<%BANNER%>

Development of High Temperature Stable Ohmic and Schottky Contacts on N-GaN

Permanent Link: http://ufdc.ufl.edu/UFE0021183/00001

Material Information

Title: Development of High Temperature Stable Ohmic and Schottky Contacts on N-GaN
Physical Description: 1 online resource (175 p.)
Language: english
Creator: Khanna, Rohit
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: boride, contact, ohmic, schottky, temperature
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: In this work the effort was made to towards develop and investigate high temperature stable Ohmic and Schottky contacts for n type GaN. Various borides and refractory materials were incorporated in metallization scheme to best attain the desired effect of minimal degradation of contacts when placed at high temperatures. This work focuses on achieving a contact scheme using different borides which include two Tungsten Borides (namely W2B, W2B5), Titanium Boride (TiB2), Chromium Boride (CrB2) and Zirconium Boride (ZrB2). Further a high temperature metal namely Iridium (Ir) was evaluated as a potential contact to n-GaN, as part of continuing improved device technology development. The main goal of this project was to investigate the most promising boride-based contact metallurgies on GaN, and finally to fabricate a High Electron Mobility Transistor (HEMT) and compare its reliability to a HEMT using present technology contact. Ohmic contacts were fabricated on n GaN using borides in the metallization scheme of Ti/Al/boride/Ti/Au. The characterization of the contacts was done using current-voltage measurements, scanning electron microscopy (SEM) and Auger Electron Spectroscopy (AES) measurements. The contacts formed gave specific contact resistance of the order of 10E-5 to 10E-6 Ohm-cm2. A minimum contact resistance of 1.5x10-6 Ohm.cm2 was achieved for the TiB2 based scheme at an annealing temperature of 850-900 degreeC, which was comparable to a regular ohmic contact of Ti/Al/Ni/Au on n GaN. When some of borides contacts were placed on a hot plate or in hot oven for temperature ranging from 200 degreeC to 350 degreeC, the regular metallization contacts degraded before than borides ones. Even with a certain amount of intermixing of the metallization scheme the boride contacts showed minimal roughening and smoother morphology, which, in terms of edge acuity, is crucial for very small gate devices. Schottky contacts were also fabricated and characterized using all the five boride compounds. The barrier height obtained on n GaN was ~0-5-0.6 eV which was low compared to those obtained by Pt or Ni. This barrier height is too low for use as a gate contact and they can only have limited use, perhaps, in gas sensors where large leakage current can be tolerated in exchange for better thermal reliability. AlGaN/GaN High Electron Mobility Transistors (HEMTs) were fabricated with Ti/Al/TiB2/Ti/Au source/drain ohmic contacts and a variety of gate metal schemes (Pt/Au, Ni/Au, Pt/TiB2/Au or Ni/TiB2/Au) and were subjected to long-term annealing at 350 degreeC. By comparison with companion devices with conventional Ti/Al/Pt/Au ohmic contacts and Pt/Au gate contacts, the HEMTs with boride-based ohmic metal and either Pt/Au, Ni/Au or Ni/TiB2/Au gate metal showed superior stability of both source-drain current and transconductance after 25 days aging at 350 degreeC. The need for sputter deposition of the borides causes problem in achieving significantly lower specific contact resistance than with conventional schemes deposited using e-beam evaporation. The borides also seem to be, in general, good getters for oxygen leading to sheet resistivity issues. Ir/Au schottky contacts and Ti/Al/Ir/Au ohmic contacts on n-type GaN were investigated as a function of annealing temperature and compared to their more common Ni-based counterparts. The Ir/Au ohmic contacts on n-type GaN with n~ 10E17 cm-3 exhibited barrier heights of 0.55 eV after annealing at 700 degreeC and displayed less intermixing of the contact metals compared to Ni/Au. A minimum specific contact resistance of 1.6 x 10E-6 Ohm.cm2 was obtained for the ohmic contacts on n-type GaN with n~10E18 cm-3 after annealing at 900 degreeC. The measurement temperature dependence of contact resistance was similar for both Ti/Al/Ir/Au and Ti/Al/Ni/Au, suggesting the same transport mechanism was present in both types of contacts. The Ir-based ohmic contacts displayed superior thermal aging characteristics at 350 degreeC. Auger Electron Spectroscopy showed that Ir is a superior diffusion barrier at these moderate temperatures than Ni.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Rohit Khanna.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Pearton, Stephen J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021183:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021183/00001

Material Information

Title: Development of High Temperature Stable Ohmic and Schottky Contacts on N-GaN
Physical Description: 1 online resource (175 p.)
Language: english
Creator: Khanna, Rohit
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: boride, contact, ohmic, schottky, temperature
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: In this work the effort was made to towards develop and investigate high temperature stable Ohmic and Schottky contacts for n type GaN. Various borides and refractory materials were incorporated in metallization scheme to best attain the desired effect of minimal degradation of contacts when placed at high temperatures. This work focuses on achieving a contact scheme using different borides which include two Tungsten Borides (namely W2B, W2B5), Titanium Boride (TiB2), Chromium Boride (CrB2) and Zirconium Boride (ZrB2). Further a high temperature metal namely Iridium (Ir) was evaluated as a potential contact to n-GaN, as part of continuing improved device technology development. The main goal of this project was to investigate the most promising boride-based contact metallurgies on GaN, and finally to fabricate a High Electron Mobility Transistor (HEMT) and compare its reliability to a HEMT using present technology contact. Ohmic contacts were fabricated on n GaN using borides in the metallization scheme of Ti/Al/boride/Ti/Au. The characterization of the contacts was done using current-voltage measurements, scanning electron microscopy (SEM) and Auger Electron Spectroscopy (AES) measurements. The contacts formed gave specific contact resistance of the order of 10E-5 to 10E-6 Ohm-cm2. A minimum contact resistance of 1.5x10-6 Ohm.cm2 was achieved for the TiB2 based scheme at an annealing temperature of 850-900 degreeC, which was comparable to a regular ohmic contact of Ti/Al/Ni/Au on n GaN. When some of borides contacts were placed on a hot plate or in hot oven for temperature ranging from 200 degreeC to 350 degreeC, the regular metallization contacts degraded before than borides ones. Even with a certain amount of intermixing of the metallization scheme the boride contacts showed minimal roughening and smoother morphology, which, in terms of edge acuity, is crucial for very small gate devices. Schottky contacts were also fabricated and characterized using all the five boride compounds. The barrier height obtained on n GaN was ~0-5-0.6 eV which was low compared to those obtained by Pt or Ni. This barrier height is too low for use as a gate contact and they can only have limited use, perhaps, in gas sensors where large leakage current can be tolerated in exchange for better thermal reliability. AlGaN/GaN High Electron Mobility Transistors (HEMTs) were fabricated with Ti/Al/TiB2/Ti/Au source/drain ohmic contacts and a variety of gate metal schemes (Pt/Au, Ni/Au, Pt/TiB2/Au or Ni/TiB2/Au) and were subjected to long-term annealing at 350 degreeC. By comparison with companion devices with conventional Ti/Al/Pt/Au ohmic contacts and Pt/Au gate contacts, the HEMTs with boride-based ohmic metal and either Pt/Au, Ni/Au or Ni/TiB2/Au gate metal showed superior stability of both source-drain current and transconductance after 25 days aging at 350 degreeC. The need for sputter deposition of the borides causes problem in achieving significantly lower specific contact resistance than with conventional schemes deposited using e-beam evaporation. The borides also seem to be, in general, good getters for oxygen leading to sheet resistivity issues. Ir/Au schottky contacts and Ti/Al/Ir/Au ohmic contacts on n-type GaN were investigated as a function of annealing temperature and compared to their more common Ni-based counterparts. The Ir/Au ohmic contacts on n-type GaN with n~ 10E17 cm-3 exhibited barrier heights of 0.55 eV after annealing at 700 degreeC and displayed less intermixing of the contact metals compared to Ni/Au. A minimum specific contact resistance of 1.6 x 10E-6 Ohm.cm2 was obtained for the ohmic contacts on n-type GaN with n~10E18 cm-3 after annealing at 900 degreeC. The measurement temperature dependence of contact resistance was similar for both Ti/Al/Ir/Au and Ti/Al/Ni/Au, suggesting the same transport mechanism was present in both types of contacts. The Ir-based ohmic contacts displayed superior thermal aging characteristics at 350 degreeC. Auger Electron Spectroscopy showed that Ir is a superior diffusion barrier at these moderate temperatures than Ni.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Rohit Khanna.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Pearton, Stephen J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021183:00001


