer. The MgCaO/GaN interface in creases in roughness from 3.24 to 96.0 for the uncapped sample and from 25.58 to 51.76 for the capped sample after annealing, (Table 7-2). There was only a small change in the roughness of the capping layer after annealing. 7.2.3 Magnesium Scandium Oxide and Scandium Magnesium Oxide Uncapped MgScO (2%Mg) showed degradatio n after annealing, (Figure 7-9). The GaN/MgScO interfacial roughness increased from 6.84 to 149 after annealing, (Table 7-2). The capped MgScO showed much less degradation after annealing (Figure 7-10), than the capped MgO and slightly less than the capped MgCaO. The addition of the scandium oxide capping laye r greatly improved the thermal stability of the MgScO. The roughness of the GaN/MgScO interface went from 35.53 to 50.93 after annealing. Due to the two-phase nature of the Sc MgO films no thermal annealing was done on any of these samples.

PAGE 135

120 Table 7-1. Ellipsometry data. Change in index of refraction, at t = 0 days, and t =20days. cap n, t = 0 days n, t= 20 days change in n MgO no cap 1.71 1.5 0.21 5 nm 1.64 1.66 0.02 10 nm 1.41 1.42 0.01 20 nm 1.17 1.14 0.03 MgCaO no cap 1.62 0.89 0.73 20 nm 1.07 1.01 0.06 MgScO no cap 1.77 1.59 0.18 20 nm 1.31 1.37 0.06

PAGE 136

121 Table 7-2. XRR data of as-grown and anneal ed samples. Data modeled using Winixa software. Anneal GaN/OxideOxide/Cap or Air Cap/Air RMS (A) RMS (A) Density (g/cm3) RMS (A) Density (g/cm3) MgO no cap Before 7.65 35.40 2.69 --After 16.84 56.33 3.33 --MgO capped Before 6.69 37.05 3.35 15.06 1.14 After 4.66 54.46 3.25 34.20 1.97 MgCaO no cap Before 7.52 3.24 3.24 --After 14.53 96.00 2.82 --MgCaO capped Before 2.70 25.78 2.62 11.60 2.97 After 3.00 151.76 1.75 22.70 2.97 MgScO no cap Before 6.84 4.20 3.97 --After 149.20 57.63 3.47 --MgScO capped Before 4.24 35.53 3.38 14.32 2.05 After 4.16 50.93 3.73 35.24 1.27

PAGE 137

122 Figure 7-1. Accelerated aging e xperimental set up. The smaller beaker of water inside the larger beaker provides elevated humidity and the hot plate provides elevated temperature. The lid is angled such that condensation will drip down the side opposite of the samples. Th e glass slide provide s protection from random drops of water. The metal block provides heat transfer to the samples while keeping them out of the water that collets on the bottom of the beaker. Hot Plate Small Beaker of Wate r Petri dish Lid Glass Slide Metal Bloc k Samples Large Beaker

PAGE 138

123 0510152025303540 1.0 1.2 1.4 1.6 1.8 refractive index, ndays MgO, uncapped MgO with 50A cap MgO with 100A cap MgO with 200A cap Figure 7-2. Ellipsometry of the degradation of MgO with and without a capping layer.

PAGE 139

124 05101520 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 refractive index, ndays MgCaO no cap MgCaO capped Figure 7-3. Ellipsometry of the degradation of MgCaO with and without capping layer.

PAGE 140

125 05101520 1.00 1.05 1.10 1.15 1.20 1.25 1.30 1.35 1.40 1.45 1.50 1.55 1.60 1.65 1.70 1.75 1.80 refractive index, ndays MgScO no cap MgScO capped Figure 7-4. Ellipsometry of the degradation of MgScO with and without capping layer.

PAGE 141

126 0.51.01.52.02.5 101102103104105106 XRR Counts MgO on GaN unannealed MgO on GaN annealed at 1000 C/2 minutes Due to MgO << GaN Figure 7-5. XRR of MgO uncapped, be fore and after annealing at 1000 C for 2 minutes.

PAGE 142

127 1234 101102103104105106 XRR counts (arb)Capped MgO as grown annealed Figure 7-6. XRR of capped MgO be fore and after annealing at 1000 C for 2 minutes.

PAGE 143

128 1234 100101102103104105106 XRR counts (arb)Uncapped MgCaO as grown annealed Figure 7-7. XRR of MgCaO uncapped be fore and after annealing at 1000 C at 2 minutes.

PAGE 144

129 1234 100101102103104105106 XRR counts (arb)Capped MgCaO as grown annealed Figure 7-8. XRR of MgCaO capped be fore and after annealing at 1000 C for 2 minutes.

PAGE 145

130 1234 100101102103104105106 XRR counts (arb)Uncapped MgScO as grown annealed Figure 7-9. XRR of MgScO uncapped, be fore and after an nealing at 1000 C for 2 minutes.

PAGE 146

131 1234 100101102103104105 XRR couts (arb)Capped MgScO as grown annealed Figure 7-10. XRR of MgScO capped, be fore and after annealing at 1000 C for 2 minutes.

PAGE 147

132 CHAPTER 8 SUMMARY AND FUTURE WORK 8.1 Magnesium Oxide Magnesium oxide was grown by gas sour ce MBE on (0001) GaN using elemental Mg and atomic oxygen supplied from an ECR pl asma source and an RF plasma source. The morphology of the ECR grown MgO wa s needle-like, while the RF grown magnesium oxide had a much smoother mo rphology. In both cases, the magnesium oxide was polycrystalline and highly textured toward the (111) direction. Breakdown fields of 1.0MV/cm for ECR grown MgO a nd 2.3 MV/cm for RF grown were obtained. Interface state densities of 4x1011 MV/cm were measured for the polycrystalline MgO/GaN heterostructure of either plasma source. RF grown MgO was found to be superior to ECR grown oxides and thus all further growths used the RF plasma for the oxygen source. Increasing the oxygen pressure during growth was found to improve the morphology and produced an Mg/O ratio closer to that obtained for single crystal MgO. By contrast, electrical char acterization of MgO/GaN diodes showed the best breakdown field and interface state density at the lower oxy gen pressures. It is believed that the superior electrical behavior at lower oxygen pressures is due to the higher ion energy obtained at the lower pressures. This is in agreement with th e poorer electrical characteristics obtained usi ng ECR oxygen plasmas, which produce significantly lower ion energies than RF plasmas. Unfortuna tely, the MgO films are environmentally and thermally unstable.

PAGE 148

133 8.2 Magnesium Calcium Oxide Magnesium calcium oxide was grown on (0001) GaN by gas-source molecular beam epitaxy. Depth profiling Auger electr on spectroscopy (AES) showed a steep increase in the Ca concentration at the surface relative to the oxide/GaN interface indicative of severe Ca segregation which was e nhanced at lower growth rate. In spite of this segregation, XRD of films deposited at 100C showed no evidence of phase separation, and the addition of Ca did increase the lattice constant of the material thus reducing the bond mismatch to GaN. A digital growth technique was used to s uppress the segregation of the Ca in the film. AES showed an improveme nt in the uniformity of the depth profile of the ternary film, especially near the surface, and enha nced oxygen incorporation in the digitally grown samples as compared to the conti nuously grown samples. HRXRD shows the same (222) peak position for both growth se quences using the same metal fluxes during growth. AFM shows that the su rface of the digital sample is smoother than that of the continuous samples. Thus all evidence s uggests that the digital growth sequence produces higher quality material than a con tinuous sequence. Th e reduction in lattice match enables the growth of an epitaxial si ngle crystalline oxide on GaN, which does not relax into nanocrystallites as film thickness increases. Changing the shutter sequence allows for more exposure of Ca during growth and resulted in increased incorporation of Ca. The standard 10 sec Mg/10 sec Ca shutter sequence produced a bond mismatch of –1.5%. The bond mismatch for the 8 sec Mg/12 sec Ca shutter sequence was +0.96 percent. The 5 sec Mg/15 sec Ca shutter sequence produced a bond mismatch of +3.04%. Thus it appears a lattice matched film would be

PAGE 149

134 produced for a shutter sequence of 9 sec Mg/ 11 sec Ca. Bond mismatch can be varied from tensile to compressive strain. Futu re experiments should include trying the 9/11 shutter sequence. 8.3 Scandium Magnesium Oxide and Magnesium Scandium Oxide Scandium magnesium oxide was grown on (0001) GaN by gas-source molecular beam epitaxy. Increasing the magnesium cell temperature during growth of ScMgO was found to increase the growth rate, have littl e effect on surface roughness, and create a second phase which was magnesium rich at Mg cell temperatures over 330C. Increasing the substrate temperature was found to have little effect on the growth rate and dramatically increased surface roughness. Th e maximum Mg concentration that could be obtained in single-phase materi al was found to be 28% Mg as determined by AES. Magnesium scandium oxide was grown on (0001) GaN by gas-source molecular beam epitaxy. The maximum amount of Sc th at could be incorporated into MgO was ~9%. TEM showed this film to be epitaxial and single crystalline. The incorporation of scandium reduced the amount of strain in th e MgO and produced a film that was single crystal. AFM showed that the MgScO grown at a substrate temperature of 100C was much smoother with the scandium oxide cap. 8.4 Environmental and Thermal Stability MgO has been found to be unstable due to reactions with moisture in the atmosphere. A scandium oxide cap of 5 nm was shown to improve the environmental stability during an experimental accelerated aging experiment. After annealing the MgO film at 1000 C for 2 minutes the film degraded and the interface roughness was

PAGE 150

135 drastically increased, probably due to recrysta llization of the MgO. The scandium oxide cap reduced the degradation of the film. MgCaO was shown to degrade faster than MgO in atmosphere. As with MgO, a scandium oxide cap improved the environmen tal degradation of the oxide. The MgCaO was annealed at 1000 C for 2 minutes. The uncapped MgCaO also degraded upon annealing. The capping layer significantly incr eased the thermal stab ility of the film. The best structural quality and stability we re obtained using the MgCaO with the Sc2O3 cap. Both uncapped and capped MgScO were aged as well at an accelerated rate. The addition of Sc was shown to have no significant effect on rate of the aging process. The capping layer prevented environmental aging as seen with the other two oxides. The thermal stability of the MgScO was better than that of the MgO. The capping layer improved this further. Further experiments are needed to determ ine the temperatures at which each of these oxide films degrade to determine a maximum processing temperature. Processing issues related to the difference s in etch rates between the Sc2O3 cap and the MgCaO must be addressed so that the el ectrical effects of the cha nge in bond mismatch can be determined.

PAGE 151

136 APPENDIX A ELLIPSOMETRY Ellipsometry is a very sensitive surface a nd thin film measurement technique that uses polarized light. It deri ves its sensitivity, which is greater than a simple reflection measurement, from the determination of the re lative phase change in a beam of reflected polarized light. Also, ellipsometry is more accurate than intensity reflectance because the absolute intensity of the reflected light doe s not have to be measured. For rotating analyzer ellipsometers, the detectors used only have to have a lin ear intensity response and do not need to be calibrated in an abso lute sense, such that no special reference samples need to be maintained. Many simple samples may be characterized by ellipsometric measurements at a single wavelength. The use of spectroscopic measurements using multiple wavelengths provides much more information about the samp le and also provides the ability to acquire data in spectral regions where the measur ed data are most sensitive to the model parameters which are to be determined. In many cases the dispersion of the optical constants of a given materials is known, or th e optical constants may be parameterized in such a way as to enforce some type of dispersion on the optical constants (Cauchy, Lorentz, and parametric semiconductor models for example). In this case the acquisition of spectroscopic data allows the user to take advantage of this knowledge to obtain more information from the analysis of the ellipso metric data than would be possible when analyzing data from a single wave length.

PAGE 152

137 Ellipsometry measures the change in the pol arization state of light reflected from the surface of a sample. Fundamentally, ellipsometry refers to the measurement of the polarization state of a light beam. However, ellipsometric measurements are usually performed in order to describe an “optical” sy stem that modifies the polarization state of a beam of light. For thin film sample an alysis, the “optical system” is simply the reflection of light from the sample. The measured values are expressed as psi ( ) and delta ( ). These values are related to the ratio of Fres nel reflection coefficients Rp and Rs for pand s-polarized light, respectively, Equation A-1. i s pe R R tan (A-1) Because ellipsometry measures the ratio of two values it can be highly accurate and very reproducible. Because the ratio is a complex number, it is also contains “phase” information ( ), which makes the measurements very sensitive. Ellipsometry can be used to determine thin film thicknesses, thin film optical constants, and in many case both for the same film. For many samples, ellipsometry is sensitive to film thickness on a submonolyer le vel. Ellipsometry has also proven to be the premier technique for determining optical c onstants in the near-UV, visible, and nearIR wavelength ranges. Ellipsometry uses polarized light for measurements. The first step in understanding polarized light is understanding an electromagne tic (EM) plane wave, which is a solution of Maxwell’s equations for electromagnetic fields. In this section we review the properties of the plane wave, the Jones matr ix calculus for describing polarization state and the various types of light polarization encountered in ellips ometric experiments.

PAGE 153

138 A.1 Maxwell’s Equations and the EM Plan Wave Maxwell’s equations from a non-conducti ng, non-dispersive medium appear as follows: 0 E (A-2) 0 B (A-3) 0 1 t B c E (A-4) 0 t E c B (A-5) Where E and B are the electric field and magne tic fields, c is the speed of light, and and are the permeability and the dielectric function, respectively. Any propagating light beam must obey these equations. These equations can be combined to yield th e wave equations for the electric field: 0 12 2 2 2 t E E (A-6) where the optical impedance is defined as c (A-7) A solution of the electric field wave e quation is the electromagnetic plane wave: t i r q n i E t r Eo exp ~ 2 exp , (A-8) Where q is a unit vector along the di rection of wave propagation, n is the complex index of refraction n + ik, is the wavelength of the light in vacuum, is the angular frequency of the wave, and Eo is a complex vector constant specifying the amplitude and

PAGE 154

139 polarization state of the wave. Such a wave propagating in a medium with no absorption (k=0), (Figure A-1). The E-field, B-field, and the directio n of propagation are all orthogonal with respect to each other. Because of the relati onship between the fields only the E-field and the direction of propagation are required to co mpletely define a plane wave. Polarization states are usually in terms of the direction and phase of the E-field vector, only. If the imaginary part of the complex index of refraction, the extinction coefficient, is non-zero, the amplitude of the wave will decay exponentially as it propagates according to the following expression E exp (-2 kz/ ) (A-8) Where k is the extinction coefficient, z is the distance of propagation in length units, and lambda is the wavelength, in the sa me length units as z. The wave will then decay to 1/e of its original amp litude after propagating a distance, Dp, known as the penetration depth, given by Dp = /2 k (A-9) This is an important concept, as many materials exhibit large values of the extinction coefficient such that the light beam may penetrate a few tens of nm or less into the material. We cannot expect to gain a ny information from a film or interface unless the light beam used in the ellipsometric experi ment penetrates to th e film on interface that we are studying, and is also able to propagate back out of the sample after reflection from the interface. For this reason it is usually not possible to measure the thickness of metal films of more than 50nm in thic kness, as very little of the incident light beam reaches the bottom of the metal films and gets back out of the top to reach the detector.

