Low Emission and High Efficiency Residential Pellet-Fired Heaters
James E. Houck and Andrew T. Scott
OMNI Environmental Services, Inc.
5465 SW Western Ave., Beaverton, OR 97005
Carol R. Purvis
National Risk Management Research Laboratory
U.S. Environmental Protection Agency
MD-63, Research Triangle Park, NC 27711
Peter H. Kariher
ARCADIS Geraghty & Miller
4915 Prospectus Dr., Ste. F, Durham, NC 27713
John Crouch and Michael J. Van Buren
Hearth Products Association
1601 N. Kent St., Ste. 1001, Arlington, VA 22209
ABSTRACT
There are an estimated 350,000 residential pellet-fired heaters currently in use in the United
States. In recent years about 30,000 to 40,000 units have been sold annually. There are two
fundamental technology types: under-feed and top-feed. Pellets originating from both hardwood
and softwood residue are available. During the 1995-1996 heating season 654,000 tons of pellets
were sold nationwide.
Nearly all pellet-fired heaters have been sold since 1989. Even with this recent introduction,
there has been significant improvement in reliability, efficiency, and air pollutant emissions in
current models as compared to the earliest models. Electronic and microprocessor control of
combustion air, fuel feed, and convection fans is primarily responsible for the improvement.
Unfortunately, air emissions and efficiency data in the open literature and in government reports
available to air quality and energy planners and regulators are still based on the performance of
the earliest models introduced ca. 1989-1990.
Even the old-technology pellet-fired heaters are more efficient than traditional cordwood stoves.
They have lower greenhouse gas and acid precipitation impacts than home heating options based
on fossil fuels, and their particulate and carbon monoxide emissions are lower than cordwood
stoves.
Air emissions testing and efficiency testing on new under-feed and top-feed commercially
available residential heaters burning hardwood- and softwood-based pellets were conducted. The
results were compared with data from earlier models. Reductions in air emissions were
documented. The data from both the old- and new-technology stoves confirm that pellet-fired
heaters offer an environmentally sound option for the utilization of wood waste for home heating.

-------
INTRODUCTION
The use of wood for home heating represents about 9% of the Nation's space heating needs'. As
with all energy options, there are environmental benefits and drawbacks associated with
residential space heating with wood. Clear environmental benefits are low greenhouse gas
emissions and low acid precipitation impacts as well as the renewable nature of wood as a fuel.
The major environmental concerns have been particulate matter (PM) and carbon monoxide
(CO) emissions.
In the late 1980's pellet stoves and low emission cordwood stoves were developed. The PM and
CO emissions from the pellet stoves were documented as being dramatically lower than from
traditional cordwood stoves2"5. Since the introduction of the first pellet stove models,
considerable improvements have been made in their design with commensurate decreases in PM
and CO emissions.
There have been four studies which have evaluated the air emissions from pellet stoves. These
are: (1) An in-home study of early-technology U.S. Environmental Protection Agency (EPA)-
certified pellet stoves conducted for the U.S. Department of Energy (DOE) during the 1989/1990
heating season2,3 — six stoves (two models) and 23 one-week long test periods make up the data
base for the study; (2) An in-home study of early-technology pellet stoves exempt from EPA
certification conducted for the DOE during the 1990/1991 heating season3,4 — six stoves (four
models) and 24 one-week long test periods make up the data base for the study; (3) Recent
laboratory testing of an early model (ca. 1990) pellet stove for the EPA under four burn rates; and
(4) Laboratory testing of new under-feed and top-feed pellet stove models using both hardwood
and softwood pellets for the Pellet Fuels Institute (PFI) and Hearth Products Association (HPA).
The results of the first two studies have been published. They are the basis for the PM and CO
emissions factors compiled by the EPA in the AP-42 emissions factor document5 and have been
generally used to represent PM with aerodynamic diameters <10 fim (PMI0) and CO emissions
characteristic of pellet stoves. The results of the latter two studies are presented here for the first
time.
Burn rates, PM emissions, and CO emissions were measured in all four studies. Particle sizing
was done as part of the two new studies. The elemental, organic, and carbonate fractions of
particles were also quantified in the PFI/HPA study.
Additional particulate emission data beyond those available from the four studies are also
obtainable from EPA certification records of pellet stoves6. The EPA certification records
provide PM emission rates at the method-prescribed weighted burn rate for 23 different models.
EXPERIMENTAL
Because much of the PM from residential wood combustion is composed of organic compounds
which are semivolatile (i.e., they are partitioned between the gas and PM phases), the method of
sample collection will affect the mass of PM emissions measured. PM emissions from pellet
2

