United States
  EPA
          EVALUATION  OF THE POLYAD* FB
AIR PURIFICATION AND  SOLVENT RECOVERY PROCESS
               FOR STYRENE REMOVAL
             control
technology center

-------
                                                 EPA-6QQ/ -  -
          EVALUATION OF THE POLY AD* FB
AIR PURIFICATION AND SOLVENT RECOVERY PROCESS
              FOR STYRENE REMOVAL
                    Prepared by:

      Larry Felix, Randy Merritt, and Ashley Williamson
              Southern Research Institute
       Environmental Sciences Research Department
                   P. O. Box 55305
              Birmingham, AL 35255-5305
           EPA Contract Number 68-D2-0062
                     Task No. 2
             Task Officer Bobby E. Daniel
     Air and Energy Engineering Research Laboratory
          U. S. Environmental Protection Agency
           Research Triangle Park, NC 27711
                    Prepared for
              Control Technology Center
          U.S. Environmental Protection Agency
           Research Triangle Park, NC 27711

-------
                             CONTROL TECHNOLOGY CENTER

                                        Sponsored by:

                                 Emission Standards Division
                          Office of Air Quality Planning and Standards
                             U.S. Environmental Protection Agency
                              Research Triangle Park, NC 27711
                        Air and Energy Engineering Research Laboratory
                             U. S. Environmental Protection Agency
                              Research Triangle Park, NC 27711
                         Center for Environmental Research Information
                              Office of Research and Development
                             U.S. Environmental Protection Agency
                                     Cincinnati, OH 45268
                                         ABSTRACT

Manufacturing processes that involve the spraying of styrene-based resins have been identified as a
possible significant source of volatile organic compound (VOC) emissions that may affect human health
and contribute to the ozone non-attainment problem. Until recently, no known technology has been
demonstrated to control such emissions of styrene. Now, two processes have been developed to
control styrene emissions and a short-term field evaluation was planned to characterize the styrene
removal efficiency of pilot-scale versions of each process.  Unfortunately, because only one system
could be made available for the proposed test period, only one of these systems could be tested. This
test was carried out at a facility (Eljer Plumbingware in Wilson, NC) that manufactures polyester
bathtubs and shower stalls by spraying styrene-based resins onto molds in vented, open, spray booths.
A side stream of air exhausted from one of the spray booths in the gel coating part of the process was
used for this test.

In this study the styrene removal efficiency of a pilot-scale version  of the Polyad* fluidized bed (FB) air
purification and solvent recovery system was quantified by continuously measuring the total hydrocarbon
(THC) content of spray booth exhaust air entering and exiting the Polyad FB device with THC analyzers
and by collecting NIOSH Method 1501/EPA Method 18 samples at the inlet and exit of the Polyad FB
device.  Styrene removal efficiencies greater than 90% were achieved.

This work was performed at the request of the Control Technology Center (CTC) steering committee to
provide information to state and local agencies for use in responding to public concerns.

-------
                                   TABLE OF CONTENTS

Abstract	fi

List of Figures	v

List of Tables	vi

Acknowledgments	vfi

Preface	vfii

Metric to Nonmetric Conversions	ix

Section	Paae

1.  Introduction	1

2.  Project Description	3
    2.1   Experimental Approach	3
    2.2  Eljer Plumbingware Facility	4
    2.3  The Polyad® FB Process	7
         2.3.1  Pilot-Scale Polyad* FB Device	11
         2.3.2  Specific Test Conditions	14
    2.4  Experimental Apparatus	16
         2.4.1  Connection to the Pilot-Scale Polyad FB Unit	16
         2.4.2  Sampling Van	16
    2.5  Experimental Methods and Procedures	21
         2.5.1  Total Hydrocarbon Analyzers	21
         2.5.2  Collection of Volatile Organic Compound Samples	22
         2.5.3  Collection of Recovered Solvent Sample	23
         2.5.4  Total Flow Rate Measurements	24

3.  Data, Results, and Discussion	25
    3.1   Total Hydrocarbon Analyzer Data	25
         3.1.1  Inlet Data	26
         3.1.2  Outlet Data	32
         3.1.3  Efficiency Data	36
         3.1.4  Estimated Styrene Emissions from Gel Coat Booth #2	46
    3.2   Volatile Organic Compound Data	48
         3.2.1  Measurements at Spray Booth Exhaust Stacks	49
         3.2.2  Measurements at the Inlet and Outlet of the Polyad FB Device	53
    3.3  Analysis of Recovered Solvent Sample	56
    3.4  Total Flow Rate Data	56
    3.5   Styrene Capture by the Polyad FB Device	-	59

4.  Costs Associated with Applying the Polyad FB System to Styrene Removal	61

5.  Summary and Conclusions	68
                                               Hi

-------
                          TABLE OF CONTENTS, CONTINUED


                                                                                 .70
Appendix A NIOSH Method 1501	

                                                                                  78
Appendix B Quality Control Evaluation Report	


                                                                                  88
Appendix C Total Hydrocarbon Analyzer Daily Results	
                                              IV

-------
                                     LIST OF FIGURES

Rgure                                                                               Page

1.  Layout of the Eljer Plumbingware Facility	5
2.  Diagramatic representation of the Polyad FB Air Purification and Solvent Recovery Process	9
3.  Schematic diagram of the pilot-scale Polyad FB unit tested at Eljer Plumbingware	12
4.  Component layout of the pilot-scale Polyad FB unit tested at Eljer Plumbingware	13
5.  Overall arrangement for sampling at the Eljer Plumbingware Facility	17
6.  Equipment arrangement used for sampling with THC analyzers	18
7.  Equipment arrangement used for NIOSH Method 1501 sampling	19
8.  Inlet hydrocarbon emissions, November 3,1992	27
9.  Inlet hydrocarbon emissions, November 4,1992	28
10. Inlet hydrocarbon emissions, Novembers, 1992	29
11. Inlet hydrocarbon emissions from 0910 to 0950, Novembers, 1992	30
12. Inlet hydrocarbon emissions from 1110to 1150, Novembers, 1992	31
13. Outlet hydrocarbon emissions, Novembers, 1992	33
14. Outlet hydrocarbon emissions, November 4,1992	34
15. Outlet hydrocarbon emissions, Novembers, 1992	35
16. Hydrocarbon removal efficiency, Novembers, 1992	37
17. Hydrocarbon removal efficiency, November 4,1992	38
18. Hydrocarbon removal efficiency, Novembers, 1992	39
19. Average inlet and outlet hydrocarbon and styrene emissions for each period of spraying	44
20. Average hydrocarbon and styrene removal efficiency for each period of spraying	45
21. Total system cost as $/ton of styrene removed for a nine year lifetime for various inlet
    concentrations of styrene. Data for 20,000 and 60,000 scfm inlet flow rates are shown
    along with values for the 200,000 scfm system proposed for Eljer Plumbingware	66
22. Total system cost as $/scfm of inlet flow rate for a nine year lifetime for various inlet
    concentrations of styrene. Data for 20,000 and 60,000 scfm inlet flow rates are shown
    along with values for the 200,000 scfm system proposed for Eljer Plumbingware	67

-------
                                    LIST OF TABLES

Table

1.    Daily Test Conditions	15
2.    Daily Averages and Grand Average of THC Analyzer Data, November 3 through 5,1992	41
3.    Summary of THC Analyzer Data, November 3 through 5.1992	42
4.    Styrene Removal from THC Data, November 3 through 5,1992	*&
B.    Summary of Estimated Styrene Emissions from Gel Coat Booth #2. THC Data	47
6.    Sampling Conditions for Measurements made at the Outlet of Gel Coat Booth #2, Rrst
     Lay-Up Booth # 5, and Second Lay-Up Booth #7,October - November, 1992. EPA Method
     18/NIOSH Method 1501 Sampling	50
7.    Styrene Concentrations Measured at the Outlet of Gel Coat Booth #2, Rrst Lay-Up Booth
     # 5, and Second Lay-Up Booth #7,October - November, 1992. EPA Method 18/NIOSH
     Method 1501 Sampling	51
8.    Results of EPA Method 18/NIOSH Method 1501 Samples Taken at the Inlet of the Polyad
     FB Device, November 3 to November 5,1992	54
9.    Results of EPA Method 18/NIOSH Method 1501 Samples Taken at the Outlet of the Polyad
     FB Device, Novembers to November 5,1992	55
10.  Compounds Identified in Liquid Sample  Recovered from the Polyad FB Device	57
11.  Flow Rate Measurements at the Inlet of the Polyad FB Device	58
12.  Estimated Styrene Capture in the Polyad FB Device	60
13.  Design Assumptions and Price Quotes for the Polyad Concentrator System and the Polyad
     Recovery System for the Eljer Facility	62
14.  Equipment List and Dimensions for Large Components for Polyad Systems Suitable for
     Installation at the Eljer Facility	63
B-1. Data Quality Indicator Goals for Critical  Measurements Estimated in QAPP	82
B-2. Data Quality Indicator Values for NIOSH Method 1501 Measurements Made at
     Eljer Plumbingware	83
B-3. Data Quality Indicator Values for NIOSH Method 1501 Charcoal Sample Tubes
     Spiked with Known Concentrations of Styrene	84
B-4. Data Quality Indicator Values for THC Analyzer Measurements Made at Eljer Plumbingware	85
C-1. THC Analyzer Results from November 3,1992. Rrst Period of Spraying	89
C-2. THC Analyzer Results from Novembers, 1992, Second Period of Spraying	90
C-3  THC Analyzer Results from November 3,1992, Third Period of Spraying	91
C-4  THC Analyzer Results from November 4,1992, Rrst Period of Spraying	92
C-5  THC Analyzer Results from November 4,1992. Second Period of Spraying	93
C-6. THC Analyzer Results from November 4,1992, Third Period of Spraying	94
C-7. THC Analyzer Results from November 5,1992, Rrst Period of Spraying	1""""™"""""" 95
C-8. THC Analyzer Results from November 5,1992. Second Period of Spraying	'..!"!'.!!!!"""!"!""" 97
C-9. THC Analyzer Results from November 5,1992, Third Period of Spraying	I."..""!!".!!".."..!98
                                            VI

-------
                                   ACKNOWLEDGMENTS
       The authors would like to thank J. N. Eason, Plant Manager at Eljer Plumbingware in Wilson,
NC, for providing a site to cany out this evaluation and Rollie Nagel, Manager of Safety and
Environmental Health at Eljer, for his support and help during the test. The authors would also like to
thank Lennart Odbratt and Magnus Danielsson of Weatheriy, Inc. for making available the Polyad® FB
unit for this test and for their help and assistance before, during, and after the test. Finally, the authors
would like to thank EPA Task Officer Bobby Daniel of the EPA Air and Energy Engineering Research
Laboratory for his help, support, and coordination throughout this work.
                                              vii

-------
                                          PREFACE







       The Control Technology Center (CTC) was established by EPA's Office of Research and



Development (ORD) and Office of Air Quality Planning and Standards (OAQPS) to provide technical



assistance to state and local air pollution control agencies.  Three levels of assistance can be accessed



through the CTC.  First, a CTC HOTLINE has been established to provide telephone assistance on



matters relating to air pollution control technology. Second, more in-depth engineering assistance can



be provided when appropriate. Third, the CTC can provide technical guidance through publication of



technical guidance documents, development of personal computer software, and presentation of



workshops on control technology matters.



       The engineering assistance projects, such as this one, focus on topics of national or regional



interest that are identified through contact with state or local agencies.
                                               VIII

-------
Metric to Nonmetric Conversions
Readers more familiar with nonmetric units may use the following factors to convert to that
system.
Metric
kPa
kPa
•c
m
m3
mmHg
kg
1000kg
m3/min
Multiplier
1450.38
4.0145
1.8T + 32
3.2808
35.3134
0.03937
2.2026
0.90802
35.3134
Yields Nonmetric
psig
in. H20
•F
ft
ft3
in. Hg
Ib
ton
ft3/min

-------
                                          SECTION 1
                                       INTRODUCTION

        The control of styrene is a major concern for many areas of the country. Up to the time of this
 evaluation, The Control Technology Center (CTC) had received over 30 calls on the topic. One area of
 styrene emissions is in the manufacture of shower stalls and bathtubs. There are approximately 200 of
 these plants operating in the U.S. emitting uncontrolled styrene to the atmosphere.  Until recently, no
 known technology had been demonstrated to control the emission of styrene.
        The CTC was contacted by the firm of Weatheriy, Inc., of Atlanta, GA, with regard to the
 possibility of evaluating a system (The Polyad® FB Process) that had been developed in Europe and is
 presently being used there to control styrene emissions. This process uses a f luidized bed adsorption
 system with proprietary macro porous polymer particles as the adsorbent.  In the Polyad FB process the
 adsorbent continuously migrates from an adsorption section to a desorption section where the polymer
 particles are regenerated and the solvent is condensed and recovered back to the adsorption section.
       After being contacted by Weatheriy, Inc., the CTC attempted to find other vendors of control
 technology for styrene removal.  One other vendor was found and the CTC contacted this vendor,
 QUAD Environmental Technologies Corporation, of Northbrook, IL. to propose the evaluation of their
 chemical scrubber process (QUAD Chemtact™ System) along with the Polyad FB process on a source
 of styrene emissions. The QUAD Chemtact process removes styrene by spraying fine droplets (a mist)
 of a diluted chemical solution into a contaminated air stream as it is injected tangentially into the top of a
 hollow cylindrical reaction chamber.  Styrene is apparently oxidized and adsorbed into the mist of
 scrubber liquor which is continuously collected and exhausted through the chamber drain. The treated
air is then exhausted tangentially through the bottom of the reaction chamber.

                                                1

-------
       The CTC initiated a proposed project to evaluate the Polyad and Chemtact processes for



controlling styrene emissions at a representative fiberglass shower stall and bath tub manufacturing



plant. Eljer Incorporated of Wilson, North Carolina was selected as a possible site and was visited by



representatives of EPA in August of 1992 and later, in October, by representatives of EPA, SRI,



Weatheriy, and QUAD. This facility was selected because at this site Eljer manufactures both fiberglass



shower stalls and bathtubs by spraying styrene-based resins onto various mold shapes in individual



spray booths that are vented to the atmosphere.  During the August visit, a tentative agreement was



reached to test both the Polyad FB and Chemtact processes on a representative source of styrene



emissions from the shower stall/bathtub construction process.  Subsequently, during testing of the



Polyad FB unit, it was found that QUAD Environmental could not supply the unit for testing within the



time frame allotted for this work.



       Vent air from the spray booths used for mold-coating that is exhausted to the atmosphere is the



major point source of emissions from the manufacture of fiberglass shower stalls and bathtubs. Thus,



the number of manufacturing steps that involve the spraying of styrene-based resins and the number of



individual spray booths in operation at a particular facility determine the level of styrene emitted to the



atmosphere.



       Any fiberglass product that during its manufacture requires the spraying of styrene is a source of



organic vapors that could affect human health both directly and indirectly. The results of this evaluation



will provide information to state and local agencies for use in responding to public concerns.

-------
                                          SECTION 2
                                    PROJECT DESCRIPTION

 2.1   EXPERIMENTAL APPROACH
        Styrene is an integral part of the industrial process that produces fiberglass bath tubs and
 shower stalls. In the first step of this manufacturing process, styrene is mixed with polyester resin and a
 pigment to create a 'gel coat* that is sprayed onto a previously prepared mold. Molds are typically
 reusable and before each use the mold is waxed and coated with a mold-release agent that also helps
 to provide a high gloss to the finished product.  In subsequent manufacturing steps, styrene and
 polyester resin are mixed with inert fillers and sprayed onto the previously coated mold along with
 chopped fiberglass.  Between each application the coated mold is set aside while the resin is allowed to
 cure. Because curing is an exothermic process, the next manufacturing step is usually not carried out
 until the coated mold has cooled. Fiberglass provides structural support for the finished article, styrene
 and polyester resin act as a glue to hold the matrix together, and the inert fillers provide additional
 structural support and can also be used as a fire retardant.  The final step of manufacture is to separate
 the finished fiberglass product from the mold.
        The purpose of this project was to evaluate the performance of two pilot-scale devices designed
 to control styrene emissions.  During this evaluation, each pilot-scale control device was to be
 configured to treat a portion of the air exhaust from a geteoat booth at a fiberglass shower stall and bath
 tub manufacturing plant operated by Eljer Plumbingware located in Wilson, North Carolina. As indicated
 above, of the two devices selected for evaluation, only the Polyad  FB device could  be tested within the
 time allotted for this work.
       To measure the styrene removal efficiency of the pilot-scale Polyad FB device, total
 hydrocarbon  (THC) analyzers equipped with flame ionization detectors (FID) were used to determine
total hydrocarbon levels at the inlet and outlet of the device on a continuous basis while charcoal-filled
sampling tubes were used to collect samples of volatile organic compounds (VOC's) at the inlet and
outlet of the device over time periods ranging from one to two hours. Styrene levels in the inlet and

-------
outlet gas streams were quantified by subsequent chromatographic analysis (with FID detection) of the
VOC's retained in the charcoal-filled sampling tubes.
2.2   ELJER PLUMBINGWARE FACILITY
        The Eljer Plumbingware facility, diagrammatically shown in Rgure 1, is located in Wilson, North
Carolina.  In this figure the location of the pilot-scale Polyad FB unit is shown along with the location of
the van used for sampling and the generator used to provide power for the Polyad device.
        Each stage of manufacture except for mold separation or 'pulling' is carried out in a spray
booth. At the Eljer facility the spray booths were not constructed in place but are prefabricated units
manufactured by Binks, Inc.  Each spray booth is approximately 3.05 m (10 ft) high. 4.11 m (13.5) ft
wide, and approximately 3.66 m (12 ft) deep. The booths are actually several feet deeper but 3.66 m
back from the mouth of the booth an expanded metal grate is mounted across the width and height of
the booth on which a large sheet of air conditioning-type filter material is mounted. The filter material is
usually changed every other day.  Each spray booth is continuously vented with air from the interior of
the plant that is pulled into the booth entrance, through the air conditioning filter mat, to a large fan unit
mounted outside of the building. Air pulled into the fan exits through ductwork that reenters the side of
the building and exhausts vertically through a 0.91 m (3 ft) diameter stack mounted on the roof of the
facility. Each exhaust fan has a nominal rated flow of 411 m3/min (14,500 acfm).
        There are three distinct manufacturing steps that are required to produce a fiberglass shower
stall or bath tub at the Eljer facility. First, a prepared  mold is mounted on a cart and wheeled into one of
the three geteoat spray booths located in the mold repair shop.  In the spray booth, the mold and cart
are designed to slide onto the arm of a permanently mounted pedestal assembly that can be elevated
above the floor of the spray booth.  The mold and cart are also designed to rotate on the arm of the
pedestal so that all parts of the mold are accessible for spraying. This mounting system is duplicated in
every spray booth at the Eljer facility.
        Geteoat is a nominal mixture of 1/3 styrene monomer, 1/3 polyester resin, and 1/3 pigment.  At
the time of this test two colors of pigment  were used: white and pink. About three minutes are required
to coat a bath tub mold with  gelcoat and five minutes are required to coat a shower stall mold with

-------
                                            •190ft-
    ACCESSAREAWAY
    FOR FILTER SERVICE
          FIBERGLASS
          ROVING CHOPPER
174.5ft
           RESIN
           MIXING
           ROOM
          GELCOAT
          STAGING
          AREA
                115 ft
                        SECOND LAY-UP
                        BOOTHM
SECOND LAY-UP
BOOTH*?     •
                        HRST LAY-UP
                        BOOTHM
                        FIRST LAY-UP
                        BOOTHM
                                                   •  MOLD SEPARATING
                                                  ]*T STATION
                                                                      PLANT OFFIC
                                                                         SPACE
                                                            MOLD REPAIR SHOP
                     SAMPLING
                       VAN
                      POLYAD
                       PILOT
                       UNIT
                       GENERATOR
          D
                       Figure 1. Layout of the Eljer Plumbingware Facility

-------
gelcoat. When spraying is complete, the mold is oriented upright and the pedestal is lowered until the
wheeled cart mounted to the mold contacts the floor. The mold and its attached cart is then wheeled out
of the booth to await the next stage of manufacture. Between each stage of manufacture the coated
mold is set aside to cure and harden for about an hour. Curing generates heat so there is a time interval
between sprayings to allow the coated mold to cool.
       The second stage of manufacture is called the first lay-up" or 'initial laminating* step. In this
stage, the mold is conveyed to one of the first lay-up booths and, as with the first step of manufacture,
mounted on a pedestal and prepared for spraying.  In this step, calcium carbonate (CaCO3) powder is
added to a mixture of 50% styrene monomer and 50% polyester resin to form a slurry that contains
approximately 50% solids. This slurry is sprayed onto the mold and during the  spraying operation,
chopped fiberglass roving is also blown at about a 30* angle into the stream of  spray as it exits the spray
nozzle.  The spray mixes with the strands of chopped  fiberglass and forms a entangled mat of resin
impregnated fiberglass on the surface of the mold. The CaCOs powder and the chopped fiberglass help
provide structural support to the finished product. As with the first stage of manufacture, this step is
brief and requires only three to five minutes to complete.  When this step is completed the coated mold
is once again set aside to cure.
       The third, and final, spraying step is called the 'second lay-up' or 'backup' step and takes place
in one of the two second lay-up booths shown in the upper left comer of Rgure 1 (Booth #7 or Booth
#8). In this step, a 60%-40% blend of powdered CaCO3 and hydrated alumina is added to a mixture of
50% styrene monomer and 50% polyester resin to form a slurry that is contains approximately 50%
solids. Hydrated alumina is added as a fire retardant. This mixture is also sprayed with chopped
fiberglass fibers and forms the final layer on the mold. As with the first stage of manufacture, this step is
brief and requires only three to five minutes to complete.  When this step is completed the coated mold
is set aside to cure for the final time.
       The final stage of manufacture is -pulling' or separation of the mold from the completed shower
stall or bath tub. After the finished fiberglass piece is  trimmed and inspected it is prepared for shipment
to its final destination.

