United States
                control Technology Center
EPA-600/R-93-213
November 19S3
EMISSIONS FROM BURNING
CABINET MAKING SCRAPS
  control  ^  technology center

-------
                             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

                                           and

                        Air and Energy Engineering Research Laboratory
                             U.S. Environmental Protection Agency
                             Research Triangle Park. NC 27711
                                   EPA REVIEW NOTICE

This report has been reviewed by the U.S. Environmental Protection Agency, and approved for
publication. Approval does not signify that the contents necessarily reflect the views and policy
of the Agency, nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.

This document  is available to the public through the National Technical Information Service
Springfield, VA  22161.

-------
                                         EPA-600/R-93-213
                                         November 1993
EMISSIONS FROM BURNING CABINET MAKING SCRAPS
                         Prepared by:

                Michael Tufts and David Natschke

                Acurex Environmental Corporation
                    4915 Prospectus Drive
                       P.O. Box 13109
               Research Triangle Park, NC 27709
                 EPA Contract No. 68-DO-0141
                  (Technical Directive 91-004)
             EPA Task Manager: Robert C. McCrillis

          Air and Energy Engineering Research Laboratory
              U.S. Environmental Protection Agency
               Research Triangle Park, NC 27711
                         Prepared for:

               U.S. Environmental Protection Agency
               Office of Research and Development
                    Washington, DC 20460

-------
                                    ACKNOWLEDGEMENT

       The authors take this opportunity to thank USEPA/AREAL for analytical support to this project.
Specifically, Roy B. Zweidinger supplied aldehyde sampling media and analysis. Robert L. Seila
supplied Summa canisters and VOC analysis.
                                         ABSTRACT

       The object of this project was to make an initial determination of differences in emissions when
burning ordinary cordwood compared to kitchen cabinet making scraps. The tests were performed in
an instrumented woodstove testing laboratory on a stove which simulated units observed in use at a
kitchen cabinet manufacturer's facility.  A series of three test bums were made using a stove made
from a 55 gallon drum and a kit  sold for that purpose. The first test bum used seasoned oak cordwood
fuel while the second test burn used particle board scraps for fuel. The third test burn used Formica6
faced particle board scraps for fuel.  The scraps for tests two and three were obtained from a kitchen
cabinet manufacturer in Vermont. In general the cordwood produced lower emissions of the  heavier
molecular weight organic compounds.  There were significant differences in burnrate between the tests
which introduced some uncertainty in interpreting the analytical results.

-------
                               TABLE OF CONTENTS


 Section                                                                      Page

 ACKNOWLEDGEMENT 	ii

 ABSTRACT 	ii

 LIST OF TABLES	iv

 LIST OF FIGURES	iv

 1.     INTRODUCTION	1

 2.     EXPERIMENTAL APPROACH	3
       2.1 PROJECT DESCRIPTION	3
       2.2 EXPERIMENTAL APPARATUS	4
       2.3 EXPERIMENTAL METHODS AND PROCEDURES	7
        2.3.1  Preparation for Sampling	7
        2.3.2  Sampling Facility Operation	8
        2.3.3  Aldehyde Analysis  	8
        2.3.4  Volatile Organics Analysis	9
        2.3.5  CEM Data 	9
        2.3.6  Sample Extraction  	9
        2.3.7  Gravimetric Analysis	10
        2.3.8  Total Chromatographable Organics Analysis	11
        2.3.9  GC/MS Analysis	11
        2.3.10 Calculations .	11

3.     PRESENTATION OF RESULTS	12

4.     DATA RESULTS AND DISCUSSION	43

5.    QUALITY ASSURANCE   	49

6.    SUMMARY AND CONCLUSIONS	 51

7.    REFERENCES	53

APPENDIX A—RELATED RECOMMENDED OPERATING PROCEDURES 	  A-1

-------
                                      LIST OF TABLES


Table
                                                                                      Page
2-1.    Sampling and Analysis Responsibilities	4
2-2.    Actual Sampling Conditions	8
3-1.    CEM Data Summary	    13
3-2.    Summary of Emission Data for All Fuels	 14
3-3.    GC/FID Volatile Organic Compound Analysis	15
3-4.    Condensed Semi-volatile Organic GC/MS Results	20
3-5.    Particle Board Semi-volatile Organic GC/MS Results	 21
3-6.    Formica® Board Semi-volatile Organic GC/MS Results	22
3-7.    Semi-volatile Organic GC/MS Results (Combined)	'23
3-8.    Chemical Groups of GC/MS Identified Compounds  	'.'.24
5-1.    Percent Blank Mass of Average Sample Mass  	      49
5-2.    Completeness of  Data	        	50
                                     LIST OF FIGURES
   lure                                                                                Page
2-1.    Barrel stove and dilution tunnel 	5
2-2.    Sampling trains	6
3-1.    Mass spec chromatograph of cordwood sample extracted from XAD-2 resin	25
3-2.    Mass spec chromatograph of particle board sample extracted from XAD-2 resin	26
3-3.    Mass spec chromatograph of Formica® board sample extracted from XAD-2 resin	27
3-4.    Mass spec chromatograph of cordwood sample extracted from quartz fiber filter	28
3-5.    Mass spec chromatograph of particle board sample extracted from quartz fiber filter	29
3-6.    Mass spec chromatograph of Formica® board sample extracted from quartz fiber filter	30
3-7.    Mass spec chromatograph subtraction of cordwood sample from particle board
       (XAD-2 resin extract)	31
3-8.    Mass spec chromatograph subtraction of cordwood sample from Formica® board
       (XAD-2 resin extract)	32
3-9.    Mass spec chromatograph subtraction of cordwood sample from particle board (filter)	33
3-10.   Mass spec chromatograph subtraction of cordwood sample from Formica® board (filter) .... 34
3-11.   CEM temperature  	35
3-12.   CEM O2	36
3-13.   CEM CO2	37
3-14.   CEM CO	38
3-15.   CEM total hydrocarbon (ppm)  	39
3-16.   CEM data for cordwood bum (data normalized for comparison)  	40
3-17.   CEM data for particle wood bum (data normalized for comparison)	41
3-18.   CEM data for Formica® bum (data normalized for comparison)	42
                                             IV

-------
                                          SECTION 1



                                       INTRODUCTION







        Under the direction of Control Technologies Center (CTC), Acurex Environmental Corporation



was contracted to characterize the emissions generated by the combustion of scrap wood composite



products at small cabinet manufacturing companies in Vermont.  The scrap is burned to heat the



facilities and reduce the companies' waste disposal costs. The state of Vermont asked for assistance



after receiving complaints from citizens about visible emissions and odors emanating from the two



facilities.



        One of the Vermont facilities (facility A) specializes in manufacturing countertops. The



laminated surface composite wood material is received ready-to-use and is then cut to specifications.



Four cylindrical steel furnaces with 0.28 m3 (10 ft3) combustion chambers are used for burning scrap.



Draft on the furnaces is regulated manually and the fuel is fed manually as needed. The smoke has a



burning plastic odor which is stronger at startup and refueling. Complaints have come mainly from



passers-by.



        Scrap produced by the other facility (facility B) consists of saw dust, small pieces of particle



board, and plywood.  The furnaces have primary and secondary air controls. Scrap chunks are fed by



hand but saw dust is fed automatically.



        Composite woods contain several types of phenolic resins including phenol-formaldehyde resin



and melamine  resin1. The chief components of phenolic resins are formaldehyde, acetaldehyde, and



phenol. Characteristics of these resins are resistance to moisture, solvents, and



heat up to 200 °C.

-------
They are also dimensionally stable, sound absorbent, and noncombustible. Chief



components of melamine resin are formakJehyes, phenols, and cyano-benzenes.

-------
                                        SECTION 2



                                EXPERIMENTAL APPROACH







2.1 PROJECT DESCRIPTION



       This project's goal was to characterize emissions from the burning of common



kitchen counter top scrap material (plain particle board and particle board laminated



with Formica®).  The conditions at Vermont facility A were emulated.  To reduce expenditures,



sampling was performed in the woodstove testing laboratory of the U.S. Environmental Protection



Agency/Air and Energy Engineering Research Laboratory (EPA/AEERL) in the Environmental Research



Center (ERC). Three varieties of wood were burned, cordwood (virgin wood), particle board, and



Formica0 board (Formicae-covered particle board). Cordwood was sampled for comparison purposes.



Both composite woods were provided by facility A. Only one test was performed per day, lasting 2-5.3



hours.  Again, to reduce expenditures, only one sample was planned for each fuel.



       Acurex Environmental performed all sampling activities, and prepared and analyzed all filter and



XAD-2 samples.  Non-volatile organic compounds (NVOC) were analyzed by gravimetric methodologies



(GRAV).  Semi-volatile organic compounds (SVOC) were  analyzed by gas chromatograph/ flame



ionization detection (GC/FID) and gas chromatograph/ mass spectroscopy (GC/MS).  Samples for



volatile organic compounds (VOC) and aldehydes were transferred to EPA/AREAL for analysis.  Table



2-1 presents the sampling and analysis responsibilities.

-------
                  TABLE 2-1. SAMPLING AND ANALYSIS RESPONSIBILITIES

Filter and XAD-2 preparation
Sampling
CEM operation
Aldehyde analysis
VOC analysis
GRAV analysis
TCO analysis
GC/MS analysis
Acurex
Environmental
X




X
X
X
Acurex
Environmental/AEERL

X
X





AREAL



X
X



2.2 EXPERIMENTAL APPARATUS



       Sampling was performed according to EPA method 5G. with modifications to include the



collection of samples for chemical analysis. The wood was burned in a barrel stove constructed from a



0.28 m3 (55-gal) steel drum and a kit purchased from McMaster Carr, Inc.  This stove provided the



manual fuel feed and air control used at facility A.  The stove was mounted on a Toledo electronic



balance with a weight capacity of 300 kg to measure fuel additions and monitor short-term fuel



consumption. An insulated 0.152 m (6 in) diameter stack ran 3.66 m (12 ft) from the top of the stove to



a dilution tunnel (Figure 2-1).  Stack exhaust enters the head of the dilution tunnel at the dilution bell.



The bell draws in air to dilute the sample and isolate the mass borne by the scale from the dilution



tunnel. The dilution process cools the sample to ambient temperature so that condensable gases



(those analytes whose vapor pressures are low at ambient temperature) can be collected with a filter.



       Samples were taken from two sample ports near the base of the dilution tunnel. Aldehydes



were drawn from one sampling port and SVOCs were drawn from the other. VOCs were drawn from a



line between the filter and the XAD-2  cartridge. NVOCs were collected from the SVOC sampling train.



Figure 2-2 illustrates the sampling trains.

-------
90*Bbow
BatflM
                                                   80'Elbow
                                                               0
                    7     7
                \
                1 CEM
                                             T
                                       Poit
                                    SamptoPort
                             Sampto Point Locaflon
                                (omwolttu*)

                     L     1^

                      \^^
                                               Damp*
                                                                     Blowrar
                                    2.54 cm
                 Figure 2-1. Barrel stove and dilution tunnel.

-------
Pump
                                                  Dilution Tunnel
                                                                        HMI«d Teflon Lin*
                                                    T*tonLin«
                                                 Duplleu*
                                                 DNPHTubM
                                                 njnm
                                   Figure 2-2. Sampling trains.

-------
       VOCs were collected in an evacuated stainless steel canister via a heated line as a side stream

 between the glass fiber filters and the XAD-2 cartridges. VOC collection post-filter ensures a

 particulate-free sample. A critical orifice at the inlet of the canister controlled the flow so that a time-

 averaged sample was collected. A dry gas meter determines the total sample volume collected. These

 canisters were delivered to Acurex Environmental ready for sample collection by EPA/AREAL and

 returned for analysis.

       The aldehyde sampling train consisted of a teflon line that ran to two pairs of

 dinitrophenylhydrazine (DNPH)-impregnated tubes parallel to the flow and split equally between the

 pairs.  A flow meter at the end  of both pairs of tubes monitored their respective flows. Aldehydes react

 with the DNPH to provide non-volatile derivatives that are ready for analysis by high performance liquid

 chromatography (HPLC). The  DNPH tubes were delivered to Acurex Environmental by AREAL ready

 for sample collection and were  returned for analysis.

       A modification to the method 5G wood heater sampling protocol was used to sample SVOCs

 and NVOCs.  This sampling train consisted of two filters run in series followed by a pair of XAD-2

 cartridges in  parallel,  with the flow split equally between them.  Equal flow
 was maintained by a flow meter and control  valve placed before one of the XAD-
 2  cartridges and a control  valve placed before the other  cartridge.   The total
 flow was  monitored by a  dry gas meter at the end of the sampling train.

       Continuous emission monitor (CEM) measurements and temperature readings were collected

from the stack 2.8 m (8 ft) from the top of the balance.

2.3 EXPERIMENTAL METHODS AND PROCEDURES

2.3.1 Preparation for Sampling

       Filters were rinsed in dichtoromethane. desiccated for at least 24 h, tared, and stored in petri

dishes in the desiccator until sampling.

       The XAD-2 resin was cleaned and quality control checked according to AEERL/ROP No. 40

(Appendix A). Approximately 170 g of cleaned XAD-2 were placed into  each stainless steel canister,

sealed in a Teflon bag, and stored in the freezer until sampling.

-------
2.3.2 Sampling Facility Operation



       Each test was started with a bed of hot cordwood coals.  Sampling data was recorded at



10-min intervals by the operator and consisted of balance readings, barometric pressure, and flow data



for the sampling trains.  Barometric pressure was recorded once per day. Interruptions for fuel



additions were recorded.



       After each test, the glass sampling probe connecting the stack and the filter was rinsed with



acetone. The probe rinse is the finsate collected by this operation.



       Ash samples were collected from the barrel stove and stored in glass sample jars with Teflon®



sealed lids at the conclusion of each test.



        SVOCs and NVOCs were collected at an average flow rate of 0.799  m3/h. The flow rate was



calculated from a dry gas meter at the end of the sample train and the elapsed time.



        Table 2-2 summarizes the sampling conditions for the three tests.



                         TABLE 2-2.  ACTUAL SAMPLING CONDITIONS
Date
Fuel type
Sampling time (h:min)
Avg. fuel consumption (kg/h)
Dilution factor
Stack gas flow rate (rrrVh)
Total gas (m3)
Total fuel consumption (kg)
Number of fuel charges
10/5/90
cordwood
5:23
5.39
6.3
9.11
49.02
28.9
3
10/10/90
particle board
2:13
8.10
22.6
0.85
1.89
19.6
3
10/11/90
Formica® board
5:21
4.71
23.5
0.82
4.38
20.5
5
 2.3.3 Aldehyde Analysis



        Roy B. Zweidinger of the EPA/AREAL supplied DNPH tubes for sampling and provided the



 aldehyde analyses after sampling was finished.



        Aldehydes were analyzed by HPLC in the laboratories of Roy Zweidinger of EPA/AREAL by



 the procedures established in that laboratory. Each tube was analyzed  individually.  The four tubes

-------
collected from each burn provided QA checks on the analysis and sample collection. Analysis of the
back tubes detected the presence of break-through during sample collection. Comparing results from
the parallel sample collections detected questionable results caused by such factors as tube overload,
clogging, etc.
2.3.4 Volatile Organics Analysis
        Bob Seila of EPA/AREAL provided Summa canisters for sampling and performed the GC/FID
analyses of VOCs by the procedures established in his laboratory.  An aliquot of gas from the Summa
canister was injected.  Compound identification was based on comparing retention time to a library of
well-characterized standard compounds.  For identified compounds, quantification was performed from
stored calibrations. Where identification was not possible, an averaged response factor was used.
2.3.5 GEM Data
        Continuous emission monitors collected data for total hydrocarbon (THC), carbon dioxide (CO2),
carbon monoxide (CO), and oxygen (O2) concentration in the stack, as well as the CO concentration in
the dilution tunnel. CO and CO2 were monitored with an 880 Rosemount Analytical instrument and O2
was monitored with a 755 Rosemount Analytical instrument. THCs were monitored with a VE7 JUM
Engineering instrument.  All CEMs were calibrated with three concentrations of span gas appropriate to
each instrument.  Thermocouples located in the stack and dilution tunnel were used to monitor
temperature.  Data were recorded at 1-min intervals, transferred to a Lotus spreadsheet,  and then
stored  to disk.
2.3.6 Sample Extraction
        After sampling, the filters were placed back into petri dishes and stored in a desiccator. Filters
were desiccated for a minimum of 24 h then weighed to determine the total sample capture before
extraction.  Both filters for each test were divided in half. Half of the front fitter and half the back filter
were combined  and extracted with dichloromethane by ultra sonic extraction. The two remaining halves
were archived.  The fitter halves from a test were placed in a Level 1 -cleaned  beaker2. One hundred
mL of reagent grade dichloromethane was added to the beaker which was sufficient to completely
submerge the fitters.  An aluminum foil cover was placed over the mouth of the beaker and the beaker

-------
was placed into an ultrasonic water bath. Liquid level in the bath was shallow enough to allow the
beaker to sit firmly on the bottom.  The ultrasonic bath was then run for 15 min.  After sonicating, the
dichloromethane was poured off into a collection flask.  These steps were repeated three times to
extract 400 ml of dichloromethane.
        After sampling, the XAD-2 cartridges were resealed in Teflon* bags and stored in a freezer until
extraction.  One  of each pair of XAD-2 cartridges was extracted by pump-through elution as described
in ROP/AEERL No. 41  (Appendix A), the other was archived.
        Before extraction, the ash samples were crushed with a mortar and pestle and passed through
a 16 mesh sieve.  Ash samples were extracted with dichloromethane in a soxhlet extraction apparatus
as described in ROP/AEERL No. 22 (Appendix A).  All remaining ash was archived.
        All dichloromethane extracts were concentrated using a Kudema-Danish apparatus as
described in AEERL/ROP No. 41. Concentration was stopped at the first evidence of saturation and
the extract was made up to a known volume. All extracts were stored in a freezer after analysis.
2.3.7 Gravimetric Analysis
        NVOCs  were collected in the probe rinse, filters, and XAD-2 and analyzed by gravimetric
methodology according to AEERL/ROP No. 12 (Appendix A).  Ash samples were also analyzed by
GRAV but all  samples  were below quantifiable limits.
        Each  sample was analyzed in duplicate and the reported result is the average of these
determinations.  A slight deviation from the protocol was implemented to conserve sample and time.  A
0.25 mL aliquot  of sample was added to  each pan rather than the standard 1.0 mL  Each GRAV test
included the analysis of blank samples to detect contamination by laboratory paniculate.
        Balance data were transferred directly to a computer spreadsheet by way of  an RS-232
interface and  Lotus Measure. This change eliminated data transfer and arithmetic errors.  QC tests are
built into the spreadsheet to ensure valid reporting of data. Any sample which fails these QC tests is
repeated with additional weighings or fresh extract in new pans until all samples pass.
        Detection limits were established at three times the smallest displayed unit of the balance (0.01
mg) and the quantifiable limit was five times the detection limit.
                                              10

-------
2.3.8 Total Chromatographable Organics Analysis



        SVOCs were collected on XAD-2 cartridges and 11 cm glass fiber filters.  The concentrated



filter and XAD-2 extracts were then analyzed by GC/FID according to AEERUROP No. 13 (Appendix



A). Ash extracts were also analyzed but contained no TCO mass. Individual peaks were not identified.