-------
stoves have been primarily measured by three techniques: (1) Dilution tunnel approaches of
which Method 5G is used in the certification process7, (2) A method based on the traditional EPA
Method 5 industrial source sampling train which is referred to as Method 5H when used for
woodstove certification7, and (3) automated in-home samplers. The emissions factors compiled
in AP-42 for pellet stoves measured in the two in-home studies2"4 were obtained by using an in-
home sampler referred to as an automated emissions sampler (AES). For the two laboratory
studies, PM emissions were determined in dilution tunnel systems similar to Method 5G.
Emissions rates listed by the EPA for certified pellet stoves are reported as Method 5H
equivalents. Equations to convert PM emissions data collected with the AES to equivalent 5G
values have been developed by the EPA8. Similarly, equations were also developed by the EPA
relating data collected by methods 5G and 5H7,8. While it is generally accepted that the
conversion equations are not highly accurate, to permit direct comparison of PM emissions, they
were used here to put all data in a 5G-like dilution tunnel format.
CO for pellet stoves in homes with the AES systems was measured using Tedlar bags which
collected a portion of the sampler's flow. CO concentrations in the bags were determined with
commercial CO analyzers. CO emissions factors for the two recent laboratory studies were
determined from periodic measurements of CO concentrations in the exhaust gas combined with
periodic measurements of stack flows.
Efficiency values were calculated by combining the flue loss method (i.e., sensible and latent
heat loss out the exhaust) and the combustion efficiency. The fraction of unburned residue was
determined gravimetrically. Greenhouse gas and acid precipitation impacts were estimated by
summing the emissions of greenhouse and acid gases in each step of the energy production
process (energy trajectory) leading to the production of space heat from pellets produced from
uncut standing trees'.
RESULTS AND DISCUSSION
CO and PM emissions from the four studies are the key results presented here. Ancillary data on
stove efficiencies and solid waste issues, along with the results of a review of greenhouse and
acid gas impacts for home space heating, are also included.
CO Emissions
CO emissions factors for new and early model pellet stoves, as well as the average value for
conventional cordwood stoves, are shown in Table 1. The studies of emissions from early
models under in-home use show that, on the average, CO is reduced by more than 75% under
actual use as compared to traditional uncertified cordwood stoves. The laboratory testing of an
early-model pellet stove at burn rates near those encountered in homes reveal similar CO
emissions factors. However, at a higher burn rate (for example, see the data for the 1.6 kg/hr
burn rate), CO emissions factors can become significantly larger for the early-model pellet
stoves. This is consistent with acknowledged difficulty in optimizing combustion conditions
with early pellet stoves with manual and/or independent controls of fuel feed rates, combustion
3

