United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EMB Report 83-CUS-20
January 1984
Air
Arsenic
Non-Ferrous Smelter

Emission Test Report
Arsenic Sampler
Comparison
Asarco, Incorporated
Tacoma, Washington

-------
            EMISSION TEST REPORT

         ARSENIC SAMPLER COMPARISON

                ASARCO, INC.
             TACOMA, WASHINGTON

           EMB REPORT NO. 83CUS20
           ESED PROJECT NO.  83/21
                     by

          PEDCo Environmental, Inc.
             11499 Chester Road
               P.O. Box 46100
        Cincinnati, Ohio  45246-0100
           Contract No. 68-02-3849
            Work Assignment No. 1
                  PN 3615-1
                Task Manager
              Mr. Winton Kelly
    U.S. ENVIRONMENTAL PROTECTION AGENCY
     EMISSION MEASUREMENT BRANCH, MD-13
 EMISSION STANDARDS AND ENGINEERING DIVISION
RESEARCH TRIANGLE PARK, NORTH CAROLINA  27711
                February 1984

-------
                           DISCLAIMER
     This report was furnished to the Environmental Protection
Agency, Emission Measurement Branch, by PEDCo Environmental,
Inc., 11499 Chester Road, P.O. Box 46100, Cincinnati, Ohio
45246-0100, in fulfillment of Contract No. 68-02-3849, Work
Assignment No. 1.  Mention of company or product names is not to
be considered as an endorsement by the Environmental Protection
Agency.
                                11

-------
                            CONTENTS

                                                            Paqe
Figures                                                       v
Tables                                                      vii
Acknowledgment                                               ix
Quality Assurance Element Finder                              x

1.   Introduction                                           1-1

2.   Summary and Discussion of Results                      2-1

     2.1  Test protocol                                     2-1

     2.2  Reverberatory furnace                             2-6

          2.2.1  Procedural comparison for arsenic          2-6
          2.2.2  Arsenic emissions and process operation    2-17
          2.2.3  Particulate emission results               2-21
          2.2.4  Summary of flue gas conditions             2-24
          2.2.5  Process sample analytical results          2-27

     2.3  Arsenic plant test results                        2-29

     2.4  Discussion of test results                        2-37

3.   Process Description and Operations                     3-1

     3.1  Process description                               3-1

     3.2  Process operations                                3-10

          3.2.1  Reverberatory furnace ESP                  3-10
          3.2.2  Arsenic plant baghouse                     3-13

4.   Sampling Locations and Test Procedures                 4-1

     4.1  Reverberatory furnace ESP outlet                  4-1

     4.2  Arsenic plant                                     4-9

     4.3  Sample and analytical procedures                  4-11

          4.3.1  Velocity and gas temperature               4-14
          4.3.2  Molecular weight                           4-14
          4.3.3  Particulate/arsenic                        4-15

5.   Quality Assurance                                      5-1
                                 111

-------
                      CONTENTS (continued)
Appendices

 A   Computer Printouts and Sample Calculations             A-l
 B   Field Data                                             B-l
 C   Laboratory Data                                        C-l
 D   Sampling and Analytical Procedures                     D-l
 E   Equipment Calibration Procedures and Results           E-l
 F   Process and Control Device Operating Data              F-l
 G   Quality Assurance Element                              G-l
 H   Project Participants and Sample Log                    H-l
                                IV

-------
                             FIGURES

Number                                                      Page

 2-1      Arsenic Emissions Versus Furnace Charge Rate      2-20

 2-2      Comparison of Calculated Arsenic Emission
           Factor and Furnace Charge Rate                   2-22

 2-3      Arsenic Plant Gas Flow Schematic                  2-36

 3-1      ASARCO-Tacoma Copper Smelter Process Flow
           Flow Diagram                                     3-3

 3-2      ASARCO-Tacoma Arsenic Plant Process Flow
           Diagram                                          3-4

 3-3      Configuration and Electrical Sectionalization
           of Reverberatory Furnace ESP's                   3-7

 4-1      No. 1 Esp Outlet Test Location                    4-2

 4-2      Sediment Profile of the ESP Outlet Duct and
           Location of Sample Points                        4-4

 4-3      Simplified ESP Outlet Velocity Profile            4-8

 4-4      Arsenic Plant Baghouse Outlet and Arsenic
           Trioxide Inlet Test Location                     4-12

 4-5      Metallic Arsenic Plant Inlet Test Location        4-13

 4-6      Sample Recovery and Analysis Flow Chart for
           Particulate/Arsenic                              4-17

 5-1      Audit Report Dry Gas Meter  (FB-1)                 5-7

 5-2      Audit Report Dry Gas Meter  (FB-2)                 5-8

 5-3      Audit Report Dry Gas Meter  (FB-3)                 5-9

 5-4      Audit Report Dry Gas Meter  (FB-4)                 5-10

 5-5      Audit Report Dry Gas Meter  (FB-5)                 5-11

 5-6      Audit Report Dry Gas Meter  (FB-9)                 5-12

-------
                       FIGURES  (continued)

Number                                                      Page

 5-7      Audit Report Dry Gas Meter  (FB-10)                5-13

 5-8      Audit Report Dry Gas Meter  (FB-11)                5-14

 5-9      Thermocouple Digital Indicator Audit Data
           Sheet  (Indicator 125)                            5-15

 5-10     Thermocouple Digital Indicator Audit Data
           Sheet  (Indicator 126)                            5-16

 5-11     Thermocouple Digital Indicator Audit Data
           Sheet  (Indicator 207)                            5-17

 5-12     Thermocouple Digital Indicator Audit Data
           Sheet  (Indicator 262)                            5-18

 5-13     Example of Field Calculation Form                 5-19
                                VI

-------
                             TABLES

Number                                                      Page

 2-1      Summary of Project Test Activity                  2-2

 2-2      Summary of ESP Outlet Arsenic Emission Data       2-9

 2-3      Summary of ESP Outlet Sample and Flue Gas
           Data - Traverse Sampling Trains                  2-11

 2-4      Summary of ESP Outlet Sample and Flue Gas
           Data - Continuous Single-Point Sampling Train    2-13

 2-5      Comparison of Arsenic Emissions Between the
           Traverse Sampling Train and Continuous Single-
           Point Sampling Train                             2-15

 2-6      Comparison of Traverse and Single-Point Sampling
           Results for Sulfur Dioxide                       2-16

 2-7      Summary of Reverberatory Furnace Calcine Charge
           Rates and Arsenic Emission Data                  2-18

 2-8      Summary of ESP Outlet Sample and Flue Gas Data
           for Particulate/Arsenic Tests                    2-19

 2-9      Summary of Particulate Emissions Data at Full
           Furnace Charge Rate                              2-23

 2-10     Summary of Particulate Results from the Contin-
           uous and Traverse Sampling Trains                2-25

 2-11     Comparison of Particulate Emissions Between the
           Traverse and Single-Point Sampling Trains and
           the ASARCO Sampler                               2-26

 2-12     Process Sample Analytical Results, Reverbera-
           tory Furnace                                     2-28

 2-13     Estimate of ESP Arsenic Collection Efficiency     2-30
                              VI1

-------
                       TABLES (continued)

Number                                                      Page

 2-14     Summary of Arsenic Plant Baghouse Sample and
           Flue Gas Data                                    2-31

 2-15     Summary of Arsenic Plant Emissions Data           2-33

 2-16     Process Sample Analytical Results, Arsenic
           Plant                                            2-38

 3-1      Summary of Reverberatory Furnace Production
           Data                                             3-12

 3-2      Process Sample Analytical Results, Reverb-
           eratory Furnace                                  3-14

 3-3      Process Sample Analytical Results, Arsenic
           Plant                                            3-16

 4-1      Summary of Initial Velocity and Temperature
           Profile Data                                     4-6

 5-1      Field Equipment Calibration                       5-3

 5-2      Results of Duplicate Arsenic Tests at ASARCO      5-21

 5-3      Statistical Data for Grouped Arsenic Runs at
           ASARCO                                           5-22

 5-4      Blank Train Analytical Results                    5-23

 5-5      Linear Regression Data (Flame)                    5-24

 5-6      Audit Results                                     5-26

 5-7      Arsenic Standard Obtained by Standard Addi-
           tion Method                                      5-27

 5-8      Field Blanks                                      5-29

 5-9      Reagent Blanks                                    5-30

 5-10     Linear Regression Data—Furnace                   5-31

 5-11     Least Squares Data  (Binominal Equation)--Furnace  5-31

 5-12     Comparison of ASARCO1s and PEDCo's Analytical
           Results                                          5-33
                               Vlll

-------
                         ACKNOWLEDGMENT
     Mr. Winton Kelly, EPA Task Manager, provided overall project
coordination and guidance.  Messrs. Kelly and Frank Clay, also of
U.S. EPA, observed the test program.  Mr. Alfred Vervaert, EPA
Lead Engineer - Industrial Studies Branch, provided project
coordination relative to process operation.  Messrs. Ed Godsey,
Charles Counts, John Richardson, and John Gordon represented
ASARCO, Inc., and provided assistance in scheduling, process
operation, and test observation.  Mr. Charles Bruffey was the
PEDCo Project Manager.  Principal authors were Messrs. Charles
Bruffey, Mark Phillips, David Osterhout, and Thomas Wagner.
                                IX

-------
                QUALITY ASSURANCE ELEMENT FINDER
(1)   Title  page

(2)   Table  of contents

(3)   Project  description

(4)   Project  organization  and  responsi-
     bilities

(5)   QA  objective  for measurement  of data
     in  terms of precision,  accuracy,  com-
     pleteness, representativeness,  and
     comparability

(6)   Sampling procedures
(7)   Sample  custody
(8)   Calibration  procedures  and frequency
(9)   Analytical  procedures
(10)  Data  reduction,  validation,  and
     reporting

(11)  Internal quality control  checks and
     frequency

(12)  Performance  and  system audits and
     frequency

(13)  Preventive maintenance procedures and
     schedules

(14)  Specific routine procedures  used to
     assess  data  precision, accuracy, and
     completeness of  specific  measurement
     parameters involved

(15)  Corrective action

(16)  Quality assurance reports to manage-
     ment
                                                       Location
                                                   Section    Page
             111
Appendix G
Appendix H

Section 5
Appendix D
Appendix G
Section 4
Appendix D
Appendix E

Appendix C
Appendix G

Appendix E
Appendix G

Appendix D
Appendix G

Appendix A
Appendix G

Section 5
Appendix G

Section 5
Appendix G
1-1

G-l
H-l

5-1
D-l
G-l
4-1
D-l
E-l

C-l
G-l

E-l
G-l

D-l
G-l

A-l
G-l

5-1
G-l

5-1
G-l
Appendix G   G-l
Section 5    5-1
Appendix G   G-l

Appendix G   G-l
Appendix G   G-l
                                x

-------
                            SECTION 1



                          INTRODUCTION






     Arsenic is listed as a hazardous air pollutant under Section



112 of the Clean Air Act (National Emission Standards for Hazard-



ous Air Pollutants).  To protect public health from unreasonable



risks associated with exposure to airborne arsenic, the U.S.



Environmental Protection Agency (EPA)  is developing standards to



decrease arsenic emissions from primary copper smelters that



process feed material with an annual average inorganic arsenic



content of 0.7 percent by weight or greater.  The only existing



primary copper smelter in the high-arsenic-throughput category is



owned and operated by ASARCO, Incorporated  (ASARCO) in Tacoma,



Washington.



     Source emissions from six major smelter processes (roasters,



reverberatory furnaces, converters, anode furnaces, arsenic



plant, and sulfuric acid/SO- plant) are vented to the atmosphere



through a single main stack at the Tacoma smelter.  Based on



material balance data and historical test results, the EPA has



estimated that the reverberatory furnace and the arsenic plant



contribute more than 90 percent of the total arsenic emitted from



the stack.  The No. 1 electrostatic precipitator  (ESP) controls



emissions from the reverberatory furnace, and a new fabric filter



(baghouse) controls emissions from the arsenic plant.
                                1-1

-------
     ASARCO measures the dust loss from each process with a

high-volume, single-point continuous sampler.  During specific

time periods, particulate (dust) samples are collected from the

gas stream exiting the No. 1 ESP and arsenic plant baghouse.  One

sample is collected every 24 hours at the ESP outlet, and samples

are collected every 4 to 15 days at the outlet of the arsenic

plant baghouse.  Collected samples are composited and analyzed

monthly for arsenic content, and ASARCO uses these data to esti-

mate monthly arsenic emission losses.

     To support the standards development process, to provide

additional arsenic data for estimating long-term emission rates,

and to resolve differences between arsenic results reported by

ASARCO and the EPA, PEDCo Environmental, Inc., performed a series

of atmospheric emissions tests on the reverberatory furnace and

arsenic plant at the Tacoma smelter for determination of concen-

tration and mass emission rates of particulate matter and

arsenic.  These tests were conducted under contract to EPA's

Emission Measurement Branch (EMB) from September 12 to 29, 1983.

The primary objectives of the test program were:

     1.   To obtain representative arsenic and particulate emis-
          sion data at the outlet of the No. 1 ESP controlling
          emissions from the No. 2 reverberatory furnace.

     2.   To obtain representative arsenic emission data at the
          inlet and outlet of the baghouse controlling emissions
          from the arsenic plant.  Testing was to be conducted so
          as to provide arsenic removal efficiency data for this
          source.

     3.   To obtain data for evaluation of the accuracy of
          arsenic results obtained with the ASARCO continuous
          sampler compared with those obtained with the EPA
          testing and analytical procedures for inorganic
          arsenic.
                                1-2

-------
     4.   To approximate the arsenic removal efficiency of the
          No. I ESP.


     Objectives 1 and 2 were met, Objective 3 was not, and Objec-


tive 4 was partially met.  A direct comparison could not be made


between the EPA sampling and analytical procedures and the ASARCO


sampler because ASARCO could not provide arsenic analyses on a


daily sample basis.  Objective 4 was only partially met because


difficulties were encountered in segregating the dust removed


from the No. 1 ESP to the holding silos.  Data on recovered dust


are available for only 3 of the 6 days of testing.


     The sample and analytical procedures used in the determina-


tion of arsenic emissions from each source followed those de-


scribed in proposed EPA Method 108.*  Particulate concentration


and other pertinent data, including gas flow rates, temperatures,


moisture content, and composition [oxygen  (0?), carbon dioxide


(C0_), and carbon monoxide  (CO)] were determined where applicable


by the procedures described in EPA Reference Methods 1 through 5


of the Federal Register.**  Personnel from either the EPA or


Pacific Environmental Services Inc.  (PES), an EPA contractor,


were present to monitor closely the operating conditions of the


process and control devices under evaluation.  Process samples


from the sources being tested were collected during the test


period and analyzed for arsenic content.
  Draft Method 108.  Determination of Particulate and Gaseous
  Arsenic Emissions from Nonferrous Smelters.
**
  40 CFR 60, Appendix A, Reference Methods 1 through 5, July
  1982.
                                1-3

-------
                            SECTION 2



             SUMMARY AND DISCUSSION OF TEST RESULTS






     This section details the results of the field sampling



project.  For the convenience of the reader, emission results are



presented in both metric and English units, and subsections are



used to define each phase of the sample effort and the corre-



sponding emission results.  Section 3 describes the operation of



the source, Section 4 describes the sample locations and test



procedures used, and Section 5 addresses project-specific data



quality assurance considerations.



     Appendix A presents sample calculations and computer print-



outs for each individual test; Appendices B and C contain the



field data sheets and laboratory analytical results, respec-



tively; Appendix D details the sample and analytical procedures



used and describes the ASARCO sampler and the techniques used to



quantify arsenic loss; Appendix E summarizes equipment calibra-



tion procedures and results for equipment used during this pro-



ject; and Appendix F contains process operating logs and process



monitoring information recorded during the performance of the



tests.





2.1  TEST PROTOCOL



     Table 2-1 presents a summary of the type and number of tests



performed during this test program.  The sample dates, test




                                2-1

-------
                 TABLE 2-1.   SUMMARY  OF  PROJECT TEST ACTIVITY

Date
(1983)
Test
(sample)
ID


Pollutant sampled


Comments
                       Reverberatory  furnace ESP outlet
9/17
9/19-20
9/20-21
9/21-22
9/24
9/25
9/26
9/26-27
9/27-28
9/28-29
AEOPA-1
AEOC-1
AEOT-1
AEOT-2
AEOT-3
AEOC-2
AEOT-4
AEOT-5
AEOT-6
AEOT-7
AEOC-3
AEOT-8
AEOT-9
AEOT-10
AEOT-11
AEOPA-2
AEOPA-3
AEOPA-4
AEOC-4
AEOT-12
AEOT-13
AEOT-14
AEOT-15
AEOC-5
AEOT-16
AEOT-17
AEOT-18
AEOT-19
AEOC-6
AEOT-20
AEOT-21
AEOT-22
AEOT-23
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Part icul ate/arsenic
Particulate/arsenic
Part icul ate/arsenic
Particulate/arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Particulate/arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Particulate/arsenic
Particulate/arsenic
Particulate/arsenic
Particulate/arsenic
Particulate/arsenic
Reverberatory furnace at
full load
Test Block No. 1 for 23-hour
continuous samples
Test Block No. 2 for 23-hour
continuous samples
Test Block No. 3 for 23-hour
continuous samples
Reverberatory furnace at
full load
Test Block No. 4 for 23-hour
continuous samples
Test Block No. 5 for 23-hour
continuous samples
Test Block No. 6 for 23-hour
continuous samples
(continued)
                                   2-2

-------
TABLE 2-1 (continued)

Date
(1983)
Test
(sample)
ID


Pollutant sampled


Comments
                                Arsenic plant
9/14
9/15
9/16
9/17
9/23
9/23
9/24
ABKI-1
AABO-1
ABKI-2
AABO-2
ABKI-3
AABO-3
ABKI-4
AABO-4
ABKI-5
ABMI-1
AABO-5
ABKI-6
ABMI-2
AABO-6
ABKI-7
ABMI-3
AABO-7
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Arsenic
Asp03 baghouse inlet
Baghouse outlet
AspO, baghouse inlet
Baghouse outlet
As203 baghouse inlet
Baghouse outlet
As203 baghouse inlet
Baghouse outlet
ASp03 baghouse inlet
Metallic plant baghouse
Baghouse outlet
As203 baghouse inlet
Metallic plant baghouse
Baghouse outlet
As^CK baghouse inlet
Metallic plant baghouse
Baghouse outlet
inlet
inlet
inlet
                                    2-3

-------
identification, and type of test are grouped by source for re-

porting purposes.  The actual sequence of test events varied

because of reverberatory furnace production curtailments and

arsenic plant production schedules.

     The sample and analytical procedures described in EPA Refer-

ence Methods 1 through 5 and 108* were used to perform a total of

33 emission tests on the off-gases from the No. 2 reverberatory

furnace for determination of the concentration and mass emission

rates of arsenic and particulate matter.  Off-gases from the

reverberatory furnace are vented to the No. 1 ESP for particulate

removal before they enter the atmosphere via the main stack.

Samples were collected in a rectangular brick breeching that

connects the ESP outlet to the main stack.  ASARCO's continuous

sampler is permanently positioned in this breeching.  The test

protocol was designed so that a Method 108  (arsenic) sampling

train could be operated with the sample nozzle positioned at a

single point near the ASARCO sampler probe, and four continuous

Method 108 cross-sectional traverse samples could be collected

during each of six 24-hour sample periods or test blocks.  The

objective in conducting both single-point and traverse tests for

arsenic was to determine if sample results varied between Method

108 and the ASARCO single-point test and if spatial variations

would invalidate single-point sample results.

     In addition to providing data for procedural comparisons,

the design of the sample schedule was such that it generated
 40 CFR 60, Appendix A, Reference Methods 1 through 5, July 1982;
 Method 108 is proposed and in draft form.


                              2-4

-------
arsenic emission data on an almost continuous basis for the

characterization of arsenic emissions relative to furnace charge

rates.

     During each test block, personnel from EPA or PES closely

monitored the operating conditions of both the control device and

the reverberatory furnace and obtained and/or recorded pertinent

operating data.  Process samples, including roaster charge mate-

rial, reverberatory furnace charge (calcine), furnace product

(copper matte), and dust samples from the ESP collection hoppers

and holding silo, were collected during each 24-hour test block,

and collected samples were composited and analyzed for arsenic.

These data were used to characterize process operation during

each test and to approximate the arsenic collection efficiency of

the ESP.

     The test objectives for this source were met, except for a

comparison of arsenic results by Method 108 and those by the

ASARCO sampler.  ASARCO's inability to provide daily arsenic

emissions data make a procedural comparison impossible.