        Each sample is analyzed in duplicate by direct injection GC/FID and the reported result is the



average of these determinations. The first and last sample from each daily test is a QC check sample.



If either QC check sample fails; the entire sample test is repeated after the problem is located and



resolved.  Any sample that fails the ROP's repeatability requirement is re-run with a fresh aliquot.



Quantitation of individual compounds was not performed by GC/MS because it would have duplicated



available information, in principle, from the GC/FID analysis at a high cost.



2.3.9 GC/MS Analysis



        GC/MS was performed to identify compounds in both the XAD-2 and filter extracts.



Compounds were identified by matching the retention times with a National Institute of Standards and



Technologies mass spectral library. Quantitation of individual compounds was not performed.



2.3.10 Calculations



        The quantity of SVOCs collected during the cordwood test was 0.069 g.  During the test, 4.33



m3 were sampled from the dilution tunnel.  The sample was drawn through two XAD-2 cartridges with



equal flow through them   The stack flow rate was 57.4 rrrVh and the bum rate was 5.39 kg/h.



        Specifications for the average gaseous concentration in the stack was as follows:



        •   (0.069 g*2/ 4.33 m3)6.3 = 0.20 g/m3



        •   Emissions/h = (0.20 g/m3)(57.4 m3/h) = 11.48 g/h



        •   Emissions/kg of fuel = (11.48 g/h)/(5.39  kg/h) = 2.13 g/kg



        Filter capture is the difference between presampling and post-sampling filter weights.  The total



capture is the sum of the filter capture, XAD-2, and the probe rinse.  SVOC is the sum of the TCOs for



the XAD-2 and filter extracts.  Extractable NVOC is the sum of the GRAV for the XAD-2 and filter



extracts.
                                              11

-------
                                       SECTION 3



                               PRESENTATION OF RESULTS







       Table 3-1 summarizes the CEM data. Table 3-2 describes the totals of OEMs, VOCs, SVOCs,



and NVOCs as a function of gaseous concentration, emission rate, and emission/fuel mass consumed.



VOC data for the cordwood and composite woods are described in Table 3-3.  Tables 3-4 through 3-6



list the compounds identified by GC/MS in each of the three wood types. Table 3-7 presents all the



compounds identified in all three wood types for ease of comparison. Table 3-8 compares some



GC/MS results by compound class.



       Figures 3-1 through 3-10 are MS chromatographs. Figures 3-1  through 3-3 are of XAD-2



extracts, Figures 3-4 through 3-6 are filter extracts, and Figures 3-7 through 3-10 are subtractions of



cordwood wood from composite wood.  Figures 3-11 through 3-15 compare the CEM data from the



three fuels.  Figures 3-16 through 3-18 relate the CEM data for each of the three fuels.
                                            12

-------
TABLE 3-1. CEM DATA SUMMARY

O2 (%) minimum
average
std.dev.
maximum
CO (ppm) minimum
average
std. dev.
maximum
CO2 (%) minimum
average
std. dev.
maximum
THC (ppm) minimum
average
std. dev.
maximum
TEMP (C) minimum
average
std. dev.
maximum
Cordwood
0.76
8.98
4.41
14.94
1328
17832
6563
33102
6.45
11.93
3.48
18.01
196
1123
1274
5102
348
641
231
1074
Particle Board
0.37
10.55
4.71
16.04
1415
19161
6757
34263
6.61
10.54
3.63
18.27
157
1048
1415
5987
349
628
291
1262
Formica® Board
1.01
12.02
3.93
16.96
1022
16548
4861
27128
3.55
9.37
3.12
17.77
1
739
1044
10071
311
567
202
1135
            13

-------
TABLE 3-2. SUMMARY OF EMISSION DATA FOR ALL FUELS
TABLE 3-2. Summary of Emission Data for All Fuels
Emission Data for Cordwood. Particle


Total Capture
Filter Capture
Non-volatile Extractabte Organics
Semi-volatile Organics
Volatile Organics
Total Hydrocarbons (ppm)
CO(ppm)
CO2 (%)
Emission Data for Cordwood. Particle


Total Capture
Fitter Capture
Non-volatile Extractabte Organics
Semi-volatile Organics
Volatile Organics
Total Hydrocarbons
CO
Emission Data for Cordwood, Particle


Total Capture
Filter Capture
Non-volatile Extractable Organics
Semi-volatile Organics
Volatile Organics
Total Hydrocarbons
CO
Board, and Formica Board

cordwood
0.75
0.41
0.31
0.20
5.26
2.21
2229
11.9
Board, and Formica Board

cordwood
4323
23.42
17.79
11.48
302.09
126.69
1279.52
Board, and Formica Board

cordwood
8.02
4.34
3.30
2.13
56.05
2350
23739
Expressed as Gaseous
particle
board
3.98
2.80
1.34
0.36
63.89
2.11
24.36
10.5
Expressed as Emission
particle
board
76.46
53.91
25.77
6.92
1228.63
40.52
468.43
Expressed as emission
particle
board
9.44
6.66
3.18
0.85
151.68
5.00

Concentration (g/m3)
formica
board
4.26
2.67
1.64
0.80

1.46
20.74
9.4
Rate (g/h)
formica
board
81.81
5128
3152
15.38

28.08
398.63
per fuel mass (g/kg)
formica
board
17.37
1039
6.69
326

5.96
M<»
                      14

-------
in
TABLE 3-3. GC/FID VOLATILE ORGANIC COMPOUND ANALYSIS (cont.)
mg/m3 mg/hr mg/kg
retention combined combined particle particle particle
compound time peaks cordwood peaks board cordwood board cordwood board
unknown 2.234 to 2.349
ethylene 2.497
acetylene 3.080
Ethane 3.592
propene 4.104
propane 4.235
unknown 4.437 to 6.470
iButane 6.73
unknown 7.020 to 8.058
nButane 8.245
unknown 8.594
t-2-Butene 8.755
unknown 9.123
1&2Butyne 9.218
c-2-Butene 9.359
unknown 1 0.283 to 1 2.069
CSOIefin 13.073
C6 Paraffin 13.58
unknown 13.837
C6Olefin 13.981
CSOIefin 14.195
unknown 1 4.274 to1 5.806
C6Olefin 15.915
nHexane 16.148
Chloroform 16.257
C60lefin 16.414
unknown 16.675
C6Olefin 16.85
unknown 17.048 to 17.205
2.2.3 TrIMeBut 17.336
(2) 0.85
19.73 28.70
8.57 25.32
1.91
9.99 3.13
3.72 0.58
(6) 2.84 (3) 1.66
0.20
(4) 5.10 (3) 4.65
0.49
0.14
0.74 1.22
0.05
0.10
0.30
(2) 1,125.96 (1) 443.00
33,278.80
2,341.95 21,028.81
1,037.40
5,661.91
3,364.83
(4) 640.88 (2) 15.36
5.88
0.04
2.26
0.06
0.16
0.04
.(2) 0.03
0.03 0.18
48.71
1,132.61 551.91
492.07 486.98
36.68
573.28 60.29
213.32 11.24
162.91 32.00
11.69
292.99 89.46
28.24
7.81
42.36 23.51
2.69
5.75
17.00
64,631.69 8,519.59
640,005.81
134,431.75 404,418.36
59,548.51
108,887.72
64,711.26
36,787.30 295.39
337.56
2.46
129.65
3.62
9.20
2.51
1.59
1.70 3.54
9.04
210.13 68.14
91.29 60.12
4.53
106.36 7.44
39.58 1.39
30.22 3.95
2.17
54.36 11.04
5.24
1.45
7.86 2.90
0.50
1.07
3.15
11,991.07 1,051.80
79,013.06
24,940.96 49,928.19
11,047.96
13,442.93
7,989.04
6,825.10 36.47
62.63
0.46
24.05
0.67
1.71
0.47
0.29
0.32 0.44
        () indicates the number of peaks found within that retention time window

-------
TABLE 3-3. GC/FID VOLATILE ORGANIC COMPOUND ANALYSIS (cont.)
mg/m3 mg/hr rag/kg
retention combined combined particle particle particle
compound time peaks cordwood peaks board cordwood board cordwood board
Benzene 17.821
3,3DiMePenta 17.997
CycloHexane 18.12
unknown 18.203
2MeHexane 18.407
unknown 18.512
C7 Paraffin 18.601
3MeHexane 18.688
unknown 18.751
1,30iMeCyPe 18.93
Tricloroeth 19.072
2,2,4 TrMePent 19.135
C7Olefin 19.202
C7Olefin 19.297
nHeptane 19.496
C8 Olefin 19.637
C8 Olefin 20.034
unknown 20.337
2,4 DiMeHexan 20.583
C8 Olefin 20.738
Toluene 21.298
2Me3EtPenta 21.488
unknown 21.668
3EtHexane 21.831
unknown 22.175
1t2DiMeCyHe 22.529
C9 Paraffin 22.618
Perdoroeth 22.632
C9 Paraffin 22.752
C9 Olefin 22.836
26.23 16.53
0.21
0.04
0.03
0.02
0.01
0.11
0.02
0.05
0.01
0.03
0.09
0.10
0.19
0.07
0.01
0.02
0.18
0.12
0.03
5.31 3.20
0.03
0.35
0.02
0.31
0.05
0.06
0.30
0.55
5.14
1,505.85 317.86
12.04
2.29
1.85
1.38
0.65 .
6.09
1.25
2.85
0.68
1.95
5.04
5.48
11.14
4.12
0.80
0.91
10.05
7.08
1.90
304.59 61.59
1.46
19.95
1.13
17.78
3.13
3.50
5.77
10.62
294.87
279.38 39.24
2.23
0.43
0.34
0.26
0.12
1.13
0.23
0.53
0.13
0.36
0.93
1.02
2.07
0.76
0.15
0.17
1.86
1.31
0.35
56.51 7.60
0.27
3.70
0.21
3.30
0.58
0.65
0.71
1.31
54.71
() indicates the number of peaks found within that retention time window

-------
TABLE 3-3. GC/FID VOLATILE ORGANIC COMPOUND ANALYSIS (cont.)
mg/m3 mg/hr mg/kg
retention combined combined particle particle particle
compound time peaks cordwood peaks board cordwood board cordwood board
2,3,5 TriMeHex 22.993
unknown 23.125
2,4DiMeHepta 23.166
4,4 DiMeHepta 23.302
unknown 23.493
1,1,3TriMCyhe 23.693
C9 Olefin 23.624
EtBenzene 23.996
unknown 24.280 to 24.588
C9 Paraffin 24.653
C9 Paraffin 24.828
unknown 24.842
Nonene-1 24.958
C9 Olefin 25.167
C9 Paraffin 25.299
unknown 25.387
C9 Olefin 25.516
C9 Paraffin 25.691
2,2 DiMeOctan 25.77
unknown 25.921
C10 Paraffin 26.093
C10 Olefin 26.254
unknown 26.343
3,6 DiMeOctan 26.375
010 Paraffin 26.474
unknown 26.551
2,3 DiMeOctan 26.776
5 MeNonane 26.797
2 MeNonane 26.948

0.02
0.03
0.09
0.05
0.01
0.05
0.08
0.71 0.38
(2) 0.76
0.02
1.07
0.79
0.33
0.01
0.03
0.04
0.04
0.01
0.10
0.07
0.02
0.90
0.52
1.43
0.02
0.08
0.44
0.39
0.05
0.04 0.73
1.33
1.98
4.95
2.83
0.40
2.87
4.37
40.89 7.22
43.46
1.05
20.62
45.40
18.84
0.49
1.87
2.43
2.40
0.63
6.00
3.85
1.37
51.48
9.96
82.31
1.02
4.63
8.51
22.10
2.79
2.56 13.97
0.25
0.37
0.92
0.52
0.08
0.53
0.81
7.59 0.89
8.06
0.20
2.55
8.42
3.49
0.09
0.35
0.45
0.44
0.12
1.11
0.71
0.25
9.55
1.23
15.27
0.19
0.86
1.05
4.10
0.52
0.48 1.72
() indicates the number of peaks found within that retention time window

-------
TABLE 3-3. GC/FID VOLATILE ORGANIC COMPOUND ANALYSIS (cont.)
mg/rn3 mg/hr mg/kg
retention combined combined particle particle particle
compound time peaks cordwood peaks board cordwood board cordwood board
C10 Paraffin 27.237
•B-Pinene 27.41
Decene-1 27.463
Unknown 27.596
nDecane 27.767
C10 Paraffin 27.644
secButBenz 26.019
ClOOIefin 28.133
ClOOIefin 28.188
1,2,3TriMeBe 28.29
C10 Paraffin 28.501
ClOAromat 28.592
nButCyHexa 28.849
1.3DiEtBenz 28.861
unknown 28.986 to 29.103
ClOAromat 29.322
C1 1 Paraffin 29.421
ClOAromat 29.548
unknown 29.68
ClOAromat 29.743
unknown 29.813
ClOAromat 29.975
unknown 30.072 to 30.302
ClOAromat 30.551
unknown 30.664 to 30.755
mDiiPropBe 31.216
CIIAromat 31.297
C11Aromat 31.472
unknown 31 .681
CIIAromat 31.781
0.13
0.14
0.02 0.21
0.68
0.02
0.05
0.08
0.02
0.03
0.28 0.30
0.41
0.04
0.48
0.27
(2) 0.48
0.01
0.07
0.04
0.02
0.03
0.03
0.10
(3) 0.44
0.02 0.21
(2) 1.21
0.07
0.11
0.10
0.04
0.02
7.23
8.04
1.07 4.13
38.77
1.23
2.75
4.45
0.91
1.51
16.12 5.76
23.56
2.27
9.22
15.54
27.84
0.81
4.15
2.11
1.08
1.80
1.85
5.49
25.46
1.37 4.08
69.25
3.83
6.14
5.48
2.33
1.07
1.34
1.49
0.20 0.51
7.19
0.23
0.51
0.83
0.17
0.28
2.99 0.71
4.37
0.42
1.14
2.68
5.16
0.15
0.77
0.39
0.20
0.33
0.34
1.02
4.72
0.25 0.50
12.85
0.71
1.14
1.02
0.43
0.20
() indicates the number of peaks found within that retention time window

-------
TABLE 3-3. GC/FID VOLATILE ORGANIC COMPOUND ANALYSIS (cont.)
mg/m3 mg/hr mg/kg
retention combined combined particle particle particle
compound time peaks cordwood peaks board cordwood board cordwood board
unknown 31.915
nDoDecene- 32.065
C11Aromat 32.192
nDodecane 32.267
C11Aromat 32.373
C11Aromat 32.496
C12 Paraffin 32.603
C1 2 Paraffin 32.707
unknown 32.867
CHAromat 33.037
C1 1 Aromat 33.258
unknown 33.388
C11 Aromat 33.511
unknown 33.761
C1 2 Aromat 34.253
unknown 34.479 to 35. 1 39
C1 3 Paraffin 35.351
C1 3 Paraffin 35.623
C1 2 Aromat 36.407
C1 3 Aromat 36.564
C1 3 Aromat 36.817
Total
0.10
0.02 0.75
3.29
0.01
0.04
0.06
0.02
0.12
0.03
0.42
0.36
0.03
0.14
0.01
0.02
(5) 0.45
0.02
0.04
0.03
0.01
0.03
5.262.70 63.885.75
5.49
1.02 14.43
189.06
0.50
2.41
3.21
1.21
6.73
1.55
24.00
20.72
1.56
8.28
0.44
1.25
25.92
1.00
2.04
1.67
0.77
1.68
302.087.33 1.228.627.47
1.02
0.19 1.78
35.08
0.09
0.45
0.60
0.23
1.25
0.29
4.45
3.84
0.29
1.54
0.08
0.23
4.81
0.19
0.38
0.31
0.14
0.31
56.045.89 151,682.40
() indicates the number of peaks found within that retention time window

