WoodCombustion_FINAL-REPORT

Literature review of emissions of modern wood
combustion devices and emissions reducing

technologies, under real-life conditions
FINAL REPORT
FLANDERS
ENVIRONMENT AGENCY en.vmm.be
FINAL REPORT
Literature review of emissions of modern wood combustion

devices and emissions reducing technologies, under real-life
conditions
Tytgat Tom1, Walpot Griet1, Cools Jan2, Lenaerts Silvia1,2
1
University of Antwerp, Research Group ‘Sustainable Energy, Air & Water Technology’
2
University of Antwerp, Institute of Environment & Sustainable Development
Study conducted by the University of Antwerp for the Flanders Environment Agency
(VMM) under contract ALMC/LUC/2017/literatuurstudie moderne houtkachels
II Literature overview of modern wood combustion devices
Executive Summary
This study attempts to provide an overview of real-life emissions of wood stoves and to
investigate the relationship between emissions from laboratory tests and emissions in real
life. Furthermore, technologies are mapped that can possibly reduce the emissions.
The first step in the study was determining a typology of the various wood combustion
appliances available on the market. In a second step, for each type of wood combustion
device information was collected regarding the emissions of particulate matter (PM),
carbon monoxide (CO), nitrogen oxides (NO x), organic gaseous carbon (OGC) and
polyaromated hydrocarbons (PAH). The emission data were gathered from tests conducted
under laboratory conditions, under simulated real life conditions (called pseudo real-life in
this study) or effectively measured values in-house (termed as real-life conditions). These
data were bundled per type of wood combustion device and are summarised in Figure S1
below.
Figure S1: Boxplot summary of all data points (number given between brackets) for each type
of wood combustion appliance. *category containing outliers above 1000 mg/MJ.
From the figure above, it becomes clear that PM emissions vary significantly within
and between different types of devices. In general, more recent devices with newer
technology show lower PM emissions compared to the older devices, which are more on
the left in the graph. It is important to note that the majority of the collected data originates
from tests in laboratory, either following standard test procedures or simulating real life
operation. This illustrates the shortage of actual real-life data. Additionally, a lack of
scientific studies on emissions from most recent stoves exists. Most tested stoves
are several years old, making it difficult to estimate emissions from stoves with most recent
technology incorporated.
III
In addition to the overview of the available wood combustion devices, an overview is also
included of technology that can result in a reduction of the emissions. Two strategies exist
to achieve the latter. First, it can be done by means of source control, in which especially
stove design and air flow prove to be crucial elements. Source control measures typically
aim to achieve optimal combustion at real operation conditions. This effect is particularly
visible in the reduced emission rates of more recent stoves, which contain these types of
measures. It is this effect that is visible in Figure S1. Secondly, end-of-pipe solutions
can also result in reduced emissions. A widely used and commercially available technology
is electrostatic precipitation. This technology is based on the collection of PM on an
electrode. In practice, this appears to reduce part of the emissions, but the achieved
reduction is highly dependent on the circumstances and the used stove. A second end-of-
pipe solution is the incorporation of a catalyst inside the combustion chamber or inside the
chimney. This is mainly effective for the reduction of CO and organic substances and is less
suitable for PM. Fluctuating efficiencies are reported and in some cases, there is a risk of
forming harmful by-products. Correct use and maintenance of such systems is therefore
important.
An important conclusion from this literature study is that a considerable knowledge gap
exists on several subjects. First of all, there is a limited amount of data collected under
real-life conditions. Additionally, the interpretation and comparison of emissions in
literature and in legislation appears to be difficult because of the difference in the use
of units. In literature, emissions are based on energy content to allow comparisons
between different types of stoves. The legislation on the other hand relies on emissions
per volume of air. The latter is highly stove specific and often not known, which makes a
comparison between different stoves very difficult.
Another aspect with a lot of ambiguity is the formation of secondary organic aerosols
(SOA). The formation of this fraction of particulate matter occurs when the exhaust gases
exit the chimney and end up in the atmosphere. Under the influence of various external
factors, the organic components present in the exhaust gases will react. Consequently, a
new fraction of particulate matter is formed. The concentration of SOA emissions can be
of the same order of magnitude as the concentration of primary PM during the combustion
process. The precise mechanism of SOA formation is very complex and dependant on
various factors, making the quantification of this fraction in real life very difficult. Since
SOA are formed after leaving the chimney, most tests do not measure the SOA fraction
which could result in a significant underestimation of real-life emissions.
The study also provides an overview of the main causes that lead to large variations
between the different testing conditions. Several aspects play a role in this variation, e.g.
used technology, test method and operational conditions. The latter is mainly focused on
the behaviour of the user of a wood combustion device. Elements such as the type of wood
used, the degree of humidity, the fire lighting procedure etc. all have a direct influence on
emissions from residential wood combustion. In the standard tests, the experimental
conditions are based on the most optimal conditions while in reality a lot of the
conditions are far from ideal, resulting in large differences between lab and real-life
tests.
An important problem with estimating real-life emissions is that consistent information of

typical or average real user behaviour is lacking. As a result, assumptions currently made
in scientific literature on (pseudo) real life emissions can proof to be completely different
IV Literature overview of modern wood combustion devices
in reality. A better understanding of these aspects is required. This can be done by

performing surveys with real-life users in order to gain knowledge about their operation of
a stove. Another problem is that a small fraction of real users (e.g. people burning wet,
contaminated wood in old stoves) could have a relatively high contribution to the total
emissions. Therefore good knowledge of best, average and worst case scenarios and their
occurrence is essential to estimate the total real-life emissions.
V
Management samenvatting
Deze studie geeft een overzicht van de effectieve uitstoot door huishoudelijke
houtverbranding en de relatie met de theoretische uitstoot gemeten in labo omgeving.
Verder worden ook technologieën in kaart gebracht die de uitstoot van houtverbranding
mogelijk kunnen reduceren.
In eerste instantie werd een typologie gemaakt van de verschillende

houtverbrandingsinstallaties op de markt. Aan de hand van deze typologie werd informatie
gezocht over de uitstoot van kachels, zowel voor fijn stof (PM) als voor andere
componenten zoals koolstofmonoxide (CO), stikstofoxides (NO x), gasvormige organische
componenten (OGC) en polycyclische aromatische koolwaterstoffen (PAH). Deze informatie
werd in kaart gebracht voor labotesten, die ofwel volgens de geijkte testprocedures ofwel
volgens gesimuleerde reële omstandigheden plaatsvonden of voor testen die effectief in
het veld bij mensen thuis gebeurden. Deze gegevens werden per type gebundeld en
samengevat in Figuur S2.
Figuur S2: Boxplot overzicht van alle datapunten (aantal tussen haakjes) voor de verschillende
types kachels. * Dit zijn categorieën die outliers bevatten boven de 1000 mg/MJ.
Uit de figuur blijkt dat de effectieve uitstoot voor PM zeer sterk varieert binnen en
tussen verschillende types van kachel. De uitstoot van meer recente kachels, met de
nieuwste technologie, is doorgaans lager dan deze van oudere kachels, die zich meer links
in de figuur bevinden. Belangrijk hierbij is wel dat de meeste data onder labo-
omstandigheden werden bepaald, volgens geijkte procedures of volgens gesimuleerde
reële omstandigheden. Om een correct beeld te krijgen van de reële uitstoot is het nodig
om houtverbrandingstoestellen ook te testen op het veld en bovendien ook de laatste
nieuwe toestellen te testen. Voor de laatste nieuwe toestellen zijn er trouwens geen of
nauwelijks wetenschappelijke studies voorhanden, wat het momenteel zeer moeilijk maakt
om hun reële prestaties in te schatten.
VI Literature overview of modern wood combustion devices
Naast de beschikbare kachels werd ook een overzicht opgenomen van verschillende
technologieën die in een vermindering van de uitstoot kunnen resulteren. Er zijn twee
verschillende strategieën om dit te bekomen: broncontrole en end-of-pipe
oplossingen. Bij broncontrole wordt door voornamelijk kacheldesign en
luchtvoorzieningen de vorming van polluenten vermeden of verminderd. Hun direct effect
is vooral zichtbaar in de verminderde uitstoot van meer recente kachels, die trachten de
verbranding optimaal te laten verlopen onder reële omstandigheden. Bij end-of-pipe
oplossingen worden gevormde polluenten aangepakt voordat zij de schouw verlaten. Een
veel gebruikte en commercieel beschikbare end-of-pipe technologie is elektrostatische
precipitatie, die vooral effectief is in het capteren van PM. In de praktijk blijkt dit ook
effectief te werken, maar de behaalde reductie is sterk afhankelijk van de
verbrandingsomstandigheden en de gebruikte kachel. Een tweede end-of-pipe methode
bestaat erin katalysatoren te incorporeren na de verbrandingskamer. Op deze manier
worden vooral de uitstoot van CO en organische stoffen verminderd en minder die van PM.
Ook hier zijn schommelingen in verwijderingsefficiëntie te zien. Bovendien bestaat in
sommige gevallen het risico op vorming van schadelijke bijproducten. Het is dan ook
belangrijk dat dergelijke systemen op een correcte manier gebruikt en onderhouden
worden.
Ondanks de talrijke beschikbare informatiebronnen in verband met huishoudelijke

houtverbranding blijken er ook een hele reeks kennislacunes te bestaan in het vakgebied.
Vooreerst is er slechts beperkte informatie beschikbaar voor uitstoottesten uitgevoerd op
het veld. Daarnaast blijkt ook de interpretatie en het vergelijken van emissies tussen
literatuur en wetgeving moeilijk te zijn vanwege het verschil in gebruik van eenheden.
In wetenschappelijke literatuur wordt er typisch gewerkt met eenheden gerelateerd aan
de energie-inhoud, waardoor vergelijkingen tussen verschillende types van kachels
mogelijk wordt. De wetgeving daarentegen baseert zich op emissies per hoeveelheid lucht.
Dit laatste is echter zeer specifiek per kachel, waardoor een onderlinge vergelijking zeer
moeilijk wordt. Omzetten van de ene eenheid in de andere is zeer moeilijk gezien er extra
informatie noodzakelijk is die vaak niet gespecifieerd wordt.
Een ander aspect waarover nog veel onduidelijkheid bestaat, is de vorming van
secundaire organische aerosolen (SOA). Deze fractie van fijn stof ontstaat in de
atmosfeer, als de uitstootgassen de schouw verlaten hebben. Onder invloed van
verschillende externe factoren zullen de organische componenten reageren, met als
resultaat de vorming van een bijkomende fractie aan fijn stof. De concentratie van deze
SOA fractie kan, afhankelijk van het soort kachel, gelijkaardig zijn aan de uitstoot van
primair fijn stof en is dus niet te onderschatten. De vormingsmechanismen die hierachter
zitten zijn zeer complex en afhankelijk van tal van factoren, waardoor het effectief in kaart
brengen van het SOA-vormend potentieel zeer moeilijk is. Omdat SOA pas gevormd
worden nadat ze de schouw verlaten, worden ze zelden gemeten in testen, wat kan
leiden tot belangrijke onderschattingen van de reële emissies.
Verder wordt er in de studie ook een overzicht gegeven van de oorzaken die ertoe leiden
dat er grote variatie bestaat tussen uitstoten onder verschillende meetomstandigheden.
Hierbij blijken een aantal aspecten een rol te spelen; de kacheltechnologie, de
meetmethode en operationele condities. Met betrekking tot de laatste, is vooral het gedrag
van de gebruiker belangrijk. Bepalende factoren zijn de gebruikte houtsoort, de
vochtigheidsgraad van het hout, de aanmaakprocedure… Bij de geijkte testprocedures
VII
gebeurt de verbranding steeds onder optimale omstandigheden die sterk afwijken van het
reële gebruik. In de praktijk zijn de verschillende factoren vaak niet optimaal. Als
gevolg hiervan kunnen grote verschillen ontstaan tussen gemeten uitstoten en werkelijke
uitstoten.
Een belangrijk probleem bij het schatten van de reële uitstoot is dat er geen eenduidige
informatie over gemiddeld of typisch gebruikersgedrag bestaat. Als gevolg hiervan worden
momenteel in de wetenschappelijke literatuur aannames gemaakt over (gesimuleerde)
reële uitstoot die volledig verschillend zouden kunnen zijn van de realiteit. Om de
noodzakelijke extra kennis hieromtrent te vergaren, kunnen enquêtes uitgevoerd worden
bij gebruikers over hun werkelijke stookgedrag en –gebruiken. Een volgend probleem is
dat een klein deel van de werkelijke gebruikers een groter dan gemiddeld aandeel hebben
in de totale uitstoot ten gevolge van houtverbranding door suboptimaal stookgedrag door
bijvoorbeeld nat of behandeld hout stoken in oude installaties. Om een realistische
schatting te maken van totale werkelijke uitstoot is het om die reden essentieel om
optimale, gemiddelde en slechte gebruiksscenario’s en hun relatieve belang te kennen.
VIII Literature overview of modern wood combustion devices
Content
Executive Summary ............................................................................................ II

Management samenvatting .................................................................................. V
Content .......................................................................................................... VIII
List of Figures ................................................................................................... IX
List of Tables ...................................................................................................... X
1. Introduction................................................................................................. 1
2. Scope and definitions .................................................................................... 3
2.1. Stove typology ....................................................................................... 3
2.2. Emissions reduction technologies ............................................................. 5
2.3. Testing conditions for measurements of emissions...................................... 5
2.4. Studied emissions .................................................................................. 6
3. Information of emissions per type .................................................................. 8
4. Strategies for emission reduction ..................................................................24
4.1. Source control ......................................................................................24
4.2. Electrostatic Precipitator.........................................................................25
4.3. Catalytic converter ................................................................................27
4.4. Ceramic filters ......................................................................................29
5. Impact of emissions on the atmospheric concentration .....................................30
5.1. Secondary Organic Aerosol Formation ......................................................30
5.2. Real-life implementations on a larger scale ...............................................31
6. Knowledge gaps & future research needs .......................................................33
6.1. Lack of data under real-life conditions ......................................................33
6.2. Comparison of data is challenging ...........................................................33
6.3. Effective impact on air quality .................................................................34
7. Towards characterisation of real-life conditions ...............................................35
8. Policy Advice...............................................................................................39
9. Conclusion ..................................................................................................41
10. References ..............................................................................................42
List of Figures IX
List of Figures
Figure 3-1: Overview of PM emissions of closed fireplaces and conventional wood stoves.
(AIR) with the reference indicates air starved experiments. ........................................ 9
Figure 3-2: Overview of PM emissions of masonry stoves. .........................................12
Figure 3-3: Overview of PM emissions of high-efficiency conventional stoves and advanced

combustion stoves. ...............................................................................................14
Figure 3-4: Overview of PM emissions of modern pellet stoves and conventional boilers.
..........................................................................................................................16
Figure 3-5: Overview of PM emissions of advanced boilers. ........................................18

