Literature review: report on indoor air quality associated with woodburning

Executive Summary

This paper reviews recent studies to assess the contribution from wood stoves to indoor particulate
matter (PM) in comparison with other common household activities. There are wide variations in
reported concentrations of PM, however the 24hr average concentrations reported are typically
below WHO recommended exposure values. Peak values of PM over short timescales are higher than
the 24hr average values, and are attributed to re-fuelling of stoves and removal of ash during cleaning.
These peak values can be mitigated if users follow manufacturer’s instructions/ best practice. Other
factors that could play a role in the PM concentrations are identified including stove construction,
climatic factors, air settings and the characteristics of the fuel burned, but these parameters are rarely
reported in the literature. No scientific evidence was found for adverse health impacts from exposure
to the indoor air typically associated with modern enclosed wood burning stoves, however data in the
context of developed world studies is extremely limited. The studies showed that other sources of
particulate matter in the home, in particular cooking derived emissions, can release much higher levels
of PM compared to enclosed wood stoves. There are limitations regarding the quantity of research
studies and consistency of the research methods used, with different analyser types reporting variable
concentrations. Standard test protocols and measurement methods for domestic indoor PM should
be devised for further research in this area. These should take into account that the length of exposure
may be different for the different cases, for instance a short cooking or ash removal activity compared
to longer stove operation activity. The rate of air changes and ventilation needs to also be defined in
a quantitative way.

1. Introduction

Exposure to high levels of particulate matter (PM) has been linked to a variety of serous health impacts
including lung cancer, Chronic obstructive pulmonary disease (COPD) and other cardiovascular (CV)
disease [Ali et al 2021]. Sources of exposure are most commonly via inhalation, but harmful
components of the particulate can also be absorbed via skin contact with contaminated surfaces and
by ingestion [Kristensen 2019]. There are many sources of particulates associated with normal
domestic activities such as cleaning, cooking, use of candles and use of fragranced personal care
products. Recent concern has been focussed on the degree to which particulate can be released from
wood burning stoves to the indoor air, and whether these concentrations are significant in terms of
health impact. This paper reviews recent studies into indoor air pollution to assess the contribution
from wood stoves in comparison with other common household activities. The scope of this paper is
focused towards developed world PM concentrations using enclosed wood burning stoves rather than
open fires in developing world areas, although an acknowledgement of issues in the wider world
context is made.

1.1 Particulate Matter

Domestic particulate matter can come from a wide range of sources such as cleaning products,
personal care products, cooking, incense, candles, human skin shedding as well as wood burning
stoves. These sources are discussed further in the sections below. PM can be formed in the air from

1
other emissions as Secondary Organic Aerosol (SOA) during complex atmospheric reactions such as
the oxidation of gas phase volatile organic compounds (VOC) The composition of particulate matter
varies according to source but can include condensed organic species, solid carbon ‘soot’ spherules
and solid inorganic material.

PM is usually measured according to its aerodynamic diameter. The definition of coarse particles or
‘PM10’ is particulate with average aerodynamic diameters below 10 µm, fine particles or ‘PM2.5’ are
diameters below 2.5µm and ultrafine particles (UFP) are <0.1µm (100nm). Figure 1 shows a
comparison to the size of a human hair [EPA, 2021]

Figure 1: Size comparisons for PM particles [EPA, USA- 2021]

When particles are first emitted from combustion sources, they are spherules of carbon typically 10-
50nm in diameter. These primary particles rapidly agglomerate into chains with organic species
condensed onto them, resulting in a larger measurable particle diameter. The solid carbon component
is either Black Carbon (BC) or Elemental Carbon (EC) depending on how the measurement is
conducted. Thermo-optical analysis is needed to determine BC whilst techniques such as
thermogravimetric analysis (TGA) will give EC. A wide variety of semi-liquid combustion products and
incompletely burned fuel fragments condense onto the solid carbon cores and contribute to the
measurable mass. These are Volatile Organic Compounds (VOC), Semi- Volatile Organic Compounds
(SVOC) and Intermediary Volatile Organic Compounds (IVOC), according to their volatility- i.e., how
readily they condense.

Another significant component of PM from combustion sources is inorganic material, which is the
incombustible ash associated with trace elements in the fuel. Wood typically contains a wide range
of inorganic material, such as calcium, silicon, magnesium, phosphorus and potassium. Whilst much
of the inorganic material will remain in the stove grate as ash, some will be carried with the
combustion flue gases- especially minerals such as potassium which are more volatile. There is
ongoing research to see whether potassium acts as a condensation nucleus for carbonaceous
particulate matter to form around.

Guidelines are given for recommended mass-based exposure to PM, corelating to 25 µg/m3 PM2.5 in
the EU as shown in Table 1 [Ali 2021]. There are still unanswered research questions regarding safe
levels of exposure to PM and the degree to which size and composition matters. With this in mind,
The World Health Organisation (WHO) have stated that there is no evidence of a safe level of exposure
to PM or a threshold below which no adverse health effects occur [WHO, 2013]. Concentrations are
measured as peak values or averages over a specific time period such as 24hrs or a year. The short

2
peak concentrations will record a maximum concentration according to specific activities such as
igniting a stove whereas the average values give an indication of ongoing exposure.

Pollutant WHO EU
PM2.5 10 µg/m3(1yr) 25 µg/m3 (24hr)
3
25 µg/m (24hr)
PM10 20 µg/m3(1yr) 50 µg/m3(1yr)
3
50 µg/m (24hr) 40 µg/m3(24hr)
Table 1: Particulate Recommended Average Exposure limits [Adapted from Ali, 2021]

2. Stove Indoor PM Exposure

2.1 Developing world

The focus of this review is the UK and developed countries with similar levels of technology, however
it is useful to place this in the context of wider Global impacts. The impact of pollution by ambient fine
particulate matter in 2015 resulted in ∼4.2 million deaths globally, representing ∼4.2% of disability-
adjusted life years [Cohen et al, 2017]. This impact is most strongly seen in Asian and African countries
where many communities rely on simple systems of solid fuel combustion for cooking and heating
such as ‘3 stone fires’ which can be made by the householder.