-------
2.3     THE POLYAD* FB PROCESS
        The Polyad FB process was developed and is marketed internationally by Nobel Chematur (now
Chernatur Engineering AB) of Karlskoga, Sweden.  Weatherty, Inc., located in Atlanta, Georgia, is a
wholly-owned subsidiary of Chematur International AB, and has the responsibility for domestic marketing
and sales of the Polyad FB process. The pilot-scale Polyad FB device that was evaluated at the Eljer
Plumbingware facility was provided and operated by Weatherty, Inc. who also supplied the information
from which the following description of the Polyad FB process was taken.
        Polyad is a collective name for adsorption processes developed by Chematur AB for the
removal of organic substances from process exhaust air and water. These purification processes are all
based on adsorption onto macro-porous polymer particles instead of activated carbon, which previously
was the adsorption agent of choice for most solvent removal processes.  Chematur indicates that
Polymer adsorbents are superior to activated carbon because polymer formulations can be optimized for
the adsorption of a specific solvent.
        For air purification applications Chematur markets their proprietary macro-porous polymer under
the trade  name of Bonopore®.  It is supplied in the form of off white to pale tan particles with an average
size of 0.5 mm, bulk density of 0.3 g/cm3, and specific surface area of approximately 800 rr^/g. The
high surface area is due to the macro-porous nature of the material. The Material Safety Data Sheet for
Bonopore indicates that it has a molecular weight > 1.000,000 and lists its  components as
divinlybenzene, ethylvinlybenzene, toluene, and water.  Chematur indicates that the adsorption
properties of Bonopore are unaffected by high humidity and that the polymer has no catalytic effect on
solvents such as styrene. It also resists abrasion which is important as the Bonopore particles are
generally  used to constitute a bed that is fluidized with solvent-laden air.  The amount of Bonopore
required for a given application is governed by total air flow through the bed,  the amount of solvent
present in the air stream, the adsorption rate of the solvent onto the Bonopore polymer, and the desired
solvent removal efficiency.
       The Polyad FB (fluidized bed) air purification and solvent  recovery process utilizes a
continuously fluidized bed for air purification and the recovery of solvents from air. Chematur claims that

-------
this process is especially well suited for recovery of solvents from air streams when the solvents have



relatively high boiling points. Typical solvents that are suitable for removal with the Polyad FB process



include aromatic and aliphatic hydrocarbons, chlorinated hydrocarbons, certain freons. alcohols.



aldehydes, ketones, and esters. Organic solvents with tow boiling points and very polar solvents, such



as methanol or methylene chloride, are not suitable candidates for removal with the Polyad process.



        At this time, two Polyad FB processes are marketed. The first of these processes is the one



that was tested at Eljer Plumbingware where solvent is recovered for reuse or disposal. In this process



(Rgure 2) solvent-laden air is purified as it passes through a fluidized bed of  Bonopore adsorbent. The



saturated Bonopore adsorbent is then pneumatically conveyed to a desorption unit where  it is



regenerated.



        The adsorption section consists of two, or more, beds of Bonopore.  The number of beds



depends on the type and concentration of the solvent and on the degree of solvent removal required.



Solvent is adsorbed by the polymer particles as air passes through and fluidizes each bed. Bonopore



continuously flows from one bed to the next. In the last bed the adsorbent, now saturated with solvent,



is removed from the bottom of the bed and pneumatically conveyed to the desorption section at the



same rate as regenerated Bonopore is fed into the first bed.



        Bonopore from the adsorption section is pneumatically conveyed to the top of a vertical moving



bed desorber or 'stripper" for regeneration.  The desorption section consists of a container mounted



above the stripper unit comprising a specially designed heat exchanger with  gas collector. As particles



of Bonopore descend through the desorber they are heated in a heat exchanger to a temperature at



which the solvent evolves. The heating medum is steam but, by using a heat exchanger,  steam and



water do not come in contact with the recovered sorbent or the Bonopore polymer. The vacuum created



in the container at the top of the desorption section by the pneumatic transport fan causes air to be



drawn in through an air intake at the bottom of the stripper.  It is then drawn  upward through the



descending adsorbent.  This air, together with the released solvent, is collected and directed into a



condenser. Condensed solvent is captured in a tank.  Air containing any uncondensed solvent is fed



back into the top of the stripper and allowed to ascend again through the heated  desorption zone.

-------
     CLEAN AIR
Figure 2. Diagrammatic representation of the Polyad FB Air
         Purification and Solvent Recovery Process.
                         9

-------
       Once past the heated section of the stripper, adsorbent is cooled by a water cooled heat
exchanger situated at the bottom of the stripper. Ordinary tap water is usually sufficient to provide for
satisfactory condensation of the recovered solvent. Regenerated adsorbent is pneumatically conveyed
from the bottom of the desorber to the top of the adsorber into the first fluidized bed, to complete the
cycle.
       In a properly designed unit, essentially all of the solvent entering the unit will be adsorbed onto
the Bonopore polymer before the process air stream carrying the solvent exits the Polyad FB unit.
Therefore, outlet emissions are primarily due to sorbent 'bleeding" from Bonopore polymer as it reenters
the adsorber after being discharged from the desorber.  This is because desorbed Bonopore is
introduced back into the adsorption section at the last fluidized bed before the system outlet. Weatheriy
engineers emphasize that such outlet emissions can be minimized by proper design and are acceptable
in light of the gain in design simplicity and reliability afforded by the use of gravity to aid in the feeding of
Bonopore polymer from one fluidized bed to the next.
       The second type of Polyad FB process is used in cases where solvent recovery is not required.
In this case, the Polyad FB process is used as a preconcentrator to increase solvent concentration in
the exhaust stream so that it can be  easily incinerated.  The net effect of such preconcentration is to
reduce the total volume of air in the exit stream while increasing the concentration of solvent. For this
application, the fluidized bed adsorber is unchanged but instead of a moving bed desorber a fluidized
bed desorber is used.  In the fluidized bed desorber the Bonopore is regenerated by blowing hot air
through the adsorbent (in a fluidized  bed) which causes the solvent to evolve. As with the first process,
regenerated adsorbent is pneumatically conveyed back to the first adsorber. The hot solvent/air mixture
leaving the desorber is passed to an incinerator where it is burned. To reduce energy consumption, a
heat exchanger can be installed to recover waste heat at the incinerator outlet.
       Among the  Polyad FB units that have been installed, two were installed specifically to remove
and recover styrene. One of these units was installed at the IFO-Sanitar AB plant in Bromolla. Sweden
in 1989. This unit treats 21,200 rrrVhr of air and can recover up to 8 kg/hr of styrene with a 90% overall
                                                10

-------
efficiency. This unit requires 52 kW of electric power, O.Srr^/h of cooling water and 150 kg of Bonopore
catalyst per year.  Another styrene removal/recovery unit was recently ordered (in Germany).
2.3.1   Pilot-Scale Polyad FB Device
     .   The Polyad mobile unit used at the Eljer Plumbingware facility was designed to accommodate
an air flow of up to 500 m3/h (294 acfm). With this unit configured as it was for this evaluation, a
maximum steady state concentration of 1000 ppm of styrene could be accomodated before the level of
outlet emissions would be affected. Short-duration peak concentrations of up to 3000 ppm of styrene
could also be experienced before outlet emissions would be expected to increase. Solvent removal
efficiency was expected to be greater than 90%. A schematic diagram of the pilot unit is shown in
Figure 3 and a component arrangement drawing of the pilot unit is shown in Figure 4.
        This portable unit consists of three separate modules designed to be assembled on site after
being shipped there by truck. The total weight of the assembled package is 2000 kg and once set up it
occupies a space 3.8 m long by 2.0 m wide by 5.0 m high. During operation the pilot-scale Polyad FB
unit requires 45 kW of 480 VAC of electrical service, primarily for steam generation, fan power, and
cooling water refrigeration. A separate generator is usually used to provide this service.  The design
cooling water pressure is 5 bar with a design volume flow of 200 l/h.
        The pilot-scale unit uses four fluidized beds (two sections of two fluidized beds). Air that enters
the unit passes first through a filter to remove foreign objects and then into the first fluidized bed.  The
filter housing also holds a heat exchanger to cool water used to condense recovered solvent.  As shown
in Figure 3, flow is upward. Clean air is passes out of the top of the adsorber unit through a cyclone to
the atmosphere. The cyclone catches any Bonopore adsorbent that escapes and feeds it back to the
third fluidized bed.
        Bonopore that has passed through the four fluidized beds is pneumatically conveyed to the top
of the moving bed desorber unit.  Excess air from the conveying process is routed back to the first
fluidized bed at the inlet of the adsorber unit along with fluidizing air that is injected at the bottom of the
desorber. The desorber incorporates two heat exchangers that are designed so that evolved solvents
and the Bonopore adsorbent do not come into contact with the source of heating or cooling.  The first

                                                 11

-------
AIR INLET      AIR OUTLET
                                                           DESORBER UNIT
                                                                                                   FAN
                                                                           STEAM M
                                                                                      COOLMd
                                                                                       WATER
                                                                           STEAM OUT


                                                                        OOCH.MQ WATER
                                                                                  AIR
                                                                                                SOLVENT
                                                                                                OUTLET
                       ADSORBER UNIT
         Figure 3. Schematic diagram of the pilot-scale Polyad FB unit tested at Eljer Plumblngware

-------
                                 2.040m-
I  STEAM  |
IGENERATOR)
       CONTROL
   j   CABINET
|CHUER
                    8L-2_
                                                                  FAN
                                                                             FAN
                                                                     PPIMCP*! eQUPOMPHTS PC TUP UHOI t PQLYtP UNIT-

                                                                     t. MLETPLENUU.NCORPORATESFI.TERANO
                                                                       HEAT EXCHANGER FOR SOLVENT RECOVERY.

                                                                     2, auraZEDBEDADSOHBER. KETATBOTTOU

                                                                     1 CYCLONE. RECOVERS BONOPORE ADSORBENT THAT
                                                                       ESCAPES FROM THE LAST FUlOOEOBEDi AIR 06-
                                                                       CHARGED FROM CYCIONETO ATMOSPHERE. CYCLONE
                                                                       CATCH ROUTED BACK TO THWD FLUKED BED.

                                                                     4. DESORBER. BONOPORE ADSORBENT FROM OVOET
                                                                       FLUIDZED BED IS PNEUMATKALLY CONVEYED TO THE
                                                                       TOP OF THE DESORBER WHERE IT B HEATED TO DE-
                                                                       SORB SOLVENT. REGENERATEOADSORBENT B
                                                                       PNEUMATCAaY CONVEYED TO THE OUTLET
                                                                       FUWXZEDBED.
Rgure 4.  Component layout of the pilot-scale Polyad FB unit tested at Eljer Plumbingware
                                                    13

-------
(three-section) heat exchanger uses 135*C steam to evolve adsorbed solvent from the Bonopore



polymer. The second (single-section) heat exchanger uses tap water to cool the Bonopore adsorbent



back to room temperature (nominally 20 - 25'C) before it passes to the bottom of the desorber and is



pneumatically conveyed back to the last section of the fluidized bed adsorber.  The average time



required for the Bonopore adsorbent to make one complete circuit of the system is one hour.



        In the desorber, air and solvent evolved from the heated bonopore polymer is drawn from



between the second and third section of the first heat exchanger and is sent to a separate water-cooled



solvent condensing unit.  Air that has passed through the condensing unit is returned to the inlet of the



desorber to mix with incoming Bonopore adsorbent. Although tap water was used for this test, if it is



required, an on-board refrigeration unit can supply chilled water (at 10'C) to the heat exchanger in the



solvent condensing unit.



2.3.2   Specific Test Conditions



        For this test the inlet air flow rate varied from 4.5 to 6.1 rrrVmin (159 to 215 acfm) at a pressure



drop of from -2 to -2.5 kPa (-8 to -10 in. H20). The flow rate of Bonopore through the fluidized beds was



kept at 10 kg/h (5.6 l/m, the lowest flow rate that can be maintained through the unit), except for one



day (November 4) when the flow was set at 17 kg/h (9.4 l/m).  The temperature of the desorber was set



at 135*C (275*F), which is near, but below, the boiling point for styrene monomer (146*C). The tap



water used for cooling the Bonopore adsorber and the heat exchanger in the solvent condensing unit



averaged 20" C throughout the test.  Table 1 summarizes the test conditions for each day of testing.



        During analysis of VOC samples collected at the inlet and outlet of the pilot unit it was



determined that while only styrene was present at the inlet of the Polyad FB device, substantial amounts



of compounds other than styrene (mainly napthalene and napthalene derivatives) were present at the



outlet.  Through conversations with Weatheriy. Inc. it was learned that this pilot unit had recently been



tested at a naptha manufacturing plant and that the Bonopore adsorbent used for that test was also



used at Eljer. At the end of that test no special attempt was made to clean the unit or to desorb naptha



from the Bonopore adsorbent. At the beginning of this test VOC emissions at the outlet of the Polyad



FB device were approximately 22% styrene.  By the end of the test VOC emissions at the outlet of the





                                               14

-------
                                 Table 1.  Daily Test Conditions*
Date
3 November
4 November
5 November
Inlet Air Temp.
CO
20-25
20-25
20-25
Ret. Humidity
(%)
N/A
62-66
N/A
Inlet Air Flow
(m*/mm)
4.47
4.81
5.63
Bonopore Flow
(kg/h)
10
17
10
Desorber Temp.
CO
135
135
135
*   System pressure drop of from -2 to -2.5 kPa (-8 to -10 in.
                                                15

-------
Polyad FB device were approximately 85% styrene. A sample of recovered solvent obtained from the



solvent condensing unit at the end of the test was approximately 26% styrene.



2.4    EXPERIMENTAL APPARATUS



2.4.1   Connection to the Pilot-Scale Polvad FB Unit



       As is shown in Figure 1, the pilot-scale Polyad FB unit was installed on the outside of the plant



as near as possible to the exhaust from gelcoat booth #2. Also shown in this drawing are the relative



locations of a generator used to provide electrical power for the pilot unit and a small van used to house



the sampling equipment used for this test. Rgure 5 shows the overall arrangement for sampling.



Approximately 19.5 m (64 ft) of flexible 15.24 cm (6 in.) diameter aluminum ducting was used to connect



the Polyad FB unit on the ground to the exhaust of gelcoat booth #2 on the roof of the plant



(approximately 4.6 m above the ground). Because outside temperatures were moderate (20 - 25"C) the



ducting was not heated or insulated. A 0.9 m (3 ft) section of the same flexible aluminum ducting was



attached to the outlet of the Polyad FB unit. All connections for the sampling equipment were made at



the actual inlet and outlet of the pilot unit through rigid duct adapters and bulkhead fittings acquired for



this test.



       At the pilot unit inlet and outlet single 9.53mm (0.375 in.) diameter Teflon® sample lines were



used to carry gas samples to the sample van for analysis. Both were about 2.13 m (7 ft) long.  An



unpublished EPA-sponsored styrene sampling effort determined that Method 18 sample lines did not



need to be maintained near the boiling point of styrene but did need to be held at least 20* F above the



local ambient temperature.  Thus, the sample lines were maintained at a temperature of 49* C (120'F).



2.4.2   Sampling Van



       Rgures 6 and 7 show how the gas sampling equipment was connected within the van used to



house the sampling equipment. Separate equivalent systems were constructed so that concurrent



samples could be obtained  at the inlet and the outlet of the Polyad FB device.  Thus, the description that



follows applies to either system.



       Shortly after each 9.53mm diameter sample line entered the sampling van it was divided into



two short 6.35 mm (0.25 in.) diameter sample lines less than 1 m in length. The smaller sampling lines





                                               16

-------
       IMliiiyiii
                                 I ROOF OF EUER FACILITY
                              GELCOAT BOOTH
                               EXHAUSTFANS
SAMPLING
  VAN
                 OUTLET
                 SAMPLE
                              FLUID BED
                             ADSORBERS
DESORBER
  UNIT
                                                           .
                                                      GELCOAT
                                                      BOOTH
                                                      EXHAUST
              POLYAD®FB
              PILOT UNIT
     Rgure 5. Overall arrangement for sampling at the Eljer Plumbingware Facility
                                    17

-------
                                INLET OR OUTLET
                                HEATED SAMPLING LINE
                         TO EPA METHOD 18
                         (NIOSH METHOD 1501)
                         SAMPLING APPARATUS
  OUTPUT TO DAS AND
  CHART RECORDER
EXHAUST
TO OTHER THC ANALYZER
SPAN AND ZERO GAS INPUTS
  I
  J.U.M. VE-7
  THC Analyzer
                             ZERO
                              AIR
  25
  ppm
49
ppm
171
ppm
                 TO OTHER THC
                   ANALYZER
                                      STYRENE CALIBRATION GAS
   Figure 6.  Equipment arrangement used lor sampling with THC analyzers
                             18

-------
                              INLET OR OUTLET
                              HEATED SAMPLING LINE
                        TO THC ANALYZER
   CHARCOAL TUBE
   CHARCOAL TUBE
   CHARCOAL TUBE
                                          EXAUST
                                      SAMPLING
                                        PUMP
                               FLOW
                            REGULATING
                              MANIFOLD
                             BYPASS PUMP
             £P
Figure 7. Equipment arrangement used for EPA Method 18 / NIOSH
       Method 1501 sampling
                            19

-------
were not heat traced.  One line was connected to a J.U.M. Instruments VE-7 THC analyzer and the



other line was connected to three Swagelock™ Tee' connectors in series from which samples of gas



could be withdrawn into charcoal-filled adsorption tubes (EPA Method 18. Section 7.4 or NIOSH Method



1501).  After the last Tee connector this sample line was connected to a small 28.3 l/m (1 cfm) capacity



sampling pump that was operated at approximately one-third of its rated flow. This pump served two



purposes. The first purpose was to assure that more sample was withdrawn from the flexible aluminum



duct than was required by the THC analyzer and the adsorption tube sample pump.  The second



purpose was to minimize the time required to convey a gas sample to the sampling equipment.  The



exhaust from this pump was diverted to the outside of the sample van.



       Rgure 6 shows the calibration gas system used for the THC analyzers. Three mixtures of



styrene in nitrogen were used for calibration (171 ppm, 49 ppm, and 25 ppm), in addition to zero air.



The bottles of calibration gas were interconnected with positive shut-off valves to a common manifold



that was itself connected to the span gas port on both THC analyzers. This system allowed both THC



analyzers to be calibrated from the same calibration gases.



       The 0 -10 V output signal from each THC analyzer was fed to one channel of a two-channel



chart recorder. The output signal from each THC analyzer was also recorded on 720 kb floppy disks



with a dedicated PC-based datalogger. Output from each THC analyzer was logged once every second.



The software used to log the data (Labtech Notebook) was configured to display the last 50 minutes of



data (last 3000 data values) from both channels on the PC monitor.  Output from the inlet THC analyzer



was displayed from 0 to 1000 ppm full scale and output from the  outlet THC analyzer was displayed on



aOto 100 ppm scale.



       Preliminary measurements with charcoal-filled adsorption tubes (using NIOSH Method 1501,



equivalent to the Adsorption Tube Procedure in Section 7.4 of EPA Method 18) made on a pre-test



survey trip to Eljer Plumbingware revealed that virtually all (99%) of the organic material exhausted from



the gelcoat spray booths was styrene monomer. During this test similar measurements were also made



at the inlet and outlet of the Polyad FB device to quantify what organic compounds were present at the



inlet and exhaust of this device. The primary intent  of these measurements was to establish time-





                                              20

-------
 averaged levels of styrene at the inlet and outlet of the Polyad FB device to determine the styrene



 removal efficiency of the Polyad FB device.  As with the pre-test visit. NIOSH Method 1501 was followed



 in obtaining these samples (see Appendix A).



        Figure 7 shows the equipment arrangement used for the NIOSH Method 1501 sampling. The



 same arrangement was used to obtain inlet and outlet samples.  Provision was made to obtain three



 replicate samples at one time by taking a sample at each tee in the sample line that led to the bypass



 pump.  The three samples were obtained with a single sampling pump connected to a manifold that, in



 turn, was connected to each tee through a standard small charcoal-filled tube. Flow through each leg of



 the manifold was set before each measurement to approximately 0.2 l/m.



 2.5    EXPERIMENTAL METHODS AND PROCEDURES



 2.5.1   Total Hydrocarbon Analyzers



        J.U.M. Instruments Model VE-7 total hydrocarbon (THC) analyzers equipped with flame



 bnization detectors (FID) were used to obtain a continuous measurement of the total hydrocarbon



 content in air that entered the Polyad FB device (air exhaust from gelcoat booth #2) and exited the



 Polyad  FB device. This analyzer extracts approximately3 l/m of sample with an internal sample pump



 and sends from 17 to 20 cm3/m of that sample to an onboard FID. The FID's in these instruments were



 set up to use hydrogen as a fuel. Two of these THC analyzers were rented from Clean Air Engineering



 of Palatine, IL for the duration of testing.  These instruments were inspected and calibrated with propane



 span gases before shipment.