This item has the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101115_AAAACI INGEST_TIME 2010-11-15T11:46:48Z PACKAGE UFE0021183_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 3928 DFID F20101115_AAAZAF ORIGIN DEPOSITOR PATH khanna_r_Page_067thm.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
f239bd27154657f0e5d6eb149211780a
SHA-1
c89ffb943798f0cef22ba0187bc5afad6c5ababf
309 F20101115_AAAYUK khanna_r_Page_174.txt
0d903df3ec71c9d10245982831434a61
6fd563cf4e67307f3ab05ebd2b03652aeaae108e
1509 F20101115_AAAYTW khanna_r_Page_159.txt
5ee12bceacc8ce09976730e67ed79d8d
4ed4cbfa6220ec1d5e7e4ae4f1bab186090379b5
26279 F20101115_AAAXRI khanna_r_Page_111.QC.jpg
389a3a5f06cd8a38e10acd18ab0ebd84
52181700420da544f419f8b2bc9c5e4f146e65a5
1437 F20101115_AAAXQU khanna_r_Page_008thm.jpg
79b7aa8b0920c9a6532b1e8488a5fbd2
7edd14b3d45d2b0df044d63969440c07000edd57
18988 F20101115_AAAZAG khanna_r_Page_084.QC.jpg
2b196b296f3fb605c06d6d44f53a83b9
70bcc8babd670b618dbb93fae246d6f112f17053
3150 F20101115_AAAYVA khanna_r_Page_157thm.jpg
8d2116ca2e3e415a38c288ac38ff1ecf
f827890e603c616b2d9a9bce6949d51dcd72ba13
1136 F20101115_AAAYUL khanna_r_Page_175.txt
6bba85543d36a2f70fe7bae99a968571
44b7d682d607a0f669cd5d7d9773e58a9fa88549
339 F20101115_AAAYTX khanna_r_Page_160.txt
67e1b1977df3c406b1d90608ef9c2c16
482cb9e5f08d33c569a76a14f396307975d03193
5561 F20101115_AAAXRJ khanna_r_Page_153thm.jpg
578f177de37a05bede599b1f1620ede8
ea2628255b72f52d1fdccdc8bccaf59da4ba1031
24657 F20101115_AAAXQV khanna_r_Page_170.QC.jpg
e196d5fe4971fc9b9187a9d10739356f
281a77441adc3aae2dfe0c1cfe3faedcab7c12e0
11090 F20101115_AAAZAH khanna_r_Page_133.QC.jpg
b51a063e10b9fb3841e06cc1f705f925
cf5b3afc3d0d773763594c5518cf08e20c569fd0
11185 F20101115_AAAYVB khanna_r_Page_081.QC.jpg
c74473a7688bfc8a6f5e9f425bcbd573
35548a69ec8a109790c12325d8a106401e0e7dff
2254 F20101115_AAAYUM khanna_r_Page_001thm.jpg
36109152f2e0760065518686ef4a87bb
b63bce61370cbe3ed6713ea21d3096216afc090a
313 F20101115_AAAYTY khanna_r_Page_161.txt
6da4dda750aea7d1b84a25609737b267
f76dafcb7777f1b54cb0d9faab6604fb10a3b8b6
73320 F20101115_AAAXRK khanna_r_Page_173.jpg
89e06aafbaffe6aafcfaaa446177805d
d2ade40db29c41573344a34323b03192dc9eb49c
49492 F20101115_AAAXQW khanna_r_Page_150.pro
911a9324f2dbff9310bae4b4561090fa
f684a69c9c63d3837b841c43e7d22b42d4e7542f
8936 F20101115_AAAZAI khanna_r_Page_157.QC.jpg
19e7080c95eddf9e8e5eb7709d2b9edb
ada8839e1cab29caa1764a002714089d8c6c6850
26020 F20101115_AAAYVC khanna_r_Page_018.QC.jpg
b9ec530de8d53546dda229b1c163860a
164bd9e8bfbaa9493fb968c14f28f90be5412340
3976805 F20101115_AAAYUN khanna_r.pdf
4939425286e36f784dfd1300ad171d0f
3d69765ddd0fbc8bab22b469e1d247595fde26b3
2148 F20101115_AAAYTZ khanna_r_Page_162.txt
48fbeca21fe646c2f84d3cf3cec712d9
bda9f794c1d9503d17640412ca38b37ca15e7c1a
104785 F20101115_AAAXRL khanna_r_Page_016.jp2
7db7ae683a4eeffea54691af5a5b5c44
2c014579ebf71c2410f120b71b4befbf5f62aaa0
21983 F20101115_AAAXQX khanna_r_Page_015.QC.jpg
bb7b87c22679ac491a0cdea55f483a64
6d9ed105e92e3c7a45532b2c692857f335c16a74
10431 F20101115_AAAXSA khanna_r_Page_022.pro
4a92caf6bcbe6c2bd135d93f516b79bf
02ca79784385bd26b7f21c14f65ea2a671c1906a
6893 F20101115_AAAZAJ khanna_r_Page_165thm.jpg
ec6b3142411c82c9c87c5db6439cdb0e
b21aa083d41e3ba9dae332320ebd3474de60da88
6569 F20101115_AAAYVD khanna_r_Page_093thm.jpg
5a6001529f168b80c322670517e3f4ca
078e663b40da27167435abc27e473fb40f9c6636
20990 F20101115_AAAYUO khanna_r_Page_032.QC.jpg
b8bfca9f38f3cc49ebba02ecab27fae4
f780c4d482ebd39e05962e42fb5385304c84d34d
6030 F20101115_AAAXRM khanna_r_Page_051.QC.jpg
9271749d72c341aa49b419441841c563
63e75ed4e1769ecd067b659f0f2eb6655c8d05ec
119591 F20101115_AAAXQY khanna_r_Page_135.jp2
961fbe99dc259594bce3dac29992a013
8571ada9484e8298f658e2ee06bfbd7959d6383b
668 F20101115_AAAXSB khanna_r_Page_143.txt
51dd542c0a068adb3a486913fc572f50
6c6f6e430a94b5230da0112aeb85438e020105b0
19481 F20101115_AAAZAK khanna_r_Page_145.QC.jpg
11bc252318ded6d02bef0ccf26d48ac7
3a0f694eacca92d1d07add0547c2aec50760d22e
21467 F20101115_AAAYVE khanna_r_Page_028.QC.jpg
373cf9d76b16ce75b886d7bcba005c41
1ea58275591f19594be2aa9555d70d395d6f1b8b
2790 F20101115_AAAYUP khanna_r_Page_063thm.jpg
3a4edf617af4a93abac4f5d0835a6376
7ef0395bce19e5777e8cec7b11df0a9860ebc995
2371 F20101115_AAAXRN khanna_r_Page_008.pro
69342bc7a78ac9b7aa23c0a7f18d4b93
5adc6073f087336a5a9cf73c863726f82aa8fd81
723 F20101115_AAAXQZ khanna_r_Page_104.txt
b0b36ea6dda3976f6091332b8060638f
cd06f7d4e85465c789b25dfe19b537951700c218
1051942 F20101115_AAAXSC khanna_r_Page_065.jp2
782ce8bd5d7f39f33142c83bf0e3551f
91def42b779c09f1fc77d08dead5dbf27890189d
11612 F20101115_AAAZAL khanna_r_Page_078.QC.jpg
7cda4763235ed5b9913ac0a8405f289d
f9989937872f685d60ab4ee1e36f78884c17bb24
23291 F20101115_AAAYUQ khanna_r_Page_090.QC.jpg
8ea022ddf744988f5fcb5b09aa2b1b4b
66d419278dc06c3e436ce1358faa95a59aa92817
70669 F20101115_AAAXRO khanna_r_Page_077.jpg
7502c3cabee924e29224f596234d49ff
eb2ff30b74d7428150107d0b22b2914c58c6a55b
10651 F20101115_AAAZBA khanna_r_Page_141.QC.jpg
52e00320b251a4fecc08b5b02a2ad0a8
26c0e96dcfe7daaf70a28678a0a9c33081545d4c
1293 F20101115_AAAZAM khanna_r_Page_003thm.jpg
6639317d5831faf27733a3a09d847206
d95e6e713f31d2e9f1ce1d9114cabfe58631e868
3915 F20101115_AAAYVF khanna_r_Page_017thm.jpg
0a6e8e25bc005f4183489f9f619a6f75
30704b830ed553486412cf3db867e08e4901d391
26488 F20101115_AAAYUR khanna_r_Page_053.QC.jpg
9950ea7ce03a4c091b525a465dd4ba2b
3d6aa3b193d0682a199309fc95c2053b45f025b7
14858 F20101115_AAAXRP khanna_r_Page_175.QC.jpg
a17f34b0a96afce70f0a82e5dc800dbc
3401b4628e4a87c400fb93777f5151a2d9c5e020
88 F20101115_AAAXSD khanna_r_Page_002.txt
30a4663e0a9bbfb5f612b27cb3ab8670
b98404508c561c87fba3678918be82274e433aa4
6776 F20101115_AAAZBB khanna_r_Page_138thm.jpg
5d3afc15458b0f0bb69136dbf4299ee4
828e888ec7561f39e56f148689e67f72c1eeffdf
2862 F20101115_AAAZAN khanna_r_Page_069thm.jpg
533fb63c19444e07a351a4f94b2926ff
43727791c7904a2bc3a1330e77f1fcfb7501f4c5
28639 F20101115_AAAYVG khanna_r_Page_012.QC.jpg
d33736be973f4ae30d55ca4104ffbd71
a2aafcf6cb064a041950f886720876ffb61564cc
22536 F20101115_AAAYUS khanna_r_Page_016.QC.jpg
c32870b0f584764e5c612a2252c449c6
45a1a0b5c7296d5e577ec7c26f0bd5b167497e58
1053954 F20101115_AAAXRQ khanna_r_Page_063.tif
3844a5b05da1c0601e7a082510635cd8
dfc52f08aec1e24d8f67476c4e36153748dd7e52
13283 F20101115_AAAXSE khanna_r_Page_071.QC.jpg
8bf62d715514ab8590a5899f87f11b5c
167da13aa349910f8419a70c9e49a597aa203430
3718 F20101115_AAAZBC khanna_r_Page_133thm.jpg
07fa7543eafded097fbd0c0c8352e54a
3ec93f7c5844687c76f41ff20342184ef8b2e0ec
6929 F20101115_AAAZAO khanna_r_Page_021thm.jpg
18965a8163f284ca6f4da6f60c9ba3d2
754e90b67d359d5cdfad4762feae073d82784b84
11274 F20101115_AAAYVH khanna_r_Page_147.QC.jpg
7c986cac002e0abe0a4abeefa06525c4
7e27f2237af25b235a3cc3e2961ef981aabc874b
3533 F20101115_AAAYUT khanna_r_Page_146thm.jpg
51354e71b32ba4a2a689355d4489e577
b9f1b662b6cb26cce0db8d40df01b243bb122ecc
18895 F20101115_AAAXRR khanna_r_Page_083.QC.jpg
31ab42a46175b1e68c901885a056cab1
6502b83c3b61c63b80720b11f8ddcf8120d1df24
26310 F20101115_AAAXSF khanna_r_Page_135.QC.jpg
b1ca301445025e559859d3460ba13132
7f21788e046c6e2a0bad0bc0c21d1f3a63693d64
2397 F20101115_AAAZBD khanna_r_Page_022thm.jpg
d958d3d74a091c6e4991f9ee37f14211
22f23557c4cd86fb68fe679923d1375f54d8fef7
11397 F20101115_AAAYVI khanna_r_Page_052.QC.jpg
b0a311a5757e001ca4ffe5c3ff281e43
5c22f8cacd836260911b0935fcc5a22e530f744f
3357 F20101115_AAAYUU khanna_r_Page_127thm.jpg
f96cba8fb3c756fef50d70888a569c4f
1ab7a2e2683d09561ca0bf7fbb8c90c804c96cf9
F20101115_AAAXRS khanna_r_Page_076.tif
f11387fcb782c6a1c25ff1acf435d09b
defd114ebd25b9126e817602ea2162a25c3899a4
5608 F20101115_AAAXSG khanna_r_Page_159thm.jpg
1c5cafda0970405b3c141c2c52412b82
f2f131bd5757146ce0cfeca2867e144fbeca649a
29371 F20101115_AAAZBE khanna_r_Page_013.QC.jpg
235992f6269249988f4fd8c9c7741377
fb2367ff1ac012f37f4c45a8ce6d35e527ae4835
15620 F20101115_AAAZAP khanna_r_Page_082.QC.jpg
3e38485518a01434831f689b80b98c30
d391eaaf7ffc2162f6ff270c0ca6d86a7d6654d1
7926 F20101115_AAAYVJ khanna_r_Page_155.QC.jpg
e100cf31d5dd02112cd6fdf9991a44ef
2ab625d3f8c2f57abe6783273ceb2745db9bfd62
6614 F20101115_AAAYUV khanna_r_Page_054thm.jpg
94591e109a46a27afd5ca3b177f7942b
959733fabf67f426a1b46c36b2c71a87f904b91f
F20101115_AAAXRT khanna_r_Page_096.tif
b748ccbced86c2708b2b7af468cfe902
7d3d3022b19f8f54bb07ac1ff3c25ffd9cf05bd2
30088 F20101115_AAAXSH khanna_r_Page_011.QC.jpg
bd0ec8556d66fa33e4446ea9dc310958
ec780cd3fe2725db17ae2f308973bb280116f36f
9590 F20101115_AAAZBF khanna_r_Page_099.QC.jpg
08569cc99d4d85ed6cdc1b0ffdac6e64
d12f5f43c9a4f57935867ce71f803409998c5985
3510 F20101115_AAAZAQ khanna_r_Page_097thm.jpg
ef6c014abf098bb80f6fa60844d0bbad
2a7faa7e972ffa1babdc29ee212b22f48d6da91d
6762 F20101115_AAAYVK khanna_r_Page_087thm.jpg
e35823e85d433432bd10a7101ab6f84c
9e26bc26c640d65929361651b696ad4be4d4ee4c
25323 F20101115_AAAYUW khanna_r_Page_024.QC.jpg
18fee5aa8494507592c478a8751688fe
52bc58cc89795c94727ddff4c700cb772bba5a17
13094 F20101115_AAAXRU khanna_r_Page_097.pro
63d473819311283ccf028f34a0aee526
f4b9a255312c884467c03c246c293f8cce4644ac
F20101115_AAAXSI khanna_r_Page_137.tif
86c422b97a7c144b1c9f47e947bbc8ed
c8bff982ac4dcbf90981f8c25730f01afa5b5d2e
7015 F20101115_AAAZBG khanna_r_Page_035thm.jpg
0d75d67bbcb0e82f046b64d553509202
60f2a25842824348635abc22c9add28df06e0065
6002 F20101115_AAAZAR khanna_r_Page_023thm.jpg
3a772592f3905aba9479d0be7a50d125
f4ef384c1196ba42b8716a373d0c9a1457c31747
6807 F20101115_AAAYWA khanna_r_Page_058thm.jpg
518851015e15b7c36f9881ce133ae204
4d34e60a48916e887f306ade8a70632a40a213c1
3590 F20101115_AAAYVL khanna_r_Page_120thm.jpg
f1a514a18c4ed2ce60e674fa60d62a84
8ffe5d7405b9efd83cbb0d7a6658e086d7779d23
25581 F20101115_AAAYUX khanna_r_Page_021.QC.jpg
a388e2ed1c2a2374395ffbaf6d9e8d5d
6147de22c7da7ecaa221738bae1dfeff79a45fe8
78106 F20101115_AAAXRV khanna_r_Page_148.jpg
b60c6089cdee766c5fe36fcba4ed71e5
e7eb827f2206ac8cd9e2da1f7f26118bdf355611
108265 F20101115_AAAXSJ khanna_r_Page_112.jp2
df52575765c041497f237331291c3d93
013a05b81440b3149dc973646bb47135772ebdd0
10094 F20101115_AAAZBH khanna_r_Page_070.QC.jpg
3af0626480fee1f9dff77cbdf51de027
bcfcb8160adecbbfd5184051db65c42a6b2bf4b0
11564 F20101115_AAAZAS khanna_r_Page_120.QC.jpg
b9953f3b7f14ee2f9198171364239a42
5ef2d5e36aea9afc4941f48c17e1981463e82645
11394 F20101115_AAAYWB khanna_r_Page_080.QC.jpg
e4ae527b4ad3c26493876437c4e0b3af
a73a824cfe4a5c220a9aa08367cbb629252d387d
6825 F20101115_AAAYVM khanna_r_Page_148thm.jpg
7d312a5f909f0c44a894ca1a8ec0ac19
b8f987d52d80aa62b53348a50b8a945a6b04096a
6907 F20101115_AAAYUY khanna_r_Page_170thm.jpg
d4072291348fc3ea074220acdfaf29ff
dd73c1508dbc2b90ec48bc2d0f2a564c2635182d
54379 F20101115_AAAXRW khanna_r_Page_021.pro
dc679dbbcf7df9325d72665319f48cc6
a9ee28f0a36061d38a929609d0df27eb9a16d66f
F20101115_AAAXSK khanna_r_Page_105.txt
53800f5400bc9f9188cd1e66ef28d5e4
343051fdc7e264884a5566c44c149cf57f96b3ef
11923 F20101115_AAAZBI khanna_r_Page_042.QC.jpg
58219b3a6fcd67660bf897a0342a18a4
e9770b79c14e89080af9ac103d984080d76cdcf3
3775 F20101115_AAAZAT khanna_r_Page_126thm.jpg
e9b9c7a91c1f679e889a05e62105020c
72e4e990f63d4a517a99cfd17569d83cf4c6fb4d
7089 F20101115_AAAYWC khanna_r_Page_109thm.jpg
918bb92b256b3211239a55ead06be2cb
1b323d36a888733fe8c5c0143873f1fd0bd0b379
6806 F20101115_AAAYVN khanna_r_Page_075thm.jpg
0eea2ecc3995e9c462afc3776ee19d0f
68ae13bdb3116aa9a3d78789731b92ce2d0a2928
10047 F20101115_AAAYUZ khanna_r_Page_140.QC.jpg
c94fb54f890dcae5f6370dcce031d2e9
478e0d5c7a243b9ace0f55c24686698c2f99120b
36697 F20101115_AAAXSL khanna_r_Page_078.jpg
630c8de0b29022c4f2f98c61a85a7b1f
2fa45bf5935318e5ef2e5e4ee895b1173dfac54f
47841 F20101115_AAAXRX khanna_r_Page_171.pro
388dda5be78efebdb22d7d4f704da6f3
a9b9aebd10b97060fb13ab7812b10622f400c6dd
7063 F20101115_AAAZBJ khanna_r_Page_025thm.jpg
204b80bc9ce3627a6e750e6616235e7b
4aaeea9915b39e30397345966a60d7c50dfd2245
23603 F20101115_AAAZAU khanna_r_Page_169.QC.jpg
0b7d72c9fc14ca24f78137e5d916f11e
3b5afc691e25b5acbdc9dd48d098a6aa86edb2b2
2239 F20101115_AAAYWD khanna_r_Page_051thm.jpg
fafba9f5ecda02db67164a1416c91ccb
ae4f7c0cfa3dbde233b00e26328a6738018526ab
11595 F20101115_AAAYVO khanna_r_Page_086.QC.jpg
3e20585543225871677b58e18f3ef987
7bbdd57576969e43bcd9c390f10cda14a7e8dd67
22818 F20101115_AAAXTB khanna_r_Page_001.jpg
339a5bd441b48f5f449fc2c33584ace5
3cc9712b6ca8d24558ccaca7fb2e9a36192c355d
340 F20101115_AAAXSM khanna_r_Page_131.txt
9eb575c3a734e22e76892801a9c25e82
ddd9bea3819bc142bcf22c9b4a786cbab55cc0ad
F20101115_AAAXRY khanna_r_Page_023.tif
bfa342105574cfbeb594f1631582bb2f
d8efdcce8fd029f9f6a543babb8c6d4aa413f344
6118 F20101115_AAAZBK khanna_r_Page_094.QC.jpg
5907a1afb19649cfaa012ce9e71dd336
799958fd4aa6dd348da60aed4c071507367682b7
26177 F20101115_AAAZAV khanna_r_Page_035.QC.jpg
6195fb8dff35fa26081f2e1897ae17f5
dbc35cc9aa39a284be9776353a589a9ee16dd4fe
24966 F20101115_AAAYWE khanna_r_Page_149.QC.jpg
7e02f7adf814d43fccf0abe295bd1f8e
7e57a0dce37ef33b6aef8bff9ddca15b13ed54a4
2446 F20101115_AAAYVP khanna_r_Page_005thm.jpg
da4ea4dac1bc53841c8bde979c41b6ab
de2a87d06a1b04424a1a91496ffc2a6bed2a1853
9453 F20101115_AAAXTC khanna_r_Page_003.jpg
19ca9cb292f1781ad7d21f07696e28bd
e8d000ee257cf6d45546c8ddbcdd144d08a9f95e
352 F20101115_AAAXSN khanna_r_Page_063.txt
a5c3bf5f6d8c483ee37400d0f7dd48a9
28848141e5636e5a7bb98d8da51873b5f043d06d
2741 F20101115_AAAXRZ khanna_r_Page_066thm.jpg
b97ef5553cc7c80341cb35c74de04f36
5401e16da1536047c6b7a7c9a9b01ff6c08b1312
6206 F20101115_AAAZBL khanna_r_Page_007thm.jpg
c04f094b152577af1156fce8266a406a
324f68f5ec34fbba2bfdc50c7e64f88d0c6cc45c
6910 F20101115_AAAZAW khanna_r_Page_034thm.jpg
775c5a894d8c42b5165aa2f3346a569d
c6d22d3ada20b1a0da6f8b912ae18013eb8958cf
3267 F20101115_AAAYWF khanna_r_Page_125thm.jpg
8908000889240c440d44814b63f107ba
a74c8e9ef12a13fc97260496ad98eaf2c6f27c0c
3622 F20101115_AAAYVQ khanna_r_Page_122thm.jpg
843e4817097bf1fcbc2d2ab3f468ee2e
ed7d9efaaa7ffb6db67c6e6c58175945fe976e1d
78931 F20101115_AAAXTD khanna_r_Page_004.jpg
e8f34ef37a61768c63d619a5748ab329
ba5aa07123cff3356870c07ec5c05f34a63210e8
25271604 F20101115_AAAXSO khanna_r_Page_012.tif
3e1a6ae452394175ca8a99e6cdd5523c
242627f30a52e72475d2aedd2dae0af5703079e7
24065 F20101115_AAAZCA khanna_r_Page_038.QC.jpg
79447d6ec8927970af9772ea6a0b0d96
042899198e549f57696543fe4f47481e5a38de40
25707 F20101115_AAAZBM khanna_r_Page_031.QC.jpg
67e967d709f494c2beaf3044b0f423a9
d879a87617457a03e5ded06ca4f329d1c6a48e60
3674 F20101115_AAAZAX khanna_r_Page_121thm.jpg
8cb136bb859bbba11314f227fc67a626
1246d2d5cc391e2f7adcb87736a3bade7691af8c
13859 F20101115_AAAYVR khanna_r_Page_143.QC.jpg
faeb06b71b9c88a7d989510f27f01104
f3f60a2065097273817d8749fb6edba04b547622
99215 F20101115_AAAXSP khanna_r_Page_151.jp2
3533e4018360c156d976f01dba33963c
e4acca136cf06dfde2f1e1d63e62846e311d846d
25561 F20101115_AAAZCB khanna_r_Page_148.QC.jpg
2d7012e94f78254197b317577c468101
ae15c39ce698fc65d974d004459b7ff0630bd649
25167 F20101115_AAAZBN khanna_r_Page_110.QC.jpg
57fd038ab9875dde23cc8b48db53b349
fd2761cbe80279d9b3ee2689fa25b3bfa35d3223
2895 F20101115_AAAZAY khanna_r_Page_154thm.jpg
21c115dae7fc23a36fa2b06e1953b54f
ee887c14506868448b024f9287b8fa367a2282f4
7308 F20101115_AAAYWG khanna_r_Page_012thm.jpg
015cbc25c147539c7075b0e49fdbb807
ef5a50d8a50e875e553c3838d82861d88e36e0ea
4511 F20101115_AAAYVS khanna_r_Page_143thm.jpg
ed0a18a3de356d1c6e4dea7da7b65f81
b686019e3f4c9f8770dd63499026954b207cc942
21744 F20101115_AAAXTE khanna_r_Page_005.jpg
972740ade236e3bcda115f35a49f909a
8484432b1241daf6c85012b47633675fe69344f1
75305 F20101115_AAAXSQ khanna_r_Page_075.jpg
4b3e668a5af667dfe919c9f3a4f27704
f4dc04476f21ba77b407bdf0bbac28ec6a145e0d
23070 F20101115_AAAZCC khanna_r_Page_074.QC.jpg
7f5894ded48247abc52f02179b2d958f
a3cbfe2856409fcd98d7582b4727de8b185f6ac1
3566 F20101115_AAAZBO khanna_r_Page_052thm.jpg
7493be0653c3f0a7cd4d3e2856d8426f
d20e5724607cf6957d226096841f949901047fe1
6865 F20101115_AAAZAZ khanna_r_Page_033thm.jpg
6f86b92a31aee1713380dec7f49276d7
79786a5960e8488b43636b331650c0436b7f9f1d
25314 F20101115_AAAYWH khanna_r_Page_165.QC.jpg
24dfcd30f54b25279a83aeceea065e82
7cfc03f5f8016f116f96613f011745a0905cf06d
7307 F20101115_AAAYVT khanna_r_Page_164thm.jpg
00aa3398c1707fff341d246454986656
08c0d4024e8ac02038e26df1b9d72bbc7323992a
85533 F20101115_AAAXTF khanna_r_Page_006.jpg
119dab41348b7e25ada84536cd0c568d
401e99e07d53a7bb2bab46ceee5fc5fc819c3cfc
308 F20101115_AAAXSR khanna_r_Page_086.txt
31d69046e38ea7aad5ccfc2eaa5ff269
fa428ec24e328b1e7d715b54d015ce0ac8d4052b
10267 F20101115_AAAZCD khanna_r_Page_117.QC.jpg
adeb7fe70e7c906e75ef041595e5f782
46b6396b6836b96abc8417dc6600228de4bc4b40
6884 F20101115_AAAZBP khanna_r_Page_038thm.jpg
b66e9eb0abe1ef43f555752a82e9173a
8d17e1b43c50b2f0b0d838440ac419ba200e6f98
25961 F20101115_AAAYWI khanna_r_Page_106.QC.jpg
e73fa9dbf374626457dce77761e18aab
9e15d8e20c1b83c23ae6c87862d40570e8e10b2e
6273 F20101115_AAAYVU khanna_r_Page_016thm.jpg
2b2b01ad113d645527575bd7d62d7b0a
cb70e5b60e0da33de8b51e6d715bead985a2b5ec
97085 F20101115_AAAXTG khanna_r_Page_007.jpg
1de9bd6e2abedf1e7f3a5db491e4eff4
4dae35827760b5959a51c4a13066cf3bc8bb063f
16142 F20101115_AAAXSS khanna_r_Page_065.QC.jpg
3e8186d2972228585cf6b6b8456a5935
6a2e49f2e79de134e84604fe9d6abe4edb7922f8
20922 F20101115_AAAZCE khanna_r_Page_030.QC.jpg
e6a3efa8b07e80e4878d56030f07649c
d289799904495d0ef4f7e142fbe2d3f16b5ec36b
15414 F20101115_AAAYWJ khanna_r_Page_044.QC.jpg
7911ea96cf7ed710a714967207f162e1
20413eafbb5e7be27c75ce96173c4b079461476b
3754 F20101115_AAAYVV khanna_r_Page_078thm.jpg
3e9008e8912da797ed9cb0278176913e
e1d62d36799f5b9f4a131ad843d613089dec157b
11690 F20101115_AAAXTH khanna_r_Page_008.jpg
667d1f82d2857ae1a3fdb59cfb1a2c0b
c159dd2a95f7a72af25b1b0420d7ebcd13d1a16d
53214 F20101115_AAAXST khanna_r_Page_082.jpg
2892e05610c4b5f7696269556e132d1d
08c2b935caa7baf236b2122b1bca81ae92fddfb0
9437 F20101115_AAAZCF khanna_r_Page_105.QC.jpg
9e64de1d352ed3d6e4d9db047349af97
0da1a1279a7b5a94270711ede9e37143b7f525a5
13089 F20101115_AAAZBQ khanna_r_Page_017.QC.jpg
78507e12b017c941576d5ae26abd051b
11acfd1449c9b9fef55c4e43016ef9c7ce7cbb43
25128 F20101115_AAAYWK khanna_r_Page_058.QC.jpg
9f67ddaaae622f5a5e62d873dac158a8
c08d40a380395e4a3a0fe03d8a503d5fc4d40bef
7007 F20101115_AAAYVW khanna_r_Page_045.QC.jpg
935c10da49cb0972da6d6d957782e96a
3005d36e95b59808c5c1a922a35631427dad6331
84574 F20101115_AAAXTI khanna_r_Page_010.jpg
ca93e0c1332538f095a04a320e6508e9
eb1eb1ec395aaddad9e5a590a72a2a024793b8bf
F20101115_AAAXSU khanna_r_Page_166.tif
83eb918f96e8d59dc3cfaa8797461655
151b54cc78e7452e27a7283a65d5da1e7cf833c8
14175 F20101115_AAAZCG khanna_r_Page_049.QC.jpg
f5ab34293d230576506e72181e98a8c2
1395903cf9793b3503566fbcf9237706cf0efc52
4031 F20101115_AAAZBR khanna_r_Page_071thm.jpg
3d6b11f83a781904c20bf4af12cbd7e5
4a6e0af1b38a99a8e7e4755583374c1a9f1ad670
5311 F20101115_AAAYXA khanna_r_Page_156thm.jpg
290cfd062de5ac7b709742e281d71531
1bfc8b7a7cc39bc9059e33c00d3b8da266851aa3
3565 F20101115_AAAYWL khanna_r_Page_008.QC.jpg
0dc793f6d92fbaecdd46036d4fb86f59
d9eba18cc5bc298d299760a5e7778383f274a8f0
25160 F20101115_AAAYVX khanna_r_Page_039.QC.jpg
dbb84c84cd60f28c63051d5e34d43e8e
7708d5645b6c408585a3d8f37eb7d652a6429919
107003 F20101115_AAAXTJ khanna_r_Page_011.jpg
8c0fed6ff53bce6d4b2a1c4468680786
aa509e5e82b40e4c5708abd8dbbb5db92727cd48
32918 F20101115_AAAXSV khanna_r_Page_128.jp2
f25515a9bd7afb3958b1a180c6319986
97152b0bcf98f00cd9e5ce72a0fbb1bed694e5b6
2064 F20101115_AAAZCH khanna_r_Page_094thm.jpg
c11a1525d8a11da7a9a5e7c0baaf158f
a933cc72376e6e16101d1e7642d5e2b0fc0ee949
6442 F20101115_AAAZBS khanna_r_Page_027thm.jpg
1d37dd01d74e131584a45b3ed33138d4
00e3dc181b2ee0ed73ed53b0fcc4ad068a224cbf
8057 F20101115_AAAYXB khanna_r_Page_066.QC.jpg
68bd6b0865ff97de8dc63d62b5f775c2
8957b5cc290eddc0d2cb055e6d1daac01d1b3fac
6839 F20101115_AAAYWM khanna_r_Page_173thm.jpg
01417f95df52b7d3d58bcbdda327b5cc
c78ba0202f79748ff4ea428720c6b7fa8fc4cdfe
15338 F20101115_AAAYVY khanna_r_Page_072.QC.jpg
8c68c3625c5232546f58556120f8eb6a
ee2f96df6b302a1154056e6c5d59d8a468d75191
100422 F20101115_AAAXTK khanna_r_Page_012.jpg
40d7776e5788c8c8cfd5110dbe60e18d
1b08efa646bb5aee343199b0505f728fc56c7196
1971 F20101115_AAAXSW khanna_r_Page_114.txt
15aa8bdad8dc31b2c1c5917976bf5ffc
0da39ffb6f73a4f70c23e5f78bc009758f1be48b
3742 F20101115_AAAZCI khanna_r_Page_147thm.jpg
a2fbf1d50e9553d590824d1d145cbf50
447f8150845b793c67627724b60b97edcb7b5555
24644 F20101115_AAAZBT khanna_r_Page_059.QC.jpg
a21f06c37a26e7a1171a9d58b566d78d
89e3fd9f5736adf50aa09967505dfb6311fdc9b7
6902 F20101115_AAAYXC khanna_r_Page_092thm.jpg
daaf89f04fbb8971c6e0c1e0d390a0a7
d5104d8621089716667242c8095f7a897f1ed90b
3953 F20101115_AAAYWN khanna_r_Page_064thm.jpg
bf9020cd3b2ac8428717363f4c62537c
509d9aa98d348a4398d9156f216a7a32d688c185
6848 F20101115_AAAYVZ khanna_r_Page_167thm.jpg
61d9ba2a38e0071e71595b04c6b3472c
2594378b44e7e8940290ed69fe26293ed4be3b2d
107007 F20101115_AAAXTL khanna_r_Page_013.jpg
27e4d636dfa87246c43ae248280d684a
56ae07d20ab6022a3c6c9d9449cc102f44bde081
8513 F20101115_AAAXSX khanna_r_Page_063.QC.jpg
e18cad23798d915315937e1ecb698b9c
7147db2fa7e76d3bb301dca526c22966f85279be
61214 F20101115_AAAXUA khanna_r_Page_032.jpg
73f7b80e73f4c9fcef9fce8fd31a1c46
4764a06e20b991b48f3f6471b7e2315b5dd4ef38
22166 F20101115_AAAZCJ khanna_r_Page_151.QC.jpg
2d30406ad17ecd71ca349eb23d334785
af3b91f8ab26cd0af4fce044c39e3b4d849e787f
5415 F20101115_AAAZBU khanna_r_Page_044thm.jpg
047dadbd27be26dd313c525731f6caf0
398a4407b6dbd6686f2f0bcc08e35674d6b643d6
7844 F20101115_AAAYXD khanna_r_Page_069.QC.jpg
75068522fe2cda79d95a2a99acecdece
2eb839a6bb3e1d6fc68cb02b66737a253b33d5a0
6677 F20101115_AAAYWO khanna_r_Page_112thm.jpg
c1862a4298c2cc72cd0a255e32954b5f
e3dc5b71e80dc9e046d394b31ce664fd69f585b5
79227 F20101115_AAAXTM khanna_r_Page_014.jpg
67387f6fe5fc83f7807d9ad5db9fa89f
e96a983505fe537a75dedb940c36d4b86a618994
201370 F20101115_AAAXSY UFE0021183_00001.mets FULL
02d8c31655885779fc7155c6b715ec56
dd77345916ea3c230dfca8f30b0985e93e878183
78287 F20101115_AAAXUB khanna_r_Page_033.jpg
8558f2631c15c2f2203b7f011d835867
1d4119eac2320494c707169a0116214da2816154
10127 F20101115_AAAZCK khanna_r_Page_104.QC.jpg
5f746a5419fb1a237369190d5a25892b
856cca53a7775a390ea1139740815d8f0a1b47af
6117 F20101115_AAAZBV khanna_r_Page_145thm.jpg
0449c1a191cb6063e081ed9952ef6149
7d281ee886596cdc0b8a1373b65a01df3b69c888
3761 F20101115_AAAYXE khanna_r_Page_086thm.jpg
862d6e449a983653da1a107143465679
91b224ea33e72c59150bceecb84876fd376111fa
6740 F20101115_AAAYWP khanna_r_Page_010thm.jpg
7b265e6c74fea0b797a87a360e6edf09
63972960ecafd36a077a7d455615966c7ee689b0
70718 F20101115_AAAXTN khanna_r_Page_015.jpg
5e7f8b86cdffc899242b399744363dd0
4bf578a748fd5a5873226ec655f4003164dbdcde
78388 F20101115_AAAXUC khanna_r_Page_034.jpg
27400c61e543c0b98d6d2aa61e01cb72
a3992c4e89992b09b627532aab3c77df0425e3fb
12954 F20101115_AAAZCL khanna_r_Page_139.QC.jpg
b82566b31c174d497afa7e271e1b5cb0
96d3d6c397bef92bd1e649cf494e7f63bdad73e5
11810 F20101115_AAAZBW khanna_r_Page_079.QC.jpg
6083206a37ecde753bfc90bd11d8f0ff
f36420846ea4d07cec662100eed45f302ce0122c
11883 F20101115_AAAYXF khanna_r_Page_123.QC.jpg
2fd72f20a701113279e4142aed92a961
45eaf4be3ccffbced6e90c93f613cf82e9fe6c78
6908 F20101115_AAAYWQ khanna_r_Page_055thm.jpg
55cecf726d057069d5f6620cd3c2c2c3
1d336fc841c8a1a0f60cd5be336a90819e6dc943
79696 F20101115_AAAXTO khanna_r_Page_019.jpg
d4b45f909b8a24d797a5845f6e26c13a
775df31049eae042ebbb85fa2281cf0abac8c0b6
79276 F20101115_AAAXUD khanna_r_Page_035.jpg
d786d11267a69cfd1b30e374180c539b
421ad83ca35c9b803854df7d648022787fd4789c
7199 F20101115_AAAZDA khanna_r_Page_022.QC.jpg
66f5af385724a6110e23814ade342479
63906368c193ab0771469f2ad042842b30785553
24339 F20101115_AAAZCM khanna_r_Page_087.QC.jpg
3967df60acd45abfa0a8428df97cf91a
3514a4b6450e072446e5aa49e7bb32bcf366cfe7
5626 F20101115_AAAZBX khanna_r_Page_083thm.jpg
6482afb2e87612adbd8a203c69d80294
268bbea8d83802c2b9403fdbb4e2545667276e14
7404 F20101115_AAAYXG khanna_r_Page_013thm.jpg
009b63d1838b2734b723505b3e733aed
0717e50a1f29b95238d9f0503709128352401645
4819 F20101115_AAAYWR khanna_r_Page_072thm.jpg
6c64e6e893fd4147e9def57a6ac24458
1a0c2a98f41f5f44afb7805f138625817ec47597
77251 F20101115_AAAXTP khanna_r_Page_021.jpg
a066289b96ba953c4e0b92bcc8b67071
13743559f73902bd1f6195bd686202020c8dc033
77150 F20101115_AAAXUE khanna_r_Page_036.jpg
ea67ab12e80a59256135841025b61d46
0ae5ba53330908093416340bdbdf5304753854b6
6790 F20101115_AAAZDB khanna_r_Page_149thm.jpg
e6cdb7d9f810b9f1d735c1eabd46a568
58c9eba910cbf7c785969d4ee1320a08976b9b4d
5812 F20101115_AAAZCN khanna_r_Page_030thm.jpg
d61ea92d67cd22333ea11b73d06e7aa5
de5ea65016b2c9404e403702b5faedb4b0a16baf
3211 F20101115_AAAZBY khanna_r_Page_070thm.jpg
42e8170fe00f7f700ab7e64f13b79325
9ca7b720d554ae2cbb599933e13af9e1772d9b42
12721 F20101115_AAAYWS khanna_r_Page_103.QC.jpg
0e03da65d5cd2aefb7fafeb2dab34041
ef9d9a5f3b1df1b3f196975276c3d75fa566b74d
22345 F20101115_AAAXTQ khanna_r_Page_022.jpg
2c6eafeb9cbdd922adef9de66f03a74d
0b1a727785126596eb75b2a592046c150486a05b
25873 F20101115_AAAZDC khanna_r_Page_033.QC.jpg
d2367bc781df65de86449175d00f9eee
c87fe57153ec0e9f2db73db4074ec2e450ddcff7
9764 F20101115_AAAZCO khanna_r_Page_132.QC.jpg
4133102d3033d73b6f441fe77eaa847e
830f905b8f346c02399ae24fe6316557b6e9c694
6698 F20101115_AAAZBZ khanna_r_Page_031thm.jpg
ed706106ccbb7b5b9e9cadb24988d498
70554e24db9332a02632110690319c243fc518ad
25262 F20101115_AAAYXH khanna_r_Page_036.QC.jpg
b5933caecce9007457ecaba414f89b59
bd3317d93f711dc9b136672d48e45dc478500613
10379 F20101115_AAAYWT khanna_r_Page_161.QC.jpg
cba6b1ccccd2b060b94a7085d84b7388
3d966d431cece6c32eff88f1d2213e5a88622b65
68555 F20101115_AAAXTR khanna_r_Page_023.jpg
e9eede6be9f3e0c6c382133cfc7d1f1c
bc790744198982dc74f6c46365e2d21176c5d38b
168873 F20101115_AAAYAA khanna_r_Page_045.jp2
a21034fce23abad68bc163fa4cc6805f
bd6e17136036b57ff1abba29f627d86c01d1b4d6
75642 F20101115_AAAXUF khanna_r_Page_037.jpg
80f6d89833d2353daccf7b4357ea1f31
647b21fe50834b13d20503f29d3f9e2326294cef
6546 F20101115_AAAZDD khanna_r_Page_090thm.jpg
6f37af37ccb7b22a9122b77b4b86537a
23b8697dda8fc1940b904e8923bf13bb6f421370
11106 F20101115_AAAZCP khanna_r_Page_126.QC.jpg
688ffbd09beb7a24459b036b4ec0125c
1f75d3b2d7482702cc172ac56edf47deec49848c
25076 F20101115_AAAYXI khanna_r_Page_088.QC.jpg
6eb4207c7d21c53a757ecfffa3a865aa
02a10978470b0eca2d7555af7fb7ec66b3ad1de3
4261 F20101115_AAAYWU khanna_r_Page_050thm.jpg
bbc9ce1169cd8e6f8d68cf69ba1d3c57
2f9b120fc94194401e37961e9a39111d9692f554
77078 F20101115_AAAXTS khanna_r_Page_024.jpg
282efa93332095c173beeab4ebe2034a
82211bba6c1ddc6b448061e482fcd2d5a69d288a
514558 F20101115_AAAYAB khanna_r_Page_046.jp2
627a9f45423f78893b36cd573c66681c
22fd54bf3464b6d7e01c6269edbb1af80fb53811
73918 F20101115_AAAXUG khanna_r_Page_038.jpg
26b55d6bf38e2bb77705afe9838c1a12
1cf18d5b320aaf70ee1056253c5de837370dba29
6857 F20101115_AAAZDE khanna_r_Page_110thm.jpg
7253f9cca3f685cc0cc1f94c781b7874
4b2ffe18f0218c07a8c7dbd07b97f575536a1066
25169 F20101115_AAAZCQ khanna_r_Page_056.QC.jpg
41a9a7198a8adff78c3acedcca75286e
6ec8924c3e489ed9bc8eb5e90a6a37ba40b71f02
3757 F20101115_AAAYXJ khanna_r_Page_098thm.jpg
cca819a1c1e1d399c22fe82d1d8b73c9
1f5f9c4114add4e272c6515afde09c71c9832bb8
6374 F20101115_AAAYWV khanna_r_Page_151thm.jpg
f7c9023d9388ec50936a4177f2499380
0783780f1f990c15b772f8d12f1071adc126e004
80460 F20101115_AAAXTT khanna_r_Page_025.jpg
e7a156d3b6d394aaf778cb004eb194d4
f956197a72603ed4b3a484203c4b529459a6627f
453884 F20101115_AAAYAC khanna_r_Page_047.jp2
de63157d8003ece4cf9d1c5b360da40f
3a5ebc923493bca07ca46fd83019aecc32dc0235
75354 F20101115_AAAXUH khanna_r_Page_039.jpg
4727fd690d956b010f3a2ce0bb65e6b0
807fd690601d233e4b801e2915cdf3c815e821a1
8391 F20101115_AAAZDF khanna_r_Page_154.QC.jpg
857cc67bf8c4317f980d28d3005a2669
af2796ac097ef63e8382007df9e251369ca5e0ce
10311 F20101115_AAAYXK khanna_r_Page_101.QC.jpg
a91ab32d9eadbb1dda88a5be24733cce
914cb7f45e73d3d79e54fda74dbb5649a6d20686
3890 F20101115_AAAYWW khanna_r_Page_123thm.jpg
4a4864b47bc01fdd5a95eed88ba9838f
5faed4d8d5b03f156541460afeaeb9a426be60bb
73454 F20101115_AAAXTU khanna_r_Page_026.jpg
e28d78c84e78d7d43ef9695ca97713df
55184d2ca892b1e6df0c4ec78dfd902bc603cea0
511313 F20101115_AAAYAD khanna_r_Page_050.jp2
65c3eafb0df06d271bc005b06dd05a23
8b3bcd749378a26e60fd4a5769c07579eb3277ef
74306 F20101115_AAAXUI khanna_r_Page_040.jpg
491270aefb23c23ced8a5a352e62da2c
cfd51fbd25125dbbc3718f638b911fac18270981
25574 F20101115_AAAZDG khanna_r_Page_091.QC.jpg
1bbf6eb4fd48e30813afba7258395a68
17179ee710db1e4865fb9c48552feedb68cc3016
10253 F20101115_AAAZCR khanna_r_Page_127.QC.jpg
5cae460eed9e68635512a5f01ea900b0
bd650aaad2c8bd28a08848fcf3a937096627e63a
3564 F20101115_AAAYYA khanna_r_Page_130thm.jpg
372005d1872e018d75b881803b3ceda3
eaff49385448512ca8de4313e90eb755c4627295
25350 F20101115_AAAYXL khanna_r_Page_007.QC.jpg
d2d736dafe0f714a1bee4453f5a97969
e6e4e0988a5ce4b9c9a6c3e35b58fb4350ff7ed2
6635 F20101115_AAAYWX khanna_r_Page_171thm.jpg
9cab94a665555c36c111d6252392d032
273e211dcca321c046d1ca0a337d133e47307508
68471 F20101115_AAAXTV khanna_r_Page_027.jpg
aaa717f7618bd0282a9762495e6763c0
f01b7f2700316c4975f3692c2844696f5873ff82
133462 F20101115_AAAYAE khanna_r_Page_051.jp2
02c02ad36df89da01a4504b8e0070b0f
88708f18a6cd42646f8d45c3f1f1776da6590910
67344 F20101115_AAAXUJ khanna_r_Page_041.jpg
db347479154196b0366e6a29a631d1ec
f524b8b8aec429ad50f69e6f2652ab1b409b46c9
16155 F20101115_AAAZDH khanna_r_Page_009.QC.jpg
1decf93086e299dad3fb61aed41a5f41
24599bdd76ce7156a52fd76aacce1a6d2b2355ff
15406 F20101115_AAAZCS khanna_r_Page_144.QC.jpg
d4219563586ae79d1747598a5f71d346
8d540c7741c73efa7954ee625cbdf1c9cf259d7f
23867 F20101115_AAAYYB khanna_r_Page_173.QC.jpg
6ee3b0e4151ff72c0015be9095e200db
ac69132aa55503c8d2d2cc151a69d3dce8d423ed
7160 F20101115_AAAYXM khanna_r_Page_019thm.jpg
270cf5e76983e740e8ebccd7178c410e
6b2168a0433b7dd5a2e60d32417ed72899dbf549
3294 F20101115_AAAYWY khanna_r_Page_132thm.jpg
66908d4e9119fa986ced47f2cc78ab45
a7720a8a413f9372fd809189abb5c4e56181e42e
66441 F20101115_AAAXTW khanna_r_Page_028.jpg
f351bb8a4dc2d771d543528b8e0115d3
d9bf1df309675d71dffb94491ec0bb93872b6872
118715 F20101115_AAAYAF khanna_r_Page_053.jp2
02fbe036bcf400eead11c77ae4880c39
b200cb923d8ce5e27ef9e885e57422e4cbc75dfd
34698 F20101115_AAAXUK khanna_r_Page_043.jpg
14b86daab68d4a7434d9ae8b3aed2698
37db22960ab2a5e8f9608303448e2def892e496d
11491 F20101115_AAAZDI khanna_r_Page_100.QC.jpg
3767f93ceaf634d3979f9ec2174bff19
3526b602976959021df673052e5a4bc7a38a31e8
11259 F20101115_AAAZCT khanna_r_Page_130.QC.jpg
2dffbff38cdf93329ae75abe3d9c3623
68c1a96569eb58cf75dedee7ecda8b3c4d30f71b
22523 F20101115_AAAYYC khanna_r_Page_014.QC.jpg
b4993dc1fa3134f1fd2b665e850e5a08
5c106db0bc5e1f37f299b5414938946ea374ac6c
3258 F20101115_AAAYXN khanna_r_Page_105thm.jpg
89d6bdb181f61347519516d6129d255b
56d2d78d80bc36c704b4010c4e135fb7d7868b63
24695 F20101115_AAAYWZ khanna_r_Page_037.QC.jpg
f61ef0c6aadbf2e2af510b27db95bf49
cfb7eab5872e80e5b27e3a97d5b8f203529ccea9
69997 F20101115_AAAXTX khanna_r_Page_029.jpg
bef88d9327132c8420145dabd591a7fb
8be2837c1401915699c73521e85d2e6a6af2ce48
113851 F20101115_AAAYAG khanna_r_Page_054.jp2
025a688e37d77a0f24ae8dfbb866c8ec
1e3dab32ccd3f8ee29dabaa8c5a01b0eee0c32a9
37132 F20101115_AAAXVA khanna_r_Page_064.jpg
9cd9120f2478f17ef238ff6c5d25e88a
c5f83e7739f09d143bbc03fdbd83e18bac1e0fbb
43710 F20101115_AAAXUL khanna_r_Page_044.jpg
7b7c13495325c9923d1c3aff3bf264a7
032cc725f27c6ff9aa1b944850cca37f1ff8c8e7
1945 F20101115_AAAZDJ khanna_r_Page_174thm.jpg
dead065b866a257c71894c70d2419dd1
561eaf026f672c708d3cc768a4862b67451323be
22599 F20101115_AAAZCU khanna_r_Page_027.QC.jpg
3dade529a871ce99c46965587dfd55f3
56b9a098b0cc3b58cef71435501bd4410370cd73
22834 F20101115_AAAYYD khanna_r_Page_115.QC.jpg
036f3251df30c3b7a8245e1709915271
4f755d861bcc5cf1f960cb663fd75fc28fcf91e8
3071 F20101115_AAAYXO khanna_r_Page_129thm.jpg
91811ec522497e9e82a92ff5ce9c7347
90924ebaf72d97c2c3c078e4cf784ee60ec51cdf
63065 F20101115_AAAXTY khanna_r_Page_030.jpg
7eedfdd92799ff849e10d73e3a911466
afc7a0e5541d4f3979ca3b9254c101617c0420ee
117496 F20101115_AAAYAH khanna_r_Page_055.jp2
37c5206bff5aa4b293643abc30952d96
fd9c07c54d5e643227712f76cbb498f762536d96
55675 F20101115_AAAXVB khanna_r_Page_065.jpg
98676bc4f68a52652e7df7c46c5dc326
2f93a6e678f575013821f54726e4d5383c2dfc39
20324 F20101115_AAAXUM khanna_r_Page_045.jpg
a4498638f2111e9216d707013893eca0
8430110196a4c7b9504e952b86c1b539eec50acb
6791 F20101115_AAAZDK khanna_r_Page_056thm.jpg
123df25c4b99331470b5f10dd619e495
4be2f5ef73f1f747674f50d1d822cc0502e0ea5c
4252 F20101115_AAAZCV khanna_r_Page_079thm.jpg
8d2eafb2dfa584cada3a79cb07994d64
28ff7265c3844088684a9f20ca7117311e8855ae
6260 F20101115_AAAYYE khanna_r_Page_089thm.jpg
6c49bf714db14fc38c01a5491996214a
29963ceff9772b86de85dea723852d3e18bdb7e4
5745 F20101115_AAAYXP khanna_r_Page_014thm.jpg
193693da480ecaf07730d44aceb35c67
59092e08bfedd8d716dad00add3d235a48e0dee5
76494 F20101115_AAAXTZ khanna_r_Page_031.jpg
04e568b94a26b480b97d34dba2fcd3fc
fb214ed6e4b49cd7154df5733267c4ef1332aae7
117614 F20101115_AAAYAI khanna_r_Page_056.jp2
8834601c4407e39ba16ed797050cec48
2f675884212ba47018c0999ecd27814501082b07
22809 F20101115_AAAXVC khanna_r_Page_066.jpg
19f43ccfa95fa83ed4e2111d9a2079b7
a62ddb8074db29a402545084c8e2734a47981e98
41473 F20101115_AAAXUN khanna_r_Page_046.jpg
69eacab19373125cde5d0d69b6dd6ecd
15c4f3883e741be042004d13dbf3a4474a790c18
3009 F20101115_AAAZDL khanna_r_Page_155thm.jpg
0d5ed078e91fd4ed51f427a3a83006b9
42c4f7e0240e252acc9548a31d29af772c2181ae
6605 F20101115_AAAZCW khanna_r_Page_074thm.jpg
0e06c832f5e8eac69c2a37ca011a18eb
05d59571a8c2e1da92deaaadd589605571b47486
7014 F20101115_AAAYYF khanna_r_Page_026thm.jpg
71ef598f47354b9ff00c4ca1a024a279
11ce097b02be2fcff98dc868820771dac00ca6c6
6784 F20101115_AAAYXQ khanna_r_Page_024thm.jpg
52a9c037e5fe7800794f70b41036d2ba
890d4249020cb553aa371297c5fe8d6e4a78eaaf
119007 F20101115_AAAYAJ khanna_r_Page_057.jp2
761044bb1eb9de710869d34eb5c379cb
9cbb568e338cfd1abc3cac3adae525214a500a38
32750 F20101115_AAAXVD khanna_r_Page_067.jpg
15a5d9345777bf90a08b352b7ba6a375
0e116fb540c7644f48da71be23c221c2eaacc16d
35078 F20101115_AAAXUO khanna_r_Page_047.jpg
4a6267e5109e04543f71de418cdd1cda
bb5790f392ac3ffd3b8c5717020e8650eddf1767
12263 F20101115_AAAZEA khanna_r_Page_064.QC.jpg
ba066bba3d1affb9751b8749adb612ad
d62dded637d29e0c8aeeee3e0f597fed1c17d9cf
3170 F20101115_AAAZDM khanna_r_Page_043thm.jpg
448eae1904b0982f093a3006ffad844a
08b8cc71736ea09dee3fcad46e2e785e1454e15d
11030 F20101115_AAAZCX khanna_r_Page_124.QC.jpg
3024ba168bf748e71a89d17e4c49ad21
617601bd5939183821c8e72b096c9ae4bf9d0591
6760 F20101115_AAAYYG khanna_r_Page_040thm.jpg
6ff1cd75dd62049cec5aaa07215fd679
3494293d4c80e79221cc140ab520717cb8710429
3467 F20101115_AAAYXR khanna_r_Page_119thm.jpg
1aab99fb1674b3e5760914451b851f8c
a814d064ed2b79a2632346229b7880e6cc81d973
116662 F20101115_AAAYAK khanna_r_Page_058.jp2
f3ce8d94735c7988f4e56eb73e69871f
aeedec14140c64f87dc3809f4d0cffc732561c51
29498 F20101115_AAAXVE khanna_r_Page_068.jpg
f2c89ef0f3b4559a86d5fc185f677ecf
fb09b9031744fccc1bc5f2ca2546ea13af95d779
42368 F20101115_AAAXUP khanna_r_Page_048.jpg
d7ecf2bfc12d80190a74588ce5cda862
a7cf1d9d198f67a25aa61b303d1bc529d464bc6f
22638 F20101115_AAAZEB khanna_r_Page_171.QC.jpg
51585059c25f0d5389849301875e9011
24beaad2c8f434a993d05d8ba72ef5c8280d6036
7058 F20101115_AAAZDN khanna_r_Page_018thm.jpg
18e9251edc832e1a8ebdd3c8f4e94d12
4bef48c774cc520b1ce334cc4f2d6ed23e9682d1
3042 F20101115_AAAZCY khanna_r_Page_062thm.jpg
275f5cd5d01b074c83538d42a9415129
657b6b0539c809606af878464afe7fd62acd9d8f
6745 F20101115_AAAYYH khanna_r_Page_169thm.jpg
dc66655852bc4c43802e6dfbfa0f1638
202d754c81e7e20e9118924d9d52a602d6f04886
11065 F20101115_AAAYXS khanna_r_Page_043.QC.jpg
9c3c9f0ca4f260a9b1c8817adda8661c
32b9913916f37a0eda6bb7c225dee71102f97bc5
112533 F20101115_AAAYAL khanna_r_Page_059.jp2
93e261dc5c8d02aa19c9a7938066729e
67e220082d5870bbd1f162bdc4e609b0f6323b4c
22899 F20101115_AAAXVF khanna_r_Page_069.jpg
e274fd1b3f67839101c2b12ec23f9b9b
0c04220d7b9db77ce8e5a2bbeed5bdbcc3cf3d0c
47165 F20101115_AAAXUQ khanna_r_Page_049.jpg
bbbff3419185f8c8626640213352a49b
7671c483b09cc810ec9e28cdaf6118afc1160b15
25340 F20101115_AAAZEC khanna_r_Page_109.QC.jpg
b30553dd275b54c97dfe60a9be8f257e
5d39f863a1fdaa53e9ef954d95515d31fb903f1a
24439 F20101115_AAAZDO khanna_r_Page_026.QC.jpg
6058866e9072beb10e3f57f1811b099c
5e685b8ddbe5a33472933420501ed99b7303956f
25217 F20101115_AAAZCZ khanna_r_Page_054.QC.jpg
2c20822acfdb84642a9cf5879f725f65
37cdc1b3e17dc2334a68bccd6e8a76f0774d1fc7
13378 F20101115_AAAYXT khanna_r_Page_050.QC.jpg
b1fc5742117cd94460d61cdc97a87ded
53dbe5c0aeb686fac948db6270ecc9edcbb309fa
23357 F20101115_AAAYAM khanna_r_Page_061.jp2
7f69060ef2186016e6b305900a1fa479
39617f209c354f693e349ff73313841d06c804fc
42531 F20101115_AAAXUR khanna_r_Page_050.jpg
0f8b2f86f0cb6aa096c31b2de3b6f66e
dc1f4f8f43b6cb764cd166eafb1ba6f893544143
33846 F20101115_AAAYBA khanna_r_Page_079.jp2
1bfd92c3e1b2ea8c6202f065bec54db6
9ec9de74ffc29a22c0ad259c8852ccbd60595317
11443 F20101115_AAAZED khanna_r_Page_146.QC.jpg
910abe13b5d0fafe4c8f760359332b46
089d190f3ffb17eadc4d69aa327e6ca48beeff4f
10859 F20101115_AAAZDP khanna_r_Page_121.QC.jpg
7a794213d9141b6d5dbb82ac0aea3668
aa06aecc23c3fefc9cd1b36d02290c42bcd422ec
25443 F20101115_AAAYYI khanna_r_Page_167.QC.jpg
1142f7a7317a568d192b0a1b0ecd26b2
2cd150d1521c643b5c336e9432813d2f42775b31
20805 F20101115_AAAYXU khanna_r_Page_060.QC.jpg
813dc4e196540a7162fbcd0eb0b12341
16a673b0ed0b6676d6adb26f03d324bc0bffe772
31401 F20101115_AAAYAN khanna_r_Page_062.jp2
22eb2bf97487efaf8fb70922bfa8c7c3
6814ad4955d5914160169924735e66fc2994f5e5
30008 F20101115_AAAXVG khanna_r_Page_070.jpg
312f5471fb041f090cc52dcad88a562e
fc4e8f28136a93e544da65b6732bd2139bfe0ecb
37600 F20101115_AAAXUS khanna_r_Page_052.jpg
3ecfdabd3f84660ba086cc4bc6721b50
7a44b12b595275af87024c33f6aee4d54ad67080
892704 F20101115_AAAYBB khanna_r_Page_082.jp2
741c4fb608ee72708911920f5d318a1d
d328bb534d6784fb8fd0e7a5dfa5c482e897f271
6624 F20101115_AAAZEE khanna_r_Page_162thm.jpg
4f0224a2ee22c5b393ff82643933eafc
a2c1a04102f33f678386cfc25dd72c7b1d3d9a52
22659 F20101115_AAAZDQ khanna_r_Page_023.QC.jpg
2318b3e44d45ad22a499babff5d115d0
a5daa60df6a17cffcda6d0bcb11270832290dc77
F20101115_AAAYYJ khanna_r_Page_131.QC.jpg
efcf7381307eb9bceccc8a15beee7ffb
3f4a8123637d777b279655724f8225116832f959
6577 F20101115_AAAYXV khanna_r_Page_114thm.jpg
bd243df126bf991d8fbcf74840c1eaef
9fd6dc4fa5b90e2274ba62f502a1c852f13dc70c
25668 F20101115_AAAYAO khanna_r_Page_063.jp2
33f024fb9bdd2bed35751c1801eabadb
e1a6f68a2e68e874e8c21e8a6630b132b865e929
41415 F20101115_AAAXVH khanna_r_Page_071.jpg
0f51f6ec7d597ee73b75c7b6ee576784
1e9b476c6e5b3f473ea2e00d40a9f6b1a3faf4c2
79680 F20101115_AAAXUT khanna_r_Page_053.jpg
ffe8d6f62d9782252a9cab82811fb742
c72ab353d5cfd7029599a73f8b87fe2db8161463
74168 F20101115_AAAYBC khanna_r_Page_083.jp2
76ea8c95bb1497e6718e3747f5a56316
24793950749ae52cc435b409e32a2740549c5646
3810 F20101115_AAAZEF khanna_r_Page_158thm.jpg
d5b51638d9c73350c3fbc3155329cb09
0980854c77ee73f613d8dcf48b36f381c8bd7d3e
25193 F20101115_AAAZDR khanna_r_Page_138.QC.jpg
3a44291895f7b5618ca6fcb0aecc3538
f5c1d92eee0e3dc3a8627142fb0826d39671d735
7124 F20101115_AAAYYK khanna_r_Page_053thm.jpg
5d1515e4e53c1313d1cd3b6f68c1481f
a22f200fe73b6f0ec3bf7bdbef7db5053eb90c03
6722 F20101115_AAAYXW khanna_r_Page_137thm.jpg
9233aa919a31083d556de66a110309a2
6104f9dd898b2ea43abd0a25ce03e2d52adccb3d
23209 F20101115_AAAYAP khanna_r_Page_066.jp2
31be4b1417d1957daf22aca948d78dbd
1fbfcfe07009886d81eba12261c147f8b316f16a
48448 F20101115_AAAXVI khanna_r_Page_072.jpg
d11fe563c168397cddcad85be2d1bbdf
99c9770d0689e42810ed99d03dd6f2f4c284eed0
75618 F20101115_AAAXUU khanna_r_Page_054.jpg
0c8f6892c258f50c197506111dccc818
3887de64769d411647c7855440d337bf38fd51aa
36884 F20101115_AAAYBD khanna_r_Page_086.jp2
2f977d6ddd29685be249d29e5054e78d
ebc595889ec3303dd4112ed836b1bc1bdd0d8477
14620 F20101115_AAAZEG khanna_r_Page_116.QC.jpg
680804ddc7f06bec0ae8c52ffdf94b80
96117899d1c5edccbc4d547b0038b64da9648ed4
F20101115_AAAYZA khanna_r_Page_128.QC.jpg
14a5ec8cc8d3c8bd1e5ee9c7e9ee34b5
d51d1ed84412d967a5c2d32ff680bfc24b873b3c
24510 F20101115_AAAYYL khanna_r_Page_137.QC.jpg
98e9a5f1feb5ea20540678523dfd3153
cc644a0adea9b59ef54bc0acfd13f3cc43023af3
3988 F20101115_AAAYXX khanna_r_Page_046thm.jpg
0240f24109dcaaaeb5e5654cb03d41ae
bc7aa39bd152c4a949689d183dc9f679e0f21e7f
76373 F20101115_AAAXVJ khanna_r_Page_073.jpg
40520367a3787ba02b34a0511d263ff2
bcb61bb8c6ac60928bd7d4f0cedda43669e21e61
78926 F20101115_AAAXUV khanna_r_Page_057.jpg
7bebe28c18133bcabb67b0397081d733
75de068f93e55890a7bc2d9ef61abbdd8811a521
112845 F20101115_AAAYBE khanna_r_Page_087.jp2
1cdff068e075b4b180e6a16b7e056029
fcc6b6bec1094ae605dd1ae1d12c57b5d72c1759
6486 F20101115_AAAZEH khanna_r_Page_077thm.jpg
cfbe744735d34c8bcc6106da1a0983e8
e1e1c44ca1fa420be08f184c80dd86e9c0bec28b
23126 F20101115_AAAZDS khanna_r_Page_150.QC.jpg
0aefb0a4690786cbb417a09179ba8b00
ec14c60953edcafbab0d3ca984445dee7eeeb1fd
10198 F20101115_AAAYZB khanna_r_Page_134.QC.jpg
7722f38943127f1e9ddb248d924da157
7319a824a1374c1e9d86f7044b2fec8459455fa5
3427 F20101115_AAAYYM khanna_r_Page_101thm.jpg
b2771a34459e01dc0949b65fb3321b5d
9cc1c2867e0ada8f3935af5676e7fee2acae0126
10490 F20101115_AAAYXY khanna_r_Page_097.QC.jpg
ab1de904e1d178377a1c9fb038f5494c
e42f2e0c93758f0cf2e267c63498315f410df3e8
352155 F20101115_AAAYAQ khanna_r_Page_067.jp2
0b84b9360945016bce60fd002570eb85
5220a86fcfe22e9666b83e7cce8627a098de0447
70953 F20101115_AAAXVK khanna_r_Page_074.jpg
caaa4b28d5e5f634037adb39d2526e03
7f931ea2df1eef7c10a2c4cc0e16bf8e6bdaee27
76379 F20101115_AAAXUW khanna_r_Page_058.jpg
ebdb6f5825f7e7cd78af4e2c29a26a4b
79c30c01e24d45b87e32d2f781c1e95c64455799
104805 F20101115_AAAYBF khanna_r_Page_089.jp2
d100a74783bc69a69fe05029482ea806
1007db168a646002481df207586d1f83dd4b3782
260843 F20101115_AAAZEI UFE0021183_00001.xml
559b0154419d565617080b296811f807
bc14ac2556c620cf0ea05fd973df9b92ab4d3c13
11682 F20101115_AAAZDT khanna_r_Page_118.QC.jpg
2cc0e737aa35df960ac03eb317571942
52f256af31db7e27078679648c5b33f003a1d487
17621 F20101115_AAAYZC khanna_r_Page_156.QC.jpg
4f1b06d310b1ea5f8834748c447fba36
be67729fc522c07f26b379023f7dcf62e337fa7e
3579 F20101115_AAAYYN khanna_r_Page_142thm.jpg
faec861bc6a590c0bb97b0366c6a37b8
bf1ba5bac0b48825b5f4fb9c53f9e342fab6ce49
7243 F20101115_AAAYXZ khanna_r_Page_001.QC.jpg
b97753edd7bff77ebeab1ab36f9346ee
21b71e14554c252a57ea37d6d8c46d8ef4c899b8
30833 F20101115_AAAXWA khanna_r_Page_096.jpg
02b04bd16029e2e449fd2b4099ef0cf7
c09bddffb308415b4e902af2189438d37d31cb22
33938 F20101115_AAAYAR khanna_r_Page_068.jp2
1d98a8372e320ef8e0e6073d7e1f6070
c02dd0e122207d697450f45e57f6b615216e4a11
74063 F20101115_AAAXVL khanna_r_Page_076.jpg
aa8f969067042d07881e977baf817732
30aff6aebe8111e2d02533677b6e78d710b72d0a
73462 F20101115_AAAXUX khanna_r_Page_059.jpg
ae6dd689cde4bb691866cf10f8796d23
87b78d2068a1cdc6d10d52d1706ebf20d4a36bcc
103678 F20101115_AAAYBG khanna_r_Page_090.jp2
8d2a7e622ded2f198c0cb5c67c23ea13
2176d6a22b3cf62bfe3d99efeb9d33a243924bc6
2964 F20101115_AAAZEJ khanna_r_Page_003.QC.jpg
a0a7606dcacc34f8fbc6ac7f0c4e1b21
154ee82d6ec63547a7ff13a3c274c552ce7055d2
6759 F20101115_AAAZDU khanna_r_Page_163thm.jpg
a70db6f727d314051fba93765fcf8923
7e97aabf89ac4058b1d85308cfe8a190d00621aa
7163 F20101115_AAAYZD khanna_r_Page_057thm.jpg
9fa13618e2b07c4708361a9ae65cff89
734156dbf001b90576146e9e7f14531eaa3db688
3544 F20101115_AAAYYO khanna_r_Page_081thm.jpg
e5c4d368a7e75ba0a463f5c24e0ce0b9
faaee88e128f82d2b7006aad85b64f669db0f666
39026 F20101115_AAAXWB khanna_r_Page_098.jpg
b2550074f24c8eed2126a070118b8e24
80fe279ed28dcad05d11c5c7579f752b60d227e8
24335 F20101115_AAAYAS khanna_r_Page_069.jp2
d2cf5ff1d9d5250902e26a5a0ed7f0c5
444c18ca3d04dc94e09c4a7060755ee30f033862
32415 F20101115_AAAXVM khanna_r_Page_080.jpg
363ebd437c73602f475e0e511549ebd6
a449a29d3f170932fe4a6e0c881515a6561a1e8a
63719 F20101115_AAAXUY khanna_r_Page_060.jpg
553e54851dc6fc340cc0d318c96f3aed
6ba1b1071a579f70e44fb0d3bae03b42a342025c
117539 F20101115_AAAYBH khanna_r_Page_091.jp2
3f665e67571d7210bab8e7d6ab7bfe91
539674a5a05671dae0ba066bb08fc6bbfae98af0
5429 F20101115_AAAZEK khanna_r_Page_006thm.jpg
b7044db39eabed872558a8afc586754f
c55cabd57b08967f0a0ec7ede647c1d4bd1e163d
4722 F20101115_AAAZDV khanna_r_Page_144thm.jpg
e6819b01e594bbec1c9269328f99d4e6
7830ffbe464d1b5ce69e27e1ce74f89a0e9c83a2
12793 F20101115_AAAYZE khanna_r_Page_098.QC.jpg
ab4836ad23cd910d20f44e5a20a5001a
4e8b1e3d2759e16b2fe00dae5e7948cac6ea9aaf
23382 F20101115_AAAYYP khanna_r_Page_114.QC.jpg
8c2bd3708594820d87c0326632b29244
19b613156b5cbcf20afc7cc88fdb9017d52143bb
30228 F20101115_AAAXWC khanna_r_Page_099.jpg
8f95ff571c86339109a52ff3e9ed6535
24eeda1022420fc4bd762aad95593b69018d0c14
497524 F20101115_AAAYAT khanna_r_Page_070.jp2
cb99a59cb1a385060659478fc7a4ac6a
079376871a2e7a8f5a27e8b2b705b170693e6c20
32744 F20101115_AAAXVN khanna_r_Page_081.jpg
2e1a430b4eb01a524b5fb7a7d98e2fe5
6eec2c08a98cf71864045be8149ceb4ee520c239
25920 F20101115_AAAXUZ khanna_r_Page_062.jpg
f7558e09510d1859c622714df7a3caff
e50fef6c1b64eccab138ea72654af12d9e539203
114459 F20101115_AAAYBI khanna_r_Page_092.jp2
4c1a3dfce100cd795a8934c5d85f1ef6
2536abebf45ed4fa7da9840ea25e0fc744254c34
26171 F20101115_AAAZEL khanna_r_Page_019.QC.jpg
cb935ae411aa913b8a81c129713001de
dda1460a0f587052b1ce37cfb7c9dafbf6364bbf
F20101115_AAAZDW khanna_r_Page_068thm.jpg
027822f663c152a3987a5d2b16ab7616
2f3b09855d2e1c0249be8ccc5a89b5738efd38af
6932 F20101115_AAAYZF khanna_r_Page_106thm.jpg
e84334607dd90870121ee10225ac3db1
832365f224a003b4d7f9b81e742a1c3d59537dcf
6747 F20101115_AAAYYQ khanna_r_Page_073thm.jpg
5ae875a22720c82fc8b088124dac917b
ba9c4e10725ce7e56a71a4cf690617a2d7b3b68f
32651 F20101115_AAAXWD khanna_r_Page_100.jpg
682ca0b7b99b62329877eb2ec966caa3
2d081f746a49b7c8beb88ab86dcc3e87f0f0f3b1
54097 F20101115_AAAYAU khanna_r_Page_071.jp2
ae6b76ce66509e905de459696f3c35df
cf2f771b980b8088dee362a3f37b65d21dd0206a
57217 F20101115_AAAXVO khanna_r_Page_083.jpg
975c2da27dd1a9032a857e249a6465a5
7fb3311d26ba266b963a79cb8d633a2ea77089ca
108398 F20101115_AAAYBJ khanna_r_Page_093.jp2
674f1c5a653bb1b1b92ae0befe23f9a5
c40a957b613cfe010452c97566605bae3830d0f8
5249 F20101115_AAAZFA khanna_r_Page_082thm.jpg
01e06b1f0f8e2dad80aaa9e3873c6846
b2855f3947826478f21a667eb89aaae7fa7ef8ef
22449 F20101115_AAAZEM khanna_r_Page_029.QC.jpg
f5103ff02a170553633a9abea995f518
99c50e1155bdb43853dab693aca5e1ed6cc0d8a7
6802 F20101115_AAAZDX khanna_r_Page_107thm.jpg
1f5b1b1b9e374eca0d9c7b24b3a3cbda
c2cd5143c7903351702bc4afff3260aaf0953a92
3576 F20101115_AAAYZG khanna_r_Page_131thm.jpg
5c9ca4e3ea3e27a19240cf4b76c65f70
83c5d81f1a902e8050f6968bc39f5d0b53ea070b
25198 F20101115_AAAYYR khanna_r_Page_073.QC.jpg
9e5b70e15298e93362fb2affae7f911f
213096e0e7a3cba60fa2133e7422887628764a84
32175 F20101115_AAAXWE khanna_r_Page_101.jpg
d43a047b1836c08a739bd0820be9ba00
bb7d65b236b450c8beacc380f3d543bde5243ae8
113127 F20101115_AAAYAV khanna_r_Page_073.jp2
b8672ec88b49b627f4a0d3597a700d24
3203b286ec06265628722f9a72e7345e7e03717e
57211 F20101115_AAAXVP khanna_r_Page_084.jpg
34fdf5efccab0269d189d77ce326cee2
6293d0237a97898f1da309d40d74f624282b126f
19180 F20101115_AAAYBK khanna_r_Page_094.jp2
d163258b50e8f57041bfa760d36b0f78
3258f06bfc0d50b6fbf8af624b64df5527722469
18091 F20101115_AAAZFB khanna_r_Page_085.QC.jpg
67dd93b9f6c453309e759de7aa76db87
109c962dfbc1dcf48fa32e5b64b7291acadc6752
5741 F20101115_AAAZEN khanna_r_Page_032thm.jpg
e39d46e2fd91f86a6d11e5b587e3f0a6
58a2567c137dfcee255f59567e17d6e8ac03aae1
5687 F20101115_AAAZDY khanna_r_Page_084thm.jpg
b379426778859d417e477993aad59153
a4af15abe1c20fad87e9529d78b457b94657d2db
26706 F20101115_AAAYZH khanna_r_Page_025.QC.jpg
fc7da7436101cd598def938ff3e1ed68
a442e6aba5dc09ca47244221bf94c52c8a47b8b9
3859 F20101115_AAAYYS khanna_r_Page_095thm.jpg
2874546cee9298242c5b99cefd063117
8727f0218cbd3ecc1ef503d1cdb8df9381433846
36442 F20101115_AAAXWF khanna_r_Page_102.jpg
88b930758e7b62ec50e2c76e7a334c97
be67d9751642d909456f1ff94857ea192434f3b8
107130 F20101115_AAAYAW khanna_r_Page_074.jp2
bfd1099d71ec917ae99b6ec3c8fc7d67
60d4f4ec167bc34b98a2cd2f3094e24b2be24974
55365 F20101115_AAAXVQ khanna_r_Page_085.jpg
5a0e806ce84d31ef4b254eedb101eca8
254b3ee46e6e4af73cf43bd9306a8c87ffb37984
471062 F20101115_AAAYBL khanna_r_Page_095.jp2
b8f31e00fb1cabe1dd32a338a4cc0994
5947b2598715c82b1bcadf2ba4a0a7a92fd75d99
5700 F20101115_AAAZFC khanna_r_Page_085thm.jpg
a01cb1cd3e4ae77e496832e35cec389f
3f7cb0b67f3227ede4b0b791c6b3ba701bc14bf5
25882 F20101115_AAAZEO khanna_r_Page_034.QC.jpg
c99189b743801f41fe3e9b2969324e8b
9f01b8e7eb24b8d1110aa6410a74f258f88cbd1d
6244 F20101115_AAAZDZ khanna_r_Page_015thm.jpg
2736c85c3998b3906ca2b24f038a3dcd
1918dcef324e4f47a804045e478f4af35e71890b
7950 F20101115_AAAYZI khanna_r_Page_061.QC.jpg
0a8eac295418246a3d4fb65c9afdcdb3
4a5307b65ddfa7a5ccbbfe4d6ed910e07975a4fe
3485 F20101115_AAAYYT khanna_r_Page_160thm.jpg
6f72434a5219f016002cfe69b1ee7d1b
3358106af9dff889e7f0a0e7ded9a5dfa4e9bfb3
39267 F20101115_AAAXWG khanna_r_Page_103.jpg
b3443cc9bb6c11d01402c15eabb26802
c1bd7382b184f6963c94e839cc563b77e2a8c695
114849 F20101115_AAAYAX khanna_r_Page_075.jp2
e2dc09fbae4276dbd0de8988ffb22450
2a87029a697e039a09bc3f83b0a5162193fd5750
33594 F20101115_AAAXVR khanna_r_Page_086.jpg
00ba54eb5ba51fe8db3f1f17ef2923a4
d0ddf94fa12a07a3d8d906cf9b2819b6004eb2be
112336 F20101115_AAAYCA khanna_r_Page_113.jp2
b06a1f318e12783517f993ad38eadf7f
5411abb54337e9bf3fbfdb82c3770a3587928fe2
34013 F20101115_AAAYBM khanna_r_Page_096.jp2
32e6382d065a1e764a9ed299d014c8ae
54949d6119a538b3f64e619ee4be9c2d7fd7e2e8
23311 F20101115_AAAZFD khanna_r_Page_089.QC.jpg
4e282e2ee81a81e12f8156d222879f25
896cb5ce3232482e662cf1862c20facbf68d1285
6746 F20101115_AAAZEP khanna_r_Page_039thm.jpg
b22a6778075b1d87739d7741d061085d
0bd9dc041ffcd17817be72b533ac49952d41b369
12412 F20101115_AAAYYU khanna_r_Page_047.QC.jpg
3460aeb5b4857a02079f2aaa79458543
891cae55893df0f39d0d5483fb764b7676db70ff
112116 F20101115_AAAYAY khanna_r_Page_076.jp2
f1d5b6b1353ed6cda492b098da330d2b
11e681fdf9dd616c3c1b810d45cdf36c9efc4582
76289 F20101115_AAAXVS khanna_r_Page_088.jpg
a8367afa1aa8788acea37db22b5d7970
f9baebc5a21ef36106abeebddcc8fa7b23ae2ea9
107383 F20101115_AAAYCB khanna_r_Page_114.jp2
f339068fc6e8e385daffe70396d7a0dd
527313cd9d315842b227568472474df249a0b19d
32375 F20101115_AAAYBN khanna_r_Page_097.jp2
6e638caaddc02a0aef60b509d7c18b6a
f4f2d06e111215ee79b286548cc7e0e3e2abd65f
6846 F20101115_AAAZFE khanna_r_Page_091thm.jpg
d6f8050a3ff9a0d737a4bfcf0f8cab1b
d6cc1aaf19c26ce1df84508519cabae846a589c1
4165 F20101115_AAAZEQ khanna_r_Page_047thm.jpg
6b2c108c96cfb556b7bc0d23c8055bdf
7a5b5de71462baddcad2a25941f9f422d2fd978c
25909 F20101115_AAAYZJ khanna_r_Page_057.QC.jpg
bb0ec8f6e9d7f784c7679aed9fe22f8e
16379fc7d15eb5fdc82d4fbfd9e6739f24a47e6a
10920 F20101115_AAAYYV khanna_r_Page_122.QC.jpg
bb74aefd95dd821ca3921a43fb5447df
142453aa7425a641e21156d62b74c1e0707f46de
29125 F20101115_AAAXWH khanna_r_Page_104.jpg
0f638f9709349259b37adaddaad0eb28
c75c89be663eca5fb7505d1d9dfd1fe795098a72
49942 F20101115_AAAYAZ khanna_r_Page_078.jp2
c7ea1aced8689291272e8e100f3d4d8f
3f37ffcedb0411c3aa4ba44292bc94ce7e44dea7
70442 F20101115_AAAXVT khanna_r_Page_089.jpg
e1af101be0b2cc6a9298c00365f7556a
88bcc3658677526fbc9ade75b34098980ea3a967
110883 F20101115_AAAYCC khanna_r_Page_115.jp2
ffd70768cd3e7017923a6791ead764d6
66c668caefd0a720fe56e64313e734150156c723
333251 F20101115_AAAYBO khanna_r_Page_100.jp2
28cb2d4c82b4381cd5a3b4283897fd83
6829ec161b96f7a644b0301a669c6e11a19916c4
12725 F20101115_AAAZFF khanna_r_Page_095.QC.jpg
d69d9b09046593c3b52d2dc1552415c8
860efe1158190658c91c1dc729491c1e527ea7ee
6968 F20101115_AAAZER khanna_r_Page_059thm.jpg
906e1ce36655c4fa0a35c1ca23bf7126
4730544774451bedb27ec9480b4d3a828898bab7
24636 F20101115_AAAYZK khanna_r_Page_108.QC.jpg
2e6a89ad83d0d587f98d18278c0ef3d9
e54ee2ebd51edaf0b9610c39e979e800c82842be
6316 F20101115_AAAYYW khanna_r_Page_041thm.jpg
f9f2cf2308fabf0ae98d3b14da075a19
b8bfb1857c51308760880565d0f195ee29f927bb
78042 F20101115_AAAXWI khanna_r_Page_106.jpg
985a31b041919eaf829085cf50b6a4da
3dc4b77ca4ef4e55c38faaec054a05dda17622a7
69548 F20101115_AAAXVU khanna_r_Page_090.jpg
5d900341bcba9af46d642b034086c78b
107375b5064909bc242eaf0ccc3dd1ed61ca9134
60066 F20101115_AAAYCD khanna_r_Page_116.jp2
f09389ac4e2dba271b8c964ee8132654
a344da0e6c16d76fdbaf72a32a2a6ff02edc2e95
372811 F20101115_AAAYBP khanna_r_Page_101.jp2
cf4d755c0837960096c60da9c397cd7c
5f674e0fc3144a0a1e9d856546d626f0e5f3d485
3807 F20101115_AAAZFG khanna_r_Page_102thm.jpg
894171af441076997ddc8998cfe0be2b
c4ffe1d4cb253231bd2f4463b15deef2fd08a5b4
5842 F20101115_AAAZES khanna_r_Page_060thm.jpg
5fac20596eadd56bd6f6f6d057f21ac0
d454a850ff4dcd4a215487076e3e5061f265fe2e
F20101115_AAAYZL khanna_r_Page_128thm.jpg
ad5ab1aa96ca18a81b445ad6848d122b
af1e4217521444259324e178f46aaedd1fb436d3
11583 F20101115_AAAYYX khanna_r_Page_102.QC.jpg
fe5415fa5e9c85df332238a8dba88b01
1baa68eef1348dc87c6e581dfe8a31d0f7aa7480
75432 F20101115_AAAXWJ khanna_r_Page_107.jpg
11a5b169626b6d48503f13d568473e6d
053e1a63614e3a934751f58c23b2601c1c51f5d1
77033 F20101115_AAAXVV khanna_r_Page_091.jpg
70699cbcdf0716d919dd087ef034647d
6b093527dbda060b7fb3241eb7a4e2ddee4f182f
34760 F20101115_AAAYCE khanna_r_Page_118.jp2
9b2e29081d6c86f164b24b101453a98d
e54886e2d19c75768b49168dfe288778a11f884e
47046 F20101115_AAAYBQ khanna_r_Page_102.jp2
5b2ff373b4512c4fdfbc8b12f361747e
a8e7039a3fae9b5e12498d18dd65d27eaf5167eb
24372 F20101115_AAAZFH khanna_r_Page_107.QC.jpg
d8ac4f7b7f8fd15ccbff40fa939a0a19
1c3b7966e1962fe7c50b97de720f585a2cf6b271
F20101115_AAAYZM khanna_r_Page_037thm.jpg
b26dc7e5686b3cc2e82fb1c0a8914383
59ccfaa02a84104a519e2ce5a455b3c78141346f
6550 F20101115_AAAYYY khanna_r_Page_150thm.jpg
1ab0da7f32a3e08d9d1e03ea30c3571a
6b5e82a96048ef04efb9182bb81cc1f18556bb92
73126 F20101115_AAAXWK khanna_r_Page_108.jpg
e9d341bce33c20c8e119e9c42fa7b30e
55d46f29ae71594bf12d7f3ef800e0b3d952a32c
75786 F20101115_AAAXVW khanna_r_Page_092.jpg
08d6d03e88ff73fe1f65ffd8f496e85e
4100396f0600f81a80f6267ddec9cc8716444bdf
45808 F20101115_AAAYCF khanna_r_Page_119.jp2
b82ba2eaf2b6098602f7d5fec08d17fd
fbb0f9d8ca2dcd4e59ae45dae9d58127476fd77a
6946 F20101115_AAAZFI khanna_r_Page_108thm.jpg
34241eb72e0cf9a6fabaa537c534d79d
18defd82ae4979f34b41b603fe8bc3f38f8319c8
8611 F20101115_AAAZET khanna_r_Page_062.QC.jpg
8f3e0e75ce097e82d9ea0ff3e1017cd0
aa9e7d99307051986e9cba082a777aeff7951ddd
6497 F20101115_AAAYZN khanna_r_Page_115thm.jpg
d015f6a46c0792efb47d46161e380e7a
478623f45a968ac79392594e3b063f142cd4c73d
23671 F20101115_AAAYYZ khanna_r_Page_112.QC.jpg
5cda3dcf2b957876b3fc3ee27ff6405d
f499e99760bd236fc02328612210f7891d495d3c
77736 F20101115_AAAXWL khanna_r_Page_109.jpg
05f5dbe8394cf9b8f6036a5608666fe3
f372eb36741a9214e5353da3d9390ce24c40167b
72498 F20101115_AAAXVX khanna_r_Page_093.jpg
e0ba8a5725ee30481df1e77b18258f43
a3ba5b83083bb9d9e9cea8bdf0cd6410dbd84327
607259 F20101115_AAAYCG khanna_r_Page_120.jp2
7812f6ea04ae2e0cc526963043d49931
25c2ade43d045846f706cc1896a87dc8f4183bb5
31582 F20101115_AAAXXA khanna_r_Page_126.jpg
d5bfff67767a28b2b8cbf4d74ed00c3f
3b6b361d0c1a4219f889818e4b7003913f07f08a
50277 F20101115_AAAYBR khanna_r_Page_103.jp2
483b1225fdc98bd428b5a3bef2df9b29
48d8b8c83703b5a52c7a1ac310ec574b7ed83e07
7093 F20101115_AAAZFJ khanna_r_Page_111thm.jpg
133f8d129001e743763ba168a81e3cb3
3ebafb41e5f5c4f72511d7ab7c6839d52f913039
4577 F20101115_AAAZEU khanna_r_Page_065thm.jpg
71777ca8d54354110432b03efbab8ccb
f2e011ae320b942412071f354e9f0eb067dc73a2
24761 F20101115_AAAYZO khanna_r_Page_092.QC.jpg
0de432ea35dcf0cbbdb2a7ff3c1699f6
e0394299df34aa5c640f843a02cd92965f5c3eae
78787 F20101115_AAAXWM khanna_r_Page_111.jpg
19c42bfea9462bba2786867e0f69f54e
ea6c9c2c47d7581b38017a29dd6189416bda0613
19678 F20101115_AAAXVY khanna_r_Page_094.jpg
4d1b3c9c1c86fd1370d88eecd62125f4
126566955c8e3e07f6828795fcaaf47a0c11951f
476192 F20101115_AAAYCH khanna_r_Page_121.jp2
232deebea6dad04d98474a0813916a44
7bcf649ca12f4e2bd415e610d496157094969cca
30749 F20101115_AAAXXB khanna_r_Page_127.jpg
890613a3de3225dfd7e01a755a28ffc7
48427a27f6981e076e17ca66730ac672efc50bf5
36780 F20101115_AAAYBS khanna_r_Page_104.jp2
728d748a6be6e5a59349ed397a514275
e1a2062f2e8d2c44aebd07fb78906b43ec7259d6
23873 F20101115_AAAZFK khanna_r_Page_113.QC.jpg
05d26fef9612ac46bce12773c1f61ee6
b75e94f13291ef7a9800404e65078cb66d9d2dbf
10671 F20101115_AAAZEV khanna_r_Page_067.QC.jpg
1997772ca94802444b32a80f9a68aae7
a9ff5cfe9d2d93caf6076fbf8939d584424db463
25562 F20101115_AAAYZP khanna_r_Page_020.QC.jpg
f30d5e9c030458a685d5c9b1edbc2ac7
8c51bd421776dc8e254000f23038aead13bd070b
74720 F20101115_AAAXWN khanna_r_Page_113.jpg
165d4998080a90838b885ad1a493ce6a
d3f9a96830f8f06e08d7440174390b3508b94308
39706 F20101115_AAAXVZ khanna_r_Page_095.jpg
a1f8252c0a6c9d80e1873aaf8bf84309
c8d02b7be3001c695b84fcecd7708c25a913d864
F20101115_AAAYCI khanna_r_Page_122.jp2
044dc0e7a76cfcb0de0ccd884fbcc0fa
1521c80e88df1fdb19e3ebf89c7322becdc78b78
32017 F20101115_AAAXXC khanna_r_Page_128.jpg
2c48a6708c63e3bcaf3f4016501dcc63
d52ecc3f108efd8cdec2be15c8fa7117cd83651d
33962 F20101115_AAAYBT khanna_r_Page_105.jp2
7ffb5340ae5cdd8322ae2622121acf0a
03507c085d0137466531bed0befeac6f1eccb8e4
3380 F20101115_AAAZFL khanna_r_Page_117thm.jpg
5f8b5dca69316b24ddc54b268ff42014
04f4600da7c9cbd50eda06844b5266afc4cb8ed7
24849 F20101115_AAAZEW khanna_r_Page_075.QC.jpg
d680db88509bf6b2507d73d8a3ec638f
d053871b42e9956aefdde6710acb31f33364b898
24865 F20101115_AAAYZQ khanna_r_Page_010.QC.jpg
f5393273ef8f6340d54c988751ab918c
ed740d262ee9668eeadc1726f799f998966c80f5
71383 F20101115_AAAXWO khanna_r_Page_114.jpg
7a4da7f7012c8ee5d11926dda27d0ffb
f5c4dec45438ec761ea12a3fc65b6fabeebc6d16
461318 F20101115_AAAYCJ khanna_r_Page_123.jp2
3375443dbaf4ae457094a578876b4071
2d21c0ac844d10423260e4a8fa71761b2215589b
29377 F20101115_AAAXXD khanna_r_Page_129.jpg
a7c29ca77260dd22a977a83700663796
90ba9f9aab5eceeea7d29dfbc36dbd034fd21ebc
116443 F20101115_AAAYBU khanna_r_Page_106.jp2
baefa338458bbd12d3257739ba1e0a54
abf48969712f23eecdea3d5a1fe4fa3b176ce2cf
9566 F20101115_AAAZFM khanna_r_Page_125.QC.jpg
3e85d90889652bfd2482fa729cad3889
ec26ab6ba0fd4a7c33b0bf3d740c4bbcc34de391
24161 F20101115_AAAZEX khanna_r_Page_076.QC.jpg
5b7d16e89cc33429790720a70e558b3a
b4a0d7d7e729a633cfd13b829da6b0a2d4cb8d19
6655 F20101115_AAAYZR khanna_r_Page_136thm.jpg
60f98c0057eee993103c2e0def1501ac
aae0d31d13e0a48d446d58ac6f5fd452452810d2
72763 F20101115_AAAXWP khanna_r_Page_115.jpg
510f12903a1dfac456f6f22a7a607a3f
75ceb38acbf9d5383217b62884a8447d2c0a3e96
46535 F20101115_AAAYCK khanna_r_Page_124.jp2
d34255ad45fa2fda5566e32b8995ba42
f7ec0e1dfd7725199ece5042abfb5e9028f7159a
32409 F20101115_AAAXXE khanna_r_Page_130.jpg
0f8d49cabd568385ea58210cb7c1999a
4d1ca5877247a396aae3896e2964afb81cd015b5
114134 F20101115_AAAYBV khanna_r_Page_107.jp2
fd854f64cde442eb94ded719c310b568
dc0f59e69d75362c71412eb20beb40ecf694ad2c
F20101115_AAAZFN khanna_r_Page_141thm.jpg
7ec4a8d9e762cf7c128f3fa2777a03a6
c206d1937a541820ce94631cf326dc95e399c50f
6801 F20101115_AAAZEY khanna_r_Page_076thm.jpg
5e288a83f623e29944d1291af278bffb
aad73b262a149bafa0e0090b542f99b696ebaf63
25876 F20101115_AAAYZS khanna_r_Page_055.QC.jpg
c3b0b2a757972745a8cd229f6bba46bc
c9a1a2081197730153b98f4ccee7d2341567d11b
46831 F20101115_AAAXWQ khanna_r_Page_116.jpg
6af3eee5efc463d393d406d127c07f5e
9432572c3294fb209b6a92f8dfd50550ccc991e9
30501 F20101115_AAAYCL khanna_r_Page_125.jp2
7cbc6772006360de17ca9ef5563564bf
f34f853a53f25a1c89f71bb4d343699ebfb94a2f
35784 F20101115_AAAXXF khanna_r_Page_131.jpg
10539cd1b6c4986a0a847c4df437ded0
d9503d7c6e8b94ff81019f537cd4289658beedbb
110941 F20101115_AAAYBW khanna_r_Page_108.jp2
5ecb5623ccd1ec8530c15a52c61a28b0
6d4b7ee7a2794e3b451b66ce50dd9507ee925322
11005 F20101115_AAAZFO khanna_r_Page_142.QC.jpg
1c3498ae98c4c88b0aef078767cb6687
77a3250d387a977204064c3b741e89f505a1eabe
23244 F20101115_AAAZEZ khanna_r_Page_077.QC.jpg
ccb4cb8c1a7c4e52719fcb0c64ab266c
acb26c1e0be2dac74532642739fc4746d664d368
7003 F20101115_AAAYZT khanna_r_Page_172thm.jpg
45f7330bb2fedba41e3bab85dff69de7
c8a6dfa0feefc97bf7a0ee58040116620206336b
30049 F20101115_AAAXWR khanna_r_Page_117.jpg
03fb2f22c61bcafeadbcece0c1e40b7c
def122017ae224281bf571b6ff6e8b32139adb86
392079 F20101115_AAAYDA khanna_r_Page_142.jp2
a746c16a67cb6cbf01cd23f523c3d484
e98a2e49cd0471b5781b187bcf73a10bd658dda6
35343 F20101115_AAAYCM khanna_r_Page_126.jp2
db0ab8bdce0fbb8d45dc15000e2f623c
86936b746439182b66ca9dfdd25e79f0ce6b458a
29892 F20101115_AAAXXG khanna_r_Page_132.jpg
5959bcc27d7b5bd67847d8964a305240
640065e83dfa68a675958045234c6157b46cff25
116432 F20101115_AAAYBX khanna_r_Page_109.jp2
303d9501d7e4817f64e216262956de34
547eb39939402af4048cdb6b1b53edb39b549fe9
25127 F20101115_AAAZFP khanna_r_Page_163.QC.jpg
549f96687db2676a122660d7d861a891
3b80e53b88a668ccdec9531d2af7ef154fa5c88e
18384 F20101115_AAAYZU khanna_r_Page_159.QC.jpg
d61568ecd680f223f6f25a94d801b384
ec0545d8e79be78653612322ee2214b33007a037
32617 F20101115_AAAXWS khanna_r_Page_118.jpg
510dfa3dc52b7bef63de802413772ebd
56feb19f2a71a1e37e87efc06e2a07cba4ce69c6
559508 F20101115_AAAYDB khanna_r_Page_143.jp2
3508ddab62567b51331134436338cdd9
6f51f4d96ff7fcc799440a1908e3f6e7f9d8b5aa
33577 F20101115_AAAYCN khanna_r_Page_127.jp2
ee08963b4ceab8cda429a18683f59262
769e8d71d614227f34963e3402e372a96ca07230
36931 F20101115_AAAXXH khanna_r_Page_133.jpg
b256b7e4196cb0320f1326e3a22fc30f
67604752178affcec0758985abf4e26fd8810c88
116025 F20101115_AAAYBY khanna_r_Page_110.jp2
f2aa46874ba222da7f9c743c7ffe2b32
759f3796f56c048d3c3f79a0f662908205ddc30b
4002 F20101115_AAAZFQ khanna_r_Page_166.QC.jpg
64ffc29ecb7bf8675a0ae972bf9b7302
29bef5510486849bb8719f3bb8ba3ccdd4387489
4721 F20101115_AAAYZV khanna_r_Page_048thm.jpg
7880cfc993bdb13b07f7555cb9f8ba01
ebc275a4b51b58146355d726b940b957cf6a04a7
34803 F20101115_AAAXWT khanna_r_Page_119.jpg
211f5968f7d76c6e7ff8be38304737a3
e204662bd024525847ce39de823b95a317e4b4e4
713869 F20101115_AAAYDC khanna_r_Page_144.jp2
76b4175881ec4bd4762b08f9a661252b
b4c8b9bc63266819525f26e2b23584ee4f3b5f90
31418 F20101115_AAAYCO khanna_r_Page_129.jp2
98715407cd114134aa1b6a738460f0c1
c949f9c62f7a4674a6805ec670a397d4bd693c0b
119153 F20101115_AAAYBZ khanna_r_Page_111.jp2
8f025a5d7dd8c2e56dfa6f3222641fff
394e607713bc39b7591644f454377bff750331d0
24767 F20101115_AAAZFR khanna_r_Page_172.QC.jpg
d18a813fc02eb9d5aa90ca65d1e96590
5723278e946f2a2b3a6cdab56f8fb5724666a697
10854 F20101115_AAAYZW khanna_r_Page_096.QC.jpg
293b026bebf79aa1ae77bc06b03cc00d
32937a8b83098878a1406ca12d662418683b6324
1051964 F20101115_AAAYDD khanna_r_Page_145.jp2
3e7af1dbe69d5754b10534dc6e158d9f
b45c99f10cd7a7c32e31f768c5844bdba6c0b15d
34255 F20101115_AAAYCP khanna_r_Page_130.jp2
849226e58fbef90b26ca089dea0a6442
9171c72b4d2d0cb8b0e4bedef39c15a796692120
30058 F20101115_AAAXXI khanna_r_Page_134.jpg
62238c3320f3c053556a0d9678ec9814
379d49354d4c27a7d842eed17637a79ec2b12940
37136 F20101115_AAAXWU khanna_r_Page_120.jpg
a7f966a3f9ee4f9c4e1a0e58fbe0ee62
753b1f695a6d2f124a0d0d4987ff3260d9e53c86
7204 F20101115_AAAYZX khanna_r_Page_135thm.jpg
6631f917e7767374382f511284a346dc
b889ada7f88ecd808c991ad432e9c529a42de0dd
479421 F20101115_AAAYDE khanna_r_Page_146.jp2
4bd4980ac3e55344687fa6db8d47e7b7
0a1c6fa046f85c3db4217dfe54579631f1109b10
46492 F20101115_AAAYCQ khanna_r_Page_131.jp2
ca31c7466c038b5e80e6b9f159586330
25c004945c052b450a2bda73e431614915f5109d
80816 F20101115_AAAXXJ khanna_r_Page_135.jpg
535be483d09ceb14ee5aae9a04dae5ab
6ae184bf688d748c02a79510424678347f198100
35204 F20101115_AAAXWV khanna_r_Page_121.jpg
4b8ba7b10d1b3f5cb10f404eab8d09d0
3b7ff6b28270f29b12873ae51aa00c138fabcd21
3365 F20101115_AAAYZY khanna_r_Page_134thm.jpg
a2e9cf9172ce8fb8dec0baa79288ea22
815ea7c92a60a9ffb997389372e5b865883037ae
436522 F20101115_AAAYDF khanna_r_Page_147.jp2
469651395a47af304a62ea60a8d78dab
a949f58b06d05c473c785c04cf35f5eb6e096f96
42160 F20101115_AAAYCR khanna_r_Page_132.jp2
1365604025b68d114b2b226c417046c2
2a3961ff7cc1b19e265f57e330d86798011414f6
73113 F20101115_AAAXXK khanna_r_Page_136.jpg
4364f988976565dde7c6e86647088c16
3e3b865686da0b01b099bad811414b77abc150e3
30613 F20101115_AAAXWW khanna_r_Page_122.jpg
ca08f491215879084f490fff07d51a62
97e135978b118ec02912c49c1807b9419778f86b
13677 F20101115_AAAYZZ khanna_r_Page_048.QC.jpg
6b634f2136f1073944c6c5b34ea53168
fc0f7038996177cc07bb3a089075d648877a13c4
116285 F20101115_AAAYDG khanna_r_Page_148.jp2
f1f5a6019d50ab3b1855f2a2edf4ccdd
e0073c36d333ad61c7b94d64d189b2712f61337f
43910 F20101115_AAAXYA khanna_r_Page_158.jpg
807e91f3b6ecc093572b6d84b8398f13
594bda7b9e497f157e30d3ee4f233fa543a605fa
74433 F20101115_AAAXXL khanna_r_Page_137.jpg
81bc832b2249b900f0b696976bf1a3eb
497e15ea490a0ed2b80509c5cc446576c4ffa7b5
36962 F20101115_AAAXWX khanna_r_Page_123.jpg
45b48ba48cf72850965b102cdffbf9cc
235bf42e371b4ded452ff99b24e70a615f2911b9
113395 F20101115_AAAYDH khanna_r_Page_149.jp2
b67704e6ca39057bda744cbccdf8b751
5b7c436f5efbd966597bfba5d6cb3cc75db18127
30389 F20101115_AAAXYB khanna_r_Page_160.jpg
9175758f8f5e381757ca4dd4e09d512b
3d89cbdab93437a93ad32de87084108a8a9f43f5
560720 F20101115_AAAYCS khanna_r_Page_133.jp2
70cec0a3b202175906a11333d85148d9
8d52de15fb86d6a530416f52cb42d31bf08bee56
37933 F20101115_AAAXXM khanna_r_Page_139.jpg
113781489dc68ba898b693d65aeed34d
a999e783a28492617d3fc3c2ed51bb43a20bc053
33921 F20101115_AAAXWY khanna_r_Page_124.jpg
9f82ee419bfe0996643d41a99434e5d2
27dd85448a882dcf93544331d077c9aa169a8a28
108262 F20101115_AAAYDI khanna_r_Page_150.jp2
c19ba785f4956de16949bcd62fca73b4
134edfc43f112c51fcdf5e76e6ba140613a4d6f5
29044 F20101115_AAAXYC khanna_r_Page_161.jpg
496b228b6b0b484cfa9456ad1148bd80
95c3234c21d23488db2652e7a1f316f301900d05
30597 F20101115_AAAYCT khanna_r_Page_134.jp2
54d98ae4ca6fb5d3b8335e7460e1fcef
a6c23412913a79fc384572e408e754b7fe4bff74
28254 F20101115_AAAXXN khanna_r_Page_140.jpg
2471b137d1dc3141a2acc1ba4b0b5af1
82c9bf8f16f4c40c0177629bbd4ba1804cda9d0a
29749 F20101115_AAAXWZ khanna_r_Page_125.jpg
3da242b27a465bf81ac1b44b83774a02
d4e8c2ca8065b5418d7324d281c02dee17511de7
120533 F20101115_AAAYDJ khanna_r_Page_152.jp2
baa80ff389ee677a42fd1a86ff80e852
bf4217bc186733555114902981daf2ce474a9a34
78936 F20101115_AAAXYD khanna_r_Page_165.jpg
1bb6f6132cd6e9b45163b5a25e776e42
1945972ffddb1fa5ca5b1dc3e1c2eaf58bfe3268
109241 F20101115_AAAYCU khanna_r_Page_136.jp2
5a2abdbad08ebba31de0caf0985e4a0a
19a7e6d3e562e94a15cbb0f8c17ea3e1a5464c8a
34186 F20101115_AAAXXO khanna_r_Page_141.jpg
b45233b11e31421e2707609c80fb37ae
4be1a0fd8fc7ebf5d21742379f3aeed2cb57f42d
88491 F20101115_AAAYDK khanna_r_Page_153.jp2
7c25101458306c8024335f3784e054be
7a9e2764a8f5ff2eb36fb4a89e079b677447d2d0
12051 F20101115_AAAXYE khanna_r_Page_166.jpg
01511521337f42a11d798ed64cd69065
5d32340dfa02d7c04bf5cef73acce1b26b566519
113506 F20101115_AAAYCV khanna_r_Page_137.jp2
8009e9abd298dfcf0ecc76e76f7aa01c
04ac1317bc1965841b63929bbedf2f64e04f3a7f
33548 F20101115_AAAXXP khanna_r_Page_142.jpg
d82dd6c66f153ade5859d13d68902be9
a2cbcaade66f3ad6dd38b3095ca045122aeb5432
21475 F20101115_AAAYDL khanna_r_Page_154.jp2
2a4a8acaac00774b73c321a34d691b61
ca19697a97188e6a7cfe68f2f83ee48efbee851a
84929 F20101115_AAAXYF khanna_r_Page_167.jpg
7487b8281b8f111404e63c3b0a70e88e
d773118740068d92e67209f23e0f8826e972b961
115860 F20101115_AAAYCW khanna_r_Page_138.jp2
78aac9355fc373b270081e8ff4c0921c
ff83afd4bf41ae321d0ee8d75c2bfc1a047f1324
46952 F20101115_AAAXXQ khanna_r_Page_144.jpg
b3c294c4a2571c5ef720cc305f5f641c
275d04c7710ea74ac21539a69cb5f6565c533145
22157 F20101115_AAAYDM khanna_r_Page_155.jp2
357ca335a3004e92b1e7758a597bae43
e75610335444703798daa3217bbd442ee889fc2f
76394 F20101115_AAAXYG khanna_r_Page_168.jpg
94f3057df446cf11d037867067f68a0c
4db8f2ece42a2155656c6c9336ae5706cbc4737d
53554 F20101115_AAAYCX khanna_r_Page_139.jp2
f3e3e42653ba4d2f7ec3fd360d3306c0
64a060cff2ecec5a9dae6e13f841f86d0ac1c891
61360 F20101115_AAAXXR khanna_r_Page_145.jpg
081cf176197eed1716460fcd6b28e11c
7b613e04e39cdcad3606e0480e8b7da7188fbd60
109275 F20101115_AAAYEA khanna_r_Page_172.jp2
d97b3d332d39d7b55d14f1787a0591ce
10f32bdaad7a85fa576065fca72e3d201e98e0d2
62538 F20101115_AAAYDN khanna_r_Page_156.jp2
239e1922976422666f80c065c6d9715d
25255d570399267cb6fcd1e1113c57aeb00aa827
73153 F20101115_AAAXYH khanna_r_Page_169.jpg
570bcd46c88a2d1bb6dc67604a7ccc08
ad190f309c24f3b51d3e437969b05eb64c9ea741
340631 F20101115_AAAYCY khanna_r_Page_140.jp2
498a7558a14113e864f59b8ee8bdca62
c781037e493b68f5c1cdfdf0d657971ef8f086cf
35714 F20101115_AAAXXS khanna_r_Page_147.jpg
3f6d97564b1d5ec08caa08a8730b8fa1
3d2eadc9088b4980347940701dc979b39145fe9e
105212 F20101115_AAAYEB khanna_r_Page_173.jp2
171914cedfebac389dd741f8bdadb0e3
c80fa85949df8f009c6936b896cf192afa202514
29772 F20101115_AAAYDO khanna_r_Page_157.jp2
25356a75830b124dcdb720246fb728bb
0227592bfa61c1e09c302d239bde715d80a38a07
79364 F20101115_AAAXYI khanna_r_Page_170.jpg
b8413e0e959ea8c9fc56ad9b0f14b358
7ffbc81a0be25c4193810b6d1139302cd8242b3e
489265 F20101115_AAAYCZ khanna_r_Page_141.jp2
6dbfa03b9598e34efb45a14efdbf8f73
b06f20de94266360f41ccd37dd8be25110ec5e28
71370 F20101115_AAAXXT khanna_r_Page_150.jpg
4097bb013f1f23d9a1b56a36bd37279a
38fff6642f344933a4fb19f6075a36fac7a00c50
63421 F20101115_AAAYEC khanna_r_Page_175.jp2
66ae2f8fe2f2e095de46d89e9475de46
2df1be65d8a65dfe88a5525332eab212e17a7b6f
849796 F20101115_AAAYDP khanna_r_Page_158.jp2
bb9b9d883dca166a7f7ff6a68d9d87b5
6f96edd6a70f37bb4513f558c61ac69f0b075586
67041 F20101115_AAAXXU khanna_r_Page_151.jpg
c456dad831574b80fc71ccde27976312
38b95a40f0ba9c1681c91bb58f3e150df267a5c1
F20101115_AAAYED khanna_r_Page_001.tif
cd14c2942b7605974f3148d4724dd4f4
dd7e5eca01377cc0e6df8aa9905a25ed7575d370
73734 F20101115_AAAYDQ khanna_r_Page_159.jp2
128c68c15a8094f55bfd684f1510a1fc
0306d69ae5b92a8f34af1a3c0c5be1e535f6ce40
69295 F20101115_AAAXYJ khanna_r_Page_171.jpg
86d4571f38149ee3f8476ba5136d2f4a
19ff817f43a793e965bf5decd76dd8318883b553
80091 F20101115_AAAXXV khanna_r_Page_152.jpg
080c43a49ff6424b19bcd65549472964
177b2b6be1efd7c4e150d781e314986fdb892678
F20101115_AAAYEE khanna_r_Page_002.tif
aa137d0c76fe71d186ab60e9be634613
2e55000f4da9d9461d82bf4886260655c9d77b90
330742 F20101115_AAAYDR khanna_r_Page_160.jp2
6569f1e8d074a9ae9a224bf2ff400808
531cda845ec68841afa7f417ff5c223e4a0f9011
76344 F20101115_AAAXYK khanna_r_Page_172.jpg
b26ec18fba812c4957181b037606172b
65b451dd64cdbda186b13794070d589258ac48b2
59266 F20101115_AAAXXW khanna_r_Page_153.jpg
f860af90f7ef8c63af2bb399051b189c
e481d60b7e8f0d137d24aadac97a0c6cef14adef
F20101115_AAAYEF khanna_r_Page_003.tif
cd169c269aee339ad74a43ece1ddda98
2a0e91590d1096eea7325b4411dce63f05b51e3c
17980 F20101115_AAAXYL khanna_r_Page_174.jpg
3549730f85dde26c27e351bb7e9a0c25
f38d02a250c4e78b9e9204189fd226e6ad6d5ddc
23196 F20101115_AAAXXX khanna_r_Page_154.jpg
725360d52832af79fc8751c394855e43
8762ced97d4fa8cd0adb771701dc8f9edeb28b86
F20101115_AAAYEG khanna_r_Page_005.tif
4047d015eb839074fb54cdb1a976261a
dc5b942c032fc00a0402a7589d08aee1961e1aba
120631 F20101115_AAAXZA khanna_r_Page_018.jp2
749be5b4fe49c345340f5cafe73eec4c
53d73706ecb4569e741b2cb7a3d6087b143c3985
31166 F20101115_AAAYDS khanna_r_Page_161.jp2
8e7c48b1cb5169c022cdbebb8b3a42ad
92c10a7fc19382338c2fa28b6e8f94e897af315f
24180 F20101115_AAAXYM khanna_r_Page_001.jp2
941d12f05f440e2085733d5a3a528f57
ccf2bd9d8c3afa525acd643d87747e201e7a2f6b
50478 F20101115_AAAXXY khanna_r_Page_156.jpg
2b530a21117ab2e321cc4ccc3b472701
4cf93d984c0a00a52ed6129ce9ba99105f8bff55
F20101115_AAAYEH khanna_r_Page_006.tif
30432d028c62379cd60d09c7fd52587e
6b0322417b8d17c872bb80607a103e3cc614c5f1
120903 F20101115_AAAXZB khanna_r_Page_019.jp2
a83aba3fdb45e95d4d76976b31c125e5
bea1b0230cae09429737137cd6e8d583e7c0a441
4899 F20101115_AAAXYN khanna_r_Page_002.jp2
58844abd3e7e892c768472b4de3ecc8d
5c5fbdb6730d8ee56f4a696c2a693a9b9d753d90
27070 F20101115_AAAXXZ khanna_r_Page_157.jpg
efc93221ff53c525245f6067b47151d2
f9eccbb6f2672b7034ff39f9448baaf38ab16b52
F20101115_AAAYEI khanna_r_Page_008.tif
c4ade6601fba66bc28d5ac0bbe230f0d
37cda34efec4d5520052a181dad415d7821ae04a
116768 F20101115_AAAXZC khanna_r_Page_020.jp2
c9706f064e12a2560d31d05161d2380b
09e8d053e3c5ee3c7ba6ce5c970b6fa81e20f36d
113323 F20101115_AAAYDT khanna_r_Page_162.jp2
ea407fe4cdde6a57180235b740c8cfdc
9b983e4689c82725d1feae54e1ec4a785ee58dd5
4342 F20101115_AAAXYO khanna_r_Page_003.jp2
f993f44e030702fc93c5e87e0770bb30
498137a9588d94cd0aec89572f4ecf27dacc1a91
F20101115_AAAYEJ khanna_r_Page_009.tif
92c0f54dd16d1a4f8012264177cc9d4c
a94ecac75bd749bf818835540fddf6b192cdf01a
117752 F20101115_AAAXZD khanna_r_Page_021.jp2
39386da2d82e4882ba2267dcb5f5be04
d514830d9861bfde390ca02c32e572a9e1448c8f
114901 F20101115_AAAYDU khanna_r_Page_163.jp2
6e0c923e156391fac79b79ec9b58d13d
931788d4fcf9755058d875f8c46ecbfd29f0365f
116245 F20101115_AAAXYP khanna_r_Page_004.jp2
c1a896ea4dc338c92d337790868406a5
7e112641b0df1dab1d9eac1c130beeaa108b7749
F20101115_AAAYEK khanna_r_Page_010.tif
9805984b567bef01fb21aa52077f9daf
3ec2f24d745e4731c3f448fe793814cfe5363436
26011 F20101115_AAAXZE khanna_r_Page_022.jp2
8bf7476e987bec35a82c1d27d0456994
a76e17a0c1884a2a8dda6931c5f3997b5806ba0f
122308 F20101115_AAAYDV khanna_r_Page_164.jp2
39c341ee302e7bab0cd9d969b1788e38
eec8ca6986f5e82cdb94cbdf0ca1f8e6ef6599bd
24882 F20101115_AAAXYQ khanna_r_Page_005.jp2
916283c1d2425b26949ea7a905fdfdc9
57b6b949bf300516cab2f6533291766271e2485b
F20101115_AAAYEL khanna_r_Page_011.tif
7cd0cd57dfe4ae00fde0602742771d03
86fc89fab57459e74782ff5a78a04f506c87f98b
100020 F20101115_AAAXZF khanna_r_Page_023.jp2
