![Source profile by unique ratio (SPUR) method applied to the InPy/(InPy + B[ghi]Pe) ratio considering ZK site data.](https://www.researchgate.net/profile/Marc-Fadel/publication/343845108/figure/fig3/AS:932212732088321@1599268243282/Source-profile-by-unique-ratio-SPUR-method-applied-to-the-InPy-InPy-BghiPe-ratio_Q320.jpg)
Content uploaded by Marc Fadel
Author content
All content in this area was uploaded by Marc Fadel on Sep 04, 2020
Content may be subject to copyright.
journal of environmental sciences 101 (2021) 98–116
Available online at www.sciencedirect.com
w w w . e l s e v i e r . c o m / l o c a t e / j e s
PM
2.5 characterization of primary and secondary
organic aerosols in two urban-industrial areas in
the East Mediterranean
Marc Fadel
1 , 2
, Frédéric Ledoux
2
, Mariana Farhat
1
, Adib Kfoury
3
,
Dominique Courcot
2
, Charbel Af
1 , 4 , ∗
1
Emissions, Measurements, and Modeling of the Atmosphere (EMMA) Laboratory, CAR, Faculty of Sciences, Saint
Joseph University, Beirut, Lebanon
2
Unité de Chimie Environnementale et Interactions sur le Vivant, UCEIV UR4492, FR CNRS 3417, University of
Littoral Côte d
Opale (ULCO), Dunkerque, France
3
Department of Environmental Sciences, University of Balamand, Al Kourah, Lebanon
4
Climate and Atmosphere Research Center, The Cyprus Institute, Nicosia, Cyprus
Article history:
Received 19 May 2020
Revised 31 July 2020
Accepted 31 July 2020
Keywords:
PM
2.5
Secondary organic aerosols
PAH s
Phthalates
Lebanon
Urban-industrial sites
Primary and secondary organic aerosols in PM
2.5
were investigated over a one-year cam-
paign at Zouk Mikael and Fiaa, Lebanon. The n-alkanes concentrations were quite similar
at both sites (26-29 ng/m
3
) and mainly explained by anthropogenic emissions rather than
natural ones. The concentrations of total Polycyclic Aromatic Hydrocarbons (PAHs) were
nearly three times higher at Zouk Mikael (2.56 ng/m
3
) compared to Fiaa (0.95 ng/m
3
), espe-
cially for indeno[1,2,3-c,d]pyrene linked to the presence of the power plant. A characteristic
indeno[1,2,3-c,d]pyrene/(indeno[1,2,3-c,d]pyrene + benzo[g,h,i]perylene) ratio in the range
0.8-1.0 was determined for heavy fuel oil combustion from the power plant. Fatty acids and
hopanes were also investigated and were assigned to cooking activities and vehicular emis-
sions respectively. Phthalates were identied for the rst time in Lebanon with high con-
centrations at Zouk and Fiaa (106.88 and 97.68 ng/m
3
respectively). Moreover, the biogenic
secondary aerosols revealed higher concentrations in summer. The total terpene concen-
tration varied between 131 ng/m
3
at Zouk Mikael in winter to 469 ng/m
3
at Fiaa in sum-
mer. Additionnally, the concentrations of the dicarboxylic acids especially for adipic and
phthalic acids were more inuenced by anthropogenic sources.The analysis of molecular
markers and diagnostic ratios indicated that the sites were strongly affected by anthro-
pogenic sources such as waste open burning, diesel private generators, cooking activities,
road transport, power plant, and industrial emissions. Moreover, results showed different
pattern during winter and summer seasons. Whereas, higher concentrations of biogenic
markers were clearly encountered during the summer period.
© 2020 The Research Center for Eco-Environmental Sciences, Chinese Academy of
Sciences. Published by Elsevier B.V.
∗Corresponding author.
E-mail: charbel.af@usj.edu.lb (C. Af).
https://doi.org/10.1016/j.jes.2020.07.030
1001-0742/© 2020 The Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences. Published by Elsevier B.V.
journal of environmental sciences 101 (2021) 98–116 99
Introduction
Atmospheric particulate matter (PM) which refers to a mixture
of solid particles and liquid droplets suspended in air, is one of
the most challenging issues in the environmental eld nowa-
days due to its chemical complexity, its measurement, and
its source apportionment leading to air quality management
( Seinfeld and Pandis, 2016 ). Recent studies have shown that
the premature mortality rate associated with exposure to am-
bient air pollution reached 8.8 million per year ( Lelieveld et al.,
2020 ) emphasizing on the seriousness of the human health
hazards ( Anderson et al., 2012 ; WHO, 2013 ; Zaheer et al., 2018 ).
Due to their very small size, PM
2.5
which are particles having
an equivalent aerodynamic diameter less than 2.5 µm, have
drawn much attention. Not only they can penetrate deeply
into the lungs, but also they can be retained inside and in-
duce respiratory ( Xing et al., 2016 ) and cardiovascular diseases
( Du et al., 2016 ).
The organic aerosol (OA) can contribute up to 50% to the
total PM
2.5
dry mass ( De Gouw and Jimenez, 2009 ) which can
be divided into Primary (POA) and Secondary (SOA) Organic
Aerosols. Primary organic compounds can serve as molecular
markers of a specic source of pollution such as hopanes for
vehicular emissions ( Rogge et al., 1996 ), fatty acids for cook-
ing activities ( Robinson et al., 2006 ), PAHs for fossil and non-
fossil fuel combustion ( Mastral et al., 2003 ). In addition, prod-
ucts of oxidation of different terpenes ( α-pinene, isoprene and
β-caryophyllene) have been used to characterize and quantify
the biogenic SOA ( Kleindienst et al., 2007 ).
Lebanon, a Middle Eastern Mediterranean country, with
a population of more than 6 million in the last few years,
faces some important pollution events. Episodically, the coun-
try is affected by long range transport of dust from deserts
( Borgie et al., 2016 ), but the main sources of pollution are local,
specially in winter ( Wak ed et al., 2013 ). Air quality in Lebanon
is heavily affected by road transport emissions caused by the
absence of the public transportation system. The road trans-
port sector is the main source of CO, NOx and Non-Methane
Volatile Organic Compounds (NMVOC) ( Abdallah et al., 2020 ;
Salameh et al., 2015 ; Waked and Af, 2012 ). Moreover, since
the national electricity company is unable to provide elec-
tricity 24/7, private diesel generators ll the gap with no law
enforcement on stack emissions ( Wak ed et al., 2012 ). On the
other hand, based on the 2010 national inventory, the main
emitter of PM is the industries followed by others like the
transport and the power generation sectors. Finally, the pop-
ulation growth hypothesis along with the refugees displace-
ment has led to an important residential solid waste genera-
tion that caused, in 2017, a substantial increase in open burn-
ing of waste in many parts of the country ( Abbas et al., 2019 ).
Main air pollution studies conducted in Lebanon focused
on the capital Beirut ( Af et al., 2008 ; Daher et al., 2013 ;
Salameh et al., 2015 ). Studies on PM conducted in Lebanon
generally focused on the inorganic composition ( Jaafar et al.,
2014 ; Kfoury et al., 2009 ; Yammine et al., 2010 ), and fewer ex-
amined certain organic families such as PAHs in urban areas
( Badran et al., 2020 ; Borgie et al., 2016 ; Daher et al., 2013 ). Fur-
thermore, Baalbaki et al. (2018) studied the PAH concentra-
tions in PM
10 samples at a site in Zouk Mikael and showed
higher values compared to two other urban sites in Beirut.
Melki et al. (2017) presented the difference of PAH s and alka-
nes concentrations and corresponding ratios between a site
under industrial inuence and a rural one in the Northern re-
gion of Lebanon. All the studies done on the PM
2.5 concen-
tration reported values exceeding the WHO daily guideline
(25 µg/m
3
) and sometimes double or triple this value during
dust storm events ( Jaafar et al., 2014 ). A more complete study
of the detailed organic characterization in summer and win-
ter has also been performed in a semi-urban area in Beirut
( Wak ed et al., 2013 ; Wake d et al., 2014 ) measuring for the rst
time SOA in Lebanon.
Although these studies bring valuable information on air
quality, they are limited to certain classes of organic com-
pounds, and a short sampling period of few days to few weeks
with a limited number of samples. In this context, this pa-
per will bring a rst detailed study of the organic compounds
in PM
2.5
collected over almost one year period in Lebanon at
a mixed industrial and heavy populated site in Zouk Mikael
and an industrial-residential site in Chekka region, Fiaa. The
present paper will focus on the composition and the seasonal
variations of the POA including PAHs, alkanes, hopanes, fatty
acids, and phthalates as well as the SOA including the oxida-
tion products of α-pinene, isoprene and β-caryophyllene, and
dicarboxylic acids. Some of these compounds, notably phtha-
lates are reported for the rst time in Lebanon. In order to
identify the possible PM sources in the sampling areas, di-
agnostic ratios for different classes of compounds were also
used. This rst one-year study in the East Mediterranean re-
gion will help in assessing the impact of the industrial emis-
sions on the organic aerosol composition.
1. Experimental
1.1. Sampling sites
PM
2.5 was collected at two sites in Lebanon: Zouk Mikael
(33 ˚57’57.07’’N 35 ˚37’09.46’’E) – Mount Lebanon district, 14 km
north-east of the capital Beirut, at the rooftop of a residen-
tial building (15 m above ground level), and in Chekka region,
specically Fiaa village (34 ˚20’47.8’’N 35 ˚47’14.0’’E)- Koura dis-
trict, 60 km north-north-east of Beirut and 10 km southwest
of Tripoli ( Fig. 1 ) .
