cVEPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EMB Report 78-NHF-8
May 1979
Air
Urea Manufacture
Emission Test Report
C. F. Industries
Donaldsonville, Louisiana
-------
REPORT ON "B" GRANULATOR AND
SYNTHESIS TOWER EMISSIONS
AT THE CF INDUSTRIES, INC.,
UREA FERTILIZER PLANT
IN DONALDSONVILLE, LOUISIANA
WILLARD A. WADE HI, P.E.
PROJECT MANAGER
REED W. CASS
PROJECT ENGINEER
TRC PROJECT 0988-E80-20
EPA CONTRACT #68-02-2820
WORK ASSIGNMENT #10
THOMAS M. BIBB
EPA PROJECT OFFICER
CLYDE E. RILEY
EPA TECHNICAL MANAGER
•TOiK
-------
TRC - ENVIRONMENTAL CONSULTANTS, INC.
Willard A. Wade, III, P. E.
Project Manager
March 1, 1980
-11-
-------
NOTE;
Mention of trade names or commercial products in this publication does not
constitute endorsement or recommendation for use by the Environmental Protection
Agency.
iii
-------
PREFACE
The work reported herein was conducted by personnel from TRC -
Environmental Consultants, Inc. (TRC), the GC A/Techno logy Division (GCA), CF
Industries, Inc. (CFI), Donaldsonville, Louisiana, and the U.S. Environmental
Protection Agency (EPA).
The scope of work issued under EPA Contract No. 68-02-2820, Work
Assignment No. 10 was under the supervision of the TRC Project Manager,
Mr. Willard A. Wade, III. Mr. Reed W. Cass of TRC served as Project Engineer and
was responsible for summarizing the test and analytical data in this report.
Analysis of the samples was performed at the CFI Donaldsonvlle, Louisiana plant
under the direction of Ms. Margaret Fox of TRC and at the TRC labs in
Wethersfield, Connecticut under the direction of Ms. Joanne 3. Marchese.
Mr. Stephen A. Capone and Mr. Timothy L. Curtin of GCA were responsible
for monitoring the process operations during the testing program. GCA personnel
were also responsible for writing the Process Description and Operation section
along with Appendix I of this report.
Members of CF Industries, Inc., Donaldsonville, Louisiana whose assistance
and guidance contributed greatly to the success of the test program include
Mr. Thomas Carville, Senior Environmental Engineer, Mr. Stephen Thompson, Area
Supervisor, and Mr. David Campo, Laboratory Supervisor.
Mr. Eric A. Noble, Office of Air Quality Planning and Standards, Industrial
Studies Branch, EPA, served as Test Process Project Engineer and was responsible
for coordinating the process operations monitoring.
Mr. Clyde E. Riley, Office of Air Quality Planning and Standards, Emission
Measurement Branch, EPA, served as Technical Manager and was responsible for
coordinating the emission test program. .
-IV-
-------
TABLE OF CONTENTS
SECTION PAGE
1. INTRODUCTION 1
2. SUMMARY OF RESULTS 12
3. PROCESS DESCRIPTION AND OPERATION 35
4. LOCATION OF SAMPLING POINTS 46
5. SAMPLING AND ANALYSIS METHODOLOGIES 55
APPENDICES
Appendix A
Computer Printout Test Results With Equations and Example
Calculations.
A-l Computer Printout Test Results
1. Inlet to "B" Granulator Scrubber (TP-1)
2. Outlet From "B" Granulator Scrubber (TP-2)
A-2 Sample Equations and Example Calculations
Appendix B
Particle Size Test Results
B-l Discussion of Particle Size Testing
B-2 Particle Size Field Data Sheets
B-3 Lab Weighting Data
Appendix C
Visible Emissions Results
C-l Visible Emissions Summary Tables for Outlet Stack "B"
Granulator
C-2 Visual Emission Recertification Certificate
C-3 Visible Emission Field Data Sheets
C-4 Guidelines for EPA Method 9
-v-
-------
APPENDICES (Continued)
Appendix D
Field Data Sheets for "B" Granulator Testing
D-l Scrubber Inlet Location (TP-1)
D-2 Scrubber Outlet Location (TP-2)
Appendix E
Field Data for Miscellaneous Samples and Operation
E-l Scrubber Water Data (Inlet-Outlet)
E-2 Scrubber Pressure Drops
E-3 Process Samples Analysis Data
E-4 Relative Humidity and Ambient Air Temperature Measurements
Appendix F
Synthesis Tower Sampling Data
F-l Draft Method for Testing Ammonium Nitrate Facilities
F-2 Field Data Sheets for Synthesis Tower Sampling
F-3 Synthesis Tower Test Results Calculations
F-
-------
APPENDICES (Continued)
Appendix K
Sampiing and Analytical Procedures
K-l EPA Procedures for Urea
K-2 EPA Procedures for Ammonia
K-3 EPA Draft Method for Formaldehyde
K-*f Discussion of Analytical Procedures
K-5 Moisture Determination Procedure
Appendix L
Cleanup Evaluation Results
Appendix M
Project Participants
Appendix N
Work Assignment
N-l Copy of Project Work Assignment
N-2 Copy of Project Technical Directives
-Vli-
-------
UST OF TABLES
TABLE PAGE
1-1 Daily Summary Log for "B" Granulator Sampling
on January 17, 1979 at CF Industries, Inc.,
Donaldsonville, Louisiana 6
1-2 Daily Summary Log for "B" Granulator Sampling
on January 18, 1979 at CF Industries, Inc.,
Donaldsonville, Louisiana
1-3 Daily Summary Log for "B" Granulator Sampling
on January 19, 1979 at CF Industries, Inc.,
Donaldsonville, Louisiana ............... 11
2-1 Overall Summary of Results of Urea, Ammonia and
and Formaldehyde Tests on January 17 and 18, 1979 of
Gases Entering and Exiting the "B" Granulator Scrubber at
CF Industries, Inc., Donaldsonville, Louisiana ....... 14
2-2 Summary of Results of Urea, Ammonia and Formaldehyde
Tests on January 17 and 18, 1979 of Gases Entering the "B"
Granulator Scrubber at CF Industries, Inc., Donaldsonville,
Louisiana ..................... . 16
2-3 Summary of Results of Urea, Ammonia, and Formaldehyde
Tests on January 17 and 18, 1979 of Gases Exiting the
"B" Granulator Scrubber at CF Industries, Inc.,
Donaldsonville, Louisiana .............. . 17
2-4 Six-Minute Arithmetic Average Opacity Readings on
"B" Granulator Scrubber Stack on January 17, 18, and 19, 1979,
at CF Industries, Inc., Donaidsonville, Louisiana ...... 22
2-5 Summary of Inlet Particle Size Test Results on "B"
Granulator Scrubber on January 18, 1979 at
CF Industries, Inc., Donaldsonviile,
Louisiana ...................... 25
2-6 Summary of "B" Granulator Scrubber Pressure Drop
Measurements on January 17 and 18, 1979 at
CF Industries, Inc., Donaldsonville,
Louisiana ...................... 26
2-7 Summary of Urea, Ammonia, and Formaldehyde
Measurements on January 17 and 18, 1979 on the
Scrubbing Liquid Entering and Exiting "B" Granulator
Scrubber at CF Industries, Inc., Donaldsonville, Louisiana. . 28
2-8 Summary of pH and Temperature Measurements on
January 17 and 18, 1979 on Individual Samples of
Scrubbing Liquid Entering and Exiting "B" Granulator
Scrubber at CF Industries,Inc., Donaidsonville, Louisiana . . 29
-vm-
-------
LIST OF TABLES (continued)
TABLE PAGE
2-9 Summary of Urea, Ammonia and Formaldehyde
Measurements on January 17 and 18, 1979 on the
"B" Granulator Melt and Product before Screening at
CF Industries, Inc., Donaldsonville, Louisiana 31
2-10 Summary of Sieve and Bulk Density Measurements
on the "B" Granulator Product Before Screening at
CF Industries, Inc., Donaldsonville, Louisiana 32
2-11 Summary of Ambient Temperature, Relative Humidity and
Barometric Pressure Measurements on January 17 and 18, 1979
for "B" Granulator at CF Industries, Inc., Donaldsonville,
Louisiana 33
2-12 Summary of Results of Urea and Ammonia Tests on
January 18 and 19, 1979 of Gases Exiting the Urea
Synthesis Tower Vent at CF Industries, Inc.,
Donaldsonville, Louisiana 34
3-1 Granulator B Feed Rate During Granulation Emissions
Testing 40
3-2. Solution Formation Rate For Production Line 1
During Solution Formation Emissions Testing 40
3-3 Relative Average Values of Operating Parameters
During Testing 42
3-4 Scrubber Operating Parameters During Scrubber
Air Inlet and Outlet Particulate Loading Tests 44
-IX-
-------
UST OF FIGURES
FIGURE PAGE
1-1 Overhead View Showing Locations of Synthesis
Tower Vent and Granulators A, B, and C
Exhaust Ducting, Scrubbers and Sampling
Points at CF Industries, Inc., Donaldsonville,
Louisiana .................... 3
2-1 Six Minute Arithmetic Averages of
January 17-19, 1979 Opacity Readings on "B"
Granulator Scrubber Stack at CF Industries, Inc.,-
Donaldsonville, Louisiana ............. 21
3-1 Typical Urea Manufacturing Process As Designed
by C & F Girdler ................. 36
4-1 Locations of "B" Granulator Scrubber Inlet Test
Ports and Points at CF Industries, Inc., Donaldsonville,
Louisiana .................... 47
4-2 . Locations of "B" Granulator Scrubber Outlet Test
Ports and Points at CF Industries, Inc., Donaldsonville,
Louisiana .................... 49
4-3 Location of Smoke Observer for
January 17-19, 1979 Opacity Readings on "B"
Granulator Scrubber Stack at C F Industries, Inc.,
Donaldsonville, Louisiana ............. 51
4-4 Locations of Scrubber Liquid Collection Taps for
January 17-19, 1979 Tests on "B" Granulator at
CF Industries, Inc., Donaldsonville, Louisiana ..... 52
4-5 Location of Synthesis Tower Solution Vent Sampling
Port at CF Industries, Inc., Donaldsonville, Louisiana. . 54
5-1 Modified EPA Particulate Sampling Train,
August 18, 1977. Federal Register .......... 57
-x-
-------
SECTION 1
INTRODUCTION
C
Section III of the Clean Air Act of 1970 charges the Administrator of the
U.S. Environmental Protection Agency (EPA) with the responsibility of establishing
Federal standards of performance for new stationary sources which may
significantly contribute to air pollution. When promulgated, these standards of
performance for new stationary sources (SPNSS) are to reflect the degree of
emission limitation achievable through application of the best demonstrated
emission control technology. To assemble this background information, EPA
utilizes emission data obtained from controlled sources in the particular industry
under considerationr
(
EPA's Office of Air Quality Planning and Standards (OAQPS) selected the
CF Industries, Inc. urea manufacturing plant at Donaldsonville, Louisiana as a site
for an emission test program. The test program was designed to provide a portion
of the emission data base required for SPNSS for the processes associated with the
production of urea.
The CF Industries, Inc. urea manufacturing plant at Donaldsonville, Louisiana
produces granulated urea for fertilizer use. The urea is made by seven
Spherodizers* (granulators) which operate continuously, 24 hours a day and 7 days a
week, as production demands.
-1-
-------
Each granulator has its own impingement type water scrubber. The
granulator exhaust is ducted through the scrubber fan and is discharged from a
stack. Flow through the granulator to the constant flow scrubber is controlled with
a dilution damper which varies the ratio of dilution air to exhaust gas. A
schematic of the granulator and control system is shown in Figure 1-1.