603135b9391db08b26ef18c054791c03
0fce0939a72bade7e05d703f2b788fe94cd803d9
117170 F20101115_AAAYDW khanna_r_Page_165.jp2
60d5c5ddd6640a661545c52aaba8f915
ec1c3a0847aa9ff4e245e664da7efd88f8a3000e
1051975 F20101115_AAAXYR khanna_r_Page_007.jp2
4d176d8cf625a51aac0b7950f357873f
bd9f0f622c8cb8aa3887fdc8108eb623ec342e05
F20101115_AAAYFA khanna_r_Page_030.tif
f8a124887e3fcf60aff97997ab361b27
73d04769bba42d5c16ca780a429e75a3c2b1fc05
F20101115_AAAYEM khanna_r_Page_013.tif
2b9fe55fc486ff9112c57e7bfbe41779
2252cb54375817ede103259af9f06dd7665bc1bb
116373 F20101115_AAAXZG khanna_r_Page_024.jp2
ce7b23f6a3e63f399a34e1451dcf6a56
107a8ce96ac46e241703c2182abb990cf7aab294
9917 F20101115_AAAYDX khanna_r_Page_166.jp2
11fde605439237462be8f7f607562adf
362a32ac98c5fabd1531ce187474657d03d57b13
90915 F20101115_AAAXYS khanna_r_Page_008.jp2
1ebdfb840ad43a46bc13cddbcd5e023c
1a575e00049aa465e5ea86ee77009329862c83b1
F20101115_AAAYFB khanna_r_Page_031.tif
74986aad79133eda85a24304c9f5bba0
d42e2d7397213f661f2b870a74fe8048861bef4a
F20101115_AAAYEN khanna_r_Page_015.tif
1496f7c32555dda0c4c9cf15b1ffc558
2537606b509077c057b0641e91cbbc0ae087b8f4
122057 F20101115_AAAXZH khanna_r_Page_025.jp2
44bf5edbcd9ef7c77365562bbc85336b
10cd73391f51ca843af22a0c7381d818a2be163a
109800 F20101115_AAAYDY khanna_r_Page_168.jp2
828478543a5e5dbcc03048d8365584ef
8d6e7aa9a5e191ba9203c9f43d1e52f06fad0704
1051953 F20101115_AAAXYT khanna_r_Page_009.jp2
e544ede00b8495e23257a258823fe6c4
f176a3473830786453c3afa0ad73761b1732b99e
F20101115_AAAYFC khanna_r_Page_033.tif
d6119e4f6ff2fd8efefa6ee1fc24a592
77e1eb715768dbd030c50fa69a03d12310e3f34f
F20101115_AAAYEO khanna_r_Page_016.tif
413ad8d1580e20090dbe9dbc41d2e06e
192fc4119c2d11cf7deb8a81dcc5d7d3976cb82f
113476 F20101115_AAAXZI khanna_r_Page_026.jp2
4805316bd3c1996f3053598de35d6c2c
0a119efed26c4d10145d1e7aef52716c7cce52bd
111208 F20101115_AAAYDZ khanna_r_Page_170.jp2
6f8452d9b34cd2e0195a845e1af40be3
194dea4d6f842e0ebe21ac554f42fcdbd28478ef
1051949 F20101115_AAAXYU khanna_r_Page_010.jp2
cbb8f7b14fe3b9444c5524a5b5aae292
0ad52a572bd84097b8a094ba33575fb3d6070678
F20101115_AAAYFD khanna_r_Page_034.tif
6cd2b28f3766f1543d5ed23bec85e832
6341cf4460acad8366af832ad3e1c9fef2528bc6
F20101115_AAAYEP khanna_r_Page_017.tif
7ce79b193fa6d8536d11e1a9538af747
e1b8f7f7776a09d0ca459ac4b6143dded0798469
106100 F20101115_AAAXZJ khanna_r_Page_027.jp2
15b2c6d5947778d1a21bfb962091123b
0440815d7ac9ee8bc4d1a334099dc67d77be6d27
1051979 F20101115_AAAXYV khanna_r_Page_011.jp2
604b290e7f7b446664ca09777256c68c
26773281440eae48901f54ce560d18c138b07222
F20101115_AAAYFE khanna_r_Page_035.tif
f1fdc935681a1b422faef4afba8ad2ab
f9f112c06ccc6a54215bdace3031f77504a6227e
F20101115_AAAYEQ khanna_r_Page_018.tif
02ceddd1aa28aa465250426448428095
0066f66484dbd0a22a2dd7342ed17e80985892e5
1051963 F20101115_AAAXYW khanna_r_Page_012.jp2
e28fc97da9f53eae911bf23076d2c51f
8073e622c10c224bf8661fb44bbd94516e1d28f2
F20101115_AAAYFF khanna_r_Page_036.tif
919e19f011e9f67aa25f88e1929b96aa
5ffbdd0b09c2fb565f0dd9849d29f0b561df8104
F20101115_AAAYER khanna_r_Page_019.tif
7e6aac841e9b041d07d56a8f7fe8331a
e594b61a3e7ba11d72f892f04d36b918af6d2fa9
98317 F20101115_AAAXZK khanna_r_Page_028.jp2
69fc8bb87360ed02367400e29fe3c0dc
4b17a094554f374c84fd35bab55e7e729536dc49
1051928 F20101115_AAAXYX khanna_r_Page_013.jp2
d8b96d3ade42f42b35dd2eced17652cb
d207a11fe459be3ca52378d72a9b773befbbeedf
F20101115_AAAYFG khanna_r_Page_037.tif
53a69de93cb123c069e3cf1824b0cd28
f84efd9ce8131203b1ce68952f8d76602ac3e30c
F20101115_AAAYES khanna_r_Page_020.tif
572301ff988828dd45f309f815fdbed9
e68c654109b96b0243f8cd8328f578ff4a6d598e
105081 F20101115_AAAXZL khanna_r_Page_029.jp2
fddd2487012410b71a493c3b4e679d7e
84beb3c4b9b2ea7aa84251f605b6a52fa4f3e00b
F20101115_AAAXYY khanna_r_Page_014.jp2
3ed703717b86cf652650d46a6097f851
4ffd61946d206b578fc05b4add28e9e9d7b184f0
F20101115_AAAYFH khanna_r_Page_038.tif
2fae60c541224a8b68d2b4c846e2e4da
84daf9fc0131f099067fb1503bab1ca813cc21d4
F20101115_AAAYET khanna_r_Page_021.tif
161085fc18de48ea014250ff637c4c1a
28cd64a45f6d7458835318a6ae0309743da110e0
94604 F20101115_AAAXZM khanna_r_Page_030.jp2
2da6792dbe25197fe7d06c27e821aa07
1948bd844c942e057e6e7c13f72c667ae6c21ce0
55877 F20101115_AAAXYZ khanna_r_Page_017.jp2
04209d3bc60778cbddd92b6668ee68f7
26bed0c66b296c7bca9e3ac51babccccc9630f98
F20101115_AAAYFI khanna_r_Page_039.tif
c7ed7bae2e2bf9b2d047b704464a9292
6b2414e9916e9286cf0a25d200a8c054d04b2282
114894 F20101115_AAAXZN khanna_r_Page_031.jp2
387522dd4e4380940f2c0e87e88394f5
d4656fd047519f70e68a0a6fd54520560ec5fdcb
F20101115_AAAYFJ khanna_r_Page_040.tif
efaf4de209335d5aa790f71925b2c7e7
05c4f5d02992c53ce9f0efbb624bde37482088c2
F20101115_AAAYEU khanna_r_Page_022.tif
72060980bb756600cdf81a9fc7e11cd2
144153d5ee4f4197f16bc1a14a5cf3fbdd61af4a
89848 F20101115_AAAXZO khanna_r_Page_032.jp2
956dc0985af7bbdf8c9abec6deeeed38
616c6838fec60b32a27a116da2dba5d8989f2890
F20101115_AAAYFK khanna_r_Page_041.tif
0c3f67ad5ef13f884a0376dd827ea113
ac60f5ca117d90a2687fa31eddd661a4ddbc83c1
F20101115_AAAYEV khanna_r_Page_024.tif
31533c5aac87a222581f708468a25c26
c8f5bd9434bea705e35e70293d1def39f1c148ce
118621 F20101115_AAAXZP khanna_r_Page_033.jp2
2eabba5ae1bee01678002824d837c0d1
c0453b8f7341d3e5b795e1b30e544ba7d223827f
F20101115_AAAYFL khanna_r_Page_042.tif
5172f117748946849cdd6222b64587a4
44a8af17958e04dbb5a08a92bbaa77294ebe5426
F20101115_AAAYEW khanna_r_Page_026.tif
2fabe73c05a7d36b5dfe6e704fc3c8aa
b8c93f52d3c05d5ece95e6e2e17d0795b229dbef
119378 F20101115_AAAXZQ khanna_r_Page_034.jp2
2302afdd8def38fba7c92d1abb267b47
72b37b8bb43dc999df9e3f19feae7cbc7d7f5350
F20101115_AAAYFM khanna_r_Page_043.tif
4a303d17bfbf4d6e2466fa371f49942e
65971a3cfc90f7e19b2b3b2b4dd7363c742be9bd
F20101115_AAAYEX khanna_r_Page_027.tif
579395aeb7d3f14375ddebb44113d656
69fd5ab0205e7f9236d511f7e3790ebaabbb4160
120814 F20101115_AAAXZR khanna_r_Page_035.jp2
a0fe3b84cf91cb2e737c32bf3f348d48
807c580abdf61d7f68d2c26b96e8f86a8387d219
F20101115_AAAYGA khanna_r_Page_060.tif
74bb2fdaca6aaeb87206c033903fe695
f2e25702ca384295234feb8ab90a56ea811a9945
F20101115_AAAYFN khanna_r_Page_044.tif
5b7a787a3e1e9175fca96f5bddc1cfbf
07af8a0ef08a97ce3f0ab99a836dfa972e0471cd
F20101115_AAAYEY khanna_r_Page_028.tif
80fe29cff779c4474107f029fbd0073c
7a778ed65c6a4f247b8008c070e660fd0382706a
118643 F20101115_AAAXZS khanna_r_Page_036.jp2
6a908f8bad607f00f459d600d6bf2185
87ef915c7daa62554bfb1dc2b1513eae043be2de
F20101115_AAAYGB khanna_r_Page_061.tif
69006a4e6cd61b4ff76720cd8ac2d605
b0b2458bfd3a863b2397d0e3796a84b827979f7e
F20101115_AAAYFO khanna_r_Page_045.tif
adc813a77992c9f14600d8d6bed5c60e
fb966ebd059d5f2715ca1a5f6d9aa97c074de614
F20101115_AAAYEZ khanna_r_Page_029.tif
838434812404ebcea2545d1360aad6d2
ef03d75c5cafca85d342e1da26e93b81d19a6de2
111981 F20101115_AAAXZT khanna_r_Page_037.jp2
4bdaa6a5be0cb6016453da394b3320af
d970bbeae15e08d58ab9258aa41cbd4b1e03d75c
F20101115_AAAYGC khanna_r_Page_062.tif
a59653c2b520e3d571618d2f24dd8654
584f7a7857308378af349d1459c0ff1f5c27180a
8423998 F20101115_AAAYFP khanna_r_Page_046.tif
c40312df8e8eb51ba018237e37a73a03
e06be1e9cedb37fd4924f5346aafd27161a2f9f4
109723 F20101115_AAAXZU khanna_r_Page_038.jp2
1e141008b7f3bfe463e88563c3dabc94
838c8dd967960ca4f99372ea811f2275b78d1408
F20101115_AAAYGD khanna_r_Page_064.tif
4c5c3a599bec42b6e7b64dd2c9f3bebf
6268e5942d3126fd6f8821f64c1ba21cfb8f71eb
F20101115_AAAYFQ khanna_r_Page_048.tif
aef4cb7e3414d72e24532d40e5a71acc
a1c769d0904fb39e138f50033ad74a512ecd28a1
112241 F20101115_AAAXZV khanna_r_Page_040.jp2
7179a40f4576cf1296e19f23eff3ae5c
0f3a058deb03b9326ed72e9bed48a957a34daf58
F20101115_AAAYGE khanna_r_Page_066.tif
b14a08e2bc42dad2d4ca849fef131760
9fc334c1ea6a6caa76a4f91955f6883b81544f0a
F20101115_AAAYFR khanna_r_Page_049.tif
1b462d3c4411cd4672299d635df66c50
fce0679fe5ecd9ab79fe3dfb5fd6d2c3f737538a
99808 F20101115_AAAXZW khanna_r_Page_041.jp2
1c30f12ac323c19e6b4f37562079332e
73d33261e1446822575422c74b7468e30398c98a
F20101115_AAAYGF khanna_r_Page_067.tif
ed90ed535581697b176b21b28adb1454
b156c358736d33b619aa95cfefe649b7a42be2e4
F20101115_AAAYFS khanna_r_Page_050.tif
954eeeaa13b9d4e876d1449eb36a1f35
f8f1860f9f81fd54365d595dfa760d0db0c6e587
494305 F20101115_AAAXZX khanna_r_Page_042.jp2
c61c647f3b38b07d4560303bd2616b2d
13aa6e73775e70c5ee89845c7c51345301b222db
F20101115_AAAYGG khanna_r_Page_068.tif
442a6f3ced0538d74a2c70e6880b1875
7362304eb7a5ef8c44953559404a761da0251b8b
F20101115_AAAYFT khanna_r_Page_051.tif
bdae351b8245bd5939a3576d6575bcc1
d098cfa14db46beb1c759dd708eb499f32a5a60a
44357 F20101115_AAAXZY khanna_r_Page_043.jp2
cd0b16acd73575a8e4897b184677e218
adadc2e24094943b3f090021eeea857613161da3
F20101115_AAAYGH khanna_r_Page_069.tif
ae38a0e34879ee7dbc671b1389f4141d
bd593f3b44911865f6d949880ee3e56ea2be2534
F20101115_AAAYFU khanna_r_Page_052.tif
a1fd5234f694d6a36fab94bc88a12a0d
45e81889c568ad71eea2d60404dd566a352b5802
637105 F20101115_AAAXZZ khanna_r_Page_044.jp2
f7afb7fb0301ef8ce42344504b3191d0
0b8d02ff67d9607df901ae617d580fe95f2f4228
F20101115_AAAYGI khanna_r_Page_070.tif
9a21f83df4359c49069e74a50317ba96
88a6569ed22b1e3628d95719ff7597bcf432a19c
F20101115_AAAYGJ khanna_r_Page_071.tif
a29f9041d0ac5365e803ea2e3084fc8a
f244cae1be818f88ae6b2a0cf4e92ceaa5470915
F20101115_AAAYFV khanna_r_Page_054.tif
201587691b51ba9eefc8a969bdfc5742
6c1e7ccbea9b409dae0311a1ebeae4eac61316c8
F20101115_AAAYGK khanna_r_Page_072.tif
586da9a46c62806dfca34d57f7bad512
a116b26cd87260a7efb7a81ad58fba69f4f388b5
F20101115_AAAYFW khanna_r_Page_056.tif
f54914dc20cac1e9927f949890b051cd
48e9207a81400e72db7af512fd929f3db55589cf
F20101115_AAAYGL khanna_r_Page_075.tif
4ad8a9193d9d6b0587b2c006ac8cd8f6
2eac6fd174cc44c73d9a89351dd3e40fe4501bac
F20101115_AAAYFX khanna_r_Page_057.tif
ccc2ae00a2b2dd340cba351544e2a414
4f9bef23a394ff1401208896f5a645d933dcdd76
F20101115_AAAYHA khanna_r_Page_098.tif
47b2f01e9104ccaf7d868a315f029536
882fa49925ba3694fca70345dca0abac54adb94e
F20101115_AAAYGM khanna_r_Page_078.tif
ef8360a3ee8a7b143cdc007155d79f46
bb46a5935bf38ea56d821c51c3ef9e563c4cda79
F20101115_AAAYFY khanna_r_Page_058.tif
53b282ad3b7378a3361a28d9de7c8317
70e476cdcb354ccee0087527869e30ffc519133f
F20101115_AAAYHB khanna_r_Page_099.tif
4399c503ebd10c263fdc9ae373909764
d7e62669ebc22534046ec31f33ab9962baea9230
F20101115_AAAYGN khanna_r_Page_079.tif
2035c02d4e135157df81e100c314be3d
6ecd84792f1c83f029d3c2966b3af50b8ad71be8
F20101115_AAAYFZ khanna_r_Page_059.tif
d997b6259a3f0f7496605afa71cbd3bd
6852cd9cdd6072a44d53f6d2a52a9491723af993
F20101115_AAAYHC khanna_r_Page_100.tif
79102fdb92a1e9995dafb0d6769c03d2
42f498c5d49618703eb3b169b17a7d5326339a16
F20101115_AAAYGO khanna_r_Page_081.tif
acd3eca62ab92f240652b97795af4fd2
5caebb4caafd032e3d7b70868d104825f634a245
F20101115_AAAYHD khanna_r_Page_101.tif
b752d7f67277fa564be83bd66fa23b5b
9628808f62ad9b811bb5d2cd36656489dba0ed93
F20101115_AAAYGP khanna_r_Page_082.tif
e270416f2f57468513b92a0384114de5
6ad57f19549ce43afb0721bd0779d7dca26c9805
F20101115_AAAYHE khanna_r_Page_102.tif
19f16c0f419f02080059146646ecf3eb
70458890e38580ee1f41e245155eec5c95cc6d45
F20101115_AAAYGQ khanna_r_Page_084.tif
226fb693e7e496109e1d95f90000b5cf
7b2e127431db6fb8f2ea5efcf7bc2e1da680618e
F20101115_AAAYHF khanna_r_Page_103.tif
24acd4c629cb7da0bee059d714c079a4
59c66d8a90b6612a4c76dc7ea46401a1e51259cb
F20101115_AAAYGR khanna_r_Page_086.tif
ded9866c3f8a2c2dd69f08bfbe1f5cfb
82a15f071a875ffb4b30c5a26e279ae73bc1ad67
F20101115_AAAYHG khanna_r_Page_104.tif
f393a9d87372b1064aa35f7dad97b672
6eaa33d8cc36144cafa8f41ebdf2066df45dbb51
F20101115_AAAYGS khanna_r_Page_087.tif
5e2a99ec7f5a365db5e4a9c280d863a4
ec5cc82cbf516a764573eabb193afae0b635b87b
F20101115_AAAYHH khanna_r_Page_105.tif
e530a70ab9148f37150be33591677ef0
c6180ce43953c3e53e8d7d9998add525ad0f0894
F20101115_AAAYGT khanna_r_Page_088.tif
fff3f5b0cb1d88a358590e026bbc9fe2
cced0936b5b5c973e09fa9601dc4250cb9aa07d1
F20101115_AAAYHI khanna_r_Page_106.tif
74276a1e58b2120cbf0e20ac54fa5389
9f4a5406e662fe8a1f405cb7caa1ccfad3f95c72
F20101115_AAAYGU khanna_r_Page_089.tif
3bfb41b233e57d24750778af8fe7ae78
121093e2cd2c9177c90ecb77eb2479536937f4e1
F20101115_AAAYHJ khanna_r_Page_107.tif
34a61a0f09f0e657ca6ebf814825b412
4ded3bd64ce282f3765f5d099fb3c54692db1509
F20101115_AAAYGV khanna_r_Page_090.tif
4a31af21575e58297f37a1b3fbf2fc22
f60bbc09c5eb21b7ab9f80d5df971126f1be3f79
F20101115_AAAYHK khanna_r_Page_109.tif
3547d186d015cee6b050c0109a45d19d
a9debfdfb2927fddd9a165230061d6491e018ebb
F20101115_AAAYHL khanna_r_Page_110.tif
3386cf7e74a657160562d1bc725e1306
a0b091854117e49e409853acda6372983911ca3b
F20101115_AAAYGW khanna_r_Page_091.tif
01f5f5b73c26a660ce2b1cd295fd21e1
e6e833fffe94092b8df74413f985ea8e8234e474
F20101115_AAAYIA khanna_r_Page_126.tif
c9025334ce0a89171cd63ea7a24e2500
e59286ae858841c6e7e4195945dca01cf9c3ac2d
F20101115_AAAYHM khanna_r_Page_111.tif
a77bc7634a17252818b9a5564c72d1f9
45c8837c1a72237677f8c03789b71437c066f5bd
F20101115_AAAYGX khanna_r_Page_092.tif
df2ffcc823ec1e923d96b5afa21fe040
714c276686c96bdce85cddfd87e4f59d43518df9
F20101115_AAAYIB khanna_r_Page_127.tif
bcd92e8a9aa1a40d176edf2bccae6a93
1dc8e41de266e88f55aaff040fce78b2dde99398
F20101115_AAAYHN khanna_r_Page_112.tif
137289f167a0eee42a5c885d49fa8f37
71de02cd3bef08486bf7739fa9ba8afc4406b4dc
F20101115_AAAYGY khanna_r_Page_093.tif
e5bdb554f1eb680eeaf8f2047b6aa8fd
f3b11fc548ecaeb33965e78fbf93aae22d35af0b
F20101115_AAAYIC khanna_r_Page_128.tif
956fca2ddf91b23dcd5e04c3c155b2e3
773fc451662debf91bd4ac800bbb2f0c3809dd27
F20101115_AAAYHO khanna_r_Page_114.tif
e40f18fc4db65a9fad5ce64142d1b7e2
c6e5b90222a90f40bc9ca583a405678cda6c830f
F20101115_AAAYGZ khanna_r_Page_094.tif
ccaf0f72710c4305fc84c452ba6f2c0b
dc051fd8f03369e13c587f6c1c8941bf485d6d58
F20101115_AAAYID khanna_r_Page_129.tif
6e4a4eae8c01b64c9ccccb60bd0fff33
35d7b5e83058aed63d73eeab05d148600b707df3
F20101115_AAAYHP khanna_r_Page_115.tif
f0811fff3581307ad1952c526cd0ed9c
6d58ed5f4a7e4b0efaa8202ed119486f6b0f3148
F20101115_AAAYIE khanna_r_Page_130.tif
c67bbbaf47dd74957bfc3fba4b883c8f
c9701c56798fd23bf2e839bdef042cbf996f3091
F20101115_AAAYHQ khanna_r_Page_116.tif
a7f5c668be3b89fbdb94dcb70fd6c70c
2dcd40e22f4d745f06d15ae35d279e521250e17f
F20101115_AAAYIF khanna_r_Page_131.tif
815fcb6c4bef748667e9bb8c85709152
65be9bf84cc88bde483b482360ef4e3faa859534
F20101115_AAAYHR khanna_r_Page_117.tif
044855f7ac9ea68e1533bcfc11adae0a
277b4073d97f8f8063b54f51fc9df842222d49bc
F20101115_AAAYIG khanna_r_Page_132.tif
48a7a7b38d78e3b65edca63fbd804919
6f4ccfee3e10bec07d70d8323b743b7fa6a71e12
F20101115_AAAYHS khanna_r_Page_118.tif
c246291c19e6d38339b2b1e8e21663ae
7a197881b9f151c109e5c34dcb2051e6c1c8101f
F20101115_AAAYIH khanna_r_Page_134.tif
0802887b06b2b84f1d34e8545e2a0b0f
d70c647f8e682164a7e22d76da7dbf270a432bd5
F20101115_AAAYHT khanna_r_Page_119.tif
35c4135dd61549df34089ca8486a5f52
a51203ea449839eda1b8d73bf37ef9f829940edd
F20101115_AAAYII khanna_r_Page_136.tif
a7ed64cac69499f4ad35e01779f315f3
fdec2e5945c7923449f4c6b80601a62aa12be97a
F20101115_AAAYHU khanna_r_Page_120.tif
0d87203756af89d41a8c8739b6c300dc
14870d24ae3cfec90a1ae7c619ae1cde83ea475a
F20101115_AAAYIJ khanna_r_Page_138.tif
186cf5410891568cb3c311afb3611edf
f9fa5d7934ad344d54dd2684d58268726f192ae3
F20101115_AAAYHV khanna_r_Page_121.tif
5df15d22dc3c225aceebfd571a1c1bda
c8c0de0356caaaa8563ab0287eee7fa943b291fd
F20101115_AAAYIK khanna_r_Page_139.tif
b1f718e706170bde7b2fb865f07bb06e
08afd797a7e04e3ca3c524f2e4b0649e33cdea06
F20101115_AAAYHW khanna_r_Page_122.tif
886174cd7a63749e742ed57e216a9998
345d1cbf426d27e44274f4a946f8317a2fa40384
F20101115_AAAYIL khanna_r_Page_140.tif
f070a6697a17ea2c9f16b6c522457912
3671a680ad34f2b6258326159f77975d1d1fc039
F20101115_AAAYIM khanna_r_Page_141.tif
11733c71b0ca856a544dabe1f5640e7c
4656fd723bcb9a7f7656dd44892d5736dd317b69
F20101115_AAAYHX khanna_r_Page_123.tif
1172e5a95d13ff7c194b857e2d07a5b9
38144db07e20669c9fb9051619013faf54c8c26b
F20101115_AAAYJA khanna_r_Page_160.tif
17095e60fb9d46367fe0d10b5e6635c8
d978ad7c6aa8e65f3fbf21be5561c49c783dfb8b
F20101115_AAAYIN khanna_r_Page_143.tif
329d52724bb045744ce572c100ad2af6
a3657b7b82ca9aa07bac7898408901191ab0bd6f
F20101115_AAAYHY khanna_r_Page_124.tif
16f1095df84123685f5f56cc56f4b2b7
34e8e47dae63fe35437c108359b3f9a2840cb3af
F20101115_AAAYJB khanna_r_Page_161.tif
8f72b4682fab2c63fd7d5d17c9312e3c
f1000f636b901e36e55761522bb7a7357c26e8ae
F20101115_AAAYIO khanna_r_Page_144.tif
7e9aaa9e484b2eb4a0d9013e09ff17b6
5747aa9922aad3c22064689588d446a2d53a0b03
F20101115_AAAYHZ khanna_r_Page_125.tif
3a5546059c7fee82edf11e45e4377294
34bc3749f63f9e7e56bcd61e45deabcf865d4095
F20101115_AAAYJC khanna_r_Page_162.tif
474ad16193fbc1bc88680c920ef8c101
fef0475e799f40f252e50bceea811927ca49bea9
F20101115_AAAYJD khanna_r_Page_163.tif
11333f5cb15b0a1a5bbc9df164dba886
ebe5c20cfd04f7d83869e23c5692ef16b2c2672e
F20101115_AAAYIP khanna_r_Page_146.tif
098c299402c0576159ee2e73d371d190
f6195f059a8d7f732d82c9a864f90221cb984f05
F20101115_AAAYJE khanna_r_Page_164.tif
a8ada338afefe8745cd12cbf9cbb1c6d
a26c65313ce30edc901b0077a28b7a95d45c873a
F20101115_AAAYIQ khanna_r_Page_147.tif
6db3a615c4749c3b5bcea271ad35a33e
51549e115309f817f664a32d137382d44e882501
F20101115_AAAYJF khanna_r_Page_165.tif
f7713e52fc5f91a0d6e3f7ea8678f1a7
70281b8335c23a02fbda1258a96a828afaefb80d
F20101115_AAAYIR khanna_r_Page_149.tif
5c22268e582c26faa596398a4e7d9298
a3c5cfa0cb438132849082ec63034d4c7830d7d6
F20101115_AAAYJG khanna_r_Page_167.tif
734b595f15ed13c43dfb6425df69bbb1
81767798121dd9a3b7a0db07d9b0f3a588c3019e
F20101115_AAAYIS khanna_r_Page_150.tif
f571f355c4d43aaca5cab77700031f1f
49c38c27bc4d870171cd3fc9911152ea0a320f73
F20101115_AAAYJH khanna_r_Page_168.tif
05b302ef2a59f3e018336333b6742a4f
851b951a6ff065215325dd0b4a209065b95445d1
F20101115_AAAYIT khanna_r_Page_151.tif
809c0859e61342f4bd6e430be69ab374
7d5bbae79bd85e614f05ef96bcfa5b86af433af8
F20101115_AAAYJI khanna_r_Page_169.tif
f8e2aae55a37d5755859c19efca788e6
73e41c5c53af9132ce97706da27849a83994734a
F20101115_AAAYIU khanna_r_Page_152.tif
e31e0b3e6f63c175e603183e81e2ec08
98022a3b8265543fbeaa1f8cbc9c8e45d8486516
F20101115_AAAYJJ khanna_r_Page_170.tif
5fba20e6e34b335cd573498d9e5ba124
2ee374cc6d5ad857a0a3b1290233d67ababce706
F20101115_AAAYIV khanna_r_Page_153.tif
9edbda31654afa92df812b703d2b833b
6846c4d3b6ef080e3a1bb8c01960a16f0ad586a0
F20101115_AAAYJK khanna_r_Page_171.tif
c2d2be5911b9873258e192b564efe2a2
50408fb50578a95996e494a58231e56bee986ed1
F20101115_AAAYIW khanna_r_Page_154.tif
3fb3e775d26fafbc792026f11e2045eb
ecf1bccc4380d0e1ee4a9f4a9a1ebe5056ba4489
F20101115_AAAYJL khanna_r_Page_172.tif
84af7996fdfa7f2f3067699719eadc23
3aa06f5e90d7247181bae2d65ae1172f7fcfef85
F20101115_AAAYIX khanna_r_Page_155.tif
441a2ad4675810a6258f10ebf8bd9882
3d568f67fad292cc18045cb60b7d1a13ef082f2a
46168 F20101115_AAAYKA khanna_r_Page_015.pro
d27b456850ef3380c227d59f621f2d21
56661f8790772f47f68aa82896957f7638ba2f53
F20101115_AAAYJM khanna_r_Page_173.tif
367658ae9daa3d7be537ff9fcff77d78
62530130312fa7bf5ce7db0a0fbe0f9489723f7f
47197 F20101115_AAAYKB khanna_r_Page_016.pro
d3819f0cf8f665b2ea3bd2c799ede272
c1daea3382739bfad7d1111ff82aa3875c82add1
F20101115_AAAYJN khanna_r_Page_174.tif
e945eddf316197fa8153e4b2be32ec67
0c2d76d54e61f3b2fd2feabe4f55a80e395a82dd
F20101115_AAAYIY khanna_r_Page_157.tif
66dbc9ffa160e3971c2f31d6e10f964e
da524aa90acdc28e1f9a19b12e793b34866c7ddf
24317 F20101115_AAAYKC khanna_r_Page_017.pro
ab1a61b673e84d2e312a5b40289b7f7c
b8384dae94c8fdfa248446e25d2ad583ef5969bb
F20101115_AAAYJO khanna_r_Page_175.tif
f450a4f60dc15ce3ac8835486dd8d7e2
f7a33dacee3cf236dc3dee457973ccf177b67ce5
F20101115_AAAYIZ khanna_r_Page_158.tif
e13999295dafa694c011196bcfb3f33d
bef2c0d563469e415eafaa9b21f9266e3e44afaf
55913 F20101115_AAAYKD khanna_r_Page_019.pro
e22a976dfad63cf11394d8352bcc03d6
fb782b04fe1c6a01eb3c86ba125d013a03c1bdd5
8008 F20101115_AAAYJP khanna_r_Page_001.pro
8ad489d87ea9d51c70baa38828d1b455
653e79a1f50a13f0b310c5fa57ddbb33c320c2a4
53951 F20101115_AAAYKE khanna_r_Page_020.pro
1a235e4d70652911a5425cb676bec250
59fe667db9cf1fc637839682f60c7944463e0a01
596 F20101115_AAAYJQ khanna_r_Page_003.pro
10040909b65b1d5d790ab78eae892d60
26d367dd77653683f5eeb914bf3768f6eef355ab
45666 F20101115_AAAYKF khanna_r_Page_023.pro
46480bcee01ab11aca4f90bcf0c2de30
58e8a84d1fb783ece34548fae4f028a4cc8c8da9
54370 F20101115_AAAYJR khanna_r_Page_004.pro
9570f387916c90c2ff0d5bad5dcbd76b
054b549be804bbbc64017f466b71e3e360804146
54037 F20101115_AAAYKG khanna_r_Page_024.pro
183263442c5c98163d3b451a1505d97b
e4d081654d6334891e7d8c71efa704d987b18758
9792 F20101115_AAAYJS khanna_r_Page_005.pro
b788e6cfcca69d85737d56bcf459eb6e
d7aa773b3ad4c2b548aae94af8d696d5b1271e60
57240 F20101115_AAAYKH khanna_r_Page_025.pro
ca4c93c57604dacd34f94e016cb2a3bd
aea560c87d524bc61825a4136bf6c2d617259745
80466 F20101115_AAAYJT khanna_r_Page_007.pro
6e97a8c1b59fafc97b489c82fe242801
08090fb7df8a0ab0f35a474594def115781f20d8
48555 F20101115_AAAYKI khanna_r_Page_027.pro
057587ecb7b3ff3b6ece9e230a1768f5
eca674271cf9f40c7d929f30b3c97020a302776e
36755 F20101115_AAAYJU khanna_r_Page_009.pro
7f77ddab91153d15acd2654189ad4858
55c7abd8428abf595cedbb1b7ee62e7049efdb36
45150 F20101115_AAAYKJ khanna_r_Page_028.pro
10936a98a68408685e93d2af24aab0b7
ee564d8efa3e7664e42f22db25892cc636567085
63377 F20101115_AAAYJV khanna_r_Page_010.pro
259371a0c0ef5e41ee8c20fa7d3f597b
5c81df6e85a076d970487142cebd3a19f72f49fc
44643 F20101115_AAAYKK khanna_r_Page_030.pro
4d1907b95e423861138d82049c526756
40b1f0e9e05612b17ccbe7fda07eef5c73186903
73270 F20101115_AAAYJW khanna_r_Page_011.pro
a4b694f946a54bb7ab7de6269e2ce799
9ec4cc68637119769aeab0ac9a21bc7826a002fb
53994 F20101115_AAAYKL khanna_r_Page_031.pro
c8a414accbc4e370c2163cd9fd4fc5c1
b5bacdaa74b2a11101253e96b8bc835df0012e16
69669 F20101115_AAAYJX khanna_r_Page_012.pro
d7837eaea732ff84e505a4a4d89bad0c
aaabdbe79a7e704451ec0fd12cf7827947cac278
41556 F20101115_AAAYKM khanna_r_Page_032.pro
e7119c4d13c27ac5db5fb1ad1a5d5472
ea80f6a0a636ea74922ebec19d9f96fbdb0b1703
74507 F20101115_AAAYJY khanna_r_Page_013.pro
6d94dcfc9490ce160cbc50448a2a4033
9f031debafa4d76fe93dc9b1ba9d1330fab4f601
4646 F20101115_AAAYLA khanna_r_Page_048.pro
b1180ea3ccb800bc5adbb801c26dd4eb
e5d0a65d10d7b1e82ba3411465970406488d91c4
55845 F20101115_AAAYKN khanna_r_Page_033.pro
2e024864504bd76d8c514ce9e0399fd6
0bdd3ac23325c474b926e5f14474b63614fc530b
14594 F20101115_AAAYLB khanna_r_Page_049.pro
dc08e8b54df84855f02f796bc18d5e06
d26306107d6b0ecb2547813b23726d81251bfe58
55979 F20101115_AAAYKO khanna_r_Page_034.pro
03c81f493467fc0086163e37276d6098
78f26d407c7c36295ba89b7cc7d9882080ab2667
52806 F20101115_AAAYJZ khanna_r_Page_014.pro
687bcca6944c6c188bd6a3414a3ca280
e7b08e68bff622719df2adf95f9981a1689e08c8
13356 F20101115_AAAYLC khanna_r_Page_050.pro
17a63a337c7613b07a9a9fbf3c7de723
c6a16115bbea6dcd24be0945ea622a697f9213bc
57056 F20101115_AAAYKP khanna_r_Page_035.pro
f3c584f603eb470882268ea877dd3e4f
fa9402a2c23d41c05d99e0dbd488a135950315b8
2609 F20101115_AAAYLD khanna_r_Page_051.pro
83db889011ab682756f9dcc6945892ad
a5126832b801a55745b86f840388ab8b5db35512
55813 F20101115_AAAYKQ khanna_r_Page_036.pro
cea52d6f2f1ea052768e91833663c552
8d6f4300cbb7b445b4d6df8619e1f2d86164c92f
55055 F20101115_AAAYLE khanna_r_Page_053.pro
6fa91fa854d85da34e417ade96eb7fc6
1c67fb287c32bb2922845547c50a7c616bb23145
51615 F20101115_AAAYKR khanna_r_Page_037.pro
951f4667078da5ba0596445a02c582ae
c02e69282fd6203e3ce39c66aa8126f85b6622a1
52234 F20101115_AAAYLF khanna_r_Page_054.pro
ee1ff663dfe1e4787dc4d8dda0cc88f4
c4bfcf22ff1996e491f828ae4f58e069db8896ba
50199 F20101115_AAAYKS khanna_r_Page_038.pro
0d1c429a4a332b4596c0b82f1098ccaf
b51401186b47888de0a4f8fcc8f09225190f83da
53833 F20101115_AAAYLG khanna_r_Page_055.pro
30b92b250faf3f5b7b340c4426905af2
0f2e71bcfc5ab07d8ee05998fa8038bd133d0841
52242 F20101115_AAAYKT khanna_r_Page_039.pro
cb3565b7b80d0e472ab65a882bc4021b
197fb1c79102ab717c40303e17836005cb61e830
54226 F20101115_AAAYLH khanna_r_Page_056.pro
1f848ccfdf65c17f12a3a6bef4701e2a
ff3b7a52ab544c267215eb3625af082620f593a2
51852 F20101115_AAAYKU khanna_r_Page_040.pro
d1cff23225ce73630363a6f85ec1eb41
6d0dfcd146f779875313b257bdd4d52553e83cd8
55947 F20101115_AAAYLI khanna_r_Page_057.pro
958a4ab81786bb2a59ebef935754e682
3c3b6e59552c013c5710cf08fada6bb8eb5d301d
45657 F20101115_AAAYKV khanna_r_Page_041.pro
6dafef5913ed3b55900f7a6274e50a30
0b7ccdf1a7d65ac768ba34707d078b6bfc3f3ec9
54349 F20101115_AAAYLJ khanna_r_Page_058.pro
839e6b8cc581b86aa1b56e3e50f1170a
25bc95f0e5a07a2c8ca0b2286ef79e35d09774b6
20277 F20101115_AAAYKW khanna_r_Page_043.pro
534ca71a118800ba59e1fef88f239315
f70ae10be463b9ef35f0beb550c71cf244c85a61
51178 F20101115_AAAYLK khanna_r_Page_059.pro
66d5ff10e0ddca335b89949d393d1241
7eee4530b9bafc50da909e528a29821dd018d4fe
2547 F20101115_AAAYKX khanna_r_Page_045.pro
2536e64f09c2ffc168e3f12a00d1947d
c4a88f89ca15e7a09450f219bc98afbee4db1498
6719 F20101115_AAAYLL khanna_r_Page_061.pro
d2b0556b88f758c6272daf4c9f12aa6b
05d393f613c99010ad6e9b330ab8d0234a09deb4
15395 F20101115_AAAYKY khanna_r_Page_046.pro
2e123f97b75d87529cb368728899984c
282e561ac7074f775a9f08d4929ff6f8ae7ddb65
8434 F20101115_AAAYMA khanna_r_Page_079.pro
1511891faa9c32809f95379dcda81d14
d4e07094315802b854b1fc8e4ae414532c294b7b
9470 F20101115_AAAYLM khanna_r_Page_062.pro
4ab8410ff72f80adbbca8b2150b99f04
6d46c2944ab90d00eb0f3ef2269c82182a62b1c5
6228 F20101115_AAAYKZ khanna_r_Page_047.pro
d83fd94b9215e9344fcea982f02c0251
82a2904164fe269925c9a0e4697d829a61ebffad
8376 F20101115_AAAYMB khanna_r_Page_081.pro
b1189de8bf073aac5a38cd939808566f
f5cba7499624722bd47f78f929d627c3514644ce
5984 F20101115_AAAYLN khanna_r_Page_063.pro
f6b6c9addffe70eb0049a8c43164e5c3
688d6f3a2812efc053209df1d362e2e5d08a3bb4
15477 F20101115_AAAYMC khanna_r_Page_082.pro
c76dde5f4cec2161206fd0e3ba192d83
74fff0e314d66b0e36599cbc23ee8464ba518fa3
5793 F20101115_AAAYLO khanna_r_Page_064.pro
71bf93d45f9bf71490903b4ff4a4fcf0
8bb881cfe812d1b3489dbb87135ee22b117d1acf
18498 F20101115_AAAYMD khanna_r_Page_083.pro
99b45287475318884a75580c6ea901b5
c10a33410a401455129385f290c033f580b708dc
21466 F20101115_AAAYLP khanna_r_Page_065.pro
01c4713b6a29d6a274e3f0a722377b73
374bf58292608b1ad6c4b2059d9cafb926e265c6
16552 F20101115_AAAYME khanna_r_Page_084.pro
cc1d6c55be7a2fd29b49a3f827a402b7
757597409502e6c1a245e74952e3e9859618f0ad
5562 F20101115_AAAYLQ khanna_r_Page_066.pro
b3f5e25e786a962c686ebc0b58da0411
40ea1cfeb0ef807ae1c98a6a90a855e41cfb1046
18716 F20101115_AAAYMF khanna_r_Page_085.pro
bcc98c0e616407195e7116312f5becae
f04a83cfd1b9156d7fc632693c8dd7c6b0e4dbc3
11203 F20101115_AAAYLR khanna_r_Page_067.pro
e79ebb9bc55bf74206c587a2271003d6
a050f1e605ad0102b715be9b97c14b63e3db3d94
6303 F20101115_AAAYMG khanna_r_Page_086.pro
3faa8def086aaf18227db5a0c6d74651
678434b11cb3e98b32563229be3ac0e95e55d4eb
9817 F20101115_AAAYLS khanna_r_Page_068.pro
2332ab7b881b2fa0308aedec66e93231
1079cc565bdba2605628e220a62e131a0d9e90ec
51211 F20101115_AAAYMH khanna_r_Page_087.pro
e6551ffa3f6efe81f37724bc98bdd9ca
86307bcee6214d8d9ae79efcc6fd225de8ce059f
5957 F20101115_AAAYLT khanna_r_Page_069.pro
b273c9b9a8fca4a4ef51b5a98ecd2d35
5f4444407249b862c95649bd7b48dd5f7a381157
52448 F20101115_AAAYMI khanna_r_Page_088.pro
b9aee226334aea60cfec674cf383493b
e2620669888ddd9c888173fff0784dd82f9875a1
4551 F20101115_AAAYLU khanna_r_Page_070.pro
ba7382faa8e1ed015dc937aca59aeb60
3d4c23cb53121636cc5a18314ba0508631b0f344
47434 F20101115_AAAYMJ khanna_r_Page_089.pro
358352e84153d1e1d784157266c15a6a
d834e1b0fc5a83adb19b240c16b0b68c00f8b547
19238 F20101115_AAAYLV khanna_r_Page_071.pro
7afc9ddf8e1058b4dd142b0e1ec91732
a43c9d16f2585741ce6cb64762cac11078dd1803
46052 F20101115_AAAYMK khanna_r_Page_090.pro
1f55578c91d7662603509bf5e9e1c516
bc70dfa6ebfeb104e57c127b2cdb4865186f3304
12453 F20101115_AAAYLW khanna_r_Page_072.pro
d6684a5d2e91f545f0e1116e609daf0c
59043e3db36b6b802f003503df4c9ebd2540f08f
53689 F20101115_AAAYML khanna_r_Page_091.pro
0986bcf11e2c099d08caa350a233201e
683e52b43d8de07f3732afcacf1f7f5acfeb60bf
48099 F20101115_AAAYLX khanna_r_Page_074.pro
d205c3e12ea535ff74ff59b5e6254807
87c53ea0af0e5407058d5ad9abf131ab7030f0dc
54480 F20101115_AAAYNA khanna_r_Page_110.pro
5f495ca7e307e1be160559b671fc575a
6ad5948ac4cf40f3a6a5e834897217821e89aa55
52418 F20101115_AAAYMM khanna_r_Page_092.pro
03dc79ada87c55aa501d386d204334b7
efb313a5131b2c9f74d8e410cb1e2b59c6a68a42
51811 F20101115_AAAYLY khanna_r_Page_076.pro
8ce6fe2d666e360bb078f8d91e295908
0a9f801041e1feac04e6a9689fdcf1a6e6614f45
55571 F20101115_AAAYNB khanna_r_Page_111.pro
9ce612edbf0f35ad19940b9dcb62fa9b
17ceb6d615e5eac03c1837f5401d0f35764ce4d5
49712 F20101115_AAAYMN khanna_r_Page_093.pro
58ee1741aa25ccd731d3852c22d1c319
a8684f95faec32f3d5be135a45d07d50d2330152
20566 F20101115_AAAYLZ khanna_r_Page_078.pro
aeebfa5e430ebf19cdd5490ea8177f82
c53a4f5931c495163ba3355c39001948e83f9471
48609 F20101115_AAAYNC khanna_r_Page_112.pro
5f81c0f3e17184483639990b8fe8702b
1c68ed60ebb0e77ca1e4081f0bfef2e1d3b9d716
6519 F20101115_AAAYMO khanna_r_Page_094.pro
6d92cc731a9bd324e4efa7c06cc067d8
b4adbfc47389b7aa4e64330854d39f410d98c042
50884 F20101115_AAAYND khanna_r_Page_113.pro
82c540e1bb4ec4180a087576d9df159a
43db79362048785bbbab83ad855e50c9d2b5f090
4872 F20101115_AAAYMP khanna_r_Page_095.pro
daf1c8483e6943ad5970465aad4452e4
98fde11522d5c878325612ef26e068c2076dea4a
49689 F20101115_AAAYNE khanna_r_Page_114.pro
4b57024caf7e0d61f7be944ec5af224f
062a529eef82afe571e93599473acb54e7aa3a8d
13553 F20101115_AAAYMQ khanna_r_Page_096.pro
9acf3ec7402cdf12d34d24f5b1242fc9
71b6c1160c99b3b95e66a8aa0a464bebcd312587
F20101115_AAAYNF khanna_r_Page_118.pro
f1f4b12b00808d71e7a49075a37d6eab
9671e53495311ddf6a58be551d2044426a25b1b7
7170 F20101115_AAAYMR khanna_r_Page_098.pro
022ade047005b16ff2f87eaf6f0945e7
974ea5a915053bb0365d659ab1a9b6bba3f12fbc
9720 F20101115_AAAYNG khanna_r_Page_119.pro
84f5ffefbcc9d0d183c039f250855c84
f8bff8c86aef9bc3880f2dd71b7b69bb79b58d1c
7219 F20101115_AAAYMS khanna_r_Page_099.pro
0dbd62961a909cc2f6c3a4fe6697c8d3
0da62daf70b8ef8d12ba9698b87dea3819281247
12333 F20101115_AAAYNH khanna_r_Page_120.pro
5bccb176a0f223d064c12d7ebde370f9
45a13dbda5baa490709a165468edede135ea818c
11860 F20101115_AAAYMT khanna_r_Page_100.pro
9fc70f0f804e453d4d2678a5a03d7245
83e6f8ac0622d4bcd4f6fd5f649655a221b4481f
5720 F20101115_AAAYNI khanna_r_Page_121.pro
58b5b55a6e8948d2a2a3e940885decb9
f8893fe3dfb61431d125327b96debc49b2a329f6
8828 F20101115_AAAYMU khanna_r_Page_101.pro
c4fa269eab6d3cb9a41daa8e9e012150
c8a691b259cdc2446d46540b6ece31b12245c7a5
9438 F20101115_AAAYNJ khanna_r_Page_122.pro
2e6db3c4cb2e0872854cb53fe14d17de
f05e7f35c4d9ac274adb65ddd2a012a7fde5f0e2
12374 F20101115_AAAYMV khanna_r_Page_103.pro
085052e1910632310cb4d380db9ec917
57f5e2ef9b3905efe3fba366335d4f97a78be2c5
5941 F20101115_AAAYNK khanna_r_Page_123.pro
efe386b3142ba87a2521c97b5e5d8396
7e3f8d262e175d6413acab41612b268abdbf966d
9190 F20101115_AAAYMW khanna_r_Page_104.pro
2b3081d6f5676b9fbf533ffb9fa76e02
c71de24413b87c4602b2e0bfa565c6786e5c1c17
5224 F20101115_AAAYNL khanna_r_Page_124.pro
93fe4887b079cc12bfe2e81d755bda5d
dea1758a323cfb98a5704a3a37ebd52fc6418c6d
9462 F20101115_AAAYMX khanna_r_Page_105.pro
fd409fa71bff5cb991f23067dec1df6f
ed8811383220a4077ba945a018c62cd1cbda6a9e
53357 F20101115_AAAYMY khanna_r_Page_106.pro
21c8ddcd27873d165001b26f739999aa
d82af0a783a5e65b6b8b7031247be3515391ecef
10344 F20101115_AAAYOA khanna_r_Page_143.pro
9e79f384bb8ee039327914153d70fc9c
e30fd2cfed634594867f4da5fdbbffe22f45086f
77144 F20101115_AAAXKK khanna_r_Page_110.jpg
dd45791f1c022c613f903f5e5b0c6c1f
074e5d27854450d617e58df15061eb7825c7c236
6877 F20101115_AAAYNM khanna_r_Page_125.pro
e7c30275457af6c8f4aa2a2b5068453a
d361d07eb9170e9c73f6200738b1a8e3b82a0c56
50715 F20101115_AAAYMZ khanna_r_Page_108.pro
1645cf27171e046eb9a4429b73db2b43
6a9f3c4fe45298c4835da7ff5be9511815b3a57a
8776 F20101115_AAAYOB khanna_r_Page_144.pro
25051f2e6a2d5f37cce9e8890dca8fc1
dc189042866281f75165233430f1ef6f0657a8d3
1914 F20101115_AAAXKL khanna_r_Page_030.txt
f3957d87089a636da7a4c11b8ed4c622
c3fddc9f18c7618984f828d82bb2abae7a6892b3
11246 F20101115_AAAYNN khanna_r_Page_126.pro
b8998d251ce8d0014c5637f96e41266b
4209dffafcbd016772a88324f8cbb73507447211
778 F20101115_AAAXLA khanna_r_Page_002.pro
aed13936323f93f3fba43df58fc12ea6
ef6368f5d39eb139ed8b574b3929ee9c9ab37223
10090 F20101115_AAAYOC khanna_r_Page_145.pro
4694459183602034cf1ae992c8db8d6c
eddded50420bc1a8998cf24ca748fcb877d2aa32
F20101115_AAAXKM khanna_r_Page_032.tif
f07f35586a0a4bb901c679947f53569c
224fa459ba0bab0fc58ab2745cacc522153e4cf4
7391 F20101115_AAAYNO khanna_r_Page_127.pro
10b88a66185dcd33d54e785abe3974dc
284d652af953573ad3d6c145024f3bf069176791
9963 F20101115_AAAXLB khanna_r_Page_068.QC.jpg
63b5b680dbe26ce6a923c271df65cc27
2aafef242c79316dece3d3562db25f15a2917e5d
9888 F20101115_AAAYOD khanna_r_Page_146.pro
c0fc45d30b63da0720454842cac0e822
bf981993175b9a0e5af87669d82cb0c758f7f383
18869 F20101115_AAAXKN khanna_r_Page_042.pro
a9b1c254f3916e03159a5f05866d4a0d
2a3e5c069e6791253aa23ca09f4902426ee2174b
11173 F20101115_AAAYNP khanna_r_Page_128.pro
fe8c24725a00f4bfde4349db58da9f9b
74f0a330111d9f5238fd6e49c85310d18d0d9708
9238 F20101115_AAAXLC khanna_r_Page_129.QC.jpg
43d68c0181a9183de4257969993fb1cf
c2c429b5d71161253ef24b82e31f2fc20e7d3c90
10523 F20101115_AAAYOE khanna_r_Page_147.pro
baee35c605e6d8458748e79cdd1e8b43
4c6e2c397e3df69065dfa8db7a29f9ecae4d9897
23364 F20101115_AAAXKO khanna_r_Page_061.jpg
bef0e81573830285f936b427ed9e8f9f
c4eb0b77a922020d2b811a43d6cd6f36c90d2cf8
7434 F20101115_AAAYNQ khanna_r_Page_129.pro
251395c129dddf849f6a60e506c35f66
d245dda2a5254f4a27b08c2f27cf5211eb70be91
177 F20101115_AAAXLD khanna_r_Page_133.txt
4151bea2dc52f284989fcf06cd00bf3a
91d2be01fc1650a7f01ff77cf49635a72761d547
53894 F20101115_AAAYOF khanna_r_Page_148.pro
87a2087f4a2fe3bac1e5cc394c03beae
bd13f3e599730cf9f0326a459566d619b5dda30e
F20101115_AAAXKP khanna_r_Page_108.tif
ebdaba2f04802064f9ad7bdbb645c9fa
94eb24e6c9f44a6b1e79673e13b4c1270e73336c
9849 F20101115_AAAYNR khanna_r_Page_130.pro
d175c6616c5d1b4a87cc5d5ce626fecf
b89d358c6431ce093746bd4414af9dda99b532a2
21587 F20101115_AAAXLE khanna_r_Page_006.QC.jpg
feb2c07c9376ff31b9857bec34e84d00
4b8c15d02ff133a3e83a64ad985a20dab627d771
51468 F20101115_AAAYOG khanna_r_Page_149.pro
ce8fc95c1e395950190db22455cd7d30
ea3fe2b316c822834404c9e811da28c4c7856aeb
3658 F20101115_AAAXKQ khanna_r_Page_140thm.jpg
de259726c57ab1f40215d31a357e6f07
f8866be86f633bc25265add288a83e4a938d12c0
7037 F20101115_AAAYNS khanna_r_Page_131.pro
6890cad364d4fd5127733712e7f6037f
85296db863fa700d7175a270f621080277c03ea1
72011 F20101115_AAAXLF khanna_r_Page_085.jp2
8e26c855e9498834a7e640163b3d03e9
68768e8789790bd510ad0980e1949455af939bfd
45353 F20101115_AAAYOH khanna_r_Page_151.pro
642968005d882106b9ff20ed3d0d37f1
1c522ccf396d42396ac15d867094d31d82b7e1c9
F20101115_AAAXKR khanna_r_Page_156.tif
5383599c27bd875ec4b3b37e326c414e
1b5c2d2fd7385aa8647fb4e18e78715b5e5b3134
7027 F20101115_AAAYNT khanna_r_Page_132.pro
714c95f0ecb791d3acced96637d3859f
b38144ec0c2a39ee3009059486077e0b65fd9a18
3770 F20101115_AAAXLG khanna_r_Page_139thm.jpg
eb75353a1a0a2b476f5f8f1ea2d2f1ca
965dc2595b6bdd40cde0234006d7b5fff2a0f1ce
55894 F20101115_AAAYOI khanna_r_Page_152.pro
7290e53a864c0a2e4afd69ac2f724bb0
4a9a94c7d521d6c8e005c6025a0d054b4d0aa181
517290 F20101115_AAAXKS khanna_r_Page_048.jp2
e3574303e5c79455e5b3ed831082d533
f29a88fbc1877016aa0609bd0cb4ef6cd3465360
1442 F20101115_AAAYNU khanna_r_Page_133.pro
3b08b215d78f114e7427ef7177c1c202
059f176d3e9a404bff6ffb96a872c7be3066a565
27963 F20101115_AAAXLH khanna_r_Page_116.pro
5522e1d709e157a4ae7c64d17e446511
963f031fe07f43d32eaf36bdf28dfc5565c4a672
39322 F20101115_AAAYOJ khanna_r_Page_153.pro
8b1e9dc4c4fada80af465fd99916eeed
e52ed84d24a091e56bbaf6f0e50f81f1bcbc5fc6
42102 F20101115_AAAXKT khanna_r_Page_143.jpg
8051287eae3f9ad3a260e828636106f7
54bdd580b0bf4727cf8d96e62a49d173c3909b3a
10967 F20101115_AAAYNV khanna_r_Page_134.pro
d0d95319e7b851a4bc11836b7ab32de6
95da0b04bc5bc2011c187e4969815213a52a0eb0
2143 F20101115_AAAXLI khanna_r_Page_110.txt
613bed0cd003b52795f9cfe0d57b078f
514d0118c0d5ed4319d82fdfa81b8f249282c852
4597 F20101115_AAAYOK khanna_r_Page_154.pro
773ee8789090bd9e1db719f646f24266
3032cb12a97649334e6a77a90e716f9192022c5a
74187 F20101115_AAAXKU khanna_r_Page_084.jp2
dd47ac32b38b78abfcd92798ef222271
4a810400fe67ce1cec0aa5cdec0c861a435601f9
54774 F20101115_AAAYNW khanna_r_Page_135.pro
d8790dec6f33e891b7f73875979bd30d
b8128447ed7f29c59ab79f29ec54e82c346f81f7
F20101115_AAAXLJ khanna_r_Page_095.tif
f7207f7497dd208341b4fee84d659acb
980722c57bfec980b6265ea4d7c9fece93ce3250
5953 F20101115_AAAYOL khanna_r_Page_155.pro
440a4e4673e08ab92dd9d83f0fcda197
42ff4bcc2261e63bdca6a01a589c746dd67dea37
54128 F20101115_AAAXKV khanna_r_Page_098.jp2
944c2af5061167242c76c9460ab2d2ce
a17fe48b6c88f58e5f27224cec3a3346549f42bc
52569 F20101115_AAAYNX khanna_r_Page_137.pro
aa49c5af3a14aede695810313185e9fe
6fb8535d409de19239d596e0cfae3d8ddc38b0e0
49817 F20101115_AAAYPA khanna_r_Page_173.pro
1536ace0476da8b8c2a2832cdebff3f5
b0f7157e24e0e026be499c04f787e38383e63ae9
6531 F20101115_AAAXLK khanna_r_Page_088thm.jpg
1006e6a1c4ba9a97e8c30fee88e68a4c
1bd101d0982921e77ee9f726e908980513f41fbd
9984 F20101115_AAAYOM khanna_r_Page_156.pro
116ecb08c2f2c8458b31036f970e3c44
c7bfe55997f687d78b3d954c1fd4ffda29a7ab5f
51330 F20101115_AAAXKW khanna_r_Page_115.pro
a4f29d8160b8d4c87b0b3b73ed86e117
1be235e42b713fba69c6d7e2f93a87865acd6ccb
22998 F20101115_AAAYNY khanna_r_Page_139.pro
3e506b643988ae5baf05ed55b2a0a328
482c4c42841b6a172014b68984aac4ccf98a5972
27579 F20101115_AAAYPB khanna_r_Page_175.pro
d9dccb8c5ad399172b4c155cc753e30b
97aaf6079ca34f79ed844297473cbd1a57df8178
34172 F20101115_AAAXLL khanna_r_Page_081.jp2
bdcc3647b7c9920563f41ee1873a838e
ef11e6063cf9f0a69f237acfd2eef0aa147096e3
7544 F20101115_AAAYON khanna_r_Page_157.pro
ffa7a0d1e06fe66893ac0bc2f1c20f24
fae75d8db24f7d9e0341d5727163bba0e3dfc700
52399 F20101115_AAAXKX khanna_r_Page_107.pro
e280aab050186737b407783405308bff
b25864c2312161b0ffc9e056a4e0b46a88db1d7f
7297 F20101115_AAAYNZ khanna_r_Page_142.pro
8c4ca67b06d7031fbbca31ef48b36861
e6767b687700b5313c01ba2ebacbfebbc25a12e2
474 F20101115_AAAYPC khanna_r_Page_001.txt
00bb4a3a5ad44b83257f2650a2bf3067
247a3108d910ab2f09c880edf189fdc5cd195367
706 F20101115_AAAXMA khanna_r_Page_147.txt
a3c24fa85a1931e0148507017376b67e
6f3ce5262aae1037cfaf36a1e939839b5587b5ff
1051985 F20101115_AAAXLM khanna_r_Page_167.jp2
4e55bff559baff268aa78e2b50cd6b36
effc2ebc0cb33fa427defd3b10aeb3c13b708a48
6026 F20101115_AAAYOO khanna_r_Page_158.pro
4aaea60186586b951ed2850bb6562219
9f9a5479f82fcda28f21df51c18f542436c42a1c
6443 F20101115_AAAXKY khanna_r_Page_141.pro
a09d1035066108d70f5052cab06d3b6e
544317e62f2188b09a4c9b9418df046dc8611fb7
2178 F20101115_AAAYPD khanna_r_Page_004.txt
df6c896a86f2a566a51ea86ef742f4b2
1d004326cf69d2c27ade389f28ca3463afc50a31
1992 F20101115_AAAXMB khanna_r_Page_038.txt
57e553332c54e63e22db53aec2b5175c
eb23ac81e643d380b71d0a6e5467ca31ac21eef7
7486 F20101115_AAAXLN khanna_r_Page_011thm.jpg
6b47af8b8bcaa071079eff3e2868d85f
22f1c03af5dd6e51c42b462a9ad8022d8e0da2ae
6943 F20101115_AAAYOP khanna_r_Page_160.pro
bc33564d18118cec9bef3109a0f70b1e
b59dcb0eb500e956f8a347c8dc2794866a5b1a4a
7080 F20101115_AAAXKZ khanna_r_Page_005.QC.jpg
ce3c714ffa3a1b8f6b5ebaab2244dd6c
e4df621b5de7808ee1e1cc47c32aa23454975972
390 F20101115_AAAYPE khanna_r_Page_005.txt
807b123a782427bdc71e4c3b83a2c634
c55afcb653502a772fc3e55180f2ad914e6093d3
6060 F20101115_AAAYOQ khanna_r_Page_161.pro
a7e0d929cbd0586a460cbf7be077a988
6f98d69d496c5f4fa094c1bbcb1494dbfdd61041
3610 F20101115_AAAXMC khanna_r_Page_096thm.jpg
8633a14ac8fb3cd22c517ba312423a41
23b6df311bf274044c0926a317f2fbec41f1724a
76502 F20101115_AAAXLO khanna_r_Page_006.pro
e1d3a6e46197bd9dc62c02d18541c327
6c137da7001a31436211022d4f7b98ac6dc50d2f
3383 F20101115_AAAYPF khanna_r_Page_006.txt
a64c46aa4496cadb7110b0c3d85fa8d5
994ccc0ff9e7a28ac6316897eac831bb6e5eb301
52607 F20101115_AAAYOR khanna_r_Page_162.pro
8cc530960d0d2dd5675ec7bb1dc07d35
83b4cec2ffc6beaae8e81eee1167b2f64c359c2c
6445 F20101115_AAAXMD khanna_r_Page_029thm.jpg
0b224ee2d21e656ca41c7d2fe078a25c
d3c37585ebe57cc3399f4cba39f4f1a6dcd4372e
F20101115_AAAXLP khanna_r_Page_073.tif
69736ec5805b9973bcecc265cc324881
7a7d9680dfead0c4ad83dd532c6d7c3aceeabc47
3278 F20101115_AAAYPG khanna_r_Page_007.txt
8eeb08a9de126a4404da31c97102b8f1
39678aeb813187af2c3588e6abfbb4b46dbdc835
52774 F20101115_AAAYOS khanna_r_Page_163.pro
6620fb0bc31215bd47167eb00a5c0839
e854aee05dc31e10484cb699153b834d8ec6d74f
F20101115_AAAXME khanna_r_Page_135.tif
8c9a3e52d02ef6d9d5a73da598217bf6
d2605278b9bc6e603fe53332202b7746ecc5138a
3368 F20101115_AAAXLQ khanna_r_Page_099thm.jpg
f0225b3abf52a6816f7cb1eb7689f4cf
20916eb34d58efaaaf1b2565e787bf12af515fcc
102 F20101115_AAAYPH khanna_r_Page_008.txt
b30019c1d572a805d056601187d58c25
5549bbc42490bbcd7a515b29489bc310cb7094fe
56451 F20101115_AAAYOT khanna_r_Page_164.pro
982ff0fd675d9a7214dbc1541ca7affc
cfa6f39159f600125ae6d5ee89e78998fb6fd047
33641 F20101115_AAAXMF khanna_r_Page_146.jpg
f833bd9a98413a6947b0e29973477238
fe52bc223ba8e34eaaaa37ebd87bc2a93d4dad63
12112 F20101115_AAAXLR khanna_r_Page_158.QC.jpg
3161864010ad50b7f00d63e69dc9fbf9
1e1ee5a411230c0076af4506d883ea9a08a83421
1530 F20101115_AAAYPI khanna_r_Page_009.txt
2063947ccb7140cc11c3dd10778e1c06
db910d4c9d5a54a20a0ee313e652efa1f1260109
54078 F20101115_AAAYOU khanna_r_Page_165.pro
4d52b80d5dd3e7b76fcc7cc8bdcb4aa8
582a8ff0d53e060f376950c683b44bb6acf833cd
13985 F20101115_AAAXMG khanna_r_Page_046.QC.jpg
82dffc7e5f1f1c558e938cbe8af5f7af
2ddb6c8be78cbb3b10d9f250f807eddc10116004
2198 F20101115_AAAXLS khanna_r_Page_152.txt
c05fe3cc0176d66d00840c6040ac1bc6
719b58a6b66320cce4249e87f2f06767f1665201
2976 F20101115_AAAYPJ khanna_r_Page_012.txt
9768063c3e5dd1aa3a1860a742dc5834
58a3447e6c333554c093fc7e9fa87cad90f61b1c
52037 F20101115_AAAYOV khanna_r_Page_167.pro
b30a621ebea6043e1837e14989671da5
91522e58bb69a0db7a48d36bd3e2288a623b05b2
103070 F20101115_AAAXMH khanna_r_Page_015.jp2
66d441eb283df6a25b33cc723a80874a
22e30dcb80abade72f776685d37ab0581bb639b4
3846 F20101115_AAAXLT khanna_r_Page_042thm.jpg
50ff3d915a3d926e2d2c4ebc2cda8c42
32ac91a984909f246a665a00e85228fbb406e2f2
3107 F20101115_AAAYPK khanna_r_Page_013.txt
d18c832465152e2ed65ba22b0533a751
bc0e28901602dd4ab93e897834498654ac24e13a
53669 F20101115_AAAYOW khanna_r_Page_168.pro
2d1a5573380565c513b1c3e46523fc21
a9a02c5ca9404605aed617789354f9ac2a1a9c66
71311 F20101115_AAAXMI khanna_r_Page_112.jpg
b38de410a4a2e9169d117f5f7c1b02ee
5baf7f9a8d0081beb3e3d5f421917da8d4aeffc6
25254 F20101115_AAAXLU khanna_r_Page_063.jpg
e6fd79af859c714d7a01abccdbf1dbe2
dd599cf2ceb8af422faac30200cf2efd9cbaa86b
F20101115_AAAYPL khanna_r_Page_014.txt
d12eb1078fc85df5cf4e232f4c8391f6
cbfb069c4f581bd5631a98395e2377049f5f2958
52257 F20101115_AAAYOX khanna_r_Page_169.pro
0f7337f9ef1d3e92351d384dfdac4397
184a475ced2dd78d32e6a9f1daa2b46a1288c65d
F20101115_AAAXMJ khanna_r_Page_113.tif
707f8b90e70a2a34f7b313cabde33ed1
ea0aa57f591a1cd4508b84f3813dd6df7c8bdf70
7577 F20101115_AAAXLV khanna_r_Page_174.pro
772436c47a92654947626615daf206a2
930a55b1dc5694e76334745fce2966b4664b793c
2189 F20101115_AAAYQA khanna_r_Page_033.txt
8a67d056a27094210f0d900651be0ba8
3c31400e1d9d811f6afa94832334db5557e3a683
2030 F20101115_AAAYPM khanna_r_Page_015.txt
bbc4fea0791b8c815c4d09e8956afdc0
acca8f2b0e05bd6e4e66de36869a9c3b016b9a3f
55137 F20101115_AAAYOY khanna_r_Page_170.pro
b6c06a35944bbb1aeb3a5184c29bf7fe
af72332dd3350b9bef99aaa3e5123755d7b48c50
1315 F20101115_AAAXMK khanna_r_Page_002thm.jpg
8ea9981ff71e3eb175162baf3b4dda67
cf80c763f0d15eb95f54e22b28823f12788fd1e2
F20101115_AAAXLW khanna_r_Page_074.tif
4bd967dfa9c18348df19d4cb4dec6a93
cfe705dee73150915f4058459ce597e526af31ca
F20101115_AAAYQB khanna_r_Page_034.txt
f5535ad9ca583b6957629efb1b6ee932
0a72e2bce69c43227f9ce9d9adf9bc48188cd46f
1874 F20101115_AAAYPN khanna_r_Page_016.txt
64ffa4c0c7e2a110c895ddcb8f259288
37a5d7d8a49c23cafd174acd26ca109246795484
52465 F20101115_AAAYOZ khanna_r_Page_172.pro
981352a9c695cd3d8ba3ce6e8e23cb48
36a4e8656179bd87312853477a84e2c206d708d1
56559 F20101115_AAAXML khanna_r_Page_018.pro
698fd2e31c5b6ec184fa228813c5bd6f
b17eba034a36bb9d0b745a0ea84a70bed9f6f0c5
417 F20101115_AAAXLX khanna_r_Page_022.txt
7d329a7e14230da99a7550a7e9e1b62c
c883c6e560c6cf294a1e8d3a58dd0f915e28445c
2234 F20101115_AAAYQC khanna_r_Page_035.txt
23c1b755bdaa4e44cfce8a1cca353382
cc75000fd0c2d086dbfbe3baa49f40f86949c36d
2294 F20101115_AAAYPO khanna_r_Page_018.txt
03771e928c2b950826466951214c2870
91ccb0e30ec463728206a11232385770e0f36927
1575 F20101115_AAAXMM khanna_r_Page_166thm.jpg
9a3300a46bade96596830815d09dff28
231de8f82c07837fc98ea6f6087fa29d0c20d8a9
2754 F20101115_AAAXLY khanna_r_Page_166.pro
8357d66bc6703c86fd4742edd69f774d
325489082114fd4d515187ae11a742a3f3fc98d8
2059 F20101115_AAAXNA khanna_r_Page_092.txt
8ca29cdd374f43bca50b03b2b49e4f86
96c9157c40be003df42ec4016684aa6433f7d8c1
2192 F20101115_AAAYQD khanna_r_Page_036.txt
14af9aa3b66a0eedc2c932f191f4ec98
8d137a06613c9b956f28afd21042e22c0664b2c2
2201 F20101115_AAAYPP khanna_r_Page_019.txt
4a9147c7abaaa35d6858042140d2b6dc
c79d8576d4989d88b9d86a96488fb475e81564cf
106894 F20101115_AAAXMN khanna_r_Page_077.jp2
be7bc8e2c8e00e940eddcd578894e3d1
d967b3e10792a530a44415d4df153af8a72526ad
2984 F20101115_AAAXLZ khanna_r_Page_011.txt
14821f6d5114406521a61f9ec227ec99
2e3178e1bf743d588abd2f77a3dd4036372714fa
114193 F20101115_AAAXNB khanna_r_Page_088.jp2
4b1bb365dc9c26b68003ed10c9140e68
cf6b6b55db8f63c13fff72fbe767965a7c2e8c1d
2080 F20101115_AAAYQE khanna_r_Page_037.txt
9453e61a7f0163219a909710f49a4b0b
513aa9b12dc80586b30b0d845fa42f25dab7fb19
2121 F20101115_AAAYPQ khanna_r_Page_020.txt
4e5ae7b5a03f71905250461330b13806
5dae5ab35a305cb04d861e2367fd1903d816bcb4
F20101115_AAAXMO khanna_r_Page_053.tif
54513db2b5cfaa6551c609432f895fc7
aa4d2b9b10ada0559d2c3cd4110eb5a2d7c493da
F20101115_AAAXNC khanna_r_Page_085.tif
6c792cb8de6f6e5f02f29b5095cfb6d8
5edd8e964634b7ca1f0dc05c4c8e3ee505d16b1a
2067 F20101115_AAAYQF khanna_r_Page_039.txt
ac1814adfe7ac58dc07fc2e99833d805
63c386f902a6680112017b305fb44b35ae371b56
2137 F20101115_AAAYPR khanna_r_Page_021.txt
5e58e9827066dae043d1290377aa4b8d
8ac755adf788298cd3267b06d6f4c8fc5e5d27a0
4377 F20101115_AAAXMP khanna_r_Page_009thm.jpg
5b2a94a69e524590534c444f8e62611e
d2db2908553e3a7a0d9129197f66011a73dbae23
6990 F20101115_AAAXND khanna_r_Page_152thm.jpg
d74b0b34af9c6f9eb5d4150ce9ab6676
9556aded389832c915f16fbb3369daf53487ab87
2048 F20101115_AAAYQG khanna_r_Page_040.txt
eea9b07108ba6cc23400db749cebc87a
510d28b35e798f8ffb619a912e15cfd8ddfebfae
1911 F20101115_AAAYPS khanna_r_Page_023.txt
8be7c471727651a2bcb5da3d582cc731
cc093e8e90c3eab16ea7f494fa47e1d9ddcdec80
39008 F20101115_AAAXMQ khanna_r_Page_042.jpg
6ef55b258ef82c72901ab06f6590d848
877e69795225312aea530e6ddf9514c7cb1a0f13
F20101115_AAAXNE khanna_r_Page_065.tif
a8eeb1711a729627bb964d34470476cd
1f75095e4a4d8b7751c7ce9ea31d06bb81d47431
1822 F20101115_AAAYQH khanna_r_Page_041.txt
ac99cf68b078b8ecef203ee6a7eba499
e82ecbde05109eb2e1166bf237a0d241314d6d50
2242 F20101115_AAAYPT khanna_r_Page_025.txt
2461e3298eb25257ad122e7ad2be5737
49a69fae8bedfb61cf10fd7ce3e7af038e1d4773
28744 F20101115_AAAXMR khanna_r_Page_105.jpg
837f8ad8aa4373f5bf92becb38433e77
12afe6095074211a5248fdba3596a89930924644
334 F20101115_AAAXNF khanna_r_Page_080.txt
c146f3736ff51192072be1f4aba11a36
3e7e53afedf4e291e265ee8a07230573b948cac3
869 F20101115_AAAYQI khanna_r_Page_042.txt
4e03782e56f28d5f080f38a392c95adf
89eb520c2a20ccd649124fb50266592f02ae5182
2090 F20101115_AAAYPU khanna_r_Page_026.txt
a9fcc7175c0aef77a66f4aae02dd4c8c
853c3c2ad8271d83554b33fca3bdfc67a409293a
F20101115_AAAXMS khanna_r_Page_025.tif
32547fd9f4543430825feb8374b0b367
8623fa6df8f4f78b717b1387bc45a8c4cda7029d
10035 F20101115_AAAXNG khanna_r_Page_160.QC.jpg
35de869ff41c1ec5fd97f7bbd36deaec
9051b961cf722e3c02998f68c0aec5764002b3b4
856 F20101115_AAAYQJ khanna_r_Page_043.txt
22bea482158775d58b4d3fedd85e8049
aad9b545b31e2be4a56e6200be4964d7fd0b8501
1927 F20101115_AAAYPV khanna_r_Page_027.txt
763bd923a73249e21ae0384dc04b938b
6f5cc4ed144967b9a1b7a0188d4d64b2384ba5ae
F20101115_AAAXMT khanna_r_Page_014.tif
c4cb7df5dbe9762afcf3305b2739930b
6ff652addac00604006d0793973ca45a336d923e
6703 F20101115_AAAXNH khanna_r_Page_080.pro
591df60c76189a5dee538070f15d194d
de5b707fd7bf4f62692df69102257e3a075a03e9
347 F20101115_AAAYQK khanna_r_Page_044.txt
9d6f1e597c03a7f1192468b8d101815b
b8744a185ec655ff1367218ea73c41f9a8027ac4
F20101115_AAAYPW khanna_r_Page_028.txt
ad4a71a663feb114792d1420a58b068b
69e710a1c104901463a7e3e877a80e23f4f69703
8813 F20101115_AAAXMU khanna_r_Page_102.pro
5adcd473ceb065444173ae089ed5c396
bbd6479f1ec26265814dbefb658188a5de97ee22
4212 F20101115_AAAXNI khanna_r_Page_116thm.jpg
b79244c771e1ecc964b38616a37f5aa9
2fb83592d6a38fb96b9eb029c8ba8c17a93f7543
337 F20101115_AAAYRA khanna_r_Page_064.txt
3cf931439e818e52fe305fb53f7e4848
64a8c812176f1e184db82120b10c221657544759
240 F20101115_AAAYQL khanna_r_Page_045.txt
764b0a2fefd5ed01a3cbf9e5a6e9ba44
9f57098466b57e8f9d83a6fc1828bdf8074d5fb1
1976 F20101115_AAAYPX khanna_r_Page_029.txt
ea772161eebcf77328687e99d86d075b
654a75c17cf552df897af2085ce89e756a358741
585720 F20101115_AAAXMV khanna_r_Page_064.jp2
d252afb3cb8d3cd76da5df2d25334913
7866f6c86cf14d78010b48119496094669bd781c
720 F20101115_AAAXNJ khanna_r_Page_082.txt
5aa8bfb3c8bd14d4ec76e3158307a14d
8985f55c39b8bed476a547da7cac6c66eea319d2
281 F20101115_AAAYQM khanna_r_Page_047.txt
c542d3ea1ad85905cfe2ec139371af7b
93491312efb00b7a22ea63077d5018c24c0792ef
2141 F20101115_AAAYPY khanna_r_Page_031.txt
cffeb3a9b5f18b36258d5a7f5350e4c6
6b80b08a980d2ce7edcb9492d514f7460cc23656
438230 F20101115_AAAXMW khanna_r_Page_052.jp2
2f8388aa8cb3740c5440bf9fad5521b4
e91e7bf4434010dd153cd202706f54e0c54e9adc
F20101115_AAAXNK khanna_r_Page_007.tif
4959491ebc5a540f752943462ef8a8c1
adb3a2ea30ca8c675091d08c7e872d94970cd132
1381 F20101115_AAAYRB khanna_r_Page_065.txt
0f8d7bca8c36ba97218a9e5a3066e190
e0ad6e967cf04fd0e93a414928e0514fb4573d3e
278 F20101115_AAAYQN khanna_r_Page_048.txt
9fe6a9455b76da0b56020bc09d477b91
018b56b0f24b2bd3bc277374fdd98a0ed964356b
1737 F20101115_AAAYPZ khanna_r_Page_032.txt
3bcc8a9e1843dc2e7437d6c8ebaf76d8
e6577db03727c6c0f752cdaec10183345b1af2fa
29174 F20101115_AAAXMX khanna_r_Page_097.jpg
73a932de45799122104e49a29aa24203
4d33c986fcdcd8914de79f3ec07a336f946d8af3
F20101115_AAAXNL khanna_r_Page_077.tif
78d904d0ff434c5baebc5c1761c2d3d1
7dbb83817c0a194fce098c1c8ee035450e24cb72
317 F20101115_AAAYRC khanna_r_Page_066.txt
84f1891c8bd5619136a68c976b2381cf
f4e93412d57707d6d1574afdb508341e2fd70a88
688 F20101115_AAAYQO khanna_r_Page_049.txt
2299a46fd6206c89f79fd49e3e7f69a7
eb5267e010200f6a22309ccd2a738bd6d91cd811
19835 F20101115_AAAXMY khanna_r_Page_153.QC.jpg
9c129d54962f38e7c0b91410f38dc140
972ddbf2c4cd4b5381845e1c3e8109fbb7121457
2074 F20101115_AAAXOA khanna_r_Page_076.txt
16f4cc9fa14b42d2f80a47ae11227dcd
e8d26a4307ac99fd7c26649a9502740a9c3c6c4a
33593 F20101115_AAAXNM khanna_r_Page_079.jpg
80bce41b5989e95c4b9ff3bd8dd797f0
cce753ea95925299dd8c0e41e1a97a09a0a2aba4
657 F20101115_AAAYRD khanna_r_Page_068.txt
6ea4be771e0eea8f3532a8513889dc44
d2ffe459cb7ec82e645bdd37b79f313ce042a2cf
776 F20101115_AAAYQP khanna_r_Page_050.txt
1c83214809a989ec01b08a6463ff5c92
1d567d8d515a9b92a30ddb40b6af2b7425a070c9
70212 F20101115_AAAXMZ khanna_r_Page_016.jpg
883bd69a6a2395ef89682c9fd85770d6
cfc22cc4cefa13b4ee73f72600292d1b0ce22808
298505 F20101115_AAAXOB khanna_r_Page_099.jp2
28fc17f59f0f1a543f698e2a97049597
37f83e8c7090cc63fa2aa84d4b1a52e80ee8017b
2136 F20101115_AAAXNN khanna_r_Page_024.txt
37dc4e5b54a4f70ee031382abf011f99
624fffa746547b0a3a3c907ded1e7969601f1ec4
343 F20101115_AAAYRE khanna_r_Page_069.txt
e2340cc5ab2e3c75b8517837540d2eae
f695e159a9cc8458b376aa14063d755b4e673dfb
155 F20101115_AAAYQQ khanna_r_Page_051.txt
03fd66b054c18baec6639af05718b008
239521e22e8df7363bcc51c7a0dd4b6e2473ae27
53313 F20101115_AAAXOC khanna_r_Page_009.jpg
6395a8852056f4a46583664350095d49
0b16ed0e3364b8cc3f98d5156c6a12cb4cc4e587
3609 F20101115_AAAXNO khanna_r_Page_104thm.jpg
12baa7270b5ca419745ee3d84ab9d2a6
d7787d322b8a0691f7823efdbd62aba4b089e90d
178 F20101115_AAAYRF khanna_r_Page_070.txt
42a0298c74d1f945b2afb47f84c466aa
5b2f6118d3eb933e39ea528d950a9fb35dd59af3
F20101115_AAAYQR khanna_r_Page_052.txt
8fbe09d3c924d480505d78dd008ef2e9
81a8f46219151f38769498ee06e15a209f907799
34660 F20101115_AAAXOD khanna_r_Page_080.jp2
1a28915a5b11b3c73a39aa522104a0ac
c97c4ed1d160d903ae13d0c08bbc24e089483654
23754 F20101115_AAAXNP khanna_r_Page_136.QC.jpg
c2a63d81fa2982f5f322685d7803ab71
43c2b207ee9c443e7cda1be829fcc4d9dd5c393f
607 F20101115_AAAYRG khanna_r_Page_072.txt
b63c239986b11215fd955ee7c6299a7c
4c96641fc5c039bd842b9dc3441558809fb07faf
2235 F20101115_AAAYQS khanna_r_Page_053.txt
9dd82a27d1acbcd8e8f1d051adfaaf20
f71413bf7e166e2afec03e598e45eea0e6f034d9
9816 F20101115_AAAXOE khanna_r_Page_002.jpg
5b1e996034f1d45746152a83febd1e7a
c9b3973a0b57b79156ff99d492092b23533fab43
972 F20101115_AAAXNQ khanna_r_Page_017.txt
6ca6335d17e2b3cbdfd02e66e7ff0b3c
0c0ae517e02dd133c5a4ce23e61a8af17cc776cc
2152 F20101115_AAAYRH khanna_r_Page_073.txt
1913664efb4af8bf970a9a2a158ca6cb
508eff23a279250d6927de182beeba778e1156fe
2091 F20101115_AAAYQT khanna_r_Page_054.txt
ed0f8f7a7d966aaa7409161b3a5e3071
b6ef14744cffaedaae08cd6965dee62813de7680
103756 F20101115_AAAXOF khanna_r_Page_169.jp2
f8d52ebf83ce88dd478e6a80fad47d50
2fe5a40615a77bc04c20f93a917e86cb0be445b3
4433 F20101115_AAAXNR khanna_r_Page_140.pro
9ad05500017e4f26ee8d4b35a4b3f667
ce12ea26ccc4c2482eccb67f4c1e719b07b6fadb
1944 F20101115_AAAYRI khanna_r_Page_074.txt
7fe8340dd3c5509968e2d837bc4a329d
b457d528bf4ad089f08ffa45a9ebe800212e89be
2146 F20101115_AAAYQU khanna_r_Page_055.txt
eb8386b9ed7b9094405ffdb9a21d7a90
daf24a6c1d99804ff76f42fda62c4892ba18667b
19829 F20101115_AAAXOG khanna_r_Page_174.jp2
ca8f1ac74cb8d1bf5c3007cc29ed26a6
a46baf7941f53b0f30ec0346446cae1fa34ff4df
6245 F20101115_AAAXNS khanna_r_Page_117.pro
b7ed1532c4ea2b545dd8c305da5abb3e
5a1063e6dde8e74cc08b128e09d67d890587d6ea
2145 F20101115_AAAYRJ khanna_r_Page_075.txt
d3b62ae69fa988acb201b6acb7d20361
0d7ed1b6c52115bf36a401f5eba328af3254f880
F20101115_AAAYQV khanna_r_Page_056.txt
75cd93b2ff33029e3b6c3294b5cb64f8
9da51f0b3a1085843ee7ecf1042d24a72ddcaa6b
6933 F20101115_AAAXOH khanna_r_Page_168thm.jpg
a8847a8ef630b39613cbc1a26c86d1f9
8cea57ccd0eaf8a8c69830171aae9df12c4a98fd
1966 F20101115_AAAXNT khanna_r_Page_173.txt
6a0c458ca4361d86fa2cc2ec5b1385e2
e866988e3c99179c36e092ceddb55fc5efe79991
1970 F20101115_AAAYRK khanna_r_Page_077.txt
1a34789600f0454359f6a57ac0cf8504
b94432f64b591d16c1033ac37b1bacc4344229dc
2195 F20101115_AAAYQW khanna_r_Page_057.txt
13874dc936602a7a74af48cddd22b8b5
7b6583ae3a5e540a7817faeab88ba4cdf0ef622a
468 F20101115_AAAXOI khanna_r_Page_127.txt
327eef040209c5548dd1eddc02ff457a
747e1a22b5f15b0d0ee2045ef678aab0a37f17e7
5178 F20101115_AAAXNU khanna_r_Page_044.pro
2571f7766835061c463dedc5e518ba45
ed728e8386dfc62ea969edc1b1020657f30c0d9f
405 F20101115_AAAYSA khanna_r_Page_098.txt
65d00bf8e4ee058befb7bdb9ca25f307
b5e62103cd8ff0bb560fd14d1d97110b34f3a9e7
948 F20101115_AAAYRL khanna_r_Page_078.txt
8f24067802c8b3d5c6536ff475097bf2
f97db23c5f5e202957a93ccd73c20afebadbd70d
2019 F20101115_AAAYQX khanna_r_Page_059.txt
57f94f61e6b30f44842c380b2a1e2fc7
5c34fedda06d0933dc4114c2dd9412384df945aa
4369 F20101115_AAAXOJ khanna_r_Page_049thm.jpg
d7ceb8eb6f0eea0a770bba81caa85adf
2b822add967dd26e81402b7eea147650c0aa3bc6
2838 F20101115_AAAXNV khanna_r_Page_061thm.jpg
4bccb22e4ee291e25a6cb9e055da3ea3
4635f63b813d488c590f3e75ad2d5f316d5823b3
444 F20101115_AAAYSB khanna_r_Page_099.txt
274a49d3e931f64651536fa514e031de
5912b8460e7d33c896bf5eb26d60dd147966a338
463 F20101115_AAAYRM khanna_r_Page_079.txt
86fc55a60f248cc0ac60224944510e2d
b80635b55c301106362452debc5d6a5a038ef78b
1720 F20101115_AAAYQY khanna_r_Page_060.txt
6317cae465f31810a7c86cb8019ef19b
b47fc0ff22424281af5b1dbda52aac35f3a9c64f
53498 F20101115_AAAXOK khanna_r_Page_075.pro
e2a1c66ebd752391cc377e0b1e286578
58c4a61f82d14d7a71b17488cd48127af29446dc
3046 F20101115_AAAXNW khanna_r_Page_002.QC.jpg
f109a2d4a411c75ffaf49e6c2d8df47e
5ac3baefdee64598dd635e48ddd43127fa779128
387 F20101115_AAAYRN khanna_r_Page_081.txt
bdc6ebb6c635a7b4962ff91d189f8428
e95d79279a984ced8e8ae0491a4870f59c581377
663 F20101115_AAAYQZ khanna_r_Page_062.txt
44b254d83caec6ad12e3ed816c93c345
a3ed274a34722708ad2d3cc63628f3cc44a3b576
27220 F20101115_AAAXOL khanna_r_Page_159.pro
cd583df43d76011c84cb2e2f31aaf841
cb62eb097efa38da03cd905eacc5496abf0e4859
3626 F20101115_AAAXNX khanna_r_Page_124thm.jpg
33a88b335139df22268ed3aa3b4a39bb
39218ec36a1b6277c4016af53de8926a81babdc4
648 F20101115_AAAYSC khanna_r_Page_100.txt
ef1a4e4d05c82fea83e8165086ddca33
9232dc43fcc1e6c9d8f195d7de40b09ce2e752c9
974 F20101115_AAAYRO khanna_r_Page_083.txt
b19bdab05eb87ff06baac9f0f59ba035
423f664b602a719d61934335d2d689ff35e72e7b
3671 F20101115_AAAXOM khanna_r_Page_080thm.jpg
876f3241eebaaf146ac40121ff9a7808
6292048bafe91779aef214075f8d38f458b38f05
55583 F20101115_AAAXNY khanna_r_Page_159.jpg
ff1eaaf67f28e00a2eeeef8ee0a3409e
a79a9933ab4d786b1b3da930d247f74015521808
4030 F20101115_AAAXPA khanna_r_Page_103thm.jpg
78ad79a3776006b37c6b286b695350db
4e8b1517335d84ca8677ed81eac346915b81bff6
467 F20101115_AAAYSD khanna_r_Page_101.txt
246ae0bd175c4ca76e3bf697e64afb1b
e02272dbe8ce7f73113875846257aaee9dedde8f
928 F20101115_AAAYRP khanna_r_Page_084.txt
d816f586558857aea9f69697ea3c9873
bb9bd2b1cf96aa0d1afc378ca15ec6454600460d
26047 F20101115_AAAXON khanna_r_Page_152.QC.jpg
c73aa22f62a5cacc1463080cc728214d
28a0796e51bba20900a4458096bb634e69a83416
75977 F20101115_AAAXNZ khanna_r_Page_087.jpg
207cbb2aaa2d7db7e89e15060ba6e2da
0171fdc15a18aa81a8df164aa679ea772ba1defe
24867 F20101115_AAAXPB khanna_r_Page_162.QC.jpg
954ddf8b84d3eb92f8e0829f20bf26af
994441d12a0c83bb6b0250d757433cad79d6afc7
773 F20101115_AAAYSE khanna_r_Page_103.txt
40bfa96e3bd75c427e0ac4d878c6ad31
feb19abb577a4556320e55acf405fc50b4d55820
979 F20101115_AAAYRQ khanna_r_Page_085.txt
f7a8e61bc3bfd7365fd38271aa525991
1513b5924d42a541798359885e7c45fb9103d636
F20101115_AAAXOO khanna_r_Page_133.tif
0ec7f8238c2ab88d2a47085e54d4b9be
dd9bc148deeae0a0a6244c58270ac840db703bbb
76817 F20101115_AAAXPC khanna_r_Page_056.jpg
1e5317eb6505699ed3341fd75e1da20b
8963b952396e9df8bc8596eae0b1b962da501ab0
F20101115_AAAYSF khanna_r_Page_106.txt
492f51cb0f954c3f780e9f1e0a4299b9
bbf5a9f3d961b928276a449cb8b532625faddf2b
2105 F20101115_AAAYRR khanna_r_Page_087.txt
962fe406bc7f8c4aa36f6b17a41cf19a
ab1be7d574b89a1a507d4923901f7f36bc1e884e
628284 F20101115_AAAXOP khanna_r_Page_049.jp2
c43f50dbe3e0b6f1fd8ea649921ef981
0bed903c682b49ec2521f812132cfa23e0c2c4c5
24131 F20101115_AAAXPD khanna_r_Page_040.QC.jpg
30a9ef18debb4b9763b9e555c308172e
dbc3ba859ee2ac48eb0d4d0ec6b2d34b22d77d19
2098 F20101115_AAAYSG khanna_r_Page_107.txt
5ac8ab6eb2ac067046869287ea691716
ce69c7405627f8cd82338d5a86351cb631846c2f
2113 F20101115_AAAYRS khanna_r_Page_088.txt
89128fd5bcbb39896806f50751ebc3b6
bded7f82236594f9b5252d3206b0af3a83da926b
926 F20101115_AAAXOQ khanna_r_Page_046.txt
02f538461238e7d1e03e5856742772c0
4c72d36f70a6e102f9ab485d043b6c1d0a112914
45760 F20101115_AAAXPE khanna_r_Page_175.jpg
0f7bf9071d6430fd2b8fc7752a583678
7cb518dd928db30261a002196c82b1ae5b2e6c3b
2051 F20101115_AAAYSH khanna_r_Page_108.txt
722fff31f5589675799a79ce52ce59ea
549a221d8af5b88de9a7271164fcd7eb8ef08564
1917 F20101115_AAAYRT khanna_r_Page_089.txt
00dace0121c4da13f2623d18619ddb1a
30bb96065a3eebda922dbdc35f09a7987d5075c1
76257 F20101115_AAAXOR khanna_r_Page_163.jpg
01e4cb4edd4510f68c2f8ef713e91c0c
12152b5b3cf6c69a12477a71c7940c1cff19573f
26275 F20101115_AAAXPF khanna_r_Page_004.QC.jpg
afe7fe83dd22aa12d0a476f3bdf5d557
f17f81fb595aca83ad467e7275418580dd94027a
2163 F20101115_AAAYSI khanna_r_Page_109.txt
238c5d2029ae4929a091a8bc3c435b0b
92f9931a7a9bdb235825aa3d70adc5dfb7e05983
1852 F20101115_AAAYRU khanna_r_Page_090.txt
1869dfe113c0367a91858ec5b2b49eb0
e1d35530df39f52bb594ee9a26a712ab1f3e1a48
1051981 F20101115_AAAXOS khanna_r_Page_006.jp2
c31001896f18ea00d9e14ddda4d3c11c
c005ef0a63226aea86ad372589eafa8979c9ef9f
7179 F20101115_AAAXPG khanna_r_Page_004thm.jpg
397a7f4e3ca1d977d22dd5c76d1b3647
d637c3cc39172760d92b5777c3da5c54c9265979
2185 F20101115_AAAYSJ khanna_r_Page_111.txt
c145ad32f9ab55cc13e2ec5830e5eb60
6f083cff5e26ab4cc01cce44e5f6938c62775739
2116 F20101115_AAAYRV khanna_r_Page_091.txt
78b34873969de79bb48e59ac86c14aab
4c30764925a1fd9f0fee58ba9b89ed265d26a8b3
53308 F20101115_AAAXOT khanna_r_Page_138.pro
45d86f1a9d9639129a56a4132793574b
6d78f4b8916448afb9c8842ee92a51a627c0871f
39755 F20101115_AAAXPH khanna_r_Page_017.jpg
002c77f672b3b39d929221f555660e0e
3a6b64ce754a5a79429500bd9cc4763f9f586118
1953 F20101115_AAAYSK khanna_r_Page_112.txt
862c3e1284367c359f87b99aee651cd2
b5a6cf96dca5aa9915b2f21ef1c873808865560a
2000 F20101115_AAAYRW khanna_r_Page_093.txt
4642a81e2dd7ddc37ef8b7e55e365218
c903bb36c346e2176dffa806075bc3b77eb5d507
81636 F20101115_AAAXOU khanna_r_Page_164.jpg
0a31078140a9fde3f6a1a4b3778044ad
ca9ea560275bb032e99bfb9c06b41da46c016391
94928 F20101115_AAAXPI khanna_r_Page_060.jp2
f6e9c755fb27d48e12d1139c77994f81
7c48260b640794f67a3c1fa9ed42f61b4bacbd95
542 F20101115_AAAYTA khanna_r_Page_130.txt
983d28191c6ec1f2c797fd4824273cfb
fa4fcbca9560033b76106548ba1dca6825a36c4f
2008 F20101115_AAAYSL khanna_r_Page_113.txt
c32d0c3f76fcd471fe6f85e8d82df8c1
7028dd92d14c05207a14b6ab05bbd333ff550e35
284 F20101115_AAAYRX khanna_r_Page_094.txt
83bc8a44759ff5e6577ef1e7199405d3
619a869593eb53ac56512826ecf346972d0a9e8b
F20101115_AAAXOV khanna_r_Page_148.tif
e10874013b5931d863e140ccbaad2007
695a593941232a563c2898ef43cd3880d8cc9c22
66496 F20101115_AAAXPJ khanna_r_Page_072.jp2
ae5c97dfa1ef94657553c6a848b302d5
6e8531ee77cfc4992dada8b1f939bca7da6db075
651 F20101115_AAAYTB khanna_r_Page_134.txt
a003ffe7acc72bb866e1d34ff5f63688
3ad8c064c463affad03dd4e17cb4b7f10593b04a
2099 F20101115_AAAYSM khanna_r_Page_115.txt
1147900033aeeb76c804370d9672fc67
1fef664b250ab1ce5bb69134a46b001a5e949326
286 F20101115_AAAYRY khanna_r_Page_095.txt
161c29252fc509d4762ed3d20f0967c6
ed24466a24fecc9f433ea197333f72754fe160cc
17524 F20101115_AAAXOW khanna_r_Page_051.jpg
1973d3d23a299f28da096e9abfda1a94
18063207f088ef1b91eb99207092879456de545b
333 F20101115_AAAXPK khanna_r_Page_132.txt
32fbcb44caefee88032c2557d7f7f9c4
45c37cd2b9c009fc24a71c682fd4668e55472ff1
2236 F20101115_AAAYTC khanna_r_Page_135.txt
2dd3bafe3178d55225b1c7a5b17d7926
ed76140f99df7914d1d0c3878a0d3fddc9efb045
1185 F20101115_AAAYSN khanna_r_Page_116.txt
47b9d175f4db8eb0ed655af24c772b7e
6dee8eaaac65962727e45a246f04cce1890ddc10
772 F20101115_AAAYRZ khanna_r_Page_096.txt
92a9d785c5dbe87a8cd1e3c932821772
6baf364c921849027cedace0257dc1eafbc5a2d5
77 F20101115_AAAXOX khanna_r_Page_003.txt
c0fa3df3a92f681eb090f61ce2d4b584
69f9b7770cbb58247e5aa83212e07ac0fc26d502
10583 F20101115_AAAXQA khanna_r_Page_119.QC.jpg
5f033612876e1267a540ca95409e10fb
4776b54e19f0855785e118b7d2fb266dd7982929
5870 F20101115_AAAXPL khanna_r_Page_174.QC.jpg
73620e67b1d80bc053b8a2e786eb39d9
ba41588e2ea599664901dbd81ad5da4c1110cf98
F20101115_AAAYSO khanna_r_Page_117.txt
38f138fe974ced5f300a05916eb8f000
2c6c51907da457f195bcb3fc0240a32ccd2c3d8c
80133 F20101115_AAAXOY khanna_r_Page_018.jpg
24cd018c4013fe93d4565d1d38fdd7b4
0b0d2d7f64a34c322f771413e4bf1e37f183a5fb
76733 F20101115_AAAXPM khanna_r_Page_138.jpg
093a96acb89e4b54ad973469fe85019b
29a29e6f543f588e52c6f9e55dfd659794e37d42
F20101115_AAAYTD khanna_r_Page_136.txt
bbfff5bc8aa2a58a8f2d09e7902e41b6
580c7a3e0514da079a7d9b9d58f193a9a4029446
793 F20101115_AAAYSP khanna_r_Page_118.txt
71b6d1055e0991c18dfe1de0129def7f
40ef4a7b7429d2ef5fa5b7575181eb03e3445972
2036 F20101115_AAAXOZ khanna_r_Page_150.txt
59dc1e5929cb80034fb3b55c845cb0aa
7aa48386c9fd379b103a57d9c8117d9c4d2954bb
75143 F20101115_AAAXQB khanna_r_Page_149.jpg
c5822f6f34a94aa4dfb1177bf4192b98
0bc211c1d102c94c2ac8335c7680bb9d9930f8bc
3741 F20101115_AAAXPN khanna_r_Page_118thm.jpg
71a58e19de7c0522101781f3de3796c8
0ec725b28828adf581ba44dd572e07cafd242304
2110 F20101115_AAAYTE khanna_r_Page_137.txt
8051860491fc0e0b2dd83b02e4d4bfd4
3f326c14b6a72454c547e532c49a29888c472d26
493 F20101115_AAAYSQ khanna_r_Page_119.txt
6c77b949ae7144cd3f9440eed131526f
97c8d198b75c30271772f803a47d7da677dc5a3e
113713 F20101115_AAAXQC khanna_r_Page_039.jp2
af1c14f6abecae047769ae70b5ce090c
737ba86684cc8248d18104c0d7cfb79cbd23d35e
77244 F20101115_AAAXPO khanna_r_Page_020.jpg
35b617b6506250a87896a7d6ae6f9e0c
77f2ac67c6bd086ef4151c9bbafd5360485675f3
2103 F20101115_AAAYTF khanna_r_Page_138.txt
04df3fe6d26d8724fee42d6e248cd994
e6bb3d479f683c9bc41dbb104b964c26fa495741
813 F20101115_AAAYSR khanna_r_Page_120.txt
87aa594208f61898df510f5282b8db5c
91c6ad507c67c3d660b6867f140e6e9ef2bdf174
718 F20101115_AAAXQD khanna_r_Page_097.txt
74451d293b59d4ed8ea3a829840cd601
9abfc9ee1970f44c17fce10e75873c474bc4a365
F20101115_AAAXPP khanna_r_Page_171.jp2
ea10b2ab6556f0164d4ff89d83e53f99
947850611cf88e62bd9645785d6ed3c24c63812f
951 F20101115_AAAYTG khanna_r_Page_139.txt
6be44cedec6f66da59b8471db60c81f9
2bd7491c47747a02178b98e1ab91da9e80aea856
323 F20101115_AAAYSS khanna_r_Page_121.txt
c11ca9fe5b5130572589eb86b1d35eff
1b3bf3aeb1fd92149469d44ade26b9756d2753c5
F20101115_AAAXQE khanna_r_Page_055.tif
051b3603ade7b00c2853c7be61f85470
623ffd3bc66056327d6f70e561d37e9cea400fc5
2570 F20101115_AAAXPQ khanna_r_Page_045thm.jpg
115657cff4f77d3ada6e4454ac38b265
40f455c0d5a63e68ee3575f2a41482903bb3deeb
216 F20101115_AAAYTH khanna_r_Page_140.txt
04f2ab37cb81ad98511c301f5f32f40f
b6a15d6c1674c0bcc51fe00bd0cb066467e87892
505 F20101115_AAAYST khanna_r_Page_122.txt
15ab2f52c4467d5ef4308f6d11a7d393
2d07c96bdf4aceabae18ff27e0e9962d6dcb4956
F20101115_AAAXQF khanna_r_Page_083.tif
e1500473fd852a325839772ab8692a5d
7cb4f9099072330cc52db20515862b859d9f977f
49502 F20101115_AAAXPR khanna_r_Page_029.pro
b9ef572bd3f47b7f191d6ed583891e9e
5a4cc649da260ff5d51b983c0f0314d382438b02
452 F20101115_AAAYTI khanna_r_Page_141.txt
b32e655cc843e226f25f1ca0ca67ff0d
9d0b71b06bb92dd110f2b446b9c1ea17f10010cd
330 F20101115_AAAYSU khanna_r_Page_123.txt
15359903fab050dda770fce318099e65
e4cb350737354f5778cb72e9dcc3bf394a4cc1ca
F20101115_AAAXQG khanna_r_Page_004.tif
42c23717b7c429b0d56e4456eef6e951
793243636be4d3dcf4454394991d986778b79b61
35576 F20101115_AAAXPS khanna_r_Page_117.jp2
dd9232c895a440b4a5b9229d031a8ee7
905f8c67f59b6ee75f8485616930cdcd50b837c1
506 F20101115_AAAYTJ khanna_r_Page_142.txt
9250a27763cfe85dc4ec41fd7a9274f5
9fb2147f370166ff9e55f3a4514cbab2dec94037
253 F20101115_AAAYSV khanna_r_Page_124.txt
5fb4dc317d8ae3cf3de19029fc9bcc0a
6603d2dad84a722673c217f17945193751c29d6a
F20101115_AAAXQH khanna_r_Page_097.tif
cfd8c145aed03ed92a493cd549a0f6ec
36b977ab6e3370b5b432d564bf9b54313e167bc4
49282 F20101115_AAAXPT khanna_r_Page_136.pro
5c5e73d568c8d6fa41096386f53308d6
6da4adf1be67e7783affeacdbdb85e7671bc1046
470 F20101115_AAAYTK khanna_r_Page_144.txt
ac76be3627138fba83318003f3924882
b20687dc1cc24bc09a05d12dbb94e7a2a7633ea1
428 F20101115_AAAYSW khanna_r_Page_125.txt
4bf28733743600f26dd7a7444822b5a8
2d02a750c9fb50381deb76cde8bd7a68d8853f8f
52365 F20101115_AAAXQI khanna_r_Page_073.pro
eb16c785408cc52f2641a096239b049b
593a99e08a5dd941f39edb1a90a03acb3a9a252e
23445 F20101115_AAAXPU khanna_r_Page_155.jpg
f0e47c01daaebc30b9c4a072b605d6c4
5090b0967058e25cafb12f49cdccb3dbf7cab259
F20101115_AAAYUA khanna_r_Page_163.txt
ec67746172dc3a12e761cab062657527
8fafe71530d2abf1b54dfec3b19df2fc65405e97
482 F20101115_AAAYTL khanna_r_Page_145.txt
052516dc8124491cf02f61d846f60bfd
4cf1af0f458ef96437d84fc8639acff90e5260be
676 F20101115_AAAYSX khanna_r_Page_126.txt
8dc80afde3e513329d7ae726f02be9cd
8b4e6e05d244bc7073fd4ed8603fb8a75eb9d5ab
23663 F20101115_AAAXQJ khanna_r_Page_093.QC.jpg
b8fee3361245e8c66cf10b53cb950de5
f6f1608c18d1cc6e4f3d0a39fd895cb676998f9d
7438 F20101115_AAAXPV khanna_r_Page_052.pro
2b29ba8972ba7ef9722a31724a894a09
f5c5a6838ce7ebf9755a0861696945ff269f016a
2226 F20101115_AAAYUB khanna_r_Page_164.txt
7134e6112c2e21a16187c532cbeee7a5
4fe284654b26cb0457fea6fc676a6efe7f4eb5ae
750 F20101115_AAAYTM khanna_r_Page_146.txt
d471d9cf6bbf7e908e77822d7d2a6f08
c93b9c5133986a81f8cf4d31eb5e0020177cbcab
543 F20101115_AAAYSY khanna_r_Page_128.txt
dc15c0fe31f67d333a53b99c72be3ca7
35a39d33bd62039f9f9cb5dc088ca41f990bea72
373 F20101115_AAAXQK khanna_r_Page_061.txt
76d7ddd3492bc45e33e58e355b6edab7
7385d6855da13da1e459ca4e696cda76d01def80
42166 F20101115_AAAXPW khanna_r_Page_060.pro
1246c6e3f1f66c0ef2fd58c0dbffe872
0fb69e658bb2d67fbfd62352b3b3a11afe8939fb
2150 F20101115_AAAYUC khanna_r_Page_165.txt
b99d0a06e2e837396724fa6895e1d5c7
2196620dff8983028fecb9647c6f4dee21af7171
2211 F20101115_AAAYTN khanna_r_Page_148.txt
1b92cc192f182e3745f1f3d9be141e1b
ec1958bb5e618df7157c2924fab417f9a0129762
780 F20101115_AAAYSZ khanna_r_Page_129.txt
1c6f1e9c7af5d2e63731cc2fd6343708
62e1b6b055e5df82c3361be92655b08037617e58
6997 F20101115_AAAXRA khanna_r_Page_036thm.jpg
022142ff2e79e7713dbf225d9b931e49
f1b4de9572a6285f5fe5277380438a22b81de4cf
F20101115_AAAXQL khanna_r_Page_145.tif
97b5bc2b9da5f4bb671c5e97862e93f5
f27328d93cd7df0a634807b763a3f5bbe60ace11
F20101115_AAAXPX khanna_r_Page_159.tif
7b1140cf356f0bdf5f5462af7c2b5a0a
248a5252cd75d0682217119a1b5bae5fd006a1d0
114 F20101115_AAAYUD khanna_r_Page_166.txt
ed503ed6d69590c4aca19051cf27f07c
18e0c297329f9087a346d801517cc6637bb04e1b
2066 F20101115_AAAYTO khanna_r_Page_149.txt
4110828575a47f3615d2ebf88fc35882
797904df9bcf8d606acc22912a017124ce97ce82
F20101115_AAAXRB khanna_r_Page_047.tif
bfec119b9825f2a89da605df671dc3e2
0e056881ebe3a8468a2faaf05cd111e69b2bc0a8
48845 F20101115_AAAXQM khanna_r_Page_077.pro
df1687791f4dcd81cf9eb0552b8f435c
b75b5cfb14321e5f63449405d593eeadccb0129f
481 F20101115_AAAXPY khanna_r_Page_102.txt
e54b8581c1edc8fc9340b799dac78f92
0779c37bf192a6e91347f98ecb399e0af73b8576
1835 F20101115_AAAYTP khanna_r_Page_151.txt
4a2103ed013b9270e68b30f5cd3e847b
c9a0c8582850d0b378e5292733f3ebbfc163b8c1
3405 F20101115_AAAXQN khanna_r_Page_161thm.jpg
ec0cd8a8dd3f2ee8128bae80ca88e072
bf64ecbe6debd258f660ebab20bb07a1475c0452
4256 F20101115_AAAXPZ khanna_r_Page_175thm.jpg
1422b16176f198c77afe5d068304666f
3a4c1849b791eb7eddc7c4c759b828125818bdae
F20101115_AAAYUE khanna_r_Page_167.txt
c84ff03ecc0ff9281f8a78d08e0ea4c6
8bcfb4cc942438512ce57eef3c4f71268e8ceba8
1622 F20101115_AAAYTQ khanna_r_Page_153.txt
3b6aeb65973f926aca2ccb94704e90d3
784728b6295c3537f6d1c99dfd9326f73e7f0ae1
53033 F20101115_AAAXRC khanna_r_Page_026.pro
957bfc1151a5d16e0ef0c9587554124d
1f1d7159a0e69176b432a3c746bd4beb2644fccb
6887 F20101115_AAAXQO khanna_r_Page_020thm.jpg
953da925cb9638c78e95c2484eb82f1c
3b7ebdc6cb3536fc253a26c9908f88b3c2157db8
6279 F20101115_AAAZAA khanna_r_Page_028thm.jpg
60204a0fb7a9c12bc404833f9b534c9d
2ce7faf869ca842d1edd5e151d88e877b4c6855e
2130 F20101115_AAAYUF khanna_r_Page_168.txt
7724bb7cf3de18485057cf126e83fea4
0f872a15805b92309ea7a74f78068824707e638d
513 F20101115_AAAYTR khanna_r_Page_154.txt
6ed429c9c400e11c888b734bd1c880b0
236763d39d7f7bf0c0fd1a0c7f6290b53652e173
598 F20101115_AAAXRD khanna_r_Page_067.txt
ee21d96f4bc7974f0a1ded329ab7b73a
27e93f0bb8d8190db68d492c1fa5b647604ad190
26669 F20101115_AAAXQP khanna_r_Page_164.QC.jpg
e67c4038a1db6a6a15cf5a038388e911
cacaa00c367bcbce97fca70cbae127e3c5bdf629
24237 F20101115_AAAZAB khanna_r_Page_168.QC.jpg
8ef53a93eb1530377b2423bbd0ac8bdc
cfe580e39428ef25fb3b31e70a193238f67c4ab3
2082 F20101115_AAAYUG khanna_r_Page_169.txt
7b36c4e68ade037f4f2dccbc1b37e347
44ac63138e8a1102759c4f58ec19c3087446645f
392 F20101115_AAAYTS khanna_r_Page_155.txt
512af77a54a4755aaad554aebb8141a3
6d7140b7528aa3a8c7614ea6b794b304086d0226
75187 F20101115_AAAXRE khanna_r_Page_162.jpg
b158e481407dc713ec2eceb5a111babb
b411843908cfe528aeba2b4b7a3340f1a83528a4
54170 F20101115_AAAXQQ khanna_r_Page_109.pro
2cb66c8f513dac3aa9a0599eceb84215
8dc9bc1ee1a61447b2756cb4419a5cb64315a526
6509 F20101115_AAAZAC khanna_r_Page_113thm.jpg
624deef7fbd43e5239935faa1df6861c
d178b4ee9ffdb44fc04ffe526372cc083cd8f3ff
2200 F20101115_AAAYUH khanna_r_Page_170.txt
4116b35e4c2cc64491227f80e74d58a3
9e5259c69f311515d4111a945acf4e2c89bcf976
494 F20101115_AAAYTT khanna_r_Page_156.txt
5e790ef70bf68523d02f24d1210606a5
06c0a876736df0249d24cd6223bb8d2210b52e2d
2643 F20101115_AAAXRF khanna_r_Page_010.txt
a0ba58991df360276ef328c9d648e78e
1963361aa785a640cf899e3b0df8748db3e3f7a3
F20101115_AAAXQR khanna_r_Page_142.tif
50a882ead04b947800995e56d8499e82
fa0e64de44dcc2551ee24a810167bfecef93df0a
3800 F20101115_AAAZAD khanna_r_Page_100thm.jpg
9fce535c17ede4fdafc3afc26c99a963
ffa89018999dc074358f6bb21519a3418192bd64
F20101115_AAAYUI khanna_r_Page_171.txt
96f7f4dc6ef0ede6e75a8e08ff4f5536
1482225a3c6c58b38cf5f550d2d4099918df7cd3
F20101115_AAAYTU khanna_r_Page_157.txt
4ef34fec98c8f9b27f7bdb737391f2bd
8f545348d1d96dcc9b631fb3a2ec76d515cd76f0
2139 F20101115_AAAXRG khanna_r_Page_058.txt
c5a99c09776d9b0c564475c89671066e
116e10e28f252853ac36b0f2f8c0d1d9e3a19f44
1007 F20101115_AAAXQS khanna_r_Page_071.txt
c0557168f014ee86ee409db2c1b9a9e6
a725dbbd7214711a7a6329ebdb68935a9e8aa549
21559 F20101115_AAAZAE khanna_r_Page_041.QC.jpg
4b1eaabf90b5459a3c64c693e77fcbec
710563856b757f420378124a23c9dfbe3270eded
2089 F20101115_AAAYUJ khanna_r_Page_172.txt
d5a3783460a9504888dd4a71db12e5a5
05c8ded3cec4e88ed3d0155fb6fd13cd53eed48b
315 F20101115_AAAYTV khanna_r_Page_158.txt
fce0867f41778e1c1e2095e521e324b4
6804a4ee7f22457857eb676532f3aaae19bb13a0
F20101115_AAAXRH khanna_r_Page_080.tif
21f7d686fce5566a82c65eb0e81c5a11
85dbf7523a18a321822686931cc071fc99f1802d
77595 F20101115_AAAXQT khanna_r_Page_055.jpg
4610459f2f7ab66a1199532f70c0a473
ffaf0f9abc8baca7f324e7a6eeeb9d4266d8f5f1







