<%BANNER%>

Detection of Microcracks in Concrete Cured at Elevated Temperature

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 E20110115_AAAAED INGEST_TIME 2011-01-15T23:35:17Z PACKAGE UFE0008824_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 77201 DFID F20110115_AACWDG ORIGIN DEPOSITOR PATH shah_v_Page_50.QC.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
b0ecc44ad9c7a6ca0b3a0a0fb6002df1
SHA-1
fecd940706afae98d71e2e3146a263fbde611418
1053954 F20110115_AACVWJ shah_v_Page_67.tif
3866dfa91e1066889b4f801d5a6648a8
708b113ba9ee76999c3fc4688e5430bdc29d564f
15105 F20110115_AACVRL shah_v_Page_02.jpg
cf9ff6cf90bc2fdb5d2e7d8c40984792
2a74f57dab4ca548f11f48758b982e5b0051c8fc
87228 F20110115_AACVMO shah_v_Page_68.jp2
b1d18f73402c1d4ac3486541f6aacdf0
09e19404e99b45d357d591810eb28a9dd1651259
45577 F20110115_AACWDH shah_v_Page_51thm.jpg
7f333465ffe1d5559d42e2075069a92d
cf48702377394e469796dfddff06194994d07d99
40086 F20110115_AACVRM shah_v_Page_03.jpg
0c63b4f8c7f0f989300320bfbd4666e6
10e8da67c130912d2beef0c32a51d77d20c82cd4
56899 F20110115_AACVMP shah_v_Page_54.pro
569393a786a5b34eed370a0a3540e5de
c4668bef9968ed57fdceeead7049ccc737afe496
F20110115_AACVWK shah_v_Page_68.tif
c1857a063a8a86b263dadc070b58deeb
e93e08a1234888b7f7b71774a444d0a5c07f8c80
88713 F20110115_AACWDI shah_v_Page_51.QC.jpg
14cea2c645aa8068cf21b4a9de6bdcff
c4effa271a24b7a15c5a73c1881453159d7b9b93
115998 F20110115_AACVRN shah_v_Page_04.jpg
f87d1f1a0a0048ea42ae47a51ac60ef3
6065bb59e2e0f0e722aa47d4f442fbefff4dc777
F20110115_AACVMQ shah_v_Page_01.tif
05ee9ee6d9a42e1b1f02299d749f5281
5c256007e029b1d3754dd59aecce82921ed66475
F20110115_AACVWL shah_v_Page_70.tif
2a3903a99a69321ecf1bb5768bdf9d6e
dd4286eb6ec3738e5b3ac464e3303dcea40e2442
98379 F20110115_AACWDJ shah_v_Page_53.QC.jpg
0bce386eb6f73d9ea07aa0c37df724cb
67495e5eef5e20c9fefbc930f2922732a8642bcb
237845 F20110115_AACVRO shah_v_Page_05.jpg
ca3ac629163a2dd176033c09a001157a
9943d032049288dcb417cca615c864a2a5d63941
1051965 F20110115_AACVMR shah_v_Page_07.jp2
76a70d221537644599a31c172a9ba7d5
9ebe355129f253f28b9d79a5d1d6f276f3142672
F20110115_AACVWM shah_v_Page_71.tif
fa8ea9e5ea5e30391aa670d566ca131b
0894879aa617319f39c8ad054db3df95e8fec1e1
86885 F20110115_AACWDK shah_v_Page_54.QC.jpg
39af24b3f5da748ea0e83ae956fe82ce
8521fc66ee5adf1e254e1e4da0f1e3225f3131a8
186779 F20110115_AACVRP shah_v_Page_07.jpg
d8772fab2fae19c41d1c72823692109c
1cc57aa397a4351a10419d230663c1fda06d4530
197963 F20110115_AACVMS shah_v_Page_47.jpg
a8300be5b502b93c1f8bd86be0272537
54056f033f77e73a90b5b70f35d1a5776f82300e
45935 F20110115_AACWDL shah_v_Page_55thm.jpg
16335c0d7f5abd310528599b5cced38d
112e12a78b131475958e5f31fc6933b4b5b598a8
153517 F20110115_AACVRQ shah_v_Page_10.jpg
8b10eaebaeadde5f6818c9123ca106af
35da1f1eb9838d21cae83a40597cb5ae690662aa
F20110115_AACVMT shah_v_Page_73.tif
8a59f7e1ca2bf4b328e8fbd0e3aa780a
b4b2abf8e60297fdd1845d39b687597e026a4c43
1249 F20110115_AACVWN shah_v_Page_02.pro
1cab610b36a9445c6d2266f07e1c7302
25ec417fd77a16d20e8af04340d556f64b845d55
87472 F20110115_AACWDM shah_v_Page_55.QC.jpg
989024438e3cec6b0cacdca69e6ec4ea
cd78f50905c5ef33fa95848a58358703e3961227
154474 F20110115_AACVRR shah_v_Page_11.jpg
83e98428a26c8f07a838e317f0dec77a
d926654c0ae4c3e63fbd212d84ef1614f11cc5c1
5918 F20110115_AACVMU shah_v_Page_02.QC.jpg
c590a2185d1c23d73f4c6bf72bf6237f
d59c5eb7daca2ce46829101ce4a23e91b664979d
7894 F20110115_AACVWO shah_v_Page_03.pro
2cc7006e213a274867ddad85c71fe8ac
fa49690ff5509e78d2ad2d5377aa80853d4f2b70
106833 F20110115_AACWDN shah_v_Page_57.QC.jpg
ca23432ac923e05505f475791b69f9c1
1c423d6e19b5a3b5a72c0ae0008374dd638da785
166337 F20110115_AACVRS shah_v_Page_12.jpg
15995019544095d52b084387aeddf2e7
5f47e1e8801d03276d8886cc8439588ca05713bb
F20110115_AACVMV shah_v_Page_17.tif
8a1bb17ae94296dd72a785f2f07efebc
b934e80a6ebad8aab4322ed8b7765f2aa8512cc0
2927 F20110115_AACVWP shah_v_Page_09.pro
b7766451bc80f00c837cab1d17328770
294d1ac5c9f24a4143c5895183effa49091b79ee
46845 F20110115_AACWDO shah_v_Page_60thm.jpg
c2695347d15d144ae3180394e0ad691d
74f5a94c20e8311ad654ec1f470300718335974a
215039 F20110115_AACVRT shah_v_Page_13.jpg
8a499380885bcd0f553d7c9072f6f060
753573fb1bed25e10be649a346c338fabb50c157
380 F20110115_AACVMW shah_v_Page_03.txt
908572ebd21a0afcf477b66adefbc457
9b555ce022143a6ae97cd94364b915c4d7660b2e
35807 F20110115_AACVWQ shah_v_Page_11.pro
c7d27679ad5fe89190d9fcaf7f2d1481
71464d67f291f628da495e41d379df4c8173686d
49082 F20110115_AACWDP shah_v_Page_61thm.jpg
ce63d9c7b3a20b9232f69d7d8b6c002a
b8a9f42a2e28c7ffebfce99ba3d6cdfec055486d
216232 F20110115_AACVRU shah_v_Page_14.jpg
631ed1b33136cdb799c82441954291a5
7a78cb5572f6f22d85c218dad78896dd2c676af8
19022 F20110115_AACVMX shah_v_Page_46thm.jpg
84bce4c0e7812a860f7aa05c78a84119
af4a6edaa0e2ef834ad9836b43df3c7d4706d2c8
38922 F20110115_AACVWR shah_v_Page_12.pro
2d4e30f078e3d320d999e1dbf510085a
722e428c541e48e664ae769b38260fef9f865232
26574 F20110115_AACWDQ shah_v_Page_62thm.jpg
4ac74ed45ce80c49f8eac1fa55e5fbe9
419c813d78a82b6a69b6508a7237e0d892f1a63e
58556 F20110115_AACVRV shah_v_Page_17.jpg
6f72afecfdf8a154fb04e8d822f64426
9fca6b1385ba744496d9939ec826ace81b0b5a45
253871 F20110115_AACVMY shah_v_Page_08.jpg
87be4e7b99f4797a9c1a3394ec66d9ce
917a335fc150493d3cd6c8ab514ed73cee71d538
52752 F20110115_AACVWS shah_v_Page_13.pro
7a938b46578459ba00173a1850f2c75a
fa200853cb54f8027170e73f01bb0f8c9163cc9f
88510 F20110115_AACWDR shah_v_Page_62.QC.jpg
d9359e461910454e8536b44768d46359
de948a5b416b079bb07cd9a43451c91c6d3ed982
184504 F20110115_AACVRW shah_v_Page_18.jpg
bc72a94111f8a8b8cf09338d292d07f9
37f208d40f8118809a08246a6488258385eb1653
89755 F20110115_AACVMZ shah_v_Page_69.jpg
73001f807d744d528b9bc57a152ce53b
fb44014561f9d1b42acbe3d8fd3f1a4b8e3c50a6
52677 F20110115_AACVWT shah_v_Page_14.pro
fbd5afb20b2bf4edd74649df9ada0654
8bea847b020c173cca0ff1cdf79a085192b302fe
45248 F20110115_AACWDS shah_v_Page_63thm.jpg
b45cc736c3bdaf4033bdaf052c44eeb4
184a09f568abae4df70d77fb79b3d09988d0a811
203527 F20110115_AACVRX shah_v_Page_19.jpg
8d30ad924745c7d291024616bc900b1b
3640610115b4e2da74d69dd3473b8ee060aef300
45989 F20110115_AACVWU shah_v_Page_15.pro
4b5c6e29f56ae60ecbebb9a41345950e
b75dba0e5caff5fab7b6b7fc2632935392c944de
16540 F20110115_AACWDT shah_v_Page_64thm.jpg
8e193df8d4cc97b51431ee07dd327107
9f80956b3fe98ed7115184a522eec629b39c8acb
209476 F20110115_AACVRY shah_v_Page_20.jpg
3fea0b50e237f5c4c2d6b58699b7ac36
4dcd37070d2c02830d81db6d159ff4cbe96eaf27
47032 F20110115_AACVWV shah_v_Page_16.pro
a20c442062287a1957cd9624e9af90f7
23734fed749c417fb5a73c8683ce2df00af4144b
44375 F20110115_AACVPA shah_v_Page_44thm.jpg
ab41e96503567391b183c5c9cab8b81b
d99ffdda0ca5861a642b1244757a2c3d4022748c
64765 F20110115_AACWDU shah_v_Page_65.QC.jpg
f96e6b6dd7b6181288ca2c8b06f9212a
13b1872eb1a6c0b5b3e393835aefda08812d2444
214408 F20110115_AACVRZ shah_v_Page_22.jpg
312089c1a532950094a4b896bdfb5c9a
78d8be45705a32e542ba3f0e17b5a8ded62f0373
11942 F20110115_AACVWW shah_v_Page_17.pro
e0c66d75c866b8304c30ad55bac2500b
48034cda9a97a01515841664e48231224c181c63
109433 F20110115_AACVPB shah_v_Page_32.jp2
b5f32243283055d56573ae111681c8c4
a87051f9efa5b2eebeb75d5a57910ccb4024ecd3
63181 F20110115_AACWDV shah_v_Page_66.QC.jpg
6928ead1254975d650e932b21bb54d43
1debf34be512bbc66f4247cdc156cfd177d5654b
43466 F20110115_AACVWX shah_v_Page_18.pro
e4a350a0c3109b6aa37d34e602ab1a34
52ed322a08b1c18623807efebccb567127434d0f
199312 F20110115_AACVPC shah_v_Page_39.jpg
37bab88ef460506c8bb1af50b809fc1d
304a3a9bbfb19a12345a121c0dbfabbdd3ccf380
79874 F20110115_AACWDW shah_v_Page_67.QC.jpg
36021a5b21f4ebf3ce2de67d620f5cbf
e3afda093f790e7d2697f5a229885d7ae54b346c
24467 F20110115_AACVKF shah_v_Page_01.jp2
8ee807ac065844f56e461d7227050950
da6e743aea4fc87cd23b778eaf1169c2e761b759
103395 F20110115_AACVUA shah_v_Page_29.jp2
acc237960ce40e38ce8cc722c0451fa7
422a1d2f9e27392e03d0c321942b4fb3ce163e29
48961 F20110115_AACVWY shah_v_Page_21.pro
cf9c8551dcb34b632cf62f33aafde66a
b9c155413433e9ffe8ef8ba98ab6c6c2c96bd8f5
52595 F20110115_AACVPD shah_v_Page_27.pro
1567731370d60a470ec9edf555b0fc5b
2bf041a8d26f8cd2f289a211d10470a587086547
64917 F20110115_AACWDX shah_v_Page_68.QC.jpg
ea66e761ad6d5b5b165adb2d55b140b1
a2bb0497e4bfb6ee8f998e0ed8ea990233af18ba
89116 F20110115_AACVKG shah_v_Page_58.QC.jpg
5a06413b6555757823400937d745297e
4a158f0d5f5bdcc72b3f268377650bc9f8005566
1051970 F20110115_AACVUB shah_v_Page_30.jp2
ecef59e593eaa2ed073c828aebcfcf5c
9515385268753225acdc1d9e4f2ef530cf6ef82f
50274 F20110115_AACVWZ shah_v_Page_22.pro
5bb636c60818a8ffcff35ab177268bac
df0796d4bc1f996e6f57e662d107fa3e8ad6295e
25271604 F20110115_AACVPE shah_v_Page_55.tif
3f9175e3a651fba7efe13e19e6342c4b
f7cd53dbdc885248d109c5ad7f061b0d4286b4c1
9359 F20110115_AACWDY shah_v_Page_69thm.jpg
3f20720d5ac2ef2464cded3dded2dc98
8c84a83288fa1e8dd2d854e9b553551c4c7e0536
F20110115_AACVKH shah_v_Page_15.tif
046bd40fe1b501c642e8f8b649af2633
7ad270eeec61cd81e4a92d801ef3c386790109d3
107263 F20110115_AACVUC shah_v_Page_31.jp2
8d625aaebac9fb776783f8cb36786da0
69cb8696165181079129a05bc82554794293c0da
19451 F20110115_AACVPF shah_v_Page_73.pro
0d3c98421ed05bf74400cee659424d38
cfdc074b8befee3d06e07ae60136fb985487e3c9
19129 F20110115_AACWBA shah_v_Page_10thm.jpg
f92fa83df6ebb0a8f19b4a19efad02eb
87916094c3d35ee7c226e5fa5c77b66e36d6d005
1964 F20110115_AACVZA shah_v_Page_32.txt
5ebce544b1c828449f3f2a58e8c9edfb
5bbc8e653027c2c74ee1a48383fbf79b32933670
85999 F20110115_AACVKI shah_v_Page_46.jp2
0489fd993712e477ca9a9fd48cd71ed0
8f7eec87e0b85a991fa44cb6eddd42f1b565c3dc
100200 F20110115_AACVUD shah_v_Page_33.jp2
15e14137d6f35135471850da0911c335
e2e7a1fc8591d230ad9362fb02d41eab75fbd7c4
28540 F20110115_AACWDZ shah_v_Page_69.QC.jpg
22e67a395691fc5874b3aa74da52a7ee
8db355d7be16277778ffc0dc476f0372f6b7abd8
76010 F20110115_AACWBB shah_v_Page_21.QC.jpg
c97e6e656b36ee257f7187b14d3261fa
0100eb59029e0fc1815359cd3c02b2a70a38a687
2118 F20110115_AACVZB shah_v_Page_34.txt
b9f36af012f30c1eb5bbe4d6854376de
6bf3dd928d7e39ac904b79b23582c934b2ba6747
28823 F20110115_AACVKJ shah_v_Page_44.pro
80d5df7e63c2fe2318d5a7d4dcb6a613
9b25a3346c32d34c19fe6aeca1a6a5b409dc8266
92978 F20110115_AACVUE shah_v_Page_36.jp2
db20e7c3bf9fcbc7ad09ac47826cc41c
233f16c5467d4019e5b14e0411c08ac2270b6c21
1271 F20110115_AACVPG shah_v_Page_42.txt
6250870f4fd5e78f4246f834ac5b1565
9cb2b75e5ceb7e3feffa62143325d902fdcc85ed
82637 F20110115_AACWBC shah_v_Page_44.QC.jpg
ec08624361e1d0107eb40202cea665e2
bc75a8b4774bcdbd5322848c11daa87277f99c30
1960 F20110115_AACVZC shah_v_Page_35.txt
02901729ab2c0658e6da0c8368c3e192
66fb3ca8b18991ce7495dce9b1c6a3b6e7fd9031
1599 F20110115_AACVKK shah_v_Page_10.txt
e7762e9170e375f83deb59a0d6247f02
4d07df720450e12e41405138d41dfb16b8abf659
101537 F20110115_AACVUF shah_v_Page_38.jp2
6c1b395aa277fa33b286d7b834697070
f3340015478cde863bdbe797142e4c48def85761
F20110115_AACVPH shah_v_Page_21.tif
69c35b99ea41d19c8d09bcd418985475
4075cf6a370390239c26f8c58bc614f4f3285a6d
78745 F20110115_AACWBD shah_v_Page_20.QC.jpg
3e88b66f97c9ba9dcd3cae8d88cf7384
63137ed56f2fc21a0302f69ef92cb6849b5d52a8
2387 F20110115_AACVZD shah_v_Page_38.txt
cb4563185161383f8646ccaea7813f0f
cd6b38b3bd140a83c153b5eb4ea81dc267d58604
53276 F20110115_AACVKL shah_v_Page_34.pro
801fed50736880f5194005b43fe37e95
94a66437ef0a6b29ac2ab489cecb7e639ffd22ca
103447 F20110115_AACVUG shah_v_Page_39.jp2
ed8d72bbd8aeafd940b3791380bfa403
3028a20254088e9d6691ec99168b1287c3d6b710
21601 F20110115_AACVPI shah_v_Page_65thm.jpg
eb8227a900139c5efc8d6a4fb55c7b38
5c945f429269f829802340c2d71c551c7b651830
88993 F20110115_AACWBE shah_v_Page_63.QC.jpg
aac96cc22de142ed23f3bc72546bf5e1
7c168e6b0d0927d849f3cb47c0b05a3fe0e95ea6
1869 F20110115_AACVZE shah_v_Page_39.txt
c99178054e647a63f7fd1a6f1f25b021
8ee7b820b82683a6de3d6af5a8b9c1978e50daac
F20110115_AACVKM shah_v_Page_57.tif
2d62109d80f39992b5ff60a456bac6fa
1f252ba75d69d8d5b2e8fdff3ba68f9f476deeba
1037834 F20110115_AACVUH shah_v_Page_42.jp2
b83b16c16a0c8a1ef28735c12f22c603
5d2979a1c84af02e1008423fe811129f61265d27
F20110115_AACVPJ shah_v_Page_39.tif
74291e13607c8c1595f97a7c77454392
68c74fa69423128fcd389f3be346d58907ad4df0
45637 F20110115_AACWBF shah_v_Page_40thm.jpg
d3e8306840d6105f89d3194cb8e0925a
f18434ede1045f082876919f8eb2e90d3f7275e3
726 F20110115_AACVZF shah_v_Page_45.txt
69dba3c9038349e384bc2f0cc97722de
19d8dc7736f2da7e1185127d9a90eea6c61be047
24347 F20110115_AACVKN shah_v_Page_20thm.jpg
885aecf9abe934d50533d5b472c254e6
185ca26f55935ac44613acd8712d1497f915b267
41393 F20110115_AACVUI shah_v_Page_45.jp2
4ef58c58348384691eb9bec0c0bbe44f
88e993316e358bf3a8a1c7acad7bb7bc827edd03
F20110115_AACVPK shah_v_Page_51.tif
8843cf6ba285bfe65dad25d755049ae5
e99e1ebc73f8701b211650f449ba467f28d24a5f
48019 F20110115_AACWBG shah_v_Page_08thm.jpg
086460d8a652b05c271d3945078b7357
ad8ed346d2788f2f04d862735a555df3f7879a05
F20110115_AACVKO shah_v_Page_33.tif
06b2ab110049d0740d1f84b3cbd1f51a
a2bf91445a3779ec1e3dbd12fb3577ec646ca600
103664 F20110115_AACVUJ shah_v_Page_47.jp2
f2cdc73392689ecf3508ddebaa052a16
6819181ddcf78940953a53e337888805705ddaf6
47295 F20110115_AACVPL shah_v_Page_23.pro
b836455ad4d59b086763c0ba8ce7680a
b9141576463cc37a7c0fcb34a0a67fc9c00d45d6
1620 F20110115_AACVZG shah_v_Page_46.txt
981a0d74cb1144b6919b8824af5a122b
02045d37555ae5f2268503f2f7859dd1375e62cf
70784 F20110115_AACWBH shah_v_Page_47.QC.jpg
e7ccfe9a0b8dd80adc129705f77aeaf8
9dd73e293e2a5808984ddfc36ebaa47cd7de92b7
103869 F20110115_AACVKP shah_v_Page_09.jp2
3715343c0085c5a0d0413cb01927e2e9
f4448a9766e27527778c8bbafb16c8093033ce80
117147 F20110115_AACVUK shah_v_Page_50.jp2
cf6808a4ca2400ba72cae5be2d72501c
96c34a080b8854a6e4a62993e1dfd2a7fac2edc9
1810 F20110115_AACVPM shah_v_Page_33.txt
b2515b7d007f802f60526c0df29287c1
a75f8f71022c22bb9029fef9853962c0ff9fc3d2
822 F20110115_AACVZH shah_v_Page_48.txt
8145d632817d44d4ee0c1aafc6bd3bf8
14fcc7b1907213cb8a705a80351821b39f264ff0
81324 F20110115_AACWBI shah_v_Page_71.QC.jpg
88842a337affa0591f3ed912c079835b
2af462fe006d7e21d86c644d6f982b2f8a3a4ad6
124957 F20110115_AACVKQ shah_v_Page_62.jp2
3675554582e30896ad844c3438763103
de4b918aac5ca9bbc3de5e07944ad50c32a6e711
F20110115_AACVPN shah_v_Page_61.tif
a1f4cad1eba654989e97d0d7e5fba059
7a9251679688d3965b553031ff83f6ec52a90f3a
811 F20110115_AACVZI shah_v_Page_49.txt
b8602a4112da7c8206796b313b2e28de
0bbba9054c71147eb0a9c3a60b0fcb10f94efca6
24921 F20110115_AACWBJ shah_v_Page_71thm.jpg
cdf7961073e486031a9b2fc3b1ebce12
ce0bdd3036266f751e0b0cedf00d1f20649fb85c
1781 F20110115_AACVKR shah_v_Page_68.txt
4cc00e55c03a06fa951cfd6c5987a792
f6b99b6d5386acbe2986248324b32eec93614a4c
911511 F20110115_AACVUL shah_v_Page_51.jp2
2f0dab9b8e5931f20105b9883f839f2a
661615a692f127b28caa403ff4c63b21a17816ba
F20110115_AACVPO shah_v_Page_66.tif
64b3ac49bdab53b0b317709822947c7e
e70e158212887dcd33ece9650c27e5e3595fdc05
515 F20110115_AACVZJ shah_v_Page_52.txt
da9cead6bc8c0770e1e812dc428d976d
d89e2214e1d21319f873618d9c9709eb19703050
18499 F20110115_AACWBK shah_v_Page_17.QC.jpg
a9abc9e79d668b1542e4b9c2efce2ef4
381cb286dca01de4270f9810bce70115c9c15f12
112111 F20110115_AACVKS shah_v_Page_67.jp2
365f4120808c875bedfc6fe3271da6b2
94852803162bf4af0197a2c27e396defbd36fea9
1051975 F20110115_AACVUM shah_v_Page_53.jp2
e18db503b2404093c081d0efed16ae38
8c36494f82d1d733b36d70587b15c1a81f4a7cb5
F20110115_AACVPP shah_v_Page_19.tif
019fee65b5af574b8283c5876b34226e
a4fc9410c6d233b9edf7a95632e35efe25fa5e1b
751 F20110115_AACVZK shah_v_Page_53.txt
68e4ccff26ab6a41a38667c216ba006c
9bb1226283629c6ff591232b2ee60832ed89664d
21404 F20110115_AACWBL shah_v_Page_36thm.jpg
93a5f249e4c172a541def021ecf301bc
cc6ab56c04ef0bf51ce6f38a5add6a8bc70e5ea7
F20110115_AACVKT shah_v_Page_37.tif
18f999ab4dde315debe51835ddc8ec77
69c64f89a9963bac1e33669f5b8ff51a6a1a9f22
910468 F20110115_AACVUN shah_v_Page_55.jp2
23229cf9227c190a77a421d2738eebea
74ec2ea37f809ade91702155ab5ad3547b559058
35246 F20110115_AACVPQ shah_v_Page_10.pro
3e45efffa85121ef659312ec8947af03
719350501b6814ce9e93a13f06fcd5012471598e
492 F20110115_AACVZL shah_v_Page_56.txt
25c4bca0209d2e27b4d6a7e42fc9462e
c3f33e6ada3e9100fadc453bb8ba042e9ffd1732
81489 F20110115_AACWBM shah_v_Page_14.QC.jpg
07fbda991d3b5a362d3e3f14013b1a31
f167a881741f43e2f0ec2da42fb872fee268c5db
171792 F20110115_AACVKU shah_v_Page_66.jpg
f73d138283b238711952c9ace735578c
f251868f3657cfffcc5561b10703c919189cd085
1051780 F20110115_AACVUO shah_v_Page_56.jp2
dd4211b3e03d1c95f2f0703d29c30335
d3b6dfb65f27e1b4ae61fb45f7aa9b73f5f6a604
120661 F20110115_AACVPR shah_v_Page_70.jp2
8b692a7215847b0fa1b2a78170d18ab9
1183937be08896db415575ac84dbacf46ec02a0e
2649 F20110115_AACVZM shah_v_Page_58.txt
df518c6eb825e02a3031886ecdd2f12b
c8a1a4f036474437aecfc0adc61c5eb94f37d908
6540 F20110115_AACWBN shah_v_Page_17thm.jpg
03b1d8856cf8bbf2681e223ca6fa85fe
788e0f61fafd784224155d1327989a7dc4a1f84b
203711 F20110115_AACVKV shah_v_Page_28.jpg
860ae06b1a7e14ce12d7d39f6bfd1d12
09bbc2f55b4eb4b7c87703a07c02b32716337a41
1051984 F20110115_AACVUP shah_v_Page_57.jp2
f43b511175ab36932b64f1e6d153b77b
3ea0a18d40e2aa0d36ec4aab36433afe295a86b9
1851 F20110115_AACVPS shah_v_Page_55.txt
e85a92c7f2d2a669990bac4792b982f4
643026ffe76fb880634e34f96f12faa39003d1d0
1717 F20110115_AACVZN shah_v_Page_59.txt
ebbbfd9b33c28f7660d620144957f92e
d0ff325a10c9aadaf87b1a05acfd53998007f6fb
24542 F20110115_AACWBO shah_v_Page_19thm.jpg
a17145ad88b78f453f059831851b48be
10067975a1dfa7dd28d13a579fb08f8aad3396f2
1083 F20110115_AACVKW shah_v_Page_04.txt
499a09dad1eadf355653684f3a2cbbf1
52b397c46a7af47a011389f7a4b08ad1607471b9
833816 F20110115_AACVUQ shah_v_Page_59.jp2
50b333bbeecd1e807d9442aeb36bd4b8
89d82c8a343b444f73002f733c3d5716d3bd2084
20193 F20110115_AACVPT shah_v_Page_01.QC.jpg
d52c68f5bae0f42dfb725cb24b2a49ce
54042b637c4e68c198710434e35cacae55f42eb8
571 F20110115_AACVZO shah_v_Page_60.txt
4596d69f1f75765e9756428dd3d7202f
5317fe18ee47b86224ada7652dc437a20b351af2
98235 F20110115_AACWBP shah_v_Page_52.QC.jpg
aaf91c3feeea92dbfe5d507e15e266a7
fee097c0135ec1688d351c85b7f697177a11950c
42204 F20110115_AACVKX shah_v_Page_06thm.jpg
80e58952e2924a615ed39a8323515230
162768e747ea1c4987def05ff61faa36ff70ec00
1051892 F20110115_AACVUR shah_v_Page_60.jp2
9c313bcb4a8ec327db56fcec3f4ebef1
4a4d28ad29cbcca3698219dbec56a66b28f87056
37884 F20110115_AACVPU shah_v_Page_07.pro
3052545ea7f65ea174f0fa1e67830bfe
e80bf6afb9061c947abe1d55d78461175bfc947d
940 F20110115_AACVZP shah_v_Page_61.txt
a7fa18f6aa5687cfffb3253735cc1b65
90914bfe5bfefcffc96c351e7dc09426de06bd2c
81424 F20110115_AACWBQ shah_v_Page_59.QC.jpg
1edfc4f1fde2da6654a07d6c27ec9af9
51e94909475028d8840eceac08811613c4bb6457
34764 F20110115_AACVKY shah_v_Page_41.pro
63fe1817372917a3f05ca400c855f528
d85706998d479701ec234167330a2967fcc61007
893912 F20110115_AACVUS shah_v_Page_63.jp2
1fabc4dc808dfdf4570628cf64393d3d
102831379b7ae57a8d4d016b20076d78ed27cfb0
1051924 F20110115_AACVPV shah_v_Page_52.jp2
3bcf3375c401c6273a47f8afd3b040c5
07dbf75e67947f62672270c963934b9d7ed90a47
102922 F20110115_AACWBR shah_v_Page_49.QC.jpg
786dc4fd4e1db05bff4f242a181d8401
ae44212c272ed2562b46d643868817ad89890e0a
65843 F20110115_AACVUT shah_v_Page_64.jp2
1d1672654b14a94b83ab28f43c829aba
52970262a74f065d353836de4c382749ec812bf3
46865 F20110115_AACVPW shah_v_Page_09.jpg
168c12b868c3cb84b97045c0224625be
52c111bedb8facec66d3d42fee970395447f775c
1686 F20110115_AACVZQ shah_v_Page_65.txt
aa3a7e9dc8b3812711379b90279ef360
697edd38cf2a6028fb004dd1ccdd3be04ed1c3bf
78173 F20110115_AACVKZ shah_v_Page_35.QC.jpg
6c01e0bca2f2ec8b7197857569766dd3
0e6d310678db879684dbb919523cdf8f8c890a23
5597 F20110115_AACWBS shah_v_Page_72thm.jpg
4402c034324a6adc2c7ed6c8f34ea277
39f8040b6c5128a900bd116b77ecdcf5527cebc0
1701 F20110115_AACVZR shah_v_Page_66.txt
2c9b83199e54b14cdc6a49297186cb4b
3733b6f6add801a522b28cbdbba02d7e05007bb8
89643 F20110115_AACVUU shah_v_Page_66.jp2
fe0d2845a72e1d5a1fc6881392b5a3f3
4b2d1fb9985a3069cf21d07b7e2e918c1374db7d
1051817 F20110115_AACVPX shah_v_Page_48.jp2
cea3b11087ce199630b45be57c8fc321
67a77e5b2448998fe6c808b955d46e468e3d868c
110043 F20110115_AACWBT UFE0008824_00001.xml FULL
78718ed486aee50ed1feb446b5c5970b
39ab8660cb528c73a768025dbb63b63e43b1a206
840 F20110115_AACVZS shah_v_Page_69.txt
8ec53a76556c0037406ddba469bc2342
11485c3a18af71f4dd07d74c7557c411d33286e7
42220 F20110115_AACVUV shah_v_Page_69.jp2
d7c78a02f3dc7d1e3e1bd18896db3ad4
15a2d41d4a5f91d2c13bede86237ebfec5e0ead8
1051974 F20110115_AACVNA shah_v_Page_49.jp2
7f94bfa50fc39e814e1d657708a3d2e3
b916c2c0fa3a2b22d2c4947e5a8c92c92a324f5d
F20110115_AACVPY shah_v_Page_22.tif
4eb3a6631ebcbba134bf4b6938d903df
29d608b127dfee6d663fc97a17c5c4a28617bf8d
42112 F20110115_AACWBU shah_v_Page_04.QC.jpg
5eb3145c573b5077b819e8c383334eb9
c75fad5b0b7df1cbfb08df74b1da57bba105c6b8
2278 F20110115_AACVZT shah_v_Page_70.txt
7b4ec9cb08a7133241d47497ea416f84
e69f85502e4d3f5b4e35f3332e2f8616a6ce2ae3
126670 F20110115_AACVUW shah_v_Page_71.jp2
8597f7922252163d783580147a458734
d941cb5e735a011ba075fc91636010f1815ff7c3
1979 F20110115_AACVNB shah_v_Page_22.txt
2abb19795c4f0119eff269fdbd257e6d
17e65061bba9dc4055431c1fef24f43a038d1e5f
F20110115_AACVPZ shah_v_Page_23.txt
5d613ae8420b6eefb05a15142cf4f0fa
0c49b4ba759d4131ba2849c089aba32a6aa07007
43541 F20110115_AACWBV shah_v_Page_05thm.jpg
55adf22ee421634787faa10b33d15206
87515c4847be42a664c0b5e61d69009c50047643
820 F20110115_AACVZU shah_v_Page_73.txt
13eabaf3951029682b995583800a7888
37c99a158e6717686fc3f6f27693c105a947de1f
23993 F20110115_AACVUX shah_v_Page_72.jp2
6f43cc18886f2070e92d900a8c9c0779
a7b0fb7549343dee0b2ac7c4f7118b4069019bc7
436 F20110115_AACVNC shah_v_Page_72.txt
f51c71c5084b88d942ed96500291cb20
3f3135cb990844ac4ffae74d02ae95d9fdbcf9a3
81031 F20110115_AACWBW shah_v_Page_07.QC.jpg
3a657b4053edbe6726887480d5724a41
0f7c0b4ec6e164980631e431b10901b744c2035c
3220530 F20110115_AACVZV shah_v.pdf
6b948576b9f7f1012f34c0a373942c16
dc09e81d700862a2be06507a37b4e82c4feeecfb
45948 F20110115_AACVUY shah_v_Page_73.jp2
c693ba75adde5910140b4da5f2f84d0c
c6c5e66667ff41616d11e7ede1c81d760bbe7ed8
2420 F20110115_AACVND shah_v_Page_71.txt
0a15c8e392736b110a57b704185e1d1d
ff9e1cef84e417340365ce7082bf8df30fd3b28a
201751 F20110115_AACVSA shah_v_Page_23.jpg
545c492200804f09dbff3e96b28b9503
32faa4f79ca97683a66673557f49e567b56d2f2e
23536 F20110115_AACVZW shah_v_Page_70thm.jpg
5abe88e43a97ff9ad8d5549d57e9b954
f28329a81c8fab7741f13322a5ae7a9895c9fc45
F20110115_AACVUZ shah_v_Page_11.tif
62e944498cc8abef617cebb097a79894
8caa8621d89d406b547773ed7d4246d227b8da0c
214942 F20110115_AACVSB shah_v_Page_24.jpg
687ef273748349dba81343a24827be5b
721748af05960254cfa053a370c72c4c9e032dd8
102416 F20110115_AACWBX shah_v_Page_08.QC.jpg
48a5e7a70161e8dc5e1321be88ba77bc
fc081edb99b566065621e5004ccb5ac33d31296d
89709 F20110115_AACVZX shah_v_Page_05.QC.jpg
babb5ad29f40c529b735ce06722afca2
b14ced49184b123b6f8fe16870ec42363b57462e
178344 F20110115_AACVNE shah_v_Page_37.jpg
443693025ea61dff1123a1a21eca1c7b
a2eb9e56f86b0ef7fe33c64582df1865caa49ff1
225143 F20110115_AACVSC shah_v_Page_26.jpg
a9bb3110d9e73cb4f87f4ceb5c8c3e3f
f4669423fda72e17d56ddf037368e84d25fffc4b
53965 F20110115_AACWBY shah_v_Page_10.QC.jpg
67a9cd0711edca19b13bb127e2b0f774
4499f04493eeea42d107c9c38fa96322c1c97ebf
25529 F20110115_AACVZY shah_v_Page_67thm.jpg
313d0125ca7975c23fdaba26e01d05e8
11f63c484afc2f615a7a5fc71ebc1b3ba7eb4c94
2052 F20110115_AACVNF shah_v_Page_51.txt
be13aab2be1cdcc4819a1d11554501d8
7a580b1e146d10cae7bec280ed5ea49c3249f5db
50744 F20110115_AACVXA shah_v_Page_24.pro
7729b9fe59636a9173ac4afaf98a90e3
6f5485bcfe540c510dc5a942fab27177e26436d9
202372 F20110115_AACVSD shah_v_Page_29.jpg
f9da1423c6081f0961d903c45efab6a1
2b43ee57ec7b54ae34af014049adce1161f84738
57924 F20110115_AACWBZ shah_v_Page_11.QC.jpg
9af968687006a3e410aa8aae1bfa60b9
1d2272dd3e53778c7ef310e21c217d383f3f3052
41767 F20110115_AACVZZ shah_v_Page_07thm.jpg
09e1a6a14510b6c3df2702952ecd399b
542fc2eab33019f2df81818fce0d3be0fdf2c9e4
250524 F20110115_AACVNG shah_v_Page_52.jpg
bf40bfbe4bfef1ff6f7590343ccbb72d
44664e644c9c1e24796635c9ea497084b751c322
53770 F20110115_AACVXB shah_v_Page_26.pro
4c51eaf2626ad86975048ab5247f744f
fba29432c6fffc9ff73ee6b305651db261fef3fd
196733 F20110115_AACVSE shah_v_Page_30.jpg
a5ca436026647f53eb1c791ca9651e33
786f96821a38e5e1d7e85360610d0d2a6b6afe0b
964 F20110115_AACVNH shah_v_Page_57.txt
0f31904ec4069181aa5294b466e25319
5083ae086aaa9ab420cbdb257d0f76333ef7cd6e
47746 F20110115_AACVXC shah_v_Page_28.pro
edfba1a258b63703e70c40fac505b81f
e25b70421d083028ace99119c522347d16b9aebc
204399 F20110115_AACVSF shah_v_Page_31.jpg
29dd5ab471e51a97e2b2824ab526e194
05b3def34f70b0d0dfe40bd2280a76ff6efd6b57
79293 F20110115_AACWEA shah_v_Page_70.QC.jpg
0657584374ae8cf0352c92f01148f997
c2f35c532b56136ed194b5712680dc2db45273c8
27502 F20110115_AACVNI shah_v_Page_45.QC.jpg
8e46d11e09851e73fdb617d6e882c154
422ad5f28ea159edb61688411a099e6dfeb547dc
46982 F20110115_AACVXD shah_v_Page_29.pro
75beef21c85f178a33d5290085af733a
24f1ed59a079d1ca5d9bcc6ae07cc60444a37e6d
210027 F20110115_AACVSG shah_v_Page_32.jpg
7a0556c583f6e6879db98e00299cf0c5
5e9b107b5cc4daea76217c9c3958a74c953da1ad
193814 F20110115_AACVNJ shah_v_Page_15.jpg
5fbc6513e3a81e3575152501f40fdcaf
1e98b73cf0c1ecea4a1387c0f1ff0259162f6f85
34876 F20110115_AACVXE shah_v_Page_30.pro
06bf8fe845dc6623816fa5d89b69c076
7df77bf0e35ffa1d6cb2db5a52aae8432fdfb8df
192811 F20110115_AACVSH shah_v_Page_33.jpg
db0b492b362b1fdb266022309b2d426c
1b0019794e38cf252c8a28ba0e43bb5c3d2ab509
48075 F20110115_AACVNK shah_v_Page_56thm.jpg
f993694a171013ddfcc336a1361934a3
84c45abe9c9624cdd289966da595fc08a4a26b8d
48071 F20110115_AACVXF shah_v_Page_31.pro
c99140fe1b586b65b75b58d4f5c68eff
b3c3c26821e231f75a0e2972e760e951f67059cc
224039 F20110115_AACVSI shah_v_Page_34.jpg
49c67b653d4eeb738e5d50859888e779
61f952fb1fdd5551fc7be77b511d29008a221bfb
F20110115_AACVNL shah_v_Page_10.tif
28b1c064d783f4a07034cf013141ceac
69c1c490c0dc94ce5c4e145dddcf521ad82dd958
49822 F20110115_AACVXG shah_v_Page_32.pro
4ad8a4a6ae739357b702a149d14f6299
76b01b344dffb15d56e24cf45e09fcb2cc64772b
25932 F20110115_AACVNM shah_v_Page_54thm.jpg
af289d7c42f44d67a9a2d25ae0f5158f
0f3fa4c2d1109239aa8edbec55af8db550fa502b
45531 F20110115_AACVXH shah_v_Page_33.pro
173ee1eb0efeb22ddbce11bc7850d090
efca498769e2925cbd178165e4511b7f5e96360b
208028 F20110115_AACVSJ shah_v_Page_35.jpg
66158829aa2c8f68a4e48de011bfb940
0cb317d100390ae274fdd752842d4069a037385b
75600 F20110115_AACVNN shah_v_Page_28.QC.jpg
64c6bdf6e1afea86bed275f7f5f9da26
f72cf6707a92102559654bd25980dc47045bf8e7
41554 F20110115_AACVXI shah_v_Page_36.pro
c7684f2a18670a9c47dffdcdcc2e474d
974346151f2b71fc67552f25d047547bee0303fa
180530 F20110115_AACVSK shah_v_Page_36.jpg
e3a56eb5615b0dc630b1d8b6d25ebc64
d0664a2ccf6a620a39b42b52867831d7d89c527a
1983 F20110115_AACVNO shah_v_Page_67.txt
84ce720f6bc52b7265c462032aadf845
710320bb9c113989ae3fa63d36958959541eb1b0
42184 F20110115_AACVXJ shah_v_Page_37.pro
c1b012600b1ccf06fe6ae8e813fe1076
84f329730b0a73ccccc34ec690e75eabf2ae655f
203494 F20110115_AACVSL shah_v_Page_38.jpg
fb0bbdb65b65d26096ab835fecd5c676
97a8bf7842b798f22c0a35a333f2bde3bc6f88a3
80968 F20110115_AACVNP shah_v_Page_06.QC.jpg
e0a1309659722806f2de2d359f12e858
b80686e3e5b94606e61d8ec06905435a44b5f2f2
51501 F20110115_AACVXK shah_v_Page_38.pro
49150d869abf470fea361a61b9a482f3
f2959208eedb1bd5787e289c1e5dfee1ea36ad10
218325 F20110115_AACVSM shah_v_Page_41.jpg
e017f96cad5c764a489a2c5ff387e1a9
a184ddf847586777e72f02a27d5e9851f38ae10c
81622 F20110115_AACVNQ shah_v_Page_27.QC.jpg
db46c9ff311c72a25178375e962c436f
c74f0aa5c525198edcced99f91e90b384e74e2de
30518 F20110115_AACVXL shah_v_Page_40.pro
0cb168304e1a450b8ff95d56297c7fd8
19918d57612caff23ec8ddfce495e3a19463557b
215274 F20110115_AACVSN shah_v_Page_42.jpg
efce6dbcc72806dd46d87ee37a1a856a
b8444faf2fdbcd629b6eb36e2285a8eb87895ebe
71349 F20110115_AACVNR shah_v_Page_33.QC.jpg
8224e319bc81047f57be34ec2f3c88b5
8c4e31190b38effd53e52aab2a31f174baa07c79
31541 F20110115_AACVXM shah_v_Page_42.pro
b58a15b440680c8e921676e33503b968
6acb0bfcc689646119b549bf1239c2283b39cc81
221770 F20110115_AACVSO shah_v_Page_43.jpg
6180ceebbed300757254541e601978f5
f53a2dbd8adcc80365759a9b6d3a53c05948c5b1
221275 F20110115_AACVNS shah_v_Page_06.jpg
8b998c682e79aa0f3fa129d738ea607c
ca55a5e7a7bcc05b4a9dabffe4b03a5894aedeef
31925 F20110115_AACVXN shah_v_Page_43.pro
34a81c40881492c832b7071adcd358f2
93d8cf0190e4b19ee482b4e0c24f9244de81329b
79479 F20110115_AACVSP shah_v_Page_45.jpg
353bde1d5238146e4a5f878f897d0d38
b948702717d2e4b79010320f768fb8a7e2d8921a
108844 F20110115_AACVNT shah_v_Page_20.jp2
0852032aa8866a9b0e374d854b2f4bfe
1a923830284a368cf6f678b70c53ad39a6fa15a8
165151 F20110115_AACVSQ shah_v_Page_46.jpg
6e4ad97fc21cb5d2f315c93192987c46
c0fd186af4013834df17295a9a91df9fa8b76f1a
22538 F20110115_AACVNU shah_v_Page_16thm.jpg
7a484e5f72fab4dc6eee1cb89bf82b62
e3d44e722c8d9ffe9619a1979aa356b17352d3ca
38005 F20110115_AACVXO shah_v_Page_46.pro
8a17629cfe7e55363ab0b423d8750d0e
7b4be627c0faf810a4288e0148b0f04c75f4ffa9
274051 F20110115_AACVSR shah_v_Page_48.jpg
7f939e556e7da6151b9baeb64cc6da4a
420e69055259599af20ec8a45f03e19ec3d11bbe
46226 F20110115_AACVXP shah_v_Page_47.pro
2a530c50ea96e0b3068e2c230b91d0d5
a8514b47a4b5a1d0e7613a21276bfad16aa4a111
255006 F20110115_AACVSS shah_v_Page_49.jpg
172649ffeabdef9d377d0e9ee6484145
b25ed2d934f320024d326288d6253c6b8e056c88
206254 F20110115_AACVNV shah_v_Page_21.jpg
8dea6081392a5a84d08ffa9e35988e1c
ac810c0b9de0f8ab20473914bf2d6177d24673e9
20463 F20110115_AACVXQ shah_v_Page_49.pro
245ae73ec8d135c3a51676e267119f0b
dc2a6a07d48803e9d18d5e6856fc44100fe4a97e
251561 F20110115_AACVST shah_v_Page_50.jpg
9e0bde45880a4c59f0642d7838e4cdaa
1d3dbb5df8a2ff0b1224d9157e268eeb73779701
37027 F20110115_AACVNW shah_v_Page_63.pro
f289774470a69cb417f677464c2f4099
0e971ffa67fd95a90f70f9a8ec1c68352bab5786
54925 F20110115_AACVXR shah_v_Page_50.pro
e17e4f3ea2110af7fb2010f80612a7fb
e017f75dc215aa22651d42fd66db33214f13ddd3
248080 F20110115_AACVSU shah_v_Page_53.jpg
9ee7fd9035536e52c3de569d83ac5d42
4022141a9c92174d2416f93d5e6f09e89d58c65c
35454 F20110115_AACVNX shah_v_Page_59.pro
36718fe8c851302c5076278b9b039fc2
e2c6afb240f98387cca753bdce2f9732839bfbbf
40221 F20110115_AACVXS shah_v_Page_51.pro
639cb1031fe2b30359b70db01c2cf0c0
cb83a75bca2860d23bf268cb971cd5d6a2b41c1d
266049 F20110115_AACVSV shah_v_Page_54.jpg
f57071164278a09897cf2e0572f256ac