Figure 3-6: Overview of PM emissions of wood/pellet boilers. .....................................20
Figure 3-7: An overview of the number of references found per emission measurement
condition and per type of appliance. ........................................................................22
Figure 3-8: Boxplot summary of all data points (number given between brackets) for each
type of wood combustion appliance. *category containing outliers above 1000 mg/MJ. .23
Figure 5-1: Overview of the burning process of wood combustion focused on PM10.
Adapted after (Nussbaumer 2017). .........................................................................30
X Literature overview of modern wood combustion devices
List of Tables
Table 1-1: Current Belgian requirements for new combustion devices regarding minimum
efficiency level and maximum CO and PM emissions, measured at 13% O2, TS 15883. .. 1
Table 1-2: European eco-design directive with maximum emissions for different pollutants
emitted by different RWC devices, at 273 K and 1 013 mbar at 13 % O2. ..................... 2
Table 3-1: Overview of closed fireplace emissions of PM, CO, OGC/PAH and NOx. .........10
Table 3-2: Overview of reported conventional wood stove emissions of PM, CO, OGC/PAH
and NOx. .............................................................................................................10
Table 3-3: Overview of masonry stoves emission factors for PM, CO, NOx and OGC/PAH at
different testing conditions. ...................................................................................13
Table 3-4: Overview of emission factors for PM, CO, OGC, PAH and NO x of high-efficiency
stoves .................................................................................................................15
Table 3-5: Overview of emission factors for PM, CO, OGC, PAH and NO x of advanced
combustion stoves ................................................................................................15
Table 3-6: Overview of PM, CO, NOx and OGC/PAH emissions of modern pellet stoves ..17
Table 3-7: Overview of PM, CO, NOx and OGC/PAH emissions of conventional boilers. ...17
Table 3-8: Overview of PM, CO, OGC/PAH and NOx emission factors of advanced boilers.
..........................................................................................................................19
Table 3-9: Overview of pellet boiler emissions of PM, CO, OGC/PAH and NOx emissions. 21
Table 3-10: Overview of emissions data collected from all different types of wood
appliances with unit mg/MJ. ...................................................................................23
Table 4-1: Overview of removal efficiency based on the IEA report of Task 32. Most tests
are conducted under laboratory conditions. Where available, real-life tests were included.
..........................................................................................................................26
Table 4-2: Achieved removal efficiencies for TSP and PM1 and effective TSP concentration
in the flue gas for the different tested wood combustion devices (Thomas Brunner,
Wuercher, and Obernberger 2017). ........................................................................27
Table 6-1: Overview of units for emitted pollutant concentrations from wood combustion
devices used in scientific literature. .........................................................................33
Table 6-2: Table containing conversion calculation based on (Skreiberg 2002). ............34
Introduction 1
1. Introduction
The Flanders Environment Agency (Vlaamse Milieumaatschappij – VMM) recently reported

that in a winter campaign in Dessel approximately one third of PM10 in the air could be
attributed to wood combustion (Vlaamse Milieumaatschappij 2016; Van Poppel et al.
2017). Besides PM, other pollutants as polyaromatic hydrocarbons (PAHs), NOx and volatile
organic compounds (VOC) are also reported to be part of the emissions due to residential
wood combustion (RWC) (Kaivosoja et al. 2012; WHO 2015). Associated with RWC caused
air pollution are health problems related to lung malfunctioning, but also carcinogenic and
cardiovascular effects are suggested (Kocbach Bølling et al. 2009; WHO 2015; Van Poppel
et al. 2017). Besides effects on human health, RWC is also known to be a large source of
black carbon, which is an important substance contributing to the problem of global
warming (Carvalho 2016; WHO 2015).
Recent awareness about the negative effects of RWC initiated the introduction of a
campaign by the Flemish Government to inform people on how to use a wood stove in a
smart and responsible way with the introduction of www.stookslim.be. Beside this, the
Flanders Environment Agency also started with an information campaign in which it asks
people not to burn wood during periods of bad air quality (www.vmm.be/stookadvies).
Furthermore, the Federal public service Health, Food Chain Safety and Environment started
checking stoves available on the Belgium market to verify their compliance with current
legislation. The Belgian legislation [KB 2010/24412] describes the requirements for new
heating devices burned with solid fuels, amongst which minimal efficiency levels and
maximum CO and PM emissions (Table 1-1).
Table 1-1: Current Belgian requirements for new combustion devices regarding minimum
efficiency level and maximum CO and PM emissions, measured at 13% O2, TS 15883.
Minimum efficiency Maximum CO Maximum particulate
level [%] emissions [%] emissions [mg/Nm³]
Operation mode Continuous Not Continuous Not Continuous Not
continuous continuous continuous
Free-standing
stove 65 75 0,8 0,1 150 40
(NBN EN 13240)
Enclosed stove
65 75 0,8 0,1 150 40
(NBN EN 13229)
Heat accumulating
device for solid
75 0,1 40
fuels
(NBN EN 15250)
Pellet device
80 0,02 30
(NBN EN 14785)
Boiler stove
75 0,1 150
(NBN EN 12809)
Boiler
75 1,5 100
(NBN EN 303-5)
Open fireplace
65 0,8 300
(NBN EN 13229)
Based on the conformity tests, non-compliant stoves are taken out of the market. Next to
the testing, the federal public service Health, Food Chain Safety and Environment also
published a list of all stoves currently sold on the Belgian market with some basic
information regarding their emissions. Currently (21/12/2017), 3047 different devices are
permitted in Belgium, as can be seen on the website of the federal public service
(https://www.health.belgium.be/nl/e-services/lijst-van-verwarmingstoestellen).
2 Literature overview of modern wood combustion devices
Beside the Belgian legislation, there is also the European eco-design directive, specifying
emissions standards for wood combustion devices based on three different types, which
will come into force in 2022. An overview of the values can be found in Table 1-2.
Table 1-2: European eco-design directive with maximum emissions for different pollutants
emitted by different RWC devices, at 273 K and 1 013 mbar at 13 % O 2.
Type of RWC device CO PM OGC NOx
(mg/Nm³) (mg/Nm³) (mg C/Nm³) (mg/Nm³)
Open front 2000 50 120 200

Close front, no pellets 1500 40 120 200
Close front, pellets 60 20 60 200
All conformity tests are currently executed under laboratory conditions, in controlled
facilities with strict control of i.a. fuel type. The question arises how representative these
tests are for emissions from stoves installed inside a household and operated by an average
end-user. This literature study tries to identify the difference between lab emissions tests
and real-life operation emissions. Beside this, an overview of emission reduction techniques
is given, also based on real-life emissions tests. Missing information in current literature is
identified and important aspects surrounding current knowledge are summarised.
Altogether, this results in policy recommendations to ensure that the current knowledge
gap related to residential wood combustion can be filled in the near future.
Scope and definitions 3
2. Scope and definitions
This report will consider different types of stoves and wood combustion technologies, so it
is important to define a uniform typology of the different systems available on the market.
Based on common literature several different divisions can be made, for example based on
the technology behind the stove (EEA 2016) or based on the typology used by the federal
public service (FOD, Health n.d.).
In this report, the typology is based on the EMEP/EEA air pollutant emission inventory
guidebook to make international comparison possible.
2.1. Stove typology
Open fireplaces
These are not considered.
Partly-closed fireplaces
These are not considered.
Closed fireplaces
Closed fireplaces are equipped with front doors and have air flow control systems, which
include the distribution of combustion air to primary routes (grate) and secondary routes
(panels), as well as a system to discharge the exhaust gases. The retention time of the
gases in the combustion zone is longer compared with open fireplaces. They are
prefabricated and installed as stand-alone units or as a fireplace inserts installed in existing
masonry fireplaces (EEA 2016).
Conventional, radiating stoves
Stoves are enclosed appliances in which hand supplied fuels are combusted to provide
useful heat, which is transmitted to the surroundings by either radiation or convection.
Convection stoves work through heat storing and accumulation. Radiating stoves can be
fired with wood and both down-burning and up-burning methods are used. These
appliances typically have poorly organised combustion process resulting in low efficiency
(40% to 50%) (EEA 2016).
Masonry stoves
Masonry stoves are always a combination of bricks and/or stones and fireproof materials
such as ceramic. Due to the large thermal capacity of masonry materials they keep a room
warm for many hours (8-12) or days (1-2) after the fire has burnt out. Their combustion
chamber can be equipped with horizontal strips or inclined, perpendicular baffles made of
steel or fireproof material, which improve combustion quality and efficiency. Because of
the increased residence time of fuels in the combustion zone there is a decrease in pollutant
emissions compared to conventional radiating stoves. Their combustion efficiency ranges
from 60% to 80% (EEA 2016).
High-efficiency conventional stoves
High-efficiency conventional stoves essentially cover traditional stoves with improved

utilization of secondary air in the combustion chamber.
As a sub-category of high-efficiency stoves, it is possible to equip stoves with a catalytic

converter in order to reduce emissions caused by incomplete combustion, this is
particularly the case for wood fuel based stoves. The catalytic converter (a cellular or
honeycomb, ceramic substrate monolith covered with a very thin layer of platinum,
rhodium, or combination of the two) is usually placed inside the flue gas channel beyond
the main combustion chamber. The catalyst efficiency in emission reduction depends on
catalyst material, its construction – active surface area, and the conditions of flue gas flow
inside the converter. Due to more complete oxidation of the fuels, energy efficiency also
increases. However the catalyst will need frequent cleaning in order to maintain its
performance (EEA 2016).
Advanced combustion stoves

Advanced combustion stoves are characterized by multiple air inlets and pre-heating of
secondary combustion air by heat exchange with hot flue gases. This design results in
increased efficiency (near 70% at full load) in comparison with the conventional stoves
(EEA 2016).
Modern pellet stoves

Modern pellet stoves are a type of advanced stove using an automatic feed for pelletized
fuels such as wood pellets, which are distributed to the combustion chamber by a fuel
feeder from small fuel storage. Modern pellets stoves are often equipped with active control
system for supply of the combustion air. They reach high combustion efficiencies by
providing the proper air/fuel mixture ratio in the combustion chamber at all times (CITEPA,
2003). As a result, they are characterised by high efficiency (between 80% and 90%) (EEA
2016).
Conventional boilers
In general, boilers are devices which heat water for indirect heating. They are mainly
intended for generation of heat for the central heating system (including hot air systems)
or hot water, or a combination of both. Solid fuel conventional boilers include both over-
fire boilers and under-fire boilers, a differentiation based on the organisation of their
combustion process. Over-fire and under-fire boilers use all types of solid fuels except
pellets and wood chips (EEA 2016).
Advanced boilers
Two types of installations are included in this category of wood combustion devices:
advanced, under-fire boilers and downdraught wood boilers. The former is similar to
conventional under-fire boilers, but in this case flow of primary and secondary air are
controlled by a fan. Downdraught wood boilers have two chambers: first one where fuel is
fed for partial devolatilisation and combustion of the fuel layer and a second one where the
released combustible gases are burnt. In downdraught boilers, combustion air and flue
gases are controlled with a fan. Some devices use lambda control probes to measure flue
gas oxygen concentration and have precise combustion air control and staged-air
combustion (EEA 2016).
Wood/pellet boilers
Automatic log-fired boilers are available, although most small boilers are wood pellet or
chip-fired. These devices have a fully automatic system for feeding of pellet or woodchip
fuels and for supply of combustion air, which is distributed into primary (beneath the grate)
and secondary (into the gas oxidation zone) air supplies. The burners can have different
design such as underfeed burners, horizontally fed burners and overfed burners (EEA
2016).
2.2. Emissions reduction technologies
In some of the described types of wood combustion devices, different emission reduction
strategies are already implemented. For example, in some of the high-efficiency stoves
catalytic converters are implemented. Another possible strategy to reduce emissions is an
automatic regulation of the air influx into the combustion chamber, based on the
temperature and oxygen content of the flue gas, so that the combustion process can be
optimised.
In general, two different approaches for emissions reduction, each based on a different
assumption, can be identified: source control and end-of-pipe solutions. Both strategies
will be taken into account and the focus will be on the aspects described below.
Source control
Reducing residential wood combustion pollutants via source control implies avoiding or
diminishing pollutant formation during the combustion process. To achieve this, it is of
upmost importance to improve the control over the combustion process. Therefore, most
innovative stoves are equipped with the necessary sensors to obtain information about
temperature and oxygen content. Using this information, air intake can be automatically
controlled to ensure an optimised combustion.
Another strategy for source control is trying to reduce end-user errors. This can be
achieved by for example incorporating smart systems, aimed at giving information about
how much wood should be added to the fire at what time, in order to ensure that the stove
will continue operation under optimal combustion conditions. Some systems even stop
functioning when the wood is too humid, in order to prevent suboptimal operation.
End-of-Pipe solutions
In end-of-pipe solutions for pollutant emission reduction, pollutants that are formed during
the combustion process are targeted and treated before they reach the atmosphere. In
this report, the following technologies are selected and studied as end-of-pipe strategies
(Obernberger and Mandl 2011)
• Electrostatic Precipitators (ESPs)
• Catalytic converters
• Filters
2.3. Testing conditions for measurements of emissions
In this report the focus lies on the emissions of stoves under real-life operation, but often
other testing procedures are used. Therefore, it is necessary to exactly define which
elements are included and which are not for the different possible procedures. In the
continuation of this report, the different testing conditions are categorised as in the
definitions below.
The first type of tests are measurements conducted in a laboratory using the methodology
and procedure described in the standard EN 13240. These tests are designated as
laboratory experiments.
Pseudo real-life conditions are the test circumstances in which measurements are
executed in a laboratory, but the methodology and procedure is based on the typical
behaviour of an end-user. For pseudo real-life conditions no testing specifications are
defined, making comparison between different experiments difficult.
Real-life conditions are measurements conducted in the field. This implies that the
experiments are executed with a stove present inside a residential building and based on
the end-users’ normal procedure of lighting, refilling and ending a heating cycle. Real-life
condition tests are the most realistic ones, but are clearly dependent on the specific
conditions which can vary largely between different use cases. This results in difficulties
with comparisons of different real-life condition experiments. The execution of these test
can be done either by the end-user themselves or by an external researcher.
2.4. Studied emissions
Particulate matter
Particulate matter (PM) is a diverse mixture of small solid particles or liquid droplets. PM
exists in different sizes and can be grouped accordingly into PM 10, PM2.5, PM1 and UFP (ultra
fine particulates). The particles have an aerodynamic diameter smaller than 10 µm, smaller
than 2.5 µm, smaller than 1 µm and smaller than 0.1 µm, respectively. Chemically, PM is
a combination of organic and inorganic particles. In the case of wood combustion, organic
carbon and soot are the two most prevalent components of PM. Another categorisation of
PM can be made according to its formation process. Primary PM is the PM that is formed
immediately during the combustion. Secondary organic aerosols on the other hand, are
formed via condensation or photochemical oxidation of organic gaseous compounds
emitted in the air. In scientific literature, particulate matter is often measured as PM 2.5 or
as TSP. TSP is defined as total suspended particles, which comprises all particles, i.e. all
sizes and all compositions.
In general, PM is associated with detrimental health effects such as decreased lung

