CEN/TR 16998:2016

SLOVENSKI STANDARD
01-julij-2017
=XQDQML]UDN3RURþLOR]DQLWURLQRNVL3$+,]YRUVWUXSHQRVWNRQFHQWUDFLMHLQ
PHULOQHPHWRGH
Ambient air - Report on nitro- and oxy-PAH - Origin, toxicity concentrations and
measurement methods
Außenluft - Nitro- und Oxy-PAHs - Herkunft, Toxizität, Konzentrationen und
Messverfahren
Air ambiant - Rapport concernant les HAP nitrés et les HAP oxygénés - Origine, toxicité,
concentrations et méthodes de mesure
Ta slovenski standard je istoveten z: CEN/TR 16998:2016
ICS:
13.040.20 Kakovost okoljskega zraka Ambient atmospheres
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

CEN/TR 16998
TECHNICAL REPORT
RAPPORT TECHNIQUE
November 2016
TECHNISCHER BERICHT
ICS 13.040.20
English Version
Ambient air - Report on nitro- and oxy-PAHs - Origin,
toxicity, concentrations and measurement methods
Air ambiant - Rapport sur les nitro- et oxy-HAP - Außenluft - Bericht über Nitro- und Oxy-PAHs -
Origine, toxicité, concentrations et méthodes de Herkunft, Toxizität, Konzentrationen und
mesure Messverfahren
This Technical Report was approved by CEN on 21 October 2016. It has been drawn up by the Technical Committee CEN/TC 264.

CEN members are the national standards bodies of Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia,
Finland, Former Yugoslav Republic of Macedonia, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania,
Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey and
United Kingdom.
EUROPEAN COMMITTEE FOR STANDARDIZATION
COMITÉ EUROPÉEN DE NORMALISATION

EUROPÄISCHES KOMITEE FÜR NORMUNG

CEN-CENELEC Management Centre: Avenue Marnix 17, B-1000 Brussels
© 2016 CEN All rights of exploitation in any form and by any means reserved Ref. No. CEN/TR 16998:2016 E
worldwide for CEN national Members.

Contents Page
European foreword . 3
Introduction . 4
1 Scope . 5
2 Symbols and abbreviations . 5
3 Literature overview . 5
4 Conclusions . 20
5 Recommendations . 20
Annex A (informative) Sampling and analysis by GC-MS of some nitro- and oxy-PAHs
associated to ambient particulate matter . 22
A.1 Sampling . 22
A.2 Analytical materials . 22
A.2.1 Glassware and sample handling . 22
A.2.2 Reagents and Solvents . 22
A.2.3 Extraction apparatus and materials . 23
A.2.4 Evaporation apparatus and materials . 24
A.2.5 Clean-up Material . 24
A.2.6 Weighting Apparatus . 24
A.2.7 Analytical system . 24
A.3 Extraction . 24
A.4 Clean-up . 24
A.5 Analysis . 25
A.6 Results . 25
A.7 Quality assurance . 30
Annex B (informative) Carcinogenicity and references to nitro- and oxy-PAHs . 31
Annex C (informative) Mutagenicity of nitro-PAHs . 34
Annex D (informative) Diesel exhaust data . 35
Annex E (informative) Structures of nitro- and oxy-PAHs referred in this Technical Report . 36
Bibliography . 43