        Five decade output ranges can be selected from 0-10 ppm to 0-100,000 ppm. A 0-10V signal is



 output at the rear of the instrument that corresponds to the decade range selected. The instrument



 used to  sample air from the inlet of the Polyad FB device was set to measure in the 0-1000 ppm range



 and the  instrument used to monitor air exhausted from the Polyad FB device was set to measure in the



 0-100 ppm range. As indicated above, the output from each of these instruments was recorded on a



two-channel chart recorder and also logged on a dedicated PC-based data acquisition system.



        These instruments are normally calibrated with propane.  However, for this test they were



calibrated with three mixtures of styrene in nitrogen (171 ppm, 49 ppm, and 25 ppm. certified by





                                              21

-------
Matheson® Gas Products. Inc) in addition to zero air (less than 0.1 ppm hydrocarbons).  When the



instruments were zeroed on zero air. Instrument response was linear with the three calibration gases.



       The instruments were calibrated and operated according to the manufacturer's instruction



manual.  Calibration and zero gas connections were made at the back panel of the instrument and



calibration and zero gas pressures were maintained at 1 bar (15 psig).  Fuel gas (hydrogen) pressure



was maintained at 1.5 bar (21 psig).  Sample pressure was maintained at 200 mbar (3 psig). Full



calibrations (all span gases, zero gas) were performed on both THC analyzers at the beginning and



middle of each day and instrument calibration  was checked at the end of each day of testing.



2.5.2  Collection of Volatile Organic Compound Samples



       The Adsorption Tube Procedure defined in Section 7.4 of EPA Method 18 (equivalent to NIOSH



Method 1501) was followed to obtain samples of volatile organic compounds (VOCs) from air that



entered the Polyad FB device (air exhaust from gelcoat booth #2) and exited the Polyad FB device.



NIOSH Method 1501 was followed for the analysis of the samples. A copy of the NIOSH procedure is



included  in Appendix A.



       As shown in Rgure 7, the heated inlet and outlet sampling lines were divided after reaching the



van housing the sampling equipment. One side was directed to a THC analyzer and the other side to a



bypass pump through three tee fittings where  VOC samples were taken. Flow through the bypass pump



was set (by restricting an internal bypass loop between the inlet and outlet of the pump) at a level only



high enough to provide a small excess flow when the THC analyzer and adsorption tube samples were



taken.  Typically, the THC  analyzer required 3 l/m and the three adsorption tube samples required a



total of 0.6 l/m.



       Three samples were taken at a time (to provide replicate samples, one sample from each tee)



and concurrent measurements were made at the inlet and outlet of the Polyad FB  device.  The three



samples were obtained with a single sample pump connected to a flow-regulating manifold. This



manifold is designed so that up to three samples can be obtained with a single sample pump (SKC



Model 224-26-03 adjustable low flow controller). Provision is made for internal regulation of each



sample flow.  At each tee,  a standard small  charcoal tube (SKC Model 226-01 charcoal-filled tube,






                                              22

-------
 NIOSH approved, Lot 120) was connected to one side of the tee with the other side connected to the
 flow regulating manifold.
        Before sampling commenced flow through each leg of the manifold was set to approximately 0.2
 l/m. This was done by inserting a 'dummy' small charcoal tube into the flow path after a bubble flow
 meter (Gilian Instruments Primary Flow Calibrator. 20 crrrVmin to 6 l/m). A dummy tube could be used
 because the flow resistance properties of the NIOSH-approved charcoal-filled sample tubes are
 designed to be uniform from tube to tube. After calibration was completed, the dummy tubes were
 replaced with previously numbered charcoal-filled tubes. To prevent contamination, these sample tubes
 are designed so that a small glass seal on either end of the tube must be broken off before a  gas
 sample can be pulled through the tube. Samples were taken over time periods ranging from one to one
 and one-half hours. When sampling ended each tube was sealed with a plastic cap provided by the
 manufacturer. Previous experience at SRI has shown that when styrene is sampled, these tubes do not
 require refrigeration to preserve the sample prior to analysis. Thus, they were kept at room temperature
 until their contents were extracted for analysis.
        The charcoal-filled sample tubes from this evaluation were returned to SRI's laboratories in
 Birmingham, Alabama for analysis.  Analysis consisted of desorption of VOC's adsorbed on the charcoal
 with carbon disulfide  (according to NIOSH Method 1501, reproduced in Appendix A) followed by injection
 into a gas chromatograph (GC) coupled to an  FID. In addition, several styrene standards were used to
 spike randomly selected charcoal-filled tubes and these samples were analyzed according to the NIOSH
 method to determine a desorption efficiency specific to this lot of charcoal-filled tubes (91.23%). From
 this analysis, styrene present in the samples was quantified. Knowledge of the amount of styrene
 present, the sample time, and the sample gas flow rate allowed the determination of a time averaged
 value for the styrene present at the inlet and outlet of the control device.
 2.5.3   Collection of  Recovered Solvent Sample
        A sample of liquid desorbed from the Bonopore adsorbent (2.6 liters) was collected at the end of
the test from the reservoir in the Polyad FB device. The liquid had the approximate color of new motor
                                               23

-------
oil but a viscosity more like that of light cooking oil.  The liquid had a pronounced unpleasant odor, not
necessarily that of styrene monomer.
       The liquid was brought back to SRI's Birmingham, Alabama laboratories for analysis. A sample
of the liquid was diluted with carbon disulfide and subjected to GC-FID analysis to determine styrene
content.  A second sample of the liquid was subjected to GC analysis with mass spectrographic
detection to determine what other compounds were present.
2.5.4  Total Flow Rate Measurements
        During each day of testing, the total flow rate into the Polyad FB unit was measured with a
thermal anemometer that had been calibrated in a wind tunnel at SRI's laboratories in Birmingham,
Alabama before being taken into the field. These measurements were made at the inlet of the Polyad
FB device at the end of a long section of straight ducting (2.5 to 3 m in length).  The inside diameter of
the flexible aluminum ducting was measured and found to be 14.6 cm (5.75 in.) which corresponds to an
area of 167.5 cm3 (0.180 ft2). To  measure flow, air velocity measurements were made at five evenly
spaced points across the small duct, the measurements were repeated (except on the first day of
testing), and the readings were averaged. The averaged air velocity measurement was then converted
to volumetric flow. The daily average results of these measurements are shown in Table  1.  More
information on these measurements is presented in Section 3.4 of this report.
                                               24

-------
                                         SECTION 3
                              DATA. RESULTS. AND DISCUSSION

3.1  TOTAL HYDROCARBON ANALYZER DATA
      The THC's were operated continuously through the three days of testing. On the last day of
testing a 15 minute power failure at 1:18 pm resulted in the loss of all data to the datalogger for the
balance of the day. Data were reconstructed from the chart recorder record (which itself was lost while
the power was off) by scanning the inlet and outlet record into computer files that were analyzed by
specialized software to determine the numerical value of the data recorded during that time. The
recovered data were then imported into the software package used to generate the experimental record
presented in this report.
      Some operational problems were encountered with the THC monitors.  On the morning of the first
day of sampling, November 3.1992. the THC used to monitor hydrocarbon emissions at the inlet of the
Polyad FB device experienced an abrupt shift in output (by a factor of 1.75).  The exact reason for this
behavior was not known but from an examination of the chart recorder record it was clear that the shift
occurred shortly after the first calibration of the morning. It was detected (and corrected) at the
lunchtime calibration. A correction factor was determined and the morning's data were corrected during
analysis of the data. The THC at the outlet of the Polyad FB unit did not  exhibit this behavior.  The fact
that both of these analyzers were calibrated at the same time with the same calibration gas allowed the
problem to be quickly isolated  and corrected. The same behavior (in the inlet THC analyzer) was
observed one other time, during the calibration check at the end of the day on November 3. The
problem did not appear again and in this instance did not affect any results.  Afterwards, the behavior of
this monitor was closely monitored.
     The THC analyzer used to monitor hydocarbon emissions at the outlet of the Polyad FB device
experienced problems with FID flame-outs.  The problem worsened until  on November 5, technicians at
Clean Air Engineering, Inc. provided guidance on how to correct the problem and the problem was fixed.
                                               25

-------
Because these periods were short and because the instruments were monitored closely no significant
data were lost.
3.1.1  Inlet Data
      VOC emissions from the spraying process can be characterized as being quite variable. At the
inlet of the Polyad FB device instantaneous hydrocarbon emissions (essentially 100% styrene) ranged
from as low as 90 ppm to as high as 400 ppm during spraying in the gelcoat booth. While molds were
being removed from the spray booth or installed in the spray booth hydrocarbon emission levels ranged
from 18 ppm to 45 ppm.  During midday lunch breaks in the production process, hydrocarbon levels
decreased to approximately 9 ppm.
      Figures 8 through 10 show output from the inlet THC analyzer that was recorded on the
datalogger for the three days of testing, November 3 through 5, 1992. For clarity. THC data taken
during periods of calibration are not shown.
      These figures show that there are three distinct "periods" of spraying per day. The first period
lasts from the start of spraying in the morning (starting at about 0845) and ends when the plant
employees have a 15 minute break at approximately 1000 hours.  The second period starts at
approximately 1015 and lasts until.the lunch break at 1145. The final period starts around 1230 and
lasts until 1415.
      In order to show the variability and the periodic nature of the emissions from this process, inlet
THC data from two 40 minute periods before the lunch break on November 5 are presented in Figures
11 and 12. Because these data have a one-second resolution in time, emissions from the spraying
process can be resolved in some detail.  With the exception of the second period of spraying shown in
Figure 11. the operator spent from four to five minutes spraying gel coat on the mold. In the case of the
second spraying period in Figure 11  and the periods of spraying shown  in Figure 12 the operator spent
from five to six minutes spraying gel coat on the mold and the pattern of emissions appears distinctly
different between the two types of spraying. Whether these differences are due to differences in the
type of mold (bath tub versus shower stall) or to the approach used by the operator cannot be
determined because no attempt was made to track what type of mold was being sprayed.

                                               26

-------
I

*-T



I
O
O





I
O
      500
      400 ~-
300 --
      200 - -
      100
          8:30
                 9:30
10:30        11:30        12:30



          Time of Day, 11/3/92
13:30
14:30
                         Figure 8. Inlet hydrocarbon emissions, November 3,1992

-------
                       400
IV)
CD
                           8:30
9:30         10:30         11:30         12:30         13:30          14:30

                     Time of Day, 11/4/92
                                          Figure 9.  Inlet hydrocarbon emissions, November 4, 1992

-------
to
10
                E
                Q.
                Q.
                C
                O

                O


                O
                S3

                rt
                         8:00
9:00
10:00       11:00       12:00

        Time of Day, 11/5/92
13:00
14:00
                                       Figure 10.  Inlet hydrocarbon emissions, Novembers, 1992

-------
                     350
                E
                Q.
                Q.
                0)
                £
                O
                O


                O
                JO

                c5
                o
                2
0>
O
100
                     50
                         9:10     9:15      9:20     9:25      9:30     9:35

                                                       Time of Day, 11/5/92
                                                             9:40
9:45
9:50
                               Figure 11. Inlet hydrocarbon emissions from 0910 to 0950, Novembers, 1992

-------
400
11:10    11:15
                      11:20     11:25    11:30    11:35
                                  Time of Day, 11/5/92
11:40    11:45
11:50
          Figure 12. Inlet hydrocarbon emissions from 1110 to 1150, November 5,1992

-------
3.1.2 Outlet Data
      Figures 13 through 15 show output from the outlet THC analyzer that was recorded on the
datalogger for the three days of testing. November 3 through 5,1992. As with THC data from the inlet.
periods of calibration are not shown. Also not shown are data from periods when the FID in the outlet
THC experienced a flame-out.
      These data show that outlet emissions from the Polyad device are essentialy uncoupled from inlet
emissions. This behavior should be expected because the pilot-scale Polyad FB unit was designed to
remove up to 1000 ppm of styrene before outlet emissions would be  expected to increase. Thus, outlet
emissions observed during this test were mainly due to the "bleeding" of solvent from Bonopore polymer
that was not completely evolved in the desorption section.
      Outlet emissions tended to decrease throughout the day.  At the start of the test, 9 am on
November 3, outlet emissions were near 22 ppm.  Outlet emissions dropped to near 4.5 ppm after
several hours and remained at that level for the balance of the day. On November 4, at the start of
testing outlet emissions averaged nearly 15 ppm.  During the day the emissions level dropped to about 6
ppm and thereafter increased slowly reaching approximately 8 ppm by the end of the day.  This rise may
have been due to the higher flow rate of Bonopore adsorbent used during that day (17 kg/h as opposed
to 10 kg/h on other days). On  November 5, the emissions level  started out at approximately 7.5 ppm.
However, at approximately 0925 outlet emissions were observed to rise quickly to about 11 ppm before
they dropped back to about 5.5 ppm by 1030. The reason for this "spike" is not known because at that
time the inlet emissions showed no such increase, as is shown in Figure 12. By the end of testing outlet
emissions were in the range of 3 to 4 ppm.
     The results shown in these figures are for total hydrocarbon emissions. Previous measurements
(and other measurements to be reviewed here) have shown that virtually all of the VOC emissions from
the Eljer facility occur as styrene.  As indicated in Section 2.3.2, ether VOC's were present in air at the
outlet of the Polyad FB device  because the device was contaminated with VOC residue from a previous
test. However, because these contaminants did not contain styrene, it was possible to determine the
portion of the outlet emissions  that were due to styrene from analysis of the charcoal tube samples.

                                               32

-------
a

I

I

I
cd
o
**
Q>
9

O
     25.0
     20.0
      15.0
      10.0
      5.0
      0.0
          9:00
                         10:00
                                                            TUC Analyzer
                                                             Calibrations
                                    22% Styrene

                                    Draeger Tube
                                     Runt 7 - g
                                    (13:00 - 14:02)
11:00          12:00          13:00

        Time of Day, 11/3/92
14:00
15:00
                         Figure 13. Outlet hydrocarbon emissions, Novembers, 1992

-------
      15.0
 I

 f    10.0
 4-»

I
 o
       5.0
0)
73

O
       0.0
          8:30
                                   THC Analyzer
                                    Calibration
26% Slyrene

Draeger Tube
Runs 13- IS
(9:05 • 10:05)
                                          THC Analyzer
                                           Calibration
                                            61%Styrene

                                           Oraeger Tube
                                            Runs 19-21
                                           (12:40 -14:10)
  9:30
10:30           11:30           12:30           13:30

         Time of Day, 11/4/92
14:30
                           Figure 14.  Outlet hydrocarbon emissions, November 4,1992

-------
en
                          12.0
                          10.0 --
                    I
o
O

i
CO
o
I
>»
CD
*3
O
                                                                THC Analyzer
                                                                 Calibration
                                                                       THC Analyzer
                                                                        Calibration
                                                                          Draeger Tube
                                                                           Run* 31 - 33
                                                                          (10:25-11:30)
                                                                                          Draeger Tube
                                                                                          Run* 37 - 39
                                                                                          (12:42 -14:02)
Oraeger Tube
 Runs 25 - 27
(8:52 -10:02)
                                8:00
                           9:00
          10:00         11:00         12:00

                     Time of Day, 11/5/92
13:00
14:00
                                                Figure 15. Outlet hydrocarbon emissions, November 5,1992

-------
Thus, in Rgures 13 through 15, periods are shown during which charcoal tube (Draeger tube) samples
were taken at the outlet of the Polyad device and the percentage of styrene measured in those samples
is indicated on these figures. Taking into account the actual amount of styrene that was present at the
outlet of the Polyad FB device, styrene emissions at the outlet of the Polyad FB device were never
greater than 8 ppm. By the end of testing emissions of styrene had fallen to less than 4 ppm.
3.1.3 Efficiency Data
      Rgures 16 through 18 show calculated hydrocarbon penetration through the Polyad FB device
and removal efficiency across the Polyad FB device for the three days of testing, November 3 through 5,
1992.  Because of the cyclic nature of the inlet emissions the calculated penetration-efficiency curves
shown in these figures exhibit the same cyclic nature. To provide the reader with visual information on
average levels of penetration and efficiency over time so that trends can be seen, average penetration-
efficiency values during each period of spraying are also shown in these figures along with curve fits
through the peak average values during each period of spraying. As indicated above, outlet emissions
appear to be essentially uncoupled from inlet emissions. Thus, the highest penetrations (lowest
efficiencies) were observed during periods between sprayings and the lowest penetrations (highest
efficiencies) were observed during periods of spraying. In terms of penetration-efficiency values
measured during periods  of spraying, efficiencies never fell below 91% and on the last day of testing an
efficiency of over 99% was measured during one period of spraying at around 1300.  Clearly, from the
data collected during this  period, the  Polyad FB process was very efficient in collecting the styrene
emissions.
      Because of the rapid variability in process inlet emissions and the level of inlet emissions it was
not possible to confirm the assertion  made by Weatherly, Inc. that any stepwise increase in inlet
emissions below 1000 ppm  would not affect the immediate outlet emissions level for the Polyad FB pilot
unit. However, the inlet and outlet emissions data presented above do suggest that within the levels of
instantaneous emissions monitored at the inlet to the Polyad FB device (up to 400 ppm of styrene) outlet
emissions are governed by "bleeding" of solvent from Bonopore polymer as it is reintroduced into the
adsorber from the desorption section.

                                                36

-------
                                                                                      99.0
                                                                   Pen. / Eff.

                                                                   Peak Average
                                                                                      90.0
                                                                         o
                                                                         o
                                                                         to

                                                                         3-
                                                                         o


                                                                         ID
                                                                         (D




                                                                         I



                                                                         i
                                                                         O
                                                                         
-------
1.0
                                     Pen. / Eff.

                                     Peak Average
                                                                                      99.0
                                                                                           f

                                                                                            o
                                                                                      90.0
                                                                                         3)
                                                                                         a>


                                                                                         1
                                                                                         w.



                                                                                         i
                                                                                         o
                                                                                         
-------
    1.0 --
o

1
I

0)
Q.

O
r>
10.0
|,
  100.0
       8:00
                                                                     99.0
                   Pen. / Eff.

                   Peak Average
                                                                                       90.0
                                                                                             •i
                                                                                             5
                                                                                             o
                                                                                             o»
                                                                           (D



                                                                           1
                                                                           ei
                                                                                             (D
9:00
                              10:00       11:00      12:00


                                     Time of Day, 11/5/92
                                                              13:00
14:00
                       Figure 18.  Hydrocarbon removal efficiency, Novembers, 1992

-------
      Because outlet emissions were effectively not coupled to inlet emissions during this test it would



 be deceptive to concentrate on the relatively low penetration-efficiency values determined during periods



 between sprayings. Likewise, it may also be deceptive to concentrate on high penetration-efficiency



 values determined only during periods of spraying. Thus, in order to obtain a conservative estimate of



 the efficiency of the Polyad FB device the following approach was adopted:  Inlet and outlet emissions



 were averaged during periods of spraying and during periods between sprayings.  Penetrations and



 efficiencies were then calculated based on these averages. Overall averages were calculated for each



 of the three periods of active spraying for each day of testing and from these averages daily averages



 and a grand average of all the data were determined.



      An inspection of the inlet emissions record suggested that an inlet hydrocarbon emissions level of



 63 ppm was a resonable cutoff point to determine if a mold was being sprayed. Thus, inlet emissions



 greater than 63 ppm were found to be typical for periods of spraying and inlet emissions of less than 63



 ppm were found to be typical for periods between sprayings. THC data during each period of active



 spraying were averaged according to this criteria.  Also, a record of the length of time of each spray/non-



 spray period was maintained as part of the averaging process. After averaging, the data were sorted by



 ascending emissions.  Overall averages (weighted by the time span of each emission) were determined



 for periods when emissions were greater than 63 ppm and for periods when emissions were less than



 63 ppm.  Finally, the percentage of time that molds were  sprayed was determined for each  period of



 active spraying.



