<%BANNER%>

Exposure Potential of Sulfuric Acid Mist at Phosphate Fertilizer Facilities

Permanent Link: http://ufdc.ufl.edu/UFE0021983/00001

Material Information

Title: Exposure Potential of Sulfuric Acid Mist at Phosphate Fertilizer Facilities
Physical Description: 1 online resource (155 p.)
Language: english
Creator: Hsu, Yu-Mei
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: aerosol, impactor, niosh, phosphate, sulfate
Environmental Engineering Sciences -- Dissertations, Academic -- UF
Genre: Environmental Engineering Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Strong inorganic acid mists containing sulfuric acid (H2SO4) were identified as a 'known human carcinogen' in a recent report on carcinogens by the National Toxicology Program where phosphate fertilizer manufacture was listed as one of many occupational exposures to strong acids. To properly assess the occupational exposure to H2SO4 mists in modern facilities, the objective of this study was to characterize the true H2SO4 mist concentration levels. Three sets of experiments were conducted. Firstly, field sampling using dichotomous samplers, silica gel tubes and cascade impactors was conducted to collect the PM2.5/PM10 H2SO4 mist concentration, total H2SO4 mist concentration, and size-resolved H2SO4 mist concentration, respectively, at phosphate fertilizer plants. The H2SO4 concentrations were found to vary significantly among these plants with H2SO4 pump tank areas having the highest concentration level. When high aerosol mass concentrations were observed, the H2SO4 mist had its mode size in the 3.8?10 ?m range that would deposit in the upper respiratory region. Secondly, SO2 adsorption and sulfur(IV) oxidation were investigated under various sampling times, SO2 concentrations and sampling flowrates. Experimental results verified that the collecting medium can adsorb SO2 gas and the extraction procedure of NIOSH Method 7903 aids the transformation of SO2 into sulfate to cause a positive artifact. The experimental data were also fitted into a deactivation model for estimating the artifact sulfate concentration. Thirdly, a honeycomb denuder system and the deactivation model were applied to minimize the artifact sulfate of NIOSH Method 7903 in a field sampling campaign. Both the system and the model were shown to effectively reduce the artifact sulfate concentration. However, the concentration thus determined was still higher than that measured by a cascade impactor which had no artifact. One possible reason is the residual sulfate in the collecting medium.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Yu-Mei Hsu.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Wu, Chang-Yu.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021983:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021983/00001

Material Information

Title: Exposure Potential of Sulfuric Acid Mist at Phosphate Fertilizer Facilities
Physical Description: 1 online resource (155 p.)
Language: english
Creator: Hsu, Yu-Mei
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: aerosol, impactor, niosh, phosphate, sulfate
Environmental Engineering Sciences -- Dissertations, Academic -- UF
Genre: Environmental Engineering Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Strong inorganic acid mists containing sulfuric acid (H2SO4) were identified as a 'known human carcinogen' in a recent report on carcinogens by the National Toxicology Program where phosphate fertilizer manufacture was listed as one of many occupational exposures to strong acids. To properly assess the occupational exposure to H2SO4 mists in modern facilities, the objective of this study was to characterize the true H2SO4 mist concentration levels. Three sets of experiments were conducted. Firstly, field sampling using dichotomous samplers, silica gel tubes and cascade impactors was conducted to collect the PM2.5/PM10 H2SO4 mist concentration, total H2SO4 mist concentration, and size-resolved H2SO4 mist concentration, respectively, at phosphate fertilizer plants. The H2SO4 concentrations were found to vary significantly among these plants with H2SO4 pump tank areas having the highest concentration level. When high aerosol mass concentrations were observed, the H2SO4 mist had its mode size in the 3.8?10 ?m range that would deposit in the upper respiratory region. Secondly, SO2 adsorption and sulfur(IV) oxidation were investigated under various sampling times, SO2 concentrations and sampling flowrates. Experimental results verified that the collecting medium can adsorb SO2 gas and the extraction procedure of NIOSH Method 7903 aids the transformation of SO2 into sulfate to cause a positive artifact. The experimental data were also fitted into a deactivation model for estimating the artifact sulfate concentration. Thirdly, a honeycomb denuder system and the deactivation model were applied to minimize the artifact sulfate of NIOSH Method 7903 in a field sampling campaign. Both the system and the model were shown to effectively reduce the artifact sulfate concentration. However, the concentration thus determined was still higher than that measured by a cascade impactor which had no artifact. One possible reason is the residual sulfate in the collecting medium.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Yu-Mei Hsu.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Wu, Chang-Yu.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021983:00001