Emissions sampling was conducted on the "B" granulator of the three older
units while its urea production rate was approximately 333 ton/day. Emission
sampling was also conducted on the main exhaust vent atop the urea synthesis
tower (see Figure 1-1). This vent combines the various solution forming process
gases into one common stack before exhausting them to the atmosphere.
EPA engaged TRC to measure urea, ammonia and formaldehyde concentra-
tions and mass flow rates; particle size distributions; and plume opacities. All
measurements made at this facility were performed during times of normal
operation of the urea production process as described in Section 3, "Process
Description and Operation."
The measurement program was conducted at the CF Industries, Inc. urea
manufacturing facility in Donaldsonville, Louisiana during the week of January 15
through January 19, 1979 and consisted of the following:
"B" Granulator Scrubber Measurements
1. Urea, Formaldehyde, and Ammonia in Gas Stream
Three runs of concurrent inlet and outlet tests were performed. The
tests were conducted in accordance with the prescribed EPA methods
for urea, ammonia, and formaldehyde and provided velocity, moisture,
ammonia, formaldehyde and urea particulate mass flow rate data.
2. Particle Size Distributions in Gas Streams
Three runs of an inlet test were conducted. The tests were performed
using the procedures provided by the cascade impactor manufacturer.
-2-
-------
FAN
VjJ
SAMPLING
PORT
SYNTHESIS
TOWER
VENT
TOWER FRAME
STACK
WET
SCRUBBER
OUTLET
SAMPLING
PORTS
WET
SCRUBBER
STACK
INLET
SAMPLING
PORTS
WET
SCRUBBER
STACK
DILUTION AIR
IIA II
GRANULATOR "A
WALL
•DILUTION AIR
GRANULATOR "n"
DILUTION AIR
-I GRANULATOR "r"
FIGURE. 1-1:
OVERHEAD VIEW SHOWING LOCATIONS OF SYNTHESIS TOWER VENT &
GRANULTORS A, B & C EXHAUST DUCTING, SCRUBBERS & SAMPLING .
AT CF INDUSTRIES, INC., . DONALDSONVILLE, LOUISIANA
-------
3. Visible Emissions from Scrubber Stack
Approximately six hours of visible emissions observations were recorded
on the scrubber stack discharge. Observations were performed in
accordance with EPA Method 9 guidelines.
4. Gas Pressure Drop Across Scrubber
Pressure drop measurements were recorded approximately every five
minutes during the testing periods.
5. Ambient Air Temperature,Pressure and Relative Humidity
Wet and dry bulb ambient air temperatures and the barometric pressure
were recorded at frequent intervals during the test days.
6. Scrubber Liquid Sampling
Samples of inlet and outlet scrubber water were taken during the urea,
ammonia and formaldehyde testing. The pH and temperature of each
sample were recorded. The samples were composited into three inlet
and three outlet samples which were analyzed for urea, ammonia, and
formaldehyde, and percent solids.
7. Urea, Ammonia, and Formaldehyde Contents of Product
A single grab sample of the urea melt was taken and analyzed for urea,
ammonia, and formaldehyde. In addition, three grab samples of the
unscreened granulator product were analyzed for sieve, bulk density,
and percent moisture. These samples were then combined for a urea
and formaldehyde determination. The percent moisture determinations
were performed by CFI personnel.
Urea Synthesis Tower Vent Measurements
Urea and Ammonia in Gas Stream
Three test runs were performed according to procedures prescribed in
EPA methods for velocity, urea, and ammonia and for sampling gas
streams with high moisture content.
TRC personnel were responsible for collecting and measuring the above
emission parameters. Concurrently, GCA was responsible for monitoring and
recording necessary process parameters. A copy of TRC's Work Assignment and
Technical Directives is included in Appendix N.
-------
The sequence of events for this sampling program is shown in Tables 1-1
through 1-3 (Daily Summary Logs).
Several of the test runs were discontinuous due to the excessive particulate
loading at the granulator scrubber inlet sampling location. These interruptions,
which also delayed the concurrent outlet sampling, occurred throughout, the test
program.
The following sections of this report cover the summary of results
(Section 2), process description and operation (Section 3), location of sampling
points (Section 4), and sampling and analysis methodologies (Section 5). In addition,
Appendix L contains the results of the cleanup evaluations performed on the
sample collectors used during the test program. Detailed descriptions of methods
and procedures, field and laboratory data, and calculations are presented in various
appendices, as noted in the Table of Contents.
-5-
-------
TABLE I-1
DAILY SUMMARY LOG FOR "B" GRANULATOR SAMPLING ON 3ANUARY 17, 1979
AT CF INDUSTRIES INC., DONALDSONVILLE, LOUISIANA
Scrubber Liquid
Production Rate "B" Granulator Scrubber Stack Inlet Outlet Pressure
Clock tons/hour Urea Participate Vis
Time Granulator Syn. Tower Inlet Outlet Emiss
ible Temp Temp Drop
ions pH (°F) pH (°F) ("H2O)
1200 Confidential Start Run 1
1201 Per CF1 I Start
1202 Request 1
1207 f
1210
1212
1215
1217
1222
1223
1226
1227
1230
i 1232
T 1237
1242
1245
1247
1252
1255
1257
1301
1306
1310
1316
1318
1320
1321
1326
1327
1331
1333
1336
1341
1343
1346
1351
1356
Pitot Plugged
Run Continued
Probe Plugged
Run Continued
Run 1
Start
21.8
r
Complete 1st
• , Traverse
1
Complete 1st
Traverse
Start 2nd
Traverse
Start 2nd
Traverse
Nozzle Plugged
Run Continued
1 '
i
21.6
9.50 135 8.15 126
20.9
20.7
20.7
20.7
21.1
21.1
20.7
9.25 154 8.10 124
20.5
20.5
20.7
9.15 164 8.10 124 20.9
20.9
20.7
20.7
20.7
20.9
Ambient Conditions
Temp R.H. Pressure Process
(°F) (%) ("HR) Sampling
69 76
69 76
65 76 30.40
69 76 30.40
Unscreened
Product
69 79 30.39
-------
TABLE 1-1 (Continued)
DAILY SUMMARY LOG FOR "B" GRANULATOR SAMPLING ON 3ANUARY 17, 1979
AT CF INDUSTRIES INC..-DONAJLDSONV1LLE, LOUISIANA
Production Rate "B" Granulator Scrubber 5ta
Clock tons/hour Urea Particulate Vis
Time Granulator Syn. Tower Inlet
1401 Confid
1406 Per
1411 Reqi
1416
1417
1422
1423
1448
1523
1548
1549
1550
1554
1559
1604
1609
1614
1619
1620
1624
1629
1634
1635
1637
1639
1644
1649
1653
1655
1657
1658
ential
CFI
jest
^
i
Complete Run I
Start Run 2
i
•
F
Complete 1st
Traverse
Start 2nd
Traverse
Probe Plugged
r
Outlet Em is
Complete Run 1
1
Scrubber Liquid
ick Inlet Outlet Pressure Ambient Conditions
iible Temp Temp Drop Temp R.H. Pressure Process
sions pH ("F) pH (°F) ("H2O) (°F) (%) ("Hg) Sampling
20.7
20.5
20.5
9.10 152 7.95 120 20.5
r '
Stop
70 74 30.39
69 79 30.39
Start Run 2 . 20.2
1
St
^
rt 9.7 166 8.3 129
20.2
20.2
20.8
20.9
20.9
20.7
9.5 183 8.2 128
20.8
20.9
. 21.0
67 85 30.39
Stop
21.0
. 20.2
Complete 1st '
Traverse
Start 2nd
Traverse
9.2 174 8.2 129
20.4
-------
TABLE 1-1 (Continued)
DAILY SUMMARY LOG FOR "B" GRANULATOR SAMPLING ON JANUARY 17, 1979
AT CF INDUSTRIES INC., DONALDSONV1LLE, LOUISIANA
oo
Scrubber Liquid
Production Rate "B" Granula tor Scrubber Stack Inlet Outlet Pressure Ambient Conditions
Clock tons/hour Urea Particulate
Visible . Temp Temp Drop Temp R.H. Pressure
Time Granula tor Syn. Tower Inlet Outlet Emissions pH (°F) pH (°F) ("H2O) (°F) (%) ("He)
1702 Confidential Run Continued
1703 Per CF1
1708 Request
1713
1715
1718
1723
1725
1728
1730
1733
1738
1743
1748
1753
1756
1800
1815
r
20.4
20.4
20.2
68 80 30.39
20.3
20.4
9.2 167 8.1 128
20.6
20.6
20.6
. 20.4
20.4
Complete Run 2 20.2
Complete Run 2
9.0 172 8.0 129
66 85 30.38
Process
Sampling
Unscreenec
Product
-------
TABLE 1-2
DAILY SUMMARY LOG FOR "B" GRANULATOR SAMPLING ON JANUARY 18, 1979
AT CF INDUSTRIES INC., DONALD5ONVILLE, LOUISIANA
Production Rate "B" Granulator Scrut
Clock tons/hour Urea Participate
Scrubber Liquid
>ber Stack Inlet Outlet Pressure
Visible Temp Temp Drop
Time Granulator Syn. Tower Inlet Outlet Emissions pH (°F) pH (°P) ("H2O)
0945 Confidential
0946 Per CFI Start Run 3 21.6
0947 Request Start Run 3
0951
0955
0956
1001
1005
1006
1011
1016
1017
1020
1021
1026
1030
1031
1036
1040
1041
1045
1046
1047
1055
1102
1105
1107
1108
1112
1115
1117
1122
.
Pitot Plugged
Continue - Probe
Plugged
Continue
. ' 21.2
9.2 180 8.3 128
21.0
21.1
21.0
21.2
21.0
20.7
20.5
9.3 172 8.25 126
20.7
20.6
20.7
i
, . Complete 1st
Traverse
Probe Plugged
Continue
Complete 1st
Traverse
9.3 177 8.25 129
Start 2nd
Traverse
Probe Plugged
Continue
(Start 2nd
Traverse
j
Ambient Conditions Synthesis
Temp R.H. Pressure Process Tower
(°F) (%) ("Hg) Sampling Sampling
62 94 30.39
62 94 30.39
64 90 30.39
64.5 87 30.38
Unscreened
Product
66 82 30.38
67.5 85 30.36
-------
TABLE 1-2 (Continued)
DAILY SUMMARY LOG FOR "B" GRANULATOR SAMPLING ON JANUARY 18, 1979
AT CF INDUSTRIES INC., DONALDSONVILLE, LOUISIANA
Production Rate "B" Granulator Scrub
Clock tons/hour Urea Particulate
Scrubber Liquid
ber Stack Inlet Outlet Pressure Ambient Conditions Synthesis
Visible Temp Temp Drop Temp R.H. Pressure Process Tower
Time Granulator Syn. Tower Inlet Outlet Emissions pH (°F) pH (°F) ("H2O) (*F) (%) ("Hg) Sampling Sampling
1127 Confidential
1132 PerCFI
1137 Request
1140
1142
1147
1152
1157
1200
1202
1207
1210
1212
1215
1217
1244
1344
1435
1500
1525
1528
1540
1543
1555
1615
1649
•
Complete Run 3
20.5
20.4
20.6
9.3 173 8.20 127 67 83 30.36
20.3
20.3
20.2
20.1
68 85 30.36
19.8
20.0
19.7
9.2 173 8.2 124
Complete Run 3 19.5
Start
Stop
69 85 30.28
71 77 30.28
73 73 30.27
Start Cascade
Run 1
.- . 71 77 30.27 Start Run 1
(Test 4)
Complete Cascade
Run 1
Complete Run!