     Table 2 shows the overall result of these calculations and includes the daily and grand test



averages referred to above. Table 3 includes the averages for each of the three test daily periods of



active spraying over the three days of testing. In these two tables no attempt was made to correct the



level of outlet emissions for actual styrene content. Table 4 provides this correction for the  times that



the amount of styrene at the outlet was determined.  In these tables average inlet and outlet emissions



and efficiency and penetration are presented for periods when molds were sprayed (emissions >  63



ppm), between sprayings (emissions < 63 ppm) and for the entire period. Figures 19 and 20 present



graphically the entire period averages shown in Tables 3  and 4 for each period of spraying. Tables





                                               40

-------
Table 2. Daily Averages and Grand Average of THC Analyzer Data,
       November 3 through 5,1992
1992
Date
3Nov
4Nov
5Nov
3-5 Nov
Inlet
Emissions
Level
>63ppm
<63ppm
ALL
>63ppm
< 63ppm
ALL
> 63ppm
<63ppm
ALL
>63ppm
<63ppm
ALL
Percent
of Time
at that Level
63.7
36.3
100.0
67.1
32.9
100.0
65.0
35.0
100.0
65.4
34.6
100.0
Average
Inlet THC
ppm
187.5
39.2
133.7
177.4
41.7
132.7
160.0
40.4
118.2
174.4
40.5
127.9
Average
Outlet THC
ppm
10.28
9.10
9.85
7.47
7.31
7.42
5.53
5.21
5.42
7.62
7.13
7.45
Average
Efficiency
%
94.51
76.79
92.63
95.79
82.47
94.41
96.55
87.10
95.42
95.63
82.37
94.18
Average
Penetration
%
5.49
23.21
7.37
4.21
17.53
5.59
3.45
12.90
4.58
4.37
17.63
5.82
                           41

-------
Table 3.  Summary of THC Analyzer Data, November 3 through 5,1992
1992
Date
3-Nov


4-Nov


5-Nov


Time
forSf
Start
9:04
10:01
12:54
8:42
10:12
12:19
8:29
10:11
12:19
Period
j raying
End
10:01
11:48
14:21
10:05
11:53
14:15
10:05
11:49
14:14
Inlet
Emissions
Level
> 63 ppm
< 63 ppm
ALL
> 63 ppm
< 63 ppm
ALL
> 63 ppm
< 63 ppm
ALL
> 63 ppm
< 63 ppm
ALL
> 63 ppm
< 63 ppm
ALL
> 63 ppm
< 63 ppm
ALL
> 63 ppm
< 63 ppm
ALL
> 63 ppm
< 63 ppm
ALL
> 63 ppm
< 63 ppm
ALL
Percent
of Time
at that Level
71.8
28.2
100.0
60.1
39.9
100.0
61.8
38.2
100.0
71.4
28.6
100.0
70.2
29.8
100.0
60.6
39.4
100.0
69.5
30.5
100.0
58.6
41.4
100.0
67.3
32.7
100.0
Average
Inlet THC
ppm
173.7
46.1
137.7
206.8
38.7
139.7
176.9
35.7
122.9
152.6
45.6
122.0
193.8
44.4
149.3
182.0
37.5
125.0
156.5
47.0
123.1
172.7
37.6
116.8
151.6
37.8
114.4
Average
Outlet THC
ppm
19.40
19.66
19.47
7.86
7.64
7.77
4.81
4.81
4.81
9.69
10.07
9.80
5.79
5.53
5.71
7.30
7.05
7.20
7.97
7.49
7.82
4.73
4.78
4.76
3.57
3.54
3.56
Average
Efficiency
%
88.83
57.40
85.86
96.20
80.28
94.44
97.28
86.53
96.09
93.65
77.92
91.97
97.01
87.56
96.17
95.99
81.21
94.25
94.91
84.05
93.65
97.26
87.28
95.93
97.65
90.65
96.89
Average
Penetration
%
11.17
42.60
14.14
3.80
19.72
5.56
2.72
13.47
3.91
6.35
22.08
8.03
2.99
12.44
3.83
4.01
18.79
5.75
5.09
15.95
6.35
2.74
12.72
4.07
2.35
9.35
3.11
                             42

-------
Table 4. Styrene Removal from THC Data, November 3 through 5,1992
1992
Date
3-Nov


4-Nov





5-Nov








Time
forSp
Start
12:54


8:42


12:19


8:29


10:11


12:19


Period
>raying
End
14:21


10:05


14:15


10:05


11:49


14:14


Inlet
Emissions
Level
> 63 ppm
< 63 ppm
ALL
> 63 ppm
< 63 ppm
ALL
> 63 ppm
< 63 ppm
ALL
> 63 ppm
< 63 ppm
ALL
> 63 ppm
< 63 ppm
ALL
> 63 ppm
< 63 ppm
ALL
Percent
of Time
at that Level
61.8
38.2
100.0
71.4
28.6
100.0
60.6
39.4
100.0
69.5
30.5
100.0
58.6
41.4
100.0
67.3
32.7
100.0
Average
Inlet THC
ppm
176.9
35.7
122.9
152.6
45.6
122.0
182.0
37.5
125.0
156.5
47.0
123.1
172.7
37.6
116.8
151.6
37.8
114.4
Average
Outlet THC
ppm
4.81
4.81
4.81
9.69
10.07
9.80
7.30
7.05
7.20
7.97
7.49
7.82
4.73
4.78
4.76
3.57
3.54
3.56
Styi
inOu
%
21.8
21.8
21.8
26.2
26.2
26.2
61.5
61.5
61.5
69.3
69.3
69.3
77.2
77.2
77.2
84.8
84.8
84.8
ene
let Air
ppm
1.05
1.05
1.05
2.54
2.64
2.57
4.48
4.33
4.42
5.53
5.19
5.43
3.66
3.69
3.67
3.03
3.00
3.02
Styrene
Efficiency
%
99.41
97.07
99.15
98.34
94.22
97.90
97.54
88.45
96.46
96.47
88.95
95.60
97.89
90.17
96.86
98.00
92.07
97.36
Styrene
Penetration
%
0.59
2.93
0.85
1.66
5.78
2.10
2.46
11.55
3.54
3.53
11.05
4.40
2.11
9.83
3.14
2.00
7.93
2.64
                               43

-------
         175
                                          Outlet Hydrocarbons

                                          Outlet Styrene
Spray Period
Day
1    2     3
 November 3
1    2     3
 November 4
1    2     3
 November 5
           Figure 19. Average inlet and outlet hydrocarbon and styrene
                     emissions for each period of spraying
                                  44

-------
        100
                                 Total Hydrocarbons

                                 Styrene
Spray Period    123123123

Day           Novembers    November 4    Novembers
              Rgure 20. Average hydrocarbon and styrene removal
                      efficiency for each period of spraying
                                45

-------
detailing averages within each period of active spraying (sorted by time) along with population standard



deviations and 95% confidence intervals are presented in Appendix C.



      Tables 2 and 3 show that, on the average, 65% of each period of active spraying activity is



occupied by spraying. During these periods the average level of styrene emissions at the inlet ranged



from 152 to 207 ppm while the average level of hydrocarbon emissions at the outlet ranged from 3.6 to



9.8 ppm (excluding the first period of active spraying on November 3). If the first period of active



spraying on November 3 is excluded the average efficiency during periods of spraying ranged from 93 to



nearly 98%,  and the average efficiency during periods between sprayings ranged from 78 to almost



91%. Overall, an average efficiency of just slightly greater than 94% was measured for the test.



      Table 4 and Figure 19 show that styrene at the outlet increased before it started to decrease.



These data also show that outlet emissions dropped by a factor of over four from the beginning to the



end of the test. This is probably because during most of the first two days of testing outlet emissions



were dominated by the evolution of napthalene-related contaminants from the Bonopore polymer.



      If data from the last period of the last day of testing can be viewed as representative of the Polyad



FB process (in the absence of contamination) then, as fable 4 and Figure 20 show, overall efficiencies



of at least 97% can be achieved.



3.1.4  Estimated Styrene Emissions from Gel Coat Booth #2



      It  is possible to estimate styrene emissions to the atmosphere from the shower stall and bath tub



gel coating process from THC data taken at the inlet of the Polyad FB device. THC data from each day



of testing were inspected to determine times during which bath tubs or shower stall molds were sprayed.



VOC emissions (assumed to be 100% styrene)  were averaged over the time required to spray each bath



tub or shower stall mold and multiplied by the time required to spray the mold to determine the



emissions rate in mg of styrene per m3/s of air flow. This value was multiplied by the nominal flow rate



of the vent exhaust fan (6.84 m3/s or 14,500 cfm)  to obtain a mass emissions rate  of styrene to the



atmosphere for each bath tub or shower stall.



     Table 5 shows the results of these calculations with population  standard deviations and 95%



confidence intervals.  Over the three days of testing sufficient data were acquired to estimate styrene





                                               46

-------
Table 5. Summary of Estimated Styrene Emissions from Gel Coat Booth #2, THC Data
Date
3-Nov

4-Nov

5-Nov

3-5 Nov
Start
9:04
10:18
12:59
Daily J
8:48
10:20
12:32
Dairy!
8:29
10:16
12:33
Daily!
Tests
End
10:01
11:42
14:07
ummary
9:57
11:50
14:08
iummary
9:49
11:45
14:07
iummary
ummary
Units
Sprayed
8
9
9
26
9
12
12
33
9
8
8
25
84
	 Time
Average
(sec)
304.5
356.6
297.6
320.1
322
328.7
290.8
313.1
373.7
410.1
340.9
374.8
333.6
o Spray Eacl
Pop. Std.
Dev. (sec)
58.9
93.8
81.1
84.2
99.1
99.2
81.5
94.7
68.8
108.7
146.7
114.6
101.8
, Unit —
95% C. I.
(sec)
40.8
61.3
53.0
32.4
64.7
56.1
46.1
32.3
44.9
75.3
101.6
44.9
21.8
— Styren
Average
(kg)
1.551
2.137
1.521
1.744
1.470
1.842
1.541
1.631
1.761
1.990
1.619
1.789
1.713
) Emissions f
Pop. Std.
Dev. (kg)
0.296
0.530
0.318
0.492
0.353
0.448
0.315
0.411
0.274
0.460
0.547
0.462
0.458
er Unit —
95% C.I.
(kg)
0.205
0.346
0.208
0.189
0.231
0.254
0.178
0.140
0.179
0.319
0.379
0.181
0.097
                                      47

-------
mass emissions for a total of 91 separate mold sprayings. On the average, 5.6 ± 1.7 minutes were
required to spray a mold and during that time approximately 1.9 ± 0.5 kg of styrene was vented to the
atmosphere, assuming that the exhaust fan in gel coat booth #2 was operated at its nominal flow rate.
The fairly high population standard deviations for these numbers are most likely due to the fact that two
different types molds were sprayed and that bath tub enclosures require a longer time to spray (with
higher emissions) than smaller shower stalls.
3.2  VOLATILE ORGANIC COMPOUND DATA
       NIOSH Method 1501 (equivalent to EPA Method 18. Adsorption Tube Procedure) was followed
to obtain charcoal tube samples at selected booth exhausts and at the inlet and outlet of the Polyad FB
device situated at gel coat booth #2.  Due to the nature of the adsorption tube sampling procedure and
the desire to sample process emissions over an extended period, sample times of from one to one and
one-half hours were typical.
       During a pretest visit on October 13,1992 charcoal tube samples were obtained at the roof
exhaust stack of gel coat boot #2.  Later, on November 6,1992, charcoal tube samples were obtained at
the roof exhaust stack of first lay-up booth #5 and second lay-up booth # 7. As described above, during
each day of sampling from November 3 -5,1992 charcoal tube samples were obtained from air
exhausted from gel coat booth #2 (the inlet of the Polyad FB device) and air exhausted from the Polyad
FB device.
       Only one operational problem was encountered during this testing. This was when a 15 minute
power failure occurred on the last day of sampling. Even though the battery-powered pumps that were
used to acquire these samples continued to function during the interval while power was lost, the sample
bypass pumps did not operate during that time (19% of the 80 minute sample period). Thus, those
particular samples may not be completely representative of the process.
       Equipment problems did significantly affect sampling.  After testing was completed it was
discovered that both sample pumps did not maintain proper sample flow under negative inlet pressures
typical of those found during this test (approximately 1 kPa above that required to sample through the
sample manifold with three charcoal tubes attached). These pumps are designed to maintain flow at

                                              48

-------
negative inlet pressures of up to 10 kPa (40 in H20). This problem was exacerbated by a crack in the
tube of the Gilian Instruments flow calibrator that prevented it from being operated at a negative
pressure of more than 0.25 kPa (1 in. HgO). Thus, although the sample pumps were calibrated before
each use, they could not be calibrated with the flow calibrator inserted into the sample line before each
charcoal tube. The net effect of this was to render the charcoal tube data taken from November 3-5,
1992 unusable except for information as to styrene content.  Fortunately, charcoal tube samples taken
at roof exhaust vents during the pretest visit on October 12 and after testing was completed on
November 6 were usable because for these measurements the sample pumps were calibrated for flow
under the actual conditions of use.
3.2.1   Measurements at Spray Booth Exhaust Stacks
        Charcoal tube samples were obtained at the exhaust stack of gel coat booth #2 during the
pretest visit on October 13,1992. Samples were also taken at the exhaust stack of first lay-up booth #5
and second lay-up booth # 7 on November 6,1992.  While samples taken at the outlet of other types of
spray booths were not required to determine the applicability of the Polyad FB device to controlling
styrene emissions from a gel-coating booth, these measurements were necessary to determine if
styrene emissions from the gel-coating operation were characteristic of other parts of the bath tub and
shower stall manufacturing process. Also, if higher levels of styrene emissions were typical of other
parts of the manufacturing process, costs to install and maintain a Polyad FB - type of control device
would be affected.  Tables 6 and 7 summarize the results of these measurements.
       Table 6 presents the sample parameters, spraying times, and number of units sprayed during
each set of runs and Table 7 presents averages with population standard deviations, emissions levels of
styrene in  ppm, and emission rate in kg of styrene per hour and kg of styrene per unit sprayed.
Because the number of molds sprayed were recorded during each period of sampling, it was possible to
determine an emission rate in terms of the amount of styrene emitted to the atmosphere per mold
sprayed. Compared to styrene emissions rates reported in Table 5 for gel coat booth #2 on November 3
- 5, the level of styrene emissions measured on October 13 with charcoal sample tubes are lower.
However, on October 13, at gel coat booth #2 only 44% of the time was spent in spraying (each period

                                               49

-------
Table 6. Sampling Conditions for Measurements made at the Outlet
        of Gel Coat Booth #2, First Lay-Up Booth # 5, and Second  '
        Lay-Up Booth #7, October - November, 1992.  EPA Method
        18/NIOSH Method 1501 Sampling.
Date
10/13
11/6

Location
Gel Coat Booth #2
First Lay-Up Booth #5
Second Lay-Up Booth #7
Sample
Numbers
1-6
40-42
43-45
Start
Tune
13:10
08:40
08:40
End
Time
14:14
09:40
09:40
Sample
Tune
min
64
60
60
Time,
Spra
min
28.17
21.08
28.33
Spent
ying
%
44.0
35.1
47.2
Number
of Units
Sprayed
8
15
27
Average Time
to Spray a Mold
sec
21 1.9 ±36.7
84.3 ±26.9
63.0 ±20.1
                              50

-------
           Table 7. Styrene Concentrations Measured at the Outlet of Gel Coat
                   Booth #2, First Lay-Up Booth * 5, and Second Lay-Up Booth #7,
                   October - November, 1992. EPA Method 18/NIOSH Method
                   1501 Sampling.
Date
10/13






11/6







Location
GC*2





Average**
FLU #5


Average
SLU#7


Average
Sample
Number
1*
2
3
4
5
6

40
41
42

43
44
45

Sample
Volume
liters
16.09
13.68
16.69
13.41
14.13
13.97

16.08
15.52
15.94

7.08
6.64
7.00

Styrene
Measured
US
141.4
6154.3
6702.7
6071.6
6102.1
5965.5

5664.6
5853.9
5544.7

3611.5
3138.5
4359.2

Styrene
Concentration
ppm
2.1
105.7
94.3
106.3
101.4
100.3
101 .6 ±4.3
84.3
90.2
83.2
85.9 ±3.1
122.3
113.4
149.1
128.3 ±15.2
Equivale
Erne
kq/h
0.22
11.08
9.89
11.15
10.64
10.52
10.66 ±0.45
8.68
9.29
8.57
8.85 ± 0.32
12.57
11.65
15.34
13.18 ±1.57
nt Styrene
kg/Unit
0.029
1.478
1.319
1.487
1.419
1.403
1.421 ±0.061
0.579
0.619
0.571
0.590 ± 0.021
0.465
0.431
0.568
0.448 ±0.01 7
Styrene
Measured
%
>99
>99
>99
>99
>99
>99
N/A
99.76
99.69
99.65
99.70 ±0.05
99.66
99.77
99.82
99.72 ±0.05
Not representative, leak around tube
Sample #1 not in averages or population standard deviations
                                           51

-------
of spraying averaged 212 seconds). As Tables 2 and 5 show, during November, molds were usually



sprayed for 65% of any period of spraying activity and the average period of spraying was approximately



334 seconds long.  Thus, styrene emissions measured for October 13 may be less than those measured



during November because of a lower production rate.



       For styrene emissions measured at first lay-up booth #5 and second lay-up booth #7 exhaust



vents on November 6.  production rates were higher and styrene emission rates per unit were tower than



that observed for the gel coating operation at any time. In first lay-up booth #5,15 units were sprayed



over an hour of sampling (84 seconds of spraying per unit) with an average emission rate of 0.59 kg/unit



sprayed. At second lay-up booth #7. 27 units were sprayed over an hour of sampling (63 seconds of



spraying per unit) with  an average emission rate of 0.45 kg/unit sprayed.  Evidently the second and third



stages of manufacture require shorter times to complete than does the initial stage of spraying with gel



coat.  Also, as Table 6 shows, the population standard deviation for average time to spray a unit can be



large compared to the  average.  These large standard deviations are almost certainly due to the mix of



units sprayed (bath tubs and shower stalls with different surface areas).



       On a per unit basis,  styrene emissions are lower for the last two stages of manufacture which



might be expected because the  mix sprayed in these last two processes contains less styrene monomer



than the mix sprayed in the gel coating process. However, on an hourly basis, at second lay up booth



#7, styrene emissions were greater than measured at gel coat booth #2 (13.2 kg/h versus 10.7 for gel



coat booth #2 and 8.9 kg/h for first lay-up booth #5).



       Compared with the results from THC measurements reported in Table 2, mass emissions



measured for first lay-up  booth #5 (86 ppm) and gel coat booth #2 (102 ppm) appear very low compared



to the average for all spraying at gel coat booth #2 (128 ppm, Table 2)  while mass emissions measured



for second lay-up booth #7 (128 ppm) agree better.  In terms of emissions of styrene per unit sprayed,



the results from gel coat booth #2 from October 13. (1.42 kg/unit) agree better with the result shown in



Table 5 (1.71 kg/unit), especially when the time of spraying is taken into account (6.7 g/sec on October



13 versus 5.1 g/sec for November 3-5).
                                               52

-------
        With respect to the amount of styrene present in the samples, while it was only verified for the
 samples taken in October that styrene was the only significant VOC present in emissions from gel coat
 booth #2, quantitative determination of styrene was performed for the samples taken in November.  As
 Table 6 shows, styrene is essentially the only VOC present in the booth exhaust air at the Eljer facility.
 3.22   Measurements at the Inlet and Outlet of the Polyad FB Device
        Table 8 summarizes the results of measurements made at the inlet of the Polyad FB device and
 Table 9 summarizes the results of measurements made at the outlet of the Polyad FB device during
 testing on November 3-5,1992.  Because of the sampling problems discussed above, Tables 8 and 9
 show the percentage of styrene and amount of styrene collected rather than concentration of styrene
 measured. Tables 8 and 9 also include averages and population standard deviations for each triplicate
 set of runs.
        Table 8 shows the amount of styrene at the inlet of the Polyad FB device from November 3 to
 November 5. This table also shows the percentage of styrene measured at the inlet as compared to the
 total amount of material detected.  While these numbers do not indicate as high a percentage of styrene
 in the exhaust of gel coat booth #2 as was reported in Table 7, an inspection of the gas chromatograms
 for each sample shown in Table 8 suggests that the samples were not contaminated by any particular
 compound.  These results also show the effects of the malfunctioning of the sample pump. The third
 sample in each set usually has much less styrene present than is present in the first two samples.
       Table 9 presents the amount of styrene at the outlet of the Polyad FB device from November 3
 to November 5 and clearly shows that the amount of styrene in the gas exiting the Polyad FB device
 increased throughout the test. This was probably due to the desorption of naphthalene-related
 contaminates from previous testing with the Polyad FB pilot unit (seeSection 2.3.2). These samples
 also show the effects of a malfunctioning sample pump.  The effect was probably more drastic at the
 outlet, compared to the inlet, because a higher sample flow rate was used. Flow through the third tube
 in each manifold was very low and so little styrene was detected that it cannot be assumed that the
amount of styrene present was reliably measured. Therefore, these samples were excluded from
condition averages and population standard deviations.

                                               53

-------
   Table 8.  Results of EPA Method 18/NIOSH Method 1501
            Samples Taken at the Inlet of the Polyad FB
            Device, November 3 to November 5,1992
Date
11/3


11/3


11/4


11/4


11/5


11/5


11/5


Run
Number
1
2
3
4
5
6
10
11
12
16
17
18
22
23
24
28
29
30
34
35
36
Start
Time
10:26


13:00


9:05


12:40


8:52


10:25


12:42


End
Time
11:26


14:02


10:05


14:10


10:02


11:30


14:02


Styrene
%
96.47
96.91
97.30
97.96
97.46
96.76
97.32
96.91
98.19
97.34
96.96
93.81
97.39
96.91
98.75
97.26
97.04
97.14
97.26
96.71
94.24
Styrene
na
1563.1
4123.8
6072.0
4909.2
1899.4
120.4
5043.3
1149.6
33.0
7614.2
4019.4
80.4
5979.9
1863.7
50.7
4332.2
1917.4
38.5
6988.4
2708.9
105.1
Average
Styrene, %*
96.89 ± 0.34


97.39 ± 0.49


97.47 ± 0.53


96.04 ±1.58


97.68 ± 0.78


97.1 4 ±0.09


96.07 ±1.31


± one population standard deviation
                                54

-------
   Table 9.  Results of EPA Method 18/NIOSH Method 1501
            Samples Taken at the Outlet of the Polyad FB
            Device, November 3 to November 5,1992
Date
11/3


11/4


11/4


11/5


11/5


11/5


Run
Number
7
8
9"
13
14
15-
19
20
21t
25
26
27"
31
32
33-
37
38
39-
Start
Time
13:00


9:05


12:40


8:52


10:25


12:42


End
Time
14:02


10:05


14:10


10:02


11:30


14:02


Styrene
%
17.67
25.83
27.69
21.78
30.61
41.29
61.52
61.39
97.32
69.22
69.42
36.55
81.12
73.32
48.78
85.87
83.78
52.72
Styrene
uq
31.9
38.8
3.4
103.5
117.8
5.1
282.1
324.5
78.2
210.1
158.0
7.2
205.7
181.5
7.3
212.8
184.0
11.3
Average
Styrene. %*
21 .75 ±4.08


26.1914.42


61 .45 ±0.06


69.32 ±0.10


77.22 ± 3.90


84.83±1.04


± one population standard deviation
Too little sample for accurate determination, excluded from average
and population standard deviation
Inconsistent result and small styrene sample, excluded from average
and population standard deviation
                                 55

-------
3.3     Analysis of Recovered Solvent Sample



        After testing had ended on November 5. the solvent recovery unit in the Polyad FB device was



drained and 2600 ml of liquid was recovered. This liquid had a strong odor, different from that of



styrene. Approximately 100 ml of this liquid was preserved in a glass container with a Teflon sealed cap



and transported to Birmingham for analysis.