-------
air blowers, and dampers. Newer models generally have microprocessor control of one or more
of these functions which permits better optimization of combustion conditions over all burn
rates. However, the effect of varying combustion conditions can still be seen in CO emissions
from new model stoves, as CO emissions are not uniform throughout the burn rate range (Figure
1). Although higher CO emissions are seen at both the low and high burn conditions, the CO
emissions factors from new-technology pellet stoves are still markedly lower than for early pellet
stove models. It should be further noted that, even with these differences between new- and old-
technology pellet stoves, the CO emissions from both new- and early-technology pellet stoves, at
typical in-home burn rates, are much lower than from traditional uncertified cordwood stoves.
For the pellet stove models and pellet fuels tested, there was no significant difference in CO
emissions from new-technology under-feed and top-feed technology types or between hardwood
and softwood pellets used in them.
PM Emissions
Like CO emissions factors, the four studies demonstrate that PM emissions factors for early-
model pellet stoves are much lower than the emissions factors for traditional uncertified
cordwood stoves, and new-technology pellet stoves show considerable reduction in their PM
emissions factors, compared to the early-model pellet stoves (Table 1).
In addition to the results from the four studies, information on particulate emissions can be
obtained from emission certification tests. As with cordwood stoves, emission certification
requirements have been promulgated for pellet stoves7. There have been 23 pellet stove models
certified since 19886. However only one model is currently (as of October 21, 1999) listed as
certified for sale9. Most pellet stoves are exempt from certification requirements because they
have a greater than 35:1 air-to-fuel ratio at one or more burn settings. The results of the
certification test do confirm low PM emissions for pellet stoves. The average 5-G adjusted
emissions factor for all 23 certified stove models, based on the 5-H rate data and a weighted burn
rate of 1.16 dry kg/hr, is 0.70 g/dry kg (Table 1).
PM emissions factors compiled in AP-42 assume that all PM emitted from pellet stoves is
smaller than 10 pm in aerodynamic diameter (PM10), and the emissions factors developed from
the total PM data generated in the two field studies are represented as PM10 data. Measurements
with the 1990 model pellet stove for the EPA and on the new under-feed and top-feed models for
the PFI/HPA study show that not all, but on the average about 84%, of the total PM emissions
are PM10. Interestingly, the same data base also shows that about 81% of the PM emissions are
smaller than 2.5 pm (PM2 5). These results are consistent with the general understanding of the
source of PM from biomass combustion. Most are submicron size particles formed from the
chemically incomplete combustion of fuels, some are large particles of entrained ash or unburned
char, and very little PM falls between the extremes. The ramification of the PM size distribution
is that emissions factors based on total PM should be reduced to 84 and 81% for PM10 and PM2 5
emissions factors, respectively.
The elemental, organic, and carbonate carbon contents of the PM emitted from the new under-
4

-------
feed and top-feed models studied for the PFI/HPA were quantified. From the organic carbon
data, the organic compound content can be estimated using a multiplier of 2.0 to account for the
mass of oxygen and hydrogen associated with carbon in organic compounds. Similarly, the
fraction of the PM emissions composed of entrained ash can be estimated from the carbonate
carbon content, using a multiplier of 3.0 since wood combustion ash is about one-third by weight
carbonates. The carbon analysis revealed a dichotomy between PM emissions from under-feed
and top-feed pellet stove models (Figure 2) and between pellet stoves in general and cordwood
stoves. A large fraction of the PM emitted from the top-feed model was elemental carbon.
Entrained ash, however, as indicated by the carbonate carbon, was not detectable. The elemental
carbon fraction increased with increasing burn rate, reaching 88% of the total PM emissions at
the highest burn rate. In contrast, virtually no elemental carbon was detected in the PM emitted
from the under-feed model, but entrained ash was estimated as comprising 26 and 8% of the PM
emissions at a medium burn rate for softwood and hardwood pellets, respectively. Visual
observation of filters used to collect the PM samples confirmed the difference in chemical
makeup of the PM. The filters used to collect PM from both the top-feed model used for the
PFI/HPA study and the 1990 top-feed model used for the EPA study were black, characteristic of
elemental carbon (also called graphitic carbon or soot). However, the filters used to collect PM
from the under-feed model were light tan. The chemical makeup of PM emitted from a
cordwood stove is unlike that from either top- or under-feed pellet stoves. Typically, particles
from a cordwood stove are composed of 10 to 20% elemental carbon and less than 1% inorganic
ash10 ". In addition, cordwood stove PM filters are generally dark brown to black.
Total PM emissions have often been used as a surrogate for air emissions of specific toxic
compounds such as polycyclic organic matter (POM). The gross chemical differences among
PM emitted from top-feed pellet stoves, under-feed pellet stoves, and traditional cordwood stoves
demonstrated that total PM emissions cannot reliably be used as a surrogate for individual or
groups of specific organic species when comparing emissions among these different technology
types.
Efficiency
The efficiencies of pellet stoves are considerably higher than those of cordwood stoves.
Efficiencies for new-model pellet stoves have been measured, by OMNI Environmental Services,
to be as high as 87%; whereas, the efficiencies for uncertified cordwood stoves are estimated as
54%s. The efficiencies of new-technology pellet stoves are much higher than the earliest models
which had efficiencies documented from the in-home studies of 56 and 68% for exempt and
certified models, respectively5. The significance of higher efficiency, beyond its obvious
desirability, is that less fuel is used by a home occupant to produce the same amount of heat,
consequently the effective reduction in air emissions per unit of heat produced is even greater
than the emissions factors, reported in units of mass pollutant per mass of fuel burned, would
imply.
Greenhouse Gas and Acid Precipitation Impacts
5