     At the arsenic plant, arsenic concentrations and mass emis-

sion rates were determined at the inlet and outlet of a fabric

filter  (baghouse) controlling emissions from the arsenic trioxide

(As_O.j) and metallic arsenic processes.  All tests were made by

the sampling and analytical procedures outlined in EPA Reference

Methods 1 through 4 and proposed Method 108.*
*40 CFR 60, Appendix A, Reference Methods 1 through 4, July 1982;
 Method 108 is proposed and in draft form.

                               2-5

-------
     The baghouse, which was installed in 1982 and started up in



1983 controls emissions from two process gas streams; one trans-



ports off-gas from the As-CU plant and metallic arsenic con-



densers and the other transports off gases from the metallic



arsenic process.  The gases exiting the baghouse are conveyed to



the main stack.



     Initially, four Method 108 tests were conducted simultane-



ously at the As_0, plant inlet and the baghouse outlet while the



metallic plant was inoperative.  Once the metallic plant came



back on line, Method 108 tests were performed at the As_0, and



metallic plant inlets and the baghouse outlet.  A total of three



Method 108 tests were conducted simultaneously at the three test



locations (two inlet and one outlet).



     These data were used to characterize arsenic emissions to



the main stack and to estimate the arsenic collection efficiency



of the baghouse.  Process operations were closely monitored



during each emission test period, and samples of Godfrey roaster



charge material and baghouse hopper catch were collected and



analyzed for arsenic content.



     The following subsections present the results of the test



program.





2.2  REVERBERATORY FURNACE



2.2.1  Procedural Comparison for Arsenic Sampling



     ASARCO's continuous single-point sampler is permanently



positioned in a breeching after the ESP but prior to the main
                                2-6

-------
stack.  Detailed information on the sampler is presented in



Appendix D.  A single-point sample is collected over a 23-hour



period; the filter bag is changed between 9:00 and 10:00 a.m.



daily.



     To establish a basis for comparison of the ASARCO sampler



and the EPA sample procedures, sampling was conducted simultane-



ously with the ASARCO sampler over the same time period ("23



hours).  Two distinct sampling systems were employed.  The first



system was a single Method 108 sampling train operated isokine-



tically and located at a fixed point in the flue as close as



possible to the ASARCO sampler probe without causing interfer-



ence.  This sampling train, designated AEOC (ASARCO ESP Outlet



Continuous Train), was operated continuously for at least 98



percent of the 23-hour sampling period.  The second system, also



a single Method 108 sampling train, was used to perform a multi-



point, isokinetic traverse of the cross-sectional area of the ESP



outlet breeching.  This sampling train, designated AEOT (ASARCO



ESP Outlet Traverse), was used to perform four consecutive tra-



verses during the 23-hour test period.  Operation of the cumula-



tive sampling train, excluding Test Block No. 1 for the traverse



trains, ranged from 80 to 85 percent of the 23-hour test period.



Start times for each test block corresponded with the start of



the ASARCO sampler.  In summary, a total of six 23-hour test



blocks representing 6 continuous and 23 traverse sample runs were



conducted during the test program.
                               2-7

-------
     Table 2-2 summarizes arsenic emission results from the six



23-hour test blocks evaluated, and Tables 2-3 and 2-4 present



pertinent sample and flue gas data for each individual test.



     Concentrations are reported in milligrams per dry normal



cubic meter (mg/dNm3) and grains per dry standard cubic foot



(gr/dscf).  Volumetric flow rates are expressed in cubic meters



per minute (m3/min) and actual cubic feet per minute  (acfm) at



stack conditions.  Flow rates corrected to standard conditions



[20°C and 760 mmHg (68°F and 29.92 in.Hg) and zero percent mois-



ture] are expressed as dry normal cubic meters per minute



(dNmVnin) and dry standard cubic feet per minute (dscfm) .



Emission rates are expressed in kilograms per hour (kg/h) and



pounds per hour  (Ib/h).  For each test, sample volumes and dry



molecular weight data were corrected for the sulfur dioxide  (S0_)



content of the gas stream.  The SO- concentration was determined



by performing a sodium hydroxide (NaOH) titration on the hydrogen



peroxide  (H-O-) fraction of the sampling train, as described in



Method 108.  As reported in Tables 2-3 and 2-4, the S02 content



typically averaged less than 0.4 percent of the total sample



volume.



     As defined in Method 108, total arsenic is the summation of



arsenic collected in the sampling nozzle and probe, on a heated



glass-fiber filter, and in distilled water contained in the first



two impingers that follow the heated filter.  The sampling probe



and filter temperatures were maintained at approximately 121°C



(250°F).
                               2-8

-------
                                      TABLE 2-2.   SUMMARY OF ESP OUTLET ARSENIC EMISSION DATA
re
i
UD
Test
No.3
1
1
Sample
ID
AEOT-1
AEOT-2
AEOT-3
AEOC-1
Date
(1983)
9/19
9/19
9/19-20
9/19-20
Average
2
2
AEOT-4
AEOT-5
AEOT-6
AEOT-7
AEOC-2
9/20
9/20
9/20-21
9/21
9/20-21
Average
3
3
AEOT-8
AEOT-9
AEOT-10
AEOT-11
AEOC-3
9/21
9/21
9/21-22
9/22
9/21-22
Average
4
4
AEOT-12
AEOT-13
AEOT-14
AEOT-1 5
AEOC-4
9/26
9/26
9/26-27
9/27
9/26-27
Average
Method 108 traverse results
Total arsenic
Concentration
mg/dNm3
4.42
3.29
2.24
3.32
2.72
2.71
2.06
2.47
2.49
2.82
1.10
4.23
2.25
2.60
8.20
14.0
10.4
10.8
10.8
gr/dscf
0.0019
0.0014
0.0010
0.0014
0.0012
0.0012
0.0009
0.0011
0.0011
0.0012
0.0005
0.0018
0.0010
0.0011
0.0036
0.0061
0.0045
0.0047
0.0047
Mass emis-
sion rate
kg/h
2.4
2.0
1.2
1.8
1.4
1.5
1.1
1.3
1.3
1.5
0.6
2.0
1.3
1.4
4.5
7.4
6.1
6.4
6.1
lb/h
5.2
4.3
2.7
4.1
3.0
3.3
2.5
2.9
2.9
3.3
1.4
4.4
2.8
3.0
10.0
16.4
13.4
14.1
13.5
Continuous single-point
Method 108 results0
Total arsenic
Concentration
mg/dNm3
1.56
1.56
1.41
1.41
1.87
1.87
6.86
6.86
gr/dscf
0.0007
0.0007
0.0006
0.0006
0.0008
0.0008
0.0030
0.0030
Mass emis-
sion rate
kg/h
0.8
0.8
0.8
0.8
1.2
1.2
4.4
4.4
lb/h
1.9
1.9
1.9
1.9
2.6
2.6
9.7
9.7
                          (continued)

-------
                                TABLE  2-2  (continued)
KJ
I



Tesj
No.




5



Sample
ID
AEOT-16
AEOT-17
AEOT-18
AEOT-19
AEOC-5



Date
(1983)
9/27
9/27
9/27-28
9/28
9/27-28
Average




6
AEOT-?0
AEOT-H
AEOT-22
AEOT-Z3
AEOC-6
9/28
9/28
9/28-29
9/29
9/28-29
Average
Method 108 traverse results
Total arsenic

Concentration
tng/dNm3
3.07
2.23
9.29
4.66
-
4.82
1.71
4.17
13.3
7.66
-
6.71
qr/dscf
0.0013
0.0010
0 . 004 1
0.0020
-
0.0021
0.0007
0.0018
0.0058
0.0033
-
0.0029
Mass emis-
sion rate
kg/h
1.7
1.2
5.4
2.9
-
2.8
1.1
Z.3
7.8
4.6
-
4.0
Ib/h
3.8
2.7
11.9
6.4
-
6.2
2.3
5.1
17.1
10.1
-
8.6
Continuous single-point
Method 108 results
Total arsenic

Concentration
mg/dNm3

-
.
-
3.82
3.82
_
-
-
-
5.56
5.56
gr/dscf

_
.
-
0.0017
0.0017
_
.
.
-
0.0024
0.0024
Mass emis-
sion rate
kg/h

.
.
-
2.5
2.5
_
.
.
-
3.7
3.7
Ib/h

_
.
-
5.4
5.4
_
.
-
-
8.0
8.0
                               aDur1ng each of six 23-hour test  blocks  (except No. 1), four multipoint traverse samples  and one
                                single-point sample were collected.

                                Total arsenic determined by the  sample  and analytical procedures described in proposed EPA Method
                                108.  Samples were collected isokinetically by use of a multipoint cross-sectional  traverse of the
                                ESP outlet breeching.   Concentrations are expressed In milligrams per dry normal  cubic meter
                                and grains per dry standard cubic  foot.  Mass emission rates are expressed in kilograms  per hour
                                and pounds per hour.

                               cTotal arsenic determined by the  procedures described 1n Proposed EPA Method 108.   Samples  were col-
                                lected by single-point, isokinetic  sampling techniques.

-------
                       TABLE  2-3.   SUMMARY OF  ESP  OUTLET SAMPLE AND FLUE GAS DATA
                                       - TRAVERSE SAMPLING TRAINS  -
Run No.
Sample
source and
location
Date
(1983)
Sampling
period
Sampling
time, mln
Sample
volume
dNm» | dscf
Percent
IsoM-
netic, t
Volumetric flow rateb
Actual
m'/mlnl acfm
Standard
dNm'/minl dscfm
Temper-
ature
°C
*F
Mois-
ture, t
Gas
composition . t
0, CO, | CO
Concentration
of SO., pop
by volume
                                           Reverberatory Furnace No. 2
AEOT-1
AEOT-2
AEOT-3
AEOT-4
AEOT-5
AfOT-6
AEOT-7
AEOT-8
AEOT-9
AEOT-10
AEOT-11
AEOT-12
AEOT-13
AEOT-14
AE01-15
ESP outlet
ESP outlet
ESP outlet
ESP outlet
ESP outlet
ESP outlet
ESP outlet
ESP outlet
ESP outlet
ESP outlet
ESP outlet
ESP outlet
ESP outlet
ESP outlet
ESP outlet
9/19
9/19
9/19
9/20
9/20
9/20
9/20
9/21
9/21
9/21
9/21
9/21
9/22
9/22
9/26
9/26
9/26
9/27
9/27
1015-1621
1635-2304
2347-0602
0948-1445
1530-2030
2120-0227
0312-0805
0948-1439
1544-2041
2119-0223
0310-0814
0940-1440
1537-2036
2115-0227
0315-0831
330
330
330
302.5
302.5
302.5
302.5
302.5
302.5
302.5
302.5
302.5
302.5
302.5
302.5
5.367
5.744
5.606
4.633
5.166
4.948
4.962
4.872
5.458
4.382
5.467
5.238
5.092
5.380
5.563
190.238
202.859
197.964
163.620
182.448
174.734
175.225
172.057
192.734
154.763
193.077
184.976
179.828
189.984
196.456
96.6
95.1
97.0
97.0
98.7
96.9
98.5
96.6
100.2
100.2
100.0
100.4
100.3
99.6
99.8
10.800
12.200
11.500
10.600
11,800
11,500
11,100
11,300
12,000
10.400
12,300
12.200
12.400
13.500
13,300
382.000
431,600
406,100
374.300
416.900
405,100
391.100
398.700
423.400
366 . 100
434.300
430.600
437,000
475.200
469.500
8.900
9.900
9,250
8.450
9,100
9,100
8,700
8,900
9.500
7,800
9,500
9,200
8.800
9,800
9,900
313.400
348,300
327,000
298.400
321 ,800
322,300
309.000
315,100
334,000
275,000
335.000
326,000
311.400
344.200
348,200
77
84
86
87
97
85
69
90
85
101
91
101
110
106
102
171
184
188
189
207
IBS
192
193
184
213
196
213
231
223
216
3.2
2.7
2.4
1.3
3.3
3.1
2.9
2.4
3.3
3.7
3.7
3.4
6.7
6.2
5.4
19.0
19.0
19.0
19.2
19.0
19.0
19.0
19.0
19.0
19.0
19.0
20.0
19.0
19.0
18.8
1.2
1.1
1.2
1.0
1.0
0.8
0.9
1.0
0.9
0.8
1.0
1.7
1.6
1.6
1.5
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
557
695
177
214
389
1614
1004
203
732
2281
1647
1446
3336
3158
2673
(continued)

-------
                   TABLE  2-3  (continued)
to
H
Run No.
AEOT-16
AEOT-17
AEOT-18
AEOT-19
AEOT-20
AEOT-21
AEOT-22
AfOT-23
Sample
source and
location
ESP outlet
ESP outlet
ESP outlet
ESP outlet
ESP outlet
ESP outlet
ESP outlet
ESP outlet
Date
(1983)
9/27
9/27
9/27
9/2B
9/28
9/28
9/28
9/28
9/29
9/29
Sampling
period
094N1459
1534-2031
2115-0219
0306-0836
0945-1443
1534-2033
2115-0216
0301-0802
Sampling
tine, nln
302. 5
302.5
302.5
302.5
302.5
302.5
302.5
302.5
Sample
volume
dNm>
5.394
5. 280
5.631
5.823
5.854
5.329
5.544
5.765
dscf
190.498
186.445
198.867
205.621
206.727
188.190
195.792
203.606
Percent
1 sokl-
net ic, I
99.9
101.7
101.2
99.0
98.8
101.7
99.0
100.1
Volumetric flow rate6
Act
m'/min
12.000
11.800
12.700
13.100
12 .600
1 1 .800
13.000
13.000
ual
acfn
425,000
417.000
449.000
462.000
445.600
417.500
460.000
460.000
Stan
dNmVnin
9,400
9,200
9.700
10.400
10.300
9,300
9.700
10.000
ard
dscfn
331.000
324,500
341.000
367,600
363.400
327,500
343 ,000
353,000
Temper-
ature
°C n
94
92
98
85
83
88
99
91
°F
201
198
208
184
181
191
210
195
Hols-
lure, t
2.9
3.5
4.2
3.7
1.8
3.8
5.6
5.1
Gas
composition1-, t
0,
18.8
19.2
19.2
19.2
19.0
19.0
19.0
19.0
CO,
1.7
1.5
1.3
1.7
1.0
0.8
1.7
1.8
CO
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Concentration
of SO., pom
by volume
556
1137
3374
1372
513
1876
2707
2082
                   *Hetered volume corrected for sulfur dioxide (SO,)  content  In gas stream and expressed  as dry normal cubic meters and dry  standard cubic feet.   Standard
                    conditions are 760 mttg and 20°C (29.92 in.Hg and  68°F).   The SO, content of the gas stream averaged less than 0.3 percent  of  the total sample volune.
                    Volumetric flow rate expressed in cubic neters per minute  and cubic feet per minute at stack conditions.  Flow rates corrected to standard conditions are
                    e«pressed as dry normal cubic meters per minute and  dry standard cubic feet per minute.
                   cFlue gas composition determined by collecting an integrated bag sample and analyzing for oxygen, carbon dioxide, and carbon monoxide Hlth an Orsat Gas
                    Analyzer.  Reported carbon dioxide results have been corrected for SO, content.
                    Concentration of sulfur dioxide in parts per million by volume.

-------
                               TABLE  2-4.    SUMMARY  OF ESP  OUTLET  SAMPLE  AND  FLUE  GAS  DATA
                                           -  CONTINUOUS SINGLE-.POINT  SAMPLING  TRAIN  -
Run No.
Sample
source and
location
Date
(1983)
Sampling
period
Sanpling
tine. Bin
Sample
volume
dNro1 | dscf
Percent
isoki-
netic. %
Volumetric flow rateb
Actual
m'/minl acfm
Standard
dNmVminl dscfm
Temper-
ature
°C [ °F
Mois-
ture. I
Gas
composition . I
0, | CO, | CO
Concentration
of SO,, pom
by volume
                                                           Reverberator/ Furnace No.  2
AEOC-1
AEOC-2
AEOC-3
AEOC-4
AEOC-S
AEOC-6
ESP outlet
Port F
ESP outlet
Port F
ESP outlet
Port F
ESP outlet
Port F
ESP outlet
Port F
ESP outlet
Port F
9/19
9/20
9/20
9/Z1
9/21
9/22
9/26
9/27
9/27
9/28
9/28
9/29
1013-0831
0945-0837
0948-0840
0940-0843
0945-0839
094S-0830
1280
1360
1360
1380
1340
1360
IS. 597
17.679
18.721
19.994
19.09S
19.S27
SS0.813
624.319
661.117
706.100
674.351
689.575
98.8
96.8
95.2
99.9
98.7
97.2
11,100
12,500
13.500
14,400
13,700
13,900
393.500
441.300
477.100
509,000
485.100
489.700
9.100
9.900
10.700
10,700
10,700
10,900
322 ,500
351.100
378.100
379.200
377,600
386.000
80
88
88
101
90
87
176
190
190
215
194
189
2.6
2.9
2.6
4.8
4.1
3.8
19.1
19.0
19.0
18.8
19.2
19.0
1.1
0.9
1.0
1.7
1.5
1.3
0.0
0.0
0.0
0.0
0.0
0.0
420
922
12S9
2643
1704
1968
 Metered volume  corrected for sulfur dioxide content  In gas stream and expressed as dry normal cubic meters and dry  standard cubic  feet.   Standard conditions
 are 760 omHg and 20°C (29.92 In.Hg and 68°F).  The SO, content of the gas stream averaged  less than 0.3 percent of  the total sample volume.

 Volumetric flow rate eipressed In cubic meters per minute and cubic feet per minute at stact conditions.  Flow rates corrected to  standard conditions are
 expressed as dry normal  cubic meters per minute and  dry standard cubic feet per minute.

cFlue gas composition determined by collecting an Integrated bag sample and analyzing for oxygen, carbon dioxide, and carbon nonoilde using an Orsat Gas
 Analyzer.  Reported carbon dioxtde results have been corrected for SO, content.

 Concentration of sulfur dioxide In parts per million by volume.

-------
     Because ASARCO could not provide daily arsenic emissions



data, a procedural comparison between Method 108 and the ASARCO




sampler was impossible.  Table 2-5, however, presents a direct



comparison between Method 108 single-point and traverse test



results.  Reported values of the traverse sampling runs are



averages of individual test results by test block number.  The



average arsenic concentration of the traverse samples was 5.1



mg/dNm3 (0.0022 gr/dscf) and the corresponding average emission



rate was 2.9 kg/h  (6.35 Ib/h).  The average arsenic concentration



during the continuous single-point Method 108 tests was 3.5



mg/dNm3 (0.0015 gr/dscf), and the average emission rate was 2.2



kg/h (4.9 Ib/h).   Both the concentration and mass emission rates



of the single-point arsenic results ranged from 7 to 55 percent



lower than the traverse arsenic results.  This trend was consist-



ent for each test block.  In contrast, the difference in SO-



concentration measured by the traverse and single-point train



averaged less than 5 percent (see Table 2-6).  In addition to the



agreement between traverse and single-point S0_ results, there



was a reproducible correlation within each test block between



volumetric flow rates, temperature, and moisture contents for the



two sampling systems.  These findings would suggest that spatial



stratification of arsenic may exist in the exit breeching.



     Because repetitive multipoint isokinetic traverse sampling



techniques tend to minimize the effects of spatial stratifica-



tion, the Method 108 traverse sampling results are more repre-



sentative of actual arsenic emissions than the single-point



results.
                               2-14

-------
              TABLE 2-5.   COMPARISON OF ARSENIC EMISSIONS BETWEEN THE  TRAVERSE  SAMPLING  TRAIN  AND  CONTINUOUS
                                                SINGLE-POINT SAMPLING  TRAIN
Date
(1983)
9/19-20
9/20-21
9/21-22
9/26-27
9/27-28
9/28-29
Test
block
No.
1
2
3
4
5
6
Average
Test
ID
AEOT-1-3
AEOC-1
AEOT-4-7
AEOC-2
AEOT-8-11
AEOC-3
AEOT-12-15
AEOC-4
AEOT-16-19
AEOC-5
AEOT-20-23
AEOC-6
AEOT
AEOC
Traverse trains
Concentration
mg/dNm3
3.33
2.49
2.60
10.8
4.82
6.71
5.13
gr/dscf
0.0014
0.0011
0.0011
0.0047
0.0021
0.0029
0.0022
Mass
emission rate
kg/h
1.8
1.3
1.6
6.1
2.8
4.0
2.9
Ib/h
4.0
2.9
2.9
13.5
6.2
8.6
6.4
Continuous single-point train
Concentration
mg/dNm3
1.56
1.41
1.87
6.86
3.82
5.56
3.51
gr/dscf
0.0007
0.0006
0.0008
0.0030
0.0017
0.0024
0.0015
Mass
emission rate
kg/h
0.8
0.8
1.2
4.4
2.5
3.7
2.2
Ib/h
1.8
1.8
2.6
9.7
5.4
8.0
4.9
Percent
difference,
traverse
versus .
single point
55 (52)
38 (45)
10 (9.3)
28 (36)
13 (19)
7 (17)
23 (30)
M
I
       Reported values are averages  of the  individual  test  results per  test block.
       Percent difference calculated  based  on  average  emission  rate data  for each test type.  The  number  in
       parentheses is  the percent difference based  on  average concentration data for each test  type.