        A sample of the liquid was diluted with carbon disulfide (1:100) and subjected to GC-FID



analysis to determine styrene content. It was determined that styrene was present in the sample and



was the largest constituent, but only at a concentration of 28.7 %. Many other hydrocarbons were



present. A second sample of the liquid was subjected to GC analysis with mass spectrographic



detection to identify what compounds were present. An NBS-certified software library was used for



compound identification. Table 10 presents the results of this analysis.



        As Table 10 shows, only two compounds could not be identified. The compounds other than



styrene appear to be contamination from the Polyad FB unit. As noted above, before being evaluated at



Eljer Plumbingware, the Polyad FB unit had been tested at a chemical plant that produced naphthalene.



The sample recovery reservoir in the Polyad FB  unit was not drained at the end of that test. When the



Polyad FB unit was prepared for this test, the reservoir was not drained nor was the Bonopore



adsorbent changed. Thus, naphthalene and naphthalene-related VOC's were present in the Polyad FB



unit at the start  of the test and throughout the test they continued to be desorbed from the Bonopore



adsorbent.



3.4     Total Flow Rate Data



        Following the methodology described in  Section 2.5.4, air flow into the Polyad FB device was



measured during each day of testing.  Table 11 presents the results of these measurements. The flow



measurements made on November 3 (at 14:45)  and November (at 08:55) 5 were the actual flows  used



for testing. On November 4,  flow rate measurements were made at the end of the day (14:45). The



flow rate indicated as 'Flow for Testing' in Table 11 was the flow rate used during that day. Flow  rates



were measured at two other fan settings to help  determine the fan setting to be used for the final day of



testing.





                                               56

-------
Table 10. Compounds Identified in Liquid Sample
         Recovered from the Polyad FB Device
Compound Identified*
Unknown • 1
Styrene
Unknown - 2
Diethyl Benzene
1, 2 -Diethyl Benzene
1,3 -Diethyl Benzene
2 - Methyl • 2 - Propenyl Benzene
Naphthalene
Bis (2 - Ethylhexyl) Phthalate
Total of Compounds Present
Percent
Present
1.439
25.737
4.578
17.028
16.052
5.071
5.809
17.342
6.944
100.000
Amount
(ml)
37.4
669.2
119.0
442.7
417.4
131.9
151.0
450.9
180.5
2600.0
                          57

-------
       Table 11. Flow Rate Measurements at the Inlet of the Polyad FB Device
Data and Time
11/3/92, 14:45




Average
11/4/92, 14:45
(Flow for testing)



Average
(Trial Row #1)




Average
. (Trial Row #2)




Average
11/5/92,08:55




Average
Distance from
Far Wall of Duct*
cm
1.3
3.8
6.4
8.9
11.4

1.3
3.8
6.4
8.9
11.4

1.3
3.8
6.4
8.9
11.4

1.3
3.8
6.4
8.9
11.4

1.3
3.8
6.4
8.9
11.4

First
Traverse
m/min
259.1
289.6
266.7
228.6
289.6

266.7
304.8
274.3
259.1
292.6

350.5
373.4
342.9
304.8
358.1

388.6
396.2
359.7
335.3
381.0

365.8
335.3
312.4
304.8
358.1

Second
Traverse
m/min
N/A
N/A
N/A
N/A
N/A

281.9
320.0
289.6
274.3
304.8

335.3
373.4
335.3
312.4
365.8

381.0
373.4
335.3
329.2
373.4

350.5
365.8
312.4
304.8
350.5

Average
Velocity
m/min
259.1
289.6
266.7
228.6
289.6
266.7 ±22.6
274.3
312.4
281.9
266.7
298.7
286.8 ±16.6
342.9
373.4
339.1
308.6
362.0
345.2 ±22.2
384.8
384.8
347.5
332.2
377.2
365.3 ±21 .5
358.1
350.5
312.4
304.8
354.3
336.0 ±22.7
Average
Row Rate"
m3/min





4.47 ±0.38





4.81 ±0.28





5.78 ± 0.37





6.12 ±0.36





5.63 ±0.38
**  The flow rate used for testing appears in bold face type.
                                        58

-------
3.5     Styrene Captured bv the Polvad FB Device



        Using the flow measured during each day of testing and the total time of spraying, it is possible



to estimate the minimum amount of air that passed through the Polyad FB device while styrene was



present. Also, if it is assumed that 95% of the styrene that entered the Polyad FB unit was captured and



completely recovered from the Bonopore adsorbent and that the average level of styrene present in the



exhaust air from gel coat booth #2 during periods of spraying was 128 ppm (from Table 2), then it is



possible to estimate the maximum amount of styrene that should have been recovered by the Polyad FB



device. Table 12  shows that amount to be 1.9 kg or 2.1 liters of liquid styrene.



        Unfortunately, it is not possible to reconcile the amount of styrene that should have been



recovered with the amount of styrene that was recovered.  This is because after testing had been



completed on November 4 (from 15:00 until 18:30 when the unit was powered down), and before testing



was started on November 5 (from 07:00 when the unit was powered up until 08:50), the return air line



from the desorber was vented to the atmosphere (as recorded in the Weatheriy, Inc. Polyad Mobile Unit



Log Book for the test at the Eljer Facility). The reason given in the Log Book for this modification to the



device was: To remove residual sorbent from previous testing." Thus, an unknown amount of styrene



that would have been otherwise recovered was lost.



        However, if even half of the styrene that could have been collected through the whole test was



lost, approximately 500 ml of styrene liquid would remain unaccounted for. Two explanations are



possible. First, some styrene did remain on the Bonopore adsorbent at the end of the test.  A sample of



the Bonopore adsorbent sent to SRI after the test retained a strong, styrene-like odor. Second, because



the recovered solvent container in the Polyad FB device is not designed to be completely emptied, some



condensed liquid may not have been recovered.
                                               59

-------
                Table 12. Estimated Styrene Capture in the Polyad FB Device
Date
11/3


11/4


11/5


Total
Spraying
Start/Stop
Times
09:04-10:01
10:01-11:48
12:54-14:21
08:42-10:05
10:12-11:53
12:19-14:15
08:29 - 10:05
10:11 -11:49
12:19-14:14

Polyad FB
Operation
hours
0.95
1.49
1.21
1.25
1.54
1.72
1.42
1.48
1.43
12.49
Polyad FB
Flow Rate
rrrVmin
4.47


4.81


5.63



Volume of Air
Sampled
rr.3
978.9


1301.6


1462.7


3743.2
Estimate of Styrene
Recovered*
Liters
0.58


0.76


0.76


2.10
Assuming daily average Styrene concentrations from Table 2
and a liquid Styrene density of 0.9074 g/ml
                                            60

-------
                                         SECTION 4
                COSTS ASSOCIATED WITH APPLYING THE POLYAD FB SYSTEM
                                   TO STYRENE REMOVAL

        In terms of styrene removal, the mobile Polyad FB pilot unit exhibited excellent performance.
 Based on the results of this evaluation, this technology can be expected to easily meet or exceed a 95%
 average removal efficiency for styrene entrained in air. While such performance is excellent, this
 technology is only suitable if it can compete with other styrene removal technologies on a cost basis.
 Thus, Weatherly, Inc. was asked to provide a price quote for a Polyad FB system suitable for the Eljer
 facility.
        Price quotes were obtained for two Polyad systems. The first system corresponds to the
 configuration tested at Eljer Plumbingware: an adsorber/desorber system with solvent recovery (the
 Polyad Recovery System). The second system incorporates a concentrator rather than a desorber and
 relies on incineration with catalytic oxidation to dispose of styrene captured in the adsorber (the Polyad
 Concentrator System).  Both systems were described in Section 2.3 of this report. Weatherly Inc.
 certifies that either system will meet a guarantee of 95% styrene removal and a maximum outlet styrene
 concentration of 5 ppm.  At the Eljer facility, installation of either Polyad system would require the
 addition of ducting to the present roof-vent system so that spray booth emissions would be directed to
 the Polyad system.  However, .these price quotes did not include stacks, manifolding, foundations,
 buildings, or weather protection.
       Both systems were quoted on the same design basis with a nine-year depreciation period.
Table 13 lists the design assumptions and price quote for each system and Table 14 provides an
equipment list for each system.  The air capacity of 200,000 scfm was based on the necessity of
accommodating the output from all of the 14 available spray booths operating near their rated capacity
of 14,500 acfm.  An average  inlet emissions  level of 110 ppm of styrene was assumed and the
temperature of the air entering the Polyad unit was assumed to be 30*C (86*F).
                                               61

-------
       Table 13. Design Assumptions and Price Quotes for the Polyad Concentrator System
                and the Polyad Recovery System for the Eljer Facility
DESIGN ASSUMPTIONS

   Air Capacity                 200,000
   Inlet Temperature                 86
   Inlet Styrene Concentration         110
   Outlet Styrene Concentration         5
   Total  System Efficiency            >95

   Operation                         1
   Hours of Operation per Year      2000
   Period of Depreciation               9
   Cost of Electrical Power             0.04
   Cost of Natural Gas                 1.50
   Cost of Steam                     2.00
   Cost of Bonopore Adsorbent         21.00
             scfm
             •F
             ppm
             ppm Maximum
             shift/day
             hours
             years
             $/kWh
             $/MMBtu
             $/1000lb
             $/lb
POLYAD RECOVERY SYSTEM'

    Installed Cost
    Total Cost

    Utility Consumption

    Electrical Power
    Steam
    Cooling Water
    Bonopore Adsorbent
5.000.000.00
    1828.00
      29.32
     690
     595
      82
    4800
$Aon of styrene removed (over 9 year life)
$/scfm (over 9 year life)
kWh
to/hour
gal/min
to/year
POLYAD CONCENTRATOR SYSTEM*
   Installed Cost
   Total Cost

   Utility Consumption

   Electrical Power
   Natural Gas
   Instrument Air
   Bonopore Adsorbent
2.340,000.00  $
     886.00  $/ton of styrene removed (over 9 year life)
      14.21  $/scfm (over 9 year life)
     690     kWh
       0.05  K^Btu/hour
      80     psig
    1200     to/year
  Delivery 8 months after receipt of order. Not included in the system
  price or equipment list is a stack, manifolding, foundations, buildings,
  or weather protection.
                                             62

-------
           Table 14. Equipment List and Dimensions for Large Components for Polyad
                    Systems Suitable for Installation at the Eljer Facility
POLYAD RECOVERY SYSTEM*

         Quantity         Equipment Description                   Foot Print

            2            Rlter Unit
            2            Fluid Bed Adsorber Unit                  16 ft x 40 ft
            2            Cyclone
            2            Main Fan
            2            Moving Bed Desorber Unit                14 ft x 14 ft
            2            Transport Fan
            2            Condenser Unit Fan
            2            Styrene Condenser Unit
            1            Styrene Recovery Tank
            2            Water Chiller
            2            Static Separator

                         Necessary pipes and valves between units
                         Electrical and Instrumental Equipment
                         Bonopore for Initial Riling


POLYAD CONCENTRATOR SYSTEM*

         Quantity         Equipment Description                   Foot Print

            2            Rlter Unit
            2            Fluid Bed Adsorber Unit                  16 ft x 40 ft
            2            Cyclone
            2            Main Fan
            2            Ruid Bed Desorber Unit
            2            Desorber Unit                           8 ft x 8.3 ft
            2            Catalytic Oxidizer Unit                    25 ft x 14 ft

                         Necessary pipes and valves between units
                         Electrical and Instrumental Equipment
                         Bonopore for Initial Riling
  Not included in the system price or equipment list is a stack, manifolding,
  foundations, buildings, or weather protection.
                                             63

-------
       The Polyad Recovery System is identical in function to the Polyad mobile unit tested at Eljer



Plumbingware. In order to recover styrene monomer, hot styrene-laden air from the moving bed



desorber is passed through a solvent recovery unit where styrene is condensed and directed to a



solvent recovery tank. Recovered solvent can be disposed of or reused. No attempt was made to



estimate how cost savings from solvent reuse affect overall system cost because it may not be possible



to use the recovered styrene monomer without distillation. In their quote for the PRS, Weatheriy, Inc.



noted that styrene has a tendency to discolor during recovery and in most cases requires distillation



before it can be reused.  This system was priced at $5.000,000.00 (installed).  Depreciated  over a nine



year lifetime, total cost (installation plus operation) is $29.32/scfm or $1828/ton of styrene removed.



       It is possible to estimate the magnitude of savings that would be realized from the recovery of



styrene over a nine-year lifetime for the Polyad Recovery System. In April of 1993, Mr. Rollie Nagel at



Eljer Plumbingware reported that Eljer's current cost for styrene was $0.68/lb,  not including  shipping or



handling costs. At an average inlet concentration of 110 ppm, system flow rate of 200,000 scfm, and



2000 hour work year, in a period of 9 years, approximately 3048 tons of styrene would be recovered,



assuming 95% recovery.  Presuming that 75% of this styrene could be reused, without escalating the



cost of styrene, a savings of approximately $3,109,000 would be realized.  If the cost of styrene



escalated at 2.5% per year, the savings would be approximately $3,437,000 over nine years. It should



be noted that this estimate of savings does not include capital or operating costs of redistilling styrene (if



needed), costs associated with disposing of spoiled or unusable recovered styrene, or manhour savings



realized from the handling and storage of less fresh styrene monomer. What this simple analysis does



indicate, however, is that solvent recovery may be cost effective.



       The Polyad Concentrator System uses the same type of fluidized bed adsorber as  was tested in



the Polyad FB pilot unit. However, rather than cooling hot air from the desorber to condense styrene as



was done in the pilot unit, hot styrene-laden air from the desorber (where styrene emissions are



concentrated by a factor of 11.8) is sent to a catalytic oxidizer to be burned. Flow to the catalytic



oxidizer is 17,000 acfm and because the styrene has been concentrated, combustion is self supporting



within the oxidizer.  Natural gas is required to start ignition and maintain a pilot light. This system was





                                                64

-------
priced at $2,340,000.00 (installed). Depreciated over a nine year lifetime, total cost (installation plus



operation) is $14.21/scfm or $886/ton of styrene removed.



       Weatheriy, Inc. also provided additional cost information for both types of Potyad systems. This



cost Information is shown in Figures 21  and 22.  Figure 21 shows total cost (for a nine year operating



life) in $/ton of styrene removed as a function of styrene concentration at the inlet for two inlet flow rates:



20,000 and 60,000 scfm. Rgure 22 also shows total system cost (for a nine year operating life) but in



terms of $/scfm as a function of the same inlet styrene concentrations for the same two inlet flow rates.



On both figures total cost information for the proposed 200,000 scfm Eljer Polyad system is shown at an



inlet styrene concentration of 110 ppm.
                                                 65

-------
                     95% Styrene  Removal
104 -
1U
•o
tr
.
CO
"o
i 103
tn '
o
0
Is
m2

0
X,

	
„ Qt, ,X i. i , ,
X. •. :X J^
X • i XC .
	 \. ..q 	 ^xr^T —
\l\ °X*
\\ x
! VX
i V*.
i
















V








1 — 1



V
s
y %

~


I I

	 O--





^§



r
^>

•




1
|



-







sx
*i

\
















N

*
s
\








^



s


1
i
•
i






-4-






i


s^
*i













! I I I I I
• 20,000 scfm Inlet Row Rate
• 60,000 scfm Inlet Row Rate
200,000 scfm, Eljer Facility


I i
|
|
N-X
s. * .
• X 0
X
X








I I
i ! i
Poly


X u
s>





ad Recovery
System
i :
'
X',
1





_J_
^ l
XK
pS
•*»!
V
N



^

Q
V



«
"n
>>
—











h
*_









3
*•
D
>
: : i
Polyad Concentrator
System
i 	 i — i — i — i i i i


  10'
               102

Inlet Styrene Concentration, ppm
Rgure 21.  Total system cost as $Aon of styrene removed for a nine year
          lifetime for various inlet concentrations of styrene. Data for 20,000
          and 60,000 scfm inlet flow rates are shown along with values for the
          200,000 scfm system proposed for Eljer Plumbingware
                                66

-------
                       95% Styrene Removal
I
to
3
CO
70
60
50
40
                       20,000 scfm Inlet Row Rate
                       60,000 scfm Inlet Row Rate
                       200,000 scfm, Eljer Facility
                                     Polyad Recovery
                                         System
                                             Polyad Concentrator
                                                   System
                  Inlet Styrene Concentration, ppm
Rgure 22.  Total system cost as $/scfm of inlet flow rate for a nine year
           lifetime for various inlet concentrations of styrene. Data for 20.000
           and 60,000 scfm inlet flow rates are shown along with values for the
           200,000 scfm system proposed for Eljer Plumbingware
                                 67

-------
                                         SECTION 5



                               SUMMARY AND CONCLUSIONS







      The purpose of this study was to evaluate the Polyad and Chemtact processes for controlling



styrene emissions at a representative fiberglass shower stall and bath tub manufacturing plant.



Because the Chemtact unit could not be made available within the time frame allotted for this work, only



the Polyad process was evaluated. The Polyad process was evaluated with the aid of a small,



transportable Polyad FB (fluidized bed) unit supplied by Weatherty, Inc., who is responsible for domestic



marketing and sales of the Polyad process. The evaluation was carried out from November 3-5,1992



at the Eljer Plumbingware facility located in Wilson, NC.



      The Polyad process uses successive beds of a proprietary regenerate adsorbent (Bonopore*)



that are fluidized by incoming solvent-laden air. Clean air is exhausted to the atmosphere from the last



bed through a cyclone (to recover bed material). Adsorbent is continually removed from the fluidized



beds, heated to evolve adsorbed solvent, and returned to the fluidized beds.  Solvent is evolved from the



bed material in a vertical moving bed desorber and recovered in a separate water-cooled condenser..



      Testing of the Polyad FB pilot unit was complicated by the fact that just before this evaluation the



Polyad FB unit was operated at a chemical plant to remove naphthalene from a contaminated air



stream. Unfortunately, at the conclusion of that test, the unit was not cleaned and residual naphthalene



and naphthalene-related compounds were not desorbed from the Bonopore adsorbent in the unit.



       Nevertheless, the Polyad FB device demonstrated that the Polyad process could achieve



styrene removal efficiencies of 99%. After the first day of testing, when molds were being sprayed the



unit operated at a styrene removal efficiency of 94% or greater. During the last day of testing, when



most of the naphthalene from the last series of tests had been evolved, the Polyad FB unit achieved an



average styrene removal efficiency of 95.4% (96.6% while molds were being sprayed).



       The possibility exists that greater styrene removal efficiencies could  have been reached. An



engineer at Weatheriy. Inc. indicated that the lowest available Bonopore mass flow rate for the Polyad





                                              68

-------
styrene monomer has a relatively high boiling point (146*C) and a longer residence time in the



adsorption unit and the desorption unit (lower flow rate) would have resulted in greater solvent recovery



and higher overall efficiency.