This item has the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101221_AAAABW INGEST_TIME 2010-12-21T21:32:44Z PACKAGE UFE0021983_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 18007 DFID F20101221_AABKOK ORIGIN DEPOSITOR PATH hsu_y_Page_063.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
b5dfafa9e7bdb9ae0c735d08fcd7b72d
SHA-1
c9abf49afe97346cb148256e0fbf588248fe9282
14603 F20101221_AABKNW hsu_y_Page_039.jpg
d3cc4d42e1a97b0da2e0d42ac42a713d
95159f96ca6364fc3e8e0465f9a3af346179356b
106789 F20101221_AABKPA hsu_y_Page_106.jpg
c868c82ff49345c52bc5ec4ff37ea373
a7956b9f3f6d9c35dac2be08c7fae5662f3ea94d
25284 F20101221_AABKOL hsu_y_Page_068.jpg
7c9f117ae9c55e0549b6aaa442d08d2d
740a6b16b2f25d14699822736b4e7067eb2ec476
16939 F20101221_AABKNX hsu_y_Page_041.jpg
356bf1b4dd656d873c0a431577e93a79
2c07cadfa316ae57d5a1fc43fe16cf629307c9ca
106359 F20101221_AABKPB hsu_y_Page_108.jpg
7b0e6b5f9ae2181633b06fcbc51a840b
29ca1f83e078fbb22ee7569a53e5fe80db622d65
45670 F20101221_AABKOM hsu_y_Page_071.jpg
24c59b43beadef8dde15f98fca9afeea
4842bfd0293237004a7a00960449022daeb71a2a
14531 F20101221_AABKNY hsu_y_Page_042.jpg
10501c886f090f6d559b50df13f0d3e2
fa9d0a5190291619367a77f32506c742e8f08489
96106 F20101221_AABKPC hsu_y_Page_109.jpg
e20a4f9c4e27487062c17c2a0bedd6aa
4078f352c7718493b1a9bfa600921b8e63f84a90
114748 F20101221_AABKON hsu_y_Page_072.jpg
6087d9ce3d0ade4b801db86129876d0f
f69ca6861eb190132c5a8ad614b8fe18f668ff95
42921 F20101221_AABKNZ hsu_y_Page_043.jpg
9c72d8caedfd279e5bf43396302e49aa
6cbcfa0ff4275c323d9decfd2cbba51eef4a04d7
19120 F20101221_AABKPD hsu_y_Page_111.jpg
f9c3d7110c0549de2c7136ffbcb04bd5
bad351ec81976e84b36fd4184bde0db8007ce56d
103574 F20101221_AABKOO hsu_y_Page_073.jpg
c51f0ff8f974d8fa6afa059931bb9bb4
109de84ee11d289976b4fbf5e1834d1a6ad5f4f8
25386 F20101221_AABKPE hsu_y_Page_114.jpg
301e472f04bb7e4521abbaa96b0052f4
de73e857b442c056d5eaf4ecfe432d4269345fe3
108636 F20101221_AABKOP hsu_y_Page_081.jpg
61641b2052ff45ddb757ffb430ba6bf7
d77e55c737c1fbe66647fc56ebfd4549762dd920
44204 F20101221_AABKPF hsu_y_Page_115.jpg
6f0f4e24f586c6412ee17f137b8f4979
f8af7f9abadad187ea605f44fa7b760f01696824
104757 F20101221_AABKOQ hsu_y_Page_084.jpg
6717cb09adb4e3f8b002da1c743d81d7
d2b260ef97f63992e48155d1f2b27c9eb688fcb4
64628 F20101221_AABKPG hsu_y_Page_116.jpg
0f4e353c8d7ef5d7dbc23060f64c8db1
578c3ed19f2e08e22510afa678e1e572f3bc3abc
83142 F20101221_AABKOR hsu_y_Page_085.jpg
92dfe134e298aa59355d4eae415e78a3
51c717249d937d9dcf5b3eb627a9a6a2a1d37f3c
37035 F20101221_AABKPH hsu_y_Page_118.jpg
85ec25f44ad0d6ccd40d897c52b82616
f97cc1fc65401d8268e50bb8c565057d6481c8f9
82910 F20101221_AABKOS hsu_y_Page_086.jpg
3fa37f0b00607de01e8bef8df3ee25d6
b5df86f0097dcda1073bd63d7bbeae135b87b59f
102457 F20101221_AABKPI hsu_y_Page_128.jpg
c58b834d9b00b712dda192c432ef6e5e
78cce88d4b6a0763c75e44ddd5b860111cd0535a
53667 F20101221_AABKOT hsu_y_Page_089.jpg
fa6b2dcad030efc249906fe3e9f24235
599f12233d15f7d708976422224cf5a8a16e479a
68785 F20101221_AABKPJ hsu_y_Page_130.jpg
60bae13fae9475025389328df09e99bd
8b94ebaa3172a16a191160d0ede6c436731099cf
15178 F20101221_AABKOU hsu_y_Page_091.jpg
ebb21c541fcf740b2cf9b0706d5baf6e
b469ea54ab6164e2525b94e2c342a2a81f100b83
50503 F20101221_AABKOV hsu_y_Page_094.jpg
67c340da5ca209a9231020b9ebe3b0e3
e7541d1a8f9a1ef1adaaa8d7f43686313819faf7
19513 F20101221_AABKPK hsu_y_Page_133.jpg
334cc29df623483daf031db64667454e
95e65d82d1079573c5491bfc40a6c4599579212d
36886 F20101221_AABKOW hsu_y_Page_095.jpg
a187dab4f8ee0451e8e3a4f907398b6e
f6655dbecad30e9d7c69715ee9a0df3ee5cd7732
43716 F20101221_AABKPL hsu_y_Page_137.jpg
773907bf50f8dfbdb9ca80129bad0148
a5ce6b0671d08b7254463aff991e9bcc35b3b107
20531 F20101221_AABKOX hsu_y_Page_097.jpg
adb5c7cfebc93ef459dcc9f54d7c4be3
5f64043e88a039eccbd9eae852d03978c6f9216f
922362 F20101221_AABKQA hsu_y_Page_018.jp2
1fcbfc89bad27261eaea59f772fcd6e5
77ba14c1f8c458eeaa49c1d77ef3b0495e132a91
98128 F20101221_AABKPM hsu_y_Page_143.jpg
f92fafe2c341aebfa5783f0894eeb5d5
27e7e058cc7726c286795c7e5101285ea5e30fcb
49800 F20101221_AABKOY hsu_y_Page_099.jpg
f5e076ca2f4dd07c801edc9378591368
e2e1054e56286ef21f74c46021fe00491bb8de2e
1047959 F20101221_AABKQB hsu_y_Page_022.jp2
4dfae9b778a4049ed0567f3c3480ed88
d9fb13e0caea5592f8db38a324431e458afde937
98800 F20101221_AABKPN hsu_y_Page_145.jpg
6dd029df07ed11718f2c43c795232b59
ca1cc86b1aa6c4429e7a646cc7768e06d254b352
116329 F20101221_AABKOZ hsu_y_Page_100.jpg
2c3ccfbad81ddb1f74f6e380e310d809
4fa239f93baead318128a8197730906f0ec67b34
1051410 F20101221_AABKQC hsu_y_Page_031.jp2
2679e845402281920371159e4384df8f
ab9d447cbbe3b465ba2c0ad0d15b56f0bd09e0e7
127922 F20101221_AABKPO hsu_y_Page_147.jpg
b724ab91495b06b497c71eb5009fa052
01dacf49fef59d00e3345ef8bf0f0cecd0843998
1051975 F20101221_AABKQD hsu_y_Page_032.jp2
893bf381867a304b096516e047d492c2
167a663bf75df005537baec0e29bef62237f3102
120655 F20101221_AABKPP hsu_y_Page_148.jpg
afff9a2eef34aa28bc713d7c52e85f4f
941b8f0f66932ed7475105403df1d3a2ec23548e
264422 F20101221_AABKQE hsu_y_Page_038.jp2
44e9c2748a80fa184201bce904abfd6e
319508988f1a5fef59b66aacda02433f40487866
128625 F20101221_AABKPQ hsu_y_Page_153.jpg
3bec85ce138f63b73c18a3f415da7ce8
d2ce566b3914e71758c42154acf2a8fa1cd62a81
222887 F20101221_AABKQF hsu_y_Page_041.jp2
908ae41e968ddcdab918399c98b14f27
2bf987d7b0d38aac71716c69a5070b24e0a33ad4
72659 F20101221_AABKPR hsu_y_Page_155.jpg
e51074dea4309ddc9253cd618c18a24d
928930795c527d7a9f8539660cfee5bd85f5870f
137087 F20101221_AABKQG hsu_y_Page_042.jp2
2f8f6e20a6a253d850003c50bbcd3a72
cc57e50b64839b271fa72b1e5099d264ca769d3a
22423 F20101221_AABKPS hsu_y_Page_002.jp2
db5a5c90e778054f2c0cf4e385e319ae
ba1682bb84bb34220ca0b3a9c28bbbaa054e066b
338006 F20101221_AABKQH hsu_y_Page_045.jp2
612f6dd9052b7b4635cbfe115d4d1566
2649c13b71f5c1f4b61db3365d9bb68d73efd619
1051984 F20101221_AABKPT hsu_y_Page_006.jp2
4824804fa3aaeea25f3df75b574b872e
a103be3f8d73d5ea0821540cb77a0ed1c1bdd3c9
353161 F20101221_AABKQI hsu_y_Page_047.jp2
a448f68235d0f6abc8ba633cc75fdb36
ba832e08e948856fc40307586fbe68035e4790e8
385954 F20101221_AABKPU hsu_y_Page_010.jp2
c33d878c76868bef95da66ace0cf4b7b
d740767fdeb22afdad0723fbfa64a3d0693bf9c3
1051919 F20101221_AABKQJ hsu_y_Page_051.jp2
9a104f691abe30b91b3976d9def73251
762245627a88ec38fbfc17e6bf48873f089267df
391098 F20101221_AABKPV hsu_y_Page_012.jp2
336a2b7bd337252615af08e322481b9e
1c7d6bbf01beeeb74c73594fba2905261e0c6c47
929917 F20101221_AABKQK hsu_y_Page_058.jp2
f9dbea72d3071a05f7053f092bf3b116
d279817ec3de681ae7f53867ab6be081272550e7
1051966 F20101221_AABKPW hsu_y_Page_014.jp2
7340e31378e77c42989e854351177604
02a625648ab61b4418eb409b606b6cebaf90467f
1051939 F20101221_AABKPX hsu_y_Page_015.jp2
bc9295be727b6b661e9d078a13c423ea
a7654f57eab146e666c44798bbfb10f06ea0ef3c
1051965 F20101221_AABKRA hsu_y_Page_103.jp2
3dc9d1da9f28a08943374982ded1690b
64a5c018c60fc2e664af32d1b6808b984b219fd0
1051962 F20101221_AABKQL hsu_y_Page_059.jp2
a25e3d709593ae76e59d20db98203d83
9d3667689bdd6c5a42b699e8fa14164ae946d675
1051943 F20101221_AABKPY hsu_y_Page_016.jp2
c27dd0b21a7b05cf40c6558c8893d460
ccee8b87e905109050ca4dae1b15916f42155e4f
855911 F20101221_AABKRB hsu_y_Page_104.jp2
03f322964d9e68b54584df9087829801
c5e6f2515c4f3c8d494430dc7dee36448ed86599
507292 F20101221_AABKQM hsu_y_Page_065.jp2
22cda610c678b243b41c98c51d32b540
5f459ba4d81a3e71d4717f79f4c8fa76bc3037f0
1039902 F20101221_AABKPZ hsu_y_Page_017.jp2
e8d842ae513e2bb0a813a61841500844
4381a7c39322ac2c5a11ca14fa35da85edf7a29b
1051977 F20101221_AABKRC hsu_y_Page_105.jp2
1543d73ba47a521a31aef0b4d50e40e3
27fa26168f83b9beeda9a2014381ee813865706b
429316 F20101221_AABKQN hsu_y_Page_067.jp2
8ce42289032dc65680414a2819881d76
2fab2a2595ec1caa5149fbcc21a2bc7c87836d5e
1044328 F20101221_AABKRD hsu_y_Page_107.jp2
bfe0c9ccdffdb35737da04679747b04b
41760d4e7f28560b33825d5bdd3828bcc8ab35cd
475195 F20101221_AABKQO hsu_y_Page_070.jp2
eabe4c754def7a12c48889ef1aaf71d4
f1e6fe6e7d32c040a0953ae971416c17207e7a9a
F20101221_AABKRE hsu_y_Page_108.jp2
cb49e2789c785080a227a9c6fcf80941
8fa95da2faae61e2e41b0c84ada1bfb6d1887cb0
1024204 F20101221_AABKQP hsu_y_Page_075.jp2
d54bff17c4b463ab9af44f47d434af2b
df86366495418ce24984230cf0d1a08687e022f6
1049039 F20101221_AABKRF hsu_y_Page_109.jp2
faed7528c0a373bd9e3f187379fd2a83
cff6963e72262a461618b4ab658f6cd629f85e17
640018 F20101221_AABKQQ hsu_y_Page_077.jp2
4f6fa2a810efe54065801ac48e2ccb95
ed686393066d02d3eedcf3a747c04a3117e0b66e
334448 F20101221_AABKRG hsu_y_Page_112.jp2
e3dcf85adcb92c0dde4132363463dde7
5aef581304daa55410221b165a3d2d8255282e17
1051956 F20101221_AABKQR hsu_y_Page_079.jp2
07cb4f2c443fc3443094778cdedfd2cb
55ae4cf5739f97f25504cc55cc9dc620a5c7b0e3
225829 F20101221_AABKRH hsu_y_Page_114.jp2
8ee9a7b91bbe109c4c65c520a2752953
6a96aff8e49e224aa7cfc95bbcbc542e799dfdcf
1051926 F20101221_AABKQS hsu_y_Page_080.jp2
731b0ba0ae82f466020a3eb388562fe0
f3026f94f46f33e821c12a98b306d7e452a85305
660384 F20101221_AABKRI hsu_y_Page_116.jp2
3a3faa80aec71ee83abca472a8f5975a
002dd70b3fe515e3e44d60adec3b64e35fb9e2d7
1051922 F20101221_AABKQT hsu_y_Page_081.jp2
9ba43cbf630eeeeab3b23420a0277b34
fa550e3f9f664cab16bcb10dccf7a507a280faeb
343989 F20101221_AABKRJ hsu_y_Page_117.jp2
0beb31b8fe8fb8bf0934b0648ae09702
84e9bec3b585e2d82f2beb07a9e5a86b8b839073
1051955 F20101221_AABKQU hsu_y_Page_083.jp2
5419129cf49113d94cc7e39b0c82bc36
617ba5d08db4072a80628951696aae21bbd37f83
1051938 F20101221_AABKRK hsu_y_Page_120.jp2
f7adba9ef973685b443179f5fa63a062
d1c75778216745200b60c49b160fa05278af39dc
891264 F20101221_AABKQV hsu_y_Page_085.jp2
49b5353c27e5a9d3a47d16f2d33eace8
f9ae8412b54320c9d4cb78873ea229d609c1bf6b
1038830 F20101221_AABKRL hsu_y_Page_127.jp2
729e9d8e1e5199da37452caaad073e01
13edd98a4b382220a0918b07c80bc873fa2d81e4
856724 F20101221_AABKQW hsu_y_Page_089.jp2
fcb864928224d8b01ee022a1bee93351
9bdae71693cfbcdc2fa6db38bfb5c43c7ec794e4
25271604 F20101221_AABKSA hsu_y_Page_013.tif
6fd2b294de2fd2170b85efe5cbcd8733
2b30a55c8651168c1063d3c926f10afc8e00cd89
502854 F20101221_AABKQX hsu_y_Page_094.jp2
f10b21b44dddebf50b21488dbc37e3f2
cdcb417c5b2063bd863abac1493575b9bf42a91e
F20101221_AABKSB hsu_y_Page_018.tif
c7c10125c5843d3eec87502f5d99ff41
4cfbe4f78028da2d35e09a301e58612b605a420c
1038104 F20101221_AABKRM hsu_y_Page_129.jp2
d5fb1545559454703f42d99119f63ad1
95c690957812efa0f09d67cc6737cc4ace95da4d
172923 F20101221_AABKQY hsu_y_Page_097.jp2
a0665bdaa758499a4214d3b1ce532bc6
0ebd634114933c1537306e00d32e8e3d2350b307
F20101221_AABKSC hsu_y_Page_019.tif
0adf296ea01d193049b67bf15eb23003
7e8ab55916566f3b575fb7e6643428cfa19940bc
304143 F20101221_AABKRN hsu_y_Page_135.jp2
a9cdce3698a410dbe770e0227d168801
2c9ed978031bd355274493961b67be38c6baa9ea
1051968 F20101221_AABKQZ hsu_y_Page_100.jp2
43c4f997541b3cbc9a12bc5e0c2d449a
e5a081215aab19c672d63f3e24222505cf8a596f
F20101221_AABKSD hsu_y_Page_020.tif
c694410ac53ebb3f45bafcb68c2fa669
ce36aa385ed197e3522e9e39e2bf6154e3a64b70
238180 F20101221_AABKRO hsu_y_Page_136.jp2
05cd2d6c4ef3d612d8d560f643476fed
02a0a12d864aa3ceb186ebf1fd2072d976b89baa
F20101221_AABKSE hsu_y_Page_022.tif
f7739783cd2b46de8ab9befa348123a8
b76efe5c06e36c9fa4db5f33f1038e866c0f8d7f
228552 F20101221_AABKRP hsu_y_Page_138.jp2
5ae98aacaa18247c812290396dccaba3
0b79ef314fe31676ac0b54b2a4b7a24aa88f58d0
47750 F20101221_AABJPC hsu_y_Page_145.pro
8614588842e3fbe052f0226731a7e1c4
04f49d00e7883a3bd84f955fb6c68c4734e52a05
F20101221_AABKSF hsu_y_Page_023.tif
0d66a6d6d0846be43fc2a1d4cb508d18
ff7c7cf11001283d07d481e6b31c7288fe48335a
431376 F20101221_AABKRQ hsu_y_Page_140.jp2
0cd7c7815c2edb32c8dc3ab651e42471
18c40eee3b30e3fc9513a344756c3046bcca4f0f
53512 F20101221_AABJPD hsu_y_Page_070.jpg
701de3abb82f4490483389faacf6fc1b
933b357dfb347122a778d12714245a83ae011dac
F20101221_AABKSG hsu_y_Page_025.tif
d04b66888e85f907f15352fb579a597f
aec4e38e66b48b32bfe872a8eb89c4dc95992c81
340477 F20101221_AABKRR hsu_y_Page_141.jp2
a0dddcbab3c1b2c48fd39b6e54c7ba0f
dd773a8fe0918fa53beadf8d7258602c3d0a3331
25217 F20101221_AABJPE hsu_y_Page_125.QC.jpg
a50e3add3803448c87d52f4a24a517ad
be4aa047b56b0299417a6205eeb6853c47ab6dc5
F20101221_AABKSH hsu_y_Page_027.tif
b875c21a16a0023c666fa6dab05792ff
6b9fcef6fc202b39cbbfd8909e1ef51dfbcfa357
F20101221_AABKRS hsu_y_Page_145.jp2
6b7edf62c3912151c2589ee049daac59
cb66d2c3e549cb549686f960f7f685b8ace3ff32
387 F20101221_AABJPF hsu_y_Page_139.txt
b554fe20091570e53ab363d386214e20
4bea38f946c6bef458ae4f00da103a9405b90c4f
F20101221_AABKSI hsu_y_Page_030.tif
02a5272bc514e7f3a51635137a23ce6b
ee60e113c747f93ff1d13952bcca62c64ae0f491
1051844 F20101221_AABKRT hsu_y_Page_148.jp2
a83f87ad7e39213cd4421555797b0ce1
874c4ec75f10813cd7cac104ac429005686709c7
1003 F20101221_AABJPG hsu_y_Page_067.txt
3cf2ef1168b3000c0cc2fd8583e8f328
656fb76ba3dd0e073b8ad38707a15484ad0655e3
1051973 F20101221_AABKRU hsu_y_Page_149.jp2
b38fee7c1c2ee47a5bcea807aed75d21
37e563b293fa905e0a976a4205820b442f9ff1ba
F20101221_AABJPH hsu_y_Page_150.tif
bddf4b3d7f4873cec3d24542f09be167
178cdbdd34e7a5737b0a1a72cc53be9a1c58cf3d
F20101221_AABKSJ hsu_y_Page_031.tif
b9e76759ae753ae5f7c78cf3489b4a81
a2f3f14c00d3bbd68a305062eb032649365131d2
1051978 F20101221_AABKRV hsu_y_Page_150.jp2
7fae717b450016f22e73484f4e141b8b
60660ed36596a24a30c69a99dd228d4ed417c9d8
5180 F20101221_AABJPI hsu_y_Page_040.QC.jpg
029b501ba600fe0874d40b54ab58a149
99a00e01786a50b42530ab6a1c3a3980fe805ab8
F20101221_AABKSK hsu_y_Page_034.tif
d7a7243bf6dffa8c077e2e430a49e6ee
1cc722644956568d2971a509df72128fdff95bd9
1051976 F20101221_AABKRW hsu_y_Page_152.jp2
7b697f8c5f662d2e665d9db6eca06ee5
d131de418cce9ba25c8a48e30b899f84c80eb40a
1942 F20101221_AABJPJ hsu_y_Page_126.txt
c25feb5e208c8b8299b3a2a93b3c676d
f586dff8dce93242a08de635ad4885a1c7f2a479
25265604 F20101221_AABKSL hsu_y_Page_036.tif
88078ddcee9eb220adca92f782c9ee87
33147515aaebf3a71ea1da00d633bf06a42a1996
770810 F20101221_AABKRX hsu_y_Page_155.jp2
dc2190564e36b87ca3b7a3a64c8fe865
c73a09dc628f211addb50cf730247e6bfadcd13d
F20101221_AABKTA hsu_y_Page_071.tif
2ef17fd73160d1fac0da98e2ea147bb5
c771e059160900adb7776c2d55c8d8f538c6397f
1039546 F20101221_AABJPK hsu_y_Page_011.jp2
4de5e70b5ec647136555c4cf65990594
e68e2063b8b8b3cc6ac45e6e10ee005535eee493
F20101221_AABKSM hsu_y_Page_037.tif
4a5b9aa818675e31b7deac53408f5fb1
6df040cede02ae85d3a23b70bdc31ccb1fc2b891
F20101221_AABKRY hsu_y_Page_001.tif
e1a88d26c253aff95b8084c9fa182009
70a0f2cb88bf123c726f7498a7f3727c70e58cee
F20101221_AABKTB hsu_y_Page_072.tif
ffb0f9a3a2b6ee9adc232d7864d3bcd6
67ffea1c938695d37adfa52a2011182d5eba3536
F20101221_AABKRZ hsu_y_Page_012.tif
707e13a055ea15db60e7a771a4ef13dd
12ce0c4ec1f66307d0bae3024ab0096cc983521f
13199 F20101221_AABJQA hsu_y_Page_135.pro
6a2f276eefa8caede56245f7268af0f6
b092f4e8ceab1b74ac3c5317a1d117da583bc31f
F20101221_AABKTC hsu_y_Page_076.tif
c02c369c8965b7179bc5a51a0da69a55
70a8ad8b0ae03d75c944db3ab120a3ee6e0a2d3c
12998 F20101221_AABJPL hsu_y_Page_141.QC.jpg
57234916b21b1b224ea4bfca1d8084f5
00075f4088c14f8adfb17f56932173cfee6227f4
F20101221_AABKSN hsu_y_Page_038.tif
dd55553f8a8cdf0677839f6ee9c81805
5a6a49db1cbe45d86d4040c11aaabc1c571fcccf
F20101221_AABJQB hsu_y_Page_097.tif
3b2c17abc7a2d68206b80abdcc9c6540
6cdf3a3d00106e71d20528081ef2570c46de95f6
F20101221_AABKTD hsu_y_Page_077.tif
86fef7975da4a32bcaf131b683c80c6f
30d463b01662a01cd73d63b1130fda8c9edcd59f
93890 F20101221_AABJPM hsu_y_Page_129.jpg
6c2242d491c521208ab76dcdc9250a10
40bc6f48fdd85d290dd67a16c38b9d15a37da1ab
F20101221_AABKSO hsu_y_Page_039.tif
a456490ebfd6dbe7413c4d73955278a6
2244253ee1cd6916633076e52e1306abd9890481
2247 F20101221_AABJQC hsu_y_Page_100.txt
c14419ff002c091afe376a7990b8aeb1
f318a62c471929ec6e5ed5cefa30d7b3b18b74a6
F20101221_AABKTE hsu_y_Page_080.tif
fb68efaae6370bc8fec65c10922f2366
4cee81aec0a353790717062cee156748306336db
93274 F20101221_AABJPN hsu_y_Page_124.jpg
4bc81dbfc86ae10d97596a44d6662b90
0ae7ddd1fcf6ab961df95a7ca49c25a53eabac66
F20101221_AABKSP hsu_y_Page_041.tif
e54247ef21502f5435057ac5ce83e8dd
daa9af341bde02270955686f2df6a7bdb5b34732
3967 F20101221_AABJQD hsu_y_Page_047thm.jpg
4430db428814d21aa48aae5cec39472f
be499e0647bbc0315ce8e4162090b9fe8680583d
F20101221_AABKTF hsu_y_Page_081.tif
967a87788bd6a3cea4ea49b6e43b2ac2
2aa2c0e1aea6d1c30658f443340409c0d8762fbc
8539 F20101221_AABJPO hsu_y_Page_141.pro
960081a2f3761218fe4112f37f27d9df
f900a8f12fdcf8db7ef55cc5086a7cbbe21401d8
F20101221_AABKSQ hsu_y_Page_044.tif
e64f947176679c63e9698b60440cead3
e6b84c601847c7138395a13a0d8f82d096c196a8
42956 F20101221_AABJQE hsu_y_Page_090.jpg
befea055f4801d83978ca75afe46f799
4d9e6cee5eb6e1beeecf739d730e6c0f9b1daf16
F20101221_AABKTG hsu_y_Page_089.tif
346ebcf7617bcfe166b5321da574759c
3e2510674df800a7d91fbd7731a3fd0305e15c03
85621 F20101221_AABJPP hsu_y_Page_053.jpg
c5e6c4e7240bf42bd0e2b2bdfbfccc1a
70e9fb10bca5ba9f92453dc52568fb524ba21fb4
F20101221_AABKSR hsu_y_Page_050.tif
69d56cadfb7c0e68fd5a78a341151558
d7ada3b4e0e7ea0e379875decbd5f0232e98e06b
35926 F20101221_AABJQF hsu_y_Page_051.QC.jpg
49aa72da29259da8a076e76decf875ed
52594fc0f0b4a6d569bec466571b464800d2fce5
F20101221_AABKTH hsu_y_Page_094.tif
0c576be73231ed5bc71e6bbe0bf4c20c
8b244af55c7c45998751cc162c6052ae1c126c2c
3129 F20101221_AABJPQ hsu_y_Page_012thm.jpg
40668d6f7637681e8db42cd305ff105c
2db684ded9d63de1d070c915255c435a7cfbafdc
F20101221_AABKSS hsu_y_Page_051.tif
a8beaf446e378a8005bbd7b19b905394
bfcdbe2ffe9225a99fe01f207763924635053a52
F20101221_AABKTI hsu_y_Page_096.tif
5daf8c1d569f450d6160e875fe49dc03
a4b9c2cc6d43d8a6bd798aeb521daeddda2bcc85
F20101221_AABJQG hsu_y_Page_072.jp2
ed96888a4ff7048a3f42919d22f39a7a
d320cb5a2d513f88b2b23178e4aa8a1a67618858
53367 F20101221_AABJPR hsu_y_Page_060.pro
3c3e02be4fdc71decf17a66283229f27
bfd8784f47ad270faafed5c7c5e7ffd6947631b3
F20101221_AABKST hsu_y_Page_052.tif
eb218a8f9c6be4a82081029e3fbe2375
87e768582e03319f7e228e8883d40bec39e61edf
F20101221_AABKTJ hsu_y_Page_099.tif
7dd0a0dc15cf6d7978c521dd66b29900
3f4c0058e6673e6d4469aade2e456f66d332602f
32923 F20101221_AABJQH hsu_y_Page_121.QC.jpg
095be193e9ec8e3ebb5c9249458e122f
8c7d05f828737bf9f8d5ab7b57693119cb1f7e4e
64715 F20101221_AABJPS hsu_y_Page_077.jpg
a57b9c9373a8e23e7602dfb872ac7853
8c7276cbef56159791c262b3741a52bfef5a2904
F20101221_AABKSU hsu_y_Page_054.tif
c0a9822a63549476419f2992e8c0f8a4
fcd6b73be139420fd80f787f51d6bd1b4b7d1fc7
F20101221_AABKTK hsu_y_Page_100.tif
84f4b229b81240db547710e0befc6695
541c687a8bfe189ddcaf945b66386f71b52e8338
4355 F20101221_AABJQI hsu_y_Page_118thm.jpg
f26b3d99b72cebdcffc37df39fb4bb90
671ba752901580adceb7e7f336960dbd94da180d
F20101221_AABJPT hsu_y_Page_083.tif
24271697d036ba7873930a09ad20f51e
b6fb04ba494218435390a422745c99282dc8f0b1
F20101221_AABKSV hsu_y_Page_056.tif
6036a3b354578d27fe5d0272f6e3875f
40c02b31bbd9384da376bf302fa4b830d1f8cfcb
F20101221_AABKTL hsu_y_Page_101.tif
c61f49ac2fede98792890c7f2ffce458
c6f69f22ad057109ae39021d1d664b7ae670824d
1051982 F20101221_AABJQJ hsu_y_Page_027.jp2
27ca0ec9a5b8b94847f027566f28270a
ef20fe64f8e8f36d2c2f0800f8e4ca43d3cc09e8
4294 F20101221_AABJPU hsu_y_Page_045thm.jpg
486e7cd00a565f6e53a8b17f137ddb2e
0cfa4c50df52ea52029c03ca87a752b4b1b874fa
F20101221_AABKSW hsu_y_Page_059.tif
d57b007a41c4f23e6347c756e941d7d4
7091c905c984e9b3431351cde76d5c4278ea139e
F20101221_AABKUA hsu_y_Page_137.tif
71df2fe653f0a95ca2db20b5c3c1ed3e
33b8fa5cf6c6395c1ffe78f201d31e07210e3d8d
F20101221_AABKTM hsu_y_Page_104.tif
95adcae1daf6a00db42e2f1ac04902ff
d6181154f5f40fe411672c1e2293d05a3c419d19
11668 F20101221_AABJQK hsu_y_Page_113.pro
99ffd6496202f964297ae3ccf0720795
ddf6bd86a114a84deb277174ade67ad03316dd78
797 F20101221_AABJPV hsu_y_Page_048.txt
db0e59a24ae178cc7652310bb3bf8fbf
c3ff611c50187849d3e38cc68b364c3a33cb2b94
F20101221_AABKSX hsu_y_Page_062.tif
7f002cb853765cb8f9fd4c622c5ae133
1c42692e58f993c29077e0cab92d31e149c93ce0
F20101221_AABKUB hsu_y_Page_138.tif
cef677d7359c5de42f43d8735616fe9a
e5a530199cf0089283be9945c5ed62bd5525d02a
F20101221_AABKTN hsu_y_Page_105.tif
d75ad857b7bc673a589add38f533cf89
b99369eb0146454defbdb35dd3c8c4ae154b9ab9
37839 F20101221_AABJQL hsu_y_Page_016.QC.jpg
366b1e8c6caaf90a5f8ab60102a11805
a7ecc44ac838ab55953e62e6ad70fb2a72e843d1
66253 F20101221_AABJPW hsu_y_Page_088.jpg
ab8083d40436860c360c7d360223cf67
952376dac7be6ac9b198f076377cc6a93a2a524d
F20101221_AABKSY hsu_y_Page_065.tif
336347a7386585ad3ebe6d0ab3710714
52702a5e10297123e8a355d4533f98a66e020ddd
F20101221_AABKUC hsu_y_Page_139.tif
7f9607e34cf68accfa99676b8d5bfedd
02016eee0061701a6e24e463a864813b8f61d0f1
9039 F20101221_AABJRA hsu_y_Page_119thm.jpg
9290cfaaff9d3aeff484d814252ef1da
c2ec00d292117869269146514af3f066c02ff030
2914 F20101221_AABJPX hsu_y_Page_138thm.jpg
59295d22ca41d2555c41035bbac4a922
94ae2931dce7a91d99b0a20ceba582607a94f39b
F20101221_AABKSZ hsu_y_Page_070.tif
34010b6bf8d50ea878afb51d145544f0
c6ee341cbc7ac7d0919746b3eb10dd2676bb5b86
F20101221_AABKUD hsu_y_Page_140.tif
8caf772bc02f28bdd49555916b16182f
fc03b4937d6ec99efcaa0ed13602df433f4649b4
F20101221_AABKTO hsu_y_Page_107.tif
33d52a44657144cbc78e6292fa8fb397
35b1d1fda43de7e0546f42cdb9263f67638bb878
1051961 F20101221_AABJRB hsu_y_Page_143.jp2
9aae8688357e43fe35a59d0ca59befd6
b25cd2478d6f281cec9356c02a91a9a9e7183012
6345 F20101221_AABJQM hsu_y_Page_077thm.jpg
ce4976b4802b491a73c6ec052e97b1b1
83094cef6b358449b13d88c778c7c6745cab3a30
96989 F20101221_AABJPY hsu_y_Page_031.jpg
f5f7ec589cd0fe7c8ea0dd12b9a73c74
201c4bd1beed6a2be92c01f5ea55993c285c910e
F20101221_AABKUE hsu_y_Page_142.tif
40ef15b53adbdba41791415c00a8a393
ccda702eeb7d1e011bf222628e1277990dbc4fd4
F20101221_AABKTP hsu_y_Page_109.tif
ed28dc8c3159b583c96e616e2360b4d1
e9b29ef298e37d498936f46f0b93e44799fcdf15
122624 F20101221_AABJRC hsu_y_Page_146.jpg
38ee1d38240be7a8ab1bd2effb580397
3cc216a29698311524895139e559520f6b816992
61688 F20101221_AABJQN hsu_y_Page_003.jp2
26f0c062fdfe969a47a9eda5d7b56431
e4d7a09839884b3ba1b96ccbadc0b21d2b4496e9
756440 F20101221_AABJPZ hsu_y_Page_007.jp2
46e4b4ea92a942081634bce82021eea0
f0d67ae4d320d13cdfbd8e2acecbbd3489585d3f
F20101221_AABKUF hsu_y_Page_143.tif
a925b56051e6c872b10f5a296cbe1ee3
f2b39f36599c19d99c654e7e68a3e6c4c12e5376
F20101221_AABKTQ hsu_y_Page_111.tif
8b6dfc6e32442fa2576c85ec97171eb9
7813c4c877aa595c188146519ff511aab5c5fed7
F20101221_AABJRD hsu_y_Page_002.tif
f3b694d687e2cf08b069884e9b401bf3
e04fd84b1d03e997fa6067496106fe189c2c5062
1710 F20101221_AABJQO hsu_y_Page_053.txt
8473ad6e0377728dfdfccc8ce861af22
aac2c5374376e1d12d27d4ab87008881254a927b
8543 F20101221_AABLAA hsu_y_Page_059thm.jpg
17de43ace45327f70990eca765a051d6
4231b5302396aad56c5c4abda0cd39a66eba2be6
F20101221_AABKUG hsu_y_Page_147.tif
d070cf946c394c686292e1b09bab3939
1c551c2c49bebd5ed6182b704b74300fe8a265bf
F20101221_AABKTR hsu_y_Page_114.tif
512934323f2bfbf6f7b44c104bcad5f5
e8b574ab76d6539b30f7353c79b7e1c6f80d6c93
420311 F20101221_AABJRE hsu_y_Page_137.jp2
70034a989064cf2c007b36a2c2e47382
742bc11a596a625281bac20a67350bb04a123eb0
F20101221_AABJQP hsu_y_Page_131.tif
6a1f7a9a59575ce7f12bbef25415ff74
224cb3f39ce5a4b2c8b6588fb91e6d6a5d0adfd3
984 F20101221_AABLAB hsu_y_Page_003thm.jpg
0998561d95e9cdc40eb6bbd103705e4e
63efa48eb69bb8bac096265c0527a9d55a4added
F20101221_AABKUH hsu_y_Page_149.tif
35efa710161ab7b4828028901b4102e4
f8688d7c0c17069c7642e557ce39c82592ea0522
F20101221_AABKTS hsu_y_Page_117.tif
e3ad0fda81be5217bf2a81619088d295
002085502802e80fb84645f6831057215403d6a7
387816 F20101221_AABJRF hsu_y_Page_066.jp2
387192e2dbfc97e995be999ae3960ba3
76d271b9c6145282be3f5887a48767ac9e988453
F20101221_AABJQQ hsu_y_Page_006.tif
42a709778800e63505c0b395170bbf10
97bf9ed1e2e2b62b8a7e61ab8a9b1f6b62c5708c
35335 F20101221_AABLAC hsu_y_Page_028.QC.jpg
053837c1a9e5aa4b5bf95471adf83d43
5a7d808c30ba0783cd3a8ff87399c41ea7729a2c
F20101221_AABKUI hsu_y_Page_153.tif
d6c33c51ff399866db95dbc13127be2f
afeea26c794db6675beb4207a3e974f16d6cad0b
F20101221_AABKTT hsu_y_Page_119.tif
ae3c3b01627edd5b67a50915ae423157
69efc90f9925b64d8db44127da675934226218e2
7735 F20101221_AABJRG hsu_y_Page_086thm.jpg
2872211c34fa8476643ec941753af574
fe5c4b80bf5615b172848753392a4b5590a2d91b
50432 F20101221_AABJQR hsu_y_Page_057.pro
fc8aec11fe35caa83c05c2350a68498a
7c609567cbdfa4688af9be909fcf91c1fac79680
37493 F20101221_AABLAD hsu_y_Page_119.QC.jpg
e253a5f52cd74cd4a9ae108159cd7a36
95f852bc9aa45c193d4d1fb226be6a1b191f37fc
F20101221_AABKUJ hsu_y_Page_154.tif
2462683a3a5b2450c7ce605f34b71f8a
a31c5009afa9e0ce1a3466b3415201fe64019962
F20101221_AABKTU hsu_y_Page_120.tif
0b33c0a1028ecd87e5e75bae4f971696
dfd0529a3147e66788fb253444bf584a92601407
818139 F20101221_AABJRH hsu_y_Page_078.jp2
c28777eb30a7310e8c6503e6d7ca371c
7282725665e717ee78c4f1e9d9d035ebbc5c1d7c
F20101221_AABJQS hsu_y_Page_014.tif
510fd2b5bc4a50b56ba892bb646847aa
1a79211e422bcbad613f49a2c6540a93ee8d294f
8289 F20101221_AABLAE hsu_y_Page_079thm.jpg
8c86a828eff3e1dfdd72ea685d20af5b
81d481485f980d9f4f8b0c66ad449557951f6b47
F20101221_AABKUK hsu_y_Page_155.tif
fbf11a551d14d157ce413d040fbc735e
9682927c33a969a473f326a1f307d91bf17b893b
F20101221_AABKTV hsu_y_Page_121.tif
67144564c0dd594535caaf1eed5813c9
e6db9bf1b690d1a6eebe358340ea80476a30675c
2246 F20101221_AABJRI hsu_y_Page_049.txt
28b9da0851be286f798a06d204e1fc91
40f5bafed2f6e60d52b6ec508c676427dae4e628
F20101221_AABJQT hsu_y_Page_141.tif
e602ad66ee71ce9b8e04e985ed6ed6cb
dcc7efc131563da62064e46661f29bf7744c5fca
8666 F20101221_AABLAF hsu_y_Page_021.QC.jpg
0f7d0d5fa1d4f15174d74d78912e42a8
ff8ff35c8d534db46d712295a147862f1740f5b2
7953 F20101221_AABKUL hsu_y_Page_001.pro
ed2ca7c7f0212b755455f2f7cc6b1569
575f3a703f68e86b6fac64043deb582db8876075
F20101221_AABKTW hsu_y_Page_124.tif
93d7e187464dda148aae58f37339211b
d2d6132523af8fe2d29fb98298f3e95e4f7225e4
1051983 F20101221_AABJRJ hsu_y_Page_147.jp2
a5def092df8f28a6d3d224e11dc96288
40e514a2b97534cb69cd780d44a57df225146bb4
F20101221_AABJQU hsu_y_Page_093.tif
269e12a5aa4aa6f9c54d276886ff8d7a
5c4a4da3c3ff5ef40ff065eadc0fcc32b62697da
15281 F20101221_AABLAG hsu_y_Page_048.QC.jpg
0c44459cdffc405909f194c3d2f744bc
42e35713bd0314d3922db9c72796708887367c4b
47458 F20101221_AABKVA hsu_y_Page_031.pro
d775c2494e8fb04b241ca130d3ad2033
cb82ef61ab79ff2831ff94c76e37252167cb6de8
723 F20101221_AABKUM hsu_y_Page_002.pro
2c466d10dc30c764dc6c74a3e5ad2c21
6ed2e3a13a81b4adae3fe45f5400a9abdb95a679
F20101221_AABKTX hsu_y_Page_126.tif
c37e798f1121d00739f5a50fe40c6d20
592adfd65b8beb9a39bb2adf19ed883b6d2e62f3
1736 F20101221_AABJRK hsu_y_Page_123.txt
b3612605f597fe2f08a5bb11f47f1dda
a19244b7203867f3b49298155be83a27580e6a60
5826 F20101221_AABJQV hsu_y_Page_155thm.jpg
67dcf3502c8572395df5e77bd7522846
77808fc4a6a3154dd981ac498fae0cec084975c7
9041 F20101221_AABLAH hsu_y_Page_150thm.jpg
df14a513520105b5dd920ee0e638ba7a
e259202012e0b44319c1ab65e0eac630d5ed82cb
53999 F20101221_AABKVB hsu_y_Page_032.pro
8ed12192a981b83a97cf5e80eeec3836
4d362746635ad6592d3c8c3fff7e6afe7152037a
50219 F20101221_AABKUN hsu_y_Page_004.pro
e383644221c976fd787d038e7ce7d667
f2fc36c48e9bf914b5761f731c362e6581fdf527
F20101221_AABKTY hsu_y_Page_127.tif
532840de385446072a0ad2c0b4cad2b3
e7f0906f8351a427fe12ab094183535114ea1df1
196631 F20101221_AABJRL hsu_y_Page_039.jp2
30a04a10ec3f05d2e4090bc090111bed
0e57c6532c3447a9a9b4109db2f89e417f4f8d67
720919 F20101221_AABJQW hsu_y_Page_088.jp2
d2bf70893d85281ff2b34aaf0ec0566c
614a0f0960e5bb167bac8c8370aac33838b61427
3057 F20101221_AABLAI hsu_y_Page_142thm.jpg
661408db70b16609ea7b7772e3317fad
6776b5a52af045e54abdd84d7c588771be42ae09
6672 F20101221_AABKVC hsu_y_Page_039.pro
206a6c64b8e504efe7134f24369a5730
f27ce0a4805d82c1cb3e9ef32a4dad448dafe97f
75196 F20101221_AABKUO hsu_y_Page_005.pro
33e3592a1690041ea62dd3ff9d4c1d4e
d203f444c4cd8b4958f31e46df1312e740c8132b
F20101221_AABKTZ hsu_y_Page_128.tif
e111ad497b251ff3fcd99600c8fccf02
b85ec5f93b7b41f46222ac2e76c88c0a5238e2ad
41683 F20101221_AABJRM hsu_y_Page_058.pro
92be0cc461fe19e9a5d6b31bc6e0296a
0ebd7ae917a0ce0d26974a9608fc1471a1c56f58
F20101221_AABJQX hsu_y_Page_043.tif
6f472585617254c9cae2ced0b5a66eef
3c2fffdd058e367355c4488f5febde0be5786d6e
7489 F20101221_AABJSA hsu_y_Page_058thm.jpg
c12827bfcda3e63dd480959749cad8a9
a7274a0c1ae1bb549a5220ab09527a8f3c924b1c
3063 F20101221_AABLAJ hsu_y_Page_037.QC.jpg
e36a12cb2b02f3bc7dcea45720b1ee87
4a235d05eada5550aa19592b225f93f5e3df07ac
1841 F20101221_AABKVD hsu_y_Page_041.pro
78aeb4a898b7aa5519709b5438ca4e9e
ed8a09856838666f6b9f45302cbefd5f5e40c81d
41487 F20101221_AABJQY hsu_y_Page_018.pro
6be4fe189b67b7603211bc37c635bd42
92511adc66726bab0f4341e364638fdfb80e7992
F20101221_AABJSB hsu_y_Page_087.tif
fe2ba57b7aa058ec905f54e3a3bbbee1
dff28df79f719f040cd3ec2936ee463d1f8a0568
7860 F20101221_AABLAK hsu_y_Page_054thm.jpg
04e72f97eff5580522f47347d5e37ad5
49504449d73af40532017774cd9098e288c97ad5
15155 F20101221_AABKVE hsu_y_Page_043.pro
da828551b5b9036e44e9dc22bd8945be
7ee16e2cbe1c2029d90679d6692f80a06e97f401
53065 F20101221_AABKUP hsu_y_Page_007.pro
b39f8efecf6f218e150c541c5c0a1793
4bfe7271e2e029188d06a2018fac88f02165f083
9402 F20101221_AABJRN hsu_y_Page_147thm.jpg
238182f49bd0048683fb04901b0e3b34
a2905dbe7e43d38d91d10e453cf4b55634e39070
F20101221_AABJQZ hsu_y_Page_009.jp2
84b4db1222925b7ee538a0e9b1b9bbd4
863f7f22a9014aa2dd16c38443d3be48d3451b02
2664 F20101221_AABJSC hsu_y_Page_021thm.jpg
b1036d07314d849023eaeeeb1fc255b4
ee362688f0a210cd4b30a00ce9b90f1f332edc4e
8395 F20101221_AABLAL hsu_y_Page_075thm.jpg
09173ca8482adea9460a20b1cf41ee7a
4ab869a8c1e486d7589dee556e59f4e89c24a5df
10044 F20101221_AABKVF hsu_y_Page_045.pro
45630f17b28e60ff5673191bd8e9b5bf
256414cb2e14b8dfc6d5ca6b871aa60ae89ffdcc
65139 F20101221_AABKUQ hsu_y_Page_008.pro
30a70babb742b1836e482264fdd389b6
067fccf3f7d71de92033659bf8b7885bbe16913d
F20101221_AABJRO hsu_y_Page_090.tif
bd45242d5934f3a9be6c5bf438aaf420
f90bdbcebdeedd9d7c9467e7881c179b407eac26
39383 F20101221_AABJSD hsu_y_Page_141.jpg
265d252ec5d7ba7615bf1d5534be02c2
01de045258eb45a02ce113794e6541198888ce8b
4820 F20101221_AABLBA hsu_y_Page_137thm.jpg
c7f123ebc05301328cc0ba9a948049f5
26073219cfd78207d6d542871ca11310504dba99
1350 F20101221_AABLAM hsu_y_Page_039thm.jpg
291e6a9dd5a31c9b3bd0d255ed3a8196
7a699c5ff9a930d634135f1cddb35467a85d9ba7
9177 F20101221_AABKVG hsu_y_Page_046.pro
3cab6f974e42508824b885892451c04d
7bf9f29ea36848573ef56a335e3f8c246fdb8f41
63354 F20101221_AABKUR hsu_y_Page_009.pro
3129da01b9108df2268fbff0ee798321
14c7f2366dbce0b69233976c6633a4e89d3f3164
61737 F20101221_AABJRP hsu_y_Page_148.pro
47599b57e9fdb14a28ef37b69bb0a7af
1ed45f3a607237290257359a2499b0aef913d10f
4756 F20101221_AABJSE hsu_y_Page_098thm.jpg
f914233045c6203588e4e4edd2c584cb
2f5eb9a7f085e7eb1985c7f4d63691984df66cd3
13262 F20101221_AABLBB hsu_y_Page_043.QC.jpg
5dbab1f1a6d2a0f546785ebf06cde687
6f967e8cfd59582fef2d17f55b0281f77d35fe2f
28344 F20101221_AABLAN hsu_y_Page_085.QC.jpg
29f3a5b6296ea2ab7b08c88236b11691
79d1164fefe039288234401d22eb39e7707b8272
12841 F20101221_AABKVH hsu_y_Page_048.pro
8d751031f3c291edfd4a6c8d062bf6dc
3d22df08451d3114530b41766aa34668b2da304d
44971 F20101221_AABKUS hsu_y_Page_011.pro
07db85964b65b8526b604e4992064e92
fab83623e2625636127f717afe83c11006160324
43465 F20101221_AABJRQ hsu_y_Page_064.jpg
39f6221118f305cde7decef6d08aae56
71493f2477547ccf602dde5282b21436e48d8d53
5329 F20101221_AABJSF hsu_y_Page_154thm.jpg
47523b8365a6fafca3fe088fcb7ad9d4
e5fa78f9d9b0d812d3779ccebadd5353066a754b
35244 F20101221_AABLBC hsu_y_Page_049.QC.jpg
a70a252421a832d0df1aad6af9a23dd2
1d33886cdb7596ac0f86dc3f62d475fae3b7a3fe
1986 F20101221_AABLAO hsu_y_Page_003.QC.jpg
95494e82ad6530fafaab0d51e60581d1
c7b117700f98600030aa65e74a34315fa76dbe71
52055 F20101221_AABKVI hsu_y_Page_052.pro
c7e9efa5f4cd8501ee42b77abb8f3149
5df3de402d29491cb13a2da3987e0cdaae5ab6c9
49839 F20101221_AABKUT hsu_y_Page_013.pro
ec35639aa9677c66516e04dff7af8d72
2adcef22e8591312a131644d2feb8887a3c62be8
961 F20101221_AABJRR hsu_y_Page_037thm.jpg
ccf6d989468eb96281e1631ffc80f5ea
923c08df4f1a9666ee3f31da8a6f8a5db6fa9c8b
2106 F20101221_AABJSG hsu_y_Page_023.txt
4574a22b537ae68f989dcd43051e7c1b
7864266bfeacedeb1b840006714ac082e20995b9
493 F20101221_AABLBD hsu_y_Page_002thm.jpg
58971015643ed2bf43bd60e5a61c3119
e21e31d78ec42ee363e4c7149a9caad35230a910
3814 F20101221_AABLAP hsu_y_Page_093thm.jpg
57108725657e39f791d7234e5eefe55d
1beb04cb5d4385928a2bc5a6d7841dcfa46fa076
41210 F20101221_AABKVJ hsu_y_Page_053.pro
d02d81d00e645fa3b3740ef07be5adfd
67e14895ad275764a8fa49e4886c0fcbf5cb369c
46408 F20101221_AABKUU hsu_y_Page_017.pro
2066df8afe8d3be6a9b7f3ce6be69b05
36ce2bc59c3e21cb2aa4fad985b8d4937b37eb6a
717 F20101221_AABJRS hsu_y_Page_068.txt
30126cdd6a6c98c8fdd7a98d5d40df30
d4d11d26e40e97630959e965eff89efa53b554bb
F20101221_AABJSH hsu_y_Page_054.jp2
37e7331051e23a352baacaead80f75e7
cd05dac5893ac8a3b6f4a0cec4abe71e928cf6e0
37151 F20101221_AABLBE hsu_y_Page_147.QC.jpg
57cf6ff0ab5c17d28d996202f67028ce
0bddbba5577828e0248fe272f1ec733b2fa0db6c
17023 F20101221_AABLAQ hsu_y_Page_094.QC.jpg
9947719c93b2ae708886ea0ae5217b7f
7f7dda2a578bae7650caa53a8bf83ea86db05f82
49321 F20101221_AABKVK hsu_y_Page_056.pro
908f52b30e255a8bc953d9ee426258a5
d4ce8e95165086d00e5f31ebacc6c5a2f4b05a2b
12573 F20101221_AABKUV hsu_y_Page_020.pro
a6be54c17465d94dff39009d0d82815b
cdfb46f4ca3c841f3de0c009cb6e9ab85715591b
47836 F20101221_AABJRT hsu_y_Page_048.jpg
c1c1feff14323597223bbd340992104a
ece9acbb490b6ff89a08cf8f8be0ec93e6c57603
106507 F20101221_AABJSI hsu_y_Page_061.jpg
10d4a435c6ef754511c1b852dd7afc9c
bb14b806e5c7670d80a8488a44b135167577a7d8
34497 F20101221_AABLBF hsu_y_Page_059.QC.jpg
aa92a69773ebae1e35884f730b63f2cb
de67ab68783b7aac84c676b1cbca36128b079d8b
2685 F20101221_AABLAR hsu_y_Page_113thm.jpg
2f3e1b42f93c2fdae47316d7705a84d2
54cf945d3a68006928f08523b3a04f85e34774e9
53798 F20101221_AABKVL hsu_y_Page_059.pro
df0ef20cb6d63002c72d547f845d9104
726cb6d453d4c36daf26bd6066023f3e037896f4
53468 F20101221_AABKUW hsu_y_Page_023.pro
094f4c5f8020a2ebc07fc1d4e2c36315
87ec15a88a72e0c7155f3d6c546c96b2dbf97412
39370 F20101221_AABJRU hsu_y_Page_010.jpg
41f9e676e01183eb30dd38f57b325cbc
ab74410ce220b76da0af1c9170859ae31266e801
124685 F20101221_AABJSJ hsu_y_Page_149.jpg
0120b6925812725806ffdde56a98de2f
02702b2e1d13bd1ce08172bd672ccbfb9333d374
22358 F20101221_AABLBG hsu_y_Page_088.QC.jpg
373c98ca19ad8658ef7dae8b95bb05c4
1dd5e295374d72c1dcec411cf29e6399a33ad1e8
4693 F20101221_AABLAS hsu_y_Page_039.QC.jpg
756b8dd908e70f5e889bb761fc1363b0
4c9b7750e6d2c5f7a0658565976f6c82a9b4f056
51054 F20101221_AABKWA hsu_y_Page_101.pro
84dbe4910959bf658f5c0627a2d2313a
47afb9cae4f63a49c560ce93c5713b02401f098e
55497 F20101221_AABKVM hsu_y_Page_062.pro
8a4b5a99edb27888dddccb7acf43fd42
15b616d545be23725f43437983ed5ac6bc788af7
49243 F20101221_AABKUX hsu_y_Page_024.pro
6dc9b8213538f9abbaefa60b3a7d2611
634d4b4aca7a0e603c59b9108af3da113e529af7
408 F20101221_AABJRV hsu_y_Page_040.txt
2773f30a77544a2525ef9d4539dc5a5c
13701436b4b072b1e4748123f13e360e0f0e0a43
52324 F20101221_AABJSK hsu_y_Page_073.pro
33fa1b5f98f75ac885ef54437b3db154
3f1deb1e882342f0c43dd68521106213fffc3fc1
21249 F20101221_AABLBH hsu_y_Page_154.QC.jpg
8965b12678c5925d620106b1f5a07dcb
7cbfba1953136c4c05a7e90d831e3e6e681a437e
11540 F20101221_AABLAT hsu_y_Page_093.QC.jpg
28959163397b86f00bd3c47870b2a9ae
2198d70eca0d948dde01b7d90b2f5eaae3ca5f4d
44993 F20101221_AABKWB hsu_y_Page_102.pro
4bce92da74cf17a5d3f88a143b29d3c1
4c81064a267935ff2917e420f863fae77608642a
20918 F20101221_AABKVN hsu_y_Page_067.pro
ea39677d822910f30d0d906a1785d045
024f3e00d8c992063f0fa70adfc1eaaa4c86fa39
47793 F20101221_AABKUY hsu_y_Page_027.pro
1fcf4d84ee5aab2d5e4036c87fb250c8
529a23f00a507f579331b6e0a8dad404ac00ae90
F20101221_AABJSL hsu_y_Page_133.tif
bb2a1d9d90e673ccde0f8a2cf6535d18
48f300bc83c7a0e84270d0976fbbef38ec2eefba
32097 F20101221_AABJRW hsu_y_Page_031.QC.jpg
a6fec2139ff1cb10837e6d28fe9f4311
ffeb7f503932873b431b53dfc1a93811e1148c25
35910 F20101221_AABLBI hsu_y_Page_120.QC.jpg
25f7c0463abdf8f676904829069279e1
a8e1c6d77d86564fe41a9bcaf71d4a1e0b1bd587
16554 F20101221_AABLAU hsu_y_Page_065.QC.jpg
7a54f955f7aa1e9615822ddf86993986
9eaed4bb594c47565eb0fed421c8aa48ad0989d8
40026 F20101221_AABKWC hsu_y_Page_104.pro
375d1d222d7aa55dfeb15769071f4dd5
d5eeca4248314352d3b387030a34ac0ac2ac5a84
10082 F20101221_AABKVO hsu_y_Page_071.pro
66f38369a4edbe0f981305dc8268b87b
cd0eb3fafde7c001daeeb6d5fd7033abd2519c7d
52975 F20101221_AABKUZ hsu_y_Page_028.pro
6e57ae92becc14469d558001465052fd
16482dcff03bd26a9dce9abf44e3e3c77b5ea431
1051969 F20101221_AABJTA hsu_y_Page_073.jp2
f15d67788d2d781ce7534f6b5351f6d4
e6e71ea08ce7571863e642bc60f3558727b29872
5253 F20101221_AABJSM hsu_y_Page_037.pro
60d5e332a8e990901aba66425bb51c98
cf7aee0c700be29da3667995e94ee7eacb04a43d
8254 F20101221_AABJRX hsu_y_Page_017thm.jpg
4e3232f3c06fc6c086ab137a654fc6b6
0aed8ae5afd7211e300f5cb8b240c0f5e1980f4a
8741 F20101221_AABLBJ hsu_y_Page_015thm.jpg
a511e0d14bfd26b03957d38680efb052
2366eefe3a920aa0ba8deb9059b73f49d294e80b
1871 F20101221_AABLAV hsu_y_Page_041thm.jpg
ba22acb75974f8dcc417680860a1cf66
dd47551d0a8a73b565406ba3517990c7f779e1d1
52132 F20101221_AABKWD hsu_y_Page_105.pro
5365662b7570474dccdac82833d6f00b
e9fe0ba9b86392a9fc58fc233ee2dd94c2a86cbc
44116 F20101221_AABKVP hsu_y_Page_075.pro
2eb5f39c9591b9a53fe29ee91c93e9f8
a919a6c675ecc8fb89ed333bd175f80fd54c6ecc
88821 F20101221_AABJTB hsu_y_Page_006.pro
bc8c16a1678d173f8efbaf33b1f4dc34
b64aa8eeec09dba1935a0aa77bb3cd5f428d9e3e
53303 F20101221_AABJSN hsu_y_Page_029.pro
5487580a79ba934fedc70ebd787815ad
9f5c477d5451c41dbe535ea8c9725b24c8854571
517 F20101221_AABJRY hsu_y_Page_020.txt
c7e32b4a4b1d2383bcc45cb8fa8b2564
5cf930dac50b9e71f9f36e58aef58c5eb2eb46ad
8601 F20101221_AABLBK hsu_y_Page_121thm.jpg
d3b74d851772f3e9f06dcbe96ca9cc46
2dc7a2f7a22c6986d1a2aec79365c9f2e6a31f7c
24090 F20101221_AABLAW hsu_y_Page_155.QC.jpg
0ae685d0c85df67085ea3e9a8ea9abcc
eae496ea0dcab5c2a982843af9a5880c1dc426a3
46236 F20101221_AABKWE hsu_y_Page_107.pro
01a424c71892a9530feff80ae1009b49
264d63048995b1273543054d4b17d7ad2d2b187e
101776 F20101221_AABJTC hsu_y_Page_079.jpg
e682d3f57983ffcfd5fa5ed2882e807c
eae76d8606a03469a8899e836f587cfa3e7b5d88
2195 F20101221_AABJRZ hsu_y_Page_144.txt
ac0bc0bb02d9752621cc7f3bb22fd52c
9153f3e31a0adf2f8b29bdc10c1379fa8eb4c11b
34971 F20101221_AABLBL hsu_y_Page_023.QC.jpg
6d9f4feb1221a7c3d156b06ea2bf69a1
b0116d04e6b5962e14549e6e49587d09ca387cad
36476 F20101221_AABLAX hsu_y_Page_062.QC.jpg
263a841400908dbeb4d66e3cfaa8d037
c28373df1083833f5330ee19c9b5bed83ec7d39b
47247 F20101221_AABKWF hsu_y_Page_109.pro
37efd452a626b438d3e9b811a1ac3693
f79653d795133a9cbc4df43f4315102e75cb39d0
24699 F20101221_AABKVQ hsu_y_Page_077.pro
8ac81323c6dcd6ce3642fe68e825e763
cc30345c1f083db8d60072c31a239552d18a061a
1580 F20101221_AABJTD hsu_y_Page_122.txt
db7ea13212859bf3be26ec6ac60e2f6f
c25fb9f768d2d9f25f8dcc4b5d4afc57e19a2dc7
103299 F20101221_AABJSO hsu_y_Page_013.jpg
1383ed47fbe58f25565c5408fec334b0
667c4bfadbca6f0b02d3a6a3d7f81e9a7dd1ac82
5573 F20101221_AABLCA hsu_y_Page_041.QC.jpg
50f1bd52315606efa6947f95a851b110
67c8f7570c2c2a295a12b4432286f7cd20d240ca
35822 F20101221_AABLBM hsu_y_Page_061.QC.jpg
2a405c9b19755efeba97f4b301287153
7325ee754876afcef21c1d2a64e88ceb2986adcd
37045 F20101221_AABLAY hsu_y_Page_019.QC.jpg
828df7a42a54b9ea88a263c911cdbd4a
0be8804fa650a58c7e36f4623fb3d4daa5b32621
41194 F20101221_AABKWG hsu_y_Page_110.pro
96ebcc808e6606a6627f18146ef5e24a
9a721cbe13c085d4dca50f626b7a67df37f677ae
54110 F20101221_AABKVR hsu_y_Page_081.pro
109cafb9c4db296ee0935d249cc01d9a
6f8a0d0e04cf1cb1ee521a12c9ad55eaa0ee49a9
2556 F20101221_AABJTE hsu_y_Page_150.txt
99d77a1a4c1a9992ddf22a8e7a1a94e3
2a8ba6bb167823d31af098576de1c68461c5544b
1756 F20101221_AABJSP hsu_y_Page_075.txt
67b16aa97c5a1dbbfd2b005ad012769e
ad467cbe3c0cdc378ae2e990f6c11aef3583c79e
28868 F20101221_AABLCB hsu_y_Page_123.QC.jpg
f2ee2d01af70ca68d9564c3177f5e16a
888bc97ac3117666ab1f1086270dbdd150881a09
8950 F20101221_AABLBN hsu_y_Page_023thm.jpg
927fde61fa809ecd090a0b24d96feb86
d1734342b5903c0c9d529e2eb493784023cd3167
28523 F20101221_AABLAZ hsu_y_Page_058.QC.jpg
9f049dbb26ef9f71f7f4634919d2b0ab
a4b66fbd639e8a214e79d96aedfbed6daea03a7c
14377 F20101221_AABKWH hsu_y_Page_112.pro
faa6ee354e6b1ea8587a3b6f19211ee6
747de7bd22358b629a8e16becf8152e5ef83be76
54976 F20101221_AABKVS hsu_y_Page_083.pro
cbe243414858087b20e572bafe62fdaa
6637985f44c0bde778d234ac8fe51ad19161179c
873521 F20101221_AABJTF hsu_y_Page_122.jp2
2c3f6368b2c2f99ef760ea6eb2f14b9e
929929bcb87de259678653c2b58077b2e9d8c9b3
1051970 F20101221_AABJSQ hsu_y_Page_008.jp2
eac11909562d5845fe308b2b29072dda
a554a9e2fd946976f3279ebc0247d2bbc717017c
2204 F20101221_AABLCC hsu_y_Page_097thm.jpg
70923b8788a9ec54fed8611f0cd009d5
a6fa7815153d67f591e5007900ee6f07a94b5717
35354 F20101221_AABLBO hsu_y_Page_073.QC.jpg
c2af4657ab4e2a3b4d2e1f42c1e754f8
cf33e1406b8447a5f3fa108cce64824f6bc2f887
23934 F20101221_AABKWI hsu_y_Page_116.pro
59c945f2089bbea414a325b63135a1ea
fa1995916c0544de94e93761959cb7595056170c
36446 F20101221_AABKVT hsu_y_Page_086.pro
308808abab8fd8f0157e44a534dadf50
5a8e9e930ede7c0bbc973b7cd5f4432fc2408a64
1051974 F20101221_AABJTG hsu_y_Page_050.jp2
43144381f4f180f2688506471c17e1f1
6730a318106b20ef73c16474ca3c5a37e6dc483f
F20101221_AABJSR hsu_y_Page_144.tif
65e3c1c3b60ba352821f684e1f0145d6
e381f0196ed8da399a499f8d9c92c0b8f046eaaa
2281 F20101221_AABLCD hsu_y_Page_096thm.jpg
ddcc07f194551fbc06a6c7d56de1761e
052857983ff471f891d01b9c44875b8e1a3a0001
33599 F20101221_AABLBP hsu_y_Page_084.QC.jpg
ac3a063699796d67c2c22b0485423dc4
146caa8dee338e074b50704c52fbc2f28f1d52ad
5532 F20101221_AABKWJ hsu_y_Page_118.pro
6ae4f2e2dc62acebfb08bc4d71fb7fbd
7ed82ae6e241e635500ef41bea4f479d4baef97f
53197 F20101221_AABKVU hsu_y_Page_087.pro
a6f491836a1afaa4a50f0c1489413ac2
dca785f49b295831f5ab4c7c772337e8314dab02
7113 F20101221_AABJTH hsu_y_Page_111.pro
7a7631cabb6fe885706f498324cfdd28
fc40ded3ebbac52ba1cbebadf016474aa5cc4d50
57763 F20101221_AABJSS hsu_y_Page_055.pro
e1f45076243112f5731da6130934f4c7
391840ffb1717eaa65690dbccf45d2dc4995ea59
7661 F20101221_AABLCE hsu_y_Page_143thm.jpg
0193f3d967908544a6a647f3412b5e28
b81b2f82fa6ab57e4b6ba1a6ea54da83fd2c95f8
7873 F20101221_AABLBQ hsu_y_Page_124thm.jpg
94389113f1010e1f65fd8c09ad19bb74
e561c32f00cc686fac91a1519632f8bd2dcb1137
37627 F20101221_AABKWK hsu_y_Page_122.pro
cbfbf658a77a7395794c85073904322d
b34f60da356d54a09cfde044b0a45e23b7f86cef
31751 F20101221_AABKVV hsu_y_Page_088.pro
2273a94c4aaf0c3988b8c71333137a42
9c8b894ff238e8d1aea685aeb9f9b3b0d77bf937
32789 F20101221_AABJTI hsu_y_Page_054.QC.jpg
2e681e123966c4890edaacf06675d9c9
1099efb9466fd4bdaf4342c9141b4076486cf271
1051952 F20101221_AABJST hsu_y_Page_025.jp2
4bce6fc47b6b3a9d21d719624091d9c3
04b46346ae1a5069052e23620879e7a6ada01c45
33621 F20101221_AABLCF hsu_y_Page_079.QC.jpg
54129c26918eb0249cd4d785b9cb7c46
c592c2ed56bd0ff7c71de4c97535cd421bfcdc54
8404 F20101221_AABLBR hsu_y_Page_126thm.jpg
b4eb4f61a86c1811aea9c78b2e4abeb1
2348173aa5e30ff8304034ac008c836c40101298
41896 F20101221_AABKWL hsu_y_Page_123.pro
08ca0966269f4b177c331528317290c0
10cd9cf5e2b3230717e83958945cc27535885453
25436 F20101221_AABKVW hsu_y_Page_090.pro
123b29e911163fd1f96e67a1dd7c0ee7
c2b5ff12dc82651cb37cd42b36dd19da1a25a4d5
88277 F20101221_AABJTJ hsu_y_Page_123.jpg
81e3d566c04c0d142bc32e3541048468
24ae5842146b36db697bbab6140b6688785ea0cc
233328 F20101221_AABJSU hsu_y_Page_021.jp2
8a04926a5b81fa18d287e37b8ed74587
4bb5c9f6d84af92dea69e4ea25e20c3571f736ff
8675 F20101221_AABLCG hsu_y_Page_081thm.jpg
b134c0b46030cc2bb2c01efda0bda303
2de68a7640f254c13ead293358ca892ddff6f654
8517 F20101221_AABLBS hsu_y_Page_052thm.jpg
a12bab7db2ac80b36e0bf3e4a04cdb47
3277f9d53b09b373984531e0b04c545a0ba9aaa2
3267 F20101221_AABKXA hsu_y_Page_005.txt
4f60798361c78b1f6cafb37338115521
d70644c2e41ef1ac8ec6dc89998105f5872b94e1
44700 F20101221_AABKWM hsu_y_Page_124.pro
737c3369134784ee91ea34ccdae1fb6f
f370b091314a3b992350c8e8819f158ac57ed27c
9123 F20101221_AABKVX hsu_y_Page_092.pro
0aaef47053a8ad64eb48092e3b39fb26
5424b664a1a6c1729a7690565ed338620f26cb9b
F20101221_AABJTK hsu_y_Page_119.jp2
d9cb25a120133ab2a238cfe98ecf9783
118a40307af4994fc5c9041bc6e548539fdbbc8e
31651 F20101221_AABJSV hsu_y_Page_145.QC.jpg
fff9775ba50c64c5e3cb00b43c825e5d
e23754d5f778c4af7719d2dfa4058953a8462ca4
36158 F20101221_AABLCH hsu_y_Page_014.QC.jpg
29be2a8f94a19163248665017beadfce
695029b763a6dc3caa3272198a0f28f100fb6e0c
1919 F20101221_AABLBT hsu_y_Page_036thm.jpg
03ae009b269a10009d66dab7020867d7
3d79bc96b47b2d69b2feaab3cb2b3c97aedff0bc
2367 F20101221_AABKXB hsu_y_Page_007.txt
5461f92edcbfb6aa3f711a4f9e21e62a
8f261b28818d79d5d775806e5ee7642ddbb246b0
49040 F20101221_AABKWN hsu_y_Page_126.pro
fa225d83e6978efcad957102d2bbf587
b7237933d8856a7fac9a4265006f0529ec781a0c
8882 F20101221_AABKVY hsu_y_Page_098.pro
2b1af1680436c835e2f2e835848de5b1
211b20b0aff236a70a8dbbce584f9cbcea635d11
30947 F20101221_AABJTL hsu_y_Page_022.QC.jpg
f15445baeb881207b155fb0ead52129d
0b44b1ac735e67b8906320239cbc99d400168ce8
F20101221_AABJSW hsu_y_Page_021.tif
4646a5bd3e999d1cd67c213af7ea5a79
19d6654285759e8570a6d6202edfabfd78abb19a
8598 F20101221_AABLCI hsu_y_Page_087thm.jpg
199ead2863ddbfeb6e1e031a55dee5af
4a6b44a620e2ae6163938ee3a2cbc8aa5c954cbc
35226 F20101221_AABLBU hsu_y_Page_026.QC.jpg
40e637c6021639c9c71120ebd6b99687
bd11f2589938a11dc2259813db3b7f35f2f9e431
2673 F20101221_AABKXC hsu_y_Page_008.txt
ea51d67e61ceede210a10482c7e2c241
9916c7712509fd5df32bada7379754fc6dd9d150
31935 F20101221_AABKWO hsu_y_Page_130.pro
f454b0c2cbc6978534b753d70558f032
3a7ae0649f4559a183206ec21f4d1aaa4d027349
55026 F20101221_AABKVZ hsu_y_Page_100.pro
4d69b0681e0d28e0193a1a73b777bbbf
99f661b816b9d11d5c7a4a96ce79dc24cb422136
5689 F20101221_AABJTM hsu_y_Page_088thm.jpg
17af7dbcedc257b9f9910c386cb3dedd
4794ffdf92790c66aa6ef20ccf00c885746b226a
1876 F20101221_AABJSX hsu_y_Page_054.txt
39e90f5d5491d0dc57be1c6ff3510ceb
94f9b0fe9f636ce676f6c8701a2285d1729daba0
34894 F20101221_AABJUA hsu_y_Page_105.QC.jpg
937a79c81d88381ee855be5fb7a1c152
0326df02c72792526198dd7516905908338adf5f
35644 F20101221_AABLCJ hsu_y_Page_106.QC.jpg
3785e7f326168d2e3ee6b26790348f03
f78b75ee88d395e3f8f9fe20f9f42afa390efd2c
2240 F20101221_AABLBV hsu_y_Page_136thm.jpg
4fd1c63af77c847adcab5fc49d5e94b2
b723acee033fc425437ddc41df2f9fd31da231dc
2003 F20101221_AABKXD hsu_y_Page_011.txt
ee85ff24805beb60c4565e5aa597d447
2e2df8e41dcdd24810121b8077613b442928fae0
46691 F20101221_AABKWP hsu_y_Page_131.pro
102d750e10b9f4d8ee031ca124600269
6e7184d16ba482762771d8276f584f1f21b41ea6
F20101221_AABJTN hsu_y_Page_092.tif
c5f37bf061dfccd098a4f64e8b323f35
600e13111e6d5839fd0bf85500757e5ab52fbafe
13584 F20101221_AABJSY hsu_y_Page_069.pro
fd7b4f952d711d29fe6a16b153b6f6e3
117f839c9c7fb359002b579953f0803c0fa00282
9117 F20101221_AABJUB hsu_y_Page_103thm.jpg
df9c0368f74caf2e2bd58e288311e7d5
8275dcd531be61e6c3a2b90d3ff01cf2055c5398
15958 F20101221_AABLCK hsu_y_Page_099.QC.jpg
810a04e60cc7598246306f4a61397a27
75b3d7ff4b027c93bbe023bb9d9a804fbf7a267b
8954 F20101221_AABLBW hsu_y_Page_153thm.jpg
4857365a6af5253812a93dd3a90d9a48
7bb3cb24468ca0b2addc02db8e8e44b8cdde2d78
688 F20101221_AABKXE hsu_y_Page_012.txt
f0beafb658fbce8f0c2b20ff4c477e61
e6ebb740a920ac01ab2c56d139144f2d328a356e
6337 F20101221_AABKWQ hsu_y_Page_139.pro
448d738794a33e3ed37e2744071924cc
6671e56495e82d2d4312ccf1ebae61d51e8b4b98
F20101221_AABJTO hsu_y_Page_057.tif
1fd6d5beb4e177da5d544fa74379ccff
8c620ce0785b123d9f8171d910e7589ba74e770a
F20101221_AABJSZ hsu_y_Page_132.tif
0d47b189e9d91bf8d6708ae37512d066
a1f1c1b8c45133231ba9d08f10cf6215577951d9
2265 F20101221_AABJUC hsu_y_Page_055.txt
bbdf13d8ade12639ca0a9cc65dcc848e
b013f6f141e907609266f23b69ec9a6f4ee1e43c
35170 F20101221_AABLCL hsu_y_Page_148.QC.jpg
f43d7cae1269c1be713f4d29973fbaca
b88dceead6a3aa4333ff0dff135ca63646313cdb
27947 F20101221_AABLBX hsu_y_Page_110.QC.jpg
de690accb09851ec1370f907f8d68c70
9c18a7ff7c4058143172a6dfc4126a7c66ee214d
2193 F20101221_AABKXF hsu_y_Page_016.txt
6d22956fa939a3a531da1e4d1ad87ab7
abff0f1069ff3babf1009e36b2ed11c150f9ad2c
1494 F20101221_AABJUD hsu_y_Page_040thm.jpg
b5b1631b0930984f36c492e88509668e
6169e1c024feac79d3ef12d91cb961dddb14035f
228450 F20101221_AABLDA UFE0021983_00001.xml FULL
c04787f73c2e08305dfa9b9f75ae7d40
aa360f2b967987e33e515e7b73ee1d9363d9a23a
5601 F20101221_AABLCM hsu_y_Page_116thm.jpg
cc87e749c47e38d4506312c02f640857
9d4a33c4d0e3f94675b1b9ae6e7e6011bd56a3ed
36499 F20101221_AABLBY hsu_y_Page_015.QC.jpg
5bf1aabe9078f0ee87c6d3967610d277
6ede94c1ca949f8a42936c2dcf055f7e7b4e624b
1898 F20101221_AABKXG hsu_y_Page_017.txt
657aa5ea8fe172d8111a8fd07d1aae9e
aa24a12eac934c4d4c1ad371fe5a4046ee273b6e
53955 F20101221_AABKWR hsu_y_Page_144.pro
13a0f841e6abb89f4a133bbc51d79657
aaa748be979fe599e9e5190b4487aad28cedb8c0
367280 F20101221_AABJTP hsu_y_Page_068.jp2
4a094b0023bf3de45a87df0a7cd102cb
efc0fd9eb0e949c7576cb6619cccca16a50e7911
F20101221_AABJUE hsu_y_Page_066.tif
bbb5dd5eeb8fd716f24216062e89d244
c3f8cc64e0cf1059c722c4838131b73f7d817fad
6742 F20101221_AABLDB hsu_y_Page_006thm.jpg
572b93864176bc066a407d958ea171e8
78bdc9fba88d17a7689332914eea837e5dbe9256
7468 F20101221_AABLCN hsu_y_Page_085thm.jpg
104f4adfcbca9362f87a697116bc631b
03d9f3fb9a539baa3af6f9e5e50dff761c2d4e23
30859 F20101221_AABLBZ hsu_y_Page_009.QC.jpg
e86205bd537a6e2d71f3d8b7b45eac78
78fa0198ce1da316da55b35a8105c17ee60643ae
1966 F20101221_AABKXH hsu_y_Page_022.txt
7d95461ad19c241301823e30d97df86b
4aa9f2bbb9430b779d4168abe815949c5a6b63dd
62554 F20101221_AABKWS hsu_y_Page_149.pro
cf476a5b3de75e27adeb9c8beffd127c
c53b59087736ee4abdc23d61775f8afafbc40169
35065 F20101221_AABJTQ hsu_y_Page_004.QC.jpg
8eeea98cfa92a03cf53d8bfecdd6ce18
0c8e4641bb507fa2f141e9a7996f8ca0a64e1387
F20101221_AABJUF hsu_y_Page_019.jp2
eb4647fa419968311d98a48970e6710d
274c09c70c08caa1568124b52d884967f735ef37
17798 F20101221_AABLDC hsu_y_Page_007.QC.jpg
d386844bebf72391b388a94bfa50ffc5
f766663901669acb55cb3535616050e99fa5a8c5
29448 F20101221_AABLCO hsu_y_Page_053.QC.jpg
09270c00c21511aef5c5815f0520a928
1ae6961aa446956bb27435b7ab08d5aaecd5f1d4
2080 F20101221_AABKXI hsu_y_Page_026.txt
e38ec68e4a709784501c7a12a6018d8c
b23ebabd346906654cc24ba97c1574dedc8504d1
63415 F20101221_AABKWT hsu_y_Page_150.pro
16adfc6b006f1ac80886496f0902f5d2
7cb8b5b4959295d9c57ba039e7309bc2b6a78897
1051964 F20101221_AABJTR hsu_y_Page_013.jp2
3840eaf8bbe1e18ae284a448d861591d
51c7b9d1dc8995566d10ceb6b2992da98bca6a7e
1368 F20101221_AABKAA hsu_y_Page_130.txt
1b7fc4e4dbe2c38551b460f52cb59187
f410b918f226adf202565bea3f9e51a126809021
26750 F20101221_AABJUG hsu_y_Page_104.QC.jpg
9f72378f4da6e6353a31dac0539f184a
c83c3440b4036652033b691ea33e2a61bf84fa4c
7743 F20101221_AABLDD hsu_y_Page_008thm.jpg
d04f4d5b27d7c750fb2e9fed93884042
322d6d4ff4125a2b5f9965b9424867a314a81b0a
7301 F20101221_AABLCP hsu_y_Page_123thm.jpg
22dd400b3b45b2c8063988fa84c03591
cf81fb55393db059c9bf652ef987af3f75c8cb34
2092 F20101221_AABKXJ hsu_y_Page_028.txt
dda6dba6694590e0c806ee28889fed9a
8123b881949400e54bf5bc7b19330a6d7d2985e6
62168 F20101221_AABKWU hsu_y_Page_151.pro
94bc7d51106094c6ba87d26c25f6b82b
e8c055321ba1ed377d9c43a1e1b63e874a637608
84246 F20101221_AABJTS hsu_y_Page_110.jpg
2edb2729a4713ee1f80053c3c9c4b80a
50ea4246cd6eee88e090a0e8c10b419cad5c4a0e
2560 F20101221_AABKAB hsu_y_Page_009.txt
94373c484a60466d274a012fc9756cff
970399dcb728e38c4e6ad432bde443245abe9914
8632 F20101221_AABJUH hsu_y_Page_105thm.jpg
5b81b6b899c95e9f66bc4feaee70e81b
c02e0eb8d687d3e926af3459ecac611aae258f25
3037 F20101221_AABLDE hsu_y_Page_010thm.jpg
d9bcc242224271d2e9138ee9b2f063af
23f98f1688fdb149b92af146f293e2ad3426dc19
8542 F20101221_AABLCQ hsu_y_Page_028thm.jpg
75c5159c70d9878951296ec2c0da604c
6f49384899a1bf519051d4a32e64df7e46f759a2
2095 F20101221_AABKXK hsu_y_Page_029.txt
afeefe78220e2c0025a0de7b9373091e
6ba7628aaf5fbb43f28a0087ba9aceae19be3d67
36553 F20101221_AABKWV hsu_y_Page_154.pro
193d2a396c1dead3e9086aea540895eb
a5a20b3eadee0a6a36842f5069f3a923bcd10e96
304125 F20101221_AABJTT hsu_y_Page_095.jp2
2769282bc5bba216db3d40ffc13a6505
2dd9ca81a7450e7de87a0c5d2519923806e04b2a
34824 F20101221_AABKAC hsu_y_Page_135.jpg
7651deeff6a44c4dd062eb6286d83baf
145b3fff34a847e761d3485e8f46b11eabe4459f
436361 F20101221_AABJUI hsu_y_Page_098.jp2
c69c2453d2c706be66c64f92851231de
b4d8ec520ed3872257b4cf470392c45796ce8b42
12993 F20101221_AABLDF hsu_y_Page_012.QC.jpg
9053b07049f397fb2ce0c6e4c6c118e7
d8df4e190c51dfe88c4df741df500c91bf71d548
8949 F20101221_AABLCR hsu_y_Page_108thm.jpg
9ce939a7531d6f66ef405a87d8ca0584
6476a2b716ed5005fb51a9a8473d2ad397f5879b
2131 F20101221_AABKXL hsu_y_Page_032.txt
164f87c1f80b9a011974462cd0db5b87
21c4eed1dce75d1f79cf31a210b37994277b06c7
490 F20101221_AABKWW hsu_y_Page_001.txt
c0c940914f3fc7c04dfebfd3616a7365
0a5cd2e61923bef7a20b3ca00ab50881b369b5ad
1717 F20101221_AABJTU hsu_y_Page_018.txt
0680bf3170c6dcb4c64433866b71ede9
beb523a68ac89c5adb4badde84ad97ee274cabd9
12863 F20101221_AABKAD hsu_y_Page_044.QC.jpg
d0afecbc0d87ff34ebe737a68e439da8
5ef22c860a7c44f06390bf657f23974e277d5866
33993 F20101221_AABJUJ hsu_y_Page_050.QC.jpg
765738c1d0c245b87990ee0f410697e0
6281aacfc3c4bbf48986296c2995c65a859b63f4
8677 F20101221_AABLDG hsu_y_Page_014thm.jpg
dc2ff9f5292de35acfdaf82fa81deb25
47664456ec6ce8f0474af377d6022b71f47cc9d5
6657 F20101221_AABLCS hsu_y_Page_097.QC.jpg
9c332816e58d29e7bb68139b867a1973
d9b8550f44cabf71eede431dd83ef9e10a5f6bfb
1249 F20101221_AABKYA hsu_y_Page_070.txt
fa6780c4e65b4ed66b4f85ef84cb1e15
f9198215a925f184727f3658d1c33ce604ab11e1
873 F20101221_AABKXM hsu_y_Page_034.txt
7305c2ce8b26576959b62983baec5de2
f24f2801ac09b5827a00e7460649917914227427
88 F20101221_AABKWX hsu_y_Page_002.txt
6d465975d1cd38d9f6d3bf828b07fefd
81f147b0a173b96631d3e662710b863731027a9f
419514 F20101221_AABJTV hsu_y_Page_064.jp2
746f05a8935e70b74495f1c44664355a
f37dde635d987843c11ea71b68eabeae4bea5a95
F20101221_AABKAE hsu_y_Page_035.tif
13a90789d5e73edeaaa2eba2e6a4644b
d70a55a7136108755b51af7349c8d3a6813c497b
243431 F20101221_AABJUK hsu_y_Page_040.jp2
04f46bea759893cdc0f845b1e84d1ec3
6ba6643ac175a309bcf93c627aba777a71db728c
8885 F20101221_AABLDH hsu_y_Page_016thm.jpg
ff268803282881f8046ecc8fa117b438
d7bec4b4593f849e6609bfda0cb91c927140171b
13018 F20101221_AABLCT hsu_y_Page_064.QC.jpg
4eb0ec88440853c78667afe0f9def651
ffc44e8a664345b98306abcbeb7ceea270ebc04e
636 F20101221_AABKYB hsu_y_Page_071.txt
d2035a55c09ffc1715b5ea6f25a2d80a
c67db3934625f7f83a546a76ccdd46ea7e714370
225 F20101221_AABKXN hsu_y_Page_037.txt
e7773296075e9b1012357197283dde57
dba8f6215b80cb289195fd8ed25eb8bbfb2178ac
158 F20101221_AABKWY hsu_y_Page_003.txt
436f0491f55e1ee8c6536c332f000bd1
8d85aa6ebb6e2e5837112fc008b39bc9c913e247
17154 F20101221_AABJTW hsu_y_Page_040.jpg
cbb521238c992a95542df0ea528ab7c5
08a906d0c89a3aa88f595be046c72589d1b47989
F20101221_AABKAF hsu_y_Page_084.tif
8a71fc0d454734905db10dc2e90f58ff
1e7d90dea6f6581c33ed5974e70923e46ebc1f5a
32505 F20101221_AABJUL hsu_y_Page_131.QC.jpg
43f9b6e7d97c898f1972c3cde467da50
3c56877154d029f69514b45578b1add7ca842363
8627 F20101221_AABLDI hsu_y_Page_019thm.jpg
f749c2dfe297462c0c90ab6b5d750b27
5411df1fa00155720e5725b044acc1ffa7bfb868
15894 F20101221_AABLCU hsu_y_Page_034.QC.jpg
efa80a1c2912e6cf3ee4a52fef892554
90a96072fa074df3f1d7825677f9660c7b6af70d
2316 F20101221_AABKYC hsu_y_Page_072.txt
6e24ebedb585de14528aa620aff5e1e3
b650cd4c24e54f62abe85b9045ad1f6d93f4f12b
264 F20101221_AABKXO hsu_y_Page_039.txt
88d2c1d2d67d38a4c84c395ae0c326e2
1af7ad1d93a4e8fc42aade09ae1a59a0bd06d078
2031 F20101221_AABKWZ hsu_y_Page_004.txt
ee0dc94c7e7c78d0ad31079b552818be
4d4f455e7c38fdae8903c5a398d45d7be7c524bc
8860 F20101221_AABJTX hsu_y_Page_148thm.jpg
9b6e10d0ad64aab47d8e38dc82bc1513
a78fc78f4516ef2d917aae0b25c15e8b0892025e
12599 F20101221_AABKAG hsu_y_Page_095.QC.jpg
6d736cfa4f8b0da13151594392ef653e
6f877fda3404f80e055bfb537ec4fff11eb13e4f
8999 F20101221_AABJVA hsu_y_Page_051thm.jpg
aac739a34ab83423ce5653839beafea7
0a0e0bc05a0d5450ec21676b3d4967bf61b309b5
30840 F20101221_AABJUM hsu_y_Page_127.QC.jpg
99edf92b6becd94876cc2a032a8e4bef
87e97f91c19126b83ba65d0d15caa15cb86b1bf0
36567 F20101221_AABLDJ hsu_y_Page_029.QC.jpg
9f2b223c5ac5aadf77f570829514faaf
c63f860bf2ac566593961a891f73acca1099bd2a
8733 F20101221_AABLCV hsu_y_Page_073thm.jpg
e5b9ba304aae2024f9ae8ff1577b8559
21d480c654c63a9d7c6247b466b78477b3d01963
1664 F20101221_AABKYD hsu_y_Page_076.txt
80b09a70ecb44e845717fae2acbef614
13dc97e637bcb2f6d13de2a734d5b97c45d76029
115 F20101221_AABKXP hsu_y_Page_041.txt
662d958d0e3d375d8b4cee175f451223
37130232ae25e84d53e7de5b79c45196dcc92e70
271066 F20101221_AABJTY hsu_y_Page_142.jp2
9c685ae456d30ae85d941767a06de00d
55aa99bd52093d89ea692d50e323eb7668580218
541 F20101221_AABKAH hsu_y_Page_038.txt
af0a91e5cdc3f2c7b3d9cb4357764bd1
2a61166c183a010f8aaa6aca0336048d7c98d0c4
5018 F20101221_AABJVB hsu_y_Page_096.pro
e395f713255905fcc562d02a7ac40cb0
5d96ba76a562397902fda6d089c7e0e1c09e66dd
F20101221_AABJUN hsu_y_Page_060.tif
2bc5f2d2496a051b1b46872f1d560cfc
ff0c7ecc32aa348b235cbb9683356fbba3e84578
8649 F20101221_AABLDK hsu_y_Page_032thm.jpg
0617ad582b2b87d2bbb2b10ce191ab2f
f6fd74ffd025d1093b68355e6fffbef1c04bf343
4186 F20101221_AABLCW hsu_y_Page_095thm.jpg
ce218e199f29f3f529e7ec0142f80a73
054979b45ab901039870135a8fd77132805522ea
1480 F20101221_AABKYE hsu_y_Page_078.txt
b849ab07a3a41b15d505d6eb75ff495a
0607d028a6039cb02d4364845cc35e330b16d379
754 F20101221_AABKXQ hsu_y_Page_043.txt
449d8a06fe7642d4a48b3817bd0d0bba
1b756914479a6e67fae0c6cd0daaa03f90072686
35716 F20101221_AABJTZ hsu_y_Page_083.QC.jpg
981c8f1771d3fbdcb796455e247648ae
61ef982e98d8cf282e2298f45ad87a6bca8613d5
9021 F20101221_AABKAI hsu_y_Page_067.QC.jpg
0925a7e3214de2c681e66a5cf2322598
795cc7c4cccad8ffe0361deb5d26a1e0b043b80b
121611 F20101221_AABJVC hsu_y_Page_037.jp2
093079fd18e785fee3d25bff960e0624
220060df99cdf893ada6b8916fbd74488ad7b573
10543 F20101221_AABJUO hsu_y_Page_112.QC.jpg
19e555f4e490a8ea6a2787d5318db848
2ee62cc38cec678861252a19e2c3c5f7a1c1e2b8
34535 F20101221_AABLDL hsu_y_Page_033.QC.jpg
d9342bf452b060628bdb0a1bc96ca48a
23df3b1abfd45255c2624b093cd1988a615c4ae0
1784 F20101221_AABLCX hsu_y_Page_042thm.jpg
92ab09054f8446ab2b50f70ef7562625
a0d39f2896caaf799e464633652d22fd4652fd69
1992 F20101221_AABKYF hsu_y_Page_079.txt
8f65f8122012853530e858d3e72e91fb
7004c78b9f79678669244b4c079fdce15e60a1fa
750 F20101221_AABKXR hsu_y_Page_044.txt
43be80821fe8a5c6e14cfeda6260b0bb
857b77388bf7a858b41b810c91733825ff2451c4
F20101221_AABKAJ hsu_y_Page_004.tif
df908fdaff6c8b4ce0067e1a988b1ff4
c254ac8a7adad355a6f4a24cc21c4735ff054e03
1021414 F20101221_AABJVD hsu_y_Page_124.jp2
3273d9f1831b96aba4cb9e18b8d0449c
e549d5a2deb5b1fbf3346384e87f1d914d2f81c9
50499 F20101221_AABJUP hsu_y_Page_079.pro
b7f4519cbbfc4dc4f098c5c6f2c8c145
2b1b1afd5a60460458cd8e89925963228a6dbd7c
4932 F20101221_AABLEA hsu_y_Page_070thm.jpg
81e97011d6aa0dbf1a72a79cbe948e0a
153c41f4476099b3069cd0a3b1af8ed53a9966ab
4192 F20101221_AABLDM hsu_y_Page_034thm.jpg
ef849236b1a72984bc3b8111a6270835
5dd904db3a1b2e2ccb8314914d317b629d821902
4325 F20101221_AABLCY hsu_y_Page_099thm.jpg
4592014260432fed79fcb8af3cfc9944
88c246361b36f1f9d800a20ef190c1be55e44a1f
2129 F20101221_AABKYG hsu_y_Page_081.txt
bfd150f2d22426144efb8cb81ca09b04
8a982c59d5b89f3d79aaed427e0826a5094c8534
F20101221_AABKAK hsu_y_Page_063.tif
182b7bf11814ef3cc956898f14db76b1
b0d29ff196aea5985f7ac861a23a956ed760451c
8142 F20101221_AABJVE hsu_y_Page_080thm.jpg
90c7e185d15f2cae1a11c0b68b0e1c88
87d2bfc1996c0046e3b296a1b95c8df5ad7699a3
31021 F20101221_AABLEB hsu_y_Page_082.QC.jpg
4f649e98c1e98b5bc15183cc41e55f9a
1e9f05defc50c010a4cadd73b5c27525b005d449
6258 F20101221_AABLDN hsu_y_Page_035.QC.jpg
8d0e5f0ea63b19cbdccf1ad4ff4ed5cd
f11ec3d8a5c916c0c0fe968e5150d73046613924
31725 F20101221_AABLCZ hsu_y_Page_017.QC.jpg
b3106696940eb01bc6972b2fbf0eef09
10abe26d98ceccd866bc3d7891a2b69786fea732
1805 F20101221_AABKYH hsu_y_Page_082.txt
9df093e0733284eddc2aabbdbaa9e189
1d53d881fd7e7d10b0c8b6f45cddf5ce13701712
666 F20101221_AABKXS hsu_y_Page_045.txt
1c92428067fc60a76e624474691f8773
cb9abbfac73e186d5533152d9547c1fd0e494594
2245 F20101221_AABKAL hsu_y_Page_019.txt
924d8b730d730539fa9f44da8b75a415
f6c674e411d580a5b2718da56ea822684d8cbc5b
18650 F20101221_AABJVF hsu_y_Page_099.pro
5c54131b2975a3eeb4f0d110bc1ea046
580f7ff96e6aea8bcbe1159fc15338210443bc11
99414 F20101221_AABJUQ hsu_y_Page_121.jpg
fead8fe334ef75cdeffbb32b7b534b66
a5adca27c5ee5e4b2f4e2e67e09fcc28ad13363a
17250 F20101221_AABLEC hsu_y_Page_089.QC.jpg
a7f16467c30bc1a131f9ce22928bf2b4
2d0c466d776246b4feb3eee088e298ded81a4c38
5136 F20101221_AABLDO hsu_y_Page_042.QC.jpg
c4020fea150d5adf368470a45aa5ab16
bc3c0188fbe48f9e836d1dac91d73918db256670
2163 F20101221_AABKYI hsu_y_Page_083.txt
3111106fa05db710d983de6cd2c9d605
27b1f2aa1dc64aed030f9ec48108d74c0a13c854
641 F20101221_AABKXT hsu_y_Page_046.txt
b3afead154c405262826c6f883a13686
75610cca43dcdc606be0b5677ed981201188b2cf
392323 F20101221_AABKAM hsu_y_Page_043.jp2
cf4775fbc956edfd0936a9180ededbd0
1f26cee216a1822fa4bc19c70deddc39ccc7ac76
37538 F20101221_AABJVG hsu_y_Page_125.pro
b34d59c7dc28fa92fa093fca1b16c876
3c2b0a99e8d7a8bda0e10c44d99264319a16ce73
52931 F20101221_AABJUR hsu_y_Page_108.pro
a523f3a52b20f9046b2a1e84e2376071
857a364ac5f2603b3b158f349bc23e5c69ad36dc
394354 F20101221_AABKBA hsu_y_Page_069.jp2
814fc96862bf55a23642de8b9741e8fb
bb0c92bb1b5e48414379441d8a77fb208881ffec
5082 F20101221_AABLED hsu_y_Page_091.QC.jpg
06d79a192bc8997b5f7f55f04fbd77b4
c72098e33c3884e2488c71d14eed3e8e6b193668
4207 F20101221_AABLDP hsu_y_Page_046thm.jpg
9b7adf77f27dcf661e7744b1c015db19
01922db0f49c336434aebb70ce1baec0fc8f7557
1726 F20101221_AABKYJ hsu_y_Page_085.txt
5079e5fee36ae6e505a65c6e8b28c004
edd38398e3d1c1d4ed0dd3ae13213fe8dfe81c5a
848 F20101221_AABKXU hsu_y_Page_047.txt
52866c629f18b629814d7ee6be57f227
e6c21b54ca232284166b9b967af66f01927f0ff5
F20101221_AABKAN hsu_y_Page_073.txt
e795c7366fabafacca7caaf4bb53a62a
7ef10f706dfe562bf7eef7f9f6a366f56d4f7c6d
7631 F20101221_AABJVH hsu_y_Page_001.QC.jpg
a1d0cb1039668a80312c43788d786b8e
d1236d3d599c21d4345b35918a0e67674c2846ad
8285 F20101221_AABJUS hsu_y_Page_129thm.jpg
0d13f02ea786094d38479fd168ae62f3
c48cb05766687cfd0c72f5008769b813aebd290c
110462 F20101221_AABKBB hsu_y_Page_144.jpg
c5aec6491cea44f0efdd71c1f9bef611
98b915a477f2b1a258e6b702060881c99cdadb68
4897 F20101221_AABLEE hsu_y_Page_092.QC.jpg
1372ea613f1a49097cb3efedfecd6245
3c22168151efc0e7f3e463421384848b9b03e7a7
8858 F20101221_AABLDQ hsu_y_Page_049thm.jpg
a9d0c264c2fa45162d937e7222430f03
3b51ab5d7e213fa31e902adbfc7f054ab8f0a510
1265 F20101221_AABKYK hsu_y_Page_088.txt
d4eee191e35b1e8b3114d6c3e86363a1
abcf2b735a76208144230fa24acb45aacdb4a49c
2047 F20101221_AABKXV hsu_y_Page_050.txt
2eb201f96dc234f73c42b16f40bd5134
abc38744984ded197fce2f1421079aeece1e7484
34321 F20101221_AABKAO hsu_y_Page_074.QC.jpg
d37b5072267664f6a304f23b123ec00a
8a65fc2f739d0247f1dec6fd7df78c389e608ef9
35489 F20101221_AABJVI hsu_y_Page_087.QC.jpg
1188d962495f7cf5d0c9cfb6b630aebb
ae4b7264907175f9136244382859b000a352d58f
2169 F20101221_AABJUT hsu_y_Page_120.txt
a83c69a385ef221a7f92c2ea8b9d07e1
ba1c55f1437e4a166926c8d3c7e363207d9a1008
44483 F20101221_AABKBC hsu_y_Page_069.jpg
50acafcf3d93fe20a6f72db820ddc6ed
6d70220bc8fd81be0e4f410a7711c71325c3cbb8
8102 F20101221_AABLEF hsu_y_Page_102thm.jpg
6feb476ebe00e28dbb9a2c3d269fcf49
d3f21350c9e2f1ec4e4a53482d7dffef3166e86a
7424 F20101221_AABLDR hsu_y_Page_053thm.jpg
cad802c02146c6aefc1c53928a758589
3f4cb1ecb932d9cde90dbe0abfc6f16f8487836f
1513 F20101221_AABKYL hsu_y_Page_089.txt
767430f798e40328010afa3c33d2850b
0963564cb1971e392144d9e83a79dbdf54af6596
2049 F20101221_AABKXW hsu_y_Page_052.txt
c631b6c8951761440ae1495264ee89cf
3b3272796dab73f7c52b958af395e942aad870fd
52228 F20101221_AABKAP hsu_y_Page_084.pro
21162add96037bd7ad81a4bbb9eb8292
85cdfd8cf3f9e224337f5af3dfbda66720d80478
91678 F20101221_AABJVJ hsu_y_Page_082.jpg
6ade42da3ac625f176dbc0543d80982d
04e3846ce7be1c0f7b68ab5f8c23ecddf8d3da2c
13232 F20101221_AABJUU hsu_y_Page_094.pro
d980881510560dad4dc71e93fa346d06
a8dc7ad54e334822c435b1f149c454a1b2d9d22e
108480 F20101221_AABKBD hsu_y_Page_033.jpg
e62a5e75b8dc43dba1b402e6b45ee85c
19095d27d02a03cbde07c681b504c30f04a36d2d
30685 F20101221_AABLEG hsu_y_Page_107.QC.jpg
d6dbf19429c8e5884c6b8b080f052104
0e8290b9cf0b82157cc2e4d6f202f4617ad4419b
37901 F20101221_AABLDS hsu_y_Page_055.QC.jpg
312417ba1108606405c03751e9dd451d
95a0999cd633943bdc6b3186baa0a6be4a049d68
547 F20101221_AABKZA hsu_y_Page_113.txt
7afbdd6cb602efd620b07e0521d76dc6
d53c31a2ae65037a71622d6bc13eeb0c86d3a59d
1054 F20101221_AABKYM hsu_y_Page_090.txt
8e0395ae95cdc997a114657c054c46b5
d02d8b1249d5a4228fff533319a65b8cdc5d5702
1996 F20101221_AABKXX hsu_y_Page_057.txt
154f174b643b588a9e3bd82e414fd4bb
c2cba04ffaa55e210e46dd2957eb8f8dfe51861b
F20101221_AABKAQ hsu_y_Page_122.tif
2e8d5b395a21a79d2f263f4f49162c57
b2ee15a86e2f33ebf41d6cc8032080f30f725aea
1883 F20101221_AABJVK hsu_y_Page_131.txt
4aa8345228ed6acae711b8172d8516cb
a9aa7b4ba116b829e39493c0508fcc549b3b6ca0
5327 F20101221_AABJUV hsu_y_Page_142.pro
4fc60f4d235a90d6cc68272d25c81565
186afc04fd1d91b550d0ecfe6fc72b58f21aec04
8178 F20101221_AABKBE hsu_y_Page_082thm.jpg
422b2f367bfdf96f39e948fbc44bbaee
59cbfef68ef0702e884ce4e4e4b2d56416bb069b
6924 F20101221_AABLEH hsu_y_Page_110thm.jpg
c7d45ef027540eae481afc921b118ffe
96d0fc46f66a062933985ef3d4905e2d10ca5e3f
9038 F20101221_AABLDT hsu_y_Page_055thm.jpg
bd4228760181e470e03454d23db2d326
8c59a11bb3eb4a1b2221e341f6d5221175c98924
643 F20101221_AABKZB hsu_y_Page_115.txt
99cf0ca7bde09f29cf0eaea92c8d3ae3
02e85648b8b824a5d529d5dc853db484c62c8cd5
383 F20101221_AABKYN hsu_y_Page_092.txt
75609c55fb93c29e475731c99761e2e3
da42e71606a970779d02ea5ea85fd13d50fbff1d
1764 F20101221_AABKXY hsu_y_Page_058.txt
c0bc14147af86197b68992272ea5a922
d8fb841a8a3cef71019b3d95bf44e9e3addb1439
31491 F20101221_AABKAR hsu_y_Page_124.QC.jpg
9945575580ad75c3e87ca2a25663e18a
867eaebf2952ac1acb0533514c1bc8ac2f83122d
1591 F20101221_AABJVL hsu_y_Page_086.txt
862e806a23bcdb6f94014b4930b6bef1
901b8fe481cd388e9b6a024408fe6e2dc18b26b3
504 F20101221_AABJUW hsu_y_Page_036.txt
4896f5e5a67b0df18f300643f76ff6e2
a762a900fc12de4d7e1736c271ab9ab1f932506e
3902 F20101221_AABKBF hsu_y_Page_007thm.jpg
3c28d3930bd7dc584bfe14614c8fa6ac
e7d8800fd2fa3c0542ccc60db0e7eb13d10dd5bf
1635 F20101221_AABLEI hsu_y_Page_111thm.jpg
09eb0532093c1e177ce128e38259d451
a2eb54a74bc6608c5678da4c09c2ca9497931046
8473 F20101221_AABLDU hsu_y_Page_056thm.jpg
67a980ed53218c662516ac9f5540ef12
16e7ce05677d8e84fbe70cf84385d0ae9922e95a
1225 F20101221_AABKZC hsu_y_Page_116.txt
9488db443e61c500dac18c2bed7cda38
0f750aa171b08b38178f5c38f7dc5b41f6ef5b29
384 F20101221_AABKYO hsu_y_Page_095.txt
e95119813b8591119c06917011672f46
09ea710fc4570c7048cc8ebc9969be73be50cbe6
712 F20101221_AABKXZ hsu_y_Page_064.txt
8db15a63704328623c4f0c1d7e396be0
2ec6893ff38ceaa1c987945a197bfb2ff151a42c
1051920 F20101221_AABJWA hsu_y_Page_101.jp2
3ff14d48eb16dc0828fb80a21ba4dfd2
e216f1b11d7cbb2b49e8c1844c33c18a331fcb0c
13057 F20101221_AABKAS hsu_y_Page_115.pro
0e015a0afac9a8dc2cad410cab15002d
85fb7ba84207e7bdbef7c71b3f47023d98a51d21
8619 F20101221_AABJVM hsu_y_Page_151thm.jpg
be40e95de8a97fcc65d9262ba6466373
a7ff4bcd75f82d5f8b3d905046f8fefa15621e0c
10942 F20101221_AABJUX hsu_y_Page_038.pro
50607fd437c35406d03d35d5d7318204
99a5d12f9d01c11dfaf7122a75f9ba7962583324
1051985 F20101221_AABKBG hsu_y_Page_033.jp2
39e2cada69b68a8a153fae57e82f882d
22a47ff101c10fa8f742e58f8bf24a7109cc59f6
9640 F20101221_AABLEJ hsu_y_Page_113.QC.jpg
05907d06bc6105bbb3a024ab1e71d935
e9064e890280a0c2a8924b4d7119990c26ae439e
8038 F20101221_AABLDV hsu_y_Page_066.QC.jpg
9e2c9301ae37835645523ac3ab0f8c8d
4d229e86aa48000de3d517dcbca6279bf1d8be8d
520 F20101221_AABKZD hsu_y_Page_118.txt
2053a8cb9651391c6331c9cedb956d2a
94a565e3eff945848bb176ef9c5cf46c3c8e0ab5
372 F20101221_AABKYP hsu_y_Page_097.txt
302bbb6dbac626b9885a2c104d1a25b2
06636078225fbe72babf9ce7ce0ac935b4670863
10106 F20101221_AABJWB hsu_y_Page_093.pro
71702524ba4c7335abe5aacf8a218fb2
39623ee657087e0a9e039c5fceab75dc4b677cdf
2088 F20101221_AABKAT hsu_y_Page_084.txt
532b1023c5ee8c9936130e6713e2a002
db2cb2aa141ef6033470aa09ab25a16edc4efe98
3145 F20101221_AABJVN hsu_y_Page_042.pro
3545b65cef38fe3a8f1a5ae4caa7dbba
f0abf18e64a61332e96ef48e601bb4fdc6a9f035
9656 F20101221_AABJUY hsu_y_Page_020.QC.jpg
25f17b03c8305563beff1f525eb9d820
569a4df8031d0754a569b3c43697e15c00e9e52e
323644 F20101221_AABKBH hsu_y_Page_118.jp2
5bbfc045f719d8844b748dbb03b6b95a
baae790729cfec4aebbf95e4040d38da2bdff384
8906 F20101221_AABLEK hsu_y_Page_114.QC.jpg
db1c0a74c3c3bee446bf0b0dbc10fd31
5c66f6b90d58a4fdfdbd960aff42eae83f2431a8
2453 F20101221_AABLDW hsu_y_Page_067thm.jpg
2e6b84854fe75073d07581a29f87796e
24fd1345f58cc093567f159dffbfd3b3dd5b5b94
2175 F20101221_AABKZE hsu_y_Page_119.txt
921083332c050981b20cf818653b17d3
5c0ba363889b52d6588563753e1ed846536d104e
478 F20101221_AABKYQ hsu_y_Page_098.txt
3873ddbe460a1f31d72302aa3d09a515
12e7ce111cbd116cb1185de23ce7d6b31f5efbbe
233519 F20101221_AABJWC hsu_y_Page_001.jp2
ed5b92ef7ef3b816d467fe5349fb9df9
f5d54fa99e3a111f0117966f9d2c5c63d58247be
4765 F20101221_AABKAU hsu_y_Page_071thm.jpg
cf2549a8a046c7f3129a1e25b7175a26
e7477501f365513ee608b92fe1ab3ea850a5e6a8
F20101221_AABJVO hsu_y_Page_151.jp2
6234a5a5e078f30d0eb31ec0bbce5ff6
07aebaf2706615639dd81ed6604ddb705dc0a5fc
452540 F20101221_AABJUZ hsu_y_Page_048.jp2
378e74c22c620daedf4a04e3b08946dc
1377af67a4d7b7119029e53c99dc8580b5cf4911
2387 F20101221_AABKBI hsu_y_Page_021.pro
86ff03818b4c7ee31a0fc2c3a4de4973
0e8c1879d92d35217a2a10308290cdf8462933c6
13384 F20101221_AABLEL hsu_y_Page_118.QC.jpg
7cfaa790c5c200a8749f26dcd640a558
9ff3b7cc3c5aff06b48f6d0be3daa825e5b15e0a
7674 F20101221_AABLDX hsu_y_Page_068.QC.jpg
e70ac1d1c3b9773b33cb647b3a556ea4
e2d313acf3b431665db55511094e2f04681b4eaa
1873 F20101221_AABKZF hsu_y_Page_127.txt
255fe81dc985730e143fa376105bd401
c31eca29946a1b78b4439b2f02c1c527ab30d8f2
939 F20101221_AABKYR hsu_y_Page_099.txt
65235b1119d8525d82c21d6ea9d3454d
d30fbd0107e12c2026b0339c959850f5d1fd6462
8647 F20101221_AABJWD hsu_y_Page_033thm.jpg
e036c708e7e79c5b00b45e32f9f8365f
60688f12c48a84261366a0272a2a7cc42c6ba3a2
F20101221_AABKAV hsu_y_Page_123.tif
c0656ab93336b84d5b3a95f19cdfad0d
e6316f19e37eba3b490e486e723cf418aa731a30
287948 F20101221_AABJVP hsu_y_Page_020.jp2
0121d71cf9bc878e0d8cba3171f0531e
4d5e43b44e3c82a6a19b025a23a05eb6ea940fd7
F20101221_AABKBJ hsu_y_Page_095.tif
771c3e10c6e0d07712ccaea43bfe39ec
096cd6cabfc47579697e50bc6985d032462cb48e
27036 F20101221_AABLEM hsu_y_Page_122.QC.jpg
a3c4c161fa17d9d6b8d1a435659ea49c
ab2500194e332aaa25b3112163da71b8d876c9ee
2061 F20101221_AABLDY hsu_y_Page_068thm.jpg
486bdaeadb9f80c5c08449c7b44bc39d
cec26264f377e9faad245ab24eddea36e9df49bc
1985 F20101221_AABKZG hsu_y_Page_128.txt
54da0116f5a71df5c1f5808e92113368
6de33eb511ad35df220a28a3cb0e21ae06f7611a
2019 F20101221_AABKYS hsu_y_Page_101.txt
5e9410e9ec715e793639796a39d0dcf5
2d1df1d333d5ddc3f84a676e243d8d5f69409b2d
F20101221_AABJWE hsu_y_Page_092thm.jpg
76d4d213ed3eaa2142d8d13d90390a4d
6329cbe1eeed3fb2fc8dfaacfe805cbfd2ee5fe1
F20101221_AABKAW hsu_y_Page_068.tif
3aabe7d9b38452365409c6f1619dde4f
8e8f53913a3b19a6b220401c092510d98c458380
28609 F20101221_AABJVQ hsu_y_Page_018.QC.jpg
e5bd84a1f894377c8e843c4b595698fe
033b20986b89472b14616afca4791a5f77e4350f
F20101221_AABKBK hsu_y_Page_058.tif
96d38b39ca55ab0782c673f96d25bb95
0885f4843f7f5cec997d7b508cadd0854e98795b
32473 F20101221_AABLEN hsu_y_Page_128.QC.jpg
8b133534761450e796134e979b4935cb
60573f54d05c0e13ff6db55c3130613366f356fb
15170 F20101221_AABLDZ hsu_y_Page_069.QC.jpg
f9d512c58c1a10ef8d32c13d564d3bd7
47212c219bb451281bde704e71a820955710657f
277 F20101221_AABKZH hsu_y_Page_133.txt
531fda2b2f0bd1f7937235a8d44f8f7a
bf0b626be2c9c2b822508fb05c3cc0d738dd91ef
8933 F20101221_AABJWF hsu_y_Page_029thm.jpg
11417161c34f5281eeb066f55369386b
153f437348489c5a526fe1501358faa4023d8368
10788 F20101221_AABKAX hsu_y_Page_047.pro
35b176ff7777a98db05a17905e70790d
f0c1b5057dde863e2364f37b114066991b325030
4521 F20101221_AABKBL hsu_y_Page_069thm.jpg
edef501c6b4c307d00e3f072ca3b836a
a297af276c9e879c12c1668b4ebe0ffd1fca08e1
6503 F20101221_AABLEO hsu_y_Page_130thm.jpg
4e5981d6594e06c21601c8f2737a88db
068c780af5506d8e07004235e123bfe5ed1a3e0c
885 F20101221_AABKZI hsu_y_Page_134.txt
355d1871e11a30eb6c59d9f593dac3fb
332d22a52892d93f365aa20334fe1ea09481c403
2121 F20101221_AABKYT hsu_y_Page_103.txt
2993311cec63067bc3b0a96e3e7aacd1
ad4e74987bbe16946c890affb41b2a67c24c9460
31001 F20101221_AABJWG hsu_y_Page_142.jpg
cb1d110a4883be9f6ddc0c06bb0f416d
d861f0e94e72bdf5fa132b468e9bed221227f591
F20101221_AABKAY hsu_y_Page_136.tif
470338bd4bee5d47fe778438ee2ef8d6
8eca1e0fac308f176430b2cc0d7e441bd4b6163f
F20101221_AABJVR hsu_y_Page_026.jp2
e5b9216cfbb055b45fb4db18f9ce8dcd
242bedddf30a0a12bd3f8196e1de9e57d9539339
23801 F20101221_AABKCA hsu_y_Page_130.QC.jpg
1fc9cbff9c2a9a59ad22acd502e6af4b
e015c9500a2f8bb8b7109c0f360269bde91cf781
27439 F20101221_AABKBM hsu_y_Page_132.pro
489fdae4e0f7d3e6d24537a1199ba937
f10183d898f1117eabf2cd29371b2f0f576da70e
1888 F20101221_AABLEP hsu_y_Page_133thm.jpg
de08997ac2baab390305b0e99c6a6a60
03d618771b6fc25a0c00404ead629a6097302b72
431 F20101221_AABKZJ hsu_y_Page_136.txt
591ed6a7eec8a12a39f634c9ca035f89
041209b6674056a499d2439d50dc675ece3da640
1797 F20101221_AABKYU hsu_y_Page_104.txt
e15a647133ac9cb154326a577d8eaf07
dfcf6a4bd5fe1a62e4e5f9dc4223922ecc940815
305 F20101221_AABJWH hsu_y_Page_138.txt
7503d5f0b46dc9a193d297b1a2a6914c
1da05015f44b5aabb204250950622f7d5f665a30
8759 F20101221_AABKAZ hsu_y_Page_120thm.jpg
aeaf182c31657a98c2b33996026c0c43
9fbeddf9173acfcc592b237d198e6fe864c31460
1880 F20101221_AABJVS hsu_y_Page_031.txt
23d7694cfb5cc85215c9223906644b37
b925edbf2536c59da71c2799c8b5ea265a5cb226
47785 F20101221_AABKCB hsu_y_Page_143.pro
05718e45647a3509d5ff8e99fdeb4820
6e847e529832702d039b9ab3f54bb6e78501ea30
10198 F20101221_AABKBN hsu_y_Page_139.QC.jpg
9f262196003577805457850b05f5c93c
0d339be979a571426c4c6805e102756cb34f7991
17014 F20101221_AABLEQ hsu_y_Page_140.QC.jpg
b5e6689602f876831ffef7e36ddc4874
e2f7dd903c50e84dbbf33022e8709c482a36b8ad
426 F20101221_AABKZK hsu_y_Page_137.txt
b0f1b768d6522c62606f37c692300b49
5506275c4a8d7cb74ba774a9dccf40ea9afa9c0d
1832 F20101221_AABKYV hsu_y_Page_107.txt
1c754fe58fb02d56cc4a8373eeb90f82
2f4d06b3e62d72c2f99c6288acfd20cc582a6482
19939 F20101221_AABJWI hsu_y_Page_116.QC.jpg
2266392bb3c9b2eca0abe15812afb12e
32471e43726fdc728b98fe661404030e973d0780
F20101221_AABJVT hsu_y_Page_135.tif
a071dd89d2eff4ed7d2abcc1298c18e1
4a5fb36ae655b7032b07229b6c4307390c3b107d
33242 F20101221_AABKCC hsu_y_Page_101.QC.jpg
ffd2959016e2f745770cb0e774d039dd
8a6bed39cc5b2011f776f206cf24e47293878a78
F20101221_AABKBO hsu_y_Page_091.tif
634774094207455d0be5c30c03cc3e61
cacf846a5318ba2bf3de00ccddf932a585753bd7
5891 F20101221_AABLER hsu_y_Page_140thm.jpg
ef748dbdf27c25cd31513b4ef25f35b9
1ee0b33f4e693db7e1aa784ef1b74788fe0e8f03
451 F20101221_AABKZL hsu_y_Page_141.txt
66bc772f52b82e7c7b2625e29c8be3df
27c42e821d4a1739b60e628166b539410af458b2
2085 F20101221_AABKYW hsu_y_Page_108.txt
5257896ff8bc8a5e87e4f977819f2712
48604c05b0cd94d800b7de99a4c2b7f6508b256f
F20101221_AABJWJ hsu_y_Page_030.jp2
af8e73ed787d91f1bfc0fa77e13ceedd
549617f7d2f3afd5f859e785e8201ed44b4bb324
16474 F20101221_AABJVU hsu_y_Page_066.pro
5357254d4cbc02e10f0d6f5e5fd6d2f1
60f95bd0c9a77ff37de71cf0bc86dee7d44d6dd9
47152 F20101221_AABKCD hsu_y_Page_054.pro
bcd491ec24dd8ee55dde32017fe82f3b
4c6ce6994cfb7211946e68b5a0ffc01c3d6ba78f
787771 F20101221_AABKBP hsu_y_Page_076.jp2
b71f2642c2398767448ae7ebdebdbbee
d49fd62a6f6170d1749350496acf27e69531706d
10721 F20101221_AABLES hsu_y_Page_142.QC.jpg
c502bc12f6fb5c1231f18cd595a3ef79
14037fb336d61b3068dbcc3a4e0a778b8d8e5696
246 F20101221_AABKZM hsu_y_Page_142.txt
6b4df241c2c06ef7aa5c358025c71adf
0df3e86a632ab7b0e36465feeff05efb30cdec77
1682 F20101221_AABKYX hsu_y_Page_110.txt
f6c716e18b29caaae3ffecdb166c8f58
7e7a42e6b4c3b8d40940793e9f0bec2b000182fc
45967 F20101221_AABJWK hsu_y_Page_129.pro
c764ea9f8cca8b8c0af190e7a7ab27e2
60b50b2036c444068c31e8ebf092208b347cc57d
F20101221_AABJVV hsu_y_Page_015.tif
e6c612ad9579785116205b7c63df3420
8b884bdffc23824a74b3e416ac6b6294c8c2a597
11506 F20101221_AABKCE hsu_y_Page_135.QC.jpg
c33547c6e813a308cd80097a5309b556
733a8851209ac7f2eaba8f5ad6635e66d17dd3a8
23999 F20101221_AABKBQ hsu_y_Page_001.jpg
f7c997ce276111c7191b6d5bc1e5e578
10a82e49fc11439c5c508f3c448f862f73f1ba8b
36493 F20101221_AABLET hsu_y_Page_144.QC.jpg
7fe9f5c5aaa4bc430dab6d154d23065a
3cbeaf21a2be24fba325dc8f7588a55d0deb4de8
2608 F20101221_AABKZN hsu_y_Page_147.txt
6e8dce6cb39021670b36962b4df5203a
5c64893abaa3504d7469cfba26e886ec38850365
282 F20101221_AABKYY hsu_y_Page_111.txt
ca6b363d4bac798532c75abdb7c26bed
efda268c70ec1675d18360f454334d30461175ec
1011762 F20101221_AABJWL hsu_y_Page_082.jp2
e4619d008232cb5819b4eb10dd34116d
0ca77fd20afb3eb96c83953ceff3aef83a1517db
53808 F20101221_AABJVW hsu_y_Page_015.pro
cfdc0e65ccac7c587e3de701eec4be0e
76c4b3b828acab467793b578794a8847969914eb
2071 F20101221_AABKCF hsu_y_Page_106.txt
298b4642617e1659cc3f3331a96d80f0
dbe0c9133c29f4da49548365c2d4a04bb3be5cdf
F20101221_AABKBR hsu_y_Page_073.tif
9bdddbaecfd4017c16f3a39afc8a076e
6e477bac85013641bcf8c3e2321f6c3d9f1fe747
8557 F20101221_AABLEU hsu_y_Page_144thm.jpg
0397ef01c3aef5b9dbb298001364a303
28d84a1694547163b532ac852b9f6e5fde59b115
2526 F20101221_AABKZO hsu_y_Page_149.txt
8ba505fb7a21d7f70f3da225df906c56
1a43ee47c549f0e11cb07ebe6b5c31773da5e653
579 F20101221_AABKYZ hsu_y_Page_112.txt
72ab997d7b6f8761241f9555293e621a
e5a1e0ffc94b8838ef79a0393de1d8156bfdfcab
32519 F20101221_AABJWM hsu_y_Page_080.QC.jpg
6d5251b6db58dfb69e1221378e06dd44
8d476d05a4efe4070ccc6659e5181031475c2023
129218 F20101221_AABJVX hsu_y_Page_006.jpg
9b54c7f6283fe8e0862602ecc7a8dbc3
d8b124fee29cc056b539bbce4c25e564b7829319
55185 F20101221_AABKCG hsu_y_Page_120.pro
767facd7e8cc486fa72a96ee5688a8b6
3e0bce184f0fd3a15056f3dc4e3c3b7193e09d85
35572 F20101221_AABJXA hsu_y_Page_081.QC.jpg
41d4d9dc87f876d4d63e06eeae42d0b7
0aa5c769ee6ae0f0e404a9ab6adeb1d20e042608
929913 F20101221_AABKBS hsu_y_Page_110.jp2
ab8da1c4964d702bbdad944f4e3b5c83
71ed10774326cdcb22e8fe01b549f0374de527b8
34821 F20101221_AABLEV hsu_y_Page_146.QC.jpg
46a953d41b1264cc1d688bcbf11dff3e
3d686b41380c3165d29c961972ab0eb098fecb6e
2333 F20101221_AABKZP hsu_y_Page_152.txt
81d5ef7474a46ebffd8157191849749a
1159a41400d4c12a2ed4a1afe464d553db7057f3
F20101221_AABJWN hsu_y_Page_069.tif
032ef795b6125cd04ba7cec87c9b95a4
8e3d709d510dbb22bcf52ff2bd736b4cb51523e2
8034 F20101221_AABJVY hsu_y_Page_145thm.jpg
678be06c1ed52875ebbc8eafe6ea84ee
3943576948bee1ad08cc38883de1282dcd245244
2091 F20101221_AABKCH hsu_y_Page_074.txt
58b2718bcceabb8badeeea53679439a7
dce8d3b6575a9e50c4ba695b8e4ce1764c31ed73
2178 F20101221_AABJXB hsu_y_Page_062.txt
1e99fcf3b6e605d8a5bb78c22f52657d
67d09d1bb1fa778b4eaff13798c39e7dc00aa9c6
29871 F20101221_AABKBT hsu_y_Page_113.jpg
4d4a9358417c5d7c2a07325c372f52e3
bff8fec58a5fd7af56f59b5db304387c7a3dfb66
35411 F20101221_AABLEW hsu_y_Page_149.QC.jpg
6ded4e4a7fba1d41c24254813bd1ecda
7deeb298d906934bdc629bda0ad1ce38477a1a2b
2635 F20101221_AABKZQ hsu_y_Page_153.txt
6e68c542c62778c7c30f942efc293775
9d6d9cfd15de3907b2c3cddba8a7ea0ea9773139
405465 F20101221_AABJWO hsu_y_Page_071.jp2
3e8c8b70484faaf16301d78393f90189
0314d8b827c6725d7545c0303665b1df65b29525
103791 F20101221_AABJVZ hsu_y_Page_101.jpg
29306961060887c114c055be58e4507e
2fa07adb761c03a66a91164feb709788347eb24f
4767 F20101221_AABKCI hsu_y_Page_114.pro
c6c8ba5cd35d1e7a54b41d97f9e7272c
6a277042fe910fa43b5e0cc5de139a60aeea8b6a
2044 F20101221_AABJXC hsu_y_Page_025.txt
9ad614ea89afa4b411c8bf67ce28ea31
1c73af917e8b9caa50635693f524507eef922d87
F20101221_AABKBU hsu_y_Page_127thm.jpg
096955558d068394c4ec4724329186e8
e922132b59bf331d9efd193e01aefa2cbf3f234d
34863 F20101221_AABLEX hsu_y_Page_151.QC.jpg
d1a5ef6cbd83189df0dea9fbb1959e25
ff43463e5077df4c3352bf0b7efeeb3196f5496f
F20101221_AABKZR hsu_y_Page_154.txt
3270e53bd0be36b4f633007afb392406
f1d11025fcf15762bd093373f9f96650fa426f48
7888 F20101221_AABJWP hsu_y_Page_131thm.jpg
5392b2b3ad24c800e8f3ae91d0a3a5b7
2030044df5fc954fd14b7f8958155e99e6483f13
3631 F20101221_AABKCJ hsu_y_Page_002.jpg
06f2b497d3c6438c1e76156d5e45a88c
6900a60aba80f07ac74436236be3782d4a825c6c
2151 F20101221_AABJXD hsu_y_Page_051.txt
a3ec308dbd23f7bb7124a2df00e4b7fa
91cd4fea74587c7a86349870e3f5cfaab458f8c4
76517 F20101221_AABKBV hsu_y_Page_154.jpg
63650a3f3fcd760067be54cd531c688f
6745efec1864c6bcb8bacfd1401fa7c84d671f7d
32559 F20101221_AABLEY hsu_y_Page_152.QC.jpg
bf4f7fe0ace52f5bd7daec043ee2c31e
4655f37da50b3140e5d6b53b37bbef931a3931e0
1327 F20101221_AABKZS hsu_y_Page_155.txt
18aacdc3db8b7ee17f6d2d1fcea937eb
8599e142de3869ffd04c9b791c992514ac681cf7
111192 F20101221_AABJWQ hsu_y_Page_119.jpg
cb0862c3da02c39c24957edeb64e4387
9d276fc8813ab00818a79703acb4243a536b7127
4251 F20101221_AABKCK hsu_y_Page_117thm.jpg
d4762ae42ef506d83cdad2f94fccb029
e8a98ba2e3ae35543ee56f9222b197faba88a00d
2376 F20101221_AABJXE hsu_y_Page_066thm.jpg
c1c9d876a7b6a7059601cddfd18fce2b
b920e1f93ecf599076b4ba0897ed5a00d54a11d5
40073 F20101221_AABKBW hsu_y_Page_046.jpg
ab17d2f039b2758cacfb99575467622e
b5202ce63d195f6f793d9d4dcd5912a776bb3dcf
8401 F20101221_AABLEZ hsu_y_Page_152thm.jpg
8e2aeeffb453c54630a81aebe7c0d793
61a15b744976dce47570be8bcdf49114578f56f2
2001 F20101221_AABKZT hsu_y_Page_001thm.jpg
7833217d4e9c3e615b27fbf40d38ebbd
06473f0ae446651a92dabd5ea101aa8e2a73664b
74749 F20101221_AABJWR hsu_y_Page_076.jpg
ae26d5c602f2fcee1e9be19b08176e24
8e174817253956c5d3dd9ed14a43a9f3302fa196
F20101221_AABKCL hsu_y_Page_102.tif
dfeaed5e51dab656264628dab6ba9e9c
19992c748e5db811f4256069927762e782833b98
2108 F20101221_AABJXF hsu_y_Page_060.txt
b10790a6e9f977e146ecceda1fe5d97c
a1a2680d29850a123f24db040e709111e7b7f90c
6137 F20101221_AABKBX hsu_y_Page_111.QC.jpg
516263b5193754d7bffe8f33ad26e9c6
02d7611a6b7d503b781628e4f8440b35a33ac147
54913 F20101221_AABKDA hsu_y_Page_051.pro
eddec5141f702bec73a9512816f805c8
29e56b03c994e98cbd0addb5b1b8a15f24a4457c
F20101221_AABKCM hsu_y_Page_023.jp2
70598dd4329b12b55e1d643273cc7b94
3e95d87b36232c7aa95cd538f70b213dc6ed4d51
27369 F20101221_AABJXG hsu_y_Page_066.jpg
b34324d9c439a48715ebbb2b8bc4afb2
d0ec0d9946b82345ac9a91057d23f79ec5edd834
F20101221_AABKBY hsu_y_Page_055.tif
0c1c14c1dfc2707d0a7e0cb7d53b5b3e
a676d458bd954b6ffe87da2ce28ab0d81f9d174e
861366 F20101221_AABKZU hsu_y.pdf
538461e62111d901cb63d33efb2c75f6
e449e7348edc74005d63dc6f52fd4d9bb59c9827
639 F20101221_AABJWS hsu_y_Page_094.txt
c07baf031e8e4aa92849609b2a6fe7c7
7c276224ed1e3949c887cd88fce83b5ee084ec30
8693 F20101221_AABKDB hsu_y_Page_025thm.jpg
b11563744523b885ee7e48d7c8dcf118
13e46e5bb9718729f4119419a336517651aff2bd
32354 F20101221_AABKCN hsu_y_Page_155.pro
93c2ffeacffb971dee33ca6651529927
71cdf0a0126d7e3791ba2860913adb4f627a84ee
8302 F20101221_AABJXH hsu_y_Page_024thm.jpg
18170054b5f0fe497d0cd617575ca634
49f0835827884deef494518ceaf0dcc94aa53520
28078 F20101221_AABKBZ hsu_y_Page_006.QC.jpg
24d862486f870aad63702396cc5f12bc
0317ffb956e5c1e1d20e32be1cf7ceb262ee7a23
3350 F20101221_AABKZV hsu_y_Page_135thm.jpg
076ef814337cc8630c44336de5e6e2bb
94ab418e60f052c9006c197678ddefa43e9ff759
3923 F20101221_AABKDC hsu_y_Page_006.txt
37e0222c6bf5589d4f994ae1f43f6e08
e734d620e9fe15dd0b8ad80d77406faadbff6bdf
1017619 F20101221_AABKCO hsu_y_Page_102.jp2
982d50cd4ea6017c8c2ce3c1219255b1
94767689af0c9255b2ae0209a544041e40d0e8cf
8540 F20101221_AABJXI hsu_y_Page_004thm.jpg
c6038a63e01980676642a05302eec22a
3219ff57f1114ebe636817d93e47e05fc205b145
773 F20101221_AABJWT hsu_y_Page_069.txt
64135c6912badf122f4041d0c3004472
ec2862ec55f543cfd3a4f997e4fbfefc489cba0a
7258 F20101221_AABKZW hsu_y_Page_018thm.jpg
40bb274b616eb1f3e7c953e0d5613ef5
b5da4522244e55a1836316df6d7914b47c441582
2543 F20101221_AABKDD hsu_y_Page_003.pro
d2e652e2c9b91149228dc8d6fedf8d1f
369055d266dc1ebdfa9b10c6bd96355cb3e6ab8f
195029 F20101221_AABKCP hsu_y_Page_091.jp2
a71491a242d9ac1ba00e034da3384c42
c6d61ea7aaeb0ed83be54a2a409c9fccb3dd02ea
110301 F20101221_AABJXJ hsu_y_Page_051.jpg
d4ba8a0af5b12d0fafa985abd47b03a7
b1e2438646d626be39e55b714e903f237a93b7a5
9697 F20101221_AABJWU hsu_y_Page_035.pro
e60f16e4b8cff3c6eab9230123d2545f
c7e55a667ca265e0bb885ff4bdc346df0a540ae1
3532 F20101221_AABKZX hsu_y_Page_090thm.jpg
10df2f63ad01e6e2de094801e6f4337a
b583d3deca72c17895e6c0cad6e32bf9f66b0793
F20101221_AABKDE hsu_y_Page_011.tif
aef065264a87d95c000914ae075be762
c9bbfc1a9cb802b181dd5f9887078cd3750f23ef
103806 F20101221_AABKCQ hsu_y_Page_050.jpg
90e7afbaeaafe2382e15d40d7b15f68a
4fb385465fdd993e0659fcdb5649f02fe81d39ff
F20101221_AABJXK hsu_y_Page_118.tif
b0da07618378b727d37be20db7f56570
c967087b396b34b95abc1fc972803190efc727be
104490 F20101221_AABJWV hsu_y_Page_004.jpg
5033bbd32074854ab502c57a3bb5f4c0
fed5ba60d631b58a7c0957e1b442087a27228228
8456 F20101221_AABKZY hsu_y_Page_027thm.jpg
2a7bc85b4d4f298f4d86ddcbb8a8690f
bc48d4e4f378419581eb1603c4d39e787cac6c85
1979 F20101221_AABKDF hsu_y_Page_030.txt
76e69cd3ed13d602e4d705a38191fecb
1c520e9ab808e77e5fbec73ddc0c2c3c9361c0b1
30502 F20101221_AABKCR hsu_y_Page_109.QC.jpg
77718b434614b797d435a94840530e4d
2d11edf74f1d5c8366f8880617a8d8fe914d5d8e
108375 F20101221_AABJXL hsu_y_Page_105.jpg
5504a4ce11147ced26c66659886c3ff8
99462a914bbd2c73919fe2606f8bd4a9fe615e61
F20101221_AABJWW hsu_y_Page_007.tif
f29809d3233c7f427a2d3e416ae9d882
4d4e06fe46d05aa6a9188768831437d8c37b2fcd
F20101221_AABKZZ hsu_y_Page_060thm.jpg
2be8ac7c66331f1de92fde7a183d349c
f5f106fd4afd31e23b4c0aa711c95ad4ec3f4508
108140 F20101221_AABKDG hsu_y_Page_074.jpg
dc390aca5f96fc8ae66660accb059fd2
63c1ac63373f9df1a1ebe8d832012f78b729acf6
32906 F20101221_AABJYA hsu_y_Page_057.QC.jpg
8472b1fc5f7656ae621e1d161530c2be
4281150d732afd08cf1f9c4e93ddaf29d5fc30dc
8847 F20101221_AABKCS hsu_y_Page_074thm.jpg
e437a790a6438125f66a1ef6ce251a68
8e548f9059c6739dcb4a4efbe61044a366dda2ae
870 F20101221_AABJXM hsu_y_Page_065.txt
af1b779640b070401919f78797e7763d
df1ee50fefaed75f35536a84a1a4f2e1dff58c5f
1939 F20101221_AABJWX hsu_y_Page_121.txt
73ffe170fb147871a09df997e5589723
bfbb4e6232b8101ef9199976267f4aaec10935ae
48140 F20101221_AABKDH hsu_y_Page_140.jpg
d4bbe37168cc9aae4ae1d3fcc414b6a7
4536ef795f0f434db24d97b3fda44a08e17d9981
110604 F20101221_AABJYB hsu_y_Page_103.jpg
2bc09ffc010768fda8eb6829bc7b1532
ee1568cd41cfee54eff130842b1e2cb88ea3809b
6231 F20101221_AABKCT hsu_y_Page_005thm.jpg
7634c41237a4ac44c3fd8b44ce852309
34e2205d8cd437311435041e9391fcac4fddb00e
F20101221_AABJXN hsu_y_Page_032.tif
5dc355ba810752e3dcc7ecacb1a53850
c75148d5ea43e27d3fbc8ebb889359c4da04aea7
7882 F20101221_AABJWY hsu_y_Page_031thm.jpg
800baaf2193c7f99c043e9e5e2f51ffd
15361a79a1ba4e7ef915809b4715094b34ace7a6
F20101221_AABKDI hsu_y_Page_112.tif
ce11f84bb5a780d6f7c0751fbaee83b8
f2041cc70613ff295992cb9c02db0a0ce2a88f23
2004 F20101221_AABJYC hsu_y_Page_038thm.jpg
0c257dda8ebde80d73f1e728cb20a4fb
6098956aa7cf1235a8b0d836bedc4bdfaa448e47
F20101221_AABKCU hsu_y_Page_110.tif
0f387167acf878cf6aa950f2bba89902
038a4b117d86966d10bfec2793d685c08ec1b55c
F20101221_AABJXO hsu_y_Page_061.tif
a1f998ca910e402577b9c8dd6d66b9e7
38f3309743de8911845f592035c356fa507270c0
109164 F20101221_AABJWZ hsu_y_Page_032.jpg
9ebed7d23dcd8107565853ad4ef0270a
00e523ab79c312248989789003592f7f56889c98
39834 F20101221_AABKDJ hsu_y_Page_117.jpg
c63d5562b33d7bb8998968482f2d8677
bff2d7cf0b045b78a80bad9ec0e79615369cc053
35296 F20101221_AABJYD hsu_y_Page_078.pro
9cd55458c1feae2540a740db91d7559d
b4809a3c0e6c0daed3844fd7b61369656e8116b0
2364 F20101221_AABKCV hsu_y_Page_146.txt
10b933a9a9bc9f4518674f528147a12c
01b37980d37886d0ef65c043945130ae11359589
1866 F20101221_AABJXP hsu_y_Page_129.txt
ec5b5d0f0c55e5a86ea51386586d3476
ddd9a22b550bead2436af290daf0633ba0f3fda8
F20101221_AABKDK hsu_y_Page_098.tif
59fca12dbc55e20cf9875b79250983eb
94e222bbdda73bdc6f65ced4bbd1a4b48a8113e9
110572 F20101221_AABJYE hsu_y_Page_120.jpg
3d5f8808b1eae51b51496a0eeab2315b
30aabee88b0cacc0c99570fcbefdee2a4b0bdb8a
8106 F20101221_AABKCW hsu_y_Page_137.pro
fe85c4798ffdb67afdab6c1007c11782
0ce323e652b983babcf19621dfd099e858f91b7d
44599 F20101221_AABJXQ hsu_y_Page_098.jpg
bbee0ca4fc714e6daf25e64c6951b024
88d7097277d1fc95b3263c33dd40266d5e5a0487
30722 F20101221_AABKDL hsu_y_Page_129.QC.jpg
d75183243d5db0580ae874bebcbe3c88
067d0741c4e2c6c3d6434db8da184ab5dfea5b51
65104 F20101221_AABJYF hsu_y_Page_153.pro
b97ffa42743a0bf21f1d5a1901054d19
e3dd5e113eabb19334e031cafb8575c9fda0e384
79069 F20101221_AABKCX hsu_y_Page_125.jpg
08d202673dc00775ab813c18afff997d
687d00c95d0c71f27d81477e1ddfe784d6af6eb8
93128 F20101221_AABJXR hsu_y_Page_102.jpg
56b29205cb31e9f3ad7615b80ad484cd
f7c4fdd943a3ef552fea276d66ad2c912e9da7a0
52608 F20101221_AABKDM hsu_y_Page_119.pro
66bcecdeb24e66990950580d6a7cedc1
884e7386ed4a81348414c28a1ae68c081d341630
64991 F20101221_AABJYG hsu_y_Page_147.pro
227acc825fd8a223cf68d3ce99260bd0
450f4562653e3b31d730889148fa2b733c1894c7
8823 F20101221_AABKCY hsu_y_Page_083thm.jpg
cc38c2c58b31ac5ee4fc64084de6c5df
f1d9bdd6c3fdfb7c978d31472c683b9e7977b6ec
13660 F20101221_AABJXS hsu_y_Page_090.QC.jpg
aa6f4e38769dad6af0a7352b994108fa
9706cb42cdc27c2432e20ce0a33315be8cdac35b
738 F20101221_AABKEA hsu_y_Page_117.txt
c332066f7583faf76d20e18af3bb02c6
1de052e0d602cf9d59119c4048cb28aa088f9af6
1051945 F20101221_AABKDN hsu_y_Page_024.jp2
fb23693c2106ca6bd2b1d87f5dfca95a
8a3b9f1268f0a313b53321fa2cee6247e3acff4b
28221 F20101221_AABJYH hsu_y_Page_086.QC.jpg
63c6c271773e1ce4f4c0efc4fb086418
06409d165fd979e910f75b8ef10020952c4deeb4
2506 F20101221_AABKCZ hsu_y_Page_151.txt
53c725a9160d5063f6e3581ba04fe813
3b60f5839f1eb81ff24eaf76f79ee27a19836f4a
8199 F20101221_AABKEB hsu_y_Page_022thm.jpg
cda46ef1cebd10df63c08bdc9149ed6a
8e007aa7750f5566a7c0292a22303e635ee897b4
102808 F20101221_AABKDO hsu_y_Page_080.jpg
b4952bc377a51e1dbb7ee989176a07b7
bcd0983cb85850cc947acacc2ef9f024323196f3
F20101221_AABJYI hsu_y_Page_049.jp2
e7348bcddf0c4fb24b0536584ed2c038
85231c412b4fa116a45e31fac28fbfba4ad5277d
F20101221_AABJXT hsu_y_Page_003.tif
0cc6567b0899b9d40ccdcb879d786a53
731032141900ba0d08076c7406bed13bc829ab5f
94915 F20101221_AABKEC hsu_y_Page_127.jpg
21cf4087ae47e792c139709672f76455
3c38be2ffb8fb846530670469fa6ed83ffd31f87
F20101221_AABKDP hsu_y_Page_060.jp2
721264cbc75564a65269f594178a41e2
8ac689f5c572df8e60ad315e1bcf53ca033e101f
1051941 F20101221_AABJYJ hsu_y_Page_144.jp2
db6a08e0af2a083844cf293469448f80
b4c7cad3fc166bcd7a78394177f9e631123341f6
21433 F20101221_AABJXU hsu_y_Page_065.pro
c9ca93106b6da02eb02b464ccbe1c659
3ea051014a8a30657a8094a6d143c610ce6ef1e6
4900 F20101221_AABKED hsu_y_Page_089thm.jpg
6dd4a6275b2e3ea6dd291372a2f1bf13
cfe076d3cecb1b752e9b52567e121c49b0fcdecc
473399 F20101221_AABKDQ hsu_y_Page_134.jp2
1332dd601f334dde0df72e9bfd14026b
87cdaceef49beee28e743c4016c7f425bd0d171f
F20101221_AABJYK hsu_y_Page_049.tif
296096de991d95f3ae1cb9fbb6209d83
d9f38ced2a34832e03287ca62896942fb6b4f416
F20101221_AABJXV hsu_y_Page_078.tif
d50e55f0edb6f2ecbe7160f2119ad5dc
d2679ed9cdf59724ff53772012e2e56295b59b2d
37330 F20101221_AABKEE hsu_y_Page_100.QC.jpg
8249faf39a9d975bb3344bc0b1703f93
ef86dfe0b1346d67907a3272c9dc02e724230012
940893 F20101221_AABJYL hsu_y_Page_123.jp2
112733c29d69bbff8663a86eb6e5a931
f22bb497e8e7b1075a7b5089907050a4fc071551
2478 F20101221_AABJXW hsu_y_Page_148.txt
4a8039333c3e72331aa1cdcdb3cdf8a0
982ab939284cd8ed63698bb0d68237944a24da60
F20101221_AABKEF hsu_y_Page_148.tif
81087de74aa0e4e1bcdcd620085888fb
a7b1980696e1ef6270f07bdf0f6aa0c775b258a6
8174 F20101221_AABKDR hsu_y_Page_009thm.jpg
c6679df4800e56a7849f738424345aee
9558e2318c0959581571a7fa281524fb3b1b60b9
7512 F20101221_AABJYM hsu_y_Page_063.pro
8d1d9ae38b1622ed45c804b147403eb4
37eeb6debc3d5858baf62488cbf5971cb9c57b9e
36086 F20101221_AABJXX hsu_y_Page_103.QC.jpg
56bdebf3045dcaca00a26bd1ad93d7fe
5e1a5053a4ada616b55a4cb8f309b6075dc17272
F20101221_AABKEG hsu_y_Page_017.tif
4746a5bb7a003b004661719c0bfba23a
063fe3062f6b6a448d713578d613b204a5a2d411
9753 F20101221_AABJZA hsu_y_Page_136.pro
5db16a5780097fdd9a85a43cda79108b
58fbd0f077417c153efcf8542bc558c366f67bc5
F20101221_AABKDS hsu_y_Page_129.tif
9d70e835f253db7238a898ff72f4d7c9
741558e9b11f8f783ec35b2e3370281fbbe7714c
393787 F20101221_AABJYN hsu_y_Page_115.jp2
6b4b69db0960fb4134a3917b288d9f9e
83c8bb5a58a8631bc8c5c36cbcab118dacc0036a
24764 F20101221_AABJXY hsu_y_Page_076.QC.jpg
4ffff9f6b4d798aeac69d3fa322c843f
5620edd7c7b1c64c25a77948a05c79cf1635f0f4
844879 F20101221_AABKEH hsu_y_Page_125.jp2
d12acafd1557fb5d6d57be79e66bf41e
59166f9ec5357de307b35b4ce940fa4a56ad32d7
32877 F20101221_AABJZB hsu_y_Page_024.QC.jpg
039b6ebcd0db58203ebdb77ed5ba71b2
328cd5fcaf0d84c4981d066b4c5aa48df945115d
12794 F20101221_AABKDT hsu_y_Page_036.pro
4d1d71bc27ffcc6df6fd3fb6a9e1c9b8
665d3a9de214152f1ab9a2dc4df278d649e1b5d9
164616 F20101221_AABJYO hsu_y_Page_096.jp2
b2111697d0009f903344444110fe4832
f25afab4cb973e7641290032fcf7f5e70e63d48a
105716 F20101221_AABJXZ hsu_y_Page_009.jpg
d8acc0eee97877e7385e847304a82ec5
40b2f2a876f723ba68904471b3c5e7c76ffcc2a6
14564 F20101221_AABKEI hsu_y_Page_117.QC.jpg
7def157d29fcb62ad14f628bd4569930
ffd0debd0bdab5dd01ca1755ae0992d0c37fb12e
48043 F20101221_AABJZC hsu_y_Page_121.pro
b98298e576fd35567acb5e4dfc9574d6
44bcf3cda18d144d3ed862683e6cc7e381e45de8
109387 F20101221_AABKDU hsu_y_Page_019.jpg
09dd76fbaf27c53894edb59342da3d18
b0fe5c84aa85ed3841ab3569949d5af4672234ba
163461 F20101221_AABJYP hsu_y_Page_111.jp2
1caa9ac1ec75bae49d6affe714ec153e
b92300432f22e29f45f923808d2f513d9d63092e
45452 F20101221_AABKEJ hsu_y_Page_082.pro
6abeb67d47b97e7eebaad45152fc4162
5f729304c0b6f8c85ae6d2e2c3787aa3e3f3090b
2066 F20101221_AABJZD hsu_y_Page_013.txt
c6027fff030ff1a71499273d33ce3cd4
412f8cf23dba5176cad56eb0d4c914a91a277b0d
F20101221_AABKDV hsu_y_Page_079.tif
ea7e1d7dfe0c062eee463733414df428
8e986234729b85fed05b5d788bdf18cad7652a3a
300801 F20101221_AABJYQ hsu_y_Page_093.jp2
1dd31a86c6f3e91a91af414ed4a4b756
594a0347aa0cf065b0da81e6fbab89d2074552cd
8587 F20101221_AABKEK hsu_y_Page_057thm.jpg
d7ccbd2850b981945d55108ab2a12117
fbf099436ca9bc257ad7fe8de28f9fe773e03620
111869 F20101221_AABJZE hsu_y_Page_049.jpg
8deca10ecae3780c8ae5772e393307b6
bb3b62af72a39cde235fc7612a7f912e060286ac
1847 F20101221_AABKDW hsu_y_Page_102.txt
ba131cc430dfe943592aa76d03cb4c04
13ef149b698293aa3502ddcdc8cbe1fddcf700d1
8502 F20101221_AABJYR hsu_y_Page_101thm.jpg
9576d2418886c794cbb3c5a8f45b30c5
5a0d795566f7015794422786de5c4761b36a2ce0
F20101221_AABKEL hsu_y_Page_005.tif
f97269f7129bf1ba0ba473774953b8ca
dfc6e8bae3ad2578c40e0b755cbb9be8d10c1f54
16068 F20101221_AABJZF hsu_y_Page_068.pro
f385cec853357b8ad31895456238b772
2f3a0d5d1b1bbf8d39d4ad2e6dc2ceb63c1e2d7c
106702 F20101221_AABKDX hsu_y_Page_087.jpg
edc7a9c6aae5f3cd722afea9c702ea4c
b75577d3e4bc1de5ada29837f5b769e725678930
F20101221_AABJYS hsu_y_Page_062.jp2
3a5b587bbabc1ed8bc25738a4589bb2c
81afad2ab669eebf98824e606b418b0b73c60274
57058 F20101221_AABKFA hsu_y_Page_072.pro
43d066c6fbb5ece0e68a286d679a44b3
ef084398be6d4246e996116a97ce7f64d05ce75a
53014 F20101221_AABKEM hsu_y_Page_033.pro
ddf7c9d2bd79d28a073ea0602e5bb47d
4f623577c009878f9d22858fad00e52e6f8ff4df
7632 F20101221_AABJZG hsu_y_Page_117.pro
dcaaf61c825c2fbf4f8acbdd12bc0692
eac9e6f557f1e2e7a807b2d24e7bbf1b0129d696
F20101221_AABKDY hsu_y_Page_029.tif
539c45defd0dbe35f084ff286b8ff629
3f0a6d646febc839780f12ef0dd3413652829be5
99237 F20101221_AABJYT hsu_y_Page_126.jpg
dc765becfddfbb3d6c8a474c2c1ce3ab
18a08425b73fede2bb56b935e89a6942b398ad6c
2058 F20101221_AABKFB hsu_y_Page_105.txt
8e7ce5122295753a2c5a307ee98c294b
c17f9e69bcf9f3df9bbca8c3f2b7f8e3faf78b86
9241 F20101221_AABKEN hsu_y_Page_134.QC.jpg
14d2f00dbe4a4732f3c0768f79f833a0
4e3dc4ad3d2db5a44ecff916e8fcb936a233f0ab
F20101221_AABJZH hsu_y_Page_053.tif
fd92473c0bc33f2c6ee478cb614de846
3dfefbfe24cae64985cb75e2484000cf7d57e820
F20101221_AABKDZ hsu_y_Page_016.tif
168a90097be2c9d849af2864c919b8be
be42627f6c64b6fcf914a96c21367d6d105a987a
34523 F20101221_AABKFC hsu_y_Page_060.QC.jpg
3735c7698ade69ac3b8871bf67096758
902dbe3d91e2a735e8924a369dd9a61202343c7c
173866 F20101221_AABKEO hsu_y_Page_063.jp2
72049d773b0edb601f5ca1919d0ecf2c
048b870c22b79cd58adedc0ea20ca329ecdbd897
34508 F20101221_AABJZI hsu_y_Page_025.QC.jpg
650f8534608fca10026bab6cb6b8983a
8f9356be6a070d3489204995573f8ade24641afd
1913 F20101221_AABJYU hsu_y_Page_109.txt
9f47204adeb7b65533dc5afe292f4b59
1274baa759b8c86d108b42f88dc7c964d5074a9b
15313 F20101221_AABKFD hsu_y_Page_137.QC.jpg
74503cb05df33bae1b69deffbb9f6aaa
36ba7c52f1beecd22b2731368d1d7c50de9b3e9c
F20101221_AABKEP hsu_y_Page_005.jp2
cc8730c961f8893ee03dbd16aee1636b
e4c846979fbff0cc80dadcae5b561c09581e13bf
1967 F20101221_AABJZJ hsu_y_Page_145.txt
bcd11e4deee0a09d5e8747d8ec1db2cb
1bad080e7b91ce26ea26a731c26041fe779ddc34
8363 F20101221_AABJYV hsu_y_Page_084thm.jpg
a59591729310e6b41de51559d80474e6
a43eb8230a1c401d9c7aaa5180ecbccc13f39967
529 F20101221_AABKFE hsu_y_Page_140.txt
6e6c7511ab6195a07e687dd71312e59c
97ea38ad778c449bd1f22d6e6be1247058cb6098
53705 F20101221_AABKEQ hsu_y_Page_014.pro
f5f71712beece1c923a85729e5dce542
64c77f9482142e50216e072746b721dd495d764a
2521 F20101221_AABJZK hsu_y_Page_134thm.jpg
06bfcb9a0d4680e2778343702e038aff
a7e2727bb109821b73a296a0a7215cbcae7739e2
54822 F20101221_AABJYW hsu_y_Page_049.pro
0aa7969eb380d081ac414a69134a8aa1
b420dfc297a40cc3226cce5102f2093427373d7a
F20101221_AABKFF hsu_y_Page_108.tif
694bc524e23aa3231506f0c4eb449b60
cc1395eb0064055b5505c9ce902df4bb63990121
50199 F20101221_AABKER hsu_y_Page_080.pro
8aa6b7709e8ec26c6951ba8b2aa74e26
b86ced961c634ac24501d0ef257ae7003aa8f1f7
6437 F20101221_AABJZL hsu_y_Page_133.pro
0234ee383226a5baf3676573ca10dad9
33fedb1f2f52a79fa51afade5786ea723f222624
F20101221_AABJYX hsu_y_Page_130.tif
f49af9dbb858ab798d8757a93e6b332e
38fbc0e41e50790c1504381c0d8f7fd590069168
15203 F20101221_AABKFG hsu_y_Page_070.pro
c14852d0734b0f86cc394e962d1e6ec4
13c69c1107c4daaaefc5621c3d6cace7a85c7abc
29093 F20101221_AABKES hsu_y_Page_011.QC.jpg
dc62f048fe0a122adb1382d4b34ca05c
c401695c7163e0653d1cfba9bb1c814b0a704b8a
58986 F20101221_AABJZM hsu_y_Page_146.pro
987eeb899b9606aaa33e7ea880bbc2c8
a79557c0d67006537aa7c7876443f433547a2482
9064 F20101221_AABJYY hsu_y_Page_100thm.jpg
3856426e02d4d69483d946c892098454
5dcba6942d00b85f2144afa8dda4bcaacdcc4df1
11486 F20101221_AABKFH hsu_y_Page_010.QC.jpg
ecb50e5808b9ae9a269855845f100d4d
3e08298c9c942ad55731c5da7892909b287aa429
132 F20101221_AABKET hsu_y_Page_021.txt
e7aa6a691dff85f44c99fca79c7eee39
00dcd4f06ac8f5928fab06efb736f0b592c694e9
275648 F20101221_AABJZN hsu_y_Page_113.jp2
d4d9dcf3674599494e45b47a8b3c0aed
c61bf8aa07b7022113b57c1c5fbb27330cc71fcd
15928 F20101221_AABJYZ hsu_y_Page_071.QC.jpg
fc8dae5147548d4cf61aae428c19823c
4c2713e1051112800f7b5f333755780798e69670
34107 F20101221_AABKFI hsu_y_Page_112.jpg
4078bffce7f9c9dff090d5f689dd7bf6
1f672472b8754147e8dce3634995f0363abb74a8
1051946 F20101221_AABKEU hsu_y_Page_121.jp2
7dea9e894c9df6ab3cf5b4017ca4185b
1f71bb03b7bd449088056ac66de8fca10e0b85ff
30021 F20101221_AABJZO hsu_y_Page_089.pro
1651fd7cfc550b98dbaaa163f36f5077
1315868fe9fc6f63ed3b129d16b952e8c194b58f
20109 F20101221_AABKFJ hsu_y_Page_096.jpg
cf677ae8be403cdcea6fc0898f69760e
6b118c5c182775ea6e1837c6d9510b9b891428fe
7867 F20101221_AABKEV hsu_y_Page_107thm.jpg
5e7e14667af410fa428512b6c8b94fa1
7702ed0f4945915df5601a31ee34f2f0c9469adb
7006 F20101221_AABJZP hsu_y_Page_122thm.jpg
8b4aa7cf9c5ab6516950d5f70b85efac
056fef5d3077b3b3744ceef2c460950163364e7c
21067 F20101221_AABKFK hsu_y_Page_134.pro
f601bd915471bf3708b1c323dad295a9
90208ef392fdbb0e746af0ede610f535c4c30037
7180 F20101221_AABKEW hsu_y_Page_104thm.jpg
8f55303fd19998409cae9b8c77656bf7
f13f68e73b889596072e0797c83dd954df9ac026
21278 F20101221_AABJZQ hsu_y_Page_010.pro
6463a03040f1674ce1fd0318ee8238b8
7fdcf6b482c4a5bb4112a764e94eeaaf19be4490
F20101221_AABKFL hsu_y_Page_106.tif
d8849d7fe208b8147cf87ad1118721b5
536f733c646626a243ad7e480d3f51f135666f34
2119 F20101221_AABKEX hsu_y_Page_059.txt
36af4511a8188341a78f7165a2269a5b
484d07a1060f7c903fce28e78f0a88424374314f
6735 F20101221_AABJZR hsu_y_Page_133.QC.jpg
9b200b7f25607558d7625ad472a506f7
29d6cd157511bcc82101116be648df47d2b28a7b
79873 F20101221_AABKGA hsu_y_Page_104.jpg
b267bcbc4636820babf0035bf662f0b1
f10e0f22089d4c31a060d3c30cf95c2fa3b63696
733182 F20101221_AABKFM hsu_y_Page_130.jp2
97b8a2cab82a70c23bfd2c8aa8d13da8
546afc1e48b2cf22036a4b5cd20ca8b90b203b43
6447 F20101221_AABKEY hsu_y_Page_063.QC.jpg
88b1fcec0753a7f82331c34c6ba19cbe
e0ae9248cb411d1ee9b7496c4ef8e6d90b07e089
25923 F20101221_AABJZS hsu_y_Page_139.jpg
0189c29c46de34261d6bf0c2cacae268
e0ced954a74d73b825c5dd3750ea48f2586dd78d
36265 F20101221_AABKFN hsu_y_Page_153.QC.jpg
f17473c069881af8c67cb740e83b04e8
420eada3670c147e6313678cdbe2170bf3106936
F20101221_AABKEZ hsu_y_Page_009.tif
b9e8e9e8a7a10dae63c0b948a7cfc6d0
398e29b61e71a606061bf1de006c603b27a9379c
12969 F20101221_AABJZT hsu_y_Page_047.QC.jpg
608e0c00044fc246d2b07bf1ce99cf45
879023524bb42d52ff4ebec88d44150d0818a4ff
5904 F20101221_AABKGB hsu_y_Page_097.pro
371de1338d83a52ea76a3800b20a093d
7b827fbf822582568b3c0f83ea683c53d606de8a
29973 F20101221_AABKFO hsu_y_Page_067.jpg
f34f169c6a8ce86d817cdb2e1200991a
efcfe1846259d98b757a1c8dc1b5c330c6cebad0
121857 F20101221_AABJZU hsu_y_Page_151.jpg
50e812b55f857ba3c76ac8b6f6504c9e
08cd684445b2085188c30017c391409d4f63b669
1252 F20101221_AABKGC hsu_y_Page_077.txt
6ced23a11275c94b95c9fadad99d2a20
22b97705a4b2d986c873f9dc247248347a1f9699
1628 F20101221_AABKFP hsu_y_Page_091thm.jpg
86672bbe55c5a5db4f2b4eeaa06ce3aa
3b9a2a5a6508ad98ca97948fa4e6fcedbc0aac55
158776 F20101221_AABKGD hsu_y_Page_133.jp2
fcbf1f855f70d828a942de439428ae17
97b4c21cbd2c705ec7c01cb2edd355dac58ab53a
8175 F20101221_AABKFQ hsu_y_Page_109thm.jpg
bc185a64f4eee15964cbac5636750a54
433667ca72d2e64f1bd82012a3eb46d75d0cb9df
33661 F20101221_AABJZV hsu_y_Page_056.QC.jpg
2891ebda1856783559a381733262b216
9f33d3034ba13a1fad812ab848daf7f2b7fd1939
8115 F20101221_AABKGE hsu_y_Page_013thm.jpg
bb9dbe8897e52d3fbba530c4bc54bf37
b82522d9e62c37c73f34771f235544b72b9c02c8
51320 F20101221_AABKFR hsu_y_Page_025.pro
bded12d4daf4bd8c6fdc21db849051ab
9106d513f0cbc6c471e53821e9bfa2beffd48683
871 F20101221_AABJZW hsu_y_Page_010.txt
3bae593889a992bbdbb6de83e7c4e2d5
290a6c014c90f13ec6a4ad427c0ad448a2c9dd03
F20101221_AABKGF hsu_y_Page_128.jp2
4d515889dd1f2dcdf93e956e48f63844
45b71af079119f0fc682855711e8d43c8992f922
1051981 F20101221_AABKFS hsu_y_Page_146.jp2
79f64a0c268b1e09623f69ef71630af9
fc45194a660ae194f4649c378d4e3a17da5f3324
37166 F20101221_AABJZX hsu_y_Page_012.jpg
5abca42bb5dae5487352009e26e24d0d
38de1b4d9f1f08cf3776e7b762444eba9b9f05b0
F20101221_AABKGG hsu_y_Page_087.jp2
8cc648afb9fe5c1e2c773767aa2f094b
7608e40665c174a52f05f245151ffb6aaf03e5fc
417 F20101221_AABKFT hsu_y_Page_096.txt
9d11c87306c880f9c48c80f4be00054e
bfe417a09ffdc5eda08fecc8fbd182cb4ce57df3
10339 F20101221_AABJZY hsu_y_Page_040.pro
dfbcb2927d486d157737290d3c37cae4
5ca460a0bcc3a38e3bf8486518b7623bbfcc0e40
1037087 F20101221_AABKGH hsu_y_Page_131.jp2
5c5913690c97922db7a69565f04a62da
7188c7d5555183d582eebf9945a2e1a79c4e76bc
F20101221_AABKFU hsu_y_Page_040.tif
4eabe5a2e745ac5631942a5f81587f2f
1b1f47414a1a3d5639caf2072c9d8160d61447a1
1051932 F20101221_AABJZZ hsu_y_Page_074.jp2
2de9dc0df12f7091c4f8f340e33e479a
c7d306727bb85451b7839c13d0fb6cd54963d984
38373 F20101221_AABKGI hsu_y_Page_047.jpg
9985ba4bc8eaff25990fb1a839e28012
32c9b979536613c12cb646d29ef490da25389d2f
2096 F20101221_AABKFV hsu_y_Page_061.txt
6271f8038813ad59ca8447ef0485389e
019c02a2e14eec9af0448203a81b3e36a4676d0d
173 F20101221_AABKGJ hsu_y_Page_042.txt
334a15fc585227c00c9b2261fa3055dc
eb0dbb3a258ad15cc20654964c7203f8a5241a27
11083 F20101221_AABKFW hsu_y_Page_044.pro
29ac9e769e91497183e59a04aa6bf10c
1484316cccf3d0dd00bec09b17234d2aa507b01e
F20101221_AABKGK hsu_y_Page_095.pro
0a7dc534b776c38b14cac57e548502b3
05a6cf16228f8bf17df39842640594472af43cfb
F20101221_AABKFX hsu_y_Page_151.tif
a3755f80d20cb529722c22c2333a0738
c536940bc6ac64e7b24f8083cce058ccf3c3e210
4287 F20101221_AABKHA hsu_y_Page_043thm.jpg
4cc3260362c6887a59d321bb3f4a4236
b63535869752c54771a360263d3b087b79ed0bcf
1887 F20101221_AABKGL hsu_y_Page_035thm.jpg
d8c841c5731a8e38ef4779d9ec9ec3e9
743d2c10362b71250fb419895734a618a4b5ec4b
26700 F20101221_AABKFY hsu_y_Page_136.jpg
16f4451166d7cda0dcb99a6a7ed5f450
c8cd7911ce30be60c15e148c93f63679e689f6aa
F20101221_AABKHB hsu_y_Page_134.tif
b0d19de52ac16ec4774d73775a7babe2
04425e0986a7eef9f081d8b63f26039c702478cc
419 F20101221_AABKGM hsu_y_Page_035.txt
3b510caaad30db6167e2f1074e71bf91
60a0bef5c2c3ffbd25dabe362d1a51b38b9a0d8a
F20101221_AABKFZ hsu_y_Page_084.jp2
282a3039f3a3311b259d3c47eb3d86f4
591f5f91b6b0168bed138255d0b827d72ffd3ed4
3674 F20101221_AABKGN hsu_y_Page_064thm.jpg
459dcb18c754c5ffbae8dd780f571d60
f597fa4d07e8ca39a393418d6df44ae6b2e1c37c
8150 F20101221_AABKHC hsu_y_Page_050thm.jpg
35a6f4c0ff02dedae80861d835a198ce
317b7a4381eeaaf062a1817b7ae1caf7d17f8d49
2173 F20101221_AABKGO hsu_y_Page_014.txt
fea4a32cb01a919a48fa85781ce0adb1
64cfd93c98f449ee15e1fe7b2d0cd6dbbc086b54
13982 F20101221_AABKHD hsu_y_Page_115.QC.jpg
b319b021d1b8d0bdc3d604529887bf27
2bc901bb84b6781e89e85f429d5e2ccb9a534bfa
33161 F20101221_AABKGP hsu_y_Page_013.QC.jpg
cc60672a7aeb3b00aca3c89d64b69b3c
d64dd4e7f4afb3f315a0a95b9b8fa5a9d2a21772
4121 F20101221_AABKHE hsu_y_Page_044thm.jpg
85cc268ec8516c04df93689ec030a492
828e94218db53b17537244a00d1bbe6ae9ad7ab9
7465 F20101221_AABKGQ hsu_y_Page_091.pro
67fe2772d1d52c1806cc4d2428aa11a1
4665d1183a4980c67f70d279f5a7c79266cf7df5
55324 F20101221_AABKHF hsu_y_Page_019.pro
890d42854e0eb4b305aff86af8da9d5b
742f8911af29eda19e7fcd5f8c833a16c6bb2ca6
809977 F20101221_AABKGR hsu_y_Page_154.jp2
b82603461f013544e7a7e14194769c76
6a9cde16e95f528f89e580049d2bb2ad510abcb3
6926 F20101221_AABKHG hsu_y_Page_078thm.jpg
ad4aca1bbd79750bb83e37f3f797104c
e8fe6d39a36f49559cdb1d259d94865ddeaf3fcc
10447 F20101221_AABKGS hsu_y_Page_140.pro
8418cc3abead43fe8c24f0cff43d995d
5424c79e6edac581948d241d58a24c9c8e6eae90
F20101221_AABKHH hsu_y_Page_046.tif
3e8c9ef0a13c84d823b2651e8e4d92b3
e9d0df893089ec1bf81c5bba2766276e6aae8501
F20101221_AABKGT hsu_y_Page_115.tif
279e3c4294e9b8a5a44aa4afcb792d37
2a3956f3cad0e01619735bdc21ef2755ac41ecd6
21743 F20101221_AABKHI hsu_y_Page_036.jpg
45302bb0f9dc2402d94bacaaf6d9c817
d13c371e0bfabd887be19383eec284b19ff980d1
40576 F20101221_AABKGU hsu_y_Page_085.pro
a07949cdce259768e7d02d629861365e
3f5756db33ba137337c3e6b7b7901cb3af1e7106
251169 F20101221_AABKHJ hsu_y_Page_035.jp2
2ae55c66171f7d7ed916cf5b9dd55a19
c805e16a6a2b28c39d7608906673e3b07b61ca3a
F20101221_AABKGV hsu_y_Page_116.tif
8d787f576caebaf2a7f1dfd4b77aec46
067fdec2ce81da17e4d7ceae9713a375aba436c4
565 F20101221_AABKHK hsu_y_Page_093.txt
a9ad864119372b94884653d42b1a68a8
8e449472d46f66c9945441a2d279d6b218c668d2
293 F20101221_AABKGW hsu_y_Page_091.txt
711ff93d9675994387d6b32f43314eae
e293a78d9966e7d1d4096fb9873c442c42267d8a
2838 F20101221_AABKIA hsu_y_Page_114thm.jpg
b76d9c3db161f2cd6431b297ad95021a
4905e9c3051e075eba6ff5d480f44462bf389e09
36297 F20101221_AABKHL hsu_y_Page_150.QC.jpg
364340cf4e692fbec0d81879d94da024
5273f845a6a8f7ec4decef489c5be116ca0bb3a4
128277 F20101221_AABKGX hsu_y_Page_150.jpg
792495ed317be9e637ee6a62141d5973
1520f4069c3ae555c5d22482f6c4fcbe1893fb59
6379 F20101221_AABKIB hsu_y_Page_038.QC.jpg
a443b00f2c668ec889c0032d00affeb9
e9b2340d747b59cb0df4ee37767855bacef7720f
77558 F20101221_AABKHM hsu_y_Page_078.jpg
bb4b59cb72cb0b291f741bcb66a0d89e
34884b3c0e2c786f00aa54e472e009ef2f566b63
F20101221_AABKGY hsu_y_Page_042.tif
ef5a6ed50461e387289eefca8b19b8ae
40ce5cb6b450939098105ec8c00bbaca8b163ad0
52353 F20101221_AABKIC hsu_y_Page_106.pro
5d749355fc3b5d4efe73bdefb66d4361
88acfd2fe983775db3198e417ba0e7c415571372
8449 F20101221_AABKHN hsu_y_Page_030thm.jpg
22e7b2086cbbcbc55c0b32af92ee5d25
84c5f02f195d5dab1ef8621bc0c6d99d9f282008
8809 F20101221_AABKGZ hsu_y_Page_061thm.jpg
3a4e3ddc26296bc491f00e09d109ddcd
4be65e94989aba70e54f62c565b33c12ee45a359
1801 F20101221_AABKHO hsu_y_Page_124.txt
58cab3fcb27fc23dad4ee3e46be261f2
499e16d1fa8d0eb7449346c65a700f40b07acc74
83790 F20101221_AABKID hsu_y_Page_122.jpg
0eeeafcdfba1c7ce9e49c0c632f99963
50ad627f69db4d2571e2c67f8e57b4dd353fdb42
3996 F20101221_AABKHP hsu_y_Page_141thm.jpg
22d3310333459aac643ea53588f5ebcb
8d05a0a2ecae2927c9eadc576267003eb8b59489
632362 F20101221_AABKIE hsu_y_Page_090.jp2
b4eb41a94961cdb2a3b44cabb4213c44
c450979638dcb16a026f8822fad340aabc1cd9f4
8746 F20101221_AABKHQ hsu_y_Page_146thm.jpg
a5137a565680651b62654e459251241c
51cb7c159ef07220df553e34ac6706e43830c0a4
F20101221_AABKIF hsu_y_Page_075.tif
1a2e5bd92b4e140008377b3ca7393adb
aff7e8633d701fe7a3e5afe5c56a041c5d497294
36485 F20101221_AABKHR hsu_y_Page_052.QC.jpg
4b3a0546daf1b6fda9eab7629baaa958
60c596e6ee1237d662046b68bb4dee42ee3eab96
95131 F20101221_AABKIG hsu_y_Page_131.jpg
5000e3f7243a8ffbed7e57776c7ad19f
123de35284156a8d5e873a034c52469a5610eedf
2418 F20101221_AABKHS hsu_y_Page_020thm.jpg
7accb01ffb80ef6dad4ba0e7f91f9fab
312a44328045fd40aeeb891af0704a9ab570f710
1729 F20101221_AABKIH hsu_y_Page_063thm.jpg
d4125f907a2ac387e93f14aecda868c6
20ebda37291245cf14f880900e9b285acb8d9ff7
F20101221_AABKHT hsu_y_Page_055.jp2
b85742f8354090c2ed8e74569156e5eb
b665dc1f61f96581a0dea4c7ccb67bf8789c014a
9372 F20101221_AABKII hsu_y_Page_138.QC.jpg
71a07b33215a60f193ee7501f5d43e5e
e01214cb207c68adf7772a16cc4049b10669d7c5
1116 F20101221_AABKHU hsu_y_Page_002.QC.jpg
112718b05f536e37ee15d3271bbbb42e
95dfa0c99fdad48dfde4af439b111fd4455f8b2b
F20101221_AABKIJ hsu_y_Page_008.tif
ee572d64baba6640f470dc16b04a58ab
6537f9d68c8965c9a5f110ee1723d1fa68a7802c
F20101221_AABKHV hsu_y_Page_146.tif
920753ac378bddd6cc716a2f415aa45e
0bc71acc7d731e0fbe01d7f32d9a21b6f895661e
337796 F20101221_AABKIK hsu_y_Page_044.jp2
d157e69b3523a93b1c08d253d5b20d6d
54c2f60cb44a490f0d1231b46674505196f8ee1b
52756 F20101221_AABKHW hsu_y_Page_026.pro
c071d5199dda62593fe7b6fba1aebe11
fcfbf044b4818ad39ffc7f28cc84590e342231b6
F20101221_AABKIL hsu_y_Page_082.tif
3cdcea3367819010937e7f84340cb8c4
0c935e7f34e6d4b4fea4c3ea7958844be1b6ac64
21139 F20101221_AABKHX hsu_y_Page_077.QC.jpg
8bce7108a859c4e2d18534ce95f2ae0a
6cea4d79b6e7596ee5adf8a6dbde9f99b175f699
54038 F20101221_AABKJA hsu_y_Page_103.pro
d0ee62e4808e90b3d4b35578ddb3619d
fd9363585bc4d5b32c3373737c15c50804fcf316
F20101221_AABKIM hsu_y_Page_074.tif
d013156eefcf83fdfe3ba31e228aabe1
41b5e6ff8067074490f64fd23d00f883b2e4e08f
483474 F20101221_AABKHY hsu_y_Page_099.jp2
06a8250be845372c8be23f66a375fd9f
c22c676ce3e39d5f540ad46edbbc127456a8a906
36959 F20101221_AABKJB hsu_y_Page_072.QC.jpg
60345b038ded78a017fca6382dc5728f
3235dc920b631ebdac999e861e168fcf534c4620
32606 F20101221_AABKIN hsu_y_Page_126.QC.jpg
b11d5fbfafd012d45965962f0906e9a4
6c7c0ff1406a8b6376ca54982128000407069588
27284 F20101221_AABKHZ hsu_y_Page_078.QC.jpg
e363023cb7c1e4ce094adb1367516621
ab6a13696f443f802d32ed789866163c2e42c68c
622 F20101221_AABKJC hsu_y_Page_135.txt
a7b77177434f954eb8257721b222b829
03f6fbeccced23d59a0757fbc2e7ab85baa6027f
1138 F20101221_AABKJD hsu_y_Page_132.txt
83dc299b6ff3992831490f204431967e
1c832ab8e2f149c6f9c8ef8969ce39d835f6b600
17681 F20101221_AABKIO hsu_y_Page_064.pro
2c4389697f06ec626946f246707c540b
8367ef99014a9fb66ec1b5216f524371b6c46a07
922047 F20101221_AABKIP hsu_y_Page_053.jp2
5894b32bfef8b7c089ee96b73bfa20d6
67d06b3fbb1a54199e15581ca1f4cbb8d2770471
1051980 F20101221_AABKJE hsu_y_Page_056.jp2
08c39f8dcd6954acb58c30cfe881306a
2a43e701a6ac5821f5ec3aa5a94b746d5012387b
353 F20101221_AABKIQ hsu_y_Page_114.txt
9c837c79940cf4cae7c968c5d9707edb
bff09686e09da879f915193b4015ce17a55be601
8429 F20101221_AABKJF hsu_y_Page_128thm.jpg
7e41dea0880469e3fea35e5a269d5ccb
d44d3d69851ce5419b2c7e565ea674071ba400ec
1051979 F20101221_AABKIR hsu_y_Page_057.jp2
7efa2448794f45d9781b15f24a6676f2
2780bdffdcb69dc85250acc78c83ad3ccbe4ec9e
95767 F20101221_AABKJG hsu_y_Page_107.jpg
06f5245ee7b20acd8b5222be55fdb97f
df59e4a1201a80071f6d41c42d23e6f746adb104
8607 F20101221_AABKIS hsu_y_Page_026thm.jpg
2878dee99884d1bc98cc8af7db8476b1
6b49d8465989a4f6727dfdd09e2ca2a4f62193cd
13603 F20101221_AABKJH hsu_y_Page_046.QC.jpg
7433935768e290cfd5b2b8d7157b9ced
e104fe37af2fd48342f5c61fa6c77394e3effe03
51210 F20101221_AABKIT hsu_y_Page_065.jpg
6e6cbdef820006cbeec47b17344c7a0e
1d2925337e6a05a35b4596411177e778d419497d
110150 F20101221_AABKJI hsu_y_Page_083.jpg
d9a408b5e4006f2ad26055f798b84aa8
b67fcfd176f3fa25bb8897f7c2cde3cb9a3211ab
4677 F20101221_AABKIU hsu_y_Page_048thm.jpg
5da9b2847095001e78f5038ebdd8b136
f8ecf0f20a16a12a161fc10cc46630a9748224e6
2936 F20101221_AABKJJ hsu_y_Page_112thm.jpg
84a32a6c3ad184e9ee5e734742be0a6f
13fc931c2bac91068d3a4f4f911ff7dfd39fc1ba
97770 F20101221_AABKIV hsu_y_Page_056.jpg
fd9eaeb176a9798529d9acdde17ed90d
e25846f7aa704c48d7bfc60a3535efdf685972d3
34985 F20101221_AABKJK hsu_y_Page_093.jpg
38deebe3be7b83f5ff28ac84ea49c8b9
92abbe4e19369578e7dfb6001c2a66bf443dafa9
7149 F20101221_AABKIW hsu_y_Page_125thm.jpg
b4183a2587ffd9ca4dd73d83b334a6b2
635197b680c5de9de2e5fc8b9617f55d1d68a041
F20101221_AABKKA hsu_y_Page_029.jp2
3cf17363ef3dbdbbb6ae89217d2cdd73
c755cbd3ddce84c767ca379c156e4fa085eeae83
7826 F20101221_AABKJL hsu_y_Page_011thm.jpg
36096675e3e0dc47865845cdaee1752a
8531d457e6713cb643a08bbd709534769e31d27d
58650 F20101221_AABKIX hsu_y_Page_132.jpg
0f90ed3346400fa7b7dba11a544fc324
9371a6360cca86e0d9b1a45014d3ff39f3cfce70
F20101221_AABKKB hsu_y_Page_033.tif
55d8dc49637337f3bddc6358d5dc76bf
d0a838e2af359999c726bb6809f7257b5554e618
4247 F20101221_AABKJM hsu_y_Page_115thm.jpg
4245698956ae986205dc28e08c5edb21
3e810b36bf577885216773783ee45d476e220424
220193 F20101221_AABKIY hsu_y_Page_092.jp2
b3e49f351a09bc094260069f6f11dfd0
d855923aa32e1835899f294dffaa6f0938911230
2124 F20101221_AABKKC hsu_y_Page_033.txt
1cdaf8f37bdf5554195471e2d7cc77c3
52a3d5125418d1f9a15ea116bf4013c401fe7206
F20101221_AABKJN hsu_y_Page_125.tif
151d350d0e0d75db775cbb3216c950d4
def415c8cd6f2b05ac4ccdfa05bb780d0bdf3467
33720 F20101221_AABKIZ hsu_y_Page_134.jpg
e589586ba9d4d7d21ef590ff51d32a77
38786ed8410fdd76e0a4e0a5b5b5bf2ffea35aeb
F20101221_AABKKD hsu_y_Page_028.tif
4811fbf56d55464339c5ac97e9f30ae1
ef0feb71fcd757e7300ac6b91f586b0883a84763
2007 F20101221_AABKJO hsu_y_Page_143.txt
09647c4ebf72d4f1c22a24feb3414804
31b89656c64a1a57b723bb921d36c59d328ffcfd
355074 F20101221_AABKKE hsu_y_Page_046.jp2
a9a0f7195d837a34d67db7f2a2a6cc6b
08088c418db2e82de2b57e28a91d3e70e0619e9a
F20101221_AABKJP hsu_y_Page_085.tif
d001b1cf062f8905dbb75ba93388d20f
ea322da645f9dc41e12d382746adcb7e00905885
95778 F20101221_AABKJQ hsu_y_Page_075.jpg
02c8782a3af91fee121d90d4016e32f7
1180904dc352e0951935bc266eeef01234e5b1e0
55178 F20101221_AABKKF hsu_y_Page_016.pro
5a200f030eccb3575b34e5d5d81c955d
fda6a19b09725a271ff7ee01d693600b1279cf39
1051949 F20101221_AABKJR hsu_y_Page_153.jp2
4affa4c916d648a2cfc532ac827f7c77
4ee67f048438833a9c4bac6d75e7186559153464
32336 F20101221_AABKKG hsu_y_Page_027.QC.jpg
75229f0ef286be6eeca70036b1d9e1ab
0bb3a629107d3d2d1cde294e4ad771c28de30525
F20101221_AABKJS hsu_y_Page_024.txt
2a08815d587b5de6797ff918d3db7994
28ab3d82eddd5f37ed7bbc7871083e49a3447185
625055 F20101221_AABKKH hsu_y_Page_132.jp2
d59e10ff8cad287f95aff4da3f77c7e5
1b025521ce4f65836c07be02f0d194f9b8c8276a
F20101221_AABKJT hsu_y_Page_026.tif
af351a6188339eec09c583715c5f382d
5ad2b9b4d8bddc59620c1669adb6bb463d579ae0
5395 F20101221_AABKKI hsu_y_Page_138.pro
af45727c7a040be7664d19f80995583e
dd76fac00f45f67bd52d4d57a7db2fb217ea4647
17076 F20101221_AABKJU hsu_y_Page_012.pro
47917d0f210232370bd406ac67f08323
e0a560199c06597c8270d38820a1fc30da5fc695
8322 F20101221_AABKKJ hsu_y_Page_136.QC.jpg
de09f98151a0b5682722035855682728
ed702fa6d665c141b2b7d9e9f96c8dc41194e508
F20101221_AABKJV hsu_y_Page_024.tif
a95c40b8a67c467d0a4bbc5c0a24a2c6
8e1e09786b555eeccac66ae5f34a686dbc6e7066
9184 F20101221_AABKKK hsu_y_Page_072thm.jpg
7c4c8b4ef0f525cdc65ae6dd1372a9ae
470502a63604cb097a29441a1d6ea843bd818caf
F20101221_AABKJW hsu_y_Page_087.txt
104ab469796375d2278f8087c83e2f97
8f33fec9895bcae5321ebcb26b66222142eacbec
F20101221_AABKKL hsu_y_Page_067.tif
6818c5c08d54fe16d043c489d99d1747
0d78bc0d0e96b162924c1e36fd9dfee1f08eb593
1051895 F20101221_AABKJX hsu_y_Page_106.jp2
75fbb456cc0b9e1d14e53bd81bb31cf3
c9a22bfa618c3418906347628d3378529d88d4d5
1051942 F20101221_AABKLA hsu_y_Page_061.jp2
4221cccd1ae38127f81e704da765ba06
b75be5b6b117fc6d98dee76e96f9ed933863d5e7
46130 F20101221_AABKKM hsu_y_Page_022.pro
f78ac847cc8222e7d666e08562b8c69e
e49dbf94e4aa8980c341c56d0e710114f34e3ff0
50447 F20101221_AABKJY hsu_y_Page_050.pro
cad2f37ba4c9bd94b9185c92b8ea5ce1
f954adb22092f7e14b3a25c99b1e9c9a253ec219
222924 F20101221_AABKLB hsu_y_Page_139.jp2
7463cb645bdc7bdddc2d99f9ed90ca81
6720e953e5fb682f6ba56fa80889a1053bab94f0
105821 F20101221_AABKKN hsu_y_Page_008.jpg
cdecd5ea1a271d77143ed476609f9019
ec90e5eaa7ec2ff989a0f660d96cf2b8b9172e22
4211 F20101221_AABKJZ hsu_y_Page_065thm.jpg
771acfb530b31cd63b9e2ba3ae4760bf
713f534019c1937072b5fc23d288e310e6b2ed54
12975 F20101221_AABKLC hsu_y_Page_045.QC.jpg
3c8f713984749fc08008122ec88d68af
2ff7f806c238b74b5e8dc44eae94d9ed514557b2
46926 F20101221_AABKKO hsu_y_Page_127.pro
0321dbf1a31d791acd095ac6490b1cc8
8b16d35d4db04a225948ff5f8e56156d7410a3d8
F20101221_AABKLD hsu_y_Page_045.tif
c85968ae2d801e2a310509f70410defe
6cdec80ec0193e64ce6008a886f49c352a720fb1
15719 F20101221_AABKKP hsu_y_Page_070.QC.jpg
e9de08a7a64fd276f2de703604ae1584
bd7e1825595408f1da28647bd2020b1e16bd30c6
F20101221_AABKLE hsu_y_Page_027.txt
0b4903e18a2bcb75e38af735ed18d03d
f5ae59ab899db956d4042b732a8dfd1d6b68faab
F20101221_AABKKQ hsu_y_Page_113.tif
f3738b64bf7712c478bd6a03f080cae0
41b71729426f01dff6d21386ef49881231ee1588
105557 F20101221_AABKLF hsu_y_Page_028.jpg
d7990674f612a1932c11b41f5ff5bf5c
cf97ce40a615c92014a0bb5e2aa3fef171543018
34448 F20101221_AABKKR hsu_y_Page_032.QC.jpg
8e8f7475da8ec4eb0264004be2e59e56
2028321fbd6b8f40a80e191acf0a39fb0fb71cd6
4856 F20101221_AABKKS hsu_y_Page_094thm.jpg
2b3b12f39d28a2e852ff422a5a582150
d033a24eff832863e333bd2f98770860a5d57057
9024 F20101221_AABKLG hsu_y_Page_062thm.jpg
5ec568af1efb87c7f5db3911aa709bcc
8327c117b6c4f5fa4216f77737e11654a37b7a71
20320 F20101221_AABKKT hsu_y_Page_038.jpg
84d057a9bfb9e06afb3a28e90b05071e
a8b249c275ef81a21922e14fda1e745acf630650
F20101221_AABKLH hsu_y_Page_103.tif
c37375d3f965052d6b04e7090baebc3e
508215d0e17e51e89f28104493afde409c95f922
94995 F20101221_AABKKU hsu_y_Page_022.jpg
1318bfd9890b6df7b0b947a4cd8af2d5
c50e1c7eec0b191d3faf8b8cc4cbb48f51ee551d
7127 F20101221_AABKLI hsu_y_Page_076thm.jpg
c4ba000195a92ce251bd774a207f74a6
60c76d7c99c447779b3439a9184e0759c5e0c9bd
30753 F20101221_AABKKV hsu_y_Page_075.QC.jpg
856415c7d36802c938077d9e78645e9b
a1cb4f9f3490b991f80de955f19fe8f55a48edc9
4510 F20101221_AABKLJ hsu_y_Page_132thm.jpg
43e22e58518da3829cce859c2ff7603e
8ed96ce236161b82b1052e4a8f969d2d8ccbb9e7
18573 F20101221_AABKKW hsu_y_Page_132.QC.jpg
8b8a82907ce11c20389c42e400222c57
16e20f1cf3b1c0ba1655798111fa483824252168
21808 F20101221_AABKLK hsu_y_Page_034.pro
47463206826a4552292837b9d075f2c7
b6a21de63ce84092749fa7095166f6bfc85df42e
F20101221_AABKKX hsu_y_Page_152.tif
9b8009546c4c66641b504edd47c058d3
c12584256f0afe8f85c9c52e47c94f7981e9bee1
31836 F20101221_AABKMA hsu_y_Page_143.QC.jpg
37aed35f63d9e663ef66b68df46b3690
a5df7017d99cc00014ffaf59b73fa1f48875ea9a
33849 F20101221_AABKLL hsu_y_Page_030.QC.jpg
ae1e3756f1cbaff90e24049a05280930
26943a466a54244a8b3bf57727e6f7bab126fdd0
1951 F20101221_AABKKY hsu_y_Page_056.txt
73fe8d474743afb9241d75806bc28224
4331f4a9460d33eaa948f6965d8fdfb8872f004e
F20101221_AABKMB hsu_y_Page_066.txt
221f6d7d0270effaa23782349df391f1
01d6574b090976587aec44e5a2a80473798b0ff1
36721 F20101221_AABKLM hsu_y_Page_076.pro
e7ac88ab649ee94940f0559b683b1513
660b197947f0eb4124f4dd2bf84da4c49e8e0864
49666 F20101221_AABKKZ hsu_y_Page_030.pro
a667da5c90b6ca13494648a3d7304788
b785798bf2ee6752b089de6cbb5093d144893630
F20101221_AABKMC hsu_y_Page_064.tif
86003386ef35f44f51b56d1c13711983
d6c6ba234044569629fddf712fceefc8926be5ed
49858 F20101221_AABKLN hsu_y_Page_128.pro
4a1947e4299d559c1f3c7364c89ac829
c50fca5406cf2ad267ea4b4d79808151e7725c17
F20101221_AABKMD hsu_y_Page_015.txt
87a11db9c5d342feff2f7aeaca7acad4
dc5fc8b6317cb1eb77e358c4dd92a4c58ced627b
35594 F20101221_AABKLO hsu_y_Page_108.QC.jpg
141c902d7b3aa18164fd5b262a7f376b
9df4dab55b6ab9282e4c1565348a53929b9a5c94
16252 F20101221_AABKME hsu_y_Page_092.jpg
5c688c46add7aa672f7ecf8b15d2e569
8e9845e2a1fbf83f409bba99f2bb1377dcf5769c
870487 F20101221_AABKLP hsu_y_Page_086.jp2
47472d40a7cca6f14478e88dea2a7e27
c259cccedab2fe369edbd83ab850e44b87aa8bda
26845 F20101221_AABKMF hsu_y_Page_138.jpg
5864a16a5ffc09635dc58570f22d31f1
06f2038446df2a13269c4dad25f0356412be8c8d
6325 F20101221_AABKLQ hsu_y_Page_036.QC.jpg
4173bc2baaec4ba8105b63c290d43eee
97f59e74955dc60eea990fd7f966b24b14ac1ffd
F20101221_AABKMG hsu_y_Page_004.jp2
cce3b255ca0e080d930a8795ac60799a
c935846681752f4225442adc532ea317944e5c64
491912 F20101221_AABKLR hsu_y_Page_034.jp2
a34147b3aa4d00c76f19ae525924ae98
0dc756e1bad031d4800e50e2351af1d8d455cc87
F20101221_AABKLS hsu_y_Page_145.tif
c24f802eee588f5c54249a3cd5f40fc8
224da57a8886580f0bae2a6a74e6993b36a43adb
8817 F20101221_AABKMH hsu_y_Page_106thm.jpg
707e8a01abb57e8648e66e73af2c51cd
54f9a611ce67f7d229e6cdb4ccc0159c8c22180a
51875 F20101221_AABKLT hsu_y_Page_061.pro
8710eaf9c20a2baf8653d38692a24779
070bf5951ecfb49c51975f18b015bdc5d822dc7f
F20101221_AABKMI hsu_y_Page_088.tif
1ffbe2f2fd4fc6ec5e29706e87f233c1
d12b6851abfe511d1960bd851e3ac60829341c67
57413 F20101221_AABKLU hsu_y_Page_152.pro
39fa88ca56f277bbbe24cc3bd285988f
9a7caf656adaf67f0c02ed640a9d492994fc88c8
F20101221_AABKMJ hsu_y_Page_047.tif
47bedfa410d091c72d1f0d2314a8d126
4e89ae344975190a9d7006b0d6f1891b630497f7
14888 F20101221_AABKLV hsu_y_Page_098.QC.jpg
56647663d76504312462f3f1ee6d3b5c
359be852f26e553f962e27a4ce6778e1e2bb3059
8526 F20101221_AABKMK hsu_y_Page_149thm.jpg
01ae411893ee8bc86e9fb4a6724c2bb2
a86e5f39113573a66afe5183f82b811d7e33be2a
7279 F20101221_AABKLW hsu_y_Page_096.QC.jpg
b9eadc58602a2fec0b6e7afe7375ba6e
2c5d7c713387eb65054cd14bc394e4ad1eab2ca4
31327 F20101221_AABKML hsu_y_Page_008.QC.jpg
356b00032d2c530f39d1d56280774cfb
568f623076f27f12fdf33bdbd30ce1fdd90427c2
316602 F20101221_AABKLX hsu_y_Page_036.jp2
26711fdee116d1386c5310b082ab01ee
d6a57bca7358a3727df43f36ffcd09acbe5584a7
7550 F20101221_AABKNB hsu_y_Page_003.jpg
3aba50de094564783768df083d2671d5
556cf510791b7af2b9414082fb0c31ffc332b9d1
F20101221_AABKMM hsu_y_Page_028.jp2
7eb8fb7c27edff2c73ac01046719eb58
ca83f3f69a4a76d6cf17b4afff6fa397548a2bdd
F20101221_AABKLY hsu_y_Page_048.tif
d45433e882d2330f2ce5cb9eaf68381c
f819977e1316ed178da81a4f6f1cc6c495759ee4
107364 F20101221_AABKNC hsu_y_Page_005.jpg
f6632828025b48bf86f73ad7c64f45b6
135f9f40b49f00edb529ce2167140e69d3b1ab9b
114276 F20101221_AABKMN hsu_y_Page_152.jpg
b9c2b5a8ef96eff4aa5577af60cd43cb
abbaa560fe2b93aaf69b6be43fda743340fc5e06
F20101221_AABKLZ hsu_y_Page_010.tif
7257abd963889dc97c8bcced5085aae9
343e6bd2094a5074cf533eb27ef1fd02275dac68
79207 F20101221_AABKND hsu_y_Page_007.jpg
fc26d4128e269ac682a7bcac45f5daa6
6ae7b9f43ab03b248ca35330b9c1d8edddbd8640
300 F20101221_AABKMO hsu_y_Page_063.txt
c0e5a8e3ff4fc50568f510b36229c0b7
966c7fa9dd0a231ef86092732eb52af30aa2f48c
95436 F20101221_AABKNE hsu_y_Page_011.jpg
5830bd2b472aab247bfdea8d0be608f0
650b3408226e2233641e7826cefe042a6f2db4d8
3262 F20101221_AABKMP hsu_y_Page_139thm.jpg
b4143cd6350b2a25525e24c1fc09ba52
5f5bd310311b2e149e410a6bc599e9afb6274898
109312 F20101221_AABKNF hsu_y_Page_014.jpg
537d8b9ebc048d49c7747bbaf88f2ad3
a93067aeb7c7c24278406b01a82f6a2ff404911e
1570 F20101221_AABKMQ hsu_y_Page_125.txt
04366ab69f3b7f5e4a6bd422d375cc7f
6e0448d4f55e6711ef6814384b4ca96bfbb8b785
110203 F20101221_AABKNG hsu_y_Page_015.jpg
1829f1efaec770d0b15962045f6f5fe1
cc6d8bf125edddce38ae460b8b85b19772dd45d2
31729 F20101221_AABKMR hsu_y_Page_102.QC.jpg
92be9b80fbc011511ce39ddcb161465a
99346a9ebb9a61463ab6e11fd9501f52064c4e65
113324 F20101221_AABKNH hsu_y_Page_016.jpg
e803af7abe6a38a5b21579a355abee02
10dd0af8ab89f5c732378224078b0e9bab5b1e24
53203 F20101221_AABKMS hsu_y_Page_074.pro
d3b4bd92c72ab0dbb9c57e42e8b51679
f8ce701663bf57cbd5d417d6756f6cdcb973e6b9
F20101221_AABKMT hsu_y_Page_052.jp2
20387e6a381016c22aa588bcbfa43247
2747b50d40d4c44b1219ae42430593bf66be873d
97632 F20101221_AABKNI hsu_y_Page_017.jpg
43a73da5c802a472c720ec999c9cb140
3dfbcc99783c6626de674bc2b3ac09bbb8879295
F20101221_AABKMU hsu_y_Page_126.jp2
830e64aa25b270403af100bd18338821
2164d34cddf1bbfa97e52c6090fff21fa8aeb8a9
84744 F20101221_AABKNJ hsu_y_Page_018.jpg
ca03827c104d69bdc859d20359924ebd
ba9ae03ae0d5e54790bdbda94e38bce9ed59d428
24221 F20101221_AABKMV hsu_y_Page_005.QC.jpg
e70b9cef9fb0435f4206e7bf48c2eee0
94d337ac7dcab5072d13d43882a2ab019639097a
28093 F20101221_AABKNK hsu_y_Page_020.jpg
fbf0c5978c934d32351d55a36e2f9024
b58f5b701f0afb0e135c7590d502c1bb146bd876
F20101221_AABKMW hsu_y_Page_080.txt
c84ebaae186fb5ec2af2ac356f9ea210
b01d471e4c9c6de81805a0cab1d93bb6c0094521
F20101221_AABKMX hsu_y_Page_086.tif
0fc515e9249140c82cd63c172b328bae
b8d83e6eae6a98cba99045afd3ca02f7cbd4e0b5
38163 F20101221_AABKOA hsu_y_Page_044.jpg
6c007e48eadbfb1213cca6f6a8329265
90fddc8f566ab3879aeb6bc525cb76c34f1e8020
26528 F20101221_AABKNL hsu_y_Page_021.jpg
e23defe9a0071ff557346cd6bbb9c009
a1b086c1217d38c635ab633048568783f8e6dbf0
176639 F20101221_AABKMY UFE0021983_00001.mets
71814a9ba66637df9a15b9afbd84df76
8e072dbcfe552dda2af57ce5172060cdeb578f28
37850 F20101221_AABKOB hsu_y_Page_045.jpg
d907c586b6fb6916015373fa8e6f0c39
d24afd11e2c3c055ad3cf7a190ab2cd019981191
106348 F20101221_AABKNM hsu_y_Page_023.jpg
5e74cf8fd25faaec15ef1bde1e434e23
9da184324ff771f9e1c823bca2c0f4cda0ec6af0
107271 F20101221_AABKOC hsu_y_Page_052.jpg
e49af31567407dfed225452ae5d7e798
950f54ecdc9e109148482455fdba2328541dcda8
100075 F20101221_AABKNN hsu_y_Page_024.jpg
e873c1d70c6da665fde35c5a667ac8bd
ac40b7891b759a7a09208da706d2f153b8a2c0c3
98404 F20101221_AABKOD hsu_y_Page_054.jpg
4f665d86da1994f4a522fd4fe9f2ae9a
f7e39e57503eaec951ff6dbebacdea25d089ab37
103940 F20101221_AABKNO hsu_y_Page_025.jpg
91d3350feb30727d0dd634636fc30843
9491a96d4642a4124f0c33c3d0055e4024128b5b
114579 F20101221_AABKOE hsu_y_Page_055.jpg
c8b7ca20766d7255fa629625eb0123b7
ede13d1229403bc7c61c57bba41f3a9486e8c716
106238 F20101221_AABKNP hsu_y_Page_026.jpg
fc17829082f5d9811c3717abb4eed6fc
2bd54bab589510a216651081d9b5ed4ee9245e1a
102906 F20101221_AABKOF hsu_y_Page_057.jpg
f89085b39ee3bb796302271539f39bfb
bad1c62f400bb387ba8f23fbccc9f286e15ad415
101013 F20101221_AABKNQ hsu_y_Page_027.jpg
7576daa5f9b566342d8e1b0dc177ced4
aaed9115be42408e7b858209d2166e209f42393d
86070 F20101221_AABKOG hsu_y_Page_058.jpg
c80a03adb579cbd97ac0d4da45a5f4fe
b577ff790dc3c9bff0ab5263f7a087e48fb7c470
106881 F20101221_AABKNR hsu_y_Page_029.jpg
79db4ad6582041b208a33fdbbad99a89
d0639bd19fbb68712b989799cbef981544353bfa
107298 F20101221_AABKOH hsu_y_Page_059.jpg
4a6038ac278601bc79f8d9d7d0a7ea6c
aead99a7428abb9c469aeb3e1c5b942ec7e99685
102561 F20101221_AABKNS hsu_y_Page_030.jpg
c4cc3dba9a1e648285f4b9b26f26a23c
6bd26c6d66d328cde2a6db8069dc2796dbc8d9fb
108431 F20101221_AABKOI hsu_y_Page_060.jpg
a2905704b62b62c6c4694308a38d0d96
552ad799b4c99e57c40000344c13485ea06fd2f4
48118 F20101221_AABKNT hsu_y_Page_034.jpg
e0ac49a22ee9a1167b1fccc736ce6fb0
c2070ffa6f4567636fc12204a1ec2c2c41884dd1
19345 F20101221_AABKNU hsu_y_Page_035.jpg
667b1a730e086a86e5eaeb94df09b373
e0ec4de8ddcf70c7f0d61225d6da7396393ed31e
112051 F20101221_AABKOJ hsu_y_Page_062.jpg
4f79bc3fa8632ca24fcd23bf5cd15b2e
0496c6268bf6d6d6f341d3c8566b4ed75152b1f7
9211 F20101221_AABKNV hsu_y_Page_037.jpg
0d0ea81d0480d48d21ef7776d8d82fbb
5326edf8aee2ee20bf9ca676eca3f8daa6fd0aeb