PAGE 155

140 A.2 Jones Vectors and Matrices One of Maxwell’s equation states that the divergence of the displacement field must equal 4 times the local charge density D = 4 (A-10) Where D is the displacement field and is the charge density. In the absence of space charges, is zero and if we assume the materi al to be isotropic the above equation reduces to E = 0 (A-11) Where E is the electric field. This requi res that he components of the polarization vector Eo must lie in the plane perpendicular to the direction of beam propagation, again assuming the material in which the beam is propa gating is isotropic. In this case, we can describe the polarization states of any b eam by specifying its components along any two orthogonal axes in the plane perpendicular to the direction of the beam propagation. In ellipsometric experiments it is common to use the pand sdi rections as the two orthogonal basis vectors use to express beam polarization states, (Figure A2-2). The pdirection is defined as lying in the plane of the incidence, defined as the plane containing the incident and reflected beams and the vect or normal tot eh sample surface. The sdirection (from Senkrecht, German for perp endicular) lies perpendicular to the pdirection such that the p-dire ction, s-direction, and the di rection of the propagation (in that order) define a right ha nd Cartesian coordinate system. We can now express any totally polarized beam by specifying the components of the electric field of the beam along the pa nd s-directions. The components are complex

PAGE 156

141 numbers, which may be conveniently written as two-component vector, known as the Jones vector s pE E E (A-12) We may now express the action of any co mponent or sample upon the polarization state of the propagating beam very simply by means of a 2x3 transfer matrix, known as the Jones matrix. The diagonal elements of the Jones matrix represent the change of amplitude and phase of the pand scom ponents of the beam, while the off-diagonal elements describe the transfer of energy fr om the p-component to the s-component, and vice versa. The Jones matrix calculus provides a ve ry convenient and powerful means of describing optical systems, such as ellipsometers, and will be used throughout the remainder of this section. A.3 Light Polarization States If one looks at the E-filed vector of linear ly polarized light in a plane perpendicular to the direction of propagation (x-y plane), one sees that the E-field lies in one line at all time. The tip of the E-fled vector traces out a line segment as a function of time. This linear polarization can be described as two co mponent waves propagati ng in phase, in the same direction, but with ort hogonal E-fields in the x and y directions. The polarization state is defined with respect to some physical frame of reference. The orientation of the total E-filed with respect to th e coordinate system is defined by the relative amplitudes of the Ex and Ey fields. The Jones vector representa tion of linearly polar ized light is i s i pe E e E, (A-13)

PAGE 157

142 where is a phase angle, which must be identified for both components. For special cases of pand s-polarized light the normalized Jones vectors have the particularly simple forms (up to and arbitrary multiplicative complex constant) 0 1 p-polarized and 1 0 s-polarized. (A-14) If the Ex and Ey fields are equal in magnitude but 90 out of phase, then the filed vector traces out a circle as a function of time. There are two types of circular polarization. If, looking into the propagating beam, the el ectrical field vector is precessing counterclockwise around the circle th e beam is left-circularly polarized, and normally the Jones vector for the beam is i1 2 1 (A-15) If the electric field vector is precessi ng clockwise around the circle, the beam is right-circularly polarized, and the corresponding Jones vector for the beam is i1 2 1 (A-16) A line segment and a circle are special types of ellipses; therefore, linearly and circularly polarized lights are just special cases of elliptically polarized light. Ellipsometry measures and by determining the polarization of ellipse of the probe beam, hence its name. A.4 Single Films on Thick Substrates The case of polarized light reflection fr om a single film on an optically thick substrate may be solved in a number of ways The most instructive is an analytical solution based on the summation of all reflect ed beam component such as the multiple

PAGE 158

143 reflections of the incident light beam which occur in the film for the case of a single film on an optically thick substrate, (Figure A-3). There are in principle an infinite number of reflected and transmitted beams, however, the splitting of the beam, into reflected and transmitted component at each reflection quickly reduce the amplitude if the subsequent reflections such that eventually the reflected and transmitted beams die out. Also, any absorption in the film will attenuate the beams as they propagate. We now take advantage of the fact that our previous derivation for the Fresnel reflection and transmission coefficients of a bulk system is locally valid for the reflection and transmission of a beam from any interface. Thus, we calculate Fresnel coefficients, which are functions for the ambient and film indices of refraction and the angle of incidence, which are valid for any beam incident on the air/film interface or the film/substrate interface. The terminology for this calculation will be as follows. First the incident beam is denoted Einc, which will represent ether a por s-polarized wave of unit amplitude. The reflected beams will be labeled refl nE, where n denotes the nth reflection and refl denotes reflected beams. Transmitted beams will be labeled trans nE, where n denotes the nth transmitted beam and trans denotes transmitted beams. Quantities relating to the ambient medium will have the subscript 0, quantities relating to the films will have the subscript 1, and quantities relating to the substrate will have the subscript 2. This notation is used because it is easily extended to any number of layers. Fresnel reflection and transmission coefficients for both interfaces will be given two numbers for a subscript, where the fi rst number denotes the region from which the

PAGE 159

144 beam is incident on the interface, and the second number denotes the region on the other side of the interface. Our calculation will be performed independently for the pand s-polarized incident beams, as any incident beams of arbitrar y polarization can be described as a linear combination of sand p-polari zation states. As a result, we do not need to label the Fresnel coefficients with p or s. For example, r01 denotes the complex Fresnel reflection coefficient for a beam incident upon the film, region 1, and from the ambient medium (region 0), while t12 denotes the Fresnel transmission coefficient for a beam incident on the film/substrate interface from the film side. In order to complete the model we must also be able to connect the propagating waves between the top and bottom interfaces of the film. A propagating wave will have the form given in Equation A-8, and it is a si mple matter to show that propagation of a wave across the film, in either directi on, yields a wave of the following form. 2 exp i E Ebefore after (A-17) where is the phase thickness or optical thickness of the film for the given wavelength and angle of incidence, given by 0 2 2 0 2 1 1 1sin 2 cos 2 n n d d n (A-18) in which d is the film thickness and is the wavelength, in the same units as the film thickness. It is now a simple matter to writ down the expressions for the successive reflected beams: incident reflE r E01 1 (A-19) incident j rE e r t t E2 12 01 10 2 (A-20)

PAGE 160

145 incident j rE e r r t t E4 2 12 10 01 10 3 (A-21) incident j rE e r r t t E6 3 12 2 10 01 10 4 (A-22) and so on. We can now identify the following general form for the nth reflected beams: ) 2 2 ( 2 12 2 10 01 10 n j n n r ne r r t t E (A-23) Now we sum the reflected beams using th is functional form, as shown below. incident n n j n n j r totalE e r r e t t r E 2 2 2 12 2 10 2 01 10 01 (A-24) We next take advantage of the following identities, 10 01r r (A-25) 2 01 01 101r t t (A-26) Inserting A-25 and A-26 into A-24 and eval uating the resulting convergent series, we find the following expression for the total reflected beam. incdnet j j r totalE e r r e r r E 2 12 01 2 12 011 (A-27) This equation is valid for pand spolarized input beams, provided the corresponding por s-polarized Fresnel re flection coefficient for the interfaces are employed. We now define the pseudo-Fresne l reflection coefficients for any arbitrary sample in terms of the incident, reflected, and transmitted beams. If we use p-polarized Fresnel coefficients for the evaluation of A-27, we can find the p-polarized pseudoFresnel reflection coefficient Rp as follows 2 12 01 2 12 011j j incident r total pe r r e r r E E R (A-28)

PAGE 161

146 with a similar equation holding for the s-pol arized case. These coefficients are easily identified from A-27. We may now calculate and from a broader definition of the ellipsometric parameters, valid for any sample exhibiting pseudo-Fresnel pand spolarized reflection coefficients Rp and Rs respectively: s p iR R e tan (A-29) The summation of multiple transmitted beam s may also be performed in a similar manner to obtain the polarization state of the transmitted beam. There are many mathematical formalisms th at may be used to calculate the pseudoreflection coefficient and/or and for multi-layered structures but that is too complex to go into at this time.

PAGE 162

147 Figure A-1. An electromagnetic plane wave. z x y Electric Field, E(z,t) Magnetic Field, B(z,t)

PAGE 163

148 Figure A-2. Geometry of an ellipsometric experiment, showing pand s-directions. E p-plane s-plane Linearly Polarized Light Reflection off sample Plane of Incidence E p-plane s-plane Elliptically polarized light

PAGE 164

149 Figure A-3. Multiple reflected and transmitted beams for a single film on an optically thick substrate. Film Substrate Incident Beam R1 R2 R3 R4 T3 T2 T1

PAGE 165

150 APPENDIX B TEM SAMPLE PREPARATION There are many ways to make TEM cross-sectional samples. For materials as brittle as GaN on sapphire the main technique has been the old fashioned method of hand grinding and polishing to elec tron transparency. A newer me thod that is being used to make cross-sectional TEM samples for many materials is focused ion beam, FIB, milling followed by lift-out. These two sample pr eparation methods ar e described below. B.1 Old-fashioned Hand GrindingSample preparation begins by cleaving thin st rips of the material, about 1 cm long. These strips are glued together using G-1 epoxy from Gatan, film side to film side. Other strips of sapphire are glued to add support, (Figure B-1). Tweezers are used to hold the strips together during th e curing process to ensure thin glue lines. This stack of strips are polished using 320 grit SiC paper to create two parallel side s, perpendicular to the glue line. This is further polished usi ng 600 grit SiC paper and a final polish with 3-5 micron diamond slurry on 800 grit SiC paper. The final thickness of the polished piece is 20 to 30 microns, (Figure B-2). This piece is glued to a nickel support grid, approximately 100mm thick, using the Gatan G-1 glue, (Figure B-3). The sample is thinned further by using a Gatan Duo ion mill. The ion mill uses an argon ion beam accelerated toward the sample to remove material by collision. The settings of the ion mill are 5 kV and 0.7 mA per beam. The milling angle is adjusted from a starting range of 18 to a final range of 13. The ion mill uses two ion guns, 180

PAGE 166

151 apart, for milling of both sides of the sample. The sample is also rotated to allow for a more uniform milling process. The fi nal sample thickness is about 100 nm. B.2 Focused Ion Beam Milling Focused Ion Beam (FIB) milling uses Ga ions to mill out a sample from a bulk piece of material. The bulk material must be coated with carbon or another conductive material to prevent charging of the surface so that it can be imaged with the ions or electrons. My samples were affixed to a FI B sample stub with carbon paint. and then carbon coated with to a thickness of about 500. Once the sample is mounted and coated, the sample is ready to be loaded into the FIB. First ensure all of the beams are off a nd the nitrogen tank is not empty before venting the chamber. Place sample on the st age and use “dog ear” tool to make a rough adjustment for the z-height of the sample Remove “dog ear” tool and pump down the chamber. Then, turn on the high tension, elec tron and ion beams, and the GIS heating (Pt source). Before depositing platinum, the eucen tric position needs to be aligned and the electron beam and the ion beam need to be li nked. The eucentric is the height at which there is minimal shift in position with a large ch ange in tilt. This ensures that the sample will not hit the needle or detectors in the cham ber. To link the beams, focus and locate a piece of debris near the desired sample area and select zero beam-shift. This ensures that the ion beam and the electron beam are always focused on the same location. The electron beam and the ion beam are 52 degrees different in tilt and will not show the same image with out this step. Focus on the piece of debris using one beam, switch to the other beam, and locate the sa me point with the x and y knobs. The Auto FIB program can be used at this point.

PAGE 167

152 The protective platinum layer (1x1x20 m) is then deposited to protect the surface of the sample during milling. The sample is now ready to be milled. The initial trenches are milled at a high curr ent (7000 pA). A series of trenches are milled to either side of the desired sample area and the sample gets progressively thinner, (Figure B-4). The beam current used for milling is reduced (2000 pA-300 pA) as well as the sample gets thinner to reduce damage a nd redeposition on the sample. At about a sample thickness of 0.5 microns, the sample is under cut which frees the bottom of the sample from the bulk of the material, (Fi gure B-5). The thinning process is continued until sample is at its desired thickness. Then the ends of the sample are cut free from bulk material, (Figure B-6). Once the sample is completely free from the bulk, it is removed from FIB chamber. Now the sample is lifted out from trench with glass rods and place on copper grid that has a thin layer of carbon to hold the samp le. Sample sticks to the glass rods and the grid by static electricity. This procedure is called lift-out or micromanipulation. Sample is now ready to be looked at in the TEM.

PAGE 168

153 Figure B-1. Strips of sample and support sa pphire glued together into a sample stack. Sample Sample Dummy strip Dummy strip Film Glue Film

PAGE 169

154 Figure B-2. TEM sample polished to 25 m thick. 3 mm 25 m

PAGE 170

155 Figure B-3. The sample glued to a 3mm Cu support ring and orientation of ion milling. Angle is the ion beam angle with respect to the plane of the sample. (B.P. Gila 2000) Ar+ Ar+ Cu Support ring, 3mm diameter

PAGE 171

156 Figure B-4. Top view of FIB thinned sample.

PAGE 172

157 Figure B-5. Tilted picture of undercut step of FIB sample preparation.

PAGE 173

158 Figure B-6. Sample is now cut free and ready for lift-out.

PAGE 174

159 LIST OF REFERENCES 1. M. Hong, M Passlack, J.P. Mannaerts, J. Kwo, S.N.G. Chu, N. Moriya, S.Y. Hou, V.J. Fratello, Journal of Vacuum Science and Technology B, 14(3), 2297 (1996) 2. F.Ren, M. Hong, W.S. Hobson, J.M. Kuo, J.R. Lothian, J.P. Mannaerts, J. Kwo, Y.K. Chen, A.Y. Cho, Tech. Dig. Int. Electron Devices Meeting, 943 (1996) 3. J. Kwo, D.W. Murphy, M. Hong, J.P. Mannaerts, R.L. Opila, R.L. Masaitis, A.M. Sergent, Journal of Vacuum Science and Technology B, 17(3), 1294 (1999) 4. J. Kwo, D.W. Murphy, M. Hong, R.L. Opila, J.P. Mannaerts, A.M. Sergent, R.L. Masaitis, Applied Physics Letters, 75(8), 1116 (1999) 5. L.W. Tu, Y.C. Lee, K.H. Lee, C.M. Lai, I. Lo, K.Y. Hsieh, M.Hong, Applied Physics Letters, 75(14), 2038 (1999) 6. A.R. Korton, M. Hong, J. Kwo, J.P. Mannaerts, N. Kopylov, Physical Review B, 60(15), 10913 (1999) 7. M. Hong, Z.H. Lu, J. Kwo, A.R. Korton, J.P. Mannaerts, J.J. Krajewski, K.C. Hsieh, L.J. Chou, K.Y. Cheng, Applied Physics Letters, 76(3), 312 (2000) 8. S.J. Pearton, F. Ren, A.P. Zhang, G. Dang, X.A. Cao, K.P. Lee, H. Cho, B.P Gila, J.W. Johnson, C. Mnier, C.R. Abernathy, J. Han, A.G. Baca, J.I. Chyi, C.M Lee, T.E. Nee, C.C. Chuo, S.N.G. Chu, Materials Science and Engineering B, 82, 227231 (2001) 9. D.R. Askeland, The Sciemce and Engineeri ng of Materials, 3rd ed., PWS, Boston MA, 1994 10. R. Wolfson and J.M. Pasachoff, Physics, Little, Brown and Company, Boston, 1987 11. J. W. Mayer, S. S. Lau, Electronic Materials Science, Macmillan, New York, 1990 12. C. T. Sah, Fundamentals of Solid -State Electronics, World Scientific, New Jersey, 1991 13. H.C. Casey Jr, G.G. Fountain, R.G. Alley, B.P Keller, S.P. Denbaars, Applied Physics Letters 68, 1850 (1996)

PAGE 175

160 14. M. Sawada, T. Sawada, Y. Yamagata, K. Im ai, H. Kumura, M. Yoshino, K. Iizuka, H. Tomozawa, Proceedings of the Second International Conference on Nitride Semiconductors, Tokushima, 482 (1997) 15. S. Aurlkumaran, T. Egawa, H. Ishikawa, T. Jimbo, M. Umeno, Applied Physics Letters, 73(6) 809-811 (1998) 16. S.L. Rumyanysev, N. Pala, M.S. Shur, R. Gaska, M.E. Levinshtein, M. Asif Khan, G. Simin, X. Hu, J. Yang, Journal of Applied Physics, 90 (1), 310-314 (2001) 17. T.E. Cook, Jr. C.C. Fulton, W.J. Mecouch, K.M. Tracey, R.F. Davis, E.H. Hurt, G. Lucovsky, R.J. Nemanich, Journal of Applied Physics 93 (7), 3995-4004(2003) 18. K.M. Chang, C.C. Cheng, C.C. Lang, El ectrical properties of SiN/GaN MIS diodes formed by ECR-CVD, Solid-State Electronics 46, 1399-1403(2002) 19. T. Hashizumu, S. Ootomo, T. Inagaki, H. Hasegawa, Journal of Vacuum Science and Technology B, 21(4), 1828-1830 (2003) 20. L.J. Guido, P. Mitev, M. Gherasimova, B. Gaffey, Applied Physics Letters, 72, 2005 (1998) 21. T.P. Ma, Microelectronics Journal 34,363-370 (2003) 22. B. Gaffey, G. Chong, L. Guido, X.W. Wang, T.P. Ma, Applied Physics Letters, 42, 2145 (1998) 23. F. Ren, C.R. Abernathy, J.D. MacKenzie, B.P. Gila, S.J. Pearton, M. Hong, M.A. Marcus, M.J. Schuman, A.G. Baca, R.J. Shul, Solid State Electronics 42, 2177 (1998) 24. H. Kawai, M. Hara, F. Nakamura, T. Asatsuma, T. Kobayashi, S. Imanaga, Journal of Crystal Growth 189/190, 738 (1998) 25. S.D. Wolter, B.P. Luther, D.L. Waltermyer, C. Onneby, S.E. Mohney, R.J. Molnar, Applied Physics Letters 70, 2156 (1997) 26. S.D. Wolter, S.E. Mohney, H. Venugopala n, A.E. Wickenden, D.D. Koleske, Journal of the Elect rochemical Society 145, 629 (1998) 27. Y. Nakano, T. Kachi, T. Jimbo, Journal of Vacuum Science and Technology B 21(5), 2220-2222 (2003) 28. M. Hong, J. Kwo, S.N.G. Chu, J.P. Mannaerts, A.R. Kortan, H.M. Ng, A.Y. Cho, K.A. Anselm, C.M. Lee, J.I. Chyi, Journal of Vacuum Science and Technology B 20 (3), 1274-1277 (May/June 2002)