-------
There are two well documented air quality advantages associated with all residential wood
combustion (not just pellets) which are important to consider when evaluating the environmental
implications of home heating options. These advantages are low greenhouse gas emissions and
low acid precipitation impacts. When all the steps involved in energy production are taken into
consideration, home heating with wood produces less than half the carbon equivalents of
greenhouse gases per unit of energy than any other home heating option1. The release of methane
and carbon dioxide into the atmosphere from the activities leading to the production of space
heat is responsible for the greenhouse impacts from home heating. In addition to the high energy
return on investment (EROI) associated with wood fuel, harvesting of mature trees for fuel
permits more rapid carbon fixation in younger replacement trees and reduces the effective
greenhouse impacts from wood fuel combustion.
Most acid precipitation impacts are produced by sulfur gases or nitrous oxide gases released
during the extraction, processing, and the higher temperature combustion of fossil fuels. Little
fossil fuel is invested in the production of space heat from wood (including pellets). A detailed
analysis of emissions from each step of the energy production process shows that residential
wood combustion produces the lowest amount of acid equivalents (a measure of the acid
precipitation potential) per unit of heat among all the home space heating options'.
Solid Waste Disposal
The residue (unburned wood char and inorganic salts) remaining after combustion of fuel in a
cordwood stove typically ranges from 1 to 5% of the fuel mass. The residue for a pellet stove
averages less than 0.5% of the fuel mass. Due to the higher efficiencies of pellet stoves, less fuel
mass is required to satisfy the same heat demand with a pellet stove than with a cordwood stove.
The combination of less fuel mass burned and a lower percent residue production makes solid
waste disposal from pellet stoves significantly less of an issue than for cordwood stoves. In
addition, wood ash (derived from either cordwood or pellets) is relatively benign. In fact, its
high calcium carbonate and potassium content makes it a good agricultural soil amendment.
CONCLUSIONS
The key conclusions that were reached by reviewing the pellet stove emission data are:
•	New-technology pellet stoves produce much less CO than uncertified cordwood stoves.
•	New-technology pellet stoves produce much less PM than uncertified cordwood stoves.
•	The emissions factors for both CO and PM are lower for new-model pellet stoves than for
earlier models.
•	Not all PM emitted from pellet stoves are PM10 or PM2 v Approximately 84% of PM is
PM10 and about 81% is PM2 5. PM10 and PM2 5 emissions factors, if based on total PM
6

-------
measurements, should be adjusted accordingly.
•	The chemical makeup of PM emitted from top-feed pellet stoves and under-feed pellet
stoves is different, and the chemical makeup of PM from both technology types is
different from that emitted from cordwood stoves. Consequently, total PM emissions are
not accurate surrogates for emissions of specific organic compounds such as those
identified as "air toxics."
•	Pellet stoves generate less solid waste than cordwood stoves.
•	The high efficiencies and low emissions of PM and CO characteristic of new-technology
pellet stoves, combined with low greenhouse gas impacts, low acid precipitation impacts,
and minimal solid waste issues, make pellet stoves an environmentally sound home space
heating option.
ACKNOWLEDGMENTS
The authors would like to thank: Dane Harman of Harman Stove Company for providing a pellet
stove, Jerry Whitfield and Doug Jenks of Pyro Industries for providing a pellet stove and
information on older model stoves, Averill Cook of Catamount Fuel Corporation for providing
pellet fuel, Robert Cary of Sunset Laboratory for conducting carbon analyses, and Robert
McCrillis, now retired from the U.S. EPA, for his participation in the pellet stove testing
program.
REFERENCES
1.	Houck, J.E., Tiegs, P.E., McCrillis, R.C., Keithley, C. and Crouch, J., 1998, Air
Emissions from Residential Heating: The Wood Heating Option Put into Environmental
Perspective, in: Proceedings of the A&WMA and EPA Specialty Conference: Emissions
Inventory: Living in a Global Environment, Vol. 1, pp. 373-384.
2.	Barnett, S.G. and Roholt, R.B., 1990, In-home Performance of Certified Pellet Stoves in
Medford and Klamath Falls, Oregon, OMNI Environmental Services Inc. report to U.S.
Department of Energy, DOE/BP-04143-1.
3.	Barnett, S.G., Houck, J.E. and Roholt, R.B., 1991, In-Home Performance of Pellet Stoves
in Medford and Klamath Falls, Oregon, presented at A&WMA 84th Annual Meeting,
Vancouver, BC, June 16-21, 1991, paper 91-129.3.
4.	Barnett, S.G. and Fields, P.G., 1991, In-home Performance of Exempt Pellet Stoves in
Medford, Oregon, OMNI Environmental Services, Inc. report to U.S. Department of
Energy, DOE/BP-04123-2.
7