EXPOSURE POTENTIAL OF SULFURIC ACID MIST AT PHOSPHATE FERTILIZER
FACILITIES























By

YU-MEI HSU


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

2008




































O 2008 Yu-Mei Hsu




































To my parents, sisters and brother for their constant love, understanding, and support.









ACKNOWLEDGMENTS

I am grateful for Dr. Chang-Yu Wu (my supervisory committee chair) for his patience,

guidance, and encouragement. I also sincerely thank Dr. Chang-Yu Wu for giving me the

opportunity to study at UF, and letting me learn how to be a good professor and to be patient

with others. He is not only my advisor, but also the paragon in my life.

I would like to express my deeply appreciation to Dr. Dale A. Lundgren for his invaluable

research experience and his inspiration. My grateful appreciation would also go to Dr. Jean M.

Andino and Dr. Wesley E. Bolch for their warmth, kindness and their good-natured support.

They have generously given their time and expertise to improve this dissertation.

This study was funded by Florida Institute of Phosphate Research (FIPR). I would also

like to express my gratitude to Dr. Brian K. Birky, Research Director for Public Health, for his

valuable guidance, advice and comments.

I am grateful to Tom McNally, Robert Ammons, and J. Wesley Nall from the Polk County

Health Unit in Winter Haven, Florida for carrying out the pre-sampling and all persons at the

phosphate fertilizer plants to help for the field sampling and they are Alan A. Pratt, Melody

Foley, Martin St. John, Debra Waters, Paul D. Holewski, Tara Crews, Todd W Smith, and Foster

Thorpe.

I thank Dr. Eric Allen for his knowledgeable instruction, and Cheng-Chuan Wang, Hsing-

Wang Li, and Shu-Hau Hsu for assisting me in the field sampling, and Joshua Kollet and

Katherine Wysocki for helping me with the lab experiments.

Many thanks go to my good friends, Ying Li, Jianmei Liu, Jin-Hwa Lee, Anadi Misra, and

Ian Liu, for their patience, kindness, love and warmth. I thank my labmates, Alex Theodore,

Jenkins Charles, Danielle Hall, Nathan Topham, Charles Michael Jenkins, Myung-Heui Woo,

Lindsey Riemenschneider, and Qi Zhang, who kindly assisted my research.












TABLE OF CONTENTS


page

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


LIST OF TABLES ................ ...............8............ ....


LIST OF FIGURES .............. ...............9.....


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


CHAPTER


1 INTRODUCTION ................. ...............13.......... ......


Sulfuric Acid Mist and Its Health Effects ................. ...............13........... ..
Sulfuric Acid Regulations ................. .. ........... ...............14......
Sulfuric Acid Mist in Manufacturing Facilities............... ..... ............1
Manufacturing Processes in Fertilizer Manufacturing Facilities ................. .....................16
Sulfuric Acid Mist Measurement .................. ........... ...............18. ....

Phosphate Fertilizer Manufacture in NTP Report ................. ...............19...............
Research Objectives............... ...............1

2 CHEMICAL CHARACTERISTICS OF AEROSOL MISTS IN PHOSPHATE
FERTILIZER MANUFACTURING FACILITIES .............. ...............21....


Background ................. ...............21.......... ......
Methods .............. .... ...............23..

Sampling Locations ............... ... ...............23.......... ......
Sampling and Analysi s Methods ................. ...............24........... ...
Results and Discussion .............. ...............26....

Background Sites ............... ... ........ .... .......... .............2
Mass Concentrations Measured in the Facilities ................. ..............................26
Ion Concentrations Measured in the Facilities .............. ...............27....
Aerosol Acidity .............. ...............3 1....
Summary ................. ...............32.................

3 SIZE-RESOLVED SULFURIC ACID MIST CONCENTRATIONS AT PHOSPHATE
FERTILIZER MANUFACTURING FACILITIES IN FLORIDA ................. ................ ..48


Back ground ................. ...............48.......... ......
M ethod s .............. ...............49....

Sampling Sites ................... ......... ... ...............49......
Sampling and Analysis Methods ................. ...............50................
Calculation of Fine Mode ................. ............. ...............52. ....
Calculation of Sulfuric Acid Mist Concentration ................. .............................52
Results and Discussion .............. ...............52....












Background Site ................ ...............52..
Plants: Cascade Impactor Samples ................. ...............53................
Attack tank area ................. ...............54........... ....
Sulfuric acid pump tank area............... ...............54..
Belt or rotating table filter floor .................. ...............55...............
Sulfuric acid truck loading/unloading station .............. ...............56....
Granulator on a scrub day .............. ...............56....
Plants: NIOSH Method Samples ................. ...... ....... .. ................ ... .............5
Comparisons of the Results from the Cascade Impactor and the NIOSH Method .........57
Comparisons of Sulfuric Acid Mist Concentrations with OSHA and ACGIH
Regulations .............. ...............60....
Summary ............ ..... ._ ...............61....


4 SIZE DISTRIBUTION, CHEMICAL COMPOSITION AND ACIDITY OF MIST
AEROSOLS IN FERTILIZER MANUFACTURING FACILITIES IN FLORIDA..........._..71


Background ............ .......__ ...............71..
M ethod s .............. ...............72....

Sam pling Sites .............. ....... ... ...............72..
Sampling and Analysis Methods ............ ......_ .. ...............73
Aerosol Thermodynamic Model............... ...............74.
Results and Discussion .............. ...............77....
Aerosol Chemical Species ............ .......__ ...............77..
Sulfuric aid pump tank areas ............ ......_ .. ...............77
Product filter floors .............. ...............79....
Attack tank areas .............. ...............80....
Granulator on a scrub day .............. ...............81....
Aerosol Acidity .............. .. ...............8 1...
Charge balance method .............. ...............8 1....
Aerosol thermodynamic model .............. ...............82....
Sum m ary ................. ...............86....... ......

5 POSITIVE SULFATE ARTIFACT FORMATION FROM SO2 ADSORPTION IN
THE SILICA GEL SAMPLER USED IN NIOSH METHOD 7903 .............. .................. 100


Background ................. ...............100......... ......
Methods .............. ....... ...............102
Sulfur(IV) Oxidation .............. ...............102....
Sulfur Dioxide Adsorption .........._.... ...............103_.._. ......
Results and Discussion ........._._.._........ ...............104....
Sulfur(IV) Oxidation .............. ...............104....
Sulfur Dioxide Ad sorption .........._.... ............... 106....._......
Sulfur dioxide concentration ........._... ....___ ........_. .............0

Sampling flow rate .............. ...............107....
Sampling time ....................... ...............10
Sulfur Dioxide Adsorption Model ....._._.__ ..... .__.. .....__._ ............0
Summary ........._._._..... ..... ...............110....












6 MINIMIZATION OF ARTIFACTS INT SULFURIC ACID MIST MEASUREMENT
USINTG NIOSH METHOD 7903 ................ ...............119...............


Background ................. ...............119................
M ethods .............. ...............121....

Field Sampling............... ...............12
Deactivation M odel .............. ...............122....

Sulfur Dioxide Adsorption ................ ...............123................
Results and Discussion ................ ...............124................

Field Sam pling............... .. .... .. ... ............2
Collection efficiency and concentration of SO2 .......... ................ ...............124
Ratio of S-SO42-/ S-SO2............... .................12
Aerosol loss of HDS............... ...............125.

Sulfur Dioxide Adsorption ................. ...............126...............
Residual Sulfate in Silica Gel Tube............... ...............127.
Minimization of Artifact Sulfate ......._._..........__ ......_._ ....__ ...........128

Aspiration Efficiency............... ..............13
Sulfate M ass Balance .............. ...............131....
Sum m ary ............. ...... ._ ...............132....


7 CONCLUSIONS .............. ...............144....


Conclusion 1 .............. ...............144....
Conclusion 2 ............. ...... __ ...............145...
Conclusion 3 .............. ...............145....
Conclusion 4 ............. ...... __ ...............146...
Conclusion 5 .............. ...............146....


LIST OF REFERENCES ............. ...... ._ ...............146...


BIOGRAPHICAL SKETCH ............. ......___ ...............155...










LIST OF TABLES


Table page

2-1 Sampling locations at phosphate fertilizer plants in Florida............... ...............34

2-2 Analysis conditions for soluble ions ................. ...............35...............

2-3 Detection limit of ion chromatography (IC S 1 500) ......_.__ .... ... .__ ........_._.....3

2-4 Median concentration (Clg/m3) Of ion species at background sites .............. ..................37

2-5 Median concentration (Clg/m3) Of aerosol chemical composition at the granulator on a
scrub day .............. ...............38....

2-6 Statistics of hydrogen ion concentrations (Cleq/m3) at each location. ................ ...............39

3-1 PM23, PMlo and PM2.5 maSs and sulfuric acid concentrations at the attack tank areas .....63

3-2 PM23, PMlo and PM2.5 maSs and sulfuric acid concentrations at the sulfuric acid
pump tank areas .............. ...............64....

3-3 Mass, sulfuric acid concentrations and sulfate/massa ratios of the impactor samples at
the sulfuric acid pump tank areas............... ...............65.

3-4 Statistics of R23, R1o and R2.5 at five types of sampling location............... ................6

3-5 Sulfuric acid concentrations and the ratios measured at two flow rates at the rotating
table filter floors using NIOSH Method 7903 .............. ...............67....

4-1 Equilibrium relations and constants ................. ...............88........... ...

4-2 Median ion species concentrations of cascade impactor samples collected at the
granulator on a scrub day (Clg/m3) .. ...90................

4-3 Aerosol deposition fractions for 3 cases .............. ...............91....

5-1 Experimental conditions of SO2 adsorption ................. ...............111.............

5-2 Rate constants for uncatalyzed oxidation reaction of sulfur(IV) by oxygen ................... 112

5-3 Rate parameters obtained using Equation 5-6 ................. ...............113........... ..

6-1 Statistics of SO2 COncentrations (ppm) .............. ...............133....

6-2 Mean and standard deviation of the sulfate loss to the total sulfate concentration ..........134

6-3 Statistical results of the residual sulfate concentrations of silica gel tubes ................... ..13 5

6-4 Relative error of 4 samplers ................. ...............136..............










LIST OF FIGURES
figure page

1-1 Monoammonium phosphate and diammonium phosphate manufacturing process ...........20

2-1 Manufacturing processes at fertilizer facilities ................ ...............40......_.._...

2-2 Geographic locations of sampling sites .............. ...............41....

2-3 Fine mode and coarse mode aerosol mass concentrations at various locations. ........._......42

2-4 Aerosol chemical species at the sulfuric acid pump tank area............... ..................4

2-5 Aerosol chemical species at the attack tank area. .............. ...............44....

2-6 Aerosol chemical species at the rotating table/belt filter floor. ............. ....................45

2-7 Aerosol chemical species at the sulfuric acid truck loading/unloading station. .............46

2-8 Relationship of cation equivalent weight and anion equivalent weight ............................47

3-1 Sulfuric acid concentrations at 5 types of locations. ................. ...._.._ ................. 68

3-2 Sulfuric acid mist and aerosol mass size distributions ........._..._... ....._._. ...............69

3-3 Comparison of PM23 Sulfuric acid concentrations from the cascade impactor and total
sulfuric acid concentrations from the NIOSH method .............. ...............70....

4-1 Sulfuric acid, phosphoric acid and fluoride concentrations at all locations ......................92

4-2 Relation between the maj or cations (ammonium and calcium) and sulfate
concentrations at the sulfuric acid pump tank areas. ............. ...............93.....

4-3 Aerosol size distributions at the product filter floors............... ...............94.

4-4 Particulate fluoride size distribution at the attack tank areas ................. .........._ .......95

4-5 Relation between ammonium and fluoride concentrations at the attack tank areas ..........96

4-6 Relationship of cation equivalent weight and anion equivalent weight. ...........................97

4-7 Aerosol hydrogen ion concentration size distribution. ............. ...............98.....

5-1 Experimental setup for sulfur dioxide adsorption .....__.___ .... ... .___ ........_._ ......1

5-2 Sulfur(IV) oxidation under four conditions. ................ ...............115..............

5-3 Artifact sulfate concentrations and time-weighted collection percentages (TWCPs). ....116











5-4 Collection index (CI) at four flow rates (Lpm) ...._ ................. ................ ...11

5-5 Relationship between sulfate concentrations from the measurement versus from the
m odel ................. ...............118................

6-1 Three sampling trains............... ...............137

6-2 S-SO2/S-SO42- as a function of SO2 COncentration. ............. ...............138....

6-3 Artifact sulfate concentration as the function of SO2 COncentration ............... .... ........._..139

6-4 Comparison of the sulfate concentrations from the CI, SG, SGHAC and SGHBC ........140

6-5 Aspiration efficiency of three samplers. ............. ...............141....

6-6 Sulfate mass balance between SG and SGHAC/SGHBC ................. .......................142









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

EXPOSURE POTENTIAL OF SULFURIC ACID MIST AT PHOSPHATE FERTILIZER
FACILITIES

By

Yu-Mei Hsu

August 2008

Chair: Chang-Yu Wu
Major: Environmental Engineering Sciences

Strong inorganic acid mists containing sulfuric acid (H2SO4) were identified as a "known

human carcinogen" in a recent report on carcinogens by the National Toxicology Program where

phosphate fertilizer manufacture was listed as one of many occupational exposures to strong

acids. To properly assess the occupational exposure to H2SO4 miSts in modern facilities, the

obj ective of this study was to characterize the true H2SO4 miSt concentration levels.

Three sets of experiments were conducted. Firstly, field sampling using dichotomous

samplers, silica gel tubes and cascade impactors was conducted to collect the PM2.5/PMlo H2SO4

mist concentration, total H2SO4 miSt concentration, and size-resolved H2SO4 miSt concentration,

respectively, at phosphate fertilizer plants.

The H2SO4 COncentrations were found to vary significantly among these plants with H2SO4

pump tank areas having the highest concentration level. When high aerosol mass concentrations

were observed, the H2SO4 miSt had its mode size in the 3.8--10 Clm range that would deposit in

the upper respiratory region.

Secondly, SO2 adsorption and sulfur(IV) oxidation were investigated under various

sampling times, SO2 COncentrations and sampling flowrates. Experimental results verified that

the collecting medium can adsorb SO2 gaS and the extraction procedure of NIOSH Method 7903










aids the transformation of SO2 into sulfate to cause a positive artifact. The experimental data

were also fitted into a deactivation model for estimating the artifact sulfate concentration.

Thirdly, a honeycomb denuder system and the deactivation model were applied to

minimize the artifact sulfate of NIOSH Method 7903 in a field sampling campaign. Both the

system and the model were shown to effectively reduce the artifact sulfate concentration.

However, the concentration thus determined was still higher than that measured by a cascade

impactor which had no artifact. One possible reason is the residual sulfate in the collecting

medium.









CHAPTER 1
INTTRODUCTION

Sulfuric Acid Mist and Its Health Effects

Strong inorganic acid mists containing sulfuric acid (H2SO4) have been reported to

correlate with the incidence of lung and laryngeal cancers in humans [Blair andKazerouni,

1997; Stubiakl\nllrl et al., 1997; Steenlan2d, 1997] and are identified as a "known human

carcinogen" as reported by the U.S. National Toxicology Program (NTP) [USDHHS, 2005].

Sulfuric acid is typically present in the air as a mist. Its chemical characteristics include low

volatility, high acidity, high reactivity, high corrosivity, and high affinity for water. Sulfuric acid

also irritates the human airways, and this irritation may potentially damage pulmonary

epithelium, causing subsequent carcinogenic effects from other inhaled substances.

Although the carcinogenetic mechanism of sulfuric acid mist is not known [Blair and

Kazerouni, 1997], a low pH environment has been reported to induce chromosomal aberrations,

gene mutation and cell transformation. Depurination, which is the removal of a purine (adenine

or guanine) from a DNA molecule, and deamination of cytidine in DNA molecules, which is the

replacement of the amine functional group by the ketone group, has been shown to be enhanced

by exposure to sulfuric acid mist [Swenberg and Beauchamp, 1997]. In a case study of workers

engaged in the manufacture of sulfuric acid, significant increases were observed in the

incidences of genotoxic effects, including sister chromatid exchange (an exchange of segments

between the sister chromatids of a chromosome), micronucleus formation, and chromosomal

aberration in peripheral lymphocytes [M~eng et al., 1995; M~eng and Zhang, 1997]. As exposure

to chemical fumes is suspected to be one of the reasons for lung cancer formation, information

regarding the concentration levels of the acid mists is of seminal importance.









Sulfuric Acid Regulations

The current Occupational Safety & Health Administration (OSHA) 8-hour time-weighted

average (TWA) of permissible exposure level (PEL) for sulfuric acid mist is currently set at 1

mg/m3 with its 15-min short-term exposure level (STEL) set at 3 mg/m3 [CFR]. It is well known

that the deposition of an aerosol in the respiratory system depends on its aerodynamic behavior.

Aerosols with an aerodynamic diameter larger than 1 Clm mainly deposit in the extrathoracic

airways, and ultrafine aerosols with particle sizes less than 0.01 Clm predominantly deposit in the

tracheobronchial region by Brownian motion, whereas aerosols of intermediate diameter deposit

in the alveolar regions [Hinds, 1999]. In considering the effects of aerosol size, the American

Conference of Governmental Industrial Hygienists (ACGIH) has adopted a threshold limit value-

time-weighted average (TLV-TWA) of 0.2 mg/m3 for the thoracic particulate fraction of sulfuric

acid mist [ACGIH, 2004].

Sulfuric Acid Mist in Manufacturing Facilities

Several studies have reported sulfuric acid concentrations in worker environments. In a

sulfuric acid plant in Sweden [Englander et al., 1988], the concentration was at 0. 1 to 3.1 mg/m3

for samples taken in 1979-1980. In a Finnish sulfuric acid plant [.\ynar~ 1978], the

concentration was measured within the range of 0 to 1.7 mg/m3. The concentration in a Russian

plant [Petrov, 1987] was found to be higher, ranging from 1.8 to 4.6 mg/m3. Samples taken in a

sulfuric acid plant at a U. S. copper smelter in 1984 showed lower concentrations ranging form

0.15 to 0.24 mg/m3. Limited information regarding the mist size distribution in sulfuric acid

production plants is available [M~uller, 1992]. It is reported that the size of mist particles ranged

from about 0. 1 Clm to greater than 10 lm.

There are no current reports characterizing sulfuric acid mist in the fertilizer

manufacturing industry. Studies cited in the NTP report were all carried out a few decades ago.









Increased rates of lung cancer in correlation with exposure to chemical fumes in some of these

studies are the reason for recent concerns. A historical cohort study [Hagmar et al., 1991] on

workers employed in a Swedish fertilizer factory was carried out for two cohorts (1236 men in

1906-62 and 2131 men in 1963-85). Significant excesses were found for cancers of the

respiratory tract and lung cancers. The results from U.S. studies, on the other hand, showed

different trends. A cohort study was carried out for men who worked in Florida phosphate

processing facilities during 1949-1978 [Checkoway et al., 1985a; 1985b]3. Lung cancer

mortality was higher than that found in for the entire U.S. population although this rate was

insignificant when compared to the Florida population. Internal comparison for mortality rates

from lung cancer was also conducted for the same population. For workers in sulfuric and

phosphoric acid production, no consistent increase in relative risk for lung cancer was found.

National Institute for Occupational Safety and Health (NIOSH) researchers conducted an

investigation at a phosphate fertilizer production facility in Polk County [Stayner et al., 1985].

Three maj or acids identified by the personal and area samplings were fluorides (mean: 3.39

mg/m3), Sulfuric acid (mean: 0. 11 mg/m3) and phosphoric acid (mean: 0.25 mg/m3). A total of

3199 subjects who had worked at the plant from 1953-1976 were studied. Overall mortality and

morbidity from all cancers were lower than expected, and the risk for lung cancers increased

only slightly. Another study [Block et al., 1988] carried out for male workers in another Florida

phosphate company between 1950 and 1979 showed a significant excess of lung cancer deaths

among white workers in comparison to both U.S. rates and Florida rates. However, when an

internal comparison of job categories was made with respect to lung cancer, no increase was

found for workers exposed to chemical fumes (sulfuric acid, sulfur dioxide and fluorides).









Limited information on acid mist concentrations in the phosphate fertilizer manufacturing

industry is available. In U.S. facilities, the mean sulfuric acid mist concentration ranges from

0.07 to 0.571 mg/m3 [Apol et al., 1987; Cassady et al., 1975; Stephenson et al., 1977]. The

concentration reported in a Finnish study [FIOH, 1990] was 8.3 mg/m3 (1951 data). A Russian

study [TadzTTT~~~~~TTTTT~~~~hibav and Gol'eva, 1976] also reported a high sulfuric acid mist concentration, 2.7

to 9.2 mg/m3. COmpared to foreign facilities, sulfuric mist concentrations in U.S. plants were

low, and the OSHA standard of 1 mg/m3 was met. Size distribution related to sulfuric acid mist

emission in the fertilizer manufacture industry, however, is not available.

Manufacturing Processes in Fertilizer Manufacturing Facilities

Phosphate rock, which is the main useful product of phosphate ore, consists of calcium

phosphate mineral apatite (Ca5(PO4)3(OH, F, Cl)) with gangue constituents including silica

(SiO2), flUOride (F), calcite (CaCO3), dolomite (CaMg(CO3)2), clay, and iron-aluminum oxide

(Fe203, Al203). Several chemical formulas are commonly used for phosphate rock which

includes fluorapatite (Ca5(PO4)3F), chlorapatite (Casg(PO4)3C1) and hydroxyapatite

(Ca5(PO4)3(OH)). The United States is the principal producer of chemical fertilizer using

phosphate rock [Hodge, 1994].

The final products from phosphate fertilizer plants in Florida are mainly diammonium

phosphate (DAP), monoammonium phosphate (MAP), and concentrated sulfuric acid solution.

Wet process using sulfuric acid to react with phosphate rock is commonly used by the phosphate

fertilizer industry in Florida to produce phosphoric acid. Figure 1-1 shows the manufacturing

process flow, which can be divided into three stages.

In the first stage, phosphate rock reacts with sulfuric acid at the attack tank to produce

phosphoric acid with 30-55 wt% P20s. The simplified reactions of the wet process at the attack

tank (also called as reactor) are shown in Reaction 1-1 [Palm, 1992]. Sulfuric acid of 93% is fed









into the attack tank. The violent reaction between phosphate rock and sulfuric acid causes heat

release in the form of vapor, which is evacuated from the attack tank with other gaseous

effluents. A cooling system is needed to maintain the temperature at 70-80 oC in this process

[Becker, 1989].

Cao(PO4 6F2 +10H2SO4 + 20H20 4 10CaLSO4 2H20 +6H3PO4 + 2HF 11

At the second stage, the reaction product from the first stage passes through a rotating table

filter or belt filter to separate phosphoric acid from its byproduct phosphogypsum (calcium

sulfate dihydrate, CaSO4- 2H20). A cooling system is applied to maintain the temperature at

70-75 OC in this process [Becker, 1989]. Some H3PO4 is lost to phosphogypsum from this

process.

At the third stage, the final product, MAP or DAP, is produced by reacting phosphoric acid

of 30-55% P20s (by weight) with ammonia at the granulator, as shown in Reaction 1-2 and 1-3.

When poor product quality is detected, weak sulfuric acid is used to scrub the granulator.

H3PO4 NH3 NH4)H2PO4 (1-2)

H3PO4 + 2NH3 7 NH4 2 HPO4 (1-3)

The manufacturing process of sulfuric acid can also be divided into three stages:

(1) The production of sulfuric acid starts from the combustion of elemental sulfur (S) to

produce sulfur dioxide (SO2). Elemental sulfur is pumped into the sulfur burner and is burned

with dry combustion air to form SO2. Sulfur dioxide and excess air leave the burner at

700-1,000 oC which need to be cooled to 425 OC to protect the converter in the next stage

[M~uller, 1992].

(2) At the second stage, SO2 is subsequently oxidized to form sulfur trioxide (SO3) when

passing through a series of catalytic converters. The reaction proceeds as Reaction 1-4 which is









an exothermic reaction. This equilibrium equation is controlled by the concentrations of SO2,

SO3 and temperature. High SO3 COncentration and temperature can favor the reverse reaction.