Figure 2 shows the extent of premature deaths worldwide attributed to indoor pollution from solid
fuels between 1981 and 2020 [Ali 2021]. The contrast between persistently high levels of impact in
developing world, especially Asia, can be contrasted with a gradually improving picture in the
developed world areas such as Europe and North America.

Figure 2; Average premature deaths attributed to diseases as a result of indoor air pollution from solid fuel
[Ali, 2021]

3
The detailed analysis of pollution studies in the developing world is outside the scope of this paper. It
can be noted that the combustion sources are usually open fires or simple cookstoves have limited
ventilation to the outside. There are many studies available in the literature regarding the levels of
indoor air pollution in these countries, many of which highlight the disproportional impact that is felt
by women in these areas, who typically spend more time adjacent to the particulate sources when
cooking [Cohen, 2017; Ali, 2021] . The combustion technology of these basic cooking systems is not
comparable with a modern developed world enclosed woodstove and hence cannot be used to inform
risk factors in a typical developed world setting.

2.2 Stove Indoor air studies in the developed world

In comparison to developing world studies, there is much less literature information available
regarding the concentrations of stove related indoor PM for developed world countries. There are
inconsistencies between measurement protocols between studies, such as whether peak values of
average values are recorded, and different analysis techniques. The most significant studies are listed
below.

Salthammer et al assessed indoor air quality associated with seven wood burning stoves/fireplaces in
private homes, located in Germany [Salthammer, 2014]. The study measured concentration of
particulate in terms of UFP (5.6-55nm), FP (0.3-20µm) and PM2.5. Selected gases and VOC were also
measured in this study. The tests were conducted over a three-day period; with day one as background
measurements, day two included ignition and 4 hours burning, then day three to evaluate residual air
quality afterwards. It is noted that no smoking or candles were permitted during the testing period
and the cumulative effects of a typical winter pattern of stove use was not studied. The study found
inconsistent PM results between the stoves, but concluded that the PM guidelines as defined by the
German Federal Environment Agency (and WHO) were not generally exceeded during the trials, with
the average PM2.5 being ~22µg/m3 (24 average) during firing. The levels of finest particles were seen
to increase and could be attributed to opening the fire chamber door for re-loading. During adverse
weather for one of the tests- a atmospheric temperature inversion resulted in reduced draft during
ignition and lead to high levels of indoor CO and NOx. The report identified that other factors could
play a role in the PM concentrations that were identified including stove construction, climatic factors,
air settings and the characteristics of the fuel burned, but that these were outside the scope of the
research study [Salthammer, 2014].

Vicente et al looked at the impact of wood combustion on indoor air quality, including a comparison
of an enclosed wood stove with an open fireplace. Samples of PM10 were taken and characterised in
terms of the inorganic species, organic carbon and elemental carbon. Vicente noted that although a
standard sequence of ignition, air setting and operation was followed, there was still a very high
degree of variability in the results.

Similarly to the Salthammer analysis, the Vincente work showed the highest peak concentrations were
associated with the ignition phase and with reloading of the wood, as shown in Figure 3 [Vicente,
2020]

4
 Figure 3: Indoor profiles of PM10, CO2 and CO measured during (A)fireplace and (B)wood stove operation
[Vicente, 2020]

A marked difference was shown in air quality between the enclosed stove and open fireplace as shown
in Figure 4. In this analysis, indoor PM10 emissions had roughly a two-fold increase compared to a
twelve-fold increase with the fireplace.

Figure 4: PM10 concentrations indoors- background air, using combustion appliances and outdoor air [Vicente,
2020]

The values of peak PM measured by Vicente et al are given in Table 2, noting that the peak values for
the woodstove correlate with refuelling. The average room concentration during operation was higher
in this study with an average PM10 of ~79 µg/m3 for the woodstove compared to 319~ µg/m3 for the
fireplace, noting a wide degree of uncertainty as shown below.

Open fire PM10 µg/m3 Woodstove PM10 µg/m3

Initial concentration µg/m3 30.5± 1.24 0.933± 0.312
Peak concentration µg/m3 2328± 1853 76.7± 28.6
Average Concentration (8hr) µg/m3 319 ± 173 78.5 ± 24.0
Emission rate µg/min 1.29± 1.25 0.049± 0.035
Table 2: Concentrations recorded for an open fire compared to wood stove. [Adapted from Vicente, 2020]

An earlier study was undertaken by Ward and Noonan in 2008, who conducted a PM2.5 sampling
program before and after a wood stove upgrade programme in Rocky Valley USA [Ward and Noonan,
2008]. The program involved replacing old stoves with improved EPA certified wood stoves. The study
looked into particulate PM2.5 mass concentration using a ‘Dustrack’ analyser, ratio of EC/OC and
various chemical markers associated with woodsmoke. As with other studies, there was a substantial
amount of variability seen in the concentrations between different households. Figure 5 shows the

5
average PM2.5 across the 20 locations investigated and it can be seen that some households had more
than double the concentrations of others.

Figure 5: PM2.5 mass concentrations before and after improved woodstove changeout [Ward and Noonan
2008]

The Ward and Noonan results demonstrated an average reduction in indoor PM2.5 of 71% once the
improved woodstoves were installed [Ward and Noonan, 2008]. The study recorded a 24-hr average
of up to 118µg/m3 before the change out, compared with most homes being below 20µg/m3 after the
changeout, which is below the WHO 24hr recommended exposure limit. The reduction demonstrates
that the age and quality of stove is an important factor when considering potential risk so
householders should be encouraged to ensure their appliances are well sealed and maintained. Only
2 homes exceeded 35µg/m3 after the changeout, but the influence of influx of outdoor air, such as
one resident known to be smoking on the patio, was not quantified. The study also measured a
reduction by 45% of levoglucosan, which is a commonly used marker for wood combustion.

Another study compared the PM2.5 concentration associated with stove use in 101 homes in rural USA
locations [Walker 2020]. The paper gave a median indoor PM2.5 concentration of 19µg/m3, noting that
there was high variability in the data as had been seen in other studies. Households that had not had
their chimney cleaned in the previous 6 months had 65% higher mean PM 2.5, however the ’grade’ of
the stove use also had a substantial impact on the indoor PM in this work also, where the stove grading
was based on the stove quality, operation and wood quality. The different between the highest grade
of stove use compared to the lowest accounted for a 186% increase in PM. This correlates with the
finding by Ward and Noonan and suggests that householders with older appliances should consider
upgrading them to newer technology stoves.