a78284ea2c3d9674807d042a9752e8edcb46e70f
F20110115_AACVLA shah_v_Page_05.jp2
74643b599e574fcbf5dadce63834492d
5fb6e789d2af3bf580505fffa7e05b68bbf35b09
24994 F20110115_AACVNY shah_v_Page_22thm.jpg
9c1e901e15e3e62a5db08305a633c6de
eaec2d63f3a3b17b55c269de5ece13b07529f5d9
13187 F20110115_AACVXT shah_v_Page_52.pro
b3b9d03aebe03943b61c48710a12872d
871de8c4abe4df2ec54960c1a5236bb68a543216
213548 F20110115_AACVSW shah_v_Page_55.jpg
f4d73e20d9afbe706c46243cb17a6c21
5eae134038ff87651dcc5d08240d93779e776a76
207039 F20110115_AACVLB shah_v_Page_51.jpg
695e9b54b3be0c3e34985dc5718f8494
abba9ea327413488ea6f12829630d41dc38bb11c
F20110115_AACVNZ shah_v_Page_59.tif
fc572237f46a42ab3b2fc6b134d1b7e3
9eff2c7d09379abbbca4ae27f7a51a97b9c2d1cb
18170 F20110115_AACVXU shah_v_Page_53.pro
7d8989b6e929363e794cfa9bf7133d63
314824872bd7c16e381c34f4e6458009b68bfb8f
265616 F20110115_AACVSX shah_v_Page_56.jpg
b2b4a82651c777df297040f4aa3947ad
0bebbb1fb8f2ce7dce8567fcb261ac8d9cdb3ed6
37346 F20110115_AACVXV shah_v_Page_55.pro
249a951f282aec0d451c7df3d4ce3f7e
c36f220b1049d2e62fe340350ef4f86afdf0e1de
5116 F20110115_AACVQA shah_v_Page_03thm.jpg
f4ea99c0bf1f29f07afb5242ae3e96d1
17fc07380a0939fddc4ad4a6c4b6bf8a15386f92
270338 F20110115_AACVSY shah_v_Page_57.jpg
b431bf1dffc5bb2e09e53c1ae982251a
0cb5f13c7c3312974b99c13e5bb3dcfc8fde2b9d
8364 F20110115_AACVLC shah_v_Page_01.pro
0e223be02cacb8518a2f2beff947a51c
7303b1d021c4593f665f798835caa219bc761346
59437 F20110115_AACVXW shah_v_Page_58.pro
d99c726912a18203c4e789c588de3b30
0bccdc58b79807915f2d81a39b15d855c3afc2f2
27858 F20110115_AACVQB shah_v_Page_17.jp2
1e5a6ee954e972d953fc4be2d3c1e693
4ea4992b2f3ad12a29d41cb40dd72284990b7d7f
268495 F20110115_AACVSZ shah_v_Page_58.jpg
1ea37443b35bcc49cb922f594c83f8e9
c2c705e1820a6823f1d55dedefdbd8fdd65a2a46
17095 F20110115_AACVLD shah_v_Page_45.pro
149298f928938701e8a93fdd633119be
76a07bf9913b6f18a6e321ab28d2220b239a203a
28988 F20110115_AACVXX shah_v_Page_64.pro
ee35ee29934075fab49399afe7bbd910
83af85f9846a19cf46fb8fc31707e7e42ce329d8
22929 F20110115_AACVQC shah_v_Page_29thm.jpg
069ba3f7aaf70c49b48e98f34c962ce6
0a1ba7234d034094f80eaa925ae6dd1ac1c23dc2
60847 F20110115_AACVLE shah_v_Page_46.QC.jpg
9c58ae294185f9ef2ae6a18f7e649e10
3372b89289363c0625f1298e9762d359b4724adf
41424 F20110115_AACVXY shah_v_Page_65.pro
73a9f211dcb00dac297167ef6638c167
916d0f52b7717db45b0d121016f29321565595f4
F20110115_AACVQD shah_v_Page_34.tif
ff7884dbecec18366df86fd4eb1a525d
9d9845ca8614cb890d1c65a34848ba4527cdca26
115928 F20110115_AACVLF shah_v_Page_34.jp2
d26aa130554c11cf4b45f28c9e2f4297
acd999f59609e1e06c27f8f3eef97cec964eb5bc
F20110115_AACVVA shah_v_Page_12.tif
44730cd7fcfa46b9c417607780f1342e
c4965aae7ebd3b970f49a03a621fcf14d9b69fa9
39826 F20110115_AACVXZ shah_v_Page_66.pro
6dad682acc6bf7cd5c45b93828641377
b65809745dcfbf3c97de22cf5337e4d39edd0821
29240 F20110115_AACVQE shah_v_Page_09thm.jpg
9f62750ffd0024954c2413d07159e658
3339b7a61f9c4f91cb5a66cb663b12da1c8e36ac
117554 F20110115_AACVLG shah_v_Page_26.jp2
02004456409c5cfb38e375354ce7b085
269a0a670e3e8357875b1364952d5074c594daa2
F20110115_AACVVB shah_v_Page_13.tif
76ae724960b8ba02e191bbed22196416
d28e9363d317d608ec7cfbd5a85644ab4134b7d3
81726 F20110115_AACVQF shah_v_Page_34.QC.jpg
614659f882682ea2fb0552440d7b2aae
78f3f19bb2fc8b9b99e20989946eb46de9af12d3
1517 F20110115_AACVLH shah_v_Page_41.txt
81e90be28188547024e4f4744ea6ada1
1934c31edcf547df86fc7c32f3275686384b8614
F20110115_AACVVC shah_v_Page_14.tif
540aa8b74d4bab4685862530499e800a
8775ec6e0dfd9765a4d4ad6271d669759b94ccb2
19362 F20110115_AACWCA shah_v_Page_12thm.jpg
36cd812ac33e4abd4f8be0337969983e
4c82d4ed873b289192045bb78b1f846dd79a6ad8
F20110115_AACVQG shah_v_Page_05.tif
8c05f3a1831794503f6c9434b2dc6acc
77dd52cbaef553624107b5162630f3393a608e7d
135482 F20110115_AACVLI shah_v_Page_64.jpg
7a0cebe72845903acef2589280589ee4
ced9342e80592f1644717a38ac721e00cb25fc93
F20110115_AACVVD shah_v_Page_16.tif
e6fe218512bdbccca3b80962c08bc097
b348c247e8b0e96a1c0dee14432e091fd48db97e
58423 F20110115_AACWCB shah_v_Page_12.QC.jpg
d9d656a199f4293426677182d367a4a3
38466ac8b80b7dc0218e9b64543c38cbaccd507c
1051973 F20110115_AACVLJ shah_v_Page_06.jp2
4f275e16491a4b8eb4edfb9d904ede10
f2215b38057d9ca1a09c3c330153e54a0ccdedb7
F20110115_AACVVE shah_v_Page_20.tif
b9ddd5e1d4eb961c20b0e4d54e8c515a
84f612b9b8a763cb32242488cd35a8c0fa595c1b
80811 F20110115_AACWCC shah_v_Page_13.QC.jpg
20a8d1ac14af379daf62170583aa3c20
12298a0c0140f65cc798a7181adcf017f3344fbe
25699 F20110115_AACVQH shah_v_Page_04.pro
f42cbfa31b19002cbb8a5f8955b0e977
fc248d612afaec286ae2f401d2424fa883ef4da3
73896 F20110115_AACVLK shah_v_Page_23.QC.jpg
e8cee4fb31184a7275374a0fa4bfaeac
342edb19fccba6e94ea7636ea6480a399e1b6d35
F20110115_AACVVF shah_v_Page_24.tif
8c185510627f5e7f13b50547dcb20a85
ae7b04de0a1dfa100cdd69420d3427f796704e7f
25159 F20110115_AACWCD shah_v_Page_14thm.jpg
0a131cf9b4a2ff6d4043a3bddf936642
9b9b57298bd27231501260d31fe5535c9ec90df5
260669 F20110115_AACVQI shah_v_Page_71.jpg
aad81a90b46fe4a5c3e5dbd21f011df3
6d06bdbe29486c304697717bf1aef93463db1891
49079 F20110115_AACVLL shah_v_Page_57thm.jpg
2391662079f97c04e5f3980c4752a119
b3d749670b2b576a8c0036a0fe1a6c240b5e5fd7
F20110115_AACVVG shah_v_Page_25.tif
691177a6468c58f66044a0859c883ddc
cbeac2977a3f18433b80bf023685895d60ead4a0
71667 F20110115_AACWCE shah_v_Page_15.QC.jpg
3901c0c09aa46ba1d6583127fb6ee55b
c38478a2a8d93d6fae3cd6ec05701eb74e3ae767
1932 F20110115_AACVQJ shah_v_Page_21.txt
2f692cdb1d035d0825773d3be8ebd41d
55df78c31eff84d47175431b3e07bbe7955cd489
195608 F20110115_AACVLM shah_v_Page_16.jpg
b5897803f67bc35deabaa16b48b71879
6fa88ec2dc7205c457a8c8c8c65f25f3b92b6355
F20110115_AACVVH shah_v_Page_26.tif
bedca20066576c446651f27e9e57aeb4
aa2721f0127200b055df26ec6b22ece8bd9ed9c1
73776 F20110115_AACWCF shah_v_Page_16.QC.jpg
32af4b4efd8c04fac278135df0f874e5
3a5db18eceaf8792c7b80fa07712c02d350159f6
176561 F20110115_AACVQK shah_v_Page_44.jpg
0790a2c294eed0f80939ba506de5de68
3107b5be22f8d5e65ac249331abab5c7250d904e
1490 F20110115_AACVLN shah_v_Page_44.txt
394b6cdebfbfc487160c5edd80af5557
99fee48e9a7e8955618af876b9d20b5c9b937b1d
F20110115_AACVVI shah_v_Page_27.tif
289c796664781da42e2a73069a8d8256
1aa347f507a1bc5cc0c5731ee957809b142eb102
68510 F20110115_AACWCG shah_v_Page_18.QC.jpg
587b73065f8036716164556e6a575ffa
cb1e6b77235546fb7ab7df1ae670bc7e2ea51261
124167 F20110115_AACVQL shah_v_Page_54.jp2
6f6197a7e9857f5dd7efbde5349e7ed2
603e0856fbd3d5febf98937e492e6d15723ee124
F20110115_AACVLO shah_v_Page_04.tif
30d059e299aeb5fa8b8735767d583b32
8dc5a6682d6dc9040aac46e5aab0c32af5a2a5fe
F20110115_AACVVJ shah_v_Page_28.tif
b5a36ccb64d1ba4e068e9d1c54838c5f
fcaa94df43c0420940145aa684ce52ca0edd0786
74297 F20110115_AACWCH shah_v_Page_19.QC.jpg
83dda9a0da55988e0bdff525dedbf8ec
fad720437175c1b07a16cedd7ba1477d0939acbd
219052 F20110115_AACVQM shah_v_Page_25.jpg
a246edd6bdab69bf7b8f1c28b8b72e79
6ba7bc9c7442aa3d0899248c097c7e1bb992bbb2
91090 F20110115_AACVLP shah_v_Page_37.jp2
a2e0038381bc29d989299819ed7da45c
c753409d7840c0fa20f96024bbce4b5df9840467
F20110115_AACVVK shah_v_Page_30.tif
dfa99634d77dd6162ae3b70d4f7fdf90
5435209e9268871b614864dca370ee974800505a
23695 F20110115_AACWCI shah_v_Page_21thm.jpg
d002c8172c7b4a5a60c7faec5d4f46b8
1111e904cc15b94aaa50c8cbb5b40163c7d89a9d
1859 F20110115_AACVQN shah_v_Page_47.txt
d35b0259616a735fed60d460f571c3e5
de4c9f21920bcdeee0ff3bc04795df70ffdc6e22
1842 F20110115_AACVLQ shah_v_Page_37.txt
87c4aab54e4c2d432a4b07c052a974cc
d0114df6e88f41ab56fd8b2151ae9dc44e1bd3e4
F20110115_AACVVL shah_v_Page_31.tif
6e462819677e74a88ff3cdbf12c3d1cd
c5bcd06eca2c98f22650c445c22975898fe1e028
77610 F20110115_AACWCJ shah_v_Page_22.QC.jpg
34e324084eb5dc8d3219034de91060bf
9a95f9cd8c3bfa7f823ed248e4473346a40b9012
2578 F20110115_AACVQO shah_v_Page_54.txt
7fa47dccf049127553949897ed493eee
5c0f4dc98f00b82a5a39c1c3a7c818bfeed3581f
57795 F20110115_AACVLR shah_v_Page_62.pro
a229d3dab36ced8abd39be4925ce0765
bf88c40b8fbb41cb0d8950d7bc0ac98491eb5c5e
24691 F20110115_AACWCK shah_v_Page_24thm.jpg
6cfa2cb19258498b1b12c2c58cb9d0a4
47c3f36d09978a7f186c03b107317e1dcbdefb29
F20110115_AACVQP shah_v_Page_58.tif
e670b689cbff0bb38c29ba4985177170
bb1d90186b18d62ffc5004edd5f436460cdee06e
94154 F20110115_AACVLS shah_v_Page_65.jp2
42eb7c5c5df9d85cecb6164c58677568
74252f708e46d3f464e06b8d05d87b7fed2cead8
F20110115_AACVVM shah_v_Page_32.tif
48d01c0c54c198eeac1b133aa7044933
7a038b47e8576859a5c8504ac5143ab496979e87
81521 F20110115_AACWCL shah_v_Page_25.QC.jpg
8ea8a8d0deab565e52cd63adad546197
ced4601b1bab9326d93c82d20a9947db5a8d4851
48947 F20110115_AACVQQ shah_v_Page_35.pro
c3a4b49a0d636b43b89a747ac7d299d4
96c580e1dc502e949778ce1c9afdc9f8a3c0384d
1051957 F20110115_AACVLT shah_v_Page_61.jp2
d5b2793e7acd36dba9097389ef0b9cc1
ee78680254e628ba2386748f918b89d90db670a4
F20110115_AACVVN shah_v_Page_35.tif
1cba6918b1fb3e15b97301823087e2c0
17dc656a4666f9453183666738e6236205779501
86144 F20110115_AACWCM shah_v_Page_26.QC.jpg
85224f0b33d4b552b67e91c8cdc3d61a
6b41d8a0574d1ebe3d2d00faca947cfaf7c8f0ea
95903 F20110115_AACVLU shah_v_Page_42.QC.jpg
53900d04777cc399999ce555ea86c57a
b3524ffaf655049079832b21773e819dca845baa
F20110115_AACVVO shah_v_Page_36.tif
483f086b39ae5b53d23dc1eff02febab
822c4f2a68f86a2fe3793d9cf24c0dd987dfc161
49588 F20110115_AACVQR shah_v_Page_20.pro
6bc79e3eb0de7ae25ca9e9cbc0691560
ac8e869f107e7d89e8bc9124e0ade2078b436155
24905 F20110115_AACWCN shah_v_Page_27thm.jpg
b0c32cf8599d8534f8b8fb400edf8ce6
9cfd029349cd9f96b1a9956d9baf35ed237d8776
109213 F20110115_AACVLV shah_v_Page_35.jp2
9c23749a9ec53b53f172d5301f4513f5
8b64b3471c4ec2add90bb0f9bc1d76fecfe341fe
F20110115_AACVVP shah_v_Page_38.tif
a4c248686a81b88c8d7a124f4ac7ecd0
037a444a7769b06b7fc088166d16cef54aeeb043
F20110115_AACVQS shah_v_Page_69.tif
b5e411d2a56d9e06b39ac8aef4b97303
c90ba5a722d46cbcdcfabcc64c5db1a82b5e3f58
76977 F20110115_AACWCO shah_v_Page_29.QC.jpg
58ffdf1d8320fbe87ff53420430b253a
8b01989e5207999a894ec8a717bcc584ba62a616
1968 F20110115_AACVLW shah_v_Page_63.txt
83d187b65793c29ee73c83da3dd5bbd3
434641da641b51d365db0b2aec58d4c1515358ff
F20110115_AACVVQ shah_v_Page_40.tif
33b59b9e882fa89ccf2678d73405fb8a
7597eeee6ab817a60788ace6533ff29c8ba0b592
1777 F20110115_AACVQT shah_v_Page_36.txt
def08baf1b2d264518dc88ddaedafadd
12ca4b31375050526b6b39a10eadfd0a24ebd846
46881 F20110115_AACWCP shah_v_Page_30thm.jpg
f98f2ce60e8aa8c8bad6dbb943883066
181376440f69a2d341c405e8d7e54278710f8908
F20110115_AACVLX shah_v_Page_07.tif
ca9bd89457635e54af1dd136cd271456
d49d49b03d0269295c8ad728d09580a858ef8127
F20110115_AACVVR shah_v_Page_41.tif
f893ae10020f52f68f724f547b9f9701
143b93c8fec2544f3a7a51a179acd67ac81821bf
F20110115_AACVQU shah_v_Page_06.tif
868d10f33e55dcec0a715b111e7b02a3
95c84f31536b188c3ec7f33048c0777c2ba198bf
24744 F20110115_AACWCQ shah_v_Page_31thm.jpg
4aae79cbfa436097d8293d2e664ed8b1
427b25b96ac83d1ce6386f5acc01429453fa650f
F20110115_AACVLY shah_v_Page_02.tif
765aff2ccc97ac589829c82fa45d73b0
9dcc762a08045e40048d8493f9f50a5cb09e3cc0
F20110115_AACVVS shah_v_Page_42.tif
4a8f703fb8c668052a26277e718d8fc2
4bad8832fd8b75d48112a4cf8bb975a9c8abfc31
14582 F20110115_AACVQV shah_v_Page_72.QC.jpg
53718d442238755f4a9f9fbf7264dd3a
81440001402e0d46ad092463b91604357fd17e47
76054 F20110115_AACWCR shah_v_Page_31.QC.jpg
586707e1aa3cb097e447b6cb4bb7f29e
d920f3a7188583cfe81fff043b18cdde9928058d
2598 F20110115_AACVLZ shah_v_Page_62.txt
a353f2f806546e25d915b77191f509c8
746dddf6503175788e8a566603c55533e09533fa
F20110115_AACVVT shah_v_Page_43.tif
afee00f5093136ff3687d4a65ef19ace
f8616c4ca07cdcd5c9bbcc6e992ed72174a87131
59484 F20110115_AACVQW shah_v_Page_06.pro
bf4629273f9f8155ba5fb52927446c43
0ec73d54762a2fbd154be7c21adb058e1515786b
22321 F20110115_AACWCS shah_v_Page_33thm.jpg
c0042a6e1fb72530e1de37fb7291c459
634787f105974303eabaf057586d18ff31e2bed0
F20110115_AACVVU shah_v_Page_44.tif
248ebdb53f7c5d30b03a81c590fa08d0
31f5e631559aafa7e19437a7129fa1bbf12cc5ce
F20110115_AACVQX shah_v_Page_18.tif
f132883ede8d6ed0fe3eb5b1856fe0f9
2c0bea4c8b412e0bc6c9555c0fe81232ff0ae072
25163 F20110115_AACWCT shah_v_Page_34thm.jpg
2fe0e7395aba1c744ade7a51aca3aa1e
b47f5179b64285d9351045ac0f52382c747ea34c
F20110115_AACVVV shah_v_Page_45.tif
7f18c68197bac98894753f1a4370d466
afc6bf88d7059d35e692fdcba62e9c334d53da44
219710 F20110115_AACVOA shah_v_Page_27.jpg
7bd19d714bc69a4b4d9af5ea938e41d8
a00ffb43ec9f9d780f5a249e7a5463fc89b4e6fa
1051978 F20110115_AACVQY shah_v_Page_43.jp2
29815a986e392dcb842d142cf1c4f59b
600c4a2e8dc2d6e6fd757f9b84395284ea93c4be
66996 F20110115_AACWCU shah_v_Page_36.QC.jpg
e7a6ad036ffee033a284fae5400fe358
499c9e3a2aa2abe1abc682e034bb4aa67f8cf84a
F20110115_AACVVW shah_v_Page_46.tif
8377f8c8401f387944a5b3921b1ef773
a38b0eb201c3c6c8b5f5bc39961280c12bc69116
1320 F20110115_AACVOB shah_v_Page_43.txt
478089027f0a364b7e1557520cd8971f
b69bf81f5db0ed7af0045ceee21f8d3e2d6f41b6
48150 F20110115_AACVQZ shah_v_Page_53thm.jpg
a647d87ea4c6612128592cca40656807
d858c565c1a11b3f1f67bd26b7c5ef2389c03bdd
21970 F20110115_AACWCV shah_v_Page_37thm.jpg
05b36732128d3a19c529f54214824642
bbf5d9131d218b3cfa5a29da6694b1f3eccb20ee
F20110115_AACVVX shah_v_Page_47.tif
579fa8330aa8e49c7798413b5b8f7a96
9f9e847335c1d09985943059bbdae9eeec4cb935
14440 F20110115_AACVOC shah_v_Page_60.pro
b42324ce477f8dbe181c16cf121c3423
ac4fae83fd936ef1e248da02d6fdc5b6c9483436
64579 F20110115_AACWCW shah_v_Page_37.QC.jpg
f57c3a829004e13b9dfae093e89eb9ad
594e415ed45258564b01e07cde8e5e9f8219761a
197408 F20110115_AACVTA shah_v_Page_59.jpg
f0f94d4cff75956a8e0f8dbb76a8446a
2481eaca481c268ec2dc9953d4a4110ccf26c7a2
F20110115_AACVVY shah_v_Page_48.tif
2bad89e7c6d059f8532230cd01c520ba
e126955f44d7abeb445942d04b216ef87fb9853f
1822 F20110115_AACVOD shah_v_Page_15.txt
5f6d90b198c7c1f6a942b103000cdaa0
f105b04bfd740995565c3316a375e5c3dc944f9c
23535 F20110115_AACWCX shah_v_Page_39thm.jpg
94d44430dfc98d5c84f179994421763c
7b9e2623393527ae7f787124841c99d9d541d580
245253 F20110115_AACVTB shah_v_Page_60.jpg
ef0b6564fc6e1cf19e804ec9f63caace
8477fcf9359c2cd3aa3c27a018bb86698af55bd5
F20110115_AACVVZ shah_v_Page_49.tif
51ab4c179d00d9d9c2bb76daf3ece6a4
5748e45e68f9e74227f28af2dae960785bb18518
55362 F20110115_AACVOE shah_v_Page_08.pro
fe3016c8cccadaa27ddb00e08d863529
5bcedb76b99f149afe3b579e23c2d66412f1da2b
263554 F20110115_AACVTC shah_v_Page_61.jpg
74f6b213738f601c45abbc2fbc4033ec
ca38efb46fa3ce278c8977b4570f01b8bbb499fb
74730 F20110115_AACWCY shah_v_Page_39.QC.jpg
43ab1f93a8433184da1fae82f2df6509
9c351c5dd1f67689ed39a5c33f8598861eb2ea9a
26495 F20110115_AACWAA shah_v_Page_58thm.jpg
1a7a985060b322e6f31dbc84c0aa2d55
e99399ef8f6eb4f4310e9e8b83a9dec8800fb192
271990 F20110115_AACVTD shah_v_Page_62.jpg
6dab3267879e10d8c024c356dc994544
d544c8d8f1222b2ac16e03057da1ab1ed0254a85
48838 F20110115_AACVOF shah_v_Page_43thm.jpg
8d1d7ebe5b5d993d8beffdc4b841e71d
df900bfd32e1274d718ea444583aa85ed0d4699a
50376 F20110115_AACVYA shah_v_Page_67.pro
d116aa74eedb518b075e3ec75162275f
4c07a2468bf98e5af874e9cfb60f11135f05f808
88871 F20110115_AACWCZ shah_v_Page_40.QC.jpg
ea966a0f8b0d4e81428765b96aaffc84
3d6987f5398ef55135d63a8646ce076d5a570ed5
88229 F20110115_AACWAB shah_v_Page_30.QC.jpg
6f7ce948e022512424716f3b5a08c565
061ba59527be4d3f484404c91ffc0143b6de58e7
206975 F20110115_AACVTE shah_v_Page_63.jpg
c26e657220a00b66fb0bf5463ee32053
b99e9f1ab571aa20f908264c22c491eca17ebae5
F20110115_AACVOG shah_v_Page_03.tif
3e4e697592853bb829defb7e6a6696c6
b304022313011d4b45b70c488c9d5a03e557e607
39182 F20110115_AACVYB shah_v_Page_68.pro
7ff8545d4e4eb7dd658075a019adffb0
ff5a01bcc98bd32236614e47cd26d2fe5ec13f5b
69675 F20110115_AACWAC shah_v_Page_38.QC.jpg
1f958455f1c4db607a9941f28013522f
14db8ba90c64c4ff6f75cd4d438027a7d83289e1
179689 F20110115_AACVTF shah_v_Page_65.jpg
0b667517f0457ccf6c46c181931e0cd7
44fb0182850bb2d4972e0b3f26d821718a43cd6f
11643 F20110115_AACVOH shah_v_Page_73thm.jpg
46fb1fb04804da58c1277c352403cbeb
15ab775e5b14ce4ae7a9eb67fa84a85b4b281bf6
19028 F20110115_AACVYC shah_v_Page_69.pro
27f7f69d47de16a40bcfe990b4de7fdf
05d858ac1d4234ebcb695dac41018fdf01ac0269
3072 F20110115_AACWAD shah_v_Page_02thm.jpg
f699bf2b6691d15722dbbb0c676aa4e7
5220806f318ffd7f39044bf937316d8892ae382c
214484 F20110115_AACVTG shah_v_Page_67.jpg
e74234c5badc41615212f8c82fcaa954
e4a2dc2480c955269984f107a2493afb718c69ba
94277 F20110115_AACVOI shah_v_Page_60.QC.jpg
731022bb5e34ce36fde3b2fa51e227d2
45c44420a6ef5d04cf8cbb237d4b72fb5d91b37a
56130 F20110115_AACVYD shah_v_Page_70.pro
f1a4573b2e03738e6d756940d00390dd
84c3f7039f4be39966b0cb601ec8b1558698022b
22687 F20110115_AACWAE shah_v_Page_18thm.jpg
6113c6fdef15fd2247d330d6beb1fe98
7a09fdbd2507c2cf0682bdcc5f3da00706103df9
175970 F20110115_AACVTH shah_v_Page_68.jpg
fa399411e8c9eb299bf0ec0f129d232a
32cb6375eb46601d5b4e70000730a85bb34521ab
22417 F20110115_AACVOJ shah_v_Page_61.pro
357e0a29a46cb0690630fa50ac200bcb
e6dcf61d7dea49c1413c002640935c83a0cdd681
59539 F20110115_AACVYE shah_v_Page_71.pro
8be58ed73ad2dcf8caff143ea32b3af3
9ac8e6125ad672f4c691ada66ab9c584e66daea8
7385 F20110115_AACWAF shah_v_Page_01thm.jpg
dac7200814a3234550a4a417252c7361
b82230c765fed6647fd94776d87c0c41118d914e
91358 F20110115_AACVTI shah_v_Page_73.jpg
5146e413ad21cfaf69194c778e1d2f99
ed8e609f6967e20167a94e1b33a1fa8ee55c27ef
1652 F20110115_AACVOK shah_v_Page_12.txt
f731b622a7fcf98cf28b64b398fd98e3
40810536c1000f059e0aa95fc1d473cce1b5cdb5
9659 F20110115_AACVYF shah_v_Page_72.pro
5c7bac50f09f074c3b4ddcec47f8491f
2b4c8f5b13be0238548d2bc0ffc6019610e31a47
10117 F20110115_AACWAG shah_v_Page_45thm.jpg
d6d679b1c2bf874e319e3f064c316f5b
d16b3e6e45b1bfefd36e7f8fd111d2a393fa02af
5933 F20110115_AACVTJ shah_v_Page_02.jp2
c307d347f721b6c5f6503e1cd4944615
1fe77d03167b7c5034a3a9ad86678b1ef2abe5fc
1012440 F20110115_AACVOL shah_v_Page_40.jp2
18d0c2e68144e5c1a5dc5ec0a0971196
7d734d1768549a5ad5a7f26f25841c07132369c1
461 F20110115_AACVYG shah_v_Page_01.txt
74455b46e6f36cc27f293eb85d0d0103
34396c2fb606ee5deaf3d5a15121bbea6061fae7
24023 F20110115_AACWAH shah_v_Page_50thm.jpg
3ef410540c01e074bf313fd1be0c7558
e1459df9ab1fd8f16c913aeb2313249ebf76765b
F20110115_AACVOM shah_v_Page_65.tif
151ef4964ac0d938087b9b6e47102205
ed3a09b8a677410401d28c260b91e2b3444c424e
104 F20110115_AACVYH shah_v_Page_02.txt
e59cc3dd15c1a5c5979ae0503edc9179
f9f21cd99bfb42f31eb0802f5597c5f8c4174f24
25766 F20110115_AACWAI shah_v_Page_26thm.jpg
25e2f9b27259bcbd8cba2309ba487fd7
0da3d2e82b8a5da53beb18d03dababf56468b34b
58609 F20110115_AACVTK shah_v_Page_04.jp2
9c7bcf1bed6375ce47434c00f4c1e996
273795f5a073ad6e9e87f5b414617cf4a336e63f
126643 F20110115_AACVON shah_v_Page_58.jp2
ea46dab51e5fc2fe99fb20f0d439e3b3
196b80ae2192ad013ff2efa89d552b44f4e1e663
2736 F20110115_AACVYI shah_v_Page_05.txt
5643a2c31a597691ec5a1ef2f25fd32c
72ce4f99b57ff27f7565f734a64e5c762212d038
23518 F20110115_AACWAJ shah_v_Page_23thm.jpg
eec29a87aabe262876cf0d5fe71da0b0
d05b04196c8933417ec51b915bf740f5db0fbe07
1051977 F20110115_AACVTL shah_v_Page_08.jp2
b0973b844884f2f4052e9475907f6a82
81517e68d6052066c33d70ea2146e0f51bbf73a6
F20110115_AACVOO shah_v_Page_72.tif
f032b1e3dcc154082691f8cba5e83779
efba9083c7d09932433ce3cf07cf747a39cb5560
2172 F20110115_AACVYJ shah_v_Page_06.txt
cf9a41f02086a47ae175e606ef874c09
fa7a32df33cec40df348f1516ef79c838ffa40f4
23802 F20110115_AACWAK shah_v_Page_35thm.jpg
b96cb6593903af1fba33f2a2ea5a6ed6
af8d04083c40f0a149e656fa480012b834981250
78716 F20110115_AACVTM shah_v_Page_10.jp2
c15cb2aa06da3889030efd35b8273bab
cb1e87751be2c54475ed346473699303837b7cbe
48552 F20110115_AACVOP shah_v_Page_72.jpg
c2f40c9b1f2af744475fbcb72e002b59
045ab36a53b01ada7ca22b0d1437fe8b4f2d01e0
1378 F20110115_AACVYK shah_v_Page_07.txt
4a1549d818d89820cdc7e26573a93d80
688ee21bdd357c0bbcc4d48e789e9576cfe149a4
13994 F20110115_AACWAL shah_v_Page_04thm.jpg
31d5c6506fa1e8ea0c353cfd722d7f84
3dce6efe634e6912698d3a9d9c6bcac82d8747e1
12674 F20110115_AACVOQ shah_v_Page_56.pro
9339550dd6d9afe39bc6d528c0b8fa2c
007b4a9660585c155adcd826015842e1a19a3550
2074 F20110115_AACVYL shah_v_Page_08.txt
549d129806daba85947fb6c42578c281
f3bf1a4a606efef60fec4ffda47d476c374832ce
82125 F20110115_AACVTN shah_v_Page_11.jp2
cb696215d8cb509f932370fee6bb9d20
1ad56783b43786881c3378f95f80b60fbcaf6761
78721 F20110115_AACWAM shah_v_Page_32.QC.jpg
451828d08f9fb2e51223743d0bbda038
901e5556972d6fa8575b4362e68899beea5cb9b0
776147 F20110115_AACVOR shah_v_Page_44.jp2
06905bc5779f2e320265ccd17262d461
58d7665c1bda61919539f6b509bdda5658770b04
F20110115_AACVYM shah_v_Page_09.txt
ccec9b60feb7519f3a598f1794cb1f53
c3f00ebeff0553b70b9e415953334e4b1d8f75de
87414 F20110115_AACVTO shah_v_Page_12.jp2
856b064c5415bcecdf924b75d896f3a9
c1b96daf33dd452cc9c22b194b553304d0043236
32997 F20110115_AACWAN shah_v_Page_73.QC.jpg
3ae184d067d58bc142f710c5aef7373b
72beb74c055a059489acb1515d3dd9d97eb16cd8
23243 F20110115_AACVOS shah_v_Page_15thm.jpg
3c24046ab38427ad8454bff390b7a499
dbf5e3683295ae1068c92936608cc57d11e6b294
1440 F20110115_AACVYN shah_v_Page_11.txt
1198c5c72dc4010e9f46fbe8745f5386
e9c27c0024c76ebefbdc2cae5aebcc8bbea6de34
113159 F20110115_AACVTP shah_v_Page_13.jp2
ed07b7530b4a32f5ca58385753c0c5da
3bf1f489e5b52cf2b738031d8da7551c330e5ec7
101480 F20110115_AACWAO shah_v_Page_61.QC.jpg
069ab738c23f209a1c1be45c3646f341
a810c94df4fa27f03efbc14aaa9440b77fab8dc8
18951 F20110115_AACVOT shah_v_Page_03.jp2
46538813e536d4fc40dc29b30e17023d
8ca722305483dcc929d44c37c7c36e0e529709e4
2102 F20110115_AACVYO shah_v_Page_13.txt
75cdc914faa332b0bad497f2e6bcad82
ce09813344da18c507ef83922b700644da746535
113238 F20110115_AACVTQ shah_v_Page_14.jp2
0c93538241b39db4890018d0c21e01b7
73973a926624dd1d3174e4d0ae8776adff8b4f50
17873 F20110115_AACWAP shah_v_Page_11thm.jpg
34d33b0d43fb6f47b1c41b063974f798
f46ef606c28973053523ef7a3ffd46bb254913ed
97031 F20110115_AACVOU shah_v_Page_43.QC.jpg
2e763df59bdd325e383dac4772952802
e483c39a4ea954bcdbffb5c25821b39b06fe0f88
100988 F20110115_AACVTR shah_v_Page_15.jp2
7c7c3ecc614d6bb55235a464ed84f3bd
cdfc0c8f9e7bf6bf4714a1353a81e2f0fd2e68af
32932 F20110115_AACWAQ shah_v_Page_09.QC.jpg
6d2b9f4e7739c07774849713a6639cd6
5cc4737959c40b1677d319c083a240325ed6d048
46786 F20110115_AACVOV shah_v_Page_39.pro
f24745f892f267a45c693c86860a17c6
cfd47c45b6492ed35d2377957d0ae07eca83b22a
2099 F20110115_AACVYP shah_v_Page_14.txt
4b8b315a377b0e1c0f9152d630e6c777
817946cc21ea613b58f934d42dff50db3327c45c
103680 F20110115_AACVTS shah_v_Page_16.jp2
fdaad98b92f605ee3456565a96840598
fa5617263f1420fc173b7a1b0cd1da1928ce8599
25318 F20110115_AACWAR shah_v_Page_13thm.jpg
6267012969d6d7efae7b23151bf08f83
7468b0e86fb3d5f05c3d8a0c8d4342b2297d1a65
48447 F20110115_AACVOW shah_v_Page_19.pro
46bfe6edae701f2cc45976a29f344998
63ea0da7ec8e25fe5d8ac9b495c6178f1bc9f508
1910 F20110115_AACVYQ shah_v_Page_16.txt
90ad222667991ce538415e87057bb0df
e9542afa10b3eeb29a4910a36ac02c623b1f89a9
95070 F20110115_AACVTT shah_v_Page_18.jp2
4bdf2438a0cfe0ac942d057b2ef1fe3e
0808b6e5ad906f0fdc464f0bb5d77340330ac5f7
102120 F20110115_AACWAS shah_v_Page_56.QC.jpg
e2ce74b545ae83cc1a07661fac1cf86b
346d088d7c170cf4bf818bac49581711e0857396
1539 F20110115_AACVOX shah_v_Page_30.txt
2be4a000484f4fdce3de611992f4b6b7
ede7c4cb17adecb96b38864ca3dd859332d09b4d
536 F20110115_AACVYR shah_v_Page_17.txt
89932b2f1195a45f4b25f17f381f5f49
edffadee33240c10075a1728e35c39ad643b2be5
106395 F20110115_AACVTU shah_v_Page_19.jp2
db47749182b6379ab906329263961b81
21ab76b62bc225e0248b33e8c42c38ec9af494b7
46692 F20110115_AACWAT shah_v_Page_64.QC.jpg
b79983c28e790116b211a5e9a78560b8
4b6c4c488394479e95c7fac87c5db00117b07d3c
20378 F20110115_AACVMA shah_v_Page_68thm.jpg
25a24114d2643a28f2d6e4976e6b1a8b
32296577d28e8c4f606318b52b795132c3c3ae51
24217 F20110115_AACVOY shah_v_Page_28thm.jpg
fea1973412850b4f7a239c5fb4e9a879
8a3502c330a9605168c30e42d52f1e9fc6db0119
1930 F20110115_AACVYS shah_v_Page_19.txt
11e4a3c251ed9190aba8f45c489a458c
7719aa226c037bed95895fe51c31606abce2011e
106214 F20110115_AACVTV shah_v_Page_21.jp2
32e3c25c8001b6765b281b2f5d4caecb
cac63e16f7528a9db9442afbd5327b5da40e6a02
22585 F20110115_AACWAU shah_v_Page_38thm.jpg
dca07f6b0665060bbb43de9952cb07f1
d8e62cec5effacaf76bd1c4e894b2a349473813c
45588 F20110115_AACVOZ shah_v_Page_59thm.jpg
88e847058f2f97847f2c4350d7fcdb3f
7ebed581bcb241f1aca553fc8ed26488862bc275
1956 F20110115_AACVYT shah_v_Page_20.txt
e3ea53bac3452158da6f47a4ef7f1942
90100c410e45c3ef738c893e4c53cbe8cf43fca1
111067 F20110115_AACVTW shah_v_Page_22.jp2
a077e97dc2f5e04c4ab8fc22b0067a6f
29a7914553cd7daada2c039b013cab64f53f938e
F20110115_AACVMB shah_v_Page_29.tif
4f8860017e0f01abfcc6c8fe97b3a06b
600b40c06925e68af21aaa7d79d713f7d19a1431
46980 F20110115_AACWAV shah_v_Page_52thm.jpg
04f5ee205f544a365f4a58b213f1daa9
87a1d8f883c14e2b8a36cbed2e4ef70001d8e20e
2035 F20110115_AACVYU shah_v_Page_25.txt
d48cd29861c02564dce1ad085cbd843d
85b94a1dadb5786bf2c2dc108e7afc86fa170fbe
104045 F20110115_AACVTX shah_v_Page_23.jp2
35759216575e6fcedaf6f2f4679c57ac
75fdb4794a8f44bc70149ed601305cb450d4c978
1428 F20110115_AACVMC shah_v_Page_64.txt
9f029e19cde5f40410aa5cbb2fbfde1c
54269b4709d208e1b3dfabc2b2ef7693cac33f44
2108 F20110115_AACVYV shah_v_Page_26.txt
7bc4935f2664098b09ea092bf71ec06e
b9da31302856109776baa75ac92d7ef5321a978b
1051964 F20110115_AACVRA shah_v_Page_41.jp2
c9ea53f5bf981d403431387b5fa8bd30
4d9a95bf965b23ba47f34009672711a630bec14c
113225 F20110115_AACVTY shah_v_Page_25.jp2
4262ae09b9ee5a94b82bff161e52bf07
0d1169cccfacbd468ee89b8fbb547c34f57a6bd7
12840 F20110115_AACWAW shah_v_Page_03.QC.jpg
366859bc14e5f5c353225e809f308398
5c907e479e58b93e035e3650da962fd96fa1ecf3
2077 F20110115_AACVYW shah_v_Page_27.txt
935bc12cd36f80780e8966e44f5dcf5d
72bae8bd25e59c663ade099023aab09a3aad8425
247965 F20110115_AACVRB shah_v_Page_70.jpg
502a0c5fd49ddb148a7b477417af3f5b
ab9759390e95b47aa908fd4ab3774b3547a44f92
114474 F20110115_AACVTZ shah_v_Page_27.jp2
d9f61ecbdfb11824ad15eb0ebac8b9b5
cc8df24bd4d8bf4dc470133c8ca4cdfb9f9420df
F20110115_AACVMD shah_v_Page_23.tif
b6dd7f3a1abfddd2842e92b989a94bd5
4f14e9408645a59c093642e690378872c2655fe4
25026 F20110115_AACWAX shah_v_Page_25thm.jpg
1cc2b585d0766b53dcf43f172b5992b8
5bea0f1bd9b6f515c520c10cd3861aaf4a3fabd4
1882 F20110115_AACVYX shah_v_Page_28.txt
c170a9ed5040c6412da0a0b6ff5b3b16
eb32f26913c010a4b47f744b11340de27af99607
201950 F20110115_AACVRC shah_v_Page_40.jpg
edc7cd8c068b6290d867e274edc74189
f45b912ba34d86dc70edbf10b89f3fc5b56dc43c
20948 F20110115_AACVME shah_v_Page_48.pro
fdaadc0653b2f91315a231feee0c83e8
3e18fe7c9c4deaefab75e1e1b1f0515d57c2050e
23874 F20110115_AACWAY shah_v_Page_32thm.jpg
c384a7ae363c85eef9a8c71e35779e15
6b1f84028d476334ebcada955b245aee5e5035c2
F20110115_AACVWA shah_v_Page_50.tif
f68de50722cac51277a2e0788eb234e2
7f179a319c34e48fe97a7eed9799d65f5300eda2
1885 F20110115_AACVYY shah_v_Page_29.txt
1cd0b2130e9ac6613e6788898fffbaa6
36346e8377e642073b5371219a19bcc31ae184d5
111140 F20110115_AACVRD shah_v_Page_24.jp2
da58bb33b0c9a373167bbb9b6c574ba9
1f096e1191cec1261eaace7bf24ee45f3c9dd3ba
82021 F20110115_AACVMF shah_v_Page_24.QC.jpg
b1b253fa46e1839494b76276bb311874
9377566cb0d514fb1625ee577023fe2defc0d78f
21868 F20110115_AACWAZ shah_v_Page_66thm.jpg
676aaed58e544c09036a12bafdaa57c2
685064eead179c739ced94bee04aeec2c2b5ea00
F20110115_AACVWB shah_v_Page_52.tif
e8071b65b9dd17f8d8b7d717e1923f1c
070b48036b183fdb6099ebcf947cf96ac4ad7786
1905 F20110115_AACVYZ shah_v_Page_31.txt
ac68ee9f3ae929aafea7d92318fd4cc3
d3ba9b8f83b246ac78544c4f7d746f2ae3272dbb
105522 F20110115_AACVRE shah_v_Page_28.jp2
d23a8c06266d506fcab50440621c88e8
3c3245c94de283e254f2b38da313d813530aa54d
2005 F20110115_AACVMG shah_v_Page_24.txt
79a473ffc31b0b3554c0610b3f798ba8
b92ab69bdf0b8022d11948bcdce49cc2219aefd8
F20110115_AACVWC shah_v_Page_53.tif
eb4b3d8babf2b238ed20394f3dab2cf5
bdca83277b31b360987e9a5eb46e3e9d462d430f
69853 F20110115_AACVRF shah_v_Page_05.pro
c1ffc696023cc7aa6ab5cfb5856ed951
5b39b556fbdbc641034b7e0512afeb21d9b6650a
2479 F20110115_AACVMH shah_v_Page_50.txt
9b9ca9e7a3ca5f7ce9f345fadb441308
0484fd4109dc243963115373eb27167e6347fa93
46918 F20110115_AACWDA shah_v_Page_41thm.jpg
8fe73beffa3b68359cd5ec8258052aaf
0a91313c50f816191129512e8ebda853dbcd2e07
F20110115_AACVWD shah_v_Page_54.tif
30ef9780c620e887830cb0552f50d1eb
797eb535448d54aa1ef3a2308d46874f135a3e61
23280 F20110115_AACVRG shah_v_Page_57.pro
873d6c4ec23e8ad61fae8244b1707aa9
1113241cfbc6d03b134db5a5da4812e5e596a870
F20110115_AACVMI shah_v_Page_08.tif
10eed89e626c5752f8281631f9ecbe6d
b1f6d694e3e0d8a28a4e309ba366898d20702213
94435 F20110115_AACWDB shah_v_Page_41.QC.jpg
39207117824f1e917264a3e5ad296da5
dc846b0ecd45ce0beeb95ff30265c529f73f7d04
F20110115_AACVWE shah_v_Page_56.tif
64fac00a70b404b94ce2581824239308
801ce971aa10efb05d6043d6ce11091e2f91ec45
85116 F20110115_AACVRH UFE0008824_00001.mets
90a461bafbacc6853835106b84bb260b
4e88f8a35c1a708bc4e1ac8c3bc1db3a10c5e467
48054 F20110115_AACVMJ shah_v_Page_42thm.jpg
6a61789e2b248544d7fdc8455e6e6f20
679d81c9e7d7d55e8e2b95bf6b1fde42b2998396
F20110115_AACWDC shah_v_Page_47thm.jpg
380bc16472228d93b22296154f91584b
417f2a25ad7932db0ca64f718008683950a47dc9
F20110115_AACVWF shah_v_Page_60.tif
9b514c0a3e1c00020cba5bc1dfea680c
2c6d79e0eb87323a32ca42e1119e0cd26f142d8c
1797 F20110115_AACVMK shah_v_Page_18.txt
a70f59a3c3888597b8960c87163aba20
2803a5b5a1e57c981235683febb896a934a0d3d7
48459 F20110115_AACWDD shah_v_Page_48thm.jpg
899a27081d2a535a360fe6ab30a3b70c
30a388acdf8f3a9336b9e2d129e63c9d340a0d60
F20110115_AACVWG shah_v_Page_62.tif
14a61d1fe15253a8d710258ea20fa085
2c90252370e8fbdcb8ede8882771f902ca54050e
1221 F20110115_AACVML shah_v_Page_40.txt
2acb0a0c1e278bb97061f88f36ca9b27
9d86070bff713acbd19b83ddd7c744b2a29afd02
103261 F20110115_AACWDE shah_v_Page_48.QC.jpg
e6feb90cede6aa738c3e510dd99cff59
d5dcb0fd7e04db4b238ae51fa175a6fe65c50205
F20110115_AACVWH shah_v_Page_63.tif
eea9259af53a2d25f4160bd8f622906f
ee950a95af05c8d917f5e275d10bcbb05e694962
F20110115_AACVMM shah_v_Page_09.tif
5d708bfe2cca0ab0b2fb681ae9650dac
363813e3d5ecec23abac1b3047e6cc6a744232aa
49330 F20110115_AACWDF shah_v_Page_49thm.jpg
0313ba3f63c202448542aff86a161396
fb15e619fc8afa8018bdbaabb26469770326cccd
F20110115_AACVWI shah_v_Page_64.tif
43f2d59e2151716567d873e55bff0b77
fd8ed9aa0e557ce51533d1df617faa7f88625aa9
58686 F20110115_AACVRK shah_v_Page_01.jpg
919646e46f772c11df51a59e92236d5c
87858295649aa9a64714681cbd06fa095ab643e3
51989 F20110115_AACVMN shah_v_Page_25.pro
665de5f612edbaa616dcce3deef00785
400c66af68a5dbab10c609b03f531214a61ddd80



