United States
                  Environmental Protection
                  Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park, NC 27711
                  Research and Development
EPA/600/S2-90/001  May 1990
v>EPA         Project  Summary
                   Effects  of  Appliance Type and
                   Operating Variables on
                   Woodstove  Emissions
                   P. G. Burnet, J. E. Houck, and R. B. Roholt
                    This project was conducted in
                  support of the Integrated Air Cancer
                  Project (IACP) to provide data on the
                  specific effects of appliance type and
                  operating variables on woodstove
                  emissions. Samples of particulate
                  material  and  volatile organic
                  compounds (VOCs) were  collected.
                  Particulate  samples were analyzed
                  for  different organic  fractions,
                  including  specific  polynuclear
                  aromatic  hydrocarbon  (PAH)
                  compounds. Inorganic matter on the
                  filters was also  analyzed.  Results
                  were  calculated  for pollutant
                  concentrations, emission rates, and
                  emission factors. Twelve  test runs
                  were conducted on a conventional
                  stove, and two runs were conducted
                  on  a catalyst-equipped  stove.
                  Operating variables included  fuel
                  type (oak and pine), altitude (80 and
                  800 m), and burn rate (high and low).
                    Test data were analyzed using
                  analysis of variance and  linear
                  regression procedures. Although the
                  data  do not show strong statistical
                  significance due  to  variability of
                  results,  some general trends  do
                  appear  to  be   present  and
                  conclusions can be drawn as to the
                  effects of various operating variables
                  on woodstove emission. At the 99%
                  confidence bound (CB), oak  fuel
                  reduced emissions of total PAH
                  compounds relative to pine fuel.
                  Emisssions of carbon monoxide (CO)
                  showed a significant decrease from
                  low  to high burn rates, and a small
decrease when oak, rather than pine,
fuel was used.  The catalytic stove
(with  a  used combustor) reduced
emissions of CO, but did not show
significant reductions of total
particulate material at the 99% CB. No
altitude effects were noted at the 99%
CB.
  At the 95% CB, oak fuel reduced
the concentration, emission rate, and
emission factor for  many individual
PAH compounds, and reduced the
emission  factor for  total PAH
compounds. Marginal decreases  in
benzene were also attributed to the
use of oak, rather than pine, fuel.
  This  Project Summary was
developed by EPA's  Air and Energy
Engineering  Research  Laboratory,
Research  Triangle  Park, NC,  to
announce key findings of the research
project that is fully documented in two
separate volumes of the  same title
(see  Project  Report  ordering
information at back).

Introduction
  During  the winter of  1986-87, the U.S.
Environmental Protection Agency (EPA)
conducted an emission measurement
program  in Boise, Idaho, as part of the
Integrated Air Cancer Project (IACP). The
program  was designed to  identify  the
potential  mutagenic impact of residential
wood burning on  ambient and indoor air.
One aspect of the  field sampling effort
involved  obtaining emission samples
from chimneys serving wood  burning
appliances in Boise. This  project was

-------
 undertaken as a  parallel project  in  an
 instrumented  woodstove test  laboratory
 to quantify woodstove emissions typical
 of Boise conditions.
  The  purpose of this laboratory project
 was  to quantify  the  effect of several
 variables on emissions from woodstoves
 so that operational conditions of stoves in
 the field could  be inferred by  existing
 data bases.  The variables  included fuel
 species,  burn  rate,  and  altitude
 (barometric  pressure).  A  conventional
 woodstove, typical of many  existing stove
 models in the Boise  area,  was  used  as
 the test  appliance. Limited testing was
 also conducted on a catalytic woodstove.
 A  modified half-factorial  test matrix was
 used to obtain test results most useful to
 the IACP.
  A woodstove dilution sampler (Wood-
 stove  Sampling System, or  WS2) was
 used to collect particulate samples, while
a modified ambient sampler was used to
collect aldehyde  and volatile  organic
compound (VOC)  samples.  Particulate
samples  were analyzed  for total mass,
polynuclear  aromatic  hydrocarbons
(PAHs), semicondensible  organics (grav),
total chromatographical organic  (TCO)
compounds, and  inorganic compounds.
Concentrations of  carbon monoxide (CO),
carbon dioxide (C02),and oxygen (02) in
flue  gases were  recorded. Samples  of
wood  fuel and   residual  ash   were
analyzed for elemental composition.