European foreword
This document (CEN/TR 16998:2016) has been prepared by Technical Committee CEN/TC 264 “Air
quality”, the secretariat of which is held by DIN.
Attention is drawn to the possibility that some of the elements of this document may be the subject of
patent rights. CEN [and/or CENELEC] shall not be held responsible for identifying any or all such patent
rights.
Introduction
Nitro-PAHs and oxy-PAHs are found in ambient air samples and there are strong indications that they
are as harmful as PAHS. Several compounds are classified as probably carcinogenic for humans (see
Table in Annex A) and nitro-PAHs are reported to be strongly mutagenic. Photooxidation of volatile
PAHs gives rise to the formation of secondary aerosols (Chan et al. 2009, Kautzman et al. 2010, Shakya
and Griffin, 2010).
1-Nitropyrene and 2-nitrofluorene are discussed as marker compounds for diesel exhaust and other
combustion processes. 2-Nitropyrene and 2-nitrofluoranthene are good marker substances for the
formation of nitro-PAHs by secondary reactions.
This Technical Report presents the state of the art of the oxy- and nitro-PAHS topics.
1 Scope
This Technical Report is focused on the presence of nitro- and oxy-PAHs in ambient air. It describes
how nitro- and oxy-PAH are formed, what typical concentrations are found, what is known about their
toxicity, and what sampling and measurement techniques are available.
The conclusions of this report are that nitro- and oxy-PAHs concentrations are present in the
atmosphere in levels that are of concern regarding their high toxicity. Information on the presence of
these compounds in ambient air is as relevant as information about PAHs. Validated techniques for the
measurement of nitro- and oxy-PAHs are available.
2 Symbols and abbreviations
DNA Deoxyribonucleic acid
EI Electron ionization
CD Chemiluminescence detection
FD Fluorescence detection
GC-MS Gas chromatography – mass spectrometry
GC-NICI-MS Gas chromatography – negative ion chemical ionization – mass spectrometry
HPLC High performance liquid chromatography
HPLC-FD HPLC – fluorescence detection
HPLC-CD HPLC – chemiluminescence detection
IARC International Agency for Research on Cancer
LC Liquid chromatography
MS Mass spectrometry
NICI Negative ion chemical ionization
Nitro-PAHs Nitrated polycyclic aromatic hydrocarbons
Oxy-PAHs Oxygenated polycyclic aromatic hydrocarbons
PAHs Polycyclic aromatic hydrocarbons
SPE Solid phase extraction
ToF-MS Time of flight mass spectrometry
3 Literature overview
3.1 Nitro-PAHs
3.1.1 Sources
3.1.1.1 General
Nitro-PAHs in the atmosphere originate mainly from combustion sources and are produced from both
gas and heterogeneous phase reactions of the parent PAHs with atmospheric oxidants such as NO ,
N O , O , OH and peroxide radicals (Arey et al., 1986; Atkinson et al., 1990; Keyte et al., 2013; Pitts et al.,
2 5 3
1985; Pitts Jr et al., 1978) in the presence of nitrogen oxides.
3.1.1.2 Direct emissions
Nitro-PAHs from direct emissions are formed by high temperature electrophilic nitration of PAHs with
NO during combustion processes (Nielsen, 1984). Nitro-PAHs have been observed in vehicle exhaust
(particularly diesel), industrial emissions, waste incinerator emissions (DeMarini et al., 1996) and
emissions from domestic residential heating/cooking (Kinouchi et al., 1988; Van Houdt, 1990). Nitro-
PAHs are also emitted by wood burning but in relative low amounts due to low emissions of NO during
this type of combustion process (Alfheim and Ramdahl, 1984, Orasche et al., 2012; Orasche et al., 2013;
Shen et al., 2011; Shen et al., 2012a; Shen et al., 2012b), (Environmental Health Criteria (EHC) 229,
2003 and references therein).
Recently, nitro-PAHs have also been quantified in exhausts of modern biodiesel engines (Karavalakis et
al., 2010a; Karavalakis et al., 2010b; Karavalakis et al., 2011). Additionally, several studies have shown
the formation nitro-PAHs in situ on catalytic diesel particulate filters as they act as chemical reactors for
the nitration of PAHs (Carrara et al., 2010; Carrara and Niessner, 2011; Heeb et al., 2008). In this case,
nitro-PAHs would be considered as primarily emitted. Gasoline emissions have also been reported but
at lower concentration levels (Alsberg et al., 1985; Hayakawa et al., 1994; IARC, 1989; Sera et al., 1994).
Overall, 1-nitropyrene, 2-nitrofluorene and 2-nitrofluoranthene are the most abundant nitro-PAHs in
diesel and gasoline exhaust (gas and particulate phases) (Beije and Möller, 1988; Environmental Health
Criteria (EHC) 229, 2003; Finlayson-Pitts and Pitts Jr, 2000; Paputa-Peck et al., 1983; Schuetzle and
Perez, 1983).
3.1.1.3 Atmospheric formation
Gas-phase reactions of parent PAHs are initiated by OH radicals during the day and by NO radicals at
night in the presence of NO producing nitro-PAHs, with subsequent partitioning to or depositing on the
x
particulate matter. (Arey et al., 1986; Atkinson et al., 1989a; Atkinson et al., 1989b; Atkinson et al.,
1990; Atkinson and Arey, 1994; Environmental Health Criteria (EHC) 229, 2003; Helmig and Harger,
1994; Keyte et al., 2013; Sasaki et al., 1997; Vione et al., 2006).
Recently, research studies reported that heterogeneous reactions may be the dominant process for loss
of atmospheric PAHs and a significant source for nitro-PAHs in the atmosphere (Keyte et al., 2013;
Kwamena et al., 2007; Perraudin et al., 2007; Pöschl et al., 2001). These reactions may dramatically
differ from the homogeneous reactions in their rates, mechanisms, and products. Numerous studies
showed results obtained with model particles (soot, sea salt, organic aerosol, silica, graphite or azelaic
acid particles) coated artificially with single or a mixture of PAHs and their reaction with various
oxidants as OH, NO , O or NO (Cazaunau et al., 2010; Esteve et al., 2003; Kwamena et al., 2007; Miet et
3 3 2
al., 2009; Perraudin et al., 2005; Zhang et al., 2011).
Few studies reported results obtained with natural soot particles laboratory generated (liquid
carburant burners) (Bedjanian et al., 2010; Kwamena and Abbatt, 2008), with natural ambient air
particles (Ringuet et al., 2012b; Zimmermann et al., 2013) or with diesel engine exhaust particles
(Esteve et al., 2006; Kamens et al., 1990; Nguyen et al., 2009; Rattanavaraha et al., 2011).
Mechanistic reaction schemes for gas phase formation of nitro-derivatives of fluoranthene and
heterogeneous formation of isomeric nitro-benzo[a]pyrenes are shown in Figure 1 and 2, respectively.
Figure 1 — Oxidation mechanisms of fluoranthene by OH during the day (Arey, 1998) and by NO
during the night (Atkinson and Arey, 1997)