PAGE 1

DETECTION OF MICROCRACKS IN CONCRETE CURED AT ELEVATED TEMPERATURE By VISHAL SANJAY SHAH A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE IN BUILDING CONSTRUCTION UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 By Vishal Sanjay Shah

PAGE 3

The thesis is dedicated to my loving parent s Sanjay and Rekha and to my best friends Reshma, Akshay, Piyush and Anushree fo r their unconditional support and caring throughout my academic endeavors. It is with the support and love of my family and friends that I am able to reach to my goals.

PAGE 4

ACKNOWLEDGMENTS The author would like to thank all of the members of his supervisory committee for their help and ideas throughout this effort especially Dr. Abdol Chini, committee chair, for his advice, patience and dedication as my advisor and also for the financial support provided by him. The author would also like to express gratitude to Dr. Larry Muszynski and Dr. Andrew Boyd for their support and contribution as members of my advisory committee. Acknowledgements are also owed to Tanya Reidhammer for her help in explaining and conducting scanning electron microscopy. In addition, I would like to thank Richard DeLorenzo for his help in the experimental process. My thanks also go to everyone within M.E.Rinker Sr. School of Building Construction for their continued help during the last two years. The author would like to thank his friends at the University of Florida for their support and help they offered during the research and writing of this thesis. iv

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................iv LIST OF TABLES ............................................................................................................vii LIST OF FIGURES .........................................................................................................viii ABSTRACT .........................................................................................................................x CHAPTER 1 INTRODUCTION............................................................................................................1 Background...................................................................................................................1 Curing of Concrete.......................................................................................................2 Heat Generation in Mass Concrete...............................................................................3 Objective and Scope.....................................................................................................5 2 LITERATURE REVIEW.................................................................................................7 Overview of Sample Preparation Techniques..............................................................8 Radiography and Acoustic Techniques.................................................................9 Replica Technique...............................................................................................10 Impregnation Techniques....................................................................................10 Dye Impregnation Methods..........................................................................11 Epoxy Impregnation Techniques.................................................................11 Thin Samples................................................................................................13 Impregnation by Woods Metal...................................................................14 High Pressure Epoxy Impregnation Technique...........................................15 Comparison of Sample Preparation Techniques.........................................................16 Microscopic Instruments............................................................................................18 Optical Microscope......................................................................................18 Scanning Electron Microscope.....................................................................19 Variable Pressure Scanning Electron Microscope.......................................21 Image Processing Techniques.....................................................................................23 Summary.....................................................................................................................24 v

PAGE 6

3. RESEARCH METHODOLOGY...................................................................................25 Introduction.................................................................................................................25 Sample Selection........................................................................................................25 Specimen Preparation Method....................................................................................27 Procedure for Sample Preparation..............................................................................28 Preparation for Image Processing...............................................................................32 Image Analysis...........................................................................................................33 Summary.....................................................................................................................34 4. IMAGE PROCESSING AND ANALYSIS..................................................................35 Introduction.................................................................................................................35 Image Processing Technique......................................................................................36 Sample 95B00P Room Temperature Sample 1................................................37 Sample 95B00P Room Temperature Sample 2................................................41 Sample 95B00P High Temperature Sample 1..................................................45 Sample 95B00P High Temperature Sample 2..................................................49 Comparative Analysis of Samples..............................................................................53 Summary.....................................................................................................................54 5 SUMMARY, RESULTS AND RECOMMENDATIONS.............................................55 Summary.....................................................................................................................55 Results and Conclusion...............................................................................................57 Recommendations.......................................................................................................57 LIST OF REFERENCES...................................................................................................59 BIOGRAPHICAL SKETCH.............................................................................................62 vi

PAGE 7

LIST OF TABLES Table page 3-1 Mix Design Data..........................................................................................................26 3-2 Adiabatic Temperature Rise Data................................................................................27 3-3 Mixing Schedule for Epoxy.........................................................................................30 4-1 Crack Quantification Analysis for Mix 95 B 00 P RT-1.............................................39 4-2 Comparison of Odd and Even Number Images...........................................................40 4-3 Crack Quantification Analysis for Mix 95 B 00 P RT-2.............................................43 4-4 Comparison of Odd and Even Number Images...........................................................44 4-5 Crack Quantification Analysis for Mix 95 B 00 P HT-1.............................................47 4-6 Comparison of Odd and Even Number Images...........................................................48 4-7 Crack Quantification Analysis for Mix 95 B 00 P HT-2.............................................51 4-8 Comparison of Odd and Even Number Images...........................................................52 4-9 Comparative Analysis of 95 B 00P RT1 and 95B00P RT2........................................53 4-10 Comparative Analysis of 95 B 00P HT1 and 95B00P HT2......................................53 5-1 Comparative Analysis of Samples...............................................................................57 vii

PAGE 8

LIST OF FIGURES Figure page 2-1 Working of Scanning Electron Microscope................................................................19 3-1 Diamond Wafer Saw...................................................................................................29 3-2 Ultra Low Epoxy Kit by SPI.......................................................................................30 3-3 Metaserv Grinder and Polisher....................................................................................31 3-4 Polished Sample..........................................................................................................31 3-5 Diamond Suspension Paste..........................................................................................32 3-6 Prepared Samples........................................................................................................32 3-7 Specimen with Copper Tape........................................................................................33 3-8 Specimens with 4x4 matrixes......................................................................................33 4-1 Representative Images of Mix 95 B 00 P RT-1 before Image Processing..................37 4-2 Representative Images of Mix 95 B 00 P RT-1 after Image Processing.....................38 4-3 Variations in Crack Density of 95 B 00 P RT1...........................................................40 4-4 Representative Images of Mix 95 B 00 P RT-2 before Image Processing..................41 4-5 Representative Images of Mix 95 B 00 P RT-2 after Image Processing.....................42 4-6 Variations in Crack Density of Sample 95 B 00 P RT2...........................................44 4-7 Representative Images of Mix 95 B 00 P HT-1 before Image Processing..................45 4-8 Representative Images of Mix 95 B 00 P HT-1 after Image Processing.....................46 4-9 Variations in Crack Density of Sample 95 B 00P HT1.............................................48 4-10 Representative Images of Mix 95 B 00 P HT-2 before Image Processing................49 4-11 Representative Images of Mix 95 B 00 P HT-2 after Image Processing...................50 viii