-------
TABLE 3-4.  CONDENSED SEMI-VOLATILE ORGANIC GC/MS RESULTS	

      Identified compounds found in combusted cordwood samples

    Retention time                  Compound

        4.59                       2,4-hexadiene-1-ol
        4.78                       1,3-dimethyl-benzene
        6.94                       benzakjehyde
        7.06                       5-methyl-2-furancarboxaldehyde
        7.72                       phenol
        9.30                       4-methyl-phenol
        9.77                       3-methyl-phenol
        9.94                       4-methoxy-phenol
        11.24                      4-ethyl-benzemethenol
        11.66                      3,5-dimethyl-phenol
        11.85                      naphthalene
        12.04                      2-methoxy-4-methyl-phenol
        12.32                      1,2-benzendiol
        13.37                      3-methoxy-1,2-benzendiol
        13.68                      2-ethyl-2-methoxy-phenol
        14.99                      2,6-dimethyl-phenol
        16.61                      1,2,3-trimethyl-benzene
        17.72                      dibenzofuran
        17.93                      1 -(2,6-dihydroxy-4-methoxyphenyl)-ethanone
        21.03                      1 -(4-hydroxy-3,5-dimethoxyphenyl)-ethanone
        21.66                      phenanthrene
        23.28                      benzo[c]cinnoline
        24.00                      2-hexadecanoic acid
        24.20                      2-phenol-naphthalene
        25.33                      pyrene
        25.60                      fluoranthene
        26.07                      benzo[b]naphtno[2,3-d]furan
        29.05                      decacene
        29.12                      benzo[ghi]fluoranthene
        29.71                      triphenylene
        29.85                      chrysene
        32.05                      1-eicosane
        32.93                      1.1-diphenyl-heptane
                                20

-------
 TABLE 3-5. PARTICLE BOARD SEMI-VOLATILE ORGANIC GC/MS RESULTS

        Identified compounds found in combusted particle board samples

	Retention time	Compound	

           4.55                      2,4-hexadiene-1-ol
           4.97                      ethyl-benzene
           5.29                      1,3,5.7-cyctotetrataene
           7.64                      4-hydroxyl-benzenesulfonic acid
           8.86                      1-propenyl-benzene
           11.80                      naphthalene
           12.82                      quinoline
           16.60                      acenaphthalene
           17.35                      2-naphthalenecarbonitrile
           18.53                      2,5-dimethyl-benzenebutanoic acid
           21.65                      phenanthrene
           25.33                      pyrene
           25.60                      fluoranthene
           26.93                      2-methyl-heptadecane
           28.04                      2-21 -dimethyldocosane
           28.59                      1-phenanthrenecarboxycyclic acid
           29.12                      2-methyl-octadecane
           29.63                      1 -methyl-octadecane
           30.13                      2-methyl-heptadecane
           31.14                      hexacosane
           32.10                      heptacosane
           33.00                      octacosane
           33.97                      nonacosane
           35.09                      tricontane
           36.41                      hentriacontane
                                  21

-------
TABLE 3-6.  FORMICA8 BOARD SEMI-VOLATILE ORGANIC GC/MS RESULTS

      Identified compounds found in combusted Formica6 board samples

      Retention time                  Compound

           4.82                      2,4-hexadiene-1-ol
           5.32                      1,3,5,7-cyctotetrataene
           6.04                      3(2H)-pyridazinone
           7.54                      isocyano-benzene
           7.88                      4-hydroxyl-benzenesutfonic acid
           8.89                      1-propenyl-benzene
           9.41                      4-methyl-phenol
           9.99                      3-methyl-phenol
           11.29                     3,5-dimethyl-phenol
           11.85                     naphthalene
           12.08                     2-methoxy-4-methyl-phenol
           13.71                     4-ethyl-2-methoxy-phenol
           13.93                     1-methyl-naphthalene
           14.34                     1-(4-methoxyphenyl)-ethanone
           15.11                     2-methoxy-5-(1 -propenyl)-phenol
           17.37                     2-naphthalenecarbonitrile
           17.71                     dibenzofuran
           18.75                     2,5-dimethyl-benzenebutanoic acid
           24.17                     6-propyl-tridecane
           25.76                     2-methyl-tetradecane
           26.93                     2-methyl-heptadecane
           28.05                     2.21-dimethyl-docosane
           29.14                     2-methyl-octadecane
           30.15                     2-methyl-heptadecane
           31.15                     hexacosane
           32.10                     heptacosane
           33.00                     octacosane
           33.99                     nonacosane
           35.08                     tricontane
           36.44                     hentriacontane
                                   22

-------
TABLE 3-7.  SEMI-VOLATILE ORGANIC GC/MS RESULTS (combined)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61

2,4-hexadiene-1-ol
1 ,3-dimethyl-benzene
benzaldehyde
5-methyl-2-furancarboxaldehyde
phenol
4-methyl-phenol
3-methyl-phenol
4-methoxy-phenol
4-ethyl-benzemethenol
3, 5-di methyl-phenol
naphthalene
2-methoxy -4-methyl-phenol
1 ,2-benzendiol
3-methoxy-1 ,2-benzendiol
2-ethyl-2-methoxy -phenol
2,6-di methyl-phenol
1 ,2,3-trimethyl-benzene
dibenzofuran
1 -(2,6-dihydroxy-4-methoxyphenyl)-ethanone
1-(4-hydroxy-3,5-dimethoxyphenyl)-e'thanone
phenanthrene
benzo[c]cinnoline
2-hexadecanoic acid
2-phenol-naphthalene
pyrene
fluoranthene
benzo[b]naphtho[2,3-d]furan
decaoene
benzo(ghi]fluoranthene
triphenylene
chrysene
1-eioosane
1 , 1 -diphenyl-heptane
ethyl-benzene
1 ,3,5.7-cydotetrataene
4-hydroxyl-benzenesulfonic acid
1 -propenyl -benzene
quinoline
acenaphthalene
2 -naphtha! en ecarbonitrile
2,5-dimethyl-benzenebutanoic
2-methyl-heptadecane
2-21 -dimethyWocosane
1-phenanthrenecarboxycydic acid
2-methyl-octadecane
1 -methyl-octadecane
2-methyl-heptadecane
hexacosane
heptacosane
octacosane
nonacosane
tricontane
hen tri aeon tane
3(2H)-pyridazinone
isocyano-benzene
4-ethy(-2-methoxy-phenol
1 -methyl-naphthalene
1 -(4-methoxypheny))-ethanone
2-methoxy-5-( 1 -propenyl)-phenol
6-propyl-tridecane
2-methyl-tetradecane
Cordwood
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X




























Particle
X









X






















X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X








Formica*
X




X
X


X
X
X





X
















X
X
X


X
X

X

X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
                          23

-------
         TABLE 3-8. CHEMICAL GROUPS OF GC/MS IDENTIFIED COMPOUNDS
                  Cordwood        Particle Board     Formica0 Board
 Oxygenated       20/33 61%         4/22  18%       11/30 37%
 Conjugated       29/33 88%         11/22 50%      17/30 57%
Fully saturated       0/33 0%          10/2245%       12/30 40%
                                     24

-------
10
in
               Abundance


          6000000
          4000000
          3000000
          2000000 -
          1000000-
                                            lon range  60.00 to 600.00 amu. from wood2.il
                                                                                                                JUU_
10       12       14        18

                 Time (min.)
18
                                                                                            20
22
                                                                                         Y: null.

                                                                                         X: null.
                          Figure 3-1.  Mass spec chromatograph of oordwood sample extracted from XAD-2 resin.

-------
ro
o>
ton rang* 60.00 to 600.00 amu. from wood4.d
Abundance
I
2400000 -
2000000 •
1600000 •



1200000 •



800000 -




400000 -



























u
1






in
Ol ^^
N in
t

M
N
en
1^

wjg
X
111,
4




1 W
*
rt
49
Wf
W fH

!^ 1
ill





*
.

rm
10
CO
<
_
*


,.
8




40
49

fen
t ^
5 oil
dLi





-------
                                 ton r«ng«  60.00 to 600.00 •mu. from woodBjl
     Abundance
6000000
4000000
3000000
2000000 -
1000000 -
                                               12
10
20
                                                                                                    24
                                                    Time (mbi.)
                                                                              Y: null.
                                                                              X: null.
              figure 3-9. Mass spec chromalograph of Fonnfca* board sample extracted from XAD-2 resin.

-------
     Abumtanca
2000000
 1600000
 1200000
 800000
                                km rang* 60.00 to 600.00 unu. from wood 14
i.
                                 CM
                                      20
30
40
                                                       (mln.)
                                                                            T:nul.
                                                                            Z: Ion range 60.00 to 600.00 am
                                                                            X:nuN.
             Flgurt 3-4. Mast iptc dwwmtoQnpli of ooidwood Mmpto extracted from quartz liber filter.

-------
                                           ton rang* 60.00 to 600.00 *mu. from wood6.4
             Abundance
ro
to
         1200000-
         1000000
          800000
          600000
          400000 -
         200000
                                                                                           8
                                                               Time (mln.)
                                                                                          Y:nuU.

                                                                                          X: nuU.
                       Figure 3-5.  Macs spec chroinatograph of particle board sample extracted from quartz fiber filter.

-------
                                  ton rang* 60.00 to 600.00 MIMI. from wood3.il
     Abundance
6000000
4000000
3000000
2000000
1000000
a.
Y:nuB.
X; nun.
           Figure 3-6.  Mass spec chromatograph of Formica0 board sample extracted from quartz fiber filter.

-------
                          Ion rang* 60.00 to 600.00 unu. from wood 44 SUBTRACTED SCALED
  Abundance
 200
-200-
-400
-600-
                                                                           particle board
               6         8        10        12
 14        16
Time (mln.)
18
22        24
                                                                                  Y:null.
                                                                                  X: nud.
       Figure 3-7. Mass spec chromatograph subtraction of cordwood sample from particle board (XAD-2 resin extract).

-------
  Abundance
 600
  400
  200
-zoo
-400-
                        ton range  60.00 to 600.00 «mi. from wood6.d SUBTRACTED SCALED
                                                             formica board
                                                                     J^wp*\Jk^
                               10        12
 14        16
Time (mln.)
18       20        22       24
ft.
                           Y:nuB.
                           X; nuB.
 Figure 3-8.  Mas* spec chromatograph subtraction of cordwood sample from Formica0 board (XAD-2 resin extract).

-------
  Abundance
                         ton range  60.00 to 600.00 unu. from wood6.4 SUBTRACTED SCALED
 600
 400
 200
-200 H
-400
-600
-BOO
           12
16
36
                                                                                 V: nuU.
                                                                                 X: null.
          Figure 3-9. Mass spec chromatograph subtraction of cordwood sample from particle board (filter).

-------
                         ton rtngo 60.00 to 600.00 unu. from wood 3 .d SUBTRACTED SCALED
•200
-400
-600
-800-1
           12
16
36

                                                                                V: nuN.
                                                                                 : nun.
          Figure 3-10.  Mass spec chromalograph subtraction of cordwood sample from Formica0 board (filler).

-------
1300
                                                                            704
 300
          i	1	1	1	1	1       r	1	r	149
00:01  00:33  01:05  01:37  02:09  02:41  03:13  03:45   04:17  04:49  05:21
  cordwood
                     particle board
                                     time
Formica board
                               Figure 3-11. OEM temperature.

-------
      o
      o»
      I
O)
                       ~1	1	1	1	1	1	1	1	1	T
              00:01  00:33  01:05  01:37   02:09  02:41  03:13  03:45   04:17  04:49  05:21
      — cordwood
+   particle board
                                                time
Formica board
                                         Figure 3-12. CEM O2.

-------

-------
   36,000
   34,000
   32,000
a.
TJ
g
o
c
o
.o
        00:01  00:33  01:05  01:37  02:09  02:41  03:13  03:45  04:17  04:49  05:21
 —  cordwood
                                         Time
particle board
Formica board
                              Figure 3-14. OEM CO.

-------
                                n	~  i	^	^~~r	1	1	1	r
        00:01   00:33  01:05  01:37  02:09  02:41   03:13  03:45  04:17  04:49  05:21
—  cordwood
particle board
                                           time
—  Formica board
                             Figure 3-15. CEM total hydrocarbon (ppm).

-------
00:00
00:14
00:28
00:43     00:57     01:12
  time since fuel addition
01:26
:40
01:55
                 temperature	CO
                                 O2 - - CO2
                                          scale wt.	THC
                 Figure 3-16.  OEM data for cordwood bum (data normalized for comparison).

-------
00:00     00:14    00:28    00:43     00:57    01:12     01:26     01:40     01:55     02:09
                                   time since fuel addition
               temperature	CO
O2	CO2
scale wt.	THC
                    Figure 3-17. OEM data lor particle wood burn (data normalized tor comparison).

-------
temperature
       time since fuel addition
CO       O2	CO2
scale wt  	THC
          Figure 3-18. CEM data for Formica6 bum (data normalized tor comparison).

-------
                                         SECTION 4



                              DATA RESULTS AND DISCUSSION







        Examination of the sampling data from Table 2-2 clearly shows that these sampling tests were



 not equivalent.  During the October 10,1992 particle board burn, high stack temperatures were



 observed when the previous day's  cordwood settings were used. These stack temperatures were high



 enough that the diluted sample presented to the sampling media was well above ambient temperature.



 Attempts to control the bum rate with the inlet air were ineffective because of the cast grating draft



 control. The flue damper was then adjusted to provide greater dilution which successfully reduced the



 diluted stack temperature.  Although Acurex Environmental could have modified the draft control to



 enhance its operation, the testing and modification of barrel stove kits was not the purpose of this



 project.



       These high stack temperatures were the result of a higher bum rate.  The composite woods



 (particle board and Formica6 board), are burned as small scraps which provide a high surface



 area/mass of fuel. This ratio increases the volatilization of gaseous components to the combustion



 zone. Additionally, a large fraction of the composite wood consists of synthetic resins that are likely to



 have lower molecular weights and, by extension, higher vapor pressures than the wood components.



 These resins may also provide some of the oxygen necessary for combustion since they are



 manufactured from phenols and aldehydes.



       Figure 3-11 compares the stack temperatures for the three fuels burned. It clearly shows the



different nature of the composite wood combustion relative to cordwood.  Manufactured wood



combustion is characterized by higher maximum temperatures, sharp peaks, and deep valleys. Figures




                                              43

-------
3-12 through 3-15 show this same periodic nature during composite wood combustion for the other



CEMs. Cordwood combustion CEM data, in contrast, show a more diffuse signature.



       Figures 3-16 through 3-18 also present this different nature of the composite wood combustion



by displaying the CEM data for the three fuels along with the scale data. The changes in slope for the



scale data suggest that combustion occurs in multiple stages.  The first stage is characterized by



volatilization.  Temperature and CO2 levels remain low while hydrocarbons are emitted.  CO tends to



follow the total hydrocarbon trend. The second stage may be described as char combustion. At the



beginning of this stage, the levels of hydrocarbons and CO go through a valley while CO2 levels and



temperature rise.  As expected, oxygen is the inverse of CO2.  Apparently,  a separate  and distinct



hydrocarbon emission is associated with each "stage" of the scale signal.



       These relationships are present, to a lesser extent, in Figure 3-14 for the cordwood burn. The



less distinct relationships are caused by overlap between fueling cycles when cordwood is burned.



During the first cordwood fueling a distinct change of slope can be  discerned in the scale data and the



described relationships are visible. The second fueling occurs before the first has ended, meaning that



the second volatilization stage overlaps the first char combustion.  Even here, separate hydrocarbon



emission phases may be observed. CO is related to hydrocarbon emission, and the stack temperature



drops somewhat until hydrocarbon emission  is finished.



       Table 3-2 summarizes the emission data for the three burns. Clearly, the two composite woods



produce less CO and total hydrocarbons than cordwood, but significantly greater amounts of the



heavier compounds. This result is seen in the total capture and the VOCs (Because the Formica6



board consists of particle board plus a laminate, its VOC emissions are likely to be similar) where the



Formica6 board total capture is twice the amount, and the particle board VOCs is nearly three times



that of cordwood on a mass/mass basis. In terms of filter capture. NVOCs, and SVOCs, Formica*



board has uniformly greater emissions than cordwood. Particle board, however, is actually closer to



cordwood in these factors than it is to the Formica6 board.  In this study, particle board emitted the



lowest SVOC levels.
                                              44

-------
       As discussed earlier, the total capture includes the gravimetric analysis of the probe rinse, the



filter capture, and the XAD-2 EOM. The reason why TCO analysis is not performed on the probe rinse



is that, just as the probe rinse is a minor constituent of the total capture value, previous experience has



shown that SVOCs are a minor constituent of the probe rinse sample which is to be expected as the



probe rinse represents those materials which condense from the diluted sample stream. In other



words, these components have boiling points greater than those which condense on the filter implying



that most SVOCs, which collect  initially on the probe, can be expected to return to the gas stream over



the sampling period.  Those remaining SVOCs are captured by occlusion in the heavier components.