A comparison of air quality using two Italian residences was conducted by Frasca et al [Frasca, 2018]
One household had a 15.4 kW pellet stove, and the other had a 25kW thermo-fireplace fuelled with
hardwood logs. This study found that infiltration from outdoors was main source of the fine
particulates whereas dust resuspension by human movement was the main source of course particles.
With regards to the PM associated with stove use, the study found that the most significant peak PM
values were due to ash removal during cleaning of the stoves rather than during their operation. The
ash was removed by a vacuum cleaning system for the pellet stove and by simple shovel for the
thermo-fireplace. The ash removal could be a health risk as it contained significant amounts of copper

6
and manganese which can have health impacts via oxidative stress induction [Charrier and Antasio,
Crobedden]. Householders should avoid inhalation of ash-dust by good ventilation during stove
cleaning, use of an ash-vacuum appliance and use of a room vacuum shortly after cleaning, allowing
for deposition time of the ash onto floors.

A study by Chakraborty et al measured PM2.5 and PM1 using low cost sensors in 20 UK residences that
had woodburning stoves [Chakraborty, 2021]. Their study identified an increase in daily average PM2.5
by 196% and PM1 by 227% compared to a non-stove control group, i.e. approximately a twofold
increase compared to background similarly to the study by Vicente [Vicente, 2020]. It can be noted
however that the actual concentration values given were fairly low, at 12.2 ±10.36 µg/m3 PM2.5 for
instance compared to 4.12±3.6 µg/m3 when a stove was not in use. The hourly peak PM2.5 during stove
use was 27.34 µg/m3compared to 12.21 µg/m3 without stove, whereas the PM1 was a maximum of
19.44 µg/m3 compared to 8.34 µg/m3. These values are below the WHO 24hr recommended exposure
limit. Their analysis of the data identified that opening the stove door for reloading of wood as the
main source of indoor PM, which has also been identified as an issue by Vicente. The ‘flooding’ of
combustion emissions into the room can be avoided if sufficient care is taken during the door opening
such that the door catch should be opened very slightly initially, allowing the air pressure to settle,
before fully opening the door for refuelling.

A general PM exposure study was conducted by Buonanno et al in 2018. They investigated the
exposure of ~100 children to UFP and black carbon using handheld particle counters and
aethalometers. The analysis showed that cooking and using transportation were recognised as the
main activities contributing to exposure [Buonanno, 2018]. The study covered both indoor and
outdoor exposure, and all the children had gas cooking in the home. Most of the children also had a
fireplace for heating with in the homes: 48% solely a fireplace, 35% had a fireplace combined with gas
heating, but the research did not specify the types of fireplaces or how often they were used. The
fireplaces were not identified as a significant source of UFP or BC in this study, nor were they included
in the results discussion. The study concluded that cooking activities presented the highest dose
intensity for UFP, whereas transportation presented the highest exposure of black carbon due to
vehicular emissions.

Research Study PM2.5 Concentration

µg/m3
Chakraborty, 2021, hourly mean 12.2 ±10.36
Salthammer, 2014. 24hr average 22
McNamara, 2013. Winter mean values 32.3±32.6
Ward and Noonan, 2008 ( post upgrade program). 24hr average <20
Walker, 2020. Median indoor value 19
Table 3: Comparison of reported PM2.5 Concentrations

Table 3 summarises PM2.5 concentrations reported by some of the literature studies, with variation in the values
reflecting the high levels on uncertainty in the results. The values given as 24 hour averages are generally below
the WHO recommended limits.

2.3 Risk of exposure evidence

Whilst some researchers have quantified the concentration of PM in households, there are relatively
few literature sources that attempt to correlate the health risks associated and this is an area where
more research is required.

7
The issues of whether adapting heating behaviours could lead to improved air quality and hence lower
risk of exposure was recently considered by a team led by VITO [Cops, 2021]. The study looked at how
specific procedures of stacking the wood in the stove, lighting the wood, the air settings and wood
moisture could impact the resultant exposure and health of participants. The test phase of the study
involved a comparison of the user’s traditional stove use behaviour vs an improved stove use
intervention following some stricter operational guidance. The intervention gave rules for loading of
the wood according to the Swiss method, whereby largest logs are stacked at the bottom and
progressively smaller towards the top, with ignition from the top of the stove. The study acknowledges
that this might not be appropriate for all stove types, for instance those with air inlet from the bottom.
Indoor and outdoor air quality was measured and a reduction in BC was measured in the homes with
the improved behaviour intervention. Participants also had biological samples taken to measure
exposure to combustion products, however the trends in the results of these were not clear-cut
because of complications due to diet, exposure to vehicle emissions and underlying infections (e.g.
having a cold). Due to this, only a limited influence of improved health impacts could be inferred.
Another limitation of the study was that only 6 households took part in the full study rather than the
8 expected due to Covid-19 impacts, although a questionnaire on a larger group was also included. In
conclusion, the researchers stated that householders should be encouraged to use alternatives to
wood burning for biggest improvements to air quality, however those using wood burners could
reduce their emissions by following improved firing and fuel practices. Recommendations for firing
included the use of kindling or natural cubes rather than paper, cardboard or newspapers; stacking
according to the Swiss method with largest logs at the bottom; wood moisture to be between 10-20%
and use of a hard wood fuel (e.g. beech, oak) rather than softwoods or prunings. This led to improved
combustion and hence lower emissions so reduced the risk of exposure.

The metric of lung deposited surface area can be used to evaluate the potential adverse health effects
related to UFP exposure. A study by Geiss et al. 2016 used LSDA to investigate risk from a variety of
environments including a house with a wood burning stove, travel by car and a factory setting.
Measurements were taken from 20 minutes before the stove was lit and for three hours during
combustion. Unfortunately, the authors had issues with another PM source in an adjacent room (gas
cooker). The cooker was used during the sampling period and the report concluded that the gas cooker
was in-fact a stronger contributor to the UFP concentrations than the woodburning stove [Geiss,
2016].