(Teit 4)
Start Moisture
Run
Complete Moisture
1 Run
-------
TABLE 1-3
DAILY SUMMARY LOG FOR "B" GRANULATOR SAMPLING ON JANUARY 19, 1979
AT CF INDUSTRIES INC., DONALDSONVILLE, LOUISIANA
Scrubber Liquid
Production Rate "B" Granulator Scrubber Stack Inlet Outlet Pressure Ambient
i
i '
t—
i
Clock tons/hour Urea Paniculate Vis
Time Granulator Syn. Tower Inlet Outlet Emis
ible Temp Temp Drop Temp
sions pH <°F) pH (°F) ("HZO) (°F)
0953 Confidential Start
1000 Per CFI Start Cascade
Request Run 2
1015
1020
1030
1040
1046
1100
1120
1125
1135
1140
1150
1153
1200
1230
Complete Cascade
Run 2
Start Cascade
Run 3
Complete Cascade
Run 3 .
68
68
70
70
70
*
75
f
Stop
77
73
R.H
(%)
90
90
86
86
86
70
66
75
Conditions
Pressure
("HK)
30.17
30.17
30.17
30.17
30.14
30.12
30.10
30.10
Synthesis
Process Tower
Sampling Sampling
Start Run 2
(Test 5)
Complete Run2
Start Run 3
(Test 6)
Complete Run 3
-------
SECTION 2
SUMMARY OF RESULTS
INTRODUCTION
This section presents the results of a testing program conducted during the
week of January 15 through 19, 1979 at the CF Industries, Inc., urea manufacturing
facility in Donaldsonville, Louisiana. Testing was performed on gas and water
streams entering and exiting the "B" Granulator Scrubber and on the gas stream
venting from the Urea Synthesis Tower.
The inlet gas sampling location for the "B" Granulator Scrubber (designated
TP-1) was in a 23-foot straight section of horizontal duct located upstream of the
scrubber. The integrated gas samples for the urea, ammonia and formaldehyde
tests were collected isokinetically from 36 traverse points located in accordance
with EPA Reference Method.1 The gas samples for the particle sizing tests were
collected from a single point located at the centroid of the duct's cross sectional
area.
The outlet gas sampling locations for the "B" Granulator Scrubber (designated
TP-2) was in the 80-foot vertical stack located downstream of the scrubber. The
integrated gas samples for the urea, ammonia and formaldehyde tests were
Standards of Performance for New Stationary Sources, Appendix A. Federal
Register, Vol. 42, No. 160-Thursday, August 18, 1977, pp. 41756-41758.
-12-
-------
collected isokinetically from 12 traverse points located in accordance with EPA
Reference Method 1.
The gas sampling location for the Synthesis Tower Vent (designated TP-1)
was in the 46-foot straight section of vertical duct. The samples were extracted
from a single point located at the centroid of the duct's cross sectional area.
/
UREA. FORMALDEHYDE AND AMMONIA TESTS ON "B" GRANULATOR
SCRUBBER
I
The overall summaries of the urea, formaldehyde and ammonia results at the
"B" Granuiator inlet (TP-1) and outlet (TP-2) are contained in Tables 2-l(a) and
2-l(b) in English and SI metric units, respectively. The calculated urea,
formaldehyde and ammonia removal efficiencies of the "B" Granuiator Scrubber
averaged 99.8%, 50.2% and 20.5%, respectively.
Tables 2-2 and 2-3 present seperately the inlet and outlet data, respectively.
In addition, these two tables show the "corrected" urea data (corrected for possible
urea loss during the Kjeldahl analysis method), and the "corrected" and
"uncorrected" ammonia data resulting from the distillation-and-Nesslerization
analysis method. The ammonia correction is for potential conversion of urea to
ammonia during distillation. This second ammonia analysis method was added
because of the potential susceptibility of the direct Nesslerization method to
interferences.2 Details of the analysis methods are contained in Section 5 and
Appendix K. Because of the uncertainties involved in correcting the urea and
ammonia data for urea conversion during analysis, the data in Table 2-1
2Standard Methods for the Analysis of Water and Wastewater, 14th Edition, APHA,
AWWA, WPCF, 1975.
-13-
-------
TABLE 2-l(a) ENGLISH
SUMMARY OF RESULTS OF UREA, AMMONIA AND FORMALDEHYDE TESTS ON GASES ENTERING AND LEAVING THE
"B" GRANUALTOR SCRUBBER ON 3ANUARY 17 AND 18, 1979 AT CF INDUSTRIES INC., DONALDSONVILLE, LOUISIANA
Run Number
Date
Location
Volume of Gas / »
Sampled - DSCFw;
Percent Moisture
by Volume1 '
Average Gas
Temperature (°F)
Stack Volumetric , *
Flowrate - DSCFMlc;
Scrubber Liquid pH
Average Percent Opacity
Production Rate* (tons/hr)
Percent Iso kinetic
UKEA DATA
Kjeldahl Analysis
Method
Total Sample Weight (mg)
gr/DSCF"57
Ib/hr
Ib/ton
collection efficiency (%)
AMMONIA DATA
Direct Nesslerization
Analysis Method
Total Sample Weight (mg)
gr/DSCF
Ib/hr
Ib/ton
collection efficiency (%)
FORMALDEHYDE DATA
Chromotropic Acid
Analysis Method
Total Weight of Sample (mg)
gr/DSCF
Ib/hr
Ib/ton
collection efficiency (%)
Run 1
01-17-79
Inlet Outlet
63.18
2.257
192.1
40180
9.25
104.5
23920
5.830
2007
111.5
410
0.0999
34.40
1.90
2.476
0.000604
0.2078
0.0115
102.5
5.522
100.5
46260
8.08
7.7
100.7
70.6
0.0106
4.205
0.232
99.8
450
0.0676
26.80
1.48
22.1
2.523
0.000379
0.1503
0.0083
27.7
Run 2
01-17-79
Inlet Outlet
65.97
2.253
190.2
41410
9.32
105.0
20690
6.090
2161
120.1
441
0.1029
36.52
2.03
1.545
0.000361
0.1280
0.0071
102.7
5.733
100.7
45590
8.16
8.7
102.4
73.1
0.0110
4.283
0.238
99.8
694
0.1041
40.17
2.26
<0
1.172
0.000176
0.0687
0.0038
46.3
Run 3
01-18-79
Inlet Outlet
66.45
2.257
183.1
41760
9.26
102.9
26258
6.085
2178
125.2
514
0.1191
42.62
2.45
3.087
0.000715
0.2560
0.0147
102.4
5.608
102.5
45760
8.24
5.5
101.6
62.3
0.0094
3.674
0.211
99.8
387
0.0582
22.82
1.31
46.5
1.296
0.000195
0.0764
0.0044
70.2
AveraRe
Inlet
65.20
2.256
188.5
41117
9.28
104.1
25423
6.005
2116
118.9
455
0.1073
, 37.85
2.13
2.369
0.000560
0.1972
0.0111
Outlet
102.5
5.621
101.2
45870
8.16
7.3
101.6
68.7
0.0103
4.058
0.228
99.8
510
0.0766
30.10
1.69
20.5
1.664
0.000250
0.0983
0.0055
50.2
Dry standard cubic feet (d 68°F, 29.92 inches Hg
(a)
; jlnlet values based on seperate moisture run
J j\Dry standard cubic feet per minute
; ;Uncorreeled for urea loss during distillation
Grains per dry standard cubic foot
* Production rate information is confidential per CF1 request
-------
TABLE 2-Kb) METRIC
SUMMARY OF RESULTS OF UREA, AMMONIA AND FORMALDEHYDE TESTS ON GASES ENTERING AND LEAVING THE "B" GRANULATOR SCRUBBER
ON 3ANUARY 17 AND 18, 1979 AT CF INDUSTRIES INC., DONALDSONVILLE, LOUISIANA
itun Number
Jate
Location
Volume of Gas .
Sarnpled-Nm"a'
Percent Moisture
by Volume1 '
Average Gas
Temperature (°C)
Stack Volumetric. .
Flowrate-Nm'M1 '
Scrubber Liquid pH
Average Percent
Opacity
Production Rate*
(Mg/Hour)
Percent Isokinetic
UREA DATA
Kjeldahl Analysis
Total Sample
WeighUmg)
g/Nm*(e)
-------
TABLE 2-2
SUMMAKY OF RESULTS OF UREA, AMMONIA AND FORMALDEHYDE TESTS ON GASES ENTERING THE
"B" GRANULATOR SCRUBBER ON JANUARY 17 AND 18, 1979 AT CF INDUSTRIES INC., DONALDSONVILLE, LOUISIANA
Kun Number
Date
Volume of Gas .
Sampled-DSCFW'
Percent Moisture
by Volume* '
Average Gas
Temperature (°Fi
Stack Volumetric ( ,
Flowrate-DSCFMvc'
Scrubber Liquid pH
Production Rate*
(Tons/hr)
Percent Isokinetic
Net Sampling
Time (Minutes)
UkEA DATA
Kjeldahl Analysis Method:
Total Sample Weight (mg)
gr/USCF^
ib/hr
Ib/ton
AMMONIA DATA
Ness lerizat ion Analysis
Method:
Total Sample Weight (mg)
gr/DSCF
Ib/hr
Ib/ton
FORMALDEHYDE DATA
Chromotropic Acid
Analysis Method:
Total Sample Weight (mg)
gr/DSCF
ib/hr
Ib/ton
Run 1 Run 2 • Run 3 Average
01-17-79 01-17-79 01-18-79
63.18
2.257
192.1
40180
9.25
104.5
111
e f
Uncorrected Corrected
23920 25594
5.830 6.238
2007 2147
111.5 119.3
g h i
Uncorrected Corrected
Direct Distilled Distilled
410 888 -127
0.0999 0.2164
34.40 74.50
1.90 4.12
2.476
0.000604
0.2078
0.0115
65.97
2.253
190.2
41410
9.32
1
105.0
110
Uncorrected Corrected
26090 27916
6.090 6.516
2161 2312
120.1 128.5
Uncorrected Corrected
Direct Distilled Distilled
>41 941 -166
0.1029 0.2196
56.52 77.93
2.03 4.33
1.545
0.000361
0.1280
0.0071
66.45
2.257
183.1
41760
9.26
102.9
111
Uncorrected Corrected
26258 28096
6.085 6.511
2178 2330
125.2 134.0
Uncorrected Corrected
Direct Distilled Distilled
)14 944 -170
).1191 0.2187
»2.62 78.27
2.45 4.50
3.087
0.000715
0.2560
0.0147
65.2
2.256
188.5
41117
9.28
104.1
111
Uncorrected Corrected
25423 27203
6.005 6.425
2116 2264
118.9 127.2
Uncorrected Corrected
Direct Distilled Distilled
(55 924 -155
0.1073 0.2179
)7.85 76.86
2.13 4.33
2.369
0.000560
0.1972
0.0111
j^Dry Standard Cubic Feet id 68°F, 29.92 inches Hg.
jbased on separate moisture run
r::Dry Standard Cubic Feet per minute
['"Grains per DSCF.
; 'Uncorrected for urea conversion during distillation.
. .Corrected for urea conversion during distillation. Corrected = Uncorrected x 1.07.
JpfDirect Nesslerization.
U)
Distillation and Nesslerization, Uncorrected lor conversion of urea to ammonia.
ft Distillation and Nesslerization, corrected for conversion of urea to ammonia. Corrected = Uncorrected -0.07 x corrected urea/1.765
Confidential
-------
TABLE 2-3
SUMMARY OF RESULTS OF UREA, AMMONIA AND FORMALDEHYDE TESTS ON GASES LEAVING THE
"B" GRANULATOR SCRUBBER ON JANUARY 17 and 18, 1979 AT CF INDUSTRIES INC., DONALDSONV1LLE, LOUISIANA
Run Number
Date
Volume of Gas , .
Sampled - DSCFU;
Percent Moisture
by Volume
Average Gas
Temperature (°F)
Stack Volumetric ,. .