PAGE 176

161 29. D.J. Fu, Y.H. Kwon, C.J. Park, K.H. Baek, H.Y. Cho, D.H. Shin, C.H. Lee, K.S. Chung, Applied Physics Letters 80 (3), 446-448 (2002) 30. L.H. Peng, C.H. Liao, Y.C. Hsu, C.S. Jong, C.N. Huang, J.K. Ho, C.C. Chiu, C.Y. Chen, Applied Physics Letters 76, 511 (2000) 31. D.J. Fu, T.W. Kang, Sh.U. Yuldashev, N.H. Kim, S.H. Park, J.S. Yun, K.S. Chung, Applied Physics Letters 78, 1309 (2001) 32. D. Fu, S.U. Yuldashev, N.H. Kim, Y.S. R yu, J.S. Yun, S.H. Park, T.W. Kang, K.S. Chung, Japanese Journal of Appl ied PhysicsPart 2 40, L10 (2001) 33. C. T. Lee, H.W. Chen, H.Y. Lee, Metal-oxide-semiconductor devices using Ga2O3 dielectrics on n-type GaN, Applied Physics Letters 82(24), 4304-4306 (2003) 34. T. Rotter, R. Ferretti, D. Mistele, F. Fedler, H. Klausing, J. Stemmer, O.K. Semschinova, J. Aderhold, J. Graul, Journal of Crystal Growth 230, 202-206 (2001) 35. C. Bae, G.B. Rayner, G. Lucovsky, Applied Surface Science 216, 119-123 (2003) 36. R. Therriem. G. Lucovsky, R. Davis, Applied Surface Science 166, 513-519 (2000) 37. C. Bae, G. Lucovsky, Surface Science 532-535, 759-763 (2003) 38. C. Bae, G. Lucovsky, Applied Surface Science 212-213, 644-648 (2003) 39. Y. Nakano, T. Kachi, T. Jimbo, Applied Physics Letter, 83(21), 4336-4338 (2003) 40. S.J. Chang, Y.K. Su, Y.Z. Chiou, J.R. Ciuo, B.R. Huang, C.S. Chang, J.F. Chen, Journal of the Elect rochemical Society, 150(2), C77-C80 (2003) 41. E.D. Readinger, S.D. Wlter, D.L. Waltermyer, J.M. Delucca, S.E. Mohney, B.I. Prenitzer, L.A. Gainnuzzi, R.J. Molnar, Journal of Electronic Materials 28, 257 (1999) 42. M. Hong, J.P. Mannaerts, J.E. Bower, J. Kwo, M. Passlack, W-Y. Hwang, L.W. Tu, Journal of Crystal Growth 175/176, 422 (1997) 43. F. Ren, M. Hong, W.S Hobson, J.M. Kuo, J.R. Lothian, J.P. Mannaerts, J. Kwo, S.N.G. Chu, Y.K. Chen, A.Y. Cho, Solid State Electronics 41, 1751 (1997) 44. Y.C. Wang, M. Hong, J.M. Kuo, J.P Mannaer ts, J. Kwo, H.S. Tsai, J.J. Krajewski, J.S. Weiner, Y.K. Chen, A.Y. Cho, Materials Research Society Symposium Proceedings 573, 219 (1999) 45. J. Kwo, M. Hong, A.R Kortan, D.W. Mur phy, J.P Mannaerts, A.M. Sergent, Y.C. Wang, K.C Hsieh, Materials Research Society Symposium Proceedings 573, 57 (1999)

PAGE 177

162 46. T.S. Lay, M. Hong, J. Kwo, J.P Mannaerts, W.H. Hung, D.J. Huang, Materials Research Society Symposium Proceedings 573, 131 (1999) 47. A.R. Mortan, M. Hong, J. Kwo, J.P Mannaerts, N. Kopylov, Materials Research Society Symposium Proceedings 573, 21 (1999) 48. F. Ren, M. Hong, S.N.G. Chu,, M.A. Marcus, A. Baca, S.J. Pearton, C.R. Abernathy, Applied Physics Letters, 73(26), 3893-3895 (1998) 49. J. Kim, R. Mehandru, B. Lou, F. Ren, B. P. Gila, A.H. Onstine, C.R. Abernathy, S.J. Pearton, Y. Irokawa, Applied Physics Letters 81(2), 373-375 50. B. Lou, J.W Johnson, B.P. Gila, A.H. Onstine, C.R. Abernathy, F. Ren, S.J. Pearton, A.G. Baca, A.M. Dabira n, A.M. Wowchack, P.P. Chow, Solid State Electronics 46, 467-476 (2002) 51. R. Mehandru, B.P. Gila, J. Kim, J.W Johnson, K.P. Lee, B. Lou, A.H. Onstine, C.R. Abernathy, S.J. Pearton, F. Ren, Electrochemical and Solid State Letters 5(7) G51-G53 (2002) 52. B. Lou, J.W Johnson, J. Kim, R. Mehandru, F. Ren, B.P. Gila, A.H. Onstine, C.R. Abernathy, S.J. Pearton, A.G. Baca, R.D. Briggs, R.J. Shul, C. Monier, J. Han, Applied Physics Letters 80(9) 1661-1663 (2002) 53. J. Kim, B.P. Gila, R. Mehandru, J.W Johns on, J.H. Shin, K.P. Lee, B. Lou, A.H. Onstine, C.R. Abernathy, S.J. Pearton, F. Ren, Journal of the Electrochemical Society 149 (8) G482-G484 (2002) 54. W.M. Yim, E.J. Stofko, P.J. Zanzucchi, J.I. Pankove, M. Ettenburg, S.L. Gilbert, Journal of Applied Physics, 44, 292 (1973) 55. V.W.L. Chin, T.L. Tansley, T. Osotchan, Journal of Applied Physics, 75, 7365 (1994) 56. M.S. Shur, M.A. Khan, Mat. Res. Bull., 22(2), 44 (1997) 57. A.I. Manakov, O.A. Esin, B.M. Lepinskikh, Russian Journal of Physic and Chemistry, 36, 1481, (1965) 58. Refractory Production, Handbook [in Russian], Izd. Metallurgiya 1965 59. G.W. Kaye, T.H. Laby, Tables of Physical and Ch emical Constants and some Mathematical Functions, 11th ed, Longmans, New York, 1956 60. S.J. Pearton, International. Journal of. Modern. Physics. B, 7, 4687 (1993) 61. M. Passlack, M. Hong, J.P. Mannaerts, Applied Physics Letters, 68, 1099 (1996)

PAGE 178

163 62. O. Kubaschewski and B. Hopkins, Oxidation of Metals and Alloys [Russian translated], Izd. Metallurgiya, 1965 63. S.S. Derbeneva, S.S. Batsano, Dokl. Chem., 175, 710 (1967) 64. S.S. Batsono, E.V. Dulepov, Soviet Physics – Solid State, 7(4), 995 (1965) 65. K.A. Gschneider, Rare Earth Alloys [Russian translation], Izd. Mir., 1965 66. Q. Xiao and J.J. Derby, Journal of Crysal Growth, 139, 147 (1994) 67. S.M. Sze, Physics of Semiconductor Devices, 2nd edition, Wiley, New York, 1981 68. D.R. Lide, Handbook of Chemistry and Physics, 71st edition, CRC Press, Boca Raton, 1990 69. R.L. Melik-Davtyan, N.F. Shvartsenau, A.I. Shelykh, Inorganic Materials, 2(1), 21 (1966) 70. J.L. Pentecost, Ceramic Age, 71, 26 (1958) 71. A.A. Vorob’ev, Physical properties of Crystalline Ionic Dielectrics [in Russian], Izd. Tomsk. Univ. 1961 72. G.V. Samsonov, The Oxide Handbook, Plenum, New York, 1973, and references therein 73. Radical Beam Source Model MDP21S Operating and Service Manual, Oxford Applied research, Witney (1989) 74. D.K. Schroder, Semiconductor Material and Device Chartacterization 2nd ed., Wiley, New York, 1998 75. B.D. Cullity, Elements of X-ray Diffraction, 2nd ed, Addison-Wesley, London, 1978 76. Y. Kuroda, E. Yasugi, H. Aoi, K. Miura, T. Morimoto, Journal of Chemical Society, Farady Tansactions, 84(7), 2421-2430 (1988) 77. Refson, Wogelius, Fraser, Payne, Lee, Milman; Physical Review B 52(15) (1995) 78. E.M. Levin, C.R. Robins, H.F. McMurdie, Phase Diagrams for Ceramists, The American Ceramic Society, Columbus, 1964 79. E.S. Hellman, E.H. Hartford, Jr., Applied Physical Letters 64 (11) 1341-1343 (1994)A. H. Onstine, B. P. Gila, J. Kim, D. Stodilka, K. Allums, C. R. Abernathy, F. Ren, S. J. Pearton submitted for publication in Sol. Stat. Elect. 2003.

PAGE 179

164 81. B.P. Gila, J.W. Johnson, R. Mehandru, B. Luo, A.H. Onstine, K.K. Allums, V. Krishnamoorthy, S. Bates, C. R. Abernathy, F. Ren, S.J. Pearton, Phys. Stat. Sol. (a) 188 (1), 239-242 (2001).

PAGE 180

165 BIOGRAPHICAL SKETCH I was born in Minneapolis, Minnesota in 1975. I received my Bachelors in Chemical Engineering from the University of Florida in May 1999. I then continued on in graduate school in Materials Science a nd Engineering and received my Masters of Science in Materials Science and Engineer ing from the University of Florida in December 2002 and Doctor of Philosophy in Ma terials Science and Engineering from the University of Florida in December 2004. Wh ile in school at UF I started the Florida Swing Dancing Club. I also lived on a horse farm, in an apartment in the barn, while in graduate school, which was wonderful.


Permanent Link: http://ufdc.ufl.edu/UFE0006644/00001

Material Information

Title: Growth and Characterization of Novel Gate Dielectrics for Gallium Nitride
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0006644:00001

Permanent Link: http://ufdc.ufl.edu/UFE0006644/00001

Material Information

Title: Growth and Characterization of Novel Gate Dielectrics for Gallium Nitride
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0006644:00001


This item has the following downloads:


Full Text












GROWTH AND CHARACTERIZATION OF NOVEL GATE DIELECTRICS FOR
GALLIUM NITRIDE














By

ANDREA H. ONSTINE


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


2004

































Copyright 2004

by

Andrea H. Onstine





















"I don't think there is any concept that you can't make understandable to the
educated lay public. I always tell my students and postdocs if you can't explain to
your grandmother what you are doing, probably you don't understand it yourself
properly."

-Nobelist Gunter Blobel









"Between the stirrup and the ground I asked for mercy and mercy there I found."

-English/Irish Proverb.















ACKNOWLEDGMENTS

This dissertation would not be possible without the help and support of many

people in my life. I would first like to thank my advisor, Dr. Cammy Abernathy, for

accepting me into the research group, for always having an encouraging word, and for

being an inspiration that you can have it all- career, life, marriage, and family. Dr.

Abernathy remembered that we were people, and encouraged us to have a life outside of

school. I thank my committee members (Dr. Steve Pearton, Dr. David Norton, Dr. Fan

Ren, Dr. Rajiv Singh) for their support and knowledge in the field of GaN device

processing and semiconductor device physics. I would also like to thank Dr. Amlan

Biswas for substituting for Dr. Ren on such short notice.

Many thanks go to Dr. Brent Gila for teaching me everything he knows (or at least

as much as I could learn) in the past 4 years. From pumps and MBE, to characterization,

to how to deal with people, Brent's help and guidance have been invaluable. People will

always enjoy working for and with him.

Thanks go to my parents and family, who supported me completely in anything and

everything that I wanted to try. Special thanks go to Aunt Janeen and my father, for all of

their help editing and proofreading my dissertation even though they may not have

understood everything. I regret that all of my grandparents, especially my paternal

grandmother, did not live long enough to see me graduate with my Ph.D.

I would also like to thank my group members and co-workers (Jerry Thaler, Rachel

Fraser, Jen Hite, Kimberly Allums, Danielle Stodilka, Mark Hlad, Andrew Hererro,

iv









Jiyhun Kim, Rishab Mehandru, and Wayne Johnson) for all of their help and support. I

thank Nina Burbure, Carrie Ross, and Erik Kuryliw for all the time they spent with the

Focused Ion Beam system. Thanks go to Kerry Siebein for all of her time spent working

in the cold with the High Resolution-Transmission Electron Microscope as well as the

conversation.

I thank my long-time friends Aamir Qaiyumi, Amy Matthews, Jennifer Babin nee

Adam, Bill Weisner, Bill O'Conner, Pete Meyer, Kathy Daly, Gloria Bergman, Jacques

Palmer, Lori Kornberg, and everyone from the Swing Dancing Club, and from Tallwood,

for all of their support. It's hard to believe that it has been so long.
















TABLE OF CONTENTS
Page

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

LIST OF TABLES ..................................................... ix

LIST OF FIGU RE S .............. .......................... ........................ .. .. .... .x

ABSTRACT ................................................... ................. xiv

CHAPTER

1 IN TR O D U C T IO N ........ .. ......................................... ..........................................1.

1.1 M otiv atio n ....................................................... ............................................. . 1
1.2 D issertation O utline ....................................................................... ...............3...

2 BACKGROUND AND LITERATURE REVIEW ................................................5...

2.1 Introduction to Dielectric Films, Capacitors, and MOS(MIS)FETs....................5...
2.1.1 D electric Film s ............................................................ ....... ................ .5
2.1.2 Metal/Oxide/Semiconductor (MOS) Capacitor......................................7...
2.1.3 Metal/Insulator/Semiconductor (MIS) Capacitor...................................8...
2.2 G aN B ased Electronic D evices.................. ....................................................9...
2 .2 .1 Silicon O xide on G aN ........................................................... ............... 10
2.2.2 Silicon N itride on G aN .................. .................................................... 10
2.2.3 A lum inum N itride on G aN .................................................... ................ 11
2.2.4 G allium O xide on G aN .......................................................... ............... 12
2.2.5 Silicon Dioxide on Gallium Oxide on Gallium Nitride.......................... 13
2.2.6 Gallium Gadolinium Oxide on GaN ............... .................................... 13
2.2.7 G adolinium O xide on G aN .................................................... ................ 14
2.2.8 Scandium O xide on G aN ....................................................... ............... 14

3 EXPERIM ENTAL APPROACH ....................... ............................................... 19

3.1 M olecular B eam E pitaxy ....................................... ....................... ............... 19
3.1.1 Sub state P reparation ............................................................. ................ 2 1
3 .1.1.1 Silicon ...................................................................................... 22
3.1.1.2 G allium N itride .......................................... ............... .............. ... 22
3.1.2 Magnesium Oxide Growth .......... ....................... 23
3.1.3 M agnesium Calcium Oxide Growth............... ..................................... 24









3.1.4 M agnesium Scandium Oxide Growth .............. .................................... 24
3.2 M materials C haracterization ............................................................... ................ 25
3.2.1 Reflection High Energy Electron Diffraction: RHEED ..........................25
3.2.2 Transmission Electron M icroscopy: TEM ...................... ..................... 26
3.2.3 X -R ay D iffraction: X R D ....................................................... ................ 27
3.2.4 Atomic Force M icroscopy: AFM ............... .................................... 28
3.2.5 Scanning Electron M icroscopy: SEM .............. .................................... 28
3.2.6 Auger Electron Spectroscopy: AES ...................................... ................ 29
3.2.7 Ellipsom etry ... ............... .. .. ...................... ............... 30
3.2.8 Current-Voltage (I-V) m easurem ents.................................... ................ 31
3.2.9 Capacitance-Voltage (C-V) measurements........................... ................ 31

4 MAGNESIUM OXIDE: RESULTS AND DISCUSSION .................................... 48

4.1 Effect of O xygen Plasm a Source ........................... .................... .....................48
4.2 E effects of O xygen Pressure ............................................................. ................ 50
4.3 Effect of Substrate Tem perature...................................................... ............... 52
4.4 Scandium Oxide Capping Layer..................................................... 53

5 MAGNESIUM CALCIUM OXIDE: RESULTS AND DISCUSSION.....................72

5 .1 G ro w th o f C aO ..................................................................................................... 7 2
5.2 G row th of Ternary M gC aO ............................................................. ................ 73

6 SCANDIUM MAGNESIUM OXIDE AND MAGNESIUM SCANDIUM OXIDE:
RESULTS AND DISCU SSION .................................................... ............... 101

6.1 Scandium M agnesium Oxide...... ............ .......... ..................... 102
6.2 M agnesium Scandium Oxide...... ............ .......... ..................... 103