PAGE 9

4-12 Variations in Crack Density of Sample HT 2............................................................52 ix

PAGE 10

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for Master of Science in Building Construction DETECTION OF MICROCRACKS IN CONCRETE CURED AT ELEVATED TEMPERATURE By Vishal Sanjay Shah December 2004 Chair: Abdol Chini Cochair: Larry Muszynski Major Department: Building Construction Proper curing is a necessity for durable concrete. Exothermic reaction during curing process generates heat in concrete creating thermal differential between outer and inner core of concrete. In mass concrete pores at Florida Department of Transportation, internal temperatures of concrete have been reported as much as 180F to 200F, which leads to a concern that higher curing temperature might initiate higher number of microcracks during its early age. Phase I of this study indicated the presence of higher curing temperature in concrete which is made from cement having higher heat of hydration, higher placing temperature and does not contain any pozzlonic material in mix design. For comparison of curing conditions, four samples were selected from this type of concrete for quantification of microcracks. Out of these four samples, two of the samples were cured at room temperature and the other two samples were cured at high temperature. x

PAGE 11

A literature review revealed that various specimen preparation methods have been used in past for viewing concrete microstructure. However, a counter diffusion method with epoxy impregnation was used in this study as this method does not require any pre heating of specimen and pressure or vacuum in sample preparation. After specimen preparation, samples were viewed under variable pressure scanning electron microscope, and a total sixteen images from each sample were taken. Image analysis was done with standard image processing software by converting images into binary mode by thresholding it until features of interest (microcracks and voids) were correctly highlighted. After differentiating voids from cracks by shape analysis method, microcracks were manually mapped and quantified by calculating crack density in each image. A statistical analysis was performed to indicate representative nature of image quantity taken in sample. The results showed no significant difference between the number of microcracks present in samples cured at room temperature and high temperature in concrete made by cement having heat of hydration up to 76.2 cal/g, placing temperature of 95F and without any pozzlonic material. Thus it could be suggested that high temperature curing upto 180F does not play any significant role in the development of microcracks. xi

PAGE 12

CHAPTER 1 INTRODUCTION Background Concrete is one of the oldest engineering material and certainly one of the key integrants in development of modern infrastructure. When properly designed and carefully produced with good quality control, concrete is inherently a durable material. However, under adverse environment and exposure conditions, concrete interacts with its surrounding environment and undergoes chemical changes. These processes involve movement of water or other gaseous substances transporting aggressive agents through pore structures of concrete, thus making concrete potentially vulnerable to deleterious attacks. Reinforced cement concrete structures exposed to hostile environment require highly durable concrete to provide long lasting performance with minimal maintenance. Modern day infrastructure has necessitated construction of numerous bridges, dams, tunnels and concrete pavements. Massive concrete structures of this nature would contain large amount of concrete, most of these structures could be termed as mass concrete work. According to American Concrete Institute Committee 116, report SP-19 (1985), mass concrete is any volume of concrete with dimensions large enough to require that measures be taken to cope with generation of heat of hydration of the cement and attendant volume change to minimize cracking. The Florida Department of Transportation (2004) further specified dimensions of mass concrete as when the 1

PAGE 13

2 minimum dimension of the concrete exceeds 36-inches and the ratio of volume of concrete to the surface area is greater than 12-inches, provide for mass concrete. Curing of Concrete Curing is widely perceived as being an important factor in achieving durable concrete structure. Curing ensures that concrete maintains a sufficient amount of water to reach a proper degree of hydration in the surface which is required to continue chemical reaction between cement and its constituents. Proper Curing allows hydration of cement to continue which is expected to reduce capillary porosity, thereby strengthening the concrete to desired strength and increasing its resistance to penetration by aggressive gases and liquids. The higher the curing temperature, the faster is the reaction between cement and water, thus shorter setting time. It can reduce durability and useful life span of structure by hardening concrete faster, reducing strength and causing excessive seepage. Curing measures are necessary to maintain a satisfactory moisture and temperature condition in concrete, because internal temperature and moisture directly influence early and ultimate concrete properties. During curing process, concrete element undergoes exothermic chemical reaction. This heat generation is of little consequences and can be easily dissipated in surrounding environments. However, in mass concrete, heat generated within the body of member does not get dissipated easily. This temperature rise causes expansion while the concrete is hardening. Temperature in inner core of concrete is much greater than outer core of concrete due to loss of heat to surroundings by exposed surfaces. If the temperature rise is significantly high and concrete undergoes non-uniform or rapid cooling, stresses due to thermal contraction in conjunction with structural restraint can result in cracking before or after the concrete eventually cools to surrounding temperature. Due to faster cooling of

PAGE 14

3 outer surface, concrete tends to shrink and causes tensile stress in outer core. When these tensile stresses are greater than tensile strength of concrete, cracks are developed. Heat Generation in Mass Concrete Heat generated in mass concrete increases temperature of concrete to a varying degree depending on surrounding temperature, size of structural element mix design configurations, and curing temperature. Current steps to counter high temperature rise in concrete includes choosing a low heat cement, external curing system, and cooling of aggregates prior to mixing. Constructing large structures in small stages, or lifts, can be done, but it is slow and the finished structure is weakened at the cold joints in the concrete. Replacement of cement by pozzlonic materials results in a decreased temperature rise in fresh concrete, which is a particularly important issue in mass concrete where a large temperature rise can lead to cracking. Currently, any mass concrete pour for FDOT requires thermal control plan which describes the measures and procedures intended to maintain temperature differential of 35F (20C) or less between interior and exterior portion of designated mass concrete element during its curing phase; there is however, no requirement on the maximum temperature rise within the mass concrete. Internal temperatures of 180F to 200F have been reported in mass concrete pours in FDOT projects. There is a concern that such higher curing temperature would initiate microcracks during its early age which would accelerate deterioration and compromise durability of concrete by microcracks acting as a conduit of harmful minerals and gases, causing reduction in durability of concrete. Study conducted by Maggenti (2001) of California Department of Transportation indicated no irregular, externally visible cracks in mass concrete work. However, it would be important to find out changes occurring at the micro level, which will obviously

PAGE 15

4 grow in future and hamper durability of concrete. The capillary effect of the microcracks increases probability of exchange of substances between the environments and microstructure of concrete. This exchange of substances may lead to transport, solution and enrichment processes in concrete microstructure which can damage the concrete by increasing the volume. Concrete, when fully hardened, contains a variety of micro structural features on a variety of scales. To the unaided eye, it appears from a polished section to be essentially a two phase composite-a matrix of hardened mortar plus coarse aggregate that occupies about 75 % of its volume. Recent studies of concrete microstructure have played important part in understanding behavior of concrete matrix. It is necessary to study the effect of elevated temperature rise in microstructure specially comparing it with normal room temperature cured samples and find out if any additional microcracks have been developed. Formation of microcracks in the surface can lead to decrease of durability, especially in exposed outdoor structures as bridge superstructures but also concrete pavement and structures in direct contact with the ground. Many methods ranging from destructive to non-destructive have been used in past to understand microstructure of concrete. A problem common with many of the techniques used to examine the microstructure of concrete is that they can alter the sample. For example, many techniques require that the specimen be prepared by mechanical polishing or to dry concrete specimen before examination by putting in vacuum or heat. Such Specimen preparation methods would invariably alter the original artifacts and induce additional microcracks.

PAGE 16

5 Advancement in science has assisted in developing equipments and techniques capable of studying the microstructure which facilitate in-depth study and analysis of specimen. While many existing techniques and equipments require drying the specimen and in a way altering the specimen, evolution of Variable Pressure Scanning Electron Microscope (VPSEM) techniques allows the study of unaltered wet specimen in a low vacuum. Objective and Scope This study is an extension of a research project Adiabatic Temperature Rise in Mass Concrete which is aimed to develop adiabatic temperature rise curves for concrete mixes used in FDOT projects and investigate properties of high temperature cured concrete. Total of 20 mixes of concrete were made as a part of the study by changing various parameters such as the source of cement, the percentage of pozzlonic materials and the initial placing temperature. Concrete samples were kept in heating curing chambers to simulate conditions of mass concrete for monitoring rise in temperature. At 14 and 28 days, samples kept in heating chambers as well as samples in normal room temperature were tested for compressive strength. The objective of this part of the study is to focus on detecting and quantifying additional microcracks developed due to concrete cured at elevated temperature. This would be done by comparing room temperature cured samples with samples cured at elevated temperature to find out if there is any difference in quantity of microcracks. The Study would be performed as given below: 1. A literature search to find out current methods used in identification of microcracks and sample preparation techniques. It should also include possible image processing techniques to enhance analysis capability.

PAGE 17

6 2. Prepare concrete specimens according to an identified sample preparation technique and view them with selected Microscopy equipment. 3. Use Image Processing Techniques for further identification and quantification of microcracks. Compare the quantity of microcracks in samples cured at elevated temperature with those cured at room temperature to determine if any additional microcracks have been developed due to high temperature curing.

PAGE 18

CHAPTER 2 LITERATURE REVIEW Presence of voids, cracks, and other defects play an important role in determining the mechanical performance of the concrete. If concrete surface is pre-damaged by microcrack formation in the near-surface area and exposed to open weathering, capillary effect would increase the transfer of substances from the environment and increase the concrete volume which can lead to several mechanisms capable of negatively influencing durability of concrete. Studies have indicated that the development of cracks and connected crack network contributes to the increase in permeability and diffusivity of concrete. The presence of preexisting microcracks of 10 m or greater on the surface can lead to reduced durability, especially of exposed outdoor structures such as bridge superstructures, but also concrete pavements and structures in direct contact with the ground. Patel et al., (1995) found that concretes cured at higher temperatures exhibited a coarser microstructure than that of a typical concrete cured at 20C, particularly with respect to ettringite. The presence of microcracks will increase moisture mobility within the concrete, which may produce density gradients within the matrix leading to further microcracking. The domination of the matrix by a network of microcracks is also conducive to the formation of secondary ettringite. Recorded curing temperature of 180F 200F inside core of FDOT Mass concrete elements has raised concerns over the initiation of microcracks. Slate and Hover (1984) defined microcracks as cracks having width of less than 100 m, while Jansen defined 7

PAGE 19

8 microcracks as extended faults with a width of less than 10 m. According to Kjellsen and Jennings (1996), these differences in definition appeared to be closely related to experimental techniques and to the orientation of the studies. For the scope of this study, reasonable and practical interpretations of microcracks were set as cracks having widths less then 10 m. While macrostructure of concrete can be seen unaided, microstructure (200 m or smaller) must be observed with the aid of a microscope. It would be fair to say, as the documented research indicates, that cracks and crack propagation is understood well from macroscopic point of view but unavailability of precise specimen preparation techniques hinder studies of concrete microstructure. Overview of Sample Preparation Techniques Sample preparation is a key to microscopic analysis of concrete. Proper preparation of concrete samples so that microcracks and voids develop a distinct contrast against the body of concrete is a pre-requisite for the application of the modern day image processing and analyzing techniques (Soroushian et al., 2003). Poor preparation methods can lead to erroneous diagnoses of problems associated with concrete specimen. According to Hornain et al., (1996), ideal specimen preparation techniques should not induce any cracks during the preparation of samples, thus techniques which involves prior drying of specimen should not be used. The ideal specimen preparation should be simple, economic, rapid, and should be able to detect very fine cracks. And finally, a sample preparation technique will not be useful without accurate image analysis and processing techniques. The sample preparation would also depend on the type of equipment used for image collection, required resolution of image, and objective of the study.

PAGE 20

9 For the past few decades, studies have been performed to find out a method which fulfills all definitions for ideal specimen preparation technique. New equipments and methods developed by technological advances have helped researchers in their analysis of concrete specimens. As reported by Ringot and Bascoul (2001), two methods are necessary for the characterization of microcracks, one for sample preparation and one for quantification of cracks. Sample preparation techniques also depend on the type of microscopy used (S. Marusin 1995). Acoustic method, ultra sonic and laser sparkle methods are useful for studies related to crack propagations but they are unequipped to monitor initial state of samples crack quantification. Thus, the rest of the sample preparation techniques can be briefly classified in three main categories: radiography techniques, replica techniques, and impregnation techniques. Radiography and Acoustic Techniques Earlier Methods by Slate and Olsefski (1963) described the use of X-radiography to study the internal features of concrete and crack formation process of mortar and concrete sample. Thin samples of 4 mm thickness were exposed to X-rays flux normal to the plane of the sample. Cracks were identified from normal constituent of concrete as penetration of X-rays was greater in surrounding area. A comparative study carried out by Najjar et al., (1986) indicated that due to poor resolution of images, X-radiography systematically overlooks thin cracks. Neutron radiography described by Samaha and Hover, (1992), which is principally quite similar to X-ray radiography approach, is in fact a better alternative for increasing image resolution. However, in this process the samples need to be air dried beforehand in order to impregnate specimen with gadolinium nitrate, which in turn might induce microcracks in samples.

PAGE 21

10 Replica Technique Replica method developed by Ollivier in 1985 is one of the methods which do not require any pre drying of the sample. In this method, the film of acetylcellulose is placed on concrete specimen by methyl acetate. The film is taken off after the solvent (methyl acetate) is evaporated in air and observed under optical or scanning electron microscope. As there is no disturbance made to the specimen surface itself, crack propagation studies can be carried out with replica method. However, in order to peel the acetylcellulose film safely, the area of film has to be limited to 2 cm 2 Thus too many replicas need to be prepared for covering significant portion of specimen, which could be time consuming. Impregnation Techniques Impregnation techniques are one of the oldest techniques used to study microstructure of concrete. In this method, concrete samples are impregnated with dye or epoxy to facilitate detection and identification of cracks. In order to reduce difficulty in viewing cracks in dense microstructure, fluorescent dye was used by Knab et al., (1984). However, pre drying of specimen in an electric oven before impregnation, which could alter the specimen condition, remained a major inhibition to this process. Methods developed by Struble et al., (1989) and Gran (1995) overcome drying of specimen by using counter diffusion method for replacement of pore water by dye impregnated organic solution. In this method, a thin sample of 15 mm was cut and kept in solution made by dissolved dye and ethanol. After 4 days of counter diffusion process between pore water present in concrete specimen and dye induced ethanol, specimen was taken out of the solution to remove excess dye present on the surface by polishing. However, this process could take several days and increases the duration of the test.

PAGE 22

11 Dye Impregnation Methods Hornain et al., (1996) modified dye impregnation techniques to reduce the time taken to prepare specimen. In this Method, 4 grams of a water soluble red powder, commercially known as Irgacete was dissolved in 100 ml ethanol solution. After keeping specimens for 5 minutes in dyed solution, a second impregnation was done. After completing these 2 steps, the specimen was taken out and polished under water with 6 m diamond paste to remove excess dye. In order to achieve proper polishing, the specimen was again polished with 3m and 1 m of diamond paste. The polished specimens were observed under optical microscope at 100X. The study reported that cracks of 1 m width and less could be easily distinguishable, providing contrast level between specimen and cracks. During sample microscopic studies, cracks going through the hydrated cement phase were distinguished from discontinuity at paste-aggregate interface. However, the author reported that dye impregnation method was not useful in finding out cracks in highly porous areas. Epoxy Impregnation Techniques Struble and Stutzman (1989) developed a new technique which involved three steps procedure to replace pore water from concrete sample. Epoxy impregnation would not only support the microstructure of specimen by filling the voids and cracks up on curing, but also support fragile pores and matrix phases by restraining it against disintegration during the different stages of preparation. This could be a major issue if the samples are to be viewed under scanning electron microscope which would generate high pressure inside the chamber. Another advantage of using epoxy is to enhance contrast between the pores, hydration products and cementitious material. The selection of the type of epoxy depends on the objective of the study. Low viscous epoxy was used for relatively highly

PAGE 23

12 permeable materials or cementitious powders, while ultra low viscosity epoxy aids in rapid infiltration for less permeable cement pastes and concrete samples. The first step for this sample preparation method was to cut the specimen of 1 inch size and immerse specimen in ethanol solution at 60C. The blades of saw were immersed in propylene glycol to prevent it from drying. A companion specimen was prepared by Struble and Stutzman (1989) which was kept in an ethanol solution with red dye to see the extent of time taken for the replacement of ethanol in pore structure. For a given concrete sample, the time taken for replacement depends essentially on the thickness of the sample. After 50% of alcohol replacement in depth by visual inspection in companion specimen, the solution is replaced by epoxy at room temperature and cured according to the manufactures specifications. After proper curing, the samples were polished and lapped with abrasion papers and diamond paste in decreasing order and viewed under Backscattered Electron imaging technique of Scanning Electron Microscope. The samples should be grinded and polished perfectly in order to remove raw saw cuts on the specimen. Insufficient polishing would leave a disturbed surface on specimen and the cracks would not be easily discernible during the microscopic observations. On the other hand, an excessive polishing would contribute to the removal of microstructure, and could affect the subsequent quantification operations. It was believed that higher heating temperature would hasten the process of pore replacement. Higher temperature for heating was not used in subsequent studies due to limited measure of quantifying the pore water alcohol exchange and quick evaporation of alcohol. Modified epoxy

PAGE 24

13 impregnation method does not require any heating or vacuum on samples and would not induce any specimen preparation related microcracks. However as reported by Soroushian (2003), epoxy impregnation can highlight more porous areas of cement paste, but generally do not yield crisp boundaries and sharp contrast between microcracks and air voids against body of concrete. A new two stage technique was developed as InkEpoxy Impregnation. In the first stage, fine capillary pores were first impregnated with parker blue ink and then microcracks and voids were impregnated with fluorescent ultra low viscosity epoxy solution. Specimen were cut to 20-30 mm or 0.8-1 inch, and lapped by abrasive liquid under 3 PSI pressure After cleaning the specimen for loose debris by first water bath and then ultrasonic bath, specimen was dried at 140 F for 3-4 hours. Ink impregnation for first stage was done by keeping the container with ink and specimen under nitrogen pressure at 280 psi for 18-24 hours. Specimen was removed and heated at 140 F for 24 hours to let ink dry. Stage 2 was performed for fluorescent epoxy penetration in samples. The samples were kept in a vacuum chamber with 0.38 psi for 1 hour. Epoxy solution was mixed and poured in a glass container and kept in nitrogen pressure of 0.0133 psi for 3-4 hours. To cure epoxy, samples underwent heat in an electric oven at 150 F for 18-24 hours following by polishing operations. While comparing with normal epoxy replacement technique, ink-epoxy replacement technique gives better results. However, the application of heat as well as vacuum pressure could instigate microcracks in samples. Thin Samples A study conducted by Anderson (1989) introduced preparing thin section of concrete samples. In this method, a larger sample of 45 mm x 30 mm x 20 mm was cut from a section of concrete, selected from the portion which was few centimeter away

PAGE 25

14 from to be investigated area. The cut section was kept under ethanol for 12 hours to reduce its tendency for cracking. The concrete specimen was kept in a vacuum oven at 85 F to 90 F to dry for two hours. A homogenized mixture of low viscosity epoxy prepared with 1.1% by volume of dye was mixed in magnetic stirrer for 24 hours in advance. Dried concrete specimen was kept in the solution for 1 hour for vacuum impregnation. It was observed that concrete was impregnated by 1-2 mm in depth. Stage 2 covered mounting of glass slide on the selected face of specimen with Ultraviolet (UV) hardening glue. The specimen was then grinded by a 15-20 m diamond disc in order to get smooth surface. After cleaning and drying the specimen, it was re-impregnated with epoxy. After curing, a thin sample was cut from the specimen by diamond saw and final grinding and polishing was done. Although the results from this method are widely used in European counties, lengthy sample preparation time and dependability on skilled and experienced technician were primary requirements. Impregnation by Woods Metal Nemati (1997) developed a new approach in impregnation techniques using Woods metal instead of epoxy. The idea of this research was to preserve the microstructure of the concrete samples which are kept under compressive stress to analyze microcracks as they exist under loading. In this three-phased study, cracks were introduced in the concrete specimen by a new test setup which allowed the application of axial stress on concrete and impregnation by woods metal simultaneously. Woods metal was impregnated within the sample for providing stability and better contrast for the identification of microcracks. However, sample preparation method involved drying of concrete cylinder by 109.5F as well as gradual heating of test assembly by 122F, 167F and 204.6F for unified initiation of molten woods metal in the specimen. As explained

PAGE 26

15 earlier, heating or drying operations before impregnation could introduce microcracks in concrete. Author reported that in no loading conditions, few cracks were observed which could be present due to cracks introduced in sample preparation or drying shrinkage. Woods Metal impregnation technique was further developed by Soroshian et al., (2003) for concrete samples without initiating microcracks through stress introduction. In this method, 2 inch thick slices were cut and washed to remove any loose debris attached. The cleaned sample was kept in electrical oven at 150 F for 24 hours to remove water present. The dried sample was than kept in a steel mold for impregnation with Woods metal. To facilitate impregnation and liquidification of Woods metal, steel mold is kept inside a vacuum pressure chamber by 0.95 psi for 3040 minutes. After keeping the oven temperature for 200 F for 1-2 hour, the air vacuum was replaced by nitrogen pressure of 280-300 psi for 3-4 hours. The specimen was allowed to cool down, followed by cutting a 6 mm sample and then the surface was prepared for viewing under Backscattered technique in Environmental Scanning Electron Microscope (ESEM). Results indicated replacement by Woods Metal gave desired contract for microcrack identification against concrete surface. However, heating and vacuum pressure application on concrete would develop additional microcracks which can be quantified with original microcracks. High Pressure Epoxy Impregnation Technique Previous methods involving florescent epoxy replacement being time consuming, Chen (2002) applied high pressure of > 20 bars to florescent epoxy impregnation in concrete sample. Samples of 2x2x1.5 cm were prepared and after initial drying of 176 F for 2 days, samples were kept in a steel cylinder in epoxy resin at least 5 cm higher than the top surface of specimen. The top and bottom of steel cylinder were blocked by Teflon, desired pressure ( 145,725, 1160, 1450, 2900 and 5800 psi) was applied on the

PAGE 27

16 blocks through a hydraulic piston and held under pressure for 90 minutes followed by curing of epoxy by heating it upto 122F for 24 hours. This robust impregnation procedure results in a uniform and vast distribution procedure of the florescent epoxy resin throughout the material structure. Prepared samples were viewed under optical microscope using UV light. The experiment proved high pressure impregnation to be a more effective and faster method to induct epoxy resin in cementitious materials. This method show higher brightness and more effective penetration of epoxy in material and when pressure and vacuum impregnated samples kept in optical microscope to analyze, high pressure impregnated samples took less time for reaching a threshold value of photon. Pressure impregnation allows higher tensile and flexural strength in samples compared to vacuum process, thus implying greater stability of cementitious specimen during the preparation phase. However, no reports clearly show or address the issue of the introduction of cracks during sample impregnation. Comparison of Sample Preparation Techniques Microscopic image is a combination of original material from which the specimen was taken, cumulative effects of all the procedures required to prepare the specimen for examination, the examination technique itself and our interpretation of the image. Specimen preparation method depends on the type of microscope used for viewing the specimen. The review of all above techniques reveals that there is no standard specimen preparation for viewing microcracks in cement and cementitious materials. Almost all researchers identified pre-drying of specimen as a major source of developing cracks prior to microscopic examination. However, most of the studies dried specimens in order to remove pore water of concrete to facilitate the introduction of dye, epoxy or woods metal. While some techniques required polished concrete section or broken fracture

PAGE 28

17 surfaces, some required thin sections dried at 122 F or oven dried in NO 2 atmosphere. Some samples are cut with a diamond impregnated wire saw without a coolant, while some samples prepared by epoxy impregnated are cut by diamond grit or lapped by alumina grit and then polished by diamond grit/ abrasive liquid of decreasing order (Marusin S., 1995). Some researchers believe that application of improper sample preparation methods would erode and destroy existing microstructure. It can also introduce elements which are not part of the original specimen which would lead to false diagnosis of damage mechanism. Most of the specimen preparation techniques requires to dry sample till a certain degree which would affect the pore structure and can induce more microcracks. As reported by Hornain et al., (1996), crack density increases with intensity of drying. Thus such methods would not be able to give true quantitative analysis of microcracks. Vacuum impregnation methods results in excessive drying of the sample and therefore cracks are introduced during sample preparations (Chen 2002). Techniques involving mechanical polishing to produce either a flat surface or thin specimen could be problematic for concrete as different phases of concrete polish at different rates. Similarly putting concrete samples into a vacuum would lead to loss of water and change of structure. Polishing and lapping operation to smooth the specimen surface would lead to spreading of components over surface, thus preventing effective use of dot map studies for elements. Polishing would also destroy delayed etterndite crystals and cracks originating from it. While pressure or vacuum would introduce more cracks in the sample, original cracks would heal or widened by epoxy impregnation techniques

PAGE 29

18 (Marusin, 1995). However, sawn and unpolished approach could be better if the study is related to DEF analysis but for proper crack identification and quantification, sample preparation techniques would be required. In brief, Bisschop and Van Mier (2002) and other studies have indicated impregnation of whole sample and then cutting thin samples from fully impregnated samples would introduce less microcracks. Specimen preparation techniques involve cutting, drying, lapping, grinding and polishing. Improper handling of these operation results in induction of additional microcracks. For the scope of this study, epoxy impregnation technique developed by Struble and Stutzman (1989) was performed as it did not involve any drying of samples in any form. In this two-stage counter diffusion process, pore water is replaced by ethanol, and ethanol is substituted by epoxy without involvement of preheating or use of pressure techniques. Microscopic Instruments Optical Microscope Optical microscopy is one of the favored techniques for concrete petrography. In this instrument, image is viewed in full color and generally at a lower magnification, thus making it a more suitable technique for observing features at the millimeter scale and larger. Concrete and cementitious specimens prepared by thin-sectioning and fluorescent microscopy techniques can be viewed under optical microscope with the use of reflected ultraviolet lights. The image acquisition is done by placing a TriCCD camera which is linked to a personal computer acquisition system. However, Optical microscopes have shallow depth of field and limited resolution capacity, so a highly smooth and polished surface is required to produce a focused image under this technique. Furthermore, the

PAGE 30

19 intensity of the vacuum-impregnated sample is not stable after 2 min of exposure to UV light, and decreases in a continuous fashion with the elapse of time (Chen 2002). Scanning Electron Microscope Scanning electron microscopy/ x-ray microanalysis is comparatively more useful in studies which investigate quantification of microstructural properties such as microcracks and voids. Scanning electron microscopes has greater depth of field and high spatial resolution which produces focused images of poor specimens and analytical data for elemental composition analysis for the features seen on those images. As scanning electron microscopy and optical microscope petrography are different tools, different sample preparations are used for these different approaches. As described in figure 2.1, SEM scans a focused beam of electron across the specimen and measures any of several signals resulting from electron beam interaction with surface of concrete specimen. Images are monochrome in nature because they reflect the electron or x-ray flux resulting from beam/specimen interaction. Three major types of signals generated as a result are Secondary electrons, backscattered electrons and X-rays analysis. Figure 2-1 Working of Scanning Electron Microscope, (Perkes, 1999)

PAGE 31

20 Secondary Electrons (SE) are low energy electrons resulting from an inelastic collision of a primary beam electron with an electron of a specimen atom. Because of their low energy, they are readily absorbed and only those produced near the surface escape, resulting in an image of surface topography. SE imaging is principally applied in the examination of early age paste microstructure, high magnification imaging of microstructural features and for examining texture. Knowledge of the morphological and compositional characteristics of the hardened cement paste constituents is invaluable for their identification. As the hardened cement paste matures, filling of the void spaces eliminates the well-formed crystals and backscattered electron and x-ray imaging are more useful in examination of these microstructures (Stutzman, 2001) Backscattered Electron are high energy beams and capable of reflecting difference in atomic numbers. It can distinguish between the particles present in matrix on the basis of the variation in brightness of their images. In Backscattered Electron imaging technique contrast is generated by different phase compositions relative to their average atomic number and is observed by differential brightness in the image. SEM analysis using backscattered electron and X ray imaging requires highly polished surface for optimum imaging and X ray microanalysis. X radiation is produced when a specimen is bombarded by high energy electrons. X-ray microanalysis systems generally employ an energy dispersive detector with the other detector type being a wavelength detector. The x-ray energy level is displaced as the number of counts of each energy interval and appears as a set of peaks on a continuous background. X-ray radiation allows elements compositions to be obtained as

PAGE 32

21 the qualitative analyses prints like a dot map images. Each element is identified on a continuous spectrum by the position of its pick. Any sample can be examined by SE imaging but best use of BE imaging is for examining flat, polished surfaces (Struble and Stutzman, 1989). One of the disadvantages of Scanning Electron Microscope is that samples must be coated to allow discharge of electron build-up on the examined surface. A thin metal coating decreases build-up of negative charge by forming a conducting path for electrons to avoid distorted images. If images are accompanied by X-ray analysis, carbon coating can not be used if a carbonation process is to be studied. A gold coating gives a peak almost at a same position as sulfur (Oberhostler 1992). Thus, in some samples, it might be appropriate to coat the specimen due to concerns for altering the microstructure. High vacuum pressure inside the specimen chamber is also one of the factors which limits the use of SEM as some samples might break or disintegrate under high vacuum pressure. Variable Pressure Scanning Electron Microscope Hanke L, (1999) described that many applications where SEM/EDS evaluation could be useful involve samples that are not electrically conductive. These samples have traditionally required pretreatment, by coating with a conductive film, before SEM examination. Nonconductive samples are subject to a buildup of electrons on the examined surface. This buildup of electrons eventually causes scattering of the incoming electron beam, which interferes with imaging and analysis. Furthermore, samples that contain substantial water or other materials that volatilize in high vacuum also present challenges for SEM examination. These samples require controlled drying to allow the SEM chamber to reach high vacuum and to prevent deformation of the sample at the SEM vacuum.

PAGE 33

22 As the SEM is increasingly used for routine evaluations, there is increasing demand for examination without pretreatment. An answer to the problems of charging and volatile samples is the development of scanning electron microscopes that operate without exposing the sample to high vacuum. These microscopes are referred to alternately as environmental, low-vacuum, or variable-pressure SEM. In variable pressure SEM, the chamber at the electron gun is maintained at high vacuum, while a controlled amount of gas is allowed into the sample chamber. A fine aperture separates the gun and sample chambers to prevent excessive gas entrance into the gun chamber. Separate vacuum systems control the vacuum in the sample chamber and at the gun. The advantages of a higher pressure in the sample chamber are obvious for wet and volatile samples. The higher pressure decreases the rate of volatilization. This decreases the drying and deformation of wet samples. For nonconductive samples, the advantage of higher pressure is less obvious. When gas molecules in the sample chamber are struck by the electron beam, the gas is ionized. These positive ions are attracted to and neutralize the negative charge build-up on the nonconductive specimens. By controlling the pressure in the sample chamber, the number of gas molecules intercepting the electron beam is maintained at a level that is sufficient to prevent charging, but does not deflect the beam sufficiently to prevent imaging and microanalysis. Thus, the use of variable pressure scanning electron microscope holds advantage over traditional scanning electron microscope as no coating is required over the samples and samples with moisture could also be imaged.

