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
------- |