Zouk Mikael area (ZK) is characterized by a high residential
density (4,200 inhabitants/km
2
), but also commercial and in-
dustrial activities. ZK has the biggest power plant in Lebanon
of 1 GW
electrical
which runs on Heavy Fuel Oil (HFO). It encom-
passes 607 MW
electrical
of boilers with 2 common stacks releas-
ing the emissions at 145 m, 198 MW
electrical of reciprocating
engines installed in 2017 with stack heights of around 40 m,
and a power barge with 11 reciprocating engines with a to-
tal capacity of 198 MW
electrical installed in 2012 with a stack
height of around 50 m. Moreover, a high number of private
generators along with small industries for plastic production,
woodworks, steel construction, aluminum extrusion, marble,
and granite production, etc. exist in this area. The Zouk Mikael
highway and the thermal power plant are respectively 1.2 and
1.5 km away from the sampling site.
100 journal of environmental sciences 101 (2021) 98–116
Fig. 1 – Location of the two sampling sites in Lebanon: Zouk Mikael (ZK) and Fiaa (FA) and the nearby industries (modied
from Google Earth).
Fiaa area (FA) is far less populated than ZK (250
residents/km
2
). It is also inuenced by private generators emi-
sisons. The main potential sources encompass chemical in-
dustries: two cement industries along with their correspond-
ing quarries in Chekka, and a sulfuric acid and phosphate fer-
tilizer industry few kilometers away ( Fig. 1 ) . In addition, the
nearest highway is 4 km away from FA with moderate trafc.
The two cement plants are 5 and 7 km away from the sampling
site.
1.2. Sample collection
The sampling of ne particles (PM
2.5
) was performed on a 24-
hour basis every three days from 13
th
of December 2018 to 15
th
of October 2019. PM
2.5
was sampled using high-volume sam-
plers (CAV-A/mb, MCV S.A., Spain) operating at 30 m
3
/hr, onto
150 mm pure quartz microbres lters (Fiorini, France). Filters
were heated for 12 hr at 550 ˚C before sampling to decrease
the organic impurities content and kept at -20 °C till sampling.
Over the sampling period, 98 samples in ZK and 95 samples in
FA have been collected. Field blanks (at least one/month) were
also considered at each site by placing a blank lter in sam-
pling conditions but without pumping. The collected lters
were sealed in aluminum foil and stored at -20 ˚C until analy-
sis. A wireless weather station (TFA-Dostmann 35.1112 OPUS)
was installed at the ZK site to collect meteorological data. Dur-
ing the whole sampling period, 93% of the samples were col-
lected under low windspeed ( < 2 m/sec) indicating that the
site was mainly under turbulent atmospheric conditions.
1.3. Organic compounds analysis
The method used for the organic compounds analysis was de-
scribed elsewhere ( Waked et al., 2013 ; 2014 ). In brief, a quar-
ter of the lter was spiked with 50 µL of 2 internal standards
(cis-ketopinic acid and bornylacetate) followed by an extrac-
tion by sonication for 30 min at 50 ˚C using 30 mL of ace-
tone/dichloromethane (50:50, V/V ). After the extraction, the
volume of the samples was reduced to 200 µL under a con-
stant ow of nitrogen gas. The obtained solution was used to
directly quantify non-polar compounds such as alkanes, PAHs,
hopanes and some phthalates while polar compounds such as
fatty acids and SOA compounds were quantied after deriva-
tization. Derivatization was achieved with 50 µL of the ex-
tract treated using N,O-bis(trimethylsilyl)-triuoroacetamide
(BSTFA) with 1% trimethylchlorosilane and 10 µL of pyridine
as a catalyst at 70 ˚C for 2 hr.
A 2-µL of the derivatized and the non-derivatized extracts
were injected using a gas chromatography coupled to a mass
spectrometer (GC/MS) in the split mode with a split ratio of
1/25. The GC consisted of an ISQ 7000 (Thermo Scientic,
United States of America) equipped with an HP 5MS UI capil-
lary column (30 m x 0.25 mm x 0.25 µm, Agilent; United States
of America). The column temperature program consisted of
an injection at 65 ˚C hold for 2 min, a ramp of temperature cor-
responding to 6 ˚C/min up to 300 ˚C followed by an isothermal
hold step at 300 ˚C for 20 min. The GC was interfaced to an ion
trap MS with an external electron ionization (EI) source (220 ˚C,
70 eV).
1.4. Identication and quantication of organic
compounds
For compounds for which an authentic standard was avail-
able in the laboratory (around 50 compounds), the identi-
cation was made by comparing the retention time and the
mass spectrum associated to the reference compound (full
journal of environmental sciences 101 (2021) 98–116 101
scan mode, range 50-500 m / z ). Additionnally, this method per-
mitted the identication of alkanes from C
12
to C
40
.
For other compounds, the identication was based on
the retention time, and the reference mass spectrum from
the literature ( Claeys et al., 2004 ; El Haddad et al., 2011b ;
Jaoui et al., 2007 ). In this case, the quantication was done
using the response factor (RF) of a surrogate compound.
For α-pinene oxidation products: (i) the RF of malic acid
was used for 3-hydroxyglutaric acid (A1), 3-acetylglutaric
acid (A2) 3-isopropylglutaric acid (A3) and 3-methyl-1,2,3-
butanetricarboxylic acid (A4); (ii) the RF of glyceric acid was as-
signed to 2-methylglyceric acid (2-MGA); (iii) the RF of threitol
was used for 2-methylthreitol (MT1), 2-methylerythritol (MT2),
and (iv) the RF of pinic acid was used for β-caryophyllinic acid
( βC). For the hopane family, the RF of 17 α(H)-21 β(H)-hopane,
an authentic standard, was used to quantify the concen-
trations of trisnorneohopane, 17 α(H)-trisnorhopane, 17 α(H)-
21 β(H)-norhopane, 17 α(H)-21 β(H)-22S, and 22R-homohopane.
The eld blanks were analyzed following the same proce-
dure as the sampling lters. The concentrations of the species
in the PM
2.5
samples were corrected by subtracting the mean
value obtained for the eld blank lters.
The detection limit was evaluated for all the compounds
and corresponds to the blank lter value plus 3 times the stan-
dard deviation calculated over 3 measurements. It ranged be-
tween 0.0003 and 0.08 ng/m
3
for non-derivatized compounds
and between 0.002 and 0.25 ng/m
3 for derivatized com-
pounds except for stearic acid with a higher detection limit
(2.3 ng/m
3
).
The coefcient of determination ( R ²) of the calibration
curves, determined several times during the analysis period,
for compounds with authentic standards, ranged between
0.93 - 0.99 except for tetracosanoic acid (0.90). Repeatability
was assessed by studying the variation in the RF of 5 con-
secutive injections of the authentic standards ( DRI, 2003 ). The
variations were less than 14%. The analytical uncertainty was
calculated using the quantication limit of the compound, the
repeatability, and the concentration of the compound. The to-
tal uncertainty including the analytical uncertainty and the
uncertainty associated to the mass ow measurement of the
sampler was in the range of 9%-30 % at 2 σ.
Recoveries were determined by spiking blank lters by
standard solutions. The values were estimated to be 80%, 82%,
92%, 90%, 85%, 82%, and 97% for alkanes, PAHs, fatty acids,
phthalates, dicarboxylic acids, pinic acid, and hopanes re-
spectively. For compounds with no authentic standard (i.e.
some SOA markers), the recovery of the surrogate compound
was determined to be 85% for glyceric acid and 95% for
threitol.
1.5. Index and diagnostic ratios calculation
Statistical diagnostic methods were used in a quantitative and
qualitative way in order to investigate the origin of the differ-
ent organic species in the PM
2.5
. The indexes for the n-alkanes
were used to differentiate the anthropogenic and the biogenic
origins. As for the PAHs, the ratios were used to separate py-
rogenic and petrogenic sources with a focus on the type of
combustion.
1.5.1. Ratios for the n-alkanes
Three methods were used to assess the contribution of the
sources for the parafns: the carbon number of the alkane
having the maximum concentration ( C
max
), the carbon prefer-
ence index (CPI), and the input of wax from plants (Wax ratio).
C
max
is used in general to differentiate between two alkane
sources: vegetation wax emissions for the high odd number of
carbons, i.e. 27, 29, and 31, and anthropogenic source for the
lower numbers ( Andreou and Rapsomanikis, 2009 ).
The Carbon Preference Index (CPI) is a measure of odd
to even alkanes predominance ( Simoneit, 1999 ) and evalu-
ates the contribution of the anthropogenic and the biogenic
source. Tw o CPI parameters were adopted: the Overall CPI
19-32
for all the alkanes and the High CPI
25-32 for the biogenic n-
alkanes and were calculated using Eq. 1 and Eq. 2 ( Bray and
Evans, 1961 ; Cooper and Bray, 1963 ).
Overall CP I
19 −32
=
odd C
19
−C
31
even C
20
−C
32
(1)
High CP I
25 −32
=
odd C
25
−C
31
even C
26
−C
32
(2)
An overall CPI value close to 1 indicates a petrogenic
source, between 2 and 5 mainly biomass burning, while a
value higher than 6 is characteristic of biogenic emissions
( Simoneit, 2002 ). For the High CPI, a value less than 1.5 in-
dicates an anthropogenic source while a value higher than 3
indicates a biogenic one ( Melki et al., 2017 ). An intermediate
value explains a mix of biogenic and anthropogenic sources.
