<%BANNER%>

Separating Dolomite from Phosphate Rock by Reactive Flotation: Fundamentals and Application

xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20110114_AAAAEM INGEST_TIME 2011-01-14T22:51:04Z PACKAGE UFE0008000_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 51329 DFID F20110114_AACUZR ORIGIN DEPOSITOR PATH elmidany_a_Page_140thm.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
8f8a02294b661452e57c5b00ddda5a03
SHA-1
6117eefe53b3866a650b228650d224eac950bb28
694856 F20110114_AACUCK elmidany_a_Page_128.jp2
18641a663c390860587ecac6c39613e3
db9e2cd1136037efec7164648ba1ad00891bb10f
199855 F20110114_AACTXE elmidany_a_Page_102.jpg
cde46353875136e18b550c0468509433
1f78ba9416cffd53509ef79ac2b9df2db70ea1af
703283 F20110114_AACUBW elmidany_a_Page_110.jp2
1d224fb37af75b347ca9724b27055ae3
40749b03a190bdf234572cd9d4b2999b131ad76b
170975 F20110114_AACTWP elmidany_a_Page_085.jpg
6026d88918ffbb13b061ea912a0c3eb8
84ea6fe9247dab92209bd33982af8aefc8f5c49c
25372 F20110114_AACUZS elmidany_a_Page_141thm.jpg
405169a67ee857072d61ca9b8626e0e0
eaf7b6523ed224f9575bb78480d90b04a9768752
98190 F20110114_AACUCL elmidany_a_Page_129.jp2
b7bb716eb614438e23bf38bdac746b6f
b7df8af303d5b45746301acf6e9512dcf291839a
134736 F20110114_AACTXF elmidany_a_Page_103.jpg
d54d2e0c778d8da15129aeb3ace51452
df977e7df930115d42f27f0753e5a6484d504b3a
95487 F20110114_AACUBX elmidany_a_Page_111.jp2
4ad1a2f3aed157e861554e395d19dd27
06bffc155ccf9cdb96d7451534cfe330c0c110fd
195409 F20110114_AACTWQ elmidany_a_Page_086.jpg
dbe0680ffa2b35c7c73a5043a0074608
c1b4f629af6281b33ec8126136e33a4a928c2bb8
84498 F20110114_AACUZT elmidany_a_Page_141.QC.jpg
dcd670ea4308ff658892b3229b03979b
d97c83ccd213efb292bf60f7896f95ac926a53ec
40963 F20110114_AACUCM elmidany_a_Page_130.jp2
e04cba2971a1f0db6e1f0ce686134fca
9b45be8af75e735b189877cf754f891425b0762a
174613 F20110114_AACTXG elmidany_a_Page_104.jpg
8437e90d93e6ed607a95f92ea1dadb32
56447f7599194740b8ad71dd3f8c9464b904b4fd
150742 F20110114_AACTWR elmidany_a_Page_087.jpg
80e3299e66bcfe4aacb2b3f6b4a0a4cb
de177744a6b7e6789ac70ceca87e6d0c14f1ed4c
1053954 F20110114_AACUDA elmidany_a_Page_004.tif
7c15acb0b680096e3f26aa2198bdd3a9
fbf7027467da346f2a6aafc84d9be1872ba7a2ed
116104 F20110114_AACUZU elmidany_a_Page_142.QC.jpg
4591aaf03a2ca00a6716b5c99749b0f1
584a9a23c669bc64c861fffa5e259bbb82e7dcc3
882022 F20110114_AACUCN elmidany_a_Page_131.jp2
502da6993633be71a86bc6be17f8dfd7
7bcf7e0019a1ed15a79beed7693174d363ebb0ee
208059 F20110114_AACTXH elmidany_a_Page_105.jpg
409682f752471ebf3d97c58fde6e1fd6
bc35c434174873ad899e0a418e96d96cec986715
58253 F20110114_AACUBY elmidany_a_Page_112.jp2
2ccf81795cdfe4b95ac03271c2e480ae
c3ab18435462941a6a1b5e72726b20d705d12808
195932 F20110114_AACTWS elmidany_a_Page_088.jpg
6f705c3f448a8d57e82e967db6951b30
d1605a7569d95b04886c82ffbc7e0c1d7da63414
F20110114_AACUDB elmidany_a_Page_005.tif
bca6ff9b09289dfaa089680e0a8557b3
7b838c9a7c37a5d537d52a2693cc4655446b6fda
50061 F20110114_AACUZV elmidany_a_Page_143.QC.jpg
ec785b12cc4afa9afdb36598e9199746
3bcadcce2cb61514b8f3c14459949be879c1927d
1051935 F20110114_AACUCO elmidany_a_Page_133.jp2
c6c1680624723a80459e4e5fd0a785eb
b72224a8c1e0f4710032598470e3e2efb9dd56c2
160115 F20110114_AACTXI elmidany_a_Page_106.jpg
28f5cd5551936c58d02fbead6e4e2301
ad9693fe4b585547e97f69e3eb959c8fef886b1d
721753 F20110114_AACUBZ elmidany_a_Page_113.jp2
6436d306ea15dc0acad5678cc3844d80
d11d8902937958792b214b7991115f40495b2166
189204 F20110114_AACTWT elmidany_a_Page_089.jpg
fcbe08f2dfb1804ac93ad48215daf8a0
8802494ad88a706a2f7ff0c14797508003bd4463
25271604 F20110114_AACUDC elmidany_a_Page_006.tif
a7c20891792cc91493efee7391e55fac
8a90275b4985d3b876bdfb1f98fa18823fca45bf
11396 F20110114_AACUZW elmidany_a_Page_144thm.jpg
16981eca0f32514f104a99207ed0c3b5
63285eab51dd662e95f0fcf9d1e671f679fc6cce
97822 F20110114_AACUCP elmidany_a_Page_134.jp2
8b153b639f6c748a7e97a0d36df56164
b5b19c3c070d8db70c9f05d88475e6196234bdf8
191799 F20110114_AACTXJ elmidany_a_Page_107.jpg
ad18ba715d4a6e4da687d39465860330
18eb19271d5db07591be7ae1809aba634f1dd011
194608 F20110114_AACTWU elmidany_a_Page_090.jpg
b9a118816c3fb453748a31356e9edefa
33ebae6a572693b6bc4a5406f222d70ad57b36b2
F20110114_AACUDD elmidany_a_Page_008.tif
66b2919e89536b83fac022343a809279
69334d55db55b0fb7a4682a28decfb600deacdde
34781 F20110114_AACUZX elmidany_a_Page_144.QC.jpg
603261894a3d574926bb15248d0f64d9
e23e2d98a5f3274ecafc1d5126e899fba03041ed
1051985 F20110114_AACUCQ elmidany_a_Page_135.jp2
9f99274317c5fdc3b657a7d9d872b32a
cd450a6d93f88c1c0cffab27ff688816506ad149
177002 F20110114_AACTXK elmidany_a_Page_108.jpg
b7d8ded68e373b252fe32c3bb9a8b2c6
51103e30f49a95214fdfaa34905b89953295c3d0
176216 F20110114_AACTWV elmidany_a_Page_091.jpg
a1a4fc65d085a395be427b03eb2b32c5
30e9e91b7ad4dc2497d38716685bda9360b0373b
F20110114_AACUDE elmidany_a_Page_009.tif
12bc32f7e357d9a56cbf8f2c3c943807
3665bc9923c57efd4eaef1d006496a9b13f5cb1a
1051971 F20110114_AACUCR elmidany_a_Page_137.jp2
e3c2952e718e1c06bdc891942ed4ff26
4ea09b74d9896fd0c69b551fb0b78375a4151542
202079 F20110114_AACTXL elmidany_a_Page_109.jpg
6a7fffcc0e6d872770d05b928cc516a3
3967a3c7f196b2d7e8a253e49933955372e7a61c
193295 F20110114_AACTWW elmidany_a_Page_092.jpg
c11c56ec9045f9a1a6f081cf753f291c
13fc99093a1a740b982ea0a38e8c5aa4df77f058
F20110114_AACUDF elmidany_a_Page_010.tif
390aa2eb6ba85e274db778e64fb8d980
fab4a38161cee10bda10c6fac7af6bc6dc650c0e
192860 F20110114_AACTYA elmidany_a_Page_129.jpg
02d7999d92da2959a365efe3c382b590
afe2bd9a3b33c766896cf7a05efed4e3eb4b99af
134409 F20110114_AACUCS elmidany_a_Page_138.jp2
4f699dd2ca2d47d7b5e86d4f11dab16b
403f3548ac8395c8abb7632089ec3b47b958c11e
162038 F20110114_AACTXM elmidany_a_Page_110.jpg
d98e5c93db79b67133789cd455136679
a5d5ef06d9201349ba438a8a87e30684177a5c59
198602 F20110114_AACTWX elmidany_a_Page_093.jpg
57ae59fa6cd987d222d55164a402f9d3
eb1e128860fafaaecb714f0f3b686fbae8e91a2b
F20110114_AACUDG elmidany_a_Page_011.tif
877d8cc793c94a9ed74b0eb6c7085306
9efe853cf0a198ca96f879f4a14bc4ca37a8f025
77801 F20110114_AACTYB elmidany_a_Page_130.jpg
f92e886dcfeaa8bf54347076d14deb85
6eede25cc3d6db4adc2995a826ab43f23d171d1d
1051968 F20110114_AACUCT elmidany_a_Page_139.jp2
f4ca7336c7c6fd5b1a9b8e43782a00b9
76fd0ae464ce0a8736c7a1d36095c0c5b63b2437
182527 F20110114_AACTXN elmidany_a_Page_111.jpg
f36edcc1fcda60f95bf4db559721747f
db2f562795faace473bb60c8e38e9c4f3e9c6a40
204401 F20110114_AACTWY elmidany_a_Page_094.jpg
720562d478eac3a6b0baa86a51ebde89
8235f69a9f057dc21f20f229b4d5bdb24007c1dc
F20110114_AACUDH elmidany_a_Page_012.tif
3f56f720be73382e61ac99f5095c9f9c
01e2096ba0f92d339984ff2d508a0843c3bc100b
191473 F20110114_AACTYC elmidany_a_Page_131.jpg
d601b2a3a6e64ff4944b264a80d9f7eb
2f41e47fdb723a079c192dfcd49b0a9328d9f6d8
1051829 F20110114_AACUCU elmidany_a_Page_140.jp2
59f236fcb3670a7dd063f163acc0986b
bdf8ba2038067eab6a82eb4d282cd46717e75a85
164298 F20110114_AACTXO elmidany_a_Page_113.jpg
bc91b02d13d3f915817d13e6d83c66f2
a2280912110bb9f986dba76abb0c042201ab9870
F20110114_AACTWZ elmidany_a_Page_096.jpg
3c2df0a98c110005433f440ef59c2ff1
364cc0a38875c92e05d113d9156ab0762f4d01f8
F20110114_AACUDI elmidany_a_Page_013.tif
1850731fe025c9df009f9a49a8ca1a88
e8524fa548ed92a65b66d964f2e8fa6565fdccc5
187269 F20110114_AACTYD elmidany_a_Page_132.jpg
00c7eeb1e6832e82281ad139c412dcb4
e75d1e98ec62d846578305c0d8b546fa1993bb2a
130764 F20110114_AACUCV elmidany_a_Page_141.jp2
2a0abb5fd2017244ffb8fee3128a517d
cb0c632de970e400bebc917336c5777501bc103c
F20110114_AACUDJ elmidany_a_Page_014.tif
e7d44a83a6534f97c9a481498b976408
1f0d3aa49b58bea5fb0485097a43b743ff2885bf
229541 F20110114_AACTYE elmidany_a_Page_133.jpg
b95142de1940e60da4a017ee16613174
5ba3913d0beb090366a97c3bb299074349bf1edb
1051986 F20110114_AACUCW elmidany_a_Page_142.jp2
eb21d49dfbf09d7d84b5561a95cba117
d9e051163fa499e39adad0d4be8616648a5dd25c
147233 F20110114_AACTXP elmidany_a_Page_114.jpg
471d1f653c11d6a34f1b3e46737320b1
a2d0769dafd27401a8550f2412ed5e844c752e85
F20110114_AACUDK elmidany_a_Page_015.tif
928732a07a84fc93c8573d53f82ce561
e1b58d069a0fd787d803bfd655739cc2d676dfee
225515 F20110114_AACTYF elmidany_a_Page_135.jpg
e4723725149844610c71ed936770ff0e
82d6986b2d5b65a75ca684ba1b1c77955c998083
45912 F20110114_AACUCX elmidany_a_Page_144.jp2
f14174aa27c6b482d4f00c47399ec0e9
266aa9a50edc4adce272db0fb9b67fc16e6db7d2
184304 F20110114_AACTXQ elmidany_a_Page_116.jpg
7357d288d32a4375becd00127e4a9cfc
0732f0b826f18b922da4aa72c20c0ddf87222571
F20110114_AACUDL elmidany_a_Page_016.tif
213928e19bd01733940868067070cbb8
9950fcf69f0b86bd37f3a05d9e0d98d8b0a1e1d4
315376 F20110114_AACTYG elmidany_a_Page_136.jpg
0a1a39cf98ed258ab21f16f6a4f80664
8dbab6c3ca8fa9e618b0b1abc0fb446500a851c4
F20110114_AACUCY elmidany_a_Page_001.tif
c2575d72bbd2ace1180d782d54d8f3fb
c02638e2266c72094a2448651071deddbd1ca0b6
164156 F20110114_AACTXR elmidany_a_Page_117.jpg
545fa5416a82248d1fe4040364a8a92b
4daac2d99980834e372e3397dbb13934875101d2
F20110114_AACUEA elmidany_a_Page_032.tif
df33a81aa969092caf050aee7942a0e2
8ee88e7f5c5cc927cbab543a6d0d8290f1adf8a4
F20110114_AACUDM elmidany_a_Page_017.tif
fef99b03d89692b5841cc776cccef4bf
84afdb4170fdbd987ecf24452c70b774d5a6db19
301930 F20110114_AACTYH elmidany_a_Page_142.jpg
4f95e863642aaf51a8a7cc3912cf8128
ac1e7c929bdbe86cdf2a01b9d644f50677ff5df5
164228 F20110114_AACTXS elmidany_a_Page_118.jpg
9bffc6e17ce01c16fa4c778d091fcb76
720f227e5557bf4bfa5f10e8e66031c4305ea025
F20110114_AACUEB elmidany_a_Page_033.tif
8fc70956b193084e406500fc2e818941
9b63f41cd5d905e4ee8581d8bc01ffac3a6ae0cf
F20110114_AACUDN elmidany_a_Page_018.tif
43c3bd8a01cf12664febe39f3013595a
d4f0b725216282a7a1a4685b60a91b404dba5ac2
163524 F20110114_AACTYI elmidany_a_Page_143.jpg
a49d0b5aeea3ebc3a395dec35edbfd36
513915d78a12089050cc098977402d2b9026534e
F20110114_AACUCZ elmidany_a_Page_003.tif
9f14d0441863599630d805cf3cfdae9b
cc75f87d19919739675f700c3fafb57ca3a4ff1b
123956 F20110114_AACTXT elmidany_a_Page_119.jpg
a35a74afa1eb9be401dab15a1114577f
3020e9f5b7e5b4bbc9170779bd9bc16eab63f43e
F20110114_AACUEC elmidany_a_Page_034.tif
4292ffa744d1a96742dd19a4677ef3af
11eb382c0c66ae4c84d2e64a8dc2ef34a315a650
F20110114_AACUDO elmidany_a_Page_019.tif
4f865825bdb85e2b9f4292a157e46c26
a76560707fc014507cc60e059f31b2c1758ec3ea
89671 F20110114_AACTYJ elmidany_a_Page_144.jpg
302b3a42c58806ba83a022336760973c
0afc48d216dfc70852faa269c2062d41e311cb8d
164320 F20110114_AACTXU elmidany_a_Page_122.jpg
b2da5a4948922799fc6a4de745ae81fb
1fe0e1563fb54d0de94f9a6b1cf9a721ed96b17d
F20110114_AACUED elmidany_a_Page_035.tif
5ef3ed6c705179377ff0b716bd86234a
1837ebaced5c7943bcd4fc0bf033e2745569ec68
F20110114_AACUDP elmidany_a_Page_020.tif
aa6f9c0e4260d153ffb76f3f0997c913
6730da74b7a8996bec38d71ae8ec6f2aa3889b2f
25110 F20110114_AACTYK elmidany_a_Page_001.jp2
3fa95af5174c37d68ddd558ac42e2a20
f3796f6611eae84193615b4627b7258727716af2
99472 F20110114_AACTXV elmidany_a_Page_123.jpg
19b5bd5b1a27b2e134408bc095421a4a
67aaf423a28ec8d409afbbdb6e74f38e5a3af7b2
F20110114_AACUEE elmidany_a_Page_036.tif
b31e8abc1fe56c30cda87b2e69d78a30
786db3735b5ae3e074966cefd92c3fb4933b652b
F20110114_AACUDQ elmidany_a_Page_021.tif
d4bb8fa99566ee3b0313f5617e10814b
06afe2d8ef4b0c68e86528bdf0a85fbd27dab979
5865 F20110114_AACTYL elmidany_a_Page_002.jp2
ba4af4e08228e42e9f7d40a030589e73
dea6c1cfb0f0d1b272eb0995e2ed7b03ed74a904
165055 F20110114_AACTXW elmidany_a_Page_124.jpg
b3282039437ef4d6f93eb97722632693
5c336e82060bb4691d2d59da4f715c65ee04776c
F20110114_AACUEF elmidany_a_Page_037.tif
00837ecf386b635c542184e6b72243fc
0faf0989d5da6e80566f42b861fa9a9cf2ff0cc2
F20110114_AACUDR elmidany_a_Page_022.tif
2a1882a4b98cdafea184665d0f1b0c11
ad4ab22263cdf83c9730e984f94425270aa20d8e
101737 F20110114_AACTYM elmidany_a_Page_004.jp2
ffa1b3ee0c4a557f9d8d52f06e21dc3f
7e7040c3cb8372137603edcaba9a1102a1de5b9b
172341 F20110114_AACTXX elmidany_a_Page_126.jpg
1eb0100bead6dc838a9407d1518ea415
9d0ac6b696d97cb87c194a066c3b8d5099bc06ce
F20110114_AACUEG elmidany_a_Page_038.tif
37debd437c96b4e853065bfada9c652a
a15f3dff5459cbd5928bdb416da64bebe12d7583
113289 F20110114_AACTZA elmidany_a_Page_019.jp2
e09988c2c013e35ed4405ccaba0d9ef9
d56df67b8b12d048bb9eb7738384e6b17588357c
F20110114_AACUDS elmidany_a_Page_023.tif
aa199b5bbf566d01f57ecba0c1251c2d
1cc8013b2e9683873bdc882f384908636be4f072
38286 F20110114_AACTYN elmidany_a_Page_005.jp2
75bbf22a5d78ea30f0761be5accda8e8
50d85ee4a9ebdb8f3a7ba600a6cf3c521348ec23
201384 F20110114_AACTXY elmidany_a_Page_127.jpg
321c8f1786d7a65a4e8eecd4e1e16fb2
d373731553a1ed2ae31009ab9ea6c360a4afc2a8
F20110114_AACUEH elmidany_a_Page_039.tif
71c01a852113e63ee36214f60ede746e
7b548c1127fc10c8f0719c87d38897b141330879
84519 F20110114_AACTZB elmidany_a_Page_020.jp2
c6fb755e226f288092421f9f4fc749db
9cf850155f999d4101bfcac36f73df20166d2520
F20110114_AACUDT elmidany_a_Page_024.tif
3c575062d32d3dfadaa0607c0879741c
0343f57066dfd017b535c34095ed9224a252c436
1051979 F20110114_AACTYO elmidany_a_Page_006.jp2
20de61fa1935bc0453fc90113d6d8876
2ac2187d055e9e7a111e15a7806fe5828d16b48f
162068 F20110114_AACTXZ elmidany_a_Page_128.jpg
fa49758aa8fbdc91fefa885064990b38
f15e0a2983092438c1b338891dcf02a4ae04e8ff
F20110114_AACUEI elmidany_a_Page_040.tif
227eab86af228d62036931f1a4a3d3d5
d020153a79968115716fbe98792617a1862c5b04
44568 F20110114_AACTZC elmidany_a_Page_021.jp2
13c5ca2ac4823723cfce86b17b4681c8
4bccc57fa31cc9d6299636c224c9d0dd00971abb
F20110114_AACUDU elmidany_a_Page_025.tif
2effe0f1d85a6d623f52f43bc25d99f4
eab7d05e4b80cd18ce241c8729cfc4b8a56628f6
1051981 F20110114_AACTYP elmidany_a_Page_007.jp2
50604966649758570859bb8e16904a1a
3152e78fecf562ae1b1d6cbd81cdb2a49e4eea5a
F20110114_AACUEJ elmidany_a_Page_041.tif
86d65869bee851494aa5e1a33dada9a4
46ec84811ae6c60acaf58d11ee1daa00b385fc06
100713 F20110114_AACTZD elmidany_a_Page_022.jp2
a306a85fa0ca37cabd42440116af32d6
5d3614ad49d93f5b6f06d6464d219ff9dc879636
F20110114_AACUDV elmidany_a_Page_027.tif
2e17e090133bca2a5f155589273fe9ea
87716e65fdb502f6dd091af12ee26838caa7e9e4
F20110114_AACUEK elmidany_a_Page_042.tif
ccf6e3c64ddb71b3faf5f7b71d364bc1
4af94ec434c4a13d071465007dffc8cb64ec336e
656013 F20110114_AACTZE elmidany_a_Page_023.jp2
767407cc6ea56b6f5a7ee43a1c1735ad
79bdb1862e03556715eb534f212c8617192d9907
F20110114_AACUDW elmidany_a_Page_028.tif
1b1e862a881fb0b347de6e12111195b7
ac89ced1e024a15380de5b379ed51ee94f8b2cf6
1051978 F20110114_AACTYQ elmidany_a_Page_008.jp2
8b2011bd68c63f47391eeed9d2487106
92854c8f5a2f19550be7a32c74b440a689f401f5
F20110114_AACUEL elmidany_a_Page_043.tif
9e2d9dd9fc5ea6642026219249728546
355478d51ee8367becef82e614841571e07d561f
63179 F20110114_AACTZF elmidany_a_Page_024.jp2
1af71a08b0238be95a3894ef09e3795a
8149c69980c0452c8629b5c5cc116c5e34667d71
F20110114_AACUDX elmidany_a_Page_029.tif
afec4a34d091e8f92fc40018bb1fd427
f1367fc1498ab1a5d6c55b83f8577d0128ebb47b
1051973 F20110114_AACTYR elmidany_a_Page_009.jp2
648ef920253ab528a24553df57efb1ed
1ae2d7fb855c868da75b6f3895ed95d81875b31d
F20110114_AACUFA elmidany_a_Page_061.tif
e36360e10ebb06c310bdaef8561be953
da918539bb88aca9dc1994adbbbef2f1169da1dd
F20110114_AACUEM elmidany_a_Page_045.tif
335cea2fd0b8781acd04469274f6677f
9e9124e8f5968e45294daed47469b424a63ecaf1
49505 F20110114_AACTZG elmidany_a_Page_025.jp2
c07498f8c405d8355b8dc53c5144a5c2
548bf678f758a61f044369c83e814999051b71c1
F20110114_AACUDY elmidany_a_Page_030.tif
5a5849b36a267f6771d0d984e4c4ad0a
d56d6892616128a539b2329fc8767fe6ed13722b
F20110114_AACUFB elmidany_a_Page_063.tif
5f341deb779ed5f2d5fe130bbda5f33b
9dc43d8587ba8bf8c6dd22cca33120383a991ce7
F20110114_AACUEN elmidany_a_Page_047.tif
1e26c3a4679aae8189b8d13cc133415b
81873cd52cd4e4b021bc825772770520c5683838
55896 F20110114_AACTZH elmidany_a_Page_026.jp2
56c6da0c5b4d2fa0b290fda9bf98c671
f219bbf78f22c6ca3f39cfeee4c1838a7f474ad3
8423998 F20110114_AACUDZ elmidany_a_Page_031.tif
83937f190cbc58e576ba33924223c75f
624de5e9c05e6e4aafcf387d83bc686e3d4156b3
1041965 F20110114_AACTYS elmidany_a_Page_010.jp2
717d453cbdef52a06852208e4c685d70
8d50cc9082f02e413b11831ac7fbff6bccf6a4cf
F20110114_AACUFC elmidany_a_Page_066.tif
c493f93f3f7a9a6db5939ebae8a2fa7f
3ca1082572ab01b4e914b22c3b643b1722f372cd
F20110114_AACUEO elmidany_a_Page_048.tif
72e28375883ca405bc4258adc065a634
7d0f7ed880d1901239180e4e5037cf0848f6669c
105038 F20110114_AACTZI elmidany_a_Page_027.jp2
10df16457e8d7a9b3f8b1156903cf1ac
c534cd8d31486054bf81df7884127a0ec7e4b928
1051977 F20110114_AACTYT elmidany_a_Page_012.jp2
609c2dd8e52685f1051799345afb6352
59e1f7ad143d6795f35a6db3cec856f2362d3fa3
F20110114_AACUFD elmidany_a_Page_067.tif
4145a1a28c9c0b0dfca19da5d9e63e44
3e9465b0c8a0d40dc6749b423ed6d0be221bd710
F20110114_AACUEP elmidany_a_Page_049.tif
b18315ae89f5ce69a9bf38373bfb70b4
0b805aa8701fdde9896cab104aabd21e2a9f6a3b
323708 F20110114_AACTZJ elmidany_a_Page_028.jp2
54a956171164c781822153a04645b593
5572a1ed46826bfe48b558bd7ce910b8150b347a
1051976 F20110114_AACTYU elmidany_a_Page_013.jp2
5d4837a3e6f48b34d02cfde7f8917d4f
b2f240027996a9142ef9e6638ab5fefe40539196
F20110114_AACUFE elmidany_a_Page_068.tif
a80f2b85a545a8789e1eb859132103d2
b50b33d74942b3939ad9370fb3be0a901afe44e1
F20110114_AACUEQ elmidany_a_Page_050.tif
4f1a36c40fd8f8257235087aa185bfb7
6e992da4ca52408ca9161f80b88ab4c81c89cecf
92179 F20110114_AACTZK elmidany_a_Page_029.jp2
ab3746aefd51451ce0584e82bb995d14
0e0fd566463fb986b3039d09ec8696e35d5cdd47
81257 F20110114_AACTYV elmidany_a_Page_014.jp2
0dcb4bd7a867fd80163d83886cccea33
dd2be3f1c0cde1f543e3db1558853c3cc4f15300
F20110114_AACUFF elmidany_a_Page_069.tif
c80a8f153316cacbb0821eba4bc74b15
20869ad390507c9c87b7ed0f64963fbb6f38b85b
F20110114_AACUER elmidany_a_Page_051.tif
017f8addb76903e67aff65dac98df102
23a05b6a48f42c52005e3f15a39affa7fb6cab7c
95798 F20110114_AACTZL elmidany_a_Page_030.jp2
c86781ed18e3ad398ae850ce1451c1f2
ac7a368ce9fb27c7d29e545bfb7b01d85ea9e116
81293 F20110114_AACTYW elmidany_a_Page_015.jp2
fa93a4c5aa76e79d07dff385f0c04b19
cc31e02d5d84043c18cdae8a94dc4764c23e6cc9
F20110114_AACUFG elmidany_a_Page_070.tif
2f644227aace23bfb991a4b4af6c0fd4
af0e8a5ceb9505a5fcfad202973903c7800032af
F20110114_AACUES elmidany_a_Page_052.tif
4230d297fd7b140db0e32267ec6d23be
ac3c8074c290d143468e6fd9f6d421cfec50e732
105024 F20110114_AACTZM elmidany_a_Page_033.jp2
51d595ab2a24d5fcaadb05fdd18b93f1
9492be04db33367c086cd0911d4791e846afa7f5
91941 F20110114_AACTYX elmidany_a_Page_016.jp2
2379999bb851f2ec9c6fdb1f47189557
4350c671b7975325f58f5644a9b31b5e30fe8f71
F20110114_AACUFH elmidany_a_Page_071.tif
6bcc368b4c5a28a0e4dcc1e462c28bdd
38ab4f1b04ddccdc73c6a1316278f5934025deec
F20110114_AACUET elmidany_a_Page_053.tif
f160615a9f6893c4a2203c6966eb1525
7f3777273721cc47be33484d1a58723feaf0be32
55743 F20110114_AACTZN elmidany_a_Page_034.jp2
af744c29330440201a6c487313415eaa
df17f7095f15c5516be0524a6962ea531e1b1435
105777 F20110114_AACTYY elmidany_a_Page_017.jp2
b772580d144085b1f1d94a079b72fd00
a8541dddaded337d1154edc3dd4ece4fe2a66f8a
F20110114_AACUFI elmidany_a_Page_072.tif
68631d7ab79e32316c778328590f7526
dfd930daf03e6fea807b427fd7f04f2ddc99f8aa
F20110114_AACUEU elmidany_a_Page_055.tif
443d944549d26e03b86864f4f60cc791
3f3930a625be8027c8ecdbdbed0d1d04c42bc45f
475455 F20110114_AACTZO elmidany_a_Page_035.jp2
d45a81101c08bb95428f059aa6bbc6b6
5fad1df9edfdc9831d09e1616cc28ea31d7fab5d
994246 F20110114_AACTYZ elmidany_a_Page_018.jp2
d736ff1f6194aae611b24df6c2ce30f8
53ea3c225780f2c193efe965860a0719e7adfd89
F20110114_AACUFJ elmidany_a_Page_073.tif
8bdd7717069bed12f2c5ac08fdb57419
3e64ea53a7ba09f9746c803d7fcaacadf2e70867
F20110114_AACUEV elmidany_a_Page_056.tif
9e85905e4c0a883633d04cca8366cb8a
841d1c739035dac8bb20d52669a759917ad84997
75435 F20110114_AACTZP elmidany_a_Page_036.jp2
b809b643469589fd8060c5eee86a2a3e
a716503b66904f35d1974c2360f9a63e14a37cc1
F20110114_AACUFK elmidany_a_Page_074.tif
6f13538c32eb752bbce65bc426646207
54acf1393a8ffce256abff94b62c537879f7c0a2
F20110114_AACUEW elmidany_a_Page_057.tif
73e016004862bb77fb43ec94cce071e6
f8288c9a9ea69c956514d72d6bce08b32e09abde
96436 F20110114_AACTZQ elmidany_a_Page_037.jp2
ea5e415b775a4d25e4737c9b1a5b1e0b
4021e1b6ae3e0c4ac2f638639c65ad767ff884a6
F20110114_AACUFL elmidany_a_Page_075.tif
9be525a39da41c881bc67c47e4d825c5
27c219af2a29dc868e99710183ba42b0169619c7
F20110114_AACUEX elmidany_a_Page_058.tif
3ed75d67065338e1a3a6820d5d72749e
ff93af791d0d51eb8c0ec11fda5aa5968ebe6896
F20110114_AACUFM elmidany_a_Page_076.tif
5fb115e1afe6983b6a04d8828c89dd5e
b8e858da3219d4e039c0386dc922c70df728998b
F20110114_AACUEY elmidany_a_Page_059.tif
c9bbc75678d49b3b82cdb9b97b982ade
6470d6e2c7f67826955585a5af96fc30497b4910
86789 F20110114_AACTZR elmidany_a_Page_038.jp2
749876adad6430762af6cef49740822b
2acc9921f3dbf1633b6fbb5975cc2a8c7776f8bd
F20110114_AACUGA elmidany_a_Page_092.tif
01f3e2ac6e24a3ec4d4136b711e4ad51
f23ad93e985995a9dc1795a579a887abcf83de5a
F20110114_AACUFN elmidany_a_Page_077.tif
d42fc22fd818aa3ee6e5da8ebf1a5616
ba6cf43a33cc3923a8f52a34a0e4b25a04d88875
F20110114_AACUEZ elmidany_a_Page_060.tif
e384789610ee6cd169c9d657e861397f
0e718a20f7cab930a828ba31675cd0d7473a7e23
1014271 F20110114_AACTZS elmidany_a_Page_040.jp2
20cee87fa2a99f2377cd2996bac3bafb
253b19b830728138d156157b9ff6b6a3b46c7d19
F20110114_AACUGB elmidany_a_Page_093.tif
9332a5f98cca32abed2532703e750ebc
c15d771eab32582e45a90cc3eaa6571320cd1c5d
F20110114_AACUFO elmidany_a_Page_078.tif
e80c8dc6f48c178e9dc32f88ed1561ba
59c1e8b5ea2a40b7e713a9e23549216a4c6a72da
886183 F20110114_AACTZT elmidany_a_Page_041.jp2
49a585e638b6b4cc3ecad4342d982950
6e3970cba0807ff49130bf03c2021c82ac7d7dee
F20110114_AACUGC elmidany_a_Page_094.tif
1f5744cd98c9d159a9f6c194fe4a9e0b
6ced6510ac48aff8f0c9281d770b4af88725c39f
F20110114_AACUFP elmidany_a_Page_080.tif
beb0eb76079b190826d26ed0e35337d7
bd68f65b2bb211b974b780ea040e58279737338e
822326 F20110114_AACTZU elmidany_a_Page_042.jp2
a03e7d97af63fd6b9f2522e4f1b1774b
8b5df4047791d914e2e7694144a7edc30c357836
F20110114_AACUGD elmidany_a_Page_095.tif
392f01b3b1eb3015be40133e695d7260
52b913d0acb84f34fe81237b0f21d21b0ceea79c
890763 F20110114_AACTZV elmidany_a_Page_043.jp2
71711d26e03ca83b389229ebacd62529
666c96f5670d7298e8ea99af85e7342fd621b85c
F20110114_AACUGE elmidany_a_Page_096.tif
3e9414e4258579477ad6ed5445933335
9aba281ec138ee570b7773fca6eddfee5b97d4ec
F20110114_AACUFQ elmidany_a_Page_081.tif
5b74e3e9f7c7cf2b8bc6ec8684590a01
378973a1098450cecb3f4b927abf7736827ec599
106433 F20110114_AACTZW elmidany_a_Page_044.jp2
183e0349f72d12870b4ce960f43ccfba
fb4a0b4450306bd188b88152f66adff5cc8465bd
F20110114_AACUGF elmidany_a_Page_097.tif
a707ec0e0eb9bd1c9e80e5be4d10228a
286a1616d663c5469414813ae8f22c2a622f40db
F20110114_AACUFR elmidany_a_Page_082.tif
d5a0607a9427020410db520c829be715
d252b845c78752e5aa20d3cdbca3536cfcbc94ef
31754 F20110114_AACTZX elmidany_a_Page_046.jp2
c52d597989b8bedaf193ea2406ab0501
cb75d70b275c50d09d2ee343627d86598c7c4617
F20110114_AACUGG elmidany_a_Page_098.tif
f874dc27d239ec6c53135ba5cf1d9377
aa40df2108d1979fe505198bde96a4c48afe887f
F20110114_AACUFS elmidany_a_Page_083.tif
11458ef677c77c7affdde468f465306d
b3b8ca236dcfc8bbd2c6ad45ea482ff7885d708a
93633 F20110114_AACTZY elmidany_a_Page_047.jp2
c7e4e45587864d2f387d64c9d300b7ec
7a21193ca5d4a33c80a2e97a41c1724e88cc6ed7
F20110114_AACUGH elmidany_a_Page_099.tif
906950bb7f1d8776bb402fc3cf4b0c70
e7442cd1c787a4784b7272d8c0f8d083200e1749
F20110114_AACUFT elmidany_a_Page_085.tif
479a9fb112e6c833c2e960958776e4e9
87e9e4dcc5f033af7bfcf237878a5e2a67ee4122
1041505 F20110114_AACTZZ elmidany_a_Page_048.jp2
1f9c0f275d668e28590543abd42b7826
daf3c1d411e05f5d90910c254c06a4ce11cf0235
F20110114_AACUGI elmidany_a_Page_100.tif
ac4dec31fa980b80bfd449761f4e8353
050045329df77e74b63a51c272dc486052765833
F20110114_AACUFU elmidany_a_Page_086.tif
d13f14453c7134ad70afe50a7df2633d
3707f85015d75d0a4f64a82aa4c7264252bf998c
F20110114_AACUGJ elmidany_a_Page_101.tif
ff5dfa927002eca69c2226b886160ad4
af81fdfec8340dcb5b22dbf6089fe1068e9845b2
F20110114_AACUFV elmidany_a_Page_087.tif
8d355d76e8f2b3041928c8251e17ea1c
b48486bff5e09cb105d4ae3e669d1534c540e700
F20110114_AACUGK elmidany_a_Page_102.tif
b512833fc56f1c1acff911d9291a5f60
d01c3e026b831b9e55791ebe9717ac77c505a4bf
F20110114_AACUFW elmidany_a_Page_088.tif
ac888ab14264eb1d2f3536aae28590d9
412dd6fbbe2b2657f1438da0d573317cca265220
F20110114_AACUGL elmidany_a_Page_103.tif
ebda2564cb63ca5c80857ff0d582dd57
10d3cf3ebd8f3e87f440bde0cdf707bc30a3ec82
F20110114_AACUFX elmidany_a_Page_089.tif
aaaff8bbdc374d36211210e5c4b16986
a3ff8ed8b2c3b139be554fbc6ba9b606ac2ad258
F20110114_AACUHA elmidany_a_Page_126.tif
b7d5f7749db4fd4a5f7d229728a09562
841e28666630420d911ab4125edba50632e2ee14
F20110114_AACUGM elmidany_a_Page_104.tif
8a1b6369e449e8398d704cb58cc0acc7
00b803c7b74b974b65a166af4a127c7241fc3685
F20110114_AACUFY elmidany_a_Page_090.tif
ca47530ded1404e45d2bf84c345de20f
72006cd4cdb9f2150725daa2d559029daee9a50a
F20110114_AACUHB elmidany_a_Page_128.tif
b40be6e32f628d2cf6b1ece9b2d0dda9
16476d8f29e20aad6ec5c1f5c42174e2cf1e0512
F20110114_AACUGN elmidany_a_Page_105.tif
23f44b784bddeb7488d1a2a86ee68814
79455f9990ecfbb36f30ef2b901e1766678ea882
F20110114_AACUFZ elmidany_a_Page_091.tif
d59b1afc267200f8ea98576df36594c3
27e4c3a20afb455423291bfe03614e34b0e10c26
F20110114_AACUHC elmidany_a_Page_129.tif
be154ec2e4aa45e93f722bcc52337bdf
361fafc57414d6f57478ed54b19a8aadafcf99e6
F20110114_AACUGO elmidany_a_Page_108.tif
4b83d395a2d4e2b5012f7cd9857f666f
5dcc9004f1535ec112edeabd89b23248495db516
F20110114_AACUHD elmidany_a_Page_130.tif
f963a15471effd889b77fda8c5f0d337
dfb9bcf2f656cd95a0e57fa9a1d77c9284c0ad72
F20110114_AACUGP elmidany_a_Page_109.tif
228f096109a684784b90c47135ffe218
6a80bc17ded307c262bffcca414e55a606e524c6
F20110114_AACUHE elmidany_a_Page_131.tif
28061f3c0fc4f2c0fc5506393f12c19e
6a47776fe49747e6a1f5a0c67857fd1d8a7b33c5
F20110114_AACUGQ elmidany_a_Page_110.tif
3fbd8836286d8f11e6b1085d444d60b7
ace65864e2c9c959d5021a8ba7b40ff0184a8063
F20110114_AACUHF elmidany_a_Page_132.tif
3b229ce0bd97d8da666f15990d1a30ae
c9dff140f672144363f52ae6571f25f3625916b5
F20110114_AACUGR elmidany_a_Page_111.tif
42a16cb25fb111a221df5ad8fffd3d68
49d1806335ca8d7e5c86ab59631da8b4ce6e7258
F20110114_AACUHG elmidany_a_Page_134.tif
b38c259a60fb663c30c81923bf22ffc1
317664fb5f7db1aa768d04ebc925e4a9859266de
F20110114_AACUGS elmidany_a_Page_112.tif
82f60c3eaf8e721c647b1383b4232dc3
63b6374f5d8f37ab89abf868d7d76b854fc48ea8
F20110114_AACUHH elmidany_a_Page_135.tif
398417c267c296e8bb0cad0e000a0c5f
585e3918de5e5ef16b030055419dbfebf77a578f
F20110114_AACUGT elmidany_a_Page_113.tif
0e7498c85277a5d41b392108a976368b
9bb36786aae031f94cc3d61ed10856540a88e407
F20110114_AACUHI elmidany_a_Page_137.tif
2dfcfd9242243fdcd2ee4109a0c176cf
ef0448501a03f3054defa0d6c1de071a22e47bf1
F20110114_AACUGU elmidany_a_Page_114.tif
f298567fd627afac39bd6abd9ae16afc
5ce77c53b7c7bf8969d611d77365ad7238fdfe87
F20110114_AACUHJ elmidany_a_Page_139.tif
83581bc8947cfad848cfaac25b0fe5c7
e0ad23c0f9424e736bc0af1030a22394add4a7d0
F20110114_AACUGV elmidany_a_Page_117.tif
b17d377b56163af92de3ac02616e68aa
c825e48185b181deb6869822747c0d6ec188f16c
F20110114_AACUHK elmidany_a_Page_140.tif
bc894aec9ac6aac01ad3925cd70de161
eab476040323c2ae5818b84b2471a511e4919ce6
F20110114_AACUGW elmidany_a_Page_119.tif
b55f411e69dd3d68cfcb436974307298
7dc694f2faed680c692b9c42de77cce325335e43
F20110114_AACUHL elmidany_a_Page_141.tif
0c1d193779608a498d3229446374ab6e
9ac079f0af25124de57a1cd1d372a786726504d7
F20110114_AACUGX elmidany_a_Page_121.tif
8445d756449b866479e492b85bd58429
c95c3ac3465ecd83123ca0ced76f6e7ec311fd8b
F20110114_AACUHM elmidany_a_Page_144.tif
67bb54af51be159d2b88fa1b920b83df
4956b5168f714e7a83dce164d61497083f692fcc
F20110114_AACUGY elmidany_a_Page_123.tif
69a24dc12ffc3d56bca8128db315756f
f8e1d4e9f181011ad403e6541198cc00b937fafa
52065 F20110114_AACUIA elmidany_a_Page_019.pro
a85fd3ca34c9780273cc07633c61a8e7
501406c0242d45f80c1b5d1d7e53bd84be3fec8d
1255 F20110114_AACUHN elmidany_a_Page_002.pro
8941f59db4dee20860c1d8dfa2f270fe
1497c97776d3e6d23594c9c7d020adc95486618b
F20110114_AACUGZ elmidany_a_Page_124.tif
2d795bbcac146b50a2e6871f3b0f8226
60f1a0ac0395774c352e969ba2ba297b02e1561a
36882 F20110114_AACUIB elmidany_a_Page_020.pro
d8b2ff9e6049d4a1443d019458e3febe
88449d03972344059a02877fb734d3af238326ec
48177 F20110114_AACUHO elmidany_a_Page_004.pro
8673749a0ea738af4e2ce2799da8cbf8
bb048da63144d6d981f93d81b5e0d81c4a4167c6
15153 F20110114_AACUIC elmidany_a_Page_021.pro
0ed47c26cee47d43ddbb293aad0224a3
8677d162e343e27a2cccad85e18f416279183a81
76173 F20110114_AACUHP elmidany_a_Page_006.pro
bfb0caff91c11b77ffae35a8390ef782
8ab93c2d816fa255a7d14403346bd97815514f1c
45729 F20110114_AACUID elmidany_a_Page_022.pro
13d1c9ab6d57e4471d14d7c803601532
5c5fb4fe221243d9903cbe1dae19def0f2b0c065
103422 F20110114_AACUHQ elmidany_a_Page_008.pro
74783d43baffa4b94da1ed6a13eab955
59a74631e2c7ca6d51cf1869303b894e4c3fe738
23792 F20110114_AACUIE elmidany_a_Page_023.pro
2ccf76389910bd679ed5acdd08b3984a
730ba9a92544fa2e1dd52d778209f5af5e57a3de
55837 F20110114_AACUHR elmidany_a_Page_009.pro
109936a400fd2016cfc7a4fe879f2335
ebd5983013472b74966fd4aafe4fee26bca21803
22744 F20110114_AACUIF elmidany_a_Page_024.pro
b2efd0859664aea6ab5f672345a827df
3deddf68f22eebdfb9fceb7fed2b521ebf8c022d
29324 F20110114_AACUHS elmidany_a_Page_010.pro
9d16828a333c8df1437262c10ebaf0ed
cdd4f2baf3f7efb674821baa47655a0e7f39a3c3
21772 F20110114_AACUIG elmidany_a_Page_025.pro
aef20cb04ba8ac445beddb896afaafc1
94eb9bfb69352933ffa8abae3d4d4d6c6993c71b
67806 F20110114_AACUHT elmidany_a_Page_012.pro
421b1429f4afd74d1fe2afb7639d31a9
465a5f0ec3003406fb20d2177da76ba51f4c293c
25070 F20110114_AACUIH elmidany_a_Page_026.pro
ba5009d06cab2332c05b3df80e95450a
c3102835f7e61d3456397ca04c4a10cfb0760634
56877 F20110114_AACUHU elmidany_a_Page_013.pro
9b488de53b91b3a055651877e02abae3
d83fb8045ed88beb5e04968d59acf89cbd9d3c0f
47594 F20110114_AACUII elmidany_a_Page_027.pro
fd72ef86be869ffe401ebfbfd6d5d294
e7d4621006c7af6001fcec37613ff1208d02eca1
36222 F20110114_AACUHV elmidany_a_Page_014.pro
8ca9c0063fb99531ee1b78aacea4018c
ad8c57d87139539afaa0a3679c9b1ff851dbb0db
6099 F20110114_AACUIJ elmidany_a_Page_028.pro
bee9479ad892f33d818bcf6f9e9812bc
8e0fde338a86e140f62c728b37e6431a980ef5d8
35994 F20110114_AACUHW elmidany_a_Page_015.pro
b56c719c3610f5077d9bf0ddcb670259
ac185d42025ba0dab626c011f432db3f8acb7f34
40227 F20110114_AACUIK elmidany_a_Page_029.pro
4296e5b3bd5c1577e6aa721d4c46d945
b72323eca93f2c9b38876a4bebee84d62d701451
42071 F20110114_AACUHX elmidany_a_Page_016.pro
46055f1be4688ab6a908c6deb7591d2e
9ec924d53c648dbac1dc196c8c5b6e3be66191f1
42364 F20110114_AACUIL elmidany_a_Page_030.pro
80d0a95e067d4cf97b65c2fe829095b4
7b171de1474b802e85714d489044b8883bb71dd7
48281 F20110114_AACUHY elmidany_a_Page_017.pro
232c9f180af552de562246ccc42a0eb6
e5b20b66895a7d565770c66eab6b15fd21b27500
43580 F20110114_AACUJA elmidany_a_Page_047.pro
709ec9144d661054349b40fcd8a15747
1ff5bff1a03e86f17c72886ee828e2c864c293fb
36536 F20110114_AACUIM elmidany_a_Page_031.pro
0388fee99763d4d7753e7d562eb564cb
b5e7232267e058e8100b6491d09077203f5bd02c
40754 F20110114_AACUHZ elmidany_a_Page_018.pro
85babc81f201782fa4e026c74d1905ff
9f59fbb2b4eff70c7bc927d6c702410efac54303
51614 F20110114_AACUJB elmidany_a_Page_048.pro
17af3fd13a5c1777b51d9fadc8f60472
1e9938edb3eb256a7562e546e3a58994f6e00d2e
35266 F20110114_AACUIN elmidany_a_Page_032.pro
17cfe9556e22859a021b00d8c60a73ce
0e0b490fb3958c4274ba5477dd5a8c29f8c72574
44918 F20110114_AACUJC elmidany_a_Page_050.pro
e252db781080c4126cb176d8dea1bcf0
c64fc49bdce62756add9922647d96b85c395afdf
36357 F20110114_AACUIO elmidany_a_Page_034.pro
61b71ebf83e9db91541a435eb321cdda
43a453bb1d26f9462aae70ffa7affd98b8e75fb4
49997 F20110114_AACUJD elmidany_a_Page_052.pro
2a3a1eb9498097207adf427714082872
d4750f931c09e8a223c8bb783a6a700b01038c64
20294 F20110114_AACUIP elmidany_a_Page_035.pro
6b122f6ec8d4a21fc9f10d6e9960eaaa
3aa970792c8709fe4a35b812a1768002727037a5
53573 F20110114_AACUJE elmidany_a_Page_055.pro
2955529810f602fb1cea819362750d70
a929392355e8ad8acd85416612f04558b954a624
43754 F20110114_AACUIQ elmidany_a_Page_036.pro
a9a8cf6aebe49ea5e7bda19ae65d538c
ffa03ff238b2856781f30ab87da3918b0b88a1bf
46744 F20110114_AACUJF elmidany_a_Page_057.pro
9c7c28ca81ba140208ec8850b5f5d5a3
4f504451f7f197d33a2e21c2e04adbd524e20a55
44102 F20110114_AACUIR elmidany_a_Page_037.pro
472fd7a6efdb3ecb9d122c80b4f03bd8
e49012a61824dea2ebadbadab82728ea19267f6b
49141 F20110114_AACUJG elmidany_a_Page_058.pro
5ded07dd5465a9fa72d1937085aeb1d8
deb9b8ad5c82a92b2949a4427c78ffb495cf0e84
40061 F20110114_AACUIS elmidany_a_Page_038.pro
39c5c9e18356970d4fb8e1bbae9d9787
e7e90542c7033fd908c82286a0806bcce47eea49
32995 F20110114_AACUJH elmidany_a_Page_059.pro
5e6792fd29aa22fbb2ecf95a56533737
79e882ebb5b61a3d13e1e1390f78de8092acd93a
43929 F20110114_AACUIT elmidany_a_Page_039.pro
b63d3445af2aff23d5b831a35f8d7d23
67de99b9d8c692ee1d79619cb1e7750684e5e84a
50196 F20110114_AACUJI elmidany_a_Page_060.pro
4d42344fedb8203fb45ede70bd06fbe7
5fdd4ec47055c573a69b57db3266f74320b87e27
34291 F20110114_AACUIU elmidany_a_Page_040.pro
e3e5f0fa04ce287ecc891d73eca3c569
858fb3add0288f8fcad8b34ff61acf1ea8cc74c0
47462 F20110114_AACUJJ elmidany_a_Page_061.pro
db7ebd45344e0eb8ee60023786f34e24
f32bb5f5d4ef37c5238a7d5552baa0164a016013
39399 F20110114_AACUIV elmidany_a_Page_041.pro
981ed9e2b8edaa188a1a4075eeb7bb39
b1b55aae2c158001818ff4dbce7a511c7691dc1b
51772 F20110114_AACUJK elmidany_a_Page_062.pro
d8c92ada45a80fc782ab9fe15d7c6ef5
fd27588390c4c739feca1d52dadca5610b074313
36878 F20110114_AACUIW elmidany_a_Page_042.pro
1b77bedbadf1d1fdfd18cad969b229dd
b0fc15e52e0e358ea3781d4927ff6228e84c69b8
18823 F20110114_AACUKA elmidany_a_Page_083.pro
d6ea138a7a47f2ac905a1633aa3a3c6e
361cd5366ec4501494bf2ac8e283f9c08332bcd9
27692 F20110114_AACUJL elmidany_a_Page_063.pro
b31019ae15c60820208f14b8f5eb9400
1cce7fb2acdcc5c0d40fc6fa7dc0a90949ddfcae
37021 F20110114_AACUIX elmidany_a_Page_043.pro
bc975e08df39927661c1c188c5daefc2
9c64afb4a2fea602c23eb26fd668b42941f1e51e
50125 F20110114_AACUJM elmidany_a_Page_066.pro
ef72a989b94e1361667c7fd8f4a3b9e4
ba146896f2e2cd9e7b6484ce6b1b2c734c70618b
42058 F20110114_AACUIY elmidany_a_Page_045.pro
4b278d87281d535a86b5d4f52cf70393
4ce2414019217b3ef04084397d7b948681b516e1
33832 F20110114_AACUKB elmidany_a_Page_084.pro
8bc60c80e6e014db4a9a3d70d8b35d4d
3d475f1f37224ce7912cd5aa208eae6bc0218cd2
48596 F20110114_AACUJN elmidany_a_Page_067.pro
110b7d1792c225f16e7f866db5f00e53
30858a121b966198aeec72fc30799d3a0e8fd39a
13450 F20110114_AACUIZ elmidany_a_Page_046.pro
f5875c7e1071f5c05c55cd3df0aa1dbe
72ced8a240d3353da436e90e68de4cf5ca06c159
30849 F20110114_AACUKC elmidany_a_Page_085.pro
01e225f772e3377c08c3731d3589c219
d796d17f9d47a4bca7158836dcb53bf7ea09fb3c
14127 F20110114_AACUJO elmidany_a_Page_068.pro
d4843eb03459bc2329e842ff6d1bda4d
c08f833e8be335498e9f7e8c3ef3f899f39da06b
37195 F20110114_AACUKD elmidany_a_Page_086.pro
9be81bb45b4e5e4e8b14fcf98e9b16d5
be77f400775e9a8fbf6974c96270b7e7d97e99e7
47720 F20110114_AACUJP elmidany_a_Page_071.pro
713bf49936e4546d2d54dc55266b8a76
f2cb9b97f3fbc966a60b51d5518b4f38dbf023e2
22154 F20110114_AACUKE elmidany_a_Page_087.pro
382c302ab81d94d74401e8ea1fa3256d
42fe74a975283a7f2d06f5d897021b0a50681ddf
51473 F20110114_AACUJQ elmidany_a_Page_072.pro
e5622c1dcc7cb4d0425c736e3789d069
3261b76d127ff83fcdba15649a60824b895c002f
37218 F20110114_AACUKF elmidany_a_Page_088.pro
ab3fd3e9e6ff0b5408b5d1908dc3345b
e9dc8227769c689c33b6c6c00902d967c7cbf860
41680 F20110114_AACUJR elmidany_a_Page_073.pro
d6ff2ae17ccf37d03587bedae413b7b8
566f157ae3e75732e1c57b1d793b9abfe13d3a32
32209 F20110114_AACUKG elmidany_a_Page_089.pro
e5c7327dcb31d0d99229098705928b92
71758f0b52a63427939fdcf66739a002bddec442
43996 F20110114_AACUJS elmidany_a_Page_074.pro
f48c9cf9c0a8b2c225d3500cdaf58a79
8c96b520fa9ca5946e6f9e31e2d858402f5bb602
32940 F20110114_AACUKH elmidany_a_Page_090.pro
a45b3f00d90828192682bf2aaf347b53
c06b8b1405eff1c0f404b3a65baf02e60e0683cf
36633 F20110114_AACUJT elmidany_a_Page_075.pro
f44c3a7af423c38ce691623c34113386
6cc6f78d0bf4807e6f5b52ed4c924c6b7e1009c0
31592 F20110114_AACUKI elmidany_a_Page_091.pro
a1d466baec4ceb844590a50026f9ca88
218ffe25ade9a1dec9e3bbcb329986b4ae0e8bd0
22996 F20110114_AACUJU elmidany_a_Page_076.pro
22286f61ab4315d87488ad8f79a972ef
0026c66d76d972e321e93224741760f9c1470d9a
39589 F20110114_AACUKJ elmidany_a_Page_092.pro
4de764148d34465e585f0c7d29eba363
b60f5db54d8484ff52c47e6c9a88cf7335bcc685
37503 F20110114_AACUJV elmidany_a_Page_077.pro
491fe361fe6f470cac23d693ce63f3a3
c305ab8436ae045621ceec6c7869cb819875192a
38201 F20110114_AACUKK elmidany_a_Page_093.pro
e263bc7488d85fae5d9622b936e4cac2
3caf77cd356b5d6d974c8cbda250ae3c864d2ea4
49020 F20110114_AACUJW elmidany_a_Page_078.pro
ada947092780cfacb9e2e0f929344660
24585686fd9ad8b1bd91b3c8c674504b93efbfa8
48053 F20110114_AACUKL elmidany_a_Page_095.pro
c2f0e2a7e449a6b2b838f4d00fb30335
e2ecbbf1f8bf4b8c8dd7c0ecbc57965fb43caa06
27511 F20110114_AACUJX elmidany_a_Page_079.pro
4c4205af2d0dc411fd3c9332a83dba17
29261cb52f1454c55793f81f563270140a5365ca
37525 F20110114_AACULA elmidany_a_Page_114.pro
6b3fb42aff22162f1aa3f44883354904
ffdf50a6bd9de22fec6b7810e6050d9653a0d13a
30063 F20110114_AACUKM elmidany_a_Page_096.pro
61ca48d973c9e9054c6275f16e328af1
53b6354cf0669d898670e536c9c9ed80c83b76dd
27436 F20110114_AACUJY elmidany_a_Page_081.pro
65cfc3158b55376839a305484aa7389c
34f9c26a104696b776a3c6fd9bbc9179984bdeb0
26495 F20110114_AACULB elmidany_a_Page_115.pro
e8b8e2f7f60abd3be7fc4cfe1620373b
77c01b72808aec056090c36fcf8e64adc5a9e67b
33061 F20110114_AACUJZ elmidany_a_Page_082.pro
5ea3e3533b3c6eedb976c9ff89afb82c
68f6309732178819e05bff7c4dc210be605a002e
32945 F20110114_AACUKN elmidany_a_Page_098.pro
a5f11cd87d88adc0fc32a0b2b2a7c034
e5bf783a76eb6632f5d5070f30cb212083f55fc7
37627 F20110114_AACULC elmidany_a_Page_116.pro
f5374386ec3c1d1510a337c63c1895cc
233181ac17212ca6e2822035eefc110aef499513
40462 F20110114_AACUKO elmidany_a_Page_099.pro
e3de4153d79d4345abab52283de4822f
539aec45b3b6f55aa9381e496d92620fb31eb6a7
35148 F20110114_AACULD elmidany_a_Page_120.pro
45235aa406039748111835437d296c51
b5e36f880323bd0ec54f90bb9a8324327c25a91a
32829 F20110114_AACUKP elmidany_a_Page_100.pro
51b16ac05069eb576a8fa8b9400a3532
417799a5034ed64544b92da6b3423dcd4a97cb91
34226 F20110114_AACULE elmidany_a_Page_121.pro
94723e729e9d3433a22c2fa326363d34
4d532496d12c84bb2d9e1c6dde003a52be236572
38128 F20110114_AACUKQ elmidany_a_Page_102.pro
920be53868583a19494cb1b440bfefde
12b062cb997466c7e7ba470c3a11e3f7624d55ff
41585 F20110114_AACULF elmidany_a_Page_122.pro
977ea6ce023b043750fe6858ebbc36dc
141a4d55e25f383bfaf3a0300fe0ab56fd1a7be6
22615 F20110114_AACUKR elmidany_a_Page_103.pro
144aca9d5458d1334d8401c5574a4a15
baf7be195d642a27d61d4caea5712696954669e9
19763 F20110114_AACULG elmidany_a_Page_123.pro
39c5c1c0ed1f22e9c893d789a86e20ad
d17f225e362cbcca3237bb0a0c890cb0df17242e
33056 F20110114_AACUKS elmidany_a_Page_104.pro
8e35398aeacc2884931b520073815cbf
dcd9c5774e261e9287be987d07ebd852505a1863
28720 F20110114_AACULH elmidany_a_Page_124.pro
2896369487d5a721d848008e937954c7
450f2744183fe6070056c21e19082b8f3cd5ba96
37048 F20110114_AACUKT elmidany_a_Page_105.pro
c412e7d51df605f02d6c4a267cc60378
930d4dc8628f0da897f9824ba3ce65c74e8a39f1
35188 F20110114_AACULI elmidany_a_Page_127.pro
e7a7629cb698045303bc536b12fbd3c6
77b19ef38df2ee532641c2690505de0e913349b9
43227 F20110114_AACUKU elmidany_a_Page_106.pro
8a161054e81729d05a98f4f425d4e536
5a2378020606c44bce0670d401bc8b53a111b601
23907 F20110114_AACULJ elmidany_a_Page_128.pro
62d721cdf4be38d98a048e716a0b5dd2
238d29900d1253b62489511f2975e088c49ef41b
29486 F20110114_AACUKV elmidany_a_Page_108.pro
46eaaa7a851f46e7c8b938fd59151fd4
354370ea9452db2203be5b44ffdba7524bb429e6
44808 F20110114_AACULK elmidany_a_Page_129.pro
bbad4aa0234e22ae893df0b261afe343
3f40202fbf4426348abda4b7779148430b1f78f1
38708 F20110114_AACUKW elmidany_a_Page_109.pro
11538a97c37560eb0a8ff7ababc029d6
4f8910226ceb2877a30307fc222d4a54b17b5d6b
1968 F20110114_AACUMA elmidany_a_Page_004.txt
6eab7af74c448891935df9cb1a84045e
5efd14064440a081d7c14d2f35a6588c39de10e3
17413 F20110114_AACULL elmidany_a_Page_130.pro
5c998420d25b5d6b954a71896f620833
cbb31a154471d7b5f2dd6f0ea2fd14a6bfe11fd5
19583 F20110114_AACUKX elmidany_a_Page_110.pro
f465911fbc107f7e289bad58761c1a9e
dea65fc008248d52a526730d4cde2b9f5d845b20
648 F20110114_AACUMB elmidany_a_Page_005.txt
6601d7da0a2b43d0bb66a9d55a9d3bc2
9d687aebe8a70cac439b28d4ebd0d234f1b69eb3
44665 F20110114_AACULM elmidany_a_Page_132.pro
d3f48d4d7ff121f7d12567d1b0ff8591
bfcc61386ed4216429449529b0da0a721c95e16c
31468 F20110114_AACUKY elmidany_a_Page_112.pro
cd7baf600bf7ee3ce966bbe3311a212d
386093e43c4fdeb5949b8892894201bf5b93ee2a
3193 F20110114_AACUMC elmidany_a_Page_006.txt
72b3f52c4d4e4c9e3edec53f0a53c97a
e612e7cdd09449f0c3a8522d56f65db6bd623893
49599 F20110114_AACULN elmidany_a_Page_133.pro
da21229570cbc7d253b1af6066463352
417ea1d19bb14aa8bc0a429dec1e6663950f77f6
20328 F20110114_AACUKZ elmidany_a_Page_113.pro
f1e6d53b7ca63e5a508c4e73dcfe2879
3a01070ee5686f32e916a8b7975e9e36c20267bb
45441 F20110114_AACULO elmidany_a_Page_134.pro
c27194fe7cb96025d7721678f786c0d0
220e42987fb27f1c177cba0a5335a97200412b09
4466 F20110114_AACUMD elmidany_a_Page_007.txt
a496c810c62b2e4deed06e1530bbb289
e420604504733aff362e72b3a75c696b3b31340b
49188 F20110114_AACULP elmidany_a_Page_135.pro
c09f5cea379104321ced251f697189aa
645c9cb6ef42d970582c2212ba1f817265bcbf8c