The temperature is 180-250 oC when gas leaves the converter [M~uller, 1992].


SO2(g) +-02 ++ SO3(g)
2 (1-4)

AH = 41,400 Btu/1b-mol

(3) Sulfur trioxide can quickly combine with water vapor to produce sulfuric acid at the

sulfuric acid pump tank. The reaction is shown as Reaction 1-5, which is also an exothermic

reaction [Ridler, 1959].


SO3(g) + H20.) H2SO4(1) (1-5)

The produced sulfuric acid is stored in tanks. Some plants produce more sulfuric acid than

needed and sell excess sulfuric acid solution to other companies. Trucks are used for the

transportation of sulfuric acid solution. The loading and unloading of sulfuric acid solution is

carried out through a nozzle at a truck station.

Sulfuric Acid Mist Measurement

Sulfuric acid exists in the mist form which can be collected by filtration. NIOSH Method

7903 is an OSHA approved method which employs a silica gel tube to collect acid mist. The

silica gel tube consists of one section of glass fiber filter plug and two sections of silica gel. The

glass fiber filter and the silica gel are designed to collect aerosols and acid gases, respectively.

NIOSH Method 7903 is the method commonly applied for personal sampling in the workplace

due to its convenience. However, both the glass fiber filter and silica gel can adsorb SO2 [Chow,

1995; Lee and2~ukund, 2001] that will lead to an overestimate of sulfate.









Phosphate Fertilizer Manufacture in NTP Report

Although phosphate fertilizer manufacture was listed in the NTP report [USDHHS, 2005]

as one of many occupational exposures to strong inorganic acids, the occupational exposures to

inorganic acid mists containing sulfuric acid existing at levels equal to or greater than the PEL

used by the International Agency for Research on Cancer (IARC) are based on results obtained

during the period 195 1 to 1976. In addition, all of the results greater than the PEL for sulfuric

acid are from outside the U.S. [USDHHS, 2005]. The significant improvement of health and

safety measures in the US fertilizer industry in the past decades is expected to have significantly

lowered current levels. Excessive respiratory protection may be costly and stressful and still not

provide any beneficial reduction in exposure. Thus, characterization of the true exposure level at

modern facilities is a necessary step to the establishment of the best policy for worker protection.

Research Objectives

Five obj ectives were set in this doctoral research study to accurately characterize sulfuric

acid mist at phosphate fertilizer facilities. The first obj ective is to characterize the maj or water

soluble ionic species of PM2.5 and PM2.5-10 at phosphate fertilizer facilities. The location with

high chemical species concentrations can be identified as well. The characterization can be

applied for the establishment of the best policy for worker protection.

The second obj ective is to determine the sulfuric acid mist concentrations with size-

resolved information by a cascade impactor and the total sulfuric acid mist concentration using

NIOSH Method 7903. The study also seeks to determine the correlation between these two

samplers at the phosphate fertilizer facilities.

The third obj ective is to determine the chemical characteristics of mist aerosols in the

current phosphate facilities with size-resolved information and to estimate the aerosol hydrogen

ion concentration at the phosphate fertilizer facilities using a thermodynamic model.









The fourth obj ective is to verify and quantify the effect of SO2 interference on the artifact

sulfate in NIOSH Method 7903. The oxidation of sulfur(IV) into sulfate and SO2 adsorption

following the NIOSH protocol were also investigated in this study.

The fifth obj ective is to investigate the effectiveness of two methods, a deactivation

model and a honeycomb denuder system, to minimize the artifact sulfate in a field sampling at

the phosphate fertilizer facilities.


























Figure 1-1. Monoammonium phosphate and diammonium phosphate manufacturing process









CHAPTER 2
CHEMICAL CHARACTERISTICS OF AEROSOL MISTS IN PHOSPHATE FERTILIZER
MANUFACTURING FACILITIES*

Background

Phosphate products are widely used around the world for fertilizer (90%), detergents

(4.5%), animal feed (3.3%), and food and beverages (0.7%) [Becker, 1989]. The United States is

the second largest producer of phosphate fertilizers in the world [Bhaskska~ran et al., 2004]. The

manufacturing process flow of phosphate fertilizer is shown in Figure 2-1. In Florida,

phosphoric acid is usually produced by a wet process by reacting H2SO4 With naturally occurring

phosphate rock in a reactor that is referred to in the industry as the "attack tank".

Phosphogypsum is a byproduct of this process. The simplified reactions of the wet process are

as Reaction 2-1 [Becker, 1989]:

Cao(PO4 6F2 +10H2SO4 + 20H20 4 10CaLSO4 2H20 +6H3PO4 + 2HF(21

Rotating table filters and belt filters are used to separate phosphoric acid and

phosphogypsum. Phosphoric acid of 30-55% P20s (by weight) reacts with ammonia to produce

MAP or DAP in the granulator, as shown in Reaction 2-2 and Reaction 2-3 [Hodge and

Popovici, 1994].

H3 PO4 NH3 ~ NH4 )H2PO4 (2-2)

H3PO4 + 2NH3 ~ NH4 2 HP~O4 (2-3)

In these facilities, H2SO4 USed for digestion of phosphate ores is produced by sulfur

combustion [Hodge andPopovici, 1994]. The sulfuric acid production process initiates in a

sulfur burner. The resulting combustion gas consists primarily of SO2 that is routed to a series of


* Reprinted with permission from Hsu, Y.-M., Wu, C.-Y., Lundgren, D. A., Nall, J. W., Birky, B. K.,
2007. Chemical Characteristics of Aerosol Mists in Phosphate Fertilizer Manufacturing Facilities. J.
Occupy Environ. Hy g. 4, 17-25.









catalytic converters to transform the sulfur dioxide into SO3. By mixing with water, sulfur

trioxide quickly forms sulfuric acid that is moved from pump tanks to storage tanks for use in the

production of phosphoric acid. Some facilities produce more sulfuric acid than needed and sell

the excess to others.

Reduced pH environments are known to enhance the depurination (i.e. removing a purine

(adenine or guanine) from a DNA molecule) rate of DNA and the deamination (that is, replacing

the amine functional group by the ketone group) rate of cytidine [USDHHS, 2005], which can

cause DNA damage or mutation. Sulfuric acid also irritates the human airway, and this irritation

may potentially damage the epithelium, causing subsequent carcinogenic effects of other

substances [ACGIH, 2004].

Phosphate fertilizer manufacture was listed in the report as one of many occupational

exposures to strong inorganic acids. However, the occupational exposures to inorganic acid

mists containing sulfuric acid existing at levels equal to or greater than PEL used by the IARC

are based on results obtained during 195 1-1976 [1992]. In addition, all of the results used by the

IARC greater than the PEL for sulfuric acid were from outside the United States [TadzTTT~~~~~TTTTT~~~~hibav

and Gol'eva, 1976]. In U. S. facilities, the sulfuric acid mist mean concentration ranged from

0.07 to 0.57 mg/m3 [Apol et al., 1987; Cassady et al., 1975; Stephenson et al., 1977]. The

significant improvement of environmental, health, and safety measures in the U.S. fertilizer

industry in the past is expected to have greatly lowered current levels.

The obj ective of this chapter was to characterize the thoracic particulate fraction of

chemical species concentration level at phosphate fertilizer facilities in Florida. The chemical

species concentration level would be applied to estimate the required sampling time for a cascade

impactor sampling. The information is critical for health risk assessment, is useful in identifying









the key sources. Such characterization is a necessary step in the establishment of the best policy

for worker protection.

Methods

Sampling Locations

Due to the different manufacturing process designs among these phosphate fertilizer plants,

the aerosol emission sources at each plant are different. Two to five locations at each plant were

selected where sulfuric acid mists may exist. These locations included the top of sulfuric acid

pump tank, attack tank (reactor), filtration floor, sulfuric acid truck loading/unloading station,

and granulator on a scrub day (Figure 2-1 and Table 2-1). In total, there were 24 sampling

locations in eight plants and two background locations included in this study. The geographic

locations of these sampling sites are shown in Figure 2-2. Winter Haven, FL and Gainesville,

FL, were employed as the background locations. Gainesville is located between the plant in

north FL and the plants in central FL. Winter Haven is located at the east of the plants in central

FL. The distance between Gainesville and the nearest plant in north FL is approximately 60

miles; the distance between Winter Haven and the nearest plant in central FL is approximately

20 miles. Three samples were obtained at each location. Due to the low particulate

concentration at the background locations, sampling was carried out for 24 h.

The sulfuric acid pump tank is where the newly produced sulfuric acid is distributed to

sulfuric acid storage tanks. The leakage of SO3 fTOm ducting can bind with moisture to form fine

aerosols. Gas duct leaks in the sulfuric acid production plant are normally repaired quickly and

are extremely difficult to sample. The sampling location chosen in the sulfuric acid production

plants was on top of the sulfuric acid pump tank because the pump tank is vented to atmosphere

and has a large throughput.









The attack tank is where sulfuric acid reacts with phosphate rock. The strong reaction

between sulfuric acid and phosphate rock causes the production of gaseous species that attach to

existing aerosols or form aerosols by reacting with other gaseous species. Therefore, in the

phosphoric acid production plants, sampling locations were chosen near the attack tanks.

Two different types of product filters are used in these plants, including a belt filter and a

rotating table (commonly the Bird type) Eilter where the reaction products are filtered and

washed under vacuum to separate phosphoric acid and phosphogypsum. The rotating table filter

is a circular table with pie-shaped fi1ter pans that rotate in a circular motion while filtering out

the phosphogypsum. The individual pan tilts and dumps the washed phosphogypsum into a

hopper for pumping to a phosphogypsum storage mound. The belt filter is a continuous belt that

washes the phosphogypsum at different sections along the belt and dumps the gypsum at the end

of the belt into a hopper for distribution to the phosphogypsum storage area. Sampling locations

were selected adj acent to both types of filters.

The sulfuric acid truck station (an outdoor location) is where trucks load or unload sulfuric

acid. Sampling was conducted near the loading nozzle as this location is where sulfuric acid is

most likely to enter the atmosphere.

The last sampling location was at the granulator on a "scrub" day when the product line is

shut down to clean the piping and ductwork by spraying a mixture of process water and small

amounts of sulfuric acid. This solution is recirculated from open-top tanks during the scrub

cycle. Sampling was conducted adj acent to these tanks.

Sampling and Analysis Methods

Dichotomous samplers (model SA241 CUM; Anderson Instruments, Atlanta, GA) were

used to separate aerosols into two sizes: fine mode (aerodynamic diameter smaller than 2.5 Clm










or PM2.5) and coarse mode (aerodynamic diameter ranges from 2.5 to 10 Clm or PM2.5-10). The

reason for using dichotomous samplers is because PMlo curve is similar to the thoracic fraction.

The sampling flow rates were 1.67 and 15 Lpm for the fine and coarse mode, respectively. The

sampling time in the production facilities was 12 h. Three samples were obtained at each

location. The total sample numbers (plants) at the pump tank area, rotating table/belt filter floor,

attack tank area, truck station, and the granulator (on a scrub day) were 27(8), 27(8), 6(2), 9(3),

and 2(1), respectively.

Zefluor, or backed Teflon, membrane filters (P5PJ001, 8 x 10 inches, 2.0 Clm, Pall Corp.,

Ann Arbor, MI) were used to collect the aerosol. Zefluor membrane filters provide high

collection efficiency and low reactivity with acidic gases [Baron and Willeke, 2001; Chow,

1995]. A microbalance (model MC 210 S, Sartorius Corp., Edgewood, N.Y.; readability 10 Gig)

was used for weighing the particle mass. Filters were placed in a desiccator at room temperature

for pre- and post-conditioning for 24 h before weighing the filters. The vapor pressures of

sulfuric acid mist and phosphoric acid are low and they remain in the mist form at this condition.

An ion chromatography (IC) system (Dionex ICS 1500; Dionex Corp., Atlanta, GA) was used to

analyze the soluble ions of the samples. The analysis conditions are shown in Table 2-2.

Fluoride, chloride, nitrate, sulfate, phosphate, sodium, potassium, calcium, and magnesium were

analyzed and the IC's detection limit for each ion is listed in Table 2-3.

Using this analysis method, it is impossible to distinguish whether the sulfate measured is

from sulfuric acid or other sulfates (e.g., ammonium sulfate, calcium sulfate). Consequently, all

particulate sulfate was assumed to be sulfuric acid mist in this study [NIOSH, 1994]. The

conversion of sulfuric acid mist concentration from sulfate concentration follows Equation 2-4

listed below:










[HO,1(~im;= [~i ],,,,dh C (g/m )98(M.W. ofH,SO4)
measredylC96(M.W. of SO~ )

Results and Discussion

Background Sites

Table 2-4 shows the median ion species concentrations at the two background sites. The

median PM2.5 and PM2.5-10 COncentrations (N = 3) in Gainesville were 15.0 and 11.2 Clg/m3, and

the maximum PMlo concentration was 29.7 Clg/m3. The median PM2.5 and PM2.5-10

concentrations (N = 3) in Winter Haven were 26.2 and 13.7 Clg/m3, and the maximum PMlo (the

sum of PM2.5 and PM2.5-10) COncentration was 50.9 Clg/m3. The mass concentrations in Winter

Haven were slightly higher than in Gainesville. The mass concentrations of the fine mode were

higher than the coarse mode. Overall, the mass concentrations at both locations were not high.

The major species of the PM2.5 in Gainesville were sulfate (median: 5.05 Clg/m3) and ammonium

(median: 4.24 Clg/m3), and the concentrations of all other species were low. The major species in

Winter Haven was ammonium. The median concentration was 5.10 Clg/m3 for PM2.5 and 7.78

Clg/m3 for PM2.5-10. Phosphate concentrations were lower than the detection limit at both

locations; however, low fluoride concentrations were observed in Winter Haven (0.086- 0.149

Clg/m3). The major fluoride emission source is fertilizer facilities in central Florida [USEPA,

1995]. Hence, the particulate fluoride might come from these facilities.

Mass Concentrations Measured in the Facilities

The aerosol mass concentrations at 24 locations are summarized in Figure 2-3 with their

maximum, minimum, mean, and median at each type of location. As shown, the mass

concentration varied significantly from one location to the other and within each type of location.

There are various factors that may affect the concentration level, such as production activity









level and types of processes. Weather and ambient aerosols may also affect outdoor locations

like the sulfuric acid pump tank and the sulfuric acid truck station.

The rotating table/belt filter floor had a high aerosol mass concentration due to mechanical

agitation and gas species condensation. The median values (range) of the Eine mode and the

coarse mode were 77.8 (27.8-381) and 79.5 (25.7-338) Clg/m3, TOSpectively. Among the various

locations, the attack tank had the highest Eine mode aerosol mass concentration. The median

value (range) of the Eine mode was 3 89 (98.6-894) Clg/m3 and its coarse mode mass

concentration (range) was 58.1 (13.6-266) Clg/m3. At the attack tank, the strong reaction of

sulfuric acid and phosphate rock releases heat and some species that have high vapor pressures.

In general, the attack tank and the rotating table/belt filter floor are the two maj or emission

sources of aerosol in these facilities whereas the aerosol mass concentrations are lower at the

sulfuric acid pump tank, the truck loading/unloading station and granulator on a scrub day.

Ion Concentrations Measured in the Facilities

The results of ion analysis are presented in Figure 2-4 to Figure 2-7 for the sulfuric acid

pump tank area, the attack tank area, the fi1ter floor, and the truck station, respectively. At the

sulfuric acid pump tank area (Figure 2-4), sulfate was the dominant species, and the median

(minimum and maximum) concentrations were 15.3 (1.33 and 77.0) and 5.1 (0.501 and 104)

Clg/m3 for the Eine mode and the coarse mode, respectively. The maximum PMlo sulfate

concentration, 181 Clg/m3, iS equivalent to a maximum sulfuric acid mist concentration of 185

Clg/m3, which is lower than 0.2 mg/m3, the TLV-TWA of the thoracic particulate fraction of

sulfuric acid recommended by ACGIH.

In the sulfuric acid manufacturing process, SO3 is present in the ductwork of the plant

before being absorbed into the acid stream. Any leak from the duct can release SO3 that can









readily react with moisture in the air to produce H2SO4 miSt [Finlayson-Pitts and Pitts, 2000].

The temperature at the pump tank is about 110 oC, which is not high enough to evaporate H2SO4

(boiling point around 330 oC) [Wea~st, 1988]. The leakage of SO3 fr0m surrounding ductwork is

likely to be the main source that causes higher relative sulfuric acid mist concentrations at the

sulfuric acid pump tank area. In addition to the SO3 emiSsion rate, the other important factor that

can influence the aerosol size of the mist is humidity. High humidity assists the condensational

growth of sulfuric acid mist, whereas a high emission rate aids the nucleation of ultrafine sulfuric

acid mist. Ammonium was also the dominant species, and the median (minimum and maximum)

concentrations were 8.7 (1.40 and 22. 1) and 2.5 (0.428 and 25.9) Clg/m3 for the fine mode and the

coarse mode, respectively.

The fast reaction of ammonia gas and sulfuric acid mist to form ammonium sulfate

((NH4)2SO4) is likely the mechanism responsible for the ammonium concentration. Although

this location is not expected to have ammonia gas emissions, ammonia comes from other

locations in these plants and also commonly exists in ambient air. The presence of ammonia gas

can neutralize the acidity of sulfuric acid or acidic aerosol. Regarding other ions such as

fluoride, chloride, phosphate, sodium, potassium, magnesium and calcium, these outdoor

locations have generally low concentrations.

At the attack tank area (Figure 2-5), fluoride was the dominant species, and the median

(minimum and maximum) concentrations were 105 (7.25 and 455) and 21.2 (1.22 and 85.9)

Clg/m3 for the fine mode and the coarse mode, respectively. The maximum PMlo fluoride

concentration was 541 Clg/m3; the TWA-PEL of fluoride is 2 mg/m3. The attack tank is where

sulfuric acid reacts with phosphate rock that mainly contains calcium, phosphate, and fluoride.

The reaction products are phosphoric acid and phosphogypsum, which will precipitate.









The violent reaction also produces heat and water vapor from this process, and gaseous

fluorides (silicon tetrafluoride (SiF4) and hydrogen fluoride (HF)) are the byproducts from this

process [Mann, 1992]. In the weak phosphoric acid production of 30% P20s, SiF4 is the major

fluoride form to volatilize because of its higher vapor pressure. The molar ratio of HF to SiF4 is

less than 2 when the phosphoric acid is below 50% P20s. Increasing the concentration of P20s to

50% makes more HF escape from liquid phase [Denzinger et al., 1979]. In the presence of

atmospheric moisture, fluorosilicic acid (H2SiF6), hydrogen fluoride and silicon oxide are created

(Reaction 2-5 to Reaction 2-8) [Hodge andPopovici, 1994]. The gaseous fluorides adsorption

process is dependent on the surface area of water droplet. Therefore, the size distribution of

water droplets may influence the partitioning of fluoride.

SiO2 + 4 HF 4 SiF4 + 2 H20 (2-5)

3 SiF4 + 2 H20 4 2 H2SiF6 + SiO2 (2-6)

SiO2 + 6 HF 4 H2SiF6 + 2 H20 (2-7)

H2SiF6 4 SiF4 + 2 HF (2-8)

For the rotating table/belt filter floor, fluoride was the dominant species for the Eine mode

(median: 3.9 Clg/m3; range: 0.173-106 Clg/m3), whereas phosphate, sulfate, and ammonium were

also important. The rotating table and belt filters are open systems used to filter intermediate

products to obtain 30% phosphoric acid. The temperature of this process is about 100 oC.

Because it is an open system operating at high temperature, water, gaseous fluoride, and

phosphoric acid evaporate to form Eine mist aerosols. Ammonia gas is easily absorbed into the

acid mist to neutralize the acid. Phosphate was the maj or constituent for the coarse mode

(median: 22.3 Clg/m3; range: 2.33-122 Clg/m3), whereas sulfate and ammonium were also

important. For PMlo, phosphate was the major species and its highest concentration was 170










Clg/m3. Filter cake formed in this filtration process, which mainly consists of phosphogypsum,

contributes to the loading of the coarse particles due to mechanical agitation. As shown, the

calcium concentration of the coarse mode is higher than the fine mode. The vaporized acids can

also condense on these mechanically agitated wet particles. Therefore, they were also present in

the coarse mode.

At the sulfuric acid truck loading and unloading stations, ammonium and sulfate were the

dominant species for the fine mode (median: 9.5 and 4.2 Clg/m3, TOSpectively) and the

concentration ranges of ammonium and sulfate were 4.77-32.9 and 0.613-9.74 Clg/m3,

respectively. Phosphate, ammonium and sulfate were the dominant species for the coarse mode

(median: 8.5, 5.2 and 3.3 Clg/m3) and their ranges were 1.26-48.4, 0.3-13.6, and 1.36-12.8

Clg/m3. For this outdoor location, the possible period to release sulfuric acid is when trucks load

and unload sulfuric acid, which is usually 2-3 h/day. This emission is not a continuous source.

Hence, most of the time, the aerosol loading was greatly influenced by the ambient condition and

likely the aerosols were from other locations in the plant. In summary, the concentrations of all

species were low.

Sampling for the granulator on a scrub day was conducted in one facility only. The

median concentrations (N = 2) are shown in Table 2-5. For the fine mode, fluoride and

ammonium were the dominant species; for the coarse mode, phosphate, fluoride, and ammonium

were the major ones. In normal operation, 30% phosphoric acid is present in this system. In this

plant, the period of scrubbing was about 4 h, and a weak sulfuric acid solution was used to scrub

the piping and ductwork. Nevertheless, sulfate concentration was low at this location.









Aerosol Acidity

The hydrogen ion plays an important role in the carcinogenetic mechanism and skin

irritation. Therefore, hydrogen ion concentration in the aerosol provides very useful data for

assessing the health risk. Assuming that charge balance of those ions exists in the aerosols in

this study, hydrogen ion concentration can be estimated by Equation 2-9:

[H ] (Cleq/m3) = [Anion] [Cation] (2-9)

Anion species include phosphate, fluoride, sulfate, chloride, and nitrate; cation species

include ammonium, sodium, potassium, calcium, and magnesium. The statistics of hydrogen ion

concentrations at each location are shown in Table 2-6. Fine mode aerosol at the attack tank area

had the highest hydrogen ion concentration, 20.7 Cleq/m3 Or 20.7 Clg/m3 with a median of 4.4

Cleq/m3, indicating they were strongly acidic. On the other hand, the average hydrogen ion

concentration at sulfuric acid pump tank areas (fine mode) and sulfuric acid truck stations were

negative (i.e., the aerosol had excess OIT), which indicated that the aerosol was somewhat basic.

The relationship of cation equivalent weight and anion equivalent weight is shown in

Figure 2-8, which can be used to determine the acidity of the aerosol. The key factors that can

control the acidity of the aerosol are the amounts of basic or acidic species. The acidic species in

these plants include fluoride, sulfate, and phosphate. The only basic gas in the plant that can

neutralize all acid species (i.e., sulfuric acid, gaseous fluoride, and phosphoric acid) is ammonia.

In Figure 2-8, the cation and anion equivalent weights can be divided into two regions: smaller

than 3 and larger than 3 Cleq/m3. In the case of low acid species loading, the aerosol acidity can

be influenced by the presence of basic species; however, at high acid species loading, the limited

amount of basic species may not be enough to neutralize the aerosol acidity. At the rotating

table/belt filter floor and the attack tank, the main acidic gases such as gaseous fluoride and









phosphoric acid are emitted from the process, and there is no source of ammonia in these indoor

locations. Therefore, the aerosols remain acidic.

On the other hand, both the sulfuric acid pump tank area and truck loading/unloading

station are outdoor locations. Wind assists in the mixing of the aerosols with ambient air. For

the sulfuric acid pump tank, the high surface area of the fine mode aerosol is easier for mass

transfer of basic gases, while the coarse mode aerosol needs a longer time to reach neutralization

[M~eng and Seinfeld, 1996]. The truck loading/unloading station does not have any significant

sulfuric acid emission source, and aerosols at this location come mainly from other locations.

Hence, the aerosol acidity depends on the wind strength/direction and aerosol emission source.

The results also imply that the aerosol becomes less hazardous as it moves away from the

emission source due to atmospheric dilution and neutralization by basic species.

Summary

Aerosol sampling using dichotomous samplers was carried out at five types of locations in

eight fertilizer facilities. The highest sulfate concentration was obtained at the sulfuric acid

pump tank area. The maximum sulfuric acid concentration measured in PMlo, including fine

mode and coarse mode, was 0.185 mg/m3. At the attack tank area where phosphoric acid is

produced by reacting sulfuric acid with phosphate rock, fluoride was the dominant species. The

maximum fluoride concentration in PMlo was 462 Clg/m3. At the rotating table/belt filter floor,

phosphoric acid is separated from phosphogypsum by rotating table/belt filter and the high

temperature is favorable for the evaporation of fluoride and phosphoric acid, which can

subsequently form fine aerosol or condense on phosphogypsum aerosol created by mechanical

agitation. The maximum phosphate concentration in PMlo was 170 Clg/m3. On a scrub day, a

weak sulfuric acid solution is used to clean the piping and ductwork of the granulator for an









average of 4 h per day. Particulate sulfate concentrations were low during the scrubbing activity.

At the truck loading/unloading station, the possible emission period is around 2-3 h/day, and this

emission is not continuous. The concentration levels at the loading/unloading station were low

and were greatly influenced by outdoor conditions.

The PMlo concentrations of sulfuric acid mist at these facilities were lower than the TLV-

TWA standard of 0.2 mg/m3 TOCOmmended by ACGIH for the thoracic fraction of sulfuric acid

aerosol. The maximum PMlo of sulfuric acid mist was observed at the sulfuric acid pump tank

area and was close to but found to be lower than the TLV-TWA. If monitoring of personal

exposures to sulfuric acid mist is to be required, these efforts should focus on workers with

activities in this area.










Table 2-1. Sampling locations at phosphate fertilizer plants in Florida
Location Sulfuric Acid Filter Floor Filter Floor Granulator Truck Attack Tank
Pump Tank (Rotating Table) (Belt) (on a Scrub day) Station
Plant A X X X X X
Plant B XX X X
Plant C X X X
Plant D X X
Plant E X X
Plant F X X
Plant G X X X X
Plant H X X
Total sample number 27 21 6 2 9 6










Table 2-2. Analysis conditions for soluble ions
Dionex ICS-1500 Cation Anion
Analyzable species K Na Ca2+, Mg2+, N4+ F-, Cl-, NO3-, SO42-, PO43-
Extraction solution volume (D.I. water) 10 ml 10 ml
Analytical column lonPac CS12A lonPac AS9-HC
Guard column lonPac CGl2A lonPac AG9-HC
Suppressor ASRS-ULTRA II CSRS-ULTRA II
Eluent 18 mM methanesulfonic acid 9.0 mM sodium carbonate
Flow rate 1.0 ml/min 1.0 ml/min
Inj section volume 50 Cll 50 Cll
Analysis time 20 min 30 min










Table 2-3. Detection limit of ion chromatography (ICS 1500)
Anion ppm Cation ppm
F- 0.10 Na' 0.05
Cl- 0.02 NH4' 0.12
NO3- 0.07 K+ 0.05
PO43- 0.06 ,gMg2+ 0.04
SO42- 0.12 Ca+ 0.08










Table 2-4. Median concentration (ug/m3) Of ion species at background sites
Mass F- Cl- PO43- SO42- Na' NH4' K' Mg"' Ca"'
Gainesville, FL
Fine 15.0
Coarse 11.2
Winter Haven, FL
Fine 15.0 0.06
Coarse 15.0









Table 2-5. Median concentration (ug/m3) Of aerosol chemical composition at the granulator on a scrub da

Mean (Fine mode) 14.1 2.11 2.53 1.93 3.75 0.389 8.77 0.2 0.064 0.391
Mean (Coarse mode) 8.69 1.31 1.46 37.0 4.75 1.22 10.1 0.238 0.527 1.16










Table 2-6. Statistics of hydrogen ion concentrations (Cleq/m3) at each location
Location Filter floor H2SO4 pump tank area Attack tank area H2SO4 truck station
Mode Fine Coarse Fine Coarse Fine Coarse Fine Coarse
Maximum 4.8 4.2 0.6 3.3 20.7 4.7 -0.1 0.3
Minimum -1.5* -1.2 -1.2 -0.9 -2.1 0.0 -1.2 -0.7
Median 0.2 0.5 0.0 0.0 4.4 1.3 -0.3 -0.2
: A ""sign indicates OH- concentration.




























Figure 2-1. Manufacturing processes at fertilizer facilities

























*Plant

o Background site


SWinter
o
1 **


Florida


Figure 2-2. Geographic locations of sampling sites





FF FC TF TC AF AC SF SC


1000


*o 1Vhnimum

*Maximum

.... Mean
Median


rci
E

v
F1
O
3
k
c,
F1
O
F1
O
O
m
m

E
3
O
3

Pc


800




600




400




200


- a


*


L~ocati on

FF Fe TF Tc AF


AC SF Sc


Figure 2-3. Fine mode and coarse mode aerosol mass concentrations at various locations. The
first letter denotes location: F- filter (rotating table or belt filter) floor, T- truck
loading/unloading station, A- attack tank area and S- sulfuric acid pump tank area.
The subscript denotes particle size: F- fine mode (PM2.5), C- COarse mode (PM2.5-10)










140
0 Minimum
120 Maximum

100 -o ...... Mean
A- Median


60

8 40

S20


F Cl PO43- SO42-Na NH4+ K+ Mg2+ Ca2+
Species

140

120 0 Minimum
CC~ I* Maximum
S100
Bo Mean

60

8 40


u 20* *

F~ Cl~ PO43- SO42-Na+ NH4+ Kf Mg2+ C2+
Species


Figure 2-4. Aerosol chemical species at the sulfuric acid pump tank area. A) For fine mode. B)
For coarse mode












0 Minimum
*Maximum
...... Mean
- Median


A

O
'3


-Q
- -


F~ Cl~ PO43- SO42- a+ NH4+ K+ Mg2+ C2+
Species


0 Minimum
*Maximum
...... Mean
-Median









F- Cl- PO43- SO42- +f NH4+ K+ Mg2+ 2+~
Species


500


400


300


200


100


0




140

120

100

80

60

40

20


B "



8
O1
0


Figure 2-5. Aerosol chemical species at the attack tank area. A) For fine mode. B) For coarse
mode



























F~ Cl~ PO43- SO42- Na NH4+ K+ Mg2+ Ca2+
Species



0 Minimum
Maximum
...... Mean
h A;,


140


120

S100
An


60

S40

u 20


-

-


140

& 120
B 4 100
s

S60

840

u 20


F C1 PO43- SO42-Na NH4+ K+ Mg2 2+"
Species


Figure 2-6. Aerosol chemical species at the rotating table/belt filter floor. A) For fine mode. B)
For coarse mode










140
o 1VImimum
S120
*Maximum
I 100 ....... Mean
-Median
A 0 0
S 60

40




F Cl PO43- SO42- Na NH4+ K+ Mg2+ Ca2+
Species

140
o Minimum
Cc) Maximum
10 to ...... Mean
-L Median
a1 80
60


a 40
u 20


F~ C1f PO43- SO42- +t NH4t Kt Mg2+ 2+"
Species


Figure 2-7. Aerosol chemical species at the sulfuric acid truck loading/unloading station. A) For
fine mode. B) For coarse mode











100


1:1

10

Basic Aerosol




vv o



J1 0.1 ~/V Ac dic Aerosol




0.01
0.01 0.1 1 3 10 100

Anion equivalent weight (peql/m3 air)

Figure 2-8. Relationship of cation equivalent weight and anion equivalent weight

Belt / rotating table filter floor (fine)
o Belt / rotating table filter floor (coarse)
r -ISO4 pump tank area (fine)
v -ISO4 pump tank area (coarse)
Attack tank area (fine)
a Attack tack area (coarse)
+ R~SO4 truck station (fine)
+ R-ISO4 truck station (coarse)









CHAPTER 3
SIZE-RESOLVED SULFURIC ACID MIST CONCENTRATIONS AT PHOSPHATE
FERTILIZER MANUFACTURING FACILITIES INT FLORIDA*

Background

Strong inorganic acid mists containing sulfuric acid have been reported to correlate with

lung and laryngeal cancer in humans [Blair andKazerouni, 1997; Stubliakl\nltinar et al., 1997;

Steenlan2d, 1997] and are identified as a human carcinogen as reported by the NTP [USDHHS,

2005]. Sulfuric acid is typically present in the air in the mist form. Its chemical characteristics

include low volatility, high acidity, high reactivity, high corrosivity and high affinity for water.

Phosphate fertilizer manufacture is listed in the NTP report as one of the industries that has

sulfuric acid mist exposure potential. The OSHA has established an 8 h TWA of PEL of sulfuric

acid mist at 1 mg/m3. It is well known that the deposition of an aerosol in the respiratory system

depends on its aerodynamic behavior. Considering the effects of aerosol size, the ACGIH has

adopted a TLV-TWA of 0.2 mg/m3 for the thoracic particulate fraction of sulfuric acid mist

[ACGIH, 2004].

NIOSH Method 7903 is an OSHA-approved method that is commonly used by the health

and safety staff in industries to measure the total sulfuric acid mist concentration. The sampler

of NIOSH Method 7903 is a silica gel tube consisting of one section of glass fiber filter plug

followed by two sections of silica gel (commercially available: ORBO-53 tube, Supelco, and

SKC silica gel tube, SKC). The glass fiber filter plug is designed to filter out the majority of

acid aerosols while the silica gel sections are used mainly to adsorb acid gases. The collected

samples are desorbed in eluent and the aliquots are analyzed by IC. NIOSH researchers who

developed the method reported ~90% collection efficiency for acidic aerosols with 0.4 Clm

* Reprinted with permission from Hsu, Y.-M., Wu, C.-Y., Lundgren, D. A., Birky, B. K., 2007. Size-
Resolved Sulfuric Acid Mist Concentrations at Phosphate Fertilizer Facilities in Florida. Ann. Occupy.
Hyg. 51, 81-89.










volume median diameter (94.8 f 4.8% for H3PO4 and 86 f 4.6% for H2SO4) when the samples

collected on the glass fiber section and the front silica gel section (400 mg) were combined

[ Cassinelli, 1986; Cassinelli and Taylor, 198 1].

Cascade impactors are commonly used for characterizing aerosol size distribution [Dibb et

al., 2002; Swietlicki et al, 1997]. Large particles are collected on a substrate by inertial

impaction, while small particles can better follow the changes in the flow direction of a curved

air stream. By adopting a series of impactor stages with increasing flow velocities, the aerosol

size distribution can be classified.

The approved NIOSH method only provides total sulfuric acid mist concentration, but not

size-dependent information. The comparison of the total mist concentration with the size-

fractionated measurement by the cascade impactor may provide a convenient tool for correlating

exposure levels based on the simpler NIOSH method. This information can be applied to

develop informed policies with respect to respiratory protection in the workplace.

To properly assess the occupational exposure of workers to sulfuric acid mist in phosphate

fertilizer manufacturing facilities, the obj ectives of this chapter were to determine the total

sulfuric acid mist concentrations using the approved NIOSH method and to characterize the size

distributions of sulfuric acid mist by cascade impactor sampling. All other chemical species will

be reported in next chapter. Furthermore, the feasibility of using a correlation factor between

these two measurements was examined.

Methods

Sampling Sites

The final products of phosphate fertilizer facilities are MAP, DAP, phosphoric acid and

sulfuric acid. The manufacturing processes have been described in Hsu et al. [2007a]. Five

types of locations with the potential of H2SO4 exposure corresponding to the manufacturing










process were selected for sampling. These locations (the number of the sampling sites) include

the sulfuric acid pump tank area (9), rotating table/belt filter floor (9), attack tank area (2), truck

station for loading/unloading sulfuric acid (3), and the granulator on a "scrub" day (1). Sampling

was carried out at eight plants. Seven of them are located in central Florida, and one of them is

located in north Florida. In addition, Gainesville, FL, and Winter Haven, FL, were chosen as the

background sites.

Sampling and Analysis Methods

A University of Washington Source Test cascade impactor (Mark III) was used as an area

sampler to sample mist aerosols for the size distribution. The inlet of the cascade impactor was

set at 1.5 m from the floor. The impactor was operated at 25 Lpm with aerodynamic cut sizes

(dso) of 0.20, 0.48, 0.98, 1.8, 3.8, 10 and 23 Clm for the seven stages, respectively. The impactor

with glass fiber filter can provide high collection efficiency for aerosols; however, glass fiber

filter is well known to adsorb acidic gas, such as sulfur dioxide [Chow, 1995; Lee and2~ukund,

2001]. Therefore, Teflon membrane filters (ZefluorTAI, 8 x 10 inches, pore size: 2 Clm, Pall

Corp., Ann Arbor, MI) that provide high collection efficiency and low reactivity with acidic

gases [Chow, 1995; Lee and2~ukund, 2001] were applied for the collection substrate. Those

filters were cut to fit onto the impactor stages. The collection efficiencies of the cascade

impactor for liquid (substrate: a glass fiber filter) and solid aerosols (substrate: an aluminum foil

with silicone coating) were 97.2 & 11.9 and 94. 1 + 17.3%, respectively [Pauhuhn, 2005]. Droplet

break-up in this instrument is negligible for large droplets (10 and 25 Clm) even when a high

sampling flow rate (28.3 Lpm) is applied [Horton and2~itchell, 1989]. A final Teflon filter

(ZefluorTAI, 47 mm, pore size: 2 Clm, Pall Corp., Ann Arbor, MI) was placed after the impactor to

collect penetrating aerosols. Filters were placed in a desiccator at room temperature for pre- and









post-conditioning for 24 h before weighing to reduce the effect of water collected by the filter.

The vapor pressure of sulfuric acid mist is low and it remains in the mist form under these

conditions. The lower limit of particle size collected on the final filter was assigned to be 0.03

Clm, a typical value of those employed in other research studies [Divita et al., 1996; Howell et

al., 1998; Wagner andLeith, 2001]. The aerodynamic diameters of collected particles were from

0.03 to 23 Clm. PM23 is the aerosol mass concentration with an aerodynamic diameter smaller

than 23 Clm, which was the largest aerosol size collected using this methodology. PM23 SUlfUTic

acid mist concentration was used to compare the total sulfuric acid mist concentration provided

by NIOSH Method 7903.

NIOSH Method 7903 was applied for the sampling of total sulfuric acid mist concentration

using a commercially available silica gel tube (ORBO 53 tube, Supelco). The sampling flow rate

was 0.45 Lpm for 72 sets of samples. Six sets of samples were also collected at 0.3 Lpm in order

to verify whether the results were the same at different flow rates in the recommended range (0.2

- 0.5 Lpm) [Cassinelli, 1986; Cassinelli andEller, 1979; Cassinelli and Taylor, 1980; NIOSH,

1994].

Gravimetric measurement for sample mass was carried out using a microbalance (model

MC 210 S, Sartorius Corp., Edgewood, NY; readability 10 Gig), and the analysis of sulfate

concentration was accomplished by using an IC system (Dionex ICS 1500, Dionex Corp.,

Atlanta, GA). The analytical columns of anion species [nitrate (NO3 ), Sulfate (SO42-), flUOride

(F-), phosphate (PO43-), and chloride (Cl-)] and cation species [potassium (K ), calcium (Ca2+)

magnesium (Mg2+), Sodium (Na ), and ammonium (NH4 )1 are I0nPac AS9-HC (Dionex Corp.,

Atlanta, GA) and lonPac CS12A (Dionex Corp., Atlanta, GA), respectively. The detection limit

for sulfate was determined to be 0.12 ppm for this system.










The sampling time was 24 h and three successive samples were obtained for each sampler

at each site. Totally, there were 72 sets of impactor samples and 78 silica gel tube samples in

those plants. In background locations, there were six sets of impactor samples and six silica gel

tube samples.

Calculation of Fine Mode

Because 2.5 Clm was not one of the cascade impactor cut-sizes, PM2.5 WAS determined by

interpolating the size bin that covers 2.5 Clm (i.e., 1.81 and 3.76 Clm). Assuming a uniform

distribution in this size range in log-scale, PM2.5 can be obtained according to the following

relationship:

log(2.5)- log(1.81) PM25 PM,, (-1
log(3.76)- log(1.81) PM376 PM,,,

Rearranging the formula, PM2.5 can be derived as:

PM25 = 0.44x[ PM376 PM,,, ] + PM,,, (3 -2)

Calculation of Sulfuric Acid Mist Concentration

According to NIOSH Method 7903, sulfuric acid mist concentration is converted from the

sulfate concentration determined by IC. Although the sulfate may not necessarily come from

sulfuric acid (i.e., it can be ammonium sulfate or calcium sulfate), any sulfate determined by this

method is conservatively assumed to be sulfuric acid. In this study, the same protocol was

followed for all samples from the fertilizer plants.

Results and Discussion

Background Site

In general, the mass concentrations and sulfate concentrations were low at both

background locations. The PM23, PMlo and PM2.5 were 20.6-53.2, 18.7-36.9 and 15.1-26.4

Clg/m3 in Gainesville, and 19.1-27.0, 16.3-27.0 and 11.5-22.8 Clg/m3 in Winter Haven. The









corresponding sulfate concentrations were 5.4-8.6 (PM23), 5.2-7.1 (PMlo) and 5.0-6.3 (PM2.5)

Clg/m3 in Gainesville, and they were lower in Winter Haven: 3.0-3.2 (PM23), 2.7-2.8 (PMlo) and

2.4-2.5 (PM2.5) Clg/m3. The ratios of sulfate concentration to the sum of all ionic species

concentrations (NO3-, SO42-, F-, PO43-, C-, K Ca2+, Mg2+, Na and NH4 ) were 0.44-0.46 in

Gainesville and 0. 16-0.39 in Winter Haven.

For NIOSH method samples, the total sulfate concentrations ranged from 6.81 to 10.5

Clg/m3 in Gainesville and much higher in Winter Haven, 31.2-46.0 Clg/m3. COmpared to the

cascade impactor results, the measurements were closer in Gainesville than those in Winter

Haven. The ratio of sulfate from the impactor to sulfate from the NIOSH method sampler

ranged from 0.67 to 0.82 in Gainesville and from 0.069 to 0.096 in Winter Haven. It should also

be noted that while the impactor results showed higher sulfate concentrations in Gainesville than

in Winter Haven, the NIOSH method measurements showed the opposite. It is suspected that the

NIOSH Method 7903 might have interference from SO2 gaS. This will be further discussed in

later sections.

Plants: Cascade Impactor Samples

The sampling results of the cascade impactor at all locations are shown in Figure 3-1A.

The highest median sulfuric acid mist concentration was observed at the sulfuric acid pump tank

areas where two sulfuric acid mist concentrations from the cascade impactor were higher than

the OSHA standard, 1 mg/m3. The size information at each type of location will be discussed as

follows.

Attack tank area

Sampling for the attack tank area was carried out at two plants. The aerosol and sulfuric

acid PM23, PMlo and PM2.5 maSs concentrations are listed in Table 3-1. Aerosols at the attack









tank areas had high mass loadings but low sulfuric acid concentrations. The violent reaction in

the attack tank releases heat and causes a significant amount of volatile species to evaporate.

These species, such as fluoride gases, condense on existing aerosols when they encounter cooler

ambient air, thus resulting in high aerosol mass concentrations. However, the temperature in the

process was not high enough for the evaporation of sulfuric acid or sulfate salt that has lower

vapor pressure. Hence, sulfuric acid mist concentrations were low at this location.

Sulfuric acid pump tank area

Sampling at the sulfuric acid pump tank area was carried out at all eight plants. The PM23,

PMlo and PM2.5 maSs concentrations are listed in Table 3-2. The geometric mean PM23, PMlo

and PM2.5 Sulfuric acid concentrations (+geometric standard deviation) were 41.7 (15.5), 37.9

(5.8), and 22.1 (14.5) Clg/m3. The highest geometric mean sulfuric acid concentration from

cascade impactor measurement among all types of locations was indeed observed at the pump

tank area. The large geometric standard deviation implies that the sulfuric acid concentrations at

these nine sulfuric acid pump tank areas differed greatly. The geometric mean ratios of sulfate

concentration to all ionic species were greater than 0.50, which indicated that sulfate was the

predominant ion accounting for the aerosol mass at this type of location.

The aerosol mass size distributions and sulfuric acid size distributions are shown in Figure

3-2. They are plotted in two ranges: larger than and smaller than 100 Clg/m3. The size

distribution maintained the same pattern at a given site, but not from site to site. It should also

be noted that most of the sulfuric acid size distributions resemble the aerosol mass size

distributions at the same site. At plants B l, D, H and B2, Figure 3-2A, both the aerosol mass

concentrations and sulfuric acid mass concentrations were higher than 100 Clg/m3, and the

aerosols were predominantly supermicron particles. The sulfate/aerosol mass ratios were greater

than 0.5 (Table 3-3). The high ratios indicate that sulfuric acid was the maj or species and that









these locations might have sulfuric acid emission sources. At the pump tank area, the possible

sulfuric acid emission source is the leakage of SO3 that will quickly combine with water

molecules to form H2SO4.

At other plants, Figure 3-2B, the sulfuric acid concentrations were lower than 100 Clg/m3

and their sulfuric acid aerosols were mainly in the submicron range or presented no specific

pattern. In the case of a very low aerosol mass loading, the sulfuric acid aerosols could very

likely be affected by the ambient aerosols at this outdoor location. The geometric mean sulfuric

acid concentration at Plant F was 6.8 Clg/m3 (Table 3-3), which is as low as that at the

background sites.

Belt or rotating table filter floor

Sampling at the belt or rotating table filter floor area was carried out at all eight plants.

The aerosol mass concentrations were high: the PM23, PMlo and PM2.5 maSs concentrations

ranged from 57.4 to 2535, 54.0 to 1857 and 28.3 to 358 Clg/m3, TOSpectively; their geometric

mean concentrations (+geometric standard deviation) were 225.3 (12.3), 187.0 (12.2) and 94.7

(11.8) Clg/m3, TOSpectively. PM23, PMlo and PM2.5 Sulfuric acid concentrations were 7.1-575,

4.9-419 and 2.4-60.0 Clg/m3; the geometric mean sulfuric acid concentrations (+geometric

standard deviation) of PM23, PMlo and PM2.5 were 17.9 (12.7), 14.4 (12.7) and 6.6 (12.1) Clg/m3.

The ratios of sulfate to all ions ranged from 0.07 to 0.32 (geometric mean: 0.16). The low

ratios indicate that sulfuric acid was not a maj or compound at the filter floor area. The fractional

size distributions of aerosol mass and sulfuric acid at nine belt/rotating table filter floors are

shown in Figure 3-2C. Sulfuric acid was dominantly present in the coarse mode. During this

process, gypsum is filtered out by belt or rotating table filter, and the aerosols are formed by









mechanical agitation. The similarity in sulfuric acid and aerosol mass size distributions indicates

that the chemical might be from the residual content in the product.

Sulfuric acid truck loading/unloading station

Sampling at the truck loading/unloading station was conducted at three plants. The aerosol

mass concentrations were low and their PM23, PMlo and PM2.5 ranged from 19.9 to 174, 15.8 to

131, and 10.0 to 69.7 Clg/m3. Sulfuric acid was the maj or species (the ratios of sulfate to total ion

mass: 0.28-0.42, median: 0.36); the concentrations ranged from 3.9 to 24.5 (PM23), 3.5 to 23.9

(PMlo) and 3.1 to 20.6 (PM2.5) Clg/m3, which were close to the measurements at the background

locations. All size distributions of sulfuric acid were similar: the mode size was 0.48-0.98 Clm.

During loading and unloading, sulfuric acid is transferred from the storage tank in liquid form to

the truck. The only possible pathway that workers can be exposed to sulfuric acid is the spray of

sulfuric acid from the truck loading nozzle. Normally, the connection is well sealed and the

workers are well protected to avoid contact with liquid sulfuric acid. The measurements verified

that the sulfuric acid concentrations were very low.

Granulator on a scrub day

The mass concentrations ranged from 126 to 835 (PM23), 90.2 to 578 (PMlo) and 59.7 to

303 (PM2.5) Clg/m3. The sulfuric acid concentration ranges were 7.63-87.9 (PM23), 5.73-59.7

(PMlo) and 3.85-51.6 (PM2.5) Clg/m3. The source of sulfuric acid mists is the spray from the

scrubbing process (a weak acid solution) which is not a steady operation. Hence, the

concentrations varied significantly, but they were all below the standards.

Plants: NIOSH Method Samples

Sampling results of the NIOSH samples at five types of locations are shown in Figure 3-

IB. Generally, the sulfuric acid pump tank area had the higher concentrations. However, the









maximum sulfuric acid concentration (11225 Clg/m3) meaSured by the NIOSH method for all

locations was obtained at the filter floor area. The largest geometric mean sulfuric acid

concentration was obtained at the pump tank area (143.2 Clg/m3), followed by the granulator on a

scrub day (122.4 Clg/m3) and then at the filter floor (108.7 Clg/m3). The geometric mean

concentrations at the granulator on a scrub day and sulfuric acid truck loading/unloading stations

were at lower levels.

Comparisons of the Results from the Cascade Impactor and the NIOSH Method

The paired measurements of PM23 Sulfuric acid concentrations from the cascade impactor

and total sulfuric acid concentrations from the NIOSH method at all sampling locations are

shown in Figure 3-3. As shown, 71 out of 72 impactor samples had a lower concentration than

the NIOSH method measurement. The ratios of sulfuric acid mist concentrations from the

NIOSH to the cascade impactor were 1.5-229.0 for 71 impactor samples. The largest difference

was over two orders of magnitude, and many of the NIOSH measurements were more than 10

times larger the impactor results. The substantial difference was quite unexpected. To

quantitatively compare these two measurements and evaluate their correlation, three ratios were

used which are defined as Equation 3-3 to Equation 3-5:


R23 23 (3-3)
C,

PM,
Rio = 0(3-4)


PM,
= 2 (3-5)
CN

where PM23, PMlo and PM2.5 are Sulfuric acid concentrations from the cascade impactor and CN

the sulfuric acid concentration by NIOSH Method 7903.









Table 3-4 displays the R23, R1o and R2.5 at five types of sampling locations. A large

variation at each type of location is observed, e.g., from 0.00 to 0.67 at the filter floor areas for

R23. A strong relationship between methods would be indicated by a constant correlation factor;

the wide variation of the ratios does not allow any meaningful correlation of these two types of

measurements. The much higher values by the NIOSH method also prompted further

investigation of the data.

In examining the data, it was found that in many cases the silica gel sections collected

more sulfuric acid than the glass fiber section (see examples in Table 3-5). The results of the

NIOSH method imply that sulfuric acid as well as sulfate is partially measured as a gas. In the

ambient condition, there is no known sulfate species (inorganic) that exists in the gas phase.

Even sulfuric acid exists in the condensed phase because it has a high boiling point of 330 oC and

a very low vapor pressure at room temperature (< 0.001 mmHg) [Wea~st, 1988]. Furthermore,

the hygroscopic sulfuric acid will rapidly pick up moisture in the air and remain in the aerosol

phase. It should be noted again that according to Ortiz and Fairchild [1976] the majority of the

aerosol mass is collected on the glass fiber filter plug. Chen et al. [2002] reported that aerosol

penetration across the filter section of the silica gel tube (SKC 226-10-03 tube) at the most

penetrating size was lower than 5%. The collection efficiency of large particles (> 3 Clm) is

higher than 99%. Sulfuric acid mists mainly exist as coarse aerosols at the pump tank area

(Figure 3-2A); hence, aerosol penetration cannot explain the high sulfuric acid concentration

sampled by silica gel. Thus, the situation that more sulfuric acid mist is collected in the silica gel

section than the glass fiber section is highly unlikely. The adsorption of a significant amount of

interfering gases by the silica gel is therefore suspected to be the reason for the observed

phenomenon.









Another unexpected phenomenon was observed when comparing the results obtained from

the NIOSH method at different sampling flow rates. In the recommended range of sampling

flow rate (0.2-0.5 Lpm), concentrations should be similar when different flow rates are used.

Two different flow rates, 0.3 and 0.45 Lpm, were employed for six sets of the NIOSH method

sampling, and the results are shown in Table 3-5. These paired samples were taken concurrently

(e.g., #1-low and #1-high were taken at the same time), and three consecutive samples were

carried out (i.e., #1 followed by #2 and then by #3). As shown, sulfuric acid concentrations at

the lower flow rate (0.3 Lpm) were higher than those at the higher flow rate (0.45 Lpm). Most of

the ratios (C@ 0.3 Lpm/ @ 0.45 Lpm) Were larger than 1, and they were much higher in the silica gel

section than those in the glass tube section. The concentrations at two different flow rates do not

exhibit any direct proportion between the gas phase and the particulate phase. Hence, systemic

errors can be neglected. The collection mechanism of acid gases on the silica gel is diffusion,

and its efficiency decreases as the flow rate increases (due to shorter residence time). The

observations support the hypothesis that the higher measurement in the silica gel section is

probably from the collection of gas.

Silica gel, a high surface area material, can adsorb SO2 [Kopac andKocaba~s, 2002;

Stratmann andBuck, 1965]. The hydrophilic property of silica gel can effectively attract

moisture which can enhance the absorption of soluble species, such as SO2 [Tsai et al., 2001].

The adsorbed or absorbed SO2 can be further oxidized to form sulfate [Lunsford, 1979] that

causes overestimation.

Annual SO2 COncentrations (2003) were monitored by the state of Florida [FDEP, 2003]

and the results indicated that central FL had higher SO2 COncentrations (~2-6 ppb) than north FL

(~2-3 ppb) If the above hypothesis is true, this might explain why sulfate concentrations in









Winter Haven measured by the NIOSH method were much higher than those in Gainesville but

the impactor results did not exhibit such a pattern.

Limited SO2 COncentrations in fertilizer facilities are available in the literature. The SO2

concentration in a fertilizer factory in India was 41.7 mg/m3 [YadyYYYYYYYYYYYYYYYYYY andKaushik, 1996] while

those in China and Sweden were 0.34-11.97 and 3.6 mg/m3, TOSpectively [Englander et al.,

1988; M~eng and Zhang, 1990]. Atmospheric dispersion can quickly reduce the concentration to

a much lower level as reported in a study near a fertilizer factory in Zimbabwe [Jonnalagadda et

al., 1991]. If the hypothesis holds true, the sulfuric acid concentration in the fertilizer facilities

can be expected to be overestimated when SO2 is present in the studied facilities. Further

investigation of this subj ect is warranted.

Comparisons of Sulfuric Acid Mist Concentrations with OSHA and ACGIH Regulations

The number of samples with concentrations higher than the OSHA regulation (total

sulfuric acid mist concentration < 1 mg/m3) was 7 of the 78 samples collected. According to the

cascade impactor sampling, 90% of the samples were lower than the ACGIH standard and 97%

of the samples were lower than the OSHA regulation. The results of the NIOSH method samples

obtained concurrently with the impactor samples had a smaller percentage of samples with

concentrations lower than the OSHA regulation. The only location where the sulfuric acid mist

concentrations from the cascade impactor exceeded both the OSHA and ACGIH standards was

the sulfuric acid pump tank area. For the NIOSH method samples, the locations included

sulfuric acid pump tank areas, belt/rotating table filter floors and the granulator on a scrub day.

Summary

In this study, the total sulfuric acid mist concentration and sulfuric acid mist size

distributions at modern phosphate fertilizer manufacturing facilities in Florida were characterized









by using NIOSH Method 7903 and a cascade impactor, respectively. Sampling was carried out

at five types of locations in eight facilities and two background sites.

Based on cascade impactor sampling, sulfuric acid pump tank areas had higher sulfuric

acid mist concentrations than other types of locations and sulfuric acid was the dominant

chemical species. When high sulfuric acid concentrations were identified, the aerosols were

dominantly in the coarse mode. The most likely cause of high sulfuric acid concentrations at this

location is the leakage of SO3. COnstant inspection of the tubing around this location and

immediate repair may provide an effective measure to further lower the concentrations.

According to the impactor sampling results, seven samples (total: 72) exceeded the ACGIH

standard (0.2 mg/m3, thoracic fraction), and two samples (total: 72) were above the OSHA

regulation (1 mg/m3, total concentration). Meanwhile, there were seven samples (total: 78) by

the NIOSH method that exceeded the OSHA regulation. It should be noted at these locations,

workers typically stay for about 1-2 h per day and respirators for sulfuric acid mist are required

in this area. Consequently, the real time-weighted exposure to sulfuric acid mist is likely to be

lower than the concentrations from the stationary sampling conducted in this study.

The results from the cascade impactor and the NIOSH method were compared to

determine if a correlation factor could be established. The sulfuric acid mist concentrations from

the NIOSH method were higher than those from the cascade impactor for the dominating

maj ority of samples. No specific trend of systemic error was observed between these two

methods. In many cases, the silica gel section collected more "sulfuric acid" than the glass fiber

filter plug. This is highly unlikely because sulfuric acid or sulfates are not known to exist in the

gas form in the ambient condition, and should not be collected in the silica gel section.

Moreover, the sulfuric acid concentrations at 0.30 Lpm were higher than the concentrations at









0.45 Lpm in the NIOSH method sampling, indicating the influence of diffusing gases. The

possible reason for the variation is the interaction between SO2 and silica gel/glass fiber filter.

Further investigation has been verified the causes and is discussed further in Chapter 5.









Table 3-1. PM23, PMlo and PM2.5 maSs and sulfuric acid concentrations at the attack tank areas
Sample Geometric Geometric standard
pg/m3 number min max mean deviation
PM23
Malssa 6 86.2 1853 341.3 3.9
Mass (IC)b 6 62.7 414.6 152.8 2.3
Sulfuric acid 6 6.7 19.0 10.1 1.5
PMio
Malssa 6 76.1 1767 308.3 4.1
Mass (IC)b 6 59.4 397 139.6 2.3
Sulfuric acid 6 6.3 12.5 8.7 1.3
PM2.5
Malssa 6 64.6 720 187.0 2.9
Mass (IC)b 6 41.6 182 90.1 1.8
Sulfuric acid 6 2.3 7.6 5.5 1.6
Massa: mass concentration from weighing the mass.
Mass (IC)b: the sum of all ionic species concentrations measured by IC.
Sulfuric acid: all sulfate is conservatively assumed to be sulfuric acid.









Table 3-2. PM23, PMlo and PM2.5 maSs and sulfuric acid concentrations at the sulfuric acid pump
tank areas
Sample Geometric Geometric standard
Clg/m3 number min max mean deviation
PM23
Massa 27 20.4 1644 121 3.3
Mass (IC)b 27 9.4 1268 69.5 3.8
Sulfuric acid 27 2.1 1187 41.7 5.5
Sulfate/ Mass (IC) 27 0.22 0.98 0.588 1.52
PMlo
Massa 27 14.7 1625 105 3.5
Mass (IC)b 27 8.7 1221 61.9 4.0
Sulfuric acid 27 2.0 1155 37.9 5.8
Sulfate/ Mass (IC) 27 0.23 0.98 0.600 1.52
PM2.5
Massa 27 12.1 648 56.4 2.8
Mass (IC)b 27 6.5 548 35.0 3.2
Sulfuric acid 27 1.8 558 22.1 4.5
Sulfate/ Mass (IC) 27 0.25 1.00 0.619 1.47
Massa: mass concentration from weighing the mass.
Mass (IC)b: the sum of whole ionic species concentrations measured by IC.
Sulfuric acid: all sulfate is conservatively assumed to be sulfuric acid.










Table 3-3. Mass, sulfuric acid concentrations and sulfate/massa ratios of the impactor samples at the sulfuric acid pump tank areas
Plant Mass (yg/m` ) Sulfuric acid (pg/m-) Sulfate/ Massa Sample size
Geometric Geometric standard Geometric Geometric standard
mean deviation mean deviation Mean
D 905.9 1.8 643.3 1.8 0.96 3
B1 550.8 3.0 298.3 3.8 0.84 3
H 177.9 2.1 114.0 2.6 0.88 3
B2 95.7 1.9 37.7 2.0 0.57 3
E 64.9 1.7 33.1 2.0 0.71 3
C 76.4 1.3 11.8 1.8 0.41 3
A 71.5 1.9 21.4 1.5 0.46 3
G 50.4 1.4 8.0 1.5 0.45 3
F 37.4 2.5 6.8 4.5 0.42 3

Massa: mass concentrations combining all ionic species concentrations










Table 3-4. Statistics of R23, Rio and R2.5 at five tvoes of sampling location


Sample
number
27
27
27
27
27
27
9
9
9
6
6
6
3
3
3


Geometric
mean
0.15
0.26
0.30
0.06
0.13
0.17
0.10
0.15
0.18
0.06
0.09
0.11
0.13
0.16
0.21


Geometric
standard deviation
2.21
2.22
2.20
2.81
2.52
2.55
1.93
1.81
1.81
1.93
2.08
2.07
3.88
4.25
3.87


Min
0.04
0.06
0.06
0.00
0.00
0.00
0.05
0.08
0.10
0.02
0.02
0.03
0.03
0.03
0.04


Max
0.58
2.65
2.74
0.41
0.64
0.67
0.36
0.42
0.44
0.10
0.22
0.23
0.28
0.40
0.54


PM2. 5
PMio
PM23
PM2. 5
PMio
PM23
PM2. 5
PMio
PM23
PM2. 5
PMio
PM23
PM2. 5
PMlo
PM23


Granulator
on a scrub
day


Sulfuric acid
pump tank
area
Belt/rotating
table filter
floor

Truck station

Attack tank
area









Table 3-5. Sulfuric acid concentrations and the ratios measured at two flow rates at the rotating table filter floors using NIOSH
Method 7903
[H2SO4 a (y8/m3)
Low flow rate High flow rate
Sampler (@ 0.3 Lpm) (@ 0.45 Lpm) C@ 0.3 Lpm C@ 0.45 Lpm
section #1 #2 #3 #1 #2 #3 #1 #2 #3
Total b90.3 141.0 118.8 80.8 87.2 89.8 1.12 1.62 1.32
Glass fiber 36.7 44.3 44.3 35.7 37.7 46.7 1.03 1.17 0.95
Silica gel 53.6 96.7 74.5 45.1 49.5 43.2 1.19 1.95 1.73
[H2SO4 a: paired samples at two flow rates taken concurrently.
Totalb: the sum of the results from the glass fiber section and the silica gel section.































OSHA standard







o O


2x104
104


h
M
E


d
o
vl
hi


102


101


2x104
104


AT PT FF
Location


TS SD


Figure 3-1. Sulfuric acid concentrations at 5 types of locations. Locations: AT- attack tank, PT-
sulfuric acid pump tank, FF- filter (rotating table or belt filter) floor, and TS- truck
station for sulfuric acid loading/unloading. Sampling methods: A) PM23 Sulfuric acid
by cascade impactor, and B) Total sulfuric acid by NIOSH Method 7903


OSHA standard





.3


Max
90%
75%
Median
25%
O10%
Min










1.0



A F
0.6

'a 0.4 --

F40.2

0.0

1.2

1.0 -75%
B Median

S0.6 10%

E 0.4 I 1

0.2 t = T.~

0.0

1.0
SSulfuric Acid
S0.8 ******* Sulfuric Acid Median
SMass
C 06-Mass Median
O0.
c 0.4

F4 0.2 ... ..

0.0
0.01 0.1 1 10 100
Aerodynamic diameter (I-m)


Figure 3-2. Sulfuric acid mist and aerosol mass size distributions. A) Sulfuric acid pump tank
areas, high aerosol mass loading (> 100 Clg/m3). B) Sulfuric acid pump tank areas,
low aerosol mass loading (< 100 Clg/m3). C) Belt/rotating table filter floors










10'


SSulfuric acid pump tank
*Belt/rotating filter floor:
STruck station
+ Scrub day
SAttack tank
1:10

*a

1:100

o *


O


o
-

0


104



103



102



10



100


100


1000


10000 100000


H2SO4 COncentration (NIOSH) (pg/m3)


Figure 3-3. Comparison of PM23 Sulfuric acid concentrations from the cascade impactor and total
sulfuric acid concentrations from the NIOSH method









CHAPTER 4
SIZE DISTRIBUTION, CHEMICAL COMPOSITION AND ACIDITY OF MIST AEROSOLS
INT FERTILIZER MANUFACTURING FACILITIES INT FLORIDA*

Background

Strong inorganic acid mists containing sulfuric acid have been reported to correlate well

with lung and laryngeal cancers in humans [Blair andKazerouni, 1997; Stubliakl\nltinar et al.,

1997; Steenlan2d, 1997]. Phosphate fertilizer manufacturing facilities, where sulfuric acid is used

to digest phosphate rock to form H3PO4, have been listed by the NTP [USDHHS, 2005] as one of

many occupational exposures to strong acid. The current OSHA 8 h TWA of PEL of H2SO4 and

H3PO4 miSts is set at 1 mg/m3. The carcinogenic mechanism of strong inorganic acid mists

containing sulfuric acid is not clearly understood yet [Blair andKazerouni, 1997]. However,

reduced pH environments are known to enhance the depurination rate of DNA and the

deamination rate of cytidine [Swenberg and Beauchamp, 1997; USDHHS, 2005], which can

cause DNA damage or mutation. Furthermore, studies have found that adverse pulmonary health

effects are related to the hydrogen ion (H ) rather than sulfate [Ostro et al., 1991; Schlesinger,

1984, 1989; Schlesinger et al., 1990a; Schlesinger et al., 1990b] and that the biological response

is a function of the total concentration of H' deliverable to the cells or the total extractable H

per particle [Schlesinger and Chen, 1994].

The aerosol acidity is an important factor that affects the amount of the extractable H i.e.,

strong acid can release more H In phosphate fertilizer facilities, H3PO4 and H2SO4 miSts are

the major acid species [Hsu et al., 2007a]. These two strong acids can contribute a significant

amount of the extractable H Currently, no method is available to directly measure the aerosol

acidity; nevertheless, several ambient aerosol thermodynamic models have been developed to

* Reprinted with permission from Hsu, Y.-M., Wu, C.-Y., Lundgren, D. A., Birky, B. K., 2007. Size
Distribution, Chemical Composition and Acidity of Mist Aerosols in Fertilizer manufacturing Facilities in
Florida. J. Aerosol Sci. doi: 10.1016/j .jaerosci.2007. 10.008.









estimate the aerosol acidity [Yao et al., 2006, 2007]. However, particulate phosphate species

have not been included in those aerosol models because the concentrations of those species are

too low to be detected in the ambient aerosols.

The hazard of aerosol depends on the chemical compositions and the sites where the

aerosol deposits. It is well known that the deposition of an aerosol in the respiratory system is

related to its aerodynamic behavior. Meanwhile, relative humidity and chemical characteristics

of the aerosol may affect its size. Sulfuric acid is hydrophilic, and the size of the mist can

increase significantly when it encounters moisture [SeinfeldandP andisPPP~~PPP~~ 1998; Wu andBiswa~s,

1998] in the respiratory system.

In considering the effects of aerosol size on human health, the ACGIH has adopted a TLV-

TWA of 0.2 mg/m3 for the thoracic particulate fraction of sulfuric acid mist [ACGIH, 2004].

Although the acid mist size distribution is an important parameter in evaluating health effects, in

the past very little data was available for the particle size distribution of phosphoric acid mists at

phosphate fertilizer facilities. The obj ectives of this chapter are to characterize the chemical

characteristics of mist aerosols in the current phosphate facilities with size-resolved information

and to establish an aerosol thermodynamic model to estimate aerosol acidity.

Methods

Sampling Sites

Five types of locations which might have high acid mist concentrations were selected,

including H2SO4 pump tank areas, product filter floors, attack tank areas, truck stations for

loading/unloading H2SO4, and a granulator on a scrub day. Sampling was conducted at seven

plants in central FL and one plant in north FL. Totally, there were 24 sampling locations at eight

fertilizer plants. The type and number of sampling locations at each plant have been described in

detail in Hsu et al. [2007a] and Chapter 2 of this dissertation.









Sampling and Analysis Methods

A University of Washington Source Test cascade impactor (Mark III) was used to collect

mists for the size distribution. The impactor was operated at 25 Lpm and the corresponding dso

was 0.20, 0.48, 0.98, 1.8, 3.8, 10, and 23 Clm for the seven stages, respectively. Aerosol bounce

from the impaction plates can result in mass loss and thus distort the sample's true size

distribution. However, a previous study [Pauluhn, 2005] showed that the mean mass recoveries

from cascade impactor analyses were 97.2 & 11.9% and 94. 1 + 17.3% for liquid and solid

aerosols, respectively; therefore, aerosol bounce problems of the cascade impactor can be

neglected.

The collection substrate used in this study was ZefluorTM, Or backed Teflon membrane

filter (pore size: 2 Clm, Pall Corp.), which provides high collection efficiency and low reactivity

with acidic gases [Chow, 1995; Lee and2\~ukund, 2001]. A final filter (ZefluorTM,47 mm, pore

size: 2 Clm, Pall Corp.) was placed after the impactors to collect any penetrating aerosols. The

lower limit of particles collected on the final filter was assigned to be 0.03 Clm, a value

commonly used in similar cascade impactor sampling studies [Divita et al., 1996; Howell et al.,

1998; Wagner andLeith, 2001]. The sampling time was 24 h and three successive samples were

obtained at each location. Totally, there were 72 sets of impactor samples in those plants. A

microbalance (model MC 210 S, Sartorius Corp.; readability: 10 Gig) was used for weighing the

aerosol mass. In order to reduce the interference of water vapor, all filters were placed in a

desiccator at room temperature for pre- and post-conditioning for 24 h before weighing the

filters. After weighing the filter, water soluble species were extracted from the filter using

deionized (DI) water in an ultrasonic bath (Bath ultrasonic timer, Fisher scientific) for 1 h. A

Nanopure Diamond system (Barnstead International) provided the DI water with a conductivity









of 18.2 megohm. The water soluble ion species concentrations were analyzed by an IC system

(Dionex ICS 1500, Dionex Corp.). Two analytical columns, lonPac CS12A (Dionex Corp.) and

lonPac AS9-HC (Dionex Corp.), were applied to analyze cations, including K Na Ca2+, Mg2+,

and NH4 and anions, including F-, Cl-, NO2-, NO3-, SO42-, and PO43-, TOSpectively.

Aerosol Thermodynamic Model

A thermodynamic model of multicomponent aerosols was developed to estimate the

acidity of aerosols with high sulfuric acid (200 Clg/m3) Or high phosphoric acid (500 Clg/m3)

concentration. The important parameters of the model chemical components, the aerosol water

amount (AWA), equilibrium reactions, reaction constants, and the activity coefficient

calculations--are described in this section.

Five chemical species larger than 1% (mass) based on the chemical analysis in this study

were selected: Ca2+, N4 Na SO42-, and PO43-. Liquid components include H Ca2+, N3,

NH4 Na SO42-, HSO4-, H2SO4, H2PO4-, H3PO4, H5P208-, H20, Na3PO4, and (N\H4)3PO4, Solid

components include Ca3(PO4)2, CaHPO4, N4H2PO4, (NH4)3PO4, Nal3PO4, NaH2PO4, Na2HPO4,

CaSO4.2H20, NH4HSO4, (N4)2SO4, NaHSO4, and Na2SO4. AWA (Cg/m3 air) was calculated

by Equation 4-1:

AWA = AMC IMC (4-1)

where AMC is the total aerosol mass concentration (Clg/m3 air) and IMC the total ionic species

mass concentration (Clg/m3 air).

Chemical equilibrium reactions and equilibrium constants are listed in Table 4-1 for 20

reactions. The equilibrium constant is expressed as a function of temperature. It should be noted

that the sampling was conducted for 24 h and temperature varied during the sampling.










Therefore, the rate constants under the standard condition, 298.15 K, were applied in the

modeling work.

The activity of ionic species in aqueous phase ( a i) is defined as Equation 4-2:

a, = ( x mi (4-2)

where ri is the activity coefficient of species i and nai is the molal concentration of species i (mole

solute/kg solvent).

Multicomponent activity coefficients were calculated following the Kusik and Meissner

method [Kint et al., 1993], which is based on binary activity coefficients:


z, zz+ z Z
logr,, = x n, logr +n?4 z,0grl 4
zz+ ,2 2


+ x na: log + ng 2 lo r3 +...,:I~ (4-3)
(zz, + z?)I 2 2

where rl2 is the activity coefficient of a 1-2 ion pair in the multicomponent solution, zi is the

absolute number of unit charges of ion species i, and l is the ionic strength of the solution, which


1S

I 2nz z, (4-4)


r~p is the mean ionic activity coefficient of the single solute solution with i-j ion pair.

To calculate multicomponent activity coefficients, binary activity coefficients need to be

available. They were calculated by a polynomial regression method [Ba~ssett and Seinfeld, 1983]

for the ionic strength smaller than 30 mole/kg and the Kusik and Meissner method [Kint et al.,

1993] for that larger than 30 mole/kg. The polynomial regression method is shown in Equation

4-5.









All/2
In ro 1+ /2 + 2 2l (4-5)


where A, B, and a, are parameter values which are available in Bassett and Seinfeld [1983].

The Kusik and Meissner method is shown in Equation 4-6 and q values used in this model

are listed in Kim et al. [1993].


T~z !i~ ~~r~l ~:0.510771'2+1/2 46


where B = 0.75 0.065 x q and C = 1 + 0.055 x q x exp(-0.023 x IS)

Currents ambient aerosol thermodynamic models have not included phosphate species,

including H3PO4, H2PO4- and HSP20s-. However, phosphate species should be considered in

this study due to high particulate phosphate concentrations and the activities were calculated by

Pitzer's equation shown in Equation 4-7 to Equation 4-9 [Jiang, 1996].


Inr, =z2 r+ m B m e20,+f fc

C a'


C8, +C m0i, 1




+C memj2 'iF1 + IZ2l, ]ca c c' cc', 2 (4-8)



In rl 1 2 2(4-9)
v, + v2



rbrr /.=1Y 31+1.21'21/2 1.22I~lI '










B = pi"' + x1-+2 ex-2


B x -1+ 1+211'+ +2)exp( -27i~







BI


17, and v, are the number of moles of ion species 1 and 2 per mole of 1-2 ion pair species

dissociating completely. The parameters A "), A'), C~ and ri/ are listed in Jiang [1996] and 86

can be neglected for most cases [Kim et al., 1993].

Results and Discussion

Aerosol Chemical Species

Sulfuric acid mists, phosphoric acid mists, and particulate fluoride were the maj or species

observed in the phosphate fertilizer plants. Their concentrations at all locations are shown in

Figure 4-1. The highest mass concentrations of these species were 1163, 1589 and 388 Clg/m3,

and they were obtained at the sulfuric acid pump tank area, the belt/rotating table filter floor, and

the attack tank area, respectively. Major chemical species concentrations and their size

information at each location are discussed in the following sections. All ion concentrations at the

truck station were very low (max: 64 Clg/m3) and are therefore not discussed further.

Sulfuric aid pump tank areas

The size distribution of sulfuric acid in this area has been reported previously [Hsu et al.,

2007b], and only the major results are summarized here. The median mass concentrations

standardd deviation) were 98 (f430), 71 (f424) and 52 (f149) Clg/m3 for PM23, PMlo and PM2.5










(aerosol aerodynamic diameter smaller than 23, 10 and 2.5 Clm), respectively. The major mode

was 3.8-10 Clm when high sulfuric acid mist concentration was observed. Sulfuric acid was the

dominant species in this area, and its size distribution was similar to the aerosol mass size

distribution at this location.

For other species at the sulfuric acid pump tank areas, ammonium had the highest

concentration, second only to sulfuric acid, followed by calcium. The sulfuric acid pump tank

area is an outdoor location where the emitted aerosols can mix with the ambient air. The

relations between sulfate and two maj or cations, calcium and ammonium, are shown in Figure 4-

2. In the case of sulfate and ammonium, the ratios of sulfate to ammonium were close to 1 for

small particles ( < 1.8 Clm, including the impactor stages of 0.03-0.20, 0.20-0.48, 0.48-1.0, 1.0-

1.8 Clm) when sulfate concentrations were low (< 50 neq/m3). Regarding the relation between

calcium and sulfate, it was the large particles ( > 1.81 Clm) that had calcium/sulfate ratios close to

1 when sulfate concentrations were low ( < 50 neq/m3). The relation between sulfate and both

basic species is shown in Figure 4-2C. It is interesting to note that both size ranges had a similar

trend where the concentrations of the two basic species were combined. Furthermore, the

combined concentrations were higher than sulfate concentrations when sulfate concentrations

were lower (< 50 neq/m3). There is a transition range between 50 and 500 neq/m3. Above 500

neq/m3, Sulfate concentrations were higher than those of the two basic species. Clearly

demonstrated, calcium and ammonium play a critical role in neutralizing the acid, and they

dominate in separate size ranges: ammonium for fine aerosols and calcium for coarse aerosols.

However, if the acid emission is high, there may not be sufficient cations in the air to balance the

acids.