Additionally, the solvent used to  prepare a probe rinse  sample is acetone.  The primary purpose of this



operation is  cleaning the probe.  Methylene chloride has been  shown to be incapable of completely



removing the residues from the probe. The more polar acetone removes these residues more



efficiently. Acetone is more reactive than methylene chloride and will modify some of the sample



components. This sample modification is not particularly significant from a mass distribution point of



view but  would lead to questionable results when extract components are identified  by GC/MS.



       Aldehyde data are not reported because all results were below the detection limit of the



analysis.   It is not clear how to interpret these results.  Several discussions were held  regarding this



analysis.  All indications during sampling operations were that  DNPH tube exposure proceeded



normally.  Sample flow through these  tubes was recorded at 10-min intervals.  These records have



been reviewed, and no abnormalities were found.  A distinct colored band was observed to form in the



front tube and gradually move down over time. The sample tubes were refrigerated for the time



between  sampling and transfer to the  analytical laboratory. Roy Zweidinger3 has confirmed that



refrigerated,  derivatized aldehydes should remain stable over the time between sampling and analysis.



No evidence that the samples were improperly collected or treated has been found.



       Aldehydes were anticipated before this study was conducted based on the use of phenols and



formaldehyde in manufacturing these  resins and on previous studies which found aldehydes during the



combustion of cordwood in air-tight woodstoves.  The CEM data make it clear that combustion occurred






                                              45

-------
very differently in this barrel stove than in an air-tight stove.  In a recent study of two non-catalytic



woodstoves, for example, oxygen varied from 13-19 percent, CO2 varied between 1-7 percent, and



stack temperature ranged from 150-600 °F. These values are quite different from those reported in



Table 3-1.  A fireplace might provide a better comparison. We are not aware of any fireplace studies



that included aldehyde results. However, based on this very limited data set, it cannot be concluded



that aldehydes are not formed during wood combustion in a barrel stove.



        VOC data were obtained only from the cordwood and particle board tests, the Formica8 board



sample was non-detectable.  However, the total hydrocarbon data presented in Table 3-1 and Figure 3-



15 for the Formica® bum suggest that this sample must have been bad, either during collection or



handling before analysis.  This VOC sample was inadvertently allowed to sit for a significant time before



delivery to AREAL for analysis.  In all likelihood, sample components reacted and condensed onto the



container walls.  Unfortunately, the limited funding of this study did not permit a repeat bum.



        Table 3-3 presents these VOC results. In terms of relative concentrations compounds past



benzene, certainly those past toluene, in retention time do not represent significant components of the



sample. Benzene represents less than 0.5 percent of the cordwood sample and less than 0.1 percent



of the particle wood sample.



        The majority of the VOC mass was reported at the retention times of 10-15 min, which is the



 range containing 4-6 carbons. During this range, the column overloaded making integration of the



peaks difficult.  Thus all of the compounds within this range were reported as a few components, the



overloading occurred for both samples. This range represents 97.8 percent of  collected mass for the



cordwood test and 99.8 percent for the particle board.  The two samples had many of the  same light



molecular weight compounds, but the cordwood had more of the heavy compounds. The



chromatograph contains many unidentified compounds. Those compounds were included in the table



to reflect the quantity and types of compounds that may be found in that range. Some differences



between the two can be attributed to the type of wood and to the binders used in the particle board.
                                              46

-------
       The filter and XAD-2 extracts were analyzed by GC/MS to identify their components. Figures
3-1 through 3-6 represent the total ion chromatograms (TICs) from the analysis of these six samples.
Figures 3-7 through 3-10 present the difference spectrum for the four composite wood samples minus
the corresponding cordwood sample. Figures 3-9 and 3-10 are the most striking showing a strong
trend to longer retention times for the filter samples from the composite woods. The fact that the GC
column used for this work separates primarily on the basis  of boiling point, suggests a trend towards
higher molecular weight components for the composite woods.  Tables 3-4 through 3-6 present the
library search results for these GC/MS tests. Each table includes both the filter and XAD-2 sample for
that wood.  Table 3-7 presents the easiest comparison of the compounds generated from the three
burns while Table 3-8 presents some observations by compound class.
       1.  Two compounds (2,4-hexadiene-1-ol and naphthalene) were observed for all three woods.
       2.  Five compounds (6,7,10,12, and 18) were not found in the particle board samples.
           However, this absence may be  due more to the lower total fuel burned during the particle
           board burn than to any real differences in combustion chemistry.
       3.  The majority (21,25,26,29, and 31) of the polycyclic aromatic hydrocarbons (PAHs) are
           observed only in the cordwood samples.
       4.  None of the saturated hydrocarbons were observed in the cordwood sample. Nearly half
           of the compounds identified in the manufactured fuel samples were saturated
           hydrocarbons.
       5.  4-Hydroxyl-benzenesulfonic acid is found in both the manufactured fuel samples while
           isocyano-benzene is found only in the Formica0 board samples.  The presence of this
           compound suggests the starting materials for Formica6 laminate  described in Hawley's
           dictionary.
       GC/FID  analysis and GC/MS analysis were performed with the same type of column and oven
temperature program.  Unfortunately, the GC/MS's vacuum changed those retention times enough to
make peak  matching between the two analyses extremely difficult.

                                             47

-------
       Ash samples were analyzed by GC/FID (TCO) and gravimetric methodology. The mass



collected from the ash samples was below the quantifiable limits of the GRAV method and the TCO



detection limit.
                                            48

-------
                                        SECTION 5
                                   QUALITY ASSURANCE'


       Field and lab blanks were collected to establish background emission levels. Field and lab
blanks were collected for XAD-2 cartridges and filters while only field blanks were collected for DNPH
tubes.  No blanks were collected for the probe or the VOC canisters because each canister is analyzed
before sampling.  Field blanks were delivered to the sampling site, opened, reseated, and returned to
the lab.  Lab blanks remained sealed until extraction.  XAD-2 and filter emission results were blank
corrected. Table 5-1 presents the percent of blank mass compared to the average of the actual sample
mass.
       Completeness for data recovery is described in Table 5-2. DNPH tubes yielded non-detectable
samples. One VOC canister failed to yield a sample, but all other samples were intact.  Conditions and
observations recorded during and after sampling indicated that samples had  been collected by these
techniques. More than two months elapsed between sampling and analysis. Samples may have been
lost or degraded during this period.

              TABLE 5-1. PERCENT BLANK MASS OF AVERAGE SAMPLE MASS
         % field blank of avg XAD-2 TCO                                  0.07
         % field blank of avg XAD-2 GRAV                                0.91
         % field blank of avg filter TCO                                   4.17
         % field blank of avg filter GRAV                                   1.46
         % field blank of avg filter total capture                            0.40
                                             49

-------
                            TABLE 5-2.  COMPLETENESS OF DATA

OEMs
Aldehydes
VOC
GRAV
GC/FID
GC/MS
Filter capture
Probe rinse
Data points
15
13
3
8
8
6
4
3
Completeness
100%
100%
67%
100%
100%
100%
100%
100%
       CEMs were calibrated before and after each test using three different concentrations of span



gas appropriate to each instrument.



       The balance used for gravimetric analysis was sensitive to 10 ug/weighing, but any mass less



than 6 mg/sample was determined as below quantifiable limits, and any mass less than 1.2 mg/sample



was considered to be below detectable limits as follows:



        •  Detection limit = (10 ug)* (3)* (sample volume 10 mL)/(aliquot volume 0.25 mL) = 1.2 mg



        •  Quantifiable limit = detection limit (1.2 mg)* (5) = 6 mg



       The GC/FID used for TCO analysis had a quantification limit of 0.014 ug and a detection limit of



0.003 u.g.  The quantification limit was set at the average mass of three hydrocarbons in our lowest



concentration calibration standard. The detection limit was established at one-fifth the quantification



limit.
                                              50

-------
                                         SECTION 6



                               SUMMARY AND CONCLUSIONS







       This study determined a number of differences between the combustion of composite woods



 and cordwood.  These composite woods burn faster than cordwood because of the higher surface area



 of these composite woods, which are burned as scraps, relative to the same mass of cordwood.



 Higher stack temperatures and oxygen concentrations, and lower CO and total hydrocarbons



 (mass/mass basis)  were observed during combustion of these composite woods versus cordwood.



       VOC levels are much higher during the combustion of these composite woods with the major



 components being in the C4-C6 region. Total emission levels (based on the total capture value) are



 also higher for these composite woods.  Higher total capture results such as these are due, in large



 part, to higher NVOC  levels. SVOC levels, on the other hand, are equivalent (Formica0) or even lower



 (particle board) than those generated by cordwood. There is a trend toward larger molecular weight



 components for these emissions. The filter extracts for these composite woods show higher



 concentrations of higher retention time analytes during the GC/MS analysis.  These components were



 primarily straight chain hydrocarbons.



       Significant differences were observed in the compounds identified from the extractable



 organics. A majority of the PAHs are associated with the cordwood  rather than the composite wood



 combustion. Additionally,  isocyano-benzene was identified from the  Formica0 samples and



4-hydroxyl-benzenesulfonic acid was found in the composite wood samples.



       No  aldehydes were detected from any of the samples collected during this study.  The meaning



of this information is not clear.  Based on air-tight woodstove studies, aldehydes were expected from at




                                             51

-------
least the cordwood samples. However, the combustion conditions during this study are probably closer



to those of a fireplace than a woodstove. No significance can be attached to these aldehyde results



without further testing.
                                              52

-------
                                       SECTION 7

                                      REFERENCES
1.      Sax, N.I., and Lewis, R.J., Hawley's Condensed Chemical Dictionary 11th ed.; Van Nostrand
       Rinehold Company, Inc., New York, 1987.

2.      Lentzen, D.E., Wagoner, D.E., Estes, E.D. and Gutknecht, W.F. (1978) IERL-RTP Procedures
       Manual; Level 1 Environmental Assessment 2nd ed.; EPA-600/7-78-201 (NTIS PB 293795).

3.      Personal communication from R. Zweidinger, USEPA/AREAL, to M. Tufts, Acurex
       Environmental, Oct. 1,1992
                                           53

-------
                                      APPENDIX A

                   RELATED RECOMMENDED OPERATING PROCEDURES


ROP No.       Title                                                                 Page

AEERL/12      Gravimetric Analysis of Organic Extracts (Interim)	 A-2
AEERL/13      Total Chromatographable Organics (TCO) Analysis (Interim)	 A-10
AEERL/22      Extraction of Filters and Solids (Interim)	 A-21
AEERL/40      Large-scale XAD-2 Resin Purification (Draft)  	 A-29
AEERL/41      Sample Recovery from XAD-2 Resin by  Pump Through Elution (Draft)	 A-39
                                         A-1

-------
              IHTCRIM PROCEDURE
  This pranhn has been autborM tar idmnci MM          Document  No:   AEERL/12
                                                       '
                                                           n    ,
  undersoottnstv.EPArw.ewpnoftofin.lization.             oJte-  9/2/86
                                                Page  1  of  8
                  RECOMMENDED OPERATING PROCEDURE  FOR
               GRAVIMETRIC ANALYSIS OF ORGANIC EXTRACTS
                                   By

                           Robert F. Martz *
                             Monica Nees **
                              Prepared for

                   The AEERL TECHNICAL SUPPORT OFFICE
 0isc1 aimer:   This recommended operating procedure  has  been prepared for
 the sole use of the Air and Energy Engineering  Researcn Laboratory,
 U.S. Environmental Protection Agency, Research  Triangle Park, North
 Carolina, and may not be specifically applicaole to  the activities of
 other organizations.
 * Chemist
   Acurex Corporation
   Research Triangle Park, NC

** Research Environmental Scientist
   Research Triangle Institute
   Research Triangle Park, NC
                                  A-2

-------
                                             Document No:  AEERL/12
                                             Status:  INTERIM
                                             Revision No:  0
                                             Date:  September 2, 19«b
                                             Page 2 of 8
       RECOMMENDED  OPERATING PROCEDURE FOR GRAVIMETRIC ANALYSIS
                          OF ORGANIC EXTRACTS
1.0  PROCEDURAL ELEMENTS
1.1  Scope and Application
     Organic compounds with boiling points of 300°C and higner, after
     extraction with methylene chloride, evaporation of the solvent,
     ana drying to constant weight, can be determined quantitatively by
     the gravimetric analysis described in this procedure.1 This
     method is applicable to organic liquids, solid sample  extracts,
     aqueous extracts, and extracts from the Source Assessment Sampling
     System or Modified Method 5 train sorbent module.  This analysis
     should be performed after enough of the sample extract has been
     concentrated to weigh accurately.2  The suggested solvent is
     methylene chloride because of its good extraction properties and
     high volatility.  Other solvents may give different results (e.g.,
     methyl alcohol  may extract polar compounds which would not be
     extracted with  methylene chloride).  All samples being driea -to
     constant weight should be stored in a desiccator.
     The range of applicability is limited by the sensitivity of the
     balance and the organic content of the sample.  The balance must
     be accurate to  +_ O.U1 mg.  If a sample of five mi Hi liters  is
     used for the analysis, then a sensitivity of O.U1 mg/5 mL or
     0.002 mg/mL of  sample can be achieved.  This can be improved by
     further concentration of more sample.
                               A-3

-------
1.2  Definitions

     o  Method Blank:
                                               Document No:  AEERL/12
                                               Status:  INTERIM
                                               Revision No:  U
                                               Date:  9/2/86
                                               Page:  3 of 8
Provides  a check on contamination resulting from
sample preparation ana measurement activities.
Typically run in the laboratory after receipt of
samples from the field by preparing a material
known not to contain the target parameter.
Addresses all chemicals and reagents used in a
method.
        Reagent Blank: Provides information on contamination due to
                       specific chemical reagents used during sample
                       preparation, plus any background from the
                       measurement system.
     o  Audit Sample:
Has known "true values,11 but is flagged for the
laboratory as a "performance evaluation (PE)
sample."  Provides information on performance,
but this information must be tempered with the
understanding that the sample may be given extra
attention by the analyst.  An internal PE sample
is created by the in.-house analytical
laboratory, while an external PE sample is
created outside of the analytical laboratory.
1.3  Interferences
     Results may be biased due to contamination of  the  solvent,  glass-

     ware, or  both.  A method blank (control)  shall  be  run  in  duplicate
     for each  run  lot of  solvent and/or set  of  samples  to  provide  a
     control check on the purity of the solvent and  the glassware
     cleaning  procedure.  The method blank,  consisting  of  a solvent
     sample  from the same lot as that used to  prepare samples,  shall  be
     prepared  and  concentrated in an identical  manner.

     Two reagent blanks  shall be analyzed each  day  samples  are run to
     ensure  results which are not biased due to solvent contamination.
     The reagent blank  shall be a solvent sample  from the  same lot used
     to prepare the samples  and shall not be concentrated  prior to
     analysis.  To minimize  error in weight  due to  moisture condensa-

-------
                                               Document No:  AEERL/12
                                               Status:  INTERIM
                                               Revision No:  0
                                               Date:  9/2/86
                                               Page:  4 of 8

tion, the pans containing the sample must appear visually dry before

being placed in a desiccator in preparation for drying to constant

weight.

1.4  Apparatus

     (1)  Analytical Balance:  Capable of weighing 0.01 mg with an
          accuracy of +_ U.U05 mg.

     (2)  Desiccating Cabinet:  Seal-tight door gasketed with gum
          ruober.   (Desiccators which use silicone sealant snail  not be
          used because of possible contamination of the sample.
          Silicone grease may interfere with subsequent analysis.)

     (3)  Oven:  Capable of operation to 175°C.

     (4)  Fume Hood:  Standard laboratory.

     (5)  Dust Cover, Plexiqlas, or equivalent:  To protect samples
          drying in hood.

1.5  Reagents and Materials

     (1)  Disposable Aluminum Weighing Pans:  Approximately 2" in
          diameter, 1/2" deep; crimped sides; weighing approximately
          1.0 grams.

     •2)  Tweezers.

     (3)  Alumi num Foi1.

     (4)  Pi pets:   1 to 5 ml  (Class A Volumetric).

     (5)  Glass Beakers:  50 to 400 ml.

     (6)  Wash Bottles, Teflon or equivalent.

     (7)  Deionized Mater.