Endotoxins are a biological component of particulate matter associated with adverse human health
effects such as respiratory illness. The levels of endotoxin and PM in 50 homes with woodstoves was
investigated by McNamara et al [McNamara 2013]. Their results showed Winter mean values of PM2.5
of 32.3±32.6, indicating wide uncertainty in the values. These figures are also higher than some of the
other studies including Vicente, Ward and Noonan and Walker [Vicente, 2020; Ward and
Noonan,2008, Walker, 2020] Significantly, although the endotoxin concentration measurements
showed a correlation with the number of residents in the household, there was no correlation with
PM2.5 or the PM10-2.5 fractions. This suggests that the particle concentration by itself does not give a
good indication of risk.

The quality of the stove appliance has been seen as an important factor in PM exposure, however the
human behaviour and stove location are also significant. Abbatt et al. 2019 considered the reactive
multiphase chemistry involved in indoor air quality. They noted that the structure of a building is a

8
barrier that inhibits the flow of air both indoors to outdoors and vice versa. The typical residence time
for indoor air highly variable but is ~1-2 hours, and air exchange is through leaky windows and walls,
driven by pressure gradients due to heating and weather [Abbatt, 2019]. Salthammer et al. 2014 also
demonstrated that room volume and air exchange rate was important. It is noted that with increasing
concern over the impacts of climate change, residences are driven to improve their insulation and
efficiency resulting in a reduction in the air exchange rate. Decreasing the amount of air changes can
cause a cumulative increase in indoor PM, regardless of the source.

Buildings can have a substantial impact on human health and wellbeing, as discussed in a report from
the American Association for the Advancement of Science symposium [Ham, 2019]. A symposium
participant from Harvard T H Chan School of Public Health showed that cognitive function and decision
making can be dramatically improved by improving residential occupants air quality. The importance
of human behaviour was highlighted, such as use of extractor fans for cooking, frequency of
vacuuming and lifestyle choices to burn incense for instance [Ham, 2019]. Correct operation of wood
stoves creates a natural draft that pulls air from the room into the appliance hence potentially drawing
cleaner outside air into a room. Householders with woodstoves should be educated regarding the
importance of ventilation in maintaining a good influx of clean air and low level of indoor PM.

The routes to exposure from stove emissions include inhalation, ingestion and permeation of skin
[Kristensen 2019]. Figure 6 shows a summary of possible health implications that have been identified
from various combustion emissions including PM [Ali 2021], noting that all the species identified can
be reduced by increasing the efficiency of the combustion appliance. There is an ongoing drive to
improve the performance of new stoves via initiatives such as clearskiesmark and introduction of
Ecodesign 2022 regulations in Europe. The quality of the fuel is also vitally important as high moisture
fuels cannot burn efficiently, hence householders in the UK should be encouraged to only used ‘Ready
to burn’ labelled wood.

Figure 6. Pollutants associated with inefficient solid fuel combustion and possible mechanisms of toxicity.
[Ali 2021]

Public Health England have conducted an extensive review of epidemiological evidence from studies
across Europe, North America, Australia and New Zealand, in order to assess any associations between

9
respiratory outcomes in children and exposure to solid fuel combustion [Guercio 2020]. Despite a
broad review of available literature, they found that results were inconsistent and limited, for instance
details were missing regarding types of fireplaces and stoves and often didn’t specify the fuels that
were used. They used statistical analysis of the data available, they found that there was no
association shown between exposure to indoor wood burning and risk of asthma in developed
countries. They found a slight but ‘non-significant’ association to risk of respiratory infections (e.g.
such as rhinitis, hay fever and influenza). A study by Petry calculated estimated exposures for different
room scenarios using candle combustion emissions, which were between 5.48µg/m3 to 90.24µg/m3
for PM2.5 [Petry 2014]. Petry concluded that under normal conditions, these PM concentrations from
scented candles do not pose known health risks to consumers. Vicente also concluded that the
carcinogenic risk due to particulate bound PAH from their woodburning stove scenario was negligible,
but could be at harmful levels for an open fireplace [Vicente 2020]. The PHE review concluded that
they could not say that exposure to indoor solid fuel emissions doesn’t affect respiratory heath, but
there is currently no strong scientific evidence to say that it does. The study recommended that
further, improved research is conducted in the area [Guercio 2020].

3. Other domestic sources of particulate matter

There are a wide range of common sources of PM in a typical household which include combustion
and heating sources (e.g. cooking, wood stove, smoking, candles, incense), semi-volatile chemicals
(cleaning products, personal care, cooking), particulates from humans themselves (skin, skin oils),
plastics (microfibres and microplastics) and minerals (personal care such as cosmetics and talcum
powder, potentially hazardous such as asbestos). A limitation in comparing literature concentration is
that studies do not use a consistent basis of measurement. Some common sources of PM in
comparison to wood stoves are discussed briefly below.

Kristensen et al, 2019 Conducted a study of semi-volatile organic compounds (SVOC) in a single family
residence in California [Kristensen 2019]. Figure 7 shows a comparison of some of these in terms of
the indoor gas‐ plus particle‐phase SVOC in a Californian household.

Figure 7; Box plot showing the outdoor and indoor SVOC concentrations (μg/m3) in a Californian study. This
compares a vacant period, and period of normal occupancy along with the total concentrations associated

10
 with indoor activities (stovetop and oven cooking, cleaning, and candle light burning) [Adapted from
Kristensen 2019]

Kristensen observed substantial increases in SVOC associated with both cooking and cleaning. It was
noticed that although thorough ventilation by opening windows could effectively reduce the indoor
air concentration of VOC, the effect was only temporary and levels returned to the previous
concentrations within hours. This could be explained by partitioning of species on to indoor surfaces,
and household materials absorbing VOC and then re-releasing them as the air concentrations
decreased to equilibrium concentrations. They found that the SVOC consisted of a complex mixture
of species including gaseous compounds, dust and surface films.