Figure 2 — Mechanism proposed for the nitration of benzo[a]pyrene (Cazaunau et al., 2010)
22-Nitrofluoranthene and 2-nitropyrene are the most abundant substances formed by gas phase
reaction of PAHs with oxidants and oxides of nitrogen. A high 2-nitrofluoranthene/1-nitropyrene ratio
is a good indicator for the secondary formation of nitro-PAHs (Albinet et al., 2007b; Albinet et al., 2008;
Arey et al., 1989; Atkinson and Arey, 1994; Bamford and Baker, 2003; Reisen and Arey, 2005; Mariano
et al., 2000; Ringuet et al., 2012a; Ringuet et al., 2012c; Zielinska et al., 1989; Zimmermann et al., 2012).
3.1.2 Concentrations, gas/particle partitioning and size distribution
Overall, in continental areas (urban, sub-urban and rural areas), nitro-PAHs atmospheric
concentrations are one or two orders lower than PAHs atmospheric concentrations. Nitro-PAHs
–3
concentrations are in the range of 0,1 to 1000 pg·m in both, gaseous and particulate phases (e.g.
Albinet et al., 2007; Albinet et al., 2008a; Bamford and Baker, 2003; Ciccioli et al., 1995; Ciccioli et al.,
1996; Feilberg et al., 2001; Feilberg and Nielsen, 2001; Hayakawa et al., 1995a; Hayakawa et al., 2002;
Maria del Rosario Sienra, 2006; Valle-Hernandez et al., 2010; Wang et al., 2011). Nitronaphthalene
isomers in gas phase and 2-nitrofluoranthene and 9-nitroanthracene in particulate phase are generally
the most abundant nitro-PAHs and account for about 15 % to 50 % of the total nitro-PAHs
concentrations.
Table 1 summarizes the reported concentration ranges of the most important nitro-PAHs.
Table 1 — List of concentration ranges of important nitro-PAHs in ambient air
Substance Concentration range ng/m
Traffic Urban Rural/remote
1-Nitronaphthalene 0,07 – 0,2 (n, o) 0,2 (h, n) 0,01 – 0,2 (h, n)
2-Nitronaphthalene 0,03 – 0,06 (n, o) 0,12 (h, n) 0,01 – 0,1 (h, n)
2-Nitrofluorene 0,001 – 0,021 (l, n) 0,05 – 0,4 (d, g, h, n) 0,001 – 0,005 (h, n)
9-Nitroanthracene 0,002 – 0,01 (n, o) 0,03 – 0,2 (a, c, d, g, h, j, l, n) 0,002 – 0,03 (g, h, n)
3-Nitrophenanthrene 0,007 – 0,1 (n, o) 0,001 – 0,02 (l, o) 0,0007 – 0,001 (o)
9-Nitrophenanthrene 0,005 – 0,05 (l, n) 0,01 – 0,3 (h, i, n) 0,0002 – 0,03 (h, n)
2-Nitrofluoranthene 0,03 – 0,2 (l, o) 0,03 – 2 (a, e, f, g, j, l, m, n) 0,02 – 0,03 (e, k, n)
3-Nitrofluoranthene 0,018 (i) 0,003 – 0,1 (d, j, l) 0,01 (e)
1-Nitropyrene 0,02 – 0,2 (l, o) 0,01 – 2 (a, b, e, f, g, h, j, l, n) 0,0006 – 0,01 (e, h, n, o)
2-Nitropyrene 0,007 – 0,2 (n, o) 0,01 – 0,04 (a, h, j, n) 0,001 – 0,08 (e, h, k, n)
4-Nitropyrene 0,02 – 0,03 (o) 0,001 (h, n) 0,0006 (h, n)
1,3-Dinitropyrene 0,0009 – 0,02 (n, o) 0,01 – 0,03 (d, h, n) 0,004 (n)
1,6-Dinitropyrene up to 0,0002 (o) 0,01 (d, h, n) 0,0001 – 0,004 (n)
7-Nitrobenzo[a]anthracene 0,005 – 0,01 (n, o) 0,004 – 0,3 (a, h, l, m) 0,002 (h, n)
6-Nitrochrysene 0,003 – 0,004 (n, o) up to 1,5 (b, h, j, l, n) 0,0003 – 0,002 (e, h, n)
6-Nitrobenzo[a]pyrene 0,007 – 0,01 (l, n) 0,001 – 0,01 (h, n) 0,0002 – 0,005 (n, o)
3-Nitrobenzanthrone 0,001 – 0,01 (l) – –
a) Atkinson et al. (1988) Glendora (USA)
b) Matsushita and Ida (1986) Tokio (Japan)
c) Hunt and Meisel (1995) Fresno (USA)
d) Tokiwa et al. (1990a) Sapporo (Japan)
e) Vasconcellos et al. (1998) Alta Floresta (Brazil)
f) Wilson et al. (1995) Houston (USA)
g) Berlincioni et al. (1995) Florence (Italy)
h) Albinet et al. (2007) Marseille (France)
i) Mücke et al. (2009) Munich (Germany)
j) Di Filippo et al. (2010) Rome (Italy)
k) Tsapakis et al. (2007) Finokalia (Greece)
l) Valle-Hernandez et al. (2010) Mexico City (Mexico)
m) Wang et al. (2011) Bejing (China)
n) Albinet et al. (2008a, 2008b) Chamonix, Maurienne (France)
o) Ringuet et al. (2012) Paris (France)
Nitro-PAHs gas/particle partitioning is poorly documented (Albinet et al., 2007; Albinet et al., 2008a;
Araki et al., 2009; Atkinson and Arey, 1994; Bamford and Baker, 2003; Dimashki et al., 2000; Huang et
al., 2014; Reisen and Arey, 2005; Wilson et al., 1995). 2-Rings nitro-PAHs (nitronaphthalenes) are
mainly associated to the gaseous phase. Nitro-PAHs with 4 or more rings are mainly bound to particles.
3-ring nitro-PAHs are partitioned in both gaseous and particulate phase. Due to their relative low
−4
vapour pressures (<10 Pa at 20 °C, Yaffe et al., 2001), nitro-PAHs resulting from gas phase reactions
condense immediately to ambient particles (Fan et al., 1995). Gas/particle partition is depending on
their vapour pressure and the ambient conditions as temperature but also on their origin (primary or
secondary) (Albinet et al., 2007; Albinet et al., 2008a; Wilson et al., 1995).
Very few papers showed results about the particle size distribution of nitro-PAHs in ambient air
(Albinet et al., 2008b; Cecinato et al., 1999; Di Filippo et al., 2010; Hayakawa et al., 1995a; Hayakawa et
al., 1995b; Jinhui and Lee, 2000; Kawanaka et al., 2004; Kawanaka et al., 2008; Ringuet et al., 2012a;
Teixeira et al., 2011). Overall, nitro-PAHs are mainly associated (>90 %) to the fine particle fraction
(D < 1 µm) and about 20 % are associated to the ultrafine particle fraction (D < 0,1 µm). These results
p p
are important information regarding the risk assessment because nitro-PAHs can thus penetrate deeply
into the lung.
Nitro-PAHs react with hydroxide and nitrate radicals, with ozone and they are decomposed by
photolysis. As a result the atmospheric half life time of nitro-PAHs ranges from less than an hour to
several days, depending on atmospheric conditions like temperature, sunlight intensity, on their
structure and on the concentrations of reactive compounds in the air (Keyte et al. 2013).