functioning, infections of the respiratory tracks and asthma. The current scientific view is
that the smaller a particle is, the more harmful it is due to the fact that smaller particles
can migrate deeper in the lungs. In Flanders, the VMM estimates that 71% of the loss of
healthy life years caused by environmental pollution can be attributed to PM10 and PM2.5. .
Besides the effects on human health, PM is also known to have an effect on to climate
change.
The main focus of the studied emissions lies on particulate matter because of its important
contribution to the total emission of PM in Flanders (VMM 2016).
CO
Carbon monoxide is an odourless and colourless gas that is formed due to incomplete
combustion, when oxygen is insufficiently present. When CO is present in higher
concentrations, it limits the amount of oxygen transported in the human body. This can
possibly result in dizziness, over unconsciousness to eventually death. Important sources
of outdoor carbon monoxide are devices that burn fossil fuels, industry and wood
combustion devices.
OGC
Organic gaseous carbon (OGC) is a collective term for all sorts of substances that exist in
ambient air as gas or as vapour. OGC originating from wood combustion include a variety
of pollutants going from benzene and formaldehyde to oxygenated organic compounds as
ketones and phenols. OGC are known to negatively affect our health as carcinogenic
compounds, but furthermore they also have a great impact on our environment. In the
presence of nitrogen oxides (NOx), OGC can result in the formation of tropospheric ozone.
OGC can also act as precursors for the formation of secondary organic aerosols.
PAH
Polycyclic aromatic hydrocarbons (PAHs) are the pollutants with condensed benzene cores
that can exist bound to particles or in the gas phase. The more volatile PAHs are a part of
what is understood as organic gaseous carbon. PAHs are mentioned separately in this
report for two reasons. First, residential wood combustion is a major source of PAH
emissions and secondly, PAHs are known to be harmful for human health. PAHs exist
naturally in coal, tar and oil, but are also formed due to incomplete combustion of organic
matter and fossil fuels. The most recent data suggest that wood combustion makes up
57% of total benzo(a)pyrene emissions in Flanders
(https://www.vmm.be/lucht/infografieken/infografiek-houtverbranding.jpg). In this
report, where possible special attention will be given to benzo(a)pyrene (B(a)P) as an
indicator of PAHs in general. This substance is well-studied and known to be carcinogenic
and therefore emphasised in regulation.
NOx
Nitrogen oxides comprise both NO and NO2. NOx are formed by reaction between nitrogen
and oxygen at high temperatures. This can happen naturally, for example by volcanic
activity or lightning, but also anthropogenically during combustion processes. In the latter
case, three different NOx formation mechanisms exist. Thermal NOx and prompt NOx
formation result from the presence of nitrogen in the air, while fuel NO x is formed from
nitrogen that is present in the fuels. In case of residential wood combustion, only fuel NO x
is formed due to the fact that combustion temperatures are not high enough to result in
other NOx-formation mechanisms. NOx are environmentally important pollutants since they
contribute to tropospheric ozone formation, acidification and photochemical smog.
Additionally, exposure to NOx can result in respiratory tract irritation.
3. Information of emissions per type
Categorisation of the investigated wood combustion devices into a certain type as

described in Paragraph 2.1 is difficult due to the minimal description of the tested devices
in scientific literature. As a consequence, it is impossible to focus the literature review
specifically on devices available on the Belgian market. To overcome this, the scope of the
examined literature is broadened to European studies, in the assumption that they report
about devices present on the European market and that these appliances are similar to the
ones present on the Belgian market. Important to remark is that Belgian legislation on
residential wood combustion devices is currently more strict than most other European
countries. As a consequence, it is possible that some of the devices listed below are not
allowed on the Belgian market.
The examined scientific papers are grouped below in tables depending on the type of wood
combustion device. In the first column the author and year of the publication are reported.
Columns two, three and four specify the circumstances in which the research has taken
place: test conditions, combustion air supply and fuel type. The next columns present the
reported emission values for PM, CO, OGC/PAH and NOx.
In general, not all studies investigated emissions of all pollutants considered in this report,
which makes comparison difficult. Furthermore, not all types of wood combustion are
examined to a similar extent. The more recent a certain technology is, e.g. advanced
combustion stoves, the less this technology is studied in scientific papers. In this report,
this results in tables of varying length.
Information of emissions per type 9
Closed fireplaces
Closed fireplaces have not been extensively studied in Europe, as can be deduced from
Table 3-1. The only study that investigated the emissions from closed fireplaces was
executed by Ozgen et al. Compared to the emissions reported in the EMEP ‘Air pollutant
emission inventory guidebook 2016’, Ozgen et al. reported lower PM and B(a)P emissions
and higher NOx and CO emissions (Table 3-10) (Ozgen et al. 2014). It has to be noted that
they tried to mimic real life operation, which could be an explanation for the variation in
emission factors per pollutant and the deviation from the EMEP values. Additionally, the
presented emission factors are averaged over the combustion cycle, but temporally higher
peak values can be expected due to fuel feeding, which does not happen automatically in
the tested closed fireplace.
Conventional wood stoves
Emission factors of conventional radiating stoves are studied by different authors. In

general, one can conclude from Figure 3-1 and Table 3-2 that wide variations exist in the
emission factors of the different pollutants that are listed. When the collected data are
compared to the values reported in the EMEP ‘Air pollutant emission inventory guidebook
2016’, it can be concluded that the emission factors reported in the EMEP guidebook fall in
the intervals of emission factors collected in this report (Table 3-10). The main reason for
the diverging emission factors is the different testing procedures and conditions used in
different studies. Other variables that can explain the deviation between different emission
factors, like fuel type or type of wood, do not lead to clear conclusions regarding their
effect on emission factors. One conclusion however can be postulated: air starvation in the
combustion chamber leads to increased emission of the different pollutants with a factor
two.
Figure 3-1: Overview of PM emissions of closed fireplaces and conventional wood stoves. (AIR)
with the reference indicates air starved experiments.
Table 3-1: Overview of closed fireplace emissions of PM, CO, OGC/PAH and NO x .
Author & Test Combustion
Fuel type Operation PM CO OGC / PAH NOx
year conditions air supply
EMEP/EEA 800 mg TSP/MJ 4000 mg/MJ 0.121 mg B(a)P/MJ 50 mg/MJ

(Ozgen et al. PSEUDO Manual HW + SW logs Normal 152-219 mg PM/MJ 3949-5030 mg/MJ 0.014 mg B(a)P/MJ 105-140 mg/MJ
2014) REAL LIFE operation
Table 3-2: Overview of reported conventional wood stove emissions of PM, CO, OGC/PAH and NO x .
(Schmidl et LAB Manual Softwood Normal 150.9 mg PM10/Nm³ 1331 mg/Nm³

al. 2011) operation briquettes
(Schmidl et LAB Manual Beech logs Normal 111.4 mg PM10/Nm³ 2779 mg/m³
al. 2011) operation
(Schmidl et LAB Manual Oak logs Normal 107.3 mg PM10/Nm³ 2948 mg/Nm³
al. 2011) operation
(Schmidl et LAB Manual Spruce logs Normal 156.6 mg PM10/Nm³ 2240 mg/Nm³
al. 2011) operation
(Schmidl et LAB Manual Spruce logs Air-starved 488.7 mg PM10/Nm³ 3971 mg/Nm³
al. 2011) operation
(Schmidl et LAB Manual Spruce logs High fuel load 178.1 mg PM10/Nm³ 2370 mg/Nm³
al. 2011) operation
(Calvo et al. LAB Manual Eucalyptus logs Normal 286-559 mg PM10/MJ 3175-3862 mg/MJ
2014) operation
(Calvo et al. LAB Manual Pine logs Normal 132-394 mg PM10/MJ 2726-2915 mg/MJ
2014) operation
(Calvo et al. LAB Manual Mixture HW logs Normal 118-955 mg PM2.5/MJ 2833-5744 mg/MJ
2015) operation
(Pettersson LAB Natural draft HW + SW logs Normal 38-170 mg TSP/MJ 1100-3400 mg/MJ 1.3-9.8 mg PAH/MJ 35-66 mg/MJ
et al. 2011)
(Pettersson LAB Natural draft HW + SW logs Air-starved 260-350 mg TSP/MJ 7100-7200 mg/MJ 45-220 mg PAH/MJ 35-38 mg/MJ
et al. 2011)
(Eriksson et LAB Natural draft Birch logs Normal 40 mg TSP/MJ 520-5600 ppm 0.1 mg PAH/MJ
al. 2014)

(Eriksson et LAB Natural draft Birch logs Air-starved 80 mg TSP/MJ 920-19000 ppm 0.7 mg PAH/MJ
al. 2014)
(Evtyugina LAB Manual Mixture HW logs Normal 3389-5778 mg/MJ
et al. 2014) operation
(Avagyan et LAB Natural draft HW + SW logs Normal 0.0325 mg PAH/MJ
al. 2016)
(Avagyan et LAB Natural draft HW + SW logs High fuel load 0.218 mg PAH/MJ
al. 2016)
(Fine, Cass, REAL LIFE Manual HW logs Normal 48-189 mg PM2.5/MJ
and Simoneit operation
2004)
(Fine, Cass, REAL LIFE Manual SW logs Normal 61-111 mg PM2.5/MJ
2004)
(Kelz et al. PSEUDO Manual Logwood Normal 55.5-74.2 mg PM1/MJ 2086-2355 mg/MJ 185.7-223.9 mg OGC/MJ
2010) REAL LIFE operation 4.561-8.786 mg PAH/MJ
(Alves et al. LAB Manal SW logs Normal 906 mg PM2.5/MJ
2011) operation
(Alves et al. LAB Manual HW logs Normal 344-839 mg PM2.5/MJ
2011) operation
(Alves et al. LAB Manual Briquettes Normal 233 mg PM2.5/MJ
2011) operation
(Ozgen et al. PSEUDO Manual HW + SW logs Normal 140-225 mg PM/MJ 6059-11131 0.122 mg B(a)P/MJ 91-110 mg/MJ
2014) REAL LIFE operation mg/MJ
(Win and LAB Unknown SW pellets High power, 43-55 mg PM2.5/MJ 16-48 mg/MJ 60-63 mg/MJ
Persson medium power &
2014) low power
Masonry stoves
In Table 3-3, the emission factors for masonry stoves from recent literature are reported.
When the collected data are compared to the emission factors from EMEP’s ‘Air pollution
emission inventory guidebook 2016’, it can be observed that the EMEP values all lie
between minimum and maximum of the collected emission factors, except for NO x (see
also in Table 3-10). The NOx emission factor in the EMEP guidebook is lower than the range
of NOx emissions reported in scientific literature. Mutually comparing the results of the
different studies is difficult since the different authors all investigated different parameters
that explain variation in emission factors. Tissari et al. focused on different operation
behaviour, Nuutinen et al. investigated different sizes and ages of masonry stoves and Fine
et al. looked for the effect of type of wood. Tissari et al. concluded that smouldering
conditions (lack of air) result in a drastic increase in PM, OGC and CO emissions compared
to normal operation (Tissari et al. 2008). Emission factors of these pollutants are up to ten
times higher under smouldering conditions. The comparison between different sizes of
masonry stoves learns that the effect of size depends on the age of the device: in older
masonry stoves, increase in size resulted in higher emissions, while this effect is not
noticed with modern stoves (Nuutinen et al. 2014). At last for type of wood, i.e. softwood
vs hardwood, no clear results are obtained by Fine et al. with respect to the lowest emission
factors.
Comparing the reported emission factors from conventional, radiating stoves with masonry
stoves learns that PM and CO emissions for both types of wood combustion devices are of
a similar order of magnitude. Reported NOx emissions from masonry stoves are higher than
those from conventional radiating stoves.
Figure 3-2: Overview of PM emissions of masonry stoves.

Table 3-3: Overview of masonry stoves emission factors for PM, CO, NO x and OGC/PAH at different testing conditions.
EMAP/EEA 800 mg TSP/MJ 4000 mg/MJ 0.121 mg B(a)P/MJ 50 mg/MJ

(Tissari et al. LAB Manual Birch wood Optimal 72-128 mg PM1/MJ 2111-2556 mg/MJ 100-144 mg OGC/MJ 72-83 mg/MJ
2008) operation
(Tissari et al. LAB Manual Birch wood Smouldering 400-833 mg PM1/MJ 5833-10 611 mg/MJ 940-2500 mg OGC/MJ 72 mg/MJ
2008) operation
(Tissari et al. LAB Manual Birch wood Normal 16-62 mg PM1/MJ 1137-1308 mg/MJ 24-276 mg OGC/MJ
2008) operation
(Fine, Cass, REAL LIFE HW logs 150-317 mg PM2.5/MJ 1.44 mg PAH/MJ
and Simoneit
2001)
(Fine, Cass, REAL LIFE SW logs 206-633 mg PM2.5/MJ 6.44 mg PAH/MJ
and Simoneit
2001)
and Simoneit
2002)
and Simoneit
2002)
and Simoneit
2004)
and Simoneit
2004)
(Kelz et al. PSEUDO Manual Logwood 28.0 – 31.3 mg PM1/MJ 1008-1207 mg/MJ 52.4-69.2 mg OGC/MJ
(Nuutinen et LAB Conventional, SW logs Normal 56.3 – 69.1 mg/MJ 1872 mg/MJ 332 mg OGC/MJ
al. 2014) small
(Nuutinen et LAB Modern, small SW logs Normal 40.2 – 44.5 mg/MJ 703 mg/MJ 96 mg OGC/MJ
al. 2014)
(Nuutinen et LAB Conventional, SW logs Normal 94.1 - 106 mg/MJ 3747 mg/MJ Out of Range mg OGC/MJ
al. 2014) medium
(Nuutinen et LAB Modern, SW logs Normal 44.5 – 44.8 mg/MJ 703 mg/MJ 83 mg OGC/MJ
al. 2014) medium
High-efficiency conventional stoves
Emission factors of high efficiency conventional stoves are studied by different authors and
are presented in Figure 3-3 and Table 3-4. Comparison between different studies is difficult
given the fact two different units for the emissions are used and data is lacking for
conversion between the two units. Secondly the different studies aimed at investigating
different parameters, which results in different testing conditions. For example, Boman et
al. examined the effect of fuel load and could conclude that emissions from low fuel loads
are higher than high fuel loads (Boman et al. 2011). Furthermore, Schmidl et al. found
that air starved operation of high efficiency stoves results in higher emissions of the
different pollutants compared to normal operation and high fuel load operation (Schmidl et
al. 2011). When the emission factor ranges are compared to those in EMEP’s ‘Air pollution
emission inventory guidebook 2016’, no clear relation is found. PM emission factors in EMEP
are higher compared to those reported in Table 3-4, for NOx the reverse is true. This can
easily be observed in Table 3-10.
At last, emission factors from high efficiency stoves are compared to those of conventional
stoves. High efficiency stoves perform better than conventional stoves when PM (and CO)
are considered. On the other hand, NOx (and PAH) emissions from high efficiency stoves
sometimes appear to be higher than in conventional radiating stoves.
Advanced combustion stoves are not often scientifically investigated, as can be deduced
from Table 3-5. One reason might be that these appliances are the most recent type of
wood combustion devices, so only in most recent studies these appliances can be
examined. PM and CO emissions reported by Tissari et al. are both lower than the emissions
in EMEP’s ‘Air pollutant emission inventory guidebook 2016’, which can also be seen in
Table 3-10 (Tissari et al. 2008).
When PM and CO emissions of advanced combustion stoves are compared to those
emissions from high efficiency stoves, it can be observed that in general emissions of
advanced combustion stoves are lower.
Figure 3-3: Overview of PM emissions of high-efficiency conventional stoves and advanced

combustion stoves.
Table 3-4: Overview of emission factors for PM, CO, OGC, PAH and NO x of high-efficiency stoves