7 ENVIRONMENTAL AND THERMAL STABILIY ................... ...................116

7 .1 E nvironm ental Stability ...................................................................................... 116
7.1.1 M agnesium O xide ......................................................... 116
7.1.2 M agnesium Calcium O xide.............................................. .................. 117
7.1.3 Magnesium Scandium Oxide and Scandium Magnesium Oxide.......... 117
7.2 T herm al Stability .............. ...... ............. ................................................ 117
7.2.1 M agnesium O xide ......................................................... 118
7.2.2 M agnesium Calcium Oxide.............................................. .................. 118
7.2.3 Magnesium Scandium Oxide and Scandium Magnesium Oxide.......... 119

8 SUMMARY AND FUTURE WORK ....................................132

8.1 M agnesium Oxide ... ............................................................... 132
8.2 M agnesium Calcium Oxide ........... .. ........................ .................. 133
8.3 Scandium Magnesium Oxide and Magnesium Scandium Oxide .....................134
8.4 Environm ental and Therm al Stability ............... .............. ..................... 134









APPENDIX

A E L L IP SO M E T R Y .................................................. ............................................ 136

A.1 M axwell's Equations and the EM Plan W ave ................................. ............... 138
A .2 Jones V ectors and M atrices ....... ........... ............ ...................... 140
A .3 L eight Polarization States................................... ...................... ............... 141
A.4 Single Films on Thick Substrates .......... ......................... 142

B TEM SAM PLE PREPAR A TION ........................................................... ............... 150

B. 1 Old-fashioned Hand Grinding- ....... .......... ......... ...................... 150
B.2 Focused Ion Beam M illing...... ............ ............ ..................... 151

LIST O F R EFEREN CE S ... ................................................................... ............... 159

BIOGRAPH ICAL SKETCH .................. .............................................................. 165



































viii















LIST OF TABLES


Table Page

2.1 Properties of dielectric materials that have been used on GaN.............................. 16

4-1 Electrical characterization of M gO/GaN diodes. ................................ ................ 54

5-1 Growth rate and AES data for M gCaO samples. ................................ ................ 78

5-2 Growth rate, AFM and AES data for MgCaO grown at 300C. ..............79

5-3 Com position and m ism atch from XRD ............................................... ................ 80

6-1 Dependence of growth rate, RMS roughness, and AES ratio..............................105

7-1 Ellipsometry data. Change in index of refraction ...................... ................... 120

7-2 XRR data of as-grown and annealed samples....... ... ..................................... 121















LIST OF FIGURES


Figure Page

2 -1 C ap acito r d iag ram s ................................................................................................... 17

2-2 Cross-section illustration of a depletion mode n-MOSFET...............................18

3-1 Typical Knudsen effusion oven (After B.P. Gila 2000) ................ ..................... 33

3-2 Riber M BE used for oxide grow th ..................................................... ................ 34

3-3 W aveM at 610 ECR plasm a head ........................................................ ................ 35

3.4 Schematic of the Oxford RF plasma source ................ .................................... 36

3-5 AFM images of as received MOCVD GaN and MBE GaN. ................................. 37

3-6 RHEED images showing treated surface of MOCVD grown GaN ......................38

3-7 RHEED photos indicating a (1x3) pattern.. ........................................ ................ 39

3-8 RHEED photos showing different types of diffraction patterns .............................40

3 -9 T E M co lu m n ........................................................................................................... 4 1

3-10 The relation between the lattice parameter and Bragg angle .............................. 42

3-11 Atomic force microscope (after K.K. Harris 2000). ...........................................43

3-12 Schem atic of SE M colum n .......................................... ....................... ............... 44

3-13 SEM operation ........................... ....................................... 45

3-14 The penetration depth and interaction of an electron beam.. ............................... 46

3-15 Ellipsometer schematic (After D.K Schroder74) ........................... ..................... 47

4-1 The M gO structure (from Cullity 197875)........................................... ................ 55

4-2 Illustration showing the symmetry between (111) NaCl and (0002) wurtzite plane.56

4-3 RHEED images indicating clean GaN surface and MgO growth ......................... 57









4-4 SEM image of the MgO surface grown at 3500C with ECR plasma source............58

4-5 AFM images of MgO grown at 3500C and MgO grown at 100C.......................59

4-6 RHEED images of MgO after 1 minute and after 90 minutes of growth at 1000C. 60

4-7 SEM and AFM of MgO grown at 100C with RF plasma source............................61

4-8 Mg/O ratio as determined by AES as a function of oxygen pressure ...................62

4-9 Dependance of growth rate on oxygen pressure ................................. ................ 63

4-10 AFM of MgO grown at 100C with oxygen pressure of Ixl0-5 or Ixl0-4 Torr.......64

4-11 SEM and TEM of M gO ........................................ ......................... ................ 65

4-12 XRD shows that the MgO film grown at 300 C vs 100C..................................66

4-13 AFM of MgO grown at 1000C and grown at 300C ............................................. 67

4-14 H R TEM of M gO grow n at 300C ........................................................... ............... 68

4-15 SEM of degraded M gO fi lm .......................................... ...................... ............... 69

4-16 I-V and C-V of MgO/n-GaN diodes, agedl5 weeks...........................................70

4-17 AFM of M gO with and without a capping layer.................................... ............... 71

5-1 Illustration of the CaO structure (from Cullity 1978) ........................................81

5-2 AES shows only Ca and 0 after growth of CaO at 300C ................................... 82

5-3 M gO -C aO phase diagram ......................................... ........................ ................ 83

5-4 HR TEM of CaO .............. ........................ ................ .. .......84

5-5 A FM of C aO grow n at 100C ....................................... .................................... 85

5-6 A FM of C aO grow n at 300C ..................................... ..................... ................ 86

5-7 AE S depth profiling of M gCaO .......................................................... ................ 87

5-8 XRD of MgCaO shows no signs of phase separation or secondary phases.............88

5-9 Powder XRD showing increase in lattice constant with the addition of Ca ..........89

5-10 HR-XRD of the relative positions of MgO (222) and MgCaO (222) peaks............90

5-11 A FM of M gC aO grow n at 100C .............................................................................91









5-12 AES shows that no Mg or Ca is present after growth time...............................92

5-13 AES of MgCaO film grown at substrate temperature of 300C...............................93

5-14 XRD shows the peak from the MgCaO film grow at 300C vs 100C .................94

5-15 AFM of M gCaO grown at 100C and 300C ........................................ ............... 95

5-16 AES scans of continuously grown and digitally grown samples ..........................96

5-17 HR-XRD showing MgCaO and MgO texturing in the (111) direction.................... 97

5-18 H igh resolution X TEM of M gO ............................................................. ............... 98

5-19 HR XRD showing change in peak position with change in shutter timing ...........99

5-20 AFM of capped MgCaO at different shutter sequences............................100

6-1 The Bixbyite crystal structure of scandium oxide. (B.P Gila 2000) ....................106

6-2 AES depth profiles taken from layers grown under different growth conditions. .107

6-3 AFM surface scans for films grown at different gorwth conditions .......... 108

6-4 XRD of ScMgO grown at TMg =3500C and Tsub = 3000C.. ............................... 109

6-5 Powder XRD showing the peak position of MgScO peak...............................110

6-6 AES of M gScO with Sc203 capping layer ............................................................. 111

6-7 AFM of MgScO grown at 100C, uncapped and with a Sc203 cap ...........112

6-8 AFM of MgScO grown at 300C without capping layer. ...... ...........113

6-9 TEM of MgScO shows epitaxial growth of MgScO on GaN ...............114

6 -10 E D S o f S cM g O ........................................................ ....................................... ... ... 1 15

7-1 Accelerated aging experimental set up............... ......................... 122

7-2 Ellipsometry of the degradation of MgO with and without a capping layer.......... 123

7-3 Ellipsometry of the degradation of MgCaO with and without capping layer .......124

7-4 Ellipsometry of the degradation of MgScO with and without capping layer......... 125

7-5 XRR of MgO uncapped, before and after annealing at 1000C for 2 minutes. ..... 126

7-6 XRR of capped MgO before and after annealing at 1000C for 2 minutes..........127









7-7 XRR of MgCaO uncapped before and after annealing at 1000C at 2 minutes..... 128

7-8 XRR of MgCaO capped before and after annealing at 1000C for 2 minutes ...... 129

7-9 XRR of MgScO uncapped, before and after annealing at 1000C for 2 minutes... 130

7-10 XRR of MgScO capped, before and after annealing at 1000C for 2 minutes. ..... 131

A-i An electrom agnetic plane wave. ...... ........... .......... ...................... 147

A-2 Geometry of an ellipsometric experiment, showing p- and s-directions.............1...48

A-3 Multiple reflected and transmitted beams for a single film.. ...... .................. 149

B-i Strips of sample and support sapphire glued together into a sample stack ..........153

B-2 TEM sample polished to 25 |tm thick........... ................154

B-3 The sample glued to a 3mm Cu support ring and orientation of ion milling .........155

B -4 Top view of FIB thinned sam ple .......................................................... ............... 156

B-5 Tilted picture of undercut step of FIB sample preparation. ..... ...........157

B-6 Sample is now cut free and ready for lift-out....... ... .................................... 158















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

GROWTH AND CHARACTERIZATION OF NOVEL GATE DIELECTRICS FOR
GALLIUM NITRIDE

By

Andrea H. Onstine

December 2004

Chair: Cammy R. Abernathy
Major Department: Material Science and Engineering

Novel crystalline dielectric materials for gate application on gallium nitride were

studied. These dielectric materials must operate at high temperatures and under high

power loads. To meet these needs, the selected dielectric materials must be thermally

stable to temperatures above 1000C for device fabrication, must be chemically stable to

prevent diffusion into the semiconductor, and must have a low defect density to reduce

the charged trap sites in the dielectric and the dielectric/semiconductor interface. The

dielectric materials studied were magnesium oxide (MgO), magnesium calcium oxide

(MgCaO), and magnesium scandium oxide (MgScO).

These materials were deposited using molecular beam epitaxy (MBE) where the

individual elements are supplied independent of each other. This technique allows for the

use of a wide range of growth conditions in order to obtain the highest quality material

and precise control of the film composition. The dielectrics were deposited on gallium

nitride for characterization and device fabrication. The samples were characterized using









a variety of techniques to determine surface roughness, crystal structure, chemical

composition, and electrical properties.

MgO deposited using the MBE approach grew epitaxially on GaN for the first 40

monolayers, after which it became polycrystalline. MgO grown at 300C using a radio

frequency (RF) oxygen plasma source showed the best properties. Optimization of the

MgO did not eliminate the problems with MgO as a gate dielectric, namely poor

environmental and thermal stability. On annealing, the interface becomes rougher as

evidenced by x-ray reflectometry.

Crystalline, single-phase MgCaO was grown by the continuous and digital MBE

methods. The digital approach showed better uniformity and morphology. From XRD it

was found that the addition of Ca could be used to vary the bond mismatch over a range

of -6.5% to +0.96%. As with MgO, environmental stability was still problematic because

of reactivity with moisture.

Bixbyite ScMgO exhibited a solid solubility limit of about 9% Mg, after which a

magnesium-rich second phase was observed by XRD. This severely limits the usefulness

of ScMgO. Rock salt MgScO was grown below this solubility limit and grew epitaxially

on GaN. The environmental and thermal stability are not significantly affected by the

addition of Sc.

The environmental and thermal stability of the oxides were also investigated. A

Sc203 capping layer was shown to improve both thermal and environmental stability of

the oxides. The most stable dielectric was found to be MgCaO with a Sc203 cap.


xv














CHAPTER 1
INTRODUCTION

1.1 Motivation

The modern microelectronics industry is largely based on solid-state silicon

technology. Compound semiconductors are becoming an increasingly important area of

research and technology because of the limitations of silicon-based technology. The

bandgap of silicon limits the temperature and power operation range of the devices.

Also, since silicon is an indirect bandgap material, it will never be an efficient light

emitter. Compound semiconductors are based on elements from Groups III and Groups

V of the periodic table, such as gallium arsenide (GaAs) and indium phosphide (InP).

These compound semiconductor devices have higher carrier mobilities, resulting in faster

devices and lower-saturation electric fields than silicon semiconductor devices. Also,

some compounds have direct band gaps that lead to efficient light emission and light

detection.

Transistor research based on compound semiconductors has led to several

breakthroughs in device performance. Recently, research in this field has produced a

GaAs metal/oxide/semiconductor (MOS) capacitor that demonstrated properties useful

for transistors.1 This discovery led to an operational GaAs-based MOS transistor

incorporating gallium gadolinium oxide as a gate dielectric.2 Further research has shown

that the gadolinium content of the oxide was responsible for surface passivation, and

improved the electrical properties of the device.3-7 There are still limits to these exciting

compound materials, such as the thermal operating limit and power-handling capabilities.

1









These obstacles can be overcome by using a semiconductor material with a wider

bandgap. The term "wide bandgap" refers to the forbidden energy gap of a material, a

region in the energy diagram that is not occupied by electrons, that is typically greater

than 2 eV, (Figure 1.1). Wide bandgap semiconductors have been researched for

decades, beginning with silicon carbide (SiC) in the middle of the 20th century. In more

recent years, research has turned to Group III-nitrides (such as gallium nitride), and

Group II-VI materials. Because of the ease of defect propagation through the II-VI

device structure during operation at room temperature, these materials have been set

aside for materials that show more potential. The III-V nitride and SiC semiconductors

are more thermally stable at device operating temperatures, and therefore do not have the

defect-propagation problems associated with II-VI semiconductors. Table 1.1 lists some

wide bandgap semiconductors and their properties, as well as those of silicon and gallium

arsenide.

Nitride-based semiconductors have become a focus for research into optoelectronic

devices. By creating ternary alloys, the III-nitride light emitting devices (LEDs) have

covered the entire visible spectrum. Other photonic devices include UV detectors and

laser diodes. Many of the lessons learned (and processes developed) by photonics

research have led to the growth of high-quality material, improved electrical contacts, and

controllable materials processing. From this base, research has been initiated to create

microwave, ultrahigh-power switches and devices that operate at high temperatures.8

Advances in SiC and GaN have led to power switches based on different configurations

like metal-semiconductor field effect transistors (MESFETs), heterojunctions field effect

transistors (HJFETs), and heterojunction bipolar transistors (HBTs). While these devices









have shown promise for a number of applications, the metal oxide (or insulator)

semiconductor field effect transistor (MOSFET or MISFET) is also a desirable structure.

Complementary devices are required for logic circuits. The MOSFET or MISFET

structure can be made into complementary metal oxide semiconductor, CMOS, logic.

The MESFET, HJFET, and HBT structures cannot be made into complementary devices.

A complementary circuit based on wide bandgap semiconductors will allow for an entire

monolithic control circuit to be constructed for high-temperature/high-power use.

For a MOS(MIS)FET to be realized, a high-quality dielectric material must be

created for the gate insulator. This material must have a bandgap wider than the

semiconductor, a dielectric constant larger than the semiconductor, and high temperature

stability similar to the semiconductor. The materials that have been previously

researched for this roll are discussed in Section 2.2. Because of deficiencies in various

areas, each of these materials is not optimum for devices on GaN. Finding new materials

to satisfy these requirements is the goal of our study. Materials selected for this study

were MgO, MgCaO, ScMgO, and MgScO.

1.2 Dissertation Outline

The objective of our study was to explore the feasibility of growing lattice-matched

oxides for GaN devices by molecular beam epitaxy and characterization of these

materials. The background and literature review is given in Chapter 2. In the

background, definitions of dielectric materials, capacitance, and MOSFET are given.

The literature review contains descriptions and results of dielectric materials used on

GaN. In Chapter 3, the growth methods of the oxides are explained along with the

characterization methods. Chapters 4, 5 and 6 describe the results of the dielectric

materials grown and discuss how the different dielectric materials compare to each other.






4


Chapter 7 covers the environmental and thermal stability of the dielectrics. Finally,

conclusions and future experiments are given in Chapter 8.














CHAPTER 2
BACKGROUND AND LITERATURE REVIEW

2.1 Introduction to Dielectric Films, Capacitors, and MOS(MIS)FETs

The following sections discuss the basics of dielectric films, their properties, and

applications. The capacitor is an important tool for testing these dielectric materials. The

metal/insulator/semiconductor transistor is a solid-state switch that is controlled by the

capacitor structure in the gate region.

2.1.1 Dielectric Films

Insulators are characterized by the absence of charge transport. Insulators have

positive and negative charges in the form of atom nucleus and electron cloud; but these

charges are bound to the atom or molecule, and are not available for conduction. When

materials are placed in an electric field, there is a shift,( or polarization) in the charge

distribution, and it is this polarization that leads to dielectric behavior in the material.9

The polarization induces dipoles within the atomic or molecular structures that are

aligned with the applied field. The ability of a material to resist the polarization of charge

is described as the dielectric constant, K, which is the ratio of the permittivity of the

material, si, to the permittivity of vacuum,Sv.