The study showed that emissions exceeding 200µg/m3 were measured with cooking activities such as
oven cooking and stove top frying, but there was no SVOC associated with boiling, microwaving or
toasting bread [Kristensen]. Analysis has shown that aerosols for cooking are mainly in the ultrafine
particle range so extremely respirable [Abdullah 2013]

Abbatt et al. 2019 compared the combustion related sources of emissions to include wood stoves, gas
stoves, candles, incense burning and cigarette smoking. Significant non combustion sources included
food and cooking, cleaning products and personal care products. Abbatt discussed that shedding of
skin and release of skin oils from humans themselves made a substantial contribution to the
household dust. It has been shown by Downing and Milstone [Downing, 1982; Milstone, 2004] that
between 10’s to 100’s of mg of matter/ per hour/ per human can be shed.

Another study looked at how human activities would influence the indoor concentration of particles
[Tian 2020]. This study also found that cooking was a major source, but high movement activities such
as mopping would also displace settled particles into the air. The study found that the dominant
source of particles was associated with oil-based cooking such as frying, whereas water-based cooking,
such as stewing and boiling, did not emit measurable particle increases. This correlates with the
findings from Kristensen. Abdullah et al identified a number of possible chemical markers that could
be used to identify PM from cooking such as carboxylic acids and levoglucosan [Abdullah, 2013].
Further research is required in this respect as the latter could be also used as marker for wood smoke.

Values of mass concentration due to cooking were measured in a manufactured test house, as part of
the HomeCHEM tests [Patel, 2020]. The mass concentration in for cooking breakfast was 35±20 µg/m 3,
stir fry 30± 10 µg/m3, toast 12± 2 µg/m3 compared to unoccupied building 2.3± 0.4 µg/m3. The cooking
activities involving frying produced higher levels of PM compared to the stove use studies reviewed in
the earlier sections.

An exposure study by Beko et al involved 60 non-smoking residents in Copenhagen. They used

portable devices to continuously log their exposure to UFP [Beko, 2015]. Participants were equipped
with a portable monitor that continuously logged particle number concentrations, along with a GPS
logger to track their locations, and they were also asked to keep a diary of their locations and activities.
This study covered both indoor and outdoor environments within the participants daily exposure. The
home environment was shown to account for an average of 50% of the daily exposure to UFP, however
significant variability was shown between the participants. There was also a discrepancy between the
GPS tracking logs and the diary entries. The results showed that homes where cooking frequently
occur have significantly higher UFP daily exposures comparted to homes that did not have cooking
activities. 90% of the participants daily exposure occurred within buildings as people were in buildings
for ~90% of the time, although the highest average particle number concentrations were measured

11
during passive transportation (cars, public transport). Lower particle number concentrations were
associated with active transport such as walking, running and cycling.

Researchers in Korea investigated personal exposures to PM2.5 relative to various

microenvironmental concentrations [Lim, 2012]. These included various indoor locations (e.g.
residential, office, restaurant), transportation and outdoor locations. The average exposure per
person was 19.8±15.3µg/m3 of PM2.5, which is similar to the indoor concentration for stove users as
discussed in the previous section. They found that different types of population group had different
average exposures, ranging from 9.8µg/m3 for office workers that went home after work, to
43.1µg/m3 for office workers that went to restaurants and bars after work. These highest
concentrations were due to permitted indoor smoking and use of ‘at table’ cooking with charcoal grills.

Burning of incense can be a significant source of PM with emission factors of PM2.5 shown to vary
between different types of incense from 0.4 (‘smokeless’ incense stick) to 44.5 mg/g [See, 2011].
Depending on the stick composition, metals are present in the smoke, for example aluminium and
iron. See et al concluded that with reference to indoor air quality guidelines, inhalation of incense
smoke can cause adverse health impacts. This could be of particular concern to those burning multiple
sticks per day, for example as part of religious practices.

Petry 2014 investigated the health risk from scented candles due to PM, VOC and SVOC. The tests
were conducted in small (2.2m2) and larger (26m2) experimental chambers with control of the air
exchange rate, temperature and humidity. Specific VOC were investigated such as aldehydes
(formaldehyde, acetaldehyde), aromatics (e.g. benzene, toluene, styrenes), PAH, and PCDD/PCDF. The
PAH and PCDD/PCDF were below the detection limits of the analysis methods. The most abundant
VOC from combustion was formaldehyde with estimated exposures of up to 63.85µg/m3 TWA.

A study at Manchester University compared indoor air associated with a 10-year-old stove, a new
Ecodesign compliant stove and various typical domestic PM sources, including cooking, candles and
incense [Lea-Langton, 2021]. Results from this study are shown in Figure 8, showing peak and average
values of indoor air for the 10-year-old stove, Ecodesign compliant new stove, 1 piece toast, cooked
breakfast, 1 scented candle, 1 incense stick and 3 puffs of hairspray. In this case, the average values
were taken over the test period only, which was typically 40 minutes, rather than as a 24-hr average.
The cooking activities including frying or grilling meat were by far the highest sources of PM and
typically had PM concentrations over ~500µg/m3, whereas the stove indoor air quality averages during
operation was below the WHO recommended limits 24-hr average of below 25 µg/m3. However higher
values of PM10 of ~500µg/m3 were seen for a short time during the ash emptying and it is
recommended that the room is ventilated during this activity. These results are consistent with the
other literature sources discussed in section 2.

12
 Figure 8: Comparison of PM sources of PM10 and PM 2.5 associated with typical domestic sources, peak values
and averages over test duration [Lea-Langton, 2021].

A study by Sain et al demonstrated that using typical tap water in an ultrasonic dehumidifier would
result in emission of inhalable mineral particles, especially if high hardness water is used. The study
recommends the use of deionised or distilled water in these devices [Sain, 2018]. Cleaning activities
can also create higher levels of airborne PM due to the unsettling of deposited material. Vicente et al
investigated the amount of PM10 associated with vacuuming and measured between 75.4 ±7.89
µg/min to 200± 99 of PM10 emitted, depending on vacuum type [Vicente, 2020b]. The particulate
collected contained a wide range of organic species as well as copper suggesting motor-related
emissions were also a contributor. Hence the increase in emission during vacuuming is partly due to
the motor and partly due to re-suspension of dust.