-------
5.	U.S. Environmental Protection Agency, 1996, Compilation of Air Pollutant Emission
Factors, AP-42, Fifth Edition, Volume I, Stationary Point and Area Sources, Chapter 1,
External Combustion Sources, Section 1.10, Residential Wood Stoves (Supplement B),
October 1996, http://www.epa.gov/ttn/chief/ap42cl.html.
6.	U.S. Environmental Protection Agency, September 21, 1998, Master Manufacturer List,
Woodheater Program, Telephone 202-564-7021, Manufacturing, Energy and
Transportation Division, Washington, D.C.
7.	U.S. Environmental Protection Agency, 1988, 40 CFR Part 60, Standards of Performance
for New Stationary Sources; New Residential Wood Heaters, Federal Register, Vol. 53,
No. 38, Friday, February 26, 1988, pp. 5860-5926.
8.	E.H. Pechan & Associates, 1993, Emission Factor Documentation for AP-42 Section
1.10, Residential Wood Stoves, report for U.S. Environmental Protection Agency's
Office of Air Quality Planning and Standards, EPA Contract No 68-D0-0120.
9.	U.S. Environmental Protection Agency, October 21, 1999, List of Certified Woodstoves,
Wood Heater Program, Telephone 202-564-7021, Manufacturing, Energy and
Transportation Division, Washington, D.C.
10.	Watson, J.G., Chow, J.C., Richards, L.W., Neff, W.D., Andersen, S.R., Dietrich, D.L.,
Houck, J.E. and Olmez, I., 1988, The 1987-88 Metro Denver Brown Cloud Study,
Volume HI: Desert Research Institute report to 1987-88 Metro Denver Brown Cloud
Study, Inc.
11.	Houck, J.E., Chow, J.C., Watson, J.G., Simons, C.A., Pritchett, L.C., Goulet, J.M. and
Frazier, C.A., 1989, Determination of Particle Size Distribution and Chemical
Composition of Particulate Matter from Selected Sources in California, OMNI
Environmental Services, Inc. report to California Air Resources Board, NTIS PB89
232805.
8

-------
Table 1
Carbon Monoxide and Particulate Emissions
Description
Pellet burn rate
(dry kg/hr)
CO emission factor
(g CO/dry kg fuel)
PM emission factor
(g PM/dry kg fuel)3
Conventional
cordwood stove
NAb
115.4
12.0
90/91 exempt pellet
stoves, avg. 24 one-
week runs, softwood
pellets
0.58
26.1
2.77
89/90 certified pellet
stoves, avg. 23 one-
week runs, softwood
pellets
0.70
22.4
1.29
Certification test
results, avg. 23
models
1.16
NDC
0.70
Lab tests 1990
model, hardwood
pellets
0.7
23.2
3.5
0.8
27.8
2.0
0.9
29.7
3.0
1.6
155
7.6
New top-feed,
softwd. pellets, test 1
0.72
7.19
0.44
New top-feed,
softwd. pellets, test 2
1.46
2.34
0.60
New top-feed,
softwd. pellets, test 3
2.45
8.17
1.0
New bottom-feed,
softwd. pellets, test 4
1.55
1.8
0.26
New bottom-feed,
hrdwd. pellets, test 5
1.55
2.7
0.40
a All PM data adjusted to U.S. EPA Method 5G (40CFR, Part 60, App. A) equivalent to permit
comparisons. Data for conventional cordwood stove from AP-42 (reference 5).
b Not applicable.
c Not detected.
9