4328 F20110114_AACUME elmidany_a_Page_008.txt
5d81eca65bb82594cab21a60ef6160ae
e94db63d9299ce8f90ba736d679a698ced75fe9f
66555 F20110114_AACULQ elmidany_a_Page_136.pro
e079b4dbc0718b0c1fdafb9a5523fd6e
5b853358ea99cf62fddba71cdbfb8bd247b4a7ae
2231 F20110114_AACUMF elmidany_a_Page_009.txt
87eb4d7f864f0cded31567f81cc238b4
56b0bdceb753c5f8ff63f83fe7de67b5e5539505
57647 F20110114_AACULR elmidany_a_Page_137.pro
9da41ec6c927de179f4b29c492587355
5b590a6167208fc543898e0dd16598cfaa963f30
2192 F20110114_AACUMG elmidany_a_Page_011.txt
1b6a0484176494b118576ffd93e54174
672a4262a4239db4c6fc9185faf0632efeb7df72
58106 F20110114_AACULS elmidany_a_Page_139.pro
35d717b0bada24657baf794c9139cde2
8c23de4c9ef6dbe439372e35ac460eed6ab49921
2638 F20110114_AACUMH elmidany_a_Page_012.txt
d667b587a7d635d52bd90c3a83d55de0
1c94f2dd5132975aa563f6addf51c3095e3b8779
63427 F20110114_AACULT elmidany_a_Page_140.pro
d7e5fcf03df8095ba49f4aa6968e038a
f2492cf1c4f00bc51559a9b46bd794f6bafaf94c
2234 F20110114_AACUMI elmidany_a_Page_013.txt
67da500b295bd430d675211ae89dbb03
fd46b75257d63ffab714e82a316779414003d889
62994 F20110114_AACULU elmidany_a_Page_141.pro
11840f3b1de8b3dcea3af503fe2f1f58
128f7d393896cae5d3a1706e48c02d002d0016f2
1610 F20110114_AACUMJ elmidany_a_Page_014.txt
26e724d64958ec28daec1ee9dc374bed
87f345d974f70d86860114b64d66944479f68ce6
38394 F20110114_AACULV elmidany_a_Page_143.pro
c09a1cfe2f00b3c86fc7ee80e1cfc8b4
b3990c0f3cd7cdac904bfbadaa517c762a42e54f
1435 F20110114_AACUMK elmidany_a_Page_015.txt
33ec0e45a6ce73a3bf70117d1efef4e2
6f122b88008c2ff0dd81881c4d78bc28942b3f6b
19223 F20110114_AACULW elmidany_a_Page_144.pro
cadfcfb4ad43a50ff96e74c13edd5723
d145ef3899a383a56875be50e512d633f715e775
1777 F20110114_AACUML elmidany_a_Page_016.txt
3e7fb4dc9da417fdcd878e300e1cb5f4
04f50df4e183495da7eb27fbcdafa0db24324084
460 F20110114_AACULX elmidany_a_Page_001.txt
2b3257ee04138d170ea098e3bd71a6cb
0cead3564ae49b19964865575ae07f9946c6b67c
1900 F20110114_AACUNA elmidany_a_Page_033.txt
ca3e79e5cb1a758cd837d55d82e623fc
e87a2e10797e926e7b2ba28be7327c4b131ddba7
1922 F20110114_AACUMM elmidany_a_Page_017.txt
877836523fdf1c8043b46e63a6cdf5a9
68296c62e81c674d99bb62a9ea756c7e88350638
116 F20110114_AACULY elmidany_a_Page_002.txt
8f8359c98749b0da65eac5164ac82008
c7d7705438b6d1be26979e6de35eab01a60aa65f
F20110114_AACUNB elmidany_a_Page_034.txt
c2a75645d4a70913d39a25f9f002d164
5d937cd4b226626d64308a0111840fd54426239b
1796 F20110114_AACUMN elmidany_a_Page_018.txt
07c9ff970b89f02de1e6d320d98f1a65
8d9331bb2f055bd66b43e9a1ceb4f4f27f0eeabb
175 F20110114_AACULZ elmidany_a_Page_003.txt
6e614b150d1fe5c0255a9e148d41697e
3a4467a55bfbadbf5a4a2a43b14d470c131955dd
1037 F20110114_AACUNC elmidany_a_Page_035.txt
cc2ce3ef7177683f1de725681b637d4a
d5db1120f0174bdbcd5fa70cc906c217cb073f6b
2105 F20110114_AACUMO elmidany_a_Page_019.txt
04153f8ce9704fb2deeab863568acc7d
147e584d6e3c27367bdfd60b651157327ace733a
2163 F20110114_AACUND elmidany_a_Page_036.txt
9c614c100079b89679c9d6e865fe4d69
02a247102d4ac9359f0159ae5e671577cd250b21
1499 F20110114_AACUMP elmidany_a_Page_020.txt
c361f2e329eecd8cc0425f8384867b94
05eb2505b41970a568f7ec324e92731ce18e0708
964 F20110114_AACUMQ elmidany_a_Page_021.txt
1dfcb883a891f68e447442ea67f39069
4aaf9df7deae5ce87ec58fc18128fc1150e57df5
1794 F20110114_AACUNE elmidany_a_Page_037.txt
7e00086da20423c9e1173a8865a14991
d5bef0c3eb9581decec20d2f415d415dac203457
1822 F20110114_AACUMR elmidany_a_Page_022.txt
b55c15e040c57c0e41904f73c8c6c8c0
f2f0c5eb02e413b1cea7a15c3d8dc1606c4371b4
1691 F20110114_AACUNF elmidany_a_Page_038.txt
5f3c3650e3a92a761d75098e8efd0b68
b001b45c4c6f75209de12f2161eab52865c695ba
995 F20110114_AACUMS elmidany_a_Page_023.txt
a80f06f8e9fd55e2fc9e3aee83ac80be
cbc6141df6c502f3a306d2947e1afee7b9c3e598
1850 F20110114_AACUNG elmidany_a_Page_039.txt
3f14d0811ce259e4439ba5a8ed74aa9c
db32b616a2d078f81d391ee7630794336d1fbb35
980 F20110114_AACUMT elmidany_a_Page_024.txt
67e67fd1ef581083e03b915e092c2bf4
a24aa4f5a527e3c27ec5db034d557917eeae23ea
1681 F20110114_AACUNH elmidany_a_Page_043.txt
4c331d670fc169a0d542f8a10bba720a
729b9e58334ba3d131c597909a5b3a7fd024fd18
1441 F20110114_AACUMU elmidany_a_Page_025.txt
f3aa50237f03a28a793bbd752140aa70
e95ca9c00474b99ad52c4f2a34eeb6e093bae8c9
1911 F20110114_AACUNI elmidany_a_Page_044.txt
ac8fb6fed99223722468665cb94d9f04
0814515cadd1c175986f5f6e395ff801f3cdfc2a
1411 F20110114_AACUMV elmidany_a_Page_026.txt
0deb24ff9e9b116d023eea4d7cc15903
3dca8b5ebb11b7f6f72f3a0d19a27367f8ac9496
2052 F20110114_AACUNJ elmidany_a_Page_045.txt
2d85d38665850bcc2389caa5fe554eeb
d42a7760f9de2d03d2582eb483f76004a8122ef3
1883 F20110114_AACUMW elmidany_a_Page_027.txt
466e754779bd58760c55a1da6db240f6
ee852d265988ccb5d4d5147b27a009b602e7ef1d
1799 F20110114_AACUNK elmidany_a_Page_047.txt
bf7c626b9f3ed09165e57e654e491aef
b92dd5500a83c55820b169a7a7f50524aebff7e4
614 F20110114_AACUMX elmidany_a_Page_028.txt
d8ccffb38ee3a7bb384096484b2e7492
c3efa7482d5019b5d42fad66f93a95d761689e68
1917 F20110114_AACUOA elmidany_a_Page_065.txt
9cd697f0ceb5023dfc040840ac539fd9
4212887da86bef8b418cba790cec66e59fa74fdf
2339 F20110114_AACUNL elmidany_a_Page_048.txt
18eb53ad2238e96f3c3272788bfe656f
409f278995b7c4c1fa4fa97bb215e1463fdf9bae
1535 F20110114_AACUMY elmidany_a_Page_031.txt
8d1942e00b326a4d0f75fd8370c69efb
d2a239f1d954098b813d040e6f0c27d3f1f46b59
2156 F20110114_AACUOB elmidany_a_Page_066.txt
50aaa7e5459cb4415d1f356bf6b2b76f
5ce913c30aec2ae117524f54a78ebf65472a9195
2159 F20110114_AACUNM elmidany_a_Page_049.txt
300974dd3442d56659f4413e354ed4d4
b12f54200c75b0a27453faf9533549e2ad880410
1743 F20110114_AACUMZ elmidany_a_Page_032.txt
96b15f5d232c1ddbdc9d21bbc88c4c85
11a0f3fe4d534b22ac4283359d7da3d327b47fb1
1952 F20110114_AACUOC elmidany_a_Page_067.txt
26651f51f9004ed9c0273ea455b738b8
7c1dfce3b6a2b5b72f5eec554b99bdf91cde58c5
2060 F20110114_AACUNN elmidany_a_Page_050.txt
61dbcc1eb7f21df343bc8cbe61f15b6d
56feaf87183cf7696f1d3e93dea18f9c4e87d5c5
970 F20110114_AACUOD elmidany_a_Page_068.txt
12949635e7453f669919681ed7d73099
a71052e4a80c5050885c6abe0b73379c47965ad9
2057 F20110114_AACUNO elmidany_a_Page_051.txt
3f2c69c98f262d1fbed59371d985171d
6e5cd4065753b877136e1e7bf6ba8f29a363ab83
1550 F20110114_AACUOE elmidany_a_Page_069.txt
86838c349a570b98c63cc1bb72746bfd
b2092788b17d468d179a37313590cac8be2f4d62
2137 F20110114_AACUNP elmidany_a_Page_052.txt
74ecbf716a079f49845700105df4b755
8d0ae249050238da22f8b4f11ae28e6ebddad60f
2071 F20110114_AACUNQ elmidany_a_Page_053.txt
0906ae92acd24ae2da777d868f0a26d0
7b789e7734db415e4178fbfb60bb9bb5187c9034
2562 F20110114_AACUOF elmidany_a_Page_070.txt
e66aef59a07bc6bbaacf2f7b1edf8bee
275d5a67de7599ef660e8ae5b2880ac03b88e2c5
2124 F20110114_AACUNR elmidany_a_Page_054.txt
d79c27dbcc969671454118c7ac94e4eb
394f74fb0787bfd9237c66933b0a8a3f2a99e6eb
2299 F20110114_AACUOG elmidany_a_Page_071.txt
4b1ed2991a040ffe96d77d010a2d82b7
16e0517904f430e1f1d8be98f70a915ee376ea9f
2355 F20110114_AACUNS elmidany_a_Page_055.txt
e5abe2a32ff9703744405aa202e8a908
1095f3a2b629785499e8da34919c595c1441d619
2284 F20110114_AACUOH elmidany_a_Page_072.txt
8c37e3bea1f9830fd4881a1a194f2385
8571fac33fbf4d0d46375a951bf6f2573f165e41
2000 F20110114_AACUNT elmidany_a_Page_056.txt
693022eeeb0a2d5b96a063a775c68548
971de658ee83b582538042dfe339691731eb7db7
1742 F20110114_AACUOI elmidany_a_Page_073.txt
ece333d8ec6a121e9551aa5562fa7928
789c2dbc3faaea323eafba5aa29bcfe1ea2f4ee5
1919 F20110114_AACUNU elmidany_a_Page_057.txt
d0cc7e66c239d15c0a398b42ca84da5a
2b717e5bb6a3e7ba8fb157ddf2e845bac0d883b0
1150 F20110114_AACUOJ elmidany_a_Page_076.txt
7303d142f5ac3683fe585ef5f59a7043
0d6eab87913e3efe09243df7cfb5634662e4bdd3
2232 F20110114_AACUNV elmidany_a_Page_058.txt
9fa40c7b86a9b2c05cc3d3486d08120e
163f999655fb378bf3cebba70d5fceb8a72f8738
1944 F20110114_AACUOK elmidany_a_Page_077.txt
851bf88a816c4f5933f017d364263245
737f1f32a4eb0e9ed013bd011a34b1c465ee74a8
1999 F20110114_AACUNW elmidany_a_Page_061.txt
9274e9ac6d8e4004bb025bc95c702bad
c6a903388cbd312457d92358f97c666876dd65a7
1948 F20110114_AACUPA elmidany_a_Page_094.txt
b20b78d3d0be9715ec566060afa140a9
ca3e4b5eeda27a2fc185aaf6dafa1208493bafef
1938 F20110114_AACUOL elmidany_a_Page_078.txt
354f1880e575f5de3d9de1de8b118c36
f28ccd1b9595fb3002cc28cc624f62b4b1198acd
2523 F20110114_AACUNX elmidany_a_Page_062.txt
4272969c7cdf69902c0e1b8136c03ce7
3da80398923cf45d9f900fab0d23390f12294c09
1910 F20110114_AACUPB elmidany_a_Page_095.txt
525aa7a519b487ef2434aec220e12cd8
677e4973440193e4912b4aa8b8e80e909cda3998
1298 F20110114_AACUOM elmidany_a_Page_079.txt
e1ffd4daef045b5148a2573a9e43ddf8
4dfca536d473a31d60273b622f325018e093c21b
1143 F20110114_AACUNY elmidany_a_Page_063.txt
f54545d47c76e1f6f0a0667cda19c5e9
b702b2204dae0e0162a97973cbe3b8c074ddbced
1412 F20110114_AACUPC elmidany_a_Page_097.txt
701f5e707ad9d4f2eac7c832fe8c6dda
344120a58eb058f9bd2c0ec143a178c0c0827e3b
1249 F20110114_AACUON elmidany_a_Page_080.txt
4d20534cfdd270e2ab6ef24c37c1ae10
b946228837cff98b5e9a9a40465216028090e6ab
1650 F20110114_AACUNZ elmidany_a_Page_064.txt
9e9c5823a1444673080a3b7cb79acbd4
65461d6d35bbd2e73f28ec32e2967ddaf4549a44
1377 F20110114_AACUPD elmidany_a_Page_098.txt
86c4a6423d8a623309dc047d2a756eee
c8e0ec241f18fc52e51967d609976b0275de5537
1279 F20110114_AACUOO elmidany_a_Page_081.txt
65969ee75eee3d63eb8064311c4fdc8b
0447691128e207d034ced7b08a609ddbc551d461
1647 F20110114_AACUPE elmidany_a_Page_099.txt
b6df08abbddd3b7d1e52a58bd256a225
1bf326771426cdc5e00077a7ae9200d022f66d0e
1421 F20110114_AACUOP elmidany_a_Page_082.txt
e34734cb9dc0d9313bac42efa0b6e15e
bf73f1e698f4ebf838c1758101e49a7aade59811
1403 F20110114_AACUPF elmidany_a_Page_100.txt
d8f1907887e2b8063e0bcd119eb6aca9
104b38e88e10516e615acabd611ce43d0523b8d4
1434 F20110114_AACUOQ elmidany_a_Page_083.txt
9697a4f9afe684404d9cb03f68105188
848a39485fa879d26c84a35c2848ffe88d0ba51e
2009 F20110114_AACUOR elmidany_a_Page_084.txt
681c13de10e0e54768757e28caf99063
d33cdde869e541dd2256bc915e0d0cd8dae2d792
1824 F20110114_AACUPG elmidany_a_Page_101.txt
0a88f315ef1d24faea3ce8f8f0bdd422
8fba818a4bbfe2a4dd9c9050998494959f6696e7
F20110114_AACUOS elmidany_a_Page_085.txt
401a93055c24be501a460a91789f2dc1
4d4e0dd439e8e9c7280820f0592ecb5d1ee6f701
1689 F20110114_AACUPH elmidany_a_Page_102.txt
c6be062352af21c27713c7cc2da0bf8b
2c5b41584bc56392bf7e3d93fd402f2159450eba
1595 F20110114_AACUOT elmidany_a_Page_086.txt
13087f0ee6c3853d9b748a2cd4529281
b444c1efaf46299be5f71b36e6d5774b87fdffef
1176 F20110114_AACUPI elmidany_a_Page_103.txt
d14643feb90da41c589a32db492d3abf
e5f715d04a2be47580b454da7d49054aeb79135a
1291 F20110114_AACUOU elmidany_a_Page_087.txt
88f26f0a040be72c10fdab8033cbd444
69b8cf3ef8cf8dbdf9adac06a2c4483d595556be
1656 F20110114_AACUPJ elmidany_a_Page_107.txt
9a18f47514f50403af0109dee7ab55bc
0acbd5aa34bb8537ae07f5833e26f5175d9c4c2a
1703 F20110114_AACUOV elmidany_a_Page_089.txt
c0d70a4099283d88db806d4e4b295430
6d4e6f633ecbce39e42254abfe9bc167666cf4e4
1562 F20110114_AACUOW elmidany_a_Page_090.txt
e13a98aafa1e166a03b52e451205fd3a
d618227576e12707e503e854e06362f2c8978298
1415 F20110114_AACUPK elmidany_a_Page_108.txt
bca458245f787d614d42a6143e6bd9ee
16afde7e33a7f983c9a929b59a20b7f62c0162aa
1416 F20110114_AACUOX elmidany_a_Page_091.txt
944323953bf559bbbb893ead8cd3a6d1
6d823b1b566223238da65a7f7b67ab9dc8985a29
1626 F20110114_AACUQA elmidany_a_Page_131.txt
38cc0c2e45d1900fd85fb5dbd62ca99a
ea6e0a997e6b50cddbf821af1b4c200a1d4f4b98
F20110114_AACUPL elmidany_a_Page_110.txt
4bf8a921a93dd23161095792dc22842a
737bfac384df97695f601ee3643cf0f042900441
1718 F20110114_AACUOY elmidany_a_Page_092.txt
3ad8096793f552b1914e485a68951f5b
e5bfed719f4d11a4942a5bdfb6de26561245a61f
1781 F20110114_AACUQB elmidany_a_Page_132.txt
e0ae74a048587558c1c88481bdbcde59
e4e6ca8274514af223e9a0ccb1abc3d0c0d1996b
1788 F20110114_AACUPM elmidany_a_Page_111.txt
e17cdc9af417140cd20bd08cae7399cd
e915f84fbbad4d844488eccf7c30b31bc144ed68
1672 F20110114_AACUOZ elmidany_a_Page_093.txt
a07c7db79821a5428a42abf9b626954c
866d02af9620024b0de3cc247f47c6713b356994
F20110114_AACTNA elmidany_a_Page_133.tif
615f637dbe0aa1e281c244b8d6aac11b
e6fe4f4140bc6ef33c1288f398cdde216768cd63
1983 F20110114_AACUQC elmidany_a_Page_133.txt
c0e5f3e83a52ec1a5fcaf304646ec715
a96d99c2b03f6b493231a5c79814bb5ca0795c2b
1620 F20110114_AACUPN elmidany_a_Page_112.txt
2c5ac77943465bac0d9ac15b1ffef41a
cc573eb00b3031a6980e801da15944c68f3f4c90
267155 F20110114_AACTNB elmidany_a_Page_139.jpg
586edc763b8737a4972814200ee5656d
c64b1e54bc63a133f64c6d3fbccd2ec38bcdd5fd
1835 F20110114_AACUQD elmidany_a_Page_134.txt
1dc58a8a6ba577947216af454ad35678
20e039f9a72416318c0f540b35c037ee75abbd87
1580 F20110114_AACUPO elmidany_a_Page_114.txt
87155f9c7275170617081b2fea978156
aa86a8493a896d7bdcf0bfde42eb9405c900740b
72724 F20110114_AACTNC elmidany_a_Page_080.QC.jpg
345ba2ba158a3d3de67f3a869f952c2e
924c105ae9c736ddc1f53befe8f0b40d55cc44d2
1928 F20110114_AACUQE elmidany_a_Page_135.txt
27eceaf24e0067552deaa86ed91b2f2c
2bcc6cb16b66e1d66474c71ad873de0530333df5
1551 F20110114_AACUPP elmidany_a_Page_115.txt
bf3ddc07339c2da35317a795f012ddb9
e5d5beeb21245def0900da5b33896a6eb11e6590
775043 F20110114_AACTND elmidany_a_Page_091.jp2
e4212504a76ccec6d4bdc3a83e5b7cc0
5537884a7238b1a0a9f1a83ad2db1679fd46c00d
2574 F20110114_AACUQF elmidany_a_Page_136.txt
90df6a020f1111001f572eacd37885de
4c667af56ff573556598f24cee331a37fee9e5ed
1561 F20110114_AACUPQ elmidany_a_Page_116.txt
3ff4041fda8e8695073072c77aca5b9a
d82f72a0c96251b3d85fdcc29598e6c0a8152173
26368 F20110114_AACTNE elmidany_a_Page_031thm.jpg
7bdaa3d60cf826db912724071d620074
35d86cf92195798bc54b18a5e16b02f59f9ed32a
2238 F20110114_AACUQG elmidany_a_Page_137.txt
ebe14561841bf6305b19b39caa374f0d
11c2031599fa81826195665454cdea3afe3dc79e
1448 F20110114_AACUPR elmidany_a_Page_117.txt
b34d259560b15842fdd97ade57008b31
571a75c98445e543f0065b94ce1f4ec76679769f
923 F20110114_AACUPS elmidany_a_Page_119.txt
6dfaa70ad43b4352921677126129a4d6
42d3dc11d0829f646e866526b8c9dd9ca8eb71c8
575 F20110114_AACTNF elmidany_a_Page_046.txt
9092ec232a41286bb5cbc659f53d36f0
5bd8d2c092347c63071ebd1ef3f91679d0368de9
2528 F20110114_AACUQH elmidany_a_Page_138.txt
ef9ae039ae73e13aa9d6244114c1fbb5
794e17131a1cc720ab08d0d51b93a9d11fd05e24
862 F20110114_AACUPT elmidany_a_Page_123.txt
43551f94c02a9bf3fc95ae75e2826813
26effa9dd960b1c894bb2a982103d593c2acd5fd
232820 F20110114_AACTNG elmidany_a_Page_095.jpg
fe37a266ee1fd77f43dab6a7e5c13026
da0e17b363934ee5f048e598276dc4885d999081
2256 F20110114_AACUQI elmidany_a_Page_139.txt
e38f2a129e42949263505fb3bc1107dc
53b09e359caa60a0af75d2639f0021c6fc9bf98e
F20110114_AACUPU elmidany_a_Page_124.txt
85b27e3c0b7cbafcae47a51590478692
6ac12ebddf72cb34d32c4d5d9d2427d0f4c228b8
1771 F20110114_AACTNH elmidany_a_Page_088.txt
8170745e3db479ee6dbb2031ab864810
9a8ffb22d14249c97c891d4f5302b6ce60af804a
2463 F20110114_AACUQJ elmidany_a_Page_140.txt
1a273d2bb7273cb33164c9314c53cf37
7aa6059ace2ca9f8b7b0b5213e52bc75e4e60e4b
1725 F20110114_AACUPV elmidany_a_Page_125.txt
ad96d3497a478b5e201c15695abd84d6
8291bcc2c134dc26b9dd5fa93a3b61154302fa4e
F20110114_AACTNI elmidany_a_Page_026.tif
8bdb7d9fdb61fd996d1c476f65b19279
f18ae99ed9670b3095ea7cde87982d4ad7ca081e
2438 F20110114_AACUQK elmidany_a_Page_141.txt
611c769e0aaf0e3c01c0a418cf4ac261
b9906a8daac7737fcf175528dfc405f917d85174
1352 F20110114_AACUPW elmidany_a_Page_126.txt
bed6014f769eecf4486ce266db7f8a0a
b9e362ddbaf1ef7020626ed41f505254eac73a72
66123 F20110114_AACURA elmidany_a_Page_049.QC.jpg
98634da7f7377bdb374bf37a6f5ac3b2
73cc023c866dfb5da4129ddf607241e85354affc
25266 F20110114_AACTNJ elmidany_a_Page_117.pro
fe3402866f1a54acb673f71007f2dc40
9669b7dae0d5afe38cb164cf3a59013b715c4e00
2623 F20110114_AACUQL elmidany_a_Page_142.txt
22a8356116e17314a3f1412feb18eae7
b20f0e99268372a0cada587bf1c4c6413088ca01
1603 F20110114_AACUPX elmidany_a_Page_127.txt
d17f76a4b10fc47471611a29808698fe
99aa8991f166f0bb3c74d22734f6c03f285d81f9
66346 F20110114_AACURB elmidany_a_Page_083.QC.jpg
bb9d60c4b794b3a040c1ae8928f086bb
489691cbba86a4014a10cae384d97ff82ad86012
49468 F20110114_AACTNK elmidany_a_Page_133thm.jpg
2dcad44057e64b5fbb8960532973be70
072d53a5ce34b94aa76feb5a6653b5ae41182904
1503 F20110114_AACUQM elmidany_a_Page_143.txt
d4a220f82749462f57ee8134b35d5a6a
8800553652374960cb8cb54197a32e671a1334a2
1261 F20110114_AACUPY elmidany_a_Page_128.txt
85d832d1d4aa76a3db22235b04b6029f
706e810ab35f05d5abb720bb16fc2dd5d9be39ad
254842 F20110114_AACTOA elmidany_a_Page_137.jpg
3fc80007fa0e386c76af68dd55dad214
6c78c2ce3b299f2b8c9b047663251b67c95f30c3
48900 F20110114_AACURC elmidany_a_Page_067thm.jpg
5e94d42ffcf445f916f87515d0204c1d
cbd744772fff70eea7e5590b67dc0b7691b65158
15181 F20110114_AACTNL elmidany_a_Page_002.jpg
602920c9f9ff3d1f9ae0d152b7e5d2ee
3827c54191ef3faa806f404e9fcd4ddf4f5c2321
812 F20110114_AACUQN elmidany_a_Page_144.txt
66a6b581d83154821676252d6a43580d
3ba86ac48f45e6b3f46f07c94476569533283111
701 F20110114_AACUPZ elmidany_a_Page_130.txt
5c58bdf8eab89d51534719229129a8ac
58f1b2f00fdae5246a433e7e88c3fe21b302d539
169294 F20110114_AACTOB elmidany_a_Page_115.jpg
a08d788a526529340822ef111e2dd156
a82d1cb75f7491e2f9bb7f258daccd02337020b8
47926 F20110114_AACURD elmidany_a_Page_100thm.jpg
9dee2ff7366df6c54221814b61bcd312
7c9058541b4e26503b7ef352cd29305cc8ea0ed7
41673 F20110114_AACTNM elmidany_a_Page_118.pro
e9f668c809c552e7d03772d70afada0f
1a83edbefef07f4c0b21f74c4179864a1fb8393a
22908 F20110114_AACUQO elmidany_a_Page_054thm.jpg
bc9d0627931edd9eb936f94e6355b85c
fc129ec4a54d5d40d3a2a6b6c639c42f69faad91
266099 F20110114_AACTOC elmidany_a_Page_011.jpg
f888c549a74f7df998abafc6221ed786
eed351eceda9a937eb07875c42c63982c671dbd4
90286 F20110114_AACURE elmidany_a_Page_089.QC.jpg
5be05ec1855f3558ecea07d5ca95e1c1
02545b986b4d537974e3da3ea82a12f3b1ed7653
45814 F20110114_AACTNN elmidany_a_Page_125thm.jpg
b7d0cf227559de2c0f37b5521c8ce7df
d977f9f3791af9424036e0f1ea8ef04da273aad1
21697 F20110114_AACUQP elmidany_a_Page_030thm.jpg
c7bce2b6aaa9956d4fffbaa731d14636
44060f867cbc43551a675fb5f5e7f4671db5d45b
2878 F20110114_AACTMZ elmidany_a_Page_003.pro
76ab871c8c5564adf65fc7bbf7488fb5
128c97f0f4438253c485c72a7789162056ad25b2
21798 F20110114_AACURF elmidany_a_Page_066thm.jpg
ec7683081457d1e8587196bffc87dec9
e622e989ce91c5a0fc0543efdcef10b44b10ecf3
F20110114_AACTOD elmidany_a_Page_002.tif
ab88866913286960a2bcaa0d6a990118
a3f3d2cafcc6d9898d3e551e5b2a977eeba22dd4
733276 F20110114_AACTNO elmidany_a_Page_124.jp2
59e65bfcea8ca3a80e0ad8542a5e6717
fc91bfd397424c2a1215c55b4792e75a89d7001f
94743 F20110114_AACUQQ elmidany_a_Page_053.QC.jpg
9852bcbb6d08bbb82114726b28274170
0513e986fba181d88349358fc4539630d8b8b682
117959 F20110114_AACURG elmidany_a_Page_008.QC.jpg
8e32b0005b6a63ed8cd8b5f5e36205ae
b79ee6661d0c4578d9e4cad87e2e77134c3a8719
1797 F20110114_AACTOE elmidany_a_Page_074.txt
ce5adf1f014adeaf37dd7138b0c8e216
5d5fb08fd2272777d4af6560a8de1e4526ec6d22
1750 F20110114_AACTNP elmidany_a_Page_030.txt
f5fa24c86b744a7a6d8910f626a64621
b5605a2b42ceb65b3c13cf377a4b45b901dee6d4
48696 F20110114_AACUQR elmidany_a_Page_053thm.jpg
2615de355df1d352d6a2f9a25524a84b
30b78f5f9c77269abd6c54413269a6a6e2f6dcd7
47518 F20110114_AACURH elmidany_a_Page_127thm.jpg
a5bf5ffe1b213af59172eed6d29acd0a
1af6ff4f650c300016a7d2cb4a9c91cca9a7200e
47233 F20110114_AACTOF elmidany_a_Page_033.pro
1d64e7476b8334ca706cfd2ec802d9e2
8a07e483b085d1e4b3db4c98892e661787a095ee
1564698 F20110114_AACTNQ elmidany_a.pdf
93b56f62349ac39903ec185c5bab7e69
e34c773d14080e07bc0258fe904fd2e7da8ae3f1
45837 F20110114_AACUQS elmidany_a_Page_006thm.jpg
65c18a3a05ff103cbb3df292d40ee4ad
4a2cdf3cc192f7be650d68c7b098c91d51f8c40c
62517 F20110114_AACTNR elmidany_a_Page_119.QC.jpg
393d3948f806d73ed4d3639042a4b781
cd2a6ea4a513a67695560ca94cfe37c47ae4a20b
21484 F20110114_AACUQT elmidany_a_Page_073thm.jpg
07c7d0d4f38cdc98ef521caaa9b8ed6f
76da25d66912099a456c3aea4dabe0b7b191f5d0
106375 F20110114_AACURI elmidany_a_Page_009.QC.jpg
058fb68918464e58da7ca9cdab86ec4f
9680db15f3bd393f189339e5380ab696d389485c
73344 F20110114_AACTOG elmidany_a_Page_061.QC.jpg
bedb24e25d4a586ead782e47d4e51ab6
bc8b3abde802957778a6f41f456a96e147026f39
182495 F20110114_AACTNS elmidany_a_Page_016.jpg
85c7e32883048768fd213a97eaaa15f2
a82ea96887ff912c602b439d9db363b9f64ec8b3
102185 F20110114_AACUQU elmidany_a_Page_067.QC.jpg
2e240243e582db1456e3129294ad20ec
c428c86aeced174e55a21fc4d88b50a61b0fef82
8041 F20110114_AACURJ elmidany_a_Page_046thm.jpg
49f72e3b9da740d201e1dc9525db0f78
c464c0aea8b74e6dd116da2862f9fafa90ab076a
744482 F20110114_AACTOH elmidany_a_Page_117.jp2
e05c25a40aab2d52b0871fccc73053bf
00a290ddd1ee2e6b01ba832683b759170f6513d4
65265 F20110114_AACTNT elmidany_a_Page_138.pro
b162047d6773cc7ef227ec48fb62de1a
794198cef5a737e373b738d3620a7af5d84c1ebe
81842 F20110114_AACUQV elmidany_a_Page_117.QC.jpg
cbf846e771ca473fee50e7bf7dd216f4
d908e8cc98cc7b31e1531a6c87882e3b90aea472
101008 F20110114_AACURK elmidany_a_Page_058.QC.jpg
0fe3992bcac4aefa717bbc49b1cafeed
fdcd9eb7d0edef6148aff88e28492c3be91b6ebb
83234 F20110114_AACTOI elmidany_a_Page_108.QC.jpg
7940428fdd67d6af4af58c4f61f68502
dcbbee8306b42e6c289a1d461cc76a83e7e6f0a7
1581 F20110114_AACTNU elmidany_a_Page_104.txt
e796395b325d58188bb7c579cb0d4f9c
2c38d36c0b46625504dc21cf9b8eb7e720fe969d
16211 F20110114_AACUQW elmidany_a_Page_143thm.jpg
8ee01bdfe3c03bfa6197eb3316f96f0b
66281059584595edb1d270b44569f490ddfc3059
61255 F20110114_AACUSA elmidany_a_Page_064.QC.jpg
87d9805df71a607cf35c56bd77fcf25c
6e9c148bc416733086ae28151825eada02615bd4
90387 F20110114_AACURL elmidany_a_Page_107.QC.jpg
9c1adc8629cd6d37e17f2e3849f18480
19ccc363c0cfe9b8bfc8edc9c73b0884d8a7669e
F20110114_AACTOJ elmidany_a_Page_062.tif
7a85649bad5ed6179930ba4e042f8372
a69a3858597f3efda9a4bff79351ecf61743a6c6
39513 F20110114_AACTNV elmidany_a_Page_103thm.jpg
32c764c2f9f2c7edb6379f360c7606e1
1c91528c423f14b9c941495c7338a727f79072fb
20819 F20110114_AACUQX elmidany_a_Page_016thm.jpg
9e33f6474402832316f4102ad9971706
1cd3ee42ed8cb152c69fa037f606d1f0c8988d32
43038 F20110114_AACUSB elmidany_a_Page_110thm.jpg
7789bc037e0b29afdcb1593eeff77dc7
d64f17c3dece6ef309d61e0f40fb14ecc2b84cf6
94496 F20110114_AACURM elmidany_a_Page_018.QC.jpg
b78851c04722c22b86128d9f9215eb17
1a47152010364c4dcf561db319f75cde59698eae
F20110114_AACTOK elmidany_a_Page_116.tif
a0de5f310990581f5f5d79e5faf4dbfc
d9c1421e841d1265867202d9ed5b6952be6bab8e
23340 F20110114_AACTNW elmidany_a_Page_061thm.jpg
3b57120cceecbeb2872544a7c41602a2
261f2cfa89eb33f7b460749eb63391a1294ff0c5
44610 F20110114_AACUQY elmidany_a_Page_115thm.jpg
2a24f1bc50fd712d2e548416a91aec68
3b564610ea58159cde619185b34f31e95b992983
81704 F20110114_AACUSC elmidany_a_Page_059.QC.jpg
5116147073573ed20533d3c9adf97b89
d764ca763a09c8fb77ed01fc6f1435fff1bb74c1
71119 F20110114_AACURN elmidany_a_Page_037.QC.jpg
2348ebc60cac64dc6d7a0564c5c2334a
02316dee770b46f57f7ba006f4b1c507c9e07402
38501 F20110114_AACTOL elmidany_a_Page_131.pro
56184709f2727305c0a89b7ffbbdeaeb
42390141c4255a4dc09eb1f1457c9f3398fa467b
8660 F20110114_AACTNX elmidany_a_Page_001.pro
2ed73c0b46cb300dfb9421b037e51f47
5221fff26700c5a56a9747b24c8843ecdb3a3745
3151 F20110114_AACUQZ elmidany_a_Page_002thm.jpg
9b7b967aadd8d0c607685ed36bcf8c2e
2f9728234cfaf4d976028a53ed8aac611ddeef5d
776297 F20110114_AACTPA elmidany_a_Page_085.jp2
ac22af84c02a96d43a3111aca6bbe47d
9a1d0d3e1e95c71e1547c523eed2a1d6381197b7
45890 F20110114_AACUSD elmidany_a_Page_128thm.jpg
b60884590b6ffb6c5a9b1addf013ce25
de85cbdf4edfb97ccb8553bb4e165d3d25c6cd9b
73626 F20110114_AACURO elmidany_a_Page_060.QC.jpg
7335afb5b07967dbbc4d193d89dd68d0
18f62dc4c95bb344bc1e00f9ce1de17305a9d7a6
51545 F20110114_AACTOM elmidany_a_Page_008thm.jpg
c13e5957b008d8b476589432ecdd7cac
c64c5f26075309aae034d121811f8e2ae0dd285f
210374 F20110114_AACTNY elmidany_a_Page_055.jpg
e7cf10e9c9ce33b24d88a5d498b7ca26
27fc5b1a18363c9abb5fe9405427ff00ad1c54c8
1627 F20110114_AACTPB elmidany_a_Page_040.txt
7a4e0befa16410f74fdf7b61d1b5ab16
ece0744b554b1055c32f7a69a531a50b40f1e8d4
15214 F20110114_AACUSE elmidany_a_Page_026thm.jpg
8c6c5ef81a4928fbb9b95ec2229b40dd
65e1a197d3e37577b03f37b8b5ef768780fbb3ba
94738 F20110114_AACURP elmidany_a_Page_006.QC.jpg
e5b36ace5a91c7b2d22140b35df23d3a
a3c83011af3e0ad88f7ff8fc22e5253195698d91
77033 F20110114_AACTON elmidany_a_Page_019.QC.jpg
56d8360bbd4f3fa3845531f7e1b59b07
09e6f487c2ebe0f5e84e8fbe659370ec8ecccb52
199180 F20110114_AACTNZ elmidany_a_Page_022.jpg
392ba3deca1e5b7a15dabc7937e7d2b6
b46f2e5f7c21f8a6d0318184a0fae343af102c33
44239 F20110114_AACTPC elmidany_a_Page_116thm.jpg
3c84dce38b02d895a3bf2d0546b29c2a
8adc93b9d3b4a31063be66775ffbffde450580ca
3269 F20110114_AACUSF elmidany_a_Page_003thm.jpg
8793cb0dad70c16c38a6c5bcdaac5d17
23986ca5ca4b9e454c687dff056e07099a5d3db6
42607 F20110114_AACURQ elmidany_a_Page_079thm.jpg
c0fd70bffa7d4eb669c93e0aa15b0077
91b096ae31ba0cad43f48cd45d4b06ddab16b378
21786 F20110114_AACTOO elmidany_a_Page_049thm.jpg
ab7abeeb0da2e1c75f7445a30ffbddcb
0b32e04b7d4510ec45d1d7c22629703420217a5c
42591 F20110114_AACTPD elmidany_a_Page_101.pro
a1839ec8d3105a10d542f4456a4b969f
27b61f3b97e3e77f5566bdeeb4e3f49c4c999b3e
39962 F20110114_AACUSG elmidany_a_Page_025.QC.jpg
94e67010a157b35d10adc85c1de047bf
1eeb6e8983f7f15bfd37a070fe9c5f73f588ae2e
89064 F20110114_AACURR elmidany_a_Page_043.QC.jpg
ab8eeff4e4343f81dff62ab0ad091c19
5546381099f193d5591356ee9d1f5db1be268e32
281627 F20110114_AACTOP elmidany_a_Page_138.jpg
5400c2f15b6f174da4b4b53be4896c00
b6900d8445fd89e8f73402065f9fee566d193dac
F20110114_AACTPE elmidany_a_Page_064.tif
b7732c9c66830c3db88f0dd5ce6319bf
0362417efec12609364f5ec967eacb32242655d2
23437 F20110114_AACUSH elmidany_a_Page_129thm.jpg
55dbe66773562b7c117804c80b9593f7
6cdb4200a51b84637dbb9abde947333d1130eac4
79656 F20110114_AACURS elmidany_a_Page_023.QC.jpg
f99802fd6c0ee6a43a3f402b2590dd7c
ba8b2bd569e9d0df3fe3f6de452a74a82876d933
F20110114_AACTOQ elmidany_a_Page_029.txt
c1e2f8d3cf0bb7478070f955a54ec39c
db95c5fd3019cdeb2006dcc8b4b95a6ced2151c0
35532 F20110114_AACTPF elmidany_a_Page_107.pro
83e78ff7c497fd56e64af802274b704f
7093c23370657d770b3ef18e7d6a6c11b175e8fd
77646 F20110114_AACUSI elmidany_a_Page_056.QC.jpg
0ba648efc17030cc2dce371073758ac0
37f52af0cd696e33ede0a226ff54e5af3b2ef000
66791 F20110114_AACURT elmidany_a_Page_039.QC.jpg
e450123a80fdc9b2a409b51e69e85e45
72898ddd256285111cd6d9eea76dc74edaab7ead
F20110114_AACTOR elmidany_a_Page_122.txt
2dea00d48741b3d5b0f1413a6680e5b8
2b15eb609e8733958421e99aafb547dec57fb704
143828 F20110114_AACTPG elmidany_a_Page_069.jpg
a479e831f01cbf519a5125dd957fa36d
c44e21d4634b131beadf1fd76fcebb4d276a1ae0
79373 F20110114_AACURU elmidany_a_Page_032.QC.jpg
0b7efdd62da0d42094150eaa574bb907
697223197dd94e6016cd86fbb99122b9a3ae848e
1442 F20110114_AACTOS elmidany_a_Page_096.txt
1b58516141c3ea9541e81b33daf6bd51
9fd34dca3d55378055b38a0746a71f9f2ed99427
65692 F20110114_AACUSJ elmidany_a_Page_029.QC.jpg
cc9dbfec171703bdb8adebaf36fb5314
58965c4fe91575e11fb021923ea348b37ddd27f4
77843 F20110114_AACURV elmidany_a_Page_113.QC.jpg
305d76d17e0d5a7e95afacaf924f86bc
2c5254ea57d30f73013d34b1250c106f330a1a5b
28555 F20110114_AACTOT elmidany_a_Page_126.pro
c0d2acbc21bbd4c4cf5b3a866d416ac4
8d9731f4ac979d65fd2dc41079ac153529e6238b
107758 F20110114_AACTPH elmidany_a_Page_007.pro
2c4787c17efc116879c935075020e807
2bd2c4447e70305deb587ce693db90e3a9974d5b
39905 F20110114_AACUSK elmidany_a_Page_010thm.jpg
2fc1338029ce9e076b3d709bdfa3d3df
944e12548af2872543e6f103ec31d768afbbf77b
28967 F20110114_AACURW elmidany_a_Page_130.QC.jpg
3f4988fde2db93a3a84e4cb9157a2196
8159e085544b1662bb4620aaaf9e38824291f72b
93408 F20110114_AACTOU elmidany_a_Page_071.jp2
e95b35da3d5c6461748c11e27dcc9524
553ccb5a202f3414b082ef006659dddf74659ff6
F20110114_AACTPI elmidany_a_Page_079.tif
582919c4a92d36533eabfa52bc64e560
e1729cad720f99c9a67db2ff7ea9c831d2f0092b
93082 F20110114_AACUTA elmidany_a_Page_074.QC.jpg
b8013c222ae0c038a157f14ab087ea64
1c4921b9af11874e729b15b678dcbdc742303df4
42620 F20110114_AACUSL elmidany_a_Page_097thm.jpg
af7454ee9058dee2354306deaf58335a
ceb238fe99e3280a6eb3e6abe486b21de3c9e50b
46679 F20110114_AACURX elmidany_a_Page_090thm.jpg
fb189d6e81631dd7c71c153330737eb1
c8d35f5130d3e282561f4d3ba1fa4b7605350b58
187654 F20110114_AACTOV elmidany_a_Page_120.jpg
502597bb6abd112b52b68d355304c6e8
e7d5cf17abd2015762788daff85bd750258b4113
516914 F20110114_AACTPJ elmidany_a_Page_068.jp2
1009c6a98c3bafd6dadc0efafe807b17
73848fb8d508c7ca7f97b69e7652c7352749289e
82038 F20110114_AACUTB elmidany_a_Page_082.QC.jpg
dd96325cfe8b66d6d083d3034c69e031
348823e6c3545c318008e7aea1613a7ae0088072
66248 F20110114_AACUSM elmidany_a_Page_073.QC.jpg
ca17e391ce9037cfbdbb98e3c56afd30
3d139c61661404694670d9bbc3193c18da5ce7b0
46405 F20110114_AACURY elmidany_a_Page_109thm.jpg
786745253cecc82acd93bffe20dfa733
8efeadec7bdd3eb090b4203f2cc4157117b59be7
F20110114_AACTOW elmidany_a_Page_065.tif
ef823118a15cf8a7bcacd25a070ebec1
2dd4293ad1145d8a1074924f0fad49e9bc5b2378
52712 F20110114_AACTPK elmidany_a_Page_013thm.jpg
dbd3911166013f74ede94d8da68c74c6
03bbeec8a00f1a17ee13bf42f81241c46dd4fe41
63967 F20110114_AACUTC elmidany_a_Page_099.QC.jpg
847d39367028da1a477e17eb3110e742
99186f1690a2cfa901557f8056069081c57c468c
16659 F20110114_AACUSN elmidany_a_Page_024thm.jpg
3526bc302bd27cfc00fb12eba2d8ba84
2c18d59e2da0cd11d574d871825065e10b99c7e9
81531 F20110114_AACURZ elmidany_a_Page_044.QC.jpg
a95e8856cf7da3672973d2da366bfd14
46859f002f325f742ee0d5349bfcca0fc6869510
92518 F20110114_AACTOX elmidany_a_Page_045.jp2
30d45a6cbb9dd5f9c85a4926514c99a3
d1a2120c578e4fff7f884d7ec00d021cff2798e3
24676 F20110114_AACTQA elmidany_a_Page_056thm.jpg
bd35ed86578631e85dc35dce15ab6582
b14ea28fbda0f40b633576a0e23d9f701621769e
F20110114_AACTPL elmidany_a_Page_136.jp2
5ceb7bcb77e0325cb7e8c9a19ea946c5
5db02683092179e6510815f6afddbf5d1a992e37
6020 F20110114_AACUTD elmidany_a_Page_002.QC.jpg
efb844198e9bb1e9fa19fcfc122f31d7
a58cca353ca1617faad4ba132a41f44c25c2574d
52287 F20110114_AACUSO elmidany_a_Page_028.QC.jpg
df9e591abe3ed506a1cecd56a7b5047e
110bad6e6df7fd00a8274cd7295ec0f3bb467b43
698821 F20110114_AACTQB elmidany_a_Page_096.jp2
7780147771a8dde3982ba69b2f614574
1c3edb4a483cd36d59dc250c042ca40e60bcf703
1120 F20110114_AACTPM elmidany_a_Page_113.txt
9376e372903ea57bcf77c185e2b3e383
958383c565d336f03a70a40fbb9d88c90c5e77e7
46576 F20110114_AACTOY elmidany_a_Page_049.pro
bef93470d98fa3bcb68b121640f99ef1
698bbad69521f45a8cb5108a9d7649ca4825f118
23238 F20110114_AACUTE elmidany_a_Page_111thm.jpg
2eefb62ed9046bb867c367581fb4e9f8
95908ed18b05077db5cff15208a8171640989719
44204 F20110114_AACUSP elmidany_a_Page_091thm.jpg
14038ae9aa8fdb872aec353a931c4c86
f80f5cc8bbe2c0ddf9b67386c4467ad088d9569e
19685 F20110114_AACTQC elmidany_a_Page_106thm.jpg
23623c8e4f644c15291c3410a4c10864
4d117ba5fe7f8b7fdbde68da678d0c3716cb6055
92210 F20110114_AACTPN elmidany_a_Page_052.jp2
7f7535385b189527df8fab52aa0b267c
2dd7657f9664eb4c2b47b419fd650f272b1c0299
45506 F20110114_AACTOZ elmidany_a_Page_042thm.jpg
6a9d4ec423a02a50f7a6f5393850914e
add1d9df78157756667dd1ec4c22eaadd8c5f916
26980 F20110114_AACUTF elmidany_a_Page_005.QC.jpg
dbd06d70a7faae83b9e7a81c4eddbd4d
5f0ae19d7b2f104a76886ae32d8733788c664cbf
74029 F20110114_AACUSQ elmidany_a_Page_081.QC.jpg
7279b62ddfdb550b7f33a1bd67434f6c
ef51be82eb44d1257de078747ac1ebb89fe88579
89709 F20110114_AACTQD elmidany_a_Page_073.jp2
6af3ce09a217d0026131f21f0b1accd6
f27dea8f33fea24131de7a574ba65c032b63bf1a
17757 F20110114_AACTPO elmidany_a_Page_069.pro
e948d7cee8eadf1b01b84bfca73ffcf1
0bda1c14e6c13beb92bc91e369ad3f163ec49d1e
43878 F20110114_AACUTG elmidany_a_Page_087thm.jpg
6ba2e027db1af6c8d955e419d186325c
96bea8cd07a056dba5441e420bd6cc7a37265b15
72361 F20110114_AACUSR elmidany_a_Page_069.QC.jpg
fe3ac3ec2b88bbe3811c7bc42a892312
18b0639430dcd5cd34d9c338753030bf4ebe3fa3
F20110114_AACTQE elmidany_a_Page_011.jp2
73053f1e27de6efce06692197a59204f
7da85dedf34a4ecaafeba0f7814ef65b4d3a9fb9
1802 F20110114_AACTPP elmidany_a_Page_118.txt
6a6f4ba98aa8367b3cff5880dc134ee1
80b32d1854777066605376a39f82de95d7d5cc6e
217757 F20110114_AACUTH UFE0008000_00001.xml FULL
233a4cee98c23edd4215fef459105ebc
9fdfefc05077de07434db29d5893c9b71487392b
82621 F20110114_AACUSS elmidany_a_Page_121.QC.jpg
de9950ce6accfa4bc718b1332984118d
f148dfebd558533c005e765fefe9ae6cb950d762
F20110114_AACTQF elmidany_a_Page_138.tif
e01e0b15688973539cfa1521086a4919
f0d681b57f508f011fe3fde30ff6930d9603df8b
26753 F20110114_AACTPQ elmidany_a_Page_080.pro
fbfce4ca2043f3368e67b45b0fef2539
9f931161d1a89f86708755165469feed995363a1
7094 F20110114_AACUTI elmidany_a_Page_001thm.jpg
15a26610b6b9964d9353f44828e24a68
323778b20e8e6f599e3bead3328dc4639b4f6adf
94482 F20110114_AACUST elmidany_a_Page_048.QC.jpg
fb22883a5b204db0da0c5b4d1d42673d
4ee46b83772beaaa78bb28e57adba639967d8e97
1973 F20110114_AACTQG elmidany_a_Page_041.txt
b247b95eb1fa89d23c69eef335356df3
7dbf86e6db6b5de3d53db473cb61161d6aeb7291
113722 F20110114_AACTPR elmidany_a_Page_112.jpg
034c4eb9b2216abd842dd42dd718e17f
25580080e69fab7343fc50ad8830f133985464dd
20635 F20110114_AACUTJ elmidany_a_Page_001.QC.jpg
41a871cd02eb0762cba79727626972c4
aacbc18c959bb09552cad06e9a880ad09133a343
46548 F20110114_AACUSU elmidany_a_Page_084thm.jpg
85f4c5bc0a7bb1ab16bce51190329aad
d3209c942d7b432e0457ac9f454ea3a78b8f2169
50749 F20110114_AACTQH elmidany_a_Page_136thm.jpg
410c0052e153804110893ea53b52e198
57f2166eb0e55f1aeb27855211e62c3d670081cd
1736 F20110114_AACTPS elmidany_a_Page_109.txt
244a3b223a4b30f2771d983c80f66f02
46d17682bf4e0a74bb75d7407994717c03f8ec23
63701 F20110114_AACUSV elmidany_a_Page_016.QC.jpg
e143d41b8661d8a1540adb17c1dbc75a
f4e7404bcea2c2e3d7bb6a30968b108f6cdd5b8d
40711 F20110114_AACTPT elmidany_a_Page_076thm.jpg
30e00ba524f81533145efb17ff833ea0
48a3663d939c5072c5176e923bd709d50c1f67a2
7056 F20110114_AACUTK elmidany_a_Page_003.QC.jpg
61a645d52cc791ef5b6e3fe97a24c215
3112afaa34808761c1180ff2486da77438b073d0
47517 F20110114_AACUSW elmidany_a_Page_048thm.jpg
769fd155a3d9cc7cb1085de724129738
388d4f477c03a88af006ef17305e746ed2311c4c
204265 F20110114_AACTQI elmidany_a_Page_125.jpg
b04ce050cb133f62c3e2aeb486957c17
92d3ad14340edde5f44513a5d38537bfb7054bfb
182895 F20110114_AACTPU elmidany_a_Page_047.jpg
bed535359c0b560cf6cfcf5fc87c12f6
946b159de38be70b9ca70d048fbb4fdd76b5e313
23001 F20110114_AACUUA elmidany_a_Page_017thm.jpg
236bb3f3e50c7f80ecaed98967c598d1
06276b6b85218af331f64c1c7d9215fea472bbbc
22562 F20110114_AACUTL elmidany_a_Page_004thm.jpg
5121c66b74673632b9926a23cea4a986
ddb0c0862773d4be93ed925cd61d9653fda36c1e
66196 F20110114_AACUSX elmidany_a_Page_038.QC.jpg
fed87cde113ace0c395589e25ba6ffb3
92cd4fc751688622f57dd83b6c5d00c9f5bf7505
68865 F20110114_AACTQJ elmidany_a_Page_111.QC.jpg
ef041063da5cb14d17723f32cbcc9ad6
452d3431a1101883ad332d5260416cd15dba9670
921304 F20110114_AACTPV elmidany_a_Page_127.jp2
1e6e1ce3c5936aaa83b817d8661ba046
92abc2db2f8626de4e489b87abeebfd0183d0b73
76268 F20110114_AACUUB elmidany_a_Page_017.QC.jpg
71f0ef3f44722fb91af3cd3c56be28b7
fe022b03426c6b684d25c5fd6cd7b19930f08d91
72798 F20110114_AACUTM elmidany_a_Page_004.QC.jpg
d4a8ee0308e0b026f45817813eb05b00
c7660585400e5c8c6d0e2d8a09ba07577f3e2a72
90750 F20110114_AACUSY elmidany_a_Page_127.QC.jpg
5a8e7da12594923b8f9dce5268188acf
fecbd3a8850025c5a810c77704a75afecf4db454
49614 F20110114_AACTQK elmidany_a_Page_095thm.jpg
7be69926a0bceb6853a95152b01db383
e6da9743099dc1afe07ffbb952ed1dc71f58ee19
F20110114_AACTPW elmidany_a_Page_120.tif
7e31f4da50d0f711f75e6c5682f5a3f3
0d144c3e0aa144b26180f98d4378ced967faa9d0
49199 F20110114_AACUUC elmidany_a_Page_018thm.jpg
b187646133c1b1a74d8773cb4b93d34f
25cdc86b2988c4092354734e0bf71e1588a83e96
9624 F20110114_AACUTN elmidany_a_Page_005thm.jpg
ae4826a37a3ca20b2450293a48b834f9
9b96a082eb2cc70ae0c5a70b2bf598ed06811f12
43369 F20110114_AACUSZ elmidany_a_Page_032thm.jpg
e5ee66feae6f313b3511ce0c37f64019
d49d71acd955c566e619cec48025dd0de9090133
22126 F20110114_AACTQL elmidany_a_Page_057thm.jpg
ad55c7f8f8bba90be24a1f1a1faedb23
b179ea91a1b28364d6b65acf22f9860c0692b68f
2076 F20110114_AACTPX elmidany_a_Page_121.txt
2f208f0f64f88bc88b10d2d5673eeb58
3d3e5126f4fa76a1dda3e0fc602af4efe53fc4a5
35376 F20110114_AACTRA elmidany_a_Page_028thm.jpg
083e6dbf7f3793ec03b040ec168358a3
8492a20706c2ce41a9e3914fb0a8b57598243bc4
23250 F20110114_AACUUD elmidany_a_Page_019thm.jpg
0db6d7a8677b405001e91684eccc8785
08cfacad5d1ff7283b65b01619bd045d77f82bbd
51513 F20110114_AACUTO elmidany_a_Page_007thm.jpg
91a31c0ee54608fccfa0f380bbf786bc
528ca36184e2b230102b2e3c53c253d25bce2962
F20110114_AACTQM elmidany_a_Page_059.txt
4a53b21acd659dd49e6d15d62b85c7a1
cde02c88b81f2440df8bc58ead4b5b9bc9737434
107154 F20110114_AACTPY elmidany_a_Page_056.jp2
ad3688758eca18f517fdce2a7cdf03c4
a79445435cff12a4a9d7d3cf87ddf9e1ab11a2b8
53906 F20110114_AACTRB elmidany_a_Page_014.QC.jpg
da3af3d174bc1c9ec115acf3fd3fd9dc
37ac81a2d3fa9d9ddf7c323694fc21f5d9bf5988
19261 F20110114_AACUUE elmidany_a_Page_020thm.jpg
86620eb691de8a046fc1fac4a5189496
c48d79243133c3a6a64354721823b5e4d412e3a8
116602 F20110114_AACUTP elmidany_a_Page_007.QC.jpg
c3c7ad9260cba72f9ceffb32066caa76
a6389ace2f0b83abd7a933a1892202682ff49f0b
69829 F20110114_AACTQN elmidany_a_Page_087.QC.jpg
8540c4df8eccb9cd2887212c7c1bb2ca
003707aea7c404fc3f64c84c7a4f39184d0f82d7
115952 F20110114_AACTPZ elmidany_a_Page_136.QC.jpg
e13b5b15adea241227c0855612d5f399
3bc57666fe0addada71b0410da944ef8de58490c
667330 F20110114_AACTRC elmidany_a_Page_097.jp2
10a18567ffd62f2affd31d250c5c8193
3d019b71006bccd7ee56a798a51146136a9445fc
61040 F20110114_AACUUF elmidany_a_Page_020.QC.jpg
fac843cd89937932a8bbd4c4e891a66b
49db5f5fa8ca1a98ef5bf8b9b9be02b87750c568
49629 F20110114_AACUTQ elmidany_a_Page_009thm.jpg
c9740e1ef0ca4d7e5218bb7cb282cfff
8a39866119b2cd95dd30a069a94500bcff8a6213
F20110114_AACTQO elmidany_a_Page_107.tif
58a53236738dafe0860897101cbf0fc7
5b1ce66ae531eb0d571219f57de099cd15f681c8
59563 F20110114_AACTRD elmidany_a_Page_065.jp2
adf5b422a9b1ce1010611ec291611c3e
59953b9da987a2f7014126f1f4daf2f2ecb84def
13926 F20110114_AACUUG elmidany_a_Page_021thm.jpg
4ba1a9ca336cc229e0a33d54f942de94
8f77069f655814a84468381b2c08914b375f9d2b
66766 F20110114_AACUTR elmidany_a_Page_010.QC.jpg
fb9a7adebeca3ba0a232d28307503c4a
cd9d80106117ad41e19a870c2fce5d326ff892af
176870 F20110114_AACTQP elmidany_a_Page_099.jpg
91614e0bcafebbb06661935505788a9a
f14a76c58a95d20ee4b517478713e9a04fd98f2c
82811 F20110114_AACTRE elmidany_a_Page_143.jp2
8effa6e95e9843db501f6a372f81afe4
3b73641e70eabdba3397d49392ce15cd7cd1d560
36794 F20110114_AACUUH elmidany_a_Page_021.QC.jpg
be0358c742b5f33d21b01bc268577c13
16e4e29e2dbd27f304392b581b4f30345641b9fc
48711 F20110114_AACUTS elmidany_a_Page_011thm.jpg
4c2e95c749b2687606ae26a5d0d6d525
0c2c6e7243d23ba7a6ec8a093b2a5cc52d0a6533
8890 F20110114_AACTQQ elmidany_a_Page_003.jp2
3f13361a2bb8d5b3ebd0bf7d32e58421
90da94a4195b0f39b17625dc85ee155b49f053e5
47144 F20110114_AACTRF elmidany_a_Page_102thm.jpg
da46b8b5c08fe48ea9a6ae6136bea98e
00e9ff5f55db4a4be23741e9c100908d047f39c1
23355 F20110114_AACUUI elmidany_a_Page_022thm.jpg
b17facbf0e1bed69e1371970cef2e81e
dea3e0e8ec7ecdec195e522bdaadd9678ba60498
107436 F20110114_AACUTT elmidany_a_Page_011.QC.jpg
35bb7090a274724e915dd8736b374bb8
c2b268f63dfe2c1f50cf5b84fe244f21f20ea50b
1679 F20110114_AACTQR elmidany_a_Page_120.txt
53dd379ddb97afdc3a7cc6b04a26364a
9cd772a8090a1a6b9fa66a906e54d56b5d0a9a89
F20110114_AACTRG elmidany_a_Page_122.tif
5626841695551a5ca4e3e555a6680353
f7d809da66b399eb9aa555e7f5b4a345db632424
72324 F20110114_AACUUJ elmidany_a_Page_022.QC.jpg
fcc263b8b7a050f429944f88be77500c
e18f5acc7820768a9f11f56f2ac51478003ac1f5
54939 F20110114_AACUTU elmidany_a_Page_012thm.jpg
655630e246ead02f42f250b628ecb004
0370360630d7e1143a2ac97c92dc6b2b43cacdd8
94297 F20110114_AACTQS elmidany_a_Page_039.jp2
62770088a3da50365511da4082b72007
605e49272998077ce975982a4779175f4f8f76f2
25761 F20110114_AACTRH elmidany_a_Page_138thm.jpg
cfd7bcebe8fc7b91d862c05a17432258
7ff26ebddd768d41be7bdb705ffcc486e4122bbf
43361 F20110114_AACUUK elmidany_a_Page_023thm.jpg
4e3f9229f89218b414cb7f803d613145
69403373361090342281612b46275bf364f22829
127994 F20110114_AACUTV elmidany_a_Page_012.QC.jpg
3eb090bd4ce85aa83bc0a1059f373318
1a27b091a508a6e024cbabb7d19d10c4628f886d
1605 F20110114_AACTQT elmidany_a_Page_042.txt
24ac4f5e55f50eb218446ab2be2dc37d
b381aa0241d85880034f43d6e52cb4db07bb233d
40677 F20110114_AACTRI elmidany_a_Page_112.QC.jpg
8a3ca1b2a74a494167fe3b34d301c888
40ca8d9a263d255f168777a49c18c9a8f93e969d
117234 F20110114_AACUTW elmidany_a_Page_013.QC.jpg
fc9f5b2b67f6ec83ed395ef8382e866e
2b8f46aedc08a31db9d3fe253b2e510cf185e9ef
86243 F20110114_AACTQU elmidany_a_Page_050.QC.jpg
eff681c4bc352eebd534acef3cee8a1a
a61b8fe9e5513ec1e0c617722a2d06e416e8ede3
19715 F20110114_AACUVA elmidany_a_Page_038thm.jpg
b8e192fc3810e6608a281b9daa40ae39
25d3d98551a75373a96325d7a7de9a2a3de3740c
45724 F20110114_AACUUL elmidany_a_Page_024.QC.jpg
78fe96afd969465527f97f9d8bf94696
0b7a8cd87c6caa81ea2ef6f0ad97b6549212b198
18072 F20110114_AACUTX elmidany_a_Page_014thm.jpg
406ed765f40048ba72a2351c1a3f5928
ba28c9821164b92c18e41560a638a25f2c11006c
49467 F20110114_AACTQV elmidany_a_Page_094.pro
84d2083ef268803b69925bcc3ebecc79
a05022d102b72aea8855d108807eedfeebffa6a6
F20110114_AACTRJ elmidany_a_Page_007.tif
3eeceb91029e413d51d89e1f07b7c926
a132f60836ab73d6901b6bc05263815860a92a16
22176 F20110114_AACUVB elmidany_a_Page_039thm.jpg
0e680931662a5281fed0333b7adcfd22
ff4c5c7eff1e38d5f22db26ce2a33781d2add529