-------
TABLE 2-6.   COMPARISON  OF  TRAVERSE AND SINGLE-POINT SAMPLING
                 RESULTS FOR  SULFUR  DIOXIDE
Date
(1983)
9/19-20
9/20-21
9/21-22
9/26-27
9/27-28
9/28-29
Average
Test
block
No.
1
2
3
4
5
6

Test
ID
AEOT-1-3
AEOC-1
AEOT-4-7
AEOC-2
AEOT-8-11
AEOC-3
AEOT-12-15
AEOC-4
AEOT-16-19
AEOC-5
AEOT-20-23
AEOC-6
AEOT
AEOC
S00 Concentrations
Traverse
train, average
ppm by volume
476
806
1216
2653
1610
1795
1426
Continuous
single-point
train, ppm
by volume
420
922
1259
2643
1704
1968
1486
Percent
difference
+12
-14
-4
+0.4
-6
-10
-4
                              2-16

-------
2.2.2  Arsenic Emissions and Process Operation



     Table 2-7 presents a summary of reverberatory furnace charge



rates and corresponding arsenic emission data.  In addition to



data on the 23 traverse runs conducted during the continuous



tests blocks, arsenic results for four traverse runs performed at



high furnace charge rates (6 to 8 loads per hour) are also pre-



sented in this table.  These runs are designated as AEOPA



(ASARCO ESP Outlet Particulate/Arsenic).  Table 2-8 summarizes



pertinent sample and flue gas data for these AEOPA test runs.



     The EPA project personnel closely monitored the furnace and



ESP operations during each individual test.  The number of loads



charged to the reverberatory furnace were recorded and divided by



the specific sampling period to obtain the calcine charge rate in



loads per hour.  The calcine charge on a mass rate basis was



calculated by multiplying the charge weight in loads per hour by



the approximate weight per load (~6.3 tons).  Emission factors



for arsenic on a pounds-per-ton basis were developed for each



test based on the calcine mass rate in tons per hour and the



corresponding arsenic mass emission rate.  Figure 2-1 presents



the result of a linear least-squares regression analysis per-



formed on the 27 data points.  The arsenic mass emission results



in pounds per hour versus the furnace charge rate in loads per



hour are plotted.  The data show that arsenic emissions are



generally proportional to the furnace charge rate, particularly



at higher furnace loads.  At low furnace charging rates, the



emission rates did not decrease in proportion to the reduction in
                                2-17

-------
   TABLE  2-7.   SUMMARY OF  REVERBERATORY  FURNACE CALCINE
           CHARGE RATES AND  ARSENIC EMISSION DATA
Date
(1963)
9/19
9/19
9/19-20
9/20
9/20
9/20-21
9/21
9/21
9/21
9/21-22
9/22
9/26
9/26
9/26-27
9/27
9/27
9/27
9/27-28
9/28
9/28
9/26
9/28-29
9/29
9/17
9/23-24
9/24
9/25-26
Test
ID
AEOT-1
AEOT-2
AEOT-3
AEOT-4
AEOT-5
AEOT-6
AEOT-7
AEOT-8
AEOT-9
AEOT-10
AEOT-11
AEOT-12
AEOT-1 3
AEOT-14
AEOT-1 5
AEOT-16
AEOT-17
AEOT-18
AEOT-19
AEOT-20
AEOT-21
AEOT-22
AEOT-23
AEOPA-1
AEOPA-2
AEOPA-3
AEOPA-4
Sampling
period
1015-1621
1635-2304
2347-0602
094B-1445
1530-2030
2120-0227
0312-0805
0946-1439
1544-2041
2119-0223
0310-0814
0940-1440
1537-2036
2115-0227
0315-0631
0947-1459
1534-2031
2115-0219
0306-0836
0945-1443
1534-2033
2115-0216
0301-0802
2021-2339
2230-0144
1851-2258
2222-0131
Calcine
charge
rate,
loads/h
0.34
0.46
0
0
0.50
2.14
1.64
0.20
0.80
4.33
2.96
2.56
7.70
7.60
4.93
0.57
1.61
7.87
3.04
0.20
3.20
7.57
6.40
7.93
8.36
6.09
7.94
Calcine
chargg
rate.
tons/h
2.17
2.92
0
0
3.15
13.51
10.33
1.27
5.07
27.28
18.63
16.12
49.14
47.89
31.08
3.62
10.14
49.61
19.12
1.26
20.16
47.69
40.32
49.94
52.67
38.36
50.02
Arsenic
emission data
Concen-
tration,
gr/dscf
0.0019
0.0014
0.0010
0.0012
0.001?
0.0009
0.0011
0.0012
0.0005
0.0018
0.0010
0.0036
0.0061
0.0045
0.0047
0.0013
0.0010
0.004]
0.0020
0.0007
0.0018
0.0056
0.0033
0.0054
0.0058
0.0042
0.0036
Mass
emission
rate.
Ib/h
5.2
4.3
2.8
3.0
3.3
2.5
2.9
3.3
1.4
4.2
2.8
10.0
16.4
13.4
14.1
3.8
2.7
11.9
6.4
2.3
5.1
17.1
10.1
15.8
16.1
11.9
10.7
Arsenic
emission
factor.
Ib/ton
2.40
1.47
NA
NA
1.05
0.19
0.28
2.60
0.28
0.25
0.15
0.62
0.33
0.28
0.45
1.05
0.27
0.24
0.33
1.83
0.25
0.36
0.25
0.32
0.30
0.30
0.22
aCalcine charge rate in loads  per hour.
     1 calcine load • 6.3 tons calcine

 Calcine charge rate in tons per hour.

cArsenic emission factor in pounds of arsenic per ton calcine charged.  This
 factor represents the total mass of arsenic emitted to the main stack per
 ton  of material  charged to the No. 2 reverberatory furnace.

Note:  All process data provided by U.S.  EPA.
                                   2-18

-------
                                            TABLE  2-8.   SUMMARY OF  ESP  OUTLET  SAMPLE AND FLUE  GAS  DATA
                                                                 FOR  PARTICULATE/ARSENIC  TESTS
NJ
 I
\->
VD
Run No.
AEOPA-]
AEOPA-2
AEOPA-3
AEOPA-4
Sa«p1e
location
ESP out-
let
ESP out-
let
ESP out-
let
ESP out-
let
Date
(1983)
9/17
9/23
9/24
9/24
9/25
9/26
Sa«p1 Ing
period
2021-2339
2230-0144
1951-22SB
2222-0131
Sailing
time, mtn
168
16S
165
165
Sample
volume
dNm>
3.096
2.833
2.960
2.971
dscf
109.321
100.054
104.52S
104.903
Percent
isokl-
netlc, »
100.0
101.8
104.5
100.2
Volumetric flow rate
Actual
mVmln
13,200
13.000
13.000
13.600
acfm
465.500
459,000
459,000
479.000
Standard
dNm'/mln
9.700
9.200
9,400
9,800
dscfm
342.000
325,300
331,000
346 ,000
Temperature
°C
101
110
110
105
°F
214
230
229
221
Noll-
ture. I
5.9
7.1
6.1
6.8
Gas b
conposttlon . I
0,
17.8
17.8
17.8
17.9
CO,
1.6
1.6
1.6
1.7
CO
0.0
0.0
0.0
0.0
Concentration
of SO,, ppm
by volume
2807
3944
2947
3738
                'Metered voluw corrected for sulfur dioxide content In gas stream and expressed as dry normal cubic meters and dry standard cubic feet.  Standard condi-
                 tions are 760 mMg and 20°C (29.92  In.Kg and 68°F).  Average SO, content  of the flue gas stream was less than 0.4 percent of the total sample volune.

                 Flue gas composition determined by  collecting an Integrated bag sample and analyzing for oxygen, carbon dioxide, and carbon monoxide with an Orsat Gas
                 Analyzer.  Reported carton dioxide  results have been corrected for SO, content.

-------
                                  Y  =  1.54X + 1.99
                                  C  =  0.895
            1     234    567    89   10
       X. REVERBERATORY FURNACE CHARGE RATE IN loads/h

Figure 2-1.  Arsenic  emissions  versus  furnace charge rate.
                             2-20

-------
charge, which indicates a residual or base arsenic emission from



the material remaining in the furnace.



     During AEOT Runs 13, 14, 18, 22, 25, and the four AEOPA



runs, furnace charge rates were representative of full smelt



conditions.  For these nine tests, the calcine charge rate aver-



aged 7.5 loads per hour  (47.3 tons/h), and the corresponding



average arsenic concentration and mass emission rates were 10.9



mg/dNm3 (0.0048 gr/dscf)  and 6.2 kg/h (13.7 Ib/h).  Arsenic



emissions for these full-smelt tests ranged from 0.22 to 0.36



Ib/ton of calcine charged and averaged 0.29 Ib/ton.  Figure 2-2



presents a plot of the arsenic emission factors calculated as



pounds per ton of calcine charged versus furnace charge rate in



tons per hour.  The plot shows that when the furnace charge rate



exceeded about 5 tons/h or one load per hour, the arsenic emis-



sion factor was relatively constant.  This emission factor ranged



from 0.15 to 0.62 Ib/ton of calcine charged and averaged 0.30 Ib



per ton.



2.2.3  Particulate Emission Results



     Five test runs for particulate emissions were conducted at



full furnace charge rate conditions  (6 to 8 loads per hour).



These data are summarized in Table 2-9.  Calcine charge rates



averaged 7.3 loads per hour  (45.8 tons/h), and the corresponding



particulate concentration and mass emission rate averaged 20.4



mg/dNm^ (0.0089 gr/dscf)  and 11.7 kg/h  (25.7 Ib/h).



     Three of the continuous single-point sampling runs  (AEOC)



and the traverse  (AEOT) sampling runs from Test Block No. 6 were
                               2-21

-------
0. U
1 2.5
.0
DC
e 2-0
^
o
1 1.5
LU
Z
tu
£ 1.0
o
*~ 0.5


I l 1 1 I I
O
o

— _
o

- o -
00
° AEOT 1 - 23
0 AEOPA 1 - 4
0 <$> °0 0 °0 &° °
o
1 1 1 1 1 1
3 10 20 30 40 50 60
                    FURNACE CHARGE RATE, tons/h
Figure 2-2.   Comparison of calculated  arsenic  emission  factor and
                        furnace  charne rate.
                               2-22

-------
  TABLE  2-9.   SUMMARY  OF  PARTICULATE EMISSION  DATA  AT  FULL  FURNACE  CHARGE RATE
"Date
(1983)
9/23-24
9/24
9/25-26
9/28-29
9/29



Test
ID
AEOPA-2
AEOPA-3
AEOPA-4
AEOT-22
AEOT-23



Charge
rate,
loads/h"
8.4
6.1
7.9
7.6
6.4
xc = 7.3
od * 1.0
RSDe! =13.7
Charge
rate,
tons/h
52.67
3B.36
50.02
47.69
40.32
x = 4S.8
o = 6.2
RSDl =13.5
Paniculate concentration
mg/dNm3
23.3
26.5
16. B
19.9
15.3
; = 20.4
o = 4.6
RSD1 =22.6
gr/dscf
0.0102
0.0116
0.0073
0.0087
0.0067
x = 0.0089
o = 0.0020
RSD5 =22.8
Paniculate
mass emission rate
kg/h
12.9
14.9
9.9
11.6
9.2
x = 11.7
o = 2.3
RSD'» =19.7
Ib/h
28.4
32.9
21.7
25.5
20.2
x = 25.7
o = 5.1
RSD% = 20
 Calcine  charge rate in loads  per hour.

     1  callrie load - 6.3 tons  calcine

 Paniculate represents material collected in the  sample probe and on  the sample filter,  both of which are
 heated to approximately 121°C (2BO°F).  Concentrations are expressed  in milligrams per dry normal cubic
 meter  {mg/dNm3) and grains  per dry standard cubic  foot (gr/dscf).  Mass emission rates are the product of
 the concentration and volumetric flow rate and are expressed in kilograms per hour (kg/h) and pounds per
 hour (Ib/h).

cMean.

 Standard deviation with (N-l) weighting of data.

 Relative standard deviation is the standard deviation expressed as a  percent of the mean.
                                              2-23

-------
analyzed for particulate matter prior to being analyzed for total



arsenic content.  Table 2-10 presents the particulate results



from Tests AEOC-4 through 6 and AEOT-20 through 23.  Although



ASARCO could not provide daily arsenic analyses on the dust



samples collected by the continuous sampler, particulate (dust)



weights were obtained daily and mass emission rates were calcu-



lated based on gas flow and temperature data recorded by the con-



tinuous sampler measurement system.  As a result, a comparison



between ASARCO's continuous sampler and the fixed-point Method



108 for particulate results is available for Test Blocks 4



through 6.  Additionally, a comparison between the traverse and



single-point Method 108 particulate results is available for Test



Block No. 6.  These data are summarized in Table 2-11.  Particu-



late mass emission rate data from Test Block No. 6 indicate less



than a 5 percent difference between the average traverse results



and results obtained by the fixed-point sampling train.  This



would suggest that spatial stratification of particulate matter



is minimal at this location.  In contrast, the particulate re-



sults reported by ASARCO for the continuous sampler ranged from



37 to 75 percent lower than results from the fixed-point continu-



ous sampling train.  The data suggest that the ASARCO continuous



sampler does not collect as much particulate matter as the EPA



type train.



2.2.4  Summary of Flue Gas Conditions



     Flue gas volumetric flow rates, temperatures, moisture



content, and composition [oxygen (02), carbon dioxide  (CO-),
                               2-24

-------
      TABLE 2-10.   SUMMARY OF PARTICULATE  RESULTS  FROM  THE  CONTINUOUS
                       AND TRAVERSE SAMPLING  TRAINS


Test
block No.
4
5
6
6





Date
(1983)
9/26-27
9/27-28
9/28-29
9/28
9/28
9/28-29
9/29


Sample
ID
AEOC-4
AEOC-5
AEOC-6
AEOT-20
AEOT-21
AEOT-22
AEOT-23
Calcine
charge
rate
loads/h
5.70
3.3
4.3
0.20
3.20
7.57
6.40
Participate
concentration


mg/dNm3
17.4
8.9
13.1
6.0
21.0
19.9
15.3

gr/dscf
0.0076
0.0039
0.0057
0.0026
0.0092
0.0087
0.0067
Participate mass
emission rate


kg/h
11.2
5.7
8.6
3.7
11.7
11.6
9.2

Ib/h
24.7
12.6
19.0
8.1
25.8
25.5
20.2
Calcine charge rate in loads per hour.   Charge  rates  for  Runs AEOC-4  through 6
represent average values for the 23-hour test block.
                                    2-25

-------
I.)
 I
NJ
CT>
                               TABLE 2-11.   COMPARISON OF PARTICULATE  EMISSIONS BETWEEN THE  TRAVERSE AND
                                              SINGLE-POINT  SAMPLING  TRAINS AND THE ASARCO  SAMPLER





(1983)
9/26-27
9/27-38
9/28-29






Test
block No.
4
5
6

__ __




Test
ID*
AEOC-4
AEOC-5
AEOI-20-23
AEOC-6
- • -• - ... _ . . , 	
Traverse trains


Concentration
mg/dNm'
-
-
15.5
-
gr/dscf
-
-
0.0068
-

Mass
emission rate
kg/h
-
-
9.1
-
Ib/h
-
-
19.9
-

Continuous sfngle-point train


Concentration
mg/dNm'
17.4
8.9
.
13.1
gr/dscf
0.0076
0.0039
_
O.OOS7

Mass
emission rate
kg/h
11.2
5.7
_
8.6
Ib/h
24.7
12.6
.
19.0
Percent
difference

traverse
and
point train
-
-
A <«

ASARCO
continuous
sampler
partlcu-
late mass
rate, lb/h°
6.42
7.92




Percent
difference
between sfngle-
ASARCO samples
74
37
7t

                 aFor Test Blocks 4 through 6. the single-point continuous train was analyzed for paniculate and arsenic.  During Test Block No. 6. the traverse
                  sampling trains were analyzed for particulate and arsenic.

                 bData provided by ASARCO. Inc.

-------
carbon monoxide (CO), and sulfur dioxide (SO-)] were determined



for each individual test by applicable EPA test procedures.  As



presented in Tables 2-3, 2-4, and Table 2-8, the data are repro-



ducible and correlate with changes in process operation.  For all



tests, volumetric flow rates typically ranged between 8000 and



10,000 dNm3/min (300,000 and 400,000 dscfm).  Temperatures ranged



from 77C to 110°C  (171° to 230°F), and the moisture content



ranged between 2 and 7 percent.  The flue gas composition was



consistent throughout the test period; oxygen, carbon dioxide,



and carbon monoxide contents averaged approximately 19.0, 1.0,



and 0.0 percent, respectively.  The SO- concentrations ranged



from 177 to 3944 ppm by volume and were generally proportional to



the furnace charge rate.



2.2.5  Process Sample Analytical Results



     Table 2-12 summarizes arsenic analytical results from sam-



ples collected by EPA and ASARCO personnel during the test pro-



ject.  These data are also presented and discussed in Section 3



of this report.  In general, the arsenic content of each sample



type compares favorably with historical data from this source and



analytical results routinely reported by ASARCO.



     An arsenic analysis of the dust collected by the No. 1 ESP



combined with an estimate of the mass of dust collected over a



23- to 24-hour period was used to estimate the mass of arsenic



entering the ESP.  These data and the results of the arsenic mass



emission rate tests conducted at the ESP outlet over the same 23-



or 24 hour period were then used to estimate the arsenic collec-



tion efficiency of the ESP.  During the test periods, ASARCO
                              2-27

-------
              TABLE 2-12.   PROCESS SAMPLE  ANALYTICAL  RESULTS,
                           REVERBERATORY  FURNACE
Sample
ID
DD687
DD688
DD689
DD690
DD692
DD693
DD694
DD691
DD696
DD697
DD698
DD695
DD700
DD701
DD702
DD699
DD704
DD705
DD706
DD707
DD703
DD778
DD779
DD780
DD781
Date
(1983)
9/19
9/19
9/19
9/20b
9/21
9/21
9/21
9/21
9/26
9/26
9/26
9/26
9/27
9/27
9/27
9/27
9/28
9/28
9/28
9/28
9/28
9/20
9/21
9/28
9/29
Sample descri
Calcine 7:00 a.m.
Calcine 3:00 p.m.
Roaster charge
Matte
Calcine 7:00 a.m.
Calcine 11:00 p.m.
Roaster charge
Matte
Calcine 7:00 a.m.
Calcine 3:00 p.m.
Roaster charge
Matte
Calcine 7:00 a.m.
Calcine 11:00 p.m.
Roaster charge
Matte
Calcine 7:00 a.m.
Calcine 3:00 p.m.
Calcine 11:00 p.m.
Roaster charge
Matte
R&R pipe and plate
R&R pipe and plate
R&R pipe and plate
R&R pipe and plate
ption
- 3:00 p.m.
- 11:00 p.m.


- 3:00 p.m.
- 7:00 a.m.


- 3:00 p.m.
- 11:00 p.m.


- 3:00 p.m.
- 7:00 a.m.


- 3:00 p.m.
- 11:00 p.m.
- 7:00 a.m.


No. 3 silo
No. 3 silo
No. 3 silo
No. 3 silo

shift
shift


shift
shift


shift
shift


shift
shift


shift
shift
shift






Arsenic, %
2.77
1.57
3.18
0.51
2.56
2.94
4.11
0.48
2.55
3.00
4.21
0.58
2.24
2.49
2.90
0.60
3.00
2.77
3.36
3.86
0.67 -
54.8
54.7
54.7
58.0
aPercent arsenic (by weight)  determined by the sample  preparation  and  analyt-
 ical  techniques described in Proposed EPA Method  108.

 No other process samples were received for arsenic  analysis  on  this date.


Note:   All samples collected  and identified by ASARCO,  Inc.
                                    2-28

-------
determined arsenic collection rates in the following manner.  For



the specific test periods, they transferred collected dust from



the ESP hoppers to a holding silo and determined the approximate



volume by sounding techniques.  They analyzed composite samples



of the dust for density, moisture, and arsenic content and, based



on these data, calculated the mass of arsenic collected on a



pound per hour basis.  Data were not obtained for all six test



periods because of dust conveying problems and operator errors in



segregating the dust collected by the No. 1 ESP.



     Table 2-13 summarizes the ESP arsenic collection efficiency



results for the three days for which complete data are available.



The arsenic analytical results reported for each time period



represent the arsenic content by weight of a composite sample of



ESP hopper dust.  The average arsenic removal efficiency of the



ESP during the specific test periods was 99.1 percent.