The Wax ratio was used to determine the distribution of
the residual wax n-alkanes when the petroleum n-alkanes are
subtracted ( Simoneit et al., 1991 ). First, Wax
Cn is calculated
by subtracting the odd average concentration C
n of the next
higher C
n + 1
and lower C
n -1
even carbon (Eq. 3) . Then, the Wax
ratio (Wax%) corresponding to the percentage of Wax related
n-alkanes, is calculated by dividing the Wax
Cn (the sum of
Wax
Cn
for odd alkanes, with negative values of Wax
Cn
taken
as a Zero) by the total concentration of all the n-alkanes in the
sample ( A) (Eq. 4) .
Wax
Cn
= C
n
−1
2
(
C
n −1
+ C
n +1
) with n : odd number (3)
Wax % =
Wax
Cn
A
×100 (4)
1.5.2. PAHs diagnostic ratios
PAH s diagnostic ratios have been used to determine the
source of particle-containing PAHs. They can help to deter-
mine the different emission sources as well the different
fuel types used in the combustion processes ( Riffault et al.,
2015 ). This methodology is based on the hypothesis that the
PAH concentration ratios remain constant between the emis-
sion source and the measuring site. This is particularly true
for isomers having similar photochemical properties consid-
ered to be affected in a similar manner by the different re-
actions occuring in the atmosphere ( Borgie et al., 2016 ). Dif-
ferent ratios were calculated considering the concentrations
of uoranthene (Fla), pyrene (Pyr), indeno[1,2,3-c,d]pyrene
(InPy), benzo[g,h,i]perylene (B[ghi]Pe), benzo[a]anthracene
(B[a]An), chrysene (Chr), and benzo[a]pyrene (B[a]P), and com-
102 journal of environmental sciences 101 (2021) 98–116
Table 1 – Atmospheric concentrations of identied n-alkanes and hopanes during the entire sampling period (Total: Dec
2018- Nov 2019), winter (Dec 2018-March 2019), and summer (June 2019-September 2019) periods at Zouk (ZK) and Fiaa
(FA) sites.
Compounds Averag e concentration (min-max) (ng/m
3
)
ZK site FA site
Tota l Winter Summer Tot al Winter Summer
n-Alkanes
Nonadecane (C19) 0.42 (0.11-0.97) 0.44 (0.13-0.84) 0.34 (0.11-0.82) 0.15 (0.01-0.78) 0.19 (0.01-0.78) 0.11 (0.01-0.77)
Heicosane (C20) 0.45 (0.01-1.50) 0.53 (0.04-1.50) 0.26 (0.01-1.06) 0.33 (0.02-1.19) 0.40 (0.02-1.19) 0.23 (0.03-0.83)
Heneicosane (C21) 0.56 (0.04-2.42) 0.80 (0.13-2.42) 0.26 (0.04-0.74) 0.49 (0.04-1.94) 0.63 (0.04-1.94) 0.33 (0.09-1.07)
Docosane (C22) 1.03 (0.11-3.61) 1.56 (0.39-3.61) 0.50 (0.11-1.22) 0.80 (0.07-3.54) 1.04 (0.07-3.54) 0.56 (0.15-1.81)
Tricosane (C23) 1.84 (0.09-6.49) 2.67 (0.84-6.49) 0.94 (0.09-3.83) 1.55 (0.24-5.57) 1.81 (0.31-5.57) 1.10 (0.24-3.95)
Tetracosane (C24) 2.55 (0.10-12.7) 3.69 (1.07-12.7) 1.26 (0.10-2.99) 2.37 (0.79-7.42) 2.70 (0.79-7.42) 1.89 (0.82-5.86)
Pentacosane (C25) 2.99 (0.45-13.2) 3.80 (0.99-13.2) 1.62 (0.45-3.77) 3.23 (0.60-10.2) 3.28 (0.73-10.2) 2.79 (0.60-8.77)
Hexacosane (C26) 2.90 (0.44-15.1) 3.69 (0.78-15.1) 1.67 (0.44-4.23) 3.23 (0.47-12.1) 3.06 (0.75-9.76) 3.14 (0.47-12.1)
Heptacosane (C27) 3.25 (0.73-16.9) 3.77 (0.73-16.9) 2.16 (0.80-5.30) 4.10 (0.86-15.1) 3.03 (0.86-9.53) 4.43 (1.09-15.1)
Octacosane (C28) 2.16 (0.21-11.6) 2.60 (0.52-11.6) 1.44 (0.21-4.77) 2.96 (0.64-12.0) 2.46 (0.64-9.32) 3.19 (0.64-12.0)
Nonacosane (C29) 2.94 (0.51-25.8) 2.79 (0.55-9.79) 2.26 (0.51-4.74) 3.93 (0.73-15.4) 2.88 (0.79-10.7) 4.02 (0.73-12.8)
Tri aco nt an e (C30) 1.65 (0.92-6.59) 1.82 (0.40-6.59) 1.28 (0.22-4.17) 2.02 (0.23-8.46) 1.52 (0.23-5.95) 2.13 (0.42-8.46)
Hentriacontane (C31) 2.66 (0.02-18.5) 2.34 (0.40-7.74) 2.42 (0.02-5.91) 2.58 (0.17-9.96) 1.72 (0.17-6.41) 2.63 (0.33-9.96)
Dotriacontane (C32) 1.27 (0.15-5.18) 1.12 (0.17-3.40) 1.18 (0.15-3.35) 1.37 (0.02-6.85) 1.05 (0.02-4.82) 1.54 (0.34-6.85)
Tota l ( A) 26.70 31.63 17.61 29.12 25.76 28.11
Hopanes
Trisnorneohopane (H1) 0.28 (0.03-1.37) 0.32 (0.04-0.89) 0.30 (0.03-1.37) - - -
17 α(H)-Trisnorhopane (H2) 0.37 (0.06-1.48) 0.46 (0.10-1.48) 0.33 (0.06-1.01) - - -
17 α(H)-21 β(H)-Norhopane (H3) 1.08 (0.24-3.51) 1.26 (0.29-3.51) 0.83 (0.34-2.19) - - -
17 α(H)-21 β(H)-Hopane (H4) 1.08 (0.28-3.39) 1.24 (0.33-3.39) 0.87 (0.28-1.94) - - -
17 α(H)-21 β(H)-22S-Homohopane (H5) 0.74 (0.13-3.19) 0.69 (0.13-1.55) 0.84 (0.20-3.19) - - -
17 α(H)-21 β(H)-22R-Homohopane (H6) 0.67 (0.10-3.74) 0.68 (0.11-3.74) 0.75 (0.18-2.27) - - -
Tota l ( Hop) 4.22 4.66 3.91 - - -
pared to the literature: Fla/(Fla + Pyr), InPy/(InPy + B[ghi]Pe),
B[a]An/(B[a]An + Chr), and B[a]P/(B[a]P + Chr).
The ratio of sum of the non-alkylated compounds
(Fluorene, pyrene, benzo[a]anthracene, chrysene,
benzo[b]uoranthene, benzo[k]uoranthene, benzo[a]pyrene,
indeno[1,2,3-c,d]pyrene and benzo[g,h,i]perylene), noted C
PAH
,
to the total concentration of the PAHs, noted T
PAH
, evaluates
the contribution of the compounds related to combustion
processes ( Ravindra et al., 2008 ). Finally, the Low Molecular
Wei gh t (LMW) PAHs to the High Molecular Weig ht (HMW)
ratio, noted (3 + 4 rings)/(5 + 6 rings) ratio, can be used as an
indicator of the local or distant origin of PAH ( Tan et al., 2011 ).
2. Results and discussions
This study was performed over almost a one-year period. We
chose to discuss the average concentrations calculated for the
overall period as well as the concentrations associated with
the winter (December 2018 till March 2019) and the summer
periods (June 2019 till September 2019) in order to assess sea-
sonal trends. The corresponding concentrations for all studied
compounds are given in Table 1 , Tab le 2 , and Table 4 . Chrono-
logical evolutions are included in the supplementary material
to support interpretations (Fig. S2, Fig. S3, Fig. S4) .
2.1. Primary Organic Aerosols (POA)
Primary emissions from biogenic and anthropogenic sources
include more than forty sources in urban areas ( Rogge et al.,
1996 ) such as road transport, road dust, tire wear, cooking
operations, industrial boilers, replaces burning woods, plant
leaf abrasion, etc.
2.1.1. n-Alkanes
The yearly average of n-alkanes concentrations at both sites is
quite similar with 26.70 ng/m
3
at ZK versus 29.12 ng/m
3
at FA
( Tab le 1 ) . Nevertheless, the distribution of the n-alkanes con-
centration over the year was different: at ZK site, the winter
period concentration (31.63 ng/m
3
) is much higher than the
summer one (17.61 ng/m
3
) whereas it is similar for both sea-
sons at FA (25.76 vs 28.11 ng/m
3
). The observed values are sim-
ilar to those reported (23 ng/m
3
) for a coastal urban-industrial
site at Dunkirk, France ( Landkocz et al., 2017 ) and much lower
than those presented for an industrial site in Tianjin, China
(136-314 ng/m
3
) ( Li et al., 2010 ). This difference gets more im-
portant with a big population in China and a site exposed to
intensive coal burning emissions for industrial and domestic
purposes.