Product filter floors

Phosphoric acid was the maj or species at the product filter floors and particulate fluoride

was the species of the next importance. The median mass concentrations standardd deviation)

of PM23, PMlo, and PM2.5 were 201 (f5 14), 178 (f372), and 95 (f68) Clg/m3, TOSpectively. The

median concentrations of phosphoric acid were 35 (PM23), 26 (PMlo), and 5 (PM2.5) Clg/m3 and

those of particulate fluoride were 19 (PM23), 17 (PMlo), and 11 (PM2.5) Clg/m3. Phosphoric acid

mists and aerosol mass size distributions are shown in Figure 4-3, and both had a similar trend.

It can be clearly seen that predominantly the phosphoric acid was present in the coarse mode

(3.8-10 Clm). The boiling point of pure phosphoric acid is 217 OC [Wea~st, 1988]. However, the

boiling point of phosphoric acid solution decreases as the concentration of phosphoric acid in the

solution decreases [M~essnaoui andBounahmidi, 2005]. The boiling point is only 108 OC when

phosphoric acid concentration is 36 wt%/ P205, which is the designated concentration in the

manufacturing process. This temperature is close to the operating temperature (100 oC) in the

manufacturing process at the plants. Hence, it is possible that phosphoric acid vaporizes at this

operating condition and then condenses to form aerosols when it encounters cooler air. The

species that had the second highest median concentration was fluoride and the size distributions

at the product filter floor areas are displayed in Figure 4-3C. The maj or mode size of fluoride

was 3.8-10 Clm. The minor mode size of fluoride was 0.03-0.2 Clm. Because the product filter

floors are semi-open space, no maj or cations are available to neutralize the phosphoric acid and

fluoride. The inside air does not mix well with ambient air. Therefore, the aerosol emission

controls the aerosol chemical species concentrations at this type of location.









Attack tank areas

Sampling was carried out at two attack tank areas and six samples were obtained.

Particulate fluoride was the maj or species at the attack tank areas. The median concentrations of

fluoride were 67 (PM23), 64 (PMlo), and 60 (PM2.5) Clg/m3, which indicated most fluoride existed

in small aerosols. Phosphate rock, which is the main useful product of phosphate ore, consists of

calcium phosphate mineral apatite with gangue constituents, which include silica, fluoride,

calcite, dolomite, clay and iron-aluminum oxide [Hodge andPopovici, 1994]. The attack tank is

where phosphate rock reacts with sulfuric acid. The violent reaction causes strong heat release in

the form of vapor, which is evacuated from the attack tank with other gaseous effluents.

Fluoride can be evaporated as gaseous fluoride, e.g., silicon tetrafluoride [Parish, 1994].

Fluoride has strong affinity with silica, and therefore fluoride generated from the process tends to

combine with silica. The high affinity of fluoride with silica causes fluorosilicic acid to remain

in phosphoric acid solution or fluorosilicate (SiF62-) to precipitate. Particulate fluoride size

distributions at the attack tank areas are shown in Figure 4-4. Two types of size distribution

were observed. When fluoride concentrations were high (two samples: 320 and 388 Clg/m3), the

mode size was 3.8-10 Clm. This occurred at a plant where the attack tank was not a closed

system. At this location with high fluoride concentrations, moisture was also high. Therefore,

the possible explanation for the size distribution is that gas phase fluoride and water were

evaporated and then condensed on existing aerosols. For the other plant where the attack tank is

a closed system, the fluoride concentrations and humidity were lower (four samples: 25-85

Clg/m3), and the mode size was 0.03-0.2 Clm.

Ammonium had the second highest concentration, and its median concentrations were 22

(PM23), 20 (PMlo) and 18 (PM2.5) Clg/m3. The relation between these two major species, fluoride









and ammonium, is shown in Figure 4-5. The results indicated that there was not enough

ammonium to neutralize fluoride when high fluoride concentrations were observed (> 250

neq/m3)

Granulator on a scrub day

Table 4-2 displays the median concentrations of PM23, PMlo and PM2.5 Of all species at the

granulator on a scrub day. Particulate fluoride, phosphate, sulfate and ammonium had higher

concentration levels than others. Nonetheless, there was no clear trend in terms of the

dominating species due to significant variation among the samples. Scrubbing is an intermittent

maintenance activity and it usually takes 3-4 h for one scrubbing. Therefore, the aerosol

emission was not stable.

Aerosol Acidity

Although the carcinogenic mechanism of strong inorganic acid mists containing sulfuric

acid is not clear yet, aerosol acidity may play a key factor. To estimate the aerosol acidity, the

charge balance method and the aerosol thermodynamic model were applied in this study.

Charge balance method

Sulfuric acid and phosphoric acid were the maj or species at the pump tank areas and the

product filter floors, respectively. Assuming that charge balance of those ions analyzed in this

study exists in the aerosols, hydrogen ion concentration can be estimated by the difference

between cation and anion equivalent weights. The relationships of cation equivalent weight and

anion equivalent weight of PM10-23, PM2.5-10, and PM2.5 at these locations are shown in Figure 4-

6. The aerosol acidity was high at the sulfuric acid pump tank areas for PM2.5-10 and PM2.5.

When anion concentration was higher than 500, 500, and 1000 neq/m3 for PM10-23, PM2.5-10, and

PM2.5, TOSpectively, most aerosols were acidic.









Whether the location is in an outdoor environment or an indoor setting affects the acidity.

In general, the maj or acid species can be neutralized by the abundant ammonia around the area in

the outdoor location; in contrast, there is not enough mixing with the ambient air in the indoor

location and the amount of ammonia is insufficient. A good example is the aerosol at the

product filter floor. All product filters are indoor locations. It can be clearly seen that most

aerosols are acidic in this type of location regardless of their particle size. In addition to

indoor/outdoor location, the other important factor is the acid loading. This can be best

illustrated using the samples at the sulfuric acid pump tank areas. At this area, the PM10-23

aerosol had a low acid loading (< 500 neq/m3) and the aerosol was more or less neutral.

Regarding PM2.5 and PM2.5-10 aroTsols, the same can be said for low anion loading cases (< 1000

and 500 neq/m3), while the aerosols were acidic at high loadings (> 1000 and 500 neq/m3) I

the case of high aerosol loadings, the aerosol remains acidic because of the limited amount of

basic species available to neutralize the particles in a short period of time after emission. In

PM2.5-10, 9 of 27 samples showed high acidity because sulfuric acid mists were dominant in this

size range. The sampling location for the granulator on a scrub day was an outdoor location

where the air could mix with ambient air. The aerosols from this location were close to neutral,

even though their concentrations were high. The maj or species to neutralize the aerosol acidity

was ammonium from the granulator where phosphoric acid reacted with ammonia to produce the

Einal products, diammonium phosphate or monoammonium phosphate.

Aerosol thermodynamic model

An aerosol thermodynamic model has been established in this study to calculate the acidity

of aerosols with high sulfuric acid and phosphoric acid mist concentrations. Sulfuric acid is a

stronger acid than phosphoric acid. Also, the former is hygroscopic in nature while the latter is









not. These two factors have strong influences on the aerosol's characteristics in a humid

environment, e.g., equilibrium size and aerosol H' concentration.

Four samples with sulfuric acid concentrations higher than 200 Clg/m3 were selected and

two samples with phosphoric acid concentrations higher than 500 Clg/m3 were selected as

representatives. Particulate fluoride concentrations were lower than 2.5 mg/m3, the OSHA

regulation. Furthermore, aerosol fluoride can exist as H2SiF6, which could not be discriminated

in this study. Therefore, the samples with high fluoride concentration at the attack tank area

were not investigated.

The results for aerosols with high sulfuric acid mist concentrations are displayed in Figure

4-7A. The modes of H+ concentration were 1.8-3.8 Clm and 3.8-10 Clm. The highest H+

concentration was 170 mole/kg (molality) and its mode was 1.8-3.8 Clm. The hygroscopic

sulfuric acid can quickly pick up moisture to form larger mist particles in the respiratory system

and the actual location where mists deposit can vary. Several assumptions were made to

estimate the deposition site of aerosols and the time to reach its equilibrium size: (1) the relative

humidity of the sampling locations and the human respiratory system were 40% and 95%,

respectively; (2) the geometric mean size (2.6 Clm for 1.8-3.8 Clm and 6.2 Clm for 3.8-10 Clm)

was considered as the representative size; (3) the cross-sectional area of human nose inlet is 2

cm2; (4) the air volume of one inhalation is 0.5 L; (5) the time for one inhalation is 1 s; (6) the

distance from nose inlet to the laryngeal region is 10 cm. The Z danov skii- Stoke s-Robi n son

(ZSR) equation [Jacobson, 1999] shown in Equation 4-10 was used to calculate the equilibrium

aerosol size at a given relative humidity, and Equation 4-11 [Hinds, 1999] was used to determine

the time for reaching the new equilibrium size.

ma = o+Y,aw +Y,a +Y,a' +Y4a r +ra 6 7~a (4-10)









where ma is the molality of solute x, aw the water activity (relative humidity expressed as a

fraction), and Y's the polynomial coefficients [Jacobson, 1999].

dI(d,) 4D~d pm
d x x #, d, > 1 (4 11)
dt Rp T,

where d, is the aerosol size, t the time, Dv the diffusion coefficient of vapor, M the molecular

weight of the liquid, R the gas constant (8.3 14 J-K- -mol )~, p the density of the liquid, p, the

partial pressure of vapor in the gas surrounding the droplet, T, the temperature away from the

droplet surface, h = 0.066 Clm, and

2il+ d
~= Fuch's correction factor

d, +5.33 +3.42il
d2

The hydrogen ion size distributions at RHs of 40% and 95% for the sample with sulfuric

acid mist concentration of 653 Clg/m3 are shown in Figure 4-7B. As the relative humidity

increases, H+ concentration decreases while the mode size slightly increases. The time required

to grow to the equilibrium size is only 0.014 s, which is smaller than the aerosol traveling time of

0.04 s for 10 cm. As shown, the inhaled mist aerosol can grow large enough to deposit in the

upper respiratory tract.

The International Commission on Radiological Protection (ICRP) model has been

developed to predict total and regional deposition of inhaled aerosol [ICRP, 1994]. Three

simplified ICRP equations, Equation 4-12 to Equation 4-14, were also applied in this study to

determine the deposition fraction (DF) for monodisperse spheres of standard density at standard

conditions [Hinds, 1999]. For the head airways, DFHA,









1 1
DFHA= IF (4-12)
1+ exp 6.84 +1.183In d, 1+exp( 0.924 -1.8851In d

where IF is the inhalable fraction given as


IF = 1- 0.5 1- (4-13)d2


For the tracheobronchial region, DFTB,

0.00352
DFTB d exip -0.234 Ind+34 +63.9exip -0.819 nd-1.61)) (4--14)

For the alveolar region, DFAL,





Three cases were considered, and the results are shown in Table 4-3. In Cases la and 2a,

the geometric mean sizes for the 2 modes discussed previously were used. These particles were

assumed to maintain their original sizes since the ICRP model does not allow consideration of

particle growth. As shown, the larger 6.2 Clm particles had a higher DF in the head airways than

the smaller 2.6 Clm particles. It should be noted, however, that the IF of 6.2 Clm particles was

slightly lower than 2.6 Clm particles. In Case 3, 11.8 Clm is the equilibrium size of 6.2 Clm. Due

to the much lower IF, the DF of 11.8 Clm particles was actually lower than the smaller 6.2 Clm

particles!

Since the growth to the equilibrium size takes much shorter time than the aerosol traveling

time as previously discussed, the equilibrium size was used to determine the deposition in Cases

lb and 2b. Different outcomes were observed! Compared with the no-growth cases, the DF

greatly increased. The results show that sulfuric acid mists sampled in this study mainly deposit









in the upper respiratory tract, which is consistent with epidemiological evidence of a correlation

between H' concentration and laryngeal cancer. The results also demonstrate the importance of

considering particle growth for hygroscopic components in assessing the particle's deposition in

the respiratory system. However, it must be noted that extensive epidemiological studies of

phosphate industry workers have concluded "no relation was found between acid mist exposures

and laryngeal cancer" [Checkoway et al., 1996].

Hydrogen ion size distributions for high phosphoric acid mist concentrations are shown in

Figure 4-7C. The mode size of hydrogen ion concentration was 1.8-3.8 Clm and the highest H

concentration was 37.4 mole-kg l. Compared to the sulfuric acid cases discussed previously,

clearly the hydrogen ion concentration of the phosphoric acid cases was lower even when its

concentration, 1589 Clg/m3, was higher than sulfuric acid mist concentration. The results show

that sulfuric acid plays the dominant role in contributing extractable H+ and phosphoric acid is

not as important as sulfuric acid in supplying extractable H .

Summary

In phosphate fertilizer facilities, phosphoric acid and sulfuric acid mists were the maj or

aerosol components for the product filter floors and the sulfuric acid pump tank areas,

respectively. The median concentration and the standard deviation of sulfuric acid at the sulfuric

acid pump tank areas were 37 & 322 Clg/m3. The median concentration & one standard deviation

for phosphoric acid was 35 + 326 Clg/m3, and it mainly existed in the coarse mode. The possible

source of phosphoric acid was evaporation and then condensation when it encountered cooler air.

The attack tank area had highest fluoride concentration, which was 25-3 88 Clg/m3 at this area.

The current OSHA 8-h TWA-PEL of phosphoric acid and sulfuric acid mist set at 1 mg/m3 was

not exceeded on average, but could be exceeded at the product filter floors and the pump tank









areas, respectively, under conservative assumptions that all phosphoric acid and sulfuric acid

came from phosphate and sulfate, respectively. It should be noted that workers spend much less

than 8 hours per day in the area, and thus the true time-weighted exposure level can be expected

to be lower.

Calcium and ammonium were the maj or species to neutralize the aerosol acidity at the

sulfuric pump tank areas when acid loading was low. The aerosol thermodynamic model

showed the modes of aerosol H+ concentration in 1.8-3.8 Clm and 3.8-10 Clm for the aerosols

with high sulfuric acid mist concentrations. These hygroscopic acid mists can grow in the high

humidity conditions of the upper respiratory system, and aerosols with high H' concentrations

mainly deposit in the upper respiratory system. Sulfuric acid was found to play a much more

prominent role than phosphoric acid and fluoride. The respiratory deposition projection of

sulfuric acid mists is consistent with that of H+ and both components mainly deposit in the

extrathoracic airways of the head and neck. However, extensive epidemiological studies of

phosphate industry workers have not shown an increased incidence of any type of cancer.









Table 4-1. Equilibrium relations and constants
Equilibrium Constant Equilibrium Constant Units Source
Equilibrium Relation Expression Key (298.15 K)


[H' HSO, H HS

[H SO, H~SOl



[H IH2PO, H H2P~O4
[H3PO4I ]H3PO4

[rOI[ H-PO, ]riP20g~p

[ NH, 3 H N~H 3H
NHNH1
[H OH




[NHIHS~O H r s



[NH, EH ISO~ rH H 2 z



[Na1 SO ,rsof-


Jacobson
mOle/kg [1999]

Jacobson
mOle/kg [1999]
Jiang
[1996]
mOle/kg

Jiang
mole- -kg [1996]

Benj amin
mole/kg [2002]
Kim et
mole/kg al.[1993]
Jacobson
mole2/kg2 [1999]
Jacobson
mOle2/kg2 [1999]
Jacobson
mole3/kg3 [1999]

mole /kg5 J10bson
Jacobson
mOle2/kg2 [1999]
mole/kg3Jacobson


HS4(aq) 4H~(aq) +Haq)


1.000x103


1.020x10-2


6.918x10-3


1.263


5.623 x 10-10


1.010x10-14

4.320x10-

1.380x102

1.820

2.930 x10

2.840 x102

4.800x10^


HSO,~aq)HSO~ (aq +H aq)


H3PO4(aq) ++ H aq) + H2PO4(aq)


H2PO4(aq) +H3PO4(aq) ++ H5P208(aq)


NH (aq) 0 NH3(aq) + H aq)

H20 mHiy + OHaq)

CaSO4 2H20 Ca2+,n~ + SOCaq) + 2H20(aq)

NH4HSO4(aq ) ~NH4aq +HSO4(aq)

( NH4 )2 SO4(s) 2 NH (aq) + SO 4aq)

(NH4 3j H (SO4 2(s) 0 3NH(aql) + 2SO~Jaq) + H aq) ,

NaHlSO4s (aq)~ +HSO(a)
NaS4(s) t2a a) + SOaq)








Ca3(PO 2(s)M3Ca2, + 2O~aq)2+NPO r 02+03- 2 109 103 m,,e5/kgS This study
(irHPO ~ ~ ~ ~ ~ C2 c' HO h HPO; r~ ,
CaP4(s) (2aq) +HPOaq) Ca+HO1.889x10-7 mole2/kg2 This study
Car (H2PO4 )2 H20s, 0 ~(Ca2+H +22O4(aq) + H2) Ca2+ IH2PO, re2+~ r2P w 6.153 x10-2 mOle3/kg3 This study
NH4H2PO4(s)r NH4(aq)+H2PO 4(aq) H2PO~:O~,,;~~ 7.106x10 mole2/kg2 This study
NH43PO(s 3N (a) PO q)NH ~iPOt H POr 9.437x101 mole4/kg4 This study
Na3PO 4(s) 0 3;Naaq) + PO a);-oi Naj(?)i PO a O 9.823 x101 mole4/kg4 This study
NaH2PO4(s) (aq), + H2P NllPO4(aq) H2lPO, 1.193 x101 mole2/kg2 This study
2irHPO4(s) 2Na aq) + HPOI (aq) 2 HPO~ ra HPO2- 6.879 mole3/kg3' This study

a"eq(298.15K)= exp kkr~vAGlo

R= 8.314 J/mole-K, T = 298. 15 K, ki = +1 for products and ki = -1 for reactants, vi = the dimensionless stoichiometric coefficient.









Table 4-2. Median ion species concentrations of cascade impactor samples collected at the granulator on a scrub day (yg/m3)
NO. Mass F- Cl- NO3- PO43 SO42 Na+ NH4t K Mg2 2a
PM23 392 54 5 4 35 26 9 66 3 2 11
PMio 331 42 5 6 23 22 8 58 3 1 7
PM2.5 263 39 4 4 9 18 6 48 2 1 3










Table 4-3. Aerosol deposition fractions for 3 cases
Size (Clm) Inhalable Deposition fraction
Case Inhalable Deposition fraction Head airway Tracheobronchial region Alveolar region
la 2.6 2.6 0.99 0.70 0.06 0.11
lb 2.6 5.0 0.99 0.89 0.04 0.06
2a 6.2 6.2 0.94 0.87 0.03 0.04
2b 6.2 11.8 0.94 0.92 0.01 0.01
3 11.8 11.8 0.78 0.77 0.01 0.01





S10 -

2 e Maximum

o 10' a95%
0 1* Median
Q 5%
O P O 0
100 .Minimum

H2SO4 H3PO4 F-

AT PT FF SD AT PT FF SD AT PT FF SD
Location
Figure 4-1. Sulfuric acid, phosphoric acid and fluoride concentrations at all locations. AT: attack
tank area, PT: sulfuric acid pump tank area, FF: product filter floor, SD: the
granulator plant on a scrub day














104


S103


102


100 I
100


10' 102 103

[SO 2] Alndp (neqna3)


104 105


mh
E

~ 103
v
a
Zi

~ 102
N
cs
U
Y


100 I
100


10' 102 103

[SO ] /Alnd, (neqna3)


104 105


~I

10' 50 102 500 10

[SO 2] Alndp (neq m )


100 I
I00


104 105


Figure 4-2. Relation between the maj or cations (ammonium and calcium) and sulfate
concentrations at the sulfuric acid pump tank areas. A) Ammonium versus sulfate. B)
Calcium versus sulfate. C) Ammonium and calcium versus sulfate





















0.1 1 10 100


-r- 95%


Median


1 5%


0.0 L
0.01

1.0

0.8

0.6r


0.2

0.0
0.01


0.1 1 10 100


0.0
0.01


0.1 1 10
Aerodynamic diameter (Clm)


Figure 4-3. Aerosol size distributions at the product filter floors. A) Phosphoric acid mist. B)
Aerosol mass. C) Particulate fluoride















1.0
******** Low conc.

0.8 High conc.

0.6

0.4

0.2:

0.0
0.01 0.1 1 10 100
Aerodynamic diameter (pCm)


Figure 4-4. Particulate fluoride size distribution at the attack tank areas
























* <1.8 plm (r2 = 4. 1x10 ')
o > 1.8 pLm (r2 = 3.5x10-2


10



103



10 1 -


100


100 10' 102 103 104 105

[F ]/Alndp (neq/m3)

Figure 4-5. Relation between ammonium and fluoride concentrations at the attack tank areas


























10' 102 103 104 10j


10i 102 103 104 105


a Product filter floor
() H SO4 pump tank area
0 Attack tank area
SGranulator on a Scrub day


10' 102 10 104 105
Anion equivalent weight (neq/m3 air)



Figure 4-6. Relationship of cation equivalent weight and anion equivalent weight. A) For
PM10-23. B) For PM2.5-10. C) For PM2.5

























































Aerodynamic diameter (pm)


Figure 4-7. Aerosol hydrogen ion concentration size distribution. A) Aerosol H' concentration
size distribution of samples with high sulfuric acid mist concentrations. B) Aerosol
H+ concentration size distributions for the sample with H2SO4 COncentration of 653
Clg/m3 at RHs of 40% and 95%. C) Aerosol H+ concentration size distribution of
samples with high phosphoric acid mist concentrations


- 1120 pg/m3
------ 1163 pg/m3
************ 653 llg/m3
.. .. .. 342 pg/m3


12 ,,

80


0.01
100 r


0I


0.1 1 10


3



2


0
0.01
50 r


I I lI _II I
0.1 1 10 100


- *A [PO43- alr
- B[PO43- al


1589 pg/m3
621.8 pg/m3


S30

-20


0.1 1


10 100









CHAPTER 5
POSITIVE SULFATE ARTIFACT FORMATION FROM SO2 ADSORPTION INT THE SILICA
GEL SAMPLER USED INT NIOSH METHOD 7903*

Background

NIOSH Method 7903 [NIOSH, 1994] is the approved method set by OSHA for measuring

the total concentration of acidic aerosols and gases, including hydrogen fluoride, hydrogen

chloride (HC1), hydrogen bromide (HBr), nitric acid (HNO3), Sulfuric acid and phosphoric acid.

It is the method commonly used by the health and safety staff in the phosphate industry, as well

as other occupational environments such as semiconductor industry, lead battery factories,

aluminum smelting, machining, electroplating processes, and even disaster response [Healy et

al., 2001; Hsu et al., 2007b; Lue et al., 1998; Tsai et al., 2001; Wallingford and Snyder, 2001].

The sampler of NIOSH Method 7903 consists of one section of glass fiber filter plug, followed

by two sections of silica gel. The glass fiber filter plug is designed to filter out the majority of

aerosols, whereas the silica gel sections are used mainly to adsorb acidic gases. The NIOSH

recommended sampling flow rate range is 0.2-0.5 Lpm (except that less than 0.3 Lpm should be

used for HF). The samples collected are extracted and then analyzed by IC. In evaluating the

method, NIOSH researchers reported ~100% collection efficiency for acidic gases (HC1, HF and

HNO3) [CGSSinelli, 1986; Cassinelli and Taylor, 1981]. For aerosols (H2SO4 and H3PO4), ~90%

efficiency (94.8f4.8% for H3PO4 and 86f4.6% for H2SO4) WAS reported when the samples

collected on the glass fiber filter section and the front silica gel section were combined. In

Chapter 3 [Hsu et al., 2007b], NIOSH Method 7903 was compared with cascade impactor

sampling for sulfuric acid mist sampling at phosphate fertilizer manufacturing facilities. The



* Reprinted with permission from Hsu, Y.-M., Kollett, J., Wysocki, K., Wu, C.-Y., Lundgren, D. A.,
Birky, B. K., 2007. Positive Artifact Sulfate Formation from SO2 Adsorption in the Silica Gel Sampler
Used in NIOSH Method 7903. Environ. Sci. Technol. 41, 6205-6209.









results indicate that sulfuric acid mist concentration from the NIOSH method might be

overestimated, because of the interference of SO2.

Past studies have reported that glass fiber filter can adsorb SO2, which can be subsequently

transformed into a sulfite (SO32-) Species [Chow, 1995; Coutant, 1977; Lee and2~ukund, 2001;

Watson and Chow, 2001] on the moist-basic surface of glass fiber filter, as listed in Reaction 5-1

and Reaction 5-2.

SO2 + OH- ++ HSO3 (5-1)

HSO3- + OH ++ SO32- + H20 (5 -2)

The glass fiber filter consists ofborosilicate glass filaments, which have high alkalinity

[Chow, 1995; Coutant, 1977; Watson et al., 1995]. It also contains high concentrations of

sodium, potassium, calcium, and other basic species that exhibit high alkalinities and pH values.

These properties aid in the adsorption of SO2, HNO3 and acidic gases [Chow, 1995; Lee and

M~ukund, 2001; Watson and Chow, 2001]. To cause artifacts in sulfate measurement, subsequent

oxidation to form sulfate is also critical. Oxidation of sulfite in solution is highly dependent on

pH, and the half-life is 4-5 min for sodium sulfite solution at room temperature, if the oxygen

supply is unlimited [Schroeter, 1963]. Coutant [1977] reported that the conversion on glass fiber

filter is 90% within 2 h when air containing SO2 paSses though the filter. Penetration occurs

when the alkalinity decreases below a certain value. Silica gel, a high surface area material, can

also adsorb sulfur dioxide [Fox and Jeffr~ies, 1979; Kopac andKocaba~s, 2002; Stratmann and

Buck, 1965]. The hydrophilic property of silica gel can effectively attract moisture, which can

enhance the absorption of soluble species such as SO2 [Tsai et al., 2001]. As mentioned, this

method is widely used in the workplace to characterize sulfuric acid mist concentrations, as a

way to evaluate protection for workers. No study has examined and quantified the artifact









sulfate from the silica gel tube. Therefore, the obj ective of this chapter was to characterize the

interference of SO2 to determine the accurate sulfuric acid mist concentration. The oxidation of

sulfur(IV) and SO2 adsorption, following the NIOSH protocol, were investigated in this study.

Methods

Two groups of experiments were conducted: sulfur(IV) oxidation and SO2 adsorption.

Sulfur(IV) includes sulfite, bisulfite (HSO3-), and sulfurous acid (H2SO3). These experiments are

described below.

Sulfur(IV) Oxidation

NIOSH Method 7903 specifies the use oflIC eluent solution as the extraction solution,

which is the 9 mM Na2CO3 Solution used in this study. In addition, a water bath at 100 OC was

used as specified in the method to enhance desorption of samples. To examine the effect of the

extraction procedures on sulfur(IV) oxidation, experiments were performed for four

combinations of eluent solution and water bath temperature, which were as follows:

(A) using DI water (Nanopure Diamond, Barnstead) of 18.2 MDZ cm as the extraction

solution without a water bath,

(B) using DI water as the extraction solution with a water bath at 100 oC to investigate the

effect of temperature,

(C) using 9 and 18 mM Na2CO3 Solutions as the extraction solution without a water bath to

examine the effect of the concentration of eluent, and

(D) using 9 mM Na2CO3 Solution as the extraction solution with a water bath at 100 oC.

Sodium sulfite (Na2SO3) WAS used as the sulfite source. 3.7% formaldehyde (HCHO)

solution was used to quench the sulfur(IV) oxidation [Dong andDasgupta, 1986; M~unger et al,

1986] at the designated time. For the procedures, a 100 mL solution (DI water, 9 or 18 mM









Na2CO3 Solution) was placed on the stirrer (Isotemp Magnetic Stirrer, Fisher) of 300 rpm at

room temperature or 100 oC. The designated amount of Na2SO3 WAS then added into the

solution. A 5 mL sample was taken out at a designated time and put into a glass tube with 5 mL

HCHO solution. Sulfate concentration was then analyzed via IC (model ICS 1500, Dionex).

Preliminary experiments were conducted to suggest that the experimental time should be 1

h for tests without water bath and 10 min for tests with water bath. Measurements indicated that

the sulfur(IV) oxidation was fast in the beginning. Hence, sulfate concentrations were measured

every minute, for the first 5 min, for oxidation without water bath and every 20 s, for the first 5

min, for oxidation with a water bath.

Sulfur Dioxide Adsorption

The experimental setup is shown in Figure 5-1. SO2 gaS from a cylinder (10 ppm, relative

uncertainty of +5%) was mixed with zero air (Thermo Electron Instrument) to obtain the desired

concentration. A bubbler was used to supply moisture, and the mixing ratio was used to adjust

the humidity. The gas stream then passed though a silica gel tube, followed by an impinger

containing 100 ml of a 9 mM Na2CO3 Solution. Preliminary experiments showed the SO2

collection efficiency of the impinger was 100%. The total molar concentration of SO2 paSsing

through the system was the sum of sulfate molar concentration from the silica gel tube and the

impinger. Hydrogen peroxide (0.6%) (1 mL) was added into the impinger after the experiment,

to oxidize collected sulfite to obtain sulfate concentration. The suggested flow rate range of

NIOSH Method 7903 is 0.2-0.5 Lpm. Thus, the experimental condition for sampling flow rate

was set in this range. According to past studies, SO2 COncentrations in sulfuric acid plants have

been 0. 12-15.9 ppm [Englander et al., 1988; M~eng and Zhang, 1990; Yady~YYYYYYYYYYYYYYYYYY andKaushik, 1996].

Hence, SO2 COncentration used in this study was set in this range. Experimental conditions are









shown in Table 5-1. The adsorption of SO2 On silica gel tube under various feed SO2

concentrations, sampling flow rates, and sampling times was examined.

Results and Discussion

Sulfur(IV) Oxidation

Aqueous sulfur(IV) uncatalyzed oxidation is a first-order reaction which can be expressed

by Equation 5-3 [Larson et al., 1978]:

dT~ SO ,,= k Sf-] (5-3)


The analytical solution to Equation 5-3 can be obtained by integrating it from time t = 0 to

time t and is shown in Equation 5-4. Sulfate was the species analyzed in this study; therefore,

sulfite mass concentration was calculated using the measured sulfate mass concentration

following Equation 5-5. The least-square fitting method (SigmaPlot, Version 8.0, SPSS Inc.,

Chicago, 1L) with Equation 5-4 was used to calculate the rate constant (k) and the pre-exponent

constant (a):


= aexp(-kt) (5-4)



[s04 ot] -[~so ]~
[SO32- t Toa x 80 (5-5)
96

where [SO32- t is the sulfite concentration at time t, [SO32- o the initial sulfite concentration,

[SO42- Total the total sulfate concentration, [SO42- t the sulfate concentration at time t, and t the

time (expressed in seconds).

Figure 5-2 illustrates the normalized sulfite concentration ([SO32- t/[SO32- o) aS a function

of time. For the conditions expressed in panels (A) and (C), sulfur(IV) oxidation reached at least

85% in 1 h and 40 min, respectively. The means and standard deviations of the rate constants for









the conditions in panels (A) and (C) were 0.0003 f 0.0001 and 0.0023 f 0.0003 s^l, respectively.

As shown, the addition of Na2CO3 enhanced the kinetics, i.e., the application of Na2CO3 Solution

as the extraction solution can result in more effective conversion. For the conditions described in

panel (B), sulfur(IV) oxidation reached at least 90% in 5-10 min and the mean f standard

deviation of the rate constants was 0.0198 f 0.0144 s^l. Comparing the results with that of

conditions expressed in panel (A), it can be observed that the kinetics was significantly increased

by the water bath designed to aid the desorption. Two Na2CO3 COncentrations (9 and 18 mM)

were tested; the results showed no discernible difference between these two. For condition (D),

sulfur(IV) oxidation reached 100% in just 2--3 min and the mean f standard deviation of the

rate constants was 0.0508 f 0.0274 s^l. The condition shown in panel (D) is the exact sample

extraction method specified by NIOSH Method 7903, which recommends a boiling time of 10

min. The results clearly show that SO2 adsorbed by silica gel and glass fiber filter can be

completely converted to sulfate during the extraction procedure.

Rate constants for uncatalyzed oxidation reactions of sulfur(IV) are summarized in Table

5-2. As shown, the reported values range widely. Indeed, many reaction rates of the uncatalyzed

oxidation of sulfur(IV) are often too high, because traces of transition metals in the water

enhance the uncatalyzed process [Huss et al., 1978]. Clark and Radojevic [Clarke and

Radojevic, 1983] obtained a rate constant that was 7 times slower for the uncatalyzed reaction,

when using Milli-R/Q water instead of distilled water. Furthermore, the oxidation rates shown in

Table 5-2 indicate that sulfur(IV) oxidation is strongly dependent on the pH. Radojevic

[Radojevic, 1984] has recommended the uncatalyzed oxidation rate be given by Equation 5-6:

d SO (M/s,)= 0.32 SO H~]. (pH < 7) (5-6)









The oxidation rates for the conditions portrayed in panels (A) and (C) in this study are

within this range. The oxidation rate for condition (C) was higher than that for condition (A) and

the pH dependence can explain this difference.

Some may have concerns about potential oxidants or catalysts present in the solution that

may change the oxidation rate. The IC extraction solution was prepared from fresh DI water; the

hydrogen peroxide (H202) and ozone (OS) COncentrations are negligible in the samples. The

concentrations of trace metals should be very low, because the DI water system provided the

fresh water with a resistivity of 18.2 MOZ-cm, which is an ion-free solution (except H' and OH-).

The IC eluent solution used in this study was prepared by using commercially available sodium

carbonate (EM Science). Based on the information provided by the supplier, the maximum iron

content is 0.0005% (w/w). The corresponding iron concentration of the IC extraction solution is

<8.5x10-s M, and the effect at this concentration level, if any, is considered to be negligible.

Sulfur Dioxide Adsorption

If artifact SO2 CauSes overestimation of sulfuric acid mist concentration, the degree of

impact for a given sampling condition must be investigated. In the second group of experiments,

SO2 COncentration, sampling flow rate, and sampling time were examined to assess their effects.

Sulfur dioxide concentration

Two runs of artifact sulfate concentrations under various SO2 COncentrations sampled at

0.3 Lpm for 2 h are displayed in Figure 5-3A. Runs A and B were conducted under the same

conditions to examine the reproducibility. Results from these two runs had the same trend and

the 10% variation of feed SO2 COncentration could explain the difference of artifact sulfate

concentrations between these runs. The sulfate concentration, or the interference of SO2,

increased as the inlet SO2 COncentration increased. Figure 5-3A indicated that, when the inlet









SO2 COncentration was 0.8 ppm (f10%), the mean artifact sulfate concentration was 190 Clg/m3

To assess the relative amount adsorbed, time-weighted collection percentage (TWCP) was

adopted, which is defined as the percentage of feed SO2 COncentration collected by the silica gel

tube over the given sampling time, i.e., (collected sulfate concentration/total sulfate

concentration) x 100%. The TWCP had the opposite trend, i.e., the TWCP was high when low

SO2 COncentration was introduced. The adsorption isotherm dictates that the adsorbed amount

increases but gradually plateaus as vapor concentration increases. For a linearly increasing

concentration, the percentage therefore decreases. When low SO2 COncentration (0.2 ppm) was

introduced, the TWCP reached 16% after 2 h of sampling at 0.3 Lpm, or equivalent to a 125

Clg/m3 Sulfate artifact. As SO2 COncentration increased to 0.8 ppm, the TWCP reached 6%,

which is equivalent to a 190 Clg/m3 Sulfate artifact after 2 h of sampling.

Sampling flow rate

The artifact sulfate concentrations and the corresponding TWCPs with an inlet SO2

concentration of 1.5 ppm (f10%) under various flow rates are exhibited in Figure 5-3B. The

sulfate concentration at a low flow rate (0.2 Lpm) was 374 Clg/m3, whereas, at the higher flow

rate, 0.5 Lpm, it was 163 Clg/m3. The TWCP at a low flow rate was also higher than that at a

high flow rate. Collection by diffusion is a function of residence time, and it increases as

residence time increases. As flow rate increased, the residence time decreased, resulting in a

lower artifact sulfate concentration. The NIOSH method recommends a flow rate lower than 0.3

Lpm if HF is present. However, as shown, a low flow rate yielded a higher amount of sulfate

artifact.









Sampling time

The artifact sulfate concentrations and the TWCP, as a function of sampling time, are

shown in Figure 5-3C. The TWCP for 2 h of sampling of 0.8 ppm SO2 were 16%, 9.2%, 5.4%,

and 4.6% for 0.2, 0.3, 0.4, and 0.5 Lpm, respectively. At 0.2 Lpm (2 h sampling), 0.8 ppm SO2

can cause artifact sulfate amount of 498 Clg/m3, which can significantly affect the accurate

determination of sulfuric acid mist concentration. Increasing the sampling time to 8 h yielded an

artifact at 0.2 Lpm of 172 Clg/m3. The artifact sulfate concentration decreased as sampling time

increased. This was due to the decrease of adsorption rate as the effective adsorption sites were

consumed over time. For the same reason, the TWCP also decreased.

Both glass fiber filter and silica gel can adsorb SO2. The adsorptions of the glass fiber

filter and the silica gel are displayed in Figure 5-4 using collection index (CI), which is defined

as the artifact sulfate amount (given in micrograms) divided by the feed SO2 COncentration

(expressed in units of pm). As shown, the adsorbed SO2 amOunt increased as the sampling time

increased for both glass fiber filter and silica gel, although the patterns were not the same. The

adsorption of SO2 On the glass fiber filter was very quick and accounted for the maj ority of the

adsorption in the first few hours. However, it was saturated within 2-3 h and the amount

increased only slightly after that. In contrast, the adsorbed amount of SO2 On silica gel increased

as the time increased.

The adsorbed SO2 amOunts under various SO2 COncentrations, flow rates, and sampling

times indicate that SO2 Can CaUSe significant interference. It can be as high as 500 Clg/m3 for 0.8

ppm SO2 at a sampling time of 2 h and flow rate of 0.2 Lpm. The OSHA regulation for sulfuric

acid mist concentration is 1 mg/m3. The result indicates that the interference caused by SO2

cannot be neglected when the mist concentration is low and SO2 COncentration is high (e.g., > 0.5









ppm). The artifact can affect the compliance status of a plant. Therefore, the silica gel sampler

containing a glass fiber filter plug and two sections of silica gel is not suitable for sampling

sulfuric acid mist under such conditions.

Sulfur Dioxide Adsorption Model

To estimate the artifact for a given sampling condition, a modified model based on a

deactivation model [Kopac andKocaba~s, 2002; Ya~syerli et al., 2001] that considers flow rate,

sampling time and SO2 COncentration was developed, which is shown in Equation 5-7. The first-

order deactivation rate constants (kd, and ksSo) can be obtained by fitting experimental data.


SO2 = 0SO O -Op- e~rxp(-kdt (5-7)


where [SOa~ is the artifact sulfate concentration (expressed in units of pLg/m3), Othe conversion

factor of SO2 to SO42- (1.67), [SO2 o the initial SO2 COncentration (expressed in unit of Clg/m3)

ksSo the observed adsorption rate (expressed in unit of cm3/min), Q the flow rate (given in units

of cm3/min), kd the first-order deactivation rate constant (given in units of min- ), and t the time

(expressed in minutes).

Results of the regression analysis of the experimental data are given in Table 5-3, which

shows a good relationship between the experimental data and the estimated values. The R2

values for the regression were 20.81 for the silica gel section. Although this model was

developed for the adsorption on silica gel, the adsorption on glass fiber filter also fit well. The

R2 ValUe WaS > 0.92.

The relation between the measured sulfate concentration from the silica gel tube and the

concentration from the model prediction is displayed in Figure 5-5. Most predictions are in the

range of +3 5% artifact sulfate concentration from the measurement. As demonstrated, the









deactivation model gives reasonably good predictions of artifact sulfate concentrations obtained

in the silica gel tube. Thus, it can be used for correcting the artifacts in sulfuric acid sampling if

the SO2 COncentration is available. The application of this model should be further investigate in

a field sampling.

Summary

NIOSH Method 7903, which uses one section of glass fiber filter and two sections of silica

gel, has been developed to determine the total concentrations of acid mists in workplace air

although certain gases are suspected to cause interference. In this study, experiments were

carried out to investigate the roles of S(IV) oxidation and sulfur dioxide (SO2) adsorption in

causing artifacts in sulfuric acid measurement. First, S(IV) oxidation under 4 combinations of

water bath temperature and Na2CO3 Solution concentration was examined to investigate the

effect of the extraction process of NIOSH Method 7903. It was shown that S(IV) oxidation to

form sulfate could reach 100% in just 2-3 min following the extraction process of NIOSH

Method 7903. The results demonstrate that using the procedure, SO2 adsorbed by the silica gel

and the glass fiber filter easily yields sulfate artifact. Sulfur dioxide adsorption under various

flow rates, SO2 COncentrations and sampling times was also investigated. The experimental data

were fitted into a deactivation model to determine the adsorption rate constant and the

deactivation rate constant. The model can serve as a tool for estimating the sulfate artifact if SO2

concentration is available.









Table 5-1. Experimental conditions of SO2 adsorption
Set Flow Rate (Lpm) Sampling Time (h) [SO2] ppm)
1 0.3 2 0.2, 0.4, 0.6 and 0.8
2 0.4 and 0.5 3 0.2, 0.6, 1.1, 1.4 and 1.6
3 0.2, 0.3, 0.4 and 0.5 3 and 8 1.5
4 0.2, 0.3, 0.4 and 0.5 2, 3, 4, 8, 10 and 12 0.8









Table 5-2. Rate constants for uncatalyzed oxidation reaction of sulfur(IV) by oxygen
Reaction conditions
k (s- ) pH Temp, T (oC) Reference
1.18 x10-' 3.0 25 Tanaka [1987]
1.72x10-' 4.0 25 Tanaka [1987]
9. 19x10-' 5.0 25 Tanaka [1987]
1.63x10-6 6.0 25 Tanaka [1987]
1.3 x10-s a 8.2-8.9 25 Clarke and Radoj evic [1983]
9.5 x10-5 b 8.2-8.9 25 Clarke and Radoj evic [1983]
1.7x10-3 6.8 25 Scott and Hobbs [1967]
3.0 x10-3 2.0-4.0 25 Miller and Pena [1972]
1.3x10-2 8.2-8.8 25 Fuller and Crist [1941]
aIn Milli-R/Q Water. bIn distilled water










Table 5-3. Rate parameters obtained using Equation 5-6
Glass fiber filter Silica gel
& ksJo kd R2 ksokd R2
Lpm mlpm min' mlpm minl
0.2 0.0254 0.0026 0.95 0.0119 0.0012 0.81
0.3 0.0226 0.0025 0.93 0.0114 0.0013 0.87
0.4 0.0169 0.0025 0.94 0.0096 0.0015 0.92
0.5 0.0160 0.0026 0.92 0.0089 0.0014 0.81
Mean 0.0202 0.0026 0.0104 0.0014
Stdeva 0.0045 5.8x10-5 0.0014 0.0001
Stdeva: Standard deviation









SO2 gaS


I400 Rotameter
.... ......Exhaus
Silica

to e Impinger


O MassBubbler
Zero Air flow
controller
Figure 5-1. Experimental setup for sulfur dioxide adsorption














0.8 -h, 0.8

O 0.6 11 0.6 -


A o 0.4B




0 10 20 30 40 50 60 0 2 4 6 8 10
1.0 1.0


0.8 1) 0.8



O 0.6 \- 0.6

C 0.4 0.4


0 10 20 30 40 50 60 0 2 4 6 8 10
Time (min) Time (min)

Figure 5-2. Sulfur(IV) oxidation under four conditions. A) DI water without a water bath. B) DI
water with a water bath. C) 9 and 18 mM Na2CO3 Solution without a water bath. D) 9
mM Na2CO3 Solution with a water bath











250


E 200

S150



50


15

10 P4
5


0


400


S300

S200

O 00


4P

2


8t 4




-*- Sulfate (A) A Sulfate (B)
-Mean sulfate ~TWCP(A)
~-TWCP (B) Mean TWCP

0 0.2 0.4 0.6 0.8 1
SO2 COncentration (ppm)




-? u


A uft(3hi Sulfate (8 h) P:~
TWCP (3 h)
STWCP (8 h)

0 0.2 0.4 0.(
Flow rate (Lpm)


+ Sulfate 0.2 Lpm
nSulfate 10.3 Lpm
*Sulfate 10.4 Lpm
ASulfate 10.5 Lpm
-e TWCP (0.2 Lpm
-~-- TWCP (0.3 Lp )
---e--- TWCP 0.4 Lp )
--A- TWCP (0.5 Lpm

e so


"E400

~300

vl200


LL


'

9


Sampling time (h)


Figure 5-3. Artifact sulfate concentrations and time-weighted collection percentages (TWCPs).
A) Various SO2 COncentrations with a flow rate of 0.3 Lpm and sampling time of 2 h.
B) Various sampling flow rates. C) Various sampling times and sampling flow rates.
Solid symbols represent artifact sulfate concentrations, and open symbols represent
their corresponding TWCPs












30

25

S20
A ,
-5a 15



5

0


30

25

20
C
-,15



5

0


30

25

20

.a-* ~15


-o-
~0 a


0 3 6 9 12 0

30

25

320



15





0 3 6 9 12 0


3 6 9 12


3 6 912
Time (h


Time (h)


Figure 5-4. Collection index (CI) at four flow rates (Lpm). A) 0.2. B) 0.3. C) 0.4. D) 0.5











600

10%
500
35%,
10%~
400


300 e
+ + 35%

O200







0 100 200 300 400 500 600

[SO42- measurd (Irg/m3)
Figure 5-5. Relationship between sulfate concentrations from the measurement versus from the
model









CHAPTER 6
MINIMIZATION OF ARTIFACTS INT SULFURIC ACID MIST MEASUREMENT USING
NIOSH METHOD 7903

Background

NIOSH Method 7903 [NIOSH, 1994] is an approved method set by the OSHA for

measuring the total concentration of acidic aerosols and gases, including HF, HC1, HBr, HNO3,

H2SO4 and H3PO4. It is the method commonly used by the health and safety staffs in the

phosphate industry [Hsu et al., 2007b] as well as other occupational environments such as the

semiconductor industry, lead battery factories, aluminum smelting, machining, electroplating

processes and even disaster response [Healy et al., 2001; Lue et al., 1998; Tsai et al., 2001;

Wallingforda~nd Snyder, 2001]. The sampler used for NIOSH Method 7903, a silica gel tube,

consists of one section of glass fiber filter plug followed by two sections of silica gel. The glass

fiber filter plug is designed to filter out the maj ority of aerosols while the silica gel sections are

used mainly to adsorb acidic gases. The NIOSH recommended sampling flow rate range is 0.2 -

0.5 Lpm, except that less than 0.3 Lpm should be used for HF. The collected samples are

desorbed in eluent and the aliquots are analyzed by IC. In evaluating the method, NIOSH

researchers reported nearly 100% collection efficiency for acidic gases (HC1, HF and HNO3)

[Cassinelli, 1986; Cassinelli and Taylor, 1981]. For aerosols (H2SO4 and H3PO4), arOund 90%

efficiency (94.8f4.8% for H3PO4 and 86f4.6% for H2SO4) WAS reported when the samples

collected on the glass fiber filter section and the front silica gel section were combined. Ortiz

and Fairchild [1976] reported approximately 70% of the aerosol mass was collected by the glass

fiber filter plug although the efficiency varied depending on the size distribution of the sampling

aerosol [Chen et al., 2002].

Sulfur dioxide is a species which can be adsorbed by the glass fiber filter [Chow, 1995;

Coutant, 1977; Lee and Mukund, 2001; Watson and Chow, 2001] and the silica gel [Fox and










Jeffr~ies, 1979; Kopac andKocaba~s, 2002; Stratmann and Buck, 1965]. Once adsorbed, it is

subsequently transformed into sulfate in the analytical process that causes an overestimate of

sulfuric acid mist concentration [Hsu et al., 2007c]. Hsu et al. [2007c] carried out experiments in

a laboratory system that verified and quantified the artifacts resulting from the adsorption and

conversion of interfering SO2 into sulfate. A deactivation model [Kopac andKocaba~s, 2002]

was modified for estimating the artifact sulfate concentration based on the various SO2

concentrations, flow rates and sampling times.

Denuder systems have been widely used for air sampling [Acker et al., 2005; Hayami,

2005; Huang et al., 2004; Pathak and Chan, 2005; Siouta~s et al., 1996; Tsai et al., 2004].

Various types of denuders have been developed and are commercially available. The denuder' s

wall is coated with pertinent adsorbents depending on the gaseous species of interest. When air

passes through the denuder, gas molecules with large diffusivity can diffuse to the wall and get

adsorbed. Basic adsorbent (e.g. sodium carbonate (Na2CO3)/glyCeoTl) can be used for acidic

gases, such as HC1, HNO3, SO2 and HNO2. Aerosols can be collected in the filter that follows.

As a consequence, this design can reduce the interference of gaseous species on aerosols. The

removal efficiency of SO2 USing an annular denuder system, depending on the operating flow

rate, has been reported to be > 99% when operated at a low flow rate [Possanzini et al., 1983].

The denuder system [Koutrakistrt~rtrt~rtrt~r et al., 1993] has also been applied for ambient air sampling and

can provide high SO2 COllection efficiency.

Accurate determination of sulfuric acid mist concentration in fertilizer manufacturing

facilities is seminal to the evaluation of its occupational exposure. Minimizing the artifact due to

SO2 gaS in the sampling process is essential in reaching this goal. The objectives of this chapter

are twofold: (1) to explore the use of a denuder for removing SO2 gaS from the sampling volume









to reduce the artifact sulfate and, (2) to assess the applicability of the deactivation model for

correcting the artifacts for the phosphate fertilizer production environment.

Methods

Two methods, a sampling system cooperating with a honeycomb denuder system (HDS)

and a deactivation model, were applied to minimize the artifact sulfate. Field sampling was

carried out to examine the artifact removal efficiency when the HDS was applied to remove the

interfering SO2 before entering the standard sampling train. Experiments were conducted to

characterize SO2 adsorption at high SO2 COncentrations encountered in the field sampling

conditions. The deactivation model was used to estimate the artifact sulfate concentration based

on known SO2 COncentrations.

Field Sampling

Field sampling was carried out on top of the sulfuric acid pump tank area at seven

phosphate fertilizer plants in Florida [Hsu et al., 2007c]. Four samples were acquired at each of

six sites and five samples were collected at the one remaining plant. Sampling time was 8 h for

each sample.

Silica gel tubes, a cascade impactor and a honeycomb denuder system were applied for the

field sampling. Three sets of sampling trains, shown in Figure 6-1, were employed. (A) A silica

gel tube This method was used for total sulfuric acid mist concentration following NIOSH

Method 7903 (N = 29). (B) The HDS followed by two silica gel tubes in parallel The HDS

(coating solution: 1% Na2CO3/glyCeoTl) was used to remove SO2 gaS before the air entered the

silica gel tubes. The measurement also provided SO2 COncentration. Two silica gel tubes were

applied for replication (N = 29 for the HDS sample; N = 58 for the silica gel tube sample). (C) A

cascade impactor (Mark III, U. Washington) with Zefluor membrane filters (P5PJ001, Pall









Corp.) This sampling system was used for sulfuric acid mist concentration with size-resolved

information (N = 29). Zefluor membrane filters can provide high aerosol collection efficiency

with low interaction with acidic gases [Chow, 1995; Watson and Chow, 2001]. All three

sampling trains were set side by side for parallel sampling.

All silica gel tube samples were analyzed following the sample preparation procedures

outlined in NIOSH Method 7903 [NIOSH, 1994], including adding 9 mM Na2CO3 Of 10 mL and

a water bath at 100 OC for 10 min. The HDS sample was extracted by 10 mL DI water

(Nanopure Diamond, Barnstead), and 1 mL hydrogen peroxide (0.6%) was then added into 1 mL

sample solution to oxidize sulfite to sulfate for the analysis. The cascade impactor sample was

extracted by 10 mL DI water with a 1-h ultrasonic bath. Sulfate concentrations of all samples

were analyzed via IC (Model ICS 1500, Dionex).

Deactivation Model

A deactivation model [Hsu et al., 2007c], Equation 6-1, has been developed to estimate the

artifact sulfate concentration. In developing this model, experiments were conducted at SO2

concentrations ranging from 0.2 to 1.6 ppm. Hence, the model can be applied for SO2

concentrations in this range.


SIO; = [SO2l~ o~ 1Op -xep(-kdt]


for 0.2 ppm < [SO2 o < 1.6 ppm (6-1I)

[SO:] : artifact sulfate concentration (pLg/m3)

6 = 1571, conversion factor

[SO2] o= feed SO2 COncentration (ppm)









ksSo = observed adsorption rate (mLpm/min), 16 for the glass fiber filter and 8.9 for

the silica gel at the flow rate of 500 mLpm

Q = flow rate (500 mLpm)

kd = first order deactivation rate constant (minl), 0.0026 for the glass fiber filter and

0.0014 for the silica gel at the flow rate of 500 mLpm

t = time (min)

Sulfur Dioxide Adsorption

High SO2 COncentrations, 1.6 ppm to 5.6 ppm, were observed during the field sampling

described in the previous section. However, the parameter values in the deactivation model have

not been validated for the artifact estimate at high SO2 COncentrations. Therefore, experiments

were conducted to quantify the SO2 adsorption by the silica gel tube under high SO2

concentrations ranging from 1.6 ppm to 5.6 ppm at 0.5 Lpm flow rate with 8 h of sampling time

(condition employed for the field sampling). The experimental setup has been described in Hsu

et al. [2007c]. SO2 gaS from a cylinder (10 ppm, relative uncertainty = +5%) was mixed with

zero air (Model 111, Thermo Electron Instrument) to obtain the desired concentration. SO2 gaS

was then passed through the silica gel tube for the adsorption. An impinger with 100 mL of 9

mM Na2CO3 Solution was employed downstream to collect the residual SO2 gaS. Sulfate

concentration was determined following the analytical procedures described in the field sampling

section. To evaluate residual sulfate of both glass fiber filter and silica gel, eighteen silica gel

tubes as received were analyzed for their sulfate concentration following the sample preparation

procedures of NIO SH Method 7903.









Results and Discussion


Field Sampling

Collection efficiency and concentration of SO2

The HDS consists of two honeycomb denuders in series. Collection efficiency, r, was

determined according to Equation 6-2:

[SO2(HDI)]-[SO2 (HD2)]
rl(%) = x 100% (6-2)
[SO2(D)

where [SO2(HDI)] and [SO2(HD2)] are SO2 COncentrations collected by the first and second

honeycomb denuders, respectively. The mean SO2 COllection efficiency f standard deviation of

the HDS was found to be 95.7 f 6.8% (N = 29), demonstrating that the deployed HDS could

effectively remove SO2 gaS from the sample gas.

Regarding the SO2 COncentration at the phosphate fertilizer plants, it ranged from 34 ppb to

5.6 ppm. Although it varied significantly from plant to plant, for each plant the SO2

concentration level varied within a limited range during the sampling period. Table 6-1 displays

the statistics of SO2 COncentrations at each plant. At plant F, sampling was conducted at two

locations where SO2 COncentrations differed greatly. Hence, SO2 COncentrations at plant F are

shown separately for each location.

Ratio of S-SO2/7 S-SO2

The deployment of the HDS was to collect SO2 gaS; however, sulfate aerosols might also

be collected by the HDS which could cause an overestimate of SO2 COncentration. To determine

if the presence of sulfate significantly affects the SO2 COncentration measurement, the S-SO42-/S-

SO2 ratios (elemental sulfur from SO42- COllected by the silica gel tubes, SGHA and SGHB in the

second sampling train, to elemental sulfur from SO2 COllected by the HDS) for all samples were

calculated and are shown in Figure 6-2. As shown, the maximum was 0.13 and it occurred at a









very low SO2 COncentration. At high SO2 COncentrations where the focus of this study is, they

were below 2%. The low ratios demonstrate that the honeycomb denuder system can be used for

measuring SO2 COncentration accurately and the interference of particulate sulfate is negligible.

Aerosol loss of HDS

The aerosol loss in the HDS causes an underestimate of sulfate concentration. Two

mechanisms causing the aerosol loss in the HDS are: (1) the aerosol with size larger than 10 Clm

can be removed by the impactor of the HDS, and (2) the aerosol can diffuse to the wall of the

denuder. The aerosol concentration with size-resolved information from the cascade impactor

(CI) can be employed to correct the first mechanism. The sulfate diffusional loss, [SOa ]DF ,

be calculated by Equation 6-3.


[SO DF= fDF, x SO ci,~ (6-3)

DFi : aerosol deposition fraction of aerosol size range i.


[SOi ci, : sulfate concentration of aerosol size range i measured by the CI.

The aerosol deposition (DF) in the honeycomb denuder due to diffusion can be calculated

according to Equation 6-4a and Equation 6-4b [Hinds, 1999].

DF, = 5.5p C,/3 3.77CL, for C1 < 0.009 (6-4a)

DF, = 1- 0.819exp(-11.54u)- 0.0975exp(- 70. 1p) for C1 > 0.009 (6-4b)

D,LN
where CL,=


Clt: dimensionless deposition parameter for particle size range i

Di: diffusion coefficient of the particle size range i (cm2/min)

L: the length of the tube (= 9.6 cm)









N : the number of tubes (= 160)

Q: the volume flow rate through the tube (= 500 mLpm)

Table 6-2 shows the mean ratio and its range of the sulfate diffusional loss, the sulfate loss due

to the HDS' inner impactor ( [SO H2SO4>~ 10p, derived from the CI measurement) and the total


sulfate loss ( SO1 ]Haios = [SO H2SO4> ~ 10p+ SO DF~) to the total sulfate concentration from


the CI ( SO: c~) at each plant.

The mean sulfate loss due to the aerosol diffusional loss at each plant ranged from 3.2% to

6.3%. For all plants, the mean loss was 4.8% demonstrating that the effect was minor. On the

other hand, the sulfate loss due to the impactor, shown in Table 6-2, was highly significant. The

mean loss ranged from 3% to 43% demonstrating that the loss varied remarkably. If the HDS

impactor sample were analyzed in this study, the information can be used to directly correct the

maj ority of the loss. The CI information is then no longer needed.

Sulfur Dioxide Adsorption

Figure 6-3 shows the artifact sulfate concentration as a function of SO2 COncentration. As

feed SO2 COncentration increased, the artifact sulfate concentration increased. This is because

more SO2 can be adsorbed by the silica gel and the glass fiber filter, and the glass fiber filter and

silica gel were not saturated yet. The relation between the feed SO2 COncentration (1.6--5.6

ppm) and the artifact sulfate concentration can be described by Equation 6-5. In this study, this

equation was applied for SO2 COncentrations between 1.6 and 5.6 ppm while the deactivation

model was applied for SO2 COncentrations from 0.2 to 1.6 ppm. The intercept in Equation 6-5

might be from the residual sulfate to be discussed in the next section. Since the ratio of the mean

amount of residual sulfate of the second silica gel section to that of the first silica gel section was









2.1, shown in Table 6-3, outliers were defined as those with ratios exceeding 2.1. Theoretically,

the first section should collect more SO2 gaS than the second section. However, the high residual

sulfate artificially yields a higher sulfate concentration in the second section. Since the mean

ratio of the sulfate concentration in the second section to that of the first section is 2. 1, outliers

are defined as exceeding this ratio.

[SOf] = 20.04 x[SO, ]o +53.32 for 1.6 ppm < [SO,]o < 5.6 ppm (6-5)


[SO: ]: artifact sulfate concentration (Irg/m3)

[SO,]o: feed SO2 COncentration (ppm)

To determine the effect of humidity, experiments were conducted for the relation between

adsorbed water amount and sampling time. RHs of 35% and 85% for various sampling times

and flowrates were investigated. It was found that the silica gel tube reached saturation quickly.

Under these sampling flowrates and sampling times, the adsorbed water amount was around 0.2

g which is one-third of the silica gel weight. When 0.2 Lpm flowrate and RH of 3 5% were

applied, the adsorbed water amount reached saturation in two hours. The results demonstrate

that the sampling tube is generally operated under the saturation condition over a wide range of

ambient RH.

Residual Sulfate in Silica Gel Tube


Due to the non-zero intercept in Equation 6-5, 18 silica gel tubes as received were analyzed

for residual sulfate concentration. The statistical results of the residual sulfate concentrations in

different sections for two batches purchased from the vendor are shown in Table 6-3. The first

batch was used for the experiment conducted in this research while the second batch was

purchased in 2004 and used for a prior study described in Chapter 3 [Hsu et al., 2007b]. For the









first batch, the mean residual sulfate concentrations from the glass fiber filter, first and second

sections of silica gel in the aliquots were 3.09, 5.23 and 10.77 Gig, respectively, which are

equivalent to 12.9, 21.8 and 44.9 Clg/m3 (in the air) at the sampling conditions of 8 h and 0.5

Lpm. Added together, the mean total sulfate concentration due to the residual was 79.5 Clg/m3

For the maximum residual sulfate concentration, the artifact sulfate concentration in the air was

439.3 Clg/m3. The total sulfate concentration in the air corresponding to the standard deviations

of the sampler was 132 Clg/m3. This is higher than the mean residual sulfate concentration,

indicating that the residual sulfate concentrations were quite variable. In contrast, the residual

sulfate of the second batch was much lower than that of the first batch. Its mean total residual

sulfate concentration was only 15.9 Clg/m3. In Summary, the residual sulfate concentration can

contribute greatly to the artifact sulfate concentration in the air and the impact of variation of

residual sulfate in the silica gel tube should not be ignored. Therefore, NIOSH Method 7903 is

not suitable for low sulfate concentration sampling, e.g., ambient air sampling.

Minimization of Artifact Sulfate

Two methods were applied to minimize the artifact sulfate and their effectiveness was

evaluated. The first method was the SO2 adsorption model described in Equation 6-1 and

Equation 6-5. This method is based on using the known SO2 COncentration to calculate the

artifact sulfate concentration. In this study, the SO2 COncentration was determined by the

denuder as discussed earlier. The corrected sulfate concentration based on the model,


[SO~ model can then be calculated according to Equation 6-6:


[SO]model= SO SG SOa~ (6-6)

where [SOI SG: .Sulfate concentration from the SG, which includes artifact sulfate.









The second method used to minimize artifact sulfate was the HDS (sampling train 2).

However, it has the potential to remove true sulfate aerosol. Therefore, the sulfate

concentrations from SGHA and SGHB need to be corrected and the correcting equation is shown

in Equation 6-7.


[SOi] SGHAC~ (or[SGHBC) = SO SGA (o SGHB + SOi H~los (6-7)

Figure 6-4 shows the sulfate concentrations from the SG, the SGHC (the average of the

corrected SGHA (SGHAC) and the corrected SGHB (SGHBC)), the model and the CI as a

function of SO2 COncentration. The sulfate concentrations from the SG were always higher than

those from other samplers. For most samples, the sulfate concentration from the CI was the

lowest and the sulfate concentrations from the model and SGHC were between the SG and the

CI. For low SO2 COncentration, the sulfate concentrations from the SG and the model were

similar due to the expected small artifact sulfate produced. The relative error (E) shown in

Equation 6-8 was applied to evaluate the difference and the results are shown in Table 6-4.


E= ;,,te;~ ], (6-8)



[SO ,,,,te : sulfate concentration measured by SG, SGHAC and SGHBC.

The relative errors of the NIOSH method and the model were relatively large when SO2

concentration was lower than 0.05 ppm; their mean values were 4.4 and 4.2, respectively.

Between 0.1 and 2.7 ppm, the sulfate concentration from the SG can be 1 to 7 times higher than

that from the CI. Meanwhile, both the model and the honeycomb denuder reduced the sulfate

concentration, making them closer to the CI results. When high SO2 COncentration, ranging from

4.0 to 5.6 ppm, was observed, relative errors of all methods were low. However, it should be









addressed that the artifact sulfate concentration from the SG was still high and needed correction.

For example, at SO2 COncentration of 4.7 ppm it was 287.6 Clg/m3.

The effectiveness, s, for reducing artifact sulfate by the SGHAC/SGHBC and the model

can be calculated by Equation 6-9.

reduced artifact sulfate [SO:]s SG-SO: iSGHAC SGHBC or model x10 69
true artifact sulfate [SO ]SG -SO: cI


They were 70 + 32% and 39 + 36% for the SGHAC/SGHBC and the model, respectively.

However, they were still higher than the CI values. One possible reason for the difference is the

residual sulfate of the silica gel tube discussed earlier. Overall, the model and the HDS can be

applied for the minimization of the artifact sulfate.

Aspiration Efficiency

The aspiration efficiencies ( rasp, calm air) for the cascade impactor, the silica gel tube and the

honeycomb denuder system were calculated by Equation 6-10 (Brockmann, 2001).



rasp, calm air =COs (7) + exp -~ (9-10)
U 1+ 2Stk





for 00 < p < 900, 10-3 < Vts/U <1, and 10-3 < Stk < 100.

Vts: terminal settling velocity

U: inlet sampling velocity

p= 00 is upward facing, 900 is horizontal

Stk: Stokes number (= T-U/d)

z: the relaxation time of the particle









d: inlet diameter

The results with two orientations are shown in Figure 6-5. For the vertical (facing up)

sampling, the aspiration efficiency of the silica gel tube is higher than others and the honeycomb

denuder system and the cascade impactor have the similar aspiration efficiencies. In the field

sampling, both the silica gel tube and the honeycomb denuder system were laid horizontally

while the cascade impactor was placed vertically. The efficiencies for both samplers in this

geometry are lower than the cascade impactor as shown in Figure 6-5 which indicates the

cascade impactor should collect more aerosols than others. However, this result was not

observed in this study. The aspiration efficiencies of three samplers were similar for aerosol

smaller than 20 Clm where most of the aerosols in the study were located. Hence, it is also

reasonable to apply the cascade impactor measurements to the aerosol information for the

honeycomb denuder system.

Sulfate Mass Balance

The sulfate mass balance between the silica gel tube in the first sampling train (SG) and the

two silica gel tubes in the second train (SGHA or SGHB) in the parallel samples was checked.

Ideally, they can be equated as Equation 6-11:

[SOi] =SOi]j o~~;~ + SOf] + SOI] j (6-11)

The sulfate mass balance is shown in Figure 6-6. Except for a few, most data points show

good mass balance, i.e., most of them were within the standard deviation of the residual sulfate

concentration of the silica gel tube. The large relative error of sulfate shown in Table 6-4

indicates the existence of artifact sulfate on the SG. The good sulfate mass balance between the

SG and the SGHAC/SGHBC supports the conclusion that SO2 gaS is the key source of artifact

sulfate.









Summary

A sampling train incorporating a honeycomb denuder system was applied for field

sampling at seven phosphate fertilizer plants to evaluate its use for reducing the artifact sulfate

concentration while preserving the actual sulfuric acid mist concentration. The denuder system

was designed to remove SO2 gaS before the air entered the silica gel tube and to monitor SO2

concentration at the same time. A deactivation model was also applied to correct for the

presence of the artifact. The denuder system had 95.7 + 6.8% collection efficiency for SO2 gaS,

and the impact of sulfate aerosol on SO2 COllection was negligible. SO2 COncentrations at the

seven plants ranged from 34 ppb to 5.6 ppm. Both the honeycomb denuder system and the

deactivation model were shown to reduce the artifact sulfate concentration by 70% and 39%,

respectively. However, they were still higher than the sulfate aerosol concentration measured by

a cascade impactor. One possible reason is the residual sulfate in the glass fiber filter and the

silica gel.










Table 6-1. Statistics of SO2 COncentrations (ppm)
Site A B C D E F-1 F-2 G
Mean 1.83 1.74 0.46 0.05 0.27 0.40 5.00 0.21
MIN 0.90 1.22 0.36 0.03 0.19 0.12 4.34 0.13
MAX 2.57 2.66 0.56 0.05 0.31 0.68 5.64 0.31

Sample no. 4 4 4 4 4 2 3 4









Table 6-2. Mean and standard deviation of the sulfate loss to the total sulfate concentration

[SO DF SO] H2SO410pm1 [SO ]HDios [SOi~ c(pg/m3
[SO ci SO~] cio SO c
A 4.8 (4.0-5.3) 15.7 (8.3-26.8) 25.4 (19.0-34.8) 92.4 (53.2-142.6)
B 3.2 (2.5-3.9) 42.8 (31.6-52.6) 49.1 (39.4-57.7) 25.7 (21.1-32.3)
C 4.0 (3.0-5.4) 26.3 (16.7-33.6) 34.4 (22.7-44.4) 31.0 (21.1-43.5)
D 6.3 (5.9-6.6) 2.8 (1.3-6.6) 15.4 (14.4-18.4) 85.3 (21.3-134.1)
E 4.8 (4.0-5.4) 15.0 (4.4-34.9) 24.7 (14.6-42.9) 32.0 (5.6-61.3)
F-1 4.8 (4.8-4.8) 13.4 (10.3-16.4) 23.0 (19.9-26.1) 16.7 (11.7-21.6)
F-2 4.7 (4.6-4.8) 11.6 (10.7-12.5) 20.9 (20.2-21.7) 578.2 (412.7-686.3)
G 5.7 (5.3-6.3) 5.8 (1.0-11.3) 17.3 (13.7-21.9) 44.3 (10.3-108.7)
Total 4.8 (4.0-5.3) 15.7 (8.3-26.8) 25.4 (19.0-34.8) 92.4 (53.2-142.6)









Table 6-3. Statistical results of the residual sulfate concentrations of silica gel tubes

Sulfate (Clg) Mean Standard deviation MAX MIN
1st Batch (N = 10)
Glass fiber filter 3.1 2.3 9.0 1.1
Silica gel 1st section 5.2 5.6 18.3 0.7
Silica gel 2nd section 10.8 23.9 78.1 0.9
2nd Section
1st section 2.1 4.3 4.3 1.3
2nd Batch (N = 8)
Glass fiber filter 0.6 0.6 1.9 0.0
Silica gel 1st section 2.0 0.5 2.6 1.2
Silica gel 2nd section 1.3 0.6 2.8 0.9
2nd Section
1st section 0.6 1.4 1.1 0.7










Table 6-4. Relative error of 4 samplers
Relative Error
SG SGHAC SGHBC Model
[SO2] < 1.6 ppm (N=21)
mean 4.46 1.73 1.86 3.30
stdev* 2.12 1.59 1.48 2.23
min 0.47 0.06 0.14 0.07
max 8.56 5.76 5.49 7.87
1.6 ppm < [SO2] < 5.6 ppm (N=8)
mean 1.90 0.48 0.63 0.82
stdev 2.00 0.23 0.44 1.02
min 0.02 0.20 0.20 0.17
max 5.78 0.94 1.30 3.10
Stdev*: standard deviation.










Air In

Silica gel tube


Rotameter


B


Air In


Honeycomb denuder



Rotameter


Silica gel tube


,- Cascade



III

I


SImpactor


Rotameter


Pump


Figure 6-1. Three sampling trains. A) NIOSH Method 7903. B) Modified NIOSH Method 7903.
C) Cascade impactor sampling system



137


Pump












0.14


0.12


0.10


0.08


0.06


0.04


0.02


0.00


1 2


SO2 (ppm)
Figure 6-2. S-SO2/S-SO42- as a function of SO2 COncentration. SGHA
gel tubes in the second sampling train


and SGHB are the silica










180

S 160 .

~- 140 *

C~120

10 -o

~ so [SO42-] t=20.04[SO2 o,+53.32
R2 =0.9463

60
1 2 3 4 5 6

SO2 (ppm)
Figure 6-3. Artifact sulfate concentration as the function of SO2 COncentration








500

400

300

200


A


0.0


800 r


0.5 1.0 1.5 2.0 2.5


3.0


SO2 (ppm)


V


OO

*I SG
O O SGHC
v Model
CI


600


400


200


B
Q


0
3.0


3.5


4.0


4.5


5.0


5.5


6.0


SO2 (ppm)

Figure 6-4. Comparison of the sulfate concentrations from the CI, SG, SGHAC and SGHBC. A)
SO2 COncentration lower than 3 ppm. B) SO2 COncentration higher than 3 ppm.










1.0



a 0.8

S0.7
d SG V'
.0 0.6
...0----- CI V L

S 0.5 -- --HD V

0.4-
-E- HD H '
0.3
0 10 20 30 40 50 60

Aerodynamic Diameter (Clm)
Figure 6-5. Aspiration efficiency of three samplers. (SG: silica gel tube, CI: Cascade impactor;
HD: honeycomb denuder system, V: vertical; H: horizontal)










1000


F9600
O

S400

20

0l


0 200


400 600
Sulfate (SG) (pg/n?)


1000


Figure 6-6. Sulfate mass balance between SG and SGHAC/SGHBC (Stdev*: standard deviation
of the residual sulfate)









CHAPTER 7
CONCLUSIONS

Aerosol sampling using the dichotomous sampler, NIOSH Method 7903, and a cascade

impactor was carried out at five types of locations at eight phosphate plants and two background

sites to determine the worker exposure to sulfuric acid mist concentration in phosphate fertilizer

plants. Artifact sulfate observed in using NIOSH Method 7903 for sampling in phosphate

fertilizer manufacture process was also investigated in this study. To minimize the artifact

sulfate, two methods were developed. Five conclusions can be drawn from this study.

Conclusion 1

The highest sulfate concentration, 0.185 mg/m3, in the plants was obtained at the sulfuric

acid pump tank area. Should monitoring of personal exposure to sulfuric acid mist be required,

efforts should focus on workers with activities in this area where concentrations approach than

the TLV-TWA standard of 0.2 mg/m3 TOCOmmended by ACGIH for the thoracic fraction of

sulfuric acid aerosol. At the attack tank area, fluoride was the dominant species and the

maximum fluoride concentration in PMlo was 462 Clg/m3. At the rotating table/belt filter floor,

phosphoric acid is separated from gypsum by rotating table/belt filter and the high temperature is

favorable for the evaporation of phosphoric acid and the maximum phosphate concentration in

PMlo was 170 Clg/m3. On a scrub day, a weak sulfuric acid solution is used to clean the piping

and ductwork of the granulator for an average of 4 hours per day. Particulate sulfate

concentrations were low during the scrubbing activity. At the truck loading/unloading station,

the possible emission period is around 2-3 h/day, and this emission is not continuous. The

concentration levels at the loading/unloading station were low and were greatly influenced by

outdoor conditions.









Conclusion 2

Based on cascade impactor sampling, sulfuric acid pump tank areas still had higher

sulfuric acid mist concentrations than other types of locations, and sulfuric acid was the

dominant chemical species. When high sulfuric acid concentrations were identified, the aerosols

were dominantly in the coarse mode. The most likely cause for high sulfuric acid concentrations

at this location is the leakage of SO3. According to the impactor sampling results, 7 samples

(total: 72) exceeded the ACGIH recommendation (0.2 mg/m3, thoracic fraction), and 2 samples

(total: 72) were above the OSHA regulation (1 mg/m3, total concentration). Meanwhile, there

were 7 samples (total: 78) by the NIOSH method that exceeded the OSHA regulation. The

sulfuric acid mist concentrations from the NIOSH method were higher than those from the

cascade impactor for the dominating majority of samples. The possible reason for this variation

is the interaction between SO2 and silica gel/glass fiber.

Conclusion 3

In phosphate fertilizer facilities, phosphoric acid and sulfuric acid mists were the maj or

aerosol components for the product filter floors and the sulfuric acid pump tank areas,

respectively. The possible source of phosphoric acid was evaporation and then condensation

when it encountered cooler air. The current OSHA 8-hour TWA PEL of phosphoric acid and

sulfuric acid mist set at 1 mg/m3 was not exceeded on average, but could be exceeded at the

product filter floors and the pump tank areas, respectively. It should be noted that workers spend

much less than 8 hours per day in the area, and thus the true time-weighted exposure level can be

expected to be lower. Calcium and ammonium were the maj or species to neutralize the aerosol

acidity at the sulfuric pump tank areas when acid loading was low. The aerosol thermodynamic

model showed the modes of aerosol H' concentration in 1.8 -3.8 Clm and 3.8 -10 Clm for the

aerosols with high sulfuric acid mist concentrations. These hygroscopic acid mists can grow in









the high humidity condition in the upper respiratory system, and aerosols with high H+

concentrations mainly deposit in the upper respiratory system. Sulfuric acid was found to play a

much more prominent role than phosphoric acid. The respiratory deposition projection of

sulfuric acid mists is consistent with that of H' ion and both components mainly deposit in the

human head airway. However, extensive epidemiological studies of phosphate industry workers

have not shown an increased incidence of any type of cancer resulting from these exposures.

Conclusion 4

The artifact sulfate of NIOSH Method 7903 originating from SO2 gaS was confirmed. SO2

can be adsorbed by the glass fiber filter and the silica gel and the adsorbed SO2 can be oxidized

and transformed into sulfate during the extraction procedures of NIOSH Method 7903. The

interference from SO2 CannOt be neglected when the sulfuric acid mist concentration is low and

SO2 COncentration is high (> 0.5 ppm). The artifact sulfate can affect the compliance status of a

facility and should be corrected. A deactivation model was developed to estimate the artifact

sulfate concentration.