     (8)  Nitric Acid/Sulfuric Acid, 50:50  (V/V);   Prepared  from
          reagent-grade acids.
                                A-5

-------
                                               Document No:  AEERL/12
                                               Status:   INTERIM
                                               Revision No:  U
                                               Date:   9/2/86
                                               Page:   5 of 8
     (9)   Methylene  Chlonae:   Burdick  and Jackson or equivalent grade.
     (10)  Methyl  Alcohol:   Burdick  and  Jackson or equivalent grade.
     (11)  Drlerite  and/or  Silica  Gel:   New Drierite or  silica gel  may
          be  usea as received.   Used  Drierite or silica gel may be
          reactivated by drying it  in an  oven for at  least two hours at
          175°C.
1.6   Samp)e Hanoiing
     All  apparatus  that  contacts  either the concentrated or evaporated
     residue  samples snaM  be  glass.  Teflon,  aluminum,  or stainless
     steel.  Evaporation of samples  shall  be  carried  out in an area
     free  of  airborne dust  and  organic  vapors that could contaminate
     the  samples.
     Ordinarily,  all glassware  coming in  contact with a sample, in
     either dilute  or concentrated  form,  must be cleaned by complete
     Level  1  procedures.2   Briefly,  this  entails sequential cleaning
     with  soapy water, deionized  water, 50:50 (V/V) nitric acid/
     sulfuric acid,  deionized  water, methyl  alcohol,  and methylene
     cnloride,  followed  by  oven drying.  The  use of deionized water for
     cleaning glassware  is  critical  when  inorganic substances are  being
     analyzed or heavy metal  contaminants  are present in high
     concentration  in tap  water.
     This  ROP,  however,  covers  only  the analysis of organic
     constituents.   Tap  water  can be  substituted for  deionized water in
     glassware  cleaning  whenever the  organic  concentration exceeds one
     mg/sample  as measured  by  this  ROP.  Experience has shown that tap
     water adds no measureable amount of organic contaminants to the
     method or  reagent blanks  under these conditions.
1.7   Sampling/Analysis Procedures
     (1)   Label aluminum sample pans  on the underside using a ballpoint
          pen or other sharp object.  Handle  dTshes only with clean
          tweezers.
                                A-6

-------
                                               Document No:  AEERL/12
                                               Status:  INTERIM
                                               Revision.No:  0
                                               Date:  9/2/86
                                               Page:  6 of 8

     (2)  Clean the weighing  pans  by first rinsing them with deionized
          water, then dipping them successively Into three beakers of
          methyl alcohol,  methylene chloride, and, finally, methyl
          alcohol again.

     (3)  Dry the cleaned  weighing pans to constant weight on a shelf
          lined with clean aluminum foil  in an oven heated to at least
          105°C.  Cool  the pans  in a desiccator for a minimum of 4 to 8
          hours or overnight.

     (4)  Weigh pans to constant weight to an accuracy of _* 0.01 mg,
          recording the pan tare weight.

     (5)  Transfer by pipet a 1.0  ml aliquot of the sample to the
          aluminum sample  pan or use 1/10 of the concentrated sample.
          Aliquot size must never  exceed  5 mL to avoid loss of sample
          througn capillary action.

     (6)  Place the sample pan on  a clean piece of aluminum foil in a
          clean fume hood.  Shield the pan from dust with a Plexiglas or
          other cover positioned to allow for adequate air circulation.
          Evaporate sample to visual dryness at room temperature.  This
          usually takes about 30 minutes.

     (7)  Place sample pan in desiccator over Orierite and/or silica gel
          for at least 8 hours.
                                V

     (8)  Weigh sample pan at approximately 4-hour intervals until three
          successive values differ by no more than +. 0.03 mg.  .If the
          residue weight is less than 0.1 mg, concentrate more sample
          1n the same sample pan.    If there is Insufficient sample
          remaining for this purpose, report the initial value obtained,
          along with an explanation.

1.8  Calculations

     The gravimetric range organics (GRAV) Is calculated in units of

     mg/sample as follows:

     GRAV «  (Sample Heightmg + Pan Ue1gntmg) - (Pan Tare Weightmg)
              Aliquot Vo)umem]/Total Concentration Sample Volumem]

     The calculated GRAV weight 1s corrected for the method blank:

               GRAV CORRECTED - GRAV MEASURED - METHOD BLANK
                               A-7

-------
                                               Document No:  AEERL/12
                                               Status:  INTERIM
                                               Revision No:  J
                                               Date:  9/2/86
                                               Page:  7 of 8
1.9  Data Reoortlna
     The results of the analysis are averaged and  reported 'in units of

     mg orgamcs/ongirial  sample.

1.10 Precision

     Duplicate analyses snail be run by the same analyst and shall be

     rejected if results differ by more than 20% from the average.  If

     insufficient material is present to rerun the  sample, both  values

     will be  reported with a qualifying statement.

1.11 Accuracy

     Dry sample weight  should be at least  1 mg per  analysis wnenever

     possible.  Accuracy of the analysis is +_ 20%  of actual value.  A
     proficiency test should be performed  by each  analyst as described

     in Section 2.0


2.0  QUALITY  CONTROL  ELEMENTS

     o  All  operators  should demonstrate proficiency with Gravimetric
        Analysis of Organic  Extracts  (GRAV) prior  to sample analysis.
         In  the  proficiency testing,  include a GRAV analysis of  a
         reagent blanic,  a  method  blanK,  and an audit sample.   The method
        or  reagent  blank  shall  be  less  than 5 mg/mL of  sample.   Results
        of  the  audit  sample  shall  be within the  precision  and  accuracy
         specifications  outlined  in this ROP.

     o   Two types  of  audit samples are  used.  The  first contains 1UO mg
        of  eicosane [C^CHghaC^]  in  250 mL of methylene  chloride.
         Concentrate this  solution  to  10 mL in a  manner  identical to
         that  used  for sample preparation  prior  to  GRAV  analysis.  The
         second  type of  audit sample  can be either  prepared  in-house or
         received  from an  independent  laboratory.  It must  contain
         organic compounds with  chain  lengths  of  more than  18  carbons
         (and boiling  points  above  30U°C)  in  sufficient  concentration to
         be  determined accurately.   Perform the  GRAV analysis  in
         duplicate  as  described  in  Section  1.7  of this  procedure.

                                A-8

-------
                                               Document No:  AEERL/12
                                               Status:  INTERIM
                                               Revision No:  0
                                               Date:  9/2/86
                                               Page:  8 of 8

     o  Determine the GRAV value of duplicate method blanks for each
        new lot of solvent and/or set of samples.  Run a method blank
        any time contamination is suspected.   Prepare the blank using
        the same lot of reagent and the same  concentration procedure as
        that used .to prepare the samples.   The solvent sample shall be
        an equivalent volume to that used  for sample preparation.  If
        the blank GRAV value is unusually  high (i.e., 5 mg/'mL of
        sample), find the cause of the contamination and repeat the
        method blank GRAV analysis.

     o  Analyze two reagent blanks for GRAV each day samples are run to
        ensure the results are not biased  due to solvent contamination.
        The reagent blanK shall consist of an aliquot of the solvent
        used to prepare the samples.  If both reagent blank GRAV values
        are high (i.e., 2 mg/mL of sample), find the cause of the con-
        tamination and reanalyze samples and  reagent blanks.

     o  Analyze all samples in duplicate.   Samples are analyzed by the
        same analyst and must agree to within 20% of the average.  In
        the event this condition is not met,  repeat the analyses.

        NOTE:   If the conditions require the  sample to be re-
               analyzed (e.g., high blank  values or poor precision)
               and insufficent sample remains, then report the
               value obtained by the initial  analysis and include
               a qualifying statement.

2.0  REFERENCES

     1.   Harris, J.C. et al.  Laboratory Evaluation Level 1 Organic
         Analysis Procedure.  EPA-60U/S7-82-048, NTIS PB 82-239, pp.
         30-36, March 1982.

     2.   Lentzen, D.E., D.E. Wagoner, E.D. Estes, and W.F. Gutknecht.
         IERL Procedures Manual:  Level 1  Environmental Assessment
         (Second Edition).  EPA-600/7-78-201, NTIS  PB 293-795,  pp.
         26-142, October 1978.
                                 A-9

-------
            iMTTPiupenrrniior                   Document  No:   AEERL/13
            INTERIM PROCEDURE                   status.   INTERIM

This procedure nas been auttoriad for reference a* a«        Revision  No:   3
KitBnmproceiur a by theAQRLQAQ. The procedure will        Date:   9/25/86
            lEPArtMMrpnartBfinalizatiofl.           Page:   1  of  11
       RECOMMENDED OPERATING  PROCEDURE FOR TOTAL
       CHROMATOGRAPHABLE  ORGANICS (TCO) ANALYSIS
                            by

                        R.  Marti**
                     Josepn D. Evans*
                       Prepared for

            The  AEERL TECHNICAL SUPPORT OFFICE
 Disclaimer:   This recommended operating  procedure has been
 prepared for the sole use of the Air  and Energy Engineering
 Research Laboratory, U.S. Environmental  Protection Agency,
 Research Triangle Park, North Carolina,  and  may not be
 specifically applicable to the activities of other organi-
 zations.
 **  Chemist
     Acurex Corporation
     Research Triangle Park,  NC

  *  Environmental Chemist
     Research Triangle Institute
     Research Triangle Park,  NC
                           A-10

-------
                                               Document No:  AEERL/13
                                               Status:  INTERIM
                                               Revisiort No:  3
                                               Date:  9/25/86
                                               Page:  2 of 11
               RECOMMENDED OPERATING PROCEDURE FOR TOTAL
               CHROMATOGRAPHABLE ORGANICS (TCO) ANALYSIS
1.0  PROCEDURAL ELEMENTS
1.1  Scope and Application
     This method provides semi-quantitative data for organic compounds
     with boiling points between 1UO°C and 30U°C.  Samples that might
     include organic compounds  in this volatility range are organic
     liquias, solid sample extracts, aqueous extracts, extracts from
     Source Assessment Samplng  System (SASS) and Modified  Method 5
     (MMb) train sorbent modules, and liquid chromatography (LC)
     fractions obtained from those samples.  This method is based on
     separating the components  of a gas or liquid mixture  in a  gas
     chromatography (GC) column and measuring the separated components
     with a suitable detector.
     The upper end of applicability is limited by column overloading
     and detector saturation.  Typical range is 1 to 20 mg/mL.   The
     operating range can be extended by dilution of samples with
     solvent (e.g., dichloromethane).  The sensitivity limit shall be
     determined by the minimum detectable concentration of
     standards.

1.2  Summary of Method
     TCO analysis quantifies chromatographable material with boiling
     points in the range of 100° to 3UO°C.  This analysis  is applied to
     all samples that might contain compounds  in this volatility  and
     boiling point range.
     For TCO analysis, a 0.9- to 3-uL portion  of the  extract is
     analyzed by gas chromatography using  a flame  ionization detector
     (F.I.D.).  Column conditions  are described  in  this document  in
     tabular form in section 1.5.
                               A-11

-------
                                                   Document No:   AEERL/13
                                                   Status:   INTERIM
                                                   Revision No:   3
                                                   Date:   9/25/86
                                                   Page:   3 of 11
     The  peak  areas  are  converted to concentration values  using quanti-

     tative  calibration  standards.

     For  more  information,  consult Lentzen  et al., IERL Procedures

     Manual:  Level  1 (reference  1).

1.3   Definitions

     0   QC Sample:
        This sample is prepared  from a  stock  solution in an identical
        manner as the calibration standard.  Its concentration is
        approximately miaway in  the  linear  working range of the GC. This
        quality control  (QC) sample  is  run  daily along with the sample
        set.

     0   Method Blank:
        Also called concentrated solvent  blank, the method blank provides a
        check on contamination resulting  from sample preparation
        activities.  It is typically prepared in the laboratory alongside a
        sample set by "extracting"  and  concentrating the  appropriate
        amount of clean solvent  in  the  desired size extraction apparatus.

1.4  Interferences

     The analytical system shall be demonstrated to be free from  internal

     contaminants on a daily basis  by  running a bakeout or a QC sample.  A

     reagent blank must be run for  each new batch of reagents used to

     determine that reagents are contaminant-free.  This is verified by  an

     instrument  response less than  the detection  limit.

     If duplicate runs of a sample  show increasing concentration  greater

     than 15%,  or if cross-contamination is  suspected  (e.g.,  high-level

     sample followed by a  low-level sample), a  reagent  blank  shall be  run

     to verify  no contamination in  the system.   If contamination  is

     evident,  the column shall  be baked out  at  approximately  25U°C for 20
     minutes  or  until the  detector is  stable,  and  the  blank check
     repeated.

                                A-12

-------
                                                      Document No:  AEERL
                                                      Status:  INTERIM
                                                      Revision No:  3
                                                      Date:  9/25/86
                                                      Page:  4 of 11

1.5   Personnel  Requirements
     This ROP  1s  written  for  Individuals  with a BS/BA degree In chemistry
     and at  least two years experience  In gas chromatography, or
     equivalent.
1.6   Facilities Requirements
     This procedure  requires  a  standard analytical  chemistry laboratory
     with counter space,  secured  areas  for compressed gas storage, and
     electricity  to  operate the equipment.  Flasks, beakers, tubing,  etc.
     customarily  found  in such  a  laboratory are also needed and assumed  to
     be readily available.  GC  tools  (e.g., wrenches, screwdrivers,  spare
     parts,  etc.) need  to also  be available in the  laboratory.
1.7   Safety  Requirements
     Routine safety  precautions required  in any analytical  chemistry
     laboratory are  applicable  here.   These include such measures as  no
     smoking while in  the laboratory; wearing safety glasses, lab coats,
     and  gloves when handling samples; handling organic solvents in a fume
     hood,  etc.  Compresseo  gases considered to be  fuels (e.g., hydrogen)
     must be stored  on a  pad  outside  the confines of the laboratory.   A
     safety  shower,  eye wash, first aid kit, and fire extinguisher must  be
     readily available inside the laboratory.
1.8   Apparatus
     1.  Gas Chromatooraoh -  GC with packed column and/or capillary column
         capaoilities, oven  temperature controller, and flame  ionization
         detector (F.I.D.).   (e.g., Perkin Elmer Sigma 115 or Hewlett
         Packard  589U.)
     2.  Autosampler - (optional) - Capable of handling methylene chloride
         extracts and  appropriate wash vials.
     3.  Autosampler vials (optional)  - Clear  glass  vials with teflon faced
         crimp caps, typically 100 microllter  or 1 ml  size.
     4.  Crimping Tool (optional) - Used  to  secure  caps  on  autosampler
         vials.
                                   A-13

-------
                                                           Document  No:   AEERL
                                                           Status:   INTERIM
                                                           Revision  No:   3
                                                           Date:   9/25/86
                                                           Page:   5  of  11
                    INSTRUMENTAL  OPERATING  CONDTIONS  FOR  GAS CHROMATOGRAPHY

Column

Fused
Silica
Capillary
Column
(15 meters
typically
DB-1, DB-5,
or equiva-
lent)
Packed
Column
(Methyl
Si 11 cone
oil coated
at 10% on
Supelcort
AW DMCS or
equivalent
1/8 In. x
6 ft.
steel)
Temperature
Program
(optional )
40°C for 3
minutes
8°C/min
increase to
250° C and
hold for
total run
time of 45
minutes
50°C for 5
minutes

20°C/min
Increase
to 250°C,
then hold •






Injector

300°C








300°C












Detector

F.I.D.
300°C







F.I.D.
300°C











Carrier
Gas
Helium
1-3 mL/min







Helium
at 30
mL/min









Split
Injector
(optional ',
10/1
split
ratio






N/A












Injection
Volume
Not to
exceed
3 ul
(Typically
1 ul)




1-5 ul












Solvent

Dichloro-
methane
(pesticide
grade,
distilled
in glass or
equivalent)


Dichloro-
methane
(pesticide
grade,
distilled
In glass or
equivalent)





N/A = Not Applicable
                                         A-14

-------
                                                      Document No:  AEERL/13
                                                      Status:  INTERIM
                                                      Revision No:  3
                                                      Date:  9/25/86
                                                      Page:  6 of 11
1.9  Reaaents and Materials
     1.  Methylene Chloride:   Burdick and Jackson or equivalent grade.

     2.  Syringe - 5 or 10 microliter, gas tight, syringe for hand
         injections.  Otherwise, 3 or 10 microliter syringes are used for
         autosampler injections.

     3.  Disposable Pasteur Pipets - Used for sample transfer.

     4.  Pipet bulbs - 1 mL,  amber.

     5.  Teflon Squeeze Bottle - 250 mL, or equivalent, used for methylene
         cnlonae nnse of vials.

1.10 Samples/Sampling Procedures

     NOTE:  All glassware coming in contact with a sample shall be cleaned
            by Level 1 procedures  (ref. 1).  Briefly, this entails
            sequential cleaning with soapy water, deionized water, bO:bU
            (V/V) nitric acid/sulfuric acid, deionized water, methyl
            alcohol, and methylene chloride, followed by oven drying.

1.1U.1  Samplinu/Analysis Procedures

     (1)  Start up by the manufacturer's suggested method.

    *(2)  Replace septum on auto-sampler and column.

    *(3)  Insure  injection needle  is  in line with  injection port.   The
          autosampler needle should be manually  "injected"  through  the
          injection port to verify alignment.

     (4)  Bakeout GC at 20U°C for  20 minutes until F.I.D.  response  is
          stable  and all evidence  of  column  contamination  is  gone  (no
          peaks)  or run an injection  of clean  solvent  as the  first
          injection of the day  to  verify  column  contamination  is
          eliminated.

    *(5)  Load auto-sampler tray with  samples.

    *(5A) Check the autosampler  flush  by  placing the  autosampler  in manual
          mode and  flushing a vial of  clean  solvent  through the needle
          assembly.