A clear seasonal pattern for the n-alkanes distribution
ranging from C
19
to C
32
was observed at ZK and FA with higher
concentrations of C
21
-C
27
during winter compared to summer
journal of environmental sciences 101 (2021) 98–116 103
Table 2 – Atmospheric concentrations of identied polycylic aromatic hydrocarbons(PAHs), fatty acids, and phthalates during the entire sampling period (Total: Dec 2018-
Nov 2019), winter (Dec 2018-March 2019), and summer (June 2019-September 2019) periods at Zouk (ZK) and Fiaa (FA) sites.
Compounds Averag e concentration (min-max) (ng/m
3
)
ZK site FA site
Tota l Winter Summer Tot al Winter Summer
PAHs
Acenaphthylene (Acy)
1 0.02 (0.01-0.08) 0.03 (0.02-0.08) 0.012 (0.01-0.04) - - -
Acenaphthene (Ace)
1 0.02 (0.01-0.09) 0.03 (0.01-0.09) 0.011 (0.01-0.04) - - -
Fluorene (Flu)
1 0.02 (0.01-0.25) 0.03 (0.01-0.25) 0.02 (0.01-0.04) - - -
Anthracene (Anth)
1 0.12 (0.01-0.42) 0.15 (0.05-0.42) 0.09 (0.01-0.21) 0.08 (0.01-0.58) 0.12 (0.01-0.58) 0.05 (0.01-0.36)
Phenanthrene (Phe)
1 0.03 (0.01-0.09) 0.04 (0.01-0.09) 0.02 (0.01-0.09) 0.02 (0.01-0.08) 0.02 (0.01-0.08) 0.02 (0.01-0.07)
Fluoranthene (Fla)
2 0.13 (0.01-0.68) 0.22 (0.05-0.68) 0.06 (0.01-0.19) 0.11 (0.01-1.16) 0.22 (0.02-1.16) 0.04 (0.01-0.31)
Pyrene (Pyr)
2 0.15 (0.01-0.76) 0.24 (0.05-0.76) 0.07 (0.01-0.19) 0.10 (0.01-0.70) 0.19 (0.01-0.69) 0.06 (0.01-0.26)
Benzo[a]anthracene (B[a]An)
2 0.15 (0.01-0.76) 0.25 (0.06-0.76) 0.07 (0.01-0.19) 0.07 (0.01-0.77) 0.16 (0.01-0.77) 0.03 (0.01-1.03)
Chrysene (Chr)
2 0.28 (0.02-1.17) 0.44 (0.11-1.17) 0.12 (0.02-0.30) - - -
Benzo[b]uoranthene (B[b]Fl)
3 0.27 (0.04-1.07) 0.44 (0.12-1.05) 0.10 (0.04-0.22) 0.14 (0.01-1.03) 0.22 (0.03-0.69) 0.08 (0.01-1.03)
Benzo[k]uoranthene (B[k]Fl)
3 0.15 (0.01-0.75) 0.27 (0.04-0.75) 0.05 (0.01-0.13) 0.07 (0.01-0.60) 0.13 (0.02-0.60) 0.03 (0.01-0.43)
Benzo[a]pyrene (B[a]P)
3 0.20 (0.01-1.26) 0.38 (0.06-1.26) 0.05 (0.01-0.20) 0.05 (0.01-0.44) 0.10 (0.02-0.35) 0.02 (0.01-0.44)
Dibenzo[a,h]anthracene (DiB[a,h]An)
3 0.45 (0.01-2.86) 0.89 (0.10-2.86) 0.08 (0.01-0.26) 0.12 (0.01-1.04) 0.22 (0.02-1.04) 0.04 (0.01-0.25)
Benzo[g,h,i]perylene (B[ghi]Pe)
4 0.07 (0.01-0.40) 0.09 (0.02-0.38) 0.04 (0.01-0.13) 0.13 (0.01-1.22) 0.22 (0.02-1.22) 0.09 (0.01-0.34)
Indeno[1,2,3-c,d]pyrene (InPy)
4 0.49 (0.03-3.11) 0.85 (0.03-3.11) 0.11 (0.03-0.50) 0.06 (0.01-0.50) 0.10 (0.01-0.50) 0.03 (0.01-0.25)
Tota l ( PAHs ) 2.56 4.35 0.88 0.95 1.70 0.50
Fatty acids
Dodecanoic acid (DDA) 5.35 (0.04-17.33) 7.00 (2.09-17.33) 3.98 (0.04-12.00) 3.93 (0.15-14.76) 4.53 (1.68-10.33) 3.76 (0.15-14.76)
Tetradecanoic acid (TDA) 8.05 (0.29-52.19) 7.15 (0.86-52.19) 8.21 (0.29-18.11) 18.93 (0.09-93.61) 30.13 (10.56-93.61) 11.07 (0.09-33.05)
Hexadecanoic acid (HDA) 259.47 (2.29-3197.75) 171.55 (20.25-1201.27) 423.99 (2.29-3197.72) 415.19 (1.47-2435.79) 521.07 (191.45-1345.41) 457.11(4.58-2435.79)
Octadecanoic acid (ODA) 175.32 (16.57-1508.02) 221.20 (16.57-730.00) 189.51(18.00-1508.02) 247.14 (3.07-1423.12) 366.36 (96.94-894.39) 229.78(3.07-1423.12)
Eicosanoic acid (EA) 8.50 (0.43-34.68) 12.08 (0.43-34.68) 7.07 (0.79-29.12) 6.23 (0.01-37.95) 6.55 (1.59-37.95) 9.03 (1.16-34.21)
Docosanoic acid (DA) 18.75 (0.10-124.74) 28.61 (4.96-107.34) 12.45 (0.10-124.74) 11.61 (0.1-103.82) 19.51 (1.27-103.82) 10.78 (1.44-29.59)
Tetracosanoic acid (TA) 15.79 (0.69-85.33) 26.84 (0.98-85.33) 10.89 (0.80-76.67) - - -
Oleic acid (OA) 17.07 (0.13-80.26) 21.34 (3.19-80.26) 14.81 (0.13-71.59) 14.61 (0.08-60.03) 17.06 (5.86-32.44) 19.00 (2.02-60.03)
Tota l ( FA) 508.29 497.24 671.68 717.63 965.23 740.53
Phthalates
Diisobutylphthalate (DIBP) 22.19 (2.12-102.10) 20.92 (3.41-102.10) 24.17 (2.12-53.75) 19.72 (0.02-78.70) 20.42 (5.51-46.30) 6.50 (0.02-41.70)
Dibutylphthalate (DnBP) 35.69 (3.86-126.93) 15.48 (3.86-60.06) 57.84 (20.09-114.06) 12.36 (0.05-57.69) 10.11 (1.80-27.34) 8.14 (0.05-38.94)
Bis(2-ethylhexylphthalate) (DEHP) 48.99 (4.93-124.05) 43.09 (4.93-104.01) 58.39 (9.13-124.05) 65.59 (0.66-432.47) 63.85 (0.66-409.61) 39.75 (5.21-265.81)
Tota l ( PAE) 106.88 79.49 140.40 97.68 94.38 54.39
Number of rings for PAHs
1 : 3 rings
2 : 4 rings
3 : 5 rings
4 : 6 rings
104 journal of environmental sciences 101 (2021) 98–116
Fig. 2 –n-Alkanes prole patterns associated with PM
2.5
in Zouk (ZK) and Fiaa (FA) in winter and summer periods.
especially for the ZK site ( Fig. 2 ) . This could be attributed to
the higher contribution of the residential heating in the cold
period. The larger difference observed for ZK reinforces this
hypothesis since ZK is much more populated than FA. In the
summer period, the n-alkanes distribution prole shifts to the
highest carbon number, ≥C
27
, ascribable to plant wax-derived
alkanes ( Rogge et al., 1993b ). The concentrations of the high
odd alkanes in FA are remarkably higher than those in ZK
showing higher contribution of the natural source at FA. This
can be explained by the fact that Chekka region, and Fiaa pre-
cisely are more densly surrounded by green lands and trees. In
both sites in winter, C
max
was at C
25
indicating the prevalence
of the anthropogenic sources ( Simoneit, 1989 ). In summer, the
most abundant n-alkanes are C
29
and C
31
at ZK and C
27
and
C
29
at FA suggesting a higher relative contribution of biogenic
aerosols during the hot season ( Li et al., 2006 ).
The CPI values were calculated as described in
section 2.5.1 for each PM
2.5
sample and presented in ( Fig. 3 ) . In
the winter period, for both sites, the obtained High CPI values,
below 1.5, indicate a quasi-exclusively anthropogenic origin
for the n-alkanes. The Overall CPI close to 1 also indicates
the contribution of a petrogenic source (petroleum residues).
In the summer period, the High CPI shifted to values be-
tween 1.5 and 3 for most of the samples in ZK, indicating a
mixed inuence of natural and anthropogenic sources. Only
few samples have the same tendency in FA site suggest-
ing the higher impact of the anthropogenic sources during
summer.
The contribution of the natural sources as the primary
biogenic source can be assumed by calculating the Wax ra-
tio (Wax%). Wax% values were similar for the whole period
(13%) at both sites ( Tab le 3 ) . This ratio is in agreement with
the value of 16% reported for a Lebanese site under industrial
inuence, Zakroun in Chekka region and lower than the one
observed in a rural site (27%), Kaftoun in Chekka region too
( Melki, 2017 ) due to the absence of primary sources near the
site.