40835 F20110114_AACUUM elmidany_a_Page_026.QC.jpg
fd3a5118d3df7afb7df4d462499b7414
058c4f8d0221ce8388e514dd5b732ea8a9c2f071
18545 F20110114_AACUTY elmidany_a_Page_015thm.jpg
5e8f5b934027a8fc9fd8f3452666fa34
20a3c47ed60db79b9e7263896c7e5d427362f098
1591 F20110114_AACTQW elmidany_a_Page_075.txt
3a88eaeda84b5c71dc988c0601c71c27
bc64ca9a824ae0160604210a358671b7ad7443b1
50317 F20110114_AACTRK elmidany_a_Page_056.pro
bf5d1f7e2054e6556f545aa086922e26
8521f4cae84c81fa1e1afd9535341e87cea6a856
46845 F20110114_AACUVC elmidany_a_Page_040thm.jpg
9c902e5ecb2932c8439f2e31905d3071
945a7ef3bec9e54be7d2dd65f8769958189c6adf
24683 F20110114_AACUUN elmidany_a_Page_027thm.jpg
3ec38e54428c2989342715eb8b6c5c45
e56a4be8d1f8b9ef84bdda5564c0bc86e80b5c61
58911 F20110114_AACUTZ elmidany_a_Page_015.QC.jpg
9a5b7cbb9506e36ae8e8d31e5c6bcadd
bf6237a6624b99326ded5a27ca73d3bb8acd1d9d
807894 F20110114_AACTQX elmidany_a_Page_031.jp2
c771fb9b577bcf64e1cbdf1d130df019
f92b8fb101c52f0364218ed2abeb7f5cf90a52d0
46319 F20110114_AACTSA elmidany_a_Page_050thm.jpg
77ce83f430e80c3e08f680a522058062
f8d94163e075f5ccbca23b70148080cb0ded38ed
45904 F20110114_AACTRL elmidany_a_Page_051.pro
67627233969bf71dfcd395d0a16dbdae
528c111e2ebf6a7c897917efbdf0d76f1a00f87e
90527 F20110114_AACUVD elmidany_a_Page_040.QC.jpg
133bf8ff5ab4a71177db2b74adcdf281
679cb2d004ca41a8fec5b9a8acaf930448498da4
75802 F20110114_AACUUO elmidany_a_Page_027.QC.jpg
88ea615ee078cc9f588477981ce68689
0072155193e0d888fee538d54ab75987cb47c05d
113537 F20110114_AACTQY elmidany_a_Page_140.QC.jpg
b5f33c8ab8cc0f79aed587175efac465
0fcef34105788ec697d924a2666eec2875fcd0e8
44175 F20110114_AACTSB elmidany_a_Page_065.QC.jpg
83ddc372c94e688ccd3cb80b235bae38
3517f629ef3c898b35528315c13258020a2b78fd
F20110114_AACTRM elmidany_a_Page_106.tif
d1fc7ccd49b2e844566e319bdf070d4f
cc9761380d805b5873be004f2e2fbf66b79ad416
46432 F20110114_AACUVE elmidany_a_Page_041thm.jpg
4fe3a9129af36f4286789c950a44fd60
2d28f5d4382c42c5a7a294e7fb3074cbc6311de4
21927 F20110114_AACUUP elmidany_a_Page_029thm.jpg
253a01696eb73c30c839d7730abedb6f
01e69afac02969923e01291b86fb954fcbe6a443
314315 F20110114_AACTQZ elmidany_a_Page_140.jpg
df7dde07f53e7ea367c9a9ae63bb461c
6a3db46709351b2e03ed9de0e42913872750f283
62133 F20110114_AACTSC elmidany_a_Page_070.pro
7ce1fa00734721b8e4f8786e2138e8f4
e8061818974f93a501e04dd97342fa22544f3850
103320 F20110114_AACTRN elmidany_a_Page_137.QC.jpg
151d1b642df043cdf4fc8b163136df0b
2cc4cf7f50a6d967397e54e2b87a8efead194cfc
89156 F20110114_AACUVF elmidany_a_Page_041.QC.jpg
231eae24d3699d9ebd6065f3fc190d45
58f39f9441b1d2353c58ccd798ea8cd28fe83f99
66357 F20110114_AACUUQ elmidany_a_Page_030.QC.jpg
bb2dd1840180e1093a625ef926f0be03
c572d3bec5efc55067dc62ccbaf4156051727312
48383 F20110114_AACTSD elmidany_a_Page_044.pro
53345c917554f492e19d8cab4f0a22f4
9dd39b270f3d4c053ab5b43a2386f483aa42b1c5
70046 F20110114_AACTRO elmidany_a_Page_054.QC.jpg
74a7e98a7eb9a37ccfd4e90626837b25
72f87a42b5f8140cdee6b853c1b9d6459ebfb6b0
85427 F20110114_AACUVG elmidany_a_Page_042.QC.jpg
805043ed9016415a09a63629401520a0
a59b06bc20707369dcc4c369775949c3226c1777
64957 F20110114_AACUUR elmidany_a_Page_031.QC.jpg
b6ef2d5e45bfbb202dbc5ed202e3b8cb
f5d2a3b5a7ec0cbd6bf8835f854e14ff3e6fc897
2146 F20110114_AACTSE elmidany_a_Page_060.txt
65086b6682f599e69fd4001ef39bf9c5
f47d514bcdf700ceedd142b6d1eca1a52b30c534
51736 F20110114_AACTRP elmidany_a_Page_142thm.jpg
06d5d8952e614f875c1f0a716e03666b
6483297ffc5f4de1ca438da9247c5df95dfb056d
46302 F20110114_AACUVH elmidany_a_Page_043thm.jpg
11c1aa7590323979a7589763227864b6
8a9007c7175728c1d82deb502b909d839ff406e2
74034 F20110114_AACUUS elmidany_a_Page_033.QC.jpg
8bafbc56ded4eaf7890a960b28f6cd7a
ceec0ee12a5a6ba2c2cedc534b6115c171148132
21025 F20110114_AACTSF elmidany_a_Page_047thm.jpg
d24f5abb8e560c1ed1ab20d42155d9cb
4adaae0e3ae5a6f8f677083aed2791f42431a061
14310 F20110114_AACTRQ elmidany_a_Page_025thm.jpg
e21454451b6ded1d23d9fe9f30391766
2b0455a039e60573971bc629f11dafeaf7e53438
24862 F20110114_AACUVI elmidany_a_Page_044thm.jpg
24b5a3de13253344c3b33426c61668ae
dd64758c224dc2421fa0867fd5a6bac6ff362c99
15073 F20110114_AACUUT elmidany_a_Page_034thm.jpg
6d3ea51add446fab3613a0b7a18caff0
cf916fe9c2838ecfde5e2e8d8cf8eaec68a74145
42789 F20110114_AACTSG elmidany_a_Page_125.pro
1ed28dcc508ad99fb6fecf386e4c006d
f3cbce0033a252de65ac08cf61dfb5f55bb33dc2
15879 F20110114_AACTRR elmidany_a_Page_005.pro
4c2777e41bee64ed7fdc06d197185bc1
ced8ff2d474bf61bcb52eb4c306e91052c0f9e78
22248 F20110114_AACUVJ elmidany_a_Page_045thm.jpg
18443792fb91312ceb838789cc2895dc
405b1f646855844f9f2e9fc465fc29fc54118cd8
43717 F20110114_AACUUU elmidany_a_Page_034.QC.jpg
2b693d0d05331ab452c845e9bf4ef6c2
c3fa0dd550c600f47fd1d9cf14fcb4849871ee24
F20110114_AACTSH elmidany_a_Page_143.tif
5780fbf97c57099512a885c3bba6d00b
e8fb7155e354d0329dfe9e074cd3e4aa33964f02
1810 F20110114_AACTRS elmidany_a_Page_129.txt
89443acba05617e847f7a9ca9b1d779b
209b27ebc5d60a6dc3e970b61e2be03c78b4a75c
66139 F20110114_AACUVK elmidany_a_Page_045.QC.jpg
9118949f0bb51cc7377514e635a40e82
ed04ecae6c723bd25d7aac7dc1843dc7bc28d54e
39581 F20110114_AACUUV elmidany_a_Page_035thm.jpg
1be7cc1506a09340a502fa8ddddb9443
ccdd1417c93b94af29a6780e3b9e52417b36493c
28585 F20110114_AACTSI elmidany_a_Page_097.pro
d9975da4b1f744f5ff053ccbb1efebf1
7dd8f2709f921a77c766f2093661a0ff4eacd690
89316 F20110114_AACTRT elmidany_a_Page_092.QC.jpg
6d3700a2c15f5ba215b7e6c51307a959
60ec80547b2d4a5f9dda768dba8ba64ec4710b29
25641 F20110114_AACUVL elmidany_a_Page_046.QC.jpg
cdf5d99559c1ae12cb6fe0c34b4195d6
aef427aae1e825c970e5e7e3f0ee46a2ec0e6f0a
63117 F20110114_AACUUW elmidany_a_Page_035.QC.jpg
6fa9b84044126c4aeee348ba6781a16c
89d94af0a07498d594f0b9ce5687e764feeff391
1599 F20110114_AACTSJ elmidany_a_Page_105.txt
49477aa7c917e3cdd66f8c813b717e16
0bb77aca8fb0c564c46b952b297ed8bff7b908a1
23102 F20110114_AACTRU elmidany_a_Page_033thm.jpg
2f5647c1d9374390376781c6a0d22d81
b63f2b8a4e721086458d8716d8fb9c5ec15bfa3d
45296 F20110114_AACUWA elmidany_a_Page_063.QC.jpg
cba0cb278d1a1f2404e29d5103f17e98
3e80c57ec2bf2c65da58d0f7bbee3a0071279849
17843 F20110114_AACUUX elmidany_a_Page_036thm.jpg
42cca6ea9368c720060b36d3f66c9b98
6a96aa48963fc90995d5ea43327695050b814f57
F20110114_AACTRV elmidany_a_Page_127.tif
36cc631cfa28dd3eca452882606d5f36
942332a785ceb76459da0bac965692779760d2fe
20383 F20110114_AACUWB elmidany_a_Page_064thm.jpg
0df8db75092daa9c956dd847172f5468
aaf7d87409b225ad773e6000eddb732d885b2817
69486 F20110114_AACUVM elmidany_a_Page_047.QC.jpg
e0ca74039e7230af615c52751be157a2
6bcbc1d2136cbcb8b96d5ddca3ddeb22b986a879
57810 F20110114_AACUUY elmidany_a_Page_036.QC.jpg
3a186e0b2e11c1695f4a60030975038b
d14468e8ec17ad220350aef7a1cde53b58ae364a
1149 F20110114_AACTSK elmidany_a_Page_010.txt
4ede2013e1f391c2de716ca495cb9e7e
0004b1368072e9e8b6b71f2d41e2ab73071acff3
47603 F20110114_AACTRW elmidany_a_Page_054.pro
0bce8e212f9e0d918948edc72c0d5d6d
3d2c7035fae7f541faa1ca095a30977b1c9ed649
15858 F20110114_AACUWC elmidany_a_Page_065thm.jpg
4a3afa9fb8b4384b20baefb3e8dbf513
2c06ad596a9dc4a7ea04dd31865ff879cacba1b8
47696 F20110114_AACUVN elmidany_a_Page_051thm.jpg
6c9087d9762f4aa235ed7ea591582ff0
54912eae54dc76d95dd7ad9a19cfca10a146e05d
21707 F20110114_AACUUZ elmidany_a_Page_037thm.jpg
99388aaed6077df5c055713700c30786
78fb048fb035a418b11912c2c1d588c8c09b5d41
80150 F20110114_AACTTA elmidany_a_Page_115.QC.jpg
405da111d457d5b3376249640d02266a
2e2575ffb79f1c2b1e9a8f55f473921aa2b84b54
F20110114_AACTSL elmidany_a_Page_125.tif
0e2e22b8f8a7b33860347ed63ea55201
ce53d1125b2753cc9ae8e3e9e688904effdaa5df
85169 F20110114_AACTRX elmidany_a_Page_118.jp2
7d773b17ebbcb33c2ccba93f51e68cc7
657ebd737ef3ac0b3b3901d4713fcbd83f388db4
71146 F20110114_AACUWD elmidany_a_Page_066.QC.jpg
8eb72be1bd71c25839d3209644bc6e6e
776a1b76935a5175365ddce1352e94e3e3b2ea22
91357 F20110114_AACUVO elmidany_a_Page_051.QC.jpg
80980e1b5c6a916103b4f8261462e4cb
1df0e7f564f115591257eb7e5df48b9e6cb0f696
144518 F20110114_AACTTB elmidany_a_Page_076.jpg
c9eba9ed53722d7ba7f93ed82ac75658
fcfca093bb3cbca1c1af5ac6c28fa2c4b7148f66
181672 F20110114_AACTSM elmidany_a_Page_121.jpg
5b2fe4500f425e4ae4d36a7d5ed8a6ef
047ee62560a1d20442e2f38c86729d15174f9478
79736 F20110114_AACTRY elmidany_a_Page_128.QC.jpg
32d3e604324b2bbc09707b9e88c03dfa
1c8f9fc86579c99aa04d86c0ad927cef5718276f
40419 F20110114_AACUWE elmidany_a_Page_068thm.jpg
a951921c8d704c2d03cd5d976dcf5555
ac88f26c9854f8d94a47e4cadc9f9ce8c1b0d1a9
20810 F20110114_AACUVP elmidany_a_Page_052thm.jpg
a356cc984f580679782fd68c88f0c6a7
75c2492582c70ab8547d2b5643a8a8a74571f65f
45728 F20110114_AACTTC elmidany_a_Page_108thm.jpg
d3b6127b9f5d27d8d4ca17f4ff4e5267
d99428931681ed50dc16b989b6cc2c409bede5a2
225216 F20110114_AACTSN elmidany_a_Page_048.jpg
630d0b59665b66431440d4a4ecba0cad
86f132a1c2b67329181b38b505d480424f698271
46789 F20110114_AACTRZ elmidany_a_Page_053.pro
fe4e6f719e91df19652f73d8de319167
1de42c5b7656ae1a962bcc9f70eb209204cc1f97
65183 F20110114_AACUWF elmidany_a_Page_068.QC.jpg
17813a833129fbb7ff29f98baa5b4a4e
3cf8c04ae66ace1b394b0ea8a514fe4b9b66394a
64949 F20110114_AACUVQ elmidany_a_Page_052.QC.jpg
cd97eaaae378e567eb9e65192c9cefcb
fe0bec70c5afce155ffafac83d19e617f32eafa4
37758 F20110114_AACTTD elmidany_a_Page_064.pro
51052c4a3f64b32619adc8e82224f634
21469744bc44f6980d66d2538fefad9f6a339e5d
849194 F20110114_AACTSO elmidany_a_Page_092.jp2
fa1e21f3ce8b5c8caec72eeb6aac913a
204a963978f5a44f9c22dbde50ad5ab12d7ab609
41647 F20110114_AACUWG elmidany_a_Page_069thm.jpg
d0d0d2dd2cf7adfc702f3bc7d853ce02
6c24921025ffc369ff348a73008b83125a05781d
22569 F20110114_AACUVR elmidany_a_Page_055thm.jpg
aa85483772b2a7971069060cbda90806
c68c8adbffbcf6a105d93dfb5f207f32a095fe1f
F20110114_AACTTE elmidany_a_Page_115.tif
9279dd163bb06e61ce0ee765ec087966
1b0b52b13ccf9803ed26a163a9b1c209b8da2d02
43965 F20110114_AACTSP elmidany_a_Page_111.pro
1c579dcbf516868553b84440b4365476
02d51ae3d60013ad717e1babfca88bedd608bfe2
23132 F20110114_AACUWH elmidany_a_Page_070thm.jpg
bee352cf406abb891b4d5eb384f5ed16
f62b7c90a6cc9b57d1ebd8e9a3bffa1edacd1a30
72035 F20110114_AACUVS elmidany_a_Page_055.QC.jpg
16c3fbff7fd508788db4f5e1f4c248ac
8161b7478397640c7689c27b243535e962fd5b5c
F20110114_AACTTF elmidany_a_Page_054.tif
e6b3a740a14c9f9309749ff679db801d
ca087163f8bdc9650a1a721fd7902a3824b14d38
F20110114_AACTSQ elmidany_a_Page_142.tif
da131c55b742764b3d6fd9004b2b4020
9cbed101a88cacb399eafb9f5d700572759dd319
82035 F20110114_AACUWI elmidany_a_Page_070.QC.jpg
8c5776953bdfd2e0b723de09b2ed0f58
79e2729d208393293b34a01c77c7749d9f513c38
70717 F20110114_AACUVT elmidany_a_Page_057.QC.jpg
967b6b6328e41fcbf1956cb5ab994775
1ef045e54e5a4d1aaf4d85d342d4ad57b0a63da5
67766 F20110114_AACTTG elmidany_a_Page_142.pro
4811e3c3ef6992e90160a092ef0ef8a4
48d5ac408d5901cec10235f2dcf094e954c2fef6
150468 F20110114_AACTSR elmidany_a_Page_010.jpg
cf28a17dca6800331eff797742770b72
fff38c772a55fbee7be456001bb52c9099e6e37d
21569 F20110114_AACUWJ elmidany_a_Page_071thm.jpg
891b5f5fb86904a6c68b8c01e6db27af
95d98d0b4c2188152fb731126abf79916a907722
F20110114_AACUVU elmidany_a_Page_058thm.jpg
73a219a4a4dee80bfbe57975b5293770
eab686635b8e8300e499fb5361d89b7f176f6ab6
920972 F20110114_AACTTH elmidany_a_Page_032.jp2
2efd29f6a21a189e4e13d982bdc608fe
e2395d689c894f5bbb911e43186c4511b872ea98
F20110114_AACTSS elmidany_a_Page_044.tif
d84e263de9e813ebcf6530c264fe84e4
ba9f4b0a7e7980483ddf62d079728cd97a938187
67772 F20110114_AACUWK elmidany_a_Page_071.QC.jpg
e8c0c23f1d27fa31b5792f4172c42518
521263dd9ed96726142ebb7e4adeb26b920103d1
44411 F20110114_AACUVV elmidany_a_Page_059thm.jpg
cc998d45425812033cd0b5d0aed4ec90
ce60722330e0e842f1b39f072ce86b334b123277
55384 F20110114_AACTTI elmidany_a_Page_011.pro
54e291d91c39c176ef111187b018c11e
348cbe03e49c9d658f772659e2e8dbdea57b29ea
189197 F20110114_AACTST elmidany_a_Page_134.jpg
984700dacf9138c71c3f87960c4d15e0
a61c8e68bb25ff2ddef4cd9370fb47c5100996cf
23816 F20110114_AACUWL elmidany_a_Page_072thm.jpg
7caff4ac1fa73361582e390b1270fb92
ff09b9003fc812f7e6e61ae647b3156413b0a5f4
23489 F20110114_AACUVW elmidany_a_Page_060thm.jpg
4415db4c888373584737167994aadc43
ac0ee19c4fa313bf44dbc0ec9aad2b0a73365a01
22584 F20110114_AACTTJ elmidany_a_Page_099thm.jpg
fc2d5af888b6be783078afb908ac3d3d
37ff221fc5f39021b676f5e4ced6efcb32110070
F20110114_AACTSU elmidany_a_Page_118.tif
d65b13e36a65ef8998c4ad1d25c6023d
93e86b5f530fc7c1aae9263d017044682eb2bffa
86791 F20110114_AACUXA elmidany_a_Page_084.QC.jpg
1ed008beae797c40b08a4136f9ad6a56
0d15018377f88bf7cfec7cff399eda2726ef68b0
78152 F20110114_AACUWM elmidany_a_Page_072.QC.jpg
d4d38bbfae4d5f5a77a7a1250673f412
6adbe4a6b2ea3afcc9cb77ede575c0ace99add08
22347 F20110114_AACUVX elmidany_a_Page_062thm.jpg
5c1fa19192496bb32d5dc7eda38a1102
b0f2e7b2477e7ee9d8903ea346f5f5ea1043d2de
F20110114_AACTTK elmidany_a_Page_046.tif
615e4568e0540fbd24325524b46a4e3f
79504a71c9c9116dc289490f1d660ac0458e1a32
F20110114_AACTSV elmidany_a_Page_084.tif
6c7619eef1b2a68c3c7c4481457cd741
f72d1a9a7eb94cca212c1a26092974c7e3b1376a
45075 F20110114_AACUXB elmidany_a_Page_085thm.jpg
a5d66205a0ffddc7acbefffc7b21d484
e3310073bd905b21c100015740071e293b1d4528
72458 F20110114_AACUVY elmidany_a_Page_062.QC.jpg
1007c7202bc32711908afcda31c245af
d1aba9b07f517e49ea8491e1f2aec9d25fb5c500
31301 F20110114_AACTSW elmidany_a_Page_065.pro
530809d534749f29da3ccd972ac1034b
3b5e80b94489f74d86077bdf01b44bbe5229772c
85364 F20110114_AACUXC elmidany_a_Page_085.QC.jpg
bedb214ed22258bf8d662a7d5efb1378
52a3ebbf9827283c3f7ff106fd7f4da693093819
46664 F20110114_AACUWN elmidany_a_Page_074thm.jpg
b3ea3636c99f1210539878e90c3c7ee2
cf1d20d978e9d06d4c8d7059b20c1af24c52bc15
14629 F20110114_AACUVZ elmidany_a_Page_063thm.jpg
fc5fc6678b022919f65c7b7be9e07b62
5fc8f4363760b5b4452af7acc4c0304a8f10ebaf
20938 F20110114_AACTTL elmidany_a_Page_119.pro
fbfce834afef9bea2640ffa74245dd11
bcf07b1136a6650d366c0c407e261923d28d1bc4
278772 F20110114_AACTSX elmidany_a_Page_141.jpg
7439c00227bb09b2327d8aadcfa2e016
117cb72740b48609b3275d58152dc74dc37be07f
259566 F20110114_AACTUA elmidany_a_Page_009.jpg
fdeb14ed42146c4e7b37a5452900f42d
eb93916b0f9ebb4af9d52fb0d0b6edf6bfba84f3
47777 F20110114_AACUXD elmidany_a_Page_086thm.jpg
86ff32d72e5d57e1eae111ac45e5fb2e
1e4ad60ef1ae8925cb1b1a2e4339e29288adef79
44318 F20110114_AACUWO elmidany_a_Page_075thm.jpg
7d05913748f6974f652346755eab1eb1
d9df5b25859d530ed1a89b509871c92749e9ca2c
911707 F20110114_AACTTM elmidany_a_Page_051.jp2
921341e1559e14b43df0628b97e76421
6d68f229e05ccc5f4ba274b7a299c10e1bf8f793
1748 F20110114_AACTSY elmidany_a_Page_106.txt
020b39b0a1d66af3e7ea447f2481b51b
3a5c95078cb4aceb2c9faf290adc49618b9f0df0
330484 F20110114_AACTUB elmidany_a_Page_012.jpg
090cf7dc479abd09b10ae58193dd3231
55c6301c1605f05c126e5642d1c7047912f5f21d
92621 F20110114_AACUXE elmidany_a_Page_086.QC.jpg
e7015508fa169b30c17e71d9ba58f9ae
3f4efc2391224da4db74f6a1731c866f7d5d8ea7
82816 F20110114_AACUWP elmidany_a_Page_075.QC.jpg
80a04df68b93c2249bd448c0f7cc5a83
bfd325e2b37108d794fc5371065f595a764e5b1d
F20110114_AACTTN elmidany_a_Page_136.tif
29d2ea20d456729ea4765381e0e36dd6
68fc5556e77b219a65c0e5b0dd87f5c804750052
797671 F20110114_AACTSZ elmidany_a_Page_075.jp2
43967a7743745f8a471dcc067f39ca8f
d0b6eba91fc4d504dc79515301c20d7a5bb4b69a
290206 F20110114_AACTUC elmidany_a_Page_013.jpg
a3fac03e983fe96777a08c246fc60306
95d17c5befa2d97e381fc9e2b7617d36ca5bcba7
46437 F20110114_AACUXF elmidany_a_Page_088thm.jpg
8ac5a02319995fb7ee608f34c9497057
4f71bc29ba4a1d0e507a0cc1b3481d064f2ffe3a
70981 F20110114_AACUWQ elmidany_a_Page_076.QC.jpg
2f54ebea6efb23d90d6e46358b734979
b25c21d6c976997331576695eaac4fb434f14698
70753 F20110114_AACTTO elmidany_a_Page_134.QC.jpg
fdcbce286ea0685d96954f2ed3df7888
e8cc1080136776544af877460f44e73a971ea5c0
165196 F20110114_AACTUD elmidany_a_Page_014.jpg
a7e48794271251eb283a3b55cef9ad31
a26ea007b3e345f05afe5c3b87fb87886368c329
90248 F20110114_AACUXG elmidany_a_Page_088.QC.jpg
b6f51d1a3725ada6364aac5d3608302f
7b702b081091fb92dd09e89e8ac4b35af3e5d29e
45407 F20110114_AACUWR elmidany_a_Page_077thm.jpg
ade4f50e1d15190350f8da67f87affea
03a1c7d12f3d3261219936757b29cfa086cc7869
98217 F20110114_AACTTP elmidany_a_Page_132.jp2
74424de49529ba2cd8b3c227920d025c
13fafe0996250f8ab93f05ecb06aa213ce2dd474
156227 F20110114_AACTUE elmidany_a_Page_015.jpg
26c208777a0a5c958af347b01503b278
d2f88f6906a4b3dcac72d96b59c7f5ba3468acce
46464 F20110114_AACUXH elmidany_a_Page_089thm.jpg
72d1f321285b4359d676ec00d6de5867
64f211dd43da22222cae69f657999c8c7cf51990
85838 F20110114_AACUWS elmidany_a_Page_077.QC.jpg
98073ea15ce2b922ea5519a6948aa6e0
161290345386e8f25cda75172167a7c773690711
167800 F20110114_AACTTQ UFE0008000_00001.mets
333fd927fe6d0f85c37a69b0917cb403
6261a8f7e927724eea912d615cbbaf9e201bb8f4
205729 F20110114_AACTUF elmidany_a_Page_017.jpg
02b69fdcb0657cab4260ea7c3d0341b0
f20622ce06187c78d0065a3a3621db5a8d5e141f
92975 F20110114_AACUXI elmidany_a_Page_090.QC.jpg
f7b861160aea48050f6a68dd14022db1
f698a81a30292522020d1ebecb8c751ac0715fa4
48883 F20110114_AACUWT elmidany_a_Page_078thm.jpg
d145ad6e73facf12334892d4b148fc57
8b839f4b742e67fea014f4e21043e4feafd61aaf
93076 F20110114_AACUAA elmidany_a_Page_049.jp2
bc509aeabbd5b150d51c8905ea2b7cb8
d819cf645d3462021ea4d1ceae97f292a4f47492
213529 F20110114_AACTUG elmidany_a_Page_018.jpg
bcc588bdfb73bf562ce3242422c54b3b
13ae9b0d087ee57b73eb13817e5639ab5d7e2530
82123 F20110114_AACUXJ elmidany_a_Page_091.QC.jpg
88076aff9472b0fb3dbe3a6bda34c8af
ce9a2fe14c3e86c027dbc2183ebd9db870d6ad20
102018 F20110114_AACUWU elmidany_a_Page_078.QC.jpg
1ed0f8ad4da80892ffd7a7301dd2187f
6c114e8da197764c884bb4ddec6edce64fd05bca
899997 F20110114_AACUAB elmidany_a_Page_050.jp2
335880bca0d65188d025b6c02a4e097a
b81e240fac8aaf1b1d97101c3efa40cc9467b068
216027 F20110114_AACTUH elmidany_a_Page_019.jpg
4b8e1e9af8174a0aef49f6d8e5f77b2d
205a1c4a7e09d53b083b6a95e62cb7ce5f8e2aba
45204 F20110114_AACUXK elmidany_a_Page_092thm.jpg
39429538c948dc14a074b88639d5b497
e7d77c679fb66ef8d5c2f6521233b7ad3151f575
74874 F20110114_AACUWV elmidany_a_Page_079.QC.jpg
14fb297e142d8666df775d64e329582c
98b091080976e889ae2f19c716c9ab153c664673
59378 F20110114_AACTTT elmidany_a_Page_001.jpg
7276cf49a898881ae344d67ac174558f
54625c0f8029c2710782e9b3bb5f01e7ecfa900d
974652 F20110114_AACUAC elmidany_a_Page_053.jp2
5e7206a401dda9ccad4d1418cdf23cfa
ef297e4476acf259cfb8edfd00eca3fc4be229ec
159314 F20110114_AACTUI elmidany_a_Page_020.jpg
b0a46d0ab77309aa50ce66dd0f0dcb6d
594299fc6af874d5b99f5388857d3066674f8e21
47584 F20110114_AACUXL elmidany_a_Page_093thm.jpg
1c221f2a55ac2d89352568b6ff918dbe
a5fb6a720864e4bea86b9af170137f6453283d64
42465 F20110114_AACUWW elmidany_a_Page_080thm.jpg
490e3ea1eb83dce498d91d1ce1023b3c
86a56fed39e3b96b5fc21b419ee8a5f8091db292
21757 F20110114_AACTTU elmidany_a_Page_003.jpg
2ad0a19e5ac871babe944a0976db619f
0daa2976ee5d72e1c46a4207beaa60e67cac4254
96538 F20110114_AACUAD elmidany_a_Page_054.jp2
7fe9d3d990e535f4b5173414f407b74d
70c7d24dd99187150f61c4726e0ef6f460516900
95394 F20110114_AACTUJ elmidany_a_Page_021.jpg
5578197e2637b7c40cc7b814e2b9223c
49bb81351b208d894da82e32049f6aa496872929
43281 F20110114_AACUYA elmidany_a_Page_104thm.jpg
aa446d50c3e5c1bab1631624dfc59f8d
c876f78db1e9eeb0af7f1f4e4c573fc3a64e00ea
92111 F20110114_AACUXM elmidany_a_Page_093.QC.jpg
3e0d6713c4d78fc3452cf4eac120dcc3
b51e72da60fc1d77babf93717fd4ec73eefae977
42873 F20110114_AACUWX elmidany_a_Page_081thm.jpg
2fb9c4d0ae0778a5188b79d9dfc32c57
0f6214fc0172c3dea604f4125dc63a8f99295e8a
108149 F20110114_AACUAE elmidany_a_Page_055.jp2
08a33dbb0eb6a07be9883f7e2fe44976
9227f7a53d757085e29237163852d5cf4776da6b
156945 F20110114_AACTUK elmidany_a_Page_023.jpg
e160553048767b7184b89c39367d7acb
be85c05642495a7ecfd7f9ba71e8ac2dbd19672b
203398 F20110114_AACTTV elmidany_a_Page_004.jpg
c3b0e207b547242e50af135e5ccec846
25de1a4eadc5846e6d64f540735959f510136578
82081 F20110114_AACUYB elmidany_a_Page_104.QC.jpg
ea73b1e743efba4929b23107b3896694
1ca4acdf5c6eeed84eae59f683a242dfb93d62dc
23438 F20110114_AACUXN elmidany_a_Page_094thm.jpg
631261634a72e63d6539ccf78f8838fb
6acb945a07d5b5d565c521d7b90eef7d0ed46b0b
44343 F20110114_AACUWY elmidany_a_Page_082thm.jpg
051193816d2b2a54711fea65d0a9d64f
aed7747503a636458757b2e7ba1fd44cdacdff82
99094 F20110114_AACUAF elmidany_a_Page_057.jp2
e1f6591ca189698f31f35429e2e71b19
3feec634732fab21bd556f0aec91e154bcf71526
121309 F20110114_AACTUL elmidany_a_Page_024.jpg
29dc1b8397c02bf1793625ba983feaed
5ca06178d1b56c543b99afd6ffffef51adb878c5
72333 F20110114_AACTTW elmidany_a_Page_005.jpg
5f86c9864d8a2ae6a7648e751546a86e
b3bf086060e3ddf78209685e829d733ee35c8e2d
47500 F20110114_AACUYC elmidany_a_Page_105thm.jpg
ff55063abb5a190fe29b25703bd305a1
90b3799a94dc8803f2673a9fcd81eb3684710a7d
40172 F20110114_AACUWZ elmidany_a_Page_083thm.jpg
d9a7d79f9e2c31de76bb0193bdcc6bd7
fd2b720f53c3727d48842a99663b6fbaaaf0168b
F20110114_AACUAG elmidany_a_Page_058.jp2
5f68d0afed4f17c0f9f6701fe682d4a7
10aaf63fd767a37b78611e0d45d47a385c5b3cc2
189611 F20110114_AACTVA elmidany_a_Page_039.jpg
230e31a5693ca52f551ed04f3a769f9c
9500f4b8ea661abd300196f21f1e5c97ae2f60cf
266964 F20110114_AACTTX elmidany_a_Page_006.jpg
2dd7fd0731f3874194d04cd6a348a91e
4d69ee5a5b388947dbcbf5df63c97bf5f8a948c3
95424 F20110114_AACUYD elmidany_a_Page_105.QC.jpg
b4398a9bd4da547d8a72245531202f0a
8102de15d429ada911505d80aa3a8ceac87d36c3
77354 F20110114_AACUXO elmidany_a_Page_094.QC.jpg
04b2a1dc59dd95aff22aac4222bae126
ec3d67afff3ed25f6f23264cc53ad5b70fc1fb08
796609 F20110114_AACUAH elmidany_a_Page_059.jp2
34c37a3db69ece7e01c0ab394a36e9c8
2710101027fc6e77f3c946638b3c81adde7da356
195317 F20110114_AACTVB elmidany_a_Page_040.jpg
6ab194e25fcd6073f7f282c2d6fc38e6
8dd1e5de012baf8ac38671e7a099add709feaa52
102849 F20110114_AACTUM elmidany_a_Page_025.jpg
537b526738c751870ba64d6eaee6fc2d
17f6ab8ce8824fc969c90d0fd9dae0cafde19515
355813 F20110114_AACTTY elmidany_a_Page_007.jpg
9d45773e365b3ea58e755aee50c05852
18d5e39a6e3a67e90c092b72878e9eb1d11925b9
60400 F20110114_AACUYE elmidany_a_Page_106.QC.jpg
25d47008924930a2939972d1dd3b103c
c8375cf0c90ac302f9a173e39d4aebc16de34aae
104275 F20110114_AACUXP elmidany_a_Page_095.QC.jpg
e2d6ff8d0058f7db356bc4f48ca72d5c
60f10a0b575dd4dd3e463003794cf8773ee160f5
105185 F20110114_AACUAI elmidany_a_Page_060.jp2
455a25ea0efe8319cc9ef4c5550a394a
b9790bdb01e4d6ff65b5145d651a75ece7f205a2
193582 F20110114_AACTVC elmidany_a_Page_041.jpg
500583ccbb0dc4eff28f3b99af5c2cd4
feacc8a690076c2186e6b5b529179992b32d6145
113720 F20110114_AACTUN elmidany_a_Page_026.jpg
3a2a32daca18e1ea39eb0f88ec30eeab
f3146b860ba779417087bf0a75f1fa7ea77c36f4
355204 F20110114_AACTTZ elmidany_a_Page_008.jpg
89c8c458cfefae4a537293f853862946
0996d3977f79c2a634afea88b9e8ce9e56eaed7e
46831 F20110114_AACUYF elmidany_a_Page_107thm.jpg
88b5186132d6199a098ecf4ce7ee95ad
9e231e6ea693def2b1d732148e87b1c3341ed18e
42190 F20110114_AACUXQ elmidany_a_Page_096thm.jpg
0925db384521b92034df40503fd79cd8
bfa9a1e289d1fd7450b22d9c94da8c7c9fab92da
101153 F20110114_AACUAJ elmidany_a_Page_061.jp2
86bc5f76df49fd2de6379c6973e789ce
8620562f193e981918376302236e15f5da6bdf13
181748 F20110114_AACTVD elmidany_a_Page_042.jpg
99ad7587ccc00923be9faff81cf5312c
9057882c4826aa79e52993e654caa72c09332a5d
203953 F20110114_AACTUO elmidany_a_Page_027.jpg
affa371a4748e4cec40235300f86988a
bb9d36beaada12cd76409dc40e5d699caae59a2c
92763 F20110114_AACUYG elmidany_a_Page_109.QC.jpg
78c68fc9f3f319f11e8e8caa71951e11
b8a2c82e03751ab01eb58740d9c72b38f2114267
78942 F20110114_AACUXR elmidany_a_Page_096.QC.jpg
fa474be1f5f6fcd4ed0e53f9b9eacad2
a94716e1ca046cd0ae15edf1a0e6aa7c4289fcb9
99929 F20110114_AACUAK elmidany_a_Page_062.jp2
d7d40d2110047700d1683566415b9e2f
607b221fc3658cb96407aeb151cc69857529632d
195135 F20110114_AACTVE elmidany_a_Page_043.jpg
33fd284a40e97639c22c41672bd97378
4c8e8a22d93ba14a5c92b7bfa007670edf1076b2
96495 F20110114_AACTUP elmidany_a_Page_028.jpg
f4ab68d9a523c7ed7a06dc7c51f18dc0
6410749e4e501d41ea574ba354005dd1eb75868f
78333 F20110114_AACUYH elmidany_a_Page_110.QC.jpg
61b83c3620ef52cd7fcc1c57bbc6b154
791f06f0875e3eb8f8659f464852d7479b111096
74838 F20110114_AACUXS elmidany_a_Page_097.QC.jpg
f2515d2031d1e587ff7e383cae285af4
6b539e2e806e3964e9916b50a5f09e678857cfc2
62810 F20110114_AACUAL elmidany_a_Page_063.jp2
72f9fbbd5275b3aaa1a5d21aba94f547
627962f53f1cdcedeea0ac343cf444751f6b11f3
211841 F20110114_AACTVF elmidany_a_Page_044.jpg
1c2ae33c0b07eb8714a96402d24c8c81
5aadfc3ea12629a67696ceaabf9cd43eb175a168
173715 F20110114_AACTUQ elmidany_a_Page_029.jpg
cc96b6fcb00487d391d1946757d2cb6a
77e47904abb6fe63037503c84949fb0f2d75684f
14446 F20110114_AACUYI elmidany_a_Page_112thm.jpg
4d44ec8997764eb6c07e27f53714598d
dc5467694881fde5803eb9dc7f6e4be03728736e
49207 F20110114_AACUXT elmidany_a_Page_098thm.jpg
b25fe25e2cfc015e94913cd55af1de41
d06c0999f84afff390f7e386444d100ecf866576
83455 F20110114_AACUAM elmidany_a_Page_064.jp2
141d88d813b014645c7e42e4318c9ddc
e334fa31c98b4201603925766b4f3d39533bb81a
179203 F20110114_AACTVG elmidany_a_Page_045.jpg
598c2aefd06cdcf870ebd732538eacf9
5c1f2172da0c66c3dd21a996daf90923c2994c4f
184034 F20110114_AACTUR elmidany_a_Page_030.jpg
05dd68deb5c6aca93c9da7f3319e21ba
ecf51a8e557ebfa9d8e4aeca65a5ae752d0786e7
545086 F20110114_AACUBA elmidany_a_Page_083.jp2
add13c346d12e4aba918fa041f73fbd4
90d5afacd2a59b4a780d547deb09eaff399ecfeb
45918 F20110114_AACUYJ elmidany_a_Page_113thm.jpg
a8b45e87d52c7faf767b279463c7c97d
249edabbea24e495093b7ab9d36a7891739aa2c0
98482 F20110114_AACUXU elmidany_a_Page_098.QC.jpg
abbd30dc57045efc56f7e2f6ebc17ca0
6098d54122a6d0589bd5fb5197c2d75290047277
104842 F20110114_AACUAN elmidany_a_Page_066.jp2
889df490deb97f4aca009aa5e3d24329
64f86be02a9a4a2a13489abb95b8f26b8a663d9f
62964 F20110114_AACTVH elmidany_a_Page_046.jpg
398de509fbb74a633f94c923263762df
176bd3fef0e46ac3b98e1e3db4015c1b30814b81
161415 F20110114_AACTUS elmidany_a_Page_031.jpg
513992f658323de4314b3d205e411dc6
bd03233e12e10afb6e40e0ed3b8ac63a9e8630e0
881032 F20110114_AACUBB elmidany_a_Page_084.jp2
19851c670aa17e13330517d598db1eb8
0e48b4ca755a14cba90c132b2c65542c444d432f
18284 F20110114_AACUYK elmidany_a_Page_114thm.jpg
9fde7391875bf1c50d9f441bc378ea52
0be60b9e99943679581270ab03912d65f5250c23
86280 F20110114_AACUXV elmidany_a_Page_100.QC.jpg
9a0659b9f49f65a7707a659f55255e14
8815a31e91594f8c6f0b0f514c263243ca95d3c0
1051926 F20110114_AACUAO elmidany_a_Page_067.jp2
4885bf80c8a2e4f341b49fb415758eef
4f9f1de5168b9dc4d8c7bd7af02c7d75086b3fe5
181634 F20110114_AACTVI elmidany_a_Page_049.jpg
208ac9178cf53ac2855e58226aae2ed5
366d82614f5b71f7071f68a70c237c3251625fa8
177137 F20110114_AACTUT elmidany_a_Page_032.jpg
ca5b8d6a99a216f78f2acac5665c211d
6a6f66c5fd318d29fe06685190aa1e6ad83ce462
898324 F20110114_AACUBC elmidany_a_Page_086.jp2
d83a2975c1430d13b8bc439a57ca7487
0705ab543b22f127ecb76f2b6c4cc8a37767ce3f
57594 F20110114_AACUYL elmidany_a_Page_114.QC.jpg
03c77e602b727dd6a6fa9f41db532277
09802e1d183048c9ecca3a208f362814bf3dff33
22792 F20110114_AACUXW elmidany_a_Page_101thm.jpg
fe98c6d376f79152404d70294aa4142d
6a334f2dc4339eb43c0bb621f839c48f4da5cfcb
633387 F20110114_AACUAP elmidany_a_Page_069.jp2
b022fb4c2dc01e48fbc95cd913931e6b
c092f224bf67d3b79e9a4dbb7c3c37caf6575d02
196363 F20110114_AACTVJ elmidany_a_Page_050.jpg
091f6679ae13b5e7d4cc62a73fbb1e51
c136624f9dd3083d6efca88dc1e1b6280daa8901
203729 F20110114_AACTUU elmidany_a_Page_033.jpg
6fe574b1fcf14e783f07e5972c9e62f3
8ba9385126cb4ea3d76ecc33edd08a0aec53afff
651732 F20110114_AACUBD elmidany_a_Page_087.jp2
538d34d9ae8c7f99831d787872a1df7c
7f1bd2d20333584ab5809cd6621373d17cb83efa
91719 F20110114_AACUZA elmidany_a_Page_125.QC.jpg
1dc0c54814e1e4928fc8f87c1a48f30c
a3e36866479aab9747469ba04b16b5de2e92f8f8
82354 F20110114_AACUYM elmidany_a_Page_116.QC.jpg
0f3706d4e182513653f3f418f47945f8
a9aaafeb389983ee4cba1280142aad15d09d75b3
64805 F20110114_AACUXX elmidany_a_Page_101.QC.jpg
1463e4de39ee1334fc4f81d90c5f502c
0d3fa86e284b0976051351e81d068c195d1bc64b
126683 F20110114_AACUAQ elmidany_a_Page_070.jp2
ab49a5852d6f689de6a4317564586de2
e48a81499ad09359e4392e6f870e3893fea64d13
201221 F20110114_AACTVK elmidany_a_Page_051.jpg
31314b820bf3210d59e49e2f20814e5d
b064f84853f6c1e03c887a9bdd33bdff532925cb
122068 F20110114_AACTUV elmidany_a_Page_034.jpg
3b75a3566bda48b93cce13785a536c71
f1bf6d56bdd1639f5ac91c54802d2a0d2fd6e9b7
897988 F20110114_AACUBE elmidany_a_Page_088.jp2
c89f00a34141ba372a54b7dedff19ec0
28b01b1f3444b60eeaf419adddbfeb6feb345aae
43844 F20110114_AACUZB elmidany_a_Page_126thm.jpg
5ed1b3e30572cf2689550e21128cf7be
a39f0590a7a8fc62aa1b9f026b8737eed507b252
44767 F20110114_AACUYN elmidany_a_Page_117thm.jpg
8c066a6428125eb5c43b091c3a7cd72b
c2c1b584b80810176b73c2dd43c7aa4069fa68ce
94429 F20110114_AACUXY elmidany_a_Page_102.QC.jpg
b39a8cf9513e9e6bb285cb93685a45ad
b7baa5ec29c4ada02fa38c08d4a4bbc842a0be68
107254 F20110114_AACUAR elmidany_a_Page_072.jp2
306b5b900bfd4d730691bdbd96989e7d
73d3c23f258124f48bac6e9d955deeb1904fd7af
171599 F20110114_AACTVL elmidany_a_Page_052.jpg
e52bf57296b82cd503b1a8adc02a7de3
f507a41611757c2f0d183f6d5d0666baf49e78b3
126067 F20110114_AACTUW elmidany_a_Page_035.jpg
3861db7589c39d4d5f19e9953887390d
ef7b47483ace5e4b0bfb0f3a5efe9c0a54a070ea
886133 F20110114_AACUBF elmidany_a_Page_089.jp2
8bc9a2741011b860893896c4615fd552
7de49ca27107855ad28fd9179de58a6faa0124fb
81390 F20110114_AACUZC elmidany_a_Page_126.QC.jpg
ee19aa06e7049e56550e6275196170fa
2ff19cf752ccf67b8178cef5f667a26a296f9f1b
19397 F20110114_AACUYO elmidany_a_Page_118thm.jpg
cf5711b4364e437b82527f9c4d7f51db
3186ddc6d9d91c5c9e27711782f1029921ada72c
68880 F20110114_AACUXZ elmidany_a_Page_103.QC.jpg
c6138c7b9ffdce169e95e861c0ce6ce0
9735c232c4f45100562bb30352911cad8c661974
951920 F20110114_AACUAS elmidany_a_Page_074.jp2
26277ef98bb4c402b131e6cefa59aa2d
85415673734056aaa3d60dc2bfa4be9a0aa13fe6
212319 F20110114_AACTVM elmidany_a_Page_053.jpg
e209cdfbd8840549fcc3f3909311fd29
3c0ddb60963ce922e6b94dda9da936b0a0248c95
153290 F20110114_AACTUX elmidany_a_Page_036.jpg
b343e4e0f6a94233be9823c207ef7fda
a91836c04ca111dd192dcf3d64ae105481df940f
938625 F20110114_AACUBG elmidany_a_Page_090.jp2
536a8dc3925b8aafd2df9ec985c32c76
841cb906c50415738328278ee3e2fec68a471e98
121636 F20110114_AACTWA elmidany_a_Page_068.jpg
eb31741e40e3bc77ad245ba915a9f3b2
4e1adc565bab1a6da29e374432dad4accdaf6764
72784 F20110114_AACUZD elmidany_a_Page_129.QC.jpg
152fc83a90d129737fb01f3c98ff1f05
994250cbadc1bc51549be9c034de155267149f44
185299 F20110114_AACTUY elmidany_a_Page_037.jpg
b3f15480945e92edf07fc443d5323029
7ab837d91ee80d48f7a378f4ba29404074701446
903157 F20110114_AACUBH elmidany_a_Page_093.jp2
82aacb7fd3cfb96d137fbe17c5ddcfb5
2cdb7f7123417bac4d5c6782d35abceb69ecd366
245623 F20110114_AACTWB elmidany_a_Page_070.jpg
c44cfcd935efd5f9122accf554b92be7
c0b3fb1b2ef836555577f79a194bb70595e881fb
608333 F20110114_AACUAT elmidany_a_Page_076.jp2
098607584c50ac3dcf94dca6a8eca1eb
4b5f40d01b612abbe6c61e681ad275f3e9511f41
10261 F20110114_AACUZE elmidany_a_Page_130thm.jpg
ae409c5b8a5b3c4418dad76780252087
c0d212d8bd47598839f3eb73d29aa8a75d827076
63050 F20110114_AACUYP elmidany_a_Page_118.QC.jpg
369c42fbb1c6ffaf749aa8c55a472beb
d25240ae67b039d7c9e559339541888f22015179
188493 F20110114_AACTVN elmidany_a_Page_054.jpg
3adb6255f02f86f78b2b2ad18b35da20
1664945b2eddf753cb3ae611f190daceb225d5ee
168077 F20110114_AACTUZ elmidany_a_Page_038.jpg
9eaa63da8993797f0af19574e8bf12df
528fd94ce1a3fb9c7577158206df313c22d9373c
106620 F20110114_AACUBI elmidany_a_Page_094.jp2
3ace1e381d1e5418c52c94ed013e5f51
b177e1c5297e47db53acb2a42d248c8a2e0bd80a
184218 F20110114_AACTWC elmidany_a_Page_071.jpg
65e5b1938f123a7c372cb241a5e07d5b
c209a86b93c11c04e12d9fd498d9f8cd25aa8b4e
841040 F20110114_AACUAU elmidany_a_Page_077.jp2
06dc1dd50da5fbd1f2335f96f66281aa
51f63f177db0094380075f886684083494188f3a
45451 F20110114_AACUZF elmidany_a_Page_131thm.jpg
decb1d70459e20ba9f7e80f316368f74
9e234a744e7a71fdd1600696439a1820d14da5b0
39295 F20110114_AACUYQ elmidany_a_Page_119thm.jpg
8d7937299d6bc4bde59ea103fadfeecc
77ab5ab3e4caae693bbfb6c73f87503ebdfd8a35
210267 F20110114_AACTVO elmidany_a_Page_056.jpg
ed2235a913b914dc802170456c94c550
0c9cd136787bc1f8bd729832ff78314ff62747b0
1051934 F20110114_AACUBJ elmidany_a_Page_095.jp2
89df6d6239e2385699b8f190cc4b3024
275b0d93ade47689021fad869ad4a98d547ac5d5
214215 F20110114_AACTWD elmidany_a_Page_072.jpg
c507463701000adb8f7d28a3bd0a651e
038d87da83de9c5f8cec155ee09d3812280405f4
1051947 F20110114_AACUAV elmidany_a_Page_078.jp2
052cdab2fe1d497a0c27afe0e91433e3
06cfe1a747757edf1f08825377bc324f6b48db55
87381 F20110114_AACUZG elmidany_a_Page_131.QC.jpg
ad309568da8f489ddcbd734b0ab70f68
e6dcd41451974624f86e6d7794a7672f2d4ec7d4
44642 F20110114_AACUYR elmidany_a_Page_120thm.jpg
7e1e4308d11d4992565b77b413508abd
77f34539ebf58ca01735fca2b40ce6f93aa0d719
191911 F20110114_AACTVP elmidany_a_Page_057.jpg
79ddece7e29d3f428a6f74c3ed6b68b0
7d6f52c60d22c8ce0bdd97901dbdea71d58542ef
1051983 F20110114_AACUBK elmidany_a_Page_098.jp2
3a6e76e827984d840ffb556fcc9f6614
68228f9265b76cbbd4c35081cb52016e114f1e90
173560 F20110114_AACTWE elmidany_a_Page_073.jpg
a1d0acd901584700c1ab47b53bee7004
85205217188eae1e56ded637e64619eae6d248e2
708150 F20110114_AACUAW elmidany_a_Page_079.jp2
8f7cd772239ec2f4e0ae0cd94e1b57fd
8c50900c97fdba4ed65f017163eec11c8c47b176
22095 F20110114_AACUZH elmidany_a_Page_132thm.jpg
67e5cbafe16b799dc27c8e43ef7739a2
029fbdc260255adda052bb67f2c9a23eb4d2dfc8
86349 F20110114_AACUYS elmidany_a_Page_120.QC.jpg
e5a86250f1d42945890ff97630a9cab9
3429c986fd758e6abf6475bfa2006e22850e5603
231906 F20110114_AACTVQ elmidany_a_Page_058.jpg
b96ebaebd9a035228ea4921fd9297311
8d7121666795b24b8b078b485044f69ab71c66f6
89935 F20110114_AACUBL elmidany_a_Page_099.jp2
f02c706edd7e61f93b35f35affe3371a
db89b73845011af6018d9cb0d56b44b65433d80b
208339 F20110114_AACTWF elmidany_a_Page_074.jpg
7e8a99b7109855041aa5b1b2b209dcfb
f57fefe2a385b78d0096ea864f95a0d60d69a7e2
69706 F20110114_AACUZI elmidany_a_Page_132.QC.jpg
9ae3c04dd27ee5572ce6ac30ae0b1af7
7849deae17a649d238ef7abe24e187f7908d53ef
44000 F20110114_AACUYT elmidany_a_Page_121thm.jpg
a5638f59c092dd12e31260215be47f4e
59f1809572106df661cc00057f3fde1da804ed6a
180561 F20110114_AACTVR elmidany_a_Page_059.jpg
a8d862cc6caf77e0fc16662764e3345b
ac63b4428c2c9af733098cff88fd77529bf364f2
77869 F20110114_AACUCA elmidany_a_Page_114.jp2
4891c267042052484c6570527739262d
5479a822026740c78acb5cdc8808852bfd10844b
818326 F20110114_AACUBM elmidany_a_Page_100.jp2
3c02e861797261833a22fc71ea2c3bf1
9b7e18639eb5a8caf9808a9406aeb04df7c9881a
179263 F20110114_AACTWG elmidany_a_Page_075.jpg
e96096b7577ebb92b7fb1bf70ffca92a
e4cc412d0fc2d5127a0184c8bb0f5c8e3d8cd44d
645029 F20110114_AACUAX elmidany_a_Page_080.jp2
e49a70b9a837c51faa3637ff5b177d9d
5c1e4f1e0fcdbc373e36b903261cbcfed340e448
99344 F20110114_AACUZJ elmidany_a_Page_133.QC.jpg
bb6d26ae3f347c8d464bf6b070a4cb48
e1c984d0d5af72af43cee25c41291a40455a3316
27035 F20110114_AACUYU elmidany_a_Page_122thm.jpg
dd4848b252db89301aaa9f22295a46a0
086348ff9ef046224e0b956b25e93d534f803fe9
205015 F20110114_AACTVS elmidany_a_Page_060.jpg
cc0110e644336d17477c8120575e40f7
fa3be6d642aff2d4a0328f1909c105b4a6132f6f
715265 F20110114_AACUCB elmidany_a_Page_115.jp2
062afed5f60e4edc6a9c1a1e2c9011ed
d4ea68035e2f6d1d5dea6196d395f87002022e69
92303 F20110114_AACUBN elmidany_a_Page_101.jp2
eae98e32766e51ae72f0f3cef5962530
0e3d06cb7963de0045c08c865d223fc0726fa6a2
185455 F20110114_AACTWH elmidany_a_Page_077.jpg
5c07bc5699bd5326b2ef0da017d92e24
3841f6663cacd3de7766e1e790251c8c6845873f
668927 F20110114_AACUAY elmidany_a_Page_081.jp2
0a02d9600d9cad89043b4aa2e02bc1a9
00d0df9078cf9a0b9bee8817271d0228e42333ce
21793 F20110114_AACUZK elmidany_a_Page_134thm.jpg
273e0e6ad802bdae08e595074c42ba3c
29a2e7e50515a7b655e94a5d646a6d8f164c956d
67208 F20110114_AACUYV elmidany_a_Page_122.QC.jpg
41e42fac6b423257582a72712de1dacb
a4a5da505e98406b60400bff2931358e75dfc529
199485 F20110114_AACTVT elmidany_a_Page_061.jpg
6870e527aff0b80df63608ee43f25aa7
c4c46485f8ab9ee245ce628ad0cc507d291f324d
803354 F20110114_AACUCC elmidany_a_Page_116.jp2
472822ba5055641eac519929b6b70b67
d6c1916dd1cfa0401bf202a96f5fc267c11c4bc6
941707 F20110114_AACUBO elmidany_a_Page_102.jp2
f8e344ea0c876db62e5bea17c6746af5
fe5822ee457eea24b206133158165dca4a72c857
229572 F20110114_AACTWI elmidany_a_Page_078.jpg
8a636942fc58394ea368339e07f9cf3b
73aec0ada9e2401951bae963b7c1ccee3d255c6d
774855 F20110114_AACUAZ elmidany_a_Page_082.jp2
0f86095ff9a0590ec592f890ca469b13
7bb63020701ce0020c83625c13125722ca362800
46190 F20110114_AACUZL elmidany_a_Page_135thm.jpg
74171e30c608d1d56914ccb8b31afb89
108ae7ca6ce43131fda841bb97450551bacd3fba
12254 F20110114_AACUYW elmidany_a_Page_123thm.jpg
01ee1007ea8a1007d1f662e55f70ea3a
1b13e8bf65ca85310d24650324ce5188a4008e74
196251 F20110114_AACTVU elmidany_a_Page_062.jpg
653061274adbb94c81573ac6d96d3995
80a393bf775072a63df5409eba706ae908e17b40
487906 F20110114_AACUCD elmidany_a_Page_119.jp2
f1ce2c93b768074f3622924db2eb5b2f
cecb90798426559f14e82ff47b00bb041963fdd9
546529 F20110114_AACUBP elmidany_a_Page_103.jp2
431db82498e3f4df9f5ed90f5f4808be
111b425036ebd41da77899e05281897303130093
157632 F20110114_AACTWJ elmidany_a_Page_079.jpg
6cdd0377a09e76e925b2f035ffc503cd
b50dbc81ea2c2b3a995005d4961fbb7e0c756722
89628 F20110114_AACUZM elmidany_a_Page_135.QC.jpg
3757080481add7ec7e5bff96ebb39b03
c52bc48ebcdbd023634138c062fc0bb14f061370
36833 F20110114_AACUYX elmidany_a_Page_123.QC.jpg
97f54bff0b1fc59fc04d96780c6a36e2
298f6e930470ea373471c14f3414e8cc49a52a2a
122400 F20110114_AACTVV elmidany_a_Page_063.jpg
b2b144be8c993747ea993b80f9a32e88
3d4087b0564999ced741e9a03cd213c853c39845
850407 F20110114_AACUCE elmidany_a_Page_120.jp2
9cee421b5ff9e82c3235f5350cbe6c62
76f59d6ccce23dcd356eaee6cf20a0f9dda88285
761990 F20110114_AACUBQ elmidany_a_Page_104.jp2
db48c2a22c2c3cbdae9037c307f5b2e9
4e8875f3be45af0cd5e483ecc02dd9a0284a995c
146932 F20110114_AACTWK elmidany_a_Page_080.jpg
006d03e282ca320fb8f47f7b59fade9a
ea9d741017ecff65c8225b71569bd618bf623fa1
49684 F20110114_AACUZN elmidany_a_Page_137thm.jpg
41e5d93908bb529af3f07447ed626c98
88e9592841e3c01b4446fda78236b22db78586c4
43851 F20110114_AACUYY elmidany_a_Page_124thm.jpg
7ddd9cd9dbdb876869c749161cf3165d
0d6aefecd82c8071cf56e2d246cbb39b51628b5f
165107 F20110114_AACTVW elmidany_a_Page_064.jpg
8d7d5b533416a3a2eab1cf776b366f0e
394eff11be390f4de4c2451c6d1fa607d5fbb5a4
857678 F20110114_AACUCF elmidany_a_Page_121.jp2
2b966099f6bb243f480dbfe9aa64a0a8
b9830357dd0b7fe2ab8f12c1d89b4d5371b4220e
987883 F20110114_AACUBR elmidany_a_Page_105.jp2
c29401cf81d35c3bc8a53d696fd8509d
8a2f6448a9cd27f4b51e9fcce68760f15a5855de
160638 F20110114_AACTWL elmidany_a_Page_081.jpg
37e530a6b146135a6e37847d2499af5d
ef43e492852f70f91badc44be699384b7da93cce
88847 F20110114_AACUZO elmidany_a_Page_138.QC.jpg
0fe3a76dadf4c78db7cc497b00318110
6d30460e14cea04705b993f36032e136d6424f75
81476 F20110114_AACUYZ elmidany_a_Page_124.QC.jpg
433a597af393b77547d02663bf98d4cc
7dd5b505a57a306ccf9d0c56a75b96643d7f42d7
117515 F20110114_AACTVX elmidany_a_Page_065.jpg
0de45df242769d2815dbee20aab3dfa8
c7923840e5756ff8343341dbf3ab07f9794a5263
807725 F20110114_AACUCG elmidany_a_Page_122.jp2
3d04e907e2a952ae8e853f5d56be0f18
f837a703f696cd8f9792f6daf7330eff15d50ccc
156303 F20110114_AACTXA elmidany_a_Page_097.jpg
0a932c771c0a73288a6e491079467766
7884bebcf2f231e8eca241136a3f4c17a1204297
80517 F20110114_AACUBS elmidany_a_Page_106.jp2
702cb87f7ad4a0a4585a600f2d1cb78f
14b8c8a1225c496a73bae365b444ef2acf59b8c4
175186 F20110114_AACTWM elmidany_a_Page_082.jpg
fe423f216aba45455f87f52eb05e7a59
5b45ac64ebcd88aea4a746dc2d6e1e1f943008a1
49878 F20110114_AACUZP elmidany_a_Page_139thm.jpg
9589cf0a6f893941b8980589413fe872
027331ac9ee8bdd4172fe56913581fa3db5a1ab3
202489 F20110114_AACTVY elmidany_a_Page_066.jpg
d76eb728028c339cd1584eb239abfd5f
1406a58528e2d2bd3adb4d7df3a328910a4fc731
50305 F20110114_AACUCH elmidany_a_Page_123.jp2
7d13895cf72b344a3baca39bd768e32a
26ab9750ff7e9a8b840e0ab39c791f8e2fe735fc
217176 F20110114_AACTXB elmidany_a_Page_098.jpg
dae5470669d830b6d01e47664f46bf41
a12152665096081f3d4621d244e15ba269743777
840718 F20110114_AACUBT elmidany_a_Page_107.jp2
a80169af9b2228b207eb6ed2a9f485ed
aeee319ce9c0193cf7dee371415909adf01a92b0
132828 F20110114_AACTWN elmidany_a_Page_083.jpg
1b702ef19539d833a1e482a8438cc080
4657fd6e7434e4231c89957d1ad3525e211c5c68
231696 F20110114_AACTVZ elmidany_a_Page_067.jpg
d15c3f3213b7da0c8934aad4e9a3aa39
75e517c972bfe687366fa316bb3637da3c0ea6ca
919673 F20110114_AACUCI elmidany_a_Page_125.jp2
699a4881dbc5fce57cbbb77c169b40cd
fad9c433e6b54a28e3e38ef01d7b3289137476c5
183471 F20110114_AACTXC elmidany_a_Page_100.jpg
6728aef635c59e3904a96ed104379cd7
cf8599dbf8cf269a7fdd4984920970fc3582c2fe
792739 F20110114_AACUBU elmidany_a_Page_108.jp2
20f2d79fc33ed1639568b51cd25dbb91
f48daadfdc65129dabf525d2b5202481f917be85
105353 F20110114_AACUZQ elmidany_a_Page_139.QC.jpg
959e7b3a5f3168b292f95a12f6fb57d2
75602526fdcf48a07427b4c9e5480959d1cac000
776490 F20110114_AACUCJ elmidany_a_Page_126.jp2
67407934bd253a325372636a6ead0961
152e1fb733d56c84770d08584b2e464128dabab7
180501 F20110114_AACTXD elmidany_a_Page_101.jpg
64cc284c94cea651de51dc7b0287dd85
d7c9f7a52611168875d83eb5b4492df925c0aed3
908510 F20110114_AACUBV elmidany_a_Page_109.jp2
fda8d7d472bd3c147c6ff5568f427db3
0db0886aec2eada1a92da080cbe62a4dca78af76
187810 F20110114_AACTWO elmidany_a_Page_084.jpg
8ad35bd22fbb881f628634a2835c0742
9a64aa46fab976c27e5a7386b84858deb175127b