PAGE 34

23 Image Processing Techniques Important steps for any microstructural images are to acquire the image and to process microstructure image to get a quantitative analysis of the sample. For this study, quantitative analysis is to identify and measure the number of cracks and crack density over number of images. The image processing should be able to differentiate between voids and cracks, and boundary cracks around aggregates or between phases. Darwin (2001) researched on quantification of microcracks of cement mortar samples by backscatter electron technique of scanning electron microscope. He reported that the use of edge detection, gradients and other filters failed to identify cracks passing through low density phases as its intensity depends on adjacent and underlying phases. In his approach, the identification of cracks was based on local changes in grey level. A line scan was taken perpendicular to the images for identification of potential cracks. Perimeter to squared area was taken to differentiate between cracks and voids. Phase identification was also carried out based on pixel intensity. He conclude that as grey level of image is affected by the density of adjacent and underlying phases, cracks identified based on change in grey level would not be correct. Thus the correct practice for crack identification should establish minimum gradient in grey level adjacent to cracks. Soroshian (2003) used segmentation of images from grey level to binary images by thresholding approach for viewing microcracks in samples. The research compared manual thresholding operations with three automated thresholding operation such as factorization, entropy and moment. In manual thresholding, low threshold level was set to zero or default for auto thresholding and best threshold level image was determined. The optimum high threshold level was determined by comparing original gray image and the image after the application of manual threshold on binary image. When features of

PAGE 35

24 interest (microcracks and voids) were correctly highlighted, the most distant contrast between microcracks/voids and concrete background was obtained. After training the program with images of different resolutions, three automated thresholding operation were performed and compared results with manual thresholding. The study reported that there was no significant difference between manual thresholding and automated thresholding operation by factorization method. There was a difference of 5% and 2% respectively, for entropy and moment methods of automated process. After identifying the cracks, form/shape factor of 3.5 was used to distinguish cracks from voids. It was shown that automated process was different by 6.39% and 3.02% from manual process. However, the author concluded that due to the ease of operation and efficiency given by automated process, this error was reasonably small. Summary Literature review of methods for preparation of concrete samples for viewing microstructure was performed. As described before, many sample preparation methods involved the use of pressure or heat, which might cause concerns over additional cracks in concrete and might lead to erroneous results. Epoxy impregnation of concrete samples was identified as a process which uses counter diffusion process as sample preparation means for this study. As optical microscope would have inadequate depth of field and resolution for microstructure, scanning electron microscopy was preferred for image acquisition, Variable pressure electron scanning microscope, which requires no special coating over samples and ability to view samples with moisture was found to be more suitable for image analysis. After acquiring images, manual thresholding approach to map microcracks was selected for quantification of microcracks in a given sampl

PAGE 36

CHAPTER 3 RESEARCH METHODOLOGY Introduction This chapter describes the methods for sample selection, specimen preparation method and image analysis techniques used in this research. As described in Chapter 2, the objective of this study was to find if the concrete cured at elevated temperature shows any additional microcracks compared to samples cured at room temperature. This investigative study was carried out in three phases. Phase one was to identify the samples which shows tendency for microcracks, while phase two was to select an appropriate sample preparation methods. Phase three comprised of sample preparation and image analysis technique to quantify and compare results. Sample Selection Samples were selected from an ongoing research project Adiabatic Temperature Rise in Mass Concrete conducted by University of Florida for Florida Department of Transportation (FDOT). In this study, twenty different mixes were prepared using AASHTO Type II cement made by two manufactures with different percentages of pozzlonic materials and two different placing temperatures of 73F and 95F were used to study adiabatic temperature rise in laboratory simulated mass concrete conditions. As higher placing temperature increases the concrete curing temperature, samples made from higher of two placing temperatures, 95F were selected for microstructural analysis. The field of investigation was further narrowed down further by selecting the mix made with cement which has higher heat of hydration. A Heat of Hydration test by Construction 25

PAGE 37

26 Technologies Ltd. (CTL) revealed higher heat of hydration in AASHTO Type II cement (Cement B) manufactured by CEMEX (78.2 cal/g at 7 days) was greater compared to AASHTO Type II cement ( Cement A) manufactured by Florida Rock (66.2 cal/g at 7 days). Thus, mixes made with Cement B and 95 F placing temperature were selected. This included five different mixes were prepared with varied pozzlonic contents. 100% Cement 20% Fly Ash + 80 % Cement 35% Fly Ash + 65 % Cement 50% Slag + 50% Cement 70% Slag + 30% Cement Table 3-1 Mix Design Data AASHTO Type II CEMEX Cement ( Cement B ) (Mix proportions per cubic yard) Mix Cement (lb) Fly Ash (lb) Slag (lb) Water (lb) Fine Aggregate (lb) Coarse Aggregate (lb) Air Entertainer (oz) Water Admixture (oz) 95B00P 760.00 0.00 0.00 279.00 1033.00 1736.00 2.30 60.86 95B20F 608.00 152.00 0.00 279.00 1033.00 1708.00 4.00 30.40 95B35F 494.00 266.00 0.00 279.00 980.00 1687.00 4.00 68.20 95B50S 380.00 0.00 380.00 279.00 1021.00 1724.00 2.30 38.06 95B70S 228.00 0.00 532.00 279.00 1021.00 1724.00 2.30 68.20 Prepared concrete samples were kept in two different types of containers. For samples to be kept at room condition, concrete was poured in 6x12 plastic containers and for samples to be kept at high temperature cured condition, specially acquired thermal curing chambers were used. Thermal curing chamber maintained temperature inside chamber in an adiabatically condition. The chambers were connected with thermo

PAGE 38

27 couples to a micro controller system for maintaining temperature and to document the temperature rise in the concrete samples. Data given in following Table 3.2 shows the first four day temperature rise in concrete specimen kept in thermal curing chambers. Table 3-2 Adiabatic Temperature Rise Data Cement B (95F Placing Temperature) Plain Cement 20% Fly Ash 35% Fly Ash 50% Slag 70% Slag Time (Day) Temp. Rise (F) Temp. Rise (F) Temp. Rise (F) Temp. Rise (F) Temp. Rise (F) 0.0 0.00 0.00 0.00 0.00 0.00 0.5 2.60 56.80 50.35 58.60 45.00 1.0 72.15 72.90 61.90 76.70 69.40 1.5 84.20 77.40 67.85 80.70 73.90 2.0 85.60 77.60 68.90 81.80 76.30 2.5 85.65 77.60 69.20 82.00 77.90 3.0 85.95 77.60 69.30 82.00 78.90 3.5 85.95 77.60 69.60 82.00 79.50 4.0 85.95 77.60 69.90 82.00 79.90 As it can be seen in Table 3-2, data indicates that sample 95 B 00 P had the highest temperature rise of 85.95 F over placing temperature of 99 F, thus effectively curing temperature in the core is 186F. Therefore, sample 95 B 00 P was selected for microscopic studies due to the following reasons. Higher placing temperature had higher curing temperature and therefore gives higher probability of presence of additional microcracks Cement B had higher heat of hydration (78.2 cal/g at 7 days) compared to Cement A (66.2 cal/g at 7 days), which might lead to higher curing temperature As indicated in Table 3.2, sample 95 B 00 P with plain cement showed higher temperature rise compared to samples with varied percentages of fly ash or blast furnace slag. Specimen Preparation Method As described in Chapter 2, many different approaches have been used in the past to view and document microcracks in cement paste or concrete. However, most of the

PAGE 39

28 methods indicate the use of heat, vacuum or pressure along the course of sample preparation or prior to impregnation, which might lead to additional microcracks due to sample preparation widely known as secondary microcracks. Samples cured at elevated temperature as well as those cured at room temperature would undergo epoxy impregnation method for Scanning Electron Microscopy as established by Struble and Stutzman (1989). In this method they used counter diffusion method to replace the pore water of concrete first by ethyl alcohol and then low viscosity epoxy. Neither of these processes requires any prior pre-heating of surface to dry, nor does it use any vacuum for impregnation of epoxy. The only drawback of this method is that sample preparation would take longer time to complete the process. Considering all methods and their advantages and disadvantages, this method was selected for sample preparation in current study. Procedure for Sample Preparation A small 1/2 thick sample was cut with diamond saw from the samples cured at room temperature and those cured in thermal curing chambers. During this operation, in order to reduce the heat generation, cooling water was circulated through water pump to diamond blade to aid cutting. Once the required thickness of the sample is obtained, other cutting operation was carried out on a diamond wafer saw machine (figure 3-1). It was used to reduce any artifacts and required polishing time in the samples during the preparation period. In this machine, blade was passed through cutting fluid (polypropylene glycol) to keep the samples from drying. After cutting by diamond wafer, average size of specimen was 2.15x 1.8. The samples then went through a two stage counter diffusion process.

PAGE 40

29 Figure 3-1 Diamond Wafer Saw In this process, first pore water of concrete is replaced by 200% Ethanol made by Sigma-Aldrich Inc., Montana, and then in the second stage, ethanol was replaced by ultra low viscosity epoxy solution made by Structure Probe Inc., Pennsylvania. For the first stage of alcohol pore water replacement stage, the specimen was placed in a lidded jar filled with ethanol (200 Proof) for the alcohol pore solution replacement stage. A companion specimen was used to determine the depth of replacement of pore water by alcohol. This companion specimen was a remnant from the specimen after trimming. This specimen was now placed in a jar filled with ethanol dyed by a deep red alcohol-miscible dye by Poly science Inc. By splitting or sawing the companion specimen after a period of time, the depth of replacement is seen by the depth of dye coloration. On the average, alcoholpore water replacement was found to be 1mm/day in concrete samples. When this depth reaches to half the section thickness, the pore solution replacement by alcohol was stopped. The section is then placed in a

PAGE 41

30 container with the ultra low-viscosity epoxy (see Figure 3-2) for the same time as alcohol pore water replacement process. Curing of epoxy was performed as specified by product manual. The epoxy was made up of 4 different chemicals as described in Table. 3-3 Figure 3-2 Ultra Low Epoxy Kit by SPI Table 3-3 Mixing Schedule for Epoxy No. Name of Chemical Mixing Schedule 1. Vinylcyclohexene dioxide (VCD) 10g 2. n-Octenyl succinic anhydride (n-OSA) 20 g 3. Butanediol Diglycidyl Ether (BDE) 0.3 g 4. Dimethylaminoethanol (DMAE) 0.3 g Ethanol was replaced by epoxy at two hour intervals with the proportion of ethanol to epoxy was changed from 3:1, 1:1, 1:3 at each interval and finally 100% epoxy solutions was used for keeping it in embedding medium overnight. For curing of epoxy, samples were kept in a heating chamber at 131F for six hours. At the end of six hours, the specimens were ready for the cutting and polishing stages. Grinding and polishing are important steps in sample preparation. Grinding is used to expose new surface layer and to remove excessive epoxy coating on the surface. However, excessive grinding would remove the depth of epoxy layer and could damage microstructure. Grinding was done on a Metaserv 2000 Grinder Polisher Machine (see

PAGE 42

31 figure 3-3 and 3-4) on 150 RPM. Abrasive papers of 180,240, 320,400 and 600 were used for the removal of material from specimen surface. Approximately 2 minute of time was given to each grinding cycle. Figure 3-3 Metaserv Grinder and Polisher Figure 3-4 Polished Sample Polishing is done for undoing the damage done due to sawing and grinding of surface. After grinding by 600 grits, surface becomes smooth and ready for further polishing. Polishing operation involves successively decreasing size of diamond suspension paste from 6m to 3m, 1m and 0.25m manufactured by Buhler (see figure 3-5). A polishing cloth was attached to rotating wheel of Metaserv 2000 Grinder Polisher. This method helps in reducing the damage done by previous suspension by removal of finer scratches and improving the surface for microscopic study. At the end of this operation, the sample is wiped with a plain cloth and kept in glass jar (See figure 3-6). During polishing process, it is important to understand that cutting by diamond saws would damage the samples up to 30% of its thickness. So polishing should be done till 30% of blade thickness in order to remove all artifacts.

PAGE 43

32 Figure 3-5 Diamond Suspension Paste Figure 3-6 Prepared Samples Preparation for Image Processing Microstructural analysis was performed at Advance Material Characteristic Lab at the Department of Civil and Coastal Engineering, University of Florida, Gainesville. The facility has a Hitachi S 3000N, variable pressure Scanning Electron Microscope (VPSEM). As described earlier, VPSEM would allow the viewing of the specimen without covering it with a conductive surface. A double sided carbon tape was used for attaching samples to specimen holder. This conductive tape avoids any charge built-up and provides a good background surface. Images were viewed at 50 Pa (0.4 Torr), accelerating voltage of 10.0 KV and on the scale of 200 m at 250 X magnification of specimen surface. At the start of this study, polished samples were viewed directly under microscope and five representative images were taken from different location of specimen. However, during the course of study, it was decided to increase the number of images to 16 to have good representative nature of the sample, which was formulated on basics of 4x4 matrixes. The following figures 3-7 and 3-8 shows the grid system.

PAGE 44

33 Figure 3-7 Specimen with Copper Tape Figure 3-8 Specimens with 4x4 matrixes 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 2.15 1.80 The samples were divided into 16 equal parts using a copper conductive tape to make grid on the sample to identify the points of interesting while viewing under microcracks. Image Analysis Images taken from 16 different square position provides information related to microstructure of concrete. A manual mapping of cracks was done in order to find out the length of the microcracks and density per micrometer for comparing all the samples. To identify proper microcracks from the image, shape factor was used. Shape factor is ratio of length to width. If this ratio was greater then 3.5, the object of consideration termed as microcracks or if the ratio was less than 3.5, the object was termed as Void. Each image was manually mapped as well as by viewing on ImageJ software to find out the exact number of cracks to avoid any bias from operator. The method of image analysis is described in more details in Chapter 4.

PAGE 45

34 Summary In this study, samples were taken from concrete mix which showed maximum probability of crack development due to high curing temperature. Specimen preparation method was selected to avoid any preheating of the samples and to avoid any use of vacuum or pressure techniques for impregnation. Once specimen were impregnated and prepared with grinding and polishing, a grid made from conductive copper tape was used to locate the points of interests, while viewing the sample under microscope. At the end, a statistical analysis was performed to investigate total representative nature of images and the rationality for use in this study.

PAGE 46

CHAPTER 4 IMAGE PROCESSING AND ANALYSIS Introduction This chapter describes the image processing and analysis carried out in this research. A total of four samples were made, two of which cured in room temperature and the other two cured at elevated temperature. As described earlier, the samples were divided into 16 equal parts by conductive copper tape. The images were taken around the center of square in 250X resolution Thus for each sample a total of 16 images were taken in 1028 x 960 pixel resolution and 50 Pa working pressure. In SEM observation, finer surface structure images can generally be obtained with lower accelerating voltages. At higher accelerating voltages, the beam penetration and diffusion area become larger, resulting in unnecessary signals (e.g., backscattered electrons) being generated from within the specimen. These signals reduce the image contrast and veils fine surface structures. It is especially desirable to use low accelerating voltage for observation of low-concentration substance. After acquiring the images, image processing was performed to view microcracks in concrete matrix. ImageJ software developed by National Institute of Health was used for the purpose of image processing in this study. Based on java platform, this software has the capability to perform wide ranges of operations such as binary thresholding, area measurement, filtering noise and mathematical operations. 35

PAGE 47

36 Image Processing Technique First step of image processing was to perform segmentation of image into binary image (black and white) by brightness thresholding approach. In this process, software would change the intensities of image pixels which satisfy the given condition, thus highlighting the regions of interests. However, as it can be imagined, there would be variations in grey level of matrix in different images, and there could be bias in operator visions. Investigating this issue, Soroshian (2003) performed a comparative analysis of manual and automated thresholding operation on images taken by Scanning Electron Microscope, where three types of automated thresholding operation (factorization, entropy and moment) were performed. However, study reported that there were no significant differences between manual and factorization mode of thresholding in automated process. The study also reported 5% and 2% difference in Entropy and Moment mode of thresholding respectively which could be regarded as less significant. Therefore for the scope of this study, manual thresholding by brightness approach was performed. Images were acquired with variable pressure scanning electron microscope and PCI-quartz picture acquisition system. Images were transferred and viewed with ImageJ software. For each image, low threshold level was set to zero or default for thresholding and best threshold level image was determined by comparing the original gray image and the image after application of manual threshold on binary image. Total of sixty-four images were converted one after another to binary mode. After binary operation, images were manually mapped to find out total number of crack lengths and crack density across specimens.

PAGE 48

37 Sample 95B00P Room Temperature Sample 1 A. Image 95 B 00 P RT 1 (2) B. Image 95 B 00 P RT 1 (10) C. Image 95 B 00 P RT 1 (11) D. Image 95 B 00 P RT 1 (16) Figure 4-1 Representative Images of Mix 95 B 00 P RT-1 before Image Processing After highlighting features of interest (microcracks and voids) to find the most distinct contrast with concrete background, image analysis was performed to distinguish microcracks and voids. As it is known that microcracks are long elongated structure, length to width (also known as aspect/ form/ shape) ratio was applied. The objects having shape factor of 3.5 or greater was termed as microcracks, while the remaining objects were termed as voids. A software plugin Shape descriptor from ImageJ was used to

PAGE 49

38 identify voids in the images. Shape descriptor takes minimum and maximum pixel values and plots the void concentration in a separate image. After completing void removal from image, microcracks are mapped manually and counted. A. Image 95 B 00 P RT 1 (2) B. Image 95 B 00 P RT -1 10 ) C. Image 95 B 00 P -RT 1 (11) D. Image 95 B 00 RT (16) Figure 4-2 Representative Images of Mix 95 B 00 P RT-1 after Image Processing As the sample was divided into 16 equal parts by copper conductive double-sided tape, 16 representative images were taken from the center of every part. After processing the image, cracks were manually mapped to find out the total length of cracks in each

PAGE 50

39 image. As it can be seen in table 4-1, density of cracks over the given area was found by converting crack lengths and dividing it by image area. Table 4-1 Crack Quantification Analysis for Mix 95 B 00 P RT-1 No. Image Scale Cracks (cm) Cracks length (m) Image Field Crack Density (m/m 2 ) 1 RT 1-1 200m/11 cm 72 cm 1309.091 183 m x 250 m 0.028614 2 RT 1-2 200m/11 cm 71 cm 1290.909 183 m x 250 m 0.028217 3 RT 1-3 200m/11 cm 65 cm 1181.818 183 m x 250 m 0.025832 4 RT 1-4 200m/11 cm 82 cm 1490.909 183 m x 250 m 0.032588 5 RT 1-5 200m/11 cm 61 cm 1109.091 183 m x 250 m 0.024242 6 RT 1-6 200m/11 cm 66 cm 1200.000 183 m x 250 m 0.026230 7 RT 1-7 200m/11 cm 64 cm 1163.636 183 m x 250 m 0.025435 8 RT 1-8 200m/11 cm 61 cm 1109.091 183 m x 250 m 0.024242 9 RT 1-9 200m/11 cm 73 cm 1327.273 183 m x 250 m 0.029011 10 RT 1-10 200m/11 cm 92 cm 1672.727 183 m x 250 m 0.036562 11 RT 1-11 200m/11 cm 72 cm 1309.091 183 m x 250 m 0.028614 12 RT 1-12 200m/11 cm 68 cm 1236.364 183 m x 250 m 0.027024 13 RT 1-13 200m/11 cm 98 cm 1781.818 183 m x 250 m 0.038947 14 RT 1-14 200m/11 cm 63 cm 1145.455 183 m x 250 m 0.025037 15 RT 1-15 200m/11 cm 65 cm 1181.818 183 m x 250 m 0.025832 16 RT 1-16 200m/11 cm 58 cm 1054.545 183 m x 250 m 0.023050 Avg. 1285.227 0.028092 SD. 203.7358 0.004453 Data analysis shows that for 95 B 00P RT-1 samples, average lengths of cracks were 1285.227 m. Maximum length of cracks in an image was found to be 1672m, while minimum length of cracks in an image was 1054.545 m, giving standard deviation of 203.227. As it can be seen in Table 4.1, average crack density in an image was 0.028092 m/m, 2 the maximum density of cracks was 0.038947 m/m 2 and minimum density of cracks was 0.02305 m/m 2 giving standard deviation of 0.004463. Figure 4.3 describes variations in crack density of room temperature cured sample by each image.

PAGE 51

40 95 B 00 P Room Temprature Sample 100.010.020.030.040.0512345678910111213141516Image NumberCrack Density Figure 4-3 Variations in Crack Density of 95 B 00 P RT1 To find out if the sample size was representative in nature, a statistical study was performed. Odd and Even numbers of images were selected for calculating average density of lengths. As it can be seen in Table 4-2, Crack density of odd number images were 0.02832 m/m 2 with standard deviation of 0.005281, while crack density of even number images were 0.02787 m/m 2 with standard deviation of 0.004982. After comparing it with average crack density of all 16 images of 0.02809 m/m 2 it can be seen that crack density is averaged almost equally. Thus it could be deduced that crack density is representative in nature throughout the sample and the number of images are adequate for giving representative portrait of the sample. Table 4-2 Comparison of Odd and Even Number Images No. Crack Density of Odd Number Image (m/m 2 ) Crack Density Even Number Images (m/m 2 ) 1 0.028614 0.028217 2 0.025832 0.032588 3 0.024242 0.026230 4 0.025435 0.024242 5 0.029011 0.036562 6 0.028614 0.027024 7 0.038947 0.025037 8 0.025832 0.023050 Avg. 0.028316 0.027869 SD 0.004643 0.004563

PAGE 52

41 Sample 95B00P Room Temperature Sample 2 A. Image 95 B 00 PRT2 (4) B. Image 95 B 00 PRT2 (10) C. Image 95 B 00 P -RT2 (15) D. Image 95 B 00 P -RT2 (16) Figure 4-4 Representative Images of Mix 95 B 00 P RT-2 before Image Processing Figure 4.4 shows images from sample number two which was cured at room temperature. It can be seen from images that extensive cracking is present in almost all parts of representative images. As described earlier, a total of 16 images were taken from each sample. The images were then processed by ImageJ software for quantification of microcracks.

PAGE 53

42 Image 95 B 00 P-RT2 (4) Image 95 B 00 P-RT2 (10) Image 95 B 00 P-RT2 (15) Image 95 B 00 P-RT2 (16) Figure 4-5 Representative Images of Mix 95 B 00 P RT-2 after Image Processing After processing the image by converting it into binary image, cracks were manually mapped to find out the total length of cracks in each image. As it can be seen in table 4-3, the density of cracks over the given area was found by converting crack lengths and dividing it by image area. Another statistical analysis was carried out to see if the number of images constituent representative area for the sample.

PAGE 54

43 Table 4-3 Crack Quantification Analysis for Mix 95 B 00 P RT-2 No. Image Scale Cracks (cm) Cracks length (m/m 2 ) Image Field Crack Density (m/m 2 ) 1 RT 2-1 200m/11 cm 64 cm 1163.636 183 m x 250 m 0.025435 2 RT 2-2 200m/11 cm 68 cm 1236.364 183 m x 250 m 0.027024 3 RT 2-3 200m/11 cm 56 cm 1018.182 183 m x 250 m 0.022255 4 RT 2-4 200m/11 cm 89 cm 1618.182 183 m x 250 m 0.035370 5 RT 2-5 200m/11 cm 90 cm 1636.364 183 m x 250 m 0.035768 6 RT 2-6 200m/11 cm 65 cm 1181.818 183 m x 250 m 0.025832 7 RT 2-7 200m/11 cm 82 cm 1490.909 183 m x 250 m 0.032588 8 RT 2-8 200m/11 cm 69 cm 1254.545 183 m x 250 m 0.027422 9 RT 2-9 200m/11 cm 83 cm 1509.091 183 m x 250 m 0.032986 10 RT 2-10 200m/11 cm 67 cm 1218.182 183 m x 250 m 0.026627 11 RT 2-11 200m/11 cm 64 cm 1163.636 183 m x 250 m 0.025435 12 RT 2-12 200m/11 cm 61 cm 1109.091 183 m x 250 m 0.024242 13 RT 2-13 200m/11 cm 67 cm 1218.182 183 m x 250 m 0.026627 14 RT 2-14 200m/11 cm 82 cm 1490.909 183 m x 250 m 0.032588 15 RT 2-15 200m/11 cm 84 cm 1527.273 183 m x 250 m 0.033383 16 RT 2-16 200m/11 cm 68 cm 1236.364 183 m x 250 m 0.027024 Avg. 1317.045 0.028788 SD. 194.692 0.004256 Data analysis shows that for 95 B 00P RT-2 samples, average lengths of cracks was 1317.045 m. Maximum length of cracks in an image was found to be 1636m, while minimum length of cracks in an image was 1018.182 m, giving standard deviation of 194.692. As it can be seen in figure 4-6, average crack density in an image was 0.028788 m/m 2 maximum density of cracks was 0.035768 m/m 2 and minimum density of cracks was 0.0222 m/m 2 giving standard deviation of 0.004256. The above data shows presence microcracks in the given sample. However, to see if the images were sufficient to draw conclusions from it, another statistical analysis was carried out. In this technique, images were divided into two formations. The formation was varied such as diagonals, odd-even or lines to prove representative nature of the samples.

PAGE 55

44 95 B 00 P Room Temperature Figure 4-6 Variations in Crack Density of Sample 95 B 00 P RT2 Figure 4-6 demonstrates variations in crack density of sample 95 B 00P RT2 by each image. The data given in Table 4-4 shows that the crack density of odd number images were 0.029309 m/m 2 with standard deviation of 0.0049, while crack densities of even number images were 0.028266m/m 2 with standard deviation of 0.0037. Table 4-4 Comparison of Odd and Even Number Images No. Crack Density of Odd Number Image (m/m 2 ) Crack Density Even Number Images (m/m 2 ) 1 0.025435 0.027024 2 0.022255 0.035370 3 0.027422 0.032588 4 0.025832 0.035768 5 0.032986 0.026627 6 0.025435 0.024242 7 0.026627 0.027024 8 0.033383 0.032588 Avg. 0.029309 0.028266 SD 0.004922 0.003737 After completing analysis of samples cured at room temperature, crack density calculations were performed on samples cured at high temperature. As described earlier, samples were divided in 16 equal parts with conductive copper tape and images were 0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Number of Images Crack Den s I T Y

PAGE 56

45 taken from the center of rectangle. Figure 4-7 shows images from high temperature cured sample one. Sample 95B00P High Temperature Sample 1 Image 95 B 00 HT -1 (1) Image 95 B 00 HT -1 (2) Image 95 B 00 HT -1 (3) Image 95 B00 HT -1 (16) Figure 4-7 Representative Images of Mix 95 B 00 P HT-1 before Image Processing Maximum heat gain during high temperature curing process of concrete reached up to 85.5 F. The adiabatic temperature condition was generated and measured in the laboratory with a microcontroller based SURE-CURE curing system.

PAGE 57

46 Image 95 B 00 P HT-1 (1) Image 95 B 00 P HT 1 (2) Image 95 B 00 P HT 1(3) Image 95 B 00 P HT 1(7) Figure 4-8 Representative Images of Mix 95 B 00 P HT-1 after Image Processing After preparing the samples and subsequent polishing and lapping operations, copper metal grid system was mounted. Samples were kept in variable pressure scanning electron microscope to analyze microscopic features. As it can be seen in Figure 4-8, images were initially processed by manual thresholding by brightness approach. Lower thresholding was kept at zero, while upper thresholding was moved till the best contrast between images and objects of interest developed. Voids were removed from the binary picture and manual mapping of microcracks were done for crack quantification.

PAGE 58

47 Table 4-5 Crack Quantification Analysis for Mix 95 B 00 P HT-1 No. Image Scale Cracks (cm) Cracks length (m) Image Field Crack Density (m/m 2 ) 1 HT 1 -1 200m/11 cm 105 cm 1909.091 183 m x 250 m 0.041729 2 HT 1-2 200m/11 cm 106 cm 1927.273 183 m x 250 m 0.042126 3 HT 1-3 200m/11 cm 94 cm 1709.091 183 m x 250 m 0.037357 4 HT 1-4 200m/11 cm 52 cm 945.4545 183 m x 250 m 0.020666 5 HT 1-5 200m/11 cm 58 cm 1054.545 183 m x 250 m 0.023050 6 HT 1-6 200m/11 cm 53 cm 963.6364 183 m x 250 m 0.021063 7 HT 1-7 200m/11 cm 48 cm 872.7273 183 m x 250 m 0.019076 8 HT 1-8 200m/11 cm 54 cm 981.8182 183 m x 250 m 0.021461 9 HT 1-9 200m/11 cm 58 cm 1054.545 183 m x 250 m 0.023050 10 HT 1-10 200m/11 cm 55 cm 1000.000 183 m x 250 m 0.021858 11 HT 1-11 200m/11 cm 71 cm 1290.909 183 m x 250 m 0.028217 12 HT 1-12 200m/11 cm 65 cm 1181.818 183 m x 250 m 0.025832 13 HT 1-13 200m/11 cm 62 cm 1127.273 183 m x 250 m 0.024640 14 HT 1-14 200m/11 cm 75 cm 1363.636 183 m x 250 m 0.029806 15 HT 1-15 200m/11 cm 65 cm 1181.818 183 m x 250 m 0.025832 16 HT 1-16 200m/11 cm 82 cm 1490.909 183 m x 250 m 0.032588 Avg. 1253.409 0.027397 SD. 339.4677 0.007420 Data analysis from Table 4-3 shows that for 95 B 00P HT-1 samples average lengths of cracks were 1253.409 m. Maximum length of cracks in an image was found to be 1927 m, while minimum length of cracks in an image was 872.73 m, giving standard deviation of 339.4677. As it can be seen in Table 4-5, average crack density in an image was 0.027397 m/m 2 maximum density of cracks was 0.042126 m/m 2 and minimum density of cracks was 0.019076 m/m 2, giving standard deviation of 0.00742. The above results determine presence of microcracks in the given sample, to find if the sample size is sufficient to draw conclusion, another statistical analysis was carried out. In this process, images were divided into two groups in various formations and average crack density was calculated for each of them. Results shown in table 4-6 shows that average crack density is present throughout the concrete sample.

PAGE 59

48 95 B 00 P High Temprature Sample 100.010.020.030.040.0512345678910111213141516Image NumberCrack Density Figure 4-9 Variations in Crack Density of Sample 95 B 00P HT1 As it can be seen in Table 4-2, Crack density of odd number images were 0.027869 m/m 2 with standard deviation of 0.007721, while crack density of even number images were 0.026925 m/m 2 with standard deviation of 0.007676. After comparing it with average crack density of all 16 images of 0.027397 m/m 2 it can be seen that crack density is averaged almost equally. Table 4-6 Comparison of Odd and Even Number Images No. Crack Density of Odd Number Image (m/m 2 ) Crack Density Even Number Images (m/m 2 ) 1 0.041729 0.042126 2 0.037357 0.020666 3 0.023050 0.021063 4 0.019076 0.021461 5 0.023050 0.021858 6 0.028217 0.025832 7 0.024640 0.029806 8 0.025832 0.032588 Avg. 0.027869 0.026925 SD 0.007751 0.007576 Coefficient of variation being less than 15% proves that results are concentrated within 15% of the mean. Thus average crack density is averaged equally all across the sample.

PAGE 60

49 Sample 95B00P High Temperature Sample 2 Image 95 B 00 HT -2 (15) Image 95 B 00 HT -2 (7) Image 95 B 00 P HT 2 (15) Image 95 B 00 P HT 2 (16) Figure 4-10 Representative Images of Mix 95 B 00 P HT-2 before Image Processing The above four images are representative images from mix 95 B 00 P cured at high temperature, taken by Hitachi S 3000 N variable pressure scanning electron microscope. A 4x4 matrix made by conductive copper tape was used to take 16 representative images across the sample.

PAGE 61

50 Image 95 B 00 P HT 2 (4) Image 95 B 00 P HT 2 (10) Image 95 B 00 P HT 2 Image 95 B 00 P HT 2 Figure 4-11 Representative Images of Mix 95 B 00 P HT-2 after Image Processing All the images were processed in a java based free source ImageJ software, developed by National Institute of Health, for image processing. Manual thresholding was done till the perfect contrast between regions of interest with background was developed. Images were then processed to distinguish voids from cracks. As cracks have higher length to width ratio, shape factor of 3.5 was used to filter out voids from the cracks. After that, cracks lengths were manually mapped on the image.

PAGE 62

51 Table 4-7 Crack Quantification Analysis for Mix 95 B 00 P HT-2 No. Image Scale Cracks (cm) Cracks length (m) Image Field Crack Density (m/m 2 ) 1 HT 2-1 200m/11 cm 58 cm 1054.545 183 m x 250 m 0.023050 2 HT 2-2 200m/11 cm 98 cm 1781.818 183 m x 250 m 0.038947 3 HT 2-3 200m/11 cm 88 cm 1600.000 183 m x 250 m 0.034973 4 HT 2-4 200m/11 cm 64 cm 1163.636 183 m x 250 m 0.025435 5 HT 2-5 200m/11 cm 57 cm 1036.364 183 m x 250 m 0.022653 6 HT 2-6 200m/11 cm 69 cm 1254.545 183 m x 250 m 0.027422 7 HT 2-7 200m/11 cm 57 cm 1036.364 183 m x 250 m 0.022653 8 HT 2-8 200m/11 cm 54 cm 981.8182 183 m x 250 m 0.021461 9 HT 2-9 200m/11 cm 75 cm 1363.636 183 m x 250 m 0.029806 10 HT 2-10 200m/11 cm 65 cm 1181.818 183 m x 250 m 0.025832 11 HT 2-11 200m/11 cm 70 cm 1272.727 183 m x 250 m 0.027819 12 HT 2-12 200m/11 cm 62 cm 1127.273 183 m x 250 m 0.024640 13 HT 2-13 200m/11 cm 65 cm 1181.818 183 m x 250 m 0.025832 14 HT 2-14 200m/11 cm 76 cm 1381.818 183 m x 250 m 0.030204 15 HT 2-15 200m/11 cm 95 cm 1727.273 183 m x 250 m 0.037755 16 HT 2-16 200m/11 cm 61 cm 1109.091 183 m x 250 m 0.024242 Avg. 1265.909 0.027670 SD. 238.8743 0.005393 Data analysis from Table 4-8 shows that for 95 B 00P HT-2 samples average lengths of cracks was 1265.909 m. Maximum length of cracks in an image was found to be 1781.818 m, while minimum length of cracks in an image was 981.8182 m, giving standard deviation of 238.8743. As it can be seen in Table 4.5, average crack density in an image was 0.027670 m/m 2 maximum density of cracks was 0.038947 m/m 2 and minimum density of cracks was 0.021461 m/m 2, giving standard deviation of 0.005393. Figure 4-12 shows variations in microcrack density in the sample made from high temperature cured concrete mix 95 B 00 P, i.e., concrete made with cement B, (manufactured by Cemex AASHTO type II, heat of hydration 78.2 cal/g at 7 days), placing temperature of 95F, and without any pozzlonic contents added to mix. The higher temperature reached was 180F.