2.3  ARSENIC PLANT TEST RESULTS



     Tables 2-14 and 2-15 summarize pertinent sample, flue gas,



and analytical data for tests performed at the arsenic plant



baghouse.



     Initially, four simultaneous tests were conducted at the



As-O..  (kitchen) inlet and baghouse outlet test locations.  During



these tests, the metallic arsenic plant was not in operation.



For the inlet tests, designated ABKI  (ASARCO Baghouse Kitchen



Inlet), the volumetric gas flow rate averaged 731 dNm3/min



(26,000 dscfm) with an average gas temperature of 74°C  (165°F)
                             2-29

-------
            TABLE 2-13.   ESTIMATE  OF  ESP  ARSENIC  COLLECTION EFFICIENCY




Date3
(1983)
9/19-20
9/26-27
9/27-28
9/28-29



Test
Block
No.
1
4
5
6




Time period3
(24-h basis)
1800-1300 (19 h)
1400-1400 (24 h)
1400-1330 (23.5 h)
1330-1330 (24 h)


Arsenic in
Dottrel la
hopper
dust, %
48.7
52.3
47.3
52.1

Estimated
arsenic
collected
in ESP,
Ib/h
477
1074
558
435
Average
ESP
outlet
arsenic
emission
rate, Ib/h
3.4
11.9
5.8
-

Estimated
arsenic
collection
efficiency,
%
99.3
98.9
99.0
-
 All  data provided by ASARCO,  Inc.

 Calculated average ESP outlet emission  data  based on  results from the traverse
 (AEOT) test means for each specific  time  period:

          9/19-20    AEOT-2, 3, and 4
          9/26-27    AEOT-13,  14,  15, and  16
          9/27-28    AEOT-17,  18,  19, and  20


Note:  Estimate of arsenic collection efficiency  is  not  available for Test
       Blocks 2 and 3 because  of a dust  conveying problem.   Estimate of
       arsenic efficiency is not available for Test  Block 6  because test
       data are available for  only 75 percent of  the test period.
                                   2-30

-------
                 TABLE 2-14.   SUMMARY  OF  ARSENIC  PLANT  BAGHOUSE  SAMPLE AND  FLUE  GAS  DATA
Run No.
ABKI-1
AABO-1
ABKI-2
AABO-2
ABKI-3
AABO-3
ASK 1-4
AABO-4
ABK1
AABO
Sample
location
Kitchen
inlet
Outlet
Kitchen
inlet
Outlet
Kitchen
Inlet
Outlet
Kitchen
Inlet
Outlet
Kitchen
Inlet
Outlet
Date
(1983)
9/U
9/14
9/15
9/15
9/16
9/16
9/17
9/17
Sampling
period
1030-1647
1030-1653
1220-1747
1220-1838
0921-1447
0921-1540
0908-1435
0908-1540
Sampling
time, min
342.9
360
176
360
176
360
176
360
Average
Average
Sample
vo 1 ume
dNm'
5.75
7.92
2.85
7.60
2.90
7.50
2.80
7.72
-
-
dscf
203.038
279.799
100.652
268.350
102.340
265.041
99.053
272.758
-
•
Percent
Isoki-
netlc, t
102.0
101.4
100.1
97.9
102.6
98.1
101.2
96.6
-
-
Volumetric flow rateb
Actual
m'/min
943
988
930
975
944
977
876
971
923
978
acfm
33,300
34,900
32.800
34,400
33.300
34.500
30,900
34.300
32.600
34,500
Standard
dNmVmln
746
784
734
779
728
768
714
802
731
783
dscfm
26,300
27,700
25.900
27,500
25.700
27.100
25,200
28,300
25 .800
27.600
Temperature
"C
77
77
75
74
78
78
65
66
74
74
"F
170
171
166
165
173
172
149
152
165
165
Mois-
ture. %
5.3
5.4
5.7C
5.6
6.7
5.8
5.3
4.4
S.8
5.3
Gas .
composition . t
0,
19.0
19.3
19.2
19.2
19.2
19.2
19.2
19.2
19.2
19.?
to,
0.6
0.4
0.4
0.5
0.3
0.4
0.5
0.6
0.5
0.5
CO
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Concentration
of SO,, ppm
by volume
3832
3695
3855
3421
4826
4387
2847
2292
3861
3477
(continued)

-------
                 TABLE  2-14  (continued)
N)

OJ
Run Ho.
ABM -5
ABMI-1
AABO-6
ABKI-6
AMI-?
AABO-6
A8KI-7
ABMI-3
AABO-7
Sample
local ion
Kitchen
inlet
Metallic
inlet
Outlet
Kitchen
Inlet
Metallic
inlet
Outlet
Kitchen
Inlet
Metallic
(nlet
Outlet
Date
(1983)
9/23
9/23
9/23
9/23
9/23
9/23
9/24
9/24
9/2«
Sampl ing
period
0945-1308
0945-1310
094S-1310
1524-1842
I525-IB4S
1525-1845
1005-1324
1005-1326
0950-1326
Sampl 1ng
time, mtn
176
180
180
176
180
180
176
180
180
Metallic
Inlet Average
Kitchen
Outlet Avenge
Sample
volume
dNm1
1.860
1.366
4.112
1.855
1.848
3.947
1.927
2.027
4.154

-
dscf
65.676
48.254
145.202
65.507
65.264
139.392
68.037
71.587
146.385

-
Percent
isoki-
netic. 1
100. fl
104.6
97.8
100.3
101.2
98.2
100.6
103.3
99.0

-
Volumetric flow rateb
Actual
m'/min
554
243
1072
569
230
1044
567
240
1040
238
563
1052
acfm
19.500
8.600
37.900
20.100
8.100
36,900
20.000
8.500
36.700
8,400
19.900
37.200
Standard
dNm'/min
476
153
641
477
147
807
494
157
840
152
482
829
dscfm
16.800
5,400
29,700
16,800
5,200
2B.500
1 7 .400
5.600
29.700
5,400
17.000
29,300
Temperature
C
48
177
79
55
172
86
44
168
74
172
49
80
°r
118
350
174
131
341
186
110
335
166
342
120
175
Mois-
ture, 1
5.2
3.4
5.3
5.5
3.2
5.0
5.5
2.3
4.5
3.0
5.4
4.9
Gas .
composition . I
0,
19.2
19.0
19.0
19.2
19.0
19.4
19.0
19.2
18.9
19.0
19.1
19.1
CO,
0.6
1.0
0.8
0.6
1.0
0.6
0.6
0.6
0.8
0.9
0.6
0.7
CO
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Concentration
of SO,, ppm
by volume
2525
245
2481
2261
72
1749
2262
66
1811
12B
2350
2014
                 *Metered volume corrected for sulfur  dioxide content In gas stream and expressed as dry normal cubic meters and dry  standard cubic feet.  Standard conditions
                  are  760 rmHg and 20"C (29.92 In.Hg and 68°F).  The SO, content of the flue  gas stream averaged less than 0.4  percent  of  the total sample volume.

                  Volumetric flow rate expressed In cubic meters per minute and cubic feet  per minute at stack conditions.  DOM rates  corrected to standard conditions are
                  e«pressed as dry normal  cubic meters  per minute and dry standard cubic feet per minute.

                 °For  ABKI-2 the measured  outlet moisture content was used In calculating the stack gas molecular weight and velocity and  1s not Included In the average
                  moisture value for the inlet tests.

                  Flue gas composition determined by collecting an integrated bag sample and analyzing for oxygen, carbon dioxide,  and  carbon monoxide with an Orsat Gas
                  Analyzer.  Reported carbon dioxide results have been corrected for SO, content.

                 Concentration of sulfur  dioxide in parts per million by volume.

-------
                                               TABLE  2-15.    SUMMARY  OF ARSENIC  PLANT  EMISSIONS  DATA
Run No.
1
7
3
4
Date
(1983)
9/14
9/15
9/16
9/17
Average
5
6
7
9/Z3
9/23
9/24
Average
Arsenic kitchen Inlet
Total arsenic*
Concent
mg/dNm>
3376C
7951
8636
7089
7892
694
1520
889
1035
"UfiP__.
gr/dscf
1.48C
3.46
3.78
3.08
3.44
0.30
0.66
0.39
0.45
Emission rate
kg/h
151C
349
378
303
343
20
44
26
30
Ib/h
333C
770
833
667
748
44
96
58
66
Metallic arsenic Inlet
Total arsenic
Concentration
mg/dNm'
-
-
33.3
35.5
19.7
29.5
gr/dscf
-
-
0.015
0.016
0.009
0.013
Emissic
" kg/h
-
0.31
0.31
0.19
0.77
n rate
Ib/h
0.67
0.69
0.41
0.59
Total Inlet
Concent
mg/dNm'
3376C
7951
8636
70B9
7892
Total ar
ration
gr/dscf
1.48C
3.47
3.78
3.08
3.44
-
-
senlc
Emissio
"»g/h
15IC
349
378
302
343
70.3
44.3
26.2
30.3
n rate
"Ib/h "
333C
770
833
667
757
44.7
96.7
58.4
66.6
Baghouse outlet
Concent
mg/dNm'
4.28
3.59
2.85
1.88
3.15
2.48
5.15
1.85
3.16
Total ar
ration
gr/dscf
0.0019
0.0016
0.0012
0.0008
0.0014
0.0011
0.0023
0.0008
0.0014
senlc___
Emisslor
kg/h
0.20
0.17
0.13
0.09
0.15
0.13
0.25
0.09
0.15

i rate
Ib/h
0.44
0.37
0.29
0.20
0.33
0.28
0.5S
0.21
0.35
Arsenic
removal
effl-.
ciency". I
99.8
99.95
99.97
99.97
99.96
99.4
99.4
99.6
99. S
Arsenic
removal
efficiency
I adjusted
-
-
99.6
99.6
99.7
99.6
ro
 I
        'Total arsenic (filterable and gaseous  fractions) determined  using sample and analytical procedures described in Proposed EPA Method 108.  Concentrations are
        expressed In milligrams per dry normal  cubic meter and grains per dry standard cubic  foot.  Mass emission rates are expressed  in kilograms per hour and
        pounds per hour.

        bThe arsenic collection efficiency of the baghouse Mas calculated using Inlet and outlet mass emission  rate data.

        'Calculation of the  adjusted arsenic collection efficiency of the baghouse was based on the outlet mass emission rate and an adjusted Inlet mass rate, assum-
        ing that all unmeasured flow was equal  In concentration to the kitchen Inlet.  The adjusted inlet mass rates

            Run 5:   64.3  Ib/h
            Run 6:  133.8  Ib/h
            Run 7:   80.7  Ib/h

        ''some sample fraction  lost during analysis and reported results are biased low.   Results from Run I (kitchen inlet) are not  Included In the group average.

-------
and moisture content of 5.8 percent.  The flue gas composition



was consistent for each test and showed oxygen, carbon dioxide,



and carbon monoxide results of 19.2, 0.45, and 0.0 percent,



respectively.  Concentrations of S02 typically averaged less than



3000 ppm or less than 0.3 percent of the total sample volume.



     The uncontrolled arsenic concentration from the As-O., plant



averaged 7892 mg/dNm3 (3.44 gr/dscf), and the corresponding mass



emission rate was 343 kg/h  (757 Ib/h).  Results from Test ABKI-1



are not included in the group average; results of this test are



biased low because of a loss of sample during analysis.  For the



baghouse outlet tests, designated AABO (ASARCO Arsenic Baghouse



Outlet), flow rates averaged 783 dNmVmin (27,700 dscfm) with an



average gas temperature of  74°C (165°F) and moisture content of



5.3 percent.  Average flue gas composition results were identical



to those reported for the kitchen inlet tests.  Outlet arsenic



concentrations and mass emission rates averaged 3.15 mg/dNm3



(0.0014 gr/dscf) and 0.15 kg/h (0.33 Ib/h), respectively.



     Based on mass emission rate results from this group of



tests, the arsenic collection efficiency of the baghouse was



greater than 99.9 percent.  Volumetric flows, temperatures,



moisture contents, and S0? concentrations measured at each loca-



tion were comparable.



     When the metallic arsenic plant began operation, this test



sequence was repeated and simultaneous tests were conducted at



three test locations—the kitchen inlet, the metallic plant



inlet, and the baghouse outlet.  At the completion of the first
                             2-34

-------
set of simultaneous tests (ABKI-5, ABMI-1, and AABO-5),  prelim-



inary calculations showed a flow imbalance between the inlet and



outlet test locations.  The cumulative inlet volumetric flow was



629 dNiWmin (22,200 dscfm)  compared with an outlet flow of 841



dNm3/min (29,700 dscfm).  The 7000-dscfm flow imbalance was



attributed to an open flow control damper located in a bypass



duct, which entered the metallic arsenic plant exhaust duct



downstream of both the metallic and kitchen inlet test locations



(See Figure 2-3).   This condition did not exist during the first



series of runs because a second flow control damper located in



the metallic plant duct downstream at the bypass duct was closed.



The flow imbalance occurred when this second damper was opened.



The negative pressure associated with the control system served



to divert a part of the flov; from the kitchen through the bypass



duct and into the baghouse,  where it was ultimately measured at



the outlet test location.



     In addition to the flow imbalance, the arsenic concentration



and the mass rate measured at the As_O, inlet test location was



significantly less than that measured during the first set of



tests (0.45 gr/dscf and 66 Ib/h versus 3.44 gr/dscf and 748



Ib/h).  No conclusive explanation can be found to account for the



significant difference in As_0., plant loading.   (A discussion of



the process operation is given in Subsection 3.2).  The arsenic



concentration and mass emission rates at the baghouse outlet



averaged 3.17 mg/dNm3 (0.0014 gr/dscf) and 0.15 kg/h  (0.35 Ib/h).
                              2-35

-------
BLOWER
                                    OUTLET
                                 SAMPLE LOCATION
                                   I
                                 P=-2"H20
                               KITCHEN INLET
                              SAMPLE LOCATION
                                     P=-3"H20
                                            DAMPER
                                          INDICATES
                                            CLOSED
                  METAL FURNACE HOODS
TO MAIN STACK


    ^ SLIDE BLIND,OPEN

    \WOOD DUCT
                     MAIN FLUE
                                                                     FROM ARSENIC
                                                                       KITCHENS
               SLIDE BLIND,CLOSED
                                                 FLOW DIVERSION
                                                METALLIC
                                               INLET SAMPLE
                                                LOCATION
               Figure 2-3.   Arsenic  plant  gas  flow schematic,


                                     2-36

-------
These values are essentially identical to those measured during



the first test series, when only the As^O., plant was operated.



     Because a malfunctioning flow control damper made the inlet



mass emission rates suspect, the arsenic collection efficiency of



the baghouse was recalculated with an adjusted arsenic inlet mass



rate for each run.  Results were adjusted by assuming the flow



imbalance was diverted to As_0- gas.  The flow difference was



assumed to have the same concentration as the kitchen inlet and



was added to the total inlet mass rate.  The arsenic collection



efficiency averaged 99.5 percent with both arsenic processes in



operation without a calculation adjustment and 99.6 percent with



an adjustment.



     Table 2-16 summarizes arsenic analytical results for Godfrey



roaster charge and baghouse dust samples collected by ASARCO



during each test.  (Personnel from EPA monitored process and



control equipment operation during each test, and these data are



presented in Section 3 and Appendix F of this report.)





2.4  DISCUSSION OF TEST RESULTS



     Overall, the test program was conducted as planned, and no



major problems were encountered with the sampling or analytical



phases of the project.  With the exception of a comparison of



arsenic results obtained by the ASARCO sampler with those ob-



tained by EPA sample and analytical procedures, all test objec-



tives were met.
                              2-37

-------
                TABLE 2-16.   PROCESS SAMPLE ANALYTICAL  RESULTS,
                                 ARSENIC PLANT
Date
(1983)
9/14
9/15
9/16
9/17
9/23
9/23
9/24
Sample description
Roaster charge
Baghouse dust
Roaster charge
Baghouse dust
Roaster charge
Baghouse dust
Roaster charge
Baghouse dust
Roaster charge 7:00 a.m. - 3:00 p.m.
Baghouse dust 7:00 a.m. - 3:00 p.m.
Roaster charge 3:00 p.m. - 11:00 p.m.
Baghouse dust 3:00 p.m. - 11:00 p.m.
Roaster charge 7:00 a.m. - 3:00 p.m.
Baghouse dust 7:00 a.m. - 3:00 p.m.
Arsenic,3 %
37.2
74.5
27.8
73.5
25.3
75.2
32.2
67.7
33.2
75.8
61.6
72.8
47.1
71.8
aPercent arsenic (by weight)  determined by the  sample  preparation  and
 analytical  techniques described in proposed EPA Method  108.

Note:  All  samples collected  and identified by  ASARCO,  Inc.
                                  2-38

-------
     Results of the comparison of traverse and single-point EPA



sampling procedures show that the single-point system consist-



ently measures less total arsenic than the traverse sampling



system does.  In contrast, data from Test Block No. 6 show that



particulate measurements by the two sample systems are compar-



able.  This would suggest spatial stratification of arsenic in



the ESP exit breeching and minimal stratification of particulate



matter.  Cross-sectional traverse sampling techniques would mini-



mize the effect of stratification, and results from these tests



are considered more representative of total arsenic emissions



than the single-point results are.  A comparison of particulate



results on a mass-rate basis obtained by the ASARCO sampler and



the fixed-point Method 108 sampling train indicate that the



ASARCO sampler system measures less particulate  over the same



approximate period of time.



     Results from tests conducted at the ESP outlet showed arsen-



ic emissions to be generally proportional to furnace charge



rates, particularly at higher furnace loads.  For tests conducted



at the full furnace charge rate (6 loads per hour or greater),



the total arsenic concentration averaged 10.9 mg/dNm3  (0.0048



gr/dscf) and the emission rate averaged 6.2 kg/h  (13.7 Ib/h).



Particulate concentrations averaged 20.4 mg/dNm3  (0.0089



gr/dscf), and the mass emission rate averaged 11.7 kg/h  (25.7



Ib/h).  For tests that yielded both particulate and arsenic



results, the total arsenic emission results were approximately 50



percent of the particulate results on both a concentration and
                            2-39

-------
mass-rate basis.  This correlates with process analytical data,



which showed an arsenic content of approximately 50 percent by



weight in dust collected from the ESP collection hoppers and



holding silo.



     Tests results from the arsenic plant showed an average total



arsenic concentration of 3.2 mg/dNm3 (0.0014 gr/dscf)  to the main



stack and an average mass rate of 0.15 kg/h (0.34 Ib/h).  Inlet



and outlet test results indicated that the arsenic collection



efficiency of the baghouse ranged from 99.4 to 99.97 percent,



depending on the operating mode.  No significant difference was



noted in the baghouse outlet emissions between tests conducted



with only the As_0.. plant in operation and tests conducted when



both the As_0^ and metallic arsenic plants were in operation.
                             2-40

-------
                            SECTION 3



               PROCESS DESCRIPTION AND OPERATIONS






     This section presents a brief description of the smelting



processes and facilities operated at the ASARCO primary copper



smelter at Tacoma,  Washington, and details on the design and



operation of the control devices tested during this test program.



As noted previously, the latter included the electrostatic pre-



cipitators serving the reverberatory smelting furnace and the



baghouse for the control of the combined process gases from the



arsenic trioxide and arsenic metal plants.  Information is also



presented on the operating conditions of the process facilities



and control devices tested during the test program.  Pertinent



operating logs and process monitoring data recorded during the



tests are included in Appendix F.





3.1  PROCESS DESCRIPTION



     The ASARCO smelter at Tacoma, Washington, is a custom smelt-



er that processes copper ore concentrates, precipitates, and



smelter byproducts from numerous domestic and foreign sources.



The smelter is capable of producing about 320 Mg  (353 tons) of



anode copper per day and houses the only arsenic production



facility in the United States.  Inorganic arsenic is recovered as



a byproduct of the flue dusts produced during the processing of
                               3-1

-------
high-arsenic copper-bearing feed materials.  Smelter operations



are subject to a Meteorological Curtailment Program designed to



prevent violations of ambient air quality standards for sulfur



dioxide.  Under the program, smelter activities are curtailed



when meteorological conditions and ambient air sulfur dioxide



measurements indicate the standards may be exceeded.



     Copper smelting facilities include 10 multihearth roasters,



two reverberatory smelting furnaces, three Peirce-Smith convert-



ers (a fourth Peirce-Smith converter generally serves only as a



holding furnace), and two anode furnaces.  Arsenic production



facilities consist of three Godfrey roasters, three arsenic



trioxide settling chambers or kitchens, storage facilities, and a



metallic arsenic plant.  Process flow diagrams of the copper



smelting and arsenic production facilities are presented in



Figures 3-1 and 3-2.



     The roaster charge, which consists of a blend of copper ore



concentrates, precipitates, lead speiss, flue dust, and fluxing



materials, typically contains 3 to 5 percent arsenic and 7 to 10



percent moisture.  At full smelt, four or five roasters are used.