Conclusion 5

Two methods, a honeycomb denuder system and a deactivation model, were applied to

minimize the artifact sulfate concentration. The honeycomb denuder system efficiently adsorbed

SO2 before it entered the silica gel tube while the deactivation model was employed to calculate

the artifact sulfate concentration. Both methods were proven to reduce the artifact. However,

the sulfate concentrations from both methods were still higher than the sulfate concentration

from the cascade impactor. One likely reason is the residual sulfate from the silica gel tube,

which yields a mean artifact sulfate equivalent to 79.5 Clg/m3 for 8-h sampling.










LIST OF REFERENCES


ACGIH (2004), Sulfuric acid. TLV Chemical Substances. 7th Edition Documentation, 1-12.

Acker, K., D. Moller, R. Auel, and D. Kalass (2005), Concentrations of nitrous acid, nitric acid,
nitrite and nitrate in the gas and aerosol phase at a site in the emission zone during
ESCOMPTE 2001 experiment, Atnzos. Res., 74, 507-524.

Apol, A. G., and M. Singal (1987), Health Hazard Evaluation, J. R. Simplot Company,
Pocatello, ID, National Institute for Occupational Safety and Health, Cincinnati, Ohio.

Baron, P. A., and K. Willeke (2001), Aerosol Mea;surement: Principles, Techniques, and
Applications, 2nd ed., 1131 pp., Wiley, New York.

Bassett, M., and J. H. Seinfeld (1983), Atmospheric equilibrium-model of sulfate and nitrate
aerosols, Atnzos. Environ., 1 7, 2237-2252.

Becker, P. (1989), Industry process chemistry. Chapter 2., in Phosphates and Phosphoric Acid:
Raw Materials, Technology, and Economics of the Wet Process, edited by P. Becker, pp.
43-177, Marcel Dekker, New York.

Becker, P. (1989), Phosphates and Phosphoric Acid: Raw Materials, Technology, and
Economics of the Wet Process, 2nd ed., 740 pp., Marcel Dekker, Inc., New York.

Benjamin, M. M. (2002), Water Chentistly, 668 pp., McGraw-Hill, Boston.

Bhaskaran, R., N. Palaniswamy, N. S. Rengaswamy, and M. Jayachandran (2004), Cost of
corrosion in fertilizer industry A case study, Mater. Corros., 55, 865-871.

Blair, A., and N. Kazerouni (1997), Reactive chemicals and cancer, Cancer Causes Control, 8,
473-490.

Block, G., G. M. Matanoski, R. Seltser, and T. Mitchell (1988), Cancer morbidity and mortality
in phosphate workers, Cancer Res., 48, 7298-7303.

Brockmann, J. E. (2001), Sampling and Transport of Aerosols, in Aerosol 2ea~surentent:
Principles, Techniques, and Applications, edited by P. A. Baron and K. Willeke, pp. 143-
195, Wiley, New York.

Cassady, M. E., H. Donaldson, and S. Gentry. (1975), Industrial hygiene survey report of
Agrico Chemical Company, Pierce, Florida, June 22-26, 1975., National Institute for
Occupational Safety and Health, Cincinnati, OH.

Cassinelli, M. E. (1986), Laboratory evaluation of silica-gel sorbent tubes for sampling
hydrogen-fluoride, Am. Ind. Hyg. Assoc. J., 47, 219-224.










Cassinelli, M. E., and D. G. Taylor (1980), Monitoring for airborne inorganic acids, Abstracts
ofPapers of the American Chemical Society, 180, 39.

Cassinelli, M. E., and D. G. Taylor (1981), Monitoring for airborne inorganic acids, Am. Chem.
Soc. Symp. Ser., 149, 137-152.

Cassinelli, M. E., and P. M. Eller (1979), lon chromatographic determination of hydrogen
fluoride, paper presented at Annual Conference, American Industrial Hygiene Conference
and Exposition, Chicago, IL, 27 May 1 June.

CFR Table Z-1 Limits for Air Contaminants, CFR (Code of Federal Regulation).

Checkoway, H., R. M. Mathew, J. L. S. Hickey, C. M. Shy, R. L. Harris, E. W. Hunt, and G. T.
Waldma (1985a), Mortality among workers in the Florida phosphate industry.1. Industry-
wide cause-specific mortality patterns, J. Occup. Environ. M~ed., 27, 885-892.

Checkoway, H., R. M. Mathew, J. L. S. Hickey, C. M. Shy, R. L. Harris, E. W. Hunt, and G. T.
Waldma (1985b), Mortality among workers in the Florida phosphate industry.2. Cause-
specific mortality relationships with work areas and exposures, J. Occup. Environ. M~ed.,
27, 893-896.

Checkoway, H., N. J. Heyer, and P. A. Demers (1996), An updated mortality follow-up study of
Florida phosphate industry workers, Am. J. Ind. M~ed., 30, 452-460.

Chen, C. C., C. H. Wu, S. H. Huang, and Y. M. Kuo (2002), Aerosol penetration through silica
gel tubes, Aerosol Sci. Technol., 36, 457-468.

Chow, J. C. (1995), Measurement methods to determine compliance with ambient air-quality
standards for suspended particles, J. Air Waste Manage. Assoc., 45, 320-3 82.

Clarke, A. G., and M. Radoj evic (1983), Chloride-ion effects on the aqueous oxidation of SO2,
Atmos. Environ., 1 7, 617-624.

Coutant, R. W. (1977), Effect of environmental variables on collection of atmospheric sulfate,
Environ. Sci. Technol., 11, 873-878.

Denzinger, H. F. J., H. J. Konig, and G. E. W. Kruger (1979), Fluorine recovery in the fertilizer
industry a review, Phosphor. Potassium, 103, 33-39.

Dibb, J. E., R. W. Talbot, G. Seid, C. Jordan, E. Scheuer, E. Atlas, N. J. Blake, and D. R. Blake
(2002), Airborne sampling of aerosol particles: Comparison between surface sampling at
Christmas Island and P-3 sampling during PEM-Tropics B, J Geophys Res-Atmos, 108,
Art. No. 8230.

Divita, F., J. M. Ondov, and A. E. Suarez (1996), Size spectra and atmospheric growth of V-
containing aerosol in Washington, DC, Aerosol Sci. Technol., 25, 256-273.










Dong, S., and P. K. Dasgupta (1986), On the formaldehyde-bisulfite-hydroxymethane-sulfnt
equilibrium, Atnzos. Environ., 20, 1635-1637.

Englander, V., A. Sjoberg, L. Hagmar, R. Attewell, A. Schutz, T. Moller, and S. Skerfying
(1988), Mortality and cancer morbidity in workers exposed to sulfur-dioxide in a sulfuric-
acid plant, hIt. Arch. Occ. Eny. Hea., 61, 157-162.

FDEP (2003), Florida Air Monitoring Report 2003, edited by A. R. Management, pp. 7.1-7.10,
Florida Department of Environmental Protection, Tallahassee, Florida.

Finlayson-Pitts, B. J., and J. N. Pitts (2000), Chemistry of the Upper and Lower Atmosphere:
Theory, Experiments and Applications, 969 pp., Academic Press, San Diego, Cali.

FIOH (1990), Industrial hygiene measurements, 1950-69 (data base), Finnish Institute of
Occupational Health, Helsinki (Finland).

Fox, D. L., and H. E. Jeffries (1979), Air-Pollution, Anal. Chent., 51, R22-R34.

Fuller, E. C., and R. H. Crist (1941), The rate of oxidation of sulfite ions by oxygen, J. Anzer.
Chent. Soc., 63, 1644 1650.

Hagmar, L., T. Bellander, C. Andersson, K. Linden, R. Attewell, and T. Moller (1991), Cancer
morbidity in nitrate fertilizer workers, Int. Arch. Occup. Environ. Health., 63, 63-67.

Hayami, H. (2005), Behavior of secondary inorganic species in gaseous and aerosol phases
measured in Fukue Island, Japan, in dust season, Atnzos. Environ., 39, 2243-2248.

Healy, J., S. D. Bradley, C. Northage, and E. Scobbie (2001), Inhalation exposure in secondary
aluminium smelting, Ann. Occup. Hyg., 45, 217-225.

Hinds, W. C. (1999), Aerosol Technology: Properties, Behavior, and Mea~surement of Airborne
Particles,~~PPP~~~PP~~~PP 2nd ed., 483 pp., John Wiley & Sons, Inc., New York.

Hod ge, C. A., and N. N. Popovici (1994), Pollution Control in Fertilizer Production, Marcel
Dekker, Inc., New York.

Horton, K. D., and J. P. Mitchell (1989), Water droplet fragmentation in an Andersen Mark-III
in-stack cascade impactor, J. Aerosol Sci., 20, 1585-1588.

Howell, S., A. A. P. Pszenny, P. Quinn, and B. Huebert (1998), A field intercomparison of three
cascade impactors, Aerosol Sci. Technol., 29, 475-492.

Hsu, Y. M., C. Y. Wu, D. A. Lundgren, and B. K. Birky (2007a), Chemical characteristics of
aerosol mists in phosphate fertilizer manufacturing facilities, J. Occup. Environ. Hyg., 4,
17-25.










Hsu, Y. M., C. Y. Wu, D. A. Lundgren, and B. K. Birky (2007b), Size-resolved sulfuric acid
mist concentrations at phosphate fertilizer manufacturing facilities in Florida, Ann. Occup.
Hyg., 51, 81-89.

Hsu, Y. M., J. Kollett, K. Wysocki, C. Y. Wu, D. A. Lundgren, and B. K. Birky (2007c),
Positive artifact sulfate formation from SO2 adsorption in the silica gel sampler used in
NIOSH Method 7903, Environ. Sci. Technol., 41, 6205-6209.

Huang, Z., R. M. Harrison, A. G. Allen, J. D. James, R. M. Tilling, and J. X. Yin (2004), Field
intercomparison of filter pack and impactor sampling for aerosol nitrate, ammonium, and
sulphate at coastal and inland sites, Atmos. Res., 71, 215-232.

Huss, A., P. K. Lim, and C. A. Eckert (1978), Uncatalyzed oxidation of sulfur(IV) in aqueous-
solutions, J. Amer. Chem. Soc., 100, 6252-6253.

IARC (1992), IARC Monograph on the Evaluation of Carcinogenic Risks to Humans, Vol. 54,
Occupational Exposures to M~ists and Vapours from Strong Inorganic Acids; and Other
Industrial Chemicals, 336 pp., International Agency for Research on Cancer (IARC).
Lyon, France.

ICRP (1994), H~uman Respiratory Tract M~odel for Radiological Protection, Elsevier Science,
Tarrytown, NY.

Jacobson, M. Z. (1999), Chemical equilibrium and dissolution processes, in Fundamentalsd~~~dd~~~ddd~~ of
Atmospheric M~odeling, edited, pp. 553-597, Cambridge University Press, New York.

Jiang, C. F. (1996), Thermodynamics of aqueous phosphoric acid solution at 25 degrees C,
Chem. Eng. Sci., 51, 689-693.

Jonnalagadda, S. B., M. Nyagani, P. Sawunyama, and R. Baloyi (1991), Studies on the levels of
sulfur-dioxide, nitrogen-dioxide, ammonia, and hydrogen-chloride in ambient air of
Harare, Zimbabwe, Environ. Int., 1 7, 461-467.

Kim, Y. P., J. H. Seinfeld, and P. Saxena (1993), Atmospheric gas-aerosol equilibrium.2.
Analysis of common approximations and activity-coefficient calculation methods, Aerosol
Sci. Technol., 19, 182-198.

Kopac, T., and S. Kocabas (2002), Adsorption equilibrium and breakthrough analysis for sulfur
dioxide adsorption on silica gel, Chem. Eng. Process., 41, 223-230.

Koutrakis, P., C. Sioutas, S. T. Ferguson, J. M. Wolfson, J. D. Mulik, and R. M. Burton (1993),
Development and evaluation of a glass honeycomb denuder filter pack system to collect
atmospheric gases and particles, Environ. Sci. Technol., 27, 2497-2501.










Larson, T. V., N. R. Horike, and H. Harrison (1978), Oxidation of sulfur-dioxide by oxygen and
ozone in aqueous-solution kinetic study with significance to atmospheric rate processes,
Atmos. Environ., 12, 1597-1611.

Lee, K. W., and R. Mukund (2001), Filter collection, in Aerosol M~easurement: Principles,
Techniques, and Applications, edited by P. A. Baron and K. Willeke, pp. 197-220, Wiley,
New York.

Lue, S. J., T. Wu, H. Hsu, and C. Huang (1998), Application of ion chromatography to the
semiconductor industry I. Measurement of acidic airborne contaminants in cleanrooms, J.
Chromatogr. A, 804, 273-278.

Lunsford, J. H. (1979), Surface Reactions of Oxides of Sulfur, U. S. Environmental Protection
Agency, Research Triangle Park, NC.

Mann, H. C. (1992), Ammonium phosphates, in Air Pollution Engineering Ma'~nual, edited by
A. J. Buonicore and W. T. Davis, pp. 574-576, Van Nostrand Reinhold, New York.

Meng, Z. Q., and B. Zhang (1997), Chromosomal aberrations and micronuclei in lymphocytes
of workers at a phosphate fertilizer factory, Mutat. Res., 393, 283-288.

Meng, Z. Q., and L. Z. Zhang (1990), Chromosomal-aberrations and sister-chromatid
exchanges in lymphocytes of workers exposed to sulfur-dioxide, Mutat. Res., 241, 15-20.

Meng, Z. Q., H. Q. Meng, and X. L. Cao (1995), Sister-chromatid exchanges in lymphocytes of
workers at a phosphate fertilizer factory, Mutat. Res., 334, 243-246.

Meng, Z. Y., and J. H. Seinfeld (1996), Time scales to achieve atmospheric gas-aerosol
equilibrium for volatile species, Atmos. Environ., 30, 2889-2900.

Messnaoui, B., and T. Bounahmidi (2005), Modeling of excess properties and vapor liquid
equilibrium of the system H3PO4 H20, Fluid Pha~se Equilib., 23 7, 77-85.

Miller, J. M., and R. G. D. Pena (1972), Contribution of scavenged sulfur-dioxide to sulfate
content of rain water, J. Geophys. Res., 77, 5905-5916.

Muller, T. L. (1992), Sulfuric acid, in Air Pollution Engineering Ma'~nual, edited by A. J.
Buonicore and W. T. Davis, pp. 469-476, Air & Waste Management Association,
Pittsburgh, PA.

Munger, J. W., C. Tiller, and M. R. Hoffmann (1986), Identification of
hydroxymethanesulfonate in fog water, Science, 231, 247-249.

NIOSH (1994), NIOSH Method No. 7903, in NIOSH2anual ofAnalytical2\~ethods, edited by
P.C. Schlecht and P.F. O'Connor, National Institute for Occupational Safety and Health
(NIOSH), Cincinnati, OH.










Ortiz, L. W., and C. I. Fairchild (1976), Aerosol Research and Development Related to Health
Hazard Analysis, US Energy Research and Development Administration.

Ostro, B. D., M. J. Lipsett, M. B. Wiener, and J. C. Selner (1991), Asthmatic responses to
airborne acid aerosols, Am. J. Public Health, 81, 694-702.

Palm, G. F. (1992), Phosphoric acid manufacturing, in Air Pollution Engineering Ma'~nual,
edited by A. J. Buonicore, et al., pp. 439-446, Van Nostrand Reinhold, New York.

Parish, W. R. (1994), Phosphoric acid by wet process: recovery of fluorine-containing gases, in
Pollution control in fertilizer production, edited by C. A. Hod ge and N. N. Popovici., pp.
197-208, Dekker, New York.

Pathak, R. K., and C. K. Chan (2005), Inter-particle and gas-particle interactions in sampling
artifacts of PM2.5 in filter-based samplers, Atmos. Environ., 39, 1 597-1607.

Pauluhn, J. (2005), Retrospective analysis of acute inhalation toxicity studies: Comparison of
actual concentrations by filter and cascade impactor analyses, Regul. Toxicol. Pharm., 42,
236-244.

Petrov, G. A. (1987), Sanation of working-conditions and environment for the process of
sulfuric acid manufacture from metallurgical gases (Russ), Gig Sanit, 70-71.

Possanzini, M., A. Febo, and A. Liberti (1983), New design of a high-performance denuder for
the sampling of atmospheric pollutants, Atmos. Environ., 1 7, 2605-2610.

Radoj evic, M. (1984), On the discrepancy between reported studies of the uncatalyzed aqueous
oxidation of SO2 by Oz, Environ. Technol. Lett., 5, 549-566.

Ridler, E. S. (1959), Chamber process for the manufacture of sulfuric acid, in The Manufacture
of Sulfuric Acid, edited by W. W. Dueker and J. R. West, Reinhold Publishing
Corporation, New York.

Sathiakumar, N., E. Delzell, Y. Amoateng-Adjepong, R. Larson, and P. Cole (1997),
Epidemiologic evidence on the relationship between mists containing sulfuric acid and
respiratory tract cancer, Crit. Rev. Toxicol., 27, 233-251.

Schlesinger, R. B. (1984), Comparative irritant potency of inhaled sulfate aerosols effects on
bronchial mucociliary clearance, Environ. Res., 34, 268-279.

Schlesinger, R. B. (1989), Factors affecting the response of lung clearance systems to acid
aerosols role of exposure concentration, exposure time, and relative acidity, Environ.
Health Perspect., 79, 121-126.










Schlesinger, R. B., and L. C. Chen (1994), Comparative biological potency of acidic sulfate
aerosols Implications for the interpretation of laboratory and field studies, Environ. Res.,
65, 69-85.

Schlesinger, R. B., L. C. Chen, I. Finkelstein, and J. T. Zelikoff (1990a), Comparative potency
of inhaled acidic sulfates speciation and the role of hydrogen ion, Environ. Res., 52, 210-
224.

Schlesinger, R. B., A. F. Gunnison, and J. T. Zelikoff (1990b), Modulation of pulmonary
eicosanoid metabolism following exposure to sulfuric acid, Fundam.ddd~~~~~ddddd~~~~ Appl. Toxicol., 15,
151-162.

Schroeter, L. C. (1963), Kinetics of air oxidation of sulfurous acid salts, J. Pharm. Sci., 52, 559-
563.

Scott, W. D., and P. V. Hobbs (1967), Formation of sulfate in water droplets, J. Atmos. Sci., 24,
54-57.

Seinfeld, J. H., and S. N. Pandis (1998), Thermodynamics of aerosols, in Atmospheric
Chemistry and Physics from Air Pollution to Climate Change, pp. 491-544, John Wiley,
New York.

Sioutas, C., P. Y. Wang, S. T. Ferguson, P. Koutrakis, and J. D. Mulik (1996), Laboratory and
field evaluation of an improved glass honeycomb denuder filter pack sampler, Atmos.
Environ., 30, 885-895.

Skytta, E. (1978), Tilasto tyoihygieenisista mittauksista v. 1971-1976 [Statistics of industrial
hygiene measurements in 1971-1976]. Institute of Occupational Health, Helsinki
(Finland) .

Stayner, L., A. B. Smith, G. Reeve, L. Blade, L. Elliott, R. Keenlyside, and W. Halperin (1985),
Proportionate mortality study of workers in the Garment Industry exposed to
formaldehyde, Am. J. Ind. M~ed., 7, 229-240.

Steenland, K. (1997), Laryngeal cancer incidence among workers exposed to acid mists (United
States), Cancer Causes Control, 8, 34-38.

Stephenson, F., M. Cassady, H. Donaldson, and T. Boyle (1977), Industrial Hygiene Survey,
CF Chemicals, Inc., Bartow, FL, August 9-12, 1976., National Institute for Occupational
Safety and Health, Cincinnati, OH.

Stephenson, F., M. Cassady, H. Donaldson, and T. Boyle (1977), Industrial Hygiene Survey,
IMC, Phosphate Chemical Complex, New Wales, Florida, National Institute for
Occupational Safety and Health, Cincinnati, OH.










Stratmann, H., and M. Buck (1965), Comparative measurements using silica-gel and TCM
procedures for determining-sulfur-dioxide in the atmosphere, Ai.r Water. Pollut., 9, 199-
218.

Swenberg, J. A., and R. O. Beauchamp (1997), A review of the chronic toxicity,
carcinogenicity, and possible mechanisms of action of inorganic acid mists in animals,
Crit. Rev. Toxicol., 27, 253-259.

Swietlicki, E., H. C. Hansson, B. Martinsson, B. Mentes, D. Orsini, B. Svenningsson, A.
Wiedensohler, M. Wendisch, S. Pahl, P. Winkler, R. N. Colvile, R. Gieray, J. Luttke, J.
Heintzenberg, J. N. Cape, K. J. Hargreaves, R. L. StoretonWest, K. Acker, W. Wieprecht,
A. Berner, Kruisz, C., M. C. Facchini, P. Laj, S. Fuzzi, B. Jones, and P. Nason (1997),
Source identification during the Great Dun Fell Cloud Experiment 1993, Atmos. Environ.,
31, 2441-2451.

Tadzhibaeva, N. S., and I. V. Gol'eva (1976), Industrial hygiene and condition of the upper
respiratory tract of people working in the production of superphosphate in Uzbekistan,
Russia, M~ed. Zh. Uzb., 4, 57-59.

Tanaka, S., K. Yamanaka, and Y. Hashimoto (1987), Measurement of concentration and
oxidation rate of S(IV) in rainwater in Yokohama, Japan, Am. Chem. Soc. Symp. Ser., 349,
158-169.

Tsai, C. J., C. H. Huang, and S. H. Wang (2001), Collection efficiency and capacity of three
samplers for acidic and basic gases, Environ. Sci. Technol., 35, 2572-2575.

Tsai, C. J., C. H. Huang, and H. H. Lu (2004), Adsorption capacity of a nylon filter of filter
pack system for HCI and HNO3 gaSes, Sep. Sci. Technol., 39, 629-643.

USDHHS (2005), Report on Carcinogens, Eleventh ed., U. S. Department of Health and Human
Services, Public Health Service, National Toxicology Program, Research Triangle Park,
NC.

USEPA (1995), Compilation of Air Pollutant Emissions Factors Volume 1: Stationary Point
and Area Sources, Fifth Edition 0I ithr Supplements, Document No. AP-42., U. S.
Environmental Protection Agency, Research Triangle Park, NC.

Wagner, J., and D. Leith (2001), Passive aerosol sampler. Part II: Wind tunnel experiments,
Aerosol Sci. Technol., 34, 193-201.

Wallingford, K. M., and E. M. Snyder (2001), Occupational exposures during the World Trade
Center disaster response, Toxicol. Ind. Health, 1 7, 247-253.

Watson, J. G., and J. C. Chow (2001), Ambient air sampling, in Aerosol M~easurement:
Principles, Techniques and Applications, edited by P. A. Baron and K. Willeke, pp. 821-
844, Wiley, New York.










Watson, J. G., G. Thurston, N. Frank, J. P. Lodge, R. W. Wiener, F. F. McElroy, M. T.
Kleinman, P. K. Mueller, A. C. Schmidt, F. W. Lipfert, R. J. Thompson, P. K. Dasgupta,
D. Marrack, R. A. Michaels, T. Moore, S. Penkala, I. Tombach, L. Vestman, T. Hauser,
and J. C. Chow (1995), 1995 Critical-review discussion measurement methods to
determine compliance with ambient air-quality standards for suspended particles, J. Air
Waste Manage. Assoc., 45, 666-684.

Weast, R. C. (1988), CRC Handbook of Chemistry and Physics, edited, CRC Press, Boca
Raton, Fla.

Wu, C. Y., and P. Biswas (1998), Particle growth by condensation in a system with limited
vapor, Aerosol Sci. Technol., 28, 1-20.

Yaday, J. S., and V. K. Kaushik (1996), Effect of sulphur dioxide exposure on human
chromosomes, Mutat. Res.-Envir. Muta., 359, 25-29.

Yao, X. H., T. Y. Ling, M. Fang, and C. K. Chan (2006), Comparison of thermodynamic
predictions for in situ pH in PM2.5, Atmos. Environ., 40, 2835-2844.

Yao, X. H., T. Y. Ling, M. Fang, and C. K. Chan (2007), Size dependence of in situ pH in
submicron atmospheric particles in Hong Kong, Atmos. Environ., 41, 3 82-3 93.

Yasyerli, S., G. Dogu, I. Ar, and T. Dogu (2001), Activities of copper oxide and Cu-V and Cu-
Mo mixed oxides for H2S removal in the presence and absence of hydrogen and
predictions of a deactivation model, Ind. Eng. Chem. Res., 40, 5206-5214.









BIOGRAPHICAL SKETCH

Yu-Mei Hsu was born in 1975 in Hsin-Chu, Taiwan. She received her B.S. degree in

environmental engineering in June 1999 at National Cheng-Kung University, Taiwan. She was

awarded the Phi Tau Phi Scholastic Honor which awarded the first ranking student among 53

students. She also earned her M.S. degree in environmental engineering sciences in June 2001 at

National Taiwan University, Taiwan. Her master's thesis, the Chloride Loss of Sea-Salt

Aerosols, was awarded the Outstanding Master' s Thesis Award from National Taiwan

University and National Science Council, Taiwan.

She j oined the research group of Dr. Chang-Yu Wu at the University of Florida in 2004

and started pursuing her Ph.D. degree in the Department of Environmental Engineering Sciences.

Her research focused on the sulfuric acid mist sampling at the phosphate fertilizer plants.

Yu-Mei Hsu was the vice-president, secretary, and webmaster of the student chapter of Air &

Waste Management Association (A&WMA) at the University of Florida from 2005 to 2007.

She was awarded the Axel Hendrickson Scholarship Award from Florida Section A&WMA in

2006, and also was awarded the AWMA Scholarship (2nd place in air quality) from A&WMA in

2007.





PAGE 1

1 EXPOSURE POTENTIAL OF SULFURIC AC ID MIST AT PHOSPHATE FERTILIZER FACILITIES By YU-MEI HSU 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 2008

PAGE 2

2 2008 Yu-Mei Hsu

PAGE 3

3 To my parents, sisters and brother for th eir constant love, understanding, and support.

PAGE 4

4 ACKNOWLEDGMENTS I am grateful for Dr. Chang-Yu Wu (my s upervisory committee chair) for his patience, guidance, and encouragement. I also sincer ely thank Dr. Chang-Yu Wu for giving me the opportunity to study at UF, and le tting me learn how to be a good professor and to be patient with others. He is not only my advisor, but also the paragon in my life. I would like to express my deep ly appreciation to Dr. Dale A. Lundgren for his invaluable research experience and his inspir ation. My grateful appreciation would also go to Dr. Jean M. Andino and Dr. Wesley E. Bolch for their wa rmth, kindness and their good-natured support. They have generously given their time and expertise to improve this dissertation. This study was funded by Florida Institute of Phosphate Research (FIPR). I would also like to express my gratitude to Dr. Brian K. Birk y, Research Director for Public Health, for his valuable guidance, advice and comments. I am grateful to Tom McNally, Robert Ammons and J. Wesley Nall from the Polk County Health Unit in Winter Haven, Fl orida for carrying out the pre-sa mpling and all persons at the phosphate fertilizer plants to he lp for the field sampling and th ey are Alan A. Pratt, Melody Foley, Martin St. John, Debra Waters, Paul D. Ho lewski, Tara Crews, Todd W Smith, and Foster Thorpe. I thank Dr. Eric Allen for his knowledgeable instruction, and Cheng-Chuan Wang, HsingWang Li, and Shu-Hau Hsu for assisting me in the field sampling, and Joshua Kollet and Katherine Wysocki for helping me with the lab experiments. Many thanks go to my good friends, Ying Li, Ji anmei Liu, Jin-Hwa Lee, Anadi Misra, and Ian Liu, for their patience, kindness, love and wa rmth. I thank my labmates, Alex Theodore, Jenkins Charles, Danielle Hall, Nathan Topha m, Charles Michael Jenkins, Myung-Heui Woo, Lindsey Riemenschneider, and Qi Zha ng, who kindly assisted my research.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES................................................................................................................ .........9 ABSTRACT....................................................................................................................... ............11 CHAPTER 1 INTRODUCTION................................................................................................................ ..13 Sulfuric Acid Mist and Its Health Effects...............................................................................13 Sulfuric Acid Regulations...................................................................................................... .14 Sulfuric Acid Mist in Manufacturing Facilities......................................................................14 Manufacturing Processes in Fertil izer Manufacturing Facilities............................................16 Sulfuric Acid Mist Measurement............................................................................................18 Phosphate Fertilizer Manufacture in NTP Report..................................................................19 Research Objectives............................................................................................................ ....19 2 CHEMICAL CHARACTERISTICS OF AEROSOL MISTS IN PHOSPHATE FERTILIZER MANUFACTURING FACILITIES...............................................................21 Background..................................................................................................................... ........21 Methods........................................................................................................................ ..........23 Sampling Locations.........................................................................................................23 Sampling and Analysis Methods.....................................................................................24 Results and Discussion......................................................................................................... ..26 Background Sites.............................................................................................................26 Mass Concentrations Measur ed in the Facilities.............................................................26 Ion Concentrations Measured in the Facilities................................................................27 Aerosol Acidity...............................................................................................................3 1 Summary........................................................................................................................ .........32 3 SIZE-RESOLVED SULFURIC ACID MI ST CONCENTRATIONS AT PHOSPHATE FERTILIZER MANUFACTURING FACILITIES IN FLORIDA........................................48 Background..................................................................................................................... ........48 Methods........................................................................................................................ ..........49 Sampling Sites................................................................................................................. 49 Sampling and Analysis Methods.....................................................................................50 Calculation of Fine Mode................................................................................................52 Calculation of Sulfuric Acid Mist Concentration............................................................52 Results and Discussion......................................................................................................... ..52

PAGE 6

6 Background Site..............................................................................................................52 Plants: Cascade Impactor Samples..................................................................................53 Attack tank area........................................................................................................54 Sulfuric acid pump tank area....................................................................................54 Belt or rotating table filter floor...............................................................................55 Sulfuric acid truck loading/unloading station..........................................................56 Granulator on a scrub day........................................................................................56 Plants: NIOSH Method Samples.....................................................................................57 Comparisons of the Results from the Ca scade Impactor and the NIOSH Method.........57 Comparisons of Sulfuric Acid Mist Concentrations with OSHA and ACGIH Regulations..................................................................................................................60 Summary........................................................................................................................ .........61 4 SIZE DISTRIBUTION, CHEMICAL COMPOSITION AND ACIDITY OF MIST AEROSOLS IN FERTILIZER MANUFAC TURING FACILITIES IN FLORIDA.............71 Background..................................................................................................................... ........71 Methods........................................................................................................................ ..........72 Sampling Sites................................................................................................................. 72 Sampling and Analysis Methods.....................................................................................73 Aerosol Thermodynamic Model......................................................................................74 Results and Discussion......................................................................................................... ..77 Aerosol Chemical Species...............................................................................................77 Sulfuric aid pump tank areas....................................................................................77 Product filter floors..................................................................................................79 Attack tank areas......................................................................................................80 Granulator on a scrub day........................................................................................81 Aerosol Acidity...............................................................................................................8 1 Charge balance method............................................................................................81 Aerosol thermodynamic model................................................................................82 Summary........................................................................................................................ .........86 5 POSITIVE SULFATE ARTIFA CT FORMATION FROM SO2 ADSORPTION IN THE SILICA GEL SAMPLER USED IN NIOSH METHOD 7903...................................100 Background..................................................................................................................... ......100 Methods........................................................................................................................ ........102 Sulfur(IV) Oxidation.....................................................................................................102 Sulfur Dioxide Adsorption............................................................................................103 Results and Discussion.........................................................................................................104 Sulfur(IV) Oxidation.....................................................................................................104 Sulfur Dioxide Adsorption............................................................................................106 Sulfur dioxide concentration..................................................................................106 Sampling flow rate.................................................................................................107 Sampling time........................................................................................................108 Sulfur Dioxide Adsorption Model.................................................................................109 Summary........................................................................................................................ .......110

PAGE 7

7 6 MINIMIZATION OF ARTIFACTS IN SU LFURIC ACID MIST MEASUREMENT USING NIOSH METHOD 7903..........................................................................................119 Background..................................................................................................................... ......119 Methods........................................................................................................................ ........121 Field Sampling...............................................................................................................121 Deactivation Model.......................................................................................................122 Sulfur Dioxide Adsorption............................................................................................123 Results and Discussion.........................................................................................................124 Field Sampling...............................................................................................................124 Collection efficiency and concentration of SO2.....................................................124 Ratio of S-SO4 2-/ S-SO2..........................................................................................124 Aerosol loss of HDS...............................................................................................125 Sulfur Dioxide Adsorption............................................................................................126 Residual Sulfate in Silica Gel Tube...............................................................................127 Minimization of Artifact Sulfate...................................................................................128 Aspiration Efficiency.....................................................................................................130 Sulfate Mass Balance....................................................................................................131 Summary........................................................................................................................ .......132 7 CONCLUSIONS.................................................................................................................. 144 Conclusion 1................................................................................................................... ......144 Conclusion 2................................................................................................................... ......145 Conclusion 3................................................................................................................... ......145 Conclusion 4................................................................................................................... ......146 Conclusion 5................................................................................................................... ......146 LIST OF REFERENCES.............................................................................................................146 BIOGRAPHICAL SKETCH.......................................................................................................155

PAGE 8

8 LIST OF TABLES Table page 2-1 Sampling locations at phosphate fertilizer plants in Florida..............................................34 2-2 Analysis conditions for soluble ions..................................................................................35 2-3 Detection limit of ion chromatography (ICS 1500)...........................................................36 2-4 Median concentration ( g/m3) of ion species at background sites....................................37 2-5 Median concentration ( g/m3) of aerosol chemical compos ition at the granulator on a scrub day...................................................................................................................... ......38 2-6 Statistics of hydrogen ion concentrations (eq/m3) at each location.................................39 3-1 PM23, PM10 and PM2.5 mass and sulfuric acid concentra tions at the attack tank areas.....63 3-2 PM23, PM10 and PM2.5 mass and sulfuric acid concentrations at the sulfuric acid pump tank areas................................................................................................................ .64 3-3 Mass, sulfuric acid c oncentrations and sulfate/massa ratios of the impactor samples at the sulfuric acid pump tank areas.......................................................................................65 3-4 Statistics of R23, R10 and R2.5 at five types of sampling location......................................66 3-5 Sulfuric acid concentrations and the rati os measured at two flow rates at the rotating table filter floors using NIOSH Method 7903...................................................................67 4-1 Equilibrium rela tions and constants...................................................................................88 4-2 Median ion species concentrations of cascade impactor samples collected at the granulator on a scrub day (g/m3).....................................................................................90 4-3 Aerosol deposition fractions for 3 cases............................................................................91 5-1 Experimental conditions of SO2 adsorption.....................................................................111 5-2 Rate constants for uncatalyzed ox idation reaction of sulfur(IV) by oxygen...................112 5-3 Rate parameters obtained using Equation 5-6.................................................................113 6-1 Statistics of SO2 concentrations (ppm)............................................................................133 6-2 Mean and standard deviation of the su lfate loss to the total sulfate concentration..........134 6-3 Statistical results of the residual sulfate concentra tions of silica gel tubes.....................135 6-4 Relative error of 4 samplers.............................................................................................136

PAGE 9

9 LIST OF FIGURES Figure page 1-1 Monoammonium phosphate and diamm onium phosphate manufacturing process...........20 2-1 Manufacturing processes at fertilizer facilities..................................................................40 2-2 Geographic locations of sampling sites.............................................................................41 2-3 Fine mode and coarse mode aerosol mass concentrations at various locations.................42 2-4 Aerosol chemical species at the sulfuric acid pump tank area...........................................43 2-5 Aerosol chemical species at the attack tank area...............................................................44 2-6 Aerosol chemical species at th e rotating table/belt filter floor..........................................45 2-7 Aerosol chemical species at the sulfur ic acid truck loading/unloading station. ...............46 2-8 Relationship of cation equivalent weight and anion equivalent weight............................47 3-1 Sulfuric acid concentrati ons at 5 types of locations...........................................................68 3-2 Sulfuric acid mist and aerosol mass size distributions.......................................................69 3-3 Comparison of PM23 sulfuric acid concentrations from the cascade impactor and total sulfuric acid concentrations from the NIOSH method......................................................70 4-1 Sulfuric acid, phosphoric acid and fl uoride concentrations at all locations......................92 4-2 Relation between the major cations (ammonium and calcium) and sulfate concentrations at the sulfuric acid pump tank areas..........................................................93 4-3 Aerosol size distributions at the product filter floors.........................................................94 4-4 Particulate fluorid e size distribution at the attack tank areas.............................................95 4-5 Relation between ammonium and fluoride concentrations at the attack tank areas..........96 4-6 Relationship of cation equivalent weight and anion equivalent weight............................97 4-7 Aerosol hydrogen ion con centration size distribution.......................................................98 5-1 Experimental setup for sulfur dioxide adsorption............................................................114 5-2 Sulfur(IV) oxidation under four conditions.....................................................................115 5-3 Artifact sulfate concentr ations and time-weighted colle ction percentages (TWCPs).....116

PAGE 10

10 5-4 Collection index (CI) at four flow rates (Lpm)................................................................117 5-5 Relationship between sulfate concentrations from the measurement versus from the model.......................................................................................................................... ......118 6-1 Three sampling trains..................................................................................................... ..137 6-2 S-SO2/S-SO4 2as a function of SO2 concentration..........................................................138 6-3 Artifact sulfate concentr ation as the function of SO2 concentration................................139 6-4 Comparison of the sulfate concentrat ions from the CI, SG SGHAC and SGHBC........140 6-5 Aspiration efficiency of three samplers...........................................................................141 6-6 Sulfate mass balance between SG and SGHAC/SGHBC................................................142

PAGE 11

11 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EXPOSURE POTENTIAL OF SULFURIC AC ID MIST AT PHOSPHATE FERTILIZER FACILITIES By Yu-Mei Hsu August 2008 Chair: Chang-Yu Wu Major: Environmental Engineering Sciences Strong inorganic acid mists c ontaining sulfuric acid (H2SO4) were identified as a known human carcinogen in a recent report on carcinog ens by the National Toxicology Program where phosphate fertilizer manufacture was listed as one of many occupationa l exposures to strong acids. To properly assess th e occupational exposure to H2SO4 mists in modern facilities, the objective of this study was to characterize the true H2SO4 mist concentration levels. Three sets of experiments were conducted. Firstly, field sampling using dichotomous samplers, silica gel tubes and cascade im pactors was conducted to collect the PM2.5/PM10 H2SO4 mist concentration, total H2SO4 mist concentration, and size-resolved H2SO4 mist concentration, respectively, at phosphate fertilizer plants. The H2SO4 concentrations were found to vary significantly among these plants with H2SO4 pump tank areas having the highest concentration level. When high aerosol mass concentrations were observed, the H2SO4 mist had its mode size in the 3.8 m range that would deposit in the upper respiratory region. Secondly, SO2 adsorption and sulfur(IV) oxidati on were investigated under various sampling times, SO2 concentrations and sampling flowrates. Experimental results verified that the collecting medium can adsorb SO2 gas and the extraction pro cedure of NIOSH Method 7903

PAGE 12

12 aids the transformation of SO2 into sulfate to cause a positive artifact. The experimental data were also fitted into a deactivation model for estimating the artifact sulfate concentration. Thirdly, a honeycomb denuder system and th e deactivation model were applied to minimize the artifact sulfate of NIOSH Met hod 7903 in a field sampling campaign. Both the system and the model were shown to effectivel y reduce the artifact su lfate concentration. However, the concentration thus determined was still higher than that measured by a cascade impactor which had no artifact. One possible re ason is the residual sulfate in the collecting medium.

PAGE 13

13 CHAPTER 1 INTRODUCTION Sulfuric Acid Mist and Its Health Effects Strong inorganic acid mists c ontaining sulfuric acid (H2SO4) have been reported to correlate with the incidence of lung and laryngeal cancers in humans [ Blair and Kazerouni 1997; Sathiakumar et al. 1997; Steenland 1997] and are identifie d as a known human carcinogen as reported by the U.S. National Toxicology Program (NTP) [ USDHHS 2005]. Sulfuric acid is typically presen t in the air as a mist. Its chem ical characteristics include low volatility, high acidity, high reactivit y, high corrosivity, and high affinity for water. Sulfuric acid also irritates the human airways, and this irritation may potentia lly damage pulmonary epithelium, causing subsequent carcinogenic e ffects from other inhaled substances. Although the carcinogenetic mechanism of sulfuric acid mist is not known [ Blair and Kazerouni 1997], a low pH environment has been re ported to induce chromosomal aberrations, gene mutation and cell transformation. Depurinati on, which is the removal of a purine (adenine or guanine) from a DNA molecule, and deaminati on of cytidine in DNA molecules, which is the replacement of the amine functional group by the ke tone group, has been shown to be enhanced by exposure to sulfuric acid mist [ Swenberg and Beauchamp 1997]. In a case study of workers engaged in the manufacture of sulfuric aci d, significant increases were observed in the incidences of genotoxic effects, including sister chromatid exchange (an exchange of segments between the sister chromatids of a chromoso me), micronucleus formation, and chromosomal aberration in peripheral lymphocytes [ Meng et al. 1995; Meng and Zhang 1997]. As exposure to chemical fumes is suspected to be one of the reasons for lung cancer formation, information regarding the concentration le vels of the acid mists is of seminal importance.

PAGE 14

14 Sulfuric Acid Regulations The current Occupational Safety & Health Administration (OSHA) 8-hour time-weighted average (TWA) of permissible exposure level (PEL) for sulfuric acid mist is currently set at 1 mg/m3 with its 15-min short-term exposure level (STEL) set at 3 mg/m3 [ CFR ]. It is well known that the deposition of an aerosol in the respir atory system depends on its aerodynamic behavior. Aerosols with an aerodynamic diameter larger th an 1 m mainly deposit in the extrathoracic airways, and ultrafine aerosols w ith particle sizes less than 0.01 m predominantly deposit in the tracheobronchial region by Browni an motion, whereas aerosols of intermediate diameter deposit in the alveolar regions [ Hinds 1999]. In considering the effect s of aerosol size, the American Conference of Governmental Indus trial Hygienists (ACG IH) has adopted a threshold limit valuetime-weighted average (TLV-TWA) of 0.2 mg/m3 for the thoracic particulate fraction of sulfuric acid mist [ ACGIH 2004]. Sulfuric Acid Mist in Ma nufacturing Facilities Several studies have reported sulfuric acid co ncentrations in worker environments. In a sulfuric acid plant in Sweden [ Englander et al. 1988], the concentrati on was at 0.1 to 3.1 mg/m3 for samples taken in 1979 1980. In a Finnish sulfuric acid plant [ Skytt 1978], the concentration was measured with in the range of 0 to 1.7 mg/m3. The concentration in a Russian plant [ Petrov 1987] was found to be higher, ranging from 1.8 to 4.6 mg/m3. Samples taken in a sulfuric acid plant at a U.S. copper smelter in 1984 showed lower concen trations ranging form 0.15 to 0.24 mg/m3. Limited information regarding the mi st size distribution in sulfuric acid production plants is available [ Muller 1992]. It is reported that th e size of mist particles ranged from about 0.1 m to greater than 10 m. There are no current reports characterizi ng sulfuric acid mist in the fertilizer manufacturing industry. Studies c ited in the NTP report were all carried out a few decades ago.

PAGE 15

15 Increased rates of lung cancer in correlation with exposure to chemical fumes in some of these studies are the reason for recent con cerns. A historical cohort study [ Hagmar et al. 1991] on workers employed in a Swedish fertilizer factor y was carried out for two cohorts (1236 men in 1906 62 and 2131 men in 1963 85). Significant excesses were found for cancers of the respiratory tract and lung cancers. The result s from U.S. studies, on the other hand, showed different trends. A cohort st udy was carried out for men who worked in Florida phosphate processing facilities during 1949 1978 [ Checkoway et al. 1985a; 1985b]. Lung cancer mortality was higher than that found in for th e entire U.S. population al though this rate was insignificant when compared to the Florida popula tion. Internal comparison for mortality rates from lung cancer was also conducted for the sa me population. For workers in sulfuric and phosphoric acid production, no consis tent increase in relative risk for lung cancer was found. National Institute for Occupational Safety a nd Health (NIOSH) researchers conducted an investigation at a phosphate fertilizer production facility in Polk County [ Stayner et al. 1985]. Three major acids identified by the personal and area samplings were fluorides (mean: 3.39 mg/m3), sulfuric acid (mean: 0.11 mg/m3) and phosphoric acid (mean: 0.25 mg/m3). A total of 3199 subjects who had worked at the plant from 1953 1976 were studied. Overall mortality and morbidity from all cancers were lower than expe cted, and the risk for lung cancers increased only slightly. Another study [ Block et al. 1988] carried out for male workers in another Florida phosphate company between 1950 and 1979 showed a significant excess of lung cancer deaths among white workers in comparison to both U.S. ra tes and Florida rates. However, when an internal comparison of job categories was made with respect to lung cancer, no increase was found for workers exposed to chemical fumes (sul furic acid, sulfur dioxide and fluorides).

PAGE 16

16 Limited information on acid mist concentrations in the phosphate fert ilizer manufacturing industry is available. In U.S. facilities, the m ean sulfuric acid mist concentration ranges from 0.07 to 0.571 mg/m3 [ Apol et al. 1987; Cassady et al. 1975; Stephenson et al. 1977]. The concentration reported in a Finnish study [ FIOH 1990] was 8.3 mg/m3 (1951 data). A Russian study [ Tadzhibaeva and Gol'eva 1976] also reported a high sulf uric acid mist concentration, 2.7 to 9.2 mg/m3. Compared to foreign facilities, sulfuric mist concentrations in U.S. plants were low, and the OSHA standard of 1 mg/m3 was met. Size distribution related to sulfuric acid mist emission in the fertilizer ma nufacture industry, however, is not available. Manufacturing Processes in Fertilizer Manufacturing Facilities Phosphate rock, which is the main useful pr oduct of phosphate ore, consists of calcium phosphate mineral apatite (Ca5(PO4)3(OH, F, Cl)) with gangue cons tituents including silica (SiO2), fluoride (F), calcite (CaCO3), dolomite (CaMg(CO3)2), clay, and iron-aluminum oxide (Fe2O3, Al2O3). Several chemical formulas are commonly used for phosphate rock which includes fluorapatite (Ca5(PO4)3F), chlorapatite (Ca5(PO4)3Cl) and hydroxyapatite (Ca5(PO4)3(OH)). The United States is the principa l producer of chemical fertilizer using phosphate rock [ Hodge 1994]. The final products from phosphate fertilizer plants in Florida are mainly diammonium phosphate (DAP), monoammonium phosphate (MAP), and concentrated sulfuric acid solution. Wet process using sulfuric acid to react with phosphate rock is commonly used by the phosphate fertilizer industry in Florida to produce phosphor ic acid. Figure 1-1 shows the manufacturing process flow, which can be divided into three stages. In the first stage, phosphate ro ck reacts with sulfuric acid at the attack tank to produce phosphoric acid with 30-55 wt% P2O5. The simplified reactions of th e wet process at the attack tank (also called as reactor) are shown in Reaction 1-1 [ Palm 1992]. Sulfuric acid of 93% is fed

PAGE 17

17 into the attack tank. The violen t reaction between phosphate rock and sulfuric acid causes heat release in the form of vapor, which is evacuat ed from the attack tank with other gaseous effluents. A cooling system is needed to maintain the temperature at 70 80 C in this process [ Becker 1989]. HF PO H O H CaSO O H SO H F PO Ca 2 6 2 10 20 104 3 2 4 2 4 2 2 6 4 10 (1-1) At the second stage, the reaction product from the first stage passes through a rotating table filter or belt filter to separate phosphoric acid from its byproduct phosphogypsum (calcium sulfate dihydrate, CaSO4 2H2O). A cooling system is applied to maintain the temperature at 70 75 C in this process [ Becker 1989]. Some H3PO4 is lost to phosphogypsum from this process. At the third stage, the fina l product, MAP or DAP, is pr oduced by reacting phosphoric acid of 30-55% P2O5 (by weight) with ammonia at the granul ator, as shown in Re action 1-2 and 1-3. When poor product quality is detected, weak sulf uric acid is used to scrub the granulator. 4 2 4 3 4 3PO H NH NH PO H (1-2) 4 2 4 3 4 32 HPO NH NH PO H (1-3) The manufacturing process of sulfuric acid can also be divided into three stages: (1) The production of sulfuric acid starts from the combustion of elemental sulfur (S) to produce sulfur dioxide (SO2). Elemental sulfur is pumped in to the sulfur burner and is burned with dry combustion air to form SO2. Sulfur dioxide and excess air leave the burner at 700 1,000 C which need to be cooled to 425 C to protect the convert er in the next stage [ Muller 1992]. (2) At the second stage, SO2 is subsequently oxidized to form sulfur trioxide (SO3) when passing through a series of catalyt ic converters. The reaction pr oceeds as Reaction 1-4 which is

PAGE 18

18 an exothermic reaction. This equilibrium equation is controlle d by the concentrations of SO2, SO3 and temperature. High SO3 concentration and temperature can favor the reverse reaction. The temperature is 180 250 C when gas leaves the converter [ Muller 1992]. ) ( 3 2 ) ( 22 1g gSO O SO (1-4) H 41,400 Btu/lb mol (3) Sulfur trioxide can quickly combine with water vapor to produce sulfuric acid at the sulfuric acid pump tank. The reaction is shown as Reaction 1-5, which is also an exothermic reaction [ Ridler 1959]. ) ( 4 2 2 ) ( 3 l g gSO H O H SO (1-5) The produced sulfuric acid is stored in tanks. Some plants produce more sulfuric acid than needed and sell excess sulfuric acid solution to other companies. Trucks are used for the transportation of sulfuric acid solution. The lo ading and unloading of sulfuric acid solution is carried out through a nozzle at a truck station. Sulfuric Acid Mist Measurement Sulfuric acid exists in the mist form which can be collected by filtration. NIOSH Method 7903 is an OSHA approved method which employs a si lica gel tube to colle ct acid mist. The silica gel tube consists of one section of glass fibe r filter plug and two sections of silica gel. The glass fiber filter and the silica gel are designed to collect aerosols and acid gases, respectively. NIOSH Method 7903 is the method commonly applie d for personal sampling in the workplace due to its convenience. However, both the gl ass fiber filter and silica gel can adsorb SO2 [Chow, 1995; Lee and Mukund, 2001] that will lead to an overestimate of sulfate.

PAGE 19

19 Phosphate Fertilizer Manufacture in NTP Report Although phosphate fertilizer manufactu re was listed in the NTP report [USDHHS, 2005] as one of many occupational exposures to strong inorganic acids, the oc cupational exposures to inorganic acid mists containing sulf uric acid existing at levels equal to or greater than the PEL used by the International Agency for Research on Cancer (IARC) are based on results obtained during the period 1951 to 1976. In addition, all of the results greater than the PEL for sulfuric acid are from outside the U.S. [USDHHS, 2005]. The significant improvement of health and safety measures in the US fertilizer industry in th e past decades is expected to have significantly lowered current levels. Excessive respiratory protection may be co stly and stressful and still not provide any beneficial reduction in exposure. Thus, characterizati on of the true exposure level at modern facilities is a necessary step to the establishment of the best policy for worker protection. Research Objectives Five objectives were set in this doctoral re search study to accurately characterize sulfuric acid mist at phosphate fertilizer facilities. The first objective is to characterize the major water soluble ionic species of PM2.5 and PM2.5-10 at phosphate fertilizer f acilities. The location with high chemical species concentrations can be iden tified as well. The characterization can be applied for the establishment of the best policy for worker protection. The second objective is to determine the sulf uric acid mist concen trations with sizeresolved information by a cascade impactor and the total sulfuric acid mist concentration using NIOSH Method 7903. The study also seeks to determine the correlation between these two samplers at the phosphate fertilizer facilities. The third objective is to determine the chemi cal characteristics of mist aerosols in the current phosphate facilities with size-resolved information and to estimate the aerosol hydrogen ion concentration at the phospha te fertilizer facilities us ing a thermodynamic model.

PAGE 20

20 The fourth objective is to verify and quantify the effect of SO2 interference on the artifact sulfate in NIOSH Method 7903. The oxida tion of sulfur(IV) in to sulfate and SO2 adsorption following the NIOSH protoc ol were also investig ated in this study. The fifth objective is to investigate the effectiveness of two methods, a deactivation model and a honeycomb denuder system, to minimize the artifact sulfate in a field sampling at the phosphate fertilizer facilities.

PAGE 21

21 Figure 1-1. Monoammonium phos phate and diammonium phos phate manufacturing process

PAGE 22

22 CHAPTER 2 CHEMICAL CHARACTERISTICS OF AEROSO L MISTS IN PHOSPHATE FERTILIZER MANUFACTURING FACILITIES Background Phosphate products are widely used around the world for fertilizer (90%), detergents (4.5%), animal feed (3.3%), a nd food and beverages (0.7%) [Becker, 1989]. The United States is the second largest producer of phospha te fertilizers in the world [Bhaskaran et al., 2004]. The manufacturing process flow of phosphate fertilizer is shown in Figure 2-1. In Florida, phosphoric acid is usually produced by a wet process by reacting H2SO4 with naturally occurring phosphate rock in a reactor that is referre d to in the industry as the attack tank. Phosphogypsum is a byproduct of this process. Th e simplified reactions of the wet process are as Reaction 2-1 [Becker, 1989]: HF PO H O H CaSO O H SO H F PO Ca2 6 2 10 20 104 3 2 4 2 4 2 2 6 4 10 (2-1) Rotating table filters and belt filters are used to separate phosphoric acid and phosphogypsum. Phosphoric acid of 30-55% P2O5 (by weight) reacts wi th ammonia to produce MAP or DAP in the granulator, as sh own in Reaction 2-2 and Reaction 2-3 [Hodge and Popovici, 1994]. 4 2 4 3 4 3PO H NH NH PO H (2-2) 4 2 4 3 4 32HPO NH NH PO H (2-3) In these facilities, H2SO4 used for digestion of phosphate ores is produced by sulfur combustion [Hodge and Popovici, 1994]. The sulfuric acid pr oduction process initiates in a sulfur burner. The resulting com bustion gas consists primarily of SO2 that is routed to a series of Reprinted with permission from Hsu, Y.-M., Wu, C.-Y ., Lundgren, D. A., Nall, J. W., Birky, B. K., 2007. Chemical Characteristics of Aerosol Mists in Phosphate Fertilizer Manufacturing Facilities. J. Occup. Environ. Hyg 4, 17-25.

PAGE 23

23 catalytic converters to transform the sulfur dioxide into SO3. By mixing with water, sulfur trioxide quickly forms sulfuric acid that is moved from pump tanks to storage tanks for use in the production of phosphoric acid. Some facilities produ ce more sulfuric acid than needed and sell the excess to others. Reduced pH environments are known to enha nce the depurination (i.e. removing a purine (adenine or guanine) from a DNA molecule) rate of DNA and the deamination (that is, replacing the amine functional group by the ke tone group) rate of cytidine [USDHHS, 2005], which can cause DNA damage or mutation. Sulfuric acid also irritates the human airway, and this irritation may potentially damage the epithelium, causing subsequent carcinogen ic effects of other substances [ACGIH, 2004]. Phosphate fertilizer manufacture was listed in the report as one of many occupational exposures to strong inorganic ac ids. However, the occupation al exposures to inorganic acid mists containing sulfuric acid existing at levels equal to or greater than PEL used by the IARC are based on results obtained during 1951-1976 [1992]. In addition, all of the results used by the IARC greater than the PEL for sulfuric acid were from outside the United States [Tadzhibaeva and Gol'eva, 1976]. In U.S. facilities, the sulfuric aci d mist mean concentration ranged from 0.07 to 0.57 mg/m3 [Apol et al., 1987; Cassady et al., 1975; Stephenson et al., 1977]. The significant improvement of environmental, health, and safety measures in the U.S. fertilizer industry in the past is expected to ha ve greatly lowered current levels. The objective of this chapter was to charac terize the thoracic pa rticulate fraction of chemical species concentration leve l at phosphate fertilizer facilities in Florida. The chemical species concentration level woul d be applied to estimate the required sampling time for a cascade impactor sampling. The information is critical for health risk assessment, is useful in identifying

PAGE 24

24 the key sources. Such characterization is a necess ary step in the establishment of the best policy for worker protection. Methods Sampling Locations Due to the different manufacturing process designs among these phospha te fertilizer plants, the aerosol emission sources at each plant are differe nt. Two to five locations at each plant were selected where sulfuric acid mists may exist. Th ese locations included the top of sulfuric acid pump tank, attack tank (reactor), filtration floor, sulfuric acid truck lo ading/unloading station, and granulator on a scrub day (Figure 2-1 and Ta ble 2-1). In total, there were 24 sampling locations in eight plants and two background locations included in this study. The geographic locations of these sampling sites are shown in Figure 2-2. Winter Haven, FL and Gainesville, FL, were employed as the backgr ound locations. Gainesville is located between the plant in north FL and the plants in central FL. Winter Have n is located at the east of the plants in central FL. The distance between Gainesville and the ne arest plant in north FL is approximately 60 miles; the distance between Winter Haven and the nearest plant in central FL is approximately 20 miles. Three samples were obtained at ea ch location. Due to the low particulate concentration at the back ground locations, sampling was carried out for 24 h. The sulfuric acid pump tank is where the newly produced sulfuric acid is distributed to sulfuric acid storage tanks. The leakage of SO3 from ducting can bind with moisture to form fine aerosols. Gas duct leaks in the sulfuric acid production plant are normally repaired quickly and are extremely difficult to sample. The sampling location chosen in the sulfuric acid production plants was on top of the sulfuric acid pump ta nk because the pump tank is vented to atmosphere and has a large throughput.

PAGE 25

25 The attack tank is where sulfuric acid reac ts with phosphate rock. The strong reaction between sulfuric acid and phosphate rock causes th e production of gaseous species that attach to existing aerosols or form aerosols by reacting w ith other gaseous species. Therefore, in the phosphoric acid production plants, sampling locations were chosen near the attack tanks. Two different types of product fi lters are used in these plants including a belt filter and a rotating table (commonly the Bird type) filter wher e the reaction products are filtered and washed under vacuum to separate phosphoric acid and phosphogypsum. The rotating table filter is a circular table with pie-shaped filter pans th at rotate in a circular motion while filtering out the phosphogypsum. The individual pan tilts and dumps the washed phosphogypsum into a hopper for pumping to a phosphogypsum storage mound. The belt filter is a continuous belt that washes the phosphogypsum at different sections along the belt and dumps the gypsum at the end of the belt into a hopper for dist ribution to the phosphogypsum stor age area. Sampling locations were selected adjacent to both types of filters. The sulfuric acid truck station (an outdoor location) is where tr ucks load or unload sulfuric acid. Sampling was conducted near the loading nozzle as this loca tion is where sulfuric acid is most likely to enter the atmosphere. The last sampling location was at the granulator on a scrub day when the product line is shut down to clean the piping and ductwork by spraying a mixture of process water and small amounts of sulfuric acid. This solution is recirculated from open-top tanks during the scrub cycle. Sampling was conducte d adjacent to these tanks. Sampling and Analysis Methods Dichotomous samplers (model SA241 CUM; Anderson Instrume nts, Atlanta, GA) were used to separate aerosols into two sizes: fine mode (aerodynamic diameter smaller than 2.5 m

PAGE 26

26 or PM2.5) and coarse mode (aerodynamic di ameter ranges from 2.5 to 10 m or PM2.5-10). The reason for using dichotomous samplers is because PM10 curve is similar to the thoracic fraction. The sampling flow rates were 1.67 and 15 Lpm for the fine and coarse mode, respectively. The sampling time in the production facilities was 12 h. Three samples were obtained at each location. The total sample numbers (plants) at th e pump tank area, rotating table/belt filter floor, attack tank area, truck station, a nd the granulator (on a scrub day) were 27(8 ), 27(8), 6(2), 9(3), and 2(1), respectively. Zefluor, or backed Teflon, membrane filters (P5PJ001, 8 10 inches, 2.0 m, Pall Corp., Ann Arbor, MI) were used to collect the aerosol. Zefluor membrane filters provide high collection efficiency and low r eactivity with acidic gases [Baron and Willeke, 2001; Chow, 1995]. A microbalance (model MC 210 S, Sartor ius Corp., Edgewood, N.Y.; readability 10 g) was used for weighing the particle mass. Filters were placed in a desiccator at room temperature for preand post-conditioning for 24 h before weighing the filters. The vapor pressures of sulfuric acid mist and phosphoric acid are low and they remain in the mist form at this condition. An ion chromatography (IC) system (Dionex ICS 1500; Dionex Corp., Atlanta, GA) was used to analyze the soluble ions of the samples. The analysis conditions are shown in Table 2-2. Fluoride, chloride, nitrate, su lfate, phosphate, sodium, potassium, calcium, and magnesium were analyzed and the ICs detection limit fo r each ion is listed in Table 2-3. Using this analysis method, it is impossible to distinguish whether th e sulfate measured is from sulfuric acid or other sulfat es (e.g., ammonium sulfate, calci um sulfate). Consequently, all particulate sulfate was assumed to be sulfuric acid mist in this study [NIOSH, 1994]. The conversion of sulfuric acid mist concentrati on from sulfate concentration follows Equation 2-4 listed below:

PAGE 27

27 323 24 244 2 measured by IC 498(M.W. of HSO) HSO g/m g/m 96(M.W. of SO) SO (2-4) Results and Discussion Background Sites Table 2-4 shows the median ion species concen trations at the two background sites. The median PM2.5 and PM2.5-10 concentrations (N = 3) in Gainesville were 15.0 and 11.2 g/m3, and the maximum PM10 concentration was 29.7 g/m3. The median PM2.5 and PM2.5-10 concentrations (N = 3) in Wi nter Haven were 26.2 and 13.7 g/m3, and the maximum PM10 (the sum of PM2.5 and PM2.5-10) concentration was 50.9 g/m3. The mass concentrations in Winter Haven were slightly higher than in Gainesville. The mass concentrations of the fine mode were higher than the coarse mode. Overall, the mass c oncentrations at both loca tions were not high. The major species of the PM2.5 in Gainesville were sulfate (median: 5.05 g/m3) and ammonium (median: 4.24 g/m3), and the concentrations of all other species were low. The major species in Winter Haven was ammonium. Th e median concentration was 5.10 g/m3 for PM2.5 and 7.78 g/m3 for PM2.5-10. Phosphate concentrations were lower than the detection limit at both locations; however, low fluoride concentrations were observed in Winter Haven (0.0860.149 g/m3). The major fluoride emission source is fe rtilizer facilities in central Florida [ USEPA 1995]. Hence, the particulate fluoride might come from these facilities. Mass Concentrations Measured in the Facilities The aerosol mass concentrations at 24 locati ons are summarized in Figure 2-3 with their maximum, minimum, mean, and median at each type of location. As shown, the mass concentration varied significantly from one location to the other and within each type of location. There are various factors that may affect the concentration level, such as production activity

PAGE 28

28 level and types of processes. Weather and ambi ent aerosols may also affect outdoor locations like the sulfuric acid pump tank and the sulfuric acid truck station. The rotating table/belt filter floor had a high aerosol mass concentration due to mechanical agitation and gas species condensation. The medi an values (range) of the fine mode and the coarse mode were 77.8 (27.8 381) and 79.5 (25.7 338) g/m3, respectively. Among the various locations, the attack tank had the highest fine mode aerosol mass concentration. The median value (range) of the fine mode was 389 (98.6 894) g/m3 and its coarse mode mass concentration (range) was 58.1 (13.6 266) g/m3. At the attack tank, the strong reaction of sulfuric acid and phosphate rock re leases heat and some species th at have high vapor pressures. In general, the attack tank and the rotating ta ble/belt filter floor are the two major emission sources of aerosol in these faci lities whereas the aerosol mass c oncentrations are lower at the sulfuric acid pump tank, the truck loading/unloa ding station and granulat or on a scrub day. Ion Concentrations Measured in the Facilities The results of ion analysis are presented in Fi gure 2-4 to Figure 2-7 for the sulfuric acid pump tank area, the attack tank area, the filter floor, and the truck station, respectively. At the sulfuric acid pump tank area (F igure 2-4), sulfate was the dom inant species, and the median (minimum and maximum) concentrations were 15.3 (1.33 and 77.0) and 5.1 (0.501 and 104) g/m3 for the fine mode and the coarse mode, respectively. The maximum PM10 sulfate concentration, 181 g/m3, is equivalent to a maximum sulfuric acid mist concentration of 185 g/m3, which is lower than 0.2 mg/m3, the TLV-TWA of the thorac ic particulate fraction of sulfuric acid recommended by ACGIH. In the sulfuric acid ma nufacturing process, SO3 is present in the ductwork of the plant before being absorbed into the acid stream. Any leak from the duct can release SO3 that can

PAGE 29

29 readily react with moisture in the air to produce H2SO4 mist [ Finlayson-Pitts and Pitts 2000]. The temperature at the pump tank is about 110 C, which is not high enough to evaporate H2SO4 (boiling point around 330 C) [ Weast 1988]. The leakage of SO3 from surrounding ductwork is likely to be the main source that causes higher relative sulfuric aci d mist concentrations at the sulfuric acid pump tank area. In addition to the SO3 emission rate, the othe r important factor that can influence the aerosol size of the mist is hum idity. High humidity assists the condensational growth of sulfuric acid mist, whereas a high emissi on rate aids the nucleati on of ultrafine sulfuric acid mist. Ammonium was also the dominant sp ecies, and the median (minimum and maximum) concentrations were 8.7 (1.40 and 22.1) and 2.5 (0.428 and 25.9) g/m3 for the fine mode and the coarse mode, respectively. The fast reaction of ammonia gas and sulfur ic acid mist to form ammonium sulfate ((NH4)2SO4) is likely the mechanism responsible fo r the ammonium concentration. Although this location is not expected to have amm onia gas emissions, ammonia comes from other locations in these plants and also commonly exists in ambient air. The presence of ammonia gas can neutralize the acidity of sulfuric acid or aci dic aerosol. Regarding other ions such as fluoride, chloride, phosphate, sodium, pota ssium, magnesium and calcium, these outdoor locations have generally low concentrations. At the attack tank area (Figure 2-5), fluorid e was the dominant species, and the median (minimum and maximum) concentrations were 105 (7.25 and 455) and 21.2 (1.22 and 85.9) g/m3 for the fine mode and the coarse mode, respectively. The maximum PM10 fluoride concentration was 541 g/m3; the TWA-PEL of fluoride is 2 mg/m3. The attack tank is where sulfuric acid reacts with phospha te rock that mainly contains calcium, phosphate, and fluoride. The reaction products are phosphoric acid and phosphogypsum, which will precipitate.

PAGE 30

30 The violent reaction also produ ces heat and water vapor from this process, and gaseous fluorides (silicon tetrafluoride (SiF4) and hydrogen fluoride (HF)) ar e the byproducts from this process [ Mann 1992]. In the weak phosphor ic acid production of 30% P2O5, SiF4 is the major fluoride form to volatilize because of its higher vapor pressure. The molar ratio of HF to SiF4 is less than 2 when the phosphoric acid is below 50% P2O5. Increasing the concentration of P2O5 to 50% makes more HF escape from liquid phase [ Denzinger et al. 1979]. In the presence of atmospheric moisture, fluorosilicic acid (H2SiF6), hydrogen fluoride and s ilicon oxide are created (Reaction 2-5 to Reaction 2-8) [ Hodge and Popovici 1994]. The gaseous fluorides adsorption process is dependent on the surface area of water droplet. Therefore, the size distribution of water droplets may influence the partitioning of fluoride. SiO2 + 4 HF SiF4 + 2 H2O (2-5) 3 SiF4 + 2 H2O 2 H2SiF6 + SiO2 (2-6) SiO2 + 6 HF H2SiF6 + 2 H2O (2-7) H2SiF6 SiF4 + 2 HF (2-8) For the rotating table/belt filter floor, fluoride was the dominant species for the fine mode (median: 3.9 g/m3; range: 0.173 106 g/m3), whereas phosphate, sulfat e, and ammonium were also important. The rotating table and belt filters are open systems used to filter intermediate products to obtain 30% phosphoric acid. The temperature of this process is about 100 C. Because it is an open system operating at hi gh temperature, water, gaseous fluoride, and phosphoric acid evaporate to form fine mist aerosol s. Ammonia gas is eas ily absorbed into the acid mist to neutralize the acid. Phosphate was the major constituent for the coarse mode (median: 22.3 g/m3; range: 2.33 122 g/m3), whereas sulfate and ammonium were also important. For PM10, phosphate was the major species a nd its highest concentration was 170

PAGE 31

31 g/m3. Filter cake formed in this filtration pr ocess, which mainly consists of phosphogypsum, contributes to the loading of the coarse particles due to mechanic al agitation. As shown, the calcium concentration of the coarse mode is high er than the fine mode. The vaporized acids can also condense on these mechanically agitated wet par ticles. Therefore, they were also present in the coarse mode. At the sulfuric acid truck loading and unloa ding stations, ammonium and sulfate were the dominant species for the fine mode (median: 9.5 and 4.2 g/m3, respectively) and the concentration ranges of amm onium and sulfate were 4.77 32.9 and 0.613 9.74 g/m3, respectively. Phosphate, ammonium and sulfate were the dominant species for the coarse mode (median: 8.5, 5.2 and 3.3 g/m3) and their ranges were 1.26 48.4, 0.3 13.6, and 1.36 12.8 g/m3. For this outdoor location, the possible period to release sulfuric acid is when trucks load and unload sulfuric acid, which is usually 2-3 h/da y. This emission is not a continuous source. Hence, most of the time, the aerosol loading wa s greatly influenced by the ambient condition and likely the aerosols were from other locations in th e plant. In summary, the concentrations of all species were low. Sampling for the granulator on a scrub day was conducted in one facility only. The median concentrations (N = 2) are shown in Table 2-5. For the fine mode, fluoride and ammonium were the dominant species; for the coarse mode, phosphate, fluoride, and ammonium were the major ones. In normal ope ration, 30% phosphoric acid is pres ent in this system. In this plant, the period of scrubbing was about 4 h, and a weak sulfuric acid solution was used to scrub the piping and ductwork. Nevertheless, sulfat e concentration was low at this location.

PAGE 32

32 Aerosol Acidity The hydrogen ion plays an important role in the carcinogenetic mechanism and skin irritation. Therefore, hydrogen ion concentration in the aerosol provides very useful data for assessing the health risk. Assuming that charge ba lance of those ions exists in the aerosols in this study, hydrogen ion concentration can be estimated by Equation 2-9: [H+] ( eq/m3) = [Anion] [Cation] (2-9) Anion species include phosphate, fluoride, sulf ate, chloride, and nitrate; cation species include ammonium, sodium, potassium, calcium, and magnesium. The statistics of hydrogen ion concentrations at each location ar e shown in Table 2-6. Fine mode aerosol at the attack tank area had the highest hydrogen ion concentration, 20.7 eq/m3 or 20.7 g/m3 with a median of 4.4 eq/m3, indicating they were strongly acidic. On the other hand, the average hydrogen ion concentration at sulfuric acid pum p tank areas (fine mode) and sulfuric acid truck stations were negative (i.e., the aerosol had excess OH), which indicated that the aerosol was somewhat basic. The relationship of cation equivalent weight and anion equivalent weight is shown in Figure 2-8, which can be used to determine the ac idity of the aerosol. The key factors that can control the acidity of the aerosol are the amounts of basic or acidic species. The acidic species in these plants include fluoride, su lfate, and phosphate. The only ba sic gas in the plant that can neutralize all acid species (i.e., su lfuric acid, gaseous fluoride, a nd phosphoric acid) is ammonia. In Figure 2-8, the cation and anion equivalent we ights can be divided into two regions: smaller than 3 and larger than 3 eq/m3. In the case of low acid species loading, the aerosol acidity can be influenced by the presence of basic species; ho wever, at high acid species loading, the limited amount of basic species may not be enough to ne utralize the aerosol acid ity. At the rotating table/belt filter floor and the attack tank, the main acidic gases such as gaseous fluoride and

PAGE 33

33 phosphoric acid are emitted from the process, and th ere is no source of ammonia in these indoor locations. Therefore, the aerosols remain acidic. On the other hand, both the sulfuric acid pump tank area and truck loading/unloading station are outdoor locations. Wi nd assists in the mixing of the aer osols with ambient air. For the sulfuric acid pump tank, the hi gh surface area of the fine mode aerosol is easier for mass transfer of basic gases, while the coarse mode aerosol needs a longer time to reach neutralization [ Meng and Seinfeld 1996]. The truck loading/unloading station does not have any significant sulfuric acid emission source, and aerosols at th is location come mainly from other locations. Hence, the aerosol acidity depends on the wind st rength/direction and aerosol emission source. The results also imply that the aerosol becomes less hazardous as it moves away from the emission source due to atmospheric dilu tion and neutralization by basic species. Summary Aerosol sampling using dichotomous samplers was carried out at five ty pes of locations in eight fertilizer facilities. Th e highest sulfate concentration wa s obtained at the sulfuric acid pump tank area. The maximum sulfuric acid concentration measured in PM10, including fine mode and coarse mode, was 0.185 mg/m3. At the attack tank area where phosphoric acid is produced by reacting sulfuric aci d with phosphate rock, fluoride was the dominant species. The maximum fluoride concentration in PM10 was 462 g/m3. At the rotating table/belt filter floor, phosphoric acid is separated fr om phosphogypsum by rotating tabl e/belt filter and the high temperature is favorable for the evaporati on of fluoride and phosphoric acid, which can subsequently form fine aeros ol or condense on phosphogypsum aer osol created by mechanical agitation. The maximum phosphate concentration in PM10 was 170 g/m3. On a scrub day, a weak sulfuric acid solution is used to clean th e piping and ductwork of the granulator for an

PAGE 34

34 average of 4 h per day. Particulate sulfate concen trations were low during the scrubbing activity. At the truck loading/unloading station, the possible emission period is around 2 3 h/day, and this emission is not continuous. The concentration levels at the loading/unloading station were low and were greatly influenced by outdoor conditions. The PM10 concentrations of sulfuric acid mist at these facilities were lower than the TLVTWA standard of 0.2 mg/m3 recommended by ACGIH for the thor acic fraction of sulfuric acid aerosol. The maximum PM10 of sulfuric acid mist was observed at the sulfuric acid pump tank area and was close to but found to be lower than the TLV-TWA If monitoring of personal exposures to sulfuric acid mist is to be required, these effo rts should focus on workers with activities in this area.

PAGE 35

35 Table 2-1. Sampling locations at phos phate fertilizer plants in Florida Location Sulfuric Acid Pump Tank Filter Floor (Rotating Table) Filter Floor (Belt) Granulator (on a Scrub day) Truck Station Attack Tank Plant A X X X X X Plant B X X X X Plant C X X X Plant D X X Plant E X X Plant F X X Plant G X X X X Plant H X X Total sample number 27 21 6 2 9 6

PAGE 36

36 Table 2-2. Analysis conditions for soluble ions Dionex ICS-1500 Cation Anion Analyzable species K+, Na+, Ca2+, Mg2+, NH4 + F-, Cl-, NO3 -, SO4 2-, PO4 3Extraction solution volume (D.I. water) 10 ml 10 ml Analytical column IonPac CS12A IonPac AS9-HC Guard column IonPac CG12A IonPac AG9-HC Suppressor ASRS-ULTRA II CSRS-ULTRA II Eluent 18 mM methanesulfonic acid 9.0 mM sodium carbonate Flow rate 1.0 ml/min 1.0 ml/min Injection volume 50 l 50 l Analysis time 20 min 30 min

PAGE 37

37 Table 2-3. Detection limit of ion chromatography (ICS 1500) Anion ppm Cation ppm F0.10 Na+ 0.05 Cl0.02 NH4 + 0.12 NO3 0.07 K+ 0.05 PO4 30.06 Mg2+ 0.04 SO4 20.12 Ca2+ 0.08

PAGE 38

38 Table 2-4. Median concentration ( g/m3) of ion species at background sites Mass FClPO4 3SO4 2Na+ NH4 + K+ Mg2+ Ca2+ Gainesville, FL Fine 15.0
PAGE 39

39 Table 2-5. Median concentration ( g/m3) of aerosol chemical composition at the granulator on a scrub day FClNO3 PO4 3SO4 2Na+ NH4 + K+ Mg2+ Ca2+ Mean (Fine mode) 14.1 2.112.531.933.750.3898.77 0.2 0.0640.391 Mean (Coarse mode) 8.691.311.46 37.0 4.75 1.22 10.1 0.238 0.527 1.16

PAGE 40

40 Table 2-6. Statistics of hydr ogen ion concentrations (eq/m3) at each location Location Filter floor H2SO4 pump tank area Attack tank area H2SO4 truck station Mode Fine Coarse FineCoarseFineCoarse FineCoarse Maximum 4.8 4.2 0.63.320.74.7 -0.1 0.3 Minimum -1.5* -1.2 -1.2-0.9-2.10.0 -1.2-0.7 Median 0.2 0.5 0.00.04.41.3 -0.3-0.2 *: A sign indicates OHconcentration.