Flowrate - DSCFMVD'
Scrubber Liquid pH
Production Rate*
(tons/hr)
Percent Isokinetic
Pressure Drop Across
Scrubber (inches H2O)
Met Sampling Time
.minutes)
UREA DATA
Kjeldahi Analysis
Method:
Total Sample
Weight Irag)
gr/DSCFvc'
ib/hr
Ib/ton
AMMONIA
DATA
Messier izat ion
Analysis
Method:
Total Sample
Weight (mg)
gr/DSCF
Ib/hr
Ib/ton
FORMALDEHYDE
OATA
Chromotropic Acid
Analysis Method
Total Sample Weight (mg)
gr/DSCF
Ib/hr
Ib/ton
Run 1
01-17-79
102.5
5.522
100.5
46260
8.08
100.7
20.8
120
d e
uncorrected corrected
70.6 75.5
0.0106 0.0113
4.205 4.499
0.232 0.248
i g h
uncorrected corrected
direct distilled distilled
450 438 435
0.0676 0.0658 0.0653
26.80 26.08 25.91
1.48 1.44 1.43
2.523
0.000379
0.1503
0.0083
Run 2
01-17-79
102.7
5.733
100.7
45590
8.16
102.4
20.5
120
uncorrected corrected
73.1 78.2
0.0110 0.0118
4.283 4.583
0.238 0.255
uncorrected corrected
direct distilled distilled
694 426 423
0.1041 0.0639 0.0635
40.17 24.96 24.79
2.26 1.39 1.38
1.172
0.000176
0.0687
0.0038
Run 3
01-17-79
102.4
5.608
102.5
45760
8.24
101.6
20.6
120
uncorrected corrected
62.3 66.7
0.0094 0.0101
3.674 3.931
0.211 0.226
uncorrected corrected
direct distilled distilled
387 388 385
0.0582 0.0584 0.0579
22.82 22.88 22.70
1.31 1.31 1.30
1 1
1.296
0.000195
0.0764
0.0044
Average
102.5
5.621
101.2
45870
8.16
101.6
20.6
120
uncorrected corrected
68.7 73.5
0.0103 0.0110
4.058 4.342
0.228 0.244
uncorrected corrected
direct distilled distilled
510 417 414
0.0766 0.0626 0.0622
30.10 24.61 24.43
1.68 1.37 1.36
1.664
0.000250
0.0983
0.0055
/
'
Standard Cubic Feet (d 68°F, 29.92 inches Hg
Dry Standard Cubic Feet per minute
^'grains per DSCF
| 'Uncorrected for urea conversion during distillation
^.Corrected for urea conversion during distillation. Corrected = uncorrected x 1.07
j 'Direct Nesslerization
^'Distillation and Nesslerizstion, uncorrected for conversion of urea to ammonia
^Distillation and Nesslerization, corrected for conversion of urea to ammonia (corrected = uncoirrected - 0.07 x corrected urea/i.765>
* Confidential Per CFI Request
-------
(uncorrected urea and direct Nesslerization ammonia) should be viewed as the most
accurate data, with the following qualification.
The direct Nesslerization outlet ammonia mass flow rate appears to be very
high in Run 2 compared to those of Runs 1 and 3 (Table 2-3). The following
observations clarify this point:
1. Runs 1 and 3 each have mass flow rates which are nearly equal for both
the direct-Nesslerization and the distillation-and-Nesslerization analy-
sis.
2. Run 2 has a mass flow rate (by distillation-and-Nesslerization)
consistent with the mass flow rates of Runs 1 and 3.
3. Run 2 has a mass flow rate (by direct-Nesslerization) more than 50%
greater than those of Runs 1 and 3.
Thus, the Run 2 direct Nesslerization ammonia outlet result is suspect.
SAMPLING AND ANALYSIS PROBLEMS WITH UREA, FORMALDEHYDE AND
AMMONIA TESTS
As noted in Section 1, some test runs were discontinuous because the high
concentration and large size of urea particles at the scrubber inlet caused plugging
of the pitot tubes and the probe and nozzle. In order to quickly remove these plugs
from the nozzle, water was squirted into the probe each time a plug occurred.
Since this additional water contributed to the total amount of water collected
during each run, a seperate run for moisture alone was performed at the scrubber
inlet. Details of these problems are contained in Section 5.
At the B granulator scrubber inlet, the large difference between the results
of the two ammonia analysis methods is believed to be due to the conversion of the
-18-
-------
urea to ammonia during distillation. The analysis by direct Nesslerization was
performed at CF Industries in Donaldsonville, and the distillation/Nesslerization
analysis was performed at the TRC laboratory in Wethersfield, CT. The
distillation/Nesslerization method was done in addition to the direct Nesslerization
because of interferences due to turbidity in the direct Nesslerization method.
However, it is documented in the literature that approximately 7 percent of the
urea is converted to ammonia during the distillation procedure. This conversion
factor is not precise, and when the concentration of urea greatly exceeds that of
ammonia, even a small deviation from the 7 percent conversion factor would mean
a considerable amount of ammonia when compared to amount of ammonia actually
measured. Details of this urea conversion are presented in Section 5. The negative
ammonia concentrations shown in Table 2-2 serve to illustrate the magnitude of
error that can result with the distilled Nesslerization method. Therefore, the
ammonia results determined by the direct Nesslerization procedure are considered
to be the more accurate values.
The average ammonia results at the scrubber outlet (Table 2-3) are contrary
to those expected because conversion of urea to ammonia during distillation should
yield a greater ammonia mass flow rate for the distillation/Nesslerization method
than for the direct Nesslerization method. The probability that the direct
Nesslerization results for Run 2 may be anomalously high, as discussed above,may
partially explain this situation.
VISIBLE EMISSIONS FROM "B" GRANULATOR SCRUBBER STACK
The opacity of the plume from the "B" Granulator Scrubber stack ranged
from 5 to 10 percent. The six-minute arithmetic averages are presented
-19-
-------
graphically in Figure 2-1 arid are summarized in Table 2-4. The observations were
made from the synthesis tower with the local terrain as background. The detailed
information on the visible emission measurements can be found in Appendix C.
PARTICLE SIZE TESTS ON "B" GRANULATOR SCRUBBER INLET (TP-1)
Particle size distribution tests were conducted on the "B" Granulator
Scrubber Inlet (TP-1). Summaries of these test results are presented in Table 2-5.
The size of the particulates entering the "B" Granulator Scrubber was 100% >5.7
ym. That is, all the particulate was collected in the cyclone precollector.
Therefore, it was not possible to plot cumulative size distribution curves.
PRESSURE DROP MEASUREMENTS ACROSS "B" GRANULATOR SCRUBBER
The pressure drop measurements across the "B" Granulator Scrubber were
made with a vertical U-tube water manometer which was connected to pressure
taps at the scrubber inlet and outlet. The pressure drop across the scrubber was
recorded at approximately 5 to 15 minute intervals during the tests for urea,
ammonia and formaldehyde.
\'
The pressure drops across the "B" Granulator Scrubber ranged from 19.5 to
21.8 inches of vertical water column and are presented in Table 2-6.
UREA, AMMONIA AND FORMALDEHYDE IN SCRUBBING LIQUID ENTERING
AND EXITING "B" GRANULATOR SCRUBBER
Half-liter samples of the scrubbing liquid streams entering and leaving the
"B" Granulator Scrubber were collected at approximately 30-minute intervals
during the test runs. The solution temperature was measured immediately after
-20-
-------
FIGURE 2-1: SIX MINUTE ARITHMETIC AVERAGES OF JANUARY 17, 18,
AND 19, 1979 OPACITY READINGS ON "B" GRANULATOR
SCRUBBER STACK At CF INDUSTRIES, INC./ :
DONALDSONVILLE, LOUISIANA
1U.U
9.0
>-
t— i
o
o 8.0
LU
1
UJ
>
< 7.0
6.0
5f\
r
i
m*
.0
1
r
i
i
r
2
I
f
j^
fl |
Jn "
1. \ .-
N
1
"
i i i i i
34567891
CM
O
CM O
CM m
• * • •
«* LO
vo
TIME (HOURS)
CM
in
LO
• •
00
o
CM
LO
1-17-79
1-18-79
1-19-79
-------
TABLE 2-4
SIX MINUTE ARITHMETIC AVERAGE'* OPACITY READINGS
ON "B" GRANULATOR SCRUBBER STACK
ON JANUARY 17, 1979, AT CF INDUSTRIES, INC..
DONALDSONVILLE, LOUISIANA
AVG. OPACITY
FOR 6 MIN.
DATE
12:02 - 12:07
12:08 - 12:13
12:14 - 12:17
12:20 - 12:25
12:26 - 12:31
12:32 - 12:37
12:38 - 12:43
12:44 - 12:49
12:50 - 12:55
12:56 - 13:01
13:02 - 13:07
13:08 - 13:13
13:14 - 13:19
13:20 - 13:25
13:26 - 13:31
13:32 - 13:37
13:38 - 13:43
13:44 - 13:40
13:50 - 14:55
13:56 - 14:01
14:02 - 14:07
14:08 - 14:13
14:14 - 14:19
14:20 - 14:22*
13:50 - 15:55
15:56 - 16:01
16:02 - 16:07
16:08 - 16:13
16:14 - 16:19
16:20 - 16:25
16:26 - 16:31
16:32 - 16:36*
8.2
6.3
5.8
5.6
5.2
6.5
6.5
6.9
3.1
8.1
3.3
7.9
9.4
7.3
9.4
8.3
7.7
3.3
8,6
7.4
7.6
3.4
3.3
9.0
9.6
8.5
8.5
8.2 -.
9.3
7.7
8.0
' 9.7
1-17-79
•
1
?
* T
Lass Chan 6 "•*" average
-22-
-------
TABLE 2-4 (Continued)
SIX MINUTE ARITHMETIC AVERAGE % OPACITY READINGS
ON "B" GRANULATOR SCRUBBER STACK
ON 3ANUARY 18, 1979, AT CF INDUSTRIES, INC.,, .
.DONALDSONVILLE, LOUISIANA
TIME
AVG. OPACITY
FOR 6 MT2T.
DATE
12 !%A - 12:49
12:50 - 12j_55_
12:36 - 13:01
13:02 - 13:07
13:08 - 13:13
13:14 - 13:19
13:20 - 13:25
13;26 - 13:31
13:32 - 13:37
13:33 - 13:43
9.0
6.0
5.0
. 5.0 .
5.0
5.0
5.0
5.0
5.0
5.0
1-1S-79
i
P
-23-
-------
TABLE 2-4 (Continued)
SIX MINUTE ARITHMETIC AVERAGE'OPACITY READINGS
ON "B" GRANULATOR SCRUBBER STACK
ON 3ANUARY 19, 1979, AT CF INDUSTRIES, INC.,
• DONALDSONVILLE, LOUISIANA
AVG. OPACITY
FOR 6 MIN.
DATS
09:53 - 09:58
09:59 - 10:04
10:05 - 10:10
10:11 - 10:16
10:17 - 10:22
10:13 - 10:28
10:29 - 10:34
10:35 - 10:40
10:41 - 10:46
10:47 - 10:52
10:53 - 10:53
10:59 - 11:04
11:05 - 11:10
11:11 - 11:16
11:17 - 11:22
11:23 - 11:23
11:29 - 11:34
11:35 - 11:40
11:41 - 11:46
11:47 - 11:52
9.3
8.5
7.4
3.3
7.4
8.1
7.1
7.3
3,5
8.5
9.0
8.2
9.6
8.3
8.3
8.5
3.3
7.9
7.7.
7.3
1-19-79
*
1
-24-
-------
TABLE 2-5
SUMMARY OF INLET PARTICLE SIZE TEST RESULTS ON "B" GRANULATOR
SCRUBBER ON JANUARY 18, 1979, AT CF INDUSTRIES INC., : -DONALDSONVILLE, LOUISIANA
I
KJ
TRC
Test
No.