DEVELOPMENT OF HIGH TEMPERATURE STABLE OHMIC AND SCHOTTKY
CONTACTS ON N-GaN





















By

ROHIT KHANNA


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

UNIVERSITY OF FLORIDA

2007



































O 2007 Rohit Khanna

































To my family









ACKNOWLEDGMENTS

I would like to thank my advisor, Prof. Stephen J. Pearton, the most important person

throughout my graduate studies, for all the opportunities, guidance, motivation, and support. I

would also like to thank my committee members, Prof. Cammy R. Abernathy, Prof. David P.

Norton, Prof. Fan Ren, and Prof. Rajiv Singh, for their time, expertise, and evaluation.

Prof Pearton has been my mentor in true sense. He provided me numerous opportunities to

give presentations, encouraged logical, problem solving thinking and gave me confidence

whenever I needed it. Prof. Pearton helped me develop wide ranging skills in semiconductor

processing area for which I am really thankful to him. I can not thank him enough. I thank Prof.

Ren for providing me with useful comments and directions in order to improve my research

work. I am grateful to them for their advice that has helped me grow professionally.

I would like to thank group members of Prof. Pearton, Prof. Ren, Prof. Abernathy, Prof

Norton, and Prof Singh research groups', Kwang Baik, Kelly Ip, Lars Voss, Jon Wright, Wantae

Lim, Rishabh Mehandru, Soohwan Jang, Byong Kang, Hung-Ta-Wang, Travis. J Anderson, J.J

Chen, Luc Stafford, Brent Gila, Seemant Rawal, Karthik Ramani, Mark Hlad and countless

others for their assistance and friendship and who have made graduate school enj oyable, learning

experience. Especially, I am grateful to have as a very good friend, Kao-Chil-Joe, who was a

visiting scholar form Taiwan and I thank him for helping me during my first year as graduate

student. I also like to thank University of Florida Nano-Fabrication Lab staff, Ivan and Bill as I

enjoyed working with them. I would also like to thank Paula, our group secretary for all the

support she has given me.

I also want to thank all my friends of school and college with whom I share special

memories forever in my life. I also want to thank my ex-roommates who made graduate life an

enjoyable experience, especially Shiv for having a lot of interesting and intellectual debates.









Most importantly, I express my deepest gratitude to my family especially to my parents (my

father, Prem Kumar Khanna and mother, Madhu Khanna) whose love and sacrifice for me is

beyond anything I will ever understand and my brother (Bhaskar) and sisters (Ela and Rashmi)

for loving and supporting me unconditionally. I thank my family for making me who I am today.












TABLE OF CONTENTS


page

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


LI ST OF T ABLE S ................. ...............9..._. .....


LIST OF FIGURES .............. ...............10....


AB S TRAC T ............._. .......... ..............._ 15...


CHAPTER


1 INTRODUCTION ................. ...............18......... .....


2 BACKGROUND AND LITERATURE REVIEW .............. ...............23....


Properties of GaN .............. ...............23....
Overview .............. ..... ....... ... .........2

Crystal Structure and Basic Properties ................ ...............23........... ...
M etal Contact .............. ...............25....
S chottky contact ................. ...............26.......... .....
Ohmic contact .............. ...............29....
Thermal stability .............. ...............32....
Common Processing Techniques ................. ...............33........... ....
Dry Plasma Etching ................. ...............33........... ....
Ion Implantation .............. ...............36....
Rapid Thermal Annealing .............. ...............37....
Characterization Techniques .............. ...............37....
Atomic Force Microscopy ................. ...............37................
Auger Electron Spectroscopy ................. ...............38................
X-ray Photoelectron Spectroscopy ................. ......... ......... .............
Electrical Measurements .............. ...............38....
Photoluminescence (PL)................ ............. ..........3
Rutherford Backscattering Spectrometry/Channeling .........._.. ......._.._ ...............39
Scanning Electron Microscopy ................. ...............40.___ ......
Secondary lon Mass Spectrometry ................. ...............40................
Stylus Profilometry ................. ...............41.__._.......

3 TUNGSTEN AND ZIRCONIUM BORIDE BASED OHMIC CONTACTS TO N-GaN....53


Introducti on ................. ...............53.................

Experimental ................. ...............54.................
Results and Discussion ............... .............. ............5

Tungten Boride Based Ohmic Contact ....._._.__ ..... ..__... ....___ ...........5
Zirconium Boride Based Ohmic Contacts ....._ .....___ .........__ ...........5
Summary and Conclusions .............. ...............60....












4 COMPARISON OF ELECTRICAL AND RELIABILITY PERFORMANCE OF TiB2,
CrB2 AND W2Bs BASED OHMIC CONTACTS ON N-GaN .............. .....................7

Introducti on ................. ...............73.................

Experimental ................. ...............74.......... ......
Results and Discussion .............. ...............75....
Summary and Conclusions .............. ...............77....


5 ZrB2 AND W2B SCHOTTKY DIODE CONTACTS ON N-GaN ................. ................ ..87


Introducti on ................. ...............87.................

Experim ental ................. ...............8.. 8..............
Results and Discussion ................... ...............90..
W2B Based Rectifying Contacts............... ...............90
ZrB2 Based Rectifying Contacts............... ...............91
Summary and Conclusions .............. ...............93....


6 ANNEALING TEMPERATURE DEPENDENCE OF TiB2 W2Bs AND CrB2
SCHOTTKY BARRIER CONTACTS ON N-GaN ................. ..............................106


Introducti on ................. ...............106................

Experimental ................. ...............107......... ......
Results and Discussion ................. ...............109..
TiB2 Based Schottky Contact ................. ...............109...............
W2B5 Based Schottky Contact ................. ...............111...............
CrB2 Based Schottky Contact ................. ...............113...............
Summary and Conclusions ................. ...............115...............


7 IMPROVED LONG-TERM THERMAL STABILITY AT 350 oC OF TiB2 BASED
OHMIC CONTACTS ON AlGaN/GaN HIGH ELECTRON MOBILITY ........................135


Introducti on ................. ...............135................

Experimental ................. ...............136......... ......
Results and Discussion .............. ...............137....
Summary and Conclusions .............. ...............139....

8 Ir BASED SCHOTTKY AND OHMIC CONTACTS ON N-GaN ................. ................. 148


Introducti on ................. ...............148................

Experimental ................. ...............149......... ......
Results and Discussion .............. ...............151....
S chottky Contacts ................. ................. 15......... 1....
Ohmic Contacts .............. ...............152....
Summary and Conclusions .............. ...............153....

9 CONCLUSIONS .............. ...............162....


LIST OF REFERENCES ........._.._.. ...._... ...............167...












BIOGRAPHICAL SKETCH ........._.... ....._.. ...............175.....










LIST OF TABLES


Table page

2-1 Electrical properties of Si, GaAs and GaN. .............. ...............41....

2-2 The physical parameters in different semiconductor materials .............. ....................42

2-3 Ionization energy of impurities for wurtzite GaN ................. ..............................42

2-4 Basic physical properties of GaN. ............. ...............43.....

2-5 Metal work function and ideal barrier heights for GaN (electron affinity: 4.1 eV)..........43

3-1 Near-surface composition of contact stack determined by AES measurements for
ZrB2 ohmic contact .............. ...............60....

4-1 Selected properties of potential boride contacts on GaN ................. ................ ...._.78

5-1 Near-surface composition data obtained from AES measurements. ............. .................94

6-1 Concentration of elements detected on the as-received surfaces of TiB2 based
schottky contacts (in Atom %"f) ................. ...............116....._._. ...

6-2 Concentration of elements detected on the as-received surfaces of W2B5 based
schottky contacts in Atom %"f) ................. ...............116........ ....

6-3 Concentration of elements detected on the as-received surfaces of CrB2 based
schottky contacts (in Atom %"f) ........._._.. ........... ...............116...










LIST OF FIGURES


Fiare page

2-1 Crystal structure of wurtzite GaN ................. ...............44...............

2-2 The III-V compound semiconductor tree ................. ...............44...............

2-3 Structure of a HEMT .............. ...............45....

2-4 Previous study of schottky contacts A) Index of interface behavior S as a function of
the electronegativity difference of the semiconductors. ............. .....................4

2-5 Lithography pattern for Schottky diode ................. ...............47...............

2-6 Lithography pattern for linear TLM A) TLM pads. B) Plot for measurement. .................47

2-7 An ICP reactor. ............. ...............48.....

2-8 Electric and magnetic fields inside the reactor. ............. ...............48.....

2-9 Chemical etching process. A) Generation of reactive species. B) Diffusion of
reactive neutral s to surface............... ...............49

2-10 Physical etching process. A) Generation of reactive species. B) Acceleration of ions
to the surface. C) Ions bombard surface .............. ...............49....

2-11 Combination of chemical and physical etching processA) Generation of reactive
species. .............. ...............50....

2-12 lon implantation system. .............. ...............50....

2-13 Simplified principle of AFM. ................ ................ ........ ......... ........ .51

2-14 Auger Process. A) An isolated atom. B) Inner core level electron dislodged. leaving
behind a vacancy. C) An outer level electron fills the vacancy ................. ................ ...52

3-1 Specific contact resistivity versus anneal temperature for Ti/Al/W2B/Ti/Au on n-
G aN .............. ...............61....

3-2 Measurement as a function of annealing temperature Ti/Al/W2B/Ti/Au on n-GaN. A)
Transfer resistance. B) Sheet resistance............... ...............6

3-3 Specific contact resistance versus measurement temperature for Ti/Al/W2B/Ti/Au on
n-GaN annealed at 800 OC. ............. ...............63.....

3-4 Secondary electron images of the Ti/Al/W2B/Ti/Au contacts on n-GaN. A) As-
deposited. B) Annealed at 500 OC. C) Annealed at 800 OC. D) Annealed at 1000 oC ......64










3-5 AES depth profies of the Ti/Al/W2B/Ti/Au on n-GaN. A) As-deposited. B)
Annealed at 500 OC. C). Annealed at 800 OC. D). Annealed at 1000 OC. ................... .......65

3-6 Contact resistance of the Ti/Al/W2B/Ti/Au on n-GaN, initially annealed at 800 OC,
as a function of subsequent time at 200 OC. .............. ...............66....

3-7 Measurement versus anneal temperature for Ti/Al/ZrB2/Ti/Au on n-GaN. A) Specific
contact resistivtiy. B) Sheet resistance............... ...............6

3-8 Measurement as a function of annealing time at 700 OC for Ti/Al/ZrB2/Ti/Au on n-
GaN. A) Specific contact resistivtiy. B) Sheet resistance............... ...............6

3-9 Specific contact resistance versus measurement temperature for Ti/Al/ZrB2/Ti/Au on
n-GaN annealed at 800 OC. ............. ...............69.....

3-10 Secondary electron images of the Ti/Al/ZrB2/Ti/Au on n-GaN. A) As-deposited. B)
Annealed at 500 OC. C) Annealed at 700 OC. D) Annealed at 1000 OC. ................... .........70

3-11 AES surface scans of the Ti/Al/ZrB2/Ti/Au on n-GaN. A) As-deposited. B)
Annealed at 500 OC. C) Annealed at 700 OC. D) Annealed at 1000 OC ................... ..........71

3-12 AES depth profies of the Ti/Al/ZrB2/Ti/Au on n-GaN. A) As-deposited. B)
Annealed at 500 OC. C) Annealed at 700 OC. D) Annealed at 1000 OC ................... ..........72

4-1 Specific contact resistance of Ti/Al/boride/Ti/Au Ohmic contacts on n-GaN as a
function of anneal temperature. ............. ...............79.....

4-2 Specific contact resistance of Ti/Al/boride/Ti/Au Ohmic contacts on n-GaN as a
function of ................. ...............80.................

4-3 Specific contact resistance of Ti/Al/boride/Ti/Au Ohmic contacts on n-GaN as a
function of measurement temperature at the optimum anneal temperatures. ................... .81

4-4 SEM micrographs. A) As-deposited Cr2B. B) As-deposited TiB2. C) As-deposited
W2Bs. D) Cr2B annealed at 800 OC............... ...............82...

4-5 AES depth profies of CrB2-based contacts. A) As-deposited. B) Annealed at 700 OC.
C) Annealed at 800 OC. D) Annealed at 1000 oC. ............. ...............83.....

4-6 AES depth profies of TiB2-based contacts. A) As-deposited. B) Annealed at 600 OC.
C) Annealed at 800 OC. D) Annealed at 1000 oC. ............. ...............84.....

4-7 AES depth profies of W2BS-based contacts. A) As-deposited. B) Annealed at 500
oC. C) Annealed at 700 OC. D) Annealed at 1000 oC. ................... ..............8

4-8 Specific contact resistance of the boride-based contacts annealed at 800 oC and the
conventional Ti/Al/Ni/Au contacts as a function of aging time at 350 oC. .......................86










5-1 SEM micrographs of W2B based schottky contacts. A) As-deposited. B) Annealed at
700 oC The inner circle is the W2B/Ti/Au while the outer ring is the Ohmic contact. ...95

5-2 Barrier height and reverse breakdown voltage as a function of measurement
temperature for as-deposited W2B/Ti/Au contacts on n-GaN. ............. .....................9

5-3 Barrier height and reverse breakdown voltage as a function of annealing temperature
for W2B/Ti/Au contacts on n-GaN. ............. ...............97.....

5-4 AES depth profiles of W2B/Ti/Au on GaN. A) Unannealed. B) After annealing at
700 oC ............. ...............98.....

5-5 I-V characteristics from ZrB2/GaN diodes as a function of post-deposition annealing
tem perature. ............. ...............99.....

5-6 Barrier height and reverse breakdown voltage as a function of annealing temperature
for ZrB2/Ti/Au contacts on n-GaN. ............. ...............100....

5-7 SEM micrographs of ZrB2 based schottky contacts. A) As-deposited. B) Annealed at
350 oC. C) Annealed at 700 OC .............. ...............101....

5-8 AES surface scans of ZrB2/Ti/Au on GaN. A) As-deposited. B) Annealed at 350 oC.
C) Annealed at 700 OC. ............. ...............102....

5-9 AES depth profiles of ZrB2/Ti/Au on GaN. A) As-deposited. B) Annealed at 350 oC.
C) Annealed at 700 OC. ............. ...............103....

5-10 Powder XRD spectrum from ZrB2 on GaN. A) Unannealed. B) After annealing at
800 oC ............. ...............104....

5-11 Glancing angle XRD spectra from ZrB2 on GaN. A) Unannealed. B) After annealing
at 800 oC. ............. ...............105....

6-1 I-V characteristics at 250 C of TiB2/Ti/Au on GaN as a function of post-deposition
annealing temperature ................. ...............117................

6-2 Barrier height and reverse breakdown voltage as a function of annealing temperature
for TiB2/Ti/Au contacts on n-GaN ................. ...............118........... ...

6-3 AES surface scans of TiB2/Ti/Au on GaN. A) As-deposited. B) Annealed at 350 oC.
C) Annealed at 700 OC. ................ ...............119........... ...

6-4 AES depth profiles of TiB2/Ti/Au on GaN. A) As-deposited. B) Annealed at 350 oC.
C) Annealed at 700 OC. ............. ...............120....

6-5 SEM micrographs of TiB2 based schottky contacts A) As-deposited. B) Annealed at
350 oC. C) Annealed at 700 OC .............. ...............121....










6-6 Barrier height and reverse breakdown voltage as a function of measurement
temperature for as-deposited TiB2/Ti/Au contacts on n-GaN ................. ................ ...122

6-7 SEM micrographs of W2B5 based schottky contacts. A) As-deposited. B) Annealed
at 350 oC. ............. ...............123....

6-8 AES depth profiles of W2BS/Ti/Au on GaN. A) As-deposited. B) Annealed at 350
oC. C) Annealed at 700 OC............... ...............124..

6-9 I-V characteristics of W2BS/Ti/Au on GaN as a function of post-deposition annealing
tem perature. ............. ...............125....

6-10 Barrier height and reverse breakdown voltage as a function of annealing temperature
for W2B5/Ti/Au contacts on n-GaN. ............. ...............126....

6-11 I-V characteristics of as-deposited W2BS/Ti/Au on GaN as a function of
measurement temperature. ............. ...............127....

6-12 Barrier height and reverse breakdown voltage as a function of measurement
temperature for as-deposited W2BS/Ti/Au contacts on n-GaN. ................... ...............12

6-13 I-V characteristics of CrB2/Ti/Au on GaN as a function of post-deposition annealing
tem perature. ............. ...............129....

6-14 Barrier height and reverse breakdown voltage as a function of annealing temperature
for CrB2/Ti/Au contacts on n-GaN. ............. ...............130....

6-15 AES depth profiles of CrB2/Ti/Au on GaN. A) As-deposited. B) Annealed at 350 oC.
C) Annealed at 700 OC. ............. ...............131....

6-16 AES surface scans of CrB2/Ti/Au on GaN. A) As-deposited. B) Annealed at 350 oC.
C) Annealed at 700 OC. ............. ...............132....

6-17 SEM micrographs of CrB2 based schottky contacts. A) As-deposited. B) Annealed at
350 oC. C) Annealed at 700 OC .............. ...............133....

6-18 Barrier height and reverse breakdown voltage as a function of measurement
temperature for as-deposited CrB2/Ti/Au contacts on n-GaN. ............. ....................134

7-1 HEMT layout used in these experiments. ............. ...............140....

7-2 Study of Raman spectra from Ti/Al/TiB2/Ti/Au contacts on HEMT wafer. A) Optical
micrograph. B) Raman spectra. ............. ...............141....

7-3 IDs-VDs characteristics from HEMT with conventional Pt/Au gate contacts and
Ti/Al/Pt/Au source/drain contacts before and after aging at 350 oC for 25 days. ...........142

7-4 IDs-VDs characteristics from HEMT with Ti/Al/TiB2/Ti/Au source/drain contacts. A)
Pt/Au gate contacts before and after aging at 350 oC for 25 days. ................ ...............143










7-5 IDs-VDs characteristics from HEMT with Ti/Al/TiB2/Ti/Au source/drain contacts. A)
Ni/Au gate contacts before and after aging at 350 oC for 25 days. ............. .................144

7-6 Optical microscopy images of HEMTs. A) Conventional contacts before aging. B)
Boride-based source/drain contacts before aging. ............. ...............145....

7-7 Percent change in saturated drain/source current from HEMTs with different
combinations of contact metal schemes as a function of aging time at 350 oC. .............146

7-8 RF performance of 1.5 x 200 Clm2 gate length HEMTs. A) HEMT with conventional
metal contacts prior to aging. ..........__...... .__ ...............147.

8-1 I-V characteristics from Ir/Au Schottky contacts on n-GaN. ............. .....................15

8-2 Schottky barrier height for Ir/Au contacts on n-GaN as a function of annealing
tem perature. ............. ...............155....

8-3 AES depth profies. A) Ir/Au after annealing at 350 oC. B) Ir/Au after annealing at
700 oC. C) Ni/Au contacts after annealing at 350 oC. ............. ...............156....

8-4 Specific contact resistance of Ti/Al//Ni/Au and Ti/Al/Ir/Au Ohmic contacts on n-
GaN as a function of anneal temperature ................. ...............157........... ..

8-5 SEM images Ir and Ni based ohmic. A) Ti/Al/Ir/Au after annealing at 500 OC. B)
Ti/Al/Ni/Au after annealing at 500 OC............... ...............158..

8-6 AES depth profies of Ir and Ni based ohmic. A) Ti/Al/Ir/Au after annealing at 500
oC. B) Ti/Al/Ni/Au after annealing at 500 OC. ............. ...............159....

8-7 Specific contact resistance of Ti/Al//Ni/Au and Ti/Al/Ir/Au Ohmic contacts on n-
GaN as a function of measurement temperature ................. ..............................16

8-8 Specific contact resistance of the Ti/Al//Ni/Au and Ti/Al/Ir/Au contacts annealed at
9000C as a function of aging time at 350 oC. ........................... ........161









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

DEVELOPMENT OF HIGH TEMPERATURE STABLE OHMIC AND SCHOTTKY
CONTACTS ON N-GaN

By

Rohit Khanna

August 2007

Chair: Stephen J. Pearton
Major: Materials Science and Engineering

In this work the effort was made to towards develop and investigate high temperature

stable Ohmic and Schottky contacts for n type GaN. Various borides and refractory materials

were incorporated in metallization scheme to best attain the desired effect of minimal

degradation of contacts when placed at high temperatures.

This work focuses on achieving a contact scheme using different borides which include

two Tungsten Borides (namely W2B, W2Bs), Titanium Boride (TiB2), Chromium Boride (CrB2)

and Zirconium Boride (ZrB2). Further a high temperature metal namely Iridium (Ir) was

evaluated as a potential contact to n-GaN, as part of continuing improved device technology

development. The main goal of this proj ect was to investigate the most promi sing boride-based

contact metallurgies on GaN, and finally to fabricate a High Electron Mobility Transistor

(HEMT) and compare its reliability to a HEMT using present technology contact.

Ohmic contacts were fabricated on n GaN using borides in the metallization scheme of

Ti/Al/boride/Ti/Au. The characterization of the contacts was done using current-voltage

measurements, scanning electron microscopy (SEM) and Auger Electron Spectroscopy (AES)

measurements. The contacts formed gave specific contact resistance of the order of 10-5 to 10-6

Ohm-cm2. A minimum contact resistance of 1.5x10-6 QZ.cm2 WaS achieved for the TiB2 based









scheme at an annealing temperature of 850-900 oC, which was comparable to a regular ohmic

contact of Ti/Al/Ni/Au on n GaN. When some of borides contacts were placed on a hot plate or

in hot oven for temperature ranging from 200 oC to 350 oC, the regular metallization contacts

degraded before than borides ones. Even with a certain amount of intermixing of the

metallization scheme the boride contacts showed minimal roughening and smoother

morphology, which, in terms of edge acuity, is crucial for very small gate devices.

Schottky contacts were also fabricated and characterized using all the five boride

compounds. The barrier height obtained on n GaN was ~0-5-0.6 eV which was low compared to

those obtained by Pt or Ni. This barrier height is too low for use as a gate contact and they can

only have limited use, perhaps, in gas sensors where large leakage current can be tolerated in

exchange for better thermal reliability.

AlGaN/GaN High Electron Mobility Transistors (HEMTs) were fabricated with

Ti/Al/TiB2/Ti/Au source/drain ohmic contacts and a variety of gate metal schemes (Pt/Au,

Ni/Au, Pt/TiB2/Au or Ni/TiB2/Au) and were subj ected to long-term annealing at 3 50 OC. By

comparison with companion devices with conventional Ti/Al/Pt/Au ohmic contacts and Pt/Au

gate contacts, the HEMTs with boride-based ohmic metal and either Pt/Au, Ni/Au or Ni/TiB2/Au

gate metal showed superior stability of both source-drain current and transconductance after 25

days aging at 350 oC.

The need for sputter deposition of the borides causes' problem in achieving significantly

lower specific contact resistance than with conventional schemes deposited using e-beam

evaporation. The borides also seem to be, in general, good getters for oxygen leading to sheet

resistivity issues.









Ir/Au schottky contacts and Ti/Al/Ir/Au ohmic contacts on n-type GaN were investigated

as a function of annealing temperature and compared to their more common Ni-based

counterparts. The Ir/Au ohmic contacts on n-type GaN with n~ 1017 cm-3 exhibited barrier

heights of 0.55 eV after annealing at 700 oC and displayed less intermixing of the contact metals

compared to Ni/Au. A minimum specific contact resistance of 1.6 x10-6 QZ.cm2 WaS obtained for

the ohmic contacts on n-type GaN with n~10"s cm-3 after annealing at 900 oC. The measurement

temperature dependence of contact resistance was similar for both Ti/Al/Ir/Au and Ti/Al/Ni/Au,

suggesting the same transport mechanism was present in both types of contacts. The Ir-based

ohmic contacts displayed superior thermal aging characteristics at 350 oC. Auger Electron

Spectroscopy showed that Ir is a superior diffusion barrier at these moderate temperatures than

Ni.









CHAPTER 1
INTRODUCTION

The microelectronics industry has grown rapidly in the past four decades and now is the

basis for our Information Age. The first semiconductor transistor was invented by the scientists

of Bell Labs in 1947. Subsequently, the concept of an Integrated Circuit (IC) was developed,

requiring a high yield of working devices that comprise the circuit. To have a 50% probability of

functionality for a 20 transistor circuit, the probability of device functionality must be (0.5)1/20

0.966 or 96.6%. This was considered wildly optimistic at the time, yet today integrated circuits

are built with billions of transistor 1. This is possible because each component or a device is

many times reliable compared to a component in any other industry. Even though the very first

semiconductor transistor was made from germanium (Ge), silicon (Si) became the semiconductor

of choice as a result of the low melting point of Ge that limits high temperature processes and the

lack of a natural occurring germanium oxide to prevent the surface from electrical leakage.

Due to the maturity of its fabrication technology, silicon continues to dominate the present

commercial market in discrete devices and integrated circuits for computing, power switching,

data storage and communication. For high-speed and optoelectronic devices such as high-speed

integrated circuits and laser diodes, gallium arsenide (GaAs) is the material of choice. It exhibits

superior electron transport properties and special optical properties. GaAs has higher carrier

mobility and higher effective carrier velocity than Si, which translate to faster devices. GaAs is a

direct bandgap semiconductor, whereas Si is indirect, hence making GaAs better suited for

optoelectronic devices. However, physical properties required for high power, high temperature

electronics and UV/blue light emitter applications are beyond the limits of Si and GaAs. It is

essential to investigate alternative materials and their growth and processing techniques in order

to achieve these deviceS 2, 3, 119. So now the focus has shifted to semiconductors having wide









bandgaps. They exhibit inherent properties such as larger bandgap, higher electron mobility and

higher breakdown field strength making them suitable for high power, high temperature

electronic devices and short wavelength optoelectronics.

Wide bandgap semiconductors offer the best technical promise for high power and high

temperature transistors. Until recently, the most promising of these materials was silicon carbide

(SiC). However, SiC has several technical shortfalls that have opened competition to the III-

nitride materials. Thermal oxides in SiC power metal oxide semiconductor field effect transistors

(MOSFETs) actually limit the temperature range of application since the gate contact degrades

and becomes electrically leaky at high temperatures. The low electron mobility of only 400

cm2/V. s yields lower PAE (<30%) for many transistors in the frequency range of 1 to 5 GHz. For

silicon and SiC, amplifier efficiency decays rapidly with increase in frequency so that it drops

below 25% for many devices operating above 2 GHz. GaN-base devices offer wider bandgap,

greater chemical inertness and higher temperature stable operation than SiC 120

Single transistor output power is the most important cost limiting issue for

commercialization of solid-state power devices. Other economic factors relating to performance

are power-added efficiency (PAE) required for lightweight portable systems, amplifier linearity

necessary to transmit digital signals without distortion or out-of-band modulation products, and

amplifier noise figure and phase noise. Output power achievable by microwave devices is

directly proportional to the breakdown voltage and sustainable current limits. For bipolar devices

under Class A operation, the maximum output power density is then

Pmax = Isat (Vcb Vknee) / 8
where Isat is the saturation current at the quiescent point, Vcb is the collector breakdown voltage

and Vknee is the saturation voltage at maximum current. Apart from high breakdown voltage, high

thermal stability, GaN-based semiconductors also benefit from a very high sustainable electron









saturation velocity of 2.7x107 cm/s. This unique property, which has been shown to significantly

benefit GaN FETs, is the result of large energetic displacement between valleys in the

conduction band profiles 120

One of the most significant problems limiting single transistor high power devices is the

heat dissipation required. Mature silicon RF power transistors are currently limited to about 125

OC junction temperature (85-100 oC ambient) with operation of little more than 1 W at 10 GHz.

Due to leaky oxides, SiC does not increase this range enough to result in significant advantage.

GaAs technology has improved on this performance to yield 50 W at 10 GHz with state-of-the-

art power FET technology. However, both silicon and GaAs devices suffer greater high

temperature de-rating than is expected from the wide bandgap GaN devices. The GaN devices

not only can operate at 400 OC or higher but also should exhibit optimal performance somewhere

near 250 oC due to improved ionization of the carriers in the material 119

While further improvements in the III-V nitride materials quality can be expected to

enhance device operation, further device advances will also require improved processing

technology. Owing to their wide bandgap nature and chemical stability, GaN and related

materials present a host of device processing challenges, including difficulty in achieving

reliable low-resistance ohmic contacts, thermally stable contacts for both n and p GaN, high

temperatures needed for implant activation, lack of efficient wet etch process, generally low dry

etch rates and low selectivity over etching masks, and dry etch damage. High thermal budget and

dry etch damage indirectly adds to the problem of having good reliable ohmic and schottky

contacts. These problems constitute a maj or obstacle to successful demonstration and

commercialization of some GaN-based devices, such as bipolar transistors and power switches,

whose performance are much more affected by the immature fabrication techniques. To fully









exploit these device applications, a number of critical advances are necessary 4,119. One of the

critical area is high temperature thermal processing, ohmic and schottky contacts which are

thermally stable and can at least sustain harsh condition which the device it self is cable of based

on its intrinsic properties.

The motivation of this work is to develop novel ohmic and schottky contacts to GaN and

AlGaN/GaN high electron mobility devices for use in high temperature application. So the

obj ective of this work is to have high temperature stable ohmic and schottky contact to n-GaN

which should circumvent or delay the problem of intermixing of metal layers and surface

roughening leading to a better and reliable contact scheme. In this proj ect, we explore a novel

metallization scheme involving borides because of the refractory nature of the borides and thus

thermal stability and very little possibility of it having solid state reactions with other metals

normally used in contact scheme. Apart from boride, a high temperature metal, namely Ir, was

also explored as part of continuing search for better contacts. The obj ective is to have an

optimized new contact for high temperature operation of AlGaN/GaN HEMTs.

The properties of GaN and background of semiconductor processing and characterizations

especially in terms of ohmic and schottky contacts are reviewed in Chapter 2. Ohmic and

schottky contacts are necessary to impart specific electrical interactions and characteristics in

achieving operating devices. The studies of ohmic and schottky contact metallization are covered

in Chapters 3 through 7. The Tungsten Boride (W2B) and Zirconium Boride (ZrB2) metal

scheme was considered first. Next a comparative study of Tungsten Boride (W2B5), Titanium

Boride (TiB2) and Chromium Boride (CrB2) is done and is presented in chapter 4. Chapter 5

discusses the rectifying nature of Tungsten Boride (W2B) and Zirconium Boride. Chapter 6

shows result of Titanium Boride, Tungsten Boride (W2B5) and Chromium Boride in usage as a









rectifying contact to n GaN. The demonstration of a High Electron Mobility Transistor using

new best boride based metallurgy is given in Chapter 7. Chapter 8 deals with high melting

temperature metal Iridium (Ir) as being explored as ohmic and schottky contact to n GaN. The

conclusion and summary of the study of new boride based and Ir contacts to n GaN and HEMTs

is given in chapter 9.









CHAPTER 2
BACKGROUND AND LITERATURE REVIEW

Properties of GaN

Overview

GaN is a wide bandgap semiconductor which has numerous properties which makes it well

suited for high temperature applications. Its electrical properties are compared to Si, SiC and

other materials in Table 2-1 3, 5-7. It has a, direct bandgap energy of 3.45 eV (h=3 59.37 nm) which

is transparent to visible light and operates in ultra violet to blue wavelengths. Hall measurements

at room temperatures show the Hall mobility of electron of 1000~1300 cm2/V-s. It has saturation

velocity little higher than GaAs. GaN like ZnO seems to be extremely stable at harsh

environment of gamma radiations. It has little change in IV characteristic even after being

irradiated by high energy proton radiation 7. This makes GaN very good candidate for outer

space and nuclear application. Sapphire or SiC substrates are generally used for growing GaN.

GaN also has different hetrostuctures available with Al, In etc. (Al, Ga, In) N forms a continuous

and direct band gap alloy from 1.92 eV (InN) to 6.2 eV (AIN) with potential for emission and

detection in spectral range between visible and the ultraviolet wavelengths s

Crystal Structure and Basic Properties

GaN is a direct bandgap semiconductor having stable form as hexagonal (wurtzite) crystal

structure, with lattice parameters a = 3.189 A+ and c = 5.178 A. The Ga (group III) atoms are

tetrahedrally coordinated with four N (group V) atoms. Alternating Ga and N layers form the

crystal structure 9 (Figure 2-1).

GaN-based semiconductors have attracted tremendous interest for their applications to blue

laser and LEDs, high temperature, high power electronics, high density optical data storage, and









electronics for the aerospace and automobile industries, telecommunication devices, and wide

band gap semiconductors in power amplifiers extends the radiation hardness of the circuit

Many of these compounds are shown graphically in Figure 2-2 10 in terms of their

crystallographic lattice constant versus the energy band gap. Especially, wide band gap

electronic devices have excellent electrical and physical characteristics. Table 2-2 shows the

physical parameters in different semiconductor materials. The high power, high frequency

operation most promising materials are GaN and SiC and the band gap energy is 3.4 eV and 3.2

eV respectively. For example, to get 10 "/cm3 intrinsic carrier concentration (ni), we need 300 oC

for the Si materials, 500 oC for the GaAs, however much higher temperatures are needed to get

the same intrinsic carrier concentration in the GaN, namely about 1000 oC. For these reasons, the

GaN is much better for use in high temperature conditions, and devices made out of it will

operate more reliably at elevated temperature.

The early unintentionally doped GaN was n-type, which at that time was believed due to

nitrogen vacancies. The high n-type background carrier concentration on the order of 101s cm-3

proved difficult to minimize and the absence of a shallow acceptor dimmed the prospects of a

production-scale GaN-based device effort.

Table 2-3 shows the ionization energy of impurities for GaN. Si is the most general n-type

dopant of for GaN since it effectively incorporates on the Ga site and forms a single shallow

donor level. Si is fully ionized at room temperature with the ionization level of ~30 meV.

Basic physical properties of GaN are listed into Table 2-4 5, 6. These values are result of

various works and some values have uncertainty because of the fact that different materials are

used for experiments and there remains certain inhomogeneity. There is also a metastable form

of GaN as Zinc Blend structure. The variations in different properties like calculated mobility,









thermal conductivity are possible because of crystal defects such as dislocations. Defects in the

materials are very critical factor in the effect. Lots of dislocations are caused by lattice mismatch,

which is 13% on the sapphire, 3% on the SiC substrate.

The successes of all GaN related devices depend largely on having excellent ohmic and

schottky contacts to these devices. A figure of a GaN HEMT is shown in Figure 2-3. There are

two types of contacts to a semiconductor. One contact is ohmic and other is schottky

Metal Contact

At present improvement in contact has become a critical factor for better technology along

with advancing the properties of the semiconductors itself. In recent years, GaN itself have been

proven to be excellent choice for high temperature, high frequency applications. The successes

of all GaN related devices for high temperature application will depend largely on having

excellent contacts to these devices. There are two types of contacts to a semiconductor. Contact

to semiconductor basically consists of region of semiconductor surface just below first metal

layer, metal semiconductor interface and few layers of metallization above it. Invariably the as

deposited contact does not give the desired properties (either low resistance or high schottky

barrier). So the contacts are annealed which results in formation of different complex

intermetallic compounds by way of solid state reaction among metal layers and semiconductor

surface. Thus a contact simply is referred to as the region of metal semiconductor interface that

leads to desirable electrical characteristic. The current transport in metal-semiconductor contact

occurs by maj ority carriers. There are two different types of contacts namely ohmic and schottky.

In ohmic contact the current-voltage relation follows Ohms law that is it should be linear. The

contact resistance should be very low so that there is negligible voltage drop across it and hence

negligible power drop. This is very important for devices and more so in power application

where minimum loss and maximum efficiency is required. Another critical requirement for high










temperature application is the need to have contact which does not degrade or rather have a high

resistance to degradation. Smooth surface morphology, sharp edge acuity and reliability and

reproducibility are other features that are desired in an ideal contact.

Another type of contact is schottky contact, or rectifying contact in which large current can

flow in one direction at small voltage and almost no current in reverse direction. High barrier

height is essential for producing rectifying effects.

Whether a metal-semiconductor interface forms an ohmic or schottky contact depends

upon the metal work function, Om, and semiconductor work function, Os. Work function is the

amount of energy required to excite an electron from Fermi energy level to the vacuum level.

Theoretically, on n-type semiconductor, ohmic contact is formed when $m < Os, and schottky

contact is formed when $m > Os. Conversely, in p-type material, Om > Os and 4m < Os produces

ohmic and rectifying contact, respectively. Selected values of work function for commonly used

metals are shown in Table 2-5 11. The semiconductor work function is sum of the electron

affinity and energy difference between Fermi energy and the bottom of the conduction band i.e.

#s = Is + 5, where Xs is the electron affinity and 5 is the energy difference between the Fermi

energy and the conduction band 3. The electron affinity for GaN is 4.1 eV 11. The work function

of Tungsten (W), Cr, Ti, Zr is 4.55, 4.5, 4.33 and 4.05 eV respectively.

Schottky contact

When an intimate contact is formed between metal and a semiconductor, the Fermi levels

in the two materials must be coincident at thermal equilibrium. This can be achieved through a

charge flow from semiconductor to metal. Thus a barrier forms at the interface and an equal and

opposite space charge is distributed over the barrier region near the semiconductor surface. With

an n-type semiconductor in the absence of surface state, the barrier height q~bn is given by









qAn = q( m X) (2-1)

where q~m is the metal work function, qX is the electron affinity of the semiconductor. For an

ideal contact between metal and a p-type semiconductor, the barrier height q~b, is given by

qAp = Eg q(~ X) (2-2)
When surface states are present on the semiconductor surface, and the density is

sufficiently large to accommodate any additional surface charges without appreciably altering

the occupation level EF, the space charge in the semiconductor will remain unaffected. As a

result, the barrier height is determined by the property of the semiconductor surface, and is

independent of the metal work function. In practice, some surface states are always present at the

semiconductor surface, and continuously distributed in energy within the energy gap. The

schottky barrier heights of metal-semiconductor systems with intimate contact are, in general,

determined by both the metal work function and the surface states.

In a simple model for all semiconductors, the schottky barrier height 75 q~b can be

expressed as

qAb = q(SX, oW) (2-3)
where Xm is metal electronegativity, 40 represents the contribution of surface states of



semiconductors, and interface index S= dXm is found to be a function of the electronegativity

difference AX between cation and anion of compound semiconductor (Figure 2-4 A). Note a

sharp transition around AX=1. For ionic semiconductors, AX>1, the index S approaches 1, and Ob

is strongly dependent of the metal electronegativity (or work function). On the other hand, for

covalent semiconductors with AX<1, S is small, Ob is affected by high density surface states from

dangling bonds and only weakly depends on metal work function. GaN has an electronegativity

difference of 1.4 (Ga: 1.6, N: 3.0), which would predict the schottky barrier heights depend on









metal work function, and are given by Equation 2-1 and Equation 2-2 for metal on n-type and p-

type material respectively. A summary of reported schottky barrier heights of a variety of

elemental metals on n-GaN is shown in Figure 2-4 B 12. It is clear that the barrier height indeed

varies with the metal work function within experimental scattering.

The current transport in metal-semiconductor contacts is mainly due to maj ority carrier, in

contrast to p-n junctions. Two maj or processes under forward bias are (1) transport of electrons

from the semiconductor over the potential barrier into the metal; (2) quantum-mechanical

tunneling of electrons through the barrier. In addition, we may have recombination current in the

space-charge region and leakage current at the contact periphery. The transport of electrons over

the potential barrier is often the dominant process for schottky diodes on moderately doped

semiconductors. It can be adequately described by thermionic emission theory for high mobility

semiconductor (for low mobility materials, the diffusion theory is also applicable), and the

electric current density over the barrier has the following expression:

qV q*1, gb qV
J = J, [exp( )-1]= A'T O Xp(- )[exp( )1
kT kT kT (2-4)
where Js is the saturation current density, A* is the effective Richardson constant. In practical

device, the barrier height dependent on bias voltage and the current-voltage characteristics is

more accurately described by:


J= A'T2 Xp(- b)[exp( q)1
kT nkT (2-5)
Factor n is called the ideality factor. The barrier height and ideality factor can be obtained from

the forward J-V characteristics (for V>3kT/q):


kT dln J (2-6)
kT A*T 2

ii .7,(2-7)









For a heavily doped semiconductor or for operation at low temperatures, the tunneling

current may become the dominant transport process. The tunneling current has an expression:


J, ~ e xp ()(2 8

where as is permittivity of semiconductor, m* is effective mass of carrier, ND is Carrief

concentration. It indicates the current will increase exponentially with NDO

Earlier work on schottky contacts to n-GaN have been done based on metals layers

consisting mainly of Ni or Pt with Au above it 37-40, 65, 66. W/Ti/Au and WSix/Ti/Au schemes have

also been used as schottky resulting in thermally stable schottky with barrier height of ~0.80 eV

which reduced to ~0.4 eV for subsequent annealing at 400 oC 68. The barrier height seems to

follow the difference in work function value with in experimental scattering.

Schottky diode can be made by depositing an inner circular schottky metal scheme with an

outer concentric ring as the ohmic metal scheme. The outer ohmic metal is first deposited and

annealed to get the desired ohmic characteristic and inner circle is realigned. A diagram of a

schottky structure is shown is Figure 2-5.

Ohmic contact

It is imperative that a semiconductor device be connected to the outside world with no

adverse change to its current-voltage characteristics. This can be accomplished through ideal

ohmic contacts to the semiconductor. An ohmic contact is defined as a metal/semiconductor

contact that has negligible contact resistance relative to the bulk or spreading resistance of the

semiconductor. A satisfactory ohmic contact can supply the required current with a voltage drop

that is sufficiently small compared with the drop across the active region of the devices. One

important figure of merit for ohmic contact is specific contact resistance re, which is defined as:










-1
re = v=0 (2-9)
For contact with lower doping concentration, at relatively high temperature, conduction

across the M/S interface is dominated by thermionic-emission over the potential barrier, as given

in Equation 2-4. Therefore,

k q~b
exp( )
re q~A**T kT (2-10)
It is obvious that low ~b should be used for small re. Ideally a metal with a lower work

function than an n-type semiconductor or higher work function than a p-type semiconductor

should be used for ohmic contact to this semiconductor. Unfortunately, very few practical

material systems satisfy this condition, and metals usually form schottky barriers at

semiconductor interface. A practical way to obtain a low resistance ohmic contact is to create a

highly doped region near the surface by ion implantation, or increase the doping by alloying the

contacts. In this case, the depletion layer cause by the schottky barrier becomes very thin, and

current transport through the barrier is enhanced by tunneling. The contact resistance can be

obtained from Equation 2-8,



re ~ A(2-11)
Note that re depends strongly on ND. Under intermediate conditions, thermionic field

emission is important, where there is enough kinetic energy for the carrier to be excited to an

energy level at which the potential barrier is thin enough for tunneling to occur. Typical ohmic

conduction is usually related to a large tunneling component.

It is difficult to make ohmic contacts on wide-bandgap semiconductors, such as GaN

(sg=3 .4 eV, X=4. 1 eV) and SiC. Generally the doping concentration is relatively low due to the

high ionization level of typical dopants.









A wide variety of metallization schemes been tried for ohmic contact for n-GaN. Some of

the earliest report of ohmic contact to n-GaN had Al as the ohmic contact metal with specific

contact resistivity of ~ 10-7 O-cm2 34. Specific contact resistivity of 8 x 10-6 O-cm2 USing Ti/Al

was achieved after 900 oC anneal for 30 sec in N2 ambient 19 Ti/Al contact with Si implantation

resulted in specific contact resistivity of 3.6 x 10-s O-cm2 and with RIE pre treatment resulted in

8.9 x 10-s O-cm2 18-21. A Specific contact resisitivity of ~ 8 x 10-5 O-cm2 WaS reported for W on

n-GaN 37. Specific contact resistivity of ~5.6 x 10-6 O-cm2 for Al/Ti contact on AlGaN /GaN

hetrostructure with Si implantation, 5.3 x 10-7 O-cm2 for Ta/Ti/Al contacts, 1.2 x 10-5 O-cm2 foT

Ti/Al/Ni/Au contacts have been also reported 24,28,30. The most common contact scheme used is

Ti/Al bilayer with Ni/Au, Ti/Au/ or Pt/Au over layer where overlayer is mainly for preventing

out diffusion, smooth morphology and Au is used for reducing sheet resistance of the layers and

to prevent oxidation during high temperature anneal 13-36. High temperature metals have also

been used for ohmic schemes to have better long term stability. Ti/Al/Mo/Au, Ti/Al/Ir/Au, W

and WSix gave specific contact resistivtiy of ~10-5 Ocm2 37-43

Contact resistance is measured and Specific contact resistance is determined by Transfer

length model (TLM), also known as transmission line model. Linear TLM patterns consist of

square or rectangular contact pads separated by different spacing. There is also a Circular TLM

patterns which has concentric circular metal patterns where either the inner radii or the outer

radii change to vary gap distance. Schematics of linear TLM and measurement plot is shown in

Figures 2-6.

Current and voltage information obtained from electrical measurements are curve fitted

with the corresponding equations to determine the specific contact resistance. For linear TLM,

the total resistance, Rs, and specific contact resistance, pe, are given by









L\
R, = 2Rc +Rs


Pc =~zR


where Rc is the contact resistance, Rs is the sheet resistance, L is the distance between two pads,

W is the width of the pad. For Circular TLM, the specific contact resistance, pe, can obtained

form the circular TLM measurements with the relationshipS 108, 120

Rs ,I R L, Ko (R, / L, )
R, = In. + +
2r Ro oI(R R, K (R /L,)
Pc = Rs L2
where RT is the total resistance, Rs is the sheet resistance, R1 is the outer radius of the annular

gap, Ro is the inner radius of the annular gap, lo, li, Ko, and K1 are the modified Bessel

functions, LT is the transfer length, and p, is the specific contact resistance.

Thermal stability

Thermal processing such as activation of ion implants and alloying of metal contacts can

be detrimental to device operation due to changes in the material, interaction, or reactions, as

also observed in GaAS 109-111, 120. It is very important to have contacts that are able to resist the

high temperature long enough to be commercially possible for high temperature applications. In

regards to this a good understanding of the degradation of the material is helpful in identifying

high temperature process limits. Reliable and stable operation of devices largely depends upon

the thermal stability of the contacts. At high temperature a lot of intermetallics compounds may

form in contacts as a result of interdiffusion of different metal. This can result in rough surface

morphology, change in stoichiometry, change in composition resulting in change in electrical

and optical properties.









Common Processing Techniques


Dry Plasma Etching

Etching refers to the crucial IC fabrication process of transferring pattern by removing

specified areas. Wet chemical etching was widely used in manufacturing until the 1960s. Even

though this technique is inexpensive, the feature size is limited to about 3 microns. The isotropic

etching results in sloped sidewalls and undercutting of the mask material. As feature dimension

decreases to microns and submicrons and device density per chip increases, anisotropic etching

is necessary. Dry etching techniques using gases as primary etch medium were developed to

meet this need. In addition to anisotropic pattern transfer, dry etching provides better uniformity

across the wafer, higher reproducibility, smoother surface morphology, and better control

capability than wet chemical etching. Three general types of dry etching include plasma etching,

ion beam milling, and reactive ion etch (RIE) 112, 113, 120. Inductively coupled plasma (ICP)

etching was used in this study and will be discussed in detail.

ICP etching is a dry etching technique where high-density plasmas are formed in a

dielectric vessel encircled by inductive coils as shown in Figures 2-7 and 2-8. When an rf power

is applied to the coil, commonly referred to as the ICP source power, the time-varying current

flowing through the coil creates a magnetic flux along the axis of the cylindrical vessel. This

magnetic flux induces an electric field inside the vacuum vessel. The electrons are accelerated

and collide with the neutral operating gas, causing the gas molecules to be ionized, excited or

fragmented, forming high-density plasma. The electrons in circular path are confined and only

have a small chance of being lost to the chamber walls, thus the dc self-bias remains low. The

plasma generated as described above consists of two kinds of active species, neutrals and ions.

The material to be etched sits on top of a small electrode that acts as parallel plate capacitor

along with the chamber as the second electrode. When an rf power, also known as electrode










power or chuck power, is applied to the sample stage, the electrons in the plasma accelerate back

and forth in the plasma from the changes in the sinusoidal field. Since electrons have much

lighter mass compared to the other species in the plasma, they respond more rapidly to the

frequency change than the other species. As the electrons impinge the chamber surfaces, the

chamber becomes slightly negative relative to the plasma. The surface area of the chamber is

larger than the sample stage, thus the negative charge is concentrated on the sample stage. This

bias attracts the ions toward the sample, bombarding the surface to remove material. In an ICP

system, the plasma density and the ion energy and are effectively decoupled in order to achieve

uniform density and energy distributions and maintain low ion and electron energy low. This

enables ICP etching to reduce plasma damage while achieving fast etch rates.

The plasma generated as described above consists of two kinds of active species: neutrals

and ions. Neutrals are chemically reactive and etch the material by chemical reactions, while

ions are usually less reactive and are responsible for removing material by physically

bombarding the sample surface. The kinetic energy of the ions is controlled by electrode bias.

The electron density and ion density are equal on average, but the density of neutrals, known as

the plasma density, is typically higher. Anisotropic profiles are obtained by superimposing an rf

bias on the sample to independently control ion energy and by using low pressure conditions to

minimize ion scattering and lateral etching.

The plasma is neutral but is positive relative to the electrode. It appears to glow due the ion

excitation from the electron movements. The recombination of charges at the boundary surfaces

surrounding the plasma creates a charge depletion layer, also known as a sheath, dark space or

dark region, resulting in diffusion of carriers to the boundaries. The diffusion of electrons is

faster than ions initially, thus an excess of positive ions is left in the plasma and assumes a










plasma potential, V,, with respect to the grounded walls. The plasma and substrate potentials

generate drift current to enhance the ion motions and hinder the electron motions until steady

state condition is achieved. The difference in electron and ion mobility also generates a sheath

near the powered electrode. The dark region, a small region in the plasma immediately above the

sample, keeps the electrons away due to the negatively charged electrode. The powered electrode

reaches a self-bias negative voltage, Vdc, with respect to the ground. Even though the voltage

drop controls the ion bombardment energy across the plasma sheath, it is difficult to measure;

therefore, it is common to monitor the Vdc. Note that the dc bias is not a basic parameter and is

characteristic to a particular piece of equipment.

Etching is accomplished by the interaction of the plasma to the substrate. The three basic

etching mechanisms, chemical etch process, physical etch process, and a combination of both

chemical and physical etching process, are shown in Figures 2-9, 2-10, and 2-11, respectively.

Chemical etch process is the chemical reaction that etches the substrate when active species

(neutrals) from the gas phase are absorbed on the surface material and react with it to form a

volatile product. The chemical etch rate is limited by the chemical reaction rate or diffusion rate

that depends on the volatility of the etch products since undesorbed products coat the surface and

prevent or hinder further reactions. Chemical etching is a purely chemical process therefore

etches isotropically, or equally in all directions. Physical process, also known as sputtering,

occurs when positive ions impinge normal to the substrate surface. If the ions have sufficiently

high energy, atoms, molecules or ions are ej ected from the substrate surface to achieve a vertical

etch profile. The etch rate of sputtering is slow, and the surface is often damaged from the ion

bombardment. A combination of both chemical and physical etching process, also known as

energy-driven, ion-enhanced mechanism, takes advantage of the effect of ion bombardment in










the presence of reactive neutral species. The energetic ions damage the surface and leave the

surface more reactive toward incident neutrals, leading to removal rates that exceed the sum of

separate sputtering and chemical etching. This process produces very fast etch rates and

anisotropic profie; therefore, it is desirable in high Eidelity pattern transfer 120

lon Implantation

lon implantation is a physical process that introduces dopants by means of high-voltage

bombardment to achieve desired electrical properties in defined areas with minimal lateral

diffusion. Inside a vacuum chamber, a filament is heated to a sufficiently high temperature where

electrons are created from the filament surface. The negatively charged electrons are attracted to

an oppositely charged anode in the chamber. As the electrons travel from the filament to the

anode, they collide and create positively charged ions from the dopant source molecules. The

ions are separated in a mass analyzer, a magnetic field that allows the passage of the desired

species of positive ions with specific characteristic arc radius based upon ion mass. The selected

ions are accelerated in an acceleration tube and then focused into a small diameter or several

parallel beams. The beam is scanned onto the wafer surface, and the ions physically bombard the

wafer. The ions enter the surface and come to rest below the surface as they lose their energy

through nuclear interactions and coulombic interactions, resulting in Gaussian distribution

concentration profile 114, 120. A schematic of an ion implantation system is illustrated in Figure 2-

12 120

During implantation, the collisions with high-energy ions cause crystal damage to the

wafer, leading to poor electrical characteristics. In most cases, the carrier lifetime and mobility

decrease drastically. Also, only a small fraction of the implanted ions are located in substitutional

sites and contribute to carrier concentration. Annealing is needed to repair the crystal damage

and to activate the dopants. To determine the depth and damage profile, Rutherford









Backscattering and Channeling (RBS/C) analytical technique is employed. Annealing process

and RBS/C will be further discussed in the subsequent sections 120

Rapid Thermal Annealing

Annealing is a thermal process used for repairing the ion implantation damage, diffusing

dopants and alloying metal contacts. After ion implantation, annealing is employed to repair the

crystal damages caused by the high-energy ion bombardment that degrade carrier lifetime and

mobility. Since the maj ority of the implanted dopants reside in the interstitial sites, the as-

implanted materials have poor electrical properties. Annealing provides thermal energy for the

dopants to migrate to the substitutional sites and contribute to the carrier concentration 115-116

Traditionally, tube furnaces were used for annealing after ion implantation. However,

furnace annealing causes the implanted atoms to diffuse laterally and requires relatively long

anneal time. Rapid thermal annealing was developed in order to overcome these drawbacks.

Rapid thermal annealing (RTA) utilizes radiation heating from arc lamps or tungsten-

halogen lamps to heat the wafer in an inert atmosphere such as N2 or Ar. It can attain higher

temperature at a shorter time period than a conventional tube furnace, and the overall anneal time

is relatively short, usually taking seconds as compared to several minutes to hours in a

conventional tube furnace. RTA allows uniform heating and cooling that reduces thermal

gradients that can lead to warping and stress-induced defects, enabling more dense design and

fewer failures due to dislocationS 120

Characterization Techniques

Atomic Force Microscopy

Atomic force microscopy (AFM) employs a microscopic tip on a cantilever that deflects a

laser beam depending on surface morphology and properties through an interaction between the

tip and the surface. The signal is measured with a photodetector, amplified and converted into an










image display on a cathode ray tube. Depending on the type of surface, AFM can be performed

in contact mode and tapping mode. A schematic diagram of AFM is shown in Figure 2-13 120

Auger Electron Spectroscopy

Auger electron spectroscopy (AES) determines the elemental composition of the few

outermost atomic layers of materials. A focused beam of electrons with energies from 3 keV to

30 keV bombards the surface of a specimen. The core-level electrons are ej ected from

approximately 1 Cpm within the sample, resulting in a vacancy in the core-level. As the atom

relaxes, an outer-level electron fills the core vacancy and releases excess energy, which in turn,

ej ects an outer electron, known as an Auger electron. This process is illustrated in Figure 2-14.

The kinetic energy of the Auger electrons is characteristic of each element, with the exception of

hydrogen and helium. Therefore, by measuring the energies of the Auger electrons, the near-

surface composition of a specimen can be identified. In addition, AES can provide compositional

depth profile from relative intensities of the elements present if the system is equipped with an

ion gun to sputter away material 117, 120

X-ray Photoelectron Spectroscopy

X-ray photoelectron spectroscopy (XPS), also known as electron spectroscopy for

chemical analysis (ESCA), provides similar information as AES. Instead of impinging the

sample surface with an electron beam, XPS utilizes a monoengergetic x-ray beam to cause

electrons to be ej ected, usually two to 20 atomic layers deep. The variation of the kinetic

energies of the ejected electrons identifies the elements present and chemical states of the

elements 118, 120

Electrical Measurements

Current-voltage (I-V) measurements were taken to characterize the electrical properties of

the contacts. These measurements are performed on an Agilent 4156 Semiconductor Parameter










Analyzer connected to a micromanipulator probe station. For diodes, the input voltage was

applied through schottky and out through ohmic for forward bias and visa versa for reverse bias

measurements. For ohmic measurements on TLM pads 4 probes were used in series, two outer

probes for applying the current and inner two probes for picking up the voltage.

Photoluminescence (PL)

Photoluminescence (PL) is an analytical technique that provides information about the

optical properties of a substrate. A light source, such as He-Cd, Ar and Kr lasers, with energy

larger than the bandgap energy of the semiconductor being studied, generates electron-hole pairs

within the semiconductor. The excess carriers can recombine via radiative and non-radiative

recombination. Photoluminescence, the light emitted from radiative recombination, is detected.

The wavelength associated with the different recombination mechanism is measured.

The luminescence from excitons, electrons and holes bound to each other, is observed only

at low temperatures in highly pure materials. As the temperature increases, the exciton breaks up

into free carriers from the thermal energy. Increase in doping also causes the dissociation of

excitons under local electric fields. Under these conditions, the electrons and holes recombine

via the band-to-band process. Since some of the electrons may not lie at the bottom of the

conduction band, their recombination and holes will produce a high-energy tail in the

luminescence spectrum. On the other hand, the band-to-band recombination will yield a sharp

cutoff at the wavelength corresponding to the band gap of the material 118, 120

Rutherford Backscattering Spectrometry/Channeling

Depth profile of implanted ions and damages can be obtained by the Rutherford

Backscattering Spectrometry/Channeling (RBS/C) technique, which measures the energy

distribution of the backscattered ions from the implanted sample surface at a specific angle. The









energy of the backscattered ion is determined by the mass of the atomic nucleus and the depth at

which the elastic collisions take place.

A beam of high-energy ions impacts the surface of the specimen. The angle of the

analyzing ions affects the penetration depth. If the ions are inj ected parallel to the crystal axis of

the specimen, they penetrate considerably deeper than if inj ected randomly, due to the lower

stopping power from channeling. Deeper penetration results in higher backscattered ions yield.

The displacement of an atom, either as host or impurities, from the crystal lattice also increases

the backscattering yield. Therefore, the distribution of displaced atoms that are caused by the

radiation damage from ion implantation can be measured by increasing the backscattered ion

yield 118, 120

Scanning Electron Microscopy

Scanning electron microscopy (SEM) generates images from electrons instead of light. A

beam of electron is produced and accelerated from an electron gun. The electron beam passes

through a series of condenser and obj ective lenses, which focus the electron beam. A scanning

coil moves the beam across the specimen surface. The electron beam interacts with the specimen,

and electrons from the surface interaction volume, such as backscattered, secondary,

characteristic x-ray continuous x-ray, and Auger, are emitted. The signals are collected,

amplified and converted to a cathode ray tube image. Depending on the specimen and the

equipment setup, the contrast in the final image provides information on the specimen

composition, topography and morphology. The main advantages of using electrons for image

formation are high magnification, high resolution and large depth of fieldS 118, 120

Secondary lon Mass Spectrometry

Secondary ion mass spectrometry (SIMS) is a highly sensitive chemical characterization

technique. Primary ions, such as Cs Ol O- and Ar, bombard the specimen in an ultra high










vacuum environment, sputtering away secondary ions from the specimen surface. A small

fraction of the ej ected atoms are ionized either positively or negatively, and they are called

secondary electrons. The composition of the surface is determined by the secondary electrons

that are individually detected and tabulated using a mass spectrometer, as a function of their

mass-to-charge ratio.

There are two modes of SIMS, static or dynamic. In the static mode, a low primary-ion

flux <1014 cm-12 iS USed, leaving the specimen surface relatively undisturbed. The maj ority of

secondary ions originate in the top one or two monolayers of the samples. The dynamic mode

monitors the selected secondary ion intensities as a function of the sputtering time, resulting in a

concentration versus depth profile. The depth resolution of this technique ranges from 5 to 20 nm

118, 120


Stylus Profilometry

Stylus profilometry is used to measure the topographical features of a specimen surface,

such as roughness, step height, width and spacing. A probe, or stylus, contacts the surface of the

specimen and follows height variation as it scans across the surface. The height variations are

converted into electrical signals, providing a cross-sectional topographical profile of the

specimen. In this work, the etch rate was calculated by the depth, as measured by the

profilometer, over a specified period of time.

Table 2-1. Electrical properties of Si, GaAs and GaN.
Property Si GaAs GaN
Bandgap energy (eV) 1.1 (indirect) 1.4 (direct) 3.4 (direct)
Electron mobility 1400 8500 1000 (bulk)
(cm2NS) 2000 (2D-gas)
Hole mobility 600 400 30
(cm2NS)
Electron effective 0.98 0.067 0.19
mass
Hole effective mass 0.16 0.082 0.60
(light)








































Table 2-3. Ionization energy of impurities for wurtzite GaN.
Impurities Ga-site (eV) N-site (eV) Remarks
Si 0.012-0.02 Donor
Native defect (VN) 0.03 Donor
C 0.11-0.14 Donor
Mg 0.14-0.21 Acceptor
Si 0.19 Acceptor
Zn 0.21-0.34 Acceptor
Native defect (v,,) 0.14 Acceptor
Hg 0.41 Acceptor


Table 2-2. The physical parameters in different semiconductor materials
Si GaAs GaN AIN 6H-SiC
Bandgap (eV) 1.1 1.4 3.4 6.2 2.9
ii300oC


1400


8500


1000 (bulk)
2000 (2D-gas)


30


2.5


Electron
mobility
(cm /V-s), RT

Hole mobility
(cm /V-s), RT

Saturation
velocity
(cm/s), 107
Breakdown
field (V/cm)
x 106
Thermal
conductivity
(W/cm)
Melting
temperature
(K)


1.5


>1700


1690


1510


3000


>2100

































Table 2-5. Metal work function and ideal barrier heights for GaN (electron affinity: 4.1 eV)
Element Work Function (eV) Ideal Barrier Height (eV)
B 4.45 0.35
Cr 4.5 0.4
Pt 5.64 1.54
Ti 4.33 0.23
W 4.55 0.45
Zr 4.05 -0.05


Table 2-4. Basic physical properties of GaN.
Property


Value
ao = 0.3189 nm
co = 0.5185 nm
6.095 g.cm-3
Wurtzite
2500
1.3, 2.2+0.2
For thick, free-standmng GaN
Along ao -5.59x10-6 K1
Along co =7.75x10-" K1
8.9
2.67 at 3.38 eV
Direct, 3.45
26
0.20


Lattice parameters at 300 K (nm)
Density (g cm-3)
Stable phase at 300 K
Melting point (oC)
Thermal conductivity (Wem- K')

Lmnear thermal expansion coefficient
Static dielectric constant
Refractive index
Energy bandgap (eV)
Exciton binding energy (meV)
Electron effective mass



























































0.17 I I I
3.3 3.6 r/ 5.43 6.05 6.5
Lattice Constant (A) at 300k

Figure 2-2. The III-V compound semiconductor tree


Figure 2-1. Crystal structure of wurtzite GaN.


I I













AlGaN

UID n+ AlGaN
space -
layer
2DEG
GaN





Substrate



Figure 2-3. Structure of a HEMT
























































4.4 4.8 4.81 5.0 5.2 5.4 5.6


Work Function (eV)

Figure 2-4. Previous study of schottky contacts A) Index of interface behavior S as a function of
the electronegativity difference of the semiconductors. B) Barrier height versus work
function of metals deposited on n-GaN reported from various groups.


S0.2 0 4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4









m *

w Pt
Ag .i

Pd
Pb Au

Cr B

Ti


I


AfN ZDS At203 SrTI103
*


CdS
GoSe
*ZnSe
SiC

` cd se
cdre
Ge GaIAs
InSb'~


SnO2
Si02
GG203


KTaOs


0.8



0.6



0.4


A


4.2















Figure 2-5. Lithography pattern for Schottky diode


WV


Slope=Rs/W


L, L2 L3
Distance L


Figure 2-6. Lithography pattern for linear TLM A) TLM pads. B) Plot for measurement.


Metal Pads


L2 L3
Sem ico nd ucto r filIm








































\ ~Rf Current




Inue EFed


2 MHz
Power
supply


Figure 2-7. An ICP reactor.


\/



Magneti Fietld


Figure 2-8. Electric and magnetic fields inside the reactor.
















eElectron
SReactive neutral
l on

*Substrate atom

Figure 2-9. Chemical etching process. A) Generation of reactive species. B) Diffusion of reactive
neutrals to surface. C) Adsorption of reactive neutrals to surface. D) Chemical
reaction with surface. E) Desorption of volatile byproducts. F) Diffusion of
byproducts into bulk gas.



A- *











Sample-Negatively biased

Figure 2-10. Physical etching process. A) Generation of reactive species. B) Acceleration of ions
to the surface. C) Ions bombard surface D) Surface atoms are ejected from the surface.









A *


B ~L


c
1
E


Figure 2-11. Combination of chemical and physical etching process. A) Generation of reactive
species. B) Diffusion of reactive neutrals to surface. C) lon bombardment to surface.
D) Adsorption of reactive neutrals to surface. E) Chemical reaction with surface. F)
Desorption of volatile byproducts. G) Diffusion of byproducts into bulk gas.


Mass Analyzer


Focus Scanner





Wafer

Acceleration
lon ,I Tube
Source



Figure 2-12. Ion implantation system.


G


8
























Laser beam


Photodetector


Specimen surface


Figure 2-13. Simplified principle of AFM.

















SElectron

O Vacancy
Auger Electron


Figure 2-14. Auger Process. A) An isolated atom. B) Inner core level electron dislodged, leaving
behind a vacancy. C) An outer level electron fills the vacancy and releases excess
energy. D) The excess energy ej ects an Auger electron.









CHAPTER 3
TUNGSTEN AND ZIRCONIUM BORIDE BASED OHMIC CONTACTS TO N-GaN

Introduction

There is a strong interest in the development of more reliable and thermally stable ohmic

contacts on GaN-based electronic devices such as high electron mobility transistors (HEMTs) 13-

45, which show outstanding potential in advanced power amplifiers for radar and communication

systems over a broad frequency range from S-band to V-band 13-15. A key aspect of operation of

nitride-based HEMTs at high powers is the need for temperature-stable high quality ohmic

contacts. There is increasing interest in the application of AlGaN/GaN HEMTs to microwave

power amplifiers capable of uncooled or high temperature operation where thermal stability of

the contact metallurgy is paramount. The most common ohmic metallization for AlGaN/GaN

HEMTs is based on Ti/Al. This bilayer must be deposited with over-layers of Ni, Ti or Pt,

followed by Au to reduce sheet resistance and decrease oxidation during the high temperature

anneal needed to achieve the lowest specific contact resistivity 16-36. For improving the thermal

stability of ohmic contacts, there is interest in higher melting temperature metals, including W

37,38,40,45, WSiX 37-39,Mo 16,V 39 and Ir 40,42,43 Another promising metallization system as the

diffusion barrier layer is based on borides of Cr, Zr, Hf, Ti or W 46-47. Stoichiometric diborides

have high melting temperatures (eg. 3200 oC for ZrB2) and thermodynamic stabilities at least as

good as comparable nitrides or silicideS 48. These metallization systems are expected to be less

reactive with GaN than the conventional Ti/Al. Previous work has shown good contact resistance

obtained with Ti/Al/Mo/Au, Ti/Al/Ir/Au, Ti/Al/Pt/WSi/Ti/Au and Ti/Al/Pt/W/Ti/Au on n-GaN

16, 39, but there is a broad range of other contact metallurgies that are promising, including those

based on borides. For example, one attractive option is ZrB2, which has a low resistivity in the

range 7-10 CLD.cm. Recently, it was shown that hexagonal ZrB2 (0001) single crystals have an in-









plane lattice constant close to that of GaN, prompting efforts at GaN heteroepitaxy on ZrB2 Of

buffer layers on Si substrates. For contacts on n-type GaN, the only related work is the study of

ZrN/Zr/n-GaN Ohmic structures 48 in which the Zr/GaN interface was found to have excellent

thermal stability. Even though we are not relying on the ZrB2 to make direct ohmic contact to

GaN,it is expected that ZrB2 COntacts on GaN will have low barrier heights, given the work

function of ZrB2 is ~3.94 eV and the electron affinity of GaN is ~4. 1 eV.

In this chapter a report on the annealing temperature dependence of contact resistance and

contact intermixing of Ti/Al/W2B/Ti/Au and of Ti/Al/ZrB2/Ti/Au on n-GaN is given. The

Tungsten Boride contacts show excellent minimum contact resistance of 7x 10-6 QZ.cm2 after

annealing at 800 OC and promising long-term stability at 200 OC. The ZrB2 based contacts show

excellent minimum contact resistance of 3x10-6 QZ.cm2 after annealing at 700 OC and retain a

good morphology even after annealing at 1000 OC.