                                  A-15

-------
                                                      Document  No:   AEEKL/13
                                                      Status:   INTERIM
                                                      Revision  No:   3
                                                      Date:   9/2b/86
                                                      Page:   7  of 11

   *(6)  Set  auto-sampler to  inject approximately 1  uL of samples.
         Capillary  column can be damaged  if  too great a volume is
         injected.

     (7)  Run  a  OC  standard  using the  specified  conditions to  verify
         that the  system  is operating  properly.  Checic the TCO window
          (Cy  -  Ci7  to  insure  the range  has  not  changed.  (Retention
         times  may  change with  column  aging.)   The TCO window  for
         calculations  should  be adjusted  as  required.

     (8)   Flusn  needle  with  solvent  (dichoromethane)  between injections.

     (9)  Run  samples  and  collect data.

     (1U) Analyze data  according to  prescribed  method.

     (11) After all  analyses are complete,  bakeout the column  at 200°C for
          20 minutes,  or run clean  solvent  as a  "sample."

     (12)   Shut down instrument  by  method  suggested by manufacturer.

* These steps  are only  applicable to automatic  injection.

1.1U.2  Preparation

     Samples for TCO analysis  arrive or are prepared as methylene chloride

     (or  occasionally  as methanol)  extracts of  environmental samples,

     filters,  resins,  or ambient sampling  components.  An aliquot of the

     extract is transferred  to a TCO vial  and loaded into the autosampler
     as required.

1.11   Sample Stability

     All  samples will  be stored  in  a refrigerator at or  below 4°C to  retard

     analyte degradation.   Samples  will be analyzed as  soon as possible

     after  sample receipt and  preparation  to avoid loss  of  sample due to
     volatilization and degradation.
                                 A-16

-------
                                                   Document No:  AEERL/13
                                                   Status:  INTERIM
                                                   Revision No:  3
                                                   Date:  9725/96
                                                   Page:  8 of 11
1.12  Calibration
     (1)  Preparation/dilution of a stock  solution:  Weigh approximately
          10U uL aliquots of each (heptane,  decane, dodecane, tetradecane,
          ana heptadecane, C7, CIO, C12,  C14,  C17) (99% + pure) into a 10
          mL volumetric flask or septum-sealed vial.  Quantitative
          calibration of the TCO procedure is  accomplished by the use of
          mixtures of known concentration  of the normal hydrocarbons
          decane, dodecane, and tetradecane.  Retention time limits
          correspond to the TCO range of  boiling points and are defined by
          the peak maxima for n-heptane (C7, B.P.   98°C) and n-heptadecane
          (C17, B.P. 303°C).  Therefore,  integration of detector response
          snould begin at the retention time of C7 and terminate at the
          retention time of C17.  The C7  and C17 peaks are not included in
          this integration.  By this procedure, the integrated area will
          cover material in the boiling range  of approximately 10U°C to
          300°C.  Weigh each hydrocarbon  successively into the vial
          starting from least volatile to  most volatile.

     (2)  Dilute the vial contents up to  approximately 3 ml with dichloro-
          methane.

     (3)  Transfer this quantitatively to  a  clean, 10-mL amber volumetric
          flask and add dichloromethane up to  the  10-mL mark.  This stock
          solution will have approximately 22  mg (C7 to C12)/mL and 15 ng
          (C14 to C17)/mL.  Several (at least  three) dilutions of the stock
          solution are made to cover the  linear working range.

1.13  Sample Analysis

     A portion of the extract is injected  into the GC under the conditions

     specified.  The peak area (F.I.D. response/uL) is summed over the TCO
     range window and corresponding TCO value  (mg/mL) is determined from

     the calibration curve.  In the event  that the TCO value is outside the
     linear working range, the sample shall  be concentrated or diluted,
     depending on the requirement, and re-analyzed.  If there is not enough

     sample to concentrate, the values are reported as found, and an
     appropriate qualifying statement is  included  in the analytical

     report.

     It i.s important that the observed values  of the total integrated area

     for samples be corrected by subtracting an appropriate solvent blank,

     prepared in the same manner as the samples.
                                   A-17

-------
                                                   Document No:  AEERL/13
                                                   Status:  INTERIM
                                                   Revision No:  3
                                                   Date:   9/25/86
                                                   Page:   9 of 11
1.14  Calculations
      The peak  area  (F.I.D.  response/uL)  is  summed over the TCO .window and

      a corresponding TCO value (mg/mL)  is  determined from the calibration

      curve.

      (1) Construct  the calibration line  by  fitting a linear regression
          equation to the results  of the  analysis of the calibration
          standard solution.  The  concentration of the standards must
          fall  within the linear working  range of the instrument and
          bracket the concentration of the  sample.  Use the CIO to
          C14 standards for calibration.

          Standard Calibration Equation:

               Rl = (M) C, + (B)

          Ri  = F.I.D. Response (total CIO to C14 Peaks)
          Ci  = Concentration mg/L (total  of CIO to C14
                 standards)
          M    = Slope of line
          B    = Intercept of line

       (2) Calculate the TCO value for the sample  (Cu, measured  value)  and
          blank  (Cg,blank value) by  summing the F.I.D.  response over  the
          TCO retention time span and calculating the concentration from
          the calibration equation.

          It is  important that the  observed values  of the  total  integrated
          area for samples  be corrected by subtracting  an  appropriate
          solvent blank  prepared in  the same  manner  as  the samples.   The
          sample  is corrected for the blank:

             Cy  corrected = C(j measured - CB


 1.15   Data Reporting

       The results of  each TCO analysis should be  reported  as  one  number (in
       milligrams), corresponding to the quantity  of material  in the 1UU°C
       to 300° boiling  range in  the  original sample  collected.   If more
       information  is  available  (e.g., cubic meters  of  gas  sampled), the
       mg/sample  value  can  then  be  easily converted  to  the  required report-
       ing units.

                                   A-18

-------
                                                   Document No:  AEERL/13
                                                   Status:  INTERIM
                                                   Revision No:  3
                                                   Date:  9/25/86
                                                   Page:  10 of 11
1.16  Corrective Action
     Corrective action procedures in this ROP are covered in the QC check

     (2.1) and QC control (2.2) sections of the document.
1.17  Precision

     Duplicate results by the same operator will be rejected if they differ

     by more than 15%.

1.18  Accuracy

     The result of a quality control sample, run daily, will be considered

     deficient if it differs by more than 15% from the preparation value.

     If this value falls outside the accepted range, the system must be
     evaluated for the probable cause and a second standard run or a new

     calibration performed over the range of interest.

2.0  QUALITY ASSURANCE/QUALITY CONTROL

2.1  QC Checks

     0  All glassware used in the TCO analysis  shall be cleaned by the
        method described in reference 1.

     0  Change the GC inlet septum daily; follow this with  a column bakeout
        at 300°C for twenty minutes or until the F.I.D. response is stable
        and all evidence of contamination is gone (no peaks) or run an
        injection of clean solvent to verify column contamination  is
        eliminated.  Repeat this procedure during the run if evidence of
        septum failure appears (e.g., increasing peak elution  time with
        each run or major loss of sensitivity).

2.2  QC Controls

     0  Run a reagent sample for each new batch of  reagent  or  lot  of
        solvent used.  If the analysis fails to show organic contaminants
        to be below detection  limits under  identical  instrument operating
        conditions as used for samples,  then the reagent  shall  be  distilled
        in glass and retested or the reagent batch  will  be  unacceptable  for
        TCO analyses.
                                  A-19

-------
                                                   Document No:  AEERL/13
                                                   Status:  INTERIM
                                                   Revision No:  3
                                                   Date:  9/25/86
                                                   Page:  11 of 11

     0   Calibrate the GC  with  standards  that generate a response/concen-
        tration curve.   The calibration  curve must be 1 and must have a
        correlation coefficient  greater  than 0.97 to b acceptable.

     0   Prepare a QC standard  that  is  approximatey mid-way in the linear
        working range.   Run this  QC  standard daily to verify the perfor-
        mance of the GC.   Determine  the  TCO value using the calibration
        curve and its value plotted  compared to the theoretical value.  If
        two runs of the QC stanqard  differ by more than 15% of the actual
        value, prepare  a  new QC  standard and repeat the test.   If the new
        sample fails the  test, determine if there is a loose column
        connection, septum, or altered split flow.  After correction, run a
        new QC standard.    If  the new  samole fails the test, recalibrate
        the instrument  ana/or  perform  a  column cnange if needed.

3.U  REFERENCES

     1.  Lentzen, D. E., D. E.  Wagoner, E.  D. Estes, and W. F.  Gutknecht.
        IERL-RTP Procedures Manual:  Level  1 Environmental Assessment
        (Second Edition).  EPA 600/7-78/201, NTIS No. PB293-795, pp.
        140-142, October  1978.
                                  A-20

-------
              INTERIM PROCEDURE

  This procedure rtas been authorized for reference as an
  interim procedure by the AEERLQAO. The procedure wi>l
  undergo extensive EPA review prior to finaiization.
Document  No:   AEERL/22
Status:   INTERIM
Revision  No:   0
Date:   Septemoer 17, 1986
Page 1  of 8
STANDARD OPERATING PROCEDURE FOR  EXTRACTION OF FILTERS AND  SOLIDS
                                 by

                          Robert F. Martz*
                           Monica Nees**
                            Prepared  for

                 The AEERL TECHNICAL  SUPPORT OFFICE
  Disclaimer:   This standard  operating procedure has  been  prepared
  for  tne  sole use of the Air and Energy Engineering  Research
  Laboratory,  J. S. Environmental  Protection Agency,  Research Triangle
  Park,  North  Carolina,  and may  not be specifically  applicable to the
  activities  of other organizations.
  *  Chemist
     Acurex Corporation
     Research Triangle Park,  NC

  ** Research Environmental  Scientist
     Research Triangle Institute
     Research Triangle Park,  NC
                               A-21

-------
                                                  Document No:  AEERL/22
                                                  Status:  INTERIM
                                                  Revision No:  0
                                                  Date:   September  17,  1986
                                                  Page 2 of 8

     STANDARD  OPERATING  PROCEDURE  FOR  EXTRACTION OF FILTERS AND SOLIDS

1.0  PROCEDURAL ELEMENTS

i.l  SCOPE AND APPLICATION
     Gaseous emissions sampling devices use a wide variety of  filter  papers
     and solids as suostrates to entrap particulates.  Filter  papers  are
     frequently made of  Teflon, glass, or quartz; solids may  be standard
     reference materials, sands, ousts, ashes, etc.  Organic  material
     adsorbed on particulates collected by these filters and  solids is
     efficiently extracted before concentration and subsequent analysis.
     Methylene chloride, because of its good extraction  properties  and  high
     volatility, is the solvent of choice.  The extraction is  performed in
     an appropriately sized Soxhlet extractor.  This standard operating
     procedure (SOP) may be used if the filter or  solid  substrate  will  fit
     into a Soxhlet extraction  thimble and if the  organic  compounds adsorbed
     on the particulates are soluble  in methylene  chloride.

 1.2  SUMMARY  OF METHOD
     Organic  material that  is  adsorbed on  particulates  entrapped  on filters
     and  solids used  in  yaseous emissions  sampling is extracted with
     methylene chloride  in  a Soxhlet  extractor.   The  extract  is then
     concentrated  to  an  appropriate volume for  subsequent  organic
     analysis.

 1.3  DEFINITIONS
     o  Method Blanic:   Provides a  check  on contamination resulting from
                        laboratory  sample  preparation activities.   Typically
                        run  in  the  laboratory after receipt of  samples  from
                        the  field.   Addresses all chemicals,  reagents,  and
                        apparatus  used in  a method.
                                     A-22

-------
                                                  Document No:   AEERL/22
                                                  Status:  INTERIM
                                                  Revision No:   0
                                                  Date:   September  17,  1986
                                                  Page  3  of  8
1.4  INTERFERENCES
     Possible contamination from an unused "clean"  filter or  solid,  solvent,
     or glassware can be determined by running a method blank  (see sections
     1.3 and 2.U).

1.5  PERSONNEL REQUIREMENTS
     This SOP is written for individuals with at least a year  of organic
     chemistry and preferably also a year of experience in an  organic
     research laboratory.

l.b  FACILITIES REQUIREMENTS
     This procedure  requires a standard wet organic chemistry  laboratory
     with balances,  a fume hood, electricity, water,  refrigerator or freezer
     for sample storage, and a 110°C drying oven.   The beakers,  flasks, ring
     stands, clamps, tubing, etc. customarily found in such a  laboratory are
     also needed and are assumed to be readily available.

1.7  SAFETY REQUIREMENTS
     Routine safety  precautions needed in any organic laboratory are
     applicable here.  These include such measures  as no  smoking in  the
     laboratory; wearing safety glasses,  lab coats, and  rubber gloves;
     handling organic solvents in a fume  hood, etc.  A safety shower,  eye
     wash,  first aid kit,  and fire extinguisher  must be  immediately
     available in the laboratory.

1.8  APPARATUS
     NOTE:  Size of  apparatus depends  on  size  of filter  or quantity of
            solid being extracted.
     1.  Soxhlet Extractor Assembly:   Flask  with appropriately sized
         extraction  tube,  tnimole, and condenser.
                                      A-23

-------
                                                  Document No:   AEERL/22
                                                  Status:  INTERIM
                                                  Revision No:.  0
                                                  Date:   September 17,  1986
                                                  Page 4 of 8


    2.  Heating Mantle:   Sized  for  Soxhlet  flask.

    3.  Rheostat.

    4.  Boiling Chips,  Teflon.

    '5.  Snvder Column,  3-Ball.

    b.  Concentrator  Tubes,  Kuderna-Oamsh.

    7.  Frit,  Sintered  Glass.


1.9  REAGENTS  AND  MATERIALS

    1.  Methylene Chloride:   Burdick  and Jackson or equivalent  grade.

    2.  Water,  Deionized.

     3.   Glass Wool.   Clean by sequential immersion in three portions of
         methylene chloride.   Dry in a 1UU°C oven.  Store in a methylene
         chloride-rinsed glass beaker covered with aluminum foil.

    4.   A1 uminum Foi1.

     5.   Pasteur Pipette, Glass, Disposable.

    •6.   Flask, Volumetric:  1U or 25 ml_.

     7.   Storage Vials, Brown or Clear, with Teflon-lined Screw Caps.


1.10  SAMPLES/SAMPLING PROCEDURES

      NOTE:   All glassware coming in contact with  a sample  shall  be
             cleaned  by Level 1 procedures.^  Briefly,  this  entails
             sequential cleaning with soapy water,  deionized water,
             50:bO (V/V)  nitric acid/sulfuric acid, deionized water,
             methyl alcohol, and methylene chloride,  followed by
             oven  drying.

I.10.1  Preparation

     Samples  for extraction  arrive  as  particulates adsorbed on  filters  or on

     solids.  The  substrate  filters  and  solids  must have  been weighed  prior

     to use in the field  if  the weight  of  particulates  is to be determined
                                     A-24

-------
                                                  Document No:   AEERL/22
                                                  Status:  INTERIM
                                                  Revision No:   0
                                                  Date:   Septemoer 17, 1986
                                                  Page 5 of 8


     in this SOP.   (This  is  not  always  the  case.)   Remove the sample from

     its container  and  prepare  it  for  insertion  into  a Soxhlet  extraction

     thimble,  using either  (1)  or  (2)  as  described below:

     (1)   Fold  filters  coated with  participates  into  a cone with the point
          down  and  particulates  facing  inward,  then place on tared aluminum
          foil.   Fold  the aluminum  foil  over  the folded  filter  to prevent
          loss  of  particulates.  'Weigh.   Record  tare  and final  weights.

                                     OR

     '2)   "Transfer  solids with  entrapped  particulates to an appropriately
          sized  tared  container.  Weign.   Record tare and final  weights.

1.10.2  Extraction

     1.   Perform this  extraction using an appropriately sized Soxhlet
         extractor  assembly.  Solid samples of 30 grams  or less and single
         filters of Teflon,  glass,  or quartz  up to 8" X 10" can be extracted
         in  a  500 ml apparatus.   Solid samples weighing between 30 and 200
         grams  and  multiple  filters require a 3-liter (Size G)  apparatus.

     2.   Use an  all-glass extraction thimble  with a coarse frit recessed
         b-15  mm aoove a crenulated ring at the thimble bottom to facilitate
         drainage.

     3.   Cover the  frit with a plug of cleaned glass wool to prevent
         particulates  from clogging the pores.

     4.   Load  the thimble with the sample prepared as described  in  section
         1.10.1.

     b.   Place a plug of cleaned glass wool on top of the  sample  to  prevent
         particulates  from floating on top of the methylene  chloride  solvent
         used  for extraction.

     6.   Assemble the Soxhlet extractor apparatus.   Fill  the round-bottomed
         flask two-thirds full  with methylene chloride.   Place  the  flask on
         a heating mantle with temperature controlled by  a  rheostat.   Place
         thimble containing the sample into the extractor  tube  and  attach
         tube  to flask.  Attach condenser to top  of  extractor  tube.   Start
         the flow of cooling water through condenser jacket.
                                  A-25

-------
                                                 Document No:   AEERL/22
                                                 Status:  INTERIM
                                                 Revision No:   0
                                                 Date:  Septe'mber  17,  1986
                                                 Page 6 of  8


    7.   Turn  on  tne  Soxhlet  and  extract with methylene chloride for  24
        hours.

    8.   Turn  off the Soxhlet.   Remove the condenser.  Depending on the size
        of  the  apparatus,  rinse  the extraction tuPe and thimole with 10 to
        50  ml of methylene chloride.  Collect and combine all  rinsings in
        the Soxhlet  flask.