Cl-1
Ci-2
CI-3
Sampling
Location
Scrubber Inlet
Scrubber Inlet
Scrubber Inlet
Test
Date
1/18/79
1/19/79
1/19/79
Test
Time
15:28-15:43
10:00-10:15
11:25-11:40
Partlculate
Concentration,
grnins/tlscf
*.5I2
3.717
0.933
Aerodynamic
Size Range, |im
> 6.0
> 5.7
> 5.8
Mass In
Size Range, %
99+
99+
99+
-------
TABLE 2-6
SUMMARY OF 3ANUARY 17 AND 18, 1979 "B" GRANULATOR
SCRUBBER PRESSURE DROP MEASUREMENTS AT CF INDUSTRIES, INC.,
DONALDSONVILLE, LOUISIANA
01/17/79
01/17/79
CLOCK
TIME
1207
1212
1217
1222
1227
1232
1237
12*2
12*7
1252
1257
1321
1326
1331
1336
13*1
13*6
1351
1356
1*01
1*06
1*11
1*16
Average
15*9
155*
1559
160*
1609
161*
1619
162*
1629
163*
1639
AP,"H2O
21.8
21,
21,
20,
20
20,
20,
21,
21,
20,
20,
20,
20,
20,
20,
20,
20,
20,
20,
20,
20,
20,
20.5
20.9
20,
20,
20,
20.8
20,
20,
20,
20.8
20.9
21.0
21.0
CLOCK
TIME
16**
1658
1703
1708
1713
1718
1723
1728
1733
1738
17*3
17*8
1753
Average
01/18/79
01/18/79
AP"H2Oi
20.2
20.*
20.*
20.*
20
20
20
20
20
20
20
20
.2
.3
.*
.6
.6
.6
.*
.5
20.2
20.5
09*6
0951
0956
1001
1006
1011
1016
1021
1026
1031
1036
10*1
1122
1127
1137
11*2
11*7
1152
1157
1202
1207
1212
1217
21.6
21.2
21.0
21.1
21.0
21.2
21.0
20.7
20.5
20.7
20.6
20.7
20.5
20.5
20.6
20.3
20.3
20.2
20.1
19.8
20.0
19.7
19.5
Average
20.6
-26-
-------
each sample was collected. The pH was measured once the sample had cooled to
approximately 70° F.
For each test run, the individual samples were combined into two composite
samples (one inlet and one outlet) which were analyzed for urea, ammonia and
formaldehyde content. Table 2-7 presents a summary of the measured urea,
ammonia and formaldehyde concentrations and the percent solids in the scrubber
liquid composite samples. Table 2-8 contains the pH and temperature readings of
each of the individual samples.
The results of the scrubber liquid analyses show that the average urea
concentration of the outlet liquid is more than 5400 times greater than that of the
inlet liquid. This would be expected because urea is being scrubbed out of the gas
stream entering the scrubber. Analysis of the outlet liquid for ammonia by direct
Nesslerization was unsuccessful because the sample solutions turned cloudy when
the Nessler reagent was added. This is indicative of interference, due perhaps to
the high urea concentration. The distillation-and-Nesslerization ammonia results
are questionable, especially the outlet liquid results, because of the possible
conversion of urea to ammonia during distillation. The uncertainty of the urea
conversion is illustrated by the negative ammonia concentrations that result when
the conventional correction is applied.
The average formaldehyde concentration of the inlet liquid is 400 times more
than that of the outlet liquid, indicating that formaldehyde may be reacting with
the urea in the outlet liquid. The lower temperature of the outlet liquid reflects
the endothermic dissolution of urea into the scrubbing medium.
-27-
-------
TABLE 2-7
SUMMARY OF UREA, AMMONIA AND FORMALDEHYDE MEASUREMENTS ON
THE SCRUBBING LIQUID ENTERING AND LEAVING THE
"B" GRANULATOR SCRUBBER AT CF INDUSTRIES INC.,
DONALDSONVILLE, LOUISIANA
Run Number
Date
Location
Urea Concentration (mg/l)e
Uncorrected3
Corrected
Ammonia Concentration (mg/1)
Direct Nesslerization
Distilled Uncorrectedc
,5o Distilled Corrected01
OC
i
Formaldehyde Concentration (mg/1)
pHg
Temperature (°F)h
Percent Solids
Run 1
01-17-79
Inlet
11*
122
180
1381
133
26
9.25
151
1.2
Outlet
516300
552400
f
1*100
-7800
0.01
8.08
12*
3.9
Run 2
01-17-79
Inlet
83
89
**
52'
85
20
9.32
172
0.9
Outlet
525100
561900
f
12800
-9500
0.08
8.16
128
4.6
Run 3
Average
01-18-79
Inlet
9*
101
23
31
27
1*
9.26
175
3.6
Outlet
538300
576000
f
13100
-9700
0.05 '
8.2*
127
*.8
Inlet
97
10*
82
7*
82
20
9.28
166
1.9
Outlet
526600
563*00
-
13330
0.05
8.16
126
*.*
. Kjeldahl analysis results uncorrected for urea conversion during distillation.
Kjeldahl analysis results corrected for urea conversion during distillation (corrected = uncorrected x 1.07).
^Distillation/Nessleration (D/N) analysis results uncorrected for urea conversion during distillation.
D/N analysis results corrected for urea conversion (corrected = uncorrected - 0.07*corrected urea/1.765).
^Milligrams per liter.
Average spectrophotometric measurement not possible due to turbidity.
§ Average of several individual liquid samples taken during each run.
. Average of individual sample temperatures taken immediately after collection.
!Average slight turbidity may have given high instrument readings. Visual estimate for this sample is 60 mg/1.
'Average slight turbidity may have given high instrument readings. Visual estimate for this sample is 30 mg/1.
-------
TABLE 2-8
SUMMARY OF pH AND TEMPERATURE
MEASUREMENTS ON INDIVIDUAL SAMPLES OF SCRUBBING LIQUID
ENTERING AND EXITING "B" GRANULATOR SCRUBBER AT
CF INDUSTRIES, INC., DONALDSONVILLE, LOUISIANA
Sampling
Run No. Date Time Scrubber Inlet Scrubber Outlet
£H TF p_H 5?
1 01/17-79 1215 9.50 135 8.15 126
1255 9.25 154 8.10 124
1331 9.15 164 8.10 124
1416 9.10 152 7.95 120
Average 9.25 151 8.08 124
2 01/17/79 1550 9.70 166 8.30 129
1620 9.50 183 8.20 128
1655 9.20 174 8.20 129
1725 9.20 167 8.10 128
1800 9.00 172 8.00 129
Average 9.32 172 8.16 129
3 01/18/79 0955 9.20 180 8.30 128
1030 9.30 172 8.25 126
1105 9.30 177 8.25 129
1140 9.30 173 8.20 127
1215 9.20 173 8.20 124
Average 9.26 175 8.24 127
-29-
-------
The higher percent solids evident in the outiet scrubber liquid compared to
the inlet may be attributed to organic and inorganic materials insoluable at room
temperature (e.g., biuret and pipescale).
"B" GRANULATOR PROCESS SAMPLES UREA. AMMONIA. AND FORMALD-
EHYDE COMPOSITION
Process samples were taken of the urea melt and the urea product before
screening. These samples were analyzed for urea, ammonia and formaldehyde and
the results of these analyses are summarized in Table 2-9. CF Industries
performed periodic checks on the moisture content of the granulated product. The
average values for the testing days is included in Table 2-9. Bulk density and sieve
analyses were performed on the granulated product samples. These results are
presented in Table 2-10.
AMBIENT TEMPERATURES. BAROMETRIC PRESSURE AND RELATIVE HUMID-
ITIES AT CF INDUSTRIES. INC.. DONALDSONVILLE, LOUISIANA
Table 2-11 summarizes the periodic meteorological measurements taken at
CF Industries, Inc., on January 17 through 19, 1979.
UREA AND AMMONIA TESTS ON SYNTHESIS TOWER MAIN VENT
A summary of the urea and ammonia test results on the Synthesis Tower Main
Vent (TP-1) are presented in Table 2-12. The average urea and ammonia mass flow
rates were 1.70 and ^31 pounds per hour, respectively.
Appendix F contains the pertinent sampling data for the synthesis tower plus
all calculations. Special calculations were required because of the high ammonia
content in the gas stream.
-30-
-------
TABLE 2-9
SUMMARY OF UREA, AMMONIA AND FORMALDEHYDE MEASUREMENTS*
ON THE "B" GRANULATOR MELT AND PRODUCT-BEFORE-SCREENING
ON 3ANUARY 17 AND 18, 1979
AT CF INDUSTRIES INC., DONALDSONVILLE LOUISIANA
Percent By Weight
Urea Urea Product
Melt Before Screening3
Urea Composition
- Uncorrectedb 90.5 . 96.1
Corrected0 96.8 102.8
Ammonia Composition
Direct Nesslerization
Distilled Uncorrected
Distilled Corrected6
Formaldehyde Composition
Moisture
g
1.7
-2.1
0.*
-
g
2.1
-2.0
0.6
0.2
aComposite sample
Kjeldahl analysis results uncorrected for urea conversion during distillation.
cKjeldahl analysis results corrected for urea conversion, (corrected = uncorrected x 1.07)
Distillation/Nesslerzation (D/N) analysis results uncorrected for urea conversion during
distillation.
eD/N analysis results corrected for urea conversion, (corrected = uncorrected - 0.07*corrected
urea/1.765.
Average of measurements by CF Industries Inc.
^Turbidity prevented sample evaluation.
*The data in this table are considered questionable, based on other analyses performed by
CFI and the fact that product samples contain many other compounds not analyzed for
here.
-31-
-------
TABLE 2-10
SUMMARY ON SIEVE AND BULK DENSITY MEASUREMENTS
ON THE "B" GRANULATOR PRODUCT BEFORE SCREENING ON JANUARY 17 and 18 1979
AT CF INDUSTRIES, INC., DONALDSONVILLE, LOUISIANA '
Tola! Sample To Sieves
Sieve No. 6
Sieve No. 8
Sieve No. 10
Sieve No. 12
Sieve No. 10
Sieve No. 16
Bottom
Sum of Mass on Sieves
Bulk Density, Grams
Per 250 ml
Granulator Unscreened
Product
Sample 01 Collected
(01303 on 01/17/79
Mass,
gin
237.36
28.25
00.50
58. '12
'15.56
39.91
13.73
6.92
237.31
213.11
Percent
of
Total
Mass
11.90
18.75
20.62
19.20
16.82
5.79
2.92
OimulaV
tivc
Percent
ol
Total
Mass
100
88.10
69.3'»
'l'i.73
25.52
8.71
2.92
Granululor Unscreened
Product
Sample 117 Collected
(01730 on 01/17/79
•
Mass
f,m
205.26
20.00
'16.77
60.7'l
08.87
00.03
10.70
7.6'l
2'l3.63
2l'».6'i
Percent
of
Total
Mass
10.03
19.20
20.93
20.06
16.59
6.05
3.10
Cumula-
tive
• Percent
of-
Total
Mass
100
89.97
70.77
05.80
25.78
9.19
3.10
Granulator Unscreened
Product
Sample //3 Collected
010*5 on 01/18/79
Mass,
gm
220.37
16.55
02.92
56.15
06.20 ,
00.58
10.37
7.70
220.07
213.53
Percent
of
Total
Mass
7.39
19.12
25.01
20.58
18.08
6.00
3.03
Cumula-
tive
Percent
of
Total
Mass
too
72.63
73.51
08.09
27.91
9.83
3.03
-------
TABLE 2-11
SUMMARY OF JANUARY 17, 18 and 19, 1979 AMBIENT TEMPERATURE
RELATIVE HUMIDITY AND BAROMETRIC PRESSURE MEASUREMENTS
FOR "B" GRANULATOR AT CF INDUSTRIES, INC., DONALDSONVILLE, LOUISIANA
DATE
01/17/79
01/18/79
01/19/79
SAMPLING
RUN TIME
1 1230
1245
1306
1320
1356
1448
Average
2 1523
1635
1715
1815
Average
3 0945
1005
1020
1040
1055
1115
1140
1200
1435
1500
1525
1540
Average
1000
1020
1040
1100
1120
1140
1200
1230
Average
AMBIENT
TEMPERATURE
°F
69
69
69
69
69
70
69
69
67
68
66
68
62
62
64
64
66
67
67
68
69
71
73
71
67
68
68
70
70
70
75
77
73
71
RELATIVE
HUMIDITY
%
76
76
76
76
76
74
76
79
85
80
85
82
94
94
90
87
82
85
83
85
85
77
73
77
84
90
90
86
86
86
70
66
75
81
BAROMETRIC
PRESSURE
INCHES Hg
.
