&EPA
            United States
            Environmental Protection
            Agency
           Office of Air Quality
           Planning and Standards
           Research Triangle Park NC 27711
EMB Report 83-CDR-4
February 1984
            Air
Emission Test
Report
(Perlite)

W. R. Grace and
Company
Irondale, Alabama

-------
                 NSPS DEVELOPMENT
PARTICULATE AND PARTICLE SIZING EMISSIONS TESTING
            PERLITE EXPANSION FURNACE
             W. R. GRACE AND COMPANY
                IRONDALE, ALABAMA
                 FEBRUARY 1, 1984
                   Compiled by:

                     Entropy
              Post Office Box 12291
       Research Triangle Park, N. C. 27709
                    68-02-3852
              Work Assignment No. 2
             EMB Project No. 81 REG 8
                   Task Manager

                 Dennis Holzschuh
           Emission Measurement Branch
   Emissions  Standards and Engineering Division
       Research  Triangle Park, N. C., 27711
   OFFICE OF  AIR QUALITY PLANNING AND STANDARDS '
       OFFICE OF AIR,  NOISE, AND RADIATION
      U.  S. ENVIRONMENTAL PROTECTION AGENCY
       RESEARCH TRIANGLE PARK, N. C., 27711

-------
                           REPORT.CERTIFICATION









     The sampling  and  analysis  performed  for  this  report was carried out




under my direction and supervision.
Date  April 25. 1984
Signature
                                                Frank Jl./Phoenix.
     I have  reviewed  all  testing details and results  in this  test report




and hereby certify that  the test report is authentic  and accurate.
Date   April 25.  1984
Signature
                                                  D.  James Grove, P.E.

-------
                                  DISCLAIMER

     Although the research described in this report has been funded wholly  or
in part by the United States Environmental Protection Agency through
Contract 68-02-3852 to Entropy, it has not been subject to the Agency's  peer
and administrative review and therefore does not necessarily reflect the
views of the Agency, and no official endorsement should be inferred.

-------
                              TABLE OF CONTENTS

1.    INTRODUCTION	1-1
     1.1    Background	1-1
     1.2    Scope of the Project	1-1
     1.3    W. R. Grace Source Testing Program  	 1-1
     1.3.1     Source Applicability 	 1-2
     1.3.2     Outline of Testing Program 	 1-2
     1.4    Report Organization 	 1-3

2.    SUMMARY AND DISCUSSION OF RESULTS	2-1
     2 .1    Summary	2-1
     2.2    Discussion	2-1

3.    PROCESS DESCRIPTION AND OPERATION  : 	 3-1
     3.1    Introduction	3-1
     3.2    Pollutant/Sampling Points 	 3-1
     3.3    Process Description 	 3-3
     3.3.1     General	3-3
     3.3.2     Control Equipment   	 3-8
     3.4    Process Conditions During Testing  	 3-8
     3.5    Discussion	3-9

4.    SAMPLING LOCATIONS 	 4-1
     4.1    Suitability of Sampling Sites 	 4-1
     4.2    Perlite Expansion Furnace Inlet (Sample Location A) 	 4-1
     4.3    Ore  Feed Belt (Observation Location A)	4-1
     4.4    Cyclones Product Outlet (Sample Location B) 	 4-1
     4.5    Baghouse Exhaust Stacks (Sample Locations C and C') 	 4-1
     4.6    Baghouse Exhaust Stacks (Sample Locations D and D') 	 4-2

5.   SAMPLING AND ANALYTICAL METHODS  	 5-1
     5.1    Sampling Objectives    	 5-1
     5.2    Particulate Emissions  Testing 	 5-1
     5.3    Trace Metals Analysis  	 5-1

                            (continued next page)

-------
                                                                        ii
                              TABLE OF CONTENTS
                                 (continued)

     5.4    Sieve and Moisture Content Analysis 	 5-1
     5.5    Plume Opacity	5-1
     5.6    Fugitive Emissions  	 5-1
     5.7    Particle Size Testing	5-1

6.   QUALITY ASSURANCE  	 6-1
     6.1    Introduction	6-1
     6.2    Sampling Train Components 	 6-1
     6.3    Preseparator and Cascade Impactors  	  .. 6-1
     6.4    Sample Collection Substrates  	 6-2
     6.5    Substrate Weighting 	 6-2
     6.6    Sample Analysis	6-2
     6.7    EPA Method 3  	6-2
     6.8    EPA Method 7  	6-2