PAGE 41

41 Figure 2-1. Manufacturing proce sses at fertilizer facilities

PAGE 42

42 Florida Gainesville Winter Haven Plant Background site Figure 2-2. Geographic locat ions of sampling sites

PAGE 43

43 Figure 2-3. Fine mode and coarse mode aerosol mass concentrati ons at various locations. The first letter denotes location: Ffilter (rotating table or belt filter) floor, Ttruck loading/unloading station, Aattack tank area and Ssulfuric acid pump tank area. The subscript denotes particle size: Ffine mode (PM2.5), Ccoarse mode (PM2.5-10) --Mean --Median Location FFFCTFTCAFACSFSC Particle mass co ncentration ( g/m 3 ) 0 200 400 600 800 1000 Minimum Maximum ...... Mean Median FF FC TF TC AF AC SF SC

PAGE 44

44 FClPO4SO4NaNH4KMgCa 0 20 40 60 80 100 120 140 Minimum Maximum Species*Concentration ( g/m 3 )...... Mean Median F-Cl-PO4 3-SO4 2-Na+NH4 +K+Mg2+Ca2+Species FClPO4SO4NaNH4KMgCa 0 20 40 60 80 100 120 140 Minimum Maximum SpeciesConcentration ( g/m 3 )...... Mean Median A B Figure 2-4. Aerosol chemical species at the sulf uric acid pump tank area. A) For fine mode. B) For coarse mode

PAGE 45

45 SpeciesConcentration ( g/m 3 ) FClPO4SO4NaNH4KMgCa 0 20 40 60 80 100 120 140 Minimum Maximum ...... Mean Median SpeciesConcentration ( g/m 3 ) FClPO4SO4NaNH4KMgCa 0 100 200 300 400 500 Minimum Maximum ...... Mean Median A B Figure 2-5. Aerosol chemical species at the atta ck tank area. A) For fine mode. B) For coarse mode

PAGE 46

46 Concentration ( g/m 3 )Species FClPO4SO4NaNH4KMgCa 0 20 40 60 80 100 120 140 Minimum Maximum ...... Mean Median F-Cl-PO4 3-SO4 2-Na+NH4 +K+Mg2+Ca2+Species FClPO4SO4NaNH4KMgCa 0 20 40 60 80 100 120 140 Concentration ( g/m 3 )Species ...... Mean Median Minimum Maximum A B Figure 2-6. Aerosol chemical species at the rotati ng table/belt filter floor. A) For fine mode. B) For coarse mode

PAGE 47

47 Species*Concentration ( g/m 3 ) FClPO4SO4NaNH4KMgCa 0 20 40 60 80 100 120 140 Minimum Maximum ...... Mean Median F-Cl-PO4 3-SO4 2-Na+NH4 +K+Mg2+Ca2+Species FClPO4SO4NaNH4KMgCa 0 20 40 60 80 100 120 140 Minimum Maximum Species*Concentration ( g/m 3 )...... Mean Median A B Figure 2-7. Aerosol chemical species at the sulfur ic acid truck loading/unloading station. A) For fine mode. B) For coarse mode

PAGE 48

48 Anion equivalent weight ( eq/m 3 air) 0.010.1110100 Cation equivalent weight ( eq/m 3 air) 0.01 0.1 1 10 100 Acidic Aerosol Basic Aerosol 1:1 3 3 Belt / rotating table filter floor (fine) Belt / rotating table filter floor (coarse) H2SO4 pump tank area (fine) H2SO4 pump tank area (coarse) Attack tank area (fine) Attack tack area (coarse) H2SO4 truck station (fine) H2SO4 truck station (coarse) Figure 2-8. Relationship of cation equivalent weight a nd anion equivalent weight

PAGE 49

49 CHAPTER 3 SIZE-RESOLVED SULFURIC ACID MIST CONCENTRATIONS AT PHOSPHATE FERTILIZER MANUFACTURING FACILITIES IN FLORIDA Background Strong inorganic acid mists cont aining sulfuric acid have been reported to correlate with lung and laryngeal cancer in humans [Blair and Kazerouni, 1997; Sathiakumar et al., 1997; Steenland, 1997] and are identified as a human carcinogen as reported by the NTP [USDHHS, 2005]. Sulfuric acid is typically present in the ai r in the mist form. Its chemical characteristics include low volatility, high acidity, high reactivity, high corrosivity a nd high affinity for water. Phosphate fertilizer manufacture is listed in th e NTP report as one of the industries that has sulfuric acid mist exposure poten tial. The OSHA has established an 8 h TWA of PEL of sulfuric acid mist at 1 mg/m3. It is well known that the deposition of an aerosol in the respiratory system depends on its aerodynamic behavior. Consideri ng the effects of aerosol size, the ACGIH has adopted a TLV-TWA of 0.2 mg/m3 for the thoracic particulate fraction of sulfuric acid mist [ACGIH, 2004]. NIOSH Method 7903 is an OSHAapproved method that is co mmonly used by the health and safety staff in industries to measure the total sulfuric acid mist con centration. The sampler of NIOSH Method 7903 is a silica gel tube consisting of one sect ion of glass fiber filter plug followed by two sections of silica gel (commerc ially available: ORBO-53 tube, Supelco, and SKC silica gel tube, SKC). The glass fiber filter plug is designed to filter out the majority of acid aerosols while the silica gel sections are used mainly to adsorb acid gases. The collected samples are desorbed in eluent and the aliquo ts are analyzed by IC. NIOSH researchers who developed the method reported ~9 0% collection efficiency for acidic aerosols with 0.4 m Reprinted with permission from Hsu, Y.-M., Wu, C. -Y., Lundgren, D. A., Birky, B. K., 2007. SizeResolved Sulfuric Acid Mist Concentrations at Phosphate Fertilizer Facilities in Florida. Ann. Occup. Hyg. 51, 81-89.

PAGE 50

50 volume median diameter (94.8 4.8% for H3PO4 and 86 4.6% for H2SO4) when the samples collected on the glass fiber sec tion and the front silica gel section (400 mg) were combined [Cassinelli, 1986; Cassinelli and Taylor, 1981]. Cascade impactors are commonly used for ch aracterizing aerosol size distribution [Dibb et al., 2002; Swietlicki et al., 1997]. Large particles are colle cted on a substrate by inertial impaction, while small particles can better follow the changes in the flow direction of a curved air stream. By adopting a series of impactor stages with increas ing flow velocities, the aerosol size distribution can be classified. The approved NIOSH method only pr ovides total sulfuric acid mi st concentration, but not size-dependent information. The comparison of the total mist concentration with the sizefractionated measurement by the cascade impactor may provide a convenient tool for correlating exposure levels based on the simpler NIOSH me thod. This information can be applied to develop informed policies with respect to respiratory protection in the workplace. To properly assess the occupationa l exposure of workers to sulf uric acid mist in phosphate fertilizer manufacturing facilities, the objectives of this chapter were to determine the total sulfuric acid mist con centrations using the a pproved NIOSH method and to characterize the size distributions of sulfuric acid mist by cascade impactor sampling. All other chemical species will be reported in next chapter. Furthermore, the feasibility of using a co rrelation factor between these two measurements was examined. Methods Sampling Sites The final products of phosphate fertilizer f acilities are MAP, DAP phosphoric acid and sulfuric acid. The manufacturing pr ocesses have been described in Hsu et al. [2007a]. Five types of locations with the potential of H2SO4 exposure corresponding to the manufacturing

PAGE 51

51 process were selected for sampling. These locati ons (the number of the sampling sites) include the sulfuric acid pump tank area (9), rotating table/belt filter floor (9), attack tank area (2), truck station for loading/unloading sulf uric acid (3), and the granulator on a scrub day (1). Sampling was carried out at ei ght plants. Seven of them are located in central Florida, and one of them is located in north Florida. In addition, Gainesville FL, and Winter Haven, FL, were chosen as the background sites. Sampling and Analysis Methods A University of Washington Source Test cascad e impactor (Mark III) was used as an area sampler to sample mist aerosols for the size distri bution. The inlet of the cascade impactor was set at 1.5 m from the floor. The impactor wa s operated at 25 Lpm with aerodynamic cut sizes (d50) of 0.20, 0.48, 0.98, 1.8, 3.8, 10 and 23 m fo r the seven stages, respectively. The impactor with glass fiber filter can provi de high collection efficiency fo r aerosols; however, glass fiber filter is well known to adsorb acidi c gas, such as sulfur dioxide [Chow, 1995; Lee and Mukund, 2001]. Therefore, Teflon membrane filters (ZefluorTM, 8 10 inches, pore size: 2 m, Pall Corp., Ann Arbor, MI) that prov ide high collection efficiency and low reactivity with acidic gases [Chow, 1995; Lee and Mukund, 2001] were applied for the co llection substrate. Those filters were cut to fit onto the impactor stag es. The collection efficiencies of the cascade impactor for liquid (substrate: a glass fiber filte r) and solid aerosols (substrate: an aluminum foil with silicone coating) were 97.2 11.9 and 94.1 17.3%, respectively [Pauluhn, 2005]. Droplet break-up in this instrument is neg ligible for large droplets (10 and 25 m) even when a high sampling flow rate (28.3 Lpm) is applied [Horton and Mitchell, 1989]. A final Teflon filter (ZefluorTM, 47 mm, pore size: 2 m, Pall Corp., Ann Arbor, MI) was placed after the impactor to collect penetrating aerosols. Filt ers were placed in a desiccator at room temperature for preand

PAGE 52

52 post-conditioning for 24 h before weighing to reduce the effect of water collected by the filter. The vapor pressure of sulfuric acid mist is low and it remains in the mist form under these conditions. The lower limit of particle size collec ted on the final filter was assigned to be 0.03 m, a typical value of those empl oyed in other research studies [Divita et al., 1996; Howell et al., 1998; Wagner and Leith, 2001]. The aerodynamic diameters of collected particles were from 0.03 to 23 m. PM23 is the aerosol mass concentration with an aerodynamic diameter smaller than 23 m, which was the largest aerosol size collected using this methodology. PM23 sulfuric acid mist concentration was used to compare the total sulfuric acid mist concentration provided by NIOSH Method 7903. NIOSH Method 7903 was applied for the sampling of total sulfuric acid mist concentration using a commercially available silica gel tube (O RBO 53 tube, Supelco). The sampling flow rate was 0.45 Lpm for 72 sets of samples. Six sets of samples were also collec ted at 0.3 Lpm in order to verify whether the results were the same at di fferent flow rates in th e recommended range (0.2 0.5 Lpm) [Cassinelli, 1986; Cassinelli and Eller, 1979; Cassinelli and Taylor, 1980; NIOSH, 1994]. Gravimetric measurement for sample mass wa s carried out using a microbalance (model MC 210 S, Sartorius Corp., Edgewood, NY; readability 10 g), and the analysis of sulfate concentration was accomplished by using an IC system (Dionex ICS 1500, Dionex Corp., Atlanta, GA). The analytical columns of anion species [nitrate (NO3 -), sulfate (SO4 2-), fluoride (F-), phosphate (PO4 3-), and chloride (Cl-)] and cation species [potassium (K+), calcium (Ca2+), magnesium (Mg2+), sodium (Na+), and ammonium (NH4 +)] are IonPac AS9-HC (Dionex Corp., Atlanta, GA) and IonPac CS12A (Dionex Corp., A tlanta, GA), respectively. The detection limit for sulfate was determined to be 0.12 ppm for this system.

PAGE 53

53 The sampling time was 24 h and three successive samples were obtained for each sampler at each site. Totally, there were 72 sets of imp actor samples and 78 silica gel tube samples in those plants. In background locations, there were six sets of impactor sa mples and six silica gel tube samples. Calculation of Fine Mode Because 2.5 m was not one of the cascade impactor cut-sizes, PM2.5 was determined by interpolating the size bin that covers 2.5 m (i.e., 1.81 and 3.76 m). Assuming a uniform distribution in this size range in log-scale, PM2.5 can be obtained according to the following relationship: 2.51.81 3.761.81PMPM log(2.5)log(1.81) log(3.76)log(1.81)PMPM (3-1) Rearranging the formula, PM2.5 can be derived as: 2.5PM= 0.44[3.761.81PMPM] +1.81PM (3-2) Calculation of Sulfuric Ac id Mist Concentration According to NIOSH Method 7903, sulfuric acid mist concentration is converted from the sulfate concentration determined by IC. Although the sulfate ma y not necessarily come from sulfuric acid (i.e., it can be ammonium sulfate or calcium sulfate), any sulfate determined by this method is conservatively assumed to be sulfuric acid. In this study, the same protocol was followed for all samples from the fertilizer plants. Results and Discussion Background Site In general, the mass concentrations and sulfate concentrations were low at both background locations. The PM23, PM10 and PM2.5 were 20.6.2, 18.7.9 and 15.1.4 g/m3 in Gainesville, and 19.1.0, 16.3.0 and 11.5.8 g/m3 in Winter Haven. The

PAGE 54

54 corresponding sulfate concen trations were 5.4.6 (PM23), 5.2.1 (PM10) and 5.0.3 (PM2.5) g/m3 in Gainesville, and they were lo wer in Winter Haven: 3.0.2 (PM23), 2.7.8 (PM10) and 2.4.5 (PM2.5) g/m3. The ratios of sulfate concentra tion to the sum of all ionic species concentrations (NO3 -, SO4 2-, F-, PO4 3-, Cl-, K+, Ca2+, Mg2+, Na+, and NH4 +) were 0.44.46 in Gainesville and 0.16.39 in Winter Haven. For NIOSH method samples, the total sulfat e concentrations ranged from 6.81 to 10.5 g/m3 in Gainesville and much hi gher in Winter Haven, 31.2.0 g/m3. Compared to the cascade impactor results, the measurements were closer in Gainesville than those in Winter Haven. The ratio of sulfate from the impact or to sulfate from the NIOSH method sampler ranged from 0.67 to 0.82 in Gainesville and from 0.069 to 0.096 in Winter Haven. It should also be noted that while the impactor results showed higher su lfate concentrations in Gainesville than in Winter Haven, the NIOSH method measurements showed the opposite. It is suspected that the NIOSH Method 7903 might have interferences from SO2 gas. This will be further discussed in later sections. Plants: Cascade Impactor Samples The sampling results of the cascade impactor at all locations are s hown in Figure 3-1A. The highest median sulfuric acid mist concentration was observed at the sulfuric acid pump tank areas where two sulfuric acid mist concentrations from the cascade impactor were higher than the OSHA standard, 1 mg/m3. The size information at each type of location will be discussed as follows. Attack tank area Sampling for the attack tank area was carried ou t at two plants. The aerosol and sulfuric acid PM23, PM10 and PM2.5 mass concentrations are listed in Ta ble 3-1. Aerosols at the attack

PAGE 55

55 tank areas had high mass loadings but low sulfuric acid concentrations. The violent reaction in the attack tank releases heat a nd causes a significant amount of volatile species to evaporate. These species, such as fluoride gases, condense on existing aerosols when they encounter cooler ambient air, thus resulting in high aerosol mass concentrations. However, the temperature in the process was not high enough for the evaporation of sulfuric acid or sulf ate salt that has lower vapor pressure. Hence, sulfuric acid mist concentrations were low at this location. Sulfuric acid pump tank area Sampling at the sulfuric acid pump tank area wa s carried out at all eight plants. The PM23, PM10 and PM2.5 mass concentrations are listed in Table 3-2. The geometric mean PM23, PM10 and PM2.5 sulfuric acid concentrations (geometric standard deviation) were 41.7 (.5), 37.9 (.8), and 22.1 (.5) g/m3. The highest geometric mean su lfuric acid concentration from cascade impactor measurement among all types of locations was indeed observed at the pump tank area. The large geometric standard deviation implies that the sulfuric acid concentrations at these nine sulfuric acid pump tank areas differe d greatly. The geometric mean ratios of sulfate concentration to all ionic species were greater than 0.50, which indicated that sulfate was the predominant ion accounting for the aerosol mass at this type of location. The aerosol mass size distributions and sulfuric acid size dist ributions are shown in Figure 3-2. They are plotted in two ranges: larger than and smaller than 100 g/m3. The size distribution maintained the same pa ttern at a given site, but not from site to site. It should also be noted that most of the sulfuric acid si ze distributions resemble the aerosol mass size distributions at the same site. At plants B1, D, H and B2, Fi gure 3-2A, both the aerosol mass concentrations and sulfuric acid mass c oncentrations were higher than 100 g/m3, and the aerosols were predominantly supermicron particles. The sulfate/aerosol mass ratios were greater than 0.5 (Table 3-3). The high ratios indicate th at sulfuric acid was the major species and that

PAGE 56

56 these locations might have sulfuric acid emissi on sources. At the pump tank area, the possible sulfuric acid emission source is the leakage of SO3 that will quickly combine with water molecules to form H2SO4. At other plants, Figure 3-2B, the sulfuric ac id concentrations were lower than 100 g/m3 and their sulfuric acid aerosols were mainly in the submicron range or presented no specific pattern. In the case of a very low aerosol mass loading, the sulf uric acid aerosols could very likely be affected by the ambient aerosols at this outdoor location. The geometric mean sulfuric acid concentration at Plant F was 6.8 g/m3 (Table 3-3), which is as low as that at the background sites. Belt or rotating table filter floor Sampling at the belt or rotating table filter fl oor area was carried out at all eight plants. The aerosol mass concentra tions were high: the PM23, PM10 and PM2.5 mass concentrations ranged from 57.4 to 2535, 54.0 to 1857 and 28.3 to 358 g/m3, respectively; their geometric mean concentrations (geometric standard devi ation) were 225.3 (.3), 187.0 (.2) and 94.7 (.8) g/m3, respectively. PM23, PM10 and PM2.5 sulfuric acid concentrations were 7.1575, 4.9419 and 2.460.0 g/m3; the geometric mean sulfuric acid concentrations (geometric standard deviation) of PM23, PM10 and PM2.5 were 17.9 (.7), 14.4 (.7) and 6.6 (.1) g/m3. The ratios of sulfate to all ions ranged from 0.07 to 0.32 (geometric mean: 0.16). The low ratios indicate that sulfuric ac id was not a major compound at the filter floor area. The fractional size distributions of aerosol mass and sulfuric acid at nine belt/rotating table filter floors are shown in Figure 3-2C. Sulfuric acid was dominantly present in the coarse mode. During this process, gypsum is filtered out by belt or rotating table filter, and the aerosols are formed by

PAGE 57

57 mechanical agitation. The similarity in sulfuric acid and aerosol mass size distributions indicates that the chemical might be from th e residual content in the product. Sulfuric acid truck loading/unloading station Sampling at the truck loading/ unloading station was conducted at three plants. The aerosol mass concentrations were low and their PM23, PM10 and PM2.5 ranged from 19.9 to 174, 15.8 to 131, and 10.0 to 69.7 g/m3. Sulfuric acid was the major species (the ratios of sulfate to total ion mass: 0.28.42, median: 0.36); the con centrations ranged from 3.9 to 24.5 (PM23), 3.5 to 23.9 (PM10) and 3.1 to 20.6 (PM2.5) g/m3, which were close to the measurements at the background locations. All size distributions of sulfuric acid were similar: the mode size was 0.48.98 m. During loading and unloading, sulfur ic acid is transferred from the storage tank in liquid form to the truck. The only possible pathway that workers can be exposed to sulfuric acid is the spray of sulfuric acid from the truck loading nozzle. Normally, the connection is well sealed and the workers are well protected to avoi d contact with liquid su lfuric acid. The measurements verified that the sulfuric acid concentrations were very low. Granulator on a scrub day The mass concentrations ranged from 126 to 835 (PM23), 90.2 to 578 (PM10) and 59.7 to 303 (PM2.5) g/m3. The sulfuric acid concen tration ranges were 7.63.9 (PM23), 5.73.7 (PM10) and 3.85.6 (PM2.5) g/m3. The source of sulfuric acid mists is the spray from the scrubbing process (a weak acid solution) whic h is not a steady operation. Hence, the concentrations varied significantly, bu t they were all below the standards. Plants: NIOSH Method Samples Sampling results of the NIOSH samples at five types of locations are shown in Figure 31B. Generally, the sulfuric ac id pump tank area had the higher concentrations. However, the

PAGE 58

58 maximum sulfuric acid concentration (11225 g/m3) measured by the NIOSH method for all locations was obtained at the filter floor area The largest geometric mean sulfuric acid concentration was obtained at the pump tank area (143.2 g/m3), followed by the granulator on a scrub day (122.4 g/m3) and then at the filter floor (108.7 g/m3). The geometric mean concentrations at the granulator on a scrub day and sulfuric acid truck loading/unloading stations were at lower levels. Comparisons of the Results from the Cascade Impactor and the NIOSH Method The paired measurements of PM23 sulfuric acid concentratio ns from the cascade impactor and total sulfuric acid concentrations from th e NIOSH method at all sampling locations are shown in Figure 3-3. As shown, 71 out of 72 im pactor samples had a lower concentration than the NIOSH method measurement. The ratios of sulfuric acid mist c oncentrations from the NIOSH to the cascade impactor were 1.5.0 for 71 impactor samples. The largest difference was over two orders of magnitude, and many of the NIOSH measurements were more than 10 times larger the impactor results. The subs tantial difference was qu ite unexpected. To quantitatively compare these two measurements and evaluate their correla tion, three ratios were used which are defined as E quation 3-3 to Equation 3-5: 23 23 NPM C R (3-3) 10 10 NPM C R (3-4) 2.5 2.5 NPM C R (3-5) where PM23, PM10 and PM2.5 are sulfuric acid concentrations from the cascade impactor and CN the sulfuric acid concentrat ion by NIOSH Method 7903.

PAGE 59

59 Table 3-4 displays the R23, R10 and R2.5 at five types of sampling locations. A large variation at each type of location is observed, e.g ., from 0.00 to 0.67 at the filter floor areas for R23. A strong relationship between methods would be indicated by a constant correlation factor; the wide variation of the ratios does not allow any meaningful correlation of these two types of measurements. The much higher values by the NIOSH method also prompted further investigation of the data. In examining the data, it was found that in many cases the silica ge l sections collected more sulfuric acid than the glass fiber section (s ee examples in Table 3-5). The results of the NIOSH method imply that sulfuric acid as well as sulfate is partially measured as a gas. In the ambient condition, there is no known sulfate species (inorganic) that exists in the gas phase. Even sulfuric acid exists in the condensed phase because it has a high boiling point of 330 oC and a very low vapor pressure at r oom temperature (< 0.001 mmHg) [ Weast 1988]. Furthermore, the hygroscopic sulfuric acid will rapidly pick up moisture in th e air and remain in the aerosol phase. It should be noted again that according to Ortiz and Fairchild [1976] the majority of the aerosol mass is collected on the gl ass fiber filter plug. Chen et al [2002] reported that aerosol penetration across the filter se ction of the silica gel tube (S KC 226-10-03 tube) at the most penetrating size was lower than 5%. The collecti on efficiency of large particles (> 3 m) is higher than 99%. Sulfuric acid mists mainly ex ist as coarse aerosols at the pump tank area (Figure 3-2A); hence, aerosol penetration canno t explain the high sulfuric acid concentration sampled by silica gel. Thus, the situation that more sulfuric acid mist is co llected in the silica gel section than the glass fiber sec tion is highly unlikely. The adso rption of a significant amount of interfering gases by the silica gel is therefor e suspected to be the reason for the observed phenomenon.

PAGE 60

60 Another unexpected phenomenon was observed wh en comparing the re sults obtained from the NIOSH method at different sampling flow ra tes. In the recommended range of sampling flow rate (0.2 0.5 Lpm), concentrations should be simila r when different flow rates are used. Two different flow rates, 0.3 and 0.45 Lpm, were employed for six sets of the NIOSH method sampling, and the results are show n in Table 3-5. These paired samples were taken concurrently (e.g., #1-low and #1-high were taken at the sa me time), and three consecutive samples were carried out (i.e., #1 followed by #2 and then by #3). As shown, sulfuric acid concentrations at the lower flow rate (0.3 Lpm) were higher than t hose at the higher flow rate (0.45 Lpm). Most of the ratios (C@ 0.3 Lpm/ C@ 0.45 Lpm) were larger than 1, and they were much higher in the silica gel section than those in the glass tube section. Th e concentrations at two different flow rates do not exhibit any direct proportion betw een the gas phase and the particulate phase. Hence, systemic errors can be neglected. The collection mechan ism of acid gases on the silica gel is diffusion, and its efficiency decreases as the flow rate increases (due to shorter residence time). The observations support the hypothesis that the higher measurement in the silica gel section is probably from the collection of gas. Silica gel, a high surface area material, can adsorb SO2 [ Kopac and Kocabas 2002; Stratmann and Buck 1965]. The hydrophilic property of silica gel can effectively attract moisture which can enhance the absorption of soluble species, such as SO2 [ Tsai et al. 2001]. The adsorbed or absorbed SO2 can be further oxidized to form sulfate [ Lunsford 1979] that causes overestimation. Annual SO2 concentrations (2003) were mon itored by the state of Florida [ FDEP 2003] and the results indicated that central FL had higher SO2 concentrations (~2 ppb) than north FL (~2 ppb) If the above hypothesis is true, this might explain why sulfate concentrations in

PAGE 61

61 Winter Haven measured by the NI OSH method were much higher th an those in Gainesville but the impactor results did not exhibit such a pattern. Limited SO2 concentrations in fertilizer facilities are available in the literature. The SO2 concentration in a fertilizer factory in India was 41.7 mg/m3 [ Yadav and Kaushik 1996] while those in China and Sweden were 0.34.97 and 3.6 mg/m3, respectively [ Englander et al. 1988; Meng and Zhang 1990]. Atmospheric dispersion can qui ckly reduce the concentration to a much lower level as reported in a study near a fertilizer factory in Zimbabwe [ Jonnalagadda et al. 1991]. If the hypothesis holds true, the sulfuric acid concentration in th e fertilizer facilities can be expected to be overestimated when SO2 is present in the studi ed facilities. Further investigation of this subject is warranted. Comparisons of Sulfuric Acid Mist Concentr ations with OSHA and ACGIH Regulations The number of samples with concentrations higher than the OSHA regulation (total sulfuric acid mist concentration < 1 mg/m3) was 7 of the 78 samples collected. According to the cascade impactor sampling, 90% of the samples were lower than the ACGIH standard and 97% of the samples were lower than the OSHA regulation. The resu lts of the NIOSH method samples obtained concurrently with the impactor samples had a smaller percentage of samples with concentrations lower than the OSHA regulation. The only location where the sulfuric acid mist concentrations from the cascade impactor ex ceeded both the OSHA and ACGIH standards was the sulfuric acid pump tank area. For the NI OSH method samples, the locations included sulfuric acid pump tank areas, belt/rotating table filter floors and the granulator on a scrub day. Summary In this study, the total sulfuric acid mist concentration and sulfuric acid mist size distributions at modern phosphate fertilizer manufacturing facilitie s in Florida were characterized

PAGE 62

62 by using NIOSH Method 7903 and a cascade impactor respectively. Sampling was carried out at five types of locations in eight facilities and two b ackground sites. Based on cascade impactor sampling, sulfuric acid pump tank areas had higher sulfuric acid mist concentrations than other types of locations and su lfuric acid was the dominant chemical species. When high sulfuric acid conc entrations were identified, the aerosols were dominantly in the coarse mode. The most likely cause of high sulf uric acid concentrations at this location is the leakage of SO3. Constant inspection of th e tubing around this location and immediate repair may provide an effective m easure to further lower the concentrations. According to the impactor sampling results, seven samples (total: 72) exceeded the ACGIH standard (0.2 mg/m3, thoracic fraction), and two samples (total: 72) were above the OSHA regulation (1 mg/m3, total concentration). Meanwhile, there were seven samples (total: 78) by the NIOSH method that exceeded the OSHA regulati on. It should be noted at these locations, workers typically stay for about 1 h per day and respirators for sulfuric acid mist are required in this area. Consequently, the real time-weighted exposure to sulf uric acid mist is likely to be lower than the concentrations from the st ationary sampling conducted in this study. The results from the cascade impactor and the NIOSH method were compared to determine if a correlation factor could be established. The sulfuric acid mist concentrations from the NIOSH method were higher than those fr om the cascade impactor for the dominating majority of samples. No specific trend of systemic error was observed between these two methods. In many cases, the silica gel section collected more sulfu ric acid than the glass fiber filter plug. This is highly unlikel y because sulfuric acid or sulfat es are not known to exist in the gas form in the ambient condition, and should not be collected in the silica gel section. Moreover, the sulfuric acid con centrations at 0.30 Lpm were hi gher than the concentrations at

PAGE 63

63 0.45 Lpm in the NIOSH method sampling, indicati ng the influence of diffusing gases. The possible reason for the variation is the interaction between SO2 and silica gel/glass fiber filter. Further investigation has been ve rified the causes and is discus sed further in Chapter 5.

PAGE 64

64 Table 3-1. PM23, PM10 and PM2.5 mass and sulfuric acid concentr ations at the at tack tank areas g/m3 Sample number min max Geometric mean Geometric standard deviation PM23 Massa 6 86.2 1853 341.3 3.9 Mass (IC)b 6 62.7 414.6 152.8 2.3 Sulfuric acidc 6 6.7 19.0 10.1 1.5 PM10 Massa 6 76.1 1767 308.3 4.1 Mass (IC)b 6 59.4 397 139.6 2.3 Sulfuric acidc 6 6.3 12.5 8.7 1.3 PM2.5 Massa 6 64.6 720 187.0 2.9 Mass (IC)b 6 41.6 182 90.1 1.8 Sulfuric acidc 6 2.3 7.6 5.5 1.6 Massa: mass concentration from weighing the mass. Mass (IC)b: the sum of all ionic species concentrations measured by IC. Sulfuric acidc: all sulfate is conservatively assumed to be sulfuric acid.

PAGE 65

65 Table 3-2. PM23, PM10 and PM2.5 mass and sulfuric acid concentr ations at the sulfuric acid pump tank areas g/m3 Sample number min max Geometric mean Geometric standard deviation PM23 Massa 27 20.4 1644 121 3.3 Mass (IC)b 27 9.4 1268 69.5 3.8 Sulfuric acidc 27 2.1 1187 41.7 5.5 Sulfate/ Mass (IC) 27 0.22 0.98 0.588 1.52 PM10 Massa 27 14.7 1625 105 3.5 Mass (IC)b 27 8.7 1221 61.9 4.0 Sulfuric acidc 27 2.0 1155 37.9 5.8 Sulfate/ Mass (IC) 27 0.23 0.98 0.600 1.52 PM2.5 Massa 27 12.1 648 56.4 2.8 Mass (IC)b 27 6.5 548 35.0 3.2 Sulfuric acidc 27 1.8 558 22.1 4.5 Sulfate/ Mass (IC) 27 0.25 1.00 0.619 1.47 Massa: mass concentration from weighing the mass. Mass (IC)b: the sum of whole ionic species concentrations measured by IC. Sulfuric acidc: all sulfate is conservatively assumed to be sulfuric acid.

PAGE 66

66 Table 3-3. Mass, sulfuric acid concentrations and sulfate/massa ratios of the impactor samples at the sulfuric acid pump tank areas Massa: mass concentrations combining a ll ionic species concentrations Mass (g/m3) Sulfuric acid (g/m3) Sulfate/ Massa Plant Geometric mean Geometric standard deviation Geometric mean Geometric standard deviation Mean Sample size D 905.9 1.8 643.3 1.8 0.96 3 B1 550.8 3.0 298.3 3.8 0.84 3 H 177.9 2.1 114.0 2.6 0.88 3 B2 95.7 1.9 37.7 2.0 0.57 3 E 64.9 1.7 33.1 2.0 0.71 3 C 76.4 1.3 11.8 1.8 0.41 3 A 71.5 1.9 21.4 1.5 0.46 3 G 50.4 1.4 8.0 1.5 0.45 3 F 37.4 2.5 6.8 4.5 0.42 3

PAGE 67

67 Table 3-4. Statistics of R23, R10 and R2.5 at five types of sampling location Sample number Min Max Geometric mean Geometric standard deviation PM2.5 27 0.04 0.58 0.15 2.21 PM10 27 0.06 2.65 0.26 2.22 Sulfuric acid pump tank area PM23 27 0.06 2.74 0.30 2.20 PM2.5 27 0.00 0.41 0.06 2.81 PM10 27 0.00 0.64 0.13 2.52 Belt/rotating table filter floor PM23 27 0.00 0.67 0.17 2.55 PM2.5 9 0.05 0.36 0.10 1.93 PM10 9 0.08 0.42 0.15 1.81 Truck station PM23 9 0.10 0.44 0.18 1.81 PM2.5 6 0.02 0.10 0.06 1.93 PM10 6 0.02 0.22 0.09 2.08 Attack tank area PM23 6 0.03 0.23 0.11 2.07 PM2.5 3 0.03 0.28 0.13 3.88 PM10 3 0.03 0.40 0.16 4.25 Granulator on a scrub day PM23 3 0.04 0.54 0.21 3.87

PAGE 68

68 Table 3-5. Sulfuric acid concen trations and the ratios measured at two flow rates at the rotating table filter floors using NIO SH Method 7903 [H2SO4]a ( g/m3) Low flow rate (@ 0.3 Lpm) High flow rate (@ 0.45 Lpm) C@ 0.3 Lpm / C@ 0.45 Lpm Sampler section #1 #2 #3 #1 #2 #3 #1 #2 #3 Total b 90.3 141.0 118.8 80.8 87.2 89.8 1.12 1.62 1.32 Glass fiber 36.7 44.3 44.3 35.7 37.7 46.7 1.03 1.17 0.95 Silica gel 53.6 96.7 74.5 45.1 49.5 43.2 1.19 1.95 1.73 [H2SO4]a: paired samples at two flow rates taken concurrently. Totalb: the sum of the results from the glass fiber section and the silica gel section.

PAGE 69

69 A B Figure 3-1. Sulfuric acid concentr ations at 5 types of locations. Locations: ATattack tank, PTsulfuric acid pump tank, FFfilter (rotating table or belt filter) floor, and TStruck station for sulfuric acid loading/ unloading. Sampling methods: A) PM23 sulfuric acid by cascade impactor, and B) Total su lfuric acid by NIOSH Method 7903 Location ATPTFFTSSD [H 2 SO 4 ] ( g/m 3 ) 101102103104 90% 10% 25% 75% Median Max Min Location ATPTFFTSSD [H 2 SO 4 ] ( g/m 3 ) 101102103104 OSHA standard2 104(b)

PAGE 70

70 Aerodynamic diameter ( m) 0.010.1110100 Fraction/ lnDp 0.0 0.2 0.4 0.6 0.8 1.0 (a) Aerodynamic diameter ( m) 0.010.1110100 Fraction/ lnDp 0.0 0.2 0.4 0.6 0.8 1.0 1.2 (b) Aerodynamic diameter ( m) 0.010.1110100 Fraction/ lnDp 0.0 0.2 0.4 0.6 0.8 1.0 (c) Sulfuric Acid Sulfuric Acid Median Mass Mass Median 90% 10% 25% 75% Median A B C Figure 3-2. Sulfuric acid mist and aerosol ma ss size distributions. A) Sulfuric acid pump tank areas, high aerosol mass loading (> 100 g /m3). B) Sulfuric acid pump tank areas, low aerosol mass loading (< 100 g/m3). C) Belt/rotating table filter floors

PAGE 71

71 1 10 100 1000 10000 100000 110100100010000100000H2SO4 concentration (NIOSH) (g/m3)H2SO4 concentration (Impactor) ( g/m3) Sulfuric acid pump tank Belt/rotating filter floor Truck station Scrub day Attack tank1:1 1:10 1:100 105 104 103 102 101 100 Figure 3-3. Comparison of PM23 sulfuric acid concentrations fr om the cascade impactor and total sulfuric acid concentrations from the NIOSH method

PAGE 72

72 CHAPTER 4 SIZE DISTRIBUTION, CHEMICAL COMPOSITION AND ACIDITY OF MIST AEROSOLS IN FERTILIZER MANUFACTURI NG FACILITIES IN FLORIDA Background Strong inorganic acid mists cont aining sulfuric acid have b een reported to correlate well with lung and laryngeal cancers in humans [ Blair and Kazerouni 1997; Sathiakumar et al. 1997; Steenland 1997]. Phosphate fertilizer manufacturing f acilities, where sulfuric acid is used to digest phosphate rock to form H3PO4, have been listed by the NTP [ USDHHS 2005] as one of many occupational exposures to strong acid. The current OSHA 8 h TWA of PEL of H2SO4 and H3PO4 mists is set at 1 mg/m3. The carcinogenic mechanism of strong inorganic acid mists containing sulfuric acid is not clearly understood yet [ Blair and Kazerouni 1997]. However, reduced pH environments are known to enha nce the depurination rate of DNA and the deamination rate of cytidine [ Swenberg and Beauchamp 1997; USDHHS 2005], which can cause DNA damage or mutation. Furthermore, st udies have found that a dverse pulmonary health effects are related to the hydrogen ion (H+) rather than sulfate [ Ostro et al. 1991; Schlesinger 1984, 1989; Schlesinger et al. 1990a; Schlesinger et al. 1990b] and that the biological response is a function of the total concentration of H+ deliverable to the cells or the total extractable H+ per particle [ Schlesinger and Chen 1994]. The aerosol acidity is an important factor th at affects the amount of the extractable H+, i.e., strong acid can release more H+. In phosphate fertilizer facilities, H3PO4 and H2SO4 mists are the major acid species [ Hsu et al. 2007a]. These two strong aci ds can contribute a significant amount of the extractable H+. Currently, no method is available to directly measure the aerosol acidity; nevertheless, several ambient aerosol thermodynamic models have been developed to Reprinted with permission from Hsu, Y.-M., Wu, C. -Y., Lundgren, D. A., Birky, B. K., 2007. Size Distribution, Chemical Composition and Acidity of Mi st Aerosols in Fertilizer manufacturing Facilities in Florida. J. Aerosol Sci. doi:10.1016/j.jaerosci.2007.10.008.

PAGE 73

73 estimate the aerosol acidity [ Yao et al. 2006, 2007]. However, particulate phosphate species have not been included in those aerosol models because the concentrations of those species are too low to be detected in the ambient aerosols. The hazard of aerosol depends on the chemical compositions and the sites where the aerosol deposits. It is well known that the deposi tion of an aerosol in the respiratory system is related to its aerodynamic behavior. Meanwhile, relative humidity and chemical characteristics of the aerosol may affect its size. Sulfuric ac id is hydrophilic, and the size of the mist can increase significantly when it encounters moisture [ Seinfeld and Pandis 1998; Wu and Biswas 1998] in the respiratory system. In considering the effects of aerosol size on human health, the ACGIH has adopted a TLVTWA of 0.2 mg/m3 for the thoracic particulate fr action of sulfuric acid mist [ ACGIH 2004]. Although the acid mist size distribution is an importa nt parameter in evaluati ng health effects, in the past very little data was avai lable for the particle size distri bution of phosphoric acid mists at phosphate fertilizer facilities. The objectives of this chapter are to characterize the chemical characteristics of mist aerosols in the current phosphate facilitie s with size-resolved information and to establish an aerosol thermodyna mic model to estimate aerosol acidity. Methods Sampling Sites Five types of locations which might have hi gh acid mist concentrations were selected, including H2SO4 pump tank areas, product filter floors, attack tank areas, tr uck stations for loading/unloading H2SO4, and a granulator on a scrub day. Sampling was conducted at seven plants in central FL and one plan t in north FL. Totally, there we re 24 sampling locations at eight fertilizer plants. The type and number of sampli ng locations at each plant have been described in detail in Hsu et al [2007a] and Chapter 2 of this dissertation.

PAGE 74

74 Sampling and Analysis Methods A University of Washington Source Test cascad e impactor (Mark III) was used to collect mists for the size distribution. The impactor was operated at 25 Lp m and the corresponding d50 was 0.20, 0.48, 0.98, 1.8, 3.8, 10, and 23 m for the seven stages, respectively. Aerosol bounce from the impaction plates can result in mass loss and thus distort the samples true size distribution. However, a previous study [ Pauluhn 2005] showed that the mean mass recoveries from cascade impactor analyses were 97.2 11.9% and 94.1 17.3% for liquid and solid aerosols, respectively; theref ore, aerosol bounce problems of the cascade impactor can be neglected. The collection substrate used in this study was ZefluorTM, or backed Teflon membrane filter (pore size: 2 m, Pall Corp.), which provides high collection efficiency and low reactivity with acidic gases [ Chow 1995; Lee and Mukund 2001]. A final filter (ZefluorTM, 47 mm, pore size: 2 m, Pall Corp.) was placed after the impactors to collect any penetrating aerosols. The lower limit of particles collected on the final filter was assigned to be 0.03 m, a value commonly used in similar cascade impactor sampling studies [ Divita et al. 1996; Howell et al. 1998; Wagner and Leith 2001]. The sampling time was 24 h and three successive samples were obtained at each location. Tota lly, there were 72 sets of impactor samples in those plants. A microbalance (model MC 210 S, Sartorius Corp.; readability: 10 g) was used for weighing the aerosol mass. In order to reduce the interference of water vapor, all filters were placed in a desiccator at room temperature for preand post-conditioning for 24 h before weighing the filters. After weighing the filter, water soluble species were extracted from the filter using deionized (DI) water in an ultras onic bath (Bath ultrasonic timer, Fisher scientific) for 1 h. A Nanopure Diamond system (Barnstead International) provided the DI wa ter with a conductivity

PAGE 75

75 of 18.2 megohm. The water solubl e ion species concentrations we re analyzed by an IC system (Dionex ICS 1500, Dionex Corp.). Two analytical columns, IonPac CS12A (Dionex Corp.) and IonPac AS9-HC (Dionex Corp.), were a pplied to analyze ca tions, including K+, Na+, Ca2+, Mg2+, and NH4 +, and anions, including F-, Cl-, NO2 -, NO3 -, SO4 2-, and PO4 3-, respectively. Aerosol Thermodynamic Model A thermodynamic model of multicomponent aerosols was developed to estimate the acidity of aerosols with high sulfuric acid (200 g/m3) or high phosphoric acid (500 g/m3) concentration. The importan t parameters of the model chemical components, the aerosol water amount (AWA), equilibrium reactions, reactio n constants, and the activity coefficient calculations are described in this section. Five chemical species larger than 1% (mass) based on the chemical analysis in this study were selected: Ca2+, NH4 +, Na+, SO4 2-, and PO4 3-. Liquid components include H+, Ca2+, NH3, NH4 +, Na+, SO4 2-, HSO4 -, H2SO4, H2PO4 -, H3PO4, H5P2O8 -, H2O, Na3PO4, and (NH4)3PO4; solid components include Ca3(PO4)2, CaHPO4, NH4H2PO4, (NH4)3PO4, Na3PO4, NaH2PO4, Na2HPO4, CaSO4 2H2O, NH4HSO4, (NH4)2SO4, NaHSO4, and Na2SO4. AWA ( g/m3 air) was calculated by Equation 4-1: AWA = AMC IMC (4-1) where AMC is the total aerosol mass concentration ( g/m3 air) and IMC the total ionic species mass concentration ( g/m3 air). Chemical equilibrium reactions and equilibr ium constants are listed in Table 4-1 for 20 reactions. The equilibrium constant is expressed as a function of te mperature. It should be noted that the sampling was conducted for 24 h and temperature varied during the sampling.

PAGE 76

76 Therefore, the rate constants under the standa rd condition, 298.15 K, were applied in the modeling work. The activity of ionic species in aqueous phase (ai) is defined as Equation 4-2: iiiarm (4-2) where ri is the activity coefficient of species i and mi is the molal concentration of species i (mole solute/kg solvent). Multicomponent activity coefficients were calculated following the Kusik and Meissner method [ Kim et al. 1993], which is based on bina ry activity coefficients: ... log 2 log 2 log0 14 2 4 1 4 0 12 2 2 1 2 2 1 2 12r z z m r z z m I z z z r + 2 2 00 32 112 112332 12loglog... 22 zz zzz mrmr zzI (4-3) where r12 is the activity coefficient of a 1-2 ion pair in the multicomponent solution, zi is the absolute number of unit charges of ion species i and I is the ionic strength of the solution, which is i i iz m I22 1 (4-4) o ijr is the mean ionic activity coefficient of the single solute solution with i j ion pair. To calculate multicomponent activity coefficients binary activity coefficients need to be available. They were calculat ed by a polynomial regression method [ Bassett and Seinfeld 1983] for the ionic strength smalle r than 30 mole/kg and the Kusik and Meissner method [ Kim et al. 1993] for that larger than 30 mole/kg. The pol ynomial regression method is shown in Equation 4-5.

PAGE 77

77 1 2 / 1 2 / 1 121 lni i i oI a BI AI r (4-5) where A B and ia are parameter values which are avai lable in Bassett and Seinfeld [1983]. The Kusik and Meissner method is shown in Equation 4-6 and q values used in this model are listed in Kim et al [1993]. 2 / 1 2 / 1 121 5107 0 exp 1 0 1 1 CI I B I B rq o (4-6) where B = 0.75 0.065 q and C = 1 + 0.055 q exp( 0.023 I3) Currents ambient aerosol thermodynamic mode ls have not included phosphate species, including H3PO4, H2PO4 and H5P2O8. However, phosphate species should be considered in this study due to high particulate phosphate concen trations and the activit ies were calculated by Pitzers equation shown in E quation 4-7 to Equation 4-9 [ Jiang 1996]. ca a a c c c a a a a rm m C mz B m f z r1 1 1 1 2 1 12 2 ln aa aa a a ca ca ca a cm m C z B z m m' 1 1 2 1 (4-7) ac c c a a a c c c c rm m C mz B m f z r2 1 2 2 2 2 22 2 ln cc cc c c ca ca ca a cm m C z B z m m' 2 2 2 2 (4-8) 1122 12 12lnrlnr ln vv r vv (4-9) where 2 / 1 2 / 1 2 / 12 1 1 ln 2 1 2 2 1 1 392 0 I I I fr

PAGE 78

78 2 / 1 2 / 1 ) 1 ( ) 0 (2 exp 2 1 1 2 I I I Bij ij ij 2 / 1 2 / 1 2 ) 1 ( '2 exp 2 2 1 1 2 I I I I Bij ij j i ij ijz z C C 2 Iij ij 1v and 2v are the number of moles of ion species 1 and 2 per mole of 1-2 ion pair species dissociating completely. The parameters C ,ij 1 0 ij ijand are listed in Jiang [1996] and 'ij can be neglected for most cases [ Kim et al. 1993]. Results and Discussion Aerosol Chemical Species Sulfuric acid mists, phosphoric acid mists, a nd particulate fluoride were the major species observed in the phosphate fertilizer plants. Their concentrations at all locations are shown in Figure 4-1. The highest mass concentrati ons of these species were 1163, 1589 and 388 g/m3, and they were obtained at the su lfuric acid pump tank area, the belt/rotating table filter floor, and the attack tank area, respectively. Major ch emical species concentrations and their size nformation at each location are discussed in the fo llowing sections. All ion concentrations at the truck station were very low (max: 64 g/m3) and are therefore not discussed further. Sulfuric aid pump tank areas The size distribution of sulfuric acid in this area has been reported previously [ Hsu et al. 2007b], and only the major results are summarized here. The median mass concentrations ( standard deviation) were 98 ( 430), 71 ( 424) and 52 ( 149) g/m3 for PM23, PM10 and PM2.5

PAGE 79

79 (aerosol aerodynamic diameter smaller than 23, 10 and 2.5 m), respectively. The major mode was 3.8 m when high sulfuric acid mist concentra tion was observed. Sulfuric acid was the dominant species in this area, and its size di stribution was similar to the aerosol mass size distribution at this location. For other species at the sulfuric acid pump tank areas, ammonium had the highest concentration, second only to su lfuric acid, followed by calcium. The sulfuric acid pump tank area is an outdoor location where the emitted aer osols can mix with the ambient air. The relations between sulfate and two major cations calcium and ammonium, are shown in Figure 42. In the case of sulfate and ammonium, the rati os of sulfate to ammonium were close to 1 for small particles ( < 1.8 m, including the impactor stages of 0.03.20, 0.20.48, 0.48.0, 1.0 1.8 m) when sulfate concentrations were low ( < 50 neq/m3). Regarding the relation between calcium and sulfate, it was the large particles ( > 1.81 m) that had calcium/sulfate ratios close to 1 when sulfate concentrations were low ( < 50 neq/m3). The relation between sulfate and both basic species is shown in Figure 4-2C. It is inte resting to note that both size ranges had a similar trend where the concentrations of the two basic species were combined. Furthermore, the combined concentrations were higher than sulfat e concentrations when sulfate concentrations were lower (< 50 neq/m3). There is a transition range between 50 and 500 neq/m3. Above 500 neq/m3, sulfate concentrations were higher than those of the two basic species. Clearly demonstrated, calcium and ammonium play a criti cal role in neutralizi ng the acid, and they dominate in separate size range s: ammonium for fine aerosols a nd calcium for coarse aerosols. However, if the acid emission is high, there may not be sufficient cations in the air to balance the acids.

PAGE 80

80 Product filter floors Phosphoric acid was the major species at the pr oduct filter floors and particulate fluoride was the species of the next importance. The median mass concentrations ( standard deviation) of PM23, PM10, and PM2.5 were 201 ( 514), 178 ( 372), and 95 ( 68) g/m3, respectively. The median concentrations of phosphoric acid were 35 (PM23), 26 (PM10), and 5 (PM2.5) g/m3 and those of particulate fluoride were 19 (PM23), 17 (PM10), and 11 (PM2.5) g/m3. Phosphoric acid mists and aerosol mass size distributions are shown in Figure 4-3, and both had a similar trend. It can be clearly seen that predominantly the phosphoric acid was presen t in the coarse mode (3.8 m). The boiling point of pure phosphoric acid is 217 C [ Weast 1988]. However, the boiling point of phosphoric acid so lution decreases as th e concentration of phosphoric acid in the solution decreases [ Messnaoui and Bounahmidi 2005]. The boiling point is only 108 C when phosphoric acid concentration is 36 wt% P2O5, which is the designated concentration in the manufacturing process. This temperature is close to the oper ating temperature (100 C) in the manufacturing process at the plants. Hence, it is possible that phosphoric acid vaporizes at this operating condition and then condenses to form ae rosols when it encounters cooler air. The species that had the second highe st median concentration was fl uoride and the size distributions at the product filter floor areas are displayed in Figure 4-3C. The major mode size of fluoride was 3.8 m. The minor mode size of fluoride was 0.03.2 m. Because the product filter floors are semi-open space, no major cations are available to neutralize the phosphoric acid and fluoride. The inside air does not mix well with ambient air. Therefore, the aerosol emission controls the aerosol chemical species concen trations at this type of location.

PAGE 81

81 Attack tank areas Sampling was carried out at two attack ta nk areas and six samples were obtained. Particulate fluoride was the major species at the a ttack tank areas. The median concentrations of fluoride were 67 (PM23), 64 (PM10), and 60 (PM2.5) g/m3, which indicated most fluoride existed in small aerosols. Phosphate rock which is the main useful produc t of phosphate ore, consists of calcium phosphate mineral apatite with gangue constituents, which include silica, fluoride, calcite, dolomite, clay a nd iron-aluminum oxide [ Hodge and Popovici 1994]. The attack tank is where phosphate rock reacts with sulfuric acid. The violent reaction causes strong heat release in the form of vapor, which is ev acuated from the attack tank w ith other gaseous effluents. Fluoride can be evaporated as gaseous fluoride, e.g., silicon tetrafluoride [ Parish 1994]. Fluoride has strong affinity with silica, and therefore fluoride generated from the process tends to combine with silica. The high affinity of fluor ide with silica causes fluor osilicic acid to remain in phosphoric acid solution or fluorosilicate (SiF6 2-) to precipitate. Particulate fluoride size distributions at the attack tank areas are shown in Figure 4-4. Two t ypes of size distribution were observed. When fluoride concentra tions were high (two samples: 320 and 388 g/m3), the mode size was 3.8 m. This occurred at a plant where the attack tank was not a closed system. At this location with high fluoride concen trations, moisture was al so high. Therefore, the possible explanation for the size distribution is that gas phase fluoride and water were evaporated and then condensed on existing aerosols. For the other plant where the attack tank is a closed system, the fluoride concentrations and humidity were lower (four samples: 25 g/m3), and the mode size was 0.03.2 m. Ammonium had the second highest concentration, and its medi an concentrations were 22 (PM23), 20 (PM10) and 18 (PM2.5) g/m3. The relation between these two major species, fluoride

PAGE 82

82 and ammonium, is shown in Figure 4-5. The results indicated that there was not enough ammonium to neutralize fluoride when high fl uoride concentrations were observed ( > 250 neq/m3). Granulator on a scrub day Table 4-2 displays the medi an concentrations of PM23, PM10 and PM2.5 of all species at the granulator on a scrub day. Particulate fluoride, phosphate, sulfate and ammonium had higher concentration levels than othe rs. Nonetheless, there was no clear trend in terms of the dominating species due to signifi cant variation among the samples. Scrubbing is an intermittent maintenance activity and it usua lly takes 3-4 h for one scrubbi ng. Therefore, the aerosol emission was not stable. Aerosol Acidity Although the carcinogenic mechanism of strong i norganic acid mists containing sulfuric acid is not clear yet, aerosol acidity may play a key factor. To estimate the aerosol acidity, the charge balance method and the aerosol thermody namic model were applied in this study. Charge balance method Sulfuric acid and phosphoric acid were the ma jor species at the pump tank areas and the product filter floors, respectively. Assuming that charge balance of those ions analyzed in this study exists in the aerosols, hydrogen ion con centration can be estimated by the difference between cation and anion equivalent weights. Th e relationships of cation equivalent weight and anion equivalent weight of PM10-23, PM2.5-10, and PM2.5 at these locations are shown in Figure 46. The aerosol acidity was high at th e sulfuric acid pump tank areas for PM2.5-10 and PM2.5. When anion concentration was higher than 500, 500, and 1000 neq/m3 for PM10-23, PM2.5-10, and PM2.5, respectively, most aerosols were acidic.

PAGE 83

83 Whether the location is in an outdoor environmen t or an indoor setting affects the acidity. In general, the major acid species can be neut ralized by the abundant a mmonia around the area in the outdoor location; in contrast, there is not enough mixing with the am bient air in the indoor location and the amount of ammonia is insuffici ent. A good example is the aerosol at the product filter floor. All product fi lters are indoor locations. It can be clearly seen that most aerosols are acidic in this type of location regardless of their pa rticle size. In addition to indoor/outdoor location, the other important factor is the acid loading. This can be best illustrated using the samples at the sulfuric acid pump tank areas. At this area, the PM10-23 aerosol had a low acid loading (< 500 neq/m3) and the aerosol was more or less neutral. Regarding PM2.5 and PM2.5-10 aerosols, the same can be said for low anion loading cases (< 1000 and 500 neq/m3), while the aerosols were acidic at high loadings (> 1000 and 500 neq/m3). In the case of high aerosol loadings, the aerosol re mains acidic because of the limited amount of basic species available to neutralize the particle s in a short period of tim e after emission. In PM2.5-10, 9 of 27 samples showed high acidity because sulfuric acid mists were dominant in this size range. The sampling location for the gra nulator on a scrub day was an outdoor location where the air could mix with ambient air. The aero sols from this location were close to neutral, even though their concentrations were high. Th e major species to neutra lize the aerosol acidity was ammonium from the granulat or where phosphoric acid reacted with ammonia to produce the final products, diammonium phosph ate or monoammonium phosphate. Aerosol thermodynamic model An aerosol thermodynamic model has been estab lished in this study to calculate the acidity of aerosols with high sulfuric ac id and phosphoric acid mist concen trations. Sulfuric acid is a stronger acid than phosphoric acid. Also, the form er is hygroscopic in natu re while the latter is

PAGE 84

84 not. These two factors have strong influences on the aerosols charac teristics in a humid environment, e.g., equilibrium size and aerosol H+ concentration. Four samples with sulfuric acid concentrations higher than 200 g/m3 were selected and two samples with phosphoric acid concentrations higher than 500 g/m3 were selected as representatives. Particulate fluoride concentrations were lower than 2.5 mg/m3, the OSHA regulation. Furthermore, aeros ol fluoride can exist as H2SiF6, which could not be discriminated in this study. Therefore, the samples with hi gh fluoride concentration at the attack tank area were not investigated. The results for aerosols with hi gh sulfuric acid mist concentr ations are displayed in Figure 4-7A. The modes of H+ concentration were 1.8.8 m and 3.8 m. The highest H+ concentration was 170 mole/kg (mol ality) and its mode was 1.8.8 m. The hygroscopic sulfuric acid can quickly pick up moisture to form larger mist particles in the respiratory system and the actual location where mists deposit can vary. Several assumptions were made to estimate the deposition site of aer osols and the time to reach its equilibrium size: (1) the relative humidity of the sampling locations and the hum an respiratory system were 40% and 95%, respectively; (2) the geometric mean size (2.6 m for 1.8.8 m and 6.2 m for 3.8 m) was considered as the representative size; (3) the cross-sectional area of human nose inlet is 2 cm2; (4) the air volume of one inhalation is 0.5 L; (5) the time for one inhalation is 1 s; (6) the distance from nose inlet to the laryngeal re gion is 10 cm. The ZdanovskiiStokesRobinson (ZSR) equation [ Jacobson 1999] shown in Equation 4-10 was us ed to calculate the equilibrium aerosol size at a gi ven relative humidity, and Equation 4-11 [ Hinds 1999] was used to determine the time for reaching the new equilibrium size. 7 7 6 6 5 5 4 4 3 3 2 2 1 0 w w w w w w w aa Y a Y a Y a Y a Y a Y a Y Y m (4-10)

PAGE 85

85 where ma is the molality of solute x, wa the water activity (relative humidity expressed as a fraction), and Ys the polynomial coefficients [Jacobson, 1999]. T p R M D dt d d dv p p4 dp > (4-11) where dp is the aerosol size, t the time, Dv the diffusion coefficient of vapor, M the molecular weight of the liquid, R the gas constant (8.314 JK-1mol-1), the density of the liquid, p the partial pressure of vapor in the gas surrounding the droplet, T the temperature away from the droplet surface, = 0.066 m, and 42 3 33 5 22 2 p p pd d d Fuchs correction factor The hydrogen ion size distributi ons at RHs of 40% and 95% fo r the sample with sulfuric acid mist concentration of 653 g/m3 are shown in Figure 4-7B. As the relative humidity increases, H+ concentration decreases while the mode si ze slightly increases. The time required to grow to the equilibrium size is only 0.014 s, whic h is smaller than the aerosol traveling time of 0.04 s for 10 cm. As shown, the inhaled mist aerosol can grow large enough to deposit in the upper respiratory tract. The International Commission on Radiologi cal Protection (ICRP) model has been developed to predict total and regi onal deposition of inhaled aerosol [ ICRP 1994]. Three simplified ICRP equations, Equation 4-12 to Equati on 4-14, were also app lied in this study to determine the deposition fraction (D F) for monodisperse spheres of standard density at standard conditions [ Hinds 1999]. For the head airways, DFHA,

PAGE 86

86 DFHA = 11 1exp6.841.183ln1exp0.9241.885lnppIF dd (4-12) where IF is the inhala ble fraction given as IF = 10.5 2.81 1 10.00076pd (4-13) For the tracheobronchial region, DFTB, DFTB = 220.00352 exp0.234ln3.4063.9exp0.819ln1.61pp pdd d (4-14) For the alveolar region, DFAL, DFAL = 220.0155 exp0.416ln2.8419.11exp0.482ln1.362pp pdd d (4-15) Three cases were considered, and the results ar e shown in Table 4-3. In Cases 1a and 2a, the geometric mean sizes for the 2 modes discusse d previously were used. These particles were assumed to maintain their original sizes since the ICRP model does not allow consideration of particle growth. As shown, the larger 6.2 m particles had a higher DF in the head airways than the smaller 2.6 m particles. It should be not ed, however, that the IF of 6.2 m particles was slightly lower than 2.6 m particles. In Case 3, 11.8 m is the equilibrium size of 6.2 m. Due to the much lower IF, the DF of 11.8 m particles was actually lower than the smaller 6.2 m particles! Since the growth to the equilibrium size takes much shorter time than the aerosol traveling time as previously discussed, the equilibrium size was used to determine the deposition in Cases 1b and 2b. Different outcomes were observed! Compared with the no-growth cases, the DF greatly increased. The results show that sulfur ic acid mists sampled in this study mainly deposit

PAGE 87

87 in the upper respiratory tract, which is consistent with epidemiological evidence of a correlation between H+ concentration and laryngeal cancer. The re sults also demonstrate the importance of considering particle growth for hygroscopic com ponents in assessing the pa rticles deposition in the respiratory system. However, it must be noted that extensive epidemiological studies of phosphate industry workers have concluded no relation was found between acid mist exposures and laryngeal cancer [ Checkoway et al. 1996]. Hydrogen ion size distributions for high phosphoric acid mist concentrations are shown in Figure 4-7C. The mode size of hydr ogen ion concentration was 1.8.8 m and the highest H+ concentration was 37.4 mole kg-1. Compared to the sulfuric acid cases discussed previously, clearly the hydrogen ion concentr ation of the phosphoric acid cas es was lower even when its concentration, 1589 g/m3, was higher than sulfuric acid mist concentration. The results show that sulfuric acid plays the dominant role in contributing extractable H+ and phosphoric acid is not as important as sulfuric acid in supplying extractable H+. Summary In phosphate fertilizer facilities, phosphoric acid and sulfuric acid mists were the major aerosol components for the product filter fl oors and the sulfuric acid pump tank areas, respectively. The median concentration and the sta ndard deviation of sulfuric acid at the sulfuric acid pump tank areas were 37 322 g/m3. The median concentrati on one standard deviation for phosphoric acid was 35 326 g/m3, and it mainly existed in the coarse mode. The possible source of phosphoric acid was evaporation and then condensation when it enc ountered cooler air. The attack tank area had highest fluori de concentration, which was 25 g/m3 at this area. The current OSHA 8-h TWA PEL of phosphoric acid and sulfur ic acid mist set at 1 mg/m3 was not exceeded on average, but could be exceeded at the product filter floors and the pump tank

PAGE 88

88 areas, respectively, under conservative assumptions that all phosphoric aci d and sulfuric acid came from phosphate and sulfate, respectively. It should be noted that workers spend much less than 8 hours per day in the area, and thus the true time-weighted exposure level can be expected to be lower. Calcium and ammonium were the major specie s to neutralize the aerosol acidity at the sulfuric pump tank areas when acid loading wa s low. The aerosol thermodynamic model showed the modes of aerosol H+ concentration in 1.8.8 m and 3.8 m for the aerosols with high sulfuric acid mist co ncentrations. These hygroscopic acid mists can grow in the high humidity conditions of the upper respir atory system, and aerosols with high H+ concentrations mainly deposit in the upper respir atory system. Sulfuric acid was found to play a much more prominent role than phosphoric acid and fluorid e. The respiratory de position projection of sulfuric acid mists is consistent with that of H+ and both components mainly deposit in the extrathoracic airways of the head and neck. However, extensive epidemiological studies of phosphate industry workers have not shown an increased incide nce of any type of cancer.

PAGE 89

89 Table 4-1. Equilibrium relations and constants Equilibrium Relation Equilibrium Constant Expression Equilibrium Constant Keq (298.15 K) Units Source ) ( ) ( 4 ) ( 4 2aq aq aqH HSO SO H 4 2 44 2 4 SO H HSO Hr SO H r r HSO H 1.000 103 mole/kg Jacobson [1999] ) ( 2 ) ( 4 ) ( 4 aq aq aqH SO HSO 4 2 44 2 4 HSO SO Hr HSO r r SO H 1.020 10-2 mole/kg Jacobson [1999] ) ( 4 2 ) ( ) ( 4 3 aq aq aqPO H H PO H 4 3 4 24 3 4 2 PO H PO H Hr PO H r r PO H H 6.918 10-3 mole/kg Jiang [1996] 24()34()528() aqaqaqHPOHPOHPO 528 2434528 2434 HPO H POHPOHPOr HPOHPOrr 1.263 mole-1 kg Jiang [1996] ) ( ) ( 3 ) ( 4 aq aq aqH NH NH 4 34 3 NH H NHr NH r r H NH 5.623 10-10 mole/kg Benjamin [2002] ) ( 2 aq aqOH H O H OH H wr r a OH H 1.010 10-14 mole/kg Kim et al.[1993] ) ( 2 2 ) ( 4 2 ) ( 2 42 2aq aq aq sO H SO Ca O H CaSO 2 2 4 22 4 2w SO Caa r r SO Ca 4.320 10-5 mole2/kg2Jacobson [1999] aq aq aqHSO NH HSO NH4 4 4 4 4 44 4 HSO NHr r HSO NH 1.380 102 mole2/kg2Jacobson [1999] 2 44()4()4() 22 s aqaqNHSONHSO 2 4 42 2 4 2 4 SO NHr r SO NH 1.820 mole3/kg3Jacobson [1999] 2 444()4()() 32()32aqaqaq sNHHSONHSOH 2 3 2 2 4 3 42 4 4 SO H NHr r r SO H NH 2.930 101 mole5/kg5Jacobson [1999] ) ( 4 ) ( ) ( 4 aq aq sHSO Na NaHSO 44 HSO Nar r HSO Na 2.840 102 mole2/kg2Jacobson [1999] 2 ) ( 4 ) ( ) ( 4 22aq aq sSO Na SO Na 2 42 2 4 2 SO Nar r SO Na 4.800 10-1 mole3/kg3Jacobson [1999]

PAGE 90

90 3 ) ( 4 2 ) ( ) ( 2 4 32 3aq aq sPO Ca PO Ca 2 3 2 3 4 3 23 4 2 PO Car r PO Ca 2.109 10-33 mole5/kg5 This studya 2 ) ( 4 2 ) ( ) ( 4 aq aq sHPO Ca CaHPO 2 4 22 4 2 HPO Car r HPO Ca 1.889 10-7 mole2/kg2 This studya 2 242()()24()2 22saqaqCaHPOHOCaHPOHO w PO H Caa r r PO H Ca2 2 4 2 24 2 2 6.153 10-2 mole3/kg3 This studya ) ( 4 2 ) ( 4 4 2 4 aq aq sPO H NH PO H NH 4 2 44 2 4 PO H NHr r PO H NH 7.106 10-1 mole2/kg2 This studya 3 ) ( 4 ) ( 4 ) ( 4 3 43aq aq sPO NH PO NH 3 4 43 3 4 3 4 PO NHr r PO NH 9.437 10-1 mole4/kg4 This studya 3 ) ( 4 ) ( ) ( 4 33aq aq sPO Na PO Na 3 43 3 4 3 PO Nar r PO Na 9.823 10-1 mole4/kg4 This studya ) ( 4 2 ) ( ) ( 4 2 aq aq sPO H Na PO NaH 4 24 2 PO H Nar r PO H Na 1.193 101 mole2/kg2 This studya 2 ) ( 4 ) ( ) ( 4 22aq aq sHPO Na HPO Na 2 42 22 4 NaHPONaHPOrr 6.879 mole3/kg3 This studya a i o i i i eqG v k T R K 1 exp K 15 298 R= 8.314 J/mole K, T = 298.15 K, ki = +1 for products and ki = 1 for reactants, vi = the dimensionless stoichiometric coefficient. o o iG Gproducts oGreactants, Go for each species is from Weast [1988].

PAGE 91

91 Table 4-2. Median ion species concentra tions of cascade impactor samples collected at the granulator on a scrub day (g/m3) NO. Mass FClNO3 PO4 3SO4 2Na+ NH4 + K+ Mg2+ Ca2+ PM23 392 54 5 4 35 26 9 66 3 2 11 PM10 331 42 5 6 23 22 8 58 3 1 7 PM2.5 263 39 4 4 9 18 6 48 2 1 3

PAGE 92

92 Table 4-3. Aerosol deposition fractions for 3 cases Size ( m) Deposition fraction Case Inhalable Deposition Inhalable fraction Head airway Tracheobronchial region Alveolar region 1a 2.6 2.6 0.99 0.70 0.06 0.11 1b 2.6 5.0 0.99 0.89 0.04 0.06 2a 6.2 6.2 0.94 0.87 0.03 0.04 2b 6.2 11.8 0.94 0.92 0.01 0.01 3 11.8 11.8 0.78 0.77 0.01 0.01

PAGE 93

93 Location 02468101214 Mist concentration ( g/m3) 0.1 1 10 100 1000 10000 Figure 4-1. Sulfuric acid, phosphoric acid and fluor ide concentrations at all locations. AT: attack tank area, PT: sulfuric acid pump tank area, FF: product filter floor, SD: the granulator plant on a scrub day

PAGE 94

94 [SO4 2-]/ lndp (neq/m3) 100101102103104105 [NH4 +]/ lndp (neq/m3) 100101102103104105 (a) 1:1 [SO4 2-]/ lndp (neq/m3) 100101102103104105 [Ca2+]/ lndp (neq/m3) 100101102103104105 (b) 1:1 [SO4 2-]/ lndp (neq/m3) 100101102103104105 ([Ca2+]+[NH4 +])/ lndp (neq/m3) 100101102103104105 <1.8 m > 1.8 m 500 500 50 50 (c) 1:1 A B C Figure 4-2. Relation between the major cati ons (ammonium and calcium) and sulfate concentrations at the sulfuric acid pump ta nk areas. A) Ammonium versus sulfate. B) Calcium versus sulfate. C) Amm onium and calcium versus sulfate

PAGE 95

95 Aerodynamic diameter ( m) 0.010.1110100 Fraction/ lndp 0.0 0.2 0.4 0.6 0.8 1.0 (a) Aerodynamic diameter ( m) 0.010.1110100 Fraction/ lndp 0.0 0.2 0.4 0.6 0.8 1.0 (b) Aerodynamic diameter ( m) 0.010.1110100 Fraction/ lndp 0.0 0.2 0.4 0.6 0.8 1.0 (c) A B C Figure 4-3. Aerosol size distribu tions at the product filter floo rs. A) Phosphoric acid mist. B) Aerosol mass. C) Particulate fluoride

PAGE 96

96 Aerodynamic diameter ( m) 0.010.1110100 Fraction/ lndp 0.0 0.2 0.4 0.6 0.8 1.0 Low conc. High conc. Figure 4-4. Particulate fluo ride size distribution at the attack tank areas

PAGE 97

97 [F-]/ lndp (neq/m3) 100101102103104105 [NH4 +]/ lndp (neq/m3) 100101102103104105 < 1.8 m (r2= 4.1 ) > 1.8 m (r2= 3.5 ) Figure 4-5. Relation between ammo nium and fluoride concentrations at the attack tank areas

PAGE 98

98 (a) PM10-23Anion equivalent weight (neq/m 3 air) 1e+11e+21e+31e+41e+5 Cation equivalent weight (neq/m 3 air) 1e+1 1e+2 1e+3 1e+4 1e+5 Basic Aerosol Acidic Aerosol 1:1 (b) PM2.5-10Anion equivalent weight (neq/m 3 air) 1e+11e+21e+31e+41e+5 Cation equivalent weight (neq/m 3 air) 1e+1 1e+2 1e+3 1e+4 1e+5 Basic Aerosol Acidic Aerosol 1:1 (c) PM2.5Anion equivalent weight (neq/m 3 air) 1e+11e+21e+31e+41e+5 Cation equivalent weight (neq/m 3 air) 1e+1 1e+2 1e+3 1e+4 1e+5 1:1 Basic Aerosol Acidic Aerosol Belt / rotating table filter floor H2SO4 pump tank area Attack tank area Granulator on a Scrub day A B C Figure 4-6. Relationship of cation equivalent weight and anion equi valent weight. A) For PM10-23. B) For PM2.5-10. C) For PM2.5

PAGE 99

99 (a) Aerodynamic diameter ( m) 0.010.1110100 [H+] (mole/kg) 0 40 80 120 160 200 1120 g/m3 1163 g/m3 653 g/m3 342 g/m3 (b) Aerodynamic diameter ( m) 0.010.1110100 [H+] (RH=40%) (mole/kg) 0 20 40 60 80 100 [H+] (RH=95%) (mole/kg) 0 1 2 3 40% 95% 0 10 20 30 40 50 0.010.1110100 Aerodynamic diameter ( m)[H+] (mole/kg) A B [PO4 3-]air= 1589 g/m3[PO4 3-]air= 621.8 g/m3(c) A B C Figure 4-7. Aerosol hydrogen ion concentr ation size distribution. A) Aerosol H+ concentration size distribution of samples with high sulfuric acid mist concentrations. B) Aerosol H+ concentration size distributions for the sample with H2SO4 concentration of 653 g/m3 at RHs of 40% and 95%. C) Aerosol H+ concentration size distribution of samples with high phosphoric acid mist concentrations

PAGE 100

100 CHAPTER 5 POSITIVE SULFATE ARTIFA CT FORMATION FROM SO2 ADSORPTION IN THE SILICA GEL SAMPLER USED IN NIOSH METHOD 7903 Background NIOSH Method 7903 [ NIOSH 1994] is the approved method set by OSHA for measuring the total concentration of aci dic aerosols and gases, includ ing hydrogen fluoride, hydrogen chloride (HCl), hydrogen bromide (HBr), nitric acid (HNO3), sulfuric acid and phosphoric acid. It is the method commonly used by the health an d safety staff in the phosphate industry, as well as other occupational environments such as se miconductor industry, lead battery factories, aluminum smelting, machining, electroplating processes, and even disaster response [ Healy et al. 2001; Hsu et al. 2007b; Lue et al. 1998; Tsai et al. 2001; Wallingford and Snyder 2001]. The sampler of NIOSH Method 7903 consists of one section of glass fiber filter plug, followed by two sections of silica gel. The glass fiber filter plug is designed to filter out the majority of aerosols, whereas the silica gel sections are used mainly to adsorb acidic gases. The NIOSH recommended sampling flow rate range is 0.2.5 Lpm (except that less than 0.3 Lpm should be used for HF). The samples collected are extrac ted and then analyzed by IC. In evaluating the method, NIOSH researchers reported ~100% collect ion efficiency for acidic gases (HCl, HF and HNO3) [ Cassinelli 1986; Cassinelli and Taylor 1981]. For aerosols (H2SO4 and H3PO4), ~90% efficiency (94.8 4.8% for H3PO4 and 86 4.6% for H2SO4) was reported when the samples collected on the glass fiber filt er section and the front silica gel section were combined. In Chapter 3 [ Hsu et al. 2007b], NIOSH Method 7903 was co mpared with cascade impactor sampling for sulfuric acid mist sampling at phosp hate fertilizer manufacturing facilities. The Reprinted with permission from Hsu, Y.-M., Kollett, J., Wysocki, K., Wu, C.-Y., Lundgren, D. A., Birky, B. K., 2007. Positive Artifact Sulfate Formation from SO2 Adsorption in the Silica Gel Sampler Used in NIOSH Method 7903. Environ. Sci. Technol. 41, 6205-6209.