Key
Winter
Summer
X Molecular weight in g/mol
Y Fraction in particulate phase
Figure 3 — Nitro-PAHs gas/particle partitioning according to their molecular weight
(Albinet et al., 2008a)
Figure 4 — Particle size distribution of 4 nitro-PAHs on a traffic (a – d) and suburban (e – h) site
in the Paris region (France) (Ringuet et al., 2012c). a, e: 1-nitronaphthalene; b, f: 9-
nitroanthracene; c, g: 2+3-nitrofluoranthene; d, h: 1-nitropyrene
3.1.3 Toxicity/mutagenicity
As they act as direct mutagens the mutagenic potential of nitro-PAHs can be 100 000 times greater than
that of PAHs (Durant et al., 1996; Durant et al., 1998; Enya et al., 1997; Hannigan et al., 1998; Lewtas et
al., 1990; Schuetzle, 1983; Landvik et al., 2007; Øvrevik et al., 2010). Four-ring nitro-PAHs seem to be
the most toxic substances (Durant et al., 1996; Durant et al., 1998; Finlayson-Pitts and Pitts Jr, 1986).
Overall, results from mutagenicity tests on bacteria (Ames test, Salmonella typhimurium) and on human
cells (h1A1v2 cells) showed that 3,6-dinitrobenzo[a]pyrene, 3,7-dinitrofluoranthene, 3,9-
dinitrofluoranthene, 6-nitrochrysene, 1- and 4-nitropyrene, 1,6-dinitropyrene and 1,8-dinitropyrene
(the most powerful mutagens described in the literature) had the highest mutagenic activities
(Environmental Health Criteria (EHC) 229, 2003; NTP, 2011; Enya et al., 1997; Pedersen et al., 2004;
Pedersen et al., 2005).
Nitro-PAHs contributions to the mutagenic and/or carcinogenic activity of atmospheric inhalable
particles were evaluated in the range 14 % to 50 % by different authors (Albinet et al., 2008a; Bandowe
et al., 2014; Finlayson-Pitts and Pitts Jr, 2000; Kawanaka et al., 2008; Taga et al., 2005).
Substances with a coplanar nitro group are more carcinogenic than those with a perpendicular nitro-
group. Dinitro-PAHs generally are more mutagenic than the mono-substituted compounds.
3.1.4 Carcinogenicity
Since the 1960s, evidence has increasingly supported the theory that chemical carcinogens (e.g. PAHs
and nitro-PAHs) are metabolized via oxidative pathways to produce electrophilic reactive
intermediates (e.g. nitrenium ions and epoxides) that react covalently with DNA and possibly with other
cellular nucleophiles. Nitro-PAHs seem to be less carcinogenic than their parent PAHs (EHC 229, 2003;
IARC, 2013; NTP, 2011; Benbrahim-Talaa et al., 2012). Different pathways seem to be possible for
carcinogenic activity. If the nitro-PAHs contain a “bay region” similar to benzo[a]pyrene and
benzo[a]anthracene their carcinogenicity is similar to the mechanisms described for the non-
substituted substances: After formation of an dihydroepoxide in the “bay region” this reactive
intermediate forms DNA adducts which are considered to be the first step causing carcinogenicity. If the
nitro group of these substances is oriented parallel to the aromatic core of the substances, their
carcinogenicity is not much lower than that of the non-substituted substances, but a perpendicular
orientation of the nitro group largely attenuates the carcinogenicity of the compounds (Fu et al., 1994;
Fu et al., 1998; Vogt et al., 2009; McDonald et al., 2004).
For some nitro-PAHs another pathway of carcinogenic action is postulated (Fu et al., 1994): The nitro
group is partly reduced to a hydroxylamine derivative which after esterification (e.g. acetylation) forms
a nitrenium ion, which reacts with DNA to form an adduct, which can cause carcinogenicity.
Additionally nitro-oxy-PAHs seem to be more toxic than oxy-PAHs or nitro-PAHs (Helmig et al., 1992a;
Helmig et al., 1992b). For example, 3-nitrobenzanthrone is described as one of the the most potent
mutagens and a potential carcinogen identified in diesel exhaust and ambient particulate matter (Arlt,
2005; Enya et al., 1997; Feilberg et al., 2002; Nagy et al., 2005; Phousongphouang and Arey, 2003). A list
of the classification of the carcinogenicity of several nitro- and oxy-PAHs is given in Annex B.
3.1.5 Measurement
3.1.5.1 Sampling from ambient air
Nitro-PAHs with 4 rings or more are particle bound. For these substances sampling procedures for
particulate matter (e.g. PM and PM , as described in EN 12341) are suitable. In order to collect more
10 2,5
volatile compounds, a combination of a filter with, e.g. PUF as described in ISO 12884 is necessary.
Reaction of PAHs with oxidants in combination with nitrous oxides may lead to positive artefacts during
sampling. Rearrangements of the nitro groups of the compounds during sampling are also possible.
The formation of nitro-PAHs via heterogeneous reactions with only nitrogen oxides has been shown to
be unfounded in case of ambient air sampling (Arey et al., 1988; Dimashki et al., 2000). Only at elevated
temperatures and extremely high concentrations of NO direct nitration of PAHs is possible (Carrara et
al., 2010; Carrara et al., 2011).
Studies delivered different results about the reactions of nitro-PAHs during sampling, possibly leading
to artefacts, but until now no clear results about reactions of nitro-PAHs during sampling and about
methods to inhibit these reactions have been published.
3.1.5.2 Analysis
A review paper on the analysis of nitro-PAHs in environmental samples was proposed by Zielinska and
Samy (Zielinska and Samy, 2006). The analytical procedure of collected samples includes an extraction
step prior to nitro-PAHs quantification. Several extraction procedures are reported (Soxhlet,
microwaves, pressurized liquid extraction, sonication) using different solvents or solvent mixtures
(dichloromethane, hexane, toluene, acetone). A purification step (e.g. SPE (solid phase extraction)) shall
also be included. Alternatively to solvent extraction methods, SFE (supercritical fluid extraction) using