PAGE 1

SEPARATING DOLOMITE FROM PHOSPHATE ROCK BY REACTIVE FLOTATION: FU NDAMENTALS AND APPLICATION By AYMAN A. EL-MIDANY A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Ayman A. El-Midany

PAGE 3

To all who made this work possible, and To all who had parts in shaping the person that I am today.

PAGE 4

ACKNOWLEDGMENTS In the Name of Allah, the Most Gracious, the Most Merciful. Oh Allah! Make useful for me what you have taught me, and teach me knowledge that will be useful to me. I entrust you with what I have read and studied. Oh Allah! Bring it back to me when I am in need of it. You are my protector, and the best of aid.. I thank Allah for blessing me with all of the good that I have, and I ask that I will be able to use such gifts to help those in need. Amen. I have fortunate to be able to study in an excellent institution surrounded by a group of extremely intelligent people. I would like to recognize the amazing staff in Department of Materials Science and Engineering at University of Florida. I would also like to thank everyone who helped me arrive at this point. Special thanks and recognition go to my advisor, Dr. Hassan El-Shall. He gave me the opportunity to do research in his group. His drive and commitment to research have inspired and motivated me through this work. He has been a constant source of ideas and encouragement over the years. Many thanks go to my committee members; especially Dr. Spyros A. Svoronos, who helped me greatly and bore with me in the modeling part of my study. I would also like to thank the other members of my committee (Drs. John J. Mecholsky, Wolfgang Sigmund, and Susan Sinnott from the Department of Materials Science and Engineering, and my external committee member, Dr. Ben Koopman, Department of Environmental Engineering). I would also like to recognize the remarkable past and present members of Dr. El-Shalls research group for their assistance, advice, and support. Finally, I cannot thank my family enough for all of their love, encouragement, and counsel. To my parents, I owe everything. They have taught me many things. Above all iv

PAGE 5

they taught me that patience, determination, persistence, and faith are the keys to success. I thank my wife for her help and patience over the years. I thank the Particle Engineering Research Center (PERC) staff for helping me to get started and for providing support and guidance along the way. Much of the experimental work in my study could not have been accomplished without their help, and for that I am extremely grateful. This work was supported by the Florida Institute of Phosphate Research (FIPR), and the National Science Foundation / Particle Engineering Research Center (NSF/PERC). v

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................iv LIST OF TABLES .............................................................................................................ix LIST OF FIGURES ...........................................................................................................xi ABSTRACT .....................................................................................................................xiv CHAPTERS 1 LITERATURE REVIEW.............................................................................................1 Introduction...................................................................................................................1 Previous Efforts to Separate Dolomite from Phosphate Rock......................................2 Anionic Flotation...................................................................................................4 Direct flotation...............................................................................................4 Reverse flotation............................................................................................5 Cationic Flotation..................................................................................................8 Nonconditioning Flotation.....................................................................................9 Two-Stage Conditioning Process........................................................................11 The CLDRI Process.............................................................................................12 Acid Leaching.....................................................................................................14 Physical Separation Methods......................................................................................14 Heavy Media Separation.....................................................................................14 Sizing, Attrition Scrubbing, and Desliming........................................................15 Problem Statement......................................................................................................15 Novelty and Scope of Study.......................................................................................15 Goal of Study..............................................................................................................16 2 MATERIALS AND METHODS...............................................................................17 Flotation Study............................................................................................................18 Screening Design (Plackett-Burman Design)......................................................18 Central Composite Design...................................................................................18 Flotation Procedures for Screening and CCD Designs.......................................21 Factorial Design for Conditioning with PVA before Flotation...........................21 Flotation Procedure for Conditioning Statistical Designs...................................21 Density and Surface Area Determination............................................................22 vi

PAGE 7

PVA Preparation..................................................................................................23 Spectrophotometric Determination of PVA........................................................23 Static (Equilibrium surface tension)....................................................................24 Dynamic Surface Tension (DST)........................................................................24 Maximum bubble pressure method..............................................................24 Dynamic surface tension setup.....................................................................26 Contact Angle......................................................................................................26 Measurement of Contact Angle...........................................................................27 Surface Characterization by FTIR.......................................................................28 Zeta Potential Measurements..............................................................................28 Adsorption...........................................................................................................28 Dissolution Kinetic Experiments.........................................................................29 Pilot Scale Experiments..............................................................................................30 The Sluice............................................................................................................30 Sluice Experiments..............................................................................................30 3 TECHNICAL ASPECTS OF REACTIVE FLOTATION: PRELIMINARY TESTS32 Introduction.................................................................................................................32 List of Potential Surfactants: Surfactants Screening..................................................33 Initial Screening Tests................................................................................................33 Specific Tests.......................................................................................................34 Acid type......................................................................................................34 Acid dosage..................................................................................................35 Surfactant type..............................................................................................36 Surfactant dosage.........................................................................................37 Particle size..................................................................................................38 Oil-Surfactant Experiments.................................................................................39 Poly-Vinyl-Alcohol (PVA) Testing....................................................................41 Effect of acid concentration.........................................................................41 Effect of particle size....................................................................................43 PVA with acid vs. PVA on the particle surface...........................................44 Preliminary Testing Using PVA Coated Particles......................................................45 Flotation of a Representative Sample..................................................................45 Fraction of Representative Sample (-4+1.19 mm)..............................................46 Effect of Time on Weight Percentage of Floated Particles.................................46 Effect of Size on the Flotation.............................................................................47 Summarizing Remarks................................................................................................48 4 OPTIMIZATION OF THE REACTIVE FLOTATION PROCESS AND PILOT SCALE TESTING................................................................................49 Introduction.................................................................................................................49 Optimization of the RF Process..................................................................................49 Screening Statistical Design................................................................................49 Central Composite Design...................................................................................52 vii

PAGE 8

Research to Reduce the PVA Consumption...............................................................52 Experimental Design of the PVA Conditioning Step..........................................56 Saturation Experiment.........................................................................................56 Pilot Scale Testing: Sluice Results.............................................................................57 5 FUNDAMENTAL STUDIES....................................................................................58 Polyvinyl alcohol-Particle Interaction........................................................................58 Introduction.........................................................................................................58 Surface Tension...................................................................................................60 Contact Angle......................................................................................................63 Work of Adhesion...............................................................................................64 Zeta Potential.......................................................................................................67 Zeta potential of minerals in supernatant solutions......................................70 Zeta potential in the presence of PVA.........................................................71 Infrared (FTIR)....................................................................................................72 Adsorption Study.................................................................................................75 Kinetics of adsorption..................................................................................76 Adsorption isotherm.....................................................................................79 Calculation of PVA adsorption and parking area.........................................82 Adsorbed layer thickness.............................................................................84 Effects of pH................................................................................................84 Adsorption in supernatants...........................................................................87 Desorption...........................................................................................................88 Desorption in Acidic Solution.............................................................................89 Bubble Formation and Stability..................................................................................90 Effect of Particle Size..........................................................................................91 Effect of PVA Coating........................................................................................94 Modeling of Bubble Formation and Dolomite-Particle Separation............................96 Blockage of the Acid Model................................................................................96 Disappearance of Active Sites Model.................................................................99 Empirical Models..............................................................................................101 First-order kinetic model............................................................................101 Acid concentration dependence model......................................................103 Estimation of CO 2 Required to Float the Dolomite Particle.............................106 Bubbles Stability...............................................................................................109 Validation of the Model Using a Different Carbonate Mineral........................112 Summarizing Remarks..............................................................................................114 6 CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK............116 Conclusions...............................................................................................................116 Recommendations for Future Work.........................................................................119 LIST OF REFERENCES.................................................................................................120 BIOGRAPHICAL SKETCH...........................................................................................129 viii

PAGE 9

LIST OF TABLES Table page 2-1. Experimental Plackett-Burman design for 11 variables...........................................19 2-2. Rotatable central composite design with 5 levels and 3 variables...........................20 2-3. Factorial design for conditioning with PVA.............................................................21 3-1. Surface-active agents used in RF tests.....................................................................33 3-2. Size analysis of the dolomitic phosphate sample.....................................................34 3-3. Chemical analyses of the dolomitic phosphate sample............................................34 3-4. Effect of sulfuric acid concentration........................................................................35 3-5. Effect of sulfuric acid concentration on the bubble formation.................................36 3-6. Preliminary tests of high dolomitic phosphate with different surfactants................37 3-7. Effect of surfactant concentration (SDSO 3 2 %).....................................................38 3-8. Effect of particle size on the dolomitic particle floatability.....................................39 3-9. Effect of presence of surfactant and oil layer on particle flotation..........................40 3-10.Effect of acid concentration......................................................................................42 3-11.Effect of acid concentration above 2 %....................................................................43 3-12.Comparison between coating the particle before adding it to acidic solution and adding the particle to acid+PVA solution without coating......................................45 3-13.Results of representative sample (4+1.19 mm particle size)..................................46 3-14.Effect of time on the flotation of 4+1.19 mm particle size....................................47 3-15.Effect of particle size on flotation............................................................................47 4-1. The MgO % removal in floated fraction for Plackett-Burman design.....................50 ix

PAGE 10

4-2. Plackett-Burman calculated t values for different variables.....................................51 4-3. Effect of saturated acid on the RF............................................................................57 4-4. Comparison of two different coating methods and their effects on MgO removal..57 5-1. Chemical analysis of the samples used in adsorption tests......................................66 5-2. Fitting parameters of second-order kinetic model....................................................77 5-3. Chemical analysis of the samples used in kinetic tests............................................91 5-4. Acid-blockage model parameters.............................................................................97 5-5. Active sites disappearance parameters...................................................................100 5-6. Reaction rate constants of first-order model...........................................................103 5-7. Acid concentration dependence parameters...........................................................105 5-8. Active sites disappearance parameters for dissolution of calcite...........................113 x

PAGE 11

LIST OF FIGURES Figure page 1-1. Simplified flowsheet for phosphoric acid production................................................3 1-2. Simplified flowsheet of USBM direct flotation.........................................................6 1-3. Flowsheet of TVAs diphosphonic depressant process..............................................8 1-4. Flowsheet for IMCF cationic flotation.......................................................................9 1-5. University of Alabama pebble processing, simplified flowsheet.............................10 1-6. Flowsheet for University of Florida two conditioning stages..................................11 1-7. Simplified flowsheet for CLDRI dolomite flotation................................................13 2-1. Strategic plan to achieve research goals...................................................................17 2-2. Five-level, rotatable central composite design.........................................................20 2-3. Characteristic bubble-pressure vs. time curve in MBP............................................25 2-4. Setup for the measurement of dynamic surface tension by MBP.............................26 2-5. Contact angle and its relation with surface tension..................................................27 2-6. Contact angle measuring apparatus..........................................................................28 2-7. Schematic diagram for separation of dolomite from phosphate in sluice................30 3-1. Effect of acid concentration on bubble formation and flotation time.......................36 3-2. Effect of the particle size on the time needed to float the particle...........................44 4-1. Effect of PVA dosage and acid conc. on MgO recovery % ....................................53 4-2. Effect of PVA dosage and acid concentration on MgO % ......................................54 4-3. Cube graph for MgO % and MgO recovery %.........................................................56 5-1. Contact angle and interfacial surface tension at solid/liquid/gas interface..............60 xi

PAGE 12

5-2. Structural formula for PVA......................................................................................61 5-3. Equilibrium Surface tension for aqueous PVA solutions at 23 C...........................61 5-4. PVA crystal structure; the dotted lines indicate hydrogen bonds.............................62 5-5. Static contact angle variation with the PVA concentration......................................64 5-6. Work of adhesion as a function of PVA concentration............................................65 5-7. Cumulative pore size distribution for apatite sample...............................................66 5-8. Cumulative pore size distribution for dolomite sample............................................67 5-9. Zeta potential of dolomite and apatite samples........................................................68 5-10. Zeta potential of dolomite in absence and presence of polymer............................68 5-11. Zeta potential of dolomite in absence and presence of polymer............................69 5-12. Effect of polymer on the shear plane position........................................................69 5-13. Zeta potential of apatite and apatite in dolomite supernatant.................................70 5-14. Zeta potential of dolomite and dolomite in apatite supernatant.............................71 5-15. Zeta potential of dolomite in apatite supernatant before/after PVA adsorption.....72 5-16. Zeta potential of apatite in dolomite supernatant before/after PVA adsorption.....72 5-17. Infrared transmittance spectra of poly vinyl alcohol (PVA)..................................73 5-18. Infrared absorbance spectra of dolomite and dolomite/PVA system.....................74 5-19. Infrared absorbance spectra of apatite and apatite/PVA system............................75 5-20. Calibration curve for PVA concentration measurements.......................................76 5-21. Kinetics of PVA adsorption on dolomite and Apatite............................................78 5-22. Adsorption isotherm of PVA on dolomite at 23C................................................81 5-23. Adsorption isotherm of PVA on apatite at 23C....................................................82 5-24. Polymer-conformation changes with increase the concentration...........................83 5-25. Effect of pH on adsorption of PVA on apatite and dolomite.................................85 5-26. Adsorption isotherm of apatite in dolomite supernatant solution, and dolomite in apatite supernatant solution......................................................................................87 xii

PAGE 13

5-27. Desorption of PVA from apatite surface................................................................88 5-28. Desorption of PVA from dolomite surface.............................................................89 5-29. The bubble formation mechanism..........................................................................90 5-30. Schematic diagram of reaction kinetics of the RF..................................................91 5-31. Dissolution kinetic of different particle sizes in terms of calcium ions ................92 5-32. Dissolution kinetic of different particle sizes in terms of magnesium ions............93 5-33. Dolomite particle (1-2 mm) before and after the reaction with 3 % H 2 SO 4 ...........93 5-34. Dissolution kinetics of coated and uncoated 0.2-0.3 mm dolomite particle ..........94 5-35. Growth of the bubbles with time on the dolomite particle surface, 50 X...............95 5-37. Acid blockage model fitting to 0.25-0.35 mm experimental data..........................98 5-38. Active sites disappearance model fitting to 1-2 mm data.....................................100 5-39. Active sites disappearance model fitting to 0.25-0.35 mm data...........................100 5-40. First-order model fitting for data of 0.25-0.35 mm particle size..........................102 5-41. First-order model fitting for data of 1-2 mm particle size....................................102 5-42. Model fitting for data of 0.25-0.35 mm particle size............................................104 5-43. Model fitting for data of 1-2 mm particle size......................................................105 5-44. Sequence of calculating the particle density and membrane elasticity................108 5-45. Density changes for different particle sizes..........................................................109 5-46. Dynamic surface tension of PVA concentration at different CO 2 flowrate.........111 5-47. Elasticity of PVA membrane vs. PVA concentration ..........................................112 5-48. Calcite dissolution: model vs. experimental data 0.25-0.35 mm particle size.....113 5-49. Calcite dissolution: model vs. experimental data for 1-2 mm particle size..........113 xiii

PAGE 14

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SEPARATING DOLOMITE FROM PHOSPHATE ROCK BY REACTIVE FLOTATION: FUNDAMENTALS AND APPLICATION By Ayman A. El-Midany December 2004 Chair: Hassan El-Shall Major Department: Materials Science and Engineering Florida is currently one of the largest phosphate producers in the world. Unfortunately, the phosphate industry in Florida is suffering from depletion of the phosphate-rich ore. Dolomite, however, is a major impurity in the ores. The presence of MgO, in dolomite, can cause several problems during production of phosphoric acid and final fertilizers. Efforts to solve the dolomite problem have been reported by several researchers throughout the world. Yet, only heavy media separation process has been applied commercially. However, the process was discontinued because of its low separation efficiency in terms of grade and recovery. In this study, we tested an innovative idea based on simple fact that dolomite, as a carbonate mineral, generates CO 2 when exposed to a slightly acidic solution, capturing CO 2 bubbles at dolomite particle surface can selectively float the dolomite and separate it from phosphate. Formation of such bubbles requires a surface-active agent(s) at xiv

PAGE 15

solution/dolomite interface. Preliminary tests indicated that polyvinyl alcohol (PVA) could serve this function. This process is called the Reactive Flotation, RF. To develop this process and to scale it up for industrial use, we conducted fundamental and applied research studies. Among these studies were optimizing of RF process by investigating the factors affecting the process (including bench and pilot scale tests) using statistical experimental designs. It was found that the acid concentration, PVA concentration, and particle size are the main factors. The optimum conditions were determined and applied to usage in a gravity separation device, (i.e., a sluice). Using such a device, a concentrate contains 0.65 % MgO with > 94 MgO % removal can be achieved. In addition, fundamental studies revealed that the hydrogen bonding is the adsorption mechanism confirmed by adsorption isotherm of PVA on dolomite and phosphate, adhesion of PVA to dolomite surface (using contact angle and surface tension), Fourier Transform InfraRed (FTIR), and Zeta potential. Moreover, PVA film thickness and polymer film elasticity (using dynamic surface tension) revealed that the optimum conditions could keep the PVA film at its highest elasticity. On the other hand, modeling was developed and used to predict CO 2 formation rate and subsequent particle density change. xv

PAGE 16

CHAPTER 1 LITERATURE REVIEW Introduction Traditionally, the Florida phosphate industry has effectively responded to the competitive pressures of the market place by implementing technological improvements to mining and beneficiation methods. With further expansion of fertilizer production capacity worldwide and depletion of Florida's high-grade deposits, competitive pressures are expected to continue. To remain the world leader, Florida must continue to seek environmentally sound technological improvements in all areas of phosphatic fertilizer production. Gu et al. [1999] estimated that the Bone Valley phosphate reserve in Central Florida (which can be cost effectively processed with available technology) may only last about 20 years at the current mining rate. In the near future, with the depletion of the low-dolomite, easy-to-process Bone Valley siliceous phosphate deposits, phosphate mining will move farther south and southeast. The phosphate matrix in the southern extension is leaner in grade and high in dolomite. Dolomite is the most troublesome, causing higher consumption of sulfuric acid, reducing filtration capacity, and lowering P 2 O 5 recovery in fertilizer manufacturing. Therefore, the MgO content of dolomite is an important index in evaluating the quality of phosphate concentrate. In Florida, the MgO content in final concentrate is usually required to be less than 1 %. Geological and mineralogical data show that about 50 % of the phosphate resource would be wasted if the high dolomitic zones were bypassed in mining, and about 13 % of 1

PAGE 17

2 the resource would be lost if the dolomite pebbles were discarded. Although heavy media separation (HMS) technology was tried commercially, to process a portion of dolomite pebbles in Florida, in most cases the high-dolomite phosphate has been bypassed during mining operations because no technology is available to process high-dolomite pebbles economically. Technically and economically feasible beneficiation technology for processing dolomite pebbles is needed to recover phosphate value, extend Florida phosphate reserves, and prolong the life of Florida phosphate industry [Geo et al., 2002]. Previous Efforts to Separate Dolomite from Phosphate Rock Phosphate rock is one of the basic materials for fertilizers. About 90 % of the phosphate rock output in the world is used for fertilizer production. The United States is one of the worlds largest producers of phosphate rock and fertilizer. Its phosphate production represents approximately one-third of the worlds total [Becker, 1989]. Phosphoric acid is an important intermediate product for production of fertilizers. Current phosphoric acid technology essentially comprises sulfuric acid attack, and separation of the phosphoric acid from the calcium sulfate crystals resulting from the reaction. Both the attack and separation, which are effected by filtration, are considerably influenced by the nature of the ore and its impurities [Becker, 1989]. Figure 1-1 shows the main processes in phosphoric acid production. As mentioned, the main impurity in Florida phosphate ore is dolomite. Dolomite, (Ca, Mg)CO 3, is the main source for calcium and magnesium. As impurities, CaO and MgO affect the phosphoric acid production by increasing sulfuric acid consumption, in addition to the MgO stays with the acid phase, and greatly increases viscosity.

PAGE 18

3 Ca 10 (PO 4 ) 6 F 2 + 10 H 2 SO 4 + 20 H 2 O 6 H 3 PO 4 + 10 CaSO 4 .2H 2 O + 2 HF Reaction Sulfuric acid Filtration Counter-Current Washing Phos p ho gyp sum Phosphoric acid Wash Water Phosphate Concentrate Figure 1-1. Simplified flowsheet for phosphoric acid production Therefore, because dolomite is the most common impurity, developing an efficient and economical method for separating dolomite from apatite is an effective way to extend phosphate mine life. Generally, acidulation of phosphate rock requires a feed of less than 0.8 % MgO, which has not been achieved at a commercial scale with ores high in MgO. Dolomite is also a worldwide problem. All of the major phosphate-producing countries have faced, or are going to face, the dolomite problem. Salt-type minerals (such as apatite, calcite, dolomite, barite, fluorite and scheelite) have been successfully separated from oxides and silicates by flotation methods. However, these minerals are often difficult to separate from each other because they contain similar alkaline earth cations that cause them to exhibit similar surface (electrical) behavior. All the zero points of charge (ZPC) of phosphates and carbonates are located in the acid pH range (i.e., 4-5.5) [Smani, 1973; Smani et al., 1975]. Systematic research in carbonate phosphate ore beneficiation started in the early 1950s [El-Shall et al., 2004]. Techniques used include the following: Direct flotation of phosphate, using sodium silicate to depress carbonate impurities.

PAGE 19

4 Reverse flotation of the carbonate impurities by depressing the phosphate using materials such as fluorides, fluosilicic acid, diphosphonic acid, phosphoric acid, aluminum sulfate-tartrate, and inorganic phosphate salts. Rapid change of conditioning parameters (such as the University of Florida two-stage conditioning process, Alabama non-conditioning process, the IMC-F process, and the low-pH conditioning process proposed by Minemet research). Physical methods: gravity concentration, sizing, desliming, heavy media separation. Calcination: although expensive because of its energy requirement, high-temperature destruction of the carbonate followed by leaching has been practiced in North Africa with low cost energy. Acid leaching: carbonates are easily dissolved by acid, and weak acids can be used to dissolve the carbonate, leaving the apatite. Anionic Flotation Direct flotation Sodium silicate is widely used as carbonate depressant [Fuerstenau, 1968; Ananthapadmanabhan and Somasundaran, 1985; Rao, 1988; Dho and Iwasaki, 1990]. Encouraging results were obtained in both batch and continuous pilot tests conducted by the United State Bureau of Mine, USBM, [Davis et al., 1984]. After 1 lb/t sodium silicate was added in each of three cleaning stages, a phosphate rougher was upgraded to about 30.7-31.3 % P 2 O 5 0.6-1.3 % MgO with a P 2 O 5 recovery from 55.1 to 89.7. The original matrix contained 5.47-10.2 % P 2 O 5 and 1.9-7.6 % MgO. Rao [1989] studied the effects of major flotation parameters on phosphate grade and recovery using sodium silicate as a calcite depressant. Maximum flotation of apatite occurred at pH 8, and phosphate recovery decreased with increasing pH. Sodium silicate at a concentration below 0.0013 M had no depressive effect on calcite. However, at pH values above 10, calcite depression was observed even in dilute silicate solution. A concentrate with 25.6 % P 2 O 5 at 80 % recovery was obtained.

PAGE 20

5 Efforts to concentrate high-dolomite Florida phosphate ores by flotation, using conventional fatty-acid/fuel-oil flotation [Liewellyn, 1982], showed that sodium silicate depression on dolomite was not enough to obtain a concentrate less than 1 % MgO. This may be due to the presence of silica in this ore in addition to dolomite. Reverse flotation Depression of phosphate: This method was used when the apatite was more abundant than calcite and dolomite (as in most phosphate ore). Among depressants used are as follow: Fluosilicic acid: USBM investigated the removal of carbonate from phosphate rock using fluosilicic acid as a depressing agent and tall oil-fatty acid as a collector for carbonate mineral flotation (Figure 1-2) [Rule, 1982;1985]. At collector dosage 0.7 kg/t, 0.45 kg/t depressant, a concentrate of 25.5 % P 2 O 5 and 0.84 % MgO with P 2 O 5 recovery 89.5 % was obtained from feed with 20.8 % P 2 O 5 and 1.1 MgO %. Abdel Khalek [2000] studied the phosphate depression using fluosilicic acid, but recovery using this process was less than 60 %. Ataly [1985] pointed out that for depressing Turkish ores fluosilicic acid was not efficient as other depressants because of the low recovery (57 %) obtained using fluosilicic acid. Acid depressants: Orthophosphoric acid was found to decrease adsorption of the fatty acid on apatite because of the increased surfacial solubility of apatite.

PAGE 21

6 Phosphoric acid also was used to depress apatite (by the USBM, in separating dolomite from a southern Florida ore) [Liewellyn, 1982]. The produced concentrates P 2 O 5 % did not exceed 28 %, and the MgO % was not less than 1.9 %. High MgO Pebble N aOH Concentrate Cleaning Phosphate Flotation Ro ugher Phosphate Flotation Conditioning Desliming 150 Mesh Scrub Desliming 150 Mesh Screening 28 Mesh Hammer Mill Sodium Silicate Fatty acid/Fuel Oil + 28 Mesh 150 Mesh Fines 150 Mesh Fines Dolomite Rough e r T ail Dolomite Cleaner T ail Figure 1-2. Simplified flowsheet of USBM direct flotation

PAGE 22

7 Lawendy [1993] found phosphoric acid to be the best among conventional depressants of phosphate. The concentrate was 32-36 % P 2 O 5 1 % MgO at 70-75 % overall P 2 O 5 recovery from feed 16-25 % P 2 O 5 and 4-11 % MgO. In other study [Hsieh and Lehr, 1985], diphosphonic acid was used to depress phosphate (TVA process), (Figure 1-3), with fatty acid collectors for beneficiating an ore containing 72 % SiO 2 1.05 % MgO and 6.3 % P 2 O 5 The resulted concentrate contained 73.3 % SiO 2 0.7 % MgO and 6.4 % P 2 O 5 Although sulfuric acid is a phosphate depressant, its ability to attack phosphate and its high need for pH control limits its usage [Elgilani and Abouzeid, 1993]. Jacob Engineering [Gruber et al., 1987] used sulfuric acid as a pH regulator to completely depress phosphate at pH 5.5. Phosphates: Rao [1979] found that the depressing action of dipotassium hydrogen phosphate is pH dependent. Phosphate can be depressed at pH 5-8, while carbonate can be depressed only at alkaline pH; which means that selective flotation can be obtained. However, the concentrate produced by this process contained no more than 24 % P 2 O 5 Sodium chloride: The effect of sodium chloride and other salt addition on dolomite and apatite flotation has been studied [Rao, 1985]. It was shown that sodium chloride acts as apatite depressant during flotation at acidic pH values with oleic acid, while dolomite flotation is not affected. It was, however, observed that dolomite particles larger than 48-mesh are difficult to float and they tend to remain in the sink fraction. Trends observed during Hallimond cell tests of single and mixed minerals were confirmed with mixtures of minerals and a natural ore at bench-scale level. Under more

PAGE 23

8 favorable conditions in the bench-scale tests, MgO content of the mineral mixtures was reduced from 5 % to 1 % or less, with at least 90 % recovery of P 2 O 5 in the sink fraction. 48+ 40 0 Mesh High MgO Pebble Diphosphonic acid/ Oleic Acid Phosphate Concentrate Silica Flotation Dolomite Flotation Scrubbing & Desli m ing Classification Ball Mill Grinding Dodecylamine Hydrochloride + 4 8 Mesh 40 0 Mesh Fines 400 Mesh Fines Float Dolomite T ail Float Silica T ail Figure 1-3. Flowsheet of TVAs diphosphonic depressant process Alizarin Red S (ARS): ARS is an organic depressant. It is used to depress either phosphate or carbonate, depending on pH. ARS can depress calcite impurities at a pH 8-10, while apatite can be depressed at a pH below 6 [Fu and Somasundaran, 1986; Xiao and Somasundaran, 1989]. Cationic Flotation The IMC-F process (Figure 1-4) is a cationic-flotation process for silica-free ores. The ore is reagentized with a cationic collector (amines or their salts) and the dolomite

PAGE 24

9 separated from phosphate at a rougher stage followed by one or two phosphate-cleaning steps [Snow, 1979; 1982; Lawver and Snow, 1980]. By using IMC-F process, a concentrate of 31 % P 2 O 5 0.86 % MgO from feed of 2 % MgO with 83 % P 2 O 5 recovery can be obtained. High MgO Pebble Amine Recleaning phosphate Flotation Cleaning Phosphate Flotation Ro ugher Phosphate Flotation Conditioning 20-30 sec Dewatering 65-70% Silica Flotation Desliming 150 Mesh Screening 24-35 Mesh Rod Milling Phosphate Conc. Acetic acid/Amine Oversize -150 Mesh Fines Dolomite Rough er Tail Dolomite Cleaner T ail Frother Dolomite Tailings Silica Tailings Figure 1-4. Flowsheet for IMCF cationic flotation Nonconditioning Flotation A research contract between FIPR and Mineral Resources Institute (MRI) at the University of Alabama aimed to develop a selective flotation process (Figure 1-5) to separate dolomitic limestone from Florida phosphate ore [Hanna and Anazia 1990]. This process depends on the difference in adsorption rate of fatty acids on both phosphate and

PAGE 25

10 carbonates. Fatty acids adsorb more rapidly on carbonate surfaces than on phosphate surfaces. At slightly acidic pH (5.5-6.0), carbonates floated immediately on addition of collector and frother. After carbonate froth collection, the pulp is conditioned briefly to float phosphate thereafter with no more addition of collector. High MgO Pebble Flotation p H 4-7 Conditioning Dolomite Flotation Conditioning Desliming 325 Mesh Scrubbing 35 % solids Desliming 150 Mesh Screening 35 Mesh Rod Milling Phosphate Conc. Fatty acid/Pine oil +35 Mesh -150 Mesh Fines Dolomite Tail Sodium silicate/Fatty acid Silica Tailings Silica Tailings -325 Mesh Figure 1-5. University of Alabama pebble processing, simplified flowsheet

PAGE 26

11 Two-Stage Conditioning Process This process was investigated through FIRP university of Florida project, in 1983 [Moudgil, 1988]. The process involves conditioning the feed at pH 10, followed by reconditioning at a lower pH before flotation, Figure 1-6. Selective flotation of dolomite from apatite was observed for both single and mixed minerals, by reconditioning at pH 4. Basic pH Modifier/Fatty acid High MgO Pebble 65x150 Mesh Feed Frother Dolomite Flotation p H 4-5 Reconditioning Flotation P H 9-10 Further processing to remove silica Conditioning pH=10 H 2 SO 4 Silica (Sink) Float Dolomite Sink (Apatite & Fine Quartz) Figure 1-6. Flowsheet for University of Florida two conditioning stages To understand the mechanisms of observed selective flotation, further studies were conducted involving electrokinetic behavior, oleate adsorption, infrared spectroscopy,

PAGE 27

12 and solubility of the minerals. Selective flotation of dolomite, by reconditioning at pH 4, is attributed to the combined effect of higher oleate adsorption on dolomite and hydrolysis of the adsorbed oleate molecules to oleic acid at lower pH values. The method has been tested in the laboratory with several dolomite-apatite mixtures and natural magnesium phosphate samples from the Florida phosphate field. Reductions of the MgO content of the samples from 1.8-4.0 % MgO to below 1 % MgO at recoveries of about 90 % P 2 O 5 in the sink fraction have been obtained in bench-scale flotation tests. Reconditioning pH was determined to be one of the most important parameters in this process. The CLDRI Process In China, over 90 % of the phosphate resources are dolomitic, with MgO content ranging from 2 % to over 10 %. For economical and rational use of domestic phosphate resources, China Lianyungang Design and Research Institute (CLDRI) has conducted studies for many years and developed some flotation technologies for removing MgO impurity from phosphate ore. These flotation technologies have successful industrial applications [El-Shall et al., 1996]. Since 1997, CLDRI has conducted extensive studies on Florida high dolomitic pebble with the objective of using CLDRIs fine particle flotation technology to recover phosphate (Figure 1-7). In 2000, pilot scale testing (400 lb/h) was carried out jointly by CLDRI, Jacobs engineering group, and IMC phosphates company at Jacobs pilot plant facility. Preliminary economic and technical analysis by Jacobs engineering (based on pilot plant test results) showed that CLDRIs fine-particle flotation technology could be used economically and technically for processing Floridas dolomitic phosphate pebbles [Geo et al., 2002].

PAGE 28

13 Figure 1-7. Simplified flowsheet for CLDRI dolomite flotation

PAGE 29

14 The CLDRIs fine-particle flotation technology (Figure 1-7) utilized for processing Florida high-dolomitic phosphate pebbles has two steps. The first step is grinding for liberation and second nondesliming after grinding, which improves flotation recovery and use of Florida resources, and lowers production cost. Acid Leaching Strong and weak acids were used in the acid leaching process. In strong-acid leaching processes, sulfuric acid was used to completely dissolve the dolomite from phosphate ore. In the early 1970s, the USBM investigated magnesium impurity removal from phosphate concentrates by acid leaching [Rule, 1970]. The MgO level was reduced to less than 0.1 % by using dilute sulfuric acid, but the phosphate loss was about 6 % P 2 O 5 Weak-acid leaching, using weak organic acids, was also studied. Pilot plant studies [Abu El-shah, 1991] using 6.7 % acetic acid solution could remove more than 80 % MgO. The advantage of using weak organic acids is that (unlike strong acids), they do not attack other minerals. The overall disadvantage is that acid leaching, in general, needs fine particle sizes produced by crushing and grinding which means more expenses. Physical Separation Methods Heavy Media Separation The only commercial process to separate dolomite from phosphate is heavy media separation [Wiegel and Hwang, 1984]. Heavy media cyclones are used in IMCF mines for processing the high-dolomite portion of the ore. The concentrate with 1.2 % MgO was obtained from feed that has 3 % MgO.

PAGE 30

15 Sizing, Attrition Scrubbing, and Desliming The dolomite is more friable than phosphate rock. Thus, dolomite can be separated from phosphate by discarding the fine fraction (mainly dolomite) by scrubbing or sizing [El-Shall and Bogan, 1990]. Problem Statement The various beneficiating techniques discussed have various limitations or disadvantages. In acid leaching, strong acids can attack phosphate as well as dolomite and can be associated with environmental problems associated with this process. In flotation processes for high dolomitic phosphate, because they could not reduce the MgO % below 1 % are not applied commercially. In CLDRI fine flotation process, by which the MgO % could be reduced to less than 1 %, concerns are as follows: handling of fine concentrate (dewatering and transport), capital cost (grinding and multiple process steps), and operating cost: increasing the MgO % in the phosphate rock increases reagent consumption, and requires grinding, thus increasing energy consumption and plant operating cost. The only process with commercial application was heavy media separation (HMS), which also suffers from high MgO % (more than 1 %) in the concentrate in addition to about 50 % phosphate loss. Novelty and Scope of Study We sought a process that separates dolomite from phosphate pebbles resulting in a phosphate concentrate with less than 1 % MgO. Since the dolomite is mostly in the pebble fraction and not attached to phosphate, the idea came from well-known reaction between carbonates and acid producing carbon dioxide gas. Dolomite, as a carbonate

PAGE 31

16 mineral, generates CO 2 when exposed to a slightly acidic solution, as shown in the following chemical reaction: Dolomite + Acid salt + CO 2 Capturing the generated CO 2 at the dolomite particle surface, into bubbles using coating agent (which is a kind of surfactant or polymer) can selectively float dolomite, and separate it from phosphate rock. Since this process depends on chemical reaction between dolomite and acidic medium to form a stable bubble at the particle surface and then to float it, this process is called the Reactive Flotation (RF) process. The RF has the following requirements (1) presence of acidic medium at low percentage, and (2) Presence of surfactant that: attaches to the dolomite, is permeable by acid, allows formation of stable bubbles at the particle surface, is impermeable to CO2 gas, and is stable enough to float the particle to the surface. Goal of Study The goal of this study is to develop and to optimize a novel and an effective process to separate the dolomite from phosphate rock. This process will not only reduce the MgO % to less than 1 % but also increase the efficiency of the subsequent processes in phosphoric acid production such as filtration and consequently reduce the phosphate losses. In addition to develop this novel process, a fundamental study needs to be done to elucidate Mechanism(s) involved in this novel process.

PAGE 32

CHAPTER 2 MATERIALS AND METHODS To achieve the goals of this research different measurements and techniques are needed to elucidate polyvinyl alcohol (PVA)/ particle interactions, and bubbles formation and stability as well as its modeling the following strategy will be followed, Figure 2-1. Film thickness Work of Adhesion Acid Diffusion CO2formation rate Bubble formation model Elasticity (Dynamic surface tension) Time required to start particle flotation Density balance Density change Adsorption Isotherm FTIR Surface Tension Contact Angle Zeta Potential 12 Film thickness Work of Adhesion Acid Diffusion CO2formation rate Bubble formation model Elasticity (Dynamic surface tension) Time required to start particle flotation Density balance Density change Adsorption Isotherm FTIR Surface Tension Contact Angle Zeta Potential 12 Figure 2-1. Strategic plan to achieve research goals As it can be seen form Figure 2-1, the first block represents the required techniques and measurements to elucidate polyvinyl alcohol (PVA)/ particle interactions. On the other hand, the second block represents another part of our goal which is bubble formation and stability as well as its modeling. 17

PAGE 33

18 Flotation Study In this study, screening and central composite statistical experimental designs are used to determine the effect of various factors on the removal of MgO and its recovery. The Plackett-Burman design was used to screen out which of the 7 factors being considered were the most statistically significant. In other words, it is a tool that helps determine which factors made the greatest impact on the MgO removal. A central composite design was used to determine if there were any interactions between the three most significant factors determined from the Plackett-Burman design. Screening Design (Plackett-Burman Design) The variables examined in the screening stage were acid concentration (1% and 3 %), Polyvinyl alcohol concentration in the coating solution (1 % and 3 %), the PVA molecular weight (60,000 and 150,000), the degree of hydrolysis (88 %-99 %) of the PVA, particle size (-0.5 mm and 5-9 mm), and drying time of the coating (0 and 2 hrs). The levels were chosen according to the exploratory test results from phase I of this project. The Plackett-Burman design used for this study had 11 factors and 12 experimental runs (Table 2-1). MgO % was determined using chemical analysis. Chemical analyses were performed using wet chemical methods, acid attack (digestion) method as described below. Central Composite Design A central composite design was used to further investigate the three most significant factors (acid concentration, PVA concentration, and particle size) determined from the screening design. Five levels were used. The central composite experimental design had 17 experimental runs (Table 2-2 and Figure 2-2). For each run, the products are chemically analyzed as above-mentioned and the MgO % and MgO removal % were

PAGE 34

19 calculated for all the experimental runs. The computer program, Design expert (SAS Institute, Inc.), was used to perform the statistical analysis for the central composite design. The data was analyzed by fitting the response variables to a second-order polynomial model. Table 2-1. Experimental Plackett-Burman design for 11 variables Run No. Variable Levels 1 2 3 4 5 6 7 8 9 10 11 1 + + + + + + 2 + + + + + + 3 4 + + + + + + 5 + + + + + + 6 + + + + + + 7 + + + + + + 8 + + + + + + 9 + + + + + + 10 + + + + + + 11 + + + + + + 12 + + + + + + Plackett-Burman Variables Used in the Reactive Flotation Study Variables Low Level (-) High Level (+) 1) Acid conc 1 % 3 % 2) PVA conc 1 % 3 % 3) dummy 4) dummy 5) PVA M.W 60,000 150,000 6) Degree of hydrolysis 88 % 99 % 7) Time of drying 0 2 h 8) dummy 9) Particle size -0.5mm 5-9 mm 10) dummy 11) dummy dummy variables are used to calculate the variance of the experimental runs

PAGE 35

20 Y Z X Figure 2-2. Five-level, rotatable central composite design. The X, Y, and Z axes represent the 3 variables, sulfuric acid concentration, PVA concentration, and particle size. The circles represent each experimental run. Table 2-2. Rotatable central composite design with 5 levels and 3 variables Run No. Acid conc PVA % Particle Size 1 -1 -1 -1 2 -1 -1 +1 3 -1 +1 -1 4 -1 +1 +1 5 +1 -1 -1 6 +1 -1 +1 7 +1 +1 -1 8 +1 +1 +1 9 -1.68 0 0 10 +1.68 0 0 11 0 -1.68 0 12 0 +1.68 0 13 0 0 -1.68 14 0 0 +1.68 15 0 0 0 16 0 0 0 17 0 0 0 Levels Variables -1.68 -1 0 +1 +1.68 Acid conc. (%) 0.32 1 2 3 3.68 PVA, (lb/t) 0.6 2 4 6 7.4 Particle size, mm -0.5 0.5-1 1-2 2-5 5-9

PAGE 36

21 Flotation Procedures for Screening and CCD Designs Using 300 ml (3 % H 2 SO 4 ) in 400 ml beaker, particles were pre-coated by spraying 3 % PVA solution (using a mist delivery nozzle) then added to the acidic solution. The experiment was left for 5 min. Dolomite floats to the surface, which was separated by decantation on a screen. The products (floated and unfloated fractions) were dried, weighed, and chemically analyzed. Factorial Design for Conditioning with PVA before Flotation This design was used to study the factors affecting the conditioning step with PVA, which is a promising step for reducing PVA consumption. Table 2-3 shows the design and levels used. Table 2-3. Factorial design for conditioning with PVA Run No. PVA concentration, % Pulp density (S/L ratio) Conditioning time, min 1 -1 -1 -1 2 -1 -1 +1 3 -1 +1 -1 4 -1 +1 +1 5 +1 -1 -1 6 +1 -1 +1 7 +1 +1 -1 8 +1 +1 +1 9 0 0 0 10 0 0 0 11 0 0 0 Levels Variables -1 0 +1 PVA concentration, % 0.25 0.625 1 Pulp density 30 % 50 % 70 % Conditioning time, min 2 11 20 Flotation Procedure for Conditioning Statistical Designs 50 gm of ore were added to 1 % PVA solution in a 600 mL beaker according to specific solid/liquid ratio. The slurry was mixed for required time at 300 rpm (using 2.5

PAGE 37

22 cm magnetic bar). After the conditioning time is elapsed the solids were screened, the solids were used for flotation in acidic solution and the filtrate solution was taken to PVA analysis to determine the PVA residual concentration. Chemical analysis: About 0.5 gm of the dried and ground representative sample were digested in 50 ml of aqua regia by boiling on hotplate until the reaction complete. After cooling, this solution was filtered through a Whatman 42 filter paper into 1000 ml volumetric flask. The filter paper and residue were then washed at least five times to remove all the traces of dissolved salts and acid. The filtrate was diluted with distilled water and thoroughly mixed and the concentrations were measured using Inductively Coupled Plasma, ICP, emission spectrometer, (Perkin Elmer Optima 3200RL Optical Emission Spectroscopy, Norwalk, CT). ICP was calibrated using AFPC rock check No.22. Standard sample concentrations were measured from time to time in the course of the analyses to check the accuracy of the calibration curve. Five replicates were taken and averaged for each element. The following wavelengths were used: Element Wave lengths, nm Ca 315.887, 317.933, 393.366, and 396.847 Mg 279.077, 285.213, and 279.553 P 213.617, and 214.914 Fe 238.204, 239.562, 259.939, and 234.349 Al 308.215 Density and Surface Area Determination Densities of powders were measured using a Quanta Chrome ultrapycnometer. The specific surface area (m 2 /g) of dolomite and apatite particles was determined using Quanta Chrome NOVA 1200 instrument. (Quantachrome Corp., Boynton Beach, FL).

PAGE 38

23 The analysis was performed using a five-point BET (0.10, 0.15, 0.20, 0.25, 0.30) using Nitrogen gas. The particles were outgased at 60C for 24 hours before analysis. PVA Preparation PVA was kindly supplied by Celanese chemicals Co., USA. Different PVA types with various ranges of molecular weights and degree of hydrolysis are used in screening factorial design, namely: Type Average Molecular Wt. Degree of Hydrolysis, % Celvol 165 150000 99 Celvol 540 150000 88 Celvol 305 60000 99 Celvol 203 60000 88 In the rest of the study, PVA, Celvol 165, was used. PVA flakes were gradually dissolved in deionized water at 90 o C with continuous mixing by magnetic stirrer and heated at 90 C for 2 h to produce different PVA concentrations (0.5, 1, 2, 3, and 4 %). Spectrophotometric Determination of PVA PVA solution of certain concentration treated with 25 ml of 4 % boric acid solution and 2 ml of iodine solution (prepared from 1.27g of iodine and 25 g of KI/l), in that order. The resulting solution is diluted to 100 ml and kept at 25C and its absorbance measured at 690 nm against a blank solution containing the same amount of boric acid and iodine solution. Beers law applies in the concentration range of 0-40 mg of PVA/l of solution [Finch, 1973; 1992]. Absorbance spectra were taken using ultraviolet/visible spectrophotometer (UV-Vis spectrophotometer, Perkin-Elmer Lambda 800). All spectra were taken at room temperature.

PAGE 39

24 Static (Equilibrium surface tension) Equilibrium surface tensions were measured for freshly prepared solutions by the Wilhelmy plate method at room temperature. The platinum plate was always cleaned and heated to a red/orange color with a Bunsen burner before use. Dynamic Surface Tension (DST) Maximum bubble pressure method The dynamic surface tension (DST) measurement is important in the dynamic interfacial processes where the equilibrium surface tension is hard to be achieved. The DST can be measured by different methods such as: the drop weight [Jho and Burke, 1983; Miller et al., 1993], oscillating jet [Thomas and Hall, 1975], capillary wave [Kelvin, 1871], growing drop technique [Macleod and Radke 1993], and the maximum bubble pressure method [Mysels, 1986; Ross et al., 1992]. The maximum bubble pressure (MBP) technique is the most commonly used technique for the measurement of DST, and has been applied to a variety of surfactant and polymer solutions [Fainerman et al., 1994; Tamura et al., 1995]. The physical principle behind the MBP measurement is the Laplace pressure; the pressure inside a curved liquid interface is higher than the ambient. This excess pressure (P) can be calculated from the Laplace equation [Adamson and Gast, 1997]. P = r 2 + gh Where is the surface tension, r is the radius of curvature of the bubble, is the liquid density, g is the gravitational constant and h is the depth of the bubble in the liquid. The first term expresses the Laplace pressure due to the curved gas/liquid interface, and the second term is the hydrostatic pressure due to the liquid height above the forming

PAGE 40

25 bubble. The first term will vary during the life cycle of the bubble, while the second term will remain constant. Figure 2-3 shows a typical cycle of a bubble and the resulting pressure within the capillary. After a bubble breaks off from the capillary tip, the pressure is the lowest. As more gas flows into the capillary, the pressure builds up as the gas is pushed out of the capillary and the radius of curvature at the tip decreases. During this expansion process, surfactant is populating the new interface and acting to lower the surface tension. At the point of minimum interface radius of curvature, where a hemisphere of gas is formed at the capillary tip, the pressure reaches maximum. Figure 2-3. Characteristic bubble-pressure vs. time curve in MBP. Both the minimum radius of curvature and the surface tension, as described by the Laplace equation, govern the maximum pressure experienced. On the incremental addition of gas, the bubble expands out of the capillary tip and the radius of curvature gradually increases. This results in a drop in the Laplace pressure, and a resulting rapid expansion of the bubble. At some point, the bubble breaks off from the capillary tip and the whole process starts again.

PAGE 41

26 Dynamic surface tension setup The maximum bubble pressure apparatus was constructed using a differential pressure transducer purchased from Omega Engineering, Inc. (Stanford, CT), with a sensitivity of 0 to 10 in. (25 cm) H 2 O (0 to 2500 Pa). A #23 steel needle was used as a capillary, with nominal 0.025 in. (0.64 mm) external diameter, 0.013 in (0.33mm) internal diameter, and a flush cut tip. The capillary diameter was chosen so that the various resistance of water to bubble growth could be ignored. Such internal and viscous effects are a potential source of error in these measurements that need to be taken into consideration [Garrett and Ward, 1989]. All measurements were conducted with the capillary tip 1 cm beneath the liquid surface. CO 2 gas was used as the bubbling gas to simulate the gas produced in the reaction of acid with dolomite and an oscilloscope connected to the pressure transducer was used to determine the bubble frequency and the dynamic surface tension (Figure 2-4). Rotamete r Polymer Solution Pressure Transducer Pressure Volta g e 1.0 cm de p th Bubble formation Oscillosco p e Air Flow Figure 2-4. Setup for the measurement of dynamic surface tension by maximum bubble pressure method. Contact Angle The angle formed at a point on the line of contact of three phases, of which at least two are condensed phases, by the tangents to the curves obtained by intersecting a plane

PAGE 42

27 perpendicular to the line of contact with each of three phases. One of the phases must be a liquid, another phase may be solid or liquid and the third phase may be gas or liquid. In other words, it is the angle included between the tangent plane to the surface of the liquid and the tangent plane to the surface of the solid, at any point along their line of contact. The surface tension of the solid will favor spreading of the liquid, but this is opposed by the solid-liquid interfacial tension and the vector of the surface tension of the liquid in the plane of the solid surface. s Solid Gas Liquid l Figure 2-5. Contact angle and its relation with surface tension Equilibrium wetting properties of liquids and solutions can be interpreted in terms of the balance of surface tensions (Figure 2-5). sg sl = lg cos Measurement of Contact Angle The contact angle goniometer, Figure 2-6, is a device mounted on the optical bench to examine a single liquid drop or gas bubble resting on the smooth, planar, solid substrate. The drop/bubble is illuminated from the rear to form a silhouette, which is viewed through a telescope connected to computer. The image of the drop/bubble appears on the computer monitor by which the brightness and contrast, focusing and amount of illumination can be adjusted. After adjusting the image the contact angle can be measured.

PAGE 43

28 Figure 2-6. Contact angle measuring apparatus Surface Characterization by FTIR The spectorometer used for FTIR was a Magna-IR760 spectrometer by Nicolet in transmission setting. As for the spectra recorded in transmission mode, KBr pellets were prepared for the background measurements. Another pellets were prepared by mixing about 1 mg of filtered powder with 50 mg of KBr. Transmission spectra were recorded using a DTGS detector, with 128 scans and 4 cm -1 resolution. Zeta Potential Measurements In the zeta potential measurement tests, 1 g of ground mineral samples was added into a 250 ml beaker in which 100 ml 0.003 M KNO 3 solutions were added. The suspension was conditioned for 3 min during which the pH was adjusted, followed by 5 min of conditioning after adding the polymer. It was then allowed to settle for 3 min, and about 10 ml of the supernatant was transferred into a standard cuvette for zeta potential measurement using a Brookhaven ZetaPlus Zeta Potential Analyzer. Solution temperature was maintained at 25 C. Ten measurements were taken and the average was reported as the measured zeta potential. Adsorption These studies were carried out at 25C using 0.5gm mineral sample in 50 ml solution having the desired PVA concentration and the ionic strength fixed at 3X10 -3 M

PAGE 44

29 using KNO 3 One set of experiments was performed first to determine the effect of conditioning time at a fixed pH (pH 7) and PVA concentration was 100 ppm (3.33X10 -8 M). Then set of experiments was also carried out as a function of pH at the same PVA concentration. Adsorption isotherm was also determined for PVA/dolomite, PVA/apatite, PVA/dolomite/apatite supernatant, PVA/apatite/dolomite supernatant systems at pH 7 and 3X10 -3 M KNO 3 as an ionic strength. After the conditioning step, the suspension was centrifuged at 3000 rpm for 5 min using an IEC-Medispin centrifuge (Model 120). Then a volume of the clear solution was withdrawn for measurement of the PVA residual concentration. In adsorption kinetic experiments, a 0.5 gm of dolomite or apatite were added to 100 ppm (3.33X10 -8 M) PVA solution a 30 ml screw-capped bottle and was shaken at room temperature using Wrist hand shaker for a given recorded time (0.05, 0.083, 0.0166, 0.25, 0.5, 1, 2, 4, 6, 8, 10, 16, 20, 24 h). After that the contents of each bottle were allowed to settle, followed by centrifugation step using an IEC-Medispin centrifuge (Model 120). The amount of PVA adsorbed was determined by difference between the initial and final concentrations measured spectrophotometrically with a Uv-vis spectrophotometer (UV-Vis spectrophotometer, Perkin-Elmer Lambda 800). These experiments were used to establish the equilibrium time for adsorption of PVA on dolomite and apatite. This equilibrium time was used as contact time in adsorption isotherm experiments. Dissolution Kinetic Experiments Dolomite specimens (1-2 mm and 0.2-0.3 mm), uncoated and coated with PVA, were added to 2000 ml aqueous medium (3 % H 2 SO 4 ) at 23 0.5C to carry out 120 min static dissolution tests. After a definite time interval, solution samples were collected

PAGE 45

30 periodically and concentrations of Mg 2+ and Ca 2+ were determined with ICP. The relations of concentration, C, with time, t was plotted to explore the dissolution kinetics of the dissolution process. Pilot Scale Experiments The Sluice The pre-conditioned material is introduced into a feed box, which contains 3 % sulfuric acid. The reaction of sulfuric acid with particles starts in the feed box forming bubbles around dolomite particles and creating the required density difference for the separation in that device. The particles soon enter the moving stream and, because of the bubbles on dolomite particles, they settle at lower rate and float to the surface when the enough bubbles formed and carried away till they reach the lip of the trough and discharged with liquid overflow. Since phosphate particles do not produce CO 2 and have no bubbles at their surfaces, they are collected at the bottom of the trough (Figure 2-7). Figure 2-7. Schematic diagram for separation of dolomite from phosphate in sluice Pre conditioned F eed Dolomite Phos p hate Inclination angle to horizontal Reject Trough angle Feed B o x 3 % Su lfuric acid C once n t r at e Sluice Experiments 250 gm of ore (a mixture of dolomite and phosphate), with particle size 1-2 mm, were conditioned in 1 % polyvinyl alcohol solution for 5 minutes, then screened to separate the solids from PVA solution. The screened solution was analyzed to determine the actual PVA consumption by depletion method. Or the material was sprayed with the

PAGE 46

31 equivalent amount of PVA and was tumbled to attain a good coating. After that, the coated solids were poured into sluice feed box which the sulfuric acid and the reaction starts to generate CO 2 bubbles. The particles move through the sluice at flow rate 125 ml/sec (2 GPM). The floated fraction, dolomite, was collected on 0.5 mm screen and the sink fraction, phosphate, was collected in the sluice trough. The floated and sink products were analyzed for MgO, P 2 O 5 CaO, and insoluble residue (I.R.).

PAGE 47

CHAPTER 3 TECHNICAL ASPECTS OF REACTIVE FLOTATION: PRELIMINARY TESTS Introduction The reactive flotation process, as mentioned earlier, requires certain conditions to give good results and separate dolomite from phosphate rock. The main requirements for RF to work properly are the acidic medium and surfactant or polymer, which attaches to the dolomite, to allow the acid to diffuse through but at the same time remain impermeable to the CO 2 gas produced by the reaction. Moreover, the formation of CO 2 will lead to stretching of the membrane-like surfactant layers, which requires an elastic behavior of that membrane. The elasticity of the membrane formed around the particle will then allow the formation of stable bubbles. The stable bubbles are the main criterion in RF to float the dolomite particles and separate them from others. At the starting point of this process, the most well behaved acids and surfactants, under RF conditions, were not known. Different acids and surfactants can be used but it was unknown which one would best fulfill the previously mentioned requirements for the RF process. For that reason, in the current study, preliminary tests were conducted using different acid types (sulfuric, hydrochloric, nitric, acetic) within a particular concentration range (0.5 4 % v/v) as well as different types of surfactants (anionic, cationic, and non-ionic), which were prepared as 1-2 % surfactant solution and used in two different ways: (1) surfactant was added to acidic solution and/or (2) surfactant was used to coat the particle before adding the particle to the acidic solution. Also, poly-vinyl32

PAGE 48

33 alcohol (PVA) was used in this study. Vegetable and corn oil were used at the acidic solution/ air interface and oil height was 0.5-0.6 cm. List of Potential Surfactants: Surfactants Screening A list of potential surfactants used in this study, are given in Table 3-1. The list contains different types of surfactants, (i.e., cationic, anionic, and non-ionic, as well as polymers). Table 3-1. Surface-active agents used in RF tests Surfactant Chemical Name Type Arquad 12-50 Dodecanaminium+isopropyl alcohol Cationic Lilamine N-3-aminopropyl-N-tallow alkyltrimethylenediCationic Armac HT Amines, hydrogenated tallow alkyl acetates Cationic Armid HT Amides Cationic Armeen 12 D dodecanamine Cationic Ethofat Tall Oil Anionic SDSO 3 Sodium dodecyl-sulphonate Anionic SDSO 4 Sodium dodecyl-sulphate Anionic Oleic acid Oleic acid Anionic Arlacel Sorbitan sesquioleate Non-ionic Span-20 Sorbitan monolautrate Non-ionic G-1096 Polyoxyethelene-sorbitol hexaoleate Non-ionic Brij-97 Polyoxyethelene oleyl ether Non-ionic Arlatone T Polyoxyethelene polyofatty acid ester Non-ionic Tween-21 Polyoxyethelene Sorbitan monolautrate Non-ionic Tween-85 Trioleate Non-ionic PVA Poly-Vinyl-Alcohol Non-ionic Initial Screening Tests A 100-pound sample of high-dolomitic phosphate oversize pebble was obtained from the IMC Four Corners mine. This is not normal high MgO rock, but the oversized high MgO rock that is removed from pebble. As such, neither the screen weight fractions nor the chemical analyses are representative of pebble. However, the rock is representative of the high-dolomitic rock that needs to be separated. The rock was screened, and each fraction is chemically analyzed in Tables 3-2 and 3-3.

PAGE 49

34 Table 3-2. Size analysis of the dolomitic phosphate sample Size, mesh Size, mm Wt. % % Cumulative (passed) % Cumulative (retained) + 3/8 in 3/8 in +3.5 3.5+6 6+9 9+16 16 + 9.5 9.5+5.6 5.6+3.36 3.36+2 2+1 1 11.62 65.36 18.14 2.05 1.6 1.23 100.00 88.38 23.02 4.88 2.83 1.23 11.62 76.98 95.12 97.17 98.77 100 100.00 Table 3-3. Chemical analyses of the dolomitic phosphate sample Size, mesh P 2 O 5 % CaO % MgO % Fe 2 O 3 % Al 2 O 3 % +3/8 in -3/8 in +3.5 -3.5+6 -6+9 -9+16 -16 19.86 21.25 17.09 24.32 20.42 13.62 32.23 36.34 27.92 38.90 28.71 22.75 4.02 3.43 2.34 2.75 2.12 2.44 0.93 1.15 1.32 1.08 1.27 1.13 0.52 0.84 0.86 1.17 1.07 0.95 Since most of the rock was relatively coarse, a portion of it was crushed and re-screened to obtain smaller fractions that could be used in the tests. Specific Tests The results of some preliminary testing are shown in the following paragraphs in terms of acid type and dosage, surfactant type and dosage, particle size, and using of oil layer at the water/acidic solution interface. Acid type Different acids (H 2 SO 4 HCl, HNO 3 and Acetic acid) were used. Inorganic acids: HCl, HNO 3 and H 2 SO 4 gave the same results at 2 % acid concentration. On the other hand, the acetic acid (RCOOH) formed a very small number of bubbles at 2 % acid concentration. Increasing the concentration of acetic acid to 5 % gave no significant variation in bubble formation, which means that the organic acids are weak acids. Consequently, a large amount of organic acids is needed relative to inorganic ones.