The chemistry associated with emission of various VOC species is complex and results in reactions
forming secondary organic aerosols (SOA) and particulates. For instance, the rapid photolysis of
chlorine from chlorine bleach followed by reaction with limonene (commonly found in foods, cleaning
products and skincare products) has been shown to produce a high yield of UFP [Abbatt, 2019].

The summary of these findings is that normal domestic behaviours such as cooking, use of personal
care products and cleaning are responsible for generating PM at levels that can exceed those from
careful use of an enclosed woodburning stove. However, to further reduce the indoor PM from stove
use; reloading of wood should be undertaken carefully to minimise any ‘flooding’ of emissions into
the room, and it’s recommended that the ash removal should be done with a specialist ash-cleaning
vacuum appliance and ventilation.

4. Limitations and research gaps

Analysis of literature shows that there are serious limitations regarding the quantity of research
studies and consistency of the research methods used. A wide variety of analysers have been used
from high specification research analysers to low-cost sensors. There is variability in the type of PM
investigated, such as PM10, PM2.5 or various VOC. The stove specifications are not typically given, nor
are details of the fuels or user behaviour. Factors such as metrological conditions and temperature
are also known to impact emissions behaviour [Ward and Noonan 2008] but are not recorded and
cannot be easily controlled. There are also inconsistencies in the exposure time to various sources- for
instance high PM activities such as ash emptying or cooking activity might only last for 10-15 mins,
whereas wood stoves would be lit for several hours at a time. The natural draft from a lit stove would
promote ingress of air from outside however there is a drive towards increasingly well sealed buildings
to improve energy efficiency. The studies reviewed had limited information on the rate of change in
the air and ventilation and this should be studied further in relation to stove indoor air quality.

Manigrasso et al. 2013, highlight some of the issues in measurement of exposure to combustion
source emissions. Most indoor sampling methods are unable to capture the transient behaviour
associated with particle formation and growth, and high short time exposures to the smallest particles
might be missed. This work demonstrated how advanced sampling methods such as a fast mobility
particle sizer (FMPS) was needed to understand dosimetry in a more detailed manner.

The quality of the measurement system will impact the results obtained. Singer 2018 compared a
selection of seven low-cost particle sensors against 2 research grade analysers to assess their
performance in assessing PM2.5 [Singer 2018]. The low-cost analysers were the AirBeam 1, Air Quality
Egg, AirVisual Node, Awair, Foobot, Purple Air PA-II and Speck. The research grade monitors were the
Thermo pDR-1500 and the MetOne BT-645. Particles from typical residential sources such as

13
combustion and cooking were generated under laboratory conditions. The study showed a wide
variation in response between the monitors. Only four of the low-cost monitors were within a factor
of 2 of the estimated true concentrations (AirBeam, AirVisual, Foobot, and Purple Air), 2 others
reported concentrations much lower than the true values (Air Quality Egg and Awair). One monitor
did not consistently respond to the source emissions (Speck). It is noted in their conclusions that the
IAQ market is dynamic in terms of new products and improvements to existing products- for example
by improving the algorithm used for the concentration calculations. It recommends the introduction
of an industry standard test method for IAQ.

The performance of low-cost environmental monitors and sensors for measuring air quality was also
assessed by Demeanega et al [Demanega 2021]. Activities included candle burning, essential oils
vacuuming and popcorn cooking. The majority of tested devices had generally good performance for
gases, but under-reported PM by up to 50% and had poor quantitative agreement for VOC. Figure 8
shows a comparison of the PM 2.5 results for the different analysers compared to the research grade
analyser (in black). The study found that whilst the monitors could detect source events happening,
they could not accurately quantify the levels and concludes that standards and guidelines for testing
are required.

Figure 8; Low cost PM sensors compared to a research grade calibrated monitor (black dashes) [Demanega
2021]

An issue is that different types of analysers respond differently to different types of PM. [McNamara
2011] found a factor of 1.65 difference between gravimetric analysis and a handheld Dustrack
analyser, whereas [Kingham 2006] found dustrack over recorded PM10 by a factor of 2.73. Ward and
Noonan 2008, Yanosky 2002 found a reasonable correlation between dustrack measurements and a
FRM sampler (R2= 0.859).

Ott et al looked at a range of PM sources using two different continuous particle monitors [Ott 2006].
The types of monitors were (a) a photo-charging ionisation technique (PC) that responds to the
particle PAH, and (b) a diffusion charging technique (DC) that measures the active surface area of fine
particles. A comparison of the readings as a ratio of PC/DC gives a mass of PAH to active surface area.
Sources include tobacco smoke (cigarettes, pipes and cigars), incense, candles, cooking, fireplaces/
woodsmoke and in vehicle exposures (California). The ratio of readings from the two detectors varied

14
substantially between different sources. The results gave similar values for incense and cigarette, low
readings for cooking and high readings for burning of cedar wood (indoors, but not within fireplace-
so no extract via chimney. A fireplace test was conducted which appeared to give a low value of PC/DC
close to zero. It should be noted that the test was not performed under normal real- world conditions
however as the chimney damper was closed for a short period (30s) to obtain a strong response on
the analysers. The paper suggests that development of this ratio technique might be used as a
signature to identify different sources.

These inconsistencies in measurement, sampling and operation of stoves should be addressed by

development of a standard test protocol. There should also be development of instrumentation to
assess leakage from the stove (either from fuel re-loading or poor seals), perhaps using laser or optical
methods. The test protocol should take into account factors such as metrological and temperature
factors, the exposure time to various sources, rate of air change, stove quality, fuel quality and user
behaviour.

15
5. Conclusions

The most significant findings of this review are as follows:

1. No scientific evidence was found for adverse health impacts from exposure to the indoor air
typically associated with modern enclosed wood burning stoves. It should also be noted
however that data in the context of developed world studies is extremely limited and
therefore this doesn’t mean there is no risk. The World Health Organisation have stated that
there is no safe level of particulate matter exposure and further research into risk is required.
2. Other sources of particulate matter in the home such as emissions from cooking can release
much higher levels of PM compared to enclosed wood stoves and could therefore have a
greater health risk potential.
3. Factors including stove quality, wood quality, location factors (e.g. ventilation, temperature,
weather) as well as user behaviour can impact indoor air quality but these are rarely reported
in the studies reviewed.
4. Standard test protocols and measurement methods for domestic indoor PM should be
devised for further research in this area. These should take into account that the length of
exposure may be different for the different cases, for instance a short cooking or ash removal
activity compared to longer stove operation activity. The rate of air changes and ventilation
needs to also be defined in a quantitative way.