Experimental

The samples used were 3 Clm thick Si-doped GaN grown by Metal Organic Chemical

Vapor Deposition on c-plane Al203 substrates. The electron concentration obtained from Hall

measurements was ~7x101 cm-3. Mesas 1.8 Cpm deep were formed by Cl2/Ar Inductively

Coupled Plasma Etching to provide electrical isolation of the contact pads. A metallization

scheme of Ti (200 A+)/Al (1000 A+)/ W2B or ZrB2 (500 A+) / Ti (200 A+) /Au (800 A+) was used in

these experiments. All of the metals were deposited by Ar plasma-assisted rf sputtering at

pressures of 15-40 mTorr and rf (13.56 MHz) powers of 200-250 W. The contacts were

patterned by liftoff and annealed at 500-1000 OC for 1 min in a flowing N2 ambient in a RTA

furnace. Auger Electron Spectroscopy (AES) depth profiling of the as-deposited contacts showed

sharp interfaces between the various metals in both types of contacts. For the AES analysis, the









samples were mounted on a stainless steel puck and placed in the system load-lock. Clean

tweezers and gloves were used for all sample handling. No additional cleaning steps were

implemented. After sufficient evacuation, the sample puck was inserted into the analytical

chamber and placed in front of the analyzer. The AES system was a Physical Electronics 660

Scanning Auger Microprobe. The electron beam conditions were 10 keV, 1 CIA beam current at

300 from sample normal. For depth proofing, the ion beam conditions were 3 keV Ar 2.0C1A (3

mm)2 raster, with sputter rate based on 110 Bkminute (SiO2) of 142 BJ minute Au (1.3*SiO2) ,64

Bkminute Ti (0.58*SiO2),123 Bkminute Al (1.12*SiO2), 44 Bkminute W (0.4*SiO2) {for W2B),55

Bkminute Al203 (0.5*SiO2) and 83 Bkminute (0.75*SiO2). Prior to AES data acquisition,

secondary electron images (SEI's) were obtained from the sample. The SEls were obtained at

magnifications of 125X, and 1,000X. The SEls were used to locate and document analysis area

locations and to document surface morphology. The quantifieation of the elements was

accomplished by using the elemental sensitivity factors. The contact properties were obtained

from linear transmission line method (TLM) measurements on 100 x100 Clm pads with spacing 5,

10, 20, 40, and 80 lm. The contact resistance Re was obtained from the relation 49RC = (RT

ps- d/Z)/2, where RT is the total resistance between two pads, ps is the sheet resistivity of the

semiconductor under the contact, d is the pad spacing, and Z is the contact width. The specific

contact resistance, pe, is then obtained from pc = RcLTZ, where LT is the transfer length obtained

from the intercept of a plot of RT vs d.

Results and Discussion

Tungsten Boride Based Ohmic Contact

Figure 3-1 shows the contact resistance as a function of anneal temperature. The as-

deposited contacts showed rectifying behavior. This is expected for any as-deposited contacts on

the wide bandgap GaN 25. The current-voltage characteristics became ohmic for anneal









temperatures > 500 oC. The contact resistance decreased up to ~800 OC, reaching a minimum

value of 7x10-6 QZ.cm2. This trend is most likely related to the formation of low resistance phases

of TiN at the interface with the GaN, as reported for conventional Ti/Al/Pt/Au contactS 19-22

However the W2B-based contacts show improved edge acuity, which is important for small gate

length HEMTs in order to reduce the possibility of shorting of the ohmic metal to the gate.

Annealing at higher temperatures leads to higher contact resistance, which as will be seen later

corresponds to extensive intermixing of the contact metallurgy. The corresponding transfer

resistance and semiconductor sheet resistance data are shown in Figure 3-2. The minimum

contact resistance obtained corresponds to a transfer resistance of 0.057 R.mm.

Figure 3-3 shows the measurement temperature dependence of the Ti/Al/W2B/Ti/Au on n-

GaN annealed at 800 OC. Over the relatively limited temperature range available for the

measurements and within the error of the measurement, we did not observe any temperature

dependence. This indicates that at this anneal temperature, the dominant current transport

mechanism is field emission 49, since thermionic emission would have significant temperature

dependence and thermionic field emission is operative at lower doping ranges (1016-101 cm-3~

Figure 3-4 shows the SEI of the as-deposited contact morphology and after annealing at

500,800 or 1000 oC. The morphology is featureless until 800 OC, which corresponds to the

minimum in contact resistance. By 1000 oC, the morphology becomes very rough and this

corresponds to the increased contact resistance.

The AES surface scans as a function of anneal temperature was also obtained. Carbon,

oxygen and gold were detected on the as-deposited surface. The carbon is adventitious and the

oxygen comes from a thin native oxide on the Au. After annealing at 500 OC, titanium was

detected on the contact surface. After 800 oC annealing, gold, titanium, aluminum, and gallium









were detected on the surface, which is consistent with the onset of extensive reaction of the

contact metallurgy. After 1000 oC annealing, titanium, aluminum, and gallium were detected on

the surface. The surface concentration of gold decreased with increasing temperature of the

annealing step. Titanium and aluminum concentrations increased with annealing temperature.

Figure 3-5 shows the AES depth profiles for the as-deposited and annealed samples. The

profile obtained from the as-deposited sample was in good agreement with the prescribed metal

layer thicknesses. Oxygen was detected at the Ti/W2B interface, in the W2B layer, at the W2B/Al

interface, and in the deep Ti layer. Note that the main (and only) boron peak overlaps with one of

the tungsten peaks. The overlap results in an overestimation ofboron in the W2B film. Also note

that nitrogen was not plotted due to a peak overlap with titanium however nitrogen should only

be present in the GaN substrate.

The profile obtained from the sample annealed at 500 OC shows diffusion of titanium

through the gold layer to the surface. The concentration of oxygen in the titanium layers is higher

in this sample compared to the as deposited sample although the distributions of oxygen

throughout the profiles are similar. The profile obtained from the sample annealed at 800 OC

shows significant diffusion or inter-diffusion of layers. The profile now shows the presence of a

thin titanium layer, followed by a thin gold layer, then an oxidized aluminum layer, the W2B

layer, another oxidized aluminum layer, a gold layer, and a final titanium layer. Layer thickness

values should be considered approximate. The sputter rates are only appropriate for pure

elements or compounds. Diffusion of other species into any given layer will probably have an

impact on the sputter rate of that material. The profile obtained from the sample annealed at 1000

oC shows titanium at the surface, followed by an oxidized aluminum layer, a gold layer, the W2B

layer, another gold layer, a titanium layer, and the GaN substrate. Some of the transport might be









attributed to grain-boundary diffusion, as reported previously for Mo-based contacts to

AlGaN/GaN heterostructures 39

Figure 3-6 shows the room temperature contact resistance of the sample annealed at 800

oC, as a function of time spent at 200 oC .This simulates the operation of an uncooled GaN-based

transistor and gives some idea of the expected stability of the contact. Within experimental error,

we did not observe any degradation of contact resistance over a period of almost 3 weeks. Future

work will establish the stability of the metallization over longer periods and at higher

temperatures.

Zirconium Boride Based Ohmic Contacts

Figure 3-7 A shows the contact resistance as a function of annealing temperature, while the

associated GaN sheet resistance under the contact is shown at the bottom of the figure. The as-

deposited contacts were rectifying, with a transition to ohmic behavior for anneal temperatures >

500 oC. The contact resistance decreased up to ~700 OC, reaching a minimum value of 3x10-6

GZ.cm2. This same basic trend is seen in most Ti/Al-based contactS 19-22 and is attributed to

formation of low resistance phases of TiN at the interface with the GaN. By comparison with the

usual Ti/Al/Pt/Au metal stack, the ZrB2-based contacts show improved edge acuity, an important

factor for small gate length HEMTs in order to reduce the possibility of shorting of the ohmic

metal to the gate. The ZrB2-based contacts show a double minimum in contact resistance versus

annealing temperature and even at 1000 OC exhibit a contact resistance below 10-5 Ocm2

Figure 3-8 shows the specific contact resistivity (A) and sheet resistance (B) as a function

of anneal time at 700 oC for Ti/Al/ZrB2/Ti/Au on n-GaN. The minimum contact resistance is

achieved for 60 s anneals. This is also consistent with the need to form a low resistance

interfacial phase whose formation kinetics is probably limited by diffusion of Ti to the GaN

interface. Figure 3 -9 shows the measurement temperature dependence of the Ti/Al/ZrB2/Ti/Au









on n-GaN annealed at 700 OC. We did not observe any significant temperature dependence,

suggesting that at this anneal temperature the dominant current transport mechanism is field

emission 4

Figure 3-10 shows scanning electron microscopy (SEM) images of the as-deposited

contact morphology and after annealing at 500,700 or 1000 oC. Even at the highest anneal

temperature, the morphology remains quite smooth on the scale accessible to the SEM. This is in

sharp contrast to the case of Ti/Al/Pt/Au, where significant roughening occurs above 800 oC and

this result suggests that the ZrB2 is an effective barrier for reducing intermixing of the contact

compared to Pt.

Figure 3-11 shows the AES surface scans as a function of anneal temperature. As expected,

only carbon, oxygen and gold were detected on the as-deposited surface. The carbon is

adventitious and the oxygen comes from a thin native oxide on the Au. After annealing at 500

oC, titanium and aluminum was detected on the contact surface and their concentrations

increased at higher annealing temperatures. The surface concentration of gold decreased with

increasing temperature of the annealing step. After 1000 oC annealing, Boron was also present on

the surface, having outdiffused from the buried ZrB2 layer. Table 3-1 summarizes the near-

surface composition data obtained from AES measurements.

Figure 3-12 shows the AES depth profiles for the as-deposited and annealed samples. The

profile obtained from the as-deposited sample was in good agreement with the prescribed metal

layer thicknesses. Oxygen was detected it the ZrB2 layer, COnsistent with past observations that

the borides are excellent getters for water vapor during deposition 47. The profile obtained from

the sample annealed at 500 OC shows extensive diffusion of titanium through the gold layer to

the surface. The intermixing of the contact metallurgy becomes more pronounced as the









annealing temperature is increased. Some of the transport might be attributed to grain-boundary

diffusion, as reported previously for Mo-based contacts to AlGaN/GaN heterostructures 39. As

noted earlier, the contact morphology does not degrade significantly even after 1000 oC

annealing.

Summary and Conclusions

Both Ti/Al/W2B/Ti/Au and Ti/Al/ZrB2/Ti/Au metallization scheme were used to form

ohmic contacts to n-type GaN. For Tungsten Boride based contact, a minimum specific contact

resistivity of 7x10-6 QZ.cm2 WaS achieved at an annealing temperature of 800 oC, which is

comparable to that achieved with conventional Ti/Al/Pt/Au on the same samples. For Zirconium

Boride based contact, a minimum specific contact resistivity of 3x10-6 QZ.cm2 WaS achieved at an

annealing temperature of 700 oC, which is comparable to that achieved with conventional

Ti/Al/Pt/Au on the same samples. The ZrB2-based contact appears to have greater thermal

stability than the conventional metallization. The Tungsten Boride based contacts showed no

change in resistance over a period of more than 450 hours at 200 OC. This approach of using

boride-based contacts looks promising for high temperature device applications.

Table 3-1. Near-surface composition of contact stack determined by AES measurements for ZrB2
ohmic contact
Sample ID C(1) O(1) Al(1) S(1) Ti(2) B(2) Au(3)
Sensitivity factors [0.076] [0.212] [0.000] [0.652] [0.000] [0.000] [0.049]
#1: As deposited 48 1 Nd 2 nd nd 49
#2: Annealed at 500 OC 41 22 16 <1 3 nd 17
#3: Annealed at 700 OC 36 29 22 <1 5 nd 8
#4: Annealed at 1000 oC 28 33 18 1 4 15 Nd











s GaN/Ti/Al/W2B/Ti/Au
2.5x105 -~ a/iAMt/iA


2.0x10-s


1.5x10-s


1.0x10-s


5.0x10-6


0.0 .....
500 600 700 800 900 1000

Temperature (oC)


Figure 3-1. Specific contact resistivity versus anneal temperature for Ti/Al/W2B/Ti/Au on n-
GaN.












5.5 -GaN/TilAl/W B/TilAu






S4.0-



F~ 3.5


2.5
500 600 700 800 900 1000
A
Temperature(oC)


0.090
E` IGaN/TilAl/H 2B/Tillu
S0.085-

S0.080-

S0.075-

~i0.070-

S0.065-

S0.060-

0.055
500 600 700 800 900 1000

Temperature(oC)
B


Figure 3-2. Measurement as a function of annealing temperature Ti/Al/W2B/Ti/Au on n-GaN. A)
Transfer resistance. B) Sheet resistance.









10-4
GaN/Ti/Al/W2B/Ti/Au





S10-6









0 10 20 30 40 50 60 70 80 90 100 110

Measurement Temperature (oC)


Figure 3-3. Specific contact resistance versus measurement temperature for Ti/Al/W2B/Ti/Au on
n-GaN annealed at 800 OC.




















I~YB~I~A B~-i iI












C


Figure 3-4. Secondary electron images of the Ti/Al/W2B/Ti/Au contacts on n-GaN. A) As-
deposited. B) Annealed at 500 OC. C).Annealed at 800 OC. D).Annealed at 1000 oC




















































Wafer #3 annealed 800oC


100 l 100
90 Au Al 90 9 G
80ggA 80
O 7 0 CO7
60 360C Ti
S50 Ga s

40 d 40
S30 ~ 1B 30
20 O O A 20 Ti B


O 50 100 150 200 200 300 500 1000 1500 2000 2500 3000 3500

Sputter Depth (A) Sputter Depth (A)
Wafer# 1 as deposited Wafer #2 annealed at 500oC




800

50
O $ 70
ii 40 6
30, i 50
Q) E Ga
O 20 O
u O 30 L
10 20
O3 1

0 500 1000 1500 2000 2500 3000 3500 400

SuSputter Depth (pute eph A D


Wafer #4 annealed 1000oC


Figure 3-5. AES depth profiles of the Ti/Al/W2B/Ti/Au on n-GaN. A) As-deposited. B)
Annealed at 500 OC. C).Annealed at 800 OC. D).Annealed at 1000 oC.








10-4

--GaNITilAI/W B/TilAu
u2







S10-6



0 5 *10 15 20
Time (Days)

Figure 3-6. Contact resistance of the Ti/Al/W2B/Ti/Au on n-GaN, initially annealed at 800 OC, as
a function of subsequent time at 200 OC.












c~
E 1%1011
O
a

U
t
ert
c
Fo
I
V)
Q)
rr: ln10~
Y
O
u
E
o
U
o

u
10~6
Ez,
r~


GaN/TilAl/ZrB /TilAu


















500 600 700 800 900 1000


Temperature (oC)


GaN/TilAl/ZrB2/Ti/Au





`


~L4
CI] 7


6
o
t:

m


d'
c, 3


V1


500 600 700 800 900 1000


1100


Temperature(oC)


Figure 3-7. Measurement versus anneal temperature for Ti/Al/ZrB2/Ti/Au on n-GaN. A) Specific
contact resistivtiy. B) Sheet resistance












S10 .
u GaN/TilAl/ZrB2/TilAu
v 700oC annealed




Q) 5




Q~10 -




,a 30sec 60sec 90sec

Annealing Time (sec)

8.0
GaN/TilAl/ZrB /TilAu
7.5 -2
700oC annealed
~-7.0-


S6.0-
gj5.5-




r~4.0-
3.5
30sec 60sec 90sec
Annealing Time



Figure 3-8. Measurement as a function of annealing time at 700 OC for Ti/Al/ZrB2/Ti/Au on n-
GaN. A) Specific contact resistivtiy. B) Sheet resistance









104


GaN/Ti/Al/ZrB2/Ti/Au





S10-s




10-6


550 500 450 400 350 300 250

Measurement Temperature (K)


Figure 3-9. Specific contact resistance versus measurement temperature for Ti/Al/ZrB2/Ti/Au on
n-GaN annealed at 800 OC.





Figure 3-10. Secondary electron images of the Ti/Al/ZrB2/Ti/Au on n-GaN. A) As-deposited. B)
Annealed at 500 OC. C) Annealed at 700 OC. D) Annealed at 1000 oC.












4000

2000 As-deposited




-200 OAu Au Au


-4000 ~1Atomic % Au
Au Au 49.1 Au
-6000
C C 47.6
-8000
400 800 1200 1600 2000 2400
Kinetic Energy (eV)


8000

6000


500oC annealed


4000


/s2000

--""":\ ( u / Atomic % YAl AuA
Ti Au 17.0
-4000 02.

-6000 Al17

-8000 0 .
~,,,JC 41.2
-100 400 800 1200 1600 2000 2400

Kinetic Energy (eV)


6000

4000~

2000


-000


1000oC Annealed





Aoic% y U
o 32.7 B
C 28.0 Al
Al 17.8
B 15.0
Ti 3.9
P 1.2
OS 1.0
Ag 0.5


-1 I I100001 D
400 800 1200 1600 2000 2400 400 800 1200 1600 2000 2400
Kinetic Energy (eV) Kinetic Energy (eV)


Figure 3-11i. AES surface scans of the Ti/Al/ZrB2/Ti/Au on n-GaN. A) As-deposited. B)
Annealed at 500 OC. C) Annealed at 700 OC. D) Annealed at 1000 oC.












100
90 500 oC annealed
Au Ga

70 AI
60
B
50

40 Z Ti
30-
20 L\Ti
10

0 500 1000 1500 2000 2500 3000
Sputter Depth (A)


~.90
S80
C~70
ct60
S50
40
S30
20
S10


500 1000 1500 2000 2500

Sputter Depth (A)


500 1000 1500 2000 2500 3000
Sputter Depth (A) ,


1000 2000 3000 4000 50(
Sputter Depth (A)D


Figure 3-12. AES depth profiles of the Ti/Al/ZrB2/Ti/Au on n-GaN. A) As-deposited. B)
Annealed at 500 OC. C) Annealed at 700 OC. D) Annealed at 1000 oC.









CHAPTER 4
COMPARISON OF ELECTRICAL AND RELIABILITY PERFORMANCE OF TiB2, CrB2
AND W2B5 BASED OHMIC CONTACTS ON N-GaN

Introduction

One of the remaining obstacles to commercialization of GaN high electron mobility

transistors (HEMTs) power amplifiers is the development of more reliable and thermally stable

ohmic contactS 13,15,20-64 .These power amplifiers show great potential for radar and

communication systems over a broad frequency range from S-band to V-band 13-15,50. GaN-based

HEMTs can operate at significantly higher power densities and higher impedance than currently

used GaAs deviceS 15, 51-63. A key aspect of operation of nitride-based HEMTs at high powers is

the need for temperature-stable high quality ohmic contactS 20-30, 32, 33, 35-45, 64. The most common

ohmic metallization for AlGaN/GaN HEMTs is based on Ti/Al. This bilayer must be deposited

with over-layers of Ni, Ti or Pt, followed by Au to reduce sheet resistance and decrease

oxidation during the high temperature anneal needed to achieve the lowest specific contact

resistivity 20-30, 32, 33, 35-45, 64. There is a lateral flow issue with these contacts due to the low

melting temperature viscous A1Au4 phase that may cause problems when the gate/source contact

separation is small. For improving the thermal stability of ohmic contacts, there is interest in

high temperature metals such as W 37, 38, 40, 45, WSiX 37-39, Mo 16, V 39 and Ir 40, 42, 43. An

unexplored class of potentially stable contacts is that of boride-based contacts such as CrB2, TiB2

and W2B5. Some of the properties of these metals are shown in Table 4-1. They have high

melting temperatures, good electrical conductivity, the heat of formation for stoichiometric

borides is comparable to silicides or nitrideS 47 and although there is a lack of information on

phase diagrams with GaN, these metals have shown corrosion resistance against molten metals

and thus should exhibit even less solubility in the solid state.









In this chapter a report on the initial results for annealing temperature dependence of

contact resistance and contact intermixing of Ti/Al/boride/Ti/Au on n-type GaN, with W2B5,

TiB2 and CrB2 as the three different selected borides is presented. These contacts show promising

long-term stability at 350 oC which is very important for devices which are going to be used in

uncooled and/or prolonged heated environment.

Experimental

The samples used were 3 Clm thick Si-doped GaN grown by Metal Organic Chemical

Vapor Deposition on c-plane Al203 substrates. The electron concentration obtained from Hall

measurements was ~7x101s cm-3. Mesas 1.8 Cpm deep were formed by Cl2/Ar Inductively

Coupled Plasma Etching to provide electrical isolation of the contact pads. A metallization

scheme of Ti (200 A+)/Al (1000 A+)/ Boride (500 A+) / Ti (200 A+) /Au (800 A+) was used in these

experiments where Borides were W2B5, CrB2 and TiB2. All of the metals were deposited by Ar

plasma-assisted rf sputtering at pressures of 15-40 mTorr and rf (13.56 MHz) powers of 200-250

W. The contacts were patterned by liftoff and annealed at 500-1000 oC for 1 min in a flowing N2

ambient in a RTA furnace.

Auger Electron Spectroscopy (AES) depth profiling of the as-deposited contacts showed

sharp interfaces between the various metals in both types of contacts. The AES system was a

Physical Electronics 660 Scanning Auger Microprobe. The electron beam conditions were 10

keV, 1 CIA beam current at 300 from sample normal. For depth profiling, the ion beam conditions

were 3 keV Ar 2.0C1A (3 mm) 2 raster. Prior to AES data acquisition, secondary electron

microscopy images (SEMs) were obtained from the sample. The SEMs were obtained at

magnifications of 125X, and 1,000X. The SEMs were used to locate and document analysis area









locations and to document surface morphology. The quantification of the elements was

accomplished by using the elemental sensitivity factors.

The contact properties were obtained from linear transmission line method (TLM)

measurements on 100 x100 Clm pads with spacing 5, 10, 20, 40, and 80 lm. The contact

resistance Rc was obtained from the relation 49 Rc = (RT PS- d/Z)/2, where RT is the total resistance

between two pads, ps is the sheet resistivity of the semiconductor under the contact, d is the pad

spacing, and Z is the contact width. The specific contact resistance, pc, is then obtained from pc =

RcLTZ, where LT is the transfer length obtained from the intercept of a plot of RT vs d.

Results and Discussion

Figure 4-1 shows the contact resistances as a function of anneal temperature for the three-

boride-based schemes. The as-deposited contacts showed rectifying behavior, as expected for

any as-deposited contacts on the wide bandgap GaN 5. The current-voltage characteristics

became Ohmic for anneal temperatures > 500 oC. The contact resistance decreased up to ~ 800-

900 oC, depending on the boride employed. The TiB2-COntaining contacts show the lowest

contact resistance of 1.6x 10-6 QZ.cm2. The minimum in the contact resistance with annealing

temperature is most likely related to the formation of low resistance phases of TiN at the

interface with the GaN, as reported for conventional Ti/Al/Pt/Au contactS 20, 22-26, 28, 30, 32 and 64

Annealing at higher temperatures leads to higher contact resistance, which as will be seen later

corresponds to extensive intermixing of the contact metallurgy. The contact properties did not

show a significant dependence on annealing time at 8500 C, as shown in Figure 4-2.

Figure 4-3 shows the measurement temperature dependence of the contacts on n-GaN

annealed at 850 oC. Within the error of the measurement, the contacts exhibited almost constant

specific contact resistance in the temperature range of25-200 oC, indicating that current flow is









dominated by tunneling. When the tunneling dominates, the specific contact resistivity (RSCR) is

dependent upon doping concentration and is given as


RSR c exp [ ( ) (4-1)

where $B is the barrier height, as the semiconductor permittivity, mp the effective mass of

electrons, h the Planck' s constant and ND is the donor concentration in the semiconductor. The

tunneling may be related to the formation of the TiNx phases, as is the case with conventional

Ohmic contacts on n-GaN.

Figure 4-4 shows the SEM of the contact morphology for the three metallization schemes

before and after annealing at 800 or 1000 oC. The morphology is featureless until 800 OC, which

corresponds to the minimum in contact resistance. By 1000 oC, the morphology becomes

rougher and this corresponds to the increased contact resistance

AES surface scans showed only the presence of carbon, oxygen and gold on the as-

deposited surface. The carbon is adventitious and the oxygen comes from a thin native oxide on

the Au. Figure 4-5 shows the AES depth profiles for the as-deposited and annealed

Ti/Al/CrB2/Ti/Au samples. The profile obtained from the as-deposited sample was in good

agreement with the prescribed metal layer thicknesses. The profile from the sample annealed at

700 oC shows diffusion of titanium through the gold layer to the surface and of the Ti layer near

the GaN through the Al above it. The profile obtained from the sample annealed at 800 OC shows

significant inter-diffusion of layers. The profile now shows the presence of a thin titanium layer,

followed by a thin gold layer on top of the Cr2B layer, an oxidized aluminum layer, a gold layer,

and a final titanium layer. The profile obtained from the sample annealed at 1000 OC shows a

similar basic structure. Some of the transport might be attributed to grain-boundary diffusion, as

reported previously for Mo-based contacts to AlGaN/GaN heterostructures 39









Similar data is shown in Figures 4-6 and 4-7 for TiB2 and W2Bs for various anneal

temperatures. Once again, the movement of the two Ti layers is the maj or effect present in both

types of contact. The W2B5 is the least thermally stable of the schemes, as j udged by the more

extensive intermixing at lower temperatures.

Figure 4-8 shows the room temperature contact resistance of the samples annealed at 800

oC, as a function of time spent at 3 50 oC. This simulates the operation of an uncooled GaN-based

transistor and gives some idea of the expected stability of the contact. We have also included

data from conventional Ti/Al/Ni/Au contacts for comparison. Note that the latter shows the

lowest initial contact resistance, but then has an increase of almost an order of magnitude after 9

days of elevated temperature operation. By sharp contrast, all of the boride-based contacts show

less change with aging time and have lower contact resistances than the Ti/Al/Ni/Au after 22

days aging at 3 50 OC. This suggests that the improved stability of the borides relative to Ni has

some beneficial effect on the long-term stability of the contacts.

Summary and Conclusions

A Ti/Al/X/Ti/Au metallization scheme, where X was eitherW2Bs, CrB2 or TiB2, WaS used

to form Ohmic contacts to n-type GaN. A minimum contact resistance of 1.5x10-6 QZ.cm2 WaS

achieved for the TiB2 based scheme at an annealing temperature of 850-900 oC. For W2Bs the

minimum contact resistance was ~1.5x10"5 G.cm2 at 800 oC while for CrB2 it was 8x10-6 QZ.cm2

at 800 oC.The latter value is comparable to that achieved with conventional Ti/Al/Ni/Au on the

same samples. The contacts showed much less change in resistance over a period of more than

22 days at 350 oC than for Ti/Al/Ni/Au and the boride-based contacts look promising for high

temperature device applications.





Table 4-1. Selected properties of potential boride


contacts on GaN.
W2B
~2670


Properties
Melting Point
(oC)
Structure
Thermal expansion
coefficients x 106
(/deg)
Phonon component
of heat conduction
(wt/m-deg.)
Elastic modulus
Ex 10- (kg/cm2)
Characteri sti c
Temperature (oK)
Density of
electronic states g x
10-2
(eV-1 cm l)
Work function (eV)
(approx.)
Heat of Formation
(Kcal/mol e)
Lattice constant(A)
Thermal
conductivity(W.m
1K1
Electrical
resi stivity(CLOhm.c
m)


TiB2
2980
~3225
hexagonal

4.6


20.6


5.6

1100


4.50


4.19(?)

71.4

3.028

26


28


ZrB2
3040
~3200
hexagonal


W2Bs
~2385


CrB2
2200


hexagonal hexagonal


10.5


10.4


18.9


726


54.6


4.76


3.94(?)


3.18(?)


76.0

3.169


31.0

2.969

32


21


2.982


unknown












-m- GaN/TilAl/W2 5/TilAu


-A GaN/TilAl/TiB /TilAu


c~l
E
O

d:
Q)
O
E

m
r(
m
Q)


O

E
O
U
O
ce~
r(
O
Q)
a


10-4














105


Temperature (oC)


Figure 4-1. Specific contact resistance of Ti/Al/boride/Ti/Au ohmic contacts on n-GaN as a
function of anneal temperature.


I


500 600 700 800 900 1000









5, 10 -4
d: I -m- GaN/Til~~l/W2 5/Ti~u
Q) -*- GaN/TilAl/CrB2/TilAu
at -r- GaN/TilAl/TiB2/TilAu

10
*


6n
10L *'
30609

Aneln Tm sc

Fiur4-.SeiccotcreitneoTiA/oieT/uhmccnatonnGNaa
fucino neltm tteotmmaneltmeauefrec yeo ea
scheme









-m- GaN/Ti/Al/W2B /Ti/Au
-*GaN/Ti/Al/CrB2/Ti/Au
-A-r GaN/Ti/Al/TiB2/T i/Au


10


10 -








450 420 390 360 330 300 270

Measurement Temperature (K)

Figure 4-3. Specific contact resistance of Ti/Al/boride/Ti/Au ohmic contacts on n-GaN as a
function of measurement temperature at the optimum anneal temperatures.









CrB2


TiB2


W2B5








W2B5 Based As-Deposited

C


C


TiB2, Based 1000oC
Annealed


TiB2 Based As-Deposited

B


TiB2 Based 800oC
Annealed


W2B5 Based 800oC
Annealed


CrB2 Based 8000C
Annealed


D


CrB2 Based 10000C
Annealed


W2B5 Based 1000oC
Annealed


Figure 4-4. SEM micrographs. A) As-deposited Cr2B. B) As-deposited TiB2. C) As-deposited
W2B5. D) Cr2B annealed at 800 OC. E) TiB2 annealed at 800 OC. F) W2B5 annealed at
800 oC. G) Cr2B annealed at 1000 oC. H) TiB2 annealed at 1000 oC. I) W2B5 annealed
at 1000 oC.


CrB2 Based As-Deposited

A











lo u0~f As-depositec

o AI
cGa
7 T Ti

0 50-
o 40 N
O sot (Cr
E 2o
S o
0 500 1000 1500 29 50
Sputter Depthpp (A)


a 90~ Annealed at 7000C
C 80

~ 70AI ,Ga
S60 B r

C40
0 so TI
2o T

0
0 1000 2000 3000 4000 5000
Sputter Depth (A~)


100
o 90 Annealed at 8000 -
80 -
To Ti G
css n B Au Ti

40O
Cr AI

20 -
O 10 -
o -
0 100 200 300 400 500 00700 800
Sputter Depth (A)


100
o 90 ~ Annealed at looo"
r 80-
70 -Ti
c, 60 A




20
O 10
oi
0 200 400 600 800
Sputter Depth (A~)
D


Figure 4-5. AES depth profiles of CrB2-based contacts. A) As-deposited. B) Annealed at 700 OC.
C) Annealed at 800 OC. D) Annealed at 1000 oC.










100 .
~] 9o Au As-deposited
S80AI
S70 ~ *
S60C \F /
Pi so -
S 40 -

a15~11 Ti Ti
*-20
O10
Ga
0 1000 2000 3000 4000 5000 6000
Sputter Depth (A)


100
~.90 600oC annealed
S80-A
70 -0
60
d soY AI a
40
0 so
O Ti_ Ti T'
S20
0 lo AI

0 1000 2000 3000 4000 5006000 7000
Sputter Depth ,W( )


55
~`so 1000oC annealed
C 45~ -Ga


S30
S25
O 20
15
lo O
O 5-

0 1000 2000 3000 4000 5000
Sputter Depth (A~)
D


0 1500 3000 4500 6000 7500 9000
Sputter Depth (A)
C


Figure 4-6. AES depth profiles of TiB2-based contacts. A) As-deposited. B) Annealed at 600 OC.
C) Annealed at 800 OC. D) Annealed at 1000 oC.










100
~.90 -A As;-depoid
C 80o \Ti AI
S70 \ Ga
6i oWTi

40 --
S3o ) B
.2 20
E 2
O 10 --
o
0 500 1000 1500 2000 2500 3000
Sputter Depth (~)


100
S90~ 500"C alnnae


so
aw W Ti
Q)50
40
O s

" 2o T
O 10
o
0 500 1000 1500 2000 2500 3000 3500
Sputter Depth (A~)


100 .,
So 9 80oC~ annealed
80 -
o
7 o- a
L 60 CAu
Q)50

340 T
( 0t BN~ AI
*-20
E
O 10 -
o -
0 1000 2000 3000 4000 5000
Sputter Depth (~)


100
S9oC 1000oC annealed
F 80-

so


50 -

O B
S20
O10

0 1000 2000 3000 4000 5000 6000
Sputter Depth (A~)


Figure 4-7. AES depth profiles of W2BS-based contacts. A) As-deposited. B) Annealed at 500
oC. C) Annealed at 700 OC. D) Annealed at 1000 oC.












a




m


10 -




10 7


15 18 21 24 27
(Days)


Figure 4-8. Specific contact resistance of the boride-based contacts annealed at 8000C and the
conventional Ti/Al/Ni/Au contacts as a function of aging time at 350 oC.


3 6 9 12
Time









CHAPTER 5
ZrB2 AND W2B SCHOTTKY DIODE CONTACTS ON N-GaN

Introduction

AlGaN/GaN high electron mobility transistors (HEMTs) have potential application in

microwave power amplifiers for radar and communication systems over a broad frequency range

from S-band to V-band .One critical requirement for commercialization of these systems is the

need for more reliable and thermally stable schottky contacts on n-type GaN 13-18,25,26,30-40. The

anticipated operation of these amplifiers under uncooled, high temperature conditions

emphasizes that the thermal stability of the contact metallurgy is paramount. An alternative

approach for the gate contact is the use of metal-oxide-semiconductor (MOS) gates, though

much of that work is in its infancy and the metal gate is still the dominant technology. The

HEMT device structure is relatively simple and the reliability is determined by the stability of

gate and source/drain contacts and surface and buffer layer trapping effects. In GaN as in other

compound semiconductor systems, the strength of interfacial reactions between the metal and

semiconductor plays a key role in determining the quality of the resultant schottky barriers. The

most common schottky metallization for AlGaN/GaN HEMTs is based on Pt/Au or Ni/Au.

Metallurgy systems with high melting temperatures and good thermodynamic stability such as W

(eg.W, WSix) 37, 38, 65, 66 show potential for improved thermal characteristics on GaN. Tungsten-

based schemes have been used for both rectifying (W/Ti/Au; WSix/Ti/Au) 35, 36, 65-67 and ohmic

(Ti/Al/Pt/ W/Ti/Au) 40 COntacts on GaN HEMTs. Sputter-deposited pure W schottky contacts on

n-GaN show as-deposited barrier heights ($b) of 0.80 eV for optimized conditionS 68. Subsequent

annealing at 500-600 oC reduces the barrier height to ~0.4 eV, its theoretical value from the

relation Ob= m-Xs (where #m is the metal work function and Xs the electron affinity of GaN).









Another promising metallization system is based on borides of Cr, Zr, Hf, Ti or W 46

Stoichiometric diborides have high melting temperatures (eg. 3200 oC for ZrB2) and

thermodynamic stabilities at least as good as comparable nitrides or silicideS 47. These have been

suggested as metal gates in Si complementary metal oxide semiconductor (CMOS) integrated

circuitS 47.One particularly attractive option is ZrB2, which has a low resistivity in the range 7-10

CLD.cm. To date, there is basically no information on the contact properties of ZrB2 on n-GaN

and W2B which is a refractory material has also not been explored as a contact to GaN. Recently,

it was shown that hexagonal ZrB2 (0001) single crystals have an in-plane lattice constant close to

that of GaN, prompting efforts at GaN heteroepitaxy on ZrB2 69-72 or buffer layers on Si

substrates 73. In terms of contacts on GaN, the only related work is the study of ZrN/Zr/n-GaN

ohmic structures 48 in which the Zr/GaN interface was found to have excellent thermal stability.

A potential drawback for application of ZrB2 COntacts on GaN is that the barrier height might be

lower than the more conventional schemes, given the work function of ZrB2 is ~3.94 eV and the

electron affinity of GaN is ~4. 1 eV.

In this chapter a report on the annealing temperature dependence of barrier height and

contact intermixing of ZrB2/Ti/Au and W2B/Ti/Au on n-GaN is presented. The ZrB2 based

contacts show a maximum barrier height of 0.55 eV and maintain a barrier height >0.5 eV to at

least 700 oC and the W2B based contacts show a maximum barrier height of 0.5 eV, with a

negative temperature coefficient, and are stable against annealing up to ~500 OC.

Experimental

The samples used were 3 Clm thick Si-doped GaN grown by Metal Organic Chemical

Vapor Deposition on c-plane Al203 substrates. The electron concentration obtained from Hall

measurements was ~ 5x1017 cm-13. A schottky metallization scheme of ZrB2 (500 A+) / Ti (200 A+)









/Au (800 A+) was used in all experiments. The Au was added to lower the contact sheet

resistance, while the Pt is a diffusion barrier. All of the metals were deposited by Ar plasma-

assisted rf sputtering at pressures of 15-40 mTorr and rf (13.56 MHz) powers of 200-250 W. The

contacts were patterned by liftoff and annealed at temperatures up to 700 OC for 1 min in a

flowing N2 ambient in a RTA furnace. For ohmic contacts, we used the standard e-beam

deposited Ti/Al/Pt/Au annealed at 850 oC for 30 secs prior to deposition of the schottky

metallization. A ring contact geometry for the diodes was employed. Figure 5-1 A shows a

scanning electron microscopy (SEM) image of the as-deposited W2B based contacts.

Auger Electron Spectroscopy (AES) depth proofing of the as-deposited contacts showed

sharp interfaces between the various metals. The AES system was a Physical Electronics 660

Scanning Auger Microprobe. The electron beam conditions were 10 keV, 1 CIA beam current at

300 from sample normal. For depth proofing, the ion beam conditions were 3 keV Ar 2.0C1A (3

mm)2 raster, with sputter rate of~160 A+ / minute. Prior to AES data acquisition, secondary

electron images were obtained from the sample. These images were used to locate and document

analysis area locations and to document surface morphology. The quantifieation of the elements

was accomplished by using the elemental sensitivity factors. The contact properties were

obtained from I-V characteristics of the ZrB2/Ti/Au diodes measured at 300 K and W2B/Ti/Au

diodes measured over the temperature range 25-150 oC using a probe station and Agilent 4145B

parameter analyzer. We fit the forward I-V characteristics to the relation for the thermionic

emission over a barrier 7

JA.~I2eeAe m eV
JF=A.2 x( exp( )5-1
kT nkT









where Jis the current density, A* is the Richardson' s constant for n-GaN, Tthe absolute

temperature, e the electronic charge, 4b the barrier height, k Boltzmann's constant, n the ideality

factor and V the applied voltage.

Results and Discussion

W2B Based Rectifying Contacts

Figure 5-2 shows the extracted barrier height and reverse breakdown voltage as a function

of measurement temperature for as-deposited W2B/Ti/Au contacts on n-GaN. From the data, 4b

was obtained as 0.55 eV for the as-deposited W2B at 25 OC and ~0.45 eV at 150 OC. The data can

be fit to yield a negative temperature coefficient for barrier height of 8 x10-4 eV/OC over the

range 25-150 oC. The forward I-V characteristics in each case showed the ideality factor was >

2, suggesting transport mechanisms other than thermionic emission, such as recombination. The

reverse breakdown voltage also shows a negative temperature coefficient, which may be due to

contributions from the reduced barrier height and also from the high defect density in the

heteroepitaxial GaN on sapphire. Defect-free GaN is expected to exhibit a positive temperature

coefficient for breakdown. The reverse leakage depends on both bias and temperature. From a

moderately doped sample of the type studied here, we would expect thermionic emission to be

the dominant leakage current mechanism 75.According to image-force barrier height lowering,

this leakage current density, JL can be written as


JL S expi (5-2)

where ABg is the image-force barrier height lowering, given by ", where EM is the


electric field strength at the metal/semiconductor interface and as is the permittivity. The

experimental dependence of JL on bias and temperature is stronger than predicted from Equation









5-2. The large bandgap of GaN makes the intrinsic carrier concentration in a depletion region

very small, suggesting that contributions to the reverse leakage from generation in the depletion

region are small. Therefore, the additional leakage must originate from other mechanisms such

as thermionic field emission or surface leakage.

Figure 5-3 shows the barrier height and reverse breakdown voltage as a function of

annealing temperature. For anneals at > 600 OC, the rectifying nature of the W2B contacts was

significantly degraded due to the formation of P-phase W2N, as reported previously for W and

WSix on GaN 37,65. The improvement in breakdown voltage at intermediate annealing

temperatures may be due to annealing of sputter damage in the near-surface of the GaN.

AES depth profies of the as-deposited and 700 oC annealed contacts are shown in Figure

5-4.The as-deposited layers (Figure 5-4, A) exhibit relatively sharp interfaces, consistent with the

excellent surface morphology evident in the SEM picture of Figure 5-1 A. By sharp contrast, the

depth resolution associated with the annealed sample (Figure 5-4, B) is poorer than the as-

deposited sample. The bubbling of the fi1m evident in the SEM of Figure 5-1 B might be the

source of the degradation in depth resolution. The Ti becomes oxidized upon annealing and

separate x-ray diffraction experiments showed the formation of P-phase W2N at this temperature.

Since this phase has been associated with improved ohmicity of W-based contacts on GaN 37, 65

it is no surprise that the rectifying nature of the contact is degraded at this temperature.

ZrB2 Based Rectifying Contacts

Figure 5-5 shows the I-V characteristics from the ZrB2/Ti/Au/GaN diodes as a function of

post-deposition annealing temperature. The as-deposited sample displays an almost symmetrical

characteristic, suggesting that sputter damage dominates the current transport. With subsequent

annealing even at 200 OC, the reverse breakdown voltage is improved and higher temperatures









increase the reverse current. Figure 5-6 shows the extracted barrier height and reverse breakdown

voltage as a function of measurement temperature for as-deposited Zr2B/Ti/Au contacts on n-

GaN. From the data, 4b was obtained as 0.52 eV for the as-deposited Zr2B at 25 OC, with a

maximum value of 0.55 eV after annealing at 200 OC. The barrier height stayed above 0.5 eV

until at least 700 oC anneal temperature. While higher barrier heights would be desirable for

HEMT operation, there may be applications for the ZrB2 where extended high temperature

operation is the most important requirement. In Figure 5-6, the improvement in breakdown

voltage at intermediate annealing temperatures may be due to annealing of sputter damage in the

near-surface of the GaN. The forward I-V characteristics in each case showed the ideality factor

was > 2, suggesting transport mechanisms other than thermionic emission, such as

recombination. The reverse leakage depended on both bias and annealing temperature. As

explained in the case for W2B schottky contact, for the doping level employed here, we would

expect thermionic emission to be the dominant leakage current mechanism 75. Similarly, given

the low intrinsic carrier concentration of GaN, the additional leakage must originate from

mechanisms such as thermionic field emission or surface leakage.

SEM micrographs of the contact stack morphology are shown in Figure 5-7 as a function

of anneal temperature. The contacts retain a smooth morphology even at 500 OC, where the

ohmic contacts already shown significant roughening. The corresponding AES surface scans are

shown in Figure 5-8. The as-deposited sample shows only Au on its surface, as expected. After

the 350 oC anneal, Ti is evident and its concentration is increased after the 500 oC anneal. Table

5-1 summarizes the near-surface composition data obtained from AES. The essential message

from this data is that the Ti outdiffuses and becomes oxidized on the surface.









AES depth profiles of the as-deposited 350 oC annealed and 700 oC annealed contacts are

shown in Figure 5-9. The as-deposited layers (Figure 5-9, A) exhibit relatively sharp interfaces,

consistent with the excellent surface morphology evident in the SEMs. There is a significant

amount of oxygen in the ZrB2 layer, COnsistent with past reports that the borides are getters for

residual water vapor in the ambient during sputter deposition 47. After the 700 oC anneal, the

extent of Ti outdiffusion is increased and the Ti becomes oxidized upon annealing. X-ray

diffraction experiments with both a powder system and glancing angle, G2=1o, crystal system did

not show any reaction of the ZrB2 with the GaN even at 800 OC, as shown in Figures 5-10 and 5-

1 1. This is in sharp contrast to the case of W2B contacts on GaN, where P-phase W2N is formed

for anneals at > 600 oC. In this latter case, the rectifying nature of the W2B contacts was

significantly degraded due to the formation of P-phase W2N, as reported previously for W and

WSix on GaN 37, 65. Since this phase has been associated with improved ohmicity of W-based

contacts on GaN 37, 65, the degradation of rectifying behavior of the contact is expected at this

temperature. However, the ZrB2 shows a much slower reaction with the GaN than W2B and the

XRD spectra show no formation of nitride or gallide phases.

Summary and Conclusions

In conclusion, ZrB2 and W2B exhibits a barrier height of ~0.5 eV on GaN. This is rather

low for HEMT gates, but it may have use in applications where thermal stability is more

important than gate leakage current such as HEMT gas sensors. The ZrB2/GaN interface is stable

against annealing at 800 OC and the contact stability is still determined by outdiffusion of Ti

from the ZrB2/Ti/Au stack. Both borides appear to be an efficient getter of water vapor during

sputter deposition.









Table 5-1. Near-surface composition data obtained from AES measurements.
Sample ID C(1) O(1) Ti(2) Au(3)
Sensitivity factors [0.076] [0.212] [0.000] [0.049]
As deposited 47 4 nd 49
Annealed @350 oC 48 16 3 33
Annealed @700 oC 45 31 14 10


















































Figure 5-1. SEM micrographs of W2B based schottky contacts. A) As-deposited. B) Annealed at
700 oC .The inner circle is the W2B/Ti/Au while the outer ring is the ohmic contact.











0.80 ... 4.5

0.75 T -- Barrier Height 4.0
-*- Breakdown Voltage
0.70 C 3.5

0.65 13.0

0.60 --2.5 O

S0.55 --2.0



0.40 .5



0.35 ***** 0.0
0 25 50 75 100 125 150 175

Measurement Temperature (oC)

Figure 5-2. Barrier height and reverse breakdown voltage as a function of measurement
temperature for as-deposited W2B/Ti/Au contacts on n-GaN.











0.80 7.0

-m- Barrier Height
0.75 --6.5
-*- Breakdown Volt 9e
0.70 -6.0

0.65 5.5 7

0.60 5.0

0.55 -gg .------ 4.5

0.50 14.0 r




0.40 3.0
0 100 200 300 400 500 600 700 800

Anneal Temperature (oC)

Figure 5-3. Barrier height and reverse breakdown voltage as a function of annealing temperature
for W2B/Ti/Au contacts on n-GaN.
















S60~


.u40~
o 30~
20
10 ;




100
S90




S60


S40
o 30


500 2000



Iple
:o a peak




Ga








2000 2500


500 1000 1~
Sputter Depth (A)


0 500 1000 1500
Sputter Depth (A~)


Figure 5-4. AES depth profiles of W2B/Ti/Au on GaN. A) Unannealed. B) After annealing at 700
oC.










-a- Control 000
-* 200oC
-A- 3500C uj 0.0006-
-Y 500oC E
-4 700oC <( 0.0004-

0.0002


-6 -5 -4 -3 -3 0- 1 2 3

.000 2 V(Volts)




Figure 5-5. I-V characteristics from ZrB2/GaN diodes as a function of post-deposition annealing
temperature.









0.65 16
-a- Barrier Height
0.60 -- Breakdown Voltage -14


0.50 -~ g1 ------- 5 -1





e 0.40 6

0.35 -i 4

0.30 2

0.21 5 I I 0
0 100 200 300 400 500 600 700 800

Anneal Temperature (oC)


Figure 5-6. Barrier height and reverse breakdown voltage as a function of annealing temperature
for ZrB2/Ti/Au contacts on n-GaN.


















































Sample 3: Annealed at 700oC C


Figure 5-7. SEM micrographs of ZrB2 based schottky contacts. A) As-deposited. B) Annealed at
350 oC. C) Annealed at 700 OC .The inner circle is the ZrB2/Ti/Au while the outer
ring is the ohmic contact.


Sample 1: As-deposited


I


A"














B


SSample 2: Annealed at 350oC



































Annealed at 350.C






A3AuAu Au

STi C 47.8 VAu
Ti 3.0 Au
S 0.8
O O 15.0


Kinetic Energy (eV)


6000r

4000 .

2000



-000
-60


-o" 400 800 1200 1600 2000 2400

Kinetic Energy (eV)


Kinetic Energy (eV)


Figure 5-8. AES surface scans of ZrB2/Ti/Au on GaN. A) As-deposited. B) Annealed at 350 oC.
C) Annealed at 700 OC.













100
ooAu As Deposited

soTi G
L70



a 4oC -\ B
30
.8 Zr
E 2o


0 500 1000 1500 2000
Sputter Depth (A~)




100
so 9 350.C Annealed

Ga
70
S60H Ti
S50
L 4oC B
30
E 20B
StoC

0 500 1000 1500 2000
Sputter Depth (A~)



100
so 700.C Annealed
~ Au
so-G
70 -0



T0

1 0

E 0

0 500 1000 1500 2000 2500
Sputter Depth (A)



Figure 5-9. AES depth profiles of ZrB2/Ti/Au on GaN. A) As-deposited. B) Annealed at 350 oC.
C) Annealed at 700 OC.










1000000
As-Deposited GaN
GaN
S apphire
100000-


10000-
a1/ Sapphire
1000 -


100-


10 A
20 40 60 80 100
26 (deg. )


1000000
GaN 80oC~ annealed GaN

10ooooo Sapphire

S apphire
10000 -
1 E CuK
1000 -


100-


10
20 40 60 80 100 B
26 (deg. )


Figure 5-10. Powder XRD spectrum from ZrB2 on GaN. A) Unannealed. B) After annealing at
800 oC.














ZrB,


~---~


Sample 1: As-Deposited


80


60 ZrB


40-


20-


0 A
20 30 40 50 60 70 80 90 100

2 Theta


100
ZrB
Sample 2: 800oC Annealed
80


60-
rn ZrB2

S 40-
U ZrB1

20-



020 30 40 50 60 70 80 90 100B

2 Theta


Figure 5-11. Glancing angle XRD spectra from ZrB2 on GaN. A) Unannealed. B) After
annealing at 800 OC.









CHAPTER 6
ANNEALING TEMPERATURE DEPENDENCE OF TiB2 W2Bs AND CrB2 SCHOTTKY
BARRIER CONTACTS ON N-GaN

Introduction

The availability of reproducible schottky contacts on GaN or AlGaN is critical to the

operation of AlGaN/GaN high electron mobility transistors (HEMTs) for advanced microwave

power amplifiers in radar and communication systems 13-40, 65-67. One of the maj or remaining

hurdles in commercializing reliable HEMT-based systems is the need for thermally stable

rectifying contacts, since it is anticipated that some power amplifiers may require operating

temperatures up to 300 oC. While it might be expected that metal-oxide-semiconductor (MOS)

gates would provide superior thermal stability compared to simple schottky metal gates, this

would degrade the rf performance due to the extra capacitance and MOS technology is still at a

relatively primitive stage for GaN devices. Some standard metallization systems such as Au on

AlGaN show an environmental aging effect 16. Typical schottky metallization for AlGaN/GaN

HEMTs are based on Pt/Au or Ni/Au, with the Au included to reduce the sheet resistance of the

contact and to prevent oxidation of the other metal 20-30, 32, 33, 35-45, 64. There is still a need to

investigate a wider range of thermally stable schottky contacts on GaN in the search for

alternatives to Pt/Au or Ni/Au. Other potentially more thermally stable metallization schemes

have been reported based on W or WSir 37, 38' 65, 66. These exhibit low barrier heights around 0.4-

0.5 eV. Another potential class of thermally stable contacts are those based on borides, which

have not attracted much attention for use on GaN. The stoichiometric diborides are thermally

stable with very high melting temperatures, well in excess of those of both Ni and Pt 68' 74. They

also exhibit good corrosion resistance but are susceptible to oxidation during thermal processing

68, 74. This may be countered by depositing an overlayer of a metal such as Au in the same

deposition chamber.









In this chapter, the electrical characteristics, annealing and measurement temperature

dependence of barrier height and stability of TiB2/Ti/Au, W2BS/Ti/Au and CrB2/Ti/Au contacts

on n-GaN were studied. The TiB2 COntacts show a maximum barrier height of 0.68 eV after

annealing at 350 oC. W2B5 and CrB2 are high temperature stable refractory materials which have

not been explored as a contact to GaN. These contacts show a maximum barrier height of 0.65

eV and 0.63 eV, respectively, with a small negative temperature coefficient, and are reasonably

stable against annealing to ~350 oC. This barrier height is lower than for Ni or Pt and thus one

would need to balance the need for improved thermal stability with the poorer rectifying

properties.

Experimental

The samples employed were 3 Clm thick Si-doped GaN grown by Metal Organic Chemical

Vapor Deposition on c-plane Al203 substrates. The electron concentration obtained from Hall

measurements was ~ 3x1017 cm-3 foT Samples used for TiB2 and ~ 5x1017 cm-3 foT Samples used

for W2B5 and CrB2 schottky contacts. The surfaces were cleaned by sequential rinsing in

acetone, ethanol and 10:1 H20: HCI prior to insertion in the sputtering chamber. A metallization

scheme of X (500 A+) / Ti (200 A+) /Au (800 A+) was used in all experiments where X was either

TiB2, W2B5 or CrB2. The Au was added to lower the contact sheet resistance, while the pure Ti is

a diffusion barrier. All of the metals or compounds were deposited by Ar plasma-assisted rf

sputtering at pressures of 15-40 mTorr and rf (13.56 MHz) powers of 200-250 W. The sputter

rates were held constant at 1.4 +. sec^l for all of the metals or compounds. The contacts were

patterned by liftoff of lithographically-defined photoresist and separate samples were annealed at

temperatures of 200,3 50,500 or 700 oC for 1 min in a flowing N2 ambient in a RTA furnace. For

ohmic contacts, we used e-beam deposited Ti (200 A+)/Al (800 A+)/Pt (400 A+)/Au (1500 A+)









annealed at 850 oC for 30 secs prior to deposition of the schottky metallization. A ring contact

geometry for the diodes was employed, with the schottky contacts of diameter 50-80 Cpm

surrounded by the ohmic contacts with diameter 250-300 pm. The ohmic-schottky spacing was

10 pm.

Auger Electron Spectroscopy (AES) depth profiling of the as-deposited contacts showed

sharp interfaces between the various metals. For the AES analysis, the samples were mounted on

a stainless steel puck and placed in the system load-lock. After chamber pump-down, the sample

puck was inserted into the analytical chamber and placed in front of the analyzer. The AES

system was a Physical Electronics 660 Scanning Auger Microprobe. The electron beam

conditions were 10 keV, 1 CIA beam current at 300 from sample normal. Charge correction was

performed by using the known position of the C-(C, H) line in the C 1s spectra at 284.8 eV. The

AES spectrometer was calibrated using a polycrystalline Au foil. The Au f7 2 peak position was

determined to be 84.00+0.02. For depth profiling, the ion beam conditions were 3 keV Ar ,

2.0p1A (3 mm) 2 raster. The quantification of the elements was accomplished by using the

elemental sensitivity factors. We also used Scanning Electron Microscopy (SEM) to examine

contact morphology as a function of annealing temperature.

The contact properties were obtained from I-V characteristics of the TiB2/Ti/Au,

W2BS/Ti/Au or CrB2/Ti/Au diodes measured over the temperature range 25-150 oC using a probe

station and Agilent 4145B parameter analyzer. We fit the forward I-V characteristics to the

relation for the thermionic emission over a barrier 74

JA.~I2eeAe m eV
JF=A.2 x( exp( )(6-1)
kT nkT
where Jis the current density, A* is the Richardson' s constant for n-GaN (26.4 A- cm-2-K-2)

the absolute temperature, e the electronic charge, 4b the barrier height, k Boltzmann's constant ,









n the ideality factor and V the applied voltage. The reverse breakdown voltage was defined as

the voltage at which the current density was 1 mA.cm-2

Results and Discussion

TiB2 Based Schottky Contact

I-V characteristics obtained from the diodes annealed at different temperatures are shown

in Figure 6-1. The extracted barrier height and reverse breakdown voltage as a function of

annealing temperature for TiB2/Ti/Au contacts on n-GaN derived from this data are shown in

Figure 6-2. From the data, 4b was obtained as 0.65 eV for the as-deposited (i.e. control sample)

TiB2 at 25 oC. The barrier height increases with anneal temperature up to 350 oC, reaching a

maximum value of 0.68 eV. The work function of sputtered TiB2 is HOt available, but both Ti and

B have work functions around 4.3 eV, while the electron affinity of GaN is 4. 1 eV and thus we

might expect a low intrinsic barrier height for the compound on GaN. The work function of

chemically vapor deposited TiB2 is reported to be in the range 4.75-5 eV 47, 85. We would also

expect that the contact properties of the as-deposited compound would be dominated by residual

sputter-damage from the deposition of the contacts (which tends to increase the near-surface n-

type conductivity) and once this is annealed out (at ~600 OC in this case), the intrinsic contact

properties are revealed. Higher anneal temperatures led to a reduction in barrier height, most

likely associated with the onset of metallurgical reactions with the GaN. The reverse breakdown

voltage shows a similar trend to the barrier height, going through a maximum where the barrier

height is also a maximum. The forward I-V characteristics in each case showed the ideality

factor was in the range 2-2.5, suggesting transport mechanisms other than thermionic emission,

such as recombination and surface contributions.

Figure 6-3 shows the AES surface scans from these same samples, confirming the onset of

Ti outdiffusion by 350 oC. After the 700 oC anneal, the more extensive Ti outdiffusion leads to









oxidation of the top surface of the contact. This is reflected in the summary of the near-surface

composition data in Table 6-1. Note that there is also a small amount of Ga outdiffusion from the

GaN to the surface. The carbon signal comes from adventitious carbon on the surface.

AES depth profies of the as-deposited and annealed contacts are shown in Figure 6-4. The

as-deposited layers (Figure 6-4, A) exhibit relatively sharp interfaces, consistent with the good

surface morphology evident in the SEM pictures described later. The depth resolution of the 3 50

oC annealed sample (Figure 6-4, B) is slightly poorer than the as-deposited sample, with clear

outdiffusion of Ti. After 700 oC annealing, the Ti shows more significant outdiffusion to the

surface where it becomes oxidized. The change in interface abruptness at the metal/GaN

interface suggests the change in effective barrier height at higher annealing temperatures may

result from reactions at the TiB2/GaN interface. This is consistent with the outdiffusion of Ga

seen in the AES surface scans. Note also that the oxygen signal increases on the annealed

contacts, consistent with the past observation that boride contacts are susceptible to oxidation 47'

s. This occurs even though the annealing environment was purified, filtered N2. It is not clear

what effect this has on the contact properties, although samples annealed in Ar environments and

left to cool completely before removal from the RTA furnace showed similar electrical contact

properties and thus to Birst order, a small amount of oxidation may not be that critical in changing

the contact properties.

Figure 6-5 shows SEM images of the contacts both before (A) and after annealing at either

350 (B) or 700 oC (C). The inner contact is the TiB2/Ti/Au, while the outer ring is the

Ti/Al/Pt/Au ohmic contact. To the resolution of the SEM, the morphology does not change over

this annealing range, although the oxidation of the Ti after the 3 50 and 700 oC anneal leads to a

darker appearance of the rectifying contact.









The I-V characteristics of the as-deposited contacts were measured as a function of

measurement temperature up to 150 oC. The extracted barrier height showed no measurable

temperature dependence in the range available to us, as shown in Figure 6-6. The reverse

breakdown voltage also shows very little temperature dependence. The reverse leakage was

found to depend on both bias and temperature. From a moderately doped sample of the type

studied here, we would expect thermionic emission to be the dominant leakage current

mechanism 74. The experimental dependence of reverse current on bias and temperature was

stronger than predicted from the TiB2 barrier height. The additional leakage must originate from

other mechanisms such as thermionic field emission or surface leakage since the large bandgap

of GaN makes the intrinsic carrier concentration in a depletion region very small implying

generation currents in the depletion region are small. While the initial results with TiB2 show

reasonable thermal stability, there is much more additional work that needs to done to establish

the long-term reliability of the contacts for HEMT power amplifier or other device applications.

This would include additional studies of the interfacial reactions occurring and the effect of bias

or environmental-aging in humid ambient.

W2Bs Based Schottky Contact

Figure 6-7 shows SEM image of the contacts both before (A) and after annealing at either

350 (B) or 700 oC (C). The inner contact is the W2BS/Ti/Au, while the outer ring is the

Ti/Al/Pt/Au ohmic contact. The morphology does not change tremendously over this annealing

range. More detailed information on contact reactions can be obtained from the AES

measurements. Table 6-2 shows the surface survey data. It is clear from this data that Ti shows

some outdiffusion through the Au at 350 oC and this is more significant after 700 oC anneals.

AES depth profiles of the as-deposited and annealed contacts are shown in Figure 6-8. The

as-deposited layers (Figure 6-8 A) exhibit relatively sharp interfaces, consistent with the good









surface morphology evident in the SEM picture of Figure 6-7 A. The depth resolution of the 3 50

oC annealed sample (Figure 6-8 B) is slightly poorer than the as-deposited sample. After 700 oC

annealing, the Ti shows significant outdiffusion to the surface where it becomes oxidized upon

annealing. Separate x-ray diffraction experiments showed the formation of P-phase W2N at this

temperature, as occurs with pure W (and also the related phase W2B). The W2N phase has been

associated with improved ohmicity of W-based contacts on GaN 38, 67

Figure 6-9 shows the I-V characteristics obtained from the diodes annealed at different

temperatures. The extracted barrier height and reverse breakdown voltage as a function of

annealing temperature for W2BS/Ti/Au contacts on n-GaN derived from this data are shown in

Figure 6-10. From the data, #b was obtained as 0.58 eV for the as-deposited W2B, at 25 OC .The

barrier height increases with anneal temperature up to 200 OC, reaching a maximum value of

0.65 eV. Higher anneal temperatures led to a reduction in barrier height, most likely associated

with the onset of metallurgical reactions with the GaN. The barrier height for the W2B5 is higher

than for pure W at these moderate anneal temperatures. The reverse breakdown voltage shows a

similar trend to the barrier height, going through a maximum where the barrier height is also a

maximum.

The I-V characteristics as a function of measurement temperature for as-deposited contacts

are shown in Figure 6-11. The reverse leakage was found to depend on both bias and

temperature. From a moderately doped sample of the type studied here, we would expect

thermionic emission to be the dominant leakage current mechanism 7. According to image-force

barrier height lowering, this leakage current density, JL can be written as 74


JL S exp(6-2)











wherre AB is the image-frce barr ierhight lowrin, given by ie where Esr is the


electric field strength at the metal/semiconductor interface and as is the permittivity. The

experimental dependence of JL on bias and temperature is stronger than predicted from equation

(2). The large bandgap of GaN makes the intrinsic carrier concentration in a depletion region

very small, suggesting that contributions to the reverse leakage from generation in the depletion

region are small. Therefore, the additional leakage must originate from other mechanisms such

as thermionic field emission or surface leakage.

The extracted barrier height shows only a slight negative temperature coefficient, almost

within the experimental error, as shown in Figure 6-12. The forward I-V characteristics in each

case showed the ideality factor was > 2, suggesting transport mechanisms other than thermionic

emission, such as recombination. The reverse breakdown voltage also shows a negative

temperature coefficient up to ~100 OC. Defect-free GaN is expected to exhibit a positive

temperature coefficient for breakdown 75, but we invariably observe negative temperature

coefficients in diodes fabricated with any kind of contact on heteroepitaxial GaN on sapphire

with its high defect density. The improvement in breakdown voltage at intermediate annealing

temperatures may be due to annealing of sputter damage in the near-surface of the GaN 37' 65

CrB2 Based Schottky Contact

Figure 6-13 shows the I-V characteristics obtained from the diodes annealed at different

temperatures. The extracted barrier height and reverse breakdown voltage as a function of

annealing temperature for CrB2/Ti/Au contacts on n-GaN derived from this data are shown in

Figure 6-14. From the data, 4b was obtained as 0.52 eV for the as-deposited CrB2 at 25 OC .The

barrier height increases with anneal temperature up to 200 OC, reaching a maximum value of










0.62 eV. Higher anneal temperatures led to a reduction in barrier height, most likely associated

with the onset of metallurgical reactions with the GaN. The reverse breakdown voltage shows a

similar trend to the barrier height, going through a maximum where the barrier height is also a

maximum.

AES depth profiles of the as-deposited and annealed contacts are shown in Figure 6-15.

The as-deposited layers (Figure 6-15 A) exhibit relatively sharp interfaces, consistent with the

good surface morphology evident in the SEM pictures described later. The depth resolution of

the 350 oC annealed sample (Figure 6-15 B) is slightly poorer than the as-deposited sample, with

clear outdiffusion of Ti. After 700 oC annealing, the Ti shows more significant outdiffusion to

the surface where it becomes oxidized.

Figure 6-16 shows the AES surface scans from these same samples, confirming the onset

of Ti outdiffusion by 350 oC. After the 700 oC anneal, the more extensive Ti outdiffusion leads

to oxidation of the top surface of the contact. This is reflected in the summary of the near-surface

composition data in Table 6-3.

Figure 6-17 shows SEM images of the contacts both before (A) and after annealing at

either 350 (B) or 700 oC (C). The inner contact is the CrB2/Ti/Au, while the outer ring is the

Ti/Al/Pt/Au ohmic contact. The morphology does not change tremendously over this annealing

range, although the oxidation of the contact after the 700 oC anneal leads to a darker appearance

of the rectifying contact.

The I-V characteristics were measured as a function of measurement temperature. The

extracted barrier height shows very little temperature dependence, almost within the

experimental error, as shown in Figure 6-18. The forward I-V characteristics in each case

showed the ideality factor was > 2, suggesting transport mechanisms other than thermionic









emission, such as recombination. The reverse breakdown voltage also shows very little

temperature dependence. Defect-free GaN is expected to exhibit a positive temperature

coefficient for breakdown, but we invariably observe negative temperature coefficients in diodes

fabricated with any kind of contact on heteroepitaxial GaN on sapphire with its high defect

density. The reverse leakage was found to depend on both bias and temperature. As explained for

the case of other two boride schottky, the additional leakage must originate from other

mechanisms such as thermionic field emission or surface leakage. The long-term reliability of

these contacts for the HEMT power amplifier applications needs is tested later in the thesis.

Summary and Conclusions

The main conclusions of our study may be summarized as follows:

* W2B5 produces an as-deposited (by sputtering) barrier height of ~0.58 eV on GaN and a
maximum value of 0.65 eV after annealing at 200 OC.

* TiB2 produces an as-deposited (by sputtering) barrier height of ~0.65 eV on GaN and a
maximum value of 0.68 eV after annealing at 350 oC.

* CrBS produces an as-deposited (by sputtering) barrier height of ~0.52 eV on GaN and a
maximum value of 0.62 eV after annealing at 200 OC. This is still lower than for Ni or Pt
HEMT gates, but it may have use in applications where thermal stability is more important
than gate leakage current such as HEMT gas sensors.

* The Boride/Ti/Au contacts show some outdiffusion of Ti at 350 oC and much more
significant reaction after 700 oC anneals.

* The as-deposited contacts show only a minor decrease in barrier height for measurement
temperatures up to 150 oC.

* Additional experiments need to done to establish the long-term reliability of the contacts
for the HEMT power amplifier applications.

* The contacts are quite susceptible to oxidation during thermal processing and care must be
used to minimize exposure to oxidizing ambient.










surfaces of TiB2 based schottky


Sensitivity factors [0.076] [0.212] [0.188] [0.240] [0.049]
Control sample 44 3 2 nd 52
350 oC anneal 38 18 9 1 34
700 oC anneal 37 23 11 2 16


Sample C O S Ti Au
As deposited 48 2 Nd Nd 50
350 oC anneal 48 13 1 7 31
700 oC anneal 29 41 1 22 7


Sample ID C(1) O(1) S(1) Ti(1) Ga(1) Au(3)
Sensitivity factors [0.076] [0.212] [0.652] [0.188] [0.240] [0.049]
As deposited 29 1 nd nd nd 69
Annealed at 350 oC 26 19 2 13 1 39
Annealed at 700 OC 33 32 1 17 nd 17
Jf AES does not detect hydrogen and helium and all concentrations are normalized to 100%.nd = element not
detected. AES detection limits range from 0.1 1.0 atomic percent


Table 6-1. Concentration of elements detected on the as-received
contacts (in Atom %Jf)
Sample ID C(1) O(1) Ti(2)


Ga(1)


Au(3)


Table 6-2. Concentration of elements detected on the as-received
schottky contacts in Atom %"f)


surfaces of W2B5 based


Table 6-3. Concentration of elements detected on the as-received
schottky contacts (in Atom %"f)


surfaces of CrB2 based











100


C4

E


-- Control
-*~- 200oC at
-r- 350oC at
-r- 500oC at
-+~- 700oC at


-2 -1 0 1


V(VolIts)

Figure 6-1. I-V characteristics at 25 OC of TiB2/Ti/Au on GaN as a function of post-deposition
annealing temperature.










0.80 ..
-m- Barrier Height -I 18
-*- Breakdown Voltage
070.75 -- 16

0.70 -14
-12



0.60 06 -' i

-6



0.50 1**

0 100 200 300 400 500 600 700 800

Anneal Temperature (oC)

Figure 6-2. Barrier height and reverse breakdown voltage as a function of annealing temperature
for TiB2/Ti/Au contacts on n-GaN.











4000
3000
2000
1000
0
3 -1000
-2000
-3000
-4000
-5000
-6000



6000

4000

2000

30

-2000

-4000

-6000


500 1000 1500 2000
Kinetic Energy (eV)


500 1000 1500 2000
Kinetic Energy (eV)


6000


500 1000 1500 2000
Kinetic Energy (eV)


Figure 6-3. AES surface scans of TiB2/Ti/Au on GaN. A) As-deposited. B) Annealed at 350 oC.
C) Annealed at 700 OC.












S90 Au Control
c so Ti
S70 Ga
z 60 B
50 -0
S40C N .n/
S30-



O 400 800 1200 1600
Sputter Depth (A


loo A
S90~ Au Annealed at 350oC.


70 Ga
S60 B

O 40 -T
S30-
20


0 500 1000 1500 2000
Sputter Depth (A)
B

100
S90 An Annealed at 7000
S 80-
70-
60 -B G
Pi50
S40 .N
o Ti
0 30-
20; Ti i


0 500 1000 1500 2000
Sputter Depth (A)
C


Figure 6-4. AES depth profiles of TiB2/Ti/Au on GaN. A) As-deposited. B) Annealed at 350 oC.
C) Annealed at 700 OC.



















































Figure 6-5. SEM micrographs of TiB2 based schottky contacts A) As-deposited. B) Annealed at
350 oC. C) Annealed at 700 OC .The inner circle is the TiB2/Ti/Au while the outer
ring is the ohmic contact.









0.80


24


20 X






12
O

8 ~


Barrier Height
-*- Breakdown Voltage




'\- 1


0.75 -








0.60 -


L
m


0.55


0.50 I4
0 25 50 75 100 125 150 175

Measurement Temperature (oC)

Figure 6-6. Barrier height and reverse breakdown voltage as a function of measurement
temperature for as-deposited TiB2/Ti/Au contacts on n-GaN.






















































Figure 6-7. SEM micrographs of W2Bs based schottky contacts. A) As-deposited. B) Annealed at
350 oC. C) Annealed at 700 OC .The inner circle is the W2BS/Ti/Au while the outer
ring is the ohmic contact.























Sputter Depth (A)


0 500 1000 1500
Sputter Depth (A)


2000


Figure 6-8. AES depth profiles of W2BS/Ti/Au on GaN. A) As-deposited. B) Annealed at 350 oC.
C) Annealed at 700 OC.











-= -- Control
-*- 200oC anneal
--r- 350oC anneal
- v 500oC anneal
-+~- 700oC anneal


0.0005


0.0004


0.0003


a


S0.0002 i1









v(V
Figure 6-9. I-V characteristics of W2BS/Ti/Au on GaN as a function of post-deposition annealing
temperature.










0.80 16
m Barrier Height
0.75 T -*- -Breakdown Voltage -1 14

0.70 -1 12

0.65 -1 10 $

S0.60 8




0.50 -4

0.45 -1 2

0.40 0
0 100 200 300 400 500 600 700 800

Anneal Temperature (oC)

Figure 6-10. Barrier height and reverse breakdown voltage as a function of annealing
temperature for W2B5/Ti/Au contacts on n-GaN.









.?c
*%",


RT
* @ 50oC
r @ 75oC
S@ 100oC
+@ 125oC
S@ 150oC


0.0009
0.0008
0.0007
0.0006


0.0005
ui0.0004


S0.0003
I I 0.0002

0.0001




-0.0002 V(Vol

Figure 6-11. I-V characteristics of as-deposited W2BS/Ti/Au on GaN as a function of
measurement temperature.









0.65 10
m- Barrier Height -1 9
0.60 -*- Breakdown Voltage


0.55 74a3C




0.45 -2 OcC




0.35 ** 0
0 25 50 75 100 125 150 175

Measurement Temperature (oC)
Figure 6-12. Barrier height and reverse breakdown voltage as a function of measurement
temperature for as-deposited W2BS/Ti/Au contacts on n-GaN.










0.0009
--I Control
-*- 200oC anneal 0. 0008
A- 350oC anneal 0.0007
-v- 500oC anneal 000
-+~- 700oC anneal
0.0005

S0.0004

0.0003
C ~0.0002

0.0001




V(Vo

Figure 6-13. I-V characteristics of CrB2/Ti/Au on GaN as a function of post-deposition annealing
temperature.










0.80 nil
--- Barrier Height -1
0.75~ -*-Breaktdown Voltage I- 16

0.70 14 I

0.65 1

0.60 10 a

S0.60 -0



0.45 -5 P


0.-0
0 100 200 300 400 500 600 700 800

Anneal Temperature (oC)

Figure 6-14. Barrier height and reverse breakdown voltage as a function of annealing
temperature for CrB2/Ti/Au contacts on n-GaN.























500 1000
Sputter Depth (A)


1500


0 200 400 600 800 1000 1200 1400 1600
Sputter Depth (A~)


0 500 1000 1500
Sputter Depth (A~)


Figure 6-15. AES depth profiles of CrB2/Ti/Au on GaN. A) As-deposited. B) Annealed at 350
oC. C) Annealed at 700 OC.



























500 1000 1500 2000
Kinetic Energy (eV)


-6000


500 1000 1500 2000
Kinetic Energy (eV)


500 1000 1500 2000
Kinetic Energy (eV)


Figure 6-16. AES surface scans of CrB2/Ti/Au on GaN. A) As-deposited. B) Annealed at 350 oC.
C) Annealed at 700 OC.





























Analysis area































ring is the ohmic contact.










0.70 ...... 8
-m- Barrier Height
0.65 -*- Break~down Voltage 11


0.60 560




S0.50



0.45 --3$


0.40 1 '.. 2
0 25 50 75 100 125 150 175

Mea sulrem en t T em ne ra ture Pon

Figure 6-18. Barrier height and reverse breakdown voltage as a function of measurement
temperature for as-deposited CrB2/Ti/Au contacts on n-GaN.









CHAPTER 7
IMPROVED LONG-TERM THERMAL STABILITY AT 3500C OF TiB2 BASED OHMIC
CONTACTS ON AlGaN/GaN HIGH ELECTRON MOBILITY TRANSISTORS

Introduction

There is significant interest in developing new metallization schemes for AlGaN/GaN

High Electron Mobility Transistors (HEMTs) intended for applications in power amplifiers and

converters with high efficiency (above 70%) for radars and communications systems, hybrid

electric vehicles, power flow control and remote sensing systems 14,87-95. The achievement of

high efficiency microwave operation at elevated temperatures is important from the viewpoint of

minimizing the weight and the volume of power stages. AlGaN/GaN HEMTs appear well-suited

to simultaneously achieving high powers, high frequencies and high efficiencies. The most

commonly used metallization scheme for source/drain contacts on these HEMTs is Ti/Al, with

over-layers of Pt, Ni or Ti and then a layer of Au to reduce oxidation problems and lower the

sheet resistance of the contact stack 16,20,22,26,28,32,36,39,44,64,96,97. These contacts produce low

specific contact resistances when annealed in the 750-900 oC range but there are concerns about

the long-term stability during high temperature operation, in part because if the metal layers

begin to intermix, a low melting temperature A1Au4 phase may form that can lead to contact

shorting at small electrode separations 44. One possible solution is to use a very high melting

point diffusion barrier in place of the Pt, Ni or Ti in the contact stack. For example, Selvanathan

et al.16,39,44,96,97 demonstrated that Ti/Al/Mo/Au contacts on n-GaN are stable at 500 and 600 oC

for 25 hours of aging, but degraded after 10 hours at 750 oC. We have shown recently that TiB2,

with a melting temperature around 3000 oC and reasonable electrical resistivity (28 CLD.cm)

shows promise as such a diffusion barrier 99

In this chapter the long-term aging characteristics at 350 oC of AlGaN/GaN HEMTs with

different combinations of TiB2-based contacts, i.e. those with Ti/Al/TiB2/Ti/Au source/drain









metal and Pt/Au or Ni/Au gates and also those with TiB2 diffusion barriers in both the

source/drain and gate contacts is studied. Compared with HEMTs with standard Ti/Al/Pt/Au

Ohmic contacts and Pt/Au gate contacts, a number of the different combinations of boride-based

contacts exhibit superior stability as judged by the change in source-drain current at zero gate

voltage and the transconductance.