pure CO is also reported in numerous studies (Castells et al., 2003; Lewis et al., 1995). The use of
solvent-free extraction techniques was also reported in different papers with thermal-desorption (TD)
coupled with GC-ToF-MS or GC × GC-ToF-MS (Fushimi et al., 2012; Orasche et al., 2011) and laser
desorption/ionization coupled to ToF-MS (LD-LI-ToF-MS) (Dotter et al., 1996). Finally, as an alternative
to traditional procedures, recently Albinet et al. (2014) reported the use of QuEChERS-like (Quick Easy
Cheap Effective Rugged and Safe) extraction approach for the analysis of nitrated and oxygenated PAHs.
Analysis of nitro-PAHs is generally achieved using GC-MS, GC-NICI-MS, HPLC-FD (fluorescence
detection), HPLC-CD (chemiluminescence detection), LC-MS, LC-MS-MS.
Because of its great sensitivity and selectivity towards the nitro group, GC-NICI-MS using methane as
the reactant gas minimizes the analytical interferences from co-eluted compounds by significantly
improving signal-to-noise ratios (Bezabeh et al., 2003). Relative to GC-MS in EI ionization mode, a
sensitivity improvement approaching two orders of magnitude could be obtained. Limits of detection in
the lower picogram and femtogram ranges have been reported for this method (Albinet et al., 2006;
Bezabeh et al., 2003). Actually, this analytical technique constitutes probably the best cost/performance
compromise for the analysis of nitro-PAHs in ambient air samples. Maintenance requirement are mainly
linked to the fouling of the MS source (due to the use of reagent gas for ionization) inducing a loss of
sensitivity, but progress has been made enhancing the lifetime of the MS source.
HPLC-FD after reduction of the compounds to the corresponding amines or HPLC-CD is also widely used
notably by Japanese research teams (e.g. Hayakawa et al., 1999; Hayakawa, 2000; Kawanaka et al.,
2008; Nassar et al., 2011; Ohno et al., 2009; Tang et al., 2005, see also references in Zielinska and Samy,
2006). Greater precision and selectivity is obtained by increased automation of these kinds of analytical
techniques. Limits of detection in the range of 1 pg to 10 pg injected have been reported for this method
(Hayakawa et al., 1999; Hayakawa, 2000; Kawanaka et al., 2008; Ohno et al., 2009; Schauer et al., 2003;
Tang et al., 2005; Zielinska and Samy, 2006 and references therein). The complexity and maintenance
requirements of these kinds of systems (i.e. time automated switching valves, multiple plumbing
components, and consumables), does implicate a need for highly specialized procedures with
customized components and protocols.
The use of LC-MS and LC-MS-MS is also reported in several papers (e.g. Mirivel et al., 2010; Schauer et
al., 2004; Zielinska and Samy, 2006 and references therein) but the sensitivity of this technique is still
not optimal for nitro-PAHs analysis. Detection limits are 3 to 100 higher (5 pg to 100 pg injected) than
those reported for GC-NICI-MS analytical systems. Only the possibilities of unknown species are a
significant advantage of this method but the investment and working costs are really higher than for the
other analytical techniques.
An example of an SOP dealing with sampling and analysis of nitro- and oxy-PAHs are given in Annex A.
3.2 Oxy-PAHs
3.2.1 Sources
3.2.1.1 General
Oxy-PAHs (ketones, aldehydes, hydroxy-PAHs) are both directly emitted from combustion processes
and formed in the atmosphere as by-products of the photolysis of parent PAHs or photochemical
reaction between parent PAHs and atmospheric oxidants (Vione et al., 2004; Yu, 2002).
3.2.1.2 Direct emissions
Major direct sources are diesel and gasoline combustions, biomass burning, waste combustion, coal and
fuel burning, and production of charcoal (Abas et al., 1995; Akimoto et al., 1997; Cho et al., 2004;
Choudhury, 1982; Leoz-Garziandia et al., 2000; Levsen, 1988; Orasche et al., 2012; Orasche et al. 2013;
Ramdhal, 1983; Schauer et al., 1999, 2001, 2002; Schulze et al., 1984; Shen et al., 2011; Shen et al.,
2012a; Shen et al., 2012b; Walgraeve et al., 2010 and references therein).
3.2.1.3 Atmospheric formation
Oxy-PAHs can be formed from PAHs via photochemical reactions and reactions with O , OH and NO
3 3
radicals Keyte et al., 2013). Given the partitioning of PAHs between the gaseous and particulate phases,
transformation processes can take place in both phases. Reactions involving OH are considered to play a
major role in gas phase reactions (Barbas et al., 1996; Bunce et al., 1997; Calvert et al., 2002; Vione et al.,
2004; Wang et al., 2007). Additionally, several studies show the formation of Oxy-PAHs via
heterogeneous reaction processes (Perraudin et al., 2007; Pöschl et al., 2001; Ringuet et al., 2012b). For
example, major identified products of the reaction between ozone and anthracene adsorbed on silica
particles were 1,1’-biphenyl-2,2’-dicarboxaldehyde, anthrone and 9,10-anthraquinone (Perraudin,
2004). Nevertheless, particle associated PAHs containing more than five rings are less susceptible to
reaction with gaseous reactive radicals. Nevertheless, they can also undergo a wide variety of
transformation processes, but often with slower kinetics than in the gas phase. Especially direct
photolysis and photolysis in the presence of photosensitizers (e.g. ketones or aromatic carbonyls) play a
significant role for these compounds (Yu, 2002). The wide electron delocalization of PAHs enables them
to absorb sunlight, so that irradiation under atmospheric conditions may lead to photooxidation (Vione
et al., 2004; Yu, 2002).
Examples of reaction pathways of benzo[a]pyrene with ozone and the photooxidation of anthracene are
presented in Figures 5 and 6.
Figure 5 — Reaction pathways of benzo[a]pyrene with ozone (Vione et al., 2004)