30.40
30.40
30.39
30.39
30.40
30.39
30.39
30.39
30.38
30.39
30.39
30.39
30.39
30.38
30.38
30.36
30.36
30.36
30.28
30.28
30.27
30.27
30.34
30.17
30.17
30.17
30.17
30.14
30.12
30.10
30.10
30.14
-33-
-------
TABLE 2-12
SUMMARY OF RESULTS OF UREA AND AMMONIA TESTS ON GASES EXITING THE UREA SYNTHESIS TOWER VENT
ON JANUARY 18 and 19, 1979 AT CF INDUSTRIES INC., DONALDSONVILLE, LOUISIANA
Run Number
Date
/olume oi Gas
Sampled (including
ammonia)-DSCFvaj
'ercent Moisture
by Volume
'Average Gas
Temperature (°F)
Stack Volumetric
Flowrate (including .
ammonia)-DSC FM""
^reduction Rate*
(tons/hr)
'ercent Isokinetic
•Jet Sampling Time
minutes)
UREA DATA
(jeldahl Analysis Methoc
Total Sample
Weight (milligrams)
jr/DSCF (irojjuding
ammonia)
b/hr
Ib/ton
AMMONIA DATA
Nesslerization
Analysis
Method:
Total Sample
Weight (milligrams)
gr/DSCF (including
ammonia)
Ib/hr
Ib/ton
Run 1 Run 2 Run 3 Average
01-18-79 01-19-79 01-19-79
5.07
67.0
193
306
84.2
15
d e
: uncorrected corrected
189 202
0.574 0.614
1.51 1.62
0.0253 0.0271
I g h
uncorrected corrected
direct distilled distilled
37720 37920 37915
114.6 115.2 115.2
300.5 302.1 302.0
5.06 5.09 5.09
5.30
57.8
186
402
66.5
15
uncorrected corrected
198 212
0.575 0.615
1.98 2.12
0.0352 0.0377
uncorrected corrected
direct distilled distilled
45840 48620 48610
133.2 141.3 141.3
458.9 486.7 486.6
8.15 8.64 8.64
5.14
64.5
191
339
77.9
15
uncorrected corrected
185 198
0.554 0.593
1.61 1.72
0.0286 0.0306
uncorrected corrected
direct distilled distilled
57810 58230 58"220
173.2 174.5 174.5
503.2 506.9 506.8
8.94 9.00 9.00
5.17
63.1
190.0
349
76.2
15
uncorrected corrected
191 204
0.580 0.619
1.70 1.82
0.0297 0.0317
uncorrected corrected
direct distilled distilled
41720 48260 48250
140.3 143.7 143.7
420.9 431.1 431.0
7.83 8.02 8.02
-p
I
I^Dry standard cubic feet (d 68°F, 29.92 inches Hg
; ;Dry standard cubic feet per minute
/^Grains per DSCF
| JUncorrected for urea conversion during distillation
, ./Corrected for urea conversion during distillation (corrected = uncorrected x 1.07)
5* jpirect Nesslerization
,° Distillation and Nesslerization - uncorrected for urea conversion during distillation
Distillation and Nesslerization - corrected for urea conversion during distillation (corrected = uncorrected - 0.07 x corrected urea/1.765)
All production rate information is confidential, per CF1 request
-------
SECTION 3
PROCESS DESCRIPTION AND OPERATION
PROCESS EQUIPMENT
Urea is produced by reacting liquid ammonia (NH.,) with carbon dioxide (CO2)
at an elevated temperature and pressure. The reaction is exothermic and
spontaneous and results in formation of liquid ammonium carbamate (NH2 COo
J- The liquid ammonium carbamate is subsequently decomposed to urea
-)^) and water. The resulting solution of urea in water is concentrated to
89+ percent urea when it is finally solidified.
The Stamicarbon CO2 Stripping Process is the urea synthesis method
employed at this facility. This particular process incorporates emission reduction
and energy recovery techniques in its standard design. The exhaust from the steam
jet ejector vacuum system on the evaporators usually consists of inerts, ammonia,
and traces of urea.
Liquid (70 percent urea) leaving the solution production area goes to a holding
tank prior to being concentrated to 99.5 percent in a two-stage vacuum evaporator.
The concentrated molten urea, referred to as melt, leaves the solution
synthesis process and is pumped to the solids formation equipment. This facility
employs rotary drum granulators, designed by C&I Girdler, as the solids forming
devices. A typical diagram displaying the process components is presented in
Figure 3-1.
-35-
-------
SAMPLE
LOCATION
STACK
SAMPLE
LOCATION
NH-
CO-
SCRUBBER LIQUOR
SOLUTIOM
SYNTHESIS
PROCESS
RECYCLE
STACK'
LIQUOR INLET
MOLTEN UREA
ADDITIVE
ROTARY DRUM
COOLING AIR
SOLID UREA
SCREENS
SOLID RECYCLE
PRODUCT
FIGURE 3-1 TYPICAL UREA MANUFACTURING PROCESS AS DESIGNED BY C & I GIRDLER
-------
The molten urea is sprayed onto a bed of solid urea "seed" particles at the
higher end of the inclined granulator. Lifting flights arrayed inside the granuiator
cause the solid urea "seed" particles to continually fall through the molten sprays
and a counter-cur rent flow of cooling air. The molten urea solidifies on these
"seed" particles, increasing their size. As the particles grow in size, they
eventually spill over a retaining dam into the cooling section of the granulator.
Cooled granules leaving the rotary drum granulator are screened. Oversize
granules are crushed, combined with undersize granules, and returned in solid form
to the bed of material at the spray end of the granulator as make-up "seed."
Product size granules are loaded and/or conveyed to a bulk storage warehouse.
This facility does not coat the product urea granules. Instead, a
formaldehyde-based additive is added to the molten urea before it is sprayed in the
granulator.
Emissions points are the various vents in the solution synthesis process and
the granulator cooling air exhaust. The solution synthesis process vents are
combined into one stack before exhausting to the atmosphere. The granulator
cooling air passes through a scrubber and fan before being exhausted out a stack.
Scrubber liquid is returned to the solution synthesis process for urea recovery.
There are two separate urea production lines at this facility. Each line has
its own urea solution production plant and rotary drum granulator units for granular
urea formation. The production lines are identical except for size and age.
Production line No. 1 was put into service in October 1974, and is rated at
998 Mg/day (1100 tons/day) of solid urea granules. Solid urea is formed in three
parallel rotary drum granulators.
-37-
-------
Production line No. 2 was put into service in 3une of 1977 and is rated at 907
Mg/day (1000 tons/day) of solid urea granules. Solid urea is formed in four parallel
rotary drum granulators. The urea solution process feeding line No. 2 is sized to
allow production of 500 TPD of solution in addition to that required for solids
formation.
EMISSION CONTROL EQUIPMENT
Emission points from each urea solution production line include the high
pressure scrubber vent, the low pressure scrubber vent, and the steam jet ejector
exhaust from the vacuum evaporator concentrators. The high pressure scrubber is
utilized to recover ammonia from various process equipment overhead streams.
The vent is necessary to remove inerts from the overall process
.aNote 3 The scrubber liquor is carbamate recycle. The low pressure
vent on the total recycle part of the plant vents ammonia, water, and inerts. The
steam jet ejectors that maintain a vacuum on the vacuum evaporators also emit
ammonia, water, and inerts. There are no controls on these three emission points
other than standard process stream controls on a total recycle plant.
The exhaust air from each granulator is ducted to separate
.Note * scrubbers °Note 5 On line No. 1, each
scrubber exhausts through its own stack.
Note 3 - See Item 3, Confidential Addendum, Contact Eric Noble, EPA (919)
. 5*1-5213.
Note*- See Item *, Confidential Addendum, Contact Eric Noble, EPA (919)
5*1-5213.
GNote5- See Item 5, Confidential Addendum, Contact Eric Noble, EPA (919)
5*1-5213.
-38-
-------
PROCESS OPERATION
In order to determine whether production line 1 was operating at
representative steady-state conditions during testing, various process and control
equipment operating parameters were monitored.
PRODUCTION RATES
Production rates were calculated using the flow rate as measured by
magnetic volumetric flow meters. These meters measured the urea melt to the
rotary drum granulators in conjunction with the insitu density of the urea melt
(1.2 gm/cc). As a check, totalizer readings on the product conveyor weigh belt
were also monitored. Average product rates for the B granulator during emissions
tests of its scrubber are presented in Table 3-1. Average product rates of the
solution production equipment during emissions testing of this equipment are
presented in Table 3-2. These were calculated using the totals of the feed rates
to all three granulators.
PRODUCTION AND CONTROL EQUIPMENT OPERATION
In addition to the process rate parameters mentioned above, many other
parameters were monitored to insure representative steady-state process
operation. During testing of B granulator emissions, inlet and outlet temperatures
of the urea and air streams through the B granulator were monitored, as was the
inhibitor addition rate. The ammonia feed rate for production line 1 was also
monitored. For the B granulator scrubber, liquor level, temperature, outlet
concentration, and inlet and outlet flow rates were monitored.
-39-
-------
TABLE 3-1 GRANULATOR B FEED RATE DURING GRANULATION EMISSIONS
TESTING
Test Granulator 8 Feed Rate
Scrubber Inlet and Cutlet Test No. 1
Scrubber Inlet and Outlet Test No. 2
Scrubber Inlet and Outlet Test No. 3
Scrubber Inlet Particle Size Test Mo. 1 " •" Note 6
Scrubber Inlet Particle Size Test No. 2
Scrubber Inlei: Particle Size Test No. 3
a'lote 6 - See Item 6, Confidential Addendum. Contact Eric Noble,
EPA (919) 541-5213.
See
TABLE 3-2 SOLUTION FORMATION RATE FOR
PRODUCTION LINE 1 DURING
SOLUTION FORMATION EMISSIONS
TESTING
Test Output as 100 Percent
- Il 7*
aMote 7 - See Item 7, Confidential
Addendum, Contact Eric Noble,
EPA (919) 541-5213.
-40-
GCA TECHNOLOGY DIVISION
-------
During emissions testing of the solution production equipment, CO-
compressor rate and vent scrubber overheads pressure (T105) were monitored in
addition to the above parameters.
Relative average values and relative standard deviations of all the above
parameters monitored during the emissions tests are shown in Table 3-3. A value
of 100 percent for a parameter represents the exact arithmetic average of all
values of that parameter monitored during all tests. Granulator and scrubber
parameters were monitored during the particle size and solution tower tests, and
these values were incorporated into the relative averages and standard deviations
shown for the granulator scrubber inlet and outlet tests. Standard deviations are
not presented for tests where the number of readings was three or less. Actual
values of monitored parameters are presented in Appendix I.
"B1-1 GRANULATOR AND SCRUBBER OPERATION
Average temperatures of the air inlet and outlet streams and the urea outlet
stream to the granulator show considerable variation between test periods. The
difference between average temperatures of a given stream for two test periods
often exceeds the standard deviations of the temperatures during the two tests.