Charging is continuous.  The calcine produced is intermittently



discharged from hoppers located below the roasters into larry



cars that transport the calcine to one of the reverberatory



furnaces.  Each larry car can transport about 5.9 Mg (6.5 tons)



of calcine.



     Offgases from the roasters, which average about 3,560 m3/min



(126,000 acfm) at 260°C (500°F), are treated in a baghouse for
                               3-2

-------
                                                             MAIN  STACK
OJ
 I
                                                                  DUST RECYCLE TO ARSENIC PLANT
            RECYCLE
              FROM
               FOB
            COPPER ORE
                                                                                                 ANODES
                                       SLAG TO  DUMP
FOB » FINE ORE BIN
                                Figure  3-1.   ASARCO-Tacoma  copper smelter process  flow diagram.

-------
                                                 MAIN STACK
                                              TEST LOCATION
                                         BAGHOUSE
                      TEST LOCATION
                              (x
SHUT OFF"
  DAMPER
     As2°3
 DUST RECYCLE
      FROM
CONTROL DEVICES
   RECYCLE
                                  RECYCLE TO
                                   ROASTERS
MALFUNCTIONING
    DAMPER
                                                          -*• BYPASS DUCT
                                                         (NO LONGER IN USE]
                  *-PRODUCT

                  **DUST TO
                    LEAD PLANT
        COPPER ORE
             As,0
                       TO MAIN STACK
                              t
    ^COPPER ANODES

      As PRODUCT

      DUST TO  LEAD PLANT
                        SLAG TO DUMP

      Figure  3-2.  ASARCO-Tacoma  arsenic  plant  process  flow diagram.
                                     3-4

-------
removal of particulate matter.  Before the roaster gases enter



the baghouse, they are tempered with dilution air to reduce the



gas stream temperature to less than 120°C (250°F).  The baghouse,



a mechanical shaker type, is designed for effective treatment of



5664 m3min  (200,000 acfm) of gas at an air-to-cloth ratio of 0.58



m3/min per m2 (1.87 cfm/ft2) to one.  Emission tests that EPA



conducted in September 1978 show that the baghouse is 99.7 per-



cent efficient in reducing arsenic emissions.  The baghouse



exhaust is vented through a flue to the smelter's main stack.



     Although the smelter has two reverberatory smelting fur-



naces, Furnace No. 2 is used almost exclusively.  This furnace is



33.5 m (110 ft)  long and 9.8  (32 ft) wide, and its smelting



capacity is about 1090 Mg (1200 tons) per day.  The furnace is



usually fired by natural gas.  Furnace charging is accomplished



by discharging the larry cars through one of four Wagstaff guns



located along the furnace sidewalls.  Typically, discharge time



for one car is about 1 minute.  At full smelt, eight cars are



charged per hour  (two every 15 minutes).  Matte from the furnace,



which normally contains 43 percent copper, is periodically tapped



into a ladle and transferred by an overhead crane to one of three



Peirce-Smith converters.  Similarly, slag produced in the furnace



is periodically tapped and transferred to a 5-pot slag train that



transports it to a slag dump.



     Process gases from the reverberatory furnace, which average



about 1,415 Mn3  (50,000 scfm), pass through a pair of waste heat



boilers, where the gases are cooled to about 400°C  (750°F).  The
                               3-5

-------
exiting gases then pass through a large rectangular brick flue,



(No. 1 flue), where additional cooling and gas stream condition-



ing take place by air infiltration and the addition of water and



sulfuric acid via sprays located in the flue.  The resultant gas



stream, about 6,100 actual mVmin (215,000 acfm)  at about 110°C



(230°F), then enters the first of two electrostatic precipitators



in series for particulate removal.  The first precipitator is a



tube or pipe design and consists of 18 sections with a total



collection area of about 6,619 m2 (71,250 ft2).  Each section



contains 84 pipes measuring 30 cm (12 inches) in diameter and 4.6



m  (15 feet) in length.  The second unit is a plate-type design



and consists of 28 sections (7 wide by 4 deep), with a total



plate collection area of 7,582 m2 (81,648 ft2).  Each section is



equipped with 16 corrugated steel electrode plates spaced 15.2 cm



(6 inches) apart and measuring about 2.3 m (7.5 ft) in length and



3.7 m  (12 ft) in height.  Power to the precipitators is supplied



by 19 transformer/rectifier sets, 9 on the pipe unit and 10 on



the plate unit.  A diagram showing the configuration and elec-



trical sectionalization of the precipitators is presented in



Figure 3-3.  Both precipitators are rapped about every 2 hours.



Dust collected in the precipitators is pulled 5 days per week.



     As noted previously, matte from the reverberatory furnace is



transferred to one of three Peirce-Smith converters.  Two of the



converters measure 4.0 m  (13 feet) in diameter by  9.1 m  (30 feet)



in length, and the third converter is 4 m  (13 feet) in diameter



and 10.7 m  (35 feet) long.  In addition to copper matte, smelter
                               3-6

-------
    * KEY:
    TOP NUMBER  IDENTIFIES
    THE SECTION WHILE THE
    LOWER NUMBER IDENTIFIES THE
    TRANSFORMER/RECTIFIER SET
    USED TO ENERGIZE THE SECTION.
                                                     PLATE  ESP
TO NO. 2 /
PLATE ESP >
/
/






ANODE FURNACE AND
VENTILATION GASES
\
V






NO. 1 FLUE
REVERB GASES 	
( \
\^/

-^
S





































1

1

2

2

6

7

7

5

5
F
1

2

3

4

5

6

7

8

9

18
3
17
3
16
4
15
4
14
6
13
8
12
8
11
9
10
9
NO. 1 CROSSOVER
^•"
x-"

NO. 2 FLUE


                                                     PIPE ESP
Figure  3-3.   Configuration  and electrical  sectionalization
                   of  reverberatory  furnace  ESP's.
                                 3-7

-------
reverts and copper scrap materials are also processed.  Typi-



cally, only one converter is on blow at any one time, and at no



more than two converters are ever operated simultaneously.  A



converter cycle normally takes from 10 to 12 hours.  With dilu-



tion air, the off-gas flow per blowing converter is about 1,130



Mm3  (40,000 scfm) and contains from 3 to 4 percent SO-.  The



blister copper produced is transferred to one of two anode fur-



naces for refining and casting.  The slag skimmed from the con-



verters is returned to the reverberatory furnace.



     Off-gases from converter blowing operations are captured by



water-cooled primary hoods and pass through a series of multi-



clones and a settling flue for removal of coarse particulates



before they enter the gas-cleaning circuits of either a liquid



S02 plant or single contact sulfuric acid plant.  Off-gases from



the anode furnaces are treated for particulate removal in a plate



type electrostatic precipitator.



     Arsenic-laden dust recovered from the particulate control



devices servicing the multihearth roasters, the reverberatory



furnace, and the converters is transferred to the arsenic recov-



ery plant, where arsenic trioxide is produced.  The dust is



charged to one of the three Godfrey roasters, where the arsenic



trioxide contained in the dust is volatilized.  The roaster



residue or calcine produced is discharged to an open freight car



and subsequently recycled through the copper smelter to recover



the copper value.



     The arsenic trioxide-laden roaster gases are passed through



one of three settling kitchens, where the gases are cooled by



                               3-8

-------
radiative cooling and the arsenic trioxide is condensed.  The



kitchens are rectangular brick structures containing multiple



baffled chambers.  Two kitchens are equipped with 15 chambers and



one is equipped with 10 chambers.  The arsenic trioxide-laden gas



enters the first chamber at approximately 205°C (400°F) and exits



the final chamber of 100°C (212°F) or less.  The condensed



arsenic trioxide product (80 to 99 percent pure) is periodically



pulled from the kitchens and is either recycled through the



Godfrey roasters or placed in storage, depending on its purity.



The stored material is shipped as bulk product, either by rail or



packed in barrels.



     ASARCO-Tacoma also operates a metallic arsenic plant, which



produces elemental arsenic from purchased, refined arsenic triox-



ide.  The metallic arsenic plant has two identical furnaces in



which arsenic trioxide is volatilized, reduced to elemental



arsenic and condensed.



     Exhaust gases from the arsenic trioxide kitchens and arsenic



metal furnaces are treated in a baghouse before they are vented



through a flue to the smelter main stack.  The baghouse, which



was installed in 1982, is a MikroPul design consisting of five



compartments that contain 540 bags each.  The bags are made of a



Homopolymer acrylic and measure 11.4 cm  (4.5 inches) in diameter



and 3.0 meters (10 feet) in length.  The total baghouse filtering



area is 1,400 m2  (15,075 ft2).  The baghouse was designed to



treat 1700 m3/min (60,000 acfm) of gas at 93°C  (200°F) and to
                               3-9

-------
operate at an air-to-cloth ratio of 1.21 m3/min per m2 (3.98



cfm/fta).   Bag cleaning is performed by pulse jet.  The cleaning



mechanism is automatically activated when the pressure drop



across a compartment exceeds 5 inches of water.





3.2  PROCESS OPERATIONS



     During the performance of each of the emission tests con-



ducted throughout the test program, personnel from EPA or Pacific



Environmental Services, Inc. (an EPA contractor) closely moni-



tored the operating conditions of both the control devices and



process facilities tested and obtained and recorded pertinent



operating data.  The information obtained and the process and



control device conditions encountered during each of the various



test series conducted are discussed in this subsection.  Actual



records and process operating data obtained are tabulated in



Appendix F.



3.2.1  Reverberatory Furnace ESP



     Information recorded during the inorganic arsenic and par-



ticulate tests conducted at the outlet of the reverberatory



furnace precipitator (pipe and plate in series) included both



electrical operating and flue gas data.  Based on electrical data



included operating values for primary voltage, primary current,



spark rate, and secondary current for each of the 19 transform-



er/rectifier sets used to energize the precipitators.  Recorded



flue gas data included the temperature and flue draft at the



inlet to the pipe unit, the temperature and flue draft at the
                               3-10

-------
inlet of the plate unit, the plate unit outlet temperature, and



the opacity of the flue gas stream at the outlet of the plate



unit.  Values were obtained from meters and chart recorders



located in the Cottrell control room, the power house, and at the



pipe precipitator inlet and plate precipitator outlet.  All



readings were recorded hourly.  A review of the electrical oper-



ating and flue gas data obtained indicates that the precipitators



and gas conditioning systems were operating within normal limits



throughout the course of the test program.



     ASARCO also provided EPA with other pertinent information,



such as production data; values for secondary voltage measure-



ments performed by ASARCO prior to the testing on September 11,



1983, and after testing on September 29, 1983; results of daily



flue dust analyses for dust acid content; 24-hour particulate



losses from the pipe and plate precipitators as measured by



ASARCO's continuous sampler; and results of arsenic analyses on



the flue dust collected by the precipitators.



     Information on the production levels achieved during the



various test runs was obtained from production logs maintained in



the reverberatory furnace department.  A summary of the produc-



tion achieved during each sample run, in terms of loads of cal-



cine charged to the reverberatory furnace per hour and tons of



calcine charged per hour, is presented in Table 3-1.  Production



levels varied substantially during the AEOT sample runs, ranging



from zero to nearly eight loads/hour.  This variability in pro-



duction was due primarily to meteorologically influenced produc-



tion curtailments to prevent the possible violation of ambient
                              3-11

-------
       TABLE 3-1.  SUMMARY OF REVERBERATORY FURNACE PRODUCTION  DATA
Date
(1983)
9/19


9/20



9/21



9/26



9/27



9/28



9/17
9/23-24
9/24
9/25-26
Correspond-
ing ID
AEOT-1
AEOT-2
AEOT-3
AEOT-4
AEOT-5
AEOT-6
AEOT-7
AEOT-8
AEOT-9
AEOT-10
AEOT-11
AEOT-12
AEOT-13
AEOT-14
AEOT-15
AEOT-16
AEOT-17
AEOT-18
AEOT-19
AEOT-20
AEOT-21
AEOT-22
AEOT-23
AEOPA-1
AEOPA-2
AEOPA-3
AEOPA-4
Sampl ing
period (hours)
1015-1603 (5.80)
1635-2304 (6.48)
2347-0602 (6.25)
0948-1445 (4.95)
1530-2130 (6.00)
2120-0228 (5.13)
0312-0805 (4.88)
0948-1145 (4.95)
1544-2042 (4.97)
2119-0224 (5.08)
0310-0814 (5.07)
0940-1445 (5.08)
1537-2037 (5.00)
2115-0223 (5.13)
0315-0831 (5.27)
0947-1500 (5.22)
1534-2032 (4.97)
2115-0220 (5.08)
0306-0842 (5.60)
0945-1444 (4.98)
1534-2034 (5.00)
2115-0216 (5.02)
0302-0802 (5.00)
2021-2339 (3.28)
2230-0144 (3.23)
1951-2258 (3.12)
2222-0131 (3.15)
Calcine
loads6
2
3
0
0
3
11
8
1
4
22
15
13
39
39
26
3
8
40
17
1
16
38
32
26
27
19
25
Calcine .
loads, h
0.34
0.46
0
0
0.50
2.14
1.64
0.20
0.80
4.33
2.96
2.56
7.80
7.60
4.93
0.57
1.61
7.87
3.04
0.20
3.20
7.57
6.40
7.93
8.42
6.25
7.74
Calcine,
tons/h
2.17
2.92
0
0
3.15
13.51
10.33
1.27
5.07
27.28
18.63
16.12
49.14
47.89
31.08
3.62
10.14
49.61
19.12
1.26
20.16
47.69
40.32
49.94
52.67
38.36
50.02
 Loads of calcine charged to the reverberatory furnace  during  the  sample
 period.

3Loads charged during the sampling period  divided  by  the  duration  of  the
 sampling period in hours.

"Tons of calcine per hour calculated assuming  6.3  tons  of calcine  per load.
Note:   Process information supplied by  ASARCO,  Inc.,  and  U.S.  EPA.
                                    3-12

-------
air quality standards for sulfur dioxide.  The AEOPA sample runs



were performed only during periods when the reverberatory furnace



was charged at its nominal maximum production rate (six to eight



loads per hour).



     Process material and dust samples were also collected.



These included composite samples of the smelter feed (roaster



charge), reverberatory furnace charge  (calcine), reverberatory



furnace matte, and dust collected by the pipe and plate precipi-



tators.  The process material samples were collected to determine



if the arsenic input to the smelter and the furnace operation



were within normal operating limits during the course of the test



program.



     All samples were collected by ASARCO personnel and submitted



to PEDCo for arsenic analysis.  The arsenic analytical results



are summarized in Table 3-2.  The arsenic content of the roaster



charge material ranged from 2.9 to 4.2 percent by weight and



averaged 3.6 percent, which is well within the 3 to 5 percent



operating range for the ASARCO-Tacoma smelter.  Arsenic content



of the calcine charged to the reverberatory furnace ranged from



1.6 to 3.4 percent for the 11 work shifts for which samples were



available.  The arsenic content of the matte samples collected



ranged from 0.5 to 0.7 percent, which is representative of normal



reverberatory furnace operation.  Precipitator dust samples



averaged 55 percent arsenic.



3.2.2  Arsenic Plant Baghouse



     Data recorded during the testing of the arsenic plant bag-



house included the baghouse inlet temperature, the temperature






                              3-13

-------
           TABLE 3-2 PROCESS SAMPLE ANALYTICAL RESULTS,
                      REVERBERATORY FURNACE
Date
(1983)
9/19
9/19
9/19
9/20
9/21
9/21
9/21
9/21
9/26
9/26
9/26
9/26
9/27
9/27
9/27
9/27
9/28
9/28
9/28
9/28
9/28
Sampl
Calcine 7:00 a.
Calcine 3:00 p.
Roaster charge
Matte
Calcine 7:00 a.
Calcine 11:00 p
Roaster charge
Matte
Calcine 7:00 a.
Calcine 3:00 p.
Roaster charge
Matte
Calcine 7:00 a.
Calcine 11:00 p
Roaster charge
Matte
Calcine 7:00 a.
Calcine 3:00 p.
Calcine 11:00 p
Roaster charge
Matte
e description
m. - 3:00 p.m.
m. - 11:00 p.m.


m. - 3:00 p.m.
.m. - 7:00 a.m.


m. - 3:00 p.m.
m. - 11:00 p.m.


m. - 3:00 p.m.
.m. - 7:00 a.m.


m. - 3:00 p.m.
m. - 11:00 p.m.
.m. - 7:00 a.m.



shift
shift


shift
shift


shift
shift


shift
shift


shift
shift
shift


Arsenic,3 %
2.77
1.57
3.18
0.51
2.56
2.94
4.11
0.48
2.55
3.00
4.21
0.58
2.24
2.49
2.90
0.60
3.00
2.77
3.36
3.86
0.67
                    ESP Hopper Dust
9/20
9/21
9/28
9/29
Pipe and plate No.
Pipe and plate No.
Pipe and plate No.
Pipe and plate No.
3 silo
3 silo
3 silo
3 silo
54.8
54.7
54.7
58.0
aPercent arsenic (by weight)  determined by use of the  sample
 preparation and analytical  techniques described in  Proposed
 EPA Method 108.

Note:  All samples collected  and identified by ASARCO, Inc.
                                3-14

-------
and pressure drop across each of the five baghouse compartments,



and information on the process status of the arsenic trioxide and



arsenic metal plants.  Process status information was limited to



a record of the number of Godfrey roasters and metallic arsenic



furnaces that were on line at any time.  This information was



recorded at half-hour intervals.  Information on actual produc-



tion rates achieved during testing was not obtainable.



     During each of the seven test runs, only four of the five



baghouse compartments were operating.  The No. 3 compartment was



off line because of a malfunctioning outlet damper.  Plant per-



sonnel indicated that this condition has existed for several



months.  The effect of the inoperative compartment is a 20 per-



cent reduction in the baghouse filtering area.  On the average,



the baghouse treated about 1020 m3/min  (36,000 cfm) of gas at



25°C (77CF) during the tests and operated at an air-to-cloth



ratio of 0.91 m3/min per m2  (2.98 cfm/ft2).



     Process samples collected during each of the seven sample



runs conducted across the baghouse included composite samples of



the Godfrey roaster charge and collected baghouse dust.  Results



of PEDCo's arsenic analyses on these samples are presented in



Table 3-3.  Feed to the roasters consisted of a blend of mate-



rials that included flue dusts from the No.  1 and 2 electrostatic



precipitators, flue dusts from the multihearth roaster baghouse,



and rerun material from the arsenic kitchens.  As shown in Table



3-3, the arsenic content of the Godfrey roaster charge ranged



from 25 to 62 percent and averaged 33.5 percent over the course
                               3-15

-------
           TABLE 3-3 PROCESS SAMPLE ANALYTICAL RESULTS
                          ARSENIC PLANT
Date
(1983)
9/14
9/15
9/16
9/17
9/23
9/23
9/24
Sample description
Roaster charge
Baghouse dust
Roaster charge
Baghouse dust
Roaster charge
Baghouse dust
Roaster charge
Baghouse dust
Roaster charge 7:00 a.m. - 3:00 p.m.
Baghouse dust 7:00 a.m. - 3:00 p.m.
Roaster charge 3:00 p.m. - 11:00 p.m.
Baghouse dust 3:00 p.m. - 11:00 p.m.
Roaster charge 7:00 a.m. - 3:00 p.m.
Baghouse dust 7:00 a.m. - 3:00 p.m.
Arsenic,3 %
37.2
74.5
27.8
73.5
25.3
75.2
32.2
67.7
33.2
75.8
61.6
72.8
47.1
71.8
 Percent arsenic (by weight)  determined by use of the sample
 preparation and analytical  techniques described in Proposed
 EPA Method 108.

Note:   All  samples collected  and identified by ASARCO,  Inc.
                               3-16

-------
of the seven sample runs.  The arsenic content of the collected



baghouse dust ranged from 68 to 76 percent and averaged 73 per-



cent.  Two Godfrey roasters (Nos. 4 and 6) were operated during



each of the seven sample runs.  No arsenic metal furnaces were



operated during the first four sample runs, whereas two were



operated during the fifth and sixth runs and one  during the



seventh run.



     As previously discussed in subsection 2.3, a flow imbalance



was discovered across the baghouse during the three sample runs



conducted with both the arsenic trioxide and arsenic metal plants



in operation.  The measured outlet flow was about 30 percent



higher than the combined flow measured at the arsenic trioxide



plant and arsenic metal plant inlet locations.



