Content uploaded by Domenico Maria Cavallo
Author content
All content in this area was uploaded by Domenico Maria Cavallo on Nov 04, 2014
Content may be subject to copyright.
Emissions of air pollutants from scented candles burning in a test chamber
Marco Derudi
a
, Simone Gelosa
a
, Andrea Sliepcevich
a
, Andrea Cattaneo
b
, Renato Rota
a
,
Domenico Cavallo
c
, Giuseppe Nano
a
,
*
a
Politecnico di Milano, Dipartimento di Chimica, Materiali e Ingegneria Chimica “G. Natta”, Via Mancinelli 7, 20131 Milano, Italy
b
Università degli Studi di Milano, Dipartimento di Medicina del Lavoro, Via San Barnaba 8, 20122 Milano, Italy
c
Università degli Studi dell’Insubria, Dipartimento di Scienze Chimiche ed Ambientali, Via Lucini 3, 22100 Como, Italy
article info
Article history:
Received 18 January 2012
Received in revised form
4 March 2012
Accepted 9 March 2012
Keywords:
Scented candles
Emission factors
Test chamber
Volatile organic compounds
PAHs
abstract
Burning of scented candles in indoor environment can release a large number of toxic chemicals.
However, in spite of the large market penetration of scented candles, very few works investigated their
organic pollutants emissions. This paper investigates volatile organic compounds emissions, with
particular reference to the priority indoor pollutants identified by the European Commission, from the
burning of scented candles in a laboratory-scale test chamber. It has been found that BTEX and PAHs
emission factors show large differences among different candles, possibly due to the raw paraffinic
material used, while aldehydes emission factors seem more related to the presence of additives. This
clearly evidences the need for simple and cheap methodologies to measure the emission factors of
commercial candles in order to foresee the expected pollutant concentration in a given indoor envi-
ronment and compare it with health safety standards.
Ó2012 Elsevier Ltd. All rights reserved.
1. Introduction
Indoor air quality is affected by a number of indoor and outdoor
pollutants sources. However, while several information are avail-
able for outdoor pollutant sources in terms of emission factors
(Ravindra et al., 2008; Estrellan and Iino, 2010), only few infor-
mation have been published for indoor pollutant sources (Ott and
Siegmann, 2006; Sarigiannis et al., 2011), which include tobacco
smoke, as well as cooking, heating, and office equipments (e.g.,
Tuckett et al., 1998; Long et al., 200 0; Fan and Zhang, 2001;
Destaillats et al., 2008).
Burning of candles in indoor environments can release a large
number of toxic chemicals, including acetaldehyde, formaldehyde,
acrolein, and polycyclic aromatic hydrocarbons (Lau et al., 1997;
USEPA, 2001; Lee and Wang, 2006; Orecchio, 2011). It is believed
that regular burning of several candles in indoor environments can
expose people to harmful amounts of organic chemicals (USEPA,
2001).
Among the huge variety of candles available on the market,
scented candles have gained popularity over the past 30 years
resulting in the current abundance of candle shops and aroma-
therapy candle products. For the sake of example, U.S. retail sales of
candles are estimated at approximately $2 billion annually, being
by far the fragrance the most important characteristic impacting
candle purchases.
In spite of this large penetration of scented candles in indoor
environments, a few works investigating pollutants emissions from
candle burning (van Alphen,1999; Fine et al.,1999; Nriagu and Kim,
2000; Guo et al., 2000; Wasson et al., 2002; He et al., 2004; Zai
et al., 2006; Lee and Wang, 2006; Pagels et al., 2009)were
mainly focused on metals and soot emissions rather than on
organic pollutants, such as Volatile Organic Compounds (VOC) or
PAHs (Lau et al., 1997; Maupetit and Squinazi, 2009; Orecchio,
2011). However, candle composition is expected to determine the
pollutants emissions, possibly leading to important emissions of
VOC. Moreover, the priority indoor pollutants identified by the
European Commission (EU) mainly refer to VOC (Kotzias et al.,
2005).
Consequently, the aim of this work has been to characterize
pollutants emissions from the burning of scented candles using
a test chamber. In particular, emission factors for some polycyclic
aromatic hydrocarbons (PAHs), aromatic species (BTEX) and some
short chain aldehydes have been measured for different scented
candles. Among the others, formaldehyde, benzene and naphtha-
lene (which are classified by the EU as “High priority chemicals”)as
well as toluene and xylenes (which are classified by the EU as
“Second priority chemicals”) have been investigated. The obtained
results have been compared with the few data available in the
*Corresponding author. Tel.: þ39 0223993134; fax: þ39 0223993180.