7.   APPENDICES
     7.1    Test Results and Example Calculations
     7.1.1     Participate Test Results, Baghouse West Stack & Example
                Calculations for Run 1
     7.1.2     Particulate Test Results, Baghouse East Stack
     7.1.3     Particle Sizing Test Results, Baghouse West Stack
     7.2    Field Data
     7.2.1     Particulate Field Data, Baghouse West Stack
     7.2.2     Particulate Field Data, Baghouse East Stack
     7.2.3     Particle Sizing Field Data, Baghouse West Stack
     7.2.4     Reactivity Field Data, Baghouse East Stack
     7.2.5     Visible Emissions, Method 9
     7.2.6     Fugitive Emissions, Method 22
     7.3    Sampling and Analytical Procedures
     7.4    Feed and Product Analytical Data
     7.5    Entropy Test Participants
     7.6    Calibration Data

-------
                                                                    iii
                           LIST OF TABLES
Number      Title

  1-1       Testing Log                                         1-2

  2-1       Baghouse Total Emission Rates and Concentrations    2-2
             and Visible Emissions Summary

  2-2       Particulate Tests Summary Baghouse West Stack       2-3

  2-3       Particulate Tests Summary Baghouse East Stack       2-4

  3-1       Emission Tests Conducted At W.R. Grace              3-2

  3-2       Data For Perlite Expansion Furnace                  3-5

  3-3       Feed and Product Specifications For Grefco NA 79    3-6

  3-4       Data For Perlite Corporation Baghouse               3-7

-------
                                                                    iv
                          LIST OF FIGURES



Number      Title                                              Page

  3-1       Perlite Expansion System Diagram                    3-4

  4-1       Perlite Expansion Process Sampling and              4-2
              Observation Locations

  4-2       Baghouse Exhaust Stack With Sampling Port           4-3
              and Point Locations (Sampling Location C)

-------
                                                                         1-1
1.    INTRODUCTION

     1.1  Background.  The Emission Measurement Branch of the U.S. EPA
(EPA EMB) is directing a project designed to generate support data for New
Source Performance Standards (NSPS) for calciners and dryers in the mineral
industries.  Emission standards for the various industries will be developed
based upon process-related emission factors determined from the testing of
controlled sources.

     1.2  Scope of the Project.  The U.S. EPA EMB is responsible for
coordinating the efforts of Entropy and Midwest Research Institute (MRI) to
achieve the goals of the testing program.  Entropy has been retained under
EMB Contract No. 68-02-3852, Work Assignment No. 2 to conduct testing
programs at designated industrial facilities.  Entropy is to perform emission
measurements at the recommended sampling locations, obtain process feed and
product samples, and, in conjunction with Research Triangle Institute (RTI),
conduct sample analyses.  MRI will monitor process and operating conditions
in order to designate suitable testing conditions for the respective
processes and to provide a record of process and operational data during the
testing.

     1.3  W. R. Grace Source Testing Program.  The present report covers
stationary source sampling performed at  the W.R. Grace & Company
manufacturing plant in Irondale, Alabama on February 1, 1984 to characterize
emissions from the perlite expansion pollution control equipment.  Clarence
Duckworth of W. R. Grace and Company and Frank Clay of EPA EMB were present
to coordinate the testing.  Amy J. Kowalski and Butch Smith of MRI monitored
the plant process and recorded operational data during the testing.

     1.3.1  Source Applicability.  W. R. Grace and Company operates a high
temperature perlite  expansion  furnace.   This furnace has the capability of
processing NA-79 crushed perlite which has a density of about  sixty pounds
per cubic foot (60 Ibs/fty).   Emissions  from the perlite expansion system
are controlled by a  baghouse,  which  represents the best available control
 technology (BACT).

-------
                                                                         1-2
     1.3.2  Outline of Testing Program.  Three sets of EPA Method 5 runs were
performed concurrently at the baghouse east and west stacks to determine the
particulate emissions from the perlite expansion process.  The impinger
reagents from one Method 5 run at each location were analyzed for trace
metals.  Concurrent with two sets of the Method 5 runs, Method 9 opacity
readings were taken at the baghouse east and west stacks, and Method 22
readings for fugitive emissions were taken at the ore feed belt.  One
particle sizing run was done at the baghouse west stack, while a simultaneous
reactivity run* was performed as a control measure at the baghouse east
stack.  Feed and product samples were taken for moisture content and sieve
analysis.  Table 1-1 presents a test log which summarizes the test dates,
sampling locations, run or sample numbers, and the types of testing
performed.