     Investigations led to the conclusion that the flow imbalance



was a result of a malfunctioning damper located in the arsenic



metal plant bypass duct.  The bypass duct connects the arsenic



metal plant baghouse inlet duct to the arsenic trioxide plant



exhaust flue.  The damper, believed to be in a closed position by



ASARCO personnel, was discovered to be stuck in the open posi-



tion.  The effect was to divert a portion of the flow in the



arsenic trioxide plant flue through the bypass duct and into the



arsenic metal plant baghouse inlet duct.  Because the sampling



location for the arsenic trioxide plant was downstream of the



bypass duct entry and the arsenic metal plant sampling location



was upstream of the bypass duct exit, the flow diverted from the



arsenic trioxide plant flue could be measured.
                               3-17

-------
     In addition to the flow imbalance problem, the arsenic



concentration measured at the arsenic trioxide plant inlet test



location was substantially lower than that measured during the



first series of tests.  The arsenic concentration measured during



the second test series averaged 1.04 g/m3 (0.45 gr/dscf) versus



7.89 g/m3 (3.44 gr/scf) during the first test series.



     The EPA and ASARCO were unsuccessful in their attempts to



explain this disparity in arsenic loading from the arsenic triox-



ide plant.
                              3-18

-------
                            SECTION 4

             SAMPLING LOCATIONS AND TEST PROCEDURES


     This section describes the sampling sites and the test

methods used to characterize arsenic and particulate emissions

from the No. 2 reverberatory furnace and the arsenic plant.


4.1  REVERBERATORY FURNACE ESP OUTLET

     During the test series, PEDCo conducted of 33 tests at this

location to determine particulate and arsenic concentrations and

mass emission rates by using procedures described in EPA Methods

1 through 5 and 108.*  As described in Sections 2 and 3, most of

these tests were performed with multipoint traverse techniques

under varying reverberatory furnace process rates.  Six continu-

ous, single-point tests were also performed.

     The existing sampling ports located on top of the rectangu-

lar brick breeching connecting the ESP outlet with the main stack

were used in all the tests.  Figure 4-1 presents a schematic of

the sampling site and the approximate location of the ASARCO

sampler.  Because of the geometric configuration of the ductwork

and the location of the flow control dampers and corresponding
 40 CFR 60, Appendix A, Reference Methods 1 through 5, July
 1982.  Proposed Method 108 is in draft form.
                               4-1

-------
             ASARCO
             CONTINUOUS
             SAMPLER
             HOUSE
 TO MAIN  STACK
                     ^
                       2 m
                    (4 ft)
                           ««—-4.9 m
                              LEDGE
(16 ft) —
	FLOW
FLOW CONTROL
DAMPER
                      NO.  1
                      ESP
  LEDGE
STRUCTURAL
 SUPPORT-^
   \9 m  (3 ft)
                                CROSS  SECTION
                  ABCDE     FGHIJ
                  nnnnn    nnnnn
                         ^2.4 m (8 ft)
   'CONTINUOUS
    SAMPLER
    LOCATION

                                6.7 m (22  ft)-
-4.9 m (16 ft)
              Figure 4-1.   No.  1  ESP  outlet  test  location.
                                   4-2

-------
structural supports (see Figure 4-1), the sampling site did not



meet the minimum requirements specified in EPA Reference Method



1.*  This site, however, was the only one available for access to



the total cross-sectional area of the ESP outlet breeching.



     Ten 6.4 x 9.5 cm (2-1/2 x 3-3/4 in.) sampling ports were



located approximately 61 cm  (24 in.) off-center in the 6.7 m  (22



ft) wide rectangular brick breeching.  The depth of each sampling



port casing varied within a range of 38 to 43 cm (15 to 17 in.).



A nominal depth of 40.6 cm  (16 in.) was used for placing the



sampling points.



     Figure 4-2 presents a schematic of the flue cross section



and relative locations of sampling points for the traverse and



single-point trains.  An uneven deposition of material on the



bottom of the flue posed problems in establishing a cross-sec-



tional sampling area and sampling point matrix that would avoid



contamination of test samples.  Therefore, the sediment depth at



each sampling port was determined by first measuring the distance



from the top of the sampling port casing to the top layer of



sediment and then subtracting the nominal depth of the port



casing.  These values are recorded in parentheses for each sample



port in Figure 4-2.



     A velocity and temperature profile was established according



to the procedures described in EPA Method 2.*  Based on the



sediment depth information contained in Figure 4-2, a 150-point
 40 CFR 60, Appendix A, Reference Methods 1 and 2, July 1982





                              4-3

-------
        16"
     ' - 0"
                                       .?   ->.«'- 0" .?
                                           CONTIGUOUS
                                           "TTHOn IOT
                                               TB«|N
                                  •ft
                                                  06     o«
                  ofl

             StDIWNT
              DtPIH
Tuvtnt row







i
3
4
S
1
7
*
DISTANCE FROM TO*
SW»lt FWT - en
71 (78)
193 (76)
?W (100)
3IS (174)
37« (148)
437 (17?)
4
-------
traverse matrix was established in which sampling points were



located at 0.3-m (1-ft) increments in each sampling port.  Fifteen



of the 150 points could not be measured because of the sediment.



Table 4-1 summarizes pertinent data from the velocity and tempera-



ture profile.  Overall, the flow and temperature characteristics



were comparable and no excessive turbulent (nonaxial) flow pat-



terns were apparent.  This is evidenced by the minimum number of



zero and/or negative flow points.  An attempt was made to charac-



terize the degree of turbulent flow according to procedures



described in EPA Reference Method 2.  In this method, the face



openings of the Type-S pitot tube are aligned perpendicularly to



the duct cross-sectional plume, designated "0-degree reference."



Null  (zero) pitot readings obtained at 0-degree reference indi-



cate an acceptable flow condition at a given point.



     If the pitot reading is not zero at 0-degree reference, a



rotation of the pitot  (up to 90 degrees ± yaw angle) is made



until a null reading is obtained.  The value of the rotation



angle (yaw) is recorded for each point and averaged across the



duct.  Method 2 criteria stipulate that average angular rotations



greater than ±10 degrees indicate turbulent flow conditions in



the duct.  Because of the size of the pitot system  [>4.6 m  (15



ft)] used and the general insensitivity of the Type-S pitot tube



in measuring yaw and pitch angles, results from this charac-



terization were inconclusive.



     As shown in Table 4-1, only Port J showed an unusual flow



pattern characterized by a number of points exhibiting zero flow.
                                4-5

-------
       TABLE 4-1.   SUMMARY  OF  INITIAL VELOCITY AND TEMPERATURE
                            PROFILE DATA
Port
A
B
C
D
E
F
G
H
I
J
Average AP,a
in.H20
0.17
0.15
0.14
0.13
0.14
0.20
0.19
0.17
0.195
0.03
Average
tempera-
ture, °F
179
175
181
182
176
173
169
170
164
166
Average
. b
Average gas
velocity,
ft/s
25.2
23.7
23.1
20.9
22.3
27.1
26.6
23.2
26.6
9.0
22.8
Average
pressure,
in.H20
-1.2
-1.25
-1.25
-1.2
-1.2
-1.2
-1.2
-1.2
-1.2
-1.2
-1.2
 Average velocity  pressure.

DA dry molecular weight  of  29.00  and a moisture content of 4.0 percent
 were  used  for  calculation  purposes, based on historical test data.
                                 4-6

-------
Based on the depth of sediment at this point, this was as ex-



pected.



     Gas velocities generally were consistent with maximum veloc-



ities near the center of each sampling port.  A decrease in gas



velocity was noted at points near the bottom of the breeching,



which reflects the effect of the flow control damper structural



support (see Figure 4-1).  Figure 4-3 presents a typical velocity



profile for Sample Ports A, D, F, and I.  The highest velocity



measurements occurred in Sample Port F, which was the port used



for the continuous, single-point sampling train.  As expected,



volumetric flow rates as measured by the single-point train were



greater than those measured by the traverse sampling train.



     Based on the sediment and velocity profile data, a 9 x 8



sample matrix, excluding Port F, was established  (see Figure



4-2).  To avoid sample contamination, sampling points located



within 40.6 cm (16 in.)  of the top layer of sediment were not



sampled.  Incremental sampling areas established for each port



were based on the sediment depth data and the distance between



sampling port centerlines.  A cross-sectional sample area of 26.1



m2 (280.5 ft2) was calculated and used in all volumetric flow and



emission rate calculations involving multiplane traverse sampling



techniques.  For the continuous, single-point sampling train, a



cross-sectional sampling area of 24.4 m2 (262.5 ft2) was used in



calculations of mass flow rates.  The sampling area for the fixed



point train was determined by excluding the incremental area of



Port J 1.7 m2 (18 ft2) from the total area to account for the



zero flow condition established at this port.
                               4-7

-------
       TOP OF DUCT
   14
   15
  16
                      AD
     - o
     i-  O
                                      DA
                                     c>   AG
a
O
   8^                           o  .   A  D

                            B        ° A

§ 10h    D                  .           A
y

  111-                 •               o  A

  12

  13
                                                     • PORT A
A    o .                                   ° PORT D
                                           A PORT F
                                           DPORT I
    0        .10        .15        .20        .25        .30
                     VELOCITY HEAD (aP), 1n.  H?0


     Figure  4-3.   Simplified ESP  outlet  velocity  profile.
                                 4-8

-------
     As shown in Figure 4-2, 55 sampling points were used to



perform each isokinetic cross-sectional traverse.  Sample times



varied with the type of test being conducted.  For example, tests



to characterize particulate and arsenic emissions at full rever-



beratory furnace charge rates (designated AEOPA) usually were run



for 3 minutes per sampling point, or 165 minutes.  Typical times



for tests conducted in conjunction with the 23-hour continuous



sampling train (designed AEOT)  were 5.5 minutes per sampling



point, or 302.5 minutes.  With the exception of tests performed



on September 17, 19, and 20, 1983 (AEOPA-1 and AOET-1 through 3),



Sampling Port J was not sampled because of the zero flow charac-



teristics exhibited and the time required to move the sample



system around several obstructions,  including electrical wiring.



These four sample points were included on the field data sheets



as zero flow points.






4.2  ARSENIC PLANT



     Arsenic concentration and mass emission rates were deter-



mined at the inlet to and outlet of the baghouse controlling



emissions from the arsenic trioxide (As_0,) and metallic arsenic



processes.



     At the baghouse outlet, existing sample ports were used for



testing.  Two sampling ports, 90 degrees off-center, were located



approximately 6.5 duct diameters upstream and more than 9 duct



diameters downstream from the nearest flow disturbances in the



95-cm (37^-in.) i.d. round duct.  The cross-sectional area of the
                              4-9

-------
duct was traversed at 12 sampling points.  Tests AABO-1 through 4



were conducted for 6 hours, and Tests AABO-5 through 7 were



conducted for 3 hours.



     At the As_0, inlet, the two existing sampling ports, 90 de-



grees off-center, were approximately 1.5 duct diameters down-



stream and less than 0.5 duct diameter upstream from the nearest



flow disturbance in the 68-cm  (26-5/8 in.) i.d. round duct.  The



location of these sampling ports did not meet the minimum re-



quirements specified in EPA Method 1.  Because a nonaxial flow at



this location could cause a bias in test results, new sampling



ports were located downstream from the existing ports.  These



ports were 4.9 duct diameters downstream and 8.1 duct diameters



upstream from flow disturbances.  The cross-sectional area of the



duct was traversed at 32 sampling points.  Because of the heavy



dust load encountered at this location during Tests 1 through 4,



sampling times were adjusted to control the pressure drop across



the sample filter.  During Test 1, which lasted nearly 6 hours, a



high pressure drop caused two filter frit supports to rupture and



a portion of the dust normally captured on the filter was col-



lected in the first impinger.  This had no effect on the arsenic



measurement because the arsenic content of the first two imping-



ers is included in the total arsenic collected by the sampling



system.  A portion of the sample was lost when rupture disks in



the acid digestion bombs failed during analysis.  As noted in



Section 2, results from Test ABKI-1 are biased low and are con-



sidered to be nonrepresentative of emissions from the As-O, plant





                             4-10

-------
during this test.  To maintain simultaneous test conditions at


the inlet and outlet and to reduce the inlet sample loading, we


started both trains simultaneously.  The outlet train ran con-


stantly, whereas the inlet train was only operated for 30 minutes


each hour, starting on the hour.  This resulted in a cumulative


inlet sampling time of 3 hours for runs ABKI-2, 3, and 4.  During


the second set of tests (ABKI-5, 6, 7; ABMI-1, 2, 3; AABO-5, 6,


7), the sampling time for all tests at the kitchen inlet, the


metallic inlet, and the baghouse outlet was 3 hours.


     At the metallic plant exit duct, test ports were installed


in the 94-cm (37-in.) round duct so that they would be at least


2.5 duct diameters upstream and 6 duct diameters downstream from


flow disturbances.  The cross-sectional area of the duct was


traversed at 24 sampling points.  Figure 4-4 depicts the baghouse


outlet and As-O, inlet sampling locations, and Figure 4-5 pre-


sents the metallic plant inlet test location.



4.3  SAMPLING AND ANALYTICAL PROCEDURES


     The sampling and analytical procedures for this test program


followed those described,  where applicable, in EPA Reference


Methods 1 through 5* and proposed Method 108.*  The procedures


are briefly summarized here.  Detailed procedures are presented


in Appendix D.
*
 40 CFR 60, Appendix A, Reference Methods 1 through 5, July 1982,
 and proposed Method 108 is in draft form.
                              4-11

-------
                                         |<	—  9.1 m (30 ft)
                            TO MAIN
                             STACK
                                                                        FLOW
                                                                                    ASARCO
                                                                                    SAMPLER
                          > 6dd UPSTREAM
TRAVERSE
POINT NO.
      1
      2
      3
      4
      5
      6
 I
!—•
to
                          DISTANCE FROM OUTSIDE
                          OF NIPPLE ()n.)
                                                                          9 dd DOWNSTREAM
                                                         8AGHOUSE OUTLET
                                                                                                      CROSS-SECTION
                            6 1/8
                           10
                           15 1/2
                           30 3/4
                           36 1/4
                           40 1/8
                                           EXISTING  PORTS
                                                                                   95 cm (37 1/4  tn.) i.d.
                                                                              NIPPLE LENGTH: 11.4  cm (4 1/2  1n.)
TRAVERSE
POINT NO.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
DISTANCE
OF NIPPLE
10 1/2
10 3/4
11 3/4
12 7/8
14
15 3/8
17
19 1/2
26 1/8
Z8 5/8
30 1/4
31 5/8
32 3/4
33 7/8
34 7/8
35 1/8
FROM OUTSIDE
( In.)

















                                                       I  	  334 C|J|     |
                                                       l*Tl31  5/8 inTTA"
                                                                                                549 on
                                                                                               (216  in.)
                                                           FLOW	»-O
                                   30.5 cm (12 in.)
                                               ~f|UU 196.5 cm (38 in.)
                                                                      '
                                            FROM
                                             PLANT
                                                               As203 INLET
                                                                                    68 cm (26 5/8 tn.)t.d.
                                                                              NIPPLE LENGTH  :  23.75 on (9 1/2  tn.)
                        Figure 4-4.   Arsenic  plant  baghouse outlet  and  arsenic
                                       trioxide  inlet  test  locations.

-------
                               2.5m (8 ft)
5.6m (18 ft)
                         TO BAGHOUSE
                                                                           7
                                UPSTREAM
                                                 •~6  dd  DOWNSTREAM-
                                                                                  •FROM METALLIC
                                                                                      PLANT
u>
TRAVERSE
POINT NO.
1
2
3
4
5
6
7
8
9
10
11
12
DISTANCE FROM OUTSIDE
OF NIPPLE (in.)
7 1/4
8 3/4
10 5/8
12 3/4
15 1/2
19 3/8
30 1/8
34
36 3/4
38 7/8
40 3/4
42 1/2
                                                                          CROSS SECTION
                                                                       94 cm (37 in.) i.d.
                                                                 NIPPLE LENGTH: 15.9cm (6 1/4 in.)
                        Figure 4-5.  Metallic arsenic plant inlet test location.

-------
4.3.1  Velocity and Gas Temperature



     A Type-S pitot tube and an induced draft gauge manometer



were used to measure the gas velocity pressures at the test



sites.  Velocity pressures were measured at each sampling point



across the duct to determine an average value.  Measurements were



taken in accordance with procedures outlined in Reference Method



2 of the Federal Register.*  The temperature at each sampling



point was measured with a thermocouple and digital readout.



4.3.2  Molecular Weight



     Flue gas composition was determined in accordance with the



basic procedures described in Reference Method 3.*  Grab samples



were collected prior to the start of any sampling to establish



baseline contents of oxygen, carbon dioxide, and carbon monoxide.



Bag samples were collected during each test at the arsenic plant



and at least twice a day at the ESP outlet during the 23-hour



continuous testing.  Samples were analyzed using an Orsat gas



analyzer.  Based on the S0_ concentration as determined from the



Method 108 sample train, carbon dioxide results were corrected



for SO- since S0_ would be included in the Orsat-generated CO-



value causing a high bias in C0_ results.  The percentage of SO-



in each gas stream sampled typically measured less than 0.4



percent by volume.  The gas composition at each test site re-



mained consistent throughout the test series.
  40 CFR 60, Appendix A, Reference Methods 2 and 3, July 1982.
                              4-14

-------
4.3.3  Particulate/Arsenic


     Methods 5* and 108** were used to measure particulate and


arsenic concentrations.  All tests were conducted isokinetically


by regulating the sample flow rate relative to the gas velocity


in the duct  (as measured by the pitot tube and thermocouple


attached to the sample probe).  The basic sampling train con-


sisted of a heated glass- or stainless steel-lined probe, a


heated 7.6-cm (3-in.)  diameter glass fiber (Whatman Reeve Angel


934AH) filter, a heated Teflon connector, and a series of six


Greenburg-Smith impingers followed by a vacuum line, vacuum


gauge, leak-free vacuum pump, dry gas meter,  thermometers, and a


calibrated orifice.


     For determination of particulate and arsenic concentrations,


the nozzle, probe, and filter holder portions were rinsed with


acetone at the end of each applicable test.  Upon completion of


the acetone rinse, an additional rinse with 0.1 N NaOH was per-


formed.  The acetone rinse and particulates caught on the filter


media were dried at room temperature, desiccated to a constant


weight, and weighed on an analytical balance.  Total filterable


particulate matter was determined by adding these two values.


     Upon completion of the gravimetric analysis, the filter,


acetone rinse, and solids contained in the 0.1 N NaOH rinse of


the front half of the sampling train were prepped, combined, and
  40 CFR 60, Appendix A, Reference Method 5, July 1982.
**
  Method 108 is a proposed method.
                               4-15

-------
analyzed for arsenic (by atomic absorption).   The only differ-



ences between the recovery and analysis of particulate/arsenic



and arsenic only are the acetone rinse and gravimetric weighing.



Tests run to determine arsenic only were rinsed with 0.1 N NaOH,



and the filter and NaOH rinse were subsequently prepped and



analyzed by atomic adsorption.  Figure 4-6 summarizes the sample



recovery and analytical procedures used.



     The volume of water collected in the impinger section of the



sampling train was measured at the end of each sampling run to



determine the moisture content of the flue gas.  The contents of



the first two impingers were transferred to a polyethylene con-



tainer.  The impingers, Teflon connector, and all connecting



glassware (including the back half of the filter holder) were



rinsed with 0.1 N NaOH, and the rinse was added to the container.



The contents of the third, fourth, and fifth impingers were



transferred to a polyethylene container.  The impingers and all



connecting glassware were rinsed with distilled water and added



to the container.  The contents of the first two impingers and



0.1 N NaOH rinse also were analyzed for arsenic by atomic absorp-



tion.  The contents of the third, fourth, and fifth impingers and



distilled water rinse were titrated on site with NaOH to deter-



mine SO» concentrations.
                             4-16

-------
                            METHOD 5-108 - PARTICULATE/ARSEN1C

PROE
1
1
FRONT HALF
E, FRONT FILTER GLASSWARE, FILTER


1 ACETONE RINSE i FILTER
1 , 1

1 EVAPORATE,
1 DESICCATE,
j AND WEIGH
1



, r- •
I 1 DESICCATE
| 1 AND WEIGH
i 1




DIGEST IN 0.1
NaOH FOLLOWED
BY CONC. HNO,



1
BACK HALF
BACK FILTER GLASSWARE, SAMPLE HEAD AND HOSE, 1MPINGERS
1
0.1 N NaOH RINSE


FILTER
1
FII

N
1
FILTER SOLUTION
1
1 FILTER




DIGEST IN HNO,
AND HF J

FILT

BOIL
EVAP
TO D
1
I AA
ADD
AND
IMP1NGERS 1 AND 2
AND 0.1 N NaOH RINSE
IMPINGERS 3-5
AND H20 RINSE

SOLUTION


TER

RATE



AND
ORATE
RYNESS

HNO,
DILUTE

FILTRATE
1
DILUTE TO 200 ml
1
TAKE ALIQUOT
1
1
DIGEST USING
HN03
[ AA ]
DILUTE TO 500 ml
DILUTE TO 1 liter

TAKE ALIQUOT
1
ADD CONC. HN03
I
BOIL AND EVAPORATE
TO DRYNESS

ADD 50% HNO,
AND DILUTEJ
1
AA

TAKE ALIQUOT

TITRATE WITH 1 N NaOH
runt r ATCC c-rrnc uni
                                                                                  INCLUDED IN METHOD  108
                                                                                  (ARSENIC ONLY) RECOVERY
                                                                                  AND ANALYTICAL PROCEDURES
Figure  4-6.   Sample recovery  and analysis flow chart  for particulate/arsenic.