PAGE 50

35 Acid dosage Since all the inorganic acids behave in the same manner, sulfuric acid was used in the rest of the experiments. Table 3-4 shows the effect of acid concentration on the bubble formation rate, amount of bubbles formed, and particle floatability that were taken as a criterion to compare the results. The surfactant used for coating the dolomitic particle was dodycyl sulphonate (SDSO 3 ). Table 3-4. Effect of sulfuric acid concentration Sulfuric acid concentration, % 0 0.25 0.5 1 2 Bubble formation rate Very slow Slow Fast Rapid Amount of bubbles formed Very small number of bubbles (about 5 bubbles for 10 minutes then no bubbles formed) The interval between formation of two bubbles was about 30 seconds Increasing the rate and number of bubbles formed Continuous formation of bubbles but at rate higher than 0.5 % Continuous and quick formation of many bubbles Flotation of dolomite particle No flotation Float after 5-10 minutes Float after 1 minute From Table 3-4, we notice that the reaction between dolomite and acid to produce CO 2 decreases noticeably with decreasing acid concentration and the flotation of dolomite particle takes longer time (5-10 min) in 1 % acid than in 2 % acid. Consequently, the 2 % sulfuric acid concentration is the most suitable acid concentration for bubble formation in terms of both bubble formation rate and amount of bubble formed. Because of the wide gap between 1 % and 2 % acid concentration another set of experiments were conducted as shown in Table 3-5 to get the optimum acid concentration.

PAGE 51

36 Table 3-5. Effect of sulfuric acid concentration on the bubble formation Sulfuric acid concentration, % 1 1.2 1.4 1.6 1.8 2 Bubble formation rate Increases Amount of bubbles formed Continuous and small No. of tiny bubbles Continuous & large No. of tiny bubbles Flotation of dolomite particle Takes 20 minutes to float, not stable (particle goes up and down) Floats and stays at water/oil (W/O) interface for 5 min. Then oscillates within 2 min intervals Floats and stays at (W/O) interface Table 3-5 shows that the flotation at 1 % & 1.2 % is not stable and flotation occurs after the particle loses part of its weight due to reaction with acid. Accordingly, this concentration deals with a small particles (smaller than 1mm) rather than large particles. The bubble formation rate is decreased with the acid percentage from 2 % to 1 %, but as the difference was narrower between 1.6-2 %, 1.6 % acid concentration was found to be optimal. The effect of acid concentration on the RF criterion was schematically summarized in Figure 3-1. Bubblformation e rate Number of bubbles formed Flotatin o A B C time Acid conc. Acid conc. Increases Increases Increases Acid conc. Figure 3-1. Effect of acid concentration on: A) Bubble formation rate, B) Number of bubbles formed, C) Flotation time. Surfactant type Table 3-6 shows the anionic, cationic and non-ionic surfactants used. According to RF criterion, the performance of Anionic surfactant was superior to Non-ionic, which was superior to Cationic. Anionic surfactants, especially SDSO 3 formed stable

PAGE 52

37 membranes that captured CO 2 and persisted for 10 minutes. This may be due to its ability to work at low pH. Therefore, SDSO 3 was used in the next experiments. Table 3-6. Preliminary tests of high dolomitic phosphate with different surfactants Surfactant Name Remarks Arquad 12-50 (1%) Lilamine (1%) Armac HT (1%) Armid HT (1%) Armeen 12 D (1%) stable bubbles for 20-30 sec, particle 2-3 mm can float less stable bubbles, stay 10 s, particle does not reach to surface of the solution (air-water interface) surfactant precipitated in the solution Ethofat (1%) SDSO 3 (1%) (2%) SDSO 4 (1%) (2%) Oleic acid Formation of bubbles on the particle surface then particle floats Bubbles form and go up to solution surface, but particle doesnt float Particle takes 2-3 sec to float Bubbles form and go up to solution surface, but particle doesnt float Flotation rate in case of SDSO 4 < Flot.rate in case of SDSO 3 Added as it is, particle stays at surface of the solution (air-water interface) with formation of many small bubbles. Arlacel Span-20 G-1096 Brij-97 Arlatone T Tween-21 Tween-85 Common observations for non-ionic surfactants: No significant change in behaviour of different non-ionic surfactants used The flotation of high dolomitic particle after 3-5 seconds and stay at the surface, while the low dolomitic does not float. Formation of stable bubbles on the floated particle surface Surfactant dosage The results of different dosages of surfactant sodium dedocyl sulfonate (SDS) are shown in Table 3-7 in terms of bare and coated particles. It is indicated from Table 3-7 that bare particles, without coating, form s bubbles quickly and in large quantity due to rapid dolomite/acid reaction but the bubbles are not stable at the interface where they can be collected.

PAGE 53

38 Table 3-7. Effect of surfactant concentration (SDSO 3 2%) SDSO 3 (2 %) 2 % 4 % 6 % Particle without coating Many bubbles formed Small particle (1-2 mm) floats Large particle (3-4 mm) goes up and down Many bubbles formed Particles (1-2 mm) and (3-4 mm) float Many bubbles formed Particles (1-2 mm) and (3-4 mm) float Particle coated with SDSO 3 Many bubbles formed -small particle floats-large particle does not float On the other hand, the coated particles take more time to float due to the diffusion of acid through the surfactant layer around each particle. Also, a greater number of bubbles are formed when no coating is used. Finally, when coating is not used each bubble forms as a separate bubble, while formation of bubbles clusters occurs when with a coating with SDS is used. Increasing the surfactant concentration increases the flotability of the particles and large particles (3-4 mm) can float at high surfactant concentrations of up to 4 %; after that, there is no noticeable change in floatability(i.e., 6 % is very much similar to 4 %). The oscillated motion of large particles (3-4 mm) for short period of time (3-5 sec) was noticed at 2 % surfactant concentration; however, the same size floats without oscillation at 4 % and 6 %. This means more surfactant is needed. Particle size The particle size effect on the RF process was studied for coated and uncoated particles. The observations of different size ranges were recorded in Table 3-8 which indicates that the coating affects the stability of the bubbles on the particle surface as well as the stability of the floated particle at the interface; however, even uncoated particles

PAGE 54

39 release many bubbles and float quickly for very short time due to their rapid reaction with acid. Table 3-8. Effect of particle size on the dolomitic particle floatability Particle Size, mm 0.5-1 1-2 2-4 4-6 Without treatment Particle floats and stays at water/air interface Particle floats for few seconds then oscillates No flotation -the particle oscillates without reaching the interface With treatment Floats and stays at interface Floats after 30 sec and stays at (w/a) interface Oscillates and finally stays at w/a interface Floats after 1 min then oscillates for 5 min This rapid rate of reaction causes rapid reduction in particle weight. In addition, it causes the production of many bubbles at a faster rate than bubbles-particle detachment, leading to presence of uncoated particle at the interface, although for shorter time than coated ones. But once the uncoated particle reaches the interface, the bubbles will not be stable anymore due to high surface tension, which is the reverse of coated particles. Large size particles need more bubbles; the bubbles formed were not enough to lift the particle causing the particle to oscillate up and down. On the contrary, the small particles 1-2 mm and smaller do not oscillate and float directly to the interface. To improve the stability of the floated particles at the interface, the reduction of the surface tension is required. Therefore, the oil layer was used at the interface for that purpose. Oil-Surfactant Experiments Addition of an oil layer, either vegetable or corn oil, to the acidic solution/ air interface was attempted with different types of surfactants to enhance the bubble stability at the interface, as shown in Table 3-9. The selected surfactants for this set of

PAGE 55

40 experiments were the best ones from Table 3-6. The particle was coated with the surfactant before its addition to acidic medium. Table 3-9. Effect of presence of surfactant and oil layer on particle flotation Surfactant Type Observations Arquad Lilamin Cationic Formation of bubbles due to dissolution of dolomite, but no flotation. 2 % sulfuric acid solution + oil layer Ethofat SDSO 3 SDSO 4 Anionic Bubble formation, Particle floats to water/oil interface for 10 min and then comes back to acidic solution. Arquad Lilamin Cationic Formation of cluster of bubbles on particle surface, but no flotation. 2 % sulfuric acid solution + 1 % surfactant Ethofat SDSO 3 SDSO 4 Anionic Bubble formation and particle oscillates and the bubble bursts at water/air interface. Arquad Lilamin Cationic The same as before Ethofat Bubbles form No flotation 2 % sulfuric acid solution + 1 % surfactant + oil layer SDSO 3 SDSO 4 Anionic Bubbles form on particle surface and particle floats to water/oil interface then to oil/air interface It is clear that, as can be seen in Table 3-9, the oil layer enhance the bubbles stability on the particle and at the interface. When the particle floats to the acidic medium/oil interface, the floated particle stays at the interface for up to 10 min. The presence of oil layer may add another layer, an oil layer, in addition to the surfactant layer on the particle surface, although this would depend on the adsorption of the surfactant to the particle as well as the type of the surfactant and its interaction with oil. From Table 3-9, it was observed that the presence of a low percentage of surfactant (1 %) in the acidic solution slightly increases the stability of bubbles in the absence of an oil layer. The presence of 1 % surfactant in the solution may reduce the fast detachment of surfactant molecules from the particle surface through reduction of the diffusion of surfactant molecule to bulk solution.

PAGE 56

41 By using acidic solution in combination with 1 % surfactant and an oil layer (Table 3-9), it was found that the oil layer causes more stable formation of bubbles on the particle when the particle floats to oil/air interface. The particle does not go back into the acid solution but stay at oil/ air interface. Consequently, large particles of up to 3 mm can float in the presence of the surfactant and oil layer. SDS is a good surfactant under these experimental conditions. However, Ethofat and cationic surfactants did not work well in the presence of surfactant mixed with acidic solution. This may be due to the interaction between the oil layer on the particle and those surfactants possibly causing the formation of a non-flexible surfactant layer, which cannot capture the CO 2 gas and cannot form the bubbles required for flotation. Poly-Vinyl-Alcohol (PVA) Testing PVA was used as a viscous polymer to coat and to float the dolomite particles. The effect of acid concentration, particle size, and different PVA concentration were studied. Effect of acid concentration PVA forms a membrane on the particle. This membrane needs a high acid concentration to penetrate it and to perform the reaction between the acid and particle surface. It was observed that there were a very small number of bubbles generated. However, most of the bubbles formed a cluster on the particle surface, which increased the volume of the bubbles at the surface and floated the particle. The floated particles were not stable at the beginning, going up and down at interface, but ultimately staying at the interface for long time (sometimes more than 2 days). On the other hand, PVA also floated the low dolomite-content particles, which were able to form bubbles. This observation is due to PVAs very low permeability by CO 2 which makes it the best among the surfactants used. By increasing the PVA

PAGE 57

42 concentration, longer time was taken to form a bubble and to float the particle, which requires high acid concentration, but in this case it became possible to float large particle sizes (up to 8 mm). Table 3-10. Effect of acid concentration Acid concentration, % 0.5 1 1.5 2 Particle coated with PVA Formation of cluster of bubbles Very low bubble formation rate No particle flotation (up and down at 1.5 %) Cluster of bubbles formed Low bubble formation rate Particle floats From Table 3-10, it is evident that the findings were that the PVA forms a membrane on the particle. This membrane needs a high acid concentration to penetrate it and to perform the reaction between the acid and particle surface to release the gas, which is responsible for floating the particle. No quick release of bubbles was observed at low acid concentrations, but rather only a very small number of bubbles at the beginning; furthermore, the rate of bubble formation and particle flotation was very slow. After a while, an increasing number of bubbles appeared and clustered on the particle surface, which then caused the particle to float. Moreover, at low acid concentrations, the floated particles were not stable, going up and down at the interface due to absence of surfactant. Even at high acid concentrations, the acid takes time to penetrate the PVA membrane and react with particle surface. For that reason, the bubble formation rate was low. But after the formation of bubbles began to cover the particle surface with clusters of bubbles, the particles started to float. The acid concentrations of less than 2 % were not effective due to the high viscosity of PVA membranes, which requires a high concentration difference to permeate

PAGE 58

43 the membrane and the reach the particles surface. Because the previous set indicated that the 2 % acid concentration was optimal, another set of experiments was conducted at higher acid concentration (> 2 %) under the premise that a concentration higher than 2 % would be even better. Therefore, additional experiments were conducted using 2.5-4 % acid concentrations. Table 3-11 shows the results of those experiments. Table 3-11. Effect of acid concentration above 2 % Sulfuric acid concentration, % 2.5 3 3.5 4 Bubble formation rate Increases with increasing acid conc. % Amount of bubbles formed Cluster of bubbles formed on the particle surface Rapid reaction no stable bubbles are seen on the surface (bubbles released in high rate from particle surface) Flotation of dolomite particle Particle Floats Is more stable than higher acid conc. Particle floats Less stable than 2.5-3 % Particle oscillates and finally goes down There was no flotation at acid concentrations above 3 % due to the high bubble formation rate and absence of stable bubbles that can attach to the surface of the particle and float it. The bubble formation rate increases with increasing acid concentration, 4 % > 3.5 % > 3 % > 2.5 %; however, 2.5-3 % resulted in the most stable bubbles. Different PVA concentrations (2 %, 3 %, and 4 %) were also used and the experiments indicated that 3-4 % concentration is most suitable for dolomite flotation. However, increasing PVA concentration increased the time it takes each particle to float. Effect of particle size Increasing the particle size increased the time needed to float the particle. This can be explained by variation of surface area to weight of the particle ratio, which decreases with increasing particle size. The results of these tests are shown in Figure 3-2.

PAGE 59

44 0102030-11-22-55-9+9Particle size, mmTime, min Figure 3-2. Effect of the particle size on the time needed to float the particle in 3 % H 2 SO 4 when the particle is pre-coated with 3 % PVA PVA with acid vs. PVA on the particle surface By using two beakers, one of them containing uncoated particles added to 3 % PVA + 3 % acid solution and the other containing PVA coated particles added to 3 % acidic solution, the comparison between the performance of different application of PVA was obtained. Table 3-12 shows the results in terms of bubble formation, rate of formation, and particle flotation. When uncoated particles were added to the PVA+acid solution, the acid attacked the particle surface. Consequently, the dissolution of dolomite is higher than in case of coating the particle with PVA before adding it to the acidic solution. Presence of micro-bubbles (cotton shape) at the particle surface indicated the dissolution of dolomite. Bubbles were vertically distributed rather than horizontally in first beaker, while the bubble distribution was horizontally uniform (spreading on the particle surface) in the second beaker. The bubble formation rate is higher in the first case than in the second.

PAGE 60

45 The flotation of the particles was better in the second case due to the presence of relatively large bubbles and uniform bubble distribution. Despite rapid dissolution of dolomite particles in the first case, flotation of the particles took a long time due to the small size of the bubbles. In both cases, the particles floated stay at interface for a long period of time (days). Table 3-12. Comparison between coating the particle before adding it to acidic solution and adding the particle to acid+PVA solution without coating First beaker Second beaker PVA + acidic solution in the beaker and the particle added without coating Particle was pre-coated with PVA before adding it to acidic solution Bubble formation rate Tiny bubbles were formed Relatively large bubbles were formed Amount of bubbles formed Cluster of the bubbles grows vertically and cover parts of the particle surface no uniform distribution of the bubbles on the particle surface. Cluster of the bubbles grows almost horizontally and uniformly distributed on the particle surface. Flotation of dolomite particle Takes long time (hours) to float Takes short time (min) to float Preliminary Testing Using PVA Coated Particles In the following preliminary tests, all particles were pre-coated with PVA Flotation of a Representative Sample Flotation of a representative sample, a mixture of phosphate and dolomite, was attempted in a one-liter beaker. The percentage of floated fraction by weight obtained after 6 hours was about 22 % and un-floated fraction was about 78 %. It was noticed that the smaller particle sizes (less than 1mm) floated immediately. It was then determined that using small size fractions of about 1 mm coated with PVA would reduce the flotation time with very low chance to dissolve the particle; it appears that good PVA coating

PAGE 61

46 prevented rapid acid-attack to the particle. On other hand, we should take into account that the PVA consumption will increase due to increased surface area. Fraction of Representative Sample (-4+1.19 mm) A bucket (11 in diameter and 15 in height) was filled with a solution of 15 liters of water containing 3 % sulfuric acid and 3 % PVA. The water height in the bucket was about 1 foot. Most of the particles were in monolayer form to avoid mechanical entrapment of the particles. After applying the finalized procedure, the MgO percentage in the representative sample was reduced to less than 1 % of the sink fraction; P 2 O 5 percentage was reduced to less than 4 % (Table 3-13). Table 3-13. Results of representative sample (4+1.19 mm particle size) Sample Chemical analysis, % Recovery, % Wt% MgO CaO P 2 O 5 I.R. MgO CaO P 2 O 5 I.R. Float Sink 14.55 85.45 13.21 0.75 32.56 34.83 2.71 19.35 6.89 10.55 75.1 24.9 13.73 86.27 2.33 97.67 10.01 89.99 Balance 100 2.56 34.50 16.93 10.02 100 100 100 100 Feed 100 2.6 34.78 16.84 10.32 100 100 100 100 Effect of Time on Weight Percentage of Floated Particles Two samples of 4+1.19 mm size were gathered at different flotation time (30 min and 60 min). The samples were immersed in a beaker containing a solution of 3 % PVA and mixed to ensure the coating of all particles with PVA before adding the sample to acidic water (3 % sulfuric acid). The same bucket as in the previous test was used and at the same conditions. Each experiment was observed for two days and the floated fraction was collected. Most of the particles (about 90 %) floated within 30 minutes and about 10 % in the remaining two days of the experiment. The results are shown in Table 3-14.

PAGE 62

47 Table 3-14. Effect of time on the flotation of 4+1.19 mm particle size Chemical Analysis Recovery Time, min Wt% MgO% CaO% P 2 O 5 % MgO% CaO% P 2 O 5 % Float Sink 16.75 83.25 13.62 0.78 24.25 31.12 0.76 18.89 77.84 22.16 13.55 86.45 0.8 99.2 Balance 100 2.93 29.97 15.73 100 100 100 30 Feed 100 3.03 30.38 15.76 100 100 100 Float Sink 17.44 82.56 13.6 0.81 26.17 30.8 0.94 19.06 78 22 15.22 84.78 1.03 98.97 Balance 100 3.04 29.99 15.90 100 100 100 60 Feed 100 3.03 30.38 15.76 100 100 100 The percentage of MgO was reduced to about 0.8 %; the percentage of P 2 O 5 was reduced to less than 1 % in the float fraction, which indicates that using PVA in a pre-coating solution is a promising process for separation of dolomite from phosphate pebbles. Effect of Size on the Flotation Two samples were crushed in a closed circuit to obtain the following sizes: 4+1.19 mm and -3.36 +1.19 mm. The conditions were the same as in the previous two tests. The results of these tests are shown in Table 3-15 within 5 minutes flotation time. The percentage of MgO was reduced to less than 1 % (0.6 % of the un-floated fraction in first experiment and 0.95 % in the second). Table 3-15. Effect of particle size on flotation Chemical analysis, % Recovery, % Size, mm Sample Wt% MgO CaO P 2 O 5 I.R. MgO CaO P 2 O 5 I.R. -4+1.19 Float Sink 20.5 79.5 12.13 0.6 30.34 30.24 3.31 19.13 8.1 9.6 83.9 16.1 20.55 79.45 4.27 95.73 17.87 82.13 Balance 100 2.96 30.26 15.89 9.29 100 100 100 100 Feed 100 3.03 30.38 15.76 9.46 100 100 100 100 -3.36+1.19 Float Sink 17.8 82.2 11.84 0.95 34.53 38.02 3.86 20.07 7.04 10.35 72.96 27.04 16.43 83.57 4 96 12.84 87.16 Balance 100 2.89 37.40 17.18 9.76 100 100 100 100 Feed 100 2.95 37.07 17.41 9.93 100 100 100 100

PAGE 63

48 Summarizing Remarks Separation of dolomite from apatite using CO 2 generated in the chemical reaction between dolomitic particles and slightly acidic media was explored by testing different parameters effecting this process, including: acid type and concentration, surfactant type and concentration, particle size, and use of polymer (PVA). The results were promising with anionic surfactants, especially SDSO 3 and Mr. Bubble solution with a concentration of 1.6 % sulfuric acid; however, low dolomite-content particles did not float. The most encouraging results were obtained with PVA, which floats the low dolomite-content particles, in addition to floatation of large particle sizes. Additional testing with a representative sample of ore mixture (phosphate and dolomite) showed the ability to reduce the MgO content to less than 1 % by using a PVA concentration of 3 % in an acidic medium concentration of 3 %. The next step will be the optimization of the RF process using PVA as well as performance of this process under dynamic conditions.

PAGE 64

CHAPTER 4 OPTIMIZATION OF THE REACTIVE FLOTATION PROCESS AND PILOT SCALE TESTING Introduction In previous chapter, it was demonstrated that when PVA coated high MgO phosphate rock is immersed in a 3 % sulfuric acid solution, effectively all the high MgO particles will float to the surface leaving a sink phosphate rock phase that is a low in MgO. In this chapter, the optimization of separation process using PVA as well as development of a separation continuous process is discussed. In this chapter, the RF is optimized using statistical designs (i.e., screening statistical design and central composite design (CCD)). Also, the reduction of PVA consumption was studied using statistical design, (i.e., factorial design). The tests show that as little as less than one pound of PVA per ton of feed rock can be effective in floating the dolomite. The tests also show that there is no significant adverse effect of having the sulfuric acid solution saturated with calcium sulfate. Furthermore, pilot scale testing of a simple sluice was carried out and demonstrated the effectiveness of this approach in giving a good separation between the phosphate and dolomite particles. Optimization of the RF Process Screening Statistical Design The MgO % removal in floated fraction, Table 4-1, was used as the response in calculating the statistical effect of variables. The effect of variables are calculated as follows: 49

PAGE 65

50 Table 4-1. The MgO % removal in floated fraction for Plackett-Burman design. Runs Response (MgO % removal) 1 3 2 3 3 3 4 3 5 86 6 90 7 3 8 12 9 3 10 20 11 3 12 3 6ReRe)()( sponsesponseEi (4.1) The variance is estimated by evaluating the effect of dummy variables and using this equation: nEVdeff22 (4.2) where Ed = effect of dummy variable and n = number of dummy variables = 5. The statistical significance of the effects is evaluated using the student t-test as calculated using: 2iEt (4.3) where the degrees of freedom = 5, confidence level = 95 % and t 5,0.05 = 2.015. Comparing the absolute t-values to 2.015 led to acceptance of variables with a higher t-value and rejecting the variables with lower t-values. In other words, the variables accepted were significant within 95 % confidence level. This comparison of the Plackett-Burman design is presented in Table 4-2. The PVA molecular weight t-value is 2.761,

PAGE 66

51 which is higher than 2.015. Nevertheless, the higher PVA molecular weight was used because of its significance. From the Plackett-Burman design, the most significant variables are acid concentration, PVA concentration, and particle size. From the screening design, a model relating the MgO % removal to the different variables could be represented by: y = -91.75 + 14.83X 1 +12X 2 3.2X 3 (4.4) where y represents MgO % removal and X 1 X 2 and X 3 represent acid concentration, PVA concentration, and particle size, respectively. All coefficients are statistically significant close to 95 % confidence level. The negative sign for particle size coefficient indicates that increase the particle size can lead to a decrease in the MgO % removal. On the other hand, the positive acid concentration and PVA concentration values indicate that an increase in both concentrations lead to higher MgO % removal. In order to further investigate the interaction among these three significant variables with respect to their effect on MgO % removal, a central composite design focusing on these three variables with five levels was used. Table 4-2. Plackett-Burman calculated t values for different variables. The decisions to accept or reject the significance of variables on the bases of comparing t-value to critical value at 95 % confidence level. Factors Effect on MgO % removal (t-value) Significant Variable Sulfuric acid concentration 2.69697 Accept PVA concentration 2.181818 Accept PVA Molecular Weight (MW) 2.969697 Accept PVA degree of hydrolysis (D.H.) 0.181818 Reject Particle size 2.45455 Accept Drying time 0.333333 Reject Statistically significant (95 % confidence level) (+ ve) sign indicates more MgO % removal (ve) sign indicates less MgO % removal

PAGE 67

52 Central Composite Design Central composite design (CDD) results of dolomite flotation experiments are given in Figures 4-1 and 4-2 in terms of MgO removal % in floated fraction and MgO % in sink (nonfloat) fraction respectively. The confidence interval is within 95 %. In other words, the effects are statistically significant at the 95 % level. The standard deviation and R-Squared for both of them are 13, 0.9287 and 0.36, 0.9287 respectively. It was indicated, from these figures, that the % MgO recovery and MgO % in concentrate are greatly and positively affected by increasing either acid concentration or PVA dosage, (i.e., % MgO recovery increases from 30 to more than 90 % and MgO % in the concentrate decreases from more than 2.1 to 0.6 respectively). However, particle size negatively affects the % MgO recovery and MgO % in concentrate. With increasing particle size the % MgO recovery decreases. At the maximum acid concentration and PVA dosage, it decreases from 90 to 75 % and % MgO concentration in the concentrate increases from 0.6 to 1.1 %. As it can be seen from these results that the PVA consumption is high (about 4.5 lb/t), which is economically unfeasible, to get a concentrate contains < 1 % MgO with high recovery > 90 %. Therefore, further study is needed to reduce that consumption, which is discussed in next section. Research to Reduce the PVA Consumption From the results obtained from the experimental designs following the screening the main factors, then studying the main factors, unfortunately, the PVA consumption was about 4-6 lb/t (calculated according to used amount of PVA to certain weight of rock including the losses on the coating device), which threaten the success of this process economically. At this point, it was necessary to find a method to reduce the consumption.

PAGE 68

MgO Removal (floated) %Acid concentration, %PVA dosage, lb/t 1.0 1.5 2.0 2.5 3.0 2 3 4 5 6 74 94 44 MgO Removal (floated) %Acid concentration, %PVA dosage, lb/t 1.0 1.5 2.0 2.5 3.0 2 3 4 5 6 73 43 93 A B MgO Removal (floated) %Acid concentration, %PVA dosage, lb/t 1.0 1.5 2.0 2.5 3.0 2 3 4 5 6 34 74 54 14 C Figure 4-1. Effect of PVA dosage and acid conc. on MgO recovery % in floated fraction at different particle size A) Particle size = -0.5 mm B) Particle size = 1-2 mm C) Particle size = 5-9 mm 53

PAGE 69

54 MgO% in unfloated fractionAcid concentration, %PVA dosage, lb/t 1.0 1.5 2.0 2.5 3.0 2 3 4 5 6 0.9 1.3 1.7 2.1 0.5 MgO% in unfloated fractionAcid concentration, %PVA dosage, lb/t 1.0 1.5 2.0 2.5 3.0 2 3 4 5 6 0.9 1.3 1.7 2.1 0.5 A B MgO% in unfloated fractionAcid concentration, %PVA dosage, lb/t 1.0 1.5 2.0 2.5 3.0 2 3 4 5 6 1.3 1.7 2.1 0.9 2.5 2.9 C Figure 4-2. Effect of PVA dosage and acid concentration on MgO % in Unfloated fraction at different particle size A) Particle size = -0.5 mm B) Particle size = 1-2 mm C) Particle size = 5-9 mm

PAGE 70

55 In this regard many tests were conducted among these tests were: Using compatible compounds with PVA like starch and cellulose. It was found that the presence of these compounds negatively affected the stability of the bubbles at the interface. Using fine particles (40 mesh) with different PVA (1, 2, 3 %) and acid conc.(1, 2, 3 %). It was found that the 3 % PVA and 3 % acid were needed which is the same condition as before and in addition the reaction was too fast due to the high surface area of the small particles. PVA + Drimax (spreading agent): to spread the PVA on the particle surface and then reduce the amount needed to cover the particles. The results of these experiments indicates that there is an interaction between the PVA and Drimax in such a way that prevented the formation of stable bubble at the particle / solution interface, probably by interfering with the crosslinking of the polymer layers that is necessary for the formation of a membrane and consequently stable bubbles. PVA + Boric acid: as it is in the literature [Finch, 1973; 1992], the boric acid and borax are used in PVA crosslinking. Using the boric acid (5 % of PVA) helps in reducing the PVA consumption due to polymer crosslinking but we did not go further in these experiments because of toxicity of boric acid and environmental reasons. PVA + Na 2 SO 4 : sulfate ions lead to coagulation of the PVA according to literature [Finch, 1973; 1992]. It was found that the coagulation is not enough to float the dolomite particle at low PVA concentration. Conditioning of the rock particles with PVA and screening it (or filtering it) to recover the excess PVA. Actually, this step was a most promising one in which the consumption of PVA was reduced to 1 lb/t instead of 4-6 lb/t in previous coating procedure. The conditioning step (with about 1 % PVA) in the same manner as a regular flotation is followed by a filtration step and calculation of the actual PVA consumption. On the other hand, it was found that the full coverage of the particle with PVA is not needed since the formation of some bubbles may be enough to create the density difference and by using sluice or spiral in which the dolomite particles will be lighter and follow the liquid stream. In this manner, the dolomite particle can be separated from the phosphate.

PAGE 71

56 Experimental Design of the PVA Conditioning Step Due to the high PVA consumption as indicated from experimental designs, different methods were tested as mentioned before, the conditioning step as in regular flotation circuits was tested since this step can reduce the consumption to one-fourth or more (The PVA consumption was reduced to 1-2 lb/t from 4-6 lb/t). For that reason statistical design with three factors, (i.e., conditioning time, PVA concentration, and pulp density) was used. It is indicated, from Figure 4-3, that the PVA concentration is the main factor and then the conditioning time while the pulp density (solid liquid ratio) has a very slightly effect on the % MgO in the concentrate and % MgO recovery in floated fraction. Cube GraphMgO recovery in floated fraction, % A: Conditioning time, minB: S/L ratioC: PVA Conc., %A-A+B-B+C-C+5.5046.835.1355.108.8820.509.3084.40Cube GraphMgO% in Concentrate A: Conditioning time, minB: S/L ratioC: PVA Conc., %A-A+B-B+C-C+2.91.852.91.62.821.12.80.7 Figure 4-3. Cube graph for MgO % and MgO recovery % Saturation Experiment The CaCO 3 was used to saturate the solution with calcium sulfate by adding CaCO 3 to 3 % sulfuric-acid solution. The formation of the precipitate of CaSO 4 indicated that the solution was saturated. It was found that 1.78gm/l of CaCO 3 would cause the precipitate to form. After the formation of the CaSO 4 precipitate, the flotation test was run as an indication of the acid performance after its saturation. Table 4-3 shows the results of these experiments before and after the saturation is reached.

PAGE 72

57 Table 4-3. Effect of saturated acid on the RF Before saturation After saturation Wt% MgO% P 2 O 5 % Wt% MgO% P 2 O 5 % Float Sink 22.1 77.9 11.1 00.8 04.2 21.1 21.3 78.7 10.95 00.85 4.15 21.2 Pilot Scale Testing: Sluice Results It was shown that the PVA consumption could be reduced to 1 lb/t on lab scale. Testing of the reactive flotation process using a simple sluice was conducted on pilot scale. Starting from the point that the formation of the bubbles on the dolomite surface creates a density difference between dolomite and phosphate particles. Therefore, the gravity separators are the promising techniques to do this separation. In that regard, two tests were run using the sluice under two different coating conditions. The first was spraying the particles with 3 % PVA solution before adding the particles to the sluice and the second was conditioning the particles in 1 % PVA solution and screening the particles to remove the excess PVA solution to reuse it (the PVA consumption is estimated to be 0.8 lb/t), then adding the particles to the sluice. The results of these two tests are given in Table 4-4. From these results, it can be seen that by applying the optimum conditions, the reactive flotation process can reduce the MgO % to less than 1 % with high recovery. In the rest of this work, the fundamental interpretations of what are observed experimentally were carried out. Table 4-4. Comparison of two different coating methods and their effects on MgO removal Coating Conditioning Sample Wt% MgO% P 2 O 5 % CaO% I.R% Wt% MgO% P 2 O 5 % CaO% I.R% Float Sink 20.74 79.26 12.3 00.65 02.9 22.3 32.82 38.33 09.8 11.22 20 80 12.41 00.68 02.85 21.90 33.55 37.93 10.01 11.14 Calc. Feed 100 03.06 18.27 37.19 10.97 100 03.03 18.09 37.05 10.93 Feed 100 3.04 18.13 37.45 11.05 100 3.04 18.13 37.45 11.05

PAGE 73

CHAPTER 5 FUNDAMENTAL STUDIES Polyvinyl alcohol-Particle Interaction Introduction Generally, adsorption of a polymer is utilized in many applications, such as: dispersion of particles, flotation, flocculation processes, treatment of surfaces, etc. If a solvent is a poor solvent (i.e., the interaction energy between the polymer segment and the solvent molecules is unfavorable when compared to polymer-polymer or solvent solvent molecular interaction energy), then the effective polymer surface interaction can be strongly attractive and the polymer will seek any opportunity to minimize the contact points with the solvent. Hence, in cases where a poor solvent is used, adsorption will increase on almost any surface, even the liquid-air interface. Adsorption is, therefore, always accentuated before precipitation will occur. The surfaces will act as a nucleation sites for polymer precipitation. This means, the poorer the solvent, the higher the adsorption and vice versa. In addition to the solvent quality, there are other factors that can affect the polymer adsorption and change its conformation, including: polymer molecular weight, polymer concentration, pH, and ionic strength. Previous studies of polymer adsorption suggest a number of qualitative trends that have been noticed in most polymer adsorption systems. The most important of them are as follows: many polymers adsorb on a variety of surfaces, Reach a plateau at few mg/m 2 a sharp isotherm sometimes is obtained, and molecular weight is a main factor, especially in a poor solvent. 58

PAGE 74

59 In this work, the polymer (PVA) is used as a membrane-forming agent to capture produced CO 2 gas on dolomite particles and makes them lighter than other particles leading to their flotation and separation from non-carbonate (apatite) particle. To understand this process and to elucidate the interaction mechanism between polyvinyl alcohol (PVA) and dolomite and apatite surfaces in the reactive flotation process, different techniques were used, such as: estimation of work of adhesion using surface tension and contact angle measurements, adsorption/desorption isotherms, Zeta potential, and surfaces characterization using Fourier Transform InfraRed (FTIR), etc.; the details of each one of these techniques are discussed in this chapter. The strength of adhesion to a solid surface can be estimated from the value of the thermodynamic work of adhesion or W a a concept first introduced by [Harkins, 1928]. In a simple system where a liquid L adheres to a solid S, Figure 5-1, the work of adhesion is defined as: slslaW .. (5.1) where s l and sl are the interfacial tensions of the solid/air, liquid/air and solid/liquid interfaces respectively. The problem with using Equation 5.1 is that, of the three interfacial tensions, only l can be measured with fluid phases, in this case the liquid and air. Those tensions involving the solid cannot be measured independently. Another approach, then, is to combine Equation 1 with the Young-Dupre equation (Equation 5.2), to provide a more useful expression for the work of adhesion [Adamson and Gast, 1997]: coslsls .. (5.2)

PAGE 75

60 where is the contact angle that the liquid makes with the solid surface. Provided that l and can be measured experimentally, it is then possible to use Equation 5.3 to calculate the work of adhesion. )cos1( laW ......(5.3) Therefore, the measurements of surface tension and contact angle are essential towards the calculation of the work of adhesion of the polymer (PVA) with solid surfaces, which will be discussed in the following sections. Figure 5-1. Contact angle and interfacial surface tension at solid/liquid/gas interface urface Tension The equilibrium surface tension values of fully hydrolyzed PVA solutions (average w =150000) as a function of the initial bulk concentration at room temperature (23 C) re shown in Figure 5-3. The surface tension of PVA decreases with increasing oncentration of PVA until it reaches a region where the surface tension (59 60 yne/cm) no longer decreases with increasing the polymer concentration. The critical oncentration at which the surface tension starts to be constant is 1 % PVA in aqueous solution. These results coincide with the literature concerning the surface tension of fully hydrolyzed polyvinyl alcohol (PVA). The fully hydrolyzed polymer has less effect on Gas l Solid Liquid s sl S M a c d c

PAGE 76

61 surface tension than partly hydrolyzed PVA [Fbiehn and Erusberger, 1951; Kim aLuckham, 1991]. The chemical structure of both is shown in Figure 5-2. nd Figure 5-2. Structural formula for PVA: A) partially hydrolyzed; B) fully hydrolyzedIt is reported that the surface tension reaches a constant level at 1-3 % PVA depending on the PVA molecular w These results are similar to several findings reported in the literature regarding polymer systems; the stabilization of the surface tension of high-Mw PEO (> 100000) solutions was associated with the saturation of the air/water interface by the polymer chains [Zeno eight and degree of hydrolysis [Finch 1973; 1992]. et al., 2004]. 5055 6065Surface Teion, dyne/cm 7075012345PVA Concentration, %ns Figure 5-3. Equilibrium Surface tension for aqueous PVA solutions at 23 C A B

PAGE 77

62 An explanation of this phenomenon is that the surface activity of a polymer mainly relies on its conformation in the aqueous solution, whereas the conformation of macromolecule depends on its chemical structure and composition, etc. At low concentrations, the chains of macromolecule are stretched in the water due to the formation of hydrogen bonding with water molecules. These long chains lying on the liquid/air interface can dramatically decrease the surface tension of water, even though the adsorbed amount on the liquid/air interface is very small. With the increase of concentration, the polymers molecules that adsorb on the surface increase and a tight hen the polymer clusters; ay also in Figure 5-4. PVA crystal structure; the dotted lines indicate hydrogen bonds. macromolecular film is formed, which makes the surface tension decrease. W ers occupy all the surface of solution, the adsorption reaches a level of saturation corresponding to the lowest value of surface tension. As this concentration furthincreases, the polymers in the solution begin to aggregate and form micelle-like at this point, the surface tension shows no more changes. In the case of PVA, it mbe related to degree of crystallinity due to formation of hydrogen bonding as indicatedFigure 5-4 [Finch 1973]. OH OHOH CHCH 2 CH OHCH 2 CH CH 2 CH OHOHOH OH CH 2

PAGE 78

63 It is expected that the surface tension will have higher values in the presence ofthe acidic media since th e SO4-2 ions increases the surface tension, affects the polymer lubility, and enhances the formation of polymer aggregates at even a slightly high acid oncentration [Sun, 1994]. ontact Angle The phenomenon of wetting or non-wetting tter nderstood by studying cct ae, which is the most universally applied method of quantifying wettability. Contact angle methods have been developed extensively over the past four decades. The contact angles of dolomite and apatite surfaces with PVA can be to the surface by hydrogen bond. As the PVA reases, more and more hydrogen bonds form; as a result of hydrogen bondslt in a modified PVA end groups [Finch 1992]. so c C of a solid by a liquid is be u onta ngl aqueous solutions at different concentrations are shown in Figure 5-5. From the figure, it seen that the contact angle increases with the increase of PVA concentration for both minerals surfaces. This phenomenon can be explained by the structure of PVA in the solution shown in Figure 5-4. When a droplet of PVA placed on the surface of dolomite or apatite, the PVA adsorbs concentration inc among its molecules, the PVA begins to form clusters. These clusters resulocal viscosity increase, which controls the droplet shape as well as its tendency to spread. This, in turn, increases the contact angle accordingly. The dynamic behavior of the molecules, their conformation at solid-liquid interface, and diffusion of macromolecules will affect the adsorption of more molecules, with a subsequent increase in contact angle. The increase in contact angle is proposed toresult from polymer cluster formation and preferential coverage of the hydrophilic sites by polymer molecules. This phenomenon is observed for PVA and also for some

PAGE 79

64 3040 1020012345Co PVA concentration, %ntact angle, degree Dolomite Apatite Figure 5-5. Static contact angle variation with the PVA concentration All measurements shown in Figure 55 were conducted immediately after application of the drop. Evaporation of water caused the drop to contract and a consequent change in the contact angle. Work of Adhesion Figure 5-6 shows the dependence of work of adhesion, Wa, on the PVA concentration. The work of adhesion decreases with increasing PVA concentration (increase in G, where Wa = G [Lloyd, 1994]). The increase in G reflects the formation of polymer clusters and increase of its crystallinity. This is due to the formation of OH-group bonds between PVA molecules causing them to favor adhesion to a solid surface. Additionally, the polymers diffusivity decreases with higher With increasing PVA concentration, the PVA molecules not only adsorb to the surface but also form hydrogen bonds among themselves. This behavior affects the concentrations.

PAGE 80

65 conforation on stage 1-3 %, at which constant surface tension and ieved. rmation of the molecules at the solid-liquid interface till they reach the satustage, when the work of adhesion becomes constant. It appears that this saturatiis the same region of PVA concentration contact angle are ach 80PVA concentration, %W 100120140orkf Adhesin, dyne/c2 160012345 oom Dolomite Apatitesence of hydrogen bonding hypothesis and to establish an adsorption mechanism additional studies were conducted and their results are discussed in the paragraphs that follow. Adsorption study: Research grade samples were purchased from Wards Natural Science and used to conduct the adsorption study. These samples were chemically analyzed by the wet chemical method and the results are listed in Table 5-1. Figure 5-6. Work of adhesion as a function of PVA concentration on dolomite and apatite substrates Additional Studies To substantiate the pre

PAGE 81

66 Table 5-1. Chemical analysis of the samples used in adsorption tests MgO% P2O5% CaO% Al2O3% Fe2O3% I.R.% L.O.I. % Dolomite 20.39 0.58 31.6 0.97 0.10 2.52 42.6 Apatite 31.73 49.50 2.64 1.04 4.95 5.0 Specific surface area measurements (BET method): The specific surface area of dolomite and apatite particles was determined by the BET in order to be used in adsorption isotherm calculations (applied in the next section). Using nitrogen as the adsorbate, the measured surface areas for dolomite and apatite were 0.9 and 1.48 m2/g respectively. Pore size distribution: The results of cumulative pore size distribution for apatite er umulative pore area, 9 m/g, than the dolomite, 1.8 m2/g. Mean pore sizes for apatite and dolom and dolomite are shown in Figures 5-7 and 5-8. It is clear that the apatite has a high2 c ite are 28 A and 42.5Arespectively. o o 02 468Cumulative pore area, m 10000re der, A2igure 5-7. Cumulative pore size distribution for apatite sample /g 20 40 600 800 1000 Po iamet F

PAGE 82

67 00. 402004006008001000 0.81.21.6Cumulative pore area, m2/g 2Pore diameter, A Figure 5-8. Cumulative pore size distribution for dolomite sample Zeta Potential In order to obtain a more detailed description of the adsorption and electrical properties of the solidsolution interface, measurements of the zeta potential were taken. The tendency of zeta potential to vary with nonionic polymer adsorption may be caused by one of two different factors: 1) the shift of the position of the shear plane, or 2) the blocking of the active sites on the surface by adsorbing polymer chains. The participation of these effects was estimated by the determining the difference in the values of the zeta potential with no polymer and with polymer present. The use of a KNO3 solution (3X10-3 M) acted as a control in these experiments. olomite in 3X10-3 KNO3 solution as function of pH are presented in Figure 5-9. It can be seen that the iso-electric points (IEP) of apatite and dolomite are about 6.8 and 9, respectively. This is in agreement with the results reported in the literature [Amankonah and Somasundaran, 1985; Chander and Furestenau, 1982; Pugh and Stenius, 1985]. The results of measuring the zeta potential of apatite and d

PAGE 83

68 -60-50-40-30-20-100102030405002468101214pHZeta Potential, mV Dolomite Apatite3X10-3M KNO3 Figure 5-9. Zeta potential of dolomite and apatite samples lts of measuring the zeta potential in the presence of a PVA solution. It is evi Zeta potential in the presence of PVA was determined as well. Figures 5-10 and 511 show the resu dent that the presence of PVA does not change the IEP of the apatite or dolomite. It does, however, cause compression in the electrical double layer. -40 -20-10020405060Zeta Potential, mV -30pH 103002468101214 Dolomite Dolomite + PVA 3X10-3M KNO3 Figure 5-10. Zeta potential of dolomite in absence and presence of polymer

PAGE 84

69 -50-40-30-20-1001020304002468101214pHZeta Potential, mV Apatite Apatite + PVA3X10-3M KNO3 Figure 5-11. Zeta potential of dolomite in absence and presence of polymer sible to potential determining ions ( VA is not due to a decrease in charge and surface potential, but rather to a shift of the shear plane (Figure 5-12). er on the shear plane position, A) without polymer, B) after polymer adsorption The results indicate that the surface sites remain acces H + and OH ) and the IEP remains unchanged during PVA adsorption, thus implyingthat the polymer does not alter the surface potential. It can be considered, as other investigators asserted [Ottewill, 1967; Mpandou and Siffert, 1984;1987; Luckham et al., 1983] that the decrease in zeta potential relating to adsorption of P Figure 5-12. Effect of polym Distance from surface Shear p lane Stern p lane Zeta Potential Surface Potential Shear p lane Stern p lane Distance from surface Zeta potential before adsorptionZeta potential after adsorption Ions counter ions Polymer chain A B

PAGE 85

70 In short, the changes of the zeta potential produced by the adsorption of substances with high molecular weight, result from blocking of the active sites on the minerals surface and as well as from the shift of the shear plane. Zeta potential of minerals in supernatant solutions In natural ores, the dolomite and apatite form a mixture. Their presence together may affect the adsorption of the polymer on both of the minerals surfaces. Therefore, the adsorption on the dolomite in presence of apatite supernatant and apatite in presence of dolomite supernatant were studied. The effect of the supernatant of dolomite on the zeta potential of apatite is shown in Figure 5-13. The results show that the apatite surface becomes more positive over the entire pH range and that the IEP has shifted from 6.8 to 8.6. Both effects are hypothesized to be due to the presence of calcium and magnesium ions in the solution. -50-40-30-20-10 50 0 10 20304002468101214pHZeta Potential, mV Apatite A p atite in Dolomite3 e and apatite in dolomite supernatant 3X10-3M KNO Figure 5-13. Zeta potential of apatit

PAGE 86

71 SP shifts from 9 to 6.5. imilarly, the zeta potential of dolomite in an apatite supernatant was measured as a function of pH. The results, given in Figure 5-14, show the zeta potential of dolomite to be reduced drastically in presence of apatite supernatant. The IE 50 -30-20040101214Zeta Potential, mV 3X10-3M KNO3 -1010203002468 -50-40pH Dolomite Dolomite in Apatite supernatant Figure 5-14. Zeta potential of dolomite and dolomite in apatite supernatant Zeta potential in the presence of PVA Moreover, the zeta potential of dolomite in apatite supernatant in presence of PVA was determined. The results in Figure 5-15 indicate that the PVA deals with surface of dolomite, when suspended in an apatite supernatant, as an apatite surface. This is due to the faster diffusion of small ions and molecules from the solution to the dolomite surface, which converts it to surface similar to apatite. As can be expected, the movement and diffusion of these ions is quite faster than that of polymer molecules. Consequently, The polymers role is applied after the surface changing and the adsorption of the polymer appears as if it is on apatite surface. The same argument mentioned in the previous paragraph can be applied in the caseof PVA present with apatite in a dolomite supernatant. The results of the zeta potential of this system are shown in Figure 5-16.

PAGE 87

72 -60-50-40-30-20-1001020 3040 3X10-3M KNO3 02468101214pHZeta Potential, mV Dolomite in apatite supernatant + PVA Dolomite in Apatite supernatant Figure 5-15. Zeta potential of dolomite in apatite supernatant before and after PVA adsorption 50 40 -40-30-200102002468101214pH -1030Zeta Potential, mV Apatite in Dolomite supernatant + PVA Apatite in Dolomite supernatantFigure 5-16. Zeta potential of apatite in dolomite supernatant before and after PVA adsorption 3X10-3M KNO3 Infrared (FTIR) The FTIR spectra of the powders, both with and without polymer adsorbed to them, previous studies [Kiselev and Rygin, 1972; were compared in order to obtain the mechanism of adsorption of PVA on dolomite and apatite surfaces. This method was applied in

PAGE 88

73 Paradip and Moudgil, 1991; Howard, and McConnel, 1967; Behl and Moudgil, 1993; Bjelopavlic et al., 2000; Zaman et al., 2002; Mathur and Moudgil, 1997; Moudgil and Prakash, 1998] to explain polymersolid interactions. The major peaks for dolomite and apatite are identified according to Ince et al. [1991] and Farmer [1974]. FTIR spectroscopic studies were carried out on the dolomite and apatite and polymer samples both before and after adsorption. The assignments of the various bands and peaks made in this study are in reasonable agreement with those reported in the literature for similar functional groups [Finch, 1973; Chanchani, 1986; Ince, 1987]. The FTIR spectrum of PVA is shown in Figure 5-17. A broad-band, is displayed at e sharp band at 2930 cm-1 is attributed to strong C-H stretching. The band at 2860 cm-1 is attributed to shoulder C-H stretching. The transmission band at 1650 cm-1 is due to the presence of moisture in the sample. The band at 1460 cm-1 can be attributed to OH and C-H bending. These results are in good agreement with the spectra of PVA reported by other workers [Finch, 1973; 1992]. 3450 cm-1 which may be attributed to hydrogen-bonded surface hydroxyl groups. Th 40080012001600200024002800320036004000Wave number, cm-11110594Figure 5-17. Infrared transmittance spectra (FTIR) of poly vinyl alcohol (PVA) 3450293016501460862 2860

PAGE 89

74 The FTIR spectrum of the bare-dolomite sample as well as those that had previously interacted with PVA is shown in Figure 5-18. The isolated hydroxyl group peak is present in dolomite powder without polymer, but the spectra with polymer shows the disappearance of this peak. Moreover, the peak intensity of the hydrogen-bonded hydroxyl groups is smaller in case of the powder with the polymer than without. These observations indicate that the adsorption of PVA on dolomite occurs by hydrogen bonding involving OH groups of the polymer. This finding is in concurrence with earlier studies on silica-PEO [Rubio and Kitchener, 1976; Cheng, 1985] and silica-Polyacrylic acid PAA [Tadros, 1978], which showed that the presence of isolated OHgroups on the surface is responsible for the adsorption of polymer molecules by hydrogen bonding. 40080012001600200024002800320036004000Wave number, cm-1 Dolomite + PVA Dolomite8831460 72818202550264029103030345036407288831460182025502640291030303450 Figure 5-18. Infrared absorbance spectra (FTIR) of A) dolomite alone, and B) dolomite/PVA system. A B

PAGE 90

75 The FTIR spectrum of the apatite sample after having interacted with PVA is shown in Figure 5-19. However, unlike the case of dolomite, this figure indicates that the adsorption mechanism of the polymer on apatite is not clarified by using FTIRno change is shown in the bands intensities or positions, either for bare apatite or apatite with polymer adsorbed on it. This finding is in agreement with the current literature [Chanchani, 1986; Ince, 1987]. It is pertinent to mention that FTIR spectroscopic studies on the dolomitePVA and apatite-PVA systems revealed the presence of hydrogen bonding only. 40080012001600200024002800320036004 000 Wave number, cm-1 Apatite Apatite + PVA866146016503450 Figure 5-19. Infrared absorbance spectra of A) apatite alone, and B) apatite/PVA system. Adsorption Study An adsorption study was used for the purpose of determining PVA adsorption on termined colometrically. Figure 5-20 shows the calibration curve and its equation, which was used in this study to calculate the residual concentration of PVA in adsorption experiments. 6071050 A B both dolomite and apatite. The calibration curve for PVA concentration was de

PAGE 91

76 The l imitation of the PVA concentration to create this curve is 40 ppm [Finch, 1973;1992]. y = 0.0036x + 0.0030.020.040.060.080.10.120.140.16Absorbance 001020304050PVA Concentration, ppm Figure 5-20. Calibration curve for PVA concentration measurements Kinetics of adsorption A study of the kinetics of adsorption is mainly to determine the equilibrium time of adsorption. The rate by which equilibrium is established during adsorption depends on the chee polymer and its molecular weight, the solvent, and the type of adsorbent. In most cases, the kinetic curves indicate an increase in the amount of adsorbed polymer over time, with an asymptotic approximation of the equilibrium value of a polymer on solid surfaces occurs by a three-step mechanism [Dijit. mical nature of th [Lipatov and Sergeeva, 1974] The adsorption et al., 1990; Somasundaran and Ramachandran, 1988]. First, the polymer is diffused from the bulk to the interface. Second, adsorption of the polymer molecules occurs on the initially bare solid surface. Finally, polymer conformation and orientation changes in the surface layer to accommodate the adsorption of more polymer molecules

PAGE 92

77 To ascertain the time required for the adsorption equilibrium to be attained, kinetests we tic re carried out at a natural pH of 7 for a period of 24 hours; PVA concentration was 100 ppm (6.7X10-7 M). Figure 5-21 shows results on the adsorption of PVA onto dolomite and apatite as a function of time. It is evident that the equilibrium is attained at about one hour. The equilibrium values are 0.04X10-6 and 0.011X10-6 mole/m2 for dolomite and apatite respectively. In consideration of these findings, a 2-hour conditioning time was selected for the adsorption studies. It was observed that the adsorption followed a typical second-order adsorption process. The second-order model predicts the behavior over the whole range of adsorption and is in agreement with an adsorption mechanism being the rate-controlling step. The second-order rate expression [Ozacar, 2003] based on solid capacity is generally expressed as follows: 2)( d eqkOr, by integration dt... (5.4). eqeqtk1..(5.5) Where is adsorption density (mole/m) and t22sorption rate constant (hr-1), and t is time (hr). By fitting the exe eq X 10 mole/mK, hr eq is equilibrium adsorption density (mole/m 2 ), K is second-order ad perimental data to Equation 5.5, the parameters eq and K were determined. Thesare listed in Table 5-2. Table 5-2. Fitting parameters of second-order kinetic model -62-1 Dolomite 0.04 361.6 Apatite 0.011 200

PAGE 93

78 It can be seen that the eq values for dolomite and apatite match the experimental values. The higher eq value of dolomite indicates the presence of more sites available fPVA adsorption than on apatite. Moreover, the K or ad for dolomite is higher than the K ad for apatite, which indicates the higher adsorption rate on dolomite. However, the plateau was reached almost at the same time for both minerals. 0.E+00 5.E-02 1.E-024.E-02PVA Adsortion density, 10-6mole/m 2.E-023.E-020152025Time, hr2 Dolomite Apatite Dolomite-2nd order model Apatite2nd order model 510 Figure 5-21. Kinomite and Apatite (Initial PVA conX10 -3 M KNO3 and 23C). Explanation of the higher adsorption rate on dolomite: The higher adsorption ll either be flat on the solid surfacegreater tendency for ttach to each other, (2) The neutrality of apatite (pH 7) while having an IEP of 6.8, whns less ions in the lution are available to act as additional driving force for PVA to adsorb on the apatite etics of PVA adsorption on dolcentration 6.7X10 -7 M, pH 7, 2 rate on dolomite may be explained by: (1) The higher solubility of dolomite and production of more ions in the solution, which affects the solutions property as a solvent(effectively converting it to poorer solvent). This property also affects the polymers conformation, which wi molecules of polymer to be adsorbed on the solid surface or a ich mea so

PAGE 94

79 surfacn ed with FTIR spectra measurement. Adsorption isotherm Most isotherms of the adsorption of polymers from dilute solutions can be visually represented as curves that reach saturation at specific concentrations [Lipatov and Sergeeva, 1974]. It is important to note that polymers may adsorb on surfaces in various conformations depending on bulk (solvent) and surface chemical properties of the substrate, as well as other physical conditions of the system [Hong et al., 2001]. At low concentrations, the polymer chain behaves like isolated coil. If the concentration of the polymer in the solution is raised, the chains become close to each influence the motion e point of chain-overlapping concentration, which is thill is e. This is especially significant as there is high competition between the interactioof PVA with water molecules and interaction of PVA with the apatite surface, and (3) OH-groups on dolomite surface (isolated and hydrogen-bonded hydroxyls) are greater than those on apatite surface (hydrogen-bonded hydroxyls). In other words, the isolated OH is exclusive to the dolomite surface, which is confirm other, which allows polymer-polymer intermolecular interactions to of the polymer chains. The effects may be expected to be cumulative until th e close packing of the polymer coils in solution. Once the overlapping concentration has been reached, the polymer motion wbe dominated by the presence of direct polymer-polymer interaction. As polymer concentration is increased, the polymer diffusivity decreases. If the concentration increases even further, a pseudo-matrix-gel will be formed. The resulting viscositygoverned mainly by intermolecular interaction, which is related to polymer chains overlap [Izmailova et al., 1975; Gotze and Sonntag, 1987; Andreyeva et al, 1973; Babayevskii et al., 1986; Koopal et al., 1988].

PAGE 95

80 For PVA, OH is the only functional group [Koopal and Lyklema, 1979; Chibowski and Paszkiewicz, 1995; Chibowski et al., 2000]. Increasing the degree of hydrolysis reduces acetate groups, which usually show preferential adsorption. Thus, thconformation of the polymer chain at the solidpolymer solution interface will depend onthe number of acetate groups in a PVA macromolecule. The lower the degree ofhydrolysis, the fl e atter the polymer chains; conversely, at high degree of hydrolysis, the be oriented perpendicularly to the surface [Santhiya et al., 1999; Sjbeon the with ental resultlly -6 ite d polymer chains should rg et al., 1999; Bajpai and Vishwakarma, 2003; Koopal and Lyklema, 1979]. This conformation of the PVA macromolecule will favor the adsorption of the polymer surfaces of dolomite and apatite, which is shown in Figures 5-22 and 5-23. Chibowski [2000] reported that the adsorption of PVA on alumina increases an increase in the degree of hydrolysis. This is also confirmed by previous experim s and polymer screening experiments. In this work, it was found that the fuhydrolysed PVA was the best among different degrees of hydrolysis PVAs (Chapter 3 in this thesis). The adsorption density (mol/m2) of fully hydrolyzed PVA (acetate groups < 1 %) on the minerals surface as a function of the initial polymer concentration (0 to 4X10mole) was determined at 23 Celsius and a pH 7 for dolomite and apatite. The results obtained are presented in an adsorption isotherm plot (Figures 5-22 and 5-23 for dolomand apatite, respectively). The analysis of these adsorption isotherms shows formation of a plateau for both apatite and dolomite. Such behavior of the macromolecules comes from a different adsorption affinity towards different surfaces. This result suggests that the observe

PAGE 96

81 increahe s his, in r dsorption isothe se in PVA adsorption is mainly caused by the change in conformation of tpolymer chain at the minerals surface. Following this change, a number of active siteon the surface become available for further adsorption of polymer molecules. As mentioned previously, the greater porosity of apatite leads to high number of adsorption sites available and the horizontal stretching of the polymer to cover these sites. Tturn, explains the lower adsorption layer thickness and higher parking area of apatite. The shape of the adsorption isotherms shown in Figures 5-22 and 5-23 is typical fophysical adsorption and is an acceptable shape for that range of polymer molecular weight. [Adamson and Gast, 1997]. The measurements were repeated three times and error bars were plotted on each graph. From the figures we can see that the a rm is S-shaped, which suggests a physical adsorption of PVA onto the dolomite andapatite surfaces. 0.00.51.0Equilibrium PVA conc., 10-7 moleAdsotion nsity,Figure 5-22. Adsorption iso 1.52.00.00.40.81.21.62.0rpde X 10-7 mo2 therm of PVA on dolomite at 23C 2.5le/mDolomite

PAGE 97

82 0.0Equilibrium PVA Conc., 10-7moleFigure 5-23. Adsorption isotherm of PVA on apatite at 23C Calculation of PVA adsorption and parking area The amount of PVA adsorption was calculated using the following formula: 0.51.52.0Adsorption density, 10 mole/m2 1.00.00.40.81.21.62.0-7Apatite SA* m V*) C-(Ci r (5.6) Where, is the amount of PVA adsorption (mole/m2), (Ci-Cr) is PVA concentration on the particle (mole/ml), Ci is initial PVA concentration, Cr is residual PVA concentration, V is the volume of solution (ml), m is the mass of dolomite or apatite particles (g), and SA is the specific surface area of particles (dolomite 0.9 and apatite 1.48 m2/g). The parking area of PVA molecules was calculated using the following formula: Parking area = Av 10118Mw Where MW = 150,000 g/mole, Av is Avogadros number (6.023 x 1023 #/mol) and parking area is nm2/#. The calculated parking areas for dolomite and apatite at the first plateau are (23.72 nm2) and (75.47 nm2) respectively. Taking into account that the (5.7)

PAGE 98

83 Dolomite / Apatite Dolomite / Apatite Dolomite / Apatite Dolomite / Apatite Dolomite / Apatite Dolomite / Apatite A B parking area per stretched molecule is 0.2 nm2 [Koopal and Lyklema, 1979], it was observed that PVA-molecule forms a wavy or zigzag shape on the solid particles with the higher thickness on the dolomite particles. Moreover, this conformation on dolomite appears at lower relative concentration range. It is worth noting that in the case of dolomite the adsorption isotherm is both steeper and reaches a plateau at lower concentrations than in case of apatite. The observed results may be explained based on the following hypothesis, which is Schematically illustrated in Figure 5-24: macromolecules are generally more tightly e figure shows the r several chains (5-24B). It may be conclfavorable and with inter Figure 5-24. Polymer-conformation changes with increase the concentration: A) molecule. bound to surfaces due to the multiple points of attachment. Th conformation of single polymer chain (5-24A) and/ o uded that without intermolecular H-bonds, spiral conformations of PVA are molecular H-bonds, zigzag conformations is expected. conformation of a single molecule, B) conformation of more than one

PAGE 99

84 Adsorbed layer thickness The thickness of PVA layer adsorbed on dolomite and/or apatite can be calculated from the adsorption isotherm by using the following relationship: h2 (5.8) Where, h is the adsorbed layer thickness, m; is adsorption density, g/m, and is PVA density, g/m3. Using this relationship, the thickness of the adsorbed layer was found to be 27 nm on dolomite and 9 nm on apatite, which indicates that the vertical polymer chains are stretched three times longer in the case of the dolomite than in the case of the apatite. This explains the greater adsorption on dolomite; since, according to their conformation, the polymer chains occupy smaller number of sites. In addition, more sites are available on dolomite than on apatite. Note: The length of the PVA molecule of this molecular weight is about 600 nm, hich indicates that the polymer conformation would be like Figure 24B. EffIonic sites at the surface generally control the surface charges, on salt-type minerals such as apatite and dolomite. Specifically, for dolomite are: Mg++, Ca++, CO3--, HCO3-, OH-++, Mg(OH)2, CaHCO++2 and for apatite are: a++, HPO4--, CaOH+, Ca(OH)2, H2PO4-, CaH2PO4+. The nature of the mineral surface, adsorption behavior. Previous studies on different minerals (i.e., alumina, silica, kaolin, clays) showed that the OH group of PVA is mainly responsible for adsorption on the w ects of pH MgHCO 3 MgOH 3 CaOH, Ca(OH) C pH, the presence of solutes, and the functional group of the polymer all affect the

PAGE 100

85 surface of a mineral [Santhiy a et al., 1999; Sjberg et al., 1999; Bajpai and Vishwakarma, 2003]studied range of 2 to 12. The effect of pH on the adsorption of PVA onto dolomite and apatite is shown in Figure 5-25. It can be seen that adsorption is pH-dependent in the 0.0E+001.0E-022.0E-023.0E-024.0E-02135791113Adsorption density, 10 5.0E-026.0E-02 mole/m pH-62 Apatite Dolomite Figure 5-25. Effect of pH on adsoStudying the adsorption of PVA on both minerals gives evidence that, due to fluctuations in pH, the adsorption is due to formation of hydrogen bonding between the tite surface from one side and OHpolymer functional group from anoth It is or rption of PVA on apatite and dolomite dolomite or apa er side. Effects of pH on Apatite: It is evident from Figure 5-25 that the adsorption density decreases with an increase of pH. The decrease in the adsorption density as a function of pH may be explained by taking into consideration the surface charge of the apatite particles as well as the surface species, both of which are dependent on pH.interesting to note that the decrease of CaH 2 PO 4 + and H 2 PO 4 species, according to the apatite activity chart [Attia and Fuerstenau, 1989] in entire pH range, are responsible f

PAGE 101

86 formation of hydrogen bonding. In addition, there is a presence of about 5 % silica in this sample, which also leads to decrease in adsorption with increasing pH. Moreover, it has been reported, as in the case of silica, that adsorption behaves thesame way as pH due to a reduction in the number of undissociated silanol groups [Rachas et al., 2000]. On the other hand, the reaction of Ca+2 or Mg+2 ions with four OH groups occurs at pH > 7 and is accompanied by compacting of the macromolecules. The degree of crystallization of the film sharply decreased (evidently because of water) after formation of a calcium or magnesium complex. ption ding capacity. It was noticed that the increase of OHspecies and formathis CH2 CH2 H2C O H O H OH2OH2 M Where M is Ca +2 or Mg +2 ions Effects of pH on dolomite: It can also be seen in Figure 5-25, that the adsordensity of dolomite increases with an increase of pH. This increase of adsorption with pH can be attributed to the hydroxylation of dolomites surface, which enhances itshydrogen-bon tion of metal (Ca and Mg) hydroxides with increasing pH are responsible for increase in adsorption. Additionally, it has been reported in some cases that, at acidic pH, the decrease in PVA adsorption, can be caused by some or all of the following factors: loss of the most active binding sites for PVA bonding, reduction of the affinity for nonionic compounds by a salting-out effect, the screening effect of hydrated counterions, or preferential adsorption of water molecules at those ionized sites [Barker and

PAGE 102

87 Garvey, 1980; Bonekamp et al., 1989; Platonov et al., 1979; Zhang et al, 1996; Hidber al., 1995; Santhiya et al, 1998]. The adsorption of PVA on both minerals giv et es evidence that th supernatants ensity lymer aggregation. The increase in dolomite, on e other hand, is due to the presence of more sites, as mentioned previously. After aching certain concentrations, both a curve become parallel, which means the same mount of the polymer has adsorbed on both surfaces. e adsorption is due to formation of hydrogen bonding between the dolomite or apatite surface from one side and OHpolymer functional group from another side. Adsorption in The results of adsorption of PVA on apatite in dolomite supernatant solution and dolomite in apatite supernatant solution are given in Figure 5-26. The adsorption dis lower than when the medium was just water, possibly due to the presence of several ionic species, which may affect the po th re a 01 5 23400.050.10.150.20.25-6Adsorption density, 109 mole Equilibrium PVA Conc., 10 moleApatite in dolomite supernatant Dolomite in apatite supernatant Figure 5-26. Adsorption isotherm of apatite in dolomite supernatant solution, and dolomite in apatite supernatant solution This can be attributed to adsorption of the newly introduced polymer to the polymer already adsorbed to the surface and the ensuing formation of an aggregate at 3*10-3 M KNO 3 the solid/liquid interface.