                                  TABLE 1-1
                            TESTING LOG OF 2/1/84
    Sampling Location
   Baghouse West  Stack
Sampling Objective
    particulate
   trace metals
   particle size
      opacity
fugitive emissions
sieving & moisture
sieving & moisture

    particulate
   trace metals
    reactivity
      opacity
 *  A reactivity run is a simultaneous duplicate run using a filter  prior  to
 the impactor stages to collect all the particulate in the gas  stream.  The
 purpose of the run is to determine if gases in the stack effluent  are
 reacting with the impactor stages and creating a high bias on  the  weights of
 the normal particle size run.   No weight gain on the impactor  stages of  the
 reactivity run indicates no reactivity bias on the normal particle size  run.
      Ore  Feed  Belt
   Furnace Inlet  Feed
 Cyclones  Product Outlet

   Baghouse East  Stack
Method
5
5
impactor
9
22
grab sample
grab sample
5
5
impactor
9
Run Numbers
1,2,3
2
SI
1 & 2
1 & 2
1,2,3
1,2,3
4,5,6
5
Rl
4 & 5

-------
                                                                         1-3
     1.4  Report Organization.  Immediately following is the "Summary of
Results" section.  Appendix 7.1 presents the complete results of each run;
field data can be found in Appendix 7.2.  The source and the process are
described in the "Process Description and Operation" section.  The next
section, "Sampling Locations" provides a comprehensive description and
illustration for each location; "Sampling and Analytical Procedures" fellows,
describing the sampling strategy used.  Descriptions of the equipment and
procedures can be found in Appendix 7.3, while Appendix 7.4 presents
analytical documentation.  The final section, "Quality Assurance," notes  the
procedures used to ensure the integrity of the sampling program; Appendix 7.6
provides pertinent calibration data.  Appendix 7.5 contains a listing of  the
Entropy test participants and their roles in the testing program.

-------
                                                                         2-1
2.   SUMMARY AND DISCUSSION OF RESULTS

     2.1  Summary.  Table 2-1 presents the total baghouse emission rates  and
concentrations and visible emissions summary for the particulate testing
performed at the baghouse west and east stacks.  A run-by-run summary of  the
particulate results for the west and east baghouse stacks is presented in
Tables 2-2 and 2-3, respectively.  Detailed results for the testing program
appear in Appendix 7.1.

     2.2  Discussion of Results.  Particle sizing Run SI data indicate a  mass
median diameter of approximately 4 microns; however, due to the low
concentrations, the catches on some stages are extremely low so, caution
should be used in interpreting the particle sizing data.  The results appear
in Appendix 7.1.3.  Due to process upsets, the particle sizing testing
scheduled for February 2 was  not completed; limited data from the aborted
runs show the mass median diameter to be  consistant with Run SI.
     Only the first set of two-hour visible and fugitive emissions runs was
completed.  The second set of readings was interrupted after one hour due to
the onset of darkness; the scheduled third set was not performed.
     Visible emissions plume  opacity was  negligible except during the
cleaning  cycles.  Fugitive emissions were observed only when the conveyer
belt was  operating.

-------
                                  TABLE 2-1

                BAGHOUSE TOTAL EMISSION RATES & CONCENTRATIONS
                                                                        2-2
Emission Rates, Ib/hr:
   Baghouse West Stack
   Baghouse East Stack

        Sum Total

Emission Rates, kg/hr:
   Baghouse West Stack
   Baghouse East Stack

        Sum Total

Air Flow Rates. SCFM:
   Baghouse West Stack
   Baghouse East Stack

        Sum Total

Concentration, Gn/DSCF;
   Baghouse West Stack
   Baghouse East Stack

    Weighted Average*

Concentration, mg/DSCM;
   Baghouse West Stack
   Baghouse East Stack

    Weighted Average**

Average Opacity, Percent:
   Baghouse West Stack
   Baghouse East Stack

Fugitive Emissions:
   Observation Time,  rain.
   Duration of Emissions, min.
                                 —	Run Numbers	
                                  1&4      2&5      3&6
       0.256
       0.321

       0.577
       0.116
       0.146

       0.262
        2,702
        1,986

        4,688
       0.0110
       0.0189

       0.0144
       25.280
       43.142

       32.948
         0.16
         0.14
        95.00
        10.60
 0.298
 0.338

 0.636
 0.135
 0.153

 0.288
 2,601
 1,986

 4,587-
0.0134
0.0198

0.0162
30.598
45.409

37.067
  0.04
  0.35
 45.00
 35.63
 0.239
 0.338

 0.577
 0.108
 0.153

 0.261
 2,607
 1,997

 4,604
0.0105
0.0198

0.0146
23.944
45.225

33.406
Average

 0.264
 0.332
 0.596
 0.120
 0.151

 0.271
 2,637
 1,990

 4,627
0.0116
0.0195

0.0150
26.607
44.592

34.550
     *  Weighted  Gn/DSCF =
Ib/hr (7000)