E-mail address: giuseppe.nano@polimi.it (G. Nano).
Contents lists available at SciVerse ScienceDirect
Atmospheric Environment
journal homepage: www.elsevier.com/locate/atmosenv
1352-2310/$ esee front matter Ó2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.atmosenv.2012.03.027
Atmospheric Environment 55 (2012) 257e262
literature in terms of emission factors, as well as with the emission
factors measured for raw paraffin used to manufacture candles in
order to identify the contribution of added fragrances and dyes to
the pollutants emissions.
These results could be used to protect people health by
comparing the expected pollutant concentration in a given indoor
environment with health safety standards, while the test chamber
designed for performing the reported measurements could be
useful to perform further measurements in an easy-to-build stan-
dardized environment.
2. Materials and methods
2.1. Experimental setup
A crucial point in the determination of the pollutants emissions
from burning candles is the simulation of realistic burning condi-
tions. Different approaches have been proposed in the literature,
ranging from sampling ambient air close to a candle burning in
a real room (Orecchio, 2011), to sampling of the exhaust air from
a real-scale ventilated room (Maupetit and Squinazi, 2009) or from
a ventilated environmental chamber (Lee and Wang, 2006) where
the burning candle is located. These methods provide some pro and
some contra, the latter being mainly related tothe cost of a real-size
instrumented room or to the lack of reproducibility of a non-
controlled environment.
To overcome these problems, a laboratory-scale test chamber
has been designed to ensure well defined and reproducible
burning conditions as well as the possibility of sampling easily
the exhausts (Gelosa et al., 2007). The final geometry, defined on
the basis of some preliminary Computational Fluid Dynamics
simulations and sketched in Fig. 1, creates a smooth air flow
around the candle an a large vortex above it, which mixes candle
fumes with the incoming air providing well-mixed conditions
and uniform concentration at the inlet of a long stack, from which
air is sampled far enough from the stack inlet to avoid entrance
effects.
As shown in Fig. 1, the test chamber consists of three parts: the
room itself is a cylinder (diameter 0.6 m, height 0.4 m) covered by
a conical cap (height 0.6 m) and a stack (internal diameter 0.07 m
and height 1.5 m). The cylindrical chamber is equipped with two
portholes to observe the candles behaviour during the tests.
The internal walls of the chamber have been blackened to
minimize radiation phenomena from the chamber walls to the
candle leading to uncontrolled rise of the candle temperature. The
test chamber has been equipped with an air sparger at the bottom
to supply air to the chamber environment with minimum turbu-
lence and very low velocity. Such an air sparger is constituted by
a perforated coil covered by a bed of small glass spheres. For all
experiments, pre-cleaned air through a charcoal trap has been
used. To verify the trap effectiveness,a blank measure on the air fed
to the chamber has been carried out before each test.
The air flow rate to the test chamber has been adjusted
to 10e15 NL min
1
to obtain realistic burning conditions with
a burning rate close to the values measured in some preliminary
burning tests in real rooms.
Four candles have been burned simultaneously for each test;
they have been placed upright at the centre of the chamber bottom,
spaced enough to avoid undesirable thermal influence from one
candle to the others. The locations of the individual candles inside
the chamber have been recorded and kept equal for all the tests.
Burning four candles simultaneously allows for both increasing the
pollutant production rate and averaging the possible differences
among candles of the same kind, thus reducing the uncertainties in
the estimation of the pollutants emission factors.
To determine the candle burning rate, each candle has been
weighted before and after the burning experiment and the corre-
sponding burning time has been recorded.
Five different kinds of coloured (about 30 mg of dye) and
scented (about 2% by weight of fragrance) commercially available
candles with a cotton wick have been tested. The candles had an
approximate weight of 150 g without any container. More details
on such candles can be found elsewhere (Gelosa et al., 2007), where
some preliminary results were also discussed.
Moreover, candles made only from three different kinds of
commercially available paraffins (which are used as raw materials
by candles makers) with a cotton wick have been investigated. They
had an approximate weight of 100 g and were contained in a glass
beaker. The results of these last experiments allow investigating the
contribution of fragrance and dye to the pollutants emissions. Main
features of the investigated candles are summarized in Table 1.