Figure 6 — Photooxidation of anthracene in aqueous solution (Yu, 2002)
3.2.2 Concentrations, gas/particle partitioning and size distribution
Oxy-PAHs atmospheric concentrations are in the same order as PAHs atmospheric concentrations in
–3
continental areas (urban, sub-urban and rural areas). Concentrations range from 10 to 10000 pg·m in
both, gaseous and particulate phases (e.g. Albinet et al., 2007; Albinet et al., 2008a; Allen et al., 1997;
Barrado et al., 2012a; Barrado et al., 2012b; Barrado et al., 2012c; Eiguren-Fernandez et al., 2008a;
Eiguren-Fernandez et al., 2008b; Enya et al., 1997; Feilberg et al., 2002; Fraser et al., 1998; Hawthorne
et al., 1992; Kojima et al., 2010; König et al., 1983; Ligocki and Pankow, 1989; Liu et al., 2006; Niederer,
1998; Schnelle-Kreis et al., 2001; Wang et al., 2011; Wilson et al., 1995; Yassaa et al., 2001). Overall,
ketones and quinones (specifically 9-fluorenone, 9,10-anthraquinone, 9,10-phenanthraquinone,
benzanthrone) are the most abundant oxy-PAHs in ambient air. Hydroxy-PAHs were detected at lower
–3 –3
concentration levels (10 pg·m to 500 pg·m ) (Barrado et al., 2012a; Barrado et al., 2012b; Barrado et
al., 2012c; Simoneit et al., 2007).
Oxy-PAHs gas/particle partitioning is poorly documented (Alam et al., 2013; Albinet et al., 2007; Albinet
et al., 2008a; Eiguren-Fernandez et al., 2008b; Fraser et al., 1998; Huang et al., 2014; Leoz-Garziandia et
al., 2000; Wilson et al., 1995). The fraction of oxy-PAHs associated with the particle phase is strongly
dependent on the molecular weight and on the ambient conditions (temperature). The lightest
compounds (number of aromatic rings ≤ 2) are detected mainly in the gas phase whereas the
compounds with a number of aromatic rings ≥ 4 are detected in the particle phase; three-ring
compounds can be found in both phases (Albinet et al., 2007; Albinet et al., 2008a).
Oxy-PAHs particle size distribution in ambient air was only reported by few authors (Albinet et al.,
2008b; Allen et al., 1997; Allen, 1997; Ladji et al., 2009; Ringuet et al., 2012a). Oxy-PAHs are mainly
associated (>90 %) to the fine particle fraction (D < 1 µm) that penetrates deeper into the organism.
p
Table 2 — List of concentration ranges of important oxy-PAHs in ambient air
Substance Traffic Urban Rural/remote
3 3 3
ng/m ng/m ng/m
1-Naphthaldehyde 0,006 – 7 (a, b, d, n) up to 2 (b, c) 0,15 – 0,6 (d)
Naphthalene-1,2-dione – up to 1,1 (e, f) –
Naphthalene-1,4-dione – up to 4 (c, e, f, g, h) –
Benzophenone – 0,2 – 0,5 (j, h) –
9H-Fluoren-9-one 0,07 – 3,6 (a, b, d, n) 0,01 – 4 (b, c, h, k) 0,2 – 11 (c, d)
1H-Phenalen-1-one – 0,4 – 2,5 (i, l, m, q) 0,03 (i)
10H-Anthracen-9-one (Anthrone) – 0,3 – 0,8 (h, i) –
Anthracene-9,10-dione 0,6 – 3,6 (d, n) 0,01 – 3,6 (d, i, k, m) 0,18 (d)
Phenanthrene-9-carboxaldehyde 0,1 – 7,2 (a, b, d, n) 0,3 (b, h) 0,01 – 0,5 (c, d)
11H-Benzo[a]fluoren-11-one 0,16 – 1,6 (b, d, n) 0,2 – 2,7 (b, i) 0,02 – 0,03 (c, d)
11H-Benzo[b]fluoren-11-one 0,15 – 1,5 (b, d, n) 0,2 – 1,7 (b, i) 0,01 – 0,5 (c, d)
7H-Benz[de]anthracen-7-one 0,05 – 2 (b, d, n) up to 2,6 (b, o, p, q) 0,02 – 1,7 (c, d)
(Benzanthrone)
Benzo[a]anthracene-7,12-dione 0,04 – 0,55 (d, n) 0,1 – 1,3 (b, e, h, l, q) 0,01 – 0,6 (c, d)
Chrysene-1,4-dione – up to 0,5 (e) –
Chrysene-5,6-dione – 0,001 – 0,8 (r) –
Pyrene-1-carboxaldehyde – 2,6 – 4,7 (h) –
Pyrene-4,5-dione – 0,3 – 0,7 (e) –
Cyclopenta[def]phenanthren-4-one – 0,01 – 0,3 (l) –
6H-benzo[cd]pyren-6-one – up to 2 (h, l) –
Benzo[a]pyrene-1,6-dione – 0,001 – 0,08 (r) –
Benzo[a]pyrene-3,6-dione – 0,001 – 0,08 (r) –
Benzo[a]pyrene-6,12-dione – 0,001 – 0,08 (r) –
a) Oda et al. (2001)  Kurashaki (Japan)
b) Albinet et al. (2007a)  Marseille (France)
c) Bandowe et al. (2014)  Xián (China)
d) Albinet et al. (2008a)  Chamonix, Maurienne (France)
e) Chung et al. (2006)  Fresno (USA)
f) Valavanidis et al. (2006) Athens (Greece)
g) Tsapakis et al. (2002)  Santiago, Temuco (Chile)
h) Sienra (2006)  Santiago (Chile)
i) Delhomme et al. (2008) Tempe (USA)
j) Neusuess et al. (2000)  Melpitz (Germany)
k) Castells et al. (2003)  Barcelona (Spain)
l) Schnelle Kreis et al. (2005) Munich (Germany)
m) Liu et al. (2006)  Augsburg (Germany)
n) Ringuet et al. (2012a)  Paris (France)
o) Yassaa et al. (2001)  Algier (Algeria)
p) Sklorz et al. (2007a)  Munich (Germany)
q) Park et al. (2000)  Seoul (Korea)
r) Lintelmann et al. (2006) Munich (Germany)