(Schmidl et LAB Manual Softwood Normal 115.5-131.3 mg PM10/Nm³ 1491 - 3074 mg/Nm³
al. 2011) operation briquettes +
logwood
(Schmidl et LAB Manual Spruce logs Normal 128.3 mg PM10/Nm³ 2161 mg/Nm³
al. 2011) operation
(Schmidl et LAB Manual Spruce logs Air starved 146.7 mg PM10/Nm³ 2841 mg/Nm³
al. 2011) operation
(Schmidl et LAB Manual Spruce logs High load 78.9 mg PM10/Nm³ 1989 mg/Nm³
al. 2011) operation
(Boman et LAB Air staged SW logs High load 15-17 mg PMtot/MJ 100-160 mg/MJ 3.3- 41 mg PAH /MJ
al. 2011) combustion
(Boman et LAB Air staged SW logs Low load 22-43 mg PMtot/MJ 310 – 770 mg/MJ 90 – 340 mg PAH/MJ
(Gonçalves LAB Manual HW logs Normal 114 mg PM10/MJ 1.65 mg PAH/MJ
(Gonçalves LAB Manual SW logs Normal 62 mg PM10/MJ 9.04 mg PAH/MJ
(Fine, Cass, REAL LIFE Manual HW logs Catalyst 122 mg PM2.5/MJ 0.000319 mg PAH/MJ
2004)
(Fine, Cass, REAL LIFE Manual SW logs Catalyst 67 mg PM2.5/MJ 0.000652 mg PAH/MJ
2004)
(Ozgen et al. PSEUDO Manual SW + HW logs Normal 120-176 mg PMtot/MJ 4885-7829 mg/MJ 0.152 mg B(a)P/MJ 99-182 mg/MJ
2014) REAL LIFE operation
(Kelz et al. PSEUDO Manual Wood log 46.1-47.2 mg PM1/MJ 1036-1048 mg/MJ 94.2-95.5 mg OGC/MJ
Table 3-5: Overview of emission factors for PM, CO, OGC, PAH and NO x of advanced combustion stoves
(Tissari et al. LAB Unique grate Birch wood Normal 9.7-68.05 mg PM1/MJ 731-824 mg/MJ 18.2-26.3 mg OGC/MJ
2008)
Pollutant emissions from different modern pellet stoves are shown in Figure 3-4 and listed in
Table 3-6. Important to remark is that not all authors expressed the pollutant emissions in
the same unit, which makes comparison of emissions difficult. From the collected data that
can be compared with emission factors in EMEP’s ‘Air pollution emission inventory guidebook
2016’, it can be concluded that all reported EMEP emission factors lie in the range of the
collected emission factors (see also Table 3-10). The comparison between pseudo real-life
data and lab data does not lead to clear relations from which general extrapolation can be
conducted. A conclusion that can be made however is that also for modern pellet stoves air
starvation conditions result in highest emissions of all pollutants compared to optimal
conditions (Eriksson et al. 2014), just like with high efficiency stoves and masonry stoves.
Conventional boiler
In Figure 3-4 and Table 3-7 the emission factors for conventional boilers are presented. As
can be deduced from the table, not much recent literature on this subject is present. Kelz’
simulated real life operation emission of PM was lower compared to the emission factor in
EMEP’s ‘Air pollutant emission inventory guidebook 2016’, but the opposite is true for CO and
PAH emissions, as can be observed in Table 3-10 (Kelz et al. 2010). No general relations
between lab data and pseudo real life data can be determined, since for each ‘category’ only
one set of data is present. The comparison between conventional boilers and conventional
wood stoves learns that PM and PAH emissions of conventional boilers are lower, and CO
emissions of boilers are higher than those of conventional wood stoves.
Figure 3-4: Overview of PM emissions of modern pellet stoves and conventional boilers.
Table 3-6: Overview of PM, CO, NO x and OGC/PAH emissions of modern pellet stoves

(Schmidl et LAB Manual SW pellets Following EU 9.9 mg PM10/Nm³ 15.17 mg/Nm³ 124.67
al. 2011) operation standard mg/Nm³
(Eriksson et LAB Not specified SW pellets Optimal 20-850 ppm <0.1 mg PAH/MJ
(Eriksson et LAB Not specified SW pellets Air starved 730 – 16000 ppm 0.5 mg PAH/MJ
al. 2014)
(Kistler LAB Wood pellets 16-31 mg PM10/MJ
2012)
(Orasche et LAB Automatic Spruce pellets Normal 22 mg PMtot/MJ 413 mg/MJ 10 mg OGC/MJ 165 mg/MJ
al. 2012) control 0.000077 mg B(a)P/MJ
(Ozgen et al. PSEUDO Automatic Pellets Normal 75-139 mg PMtot/MJ 73-108 mg/MJ 0.0015 mg B(a)P/MJ 32-90 mg/MJ
2014) REAL LIFE control
(Vicente et LAB Pellets 26.6 – 102 mg/MJ
al. 2015)
(Win, PSEUDO Unknown Pellets 80 % of nominal 31 mg PM/MJ 192 mg/MJ 61 mg/MJ
Persson, and REAL LIFE power (12 kW)
Bales 2012)
Table 3-7: Overview of PM, CO, NO x and OGC/PAH emissions of conventional boilers.


(Kelz et al. PSEUDO Manual Logwood Real life 98.6-106.1 mg 8969-12 632 mg/MJ 650.8-1143.8 mg
2010) REAL LIFE operation simulation PM1/MJ OGC/MJ
3.39-18.85 mg PAH/MJ
Advanced boiler
Collected PM, CO, OGC/PAH and NOx emission factors of advanced boilers are presented in
Figure 3-5 and Table 3-8. In general, it can be concluded that the PM and CO emission factors
in EMEP’s ‘Air pollutant emission inventory guidebook 2016’ are higher than the collected PM
and CO emissions data, while EMEP’s emission factors for PAH and NOx are in the collected
interval of emission factors (see also in Table 3-10). In general, the only clear conclusion that
can be made from comparing the pseudo real life experiments with the laboratory
experiments is that PAH emissions from lab tests are lower than those from pseudo real life
tests. Furthermore, Kelz et al. investigated the effect of fuel type on pollutant emissions for
an advanced boiler operated under simulated real life conditions. From these experiments it
can be concluded that the use of pellets results in lower emissions than wood chips and log
wood (Kelz et al. 2010). Lamberg’s experiments on effect of fuel load did not lead to very
sharp differences between the tested fuel loads, although it could be stated that lower fuel
loads result in slightly higher emission factors (Lamberg et al. 2011).
At last, the comparison between advanced boilers and advanced combustion stoves learns
that generally PM and CO emission factors from advanced boilers are the lowest.
Figure 3-5: Overview of PM emissions of advanced boilers.

Table 3-8: Overview of PM, CO, OGC/PAH and NO x emission factors of advanced boilers.

(Kelz et al. PSEUDO Manual Pellets Real life 6.0-6.2 mg PM1/MJ 45.4-47.1 mg/MJ 1.7-2.5 mg OGC/MJ
2010) REAL LIFE operation simulation 0.006-0.014 mg PAH/MJ
(Kelz et al. PSEUDO Manual Wood chips Real life 13.6-15.3 mg PM1/MJ 168.1-182.2 mg/MJ 3.0-5.4 mg OGC/MJ
(Kelz et al. PSEUDO Manual Logwood Real life 14.2-17.6 mg PM1/MJ 700.4-793.1 mg/MJ 62.4-78.7 mg OGC/MJ
(Orasche et LAB Automatic SW logs Normal 17 mg PMtot/MJ 27 mg/MJ 2 mg OCG/MJ 143 mg/MJ
(Orasche et LAB Automatic HW logs Normal 14 mg PMtot/MJ 15 mg/MJ 1 mg OCG/MJ 157 mg/MJ
(Lamberg et LAB Automatic Pellets Normal 9.7-45.8 mg PM1/MJ 94.8-455 mg/MJ 0.76-4.66 mg OGC/MJ 50.2-168 mg/MJ
al. 2013) control
(Lamberg et LAB Automatic Pellets Normal 11.2-13.2 mg PM1/MJ 7-118 mg/MJ 77.2-84.8 mg/MJ
al. 2011) control
(Lamberg et LAB Automatic Pellets Medium load 14.2-16.4 mg PM1/MJ 19-245 mg/MJ 69.6-82.4 mg/MJ
al. 2011) control
(Lamberg et LAB Automatic Pellets Low load 15.4-17.2 mg PM1/MJ 66-280 mg/MJ 72.2-85.8 mg/MJ
al. 2011) control
Wood/pellet boilers
Emission factors of PM, CO, PAH/OGC and NOx from wood/pellet boilers are shown in Figure
3-6 and summarised in Table 3-9. When one compares the emission factors of the different
listed pollutants with those from EMEP’s ‘Air pollutant emission inventory guidebook 2016’,
one can observe that EMEP’s emission factor of PM, CO and NOx lie in the interval of collected
emission factor data, while for PAH, the collected emission factors are lower than the one in
the EMEP guidebook (see also Table 3-10). As a general trend, one could state that PM, CO
and NOx emissions from the pseudo real life experiments are higher than those from
laboratory experiments. A more precise and generally valid correlation between emission
factors of both types of experiments could not be found.
The comparison between emission factors of modern pellet stoves and wood/pellet boilers
learns that emission factors of both types of devices are in a similar order of magnitude.
Figure 3-6: Overview of PM emissions of wood/pellet boilers.

Table 3-9: Overview of pellet boiler emissions of PM, CO, OGC/PAH and NO x emissions.

(Schmidl et LAB Automatic air SW pellets Following EU standard 33.6 mg PM10/Nm³ 190 mg/Nm³
al. 2011) control
(Ozgen et al. PSEUDO Automatic Pellets Normal operation 30-103 PMtot/MJ 350 mg/MJ 0.00006 mg B(a)P/MJ 71 mg/MJ
2014) REAL LIFE control
(Win, PSEUDO Unknown Pellets 80 % of nominal power 84 mg PM2.5/MJ 547 mg/MJ 64 mg/MJ
Persson, and REAL LIFE (20 kW)
Bales 2012)
Bales 2012)
Bales 2012)
Bales 2012)
Bales 2012)
(Win and LAB Unknown SW pellets High power, medium 51-62 mg PM2.5/MJ 54-485 mg/MJ 62-67 mg/MJ
Persson power & low power
2014)
2014)
2014)
(Orasche et LAB Automatic SW pellets Normal operation 11 mg PMtot/MJ 17 mg/MJ 2 mg OGC/MJ 81 mg/MJ
(Orasche et LAB Automatic SW chips Normal operation 31 mg PMtot/MJ 75 mg/MJ 1 mg OGC/MJ 127 mg/MJ
Overview
In Figure 3-7, an overview is given of the number of references found for every type based
on the different measurement conditions. Based on these results it becomes clear that
most data are lab based with only a small fraction being measurements in real-life
conditions.
Wood/pellet boilers Lab

Pseudo real-life
Advanced boiler Real-life
Conventional boiler
High-efficiency conventional…
Masonry stoves
Conventional wood stoves
Closed fireplaces
0 2 4 6 8 10 12
Number of relevant studies
Figure 3-7: An overview of the number of references found per emission measurement
condition and per type of appliance.
A summary of the data mentioned above is given in Figure 3-8 containing a boxplot of all
available data points using the unit mg/MJ. From this overview, it is clear that newer types
of wood combustion appliances generally have lower emissions. This information is also
confirmed in Table 3-10 which contains the EMEP’s ‘Air pollutant emission inventory
guidebook 2016’ values and the minimum and maximum values as found from Table 3-1
to Table 3-9. It is important to note that there are exceptions to the general finding and
that sometimes higher emissions are reported for newer stoves types than for certain older
types. This is probably due to the operating conditions in the test which, as will be shown
further in this report, can have a large impact on the emissions.
Figure 3-8: Boxplot summary of all data points (number given between brackets) for each type
of wood combustion appliance. *category containing outliers above 1000 mg/MJ.
Table 3-10: Overview of emissions data collected from all different types of wood appliances
with unit mg/MJ.
PM (mg/MJ) CO (mg/MJ) PAH (mg/MJ) NOx (mg/MJ)
EEA (TSP) Literature EEA Literature EEA (B(a)P) Literature EEA (TSP) Literature
Closed
800 152-219 4000 3949-5030 0.121 0.014 50 105-140
fireplace
Conventional
800 38 - 955 4000 1100-7200 0.121 0.0325-220 50 35 - 66
wood stove
Masonry
800 16 - 833 4000 703-10611 0.121 0.081-14.10 50 72 - 83
stove
High
efficiency 400 15 - 176 4000 100-7829 0.121 0.0003-340 80 99-182
stove
Advanced
100 9.7-68.05 2000 731-824 0.010 NO DATA 95 NO DATA
stove
Modern
62 16 – 139 300 73-413 0.010 0.000077-0.5 80 32-165
pellet stove
Conventional
500 98.6-106.1 4000 8969-12 632 0.121 3.39-18.85 80 NO DATA
boiler
Advanced
100 6.0-45.8 2000 7-793.1 0.010 0.00012-0.105 95 50.2-168
boiler
Wood/pellet 0.00003-
62 11-116 300 12-547 0.010 80 59-127
boiler 0.00015
4. Strategies for emission reduction
Two strategies are possible to reduce emissions namely source control and end-of-pipe
solutions. Most end-of-pipe measures for residential wood combustion are based on
industrially available techniques. In those applications high removal efficiencies of
emissions can be obtained (VITO 2017). In order to reach these removal efficiencies, it is
important that process conditions remain constant and are constantly followed up. In
residential wood combustion these aspects are much harder to achieve, since they are
often based on the quality of the wood, the installation of the chimney and the stove,
maintenance and the end user. As a consequence it is important to check the real emission
reduction reached with these techniques in comparison with the theoretically achievable
reduction.
4.1. Source control
In this report the effects of source control measures are included in the typology of the
different stoves. Overall the following technologies are integrated inside the newest
appliances.
• Air supply strategies: For an optimal combustion, air supply is crucial. Both
excessive air and a shortage of air can create an inefficient combustion with a lot
of emissions. An important aspect to improve emission characteristics of wood
combustion devices is based on the concept of air staging. It means that the air
supply is split into two flows; one directly into the primary combustion chamber and
the second one above the primary combustion zone. In air staging, the first flow is
kept low so that the fuel just continues to burn while the second flow is present to
completely oxidise the gaseous compounds (T. Brunner, Obernberger, and Scharler
2009).
• Fuel feeding concepts: Using an automatic system for feeding the fuel to the
combustion chamber will ensure the correct load of wood. These system are
especially useful and convenient when pellets are the fuel. Using such a system,
the correct amount of fuel can be delivered and overloading can be eliminated. In
wood log combustion devices, fuel feeding can be optimised through design of the
combustion chamber or by indicating the user when to add fuel to the fire (T.
Brunner, Obernberger, and Scharler 2009).
• Combustion chamber and air supply design: Designing the combustion chamber
proofs to be a key element in reducing pollutant emissions. Innovation lies in the
geometries as well as the air injection used inside the combustion chamber.
Simulation with computational fluid dynamics can help significantly during the
development phase to eliminate death zones and optimise the combustion (T.
Brunner, Obernberger, and Scharler 2009).
• Process control concepts: A smart system is focused on ensuring that the right
temperature and air flow is present in the combustion chamber. This can be
achieved by automatically regulating the air intake to optimise the burning process
(Illerup et al. 2014).
Strategies for emission reduction 25
4.2. Electrostatic Precipitator
An electrostatic precipitator (ESP) is a typical instrument to remove all sizes of particulate

matter as long as the pollutant can be electrically charged. Organic compounds, NOx and
CO are typically not removed, which is why literature often does not mention them.
Mechanism
In ESPs a high voltage is applied between two electrodes, a discharge electrode and a
grounded electrode. Due to this high voltage, ionisation of the gas and mainly particulate
pollutants present in the gas occurs. As a consequence of the electrical field between the
electrodes, charged pollutants are attracted to the grounded electrode and deposit onto it.
As a result, particulate pollutants are removed from the gas stream.
Lab performance
In the IEA report of Task 32, 12 different ESP systems available on the market were
considered and tested, most of the time under lab conditions (Obernberger and Mandl
2011). The focus of ESP lies in the reduction of TSP, but in some cases PM1 is also taken
into consideration.
Two different concepts are present in the mentioned IEA report. The first one are ESP
models that can be used as a retrofit on top of the chimney and the others are placed
between stove and chimney to ensure an early removal of TSP. Both concepts have their
merits, but during design several aspects need to be considered, e.g. flue gas temperature
or robustness.
In Table 4-1 the available removal efficiencies for TSP are given based on the value
measured under laboratory conditions. If available, the value for removal efficiencies of
PM1 are also presented. When real-life tests were conducted, these data were also added
(columns containing label real-life). Important to take into account is that the tests
conducted in these studies are performed by different laboratories. This results in different
protocols that are used and data that is presented in different fashions. The values
presented here are a compilation of these data. Table 4-1 clearly shows that the removal
efficiency of ESPs ranges from 11 to 99%. This wide range is due to several issues which
should be taken into account:
• Type and age of stove