    9.   In  a  clean fume hood,  place the flask containing the  methylene
        chlonae extract and rinsings on a heating mantle with  temperature
        controlled by a rheostat.  Add Teflon boiling chips to  the flask,
        if  necessary, to prevent bumping.

    10. Attach a 3-ball Snyder column pre-wetted with methylene chloride to
        the flask.  To prevent any  foreign material  from entering  the
        flask,  fit the top of the column with a  ground  glass  adapter
        attached to a bent glass tube.  Direct the open end of  the tube
        towards the  rear  of the hood.

    11. Concentrate  the extract to  the appropriate volume by  maintaining a
        temperature  just  high enough to boil  off the  methylene  chloride.

    12. Use methylene  chloride to quantitatively transfer the concentrate
        to Kuderna-Oanish tubes for further  concentration,  if necessary.
        Attach  the  same 3-ball Snyder  column used  in  step  1U to the
        Kuderna-Oanish  tube and concentrate  to the appropriate  volume.
        Remove  any  contaminating particulates  by filtering  the concentrate
        through a sintered  glass frit  into  a small  flask.

     13. Rinse the Snyder  column with  small  portions  of  methylene chloride.
        Collect and  combine all  rinsings.   Combine  rinsings with the
        concentrate  from  step 12.

     14. using a Pasteur pipette, transfer  the sample quantitatively to a
        volumetric  flask  and  dilute to  the  mark  with methylene chloride.
        (A  10 ml  or  25 ml flask  is  the  size used most frequently.)

1.10.3   Storage

     Store  the  sample in either  a Teflon-taped volumetric flask or a brown
     or clear  vial with a  Teflon-lined screw cap.  Place in a refrigerator
     or freezer.
1.11   CALIBRATION/STANDARDIZATION

      Calibration and standardization are not applicable to  this  SOP  which
      covers  only extraction and not analytical procedures.
                                    A-26

-------
                                                  Document No:   AEERL/22
                                                  Status:   INTERIM
                                                  Revision No:   0
                                                  Date:   September 17,  1986
                                                  Page 7 of 8
1.12  ANALYSIS PROCEDURES
      This extraction SOP does not stand alone, but is used in conjunction
      with numerous other SOPs which describe analysis procedures.   Consult
      an appropriate analytical SOP (e.g., AEERL/12,  Standard Operating
      Procedure for the Gravimetric Analysis of Organic Extracts)  for
      analytical details.

1.13  CALCULATIONS
               Let P  = particulates(mg\
                   Wg = (particulates + substrate + tare)/mg\
                   Wi = substrate(mg)

                   T  = tare(mg)

               Then P = Wg - Wi - T

1.14  DATA REPORTING
      The results are reported in units of mg particulates/sample.

1.15  CORRECTIVE ACTION
      This SOP does not stand alone, but is used as a forerunner to numerous
      analytical SOPs.  Consult the appropriate analytical SOP for
      corrective action procedures.

1.16  METHOD PRECISION/ACCURACY
      This SOP does not stand alone, but is used as a forerunner to numerous
      analytical SOPs.  Consult the appropriate analytical SOP for precision
      and accuracy requirements.
                                   A-27

-------
                                                  Document No:   AEERL/22
                                                  Status:  INTERIM
                                                  Revision No:   U
                                                  Date:  Septemoer  17,  1986
                                                  Page b of 8
2.U  QUALITY CONTROL ELEMENTS

     1.  Determine a method blank by performing an extraction using  the  same
         size of Soxnlet apparatus and the same amount of glass wool,
         methylene chlonae, and, if possible, unused  "clean" filter or
         solid as employed in the extraction of the field sample.  This
         method blank provides a check on contamination  resulting  from all
         sample preparation activities in the laboratory.

     2.  Perform a method blank along with each set of samples  run.   Any
         method blank value will eventually be subtracted from  the sample
         value found in subsequent organic analysis.


2.U  REFERENCES

     1.  Lentzen, D.E., D.E. Wagoner, E.D. Estes, and W.F. Gutknecht,
         "IERL-RTP Procedures Manual:  Level 1 Environmental Assessment
         (Second Edition), EPA 600/7-78-201, NTIS No. PB-293-795,  pp. 26,
         136-142, October 1978.
                                   A-28

-------
                                                       Document No:  AEERL/40
                                                       Status:  Draft
                                                       Revision No:  1
                                                       Date:  June 1987
                                                       Page:  1 of 10
                       RECCfMBNDED OPERATING PROCEDURE
                   FOR  LARGE-SCALE XAD-2 RESIN PURIFICATION
                                     by

                             David F. Natschke*
                                Monica Nees**
                                 Prepared  for

                         THE AEERL AIR TOXICS  BRANCH
          Disclaimer:   This recommended operating procedure has
          been prepared for the sole use of the Air Toxics Branch
          of  the Air and Energy Engineering Research Laboratory,
          U.S.  Environmental Protection Agency, Research Triangle
          Park,  NC,  and may not be specifically applicable to the
          activities of other organizations.
  Chemist
  Acurex Corporation
  Research Triangle Park,  NC
**Research Environmental Scientist
  Research Triangle Institute
  Research Triangle Park, NC
                                   .

-------
                                                        Document No:   AEERL/40
                                                        Status:  Draft
                                                        Revision No:   1
                                                        Date:   June 1987
                                                        Page:   2 of 10
                        STANDARD OPERATING  PROCEDURE FOR
                      LARGE-SCALE  XAD-2 RESIN  PURIFICATION
1.0  PROCEDURAL ELEMENTS

1.1  Scope and Application

     This recommended operating  procedure  (HOP)  has been developed as an

alternative to the small-scale XAD-2  resin purification procedure described in

AEERL/25.  It describes the purification of large amounts of XAD-2 resin for

subsequent use in gaseous emission sampling.   Commercial XAD-2 resin is

impregnated with a bicarbonate solution to inhibit microbial growth during

storage.  The bicarbonate solution, any residual extractable monomer or

polymer, and other residual organic material are removed by a series of

aqueous and organic extractions.  This ROP differs form AEERL/25 in that a

chromatographic elution rather than a Soxhlet extraction is performed.

     This ROP can also be employed to recycle resin used in field sampling,

provided the resin has not been permanently discolored.  Experience has shown

that badly discolored resin cannot be purified well enough to pass the quality

control checks described later in Section 2.1.  Purification of recycled XAD-2

resin can begin at Step 4 of Section 1.8.   The prior aqueous washings have

been shown to be unnecessary to recycle the resin.  This procedure should be

used on an "as needed" basis.  The purified resin  should not be stored «ore

than three weeks before use.

     This procedure may not produce material suitable  for ultra-trace  level

sampling and analysis since the allowable contaminated level  is  1.75 mg/175

gram cartridge.

                                     A-30

-------
                                                        Document No:  AKERL/40
                                                        Status:  Draft
                                                        Revision No:  1
                                                        Date:- June 1987
                                                        Page:  3 of 10

1.2  Summary of Method

     This method is a chromatographable elution.  Bicarbonate is first reaoved

by soaking the resin in water.  The wet resin is slurry transferred to an

extraction tube and organic contaminants are eluted by sequentially puBping

•ethanol, nethylene chloride, methanol, and methylene chloride through the

resin bed. The resin is ready for use in sampling after it has been dried

under nitrogen and passed the quality control tests.

1.3  Personnel Requirements

     This procedure requires one chemist or technician trained on this ROP.

1.4  Facilities

     This procedure requires a laboratory set up for organic sample analysis.

This laboratory should include a fume hood, a source of de-ionized water,

solvent storage, glassware, and cleaning facilities.  Because flammable

solvents are used, the laboratory should be free of sources of flames or

sparks when this purification procedure is performed.

1.5  Safety Precautions

     This procedure uses flammable and halogenated organic solvents.  There

are known hazards of fire and of poisoning due  to  ingest ion.   There may  also

be hazards due to long-term exposure to methylene chloride fumes.   There are

no known hazards due to contact with XAD-2 resin.   This procedure  should be

performed in a well-ventilated, no-smoking area.   Sources of sparks or  flame*

should be reaoved from the area.   Personnel  protection should include safety

glasses, lab coats, and disposable gloves.   Atmospheric monitoring for

aethylene chloride should alao be  considered.

1.6  Apparatus

     (1)  Extraction apparatus;  see  Figure  1

     (2)  Garbage pails, plastic.  25-gallon  capacity
                                     A-31

-------
 Drain valve
                                              Document No:   AKKRL/40
                                              Status:  Draft
                                              Revision No:   1
                                              Date:. June  1987
                                              Pa«e:  4 of  10
                                     To pump
                      I
           Figure 1.  Cleanup Apparatus

(Dimensions withheld pending patent or publication)

                       A-32

-------
                                                        Document No:  AEERL/40
                                                        Status:  Draft
                                                        Revision No:  1
                                                        Dater  June 1987
                                                        Page:  5 of 10

     (3)   Autotransfomer.  variable:   "Variac"

     f4)   Heat  exchanger coil:   see  Figure  2

     (5)   Pump,  variable speed:  Teflon and  stainless steel construction,

          capable of 0-3L/hr flow rate

     (6)   Dessicator. with rubber gasket

     (7)   Analytical balance:   0.01  ng resolution

     (8)   Gas chromatograph:  with flame  ionization detector

     (9)   Snvder Column. 3-ball

     (10) Flask. Round-bottomed. 500 mL

     (11) Flask. Round-bottomed. 500 mL

     (12) Flask. Volumetric. 10 mL

1.7  Reagents and Materials

     (1)   Amber lite XAD-2 Resin:  a» supplied by Roh» fc Baas, Co.,

          Philadelphia, Pennsylvania; 7.5 kg

     (2)   Water. Deionized

     (3)   Methaaol:  Burdick and Jackson or equivalent grade

     (4)   Methylene Chloride:  Burdick and Jackson  or equivalent grade

     (5)   Nitrogen. Liquified:  low pressure  tank,  National Welders, Airco, or

          equivalent grade

     (6)   Storage Bottles.  Solvent;  brown, gallon-sized,  with Teflon-lined

          screw cap

     (7)   Toluene;  Burdick and Jackson or equivalent  grade

     (8)  Boiling Chips. Teflon;  solvent  rinsed

     (9)   Teflon Tape

     (10) Disposable Aluminum Weighing Pans;   approximately 2" in diameter,

          1/2"  deep; crimped sides;  weighing approximately 1.0 grams
                                    A-33

-------
Coils formed  from  1/4-in.  00 copper tubing
  Supelco
€
              Flow
 From liquid outlet

 of low pressure  liquid

 nitrogen tank
  Carrier gas drying tube
                                                                                 To inlet of  XAD

                                                                                 cleaning apparatus
9
t«.

2
                                                       l-in. wide heat  tape  wrapped

                                                       around the entire  coil
                                                                             Variac
                                                                                                    « r* < »

                                                                                                     ?. ?. 5 £
                                                                                                         M. K
                                                                                                         o ••
                                                                                                     en <_i a   »*

                                                                                                     o § z o z
                                                                                                     •-S <» O H O
                                                                                                         .. ft ..
                                                                                                     !-•>—•  M>
                                                                                                     O U>  c*
                                                                                                       00 <-•
                                  Figure 2.  Heat Exchanger

-------
                                                        Document No:  AEERL/40
                                                        Status:  Draft
                                                        Revision No:  1
                                                        Date:  June 1987
                                                        Page:  7 of 10

1.8  Extraction Procedure

     i'l)  Pour 7.5 kg of resin into a 24-gallon plastic garbage pail;  cover

          pail.  Add sufficient de-ionized water so that the entire resin bed

          floats.  Allow it to soak for at least 7 days before proceeding.

     (2)  Transfer the resin-water mixture to the extraction apparatus by

          pouring it in the top.  Drain the aqueous waste through the bottom

          valve  into any suitable sized container.  The aqueous waste is

          known  to be non-hazardous and may be disposed of by pouring down a

          sink drain.

     (3)  Pour deionized water in the top of the extractor with the bottom

          drain  valve open.  Continue until the eluent is clear.

     (4)  Pour 4 gallons of nethaool in the top of the extractor with the

          bottom drain open.  Close the valve.  The excess water will have

          been removed. Fill the resin bed with methanol.  Replace  the top

          cap.   Allow  it to soak overnight before proceeding.

          NOTE:  Redistilled, used Methanol may be used  in this extraction

                 step.

     (5)  Pump 5 gallons of methanol through  the extractor over a period of

          1.5 hour.  Stop  the pump.  Close  the inlet  valve.   Drain the bed

          through the bottom valve.  Close  the bottom valve.

     (6)  Change receiver  vessels.  Open the inlet valve.  Pump 5 gallons of

          methylene  chloride  through  the extractor  over a period of

          1.5  hours.   Stop the  pump.   Close the inlet valve.  Drain the bed

          through  the  bottom valve.   Close the bottom valve.

          NOTE:   Redistilled,  used methylene chloride may be used  in this

                  extraction step.   A"35

-------
                                                   Document No:  AEERL/40
                                                   Status:   Draft
                                                   Revision No:  1
                                                   Date:   June 1987
                                                   Page:   8 of 10

(7)   Change receiver vessels.   Open the inlet valve.  Pump 5 gallons of

     new or redistilled methanol  through the extractor over a period of

     1.5 hours.   Stop the pump.  Close the inlet valve.  Drain the bed

     through the bottom valve.   Close the bottom valve.

(8)  Change receivers.  Open the inlet valve.  Pump 5 gallons of new or

     redistilled methylene chloride  through the extractor over a period

     of  1.5 hours.  Collect the final 2 liters as 2 1-liter aliquots for

     the preparation of quality control (QC) samples.  Stop the pump.

     close the  inlet valve.  Drain the bed through  the bottom valve.

 (9)  Connect  the heat  exchanger to the liquid outlet of the liquid

     nitrogen tank.  Connect the outlet of the liquid nitrogen tank.

     connect  the Outlet  of  the heat exchanger to the bottom valve of the

     extractor.   Connect a  Variac  to  the  heat exchanger.

     NOTE:  Warming the extractor  with an extra heat  tape  may  speed up

             the drying process.  The  heated area  should not  be more than

             slightly warm to  the touch.

                               WARNING
           Do not exhaust the fumes directly into the room.


 (10) Turn on the Variac.  Open the liquid nitrogen valve to a low flow.

      The Nz flow should be the maximum flow that does not entrain resin.

      Adjust the Variac so that the output of the heat exchanger  is

      gaseous nitrogen at a temperature somewhat above ambient (30-40°C is

      satisfactory).  Continue until  the  resin  is dry.  This should  take

      around 48 hours.           _

-------
                                                         Document  No:   ABKRL/40
                                                         Status:   Draft
                                                         Revision  No:   1
                                                         Date:  June  1987
                                                         Page:" 9  of  10

     (11) Transfer the dried resin to brown glass solvent  bottles, cleaned

          according to Level 1 procedures.  Wrap Teflon  tape  around  the cap.

     (12) Store in an area free of organic materials.

2.0  QUALITY CONTROL ELEMENTS

2.1  Quality Control (QC) Checks

     (1)  Transfer the two 1-L aliquots of methylene chloride reserved in

          Step 7 of Section 1.8 to 2-L round-bottomed flasks.  Add Teflon

          boiling chips.  Add a pre-wetted Snyder column and adapter to each

          flask.  In a hood, concentrate  these QC samples  to  less than 100 ml.

          Transfer the concentrates, Snyder columns, and adapters to 500 ML

          round-bottomed flasks.  Continue concentrating the QC samples to

          less than 5 ml.  Cool.  Transfer the concentrates to 10 mL

          volumetric flasks.  Dilute to volume with fresh  methylene chloride.

     (2)  Perform duplicate GRAV analysis using procedure  AEER1/12 on each OC

          sample using 1 ml aliquots.  Refer to ABERL/12,  Standard Operating

          Procedure for Gravimetric Analysis of Organic Extracts, for details.

          Calculate the GRAV in units of  mg GRAV material/sampling cartridge,

          where the 1-L methylene chloride AC sample is  assumed to be

          equivalent to 8 sampling cartridges.  The pass/fail value  is

          5 mg/cartridge.

     (3)  Perform duplicate TOO analysis  on each QC sample.   Refer to

          AEERL/13, Standard Operating Procedure  for Total Chromatographable

          Organic*, for details.  Calculate the TOO in units  of mg TOO

          material/sampling cartridge, where  the  1-L methylene  chloride QC

          sample is assumed to be equivalent  to 8 sampling cartridges.  The

          pass/fail value  is  1.75 mg/cartridge.
                                   A-37

-------
                                                       Document  No:   AEERL/40
                                                       Status:   Draft
                                                       Revision  No:   1
                                                       Date:   June 1987
                                                      Page:   10  of 10

    (4)  Perform a residual methylene chloride (RMC)  test.   Transfer l.+O.l g

         of dried, cleaned resin to a vial with a screw  cap  containing  a

         Teflon-lined septum.  Add 3.0 ml of toluene.  Cap the vail  and shake

         well.  Analyze for residual methylene chloride  by gas

         chromatography. Calculate the RMC as micrograms RMC/gram of resin.