2.2. Analytical methods
At the beginning of each experimental run the candles were
burnt for about 15 min before to start the exhaust sampling. This
initial burning period was used to check that individual candles
reached proper burning conditions, without atypical burning
Fig. 1. Sketch of the test chamber. Arrows inside the chamber roughly represent air
patterns.
Table 1
Main features of the investigated candles.
Sample Container Color Fragrance
A No Brown Cedarwood
B No Blue Plumeria
C No Navy blue Oriental spices
D No Red Rhubarb
E No Pale green Aloe vera
W1 Glass No No
W2 Glass No No
W3 Glass No No
M. Derudi et al. / Atmospheric Environment 55 (2012) 257e262258
behaviour, and to ensure that steady conditions are obtained within
the chamber; for this reason, an on-line measure of the residual
oxygen concentration within the stack was performed. Once
a steady oxygen concentration was obtained the sampling was
started. Then, exhaust gases were sampled and analysed to evaluate
the concentrations of PAHs, BTEX, as well as short chain aldehydes
such as formaldehyde, acetaldehyde, propionaldehyde and benz-
aldehyde. Considering both the initial transient phase and the
sampling period, the overall duration of each experiment was less
than 4.5 hours.
For the short chain aldehydes the DNPH (2,4-
dinitrophenylhydrazine) method was used (Sesana et al., 1991). By
means of a gas sampling pump a well defined volume of exhaust is
passed through a cartridge containing the DNPH sorbent (LpDNPH
S10 cartridge). The aldehydes react quantitatively with DNPH to the
corresponding hydrazone compounds which can be measured
through HPLC. According to the sampling capacity of the DNPH
cartridge, a sampling flow of about 1.5 L min
1
allows for collecting
a total volume of about 45 L in 30 min. The DNPH cartridge can be
desorbed using acetonitrile, and aldehydes analysed using HPLC
(C18 column, 5
m
m 250 mm, detector UV@360 nm). The extraction
yield of the aldehydes has been always close to 100% and the rela-
tive standard deviations on the corresponding measurements were
less than 5%. The detection limit for the various aldehydes has been
estimated equal to about 0.1
m
gg
1
of candle burnt.
For measuring benzene, toluene, ethylbenzene and xylenes,
awelldefined volume of exhaust gas was sampled by means of
a gas pump and passed through a charcoal cartridge (Carbotrap
349). According to the sampling capacity of the charcoal cartridge,
a sampling flow of about 0.05 L min
1
allows for collecting a total
volume of about 12 L in 4 h. After sampling, the charcoal cartridge
was thermally desorbed and analysed with gas chromatography e
mass spectroscopy (CG/MS) equipped with Restek Rxi-5Sil-MS
chromatographic column. The average extraction yields of the
BTEX ranged from 86% to 100% and the relative standard deviations
on the corresponding measurements were less than 10%. The
detection limit for the various BTEX has been estimated equal to
about 0.01
m
gg
1
of candle burnt.
PAHs were sampled with a combined particle/gas phase system:
awelldefined volume of exhaust gas was sampled by means of a gas
pump and passed through a PTFE filter (TE 35, pore size of 0.2
m
m)
and an adsorption Tenax cartridge (Supelco XAD Orbo Tube).
Particles were collected on the PTFE filter surface, while gas phase
PAHs were sampled on the XAD2 sorbent. Both filter and cartridge
were extracted with dichloromethane coupled with a sonication
treatment of 30 min. According to the sampling capacity of both the
filter and the cartridge, a sampling flow of about 0.5 L min
1
allows
for collecting a total volume of about 120 L in 4 h. After vacuum
concentration, the desorbed solution was analysed using gas
chromatography coupled with a mass-spectrometry (GC/MS).
Considering the measurements of all the samples, extraction yields
of the PAHs were never less than 85% and in most cases almost
100%, while the relative standard deviations on the corresponding
measurements were less than 15%. The detection limit for the
Fig. 2. Aldehydes emission factors for the investigated scented candles (filled bars) and pure paraffin candles (empty bars). Dashed line represents the quantification limit.
Fig. 3. BTEX emission factors for the investigated scented candles (filled bars) and pure paraffin candles (empty bars). Dashed line represents the quantification limit.