 SCFM (60)
                               453,592
    ** Weighted mg/DSCM =  	  *  Weighted Gn/DSCF
                           7000 (0.02832)

-------
                             TABLE  2-2

                PARTICULATE TESTS  SUMMARY OF RESULTS

               Expansion Furnace Baghouse West Stack
                                                                   2-3
                                     1
Run Date

Test Train Parameters:
  Volume of Dry Gas
    Sampled, SCF*

  Percent Isokinetic

Stack Parameters:
  Temperature, Deg. F

  Air Flow Rates
    SCFM*. Dry

    ACFM, Wet

Method 5 Test Results:
  Catch, Milligrams

  Concentration,
    Grains Per DSCF*

    Milligrams Per DSCM

  Emission Rate,
    Pounds Per Hour

    Kilograms Per Hour
02/01/84     02/01/84
  90.792


    99.1


     370


   2,702

   4,525


    65.0


  0.0110

  25.280


   0.256

   0.116
89.090


 101.0


   383


 2,601

 4,463


  77.2


0.0134

30.598


 0.298

 0.135
02/01/84


  90.990


   100.5


     378


   2,670

   4,568


    61.7


  0.0105

  23.944


   0.239

   0.108
* 68 Deg. F. - 29.92 in. Hg.

-------
                             TABLE 2-3

                PARTICIPATE TESTS SUMMARY  OF  RESULTS

               Expansion Furnace Baghouse  East  Stack
                                                                   2-4
Run Date

Test Train Parameters;
  Volume of Dry Gas
    Sampled, SCF*

  Percent Isokinetic

Stack Parameters:
  Temperature, Deg. F

  Air Flow Rates
    SCFM*. Dry

    ACFM, Wet

Method 5 Test Results:
  Catch, Milligrams

  Concentration,
    Grains Per DSCF*

    Milligrams Per DSCM

  Emission Rate,
    Pounds Per Hour

    Kilograms Per  Hour
02/01/84     02/01/84     02/01/84
 106.647

   104.6


     371

   1,986

   3,346


   130.3


  0.0189

  43.142


   0.321

   0.146
108.400

  106.3


    376

  1,986

  3,380

  139.4


 0.0198

 45.409


  0.338

  0.153
107.826

  105.2


    372

  1,997

  3,389

  138.1


 0.0198

 45.225


  0.338

  0.153
* 68 Deg. F. - 29.92 in. Hg.

-------
                                                                         3-1
3.   PROCESS DESCRIPTION AND OPERATION

     3.1 Introduction.  Source emission tests were conducted on the perlite
expansion furnace at the W. R. Grace & Co. (Grace) plant in Irondale, Alabama
from January 30 through February 2, 1984.  Mr. Frank Clay of EPA/EMB was
present during the testing to observe emission testing procedures.  The tests
were conducted by an eight-person test crew headed by Mr. Frank Phoenix of
Entropy.  The process was monitored by Stacy Smith and Amy Kowalski
from MRI.  Mr. Clarence Duckworth of Grace coordinated the testing with plant
personnel and provided process information.

     3.2 Pollutants/Sampling Points.  The primary objectives of the emission
test were to obtain the following data for the outlet of the perlite
expansion furnace baghouse:  (1) particulate matter (PM) concentrations, (2)
PM mass emission rates, and (3) particle size distributions (PS) for the PM.
Visible emission (VE) observations were made at  the two outlet stacks
simultaneously with the emission testing.  VE measurements were also made of
the process fugitive emissions at  the expansion  furnace feed inlet.  The feed
inlet  consists of a covered belt conveyer which  dumps into a bucket
elevator.  No VE measurements were taken on the  product outlet because the
system was totally enclosed with no visible leaks.  Grab samples of  the feed
and product material  and  of the baghouse fines were taken for particle size
sieve  and moisture content  analyses.  Table 1 presents  the tests conducted at
this facility.

-------
                                                                              3-2
         TABLE 3-1. EMISSION TESTS CONDUCTED AT W. R. GRACE & CO.'
Sampling point
Furnace feed inlet
Furnace product
outlet
Test type
Visible emissions
Moisture content
Particle size sieve
Moisture content
Particle size sieve
Test No. of
method(s) samples/run
EPA-9,22
ASTM D2216
ASTM D422
ASTM D2216
ASTM D422
1
1
1
1
1
Baghouse outlet
  stacks (2)
Particle concentration
Particle size
Visible emissions
EPA-5             1
EPA draft method  1
EPA-9             1
 EPA Method 5 tests consisted of three runs, all on February 1, 1984.
 Particle size testing consisted of one run also on February 1, 1984.
 Additional particle size runs were not possible because of fan malfunc-
 tions.  Simultaneous particle size and particulate testing was not
 feasible because of the small stack diameter.