Key
Winter
Summer
X Molecular weight in g/mol
Y Fraction in particulate phase
Figure 7 — Oxy-PAHs gas/particle partitioning according to their molecular weight
(Albinet et al., 2008a)
Figure 8 — Particle size distribution of 4 oxy-PAHs on a traffic (a – d) and suburban (e – h) site in
the Paris region (France) (Ringuet et al., 2012c). a, e: 9,10-Anthraquinone; b, f:
Benzo[a]fluorenone; c, g: Benzo[b]fluorenone; d, h: Benz[a]anthracene-7,12-dione
a)
b)
Key
1a, 1b PAHs Chamonix valley 0,01 µm – 0,39 µm

1c, 1d PAHs Maurienne valley 0,39 µm – 1,3 µm

2a, 2b oxy-PAHs Chamonix valley 1,3 µm – 4,2 µm

2c, 2d oxy-PAHs Maurienne valley 4,2 µm – 50 µm

3a, 3b nitro-PAHs Chamonix valley
3c, 3d nitro-PAHs Maurienne valley
Y Fraction in size class
Figure 9 — Average PAHs, oxy-PAHs and nitro-PAHs concentration fractions in the different
particle size classes for different sampling site typologies located in the French Alpine valleys
(upper figure: winter 2002/2003, lower figure: summer 2003). The error bars show the
standard deviation from the weekly average of the sampling campaign (n = 7) (Albinet et al.,
2008b)
3.2.3 Toxicity
Only few toxicological impacts of oxy-PAHs are documented. Studies mainly focus on compounds used
for industrial or medical applications. The mechanisms underlying their toxicity are complex and far
from fully understood (Lundstedt et al., 2007; Walgraeve et al., 2010 and references therein; Xia et al.,
2004). Data on mutagenic effects mainly resulted from tests on bacteria (Ames test) and human cells
(h1A1v2) (Durant et al., 1996; Durant et al., 1998; Pedersen et al., 2004; Pedersen et al., 2005). With
respect to their human health effects, oxy-PAHs are considered to be more acute-toxic than their parent
PAHs because of their direct mutagenic potency, whereas PAHs require first an enzymatic activation
(Durant et al., 1996; Durant et al., 1998; Pedersen et al., 2004; Pedersen et al., 2005; Yu, 2002). Among
the studied oxy-PAHs, ketones, quinones and three-ring oxy-PAHs seemed mutagens (e.g.
benzanthrone, benzo[a]fluorenone, benzo[b]fluorenone). Tests on human cells showed that 6H-
benzo[cd]pyren-6-one and anthanthrenequinone seemed highly mutagenic (Pedersen et al., 2004;
Pedersen et al., 2005). 9,10-Anthraquinone is classified as possibly carcinogenic to humans (2B, IARC,
2012).
Some nitro-oxy-PAHs are known as strong mutagenes such as 3-nitrobenzanthrone.
Nitro-oxy-PAHs seem to be more toxic than nitro-PAHs or oxy-PAHs (Helmig et al., 1992a; Helmig et al.,
1992b). As for example, 3-nitrobenzanthrone was described as one of the most potent mutagens and a
potential carcinogen identified in diesel exhaust and ambient particulate matter (Arlt, 2005; Enya et al.,
1997; Feilberg et al., 2002; Nagy et al., 2005; Phousongphouang and Arey, 2003). However, 3-
nitrobenzanthrone itself undergoes extensive atmospheric rearrangements to the less mutagenic
isomer 2-nitrobenzanthrone. Thus the level of 3-nitrobenzanthrone is highest near the source of
emission, but only relatively low in general ambient air (Phousongphouang and Arey, 2003). A list of
the classification of the carcinogenicity of several nitro- and oxy-PAHs is given in Annex B.
3.2.4 Measurement
A recent review on the analysis of oxy-PAHs in environmental samples was published by Walgraeve et
al. (Walgraeve et al., 2010). As for nitro-PAHs, oxy-PAHs analytical procedures are generally solvent
based methods including extraction and purification steps. Various extraction methods have been
reported (Soxhlet extraction, microwave extraction, pressurized liquid extraction, sonication) with
different solvents or solvent mixtures (dichloromethane, methanol, acetone, hexane).
The use of thermal desorption (TD) methods for extraction, coupled to chromatographic methods for