M. Derudi et al. / Atmospheric Environment 55 (2012) 257e262 259
various PAHs has been estimated equal to about 0.01 ng g
1
of
candle burnt.
Sampling was always stopped before the blowout of the candles
so as to avoid the collection of pollutants produced during the last
transient stage of the burning process.
3. Results and discussion
As previously mentioned, four scented candles have been
burned in the test chamber for each test, leading to an average
candle burning rate of about 4.5 g h
1
for all the investigated
candles.
Apart from the beginning of the test, when a calm flame
gradually formed a cup rim surrounding the so-called burn bowl,
the flame burned without visible release of smoke and the candles
did not drip. In particular, scented candles have been burnt in
a single step, but to verify if the emissions were constantly
released few experiments have been also carried out in different
burning cycles of about 4 hours, followed by a 1 hour stop
between the cycles. No significant differences were found for the
two procedures.
Measured emissions have been calculated as emission factors
with reference to 1 g of candle burnt. These values allow a direct
comparison between different types of candles; as shown in
Figs. 2e4, emission factors can change significantly from one candle
type to another one: variation as large as one order of magnitude
have been recorded for individual pollutants.
The measured emission factors show that large differences can
be found also in similar candles; moreover, no clear correlations can
be deduced from these data, because one candle can show large
emission factor for one pollutant and small emission factor for
another one. Moreover, we can see that large emissions of BTEX are
not necessarily related to high levels of PAHs, and even within the
BTEX no clear relations exist. One candle clearly show lower levels
of BTEX and PAHs, while evidencing emission factors for aldehydes
similar to the other candles. The other candles evidenced that more
than 34% of the emitted BTEX is constituted by ethylbenzene while
xylenes formation is negligible.
For what concerns PAHs, two scented candles show emission
factors close to the detection limit of the analytical procedures,
while the other ones show not negligible emission factors (Fig. 4).
Most of the PAHs detected are constituted by 2- to 4-ring PAHs; in
particular, naphthalene (2-ring PAH), phenanthrene (3-ring PAH),
fluoranthene and pyrene (4-ring PAHs), which are known as
precursors of particulate matter, evidenced average emission
factors above the detection limit.
Finally, aldehydes emissions are quite similar for all the candles,
as expected if such emissions would be mainly related to the
presence of a fragrance rather than to the other candle parameters.
In particular, formaldehyde shows always the highest emission
factors, followed generally by acetaldehyde; on the other hand,
other short chain aldehydes do not show a clear behaviour,
evidencing several individual emission factors below the detection
limit (Fig. 2).
This has been confirmed by the experiments carried out using
candles made by pure paraffin and enclosed in a glass beaker,
whose results are also summarized in Figs. 2e4. These candles
showed an average burning rate lower than the previous one and
equal to about 2.5 g h
1
, as expected due to the presence of the
glass container that hinders the air flow towards the flame.
The results summarized in Fig. 2 clearly show that burning pure
paraffin candles does not produce any detectable amount of alde-
hydes, whose production in the commercial candles investigated
should be consequently ascribed to the presence of additives.
Moreover, a data scattering among the three paraffins investi-
gated even larger than that found for the scented candles is evident
for BTEX and PAHs emission factors. Differences as large as two
orders of magnitude can be seen from Figs. 3 and 4, especially for
2e3 ring PAHs. This seems to indicate that the kind of raw material
rather than the additives determines BTEX and PAHs emissions.
Studying the distribution of different PAHs isomers into the
exhausts it is possible to hypothesize if they are released by evap-
oration and pyrolysis phenomena at relatively low temperature,
due to the overheating of the candles constituents, or produced by
an incomplete combustion (Yunker et al., 2002; Orecchio, 2010).
Usually, values of anthracene to anthracene plus phenanthrene
Fig. 4. PAHs emission factors for the investigated scented candles (filled bars) and pure paraffin candles (empty bars). Dashed line represents the quantification limit.
Table 2
Isomeric ratios for the investigated candles (legend as in Table 1).