PAGE 103

88 Desorption To ascertain whether the adsorption process is reversible, desorption tests were carried out by interacting the sample containing the adsorbed polymer with distilled wfor a specified period and analyzing the supernatant solution for residual polymer ater in Figures 5-27 and 5-28. From Figures 5-27 and 5-28 it is evif content. The results are shown dent that, in the case of PVA, desorption characteristics follow a trend in reverse othe adsorption behavior. However, only about 20 % of the polymer could be desorbed, confirming the relatively weaker nature of this particular interaction with dolomite and apatite. 00.040.08 0.12-6/m 0.1600.050.10.150.2Equilibrium PVA Conc., 10-6 molePVA adsorption density, 10 mole2 Adsorption Desorption Figure 5-27. Desorption of PVA from apatite surface

PAGE 104

89 00.05 0.10.150.25A Asorpion dnsit 10-6 molem2 0.200.050.10.150.2Equilibrium PVA Conc., 10-6 molePVdtey,/ Adsorption Desorption Figure 5-28. Desorption of PVA from dolomite surface Desorption in Acidic Solution Since the RF process requires adding PVA coated particles to an acidic media, desorption of PVA in an acidic solution was studied to simulate the RF process. Desorption tests show almost no desorption in both the cases of dolomite and apatite. This behavior confirms that the formation of polymer cluster at the particle/ solution interface is due to the presence of SO4-2 ions, which is one of the most ionic species leading to cross-linking of the polymer chains [Finch, 1992; Garba et al., 1996]. Consequently, this increases the viscosity of the polymer-coated layer in addition to strengthening the adhesion forces among polymer molecules thus affecting its viscoelastic properties, its hydrodynamic volume, and its stretching when CO2 gas is produced as a result of the acid-carbonate reaction. Moreover, it is mentioned in the literature that even H2SO4, a strong acid, does not lead to salting out of the PVA; rather, its effect is extremely weak [Finch 1973; 1992].

PAGE 105

90 Bubble Formation and Stability The reactive flotation (RF) process, as mentioned earlier, includes two main steps: (1) the coating step, which is discussed in detail in the previous chapter and (2) the reaction step. In the second step, the sulfuric acid penetrates the polyvinyl alcohol (PVA) membrane and reacts with the dolomite surface. CO2 gas is one of the reaction products according to the following reaction: Ca Mg (CO3)2 +2H2SO4 CaSO4+MgSO4 +2CO2+2H2OBecause PVA is known for its very low permeability by gases (Finch, 1992), especially CO2, the CO2 will accumulate between the particle and membrane creating a density difference between the dolomite and phosphate rocks (i.e., lowering the dolomite tration of the mechanism of bubble formation during the reof the acid f ts to predict the membrane elasticity at particle density). A schematic illus action and how the particle density changes is given in Figure 5-29. Figure 5-29. The bubble formation mechanism To predict the required amount of CO 2 to float the particle, the modeling reaction with the dolomite surface and bubble formation was essential. In this chapter,different modeling efforts are discussed and the CO 2 production rate and the amount oCO 2 required to float the particle are calculated. Furthermore, the results are correlated with dynamic surface tension (DST) measuremen Acid diffusion Acid diffusion CO2 Dolomite Dolomite Dolomite CO 2BubbleCO2 Dolomite Dolomite Dolomite CO 2BubbleCO2 Dolomite Dolomite Dolomite CO 2Bubble

PAGE 106

91 a certain CO2 flowrate. Figu of the requirements to achie re 5-30 shows a schematic diagram ve flotation by RF process. CO2 re quired to float the particle Dolomite particle density changes Membrane elasticity CO2 production rate Dissolved Species (Ca and/or Mg) Dolomite-Acid reaction Figure 5-30. Schematic diagram of reaction kinetics of the RF Chemical Analysis of Samples Used in Kinetic Study: The experiments were carried out with two particle sizes (0.2 0.3, and 1-2 mm). The chemical analyses of these two fractions are given in Table 5-3. able 5-3. Chemical analysis of the samples used in kinetic tests Size, mm MgO% P2O5% CaO% Al2O3% Fe2O3% I.R.% L.O.I. % T -2+1 14.48 2.8232.61.021.5611.2 32.12 -0.3+0.2 15.36 2.45 31.47 34.37 0.99 1.34 10.95 It can be seen from the Table 5-3 that the analysis of the two samples are approximately the same. It is expected that the very small differences, as shown in the tic tests results. Effec. analysis, will have a negligible effect on the kine t of Particle Size Different particle sizes were tested in 3 % sulfuric acid, H 2 SO 4 solution (0.33 M)The results of reacting uncoated particles with acid at room temperature (23 1C), are shown in Figures 5-31 and 5-32 in terms of Ca and Mg dissolved, respectively. In all of the experiments, the solid loading in the reactor was 10 %.

PAGE 107

92 In the course of each dissolution process, concentrations of both the calcium and magnesium ions were determined as a function of time following the reaction between the dolomite and the sulfuric acid. The dissolution process of dolomite in aqueous sulfuric acid was found to depend on the solution acidity, as indicated in statistical design study (Chapter 4). It was also indicated in the literature that the reaction of calcite with sulfuric acid follows first-order kinetics and it depends only on acid concentration at pH< 5 [Booth et al, 1997]. Generally, calcium and magnesium ion concentration in the liquid phase first nderwent an exponential increase and then slowly approached an asymptotic value. rapidly due to the higher surface area available for H ions after which the dissolution reached a plateau ace area and did not reach equilibrium even u For the small particles, 0.2-0.3 mm, the reaction began very +(equilibrium) after 30 minutes. In the case of large particles, 1-2 mm, the reaction increased gradually as a result of lower surf after 2 hours. 00.0050.01 0.025 0.0150.02Ca dissolved, M Large particles 1-2 mm 020406080100120140Time, min Small particles 0.2-0.3 mm % sulfuric acid solution at 23C. Figure 5-31. Dissolution kinetic of different particle sizes in terms of calcium ions in 3

PAGE 108

93 00.0050.01 0.015Dissolved Mg, M 0.020.025020406080100120140Time, min Small particle (0.2-0.3 mm) Large particle (1-2 mm) Figurns in 3 % sulfuric acid solution at 23C. Formation of an insulating layer: The H+ transport is perturbed by the presence of the reaction products and since the H+ has a higher diffusivity, the reaction continued until it reached the point of equilibrium. At this point, the H+ can no longer reach the surface due to the formation of an insulating layer of reaction products, most probably gypsum, CaSO.2HO and/or a CO2 gas layer around the particle. This was especially true in the case of coated particles. To confirm the formation of a passivation layer of gypsum, a picture of a particle both before and after the reaction was taken (Figure 5-33). The variation of the particle color from yellowish to white is an indication of the formation of the gypsum layer at the surface of the particle. Figure 5-33. Dolomite particle (1-2 mm) before and after the reaction with 3 % H2SO4 e 5-32. Dissolution kinetic of different particle sizes in terms of magnesium io 4 2 After reaction Before reaction

PAGE 109

94 In the rest of this chapter, the Mg concentration will be taken as the dissolution kinetic indicator since both the calcium and magnesium follow the same trend. Effect of PVA Coating The reaction was tested with the particles having been coated with PVA. Figure 5-34 shows the results of both coated and uncoated particles. The curve is steeper in the case of uncoated particles and the curves for both coated and uncoated reach a plateau after 30 and 40 minutes respectively (Figure 5-34). This can be explained by the and then to the bulk solution. In addition, formation of the products occurs in a confined place (between the membrane and the particle surface), which also leads to a decrease in the saturation point and helps in rapid nucleation and growth of hydrated calcium sulfate, 22423ion of CO2 works as a diffusion barrier as well. presence of a polymer, which controls the transfer of ions to and from the boundary layer gypsum, (CaSO.2HO). Such a possibility exists in the case of a calcite-sulfuric acid system, providing that the solubility products are 2.5 x10-5 M2 for (CaSO.2HO) [Culberson et al., 1978] and 4.8 x 10-9 M2 for (CaCO) [Stark and Wallace, 1982]. Moreover, the accumulat 00.0050.01 0.0150.020.025 080100120140Time, minMg Dissolved, M uncoated coated 204060 igure 5-34. Dissolution kinetics of coated and uncoated 0.2-0.3 mm dolomite particle in F 3 % sulfuric acid solution at 23C.

PAGE 110

95 Bubble formation and growth: Figure 5-35 (a-e) shows the growth of bubbleon the dolomite particle surface with time. The reaction starts immediately after s adding solution. The bubbles then grow as a result of the evolution of CO2, id the particles to the acidic the impermeability of the PVA membrane, and the stretching of that membrane. If this membrane is not flexible enough, it will rupture and CO 2 gas will be released, leaving the particle without causing any flotation. It is also interesting to notice the rapgrowth of bubbles within a few seconds. 2s 5s 10s 15s 20s Figure 5-35. Growth of the bubbles with time on the dolomite particle surface, 50 X

PAGE 111

96 A echanism(s) controlling the dissolution process. The two mechanismge of acid and b) depletion of the active sites. Moreo empirical models are given fitted to experimental data. Those empirical models depend on reaching the saturation concentration in the bulk solution and the depletion of acid as the reaction products are formed. The main purpose of these models is to predict the amount of CO2 evolved and the consequent density changes of the dolomite particles, which leads to its separation. The details of each model are explained in the following paragraphs. Blockage of the Acid Model In this model, the reaction depends on the transfer of the acid to the particle surface; the products of the reaction then diffuse out to the bulk solution. The formation of the reaction products, (i.e., Ca-sulfate and/or CO2 gas), prevents the acids accessibility to the surface and stops the reaction. For simplicity of calculations, the sphericity of the particle, homogeneity of the coating around the particle, and presence of a barrier to CO2 gas transfer are assumed. lso, Figure 5-35 illustrates that presence of certain sites that can be considered asactive sites for reaction to form CO 2 bubbles. In other words, if the entire surface is reacting, the polymer will form a uniform and homogeneous membrane around the particle. Modeling of Bubble Formation and Dolomite-Particle Separation In the present study, two models were developed to explain the phenomena associated with the RF process. These models depend on the m s studied are a) blocka over, tw and The equations of this model are: )(momimairmiCCAKACKdt dC (5.9)

PAGE 112

97 airaiACKdt)(aiaoaCCAK dC (5.10) )(momimmoCCAK dtdC (5.11) aiAC (5.12) mdtdK aiaACdt (5.13) Where: C dKCmo : concentration of Mg in the bulk, M -2-1Ka : transfer coefficient, cm.min mi : concentration of Mg at the surface, M C ai : concentration of acid at the surface, M A : surface area, cm 2 K r : reaction rate coefficient, cm.min-2-1K m : transfer coefficient, cm -2 .min -1 and : coefficients, M -1 cm -4 min -2 : ratio between the boundary layer volume and total solution volume (Vi/Vo)37 for Large Small Fitting of this model to the experimental data is given in Figures 5-36 and 5-both size fractions. The fitting parameters are given in Table 5-4. Table 5-4. Acid-blockage model parameters Uncoated Coated Uncoated Coated *P-1 3 3 8.99 8.99 0.37 0.46 8.3 8.32 *10.1 3.29 9.27 9.26 0.67 0.67 9 9 0.00038 0.00085 0.0012 0.002 K m 5.97 5.97 10.04 K a 3.29 P-1 = Kr A Km= Km A Ka= Ka A

PAGE 113

98 0 0.0050.010.0150.020.02580100120140160Time, minMg Dissolved, M 0204060 Data-uncoated Model-uncoated Model-coated Data-coated1-2 mm experimental data Figure 5-36. Acid blockage model fitting to 00.005 0.025 0.02 0.015 0.01 001140Time, minMg Dissolved, M Data-uncoated Model-uncoat ed Model-coated 2040 6 80 00120 160 Data-coated ure 5-37. Acid model fi.25-0.35 rimental This model shows no difference in all its factors between coated and uncoated particlee can be explained by the swelling of the membrane due to the formation of CO2 gas, which Fig block age tting to 0 mm e xpe data s except which increases in the case of coated particles. The increas

PAGE 114

99 increases the internal boundary layer volume at constant external volume. The increase in P-1 for small particles is due to the increase in the number of sites that the acid can react with. Disappearance of Active Sites Model In this model, the reaction depends on the availability of the active sites at the particle surface. The depletion and disappearance of these sites will lead to cessation of the reaction. The disappearance of active sites may be due to formation of a passive layer from the reaction products, (i.e., Ca-sulfate and/or CO2 gas). In this way the acid is available to react, but there are no active sites to react with. arrier, constancy of acid concentration, homogeneity of the coating around the particle, and presence of a barrier to CO2 gas transfer are assumed. For simplicity of calculations, the sphericity of the particle, absence of an acid b )(mimomaormiCCAKACKdtdC (5.14) )(momimmoCCAKdtdC (5.15) aorACKdtdA (5.16) Where: Cmi : concentration of Mg at the surface, M Cao : concentration of acid in the bulk, M (assuming Ci=Co) A : surface area, cm2 concentration of Mg at the surface, M Km : t-2-1Kr : reaction rate coefficient, cm.min C mo : concentration in the bulk, M ransfer coefficient, cm.min-2-1 : ratio of volume of boundary layer to the volume of the bulk (Vi/Vo)

PAGE 115

100 Fitting of this model to the experimental data is given in Figures 5-38 and 5-39 for both size fractions. The fitting parameters are given in Table 5-5. 00.0050.01020406080100120Mg Dissolved, M 0.0150.020.025140160 Data-uncoa ted Model-uncoated Model-coat Time, min ed Data-coatedFigure 5-38. Active sites disappearance model fitting to 1-2 mm data 00.0050.010.0150.02 0.025 020406080100120140160Time, minMg Dissolved, M Data-uncoated Model-uncoated Model-coated Data-coated l fitting to 0.25-0.35 mm data Figure 5-39. Active sites disappearance mode Table 5-5. Active sites disappearance parameters Large Small Uncoated Coated Uncoated Coated 3 8.99 8.99 P-1 3 0.00038 0.00085 0.0012 0.002 7 10.04 10.1 9.26 K m 5.97 5.9 Acini 3.29 3.29 9.27 P-1 = KA K= KA rmm

PAGE 116

101 This model shows no difference in P-1, K m and K a between coated and uncoated particles. The only difference is in the value of which increases in the case of coated particles. The reason behind this increase is the constant external volume and increase of the internal volume since the polymer does not permeate the CO2. The increase in P-1 in for small particles is due to the increase in the number of sites that acid can react with. Empirical Models There are many models that can fit empirically to experimental data. The Two models chosen to fit the experimental data in this study were widely used in the literature for calcite dissolution [Sjoberg and Rickard, 1984; Sjoberg, 1978; Lund et al., 1973]. ndent model. However, these models do not reflect the physics associated with RF process First-order kinetic model For the phase of the study involving treatment with Sulfuric acid, a kinetic model [Cameselle et al., 1997; Froment and Bischo, 1979; Blancarte et al., 1987] was employed, the equation of which would be: These models are the First Order kinetic and Acid Concentration Depe )(CCeqkdtdc (5.17) Where: C : dissolved Mg concentration, M ; : treatment timBy integrating Equation 5.17, the value of the kinetic constant can be deduced: q (5.1 k : kinetic coefficient, min -1 ; C eq : Equilibrium concentration M t e, min. )1(kte CeC 8)

PAGE 117

102 Figures 5-40 and 5-41 show the fitting to this model to the results from experimenwith 3 % (0.33 M) sulfuric acid for two different particle sizes. The reaction rate constants are given in Table 5-6. The reaction rate constant, K, is 2.5X10 ts m reaching the surface and/or covering the active sites. -2 min -1 The variation in the equilibrium concentration at which the curves are leveling off is the presence of CO 2 as well as the development of a gypsum passivation layer, which prevents the acid fro 0 0.020.03 dissold, M 0.01M 020406080100120140 Time, mingve model coated data coated data uncoated model uncoatedFigure 5-40. First-order model fit ting for data of 0.25-0.35 mm particle size 0.0050.020.02 5 0 0.01g diss 0.015olved, M M 020406080100120140Time, min data uncoated model uncoated model coated data coated igure 5-41. First-order model fitting for data of 1-2 mm particle size F

PAGE 118

103 Table K Ceq 5-6. Reaction rate constants of first-order model Large particles Uncoated 0.025 0.02 (1-2 mm) Coated 0.0115 Small particles Uncoated 0.25 0.02 (0.25-0.35 mm) Coated 0.0115 Acid concentration dependence model Chemical reaction rate equations are used to define a conversion of the material or change in concentration due to dissolution and involve both a rate constant and a reaction order. However, a third parameter is also included in this model based on fact that the reaction product forms a barrier to acid diffusion. The circumstances under which this barrier is created are as follows: First, the acid concentration decreases with time as an evolution of reaction products. The reaction products then form a barrier between the acid and the particle surface. The thickness of this barrier increases with time until it reaches a certain stage where no acid can diffuse through and/ or the saturation stage is reached. Compton [1989] found that under very acidic (pH < 4) conditions, dissolution control passed from H+ diffusion control to a first-order heterogeneous reaction of H+ at the surface. This model can be presented by the following equation mAKCd t dc .(5.19) But the acid concentration changes with increasing the product concentrations according to the following relationship: (5.20) al form, which is: iAoAnCCC By plugging Equation 5.19 into Equation 5.20, we arrive at a differenti

PAGE 119

104 mnCCKdc)( iAdt5.21) By integrating.21, the ression will b .( Equation 5 final exp e: n CKtmmA1)1( nCCAi1][Wher m(5.22) 1 e, C A : acid concentration, M C AO : initial acid concentration, M C i : ions concentration, M n : number of ions evolved as reaction products K : reaction rate constant, min -1 t : time, min m: reaction order Figures 5-42 and 5-43 show the fitting of this model to experimental data. Its parameters are listed in Table 5-7. 00.01 0.03 0.02ed, M 0204060801001201 40Time, minMg dissolv model coated data coat ed data uncoated model uncoated Figur e 5-42. Model fitting for data of 0.25-0.35 mm particle size

PAGE 120

105 0 0.0050.010.0150.02508Time, minMg dissolved, 0.02 M 020406 0100120140 data uncoated model uncoated model coated data coated p Figure 5-43. Model fitting for data of 1-2 mm article size Table 5-7. Acid concentration dependence parameters Size, mm K n m 1-2 Coated Uncoated 1.65X10-21.65X10-211.9 11.9 3.97 2.85 0.25-0.35 Coated Uncoated 1.65X10-21.65X10-229.0 16.7 1.79 1.68 The reaction rate constant, K, is 1.65X10-2 min-1. For the small particles, the number of ions, n, in the case of coated particles is higher than uncoated because of a lesser volume around each particle in the presence of the polymer. This is not the situation in the case of the large particlesbecause of the lower surface area, the generation of ions is less and there is no competition among ions to diffuse out. Therefore, the n is the same, 11.9, in both cases. The order of the reaction, m, is primarily used to express the availability of acid. he acid, wile for oated particles, the polymer causes the formation of a barrier and limits both the accessibility of acid as well as the outgoing transfer of reaction products. Consequently, a higher m is needed to keep the reaction going at a constant rate. For uncoated particles m is smaller because there is no barrier for t h c

PAGE 121

106 Estimation of CO2 Required to Float the Dolomite Particle Simple calculations were used to predict the size of the particle with its surrounding envelope of gas and to estimate the relation between the radius of the particle before and after the formation of the gas bubbles around the particles. For simplicity, the following conditions were assumed: spherical particle, uniform gas layer around the particle, and at equilibrium condition: The part of the particle dissolved in the reaction was also neglected in the calculations. s to float when the density of nsity of acidic solution. The w As a starting point, we assumed that the particle start the particle and the gas layer around it is less than the de calcu lations are as follo s: Coating a g ent CO2 g asDolomite p article r rM 1Where: M : mass of CO gas Vd : volume of particle Vw : volume of water R1 : total radius (particle + gas) e area of the d : mass of dolomite particle g2M w : mass of water V g : volume of CO 2 gas R: particle radius S d : surfac p article r 1.4 r3434 2.9)r-(r34 r34 r 34 2.9VV MVVV1133133333wgd r )r 2.9(r r 0) (M gm/cm M M(solution) (acidicdensity Omide1gddg3 1w ) M gd gas) 2 Cte (Dolonsity

PAGE 122

107 In addition, the ratio between gas volume and particle vol ume and the ratio between the ga From the previous calculations, we can deduce that: The total radius is 140 % of the original radius of the particle (without taking the dissolved part of the particle and the pores volume into account). Also, the volume of gas around the particle is approximately twice the original particle radius. A larger particle needs more volume of gas to float due to its lower surface area. Change in Density: The density change can be calculated according to the sequence in Figure 5-44. The model will be used to calculate the number of moles of Mg, ntly will give the number of moles of CO2 through the equation of hemical reaction. Using thety, the volume olumgenated in certain pee can then beerage owr ofO2 as it evolvee used to calculate the change in the ensity awell as to correlate that change to membranelow rates. s volume and particle surface area were calculated as follows: which conseque c CO 2 densi e of CO ga 2 s can be calculated. Th v e ate er C riod of tim used to calculate the av fl s. This flowrate can b d s elasticity at different f rr58.0)1((1.4) V3rS3V3g2d1g rrareasurfaceparticlevolumeGas)r((1.4) Vr4)(r4 S:333g233d 3Sd dg31.75V V75.11(1.4) dgg1g VV]r[(1.4)V3Vr 3333d1333dr V1.4r rr34)(r4 V:andrvolumeparticlevolumeGas

PAGE 123

108 igure 5-44. Sequence of calculating the particle density and membrane elasticity Figure 5-45 shows the change in dolomite particle density as a function of reaction time. It is very rapid due to the CO2 build up in the RF process. The density of the dolomite particles changes rapidly due to intense reaction in the first 20-30 seconds; after that the rate of change decreases dramatically (Figure 5-45). The change is sharper in the case of the smaller particles due to their higher surface area and greater amount of CO2 generated. The density can be calculated by: ure 5-44. Sequence of calculating the particle density and membrane elasticity Figure 5-45 shows the change in dolomite particle density as a function of reaction time. It is very rapid due to the CO2 build up in the RF process. The density of the dolomite particles changes rapidly due to intense reaction in the first 20-30 seconds; after that the rate of change decreases dramatically (Figure 5-45). The change is sharper in the case of the smaller particles due to their higher surface area and greater amount of CO2 generated. The density can be calculated by: Model Mg Moles CO2 Moles Density CO2 flowrate DST Elasticity F gp ppfVV V (5.23) Where: : floated particle density, g/cm fp 3 p : particle density, g/cm 3 V: particle volume, cm3, and V g : gas volume, cm 3

PAGE 124

109 0.01.02.03.0051015Time, minDensity, gm/cm3 0.2-03 mm 1-2 mm 2-5 mm 5-9 mm Figure 5-45. Density changes for different particle sizes Bubbles Stability As the reaction transpires at the dolomite particle surface, the CO2 gas produced affects the growing bubbles in terms of their stability, (i.e., the stretching and elasticity of the polymer due to gas-pressure on the adsorbed polymer layer). Presence of the surfactant or polymer adsorbed at either the gasliquid or liquidliquid interface leads to the viscoelastic properties of that interface. They may be characterized by a surface dilatational modulus, which is defineaccorg to Gibbss equation [Johnson and Stebe, nkenheimer et al., 1996; Pikhitsa and Tsargorodskaya, 2000]: d din 1 996; Lu AdE dln (5.24)ic surface tension (DST), and A is the geometric area of the surface. For that reason, the Dynamic Surface Tension (DST) measurements using the Maximum Bubble Pressure Technique were determined at different gas flow rates: 0.23, Where is the dynam

PAGE 125

110 0.47, 0.78, and 1.55 cm3/min. In Figure 5-46, the experimental data of DST for PVA solutions of different concentrations are given. These results show the change in Dynamic Surface Tension as a function of the increase in polymer concentration. Clearly, at PVA polymer concentrations of up to 1 %, the surface tension significantly decreases as compared to the surface tension of water. Over the concentration range from to 1 % to 3 %, the surface tension decreases gradually but at greater concentration levels the surface tension remains nearly constant. The decrease in surface tension can be described, according to the Ward and Tordai model [Pefferkorn et al., 1985; Nahringbauer, 1995; Fainerman and Miller, 1996], as a orption process. For a short time scale (from milliseconds to seconds) the equation can be written as diffusion-controlled ads 21)(2)(Dtcto (5.25) surface concentration at any time, (g m Where, Co bulk concentration, (g m), : commonly used value ore pronounced at lower flow ratesthe greate the variation in This can be referred to as the time needed for the polymer to get onto the newly formetes, the bubbles surfaces remain almos an active agent at the interfacial region will be lese 2 ), 2D monomer diffusion coefficient, (m 2 s 1 ). t: time, (s) It is interesting to note that these decreases in surface tension are m r the increase in the flowrate, the smaller the DST d interfacial region. Therefore, at high flow ra t clean and the effect of the polymer as s because the polymer molecules do not have enough time to adsorb onto it. Th

PAGE 126

111 lower the flow rate, the higher the chance for the polymer molecules to adsorb onto thinterface of the generated bubbles. e 60 66687072DST,cm 6264012345 74PVA conc, % dyne/ Flowrate, cm3/min 1.55 0.78 0.47 0.23Figure 5-46. Dynamic surface tension of PVA concentration at different CO2 flowrate Figure 5-47icity in terms of PVA concentration at a different CO2 gas flowrate. It is interesting to note that This is due to formation of a membrane ration, which is able to resist the pressure he CO2 gas generated. However, the higher the bubble formation rate (flow shows the results of using Gibbs equation to determine the elast the elasticity increases with PVA concentration as a result of increasing the polymer concent exerted by t rate), the lower the elasticity. This can be referred to as the lower diffusivity of the polymer to the bubble surface and, consequently, DST at high bubble formation rates. The flow rate of CO 2 generation was found to be very low due to the reaction, which is closer to equilibrium conditions. This also indicates higher membrane elasticity.

PAGE 127

112 024681012 14 16012345PVA conc, %Elasticity, dyne/cm2 1.55 Flowrate, cm3/min 0.78 0.47 0.23 Figure 5-47. Elasticity of PVA membrane in terms of PVA concentration at different CO2 flowrate. -39). and 5-49 for both two different fractions. The fitting parameters are given in Table 5-8. It should be noted that these parameters are different from the dolomite system due to differences in the number of active sites on both cases. Examining the data in Figures 5Validation of the Model Using a Different Carbonate Mineral The calcite was used as another carbonate material to validate the developed models. The calcite dissolution tests were conducted for coated and uncoated calcite particles in 3 % sulfuric acid solution at room temperature. Figures 5-48 and 5-49 show the results of both coated and uncoated particles. As expected, the curves show similartrends as that obtained in the dolomite dissolution experiments (Figures 5-38 and 5In other words, lower dissolution is obtained in presence of polymer coating. Once again this may be explained on the basis of the presence of a polymer, which controls the transfer of ions to and from the boundary layer and then to the bulk solution. Fitting of active site model to this experimental data is given in Figures 5-48

PAGE 128

113 48 and 5-49 indicate an excellent fit of the model to the calcite system, which is a good verification for the model. Table 5-8. Active sites disappearance parameters for dissolution of calcite Large Small Uncoated Coated Uncoated Coated *P-1 0.016 0.008 0.06 0.06 0.0051 0.0075 0.0013 0.0026 *Km5.97 5.97 10 10 Acini 5.9 2.5 30 9.3 P-1 = KrA Km= KmA 0.01Ca dissolve 00.020.025 .015d, M 00.005020406080100120140160Time, Min. Model-uncoated Exp-uncoated Model-coated Exp-coated Figure 5-48. Calcite dissolution: model vs. experimental data 0.25-0.35 mm particle size 00.005Time, Min.Ca 0.01020406080100120140160 die 0.0150.020.025ssolvd, M Model-uncoated Exp-uncoated Model-coated Exp-coated Figure 5-49. Calcite dissolution: model vs. experimental data for 1-2 mm particle size

PAGE 129

114 Summarizing Remarks It was found that wor k of adhesion decreases with increasing PVA concentration. ill create om for bubbles increaA membranes ability to expand. On the isolated hyite, which can be used evidence that the adsorption o interaction o with dolomigen bonding. However, the FTIR did not give a clear explanation for the adsorption of PVA on apatite. As a physical adsorption confirmation, zeta potential measurements indicated that the PVA does not change the IEP for either apatite or dolomite. Also, the adsorption of PVA in the presence of supernatants indicates that the PVA adsorption is not affected by the supernatants even though the PVA may tend to form aggregates in highly ionic strength (as reported in literature). Moreover, The adsorption isotherm is S-Shaped, which means adsorption is physical on both minerals. The adsorption on dolomite is more stretched than on apatite. Also, the parking area of PVA on dolomite is less than on apatite. Furthermore the adsorbed layer thickness is higher in the case of dolomite than apatite, desorption tests indicated that the process of PVA adsorption on the minerals surfaces (dolomite and apatite) is a slowly reversible process, and the PVA membrane elasticity was determined using dynamic surface tension. The elasticity modulus increases with PVA concentration and decreases with increasing CO2 flowrate. More Low work of adhesion is favorable for the reactive flotation processthis w ro to form and se the PV other hand, FTIR shows that the presence of droxy l groups is exclusive for dolom r f PV A te depends mainly on hydro higher than the adsorption on apatite and the conformation of the polymer on dolomite is

PAGE 130

115 interestingly, the rate of CO2 evoce is very low which keeps the membcalcite system, which is a good verifi lution at the particle surfa rane at its higher elasticity. From modeling point of view, two models were developed to predict the CO 2 formation rate and these models agreed well with the experimental data. The density change of the particle was predicted and verified by single particle beaker tests. As another carbonate mineral, calcite was used. Calcite dissolution tests showed similar trends as that obtained in the dolomite dissolution experiments. Moreover, the model showed an excellent fit to experimental data of the cation for the model.

PAGE 131

116 rolling factors were screened using the Placket Burman design and three factors were determined to be significant within a 95 % confidence vel. These factors were PVA concentration, acid concentration, and particle size. To urther explore the effect of these factors, Central Composite Design (CCD) was used. he surface responses of CCD for different particle sizes indicated that the percentage of gO could be lowered to less than 1 % with recovery increasing with decreasing the article size. Additionally, it was noticed that the PVA consumption in the original RF process as about four to six pounds per ton, which is very high and unfeasible. Many attempts decrease the PVA consumption were conducted and it was found that using a spray to re-condition the rock can decrease the consumption to as little as one pound per ton. his was determined by running a factorial design in terms of particle size, conditioning me, and PVA. The optimized results were applied for process enhancement using a sluice, a ravity concentrator, which separates particles depending on the differences in the gravity CHAPTER 6 CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK Conclusions In this study, the reactive flotation (RF) process was developed with a high potential to separate dolomite from phosphate rock at high recoveries. The optimization of this process was carried out using screening and central composite designs. The cont le f T M p w to p T ti g 116

PAGE 132

117 o f individual particles. The results were very encouraging and similar to the data obtained in beaker tests. With the RF process developed and optimized, the next step was understood the mechanismerms of dolomite/PVA and phosphate/PVA well as bubble formation and stabilstudied using severth sion VA ease in surface tension indicates that PVA is a surface-active molecule, whicheen ydroxyl (OH) is the functional group in PVA molecules, this polymer (PVA s involved in this process. the process was explored in t Consequently, interactions as ity. The interactions of both dolomite and phosphate with PVA were al techniques in order to elucidate the mechanism of adsorption of PVA on bominerals. Work of adhesion was determined by measuring the equilibrium surface tenand contact angle. It was found that the work of adhesion decreases with increasing Pconcentration, which indicates film formation on both Dolomite and Phosphate particlesThis decrease is due to the decrease in surface tension and increase in contact angle (making the cosine of the contact angle smaller). The final result is an overall decrease in the work of adhesion. The decr satisfies the water molecules at the water/air interface. Moreover, the surface tension remains at a constant value after 1 % PVA concentration is reached. This indicates the formation of a tight molecular film at the interface, which, consequently, increases the viscosity and contact angle due to the formation of hydrogen bonds betwPVA molecules. Since the h ) was expected to adhere to surfaces by hydrogen bonding. To confirm this hypothesis, the FTIR was used to characterize the surface of both minerals. Results of the

PAGE 133

118 FTIR indicated the presence of isolated OH groups on dolomite only and hydrogen bonded OH grou ps on both minerals. adsorption isotherms were determined for both minerals as well. Isotherms were s s: The adsorption on dolomite is higher than on apatite O2 bubbles will burst. Accordingly, the low flow rate is required to keep ce ound rmation rate to be very slow relative to the data measured by DST. rticle Secondary evidence, indicating the physical nature of PVA adsorption on Dolomite and/ or Apatite particles, was also considered. By using zeta potential measurements, results indicated no shift in the isoelectric point of PVA. The studied as a function of the equilibrium time, which was determined from previouadsorption kinetic studies to be 2 hours. Characteristics of the isotherms were as followThe isotherms were S-shaped The adsorption process resulted in the formation of multiple layers on dolomite. The thickness of the adsorbed layer on dolomite was calculated to be about 3 times thicker than apatite. The importance of the thickness of adsorbed layer lies in its resistance to the C formation rate. If the CO 2 formation rate is very high the stability of the PVA membrane will be affected and the the bubbles at maximum stability. The previously discussed finding was revealed using measuring dynamic surfatension (DST) maximum bubble pressure method at different flow rates. It was fthat increasing the PVA concentration increases the elasticity modulus. However, increasing the flowrate decreases the elasticity modulus. The actual experimental results showed the CO 2 fo In the kinetic study, the CO 2 rate was calculated in terms of Mg +2 ions dissolved. From the reaction equation, the CO 2 was determined stoichiometrically for two pasize fractions. The results of the kinetics experiments revealed that bare (uncoated)

PAGE 134

119 particles dissolved faster than coated ones. However, both of them reached a constanlevel of Mg t period of time. These observations were lubility than uncoated particles. In addticle. The second is the disappearance of tiveill continprodule densitythe smaller the particle size, the shBuild a new sluice design capable of continuous operation performance in terms of grade and recovery +2 dissolved concentration after certain attributed to the formation of product layers such as CO 2 or gypsum, which act as barriers. I was noted that the coated particles had a higher so ition, both types of particles reached a plateau of solubility after a certain period of time. This distinct behavior of the particles was explained by two controlling mechanisms. The first is the acid blockage mechanism in which CO 2 formation prevents the acid from reaching the surface of a par ac sites to which the magnesium carbonate can adsorb. In both cases, the reaction w ue until all active sites have been used or the sites are covered with reaction cts. The modeling of these two mechanisms were carried out and fitted to experimental data, to which they conformed well. The same models used to predict the CO 2 formation rate were used to predict the decrease of partic arper the change in the density. To validate this model, the predicted values for several size fractions were compared with experimental beaker tests. These showed a good agreement as well. Recommendations for Future Work There are several suggestions for future work: Investigate a lower polymer dosage with fine fractions Test a lower molecular PVA polymer Extend the model derived in this work for prediction of gravity separator

PAGE 135

LIST OF REFERENCES Abdewith sate. Amankonah, J.O., Somasundaran, P., 1985. Effects of dissolved mineral species on the ganic and surfactants and its role in adsorption and flotation. Colloids and Surfaces 13, 151-Andresolutions. Polymer Science USSR 15 (8), 1991-1998. Atalay, V., 1985. Beneficiation of low grade Tasit phosphate ore from Turkey. World elastic properties of sparse-network polyvinyl alcohol. Polymer -482. Bajpai, A., Vishwakarmahol onto Fuller's earth surfacand Engineering Aspects 220 (1-3), Science 74 (2), 331-0. l Khalek, N.A., 2000. Evaluation of flotation strategies for sedimentary phosphate iliceous and carbonates gangues. Journal of Minerals Engineering 13 (7), 789-793. Abu El-shah, S.I., 1991. Beneficiation of calcareous phosphate rock using dilute acetic acid solution: optimization of operating conditions for Ruseifa (Jordan) phosphInternational Journal of Mineral Processing 31, 115-126. Adamson, A., Gast, A., 1997. Physical Chemistry of surfaces. Wiley (6 th edition), New York electrokinetic of calcite and apatite. Colloids and Surfaces 15, 335-353. Ananthapadmanabhan, K.P., Somasundaran, P., 1985. Surface precipitation of inor167. yeva, V., Anikeyeva, A., Lirova, B., Tager, A., 1973. Investigation of PVA Congress on non-metallic Minerals Proceedings, Belgrade, 389-396. Attia, Y., Fuerstenau, D.W., 1989. The equilibrium composition of hydroxyapatite and fluorapatitewater interface. Colloids and Surfaces 34, 271-285. Babayevskii, P.G., Kozlov, N.A., Trostyasnkaya, F., 1986. Effect of water on physical transformations and viscoScience USSR 28 (2), 475 N., 2003. Adsorption of polyvinyl alco es. Colloids and Surfaces A: Physicochemical 117-130. Barker, M., Garvey, M., 1980. The effect of solvency on the adsorption of poly(vinyl alcohol) onto polystyrene latex. Journal of Colloids and Interface 34 120

PAGE 136

121 Becker, P.,1989. Phosphates and phosphoric acid: Raw materials technology and economics of the wet process. Fertilizer Science and Technical Series, 2nd ed., 6, Marcel Dekker Inc. Behl, S., Moudgil, B., 1993. Mechanisms of PEO interactions with dolomite and apatite. Journal of Colloids and Interface Science 16443 Bjelopavlic, M., Singh, P., El-Shall, H., Moudgil, B., 2000. Role of surface molecular and Branion & H.G.Ebner. amp, B., Schoute, M., Veringa, H., 1989. Elsevier Science, London, G. de With, R. vergrowths ic acids. Journal of Chemical Technology and Biotechnology 70, 349-354 te as f Florida. ainesville, University Chibowski, S., Paszkiewicz, M., Krupa, M., 2000. Investigation of the influence of the Compton, R.G., Pritchard, K.L., Unwin, P.R., Grigg, G., Silvester, S., Lees, M., House, rson, C., Latham, G., Bates, R., 1978. Solubilities and activity coefficients of calcium and strontium sulfates in synthetic sea water at 0.5 and 25C. Journal of Physical Chemistry 82 (25), 2693-2699. 1, architecture and energetics of hydrogen bonding sites in adsorption of polymers and surfactants. Journal of Colloid and Interface Science 226 (1), 159-165. Blancarte-Zurita, M.A., Branion, R.M., Lawrence, R.W., 1987. Application of a shrinking particle model to the kinetics of microbiological leaching. In FundamentalApplied Biohydrometallurgy, eds R.W.Lawrence, R.M.R. Elsevier Science Publishers, Amsterdam, 243-253. Bonek A. Terpstra, R. Metselaar, Eds., EuroCeramics 1, 145. Booth, J., Hong, O., Compton, R., Prout, K., Payne, R., 1997. Gypsum o passivate calcite to acid attack. Journal of Colloid and Interface Science 192, 207. Cameselle, C., Nunez, M.J., Lema, J.M, 1997. Leaching of kaolin iron-oxides with organ Chanchani, R., 1986. Selective flotation of dolomite from apatite using sodium oleacollector. Ph.D. Dissertation, Gainesville, University o Chander, S., Furestenau, D., 1982. On the dissolution and interfacial properties of hydroxyapatite. Colloids and Surfaces 4, 101-120. Cheng, Y.C., 1985. The effect of surface hydration on the adsorption and flocculation ofa model silica suspension using polyethylene oxide. M.S.Thesis, G of Florida. Chibowski, S., Paszkiewicz, M., 1995. Journal of Polymer Chemistry 69, 461. polyvinyl alcohol adsorption on the electrical properties of Al 2 O 3 solution interface, thickness of the adsorption layers of PVA. Powder Technology 107 (3), 251-255. W.A., 1989. Journal of Chemical Society, Faraday Transactions I 85, 4335. Culbe

PAGE 137

122 Davis, B.E., Liewellyn, T.O., Smith, C.W, 1984. Continuous beneficiation of dolophosphate rocks. RI 8903, United States Bureau of Mines. mitic Dho, H, Iwasaki, I., 1990. Role of sodium silicate in phosphate flotation. Minerals and Dijit, J., Stuart, M., Hofman, J., Fleer, G, 1990. Kinetics of polymer adsorption in osphate ores in acidic media. International Journal of Mineral Processing 38, 235-256. gy ience 178 (1), 168-175. Fbiehn, G., Erusberger, M., 1951. Industrial Engineering Chemistry 43, 1108. y Fu, E., Somasundaran, P., 1986. Alizarin red S as a flotation modifying agent in calcite-Fuerstenau, M.C., 1968. The influence of sodium silicate in non-metallic flotation Metallurgical Processing 7, 215-235. stagnation point flow. Colloids and Surfaces 51, 141-158. Elgilani, D.A., Abouzeid, A-Z.M., 1993. Flotation of carbonates from ph El-Shall, H., Bogan, M., 1990. Characterization of future Florida phosphate resources. 2 ndprogress report, FIPR Project 89-02-082R. El-Shall, H., Zhang, P., Abdel Khalek, N., El-Mofty, S., 2004. Beneficiation technolo of phosphates: challenges and solutions. Minerals and Metallurgical Processing 21 (1), 17-26. El-Shall, H., Zhang, P., Snow, R., 1996. Comparative analysis of dolomite/francoliteflotation techniques. Minerals and Metallurgical processing 7, 135-140. Fainerman, V., Miller, R., Joos, P., 1994. Colloid and Polymer Science 272, 731-740Fainerman, V.B., Miller, R., 1996. Adsorption kinetics of short-chain alcohols at the water/air interface: diffusion-controlled adsorption under the conditions of a nonequilibrium surface layer. Journal of Colloid and Interface Sc Farmer, V., 1974. The layer silicate. The Mineral Society, London. Finch, C.A., 1973. Polyvinyl alcohol; properties and applications. John Wiley and Sons, New York. Finch, C.A., 1992. PVA-Developments. John Wiley and Sons, New York. Froment, G. F., Bischo, K. B., 1979. Chemical Reactor Analysis and Design. John Wileand Sons, New York, USA. apatite systems. International Journal of Mineral Processing 18 (3-4), 287-296. systems. Transactions of American Institute of Mining, Metallurgical, and PetroleumEngineers 241, 319-323.

PAGE 138

123 Garba, O., Yahya, S., Ali, A., Hamad, E., 1996. Surface and interfacial activities of hydrophobically modified poly (vinyl alcohol) (PVA). Poly mer 37 (7), 1183-1188. ressure method. Journal of Colloids and Interface Sciences 132 (2), 475-490. echnology for Florida high dolomite pebble. In: Beneficiation of Phosphates: Fundamentals and Technology. urfaces 25, 77-90. Gu, Z., Geo, Z., Hwang, C.L., 1999. Development a new technology for beneficiation of Hanna J., Anazia I., 1990. Selective flotation of dolomitic limestone impurities from 691. Hidber, P., Graule, T., Gauckler, L., 1995. Competitive adsorption of citric acid and ration during drying. urnal of American Ceramic Society 78 (7), 1775-1780. Hong, P., Chou, C., He, C., 2001. Solvent effects on aggregation behavior of PVA Howard, G., McConnel, P., 1967. Adsorption of polymers at the solution-solid interface. rnal of Physical Chemistry 71 (9), 2981-2990. 10-13. dy Garrett, P., Ward, D., 1989. A reexamination of the measurement of dynamic surface tensions using the maximum bubble p Geo, Z., Zheng, S., Gu, Z., 2002. Review of beneficiation t P.Zhang, H.El-Shall, P.Somasundaran, and R.Stana (Editors), SME, USA, Chapter 21, 235-245. Gotze, Th., Sonntag, H., 1987. The effect of solvent quality on the interaction between two layers of adsorbed PVA. Colloids and S Gruber, G.A., Raulerson, J.D, Farias, R.P., 1987. Adapting technology to beneficiate a low-grade phosphate ores. Minerals and Metallurgical Processing 7, 14-18. Florida dolomitic phosphate resources. Final Report, FIPR-02-129-167, Florida, USA. Florida phosphates. University of Alabama, FIPR report No. 002-066-089. Harkins, W., 1928. Surface energy and orientation of the molecules in surfaces are revealed by surface energy relations. Journal of Physical Chemistry 139, 647poly(vinyl alcohol) onto alumina and its influence on the binder mig Jo solutions. Polymer 42 (14), 6105-6112. I. Polyethers on silica. Journal of Physical Chemistry 71 (9), 2974-2981. Howard, G., McConnel, P., 1967. Adsorption of polymers at the solution-solid interface. II. Polyethers on carbon. Jou Hsieh, S.S., Lehr, J.R., 1985. Beneficiation of dolomite Idaho phosphate rock by the TVA diphosphonic depressant process. Mineral and Metallurgical Processing 2, Ince, D., Johnston, C., Moudgil, B., 1991. Fourier transform infrared spectroscopic stuof adsorption of oleic acid/oleate on surfaces of apatite and dolomite. Langmuir 7 (7), 1453-1457.

PAGE 139

124 Ince, D.E., 1987. Effect of sodium chloride on the selective flotation of dolomite froapatite. Ph.D. Dissertation, Gainesville, University of Florida. m ), ce .O., Stebe, K.J., 1996. An oscillating bubble technique to determine surfactant mass transfer kinetics. Colloids and Surfaces A: Physicochemical and Engineering Kelvin, L., 1871. Physics Magazine 42, 368. tex-particles bearing an adsorbed layer of poly (vinyl alcohol). Journal of Colloids and Interface Kiselev, A.V., Rygin, V.I., 1972. Infrared spectra of surface compounds. John Wiley, Koopal, L., Hlady, V., Lyklema, J., 1988. Electrophoretic study of polymer adsorption: ids yer 2. ering Phosphate: Theory and Practice, Dec 5-10, alm Coast, Florida, Chapter 5, 60-72. U.S. Patent ,189,103. Florida. RI 8609, United States Bureau of Mines. Izmailova, V.N., Alekseyeva, I.G., Tulovskaya, Z.D., 1975. Rheological properties of interphase adsorption layers of PVA on liquid boundaries. Polymer Science USSR 17 (92422-2428. Jho, C., Burke, R., 1983. Drop weight technique for the measurement of dynamic surfatension. Journal of Colloids and Interface Sciences 95 (1), 61-71. Johnson, D Aspects 114, 41-51. Kim, I.T., Luckham, J.P., 1991. The viscoelastic properties of polystyrene la Sciences 144 (1), 174-184. New York. Dextrin, polyethylene oxide and polyvinyl alcohol on silver iodide. Journal of Colloand Interface Science 121 (1), 49-62. Koopal, L., Lyklema, J., 1979. Characterization of adsorbed polymers from double laexperiments: The effect of acetate groups in polyvinyl alcohol on its adsorption on silver iodide. Journal of Electroanalytical Chemistry 100, 895-91 Lawendy, T.A., 1993. Flotation of dolomitic and calcareous phosphate ores. EngineFoundation Conference. Beneficiation of PLawver, L.E., Snow, R.E., 1980. Method of beneficiating phosphate ores. 4 Liewellyn, T.D.,1982. Beneficiation of high-magnesium phosphate from southern Lipatov, Y., Sergeeva, L., 1974. Adsorption of polymers. Halsted Press, John Wiley andSons Inc., New York. Lloyd, T.B., 1994. Experimental procedures to characterize acid-base and dispersion force contributions to solid wettability: A Review. Colloids and Surfaces 93, 25-37.

PAGE 140

125 Luckham, P., Vincent, B., Tadros, Th., 1983. The controlled flocculation of particulate dispersions using small particles of opposite charge. IV. Effect of surface coverage of adsorbed polymer on heteroflocculation. Colloids and Surfaces 6 (2), 119-133. acid. Chemical Engineering Science 28, 691-700. lmonary surfactant layers. Colloids and Surfaces A: Physicochemical and Engineering Aspects 114, 199-C., 1993. A growing drop technique for measuring dynamic interfacial tension. Journal of Colloids and Interface Sciences 160 (2), 435-448. Mathur, S., Moudgil, B., 1997. Adsorption mechanism(s) of poly(ethylene oxide) on o, K., Neuman, A., 1993. Relaxation of adsorption layers at solution/air interfaces using axisymmetric drop-shape analysis. Colloids and Surfaces 69 Moudgil, B.M., 1988. Separation of dolomite from the south Florida phosphate rock. petitive adsorption of polymer and surfactants on solid substrates. Colloids and Surfaces 133, 93-97. 72. terface Science 102 (1), 138-145. g ion. Langmuir 2 (4), 428-432. thyl Journal of Colloid and Interface Science 176, 318. ionic surfactant. In Martin J.Schick (Ed.), Marcel Dekker, New York, 1, 627-682. calcined alunite. Adsorption 9, 125-132. Lund, K., Fogler, H., McCune, C., 1973. Acidization-I: The dissolution of dolomite in hydrochloric Lunkenheimer, K. Winsel, K., Fruhner, H., Fang, J., Wantke, K.-D., Siegler K., 1996. Dynamic surface tension and surface area elasticity of adsorbed pu 210. Macleod, C., Radke, oxide surfaces. Journal of Colloid and Interface Science 196, 92-98. Miller, R., Sedev, R., Schan (4), 209-216. University of Florida, FIPR report No. 02-023-066. Moudgil, B., Prakash, T.S., 1998. Com Mpandou, A., Siffert, B., 1987. Polyethyleneglycol adsorption at TiO 2 -H 2 O interface: Distortion of ionic structure and shear plane position. Colloids and Surfaces 24, 159-1 Mpandou, A., Siffert, B., 1984. Sodium carboxylate adsorption onto TiO 2 : Shortest chain length allowing hemimicelle formation and shear plane position in the electric double layer. Journal of Colloids and In Mysels, K, 1986. Improvements in the maximum-bubble-pressure method of measurinsurface tens Nahringbauer, I, 1995. Dynamic surface tension of aqueous polymer solutions, I: E(hydroxyethyl) cellulose (BERMOCOLL cst-103) Ottewill, R., 1967. Non Ozacar, M., 2003. Equilibrium and kinetic modeling of adsorption of phosphorus on

PAGE 141

126 Paradip, Moudgil, B., 1991. Selective flocculation of tribasic calcium phosphate from mixtures with quartz using polyacrylic acid flocculant. International Journal of MineralProcessing 32 (3-4), 271-281. Macromolecules 18, 2252-2258. of l gineering Aspects 167 (3), 287-291. Pugh, R., Stenius, P., 1985. Solution chemistry studies and flotation of apatite, calcite and al and Engineering Aspects 161, 307-319. 7. Processing 26, 123-140. and etasilicate. Colloids and Surfaces 34, 227-239. ce tension measurements of benzyl benzoate using the Sugden maximum bubble pressure method. Langmuir 8 (11), 2644-2648. e) nal of Colloids and Interface Science 57 (1), 132-142. tes Bureau of Mines. es Bureau of Mines. Pefferkorn, E., Carroy, A., Varoqui, R., 1985. Adsorption of polyacrylamide on solid surfaces: Kinetics of the establishment of adsorption equilibrium Pikhitsa, P., Tsargorodskaya, A., 2000. Possible mechanism for multistage coalescencea floating droplet on the air/liquid interface. Colloids and Surfaces A: Physicochemicaand En Platonov, B., Baran, A., Polischuk T., 1979. Acta Physica Chemica 25, 201. fluorite minerals with oleate collector. International Journal of Mineral Processing 15, 193-218. Rachas, I., Tadros, Th.F., Taylor, P., 2000. The displacement of adsorbed polymer from silica surfaces by the addition of a nonionic surfactant. Colloids and Surfaces A: Physicochemic Rao, D.V., 1979. Flotation of low-grade phosphate ore. International Symposium on Resources Engineering and Technology, Jan 8-11, 1Rao, D.V., 1985. Flotation of calcareous Mussori phosphate ore. International Journal of Mineral Processing 14 (1), 57-66. Rao H.K., 1989. Flotation of phosphatic material containing carbonatic gangue using sodium oleate as collector and sodium silicates modifier. International Journal of Mineral Rao, H.K., 1988. Mechanism of oleate interaction on salt type minerals, I-Adsorptionelectrokinetic studies of calcite in presence of sodium oleate and sodium m Ross, J., Bruce, W., Janna, W., 1992. Surfa Rubio, J., Kitchener, J.A., 1976. The mechanism of adsorption of poly(ethylene oxidflocculant on silica. Jour Rule, A.R., 1970. Removal of magnesium impurities from phosphate rock concentrates.RI 7362, United Sta Rule, A.R., 1982. Application of carbonate-silica flotation techniques to Western phosphate materials. RI 8728, United Stat

PAGE 142

127 Rule, A.R.,1985. Beneficiation of complex phosphate ores containing carbonate and silica gangue. XV International Mineral Processing Conference proceedings 3, 380-389. Santhiya, D., Subramanian, S., Natarajan, K.A., Malghan S.G., 1999. Mineral and Subramanian, S., Natarajan, K.A., Malghan S.G., 1999. Surface chemical studies on the competitive adsorption of poly(acrylic acid) and poly(vinyl alcohol) onto atarajan, K., Malghan, S., 1998. Surface chemical studies on the competitive adsorption of PAA and PVA onto alumina. Colloid nd Surfaces A: Physicochemical and Engineering Aspects 159 (1), 197-208. Sjoberg, E.L., 1978. Kinetics and mechanism of calcite dissolution in aqueous solutions Smani, S. M., 1973. Beneficiation of sedimentary Moroccan phosphate ore: Part 4 leum Engineers 258, 181-182. and Petroleum Engineers 258, 176-180. hate ore containing dolomite. U.S.Patent 4,364,824. ds., Engineering Foundation Conference, Palm Coast, Florida. on. 9 p. and ds and Interface Science 64 (1), 36-47. Metallurgical Processing 16, 51. Santhiya, D., alumina. Journal of Colloid and Interface Science 216 (1), 143-153. Santhiya, D., Nandini, G., Subramanian, S., N and Surfaces 133, 157. Sjberg, M., Bergstrm, L., Larsson, A., Sjstrm, E., 1999. The effect of polymer and surfactant adsorption on the colloidal stability and rheology of kaolin dispersions. Colloids a Sjoberg, E.L., Rickard, D.T., 1984, Calcite dissolution kinetics: surface speciation and the origin of the variable pH dependence. Chemical Geology 42, 119-136. at low temperatures. Stockholm Contribution in Geology 32, 1-96. depression of phosphate oolites and calcite flotation. Transactions of American Institute of Mining, Metallurgical, and Petro Smani, S. M., Cases, J.M., Blazy, P., 1975. Beneficiation of sedimentary Moroccan phosphate ore: Part 3 selective flotation and recovery. Transactions of American Institute of Mining, Metallurgical, Snow, R.E., 1979. Beneficiation of phosphate ore. U.S.Patent 4,144,969. Snow, R.E., 1982. Flotation of phosp Somasundaran, P., Ramachandran, R., 1988. Innovative approaches to elucidate floc structure and polymer conformations at interfaces. B.M.Moudgil and B.J.Scheiner, E Stark, G., Wallace, H., 1982. Chemistry Data Book. 2nd ed., John Murray, Lond Sun, S. F., 1994. Physical chemistry of macromolecules: Basic principles and issues. John Wiley and Sons, Inc., New York, 46 Tadros, Th.F., 1978. Adsorption of polyvinyl alcohol on silica at various ph valuesits effect on the flocculation of the dispersion. Journal of Colloi

PAGE 143

128 Tamura, T., Kaneko, Y., Ohyama, M., 1995. Dynamic surface tension and foaming properties of aqueous polyoxyethylene n-dodecyl ether solutions. Journal of Colloids and Interface Sciences, 173 (2), 493-499. tion/air interfaces. II. Adsorption of surfactants in dilute aqueous solutions. Anomalous oscillating jet phenomena. Journal of Colloids and cyclone separation f phosphate pebble from dolomite. Society of Mining Engineers American Institute of P., 1989. Interaction between oleate collector and Alizarin modifier in dolomite / francolite flotation system. Minerals and Metallurgical Processing polyacrylic acid on silica, alumina, and kaolin. Journal of Colloid and Interface Science Zeno, E., Beneventi, D., Carr, B., 2004. Interactions between poly(ethylene oxide) and ffect Thomas, W., Hall, D., 1975. Solu Interface Sciences 51 (2), 328-332. Wiegel, R.L., Hwang,C.L., 1984. A predictive model for heavy mediao Mining, Metallurgical, and Petroleum Engineers Meeting, preprint # 84-648. Xiao, L., Somasundaran 5, 100-103. Zaman, A., Tsuchiya, R., Moudgil, B., 2002. Adsorption of a low-molecular-weight 256 (1), 73-78. fatty acids sodium salts studied by surface tension measurements. Journal of Colloid and Interface Science 277 (1), 215-220. Zhang, Y., Suga, T., Kawasaki, M., Xiao-Xia, T., Uchida, N., Uematsu, K., 1996. Eof poly(vinyl alcohol) adsorption on binder segregation during drying. Journal of American Ceramic Society 79 (2), 435.

PAGE 144

BIOG RAPHICAL SKETCH a ning engineering at Cairo University. He completed masters degree in May ), researcher assistant in 1993. In 2001, he came to the University of ch assistant and performed research on minerals Ayman El-Midany was born in Cairo, Egypt, in February 1969. He attended the College of Engineering at Cairo University from 1986 to 1991. He completed bachelors degree in mining engineering in 1991. In 1992, he enrolled for graduate studies in mi 1997. He Joined Central Metallurgical Research and Development Institute (CMRDICairo, Egypt, as Florida to pursue a Ph.D degree in materials science and engineering. He worked with Professor Hassan El-Shall as a resear separation and particle science technology. 129


Permanent Link: http://ufdc.ufl.edu/UFE0008000/00001

Material Information

Title: Separating Dolomite from Phosphate Rock by Reactive Flotation: Fundamentals and Application
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: UFE0008000:00001

Permanent Link: http://ufdc.ufl.edu/UFE0008000/00001

Material Information

Title: Separating Dolomite from Phosphate Rock by Reactive Flotation: Fundamentals and Application
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: UFE0008000:00001


This item has the following downloads:


Full Text












SEPARATING DOLOMITE FROM PHOSPHATE ROCK
BY REACTIVE FLOTATION: FUNDAMENTALS AND APPLICATION

















By

AYMAN A. EL-MIDANY


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

UNIVERSITY OF FLORIDA


2004

































Copyright 2004

by

Ayman A. El-Midany


































To all who made this work possible, and
To all who had parts in shaping the person that I am today.















ACKNOWLEDGMENTS

In the Name of Allah, the Most Gracious, the Most Merciful.
"Oh Allah! Make useful for me what you have taught me, and teach me knowledge
that will be useful to me. I entrust you with what I have read and studied. Oh Allah!
Bring it back to me when I am in need of it. You are my protector, and the best of
aid.". I thank Allah for blessing me with all of the good that I have, and I ask that I
will be able to use such gifts to help those in need. Amen.