-------
                                                                           3-3
     3.3 Process Description

     3.3.1 General.  Grace expands one size of perlite ore, for horticultural
uses, in its Perlite Corporation Model VS-450 vertical expansion furnace.
During emission testing, Grefco NA 79 ore was processed.  The pre-sized
perlite concentrate is fed into the 0.7-m (28-in.) -diameter stainless steel
furnace through a surge preheater at a rate of about 0.91 mg/h (1 ton/h).
Four variable speed screw feeders on the side of the furnace inject the
perlite 1.5 to 1.8 m (5 to 6ft) above the natural gas-fired combustion
burner.  The fuel usage rate for this furnace is 198 to 227 m3/h (7,000 to
8,000 ft3/h).
     The perlite is expanded 4 to 20 times its original volume in the hot
zone of the furnace at temperatures of 982 degrees to 1093 degrees Celsius
(1800 to 2000 degrees Fahrenheit).  The four baghouse exhaust fans convey the
expanded perlite particles through 30.5-cm (1-ft) -diameter circular ductwork
to a 1.75-m (69-in.) -diameter product collection cyclone.  The collected
perlite falls through a cooler/classifier unit prior to bagging.  The air
stream from the product cyclone and cooler/classifier cyclone passes through
a four-compartment Perlite Corporation baghouse  prior to being emitted to the
atmosphere.  Figure 1 shows  an expansion furnace with the Perlite Corporation
cyclone and baghouse that is similiar to the system used at Grace.  The
controlled expansion temperature of the furnace, as indicated by the control
panel monitor, ranges from 799 to 832 degrees Celsius (1470 to 1530 degrees
Fahrenheit).  The  furnace has  a retention time of two to three seconds.
Table 2 presents the design  and operating parameters for the furnace.
Table 3 presents the specifications for the feed and product material
provided  by Mr. Frederick Eaton of Grace.

-------
                                                             3-4
Figure 3-1. Perlite  expansion   system
          W. R. Grace & Co.

-------
                                                                               3-5
                   TABLE 3-2. DATA FOR PERLITE EXPANSION
              FURNACE AT W. R.  GRACE & CO., IRONDALE, ALABAMA
Manufacturer

Model

Date of installation

Design production rate, Mg/h (ton/h)

Actual production rate, Mg/h (ton/h)


Hours of operation
  hours/day
  days/week

Retention time, s

Maximum gas temperature, °C (°F)

Fuel used

Fuel usage rate, mVh (ft3/h)


Feed moisture content, %

Feed particle size

Feed density, kg/m3 (lb/ft3)

Product moisture content, %

Product density, kg/m3 (lb/ft3)
Perlite Corporation

VS-450

1977

1.09 (1.20)

See Confidential
Addendum Item No.  2


8-16
5

2-3

982-1093 (1800-2000)

Natural gas

See Confidential
Addendum Item No.  3

2-5

90 percent £20 mesh

881+ (55+)

0

64-112 (4-7)

-------
                                                                        3-6
TABLE 3-3. FEED AND PRODUCT SPECIFICATIONS FOR GREFCO NA79 PERLITE3
Mesh size
+4
+8
+12
+16
+20
+50
+100
Cumulative
Raw ore
0
20
60
85
90
—
—
percent retained
Expanded
product
--
50% minimum
—
—
90% minimum
—
96% minimum
        aTelecons.  A. J. Kowalski, MRI, with Mr. Frederick
         Eaton, W. R. Grace & Co. on  February 1 and April 4,
        b!984.
         W. R. Grace & Co. quality control purchase
         specifications.

-------
                                                                             3-7
              TABLE 3-4. DATA FOR PERLITE CORPORATION BAGHQUSE
                 W. R. GRACE & COMPANY, IRONDALE, ALABAMA3
Manufacturer
Bag material
Air-to-cloth ratio
Inlet temp., °C (°F)
Actual gas flow rate, m3/s (acfm)
Pressure drop, kPa (in. w.c.)
Number of stacks
Stack height (from roof), m (ft)
Stack diameter, m (ft)
Number of compartments
Design outlet grain loading, g/dsm3
  (gr/dscf)
Design efficiency
Perlite Corporation
Glass graphite
2.6:1
232-243 (450-470)
170 @ 177°C (6,000 @ 350°F)'
1.5 (6)
2
11.3 (27)
0.4 (1.25)
4
1.8 xlO-5 (0.01)

>99%
 Telecon, 6/7/83 with Mohammed Huda, Jefferson County Dept.  of Health,
.Birmingham, Ala., and Section 114 response.
 Stack temperature.