Isomeric ratios Scented samples Waxes Mean scented Mean waxes
ABC DEW1W2W3
Ant/(Ant þPhe) 0.19 0.33 0.07 0.27 0.25 0.95 0.63 0.83 0.22 0.80
Fla/(Fla þPyr) 0.50 0.50 0.40 0.43 0.60 0.09 0.20 0.11 0.49 0.13
B[a]A/(B[a]A þChr) 0.76 0.67 0.50 0.01 0.50 0.50 0.50 0.50 0.49 0.50
Total index 6.96 7.92 4.18 3.86 6.50 12.19 9.25 11.11 5.89 10.85
M. Derudi et al. / Atmospheric Environment 55 (2012) 257e262260
Ant/(Ant þPhe) ratio <0.10 are an index of low temperature
sources while values larger than 0.10 indicates a dominance of
combustion (Yunker et al., 2002). As reported in Table 2, most of the
investigated samples evidenced high values of this index; consid-
ering the average values, Ant/(Ant þPhe) is equal to 0.22
for scented candles and to 0.8 for the scentless ones. Concerning
Fla/(Fla þPyr) and B[a]A/(B[a]A þChr) ratios, values of 0.4 and 0.2
respectively can be assumed as common threshold between low
temperature sources and intermediate conditions or combustion
sources (liquid and solid fossil fuels). For the samples analysed in
this paper, scented candles exhibited average values equal to 0.49
for both the isomeric ratios confirming that PAHs are mainly
emitted by combustion, while the paraffinic waxes evidenced
discordant average values of Fla/(Fla þPyr) <0.4 and B[a]A/(B[a]
AþChr) >0.2. To overcome this problem, as suggested by Orecchio
(2010), a total index, defined as the sum of the above mentioned
isomeric ratios, normalized to the limit values reported in literature
(Yunker et al., 2002), has been evaluated as:
Total index ¼Ant
ðAnt þPheÞ$0:1þFla
ðFla þPyrÞ$0:4
þB½aA
ðB½aAþChrÞ$0:2(1)
When the total index is >4, PAHs are mainly originated from
high temperature processes (combustion) while lower values
indicate prevalently low temperature emissions. In this study, the
average total index is equal to 5.89 for scented candles and 10.85 for
paraffins, thus confirming that PAHs are emitted by high temper-
ature processes.
Finally, the scented candles results have been compared with
the few data concerning scented candles available in the literature
(Maupetit and Squinazi, 2009; Orecchio, 2011), as summarized in
Table 3. We can see that most of the emission factors reported in
Figs. 2e4lie inside the ranges measured in these previous works.
However, we can also see that such ranges are quite wide,
therefore confirming that emission factors can change significantly
from one candle type to another one without any self-explaining
reason. This clearly evidences the need for simple and cheap
methodologies to measure the emission factors of commercial
candles in order to foresee their health hazard level, possibly before
they can reach the market.
Emissions of few high priority pollutants have been then
compared to the corresponding ambient air quality standards; on
the basis of the determined emission rates of the candles, a simple
indoor air scenario was used to determine resulting indoor air
levels when four scented candles are burnt simultaneously in
a30m
3
room, considering an air exchange rate of 0.5 h
1
. In a well-
mixed environment, the steady-state concentration for the i-th
pollutant (C
ss
) has been computed as follows:
C
ss;i
¼n$m$EF
i
V$AER (2)
with n¼number of candles, m¼candle burnig rate, EF
i
¼emission
factor of the i-th pollutant, V¼room volume, and AER ¼air
exchange rate.
The resulting indoor air concentrations, computed for the
maximum emission factors found for formaldehyde, benzene and
benzo[a]pyrene, respectively, are summarized in Table 4, together
with the corresponding air quality standards. It can be seen that
formaldehyde and benzene concentrations are well below the
standards considered, while the concentration of benzo[a]pyrene is
about 40% of the recommended value; obviously a much more
detailed toxicological study should be performed to critically
evaluate the obtained data.
4. Conclusions
The main aim of this work has been to characterize pollutants
emission factors, with particular regard to VOC identified by the EU
as priority indoor pollutants, from the burning of scented candles.
All the experiments have been carried out using a laboratory-
scale test chamber, which allows for performing these measure-
ments in a controlled environment avoiding the large investments
required by a full-scale environmental room.
It has been found that the BTEX and PAHs emission factors show
large differences in similar candles without any clear correlations.
On the other hand, aldehydes emission factors are quite similar for
all the candles, leading to the conclusion that such emissions are
mainly related to the presence of a fragrance rather than to the
other candle parameters. This has been confirmed by the experi-
ments carried out using candles made by pure paraffin, where
almost no emissions of aldehydes have been found. Moreover,
a data scattering among the three paraffins investigated even larger
than that found for the scented candles is evident for BTEX and
PAHs emission factors. This seems to indicate that the kind of raw
material rather than the additives determines BTEX and PAHs
emissions.