Experimental

The layer structures were grown on sapphire substrates by Molecular Beam Epitaxy and

employed a low temperature AIN (300 A+ thick) buffer, 2Clm of undoped GaN grown at 750 oC

under Ga-rich conditions, 250 A+ of undoped Al0.2Gao.sN and a 30 A+ undoped GaN cap. A growth

rate of 0.5-1.0 lm-hrl was used for all depositions. Mesa isolation was achieved with Cl2/Ar

inductively coupled plasma etching (300 W source power, 40 W rf chuck power). Ohmic

contacts were formed by lift-off of e-beam deposited Ti/Al/Pt/Au subsequently annealed at 900

oC for 1 min in a flowing N2 atmosphere in an RTA furnace. A metallization scheme of Ti (200

A+)/Al (1000 A+)/ TiB2 (500 +) / Ti (200 A+) /Au (800 A+) was used for comparison in these

experiments. All of the metals were deposited by Ar plasma-assisted rf sputtering at pressures of

15-40 mTorr and rf (13.56 MHz) powers of 200-250 W. The contacts were patterned by liftoff

and also annealed at 900 OC for 1 min. The specific contact resistance derived from separate

Transmission Line Method measurements was ~2x10-6 QZ.cm2. Schottky gates (1.5 x200Clm2) Of

sputter-deposited Ni/Au, Ni/TiB2/Au, Pt/Au or Pt/TiB2/Au were also patterned by lift-off. The

HEMT layout is shown in the schematic of Figure 7-1. The HEMT dc characteristics were

measured in dc mode using a HP 4145B parameter analyzer. The rf performance of the HEMTs

was characterized with a HP 8723C network analyzer using cascaded probes.









The HEMTs were aged for a period of 25 days at 350 oC on a heater plate in air, with the

electrical characteristics measured every 2-3 days. The samples were removed from the heater

block, allowed to cool to room temperature and measured before being returned for further aging

on the heater. The temperature was measured with a thermocouple attached to the heater.

Micro-Raman scattering measurements to determine stress state in the boride-based ohmic

contacts were performed in a backscattering geometry with the 488 nm line of an Ar-ion laser.

The laser spot size was ~0.8 Clm and the laser power at the sample was ~ 9 mW. The bandgap of

GaN is larger than the incident photon energy, which minimizes laser-induced heating. Because

of its higher relative intensity in this scattering geometry and its sensitivity to stress, we selected

the E22 phonon as a probe to monitor the film stress 98

Results and Discussion

Since the contact metal reflected the laser light, it was difficult to get enough signal by

Raman Scattering to get a quantitative measure of the residual stress in the contact structure.

Figure 7-2 A shows an optical image labeled with the four positions from which Raman spectra

were acquired (Figure 7-2 B). At the edge of metal contact, we found that there was tensile stress

between the Ti/Al/TiB2/Au and the GaN, but it was of a similar magnitude to the conventional

Ti/Al/Pt/Au metallization. No peeling or other manifestations of large residual stress were

observed in any of our experiments.

Figure 7-3 shows drain-source current as a function of drain-source voltage (IDs-VDS)

characteristics from an AlGaN/GaN HEMT with conventional Ti/Al/Pt/Au source/drain contacts

and Pt/Au gate metal before and after aging for 25 days at 350 oC. These devices showed a

significant decrease (~30%) in source-drain current within even 2 days of aging and a similar

decrease (~3 5%) in transconductance. The decrease in current in some of the curves at high









voltages in the un-aged sample appears to be due to discharge of traps. The work on

Ti/Al/Mo/Au contacts on n-GaN showed that they degraded by in-diffusion of Mo and Au into

the semiconductor and also by oxidation of the contactS 16, 39, 44, 96, and 97. Previous work in this

regard on ohmic contacts on GaN have also shown that they degrade by intermixing of the

contact metallurgy, probably aided in some cases by grain boundary transport.

Figures 7-4 and 7-5 shows similar IDs-VDs characteristics from HEMTs with the

Ti/Al/TiB2/Ti/Au ohmic contacts and either Pt/Au or Pt/TiB2/Au gates (Figure 7-4) or Ni/Au or

Ni/TiB2/Au gates (Figure 7-5). Several of these combinations show superior thermal stability

over the 25 day aging period to the devices with conventional metallization. The boride-based

Ohmic contacts retained a smoother morphology than the Ti/Al/Pt/Au both before and after

aging, as measured by both atomic force microscopy and optical microscopy. An example is

shown in the optical microscope images of Figure 7-6.

Figure 7-7 summarizes the data for percentage change in saturated drain-source current as

a function of aging time for all of the HEMTs with different combinations of contacts. All of the

devices with boride-based contacts show superior aging characteristics compared to the

conventional devices, with the exception of the HEMTs with Pt/TiB2/Au gate contacts. The

devices with Pt/Au gate metal and Ti/Al/TiB2/Ti/Au source/drain contacts show only a 10%

decrease in IDs after 25 days at 3 50 OC. Note that all of the devices appear to show a decrease in

current that saturates at different levels depending on the metal. This may indicate that the

reaction between the GaN and the gate and source/drain metals is limited by both the

thermodynamics of each system and by the thickness of the reacting metal.

Figure 7-8 shows rf data for the devices with conventional Pt/Au gate contacts and

Ti/Al//Pt/Au ohmic contacts before aging (A) and for a device with Pt/Au gates and









Ti/Al/TiB2/Ti/Au ohmics after (B) 25 days aging at 350 oC. The cutoff frequency, fr, remains at

~5.5 GHz with the maximum frequency of oscillation, fMAX, at ~-30 GHz in both cases. After

aging the conventional HEMT, we could not obtain reproducible rf data due to difficulties in

making consistent contact to the roughened metal surface.

Summary and Conclusions

Preliminary aging data on AlGaN/GaN HEMTs with TiB2 diffusion barriers in the

source/drain contacts show a higher resistance to degradation than devices with conventional

Ti/Al/Pt/Au contacts. Much more work is needed to determine the contact degradation

mechanisms and their activation energies and whether aging under bias makes a difference in the

contact reliability. The borides are known to be susceptible to oxidation but the presence of the

capping layers may reduce the significance of this issue.





Drain Drain
I I I I L

Source Gdin Source


Device Geometry
L,= 1.5 pm", Lgd+Ls+L, = 18 pm,
Width= 200 pm


Figure 7-1. HEMT layout used in these experiments.





-50

-40

-30

-20

-10




3 0






50


-60 -40 -20 0
X ( pm)


20 40 60


1
2
3







i"4


Raman Shift (cm-1)


Figure 7-2. Study of Raman spectra from Ti/Al/TiB2/Ti/Au contacts on HEMT wafer. A) Optical
micrograph. B) Raman spectra.










400


300 Day
*Day 25


200-


100~~ V=-


=-2
V =-2

0 24 6 810


VDS (Volts)


Figure 7-3. IDs-VDs characteristics from HEMT with conventional Pt/Au gate contacts and
Ti/Al/Pt/Au source/drain contacts before and after aging at 350 oC for 25 days.























E =V =_-2


0246810~




VDS (Volts)


Gate: Pt/TiB /Au; Source,Drain: TilAll/TiB /TilAu
a Day 0
*Day 25 V =
300-



200-



V r V =-3
V =-2


V~~ (Vlt)





Figure 7-4. IDs-VDs characteristics from HEMT with Ti/Al/TiB2/Ti/Au source/drain contacts. A)
Pt/Au gate contacts before and after aging at 350 oC for 25 days. B) Pt/TiB2/Au gate
contacts before and after aging at 350 oC for 25 days.























0 2 4 6 8 10

VDS (Volts)


0 2 4 6 8 10 B

VDS(Volts)


Figure 7-5. IDs-VDs characteristics from HEMT with Ti/Al/TiB2/Ti/Au source/drain contacts. A)
Ni/Au gate contacts before and after aging at 350 oC for 25 days. B) Ni/TiB2/Au gate
contacts before and after aging at 350 oC for 25 days.
























Regular after Day 0


Boride after Day 0 B


I I1~1 1


Regular after Day 25 C Boride after Day 25 D


Figure 7-6. Optical microscopy images of HEMTs. A) Conventional contacts before aging. B)
Boride-based source/drain contacts before aging. C) Conventional contacts after
aging at 350 oC for 25 days. D) Boride-based source/drain contacts after aging at 350
oC for 25 days.









Boride Source,Drain
-*- Pt/Au
10-
Ni/Au
-r- Pt/Boride/Au
Ni/Boride/Au

-10 'S g---+g

-20-



do-40-

-50 -i


-60-7 i i


O 2 4 6 8 10 12 14 16 18 20 22 24 26

Time (Days)





Figure 7-7. Percent change in saturated drain/source current from HEMTs with different
combinations of contact metal schemes as a function of aging time at 350 oC.



















S40-


20 -*



0.1 1 10 100
A
Frequency (GHz)

80



60 U
H21
~ I Gmax

S40







20 ., ,.... ~,, ,~ ,,,, B
0.1 1 10 100

Frequency (GHz)


Figure 7-8. RF performance of 1.5 x 200 ym2 gate length HEMTs. A) HEMT with conventional
metal contacts prior to aging. B) HEMT with Pt/Au gates and Ti/Al/TiB2/Ti/Au
source/drain contacts after aging at 350 OC for 25 days.









CHAPTER 8
Ir BASED SCHOTTKY AND OHMIC CONTACTS ON N-GaN

Introduction

GaN high electron mobility transistor (HEMT) power amplifiers have now entered the

commercialization stage for use in wireless communications and military applications. There are

also possible applications in improved automotive radar and power electronics for hybrid electric

vehicles and in advanced satellite communication systems. GaN HEMTs can provide the high-

power, high-efficiency, high-linearity RF power transistors required in base stations for mobile

data network services 13,14,23,24,50,52-57,61-64.One of the maj or issues with some applications for

these power amplifiers is the need for very stable ohmic and schottky metal contacts, capable of

extended operation at elevated temperatures (typically 200oC or higher) 25-45,101-107. Most ohmic

contact schemes for AlGaN/GaN HEMTs use Ti/Al, with over-layers of Ni, Ti or Pt, followed by

Au to reduce sheet resistance and decrease oxidation during annealing to achieve the lowest

contact resistance needed to achieve the lowest specific contact resistivity. The formation of

TiNx phases is integral to the contact formation mechanism. One drawback is the often poor

lateral edge definition of the contacts and potential shorting to the gate contact because of flow

of the low melting temperature viscous A1Au4 phase. Similarly, the gate metal must be stable

during elevated temperature operation and alternatives to the usual Ni, Pd or Pt with overlayers

of Au are attractive. There is continued interest in use of high temperature metals such as W,

WSix, W2B5,Mo ,V,Ir, Cu, CrB2, ZrB2 and TiB2 in both schottky and ohmic contactS 25-45, 101-107

in an attempt to improve the long term stability at elevated temperatures.

In this chapter the annealing temperature dependence of contact resistance and contact

intermixing of Ti/Al/Ir/Au ohmic and Ir/Au schottky metals on n-type GaN is studied. These

contacts show promising long-term stability compared to the existing standard metal contact









schemes for n-GaN. Preliminary studies on Ir-based contacts on HEMTs have shown the

potential for improved contact performance 42, 43

Experimental

For schottky contact studies, the GaN samples consisted of 3 Clm thick Si-doped GaN

grown by Metal Organic Chemical Vapor Deposition (MOCVD) on c-plane Al203 substrates.

The electron concentration obtained from Hall measurements was ~ 5x1017 cm-3. A metallization

scheme of Ir (500 A+) /Au (800 A+) was used in all experiments. The Au was added to lower the

contact sheet resistance. For comparison to a more conventional metal scheme, we also

fabricated samples with Ni/Au contacts with the same layer thicknesses as the Ir/Au. For ohmic

contacts to these samples for schottky studies, we used the standard e-beam deposited Ti (200

A+)/Al (400 A+)/Pt (200 BA/u (800 A+) annealed at 850 oC for 30 secs prior to deposition of the

schottky metallization. A ring-contact geometry for the diodes was employed, with the schottky

contacts surrounded by the ohmic contacts. The inner contact diameter was 75 pm.

For ohmic studies, the samples used were also 3 Clm thick Si-doped GaN grown by

MOCVD on c-plane Al203 substrates. The electron concentration obtained from Hall

measurements was ~7x101s cm-3. Mesas 1.8 Cpm deep were formed by Cl2/Ar Inductively

Coupled Plasma Etching to provide electrical isolation of the contact pads. All of the metals were

deposited by Ar plasma-assisted rf sputtering at pressures of 15-40 mTorr and rf (13.56 MHz)

powers of 200-250 W. The contacts were patterned by liftoff and annealed at 500-1000 OC for 1

min in a flowing N2 ambient in a RTA furnace. Conventional Ti (200 A+)/Al (400 A+)/Ni (200

A+)/Au (800 A+) was compared with a scheme in which the Ni was replaced with Ir.

Auger Electron Spectroscopy (AES) depth profiling of the as-deposited contacts showed

sharp interfaces between the various metals in all contacts. The AES system was a Physical









Electronics 660 Scanning Auger Microprobe. The electron beam conditions were 10 keV, 1 CIA

beam current at 300 from sample normal. For depth profiling, the ion beam conditions were 3

keV Ar 2.0C1A (3 mm) 2 raster. Prior to AES data acquisition, secondary electron microscopy

images (SEMs) were obtained from the sample. The SEMs were obtained at magnifications of

125X, and 1,000X. The SEMs were used to locate and document analysis area locations and to

document surface morphology. The quantification of the elements was accomplished by using

the elemental sensitivity factors.

The schottky contact properties were obtained from current-voltage (I-V) characteristics of

the Ir/Au and Ni/Au diodes measured over the temperature range 25-150 oC using a probe station

and Agilent 4145B parameter analyzer. Fitting is done to the forward I-V characteristics to the

relation for the thermionic emission over a barrier 74

eA eV
JF = A*.T2 exp~(- )exp( )(8-1)
kT nkT
where Jis the current density, A* is the Richardson' s constant for n-GaN, Tthe absolute

temperature, e the electronic charge, 4b the barrier height, k Boltzmann's constant n the ideality

factor and V the applied voltage. For ohmic studies, the contact properties were obtained from

Circular transmission line method (CTLM) measurements on circular rings with spacing 5, 10,

15, 20, 25, 30, 35, and 40 lm. The outer diameter of the circular pads was fixed at 300 Clm and

inner diameter varied from 220 to 290 lm. The specific contact resistance, pe, was obtained form

the circular TLM measurements with the relationships 97

Rs R L o(o/L)L o( ,
R,27r o R I(Ro/ L) K (R / L,)

pc = Rs LT
where RT is the total resistance, Rs is the sheet resistance, R1 is the outer radius of the annular

gap, Ro is the inner radius of the annular gap, lo, lI, Ko, and K1 are the modified Bessel










functions, LT is the transfer length, and p, is the specific contact resistance. The sheet resistance

can be obtained by iterative mathematical process.

Results and Discussion

Schottky Contacts

Figure 8-1 shows the I-V characteristics obtained from the Ir/Au diodes annealed at

different temperatures. The extracted barrier heights as a function of annealing temperature

derived from this data are shown in Figure 8-2. From the data, 4b was obtained as 0.42 eV for the

as-deposited Ir at 25 OC .The barrier height increases with anneal temperature up to 500 OC,

reaching a maximum value of 0.55 eV in the range 500-700 oC. The barrier height of Ni/Au

contacts was of similar magnitude in this annealing range (0.52 to 0.56 eV). Higher anneal

temperatures led to high leakage currents in the Ni/Au contacts, associated with the onset of

metallurgical reactions with the GaN. By contrast, the Ir contacts did not show the onset of

leakage until anneal temperatures of >900 oC.The forward I-V characteristics showed the ideality

factor was always higher for Ni, ranging from 1.75 for anneals below 350 oC to > 2 at higher

temperatures, compared to 1.3 for Ir at anneal temperatures below 350 oC and 1.8 between 500-

700 oC.

AES depth profiles of the annealed Ir/Au and Ni/Au contacts are shown in Figure 8-3.The

as-deposited layers exhibited sharp interfaces in both cases. The depth resolution for the 350 oC

annealed samples is similar to that of the as-deposited samples. After 700 oC annealing, the

Ni/Au contact shows significant outdiffusion of Ni to the surface, corresponding to roughening

of the contact morphology. By sharp contrast, the Ir/Au contact shows very little change after

700 oC annealing.









Ohmic Contacts

Figure 8-4 shows the specific contact resistances as a function of anneal temperature for

the two ohmic metal schemes. The contact resistance decreased up to ~ 900 OC in both cases,

with a lowest contact resistance of 1.6x 10-6 QZ.cm2. The minimum in the contact resistance with

annealing temperature is most likely related to the formation of low resistance phases of TiN at

the interface with the GaN, as reported for conventional contacts previously 25-38. Annealing at

higher temperatures leads to higher contact resistance, which as will be seen later corresponds to

extensive intermixing of the contact metallurgy. The contact properties did not show a significant

dependence on annealing time at 900 OC.

Figure 8-5 shows the SEM of the contact morphology for the two metallization schemes

after annealing at 500 or 900 OC. The morphology is featureless until 900 OC, which corresponds

to the minimum in contact resistance. AES surface scans showed only the presence of carbon,

oxygen and gold on the as-deposited surface. The carbon is adventitious and the oxygen

originated from a thin native oxide on the Au. Figure 8-6 shows the AES depth profies for the

samples corresponding to the SEM images in the previous figure. The profies from the samples

annealed at 500 OC shows significantly less diffusion of Al through the gold layer to the surface

in the case of an Ir interlayer and the Ni itself is mobile at 500 OC. The profiles obtained from the

samples annealed at 900 OC shows significant inter-diffusion of all the layers.

Figure 8-7 shows the measurement temperature dependence of the contacts on n-GaN

annealed at 850 oC. The contacts showed an increased specific contact resistance in the

temperature range 320-500 K, most likely due to increased sheet resistivity of the GaN as the

carrier mobility decreases. The doping in the n-GaN is not high enough to have the current flow

dominated by tunneling. When the tunneling dominates, the specific contact resistivity (RSCR) is

dependent upon doping concentration and is basically independent of temperature, i.e.











RSR C exp [ ( )

where $B is the barrier height, as the semiconductor permittivity, mp the effective mass of

electrons, h the Planck' s constant and ND is the donor concentration in the semiconductor.

Figure 8-8 shows the room temperature contact resistance of the samples annealed at 900

oC, as a function of aging time spent at 350 oC. This simulates the operation of an uncooled

GaN-based transistor and gives some idea of the expected stability of the contact. The

conventional Ti/Al/Ni/Au has an increase in specific contact resistance approximately one and a

half orders of magnitude afterl3.5 days of elevated temperature operation. The contacts showed

high resistance (rectifying) behavior beyond this point. By sharp contrast, the Ir-containing

contacts show less change with aging time and exhibited a stable specific contact resistance of

~105 GZ.cm2 after 22 days aging at 3 50 OC. This suggests that the Ir restricts some of the contact

reaction at 3 50 oC relative to Ni and has a beneficial effect on the long-term stability of the

ohmic contacts.

Summary and Conclusions

The replacement of Ni by Ir in both ohmic and schottky contacts to n-GaN improves the

thermal stability of both types of contacts. The ohmic contacts exhibit superior stability during

aging at 3 50 OC while the schottky contacts show less intermixing of the metals after annealing

at 700 oC. The Ir may be a superior choice to boride contact schemes for GaN, since the latter are

prone to oxidation.










0.006-


Ir/Au
--- As depo 0.004-
-*- -200
-r- 350
-v- 500
0.002-
-+~- 700 <






v (V)
-0.002-
Figure 8-1. I-V characteristics from Ir/Au Schottky contacts on n-GaN.










Barrier Height
0.6 -- I-r/Au


0.5 ~~








0.4-



O 100 200 300 400 500 600 700 800

Anneal Temperature (oC)

Figure 8-2. Schottky barrier height for Ir/Au contacts on n-GaN as a function of annealing
temperature.


























500 1000
Sputter Depth (A)


400 800
Sputter Depth (A)


100
90
00
70
60
Q)50
040
S30
-- 20
0 10
0


500 1000
Sputter Depth (A~)


1500


500 1000
Sputter Depth (A~)


Figure 8-3. AES depth profiles. A) Ir/Au after annealing at 350 oC. B) Ir/Au after annealing at
700 oC. C) Ni/Au contacts after annealing at 350 oC. D) Ni/Au contacts after
annealing at 700 OC.









I/


II I


GaN/Ti/Al/Ir/Au
GaN/Ti/Al/Ni/Au


10 -





1060


#"" "'~'~
500 600 700 800 900 1000


Temperature (oC)


Figure 8-4. Specific contact resistance of Ti/Al//Ni/Au and Ti/Al/Ir/Au Ohmic contacts on n-
GaN as a function of anneal temperature.


1
i \





































Figure 8-5. SEM images Ir and Ni based ohmic. A) Ti/Al/Ir/Au after annealing at 500 OC. B)
Ti/Al/Ni/Au after annealing at 500 OC. C) Ti/Al/Ir/Au after annealing at 900 OC. D)
Ti/Al/Ni/Au after annealing at 900 OC.


A












100 100
90 -o 1 r-5000C 90 Ni-500oC
Ir
S 80 -80-
O O
70 --70-
60 uGa60 -N| G
Q)50 Q)50 .Au
o a
40-40 Al
O 30 -TN 0 30
o I \ o
S20 E20 T
O 10 -O 10
Q~~~ ~ ~ ~ gh --~~~~~kpeo
0 1000 2000 3000 4000 0 1000 2000 3000
Sputter Depth (Ai) Sputter Depth (A)

A B

100 100
90 Ir-9000C 0Ni-9000C
80 -- C 80-
70 AI 70-
2 60Au 60-
50 -\ 1 50 c Au
o 0-40 0- AI
S30 -Ti N 30-
"g20 o 2
o 10 .Ir ET
O 10
< Ni
0 1000 2000 3000 4000 0
0 1000 2000 3000
Sputter Depth (Ai)SuteDeh()

C D


Figure 8-6. AES depth profiles of Ir and Ni based ohmic. A) Ti/Al/Ir/Au after annealing at 500
oC. B) Ti/Al/Ni/Au after annealing at 500 OC. C) Ti/Al/Ir/Au after annealing at 900
oC. D) Ti/Al/Ni/Au after annealing at 900 OC.










--- GaN/Ti/Al/Ir/Au
-*- GaN/Ti/Al/Ni/Au


10"


10 -




510 480 450 420 390 360 330 300

Measurement Temperature (K)

Figure 8-7. Specific contact resistance of Ti/Al//Ni/Au and Ti/Al/Ir/Au ohmic contacts on n-GaN
as a function of measurement temperature.









1


C,







C
Oa
C


10-4




105




10-6


0 3 6 9 1 2 1 5
Time (Days)


1 8 2 1 24


Figure 8-8. Specific contact resistance of the Ti/Al//Ni/Au and Ti/Al/Ir/Au contacts annealed at
9000C as a function of aging time at 350 oC.









CHAPTER 9
CONCLUSIONS

As discussed earlier, while further improvements in the III-V nitride materials quality can

be expected to enhance device operation, further device advances will also require improved

processing technology. Owing to their wide bandgap nature and chemical stability, GaN and

related materials present a host of device processing challenges. One of the critical area is

thermal processing, high temperature ohmic and schottky contacts which are thermally stable

and can at least sustain harsh condition which the device itself is cable of based on its intrinsic

properties.

The problem tackled in this work is reliable low resistance, high temperature operational

ohmic contact and reliable high temperature stable rectifying contacts. To this end, new material

and metallization scheme were explored which would give better ohmic and schottky contacts. In

this context, the contacts being better not only mean low ohmic contact resistance or high

schottky barrier height, as it used to mean in earlier work but it shall also mean less rouging of

contacts, sharp edge acuity, less intermixing of the metallization or even if the intermixing

occurs minimal decrease in specific contact resistance. Some contacts fabricated were tested with

prolonged heating over a hot plate or in some cases in hot oven.

In first section, ohmic contact formation on n-GaN using novel Titanium/Aluminum/

Tungsten Boride / Titanium / Gold and Titanium / Aluminum / Zirconium Boride / Titanium /

Gold metallization schemes were studied using contact resistance, scanning electron microscopy

and Auger Electron Spectroscopy measurements. For the case of Tungsten Boride based contact,

a minimum specific contact resistivity of 7x10-6 QZ.cm2 WaS achieved at an annealing temperature

of 800 oC. For the case of, Zirconium Boride based a minimum specific contact resistivity of

3x10-6 QZ.cm2 WaS achieved at an annealing temperature of 700 oC. This order of specific contact









resistance is comparable to that achieved with conventional Ti/Al/Pt/Au on the same samples.

The lowest contact resistance was obtained for 60 s anneals. The contact resistance was

essentially independent of measurement temperature, indicating that field emission plays a

dominant role in the current transport .The Ti began to out diffuse to the surface at temperatures

of ~500 oC, while at 800 OC the Al also began to intermix within the contact. By 1000 oC, the

contact showed a reacted appearance and AES showed almost complete intermixing of the

metallization. The contact resistance showed excellent stability for extended periods at 200 OC,

which simulates the type of device operating temperature that might be expected for operation of

GaN-based power electronic devices.

Keeping the contact at 2000C for prolonged duration was an important step and must be

looked as important step towards achieving the goal for all the future study in this area of contact

study.

In another section, three different metal borides (TiB2, CrB2 and W2B5) were examined for

use in Ti/Al/boride/Ti/Au ohmic contacts on n-type GaN and the reliability compared to the

more usual Ti/Al/Ni/Au metal scheme. A minimum contact resistance of 1.5x10-6 Q2~cm12 WaS

achieved for the TiB2-based scheme at an annealing temperature of 850-900 oC. For W2B5 the

minimum contact resistance was ~1.5x10"5 G.cm2 at 800 oC while for CrB2 it was 8x10-6 QZ.cm2

at 800 oC. Thus, minimum specific contact resistance obtained with TiB2 WaS approximately an

order of magnitude lower than with CrB2 and W2B5. In all cases, the minimum contact resistance

is achieved after annealing in the range 700-900 oC. The contact resistance did not change

significantly with changing temperature at which the I-V measurements were done. The TiB2 and

CrB2 COntacts retain smooth morphology even after annealing at 1000 OC. Auger Electron

Spectrosopy depth profiling indicated that formation of an interfacial TiNx layer is likely









responsible for the ohmic nature of the contact after annealing. All three boride-based contacts

were tested at even harsher condition this time, being placed over a hot plate for period of more

than 22 days at temperature of 350 oC. After extended aging the boride based contacts, in

general, show less change in specific contact resistance than Ti/Al/Ni/Au even as the intermixing

of the metallization scheme occurs.

Then, Schottky contact formation on n-GaN using a novel W2B based and Zirconium

based metallization scheme was studied using current-voltage, scanning electron microscopy and

Auger Electron Spectroscopy measurements. A maximum barrier height of ~0.55 eV was

achieved on as-deposited samples for Tungsten boride based and after 200 oC anneal for

Zirconium Boride based, with a negative temperature coefficient of 8 x10-4 eV/OC over the range

25-150 oC. The barrier height was essentially independent of annealing temperature up to 500 OC

for W2B and 700 oC for ZrB2 and decreased thereafter due to the onset of metallurgical reactions

with the GaN. The Ti began to outdiffuse to the surface at temperatures of >500 oC. In

conclusion, two borides produces only a low barrier height of ~0.5 eV on n-GaN. This is rather

low for HEMT gates, but it may have use in applications where thermal stability is more

important than gate leakage current such as HEMT gas sensors.

In the next section, the annealing temperature (25-700 oC) dependence of schottky contact

characteristics on n-GaN using TiB2, CrB2 and W2B5 based metallization scheme deposited by

sputtering are reported. The main conclusions of this study may be summarized as follows:

* W2B5 produces an as-deposited (by sputtering) barrier height of ~0.58 eV on GaN and a
maximum value of 0.65 eV after annealing at 200 OC.

* TiB2 produces an as-deposited (by sputtering) barrier height of ~0.65 eV on GaN and a
maximum value of 0.68 eV after annealing at 350 oC.

* CrB2 produces an as-deposited (by sputtering) barrier height of ~0.52 eV on GaN and a
maximum value of 0.62 eV after annealing at 200 OC. This is still lower than for Ni or Pt









HEMT gates, but it may have use in applications where thermal stability is more important
than gate leakage current such as HEMT gas sensors.

* The Boride/Ti/Au contacts show some outdiffusion of Ti at 350 oC and much more
significant reaction after 700 oC anneals.

* The as-deposited contacts show only a minor decrease in barrier height for measurement
temperatures up to 150 oC.

* Additional experiments need to done to establish the long-term reliability of the contacts
for the HEMT power amplifier applications.

* The contacts are quite susceptible to oxidation during thermal processing and care must be
used to minimize exposure to oxidizing ambient.

AlGaN/GaN High Electron Mobility Transistors (HEMTs) were fabricated with

Ti/Al/TiB2/Ti/Au source/drain ohmic contacts and a variety of gate metal schemes (Pt/Au,

Ni/Au, Pt/TiB2/Au or Ni/TiB2/Au) and subj ected to long-term annealing at 3 50 OC. By

comparison with companion devices with conventional Ti/Al/Pt/Au ohmic contacts and Pt/Au

gate contacts, the HEMTs with boride-based ohmic metal and either Pt/Au, Ni/Au or Ni/TiB2/Au

gate metal showed superior stability of both source-drain current and transconductance after 25

days aging at 350 oC.

Ir/Au schottky contacts and Ti/Al/Ir/Au ohmic contacts on n-type GaN were investigated

as a function of annealing temperature and compared to their more common Ni-based

counterparts. The Ir/Au ohmic contacts on n-type GaN with n~ 1017 cm-3 exhibited barrier

heights of 0.55 eV after annealing at 700 oC and displayed less intermixing of the contact metals

compared to Ni/Au. A minimum specific contact resistance of 1.6 x10-6 QZ.cm2 WaS obtained for

the ohmic contacts on n-type GaN with n~10"s cm-3 after annealing at 900 oC. The measurement

temperature dependence of contact resistance was similar for both Ti/Al/Ir/Au and Ti/Al/Ni/Au,

suggesting the same transport mechanism was present in both types of contacts. The Ir-based

ohmic contacts displayed superior thermal aging characteristics at 350 oC. Auger Electron










Spectroscopy showed that Ir is a superior diffusion barrier at these moderate temperatures than









LIST OF REFERENCES

Stephen A Campbell, The Science and Engineering of Microelectronic Fabrication second
edition, Oxford University Press, Oxford, NY (2001).

2. J. C. Bean, High-Speed Semiconductor Devices, edited by S.M. Sze, John Wiley & Sons,
New York, NY (1990).

3. S. M. Sze, Physics of Semiconductor Devices, John Wiley & Sons, New York, NY (1981).

4. D. C. Look, Mater. Sci., Eng. B 80, 383 (2001).

5.National Compound Semiconductor Roadmap/Basic Material Properties:
http://www.onr.navy.mil/sci~tech/informatio/12_electronics/ncsr/properties.asp, retrieved
June 2005, Gainesville, FL.

6.D. W. Palmer, The Semiconductor Information Website:
http://www. semiconductorsdirect. com, retrieved June 2005, Gainesville, FL.

7.R. Khanna, K. K. Allums, C. R.Abernathy, and S. J. Pearton, J. Kim, F. Ren, R. Dwivedi, T.
N. Fogarty, and R. Wilkins, Appl. Phys. Lett. 85, 3131 (2004)

s.S. Nakamura and G. Fasol, The Blue Laser Diode, Springer, Heidelberg (1997).

9. http://www.webelements.com/webelements/comonstxtG/aN-25617974.html,
retrieved June 2007, St Petersburg, Fl.

10. http://www.emf.co.uk/, retrieved June 2007, St Petersburg, Fl.

"1 CRC Handbook of Chemistry and Physics, edited by David R. Lide, Chemical Rubber
Publishing Company, Boca Raton, FL (2000).

12. B. E. Fortz, S. K. O'Leary, M. S. Shur, L. F. Eastman, U. V. Bhaphkar, Proc. MRS 512, 549
(1998).

13. J. Sun, H. Fatima, A. Koudymov, A. Chitnis, X. Hu, H.-M. Wang, J. Zhang, G. Simin, J.
Yang, and M. A. Khan, IEEE Electron Dev. Lett., 24, 375 (2003).

14. A. P. Zhang, L. B. Rowland, E. B. Kaminsky, J. W. Kretchmer, R. A. Beaupre, J. L. Garrett ,
J. B. Tucker, B. J. Edward, J. Foppes and A. F. Allen, Solid-State Electron 47, 821 (2003).

A. P. Zhang, L. B. Rowland, E. B. Kaminsky, V. Tilak, J. C. Grande, J. Teetsov, A.
Vertiatchikh and L. F. Eastman, J. Electron.Mater. 32, 388 (2003).

16. V. Kumar, L. Zhou, D. Selvanathan, and I. Adesida, J. Appl. Phys. 92, 1712 (2002).









Abhishek Motayed, Ravi Bathe, Mark C. Wood, Ousmane S. Diouf, R. D. Vispute, and S.
Noor Mohammad, J. Appl. Phys. 93, 1087 (2003).

I Jinwook Burm, Kenneth Chu, William A. Davis, William J. Schaff, Lester F. Eastman, and
Tyler J. Eustis, Appl. Phys. Lett. 70, 464 (1997).

19. M. E. Lin, Z. Ma, F. Y. Huang, Z. F. Fan, L. H. Allen, and H. Morkog, Appl. Phys. Lett. 64,
1003 (1994).

20. B. P. Luther, S. E. Mohney, T. N. Jackson, M. Asif Khan, Q. Chen, and J. W. Yang, Appl.
Phys. Lett. 70, 57 (1997).

21. Zhifang Fan, S. Noor Mohammad, Wook Kim, ijzgiir Aktas, Andrei E. Botchkarev, and
Hadis Morkog, Appl. Phys. Lett. 68, 1672 (1996).

22. S. Ruvimov, Z. Liliental-Weber, J. Washburn, D. Qiao, S. S. Lau, and Paul K. Chu, Appl.
Phys. Lett. 73, 2582 (1998).

23. A. N. Bright, P. J. Thomas, M. Weyland, D. M. Tricker, C. J. Humphreys, and R. Davies, J.
Appl. Phys. 89, 3143 (2001).

24. N. A. Papanicolaou, M. V. Rao, J. Mittereder, and W. T. Anderson, J. Vac. Sci. Technol. B
19, 261 (2001).

25. Q. Z. Liu and S. S. Lau, Solid-State Electronics, 42, 677-691 (1998).

26. R. Li, S. J. Cai, L. Wong, Y. Chen, K. L. Wang, R. P. Smith, S. C. Martin, K. S. Boutros, J.
M. Redwing, IEEE Electron Device Letters, 20, 323 (1999).

27. S. T. Sheppard, K. Doverspike, W. L. Pribble, S. T. Allen, J. W. Palmour, L. T. Kehias, T. J.
Jenkins, IEEE Electron Device Letters, 20, 161 -163 (1999).

28. D. Qialo, Z. F. Guan, J. Carlton, S. S. Lau, and G. J. Sullivan, Appl. Phys. Lett. 74, 2652
(1999).

29. V. Tilalk, R. Dimitov, M. Murphy, B. Green, J. Smart, W. J. Schaft, J. R. Shealy and L. F.
Eastman, Mat. Res. Soc. Symp. Proc., 622, T7.4.1 (2000).

30. S.-H. Lim, J. Washburn, Z. Liliental-Weber, and D. Qiao, Appl. Phys. Lett. 78, 3797 (2001).

31. M. Asif Khan, M. S. Shur and Q. Chen, Appl. Phys. Lett., 68, 3022 (1996).

32. S. Murai, H. Masuda, Y. Koide, and Masanori Murakami, Appl. Phys. Lett. 80, 2934 (2002).

33. Changzhi Lu, Hongnai Chen, Xiaoliang Lv, Xuesong Xie, and S. Noor Mohammad, J. Appl.
Phys. 91, 9218 (2002).









34. J. S. Foresi and T. D. Moustakus, Appl. Phys. Lett., 62, 2859 (1993).


35. M. W. Fay, G. Moldovan, P. D. Brown, I. Harrison, J. C. Birbeck, B. T. Hughes, M. J. Uren,
and T. Martin, J. Appl. Phys. 92, 94 (2002).

36. K. O. Schweitz, P. K. Wang, S. E. Mohney, and D. Gotthold, Appl. Phys. Lett. 80, 1954
(2002).

37. M. W. Cole, D. W. Eckart, W. Y. Han, R. L. Pfeffer, T. Monahan, F. Ren, C. Yuan, R. A.
Stall, S. J. Pearton, Y. Li and Y. Lu, J. Appl. Phys. 80, 278 (1996).

38. A. Zeitouny, M. Eizenberg, S. J. Pearton, and F. Ren, J. Appl. Phys. 88, 2048 (2000).

39. D. Selvanathan, F. M. Mohammed, A. Tesfayesus and I. Adesida, J. Vac.Sci.Technol. B22,
2409 (2004).

40. B.Luo, F.Ren, R. C. Fitch, J. K. Gillespie, T. Jenkins, J. Sewell, D. Via, A. Crespo, A. G.
Baca, R. D. Briggs, D. Gotthold, R. Birkhahn, B. Peres and S. J. Pearton, Appl. Phys. Lett.
82, 3910 (2003).

41. H. W. Jang and J. L. Lee, J.Appl.Phys. 93, 5416 (2003).

42. R. Fitch, J. K. Gillespie, N. Moser, T. Jenkins, J. Sewell, D. Via, A. Crespo, A. M. Dabiran,
P. P. Chow, A. Osinsky, J. R. LaRoche, F. Ren and S. J. Pearton, Appl. Phys. Lett. 84, 1495
(2004).

43. R. C. Fitch, J. K. Gillespie, N. Moser, T. Jenkins, J. Sewell, D. Via, A. Crespo, A. M.
Dabiran, P. P. Chow,A. Osinsky, J. R. LaRoche, F. Ren and S. J. Pearton, J. Vac. Sci.
Technol. B22, 619 (2004).

44. D. Selvanathan, L. Zhou, V. Kumar, I. Adesida and N. Finnegan, J.Electron.Mater. 32, 335
(2003).

45. X. A. Cao, S. J. Pearton, G. Dang, A. P. Zhang, F. Ren, and J. M. Van Hove, Appl. Phys.
Lett. 75, 4130 (1999).

46. L. Voss, R. Khanna, S.J. Pearton, F. Ren and I .I. Kravehenko, Appl. Phys. Lett. 88, 012104
(2006).

47. W. Zalgozdzon-Wosik, C. Darne, D. Radhakrishnan, I. Rusakova, P. Van der Heide, Z. H.
Zhang, J. Bennett, L. Trombetta, P. Majhi and D. Matron, Rev. Adv. Mater. Sci. 8, 185
(2004).

48. S. D. Welter, B. P Luther, S. E. Mohney, R. Karlicek and R. S. Kern, Electrochem. Solid-
State Lett. 2, 151 (1999).









49. F. A. Padovani and R. Stratton, Solid-State Electron. 9 695 (1966).


so. I. Ahmad, V. Kasisomayajula, M. Holtz, J. M. Berg, S. R. Kurtz, C. P. Tigges, A. A.
Allerman, and A. G. Baca, Appl. Phys. Lett. 84, 371-373 (2004).

5 A. P. Zhang, L. B. Rowland, E. B. Kaminsky, J. B. Tucker, J. W. Kretchmer, A. F. Allen, J.
Cook, and B. J. Edward, Electron. Lett. 39, 245 (2003).

52. W. Saito, Y. Takada, M. Kuraguchi, K. Tsuda, I. Omura, T. Ogura and H. Ohashi, IEEE
Trans Electron Dev. 50, 2528 (2003).

53. Wu Lu, V. Kumar, E. L. Piner and I. Adesida, IEEE Trans. Electron Dev. 50,1069 (2003).

54. P. Valizadeh and D. Pavlidis, IEEE Trans. Electron. Devices, 52, 933 (2005).

5 M. Hikita, M. Yanagihara, K. Nakazawa, H. Ueno, Y. Hirose, T. Ueda, Y. Uemoto, T.
Tanaka, D. Ueda and T. Egawa, IEEE Trans. Electron Devices, 52, 1963 (2005).

56. S. Nakazawa, T. Ueda, K. Inoue, T. Tanaka, H. Ishikawa and T. Egawa, IEEE Trans.
Electron Devices, 52, 2124 (2005).

5 T. Palacios, S. Rajan, A. Chakraborty, S. Heikman, S. Keller, S. P. DenBaars and U. K.
Mishra, IEEE Trans. Electron Dev. 52,2117 (2005).

5 U. K. Mishra, P. Parikh, and Y. F. Wu, Proc. IEEE 90, 1022 (2002).

59. L. F. Eastman, V. Tilak, J. Smart, B. M. Green, E. M. Chumbes, R. Dimitrov, H. Kim, O. S.
Ambacher, N. Weimann, T. Prunty, M. Murphy, W. J. Schaff, J. R. Shealy, IEEE Trans.
Electron Dev. 48, 479 (2001).

60. S. Keller, Y.-F. Wu, G. Parish, N. Ziang, J. J. Xu, B. P. Keller, S. P. DenBaars, U. K. Mishra,
IEEE Trans. Electron. Dev. 48, 552 (2001).

61. V. Adivarahan, M.Gaevski, W. H. Sun, H. Fatima, A. Koudymov, S. Saygi, G. Simin, J.
Yang, M. A. Khan, A. Tarakji, M. S. Shur and R. Gaska, IEEE Electron Device Letters
24,514 (2003).

62. A. Tarakji, H. Fatima, X. Hu, J. P. Zhang, G. Simin, M. A. Khan, M. S. Shur and R. Gaska,
IEEE Electron Dev. Lett. 24, 369 (2003).

63. S. Iwakami, M. Yanagihara, O. Machida, E. Chino, N. Kaneko, H. Goto and K. Ohtsuka,
Jpn. J. Appl. Phys.43, L831 (2004).

64. A. G. Balca, F. Ren, J. C. Zolper, R. D. Briggs, and S. J. Pearton, Thin Solid Films, 308-309,
599-606 (1997).









65. M. W. Cole, F. Ren, and S. J. Pearton, J. Electrochem. Soc. 144, L275 (1997).

66. X. A. Cao, S. J. Pearton, F. Ren, and J. R. Lothian, Appl. Phys. Lett. 73, 942 (1998).

67. J. Kim, F. Ren, A. G. Baca, and S. J. Pearton, Appl. Phys. Lett. 82, 3263 (2003).

68. R. Mehandru, S. Kang, S. Kim, F. Ren, I. Kravehenko, W. Lewis and S. J. Pearton, Mater.
Sci. Semicond. Proc. 7 95 (2004).

69. J. Suda and H. Matsunami, J. Cryst. Growth 237-239, 1114 (2002).

70. R. Liu, A. Bell, F. A. Ponce, S. Kamiyama, H. Amano and I. Akasaki, Appl. Phys. Lett. 81,
3182 (2002).

7 J. I. Iwata, K. Shiraishi and A. Oshiyama, Appl. Phys. Lett. 83, 2560 (2003).

72. P. L. Liu, A. V. G. Chizmeshya, J. Kouvetakis and I. S. T. Tsong, Phys. Rev. B72, 245335
(2005).

73. J. Tolle, R. Roucka, I. S. T. Tong, C. Ritter and J. Kouvetakis, Appl. Phys. Lett. 82, 2398
(2003).

74. D. K. Schroder, Semiconductor material and device characterization, Wiley and Sons, NY
(1990).

7 S. J. Pearton, J. C. Zolper, R. J. Shul and F. Ren, J. Appl. Phys., Applied Physics Reviews
86, 1 (1999).

76. L. Zhou, A. T. Ping, K. Boutros, J. Redwing and I. Adesida, Electron Lett. 35, 745 (1999).

7 J. Hilsenbeck,W. Reiger, E. Nebauer,W. John, G. Trankle, J. Wurfl, A. Ramakrishan and H.
Obloh, Phys. Stat. Solidi. A176, 183 (1999).

7 L. S.Yu, D. J. Qiao, Q. J. Xing, S. S. Lau, K. S. Boutros and J. M. Redwing, Appl. Phys.
Lett. 73, 3917 (1998).

79. J. Hilsenbeck, E. Nebauer, J. Wurf1, G. Tarnkel and H. Obloh, Electron. Lett. 36, 980 (2000).

so. J. Wurfl, J. Hilsenbeck, E. Nebauer, G. Trankel, H. Obloh and W. Osterle, Microelectron.
Rel. 40, 1689 (2000).

8 C. M. Jeon, H. W. Jang and J. L. Lee, Appl. Phys. Lett. 82, 391 (2003).

82. S. Arulkumaran, T. Egawa, H. Ishikawa, M. Umeno and T. Jimbo, IEEE Trans. Electron.
Dev. 48, 573 (2001).









83. E. D. Readinger and S. E. Mohney, J. Electron. Mater. 34, 375 (2005).

84. V. Kumar, D. Selvanathan, A. Kuliev, S. Kim, J. Flynn and I. Adesida, Electron. Lett. 39,747
(2003).

s R. Ranjit, W. Zagozdzon-Wosik, I. Rusakova, P. van der Heide, Z.-H. Zhang, J. Bennett,
Reviews on Advanced Materials Science 8, 176 (2004).

86. E. D. Readinger, B. P. Luther, S. E. Mohney and E. L. Piner, J. Appl. Phys. 89,7983 (2001).

s C. Chen, J. Zhang, V. Adivarahan, A. Koudymov, H. Fatima, G. Simin, J. Yang and M. Asif
Khan, Appl. Phys. Lett. 82, 4593 (2003).

s H. Xu, S.Gao, S.Heikman, S. I. Long, U. K. Mishra, and R. A. York, IEEE MWC Lett. V.16,
22-24 (2006).

89. W. Saito, T. Domon, I. Omura, M. Kuraguchi, Y. Takada, K. Tsuda and M. Yamaguchi,
IEEE Electron. Dev. Lett 27, 326 (2006).

90. G. Simin, V. Adivarahan, J Yang, A. Koudymov, S. Rai and M. Asif Khan, Electron. Lett.
41, 774 (2005).

91. I. Ahmad, V. Kasisomayajula, M. Holtz, J. M. Berg, S. R. Kurtz, C. P. Tigges, A. A.
Allerman, and A. G. Baca, Appl. Phys. Lett. 86, 173503 (2005).

92. T. Makimoto, Y. Yamauchi, and K. Kumakura, Appl. Phys. Lett. 84, 1964 (2004).

93. A. Chini, D. Buttari, R. Coffle, S. Heikman, S. Keller, and U. K. Mishra, Electron. Lett. 40,
73 (2004).

94. Z. Yang, A. Koudymov, V. Adivarahan, J. Yang, G. Simin, M. A. Khan, IEEE Microwave
and Wireless Components Lett. V.15, 850-852 (2005).

95. A. Koudymov, S. Rai, V. Adivarahan, M. Gaevski, J. Yang, G. Simin, M. A. Khan, IEEE
Microwave and Wireless Components Lett. 14 12, 560-562 (2004).

96. D. Selvanathan, L. Zhou, V. Kumar, J. P. Long, M. A. L. Johnson, J. F. Schetzina and I.
Adesida, Electron. Lett. 38, 755 (2002).

97. D. Selvanathan, L. Zhou, V. Kumar, and I. Adesida, Phys. Status Solidi A194, 583 (2002).

98. J. Kim, J. A. Freitas, J. Mittereder, R. Fitch, B. S. Kang, S. J. Pearton and F. Ren, Solid-State
Electron 50 408 (2006).

99. V. I .Matkovich, Boron and Refractory Borides (Spriger-Verlag, Berlin Heidelberg NY
(1977).










100. N. Mahadik, M. V. Rao and A. V. Davydov, J. Electron. Mater. 35, 2035 (2006).

101. L. Wang, F. M. Mohammed, and I. Adesida, J. Appl. Phys. 101, 013702 (2007).

102. F. M. Mohammed, L. Wang, I. Adesida, and E. Piner, J. Appl. Phys. 100, 023708 (2006).

103. F. M. Mohammed, L. Wang, and I. Adesida, Appl. Phys. Lett. 88, 212107 (2006).

104. A. Basu, F. M. Mohammed, S. Guo, B. Peres, and I. Adesida, J. Vac. Sci. Technol. B 24, L16
(2006).

'os. F. M. Mohammed, L. Wang, and I. Adesida, Appl. Phys. Lett. 87, 262111 (2005).

106. L. Wang, F. M. Mohammed, and I. Adesida, J. Appl. Phys. 98, 106105 (2005).

107. F. M. Mohammed, L. Wang, D. Selvanathan, H. Hu, and I. Adesida, J. Vac. Sci. Technol. B
23, 2330 (2005).

1os. G. S. Marlow and M. B. Das, Solid-State Electron. 25, 91 (1982).

109. J. Kasahara, M. Arai and N. Watanabe, J. Appl. Phys. 50, 541 (1979).

110. C. B Vartuli, S. J. Pearton, C. R. Abernathy, J. D. MacKenzie, E. S. Lambers and J. C.
Zolper, J. Vac. Sci. Technol. B14, 3523 (1996).

111 M. K. Puchert, P. Y. Timbrell and R. N. Lamb, J. Vac. Sci. Technol. A14, 2200 (1996).

112. M. Sugawara, Plasma Etching: Fundamentals and Applications, Oxford University Press,
New York, NY (1998).

113. R. J. Shul, Plasma Etching and RIE: Fundamentals and Applications, AVS Short Course
Program, Orlando, FL (2004).

114. H. Ryssel and I. Ruge, lon Implantation, John Wiley & Sons, New York, NY (1986).

'15 P. Van Zant, Microchip Fabrication, McGraw-Hill, New York, NY (2000).

116. S. Mahajan and K. S. Sree Harsha, Principles of Growth and Processing of Semiconductor,
McGraw-Hill, Boston, MA (1999).

117 C. R. Brundle, C. A. Evans, Jr. and S. Wilson, Encyclopedia of Materials Characterization:
Surfaces, Interfaces, Thin Films, Butterworth Heinemann, Stoneham, MA (1992).

'Is J. G. E Gardeniers, Z. M. Rittersma and G. J. Burger, J. Appl. Phys. 83, 7844 (1998).









119. Kyu-Pil Lee, Design and fabrication of GaN-based hetrojunction bipolar transistor, Ph.D.
dissertation, University of Florida, Gainesville (2003).

120. Kelly Pui-Sze Ip, Process development for ZnO-based devices, Ph.D. dissertation, University
of Florida, Gainesville (2005).









BIOGRAPHICAL SKETCH

Rohit Khanna was born on 25th Feburary, 1981 in Lucknow, Uttar Pradesh, India. He grew

up and spent his high school years in Varanasi, Uttar Pradesh, India. On graduating from high

school in 1999, he secured a position in Indian Institute of Technology-Joint Entrance

Examination (IIT-JEE), earning an admission to the prestigious Institute of Technology, Banaras

Hindu University (IT-BHU), Varanasi, India.

He obtained his Bachelor of Technology (B.Tech.) from the Department of Ceramic

Engineering at IT-BHU in 2003. Then he applied for graduate studies and was accepted in the

Department of Materials Science and Engineering (MSE) at UCLA and UFL (USA). In fall 2003

he j oined the doctoral program at the MSE at the University of Florida. He j oined Prof. Dr.

Pearton's research group from spring 2004. While continuing his doctorate program under Prof.

Dr. Pearton, he was offered a job with Oerlikon USA Inc (earlier Unaxis and Plasma Therm)

which he j oined as an Associate Applications Lab Engineer in January 2007.





PAGE 1

1 DEVELOPMENT OF HIGH TEMPERATURE STABLE OHM IC AND SCHOTTKY CONTACTS ON N Ga N By ROHIT KHANNA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DE GREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2 2007 Rohit Khanna

PAGE 3

3 To my family

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my advisor, Prof. Stephen J. Pearton, the most important person throughout my graduate studies, for all the opportunities, guidance, motivation, and support. I would also like to thank my committee members, Prof. Cammy R. Abernathy, Prof. David P. Norton, Prof. Fan Ren, and Prof. Rajiv Singh, for their time, expertise, and evaluation. Prof Pearton has been my m entor in true sense. He provided me numerous opportunities to give presentations, encouraged logical, problem solving thinking and gave me confidence whenever I needed it. Prof. Pearton helped me develop wide ranging skills in semiconductor processing area for which I am really thankful to him. I can not thank him enough. I thank Prof. Ren for providing me with useful comments and directions in order to improve my research work. I am grateful to them for their advice that has helped me grow professionally. I would like to thank group members o f Prof. Pearton, Prof. Ren, Prof. Abernathy Prof Norton, and Prof Singh research groups Kwang Baik, Kelly Ip, Lars Voss, Jon Wright, Wantae Lim, Rishabh Mehandru, Soohwan Jang, Byong Kang, Hung Ta Wang, Travis. J And erson, J.J Chen, Luc Stafford, Brent Gila, Seemant Rawal, Karthik Ramani Mark Hlad and countless others for their assis tance and friendship and who have made grad uate school enjoyable learning experience. Especially, I am grateful to have as a very good friend, Kao Chil Joe, who was a visiting scholar form Taiwan and I thank him for helping me during my first year as graduate student. I also like to thank University of Florida Nano Fabrication Lab staff, Ivan and Bill as I enjoyed working with them. I wou ld also like to thank Paula, our group secretary for all the support she has given me. I also want to thank all my friends of school and college with whom I share special memories forever in my life. I also want to thank my ex roommate s who made graduate l if e an enjoyable experience, especially Shiv for having a lot of interesting and intellectual debates.

PAGE 5

5 Most importantly, I express my deepest gratitude to my family especially to my parents (my father, Prem Kumar Khanna and mother, Madhu Khanna) whose love and sacrifice for me is beyond anything I will ever understand and my brother (Bhaskar) and sisters (Ela and Rashmi) for loving and supporting me unconditionally. I thank my family for making me who I am tod ay.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 9 LIST OF FIGURES ................................ ................................ ................................ ....................... 10 ABSTRACT ................................ ................................ ................................ ................................ ... 15 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 18 2 BACKGROUND AND LITERATURE REVIEW ................................ ................................ 23 Properties of GaN ................................ ................................ ................................ ................... 23 Overview ................................ ................................ ................................ ......................... 23 Crystal S tructure and Basic Properties ................................ ................................ ............ 23 Metal Contact ................................ ................................ ................................ .................. 25 Schottky c ontact ................................ ................................ ................................ ....... 26 Ohmic c ontact ................................ ................................ ................................ .......... 29 Thermal s tability ................................ ................................ ................................ ...... 32 Common Processing Techniques ................................ ................................ ............................ 33 Dry Plasma Etching ................................ ................................ ................................ ......... 33 Ion Implantation ................................ ................................ ................................ .............. 36 Rapid Thermal Annealing ................................ ................................ ............................... 37 Characterization Techniques ................................ ................................ ................................ .. 37 Atomic Force Microscopy ................................ ................................ ............................... 37 Auger Electron Spectroscopy ................................ ................................ .......................... 38 X ray Photoelectron Spect roscopy ................................ ................................ .................. 38 Electrical Measurements ................................ ................................ ................................ 38 Photoluminescence (PL) ................................ ................................ ................................ .. 39 Rutherford Backscattering Spectrometry/Channeling ................................ ..................... 39 Scanning Electron Microscopy ................................ ................................ ........................ 40 Secondary Ion Mass Spectrometry ................................ ................................ .................. 40 Stylus Profilometry ................................ ................................ ................................ .......... 41 3 TUNGSTEN AND ZIRCONIUM B ORIDE BASED OHMIC CONTACTS TO N Ga N .... 53 Introduction ................................ ................................ ................................ ............................. 53 Experimental ................................ ................................ ................................ ........................... 54 Results and Discussion ................................ ................................ ................................ ........... 55 Tungten Boride Based Ohmic C ontact ................................ ................................ ............ 55 Zirconium Boride Based Ohmic C ontacts ................................ ................................ ....... 58 Summary a nd Conclusions ................................ ................................ ................................ ..... 60

PAGE 7

7 4 COMPARISON OF ELECTRICAL AND RELIABILITY PERF ORMANCE OF Ti B 2 Cr B 2 AND W 2 B 5 BASED OHMIC CONTACTS ON N Ga N ................................ .............. 73 Introduction ................................ ................................ ................................ ............................. 73 Experimental ................................ ................................ ................................ ........................... 74 Results and Discussion ................................ ................................ ................................ ........... 75 Summary and Conclusions ................................ ................................ ................................ ..... 77 5 Zr B 2 AND W 2 B SCHOTTKY DIODE CONTACTS ON N Ga N ................................ ......... 87 Introduction ................................ ................................ ................................ ............................. 87 Exper imental ................................ ................................ ................................ ........................... 88 Results and Discussion ................................ ................................ ................................ ........... 90 W 2 B Based Rectifying C ontacts ................................ ................................ ...................... 90 ZrB 2 Based Rectifying C ontacts ................................ ................................ ...................... 91 Summary and Conclusions ................................ ................................ ................................ ..... 93 6 ANNEAL ING TEMPERATURE DEPENDENCE OF Ti B 2 W 2 B 5 AND Cr B 2 S CHO TTKY BARRIER CONTACTS ON N Ga N ................................ ............................. 106 Introduction ................................ ................................ ................................ ........................... 106 Experimental ................................ ................................ ................................ ......................... 107 Results and Discussion ................................ ................................ ................................ ......... 109 TiB 2 Based Schottky C ontact ................................ ................................ ........................ 109 W 2 B 5 Based Schottky C ontact ................................ ................................ ....................... 111 CrB 2 Based Schottky C ontact ................................ ................................ ........................ 113 Summary and Conclusions ................................ ................................ ................................ ... 115 7 IMPROVED LONG TERM THERMAL STABILI TY AT 350 o C OF Ti B 2 BASED OHMIC CONTACTS ON AlGaN/Ga N HIGH ELECTRON MOBILITY ........................ 135 Introduction ................................ ................................ ................................ ........................... 135 Experimental ................................ ................................ ................................ ......................... 136 Results and Discussion ................................ ................................ ................................ ......... 137 Summary and Conclusions ................................ ................................ ................................ ... 139 8 Ir BASED SCH OTTKY AND OHMIC CONTACTS ON N Ga N ................................ ...... 148 Introduction ................................ ................................ ................................ ........................... 148 Experimental ................................ ................................ ................................ ......................... 149 Results and Discussion ................................ ................................ ................................ ......... 151 Schottky Contacts ................................ ................................ ................................ .......... 151 Ohmic Contacts ................................ ................................ ................................ ............. 152 Summary and Conclusions ................................ ................................ ................................ ... 153 9 CONCLUSIONS ................................ ................................ ................................ .................. 162 LIST OF REFERENCES ................................ ................................ ................................ ............. 167

PAGE 8

8 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 175

PAGE 9

9 LIST OF TABLES Table page 2 1 Electrical properties of Si GaAs and GaN. ................................ ................................ ....... 41 2 2 The physical parameters in different semiconductor materials ................................ ......... 42 2 3 Ionization energy of impurities for wurtzite GaN. ................................ ............................. 42 2 4 Basic physical properties of GaN. ................................ ................................ ..................... 43 2 5 Metal work function and ideal barrier heights for GaN (electron affinity: 4.1 eV) .......... 43 3 1 Near surface composition of contact stack determined by AES measurements for ZrB 2 ohmic contact ................................ ................................ ................................ ............ 60 4 1 Selected properties of potential boride con tacts on GaN. ................................ .................. 78 5 1 Near surface composition data obtained from AES measurements. ................................ 94 6 1 Concentration of elements detected on the as received surfaces of TiB 2 based ................................ ................................ ....................... 116 6 2 Concentration of elements detected on the as received surfaces of W 2 B 5 based ................................ ................................ ........................ 116 6 3 Concentration of elements detected on the as received surfaces of CrB 2 based ................................ ................................ ....................... 116

PAGE 10

10 LIST OF FIGURES Figure page 2 1 Crystal structure of wurtzite GaN. ................................ ................................ ..................... 44 2 2 The III V compound semiconductor tree ................................ ................................ ........... 44 2 3 Structure of a HEMT ................................ ................................ ................................ ......... 45 2 4 Previous study of schottky contacts A) Index of interface behavior S as a function of the electronegativity difference of the sem iconductors. ................................ .................... 46 2 5 Lithography pattern for Schottky diode ................................ ................................ ............. 47 2 6 Lithography pattern for linear TLM A) TLM pads. B) Plot for measurement. ................. 47 2 7 An ICP reactor. ................................ ................................ ................................ .................. 48 2 8 Electric and magnetic fields inside the reactor. ................................ ................................ 48 2 9 Chemical etching process. A) Generation of reactive species. B) Diffusion of reactive neutrals to surface. ................................ ................................ ................................ 49 2 10 Physical etching process. A) Gener ation of reactive species. B) Acceleration of ions to the surface. C) Ions bombard surface ................................ ................................ ............ 49 2 11 Combination of chemical and physical etching process A) Generation of reactive species. ................................ ................................ ................................ ............................... 50 2 12 Ion implantation system. ................................ ................................ ................................ .... 50 2 13 Simplified principle of AFM. ................................ ................................ ............................ 51 2 14 Auger Process. A) An isolated atom. B) Inner core level electron dislodged. leaving behind a vacancy. C) An outer level electron fills the vacancy ................................ ......... 52 3 1 Specific contact r esistivity versus anneal temperature for Ti/Al/W 2 B/Ti/Au on n GaN. ................................ ................................ ................................ ................................ ... 61 3 2 Measurement as a function of annealing temperature Ti/Al/W 2 B/Ti/Au on n GaN. A) Transfer resistance. B) Sheet resist ance. ................................ ................................ ............ 62 3 3 Specific contact resistance versus measurement temperature for Ti/Al/W 2 B/Ti/Au on n GaN annealed at 800 C. ................................ ................................ ................................ 63 3 4 Secondary electron images of the Ti/Al/W 2 B/Ti/Au contacts on n GaN. A) As deposited. B) Annealed at 500 C. C) Annealed at 800 C D) Annealed at 1000 C ...... 64

PAGE 11

11 3 5 AES depth profiles of the Ti/Al/W 2 B/Ti/Au on n GaN. A) As deposited. B) Annealed at 500 C. C). Annealed at 800 C D). Annealed at 1000 C .......................... 65 3 6 Contact resistance of the Ti/Al/W 2 B/Ti/Au on n GaN, initially annea led at 800 C as a function of subsequent time at 200 C ................................ ................................ ....... 66 3 7 Measurement versus anneal temperature for Ti/Al/ZrB 2 /Ti/Au on n GaN. A) Specific contact resistivtiy. B) Sheet resistance ................................ ................................ ............... 67 3 8 Measurement as a function of annealing time at 700 C for Ti/Al/ZrB 2 /Ti/Au on n GaN. A) Specific contact resistivtiy. B) Sheet resistance ................................ .................. 68 3 9 Specific contact resistance versus measurement temperature for Ti/Al/ZrB 2 /Ti/Au on n GaN annealed at 800 C ................................ ................................ ................................ 69 3 10 Secondary electron images of the Ti/Al/ZrB 2 /Ti /Au on n GaN. A) As deposited. B) Annealed at 500 C C) Annealed at 700 C D) Annealed at 1000 C ............................ 70 3 11 AES surface scans of the Ti/Al/ZrB2/Ti/Au on n GaN. A) As deposited. B) Annealed at 500 C C) Annealed at 700 C D) Annealed at 1000 C ............................ 71 3 12 AES depth profiles of the Ti/Al/ZrB 2 /Ti/Au on n GaN. A) As deposited. B) Annealed at 500 C C) Annealed at 700 C D) Annea led at 1000 C ............................ 72 4 1 Specific contact resistance of Ti/Al/boride/Ti/Au Ohmic contacts on n GaN as a function of anneal temperature. ................................ ................................ ......................... 79 4 2 Specific contact resistance of Ti/Al/boride/Ti/Au Ohmic contacts on n GaN as a function of ................................ ................................ ................................ .......................... 80 4 3 Specific contact resistance of Ti/Al/boride/Ti/Au Ohmic contacts on n GaN as a function of measurement temperature at the optimum anneal temperatures. .................... 81 4 4 SEM micrographs. A) As deposited Cr 2 B. B) As deposited TiB 2 C) As deposited W 2 B 5 D) Cr 2 B annealed at 800 C ................................ ................................ ................... 82 4 5 AES depth profiles of CrB 2 based contacts. A) As deposited. B) Annealed at 700 C C) Annealed at 800 C D) Annealed at 1000 C ................................ ............................. 83 4 6 AES depth profiles of TiB 2 based contacts. A) As deposited. B) Annealed at 600 C C) Annealed at 800 C D) Annealed at 1000 C ................................ ............................. 84 4 7 AES depth profiles of W 2 B 5 based c ontacts. A) As deposited. B) Annealed at 500 C C) Annealed at 700 C D) Annealed at 1000 C ................................ ....................... 85 4 8 Specific contact resistance of the boride based contacts annealed at 800 o C and the con ventional Ti/Al/Ni/Au contacts as a function of aging time at 350 C ....................... 86

PAGE 12

12 5 1 SEM micrographs of W 2 B based schottky contacts. A) As deposited. B) Annealed at 700 C The inner circle is the W 2 B/Ti/Au while the outer ring is the Ohmic contact. ... 95 5 2 Barrier height and reverse breakdown voltage as a function of measurement temperature for as deposited W 2 B/Ti/Au contacts on n GaN. ................................ .......... 96 5 3 Barrier height and reverse breakdown voltage as a function of annealing temperature for W 2 B/Ti/Au contacts on n GaN. ................................ ................................ ................... 97 5 4 AES depth profiles of W 2 B/Ti/Au on GaN. A) Unannealed. B) After annealing at 700 C ................................ ................................ ................................ ............................... 98 5 5 I V characteristics from ZrB 2 /GaN diodes as a function of post deposition annealing temper ature. ................................ ................................ ................................ ....................... 99 5 6 Barrier height and reverse breakdown voltage as a function of annealing temperature for ZrB 2 /Ti/Au contacts on n GaN. ................................ ................................ ................. 100 5 7 SEM micrographs of ZrB 2 based schottky contacts. A) As deposited. B) Annealed at 350 C C) Annealed at 700 C ................................ ................................ ...................... 101 5 8 AES surface scans of ZrB 2 /Ti/Au on GaN. A) As deposi ted. B) Annealed at 350 C C) Annealed at 700 C ................................ ................................ ................................ .... 102 5 9 AES depth profiles of ZrB 2 /Ti/Au on GaN. A) As deposited. B) Annealed at 350 C C) Annealed at 700 C ................................ ................................ ................................ .... 103 5 10 Powder XRD spectrum from ZrB 2 on GaN. A) Unannealed. B) After annealing at 800 C ................................ ................................ ................................ ............................. 104 5 11 Glancing angle XRD spectra from ZrB 2 on GaN. A) Una nnealed. B) After annealing at 800 C ................................ ................................ ................................ ......................... 105 6 1 I V characteristics at 25 C of TiB 2 /Ti/Au on GaN as a function of post deposition annealing temperature. ................................ ................................ ................................ ..... 117 6 2 Barrier height and reverse breakdown voltage as a function of annealing temperature for TiB 2 /Ti/Au contacts on n GaN. ................................ ................................ .................. 118 6 3 AES surface scans of TiB 2 /T i/Au on GaN. A) As deposited. B) Annealed at 350 C C) Annealed at 700 C ................................ ................................ ................................ .... 119 6 4 AES depth profiles of TiB 2 /Ti/Au on GaN. A) As deposited. B) Annealed at 350 C C) Annealed at 700 C ................................ ................................ ................................ .... 120 6 5 SEM micrographs of TiB 2 based schottky contacts A) As deposited. B) Annealed at 350 C C) Annealed at 700 C ................................ ................................ ...................... 121

PAGE 13

13 6 6 Barrier height and reverse breakdown voltage as a function of measurement temperature for as deposited TiB 2 /Ti/Au contacts on n GaN. ................................ ......... 122 6 7 SEM micrographs of W 2 B 5 based schottky contacts A) As deposited. B) Annealed at 350 C ................................ ................................ ................................ ......................... 123 6 8 AES depth profiles of W 2 B 5 /Ti/Au on GaN. A) As deposited. B) Annealed at 350 C C) Annealed at 700 C ................................ ................................ .............................. 124 6 9 I V characteristics of W 2 B 5 /Ti/Au on GaN as a function of post deposition annealing temperature. ................................ ................................ ................................ ..................... 125 6 10 Barrier height and reverse breakdown voltage as a function of annealing temperature for W 2 B 5 /Ti/Au contacts on n GaN. ................................ ................................ ................ 1 26 6 11 I V characteristics of as deposited W 2 B 5 /Ti/Au on GaN as a function of measurement temperature. ................................ ................................ ............................... 127 6 12 Barrier height and reverse breakdown voltage as a function of measurement temperature for as deposited W 2 B 5 /Ti/Au contacts on n GaN. ................................ ....... 128 6 13 I V characteristics of CrB 2 /Ti/Au on GaN as a function of post deposition annealing temperature. ................................ ................................ ................................ ..................... 129 6 14 Barrier height and reverse breakdown voltage as a function of annealing temperature for CrB 2 /Ti/Au contacts on n GaN. ................................ ................................ ................. 130 6 15 AES depth profiles of CrB 2 /Ti/Au on GaN. A) As deposited. B) Annealed at 350 C C) Annealed at 700 C ................................ ................................ ................................ .... 131 6 16 AES surface scans of CrB 2 /Ti/Au on GaN. A) As deposited. B) Annealed at 350 C C) Annealed at 700 C ................................ ................................ ................................ .... 132 6 17 SEM micrographs of CrB 2 based sch ottky contacts. A) As deposited. B) Annealed at 350 C C) Annealed at 700 C ................................ ................................ ...................... 133 6 18 Barrier height and reverse breakdown voltage as a function of measurement temperature for as deposite d CrB 2 /Ti/Au contacts on n GaN. ................................ ........ 134 7 1 HEMT layout used in these experiments. ................................ ................................ ........ 140 7 2 Study of Raman spectra from Ti/Al/TiB 2 /Ti/Au contacts on HEMT wafer. A) Optical micrograph. B) Raman spectra. ................................ ................................ ....................... 141 7 3 I DS V DS characteristics from HEMT with conventional Pt/Au gate contacts and Ti/Al/Pt/Au source/drain contacts before and after aging at 350 C for 25 days. ........... 142 7 4 I DS V DS characteristics from HEMT with Ti/Al/TiB 2 /Ti/Au source/drain contacts. A) Pt/Au gate contacts before and after aging at 350 C f or 25 days. ................................ .. 143

PAGE 14

1 4 7 5 I DS V DS characteristics from HEMT with Ti/Al/TiB 2 /Ti/Au source/drain contacts. A) Ni/Au gate contacts before and after aging at 350 C for 25 days. ................................ 144 7 6 Optical microscopy images of HEMTs. A) Conventional contacts before aging. B) Boride based source/drain contacts before aging. ................................ ........................... 145 7 7 Per cent change in saturated drain/source current from HEMTs with different combinations of contact metal schemes as a function of aging time at 350 C .............. 146 7 8 RF performance of 1.5 200 m 2 g ate length HEMTs. A) HEMT with conventional metal contacts prior to aging.. ................................ ................................ .......................... 147 8 1 I V characteristics from Ir/Au Schottky contacts on n GaN. ................................ .......... 154 8 2 Schottky barrier height for Ir/Au contacts on n GaN as a function of annealing temperature. ................................ ................................ ................................ ..................... 155 8 3 AES depth profiles. A) Ir/Au after annealing at 350 C B) Ir/Au after annealing at 700 C C) Ni/Au contacts after annealing at 350 C ................................ ..................... 156 8 4 Specific contact resistance of Ti/Al//Ni/Au and Ti/Al/Ir/Au Ohmic contacts on n GaN as a function of anneal t emperature. ................................ ................................ ........ 157 8 5 SEM images Ir and Ni based ohmic. A) Ti/Al/Ir/Au after annealing at 500 C B) Ti/Al/Ni/Au after annealing at 500 C ................................ ................................ ............ 158 8 6 AES depth profiles of Ir and Ni based ohmic. A) Ti/Al/Ir/Au after annealing at 500 C B) Ti/Al/Ni/Au after annealing at 500 C ................................ ................................ 159 8 7 Specific contact resistance of Ti/Al//Ni/Au and Ti/Al/Ir/Au Ohmic contacts on n GaN as a function of measurement temperature. ................................ ............................. 160 8 8 Specific contact resistance of the Ti/Al//Ni/Au and Ti/Al/Ir/Au contacts annealed at 90 0 o C as a function of aging time at 350 C ................................ ................................ ... 161

PAGE 15

15 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 DEVELOPMENT OF HIGH TEMPERATURE STABLE OHM IC AND SCHOTTKY CONTACTS ON N Ga N By Rohit Khanna August 2007 Chair: Stephen J. Pearton Major: Materials Science and Engineering In this work the effort was made to towards develop and investigate h igh temperature stable Ohmic and Scho ttky contacts for n type GaN. Various borides and refractory materials were incorporated in metallization scheme to best attain the desired effect of minimal degradation of contacts when placed at high temperatures. Th is work focuses on achieving a contact scheme using different borides which include two Tungsten Borides (namely W 2 B, W 2 B 5 ), Titanium Boride (TiB 2 ), Chromium Boride (CrB 2 ) and Zirconium Boride (ZrB 2 ). Further a high temperatur e metal namely Iridium (Ir) wa s ev aluated as a potential contact to n GaN, as part of continuing improved device tec hnology development. The main goal of this project was to investigate the most promising boride based contact metallurgies on GaN, and finally to fabricate a High Electro n Mobility T ransistor (HEMT) and compare it s reliability to a HEMT using present technology contact. Ohmic contacts were fabricated on n GaN using borides in the metallization scheme of Ti/Al/boride/Ti/Au. The characterization of the contacts was done usi ng current voltage measurements, scanning electron microscopy (SEM) and Auger Electron Spectroscopy (AES) measurements. The contacts formed gave specific contact resistance of the order of 10 5 to 10 6 Ohm cm 2 A minimum contact resistance of 1.5x10 6 2 was achieved for the TiB 2 based

PAGE 16

16 scheme at an annealing temperature of 850 900 C which was comparable to a regular ohmic contact of Ti/Al/Ni/Au on n GaN. When some of borides contacts were placed on a hot plate or in hot oven for temperature ranging fro m 200 o C to 350 o C, the regular metallization contacts degraded before than borides ones. Even with a certain amount of intermixing of the metallization scheme the boride contacts showed minimal roughening and smoother morphology, which, in terms of edge a cuity, is crucial for very small gate devices. Schottky contacts were also fabricated and characterized using all the five boride compounds. The barrier hei ght obtained on n GaN was ~0 5 0.6 eV which was low compared to those obtained by Pt or Ni. This bar rier height is too low for use as a gate contact and they can only have limited use, perhaps, in gas sensors where large leakage current can be tolerated in exchange for better thermal reliability. AlGaN/GaN High Electron Mobility Transistors (HEMTs) were fabricated with Ti/Al/TiB 2 /Ti/Au source/drain o hmic contacts and a variety of gate metal schemes (Pt/Au, Ni/Au, Pt/TiB 2 /Au or Ni/TiB 2 /Au) and were subjected to long term annealing at 350 C By comparison with companion devices with conventional Ti/Al/Pt/A u o hmic contacts and Pt/Au gate contact s, the HEMTs with boride based o hmic metal and either Pt/Au, Ni/Au or Ni/TiB 2 /Au gate metal showed superior stability of both source drain current and transconductance after 25 days aging at 350 C The need for sputt er deposition of the borides problem in achieving significantly lower specific contact resistance than with conventional schemes deposited using e beam evaporation. The borides also seem to be, in general, good getters for oxygen leading to shee t r esistivity issues

PAGE 17

17 Ir/Au s ch ottky contacts and Ti/Al/Ir/Au o hmic contacts on n type GaN were investigated as a function of annealing temperature and compared to their more common Ni based counterparts. The Ir/Au o hmic contacts on n type GaN with n~ 10 17 c m 3 exhibited barrier heights of 0.55 eV after annealing at 700 C and displayed less intermixing of the contact metals compared to Ni/Au. A minimum specific contact resistance of 1.6 x10 6 2 was obtained for the o hmic contacts on n type GaN with n~10 1 8 cm 3 after annealing at 900 C. The measurement temperature dependence of contact resistance was similar for both Ti/Al/Ir/Au and Ti/Al/Ni/Au, suggesting the same transport mechanism was present in both t ypes of contacts. The Ir based o hmic contacts disp layed superior thermal aging characteristics at 350 C. Auger Electron Spectroscopy showed that Ir is a superior diffusion barrier at thes e moderate temperatures than Ni.