-------
                                                                         3-8
     3.3.2 Control Equipment.  Particulate emissions from the perlite
expansion furnace are controlled by a baghouse.  Data for the baghouse are
shown in Table 4.  The baghouse was manufactured by the Perlite Corporation.
It has four compartments and is equipped with 100 glass graphite bags.  The
design air-to-cloth ratio is 2.6:1.  The temperature of the inlet gas is 232
to 243 degrees Celsius (450 to 470 degrees Fahrenheit).  The collected
material from the baghouse is bagged separately and landfilled. The baghouse
is equipped with four fans, one on each compartment.  The baghouse is cleaned
by compartment when the respective fan is shut off for a few seconds.  The
cleaning cycle is set on seven minutes per compartment.

     3.4 Process Conditions During Testing.  All processes were operating
normally during  the emission testing.  The furnace operation is monitored
from a control panel that contains gauges for the furnace outlet temperature,
the baghouse internal temperature, and the ore feeder speed.  To achieve the
required amount  of exfoliation for final product quality, the hot section of
the furnace must be 982 to 1093 degrees Celsius (1800 to 2000 degrees
Fahrenheit).  This corresponds to  a  furnace  control temperature of
approximately 816 degrees Celsius  (1500 degrees Fahrenheit) according to Mr.
Duckworth.  Throughout  the testing the furnace control temperature was 799  to
832 degrees Celsius (1470 to 1530  degrees Fahrenheit) indicating normal
furnace  operation.
     The design  capacity  of  the vertical expansion furnace is 1.09 mg/h (1.2
ton/h).  The normal actual production  rate is 0.9 mg/h (1 ton/h).  Natural
gas with a  heating value  of  1,000  Btu/ft3 was  used  to  fire the  furnace
during  the  testing.   The  gas  firing  system was operating  normally during  the
test.
     Additional  information  about  the  process  operation  during  testing  is
provided in a  Confidential Addendum  to this  report.

-------
                                                                         3-9
     35 Discussion.  Due to the small diameter of the baghouse outlet
stacks [0.4 m (1.25 ft)}, simultaneous particle size and Method 5 testing was
not possible.  In addition, two of the four baghouse fans malfunctioned on
three separate occasions.  None of the malfunctions affected any of the
testing runs.  As a result of  these breakdowns, however, Method 5 Run No. 3
was performed at night, and VE readings  could not be taken.  Also, only one
2-hour particle size run was completed due to the fan breakdowns.
     Additional particle size  testing was scheduled on Thursday.
February 2, 1984, to complete  the testing requirements.  Originally, the
three particle size runs were  to be performed in only one stack.  Due to time
constraints, EMB personnel approved simultaneous particle size testing using
both stacks.  After approximately one hour of testing, the  third fan
malfunction occurred, and testing was suspended.  Because acceptable Method 5
runs had already been obtained and because the outlet particle size data
could not  be obtained concurrently and were not critical to the calciners and
dryers project, it was  decided that the  testing would be terminated without
the final  two particle  size runs.
Confidential Attachment

-------
                                                                        4-1
4.   SAMPLING LOCATIONS

     4.1  Suitability of Sampling Sites.  The primary goal  of  the  testing
program was to characterize emissions from the perlite expansion pollution
control equipment.  Additionally, physical properties of the process  feed
stock and the product were investigated.  Sampling sites appropriate  to  these
objectives were approved prior to testing.  Each sampling location is
discussed individually, and the position of each site within the system  is
illustrated in Figure 4-1.

     4.2  Perlite Expansion Furnace Inlet (Sampling Location A).  Grab
samples of the feed to the expansion furnace were collected for sieve
analysis and moisture content determination.  Samples were taken during  each
set of  simultaneous Method 5 runs.

     4.3  Ore Feed Belt (Observation Location E).  Fugitive emissions were
determined at the ore feed belt  according to EPA Reference Method 22.

     4.4  Cyclones Product Outlet (Sampling Location B).  Grab samples of
product from the  cyclones product outlet were collected for sieve analysis
and moisture content determination.  Samples were taken during each set  of
simultaneous Method 5 runs.