       In addition to the evaluation of the Polyad process, it was possible to quantify styrene emissions



in the spray booth exhaust to which the Polyad FB device was connected and from the exhausts of two



other spray booths at the Eljer Facility.  These measurements showed that styrene was the only volatile



organic compound present in the spray booth exhausts at this facility and that time averaged



concentrations of styrene ranged from 85 to 150 ppm.
                                                69

-------
   APPENDIX A



NIOSH METHOD 1501
       70

-------
-.-.-.^  _  ..   .                                          HYDROCARBONS, AROMATIC
FORMULA: Table i                            	
                                                                                METHOD:
M.W.: Table 1                                                                   ISSUED:  2/15/84
OSHA. NIOSH,  ACGIH: Table 2                 PROPERTIES: Table 1

COMPOUNDS:     benzene                 cunane         o-methylstyrene    styrene    vinyl toluene
(Synonyms      p-tert-butyltoluene     ethylbenzene    naphthalene        toluene    xylene
in Table  1)

""                   SAMPLING                                      MEASUREMENT

SAMPLER: SOLID SORBENT TUBE                    !TECHNIQUE:  GAS CHROMATOGRAPHY, FID
          (coconut shell charcoal,              !
          100 mg/50 mg)                         !ANALYTES: hydrocarbons  listed above

FLOW RATE.  VOLUME: Table 3                     JOESORPTION: 1 mL CS2; stand 30 min

SHIPMENT:  no special precautions               !INJECTION VOLUME:  5 uL
                                                i
SAMPLE STABILITY: not determined             ' !TEMPERATURE-IN3ECTION: 225 *C
                                                !            -OETECTOR: 225 *C
BLANKS:  2 to 10 field blanks per set           !              -COLUMN: see step  11
                                                i
BULK  SAMPLE: desirable. 1  to 10 mL;  ship in    !CARRIES GAS: »>2 or He. 25 mL/min
           •   separate containers from samples  !
                                                iCOLUHN: glass, 3.0 m x 2 am, lot OV-275 on
__ ____^_^_^__________^__   !        100/120 mesh Chranosorb. w-AW
	ACCURACY                     !        or equivalent

RANGE STUDIED.                                  !CALIBRATION: analytes in C^
BIAS  and OVERALL PRECISION (sr): Table 3       !
                                                'RANGE AND PRECISION (s,.): Table 4

                                                !£STTHATED LOO: 0.001 to 0.01 mg per sample
                                                !               with capillary column [1]
                                                i
APPLICABILITY:   This method is for peak, ceiling and TWA determinations of aromatic hydrocarbons
It may be used  for simultaneous measurements, though there is the possibility that interactions
between  analytes may reduce the breakthrough volumes and change desorption efficiencies.
INTERFERENCES:   Use of the recatmended column will prevent interference by alkanes (£C10).
Under conditions of high humidity, the breakthrough volumes may be reduced by as much as 501.
Other volatile  organic solvents, e.g.,  alcohols, tetones, ethers and halogenated hydrocarbons.
are possible interferences.   If interference is suspected, use a less polar'column or change
column temperature.
OTHER METHODS:   This method is based on and super-cedes Methods PSCAM 127. benzene, styrene.
toluene  and  xylene [2]; S311,  benzene [4];  S22, p-tert-butyltoluene [3]; S23, cumene [31; S29,
ethylbenzene [3]; S26. a-methylstyrene [3]; S292,  naphthalene [4]; S30,  styrene [3]; S343,
toluene  f4l:  S25, vinyltoluene [31:  S318. xvlene [41.	
2/15/84                                      1501-1
                                                   71

-------
HYDROCARBONS.  AROMATIC
                                                                                   METHOD:   1501
REAGENTS:
1. Eluent:   Carton disulfide*.
   chromatographic quality containing
   (optional)  suitable internal
   standard.
2. Analytes. reagent grade*
3. Mitrogen  or helium, purified
4. Hydrogen, prepurified.
S. Air.  filtered.
6. Naphthalene calibration stock
   solution, 0.40 g/mL in
   *See  Special  Precautions.
EQUIPMENT:
1. Sampler:  glass tube. 7 on long. 6 on CD.  4 nn 10,
   flame-sealed ends, containing two sections of
   activated (600 *C) coconut shell charcoal  (front
   a 100 mg, bade » 50 rag) separated by a 2-am urethane
   foam plug.  A silylated glass wool plug precedes the
   front section and a 3-mn urethane foam plug follows
   the back section.  Pressure drop across the tube at
   1 L/min airflow must be less than 3.4 kf»a.   Tubes
   are coamercially available.
2. Personal sampling pumps, 0.01 to 1 L/min
   (Table 3), with flexible connecting
   tubing.
3. Gas chramatograph, FID, integrator,  and column
   (page 1501-1).
4. Vials, glass, 1-nL. with PTFE-lined caps.
5. Pipet, ]-oL, and pipet bulb.
6. Syringes, S-, 10-, 25- and 100-uL.
7. Volumetric flasks.  10-oL.
SPECIAL  PRECAUTIONS:  Carbon disulfide is toxic and extremely Damnable (flash point » -30 *C);
benzene  is 3 suspect carcinogen.   Prepare samples and standards in a well-ventilated hood.
SAMPLING:
  1. Calibrate each personal  sapling pump with a representative sampler in line.
  2. Break  the ends of the sampler iomediately before sampling.  Attach sampler to personal
    sampling pump with flexible tubing.
  3. Sample at an accurately  known How rate  between 0.01 and 0.2 L/min (to 1 L/milTVbr
    naphthalene or styrene)  for a total  sample size as shown in Table 3.
  4. Cap  the samplers with plastic (not rubber)  caps and pack securely for shipment.

SAMPLE PREPARATION:
  5. Place  the front and back sorbent sections of the sampler tube  in separate vials.  Discard
    the  glass wool and foam  plugs.
  6. Add  1.0 mL eluent to each vial.   Attach  crimp cap to each vial  iomediately.
  7. Allow  to stand at least  30 nrin with occasional agitation.

CALIBRATION AND QUALITY CONTROL:
  8. Calibrate daily with at  least five working standards over the  appropriate range (ca. 0.01
    to 10  ng analyte per sample; see Table 4).
    a. Add known amounts of  analyte  (calibration stock solution for naphthalene) to eluent  in
       10-mL volumetric flasks and dilute to the mark.
    b. Analyze together with samples and blanks (steps  11,  12 and  13).
    c. Prepare calibration graph (peak area  of analyte vs.  mg analyte).
2/15/84
                                             1501-2
                                                72

-------
METHOD:   ISO!	HYDROCARBONS.  AROMATIC

 9. Determine desorption efficiency (DE)  at least once  for each batch of charcoal  used for
    sampling in the calibration range (step 8).   Prepare  three tubes at  each  of five levels
    plus  three media blanks.
    a. Remove and discard bade sorbent section of a media blank sampler.
    b. Inject a known amount of analyte (calibration  stock solution for  naphthalene)  directly
       onto front sorbent section with a micro liter syringe.
    c. Cap the tube.  Allow to stand overnight.
    d. Oesorb (steps S through 7) and analyze together  with working standards (steps 11,  12
       and 13).
    e. Prepare a graph of 06 vs. ng analyte recovered.
 10. Analyze three quality control blind spikes and three  analyst spikes  to insure  that the
    calibration graph and OE graph are in control.

 MEASUREMENT:
 11. S«t gas chronatograph according to manufacturer's reconnendatians and to  conditions given
    on page 1501-1.  Select appropriate column temperature:

                           Approximate Retention  Time  (min). at Indicated  Column Temperature
                             50 *C           100  *C          ISO *C            Programmed0
benzene
toluene
xylene (para)
ethylbenzene
xylene (meta)
cumene
xylene (grtho)
styrene
o-methylstyrene
vinyltoluene (meta)
naphthalene
2.5
4.3
7.0
7.0
7.2
8.3
10
16




1.1
1.4
1.4
1.5
1.6
1.9
2.5
3.2 1.0
3.8 1.2
25 4.3
2.5
4.2
5.2
5.5
5.5
6.0
6.5
7.6
8.1
8.5
12
    aOata not available for E-tert-butyltoluene and g-vinyltoluene.
    ^Temperature program:  50 *C for 3 min, then 15 •C/min to  200  *C.

    NOTE:  Alternatively, column and temperature nay be  taken from  Table 4.

12. Inject sample aliquot manually using solvent flush  technique or with autosanpler.
    NOTE:  If peak area is above the linear range of the working standards,  dilute with  eluent,
           reanalyze and apply the appropriate dilution  factor  in calculations.
13. Measure peak area.

CALCULATIONS:
14. Determine the mass, mg (corrected for OE) of analyte found in  the sample front  (Wf) and
    bade  (VAj) sorbent sections, and in the average media blank front  (Bf)  and bade
    sorbent sections.
    NOTE:  If Wjj > Wf/10, report breakthrough and possible sample loss.
2/15/84                                     1501-3
                                                 73

-------
 HYDROCARBONS. AROMATIC	METHOO:  1501

 15. Calculate concentration, C, of analyte in the air volume sampled, V (L):
 EVALUATION OF METHOD:
 Precisions and biases  listed in Table 3 were determined by analyzing generated atmospheres
 containing one-half, one. and two times the OSHA standard.  Generated concentrations were
 independently verified.  Breakthrough capacities were determined  in  dry air.   Storage  stability
 was not assessed.  Measurement precisions given in Table 4 were determined by  spiking  sampling
 media with amounts corresponding to one-half, one, and two times  the OSHA standard  for nominal
 air volumes.  Oesorption efficiencies for spiked samplers containing only one  compound exceeded
 751.  Reference [12] provides more specific information.

 REFERENCES:
  [1] User check. UBTl. NIOSH Sequence M121-S (unpublished, December 7.  1983).
  [2] NIOSH Manual of Analytical  Methods, 2nd. ed., V.  1,  PSCAH 127,  U.S.  Department of Health,
      Education, and Welfare. Pub).  (NIOSH) 77-157-A (1977).
  [3] Ibid. V. 2. S22.  S23. S2S.  S26.  S29,  S30.  U.S.  Department of Health.  Education, and
      Welfare. Publ. (NIOSH) 77-157-8 (1977).
  [4] Ibid. v. 3. S292. S311, S318,  S343. U.S. Department  of Health,  Education, and Welfare.
      Publ. (NIOSH) 77-1S7-C (1977).
  [5] R.  D. Oreisbach.  "Physical  Properties of Chemical Compounds";  Advances in Chemistry
      Series, No. 15; American Chemical  Society.  Washington  (1955).
  [6] Code of Federal Regulations;  Title 29 (Labor),  Parts  1900 to 1910; U.S. Government
      Printing Office. Washington  (1980); 29 CFR 1910.1000.
  [7] Update Criteria and Reconnendations for a Revised Benzene Standard, U.S. Department of
      Health, Education, and Welfare,  (August 1976).
  [8] Criteria for a Recommended Standard	Occupational Exposure to Toluene. U.S. Department
      of  Health, Education, and Welfare,  Publ.  (NIOSH)  73-11023 (1973).
  [9] Criteria for a Recoimended Standard....Occupational Exposure to Xylene, U.S. Department of
      Health, Education, and Welfare,  Publ. (NIOSH)  75-168  (1975).
 [10] TLVs - Threshold Limit Values  for  Chemical  Substances and Physical Agents in the Wbrte
      Environment with Intended Changes  for 1983-84.  ACGIH, Cincinnati, OH  (1983).
 [11] Criteria for a Reconnended Standard...Occupational Exposure  to  Styrene, U.S. Department of
      Health and Human Services, Publ.  (NIOSH) 83-119 (1983).
 [12] Documentation of the NIOSH Validation Tests,  S22. S23, S25.   S26, S29. S30, S292,  S311.
      S318,  S343,  U.S.  Department of Health,  Education, and Welfare;  Publ.  (NIOSH) 77-185  (1977).

 METHOD REVISED BY:   R.  Alan Lunsford. Ph.D.,  and Julie R. OVenfuss;  based on results of NIOSH
                     Contract CDC-99-74-45.
2/15/84
                                              74

-------
gTHOQ:   ISO!
                                                                         HYDROCARBONS. /WOHATTC
Table  1.  Synonyms,  formula, molecular'weight, properties


                                     Empirical
                         Structure     Formula
Name/Synonyms

benzene
   CAS #71-43-2
p-tert-butv1to1uene
   CAS #98-51-1
   1-tert-butyl -4-methylbenzene
cumene
   CAS #98-82-8
   i sopropy1benzene

ethylbenzene
   CAS #100-41-4
a-metliy 1 styrene
   CAS  #98-33-9
   i sopropeny 1 benzene
   ( 1 wnetny 1 etheny 1 ) -benzene
naphthalene
   CAS #9.1-20-3

styrene
   CAS #100-42-5
   vinylbenzene

toluene
   CAS #108-88-3
   •ethylbenzene
vinyltoluene^
   CAS #25013-15-4
   nethylstyrene     (p.-¥iny1 toluene)
   methyl vi nyl benzene
                                       CnH16
                                       CgH10
                                                  Molec-
                                                   ular
                                                  Height
                                                   78.11
CS].

  Boiling
   Point
   eg

   80.1
                                                  148.2S     192.8
                                                  120.20      152.4
                                                  106.17
                                                  118.18
                                                  128.18
                                                  104.15
                                                  118.18
   136.2


   165.4
                                                             80.2*


                                                             145.2
                                                   92.14     110.6
                                                             167.7
                                                             171.6
                                                             172.8
                                                             169.8
Vapor Pressure    Density
   9 25 *C        9 20 «C
Oim Hg)  (kPa)     (q/mO
                                                                       95.2    12.7      0.879
               0.7    0.09     0.861
                                                                        4.7     0.62     0.862
  9.6     1.28     0.867
                                                                        2.5     0.33    0.911
               0.2    0.03     1.025


               6.1    0.81     0.906



              28.4    3.79     0.867
               1.6
               1.9
               1.8
               1.8
          0.22
          0.26
          0.24
          0.24
0.898
0.911
0.911
0.904
xylenec
   CAS #1330-20-7
   dimethy1benzene
                         (p.-xylene)
                                              fortho)
                                              Owta)
                                              (Bara)
   144.4       6.7     0.89     0.830
   139.1       8.4     1.12     0.864
   138.4       8.8     1.18     0.861
         point.
bcosmercial  mixture of meta and para  i
fixture of isomers.
2/15/84
                                             1501-5
                                                 75

-------
HYDROCARBONS. ARCTWTTC

Table 2.   Permissible exposure  limits, ppra [6-11].

Substance
benzene
p-tert-butv 1 to 1 uene
cunene
ethylbenzene
«-«ethyl styrene
naphthalene
styrene
toluene
vinyl toluene
xylene

TW
10
10
so
100

10
100
200
IOC
100
OSHA
C
25
(skin)

100

200
300



Peak
50*




600"
500*


NIOSH
TVM C
1




50 100
100 200*

100 200*
ACGIH
TLV !
10**
10
50
100
SO
10
SO
100
50
100
[TEL
25**
20
75 (skin)
I2S
100
IS
100
ISO (skin)
100
ISO
ag/tift
per pom
3.19
6.06
4.91
4.34
4.83
S.24
4.26
3.77
4.33
4.34
 aHaxinun duration 10 oin in 8 fir.
 ^Maximum duration 5 nin in any 3 hr.
                          **ACGIH:  suspect carcinogen [10].
                          * 10-min  sample.
 Table 3.    Sampling Flowrate*.  volume,  capacity, range, overall bias and precision [3,4,12].

                                                    Breakthrough
 Substance
                              Sampling
Flowrate
                                  Volume (I)
 (L/nrin)   VOL-NOH
                     volume 9
                   .Concentration
                    (I)
                  (mg/m3)    (mg/m3)
                                                               Overall
                             Bias   Precision
                              CO
 benzene
 p-tett-butvl to 1 uene
 ethylbenzene
 o-nethy1styrene
 naphthalene4
 styrene
 toluene
 vinyltoluene
 xylene
  £0.20
  20.20
  20.20
  £0.20
  20.20
  21.0
  21.0
  20.20
  20.20
  20.20
 10
 10
 10
  *
200
  S3
  2C
 10
 12
 30
 29
 30
 24
 30
200
 14
  8
 24
 23
 XS
  44
 >45
  35
 >45
>240
  21
  12
  36
  35
 149
 112
 480
 917
 940
  81
1710
2294
 952
 870
42- 165
29- 119
120- 480
222- 884
236- 943
19- 33
426-1710
548-2190
256- 970
218- 870
0.8
-10.4
4.6
-8.1
-10.8
-0.5
-10.7
3.8
-9.5
-2.1
              imenoed flow is 0.01 L/min.
 ^Approximately two-thirds the breakthrough volume,  except  for naphthalene.
 ciO-min sample.
 ^Corrected value, calculated from data in Reference 12.
 ^Naphthalene shows poor desorption efficiency at low loading; 100-4. minimum volume is
  reaamended.
 ^15-min sample.
 95-min sample.
0.059
0.071d
0.059
O.OB9d
0.061d
0.055
0.053d
0.052
0.061d
0.060
 2/15/84
                                             1501-6
                                               76

-------
METHOD:
         1501
HYDROCARBONS. AROMATIC
Table 4. Measurement


Substance
benzene
p-tert-butvl toluene
cunene
ethylbenzene
a-methy 1 styrene
naphthalene
styrene
toluene
vinyltoluene
xylene
range, precisian and conditions4 [3.4,12].
Oesorption
Volume
C«O
1.0
O.S
O.S
O.S
O.S
1.0
O.S
1.0
O.S
1.0
Measurement
Range Precision
fog)
0.09- 0.3S
0.27- 1.09
0.86- 3.46
2. 17- 8.67
0.69- 3.S7
4.96-19.7
2.17-8.49
1.13- 4.S1
2.41- 9.64
2.60-10.4

0.036
0.021d
0.010
0.010
0.011
0.019
0.013d
0.011
0.008
0.010
Carrier
Flow
(mL/min)
SO
SO
50
SO
SO
30
SO
50
SO
SO
Column Parameters''
t
CO
115
us
99
85
US
125
109
155
120
180
length
(n)
0.9
3.0
3.0
3.0
3.0
3.0
3.0
0.9
3.0
0.9

Packing^
A
B
B
B
B
C
B
0
B
0
'Injection volume, 5.0  uL;  nitrogen carrier gas.
     columns stainless  steel, 3.2 on outside du
    SO/80 mesh Porapak  P; B,  101 FFAP on 80/100 mesh Chn
                                                              rt> U AU-ONCS;
 C, 101 OV-101 on  100/120 mesh Supelcoport; 0. 50/80 atsh Porapak Q.
^Corrected value,  calculated  from data  in  [12].
2/15/84
                                              1501-7
                                                 77

-------
           APPENDIX B



QUALITY CONTROL EVALUATION REPORT
               78

-------
SUMMARY
       A Quality Assurance Project Plan (QAPP) was written and approved for this project before
testing began. No field audits were planned or performed. However, as stated in the QAPP, certified
calibration gases (28, 55, and 192 ppm of styrene in nitrogen and zero air with less than 0.1 ppm THC
content) served as field performance audit samples for NIOSH Method 1501 and THC sampling. Also,
EPA personnel were on site to oversee diagnostic measurements. In the field, QC was addressed by
strict adherence to standard sampling protocols either as specified for NIOSH Method 1501 or by
following a standard operating procedure (modified as needed for this particular sampling task) with the
THC analyzers as specified in the THC instruction manual.
       In SRI's Analytical Chemistry facilities, QC is addressed by strict adherence to standard
operating procedures (SOP) previously defined and implemented.  Pertinent SOP's for the analyses
required on this project were included in the QAPP. While random audits can occur while the field
samples from any project are being analyzed, and audits are regularly performed by the QA officer at
this facility, no audit was planned or performed as part of this project.
       Overall, data quality indicator (DQI) goals were achieved.  However, some data were lost due to
an equipment malfunction and because of that malfunction much less data were obtained from the
NIOSH Method 1501 samples than was planned. Since these measurements were duplicated by the
THC analyzer measurements, overall objectives of the project were not compromised.

SIGNIFICANT QA/QC PROBLEMS
       One significant QA/QC problem was encountered. Styrene concentrations determined from
charcoal tube samples taken at the inlet and outlet of the Polyad FB mobile pilot unit were not reliable
because the sample pumps used to draw air samples through the charcoal tubes were unable to
maintain correct flow at inlet pressures negative with respect to ambient. This problem was not
discovered until after all field sampling was completed primarily because the bubble flow meter used to
calibrate the sample pumps could not be operated at inlet pressures negative with respect to ambient.
After this problem was discovered, numerous attempts were made to simulate the sampling conditions

                                              79

-------
under which the samples were taken so that sample flow rates could be corrected. However, these
attempts were not sucessful. No other corrective actions were required or taken during the collection of
samples and data or during subsequent analysis of samples collected during testing.
     i
DATA QUALITY
       The following procedures were used to determine how well data quality indicator (DQI) goals
were met:
     •   Precision is expressed as percent coefficient of variation:

                                %CV = 100x(Sx/XaVg)

         where Sx is the standard deviation of x number of data values from the data set and
         Xavg is the mean or average of the x number of data values from the data set.

     •   Bias is expresses as a difference or percent difference between measured and known
         values:

                                Bias = (X-T)

                                %RPD = 100 x [(X-T)/T]

         where T is the true value (reference standard) and X is the mean sample
         concentration. %RPD is the relative percent difference.

     •    Completeness is expressed as a percent between successful analyses and total
         attempts:
                                              80

-------
                                Completeness -100 x S/A

         where S is the number of successful analyses and A is the total number of attempts.

      •  Comparability is expressed as a percent difference (%Diff) between the results for two
         methods:

                                %Diff = 100 x (Ri-R2)/l(Rl+R2)/2]

         where R-j is the result for one method and R2 is the result for the second method.