• Temperature of the flue gasses
• Fuel
• Ignition procedure
• Maintenance of the ESP
Table 4-1: Overview of removal efficiency based on the IEA report of Task 32 . Most tests are
conducted under laboratory conditions. Where available, real-life tests were included.
Device Test TSP Real-life TSP Test PM1 Real-life PM1

Type A No data 97.50%
Type B 11-17% 41% 53%
Type C 26-94%
Type D 60-80%
Type E 68-78% 71-83%
Type F 69%
Type G >70% 68% 60%
Type H >70%
Type I 82%
Type J 85-99% 54-61%
Type K 87%
Type L <93%
It is also shown that in two of the three cases the real-life values are slightly below the
results measured under laboratory conditions. In one case, the real-life test was
significantly better than the laboratory test result, but in that case the measurements were
conducted in two different labs, using different appliances and test methods. This clearly
indicates difficulties in comparing emission reduction techniques (Obernberger and Mandl
2011).
Around the same time the Danish Ministry ordered a study to compare 5 different ESP
technologies. These were all tested under laboratory conditions and in real-life. In both
operation conditions an old stove (sold between 1990 and 2007), an eco-labelled stove
(building year unknown, still on the market in 2012-2013) and a boiler were used. The
results showed a reduction in PM2.5 emissions during the test in the lab but removal
efficiencies were rather low. Using the eco-labelled stove, which is a more recent appliance,
the tests resulted in similar or lower emissions than when ESP systems were introduced
on older appliances. Additionally, implementation was challenging due to several
operational constraints and side effects (Schleicher et al. 2011).
Real-life performance
More recent a 2-year study under real-life conditions was executed by Brunner et al. using
a specific ESP system (Oekotube). This system was selected because of its potential to be
used in older biomass burning appliances. The year of manufacture of the used wood
combustion appliances were 2010 (logwood boiler), 1997 (logwood boiler) and 2009
(logwood stove) (Thomas Brunner, Wuercher, and Obernberger 2017).
The results of this study are shown in Table 4-2. While reduction efficiency is rather high,
the ESP is not able to reduce the effective concentration of TSP below the European eco-
design legislation on a permanent base.
Table 4-2: Achieved removal efficiencies for TSP and PM 1 and effective TSP concentration in
the flue gas for the different tested wood combustion devices (Thomas Brunner, Wuercher,
and Obernberger 2017).
2010 boiler 1997 boiler 2009 stove

TSP removal efficiency [%] 30-93 35-83 57-93
PM1 removal efficiency [%] 55-96 44-93 50-97
Concentration TSP after ESP [mg/Nm³] 4-174 22-154 14-46
Risks and disadvantages
The data above show that under the right conditions ESPs can realise significant reductions
in PM emissions. Nonetheless, several risks and issues are described in literature related
with this technology. These are listed below (Obernberger and Mandl 2011; Schleicher et
al. 2011):
• Sparkover effects: one of the risks of ESP is sparkover in the reactor. Sparkover can
occur due to the higher temperature present in the ESP where electrons are moving
directly between electrode and collector. When this occurs frequently, a reduction in
power or a shutdown of the ESP is needed and the effectiveness of the system
decreases.
• Noise: typically associated with sparkover is a certain sound, which can be annoying.
• Temperature effect on the electrode: in some cases the electrode can deteriorate under
influence of the higher temperature of the flue gas and thereby thus significantly reduce
the removal efficiency.
• TSP deposition on the collector electrode: by collecting the TSP on the electrodes the
effect of the charging and attraction towards the collector is reduced, resulting in lower
efficiencies. Another possible side effect is the re-entrainment of the particles in the
air, especially with agglomerated particles this is possible due to their lower resistivity.
• Price: most technologies available on the market are relative expensive, especially
compared with the overall cost of a wood combustion device.
• Resistivity and toxicity: particles going through an ESP should have the right resistivity.
In cases where this is too high or too low, the attraction of the particle on the collector
electrode decreases and low efficiencies are achieved.
A possible solution to overcome some of the described drawbacks, is the introduction of

novel technologies based on ESP such as plasma catalysis. In such technologies the
collector electrode is treated with a photocatalytic coating making degradation of the
deposited pollutants on the electrode possible. Another effect is that the plasma results in
radical formation in the reactor. These radicals are capable of reducing the concentration
of OGC and NOx (Van Wesenbeeck, Hauchecorne, and Lenaerts 2017).
4.3. Catalytic converter
The use and integration of catalytic converters in wood stoves in Europe is rather limited.
In recent years, most new developments of European wood stove manufactures focused
on the optimisation of the primary combustion process (Reichert, Schmidl, et al. 2017). In
Northern America the focus lies more on the active integration of catalytic converters in
wood stoves (15% of the approved stoves contains a catalytic converter) (US EPA 2015).
Most available data is thus related towards that market (US EPA 2015; Kaivosoja et al.
2012).
Mechanism
The principle behind a catalytic converter is that it decreases the temperature needed to
oxidize pollutants to achieve complete mineralisation. To achieve this, a catalyst, often a
noble metal like platina or palladium, is coated onto a honeycomb structure which can be
ceramic or metallic. This is placed inside the combustion chamber or inside the chimney.
Especially for volatile organic compounds, CO and NOx this method has already proven it
effectiveness in other applications, e.g. as the three-way-catalyst in cars or in large scale
facilities (Hukkanen et al. 2012).
Performance
A limited amount of literature is available with regards to reduction efficiencies of exhaust

emissions by catalytic converters.
The IEA report of 2011 describes two commercially available systems. The first system
reports efficiencies up to 85% for reduction of TSP according to the manufacturer, but no
independent tests are available. Another commercial catalytic converter claims to achieve
a reduction of 35% for TSP, up to 75% for hydrocarbons, up to 82% for CO and up to 95%
for soot. The Graz University of Technology performed real-life tests with these appliances
and obtained the following results: 0-14% reduction of CO, 0-15% reduction of
hydrocarbons and increased PM1 emissions due to poor combustion behaviour of the stove
(Obernberger and Mandl 2011).
A similar conclusion about catalytic converters is found by Kaivosoja et al. where a limited
reduction is achieved for CO (around 25%), hydrocarbons and PAH while highly toxic
PCDD/F by-products are formed (Kaivosoja et al. 2012).
More recent literature nonetheless shows significant improvement in the pollutant removal
potential of catalytic converters. This can be attributed to an increased knowledge of the
catalyst itself but also to an improvement in stove design with integrated catalytic
converters. For example, the research of Reichert and co-workers reported that
conversions of 95% for CO, 60% for OGC and 30% for PM could be achieved using a
catalytic converter. Beside these high conversion rates, quick responses were observed
due to the elevated temperature at which the catalyst was operated. Important in this
research was the type of catalyst used. The metallic honeycomb has a higher conversion
rate than the ceramic one, which demonstrates the importance of the type of catalyst used.
The experiments were performed under pseudo real-life conditions that are highly relevant
for real-life operation (Reichert et al. 2018).
Similar conclusions are reported by Wöhler and co-workers. Their experiments were
conducted under pseudo real-life conditions using an advanced stove built in 2016 with air
staging and separate control of primary and secondary air. The stove was originally
equipped with a ceramic filter but for the test it was replaced with a ceramic catalytic filter
and a dummy to keep the same operational conditions during the experiments. The test
conducted with the ceramic filter shows limited reductions, while the use of the metallic
honeycomb structure gives similar results as reported by Reichert et al. when tested under
nominal load. During start-up phase, lower reductions are achieved, illustrating the
importance to focus on this phase in experiments (Wöhler et al. 2017).
All tests mentioned above are conducted in short time periods with fresh catalysts. In order
to study the stability and reliability of the catalyst, longer tests should be performed. As
we know, catalysts regularly need to be re-activated by burning off all deposits. The optimal
window of operation and the long term performance of the catalyst require further study
(Reichert, Schmidl, et al. 2017).
Risks and disadvantages

Similar to ESP, a positive effect regarding pollutant emission reduction by using a catalyst
can be observed. Nonetheless, several risks are found when using a catalytic converter:
• During start-up and end phase the temperature inside the catalyst can be too low
to obtain an optimal oxidation (Ozil et al. 2009; Carnö, Berg, and Järås 1996;
Hukkanen et al. 2012).
• Due to poisoning, thermal deactivation and fouling deactivation of the catalyst can
occur (Ozil et al. 2009; Carnö, Berg, and Järås 1996).
• Information about the impact on particle reduction is limited (Hukkanen et al.

2012).
• Creation of highly toxic compounds such as PCDD/F (Kaivosoja et al. 2012) could
be problematic and needs to be further investigated.
4.4. Ceramic filters
Integrating a purely ceramic filter without catalytic activity is rarely found in literature.
When present, the reduction rates were rather low or non-existent, illustrating the limited
potential of this technique (Obernberger and Mandl 2011; Wöhler et al. 2017). The reason
for the limited testing is related to the fact that in most cases such a filter is directly
incorporated in the design of the wood combustion device. By removing it from the stove,
combustion is altered, which makes comparison very difficult. However, pollutant
emissions are very similar to comparable stoves without a ceramic filter (Obernberger and
Mandl 2011).
5. Impact of emissions on the atmospheric concentration
Knowledge about emissions associated with residential wood combustion with or without
emission reduction technologies is important, but the ultimate goal should be the
determination of their effect on the overall air quality. Besides pollutants that are formed
during the combustion, also secondary pollutants can be formed from precursors emitted
by residential wood combustion.
5.1. Secondary Organic Aerosol Formation
Current Belgian legislation focuses on CO and PM concentrations measured at the exit of a

chimney. In European eco-design legislation organic gaseous carbon (OGC) and NOx
concentrations measured at the exit of the chimney are also included. Secondary pollution
formation in the atmosphere is not taken into account. Recent literature shows nonetheless
that there is large potential for the formation of secondary organic aerosol (SOA) (Denier
Van Der Gon et al. 2015; Vicente and Alves 2018).
The origin of these SOA lies in the incomplete combustion and the formation of condensable
organic compounds (COC) and volatile organic compounds (VOC). In Figure 5-1 an
overview by Nussbaumer on the formation of organic aerosols and other products is
presented (Nussbaumer 2017). The COC and VOC fraction and their potential to form SOA
are important topics for further research (Vicente and Alves 2018; Keller and Burtscher
2017; Bruns et al. 2016). The lack of in depth knowledge on the faith of this fraction is
also an important object of study giving its effects on the environment and human health
(Vicente and Alves 2018).
Figure 5-1: Overview of the burning process of wood combustion focused on PM10. Adapted
after (Nussbaumer 2017).
Impact of emissions on the atmospheric concentration 31
In one of the studies by Keller and co-workers a micro smog chamber (MSC) is used to
simulate SOA formation. He describes a lab experiment with several different small wood
stoves and pellet stoves, as well as real-life tests. The real-life tests were executed with
medium sized boilers for heating warm water at nominal power during the winter of
2011/2012. The results of this study show that when using small combustion installations
fired with log wood, concentrations of 53 mg/m³ SOA potential could be observed during
the start phase and 8.8 mg/m³ during stable operation. For pellets, these values were 42
mg/m³ and 1.6 mg/m³, respectively. Especially in the start-up phase the observed
concentrations are at the level of current PM10 legislation thresholds (Keller and Burtscher
2017).
Precursors
The importance of SOAs in the overall exhaust emission confirms the need for further
investigation. Because this subject has proven to be very complex, it might be better to
focus on the precursors that are responsible for the formation of SOA (Keller and Burtscher
2017). Research by Grieshop has evidence that 85% of all SOA are related to low volatile
species (Grieshop et al. 2009). This is further specified by Bruns, who identified 22
precursor molecules for SOA formation, with phenol, naphthalene and benzene being the
three most important ones (Bruns et al. 2016).
Mitigation strategy
A well-known technology for removal of particles in flue gas cleaning is the use of ESP. As
described earlier, the removal efficiency for other pollutants than PM is rather low, including
the precursors of SOAs. A possible solution might be the removal of the least volatile OGC
molecules. This can be achieved by a gas cooling in front of the ESP in order to condensate
the least volatile OGCs into particles.
Keller and co-workers studied a system currently available on the market, under the name
NOSMOG, using a logwood stove equipped with a cool down system. Results showed that
the SOA potential dropped significantly by combining the ESP with a cooling system. This
illustrates the potential of flue gas condensing to remove SOA precursors and thus reduce
the SOA formation (Keller and Burtscher 2017).
5.2. Real-life implementations on a larger scale
Small estate
A study for the Danish Ministry of the Environment has been executed in a town called
Hillerød, in a small estate called Slåenbækken. In this estate, consisting of 20 houses, 7
wood stoves were present that were regularly used during winter period. Data was
collected both inside the estate as 27 km away in a background measurement station.
During the measurement campaign no clear impact on the outdoor concentrations could
be observed when wood combustion devices were equipped with end-of-pipe technology.
A reason for this could not be identified and further research is needed (Schleicher et al.
2011).
City scale
To investigate the impact on air pollution on a larger scale, two other studies have been
done on city scale. In the first study a change-out program was used, while in the second
case end-of-pipe solutions were implemented on wood combustion devices.
The change-out program was done in 2005-2007, so the appliances used are currently no
longer state of the art. This should be taken into account when interpreting the results.
The overall effect was that a reduction of 20 to 28% could be achieved by the changeout
at that time (Ward, Palmer, and Noonan 2010).
In a more recent study executed in Australia a catalytic converter as end-of-pipe