-------
                            SECTION 5



                        QUALITY ASSURANCE






     Because the desired end product of testing is representative



emission results, quality assurance is one of the main facets of



stack sampling.  Quality assurance guidelines provide the de-



tailed procedures and actions necessary for defining and pro-



ducing acceptable data.  Four such documents were used in this



test program to ensure the collection of acceptable data and to



provide a definition of unacceptable data.  The following docu-



ments comprise the source-specific test plan prepared by PEDCo



and reviewed by the Emission Measurement Branch:  the Quality



Assurance Handbook Volume III, EPA-600/4-77-027; the PEDCo Envi-



ronmental Emission Test Quality Assurance Plan; and the PEDCo



Environmental Laboratory Quality Assurance Plan.  The last two,



which are PEDCo's general guideline manuals, define the company's



standard operating procedures and are followed by the emission



testing groups and the laboratory groups.



     Relative to this specific test program, the following steps



were taken to ensure that the testing and analytical procedures



produced quality data.



     0    Calibration of all field sampling equipment.



     0    Checks on train configuration and calculations.
                               5-1

-------
     0    Onsite quality assurance checks (i.e., sampling train,
          pitot tube, and Orsat line leak checks)  and quality
          assurance checks of all test equipment prior to use.

     0    Use of designated analytical equipment and sampling
          reagents.

     0    Internal and external audits to ensure accuracy in
          sampling and analysis.

     Table 5-1 lists the sampling equipment used to perform the

particulate and arsenic tests as well as the calibration guide-

lines and limits.  In addition to the pre- and post-test cali-

brations, a field audit was performed on the metering systems and

thermocouple digital indicators used for the particulate and

arsenic sample runs.  PEDCo-constructed critical orifices were

used for dry gas meter audits.  Figures 5-1 through 5-8 show

example audit runs for each dry gas meter used for particulate

and arsenic tests.  Figures 5-9 through 5-12 show example audit

results for the digital temperature indicators used during test-

ing.

     Emission rate calculations were performed on site by EPA and

PEDCo personnel.  Calculations were performed independently for

data validation purposes.  Computer programming was used to

recheck and validate the data at the end of the test program.

The minor discrepancies that exist between the hand calculations

and computer printouts are due primarily to round-off error.

Figure 5-13 presents an example calculation form used by the EPA

during this test program.  Computerized example calculations are

presented in Appendix A.
                                5-2

-------
                                      TABLE  5-1.   FIELD EQUIPMENT CALIBRATION
Equipment
Meter box




















I.D.
No.
FB-1


FB-2


FB-3


FB-4


FB-5


FB-9


FB-10


Calibrated
against
Uet test meter




















Allowable
deviation
Y +0.02 Y
AH @ +0.15
(Y +0.15 Y post-test)


















Actual
deviation
+0.004
-0.09
+0.012
-0.012
-0.08
+0.005
+0.016
+0.05
-0.003
-0.005
-0.10
+0.042
-0.002
+0.05
+0.009
-0.006
-0.09
-0.003
+0.003
-0.09
+0.007
Within
allowable
1 imits
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Comments





















u>
      (continued)

-------
        TABLE 5-1 (continued)
Equipment
Meter box
(continued)

Pi tot tube





Digital indi-
cator



Thermocouple






Orsat analyzer
I.D.
No.
FB-11


026
028
032
187
284
387
125
126
207
220
262
130
138
175
178
202
203
204
141
Calibrated
against
Wet test meter


Standard pi tot
tube




Mil 1 ivol t signal s




ASTM-2F






Standard gas
Allowable
deviation
Y +0.02 Y
AH @ +0.15
(Y +0.15 Y post-test)
Cp +0.01





+0.5%




+1.5%
(+2% saturated)





+0.5%
Actual
deviation
+0.007
-0.03
-0.004
_
-
-
-
-
-
+0.77%
-0.50%
+0 . 30%
-0.50%
+0.38%
+0.61%
+0 . 20%
+0.22%
+0.47%
+0.47%
-0.37%
+0.47%
-0.3%
Within
allowable
1 imits
X
X
X
X
X
X
X
X
X
No
X
X
X
X
X
X
X
X
X
X
X
X
Comments



Visually inspected
on site

















en
I
        (continued)

-------
TABLE 5-1 (continued)
Equipment
Impinger
thermometer


Trip balance
Barometer

Dry gas
thermometer














I.D.
No.
288
290
382
384
195
229

FB-1
FB-1
FB-2
FB-2
FB-3
FB-3
FB-4
FB-4
FB-5
FB-5
FB-9
FB-9
FB-lU
FB-10
FB-11
FB-11
Cal ibrated
against
ASTM-2F



Type S weights
NBS traceable
barometer
ASTM-2F















Allowable
deviation
+2°F



+0.5 g
+0.10 in.Hq
(0.20 in.Hg post-test)
+5°F















Actual
deviation
0°F
0°F
0°F
+ 1°F
0.0 g
-0.01 in.Hg
(-0.11 in.Hg
-4.0°F
-5.0°F
-3.5°F
-2.0°F
-3.0°F
-2.7°F
-3.0°F
+1.3°F
+4.0°F
-2.5°F
+2.0°F
-2.0°F
-2.0°F
-2.0°F
-2.0°F
+2.0°F
Within
allowable
1 imits
X
X
X
X
X
X
> x
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Comments







Inlet
Outlet
Inlet
Outlet
Inlet
Outlet
Inlet
Outlet
Inlet
Outlet
Inlet
Outlet
Inlet
Outlet
Inlet
Outlet
(continued)

-------
          TABLE 5-1  (continued)
Equipment
Probe nozzle





I.D.
No.
3-108
2-106
Setup 1
Setup 2
5-104
5-103
Calibrated
against
Cal ioer





Allowable
deviation
Dn + 0.004 in.





Actual
deviation
0.002 in.
0.000 in.
0.001 in.
0.003 in.
0.002 in.
0.001 in.
Within
allowable
1 imits
X
X
X
X
X
X
Comments






t_n
 I

-------
                        AUDIT  REPORT DRY GAS METER
                       	   CLIENT:
DATE:   7-/1 -fr ^	          	
BAROMETRIC PRESSURE (Pbar):  "3Q.l^in.  Hg  METER BOX NO.     /=~/5 -  /
ORIFICE NO.  ft  \	   PRETEST  Y:      H. 1C,S
ORIFICE K FACTOR:     M.kTJyiQ"11        AUDITOR:
Orifice
manometer
reading
AH
in H20

,CO
Dry gas
meter
reading
Vvf
ft3
Hi .000
I5?^oo
Temperatures
Ambient
Tai/Taf
°F
no
-?0
Dry gas meter
Inlet
w
°F
1^
"75
Outlet
vv
°F
(*%
6-fc
Duration
of
run
0
min
1^ Q2-
13 -^o

Dry gas
meter
volume
Vm
ft3
//.3>rc~
Average temperatures
Ambient
Ta
°F
•?r
Dry gas
meter
m
°F
^/.T.5
mstd
ft3
/f.3^
\ci
ft3
//. 247
Audit
Y
rW%
Y
deviation
%
3trf .^
\td
(17.647)( Vm )(Pbar + AH/13.6)
(Tm + 460)
//3fi

Audit y
Vm .
act
V["std
aWZ-
Xct
(1203)( 0 )( K )(P )
x, . -,««, /2
a
//.I?*?

Y deviation, %
(Y audit - Y pre-test)(100£)

(Y audit)
3#t-%
Audit Y niust be in the range,  pre-test  Y ^0.05
                  Figure 5-1.  Audit report dry gas meter (FB-1)

                                     5-7

-------
                        AUDIT REPORT DRY GAS METER
DATE:    9/22* //3	  CLIENT: 	
BAROMETRIC PRESSURE (Pbjr):£?.%5  in. Hg   METER BOX NO.
ORIFICE NO.   |Z-	^        PRETEST  Y: _
ORIFICE K FACTOR:    f
AUDITOR:    PlJI
Orifice
manometer
reading
AH
in H20
U
Dry gas
meter
reading
Vvf
•a
ft3
I33.-/01/
/-o,:tte
Temperatures
Ambient
Tai/Taf

°F
.5*
38
Dry gas meter
Inlet
w
°F
^
(oO
Outlet
Toi/Tof
CF
-51
bl
Duration
of
run
0
min

/5.0

Dry gas
meter
volume
Vm
ft3
\l&\\
Average temperatures
Ambient

Ta
°F
5$
Dry gas
meter
Tm
°F
>5?

v_
mstd

ft3
//.J?7//

v_
mact

ft3
//.3ML

Audit

•y

.Itf


Y
deviation
%
•"iiu?
Vmstd
(17.647)( Vm )(Pbar + AH/13.6)
(Tm + 460)
//•«7^

Audit y
V%ct
mstd
• ?55
\ct
(1203)( 0 )( K )(Pbar)
(Ta + 460)1/2
G
//.33?

Y deviation, %
(y audit - Y pre-test)(100%)
(y audit)
-3.c*
Audit Y must be in the range, pre-test Y ±0.05
                  Figure 5-2.  Audit report dry gas meter (FB-2),
                                      5-8

-------
                        AUDIT REPORT DRY GAS METER
                        	   CLIENT:
DATE: 	          	
BAROMETRIC PRESSURE (P^^e.^3  in.  Hg   METER BOX  NO.
ORIFICE NO.       /2>	   PRETEST Y:      	
ORIFICE K FACTOR:     V. 7^V/l> "*'         AUDITOR:     ^L/^
Orifice
manometer
reading
AH
in H20
r."
Dry gas
meter
reading
v./vf
ft3
y/#.^o
Y?9 '-3c
Temperatures
Ambient
w
°F
60
bt
Dry gas meter
Inlet
°F
62
£,9
Outlet
To1/Tof
°F
•3 1
&c
Duration
of
run
9
min
ir.s

Dry gas
meter
volume
ft3
//,/
Average temperatures
Ambient
Ta
°F
fpD
Dry gas
meter
Tm
°F
£V
mstd
ft3
//.tt-tT
\ct
ft3
H.601
Audit
l.trt
Y
deviation
-3 (,^'^J


Vmstd
(17.647)( Vm )(Pbar + AH/13.6)
(Tm + 460)


Audit Y
Vm t
mstd
f.OOW
\ct
(1203)( 0 )( K )(P. )
i I n . , Ha ' ..
(T. + 460)1/2
a


Y deviation, %
(Y audit - Y pre-test)(100%)
(Y audit)

Audit Y must be in the range, pre-test Y ±0.05 Y

               Figure 5-3.  Audit report dry gas meter (FB-3)
                                      5-9

-------
                        AUDIT REPORT DRY GAS METER
DATE:    ?Ac/l3	   CLIENT:   |
BAROMETRIC PRESSURE (Pbjr);^o./5 1n.  Hg    METER BOX NO.   F,B - *f
ORIFICE NO.  3	    .       PRETEST  y:   /,03O
ORIFICE K FACTOR:  ^.S^A/O~^           AUDITOR:
Orifice
manometer
reading
AH
in H20
*.GO
Dry gas
meter
reading
Vvf
ft3
&VJ.IIO
a
'6-56"
Temperatures
Ambient
VTaf
°F
6f
6f
Dry gas meter
Inlet
Tii/Tif
°F
7o>
7G,
Outlet
VTof
°F
6?
6,?
Duration
of
run
ID
min
/s.O

Dry gas
meter
volume
Vm
ft3
JZ.455
Average temperatures
Ambient
Ta
°F
^
Dry gas
meter
Tm
°F
~73
V[T1std
ft3
U-5'Z-
Set
ft3
M.m
Audit
Y
/ 053
y
deviation
%
2.L
Xtd
(17.647K Vm )(Pbar + AH/13.6)
(Tm + 460)
)<.SI^-
Xct
(1203)( 0 )( K )(Pbar)
(T + 460)'^
a

/s.m

Audit Y
"m .
act
mstd
1-053
Y deviation, %
(y audit - y pre-test)(l
00?)
(Y audit)
+ 2.7L
Audit Y must be in the range, pre-test Y ±0.05
                 Figure  5-4.  Audit report dry gas meter (FB-4)
                                     5-10

-------
DATE:
AUDIT REPORT DRY  GAS  METER
                   CLIENT:
.  r.PA &
BAROMETRIC
ORIFICE NO.
PRESSURE (P
U
hn r ' ' -'*" ' ' -

in. Hg METER BOX NO. &
PRETEST Y:
ORIFICE K FACTOR: Z.HlJ- "<"" ^ AUDITOR: .'

Orifice
manometer
reading
AH
in H20
,/'- -'-•

Dry gas
meter
reading
Vvf
.:,•:••/ .:'; v.
3 ?;,:-. J-ev.-

A-7^~>
'K Ine

Temperatures
Ambient
Tai/Taf
°F
b$
t£>
Dry gas meter
Inlet
f,:T
f;V
Outlet
W
°F
&(.->
&;
Duration
of
run
min
x=r-/

Dry gas
meter
volume
Vm
ft3
Xj$,.T'

-------
                        AUDIT REPORT DRY GAS METER
                                           CLIENT:     U5EPA-
DATE: 	
BAROMETRIC PRESSURE (PhaJ: ID-IS  in.  Hg   METER BOX NO.
ORIFICE NO.
                 //
                      bar
ORIFICE K FACTOR:
                            VlO
                                '^
PRETEST  Y:
AUDITOR:
Orifice
manometer
reading
AH
in HpO
,5.15
Dry gas
meter
reading
Vvf
ft3
~7iO,fOO
7^.-ioo
Temperatures
Ambient
°F
•70
70
Dry gas meter
Inlet
°F
7S
7rf
Outlet
Toi/Tof
°F
"7o
-70
Duration
of
run
min
j£ ^C^'
x^O

Dry gas
meter
volume
Vm
ft3
/J.*/c-
Average temperatures
Ambient
Ta
°F
?^
Dry gas
meter
m
°F
?
-------
                        AUDIT  REPORT DRY GAS METER
                        	  CLIENT:
DATE:    9'	          	
BAROMETRIC PRESSURE (Pbar): 3O-/5 in.  Hg   METER  BOX NO.
ORIFICE NO.  -ff  JQ	   PRETEST Y: _
ORIFICE K FACTOR:     V-7/3X/C'^         AUDITOR:
Orifice
manometer
reading
AH
in H20
1. 10
Dry gas
meter
reading
Vvf
ft3
5^1-lco
5^0.500
Temperatures
Ambient
Tai/Taf
°F
"70
10
Dry gas meter
Inlet
T../T.r
11 if
°F
71
18"
Outlet
Toi/Tof
°F
~7^
~)^
Duration
of
run
min
IS^o

Dry gas
meter
volume
Vm
ft3
/J.^or.
Average temperatures
Ambient
Ta
°F
3o
Dry gas
meter
m
°F
95.0
Vfnstd
ft3
/P.5*/?
Xct
/3-2V6
Audit
Y
P/?'-^
Y
deviation
0%?-
V
mstd
(17.647)( Vm )(Pbar + AH/13.6)
(Tm + 460)
/S. &*3
-------
                        AUDIT REPORT DRY  GAS METER
                                           CLIENT:
DATE: 	           	
BAROMETRIC PRESSURE (pbjr):  51/5 in.  Hg   METER BOX NO.
ORIFICE NO. -3?/2~	       PRETEST  Y:  _
ORIFICE K FACTOR:    */.£'?*///&'       AUDITOR:
                                                                 //
Orifice
manometer
reading
AH
in H20
//5
Dry gas
meter
reading
Vvf
ft3
tffo, 500
rr?5,ffoo
Temperatures
Ambient
Tai/Taf
°F
9^
7^
Dry gas meter
Inlet
^5
~7k
Outlet
Toi/Tof
°F
£g
11
Duration
of
run
0
min
1 £\ ~)'.\ 1& 	
J O.-oc* 1 oL"

Dry gas
meter
volume
Vm
ft3
/J,2>rr
Average temperatures
Ambient
Ta
°F
9-r>
Dry gas
meter
m
°F
17.6
V|T1std
ft3
X?,37Z-
ft3
//*?£
Audit
Y
&.
-------
                THERMOCOUPLE DIGITAL  INDICATOR
                        AUDIT DATA SHEET
Date
Indicator  ho. fr
                                                 Operator
Test Point
No.
1
2
3
4
Millivolt
signal*




Equivalent
temperature,
op*
^ ~
ado
5MO
1 n^l '
Digital Indicator
temperature reading,
•F
3\M
\^6^
531.6,
mo .0
Difference,
I
&.IZ-
o. n
P. 7-4
C.?.*f
Percent difference must be less than or equal  to  0.54.

Percent difference:

  (Equivalent  temperature °R- Digital Indicator  temperature reading °R)(100°;)
                             (Equivalent temperature  BR)
Where *R «= °F 4  460°F
  These values are to be obtained from the calibration data sheet for the
  calibration device.
           Figure 5-9.  Thermocouple  digital indicator audit  data sheet
                               (Indicator 125).
                                    5-15

-------
                THERMOCOUPLE DIGITAL INDICATOR
                        AUDIT DATA SHEET
Date   9/q/g.3     Indicator No.
                                                 Operator
Test Point
No.
1
2
3
4
Millivolt
signal*




Equivalent
temperature,
°F*
— s-.
^ • >*^
,-ii!-V-'
»"^~V -
// V'~"
Digital Indicator
temperature reading,
°F
So
/'-/'/
£z£
//&7
Difference,
X
Grf
O,l£
L>. .T
t>.+
Percent difference must be less than or equal to 0.51.
Percent difference:
  (Equivalent  temperature BR - Digital  Indicator temperature reading °R)(10D%)
                             (Equivalent  temperature *R)
Where *R «  eF  + 460°F
  These values are to be obtained from the calibration data sheet for the
  calibration device.
           Figure 5-10.  Thermocouple digital  indicator audit data  sheet
                               (Indicator 126).
                                     5-16

-------
                THERMOCOUPLE DIGITAL  INDICATOR
                        AUDIT DATA SHEET
Date '•/ -
Indicator No.
                                                 Operator
Test Point
No.
1
2
3
4
Millivolt
signal*




Equivalent
temperature,
OF*
~2?-s
32-00
£<-jo
./'W
Digital Indicator
temperature reading.
•F
^y
.:?£"/
^r«-/c?
//fz.
Difference,
X
- ^^ y^
0*1
O.oo
&1Z-
Percent difference must be less than or equal  to 0.5JL

Percent difference:

  (Equivalent  temperature °R • Digital Indicator temperature reading °R)(1DO%)
                             (Equivalent temperature   R)
Where °R « °F + 460°F
  These values  are to be obtained from the calibration data sheet for the
  calibration device.
          Figure 5-11.  Thermocouple  digital indicator audit data sheet
                              (Indicator 207).
                                      5-17

-------
                THERMOCOUPLE DIGITAL  INDICATOR
                        AUDIT DATA SHEET
Date
Indicator  ho.
Operator
o$
Test Point
No.
1
2
3
4
Millivolt
signal*




Equivalent
temperature,
op*
3^
2oO
5^d -O.Z.
--^^-0.5
* *0.l
«^ -.^
Percent difference must be less than or equal  to  0.5%.

Percent difference:

  (Equivalent  temperature °R -Digital  Indicator  temperature reading °R)(100%)
                             (Equivalent temperature  CR)

Where *R *  *F  + 460°F
  These values are to be obtained from the  calibration data sheet for the
  calibration device.
             Figure 5-12.  Thermocouple  digital indicator audit data  sheet
                                 (Indicator 262).
                                    5-18

-------
Ui
 I
vo
                      PLANT.
                            SAMPLING SUMMARY SHEET

                                                   LOCATION   TACOHA. WASHINGTON
SAMPLED SOURCE
tun Date
ifOT-l
AloT 1
nrwi



totems

REVERBERATOR? FURNACE ESP OUTLET TRAVERSING RUNS
^.^ vv ___ ^
© © © © © \?/ Cv © 17 © Ol' © 21 22 23
N * All P Bar. V T V Std . * P . P T VP V V ZH M.
p m n m m tt » t w w d
«<»
55
IT1)

.9*9
.9*5
.3*5-

1.20
1-3*
I.JO

30.4)
JO. 34
30.3fc

.9*. 0.8
3o».«77
HV107

@ (25) 26 1^7) (tBi 29 30
Ron CO, // 0, N, CO SO, HW^ HW
65
fc«
fc^

199. lOfl
io> «*o
197.934

,047;
.0*24
.0/97

® ® ©
C_-J«''.tT. * '•60)' T
-M
- I.J
- 1.7

3o, ^
30.15
30.27

n i
IB*
109

^^-^
^^^
^^

l».7
,,,.,
105.7

f-«
5. ft/
4.99

yj
2.7
2.5

.9tt
•fa
.175

(5ft) 35 (36) »3?i 38 ^9/ «0
DVD Are* ACFII DSCFM XI
Koi-l
Mm
/ron

0.99
0.94
l.lg

\9.1
1*,*
/9.0

79.B
7».»
7».«


0.0
o.o

.0*
.06
.02

a».->4-
a. .94
28.96

a».s9
JL8.*5
28.69

,94
.*¥
•ff<

'0-44.01
10- 602B
9. 3oo3

294
??9
390

••3"?
.3«9
.112.

a, .$
3t. 1
24-1

ixx
/s
//

19o. 5
309.5
a»c.?