The Wax% increased in the summer period, 20% and 14%
respectively in ZK and FA (compared to 7% and 9% during
winter respectively) emphasizing on the increase of natural
source contributions in the summer period. The winter Wax%
at ZK and FA are in agreement with the value of 7% obtained in
Sin El Fil, an urban site in Lebanon during a winter campaign
in 2017 ( Badran et al., 2020 ). The site was characterized by im-
portant road trafc, diesel generators, and waste incineration
emissions. Additionnally, the values appear to be in the range
of those measured at an industrial site in Tianjin of 10% in
winter and 30% in summer ( Li et al., 2010 ). A good correlation
( R
2
> 0.90) was observed between the Wax% and the Overall
CPI at both sites for the whole sampling period which is also
noted in other urban areas ( Andreou and Rapsomanikis, 2009 ;
Kotianova et al., 2008 ).
Despite the fact that biogenic related alkanes have higher
contributions in the hot season at both sites, low wax per-
centages and CPI values close to unity show that plant waxes
are not the major source for these alkanes. Hence, anthro-
pogenic combustion sources related to diesel generators, road
trafc, industrial processes and to residential heating in win-
ter can be assumed to account to the majority of emissions of
n-alkanes in the studied areas.
journal of environmental sciences 101 (2021) 98–116 105
Fig. 3 – Source identication of alkanes using the Carbon Preference index (CPI) during winter and summer period at Zouk
(ZK) and Fiaa (FA) sites.
Table 3 –Alkanes wax ratio for Zouk (ZK) and Fiaa (FA) sites during total, winter and summer periods.
Wax % ZK site FA site
Tota l period Winter Summer Tot al period Winter Summer
Averag e 13% 7% 20% 13% 9% 14%
Minimum 0% 1% 0% 1% 1% 4%
Maximum 52% 24% 36% 35% 25% 35%
2.1.2. Polycyclic aromatic hydrocarbons (PAHs)
The concentration of the 16 priority PAHs listed by the United
States Environmental Protection Agency (US EPA) in relation
with their toxicity, mutagenicity and/or carcinogenic proper-
ties has been investigated. The main source of emission of
PAH s in the atmosphere is the combustion of fossil and non-
fossil fuels ( Mastral et al., 2003 ).
The total PAH concentration was higher in ZK (2.56 ng/m
3
)
than in FA (0.95 ng/m
3
). The concentration of naphthalene
was below the detection limit which has been also observed
in other studies since the partitionning coefcient favors the
gas phase of this compound ( Waked et al., 2013 ; 2014 ). The
total particulate phase PAH s measured values are in agree-
ment with the 1.16 ng/m
3 reported for a background ur-
ban site in Lebanon ( Borgie et al., 2015 ) but are lower than
those reported in both paticulate and gaseous phases by
Daher et al. (2013) near Jal el Dib freeway, Lebanon for PM
2.5
(12.2 ng/m
3
) and by Baalbaki et al. (2018) in Zouk Mikael (25.1
and 27.7 ng/m
3
in winter and summer respectively for PM
10
).
Compared to other industrial urban sites, these values are
higher than those observed (0.224 ng/m
3
) in the Czech Re-
public ( Mikuška et al., 2015 ) but lower than those in North-
ern France (7.7 ng/m
3
) ( Landkocz et al., 2017 ), and much lower
than those measured at an industrial region in China (235
ng/m
3
) ( Bi et al., 2020 ).
In this study, PAH concentrations tend to be 3 (at FA) to
5 (at ZK) times higher in the winter period (4.35 and 1.70
ng/m
3 at ZK and FA respectively) compared with the sum-
mer period (0.88 and 0.50 ng/m
3 respectively). This observa-
tion could be due to increasing primary source emissions spe-
cially combustion activities alongside the atmospheric stabil-
ity in winter, partitioning between particulate and gaseous
phases, and greater photochemical degradation during sum-
mer ( Pindado et al., 2009 ). The ratio of (3 + 4 rings)/(5 + 6 rings)
PAH s was calculated for both seasons and showed the same
average of 0.76 at FA while a seasonal difference was evi-
denced in ZK (0.59 in winter and 1.21 in summer). The lower
ratio in winter at ZK could indicate that the contribution of
local sources to PAH concentration was higher in this period
compared with the summer one ( Tan et al., 2011 ).
Looking at the yearly average concentrations, and compar-
ing the PAH s distribution, a specicity appears for the ZK site.
106 journal of environmental sciences 101 (2021) 98–116
Fig. 4 – Bi-variate plots of InPy/(InPy + B[ghi]Pe) vs. Fla/(Fla + Pyr) ratios at Zouk (ZK) and Fiaa (FA) sites during winter and
summer respectively.
While no predominant compound is clearly evidenced at FA
site, at ZK site InPy and DiB[a,h]An appear to be the main
PAHs. Moreover, a strong correlation during the total sampling
period exists for these latter only at ZK ( R
2
= 0.87 at ZK versus
R ²= 0.40 at FA). This suggests that these compounds are lo-
cally emitted and are related to a common source of emission
in the ZK study area.
To go further, the contribution of the different sources of
PAH s was investigated by the study of the PAHs diagnostic
ratios. Using the Fla/(Fla + Pyr) and the InPy/(InPy + B[ghi]Pe)
isomer ratios makes possible the evidencing of the contri-
bution of petrogenic, wood and coal combustion, fuel com-
bustion, and also diesel and gasoline sources. According to
the literature, the Fla/(Fla + Pyr) suggests a petrogenic source
for a value lower than 0.2, a liquid fossil fuel combus-
tion for 0.4-0.5, and wood or coal combustion for a value
higher than 0.5 ( Cazier et al., 2016 ; Ravindra et al., 2008 ). The
InPy/(InPy + B[ghi]Pe) ranging between 0.2 and 0.5 is consid-
ered as a marker for gasoline source, 0.35-0.7 for diesel source,
and wood and coal combustion for values higher than 0.5
( Bi et al., 2020 ; Riffault et al., 2015 ).
At ZK and in both seasons, Fla/(Fla + Pyr) ratio ranges be-
tween 0.4 and 0.5 indicating that fuel combustion emissions
were predominant (Fig. 4) . Moreover, InPy/(InPy + B[ghi]Pe) ra-
tio shows values mainly above 0.8 and only few values in the
0.2-0.7 range. Hence, these observations allow to conclude that
at ZK site (i) road trafc is not the predominant source of InPy
and B[ghi]Pe since it has a characteristic InPy/(InPy + B[ghi]Pe)
value of 0.31 in Lebanon ( Daher et al., 2013 ) in line with
Bi et al. (2020) and Riffault et al. (2015) as the trafc in Lebanon
is dominated by gasoline ( Abdallah et al., 2020 ), (ii) Diesel pri-
vate generators are not the main source since the observed
ratio values are higher than 0.8, and (iii) this source cannot be
associated to biomass burning which has ratio values above
0.6 ( Ravindra et al., 2008 ; Tobiszewski and Namie
´
snik, 2012 )
since it contradicts the observed Fla/(Fla + Pyr) values. There-
fore, a deepen interpretation of PAH s concentrations is needed
to identify a potential source with a characteristic ratio above
0.8 specic to liquid fuel combustion.
High values of the InPy/(InPy + B[ghi]Pe) ratio were obtained
by Manoli et al. (2004) for oil burning from residential heat-
ing appliance chimneys (0.82), cement plants (0.90-0.96) and
diesel emissions from taxis and buses (0.96). However, none of
these sources are present in the surroundings of the ZK site.
Consequently, the thermal power plant located in ZK is a po-
tential source since it uses a third type of liquid fuel; the Heavy
Fuel Oil (HFO). To investigate this hypothesis, the samples
when the site was predominantly down wind of the power
plant encompassing wind speeds above 2m/s were exam-
ined. Only two lters met clearly the selection criteria (Fig. S1)
and showed InPy/(InPy + B[ghi]Pe) ratio values of 0.84 and 0.93
which are in line with our hypothesis. Ratio values varying
around 0.9 were encountered in most of the samples which
can be explained by the fact that HFO combustion emissions
are released at 145 m of height by the 607 MW boilers result-
ing in an enhanced dispersion affecting a broader area like the
surrounding cities ie. Jounieh, but also at around 50 m from
the ground by the 396 MW reciprocating engines with con-
sequently less enhanced dispersion impacting a smaller area
concentrated on Zouk Mikael entirely and nearest surround-
ings. This observation results in affected air masses reaching
the site from all directions.
journal of environmental sciences 101 (2021) 98–116 107
Fig. 5 – Source prole by unique ratio (SPUR) method applied to the InPy/(InPy + B[ghi]Pe) ratio considering ZK site data.
Up to our knowledge, the literature is scarce regarding
characteristic InPy/(InPy + B[ghi]Pe) ratio value for the heavy
fuel oil combustion occurring in power plants. Cecinato
et al. (2014) indicates a value of 0.35 based on the emis-
sion factors. This value won’t vary much for the particulate
phase only since partitioning coefcients between particulate
and gaseous phases for the two compounds are very close
( Kim and Kim, 2015 ). Va lu e s of 0.35 and 0.5 can be also ob-
tained from a study conducted by Yan g and co-workers and
from the US EPA AP-42 respectively ( USEPA, 2010 ; Ya ng et al.,
1998 ). However, the representativeness of these values is poor,
as several authors in the literature stress on the idea that the
PAH emissions strongly depend on the combustion conditions
and quality of fuel that might change from a site to another
( Masclet et al., 1987 ; Mastral and Callén, 2000 ; Ravindra et al.,
2008 ; Tobiszewski and Namie
´
snik, 2012 ).