technology was implemented in existing wood combustion devices. A comparison between
the concentrations in ambient air was done for four towns that are similar in topographical
and population size and in the number of wood stoves. In one of them, people were asked
to use a commercially available catalyst free of charge inside their stove. Although 80% of
them accepted the offer, no significant reduction could be observed in the emissions of
PM2.5 in the atmosphere, while under laboratory conditions a reduction could be observed.
Possible explanations for the observations are meteorological reasons and changes in
background concentrations, but further investigation is still needed. This should include
the effect the community can have on the use and therefore effectiveness of the catalyst
(Johnston et al. 2016).
Knowledge gaps & future research needs 33
6. Knowledge gaps & future research needs
The literature review identified several important knowledge gaps. An overview is given
below. A brief overview of the research needs to address the knowledge gaps is described
as well.
6.1. Lack of data under real-life conditions
Based on the information provided in Chapter 4, it becomes clear that there is a lack of
data measured under real-life conditions, i.e. measured at the premises where wood
combustion appliances are used in-house. The vast majority of data has been collected
under standardised test conditions. Some studies demonstrate the effects of combustion
practices on emissions under controlled test conditions, e.g. by using various wood types.
Real-life conditions are rarely tested, and the methods used are insufficiently detailed in
literature. The use of different methods makes it difficult to find a correlation between data
from different types of measurements (Vicente and Alves 2018).
Besides this, it is also observed that most literature refers to older appliances. Indications
on the performance of modern stoves and boilers are scarce. A first characterisation of the
important elements of real-life testing is elaborated in Chapter 7.
6.2. Comparison of data is challenging
When studying the literature there is a clear lack of consistency in the use of units for
describing emissions of wood combustion. This creates an obstacle to objectively compare
data between different experiments and stoves.
In Table 6-1 an overview is presented of the different units found in literature. In general,
the units can be divided into two types: energy and air volume related units. The energy
related units express emissions per amount of released heat or per mass of wood
combusted. Emissions shown in air volume related units are expressed per volume of
combustion air, which depends on the characteristics of the stove. As a result, air volume
related units hamper comparison between different wood combustion devices.
Nonetheless, the unit mg/Nm³ is used in standard tests.
Table 6-1: Overview of units for emitted pollutant concentrations from wood combustion
devices used in scientific literature.
Energy related Air volume related

g/kgdry_wood mg/Nm³
g/GJ mg/Nm³ in STP 13% O2
g/u ppm
vol%
A realistic conversion between energy and air volume should be made to compare data on
a same unit base.
Skreiberg describes a methodology to convert between the different units of emissions. An
overview of these conversions is presented in Table 6-2. A number of conversions are
relative easy to make, but for the conversion of MJ to Nm³, knowledge about the volume
of flue gas is needed (Skreiberg 2002).
Table 6-2: Table containing conversion calculation based on (Skreiberg 2002).
To E1 mg/Nm³ E2 mg/Nm³ E3 mg/kg E4 mg/MJ

From at O2 at 13% O2
𝑉𝐹𝐺 𝑉𝐹𝐺
E1 mg/Nm³ at O2 E1 𝐸1. 𝐸1 . 𝑉𝐹𝐺 𝐸1 .
𝑉𝐹𝐺_13% 𝑈𝐻𝑉
𝑉𝐹𝐺_13% 𝑉𝐹𝐺_13%
E2 mg/Nm³ at 13% O2 𝐸2. E2 𝐸2 . 𝑉𝐹𝐺_13% 𝐸2 .
𝑉𝐹𝐺 𝑈𝐻𝑉
𝐸3 𝐸3 𝐸3
E3 mg/kg E3
𝑉𝐹𝐺 𝑉𝐹𝐺_13% 𝑈𝐻𝑉
𝑈𝐻𝑉 𝑈𝐻𝑉
E4 mg/MJ 𝐸4 . 𝐸2 . 𝐸4 . 𝑈𝐻𝑉 E4
𝑉𝐹𝐺 𝑉𝐹𝐺_13%
with
𝑉𝐹𝐺 = Nm³ dry flue gas per kg dry fuel
𝑉𝐹𝐺13% = Nm³ dry flue gas per kg dry fuel based on STP of 13% O2
𝑈𝐻𝑉 = Upper heating value = the amount of energy you have per kg dry fuel (MJ/kg)
To convert mg/Nm³ to mg/MJ the type of fuel should be known to determine the UHV. The
heating values that are used range between 15.6 and 19.25 MJ/kg. In the EMEP/EEA air
pollutant emission inventory guidebook it is suggested to use 18 MJ/kg for wood logs and
19 MJ/kg for wood pellets in case no details are provided about the fuel. Additionally the
volume of dry flue gas of the used stove should be known. This last number cannot be
determined based on stoichiometric values because an excessive amount or air is typically
used. During measurements, it is therefore important to include this parameter in the
monitoring (Skreiberg 2002).
A special program, called FuelSim – Average, is developed that can be used to do the
conversion. With the program it is possible to follow the mass, volume and energy balance
of a continuous combustion process (Skreiberg 2002).
6.3. Effective impact on air quality
As shown in Chapter 5, the direct impact of emissions from residential wood combustion
on the air quality is not completely clear. One of the important aspects in this regard is the
lack of knowledge about the SOAs potential due to residential wood combustion.
Uncertainty remains about the exact formation pathway of SOAs, which precursors are
responsible for their formation and how high the ambient contribution can become. Further
studies are needed that are focussed on the formation of SOA.
Directly related to this problem is the limited knowledge on the impact of different emission
reducing strategies on local air quality. This type of studies proofs to be very difficult due
to several external factors like long range transport, topography, weather and atmospheric
stability. Comparison of different results is therefore difficult, making it hard to know the
effective potential of the different approaches to reduce primary emissions and the
formation of secondary organic aerosols.
Towards characterisation of real-life conditions 35
7. Towards characterisation of real-life conditions
The results of Chapter 3 confirm that a clear difference between real-life and lab emissions
exists. Therefore, while studying exhaust emissions from wood combustion appliances, five
different elements will influence the final results. These are the following:
• Technological factors
• Test-related factors
• Operating conditions – non-user behaviour
• User behaviour
• Fuel
Technological factors
Chapter 3 illustrates the impact of technology on exhaust emission levels. First, a clear
difference between traditional wood log stoves and pellets stoves is observed.
Furthermore, the level of technology implemented with the stove is of upmost importance.
The implementation of technologies like air staging, optimising the dimensioning of the
combustion chamber or improving the air tightness of the combustion chamber significantly
reduces exhaust emissions (Reichert et al. 2016) and also influences the difference
between lab and real-life conditions.
From the earlier results, it can be concluded that more recent appliances perform better
regarding pollutant emissions, but around 80% of appliances in Belgium is older than 2010
(Agoria-CIV 2017). As a result, the majority of the wood combustion devices currently
installed in Belgium is not equipped with state-of-the-art technology. A potential solution
might be the retrofit of emissions abatement technologies under strictly controlled
conditions (Obernberger and Mandl 2011).
Test-related factors
As described, three different types of tests are performed on stoves. In current legislation
the standard lab test employs optimal steady-state situations resulting in limited variability
of optimal burning performance. In practice, testing needs to be performed in more realistic
cycles, similar to the more realistic testing cycles for determination of exhaust emission by
cars. Of course, real-life testing with realistic conditions is the best solution (Reichert et al.
2016).
Operating conditions – non-user behaviour
Operating conditions are typically divided into two distinct categories based on the
possibility of user interference: user behaviour and non-user, operating conditions. The
non-user operating conditions are related to flow conditions induced by natural draught
(Reichert et al. 2016).
Reichert and co-workers investigated the impact of draught. They observed that the effect
of draught depends on wood combustion device and that draught can influence emissions
in either positive or negative way. Around the typical draught under real-life conditions
(20-30 Pa), small differences were observed. Still, an effect on concentration is present
since draught affects combustion efficiency. Higher draught leads to lower efficiency and
thus a longer heating period that increases exhaust emissions (Reichert, Hartmann, et al.
2017).
User behaviour
User behaviour factors highly influence real life emissions, since they have a direct impact
on operating conditions. One of the most important variables is fuel properties, which is
described in the next section. Other influences are:
• Ignition method: in wood combustion there are typically two strategies to ignite
the fire: top-down or bottom-up. In the first case the firewood is placed below,
smaller material is placed above and the ignition aid is placed on the top. The fire
thus starts on top and needs to migrate downwards. In a bottom-up strategy, the
starting aid is placed on the bottom of the wood combustion device and firewood
is placed on top of it. Described by Hartmann in 2012 and Reichert in 2016, the
top-down method results in significantly lower emissions of PM, OGC and CO when
compared with the bottom-up method (Hartmann et al. 2012; Reichert et al.
2016). More recent reports were less conclusive in relation to the ignition method.
No clear results could be found using the different wood combustion devices
studied, making the impact of the ignition method less clear (Reichert, Hartmann,
et al. 2017).
• Air settings: another important aspect is the air-to-fuel ratio. While modern
appliances are mostly equipped with automatic air control, it is more common to
manually control the air flow. Fachinger and co-workers found that by using small
and very dry logwood with excessive air, emissions could significantly rise (up to
six-fold) due to reduced residence time (Fachinger et al. 2017). On the other hand,
a limited amount of air results in a smouldering phase, something that typically
occurs at the end phase of a burning cycle and shows significant increase in
emissions of all pollutants (Tissari et al. 2008; Schmidl et al. 2011; Reichert et al.
2016).
• Maintenance: it is important to have regular maintenance of the appliance to keep

it functioning as optimal as possible. Effects like soot deposition or air leakages will
have a direct negative impact on the emissions. Especially the impact of air leakage
by damaged door gaskets and air-tightness will have significant negative impact
on the emissions of TSP (Sevault et al. 2015).
Towards characterisation of real-life conditions 37
Fuel
One of the most important user behaviour factors affecting emissions is the fuel that is
used. From literature it is clear that the use of fuels other than solid wood, briquettes or
pellets has a significant negative influence on the emissions. This is especially the case if
treated wood with paint, paper or cardboard is used (Fachinger et al. 2017).
But even when only solid wood is considered as fuel, significant differences in emissions
using the same appliance can be observed. The impact of fuel type is one of the most
studied topics under real-life conditions. However, different studies are difficult to compare
due to the difference in measuring methods, stoves and settings used. In what follows, a
general overview of the important aspects is presented:
• Type of wood: an important factor in the emissions is the kind of wood used. In
most cases this is divided into two groups being hardwood and softwood. Different
studies lead to different conclusions. Both Calvo and Gonçalves have found that
softwood is better than hardwood (Calvo et al. 2014; Gonçalves et al. 2010).
Hartmann on the other hand showed that the use of hardwood is preferential and
recommended (Hartmann et al. 2012). The results in the overview in Chapter 3
confirm these mixed results.
• Moisture content: different studies compared pollutant emissions between high

(moisture content > 20%) and moderate moisture contents (between 10 and 20%).
From these studies it can be concluded that emissions of PM and PAHs are higher
when the moisture content of the fuel is higher (Shen et al. 2013; Hays et al. 2003;
Fernandes et al. 2011; Orasche et al. 2013; Mcdonald et al. 2000). One report
states that the optimal firewood moisture content ranges between 7.5 and 20%
(Hartmann 2012). Although not much research has focused on pollutant emissions
from firewood with moisture contents lower than 7.5%, one study shows that PAH
emissions are higher when very dry wood (moisture content = 1.6%) is used
instead of moderately dry wood (Orasche et al. 2013).
• Fuel refilling: refilling a wood combustion device during operation can result in
significant negative effects on pollutant emissions when this is badly timed. Surveys
with real-life users however learned that mostly refilling occurs at the right moment.
Most users recharge wood at the moment the flames are almost extinguished, the
moment that is actually the most optimal one to ensure limited emissions. A second
aspect related to fuel refilling is the amount of fuel that is added. Based on surveys
this is often not optimal and people tend to overload their stove, which results in
decreased burning efficiency and increased emissions (Reichert et al. 2016).
• Quality of the fuel: even when regulated pellets are used, significant differences in
emissions are observed. Venturini and co-workers investigated three different types
of pellets based on the ISO 17225-2 technical standard: pellets from the highest
class (A1), pellets from the lowest class (B) and pellets that do not comply with the
legislation. The results showed that lower quality pellets resulted in increased TSP
emissions two to five times (Venturini et al. 2018).
A big problem with all the described factors is that consistent information of typical or
average real user behaviour is lacking. As a result, assumptions currently made in scientific
literature on (pseudo) real life emissions can proof to be completely different in reality. A
better understanding of these aspects is required. This can be done by performing surveys
with real-life users in order to gain knowledge about their operation of a stove (Wöhler et
al. 2016; Reichert et al. 2016). Another problem is that a small fraction of real users (e.g.
people burning wet, contaminated wood in old stoves) could have a relative high
contribution to the total emissions. Therefor good knowledge of best, average and worst
case scenarios and their occurrence is essential to estimate the total real-life emissions.
Policy Advice 39
8. Policy Advice
After the findings of the literature review, the following policy recommendations are
proposed to better control the emissions from wood combustion under real-life conditions.
The recommendations are grouped under the following key messages:
• Better understand the use of wood stoves in practice, and the impact on regional
air quality
• Develop test protocols to monitor emissions under real-life conditions, while
ensuring data comparability
• Develop a governmental strategy for reducing and better controlling residential
wood smoke
• Online tool to guide potential buyers, users, retailers and vendors to the cleanest
stoves and good practices in using them
1. Better understand the use of wood stoves in practice, and the impact on
regional air quality
At the moment, there is no exact knowledge of the number and the types of stoves in
active operation. Questions that are to be answered include: ‘Is wood burning mostly
relevant in rural areas, or also in urban areas? How often are stoves used in practice?
Which burning practices and wood type do users have? To which extent are the emissions
measured under test conditions valid for an assessment of the cumulative impact of wood
burning on local and regional scale?’
A survey could be undertaken to approximate the number of households that use wood
burning. Statistics from vendors and/or retailers could potentially be of use to cross-check
the survey results. Additional data that can be collected in the survey are a breakdown on
the type of stoves used, burning practices and fuel type and quality. A broader survey can
also be undertaken to better understand the knowledge of citizens on the adverse effects
of wood burning to the environment and human health and on the nuisance stove users
create to neighbours.
2. Develop test protocols to monitor emissions under real-life conditions, while

ensuring data comparability
The lack of data on emissions under real-life operation, and the challenges to data
comparability are important barriers found in this study. To better understand the real-life
emissions from wood burning appliances, a set of test protocols can be developed to
systematically monitor real-life and pseudo real-life conditions, preferentially at European
level. A separate testing protocol is to be developed for wood log appliances and pellet
appliances. Under pseudo real-life conditions, the emissions from a range of burning
practices, including best, average and worst case scenarios, can be monitored under
controlled conditions. To enable the comparison of inherently heterogeneous burning
practices under real-life conditions, protocols need to be developed that consider the
various technologies of which a stove/boiler is composed, user behaviour (e.g. ignition
method, air settings, maintenance) and fuel types (e.g. wood species, moisture content of
wood, fuel amount and sequence of loading).
3. Develop a strategy for reducing and better controlling residential wood smoke
In addition to the general air quality strategy for Flanders (under development), a specific
strategy can be developed to reduce residential wood smoke. The strategy can hold actions
for the regional and local government, vendors, retailers and users. Specific actions need
to be tailored to households having older wood stoves compared to household owning more
recent stoves. The strategy can be composed of specific actions, for which a fact sheet can
be developed, explaining potential impacts, time schedules, transitional arrangements,
budgetary aspects etc.
Potential actions that can be included in a strategy to reduce residential wood smoke can
include hard/compulsory measures such as changeout campaign for older stoves/boilers,
compulsory maintenance of stoves/boilers, burn bans on days with bad air quality and
softer measures aiming to change behaviour, such as information campaigns on best
practices for users, vendors and retailers.
Compulsory actions can be useful for actions that users, retailors or vendors would not
accept without a binding government decision and/or legislation. A disadvantage of
compulsory action is that the setting up of compulsory actions could take more time and
may generate resistance from the general public. Voluntary actions and information actions
might engage stakeholders more efficiently. A combination of compulsory and voluntary
measures can also be considered.
A phase-out campaign of stoves/boilers could be a potential action. Older, low performing