K= Si/ Sv (2-1)

The dielectric constant can also be related to the internal field created within the

material and the external applied field, through the equation 2-2.10

Einternal = Eapplied/ K (2-2)









The polarization, P, of the material is related to the dielectric constant by the

Equation 2-3, where is the strength of the electric field (V/m). It can be assumed from

this relation that the polarization increases as the electric field strength increases, until all

the dipoles are aligned such that

P=(K-1) FSo (2-3)

There are several applications for dielectric materials. Passivation of high-voltage

junctions, isolation of devices and interconnects, and gate insulation of field-effect

transistors are a few applications that are relevant to this discussion. For a material to be

a successful dielectric, it must meet certain criteria. Desirable characteristics include

chemical stability over the lifetime of the device, immobile charge traps (to avoid

shorting and frequency limits), and a dielectric constant higher that that of the

semiconductor (to avoid generating a high electric field in the dielectric). In the case of

the wide-bandgap semiconductor devices, the dielectric materials must also have

excellent thermal stability, since the high-power applications will result in elevated

operating temperatures. Another criterion important to wide bandgap semiconductors is

that the band gap of the dielectric must be greater than that of the semiconductor. The

ideal dielectric would keep the electrons and holes in the semiconductor, and out of the

dielectric. To achieve these properties, the bandgap of the dielectric must be larger than

the semiconductor, and the electron affinity nearer to the vacuum level, according to the

electron affinity rule, Equation 2-4.

AEc +AEv = AEg (2-4)

The dielectric/semiconductor interface is also an important focus of research in the

area of device processing. The interface state density of carrier traps must be









<1011 eV-1cm-2 for a device to be considered successful. Another important focus is the

fixed dielectric charge density, or carrier trap density within the dielectric. To date,

several dielectric materials have been researched for use in wide-bandgap semiconductor

switches, including AIN, A1203, Ga203, Gd203, Ga203(Gd203), SiOx and Si3N4, Table

2.1.

2.1.2 Metal/Oxide/Semiconductor (MOS) Capacitor

Since the gate is the actual on/off switch in the transistor structure, defining the

properties of the gate and its operation are extremely important. The gate structure is

identical to the metal/oxide/semiconductor capacitor, (Figure 2-1). A capacitor is a

device made of two parallel, conducting plates separated by an insulating material. When

a direct current (DC) voltage is applied to one side of a capacitor, an equal and opposite

charge forms on the other side of the capacitor. In most cases, the DC voltage is applied

to the metal side of the capacitor, and the charge is formed in the semiconductor. The

amount of capacitance that the capacitor can hold is directly related to the dielectric

constant of the dielectric material,11 (Equation 2-5) where C is the capacitance, so is the

permittivity of vacuum (o0=8.854x10-14 F/cm), si is the permittivity of the material, A is

the area of the metal contact, and d is the thickness of the dielectric material.

C= So Si A/d (2.5)

The capacitance is independent of the applied voltage to the gate, but is completely

dependant on the geometry and the dielectric constant. This gives a theoretical

capacitance for a given device geometry. If the applied electric field becomes too great,

the charges are ripped from the material and conducted to the charged plates. This leads

to a short in the material, and is termed dielectric breakdown.









However, as usually happens in practice, the theoretical value and the measured

value rarely match. The measured capacitance value is actually the sum of two

capacitors in series. These are the dielectric capacitance and the semiconductor space-

charge layer capacitance. The semiconductor capacitance, Cs, is responsible for deviation

in the measured capacitance, and is voltage-dependant.12 As the bias is applied to the

metal, the majority carriers in the semiconductor are repelled from the

oxide/semiconductor interface, resulting in a space-charge layer. This is assuming that

the applied bias is the same charge as the majority carriers in the semiconductor. The

space-charge layer is populated by minority carriers, and given the name majority-carrier

depletion layer.11

The frequency of the field applied to the capacitor can also affect the trap layer in

the semiconductor. Traps within the semiconductor material become filled with carriers

that are attracted to the dielectric/semiconductor interface. When the bias is released, the

traps are emptied. If the frequency becomes too high, the traps do not have sufficient

time to empty and form a charge layer. At low frequencies, this layer is almost

nonexistent. As the frequency is increased, however, this trap layer becomes thicker and

adds to the total capacitance. Capacitance measurements must be made at extremely low

frequencies, called quasi-static frequencies, to obtain the capacitance that is purely within

the dielectric material.

2.1.3 Metal/Insulator/Semiconductor (MIS) Capacitor

Modem silicon technology is based on complementary pairs of metal oxide

semiconductor (CMOS) transistors. This is one of the most common devices found in

logic and memory circuits. The gate region of the transistor determines the capabilities

of the device. The two types of metal oxide semiconductor field effect transistor









(MOSFET) devices are depletion mode and enhancement mode. In the depletion-mode

device, the material type under the source, gate and drain regions is the same. This

device is in the "on" state when no gate voltage is applied. In the enhancement-mode

device, the material type under the gate is opposite that under the source and drain. This

device is in the "off" state when no applied gate voltage is applied. Figure 2-2 shows a

cross-section of a depletion-mode MOSFET. Because of the relatively low p-type carrier

concentration available for p-GaN, only n-type depletion-mode devices were considered.

The n-type MOSFET was used to describe the operation of the gate in the

transistor. For a p-type MOSFET, the gate voltage is reversed. When there is a zero gate

voltage, carriers are free to flow from the source to the drain in the MOSFET structure.

The switch is "on". As a negative voltage is applied to the gate contact, electrons under

the gate dielectric are repelled (like charges repel) and a positive charge is induced in the

semiconductor under the gate region. This positive charge hinders the ability for

electrons to flow from the source region to the drain region. As the gate voltage is

increased, more positive charges collect under the gate until the flow from source to drain

is completely stopped. The voltage is called the pinch-off voltage, since it effectively

pinches the channel shut. The transistor is now "off". As the current through the source-

drain is increased, it requires more gate voltage to successfully pinch-off the carrier flow.

Thus the maximum operating parameters of the device are determined by the amount of

electric field that can be applied to the gate before the dielectric breakdown occurs.

2.2 GaN Based Electronic Devices

Gallium nitride research has resulted in long-lifetime, room temperature operation

of photonic devices. These include LEDs that cover the visible spectrum, laser diodes in

the blue and blue-green regime, and UV detectors. These devices are just recently









reaching production levels with problems still to be solved in the fields of n-Ohmic and

p-Ohmic contacts, p-type doping issues, Schottky contacts, and dielectric materials.

Also, with the lack of availability of high-quality GaN substrates, research of epitaxial

growth and substrate selection is still ongoing. From material and processing advances

learned from the photonics research, high-power and high-temperature switches have

been realized. The following summarizes some of the dielectrics research to date.

2.2.1 Silicon Oxide on GaN

Silicon oxide is a very attractive choice for a dielectric material since is has been

well studied and the processing is well established. Silicon oxide deposited by plasma

enhanced chemical vapor deposition (PECVD)13-16 has been reported to give interface

state densities on the order of low 1011 eV-cm-2. Silicon oxide deposited by electron

beam (EB) evaporation has shown interface state density of 5.3xl011 eV-cm-2.15 After

annealing the EB deposited SiOx at 650C, the valance band offset was measured to be

2eV and the conduction band offset was measured to be 3.6 eV.17 The EB evaporated

SiOx shows a silicon rich stoichiometry when compared to the PECVD SiOx.

There are several inherent problems with SiOx as a dielectric material for wide

bandgap semiconductors. The high Dit of SiOx is due to uncontrolled oxidation of the

surface18. The most significant limitation is that SiOx has a dielectric constant (s) of 3.9,

which is significantly lower than that of GaN (s = 8.9). This will create a very large

electric field in the dielectric, leading to further breakdown.

2.2.2 Silicon Nitride on GaN

Silicon nitride deposition by PECVD15'16 reported an interface state density of

6.5x101 eV- cm-2. This value is reasonable for a first attempt. But when compared to









GaN MESFETs, the Si3N4 MISFET was out performed. Electrical measurements showed

the MISFET structure had a large flat band voltage shift (3.07 V) and a low breakdown

voltage (1.5 MV/cm). There was no microstructural analysis performed on the deposited

Si3N4 films. SiN deposited by ECR plasma CVD showed a Dit of x101cm-2eV-1 but has

excess leakage current due to small conduction band offset.18 The ECR-CVD MIS diode

showed a Dit of 4x1011 cm-2eV-1, fixed oxide charge of 1.lx101 cm-2 and breakdown

5.7 MV/cm with a dielectric constant of 6.19

A unique dielectric structure of Si02/Si3N4/SiO2 (ONO) was reported to have

breakdown field strength of 12.5 MV/cm for temperatures as high as 3000C.20 The ONO

structure was deposited by jet vapor deposition to a thickness of 10 nm /20 nm /10 nm.

The multiplayer structure does allow for unique engineering of a dielectric, but multiple

interfaces can lead to an extremely large number of interface state traps and increased

processing. The Dit for the ONO structure was shown to be less than 5x1010 eV- cm-2

with breakdown fields greater than 12 MV/cm.21

2.2.3 Aluminum Nitride on GaN

Aluminum nitride deposited by MBE and MOCVD has been used to create

MISFET devices and insulated gate heterostructure field effect transistors (IG-HFET)

devices.22'23 The AIN MISFET structure grown at 4000C was polycrystalline. From x-

ray reflectivity measurements, the AIN/GaN interface showed a roughness of 2.0 nm.

This may be due to the polycrystalline nature of the film or from intermixing of the A1N

and GaN at the interface. The dielectric breakdown field was calculated to be

1.2 MV/cm. The AIN IG-HFET structure was grown at 9900C, forming a single crystal

film of 4.0 nm. This device operated in enhancement mode and had a pinch-off voltage









of 0 V. Hexagonal aluminum nitride has a 2.4% lattice mismatch with hexagonal GaN

on the (0001) plane. The 4.0 nm film thickness is greater than the critical thickness

allowed for elastic deformation leading to threading dislocations forming from plastic

deformation. Single crystal AIN and polycrystalline AIN films suffer from defects and

grain boundaries that cause shorting.

2.2.4 Gallium Oxide on GaN

Gallium nitride forms a stable native oxide. This oxide has been considered as a

dielectric material, like the native oxide of silicon. Thermal oxidation of the GaN surface

has lead to some interesting research. Oxidation was performed in dry24'25 and wet26

atmospheres. Dry oxidation of GaN epilayers at temperatures below 9000C showed

minimal oxidation. Dry oxidation at 880C for 5 hours showed 1110 nm of p-Ga203 with

a Dit of xl10 eV-cm-2 and showed inversion.27 At temperatures above 9000C, a

polycrystalline monoclinic Ga203 forms at a rate of 5.0 nm/hr. This oxidation rate is too

slow to be viable as a processing step. Wet oxidation of GaN also forms polycrystalline

monoclinic Ga203, but at a rate of 50.0 nm/hr at 9000C. From cross-sectional

transmission electron microscopy, the interface between the oxide and the GaN is found

to be non-uniform. Scanning electron microscopy shows that both films are rough and

faceted. Electrical characterization of the oxide shows the dry oxide dielectric field

strength of 0.2 MV/cm and the wet oxide dielectric field strength of 0.05 to 0.1 MV/cm.

Some limits to the thermal oxidation of GaN are that only one microstructure has been

formed from this process and GaN is consumed in the process. XRD shows this to be a

high temperature hexagonal phase28. Ga203 passivates the surface,28-31 and has a Dit of

101 eV-cm-2 for GaN MOS. A negative oxide charge as well as high capacitance and









reduced reverse leakage where shown for thicker oxides grown by PEC.29 Using PEC

and HeCd laser, a low reverse leakage current of 200 pA at 20 Vm has been achieved.

For this oxide the forward breakdown, Efb, is 2.8 MV/cm and the reverse breakdown, Erb,

is 5.7 MV/cm with a Dit of 2.53 x 1011 cm-2eV-1. The dielectric constant of Ga203 grown

under these conditions is estimated at 10.6.33 With amorphous GaO deposited by PEC,

low leakage currents of <10 x 10-8 A/cm at -15 V have been measured. Using Ga203 as

both gate dielectric and passivation layer, a breakdown field of 0.4 MV/cm was observed.

The bandgap, Eg, of Ga203 was measured to be 4.4 eV.34

2.2.5 Silicon Dioxide on Gallium Oxide on Gallium Nitride

Depending on the growth technique, the interface between the SiO and the GaN can

vary drastically. When SiO is deposited by RPECVD, a parasitic subcutaneous layer of

native gallium oxide is grown on the GaN surface. This layer as been shown to have a

direct effect on the device performance. When the thickness of the initial GaO layer is

controlled by a pre-oxidation step the device characteristics improve markedly. Remote

Plasma Assisted Oxidation first followed by RPECVD gave a lower Dit and a smaller flat

band shift over the RPECVD of the SiOx alone.35 Real and ideal CV curves are nearly

identical.36 After an anneal in an RTA for 1 min at 900C in Ar the Dit is

2-3x101 cm-2.37'38 Another group used a similar oxide growth technique and measured a

Dit of 3.9x1010 eV- cm-2 and a low leakage current.39'40

2.2.6 Gallium Gadolinium Oxide on GaN

Due to the recent success of gallium gadolinium oxide (GGG) as a dielectric in

GaAs MOSFETs41 46, attention has turned toward this as a dielectric material for GaN.

The GGG dielectric was deposited on a GaN epilayer by EB evaporation of a single

crystal GGG source.22 The substrate temperature was 5500C. The interface roughness









was calculated to be 0.3 nm from x-ray reflectivity. Metal oxide semiconductor (MOS)

capacitors were formed and tested. A breakdown field of greater than 12 MV/cm was

estimated. More recently, the thermal stability of the film and the interface has been

proven to temperatures as high as 9500C and operation of a depletion mode MOSFET has

been performed at temperatures up to 4000C.47 The EB evaporated GGG stoichiometry

is heavily dependant upon the substrate temperature. Changes in temperature lead to

changes in the stoichiometry.48 This limits the available microstructure obtainable within

the stoichiometric limits of GGG.

2.2.7 Gadolinium Oxide on GaN

GaN based MOSFETs have been made that used a stacked gate oxide consisting of

single crystal gadolinium oxide and amorphous SiO2.49 The gadolinium oxide provides a

good oxide /semiconductor interface and the SiO2 reduces the gate leakage current and

enhances oxide breakdown voltages. The dislocations in the Gd203 film limit the

breakdown field that can be sustained in the dielectric.

2.2.8 Scandium Oxide on GaN

Scandium oxide grown by MBE has been used as a gate dielectric and passivation

layer for GaN based devices. Scandium oxide has the bixbyite crystal structure, a

reasonable band gap of 6.3 eV and a lattice mismatch to GaN of 9.2%. The scandium

oxide was grown by MBE using an RF oxygen plasma, substrate temperature of 6500C

and an effusion cell temperature of 11300C. The surface state density is 8.2x1012eV cm-2

and showed inversion for gated diodes.50 This oxide was also grown under the same

conditions except the substrate temperature was lower to 1000C, which resulted in an

interface state density of 5x101 eV- cm-2.5






15


Scandium oxide has also been used as a field passivation layer for GaN devices. It

has been shown to reduce the reverse leakage current an increase the fT and fMAX. The

passivation films have been grown with a substrate temperature of 1000C and a cell

temperature of 1130C.52 Scandium oxide has better long-term stability than SiNx as a

passivation film for GaN based HEMTs. It has been shown to dramatically reduce the

gate lag problems due to surface states on AlGaN/GaN HEMTs.53









Table 2.1. Properties of dielectric materials that have been used on GaN



Material Bandgap (eV) Dielectric (s) Melting Point References
(K)


A1N

A1203

Ga203

Gd203

Ga203(Gd203)

SiOx

Si3N4

MgO

Sc203


8.5

12

10

11.4

14.2

3.9

7.5

9.8

11.4


3273

2319

2013

2668

2023

1993

2173

3073

2678


54,55,56

57,58,59

60,61,62

63,64,65

2,66

12,59,67

12,67,68

59,69,70,71

72


6.2

5.75

4.4

8.5

4.7

8-9

5.0

7.3

6.3











VG

Metal
I I M t Insulator
V


Ohmic contact


DIELECTRIC


SEN IICONDU'TOR u-fl j)C


Figure 2-1. Capacitor diagrams. A) Typical metal-insulator-semiconductor capacitor
with backside ohmic contact. B) Typical planar capacitor produced in our
study.