PAGE 18

18 CHAPTER 1 INTRODUCTION The micro electronic s industry has grown rapidly in the past f our decades and now is the basis for our Information Age The first semiconductor transistor was invented by the scientists of Bell Labs in 1947. Subsequently, the concept of an Integrated Circuit (IC) was developed, requiring a high yield of working devic es that comprise the circuit. T o have a 50% probability of functionality for a 20 transistor circuit, the probability of device functionality must be (0.5) 1/20 = 0.966 or 96.6%. This was considered wildly optimistic at the time, yet today integrated circui ts are bu ilt with billions of transistor 1 This is possible because each component or a device is many times reliable compared to a component in any other industry. Even though the very first semiconductor transistor was made from germanium (Ge), silicon (Si) became the semiconductor of choice as a result of the low melting point of Ge that limits high temperature processes and the lack of a natural occurring germanium oxide to prevent the surface from electrical leakage. Due to the maturity of its fabrica tion technology, silicon continues to dominate the present commercial market in discrete devices and integrated circuits for computing, power switching, data storage and communication. For high speed and optoelectronic devices such as high speed integrated circuits and laser diodes, gallium arsenide (Ga As) is the material of choice. It exhibits superior electron transport properties and special optical properties. GaAs has higher carrier mobility and higher effective carrier velocity than Si, whic h translat e to faster devices. GaAs is a direct bandgap semiconductor, whereas Si is indirect, hence making GaAs better sui ted for optoelectronic devices. However, physical properties required for high power, high temperature electronics and UV/blue light emitter ap plications are be yond the limits of Si and GaAs. It is essential to investigate alternative materials and their growth and processing techniques in or der to achieve these devices 2, 3 119 So now the focus ha s shifted to semiconductors having wide

PAGE 19

19 bandgap s. They exhibit inherent properties such as larger bandgap, higher electron mobility and higher breakdown field strength making them suitable for high power, high temperature electronic devices and short wavelength optoelectronics. Wide bandgap semiconduct ors offer the best technical promise for high power and high temperature transistors. Until recently, the most promising of these materials was silicon carbide (SiC). However, SiC has several technical shortfalls that have opened competition to the III nit ride materials. Thermal oxides in SiC power metal oxide semiconductor field effect transistors (MOSFETs) actually limit the temperature range of application since the gate contact degrades and becomes electrically leaky at high temperatures. The low electr on mobility of only 400 cm 2 /V.s yields lower PAE (<30%) for many transistors in the frequency range of 1 to 5 GHz. For silicon and SiC, amplifier efficiency decays rapidly with increase in frequency so that it drops below 25% for many devices operating abo ve 2 GHz. GaN base devices offer wider bandgap, greater chemical inertness and higher temperature stable operation than SiC 120 Single transistor output power is the most important cost limiting issue for commercialization of solid state power devices. O ther economic factors relating to performance are power added efficiency (PAE) required for lightweight portable systems, amplifier linearity necessary to transmit digital signals without distortion or out of band modulation products, and amplifier noise f igure and phase noise. Output power achievable by microwave devices is directly proportional to the breakdown voltage and sustainable current limits. For bipolar devices under Class A operation, the maximum output power density is then P max = I sat (V cb V knee ) / 8 where I sat is the saturation current at the quiescent point, V cb is the collector breakdown voltage and V knee is the saturation voltage at maximum current. Apart from high breakdown voltage, high thermal stability, GaN based semiconductors also b enefit from a very high sustainable electron

PAGE 20

20 saturation velocity of 2.7x10 7 cm/s. This unique property, which has been shown to significantly benefit GaN FETs, is the result of large energetic displacement between valleys in the conduction band profiles 12 0 One of the most significant problems limiting single transistor high power devices is the heat dissipation required. Mature silicon RF power transistors are currently limited to about 125 C junction temperature (85 100 C ambient) with operation of lit tle more than 1 W at 10 GHz. Due to leaky oxides, SiC does not increase this range enough to result in significant advantage. GaAs technology has improved on this performance to yield 50 W at 10 GHz with state of the art power FET technology. However, both silicon and GaAs devices suffer greater high temperature de rating than is expected from the wide bandgap GaN devices. The GaN devices not only can operate at 400 C or higher but also should exhibit optimal performance somewhere near 250 C due to improv ed ionization of the carriers in the material 119 While further improvements in the III V nitride materials quality can be expected to enhance device operation, further device advances will also require improved processing technology. Owing to their wide bandgap nature and chemical stability, GaN and related materials present a host of device processing challenges, including difficulty in achieving reliable low resistance o hmic contacts, thermally stable contacts for both n and p GaN, high temperatures nee ded for implant activation, lack of efficient wet etch process, generally low dry etch rates and low selectivity over etching masks, and dry etch damage. High thermal budget and dry etch damage indirectly adds to the problem of having good reliable ohmic a nd s chottky contacts. These problems constitute a major obstacle to successful demonstration and commercialization of some GaN based devices, such as bipolar transistors and power switches, whose performance are much more affected by the immature fabricati on techniques. To fully

PAGE 21

21 exploit these device applications, a number of critical advances are necessary 4 ,119 One of the critical area is high temperature thermal processing, ohmic and s chottky contacts which are thermally stable and can at least sustain h arsh condition which the device it self is cable of base d on its intrinsic properties. The motivation of this work is to develop novel ohmic and s chottky contacts to GaN and AlGaN/GaN high electron mobility devices for use in high temperature application. So the objective of this work is to have high temperature stable ohmic and schottky contact to n GaN which should circumvent or delay the problem of intermixing of metal layers and surface roughening leading to a better and reliable contact scheme. In this project, we explore a novel metallization scheme involving borides because of the refractory nature of the borides and thus thermal stability and very little possibility of it having solid state reactions with other metals normally used in contact scheme. Apart from boride, a high temperature metal, namely Ir, was also explored as part of continuing search for better contacts. The objective is to have an optimized new contact for high temperature operation of AlGaN/GaN HEMT s The properties of GaN and back ground of semiconductor processing and characterizations especially in terms of ohmic and schottky contacts are reviewed in Chapter 2. Ohmic and s chottky contacts are necessary to impart specific electrical interactions and characteristics i n achieving ope rating devices. The studies of ohmic and s chottky contact metallization are covered in Chapters 3 through 7. The Tungsten Boride (W 2 B) and Zirconium Boride (ZrB 2 ) metal scheme was considered first. Next a comparative study of Tungsten Boride (W 2 B 5 ), Titani um Boride (TiB 2 ) and Chromium Boride (CrB 2 ) is done and is presented in chapter 4. Chapter 5 discusses the rectifying nature of Tungsten Boride (W 2 B) and Zirconium Boride. Chapter 6 shows result of Titanium Boride, Tungsten Boride (W 2 B 5 ) and Chromium Borid e in usage as a

PAGE 22

22 rectifying contact to n GaN. The demonstration of a High Electron Mobility Transistor using new best boride based metallurgy is given in Chapter 7. Chapter 8 de als with high melting temperature metal I ridium (Ir) as being explored as o hmic and s chottky contact to n GaN. The conclusion a nd summary of the study of new boride based and Ir contacts to n GaN and HEMTs is given in chapter 9

PAGE 23

23 CHAPTER 2 BACKGROUND AND LITERATURE REVIEW Properties of GaN Overview GaN is a wide bandgap semiconducto r which has numerous properties which makes it well suited for high temperature applications. Its electrical properties are compared to Si, SiC and othe r materials in Table 2 1 3, 5 7 It has a direct bandgap energy of 3.45 is transp arent to visible light and operates in ultra violet to blue wavelengths. Hall measurements at room temperatures show the Hall mobility of electron of 1000~1300 cm 2 /V s. It has saturation velocity little higher than GaAs. GaN like ZnO seems to be extremely stable at harsh e nvironment of gamma radiations It has little change in IV characteristic even after being irradiated b y high energy proton radiat ion 7 This makes GaN very good candidate for outer space and nuclear application. Sapphire or SiC substrates are generally used for growing GaN. GaN also has different hetrostuctures available with Al, In etc. (Al, Ga, In) N forms a continuous and direct band gap alloy from 1.92 eV (InN) to 6.2 eV (AlN) with potential for emission and detection in spectral range between visible an d the ultraviolet wavelengths 8 Crystal S tructure and Basic Properties GaN is a direct bandgap semiconductor having stable form as hexagonal (wurtzite) crystal structure, with lattice parameter s a = 3.189 and c = 5.178 A. The Ga (gro up III) atoms are tetrahedrally coordinated with four N (group V) atoms. Alternating Ga and N layers form the crystal s tructure 9 ( Figu re 2 1 ) GaN based semiconductors have attracted tremendous interest for their applications to blue laser and LEDs, high temperature, high power electronics, high density optical data storage, and

PAGE 24

24 electronics for the aerospace and automobile industries, telecommunication devices, and wide band gap semiconductors in power amplifiers extends the radi ation hardness of the circu it 7 Many of these compounds are shown graphically in Fig ure 2 2 10 in terms of their crystallographic lattice constant versus the energy band gap. Especially, wide band gap electronic devices have excellent electrical and physical characteristics. Table 2 2 shows the physical parameters in different semiconductor materials. The high power, high frequency operation most promising materials are GaN and SiC and the band gap energy is 3.4 eV and 3.2 eV respectively. For example, to get 10 15 /cm 3 intrinsic carr ier concentration (n i ), we need 300 C for the Si materials, 500 C for the GaAs, however much higher temperature s are needed to get the same intrinsic carrier concentration in the GaN, namely about 1000 C For these reasons, the GaN is much better for us e in high temperature conditions, and devices made out of it will operate more reliably at elevated temperature. The early unintentionally doped GaN was n type, which at that time was believed due to nitrogen vacancies. The high n type background carrier c oncentration on the order of 10 18 cm 3 proved difficult to minimize and the absence of a shallow acceptor dimmed the prospects of a production scale GaN based device effort. Table 2 3 shows the ionization energy of impurities for GaN. Si is the most gener al n type dopant of for GaN since it effectively incorporates on the Ga site and forms a single shallow donor level. Si is fully ionized at room temperature with t he ionization level of ~30 meV. Basic physical properties of GaN are listed into Table 2 4 5, 6 These values are result of various works and some values have uncertainty because of the fact that different materials are used for experiments and there remains certain inhomogeneity. There is also a m etastable form o f GaN as Zinc Blend structure. The variations in different properties like calculated mobility,

PAGE 25

25 thermal conductivity are possible because of crystal defects such as dislocations. Defects in the materials are very critical factor in the effect. Lots of dislocations are caused by lattice mis match, which is 13% on the sapphire, 3% on the SiC substrate. The successes of all GaN related devices depend largely on having excellent ohmic a nd schottk y contacts to these devices. A figure of a GaN HEMT is shown in Figure 2 3. There are two types of co ntacts to a semiconductor. One contact is o hmic and other is schottk y Metal Contact At present improvement in contact has become a critical factor for better technology along with advancing the properties of the semiconductors itself. In recent years, GaN itself have been proven to be excellent choice for high temperature, high frequency applications. The successes of all GaN related devices for high temperature application will depend largely on h aving excellent contacts to these devices. There are two typ es of contacts to a semiconductor. Contact to semiconductor basically consists of region of semiconductor surface just below first metal layer, metal semiconductor interface and few layers of metallization above it. Invariably the as deposited contact does not give the desired properties (either low resistance or high schottky barrier). So the contacts are annealed which results in formation of different complex intermetallic compounds by way of solid state reaction among metal layers and semiconductor surf ace. Thus a contact simply is referred to as the region of metal semiconductor interface that leads to desira ble electrical characteristic. The current transport in metal semiconductor contact occurs by majority carriers. There are two different types of c ontacts namely ohmic and schottky In ohmic contact the current voltage relation follows Ohms law that is it should be linear. The contact resistance should be very low so that there is negligible voltage drop across it and hence negligible power drop. Thi s is very important for devices and more so in power application where minimum loss and maximum efficiency is required. Another critical requirement for high

PAGE 26

26 temperature application is the need to have contact which does not degrade or rather have a high r esistance to degradation. Smooth surface morphology, sharp edge acuity and reliability and reproducibility are other features that are desired in an ideal contact. Another type of contact is schottky contact, or rectifying contact in which large current ca n flow in one direction at small voltage and almost no current in reverse direction. High barrier height is essential for producing rectifying effects. Whether a metal semiconductor interface forms an ohmic or schottky contact depends upon the metal work function, m, and semiconductor work function, s. Work function is the amount of energy required to excite an electron from Fermi en ergy level to the vacuum level. Theoretically, on n type semiconductor, ohmic contact is formed when m < s, and schottky contact is formed when m > s. Conversely, in p type material, m > s and m < s produces ohmic and rec tifying contact, respectively. Selected values of work function for commonly used metals are shown in Table 2 5 11 The semiconductor work function is sum of the electron affinity and energy difference between Fermi energy and the bottom of the conduction band i.e. s = s + where s is the electron affinity and is the energy difference between the Fermi energy and the conduction band 3 The elec tr on affinity for GaN is 4.1 eV 11 The work function of Tungsten (W), Cr, Ti, Zr is 4.55, 4.5, 4.33 and 4.05 eV respectively. Schottky c ontact When an intimate contact is formed between metal and a semiconductor, the Fermi levels in the two materials must b e coincident at thermal equilibrium. This can be achieved through a charge flow from semiconductor to metal. Thus a barrier forms at the interface and an equal and opposite space charge is distributed over the barrier region near the semiconductor surface. With an n type semiconductor in the absence of surface state, the barrier height q bn is given by

PAGE 27

27 (2 1 ) where q m is the metal work function, q is the electron affinity of the semiconductor. For an ideal contact between m etal and a p type semiconductor, the barrier height q bp is given by (2 2 ) When surface states are present on the semiconductor surface, and the density is sufficiently large to accommodate any additional surface charges wit hout appreciably altering the occupation level E F the space charge in the semiconductor will remain unaffected. As a result, the barrier height is determined by the property of the semiconductor surface, and is independent of the metal work function. In p ractice, some surface states are always present at the semiconductor surface, and continuously distributed in ene rgy within the energy gap. The s c hottky barrier heights of metal semiconductor systems with intimate contact are, in general, determined by bot h the metal work f unction and the surface states. In a simple mod el for all semiconductors, the s chottky barrier height 75 q b can be expressed as (2 3 ) where m is metal electronegativity, 0 represents the contribution of surface states of semiconductors, and interface index S= is found to be a function of the electronegativity difference between cation and anion of compound sem i conductor ( Figure 2 4 A ). Note a sharp transition around =1. For ionic semiconductors, >1, the index S approaches 1, and b is strongly dependent of the metal electronegativity (or work function). On the other hand, for covalent semiconductors with <1, S is small, b is affected by high density surface states from dangling bonds and only w eakly depends on metal work function. GaN has an electronegativity difference of 1.4 (Ga: 1.6, N: 3.0), which would predict the s chottky barrier heights depend on

PAGE 28

28 metal work fu nction, and are given by Equation 2 1 and Equation 2 2 for metal on n type and p type material respe ctively. A summary of reported s chottky barrier heights of a variety of elemental meta ls on n GaN is shown in Figure 2 4 B 12 It is clear that the barrier height indeed varies with the metal work function within experimental scattering The current transport in metal semiconductor contacts is mainly due to majority carrier, in contrast to p n junctions. Two major processes under forward bias are (1) transport of electrons from the semiconductor over the potential barrier into the metal ; (2) quantum mechanical tunneling of electrons through the barrier. In addition, we may have recombination current in the space charge region and leakage current at the contact periphery. The transport of electrons over the potential barrier is often the dominant process for s chottky diodes on moderately doped semiconductors. It can be adequately described by thermionic emission theory for high mobility semiconductor (for low mobility materials, the diffusion theory is also applicable), and the electric cu rrent density over the barrier has the following expression: (2 4 ) where J s is the saturation current density, is the effective Richardson constant. In practical device, the barrier height dependent on bias voltage and the current voltage characteri stics is more accurately described by: (2 5 ) Factor n is called the ideality factor. The barrier height and ideality factor can be obtained from the forward J V characteristics (for V>3kT/q): (2 6 ) (2 7 )

PAGE 29

29 For a heavily doped sem iconductor or for operation at low temperatures, the tunneling current may become the dominant transport process. The tunneling current has an expression: (2 8 ) where s is permittivity of semiconductor, m* is effective mass of carrier, N D is car rier concentration. It indicates the current will increase exponentially with N D 0.5 Earlier work on schottky contacts to n GaN have been done based on metals layers consisting mainly of Ni or Pt with Au above it 37 40 65, 66 W/Ti/Au and WSi x /Ti/Au schem es have also been used as schottky resulting in thermally stable schottky with barrier height of ~0.80 eV which reduced to ~0.4 eV for subsequent annealing at 400 o C 68 The barrier height seems to follow the difference in work function value w ith in exper imental scattering. Schottky diode can be made by depositing an inner circular schottky metal scheme with an outer concentric ring as the ohmic metal scheme. The outer ohmic metal is first deposited and annealed to get the desired ohmic characteristic and inner circle is realigned. A diagram of a schottky st ructure is shown is Figure 2 5. Ohmic c ontact It is imperative that a semiconductor device be connected to the outside world with no adverse change to its current voltage characteristics. This can be acc omplished through ideal ohmic contacts to the semiconductor. An ohmic contact is defined as a metal/semiconductor contact that has negligible contact resistance relative to the bulk or spreading resistance of the semiconductor. A satisfactory ohmic contact can supply the required current with a voltage drop that is sufficiently small compared with the drop across the active region of the devices. One important figure of merit for ohmic contact is specific contact resistance r c which is defined as:

PAGE 30

30 r c = (2 9 ) For contact with lower doping concentration, at relatively high temperature, conduction across the M/S interface is dominated by thermionic emission over the potent ial barrier, as given in Eq uation 2 4 Therefore, r c (2 10 ) It is ob vious that low b should be used for small r c Ideally a metal with a lower work function than an n type semiconductor or higher work function than a p type semiconductor should be used for ohmic contact to this semiconductor. Unfortunately, very few pract ical material systems satisfy this cond ition, and metals usually form s chottky barriers at semiconductor interface. A practical way to obtain a low resistance ohmic contact is to create a highly doped region near the surface by ion implantation, or increas e the doping by alloying the contacts. In this case, th e depletion layer cause by the s chottky barrier becomes very thin, and current transport through the barrier is enhanced by tunneling. The contact resistan ce can be obtained from Eq uation 2 8 r c ~ (2 11 ) Note that r c depends strongly on N D Under intermediate conditions, thermionic field emission is important, where there is enough kinetic energy for the carrier to be excited to an energy level at which the potential barrier is thin enough f or tunneling to occur. Typical ohmic conduction is usually related to a large tunneling component. It is difficult to make ohmic contacts on wide bandgap semiconductors, such as GaN ( g =3.4 eV, =4.1 eV) and SiC. Generally the doping concentration is rela tively low due to the high ioniz ation level of typical dopants.

PAGE 31

31 A wide variety of metallization schemes been tr ied for ohmic contact for n GaN Some of the earliest report of ohmic contact to n GaN had Al as the ohmic contact metal with specific contact re sistivity of ~ 10 7 cm 2 34 Specific contact resistivity of 8 x 10 6 cm 2 using Ti/Al was achieved after 900 o C anneal for 30 sec in N 2 ambient 19 Ti/Al contact with Si implantation resulted in specific contact resistivity of 3.6 x 10 8 cm 2 and with R IE pre treatment resulted in 8.9 x 10 8 cm 2 18 21 A Specific contact resisitivity of ~ 8 x 10 5 cm 2 was reported for W on n GaN 37 Specific contact resistivity of ~5.6 x 10 6 cm 2 for Al/Ti contact on AlGaN /GaN hetrostructure with Si implantation, 5.3 x 10 7 cm 2 for Ta/Ti/Al contacts, 1.2 x 10 5 cm 2 for Ti/Al/Ni/Au cont acts have been also reported 24,28,30 The most common contact scheme used is Ti/Al bi layer with Ni/Au Ti/Au/ or Pt/Au over layer where overlayer is mainly for preventing out dif fusion, smooth morphology and Au is used for reducing sheet resistance of the layers and to prevent oxidation during high tempe rature anneal 13 36 High temperature metals have also been used for ohmic schemes to have better long term stability. Ti/Al/Mo/A u, Ti/Al/Ir/Au, W and WSi x gave specific contact resis tivtiy of ~10 5 cm 2 37 43 Contact resistance is measured and Specific contact resistance is determined by Transfer length model (TLM), also kno wn as transmission line model. Linear TLM patterns consi st of square or rectangular contact pads s eparated by different spacing. There is also a Circular TLM patterns which has concentric circular metal patterns where either the inner radii or the outer radi i change to vary gap distance. Schematics of linear TL M and measurement plot is shown in Figures 2 6. Current and voltage information obtained from electrical measurements are curve fitted with the corresponding equations to determine th e specific contact resistance. For linear TLM, the total resistance, R s and specific contact resistance, c are given by

PAGE 32

32 where R C is the contact resistance, R s is the sheet resistance, L is the distance between two pads, W is the width of the pad. For Circular TLM, t he specific contact resistance, c can obtained form the circular TLM measurements with the relationships 10 8 120 where R T is the total resistance, R S is the sheet resistance, R 1 is the outer radius of the annular gap, R O is the inner radi us of the annular gap, I O I 1 K O and K 1 are the modified Bessel functions, L T is the transfer length, and c is th e specific contact resistance. Thermal s tability Thermal processing such as activation of ion implants and alloying of metal contacts can b e detrimental to device operation due to changes in the material, interaction, or reactions, as also observed in GaAs 109 11 1 120 It is very important to have contacts that are able to resist the high temperature long enough to be commercially possible f or high temperature applications. In regards to this a good understanding of the degradation of the material is helpful in identifying high temperature process limits. Reliable and stable operation of devices largely depends upon the the rmal stability of t he contacts. At high temperature a lot of intermetallics compounds may form in contacts as a result of interdiffusion of different metal. This can result in rough surface morphology, change in stoichiomet r y, change in composition resulting in change in ele ctrical and optical properties

PAGE 33

33 Common Processing Techniques Dry Plasma Etching Etching refers to the crucial IC fabrication process of transferring patter n by removing specified areas. Wet chemical etching was widely used in manufacturing until the 1960s. Even though this technique is inexpensive, the feature size is limited to about 3 microns. The isotropic etching results in sloped sidewalls and unde rcutting of the mask material. As feature dimension decreases to microns and submicrons and device density per chip increases, ani sotropic etching is necessary. Dry etching techniques using gases as primary etch medium were developed to meet this need. In addition to anisotropic pattern transfer, dry etching provides better uniformity across the wafer, higher reproducibility, smoother surface morphology, and better control capabil ity than wet chemical etching. Three general types of dry etching include plasma etching, ion beam milling, and reactive io n etch (RIE) 112, 11 3 120 Inductively coupled plasma (ICP) etching was used in this study and will be discussed in detail. ICP etching is a dry etching technique where high density plasmas are formed in a dielectric vessel encircled by inductive coils as shown in Figures 2 7 and 2 8. When an rf power is applied t o the coil, commonly referred to as the ICP source power, the time varying current flowing through the coil creates a magnetic flux along the axi s of the cylindrical vessel. This magnetic flux induces an electric f ield inside the vacuum vessel. The electro ns are accelerated and collide with the neutral operating gas, causing the gas molecules to be ionized, excited or fragmented, forming high den sity plasma. The electrons in circular path are confined and only have a small chance of being lost to the chambe r walls, thus the dc self bias remains low. The plasma generated as described above consists of two kinds of active species, neutrals and ion s. The material to be etched sits on top of a small electrode that acts as parallel plate capacitor along with the ch amber as the second electrode. When an rf power, also known as electrode

PAGE 34

34 power or chuck power, is applied to the sample stage, the electrons in the plasma accelerate back and forth in the plasma from the ch anges in the sinusoidal field. Since electrons h ave much lighter mass compared to the other species in the plasma, they respond more rapidly to the frequency change than the other specie s. As the electrons impinge the chamber surfaces, the chamber becomes slightly ne gative relative to the plasma. The su rface area of the chamber is larger than the sample stage, thus the negative charge is con centrated on the sample stage. This bias attracts the ions toward the sample, bombarding t he surface to remove material. In an ICP system, the plasma density and the ion energy and are effectively decoupled in order to achieve uniform density and energy distributions and maintain low ion and electro n energy low. This enables ICP etching to reduce plasma damage wh ile achieving fast etch rates. The plasma generated as de scribed above consists of two kinds of active species: neutrals and ions. Neutrals are chemically reactive and etch the material by chemical reactions, while ions are usually less reactive and are responsible for removing material by physically bombarding the sample surface. The kinetic energy of the ions i s controlled by electrode bias. The electron density and ion density are equal on average, but the density of neutrals, known as the plasm a density, is typically higher. Anisotropic profiles are obtained by superimposing an rf bias on the sample to independently control ion energy and by using low pressure conditions to minimize ion scattering and lateral etching. The plasma is neutral but is posi tive relative to the electrode. It appears to glow due the ion excitati on from the electron movements. The recombination of charges at the boundary surfaces surrounding the plasma creates a charge depletion layer, also known as a sheath, dark space or dark region, resulting in diffusion of carriers to the boundari es. The diffusion of electrons is faster than ions initially, thus an excess of positive ions is left in the plasma and assumes a

PAGE 35

35 plasma potential, V p with respect to the grounded walls. The plasma and substrate potentials generate drift current to enhanc e the ion motions and hinder the electron motions until stea dy state condition is achieved. The difference in electron and ion mobility also generates a she ath near the powered electrode. The dark region, a small region in the plasma immediately above the sample, keeps the electrons away due to th e negatively charged electrode. The powered electrode reaches a self bias negative voltage, V dc with respect to the ground. Even though the voltage drop controls the ion bombardment energy across the plasma sheath it is difficult to measure; therefore, it is common to monitor the V dc Note that the dc bias is not a basic parameter and is characteristic to a particular piece of equipment. Etching is accomplished by the interaction of the plasma to the substrate. Th e three basic etching mechanisms, chemical etch process, physical etch process, and a combination of both chemical and physical etching p rocess, are shown in Figure s 2 9 2 1 0, and 2 11 respectively. Chemical etch process is the chemical reaction that etc hes the substrate when active species (neutrals) from the gas phase are absorbed on the surface material and react with it to form a volatile product. The chemical etch rate is limited by the chemical reaction rate or diffusion rate that depends on the vol atility of the etch products since undesorbed products coat the surface and preve nt or hinder further reactions. Chemical etching is a purely chemical process therefore etches isotropically or equally in all directions. Physical process, also known as spu ttering, occurs when positive ions impinge n ormal to the substrate surface. If the ions have sufficiently high energy, atoms, molecules or ions are ejected from the substrate surface to a chieve a vertical etch profile. The etch rate of sputtering is slow, and the surface is often da maged from the ion bombardment. A combination of both chemical and physical etching process, also known as energy driven, ion enhanced mechanism, takes advantage of the effect of ion bombardment in

PAGE 36

36 the presen ce of reactive neutra l species. The energetic ions damage the surface and leave the surface more reactive toward incident neutrals, leading to removal rates that exceed the sum of separate s puttering and chemical etching. This process produces very fast etch rates and anisotro pic profile; therefore, it is desirable in hig h fidelity pattern transfer 120 Ion Implantation Ion implantation is a physical process that introduces dopants by means of high voltage bombardment to achieve desired electrical properties in defined areas wi th minimal lateral diffusion. Inside a vacuum chamber, a filament is heated to a sufficiently high temperature where electrons are cre ated from the filament surface. The negatively charged electrons are attracted to an oppositely charged anode in the chamb er. As the electrons travel from the filament to the anode, they collide and create positively charged ions fr om the dopant source molecules. The ions are separated in a mass analyzer, a magnetic field that allows the passage of the desired species of posi tive ions with specific characteristic arc radius based upon ion mass. The selected ions are accelerated in an acceleration tube and then focused into a small diam eter or several parallel beams. The beam is scanned onto the wafer surface, and the ion s phys ically bombard the wafer. The ions enter the surface and come to rest below the surface as they lose their energy through nuclear interactions and coulombic interactions, resulting in Gaussian distr ibution concentration profile 11 4 120 A schematic of an ion implantation system is illustrated in Figure 2 12 120 During implantation, the collisions with high energy ions cause crystal damage to the wafer, leading to p oor electrical characteristics. In most cases, the carrier lifetime and mobility decrease dr astically. Also, only a small fraction of the implanted ions are located in substitutional sites and contr ibute to carrier concentration. Annealing is needed to repair the crystal dama ge and to activate the dopants. To determine the depth and damage profil e, Rutherford

PAGE 37

37 Backscattering and Channeling (RBS/C) an alytical technique is employed. Annealing process and RBS/C will be further discussed in the subsequent sections 120 Rapid Thermal Annealing Annealing is a thermal process used for repairing the ion im plantation damage, diffusing dopan ts and alloying metal contacts. After ion implantation, annealing is employed to repair the crystal damages caused by the high energy ion bombardment that degrade carrier lifetime and mobility. Since the majority of the im planted dopants reside in the interstitial sites, the as implanted materials have poor electrical prop erties. Annealing provides thermal energy for the dopants to migrate to the substitutional sites and contribute to the carrier concentration 115 11 6 Trad itionally, tube furnaces were used for an nealing after ion implantation. However, furnace annealing causes the implanted atoms to diffuse laterally and requir es relatively long anneal time. Rapid thermal annealing was developed in orde r to overcome these d rawbacks. Rapid thermal annealing (RTA) utilizes radiation heating from arc lamps or tungsten halogen lamps to heat the wafer in an inert atmosphere such as N 2 or Ar. It can attain higher temperature at a shorter time period than a conventional tube furnac e, and the overall anneal time is relatively short, usually taking seconds as compared to several minutes to hours i n a conventional tube furnace. RTA allows uniform heating and cooling that reduces thermal gradients that can lead to warping and stress ind uced defects, enabling more dense design and fewe r failures due to dislocations 120 Characterization Techniques Atomic Force Microscopy Atomic force microscopy (AFM) employs a microscopic tip on a cantilever that deflects a laser beam depending on surface morphology and properties through an interaction be tween the tip and the surface. The signal is measured with a photodetector, amplified and converted into an

PAGE 38

38 image display on a cathode ray tube. Depending on the type of surface, AFM can be performed in c ontact mode and tapping mode. A schematic diagram o f AFM is shown in Figure 2 13 120 Auger Electron Spectroscopy Auger electron spectroscopy (AES) determines the elemental composition of the few outerm ost atomic layers of materials. A focused beam of ele ctrons with energies from 3 keV to 30 keV bomb ards the surface of a specimen. The core level electrons are ejected from in a vacancy in the core level. As the atom relaxes, an outer level electron fills the core vacancy and releases excess energy, which in turn, ejects an outer electr on, known as a n Auger electron. This process is illustrated in Figure 2 14. The kinetic energy of the Auger electrons is characteristic of each element, with the ex ception of hydrogen and helium. Therefore, by measuring the energies of the Auger electrons, the near surface composition o f a specimen can be identified. In addition, AES can provide compositional depth profile from relative intensities of the elements present if the system is equipped with an ion gun to sputter away material 11 7 120 X ray Pho toelectron Spectroscopy X ray photoelectron spectroscopy (XPS), also known as electron spectroscopy for chemical analysis (ESCA), provi des similar information as AES. Instead of impinging the sample surface with an electron beam, XPS utilizes a monoengerge tic x ray beam to cause electrons to be ejected, usuall y two to 20 atomic layers deep. The variation of the kinetic energies of the ejected electrons identifies the elements present and ch emical states of the elements 11 8 120 Electrical Measurements Curr ent voltage (I V) measurements were taken to characterize the electrical properties of the contacts. These measurements are performed on an Agilent 4156 Semiconductor Parameter

PAGE 39

39 Analyzer connected to a micromanipulator probe station. For diodes, the input voltage was applied through schottky and out through ohmic for forward bias and visa versa for reverse bias measurements. For ohmic measurements on TLM pads 4 probes were used in series, two outer probes for applying the current and inner two probes for pi cking up the voltage. Photoluminescence (PL) Photoluminescence (PL) is an analytical technique that provides information about the optical prop erties of a substrate. A light source, such as He Cd, Ar and Kr lasers, with energy larger than the bandgap energ y of the semiconductor being studied, generates electron hole p airs within the semiconductor. The excess carriers can recombine via radiative a nd non radiative recombination. Photoluminescence, the light emitted from radiat ive recombination, is detected. T he wavelength associated with the different recomb ination mechanism is measured. The luminescence from excitons, electrons and holes bound to each other, is observed only at low tempera tures in highly pure materials. As the temperature increases, the excit on breaks up into free ca rriers from the thermal energy. Increase in doping also causes the dissociation of excitons under local electr ic fields. Under these conditions, the electrons and holes recombin e via the band to band process. Since some of the elec trons may not lie at the bottom of the conduction band, their recombination and holes will produce a high energy tail in the luminesc ence spectrum. On the other hand, the band to band recombination will yield a sharp cutoff at the wavelength corresponding to the band gap of the material 11 8 120 Rutherford Backscattering Spectrometry/Channeling Depth profile of implanted ions and damages can be obtained by the Rutherford Backscattering Spectrometry/Channeling (RBS/C) technique, which measures the energy di stribution of the backscattered ions from the implanted samp le surface at a specific angle. The

PAGE 40

40 energy of the backscattered ion is determined by the mass of the atomic nucleus and the depth at which the elastic collisions take place. A beam of high energy ions impac ts the surface of the specimen. The angle of the analyzing ions affects the penetration depth. If the ions are injected parallel to the crystal axis of the specimen, they penetrate considerably deeper than if injected randomly, due to the lower s topping power from channeling. Deeper penetration results in h igher backscattered ions yield. The displacement of an atom, either as host or impurities, from the crystal lattice also incr eases the backscattering yield. Therefore, the distribution of displa ced atoms that are caused by the radiation damage from ion implantation can be measured by increasin g the backscattered ion yield 11 8 120 Scanning Electron Microscopy Scanning electron microscopy (SEM) generates images f rom electrons instead of light. A beam of electron is produced and ac celerated from an electron gun. The electron beam passes through a series of condenser and objective lenses, which focus the electron beam. A scanning coil moves the be am across the specimen surface. The electron beam int eracts with the specimen, and electrons from the surface interaction volume, such as backscattered, secondary, characteristic x ray continuous x ray, and Auger, are emitted. The signals are collected, amplified and convert ed to a cathode ray tube image. De pending on the specimen and the equipment setup, the contrast in the final image provides information on the specimen composit ion, topography and morphology. The main advantages of using electrons for image formation are high magnification, high resolut ion and large depth of fields 11 8 120 Secondary Ion Mass Spectrometry Secondary ion mass spectrometry (SIMS) is a highly sensitive chemi cal characterization technique. Primary ions, such as Cs + O 2 + O and Ar, bombard the specimen in an ultra high

PAGE 41

41 vacuum environment, sputtering away secondary ions from the specimen surface. A small fraction of the ejected atoms are ionized either positively or negatively, and they are called secondary electrons. The composition of the surface is determined by the secondary electrons that are individually detected and tabulated using a mass spectrometer, as a function of their ma ss to charge ratio. There are two mo des of SIMS, static or dynamic. In the static mode, a low primary ion flux < 10 14 c m 2 is used, leaving the speci men surface relatively undisturbed. The majority of secondary ions originate in the top one or two monolayers of the samples. The dynamic mode monitors the selected secondary ion intensities as a function of the sputtering time, resulting in a conc entratio n versus depth profile. The depth resolution of this tec hnique ranges from 5 to 20 nm 11 8 120 Stylus Profilometry Stylus profilometry is used to measure the topographical features of a specimen surface, such as roughness, step height, width and spacing. A probe, or stylus, contacts the surface of the specimen and follows height variation as it scans acr oss the surface. The height variations are converted into electrical signals, providing a cross sectional topogra phical profile of the specimen. In this wo rk, the etch rate was calculated by the depth, as measured by the profilometer, over a specified period of time. Table 2 1 Electrical properties of Si, GaAs and GaN. Property Si GaAs GaN Bandgap energy (eV) 1.1 (indirect) 1.4 (direct) 3.4 (direct) Elect ron mobility (cm 2 /Vs) 1400 8500 1000 (bulk) 2000 (2D gas) Hole mobility (cm 2 /Vs) 600 400 30 Electron effective mass 0.98 0.067 0.19 Hole effective mass (light) 0.16 0.082 0.60

PAGE 42

42 Table 2 2 The physical parameters in different semiconductor materials Si GaAs GaN AlN 6H SiC Bandgap (eV) @300 o C 1.1 1.4 3.4 6.2 2.9 Electron m obility (cm 2 /V s), RT 1400 8500 1000 (bulk) 2000 (2D gas) 135 600 Hole mobility (cm 2 /V s), RT 600 400 30 14 40 Saturation v elocity ( cm/s), 10 7 1 2 2.5 1.4 2 Breakdown field (V/cm) x 10 6 0.3 0.4 >5 4 Thermal c onductivity (W/cm) 1.5 0.5 1.5 2 5 Melting t emperature (K) 1690 1510 >1700 3000 >2100 Table 2 3 Ionization energy of i mpurit ies for w urtzite GaN. Impurities G a site ( e V) N site ( e V) Remarks Si 0.012 0.02 Donor Native d efect (V N ) 0.03 Donor C 0.11 0.14 Donor M g 0.14 0.21 Acceptor Si 0.19 Acceptor Zn 0.21 0.34 Acceptor Native d efect ( v ga ) 0.14 Acceptor Hg 0.41 Acceptor

PAGE 43

43 Table 2 4 Basic physical properties of GaN. Property Value Lattice parameters at 300 K (nm) a 0 = 0.3189 nm c 0 = 0.5185 nm Density (g cm 3 ) 6.095 g.cm 3 Stable phase at 300 K Wurtzite Melting point (C ) 2500 Thermal conductivity ( Wcm 1 K 1 ) 1.3, 2.20.2 F or thick, free standing GaN Linear thermal expansion coefficient Along a 0 =5.59x10 6 K 1 Along c 0 =7.75x10 6 K 1 Static dielectric c onstant 8.9 Refractive index 2.67 at 3.38 eV Energy bandgap (e V) Direct, 3.45 Exciton binding energy (meV) 26 Electron effective mass 0.20 Table 2 5. Metal work function and ideal barrier heights for GaN (electron affinity: 4.1 eV) Element Work Function (eV) Ideal Barrier Height (eV) B 4.45 0.35 Cr 4.5 0.4 Pt 5.64 1.54 Ti 4.33 0.23 W 4.55 0.45 Zr 4.05 0.05

PAGE 44

44 Figure 2 1. Crystal structure of wurtzite GaN Figure 2 2 The III V compound semiconductor tree

PAGE 45

45 Figure 2 3. S tructure of a HEMT

PAGE 46

46 Figure 2 4. Previous study of schot t k y contacts A) Index of interface behavior S as a function of the electronegativity diffe rence of the semiconductors. B) Barrier height versus work function of metals deposited on n GaN reported from various groups. B A

PAGE 47

47 Fi gure 2 5. Lithography p attern fo r S c hottky diode Figure 2 6 Lithography pattern for linear TLM A) TLM pads. B) P lot for measurement. A B

PAGE 48

48 Figure 2 7. A n ICP reactor Figure 2 8 Electric and magnetic fields inside the reac tor ~ Plasma Powered electrode Sample Gas distribution 2 MHz Power supply 13.56 MHz Rf power source Alumina chamber ~ Gas outlet Induced E Field Rf Current Magnetic Field

PAGE 49

49 Figure 2 9 Chemical etching process. A) G eneration of reactive species. B) Diffusion of reactive neutrals to surface. C) Adsorption of r eactive neutrals to surface. D) Che mical reaction with surface. E) Desorpt ion of volatile byproducts. F) Diffusi on of byproducts into bulk gas Figure 2 10 Physical etching process. A) G eneration of reactive species. B) Accele ration of ions to the surface. C) Ions bombard sur face D) Surface atoms a re ejected from the surface + A B C D E F + Electron Reactive neutral Ion Substrate atom + + + A B C D Sample Negatively biased + + + +

PAGE 50

50 Figure 2 11 Combination of chemical an d physical etching process. A) Gener ation of reactive species. B) Diffusion of rea ctive neutrals to surface. C) Ion bombardment to surf ace. D) Adsorption of reactive neutrals to surf ace. E) Che mical reaction with surface. F) Desorpt ion of volatile byproducts. G) Diffusi on of byproducts into bulk gas Figure 2 12 I on implantation system. + A B D E F G + C Acceleration Tube Ion Source Mass Analyzer Scan ner Focus Wafer

PAGE 51

51 Figure 2 13 S implified principle of AFM. AFM tip Photodetector Laser beam Specimen surface

PAGE 52

52 Figure 2 14 Auger Process. A) An i solated atom B) I nne r core level electron dislodged, leaving behind a vacancy C) A n outer level electron fills the vacancy and releases excess energy D) T he excess ene rgy ejects an Auger electron. Electron Vacancy Auger Electron

PAGE 53

53 CHAPTER 3 TUNGSTEN AND ZIRCONIUM BORIDE BASED OHMIC CONTACTS TO N Ga N Introduction There is a strong interest in the development of more reliable and thermally stable o hmic contacts on GaN b ased electronic devices such as high electron mobility transistors (HEMTs) 13 45 which show outstanding potential in advanced power amplifiers for radar and communication systems over a broad frequency range from S band to V band 13 15 A key aspect of op eration of nitride based HEMTs at high powers is the need for temp erature stable high quality o hmic contacts. There is increasing interest in the application of AlGaN/GaN HEMTs to microwave power amplifiers capable of uncooled or high temperature operation where thermal stability of the contact metallurgy is paramount. T he most common o hmic metallization for AlGaN/GaN HEMTs is based on Ti/Al. This bilayer must be deposited with over layers of Ni, Ti or Pt, followed by Au to reduce sheet resistance and decr ease oxidation during the high temperature anneal needed to achieve the lowest specific contac t resistivity 16 36 For impr oving the thermal stability of o hmic contacts, there is interest in higher melting t em perature metals including W 37,38,40,45 WSi X 37 39 ,Mo 16 ,V 39 and Ir 40,42,43 Another promising metallization system as the diffusion barrier layer is based on borides of Cr, Zr, Hf, Ti or W 46 47 Stoichiometric diborides have high melting temperatures (eg. 3200 C for ZrB 2 ) and thermodynamic stab ilities at least as good as comparable nitrides or silicides 48 These metallization systems are expected to be less reactive with GaN than the conventional Ti/Al. Previous work has shown good contact resistance obtained with Ti/Al/Mo/Au, Ti/Al/Ir/Au, Ti/A l/Pt/WSi/Ti/Au and Ti/Al/Pt/W/Ti/Au on n GaN 16 39 but there is a broad range of other contact metallurgies that are promising, including those based on borides. For example, one attractive option is ZrB 2 which has a low resistivity in the range 7 .cm. Recently, it was shown that hexagonal ZrB 2 (0001) single crystals have an in

PAGE 54

54 plane lattice constant close to that of GaN, prompting efforts at GaN heteroepitaxy on ZrB 2 or buffer layers on Si substrates. For contacts on n type GaN, the only related wo rk is the study of ZrN/Zr/n GaN Ohmic structures 48 in which the Zr/GaN interface was found to have excellent thermal stability. Even though we are not relying on the ZrB 2 to make direct o hmic contact to GaN,it is expected that ZrB 2 contacts on GaN will ha ve low barrier heights, given the work function of ZrB 2 is ~3.94 eV and the electron affinity of GaN is ~4.1 eV. In this chapter a report on the annealing temperature dependence of contact resistance and contact intermixing of Ti/Al/W 2 B/Ti/Au and of Ti/Al /ZrB 2 /Ti/Au on n GaN is given. The Tungsten Boride contacts show excellent minimum contact resistance of 7x10 6 2 after annealing at 800 C and promising long term stability at 200 C The ZrB 2 based contacts show excellent minimum contact resistance o f 3x10 6 2 after annealing at 700 C and retain a good morphology even after annealing at 1000 C Experimental The samples used were 3 m thick Si doped GaN grown by Metal Organic Chemical Vapor Deposition on c plane Al 2 O 3 substrates. The electron con centration obtained from Hall measurements was ~7x10 18 cm 3 2 /Ar Inductively Coupled Plasma Etching to provide electrical isolation of the contact pads. A metallization scheme of Ti ( 200 )/ Al ( 1000 )/ W 2 B or ZrB 2 ( 500 ) / Ti ( 200 ) / Au ( 800 ) was used in these experiments. All of the metals were deposited by Ar plasma assisted rf sputtering at pressures of 15 40 mTorr and rf (13.56 MHz) powers of 200 250 W. The contacts were patterned by liftoff and annealed at 500 1 000 C for 1 min in a flowing N 2 ambient in a RTA furnace. Auger Electron Spectroscopy (AES) depth profiling of the as deposited contacts showed sharp interfaces between the various metals in both types of contacts. For the AES analysis, the

PAGE 55

55 samples were m ounted on a stainless steel puck and placed in the system load lock. Clean tweezers and gloves were used for all sample handling. No additional cleaning steps were implemented. After sufficient evacuation, the sample puck was inserted into the analytical c hamber and placed in front of the analyzer. The AES system was a Physical Electronics 660 Scanning Auger Microprobe. The electron beam conditions were 10 keV, 1 A beam current at 30 from sample normal. For depth profiling, the ion beam conditions were 3 keV Ar + 2.0 A ( 3 mm) 2 raster, with sputter rate based on 110 /minute (SiO 2 ) of 142 / minute Au (1.3*SiO 2 ) ,64 /minute Ti (0.58*SiO 2 ),123 /minute Al (1.12*SiO 2 ), 44 /minute W (0.4*SiO 2 ){for W 2 B},55 /minute Al 2 O 3 (0.5*SiO 2 ) and 83 /minute (0.75*SiO 2 ) Prior to AES data acquisition, magnifications of 125X, and 1,000X. The SEIs were used to locate and document analysis area locations and to document surface morpho logy. The quantification of the elements was accomplished by using the elemental sensitivity factors. The contact properties were obtained from linear transmission line method (TLM) measurements on 100100 m pads with spacing 5, 10, 20, 40, and 80 m. The contact resistance R C was obtained from the relation 49 R C = ( R T S d / Z )/2, where R T is the total resistance between two pads, S is the sheet resistivity of the semiconductor under the contact, d is the pad spacing, and Z is the contact width. The specific C is then obtained C = R C L T Z where L T is the transfer length obtained from the intercept of a plot of R T vs d Results and Discussion Tungsten Boride Based Ohmic C ontact Figure 3 1 shows the contact resistance as a function of anneal temperature. The as deposited contacts showed rectifying behavior. This is expected for any as deposited contacts on the wide bandgap GaN 25 The current voltage characteristics became o hmic for anneal

PAGE 56

56 500 C The contact resistance decreased up to ~800 C reaching a minimum value of 7x10 6 2 This trend is most likely related to the formation of low resistance phases of TiN at the interface with the GaN, as reported for conventional Ti/Al/Pt/Au contacts 19 22 However the W 2 B based contacts show improved edge acuity, which is important for small gate length HEMTs in order to reduce the possibility of shorting of the o hmic metal to the gate. Annealing at higher temperatures leads to hi gher contact resistance, which as will be seen later corresponds to extensive intermixing of the contact metallurgy. The corresponding transfer resistance and semiconductor sheet resistance data are shown in Figure 3 2. The minimum contact resistance obtai ned Figure 3 3 shows the measurement temperature dependence of the Ti/Al/W 2 B/Ti/Au on n GaN annealed at 800 C Over the relatively limited temperature range available for the measurements and within the error of the measurement, we did not observe any temperature dependence. This indicates that at this anneal temperature, the dominant current transport mechanism is field emission 49 since thermionic emission would have significant temperature dependence and thermionic field emission is operative at lower doping ranges (10 16 10 18 cm 3 ). Figure 3 4 shows the SEI of the as deposited contact morphology and after annealing at 500,800 or 1000 C The morphology is featureless until 800 C which corresponds to the minimum in contact resistance. By 1000 C the morphology becomes very rough and this corresponds to the increased contact resistance. T he AES surface scans as a function of anneal temperature was also obtained Carbon, oxygen and gold were detected on the as deposited surface. The carbon is adventitious and the oxygen comes from a thin native oxide on the Au. After annealing at 500 C titanium was detected on the contact surface. After 800 C annealing, gold, titanium, aluminum, and gallium

PAGE 57

57 were detec ted on the surface, which is consistent with the onset of extensive reaction of the contact metallurgy. After 1000 o C annealing, titanium, aluminum, and gallium were detected on the surface. The surface concentration of gold decreased with increasing tempe rature of the annealing step. Titanium and aluminum concentrations increased with annealing temperature. Figure 3 5 shows the AES depth profiles for the as deposited and annealed samples. The profile obtained from the as deposited sample was in good agreem ent with the prescribed metal layer thicknesses. Oxygen was detected at the Ti/W 2 B interface, in the W 2 B layer, at the W 2 B/Al interface, and in the deep Ti layer. Note that the main (and only) boron peak overlaps with one of the tungsten peaks. The overlap results in an overestimation of boron in the W 2 B film. Also note that nitrogen was not plotted due to a peak overlap with titanium however nitrogen should only b e present in the GaN substrate. The profile obtained from the sample annealed at 500 C shows diffusion of titanium through the gold layer to the surface. The concentration of oxygen in the titanium layers is higher in this sample compared to the as deposited sample although the distributions of oxygen throughout the profiles are similar. The profi le obtained from the sample annealed at 800 C shows significant diffusion or inter diffusion of layers. The profile now shows the presence of a thin titanium layer, followed by a thin gold layer, then an oxidized aluminum layer, the W 2 B layer, another oxi dized aluminum layer, a gold layer, and a final titanium layer. Layer thickness values should be considered approximate. The sputter rates are only appropriate for pure elements or compounds. Diffusion of other species into any given layer will probably ha ve an impact on the sputter rate of that material. The profile obtained from the sample annealed at 1000 C shows titanium at the surface, followed by an oxidized aluminum layer, a gold layer, the W 2 B layer, another gold layer, a titanium layer, and the Ga N substrate. Some of the transport might be

PAGE 58

58 attributed to grain boundary diffusion, as reported previously for Mo based contacts to AlGaN/GaN heterostructures 39 Figure 3 6 shows the room temperature contact resistance of the sample annealed at 800 C as a function of time spent at 200 C .This simulates the operation of an uncooled GaN based transistor and gives some idea of the expected stability of the contact. Within experimental error, we did not observe any degradation of contact resistance over a p eriod of almost 3 weeks. Future work will establish the stability of the metallization over longer periods and at higher temperatures. Zirconium Boride Based Ohmic C ontacts Figure 3 7 A shows the contact resistance as a function of annealing temperature, while the associated GaN sheet resistance under the contact is shown at the bottom of the figure. The as deposited contacts were re ctifying, with a transition to o hmic behavior for anneal 500 C The contact resistance decreased up to ~700 C reaching a minimum value of 3x10 6 2 This same basic trend is seen in most Ti/Al based contacts 19 22 and is attributed to formation of low resistance phases of TiN at the interfac e with the GaN. By comparison with the usual Ti/Al/Pt/Au metal stack, the ZrB 2 based contacts show improved edge acuity, an important factor for small gate length HEMTs in order to reduce the possibility of shorting of the o hmic metal to the gate. The ZrB 2 based contacts show a double minimum in contact resistance versus annealing temperature and even at 1000 C exhibit a contact resistance below 10 5 2 Figure 3 8 shows the s pecific contact resistivity (A) and sheet resistance (B ) as a function of anneal time at 700 C for Ti/Al/ZrB 2 /Ti/Au on n GaN. The minimum contact resistance is achieved for 60 s anneals. This is also consistent with the need to form a low resistance interfacial phase whose formation kinetics is probably limited by diffusion of Ti to the GaN interface. Figure 3 9 shows the measurement temperature dependence of the Ti/Al/ZrB 2 /Ti/Au

PAGE 59

59 on n GaN annealed at 700 C We did not observe a ny significant temperature dependence, suggesting that at this anneal temperature the dominant current transport mechanism is field emission 49 Figure 3 10 shows scanning electron microscopy (SEM) images of the as deposited contact morphology and after an nealing at 500,700 or 1000 C Even at the highest anneal temperature, the morphology remains quite smooth on the scale accessible to the SEM. This is in sharp contrast to the case of Ti/Al/Pt/Au, where significant roughening occurs above 800 C and this r esult suggests that the ZrB 2 is an effective barrier for reducing intermixing of th e contact compared to Pt. Figure 3 11 shows the AES surface scans as a function of anneal temperature. As expected, only carbon, oxygen and gold were detected on the as depo sited surface. The carbon is adventitious and the oxygen comes from a thin native oxide on the Au. After annealing at 500 C titanium and aluminum was detected on the contact surface and their concentrations increased at higher annealing temperatures. The surface concentration of gold decreased with increasing temperature of the annealing step. After 1000 o C annealing, Boron was also present on the surface, having outdiffused from the buried ZrB 2 layer. Table 3 1 summarizes the near surface composition dat a obtained from AES measurements. Figure 3 12 shows the AES depth profiles for the as deposited and annealed samples. The profile obtained from the as deposited sample was in good agreement with the prescribed metal layer thicknesses. Oxygen was detected i t the ZrB 2 layer, consistent with past observations that the borides are excellent getters for water vapor during deposition 47 The profile obtained from the sample annealed at 500 C shows extensive diffusion of titanium through the gold layer to the sur face. The intermixing of the contact metallurgy becomes more pronounced as the

PAGE 60

60 annealing temperature is increased. Some of the transport might be attributed to grain boundary diffusion, as reported previously for Mo based contacts to AlGaN/GaN heterostruct ures 39 As noted earlier, the contact morphology does not degrade significantly even after 1000 C annealing. Summary a nd Conclusions Both Ti/Al/W 2 B/Ti/Au and Ti/Al/ZrB 2 /Ti/Au metalliz ation scheme were used to form o hmic contacts to n type GaN. For Tungst en Boride based contact, a minimum specific contact resistivity of 7x10 6 2 was achieved at an annealing temperature of 800 C, which is comparable to that achieved with conventional Ti/Al/Pt/Au on the same samples. For Zirconi um Boride b ased contact, a minimum specific contact resistivity of 3x10 6 2 was achieved at an annealing temperature of 700 C, which is comparable to that achieved with conventional Ti/Al/Pt/Au on the same samples. The ZrB 2 based contact appears to have greater thermal stabil ity than the conventional metallization. The Tungsten Boride based contacts showed no change in resistance over a period of more than 450 hours at 200 C This approach of using boride based contacts looks promising for high temperature device applications Table 3 1 Near surface composition of contact stack determined by AES measurements for ZrB 2 ohmic contact Sample ID C(1) O(1) Al(1) S(1) Ti(2) B(2) Au(3) Sensitivity factors [0.076] [0.212] [0.000] [0.652] [0.000] [0.000] [0.049] #1: As deposited 48 1 N d 2 nd nd 49 #2: A nnealed at 500 C 41 22 16 <1 3 nd 17 #3: A nnealed at 700 C 36 29 22 <1 5 nd 8 #4: A nnealed at 1000 C 28 33 18 1 4 15 N d

PAGE 61

61 Figure 3 1 S pecific contact resistivity vers us anneal temperature for Ti/Al/W 2 B/Ti/Au on n GaN.

PAGE 62

62 Figure 3 2 Measurement as a function of anneal ing temperature Ti/Al/W 2 B /Ti/Au on n GaN A) T ransfer resistance. B) S heet resistan ce A B

PAGE 63

63 Figure 3 3. Specific contact resistance versus mea surement temperature for Ti/Al/W 2 B/Ti/ Au on n GaN annealed at 800 C

PAGE 64

64 Figure 3 4 Secondary electron i mages of the Ti/Al/W 2 B/Ti/Au contacts on n GaN A) A s d eposited. B) Annealed at 500 C C) Annealed at 800 C D) Annealed at 1000 C A B C D

PAGE 65

65 Figure 3 5 AES depth profiles of the Ti/Al/W 2 B/Ti/Au on n GaN A) A s d eposited. B) Annealed at 500 C C).Annealed at 800 C D).Annea led at 1000 C A B C D

PAGE 66

66 Figure 3 6 Contact resistance of the Ti/Al/W 2 B/Ti/Au on n GaN, initially annealed at 800 C as a function of subsequent time at 200 C

PAGE 67

67 Figure 3 7 Measurement versus anneal temperature f or Ti/Al/ZrB 2 /Ti/Au on n GaN. A) S pecific contact resistivtiy. B) S heet resistance A B

PAGE 68

68 Figure 3 8 Measurement as a function of annealing time at 700 C for Ti/Al/ZrB 2 /Ti/Au on n GaN. A) Sp ecific c o ntact resistivtiy. B) S heet resistance A B

PAGE 69

69 Figure 3 9 Specific contact resistance versus mea surement temperature for Ti/Al/ ZrB 2 /Ti/ Au on n GaN annealed at 800 C

PAGE 70

70 Figure 3 10 Secondary electron i m ages of the Ti/Al/ZrB 2 /Ti/Au on n GaN A) A s dep osited B) Annealed at 500 C C) Annealed at 700 C D) Annealed at 1000 C A B C D

PAGE 71

71 Figure 3 11 AES surface scans of the Ti/Al/ZrB 2 /Ti/Au on n GaN A) A s dep osited. B) Annealed at 500 C C) Annealed at 700 C D) Annealed at 1000 C A B C D

PAGE 72

72 Figure 3 12 AES depth profiles of the Ti/Al/ZrB 2 /Ti/Au on n GaN A) A s dep osited. B) Annealed at 500 C C) Annealed at 700 C D) Annealed at 1000 C A B C D

PAGE 73

73 CHAPTER 4 COMPARISON OF ELECTRICAL A ND RELIABIL ITY PERFORMANCE OF Ti B 2 Cr B 2 AND W 2 B 5 BASED OHMIC CONTACTS ON N Ga N Introduction One of the remaining obstacles to commercialization of GaN high electron mobility transistors (HEMTs) power amplifiers is the development of more reliable and thermall y stabl e o hmic contacts 13,15,20 64 .These power amplifiers show great potential for radar and communication systems over a broad frequency range from S band to V band 1 3 15,50 GaN based HEMTs can operate at significantly higher power densities and higher impeda nce than currently used GaAs devices 15, 51 63 A key aspect of operation of nitride based HEMTs at high powers is the need for t emperature stable high quality o hmic contacts 20 30, 32, 33, 35 45, 64 The most common o hmic metallization for AlGaN/GaN HEMTs is based on Ti/Al. This bilayer must be deposited with over layers of Ni, Ti or Pt, followed by Au to reduce sheet resistance and decrease oxidation during the high temperature anneal needed to achieve the lowest specific contact resistivity 20 30, 32, 33 35 45, 64 There is a lateral flow issue with these contacts due to the low melting temperature viscous AlAu 4 phase that may cause problems when the gate/source contact separation is small. For impr oving the thermal stability of o hmic contacts, there is interest in high temperature metals such as W 37, 38, 40, 45 WSi X 37 39 Mo 16 V 39 and Ir 40, 42, 43 An unexplored class of potentially stable contacts is that of boride based contacts such as CrB 2 TiB 2 and W 2 B 5 Some of the properties of these metals are shown in Table 4 1. They have high melting temperatures, good electrical conductivity, the heat of formation for stoichiometric borides is comparable to silicides or nitrides 47 and although there is a lack of information on phase diagrams with GaN, t hese metals have shown corrosion resistance against molten metals and thus should exhibit even less solubility in the solid state.

PAGE 74

74 In this chapter a report on the initial results for annealing temperature dependence of contact resistance an d contact interm ixing of Ti/Al/boride /Ti/Au on n type GaN, with W 2 B 5 TiB 2 and CrB 2 as the t hree different selected borides is presented. These contacts show promising l ong term stability at 350 o C which is very important for devices which are going to be used in uncooled and/or prolonged heated environment. Experimental The samples used were 3 m thick Si doped GaN grown by Metal Organic Chemical Vapor Deposition on c plane Al 2 O 3 substrates. The electron concentration obtained from Hall measurements was ~7x10 18 cm 3 Mesa 2 /Ar Inductively Coupled Plasma Etching to provide electrical isolation of the contact pads. A metallization scheme of Ti (200 )/Al (1000 )/ Boride (500 ) / Ti (200 ) /Au (800 ) was used in these experiments where Boride s were W 2 B 5 CrB 2 and TiB 2 All of the metals were deposited by Ar plasma assisted rf sputtering at pressures of 15 40 mTorr and rf (13.56 MHz) powers of 200 250 W. The contacts were patterned by liftoff and annealed at 500 1000 C for 1 min in a flowing N 2 ambient in a RTA furnace. Auger Electron Spectroscopy (AES) depth profiling of the as deposited contacts showed sharp interfaces between the various metals in both types of contacts. The AES system was a Physical Electronics 660 Scanning Auger Microprobe The electron beam conditions were 10 keV, 1 A beam current at 30 from sample normal. For depth profiling, the ion beam conditions were 3 keV Ar + 2.0 A ( 3 mm ) 2 rast er Prior to AES data acquisition, secondary electron microscopy images (SEM s) were o bt ained from the sample. The SEM s were obtained at magnificati ons of 125X, and 1,000X. The SEM s were used to locate and document analysis area

PAGE 75

75 locations and to document surface morphology. The quantification of the elements was accomplished by using the elem ental sensitivity factors. The contact properties were obtained from linear transmission line method (TLM) measurements on 100100 m pads with spacing 5, 10, 20, 40, and 80 m. The contact resistance R C was obtained from the relation 49 R C = ( R T S d / Z )/2 where R T is the total resistance between two pads, S is the sheet resistivity of the semiconductor under the contact, d is the pad spacing, and Z is the C is then obtained C = R C L T Z where L T is th e transfer length obtained from the intercept of a plot of R T vs d Results and Discussion Figure 4 1 shows the contact resistances as a function of anneal temperature for the three boride based schemes. The as deposited contacts showed rectifying behavior as expected for any as deposited contacts on the wide bandgap GaN 58 The current voltage characteristics became Ohmic for anneal 500 C The contact resistance decreased up to ~ 800 900 C depending on the boride employed. The TiB 2 containing contacts show the lowest contact resistance of 1.6x 10 6 2 The minimum in the contact resistance with annealing temperatu re is most likely related to the formation of low resistance phases of TiN at the interface with the GaN, as reported for conventional Ti/Al/Pt/Au contacts 20 22 26 28, 30, 32 and 64 Annealing at higher temperatures leads to higher contact resistance, w hich as will be seen later corresponds to extensive intermixing of the contact metallurgy. The contact properties did not show a significant dependence on annealing time at 850 C, as shown in Figure 4 2. Figure 4 3 shows the measurement tempe rature depend ence of the contacts on n GaN annealed at 850 C W ithin the error of the measurement, the contacts exhibited almost constant specific contact resistance in the temperature range of 25 200 o C, indicating that current flow is

PAGE 76

76 dominated by tunneling. When th e tunneling dominates, the specific contact resistivity (R SCR ) is dependent upon doping concentration and is given as (4 1) where B is the barrier height, S the semiconductor permittivity, the effective mass of electrons, D is the donor concentration in the semiconductor. The tunneling may be related to the formation of the TiN X phases, as is the case with conventional Ohmic contacts on n GaN. Figure 4 4 shows the SEM of the contact morphology for the three metallization schemes before and after annealing at 800 or 1000 C The morphology is featureless until 800 C which corresponds to the minimum in contact resistance. By 100 0 C the morphology becomes rough er and this corresponds to the increased contact resistance AES surface scans showed only the presence of c arbo n, oxygen and gold on the as deposited surface. The carbon is adventitious and the oxygen comes from a thin native oxide on the Au. Figure 4 5 shows the AES depth profiles for the as deposited and annealed Ti/Al/CrB 2 /Ti/Au samples. The profile obtained from the as deposited sample was in good agreement with the prescribed metal layer thicknesses. The profile from the sample annealed at 7 00 C shows diffusion of titanium through the gold layer to the surface and of the Ti layer near the GaN through the Al above it. The profile obtained from the sample annealed at 800 C shows significant inter diffusion of layers. The profile now shows the pres ence of a thin titanium layer, followed by a thin gold layer on top of the Cr 2 B layer, an oxidized aluminum layer, a gold layer, and a final titanium layer. The profile obtained from the sample annealed at 1000 C shows a similar basic structure. Some of t he transport might be attributed to grain boundary diffusion, as reported previously for Mo based contacts to AlGaN/GaN heterostructures 39

PAGE 77

77 Similar data is shown in Figures 4 6 and 4 7 for TiB 2 and W 2 B 5 for various anneal temperatures. Once again, the mov ement of the two Ti layers is the major effect present in both types of contact. The W 2 B 5 is the least thermally stable of the schemes, as judged by the more extensive intermixing at lower temperatures. Figure 4 8 shows the room temperature contact resist ance of the sample s annealed at 80 0 C as a fun ction of time spent at 350 C This simulates the operation of an uncooled GaN based transistor and gives some idea of the expected stability of the contact. We have also included data from conventional Ti/Al /Ni/Au contacts for comparison. Note that the latter shows the lowest initial contact resistance, but then has an increase of almost an order of magnitude after 9 days of elevated temperature operation. By sharp contrast, all of the boride based contacts s how less change with aging time and have lower contact resistances than the Ti/Al/Ni/Au after 22 days aging at 350 C This suggests that the improved stability of the borides relative to Ni has some beneficial effect on the long term stability of the cont acts. Summary and Conclusions A Ti/Al/X/Ti/Au metallization scheme, where X was either W 2 B 5 CrB 2 or TiB 2 was used to form Ohmic contacts to n type GaN. A minimum contact resistance of 1.5x10 6 2 was achieved for the TiB 2 based scheme at an annealing temperature of 850 900 C For W 2 B 5 the minimum contact resistance was ~1.5x10 5 2 at 800 o C while for CrB 2 it was 8x10 6 2 at 800 o C.The latter value is comparable to that achieved with conventional Ti/Al/Ni/Au on the same samples. The contacts showed much less change in resistance over a period of more than 22 days at 350 C than for Ti/Al/Ni/Au and the boride based contacts look promising for high temperature device applications.

PAGE 78

78 Tabl e 4 1. Se lected properties of potential boride c ontacts on GaN. Properties TiB 2 ZrB 2 W 2 B W 2 B 5 CrB 2 Melting Point ( o C) 2980 ~3225 3040 ~3200 ~2670 ~2385 2200 Structure hexagonal hexagonal hexagonal hexagonal Thermal expansion coefficients x 10 6 ( /deg) 4.6 5.9 10.5 Phonon component of heat conduction (wt/m deg.) 20.6 18.9 10.4 Elastic modulus E x 10 6 (kg/cm 2 ) 5.6 4.3 2.5 Characteristic Temperature ( o K) 1100 765 726 Density of electronic states g x 10 21 (eV 1 cm 1 ) 4.50 4.76 54 .6 Work function (eV) (approx.) 4.19(?) 3.94(?) 3.18(?) Heat of Formation (Kcal/mole) 71.4 76.0 31.0 Lattice constant(A) 3.028 3.169 2.982 2.969 Thermal conductivity(W.m 1 K 1 ) 26 80 unknown 32 Electrical resistivity( Ohm.c m) 28 4.6 19 21

PAGE 79

79 Figure 4 1 Specific contact re sistance of Ti/Al/boride/Ti/Au o hmic contacts on n GaN as a function of anneal temperature.

PAGE 80

80 Figure 4 2 Specific contact re sistance of Ti/Al/boride/Ti/Au o hmic contacts on n GaN as a function of anneal time at the optimum anneal temperature for each type of metal scheme.

PAGE 81

81 Figure 4 3 Specific contact re sistance of Ti/Al/boride/Ti/Au o hmic contacts on n GaN as a function of measurement temperature at the optimum anneal temperatures.

PAGE 82

82 Figure 4 4 SEM micrographs. A) As deposited Cr 2 B. B) As deposited TiB 2 C) As deposited W 2 B 5 D) Cr 2 B annealed at 800 C E) TiB 2 anneale d at 800 C F) W 2 B 5 annealed at 800 C G) Cr 2 B annealed at 1000 C H) TiB 2 annealed at 1000 C I) W 2 B 5 annealed at 1000 C A B C D E F G H I

PAGE 83

83 Figure 4 5 AES depth profiles of CrB 2 based contacts. A) As deposited. B) Annealed at 700 C C) Annealed at 800 C D) Annealed at 1000 C A B C D

PAGE 84

84 Figure 4 6 AES depth p rofiles of Ti B 2 based contacts. A) As deposited. B) Annealed at 600 C C) Annealed at 800 C D) Annealed at 1000 C A B C D

PAGE 85

85 Figure 4 7 AES depth p rofiles of W 2 B 5 based contacts A) As deposited. B) Annealed at 500 C C) Annealed at 700 C D) Annealed at 1000 C A B C D

PAGE 86

86 Figure 4 8 Specific contact resistance of the boride based contacts annealed at 800 o C and the convent ional Ti/Al/Ni/Au contacts as a function of aging time at 350 C

PAGE 87

87 CHAPTER 5 Zr B 2 AND W 2 B SCHOTTKY DIODE CONTACTS ON N G a N Introduction AlGaN/GaN high electron mobility transistors (HEMTs) have potential application in microwave power amplifiers for rada r and communication systems over a broad frequency range from S band to V band .One critical requirement for commercialization of these systems is the need for more reliable and thermally stab le s chottky contacts on n type GaN 1 3 18,25,26,30 40 The antici pated operation of these amplifiers under uncooled, high temperature conditions emphasizes that the thermal stability of the contact metallurgy is paramount. An alternative approach for the gate contact is the use of metal oxide semiconductor (MOS) gates, though much of that work is in its infancy and the metal gate is still the dominant technology. The HEMT device structure is relatively simple and the reliability is determined by the stability of gate and source/drain contacts and surface and buffer layer trapping effects. In GaN as in other compound semiconductor systems, the strength of interfacial reactions between the metal and semiconductor plays a key role in determining the quality of t he resultant s cho ttky barriers. The most common s chottky metalli zation for AlGaN/GaN HEMTs is based on Pt/Au or Ni/Au. Metallurgy systems with high melting temperatures and good thermodynamic stability such as W (eg.W, WSi X ) 37, 38, 65, 66 show potential for improved thermal characteristics on GaN. Tungsten based schem es have been used for both rectifying (W/Ti/Au; WSi X /Ti/Au) 35, 36, 65 67 and o hmic (Ti/Al/Pt/ W/Ti/Au) 40 contacts on GaN H EMTs. Sputter deposited pure W s chottky contacts on n GaN show as deposited barrier heights ( b ) of 0.80 eV for optimized conditions 68 Subsequent annealing at 500 600 C reduces the barrier height to ~0.4 eV, its theoretical value from the relation b = m s (where m is the metal work function and s the electron affinity of GaN).

PAGE 88

88 Another promising metallization system is based on bo rides of Cr, Zr, Hf, Ti or W 46 Stoichiometric diborides have high melting temperatures (eg. 3200 C for ZrB 2 ) and thermodynamic stabilities at least as good as comparable nitrides or silicides 47 These have been suggested as metal gates in Si complement ary metal oxide semiconductor (CMOS) integrated circuits 47 .One particularly attractive option is ZrB 2 which has a low resistivity in the range 7 10 2 on n GaN and W 2 B which is a refractory material has also not been explored as a contact to GaN. Recently, it was shown that hexagonal ZrB 2 (0001) single crystals have an in plane lattice constant close to that of GaN, prompting efforts at GaN heteroepitaxy on ZrB 2 69 72 or buff er layers on Si substrates 73 In terms of contacts on GaN, the only related wor k is the study of ZrN/Zr/n GaN o hmic structures 48 in which the Zr/GaN interface was found to have excellent thermal stability. A potential drawback for application of ZrB 2 con tacts on GaN is that the barrier height might be lower than the more conventional schemes, given the work function of ZrB 2 is ~3.94 eV and the electron affinity of GaN is ~4.1 eV. In this chapter a report on the annealing temperature dependence of barrier height and contact intermixing of ZrB 2 /Ti/Au and W 2 B/Ti/Au on n GaN is presented. The ZrB 2 based contacts show a maximum barrier height of 0.55 eV and maintain a barrier height >0.5 eV to at least 700 C and the W 2 B based contacts show a maximum barrier he ight of 0.5 eV, with a negative temperature coefficient, and are stable against annealing up to ~500 C. Experimental The samples used were 3 m thick Si doped GaN grown by Metal Organic Chemical Vapor Deposition on c plane Al 2 O 3 substrates. The electron c oncentration obtained from Hall measurements was ~ 5x10 17 cm 3 A s chottky metallizatio n scheme of ZrB 2 (500 ) / Ti ( 200 )

PAGE 89

89 / Au ( 8 00 ) was us ed in all experiments. The Au was added to lower the contact sheet resistance, while the Pt is a diffusion barrie r. All of the metals were deposited by A r plasma assisted rf sputtering at pressures of 15 40 mTorr and rf (13.56 MHz) powers of 200 250 W. The contacts were patterned by liftoff and annealed at temperatures up to 700 C for 1 min in a flowing N 2 ambient i n a RTA furnace. For o hmic contacts, we used the standard e beam deposited Ti/Al/Pt/Au annealed at 850 C for 30 s ecs prior to deposition of the s chottky metallization. A ring contact geometry for the diod es was employed. Figure 5 1 A shows a scanning elec tron microscopy (SEM) image of the as deposited W 2 B based contacts. Auger Electron Spectroscopy (AES) depth profiling of the as deposited contacts showed sharp interfaces between the various metals The AES system was a Physical Electronics 660 Scanning Au ger Microprobe. The electron beam conditions were 10 keV, 1 A beam current at 30 from sample normal. For depth profiling, the ion beam conditions were 3 keV Ar + 2.0 A ( 3 mm ) 2 raster, with sputter rate of~160 / minute Prior to AES data acquisition, s econdary electron images were ob tained from the sample. These images were used to locate and document analysis area locations and to document surface morphology. The quantification of the elements was accomplished by using the elemental sensitivity factors The contact properties were obtained from I V characte ristics of the ZrB 2 /Ti/Au diodes measured at 300 K and W 2 B/Ti/Au diodes measured over the temperature range 25 150 C using a probe station and Agilent 4145B parameter analyzer. We fit the forward I V characteristics to the relation for the thermionic emission over a barrier 74 5 1

PAGE 90

90 where J is the current density, A GaN, T the absolute temperature, e the electronic charge, b the barrier fa ctor and V the applied voltage. Results and Discussion W 2 B Based Rectifying C ontacts Figure 5 2 shows the extracted barrier height and reverse breakdown voltage as a function of measurement temperature for as deposited W 2 B/Ti/Au contacts on n GaN. From the data, b was obtained as 0.55 eV for the as deposited W 2 B at 25 C and ~0.4 5 eV at 150 C. The data can be fit to yield a negative temperature coefficient for barrier height of 8 x10 4 eV/C over the rang e 25 150 C The forward I V characteristics in each case showed the ideality factor was > 2 suggesting transport mechanisms other than thermionic emission, such as recombination. The reverse breakdown voltage also shows a negative temperature coefficient which may be due to contributions from the reduced barrier height and also from the high defect density in the heteroepitaxial GaN on sapphire. Defect free GaN is expected to exhibit a positive temperature coefficient for breakdown The reverse leakage d epends on both bias and temperature. From a moderately doped sample of the type studied here, we would expect thermionic emission to be the dominant leakage current mechanism 75 According to image force barrier height lowering, this leakage current density J L can be written as ( 5 2 ) where B is the image force barrier height lowering, given by where E M is the electric field strength at the metal/semiconductor interface and S is the permittivity. The experimental dependence of J L on bias and temperature is stronger than predicted from Eq uation

PAGE 91

91 5 2 The large bandgap of GaN makes the intrinsic carrier concentration in a depletion region very small, suggesting that contributions to the reverse leakage f rom generation in the depletion region are small. Therefore, the additional leakage must originate from other mechanisms such as thermionic field emission or surface leakage. Figure 5 3 shows the barrier height and reverse breakdown voltage as a function o f annealing temperature. For anneals at 00 C, the rectifying nature of the W 2 B contacts was phase W 2 N, as reported previously for W and WSi X on GaN 37,65 The improvement in breakdown voltage at intermediate annealing temperatures may be due t o annealing of sputter damage in the near surface of the GaN. AES depth profiles of the as deposited and 700 C annealed contacts are shown in Figure 5 4.The as deposited layers (Figure 5 4, A ) exhibit relatively sharp interfaces, consistent with the excel lent surface morphology evident in the SEM picture of Figure 5 1 A By sharp contrast, the depth resolution associated with the anneal ed sample ( Figure 5 4 B ) is poorer than the as deposited sample. The bubbling of the film evident in the SEM of Figure 5 1 B might be the source of the degradation in depth resolution. The Ti becomes oxidized upon annealing and separate x ray diffraction experiments showed the formation of phase W 2 N at this temperature. Since this phase has been associated with improved oh micity of W based contacts on GaN 37, 65 it is no surprise that the rectifying nature of the contact is degraded at this temperature. ZrB 2 Based Rectifying C ontacts Figure 5 5 shows the I V characteristics from the ZrB 2 /Ti/Au/GaN diodes as a function of p ost deposition annealing temperature. The as deposited sample displays an almost symmetrical characteristic, suggesting that sputter damage dominates the current transport. With subsequent annealing even at 200 C the reverse breakdown voltage is improved and higher temperatures

PAGE 92

92 increase the reverse current. Figure 5 6 shows the extracted barrier height and reverse breakdown voltage as a function of measurement temperature for as deposited Zr 2 B/Ti/Au contacts on n GaN. From the data, b was obtained as 0.5 2 eV for the as deposited Zr 2 B at 25 C with a maximum value of 0.55 eV after annealing at 200 C The barrier height stayed above 0.5 eV until at least 700 C anneal temperature. While higher barrier heights would be desirable for HEMT operation, there m ay be applications for the ZrB 2 where extended high temperature operation is t he most important requirement. In Figure 5 6, the improvement in breakdown voltage at intermediate annealing temperatures may be due to annealing of sputter damage in the near su rface of the GaN. The forward I V characteristics in each case showed the ideality factor was > 2 suggesting transport mechanisms other than thermionic emission, such as recombination The reverse leakage depended on both bias and annealing temperatur e. A s explained in the case for W 2 B schottky contact, for the doping level employed here we would expect thermionic emission to be the dominant leakage current mechanism 75 Similarly, given the low intrinsic carrier concentration of GaN, t he additional leaka ge must originate from mechanisms such as thermionic field emission or surface leakage. SEM micrographs of the contact stack morphology are shown in Figure 5 7 as a function of anneal temperature. The contacts retain a smooth morphology even at 500 C whe re the o hmic contacts already shown significant roughening. The corresponding AES surface scans are shown in Figure 5 8. The as deposited sample shows only Au on its surface, as expected. After the 350 C anneal, Ti is evident and its concentration is incr eased after the 500 C anneal. Table 5 1 summarizes the near surface composition data obtained from AES. The essential message from this data is that the Ti outdiffuses and becomes oxidized on the surface.