separation and analysis of oxy-PAHs, was also reported in the literature. Advantages are reduced
extraction times, the elimination of solvent use, which fits within the concept of green analytical
chemistry and the elimination of the sample preparation steps (Orasche et al., 2011; Fushimi et al.,
2012). Recently Albinet et al. (2014) report the use of QuEChERS-like extraction approach as an
alternative procedure for the analysis of nitrated and oxygenated PAHs.
Analysis is generally achieved using GC-MS (e.g. Allen et al., 1997; Kojima et al., 2010), GC-NICI-MS, LC-
MS or LC-MS-MS analytical systems) and includes a purification step on solid phase extraction or by
HPLC. HPLC-Fluorescence detection analytical procedures are also reported in the literature (e.g.
Barrado et al., 2008). For gas chromatography methods, higher sensibility and selectivity is obtained
using negative chemical ionization in mass spectrometry (GC-NICI-MS) (Albinet et al., 2006; Wang et al.,
2011). Femtograms injected (0,03 pg to 0,07 pg) have been reported as limits of detection for this
method (Albinet et al., 2006). For GC-MS using electron impact ionization, a derivatization step (e.g.
using BSTFA reagent (N,O-bis-trimethylsilyltrifluoroacetamide)) could be included in order to improve
the sensitivity and the separation of hydroxy-PAHs (Simoneit et al., 2007). Identification of unknown
species could be achieved using GC-MS in electron impact mode (with or without derivatization step
procedure) while it is not possible using a soft ionization technique as chemical ionization. Compared to
GC-MS techniques, LC-MS and LC-MS-MS offer several advantages, especially if the target compounds
are thermally labile, have low vapour pressures or have highly polar functional groups (e.g. Delhomme
et al., 2008; Mirivel et al., 2010; Schauer et al., 2004; Walgraeve et al., 2012). Limits of detection
reported for these methods are 10 to 1000 higher than those for GC-NICI-MS analytical technique
(10 pg to 1500 pg injected). As for Nitro-PAHs, the possibilities of unknown species are an advantage of
these methods but the investment cost is quite significant. Finally, as for nitro-PAHs, GC-NICI-MS
analytical technique constitutes probably the best compromise cost/performance for the analysis of
oxy-PAHs in ambient air samples.
An example of an SOP for the analysis of nitro- and oxy-PAHs is given in Annex A.
4 Conclusions
Nitro-PAHs are about as carcinogenic as their parent PAHs. They are also known as strong mutagens.
The concentrations of nitro-PAHs in ambient air are a magnitude lower than those of the parent PAHs. A
lot of data for ambient air concentrations of these compounds have been published (see Tables 1 and 2),
but adequate systematic long-term studies at different types of stations (traffic, urban background,
rural) are still missing.
Nitro-PAHs can be directly formed by combustion processes, 1-nitropyrene and 2-nitrofluorene being
the most abundant compounds. Thus these two compounds are good markers for diesel exhaust and
diesel soot in PM (see 3.1.1.2). Emission factors for these substances in diesel exhaust and in other
combustion aerosols should be established.
The nitration of three- and four-ring PAHs by oxidative agents in the presence of nitrogen oxides is an
important reaction for the formation of secondary atmospheric aerosol. 2-Nitrofluoranthene and 2-
nitropyrene are the most abundant substances formed by gas phase reaction of PAHs with oxidants and
oxides of nitrogen.
The main nitro-PAHs formed by combustion processes and secondary reactions are different. Therefore
the ratios 1-nitropyrene/2-nitrofluoranthene and 1-nitropyrene/2-nitropyrene are a good measure for
the relation of primary versus secondary aerosols. 2-Nitrofluoranthene and 2-nitropyrene are good
tracer sub
...