-------
450 t



¦^^M.
1.58
2.40	3.22
Burn Rate (kg/hr)
4.06
4.90
Figure 1. Carbon monoxide concentration in exhaust gas versus burn rate with a new top-feed
pellet stove model burning softwood pellets.
1.0
0.9
0.8 -
0.7
o>
•X
(0
.2 0.6
(O
<0
E 0.5
UJ
©
4-*
ro
3
O
t
TO
0.2
0.4
0.3
0.1
0.0
~	Organic Compounds
¦ Elemental Carbon
~	Ash
J i
3
Test#
Figure 2. Composition of PM emissions. Tests 1,2, & 3 are for a top-feed model burning
softwood pellets at 0.72, 1.46, and 2.45 kg/hr. Tests 4 & 5 are for an under-feed model. Softwood
pellets at 1.55 kg/hr were burned in test 4. Hardwood pellets at 1.55 kg/hr were burned in test 5.
10

-------
TECHNICAL REPORT DATA
N RM RL~ RT P" P~ 510 (Please read Instructions on the reverse before completing)

1. REPORT NO. 2.
EPA/600/A-00/102
3. RECIPte
4. TITLE AND SUBTITLE
Low Emission and High Efficiency Residential
Pellet-fired Heaters
5. REPORT OATE
6. PERFORMING ORGANIZATION CODE
7.AUTHOR(s)JtHouck and Scott (OMNI), C. Purvis (EPA),
P.Kariher (ARCADIS), and J. Crouch and M. Van
Buren (Hearth Products)
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESSq Environmen-
tal Services Inc., Beaverton, OR 97005; ARCADIS
Geragthy and Miller, Durham, NC 27713; and Hearth
Products Assoc., Arlington, VA 22209
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-D4-0005, WA 4-002
(ARCADIS)
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Air Pollution Prevention and Control Division
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOQ COVERED
Published paper; 107 9 7-9/98
14. SPONSORING AGENCY CODE
EPA/600/13
is.supplementary notes^pp££) project officer is Carol R. Purvis, Mail Drop 63, 919/-
541-7519. For presentation at Bioenergy 2000 Conference, Buffalo, NY, 10/15-19/00.
16. abstract The paper gives results of air emissions testing and efficiency testing on
new commercially available under~feed and top-feed residential heaters burning
hardwood- and softwood-based pellets. The results were compared with data from
earlier models. Reductions in air emissions were documented. The data frOm both
the old and new technology stoves confirm that pellet-fired heaters offer an environ-
mentally sound option for the utilization of wood waste for home heating. (NOTE:
An estimated 350,000 residential pellet-fired heaters are currently in use in the
U. S. In recent years, about 30, 000 to 40, 000 units have been sold annually. There
are two fundamental technology types: under-feed and top-feed. Pellets originating
from both hardwood and softwood residue are available. During the 1995-1996 heat-
ing season, 654,000 tons of pellets were sold nationwide. Nearly all pellet-fired
heaters have been sold since 1989. Even with this recent introduction, there has been
significant improvement in reliability, efficiency, and air pollutant emissions in
current models, compared to the earliest models. Electronic and microprocessor
control of combustion air, fuel feed, and convection fans is primarily responsible
for the improvement. Unfortunately, air emissions and efficiency data in the open
literature and in government reports are still based on earlier-model performance.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b.lOENTIFIERS/OPEN ENDED TERMS
c. COS ATI Field/Group
Pollution
Wood
Pellets
Combustion
Emission
Stoves
Pollution Qontrol
Stationary Sources
Wood Pellets
Residential Heaters
13	B
11L
11G
2 IB
14	G
13 A
18. DISTRIBUTION STATEMENT
19. SECURITY CLASS (This Report)
"ad
21. NO. OF PAGES
10
Release to Public
_ASS (This page)
ed
22. PRICE
EPA Form 2220-1 (9-73)

-------