A comparison with the few data concerning scented candles
available in the literature confirmed that the emission factors of the
scented candles lie inside the ranges measured in these previous
works, which are indeed quite wide.
This clearly evidences the need for simple and cheap method-
ologies to measure the emission factors of commercial candles in
order to foresee the expected pollutant concentration in a given
indoor environment and compare it with health safety standards.
In this regard, the laboratory-scale test chamber realized in this
study could be useful to perform measurements in an easy-to-build
standardized environment.
Table 4
Airborne concentrations of pollutants estimated for the considered exposure
scenario.
Pollutant Scented candles
(Max. value)
Guidance
limit
Reference
Formaldehyde,
m
gm
3
3.49 100 (World Health
Organization, 2010)
Benzene,
m
gm
3
0.15 5 (EU, 2000)
Benzo(a)pyrene, ng m
3
0.41 1 (EU, 2004)
Table 3
Comparison among emission factors measured for different scented candles.
This work
(maximum value)
Maupetit and
Squinazi, 2009
Orecchio, 2011
Formaldehyde,
m
gg
1
2.91 7.09e141.30
Acetaldehyde,
m
gg
1
1.12 1.77e8.48
Benzaldehyde,
m
gg
1
0.55 0.00e11.30
Benzene,
m
gg
1
0.13 <1
Toluene,
m
gg
1
0.23 <1
Xylenes,
m
gg
1
0.04 <1
Naphthalene, ng g
1
0.04 <2800 0.48e15.00
Fluorene, ng g
1
0.04 0.06e4.70
Phenanthrene, ng g
1
0.14 0.05e3.00
Anthracene, ng g
1
0.02 <15
Fluoranthene, ng g
1
0.07 0.005e1.20
Pyrene, ng g
1
0.10 0.005e1.00
Chrysene, ng g
1
0.75 0.007e3.00
Benzo(a)anthracene, ng g
1
0.03 0.01e0.40
Benzo(a)pyrene, ng g
1
0.33 0.01e1.40
M. Derudi et al. / Atmospheric Environment 55 (2012) 257e262 261
Acknowledgements
Financial support from Cereria Lumen srl, as well as from the
Associazione Cerai d’Italia is gratefully acknowledged.
References
Destaillats, H., Maddalena, R.L., Singer, B.C., Hodgson, A.T., McKone, T.E., 2008.
Indoor pollutants emitted by office equipment: a review of reported data and
information needs. Atmospheric Environment 42, 1371e1388.
Estrellan, C.R., Iino, F., 2010. Toxic emissions from open burning. Chemosphere 80,
193 e207.
EU, 2000. EC Directive relating to limit values for benzene and carbon monoxide in
ambient air. Official Journal of the European Communities, 12e21.
EU, 2004. Directive 2004/107/EC of the European Parliament and of the Council of
15 December 2004 relating to arsenic, cadmium, mercury, nickel and Polycyclic
Aromatic Hydrocarbons in ambient air. Official Journal of the European Union,
3e16.
Fan, C.W., Zhang, J., 2001. Characterization of emissions from portable household
combustion devices: particle size distributions, emission rates and factors, and
potential exposures. Atmospheric Environment 35, 1281e1290.
Fine, P.M., Cass, G.R., Simoneit, B.T., 1999. Characterization of fine particle emissions
from burning church candles. Environmental Science Technology 33,
2352e2362.
Gelosa, S., Derudi, M., Sliepcevich, A., Gelosa, D., Nano, G., Rota, R., 2007. Charac-
terization of pollutants emissions from burning candles. In: 30th Meeting of the
Italian Section of the Combustion Institute Ischia, Italy, pp. 1e6.
Guo, Z., Mosley, R., McBrian, J.A., Fortmann, R., 2000. Fine particulate matter
emissions from candles. In: Engineering Solutions to Indoor Air Quality Prob-
lems, Raleigh, NC, July 17e19.
He, C., Morawska, L., Hitchins, J., Gilbert, D., 2004. Contribution from indoor sources
to particle number and mass concentrations in residential houses. Atmospheric
Environment 38, 3405e3415.