PAGE 63

52 95 B 00 P High Temprature00.0050.010.0150.020.0250.030.0350.040.04512345678910111213141516Number of ImagesCrack Density Figure 4-12 Variations in Crack Density of Sample HT 2 Figure 4-12 shows maximum value of crack density was 0.3894 with lowest crack density appeared to be 0.214. Table 4.8 shows crack density of odd number images from the sample having crack density of 0.028068 m/m 2 while crack density of even number images from samples was 0.027273 m/m 2 From Statistical analysis of image patterns and coefficient of ratio within 20% suggests that, average image density is present throughout the images and given sample size is true representative in nature. Similar results were found with random or diagonal image patterns with same images. Table 4-8 Comparison of Odd and Even Number Images No. Crack Density of Odd Number Image m/m 2 Crack Density Even Number Images m/m 2 1 0.023050 0.038947 2 0.034973 0.025435 3 0.022653 0.027422 4 0.022653 0.021461 5 0.029806 0.025832 6 0.027819 0.024640 7 0.025832 0.030204 8 0.037755 0.024242 Avg. 0.028068 0.027273 SD 0.005775 0.005349

PAGE 64

53 Comparative Analysis of Samples To analyze the differences between both samples, which are cured at room temperature, a comparative analysis was performed. Table 4-9 Comparative Analysis of 95 B 00P RT1 and 95B00P RT2 Sample Avg. Crack Length (m) Avg. Crack Density (m/m 2 ) RT1 1285.227 0.028092 RT2 1317.045 0.028788 Average 1301.136 0.02844 % Change 2.48% 2.48% The results shown in Table 4-9 show that the difference between average crack lengths of samples cured at room temperatures was 32 m/m 2 ; the difference between average crack densities was 0.0006. The differences between average crack lengths in high temperature cured samples were 12.5 m/m2 and difference in their crack density were 0.0002. Table 4-10 Comparative Analysis of 95 B 00P HT1 and 95B00P HT2 Sample Avg. Crack Length (m) Avg. Crack Density (m/m 2 ) HT1 1253.409 0.027397 HT2 1265.909 0.027670 Average 1259.659 0.027535 % Change 0.957% 0.99%

PAGE 65

54 Summary Four samples were taken from concrete mix 95 B 00 P, two of which were cured at room temperature and the two remaining samples were cured at high temperature. After sample preparation, the samples were viewed under Hitachi S 3000 N, variable pressure scanning electron microscope. Each sample was coated with copper conductive tape as 4x4 matrixes which divided the samples in sixteen equal parts. Thus, a total of 16 images were taken, each one from the center of the parts. The images were later analyzed with ImageJ, image processing software. Segmentation of images was done by manually thresholding the image till the best contrast developed between background and points of interests. After thresholding, cracks were manually mapped for quantification of microcracks in sample. The results show that the average crack density of concrete sample cured at room temperature was 0.02844 mm/mm 2 while crack density of sample cured at elevated temperature was 0.027535 mm/mm 2 These results are in accordance with research study conducted by Ammouche et al., (2000) in which average crack density of undamaged samples was found to be 0.00271 mm/mm 2 From the above analysis, it could be concluded that room temperature cured sample had approximately 3% more microcracks than the sample cured at high temperature. Thus it could be said that there is no or very small difference between the quantity of microcracks found in concrete cured at room temperature and concrete cured at elevated temperature for concrete samples made without pozzlonic materials.

PAGE 66

CHAPTER 5 SUMMARY, RESULTS AND RECOMMENDATIONS Summary The purpose of this research was to identify if any additional microcracks are developed due to concrete cured at elevated temperature. To accomplish that following tasks were performed. Identification of the samples which had highest curing temperature and therefore may develop additional microcracking. Performed a literature review on methods available for diction of microcracks and finding out their advantages and disadvantages. Prepared specimens in order to view them under Scanning Electron Microscope for investigating microcracks. Processed images to quantify and compare the intensity of microcracks in both samples Samples were selected out of 20 different concrete mixes made with different percentages of pozzlonic materials, two AASHTO type II cement with two different placing temperatures. Phase I of this project showed that, concrete made from a cement with higher heat of hydration and without adding any pozzlonic material generate highest curing temperature. Therefore, a mix placed at 95F and made with AASTO type II cement having heat of hydration of 78.2 cal/g at 7 days without any pozzlonic material was selected for this study. Selection of appropriate sample preparation method was one of the important issues as improper sample preparation would cause additional microcracks in sample and lead to erroneous results. Most of the sample preparation methods used in past involved use of 55

PAGE 67

56 preheating the specimen for removal of pore water or putting specimen under vacuum for impregnation. There is a concern that by use of heat or vacuum, samples might develop additional microcracks. Epoxy impregnation method by Struble and Stutzman (1989) was used in this study as this method uses counter diffusion process to impregnate samples. Apart from supporting the microstructure of specimen, Epoxy impregnation also provides better contrast in cementitious matrix. Longer sample preparation time is the only drawback for performing this experiment. Images were acquired by a Hitachi S 3000 N, variable pressure scanning electron microscope (VP-SEM) in backscattered scattered electron (BSE) mode which is a high energy beam capable of distinguishing the particles present in matrix by variation of brightness in images. The specimens were divided as 4x4 matrixes by a copper conductive tape to aid viewing of points of interest from predefined positions. Total of 16 images were taken from each sample and analyzed in ImageJ software for quantification of microcracks. Images were converted to binary mode by manual thresholding operation. When features of interest (microcracks and voids) were correctly highlighted, the most distinct contrast between microcracks/voids and concrete background was obtained. After getting binary image, next task performed was to distinguish between voids and microcracks. As its known that microcracks are long elongated structure, thus length to width (also known as aspect/ form/ shape) ratio was applied. Objects having shape factor of 3.5 or greater was termed as microcracks and rest other objects were termed as voids. A software plugin Shape descriptor from ImageJ was used to identify voids in the images. Shape descriptor takes minimum and maximum pixel values and plots the void concentration in a separate image. After completing void

PAGE 68

57 removal from image, microcracks are mapped manually and counted. Results of quantification analysis were later converted into crack density per area to give comparative analysis. Results and Conclusion Results from this studies showed that, crack density of concrete sample cured at room temperature was 0.02844 while, crack density of sample cured at elevated temperature was 0.027535. Table 5-1 Comparative Analysis of Samples Sample Avg. Crack Length(m) Avg. Crack Density (m/m 2 ) Room Temperature Cured samples 1301.136 0.028440 High Temperature Cured samples 1259.659 0.027535 % Change 3.3% 3.28% This analysis indicated that there is very small difference between quantity of microcracks found in concrete cured at room temperature or concrete cured at elevated temperature for samples made with no pozzlonic materials and placed at higher temperature. Thus it can be concluded that high temperature curing conditions upto 180F may not play significant part in formation of additional cracks. Recommendations From research of this study, the following recommendations were made for conducting future work in this area. In future, studies involving mixes made with different percentage of pozzlonic materials, water/cement ratios, aggregates and placing temperature should be investigated to find out if mix design changes have any impact on density of microcracks in concrete samples.

PAGE 69

58 Epoxy solution used in this study had to be cured at 131F according to manufacturers instruction. Curing of the sample at high temperature might alter the microstructure of sample, thus efforts should be directed toward finding an ultra low viscosity epoxy solution, which does not require heating during its curing process. As cutting operations would create damaged zone in concrete microstructure upto 30% of blade thickness in sample, proper grinding and polishing operations should be followed to expose an undamaged surface. A more intrinsic grid system of conductive tape should be created to acquire more images from concrete sample which can be useful if automated image processing is used.

PAGE 70

LIST OF REFERENCES American Concrete Institute (ACI) Committee 116, Report SP-19, Farmington Hills, Michigan, 1995 Advances in Materials Problem solving with Electron Microscope Proceedings of Material Research Society, Volume 589, 1999 Ammouche, A. Breysse, D. Hornain, H. Didry, O., and Marchand, J., New image analysis technique for the quantitative assessment of microcracks in cement-based material, Cement and Concrete Research Volume 30, No.1, Page 25-35, 2000 Bisschop, J., and Van Mier, J.G.M., How to study drying shrinkage microcracking in cement-based materials using optical and scanning electron microscopy, Cement and Concrete Research Volume 32, No. 2, Page 279-287, February 2002 Florida Department of Transportation, Structural Design Guide, Section 3.9 Mass Concrete, Tallahassee, Florida Gran, H. C., Fluorescent liquid replacement technique: A means of crack detection and water-binder ratio determination in high-strength concretes. Cement and Concrete Research Volume 25, No 5, Page 1063-1074, 1995 Hanke, L. Variable Pressure Scanning Electron Microscopy for Nonconductive and Volatile Samples, Handbook of Analytical Methods, 1999 Hornain H., Marchand J., Ammouche A., Commene J.P., and Moranville M., "Microscopic observation of cracks in concrete-a new sample preparation technique using dye impregnation. Cement and Concrete Research, Volume 26, No.4 Page 573-583, 1996 Kjellsen, K. O., and Jennings, H M., Observations of microcracking in cement paste upon drying and rewetting by environmental scanning electron microscopy, Advanced Cement Based Materials Volume 3, No.1, Page 14-19, January 1996 Knab L.I., Walker H.N., Clifton J.R., and Fuller E.R., Fluorescent thin sections to observe the fracture zone in mortar, Cement and Concrete Research, Volume 14, No. 3, Page 339-344, 1984 Maggenti R., Mass Concrete Report, San Francisco-Oakland Bay Bridge East Spans Safety Project. 04-Ala/SF-80-Var, E.A. 04-012021, Available online at http://www.dot.ca.gov/hq/esc/tollbridge/SFOBB/EastSpan/012024/MaterialsHandout/cd2/concret1.pdf, last seen, Oct 12, 2004 59

PAGE 71

60 Marusin S., Sample preparation-the key to ESM studied of failed concrete, Cement and Concrete. Composites Volume 17, No.4, Page 311, 1995 Najjar, W.S., Aderhold, H.C., and Hover, K.C., Application of neutron radiography to the study of microcracking in concrete Cement, Concrete and Aggregates, Volume 8, No. 2, Page 103-109, 1986 Oberholster, R. E., Maree, H. and Brand, J. H. B, Cracked prestressed concrete railway sleepers: alkali silica reaction or delayed ettringite formation, Proceedings of the 9th International Conference on Alkali-Aggregate Reaction in Concrete, London, Page 740-749, 1992 Ollivier, J.P., Cement and Concrete Research, Volume 15, No.6, Page 1055-1060, 1985 Patel, H.H., Bland, C.H., and Poole, A.B., Microstructure of concrete cured at elevated temperatures, Cement and Concrete Research Volume 25 No. 3 Page 485-490, April 1995 Perkes, P., ACEPT W group, Department of Physics and Astronomy, Arizona State University, http://acept.la.asu.edu/PiN/rdg/elmicr/elmicr.shtml Last Visited October 2004 Ringot, E., and Bascoul, A., About the analysis of microcracking in concrete, Cement and Concrete Composites Volume 23, No. 2-3, Page 261-266, April 2001 Samaha, H.R., and Hover, K.C., Influence of microcracking on the mass transport properties of concrete, ACI materials Journal, Volume 89, No.4, Page 416-424, 1992 Slate, F.O., and Hover, K.C., Fracture Mechanics of Concrete: Edited by F.H. Whitmann, Published by Elsevier Science Page 85-93, 1984 Slate F.O., and Olsefski, S., X-Rays for Study of Internal Structure and Microcracking of Concrete, Journal of American construction Institute, Volume 60, No. 5, Page 575-588, 1963 Soroushian, P., Elzafraney, M., and Nossoni, A., Specimen preparation and image processing and analysis techniques for automated quantification of concrete microcracks and voids, Cement and Concrete Research Volume 33, No.12, Page 1949-1962, December 2003 Saint John, D., Poole A., and Sims A, Concrete Petrography-A Handbook of Investigative Techniques Published by Arnold, 1998 Struble L. J., and Stutzman P.E., Epoxy impregnation of hardened cement for microstructural characterization, Journal of Material Science Letters, Volume 8, Page 632-634, 1989

PAGE 72

61 Stutzman P, Scanning Electron Microscopy in Concrete Petrography, Materials Science of Concrete Special Volume: Calcium Hydroxide in Concrete (Workshop on the Role of Calcium Hydroxide in Concrete). Proceedings of The American Ceramic Society, edited by J. Skalny, J. Gebauer and I. Odler, November 1-3, 2000, Anna Maria Island, Florida, Page 59-72, 2001.

PAGE 73

BIOGRAPHICAL SKETCH Vishal Shah was born in Surendranagar, India, on March 21 st 1980, to Sanjay and Rekha Shah and grew up in Mumbai, located in Western India. He graduated in April of 1997 from Mithibai College and joined the Maharashtra Institute of Technology, University of Pune, to complete his undergraduate degree in civil engineering. After graduating with first class in 2002, Vishal continued his graduate studies at the M.E.Rinker Sr. School of Building Construction, University of Florida. During his stay at the University of Florida, Vishal worked as a graduate assistant for Dr. Abdol Chini. Upon his graduation in December 2004, Vishal plans to pursue a career in the exciting and challenging field of construction. 62


Permanent Link: http://ufdc.ufl.edu/UFE0008824/00001

Material Information

Title: Detection of Microcracks in Concrete Cured at Elevated Temperature
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0008824:00001

Permanent Link: http://ufdc.ufl.edu/UFE0008824/00001

Material Information

Title: Detection of Microcracks in Concrete Cured at Elevated Temperature
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0008824:00001


This item has the following downloads:


Full Text












DETECTION OF MICROCRACKS IN CONCRETE CURED AT ELEVATED
TEMPERATURE















By

VISHAL SANJAY SHAH


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE IN BUILDING CONSTRUCTION

UNIVERSITY OF FLORIDA


2004



























Copyright 2004

By

Vishal Sanjay Shah

































The thesis is dedicated to my loving parents Sanjay and Rekha and to my best friends
Reshma, Akshay, Piyush and Anushree for their unconditional support and caring
throughout my academic endeavors. It is with the support and love of my family and
friends that I am able to reach to my goals.















ACKNOWLEDGMENTS

The author would like to thank all of the members of his supervisory committee for

their help and ideas throughout this effort especially Dr. Abdol Chini, committee chair,

for his advice, patience and dedication as my advisor and also for the financial support

provided by him.

The author would also like to express gratitude to Dr. Larry Muszynski and Dr.

Andrew Boyd for their support and contribution as members of my advisory committee.

Acknowledgements are also owed to Tanya Reidhammer for her help in explaining and

conducting scanning electron microscopy. In addition, I would like to thank Richard

DeLorenzo for his help in the experimental process.

My thanks also go to everyone within M.E.Rinker Sr. School of Building

Construction for their continued help during the last two years. The author would like to

thank his friends at the University of Florida for their support and help they offered

during the research and writing of this thesis.
















TABLE OF CONTENTS

page

A CK N O W LED G M EN T S ....................................................................... iv

LIST OF TABLES .................. ........................................... ...... .. ........vii

LIST OF FIGURES .................. .......................................... ...... ........ viii

ABSTRAC T ............. ............ ................. ........ ........x

CHAPTER

1 IN TRODU CTION .............. ............................... ............ ..............1

Background......................................................... .................. ...............1
Curing of Concrete .............. ...... ................................2
H eat G generation in M ass Concrete..........................................................................3
Objective and Scope ................................................ .... ....5

2 LITERATURE REVIEW .............................................................7

Overview of Sam ple Preparation Techniques ........................................ .............. 8
Radiography and Acoustic Techniques................... .......................9
Replica Technique ........... .......... .... .......................... .................. 10
Impregnation Techniques .............................. ....................... ...10
Dye Impregnation Methods.............................. ..........................11
Epoxy Impregnation Techniques ............................................11
Thin Samples................................. ........ ........ 13
Impregnation by Wood's Metal .................................. .. ......14
High Pressure Epoxy Impregnation Technique ........................................ 15
Comparison of Sample Preparation Techniques............................ ...............16
M icroscopic Instrum ents ................................................ ............... 18
O ptical M icroscope ............................................... .............. 18
Scanning Electron M icroscope............. ................ .................... .......... 19
Variable Pressure Scanning Electron M icroscope .....................................21
Image Processing Techniques..............................................23
Summary ............................... ..................24






v










3. RESEARCH M ETHODOLOGY ......................................................... ... ............ 25

Introduction .................. .... ................. ........ .25
Sample Selection .............................................................25
Specim en Preparation M ethod.................................. ....................... ..............27
Procedure for Sample Preparation......................................... ..........28
Preparation for Image Processing............................ ......... ......... 32
Image Analysis .......................... ...........................33
Sum m ary .................. ........... ....................................... ........ 34

4. IMAGE PROCESSING AND ANALYSIS ..................................... .....35

Introduction .................. ............................................................... 35
Im age Processing T technique .................... ..........................................36
Sample 95BOOP Room Temperature Sample 1 ..........................................37
Sample 95BOOP Room Temperature Sample 2 .....................................41
Sample 95BOOP High Temperature Sample 1 ...................................... 45
Sample 95BOOP High Temperature Sample 2..................................... 49
Com parative A analysis of Sam ples................................................................... 53
Sum m ary ..................................................................................... 54

5 SUMMARY, RESULTS AND RECOMMENDATIONS ...........................................55

Sum m ary ................................................................................. 55
Results and Conclusion................ .... ............... 57
Recommendations.......................... .......... 57

LIST OF REFERENCES ..................................... ............... .....................59

BIOGRAPHICAL SKETCH ............................................... 62
















LIST OF TABLES


Table page

3-1 Mix Design Data................. ............................. 26

3-2 Adiabatic Temperature Rise Data.................................... .................. 27

3-3 Mixing Schedule for Epoxy........................................ 30

4-1 Crack Quantification Analysis for Mix 95 B 00 P RT-1 ................... ..............39

4-2 Comparison of Odd and Even Number Images.......... ............................40

4-3 Crack Quantification Analysis for Mix 95 B 00 P RT-2................................43

4-4 Comparison of Odd and Even Number Images................... ...............44

4-5 Crack Quantification Analysis for Mix 95 B 00 P HT-1.................................47

4-6 Comparison of Odd and Even Number Images.................................48

4-7 Crack Quantification Analysis for Mix 95 B 00 P HT-2................................51

4-8 Comparison of Odd and Even Number Images............................52

4-9 Comparative Analysis of 95 B OOP RT1 and 95BOOP RT2 .....................................53

4-10 Comparative Analysis of 95 B OOP HT1 and 95BOOP HT2.................................53

5-1 Comparative Analysis of Samples....................... ... .....................57
















LIST OF FIGURES


Figure page

2-1 Working of Scanning Electron Microscope .................................... .....19

3-1 D iam ond W after Saw ..............................................................29

3-2 U ltra Low Epoxy K it by SPI .............................................. ............... 30

3-3 Metaserv Grinder and Polisher............... ....... ......... ..........................31

3-4 Polished Sam ple ................................................. ....... .31

3-5 D iam ond Suspension Paste.................................................... 32

3-6 Prepared Samples ....................................... .... ... .... ...............32

3-7 Specim en w ith Copper Tape...................................................... 33

3-8 Specim ens with 4x4 m atrixes .......................................................... .............33

4-1 Representative Images of Mix 95 B 00 P RT-1 before Image Processing ..................37

4-2 Representative Images of Mix 95 B 00 P RT-1 after Image Processing....................38

4-3 Variations in Crack Density of 95 B 00 P RT1 ................. ................. ...........40

4-4 Representative Images of Mix 95 B 00 P RT-2 before Image Processing .................41

4-5 Representative Images of Mix 95 B 00 P RT-2 after Image Processing.....................42

4-6 Variations in Crack Density of Sample 95 B 00 P RT2.......................................44

4-7 Representative Images of Mix 95 B 00 P HT-1 before Image Processing................45

4-8 Representative Images of Mix 95 B 00 P HT-1 after Image Processing.....................46

4-9 Variations in Crack Density of Sample 95 B OOP HT- 1.......................................48

4-10 Representative Images of Mix 95 B 00 P HT-2 before Image Processing................49

4-11 Representative Images of Mix 95 B 00 P HT-2 after Image Processing................ 50









4-12 Variations in Crack Density of Sample HT 2......................................52
















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for Master of Science in Building Construction

DETECTION OF MICROCRACKS IN CONCRETE CURED AT ELEVATED
TEMPERATURE

By

Vishal Sanjay Shah

December 2004

Chair: Abdol Chini
Cochair: Larry Muszynski
Major Department: Building Construction

Proper curing is a necessity for durable concrete. Exothermic reaction during curing

process generates heat in concrete creating thermal differential between outer and inner

core of concrete. In mass concrete pores at Florida Department of Transportation,

internal temperatures of concrete have been reported as much as 1800F to 2000F, which

leads to a concern that higher curing temperature might initiate higher number of

microcracks during its early age.

Phase I of this study indicated the presence of higher curing temperature in

concrete which is made from cement having higher heat of hydration, higher placing

temperature and does not contain any pozzlonic material in mix design. For comparison

of curing conditions, four samples were selected from this type of concrete for

quantification of microcracks. Out of these four samples, two of the samples were cured

at room temperature and the other two samples were cured at high temperature.









A literature review revealed that various specimen preparation methods have been

used in past for viewing concrete microstructure. However, a counter diffusion method

with epoxy impregnation was used in this study as this method does not require any pre

heating of specimen and pressure or vacuum in sample preparation. After specimen

preparation, samples were viewed under variable pressure scanning electron microscope,

and a total sixteen images from each sample were taken.

Image analysis was done with standard image processing software by converting

images into binary mode by thresholding it until features of interest (microcracks and

voids) were correctly highlighted. After differentiating voids from cracks by shape

analysis method, microcracks were manually mapped and quantified by calculating crack

density in each image. A statistical analysis was performed to indicate representative

nature of image quantity taken in sample.

The results showed no significant difference between the number of microcracks

present in samples cured at room temperature and high temperature in concrete made by

cement having heat of hydration up to 76.2 cal/g, placing temperature of 950F and

without any pozzlonic material. Thus it could be suggested that high temperature curing

upto 1800F does not play any significant role in the development of microcracks.















CHAPTER 1
INTRODUCTION

Background

Concrete is one of the oldest engineering material and certainly one of the key

integrants in development of modern infrastructure. When properly designed and

carefully produced with good quality control, concrete is inherently a durable material.

However, under adverse environment and exposure conditions, concrete interacts with its

surrounding environment and undergoes chemical changes. These processes involve

movement of water or other gaseous substances transporting aggressive agents through

pore structures of concrete, thus making concrete potentially vulnerable to deleterious

attacks.

Reinforced cement concrete structures exposed to hostile environment require highly

durable concrete to provide long lasting performance with minimal maintenance. Modern

day infrastructure has necessitated construction of numerous bridges, dams, tunnels and

concrete pavements. Massive concrete structures of this nature would contain large

amount of concrete, most of these structures could be termed as "mass concrete" work.

According to American Concrete Institute Committee 116, report SP-19 (1985), mass

concrete is "any volume of concrete with dimensions large enough to require that

measures be taken to cope with generation of heat of hydration of the cement and

attendant volume change to minimize cracking." The Florida Department of

Transportation (2004) further specified dimensions of mass concrete as "when the









minimum dimension of the concrete exceeds 36-inches and the ratio of volume of

concrete to the surface area is greater than 12-inches, provide for mass concrete."

Curing of Concrete

Curing is widely perceived as being an important factor in achieving durable

concrete structure. Curing ensures that concrete maintains a sufficient amount of water to

reach a proper degree of hydration in the surface which is required to continue chemical

reaction between cement and its constituents. Proper Curing allows hydration of cement

to continue which is expected to reduce capillary porosity, thereby strengthening the

concrete to desired strength and increasing its resistance to penetration by aggressive

gases and liquids. The higher the curing temperature, the faster is the reaction between

cement and water, thus shorter setting time. It can reduce durability and useful life span

of structure by hardening concrete faster, reducing strength and causing excessive

seepage. Curing measures are necessary to maintain a satisfactory moisture and

temperature condition in concrete, because internal temperature and moisture directly

influence early and ultimate concrete properties.

During curing process, concrete element undergoes exothermic chemical reaction.

This heat generation is of little consequences and can be easily dissipated in surrounding

environments. However, in mass concrete, heat generated within the body of member

does not get dissipated easily. This temperature rise causes expansion while the concrete

is hardening. Temperature in inner core of concrete is much greater than outer core of

concrete due to loss of heat to surroundings by exposed surfaces. If the temperature rise is

significantly high and concrete undergoes non-uniform or rapid cooling, stresses due to

thermal contraction in conjunction with structural restraint can result in cracking before

or after the concrete eventually cools to surrounding temperature. Due to faster cooling of









outer surface, concrete tends to shrink and causes tensile stress in outer core. When these

tensile stresses are greater than tensile strength of concrete, cracks are developed.

Heat Generation in Mass Concrete

Heat generated in mass concrete increases temperature of concrete to a varying

degree depending on surrounding temperature, size of structural element, mix design

configurations, and curing temperature. Current steps to counter high temperature rise in

concrete includes choosing a low heat cement, external curing system, and cooling of

aggregates prior to mixing. Constructing large structures in small stages, or lifts, can be

done, but it is slow and the finished structure is weakened at the cold joints in the

concrete. Replacement of cement by pozzlonic materials results in a decreased

temperature rise in fresh concrete, which is a particularly important issue in mass

concrete where a large temperature rise can lead to cracking.

Currently, any mass concrete pour for FDOT requires thermal control plan which

describes the measures and procedures intended to maintain temperature differential of

350F (200C) or less between interior and exterior portion of designated mass concrete

element during its curing phase; there is however, no requirement on the maximum

temperature rise within the mass concrete. Internal temperatures of 1800F to 2000F have

been reported in mass concrete pours in FDOT projects. There is a concern that such

higher curing temperature would initiate microcracks during its early age which would

accelerate deterioration and compromise durability of concrete by microcracks acting as a

conduit of harmful minerals and gases, causing reduction in durability of concrete.

Study conducted by Maggenti (2001) of California Department of Transportation

indicated no irregular, externally visible cracks in mass concrete work. However, it

would be important to find out changes occurring at the micro level, which will obviously









grow in future and hamper durability of concrete. The capillary effect of the

microcracks increases probability of exchange of substances between the environments

and microstructure of concrete. This exchange of substances may lead to transport,

solution and enrichment processes in concrete microstructure which can damage the

concrete by increasing the volume.

Concrete, when fully hardened, contains a variety of micro structural features on a

variety of scales. To the unaided eye, it appears from a polished section to be essentially a

two phase composite-a matrix of hardened mortar plus coarse aggregate that occupies

about 75 % of its volume. Recent studies of concrete microstructure have played

important part in understanding behavior of concrete matrix. It is necessary to study the

effect of elevated temperature rise in microstructure specially comparing it with normal

room temperature cured samples and find out if any additional microcracks have been

developed. Formation of microcracks in the surface can lead to decrease of durability,

especially in exposed outdoor structures as bridge superstructures but also concrete

pavement and structures in direct contact with the ground.

Many methods ranging from destructive to non-destructive have been used in past

to understand microstructure of concrete. A problem common with many of the

techniques used to examine the microstructure of concrete is that they can alter the

sample. For example, many techniques require that the specimen be prepared by

mechanical polishing or to dry concrete specimen before examination by putting in

vacuum or heat. Such Specimen preparation methods would invariably alter the original

artifacts and induce additional microcracks.









Advancement in science has assisted in developing equipment and techniques

capable of studying the microstructure which facilitate in-depth study and analysis of

specimen. While many existing techniques and equipment require drying the specimen

and in a way altering the specimen, evolution of Variable Pressure Scanning Electron

Microscope (VPSEM) techniques allows the study of unaltered wet specimen in a low

vacuum.

Objective and Scope

This study is an extension of a research project "Adiabatic Temperature Rise in

Mass Concrete" which is aimed to develop adiabatic temperature rise curves for concrete

mixes used in FDOT projects and investigate properties of high temperature cured

concrete. Total of 20 mixes of concrete were made as a part of the study by changing

various parameters such as the source of cement, the percentage of pozzlonic materials

and the initial placing temperature. Concrete samples were kept in heating curing

chambers to simulate conditions of mass concrete for monitoring rise in temperature. At

14 and 28 days, samples kept in heating chambers as well as samples in normal room

temperature were tested for compressive strength.

The objective of this part of the study is to focus on detecting and quantifying

additional microcracks developed due to concrete cured at elevated temperature. This

would be done by comparing room temperature cured samples with samples cured at

elevated temperature to find out if there is any difference in quantity of microcracks.

The Study would be performed as given below:

1. A literature search to find out current methods used in identification of microcracks
and sample preparation techniques. It should also include possible image
processing techniques to enhance analysis capability.






6


2. Prepare concrete specimens according to an identified sample preparation
technique and view them with selected Microscopy equipment.

3. Use Image Processing Techniques for further identification and quantification of
microcracks. Compare the quantity of microcracks in samples cured at elevated
temperature with those cured at room temperature to determine if any additional
microcracks have been developed due to high temperature curing.














CHAPTER 2
LITERATURE REVIEW

Presence of voids, cracks, and other defects play an important role in determining

the mechanical performance of the concrete. If concrete surface is pre-damaged by

microcrack formation in the near-surface area and exposed to open weathering, capillary

effect would increase the transfer of substances from the environment and increase the

concrete volume which can lead to several mechanisms capable of negatively influencing

durability of concrete.

Studies have indicated that the development of cracks and connected crack network

contributes to the increase in permeability and diffusivity of concrete. The presence of

preexisting microcracks of 10 [im or greater on the surface can lead to reduced durability,

especially of exposed outdoor structures such as bridge superstructures, but also concrete

pavements and structures in direct contact with the ground. Patel et al., (1995) found that

concretes cured at higher temperatures exhibited a coarser microstructure than that of a

typical concrete cured at 200C, particularly with respect to ettringite. The presence of

microcracks will increase moisture mobility within the concrete, which may produce

density gradients within the matrix leading to further microcracking. The domination of

the matrix by a network of microcracks is also conducive to the formation of secondary

ettringite.

Recorded curing temperature of 1800F 2000F inside core of FDOT Mass concrete

elements has raised concerns over the initiation of microcracks. Slate and Hover (1984)

defined microcracks as cracks having width of less than 100 [im, while Jansen defined









microcracks as extended faults with a width of less than 10 rim. According to Kjellsen

and Jennings (1996), these differences in definition appeared to be closely related to

experimental techniques and to the orientation of the studies. For the scope of this study,

reasonable and practical interpretations of microcracks were set as cracks having widths

less then 10 jim. While macrostructure of concrete can be seen unaided, microstructure

(200 rm or smaller) must be observed with the aid of a microscope. It would be fair to

say, as the documented research indicates, that cracks and crack propagation is

understood well from macroscopic point of view but unavailability of precise specimen

preparation techniques hinder studies of concrete microstructure.

Overview of Sample Preparation Techniques

Sample preparation is a key to microscopic analysis of concrete. Proper preparation

of concrete samples so that microcracks and voids develop a distinct contrast against the

body of concrete is a pre-requisite for the application of the modern day image processing

and analyzing techniques (Soroushian et al., 2003). Poor preparation methods can lead to

erroneous diagnoses of problems associated with concrete specimen. According to

Hornain et al., (1996), ideal specimen preparation techniques should not induce any

cracks during the preparation of samples, thus techniques which involves prior drying of

specimen should not be used. The ideal specimen preparation should be simple,

economic, rapid, and should be able to detect very fine cracks. And finally, a sample

preparation technique will not be useful without accurate image analysis and processing

techniques. The sample preparation would also depend on the type of equipment used for

image collection, required resolution of image, and objective of the study.









For the past few decades, studies have been performed to find out a method which

fulfills all definitions for ideal specimen preparation technique. New equipment and

methods developed by technological advances have helped researchers in their analysis

of concrete specimens. As reported by Ringot and Bascoul (2001), two methods are

necessary for the characterization of microcracks, one for sample preparation and one for

quantification of cracks. Sample preparation techniques also depend on the type of

microscopy used (S. Marusin 1995). Acoustic method, ultra sonic and laser sparkle

methods are useful for studies related to crack propagations but they are unequipped to

monitor initial state of samples crack quantification. Thus, the rest of the sample

preparation techniques can be briefly classified in three main categories: radiography

techniques, replica techniques, and impregnation techniques.

Radiography and Acoustic Techniques

Earlier Methods by Slate and Olsefski (1963) described the use of X-radiography to

study the internal features of concrete and crack formation process of mortar and concrete

sample. Thin samples of 4 mm thickness were exposed to X-rays flux normal to the plane

of the sample. Cracks were identified from normal constituent of concrete as penetration

of X-rays was greater in surrounding area. A comparative study carried out by Najjar et

al., (1986) indicated that due to poor resolution of images, X-radiography systematically

overlooks thin cracks. Neutron radiography described by Samaha and Hover, (1992),

which is principally quite similar to X-ray radiography approach, is in fact a better

alternative for increasing image resolution. However, in this process the samples need to

be air dried beforehand in order to impregnate specimen with gadolinium nitrate, which

in turn might induce microcracks in samples.









Replica Technique

Replica method developed by Ollivier in 1985 is one of the methods which do not

require any pre drying of the sample. In this method, the film of acetylcellulose is placed

on concrete specimen by methyl acetate. The film is taken off after the solvent (methyl

acetate) is evaporated in air and observed under optical or scanning electron microscope.

As there is no disturbance made to the specimen surface itself, crack propagation studies

can be carried out with replica method. However, in order to peel the acetylcellulose film

safely, the area of film has to be limited to 2 cm2. Thus too many replicas need to be

prepared for covering significant portion of specimen, which could be time consuming.

Impregnation Techniques

Impregnation techniques are one of the oldest techniques used to study

microstructure of concrete. In this method, concrete samples are impregnated with dye or

epoxy to facilitate detection and identification of cracks. In order to reduce difficulty in

viewing cracks in dense microstructure, fluorescent dye was used by Knab et al., (1984).

However, pre drying of specimen in an electric oven before impregnation, which could

alter the specimen condition, remained a major inhibition to this process. Methods

developed by Struble et al., (1989) and Gran (1995) overcome drying of specimen by

using counter diffusion method for replacement of pore water by dye impregnated

organic solution. In this method, a thin sample of 15 mm was cut and kept in solution

made by dissolved dye and ethanol. After 4 days of counter diffusion process between

pore water present in concrete specimen and dye induced ethanol, specimen was taken

out of the solution to remove excess dye present on the surface by polishing. However,

this process could take several days and increases the duration of the test.