Author

Dr Amanda Lea-Langton

Senior Lecturer in Bioenergy Engineering, The University of Manchester

Independent Peer Reviewer

Prof Alan Williams, CBE FREng FRSC

Emeritus Professor in Combustion, The University of Leeds

16
Nomenclature

BC Black carbon, measured using

COPD Chronic Obstructive Pulmonary Disease

CV Cardio- vascular

DC Diffusion Charging technique (DC)

EC Elemental carbon

FP Fine particle also known as PM2.5, diameters that are < 2.5 µm.

IAQ Indoor Air Quality

OC Organic Carbon, condensable fraction of PM that contains complex mix of SVOC,

VOC

PAH Polycyclic Aromatic Hydrocarbons

PC Photo-Charging ionisation technique (PC)

PCDD Polychlorinated dibenzodioxin

PCDF Polychlorinated Dibenzofuran

PM2.5 Fine particles that are generally <2.5 µm diameter

PM10 Inhalable particles that are generally <10µm diameters

SOA Secondary Organic Aerosol

SVOC Semi-volatile Organic Compounds

TWA Time weighted average

UFP Ultra Fine Particulates

VOC Volatile Organic Compounds

17
References

Abbatt, Jonathan P. D. and Wang, Chen., (2020), The atmospheric chemistry of indoor
environments, Environ. Sci.: Processes Impacts, Pages 22-25

Abdullahi, Karimatu L., Delgado-Saborit, Juana Maria, Harrison, Roy M. (2013) Emissions and
indoor concentrations of particulate matter and its specific chemical components from
cooking: A review, Atmospheric Environment 71 p260-294

Ali, Muhammad Ubaid., Yu, Yangmei., Yousaf, Balal., Munir, Mehr Ahmed Mujtaba., Ullah, Sami.,
Zheng, Chunmiao., Kuang, Xingxing., and Wong, Ming Hung. (2021) Health impacts of
indoor air pollution from household solid fuel on children and women. Journal of Hazardous
Materials 416

Beko, Gabriel., Kjeldsen, Birthe Uldahl., Olsen, Yulia., Schipperijn, Jasper., Wierzbicka, Aneta.,
Karottki, Dorina Gabriela., Toftum, Jørn., Loft, Steffen., and Clausen, Geo. (2015)
Contribution of various microenvironments to the daily personal exposure to ultrafine
particles: Personal monitoring coupled with GPS tracking, Atmospheric Environment 110,
pages 122-129

Buonanno, G., Stabile, L., Morawska, L. and Russi, A. (2013). Children exposure assessment to
ultrafine particles and black carbon: The role of transport and cooking activities.
Atmospheric Environment 79 53-58.

Carvalho, Ricardo L., Jensen, Ole Michael., Afshari, Alireza., Bergsøe, Niels C., (2013). Wood-
burning stoves in low-carbon dwellings, . Energy and Buildings, Volume 59, Pages 244-251,
ISSN 0378-7788, https://doi.org/10.1016/j.enbuild.2012.12.006.

Calvo, A.I., Tarelho, L.A.C., Alves, C.A., Duarte, M., Nunes, T., (2014). Characterization of
operating conditions of two residential wood combustion appliances. Fuel Process. Technol.
126, 222–232.

Calvo, A.I., Martins, V., Nunes, T., Duarte, M., Hillamo, R., Teinila, K., Pont, V., Castro, A., Fraile,
R., Tarelho, L., Alves, C., (2015). Residential wood combustion in two domestic devices:
relationship of different parameters throughout the combustion cycle. Atmos. Environ. 116,
72–82.

Chakraborty, Rohit., Heydon, James., Mayfield, Martin. and Mihaylova, Lyudmila. (2020) Indoor
Air Pollution from Residential Stoves: Examining the Flooding of Particulate Matter into
Homes during Real-World Use. Atmosphere, 11

Cohen, A.J., Brauer, M., Burnett, R., Anderson, H.R., Frostad, J., Estep, K., Balakrishnan, K.,
Brunekreef, B. Dandona, L. et al. (2017). Estimates and 25-year trends of the global burden
of disease attributable to ambient air pollution: an analysis of data from the Global Burden
of Diseases Study 2015 Lancet, 389 pp. 1907-1918

Cops, Jirka., Debrouwere, Katleen., Koppen, Gudrun., Stranger, Marianne., Van Poppel,
Martine., Franken, Carmen., Den Hond, Elly. and Van Dyck, Lize. (2021) Intervention study:

18
 from a traditional to consciously better use of wood-burning stoves: Impact on air quality
and health. 03-02-2021 Published by VITO, PIH and Vlaams Instituut Gezond Leven

Demanega, Ingrid., Mujan, Igor., Singer, Brett C., Anđelkovi, Aleksandar S., Babich, Francesco
and Licina, Dusan. (2021) Performance assessment of low-cost environmental monitors and
single sensors under variable indoor air quality and thermal conditions. Building and
Environment 187

Downing, D. T., Stranieri, A. M., and Strauss, J. S. [1982] The Effect of Accumulated Lipids on
Measurements of Sebum Secretion in Human Skin, J. Invest. Dermatol., 79, p226–228.

EPA, USA (2021) https://www.epa.gov/pm-pollution/particulate-matter-pm-basics

Frasca, Daniele., Marcoccia, Melissa., Tofful, Luca., Simonetti, Giulia., Perrino, Cinzia. and
Canepari, Silvia. (2018) Influence of advanced wood-fired appliances for residential heating
on indoor air quality. Chemosphere 211 p62-71

Geiss, Otmar., Bianchi, Ivana., Barrero-Moreno, Josefa. (2016) Lung-deposited surface area
concentration measurements in selected occupational and non-occupational
environments, Journal of Aerosol Science 96 p24–37

Guercio, Valentina., Pojum, Iulia C., Leonardi, Giovanni S., Shrubsole, Clive., Gowers, Alison M.,
Dimitroulopoulou, Sani., Exley, Karen S. (2020) Exposure to indoor and outdoor air pollution
from solid fuel combustion and respiratory outcomes in children in developed countries: a
systematic review and meta-analysis. Science of the Total Environment 755.