I have fortunate to be able to study in an excellent institution surrounded by a group

of extremely intelligent people. I would like to recognize the amazing staff in Department

of Materials Science and Engineering at University of Florida. I would also like to thank

everyone who helped me arrive at this point. Special thanks and recognition go to my

advisor, Dr. Hassan El-Shall. He gave me the opportunity to do research in his group. His

drive and commitment to research have inspired and motivated me through this work. He

has been a constant source of ideas and encouragement over the years. Many thanks go to

my committee members; especially Dr. Spyros A. Svoronos, who helped me greatly and

bore with me in the modeling part of my study. I would also like to thank the other

members of my committee (Drs. John J. Mecholsky, Wolfgang Sigmund, and Susan

Sinnott from the Department of Materials Science and Engineering, and my external

committee member, Dr. Ben Koopman, Department of Environmental Engineering). I

would also like to recognize the remarkable past and present members of Dr. El-Shall's

research group for their assistance, advice, and support.

Finally, I cannot thank my family enough for all of their love, encouragement, and

counsel. To my parents, I owe everything. They have taught me many things. Above all









they taught me that patience, determination, persistence, and faith are the keys to success.

I thank my wife for her help and patience over the years.

I thank the Particle Engineering Research Center (PERC) staff for helping me to

get started and for providing support and guidance along the way. Much of the

experimental work in my study could not have been accomplished without their help, and

for that I am extremely grateful.

This work was supported by the Florida Institute of Phosphate Research (FIPR),

and the National Science Foundation / Particle Engineering Research Center

(NSF/PERC).
















TABLE OF CONTENTS

page

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

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

LIST OF FIGURES ......... ......................... ...... ........ ............ xi

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

CHAPTERS

1 LITERATURE REVIEW ..............................................................................1

In tro du ctio n ......................... ...... .... ...... .... ............ .......... ....................
Previous Efforts to Separate Dolomite from Phosphate Rock............... ..................
A nionic F location .............................................. ................ .. 4
D irect flotation ....................... .............................. .. ......... ............ ..
R everse flotation .......................... ........ .................... .. ..
C ationic F location ....................................................... 8
N onconditioning Flotation.......................................................... ............... 9
Two-Stage Conditioning Process ..... .................... ...................11
The CLD R I Process .......... ........ ....... ................... .... .........12
A cid L each in g ................................................... ................ 14
Physical Separation Methods ........... .. ......... ....... ............... 14
H eavy M edia Separation ............................................. ............................ 14
Sizing, Attrition Scrubbing, and Desliming ................................................15
Problem Statement........ ...................................... .. ........ .. .... .......... 15
N ovelty and Scope of Study .............. ..........................................................15
G o al o f S tu d y ...................... .. ............. .. .................................................16

2 M ATERIALS AND M ETHOD S ........................................ ......................... 17

Flotation Study......................................................... .. .. ............. 18
Screening Design (Plackett-Burman Design).......................... ................18
C central C om posite D esign ................................................ ............... ... 18
Flotation Procedures for Screening and CCD Designs ....................................21
Factorial Design for Conditioning with PVA before Flotation...........................21
Flotation Procedure for Conditioning Statistical Designs...............................21
Density and Surface Area Determination....................................22









PV A Preparation ........ .......................................... .......... ........ .......... 23
Spectrophotometric Determination of PVA...................................................23
Static (Equilibrium surface tension)................... ...... ..............24
Dynam ic Surface Tension (D ST) ............................................. ............... 24
Maximum bubble pressure method............................................................24
Dynamic surface tension setup ............... ............................................. 26
C o n tact A n g le ................................................................... ............... 2 6
M easurem ent of Contact Angle............................................... ........ ....... 27
Surface Characterization by FTIR .................................................................... 28
Zeta Potential M easurem ents ........................................ ......................... 28
A d so rp tio n .................. ................... ............................................. 2 8
Dissolution Kinetic Experim ents...................................................................... 29
Pilot Scale Experim ents .................. ..................................... ..... .. ........ 30
T he Sluice ............................................................................................. ....... 30
Sluice Experim ents .................. .......................... .... .... ... ........ .... 30

3 TECHNICAL ASPECTS OF REACTIVE FLOTATION: PRELIMINARY TESTS32

In tro d u ctio n ............. ...... ... ........................ ......................... 3 2
List of Potential Surfactants: Surfactants Screening ..............................................33
Initial Screening Tests ............................ ............ ................... ..... .... 33
S p ecific T e sts................................................. ................ 3 4
A cid ty p e ......................................................................34
A cid d o sag e .............................................................. ............... 3 5
Su rfactant ty p e ................................................. ........... ............... 3 6
Surfactant dosage ................... .... .......... ............. .... ........37
P article siz e .............................................................3 8
Oil-Surfactant Experiments ................................................... .................. 39
Poly-Vinyl-Alcohol (PVA) Testing ....................................... ............... 41
Effect of acid concentration ........................................ ...... ............... 41
Effect of particle size............ .. .............. ................ ......... .. ............. 43
PVA with acid vs. PVA on the particle surface .......................................44
Preliminary Testing Using PVA Coated Particles ................................................45
Flotation of a Representative Sam ple....................................... ............... 45
Fraction of Representative Sample (-4+1.19 mm) ............................................46
Effect of Time on Weight Percentage of Floated Particles .................................46
Effect of Size on the Flotation............ .............................. ............... 47
Sum m arizing R em arks........................................................................ ..................48

4 OPTIMIZATION OF THE REACTIVE FLOTATION PROCESS
AND PILOT SCALE TESTING........................................... .................... 49

Introduction ......... ..... .... ......................................................... ..... 49
Optimization of the RF Process ............ ..... ........ ................... 49
Screening Statistical Design .....................................................49
C central C om posite D esign........................................................ ............... 52









Research to Reduce the PVA Consumption .................................... ............... 52
Experimental Design of the PVA Conditioning Step........................................56
Saturation Experiment ............ ..... ............... ............... 56
Pilot Scale Testing: Sluice R results ........................................ ......... ............... 57

5 FUNDAM ENTAL STUDIES ............................................................................. 58

Polyvinyl alcohol-Particle Interaction ..... ................ ...............58
In tro d u ctio n ................................................................................................... 5 8
Surface Tension ......... ......... .................. ............... .... ........... 60
Contact Angle ............ ...... ..................... ...... ........ ... .......... 63
W ork of A dhesion .............................. ........................ .. ...... .... ...... ...... 64
Zeta Potential .............................................67
Zeta potential of minerals in supernatant solutions ............. ...............70
Zeta potential in the presence of PVA ............. ..................... ......... ...... 71
Infrared (FTIR) ....... ........ ....... ............................... .... ........72
Adsorption Study ......... ..... ...... ....... ......... .. .......... ................. 75
K inetics of adsorption ............................................................................ 76
Adsorption isotherm ................... ... ................................ 79
Calculation of PVA adsorption and parking area .............. .............82
A dsorbed layer thickness ........................................ ......... ............... 84
Effects of pH .............................. ..... ........ ..... .............. ................ .....84
Adsorption in supernatants .......................... 87
D e so rp tio n ...................................................... ................ 8 8
Desorption in Acidic Solution ...... ................ .. ..................89
Bubble Formation and Stability.......................................... 90
Effect of Particle Size ........... .. ..... ......... .................. ............... 91
Effect of PVA Coating ............................................ ......................94
Modeling of Bubble Formation and Dolomite-Particle Separation .........................96
B lockage of the A cid M odel......... ......... ...... ........................... ..... ......... 96
Disappearance of Active Sites M odel .......... .......... ......... .. ............ 99
Em pirical M odels .............................. .................. .. .......... .... ............. 101
First-order kinetic m odel ........................................... ............ ............... 101
Acid concentration dependence model .................................. 103
Estimation of CO2 Required to Float the Dolomite Particle ........................... 106
B bubbles Stability ................ ..... ...... .. .. .... .. .................. .. .............109
Validation of the Model Using a Different Carbonate Mineral .....................12
Sum m arizing R em arks......... ................. ...................................... ............... 114

6 CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK ............116

C o n c lu sio n s.................... ......... ........................................................................ 1 1 6
Recommendations for Future W ork .............. ............ ........... .... ............... 119

L IST O F R EFER EN CE S ....................... ................. .............................. ............... 120

BIOGRAPH ICAL SKETCH ....................... .......... ........................................... 129


viii
















LIST OF TABLES


Table pge

2-1. Experimental Plackett-Burman design for 11 variables................ .............. ....19

2-2. Rotatable central composite design with 5 levels and 3 variables .........................20

2-3. Factorial design for conditioning with PVA............... .........................................21

3-1. Surface-active agents used in RF tests ........................................ .....................33

3-2. Size analysis of the dolomitic phosphate sample ............................................34

3-3. Chemical analyses of the dolomitic phosphate sample ........................................34

3-4. Effect of sulfuric acid concentration ............................................. ............... 35

3-5. Effect of sulfuric acid concentration on the bubble formation..............................36

3-6. Preliminary tests of high dolomitic phosphate with different surfactants ..............37

3-7. Effect of surfactant concentration (SDSO3, 2 %) ................... ...................... 38

3-8. Effect of particle size on the dolomitic particle floatability ...................................39

3-9. Effect of presence of surfactant and oil layer on particle flotation ..........................40

3-10.Effect of acid concentration... ........................................................... ............... 42

3-11.Effect of acid concentration above 2 %.............................................. 43

3-12.Comparison between coating the particle before adding it to acidic solution and
adding the particle to acid+PVA solution without coating ....................................45

3-13.Results of representative sample (- 4+1.19 mm particle size)...............................46

3-14.Effect of time on the flotation of- 4+1.19 mm particle size ...............................47

3-15.Effect of particle size on flotation ..................................... ............ ....................47

4-1. The MgO % removal in floated fraction for Plackett-Burman design ...................50









4-2. Plackett-Burman calculated t values for different variables..............................51

4-3. Effect of saturated acid on the RF ................... ............ 57

4-4. Comparison of two different coating methods and their effects on MgO removal..57

5-1. Chemical analysis of the samples used in adsorption tests ....................................66

5-2. Fitting parameters of second-order kinetic model............................... ............... 77

5-3. Chemical analysis of the samples used in kinetic tests ........................................91

5-4. A cid-blockage m odel param eters ........................................ ......................... 97

5-5. Active sites disappearance parameters ....................................... ............... 100

5-6. Reaction rate constants of first-order m odel................................... ... ................ 103

5-7. Acid concentration dependence parameters ................................. ............... 105

5-8. Active sites disappearance parameters for dissolution of calcite .........................113
















LIST OF FIGURES


Figure pge

1-1. Simplified flowsheet for phosphoric acid production .............................................3

1-2. Simplified flowsheet of USBM direct flotation .............................. ...............6

1-3. Flowsheet of TVA's diphosphonic depressant process..............................................8

1-4. Flowsheet for IMCF cationic flotation......... ........................................9

1-5. University of Alabama pebble processing, simplified flowsheet...........................10

1-6. Flowsheet for University of Florida two conditioning stages .............................. 11

1-7. Simplified flowsheet for CLDRI dolomite flotation ..........................................13

2-1. Strategic plan to achieve research goals................... ....................17

2-2. Five-level, rotatable central composite design.. ................................... ............... 20

2-3. Characteristic bubble-pressure vs. time curve in MBP. .........................................25

2-4. Setup for the measurement of dynamic surface tension by MBP...........................26

2-5. Contact angle and its relation with surface tension ............................. .................27

2-6. Contact angle measuring apparatus ...................................................................... 28

2-7. Schematic diagram for separation of dolomite from phosphate in sluice ................30

3-1. Effect of acid concentration on bubble formation and flotation time.....................36

3-2. Effect of the particle size on the time needed to float the particle ...........................44

4-1. Effect of PVA dosage and acid conc. on MgO recovery % ...................................53

4-2. Effect of PVA dosage and acid concentration on MgO % ....................................54

4-3. Cube graph for MgO % and MgO recovery %................................................ 56

5-1. Contact angle and interfacial surface tension at solid/liquid/gas interface ..............60









5-2. Structural formula for PVA ...... ........................... ......................................61

5-3. Equilibrium Surface tension for aqueous PVA solutions at 23 C.........................61

5-4. PVA crystal structure; the dotted lines indicate hydrogen bonds.............................62

5-5. Static contact angle variation with the PVA concentration.................................64

5-6. Work of adhesion as a function of PVA concentration................ .............. ....65

5-7. Cumulative pore size distribution for apatite sample ............................................66

5-8. Cumulative pore size distribution for dolomite sample.........................................67

5-9. Zeta potential of dolomite and apatite samples ................................ ............... 68

5-10. Zeta potential of dolomite in absence and presence of polymer ............................68

5-11. Zeta potential of dolomite in absence and presence of polymer ............................69

5-12. Effect of polymer on the shear plane position.............. .................................69

5-13. Zeta potential of apatite and apatite in dolomite supernatant..............................70

5-14. Zeta potential of dolomite and dolomite in apatite supernatant ...........................71

5-15. Zeta potential of dolomite in apatite supernatant before/after PVA adsorption.....72

5-16. Zeta potential of apatite in dolomite supernatant before/after PVA adsorption.....72

5-17. Infrared transmittance spectra of poly vinyl alcohol (PVA) .............................73

5-18. Infrared absorbance spectra of dolomite and dolomite/PVA system ...................74

5-19. Infrared absorbance spectra of apatite and apatite/PVA system. .........................75

5-20. Calibration curve for PVA concentration measurements................... ..............76

5-21. Kinetics of PVA adsorption on dolomite and Apatite................ .............. ....78

5-22. Adsorption isotherm of PVA on dolomite at 23C ..............................................81

5-23. Adsorption isotherm of PVA on apatite at 230C............................... ..............82

5-24. Polymer-conformation changes with increase the concentration ..........................83

5-25. Effect of pH on adsorption of PVA on apatite and dolomite ...............................85

5-26. Adsorption isotherm of apatite in dolomite supernatant solution, and dolomite in
apatite supernatant solution ............... ................ .................... 87









5-27. Desorption of PVA from apatite surface ..... ......... ..................................88

5-28. Desorption of PVA from dolomite surface................................. ............... 89

5-29. The bubble form ation m echanism ........................................ ....... ............... 90

5-30. Schematic diagram of reaction kinetics of the RF ...............................................91

5-31. Dissolution kinetic of different particle sizes in terms of calcium ions ..............92

5-32. Dissolution kinetic of different particle sizes in terms of magnesium ions............93

5-33. Dolomite particle (1-2 mm) before and after the reaction with 3 % H2SO4...........93

5-34. Dissolution kinetics of coated and uncoated 0.2-0.3 mm dolomite particle ..........94

5-35. Growth of the bubbles with time on the dolomite particle surface, 50 X..............95

5-37. Acid blockage model fitting to 0.25-0.35 mm experimental data..........................98

5-38. Active sites disappearance model fitting to 1-2 mm data..................................100

5-39. Active sites disappearance model fitting to 0.25-0.35 mm data.........................100

5-40. First-order model fitting for data of 0.25-0.35 mm particle size..........................102

5-41. First-order model fitting for data of 1-2 mm particle size..................................102

5-42. Model fitting for data of 0.25-0.35 mm particle size.........................................104

5-43. M odel fitting for data of 1-2 mm particle size.................. .................. ................105

5-44. Sequence of calculating the particle density and membrane elasticity ..............108

5-45. Density changes for different particle sizes............................... ............... 109

5-46. Dynamic surface tension of PVA concentration at different CO2 flowrate .........111

5-47. Elasticity of PVA membrane vs. PVA concentration ............... .................... 112

5-48. Calcite dissolution: model vs. experimental data 0.25-0.35 mm particle size .....113

5-49. Calcite dissolution: model vs. experimental data for 1-2 mm particle size..........113















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

SEPARATING DOLOMITE FROM PHOSPHATE ROCK
BY REACTIVE FLOTATION: FUNDAMENTALS AND APPLICATION

By

Ayman A. El-Midany

December 2004

Chair: Hassan El-Shall
Major Department: Materials Science and Engineering

Florida is currently one of the largest phosphate producers in the world.

Unfortunately, the phosphate industry in Florida is suffering from depletion of the

phosphate-rich ore. Dolomite, however, is a major impurity in the ores. The presence of

MgO, in dolomite, can cause several problems during production of phosphoric acid and

final fertilizers. Efforts to solve the dolomite problem have been reported by several

researchers throughout the world. Yet, only heavy media separation process has been

applied commercially. However, the process was discontinued because of its low

separation efficiency in terms of grade and recovery.

In this study, we tested an innovative idea based on simple fact that dolomite, as a

carbonate mineral, generates CO2 when exposed to a slightly acidic solution, capturing

CO2 bubbles at dolomite particle surface can selectively float the dolomite and separate it

from phosphate. Formation of such bubbles requires a surface-active agents) at









solution/dolomite interface. Preliminary tests indicated that polyvinyl alcohol (PVA)

could serve this function. This process is called the Reactive Flotation, RF.

To develop this process and to scale it up for industrial use, we conducted

fundamental and applied research studies. Among these studies were optimizing of RF

process by investigating the factors affecting the process (including bench and pilot scale

tests) using statistical experimental designs. It was found that the acid concentration,

PVA concentration, and particle size are the main factors. The optimum conditions were

determined and applied to usage in a gravity separation device, (i.e., a sluice). Using such

a device, a concentrate contains 0.65 % MgO with > 94 MgO % removal can be

achieved. In addition, fundamental studies revealed that the hydrogen bonding is the

adsorption mechanism confirmed by adsorption isotherm of PVA on dolomite and

phosphate, adhesion of PVA to dolomite surface (using contact angle and surface

tension), Fourier Transform InfraRed (FTIR), and Zeta potential. Moreover, PVA film

thickness and polymer film elasticity (using dynamic surface tension) revealed that the

optimum conditions could keep the PVA film at its highest elasticity. On the other hand,

modeling was developed and used to predict CO2 formation rate and subsequent particle

density change.














CHAPTER 1
LITERATURE REVIEW

Introduction

Traditionally, the Florida phosphate industry has effectively responded to the

competitive pressures of the market place by implementing technological improvements

to mining and beneficiation methods. With further expansion of fertilizer production

capacity worldwide and depletion of Florida's high-grade deposits, competitive pressures

are expected to continue. To remain the world leader, Florida must continue to seek

environmentally sound technological improvements in all areas of phosphatic fertilizer

production.

Gu et al. [1999] estimated that the Bone Valley phosphate reserve in Central

Florida (which can be cost effectively processed with available technology) may only last

about 20 years at the current mining rate. In the near future, with the depletion of the

low-dolomite, easy-to-process Bone Valley siliceous phosphate deposits, phosphate

mining will move farther south and southeast. The phosphate matrix in the southern

extension is leaner in grade and high in dolomite. Dolomite is the most troublesome,

causing higher consumption of sulfuric acid, reducing filtration capacity, and lowering

P205 recovery in fertilizer manufacturing. Therefore, the MgO content of dolomite is an

important index in evaluating the quality of phosphate concentrate. In Florida, the MgO

content in final concentrate is usually required to be less than 1 %.

Geological and mineralogical data show that about 50 % of the phosphate resource

would be wasted if the high dolomitic zones were bypassed in mining, and about 13 % of









the resource would be lost if the dolomite pebbles were discarded. Although heavy

media separation (HMS) technology was tried commercially, to process a portion of

dolomite pebbles in Florida, in most cases the high-dolomite phosphate has been

bypassed during mining operations because no technology is available to process high-

dolomite pebbles economically. Technically and economically feasible beneficiation

technology for processing dolomite pebbles is needed to recover phosphate value, extend

Florida phosphate reserves, and prolong the life of Florida phosphate industry [Geo et al.,

2002].

Previous Efforts to Separate Dolomite from Phosphate Rock

Phosphate rock is one of the basic materials for fertilizers. About 90 % of the

phosphate rock output in the world is used for fertilizer production. The United States is

one of the world's largest producers of phosphate rock and fertilizer. Its phosphate

production represents approximately one-third of the world's total [Becker, 1989].

Phosphoric acid is an important intermediate product for production of fertilizers.

Current phosphoric acid technology essentially comprises sulfuric acid attack, and

separation of the phosphoric acid from the calcium sulfate crystals resulting from the

reaction. Both the attack and separation, which are effected by filtration, are considerably

influenced by the nature of the ore and its impurities [Becker, 1989]. Figure 1-1 shows

the main processes in phosphoric acid production.

As mentioned, the main impurity in Florida phosphate ore is dolomite. Dolomite,

(Ca, Mg)C03, is the main source for calcium and magnesium. As impurities, CaO and

MgO affect the phosphoric acid production by increasing sulfuric acid consumption, in

addition to the MgO stays with the acid phase, and greatly increases viscosity.









Calo(PO4)6F2 + 10 H2SO4 + 20 H20 6 H3PO4 + 10 CaSO4.2H20 + 2 HF

Phosphate
Phd o ate Wash Water
Concentrate
Sulfuric acid *

I


Reaction


J* Filration


Phosphoric acid rnospnovDosum

Figure 1-1. Simplified flowsheet for phosphoric acid production

Therefore, because dolomite is the most common impurity, developing an efficient

and economical method for separating dolomite from apatite is an effective way to extend

phosphate mine life. Generally, acidulation of phosphate rock requires a feed of less than

0.8 % MgO, which has not been achieved at a commercial scale with ores high in MgO.

Dolomite is also a worldwide problem. All of the major phosphate-producing countries

have faced, or are going to face, the dolomite problem.

Salt-type minerals (such as apatite, calcite, dolomite, barite, fluorite and scheelite)

have been successfully separated from oxides and silicates by flotation methods.

However, these minerals are often difficult to separate from each other because they

contain similar alkaline earth cations that cause them to exhibit similar surface (electrical)

behavior. All the zero points of charge (ZPC) of phosphates and carbonates are located in

the acid pH range (i.e., 4-5.5) [Smani, 1973; Smani et al., 1975].

Systematic research in carbonate phosphate ore beneficiation started in the early

1950s [El-Shall et al., 2004]. Techniques used include the following:

* Direct flotation of phosphate, using sodium silicate to depress carbonate impurities.









* Reverse flotation of the carbonate impurities by depressing the phosphate using
materials such as fluorides, fluosilicic acid, diphosphonic acid, phosphoric acid,
aluminum sulfate-tartrate, and inorganic phosphate salts.

* Rapid change of conditioning parameters (such as the University of Florida two-
stage conditioning process, Alabama non-conditioning process, the IMC-F process,
and the low-pH conditioning process proposed by Minemet research).

* Physical methods: gravity concentration, sizing, desliming, heavy media
separation.

* Calcination: although expensive because of its energy requirement, high-
temperature destruction of the carbonate followed by leaching has been practiced in
North Africa with low -cost energy.

* Acid leaching: carbonates are easily dissolved by acid, and weak acids can be used
to dissolve the carbonate, leaving the apatite.

Anionic Flotation

Direct flotation

Sodium silicate is widely used as carbonate depressant [Fuerstenau, 1968;

Ananthapadmanabhan and Somasundaran, 1985; Rao, 1988; Dho and Iwasaki, 1990].

Encouraging results were obtained in both batch and continuous pilot tests conducted by

the United State Bureau of Mine, USBM, [Davis et al., 1984]. After 1 lb/t sodium silicate

was added in each of three cleaning stages, a phosphate rougher was upgraded to about

30.7-31.3 % P205, 0.6-1.3 % MgO with a P205 recovery from 55.1 to 89.7. The original

matrix contained 5.47-10.2 % P205, and 1.9-7.6 % MgO.

Rao [1989] studied the effects of major flotation parameters on phosphate grade

and recovery using sodium silicate as a calcite depressant. Maximum flotation of apatite

occurred at pH 8, and phosphate recovery decreased with increasing pH. Sodium silicate

at a concentration below 0.0013 M had no depressive effect on calcite. However, at pH

values above 10, calcite depression was observed even in dilute silicate solution. A

concentrate with 25.6 % P205 at 80 % recovery was obtained.









Efforts to concentrate high-dolomite Florida phosphate ores by flotation, using

conventional fatty-acid/fuel-oil flotation [Liewellyn, 1982], showed that sodium silicate

depression on dolomite was not enough to obtain a concentrate less than 1 % MgO. This

may be due to the presence of silica in this ore in addition to dolomite.

Reverse flotation

Depression of phosphate: This method was used when the apatite was more

abundant than calcite and dolomite (as in most phosphate ore). Among depressants used

are as follow:

Fluosilicic acid: USBM investigated the removal of carbonate from phosphate

rock using fluosilicic acid as a depressing agent and tall oil-fatty acid as a collector for

carbonate mineral flotation (Figure 1-2) [Rule, 1982; 1985]. At collector dosage 0.7 kg/t,

0.45 kg/t depressant, a concentrate of 25.5 % P205 and 0.84 % MgO with P205 recovery

89.5 % was obtained from feed with 20.8 % P205 and 1.1 MgO %.

Abdel Khalek [2000] studied the phosphate depression using fluosilicic acid, but

recovery using this process was less than 60 %. Ataly [1985] pointed out that for

depressing Turkish ores fluosilicic acid was not efficient as other depressants because of

the low recovery (57 %) obtained using fluosilicic acid.

Acid depressants: Orthophosphoric acid was found to decrease adsorption of the

fatty acid on apatite because of the increased surfacial solubility of apatite.









Phosphoric acid also was used to depress apatite (by the USBM, in separating

dolomite from a southern Florida ore) [Liewellyn, 1982]. The produced concentrate's

P205 % did not exceed 28 %, and the MgO % was not less than 1.9 %.

High MgO Pebble


Hammer Mill


Screening I + 28 Mesh
28 Mesh


Desliming
150 Mesh I- -150 Mesh Fines


NaOH Scrub


Desliming -150 Mesh Fines
150 Mesh


Fatty acid/Fuel Oil I ..


Dolomite
Rougher Tail


Dolomite
Cleaner Tail


Figure 1-2. Simplified flowsheet of USBM direct flotation


Sodium Silicate









Lawendy [1993] found phosphoric acid to be the best among conventional

depressants of phosphate. The concentrate was 32-36 % P205, 1 % MgO at 70-75 %

overall P205 recovery from feed 16-25 % P205 and 4-11 % MgO. In other study [Hsieh

and Lehr, 1985], diphosphonic acid was used to depress phosphate (TVA process),

(Figure 1-3), with fatty acid collectors for beneficiating an ore containing 72 % SiO2,

1.05 % MgO and 6.3 % P205. The resulted concentrate contained 73.3 % SiO2, 0.7 %

MgO and 6.4 % P205.

Although sulfuric acid is a phosphate depressant, its ability to attack phosphate and

its high need for pH control limits its usage [Elgilani and Abouzeid, 1993]. Jacob

Engineering [Gruber et al., 1987] used sulfuric acid as a pH regulator to completely

depress phosphate at pH 5.5.

Phosphates: Rao [1979] found that the depressing action of dipotassium hydrogen

phosphate is pH dependent. Phosphate can be depressed at pH 5-8, while carbonate can

be depressed only at alkaline pH; which means that selective flotation can be obtained.

However, the concentrate produced by this process contained no more than 24 % P205.

Sodium chloride: The effect of sodium chloride and other salt addition on

dolomite and apatite flotation has been studied [Rao, 1985]. It was shown that sodium

chloride acts as apatite depressant during flotation at acidic pH values with oleic acid,

while dolomite flotation is not affected. It was, however, observed that dolomite particles

larger than 48-mesh are difficult to float and they tend to remain in the sink fraction.

Trends observed during Hallimond cell tests of single and mixed minerals were

confirmed with mixtures of minerals and a natural ore at bench-scale level. Under more









favorable conditions in the bench-scale tests, MgO content of the mineral mixtures was

reduced from 5 % to 1 % or less, with at least 90 % recovery of P205 in the sink fraction.


High MgO Pebble


+ 48 Mesh
-400 Mesh Fines



-400 Mesh Fines


Float Dolomite Tail


Diphosphonic acid/
Oleic Acid


Dodecylamine
Hydrochloride


Float Silica Tail


Figure 1-3. Flowsheet of TVA's diphosphonic depressant process

Alizarin Red S (ARS): ARS is an organic depressant. It is used to depress either

phosphate or carbonate, depending on pH. ARS can depress calcite impurities at a pH 8-

10, while apatite can be depressed at a pH below 6 [Fu and Somasundaran, 1986; Xiao

and Somasundaran, 1989].

Cationic Flotation

The IMC-F process (Figure 1-4) is a cationic-flotation process for silica-free ores.

The ore is reagentized with a cationic collector aminess or their salts) and the dolomite










separated from phosphate at a rougher stage followed by one or two phosphate-cleaning

steps [Snow, 1979; 1982; Lawyer and Snow, 1980]. By using IMC-F process, a

concentrate of 31 % P205, 0.86 % MgO from feed of 2 % MgO with 83 % P205 recovery

can be obtained.


High MgO Pebble


Acetic acid/Amine


Frother





Phosphate Conc.


- -150 Mesh Fines

--- Silica Tailings









Dolomite
Rougher Tail

Dolomite
Cleaner Tail


Figure 1-4. Flowsheet for IMCF cationic flotation

Nonconditioning Flotation

A research contract between FIPR and Mineral Resources Institute (MRI) at the

University of Alabama aimed to develop a selective flotation process (Figure 1-5) to

separate dolomitic limestone from Florida phosphate ore [Hanna and Anazia 1990]. This

process depends on the difference in adsorption rate of fatty acids on both phosphate and









carbonates. Fatty acids adsorb more rapidly on carbonate surfaces than on phosphate

surfaces. At slightly acidic pH (5.5-6.0), carbonates floated immediately on addition of

collector and brother. After carbonate froth collection, the pulp is conditioned briefly to

float phosphate thereafter with no more addition of collector.


High MgO Pebble


screening I +35 Mesh
35 Mesh


Desliming
150 Mesh Io -150 Mesh Fines


Scrubbing --- Silica Tailings
35% solids

Desliming -
325 Mesh -325 Mesh


Fatty acid/Pine oil
Conditioning


Dolomite Dolomite
Flotation Tail

Sodium silicate/Fatty Condi
acid Conditioning
acid

Phosphate Conc.
Flotation
pH 4-7 Silica Tailings

Figure 1-5. University of Alabama pebble processing, simplified flowsheet









Two-Stage Conditioning Process

This process was investigated through FIRP university of Florida project, in 1983

[Moudgil, 1988]. The process involves conditioning the feed at pH 10, followed by

reconditioning at a lower pH before flotation, Figure 1-6. Selective flotation of dolomite

from apatite was observed for both single and mixed minerals, by reconditioning at pH 4.


High MgO Pebble
65x150 Mesh Feed
Basic pH
Modifier/Fatty Condit
acid Conditioning
pH=10



Flotation 1 Silica (Sink)
PH 9_10


H SO
2 4 Reconditioning




Frother Dolomite Flotation Float Dolomite
pH 4_5


Sink (Apatite & Fine Quartz)


Further processing
to remove silica


Figure 1-6. Flowsheet for University of Florida two conditioning stages

To understand the mechanisms of observed selective flotation, further studies were

conducted involving electrokinetic behavior, oleate adsorption, infrared spectroscopy,









and solubility of the minerals. Selective flotation of dolomite, by reconditioning at pH 4,

is attributed to the combined effect of higher oleate adsorption on dolomite and

hydrolysis of the adsorbed oleate molecules to oleic acid at lower pH values.

The method has been tested in the laboratory with several dolomite-apatite

mixtures and natural magnesium phosphate samples from the Florida phosphate field.

Reductions of the MgO content of the samples from 1.8-4.0 % MgO to below 1 % MgO

at recoveries of about 90 % P205 in the sink fraction have been obtained in bench-scale

flotation tests. Reconditioning pH was determined to be one of the most important

parameters in this process.

The CLDRI Process

In China, over 90 % of the phosphate resources are dolomitic, with MgO content

ranging from 2 % to over 10 %. For economical and rational use of domestic phosphate

resources, China Lianyungang Design and Research Institute (CLDRI) has conducted

studies for many years and developed some flotation technologies for removing MgO

impurity from phosphate ore. These flotation technologies have successful industrial

applications [El-Shall et al., 1996].

Since 1997, CLDRI has conducted extensive studies on Florida high dolomitic

pebble with the objective of using CLDRI's fine particle flotation technology to recover

phosphate (Figure 1-7). In 2000, pilot scale testing (400 lb/h) was carried out jointly by

CLDRI, Jacobs engineering group, and IMC phosphate's company at Jacob's pilot plant

facility. Preliminary economic and technical analysis by Jacob's engineering (based on

pilot plant test results) showed that CLDRI's fine-particle flotation technology could be

used economically and technically for processing Florida's dolomitic phosphate pebbles

[Geo et al., 2002].






































































TO POND
OFloaw OF~
CYCLONES IV THICKENER -
U'Fow |

I ------ ----- I Of Flow
FILTER |io

STOPOND
FINAL CONCENTRATE


Figure 1-7. Simplified flowsheet for CLDRI dolomite flotation









The CLDRI's fine-particle flotation technology (Figure 1-7) utilized for processing

Florida high-dolomitic phosphate pebbles has two steps. The first step is grinding for

liberation and second nondesliming after grinding, which improves flotation recovery and

use of Florida resources, and lowers production cost.

Acid Leaching

Strong and weak acids were used in the acid leaching process. In strong-acid

leaching processes, sulfuric acid was used to completely dissolve the dolomite from

phosphate ore.

In the early 1970s, the USBM investigated magnesium impurity removal from

phosphate concentrates by acid leaching [Rule, 1970]. The MgO level was reduced to

less than 0.1 % by using dilute sulfuric acid, but the phosphate loss was about 6 % P205.

Weak-acid leaching, using weak organic acids, was also studied. Pilot plant studies

[Abu El-shah, 1991] using 6.7 % acetic acid solution could remove more than 80 %

MgO. The advantage of using weak organic acids is that (unlike strong acids), they do

not attack other minerals. The overall disadvantage is that acid leaching, in general, needs

fine particle sizes produced by crushing and grinding which means more expenses.

Physical Separation Methods

Heavy Media Separation

The only commercial process to separate dolomite from phosphate is heavy media

separation [Wiegel and Hwang, 1984]. Heavy media cyclones are used in IMCF mines

for processing the high-dolomite portion of the ore. The concentrate with 1.2 % MgO was

obtained from feed that has 3 % MgO.









Sizing, Attrition Scrubbing, and Desliming

The dolomite is more friable than phosphate rock. Thus, dolomite can be separated

from phosphate by discarding the fine fraction (mainly dolomite) by scrubbing or sizing

[El-Shall and Bogan, 1990].

Problem Statement

The various beneficiating techniques discussed have various limitations or

disadvantages. In acid leaching, strong acids can attack phosphate as well as dolomite

and can be associated with environmental problems associated with this process.

In flotation processes for high dolomitic phosphate, because they could not reduce the

MgO % below 1 % are not applied commercially.

In CLDRI fine flotation process, by which the MgO % could be reduced to less

than 1 %, concerns are as follows: handling of fine concentrate (dewatering and

transport), capital cost (grinding and multiple process steps), and operating cost:

increasing the MgO % in the phosphate rock increases reagent consumption, and requires

grinding, thus increasing energy consumption and plant operating cost. The only process

with commercial application was heavy media separation (HMS), which also suffers from

high MgO % (more than 1 %) in the concentrate in addition to about 50 % phosphate

loss.

Novelty and Scope of Study

We sought a process that separates dolomite from phosphate pebbles resulting in a

phosphate concentrate with less than 1 % MgO. Since the dolomite is mostly in the

pebble fraction and not attached to phosphate, the idea came from well-known reaction

between carbonates and acid producing carbon dioxide gas. Dolomite, as a carbonate









mineral, generates CO2 when exposed to a slightly acidic solution, as shown in the

following chemical reaction:

Dolomite + Acid salt + CO2

Capturing the generated CO2, at the dolomite particle surface, into bubbles using

coating agent (which is a kind of surfactant or polymer) can selectively float dolomite,

and separate it from phosphate rock. Since this process depends on chemical reaction

between dolomite and acidic medium to form a stable bubble at the particle surface and

then to float it, this process is called the Reactive Flotation (RF) process.

The RF has the following requirements (1) presence of acidic medium at low

percentage, and (2) Presence of surfactant that: attaches to the dolomite, is permeable by

acid, allows formation of stable bubbles at the particle surface, is impermeable to CO2

gas, and is stable enough to float the particle to the surface.

Goal of Study

The goal of this study is to develop and to optimize a novel and an effective

process to separate the dolomite from phosphate rock. This process will not only reduce

the MgO % to less than 1 % but also increase the efficiency of the subsequent processes

in phosphoric acid production such as filtration and consequently reduce the phosphate

losses. In addition to develop this novel process, a fundamental study needs to be done to

elucidate Mechanism(s) involved in this novel process.















CHAPTER 2
MATERIALS AND METHODS

To achieve the goals of this research different measurements and techniques are

needed to elucidate polyvinyl alcohol (PVA)/ particle interactions, and bubbles formation

and stability as well as its modeling the following strategy will be followed, Figure 2-1.


SContact Angle Surface Tension


Adsorption Isotherm Work of Adhesion FTIR Zeta Potential


Film thickness

Elasticity (Dynamic -
surface tension) Acid Diffusion

CO2 formation rate Bubble
formation
model
Density balance Time required to
start particle flotation

.............................................................................................................................................................. I ..........................................................................................................................................................
Density change


Figure 2-1. Strategic plan to achieve research goals

As it can be seen form Figure 2-1, the first block represents the required

techniques and measurements to elucidate polyvinyl alcohol (PVA)/ particle interactions.

On the other hand, the second block represents another part of our goal which is bubble

formation and stability as well as its modeling.









Flotation Study

In this study, screening and central composite statistical experimental designs are

used to determine the effect of various factors on the removal of MgO and its recovery.

The Plackett-Burman design was used to screen out which of the 7 factors being

considered were the most statistically significant. In other words, it is a tool that helps

determine which factors made the greatest impact on the MgO removal. A central

composite design was used to determine if there were any interactions between the three

most significant factors determined from the Plackett-Burman design.

Screening Design (Plackett-Burman Design)

The variables examined in the screening stage were acid concentration (1% and 3

%), Polyvinyl alcohol concentration in the coating solution (1 % and 3 %), the PVA

molecular weight (60,000 and 150,000), the degree of hydrolysis (88 %-99 %) of the

PVA, particle size (-0.5 mm and 5-9 mm), and drying time of the coating (0 and 2 hrs).

The levels were chosen according to the exploratory test results from phase I of this

project. The Plackett-Burman design used for this study had 11 factors and 12

experimental runs (Table 2-1). MgO % was determined using chemical analysis.

Chemical analyses were performed using wet chemical methods, acid attack (digestion)

method as described below.

Central Composite Design

A central composite design was used to further investigate the three most

significant factors (acid concentration, PVA concentration, and particle size) determined

from the screening design. Five levels were used. The central composite experimental

design had 17 experimental runs (Table 2-2 and Figure 2-2). For each run, the products

are chemically analyzed as above-mentioned and the MgO % and MgO removal % were










calculated for all the experimental runs. The computer program, Design expert (SAS

Institute, Inc.), was used to perform the statistical analysis for the central composite

design. The data was analyzed by fitting the response variables to a second-order

polynomial model.

Table 2-1. Experimental Plackett-Burman design for 11 variables

Run No. Variable Levels
1 2 3 4 5 6 7 8 9 10 11
1 + + + + + +
2 + + + + + +
3 -
4 + + + + + + -
5 + + + + + +
6 + + ++ + +
7 + + + + + +
8 + + + + + +
9 + + + + + + -
10 + + + + + +
11 + + + + + +
12 + + + + + +

Plackett-Burman Variables Used in the Reactive Flotation Study
Variables Low Level (-) High Level (+)
1) Acid conc 1 % 3 %
2) PVA conc 1 % 3 %
3) dummy *
4) dummy
5) PVA M.W 60,000 150,000
6) Degree of hydrolysis 88 % 99%
7) Time of drying 0 2 h
8) dummy
9) Particle size -0.5mm 5-9 mm
10) dummy
11) dummy
dummy variables are used to calculate the variance of the experimental
runs


















-_X
*--------------






Figure 2-2. Five-level, rotatable central composite design. The X, Y, and Z axes
represent the 3 variables, sulfuric acid concentration, PVA concentration, and
particle size. The circles represent each experimental run.

Table 2-2. Rotatable central composite design with 5 levels and 3 variables
Run No. Acid conc PVA % Particle Size


-1
-1
-1
-1
+1
+1
+1
+1
-1.68
+1.68
0
0
0
0
0
0
0


-1
-1
+1
+1
-1
-1
+1
+1
0
0
-1.68
+1.68
0
0
0
0
0


+1
-1
+1
0
0
0
0
-1.68
+1.68
0
0
0


Levels
Variables -1.68 -1 0 +1 +1.68
Acid conc. (%) 0.32 1 2 3 3.68
PVA, (lb/t) 0.6 2 4 6 7.4
Particle size, mm -0.5 0.5-1 1-2 2-5 5-9









Flotation Procedures for Screening and CCD Designs

Using 300 ml (3 % H2SO4) in 400 ml beaker, particles were pre-coated by spraying

3 % PVA solution (using a mist delivery nozzle) then added to the acidic solution. The

experiment was left for 5 min. Dolomite floats to the surface, which was separated by

decantation on a screen. The products (floated and unfloated fractions) were dried,

weighed, and chemically analyzed.

Factorial Design for Conditioning with PVA before Flotation

This design was used to study the factors affecting the conditioning step with PVA,

which is a promising step for reducing PVA consumption. Table 2-3 shows the design

and levels used.

Table 2-3. Factorial design for conditioning with PVA
Run No. PVA Pulp density Conditioning time, min
concentration, % (S/L ratio)
1 -1 -1 -1
2 -1 -1 +1
3 -1 +1 -1
4 -1 +1 +1
5 +1 -1 -1
6 +1 -1 +1
7 +1 +1 -1
8 +1 +1 +1
9 0 0 0
10 0 0 0
11 0 0 0

Levels
Variables -1 0 +1
PVA concentration, % 0.25 0.625 1
Pulp density 30 % 50 % 70 %
Conditioning time, min 2 11 20

Flotation Procedure for Conditioning Statistical Designs

50 gm of ore were added to 1 % PVA solution in a 600 mL beaker according to

specific solid/liquid ratio. The slurry was mixed for required time at 300 rpm (using 2.5









cm magnetic bar). After the conditioning time is elapsed the solids were screened, the

solids were used for flotation in acidic solution and the filtrate solution was taken to PVA

analysis to determine the PVA residual concentration.

Chemical analysis: About 0.5 gm of the dried and ground representative sample

were digested in 50 ml of aqua regia by boiling on hotplate until the reaction complete.

After cooling, this solution was filtered through a Whatman 42 filter paper into 1000 ml

volumetric flask. The filter paper and residue were then washed at least five times to

remove all the traces of dissolved salts and acid. The filtrate was diluted with distilled

water and thoroughly mixed and the concentrations were measured using Inductively

Coupled Plasma, ICP, emission spectrometer, (Perkin Elmer Optima 3200RL Optical

Emission Spectroscopy, Norwalk, CT). ICP was calibrated using AFPC rock check

No.22.


Standard sample concentrations were measured from time to time in the course of

the analyses to check the accuracy of the calibration curve. Five replicates were taken and

averaged for each element. The following wavelengths were used:

Element Wave lengths, nm
Ca 315.887, 317.933, 393.366, and 396.847
Mg 279.077, 285.213, and 279.553
P 213.617, and 214.914
Fe 238.204, 239.562, 259.939, and 234.349
Al 308.215

Density and Surface Area Determination

Densities of powders were measured using a Quanta Chrome ultrapycnometer. The

specific surface area (m2/g) of dolomite and apatite particles was determined using

Quanta Chrome NOVA 1200 instrument. (Quantachrome Corp., Boynton Beach, FL).









The analysis was performed using a five-point BET (0.10, 0.15, 0.20, 0.25, 0.30) using

Nitrogen gas. The particles were outgased at 600C for 24 hours before analysis.

PVA Preparation

PVA was kindly supplied by Celanese chemicals Co., USA. Different PVA types

with various ranges of molecular weights and degree of hydrolysis are used in screening

factorial design, namely:

Type Average Molecular Wt. Degree of Hydrolysis, %
Celvol 165 150000 99
Celvol 540 150000 88
Celvol 305 60000 99
Celvol 203 60000 88

In the rest of the study, PVA, Celvol 165, was used. PVA flakes were gradually

dissolved in deionized water at 90 C with continuous mixing by magnetic stirrer and

heated at 90 OC for 2 h to produce different PVA concentrations (0.5, 1, 2, 3, and 4 %).

Spectrophotometric Determination of PVA

PVA solution of certain concentration treated with 25 ml of 4 % boric acid solution

and 2 ml of iodine solution (prepared from 1.27g of iodine and 25 g of KI/1), in that

order. The resulting solution is diluted to 100 ml and kept at 250C and its absorbance

measured at 690 nm against a blank solution containing the same amount of boric acid

and iodine solution. Beer's law applies in the concentration range of 0-40 mg of PVA/1 of

solution [Finch, 1973; 1992]. Absorbance spectra were taken using ultraviolet/visible

spectrophotometer (UV-Vis spectrophotometer, Perkin-Elmer Lambda 800). All spectra

were taken at room temperature.









Static (Equilibrium surface tension)

Equilibrium surface tensions were measured for freshly prepared solutions by the

Wilhelmy plate method at room temperature. The platinum plate was always cleaned and

heated to a red/orange color with a Bunsen burner before use.

Dynamic Surface Tension (DST)

Maximum bubble pressure method

The dynamic surface tension (DST) measurement is important in the dynamic

interfacial processes where the equilibrium surface tension is hard to be achieved. The

DST can be measured by different methods such as: the drop weight [Jho and Burke,

1983; Miller et al., 1993], oscillating jet [Thomas and Hall, 1975], capillary wave

[Kelvin, 1871], growing drop technique [Macleod and Radke 1993], and the maximum

bubble pressure method [Mysels, 1986; Ross et al., 1992]. The maximum bubble pressure

(MBP) technique is the most commonly used technique for the measurement of DST, and

has been applied to a variety of surfactant and polymer solutions [Fainerman et al., 1994;

Tamura et al., 1995]. The physical principle behind the MBP measurement is the Laplace

pressure; the pressure inside a curved liquid interface is higher than the ambient. This

excess pressure (P) can be calculated from the Laplace equation [Adamson and Gast,

1997].

2y
P = 27 + pgh
r

Where y is the surface tension, r is the radius of curvature of the bubble, p is the

liquid density, g is the gravitational constant and h is the depth of the bubble in the liquid.

The first term expresses the Laplace pressure due to the curved gas/liquid interface, and

the second term is the hydrostatic pressure due to the liquid height above the forming









bubble. The first term will vary during the life cycle of the bubble, while the second term

will remain constant. Figure 2-3 shows a typical cycle of a bubble and the resulting

pressure within the capillary. After a bubble breaks off from the capillary tip, the pressure

is the lowest. As more gas flows into the capillary, the pressure builds up as the gas is

pushed out of the capillary and the radius of curvature at the tip decreases. During this

expansion process, surfactant is populating the new interface and acting to lower the

surface tension. At the point of minimum interface radius of curvature, where a

hemisphere of gas is formed at the capillary tip, the pressure reaches maximum.


Pressure




Capillary
immersed in
.,e liquid


Point of PM U


Figure 2-3. Characteristic bubble-pressure vs. time curve in MBP.

Both the minimum radius of curvature and the surface tension, as described by the

Laplace equation, govern the maximum pressure experienced. On the incremental

addition of gas, the bubble expands out of the capillary tip and the radius of curvature

gradually increases. This results in a drop in the Laplace pressure, and a resulting rapid

expansion of the bubble. At some point, the bubble breaks off from the capillary tip and

the whole process starts again.










Dynamic surface tension setup

The maximum bubble pressure apparatus was constructed using a differential

pressure transducer purchased from Omega Engineering, Inc. (Stanford, CT), with a

sensitivity of 0 to 10 in. (25 cm) H20 (0 to 2500 Pa). A #23 steel needle was used as a

capillary, with nominal 0.025 in. (0.64 mm) external diameter, 0.013 in (0.33mm)

internal diameter, and a flush cut tip. The capillary diameter was chosen so that the

various resistance of water to bubble growth could be ignored. Such internal and viscous

effects are a potential source of error in these measurements that need to be taken into

consideration [Garrett and Ward, 1989]. All measurements were conducted with the

capillary tip 1 cm beneath the liquid surface. CO2 gas was used as the bubbling gas to

simulate the gas produced in the reaction of acid with dolomite and an oscilloscope

connected to the pressure transducer was used to determine the bubble frequency and the

dynamic surface tension (Figure 2-4).




Rotameter
Pressure Voltage
Pressure
___ -'- Transducer

Oscilloscope
1.0 cm depth
A ir F low........................... .

Polymer Bubble formation
Solution

Figure 2-4. Setup for the measurement of dynamic surface tension by maximum bubble
pressure method.

Contact Angle

The angle formed at a point on the line of contact of three phases, of which at least

two are condensed phases, by the tangents to the curves obtained by intersecting a plane









perpendicular to the line of contact with each of three phases. One of the phases must be

a liquid, another phase may be solid or liquid and the third phase may be gas or liquid.

In other words, it is the angle included between the tangent plane to the surface of

the liquid and the tangent plane to the surface of the solid, at any point along their line of

contact. The surface tension of the solid will favor spreading of the liquid, but this is

opposed by the solid-liquid interfacial tension and the vector of the surface tension of the

liquid in the plane of the solid surface.



Gas


Ys Liquid
Solid


Figure 2-5. Contact angle and its relation with surface tension

Equilibrium wetting properties of liquids and solutions can be interpreted in terms

of the balance of surface tensions (Figure 2-5).

Ysg -Ysl= Yg COS 0

Measurement of Contact Angle

The contact angle goniometer, Figure 2-6, is a device mounted on the optical bench

to examine a single liquid drop or gas bubble resting on the smooth, planar, solid

substrate. The drop/bubble is illuminated from the rear to form a silhouette, which is

viewed through a telescope connected to computer. The image of the drop/bubble appears

on the computer monitor by which the brightness and contrast, focusing and amount of

illumination can be adjusted. After adjusting the image the contact angle can be

measured.









microscope videocamera


light source



frame-
grabber
motor-driven syringe computer

Figure 2-6. Contact angle measuring apparatus

Surface Characterization by FTIR

The spectorometer used for FTIR was a Magna-IR760 spectrometer by Nicolet in

transmission setting. As for the spectra recorded in transmission mode, KBr pellets were

prepared for the background measurements. Another pellets were prepared by mixing

about 1 mg of filtered powder with 50 mg of KBr. Transmission spectra were recorded

using a DTGS detector, with 128 scans and 4 cm-1 resolution.

Zeta Potential Measurements

In the zeta potential measurement tests, 1 g of ground mineral samples was added

into a 250 ml beaker in which 100 ml 0.003 M KNO3 solutions were added. The

suspension was conditioned for 3 min during which the pH was adjusted, followed by 5

min of conditioning after adding the polymer. It was then allowed to settle for 3 min, and

about 10 ml of the supernatant was transferred into a standard cuvette for zeta potential

measurement using a Brookhaven ZetaPlus Zeta Potential Analyzer. Solution temperature

was maintained at 25 C. Ten measurements were taken and the average was reported as

the measured zeta potential.

Adsorption

These studies were carried out at 250C using 0.5gm mineral sample in 50 ml

solution having the desired PVA concentration and the ionic strength fixed at 3X10-3 M









using KNO3. One set of experiments was performed first to determine the effect of

conditioning time at a fixed pH (pH 7) and PVA concentration was 100 ppm (3.33X10-8

M). Then set of experiments was also carried out as a function of pH at the same PVA

concentration. Adsorption isotherm was also determined for PVA/dolomite, PVA/apatite,

PVA/dolomite/apatite supernatant, PVA/apatite/dolomite supernatant systems at pH 7

and 3X10-3 M KNO3 as an ionic strength. After the conditioning step, the suspension

was centrifuged at 3000 rpm for 5 min using an IEC-Medispin centrifuge (Model 120).

Then a volume of the clear solution was withdrawn for measurement of the PVA residual

concentration.

In adsorption kinetic experiments, a 0.5 gm of dolomite or apatite were added to

100 ppm (3.33X10-8 M) PVA solution a 30 ml screw-capped bottle and was shaken at

room temperature using Wrist hand shaker for a given recorded time (0.05, 0.083,

0.0166, 0.25, 0.5, 1, 2, 4, 6, 8, 10, 16, 20, 24 h). After that the contents of each bottle

were allowed to settle, followed by centrifugation step using an IEC-Medispin centrifuge

(Model 120). The amount of PVA adsorbed was determined by difference between the

initial and final concentrations measured spectrophotometrically with a Uv-vis

spectrophotometer (UV-Vis spectrophotometer, Perkin-Elmer Lambda 800). These

experiments were used to establish the equilibrium time for adsorption of PVA on

dolomite and apatite. This equilibrium time was used as contact time in adsorption

isotherm experiments.

Dissolution Kinetic Experiments

Dolomite specimens (1-2 mm and 0.2-0.3 mm), uncoated and coated with PVA,

were added to 2000 ml aqueous medium (3 % H2SO4) at 23 + 0.50C to carry out 120 min

static dissolution tests. After a definite time interval, solution samples were collected









periodically and concentrations of Mg2+ and Ca2+ were determined with ICP. The

relations of concentration, C, with time, t was plotted to explore the dissolution kinetics

of the dissolution process.

Pilot Scale Experiments
The Sluice

The pre-conditioned material is introduced into a feed box, which contains 3 %

sulfuric acid. The reaction of sulfuric acid with particles starts in the feed box forming

bubbles around dolomite particles and creating the required density difference for the

separation in that device. The particles soon enter the moving stream and, because of the

bubbles on dolomite particles, they settle at lower rate and float to the surface when the

enough bubbles formed and carried away till they reach the lip of the trough and

discharged with liquid overflow. Since phosphate particles do not produce C02, and have

no bubbles at their surfaces, they are collected at the bottom of the trough (Figure 2-7).


Pre-conditinned Feed
3% Sulfuric acid


O Dolomite Reiect F
O Feed Box
O O
* Phosphate
Trough angle
Inclination angle to horizontal
Concentrate
Figure 2-7. Schematic diagram for separation of dolomite from phosphate in sluice

Sluice Experiments

250 gm of ore (a mixture of dolomite and phosphate), with particle size 1-2 mm,

were conditioned in 1 % polyvinyl alcohol solution for 5 minutes, then screened to

separate the solids from PVA solution. The screened solution was analyzed to determine

the actual PVA consumption by depletion method. Or the material was sprayed with the






31


equivalent amount of PVA and was tumbled to attain a good coating. After that, the

coated solids were poured into sluice feed box which the sulfuric acid and the reaction

starts to generate CO2 bubbles. The particles move through the sluice at flow rate 125

ml/sec (2 GPM). The floated fraction, dolomite, was collected on 0.5 mm screen and the

sink fraction, phosphate, was collected in the sluice trough. The floated and sink products

were analyzed for MgO, P205, CaO, and insoluble residue (I.R.).














CHAPTER 3
TECHNICAL ASPECTS OF REACTIVE FLOTATION: PRELIMINARY TESTS

Introduction

The reactive flotation process, as mentioned earlier, requires certain conditions to

give good results and separate dolomite from phosphate rock. The main requirements for

RF to work properly are the acidic medium and surfactant or polymer, which attaches to

the dolomite, to allow the acid to diffuse through but at the same time remain

impermeable to the CO2 gas produced by the reaction. Moreover, the formation of CO2

will lead to stretching of the membrane-like surfactant layers, which requires an elastic

behavior of that membrane. The elasticity of the membrane formed around the particle

will then allow the formation of stable bubbles. The stable bubbles are the main criterion

in RF to float the dolomite particles and separate them from others.

At the starting point of this process, the most well behaved acids and surfactants,

under RF conditions, were not known. Different acids and surfactants can be used but it

was unknown which one would best fulfill the previously mentioned requirements for the

RF process. For that reason, in the current study, preliminary tests were conducted using

different acid types (sulfuric, hydrochloric, nitric, acetic) within a particular

concentration range (0.5 4 % v/v) as well as different types of surfactants (anionic,

cationic, and non-ionic), which were prepared as 1-2 % surfactant solution and used in

two different ways: (1) surfactant was added to acidic solution and/or (2) surfactant was

used to coat the particle before adding the particle to the acidic solution. Also, poly-vinyl-









alcohol (PVA) was used in this study. Vegetable and corn oil were used at the acidic

solution/ air interface and oil height was 0.5-0.6 cm.

List of Potential Surfactants: Surfactants Screening

A list of potential surfactants used in this study, are given in Table 3-1. The list

contains different types of surfactants, (i.e., cationic, anionic, and non-ionic, as well as

polymers).

Table 3-1. Surface-active agents used in RF tests
Surfactant Chemical Name Type
Arquad 12-50 Dodecanaminium+isopropyl alcohol Cationic
Lilamine N-3-aminopropyl-N-tallow alkyltrimethylenedi- Cationic
Armac HT Amines, hydrogenated tallow alkyl acetates Cationic
Armid HT Amides Cationic
Armeen 12 D dodecanamine Cationic
Ethofat Tall Oil Anionic
SDSO3 Sodium dodecyl-sulphonate Anionic
SDSO4 Sodium dodecyl-sulphate Anionic
Oleic acid Oleic acid Anionic
Arlacel Sorbitan sesquioleate Non-ionic
Span-20 Sorbitan monolautrate Non-ionic
G-1096 Polyoxyethelene-sorbitol hexaoleate Non-ionic
Brij-97 Polyoxyethelene oleyl ether Non-ionic
Arlatone T Polyoxyethelene polyofatty acid ester Non-ionic
Tween-21 Polyoxyethelene Sorbitan monolautrate Non-ionic
Tween-85 Trioleate Non-ionic
PVA Poly-Vinyl-Alcohol Non-ionic

Initial Screening Tests

A 100-pound sample of high-dolomitic phosphate oversize pebble was obtained

from the IMC Four Corer's mine. This is not "normal" high MgO rock, but the

oversized high MgO rock that is removed from pebble. As such, neither the screen

weight fractions nor the chemical analyses are representative of pebble. However, the

rock is representative of the high-dolomitic rock that needs to be separated. The rock was

screened, and each fraction is chemically analyzed in Tables 3-2 and 3-3.









Table 3-2. Size analysis of the dolomitic phosphate sample
Size, mesh Size, mm Wt. % % Cumulative % Cumulative
(passed) (retained)
+ 3/8 in +9.5 11.62 100.00 11.62
- 3/8 in +3.5 9.5+5.6 65.36 88.38 76.98
-3.5+6 -5.6+3.36 18.14 23.02 95.12
-6+9 -3.36+2 2.05 4.88 97.17
-9+16 -2+1 1.6 2.83 98.77
-16 1 1.23 1.23 100
100.00

Table 3-3. Chemical analyses of the dolomitic phosphate sample
Size, mesh P205 % CaO % MgO % Fe203 % A1203 %
+3/8 in 19.86 32.23 4.02 0.93 0.52
-3/8 in +3.5 21.25 36.34 3.43 1.15 0.84
-3.5+6 17.09 27.92 2.34 1.32 0.86
-6+9 24.32 38.90 2.75 1.08 1.17
-9+16 20.42 28.71 2.12 1.27 1.07
-16 13.62 22.75 2.44 1.13 0.95

Since most of the rock was relatively coarse, a portion of it was crushed and re-

screened to obtain smaller fractions that could be used in the tests.

Specific Tests

The results of some preliminary testing are shown in the following paragraphs in

terms of acid type and dosage, surfactant type and dosage, particle size, and using of oil

layer at the water/acidic solution interface.

Acid type

Different acids (H2SO4, HC1, HNO3, and Acetic acid) were used. Inorganic acids:

HC1, HNO3, and H2SO4 gave the same results at 2 % acid concentration. On the other

hand, the acetic acid (RCOOH) formed a very small number of bubbles at 2 % acid

concentration. Increasing the concentration of acetic acid to 5 % gave no significant

variation in bubble formation, which means that the organic acids are weak acids.

Consequently, a large amount of organic acids is needed relative to inorganic ones.









Acid dosage

Since all the inorganic acids behave in the same manner, sulfuric acid was used in

the rest of the experiments. Table 3-4 shows the effect of acid concentration on the

bubble formation rate, amount of bubbles formed, and particle floatability that were taken

as a criterion to compare the results. The surfactant used for coating the dolomitic particle

was dodycyl sulphonate (SDS03).

Table 3-4. Effect of sulfuric acid concentration
Sulfuric acid concentration, %
0 0.25 0.5 1 2
Bubble Very slow Slow Fast IRpid
formation rate

Amount of Very small The interval Increasing the Continuous Continuous
bubbles number of between formation rate and formation of and quick
formed bubbles (about 5 of two bubbles number of bubbles but at formation of
bubbles for 10 was about 30 bubbles rate higher than many
minutes then no seconds formed 0.5 % bubbles
bubbles formed)

Flotation of No flotation Float after 5-10 Float after 1
dolomite minutes minute
particle

From Table 3-4, we notice that the reaction between dolomite and acid to produce

CO2 decreases noticeably with decreasing acid concentration and the flotation of

dolomite particle takes longer time (5-10 min) in 1 % acid than in 2 % acid.

Consequently, the 2 % sulfuric acid concentration is the most suitable acid concentration

for bubble formation in terms of both bubble formation rate and amount of bubble

formed. Because of the wide gap between 1 % and 2 % acid concentration another set of

experiments were conducted as shown in Table 3-5 to get the optimum acid

concentration.










Table 3-5. Effect of sulfuric acid concentration on the bubble formation


Sulfuric acid concentration, %
1 1.2 1.4 1.6 1.8 2


Bubble Increases
formation rate

Amount of
bubbles Continuous and small No. of tiny bubbles Continuous & large No. of tiny bubb
formed

Flotation of Takes 20 minutes to Floats and stays at water/oil Floats and stays
dolomite float, not stable (particle (W/O) interface for 5 min. Then (W/O) interfac
particle goes up and down) oscillates within 2 min intervals

Table 3-5 shows that the flotation at 1 % & 1.2 % is not stable and flotation

occurs after the particle loses part of its weight due to reaction with acid. Accordingly,

this concentration deals with a small particles (smaller than 1mm) rather than large

particles. The bubble formation rate is decreased with the acid percentage from 2 % to 1

%, but as the difference was narrower between 1.6-2 %, 1.6 % acid concentration was

found to be optimal. The effect of acid concentration on the RF criterion was

schematically summarized in Figure 3-1.


les


at
e


A B C
Bubble Number of Flotation
formation bubbles time
rate formed




Acid conc. Increases Acid conc. Increases Acid conc.

Figure 3-1. Effect of acid concentration on: A) Bubble formation rate, B) Number of
bubbles formed, C) Flotation time.