Dye Impregnation Methods

Hornain et al., (1996) modified dye impregnation techniques to reduce the time

taken to prepare specimen. In this Method, 4 grams of a water soluble red powder,

commercially known as Irgacete was dissolved in 100 ml ethanol solution. After keeping

specimens for 5 minutes in dyed solution, a second impregnation was done. After

completing these 2 steps, the specimen was taken out and polished under water with 6

lm diamond paste to remove excess dye. In order to achieve proper polishing, the

specimen was again polished with 3[im and 1 [im of diamond paste. The polished

specimens were observed under optical microscope at 100X. The study reported that

cracks of 1 .im width and less could be easily distinguishable, providing contrast level

between specimen and cracks. During sample microscopic studies, cracks going through

the hydrated cement phase were distinguished from discontinuity at paste-aggregate

interface. However, the author reported that dye impregnation method was not useful in

finding out cracks in highly porous areas.

Epoxy Impregnation Techniques

Struble and Stutzman (1989) developed a new technique which involved three steps

procedure to replace pore water from concrete sample. Epoxy impregnation would not

only support the microstructure of specimen by filling the voids and cracks up on curing,

but also support fragile pores and matrix phases by restraining it against disintegration

during the different stages of preparation. This could be a major issue if the samples are

to be viewed under scanning electron microscope which would generate high pressure

inside the chamber. Another advantage of using epoxy is to enhance contrast between the

pores, hydration products and cementitious material. The selection of the type of epoxy

depends on the objective of the study. Low viscous epoxy was used for relatively highly









permeable materials or cementitious powders, while ultra low viscosity epoxy aids in

rapid infiltration for less permeable cement pastes and concrete samples.

The first step for this sample preparation method was to cut the specimen of 1 inch

size and immerse specimen in ethanol solution at 600C. The blades of saw were

immersed in propylene glycol to prevent it from drying. A companion specimen was

prepared by Struble and Stutzman (1989) which was kept in an ethanol solution with red

dye to see the extent of time taken for the replacement of ethanol in pore structure. For a

given concrete sample, the time taken for replacement depends essentially on the

thickness of the sample. After 50% of alcohol replacement in depth by visual inspection

in companion specimen, the solution is replaced by epoxy at room temperature and cured

according to the manufacture's specifications. After proper curing, the samples were

polished and lapped with abrasion papers and diamond paste in decreasing order and

viewed under Backscattered Electron imaging technique of Scanning Electron

Microscope.

The samples should be grinded and polished perfectly in order to remove raw saw

cuts on the specimen. Insufficient polishing would leave a disturbed surface on specimen

and the cracks would not be easily discernible during the microscopic observations. On

the other hand, an excessive polishing would contribute to the removal of microstructure,

and could affect the subsequent quantification operations. It was believed that higher

heating temperature would hasten the process of pore replacement. Higher temperature

for heating was not used in subsequent studies due to limited measure of quantifying the

pore water alcohol exchange and quick evaporation of alcohol. Modified epoxy









impregnation method does not require any heating or vacuum on samples and would not

induce any specimen preparation related microcracks.

However as reported by Soroushian (2003), epoxy impregnation can highlight more

porous areas of cement paste, but generally do not yield crisp boundaries and sharp

contrast between microcracks and air voids against body of concrete. A new two stage

technique was developed as Ink- Epoxy Impregnation. In the first stage, fine capillary

pores were first impregnated with parker blue ink and then microcracks and voids were

impregnated with fluorescent ultra low viscosity epoxy solution. Specimen were cut to

20-30 mm or 0.8-1 inch, and lapped by abrasive liquid under 3 PSI pressure After

cleaning the specimen for loose debris by first water bath and then ultrasonic bath,

specimen was dried at 1400 F for 3-4 hours. Ink impregnation for first stage was done by

keeping the container with ink and specimen under nitrogen pressure at 280 psi for 18-24

hours. Specimen was removed and heated at 1400 F for 24 hours to let ink dry.

Stage 2 was performed for fluorescent epoxy penetration in samples. The samples

were kept in a vacuum chamber with 0.38 psi for 1 hour. Epoxy solution was mixed and

poured in a glass container and kept in nitrogen pressure of 0.0133 psi for 3-4 hours. To

cure epoxy, samples underwent heat in an electric oven at 150 OF for 18-24 hours

following by polishing operations. While comparing with normal epoxy replacement

technique, ink-epoxy replacement technique gives better results. However, the

application of heat as well as vacuum pressure could instigate microcracks in samples.

Thin Samples

A study conducted by Anderson (1989) introduced preparing thin section of

concrete samples. In this method, a larger sample of 45 mm x 30 mm x 20 mm was cut

from a section of concrete, selected from the portion which was few centimeter away









from to be investigated area. The cut section was kept under ethanol for 12 hours to

reduce its tendency for cracking. The concrete specimen was kept in a vacuum oven at

85 'F to 90 OF to dry for two hours. A homogenized mixture of low viscosity epoxy

prepared with 1.1% by volume of dye was mixed in magnetic stirrer for 24 hours in

advance. Dried concrete specimen was kept in the solution for 1 hour for vacuum

impregnation. It was observed that concrete was impregnated by 1-2 mm in depth. Stage

2 covered mounting of glass slide on the selected face of specimen with Ultraviolet (UV)

hardening glue. The specimen was then grinded by a 15-20 [im diamond disc in order to

get smooth surface. After cleaning and drying the specimen, it was re-impregnated with

epoxy. After curing, a thin sample was cut from the specimen by diamond saw and final

grinding and polishing was done. Although the results from this method are widely used

in European counties, lengthy sample preparation time and dependability on skilled and

experienced technician were primary requirements.

Impregnation by Wood's Metal

Nemati (1997) developed a new approach in impregnation techniques using

Wood's metal instead of epoxy. The idea of this research was to preserve the

microstructure of the concrete samples which are kept under compressive stress to

analyze microcracks as they exist under loading. In this three-phased study, cracks were

introduced in the concrete specimen by a new test setup which allowed the application of

axial stress on concrete and impregnation by wood's metal simultaneously. Wood's metal

was impregnated within the sample for providing stability and better contrast for the

identification of microcracks. However, sample preparation method involved drying of

concrete cylinder by 109.50F as well as gradual heating of test assembly by 1220F, 1670F

and 204.60F for unified initiation of molten wood's metal in the specimen. As explained









earlier, heating or drying operations before impregnation could introduce microcracks in

concrete. Author reported that in no loading conditions, few cracks were observed which

could be present due to cracks introduced in sample preparation or drying shrinkage.

Wood's Metal impregnation technique was further developed by Soroshian et al.,

(2003) for concrete samples without initiating microcracks through stress introduction. In

this method, 2 inch thick slices were cut and washed to remove any loose debris attached.

The cleaned sample was kept in electrical oven at 150 OF for 24 hours to remove water

present. The dried sample was than kept in a steel mold for impregnation with Wood's

metal. To facilitate impregnation and liquidification of Wood's metal, steel mold is kept

inside a vacuum pressure chamber by 0.95 psi for 30- 40 minutes. After keeping the oven

temperature for 200 'F for 1-2 hour, the air vacuum was replaced by nitrogen pressure of

280-300 psi for 3-4 hours. The specimen was allowed to cool down, followed by cutting

a 6 mm sample and then the surface was prepared for viewing under Backscattered

technique in Environmental Scanning Electron Microscope (ESEM). Results indicated

replacement by Wood's Metal gave desired contract for microcrack identification against

concrete surface. However, heating and vacuum pressure application on concrete would

develop additional microcracks which can be quantified with original microcracks.

High Pressure Epoxy Impregnation Technique

Previous methods involving florescent epoxy replacement being time consuming,

Chen (2002) applied high pressure of> 20 bars to florescent epoxy impregnation in

concrete sample. Samples of 2x2x1.5 cm were prepared and after initial drying of 176 'F

for 2 days, samples were kept in a steel cylinder in epoxy resin at least 5 cm higher than

the top surface of specimen. The top and bottom of steel cylinder were blocked by

Teflon, desired pressure ( 145,725, 1160, 1450, 2900 and 5800 psi) was applied on the









blocks through a hydraulic piston and held under pressure for 90 minutes followed by

curing of epoxy by heating it upto 1220F for 24 hours. This robust impregnation

procedure results in a uniform and vast distribution procedure of the florescent epoxy

resin throughout the material structure. Prepared samples were viewed under optical

microscope using UV light. The experiment proved high pressure impregnation to be a

more effective and faster method to induct epoxy resin in cementitious materials. This

method show higher brightness and more effective penetration of epoxy in material and

when pressure and vacuum impregnated samples kept in optical microscope to analyze,

high pressure impregnated samples took less time for reaching a threshold value of

photon. Pressure impregnation allows higher tensile and flexural strength in samples

compared to vacuum process, thus implying greater stability of cementitious specimen

during the preparation phase. However, no reports clearly show or address the issue of

the introduction of cracks during sample impregnation.

Comparison of Sample Preparation Techniques

Microscopic image is a combination of original material from which the specimen

was taken, cumulative effects of all the procedures required to prepare the specimen for

examination, the examination technique itself and our interpretation of the image.

Specimen preparation method depends on the type of microscope used for viewing the

specimen. The review of all above techniques reveals that there is no standard specimen

preparation for viewing microcracks in cement and cementitious materials. Almost all

researchers identified pre-drying of specimen as a major source of developing cracks

prior to microscopic examination. However, most of the studies dried specimens in order

to remove pore water of concrete to facilitate the introduction of dye, epoxy or wood's

metal. While some techniques required polished concrete section or broken fracture









surfaces, some required thin sections dried at 122 'F or oven dried in NO2 atmosphere.

Some samples are cut with a diamond impregnated wire saw without a coolant, while

some samples prepared by epoxy impregnated are cut by diamond grit or lapped by

alumina grit and then polished by diamond grit/ abrasive liquid of decreasing order

(Marusin S., 1995).

Some researchers believe that application of improper sample preparation methods

would erode and destroy existing microstructure. It can also introduce elements which are

not part of the original specimen which would lead to false diagnosis of damage

mechanism. Most of the specimen preparation techniques requires to dry sample till a

certain degree which would affect the pore structure and can induce more microcracks.

As reported by Hornain et al., (1996), crack density increases with intensity of drying.

Thus such methods would not be able to give true quantitative analysis of microcracks.

Vacuum impregnation methods results in excessive drying of the sample and therefore

cracks are introduced during sample preparations (Chen 2002). Techniques involving

mechanical polishing to produce either a flat surface or thin specimen could be

problematic for concrete as different phases of concrete polish at different rates. Similarly

putting concrete samples into a vacuum would lead to loss of water and change of

structure.

Polishing and lapping operation to smooth the specimen surface would lead to

spreading of components over surface, thus preventing effective use of dot map studies

for elements. Polishing would also destroy delayed ettemdite crystals and cracks

originating from it. While pressure or vacuum would introduce more cracks in the

sample, original cracks would heal or widened by epoxy impregnation techniques









(Marusin, 1995). However, sawn and unpolished approach could be better if the study is

related to DEF analysis but for proper crack identification and quantification, sample

preparation techniques would be required.

In brief, Bisschop and Van Mier (2002) and other studies have indicated

impregnation of whole sample and then cutting thin samples from fully impregnated

samples would introduce less microcracks. Specimen preparation techniques involve

cutting, drying, lapping, grinding and polishing. Improper handling of these operation

results in induction of additional microcracks. For the scope of this study, epoxy

impregnation technique developed by Struble and Stutzman (1989) was performed as it

did not involve any drying of samples in any form. In this two-stage counter diffusion

process, pore water is replaced by ethanol, and ethanol is substituted by epoxy without

involvement of preheating or use of pressure techniques.

Microscopic Instruments

Optical Microscope

Optical microscopy is one of the favored techniques for concrete petrography. In

this instrument, image is viewed in full color and generally at a lower magnification, thus

making it a more suitable technique for observing features at the millimeter scale and

larger. Concrete and cementitious specimens prepared by thin-sectioning and fluorescent

microscopy techniques can be viewed under optical microscope with the use of reflected

ultraviolet lights. The image acquisition is done by placing a Tri-CCD camera which is

linked to a personal computer acquisition system. However, Optical microscopes have

shallow depth of field and limited resolution capacity, so a highly smooth and polished

surface is required to produce a focused image under this technique. Furthermore, the










intensity of the vacuum-impregnated sample is not stable after 2 min of exposure to UV

light, and decreases in a continuous fashion with the elapse of time (Chen 2002).

Scanning Electron Microscope

Scanning electron microscopy/ x-ray microanalysis is comparatively more useful in

studies which investigate quantification of microstructural properties such as microcracks

and voids. Scanning electron microscopes has greater depth of field and high spatial

resolution which produces focused images of poor specimens and analytical data for

elemental composition analysis for the features seen on those images. As scanning

electron microscopy and optical microscope petrography are different tools, different

sample preparations are used for these different approaches.

As described in figure 2.1, SEM scans a focused beam of electron across the

specimen and measures any of several signals resulting from electron beam interaction

with surface of concrete specimen. Images are monochrome in nature because they

reflect the electron or x-ray flux resulting from beam/specimen interaction. Three major

types of signals generated as a result are Secondary electrons, backscattered electrons and

X-rays analysis.





1e-sys(= l









specimen d&t1or
to pumps

Figure 2-1 Working of Scanning Electron Microscope, (Perkes, 1999)









Secondary Electrons (SE) are low energy electrons resulting from an inelastic

collision of a primary beam electron with an electron of a specimen atom. Because of

their low energy, they are readily absorbed and only those produced near the surface

escape, resulting in an image of surface topography. SE imaging is principally applied in

the examination of early age paste microstructure, high magnification imaging of

microstructural features and for examining texture. Knowledge of the morphological and

compositional characteristics of the hardened cement paste constituents is invaluable for

their identification. As the hardened cement paste matures, filling of the void spaces

eliminates the well-formed crystals and backscattered electron and x-ray imaging are

more useful in examination of these microstructures (Stutzman, 2001)

Backscattered Electron are high energy beams and capable of reflecting difference

in atomic numbers. It can distinguish between the particles present in matrix on the basis

of the variation in brightness of their images. In Backscattered Electron imaging

technique contrast is generated by different phase compositions relative to their average

atomic number and is observed by differential brightness in the image. SEM analysis

using backscattered electron and X ray imaging requires highly polished surface for

optimum imaging and X ray microanalysis.

X radiation is produced when a specimen is bombarded by high energy electrons.

X-ray microanalysis systems generally employ an energy dispersive detector with the

other detector type being a wavelength detector. The x-ray energy level is displaced as

the number of counts of each energy interval and appears as a set of peaks on a

continuous background. X-ray radiation allows elements compositions to be obtained as









the qualitative analyses prints like a dot map images. Each element is identified on a

continuous spectrum by the position of its pick.

Any sample can be examined by SE imaging but best use of BE imaging is for

examining flat, polished surfaces (Struble and Stutzman, 1989). One of the disadvantages

of Scanning Electron Microscope is that samples must be coated to allow discharge of

electron build-up on the examined surface. A thin metal coating decreases build-up of

negative charge by forming a conducting path for electrons to avoid distorted images. If

images are accompanied by X-ray analysis, carbon coating can not be used if a

carbonation process is to be studied. A gold coating gives a peak almost at a same

position as sulfur (Oberhostler 1992). Thus, in some samples, it might be appropriate to

coat the specimen due to concerns for altering the microstructure. High vacuum pressure

inside the specimen chamber is also one of the factors which limits the use of SEM as

some samples might break or disintegrate under high vacuum pressure.

Variable Pressure Scanning Electron Microscope

Hanke L, (1999) described that many applications where SEM/EDS evaluation

could be useful involve samples that are not electrically conductive. These samples have

traditionally required pretreatment, by coating with a conductive film, before SEM

examination. Nonconductive samples are subject to a buildup of electrons on the

examined surface. This buildup of electrons eventually causes scattering of the incoming

electron beam, which interferes with imaging and analysis. Furthermore, samples that

contain substantial water or other materials that volatilize in high vacuum also present

challenges for SEM examination. These samples require controlled drying to allow the

SEM chamber to reach high vacuum and to prevent deformation of the sample at the

SEM vacuum.









As the SEM is increasingly used for routine evaluations, there is increasing demand

for examination without pretreatment. An answer to the problems of charging and

volatile samples is the development of scanning electron microscopes that operate

without exposing the sample to high vacuum. These microscopes are referred to

alternately as environmental, low-vacuum, or variable-pressure SEM. In variable pressure

SEM, the chamber at the electron gun is maintained at high vacuum, while a controlled

amount of gas is allowed into the sample chamber. A fine aperture separates the gun and

sample chambers to prevent excessive gas entrance into the gun chamber. Separate

vacuum systems control the vacuum in the sample chamber and at the gun. The

advantages of a higher pressure in the sample chamber are obvious for wet and volatile

samples. The higher pressure decreases the rate of volatilization. This decreases the

drying and deformation of wet samples.

For nonconductive samples, the advantage of higher pressure is less obvious. When

gas molecules in the sample chamber are struck by the electron beam, the gas is ionized.

These positive ions are attracted to and neutralize the negative charge build-up on the

nonconductive specimens. By controlling the pressure in the sample chamber, the number

of gas molecules intercepting the electron beam is maintained at a level that is sufficient

to prevent charging, but does not deflect the beam sufficiently to prevent imaging and

microanalysis.

Thus, the use of variable pressure scanning electron microscope holds advantage

over traditional scanning electron microscope as no coating is required over the samples

and samples with moisture could also be imaged.









Image Processing Techniques

Important steps for any microstructural images are to acquire the image and to

process microstructure image to get a quantitative analysis of the sample. For this study,

quantitative analysis is to identify and measure the number of cracks and crack density

over number of images. The image processing should be able to differentiate between

voids and cracks, and boundary cracks around aggregates or between phases. Darwin

(2001) researched on quantification of microcracks of cement mortar samples by

backscatter electron technique of scanning electron microscope. He reported that the use

of edge detection, gradients and other filters failed to identify cracks passing through low

density phases as its intensity depends on adjacent and underlying phases. In his

approach, the identification of cracks was based on local changes in grey level. A line

scan was taken perpendicular to the images for identification of potential cracks.

Perimeter to squared area was taken to differentiate between cracks and voids. Phase

identification was also carried out based on pixel intensity. He conclude that as grey level

of image is affected by the density of adjacent and underlying phases, cracks identified

based on change in grey level would not be correct. Thus the correct practice for crack

identification should establish minimum gradient in grey level adjacent to cracks.

Soroshian (2003) used segmentation of images from grey level to binary images by

thresholding approach for viewing microcracks in samples. The research compared

manual thresholding operations with three automated thresholding operation such as

factorization, entropy and moment. In manual thresholding, low threshold level was set to

zero or default for auto thresholding and best threshold level image was determined. The

optimum high threshold level was determined by comparing original gray image and the

image after the application of manual threshold on binary image. When features of









interest (microcracks and voids) were correctly highlighted, the most distant contrast

between microcracks/voids and concrete background was obtained. After training the

program with images of different resolutions, three automated thresholding operation

were performed and compared results with manual thresholding. The study reported that

there was no significant difference between manual thresholding and automated

thresholding operation by factorization method. There was a difference of 5% and 2%

respectively, for entropy and moment methods of automated process. After identifying

the cracks, form/shape factor of 3.5 was used to distinguish cracks from voids. It was

shown that automated process was different by 6.39% and 3.02% from manual process.

However, the author concluded that due to the ease of operation and efficiency given by

automated process, this error was reasonably small.

Summary

Literature review of methods for preparation of concrete samples for viewing

microstructure was performed. As described before, many sample preparation methods

involved the use of pressure or heat, which might cause concerns over additional cracks

in concrete and might lead to erroneous results. Epoxy impregnation of concrete samples

was identified as a process which uses counter diffusion process as sample preparation

means for this study. As optical microscope would have inadequate depth of field and

resolution for microstructure, scanning electron microscopy was preferred for image

acquisition, Variable pressure electron scanning microscope, which requires no special

coating over samples and ability to view samples with moisture was found to be more

suitable for image analysis. After acquiring images, manual thresholding approach to

map microcracks was selected for quantification of microcracks in a given sample














CHAPTER 3
RESEARCH METHODOLOGY

Introduction

This chapter describes the methods for sample selection, specimen preparation

method and image analysis techniques used in this research. As described in Chapter 2,

the objective of this study was to find if the concrete cured at elevated temperature shows

any additional microcracks compared to samples cured at room temperature. This

investigative study was carried out in three phases. Phase one was to identify the samples

which shows tendency for microcracks, while phase two was to select an appropriate

sample preparation methods. Phase three comprised of sample preparation and image

analysis technique to quantify and compare results.

Sample Selection

Samples were selected from an ongoing research project "Adiabatic Temperature

Rise in Mass Concrete" conducted by University of Florida for Florida Department of

Transportation (FDOT). In this study, twenty different mixes were prepared using

AASHTO Type II cement made by two manufactures with different percentages of

pozzlonic materials and two different placing temperatures of 730F and 950F were used

to study adiabatic temperature rise in laboratory simulated mass concrete conditions. As

higher placing temperature increases the concrete curing temperature, samples made from

higher of two placing temperatures, 950F were selected for microstructural analysis. The

field of investigation was further narrowed down further by selecting the mix made with

cement which has higher heat of hydration. A Heat of Hydration test by Construction










Technologies Ltd. (CTL) revealed higher heat of hydration in AASHTO Type II cement

(Cement B) manufactured by CEMEX (78.2 cal/g at 7 days) was greater compared to

AASHTO Type II cement ( Cement A) manufactured by Florida Rock (66.2 cal/g at 7

days). Thus, mixes made with Cement B and 95 'F placing temperature were selected.

This included five different mixes were prepared with varied pozzlonic contents.

* 100% Cement

* 20% Fly Ash + 80 % Cement

* 35% Fly Ash + 65 % Cement

* 50% Slag + 50% Cement

* 70% Slag + 30% Cement

Table 3-1 Mix Design Data


AASHTO Type II CEMEX Cement ( Cement B ) (Mix proportions per cubic yard)

Cement Fly Slag Water Fine Coarse Air Water
Cement Slag Water
Mix Ash Aggregate Aggregate Entertainer Admixture
(lb) (lb) (lb) (oz) (oz)
95BOOP 760.00 0.00 0.00 279.00 1033.00 1736.00 2.30 60.86

95B20F 608.00 152.00 0.00 279.00 1033.00 1708.00 4.00 30.40

95B35F 494.00 266.00 0.00 279.00 980.00 1687.00 4.00 68.20

95B50S 380.00 0.00 380.00 279.00 1021.00 1724.00 2.30 38.06

95B70S 228.00 0.00 532.00 279.00 1021.00 1724.00 2.30 68.20


Prepared concrete samples were kept in two different types of containers. For

samples to be kept at room condition, concrete was poured in 6x12 plastic containers and

for samples to be kept at high temperature cured condition, specially acquired thermal

curing chambers were used. Thermal curing chamber maintained temperature inside

chamber in an adiabatically condition. The chambers were connected with thermo










couples to a micro controller system for maintaining temperature and to document the

temperature rise in the concrete samples.

Data given in following Table 3.2 shows the first four day temperature rise in

concrete specimen kept in thermal curing chambers.

Table 3-2 Adiabatic Temperature Rise Data

Cement B (950F Placing Temperature)
Time Plain Cement 20% Fly Ash 35% Fly Ash 50% Slag 70% Slag
(Day) Temp. Rise Temp. Rise Temp. Rise Temp. Rise Temp. Rise
(OF) (OF) (OF) (OF) (OF)
0.0 0.00 0.00 0.00 0.00 0.00
0.5 2.60 56.80 50.35 58.60 45.00
1.0 72.15 72.90 61.90 76.70 69.40
1.5 84.20 77.40 67.85 80.70 73.90
2.0 85.60 77.60 68.90 81.80 76.30
2.5 85.65 77.60 69.20 82.00 77.90
3.0 85.95 77.60 69.30 82.00 78.90
3.5 85.95 77.60 69.60 82.00 79.50
4.0 85.95 77.60 69.90 82.00 79.90

As it can be seen in Table 3-2, data indicates that sample 95 B 00 P had the highest

temperature rise of 85.95 'F over placing temperature of 99 'F, thus effectively curing

temperature in the core is 1860F. Therefore, sample 95 B 00 P was selected for

microscopic studies due to the following reasons.

* Higher placing temperature had higher curing temperature and therefore gives
higher probability of presence of additional microcracks

* Cement B had higher heat of hydration (78.2 cal/g at 7 days) compared to Cement
A (66.2 cal/g at 7 days), which might lead to higher curing temperature

* As indicated in Table 3.2, sample 95 B 00 P with plain cement showed higher
temperature rise compared to samples with varied percentages of fly ash or blast
furnace slag.

Specimen Preparation Method

As described in Chapter 2, many different approaches have been used in the past to

view and document microcracks in cement paste or concrete. However, most of the









methods indicate the use of heat, vacuum or pressure along the course of sample

preparation or prior to impregnation, which might lead to additional microcracks due to

sample preparation widely known as secondary microcracks. Samples cured at elevated

temperature as well as those cured at room temperature would undergo epoxy

impregnation method for Scanning Electron Microscopy as established by Struble and

Stutzman (1989). In this method they used counter diffusion method to replace the pore

water of concrete first by ethyl alcohol and then low viscosity epoxy. Neither of these

processes requires any prior pre-heating of surface to dry, nor does it use any vacuum for

impregnation of epoxy. The only drawback of this method is that sample preparation

would take longer time to complete the process. Considering all methods and their

advantages and disadvantages, this method was selected for sample preparation in current

study.

Procedure for Sample Preparation

A small 1/2" thick sample was cut with diamond saw from the samples cured at

room temperature and those cured in thermal curing chambers. During this operation, in

order to reduce the heat generation, cooling water was circulated through water pump to

diamond blade to aid cutting. Once the required thickness of the sample is obtained, other

cutting operation was carried out on a diamond wafer saw machine (figure 3-1). It was

used to reduce any artifacts and required polishing time in the samples during the

preparation period. In this machine, blade was passed through cutting fluid

(polypropylene glycol) to keep the samples from drying. After cutting by diamond

wafer, average size of specimen was 2.15"x 1.8". The samples then went through a two

stage counter diffusion process.













i I I


Figure 3-1 Diamond Wafer Saw

In this process, first pore water of concrete is replaced by 200% Ethanol made by Sigma-

Aldrich Inc., Montana, and then in the second stage, ethanol was replaced by ultra low

viscosity epoxy solution made by Structure Probe Inc., Pennsylvania.

For the first stage of alcohol pore water replacement stage, the specimen was

placed in a lidded jar filled with ethanol (200 Proof) for the alcohol pore solution

replacement stage. A companion specimen was used to determine the depth of

replacement of pore water by alcohol. This companion specimen was a remnant from the

specimen after trimming. This specimen was now placed in a jar filled with ethanol dyed

by a deep red alcohol-miscible dye by Poly science Inc. By splitting or sawing the

companion specimen after a period of time, the depth of replacement is seen by the depth

of dye coloration. On the average, alcohol- pore water replacement was found to be

Imm/day in concrete samples. When this depth reaches to half the section thickness, the

pore solution replacement by alcohol was stopped. The section is then placed in a









container with the ultra low-viscosity epoxy (see Figure 3-2) for the same time as alcohol

pore water replacement process. Curing of epoxy was performed as specified by product

manual. The epoxy was made up of 4 different chemicals as described in Table. 3-3

















Figure 3-2 Ultra Low Epoxy Kit by SPI

Table 3-3 Mixing Schedule for Epoxy
No. Name of Chemical Mixing Schedule
1. Vinylcyclohexene dioxide (VCD) 10g
2. n-Octenyl succinic anhydride (n-OSA) 20 g
3. Butanediol Diglycidyl Ether (BDE) 0.3 g
4. Dimethylaminoethanol (DMAE) 0.3 g

Ethanol was replaced by epoxy at two hour intervals with the proportion of ethanol

to epoxy was changed from 3:1, 1:1, 1:3 at each interval and finally 100% epoxy

solutions was used for keeping it in embedding medium overnight. For curing of epoxy,

samples were kept in a heating chamber at 1310F for six hours. At the end of six hours,

the specimens were ready for the cutting and polishing stages.

Grinding and polishing are important steps in sample preparation. Grinding is used

to expose new surface layer and to remove excessive epoxy coating on the surface.

However, excessive grinding would remove the depth of epoxy layer and could damage

microstructure. Grinding was done on a Metaserv 2000 Grinder Polisher Machine (see









figure 3-3 and 3-4) on 150 RPM. Abrasive papers of 180,240, 320,400 and 600 were

used for the removal of material from specimen surface. Approximately 2 minute of time

was given to each grinding cycle.


















Figure 3-3 Metaserv Grinder and Polisher Figure 3-4 Polished Sample


Polishing is done for undoing the damage done due to sawing and grinding of

surface. After grinding by 600 grits, surface becomes smooth and ready for further

polishing. Polishing operation involves successively decreasing size of diamond

suspension paste from 6 m to 3pm, lm and 0.25 im manufactured by Buhler (see figure

3-5). A polishing cloth was attached to rotating wheel of Metaserv 2000 Grinder Polisher.

This method helps in reducing the damage done by previous suspension by removal of

finer scratches and improving the surface for microscopic study. At the end of this

operation, the sample is wiped with a plain cloth and kept in glass jar (See figure 3-6).

During polishing process, it is important to understand that cutting by diamond saws

would damage the samples up to 30% of its thickness. So polishing should be done till

30% of blade thickness in order to remove all artifacts.























Figure 3-5 Diamond Suspension Paste


Figure 3-6 Prepared Samples


Preparation for Image Processing

Microstructural analysis was performed at Advance Material Characteristic Lab at

the Department of Civil and Coastal Engineering, University of Florida, Gainesville. The

facility has a Hitachi S 3000N, variable pressure Scanning Electron Microscope

(VPSEM). As described earlier, VPSEM would allow the viewing of the specimen

without covering it with a conductive surface. A double sided carbon tape was used for

attaching samples to specimen holder. This conductive tape avoids any charge built-up

and provides a good background surface.

Images were viewed at 50 Pa (0.4 Torr), accelerating voltage of 10.0 KV and on the

scale of 200 [im at 250 X magnification of specimen surface. At the start of this study,

polished samples were viewed directly under microscope and five representative images

were taken from different location of specimen. However, during the course of study, it

was decided to increase the number of images to 16 to have good representative nature of

the sample, which was formulated on basics of 4x4 matrixes. The following figures 3-7

and 3-8 shows the grid system.











1 2 3 4




2.15"

9 10 11 12


13 14 15 16


1.80"1


Figure 3-7 Specimen with Copper Tape Figure 3-8 Specimens with 4x4 matrixes




The samples were divided into 16 equal parts using a copper conductive tape to

make grid on the sample to identify the points of interesting while viewing under

microcracks.

Image Analysis

Images taken from 16 different square position provides information related to

microstructure of concrete. A manual mapping of cracks was done in order to find out the

length of the microcracks and density per micrometer for comparing all the samples. To

identify proper microcracks from the image, shape factor was used. Shape factor is ratio

of length to width. If this ratio was greater then 3.5, the object of consideration termed as

microcracks or if the ratio was less than 3.5, the object was termed as Void. Each image

was manually mapped as well as by viewing on ImageJ software to find out the exact

number of cracks to avoid any bias from operator. The method of image analysis is

described in more details in Chapter 4.









Summary

In this study, samples were taken from concrete mix which showed maximum

probability of crack development due to high curing temperature. Specimen preparation

method was selected to avoid any preheating of the samples and to avoid any use of

vacuum or pressure techniques for impregnation. Once specimen were impregnated and

prepared with grinding and polishing, a grid made from conductive copper tape was used

to locate the points of interests, while viewing the sample under microscope. At the end, a

statistical analysis was performed to investigate total representative nature of images and

the rationality for use in this study.














CHAPTER 4
IMAGE PROCESSING AND ANALYSIS

Introduction

This chapter describes the image processing and analysis carried out in this

research. A total of four samples were made, two of which cured in room temperature

and the other two cured at elevated temperature. As described earlier, the samples were

divided into 16 equal parts by conductive copper tape. The images were taken around the

center of square in 250X resolution Thus for each sample a total of 16 images were taken

in 1028 x 960 pixel resolution and 50 Pa working pressure.

In SEM observation, finer surface structure images can generally be obtained with

lower accelerating voltages. At higher accelerating voltages, the beam penetration and

diffusion area become larger, resulting in unnecessary signals (e.g., backscattered

electrons) being generated from within the specimen. These signals reduce the image

contrast and veils fine surface structures. It is especially desirable to use low accelerating

voltage for observation of low-concentration substance.

After acquiring the images, image processing was performed to view microcracks

in concrete matrix. ImageJ software developed by National Institute of Health was used

for the purpose of image processing in this study. Based on java platform, this software

has the capability to perform wide ranges of operations such as binary thresholding, area

measurement, filtering noise and mathematical operations.









Image Processing Technique

First step of image processing was to perform segmentation of image into binary

image (black and white) by brightness thresholding approach. In this process, software

would change the intensities of image pixels which satisfy the given condition, thus

highlighting the regions of interests. However, as it can be imagined, there would be

variations in grey level of matrix in different images, and there could be bias in operator

visions. Investigating this issue, Soroshian (2003) performed a comparative analysis of

manual and automated thresholding operation on images taken by Scanning Electron

Microscope, where three types of automated thresholding operation factorizationn,

entropy and moment) were performed. However, study reported that there were no

significant differences between manual and factorization mode of thresholding in

automated process. The study also reported 5% and 2% difference in Entropy and

Moment mode of thresholding respectively which could be regarded as less significant.