       Table B-1 shows the DQI goals that were estimated for critical measurements in the QAPP.
Tables B-2 and B-3 show DQI values for measurements carried out with charcoal tubes (NIOSH Method
1501). Table B-4 shows DQI values for THC analyzer measurements.  Below, the precision, accuracy,
and completeness of the data that were obtained in this project are reviewed.
Precision
       Triplicate charcoal tube samples taken on October 13,1992 and on November 6,1992 provided
a total of 12 samples from which precision for this method could be estimated. Details and results of
these measurements are shown in Tables 6 and 7 of this  report.  Three series of runs were made. Six
separate runs were made at the same time at gel coat booth #2 on October 13,1992 and two sets of
triplicate runs were made on November 6,1992 after sampling was completed at gel coat booth #2 (one
set at first lay-up booth #5 and one set at second lay-up booth #7). An outlier was identified in the first
set of 6 runs made on October 13,1992 and was not included in results reported in Table B-2.
Precision for the three distinct sets of runs averaged 6.6%, greater than the 5.8% expected for NIOSH
Method 1501.  However, if the third sample taken at second lay-up booth #7 on November 6,1992 is
excluded, the average precision drops to approximately 4%.
                                              81

-------
     Table B-1. Data Quality Indicator Goals for Critical Measurements Estimated in QAPP
Method and
Reference
NIOSH 1501
Total
Hydrocarbon
Analyzer with
FID.2
Measurement
Parameter
Styrene
Content
Hydrocarbon
compounds in
air.
Experimental
Condition
1. Inlet and Outlet
of control
device,
2. Calibration gas
samples.
1. Inlet and Outlet
of control
device,
2. Calibration gas
samples.
Expected
Precision
(Rel. Std. Dev., %)
5.8i
±103
Expected
Accuracy
(% Bias)
-10.71
±53
Completeness
(%)
85
85
1 Precision and bias for sampling with charcoal-filled tube.
2 J.U.M. Model VE-7 THC Analyzer.
3 Estimated values. Precision and bias will be determined for each instrument
                                               82

-------
             Table B-2.  Data Quality Indicator Values for NIOSH Method 1501
                        Measurements Made at Eljer Plumbingware
Method and
Reference
NIOSH
1501


Measurement
Parameter
Styrene
Content


Experimental
Condition
1 Gel Coat Booth #2
Exhaust, 10/13/92
2. First Lay-Up Booth
#5 Exhaust, 11/6/92
3. Second Lay-Up
Booth #7 Exhaust,
11/6/92
Precision
(Rel. Std. Dev.. %)*
4.3
3.6
11.9
Accuracy
(% Bias)
N/A
N/A
N/A
Completeness
(%)
83.3
100
100
Precision for sampling with charcoal-filled tube.  Average of three conditions = 6.6%
                                             83

-------
Table B-3.  Data Quality Indicator Values for NIOSH Method
           1501 Charcoal Sample Tubes Spiked with Known
           Concentrations of Styrene
Known Styrene Content
(ug/ml)
90.6
181.2
181.2
453.0
1812
1812
1812
3624
3624
3624
7248

Recovered Styrene Content
(gq/mO
73.7
181.1
231.0
338.7
1750
1938
1643
3831
3590
3764
7103
Average
Bias*
(%)
-18.65
•0.06
27.48
-25.23
-3.42
6.95
-9.33
5.71
-0.94
3.86
-2.00
-1.42
Precision
(%)
—
-
-
-


8.4


3.3
-
5.9
Expressed as Relative Percent Difference
                            84

-------
                Table B-4. Data Quality Indicator Values for THC Analyzer
                         Measurements Made at Eljer Plumbingware
INLET THC ANALYZER
Span Gas/% Bias






Precision (% CV)
Average % Bias
25 ppm Span Gas
(Measured Value)
25.0
27.0
25.0
27.6
27.8
24.1
6.0
Bias
%
0.0
8.0
0.0
10.4
11.2
-3.6
4.3
49 ppm Span Gas
(Measured Value)
53.1
51.1
50.8



2.4
Bias
%
8.0
4.0
3.4



5.1
171 ppm Span Gas
(Measured Value)
172.5
171.7
164.8
170.0


2.0
Bias
%
0.5
0.0
-4.0
-1.0


-1.1
OUTLET THC ANALYZER
Span Gas/% Bias






Precision (% CV)
Average % Bias
25 ppm Span Gas
(Measured Value)
25.0
26.0
25.0
24.3
26.0
26.2
3.0

Bias
%
0.0
4.0
0.0
-2.8
4.0
4.8

1.7
49 ppm Span Gas
(Measured Value)
53.1
48.1
51.1



5.0

Bias
%
8.0
-2.2
4.0




3.3
171 ppm Span Gas
(Measured Value)
168.2
175.1
159.7
169.1
171.7

3.8

Bias
%
-2.0
2.0
-7.0
-1.5
0.0


-1.7
                                         85

-------
       Table B-3 shows the results of measurements made with charcoal tubes spiked with a known



amount of styrene monomer. These data were fit with a linear regression (R2 = 0.9969) to quantify the



results of GC-FID measurements made on styrene desorbed from the charcoal tube samples taken



during testing. Two concentrations were spiked in triplicate (1812 and 3624 ug/ml). The precision



measured for these two sets averaged 5.9%.



       For THC analyzer measurements, precision was determined from repeated measurements of



the three span gases. Averaged precision values for the inlet and outlet THC analyzers reported in



Table B-3 (3.5% for the inlet THC and 3.9% for the outlet THC) are approximately one-third of that



originially estimated in Table B-1.



Bias



       Bias could  not be determined from charcoal tube samples obtained during testing at Eljer



Plumbingware because of the problems in sampling reported above.  However charcoal tubes spiked



with known concentrations of styrene were prepared and analyzed at SRI. The results of this analysis



are reported in Table B-3.  For some of the lower concentrations listed, several bias values (as relative



percent difference) in this table are rather high. However, on the average, bias was low (-1.4%).



       For THC analyzer measurements, bias was determined for each measurement made of the



three span gases and are reported in Table B-3. Average bias values for the inlet THC ran from -1,1%



(192 ppm styrene) to 5.1% (55 ppm styrene).  Average bias values for the outlet THC ran from -1.7%



(192 ppm styrene) to 3.3% (55 ppm styrene).  Overall, bias values were within the ±5% estimated in



Table B-1.



Completeness



       For the NIOSH Method 1501 samples taken at the inlet and outlet of the Polyad FB device,



completeness was 0% because styrene concentration could not be determined from any of the



samples.  From the NIOSH Method 1501 samples taken before and after testing the Polyad FB device



completeness was 91.7%.



       For THC analyzer measurements, completeness was much higher. Some data was lost during



FID flame-outs and other data was lost during a power failure on the last day of testing. Out of





                                              86

-------
approximately 12.3 hours of data (at one data point per second) less than 5 minutes worth of data were



lost due to FID flame-outs (completeness of 99.3%).  Including in a power failure of 15 minutes when no



data were obtained, completeness was 97.3%.



Representativeness



       The design of the portable Polyad FB device dictated much of the sampling strategy and



sampling methodology practiced during this evaluation to obtain representative samples.  The use of a



large, flexible aluminum sampling line avoided contamination from plasticizers in a flexible plastic line.



Location of the sampling line inlet (within the vent exhaust duct) and the high flow velocity into the



Polyad FB unit (4-6 m/sec) assured that the sample extracted from the gel coat booth #2 exhaust was



representative.  Following the sample methodology recommended in NIOSH Method 1501 and EPA



Method 18 (Section 7.4) also assured that representative samples were obtained.



Comparability



       The sampling plan for this project made provision for simultaneous sampling using the two



measurement methods of this study which would allow comparison of the results when suitably



averaged over the same sampling period. However, because the NIOSH 1501 samples taken



coincident with the THC sampling periods were unusable, no such comparison can be made.



Nonetheless, since measured precision and accuracy of the THC data were both well within our DQI



objectives, we feel confident that the overall objectives of the project were not compromised.
                                               87

-------
              APPENDIX C



TOTAL HYDROCARBON ANALYZER DAILY RESULTS
                   88

-------
Table C-1. THC Analyzer Results from November 3,1992, First Period of Spraying
Novembs
Start
Time
9:04
9:08
9:09
9:09
9:09
9:09
9:12
9:16
9:16
922
925
9:30
9:32
9:33
9:33
9:39
9:40
9:41
9:47
9:48
9:54
9:55
9:55
9:55
9:55
9:55
9:57
Emissions
ir 3. 1992
End
Tuna
9:08
9:08
9:09
9:09
9tf9
9:12
9:16
9:16
922
925
9:30
9:32
9:33
9:33
9:39
9:40
9:41
9:47
9:48
9:54
9:55
9:55
9:55
9:55
9:55
9:57
10:01
> 63 PPM
Emissions < 63 PPM
All Emissions
Elapsed
Time
(seconds)
262
8
10
6
3
218
217
23
306
182
304
145
31
30
359
42
100
302
64
409
9
6
3
6
3
134
246
2460
968
3428
_______ ||L
^^^^^^^M* ||i
Average
165.7
61.1
60.5
63.7
63.3
41.0
176.9
55.9
195.7
45.8
171.9
45.5
99.9
53.4
202.3
54.2
49.1
190.4
52.5
131.0
64.9
59.1
63.6
59.6
64.5
41.6
183.3
173.7
46.1
137.7
LET THC (pp
Population
Std. Dav.
81.8
0.8
12
0.6
0.5
5.7
58.1
3.7
59.4
5.6
55.0
5.4
22.3
4.0
80.1
4.8
4.3
87.6
5.2
68.3
1.3
1.7
02
1.4
12
9.1
59.7



1) '•"•'•"-
95%Conf.
Interval
9.9
0.5
0.7
0.4
0.6
0.8
7.7
1.5
6.6
0.8
62
0.9
7.9
1.4
8.3
1.4
0.8
9.8
1.3
6.6
0.9
1.3
02
12
1.3
1.5
7.4



	 OL
Average
21.52
21.37
21.31
2124
2124
21.09
20.64
20.50
20.39
2020
19.87
19.61
19.56
19.59
19.53
19.38
1921
18.94
18.57
18.14
17.68
17.67
17.65
17.65
17.71
17.46
16.71
19.40
19.66
19.47
TL£TTHC(
Population
Std. Dav.
0.14
0.04
0.06
0.01
0.04
0.13
0.11
0.04
0.08
0.10
0.14
0.05
0.05
0.05
0.08
0.06
0.05
0.17
0.05
023
0.03
0.03
0.02
0.04
0.00
0.15
023



)pm) 	
95%Conf.
Interval
0.02
0.04
0.04
0.01
0.04
0.02
0.02
0.02
0.01
0.02
0.02
0.01
0.02
0.02
0.01
0.02
0.01
0.02
0.02
0.03
0.02
0.02
0.02
0.03
0.01
0.03
0.03



71.8 % of time spent spraying. 9:04 • 10:01 AM
                                         89

-------
Table C-2. THC Analyzer Results from November 3,1992. Second Period of Spraying
Novemtx
Start
Time
10:01
10:18
1023
1023
10:26
1027
10:27
10:28
10:32
10:32
10:36
10:37
10:38
10:38
10:38
10:38
10:42
10:49
10:49
10:50
10:56
11:00
11:07
11:07
11:07
11:08
11:14
11:15
11:15
11:18
1121
11:22
1127
11:35
11:35
11:35
11:38
11:42
11:47
Emissions
»3. 1992
End
10:18
10:23
1023
1025
1027
1027
1028
10:32
10:32
10:36
10:37
10:38
10:38
10:38
10:38
10:42
10:49
10:49
10:50
10:56
11:00
11:07
11:07
11:07
11:08
11:14
11:15
11:15
11:18
1121
1122
1127
11:35
11:35
11:35
11:38
11:42
11:47
11:48
> 63 PPM
Emissions < 63 PPM
All Emissions
Elapsed
TVrw
(seconds)
1014
287
5
150
78
5
70
227
10
226
81
56
4
5
13
234
375
42
25
393
195
433
4
3
67
375
32
5
186
208
5
337
449
3
6
206
232
295
37
3223
2141
5364
_^ 	 ._ II
Average
34.9
205.3
60.4
33.2
28.3
70.3
29.8
218.9
60.3
206.6
49.7
131.0
58.7
58.6
54.5
224.3
372
127.6
52.8
205.0
36.3
205.1
612
63.9
44.7
214.1
51.5
67.4
40.1
216.5
67.4
39.3
202.9
61.1
67.0
34.3
220.5
42.0
199.5
206.8
38.7
139.7
4LETTHC(pp
Population
StUDev.
9.4
55.7
1.3
7.7
6.9
3.5
8.9
74.2
1.9
73.0
42
60.9
2.1
2.0
4.0
882
8.0
52.9
4.8
77.8
5.7
70.6
0.6
1.1
5.6
78.1
4.9
22
4.0
79.6
2.2
8.8
742
0.4
2.9
4.5
66.1
6.7
69.6



m) 	
95%Conf.
Interval
0.5
6.4
12
1.3
1.5
3.0
Z1
9.7
12
9.5
0.9
15.9
£1
1.7
£1
11.3
0.8
16.0
1.9
7.7
0.8
6.6
0.6
1.3
1.3
7.9
1.7
2.0
0.5
10.8
2.0
1.0
6.9
0.4
2.3
0.6
8.5
0.8
22.4



	 Ol
Average
14.28
12.04
11.66
11.35
10.96
10.87
10.69
10.32
10.06
9.74
9.32
9.18
9.17
9.14
9.10
8.81
8.07
7.71
7.69
7.39
6.93
6.63
6.51
6.47
6.44
6.54
6.79
6.80
6.90
7.02
6.89
6.92
6.54
629
627
624
6.10
6.02
6.06
7.86
7.64
7.77
rTlETTHC(
Population
StdLDev.
1.07
022
0.03
0.16
0.07
0.02
0.06
0.14
0.02
0.19
0.08
0.04
0.02
0.03
0.03
0.18
023
0.04
0.04
0.19
0.08
0.12
0.02
0.01
0.04
0.15
0.03
0.03
0.08
0.06
0.01
0.11
0.14
0.03
0.01
0.04
0.05
0.06
0.03



>pm) 	
95%Conf.
Interval
0.06
0.03
0.03
0.03
0.02
0.02
0.02
0.02
0.01
0.03
0.02
0.01
0.02
0.03
0.02
0.03
0.03
0.01
0.02
0.02
0.01
0.01
0.02
0.02
0.01
0.02
0.01
0.03
0.01
0.01
0.01
0.01
0.01
0.04
0.01
0.01
0.01
0.01
0.01



 60.1 % of time spent spraying. 10:01 -11:48 AM
 Outlier, from period at beginning of spraying.
                                            90

-------
Table C-3.  THC Analyzer Results from November 3,1992, Third Period of Spraying
Novembi
Start
Tim*
12:54
1255
12:59
13:04
13:04
13:05
13:08
13:09
13:10
13:16
13:19
13:23
13:26
13:29
13:33
13:37
13:37
13:40
13:44
13:48
13:53
13:55
14:01
14:02
14:07
Emissions
1 3. 1992
End
Tlm0
12:55
12:59
13:04
13:04
13:05
13:08
13:09
13:10
13:16
13:19
1323
13:26
13:29
13:33
13:37
13:37
13:40
13:44
13:48
13:53
13:55
14:01
14:02
14:07
1421
> 63 PPM
Emissions < 63 PPM
All Emissions
Elapsed
Time
(seconds)
26
263
273
3
45
231
54
9
411
151
262
192
174
220
218
3
184
265
207
336
126
342
78
286
729
2692
1667
4359
	 IK
Average
2812
28.5
186.6
61.9
752
28.1
121.8
57.3
160.1
28.3
177.5
31.0
192.3
39.4
181.0
61.0
48.4
190.5
42.4
179.0
38.5
164.6
41.5
195.7
27.4
176.9
35.7
122.9
LETTHC(ppi
Population
StdDev.
11.0
2.6
882
0.6
5.5
6.8
412
5.5
80.3
7.0
55.4
62
65.6
6.5
67.9
12
4.9
80.0
5.5
77.1
4.6
78.4
8.0
68.4
10.4


'
H) 	
95%Conf.
Interval
42
0.4
10.5
0.7
1.6
0.9
11.0
3.6
7.8
1.1
6.7
0.9
9.7
0.9
9.0
1.3
0.7
9.7
0.7
82
0.8
8.3
1.8
8.0
0.7



	 OL
Average
•
4.92
4.82
4.69
4.70
4.67
4.63
•
.4.79
4.77
4.73
4.67
4.67
4.69
4.71
4.75
4.77
4.91
4.93
4.95
5.09
5.08
5.05
4.52
5.16
4.81
4.81
4.81
TLETTHC(
Population
StdDev.
•
0.04
0.05
0.02
0.03
0.04
0.04
•
0.04
0.03
0.04
0.03
0.04
0.04
0.04
0.03
0.06
0.04
0.04
0.04
0.03
0.04
0.04
1.69
0.04



>pm) 	
95%Conf.
Interval
•
0.01
0.01
0.03
0.01
0.00
0.01
•
0.00
0.01
0.01
0.00
0.01
0.00
0.00
0.03
0.01
0.00
0.01
0.01
0.01
0.00
0.01
_020
0.00



61.8 % of time spent spraying. 12:54 - 221 PM
Data tost due to flame out of RD in outlet THC analyzer.
Outlier, from period at end of spraying.
                                            91

-------
Table C-4. THC Analyzer Results from November 4.1992, First Period of Spraying
NovemtM
Start
Time
8:42
8:45
8:48
8:53
8:54
8:54
8:54
8:54
8:55
8:55
8:55
8:55
8:55
8:55
8:55
8:56
8:56
8:56
8:58
8:59
8:59
9:08
9:09
9:13
9:15
.921
924
928
929
9:29
9:30
9:32
9:35
9:36
9:36
9:36
9:36
9:37
9:37
9:37
9:43
9:46
9:50
9:50
9:51
9:57
Emissions
ir 4, 1992
End
Time
8:45
8:48
8:53
8:54
8:54
8:54
8:54
8:55
8:55
8:55
8:55
8:55
8:55
8:55
8:55
8:56
8:56
8:58
8:59
8:59
9:08
9:09
9:13
9:15
921
924
9:28
929
929
9:30
9:32
9:35
9:36
9:36
9:36
9:36
9:37
9:37
9:37
9:43
9:46
9:50
9:50
9:51
9:57
10:05
> 63 PPM
Emissions < 63 PPM
All Emissions
Elapsed
Time
(seconds)
169
208
317
3
5
22
6
28
3
5
6
4
3
17
15
49
7
120
46
7
502
54
283
108
336
217
245
33
27
6
155
195
28
18
4
9
9
15
15
355
174
240
5
75
372
480
3229
1291
4520
— ••— - IF
Average
164.8
41.3
136.4
61.6
65.9
69.6
61.1
65.8
62.1
64.9
61.5
64.7
632
60.0
77.8
58.3
68.3
36.7
1122
58.1
139.3
44.8
167.6
46.7
165.8
39.9
176.0
58.1
66.6
642
53.6
190.0
57.6
68.3
612
65.0
60.6
66.6
58.5
181.9
42.9
155.3
652
45.0
134.7
37.5
152.6
45.6
122.0
4LETTHC(ppi
Population
StUDev.
60.1
6.4
63.3
0.8
2.5
3.6
0.8
2.1
0.4
0.8
0.8
12
0.6
1.3
6.8
22
3.9
6.6
39.4
2.1
64.7
7.1
672
5.7
59.9
5.3
54.1
2.0
1.7
0.8
3.6
57.1
2.7
2.6
0.6
1.5
1.0
2.S
2.B
68.6
7.4
67.7
1.4
4.3
68.0
4.0



m) 	
95%Conf.
Interval
9.0
0.9
7.0
0.9
2.1
1.5
0.7
0.8
0.4
0.7
0.6
12
0.7
0.6
3.4
0.6
3.0
12
11.4
1.6
5.6
1.9
7.9
1.1
6.4
0.7
6.8
0.7
0.6
0.6
0.5
8.0
1.0
12
0.6
1.0
0.6
1.3
1.4
72
1.1
8.6
1.3
1.0
6.9
0.4



	 OU
Average
14.68
14.18
13.50
13.02
12.99
12.96
12.93
12,88
12,87
12.85
12.84
12.83
12.83
12.86
12.80
12.75
12.70
1Z48
1227
1221
11.52
10.77
10.39
10.00
9.68
9.36
9.01
8.74
8.66
8.61
8.46
8.09
7.66
7.53
7.45
7.44
7.40
7.38
7.34
7.14
6.88
6.61
6.41
6.39
622
5.90
9.69
10.07
9.80
TLETTHC(
Population
Ski Dm.
0.17
0.12
026
0.04
0.02
0.03
0.02
0.03
0.00
0.02
0.01
0.00
0.02
0.03
0.04
0.04
0.02
0.11
0.04
0.02
0.38
0.04
021
0.06
0.11
0.09
0.13
0.04
0.03
0.02
0.10
0.19
0.05
0.04
0.02
0.03
0.03
0.02
0.02
0.12
0.06
0.13
0.04
0.03
0.09
0.11



jpm) 	
95%Conf.
Interval
0.03
0.02
0.03
0.04
0.02
0.01
0.02
0.01
-
0.02
0.01
0.00
0.02
0.01
0.02
0.01
0.02
0.02
0.01
0.01
0.04
0.01
0.03
0.01
0.01
0.01
0.02
0.01
0.01
0.01
0.02
0.03
0.02
0.02
0.02
0.02
0.02
0.01
0.01
0.01
0.01
0.02
0.03
0.01
0.01
0.01



71.4 % of time spent spraying, 8:42 -10:05 AM
Outlier, from period at end of spraying.
                                           92