stoves could be banned and/or be replaced by newer, cleaner devices. To decide which
stoves would need to be phased out, thresholds values can be setup, based on e.g. age or
real-life emissions. A phase-out campaign can speed up the retirement of older appliances.
Yet, substantial logistical effort is needed to firstly identify low performing stoves and
secondly achieve replacement or removal. In analogy with the energy certificate, house
owners could be obliged to obtain a certificate on their appliance when the property is sold.
Another strategy might be the obliged registration of each newly bought wood combustion
device, e.g. in the existing home pass. An advantage of registration is that step by step an
overview of the different devices in use can be set up. Beside the registration of new
devices, a registration process for all currently used devices could be considered, as is
already done in Montreal. Furthermore, obligatory maintenance of appliances can also be
considered, potentially associated with an assessment on the emissions.
4. Online tool to guide potential buyers, users, retailers and vendors to the
cleanest stoves and good practices in using them
Informative and educational material can be developed on the best wood burning practices
to reduce emissions, complementary to, or included in existing campaigns in Flanders, e.g.
the ‘burn wise’ campaign (www.stookslim.be). Actions could include:
• Develop an online tool and associated guidance where potential consumers are
guided to the cleanest wood stoves/boilers, including tips on how to use them.
• Introduce a voluntary label for vendors/retailers to advertise that their stoves are
clean (analogy with energy efficiency labelling)
• Avoid wood burning on days with bad air quality and/or unfavourable
meteorological conditions (e.g. fog)
Conclusion 41
9. Conclusion
The aim of this literature overview was to map real life emissions from residential wood
combustion devices. It appears that the amount of tests conducted under real life
conditions is rather limited. Differences between emissions under lab and real life
conditions exist and are determined by multiple parameters. Most important in this regard
are user behaviour and more specifically fuel characteristics.
Emission data from lab and pseudo real life tests indicate that significant variation exists
in emissions from different types of wood combustion devices. A general trend that is
observed, is that more recent devices emit less particulate matter compared to older
devices, indicating that source control as emission reduction strategy is effective. Another
approach for reducing emissions are end-of-pipe solutions. Different options exist, each
with varying removal efficiencies and issues. Installing end-of-pipe solutions in retrofit on
existing wood combustion devices, could be a possibility to reduce emissions, but operation
conditions have to be carefully controlled.
Next to the primary emissions, wood combustion devices are also responsible for the
formation of secondary organic aerosols. The precursors of these pollutants and the extent
of their emissions due to residential wood combustion and their exact impact on air quality
should be further investigated. Another issue that needs to be clarified is the unit for
expression of emissions from wood combustion devices. Currently, two different types of
units that are difficult to compare are used in scientific literature and legislation.
Four main policy recommendations are suggested from this literature overview. At first, a
better understanding of emissions from wood combustion devices operated in real life
should be obtained. Secondly, standard protocols for test procedures mimicking real life
operation should be developed so that objective and realistic comparison between different
wood combustion devices can be done. Furthermore, a national or regional strategy should
be determined for the reduction of pollutant emissions from residential wood combustion
and its impact on local air quality. At last, citizens should be informed about best practices
concerning wood combustion.
10. References
Agoria-CIV. 2017. “Feiten En Cijfers over Houtverbranding : Hernieuwbare Energie Voor

Huishoudelijke Verwarming.”
Alves, Célia, Cátia Gonçalves, Ana Patrícia Fernandes, Luís Tarelho, and Casimiro Pio. 2011.
“Fireplace and Woodstove Fine Particle Emissions from Combustion of Western
Mediterranean Wood Types.” Atmospheric Research 101 (3). Elsevier:692–700.
https://doi.org/10.1016/J.ATMOSRES.2011.04.015.
Avagyan, Rozanna, Robin Nyström, Robert Lindgren, Christoffer Boman, and Roger
Westerholm. 2016. “Particulate Hydroxy-PAH Emissions from a Residential Wood Log
Stove Using Different Fuels and Burning Conditions.” Atmospheric Environment 140
(September). Pergamon:1–9. https://doi.org/10.1016/J.ATMOSENV.2016.05.041.
Boman, Christoffer, Esbjörn Pettersson, Roger Westerholm, Dan Boström, and Anders
Nordin. 2011. “Stove Performance and Emission Characteristics in Residential Wood
Log and Pellet Combustion, Part 1: Pellet Stoves.” Energy & Fuels 25 (1). American
Chemical Society:307–14. https://doi.org/10.1021/ef100774x.
Brunner, T., I. Obernberger, and R. Scharler. 2009. “Primary Measures for Low-Emission
Residential Wood Combustion - Comparison of Old with Optimised Modern Systems.”
Proceedings of the 17th European Biomass Conference & Exhibition.
Brunner, Thomas, Gerald Wuercher, and Ingwald Obernberger. 2017. “2-Year Field
Operation Monitoring of Electrostatic Precipitators for Residential Wood Heating
Systems.” Biomass and Bioenergy, February. Pergamon.
https://doi.org/10.1016/J.BIOMBIOE.2017.01.025.
Bruns, Emily A, Imad El Haddad, Jay G Slowik, Dogushan Kilic, Felix Klein, Urs
Baltensperger, and André S H Prévôt. 2016. “Identification of Significant Precursor
Gases of Secondary Organic Aerosols from Residential Wood Combustion.” Scientific
Reports 6 (June). Nature Publishing Group:27881.
https://doi.org/10.1038/srep27881.
Calvo, A. I., V. Martins, T. Nunes, M. Duarte, R. Hillamo, K. Teinilä, V. Pont, et al. 2015.
“Residential Wood Combustion in Two Domestic Devices: Relationship of Different
Parameters throughout the Combustion Cycle.” Atmospheric Environment 116:72–82.
https://doi.org/10.1016/j.atmosenv.2015.06.012.
Calvo, A.I., L.A.C. Tarelho, C.A. Alves, M. Duarte, and T. Nunes. 2014. “Characterization
of Operating Conditions of Two Residential Wood Combustion Appliances.” Fuel
Processing Technology 126 (October). Elsevier:222–32.
https://doi.org/10.1016/J.FUPROC.2014.05.001.
Carnö, Johanna, Magnus Berg, and Sven Järås. 1996. “Catalytic Abatement of Emissions
from Small-Scale Combustion of Wood: A Comparison of the Catalytic Effect in Model
and Real Flue Gases.” Fuel 75 (8). Elsevier:959–65. https://doi.org/10.1016/0016-
2361(96)00047-6.
Carvalho, Ricardo Luis Teles de. 2016. “Wood-Burning Stoves Worldwide: Technology,
Innovation and Policy,” June. Aalborg Universitetsforlag.
https://doi.org/10.5278/VBN.PHD.ENGSCI.00122.
Denier Van Der Gon, H A C, R Bergström, C Fountoukis, C Johansson, S N Pandis, D
Simpson, and A J H Visschedijk. 2015. “Particulate Emissions from Residential Wood
Combustion in Europe – Revised Estimates and an Evaluation.” Atmos. Chem. Phys
15:6503–19. https://doi.org/10.5194/acp-15-6503-2015.
EEA. 2016. “EMEP/EEA Air Pollutant Emission Inventory Guidebook 2016. Railways,” no.
21:1–20. https://doi.org/10.2800/247535.
Eriksson, A. C., E. Z. Nordin, R. Nyström, E. Pettersson, E. Swietlicki, C. Bergvall, R.
Westerholm, C. Boman, and J. H. Pagels. 2014. “Particulate PAH Emissions from
Residential Biomass Combustion: Time-Resolved Analysis with Aerosol Mass
Spectrometry.” Environmental Science & Technology 48 (12). American Chemical
Society:7143–50. https://doi.org/10.1021/es500486j.
Evtyugina, Margarita, Célia Alves, Ana Calvo, Teresa Nunes, Luís Tarelho, Márcio Duarte,
Sónia O. Prozil, Dmitry V. Evtuguin, and Casimiro Pio. 2014. “VOC Emissions from
Residential Combustion of Southern and Mid-European Woods.” Atmospheric
References 43
Environment 83 (February). Pergamon:90–98.