PAGE 101

101 results indicate that sulfuric acid mist concentration from the NIOSH method might be overestimated, because of the interference of SO2. Past studies have reported that glass fiber filter can adsorb SO2, which can be subsequently transformed into a sulfite (2 3SO ) species [ Chow 1995; Coutant 1977; Lee and Mukund 2001; Watson and Chow 2001] on the moist-basic surface of glass fiber filter, as listed in Reaction 5-1 and Reaction 5-2. 3 2HSO OH SO (5-1) O H SO OH HSO2 2 3 3 (5-2) The glass fiber filter consists of borosilicat e glass filaments, whic h have high alkalinity [ Chow 1995; Coutant 1977; Watson et al. 1995]. It also contains high concentrations of sodium, potassium, calcium, and other basic species that exhibit high alkalin ities and pH values. These properties aid in the adsorption of SO2, HNO3 and acidic gases [ Chow 1995; Lee and Mukund 2001; Watson and Chow 2001]. To cause artifacts in sulfate measurement, subsequent oxidation to form sulfate is also critical. Oxid ation of sulfite in solutio n is highly dependent on pH, and the half-life is 4 5 min for sodium sulfite solution at room temperature, if the oxygen supply is unlimited [ Schroeter 1963]. Coutant [1977] reported that the conversion on glass fiber filter is 90% within 2 h when air containing SO2 passes though the filter Penetration occurs when the alkalinity decreases below a certain value. Silica gel, a high surface area material, can also adsorb sulfur dioxide [ Fox and Jeffries 1979; Kopac and Kocabas 2002; Stratmann and Buck 1965]. The hydrophilic property of silica ge l can effectively attract moisture, which can enhance the absorption of soluble species such as SO2 [ Tsai et al. 2001]. As mentioned, this method is widely used in the workplace to characterize sulfuric acid mist concentrations, as a way to evaluate protection for workers. No study has examined and quantified the artifact

PAGE 102

102 sulfate from the silica gel tube. Therefore, the objective of this chapter was to characterize the interference of SO2 to determine the accurate sulfuric acid mist concentration. The oxidation of sulfur(IV) and SO2 adsorption, following the NIOSH protocol, were investigated in this study. Methods Two groups of experiments were co nducted: sulfur(IV) oxidation and SO2 adsorption. Sulfur(IV) includes su lfite, bisulfite (HSO3 -), and sulfurous acid (H2SO3). These experiments are described below. Sulfur(IV) Oxidation NIOSH Method 7903 specifies the use of IC el uent solution as the extraction solution, which is the 9 mM Na2CO3 solution used in this study. In addition, a water bath at 100 C was used as specified in the method to enhance desorption of samples. To examine the effect of the extraction procedures on sulf ur(IV) oxidation, experiments were performed for four combinations of eluent solution and water bath temperature, which were as follows: (A) using DI water (Nanopure Diamond, Barnstead) of 18.2 M cm as the extraction solution without a water bath, (B) using DI water as the extractio n solution with a water bath at 100 oC to investigate the effect of temperature, (C) using 9 and 18 mM Na2CO3 solutions as the extraction so lution without a water bath to examine the effect of the co ncentration of eluent, and (D) using 9 mM Na2CO3 solution as the extraction solution with a water bath at 100 oC. Sodium sulfite (Na2SO3) was used as the sulfite sour ce. 3.7% formaldehyde (HCHO) solution was used to quench the sulfur(IV) oxidation [ Dong and Dasgupta 1986; Munger et al. 1986] at the designated time. For the proced ures, a 100 mL solution (D I water, 9 or 18 mM

PAGE 103

103 Na2CO3 solution) was placed on the stirrer (Isotem p Magnetic Stirrer, Fisher) of 300 rpm at room temperature or 100 C. The designated amount of Na2SO3 was then added into the solution. A 5 mL sample was taken out at a desi gnated time and put into a glass tube with 5 mL HCHO solution. Sulfate concen tration was then analyzed via IC (model ICS 1500, Dionex). Preliminary experiments were conducted to s uggest that the experime ntal time should be 1 h for tests without water bath and 10 min for tests with water bath. Measurements indicated that the sulfur(IV) oxidation was fast in the beginning. Hence, sulfate concentrations were measured every minute, for the first 5 min, for oxidation w ithout water bath and every 20 s, for the first 5 min, for oxidation with a water bath. Sulfur Dioxide Adsorption The experimental setup is shown in Figure 5-1. SO2 gas from a cylinder (10 ppm, relative uncertainty of 5%) was mixed with zero air (Thermo Elect ron Instrument) to obtain the desired concentration. A bubbler was used to supply moisture, and the mixing ratio was used to adjust the humidity. The gas stream then passed th ough a silica gel tube, followed by an impinger containing 100 ml of a 9 mM Na2CO3 solution. Preliminary experiments showed the SO2 collection efficiency of the impinger was 100%. The total molar concentration of SO2 passing through the system was the sum of sulfate molar concentration from the silica gel tube and the impinger. Hydrogen peroxide (0.6%) (1 mL) wa s added into the impinger after the experiment, to oxidize collected sulfite to obtain sulfate concentration. Th e suggested flow rate range of NIOSH Method 7903 is 0.2 0.5 Lpm. Thus, the experimental condition for sampling flow rate was set in this range. According to past studies, SO2 concentrations in sulfuric acid plants have been 0.12.9 ppm [ Englander et al. 1988; Meng and Zhang 1990; Yadav and Kaushik 1996]. Hence, SO2 concentration used in this study was set in this range. Experimental conditions are

PAGE 104

104 shown in Table 5-1. The adsorption of SO2 on silica gel tube under various feed SO2 concentrations, sampling flow rates, and sampling times was examined. Results and Discussion Sulfur(IV) Oxidation Aqueous sulfur(IV) uncatalyzed oxidation is a first-order reaction which can be expressed by Equation 5-3 [ Larson et al. 1978]: 2 3 2 3SO k dt SO d (5-3) The analytical solution to Equation 5-3 can be obtained by integrating it from time t = 0 to time t and is shown in Equation 5-4. Sulfate was th e species analyzed in th is study; therefore, sulfite mass concentration was calculated using the measured sulfate mass concentration following Equation 5-5. The least-square fit ting method (SigmaPlot, Version 8.0, SPSS Inc., Chicago, IL) with Equation 5-4 was used to calculate the rate constant ( k ) and the pre-exponent constant (a): 2 3 2 3expt oSO akt SO (5-4) [SO3 2-]t = 22 4480 96TotaltSOSO (5-5) where [SO3 2-]t is the sulfite concentration at time t [SO3 2-]o the initial sulfite concentration, [SO4 2-]Total the total sulfate concentration, [SO4 2-]t the sulfate concentration at time t and t the time (expressed in seconds). Figure 5-2 illustrates the normalized sulfite concentration ([SO3 2-]t/[SO3 2-]o) as a function of time. For the conditions expressed in panels (A) and (C), sulfur(IV) ox idation reached at least 85% in 1 h and 40 min, respectively. The means a nd standard deviations of the rate constants for

PAGE 105

105 the conditions in panels (A) and (C) were 0.0003 0.0001 and 0.0023 0.0003 s-1, respectively. As shown, the addition of Na2CO3 enhanced the kinetics, i.e., the application of Na2CO3 solution as the extraction solution can resu lt in more effective conversion. For the conditions described in panel (B), sulfur(IV) oxidati on reached at least 90% in 5 10 min and the mean standard deviation of the rate constants was 0.0198 0.0144 s-1. Comparing the results with that of conditions expressed in panel (A), it can be observed that the kine tics was significantly increased by the water bath designed to aid the desorption. Two Na2CO3 concentrations (9 and 18 mM) were tested; the results showed no discernible di fference between these two. For condition (D), sulfur(IV) oxidation reached 100% in just 2 min and the mean standard deviation of the rate constants was 0.0508 0.0274 s-1. The condition shown in pane l (D) is the exact sample extraction method specified by NIOSH Method 7903, which recommends a boiling time of 10 min. The results clearly show that SO2 adsorbed by silica gel and glass fiber filter can be completely converted to sulfate dur ing the extraction procedure. Rate constants for uncatalyzed oxidation reactio ns of sulfur(IV) are summarized in Table 5-2. As shown, the reported values range wide ly. Indeed, many reaction rates of the uncatalyzed oxidation of sulfur(IV) are often too high, because traces of transition metals in the water enhance the uncatalyzed process [ Huss et al. 1978]. Clark and Radojevic [ Clarke and Radojevic 1983] obtained a rate consta nt that was 7 times slower for the uncatalyzed reaction, when using Milli-R/Q water instead of distilled wa ter. Furthermore, the oxidation rates shown in Table 5-2 indicate that sulfur(IV) oxidation is strongly dependent on the pH. Radojevic [ Radojevic 1984] has recommended the uncatalyzed oxida tion rate be given by Equation 5-6: 2 / 1 2 3 2 4H SO 32 0 M/s d SO d t (pH 7) (5-6)

PAGE 106

106 The oxidation rates for the c onditions portrayed in panels (A) and (C) in this study are within this range. The oxidation rate for cond ition (C) was higher than that for condition (A) and the pH dependence can explai n this difference. Some may have concerns about potential oxidants or catalysts present in the solution that may change the oxidation rate. Th e IC extraction solution was prepar ed from fresh DI water; the hydrogen peroxide (H2O2) and ozone (O3) concentrations are negligible in the samples. The concentrations of trace metals should be very low, because the DI water system provided the fresh water with a resistivity of 18.2 M -cm, which is an ion-free solution (except H+ and OH-). The IC eluent solution used in this study was prepared by using commercially available sodium carbonate (EM Science). Based on the informati on provided by the supplier, the maximum iron content is 0.0005% (w/w). The corresponding iron c oncentration of the IC extraction solution is <8.5 10-8 M, and the effect at this concentration level, if any, is considered to be negligible. Sulfur Dioxide Adsorption If artifact SO2 causes overestimation of sulfuric aci d mist concentration, the degree of impact for a given sampling condition must be inve stigated. In the second group of experiments, SO2 concentration, sampling flow rate, and sampli ng time were examined to assess their effects. Sulfur dioxide concentration Two runs of artifact sulfate concentrations under various SO2 concentrations sampled at 0.3 Lpm for 2 h are displayed in Figure 5-3A Runs A and B were conducted under the same conditions to examine the reproducibility. Results fr om these two runs had the same trend and the 10% variation of feed SO2 concentration could explain th e difference of artifact sulfate concentrations between these runs. The sulf ate concentration, or the interference of SO2, increased as the inlet SO2 concentration increased. Figure 53A indicated that, when the inlet

PAGE 107

107 SO2 concentration was 0.8 ppm ( 10%), the mean artifact sulfate concentration was 190 g/m3. To assess the relative amount adsorbed, time-we ighted collection percentage (TWCP) was adopted, which is defined as the percentage of feed SO2 concentration collected by the silica gel tube over the given sampling time, i.e., (collected sulfate concentration/total sulfate concentration) 100%. The TWCP had the opposite tre nd, i.e., the TWCP was high when low SO2 concentration was introduced. The adsorption isotherm dictates that the adsorbed amount increases but gradually plateaus as vapor concen tration increases. For a linearly increasing concentration, the percentage therefore decreases. When low SO2 concentration (0.2 ppm) was introduced, the TWCP reached 16% after 2 h of sampling at 0.3 Lpm, or equivalent to a 125 g/m3 sulfate artifact. As SO2 concentration increased to 0.8 ppm, the TWCP reached 6%, which is equivalent to a 190 g/m3 sulfate artifact after 2 h of sampling. Sampling flow rate The artifact sulfate concentrations a nd the corresponding TWCPs with an inlet SO2 concentration of 1.5 ppm ( 10%) under various flow rates are exhibited in Figure 5-3B. The sulfate concentration at a low flow rate (0.2 Lpm) was 374 g/m3, whereas, at the higher flow rate, 0.5 Lpm, it was 163 g/m3. The TWCP at a low flow rate was also higher than that at a high flow rate. Collection by diffusion is a f unction of residence time, and it increases as residence time increases. As flow rate increase d, the residence time d ecreased, resulting in a lower artifact sulfate co ncentration. The NIOSH method recommends a flow rate lower than 0.3 Lpm if HF is present. However, as shown, a low flow rate yielded a higher amount of sulfate artifact.

PAGE 108

108 Sampling time The artifact sulfate concentr ations and the TWCP, as a f unction of sampling time, are shown in Figure 5-3C. The TWCP for 2 h of sampling of 0.8 ppm SO2 were 16%, 9.2%, 5.4%, and 4.6% for 0.2, 0.3, 0.4, and 0.5 Lpm, resp ectively. At 0.2 Lpm (2 h sampling), 0.8 ppm SO2 can cause artifact su lfate amount of 498 g/m3, which can significantly affect the accurate determination of sulfuric acid mist concentration. Increasing the sampling time to 8 h yielded an artifact at 0.2 Lpm of 172 g/m3. The artifact sulfate concentr ation decreased as sampling time increased. This was due to the decrease of adsorp tion rate as the effective adsorption sites were consumed over time. For the same reason, the TWCP also decreased. Both glass fiber filter and silica gel can adsorb SO2. The adsorptions of the glass fiber filter and the silica gel are displayed in Figure 5-4 using collection index (CI), which is defined as the artifact sulfate amount (given in micrograms) divided by the feed SO2 concentration (expressed in units of pm). As shown, the adsorbed SO2 amount increased as the sampling time increased for both glass fiber filte r and silica gel, although the patt erns were not the same. The adsorption of SO2 on the glass fiber filter was very quick and accounted for the majority of the adsorption in the first few hours. However, it was saturated within 2 3 h and the amount increased only slightly af ter that. In contrast, the adsorbed amount of SO2 on silica gel increased as the time increased. The adsorbed SO2 amounts under various SO2 concentrations, flow rates, and sampling times indicate that SO2 can cause significant interferen ce. It can be as high as 500 g/m3 for 0.8 ppm SO2 at a sampling time of 2 h and flow rate of 0.2 Lpm. The OSHA regulation for sulfuric acid mist concentration is 1 mg/m3. The result indicates that the interference caused by SO2 cannot be neglected when the mist concentration is low and SO2 concentration is high (e.g., > 0.5

PAGE 109

109 ppm). The artifact can affect the compliance status of a plant. Therefor e, the silica gel sampler containing a glass fiber filter plug and two sections of silica gel is not suitable for sampling sulfuric acid mist under such conditions. Sulfur Dioxide Adsorption Model To estimate the artifact for a given sampling condition, a modified model based on a deactivation model [ Kopac and Kocabas 2002; Yasyerli et al. 2001] that considers flow rate, sampling time and SO2 concentration was developed, which is shown in Equation 5-7. The firstorder deactivation rate constants ( kd, and ksSo) can be obtained by fitting experimental data. t k Q S k SO SOt d o s O 2 2 4exp exp 1 (5-7) where t 2 4SO is the artifact sulfate con centration (expresse d in units of g/m3), the conversion factor of SO2 to SO4 2(1.67), [SO2]o the initial SO2 concentration (expressed in unit of g/m3), ksSo the observed adsorption ra te (expressed in unit of cm3/min), Q the flow rate (given in units of cm3/min), kd the first-order deactivation rate constant (given in units of min-1), and t the time (expressed in minutes). Results of the regression analys is of the experimental data are given in Table 5-3, which shows a good relationship between the experimental data and the estimated values. The R2 values for the regression were 0.81 for the silica gel secti on. Although this model was developed for the adsorption on sili ca gel, the adsorption on glass fiber filter also fit well. The R2 value was > 0.92. The relation between the measur ed sulfate concentration from the silica gel tube and the concentration from the model predic tion is displayed in Figure 5-5. Most predictions are in the range of 35% artifact sulfate concentration from the measurement. As demonstrated, the

PAGE 110

110 deactivation model gives reasonably good predictions of artifact sulfate concentrations obtained in the silica gel tube. Thus, it can be used for correcting the artifacts in sulfuric acid sampling if the SO2 concentration is available. The application of this model s hould be further investigate in a field sampling. Summary NIOSH Method 7903, which uses one section of gl ass fiber filter and tw o sections of silica gel, has been developed to determine the total concentrations of acid mists in workplace air although certain gases are suspected to cause inte rference. In this study, experiments were carried out to investigate the roles of S(IV) oxidation and sulfur dioxide (SO2) adsorption in causing artifacts in sulfuric acid measurement. First, S(IV) oxidation under 4 combinations of water bath temperature and Na2CO3 solution concentration was ex amined to investigate the effect of the extraction process of NIOSH Met hod 7903. It was shown th at S(IV) oxidation to form sulfate could reach 100% in just 2-3 min following the extraction process of NIOSH Method 7903. The results demonstr ate that using the procedure, SO2 adsorbed by the silica gel and the glass fiber filter easily yields sulfate artifact. Sulfur dioxide adsorption under various flow rates, SO2 concentrations and sampling times was also investigated. The experimental data were fitted into a deactivation model to determine the adsorption rate constant and the deactivation rate constant. The model can serve as a tool for estimating the sulfate artifact if SO2 concentration is available.

PAGE 111

111 Table 5-1. Experiment al conditions of SO2 adsorption Set Flow Rate (Lpm) Sampling Time (h) [SO2] (ppm) 1 0.3 2 0.2, 0.4, 0.6 and 0.8 2 0.4 and 0.5 3 0.2, 0.6, 1.1, 1.4 and 1.6 3 0.2, 0.3, 0.4 and 0.5 3 and 8 1.5 4 0.2, 0.3, 0.4 and 0.5 2, 3, 4, 8, 10 and 12 0.8

PAGE 112

112 Table 5-2. Rate constants for uncataly zed oxidation reaction of sulfur(IV) by oxygen Reaction conditions k (s-1) pH Temp, T (oC) Reference 1.1810-7 3.0 25 Tanaka [1987] 1.7210-7 4.0 25 Tanaka [1987] 9.1910-7 5.0 25 Tanaka [1987] 1.6310-6 6.0 25 Tanaka [1987] 1.310-5 a 8.28.9 25 Clarke and Radojevic [1983] 9.510-5 b 8.28.9 25 Clarke and Radojevic [1983] 1.710-3 6.8 25 Scott and Hobbs [1967] 3.010-3 2.04.0 25 Miller and Pena [1972] 1.310-2 8.28.8 25 Fuller and Crist [1941] a In Milli-R/Q Water. b In distilled water

PAGE 113

113 Table 5-3. Rate parameters obtained using Equation 5-6 Glass fiber filter Silica gel Q ksSo kd R2 ksSo kd R2 Lpm mlpm min-1 mlpm min-1 0.2 0.0254 0.0026 0.95 0.0119 0.0012 0.81 0.3 0.0226 0.0025 0.93 0.0114 0.0013 0.87 0.4 0.0169 0.0025 0.94 0.0096 0.0015 0.92 0.5 0.0160 0.0026 0.92 0.0089 0.0014 0.81 Mean 0.0202 0.0026 0.0104 0.0014 Stdeva 0.0045 5.810-5 0.0014 0.0001 Stdeva: Standard deviation

PAGE 114

114 SO2 gas Bubbler 200 Mass flow controller SO2 10 ppm Zero Air 0.8 75 400 Silica gel tube Rotameter Exhaust Impinger Figure 5-1. Experimental setup for sulfur dioxide adsorption

PAGE 115

115 (b) Time (min) 0246810 [SO3 2-]t/[SO3 2-]o 0.0 0.2 0.4 0.6 0.8 1.0 (d) Time (min) 0246810 [SO3 2-]/[SO3 2-]o 0.0 0.2 0.4 0.6 0.8 1.0 Median Mean (a) Time (min) 0102030405060 [SO3 2-]t/[SO3 2-]o 0.0 0.2 0.4 0.6 0.8 1.0 (c) Time (min) 0102030405060 [SO3 2-]t/[SO3 2-]o 0.0 0.2 0.4 0.6 0.8 1.0 A C Figure 5-2. Sulfur(IV) oxidation under four conditions. A) DI water without a water bath. B) DI water with a water bath. C) 9 and 18 mM Na2CO3 solution without a water bath. D) 9 mM Na2CO3 solution with a water bath B D

PAGE 116

116 0 100 200 300 400 00.20.40.6 Flow rate (Lpm)[SO4 2-] ( g/m3)0 2 4 6 8TWCP (%) Sulfate (3 h) Sulfate (8 h) TWCP (3 h) TWCP ( 8 h ) (b) 0 100 200 300 400 500 600 051015 Sampling time (h)[SO4 2-] ( g/m3)0 3 6 9 12 15 18TWCP (%) Sulfate (0.2 Lpm) Sulfate (0.3 Lpm) Sulfate (0.4 Lpm) Sulfate (0.5 Lpm) TWCP (0.2 Lpm) TWCP (0.3 Lpm) TWCP (0.4 Lpm) TWCP (0.5 Lpm)(c) 0 50 100 150 200 250 00.20.40.60.81 SO2 concentration (ppm)[SO4 2-] ( g/m3)0 5 10 15 20TWCP (%) Sulfate (A) Sulfate (B) Mean sulfate TWCP (A) TWCP (B) Mean TWCP(a) A B C Figure 5-3. Artifact sulfate conc entrations and time-weighted co llection percentages (TWCPs). A) Various SO2 concentrations with a flow rate of 0.3 Lpm and sampling time of 2 h. B) Various sampling flow rate s. C) Various sampling times and sampling flow rates. Solid symbols represent artifact sulfate c oncentrations, and open symbols represent their corresponding TWCPs

PAGE 117

117 0 5 10 15 20 25 30 036912Time (h)CI ( g/ppm)(b) 0 5 10 15 20 25 30 036912 Time (h)CI ( g/ppm) Glass fiber filter Silica gel Total (d) 0 5 10 15 20 25 30 036912 Time (h)CI ( g/ppm)(a) 0 5 10 15 20 25 30 036912 Time (h)CI ( g/ppm)(c) A B C D Figure 5-4. Collection index (CI) at four flow rates (Lpm). A) 0. 2. B) 0.3. C) 0.4. D) 0.5

PAGE 118

118 0 100 200 300 400 500 600 0100200300400500600 [SO4 2-]measured ( g/m3)[SO4 2-]simulated ( g/m3)35% 35% 10% 10% Figure 5-5. Relationship between sulfate concentrations from the measurement versus from the model

PAGE 119

119 CHAPTER 6 MINIMIZATION OF ARTIFACT S IN SULFURIC ACID MIST MEASUREMENT USING NIOSH METHOD 7903 Background NIOSH Method 7903 [NIOSH, 1994] is an approved method set by the OSHA for measuring the total concentrati on of acidic aerosols and gases, including HF, HCl, HBr, HNO3, H2SO4 and H3PO4. It is the method commonly used by th e health and safety staffs in the phosphate industry [Hsu et al., 2007b] as well as other occupati onal environments such as the semiconductor industry, lead batte ry factories, aluminum smelti ng, machining, electroplating processes and even disaster response [Healy et al., 2001; Lue et al., 1998; Tsai et al., 2001; Wallingford and Snyder, 2001]. The sampler used for NI OSH Method 7903, a silica gel tube, consists of one section of glass fiber filter plug followed by two sections of silica gel. The glass fiber filter plug is designed to filter out the majo rity of aerosols while the silica gel sections are used mainly to adsorb acidic gases. The NIOS H recommended sampling flow rate range is 0.2 0.5 Lpm, except that less than 0.3 Lpm should be used for HF. The collected samples are desorbed in eluent and the a liquots are analyzed by IC. In evaluating the method, NIOSH researchers reported nearly 100% collection efficiency for acidic gases (HCl, HF and HNO3) [Cassinelli, 1986; Cassinelli and Taylor, 1981]. For aerosols (H2SO4 and H3PO4), around 90% efficiency (94.84.8% for H3PO4 and 864.6% for H2SO4) was reported when the samples collected on the glass fibe r filter section and the front silica gel section were combined. Ortiz and Fairchild [1976] reported ap proximately 70% of the aerosol mass was collected by the glass fiber filter plug although the efficiency varied depending on the size distribution of the sampling aerosol [Chen et al., 2002]. Sulfur dioxide is a species which can be adsorbed by the glass fiber filter [Chow, 1995; Coutant, 1977; Lee and Mukund, 2001; Watson and Chow, 2001] and the silica gel [Fox and

PAGE 120

120 Jeffries, 1979; Kopac and Kocabas, 2002; Stratmann and Buck, 1965]. Once adsorbed, it is subsequently transformed into sulfate in the an alytical process that causes an overestimate of sulfuric acid mist concentration [Hsu et al., 2007c]. Hsu et al. [2007c] carried out experiments in a laboratory system that verifi ed and quantified the artifacts resulting from the adsorption and conversion of interfering SO2 into sulfate. A deactivation model [Kopac and Kocabas, 2002] was modified for estimating the artifact su lfate concentration based on the various SO2 concentrations, flow rates and sampling times. Denuder systems have been widely used for air sampling [Acker et al., 2005; Hayami, 2005; Huang et al., 2004; Pathak and Chan, 2005; Sioutas et al., 1996; Tsai et al., 2004]. Various types of denuders have been developed and are commercially available. The denuders wall is coated with pertinent adsorbents dependin g on the gaseous species of interest. When air passes through the denuder, gas mo lecules with large diffusivity can diffuse to the wall and get adsorbed. Basic adsorbent (e.g. sodium carbonate (Na2CO3)/glycerol) can be used for acidic gases, such as HCl, HNO3, SO2 and HNO2. Aerosols can be collected in the filter that follows. As a consequence, this design can reduce the in terference of gaseous species on aerosols. The removal efficiency of SO2 using an annular denuder syst em, depending on the operating flow rate, has been reported to be > 99% when operated at a low flow rate [Possanzini et al., 1983]. The denuder system [Koutrakis et al., 1993] has also been applied for ambient air sampling and can provide high SO2 collection efficiency. Accurate determination of sulfuric acid mi st concentration in fertilizer manufacturing facilities is seminal to the evalua tion of its occupational exposure. Minimizing the artifact due to SO2 gas in the sampling process is essential in reaching this goal. The objectives of this chapter are twofold: (1) to explore th e use of a denuder for removing SO2 gas from the sampling volume

PAGE 121

121 to reduce the artifact sulfate and, (2) to assess the applicability of the deactivation model for correcting the artifacts fo r the phosphate fertilizer production environment. Methods Two methods, a sampling system cooperati ng with a honeycomb denuder system (HDS) and a deactivation model, were applied to mini mize the artifact sulfate. Field sampling was carried out to examine the artifact removal effi ciency when the HDS was applied to remove the interfering SO2 before entering the standard sampling train. Experiments were conducted to characterize SO2 adsorption at high SO2 concentrations encountered in the field sampling conditions. The deactivation model was used to es timate the artifact sulfate concentration based on known SO2 concentrations. Field Sampling Field sampling was carried out on top of the sulfuric acid pump tank area at seven phosphate fertilizer pl ants in Florida [Hsu et al., 2007c]. Four samples were acquired at each of six sites and five samples were collected at the one remaining plant. Sampling time was 8 h for each sample. Silica gel tubes, a cascade impactor and a honeycomb denuder system were applied for the field sampling. Three sets of sampling trains, sh own in Figure 6-1, were employed. (A) A silica gel tube This method was used for total sulfuric acid mist concentration following NIOSH Method 7903 (N = 29). (B) The HDS followe d by two silica gel tubes in parallel The HDS (coating solution: 1% Na2CO3/glycerol) was used to remove SO2 gas before the air entered the silica gel tubes. The measurement also provided SO2 concentration. Two silica gel tubes were applied for replication (N = 29 for the HDS sample ; N = 58 for the silica gel tube sample). (C) A cascade impactor (Mark III, U. Washington) with Zefluor membrane filters (P5PJ001, Pall

PAGE 122

122 Corp.) This sampling system was used for sulfuric acid mist concentration with size-resolved information (N = 29). Zefluor membrane filter s can provide high aerosol collection efficiency with low interaction with acidic gases [Chow, 1995; Watson and Chow, 2001]. All three sampling trains were set side by side for parallel sampling. All silica gel tube samples were analyzed fo llowing the sample preparation procedures outlined in NIOSH Method 7903 [NIOSH, 1994], including adding 9 mM Na2CO3 of 10 mL and a water bath at 100 C for 10 min. The HDS sample was extracted by 10 mL DI water (Nanopure Diamond, Barnstead), an d 1 mL hydrogen peroxide (0.6%) was then added into 1 mL sample solution to oxidize sulfite to sulfate fo r the analysis. The cascade impactor sample was extracted by 10 mL DI water with a 1-h ultrasonic bath. Sulfat e concentrations of all samples were analyzed via IC (Model ICS 1500, Dionex). Deactivation Model A deactivation model [Hsu et al., 2007c], Equation 6-1, has b een developed to estimate the artifact sulfate concentration. In developi ng this model, experime nts were conducted at SO2 concentrations ranging from 0.2 to 1.6 ppm. Hence, the model can be applied for SO2 concentrations in this range. 2so 42d o tkS SO SO1expexpkt Q for 0.2 ppm < 2 oSO< 1.6 ppm (6-1) 24 tSO : artifact sulfate concentration (g/m3) = 1571, conversion factor 2 oSO= feed SO2 concentration (ppm)

PAGE 123

123 ksSo = observed adsorption rate (mLpm/min), 16 for the glass fiber filter and 8.9 for the silica gel at the flow rate of 500 mLpm Q = flow rate (500 mLpm) kd = first order deactiva tion rate constant (min-1), 0.0026 for the glass fiber filter and 0.0014 for the silica gel at th e flow rate of 500 mLpm t = time (min) Sulfur Dioxide Adsorption High SO2 concentrations, 1.6 ppm to 5.6 ppm, were obs erved during the field sampling described in the previous section. However, th e parameter values in the deactivation model have not been validated for the ar tifact estimate at high SO2 concentrations. Therefore, experiments were conducted to quantify the SO2 adsorption by the silic a gel tube under high SO2 concentrations ranging from 1.6 ppm to 5.6 ppm at 0.5 Lpm flow ra te with 8 h of sampling time (condition employed for the field sampling). The e xperimental setup has been described in Hsu et al. [2007c]. SO2 gas from a cylinder (10 ppm, relative uncertainty = 5%) was mixed with zero air (Model 111, Thermo Electron Instrument ) to obtain the desired concentration. SO2 gas was then passed through the silica gel tube for the adsorption. An impinger with 100 mL of 9 mM Na2CO3 solution was employed downstrea m to collect the residual SO2 gas. Sulfate concentration was determined following the analyti cal procedures described in the field sampling section. To evaluate residual sulfate of both gl ass fiber filter and silica gel, eighteen silica gel tubes as received were analyzed for their sulf ate concentration following the sample preparation procedures of NI OSH Method 7903.

PAGE 124

124 Results and Discussion Field Sampling Collection efficiency and concentration of SO2 The HDS consists of two honeycomb denuders in series. Collection efficiency, was determined according to Equation 6-2: 22 2SO(HD1)SO(HD2) (%)100% SO(HD1) (6-2) where [SO2(HD1)] and [SO2(HD2)] are SO2 concentrations collected by the first and second honeycomb denuders, respectively. The mean SO2 collection efficiency standard deviation of the HDS was found to be 95.7 6.8% (N = 29), demonstrati ng that the deployed HDS could effectively remove SO2 gas from the sample gas. Regarding the SO2 concentration at the phosphate ferti lizer plants, it ranged from 34 ppb to 5.6 ppm. Although it varied significantly fr om plant to plant, for each plant the SO2 concentration level varied within a limited range during the sampling period. Table 6-1 displays the statistics of SO2 concentrations at each plant. At plant F, sampling was conducted at two locations where SO2 concentrations differed greatly. Hence, SO2 concentrations at plant F are shown separately for each location. Ratio of S-SO4 2-/ S-SO2 The deployment of the HDS was to collect SO2 gas; however, sulfat e aerosols might also be collected by the HDS which could cause an overestimate of SO2 concentration. To determine if the presence of sulfate significantly affects the SO2 concentration measurement, the S-SO4 2-/SSO2 ratios (elemental sulfur from SO4 2collected by the silica gel tubes, SGHA and SGHB in the second sampling train, to elemental sulfur from SO2 collected by the HDS) for all samples were calculated and are shown in Figur e 6-2. As shown, the maximum was 0.13 and it occurred at a

PAGE 125

125 very low SO2 concentration. At high SO2 concentrations where the focus of this study is, they were below 2%. The low ratios demonstrate that the honeycomb denuder system can be used for measuring SO2 concentration accurately and the interfer ence of particulate sulfate is negligible. Aerosol loss of HDS The aerosol loss in the HDS causes an undere stimate of sulfate concentration. Two mechanisms causing the aerosol lo ss in the HDS are: (1) the aero sol with size larger than 10 m can be removed by the impactor of the HDS, and (2) the aerosol can diffuse to the wall of the denuder. The aerosol concentrati on with size-resolved informa tion from the cascade impactor (CI) can be employed to correct the first mechanism. The sulfate diffusional loss, 2 4 DFSO can be calculated by Equation 6-3. i224i4 DFCISODFSOi (6-3) DFi : aerosol deposition fraction of aerosol size range i. i24 CISO : sulfate concentration of aerosol size range i measured by the CI. The aerosol deposition (DF) in the honeycom b denuder due to diffusion can be calculated according to Equation 6-4a and Equation 6-4b [Hinds, 1999]. 2/3 iiDF= 5.5 i3.77 for < 0.009 (6-4a) iiiDF= 10.819exp(11.5)0.0975exp(70.1) for 0.009 (6-4b) where i iDLN Q : dimensionless deposition parame ter for particle size range i Di: diffusion coefficient of th e particle size range i (cm2/min) L: the length of the tube (= 9.6 cm)

PAGE 126

126 N : the number of tubes (= 160) Q: the volume flow rate th rough the tube (= 500 mLpm) Table 6-2 shows the mean ratio and its range of the sulfate diffusional loss, the sulfate loss due to the HDS inner impactor (2424 HSO> 10 mSO derived from the CI measurement) and the total sulfate loss (loss24 HDSO = 2424 HSO> 10 mSO +2 4 DFSO ) to the total sulfate concentration from the CI (24 CISO ) at each plant. The mean sulfate loss due to the aerosol diffusi onal loss at each plant ranged from 3.2% to 6.3%. For all plants, the mean loss was 4.8% de monstrating that the effect was minor. On the other hand, the sulfate loss due to the impactor, shown in Table 6-2, was highly significant. The mean loss ranged from 3% to 43% demonstrating that the loss varied remarkably. If the HDS impactor sample were analyzed in this study, the information can be used to directly correct the majority of the loss. The CI information is then no longer needed. Sulfur Dioxide Adsorption Figure 6-3 shows the artifact sulfat e concentration as a function of SO2 concentration. As feed SO2 concentration increased, the artifact sulfate concentration increased. This is because more SO2 can be adsorbed by the silica gel and the gla ss fiber filter, and the glass fiber filter and silica gel were not saturated yet. The relation between the feed SO2 concentration (1.6.6 ppm) and the artifact sulfate con centration can be described by Equa tion 6-5. In this study, this equation was applied for SO2 concentrations between 1.6 a nd 5.6 ppm while the deactivation model was applied for SO2 concentrations from 0.2 to 1.6 ppm. The intercept in Equation 6-5 might be from the residual sulfate to be discussed in the next section. Since the ratio of the mean amount of residual sulfate of the s econd silica gel section to that of the first silica gel section was

PAGE 127

127 2.1, shown in Table 6-3, outliers were defined as those with ratios exceeding 2.1. Theoretically, the first section should collect more SO2 gas than the second section. However, the high residual sulfate artificially yields a highe r sulfate concentrati on in the second section. Since the mean ratio of the sulfate concentration in the second sec tion to that of the first section is 2.1, outliers are defined as exceeding this ratio. 2 42o tSO20.04[SO]53.32 for 1.6 ppm < 2o[SO]< 5.6 ppm (6-5) 2 4 tSO : artifact sulfate concentration ( g/m3) 2o[SO]: feed SO2 concentration (ppm) To determine the effect of humidity, experi ments were conducted for the relation between adsorbed water amount and sampling time. RH s of 35% and 85% for various sampling times and flowrates were investigated. It was found that the silica gel tube reached saturation quickly. Under these sampling flowrates and sampling tim es, the adsorbed water amount was around 0.2 g which is one-third of the silica gel weight. When 0.2 Lpm flowrate and RH of 35% were applied, the adsorbed water am ount reached saturation in two hour s. The results demonstrate that the sampling tube is generally operated under the saturation condition over a wide range of ambient RH. Residual Sulfate in Silica Gel Tube Due to the non-zero intercept in Equation 6-5, 18 silica gel tubes as received were analyzed for residual sulfate concentration. The statistical results of the residual sulfate concentrations in different sections for two batches purchased from the vendor are shown in Table 6-3. The first batch was used for the experiment conducted in this research while the second batch was purchased in 2004 and used for a pr ior study described in Chapter 3 [Hsu et al., 2007b]. For the

PAGE 128

128 first batch, the mean residual su lfate concentrations fr om the glass fiber filter, first and second sections of silica gel in the a liquots were 3.09, 5.23 and 10.77 g, respectively, which are equivalent to 12.9, 21.8 and 44.9 g/m3 (in the air) at the sampling conditions of 8 h and 0.5 Lpm. Added together, the mean total sulfat e concentration due to the residual was 79.5 g/m3. For the maximum residual sulfate co ncentration, the artifact sulfat e concentration in the air was 439.3 g/m3. The total sulfate concentration in the ai r corresponding to the st andard deviations of the sampler was 132 g/m3. This is higher than the mean residual sulfate concentration, indicating that the residual sulfat e concentrations were quite vari able. In contrast, the residual sulfate of the second batch was much lower than that of the first batch. Its mean total residual sulfate concentration was only 15.9 g/m3. In summary, the residual sulfate concentration can contribute greatly to the artifact sulfate concentration in the air and the impact of variation of residual sulfate in the silica ge l tube should not be ignored. Therefore, NIOSH Method 7903 is not suitable for low sulfate concentration sampling, e.g., ambient air sampling. Minimization of Artifact Sulfate Two methods were applied to minimize the ar tifact sulfate and th eir effectiveness was evaluated. The first method was the SO2 adsorption model described in Equation 6-1 and Equation 6-5. This method is based on using the known SO2 concentration to calculate the artifact sulfate concentrati on. In this study, the SO2 concentration was determined by the denuder as discussed earlier. The corrected sulfate concentration based on the model, 24 modelSO can then be calculated according to Equation 6-6: 2-2-2444 modelSGtSO=SOSO (6-6) where 24 SGSO: sulfate concentration from the SG which includes artifact sulfate.

PAGE 129

129 The second method used to minimize artifact sulfate was the HDS (sampling train 2). However, it has the potential to remove true sulfate aerosol. Therefore, the sulfate concentrations from SGHA and SGHB need to be corrected and the correcting equation is shown in Equation 6-7. loss2-2-2444 SGHAC (or SGHBC)SGHA (or SGHB)HDSO = SO + SO (6-7) Figure 6-4 shows the sulfate concentrations fr om the SG, the SGHC (the average of the corrected SGHA (SGHAC) and the corrected S GHB (SGHBC)), the model and the CI as a function of SO2 concentration. The sulfat e concentrations from the SG were always higher than those from other samplers. For most samples, the sulfate concentration from the CI was the lowest and the sulfate concentrations from th e model and SGHC were between the SG and the CI. For low SO2 concentration, the sulfate concentrations from the SG and the model were similar due to the expected small artifact su lfate produced. The rela tive error (E) shown in Equation 6-8 was applied to evaluate the differe nce and the results are shown in Table 6-4. 2-244 CI 24 CISOSO E= SO sampler (6-8) 24SO s ampler : sulfate concentration measur ed by SG, SGHAC and SGHBC. The relative errors of the NIOSH method a nd the model were relatively large when SO2 concentration was lower than 0.05 ppm; their me an values were 4.4 and 4.2, respectively. Between 0.1 and 2.7 ppm, the sulfate concentration from the SG can be 1 to 7 times higher than that from the CI. Meanwhile, both the mode l and the honeycomb denude r reduced the sulfate concentration, making them closer to the CI results. When high SO2 concentration, ranging from 4.0 to 5.6 ppm, was observed, relative errors of all methods were low. However, it should be

PAGE 130

130 addressed that the artifact sulfat e concentration from the SG was stil l high and needed correction. For example, at SO2 concentration of 4.7 ppm it was 287.6 g/m3. The effectiveness, for reducing artifact sulfate by the SGHAC/SGHBC and the model can be calculated by Equation 6-9. 2-244 SGSGHAC/SGHBC or model 2-244 SGCISOSO reduced artifact sulfate %100% true artifact sulfate SOSO (6-9) They were 70 32% and 39 36% for the SGHAC/SGHBC and the model, respectively. However, they were still higher than the CI valu es. One possible reason for the difference is the residual sulfate of the silica gel tube discussed earlier. Overa ll, the model and the HDS can be applied for the minimization of the artifact sulfate. Aspiration Efficiency The aspiration efficiencies (asp, calm air ) for the cascade impactor, th e silica gel tube and the honeycomb denuder system were calculated by Equation 6-10 (Brockmann, 2001). 1 asp, calm air4 cosexp 12tsV U tsV Stk UStk (9-10) for 0 90 10-3 Vts/U 1, and 10-3 Stk 100. Vts: terminal settling velocity U: inlet sampling velocity = 0 is upward facing, 90 is horizontal Stk: Stokes number (= U/d) : the relaxation time of the particle

PAGE 131

131 d: inlet diameter The results with two orientations are shown in Figure 6-5. For th e vertical (facing up) sampling, the aspiration efficiency of the silica ge l tube is higher than others and the honeycomb denuder system and the cascade impactor have the similar aspiration efficiencies. In the field sampling, both the silica gel tube and the hone ycomb denuder system were laid horizontially while the cascade impactor was placed vertically. The efficiencies for both samplers in this geometry are lower than the cascade impactor as shown in Figure 65 which indicates the cascade impactor should collect more aerosols than others. However, this result was not observed in this study. The as piration efficiencies of three samplers were similar for aerosol smaller than 20 m where most of the aerosols in the st udy were located. Hence, it is also reasonable to apply the cascade impactor measur ements to the aerosol information for the honeycomb denuder system. Sulfate Mass Balance The sulfate mass balance between the silica gel t ube in the first sampling train (SG) and the two silica gel tubes in the second train (SGHA or SGHB) in the parallel samples was checked. Ideally, they can be equated as Equation 6-11: loss2-2-2-24444 SGSGHA (or SGHB)tHDSO = SO + SO + SO (6-11) The sulfate mass balance is shown in Figure 6-6. Except for a few, most data points show good mass balance, i.e., most of them were within the standard deviation of the residual sulfate concentration of the silica gel tube. The larg e relative error of sulfate shown in Table 6-4 indicates the existence of artifact sulfate on the SG. The good sulfate mass balance between the SG and the SGHAC/SGHBC supports the conclusion that SO2 gas is the key source of artifact sulfate.

PAGE 132

132 Summary A sampling train incorporating a honeycom b denuder system was applied for field sampling at seven phosphate fertilizer plants to evaluate its use for reduc ing the artifact sulfate concentration while preserving th e actual sulfuric acid mist c oncentration. The denuder system was designed to remove SO2 gas before the air entered the silica gel tube and to monitor SO2 concentration at the same time. A deactivati on model was also applied to correct for the presence of the artifact. The denuder system had 95.7 6.8% collection efficiency for SO2 gas, and the impact of sulfate aerosol on SO2 collection was negligible. SO2 concentrations at the seven plants ranged from 34 ppb to 5.6 ppm. Both the honeycomb denuder system and the deactivation model were shown to reduce the artifact sulfate concentration by 70% and 39%, respectively. However, they were still higher th an the sulfate aerosol concentration measured by a cascade impactor. One possible re ason is the residual sulfate in the glass fiber filter and the silica gel.

PAGE 133

133 Table 6-1. Statistics of SO2 concentrations (ppm) Site A B C D E F-1 F-2 G Mean 1.83 1.74 0.46 0.05 0.27 0.40 5.00 0.21 MIN 0.90 1.22 0.36 0.03 0.19 0.12 4.34 0.13 MAX 2.57 2.66 0.56 0.05 0.31 0.68 5.64 0.31 Sample no. 4 4 4 4 4 2 3 4

PAGE 134

134 Table 6-2. Mean and standard deviation of the su lfate loss to the total sulfate concentration 24 DF 24 CISO SO (%) 2424 HSO> 10 m 24 CISO SO (%) loss24 HD 24 CISO SO (%) 24 CISO ( g/m3) A 4.8 (4.0 5.3) 15.7 (8.3 26.8) 25.4 (19.0 34.8)92.4 (53.2 142.6) B 3.2 (2.5 3.9) 42.8 (31.6 52.6)49.1 (39.4 57.7)25.7 (21.1 32.3) C 4.0 (3.0 5.4) 26.3 (16.7 33.6)34.4 (22.7 44.4)31.0 (21.1 43.5) D 6.3 (5.9 6.6) 2.8 (1.3 6.6) 15.4 (14.4 18.4)85.3 (21.3 134.1) E 4.8 (4.0 5.4) 15.0 (4.4 34.9) 24.7 (14.6 42.9)32.0 (5.6 61.3) F-1 4.8 (4.8 4.8) 13.4 (10.3 16.4)23.0 (19.9 26.1)16.7 (11.7 21.6) F-2 4.7 (4.6 4.8) 11.6 (10.7 12.5)20.9 (20.2 21.7)578.2 (412.7 686.3) G 5.7 (5.3 6.3) 5.8 (1.0 11.3) 17.3 (13.7 21.9)44.3 (10.3 108.7) Total 4.8 (4.0 5.3) 15.7 (8.3 26.8) 25.4 (19.0 34.8)92.4 (53.2 142.6)

PAGE 135

135 Table 6-3. Statistical results of the residual sulfate concentrations of silica gel tubes Sulfate ( g) Mean Standard deviation MAX MIN 1st Batch (N = 10) Glass fiber filter 3.1 2.3 9.0 1.1 Silica gel 1st section 5.2 5.6 18.3 0.7 Silica gel 2nd section 10.8 23.9 78.1 0.9 nd st2 section 1 section 2.1 4.3 4.3 1.3 2nd Batch (N = 8) Glass fiber filter 0.6 0.6 1.9 0.0 Silica gel 1st section 2.0 0.5 2.6 1.2 Silica gel 2nd section 1.3 0.6 2.8 0.9 nd st2 section 1 section 0.6 1.4 1.1 0.7

PAGE 136

136 Table 6-4. Relative error of 4 samplers Relative Error SG SGHAC SGHBCModel [SO2] < 1.6 ppm (N=21) mean 4.46 1.73 1.86 3.30 stdev* 2.12 1.59 1.48 2.23 min 0.47 0.06 0.14 0.07 max 8.56 5.76 5.49 7.87 1.6 ppm < [SO2] < 5.6 ppm (N=8) mean 1.90 0.48 0.63 0.82 stdev 2.00 0.23 0.44 1.02 min 0.02 0.20 0.20 0.17 max 5.78 0.94 1.30 3.10 Stdev*: standard deviation.

PAGE 137

137 Silica gel tube Air In A Honeycomb denuder Silica gel tube Air In B C Collection plate No.1 Jet plate No. 2 Collection plate No. 2 Jet plate No. 3 Collection plate No. 3 Jet plate No. 4 Collection plate No. 4 Jet plate No. 5 Collection plate No. 5 Jet plate No. 6 Collection plate No. 6 Jet plate No. 7 Collection plate No. 7 Filter Inlet nozzle Outlet Collection plate No.1 Jet plate No. 2 Collection plate No. 2 Jet plate No. 3 Collection plate No. 3 Jet plate No. 4 Collection plate No. 4 Jet plate No. 5 Collection plate No. 5 Jet plate No. 6 Collection plate No. 6 Jet plate No. 7 Collection plate No. 7 Filter Inlet nozzle Outlet Rotameter Pump Rotameter Pump Rotameter Pump Cascade Impactor Figure 6-1. Three sampling trains. A) NIOS H Method 7903. B) Modifi ed NIOSH Method 7903. C) Cascade impactor sampling system

PAGE 138

138 SO2 (ppm) 0123456 [S-SO4 2-]/[S-SO2] 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 SGHA SGHB Figure 6-2. S-SO2/S-SO4 2as a function of SO2 concentration. SGHA and SGHB are the silica gel tubes in the second sampling train

PAGE 139

139 SO2 (ppm) 123456 Artifact Sulfate ( g/m3) 60 80 100 120 140 160 180 24t2o 2[SO]=20.04[SO]+53.32 R=0.9463Figure 6-3. Artifact sulfate conc entration as the function of SO2 concentration

PAGE 140

140 SO 2 (ppm) 0.00.51.01.52.02.53.0 Sulfate ( g/m 3 ) 0 100 200 300 400 500 SO 2 (ppm) 3.03.54.04.55.05.56.0 Sulfate ( g/m 3 ) 0 200 400 600 800 SG SGHC Model CI Figure 6-4. Comparison of the sulfate concentrat ions from the CI, SG, SGHAC and SGHBC. A) SO2 concentration lower than 3 ppm. B) SO2 concentration higher than 3 ppm. A B

PAGE 141

141 Aerodynamic Diameter ( m) 0102030405060 Aspiration Efficiency 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 SG V CI V HD V SG H HD H Figure 6-5. Aspiration effi ciency of three samplers. (SG: si lica gel tube, CI: Cascade impactor; HD: honeycomb denuder system, V: vertical; H: horizontal)

PAGE 142

142 0 200 400 600 800 1000 02004006008001000 Sulfate (SG) ( g/m3)Sulfate (SGHAC or SGHBC) ( g/m3) SGHAC SGHBC 1:1 Stdev Figure 6-6. Sulfate mass balance between SG a nd SGHAC/SGHBC (Stdev*: standard deviation of the residual sulfate)

PAGE 143

143 CHAPTER 7 CONCLUSIONS Aerosol sampling using the dichotomous sampler, NIOSH Method 7903, and a cascade impactor was carried out at five types of loca tions at eight phosphate plants and two background sites to determine the worker e xposure to sulfuric acid mist con centration in phosphate fertilizer plants. Artifact sulfate observed in usi ng NIOSH Method 7903 for sampling in phosphate fertilizer manufacture process was also investigat ed in this study. To minimize the artifact sulfate, two methods were developed. Five conclusions can be drawn from this study. Conclusion 1 The highest sulfate c oncentration, 0.185 mg/m3, in the plants was obtained at the sulfuric acid pump tank area. Should monitoring of personal exposure to sulfuric acid mist be required, efforts should focus on workers with activities in this area where concentrations approach than the TLV-TWA standard of 0.2 mg/m3 recommended by ACGIH for the thoracic fraction of sulfuric acid aerosol. At th e attack tank area, fluoride wa s the dominant species and the maximum fluoride concentration in PM10 was 462 g/m3. At the rotating table/belt filter floor, phosphoric acid is separated from gypsum by rotati ng table/belt filter and the high temperature is favorable for the evaporation of phosphoric ac id and the maximum phosphate concentration in PM10 was 170 g/m3. On a scrub day, a weak sulfuric ac id solution is used to clean the piping and ductwork of the granulator for an aver age of 4 hours per day. Particulate sulfate concentrations were low during the scrubbing activity. At the truck loading/unloading station, the possible emission period is around 2-3 h/day, and this em ission is not continuous. The concentration levels at the loading/unloading st ation were low and were greatly influenced by outdoor conditions.

PAGE 144

144 Conclusion 2 Based on cascade impactor sampling, sulfur ic acid pump tank areas still had higher sulfuric acid mist concentrations than othe r types of locations, a nd sulfuric acid was the dominant chemical species. When high sulfuric ac id concentrations were identified, the aerosols were dominantly in the coarse mode. The most lik ely cause for high sulfuric acid concentrations at this location is the leakage of SO3. According to the impactor sampling results, 7 samples (total: 72) exceeded the ACGIH recommendation (0.2 mg/m3, thoracic fraction), and 2 samples (total: 72) were above th e OSHA regulation (1 mg/m3, total concentration). Meanwhile, there were 7 samples (total: 78) by the NIOSH method that exceeded the OSHA regulation. The sulfuric acid mist concentra tions from the NIOSH method were higher than those from the cascade impactor for the dominating majority of samples. The po ssible reason for this variation is the interaction between SO2 and silica gel/glass fiber. Conclusion 3 In phosphate fertilizer facilities, phosphoric acid and sulfuric acid mists were the major aerosol components for the product filter fl oors and the sulfuric acid pump tank areas, respectively. The possible source of phosphoric acid was evaporation and then condensation when it encountered cooler air. The curre nt OSHA 8-hour TWA PEL of phosphoric acid and sulfuric acid mist set at 1 mg/m3 was not exceeded on average, but could be exceeded at the product filter floors and the pump ta nk areas, respectively. It shoul d be noted that workers spend much less than 8 hours per day in the area, and th us the true time-weighted exposure level can be expected to be lower. Calcium and ammonium were the major species to neutralize the aerosol acidity at the sulfuric pump tank areas when acid loading was low. The aerosol thermodynamic model showed the modes of aerosol H+ concentration in 1.8 .8 m and 3.8 m for the aerosols with high sulfuric acid mist concentratio ns. These hygroscopic acid mists can grow in

PAGE 145

145 the high humidity condition in the upper re spiratory system, and aerosols with high H+ concentrations mainly deposit in the upper respiratory system. Sulf uric acid was found to play a much more prominent role than phosphoric ac id. The respiratory deposition projection of sulfuric acid mists is consistent with that of H+ ion and both components mainly deposit in the human head airway. However, extensive epidem iological studies of phosphate industry workers have not shown an increased incidence of any t ype of cancer resulting from these exposures. Conclusion 4 The artifact sulfate of NIOS H Method 7903 originating from SO2 gas was confirmed. SO2 can be adsorbed by the glass fiber filter and the silica gel and the adsorbed SO2 can be oxidized and transformed into sulfate during the extract ion procedures of NI OSH Method 7903. The interference from SO2 cannot be neglected when the sulfuric acid mist concentration is low and SO2 concentration is high (> 0.5 ppm). The artifact sulfate can affect the compliance status of a facility and should be corrected. A deactivati on model was developed to estimate the artifact sulfate concentration. Conclusion 5 Two methods, a honeycomb denuder system and a deactivation model, were applied to minimize the artifact sulfate concentration. The honeycomb denuder system efficiently adsorbed SO2 before it entered the silica gel tube while th e deactivation model was employed to calculate the artifact sulfate concentration. Both methods were proven to reduce the artifact. However, the sulfate concentrations from both methods we re still higher than th e sulfate concentration from the cascade impactor. One likely reason is the residual sulfate from the silica gel tube, which yields a mean artifact sulfate equivalent to 79.5 g/m3 for 8-h sampling.

PAGE 146

146 LIST OF REFERENCES ACGIH (2004), Sulfuric acid. TLV Chemical S ubstances. 7th Edition Documentation, 1-12. Acker, K., D. Moller, R. Auel, and D. Kalass (2005) Concentrations of nitr ous acid, nitric acid, nitrite and nitrate in the gas and aerosol pha se at a site in th e emission zone during ESCOMPTE 2001 experiment, Atmos. Res., 74, 507-524. Apol, A. G., and M. Singal (1987), Health H azard Evaluation, J. R. Simplot Company, Pocatello, ID, National Institute for Occupati onal Safety and Health, Cincinnati, Ohio. Baron, P. A., and K. Willeke (2001), Aerosol Measurement: Principles, Techniques, and Applications, 2nd ed., 1131 pp., Wiley, New York. Bassett, M., and J. H. Seinfeld (1983), Atmos pheric equilibrium-model of sulfate and nitrate aerosols, Atmos. Environ., 17, 2237-2252. Becker, P. (1989), Industry proce ss chemistry. Chapter 2., in Phosphates and Phosphoric Acid: Raw Materials, Technology, and Economics of the Wet Process, edited by P. Becker, pp. 43-177, Marcel Dekker, New York. Becker, P. (1989), Phosphates and Phosphoric Acid: Ra w Materials, Technology, and Economics of the Wet Process, 2nd ed., 740 pp., Marcel Dekker, Inc., New York. Benjamin, M. M. (2002), Water Chemistry, 668 pp., McGraw-Hill, Boston. Bhaskaran, R., N. Palaniswamy, N. S. Renga swamy, and M. Jayachandran (2004), Cost of corrosion in fertilizer industry A case study, Mater. Corros., 55, 865-871. Blair, A., and N. Kazerouni (1997) Reactive chemicals and cancer, Cancer Causes Control, 8, 473-490. Block, G., G. M. Matanoski, R. Seltser, and T. Mitchell (1988), Cancer morbidity and mortality in phosphate workers, Cancer Res., 48, 7298-7303. Brockmann, J. E. (2001), Sampling and Transport of Aerosols, in Aerosol Measurement: Principles, Techniques, and Applications, edited by P. A. Baron and K. Willeke, pp. 143195, Wiley, New York. Cassady, M. E., H. Donaldson, and S. Gentry (1975), Industrial hygien e survey report of Agrico Chemical Company, Pi erce, Florida, June 22-26, 1975., National Institute for Occupational Safety and Health, Cincinnati, OH. Cassinelli, M. E. (1986), Laboratory evaluatio n of silica-gel sorbent tubes for sampling hydrogen-fluoride, Am. Ind. Hyg. Assoc. J., 47, 219-224.

PAGE 147

147 Cassinelli, M. E., and D. G. Taylor (1980) Monitoring for airbor ne inorganic acids, Abstracts of Papers of the American Chemical Society, 180, 39. Cassinelli, M. E., and D. G. Taylor (1981) Monitoring for airbor ne inorganic acids, Am. Chem. Soc. Symp. Ser., 149, 137-152. Cassinelli, M. E., and P. M. Eller (1979), Ion chromatographic determination of hydrogen fluoride, paper presented at Annual Conferen ce, American Industrial Hygiene Conference and Exposition, Chicago, IL, 27 May 1 June. CFR Table Z-1 Limits for Air Contaminan ts, CFR (Code of Federal Regulation). Checkoway, H., R. M. Mathew, J. L. S. Hickey, C. M. Shy, R. L. Harris, E. W. Hunt, and G. T. Waldma (1985a), Mortality among workers in the Florida phosphate industry.1. Industrywide cause-specific mortality patterns, J. Occup. Environ. Med., 27, 885-892. Checkoway, H., R. M. Mathew, J. L. S. Hickey, C. M. Shy, R. L. Harris, E. W. Hunt, and G. T. Waldma (1985b), Mortality among workers in the Florida phosphate industry.2. Causespecific mortality relationships with work areas and exposures, J. Occup. Environ. Med., 27, 893-896. Checkoway, H., N. J. Heyer, and P. A. Deme rs (1996), An updated mortality follow-up study of Florida phosphate industry workers, Am. J. Ind. Med., 30, 452-460. Chen, C. C., C. H. Wu, S. H. Huang, and Y. M. Kuo (2002), Aerosol penetration through silica gel tubes, Aerosol Sci. Technol., 36, 457-468. Chow, J. C. (1995), Measuremen t methods to determine complia nce with ambien t air-quality standards for suspended particles, J. Air Waste Manage. Assoc., 45, 320-382. Clarke, A. G., and M. Radojevic (1983), Chlo ride-ion effects on the a queous oxidation of SO2, Atmos. Environ., 17, 617-624. Coutant, R. W. (1977), Effect of environmental variables on collection of atmospheric sulfate, Environ. Sci. Technol., 11, 873-878. Denzinger, H. F. J., H. J. Konig, and G. E. W. Kruger (1979), Fluorine rec overy in the fertilizer industry a review, Phosphor. Potassium, 103, 33-39. Dibb, J. E., R. W. Talbot, G. Seid, C. Jordan, E. Scheuer, E. Atlas, N. J. Blake, and D. R. Blake (2002), Airborne sampling of aerosol particle s: Comparison between surface sampling at Christmas Island and P-3 sampling during PEM-Tropics B, J Geophys Res-Atmos, 108, Art. No. 8230. Divita, F., J. M. Ondov, and A. E. Suarez ( 1996), Size spectra and atmo spheric growth of Vcontaining aerosol in Washington, DC, Aerosol Sci. Technol., 25, 256-273.

PAGE 148

148 Dong, S., and P. K. Dasgupta (1986), On the fo rmaldehyde-bisulfite-hydroxymethane-sulfonate equilibrium, Atmos. Environ., 20, 1635-1637. Englander, V., A. Sjoberg, L. Hagmar, R. Atte well, A. Schutz, T. Moller, and S. Skerfving (1988), Mortality and cancer morbidity in worker s exposed to sulfur-dio xide in a sulfuricacid plant, Int. Arch. Occ. Env. Hea., 61, 157-162. FDEP (2003), Florida Air Monito ring Report 2003, edited by A. R. Management, pp. 7.1-7.10, Florida Department of Environmenta l Protection, Tallahassee, Florida. Finlayson-Pitts, B. J., and J. N. Pitts (2000), Chemistry of the Upper and Lower Atmosphere: Theory, Experiments and Applications, 969 pp., Academic Press, San Diego, Cali. FIOH (1990), Industrial hygiene measurements, 1950-69 (data base), Finnish Institute of Occupational Health, Helsinki (Finland). Fox, D. L., and H. E. Je ffries (1979), Air-Pollution, Anal. Chem., 51, R22-R34. Fuller, E. C., and R. H. Crist (1941), The rate of oxidation of sulfite ions by oxygen, J. Amer. Chem. Soc., 63, 1644 1650. Hagmar, L., T. Bellander, C. Andersson, K. Li nden, R. Attewell, and T. Moller (1991), Cancer morbidity in nitrate fertilizer workers, Int. Arch. Occup. Environ. Health., 63, 63-67. Hayami, H. (2005), Behavior of secondary inorganic species in gaseous and aerosol phases measured in Fukue Island, Japan, in dust season, Atmos. Environ., 39, 2243-2248. Healy, J., S. D. Bradley, C. Northage, and E. Scobbie (2001), Inhalation exposure in secondary aluminium smelting, Ann. Occup. Hyg., 45, 217-225. Hinds, W. C. (1999), Aerosol Technology: Properties, Behav ior, and Measurement of Airborne Particles, 2nd ed., 483 pp., John Wiley & Sons, Inc., New York. Hodge, C. A., and N. N. Popovici (1994), Pollution Control in Fertilizer Production, Marcel Dekker, Inc., New York. Horton, K. D., and J. P. Mitchell (1989), Water droplet fragmentation in an Andersen Mark-III in-stack cascade impactor, J. Aerosol Sci., 20, 1585-1588. Howell, S., A. A. P. Pszenny, P. Quinn, and B. Huebert (1998), A field in tercomparison of three cascade impactors, Aerosol Sci. Technol., 29, 475-492. Hsu, Y. M., C. Y. Wu, D. A. Lundgren, and B. K. Birky (2007a), Chemi cal characteristics of aerosol mists in phosphate fertili zer manufacturing facilities, J. Occup. Environ. Hyg., 4, 17-25.

PAGE 149

149 Hsu, Y. M., C. Y. Wu, D. A. Lundgren, and B. K. Birky (2007b), Sizeresolved sulfuric acid mist concentrations at phosphate fertiliz er manufacturing facilities in Florida, Ann. Occup. Hyg., 51, 81-89. Hsu, Y. M., J. Kollett, K. Wysocki, C. Y. Wu, D. A. Lundgren, and B. K. Birky (2007c), Positive artifact sulfate formation from SO2 adsorption in the silica gel sampler used in NIOSH Method 7903, Environ. Sci. Technol., 41, 6205-6209. Huang, Z., R. M. Harrison, A. G. Allen, J. D. James, R. M. Tilling, and J. X. Yin (2004), Field intercomparison of filter pack and impactor sampling for aerosol nitrate, ammonium, and sulphate at coastal and inland sites, Atmos. Res., 71, 215-232. Huss, A., P. K. Lim, and C. A. Eckert (1978) Uncatalyzed oxidation of sulfur(IV) in aqueoussolutions, J. Amer. Chem. Soc., 100, 6252-6253. IARC (1992), IARC Monograph on the Evaluation of Carcinogenic Risks to Humans, Vol. 54, Occupational Exposures to Mists and Vapou rs from Strong Inorganic Acids; and Other Industrial Chemicals, 336 pp., International Agency for Research on Cancer (IARC). Lyon, France. ICRP (1994), Human Respiratory Tract Mode l for Radiological Protection, Elsevier Science, Tarrytown, NY. Jacobson, M. Z. (1999), Chemical equilibrium and dissolution processes, in Fundamentals of Atmospheric Modeling, edited, pp. 553-597, Cambridge University Press, New York. Jiang, C. F. (1996), Thermodynamics of aque ous phosphoric acid solution at 25 degrees C, Chem. Eng. Sci., 51, 689-693. Jonnalagadda, S. B., M. Nyagani, P. Sawunyama, and R. Baloyi (1991), St udies on the levels of sulfur-dioxide, nitrogen-dioxide, ammonia, and hydrogen-chloride in ambient air of Harare, Zimbabwe, Environ. Int., 17, 461-467. Kim, Y. P., J. H. Seinfeld, and P. Saxena (1993), Atmospheric gas-aerosol equilibrium.2. Analysis of common approximations and activity-coefficient calculation methods, Aerosol Sci. Technol., 19, 182-198. Kopac, T., and S. Kocabas (2002), Adsorption equ ilibrium and breakthrough analysis for sulfur dioxide adsorption on silica gel, Chem. Eng. Process., 41, 223-230. Koutrakis, P., C. Sioutas, S. T. Ferguson, J. M. Wolfson, J. D. Mulik, and R. M. Burton (1993), Development and evaluation of a glass honeycom b denuder filter pack system to collect atmospheric gases and particles, Environ. Sci. Technol., 27, 2497-2501.

PAGE 150

150 Larson, T. V., N. R. Horike, and H. Harrison ( 1978), Oxidation of sulf ur-dioxide by oxygen and ozone in aqueous-solution kine tic study with significance to atmospheric rate processes, Atmos. Environ., 12, 1597-1611. Lee, K. W., and R. Mukund (2001), Filter collection, in Aerosol Measurement: Principles, Techniques, and Applications, edited by P. A. Baron and K. Willeke, pp. 197-220, Wiley, New York. Lue, S. J., T. Wu, H. Hsu, and C. Huang ( 1998), Application of ion chromatography to the semiconductor industry I. Measurement of acidi c airborne contaminants in cleanrooms, J. Chromatogr. A, 804, 273-278. Lunsford, J. H. (1979), Surface Reactions of Oxid es of Sulfur, U.S. Environmental Protection Agency, Research Triangle Park, NC. Mann, H. C. (1992), Ammonium phosphates, in Air Pollution Engineering Manual, edited by A. J. Buonicore and W. T. Davis, pp. 574-576, Van Nostrand Reinhold, New York. Meng, Z. Q., and B. Zhang (1997), Chromosoma l aberrations and micronuclei in lymphocytes of workers at a phosphate fertilizer factory, Mutat. Res., 393, 283-288. Meng, Z. Q., and L. Z. Zhang (1990), Chro mosomal-aberrations and sister-chromatid exchanges in lymphocytes of work ers exposed to sulfur-dioxide, Mutat. Res., 241, 15-20. Meng, Z. Q., H. Q. Meng, and X. L. Cao (1995), Sister-chromatid exchanges in lymphocytes of workers at a phosphate fertilizer factory, Mutat. Res., 334, 243-246. Meng, Z. Y., and J. H. Seinfeld (1996), Time scales to achieve atmospheric gas-aerosol equilibrium for volatile species, Atmos. Environ., 30, 2889-2900. Messnaoui, B., and T. Bounahmidi (2005), Mode ling of excess properties and vapor liquid equilibrium of the system H3PO4 H2O, Fluid Phase Equilib., 237, 77-85. Miller, J. M., and R. G. D. Pena (1972), Contri bution of scavenged sulfur-dioxide to sulfate content of rain water, J. Geophys. Res., 77, 5905-5916. Muller, T. L. (1992), Sulfuric acid, in Air Pollution Engineering Manual, edited by A. J. Buonicore and W. T. Davis, pp. 469-476, Air & Waste Management Association, Pittsburgh, PA. Munger, J. W., C. Tiller, and M. R. Hoffmann (1986), Identification of hydroxymethanesulfonate in fog water, Science, 231, 247-249. NIOSH (1994), NIOSH Method No. 7903, in NIOSH Manual of Analytical Methods, edited by P.C. Schlecht and P.F. O'Connor, National Institute for Occupational Safety and Health (NIOSH), Cincinnati, OH.

PAGE 151

151 Ortiz, L. W., and C. I. Fairchild (1976), Aerosol Research and Development Related to Health Hazard Analysis, US Energy Research and Development Administration. Ostro, B. D., M. J. Lipsett, M. B. Wiener, a nd J. C. Selner (1991), Asthmatic responses to airborne acid aerosols, Am. J. Public Health, 81, 694-702. Palm, G. F. (1992), Phosphoric acid manufacturing, in Air Pollution Engineering Manual, edited by A. J. Buonicore, et al., pp. 439-446, Van Nostrand Reinhold, New York. Parish, W. R. (1994), Phosphoric acid by wet proce ss: recovery of fluorinecontaining gases, in Pollution control in fertilizer production, edited by C. A. Hodge and N. N. Popovici., pp. 197-208, Dekker, New York. Pathak, R. K., and C. K. Chan (2005), Inter-part icle and gas-particle in teractions in sampling artifacts of PM2.5 in filter-based samplers, Atmos. Environ., 39, 1597-1607. Pauluhn, J. (2005), Retrospective analysis of acu te inhalation toxicity studies: Comparison of actual concentrations by filter and cascade impactor analyses, Regul. Toxicol. Pharm., 42, 236-244. Petrov, G. A. (1987), Sanation of working-cond itions and environment for the process of sulfuric acid manufacture from metallurgical gases (Russ), Gig Sanit, 70-71. Possanzini, M., A. Febo, and A. Liberti (1983) New design of a high-performance denuder for the sampling of atmo spheric pollutants, Atmos. Environ., 17, 2605-2610. Radojevic, M. (1984), On the discrepancy between reported studies of the uncatalyzed aqueous oxidation of SO2 by O2, Environ. Technol. Lett., 5, 549-566. Ridler, E. S. (1959), Chamber process fo r the manufacture of sulfuric acid, in The Manufacture of Sulfuric Acid, edited by W. W. Dueker and J. R. West, Reinhold Publishing Corporation, New York. Sathiakumar, N., E. Delzell, Y. AmoatengAdjepong, R. Larson, and P. Cole (1997), Epidemiologic evidence on the relationship be tween mists containing sulfuric acid and respiratory tract cancer, Crit. Rev. Toxicol., 27, 233-251. Schlesinger, R. B. (1984), Compar ative irritant potency of inha led sulfate aerosols effects on bronchial mucociliary clearance, Environ. Res., 34, 268-279. Schlesinger, R. B. (1989), Factors affecting the response of lung clea rance systems to acid aerosols role of exposur e concentration, exposure time, and relative acidity, Environ. Health Perspect., 79, 121-126.

PAGE 152

152 Schlesinger, R. B., and L. C. Chen (1994), Co mparative biological pote ncy of acidic sulfate aerosols Implications for the interpre tation of laboratory and field studies, Environ. Res., 65, 69-85. Schlesinger, R. B., L. C. Chen, I. Finkelstein and J. T. Zelikoff (1990a), Comparative potency of inhaled acidic sulfates specia tion and the role of hydrogen ion, Environ. Res., 52, 210224. Schlesinger, R. B., A. F. Gunnison, and J. T. Zelikoff (1990b), Modulation of pulmonary eicosanoid metabolism following exposure to sulfuric acid, Fundam. Appl. Toxicol., 15, 151-162. Schroeter, L. C. (1963), Kinetics of air oxidation of sulfurous acid salts, J. Pharm. Sci., 52, 559563. Scott, W. D., and P. V. Hobbs (1967), Formation of sulfate in water droplets, J. Atmos. Sci., 24, 54-57. Seinfeld, J. H., and S. N. Pandis (1998), Thermodynamics of aerosols, in Atmospheric Chemistry and Physics from Ai r Pollution to Climate Change, pp. 491-544, John Wiley, New York. Sioutas, C., P. Y. Wang, S. T. Ferguson, P. K outrakis, and J. D. Mu lik (1996), Laboratory and field evaluation of an improved glass honeycomb denuder filter pack sampler, Atmos. Environ., 30, 885-895. Skytt, E. (1978), Tilasto tyhygieenisist mittauksista v. 1971-1976 [Statistics of industrial hygiene measurements in 1971-1976]. Institut e of Occupational Health, Helsinki (Finland). Stayner, L., A. B. Smith, G. Reeve, L. Blade, L. Elliott, R. Keenlyside, and W. Halperin (1985), Proportionate mortality study of workers in the Garment Industry exposed to formaldehyde, Am. J. Ind. Med., 7, 229-240. Steenland, K. (1997), Laryngeal cancer incidence among workers e xposed to acid mists (United States), Cancer Causes Control, 8, 34-38. Stephenson, F., M. Cassady, H. Donaldson, a nd T. Boyle (1977), Industrial Hygiene Survey, CF Chemicals, Inc., Bartow, FL, August 9-12, 1976., National Institu te for Occupational Safety and Health, Cincinnati, OH. Stephenson, F., M. Cassady, H. Donaldson, a nd T. Boyle (1977), Industrial Hygiene Survey, IMC, Phosphate Chemical Complex, New Wales, Florida, National Institute for Occupational Safety and Health, Cincinnati, OH.

PAGE 153

153 Stratmann, H., and M. Buck (1965), Comparat ive measurements using silica-gel and TCM procedures for determining-sulf ur-dioxide in the atmosphere, Ai.r Water. Pollut., 9, 199218. Swenberg, J. A., and R. O. Beauchamp ( 1997), A review of the chronic toxicity, carcinogenicity, and possible mechanisms of ac tion of inorganic acid mists in animals, Crit. Rev. Toxicol., 27, 253-259. Swietlicki, E., H. C. Hansson, B. Martinsson, B. Mentes, D. Orsini B. Svenningsson, A. Wiedensohler, M. Wendisch, S. Pahl, P. Winkler R. N. Colvile, R. Gieray, J. Luttke, J. Heintzenberg, J. N. Cape, K. J. Hargreaves, R. L. StoretonWest, K. Acker, W. Wieprecht, A. Berner, Kruisz, C., M. C. Facchini, P. Laj, S. Fuzzi, B. Jones, and P. Nason (1997), Source identification during the Gr eat Dun Fell Cloud Experiment 1993, Atmos. Environ., 31, 2441-2451. Tadzhibaeva, N. S., and I. V. Gol'eva (1976), Industrial hygiene and condition of the upper respiratory tract of people working in the production of superphospha te in Uzbekistan, Russia, Med. Zh. Uzb., 4, 57-59. Tanaka, S., K. Yamanaka, and Y. Hashimot o (1987), Measurement of concentration and oxidation rate of S(IV) in ra inwater in Yokohama, Japan, Am. Chem. Soc. Symp. Ser., 349, 158-169. Tsai, C. J., C. H. Huang, and S. H. Wang ( 2001), Collection efficiency and capacity of three samplers for acidic and basic gases, Environ. Sci. Technol., 35, 2572-2575. Tsai, C. J., C. H. Huang, and H. H. Lu ( 2004), Adsorption capacity of a nylon filter of filter pack system for HCl and HNO3 gases, Sep. Sci. Technol., 39, 629-643. USDHHS (2005), Report on Carcinogens, Eleventh ed., U.S. Department of Health and Human Services, Public Health Service, National Toxicology Program, Res earch Triangle Park, NC. USEPA (1995), Compilation of Air Pollutant Emissions Factors Volume 1: Stationary Point and Area Sources, Fifth Edition with Supplements, Document No. AP-42., U.S. Environmental Protection Agency, Research Triangle Park, NC. Wagner, J., and D. Leith (2001), Passive aeroso l sampler. Part II: Wind tunnel experiments, Aerosol Sci. Technol., 34, 193-201. Wallingford, K. M., and E. M. Snyder (2001), Occupational exposures during the World Trade Center disaster response, Toxicol. Ind. Health, 17, 247-253. Watson, J. G., and J. C. Chow (2001), Ambient air sampling, in Aerosol Measurement: Principles, Techniques and Applications, edited by P. A. Baron and K. Willeke, pp. 821844, Wiley, New York.

PAGE 154

154 Watson, J. G., G. Thurston, N. Frank, J. P. Lodge, R. W. Wiener, F. F. McElroy, M. T. Kleinman, P. K. Mueller, A. C. Schmidt, F. W. Lipfert, R. J. Thompson, P. K. Dasgupta, D. Marrack, R. A. Michaels, T. Moore, S. Pe nkala, I. Tombach, L. Vestman, T. Hauser, and J. C. Chow (1995), 1995 Critical-review discussion measurement methods to determine compliance with ambient air-qua lity standards for suspended particles, J. Air Waste Manage. Assoc., 45, 666-684. Weast, R. C. (1988), CRC Handbook of Chemistry and Physics, edited, CRC Press, Boca Raton, Fla. Wu, C. Y., and P. Biswas (1998) Particle growth by condensation in a system with limited vapor, Aerosol Sci. Technol., 28, 1-20. Yadav, J. S., and V. K. Kaushik (1996), Effect of sulphur di oxide exposure on human chromosomes, Mutat. Res.-Envir. Muta., 359, 25-29. Yao, X. H., T. Y. Ling, M. Fang, and C. K. Chan (2006), Comparison of thermodynamic predictions for in situ pH in PM2.5, Atmos. Environ., 40, 2835-2844. Yao, X. H., T. Y. Ling, M. Fang, and C. K. Chan (2007), Size dependence of in situ pH in submicron atmospheric particles in Hong Kong, Atmos. Environ., 41, 382-393. Yasyerli, S., G. Dogu, I. Ar, and T. Dogu (2001) Activities of copper oxide and Cu-V and CuMo mixed oxides for H2S removal in the presence and absence of hydrogen and predictions of a deactivation model, Ind. Eng. Chem. Res., 40, 5206-5214.

PAGE 155

155 BIOGRAPHICAL SKETCH Yu-Mei Hsu was born in 1975 in Hsin-Chu, Ta iwan. She received her B.S. degree in environmental engineering in June 1999 at Na tional Cheng-Kung University, Taiwan. She was awarded the Phi Tau Phi Scholastic Honor whic h awarded the first ra nking student among 53 students. She also earned her M.S. degree in en vironmental engineering sciences in June 2001 at National Taiwan University, Taiwan. Her masters thesis, the Chloride Loss of Sea-Salt Aerosols, was awarded the Outstanding Masters Thesis Award from National Taiwan University and National Science Council, Taiwan. She joined the research group of Dr. Chang-Yu Wu at the University of Florida in 2004 and started pursuing her Ph.D. degree in the Depart ment of Environmental Engineering Sciences. Her research focused on the sulfuric acid mist sampling at the phosphate fertilizer plants. Yu-Mei Hsu was the vice-president, secretary, and webmaster of the student chapter of Air & Waste Management Association (A&WMA) at the University of Florida from 2005 to 2007. She was awarded the Axel Hendrickson Scholarsh ip Award from Florida Section A&WMA in 2006, and also was awarded the AWMA Scholarship (2nd place in air quality) from A&WMA in 2007.