Kotzias, D., Koistinen, K., Kephalopoulos, S., Schlitt, C., Carrer, P., Maroni, M.,
Jantunen, M., Cochet, C., Kirchner, S., Lindvall, T., McLaughlin, J., Mølhave, L., de
Oliveira Fernandes, E., Seifert, B., 2005. Final Report eThe INDEX project e
Critical Appraisal of the Setting and Implementation of Indoor Exposure Limits
in the EU, EUR 21590 EN. Available at: http://publications.jrc.ec.europa.eu/
repository/handle/111111111/5418.
Lau, C., Fiedler, H., Hutzinger, O., Schwind, K.H., Hosseinpour, J., 1997. Levels of
selected organic compounds in materials for candle production and human
exposure to candle emissions. Chemosphere 34, 1623e1630.
Lee, S.C., Wang, B., 2006. Characteristics of emissions of air pollutants from
mosquito coils and candles burning in a large environmental chamber.
Atmospheric Environment 40, 2128e2138.
Long, C.M., Suh, H.H., Koutrakis, P., 2000. Characterization of indoor particle sources
using continuous mass and size monitors. Journal of the Air & Waste
Management Association 50, 1236e1250.
Maupetit, F., Squinazi, F., 2009. Caractérisation des émissions de benzène et de
formaldéhyde lors de la combustion d’encens et de bougies d’intérieur: elab-
oration de scénarios d’exposition et conseils d’utilisation. Environnement,
Risques & Santé 8, 109e118 .
Nriagu, J.O., Kim, M.-J., 2000. Emissions of lead and zinc from candles with metal-
core wicks. The Science of the Total Environment 250, 37e41.
Orecchio, S., 2010. Analytical method, pattern and sources of polycyclic aromatic
hydrocarbons (PAHs) in the stone of the Temples of Agrigento (Italy). Journal of
Hazardous Materials 176, 339e347.
Orecchio, S., 2011. Polycyclic aromatic hydrocarbons (PAHs) in indoor emission from
decorative candles. Atmospheric Environment 45, 1888e1895.
Ott, W.R., Siegmann, H.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, 821e843.
Pagels, J., Wierzbicka, A., Nilsson, E., Isaxon, C., Dahl, A., Gudmundsson, A.,
Swietlicki, E., Bohgard, M., 2009. Chemical composition and mass emission
factors of candle smoke particles. Journal of Aerosol Science 40, 193e208.
Ravindra, K., Sokhi, R., van Grieken, R., 2008. Atmospheric polycyclic aromatic
hydrocarbons: source attribution, emission factors and regulation. Atmospheric
Environment 42, 2895e2921.
Sarigiannis, D.A., Karakitsios, S.P., Gotti, A., Liakos, I.L., Katsoyiannis, A., 2011.
Exposure to major volatile organic compounds and carbonyls in European
indoor environments and associated health risk. Environment International 37,
743 e765.
Sesana, G., Nano, G., Baj, A., Balestreri, S., 1991. New sampling tool for airborne
volatile aldehydes. Fresenius Journal of Analytical Chemistry 339, 485e487.
Tuckett, C.J., Holmes, P., Harrison, P.T., 1998. Airborne particles in the home. Journal
of Aerosol Science 29, 293e294.
USEPA, 2001. Candles and Incense as Potential Sources of Indoor Air Pollution:
Market Analysis and Literature Review. USEPA-600/R-01-001.
van Alphen, M., 1999. Emission testing and inhalational exposure-based risk
assessment for candles having Pb metal wick cores. The Science of the Total
Environment 243, 53e65.
Wasson, S.J., Guo, Z., McBrian, J.A., Beach, L.O., 2002. Lead in candle emissions. The
Science of the Total Environment 296, 159e174.
Zai, S., Zhen, H., Jia-song, W., 2006. Studies on the size distribution, number and
mass emission factors of candle particles characterized by modes of burning.
Journal of Aerosol Science 37, 1484e1496.
World Health Organization, 2010. WHO Guidelines for Indoor Air Quality: Selected
Pollutants. WHO Regional Office for Europe, Copenhagen, Denmark.
Yunker, M.B., Macdonald, R.W., Vingarzan, R., Mitchell, R.H., Goyette, D.,
Sylvestre, S., 2002. PAHs in the FraserRiver basin: a critical appraisal of PAH
ratios as indicators of PAH source and composition. Organic Geochemistry 33,
489e515.
M. Derudi et al. / Atmospheric Environment 55 (2012) 257e262262