https://doi.org/10.1016/J.ATMOSENV.2013.10.050.
Fachinger, Friederike, Frank Drewnick, Reto Gieré, and Stephan Borrmann. 2017. “How
the User Can Influence Particulate Emissions from Residential Wood and Pellet Stoves:
Emission Factors for Different Fuels and Burning Conditions.” Atmospheric
Environment 158 (June). Pergamon:216–26.
Fernandes, A. P., C. A. Alves, C. Gonçalves, L. Tarelho, C. Pio, C. Schimdl, and H. Bauer.
2011. “Emission Factors from Residential Combustion Appliances Burning Portuguese
Biomass Fuels.” Journal of Environmental Monitoring 13 (11):3196.
https://doi.org/10.1039/c1em10500k.
Fine, Philip M., Glen R. Cass, and Bernd R. T. Simoneit. 2001. “Chemical Characterization
of Fine Particle Emissions from Fireplace Combustion of Woods Grown in the
Northeastern United States.” American Chemical Society.
https://doi.org/10.1021/ES001466K.
Fine, Philip M., Glen R. Cass, and Bernd R. T.. Simoneit. 2004. “Chemical Characterization
of Fine Particle Emissions from the Fireplace Combustion of Wood Types Grown in the
Midwestern and Western United States.” Environmental Engineering Science 21
(3):387–409. https://doi.org/10.1089/109287504323067021.
Fine, Philip M., Glen R. Cass, and Bernd R. T . Simoneit. 2002. “Chemical Characterization
of Fine Particle Emissions from the Fireplace Combustion of Woods Grown in the
Southern United States.” American Chemical Society.
https://doi.org/10.1021/ES0108988.
FOD, Health, Food Chain safety and Enviroment. n.d. “Website Containing All Approved
Stoves on the Belfium Merket.” Accessed December 13, 2017.
https://apps.health.belgium.be/ordsm/02/f?p=500:6005:::NO:::
Gonçalves, Cátia, Célia Alves, Margarita Evtyugina, Fátima Mirante, Casimiro Pio,
Alexandre Caseiro, Christoph Schmidl, Heidi Bauer, and Fernando Carvalho. 2010.
“Characterisation of PM10 Emissions from Woodstove Combustion of Common Woods
Grown in Portugal.” Atmospheric Environment 44 (35). Pergamon:4474–80.
Grieshop, A. P., J. M. Logue, N. M. Donahue, and A. L. Robinson. 2009. “Laboratory
Investigation of Photochemical Oxidation of Organic Aerosol from Wood Fires 1:
Measurement and Simulation of Organic Aerosol Evolution.” Atmospheric Chemistry
and Physics 9 (4):1263–77. https://doi.org/10.5194/acp-9-1263-2009.
Hartmann, Hans. 2012. “Technologie-Und Förderzentrum Im Kompetenzzentrum Für
Nachwachsende Rohstoffe Status on Emissions, Regulation and Technical
Improvements and Future Developments for Residential Wood Burning Appliances in
Germany.”
Hartmann, Hans, Claudia Schön, Peter Turowski, Friedrich Biedermann, Thomas Brunner,
Linda Bäfver, John Finnan, and John Carroll. 2012. “Low Emission Operation Manual
for Chimney Stove Users.”
Hays, Michael D., N. Dean Smith, John Kinsey, Yuanji Dong, and Peter Kariher. 2003.
“Polycyclic Aromatic Hydrocarbon Size Distributions in Aerosols from Appliances of
Residential Wood Combustion as Determined by Direct Thermal Desorption - GC/MS.”
Journal of Aerosol Science 34 (8):1061–84. https://doi.org/10.1016/S0021-
8502(03)00080-6.
Hukkanen, A., T. Kaivosoja, O. Sippula, K. Nuutinen, J. Jokiniemi, and J. Tissari. 2012.
“Reduction of Gaseous and Particulate Emissions from Small-Scale Wood Combustion
with a Catalytic Combustor.” Atmospheric Environment 50 (April). Pergamon:16–23.
Illerup, J.B., J. Nickelsen, B.B. Hansen, W. Lin, and K. Dam-Johansen. 2014. “Intelligent
Heat System - High-Energy Efficient Wood Stoves with Low Emissions. Field Tests.”
European Biomass Conference and Exhibition Proceedings. ETA-Florence Renewable
Energies, 689–94. https://doi.org/10.5071/22ndeubce2014-2bv.2.33.
Johnston, Olivia, Fay Johnston, John Todd, and Grant Williamson. 2016. “Community-Wide
Distribution of a Catalytic Device to Reduce Winter Ambient Fine Particulate Matter
from Residential Wood Combustion: A Field Study.” Edited by Wenjun Ding. PLOS ONE
11 (11). Public Library of Science:e0166677.
https://doi.org/10.1371/journal.pone.0166677.
Kaivosoja, T., A. Virén, J. Tissari, J. Ruuskanen, J. Tarhanen, O. Sippula, and J. Jokiniemi.
2012. “Effects of a Catalytic Converter on PCDD/F, Chlorophenol and PAH Emissions
in Residential Wood Combustion.” Chemosphere 88 (3). Pergamon:278–85.
https://doi.org/10.1016/J.CHEMOSPHERE.2012.02.027.
Keller, Alejandro, and Heinz Burtscher. 2017. “Characterizing Particulate Emissions from
Wood Burning Appliances Including Secondary Organic Aerosol Formation Potential.”
Journal of Aerosol Science 114 (December). Pergamon:21–30.
https://doi.org/10.1016/J.JAEROSCI.2017.08.014.
Kelz, J., T. Brunner, I. Obenberger, P. Jalava, and M.-R. Hirvonen. 2010. “PM Emissions
from Old and Modern Biomass Combustion Systems and Their Health Effects.” In 18th
European Biomass Conference, 1231–43.
Kistler, Magdalena. 2012. “Particulate Matter and Odor Emission Factors from Small Scale
Biomass Combustion Units.”
Kocbach Bølling, Anette, Joakim Pagels, Karl Yttri, Lars Barregard, Gerd Sallsten, Per E
Schwarze, and Christoffer Boman. 2009. “Health Effects of Residential Wood Smoke
Particles: The Importance of Combustion Conditions and Physicochemical Particle
Properties.” Particle and Fibre Toxicology 6 (1). BioMed Central:29.
https://doi.org/10.1186/1743-8977-6-29.
Lamberg, Heikki, Olli Sippula, Jarkko Tissari, and Jorma Jokiniemi. 2011. “Effects of Air
Staging and Load on Fine-Particle and Gaseous Emissions from a Small-Scale Pellet
Boiler.” Energy & Fuels 25 (11). American Chemical Society:4952–60.
https://doi.org/10.1021/ef2010578.
Lamberg, Heikki, Jarkko Tissari, Jorma Jokiniemi, and Olli Sippula. 2013. “Fine Particle and
Gaseous Emissions from a Small-Scale Boiler Fueled by Pellets of Various Raw
Materials.” Energy & Fuels 27 (11). American Chemical Society:7044–53.
https://doi.org/10.1021/ef401267t.
Mcdonald, Jacob D., Barbara Zielinska, Eric M. Fujita, John C. Sagebiel, Judith C. Chow,
and John G. Watson. 2000. “Fine Particle and Gaseous Emission Rates from Residential
Wood Combustion.” Environmental Science and Technology 34 (11):2080–91.
https://doi.org/10.1021/es9909632.
Nussbaumer, Thomas. 2017. “Aerosols from Biomass Combustion.”
Nuutinen, K., J. Jokiniemi, O. Sippula, H. Lamberg, J. Sutinen, P. Horttanainen, and J.
Tissari. 2014. “Effect of Air Staging on Fine Particle, Dust and Gaseous Emissions from
Masonry Heaters.” Biomass and Bioenergy 67 (August). Pergamon:167–78.
Obernberger, Ingwald, and Christoph Mandl. 2011. “Survey on the Present State of Particle
Precipitation Devices for Residential Biomass Combustion with a Nominal Capacity up
to 50 kW in IEA Bioenergy Task32 Member Countries Final Version.” IEA Task 32:
Bioenergy Workshop, no. December:1–105.
Orasche, Jürgen, Jürgen Schnelle-Kreis, Claudia Schön, Hans Hartmann, Hans Ruppert,
Jose M. Arteaga-Salas, and Ralf Zimmermann. 2013. “Comparison of Emissions from
Wood Combustion. Part 2: Impact of Combustion Conditions on Emission Factors and
Characteristics of Particle-Bound Organic Species and Polycyclic Aromatic
Hydrocarbon (PAH)-Related Toxicological Potential.” Energy and Fuels 27 (3):1482–
91. https://doi.org/10.1021/ef301506h.
Orasche, Jürgen, Torben Seidel, Hans Hartmann, Jürgen Schnelle-Kreis, Judith C. Chow,
Hans Ruppert, and Ralf Zimmermann. 2012. “Comparison of Emissions from Wood
Combustion. Part 1: Emission Factors and Characteristics from Different Small-Scale
Residential Heating Appliances Considering Particulate Matter and Polycyclic Aromatic
Hydrocarbon (PAH)-Related Toxicological Potential of Particle-Bound Organic
Species.” Energy & Fuels 26 (11). American Chemical Society:6695–6704.
https://doi.org/10.1021/ef301295k.
Ozgen, Senem, Stefano Caserini, Silvia Galante, Michele Giugliano, Elisabetta Angelino,
Alessandro Marongiu, Francesca Hugony, Gabriele Migliavacca, and Carmen Morreale.
References 45
2014. “Emission Factors from Small Scale Appliances Burning Wood and Pellets.”
Atmospheric Environment 94 (September). Pergamon:144–53.
Ozil, Fabien, V. Tschamber, Frédéric Haas, and G. Trouvé. 2009. “Efficiency of Catalytic
Processes for the Reduction of CO and VOC Emissions from Wood Combustion in
Domestic Fireplaces.” Fuel Processing Technology 90 (9). Elsevier:1053–61.
Pettersson, Esbjörn, Christoffer Boman, Roger Westerholm, Dan Boström, and Anders
Nordin. 2011. “Stove Performance and Emission Characteristics in Residential Wood
Log and Pellet Combustion, Part 2: Wood Stove.” Energy & Fuels 25 (1). American
Chemical Society:315–23. https://doi.org/10.1021/ef1007787.
Poppel, Martine Van, Wim Aerts, Bart Bayens, Rob Brabers, Jan Peters, Maarten Spruyt,
and Jo Van Laer. 2017. “Inschatting van de Bijdrage van Houtverbranding Door
Burgers Aan Luchtverontreiniging in Vlaanderen.”
Reichert, G., H. Hartmann, W. Haslinger, H. Oehler, R. Mack, C. Schmidl, C. Schön, et al.
2017. “Effect of Draught Conditions and Ignition Technique on Combustion
Performance of Firewood Roomheaters.” Renewable Energy 105 (May).
Pergamon:547–60. https://doi.org/10.1016/J.RENENE.2016.12.017.
Reichert, G., C. Schmidl, W. Haslinger, M. Schwabl, W. Moser, S. Aigenbauer, M. Wöhler,
and C. Hochenauer. 2016. “Investigation of User Behavior and Assessment of Typical
Operation Mode for Different Types of Firewood Room Heating Appliances in Austria.”
Renewable Energy 93 (August). Pergamon:245–54.
https://doi.org/10.1016/J.RENENE.2016.01.092.
Reichert, G., C. Schmidl, W. Haslinger, H. Stressler, R. Sturmlechner, M. Schwabl, N.
Kienzl, and C. Hochenauer. 2017. “Long Term Durability and Safety Aspects of
Oxidizing Honeycomb Catalysts Integrated in Firewood Stoves.” Biomass and
Bioenergy 105 (October). Pergamon:428–42.
Reichert, G., C. Schmidl, W. Haslinger, H. Stressler, R. Sturmlechner, M. Schwabl, M.
Wöhler, and C. Hochenauer. 2018. “Catalytic Efficiency of Oxidizing Honeycomb
Catalysts Integrated in Firewood Stoves Evaluated by a Novel Measuring Methodology
under Real-Life Operating Conditions.” Renewable Energy 117 (March).
Pergamon:300–313. https://doi.org/10.1016/J.RENENE.2017.10.065.
Schleicher, Ole, Karsten Fuglsang, Peter Wåhlin, Helge Rørdam Olesen, Jacob Klenø
Nøjgaard, and Max Bjerrum. 2011. “Test of Technologies for Flue Gas Cleaning and
Combustion Improvement for Existing Residential Wood Burning Appliances.”
Schmidl, Christoph, Markus Luisser, Emmanuel Padouvas, Leopold Lasselsberger,
Magdalena Rzaca, Carlos Ramirez-Santa Cruz, Markus Handler, Ge Peng, Heidi Bauer,
and Hans Puxbaum. 2011. “Particulate and Gaseous Emissions from Manually and
Automatically Fired Small Scale Combustion Systems.” Atmospheric Environment 45
(39). Pergamon:7443–54. https://doi.org/10.1016/J.ATMOSENV.2011.05.006.
Sevault, Alexis, Morten Seljeskog, Line Rydsa, Asbjorn Ostnor, Birger Ronning, Magnus
Rishaug, and Norwegian Standard Ns. 2015. “Effect of Maintenance on Particulate
Emissions from Residential Woodstoves.”
Shen, Guofeng, Miao Xue, Siye Wei, Yuanchen Chen, Qiuyue Zhao, Bing Li, Haisuo Wu,
and Shu Tao. 2013. “Influence of Fuel Moisture, Charge Size, Feeding Rate and Air
Ventilation Conditions on the Emissions of PM, OC, EC, Parent PAHs, and Their
Derivatives from Residential Wood Combustion.” Journal of Environmental Sciences
(China) 25 (9):1808–16. https://doi.org/10.1016/S1001-0742(12)60258-7.
Skreiberg, Øyvind. 2002. “Fuelsim -Average v1.1: A Mass, Volume and Energy Balance
Spreadsheet for Continuous Combustion Applications.”
Tissari, J., J. Lyyränen, K. Hytönen, O. Sippula, U. Tapper, A. Frey, K. Saarnio, et al. 2008.
“Fine Particle and Gaseous Emissions from Normal and Smouldering Wood Combustion
in a Conventional Masonry Heater.” Atmospheric Environment 42 (34).
Pergamon:7862–73. https://doi.org/10.1016/J.ATMOSENV.2008.07.019.
US EPA. 2015. “Environmental Protection Agency 40 CFR Part 60 Standards of Performance
for New Residential Wood Heaters, New Residential Hydronic Heaters and Forced-Air
Furnaces; Final Rule.” Federal Register 80 (50).

Venturini, Elisa, Ivano Vassura, Francesca Agostini, Andrea Pizzi, Giuseppe Toscano, and
Fabrizio Passarini. 2018. “Effect of Fuel Quality Classes on the Emissions of a
Residential Wood Pellet Stove.” Fuel 211 (January). Elsevier:269–77.
https://doi.org/10.1016/J.FUEL.2017.09.017.
Vicente, E.D., and C.A. Alves. 2018. “An Overview of Particulate Emissions from Residential
Biomass Combustion.” Atmospheric Research 199 (January). Elsevier:159–85.
https://doi.org/10.1016/J.ATMOSRES.2017.08.027.
Vicente, E.D., M.A. Duarte, A.I. Calvo, T.F. Nunes, L. Tarelho, and C.A. Alves. 2015.
“Emission of Carbon Monoxide, Total Hydrocarbons and Particulate Matter during
Wood Combustion in a Stove Operating under Distinct Conditions.” Fuel Processing
Technology 131 (March). Elsevier:182–92.
VITO. 2017. “Home | EMIS.” 2017. https://emis.vito.be/nl.
Vlaamse Milieumaatschappij. 2016. “Lozingen in de Lucht 2000-2016,” 329.
VMM. 2016. “Luchtkwaliteit in Het Vlaamse Gewest. Jaarverslag Immissiemeetnetten -
2015.”
Ward, Tony J, Christopher P Palmer, and Curtis W Noonan. 2010. “Fine Particulate Matter
Source Apportionment Following a Large Woodstove Changeout Program in Libby,
Montana.” Journal of the Air & Waste Management AssociationOnline) Journal Journal
of the Air & Waste Management Association 606:1096–2247.
https://doi.org/10.3155/1047-3289.60.6.688doi.org/10.3155/1047-3289.60.6.688.
Wesenbeeck, K. Van, B. Hauchecorne, and S. Lenaerts. 2017. “Study of Positive and
Negative Plasma Catalytic Oxidation of Ethylene.” Environmental Technology 38 (12).
Taylor & Francis:1554–61. https://doi.org/10.1080/09593330.2016.1237553.
WHO. 2015. “Residential Heating with Wood and Coal: Health Impact and Policy Options
in Europe and North America Impact.”
Win, Kaung Myat, and Tomas Persson. 2014. “Emissions from Residential Wood Pellet
Boilers and Stove Characterized into Start-Up, Steady Operation, and Stop
Emissions.” Energy & Fuels 28 (4). American Chemical Society:2496–2505.
https://doi.org/10.1021/ef4016894.
Win, Kaung Myat, Tomas Persson, and Chris Bales. 2012. “Particles and Gaseous Emissions
from Realistic Operation of Residential Wood Pellet Heating Systems.” Atmospheric
Environment 59 (November). Pergamon:320–27.
Wöhler, Marius, Jes Sig Andersen, Gero Becker, Henrik Persson, Gabriel Reichert, Claudia
Schön, Christoph Schmidl, Dirk Jaeger, and Stefan K. Pelz. 2016. “Investigation of
Real Life Operation of Biomass Room Heating Appliances – Results of a European
Survey.” Applied Energy 169 (May). Elsevier:240–49.
https://doi.org/10.1016/J.APENERGY.2016.01.119.
Wöhler, Marius, Dirk Jaeger, Stefan K. Pelz, and Harald Thorwarth. 2017. “Potential of
Integrated Emissions Reduction Systems in a Firewood Stove under Real Life
Operation Conditions.” Energy & Fuels 31 (7). American Chemical Society:7562–71.
https://doi.org/10.1021/acs.energyfuels.7b00803.
Flanders Environment Agency
Dokter De Moorstraat 24-26
B-9300 Aalst
en.vmm.be

WoodCombustion_FINAL-REPORT

Uploaded by

Document Informationclick to expand document information

Document Informationclick to expand document information

Copyright:

Available Formats

WoodCombustion_FINAL-REPORT

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

WoodCombustion_FINAL-REPORT

Uploaded by

Copyright:

Available Formats

Literature review of emissions of modern wood

combustion devices and emissions reducing

Literature review of emissions of modern wood combustion

Tytgat Tom1, Walpot Griet1, Cools Jan2, Lenaerts Silvia1,2

An important problem with estimating real-life emissions is that consistent information of

in reality. A better understanding of these aspects is required. This can be done by

In eerste instantie werd een typologie gemaakt van de verschillende

Ondanks de talrijke beschikbare informatiebronnen in verband met huishoudelijke

Executive Summary ............................................................................................ II

Figure 3-3: Overview of PM emissions of high-efficiency conventional stoves and advanced

Figure 3-5: Overview of PM emissions of advanced boilers. ........................................18

The Flanders Environment Agency (Vlaamse Milieumaatschappij – VMM) recently reported

Open front 2000 50 120 200

2. Scope and definitions

2.1. Stove typology

These are not considered.

These are not considered.

Conventional, radiating stoves

High-efficiency conventional stoves

High-efficiency conventional stoves essentially cover traditional stoves with improved

As a sub-category of high-efficiency stoves, it is possible to equip stoves with a catalytic

Advanced combustion stoves

Modern pellet stoves

2.2. Emissions reduction technologies

2.3. Testing conditions for measurements of emissions

2.4. Studied emissions

In general, PM is associated with detrimental health effects such as decreased lung

3. Information of emissions per type

Categorisation of the investigated wood combustion devices into a certain type as

Conventional wood stoves

Emission factors of conventional radiating stoves are studied by different authors. In

EMEP/EEA 800 mg TSP/MJ 4000 mg/MJ 0.121 mg B(a)P/MJ 50 mg/MJ

EMEP/EEA 800 mg TSP/MJ 4000 mg/MJ 0.121 mg B(a)P/MJ 50 mg/MJ

(Schmidl et LAB Manual Softwood Normal 150.9 mg PM10/Nm³ 1331 mg/Nm³

Author & Test Combustion

Figure 3-2: Overview of PM emissions of masonry stoves.

EMAP/EEA 800 mg TSP/MJ 4000 mg/MJ 0.121 mg B(a)P/MJ 50 mg/MJ

High-efficiency conventional stoves

Advanced combustion stoves

Figure 3-3: Overview of PM emissions of high-efficiency conventional stoves and advanced

EMEP/EEA 400 mg TSP/MJ 4000 mg/MJ 0.121 mg B(a)P/MJ 80 mg/MJ

EMEP/EEA 100 mg TSP/MJ 2000 mg/MJ 0.010 mg B(a)P/MJ 95 mg/MJ

Modern pellet stoves

EMEP/EEA 62 mg TSP/MJ 300 mg/MJ 0.010 mg B(a)P/MJ 80 mg/MJ

Author & Test Combustion

EMEP/EEA 500 mg TSP/MJ 4000 mg/MJ 0.121 mg B(a)P/MJ 80 mg/MJ

Figure 3-5: Overview of PM emissions of advanced boilers.

EMEP/EEA 100 mg TSP/MJ 2000 mg/MJ 0.010 mg B(a)P/MJ 95 mg/MJ

Figure 3-6: Overview of PM emissions of wood/pellet boilers.

EMEP/EEA 62 mg TSP/MJ 300 mg/MJ 0.010 mg B(a)P/MJ 80 mg/MJ

Wood/pellet boilers Lab

PM (mg/MJ) CO (mg/MJ) PAH (mg/MJ) NOx (mg/MJ)

4. Strategies for emission reduction

4.1. Source control

4.2. Electrostatic Precipitator

An electrostatic precipitator (ESP) is a typical instrument to remove all sizes of particulate

• Type and age of stove

Device Test TSP Real-life TSP Test PM1 Real-life PM1