PAGE 93

93 AES depth profiles of the as deposited 350 o C anne aled and 700 C anneale d contacts are shown in Figure 5 9 The as deposited layers (Figure 5 9, A ) exhibit relatively sharp interfaces, consistent with the excellent surface morphology eviden t in the SEMs There is a significant amount of oxygen in the ZrB 2 layer consistent with past reports that the borides are getters for residual water vapor in the ambient during sputter deposition 47 After the 700 C anneal, the extent of Ti outdiffusion is increased and the Ti becomes oxidized upon annealing. X ray d o crystal system did not show any reaction of the ZrB 2 with the GaN even at 800 C as shown in Figures 5 10 and 5 11. This is in sharp contrast to the case of W 2 B contacts on GaN, w here phase W 2 N is formed f or anneals at 00 C. In this latter case, the rectifying nature of the W 2 B contacts was phase W 2 N, as reported previously for W and WSi X on GaN 37, 65 Since this phase has b ee n associated with improved o hmicity of W based contacts on GaN 37, 65 the degradation of rectifying behavior of the contact is expected at this temperature. However, the ZrB 2 shows a much slower reaction with the GaN than W 2 B and the XRD spectra show no f ormation of nitride or gallide phases. Summary and Conclusions In conclusion, ZrB 2 and W 2 B exhibits a barrier height of ~0.5 eV on GaN. This is rather low for HEMT gates, but it may have use in applications where thermal stability is more important than ga te leakage current such as HEMT gas sensors. The ZrB 2 /GaN interface is stable against annealing at 800 C and the contact stability is still determined by outdiffusion of Ti from the ZrB 2 /Ti/Au stack. Both borides appear to be an efficient getter of water vapor during sputter deposition.

PAGE 94

94 Table 5 1 Near surface composition data obtained from AES measurements. Sample ID C(1) O(1) Ti(2) Au(3) Sensitivity factors [0.076] [0.212] [0.000] [0.049] As deposited 47 4 nd 49 Annealed @350 C 48 16 3 33 Ann ealed @700 C 45 31 14 10

PAGE 95

95 Figure 5 1 SEM micrograph s of W 2 B based schottky contacts. A) A s deposited B) Annealed at 700 C .The inner circle is the W 2 B/Ti/ Au while the outer ring is the o hmic contact. A B

PAGE 96

96 Figure 5 2 Barrier height and reverse breakdown voltage as a function of measurement temperature for as deposited W 2 B/Ti/Au contacts on n GaN

PAGE 97

97 Figure 5 3. Barrier height and reverse breakdown voltage as a function of annealing temperature for W 2 B/Ti/Au contacts on n GaN.

PAGE 98

98 Figure 5 4 AES depth profiles of W 2 B/Ti/Au on GaN. A) Unannealed. B) A fter annealing at 700 C A B

PAGE 99

99 Figure 5 5 I V characteristics from ZrB 2 /GaN diodes as a function of post deposition annealing temperature.

PAGE 100

100 Figure 5 6 Barrier height and reverse breakdown voltage as a function of annealing temperature for ZrB 2 /Ti/Au contacts on n GaN.

PAGE 101

101 Figur e 5 7 SEM micrograp hs of ZrB 2 based schottky contacts. A) A s deposited B) Annealed at 350 C C) Annealed at 700 C .The inner circle is the ZrB 2 /Ti/ Au while the outer ring is the o hmic contact. Sample 1: As d eposited Sample 2: Annealed at 350 o C Sample 3 : Annealed at 70 0 o C A B C

PAGE 102

102 Figure 5 8 AES surface s cans of ZrB 2 /Ti/Au on GaN A) A s deposited B ) Annealed at 350 C C) Annealed at 700 C A B C

PAGE 103

103 Figure 5 9 AES depth profiles of ZrB 2 /Ti/Au on GaN A) As deposited. B) Annealed at 350 C C) Annealed at 700 C A B C

PAGE 104

104 Figure 5 1 0 Powder XRD spectrum from ZrB 2 on GaN. A) Unannealed. B) A fter annealing at 800 C A B

PAGE 105

105 Figure 5 11 Glancing angle XRD spectra from ZrB 2 on GaN. A) Unannealed. B) A fter annealing at 800 C A B

PAGE 106

106 CHAPTER 6 ANNEALING TEMPERATURE DEPENDENCE OF Ti B 2 W 2 B 5 AND Cr B 2 S CHOTTKY BARRIER CONTACTS ON N Ga N Introduction Th e availability of reproducible s chottky contacts on GaN or AlGaN is critical to the operation of AlGaN/GaN high electron mobility transistors (HEMTs) for advanced microwave power amplifiers in ra dar and communication systems 1 3 40, 65 67 One of the major remaining hurdles in commercializing reliable HEMT based systems is the need for thermally stable rectifying contacts, since it is anticipated that some power amplifiers may require operating tem peratures up to 300 C While it might be expected that metal oxide semiconductor (MOS) gates would provide superior therma l stability compared to simple s chottky metal gates, this would degrade the rf performance due to the extra capacitance and MOS techn ology is still at a relatively primitive stage for GaN devices. Some standard metallization systems such as Au on AlGaN show an environmental aging effect 16 Typical s chottky metallization for AlGaN/GaN HEMTs are based on Pt/Au or Ni/Au, with the Au inclu ded to reduce the sheet resistance of the contact and to prevent oxidation of the other metal 20 30, 32, 33, 35 45, 64 There is still a need to investigate a w ider range of thermally stable s chottky contacts on GaN in the search for alternatives to Pt/Au or Ni/Au. Other potentially more thermally stable metallization schemes have been reported based on W or WSi x 37, 38, 65, 66 These exhibit low barrier heights around 0.4 0.5 eV. Another potential class of thermally stable contacts are those based on borid es, which have not attracted much attention for use on GaN. The stoichiometric diborides are thermally stable with very high melting temperatures, well in excess of those of both Ni and Pt 68, 74 They also exhibit good corrosion resistance but are suscept ible to oxidation during thermal processing 68, 74 This may be countered by depositing an overlayer of a metal such as Au in the same deposition chamber.

PAGE 107

107 In this chapter the electrical characteristics, annealing and measurement temperature dependence of barrier height and stability of TiB 2 /Ti/Au, W 2 B 5 /Ti/Au and CrB 2 /Ti/Au contacts on n GaN were studied. The TiB 2 contacts show a maximum barrier height of 0.68 eV after annealing at 350 C. W 2 B 5 and CrB 2 are high temperature stable refractory materials whi ch have not been explored as a contact to GaN. These contacts show a maximum barrier height of 0.65 eV and 0.63 eV, respectively, with a small negative temperature coefficient, and are reasonably stable against annealing to ~350 C. This barrier height is lower than for Ni or Pt and thus one would need to balance the need for improved thermal stability with the poorer rectifying properties. Experimental The samples employed were 3 m thick Si doped GaN grown by Metal Organic Chemical Vapor Deposition on c p lane Al 2 O 3 substrates. The electron concentration obtained from Hall measurements was ~ 3x10 17 cm 3 for samples used for TiB 2 and ~ 5x10 17 cm 3 for samples used for W 2 B 5 and CrB 2 schottky contacts. The surfaces were cleaned by sequential rinsing in acetone ethanol and 10:1 H 2 O : HCl prior to insertion in the sputtering chamber. A metallization scheme of X ( 500 ) / Ti ( 200 ) / Au ( 800 ) was used in all experiments where X was either TiB 2 W 2 B 5 or CrB 2 The Au was added to lower the contact sheet resistance while the pure Ti is a diffusion barrier. All of the metals or compounds were deposited by Ar plasma assisted rf sputtering at pressures of 15 40 mTorr and rf (13.56 MHz) powers of 200 250 W. The sputter rates were held constant at 1.4 .sec 1 for all of the metals or compounds. The contacts were patterned by liftoff of lithographically defined photoresist and separate samples were annealed at temperatures of 200,350,500 or 700 C for 1 min in a flowing N 2 ambient in a RTA furnace. For o hmic contacts, we used e beam deposited Ti ( 200 )/ Al ( 800 )/ Pt ( 400 )/ Au ( 1500 )

PAGE 108

108 annealed at 850 C for 30 s ecs prior to deposition of the s chottky metallization. A ring contact geometry for the diodes was employed, with the s chottky contacts of diame ter 50 surrou nded by the o hmic contacts with diameter 250 s chottky spacing was Auger Electron Spectroscopy (AES) depth profiling of the as deposited contacts showed sharp interfaces between the various metals. For the AES analysis, the samples were mounted on a stainless steel puck and placed in the system load lock. After chamber pump down, the sample puck was inserted into the analytical chamber and placed in front of the analyzer. The AES system was a Physical Electronics 660 Scanning Auger M icroprobe. The electron beam conditions were 10 keV, 1 A beam current at 30 from sample normal. Charge correction was performed by using the known position of the C (C, H) line in the C 1s spectra at 284.8 eV. The AES spectrometer was calibrated using a polycrystalline Au foil. The Au f 7/2 peak position was determined to be 84.000.02 For depth profiling, the ion beam conditions were 3 keV Ar + 2.0 A ( 3 mm) 2 raster. The quantification of the elements was accomplished by using the elemental sensitivity f actors. We also used Scanning Electron Microscopy (SEM) to examine contact morphology as a function of annealing temperature. The contact properties were obtained from I V characteristics of the TiB 2 /Ti/Au, W 2 B 5 /Ti/Au or CrB 2 /Ti/Au diodes measured over the temperature range 25 150 C using a probe station and Agilent 4145B parameter analyzer. We fit the forward I V characteristics to the relation for the thermionic emission over a barrier 74 ( 6 1) where J is the current density A GaN (26.4 Acm 2 K 2 ) T the absolute temperature, e the electronic charge, b

PAGE 109

109 n the ideality factor and V the applied voltage. The reverse breakdown voltage was defined as the voltage at which the current density was 1 mA.cm 2 Results and Discussion TiB 2 Based S chottky C ontact I V characteristics obtained from the diodes annealed at different temperatures are shown in Figure 6 1. The extracted barrier height and reverse breakdown voltage as a function of annealing temperature for TiB 2 /Ti/Au contacts on n GaN derived from this data are shown in Figure 6 2. From the data, b was obtained as 0.65 eV for the as deposited (i e. control sample) TiB 2 at 25 C The barrier height increases with anneal temperature up to 350 C reaching a maximum value of 0.68 eV. The work function of sputtered TiB 2 is not available, but both Ti and B have work functions around 4.3 eV, while the electron affinity of GaN is 4.1 eV and thus we might expect a low intrinsic barrier height for the compound on GaN. The work function of chemically vapor deposited TiB 2 is reported to be in the range 4.75 5 eV 47, 85 We would also expect that the contact properties of the as deposited compound would be domi nated by residual sputter damage from the deposition of the contacts (which tends to increase the near surface n type conductivity) and once this is annealed out (at ~600 C in this case), the intrinsic c ontact properties are revealed. Higher anneal temper atures led to a reduction in barrier height, most likely associated with the onset of metallurgical reactions with the GaN. The reverse breakdown voltage shows a similar trend to the barrier height, going through a maximum where the barrier height is also a maximum. The forward I V characteristics in each case showed the ideality factor was in the range 2 2.5, suggesting transport mechanisms other than thermionic emission, such as recombination and surface contributions. Figure 6 3 shows the AES surface sca ns from these same samples, confirming the onset of Ti outdiffusion by 350 C After the 700 C anneal, the more extensive Ti outdiffusion leads to

PAGE 110

110 oxidation of the top surface of the contact. This is reflected in the summary of the near surface compositio n data in Table 6 1. Note that there is also a small amount of Ga outdiffusion from the GaN to the surface. The carbon signal comes from adventitious carbon on the surface. AES depth profiles of the as deposited and annealed contacts are shown in Figure 6 4. The as deposited layers (Figure 6 4, A ) exhibit relatively sharp interfaces, consistent with the good surface morphology evident in the SEM pictures described later. The depth resolution of the 350 C annealed sample (Figure 6 4, B ) is slightly poorer t han the as deposited sample, with clear outdiffusion of Ti. After 700 C annealing, the Ti shows more significant outdiffusion to the sur face where it becomes oxidized. The change in interface abruptness at the metal/GaN interface suggests the change in ef fective barrier height at higher annealing temperatures may result from reactions at the TiB 2 /GaN interface. This is consistent with the outdiffusion of Ga seen in the AES surface scans. Note also that the oxygen signal increases on the annealed contacts, consistent with the past observation that boride contacts are susceptible to oxidation 47, 85 This occurs even though the annealing environment was purified, filtered N 2 It is not clear what effect this has on the contact properties, although samples ann ealed in Ar environments and left to cool completely before removal from the RTA furnace showed similar electrical contact properties and thus to first order, a small amount of oxidation may not be that critical in changing the contact properties. Figure 6 5 shows SEM images of the contacts both before (A ) and after annealing at either 350 (B ) or 700 C (C ). The inner contact is the TiB 2 /Ti/Au, while the outer ring is the Ti/Al/Pt/Au o hmic contact. To the resolution of the SEM, the morphology does not chan ge over this annealing range, al though the oxidation of the Ti after the 350 and 700 C anneal leads to a darker appearance of the rectifying contact.

PAGE 111

111 The I V characteristics of the as deposited contacts were measured as a function of measurement temperatu re up to 150 C The extracted barrier height showed no measurable temperature dependence in the range available to us, as shown in Figure 6 6 The reverse breakdown voltage also shows very little temperature dependence. The reverse leakage was found to de pend on both bias and temperature. From a moderately doped sample of the type studied here, we would expect thermionic emission to be the dominant leakage current mechanism 74 The experimental dependence of reverse current on bias and temperature was stro nger than predicted from the TiB 2 barrier height. The additional leakage must originate from other mechanisms such as thermionic field emission or surface leakage since the large bandgap of GaN makes the intrinsic carrier concentration in a depletion regio n very small implying generation currents in the depletion region are small. While the initial results with TiB 2 show reasonable thermal stability, there is much more additional work that needs to done to establish the long term reliability of the contacts for HEMT power amplifier or other device applications. This would include additional studies of the interfacial reactions o ccurring and the effect of bias or environmental aging in humid ambient W 2 B 5 Based Schottky C ontact Figure 6 7 shows SEM image of t he contacts both before (A ) and after annealing at either 350 (B ) or 700 C ( C ). The inner contact is the W 2 B 5 /Ti/Au, while the outer ring is the Ti/Al/Pt/Au o hmic contact. The morphology does not change tremendously over this annealing range. More detail ed information on contact reactions can be obtained from the AES measurements. Table 6 2 shows the surface survey data It is clear from this data that Ti shows some outdiffusion through the Au at 350 C and this is more significant after 700 C anneals. A ES depth profil es of the as deposited and anneale d contacts are shown in Figure 6 8 The as deposited layers (Figure 6 8 A ) exhibit relatively sharp interfaces, consisten t with the good

PAGE 112

112 surface morphology evident in the SEM picture of Figure 6 7 A T he dep th resolution of the 350 C anneal ed sample (Figure 6 8 B ) is slightly poorer than the as deposited sample. After 700 C annealing, the Ti shows significant outdiffusion to the surface where it becomes oxidized upon annealing. Separate x ray diffraction ex periments showed the formation of phase W 2 N at this temperature, as occurs with pure W ( and also the related phase W 2 B). The W 2 N phase has been associated with improved o hmic ity of W based contacts on GaN 38, 67 Figure 6 9 shows the I V characteristics obtained from the diodes annealed at different temperatures. The extracted barrier height and reverse breakdown voltage as a function of annealing temperature for W 2 B 5 /Ti/Au contacts on n GaN derived from this data are shown in Figure 6 10. From the data, b was obtained as 0.58 eV for th e as deposited W 2 B 5 at 25 C .The barrier height increases with anneal temperature up to 200 C reaching a maximum value of 0.65 eV. Higher anneal temperatures led to a reduction in barrier height, most likely associated with the onset of metallurgical re actions with the GaN. The barrier height for the W 2 B 5 is higher than for pure W at these moderate anneal temperatures. The reverse breakdown voltage shows a similar trend to the barrier height, going through a maximum where the barrier height is also a max imum. The I V characteristics as a function of measurement temperature for as deposited contacts are shown in Figure 6 11. The reverse leakage was found to depend on both bias and temperature. From a moderately doped sample of the type studied here, we wou ld expect thermionic emission to be the dominant leakage current mechanism 75 According to image force barrier height lowering, this leakage current density, J L can be written as 74 ( 6 2 )

PAGE 113

113 where B is the image force barrie r height lowering, given by where E M is the electric field strength at the metal/semiconductor interface and S is the permittivity. The experimental dependence of J L on bias and temperature is stronger than predicted from e quation ( 2). The large bandgap of GaN makes the intrinsic carrier concentration in a depletion region very small, suggesting that contributions to the reverse leakage from generation in the depletion region are small. Therefore, the additional leakage must originat e from other mechanisms such as thermionic field emission or surface leakage. The extracted barrier height shows only a slight negative temperature coefficient, almost within the experimental error, as shown in Figure 6 12. The forward I V characteristics in each case showed the ideality factor was > 2 suggesting transport mechanisms other than thermionic emission, such as recombination. The reverse breakdown voltage also shows a negative temperature coefficient up to ~100 C Defect free GaN is expected t o exhibit a positive tempera ture coefficient for breakdown 75 but we invariably observe negative temperature coefficients in diodes fabricated with any kind of contact on heteroepitaxial GaN on sapphire with its high defect density. The improvement in bre akdown voltage at intermediate annealing temperatures may be due to annealing of sputter damage in the near surface of the GaN 37 65 CrB 2 Based Schottky C ontact Figure 6 13 shows the I V characteristics obtained from the diodes annealed at different temp eratures. The extracted barrier height and reverse breakdown voltage as a function of annealing temperature for CrB 2 /Ti/Au contacts on n GaN derived from this data are shown in Figure 6 14. From the data, b was obtained as 0.52 eV for the as deposited CrB 2 at 25 C .The barrier height increases with anneal temperature up to 200 C reaching a maximum value of

PAGE 114

114 0.62 eV. Higher anneal temperatures led to a reduction in barrier height, most likely associated with the onset of metallurgical reactions with the G aN. The reverse breakdown voltage shows a similar trend to the barrier height, going through a maximum where the barrier height is also a maximum. AES depth profil es of the as deposited and anneale d contacts are shown in Figure 6 15 The as deposited layer s (Figure 6 15 A ) exhibit relatively sharp interfaces, consisten t with the good surface morphology eviden t in the SEM pictures described later. T he depth resolution of the 350 C anneal ed sample (Figure 6 15 B ) is slightly poorer than the as deposited samp le, with clear outdiffusion of Ti. After 700 C annealing, the Ti shows more significant outdiffusion to the surface where it becomes oxidized. Figure 6 16 shows the AES surface scans from these same samples, confirming the onset of Ti outdiffusion by 350 C After the 700 C anneal, the more extensive Ti outdiffusion leads to oxidation of the top surface of the contact. This is reflected in the summary of the near surface composition data in Table 6 3. Figure 6 17 shows SEM images of the contacts bot h befo re (A ) and after annealing at either 350 ( B ) or 700 C ( C ). The inner contact is the Cr B 2 /Ti/Au, while the outer ring is the Ti/Al/Pt/Au o hmic contact. The morphology does not change tremendously over this annealing range, although the oxidation of the co ntact after the 700 C anneal leads to a darker appearance of the rectifying contact. The I V characteristics were measured as a function of measurement temperature. The extracted barrier height shows very little temperature dependence, almost within the e xperimental error, as shown in Figure 6 18. The forward I V characteristics in each case showed the ideality factor was > 2 suggesting transport mechanisms other than thermionic

PAGE 115

115 emission, such as recombination. The reverse breakdown voltage also shows ver y little temperature dependence. Defect free GaN is expected to exhibit a positive temperature coefficient for breakdown, but we invariably observe negative temperature coefficients in diodes fabricated with any kind of contact on heteroepitaxial GaN on sa pphire with its high defect density. The reverse leakage was found to depend on both bias and temperature. As explained for the case of other two boride schot t ky, t he additional leakage must originate from other mechanisms such as thermionic fi eld emission or surface leakage. The long term reliability of these contacts for the HEMT power amplifier applications needs is tested later in the thesis. Summary and Conclusions The main conclusions of our study may be summarized as follows: W 2 B 5 produces an as depo sited (by sputtering) barrier height of ~0.58 eV on GaN and a maximum value of 0.65 eV after annealing at 200 C TiB 2 produces an as deposited (by sputtering) barrier height of ~0.65 eV on GaN and a maximum value of 0.68 eV after annealing at 350 C CrB 5 produces an as deposited (by sputtering) barrier height of ~0.52 eV on GaN and a maximum value of 0.62 eV after annealing at 200 C This is still lower than for Ni or Pt HEMT gates, but it may have use in applications where thermal stability is more imp ortant than gate leakage current such as HEMT gas sensors. The Boride/Ti/Au contacts show some outdiffusion of Ti at 350 C and much more significant reaction after 700 C anneals. The as deposited contacts show only a minor decrease in barrier height for measurement temperatures up to 150 C Additional experiments need to done to establish the long term reliability of the contacts for the HEMT power amplifier applications. The contacts are quite susceptible to oxidation during thermal processing and care must be used to minimize exposure to oxidizing ambient.

PAGE 116

116 Table 6 1 Concentration of elements detected on the as received s urfaces of TiB 2 based schottky contacts ( Sample ID C(1) O(1) Ti(2) Ga(1) Au(3) Sensitivity factors [0.076] [0.212] [0.188] [0.240] [0.049] Control sample 44 3 2 nd 52 350 C anneal 38 18 9 1 34 700 C anneal 37 23 11 2 16 Table 6 2 C oncentration of elements detected on the as received s urfaces of W 2 B 5 based schottky contacts Sample C O S Ti Au As deposited 48 2 N d N d 50 350 C a nneal 48 13 1 7 31 700 C a nneal 29 41 1 22 7 Table 6 3 Concentration of elements detected on the a s received s urfaces of CrB 2 based schottky contacts Sample ID C(1) O(1) S(1) Ti(1) Ga(1) Au(3) Sensitivity factors [0.076] [0.212] [0.652] [0.188] [0.240] [0.049] As deposited 29 1 nd nd nd 69 Annealed at 350 C 26 19 2 13 1 39 Annealed at 700 C 33 32 1 17 nd 17 ions are normalized to 100%.nd = element not detected. AES detection limits range from 0.1 1.0 atomic percent

PAGE 117

117 Figure 6 1. I V characteristics at 25 C of TiB 2 /Ti/Au on GaN as a function of post deposition annealing temperatu re.

PAGE 118

118 Figure 6 2. Barrier height and reverse breakdown voltage as a function of annealing temperature for TiB 2 /Ti/Au contacts on n GaN.

PAGE 119

119 Figure 6 3. AES surface scans of TiB 2 /Ti/Au on GaN A) As deposited. B) Annealed at 350 C C) Annealed at 700 C A B C

PAGE 120

120 Figure 6 4. AES depth profiles of TiB 2 /Ti/Au on GaN A) As deposited. B) Annealed at 350 C C) Annealed at 700 C A B C

PAGE 121

121 Figure 6 5. SEM micrographs of TiB 2 based schottky contacts A) As deposited. B) Annealed at 350 C C) Annealed at 700 C The inner circle is the TiB 2 /Ti/ Au while the outer ring is the o hmic contact. A B C

PAGE 122

122 Figure 6 6 Barrier height and reverse breakdown voltage as a function of measurement temperature for as deposited TiB 2 /Ti/Au contacts on n GaN. B

PAGE 123

123 Figure 6 7 SEM micrographs of W 2 B 5 based schottky contacts. A) As deposited. B) Annealed at 350 C C) Annealed at 700 C .The inner circle is the W 2 B 5 /Ti/Au while the outer ring is the ohmic contact A B C

PAGE 124

124 Figure 6 8 AES depth profiles of W 2 B 5 /Ti/Au on GaN A) As deposited. B) Annealed at 350 C C) Annealed at 700 C A B C

PAGE 125

125 Figure 6 9 I V characteristics of W 2 B 5 /Ti/Au on GaN as a function of post deposition annealing temperature.

PAGE 126

126 Figure 6 10 Barrier height and reverse breakdown voltage as a function of annealing temperature for W 2 B 5 /Ti/Au contacts on n GaN.

PAGE 127

127 Figure 6 11 I V characteristics of as deposited W 2 B 5 /Ti/Au on GaN as a function of measurement temperature.

PAGE 128

1 28 Figure 6 12 Barrier height and re verse breakdown voltage as a function of measurement t emperature for as deposited W 2 B 5 /Ti/Au contacts on n GaN.

PAGE 129

129 Figure 6 13 I V characteristics of CrB 2 /Ti/Au on GaN as a function of post deposition annealing temperature.

PAGE 130

130 Figure 6 14 Barrier height and reverse breakdown voltage as a function of annealing temperature for CrB 2 /Ti/Au contacts on n GaN.

PAGE 131

131 Figure 6 15 AES depth prof iles of CrB 2 /Ti/Au on GaN A) As deposited. B) Annealed at 350 C C) Annealed at 700 C A B C

PAGE 132

132 Figure 6 16 AES surface scans of CrB 2 /Ti/Au on GaN A) As deposited. B) Annealed at 350 C C) Annealed at 700 C A B C

PAGE 133

133 Figu re 6 17 SEM micrograp hs of CrB 2 based schottky contacts. A) As deposited. B) Annealed at 350 C C) Annealed at 700 C .The inner circle is the CrB 2 /Ti/Au while the outer ring is the ohmic contact A B C

PAGE 134

134 Figure 6 18 Barrier height and reverse breakdow n voltage as a function of measurement temperature for as deposited CrB 2 /Ti/Au contacts on n GaN.

PAGE 135

135 CHAPTER 7 IMPROVED LONG TERM THERMAL STABILITY AT 350 o C OF Ti B 2 BASED OHMIC CONTACTS ON AlGaN/Ga N HIGH ELECTRON MOBILITY TRANSISTORS Introduction There is significant interest in developing new metallization schemes for AlGaN/GaN High Electron Mobility Transistors (HEMTs) intended for applications in power amplifiers and converters with high efficiency (above 70%) for radars and communications systems, hybri d electric vehicles, power flow control and remote sensing systems 1 4,87 95 The achievement of high efficiency microwave operation at elevated temperatures is important from the viewpoint of minimizing the weight and the volume of power stages. AlGaN/GaN HEMTs appear well suited to simultaneously achieving high powers, high frequencies and high efficiencies. The most commonly used metallization scheme for source/drain contacts on these HEMTs is Ti/Al, with over layers of Pt, Ni or Ti and then a layer of Au to reduce oxidation problems and lower the sheet resistance of the contact stack 16, 20,22,26, 28,32,36,39,44, 64 ,96,97 These contacts produce low specific contact resistances when annealed in the 750 900 C range but there are concerns about the long ter m stability during high temperature operation, in part because if the metal layers begin to intermix, a low melting temperature AlAu 4 phase may form that can lead to contact shorting at small electrode separations 44 One possible solution is to use a very high melting point diffusion barrier in place of the Pt, Ni or Ti in the contact stack. For example, Selvanathan et al. 16,39,44,96,97 demonstrated that Ti/Al/Mo/Au contacts on n GaN are stable at 500 and 600 C for 25 hours of aging, but degraded after 10 hours at 750 C We have shown recently that TiB 2 with a melting temperature around 3000 C and reasonabl .cm) shows prom i se as such a diffusion barrier 99 In this chapter the long term aging characteristics at 350 C of Al GaN/GaN HEMTs with different combinations of TiB 2 based contacts, i.e. those with Ti/Al/TiB 2 / Ti/ Au source/drain

PAGE 136

136 metal and Pt/Au or Ni/Au gates and also those with TiB 2 diffusion barriers in both the source/drain and gate contacts is studied. Compared with HEMTs with standard Ti/Al/Pt/Au Ohmic contacts and Pt/Au gate contacts, a number of the different combinations of boride based contacts exhibit superior stability as judged by the change in source drain current at zero gate vo ltage and the transconductance Experimental The layer structures were grown on sapphire substrates by Molecular Beam Epitaxy and employed a low temperature AlN (300 thick) buffer, 2 m of undoped GaN grown at 750 C under Ga rich conditions, 250 of undoped Al 0.2 Ga 0.8 N and a 30 un doped GaN cap. A growth rate of 0.5 1.0 m hr 1 was used for all depositions. Mesa isolation was achieved with Cl 2 /Ar inductively coupled plasma etching (300 W source power, 40 W rf chuck power). Ohmic contacts were formed by lift off of e beam deposited T i/Al/Pt/Au subsequently annealed at 9 0 0 C for 1 min in a flowing N 2 atmosphere in an RTA furnace. A metallization scheme of Ti (200 )/Al (1000 )/ TiB 2 (500 ) / Ti (200 ) /Au (800 ) was used for comparison in these experiments All of the metals were deposited by Ar plasma assisted rf sputtering at pressures of 15 40 mTorr and rf (13.56 MHz) powers of 200 250 W. The contacts were patterned by liftoff and also annealed at 900 C for 1 min. The specific contact resistance derived from separate Transmissi on Line Method measurements was ~2x10 6 2 Schottky gates (1.5 2 00 m 2 ) of sputter deposited Ni/Au, Ni/TiB 2 /Au, Pt/Au or Pt/TiB 2 /Au were also patterned by lift off. The HEMT layout is shown in the schematic of Figure 7 1. The HEMT dc characteristics wer e measured in dc mode using a HP 4145B parameter analyzer. The rf performance of the HEMTs was characterized with a HP 8723C network analyzer using cascaded probes.

PAGE 137

137 The HEMTs were aged for a period of 25 days at 350 C on a heater plate in air, with the el ectrical characteristics measured every 2 3 days. The samples were removed from the heater block, allowed to cool to room temperature and measured before being returned for further aging on the heater. The temperature was measured with a thermocouple attac hed to the heater. Micro Raman scattering measurements to determine st ress state in the boride based o hmic contacts were performed in a backscattering geometry with the 488 nm line of an Ar ion laser. The laser spot size was ~0.8 m and the laser power at the sample was ~ 9 mW. The bandgap of GaN is larger than the incident photon energy, which minimizes laser induced heating. Because of its higher relative intensity in this scattering geometry and its sensitivity to stress, we selected the E 2 2 phonon as a probe to monitor the film stress 98 Results and Discussion Since the contact metal reflected the laser light, it was difficult to get enough signal by Raman Scattering to get a quantitative measure of the residual stress in the contact structure. Figure 7 2 A shows an optical image labeled with the four positions from which Ram an spectra were acquired (Figure 7 2 B ). At the edge of metal contact, we found that there was tensile stress between the Ti/Al/TiB 2 /Au and the GaN, but it was of a similar magnitude to the conventional Ti/Al/Pt/Au metallization. No peeling or other manifestations of large residual stress were observed in any of our experiments. Figure 7 3 shows drain source current as a function of drain source voltage (I DS V DS ) characteristics from an AlGaN/GaN HEMT with conventional Ti/Al/Pt/Au source/drain contacts and Pt/Au gate metal before and after aging for 25 days at 350 C These devices showed a significant decrease (~30%) in source drain current within even 2 days of aging and a similar de crease (~35%) in transconductance. The decrease in current in some of the curves at high

PAGE 138

138 voltages in the un aged sample appears to be due to discharge of traps. The work on Ti/Al/Mo/Au contacts on n GaN showed that they degraded by in diffusion of Mo and A u into the semiconductor and also by oxidation of the contacts 16 39, 44, 96 and 97 Previous work in this regard on o hmic contacts on GaN have also shown that they degrade by intermixing of the contact metallurgy, probably aided in some ca ses by grain b oundary transport Figures 7 4 and 7 5 shows similar I DS V DS characteristics from HEMTs with the Ti/Al/TiB 2 /Ti/Au o hmic contacts and either Pt/Au or Pt/TiB 2 /Au gates (Figure 7 4) or Ni/Au or Ni/TiB 2 /Au gates (Figure 7 5). Several of these combinations show superior thermal stability over the 25 day aging period to the devices with conventional metallization. The boride based Ohmic contacts retained a smoother morphology than the Ti/Al/Pt/Au both before and after aging, as measured by both atomic force micro scopy and optical microscopy. An example is shown in the optical microscope images of Figure 7 6. Figure 7 7 summarizes the data for percentage change in saturated drain source current as a function of aging time for all of the HEMTs with different combina tions of contacts. All of the devices with boride based contacts show superior aging characteristics compared to the conventional devices, with the exception of the HEMTs with Pt/TiB 2 /Au gate contacts. The devices with Pt/Au gate metal and Ti/Al/TiB 2 /Ti/Au source/drain contacts show only a 10% decrease in I DS after 25 days at 350 C Note that all of the devices appear to show a decrease in current that saturates at different levels depending on the metal. This may indicate that the reaction between the GaN and the gate and source/drain metals is limited by both the thermodynamics of each system and by the thickness of the reacting metal. Figure 7 8 shows rf data for the devices with conventional P t/Au gate contacts and Ti/Al//Pt /Au o hmic contacts before agi ng (A ) and for a device with Pt/Au gates and

PAGE 139

139 Ti/Al/TiB 2 /Ti/Au o hmics after (B ) 25 days aging at 350 C The cutoff frequency, f T remains at ~5.5 GHz with the maximum frequency of oscillation, f MAX at ~30 GHz in both cases. After aging the conventional HE MT, we could not obtain reproducible rf data due to difficulties in making consistent contact to the roughened metal surface. Summary and Conclusions Preliminary aging data on AlGaN/GaN HEMTs with TiB 2 diffusion barriers in the source/drain contacts show a higher resistance to degradation than devices with conventional Ti/Al/Pt/Au contacts. Much more work is needed to determine the contact degradation mechanisms and their activation energies and whether aging under bias makes a difference in the contact re liability. The borides are known to be susceptible to oxidation but the presence of the capping layers may reduce the significance of this issue.

PAGE 140

140 Figure 7 1 HEMT layout used in these experiments.

PAGE 141

141 Figure 7 2 Study of Raman spectra from Ti/Al/TiB 2 /Ti/Au contacts on HEMT wafer. A ) Optical micrograph B ) Raman spectra A B

PAGE 142

142 Figure 7 3 I DS V DS characteristics from HEMT with conventional Pt/Au gate contacts and Ti/Al/Pt/Au source/drain contacts before and after aging at 350 C for 25 d ays.

PAGE 143

143 Figure 7 4. I DS V DS characteristics from HEMT with Ti/Al/TiB 2 /Ti/Au s ource/drain contacts. A ) Pt /Au gate contacts before and a fter aging at 350 C for 25 days. B ) Pt/TiB 2 /Au gate contacts before and after aging at 350 C fo r 25 days. A B

PAGE 144

144 Figure 7 5 I DS V DS characteristics from HEMT with Ti/Al/TiB 2 /Ti/Au source/drain contacts. A ) Ni/Au gate contacts before and a fter aging at 350 C for 25 days. B ) Ni/TiB 2 /Au gate contacts before and after aging at 350 C for 25 d ays. A B

PAGE 145

145 Figure 7 6 Opti cal microscopy images of HEMTs. A) C onventional contacts before aging. B ) B oride bas ed source/drain contacts before aging. C) C onventional contacts after aging at 350 C for 25 days D) B oride based source/drain contacts after aging at 350 C for 25 days Regular after Day 0 Boride after Day 0 Regular after Day 25 Boride after Day 25 A B C D

PAGE 146

146 Figure 7 7 Percent change in saturated drain/source current from HEMTs with different combinations of contact metal schemes as a function of aging time at 350 C

PAGE 147

147 Figure 7 8. RF performance of 1.5 200 m 2 gate length HEMTs. A ) HEMT with c onventional met al contacts prior to aging B) HEMT w ith Pt/Au gates and Ti/Al/TiB 2 /Ti/Au source/drain contacts after ag ing at 350 C for 25 day s. A B

PAGE 148

148 CHAPTER 8 Ir BASED SCHOTTKY AND OHMIC CONTACTS ON N Ga N Introduction GaN hi gh electron mobility transistor (HEMT ) power amplifiers have now entered the commercialization stage for use in wireless communications and military applications. There are a lso possible applications in improved automotive radar and power electronics for hybrid electric vehicles and in advanced satellite communication systems. GaN HEMTs can provide the high power, high efficiency, high linearity RF power transistors required i n base stations fo r mobile data network services 13,14,23,24,50,52 57,61 64 One of the major issues with some applications for these power amplifie rs is the need for very stable ohmic and s chottky metal contacts, capable of extended operation at elevated t emperatur es (typically 200C or higher) 25 45,101 107 Most o hmic contact schemes for AlGaN/GaN HEMTs use Ti/Al, with over layers of Ni, Ti or Pt, followed by Au to reduce sheet resistance a nd decrease oxidation during anneal ing to achieve the lowest conta ct resistance needed to achieve the lowe st specific contact resistivity The formation of TiN X phases is integral to the contact formation mechanism. One drawback is the often poor lateral edge definition of the contacts and potential shorting to the gate contact because of flow of the low melting temperature viscous AlAu 4 phase. Similarly, the gate metal must be stable during elevated temperature operation and alternatives to the usual Ni, Pd or Pt with overlayers of Au are attractive. T here is continued i nterest in use of high temperature metals such as W WSi X W 2 B 5 Mo ,V Ir Cu, CrB 2 ZrB 2 and TiB 2 in bo th schottky and o hmic contacts 25 45, 101 107 in an attempt to improve the long term stability at elevated temperatures. In this chapter the annealing t emperature dependence of contact resistance an d contact intermixing of Ti/Al/Ir/ Au o hmic a nd Ir/Au s chottky metals on n type GaN is studied. These contacts show promising l ong term stability compared to the existing standard metal contact

PAGE 149

149 schemes for n GaN Preliminary studies on Ir based contacts on HEMTs have shown the potential fo r improved contact performance 42, 43 Experimental For s chottky contact studies, the GaN samples consisted of 3 m thick Si doped GaN grown by Metal Organic Chemical Vapor Depo sition (MOCVD) on c plane Al 2 O 3 substrates. The electron concentration obtained from Hall measurements was ~ 5x10 17 cm 3 A metallizatio n scheme of Ir (500 ) /Au (800 ) was used in all experiments. The Au was added to lower the contact sheet resistance. For comparison to a more conventional metal scheme, we also fabricated samples with Ni/Au contacts with the same layer thicknesses as the Ir/Au. For o hmic contacts to th ese samples for s chottky studies, we used the standard e beam deposited Ti ( 200 )/ Al ( 400 )/ Pt ( 200 /Au (800 ) annealed at 850 C for 30 s ecs prior to deposition of the s chottky metallization. A ring contact geometry for the diodes was employed, w ith the s chot tky contacts surrounded by the o hmic contacts. The inner contact diameter was 7 For o hmic studies, t he samples used were also 3 m thick Si dope d GaN grown by MOCVD on c plane Al 2 O 3 substrates. The electron concentration obtained from Hall measurements was ~7x10 18 cm 3 2 /Ar Inductively Coupled Plasma Etching to provide electrical isolation of the contact pads. All of the metals were deposited by Ar plasma assisted rf sputtering at pressures of 15 40 mTorr and rf (13.56 MHz) powers of 200 250 W. The contacts were patterned by liftoff and anneale d at 500 1000 C for 1 min in a flowing N 2 ambient in a RTA furnace. C onventional Ti (200 )/Al (400 )/Ni (200 ) /Au (800 ) was compared with a sc heme in wh i c h the Ni was replaced with Ir. Auger Electron Spectroscopy (AES) depth profiling of the as depos ited contacts showed sharp interfaces between the various metals in all contacts. The AES system was a Physical

PAGE 150

150 Electronics 660 Scanning Auger Microprobe. The electron beam conditions were 10 keV, 1 A beam current at 30 from sample normal. For depth prof iling, the ion beam conditions were 3 keV Ar + 2.0 A (3 mm ) 2 rast er Prior to AES data acquisition, secondary electron microscopy images (SEM s) were o btained from the sample. The SEM s were obtained at magnificati ons of 125X, and 1,000X. The SEM s were used to locate and document analysis area locations and to document surface morphology. The quantification of the elements was accomplished by using the elemental sensitivity factors. The s chottky contact properties were obtained from current voltage ( I V ) ch aracte ristics of the Ir/Au and Ni/Au diodes measured over the temperature range 25 150 C using a probe station and Agilent 4145B parameter analyzer. F it ting is done to the forward I V characteristics to the relation for the thermionic emission over a barr ier 74 ( 8 1) where J is the current density, A GaN, T the absolute temperature, e the electronic charge, b factor and V the appli ed voltage. For o hmic studies, t he contact properties were obtained from Circular transmission line method ( C TLM) measurements on circular rings with spacing 5, 10, 15, 20 25, 30 35 and 40 m The outer diameter of the circular pads was fixed at 300 m and inner diameter varied from 220 to 290 m. The specific contact resistance, c was obtained form the circular TLM measurements with the relationships 97 C = R S L T 2 where R T is the total resistance, R S is the sheet resistance, R 1 is the outer radius of the annular gap R O is the inner radius of the annular gap, I O I 1 K O and K 1 are the modified Bessel

PAGE 151

151 functions, L T is the transfer length, and c is the specific contact resistance. The sheet resistance can be obtained by it erative mathematical process. Results and Discussion Schottky Contacts Figure 8 1 shows the I V characteristics obtained from the Ir/Au diodes annealed at different temperatures. The extracted barrier heights as a function of annealing temperature derived from this data are shown in Figure 8 2. From the data, b was obtained as 0.42 eV for the as deposited Ir at 25 C .The barrier height increases with anneal temperature up to 500 C reaching a maximum value of 0.55 eV in the range 500 700 C. The barrier height of Ni/Au contacts was of similar magnitude in this annealing range (0.52 to 0.56 eV). Higher anneal temperatures led to high leakage currents in the Ni/Au contacts, associated with the onset of metallurgical reactions with the GaN. By contrast, the Ir contacts did not show the onset of leakage until anneal temperatures of >900 C. The forward I V characteristics s howed the ideality factor was always higher for Ni ranging from 1.75 for anneals below 350 C to > 2 at higher temperatures, compared to 1. 3 for Ir at anneal temperatures below 350 C and 1.8 between 500 700 C AES depth profil es of the anneale d Ir/Au and Ni/Au contacts are shown in Figure 8 3 .The as deposited layers exhibited sharp interfaces in both cases. T he depth resolution for the 350 C anneal ed samples is similar to that of the as deposited sample s After 700 C annealing, the Ni/Au contact shows significant outdiffusion of Ni to the surfa ce, corresponding to roughening of the contact morphology. By sharp contrast, the Ir/Au contact shows very littl e change after 700 C annealing.

PAGE 152

152 Ohmic C ontacts Figure 8 4 shows the specific contact resistances as a function of anneal temperature for the two o hmic metal schemes The contact resistance decreased up to ~ 900 C in both cases, with a lo west contact resistance of 1.6x 10 6 2 The minimum in the contact resistance with annealing temperature is most likely related to the formation of low resistance phases of TiN at the interface with the GaN, as repor ted for conventional contacts previously 25 38 Annealing at higher tempe ratures leads to higher contact resistance, which as will be seen later corresponds to extensive intermixing of the contact metallurgy. The contact properties did not show a significant depen dence on annealing time at 900 C. Figure 8 5 shows the SEM of th e contact morphology for the two metallization schemes after annealing at 500 or 9 00 C The m orphology is featureless until 9 00 C which corresponds to the minimum in contact resistance. AES surface scans showed only the presence of c arbo n, oxygen and go ld on the as deposited surface. The carbon is adventitiou s and the oxygen originated from a thin native oxide on the Au. Figure 8 6 shows the AES depth p rofiles for the samples corresponding to the SEM images in the previous figure. The profiles from the s amples annealed at 5 00 C shows significantly less diffusion of Al through the gold layer to the surface in the case of an Ir interlayer and the Ni itself is mobile at 500 C The profile s obtained from the sample s annealed at 9 00 C shows significant inte r diffusion of all the layers Figure 8 7 shows the measurement tempe rature dependence of the contacts on n GaN annealed at 850 C The contacts showed an increased specific contact resistance in the temperature range 320 500 K, most likely due to increase d sheet resistivity of the GaN as the carrier mobility decreases. The doping in the n GaN is not high enough to have the current flow dominated by tunneling. When the tunneling dominates, the specific contact resistivity (R SCR ) is dependent upon doping con centration and is basically independent of temperature, i e.

PAGE 153

153 where B is the barrier height, S the semiconductor permittivity, the effective mass of electrons, N D is the donor concentration in the semiconductor. Figure 8 8 shows the room temperature contact resistance of the sample s annealed at 90 0 C as a fun ction of aging time spent at 350 C This simulates the operation of an uncooled GaN based transistor a nd gives some idea of the expected stability of the contact. The conventional Ti/Al/Ni/Au has an increase in specific contact resistance approximately one and a half orders of magnitude after13.5 days of elevated temperature operation. The contacts showed high resistance (rectifying) behavior beyond this point. By sharp contrast, the Ir containing contacts show less change with aging time and exhibited a stable specific contact resistance of ~10 5 2 after 22 days aging at 350 C This suggests that the Ir restricts some of the contact reaction at 350 C relative to Ni and has a beneficial effect on the long term stability of the o hmic contacts. Summary and Conclusions The replacement of Ni by I r in both ohmic and s chottky contacts to n GaN improves the thermal stability of both types of contacts. The o hmic contacts exhibit superior stability during aging at 350 C while the s chottky contacts show less intermixing of the metals after annealing at 700 C The Ir may be a superior choice to boride contact schemes for GaN, since the latter are prone to oxidation

PAGE 154

154 Figure 8 1 I V characteristics from Ir/Au Schottky contacts on n GaN.

PAGE 155

155 Figure 8 2 Schottky barr ier height for Ir/Au contacts on n GaN as a function of annealing temperature.

PAGE 156

156 Figure 8 3 AES depth profiles. A ) Ir/Au after annealing at 350 C B ) Ir/Au after annealing at 700 C C ) N i/Au contact s after annealing at 350 C D ) N i/Au contacts after annealing at 700 C A B C D

PAGE 157

157 Figure 8 4. Specific contact resistance of Ti/Al//Ni/Au and Ti/Al/Ir/Au Ohmic contacts on n GaN as a function of anneal temperature.

PAGE 158

158 Figure 8 5 SEM images Ir and Ni based ohmic. A ) Ti/Al/Ir/Au after annealing at 500 C B ) Ti/Al/Ni/Au after annealing at 500 C C ) Ti/Al/Ir/Au after annealing at 900 C D ) Ti/Al/Ni/Au after annealing at 900 C A B C D

PAGE 159

159 Figure 8 6 AES depth profiles of Ir and Ni based ohmic. A ) Ti/Al/Ir/Au after annealing at 500 C B ) Ti/Al/Ni/Au after annealing at 500 C C ) Ti/Al/Ir/Au after annealing at 900 C D ) Ti/Al/Ni/Au a fter annealing at 900 C A B C D

PAGE 160

160 Figure 8 7 Specific contact resistance o f Ti/Al//Ni/Au and Ti/Al/Ir/Au o hmic contacts on n GaN as a function of measurement temperature.

PAGE 161

161 Figure 8 8 Specific contact resistance of the Ti/Al//Ni/Au and Ti/Al/Ir/Au contacts annealed at 900 o C as a function of aging time at 350 C

PAGE 162

162 CHAPTER 9 CONCLUSIONS As discussed earlier, w hile further improvements in the III V nitride materials quality can be expected to enhance device operation, further device advances will also require improved processing technology. Owing to their wide bandgap nature and chemical stability, GaN and related materials present a host of device processing challenges. One of the critical area is thermal processing, high temperature ohmic and schottky contacts which are thermally stable and can at least sustain harsh condition which the device it self is cable of bas ed on its intrinsic properties. The problem tackled in this work is reliable low resistance, high temperature operationa l ohmic contact and reliable high temperature stable rectifying contacts. To this end, new material and metallization scheme were explored which would give better ohmic and schottky contacts. In this context, t he contacts being better not only mean low ohm ic contact resistance or high schottky barrier height, as it used to mean in earlier work but it shall also mean less rouging of contacts, sharp edge acuity, less intermixing of the metallization or even if the intermixing occurs minimal decrease in specif ic contact resistance. Some contacts fabricated were tested with prolonged heating over a hot plate or in some cases in hot oven. In first section, ohmic contact formation on n GaN using novel Titanium/Aluminum/ Tungsten Boride / Titanium / Gold and Titan ium / Aluminum / Zirconium Boride / Titanium / Gold metallization schemes were studied using contact resistance, scanning electron microscopy and Auger Electron Spectroscopy measurements. For the case of Tungsten Boride based contact, a minimum specific co ntact resistivity of 7x10 6 2 was achieved at an annealing temperature of 800 C For the case of, Zirconium Boride based a minimum specific contact resistivity of 3x10 6 2 was achieved at an annealing temperature of 700 C. This order of specific contact

PAGE 163

163 resistance is comparable to that achieved with conventional Ti/Al/Pt/Au on the same samples. The lowest contact resistance was obtained for 60 s anneals. The contact resistance was essentially independent of measurement temperature, indicating that field emission plays a dominant role in the current transport .The Ti began to out diffuse to the surface at temperatures of ~500 C while at 800 C the Al also began to intermix within the contact. By 1000 C the contact showed a reacted appearance and AES showed almost complete intermixing of the metallization. The contact resistance showed excellent stability for extended periods at 200 C which simulates the type of device operating temperature that might be expected for operation of GaN based power electronic devices. Keeping the contact at 200 o C for prolonged duration was an important step and m ust be looked as important step towards achieving the goal for all the future study in this area of contact study. In another section, three different metal borides (TiB 2 CrB 2 and W 2 B 5 ) were examined for use in Ti/Al/boride/Ti/Au o hmic contacts on n type GaN and the reliability compared to the more usual Ti/Al/Ni/Au metal scheme. A minimum contact resistance of 1.5x10 6 2 was achieved for the TiB 2 based scheme at an annealing temperature of 850 900 C. For W 2 B 5 the minimum contact resistance was ~1.5x10 5 2 at 800 o C while for CrB 2 it was 8x10 6 2 at 800 o C. Thus, minimum specific contact resistance obtained with TiB 2 was approximately an order of magnitude lower than with CrB 2 and W 2 B 5 In all cases, the minimum contact resistance is achieved after annealing in the range 700 900 C The contact resistance did not change significantly with changing temperature at which the I V measurements were don e. The TiB 2 and CrB 2 contacts retain smooth morphology even after annealing at 1000 C Auger Electron Spectrosopy depth profiling indicated that formation of an interfacial TiN X layer is likely

PAGE 164

164 responsible for the o hmic nature of the contact after anneal ing. All three boride based contacts were tested at even harsher condition this time, being placed over a hot plate for period of more than 22 days at temperature of 350 o C. After extended aging the boride based contacts, in general, show less change in sp ecific contact resistance than Ti/Al/Ni/Au even as the intermixing of the met allization scheme occurs. Then, Schottky contact formation on n GaN using a novel W 2 B based and Zirconium based metallization scheme was studied using current voltage, scanning e lectron microscopy and Auger Electron Spectroscopy measurements. A maximum barrier height of ~0.55 eV was achieved on as deposited samples for Tungsten boride based and after 200 o C anneal for Zirconium Boride based, with a negative temperature coefficient of 8 x10 4 eV/C over the range 25 150 C The barrier height was essentially independent of annealing temperature up to 500 C for W 2 B and 700 o C for ZrB 2 and decreased thereafter due to the onset of metallurgical reactions with the GaN. The Ti began to outdiffuse to the surface at temperatures of >500 C In conclusion, two borides produces only a low barrier height of ~0.5 eV on n GaN. This is rather low for HEMT gates, but it may have use in applications where thermal stability is more important than g ate leakage current such as HEMT gas sensors. In the next section, the annealing temperature (25 700 C ) dependence of s chottky contact characteristics on n GaN using TiB 2 CrB 2 and W 2 B 5 based metallization scheme deposited by sputtering are reported. The main conclusions of this study may be summarized as follows: W 2 B 5 produces an as deposited (by sputtering) barrier height of ~0.58 eV on GaN and a maximum value of 0.65 eV after annealing at 200 C TiB 2 produces an as deposited (by sputtering) barrier hei ght of ~0.65 eV on GaN and a maximum value of 0.68 eV after annealing at 350 C CrB 2 produces an as deposited (by sputtering) barrier height of ~0.52 eV on GaN and a maximum value of 0.62 eV after annealing at 200 C This is still lower than for Ni or P t

PAGE 165

165 HEMT gates, but it may have use in applications where thermal stability is more important than gate leakage current such as HEMT gas sensors. The Boride/Ti/Au contacts show some outdiffusion of Ti at 350 C and much more significant reaction after 700 C anneals. The as deposited contacts show only a minor decrease in barrier height for measurement temperatures up to 150 C Additional experiments need to done to establish the long term reliability of the contacts for the HEMT power amplifier applications The contacts are quite susceptible to oxidation during thermal processing and care must be used to minimize exposure to oxidizing ambient. AlGaN/GaN High Electron Mobility Transistors (HEMTs) were fabricated with Ti/Al/TiB 2 /Ti/Au source/drain o hmic conta cts and a variety of gate metal schemes (Pt/Au, Ni/Au, Pt/TiB 2 /Au or Ni/TiB 2 /Au) and subjected to long term annealing at 350 C By comparison with companion devices with conventional Ti/Al/Pt/Au o hmic contacts and Pt/Au gate contact s, the HEMTs with borid e based o hmic metal and either Pt/Au, Ni/Au or Ni/TiB 2 /Au gate metal showed superior stability of both source drain current and transconductance after 25 days aging at 350 C Ir/Au s chottky contacts and Ti/Al/Ir/Au o hmic contacts on n type GaN were invest igated as a function of annealing temperature and compared to their more common Ni based counterparts. The Ir/Au o hmic contacts on n type GaN with n~ 10 17 cm 3 exhibited barrier heights of 0.55 eV after annealing at 700 C and displayed less intermixing of the contact metals compared to Ni/Au. A minimum specific contact resistance of 1.6 x10 6 2 was obtained for the o hmic contacts on n type GaN with n~10 18 c m 3 after annealing at 900 C. The measurement temperature dependence of contact resistance was s imilar for both Ti/Al/Ir/Au and Ti/Al/Ni/Au, suggesting the same transport mechanism was present in both t ypes of contacts. The Ir based o hmic contacts displayed superior thermal aging characteristics at 350 C. Auger Electron

PAGE 166

166 Spectroscopy showed that Ir i s a superior diffusion barrier at these moderate temperatures than Ni.

PAGE 167

167 LIST OF REFERENCES 1. Stephen A Campbell, The Science and Engineering of Microelectronic Fabrication second edition, Oxford University Press, Oxf o rd, NY ( 2001 ). 2. J. C. Bean, High Speed Semiconductor Devices, edited by S.M. Sze, John Wiley & Sons, New York, NY ( 1990 ) 3. S. M. Sze, Physics of Semiconductor Devices, John Wiley & Sons, New York, NY ( 1981 ) 4. D. C. Look, Mater. Sci., Eng. B 80 383 (2001). 5. National Compound Semiconductor Roadm ap/Basic Material Properties: http://www.onr.navy.mil/sci_tech/information/312_electronics/ncsr/properties.asp retrieved June 2005, Gainesville, FL. 6. D. W. P almer, The Semiconductor Information Website: http://www.semiconductorsdirect.com retrieved June 2005, Gainesville, FL. 7. R. Khanna, K. K. Allums, C. R.Abernathy, and S. J. Pearton, J. Kim, F. Ren, R. Dw ivedi, T. N. Fogarty, and R. Wilkins, Appl. Phys. Lett. 85, 3131 (2004) 8. S. Nakamura and G. Fasol, The Blue Laser Diode, Springer, Heidelberg ( 1997 ). 9. http://www.w ebelements.com/webelements/compounds/text/Ga/Ga1N1 25617974.html retrieved June 2007, St Petersburg, Fl. 10. http://www.emf.co.uk/ retrieved June 2007, St Petersburg, Fl. 11. CRC Handbook of Chemistry and Physics, edited by David R. Lide, Chemical Rubber Publishing Company, Boca Raton, FL ( 2000 ) 12. B. E. Fortz, S. S. Shur, L. F. Eastman, U. V. Bhaphkar, Proc. MRS 512 549 (1998). 13. J. Sun, H. Fatima, A. Koudymov, A. Chitnis, X. Hu, H. M. Wang, J. Zhang, G. S imin, J. Yang, and M. A. Khan, IEEE Electron Dev. Lett ., 24 375 (2003). 14. A. P. Zhang, L. B. Rowland, E. B. Kaminsky, J. W. Kretchmer, R. A. Beaupre, J. L. Garrett J. B. Tucker, B. J. Edward, J. Foppes and A. F. Allen, Solid State Electron 47 821 (2003) 15. A. P. Zhang, L. B. Rowland, E. B. Kaminsky, V. Tilak, J. C. Grande, J. Teetsov, A. Vertiatchikh and L. F. Eastman, J. Electron.Mater. 32 388 (2003). 16. V. Kumar, L. Zhou, D. Selvanathan, and I. Adesida, J. Appl. Phys. 92 1712 (2002)

PAGE 168

168 17. Abhishek Motayed, Ravi Bathe, Mark C. Wood, Ousmane S. Diouf, R. D. Vispute, and S. Noor Mohammad, J. Appl. Phys. 93 1087 (2003) 18. Jinwook Burm, Kenneth Chu, William A. Davis, William J. Schaff, Lester F. Eastman, and Tyler J. Eustis, Appl. Phys. Lett. 70 464 (1997) 19. M. E. Lin, Z. Ma, F. Y. Huang, Z. F. Fan, L. H. Allen, and H. Morko, Appl. Phys. Lett. 64 1003 (1994). 20. B. P. Luther, S. E. Mohney, T. N. Jackson, M. Asif Khan, Q. Chen, and J. W. Yang, Appl. Phys. Lett. 70 57 (1997). 21. Zhifang Fan, S. Noor Mohammad, Wook K im, zgr Aktas, Andrei E. Botchkarev, and Hadis Morko, Appl. Phys. Lett. 68 1672 (1996) 22. S. Ruvimov, Z. Liliental Weber, J. Washburn, D. Qiao, S. S. Lau, and Paul K. Chu, Appl. Phys. Lett. 73 2582 (1998). 23. A. N. Bright, P. J. Thomas, M. Weyland, D. M. Tricker, C. J. Humphreys, and R. Davies, J. Appl. Phys. 89 3143 (2001). 24. N. A. Papanicolaou, M. V. Rao, J. Mittereder, and W. T. Anderson, J. Vac. Sci. Technol. B 19 261 (2001). 25. Q. Z. Liu and S. S. Lau, Solid State Electronics 42 677 691 ( 1998). 26. R. Li, S. J. Cai, L. Wong, Y. Chen, K. L. Wang, R. P. Smith, S. C. Martin, K. S. Boutros, J. M. Redwing, IEEE Electron Device Letters, 20 323 (1999). 27. S. T. Sheppard, K. Doverspike, W. L. Pribble, S. T. Allen, J. W. Palmour, L. T. Kehias, T. J. Jenkins, IEE E Electron Device Letters, 20 161 163 (1999). 28. D. Qiao, Z. F. Guan, J. Carlton, S. S. Lau, and G. J. Sullivan, Appl. Phys. Lett. 74 2652 (1999) 29. V. Tilak, R. Dimitov, M. Murphy, B. Green, J. Smart, W. J. Schaft, J. R. Shealy and L. F. Eastman, Mat. Res Soc. Symp. Proc., 622 T7.4.1 ( 2000). 30. S. H. Lim, J. Washburn, Z. Liliental Weber, and D. Qiao, Appl. Phys. Lett. 78 3797 (2001). 31. M. Asif Khan, M. S. Shur and Q. Chen, Appl. Phys. Lett., 68 3022 (1996). 32. S. Murai, H. Masuda, Y. Koide, and Masanori Mur akami, Appl. Phys. Lett. 80 2934 (2002) 33. Changzhi Lu, Hongnai Chen, Xiaoliang Lv, Xuesong Xie, and S. Noor Mohammad, J. Appl. Phys. 91 9218 (2002)

PAGE 169

169 34. J. S. Foresi and T. D. Moustakus, Appl. Phys. Lett., 62 2859 (1993). 35. M. W. Fay, G. Moldovan, P. D. Brow n, I. Harrison, J. C. Birbeck, B. T. Hughes, M. J. Uren, and T. Martin, J. Appl. Phys. 92 94 (2002). 36. K. O. Schweitz, P. K. Wang, S. E. Mohney, and D. Gotthold, Appl. Phys. Lett. 80 1954 (2002). 37. M. W. Cole, D. W. Eckart, W. Y. Han, R. L. Pfeffer, T. Mon ahan, F. Ren, C. Yuan, R. A. Stall, S. J. Pearton, Y. Li and Y. Lu, J. Appl. Phys. 80 278 (1996). 38. A. Zeitouny, M. Eizenberg, S. J. Pearton, and F. Ren, J. Appl. Phys. 88 2048 (2000). 39. D. Selvanathan, F. M. Mohammed, A. Tesfayesus and I. Adesida, J. Vac. Sci.Technol. B22 2409 (2004). 40. B.Luo, F.Ren, R. C. Fitch, J. K. Gillespie, T. Jenkins, J. Sewell, D. Via, A. Crespo, A. G. Baca R. D. Briggs, D. Gotthold, R. Birkhahn, B. Peres and S. J. Pearton, Appl. Phys. Lett. 82, 3910 (2003). 41. H. W. Jang and J. L. Lee, J.Appl.Phys. 93 5416 (2003). 42. R Fitch, J. K. Gillespie, N. Moser, T. Jenkins, J. Sewell, D. Via, A. Crespo, A. M. Dabiran, P. P. Chow, A. Osinsky, J. R. LaRoche, F. Ren and S. J. Pearton, Appl. Phys. Lett. 84, 1495 (2004). 43. R. C. Fitch, J. K. Gillespie, N. Moser, T. Jenkins, J. Sewell, D. Via, A. Crespo, A. M. Dabiran, P. P. Chow,A. Osinsky, J. R. LaRoche, F. Ren and S. J. Pearton J. Vac. Sci. Technol. B22 619 (2004). 44. D. Selvanathan, L. Zhou, V. Kumar, I. Adesida and N. Finnegan, J.Electron .Mater 32 335 (2003). 45. X. A. Cao, S. J. Pearton, G. Dang, A. P. Zhang, F. Ren, and J. M. Van Hove, Appl. Phys. Lett. 75 4130 (1999). 46. L. Voss, R. Khanna, S.J. Pearton, F. Ren and I .I. Kravchenko, Appl. Phys. Lett. 88 012104 (2006). 47. W. Zagozdzon Wosik C. Darne, D. Radhakrishnan, I. Rusakova, P. Van der Heide, Z. H. Zhang, J. Bennett, L. Trombetta, P. Majhi and D. Matron, Rev. Adv. Mater. Sci. 8 185 (2004). 48. S. D. Wolter, B. P Luther, S. E. Mohney, R. Karlicek and R. S. Kern, Electrochem. Solid State Lett. 2 151 (1999).

PAGE 170

170 49. F. A. Padovani and R. Stratton, Solid State Electron. 9 695 (1966). 50. I. Ahmad, V. Kasisomayajula, M. Holtz, J. M. Berg, S. R. Kurtz, C. P. Tigges, A. A. Allerman, and A. G. Baca, Appl. Phys. Lett. 84 371 373 (2004) 51. A. P. Zhang, L. B. Rowland, E. B. Kaminsky, J. B. Tucker, J. W. Kretchmer, A. F. Allen, J. Cook, and B. J. Edward, Electron. Lett. 39 245 (2003). 52. W. Saito, Y. Takada, M. Kuraguchi, K. Tsuda, I. Omura, T. Ogura and H. Ohashi, IEEE Trans Electron Dev. 50 25 28 (2003 ). 53. Wu Lu, V. Kumar, E. L. Piner and I. Adesida, IEEE Trans. Electron Dev. 50 ,1069 (2003). 54. P. Valizadeh and D. Pavlidis, IEEE Trans. Electron. Devices, 52 933 (2005). 55. M. Hikita, M. Yanagihara, K. Nakazawa, H. Ueno, Y. Hirose, T. Ueda, Y. Uemoto, T. Tanaka, D. Ueda and T. Egawa, IEEE Trans. Electron Devices, 52 1963 (2005) 56. S. Nakazawa, T. Ueda, K. Inoue, T. Tanaka, H. Ishikawa and T. Egawa, IEEE Trans. Electron Devices, 52 2124 (2005 ). 57. T. Palacios, S. Rajan, A. Chakraborty, S. Heikman, S. Keller, S. P. Den Baars and U. K. Mishra, IEEE Trans. Electron Dev. 52 ,2117 (2005 ) 58. U. K. Mishra, P. Parikh, and Y. F. Wu, Proc. IEEE 90 1022 (2002). 59. L. F. Eastman, V. Tilak, J. Smart, B. M. Green, E. M. Chumbes, R. Dimitrov, H. Kim, O. S. Ambacher, N. Weimann, T. Prunty M. Murphy, W. J. Schaff, J. R. Sh ealy, IEEE Trans. Electron Dev. 48 479 (2001). 60. S. Keller, Y. F. Wu, G. Parish, N. Ziang, J. J. Xu, B. P. Keller, S. P. DenBaars, U. K. Mis hra, IEEE Trans. Electron. Dev. 48 552 (2001). 61. V. Adivarahan, M.Gaevski, W. H. Sun, H. Fatima, A. Koudymov, S. Saygi, G. Simin, J. Yang, M. A. Khan, A. Tarakji, M. S. Shur and R. Gaska, IEEE Electron Device Letters 24 ,514 ( 2003). 62. A. Tarakji, H. Fatima, X. Hu, J. P. Zhang, G. Simin, M. A. Khan, M. S. Shur and R. Gaska, IEEE Electron Dev. Lett. 24 369 (2003). 63. S. Iwakami, M. Yanagihara, O. Machida, E. Chino, N. Kaneko, H. Goto and K. Ohtsuka, Jpn. J. Appl. Phys. 43 L831 (2004). 64. A. G. Baca, F. Ren, J. C. Zolper, R. D. Briggs, and S. J. Pearton, Thi n Solid Films, 308 309 599 606 (1997 ).

PAGE 171

171 65. M. W. Cole, F. Ren, and S. J. Pearton, J. Electrochem. Soc. 144 L275 (1997). 66. X. A. Cao, S. J. Pearton, F. Ren, and J. R. Lothian, Appl. Phys. Lett. 73 942 (1998). 67. J. Kim, F. Ren, A. G. Baca, and S. J. Pearton, Appl. Phys. Lett. 82 3263 (2003). 68. R. Mehandru, S. Kang, S. Kim, F. Ren, I. Kravchenko, W. Lewis and S. J. Pearton, Mater. Sci. Semicond. Proc. 7 95 (2004). 69. J. Suda and H. Matsunami, J. Cryst. Growth 237 239 1114 (2002). 70. R. Liu, A. Bell, F. A. Ponce, S. Kamiyama, H. Amano and I. Akasaki, A ppl. Phys. Lett. 81 3182 (2002). 71. J. I. Iwata, K. Shiraishi and A. Oshiyama, Appl. Phys. Lett. 83 2560 (2003). 72. P. L. Liu, A. V. G. Chizmeshya, J. Kouvetakis and I. S. T. Tsong, Phys. Rev. B72 245335 (2005). 73. J. Tolle, R. Roucka, I. S. T. Tong, C. Ritte r and J. Kouvetakis, Appl. Phys. Lett. 82 2398 (2003) 74. D. K. Schroder, Semiconductor materia l and device characterization, Wiley and Sons, NY ( 1990 ) 75. S. J. Pearton, J. C. Zolper, R. J. Shul and F. Ren, J. Appl. Phys., Applied Physics Reviews 86 1 (1999 ). 76. L. Zhou, A. T. Ping, K. Boutros, J. Redwing and I. Adesida, Electron Lett. 35 745 (1999). 77. J. Hilsenbeck,W. Reiger, E. Nebauer,W. John, G. Trankle, J. Wurfl, A. Ramakrishan and H. Obloh, Phys. Stat. Solidi. A176 183 (1999). 78. L. S.Yu, D. J. Qiao, Q. J Xing, S. S. Lau, K. S. Boutros and J. M. Redwing, Appl. Phys. Lett. 73 3917 (1998). 79. J. Hilsenbeck, E. Nebauer, J. Wurfl, G. Tarnkel and H. Obloh, Electron. Lett. 36 980 (2000). 80. J. Wurfl, J. Hilsenbeck, E. Nebauer, G. Trankel, H. Obloh and W. Osterle, Microelectron. Rel. 40 1689 (2000). 81. C. M. Jeon, H. W. Jang and J. L. Lee, Appl. Phys. Lett. 82 391 (2003). 82. S. Arulkumaran, T. Egawa, H. Ishikawa, M. Umeno and T. Jimbo, IEEE Trans. Electron. Dev. 48 573 (2001).

PAGE 172

172 83. E. D. Readinger and S. E. Mohney, J. E lectron. Mater. 34 375 (2005). 84. V. Kumar, D. Selvanathan, A. Kuliev, S. Kim, J. Flynn and I. Adesida, Electron. Lett. 39 ,747 (2003). 85. R. Ranjit, W. Zagozdzon Wosik, I. Rusakova P. van der Heide, Z. H. Zhang, J. Bennett, Review s on Advanced Materials Scie nce 8 176 (2004). 86. E. D. Readinger, B. P. Luther, S. E. Mohney and E. L. Piner, J. Appl. Phys. 89 ,7983 (2001). 87. C. Chen, J. Zhang, V. Adivarahan, A. Koudymov, H. Fatima, G. Simin, J. Yang and M. Asif Khan, Appl. Phys. Lett. 82 4593 (2003) 88. H. Xu, S.Gao, S.Heikman, S. I. Long, U. K. Mishra and R. A. York, IEEE MWC Lett. V.16 22 24 (2006) 89. W. Saito, T. Domon, I. Omura, M. Kuraguchi, Y. Takada, K. Tsuda and M. Yam aguchi, IEEE Electron. Dev. Lett 27 326 (2006). 90. G. Simin, V. Adivarahan, J Yang, A. Koudym ov, S. Rai and M. Asif Khan, Electron. Lett. 41 774 (2005). 91. I. Ahmad, V. Kasisomayajula, M. Holtz, J. M. Berg, S. R. Kurtz, C. P. Tigges, A. A. Allerman, and A. G. Baca Appl. Phys. Lett. 86 173503 (2005). 92. T. Makimoto, Y. Yamauchi, and K. Kumakura App l. Phys. Lett. 84 1964 (2004) 93. A. Chini, D. Buttari, R. Coffie, S. Heikman, S. Keller, and U. K. Mishra, Electron. Lett. 40 73 (2004) 94. Z. Yang, A. Koudymov, V. Adivarahan, J. Yang, G. Simin, M. A. Khan, IEEE Microwave and Wireless Components Lett. V. 15 850 852 (2005) 95. A. Koudymov, S. Rai, V. Adivarahan, M. Gaevski, J. Yang, G. Simin, M. A. Khan, IEEE Microwave and Wi reless Components Lett. 14 12, 560 562 (2004). 96. D. Selvanathan, L. Zhou, V. Kumar, J. P. Long, M. A. L. Johnson, J. F. Schetzina and I. Adesida, Electron. Lett. 38 755 (2002). 97. D. Selvanathan, L. Zhou, V. Kumar, and I. Adesida, Phys. Status Solidi A 194 583 (2002). 98. J. Kim, J. A. Freitas, J. Mittereder, R. Fitch, B. S. Kang, S. J. Pearton and F. Ren, Solid State Electron 50 408 (2006). 99. V. I .Matkovich, Boron and Refractory Borides ( Spriger Verlag, Berlin Heidelberg NY ( 1977)

PAGE 173

173 100. N. Mahadik, M. V. Rao and A. V. Davydov, J. Electron. Mater. 35 2035 (2006). 101. L. Wang, F. M. Mohammed, and I. Adesida J. Appl. Phys. 101 013702 (2007). 102. F. M. M ohammed, L. Wang, I. Adesida, and E. Piner J. Appl. Phys. 100 023708 (2006). 103. F. M. Mohammed, L. Wang, and I. Adesida Appl. Phys. Lett. 88 212107 (2006). 104. A. Basu, F. M. Mohammed, S. Guo, B. Peres, and I. Adesida J. Vac. Sci. Technol. B 24 L16 (2006) 105. F. M. Mohammed, L. Wang, and I. Adesida Appl. Phys. Lett. 87 262111 (2005). 106. L. Wang, F. M. Mohammed, and I. Adesida J. Appl. Phys. 98 106105 (2005). 107. F. M. Mohammed, L. Wang, D. Selvanathan, H. Hu, and I. Adesida J. Vac. Sci. Technol. B 23 2330 ( 2005). 108. G. S. Marlow and M. B. Das, Solid State Electron. 25 91 (1982). 109. J. Kasahara, M. Arai and N. Watanabe, J. Appl. Phys. 50 541 (1979). 110. C. B Vartuli, S. J. Pearton, C. R. Abernathy, J. D. MacKenzie, E. S. Lambers and J. C. Zolper, J. Vac. Sci. Tech nol. B14 3523 (1996). 111. M. K. Puchert, P. Y. Timbrell and R. N. Lamb, J. Vac. Sci. Technol. A14 2200 (1996). 112. M. Sugawara, Plasma Etching: Fundamentals and Applications, Oxford University Press, New York, NY ( 1998 ) 113. R. J. Shul, Plasma Etchin g and RIE: Fu ndamentals and Applications, AVS Short Course Program, Orlando, FL ( 2004 ) 114. H. Ryssel and I. Ruge, Ion Implantation, John Wiley & Sons, New York, NY ( 1986 ) 115. P. Van Zant, Microchip Fabrication, McGraw Hill, New York, NY ( 2000 ) 116. S. Mahajan and K. S. Sree H arsha, Principles of Growth and Processing of Semiconductor, McGraw Hill, Boston, MA ( 1999 ). 117. C. R. Brundle, C. A. Evans, Jr. and S. Wilson, Encyclopedia of Materials Characterization: Surfaces, Interfaces, Thin Films, Butterworth Heinemann, Stoneham, MA ( 1992 ) 118. J. G. E Gardeniers, Z. M. Rittersma and G. J. Burger, J. Appl. Phys. 83 7844 (1998).

PAGE 174

174 119. Kyu Pil Lee Design and fabrication of GaN based hetrojunction bipolar transistor, Ph.D. dissertation University of Florida, Gainesville ( 2003 ) 120. Kelly Pui Sze Ip Process development for ZnO based devices, Ph.D. dissertation University of Florida, Gainesville ( 2005 )

PAGE 175

175 BIOGRAPHICAL SKETCH Rohit Khanna was born on 25 th Feburary, 1981 in Lucknow, Uttar Pradesh, India. He grew up and spent his high school years in Varanasi, Uttar Pradesh, India. On graduating from high school in 1999, he secured a position in Indian Institute of Technology Joint Entrance Examination (IIT JEE), earning an admission to the prestigious Institute of Technology, Banaras Hindu Univers ity (IT BHU), Varanasi, India. He obtained his Bac helor of Technology (B.Tech.) from the Department of Ceramic Engineering at IT BHU in 2003. Then he applied for graduate studies and was accepted in the Department of Material s Science and Engineering (MSE) at UCLA and UFL (USA). In fall 2003 he joined the doctoral program at the MSE at the University of Florida. He joined Prof. Dr. Dr. Pearton, he was offered a job with Oerlikon USA Inc (earlier Unaxis and Plasma Therm) which he joined as an Associate Applications Lab Engineer in January 2007.