The effect of these variations on the particulate concentration entering the 'B1
granulator scrubber cannot be predicted. Inhibitor addition rate was gradually
reduced between January 17, when the first scrubber air inlet and outlet tests were
made, and January 18, when the last tests were made. It was maintained at a
constant value during particle size characterization tests, although this rate was
somewhat lower than the rates used during scrubber inlet and outlet tests.
-------
TABLE 3-3 RELATIVE AVERAGE VALUES OF OPERATING PARAMETERS DURING TESTING
PAKAIIKTKR
Production K.itr Parii:neteru
Ure.i Helt Feed to Gr.-inulntoi A
Urva Helt Feed to Cranulalor B
Urnn Mvlt Feed tt< Cranulatcir C
Tot.il Solids from A, B, and C
B firnnuliiror Operating Parameters
Air Inlet Temperature
Air Outlet lYwpeiiilure
Ureii lulel Temper at >ire
Siil Ids Outlet Temperature
.p- Formaldehyde Addition Kate
to
Scrubber lli-rrntlng Parameters
Liquor Level
Liquor Teuiperoture
Liquor Feed Rnte
Liquor Dlowdoun Rate
B loud own Urea Content
Oi
>2.l?i
100
101
105
ICO
101
100
100
101
10'.
101
99
105
70
101
.mill. it "I
1200 I-'
SlT. U~t
().')
ll.il
11.5
-
1.0
0.1
0.7
t.l
4.;
1.0
0.5
0.0
(ill.
o.a
r Scrubber
J5 1/JJ.j..
!V. AVR.
HI
100
9*1
9.'
IOU
JUO
IOU
99
104
101
IO-1,
107
11!0
9B
Inlet d
Tt'a'llk'v'r
n.e>
0.0
0.0
1.7
0.6
0.6
0.7
0.0
1.0
0.5
5.b
61.
2.0
Out let
i/iSi.
111
9;
lOb
104
92
9B
IOU
too
101
99
111
1 16
1 (if'
Sl-l. IK'V.
0. 7
h. 1
•i.'i
-
1.5
0.5
0.3
0.7
3.1
1.0
0.0
9.2
1 J6.
4. 'I
Particle Size Tests
liiauulator F-wlnuions
KM- lOu- Hit
1545 1018
11} 100
98 100
105 94
119 98
1147
100
100
•J'l
97
Solution Tower tjnls-
slonB Teslc
1/IH j/19 1/19
1545- 1030- 1110-
IdOl 1047 120!)
Ill 100 100
•iH ]<>•> 100
lo.: '>>• «*'•
110 Mil . 102
Solution J_£i|
-------
Scrubber liquor level was maintained constant during tests of scrubber air
outlet. There was some variation of average liquor feed and blow-down rates,
temperature, and concentration between the three tests, as can be seen in Table 3-
4. The effect of variations in these last three parameters on emissions cannot be
predicted.
SOLUTION PRODUCTION EQUIPMENT OPERATION
Average values of the ammonia feed pump RPM, CO2 compressor rate, and
urea solution feed to the evaporators were fairly uniform for the various test
periods. Vent scrubber overheads pressure also showed little variation.
-43-
-------
TABLE 3-* SCRUBBER OPERATING PARAMETERS DURING SCRUBBER
AIR INLET AND OUTLET PARTICULATE LOADING TESTS
Number See "<** **
I
2 See Note 8a
3 .
aNote 8 - See Item 3, Confidential Addendum, Contact Eric Noble,
EPA (919) 541-5213.
On the morning of January 18, the plant was having trouble maintaining
the correct airmonia to CO? feed ratio. No testing of solution formation
equipment was conducted that morning, and the feed ratio problem is not expec-
ted to have affected emissions from solids formation.
GCA.- TECHNOLOGY DIVISION
-------
REFERENCES
1. Letter to T. Curtin, GCA/Technology Division, from M. Dial, Vice President,
Manufacturing, Lousiana Region, CF Industries, Inc., Donaldsonviile,
Louisiana, May 28, 1979.
-------
SECTION 4
LOCATION OF SAMPLING POINTS
Testing was conducted on the "B" Granulator Scrubber and the Synthesis
Tower Vent at the CF Industries, Inc., urea plant in Donaldsonviile, Louisiana. This
section presents the detailed descriptions of the sampling locations for the urea,
ammonia, formaldehyde, particle size and the opacity measurements.
SCRUBBER INLET. TP-1
The scrubber inlet sampling site was located in a 49 1/2 inch I.D. horizontal
section of steel duct. A schematic of the sampling site including the traverse point
sampling locations and duct dimensions is presented in Figure 4-1. Two 4-inch
pipe-flange sampling ports positioned 90° apart were located 6 feet (1.45 duct
diameters) upstream of a short radius right-angle bend. The distance from the
ports to the nearest upstream disturbance (another right-angle bend) was 17 feet
(4.12 duct diameters).
This inlet sampling location did not meet the "eight and two diameter"
\
criterion as outlined in EPA Method 1; consequently 18 sampling points were
chosen for each axis traverse for a total of 36 sampling points as specified in the
method. Figure 4-1 shows a cross sectional view of the duct at the sampling
location and lists the exact distance of each traverse point from the outside flange
edge.
-46-
-------
SCRUBBER
SCRUBBER
INLET PORTS
6'
17'
DILUTION
AIR
WEST
TRAVERSE
POINT NUMBER
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
TRAVERSE POINT
DISTANCE FROM OUTSIDE
EDGE OF FLANGE, INCHES
5"
6 V8 "
7 K "
9V
11 V
13V
15 5/g "
18 5/a "
22 7/3 "
34 % "
38 7/8 "
41 7/a "
44 V
46 !/„ "
48 i/8 "
49 \ "
51 3/9 "
521/211
FIGURE 4-1:
LOCATIONS OF "B" GRANULATOR SCRUBBER INLET
TEST PORTS AND POINTS AT CF INDUSTRIES INC.,
DONALDSONVILLE, LOUISIANA
-47-
-------
SCRUBBER OUTLET, TP-2
The cleaned gases exiting the scrubber unit are ducted to an induced draft
fan adjacent to the scrubber system. The fan discharge is directed verticaliy
through a steel stack to the atmosphere.
The "B" scrubber 60 inch I.D. outlet stack was fitted with two 3-inch pipe
flange sampling ports positioned 90° apart in a horizontal plane. The two ports
were located 40 feet (8 stack diameters) downstream from the fan discharge, and
20 feet (4 stack diameters) upstream of the top of the stack. See Figure 4-2. The
port locations met the "eight and two diameter" criterion as delineated in EPA
Method 1; consequently 6 sampling points were used for each axis traverse for a
total of 12 sampling points as specified in the method.
INLET PARTICLE SIZE SAMPLING LOCATION, TP-1
Particle sizing test runs were performed in the "B" granulator scrubber inlet
gas stream. An in-stack cascade impactor was positioned in the duct through the
test ports used for the emissions tests. The impactor nozzle was located at the
geometric center of the duct for each impactor run. Only this one central
sampling point was used in order to maintain isokinetic sampling, per standard
procedures established by,the cascade impactor manufacturer.
VISIBLE EMISSIONS OBSERVATION LOCATIONS
The white scrubber stack plume was observed from a position 90 feet above
ground level and about 10 feet above the top of the stack. Observations were made
looking north to the stack, 150 feet away. The plume was observed against a
background consisting of a railroad bed and the ground nearby. This position was
-48-
-------
20'
5'
. sciumuF.it
OUTLET PORTS
40'
-p-
NO
I
I
I
SOUTH
SCRUUDER
TRAVERSE POINT NO.
1
2
3
4
5
6
TRAVERSE POINT DISTANCE
FROM OUTSIDE EDGE OE NIPPLE
(INCHES)
47a'
10 3/.,'
1Q 3/ I
* •* /*|
44 V.,
53 Vi,
59 V«
EAH
FIGURE 4-2:
LOCATIONS OF "B" GRANWLATOR SCRUBBER OUTLET
TEST PORTS AND POINTS A7 CF INDUSTRIES, INC.,
DONAtDSONVILLE. LOUISIANA
-------
selected to conform to the EPA Method 9 requirements. The location of the smoke
observer is shown in Figure ^-3.
SCRUBBER PRESSURE DROP MEASUREMENT LOCATIONS
Pressure drop across the "B" granulator scrubber was measured with a
vertical U-tube water manometer which was connected to pressure taps at the
scrubber inlet and outlet. The inlet pressure tap was located in the inlet duct 10
feet upstream of the scrubber. The outlet pressure tap consisted of a hole drilled
through the transition ducting between the scrubber outlet and the inlet of the fan.
SCRUBBER LIQUID COLLECTION LOCATIONS
Scrubber solution samples were collected from the streams entering and
leaving the "B" granulator scrubber. See Figure 4-^. The inlet sample was tapped
from the solution line just before it entered the scrubber. The outlet sample was
tapped from the pump discharge.
\.
PROCESS SAMPLE COLLECTION LOCATIONS
Throughout the testing program, process samples were collected directly
from their applicable process units/operations. These consisted of samples of urea
melt and the granuiator unscreened product.
-50-
-------
SOLUTION
TOWER
VENT STACK
N
OBSERVER LEVEL
"B" GRANULATOR
SCRUBBER STACK
-90'
-80'
t
FIGURE 4-3:
LOCATION OF SMOKE OBSERVER FOR JANUARY 17 - 19.
1979 OPACITY READINGS ON "B" GRANULATOR SCRUBBER
STACK AT CF INDUSTRIES INC.,
DONALDSONVILLE, LOUISIANA
-------
H
INLET
TAP
"B" GRANULATOR
SCRUBBER
PUMP
OUTLET
TAP
FIGURE 4-4:
LOCATIONS OF SCRUBBER LIQUID COLLECTION
TAPS FOR JANUARY 17 - 19, 1979 TESTS ON
"8" GRANULATOR AT CF INDUSTRIES INC.,
DONALDSONVILLE, LOUISIANA
-52-
-------
UREA SYNTHESIS TOWER VENT LOCATION
The urea synthesis tower sampling site was located in a 14 inch I.D. vertical
section of heavy gauge stainless steel pipe. A schematic of the sampling site is
presented in Figure 4-5. One 6 inch I.D. pipe flange sampling port was positioned
approximately 16 feet (13.7 stack diameters) upstream of a connecting exhaust
vent. The distance between the sampling point and the stack vertex was
approximately 30 feet (25.7 stack diameters). The port location met the EPA
Method 1 criterion. The use of an in-stack orifice in the 14 inch I.D. stack
restricted the sampling to a single point. The nozzle was positioned at its centroid
of the stack for the three runs.
AMBIENT AIR TEMPERATURE AND RELATIVE HUMIDITY MEASUREMENT
LOCATION
The ambient air temperature and relative humidity measurements were made
on the ground next to the CFI chemical laboratory, within 300 yards of the "B"
Granulator.
-53-
-------
14"
'l.O.
FLOW
ft
ft
ft
SOUTH
16'
1
FIGURE 4-5:
LOCATION OF SYNTHESIS TOWER SOLUTION
VENT SAMPLING PORT AT OF INDUSTRIES INC.
OONALDSONVILLE, LOUISIANA
-------
SECTION 5
SAMPLING AND ANALYSIS METHODOLOGIES
This section presents the description of the sampling and analysis
methodologies employed at the CF Industries, Inc. urea manufacturing facility in
Donaldsonville, Louisiana during January 15-19, 1979. Sampling and Analysis
Methodologies are categorized as:
A - Urea, Formaldehyde, and Ammonia "B" Granulator Scrubber
B - Visible Emissions "B" Granulator Scrubber
C - Particle Size "B" Granulator System Inlet
D - Scrubber Liquid "B" Granulator
E - Process Samples "B" Granulator
F - Urea and Ammonia Synthesis Tower Main Vent
The EPA designated methods and any deviations from these methods are
contained in Appendices J and K. This section presents general descriptions of the
methods along with discussions of problems encountered during the test program.