     4.5  Baghouse Exhaust Stacks (Sampling Locations C and C').  Measures  of
particulate  emissions were made  at  the  identical baghouse east and west
stacks.  A determination  of particle size distribution was made only at  the
baghouse west stack, while a  reactivity particle size run was conducted  for
reference at the  baghouse east stack.   Figure 4-2 illustrates the dimensions
and  sampling points which are typical for both  stacks.
     Both stacks  are vertical with  inside diameters of  15 inches.   Each  stack
has  two sampling  ports  spaced 90° apart and  located 12  inches (0.8 stack
diameters) upstream from  the  stack  outlet and 60 inches (4 stack diameters)
downstream from  the nearest flow disturbance.   In accordance with EPA
Method  1 (promulgated on  September  30,  1983), 24 sampling points were used,

-------
           OBSERVATION
           LOCATION E
         PERLITE
        EXPANSION
         FURNACE
 PRIMARY
CYCLONE
         F.D.  FAN
r
                         COOLER
                       CLASSIFIER
              SECONDARY
               CYCLONE
                                    >B
                              TO BAGGERS
                                                   .	»-  PRODUCT FLOW
                                                   — •*-  AIR FLOW
                                                     <8>  GRAB SAMPLES
                                                                           OBSERVATION
                                                                           LOCATION D. D1
BAGHOUSE
COMPARTMENT
BAGHOUSE
COMPARTMENT
BAGHOUSE
COMPARTMENT
BAGHOUSE
COMPARTMENT
1
2
3
4
                                                                                           STACK
                                                                                             C'
                                                 A,B
                                                   C
                                                   C'
                                                 D,D'
                                                   E
                                     ASTM METHOD
                                     METHOD 5,  IMPACTOR
                                     METHOD 5
                                     METHOD 9
                                     METHOD 22
^IGURE 4-1. PERLITE EXPANSION  PROCESS SAMPLING LOCATIONS
            W. R. GRACE,  IRONDALE,  ALABAMA
                                                                                                        I
                                                                                                        ro

-------
                                                                 4-3
        M
                 AXES: 2
                 POINTS/AXIS: 12
                 TOTAL POINTS:  24
            12"
             60"
                     15"
                    A  ;..B
                    6
                                           SECTION M-M
                                      2 SAMPLING PORTS
                                                          15"  DIA.
                                      NOTE:  EAST AND WEST
                                             BAGHOUSE STACKS
                                             ARE IDENTICAL
             FROM
             FAN
FROM
FAN
FIGURE .'4-2. EAST OR WEST BAGHOUSE  OUTLET  STACK  DIMENSIONS  WITH
           '•SAMPLING PORT AND POINT" LOCATIONS

-------
                                                                        4-4
i.e., 12 sampling points on each of two traverse axes {labeled A and B).
Each point was sampled for five minutes resulting in a net test time of  120
minutes.
     Particle size determination at the baghouse west stack was performed
using the four-point scheme recommended by the Industrial Environmental
Research Laboratory (IERL).  For the particle size testing the stack cross
section was divided into four equal areas, i.e., two traverse axes with  two
sample points per axis.  The centroid of each equal area was sampled for 30
minutes resulting in a net sample time of 120 minutes.

     4.6  Baghouse Exhaust Stacks (Observation Locations D and D1).  Plume
opacity observations for both baghouse exhaust stacks were performed
according to procedures outlined in EPA Reference Method 9.

-------
                                                                        5-1
5.   SAMPLING AND ANALYTICAL METHODS

     5.1  Sampling Objectives.  This section describes the  sampling  and
analytical procedures which were employed at the W.  R. Grace plant in order
to gather data concerning emissions from the perlite expansion pollution
control equipment and to investigate physical properties of the process feed
stock and the product.  The sampling program included outlet tests for
particulate emissions, plume opacity, and, at the baghouse  west stack only,
particle size distribution.

     5.2  Particulate Emissions Testing.  Particulate emissions sampling
conformed to the standards and procedures set forth by EPA  Reference
Methods 1-5 and described in 40 CFR Part 60, Appendix 7.3.

     5.3  Trace Metals Analysis.  For one run at each location, the  Method 5
particulate catch and the distilled water reagent from the  impingers were
analyzed for trace metals by using atomic absorption or inductively  coupled
Plasma  Spectroraetry.  These metals are  zinc, nickel, iron,  manganese,
vanadium, calcium, silicon, aluminum, magnesium, fluorine,  beryllium,
uranium, lead, and mercury.