With the idea to better dene this new characteristic ra-
tio value, the “Source Prole by Unique Ratio” (SPUR) method
was applied ( Annegarn et al., 1992 ; Ledoux et al., 2017 ). This
method consists to plot a ratio involving the characteristic
species of the source versus the concentration of the charac-
teristic species. The limit of the ratio obtained for the high-
est concentrations of the characteristic species can be as-
sumed as the characteristic ratio of the source. It has been
applied to the InPy/(InPy + B[ghi]Pe) ratio versus InPy con-
centrations ( Fig. 5 ) as InPy shows particularly high concen-
trations at ZK site. The SPUR method allows to suggest an
InPy/(InPy + B[ghi]Pe) characteristic ratio between 0.8 and 1 for
the Heavy Fuel Oil combustion occurring in a thermal power
plant.
At FA, the InPy/(InPy + B[ghi]Pe) ratio values appear mainly
in the 0.2-0.7 range (diesel and gasoline combustion) and no
value higher than 0.75. In addition, the Fla/(Fla + Pyr) ratio val-
ues indicate a mix of fuel combustion from vehicular emis-
sions and diesel generators along with wood and coal com-
bustion. The cement industries in Chekka use coke as their
primary combustion source which explains our results.
Then, the values of B[a]An/(B[a]An + Chr) ratios in ZK were
0.36 and 0.37 for winter and summer respectively with no sig-
nicant difference between the seasons meaning that these
compounds come from a constant activity throughout the
sampling period. In addition to that, the B[a]P/(B[a]P + Chr) ra-
tio was 0.46 in winter and 0.28 in summer . According to the
literature, the ratio B[a]An/(B[a]An + Chr) refers to a pyrogenic
source generated from the combustion of fossil fuel (coal and
petroleum) and/or biomass for values higher than 0.35 and a
petrogenic one from unburned crude oil and petroleum prod-
ucts for values lower than 0.2 ( Boonyatumanond et al., 2007 ;
Wu et al., 2014 ). On the other hand, the ratio B[a]P/(B[a]P + Chr)
is generally used to assess the contribution of vehicular
emissions. It was reported as 0.33 at an urban environment
( Guo, 2003 ), 0.49 for gasoline emissions, and 0.73 for diesel
emissions ( Khalili et al., 1995 ). In Lebanon, the on-road eet
runs to a high extenton gasoline ( Abdallah et al., 2020 ). Con-
sequently, these two ratios suggest that the road trafc is an
important source of B[a]An, B[a]P, and Chr in ZK during both
seasons. These ratios were not calculated in FA due to the val-
ues of Chr below quantitation limit.
Finally, the combustion PAHs (CPAHs) accounted for 71%
and 77% during winter and 69% and 72% of the total PAH con-
centration (TPAH) during summer in ZK and FA, respectively.
Despite the difference in the concentrations between the two
sites, the combustion source impact is remarkably important
and constant during both seasons suggesting that these PAHs
are mainly emitted from sources that do not have any sea-
sonal pattern such as diesel private generators, road trafc,
and industrial emissions (HFO and coke combustion at ZK and
FA, respectively) rather than unburned fossil fuels. These re-
sults are in agreement with the dominant anthropogenic ori-
gin of the alkanes emissions.
108 journal of environmental sciences 101 (2021) 98–116
2.1.3. Hopanes
Hopanes are fossil fuel compounds present in unburned lu-
bricating oils and are not found in diesel and gasoline because
they belong to the higher boiling fraction of crude petroleum
( Henry et al., 1984 ; Rogge et al., 1993a ).
At ZK, the total concentration during the sampling period
equals to 4.18 ng/m
3 ( Tab le 1 ) and is close to that reported
in Ostrava, Czech Republic equal to 3.79 ng/m
3 after smog
episode ( Mikuška et al., 2015 ). The average concentrations of
hopanes were 4.66 and 3.91 ng/m
3
in PM
2.5
for ZK in the winter
and the summer periods respectively. The seasonal difference
could be mainly due to the fact that the hopanes are more
volatile during the hot season ( Ruehl et al., 2011 ) as well as
the lower average road trafc intensity in summer due to the
closing of education institutions till September ( Wake d and
Af, 2012 ). These values are higher than 1.2 and 1.65 ng/m
3
re-
ported for an urban area in Guangzhou, China for winter and
summer, respectively ( Wang et al., 2016 ).
The two most abundant hopanes were 17 α(H)-21 β(H)-
norhopane and 17 α(H)-21 β(H)-hopane accounting for 50% of
the total hopane concentration and exhibit a good correla-
tion ( R
2
= 0.85). The S/(S + R) epimers ratio for 17 α(H)-21 β(H)-
homohopane could indicate that compounds are either emit-
ted from road trafc or from coal combustion with 0.5 as the
cutoff value between the two sources ( Mikuška et al., 2015 ). At
ZK site, higher concentrations for the S epimer compared with
the R epimer were found (S/S + R > 0.5) showing the major in-
uence of road trafc. All these ndings were in agreement
with El Haddad et al. (2009) who concluded that these com-
pounds were emitted from vehicular emissions.
The observations also show a decrease of 25 % on average
in hopanes concentration on Sundays (3.27 ng/m
3
) compared
to weekdays (4.40 ng/m
3
) (Table S1) . The weekend in general,
but specially Sundays exhibit lower trafc related activities.
At the FA site, the PM
2.5
hopane content was below the de-
tection limit. This is in agreement with the fact that the vehic-
ular emissions in Chekka region are much lower than those in
ZK which is also highlighted by the chrysene concentrations
which are below the detection limit at FA site.
2.1.4. Fatty acids
In this study, the fatty acids class were the most abun-
dant detected organic compounds with average concentra-
tions over the entire period of 508.29 and 717.63 ng/m
3 at
ZK and FA, respectively. The concentrations are in range of
the 644 ng/m
3 reported for the semi urban site in Beirut in
summer ( Wak ed et al., 2014 ) but much higher than the 234
ng/m
3 reported in winter ( Wa ke d et al., 2013 ) for the same
site and those reported for 3 urban Indian sites (234-583
ng/m
3
) ( Gadi et al., 2019 ). Ave rage concentrations of hexade-
canoic acid (HDA) and octadecanoic acid (ODA) were 259.47
and 175.32 ng/m
3 in ZK, and 415.19 and 247.14 ng/m
3 in FA,
respectively. Concentrations of these species accounted for 86
and 92% of the total alkanoic acids in ZK and FA , respectively.
In addition to that, oleic acid which is an alkenoic acid was an-
alyzed and showed concentrations of 17.07 ng/m
3
at ZK and
14.61 ng/m
3
at FA.
Generally, this class of compounds is the most abundant
in the organic fraction ( Rogge et al., 1991 ). The main compo-
nents are hexadecanoic and octadecanoic acids as saturated
fatty acids and oleic acid as unsaturated fatty acid. These com-
pounds have multiple sources but are mostly emitted from
cooking activities in urban areas ( Rogge et al., 1991 ) when glyc-
erides present in seed oils are pyrolyzed ( Schauer et al., 2002 ).
HDA and ODA were well correlated in both sites ( R
2
= 0.94
in ZK and R
2
= 0.96 in FA) but no correlation was found with
oleic acid ( R
2
= 0.15 in ZK and R
2
= 0.20 FA). These results are
in agreement with Robinson et al. (2006) who concluded that
saturated and unsaturated fatty acids have different cooking
dominant sources with the assumption that these compounds
are stable in the atmosphere.
The high concentrations observed for fatty acids are
mainly due to the residential typology of the sites. The cook-
ing activities are abundant especially with the usage of Canola
and Soybean seed oil for frying. The concentrations for the
fatty acids are higher in FA than in ZK due to the fact that
the sampling site in FA was closer to houses. The seasonal
variation in ZK show higher concentrations in summer mainly
due to more outdoor cooking activities (i.e. charcoal grilling of
meat and chicken) as well as for the restaurants that are more
abundant in the area.
2.1.5. Phthalates
Phthalates are a group of man-made chemical compounds
with esters of phthalic acid used as plasticizers in industrial
nal products and building materials ( Lu et al., 2018 ).
To our knowledge, this study is the rst dealing with the
quantication of phthalates in Lebanon. The average concen-
tration of phthalates was 106.88 and 97.68 ng/m
3
respectively
at ZK and FA. These values are lower than those reported in 3
Indian urban sites (211, 159, and 130 ng/m
3
) ( Gadi et al., 2019 )
but similar to the 89 ng/m
3
reported in an urban site in North-
ern Vietnam ( Nguyen et al., 2016 ).
Bis(2-ethylhexylphthalate) (DEHP) has the highest concen-
tration between the compounds during the different seasons
at both sites and in all the samples. It is generally found in the
particulate phase while other phthalates like diisobutylphtha-
late (DIBP) and dibutylphthalate (DnBP) are predominant in
the gaseous fraction ( Pei et al., 2013 ).
A particular attention was given to DIBP and DnBP at both
sites because they show different time series (Fig. S2) . At ZK
site, similar values and trends were observed between the two
species until the 15
th
of May after which the trend and values
became different with higher concentrations of DIBP. While in
FA, the contrary occurred with higher concentrations of DIBP
than DnBP until the 1
st of June 2019 after which concentra-
tions became lower and similar. The comparison of the time
series between the sites suggests that the emission sources
were not the same and probably mainly related to local inu-
ence. Generally, it is known that these compounds are emitted
during plastic burning ( Simoneit et al., 2005 ). At FA, the high
concentrations of these two phthalates were probably caused
by the open waste burning in the North Governorate-which
FA is part of- during the whole sampling time in several lo-
cations. This phenomenon increased during the March –May
period (author eld observation).