T
Semiconductor


I












VG=0O


source


gate


drain


+n-GaN -I I. ~




n-GaN





VG gate


source drain


+n-GaN :--




n-GaN


Figure 2-2. Cross-section illustration of a depletion mode n-MOSFET. In the top figure,
the device is in the "ON" state with VG=0. The bottom figure is the device in
the "OFF" state with VG<0, notice the conduction channel is pinched-off.














CHAPTER 3
EXPERIMENTAL APPROACH

3.1 Molecular Beam Epitaxy

Molecular beam epitaxy (MBE) was the growth method employed in this work to

produce the dielectric films. The dielectric materials were deposited in an ultra-high

vacuum environment from the purest attainable elements. In MBE, the individual

elements of the compound are provided to the growth surface independent of each other,

allowing for a high degree of control over the stoichiometry of the dielectric material that

conventional sputtering and electron beam evaporation do not allow. Also, MBE allows

for precise control of the substrate temperature, which in turn helps to control the

microstructure of these materials. The growth rate is dependent upon the substrate

temperature, the ratio of the elements, and also the rate at which the elements are

supplied to the substrate.

In conventional MBE, beams of atomic or diatomic species are produced from

ovens called Knudsen cells (Figure 3-1). The purity of the atomic beam depends upon

the vacuum level in the chamber and the purity of the source material. The number of

atoms emitted from the Knudsen cell is related to the temperature of the cell and the

relative atomic mass of the material in the cell. This relation is modeled by the Equation

3-1 where F is the flux of the Knudsen cell in atoms/cm2-s, p is the vapor pressure in the

cell in Torr, a is the orifice area in cm2, d is the cell to substrate distance in cm, M is the

atomic mass of the element in amu, and T is the temperature of the cell in degrees Kelvin.









F 1.18x1022(p)(a) (atoms (3-1)
d2(MT)1/2 cm S)

For these oxides, a Riber model 2300 MBE was modified to perform the growth. A

sketch of this system is shown in Figure 3.2. The main growth chamber was pumped on

with a cryopump allowing for a base pressure of 2xl0-9 Torr. This modified system is

equipped with a Reflection High-Energy Electron Diffraction (RHEED) system,

described in Section 3.2.1. This allows for the arrangement of the top few monolayers of

atom to be determined. This is extremely important in determining the atomic

arrangement of the starting growth surface and in determining the structure of the film.

The oxygen source for the oxide growth was a WaveMat model 610 electron

cyclotron resonance (ECR) plasma source, (Figure 3-3) operating at a frequency of 2.54

GHz and powers ranging from 100 to 200 watts or an Oxford Applied Research radio

frequency (RF) oxygen plasma source, Figure 3-4, operating at 13.56 MHz with the RF

power set at 300W. Oxygen is supplied to the plasma generator through a leak valve

using a 99.995% oxygen source. The ECR plasma sources works by microwave energy

that is guided into the source chamber and coupled into the oxygen molecule electron

cloud. A series of permanent magnets around the source chamber create a magnetic field,

which accelerates the electron motion into helical paths that collide and ionize the source

gas molecules. This creates a dense plasma in the source chamber. The plasma contains

the atomic species for growth, as well as ionic and molecular species. The RF plasma

source operates by means of an electrical discharge created from inductively-coupled RF

excitation.73 Atoms produced by dissociation in the discharge tube can escape into the

vacuum environment along with the undissociated molecules via an array of fine holes in

the aperture plate. The electrical potentials are such that negligible currents of ions or









electrons escape form the discharge during normal operation of the source. Dissociated

atoms undergoing wall collisions in the discharge chamber exhibit a low recombination

coefficient and may also ultimately contribute to the radical beam flux. The pressure in

the discharge chamber is sufficiently low that atom-gas collisions are minimal over the

dimensions of the discharge chamber.73

The substrate temperature was determined by a backside thermocouple in close

proximity with the substrate holder. The substrate thermocouple was calibrated by using

the melting points of gallium antimonide (GaSb) at 707C and indium antimonide (InSb)

at 5250C. Pieces of GaSb and InSb were heated in the growth position with a nitrogen

plasma impinging on the surface. This reduces the chance for loss of the Group V, Sb,

species during the heating process, which would result in an incorrect melting

temperature. With out the nitrogen over pressure, the InSb and GaSb would degrade by

loss of Sb, the more volatile Group V species, before melting.

3.1.1 Substrate Preparation

Prior to any epitaxial growth, the substrates receive an ex-situ and in-situ surface

treatment to remove any contamination and the native oxide. The surface of the

semiconductor must be as clean and planar as possible to ensure high quality dielectric

film deposition. Surface contamination leads to impurities at the

dielectric/semiconductor interface, which ultimately results in creating interface traps, as

described in Section 2.1.2. The substrates were visually inspected as well as scanned

using atomic force microscopy (AFM), described in Section 3.2.4. This gave a reference

for the surface roughness to compare to the final product. The substrates used in this

work are silicon (Si) and gallium nitride (GaN).









3.1.1.1 Silicon

All of the initial experiments were carried out on Si single crystal substrates

oriented in the <001> direction. This was due to the wide availability of Si substrates and

their low cost. The data gathered from the Si substrates is used to calibrate the growth

rate and composition of the dielectric films. The microstructure of the dielectric films

grown on Si will be different from that on GaN since the surface atomic spacing and the

crystal structure of the substrates are different.

Silicon wafers received as ex-situ treatment of a 30 second wet etch in a buffered

oxide solution consisting of 6 parts ammonium fluoride and 1 part hydrofluoric acid,

rinsed in deionized water (DI), and dried under nitrogen gas. This treatment results in a

smooth surface with a surface roughness root mean square (RMS) value of 0.08 nm, as

seen in AFM. This surface is oxide free and stable for a period of up to one hour. The

in-situ cleaning consists of heating the silicon up to 200C in order to drive off any

moisture from exposure to the atmosphere.

3.1.1.2 Gallium Nitride

Since gallium nitride wafers substrates are not currently available, gallium nitride

grown on sapphire wafers oriented <0001> were used. These will be referred to

throughout this work as GaN substrates. Two different types of growth of the GaN

substrates were employed in this work, MBE and metal-organic chemical vapor

deposition (MOCVD). The MBE GaN substrates were provided by SVT Associates and

the MOCVD substrates were provided by Epitronics, QinetiQ, and Uniroyal

Optoelectronics (UOE). From AFM, there is a large difference in surface roughness

between the two types of GaN. The MOCVD substrates are 1-3 nm RMS roughness and

the MBE substrates are ~6 nm RMS roughness (Figure 3-5).









The GaN substrates received an ex-situ treatment consisting of a 3-minute etch in

(1:1) hydrochloric acid: water, followed by a DI rinse and blown dry by nitrogen. This

was use to remove any organic residue from the surface. Then a 25-minute exposure to

ozone produced by an ultraviolet lamp in a UVOCS UVO Cleaner model number 42-220

was used to oxidize the carbon on the surface and create a thin native oxide layer. Next,

the substrates received another etch in buffered oxide etch for 5 minutes, in the solution

described for silicon, to remove the native oxide. This is shown by observing the

reflection high-energy electron diffraction (RHEED) pattern produced from the surface,

described in Section 3.2.1. The RHEED pattern produced by the native oxide is more

diffuse than the pattern produced by the buffered oxide etched surface (Figure 3-6).

The GaN substrates were then mounted to molybdenum blocks using indium

solder, and then loaded into the load-lock of the MBE. Room temperature RHEED

measurements showed a reasonably clean (lxl) surface (Figure 3-6). Two different

crystal directions are observed in RHEED to create a more complete understanding of the

surface. Here, the <1-100> and the <11-20> directions are observed. An in-situ thermal

treatment was employed to further remove any oxide or contamination left on the surface.

The substrates were heated to 700C in vacuum and no overpressure of nitrogen was

used. The RHEED patterns recorded at this temperature indicate a sharp (1x3) pattern

(Figure 3-7).

3.1.2 Magnesium Oxide Growth

The magnesium oxide samples were grown from 99.99% pure magnesium and

Knudsen cell temperatures ranging from 3500C to 4000C. Substrate temperatures

between 1000C and 3400C were used. The oxygen was provided by an Oxford RF









plasma source at 300 W forward power or an ECR Wavemat plasma source set to 200 W

forward power. There was a significant difference in the properties of the films between

the 2 plasma sources. The RF plasma source produced the superior films. Oxygen

pressure was varied from 8x10-6 up to Ixl105 Torr. In all cases, the sample rotation was

kept at 15 rpm.

3.1.3 Magnesium Calcium Oxide Growth

The magnesium calcium oxide samples were grown from the same Mg sources as

that used for the MgO with the addition of 99.999% pure calcium, Ca, from another

Knudsen cell with temperatures ranging from 4500C to 5000C. Substrate growth

temperatures were between 1000C and 3000C. The two growth methods that were used

are continuous where all shutters open at once and exposed to the substrate and digital

alloying where alternating layers of MgO and CaO. Changing the flux of the metal

sources during a continuous growth or changing the timing of the shutter sequences

during digital growth varied the composition of the film. Oxygen pressure was held at

8x10-6 Torr and used only the RF plasma source. As in the MgO growth, the sample

rotation was kept at 15 rpm.

3.1.4 Magnesium Scandium Oxide Growth

The magnesium scandium oxide films were grown using the same Mg source with

the addition of a Sc metal Knudsen cell with temperatures ranging from 10900C to

11900C due to the extremely low vapor pressure of scandium. Substrate temperatures

were between 1000C and 3000C. Growth methods for this ternary oxide were continuous

and digital as in the MgCaO films. It was not possible to take a flux reading of the

scandium due to severe fluctuations in readings, the needle bounced around, when the









cell at temperature due to out-gassing of He from the Sc source metal. The out-gassing

of He was shown by the mass spectrometer. The composition of the film was varied by

changing the temperature of the source metals. As a convention, films with a higher

amount of scandium then magnesium are referred to as scandium magnesium oxides,

ScMgO. Films with a greater amount of magnesium than scandium are referred to as

magnesium scandium oxides, MgScO. Oxygen pressure was held at 8x10-6 Torr and

used only the RF plasma source. As in the MgO growth, the sample rotation was kept at

15 rpm.

3.2 Materials Characterization

The films were heavily characterized after growth. Emphasis of the research was

placed on, but not limited to, the microstructure and the stoichiometry of the epitaxial

films and how these properties related to the electrical properties, environmental and

thermal stability of the dielectric materials. The dielectric films were annealed to

temperatures as high as 10000C by a rapid thermal anneal (RTA) process to determine the

thermal stability of the films.

3.2.1 Reflection High Energy Electron Diffraction: RHEED

In-situ structural characterization can be done in the growth chamber via reflection

high-energy electron diffraction (RHEED). A RHEED system consists of an electron

gun, typically 5 to 30 kV, and a phosphorescent screen. The electrons from a filament

are collimated, accelerated, and reflected off the surface of the sample. A diffraction

pattern is seen on the phosphorous screen. From this diffraction pattern, single crystal,

polycrystalline and amorphous films can be differentiated. This technique was used to

determine the starting substrate surface quality and the quality of the films grown while









in the ultra-high vacuum system. Also, the method of growth initiation, which has an

enormous impact on the overall film quality, can be determined from RHEED.

RHEED reflections are created by diffraction from the surface of the substrate. The

incoming electron beam has a grazing incident angle of 1 to 2 degrees. Diffraction

occurs only along certain crystal directions in a single crystal material. From the type of

pattern, intensity, and spacing between different diffraction events, a 2-dimensional map

of the surface can be created. It is this map that will help determine the condition of the

starting substrate as well as of the grown film. A surface growing layer-by-layer (2D)

will produce a pattern with streaky lines, whereas a surface growing by islanding (3D)

will produce a pattern that is spotty. Polycrystalline surfaces show a ringed pattern and

amorphous surfaces show almost no pattern at all (Figure 3-8).

3.2.2 Transmission Electron Microscopy: TEM

One of the most powerful microstructural analysis techniques available is

transmission electron microscopy (TEM). From TEM, not only can the microstructure of

an epitaxial film be determined, but also detailed analysis of defects in the film, atomic

imaging of the interface, and an accurate calculation of lattice spacing is determined.

TEM uses a beam of electrons that pass through and interact with a very thin sample to

form an image on the other side of the sample (Figure 3-9). The interactions between the

atoms in the sample and the electrons produce the contrast seen in the image. One of the

major drawbacks of TEM is sample preparation required to obtain the images. The

sample must be cut, polished, and thinned to electron transparency (-100 nm) via hand

polishing and ion beam milling or by using a focused ion beam (FIB) system. A

complete description of the sample preparation is given in Appendix B. This is

especially difficult for the nitride materials due to their hardness. The FIB used to make









these sample was a FEI Strata DB (Dual Beam) 235 FIB. A JOEL 200CX operating at

200 keV was used for film analysis and JOEL 2010FX operating at 400 keV was used for

high-resolution analysis of the interface.

3.2.3 X-Ray Diffraction: XRD

Another structural analysis technique is x-ray diffraction (XRD). This technique

had virtually no sample preparation when compared to TEM. X-rays are diffracted off

the sample to produce characteristic peaks of the atomic planes in the sample. The full

width at half maximum (FWHM) of these characteristic peaks is used to determine the

crystalline quality of the films. Powder x-ray diffraction can be used if the samples are

polycrystalline or polycrystalline/amorphous. This was used in preliminary sample

analysis to determine if second phases were present. X-ray reflectivity (XRR) from the

air/film interface and the film/substrate interface help determine the roughness of these

interfaces and the thickness of the film. The powder system is a Phillips 3720. The high-

resolution system is a Phillips MPD 1880/HR with a 5-crystal analyzer. The x-ray source

for both systems is a copper (Cu) Ka x-ray source. Figure 3-10 shows an illustration of

the sample geometry for the x-ray diffraction. Samples were scanned measuring the Q-

20 with the GaN peak optimized for the high-resolution system. For the powder system,

peak intensity- 20 was measured. For the x-ray scan, the 20 peak positions are obtain

and using Bragg's Law, Equation 3-2, the d-spacing between the corresponding planes

are calculated.

nA= 2dsin (3-2)

Here, X is the wavelength of the incident x-ray, d is the spacing between the planes,

and 0 is the measured peak position. The d-spacings are compared to known values of

the material to determine crystal orientation.









3.2.4 Atomic Force Microscopy: AFM

The surfaces of the grown films were characterized using atomic force microscopy

(AFM) to give a quantitative analysis of film morphology or surface roughness. Tapping

mode AFM was used to obtain a root mean square (RMS) roughness of the surface. In

tapping mode, the tip of a stylus, made from Si3N4, is brought into close proximity to the

surface, close enough to be defected by van der Waals forces of the surface atoms. A

laser is reflected off the AFM tip and collected into a photodiode (Figure 3-11). The

intensity of the reflected light is read as height. The tip is rastered across the surface and

each point is read as a height, creating a 3-dimensional map of the surface. From this 3-

dimensional map, a surface roughness is calculated and from this a RMS roughness. This

is very useful in characterizing the starting substrate, the dielectric film, and the effects of

the various growth and processing steps. The sensitivity of the AFM is largely dependent

on the sharpness of the tip and the sensitivity of the deflection. The tapping mode tip

used here has a tip radius of 5 nm and the deflection sensitivity is about 0.01 nm. This

makes tapping mode extremely sensitive to surface roughness. An alternative mode of

operation is contact mode. However, the tip radius is about 20 nm, which greatly reduces

the resolution. The AFM used in this study was a Digital Instruments Nanoscope III.

3.2.5 Scanning Electron Microscopy: SEM

The surface morphology of the samples is characterized on the macro level by

using the secondary electron microscope (SEM). The SEM technique enables an image

of the surface to be taken at very high magnifications, from between 5000X to 100,OOOX.

In this technique, an electron beam of energy between 5 keV and 30 keV is rastered

across the sample surface. A schematic of the SEM is shown in Figure 3-12. One

requirement of an SEM sample is that the surface be conducting in order to prevent









surface charging due to interaction of the beam with the sample. In the case of insulating

samples a thin carbon film is applied or beam conditions are used that reduce surface

charging. The interaction of the beam with the sample produces several different species

at differing depth within the sample including secondary electrons, Auger electrons,

backscattered electrons, and characteristic x-rays. Secondary electrons have energies

below 50 eV, and due to this low energy can only escape from the sample if they are

produce within a few nm of the surface. As the electron beam is rastered, a detector

picks up the secondary electron signal. This signal is fed into a cathode ray tube that is

scanned at the same rate as the beam, producing an image (Figure 3-13).

The SEM used in this research is a JOEL 6400. The technique gives a qualitative

analysis of the surface, indicating the overall surface morphology of the film. This is

important for future processes in fabricating capacitors and MOS(MIS)FET's, since

device processing requires annealing, etching, and metal deposition, all of which are

sensitive to surface morphology. SEM can also be used to look at the surface topography,

which should determine how well the surface is covered and see any obvious defects and

pin holes.