     5.4  Sieve Analysis and Moisture Content.  Sieve analysis and moisture
content determinations were performed on  all feed and product samples. ASTM
Method  D 2216 was used to analyze  the samples for moisture content,  while
ASTM Method D 422 was used for sieve analysis.

     5.5  Plume Opacity.  Visible  emissions  observations were performed in
accordance with EPA  Reference Method 9  as described in 40 CFR Part 60.

     5.6  Fugitive Emissions.  Fugitive emissions were determined in
accordance with EPA  Reference Method 22 as  described  in 40 CFR Part  60.

     5.7  Particle Size Tests.  Particle  size determinations were made using
a right angle  inlet  preseparator,  followed  by an Andersen Mark III cascade
impactor.  The  test  procedures were  based upon  the  publication, "Procedures
for Cascade  Irapactor Calibration  and Operation  in Process  Streams -  Revised
1979,"  developed  by  the Industrial Environmental Research Laboratory (IERL)
and Southern Research Institute.

-------
                                                                        6-1
6.  QUALITY ASSURANCE

     6.1  Introduction.  The goal of quality assurance for the project was  to
ensure the accuracy of all data  collected.  The procedures used are contained
in Entropy's "Quality Assurance Program Plan," which was approved by the U.S.
EPA EMB in the  contract agreement governing the project.
     In order to ensure continuity among field testing personnel, daily
meetings were held before each day of the field testing.  At the meetings,
results from the testing conducted on the previous day were reviewed.
Responsibilities were  clearly delineated for each member of the testing team,
and questions were addressed and resolved immediately.  In situations where
more than one person was performing similar activities, consistency was
ensured through communication at the meetings.
     In addition to the general  quality assurance measures, specific quality
assurance activities were conducted for several of the individual test
methods performed.

     6.2  Sampling Train Components.  Entropy's sampling equipment, including
nozzles, pitot  tubes,  dry gas meters, orifices, and thermocouples, was
uniquely identified and  calibrated in accordance with documented procedures
and acceptance  criteria prior to and at the completion of  the field testing
program.  All sampling equipment was manufactured by Nutech Corporation,
Andersen 2000 or by Entropy.  Calibration data for the sampling  equipment are
contained in Appendix  7.6.

     6.3  Preseparator and  Cascade Impactors.  All internal components and
surfaces of  the impactors were  cleaned  in an ultrasonic bath  to  ensure that
all  surface  impurities were removed, and visual inspections for  cleanliness
were made  prior to  shipment to  the field.  After each  sample  recovery, the
preseparator,  the  impactor  body, and the  plates were  rinsed with acetone to
ensure that  all organic  residuals  and/or  particulate matter were removed.

-------
                                                                        6-2
     6.4  Sample Collection Substrates.  Schleicker & Schuell #30  glass  fiber
sample collection substrates were used for particle size testing.   To  prevent
contamination of the substrate surface, all filters were handled with
laboratory tweezers.  This procedure was used during impactor assembly,
sample recovery, and weighing of the substrates.

     6.5  Substrate Weighing.  An analytical balance capable of weighing to
the nearest 0.01 milligram (mg) was used.  To ensure that no weight bias was
produced from the preparation, transportation, recovery, or weighing
procedures, two control samples were obtained during the test program.  A
reactivity run was performed to ensure that the flue gases did not interact
with the substrate to produce extraneous results.  For the reactivity  run,  a
solid filter was placed in the front section of the impactor, and  the
impactor was introduced into the stack, and a sample was pulled through the
head using the parameters outlined for a normal particle sizing run.  The
average difference between the pre- and post-test weights, as shown in
Appendix 7.6., was 0.08 milligrams, based upon weight differences  ranging
from 0.01 to 0.19 milligrams.  A blank run was also performed to demonstrate
that the impactor could be assembled and disassembled without affecting the
weight of the substrate.  The average difference between the pre-  and
post-assembly weights was 0.03 milligrams, based upon a difference ranging
from 0.0 to 0.03 milligrams.

     6.6  Sample Analysis.   In order to  reduce  the probability of errors or
inconsistent  results, one member of the  field crew had  sole  responsibility
for  the sample  analysis procedure.  Sample analysis was performed in a room
dedicated exclusively to filter weighing.

     6.7  EPA Method 3.  All Method 3  analyses  were performed in triplicate.
Each analyzer was checked for  leaks prior  to  any analysis as specified in the
method.  Samples were analyzed within  four hours of  collection.

     6.8  EPA Method 9.  The visible emissions  observers held current
 certifications  issued within the last  6  months.  Documentation verifying the
 observer's  certifications are  provided in Appendix 7 .2.4.

-------