In order to further investigate this matter, DIBP and DnBP
concentrations were plotted versus the sampling date ( Fig. 6 ) .
Different correlation trends were observed at both sites. For
the FA site, two separate trends with slopes of 1.36 (before June
journal of environmental sciences 101 (2021) 98–116 109
Fig. 6 – Correlations between DIBP and DnBP at Fiaa(FA) and Zouk(ZK) plotted versus the sampling date.
1
st
, 2019) and 1.01 (after June 1
st
, 2019) can be distinguished.
The difference might be explained by the different composi-
tion of waste that was burned.
A different scenario is shown in ZK between DIBP and
DnBP, where the 15
th
of May 2019 is considered as the cutoff
date between two trends for these 2 compounds. Before this
date, the compounds are well correlated and show mostly a
slope of 1.42 similar to those reported for Chekka region sug-
gesting open waste burning with different waste composition.
This activity might have been reduced in ZK after mid-May
because the region is considered as a touristic destination in
Lebanon during summer. However, a different slope value and
higher DnBP concentrations were reported after this date sug-
gesting higher production rates at the plastic industries lo-
cated in ZK area and its surroundings in the south-west sector.
Considering the high concentrations of the phthalates in
ZK and FA, further investigations should be focused on their
emissions in the Lebanon and the middle eastern region.
Wet he r they are emitted from municipal open waste burn-
ing or plastic industries, studies have shown that they might
cause diverse health effects specially on the endocrine sys-
tem, ( Ji et al., 2014 ) the reproductive systems and children’s
intelligence ( Lu et al., 2018 ).
2.2. Secondary Organic Aerosols (SOA)
The secondary organic aerosols (SOA), an important fraction
of the particulate matter, encompasses compounds produced
from the transfomation of organic species in the atmosphere
in the gas or condensate phase ( Kroll and Seinfeld, 2008 ).
SOA from gas-phase reactions can result from oxidation of
Volatile Organic Compounds (VOC) by atmospheric oxidants
such as ozone O
3
, hydroxyl radicals OH and nitrate radicals
NO
3
( Atkinson, 2008 ). Isoprene, α-pinene, and β-caryophyllene
are mainly emitted from deciduous trees, pine forests, and
vegetation respectively as primary emissions. They can be ox-
idized through photochemical reactions to give Biogenic Sec-
ondary Organic Aerosol (BSOA). The formation of these com-
pounds depend largely on temperature variations and photo-
chemical processes of the precursors ( Feng et al., 2013 ). In this
part, we will focus on the BSOA as well as dicarboxylic acids
which are generally produced by the gas phase photochemical
reactions including a variety of anthropogenic and biogenic
precursors.
2.2.1. Isoprene oxidation products
Isoprene is mainly emitted by broadleaf vegetation
( Guenther et al., 1995 ) and is considered as a highly re-
active compound due to its two double bonds ( Carlton et al.,
2009 ). The major SOA tracers identied in the case of iso-
prene are mainly the 2-methylglyceric acid (2-MGA) and
the two diastereoisomers 2-methylthreitol (MT1) and 2-
methylerythritol (MT2).
The average concentrations of MT1 and MT2 in summer
were respectively 2.64 and 6.53 ng/m
3
at ZK and 3.68 and 10.40
ng/m
3
at FA ( Tabl e 4 ). They were at least 5 times higher than
those measured in winter. This phenomenon is mainly due to
the fact that the emissions of the precursor depend largely on
the temperature variation and solar radiation as it enhances
the photochemical reactions in the atmosphere ( Feng et al.,
2013 ).
The concentrations of the 2-methyltetrols (MT) were in
range of those reported by Feng et al. (2013) for an urban and a
suburban sites in Shanghai and lower than those reported for
110 journal of environmental sciences 101 (2021) 98–116
Table 4 – Atmospheric concentrations of identied secondary organic compounds during the entire sampling period (Total:
Dec 2018- Nov 2019), winter (Dec 2018-March 2019), and summer (June 2019-September 2019) periods at Zouk (ZK) and
Fiaa (FA) sites.
Compounds Averag e concentration (min-max) (ng/m
3
)
ZK site FA site
Tota l Winter Summer Tot al Winter Summer
Isoprene oxidation products
2-Methylglyceric acid (2-MGA) 0.47 (0.04-5.19) 0.22 (0 .05-0.93) 0.77 (0.04-5.19) 0.86 (0.01-2.64) 0.32 (0.01-1.40) 1.30 (0.17-2.64)
2-Methylthreitol (MT1) 1.41 (0.10-5.81) 0.37 (0.10-1.13) 2.64 (0.73-5.81) 2.19 (0.08-13.71) 0.68 (0.08-3.00) 3.68 (0.54-13.71)
2-Methylerythritol (MT2) 3.78 (0.53-14.61) 1.43 (0.53-4.69) 6.53 (1.04-14.61) 6.14 (0.13-39.79) 1.57 (0.13-8.09) 10.40 (0.17-2.64)
Tota l 5.67 2.01 9.93 9.19 2.56 15.38
α-pinene oxidation products
Pinic acid (PA) 0.73 (0.09-3.54) 0.50 (0.09-1.75) 0.99 (0.11-3.54) 1.63 (0.10-9.21) 1.24 (0.10-3.96) 1.50 (0.38-3.04)
3-Hydroxyglutaric acid (A1) 1.63 (0.01-9.90) 0.65 (0.01-3.28) 2.94 (0.13-9.90) 5.51 (0.02-21.90) 4.31 (0.06-20.02) 7.56 (0.87-21.90)
3-Acetylglutaric acid (A2) 1.64 (0.01-8.96) 0.62 (0.01-2.10) 2.74 (0.43-8.96) 3.59 (0.18-13.00) 2.89 (0.18-13.00) 4.65 (0.87-12.16)
3- i sopropylglutaric acid (A3) 1.57 (0.05-7.19) 0.74 (0.05-3.03) 2.40 (0.10-7.19) 3.44 (0.09-13.50) 2.79 (0.21-9.62) 4.14 (0.09-13.50)
3-Methyl-1,2,3-butanetricarboxylic
acid (A4)
1.26 (0.01-8.58) 0.41 (0.01-1.82) 2.34 (0.07-8.58) 3.51 (0.05-15.58) 2.45 (0.07-11.69) 5.03 (0.11-15.58)
Tota l 6.82 2.92 11.41 17.68 13.68 22.88
β-Caryophyllene oxidation product
β-Caryophyllinic acid ( βC) 0.71 (0.01-3.29) 0.54 (0.01-1.86) 0.62 (0.03-2.13) 0.44 (0.01-1.66) 0.42 (0.01-1.53) 0.47 (0.02-1.34)
Tota l 0.71 0.54 0.62 0.44 0.42 0.47
Dicarboxylic acids
Oxalic acid (diC
2
) 1.66 (0.02-6.74) 1.33 (0.02-4.44) 2.05 (0.26-6.74) 3.48 (0.03-70.59) 2.33 (0.19-10.04) 5.34 (0.71-70.59)
Adipic acid (diC
6
) 1.67 (0.08-7.48) 1.04 (0.08-4.86) 1.88 (0.17-3.62) 2.89 (0.85-17.75) 4.14 (0.92-17.75) 2.29 (0.85-8.48)
Azelaic acid (diC
9
) 10.30 (0.21-47.10) 12.94 (3.63-46.56) 8.03 (0.21-47.10) 5.48 (0.07-47.83) 6.64 (1.10-47.83) 5.36 (1.63-26.78)
Phthalic acid (PhA) 3.41 (0.52-13.73) 4.26 (0.75-7.71) 2.61 (0.52-7.71) 6.65 (0.29-39.12) 13.20 (3.07-39.12) 3.35 (0.50-8.22)
Tota l 17.04 19.57 14.57 18.50 26.31 16.34
two industrial sites in Ohio for the summer since the authors
could not identify these compounds during the cold season
( Rutter et al., 2014 ). In ZK and FA , MT1 and MT2 showed a very
good correlation ( R
2 = 0.94 at ZK and R
2 = 0.96 at FA) along
with an MT2/MT1 ratio equal to 2.8 at FA and 2.6 at ZK (Fig. S3) .
These values are consistent with the value of 2.77 reported by
Ion et al. (2005) suggesting that the 2-methyltetrols have the
same photochemical reaction scheme, originating from the
direct oxidation of isoprene, and their formation rate is rel-
atively constant during the sampling period ( Ding et al., 2011 ;
Zhu et al., 2018 ).
The levels of 2-MGA averaging at 0.47 and 0.86 ng/m
3 at
ZK and FA respectively were lower than the 2-methyltetrols
( Tab le 4 ) at both sites and through the seasons. They are in
the range of the concentrations reported for a Mediterranean
urban industrial site (0.027-5.9 ng/m
3
) in Marseille ( El Haddad
et al., 2011a ) and lower than the 1.29 ng/m
3 observed at an
urban site in Shanghai, China ( Zhu et al., 2018 ).