-------
Table C-5. THC Analyzer Results from November 4,1992, Second Period of Spraying
Novambc
Start
Time
10:12
1020
1024
1024
1028
10:33
10:38
10:51
10:52
10:52
10:59
10:59
10:59
11:00
11:00
11:00
11:00
11:01
11:01
11:01
11:07
11:09
11:14
11:14
11:14
11:14
11:18
11:18
11:18
11:18
11:18
11:18
11:18
11:18
11:18
11:19
11:19
11:19
1127
1127
11:31
11:31
11:36
11:36
11:38
11:38
11:38
11:44
11:45
11:46
11:50
11:50
Emissions
if 4, 1992
End
Time
1020
1024
1024
1028
10:33
10:38
10:51
10:52
10:52
10:59
10:59
10:59
10:59
11:00
11:00
11:00
11:01
11:01
11:01
11:07
11:09
11:14
11:14
11:14
11:14
11:18
11:18
11:18
11:18
11:18
11:18
11:18
11:18
11:18
11:19
11:19
11:19
11:27
1127
11:31
11:31
11:36
11:36
11:38
11:38
11:38
11:44
11:45
11:46
11:50
11:50
11:53
> 63 PPM
Emissions < 63 PPM
All Emissions
Elapsed
Tim*
(seconds)
501
221
26
259
285
300
752
56
34
393
4
6
38
50
4
5
4
3
31
335
111
322
3
5
14
206
4
6
6
3
7
4
6
6
39
13
5
437
16
252
3
308
8
68
7
4
392
18
74
235
13
162
3905
1658
5563
	 IK
Average
19.0
218.1
50.4
217.6
40.4
192.0
37.8
138.7
54.9
187.9
60.8
68.6
49.2
124.5
59.7
65.5
62.1
642
54.0
177.6
49.7
21£8
61.9
63.8
56.4
186.6
60.8
64.5
612
64.4
61.5
65.6
59.8
64.7
58.3
72.6
57.6
186.8
59.9
206.6
60.5
225.5
612
107.1
64.1
60.5
203.2
602
55.5
194.0
64.6
542
193.8
44.4
149.3
ILETTHC(ppc
Population
StdDev.
2.8
73.6
5.7
69.4
5.9
62.4
4.8
58.9
3.8
69.3
1.4
3.4
4.8
43.8
1.3
0.7
0.4
1.0
3.0
72.5
4.9
81.5
0.4
0.4
2.\
69.1
1.3
0.4
0.9
1.0
0.9
1.9
1.8
1.1
1.5
6.5
2.9
72.6
1.4
72.8
0.5
69.0
0.6
47.9
0.6
1.4
78.6
0.7
Z1
85.7
1.0
3.1



n) 	
95%Conf.
Interval
0.3
9.7
2.2
8.4
0.7
7.1
0.4
15.5
1.3
6.9
1.4
2.7
1.5
1Z2
1.3
0.6
0.4
12
1.1
7.8
0.9
8.9
0.4
0.4
1.1
9.5
1.3
0.4
0.7
1.1
0.6
1.9
1.4
6.9
0.4
3.6
2.S
6.8
0.7
9.0
0.6
7.7
0.4
11.4
0.4
1.3
7.8
0.4
0.4
10.9
0.5
0.4



	 OU
Average
5.36
528
5.33
5.31
5.33
5.36
5.30
5.41
5.41
5.41
5.43
5.40
5.40
5.37
5.34
5.36
5.33
5.35
5.36
5.41
5.48
5.67
5.80
5.78
5.81
5.83
5.86
5.87
5.89
5.93
5.92
5.89
5.88
5.88
5.92
5.95
5.94
6.03
6.06
6.09
6.09
6.17
625
629
6.35
6.31
6.47
6.46
6.42
6.35
6.34
6.31
5.79
5.53
5.71
TLETTHCJ
Population
StdDev
0.10
0.03
0.03
0.04
0.04
0.04
0.06
0.04
0.03
0.04
0.02
0.02
0.03
0.03
0.03
0.02
0.01
0.01
0.04
0.04
0.05
0.06
0.04
0.01
0.04
0.04
0.01
0.04
0.04
0.01
0.03
0.01
0.02
0.02
0.04
0.04
0.03
0.08
0.03
0.04
0.05
0.06
0.02
0.05
0.01
0.02
0.08
0.03
0.04
0.04
0.02
0.07



xn) ---
95%Conf.
Interval
0.01
0.00
0.01
0.00
0.00
0.00
0.00
0.01
0.01
0.00
0.02
0.02
0.01
0.01
0.03
0.02
0.01
0.01
0.01
0.00
0.01
0.01
0.05
0.01
0.02
0.01
0.01
0.03
0.03
0.01
0.03
0.01
0.01
0.02
0.01
0.02
0.02
0.01
0.01
0.00
0.06
0.01
0.01
0.01
0.01
0.02
0.01
0.01
0.01
0.01
0.01
0.01



 702 % of time spent spraying. 10:12 -11:53 AM
 Outlier, from period at beginning of spraying.
                                           93

-------
Table C-6.  THC Analyzer Results from November 4,1992, Third Period of Spraying
Novembc
Start
Time
12:19
12:32
1234
12:34
12:34
12:39
12:39
12:39
12:47
12:48
12:48
12:48
12:54
12:56
13:05
13:10
13:12
13:15
13:16
13:19
13:20
13:25
13:25
1329
13:30
13:34
13:34
13:34
13:38
13:42
13:46
13:52
13:54
14:00
14:00
14:04
14:08
Emissions
H4, 1992
End
Time
1232
12:34
12:34
12:34
1239
12:39
1239
1247
1248
1248
1248
1254
1256
13:05
13:10
13:12
13:15
13:16
13:19
13:20
1325
13:25
1329
13:30
13:34
13:34
13:34
13:38
13:42
13:46
13:52
13:54
14:00
14:00
14:04
14:08
14:15
> 63 PPM
Emissions < 63 PPM
All Emissions
Elapsed
Time
(seconds)
739
123
10
6
285
28
4
476
33
3
4
347
152
518
312
105
177
69
201
38
283
27
218
45
264
6
7
211
273
237
347
103
370
11
254
206
416
3487
2266
5753
	 If
Average
11.5
173.3
69.6
54.6
156.9
73.8
60.9
27.1
136.0
64.1
58.3
176.8
37.2
327
204.0
43.3
185.6
472
188.0
55.4
186.3
46.6
200.9
502
2102
61.5
64.4
44.3
221.3
46.5
1552
462
144.8
528
35.9
218.3
40.0
1820
37.5
125.0
4LETTHC(ppi
Population
Std-Dev.
5.8
57.0
3.1
3.7
66.8
7.8
12
5.7
429
1.0
24
68.1
6.3
3.8
73.9
4.6
60.3
4.3
68.6
24
59.9
4.5
69.3
4.2
75.0
0.7
0.9
5.9
832
7.0
662
6.9
64.6
5.5
3.3
76.7
8.0



n) "
95%Conf.
Interval
0.4
10.1
20
3.0
7.8
29
12
0.5
14.7
12
23
72
1.0
0.4
82
0.9
8.9
1.0
9.5
0.8
7.0
1.7
92
1.3
9.0
0.5
0.6
0.8
9.8
0.9
7.0
1.3
6.6
32
0.4
10.5
0.8



	 01
Average
6.10
5.92
5.90
5.87
5.87
5.84
5.87
5.87
.5.93
5.94
5.92
6.01
6.13
6.41
6.83
6.97
7.05
7.16
7.24
7.30
7.30
7.35
7.46
7.58
7.72
7.93
7.95
8.07
8.18
827
826
828
828
8.30
8.39
8.37
8.44
7.30
7.05
7.20
TLETTHC(
Population
StdDev.
0.17
0.04
0.05
0.02
0.04
0.04
0.03
0.04
0.04
0.01
0.04
0.07
0.04
0.14
0.10
0.04
0.04
0.04
0.08
0.04
0.05
0.04
0.07
0.04
0.12
0.02
0.03
0.10
0.06
0.04
0.04
0.04
0.04
0.04
0.08
0.04
0.09



ipm) 	
95%Conf.
Interval
0.01
0.01
0.04
0.01
0.01
0.02
0.03
0.00
0.01
0.01
0.04
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.02
0.01
0.02
0.01
0.01
0.01
0.00
0.01
0.00
0.02
0.01
0.01
0.01



60.6 % of time spent spraying. 1219 -14:15
Outlier, from periods at beginning and end of spraying.
                                           94

-------
Table C-7. THC Analyzer Results from November 5,1992. First Period of Spraying
Novembt
Start
Tune
829
8:37
8:37
8:37
8:38
8:38
8:38
8:46
8:49
8:55
8:55
8:55
8:56
8:56
8:56
8:58
9:03
9:03
9:03
9:10
9:14
9:14
9:14
9:15
9:17
.923
927
9:32
9:32
9:33
9:33
9:33
9:33
9:35
9:42
9:42
9:43
9:43
9:49
9:49
9:50
9:50
9:50
9:50
9:50
9:50
9:50
9:50
9:51
9:51
9:51
9:51
9:51
9:51
9:51
9:51
ir 5. 1992
End
Time
8:37
8:37
8:37
8:38
8:38
8:38
8:46
8:49
8:55
8:55
8:55
8:56
8:56
8:56
8:58
9:03
9:03
9:03
9:10
9:14
9:14
9:14
9:15
9:17
923
927
9:32
9:32
9:33
9:33
9:33
9:33
9:35
9:42
9:42
9:43
9:43
9:49
9:49
9:50
9:50
9:50
9:50
9:50
9:50
9:50
9:50
9:51
9:51
9:51
9:51
9:51
9:51
9:51
9:51
9:52
Elapsed
Time
(seconds)
467
3
5
26
11
41
440
165
386
5
6
41
3
10
113
309
5
12
398
241
10
7
14
134
391
227
318
3
28
7
4
9
83
438
6
41
16
373
4
18
13
4
5
4
3
4
3
24
3
5
6
3
3
4
4
20
•••••»-•»•• its
Average
148.0
612
632
55.7
65.0
512
163.8
46.3
154.6
58.7
652
52.9
64.0
602
48.3
190.1
61.9
652
39.8
180.1
58.6
64.8
57.8
432
184.6
45.1
163.4
60.9
59.3
64.4
612
65.7
45.8
147.7
612
72.5
54.7
140.6
60.1
67.9
68.7
67.3
59.8
64.6
59.4
67.1
612
67.8
64.3
67.9
59.7
61.5
64.5
60.4
60.7
58.3
ILETTHC(ppi
Population
Std. Dev.
66.1
0.7
0.5
3.6
2.1
6.9
64.6
5.1
77.1
1.5
2.5
3.7
0.7
1.3
4.4
69.7
0.3
1.4
72
76.7
2.1
1.6
3.1
8.6
56.8
5.5
76.1
0.6
2.6
1.3
0.8
£1
4.6
62.0
1.1
6.3
4.7
722
0.5
4.5
Z5
2.8
1.9
0.5
1.3
0.9
12
22
1.4
3.0
1.4
0.8
1.0
1.1
1.4
2.1
n) 	
95%Con(.
Interval
6.0
0.8
0.4
1.3
1.3
2.1
6.0
0.8
7.7
1.3
2.0
1.1
0.8
0.9
0.8
7.8
0.3
0.8
0.7
9.7
1.3
12
1.6
1.4
5.8
0.7
8.4
0.7
1.0
1.0
0.8
1.4
1.0
5.8
0.8
1.9
2.3
7.3
0.5
Z1
1.3
2.7
1.6
0.5
1.4
0.9
1.3
0.9
1.6
2.7
12
0.9
12
1.1
1.3
1.0
	 01
Average
7.41
6.99
6.97
6.98
6.95
6.90
6.73
6.66
6.83
7.03
7.04
7.06
7.12
7.09
7.14
7.24
727
728
725
6.97
6.90
6.87
6.88
6.86
6.75
7.69
9.74
9.75
10.17
10.72
10.94
11.11
10.97
10.89
9.92
9.89
10.09
8.77
7.85
7.81
7.77
7.71
7.67
7.70
7.68
7.64
7.64
7.60
7.59
7.56
7.53
7.50
7.49
7.48
7.47
7.42
TLETTHC(
Population
Std. Dev
0.30
0.04
0.01
0.02
0.03
0.03
0.06
0.03
0.15
0.02
0.02
0.03
0.02
0.03
0.04
0.05
0.02
0.02
0.07
0.09
0.02
0.02
0.03
0.05
0.04
0.80
0.11
0.03
024
0.09
0.09
0.11
0.31
0.37
0.02
0.04
0.04
0.69
0.03
0.03
0.04
0.01
0.03
0.01
0.04
0.01
0.01
0.03
0.03
0.04
0.01
0.04
0.03
0.02
0.02
0.04
>pm) 	
95%Conf.
Interval
0.03
0.04
0.01
0.01
0.02
0.01
0.01
0.01
0.02
0.02
0.02
0.01
0.02
0.02
0.01
0.01
0.02
0.01
0.01
0.01
0.01
0.02
0.01
0.01
0.01
0.11
0.01
0.04
0.09
0.06
0.09
0.07
0.06
0.04
0.02
0.02
0.03
0.07
0.03
0.01
0.03
0.01
0.03
0.01
0.04
0.01
0.01
0.01
0.03
0.03
0.01
0.04
0.03
0.02
0.02
0.02
                                                            (Continued)
                                       95

-------
                             Table C-7 Continued
Novembi
Start
Time
9:52
9:52
9:54
9:55
9:55
Emissions
>f5, 1992
End
Tim*
9:52
9:54
9:54
9:55
10:05
> 63 PPM
Emissions < 63 PPM
All Emissions
Elapsed
Tuna
(seconds)
4
131
23
42
599
3565
1561
5126
	 If
Average
59.4
53.6
67.8
58.9
32.8
156.5
47.0
123.1
U£TTHC(pp
Population
StcLDav.
2.1
3.8
3.5
3.0
7.4



m) 	
95%Conf.
Interval
Z^
0.6
1.4
0.9
0.6



	 OL
Average
7.36
7.30
7.13
7.02
6.34
7.97
7.49
7.82
rOETTHC(
Population
StdDev.
0.03
0.04
0.03
0.04
0.33



ipm) 	
95%Conf.
Interval
0.03
0.01
0.01
0.02
0.03



69.5 % of time spent spraying, 829 • 10:05 AM
Outlier, from period at end of spraying.
                                             96

-------
Table C-8.  THC Analyzer Results from November 5,1992, Second Period of Spraying
Novemtx
Start
Time
10:11
10:16
1020
1020
10:20
10:20
10:23
1024
10:24
10:25
1028
10:28
10:28
1029
10:29
10:34
10:35
10:35
10:35
10:35
10:42
10:48
10:52
10:52
10:52
10:54
11:02
11:02
11:03
11.-O4
11:13
11:14
11:14
11:14
11:15
11:15
11:15
1121
11:21
1121
1124
1125
1126
11:33
11:37
11:37
11:37
11:39
11:45
11:45
Emissions
ir 5, 1992
End
Time
10:16
1020
10:20
1020
1020
10:23
1024
10:24
1025
10:28
1028
10:28
1029
10:29
10:34
10:35
10:35
10:35
10:35
10:42
10:48
10:52
10:52
10:52
10:54
11:02
11:02
11:03
11:04
11:13
11:14
11:14
11:14
11:15
11:15
11:15
11:21
11:21
1121
11:24
1125
11:26
11:33
11:37
11:37
11:37
11:39
11:45
11:45
11:49
> 63 PPM
Emissions < 63 PPM

Elapsed
Time
(seconds)
281
258
13
5
5
162
50
5
28
228
4
6
18
4
282
87
3
3
3
418
326
253
3
8
112
484
5
83
13
570
36
23
18
4
3
25
343
8
3
200
45
42
440
211
24
14
82
362
8
251
3123
2207
5330
	 IN
Average
15.6
182.8
69.0
59.7
68.7
32.5
136.8
60.9
53.8
191.1
61.1
66.6
55.5
63.7
36.7
1192
59.9
662
59.4
172,4
38.6
180.9
61.9
68.8
44.3
165.9
59.8
412
65.8
32.8
142.6
53.6
93.0
612
61.5
482
192.3
56.1
64.0
37.6
120.9
51.8
156.9
322
123.5
97.7
45.4
214.5
71.4
37.5
172.7
37.6
116.8
L£TTHC(ppr
Population
StdDev.
22
67.9
22
1.1
3.6
9.7
46.3
0.9
52
58.9
0.4
1.9
4.6
0.7
6.9
53.4
0.3
1.7
1.5
85.9
6.7
75.1
0.6
2.0
4.7
722
12
6.3
22
5.9
54.5
4.6
25.0
0.7
0.4
4.6
762
3.0
0.9
5.7
38.8
5.1
79.6
6.3
47.8
24.0
72
65.0
6.3
5.8



n) 	
95%Conf.
Interval
0.3
8.3
1.3
1.0
3.1
1.5
12.9
0.8
2.0
7.7
0.4
1.5
^1
0.7
0.8
112
0.3
2.0
1.7
82
0.7
9.3
0.7
1.3
0.9
6.4
1.0
1.3
1.3
0.4
17,8
1.9
11.5
0.6
0.5
1.8
8.0
2.1
1.0
0.8
11.4
1.5
7.4
0.9
19.1
12.5
1.5
6.7
4.4
0.7



•»••••••» OU
Average
5.42
520
5.13
5.13
5.11
5.09
5.02
4.98
5.00
4.96
4.93
4.97
4.96
4.94
4.94
4.90
4.94
4.94
4.93
4.98
5.02
4.86
••
••
4.77
4.75
4.86
4.84
4.87
4.79
4.70
4.71
4.69
4.69
4.67
4.67
4.62
4.52
4.52
4.57
••
"*
4.43
4.31
4.31
4.31
430
427
425
4.18
4.73
4.78
4.76
TLETTHC(
Population
Std. Dev
0.13
0.04
0.02
0.03
0.02
0.04
0.03
0.01
0.04
0.03
0.01
0.04
0.04
0.02
0.05
0.03
0.02
0.03
0.02
0.06
0.04
0.05
«•
•*
0.04
0.04
0.01
0.03
0.04
0.05
0.02
0.03
0.03
0.03
0.02
0.02
0.05
0.02
0.01
0.04
**
••
0.11
0.04
0.03
0.04
0.03
0.04
0.02
0.06



ipm)
95%Con(.
Interval
0.02
0.01
0.01
0.03
0.02
0.01
0.01
0.01
0.01
0.00
0.01
0.03
0.02
0.02
0.01
0.01
0.03
0.03
0.02
0.01
0.01
0.01
"•
*•
0.01
0.00
0.01
0.01
0.02
0.00
0.01
0.01
0.01
0.03
0.02
0.01
0.01
0.02
0.01
0.01
M
«•
0.01
0.01
0.01
0.02
0.01
0.01
0.02
0.01



 58.6 % of time spent spraying. 10:11 -11:49 AM
 Outlier, from period at beginning or end of spraying.
 Data tost due to flame out of RD in outlet THC analyzer.
                                             97

-------
  Table C-9.  THC Analyzer Results from November 5,1992, Third Period of Spraying
NovwntM
Start
1219
1233
1233
1234
1238
1242
1248
1248
1248
1248
1248
1248
1248
1253
1253
1254
1255
1255
1255
1255
1256
1257
1257
13:00
13:06
13:11
13:12
13:13
13:17
13:33
13:34
13:38
13:50
13:54
13:59
14:03
14:07
Emissions
ir 5. 1992
End
Tlma
1233
1233
1234
1238
1242
1247
1248
1248
1248
12:48
1248
1248
1253
1253
1254
1254
1255
1255
1255
1256
1257
1257
13:00
13:06
13:11
13:12
13:13
13:17
13:18
13:34
13:38
13:50
13:54
13:59
14:03
14:07
14:14
> 63 PPM
Emissions < 63 PPM
AH Emissions
Elapsed
Time
(seconds)
763
26
11
245
245
344
4
9
5
4
4
9
307
8
60
4
4
6
10
66
36
3
217
365
271
90
50
247
58
114
225
723
252
292
223
262
370
3229
1570
4799
••••«••_» ftfc
Average
6.6
135.5
53.2
168.1
25.7
1429
67.4
65.4
60.4
64.8
60.4
58.4
84.9
65.4
68.1
60.5
60.4
61.5
65.7
48.8
141.2
60.1
204.2
40.9
1942
48.0
116.4
188.7
48.8
116.1
34.9
144.8
36.9
163.8
35.9
164.5
27.7
151.6
37.8
114.4
JLETTHC(ppt
Population
SklDav.
3.4
43.5
5.3
71.6
7.4
68.8
3.0
21
1.4
0.4
0.9
22
8.6
24
27
0.8
0.9
0.7
1.7
4.7
53.6
20
70.7
4.6
629
5.1
43.6
58.4
4.8
55.6
5.5
73.6
15.7
59.6
10.7
55.7
72



m) 	
95%Conf.
Interval
0.3
16.7
3.1
8.9
0.9
72
3.0
1.3
1.3
0.4
0.9
1.4
1.0
1.7
0.7
0.8
0.9
0.5
1.1
12
17.5
21
9.4
0.4
7.5
1.1
121
72
1.3
102
0.7
5.4
20
6.8
1.4
6.7
0.7



	 Ol
Avaraga
3.34
3.18
3.17
3.13
3.10
3.16
3.16
3.14
3.14
3.16
3.16
3.15
3.15
3.09
3.08
3.06
3.04
3.07
3.07
3.08
3.09
3.15
3.06
3.09
325
3.40
3.44
3.61
3.72
3.99
4.05
4.13
3.98
3.88
3.93
3.96
3.89
3.57
3.54
3.56
nXETTHC(
Population
Std. Oev.
0.10
0.04
0.02
0.04
0.03
0.04
0.03
0.02
0.02
0.02
0.04
0.03
0.04
0.02
0.03
0.02
0.02
0.03
0.03
0.03
0.04
0.03
0.03
0.04
0.08
0.04
0.03
0.09
0.03
0.01
0.04
0.07
0.06
0.02
0.01
0.02
0.02



spm) 	
95%Con(.
Interval
0.01
0.01
0.01
0.01
0.00
0.00
0.03
0.01
0.02
0.02
0.04
0.02
0.00
0.02
0.01
0.02
0.02
0.02
0.02
0.01
0.01
0.03
0.00
0.00
0.01
0.01
0.01
0.01
0.01
0.00
0.01
0.01
0.01
0.00
0.00
0.00
0.00



  67.3 % of tima spent spraying, 1219 - 214 PM
  Outlier, from period at beginning or and of spraying.
' THC data lost due to power failure.
                                              98

-------