3.2.6 Auger Electron Spectroscopy: AES

Auger electron spectroscopy (AES) was used to determine qualitatively the

elements present in the grown dielectrics. Auger electrons are also emitted from the

sample during the electron/sample interaction (Figure 3-14). An incident electron strikes

an inner shell electron of an atom and ejects that electron. An upper shell electron fills

the void and energy is given off in the form of an Auger electron. The Auger electron is

specific in energy to the element it came from and is thus a characteristic electron to that

element. The Auger electrons are collected and an elemental analysis of the surface is









obtained. The AES technique can detect elements down to the alloy level (-1%) within

the top 1.0 nm of the surface. Auger electrons are produced throughout the interaction

volume of the incident electrons, however, because of their low energies, only those

produced near or at the surface can escape. From the ratio of the peak heights from each

element and published sensitivity factors, an approximate ratio of elements can be

determined. A Perkin Elmer 660 AES was used for these measurements. This system is

also equipped with an ion gun for creating depth profile Auger electron spectra. From

this, changes in the element ratios perpendicular to the interface and the interface itself

can be studied. This will help in determining if film composition is constant throughout

the film and if there is segregation in the film. Also, approximate film thickness can be

determined from known standards.

3.2.7 Ellipsometry

Ellipsometry is used primarily to determine the thickness of thin dielectric films on

highly absorbing substrates but can also be used to determine the optical constants of

films or substrates. Ellipsometry is based on measuring the state of polarization of

polarized light. When light is reflected from a single surface it will generally be reduced

in amplitude and shifted in phase. For multiple reflecting surfaces, the various reflecting

beams will further interact and give maxima and minima as a function of wavelength or

incident angle. Since ellipsometry depends on angle measurements, optical variables can

be measured with great precision, being independent of light intensity, total reflectance,

and detector-amplitude sensitivity. A general schematic of the machine is in Figure 3-15.

For further explanation of ellipsometry see Appendix A. The ellipsometer used for this

experiment is a JA Woollam Variable Angle Spectroscopic Ellipsometer.









3.2.8 Current-Voltage (I-V) measurements

Current-voltage measurements were made using a Hewlett Packard Model 4145.

In these measurements, the current is set to an upper and lower limit, typically 5[A

tolOA and the voltage is swept from negative to positive. The voltage range is

increased until the forward and reverse breakdowns are reached. The current limit is then

set to ImA and the voltage is measured again. The voltage at 1 mA is divided by the

dielectric thickness and the breakdown field strength is obtained. This is one parameter

used in defining the quality of the dielectric film. This helps to determine the breakdown

field of the dielectric at elevated temperatures. There is a heated stage with the current-

voltage measurement equipment with a maximum temperature for testing of the 300C.

This was used to study the dependence of the temperature with the breakdown of the

dielectric.

Ohmic contacts were made to the silicon substrates using a Pt/Au (300 A /1000 A)

bilayer structure using electron beam evaporation. Ohmic contacts made to the gallium

nitride were made using a multiplayer structure of Ti/Al/Pt/Au with the following

thicknesses- 20 nm Ti /70 nm Al/40 nm Pt /100 nm Au. The contacts on the GaN were

annealed for 30 seconds at a temperature of 450C in a nitrogen ambient. Contacts on the

dielectric were Pt/Au and were deposited through a shadow mask with varying contact

sizes. The most commonly used contact sizes were 50 [tm and 80 am.

3.2.9 Capacitance-Voltage (C-V) measurements

Capacitance-voltage measurements were made using a Hewlett Packard Model

4284. Here a bias of 2 to 20 volts is applied across the capacitor and cycled at a selected

frequency, typically 100 Hz to 1 MHz, and the resulting capacitance is recorded. From

capacitance-voltage plots, the flat band voltage shift, dielectric constant, and interface






32


state density can be calculated. Calculating the carrier concentration of the substrate

from the data obtained can test the accuracy of these measurements. This carrier

concentration should match the quoted value from the manufacturer.












Atoms or atom
clusters






PBN crucible

Source material

Ta Heater element



Thermocouple


Figure 3-1. Typical Knudsen effusion oven (After B.P. Gila 2000).











PLASMA
SOURCE



















Buffer chamber


SOLID
SOURCES


REED Gun


Load/lock


Figure 3-2. Riber MBE used for oxide growth.












Active species


Tuning antenna


Permanent magnets


Source gas inlet


Cooling gas inlet and
microwave power lead


Figure 3-3. WaveMat 610 ECR plasma head.















PBN
.Apeture
Photodiode A, tu e
P Water Cooled RF Coils Plate
Optical Filter







SaoDhire Window F l Discharge Tube
Flange Aperture
Retaining
r Ring

Tuning R__ -g RF Coupling
Knob




Location Button
Retaining Knob

Rf Tuning Box

Figure 3.4. Schematic of the Oxford RF plasma source (After MDP21S Operating and Service Manual 198973).





































A 0.8 X 0.200 pM/div
2 10.000 nM/div





















0.2

0.4

0.6 /

0.8 U X 0.200 jM/div
B 2 50.000 nM/div



Figure 3-5. AFM images of as received A) MOCVD GaN from Epitronics, and B) as
received MBE GaN from SVT.











































Figure 3-6. RHEED images showing A) the UV-ozone treated surface of MOCVD
grown GaN and B) the buffered oxide etched UV-ozone surface of GaN.













































Figure 3-7. RHEED photos indicating a (1x3) pattern. The top photo is along the
<11-20> crystal direction and the bottom photo is along the <1-100> crystal
direction.






40















A














B














C

Figure 3-8. RHEED photos showing A) an amorphous diffraction pattern, B) a
polycrystalline diffraction pattern, and C) a single crystal diffraction pattern.











Electron Source


Condenser Lens




Specimen



Objective Lens









Back Focal Plane


First Intermediate Image


Intermediate Lens



Second Intermediate Image


Projector Lens







Image


Figure 3-9. TEM column.






42




Diffracted x-rays Incident x-rays





Osub Ofilsm film ub


Film



Substrate




dsub < dfilm Osub > Ofilm

Figure 3-10. The relation between the lattice parameter and Bragg angle for film and
substrate.












A way of sensing the
vertical position of the tip


A feedback
system to
control the
vertical position
of the tip


A piezoelectric scanner
which moves the tip over
the sample (or the sample
under the tip) in a raster
pattern


, A snarp tip



Sample


A course positioning system to
bring to tip into the general
vicinity of the sample.














A computer system that
drives the scanner,
measures data and
converts the data into an
image.


Figure 3-11. Atomic force microscope (after K.K. Harris 2000).


0


I---------------









Optic Axis


Spot Size


Figure 3-12. Schematic of SEM column.


Electron Source



Initial Beam Size











Condenser Lens


Intermediate
Beam Size










Objective Aperture



I Objective Lens












Electron
source

---- DEFLECTION




election
o l W coils
Detector giving
modulating signal

/... Secondary Cathode ray
./ electron signal tube




S;iinll)e Screen




Figure 3-13. SEM operation. Electron beam is rastered over the sample producing
secondary electrons (after S.M. Donovan 1999).



















Sample Surface
50-500A


E = Ec


Continuum -
X-Rays


IB.S.E. Resolution I
I I

X-Ray Resolution


Figure 3-14. The penetration depth and interaction of an electron beam in a material.
Notice that Auger only escape the top 1.0 nm (after Goldstein).















Unpolarized


Linearly
Polarized


Elliptically
Polarized


Extinguished


Linearly
Polarized


Analyzer


Compensator


Figure 3-15. Ellipsometer schematic (After D.K Schroder74)


Light
Source


Detector


Polarizer














CHAPTER 4
MAGNESIUM OXIDE: RESULTS AND DISCUSSION

Magnesium oxide was chosen as a possible dielectric material for GaN because of

its high melting point (28500C), large band gap (7.3 eV), large dielectric constant (9.8)

and good band offsets with GaN. The structure of MgO is the NaCl crystal structure75

(Figure 4-1). This is a cubic structure with a lattice constant of 4.20 A. The symmetry

alignment for MgO:GaN is the MgO (111) and the GaN (0001). The cation spacing for

the MgO (111) plane is 2.97 A and for the GaN (0001) plane is 3.19 A which gives a

mismatch of -6.9% (Figure 4.2).

4.1 Effect of Oxygen Plasma Source

Many factors affect the growth of MgO such as oxygen source, oxygen pressure,

substrate temperature, and plasma power. Initial work was carried out using an ECR

plasma head for the oxygen source. Samples grown using a substrate temperature of

350C and a cell temperature, TMg, of 340C, showed spotty RHEED diffraction patterns

(Figure 4-3). As seen in this series of images, the pattern changed from streaky pattern,

indicative of the single crystal GaN surface, to a broken line pattern indicative of a

roughened surface. The surface diffraction also showed 6-fold symmetry indicating that

the MgO grew with the (111) plane parallel to the GaN (0001) basal plane. During

growth the pattern changed slowly until at the end of 120 minutes of growth, the pattern

was highly three-dimensional and showed arcs associated with a polycrystalline pattern.

This indicates that the initial layers were single crystal and that the microstructure

gradually changed to polycrystalline as more layers were deposited. Using a

48









profilometer, the film thickness was measured to be 150 nm, yielding a growth rate of

-1.25 nm/min. From XRD results, the MgO film was found to be cubic and oriented

with the (111) direction normal to the surface. Needle-like structures were observed on

the surface by SEM (Figure 4-4) and the sample surface had an AFM root-mean-square

(RMS) roughness of 4.07 nm (Figure 4-5). These needle-like features were found to

decrease with increasing film thickness.

Samples grown at a substrate temperature of 100C using the ECR plasma head

were significantly different from those grown at 350C. Initiating growth on the same

GaN surface, the films remained single crystal for 1-2 minutes (-2.5-5.0 nm), then a

polycrystalline RHEED pattern began to form, Figure 4-6. This pattern remains for the

duration of the growth. The polycrystalline RHEED pattern shows six-fold rotational

symmetry, indicating that the film is textured towards the <111>. This polycrystalline

pattern differs from the pattern obtained at 350C, in that the arcs are more pronounced

and the spots are not visible. This suggests a less textured film with a smaller grain size.

From an etch step measurement, a growth rate of 2.5 nm/min was calculated. The sample

surface has an AFM RMS roughness value of 1.26 nm and a reduced number of needle-

like features. These needle-like features are not visible on films with a thickness of 100

nm or more. From XRD measurements, the polycrystalline MgO films were confirmed

to be highly oriented toward the (111) direction.

Diodes were fabricated from these materials to measure the electrical properties of

the dielectric film and the dielectric/GaN interface. Ohmic windows were created by

etching the MgO with dilute phosphoric acid. Ohmic contacts were made by e-beam

evaporating Ti/Al/Pt/Au. The gate metal was Pt/Au with contact sizes ranging from 100-









50 am. Because of the leaky nature of the single crystal MgO grown at temperatures of

350C, no C-V results could be measured. The leakage in these films could be due to the

needle-like microstructure. For the polycrystalline MgO grown at 100C, a forward

breakdown field of 2.3 MV/cm was measured at a current density of 5.1 mA/cm2. From

C-V measurements a corresponding interface state density of 4 x 1011 cm-2eV1 was

calculated for the diode using the Terman method. It is clear that the different

microstructure obtained at the 100C is critical to improving the electrical behavior of the

MgO dielectric. The combination the single crystalline interface and the polycrystalline

material on top appears to produce a low interface trap density while at the same time

eliminating the shorting paths obtained in the films which are mostly single crystal but

highly defective.

When an RF plasma head was used as the oxygen source for MgO growth the

materials properties improved further. For samples grown under the same conditions but

with an RF plasma oxygen source, the needle-like microstructure was not observed.

SEM and AFM also show a smoother surface, Figure 4-7. For these reasons, all further

growths used the RF plasma head for the oxygen source and the ECR plasma head was

removed from the system.

4.2 Effects of Oxygen Pressure

To study the effect of oxygen pressure on the properties of the growth of MgO,

several different pressures were investigated- lxl105, 3x105, 7x105, and Ixl0-4 Torr as

measured by the beam flux monitor. The MgO flux was held constant. The higher

oxygen pressures produced a decrease in the growth rate (Figure 4.9). From AES, the

films were shown to contain only Mg and oxygen, with the Mg/O decreasing from 0.72

to 0.63 as the pressure was increased, (Figure 4.8). By comparison, the ratio of peak









heights of a standard single crystalline sample of MgO was measured and showed the

ratio to be 0.60. The reduction in growth rates at higher pressures may be an indication

of a reduction in the concentration of reactive oxygen species at the surface or of site

blocking due to the higher concentration of oxygen incident on the surface. Both cases

would result in a reduction in the Mg sticking coefficient. The lower Mg/O ratio at the

higher pressures would seem to favor the site blocking explanation. AFM analysis

indicated that as the pressure was increased the surface morphology became smoother, as

evidenced by the decrease in RMS roughness from 0.998 nm at a pressure of Ixl0-5 Torr

to 0.247 nm at Ix10-4 Torr, (Figure 4-10). All of the films appeared smooth when

examined by SEM. XTEM of the MgO grown at the lowest pressure showed that the

initial 40 monolayers were epitaxial, with the remainder of the layer appearing to be fine-

grained polycrystalline, (Figure 4-11). The precise microstructures of the films grown at

higher pressure are not yet known. It is quite likely that given the superior morphology,

these films retain their single crystal nature for a greater percentage of their thickness

before becoming poly-crystalline.

From structural and compositional analysis it would appear that the higher

pressures are beneficial to the growth of MgO layers. However, electrical

characterization of the MgO/GaN diodes suggests the opposite, (Table 4-1). The

breakdown field, Vbd, and the interface state density, Dit, improve with decreasing

pressure. Ironically, the reduction in the dielectric strength may be due to the superior

but not perfect microstructure in the films grown at higher pressures. Previous work with

other dielectrics such as Gd20349 has shown that if the layer does not contain a substantial

polycrystalline region, then the breakdown field will be substantially lowered due to









leakage through the defects, which propagate through the layer. The presence of a

nanocrystalline layer on top of the single crystal material at the interface appears to

improve the breakdown strength of the layer in spite of the presence of numerous grain

boundaries.

The effect of the pressure on the interface and the bulk charge densities suggests

that the electrical behavior of the layer is enhanced by the presence of higher ion energy

species at the surface. Since the total power to the plasma remains constant, increasing

the oxygen flow will decrease the average energy per ion, and possibly decrease the

concentration of ionized oxygen species as well. Studies with ECR plasmas suggest that

the average ion energy is a critical parameter. ECR plasmas typically exhibit very low

ion energies. MgO films grown using an ECR plasma with similar oxygen pressures to

those in the RF grown films exhibit breakdown fields which are up to four times lower

than those obtained with the RF plasmas. As the ion energy is increased, damage of the

interface will eventually become a factor and begin to increase the density of interface

states. Clearly, however, this does not occur at the standard pressures and powers used in

this study, making the RF plasmas the optimum choice for the deposition of MgO

dielectric on GaN.

4.3 Effect of Substrate Temperature

For MgO samples grown with the RF plasma at 300C, XRD shows a sharper peak,

which indicates that the film grown at a substrate temperature of 300C is more

crystalline than that grown at a substrate temperature of 100C, (Figure 4-12). AFM of

films grown at 300C are rougher than films grown at 100C, (Figure 4-13). The RMS

roughness of the 300C film is 1.998 nm while the 100C grown samples have RMS

roughnesses of 1.334 nm. AES shows that the Mg/O ratio of the films deposited at 100C









is 0.75 while the 300C grown samples have an Mg/O ratio of 0.66. Single crystal MgO

has a Mg/O ratio of 0.60 when investigated in the Auger system. This shows more

incorporation of oxygen at a substrate temperature of 300C under the same oxygen flux

and Mg flux. TEM of MgO grown at 300C (Figure 4-14) shows that the films is still

epitaxial and is single crystalline for a greater thickness before rotation as compared to

MgO grown at 100C, (Figure 4-11).

4.4 Scandium Oxide Capping Layer

Despite the positive aspects of MgO, it has been found to be environmentally

unstable as shown by SEM, (Figure 4-15). The MgO films degrade over time in

atmosphere due to the presence of water vapor.76'77 From device processing, it does not

appear that the MgO under the metal contacts degrades, only the areas exposed to

atmosphere, (Figure 4-16). Capping layers of scandium oxide of various thicknesses

have been tried in order to prevent this degradation. AFM shows that the capping layer

smoothes the surface for both MgO grown at 100C and 300C. The RMS for MgO

grown at 300C goes from 1.921 nm uncapped to 0.337 nm with a Sc203 cap of 20 nm,

(Figure 4-17). For the MgO grown at 100C the RMS goes from 1.539 nm uncapped to

0.869 nm with a 20 nm Sc203 cap. The effect of the capping layer on the environmental

stability of MgO is discussed further in Chapter 7.