         The  pass/fail value  is  1000 ug/g.

     (5)  The  resin must pass  all three tests before  it may be used for

         sampling purposes,   if  it fails only the RMC test,  redry the resin

          as described in  Section 1.8 and retest for  RMC  as described in the

          previous step.

3.0  REFERENCES

     1.  Lentzen,  D.E.,  D.E.  Wagoner, E.D. Estes,  and W.F. Gutknecht.

         IERL-RTP Procedures Manual:   Level  1  Environmental Assessment  (Second

         Editon),  EPA 600/7-78-201,  NTIS  No. PB-293-795,  pp.  26-32, 136-142

         and Appendix B,  October 1978.

     2.  Hammersaa,  J.W.,  D.G. Ackerman,  M.M.  Yamada, C.A. Zee, C.Y.  Ung, K.T.

         McGregor, J.F. Clausen,  M.L.  Draft,  J.S.  Shapiro, and E.L. Moon.

         emission Assessment of Conventional Stationary Combustion Systems:

         Methods and Procedures Manual for Sampling  and  Analysis.  EPA

         600/7-74-0244, Appendix EO, January 1979.
                                     A-38

-------
                                                       Document No. AEERL/41
                                                       Status:       Draft
                                                       Revision No.   0
                                                       Date:        June 1987
                                                       Page:        Iof9
            'RECOMMENDED OPERATING PROCEDURE FOR SAMPLE RECOVERY FROM
                      XAD-2 RESIN BY PUMP THROUGH BLUTION"
                                       by

                                 Robert F. Martz*
                               Theodore X. Koinis*
                                David F. Natschke*
                                  Prepared for

                           THE AEERL AIR TOXICS BRANCH
              Disclaimer:   This Recommended Operating Procedure has
              been prepared for the sole use of the Air and Energy
              Engineering Research Laboratory, U. S. Environmental
              Protection Agency, Research Triangle Park, North
              Carolina, and may not be specifically applicable to
              the activities of other organizations.
*Chemist
Acurex Corporation
Research  Triangle Park, N.C.
                                           A-39

-------
                                                     Document  No:   AEERL/41
                                                     Status:         Draft
                                                     Revision  No:   0
                                                     Date:           June 1987
                                                     Page:           2  of 9

RECOMMENDED OPERATING PROCEDURE  FOR SAMPLE  RECOVERY FROM IAD-2  RESIN BY
                             PUMP-THROUGH ELUTION

1.   PROCEDURAL ELEMENTS

1.1  Scope and Application

     This recommended operating  procedure (ROP)  has been developed as an

alternative procedure to AEERL/22  for  the recovery of semi-volatile organic

samples collected on Amberlite XAD-2 resin.   It  has been shown  to be

applicable to the recovery  of both ambient  and source samples collected on

XAD-2 resin.

     This procedure has not been shown to be applicable to the recovery of

samples collected on any other sorbent.   It is applicable to the recovery of

compounds soluble in methylene chloride or  methanol.  Caution must be used in

the interpretation of analytical results where methanol was used in the workup

since  it  is known tn r»*ct  with  certain classes of compounds.  Extraction

solvent volumes  and  flow rates were developed for sampling cartridges used for

 IACP studies.  Other cartridge designs may require different conditions.

 1.2  Summary  of  Method

      This  procedure  is  basically a chromatographic technique.  The sample is

 eluted from the  resin by pumping solvent through  the sample cartridge  against

 the force of  gravity.   Depending upon purpose, one or more solvents may be

 used sequentially.   The eluent  is collected  in one  or more round bottom flasks

 and concentrated by solvent  distillation using a  Snyder column.   Multiple

 sample cartridges may be manifolded for efficiency.

 1.3  Personnel Requirements

      This procedure requires one  chemist or  trained technician comfortable

 with solvent handling techniques.   In addition,  the person must have refined

 mechanical skills for fittings  and  glassware manipulation.
                                   A-40

-------
                                                      Document No:    AEERL/41
                                                      Status:         Draft
                                                      Revision No:    0
                                                      Date:           June 1987
                                                      Page:           3 of 9

1.4  Facilities

     This procedure requires one standard laboratory set up for organic sample

analysis.  This laboratory should include a fume hood, solvent storage, and

sample storage.  A source of glassware cleaned by AEERI Level 1 procedures  is

required.  If flammable solvents are used, the laboratory should be free of

sources of flames or sparks.

1.5  Safety Precautions

     This procedure uses halogenated organic and/or flammable solvents.  There

are known hazards of poisoning due to inhalation, dermal exposure or ingestion

and fire.  There may be hazards due to longterm exposure to fumes from

methylene chloride.  The concentration step must be done in a  fume hood or

using  some other form of vapor extraction.  There are no known hazards due to

contact  with XAD-2  resin.

     This procedure should be performed  in a well-ventilated,  no smoking area.

Sources  of sparks  or flames should be removed from  the  area.   Personal

protection should  include safety glasses, gloves,  lab coats,  and organic vapor

mask.   Disposable gloves should be worn  during the manipulation of

 concentrated sample extracts unless  all  components of the sample are known to

 be non-hazardous.

 1.6  Apparatus

      (1)  Extraction  apparatus;  see Figure 1

       (2)  Flask. Round Bottom. 2.000 mL with 24/40 ground glass joint

       (3)  Flask. Round Bottom, SQQ »T. with 24/40 ground glass joint

       (4)  Flask. Volwetric. 10 mL

       (5)  Autotransformer, variable;  "Variac"

       (6)  3 ball Snyder Column with  24/40 ground glass joint

       (7)  Heating  Mantle,  quartz;  for  2000 ml RB
                                    A-41

-------
                                           Document No:
                                           Status:
                                           Revision No:
                                           Date:
                                           Page:
AEERL/41
Draft
0
June  1987
4 of  9
1,  SAMPLE  CARTRIDGE
2,  RECEIVER-ROWD BOTTCM FLASK
3,  PIMP
4,  SOLVENT tesEfwoiR
5,  VEKTHHTO-ATMOSPHERE ON/OFF VALVE.
6,  FLOW RATE CONTROLLER VALVE
7,  DRAIN VALVE
8.  DRAIN tesERvoiR
9,  ATMOSPHERE VENT VALVE
                  FIGURE 1.  DIAGRAM OF EXTRACTION
                               A-42

-------
                                                    Document No:   .AEERL/41
                                                    Status:          Draft
                                                    Revision No:    0
                                                    Date:            June 1987
                                                    Page:            5 of 9

     (8)   Heating Mantle, quartz;  for 500 ml EB

     (9)   Support Ring: for round bottoa flasks

     (10)  Pump,  variable speed:  Teflon and stainless steel construction

     (11)  Squeeze bottle. Teflon

     (12)  Adapter;  with 24/40 ground glass joint

     (13)  Syringe,  glass, with luer lock fitting. 10 ml

     (14)  Glass wool

     (15)  Aluminum  foil

1.7  Reagents and Supplies

     (1)   Methanol;   distilled in glass or equivalent, dependent upon expected

          sample BBSS levels

     (2)   Methylene Chloride;  distilled  in  glass or equivalent, depending on

          expected sample mass levels

     (3)   Pipet. Pasteur, disposable

     (4)   Bulb Pipet. ImL

     (5)   Boiling chips. Teflon; solvent  rinsed

     (6)  Sample vials.  4  dram,  glass:   with screw cap and Teflon coated

          septua

     (7)  Filter units.  .45 micron,  disposable;  to fit luer lock syringe,

          Supelco 45-8072 or equivalent

 1.8  Extraction Procedure

     Note:   This procedure is written based upon sequential  elutioo with MeCla

             and MeOH of several manifolded samples in cartridges.   This

             procedure may be used with a single solvent,  dependent upon

             project  requirements, provided  that proper  recovery of the desired

             sample components can be  independently proven.
                                   A-43

-------
                                               Document No:    AEERL/41
                                               Status:         Draft
                                               Revision No:    0
                                               Date:           J»™ *987
                                               Page:           6 of 9

(1)    Connect  the sample  cartridge to the manifold with the sample inlet

      up,  if known  (Refer to Figure 1).   Place the manifold inlet  into a

      reservoir  of  nethylene chloride solvent (reservoir should contain

      800  mL of  sorbent/cartridge).  Place the outlet into  a 1  liter

      round bottom  flask  which is sitting on a support ring and labeled

      with the appropriate sample number.

(2)   Close the  drain valve.  Open valves 5 & 6 for all lines.   Turn on

      the punp.   Adjust  the regulating valve, 6,  for approximately even

      flows to all  cartridges.  Adjust the pump speed to yield  a flow

      rate of around 100-150 ml/minute through each branch  of the

      manifold.

(3)   Pump 500 mL,  as measured in the collection flask, of  methylene

      chloride through each cartridge.  Close valve 5 for each  cartridge

      as that amount is reached.  Adjust pump as needed to  maintain a

      flow of 100-150 ml/minute through each branch of the manifold.

      Because of the output vented to atmosphere feature of the valves,

      solvent in the resin  dead volume (**250aL) will drain into the

      round bottom flask.   However,  it may be necessary to temporarily

      invert  the cartridge  or to  disconnect  the cartridge  inlet to

      complete  the drainage.

 (4)   When  all the  cartridges  on  the  manifold have  been pumped according

      to step  3, turn off the  pump.   Drain the manifold lines by opening

      the  drain  and the vent valve.   Remove the MeCla  reservoir.

      Note:  Project  requirements may call for a single solvent

            extraction.   In this case, proceed from step 4 directly to

            Section 1.9, Sample Concentration.
                               A-44

-------
                                                    Document No:   AEERL/41
                                                    Status:         Draft
                                                    Revision No:   0
                                                    Date:           June 1987
                                                    Page:           7  of 9
     (5)  Close the vent and drain valves.   Place  the Manifold inlet into a

          methanol reservoir.   Place  the outlet  into a  21 round bottom flask

          sitting on a support ring and labeled  with the appropriate sample

          number.

     (6)   Follow steps 2,  3, & 4  using nethanol  in place of MeCla.

          Note:   Project requirements may call for composited or combined

                 extracts.   If this is so, you may choose to elute the sample

                 composite into a single 2 liter round  bottom flask  (or

                 whatever size would be appropriate).

1.9  Sample Concentration

     (1)  Connect the large heating mantles  to variacs  in a fume hood.

     (2)  Drop some clean Teflon  boiling chips  into each flask.  Place*a round

          bottom flask containing extracted  sample on each heating mantle.

          Place a presetted Snyder column onto  each flask.  Add an  adapter to

          the top of each column.

    (3)   Turn on the variac.   Adjust the voltage  so as to reflux  the  solvent.

          Concentrate the extract to  less  than 100 mi..   Turn  off the variac.

          Note:   It is beneficial to  insulate the outside of  the column

                  with aluminum foil or glass wool.

     (4)  Transfer  the  concentrated extract to a 500 mi. (or other appropriate

          size)  round bottom  flask, using additional solvent to quantitatively

          perform the transfer.   Repeat the experimental setup of

          concentration steps 1  and 2 using small beating mantles and the

          500 ml flasks.

                                            CAUTION

           The final concentration of the extracts calls for the exercise of

           judgement.  It may not be  possible to keep the extract of a high
                                    A-45

-------
                                              Document No:    AEERL/41
                                              Status:          Draft
                                              Revision No:    0
                                              Date:            June I?87
                                              Page:            8 of 9

    solids sample, such as a source dilution sample,  ID solution if the

    total volume is reduced to 10  ml.   If visual or historical  evidence

    indicates that a given extract has  concentrated enough for  the

    analytical purpose of the sample, stop the process and dilute to a

    known volume in a volumetric flask.

f5)  Turn on the variacs.  Adjust the voltage so as to reflux the

    solvent.  Concentrate the sample to less than 10 ml.  Turn  off the

    variac.  Remove the adapter.

(6)   Rinse  the Snyder column with 1-2 mLs of the appropriate solvent into

     the round bottom flask.  A clean Teflon squeeze bottle of  solvent is

     adequate. Remove the column.

(7)   Get a  sample  vial,  a 10-mI volumetric flask or flask  of appropriate

     size,  a Pasteur pipet,  a  10 ml  luer  lock syringe,  a filter unit, and

     a Teflon squeeze bottle of  the  appropriate solvent.

(8)  Remove the  plunger from the syringe.  Attach  the filter unit to  the

     syringe. Using a  Pasteur pipet, transfer the concentrated extract

     to the syringe.  Replace the  plunger and filter  the extract into the

     volumetric  flask.

(9)  Remove the plunger and use  the same Pasteur pipet  to transfer flask

     washes to the syringe.  Again, replace the plunger and filter into

     the volumetric flask.   Hake up to  exact volume with fresh  solvent.

     Transfer to a sample vial.   Seal with septum and cap.  Wrap the cap

     joint with Teflon tape.  Mark the  vial with the sample code.  Store

      in a  refrigerator  or freezer.  Record the sample  code, date of

     extraction,  extraction solvent, and final volume.

      Note:   The extraction  of a sample collected  on  XAD-2 with methanol

             frequently  results  in  a cloudy extract due to resin
                             A-46

-------
                                                     Document No. AEERL/41
                                                     Status:        Draft
                                                     Revision No.    0
                                                     Date:         June 1987
                                                     Page:         9 of 9

                •icroparticulate.  This microparticulate is  innocuous as far

                as  the sample preparation is concerned,  though  it does need

                to  be removed before any analysis  is performed  on the

                extract.   This nicroparticulate should not be confused with a

                saturated sample.  The concentration step should  not be cut

                short simply because microparticulate is  present.

2.  QUALITY CONTROL  ELEMENTS

    It is assumed that the sample set includes the desired quality control

samples.  If the sample set is not known to include a laboratory blank (it may

be included as a blind sample,  for example) one should be  prepared as part of

the sample set.  No  special procedure blanks are run.  The blank value for

this procedure is included in the XAD lab blank.  The XAD  lab blank is

determined for each  batch of XAD.
                                          A-47

-------
                                 TECHNICAL REPORT DATA
                          (Please read Instructions on the reverse before completing)
 1. REPORT NO.
  EPA-600/R-93-213
                            2.
                                                       3. RECIPIENT'S ACCESSION"NO.
 4. TITLE AND SUBTITLE
 Emissions from Burning Cabinet Making Scraps
                                    5. REPORT DATE
                                     November 1993
                                                       6. PERFORMING ORGANIZATION CODE
 7. AUTHOR(S)
 Michael Tufts and David Natschke
                                                       8. PERFORMING ORGANIZATION REPORT NO.
 9. PERFORMING ORGANIZATION NAME AND ADDRESS
 Acurex Environmental Corporation
 P. O.  Box 13109
 Research Triangle Park, North Carolina 27709
                                                       10. PROGRAM ELEMENT NO.
                                                      91-004
 12. SPONSORING AGENCY NAME AND ADDRESS
 EPA, Office of Research and Development
 Air and Energy Engineering Research Laboratory
 Research Triangle Park, NC  27711
                                    13. TYPE OF REPORT AND PERIOD COVERED
                                    Task Final; 9/90 - 6/91
                                    14. SPONSORING AGENCY CODE
                                     EPA/600/13
 is. SUPPLEMENTARY NOTES AEERL project officer is Robert C. McCrillis, Mail Drop 61. 919/
 541-2733.
 16. ABSTRACT
          The report gives results df an initial determination of differences in emis-
 sions when burning ordinary cordwood compared to kitchen cabinet making scraps.
 The tests were performed in an instrumented woodstove testing laboratory on a
 stove that simulated units observed in use at a kitchen cabinet manufacturer's facil-
 ity. Three test burns were made,  using a stove made from a 55 gal.  (0.208 cu m)
 drum and a kit sold for that purpose. Test burn 1 used seasoned oak cordwood fuel,
 test burn 2 used particle board scraps,  and test burn 3 used Formica-faced particle
 board scraps. The scraps for  tests 2 and 3 were obtained from a kitchen cabinet
 manufacturer in Vermont. In general,  the cordwood produced higher emissions of
 carbon monoxide and total hydrocarbons,  while the composite woods produced higher
 emissions of the heavier molecular weight organic compounds. There were signifi-
 cant  differences in burnrate between the tests, which introduced some uncertainty
 in interpreting the analytical results.
 7.
                              KEY WORDS AND DOCUMENT ANALYSIS
                 DESCRIPTORS
                                           b.lDENTIFIERS/OPEN ENDED TERMS
                                                 c.  COSATI Field/Group
 Pollution
 Combustion
 Emission
 Scrap
 Wood Products
 Particle Boards
Formica
Pollution Control
Stationary Sources
Cabinet Making
13 B
21B
14G
11G
11L
11D, 111
 3. DISTRIBUTION STATEMENT
 Release to Public
                                           19. SECURITY CLASS (ThisReport)
                                           Unclassified
                                                                    21. NO. OF PAGES
                                                     105
                        2O. SECURITY CLASS (Thispage)
                        Unclassified
                                                 22. PRICE
EPA Form 2220-1 (9-73)
                                        A-48

-------