Therefore for the scope of this study, manual thresholding by brightness approach

was performed. Images were acquired with variable pressure scanning electron

microscope and PCI-quartz picture acquisition system. Images were transferred and

viewed with ImageJ software. For each image, low threshold level was set to zero or

default for thresholding and best threshold level image was determined by comparing the

original gray image and the image after application of manual threshold on binary image.

Total of sixty-four images were converted one after another to binary mode. After binary

operation, images were manually mapped to find out total number of crack lengths and

crack density across specimens.











Sample 95BOOP Room Temperature Sample 1


A. Image 95 B 00 P -RT 1 (2)


C. Image 95 B 00 P -RT 1 (11)


B. Image 95 B 00 P -RT 1 (10)


D. Image 95 B 00 P -RT 1 (16)


Figure 4-1 Representative Images of Mix 95 B 00 P RT-1 before Image Processing

After highlighting features of interest (microcracks and voids) to find the most

distinct contrast with concrete background, image analysis was performed to distinguish

microcracks and voids. As it is known that microcracks are long elongated structure,

length to width (also known as aspect/ form/ shape) ratio was applied. The objects having

shape factor of 3.5 or greater was termed as microcracks, while the remaining objects

were termed as voids. A software plugin Shape descriptor from ImageJ was used to










identify voids in the images. Shape descriptor takes minimum and maximum pixel values

and plots the void concentration in a separate image. After completing void removal from

image, microcracks are mapped manually and counted.


A. Image 95 B 00 P RT 1 (2)


C. Image 95 B 00 P -RT 1 (11)


B. Image 95 B 00 P RT -1 10 )


SUPA 21-OCt-U4 RT1 WDUl.I2m 1UURkV X25U ZIJUU7


D. Image 95 B 00 RT -1 (16)


Figure 4-2 Representative Images of Mix 95 B 00 P RT-1 after Image Processing

As the sample was divided into 16 equal parts by copper conductive double-sided

tape, 16 representative images were taken from the center of every part. After processing

the image, cracks were manually mapped to find out the total length of cracks in each


i









image. As it can be seen in table 4-1, density of cracks over the given area was found by

converting crack lengths and dividing it by image area.

Table 4-1 Crack Quantification Analysis for Mix 95 B 00 P RT-1

No. Image Scale Cracks Cracks Image Field Crack
(cm) length Density
(lm)_ (m/4m2
1 RT 1-1 200Om/11 cm 72 cm 1309.091 183 pm x 250 im 0.028614
2 RT 1-2 200Om/11 cm 71 cm 1290.909 183 pm x 250 im 0.028217
3 RT 1-3 200nm/11 cm 65 cm 1181.818 183 mr x 250 mr 0.025832
4 RT 1-4 200im/11 cm 82 cm 1490.909 183 rm x 250 pm 0.032588
5 RT 1-5 200nm/11 cm 61 cm 1109.091 183 mr x 250 mr 0.024242
6 RT 1-6 200Cim/11 cm 66 cm 1200.000 183 rm x 250 rm 0.026230
7 RT 1-7 200nm/11 cm 64 cm 1163.636 183 rm x 250 pm 0.025435
8 RT 1-8 200rm/11 cm 61 cm 1109.091 183 mx250 m 0.024242
9 RT 1-9 200rm/11 cm 73 cm 1327.273 183 rmx 250 pm 0.029011
10 RT 1-10 200um/11 cm 92 cm 1672.727 183 mr x 250 mr 0.036562
11 RT 1-11 200pm/11 cm 72 cm 1309.091 183 pm x 250 pm 0.028614
12 RT 1-12 200m/11 cm 68 cm 1236.364 183 mr x 250 mr 0.027024
13 RT 1-13 200um/11 cm 98 cm 1781.818 183 mx250 m 0.038947
14 RT 1-14 200um/11 cm 63 cm 1145.455 183 mr x 250 mr 0.025037
15 RT 1-15 200pm/11 cm 65 cm 1181.818 183 pm x 250 pm 0.025832
16 RT 1-16 200pm/11 cm 58 cm 1054.545 183 pm x 250 pm 0.023050
Avg. 1285.227 0.028092
SD. 203.7358 0.004453

Data analysis shows that for 95 B OOP RT-1 samples, average lengths of cracks

were 1285.227 rim. Maximum length of cracks in an image was found to be 1672[im,

while minimum length of cracks in an image was 1054.545 rim, giving standard deviation

of 203.227. As it can be seen in Table 4.1, average crack density in an image was

0.028092 nm/pm,2 the maximum density of cracks was 0.038947 im/m2 and minimum

density of cracks was 0.02305 rm/pm2, giving standard deviation of 0.004463. Figure

4.3 describes variations in crack density of room temperature cured sample by each


image.










96 B 00 P Room Temprature Sample 1
0.05
> 0.04
S0.03
0.02 -
S0.01

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Image Number


Figure 4-3 Variations in Crack Density of 95 B 00 P RT1

To find out if the sample size was representative in nature, a statistical study was

performed. Odd and Even numbers of images were selected for calculating average

density of lengths. As it can be seen in Table 4-2, Crack density of odd number images

were 0.02832 gm/gm2 with standard deviation of 0.005281, while crack density of even

number images were 0.02787 gm/gm2 with standard deviation of 0.004982. After

comparing it with average crack density of all 16 images of 0.02809 gm/gm2, it can be

seen that crack density is averaged almost equally. Thus it could be deduced that crack

density is representative in nature throughout the sample and the number of images are

adequate for giving representative portrait of the sample.

Table 4-2 Comparison of Odd and Even Number Images

No Crack Density of Odd Number Image Crack Density Even Number Images
No.(m/gm2) (m/gm2)
1 0.028614 0.028217
2 0.025832 0.032588
3 0.024242 0.026230
4 0.025435 0.024242
5 0.029011 0.036562
6 0.028614 0.027024
7 0.038947 0.025037
8 0.025832 0.023050
Avg. 0.028316 0.027869
SD 0.004643 0.004563









Sample 95BOOP Room Temperature Sample 2


Figure 4-4 Representative Images of Mix 95 B 00 P RT-2 before Image Processing

Figure 4.4 shows images from sample number two which was cured at room

temperature. It can be seen from images that extensive cracking is present in almost all

parts of representative images. As described earlier, a total of 16 images were taken from

each sample. The images were then processed by ImageJ software for quantification of

microcracks.




























50Pa 15-Oct-04 wD17,0o 10.0kV x250 200um


Image 95 B 00 P-RT2 (4)


Image 95 B 00 P-RT2 (15)


Image 95 B 00 P-RT2 (10)


Image 95 B 00 P-RT2 (16)


Figure 4-5 Representative Images of Mix 95 B 00 P RT-2 after Image Processing

After processing the image by converting it into binary image, cracks were

manually mapped to find out the total length of cracks in each image. As it can be seen in

table 4-3, the density of cracks over the given area was found by converting crack lengths

and dividing it by image area. Another statistical analysis was carried out to see if the

number of images constituent representative area for the sample.


I









Table 4-3 Crack Quantification Analysis for Mix 95 B 00 P RT-2


No. Image Scale Cracks Cracks Image Field Crack
length Density
(cm) (nm/m2) (Gm/4m2)
1 RT 2-1 200pm/11 cm 64 cm 1163.636 183 pm x 250 pm 0.025435
2 RT 2-2 200Om/11 cm 68 cm 1236.364 183 pm x 250 Lm 0.027024
3 RT 2-3 200m/l 1 cm 56 cm 1018.182 183 m x 250 m 0.022255
4 RT 2-4 200m/11 cm 89 cm 1618.182 183 mx250 m 0.035370
5 RT 2-5 200m/11 cm 90 cm 1636.364 183 pm x 250 pm 0.035768
6 RT 2-6 200m/11 cm 65 cm 1181.818 183 pm x 250 Lm 0.025832
7 RT 2-7 200m/11 cm 82 cm 1490.909 183 pm x 250 pm 0.032588
8 RT 2-8 200(m/11 cm 69 cm 1254.545 183 mx250 m 0.027422
9 RT 2-9 200um/11 cm 83 cm 1509.091 183 pmx250 pm 0.032986
10 RT 2-10 200um/11 cm 67 cm 1218.182 183 pm x 250 pm 0.026627
11 RT 2-11 200pm/11 cm 64 cm 1163.636 183 pm x 250 pm 0.025435
12 RT 2-12 200um/11 cm 61 cm 1109.091 183 pm x 250 pm 0.024242
13 RT 2-13 200um/11 cm 67 cm 1218.182 183 mx250 m 0.026627
14 RT 2-14 200um/11 cm 82 cm 1490.909 183 pmx250 pm 0.032588
15 RT 2-15 200um/11 cm 84 cm 1527.273 183 pm x 250 pm 0.033383
16 RT 2-16 200pm/11 cm 68 cm 1236.364 183 pm x 250 pm 0.027024
Avg. 1317.045 0.028788
SD. 194.692 0.004256

Data analysis shows that for 95 B OOP RT-2 samples, average lengths of cracks was

1317.045 rim. Maximum length of cracks in an image was found to be 1636km, while

minimum length of cracks in an image was 1018.182 rim, giving standard deviation of

194.692. As it can be seen in figure 4-6, average crack density in an image was 0.028788

inm/im2, maximum density of cracks was 0.035768 im/m2 and minimum density of

cracks was 0.0222 im/pm2, giving standard deviation of 0.004256.

The above data shows presence microcracks in the given sample. However, to see

if the images were sufficient to draw conclusions from it, another statistical analysis was

carried out. In this technique, images were divided into two formations. The formation

was varied such as diagonals, odd-even or lines to prove representative nature of the


samples.




























Figure 4-6 Variations in Crack Density of Sample 95 B 00 P RT2

Figure 4-6 demonstrates variations in crack density of sample 95 B OOP RT2 by

each image. The data given in Table 4-4 shows that the crack density of odd number

images were 0.029309 gm/gm2 with standard deviation of 0.0049, while crack densities

of even number images were 0.028266gm/gm2 with standard deviation of 0.0037.

Table 4-4 Comparison of Odd and Even Number Images

No Crack Density of Odd Number Image Crack Density Even Number Images
(gm/gm2) (gm/gm2
1 0.025435 0.027024
2 0.022255 0.035370
3 0.027422 0.032588
4 0.025832 0.035768
5 0.032986 0.026627
6 0.025435 0.024242
7 0.026627 0.027024
8 0.033383 0.032588
Avg. 0.029309 0.028266
SD 0.004922 0.003737



After completing analysis of samples cured at room temperature, crack density

calculations were performed on samples cured at high temperature. As described earlier,

samples were divided in 16 equal parts with conductive copper tape and images were


C 95 B 00 P Room Temperature

a 0.04
c 0.035
k 0.03
0.025
D 0.02 -
e 0.015 -
0.01
s
0.005 -
T 0
Y 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Number of Images









taken from the center of rectangle. Figure 4-7 shows images from high temperature cured

sample one.

Sample 95BOOP High Temperature Sample 1


Figure 4-7 Representative Images of Mix 95 B 00 P HT-1 before Image Processing

Maximum heat gain during high temperature curing process of concrete reached up

to 85.5 'F. The adiabatic temperature condition was generated and measured in the

laboratory with a microcontroller based SURE-CURE curing system.

























Image 95 B 00 P HT 1 (2)


UvFa 2-O- GU WD01.M 1U.U'W ZwU MVUW
Image 95 B 00 P -HT 1(3)


dUFA 2U-uct-U GCU W OIU.J=n laU.wCY LV X UUuM
Image 95 B 00 P -HT 1(7)


Figure 4-8 Representative Images of Mix 95 B 00 P HT-1 after Image Processing

After preparing the samples and subsequent polishing and lapping operations,

copper metal grid system was mounted. Samples were kept in variable pressure scanning

electron microscope to analyze microscopic features. As it can be seen in Figure 4-8,

images were initially processed by manual thresholding by brightness approach. Lower

thresholding was kept at zero, while upper thresholding was moved till the best contrast

between images and objects of interest developed. Voids were removed from the binary

picture and manual mapping of microcracks were done for crack quantification.


Image 95 B 00 P HT-1 (1)









Table 4-5 Crack Quantification Analysis for Mix 95 B 00 P HT-1

No. Image Scale Cracks Cracks Image Field Crack
length Density
(cm) (Gm) (Gm/4m2
1 HT 1 -1 200m/11 cm 105 cm 1909.091 183 pm x 250 Lm 0.041729
2 HT 1-2 200m/11 cm 106 cm 1927.273 183 pm x 250 Lm 0.042126
3 HT 1-3 200m/11 cm 94 cm 1709.091 183 mx250 m 0.037357
4 HT 1-4 200im/11 cm 52 cm 945.4545 183 pm x 250 pm 0.020666
5 HT 1-5 200um/11 cm 58 cm 1054.545 183 pm x 250 pm 0.023050
6 HT 1-6 200pm/11 cm 53 cm 963.6364 183 pm x 250 pm 0.021063
7 HT 1-7 200m/11 cm 48 cm 872.7273 183 pm x 250 pm 0.019076
8 HT 1-8 200(m/11 cm 54 cm 981.8182 183 mx250 m 0.021461
9 HT 1-9 200um/11 cm 58 cm 1054.545 183 mx250 m 0.023050
10 HT 1-10 200um/11 cm 55 cm 1000.000 183 pm x 250 pm 0.021858
11 HT 1-11 200pm/11 cm 71 cm 1290.909 183 pm x 250 pm 0.028217
12 HT 1-12 200m/11 cm 65 cm 1181.818 183 pm x 250 pm 0.025832
13 HT 1-13 200um/11 cm 62 cm 1127.273 183 mx250 m 0.024640
14 HT 1-14 200um/11 cm 75 cm 1363.636 183 mx250 m 0.029806
15 HT 1-15 200um/ll cm 65 cm 1181.818 183 pm x 250 pm 0.025832
16 HT 1-16 200pm/11 cm 82 cm 1490.909 183 pm x 250 pm 0.032588
Avg. 1253.409 0.027397
SD. 339.4677 0.007420

Data analysis from Table 4-3 shows that for 95 B OOP HT-1 samples average

lengths of cracks were 1253.409 rim. Maximum length of cracks in an image was found

to be 1927 rim, while minimum length of cracks in an image was 872.73 rim, giving

standard deviation of 339.4677. As it can be seen in Table 4-5, average crack density in

an image was 0.027397 [im/pm2, maximum density of cracks was 0.042126 im/m2 and

minimum density of cracks was 0.019076 rm/pm2' giving standard deviation of 0.00742.

The above results determine presence of microcracks in the given sample, to find if

the sample size is sufficient to draw conclusion, another statistical analysis was carried

out. In this process, images were divided into two groups in various formations and

average crack density was calculated for each of them. Results shown in table 4-6 shows

that average crack density is present throughout the concrete sample.











95 B 00 P High Temprature Sample 1


0.05
S0.04
C 0.03
g 0.02
0 0.01
0


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Image Number


Figure 4-9 Variations in Crack Density of Sample 95 B OOP HT- 1

As it can be seen in Table 4-2, Crack density of odd number images were 0.027869

nm/gm2 with standard deviation of 0.007721, while crack density of even number images

were 0.026925 gm/gm2 with standard deviation of 0.007676. After comparing it with

average crack density of all 16 images of 0.027397 nm/gm2, it can be seen that crack

density is averaged almost equally.

Table 4-6 Comparison of Odd and Even Number Images

Crack Density of Odd Number Image Crack Density Even Number Images
No. (nm/nm2) (mnm/nm2)
1 0.041729 0.042126
2 0.037357 0.020666
3 0.023050 0.021063
4 0.019076 0.021461
5 0.023050 0.021858
6 0.028217 0.025832
7 0.024640 0.029806
8 0.025832 0.032588
Avg. 0.027869 0.026925
SD 0.007751 0.007576



Coefficient of variation being less than 15% proves that results are concentrated within

15% of the mean. Thus average crack density is averaged equally all across the sample.










Sample 95BOOP High Temperature Sample 2


Image 95 B 00 HT -2 (15)


Image 95 B 00 P -HT 2 (15)


Image 95 B 00 HT -2 (7)


Image 95 B 00 P -HT 2 (16)


Figure 4-10 Representative Images of Mix 95 B 00 P HT-2 before Image Processing

The above four images are representative images from mix 95 B 00 P cured at

high temperature, taken by Hitachi S 3000 N variable pressure scanning electron

microscope. A 4x4 matrix made by conductive copper tape was used to take 16

representative images across the sample.


I




























WD17.0mB 10.OWkV 250 200Dum


Image 95 B 00 P -HT 2 (4)


Un


Image 95 B 00 P -HT 2 (10)


Image 95 B 00 P -HT 2


Image 95 B


UOL.1m HTU.U2V xzU zUum


00 P -HT 2


Figure 4-11 Representative Images of Mix 95 B 00 P HT-2 after Image Processing

All the images were processed in a java based free source ImageJ software,


developed by National Institute of Health, for image processing. Manual thresholding


was done till the perfect contrast between regions of interest with background was


developed. Images were then processed to distinguish voids from cracks. As cracks have


higher length to width ratio, shape factor of 3.5 was used to filter out voids from the


cracks. After that, cracks lengths were manually mapped on the image.


50P 15-oct-04


50Pa 15-OGt-04


W17.2- 10O.kV .250 20Dum


buFa i5-ut-UG









Table 4-7 Crack Quantification Analysis for Mix 95 B 00 P HT-2


No. Image Scale Cracks Cracks Image Field Crack
length Density
(cm) (m) (Gm/4m2)
1 HT 2-1 200Cim/11 cm 58 cm 1054.545 183 rm x 250 rm 0.023050
2 HT 2-2 200rm/11 cm 98 cm 1781.818 183 pm x 250 im 0.038947
3 HT 2-3 200nm/11 cm 88 cm 1600.000 183 mr x 250 rm 0.034973
4 HT 2-4 200m/11 cm 64 cm 1163.636 183 mx250 m 0.025435
5 HT 2-5 200nm/11 cm 57 cm 1036.364 183 mr x 250 mr 0.022653
6 HT 2-6 200m/11 cm 69 cm 1254.545 183 pm x 250 im 0.027422
7 HT 2-7 200nm/11 cm 57 cm 1036.364 183 mr x 250 mr 0.022653
8 HT 2-8 200(m/11 cm 54 cm 981.8182 183 mx250 m 0.021461
9 HT 2-9 200rm/11 cm 75 cm 1363.636 183 rmx250 pm 0.029806
10 HT 2-10 200nm/11 cm 65 cm 1181.818 183 mr x 250 mr 0.025832
11 HT 2-11 200rm/11 cm 70 cm 1272.727 183 pm x 250 pm 0.027819
12 HT 2-12 200nm/11 cm 62 cm 1127.273 183 mr x 250 mr 0.024640
13 HT 2-13 200rm/11 cm 65 cm 1181.818 183 mx250 m 0.025832
14 HT 2-14 200rm/11 cm 76 cm 1381.818 183 mx250 m 0.030204
15 HT 2-15 200nm/11 cm 95 cm 1727.273 183 mr x 250 mr 0.037755
16 HT 2-16 200rm/11 cm 61 cm 1109.091 183 pm x 250 pm 0.024242
Avg. 1265.909 0.027670
SD. 238.8743 0.005393

Data analysis from Table 4-8 shows that for 95 B OOP HT-2 samples average

lengths of cracks was 1265.909 rim. Maximum length of cracks in an image was found to

be 1781.818 rim, while minimum length of cracks in an image was 981.8182 rim, giving

standard deviation of 238.8743. As it can be seen in Table 4.5, average crack density in

an image was 0.027670 [im/pm2, maximum density of cracks was 0.038947 im/m2 and

minimum density of cracks was 0.021461 rm/pm2, giving standard deviation of

0.005393. Figure 4-12 shows variations in microcrack density in the sample made from

high temperature cured concrete mix 95 B 00 P, i.e., concrete made with cement B,

(manufactured by Cemex AASHTO type II, heat of hydration 78.2 cal/g at 7 days),

placing temperature of 950F, and without any pozzlonic contents added to mix. The


higher temperature reached was 1800F.











95 B 00 P High Temprature

0.045
0.04
>, 0.035
*3 0.03
0.025 -
0.02
8 0.015
6 0.01 -
0.005

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Number of Images

Figure 4-12 Variations in Crack Density of Sample HT 2

Figure 4-12 shows maximum value of crack density was 0.3894 with lowest crack

density appeared to be 0.214.

Table 4.8 shows crack density of odd number images from the sample having crack

density of 0.028068 gm/gm2, while crack density of even number images from samples

was 0.027273 gm/gm2. From Statistical analysis of image patterns and coefficient of

ratio within 20% suggests that, average image density is present throughout the images

and given sample size is true representative in nature. Similar results were found with

random or diagonal image patterns with same images.

Table 4-8 Comparison of Odd and Even Number Images

Crack Density of Odd Number Image Crack Density Even Number Images
No. m/m2 m/m2

1 0.023050 0.038947
2 0.034973 0.025435
3 0.022653 0.027422
4 0.022653 0.021461
5 0.029806 0.025832
6 0.027819 0.024640
7 0.025832 0.030204
8 0.037755 0.024242
Avg. 0.028068 0.027273
SD 0.005775 0.005349









Comparative Analysis of Samples

To analyze the differences between both samples, which are cured at room

temperature, a comparative analysis was performed.

Table 4-9 Comparative Analysis of 95 B OOP RT1 and 95BOOP RT2

Sample Avg. Crack Length (jim) Avg. Crack Density (inm/jim2)

RT1 1285.227 0.028092

RT2 1317.045 0.028788

Average 1301.136 0.02844

% Change 2.48% 2.48%


The results shown in Table 4-9 show that the difference between average crack

lengths of samples cured at room temperatures was 32 inm/jm2; the difference between

average crack densities was 0.0006. The differences between average crack lengths in

high temperature cured samples were 12.5 jim/jim2 and difference in their crack density

were 0.0002.

Table 4-10 Comparative Analysis of 95 B OOP HT1 and 95BOOP HT2

Sample Avg. Crack Length (jim) Avg. Crack Density (inm/inm2)


HT1 1253.409 0.027397

HT2 1265.909 0.027670

Average 1259.659 0.027535

% Change 0.957% 0.99%









Summary

Four samples were taken from concrete mix 95 B 00 P, two of which were cured at

room temperature and the two remaining samples were cured at high temperature. After

sample preparation, the samples were viewed under Hitachi S 3000 N, variable pressure

scanning electron microscope. Each sample was coated with copper conductive tape as

4x4 matrixes which divided the samples in sixteen equal parts. Thus, a total of 16 images

were taken, each one from the center of the parts. The images were later analyzed with

ImageJ, image processing software. Segmentation of images was done by manually

thresholding the image till the best contrast developed between background and points of

interests. After thresholding, cracks were manually mapped for quantification of

microcracks in sample.

The results show that the average crack density of concrete sample cured at room

temperature was 0.02844 mm/mm2, while crack density of sample cured at elevated

temperature was 0.027535 mm/mm2. These results are in accordance with research study

conducted by Ammouche et al., (2000) in which average crack density of undamaged

samples was found to be 0.00271 mm/mm2. From the above analysis, it could be

concluded that room temperature cured sample had approximately 3% more microcracks

than the sample cured at high temperature. Thus it could be said that there is no or very

small difference between the quantity of microcracks found in concrete cured at room

temperature and concrete cured at elevated temperature for concrete samples made

without pozzlonic materials.














CHAPTER 5
SUMMARY, RESULTS AND RECOMMENDATIONS

Summary

The purpose of this research was to identify if any additional microcracks are

developed due to concrete cured at elevated temperature. To accomplish that following

tasks were performed.

* Identification of the samples which had highest curing temperature and therefore
may develop additional microcracking.

* Performed a literature review on methods available for diction of microcracks and
finding out their advantages and disadvantages.

* Prepared specimens in order to view them under Scanning Electron Microscope for
investigating microcracks.

* Processed images to quantify and compare the intensity of microcracks in both
samples

Samples were selected out of 20 different concrete mixes made with different

percentages of pozzlonic materials, two AASHTO type II cement with two different

placing temperatures. Phase I of this project showed that, concrete made from a cement

with higher heat of hydration and without adding any pozzlonic material generate highest

curing temperature. Therefore, a mix placed at 950F and made with AASTO type II

cement having heat of hydration of 78.2 cal/g at 7 days without any pozzlonic material

was selected for this study.

Selection of appropriate sample preparation method was one of the important issues

as improper sample preparation would cause additional microcracks in sample and lead to

erroneous results. Most of the sample preparation methods used in past involved use of









preheating the specimen for removal of pore water or putting specimen under vacuum for

impregnation. There is a concern that by use of heat or vacuum, samples might develop

additional microcracks. Epoxy impregnation method by Struble and Stutzman (1989) was

used in this study as this method uses counter diffusion process to impregnate samples.

Apart from supporting the microstructure of specimen, Epoxy impregnation also provides

better contrast in cementitious matrix. Longer sample preparation time is the only

drawback for performing this experiment.

Images were acquired by a Hitachi S 3000 N, variable pressure scanning electron

microscope (VP-SEM) in backscattered scattered electron (BSE) mode which is a high

energy beam capable of distinguishing the particles present in matrix by variation of

brightness in images. The specimens were divided as 4x4 matrixes by a copper

conductive tape to aid viewing of points of interest from predefined positions.

Total of 16 images were taken from each sample and analyzed in ImageJ software

for quantification of microcracks. Images were converted to binary mode by manual

thresholding operation. When features of interest (microcracks and voids) were correctly

highlighted, the most distinct contrast between microcracks/voids and concrete

background was obtained. After getting binary image, next task performed was to

distinguish between voids and microcracks. As its known that microcracks are long

elongated structure, thus length to width (also known as aspect/ form/ shape) ratio was

applied. Objects having shape factor of 3.5 or greater was termed as microcracks and rest

other objects were termed as voids. A software plugin Shape descriptor from ImageJ was

used to identify voids in the images. Shape descriptor takes minimum and maximum

pixel values and plots the void concentration in a separate image. After completing void









removal from image, microcracks are mapped manually and counted. Results of

quantification analysis were later converted into crack density per area to give

comparative analysis.

Results and Conclusion

Results from this studies showed that, crack density of concrete sample cured at

room temperature was 0.02844 while, crack density of sample cured at elevated

temperature was 0.027535.

Table 5-1 Comparative Analysis of Samples
Avg. Crack Density
Sample Avg. Crack Length(rm) (&m m2)

Room Temperature Cured samples 1301.136 0.028440

High Temperature Cured samples 1259.659 0.027535

% Change 3.3% 3.28%



This analysis indicated that there is very small difference between quantity of

microcracks found in concrete cured at room temperature or concrete cured at elevated

temperature for samples made with no pozzlonic materials and placed at higher

temperature. Thus it can be concluded that high temperature curing conditions upto

1800F may not play significant part in formation of additional cracks.

Recommendations

From research of this study, the following recommendations were made for

conducting future work in this area.

* In future, studies involving mixes made with different percentage of pozzlonic
materials, water/cement ratios, aggregates and placing temperature should be
investigated to find out if mix design changes have any impact on density of
microcracks in concrete samples.






58


* Epoxy solution used in this study had to be cured at 1310F according to
manufacturer's instruction. Curing of the sample at high temperature might alter
the microstructure of sample, thus efforts should be directed toward finding an ultra
low viscosity epoxy solution, which does not require heating during its curing
process.

* As cutting operations would create damaged zone in concrete microstructure upto
30% of blade thickness in sample, proper grinding and polishing operations should
be followed to expose an undamaged surface.

* A more intrinsic grid system of conductive tape should be created to acquire more
images from concrete sample which can be useful if automated image processing is
used.













LIST OF REFERENCES


American Concrete Institute (ACI) Committee 116, Report SP-19, Farmington Hills,
Michigan, 1995

Advances in Materials Problem solving with Electron Microscope Proceedings of
Material Research Society, Volume 589, 1999

Ammouche, A. Breysse, D. Hornain, H. Didry, 0., and Marchand, J., "New image
analysis technique for the quantitative assessment of microcracks in cement-based
material," Cement and Concrete Research Volume 30, No.1, Page 25-35, 2000

Bisschop, J., and Van Mier, J.G.M., "How to study drying shrinkage microcracking in
cement-based materials using optical and scanning electron microscopy," Cement
and Concrete Research Volume 32, No. 2, Page 279-287, February 2002

Florida Department of Transportation, Structural Design Guide, Section 3.9 Mass
Concrete, Tallahassee, Florida

Gran, H. C., "Fluorescent liquid replacement technique: A means of crack detection and
water-binder ratio determination in high-strength concretes." Cement and Concrete
Research Volume 25, No 5, Page 1063-1074, 1995

Hanke, L. Variable Pressure Scanning Electron Microscopy for Nonconductive and
Volatile Samples, Handbook of Analytical Methods, 1999

Hornain H., Marchand J., Ammouche A., Commene J.P., and Moranville M.,
"Microscopic observation of cracks in concrete-a new sample preparation technique
using dye impregnation." Cement and Concrete Research, Volume 26, No.4 Page
573-583, 1996

Kjellsen, K. 0., and Jennings, H M., "Observations of microcracking in cement paste
upon drying and rewetting by environmental scanning electron microscopy,"
Advanced Cement Based Materials Volume 3, No.1, Page 14-19, January 1996

Knab L.I., Walker H.N., Clifton J.R., and Fuller E.R., "Fluorescent thin sections to
observe the fracture zone in mortar," Cement and Concrete Research, Volume 14,
No. 3, Page 339-344, 1984

Maggenti R., Mass Concrete Report, San Francisco-Oakland Bay Bridge East Spans
Safety Project. 04-Ala/SF-80-Var, E.A. 04-012021, Available online at
http://www.dot.ca.gov/hq/esc/tollbridge/SFOBB/EastSpan/012024/MaterialsHando
ut/cd2/concretl.pdf, last seen, Oct 12, 2004









Marusin S., "Sample preparation-the key to ESM studied of failed concrete," Cement and
Concrete. Composites Volume 17, No.4, Page 311-318, 1995

Najjar, W.S., Aderhold, H.C., and Hover, K.C., "Application of neutron radiography to
the study of microcracking in concrete Cement", Concrete and Aggregates, Volume
8, No. 2, Page 103-109, 1986

Oberholster, R. E., Maree, H. and Brand, J. H. B, "Cracked prestressed concrete railway
sleepers: alkali silica reaction or delayed ettringite formation," Proceedings of the
9th International Conference on Alkali-Aggregate Reaction in Concrete, London,
Page 740-749, 1992

Ollivier, J.P., Cement and Concrete Research, Volume 15, No.6, Page 1055-1060, 1985

Patel, H.H., Bland, C.H., and Poole, A.B., "Microstructure of concrete cured at elevated
temperatures," Cement and Concrete Research Volume 25 No. 3 Page 485-490,
April 1995

Perkes, P., ACEPT W group, Department of Physics and Astronomy, Arizona State
University, http://acept.la.asu.edu/PiN/rdg/elmicr/elmicr.shtml Last Visited
October 2004

Ringot, E., and Bascoul, A., "About the analysis of microcracking in concrete," Cement
and Concrete Composites Volume 23, No. 2-3, Page 261-266, April 2001

Samaha, H.R., and Hover, K.C., "Influence of microcracking on the mass transport
properties of concrete," ACI materials Journal, Volume 89, No.4, Page 416-424,
1992

Slate, F.O., and Hover, K.C., Fracture Mechanics of Concrete: Edited by F.H. Whitmann,
Published by Elsevier Science Page 85-93, 1984

Slate F.O., and Olsefski, S., "X-Rays for Study of Internal Structure and Microcracking
of Concrete," Journal of American construction Institute, Volume 60, No. 5, Page
575-588, 1963

Soroushian, P., Elzafraney, M., and Nossoni, A., "Specimen preparation and image
processing and analysis techniques for automated quantification of concrete
microcracks and voids," Cement and Concrete Research Volume 33, No. 12, Page
1949-1962, December 2003

Saint John, D., Poole A., and Sims A, Concrete Petrography-A Handbook of
Investigative Techniques Published by Arnold, 1998

Struble L. J., and Stutzman P.E., "Epoxy impregnation of hardened cement for
microstructural characterization," Journal of Material Science Letters, Volume 8,
Page 632-634, 1989






61


Stutzman P, Scanning Electron Microscopy in Concrete Petrography, Materials Science
of Concrete Special Volume: Calcium Hydroxide in Concrete (Workshop on the
Role of Calcium Hydroxide in Concrete). Proceedings of The American Ceramic
Society, edited by J. Skalny, J. Gebauer and I. Odler, November 1-3, 2000, Anna
Maria Island, Florida, Page 59-72, 2001.















BIOGRAPHICAL SKETCH

Vishal Shah was born in Surendranagar, India, on March 21st, 1980, to Sanjay and

Rekha Shah and grew up in Mumbai, located in Western India. He graduated in April of

1997 from Mithibai College and joined the Maharashtra Institute of Technology,

University of Pune, to complete his undergraduate degree in civil engineering. After

graduating with first class in 2002, Vishal continued his graduate studies at the

M.E.Rinker Sr. School of Building Construction, University of Florida. During his stay at

the University of Florida, Vishal worked as a graduate assistant for Dr. Abdol Chini.

Upon his graduation in December 2004, Vishal plans to pursue a career in the exciting

and challenging field of construction.