Ham, Becky [2019]. ‘Indoor chemical pollution impacts often remain invisible’, report from
American Association for the Advancement of Science symposium.
http://science.sciencemag.org/

Kristensen, Kasper., Lunderberg, David M., Liu, Yingjun., Misztal, Pawel K., Tian, Yilin., Arata,
Caleb., Nazaroff, William W., and Goldstein, Allen H. (2019) Sources and dynamics of
semivolatile organic compounds in a single‐family residence in northern California, Indoor
Air.29: p645–655.

Lea-Langton, A.R., Allan, J., Williams, P.I., McFiggans, G. [2021] Evaluation of indoor air quality
from wood burning stoves in comparison to other domestic PM sources. Conference paper-
Cambridge Particle Meeting, Cambridge UK.

Li, Zhisheng., Wen, Qingmei., Zhang, Ruilin. (2017) Sources, health effects and control
strategies of indoor fine particulate matter (PM2.5): A review, Science of the Total
Environment 586 p610–622

Lim, Soogil., Kim, Jeonghoon., Kim, Taehyun., Lee, Kiyoung., Yang, Wonho., Jun, Sangil., Yu
Seungdo. (2012) Personal exposures to PM2.5 and their relationships with
microenvironmental concentrations, Atmospheric Environment 47 p407-412

Manigrasso, M., Stabile, L., Avino, P. and Buonanno, G. (2013). Influence of measurement
frequency on the evaluation of short-term dose of sub-micrometric particles during indoor
and outdoor generation events, Atmospheric Environment 67

19
Milstone, L. M. [2004] Epidermal desquamation, J. Dermatol. Sci., 36, p131–140.

McNamara, M., Thornburg, J., Semmens, E., Ward, T. and Noonan, C. (2013) Coarse particulate
matter and airborne endotoxin within wood. stove homes. Indoor Air 23: p498–505

Ott, Wayne R., Siegmann, Hans C. (2006) Using multiple continuous fine particle monitors to
characterize tobacco, incense, candle, cooking, wood burning, and vehicular sources in
indoor, outdoor, and in-transit settings. Atmospheric Environment 40 p821–843

Patel, Sameer., Sankhyan, Sumit., Boedicker, Erin K., DeCarlo, Peter F., Farmer, Delphine
K.,Goldstein, Allen H., Katz, Erin F., Nazaroff, William W., Tian, Yilin., Vanhanen, Joonas., and
Vance. Marina E. (2020) Indoor Particulate Matter during HOMEChem: Concentrations, Size
Distributions, and Exposures, Environ. Sci. Technol., 54, p7107−7116

Petry, Thomas., Vitale, Danielle., Joachim, Fred J., Smith, Ben., Cruse, Lynn., Mascarenhas,
Reuben., Schneider, Scott. and Singal, Madhuri. (2014) Human health risk evaluation of
selected VOC, SVOC and particulate emissions from scented candles.

Sain A. E., Zook,. J. Davy, B. M., Marr L. C. and Dietrich., A. M. (2018) Size and mineral
composition of airborne particles generated by an ultrasonic humidifier. Indoor Air.
2018;28:80–88.

Salthammer, Tunga., Schripp, Tobias., Wientzek, Sebastian., l Wensing, MichaeI. 2014. Impact
of operating wood-burning fireplace ovens on indoor air quality, Chemosphere, Volume
103, Pages 205-211, ISSN 0045-6535, https://doi.org/10.1016/j.chemosphere.2013.11.067.

See, S.W. and Balasubramanian, R. (2011) Characterization of fine particle emissions from
incense burning. Building and Environment 46 p1074-1080

Singer, B. C. and Delp, W. W. (2018) Response of consumer and research grade indoor air
quality monitors to residential sources of fine particles, Indoor Air. 2018; 28:624–639.

Tian, Yilin., Arata, Caleb., Boedicker, Erin., Lunderberg, David M., Patel, Sameer., Sankhyan,
Sumit., Kristensen, Kasper., Misztal, Pawel K., Farmer, Delphine K., Vance, Marina., Novoselac,
Atila., Nazaroff., William W. and Goldstein. Allen H. (2020) Indoor emissions of total and
fluorescent supermicron particles during HOMEChem, Indoor Air. 00:1–11.

Vicente, E.D., Vicente, A.M., Evtyugina, M., Oduber, F.I., Amato, F., Querol, X., Alves, C. (2020)
Impact of wood combustion on indoor air quality, Science of the Total Environment 705

Vicente, Estela D., Vicente, Ana M., Evtyugina, Margarita., Calvo, Ana I., Oduber, Fernanda.,
Alegre, Carlos Blanco., Castro, Amaya., Fraile, Roberto., Nunes, Teresa., Lucarelli, Franco.,
Calzolai, Giulia., Nava, Silvia., A. Alves C´elia, (2020) Impact of vacuum cleaning on indoor
air quality, Building and Environment 180

Walker, Ethan S., Noonan, Curtis W., Semmens, Erin O., Ware, Desirae., Smith, Paul. Boyer, Bert
B., Erdei, Esther., Hopkins, Scarlett E., Lewis, Johnnye., Belcourt., Annie and Ward, Tony J.
(2021) Indoor fine particulate matter and demographic, household, and wood stove
characteristics among rural US homes heated with wood fuel. Indoor Air

20
Ward, T. and Noonan, C. (2008), Results of a residential indoor PM2.5 sampling program before
and after a woodstove changeout. Indoor Air, 18: 408-415.

Win, Kaung Myat., Persson, Tomas., Bales, Chris., (2012). Particles and gaseous emissions from
realistic operation of residential wood pellet heating systems, Atmospheric Environment,
Volume 59, Pages 320-327,

21