54


Table 4-1. Electrical characterization of MgO/GaN diodes. VBD was the applied voltage,
which produced a leakage current of 1 mA/cm2. Dit was the defect value at
0.4 eV below the conduction band calculated using the Terman method. The
diode grown at Ixl0-4 Torr was too leaky to be measured.


Oxygen Pressure VBD Dit
(Torr) (MV/cm) (eV-cm-2)

1xl0-5 4.4 3.4x1011
3x10-5 4 7.1x101
7x10-5 1.2 1.8x1012










o-0
Mg -





[010]


Figure 4-1. The MgO structure (from Cullity 197875)










B\/B
0A A
CB\C
A A


(111)


(0002)


FCC


HCP


Figure 4-2. Illustration showing the symmetry between the (111) NaCi plane and the
(0002) wurtzite plane. The spacing between atoms marked A is 2.97A for
NaCl MgO and 3.19A for wurtzite GaN. (from Cullity 1978)



















A)











B)











C)


Figure 4-3. RHEED images indicating A) GaN at 3500C before growth, B) MgO after 1
minute of growth, and C) MgO after 120 minutes of growth at 3500C.







.,



6


/


V
'a


p


S


I


. ,, .... .


Figure 4-4. SEM image of the MgO surface grown at 3500C with ECR plasma source.
Scale bar is 5pm.


r<


On
i.


M"


...'." 0


t


SJ .....


























A) B)


Figure 4-5. AFM images of A) MgO grown at 3500C (4.07 nm RMS) and B) MgO
grown at 1000C (1.26 nm RMS). Both films are grown with the ECR oxygen
plasma source. Images are 5jtm per side.


















A)














B)





Figure 4-6. RHEED images indicating MgO after A) 1 minute of growth, and B) after
90 minutes of growth at 100C.




































1.00





0.75





0.50





0.25


0 0.25 0.50 0.75


1. 00 Ji


Figure 4-7. SEM, above, and AFM, below, of MgO grown at 1000C with RF plasma
source, RMS = 1.334 nm.


jeol Sri )-()kv X-10,000 11mi WI ) 14.4rilm












0.78-

0.76-

0.74-

0.72-

0.70

0 0.68-

0.66 \

0.64 \

0.62-

0.60- i i
lx10-5 3x10-5 7x10-5

Oxygen Pressure (torr)

Figure 4-8. Mg/O ratio as determined by AES as a function of oxygen pressure.






63

250

. 200

150

- 100

2 50

0
1x10-5 3x10-5 7x10-5
Oxygen Pressure (torr)

Figure 4-9. Dependance of growth rate on oxygen pressure


1x10-4







64




1.00




0.75




0.50




0.25


A)

0
0 0.25 0.50 0.75 1.00nM






0.75




0.50




0.25

B)


0
0 0.25 0.50 0.75 1.00pm


Figure 4-10. AFM scans of MgO grown at 100C using an oxygen pressure of A) Ixl05
Torr, or B) Ixl0-4 Torr. The RMS roughnesses were 0.998 nm and 0.247 nm
respectively.


























MgO


, Interface
^-^*^^ ^Aae--


Figure 4-11. SEM and TEM of MgO. An SEM image (10,000x) of the MgO layer
grown at Ixl05 Torr is shown above and an XTEM image of the same layer is
shown below.


jeol SLI !).OkV X10.000 Ilym WU M.Arnrn







66





Top MgO grown at 300C
Bottom MgO grown at 100C










0






I I I
74 76 78 80
2-0 (degrees)


Figure 4-12. XRD shows that the MgO film grown at 300C has a much sharper peak
than the MgO film grown at 100C.
















0..2 0.75 75




0.50




U.25


A)

0 0.25 0.50 0.75 1.0 [m


1.00




0.75









0.25


B)

0 0.25 0.50 0.75 1, 00 um

Figure 4-13. AFM of MgO A) grown at 1000C with and RMS roughness of 1.334 nm,
and B) grown at 3000C with an RMS roughness of 1.998 nm.



































Figure 4-14. HRTEM of MgO grown at 300C.














7 N
Bifm^^S


Figure 4-15. SEM of degraded MgO film.















1 .0xi 07


8.0xl 08


6.0xl 08

- 4.0x108


o 2.0x108


0.0


-2.0xl 0-8


as-grown
aged w/gate
aged w/o gate


-4 -3 -2 -1 0
voltage (V)


1 2 3


Figure 4-16. I-V and C-V of MgO/n-GaN diodes, agedl5 weeks with and without
processing


I I I I I
-40 -20 0 20 40
voltage (V)


= as-grown
-e- aged w/gate
- aged w/o gate


5.00E-01 2-



4.00E-01 2-



3.00E-01 2-



2.00E-01 2-



1.00E-012-







71



5.00









2.50







A)
0
0 2.50 5. 00 UM


5.00









2.50







B)
0
0 2.50 5. 00 Jm

Figure 4-17. AFM of MgO with and without a capping layer. A) MgO without a cap,
RMS = 1.921 nm, B) MgO with a 20 nm Sc203 cap, RMS = 0.337 nm. The
MgO is grown at 300C. The Sc203 cap is grown at 100C.














CHAPTER 5
MAGNESIUM CALCIUM OXIDE: RESULTS AND DISCUSSION

It is expected that by lattice matching the oxide to the GaN, the interfacial trap

density can be reduced. One method of accomplishing this is to alloy the MgO with an

oxide of larger lattice constant. Calcium oxide was chosen for this purpose because CaO

is also a rock salt dielectric, (Figure 5-1) like MgO, but has a larger lattice constant, 4.779

A. The CaO dielectric constant (11.8) and bandgap (7.1eV) are similar to those of MgO.

From Vegard's Law, the optimum ratio of Mg:Ca should be 1:1 in order to get material

lattice matched to GaN. Though the phase diagram78 shows poor miscibility between

MgO and CaO, (Figure 5-2) single-phase thin films on GaAs whose compositions span

the entire composition range have been reported using LT-MBE as the deposition

method.79

5.1 Growth of CaO

First, to prove that CaO can be grown by MBE, starting with conditions similar to

those used for the growth of MgO, CaO films were grown on GaN. The solid metal Ca

source washeld at temperature of 405C to achieve a flux of 8x10-8 Torr. The same

fluxwas used for the growth of MgO. The RF oxygen plasma was set at a pressure of

2x105 Torr. The substrate temperature was 100C or 300C. AES shows that the film

was comprised of Ca and 0, (Figure 5-3) but XRD showed no evidence of CaO or Ca

diffraction peaks. TEM confirmed that the film was poly-crystalline at the interface but

amorphous after 2 nm, (Figure 5-4). The amorphous nature of the bulk of the film

explains the absence of peaks in XRD. The CaO film is very sensitive to the electron

72









beam in the TEM making it difficult to obtain good images. The growth rate of CaO at

100C was 25.8 A/min, while at 300C the growth rate was 2.2 nm/min. This is to be

expected since an increase in temperature should decrease the sticking coefficient. AFM

shows no change in RMS roughness of CaO grown at 100C versus 300C but stays

constant at about 0.667 nm, Figure 5-5 and Figure 5-6. Unfortunately, CaO etches in

deionized water, which makes processing difficult so no devices were made with CaO.

5.2 Growth of Ternary MgCaO

The standard MgO growth conditions that have produced oxide/GaN interfaces

with low Dit consist of a Mg beam equivalent pressure, BEP, of 10x10-8 Torr. The

addition of Ca to this beam at a comparable Ca BEP of 8x10-8 Torr, produced an increase

of less than 50% in growth rate. This suggests that the sticking coefficient of the Ca is

significantly lower than that of the Mg. This is further confirmed by AES analysis,

which shows an Mg/Ca ratio less than that expected for a 50/50 composition film, (Table

5-1). In addition, AES depth profiling analysis shows that the Ca has severely segregated

to the surface, (Figure 5-7). This would also indicate a low Ca sticking coefficient.

In spite of the apparent segregation of the Ca, XRD analysis of the MgCaO layer

shows no evidence of phase separation, (Figure 5-8). MgO layers grown under similar

conditions typically show primarily a (222) peak due to the texturing of the film. The

MgCaO layer shows no evidence of either the MgO or the CaO (222) peaks suggesting

that phase separation into the two binaries has not occurred. Instead there appears to be a

shoulder on the GaN (004) peak that is not observed in spectra taken from either GaN

substrates or MgO layers grown on GaN. This peak is most likely the (222) peak from

the ternary MgCaO. The peak position is shifted to larger plane spacing relative to the

MgO peak as would be expected from the addition of Ca, (Figure 5-9). The proximity of









this peak to the GaN (004) peak is encouraging and suggests that the addition of Ca may

be useful in reducing the lattice mismatch between the dielectric and the GaN.

Unfortunately due to the severe segregation, the peak is broadened indicative of a range

of compositions present in the film, (Figure 5-10).

AFM analysis of the ternary grown at the highest flux rates shows a very rough

surface morphology, (Figure 5-11). MgO layers grown under similar conditions show

smooth morphology with an RMS of -1 nm, suggesting that the Ca addition has

dramatically altered the microstructure. It was thought that the higher growth rate used

for the deposition of the ternary might have caused the rough morphology. To

investigate a lower growth rate, the fluxes were reduced to BEP ~ 5.7x10-8 Torr. This did

produce a substantial drop in the growth rate to ~1 nm/min and was successful in

improving the surface morphology. Reducing the growth rate did not, however, suppress

the Ca segregation. In fact there appears to be even greater disparity in the Ca surface

and interface concentrations. This is not surprising since lower growth rates are usually

found to enhance segregation.

Substrate temperature also affects the growth of MgCaO. At a substrate

temperature of 340C, no film was grown even after 20 minutes of growth time as shown

by AES, (Figure 5-12). A slightly lower substrate temperature of 300C did result in

deposition of a film as determined by AES, (Figure 5-13). This film has a reasonable

growth rate of 51.6 A/min, and enhanced crystallinity as shown by XRD, (Figure 5-14).

Though for this substrate temperature, AFM does show a slightly rougher surface than

that produced with a substrate temperature of 100C, (Figure 5-15). Due to









improvements in film quality with the increased substrate temperature of 300C, all

further growths were done at a substrate temperature of 300C.

In order to reduce the segregation, a digital growth technique was used. This

showed superior compositional and structural properties as compared to the continuous

growth. To study the effects of growth technique on the characteristics of MgCaO

deposited on GaN, several films were grown under different conditions. Initially

conditions were set so that the Mg and Ca fluxes were equal. The timing sequence was

10 seconds Mg followed by 10 seconds Ca (10/10) with continuous operation of the

oxygen plasma. The intended thickness of each layers was 3 A and the substrate

temperature was 3000C. The resulting films were Mg-rich, which showed that Ca has a

lower sticking coefficient than Mg. Subsequent samples were grown using a

progressively higher Ca flux in order to incorporate more Ca into the film and reduce the

lattice mismatch. Digital samples were grown at the same fluxes and oxygen pressures as

the continuous samples. AFM shows that the digital samples have a slightly smoother

surface than the continuous samples. Also the growth rate of the continuous samples is

about twice that of the digital samples, which was expected. Because of a combination of

the growth rate and the growth sequence, the digital samples all showed a much more

uniform depth profile in AES especially near the surface, (Figure 5-16). In the

continuous samples there is a dip in the oxygen concentration near the surface as well as

in the Ca profile. The ratio of oxygen to total metal shows a greater incorporation of

oxygen in the digital samples than the continuous samples, (Table 5-2).

Powder XRD shows no oxide peaks other than those expected for the MgCaO. The

oxide (222) peak is found to shift toward the GaN peak as the amount of Ca incorporated









into the film is increased for both the digital and the continuous growth sequences,

(Figure 5-9). This peak position is the same for either growth method grown under the

same fluxes, (Figure 5-17). At the highest Ca concentration, the lattice mismatch has

been reduced from -6.5% for MgO to -2.05% for the ternary, (Table 5-3). High-

resolution XRD shows a full-width-at-half-maximum (FWHM) of 3542 arcseconds,

(Figure 5-17). Though this is substantially higher than the GaN FWHM of 507

arcseconds, it is a significant improvement over the value of 4327 arcseconds measured

for MgO grown on GaN using similar growth conditions.

Similar to the XRD data, XTEM shows improved crystal quality in the ternary as

compared to the binary, (Figure 5-18). In both cases, the oxide /GaN interface is

epitaxial. For the binary, continued growth produces a change in microstructure

indicative of a nanocrystalline film. For the ternary, this transition is not observed and

the overall defect density appears to be significantly lower. This improvement in

structural quality is most likely due to the reduction in mismatch for the ternary relative

to the binary MgO.

In order to increase the amount of Ca incorporated into the films and reduce the

lattice mismatch further, the fluxes of the Mg and Ca were held constant and the length of

time the shutters were open was varied. The standard procedure was 10 seconds of Mg

followed by 10 seconds of Ca(10/10) with continuous operation of the oxygen plasma.

This shutter sequence was continued for the entire growth time. Keeping the cycle time

constant at 20 seconds, shutter times of 8 seconds Mg followed by 12 seconds Ca (8/12),

and 5 seconds Mg followed by 15 seconds of Ca (5/15) were tried. All samples began

with an Mg layer as the first layer.









HRXRD shows thel0/10 sample to have the expected peak position to the right of

the GaN (004) peak at 73.9 degrees. This corresponds to a composition of 60% Mg and -

1.5 lattice mismatch. For the 8/12 shutter sequence more Ca is incorporated and the film

is slightly Ca rich. The Mg content goes down to 40.5% Mg and the mismatch is now

+0.96%. The 8/12 sample peak shows up as a shoulder on the GaN (004) peak. The 5/15

shutter sequence incorporates even more Ca and has a peak position of 70.175 degrees.

This results in a Mg content of 24% and a lattice mismatch of +3.04%. Figure 5-19

shows the XRD plots of these 3 samples.

AFM shows a slight difference in surface roughness for the different shutter

sequences, (Figure 5-20). For 10/10 RMS is 0.790 nm, for 8/12 RMS is 1.081 nm, for

5/15 RMS is 0.979 nm. All of these samples have a 20 nm Sc203 cap. HRTEM shows

the oxide to be crystalline and the capping layer to be polycrystalline for the 10/10 shutter

sequence.






78


Table 5-1. Growth rate and AES data for MgCaO samples. AES data taken from an
MgO single crystal is shown for comparison.


Mg Beam Equivalent Ca Beam Equivalent Growth Mg/Ca (Mg+Ca)/O
Pressure Pressure Rate Ratio Ratio
(Torr) (Torr) (nm/min)

5.7x10-8 5.7x10-8 1.0 0.81 0.92

10x10-8 8x10-8 5.3 1.27 0.83

10x10-8 3.4 0.72 0.72

MgO Ref. 0.6






79


Table 5-2. Growth rate, AFM and AES data for MgCaO grown at 300C.

Growh MgfBeam CafBeam
Growth Mg Beam Ca Beam Growth Rate RMS roughness O/(Mg+Ca)
Equivalent Equivalent .
Method Euvln (nm/min) (nm) from AES
Pressure (Torr) Pressure (Torr)
Continuous 10x10-8 9.0x10-8 5.6 1.931 0.52
Digital 9.6x10-8 8.6x10-8 3 0.712 1.40
Continuous 8.0x10-8 9.0x10-8 5.3 0.774 1.29
Digital 8.0x10-8 9.0x10-8 1.8 0.554 1.81
MgO ref. 12x10-8 --- 2.4 1.998 0.50






80


Table 5-3. Composition and mismatch from XRD

layer %
C- continuous mismatch
2-theta % Mg
D- digital to GaN

MgO 78.59 100 -6.45
MgCaO-C 75.43 72.8 -3.11
MgCaO-D 75.41 72.8 -3.09
MgCaO-C 74.79 74.79 -2.40
MgCaO-D 74.47 74.47 -2.05
50/50 -72.9 50 -0.23
CaO 67.375 0 6.86









0-0
Ca-O





[010]


Figure 5-1. Illustration of the CaO structure (from Cullity 1978)







82










Ca 0














I I I I
500 1000 1500 2000
Kinetic Energy (eV)


Figure 5-2. AES shows only Ca and 0 after growth of CaO at 300C



































CaO


MgO


FIG. 229.-System CaO-MgO.


R. C. Doman,
and A. M. Alper,
(1963).


J. B. Barr, R. N. McNally,
J. Am. Ceram. Soc., 46 [7] 314


Figure 5-3. MgO-CaO phase diagram.78


CaO-MgO













Amorphous
CaO


Figure 5-4. HR TEM of CaO.


































0 2.5


Figure 5-5. AFM of CaO grown at 100C, RMS


-10.0




7.5




5.0




2.5




0


0.667 nm.