Surfactant type

Table 3-6 shows the anionic, cationic and non-ionic surfactants used. According to

RF criterion, the performance of Anionic surfactant was superior to Non-ionic, which

was superior to Cationic. Anionic surfactants, especially SDSO3, formed stable


Increases


\










"membranes" that captured CO2 and persisted for 10 minutes. This may be due to its

ability to work at low pH. Therefore, SDSO3 was used in the next experiments.

Table 3-6. Preliminary tests of high dolomitic phosphate with different surfactants
Surfactant Name Remarks
Arquad 12-50 (1%) stable bubbles for 20-30 sec, particle 2-3 mm can float

Lilamine (1%) less stable bubbles, stay 10 s, particle does not reach to surface of the solution
(air-water interface)

Armac HT (1%)

Armid HT (1%) surfactant precipitated in the solution
Armi HT 10/o


Armeen 12 D (1%)

Ethofat (1%)
SDSO3 (1%)
(2%)
SDSO4 (1%)
(2%)
Oleic acid


Formation of bubbles on the particle surface then particle floats

Bubbles form and go up to solution surface, but particle doesn't float
Particle takes 2-3 sec to float

Bubbles form and go up to solution surface, but particle doesn't float
Flotation rate in case of SDSO4 < Flot.rate in case of SDSO3

Added as it is, particle stays at surface of the solution (air-water interface) with
formation of many small bubbles.


Arlacel Common observations for non-ionic surfactants:
Span-20 No significant change in behaviour of different non-ionic surfactants used
G-1096
Brij-97 The flotation of high dolomitic particle after 3-5 seconds and stay at the surface,
Arlatone T while the low dolomitic does not float.
Tween-21
Tween-85 Formation of stable bubbles on the floated particle surface

Surfactant dosage

The results of different dosages of surfactant sodium dedocyl sulfonate (SDS) are

shown in Table 3-7 in terms of bare and coated particles. It is indicated from Table 3-7

that bare particles, without coating, forms bubbles quickly and in large quantity due to

rapid dolomite/acid reaction but the bubbles are not stable at the interface where they can

be collected.









Table 3-7. Effect of surfactant concentration (SDSO3, 2%)
SDSO3 (2 %)
2% 4% 6%
Many bubbles formed Many bubbles Many bubbles
Particle Small particle (1-2 mm) formed formed
without floats Particles (1-2 Particles (1-2
coating Large particle (3-4 mm) mm) and (3-4 mm) mm) and (3-4 mm)
goes up and down float float

Particle
coated with Many bubbles formed -small particle floats-large particle does not float
SDSO3

On the other hand, the coated particles take more time to float due to the diffusion

of acid through the surfactant layer around each particle. Also, a greater number of

bubbles are formed when no coating is used. Finally, when coating is not used each

bubble forms as a separate bubble, while formation of bubbles clusters occurs when with

a coating with SDS is used.

Increasing the surfactant concentration increases the flotability of the particles and

large particles (3-4 mm) can float at high surfactant concentrations of up to 4 %; after

that, there is no noticeable change in floatability-(i.e., 6 % is very much similar to 4 %).

The oscillated motion of large particles (3-4 mm) for short period of time (3-5 sec) was

noticed at 2 % surfactant concentration; however, the same size floats without oscillation

at 4 % and 6 %. This means more surfactant is needed.

Particle size

The particle size effect on the RF process was studied for coated and uncoated

particles. The observations of different size ranges were recorded in Table 3-8 which

indicates that the coating affects the stability of the bubbles on the particle surface as well

as the stability of the floated particle at the interface; however, even uncoated particles









release many bubbles and float quickly for very short time due to their rapid reaction with

acid.

Table 3-8. Effect of particle size on the dolomitic particle floatability
Particle Size, mm

0.5-1 1-2 2-4 4-6
Without Particle floats and Particle floats for few No flotation -the particle oscillates
treatment stays at water/air seconds then oscillates without reaching the interface
interface

With Floats and stays at Floats after 30 sec and Oscillates and Floats after 1
treatment interface stays at (w/a) interface finally stays at min then
w/a interface oscillates for 5
min


This rapid rate of reaction causes rapid reduction in particle weight. In addition, it

causes the production of many bubbles at a faster rate than bubbles-particle detachment,

leading to presence of uncoated particle at the interface, although for shorter time than

coated ones. But once the uncoated particle reaches the interface, the bubbles will not be

stable anymore due to high surface tension, which is the reverse of coated particles.

Large size particles need more bubbles; the bubbles formed were not enough to lift

the particle causing the particle to oscillate up and down. On the contrary, the small

particles 1-2 mm and smaller do not oscillate and float directly to the interface.

To improve the stability of the floated particles at the interface, the reduction of the

surface tension is required. Therefore, the oil layer was used at the interface for that

purpose.

Oil-Surfactant Experiments

Addition of an oil layer, either vegetable or corn oil, to the acidic solution/ air

interface was attempted with different types of surfactants to enhance the bubble stability

at the interface, as shown in Table 3-9. The selected surfactants for this set of









experiments were the best ones from Table 3-6. The particle was coated with the

surfactant before its addition to acidic medium.

Table 3-9. Effect of presence of surfactant and oil layer on particle flotation
Surfactant Type Observations
2 % sulfuric Arquad Cationic Formation of bubbles due to dissolution of
acid solution + Lilamin dolomite, but no flotation.
oil layer Ethofat Bubble formation,
SDSO3 Anionic Particle floats to water/oil interface for 10 min
SDSO4 and then comes back to acidic solution.
2 % sulfuric Arquad Cationic Formation of cluster of bubbles on particle
acid solution + Lilamin surface, but no flotation.
1 % surfactant Ethofat Bubble formation and particle oscillates and
SDSO3 Anionic the bubble bursts at water/air interface.
SDSO4
2 % sulfuric Arquad Cationic The same as before
acid solution + Lilamin
1 % surfactant Ethofat Bubbles form No flotation
+ oil layer SDSO3 Anionic Bubbles form on particle surface and particle
SDSO4 floats to water/oil interface then to oil/air
interface

It is clear that, as can be seen in Table 3-9, the oil layer enhance the bubbles'

stability on the particle and at the interface. When the particle floats to the acidic

medium/oil interface, the floated particle stays at the interface for up to 10 min. The

presence of oil layer may add another layer, an oil layer, in addition to the surfactant

layer on the particle surface, although this would depend on the adsorption of the

surfactant to the particle as well as the type of the surfactant and its interaction with oil.

From Table 3-9, it was observed that the presence of a low percentage of surfactant

(1 %) in the acidic solution slightly increases the stability of bubbles in the absence of an

oil layer. The presence of 1 % surfactant in the solution may reduce the fast detachment

of surfactant molecules from the particle surface through reduction of the diffusion of

surfactant molecule to bulk solution.









By using acidic solution in combination with 1 % surfactant and an oil layer (Table

3-9), it was found that the oil layer causes more stable formation of bubbles on the

particle when the particle floats to oil/air interface. The particle does not go back into the

acid solution but stay at oil/ air interface. Consequently, large particles of up to 3 mm

can float in the presence of the surfactant and oil layer.

SDS is a good surfactant under these experimental conditions. However, Ethofat

and cationic surfactants did not work well in the presence of surfactant mixed with acidic

solution. This may be due to the interaction between the oil layer on the particle and

those surfactants possibly causing the formation of a non-flexible surfactant layer, which

cannot capture the CO2 gas and cannot form the bubbles required for flotation.

Poly-Vinyl-Alcohol (PVA) Testing

PVA was used as a viscous polymer to coat and to float the dolomite particles. The

effect of acid concentration, particle size, and different PVA concentration were studied.

Effect of acid concentration

PVA forms a membrane on the particle. This membrane needs a high acid

concentration to penetrate it and to perform the reaction between the acid and particle

surface. It was observed that there were a very small number of bubbles generated.

However, most of the bubbles formed a cluster on the particle surface, which increased

the volume of the bubbles at the surface and floated the particle. The floated particles

were not stable at the beginning, going up and down at interface, but ultimately staying at

the interface for long time (sometimes more than 2 days).

On the other hand, PVA also floated the low dolomite-content particles, which

were able to form bubbles. This observation is due to PVA's very low permeability by

CO2, which makes it the best among the surfactants used. By increasing the PVA









concentration, longer time was taken to form a bubble and to float the particle, which

requires high acid concentration, but in this case it became possible to float large particle

sizes (up to 8 mm).

Table 3-10. Effect of acid concentration
Acid concentration, %
0.5 1 1.5 2
Particle Formation of cluster of bubbles Cluster of bubbles formed
coated with Very low bubble formation rate Low bubble formation rate
PVA No particle flotation (up and down at Particle floats
1.5%)

From Table 3-10, it is evident that the findings were that the PVA forms a

membrane on the particle. This membrane needs a high acid concentration to penetrate it

and to perform the reaction between the acid and particle surface to release the gas, which

is responsible for floating the particle. No quick release of bubbles was observed at low

acid concentrations, but rather only a very small number of bubbles at the beginning;

furthermore, the rate of bubble formation and particle flotation was very slow. After a

while, an increasing number of bubbles appeared and clustered on the particle surface,

which then caused the particle to float.

Moreover, at low acid concentrations, the floated particles were not stable, going

up and down at the interface due to absence of surfactant. Even at high acid

concentrations, the acid takes time to penetrate the PVA membrane and react with

particle surface. For that reason, the bubble formation rate was low. But after the

formation of bubbles began to cover the particle surface with clusters of bubbles, the

particles started to float.

The acid concentrations of less than 2 % were not effective due to the high

viscosity of PVA membranes, which requires a high concentration difference to permeate









the membrane and the reach the particles' surface. Because the previous set indicated that

the 2 % acid concentration was optimal, another set of experiments was conducted at

higher acid concentration (> 2 %) under the premise that a concentration higher than 2 %

would be even better. Therefore, additional experiments were conducted using 2.5-4 %

acid concentrations. Table 3-11 shows the results of those experiments.

Table 3-11. Effect of acid concentration above 2 %
Sulfuric acid concentration, %
2.5 3 3.5 4
Bubble formation Increases with increasing acid conc. %
rate
Amount of bubbles Cluster of bubbles formed on the Rapid reaction no stable
formed particle surface bubbles are seen on the
surface (bubbles released
in high rate from particle
surface)
Flotation of Particle Floats Particle floats Particle oscillates and
dolomite particle Is more stable than Less stable finally goes down
higher acid conc. than 2.5-3 %

There was no flotation at acid concentrations above 3 % due to the high bubble

formation rate and absence of stable bubbles that can attach to the surface of the particle

and float it. The bubble formation rate increases with increasing acid concentration, 4 %

> 3.5 % > 3 % > 2.5 %; however, 2.5-3 % resulted in the most stable bubbles.

Different PVA concentrations (2 %, 3 %, and 4 %) were also used and the

experiments indicated that 3-4 % concentration is most suitable for dolomite flotation.

However, increasing PVA concentration increased the time it takes each particle to float.

Effect of particle size

Increasing the particle size increased the time needed to float the particle. This can

be explained by variation of surface area to weight of the particle ratio, which decreases

with increasing particle size. The results of these tests are shown in Figure 3-2.










30


20


10-


0
-1 1-2 2-5 5-9 +9

Particle size, mm


Figure 3-2. Effect of the particle size on the time needed to float the particle in 3 %
H2SO4 when the particle is pre-coated with 3 % PVA

PVA with acid vs. PVA on the particle surface

By using two beakers, one of them containing uncoated particles added to 3 %

PVA + 3 % acid solution and the other containing PVA coated particles added to 3 %

acidic solution, the comparison between the performance of different application of PVA

was obtained. Table 3-12 shows the results in terms of bubble formation, rate of

formation, and particle flotation.

When uncoated particles were added to the PVA+acid solution, the acid attacked

the particle surface. Consequently, the dissolution of dolomite is higher than in case of

coating the particle with PVA before adding it to the acidic solution. Presence of micro-

bubbles (cotton shape) at the particle surface indicated the dissolution of dolomite.

Bubbles were vertically distributed rather than horizontally in first beaker, while

the bubble distribution was horizontally uniform (spreading on the particle surface) in the

second beaker. The bubble formation rate is higher in the first case than in the second.









The flotation of the particles was better in the second case due to the presence of

relatively large bubbles and uniform bubble distribution. Despite rapid dissolution of

dolomite particles in the first case, flotation of the particles took a long time due to the

small size of the bubbles. In both cases, the particles floated stay at interface for a long

period of time (days).

Table 3-12. Comparison between coating the particle before adding it to acidic solution
and adding the particle to acid+PVA solution without coating
First beaker Second beaker
PVA + acidic solution in the beaker Particle was pre-coated with PVA
and the particle added without coating before adding it to acidic solution
Bubble Tiny bubbles were formed Relatively large bubbles were
formation rate formed

Amount of Cluster of the bubbles grows vertically Cluster of the bubbles grows almost
bubbles and cover parts of the particle horizontally and uniformly
formed surface- no uniform distribution of distributed on the particle surface.
the bubbles on the particle surface.

Flotation of Takes long time (hours) to float Takes short time (min) to float
dolomite
particle

Preliminary Testing Using PVA Coated Particles

In the following preliminary tests, all particles were pre-coated with PVA

Flotation of a Representative Sample

Flotation of a representative sample, a mixture of phosphate and dolomite, was

attempted in a one-liter beaker. The percentage of floated fraction by weight obtained

after 6 hours was about 22 % and un-floated fraction was about 78 %. It was noticed that

the smaller particle sizes (less than 1mm) floated immediately. It was then determined

that using small size fractions of about 1 mm coated with PVA would reduce the flotation

time with very low chance to dissolve the particle; it appears that good PVA coating









prevented rapid acid-attack to the particle. On other hand, we should take into account

that the PVA consumption will increase due to increased surface area.

Fraction of Representative Sample (-4+1.19 mm)

A bucket (11 in diameter and 15 in height) was filled with a solution of 15 liters of

water containing 3 % sulfuric acid and 3 % PVA. The water height in the bucket was

about 1 foot. Most of the particles were in monolayer form to avoid mechanical

entrapment of the particles.

After applying the finalized procedure, the MgO percentage in the representative

sample was reduced to less than 1 % of the sink fraction; P205 percentage was reduced to

less than 4 % (Table 3-13).

Table 3-13. Results of representative sample (- 4+1.19 mm particle size)
Sample Chemical analysis, % Recovery, %
Wt% MgO CaO P205 I.R. MgO CaO P205 I.R.
Float 14.55 13.21 32.56 2.71 6.89 75.1 13.73 2.33 10.01
Sink 85.45 0.75 34.83 19.35 10.55 24.9 86.27 97.67 89.99
Balance 100 2.56 34.50 16.93 10.02 100 100 100 100
Feed 100 2.6 34.78 16.84 10.32 100 100 100 100

Effect of Time on Weight Percentage of Floated Particles

Two samples of- 4+1.19 mm size were gathered at different flotation time (30 min

and 60 min). The samples were immersed in a beaker containing a solution of 3 % PVA

and mixed to ensure the coating of all particles with PVA before adding the sample to

acidic water (3 % sulfuric acid). The same bucket as in the previous test was used and at

the same conditions. Each experiment was observed for two days and the floated fraction

was collected. Most of the particles (about 90 %) floated within 30 minutes and about 10

% in the remaining two days of the experiment. The results are shown in Table 3-14.









Table 3-14. Effect of time on the flotation of- 4+1.19 mm particle size
Time, Chemical Analysis Recovery
min Wt% MgO% CaO% P205% MgO% CaO% P205%
30 Float 16.75 13.62 24.25 0.76 77.84 13.55 0.8
Sink 83.25 0.78 31.12 18.89 22.16 86.45 99.2
Balance 100 2.93 29.97 15.73 100 100 100
Feed 100 3.03 30.38 15.76 100 100 100

60 Float 17.44 13.6 26.17 0.94 78 15.22 1.03
Sink 82.56 0.81 30.8 19.06 22 84.78 98.97
Balance 100 3.04 29.99 15.90 100 100 100
Feed 100 3.03 30.38 15.76 100 100 100

The percentage of MgO was reduced to about 0.8 %; the percentage of P205 was

reduced to less than 1 % in the float fraction, which indicates that using PVA in a pre-

coating solution is a promising process for separation of dolomite from phosphate

pebbles.

Effect of Size on the Flotation

Two samples were crushed in a closed circuit to obtain the following sizes: -

4+1.19 mm and -3.36 +1.19 mm. The conditions were the same as in the previous two

tests. The results of these tests are shown in Table 3-15 within 5 minutes flotation time.

The percentage of MgO was reduced to less than 1 % (0.6 % of the un-floated fraction in

first experiment and 0.95 % in the second).

Table 3-15. Effect of particle size on flotation
Size, mm Sample Wt% Chemical analysis, % Recovery, %
MgO CaO P205 I.R. MgO CaO P20 I.R.
-4+1.19 Float 20.5 12.13 30.34 3.31 8.1 83.9 20.55 4.27 17.87
Sink 79.5 0.6 30.24 19.13 9.6 16.1 79.45 95.73 82.13
Balance 100 2.96 30.26 15.89 9.29 100 100 100 100
Feed 100 3.03 30.38 15.76 9.46 100 100 100 100

-3.36+1.19 Float 17.8 11.84 34.53 3.86 7.04 72.96 16.43 4 12.84
Sink 82.2 0.95 38.02 20.07 10.35 27.04 83.57 96 87.16
Balance 100 2.89 37.40 17.18 9.76 100 100 100 100
Feed 100 2.95 37.07 17.41 9.93 100 100 100 100









Summarizing Remarks

Separation of dolomite from apatite using CO2 generated in the chemical reaction

between dolomitic particles and slightly acidic media was explored by testing different

parameters effecting this process, including: acid type and concentration, surfactant type

and concentration, particle size, and use of polymer (PVA). The results were promising

with anionic surfactants, especially SDSO3, and Mr. Bubble solution with a concentration

of 1.6 % sulfuric acid; however, low dolomite-content particles did not float.

The most encouraging results were obtained with PVA, which floats the low

dolomite-content particles, in addition to floatation of large particle sizes. Additional

testing with a representative sample of ore mixture (phosphate and dolomite) showed the

ability to reduce the MgO content to less than 1 % by using a PVA concentration of 3 %

in an acidic medium concentration of 3 %. The next step will be the optimization of the

RF process using PVA as well as performance of this process under dynamic conditions.














CHAPTER 4
OPTIMIZATION OF THE REACTIVE FLOTATION PROCESS
AND PILOT SCALE TESTING

Introduction

In previous chapter, it was demonstrated that when PVA coated high MgO

phosphate rock is immersed in a 3 % sulfuric acid solution, effectively all the high MgO

particles will float to the surface leaving a sink phosphate rock phase that is a low in

MgO. In this chapter, the optimization of separation process using PVA as well as

development of a separation continuous process is discussed.

In this chapter, the RF is optimized using statistical designs (i.e., screening

statistical design and central composite design (CCD)). Also, the reduction of PVA

consumption was studied using statistical design, (i.e., factorial design). The tests show

that as little as less than one pound of PVA per ton of feed rock can be effective in

floating the dolomite. The tests also show that there is no significant adverse effect of

having the sulfuric acid solution saturated with calcium sulfate. Furthermore, pilot scale

testing of a simple sluice was carried out and demonstrated the effectiveness of this

approach in giving a good separation between the phosphate and dolomite particles.

Optimization of the RF Process

Screening Statistical Design

The MgO % removal in floated fraction, Table 4-1, was used as the response in

calculating the statistical effect of variables. The effect of variables are calculated as

follows:









Table 4-1. The MgO % removal in floated fraction for Plackett-Burman design.
Runs Response
(MgO % removal)
1 3
2 3
3 3
4 3
5 86
6 90
7 3
8 12
9 3
10 20
11 3
12 3


Re sponse(+ Re sponse
Ei = (4.1)
6

The variance is estimated by evaluating the effect of dummy variables and using

this equation:


0"2 = ef E (4.2)
n

where Ed = effect of dummy variable and n = number of dummy variables = 5.

The statistical significance of the effects is evaluated using the student t-test as calculated

using:


t- (4.3)
/ 2

where the degrees of freedom = 5, confidence level = 95 % and t,0.05 = 2.015.

Comparing the absolute t-values to 2.015 led to acceptance of variables with a higher t-

value and rejecting the variables with lower t-values. In other words, the variables

accepted were significant within 95 % confidence level. This comparison of the Plackett-

Burman design is presented in Table 4-2. The PVA molecular weight t-value is 2.761,









which is higher than 2.015. Nevertheless, the higher PVA molecular weight was used

because of its significance. From the Plackett-Burman design, the most significant

variables are acid concentration, PVA concentration, and particle size.

From the screening design, a model relating the MgO % removal to the different

variables could be represented by:

y = -91.75 + 14.83X1 +12X2 3.2X3, (4.4)

where y represents MgO % removal and Xi, X2, and X3 represent acid concentration,

PVA concentration, and particle size, respectively. All coefficients are statistically

significant close to 95 % confidence level. The negative sign for particle size coefficient

indicates that increase the particle size can lead to a decrease in the MgO % removal. On

the other hand, the positive acid concentration and PVA concentration values indicate

that an increase in both concentrations lead to higher MgO % removal. In order to

further investigate the interaction among these three significant variables with respect to

their effect on MgO % removal, a central composite design focusing on these three

variables with five levels was used.

Table 4-2. Plackett-Burman calculated t values for different variables. The decisions to
accept or reject the significance of variables on the bases of comparing t-value
to critical value at 95 % confidence level.
Factors Effect on MgO % removal Significant Variable
(t-value)*
Sulfuric acid concentration 2.69697* Accept
PVA concentration 2.181818* Accept
PVA Molecular Weight (MW) 2.969697' Accept
PVA degree of hydrolysis (D.H.) 0.181818 Reject
Particle size 2.45455* Accept
Drying time 0.333333 Reject
*Statistically significant (95 % confidence level)
"(+ ve) sign indicates more MgO % removal
"(- ve) sign indicates less MgO % removal









Central Composite Design

Central composite design (CDD) results of dolomite flotation experiments are

given in Figures 4-1 and 4-2 in terms of MgO removal % in floated fraction and MgO %

in sink (nonfloat) fraction respectively. The confidence interval is within 95 %. In other

words, the effects are statistically significant at the 95 % level. The standard deviation

and R-Squared for both of them are 13, 0.9287 and 0.36, 0.9287 respectively.

It was indicated, from these figures, that the % MgO recovery and MgO % in

concentrate are greatly and positively affected by increasing either acid concentration or

PVA dosage, (i.e., % MgO recovery increases from 30 to more than 90 % and MgO % in

the concentrate decreases from more than 2.1 to 0.6 respectively). However, particle size

negatively affects the % MgO recovery and MgO % in concentrate. With increasing

particle size the % MgO recovery decreases. At the maximum acid concentration and

PVA dosage, it decreases from 90 to 75 % and % MgO concentration in the concentrate

increases from 0.6 to 1.1 %.

As it can be seen from these results that the PVA consumption is high (about 4.5

lb/t), which is economically unfeasible, to get a concentrate contains < 1 % MgO with

high recovery > 90 %. Therefore, further study is needed to reduce that consumption,

which is discussed in next section.

Research to Reduce the PVA Consumption

From the results obtained from the experimental designs following the screening

the main factors, then studying the main factors, unfortunately, the PVA consumption

was about 4-6 lb/t (calculated according to used amount of PVA to certain weight of rock

including the losses on the coating device), which threaten the success of this process

economically. At this point, it was necessary to find a method to reduce the

consumption.












6 MgO Removal (floated) %



5-


4- A
74

3-



2--A
1.0 1.5 2.0 2.5

Acid concentration, %




6 MgO R







4- 14 34
5
-n\


S M RO Removal (floated) %



5- 93
5-K


-73 B


3 43
3-


2-
1.0 1.5 2.0 2.5


Acid concentration, %


1.0 1.5 2.0 2.5 3.0


Acid concentration, %



Figure 4-1. Effect of PVA dosage and acid cone. on MgO recovery % in floated fraction
at different particle size
A) Particle size = -0.5 mm
B) Particle size = 1-2 mm
C) Particle size = 5-9 mm








53






54


Mg O% in unfloated fraction MvgO% in unfloated fra@


I\ ) \
5- 5-

A B
4- 0.9 4-
1.3

1.3 3

3- 3-

1.0 1.5 2.0 2.5 3.0 1.0 1.5 2.0 2.
Acid concentration, % Acid concentration, %

S MgO% in unfloated fraction


5- C
C


4- 2.5 2.1 1.7 1.3 0.9

3-\
2.9


1.0 1.5 2.0 2.5 3.0
Acid concentration, %

Figure 4-2. Effect of PVA dosage and acid concentration on MgO % in Unfloated
fraction at different particle size
A) Particle size = -0.5 mm
B) Particle size = 1-2 mm
C) Particle size = 5-9 mm









In this regard many tests were conducted among these tests were:

* Using compatible compounds with PVA like starch and cellulose. It was found that
the presence of these compounds negatively affected the stability of the bubbles at
the interface.

* Using fine particles (- 40 mesh) with different PVA (1, 2, 3 %) and acid conc.(1, 2,
3 %). It was found that the 3 % PVA and 3 % acid were needed which is the same
condition as before and in addition the reaction was too fast due to the high surface
area of the small particles.

* PVA + Drimax (spreading agent): to spread the PVA on the particle surface and
then reduce the amount needed to cover the particles. The results of these
experiments indicates that there is an interaction between the PVA and Drimax in
such a way that prevented the formation of stable bubble at the particle / solution
interface, probably by interfering with the crosslinking of the polymer layers that is
necessary for the formation of a membrane and consequently stable bubbles.

* PVA + Boric acid: as it is in the literature [Finch, 1973; 1992], the boric acid and
borax are used in PVA crosslinking. Using the boric acid (5 % of PVA) helps in
reducing the PVA consumption due to polymer crosslinking but we did not go
further in these experiments because of toxicity of boric acid and environmental
reasons.

* PVA + Na2SO4: sulfate ions lead to coagulation of the PVA according to literature
[Finch, 1973; 1992]. It was found that the coagulation is not enough to float the
dolomite particle at low PVA concentration.

* Conditioning of the rock particles with PVA and screening it (or filtering it) to
recover the excess PVA. Actually, this step was a most promising one in which the
consumption of PVA was reduced to 1 lb/t instead of 4-6 lb/t in previous coating
procedure. The conditioning step (with about 1 % PVA) in the same manner as a
regular flotation is followed by a filtration step and calculation of the actual PVA
consumption.

On the other hand, it was found that the full coverage of the particle with PVA is

not needed since the formation of some bubbles may be enough to create the density

difference and by using sluice or spiral in which the dolomite particles will be lighter and

follow the liquid stream. In this manner, the dolomite particle can be separated from the


phosphate.







56


Experimental Design of the PVA Conditioning Step

Due to the high PVA consumption as indicated from experimental designs,

different methods were tested as mentioned before, the conditioning step as in regular

flotation circuits was tested since this step can reduce the consumption to one-fourth or

more (The PVA consumption was reduced to 1-2 lb/t from 4-6 lb/t). For that reason

statistical design with three factors, (i.e., conditioning time, PVA concentration, and pulp

density) was used. It is indicated, from Figure 4-3, that the PVA concentration is the main

factor and then the conditioning time while the pulp density (solid liquid ratio) has a very

slightly effect on the % MgO in the concentrate and % MgO recovery in floated fraction.



Cube Graph Cube Graph
MgO% in Concentrate ube
MgO recovery in floated fraction, %






0 o

m m
S................ 2C+ 846583 2 C+

C PVAConc, % C PVAConc, %

B- 29 282 C- B- 550 888 C-
A- A+ A- A+
A Conditioning time, mln A Conditioning time, mm


Figure 4-3. Cube graph for MgO % and MgO recovery %

Saturation Experiment

The CaCO3 was used to saturate the solution with calcium sulfate by adding CaCO3

to 3 % sulfuric-acid solution. The formation of the precipitate of CaSO4 indicated that the

solution was saturated. It was found that 1.78gm/l of CaCO3 would cause the precipitate

to form. After the formation of the CaSO4 precipitate, the flotation test was run as an

indication of the acid performance after its saturation. Table 4-3 shows the results of

these experiments before and after the saturation is reached.









Table 4-3. Effect of saturated acid on the RF
Before saturation After saturation
Wt% MgO% P205% Wt% MgO% P205%
Float 22.1 11.1 4.2 21.3 10.95 4.15
Sink 77.9 0.8 21.1 78.7 0.85 21.2

Pilot Scale Testing: Sluice Results

It was shown that the PVA consumption could be reduced to 1 lb/t on lab scale.

Testing of the reactive flotation process using a simple sluice was conducted on pilot

scale. Starting from the point that the formation of the bubbles on the dolomite surface

creates a density difference between dolomite and phosphate particles. Therefore, the

gravity separators are the promising techniques to do this separation.

In that regard, two tests were run using the sluice under two different coating

conditions. The first was spraying the particles with 3 % PVA solution before adding the

particles to the sluice and the second was conditioning the particles in 1 % PVA solution

and screening the particles to remove the excess PVA solution to reuse it (the PVA

consumption is estimated to be 0.8 lb/t), then adding the particles to the sluice. The

results of these two tests are given in Table 4-4. From these results, it can be seen that by

applying the optimum conditions, the reactive flotation process can reduce the MgO % to

less than 1 % with high recovery. In the rest of this work, the fundamental interpretations

of what are observed experimentally were carried out.

Table 4-4. Comparison of two different coating methods and their effects on MgO
removal
Sample Coating Conditioning
Wt% MgO% P205% CaO% I.R% Wt% MgO% P205% CaO% I.R%
Float 20.74 12.3 2.9 32.82 9.8 20 12.41 2.85 33.55 10.01
Sink 79.26 0.65 22.3 38.33 11.22 80 0.68 21.90 37.93 11.14
Calc. Feed 100 3.06 18.27 37.19 10.97 100 3.03 18.09 37.05 10.93
Feed 100 3.04 18.13 37.45 11.05 100 3.04 18.13 37.45 11.05














CHAPTER 5
FUNDAMENTAL STUDIES

Polyvinyl alcohol-Particle Interaction
Introduction

Generally, adsorption of a polymer is utilized in many applications, such as:

dispersion of particles, flotation, flocculation processes, treatment of surfaces, etc. If a

solvent is a poor solvent (i.e., the interaction energy between the polymer segment and

the solvent molecules is unfavorable when compared to polymer-polymer or solvent -

solvent molecular interaction energy), then the effective polymer -surface interaction can

be strongly attractive and the polymer will seek any opportunity to minimize the contact

points with the solvent. Hence, in cases where a poor solvent is used, adsorption will

increase on almost any surface, even the liquid-air interface. Adsorption is, therefore,

always accentuated before precipitation will occur. The surfaces will act as a nucleation

sites for polymer precipitation. This means, the poorer the solvent, the higher the

adsorption and vice versa. In addition to the solvent quality, there are other factors that

can affect the polymer adsorption and change its conformation, including: polymer

molecular weight, polymer concentration, pH, and ionic strength.

Previous studies of polymer adsorption suggest a number of qualitative trends that

have been noticed in most polymer adsorption systems. The most important of them are

as follows: many polymers adsorb on a variety of surfaces, Reach a plateau at few

mg/m2, a sharp isotherm sometimes is obtained, and molecular weight is a main factor,

especially in a poor solvent.









In this work, the polymer (PVA) is used as a membrane-forming agent to capture

produced CO2 gas on dolomite particles and makes them lighter than other particles

leading to their flotation and separation from non-carbonate (apatite) particle.

To understand this process and to elucidate the interaction mechanism between

polyvinyl alcohol (PVA) and dolomite and apatite surfaces in the reactive flotation

process, different techniques were used, such as: estimation of work of adhesion using

surface tension and contact angle measurements, adsorption/desorption isotherms, Zeta

potential, and surfaces characterization using Fourier Transform InfraRed (FTIR), etc.;

the details of each one of these techniques are discussed in this chapter.

The strength of adhesion to a solid surface can be estimated from the value of the

thermodynamic 'work of adhesion' or Wa, a concept first introduced by [Harkins, 1928].

In a simple system where a liquid L adheres to a solid S, Figure 5-1, the work of adhesion

is defined as:

Wa = Yi + -Y 7sl ............................ (5.1)

where Ys y and Ysl are the interfacial tensions of the solid/air, liquid/air and solid/liquid

interfaces respectively.

The problem with using Equation 5.1 is that, of the three interfacial tensions, only

y1 can be measured with fluid phases, in this case the liquid and air. Those tensions

involving the solid cannot be measured independently. Another approach, then, is to

combine Equation 1 with the Young-Dupre equation (Equation 5.2), to provide a more

useful expression for the work of adhesion [Adamson and Gast, 1997]:

Ys 7sl = yl cos ........... .............. (5.2)









where 0 is the contact angle that the liquid makes with the solid surface. Provided

that y1 and 0 can be measured experimentally, it is then possible to use Equation 5.3 to

calculate the work of adhesion.

W a = y i(1 + cos0) ............... .......... (5.3)

Therefore, the measurements of surface tension and contact angle are essential

towards the calculation of the work of adhesion of the polymer (PVA) with solid

surfaces, which will be discussed in the following sections.


Y1 Gas






Ys g Liquid Ysi
^ ------- -v ----- [ --------
Solid


Figure 5-1. Contact angle and interfacial surface tension at solid/liquid/gas interface

Surface Tension

The equilibrium surface tension values of fully hydrolyzed PVA solutions (average

Mw =150000) as a function of the initial bulk concentration at room temperature (23 OC)

are shown in Figure 5-3. The surface tension of PVA decreases with increasing

concentration of PVA until it reaches a region where the surface tension (59 60

dyne/cm) no longer decreases with increasing the polymer concentration. The critical

concentration at which the surface tension starts to be constant is 1 % PVA in aqueous

solution. These results coincide with the literature concerning the surface tension of fully

hydrolyzed polyvinyl alcohol (PVA). The fully hydrolyzed polymer has less effect on









surface tension than partly hydrolyzed PVA [Fbiehn and Erusberger, 1951; Kim and

Luckham, 1991]. The chemical structure of both is shown in Figure 5-2.

A B


OHi




Figure 5-2. Structural formula for PVA: A) partially hydrolyzed; B) fully hydrolyzed.

It is reported that the surface tension reaches a constant level at 1-3 % PVA

depending on the PVA molecular weight and degree of hydrolysis [Finch 1973; 1992].

These results are similar to several findings reported in the literature regarding polymer

systems; the stabilization of the surface tension of high-Mw PEO (> 100000) solutions

was associated with the saturation of the air/water interface by the polymer chains [Zeno

et al., 2004].


75

S70

S65

S60

55

50
0 1 2 3 4 5
PVA Concentration, %

Figure 5-3. Equilibrium Surface tension for aqueous PVA solutions at 23 C










An explanation of this phenomenon is that the surface activity of a polymer mainly

relies on its conformation in the aqueous solution, whereas the conformation of

macromolecule depends on its chemical structure and composition, etc. At low

concentrations, the chains of macromolecule are stretched in the water due to the

formation of hydrogen bonding with water molecules. These long chains lying on the

liquid/air interface can dramatically decrease the surface tension of water, even though

the adsorbed amount on the liquid/air interface is very small. With the increase of

concentration, the polymer's molecules that adsorb on the surface increase and a tight

macromolecular film is formed, which makes the surface tension decrease. When the

polymers occupy all the surface of solution, the adsorption reaches a level of saturation

corresponding to the lowest value of surface tension. As this concentration further

increases, the polymers in the solution begin to aggregate and form micelle-like clusters;

at this point, the surface tension shows no more changes. In the case of PVA, it may also

be related to degree of crystallinity due to formation of hydrogen bonding as indicated in

Figure 5-4 [Finch 1973].




OH CH
CH,

CH,
CH > OH ... ...... O
CH ? <. .............
OH
OH CH
..' CH,
CH OH
OH
CH, ....
OH





Figure 5-4. PVA crystal structure; the dotted lines indicate hydrogen bonds.









It is expected that the surface tension will have higher values in the presence of

the acidic media since the S04-2 ions increases the surface tension, affects the polymer

solubility, and enhances the formation of polymer aggregates at even a slightly high acid

concentration [Sun, 1994].

Contact Angle

The phenomenon of wetting or non-wetting of a solid by a liquid is better

understood by studying contact angle, which is the most universally applied method of

quantifying wettability. Contact angle methods have been developed extensively over the

past four decades. The contact angles of dolomite and apatite surfaces with PVA

aqueous solutions at different concentrations are shown in Figure 5-5. From the figure, it

can be seen that the contact angle increases with the increase of PVA concentration for

both minerals' surfaces. This phenomenon can be explained by the structure of PVA in

the solution shown in Figure 5-4. When a droplet of PVA placed on the surface of

dolomite or apatite, the PVA adsorbs to the surface by hydrogen bond. As the PVA

concentration increases, more and more hydrogen bonds form; as a result of hydrogen

bonds among its molecules, the PVA begins to form clusters. These clusters result in a

local viscosity increase, which controls the droplet shape as well as its tendency to

spread. This, in turn, increases the contact angle accordingly.

The dynamic behavior of the molecules, their conformation at solid-liquid

interface, and diffusion of macromolecules will affect the adsorption of more molecules,

with a subsequent increase in contact angle. The increase in contact angle is proposed to

result from polymer cluster formation and preferential coverage of the hydrophilic sites

by polymer molecules. This phenomenon is observed for PVA and also for some

modified PVA end groups [Finch 1992].











40



S30



S20
-- Dolomite
-*- Apatite
10
0 1 2 3 4 5

PVA concentration, %

Figure 5-5. Static contact angle variation with the PVA concentration

All measurements shown in Figure 5-5 were conducted immediately after

application of the drop. Evaporation of water caused the drop to contract and a

consequent change in the contact angle.

Work of Adhesion

Figure 5-6 shows the dependence of work of adhesion, W,, on the PVA

concentration. The work of adhesion decreases with increasing PVA concentration

(increase in AG, where W, = AG [Lloyd, 1994]). The increase in AG reflects the

formation of polymer clusters and increase of its crystallinity. This is due to the

formation of OH-group bonds between PVA molecules causing them to favor adhesion to

a solid surface. Additionally, the polymer's diffusivity decreases with higher

concentrations.

With increasing PVA concentration, the PVA molecules not only adsorb to the

surface but also form hydrogen bonds among themselves. This behavior affects the









conformation of the molecules at the solid-liquid interface till they reach the saturation

stage, when the work of adhesion becomes constant. It appears that this saturation stage

is the same region of PVA concentration, 1-3 %, at which constant surface tension and

contact angle are achieved.




E 160
S-4- Dolomite
140 --- Apatite


1 120


too
S 100


80 i ,
0 1 2 3 4 5
PVA concentration, %

Figure 5-6. Work of adhesion as a function of PVA concentration on dolomite and
apatite substrates

Additional Studies

To substantiate the presence of hydrogen bonding hypothesis and to establish an

adsorption mechanism additional studies were conducted and their results are discussed

in the paragraphs that follow.

Adsorption study: Research grade samples were purchased from Wards Natural

Science and used to conduct the adsorption study. These samples were chemically

analyzed by the wet chemical method and the results are listed in Table 5-1.










Table 5-1. Chemical analysis of the samples used in adsorption tests
MgO% P205% CaO% A1203% Fe203% I.R.% L.O.I. %
Dolomite 20.39 0.58 31.6 0.97 0.10 2.52 42.6
Apatite 31.73 49.50 2.64 1.04 4.95 5.0

Specific surface area measurements (BET method): The specific surface area

of dolomite and apatite particles was determined by the BET in order to be used in

adsorption isotherm calculations (applied in the next section). Using nitrogen as the

adsorbate, the measured surface areas for dolomite and apatite were 0.9 and 1.48 m2/g

respectively.

Pore size distribution: The results of cumulative pore size distribution for apatite

and dolomite are shown in Figures 5-7 and 5-8. It is clear that the apatite has a higher

cumulative pore area, 9 m2/g, than the dolomite, 1.8 m2/g. Mean pore sizes for apatite and

0 o
dolomite are 28 A and 42.5 A respectively.



10













0
0 200 400 600 800 1000
Pore diameter, A


Figure 5-7. Cumulative pore size distribution for apatite sample











2


S1.6


1.2


0.8


3 0.4


0
0 200 400 600 800 1000
Pore diameter, A


Figure 5-8. Cumulative pore size distribution for dolomite sample

Zeta Potential

In order to obtain a more detailed description of the adsorption and electrical

properties of the solid-solution interface, measurements of the zeta potential were taken.

The tendency of zeta potential to vary with nonionic polymer adsorption may be caused

by one of two different factors: 1) the shift of the position of the shear plane, or 2) the

blocking of the active sites on the surface by adsorbing polymer chains. The participation

of these effects was estimated by the determining the difference in the values of the zeta

potential with no polymer and with polymer present. The use of a KNO3 solution (3X10-

3 M) acted as a control in these experiments.

The results of measuring the zeta potential of apatite and dolomite in 3X10-3 KNO3

solution as function of pH are presented in Figure 5-9. It can be seen that the iso-electric

points (IEP) of apatite and dolomite are about 6.8 and 9, respectively. This is in

agreement with the results reported in the literature [Amankonah and Somasundaran,

1985; Chander and Furestenau, 1982; Pugh and Stenius, 1985].








68




50
40 3X10 3M KNO3
30
20

0

1 -10 2 4 6 8 10 12
S20
-30
-40 DDolomite
-50 Apatite
-60
pH




Figure 5-9. Zeta potential of dolomite and apatite samples


Zeta potential in the presence of PVA was determined as well. Figures 5-10 and 5-


11 show the results of measuring the zeta potential in the presence of a PVA solution. It


is evident that the presence of PVA does not change the IEP of the apatite or dolomite. It


does, however, cause compression in the electrical double layer.



60

50 3X10 3M
KNO,
40




a 10
0
2 4 6 8 1
-10
-20
Dolomite
-30 4Dolomite + PVA
-40


Figure 5-10. Zeta potential of dolomite in absence and presence of polymer











40
30 3X103M KNO,
20
10
0
-10 2 4 6 8 0 12 1
-20

30 Apatite
Apatite + PVA
-40
-50
pH

Figure 5-11. Zeta potential of dolomite in absence and presence of polymer

The results indicate that the surface sites remain accessible to potential determining

ions (H+ and OH-) and the IEP remains unchanged during PVA adsorption, thus implying

that the polymer does not alter the surface potential. It can be considered, as other

investigators asserted [Ottewill, 1967; M'pandou and Siffert, 1984; 1987; Luckham et al.,

1983] that the decrease in zeta potential relating to adsorption of PVA is not due to a

decrease in charge and surface potential, but rather to a shift of the shear plane (Figure 5-

12).


Ster plane Ster plane
Surface Potential A
i Shear plane Shear plane B



ZeaZP l -- eta potential before adsorption
Zeta Potential -_-- oO lS --
S- Zeta potential after adsorption
*S* e

Distance from surface Distance from surface



0 Ions counter ions e\ Polymer chain

Figure 5-12. Effect of polymer on the shear plane position, A) without polymer, B) after
polymer adsorption










In short, the changes of the zeta potential produced by the adsorption of

substances with high molecular weight, result from blocking of the active sites on the

minerals' surface and as well as from the shift of the shear plane.

Zeta potential of minerals in supernatant solutions

In natural ores, the dolomite and apatite form a mixture. Their presence together

may affect the adsorption of the polymer on both of the minerals' surfaces. Therefore, the

adsorption on the dolomite in presence of apatite supernatant and apatite in presence of

dolomite supernatant were studied.

The effect of the supernatant of dolomite on the zeta potential of apatite is shown in

Figure 5-13. The results show that the apatite surface becomes more positive over the

entire pH range and that the IEP has shifted from 6.8 to 8.6. Both effects are

hypothesized to be due to the presence of calcium and magnesium ions in the solution.


50
40 3X10-M KNO3
30 6
>20
10
0 .
-10 2 4 6 8 10 12
-10
-20

--30 -Apatite
-40
Apatite in Dolomite
-50
pH

Figure 5-13. Zeta potential of apatite and apatite in dolomite supernatant










Similarly, the zeta potential of dolomite in an apatite supernatant was measured as

a function of pH. The results, given in Figure 5-14, show the zeta potential of dolomite

to be reduced drastically in presence of apatite supernatant. The IEP shifts from 9 to 6.5.


50
40 3X10MM KNO3
30
20
-. 10


2 4 6 8 10 12
M-10
N-20

-30 -Dolomite
-40 -Dolomite in Apatite supernatant
-50
pH
Figure 5-14. Zeta potential of dolomite and dolomite in apatite supernatant

Zeta potential in the presence of PVA

Moreover, the zeta potential of dolomite in apatite supernatant in presence of PVA

was determined. The results in Figure 5-15 indicate that the PVA deals with surface of

dolomite, when suspended in an apatite supernatant, as an apatite surface. This is due to

the faster diffusion of small ions and molecules from the solution to the dolomite surface,

which converts it to surface similar to apatite. As can be expected, the movement and

diffusion of these ions is quite faster than that of polymer molecules. Consequently, The

polymer's role is applied after the surface changing and the adsorption of the polymer

appears as if it is on apatite surface.

The same argument mentioned in the previous paragraph can be applied in the case

of PVA present with apatite in a dolomite supernatant. The results of the zeta potential of

this system are shown in Figure 5-16.














40
3X10 M KNO,
30
20



S10 2 4 6 12




-40
N -30

-40 Dolomite in apatite supernatant + PVA
-50 Dolomite in Apatite supernatant
-60
pH

Figure 5-15. Zeta potential of dolomite in apatite supernatant before and after PVA
adsorption


40 3X10MM KNO3

30

> 20

10
0 T
S 2 4 6 8 11

-20
-30 "4'Apatite in Dolomite supernatant + PVA
Apatite in Dolomite supernatant
-40
pH

Figure 5-16. Zeta potential of apatite in dolomite supernatant before and after PVA
adsorption

Infrared (FTIR)

The FTIR spectra of the powders, both with and without polymer adsorbed to them,

were compared in order to obtain the mechanism of adsorption of PVA on dolomite and

apatite surfaces. This method was applied in previous studies [Kiselev and Rygin, 1972;










Paradip and Moudgil, 1991; Howard, and McConnel, 1967; Behl and Moudgil, 1993;

Bjelopavlic et al., 2000; Zaman et al., 2002; Mathur and Moudgil, 1997; Moudgil and

Prakash, 1998] to explain polymer-solid interactions. The major peaks for dolomite and

apatite are identified according to Ince et al. [1991] and Farmer [1974].

FTIR spectroscopic studies were carried out on the dolomite and apatite and

polymer samples both before and after adsorption. The assignments of the various bands

and peaks made in this study are in reasonable agreement with those reported in the

literature for similar functional groups [Finch, 1973; Chanchani, 1986; Ince, 1987].

The FTIR spectrum of PVA is shown in Figure 5-17. A broad-band, is displayed at

3450 cm-1, which may be attributed to hydrogen-bonded surface hydroxyl groups. The

sharp band at 2930 cm-1 is attributed to strong C-H stretching. The band at 2860 cm-1 is

attributed to shoulder C-H stretching. The transmission band at 1650 cm-1 is due to the

presence of moisture in the sample. The band at 1460 cm-1 can be attributed to O-H and

C-H bending. These results are in good agreement with the spectra of PVA reported by

other workers [Finch, 1973; 1992].










'2930

81650
3450 1460 V862
594
1110

4000 3600 3200 2800 2400 2000 1600 1200 800 400
Wave number, cm-1

Figure 5-17. Infrared transmittance spectra (FTIR) of poly vinyl alcohol (PVA)











The FTIR spectrum of the bare-dolomite sample as well as those that had

previously interacted with PVA is shown in Figure 5-18. The isolated hydroxyl group

peak is present in dolomite powder without polymer, but the spectra with polymer shows

the disappearance of this peak. Moreover, the peak intensity of the hydrogen-bonded

hydroxyl groups is smaller in case of the powder with the polymer than without. These

observations indicate that the adsorption of PVA on dolomite occurs by hydrogen

bonding involving -OH groups of the polymer. This finding is in concurrence with

earlier studies on silica-PEO [Rubio and Kitchener, 1976; Cheng, 1985] and silica-

Polyacrylic acid PAA [Tadros, 1978], which showed that the presence of isolated OH-

groups on the surface is responsible for the adsorption of polymer molecules by hydrogen

bonding.



Dolomite + PVA 1460
Dolomite 883

728
1820

3450 1460
3640 3030 2910 2550
A 2640 883
728
1820

2550
3450 2910 26402550
3030

4000 3600 3200 2800 2400 2000 1600 1200 800 400
Wave number, cm-1

Figure 5-18. Infrared absorbance spectra (FTIR) of A) dolomite alone, and B)
dolomite/PVA system.









The FTIR spectrum of the apatite sample after having interacted with PVA is

shown in Figure 5-19. However, unlike the case of dolomite, this figure indicates that the

adsorption mechanism of the polymer on apatite is not clarified by using FTIR-no

change is shown in the bands' intensities or positions, either for bare apatite or apatite

with polymer adsorbed on it. This finding is in agreement with the current literature

[Chanchani, 1986; Ince, 1987]. It is pertinent to mention that FTIR spectroscopic studies

on the dolomite-PVA and apatite-PVA systems revealed the presence of hydrogen

bonding only.


1050


-Apatite 607
-Apatite + PVA 1460

866
3450 1650

A






4000 3600 3200 2800 2400 2000 1600 1200 800 400
Wave number, cm1

Figure 5-19. Infrared absorbance spectra of A) apatite alone, and B) apatite/PVA system.

Adsorption Study

An adsorption study was used for the purpose of determining PVA adsorption on

both dolomite and apatite. The calibration curve for PVA concentration was determined

colometrically. Figure 5-20 shows the calibration curve and its equation, which was used

in this study to calculate the residual concentration of PVA in adsorption experiments.









The limitation of the PVA concentration to create this curve is 40 ppm [Finch, 1973;

1992].


0.16
0.14

S0.1
0.08 -
I 0.06
0.04 -
0.02
0
0 10 20 30 40 50

PVA Concentration, ppm

Figure 5-20. Calibration curve for PVA concentration measurements

Kinetics of adsorption

A study of the kinetics of adsorption is mainly to determine the equilibrium time of

adsorption. The rate by which equilibrium is established during adsorption depends on

the chemical nature of the polymer and its molecular weight, the solvent, and the type of

adsorbent. In most cases, the kinetic curves indicate an increase in the amount of

adsorbed polymer over time, with an asymptotic approximation of the equilibrium value

[Lipatov and Sergeeva, 1974]

The adsorption of a polymer on solid surfaces occurs by a three-step mechanism

[Dijit et al., 1990; Somasundaran and Ramachandran, 1988]. First, the polymer is

diffused from the bulk to the interface. Second, adsorption of the polymer molecules

occurs on the initially bare solid surface. Finally, polymer conformation and orientation

changes in the surface layer to accommodate the adsorption of more polymer molecules.









To ascertain the time required for the adsorption equilibrium to be attained, kinetic

tests were carried out at a natural pH of 7 for a period of 24 hours; PVA concentration

was 100 ppm (6.7X10-7 M). Figure 5-21 shows results on the adsorption of PVA onto

dolomite and apatite as a function of time. It is evident that the equilibrium is attained at

about one hour. The equilibrium values are 0.04X10-6 and 0.011X10-6 mole/m2 for

dolomite and apatite respectively. In consideration of these findings, a 2-hour

conditioning time was selected for the adsorption studies.

It was observed that the adsorption followed a typical second-order adsorption

process. The second-order model predicts the behavior over the whole range of

adsorption and is in agreement with an adsorption mechanism being the rate-controlling

step. The second-order rate expression [Ozacar, 2003] based on solid capacity is

generally expressed as follows:

= k(Feq F)2 ....... .............. (5.4).
dt
Or, by integration

F t ................. (5.5)
1 t
+
kF2 r
keq Feq

Where F is adsorption density (mole/m2) and Feq is equilibrium adsorption density

(mole/m2), K is second-order adsorption rate constant (hr-1), and t is time (hr). By fitting

the experimental data to Equation 5.5, the parameters Feq and K were determined. These

are listed in Table 5-2.

Table 5-2. Fitting parameters of second-order kinetic model
Feq X 10-6 mole/m2 K, hr-1
Dolomite 0.04 361.6
Apatite 0.011 200










It can be seen that the Feq values for dolomite and apatite match the experimental

values. The higher Feq value of dolomite indicates the presence of more sites available for

PVA adsorption than on apatite. Moreover, the Kad for dolomite is higher than the Kad

for apatite, which indicates the higher adsorption rate on dolomite. However, the plateau

was reached almost at the same time for both minerals.




5.E-02


4.E-02
2

h 3.E-02


~ 2.E-02


1.E-02


O.E+00


0 5 10 15 20 25
Time, hr


Figure 5-21. Kinetics of PVA adsorption on dolomite and Apatite
(Initial PVA concentration 6.7X107 M, pH 7, 2X10 -3 M KNO3 and 230C).

Explanation of the higher adsorption rate on dolomite: The higher adsorption

rate on dolomite may be explained by: (1) The higher solubility of dolomite and

production of more ions in the solution, which affects the solution's property as a solvent

(effectively converting it to poorer solvent). This property also affects the polymer's

conformation, which will either be flat on the solid surface-greater tendency for

molecules of polymer to be adsorbed on the solid surface or attach to each other, (2) The

neutrality of apatite (pH 7) while having an IEP of 6.8, which means less ions in the

solution are available to act as additional driving force for PVA to adsorb on the apatite


Dolomite
A Apatite
Dolomite-2nd order model
-Apatite- 2nd order model

-A A AA


f









surface. This is especially significant as there is high competition between the interaction

of PVA with water molecules and interaction of PVA with the apatite surface, and (3)

OH-groups on dolomite surface (isolated and hydrogen-bonded hydroxyls) are greater

than those on apatite surface (hydrogen-bonded hydroxyls). In other words, the isolated

OH is exclusive to the dolomite surface, which is confirmed with FTIR spectra

measurement.

Adsorption isotherm

Most isotherms of the adsorption of polymers from dilute solutions can be visually

represented as curves that reach saturation at specific concentrations [Lipatov and

Sergeeva, 1974]. It is important to note that polymers may adsorb on surfaces in various

conformations depending on bulk (solvent) and surface chemical properties of the

substrate, as well as other physical conditions of the system [Hong et al., 2001].

At low concentrations, the polymer chain behaves like isolated coil. If the

concentration of the polymer in the solution is raised, the chains become close to each

other, which allows polymer-polymer intermolecular interactions to influence the motion

of the polymer chains. The effects may be expected to be cumulative until the point of

chain-overlapping concentration, which is the close packing of the polymer coils in

solution. Once the overlapping concentration has been reached, the polymer motion will

be dominated by the presence of direct polymer-polymer interaction. As polymer

concentration is increased, the polymer diffusivity decreases. If the concentration

increases even further, a pseudo-matrix-gel will be formed. The resulting viscosity is

governed mainly by intermolecular interaction, which is related to polymer chains

overlap [Izmailova et al., 1975; Gotze and Sonntag, 1987; Andreyeva et al, 1973;

Babayevskii et al., 1986; Koopal et al., 1988].









For PVA, -OH is the only functional group [Koopal and Lyklema, 1979;

Chibowski and Paszkiewicz, 1995; Chibowski et al., 2000]. Increasing the degree of

hydrolysis reduces acetate groups, which usually show preferential adsorption. Thus, the

conformation of the polymer chain at the solid-polymer solution interface will depend on

the number of acetate groups in a PVA macromolecule. The lower the degree of

hydrolysis, the flatter the polymer chains; conversely, at high degree of hydrolysis, the

polymer chains should be oriented perpendicularly to the surface [Santhiya et al., 1999;

Sjoberg et al., 1999; Bajpai and Vishwakarma, 2003; Koopal and Lyklema, 1979]. This

conformation of the PVA macromolecule will favor the adsorption of the polymer on the

surfaces of dolomite and apatite, which is shown in Figures 5-22 and 5-23.

Chibowski [2000] reported that the adsorption of PVA on alumina increases with

an increase in the degree of hydrolysis. This is also confirmed by previous experimental

results and polymer screening experiments. In this work, it was found that the fully

hydrolysed PVA was the best among different degrees of hydrolysis PVAs' (Chapter 3 in

this thesis).

The adsorption density (mol/m2) of fully hydrolyzed PVA (acetate groups < 1 %)

on the minerals' surface as a function of the initial polymer concentration (0 to 4X10-6

mole) was determined at 230 Celsius and a pH 7 for dolomite and apatite. The results

obtained are presented in an adsorption isotherm plot (Figures 5-22 and 5-23 for dolomite

and apatite, respectively).

The analysis of these adsorption isotherms shows formation of a plateau for both

apatite and dolomite. Such behavior of the macromolecules comes from a different

adsorption affinity towards different surfaces. This result suggests that the observed







81


increase in PVA adsorption is mainly caused by the change in conformation of the

polymer chain at the minerals' surface. Following this change, a number of active sites

on the surface become available for further adsorption of polymer molecules. As

mentioned previously, the greater porosity of apatite leads to high number of adsorption

sites available and the horizontal stretching of the polymer to cover these sites. This, in

turn, explains the lower adsorption layer thickness and higher parking area of apatite.

The shape of the adsorption isotherms shown in Figures 5-22 and 5-23 is typical for

physical adsorption and is an acceptable shape for that range of polymer molecular

weight. [Adamson and Gast, 1997]. The measurements were repeated three times and

error bars were plotted on each graph. From the figures we can see that the adsorption

isotherm is S-shaped, which suggests a physical adsorption of PVA onto the dolomite and

apatite surfaces.




@ 2.5
S Dolomite
2.0

^1.5

5 1.0

I 0.5

0.0
0.0 0.4 0.8 1.2 1.6 2.0
Equilibrium PVA cone., 107 mole

Figure 5-22. Adsorption isotherm of PVA on dolomite at 230C











2.0
~ Apatite
1.5


S1.0


0.5

0.0
0.0 0.4 0.8 1.2 1.6 2.0
Equilibrium PVA Conc., 10 mole
Figure 5-23. Adsorption isotherm of PVA on apatite at 230C

Calculation of PVA adsorption and parking area

The amount of PVA adsorption was calculated using the following formula:

(C -C )* V
S= (C -C)* (5.6)
m*SA

Where, F is the amount of PVA adsorption (mole/m2), (Ci-Cr) is PVA concentration on

the particle (mole/ml), Ci is initial PVA concentration, Cr is residual PVA concentration,

V is the volume of solution (ml), m is the mass of dolomite or apatite particles (g), and

SA is the specific surface area of particles (dolomite 0.9 and apatite 1.48 m2/g).

The parking area of PVA molecules was calculated using the following formula:

ix 10O xxMw}

Parking area = (5.7)
Av

Where MW = 150,000 g/mole, Av is Avogadro's number (6.023 x 1023 #/mol) and

parking area is nm2/#. The calculated parking areas for dolomite and apatite at the first

plateau are (23.72 nm2) and (75.47 nm2) respectively. Taking into account that the









parking area per stretched molecule is 0.2 nm2 [Koopal and Lyklema, 1979], it was

observed that PVA-molecule forms a wavy or zigzag shape on the solid particles with the

higher thickness on the dolomite particles. Moreover, this conformation on dolomite

appears at lower relative concentration range. It is worth noting that in the case of

dolomite the adsorption isotherm is both steeper and reaches a plateau at lower

concentrations than in case of apatite.

The observed results may be explained based on the following hypothesis, which is

Schematically illustrated in Figure 5-24: macromolecules are generally more tightly

bound to surfaces due to the multiple points of attachment. The figure shows the

conformation of single polymer chain (5-24A) and/ or several chains (5-24B). It may be

concluded that without intermolecular H-bonds, spiral conformations of PVA are

favorable and with intermolecular H-bonds, zigzag conformations is expected.








Dolomite/ Apatite Dolomite / Apatite Dolomite/ Apatiti





DolomiteDolomitApDolomiteite ite i Dolomite/ Apatit(


Figure 5-24. Polymer-conformation changes with increase the concentration: A)
conformation of a single molecule, B) conformation of more than one
molecule.









Adsorbed layer thickness

The thickness of PVA layer adsorbed on dolomite and/or apatite can be calculated

from the adsorption isotherm by using the following relationship:


h = -(5.8)

Where, h is the adsorbed layer thickness, m; F is adsorption density, g/m2, and p is

PVA density, g/m3.

Using this relationship, the thickness of the adsorbed layer was found to be 27 nm

on dolomite and 9 nm on apatite, which indicates that the vertical polymer chains are

stretched three times longer in the case of the dolomite than in the case of the apatite.

This explains the greater adsorption on dolomite; since, according to their conformation,

the polymer chains occupy smaller number of sites. In addition, more sites are available

on dolomite than on apatite.

Note: The length of the PVA molecule of this molecular weight is about 600 nm,

which indicates that the polymer conformation would be like Figure 24B.


Effects of pH

Ionic sites at the surface generally control the surface charges, on salt-type minerals

such as apatite and dolomite. Specifically, for dolomite are: Mg+, Ca+, CO3--, HC03-,

OH-, MgHCO3+, MgOH+, Mg(OH)2, CaHCO3+, CaOH+, Ca(OH)2 and for apatite are:

Ca+, HP04--, CaOH Ca(OH)2, H2PO4-, CaH2PO4 The nature of the mineral surface,

pH, the presence of solutes, and the functional group of the polymer all affect the

adsorption behavior. Previous studies on different minerals (i.e., alumina, silica, kaolin,

clays) showed that the OH group of PVA is mainly responsible for adsorption on the










surface of a mineral [Santhiya et al., 1999; Sjoberg et al., 1999; Bajpai and Vishwakarma,

2003]. The effect of pH on the adsorption of PVA onto dolomite and apatite is shown in

Figure 5-25. It can be seen that adsorption is pH-dependent in the studied range of 2 to

12.


6.0E-02

S5.E0-02

S4.0E-02

S3.0E02

5 2.0E02

1.0E02

0.0E+00
1 3 5 7 9 11 13
pH

Figure 5-25. Effect of pH on adsorption of PVA on apatite and dolomite

Studying the adsorption of PVA on both minerals gives evidence that, due to

fluctuations in pH, the adsorption is due to formation of hydrogen bonding between the

dolomite or apatite surface from one side and OH- polymer functional group from

another side.

Effects of pH on Apatite: It is evident from Figure 5-25 that the adsorption

density decreases with an increase of pH. The decrease in the adsorption density as a

function of pH may be explained by taking into consideration the surface charge of the

apatite particles as well as the surface species, both of which are dependent on pH. It is

interesting to note that the decrease of CaH2PO4 and H2PO4- species, according to the

apatite activity chart [Attia and Fuerstenau, 1989] in entire pH range, are responsible for