42P44fl
439583
406035

3f/33T
3« 792
32i m, «cr>
Bl » fn CM* Ntlitm I
                                                                  11.1     •
                                                                        * Ml. • f>l«v •• »TT t»« 41
                                                                                m. ner*
                                                                                                                                    f. • rr 0.. Ml.
                                                                            IIM* ••• rn
                                                                                 flM* TMfiriKr*! *f
                                               !«! • f>|»> I ITT
                                               1C* • T>|M
                                                                                                    ». • f
                                                                                                               N»III<|MI
                                                                                               It fMl f*r •<•
                                                                              it • r» «~i IH*|MII«
                                                                                       i a*i> '«•• » 
-------
     Tables 5-2 and 5-3 summarize results obtained with duplicate



train tests performed at the ESP outlet test location.  These



data were collected to assess the precision and accuracy of EPA



Method 108 as applied to this particular source.  We conducted



seven paired-train tests in a single sampling port and at essen-



tially a single point in the outlet breeching.  The duration of



each duplicate test ranged from 1 to 2 hours.  Collected samples



were recovered and shipped to our Cincinnati laboratory for



arsenic analysis.



     The distribution of arsenic in the sample train  [front



(probe and filters) versus back (impingers)] is variable and



probably a function of the chemical properties of arsenic triox-



ide  (As-O.,) relative to the sample temperature.  As shown in



Table 5-2, the relative standard deviation for total arsenic of



all paired-runs averaged 12.5 percent, which is considered ac-



ceptable based on the complexity of the sample preparation and



analytical techniques required and the overall configuration of



the sample location (see Section 4).



     As an additional check of the reliability of the method used



to analyze the samples, a blank train was assembled and tran-



sported to the test site.  The sampling train was capped off and



remained at the test site for about 2 hours.  It was then re-



turned to the recovery area, where it was recovered in the same



manner as the test samples were recovered.  The blank train



samples were shipped to the laboratory and analyzed according to
                               5-20

-------
                                    TABLE 5-2.  RESULTS OF DUPLICATE ARSENIC TESTS AT ASARCO
en
 I
AEOD
Run No.
1A
IB
2A
2B
3A
3B
4A
4B
5A
5B
6A
6B
7A
7B
Sample
volume,
dNm3
1.371
1.360
2.332
2.251
1.294
1.250
1.286
1.248
1.262
1.205
1.310
1.271
1.130
1.088
Sample weights, mg
Filterb
3.88
4.91
1.68
1.73
4.74
10.15
3.82
4.12
3.91
4.40
4.44
3.74
2.53
5.62
Probe
rinse
2.63
3.06
2.15
1.05
5.81
5.97
3.74
3.11
4.34
4.85
3.53
2.77
1.70
2.06
Total
front
half
6.51
7.97
3.83
2.78
10.55
16.12
7.56
7.23
8.25
9.25
7.97
6.51
4.23
7.68
Back
half
4.93
2.51
3.94
3.58
9.42
10.3.6
7.65
8.33
8.36
5.20
2.96
6.60
5.52
5.32
Total
11.44
10.48
7.77
6.36
19.97
26.48
15.21
15.56
16.61
14.45
10.93
13.11
9,75
13.00
Concentration, mg/dNm3
Front
half
4.75
5.86
1.64
1.24
8.15
12.89
5.88
5.79
6.54
7.68
6.09
5.12
3.74
7.06
Back
half
3.60
1.84
1.69
1.59
7.28
8.29
5.95
6.68
6.63
4.32
2.26
5.19
4.88
4.89
Total
8.34
7.70
3.33
2.82
15.43
21.18
11.83
12.47
13.17
11.99
8.35
10.31
8.63
11.95
Iso-
kinetic
rate, %
108
108
106
103
101
100
101
98
102
99
102
99
103
100
                  'includes volume of S02, which represented less than 0.3 percent of the sample volume.
                  'includes PARR bomb  results, which constituted approximately 5 to 20 percent of the  total
                  arsenic on  the  filter.

-------
                               TABLE 5-3.  STATISTICAL DATA FOR GROUPED ASRSENIC RUNS AT ASARCO
AEOD
Run No.
1A
IB
2A
2B
3A
3B
4A
4B
5A
5B
6A
6B
7A
7B
Overall
means
Front-half arsenic
Concentra-
tion, mg/dNm3
Run
value
4.75
5.86
1.64
1.24
8.15
12.89
5.88
5.79
6.54
7.68
6.09
5.12
3.74
7.06

Paired
mean
5.31
1.44
10.53
5.84
7.11
5.60
5.40
5.89
a
a,
mg/dNm3
0.78
0.28
3.36
0.06
0.81
0.69
2.35
1.19
RSD,b
%
14.7
19.4
31.9
1.0
11.4
12.3
43.5
19.2
Back-half arsenic
Concentra-
tion, mg/dNm3
Run
value
3.60
1.84
1.69
1.59
7.28
8.28
5.95
6.68
6.63
4.32
2.26
5.19
4.88
4.89

Pai red
mean
2.72
1.64
7.78
6.32
5.48
3.72
4.88
4.65
o,
mg/dNm3
1.24
0.07
0.71
0.52
1.63
2.07
0.01
0.89
RSD,
%
45.6
4.3
9.1
8.2
29.7
55.6
0.2
21.8
Total arsenic
Concentra-
tion, mg/dNm3
Run
value
8.34
7.70
3.33
2.82
15.43
21.17
11.83
12.47
13.17
11.99
8.35
10.31
8.63
11.95

Paired
mean
8.02
3.08
18.30
13.15
12.58
9.33
10.29
10.68
o,
mg/dNm3
0.45
0.36
4.06
0.45
0.83
1.39
2.35
1.41
RSD,
%
5.6
11.7
22.2
3.7
6.6
14.9
22.8
12.5
 I
NJ
to
        Standard deviation with N-l weighting for sample data.


        Relative standard deviation is the standard deviation expressed as a percent of the mean concentration.
        cc,-
         Simple averages of tabulated data.

-------
the same procedures used for the actual emission samples.  Table



5-4 presents the results of the blank sample train analysis.



These results are considered reasonable and indicate background



arsenic contamination was not a problem in the sample recovery



area.
               TABLE 5-4.  BLANK TRAIN ANALYTICAL RESULTS

lab No.
DC577
DC578
DC579
Blank train
component
Filter
Rinse
Impinger
Total
mg As
0.55
0.04
0.002

Volume, ml
_
229
500
     The following is a summary of the quality assurance activi-



ties performed during the analytical phase of this project.



Twenty-two batches of samples were analyzed by flame atomic



absorption.  Fifty-six sets of standards were analyzed with the



22 batch of samples.  The linear regression data for all the



standards analyzed with a given batch of samples are presented in



Table 5-5.  The average correlation coefficient is 0.9985; the



range was 0.9997 to 0.9944.  The average detection limit is 3.3



ppm.  The calculated detection limit was based a value of twice



the range of the 0 ppm standard above the Y-intercept.  A stan-



dard reference solution independently prepared from As_0^ with a



nominal value of 150 ppm was analyzed (1-2 dilution) with each



set of standards.   (Standards were prepared from a commercially
                             5-23

-------
TABLE 5-5.   LINEAR REGRESSION DATA (FLAME)
Date
(1983)
9/13
9/17
9/20
9/21
9/22
9/23
9/24
9/24
9/27
9/28
9/29
9/30
10/3
10/4
10/5
10/7
10/7
10/10
10/11
10/13
10/14
10/14
No. of
standard
curves
2
3
3
2
5
4
2
2
2
4
2
3
2
2
3
3
1
2
2
3
2
2
Y-intercept
-0.0031
-0.0007
-0.0058
-0.0064
+0.0020
0.0010
-0.0055
0.0018
0.0020
0.0000
-0.0008
0.0026
-0.0005
-0.0055
0.0016
0.0040
0.0002
0.0004
0.0035
0.0011
0.0010
-0.0062
Slope
0.00287
0.00265
0.00337
0.00332
0.00298
0.00319
0.00276
0.00272
0.00264
0.00344
0.00292
0.00448
0.00282
0.00304
0.00317
0.00290
0.00278
0.00273
0.00291
0.00306
0.00323
0.00297
Correction
0.9995
0.9990
0.9968
0.9995
0.9990
0.9985
0.9988
0.9976
0.9994
0.9995
0.9993
0.9997
0.9977
0.9990
0.9959
0.9983
0.9993
0.9990
0.9995
0.9944
0.9983
0.9984
Detection
limit, ppm
4.18
4.53
4.16
0.61
3.34
6.27
4.36
1.48
1.52
1.16
5.48
4.03
4.97
2.62
5.05
3.44
2.86
1.45
0.69
3.26
1.87
4.70
                  5-24

-------
available 100-ppm standard solution.)  The average value obtained



for the 56 analyses of this standard reference solution (SRS) is



157.1 ppm; the standard deviation  (SD) was 4.75 ppm [3.02 percent



relative standard deviation (RSD)].  Of the 56 determinations



made, only two fell outside the range of mean ± 2 SD (one above



and one below).  No significant historical data are available for



this level of arsenic analysis.



     These data indicate that the precision and accuracy of the



flame atomic absorption analyses are well within acceptable



limits.  The percent difference of the average measured value of



the SRS and its predicted value is 4.7 percent; the RSD of the



measured value is 3.0 percent.



     The results of the audit samples supplied by EPA and deter-



mined by flame atomic absorption are listed in Table 5-6.  These



results are consistent with the above data.  The relatively large



difference at 10 ppm is predictable inasmuch as it is only 3



times the detection limit.



     Seven samples were checked by the method of standard addi-



tion.  These results are contained in Table 5-7.  These analyses



were performed early during the project to ensure that no sig-



nificant matrix effects were present.  The standard addition of



Sample DC583 was conducted on September 17, 1983; the others were



performed on September 20, 1983.  The slopes of all the standard



addition analyses are between 0.9 and 1.1 except for the first



sample, which is probably due to an error in the spiking solution



on the first day.  An analysis of the results of the unspiked
                               5-25

-------
                         TABLE  5-6.  AUDIT RESULTS0
EPA No.
B-3-I
B-4-I
G-l-I
G-3-I
H-l-I
H-2-I
Lab No.
DC329
DC330
DC331
DC332
DC333
DC334
Volume, ml
500
500
500
500
500
500
EPA values,
mg As/liter
10
10
100
100
40
40
As cone.
measured,
mg/liter
13.6, 12.6
10.8, 12.6
103, 105
107, 106
45.4, 46.1
45.7, 48.5
Total
As, mg
6.82, 6.29
5.42, 6.29
51.3, 52.7
53.4, 53.1
22.7, 23.0
22.9, 24.3
Reported to and accepted  by  EPA  on  9/14/83.
                                   5-26

-------
TABLE 5-7.  ARSENIC RESULTS  OBTAINED BY STANDARD ADDITION METHOD
Lab No.
DC583 impinger

DC581 filter
bomb

DC588 filter
extract

DC588 filter
bomb

DC590 impinger

DC591 filter
extract

DC599 impinger

Spike,
ppm
0
10
20
30
0
50
100
200
0
20
40
80
0
10
20
40
0
10
20
40
0
20
40
80
0
10
20
40
Previously
determined
flame, ppm
17.61

a
274.7

a
70.38

22.9

12.3

92.62

10.07

Measured,
ppm
17.51
23.84
32.89
40.58
267.2
301.5
353.7
451.6
76.23
95.81
111.47
149.97
15.24
26.28
39.44
55.82
13.46
24.49
32.33
52.26
99.07
115.39
144.76
176.73
11.33
24.85
35.88
50.83
Linear
regression analysis
Slope = 0.783
Y intercept = 16.97
Corr. = 0.9979
X intercept = -21.68
Slope = 0.9390
Y intercept = 261.3
Corr. = 0.9977
X intercept = -278.3
Slope = 0.9145
Y intercept = 76.36
Corr. = 0.9994
X intercept = -83.50
Slope = 1.018
Y intercept = 16.38
Corr. = 0.9944
X intercept = -16.09
Slope = 0.9571
Y intercept = 13.89
Corr. = 0.9989
X intercept = -14.51
Slope = 0.9938
Y intercept = 99.20
Corr. = 0.9927
X intercept = -99.82
Slope = 0.9700
Y intercept = 13.75
Corr. = 0.9891
X intercept = -14.17
                              5-27

-------
samples and the X-intercepts (standard addition values) showed



less than a 10 percent difference except for Samples DC583 and



DC99.  The results for DC583 were expected because of the slope,



and the results for DC599 were not unreasonable inasmuch as this



value is at the low end of the calibration curve.



     The results of field blanks and reagents blanks are listed



in Tables 5-8 and 5-9.  These results show that contamination was



not a problem.  Because the results of the blanks are relatively



small and vary considerably, no blank corrections were applied to



the reported data.



     Sixteen batches of samples were analyzed by atomic absorp-



tion with graphite furnace techniques.  All samples below 30 ppm



were analyzed by furnace techniques.  Samples between 10 and 30



ppm that did not agree within approximately 20 percent of the



flame value were reanalyzed by both flame and furnace techniques.



Forty-three sets of standards were analyzed with the 16 batches



of samples.  Twelve of the data sets were reduced by linear



regression analysis (see Table 5-10); the remaining four sets



were fit to a second-degree polynomial  (see Table 5-11).  The



average correlation coefficient for the linear regression anal-



ysis is 0.9984; the range was 0.9996 to 0.9938.  The average



correlation coefficient for the polynomial fit is 0.9985; the



range was 0.9987 to 0.9983.  The average detection limit for the



graphite furnace is 0.0031 ppm.  The calculated detection limit



was based on a value of twice the range of the 0 ppm standard



above the Y-intercept.
                               5-28

-------
                           TABLE 5-8.   FIELD  BLANKS
Date
received
9/16/83
9/16/83
9/20/83
9/22/83
9/22/83
9/27/83
10/2/83
9/16/83
9/16/83
9/20/83
9/20/83
9/27/83
10/2/83
9/16/83
9/16/83
9/20/83
9/20/83
9/27/83
10/2/83
Lab No.
DC577
DC572
DC734
DC862
DC863
DD041
DD214
DC578
DC573
DC736
DC769
DD043
DD215
DC579
DC574/DC573
DC735/DC736
DC768/DC769
DD042/DD043
DD215/DD216
Sample
description
Filter
Filter
Filter
Filter
Filter
Filter
Filter
Rinse
Rinse
Rinse
Rinse
Rinse
Rinse
Impinger
Impinger
Impinger
Impinger
Impinger
Impinger
Volume,3
ml
NAb
NA
NA
NA
NA
NA
NA
229
338
374
374
399
297
500
625
610
610
719
762
Total
arsenic, mg
0.55
0.33
0.81
0.43
0.34
0.13
0.07

0.015
0.094
0.12r
1.73C
0.018

0.044
0.091
0.12.
1.18C
0.058
 Actual  volume or based on the average volumes  received  for these  fractions
 on the  day the blanks were received.

 Not applicable.

cLaboratory contamination of DD043 suspected.
                                     5-29

-------
                    TABLE 5-9.   REAGENT BLANKS
Date
9/22/83
9/22/83
9/22/83
9/27/83
9/27/83
9/27/83
9/27/83
9/27/83
9/27/83
9/29/83
9/29/83
9/29/83
9/29/83
9/29/83
9/29/83
9/29/83
9/29/83
9/29/83
Description
Filter NaOH extract
Bomb fraction
Rinse or impinger solution
Filter NaOH extract
Filter NaOH extract
Bomb fraction
Bomb fraction
Rinse or impinger solution
Rinse or impinger solution
Filter NaOH extract
Filter NaOH extract
Filter NaOH extract
Bomb fraction
Bomb fraction
Bomb fraction
Rinse or impinger solution
Rinse or impinger solution
Rinse or impinger solution
Total
arsenic, mg
0.03a
1.70a
0.02
0.08
0.57
0.05
0.02
0.001
0.003
0.01
0.01
0.01
0.01
0.01
0.01
0.007
0.001
0.005
Laboratory contamination  suspected.
                              5-30

-------
        TABLE 5-10.   LINEAR  REGRESSION  DATA—FURNACE
Date
(1983)
9/20
9/23
9/26
9/29
9/30
10/3
10/5
10/7
10/10
10/11
10/14
10/17
No. of
standard
curves
2
4
5
2
3
2
3
3
3
2
2
2
Y-intercept
0.0027
0.0025
0.0026
0.0017
0.0056
-0.0040
-0.0062
-0.0007
0.0025
-0.0036
0.0017
0.0011
Slope
2.590
3.547
3.557
3.657
3.707
3.779
3.719
3.616
3.673
3.733
3.841
3.849
Correction
0.9938
0.9995
0.9979
0.9996
0.9994
0.9995
0.9988
0.9976
0.9983
0.9990
0.9994
0.9979
Detection
limit, ppm
0.0008
0.0022
0.0062
0.0006
0.0021
0.0037
0.0038
0.0039
0.0011
0.0027
0.0042
0.0036
TABLE 5-11.   LEAST SQUARES  DATA  (BINOMIAL  EQUATION)-FURNACE

Date
(1983)
9/21
9/27
9/27
9/28
No. of
standard
curves
2
2
3
3


Y-intercept
-0.0091
-0.0124
-0.0094
0.0049

1st
degree
4.088
4.052
4.128
3.623

2nd
degree
-4.810
-3.570
-4.659
-2.922


Correction
0.9987
0.9987
0.9983
0.9984

Detection
limit, ppm
0.0025
0.0020
0.0039
0.0056
                            5-31

-------
     A standard reference solution  (SRS), which was independently



prepared from As-O. with a nominal value of 0.0750 ppm, was



analyzed with each set of standards.   (Standards were prepared



from a commercially available 1000-ppm standard solution.)  The



average value of the 43 analyses of this SRS was 0.0759 ppm, with



a standard deviation (SD) of 0.0036  (4.7 percent relative stan-



dard deviation RSD).  All of the 43 determinations were within ±2



SD of the mean.  Historically, the mean value for this SRS is



0.0762, with a standard deviation of 0.0027.  The values obtained



for the SRS solution during this project are in good agreement



with these historical data.  These data indicate that the preci-



sion and accuracy of the furnace atomic absorption analyses are



well within acceptable limits.  The difference between the aver-



age measured value of the SRS and its predicted value is 1.2



percent; the RSD of the measured value is 4.7 percent.



     It was not practical to perform a standard addition with the



furnace method because of the large dilutions required.  Furnace



dilutions usually ranged between 1:40 and 1:400.



     A final quality check of the data involved splitting the



sample extracts with ASARCO, who analyzed all samples above 20



.ppm by flame atomic absorption and all samples below 20 ppm by a



pyridine-SDDC colorimetric method.  The data generated by the two



labs are presented in Table 5-12.  These data are in good agree-



ment.
                               5-32

-------
                   TABLE 5-12.   COMPARISON OF ASARCO'S  AND
                         PEDCo'S ANALYTICAL RESULTS
Run No.
AABO-1
AABO-1
AABO-1
AABO-1
AEOD-2A
AEOD-2A
AEOD-2A
AEOD-2A
ABKI-3
ABKI-3
ABKI-3
ABKI-3
ABKI-3
ABKI-3
ABKI-3
ABKI-3
ABKI-3
ABKI-3
AEOD-1A
AEOD-1A
AEOD-1A
AEOD-1A
Lab No.
DC581
DC581 bomb
DC582
DC583
DC594
DC594 bomb
DC595
DC596
DC726 1st
DC726 2nd
DC726 bomb
DC727 1st
DC727 2nd
DC727 bomb
DC726/727 solid 1st
DC728
DC729
DC729 bomb solids
DC588
DC588 bomb
DC589
DC590
ASARCO, mg
0.26
13.35
1.62
13.20
1.60
0.13
2.10
4.05
2500
423
7000
1866
210
2025
a
5372
3589
a
3.50
0.35
2.34
5.25
PEDCo, mg
0.26
13.83
1.52
18.3
1.56
0.12
2.15
3.94
2571
430
7963
1868
188
2144
1199
5597
2995
90
3.52
0.36
2.63
4.93
ASARCO did not receive these samples.
                                    5-33

-------