Summary of Test Results
  Total  particulate capture values  were
substantial, ranging from about 2-16g  of
material.  Although  the grav-plus-TCO
values were expected to be similar to the
total  capture values because of the high
organic  content  of woodsmoke,  these
values  were typically  45-70%  of total
capture  values. It is  not  clear why the
unextractable  fractions from the  filter
catches  were so high.
  Table  1  lists  particulate  emissions  as
total  capture, grav,  and TCO fractions.
Total capture  emission  rates  for the
conventional stove ranged from 10.9 g/hr
(pine fuel, low elevation, low burn) to 50.7
g/hr (pine fuel, high elevation, low  burn).
While some replicate runs  showed close
agreement (Runs 1 and 3,  and 7 and  8),
other  runs  showed  considerable
differences: Runs 2 and 4,  6 and 14, and
9  and 10 varied by  factors  of  3 to 4
(Figure  1). Agreement and discrepancies
between the  replicate runs were  similar
when reported as grams per kilogram.
  Most of the measured PAH compounds
were  lower-molecular weight compounds,
with low emission rates of high molecular
weight  compounds.  Naphthalene
accounted  for  almost  half  of  the
emissions of measured PAH compounds
in  many samples (Runs 4,  6,  14, 16,  10,
7, 8, 12, 13). ,A very large fraction of total
PAH  compounds were recovered from
the  filter and  XAD-2  resin  extract
fractions,  with  a  relatively  small
contribution from the probe  rinse  fraction.
  Total  VOC  emissions  (C-|-C7
hydrocarbons) ranged from 9  g/hr (Runs
4 and 6) to 40  g/hr (run9). Replicate VOC
samples for  Runs 5 and 16 and Runs 7
and 8 were in relatively close agreement.
Other  replicate   tests  showed
considerable variation. Methane  was the
largest single  fraction of most samples,
although Run 1 had no reported methane.
                        -C
                        0>

                        05
                        (C
                        QC
                            20
                            10
                                SOLL-1   SOLH-1  SPLL-1   SPLH-1  SPHL-1  SPHH-1  EPHL-1
                                   SOLL-2  SOLH-2  SPLL-2  SPLH-2  SPHL-2  SPHH-2  EPHH-1
                                                         Test Code
                    Figure 1.    Emission rates for particulate capture fractions.

-------
   Table 1. Paniculate Matter Emissions
                                            Particulate Concentration
                                             Paniculate Emission Rate
                                                                                              Paniculate Emission Factor
Run
No.
1
3
21
4
6
14
5
16
9
10
7
8
12
13
Sample
Code3
SOLL-1
SOLL-2
SOLH-1
SOLH-2
SPLL-1
SPLL-2
SPLH-1
SPLH-2
SPHL-1
SPHL-2
SPHH-1
SPHH-2
EPHL-1
EPHH-1
Flue Gas
Flow
(dscmh)b
15.368
14.042
18.872
17.248
11.589
10.642
14.535
20.252
11.077
12.018
18.986
18.652
10.305
13.384
Sample
Volume
(1)c
3,315.6
3,330.8
1,590.0
3,385.6
3,080.9
3,344.0
2,979.3
2,965.9
3,579.0
2,870.0
2,800.6
2,821.5
3,661.4
3,518.8

Totals
2.052
2.471
2.309
0.919
0.941
2.886
k
1.113
4.574
0.928
0.678
0.608
0.601
0.601
\yiiu /
Grav.h
1.218
1.687
1.522
0.445
0.430
1.634
k
0.599
2.844
0.510
0.341
0.351
0.344
0.166

TOO'
0.306
0.220
0.312
0.166
0.190
0.668
0.339
0.334
0.817
0.303
0.188
0.198
0.140
0.367

Total9
31.5
34.7
43.6
15.6
10.9
30.7
k
22.5
50.7
11.2
12.9
11.3
6.2
8.0
[yini r
Grav.h
18.7
23.7
28.7
7.7
5.0
17.4
k
12.1
31.5
6.1
6.5
6.5
3.5
2.2

TOO'
4.7
3.1
5.9
2.9
2.2
7.1
4.9
6.8
9.0
3.6
3.6
3.7
1.4
4.9

Totals
23.3
27.1
16.8
7.5
6.6
18.4
k
7.7
43.0
6.9
4.1
3.6
4.7
4.1

-------
  P. Burnet, J. Houck, and R. Roholt are with OMNI Environmental Services, Inc.,
        Beaverton, OR 97005.
  Robert C. McCrillis is the EPA Project Officer (see below).
  The  complete report,  entitled  "Effects  of  Appliance  Type and Operating
        Variables on Woodstove Emissions,"
  Volume I.  Report and Appendices  AC: (Order No.  PB90 151457 'AS; Cost:
        $23.00, subject to change).
  Volume II.  Appendices D-F: (Order Nc. PB90  151465/AS; Cost: $31.00, subject
        to change) will be available only from:
            National Technical Information Service
            5285 Port Royal Road
            Springfield,  VA22161
            Telephone: 703-487-4650
  The EPA Project Officer can be contacted at:
            Air and Energy Engineering Research Laboratory
            U.S. Environmental Protection Agency
            Research Triangle Park, NC 27711
United States                    Center :or Environmental Research
Environmental Protection         Information
Agency                         Cincinnati OH 45268
Official Business
Penalty for Private Use $300

EPA/600/S2-90/001

-------