UREA, FORMALDEHYDE, AND AMMONIA METHODS USED ON "B" GRANU-
LATOR SCRUBBER
Urea, formaldehyde, and ammonia in the "B" Granulator Scrubber inlet and
outlet gas streams were sampled at points identified by EPA Method 1 in
accordance with the relationship of the sampling ports to upstream and
downstream flow disturbances. The velocity of the duct gas was measured using
calibrated S-type pitot tubes in accordance with EPA Method 2. Construction and
calibration of the S-type pitot tubes was consistent with EPA Method 2. The
complete methods for sampling and analysis for urea, formaldehyde and ammonia
are contained in Appendix K.
-55-
-------
The sampling train is shown schematically in Figure 5-1 and consists of
nozzle, probe, Teflon line, six impingers, vacuum pump, dry gas meter, and an
orifice flow meter. The nozzle (1) is stainless steel and is of buttonhook shape. It
was connected to a 5/8" stainless steel glass lined probe (2) that is wrapped with
nichrome heating wire and jacketed. The probe temperature was maintained at
approximately 190°F at the inlet and approximately 110°F at the outlet in order to
prevent condensation of the sampled gas. Following the probe, the gas stream
passed through a 3/8" I.D. Teflon line (3) into an ice bath/impinger system (**).
The first and second impingers in the ice bath contained deionized distilled
water (100 ml), the third and fourth contained 100 ml of IN H2SO^ each, the fifth
was dry, and the sixth contained silica gel (200 grams) to remove any remaining
moisture. Leaving the last impinger, the sample stream flowed through flexible
tubing (5), a vacuum gauge (6), needle valve (7), pump (8), and a dry gas meter (9).
A calibrated orifice and inclined manometer (10) complete the sampling train. The
stack velocity pressure was measured using a pitot tube (11) and inclined
manometer. Stack temperature was monitored by a thermocouple attached to the
probe and connected to potentiometer (12). A nomograph was used to rapidly
determine the orifice pressure drop required for any pitot velocity pressure and
stack temperature in order to maintain isokinetic sampling conditions.
The recorded test data included test time, sampling duration at each traverse
point, pitot pressure, stack temperature, meter volume, meter inlet-outlet
temperature and orifice pressure drop. At completion of each run, the nozzle and
probe contents were tripled washed and brushed into a glass sample jar. Next, the
Teflon line between the probe and first impinger was also rinsed three times with
distilled, deionized water into the same glass sample jar which was then sealed
-56-
-------
STACK WALL
THERMOMETER
LEGEND
1 - NOZZLE 7
2 - PROBE , 8
3 - TEFLON LINE 9
4 - ICE BATH 10
5 - FLEXABLE LINE 11
6 - VACUUM GAGE 12
NEEDLE VALVE
PUMP
DRY GAS METER
ORIFICE
PITOT TUBE & INCLINED MANOMETER
POTENTIOMETER
FIGURE 5-Is
MODIFIED £PA PARTICIPATE SAMPLING TRAIN
AUGUST 18,1977, FEDERAL REGISTER
-------
with a Teflon lined cap. The sample jars and impingers were returned to the sample
train prep-and-clean-up room.
At the same prep-and-clean-up room, the contents of the impingers were
transferred to tare-weighed sample jars. The sample jars were used as follows:
Jar #1 - contents of the first and second water impingers, the water wash
of these impingers and their connecting glassware, and the nozzle,
probe and teflon line washes.
Jar //2 - contents of third and fourth acid impingers, the water wash of
these impingers, and their connecting glassware.
*
Jar #3 - silica gel from the sixth impinger.
The contents of jars 1 and 2 were analyzed by a TRC chemist,
Ms. Margaret Fox, at the CF Industries laboratories and also at the TRC
laboratories. Ammonia concentrations were determined by two methods: Ddirect
Nesslerization (on-site) and 2) preliminary distillation (on-site) followed by
•
Nesslerization (at TRC). Urea concentrations were determined by the Kjeldehl
analysis method, as follows: the portion of sample remaining after the preliminary
distillation was then digested and redistilled (on-site) and then brought to TRC for
completion of the urea analysis. Urea and ammonia concentrations were in all
cases determined colorimetrically with a spectrophotometer at a wave length of
405 nm.
The analysis for formaldehyde, done after transporting the samples "on-ice"
back to the TRC laboratories, consisted of mixing an aliquot of the sample solution
with the chromatropic-sulfuric acid reagent to form a purple chromogen. This
colored solution was analyzed colorimetrically using a spectrophotometer at
580 nm.
-58-
-------
In all these analyses, standard solutions were prepared and analyzed, and
calibration curves of absorbance vs. concentration were drawn. These calibration
curves were used to determine the urea, formaldehyde and ammonia concentrations
of the samples. The sample concentration indicated by the calibration curve was
multiplied by the sample volume to determine the total mass of urea, formaldehyde
or ammonia collected.
The distillation-and-Nesslerization analysis method was done in addition to
the direct-Nesslerization method for ammonia analysis because of potential
interference from several species (including formaldehyde and urea) on ammonia
concentrations determined by the direct Nesslerization method. But there also
exists evidence that about 7 percent of any urea in a sample is converted to
ammonia during distillation. Uncertainty in the exact mechanism and rate of urea
conversion, however, leads to uncertainty in indicated ammonia concentrations,
especially so in samples with high urea concentrations. This is evident in the
analyses of the gases entering the "B" granulator scrubber (Table 2-2) and the
liquid exiting this scrubber (Table 2-7). In these cases, the application of the
standard 7 percent correction factor yielded negative "corrected" ammonia
concentrations. Because of the uncertainty in the urea conversion factor, more
confidence is placed in the direct-Nesslerization results.
The main sampling problem during these scrubber gas test runs occurred at
the scrubber inlet. The large size and high concentration of the urea particles at
this location caused plugging of the pitot, nozzle and probe. Plugs were removed
from the pitot by pumping air via a purge line through the pitot tube side facing
Standard Methods for the Examination of Water and Wastewater, 14th Edition,
APHA, AWWA, WPCF, 1975, p. 408.
-59-
-------
upstream; this purge line would then be removed and the pitot reconnected to the
manometer. The frequent nozzle and probe plugging required shutting down the
sampling train and cleaning the nozzle and probe. Cleaning was quickly
accomplished by squirting water into the nozzle to dissolve the plug, as
recommended by the EPA Technical Manager. Plugging was minimized by inserting
the probe into the duct with the nozzle facing downstream. The probe was then
rotated 180 degrees into the flow stream immediately before the initiation of
sampling. The probe was removed from the duct in reverse order. In addition, the
largest possible nozzle was used as a precaution to minimize plugging problems.
The water used to clean the nozzle and probe did, however, contribute to
the total volume of water collected by the impingers during the inlet test runs.
For this reason and because of the large mass of urea collected at the inlet, a
separate test run was performed at the scrubber inlet to determine the moisture
content of the scrubber inlet gas stream. Details of this moisture run and
calculations are contained in Appendix A.
VISIBLE EMISSIONS METHODS USED ON "B" GRANULATOR
The visible emissions measurements were conducted by a certified visible
emission evaluator in accordance with EPA Reference Method 9. The readings were
taken at 15 second intervals. Since the plume was white, it was necessary to read
the emissions against a dark background. The dark background selected was the
railroad tracks and the dark dirt and rock ground surface in that area.
PARTICLE SIZE METHODS USED ON "B" GRANULATOR INLET
A Sierra Model 266 multi-stage cascade impactor was operated in its in-stack
mode. Sampling was performed isokinetically from a single point at the center of
-60-
-------
the scrubber inlet duct. Prior to the initiation of sampling the impactor was leak
tested and placed in the duct for 20 minutes to allow it to heat to duct
temperature to prevent condensation. Sampling was initiated immediately upon
rotation of the nozzle into the flow stream. The brief sampling time at the inlet
necessitated presetting the sample valve so that when the nozzle was pointed into
the flow stream and the pump started, only a minimal adjustment of the valve was
necessary to achieve isokinetic flow.
The impactor was loaded before each run with preweighed glass fiber
collection substrates. After sampling the impactor was removed to the clean up
room and the substrates placed in plastic petri dishes and sealed. The cyclone
contents were brushed into a sample jar and sealed. These samples were weighed
on an analytical balance to 0.1 mg. Additional information on the particle sizing is
presented in Appendix B.
SCRUBBER LIQUID SAMPLING METHODS USED ON "B" GRANULATOR
Half-liter samples of the scrubbing liquid streams entering and exiting the
"B" granulator scrubber were collected at approximately 30 minute intervals during
the granulator scrubber tests. Sample temperatures were recorded immediately
upon collection. The pH of each sample was measured using a pH meter.
All the individual samples collected during a run were combined into one inlet
and one outlet sample. The composite samples were analyzed for urea, ammonia,
and formaldehyde concentration in accordance with the procedure described in
Appendices 3 and K.
-61-
-------
PROCESS SAMPLING AND PRODUCT ANALYSIS
Grab samples of the urea melt and unscreened product were collected at
their respective locations in the process.
The urea melt and unscreened product samples were prepared for analysis by
dissolving a known weight (approximately l.Og) into 100 ml of distilled deionized
water. The solution was then analyzed for urea, ammonia and formaldehyde using
the procedures previously described above and delineated in Appendices 3 and K. A
portion of each of the unscreened product samples were also subjected to bulk
density and sieve analyses. Six sieve sizes were used in the sieve analysis, and the
mass of material required to fill a 250 ml volume was measured for the bulk
density analysis.
UREA AND AMMONIA SAMPLING METHODS USED ON SYNTHESIS TOWER
MAIN VENT
Urea and ammonia in the Synthesis Tower Main Vent Stack were sampled and
analyzed using the methodology and equipment described in the prior subsection
entitled Urea and Ammonia "B" Granulator with the following modifications:
1. An in-stack orifice was used to permit isokinetic sampling of a stream
with a moisture content greater than 50%. The orifice measures the
sample stream at the same moisture conditions as exist in the stack
and therefore changes in moisture will not affect isokinetic sampling.
2. A filter was not used in the sampling train.
3. Two extra impingers were added to achieve^ complete condensation of
the moisture and collection of ammonia. Impingers one through three
contained 100 ml each of distilled deionized water, impingers four
through six each contained 100 ml of 10 N H2SOi», impinger seven was
empty, and impinger eight contained silica gel. The 10 N H2SOi» was
necessary because of the high ammonia concentration and served to
trap ammonia by condensation and neutralization.
4. Only one sampling point (at center of stack) was used because of
physical restrictions imposed by the in-stack orifice nozzle.
-62-
-------
The total volume of water collected in these synthesis tower test runs
consisted of water condensed from vapor and liquid water droplets that were
extracted from the gas stream. The existence of liquid water in the gas stream
was confirmed by observation and by the data shown in Appendix F. These data
show that the total, volume of water collected during each run exceeds the volume
of water present in a saturated gas stream at the stack temperatures.
PRESSURE DROP MEASUREMENTS ACROSS "B" GRANULATOR SCRUBBER
The pressure drop measurements across the "B" granulator scrubber were
made with a vertical U-tube water manometer which was connected to pressure
taps at the scrubber inlet and outlet. The pressure drop across the scrubber was
recorded at 5 to 15 minute intervals during the tests for urea, ammonia and
formaldehyde at the B granulator scrubber.
AMBIENT AIR TEMPERATURE AND RELATIVE HUMIDITY MEASUREMENTS
Ambient air temperature and relative humidity measurements were taken at
the CF1 chemical laboratory with a Bendix psychron at 15 to 30 minute intervals
during each test run.
-63-
------- |