In addition to that, 2-MGA showed a different time series
from the two diastereoisomers MT1 and MT2 at both sites
(Fig. S4) but also higher concentrations during summer and
a low correlation with their sum ( R
2 = 0.50 in ZK and 0.57
in FA) . Originating from the oxidation of an isoprene rst-
generation product methacrolein, the formation of 2-MGA in
the atmosphere strongly depends among others on humid-
ity, acidity, and NO
x
conditions. Oxidation of isoprene under
low NOx conditions preferentially lead to the formation of the
MT while high NOx conditions favors the 2-MGA formation
( Hallquist et al., 2009 ). Thus, the MT/2-MGA ratio might give
an insight on the variation of local conditions. At ZK and FA,
the MT/2-MGA ratio increased respectively from 11.9 and 10.9
in winter to 16.7 and 13.5 in summer. The high MT/2-MGA ratio
value at both sites let suggest a predominance of the MT for-
mation pathway. The slight increase at both sites in the sum-
mer period might be associated to the change of conditions in
aerosol acidity or the gas-phase partitioning of the precursors
of these compounds ( Fu et al., 2014 ).
2.2.2. α-Pinene oxidation products
α-Pinene is considered as the main species in the monoter-
pene class and is emitted from conifers ( Guenther et al., 1995 ).
The average total concentration of α-pinene derived SOA were
2.92 and 11.41 ng/m
3 at ZK and 13.68 and 22.88 ng/m
3 at
FA during winter and summer periods respectively. These
values are higher than those presented for two urban sites
in Shanghai (0.8-0.9 ng/m
3 in January and 8.0-10.0 ng/m
3 in
July) ( Feng et al., 2013 ). The concentrations in summer at FA
are also higher than the 16.8 ng/m
3 found in a study con-
ducted in July in western Germany while those at ZK are lower
( Kourtchev et al., 2008 ). Generally, higher concentrations were
journal of environmental sciences 101 (2021) 98–116 111
recorded at FA at both seasons due a greater coverage of plants
and vegetation in the Chekka region.
The oxidation of α-pinene by OH radicals or its ozonoly-
sis leads to pinic acid (PA) and pinonic acid which are con-
sidered as lower -or rst- generation oxidation products. The
concentrations of PA ( Tabl e 4 ) at both sites were found lower
than the 16 ng/m
3
reported at a semi-urban site during sum-
mer ( Wake d et al., 2014 ) and winter (9.87 ng/m
3
) in Lebanon
( Wak ed et al., 2013 ). It is also lower than 4.2 ng/m
3 found in
western Germany ( Kourtchev et al., 2008 ).
For A1, A2, A3, and A4, the average concentration in ZK
PM
2.5
were 0.65, 0.62, 0.74, and 0.41 ng/m
3
respectively in win-
ter and about 5 times higher in the summer period. The same
observation was highlighted in FA with twice higher concen-
trations in the hot period ( Tabl e 4 ) . Ty p i c a l l y, higher tempera-
ture and more solar radiations during the summer period en-
hance photochemical reactions. In addition to that, A1, A2, A3
and A4 concentrations were highly inter-correlated ( R
2
= 0.77-
0.85 at ZK and 0.78-0.85 at FA) during the total sampling pe-
riod (Table S2) emphasizing their origins as second generation
oxidation products of α-pinene. It is well established that A1
and A4 are generated by further reaction of pinonic acid in-
cluding OH radical and NO
x
( Hallquist et al., 2009 ). El Haddad
et al. (2011a) reported a good correlation between these four
compounds suggesting a similar formation process. However,
even though belonging to the same class, these compounds
did not show any correlation with pinic acid ( R
2
= 0.10-0.25)
probably due to the different formation pathway and the ki-
netics of the oxidation processes ( El Haddad et al., 2011a ).
2.2.3. β-Caryophyllene oxidation product
The β-caryophyllene ( βC) is considered as the most abundant
species in the sesquiterpene class emitted from plants. The
sesquiterpenes are characterized by their high reactivity and
their low vapor pressure ( Fu et al., 2010 ). β-caryophyllinic acid
was rst identied in smog chambers then in ambient sam-
ples ( Jaoui et al., 2007 ). It is considered as the major oxidation
product and the molecular marker of β-caryophyllene emitted
by terrestrial vegetations.
In this study, the levels of β-caryophyllinic acid were higher
in ZK (0.71 ng/m
3
) than FA (0.44 ng/m
3
). βC did not show any
seasonal variations at both sites. The same observation was
made during an aircraft campaign over central China ( Fu et al.,
2014 ) and in an urban and suburban site in Shanghai, China
( Feng et al., 2013 ). The total average concentrations at both
sites are slightly higher than the 0.27 ng/m
3
reported in Mar-
seille, France ( El Haddad et al., 2011a ) in summer 2008 and
comparable with the 0.7 ng/m
3 observed at an urban site in
Shanghai ( Fu et al., 2014 ). However, much higher concentra-
tions were observed at a suburban site in Lebanon with 10.59
ng/m
3 in the summer and 1.21 ng/m
3 in the winter period
( Wak ed et al., 2013 ; 2014 ). This difference might be explained
by the biomass burning activities in the suburban site leading
to higher β-caryophyllinic acid concentrations ( Fu et al., 2014 ).
2.2.4. Contribution of biogenic sources to the secondary or-
ganic carbon
The apportionment of the Secondary Organic Carbon (SOC) to
the different BSOA is based on the SOA tracer method pro-
posed by Kleindienst et al. (2007) in order to determine the
highest contributor to the the organic carbon ( Wak ed et al.,
2014 ) among the isoprene, α-pinene, and β-caryophyllene.
Briey, the measured concentrations of tracer compounds
derived from a given hydrocarbon precursor were con-
verted into SOC concentrations by using laboratory gener-
ated mass fractions of the same tracers (ratio of the trac-
ers/SOC determined in smog chambers). The considered
values for the mass fractions were 0.155 ±0.039 for iso-
prene, 0.231 ±0.111 for α-pinene and 0.023 ±0.005 for β-
caryophyllene ( Kleindienst et al., 2007 ). In addition to that, to
assess the SOC contribution of monoterpenes and sesquiter-
penes, the SOC estimates of α-pinene and β-caryophyllene
were multiplied by 3.2 and 3.6 respectively ( Geron and
Arnts, 2010 ; Ormeño et al., 2007 ; Wak ed et al., 2014 ). This
method holds high uncertainties due to limiting the complex
chemistry behind the SOA formation to a simplied single
value for each precursor. This replacement cannot cover nei-
ther the whole range of compounds emitted from the precur-
sor nor the meteorological conditions ( Kleindienst et al., 2007 ;
Wak ed et al., 2014 ). However, the method remains a valuable
approach to give insights into apportioning BSOC fractions.
Fig. 7 shows the contribution of isoprene, monoterpenes,
and sesquiterpenes to the SOC at both sites. A clear seasonal
pattern is evidenced for the SOC concentrations and could be
related to the higher concentrations of the precursors in sum-
mer. The total SOC concentrations in FA account for 469 ng/m
3
in the summer period and 255 ng/m
3
in winter. These values
are higher than ZK (305 ng/m
3
in summer and 131 ng/m
3
in
winter) but lower than those reported for a suburban site in
Lebanon (3408 ng/m
3
in summer 2011 and 462 ng/m
3
in win-
ter 2012) ( Waked et al., 2013 ; 2014 ). In the latter study, the site
was located in the suburb of the capital Beirut and was mainly
characterized by a high residential density, commercial activ-
ities and by forested trees surrouding the sampling site.
As discussed before, the anthropogenic contribution of
all the primary species is more important than the biogenic
ones in ZK and FA due to the industrial typology of the sites
while the Chekka area is more affected by forested pine lands.
Monoterpenes are found to be the largest contributors to SOC
especially in summer accounting for 47% and 63% in ZK and
FA respectively.
2.2.5. Dicarboxylic acids
Dicarboxylic acids are part of the water-soluble organic com-
pounds. Due to their low vapor pressures, they are mainly
present in the particulate phase ( Li et al., 2006 ). The concen-
trations of oxalic acid recorded at both sites were 1.33 ng/m
3
and 2.33 ng/m
3
in winter, 2.05 ng/m
3
and 5.34 ng/m
3
in sum-
mer at ZK and FA respectively. The higher concentrations in
summer could be due, as for other SOA, to enhanced photo-
chemical reactions by more intense solar radiations, and fa-
vorizing the decomposition of succinic acid to malonic and
oxalic acid ( Hsieh et al., 2007 ). Wake d et al. (2013) ; (2014) re-
ported much higher concentrations of this compound in sum-
mer (67.8 ng/m
3
) and in winter (14.1 ng/m
3
) in Mansourieh,
Lebanon: a suburban site located in the suburbs of Beirut.
Generally, oxalic acid is considered as the dominant species
between the dicarboxylic acids because of its stability and
its production by atmospheric oxidation of other dicarboxylic
acids with higher number of carbons ( Yu et al., 2019 ).
112 journal of environmental sciences 101 (2021) 98–116
Fig. 7 –The SOC contributions (in ng/m
3
) of isoprene, monoterpenes, and sesquiterpenes at Zouk(ZK) and Fiaa(FA) during
summer and winter periods.
In this study, Azelaic acid (diC
9
) is the most abundant dicar-
boxylic acid at ZK with concentrations of 12.9 and 8.03 ng/m
3
and the second most abundant with 6.64 and 5.36 ng/m
3
at FA
in the winter and the summer periods respectively. The sum-
mer values are higher than the 4.2 ng/m
3
observed at central
Alaska during the hot season ( Deshmukh et al., 2018 ). Azelaic
acid is generated by the oxidation process of biogenic unsat-
urated fatty acids such as oleic acid. The high concentrations
might be explained by the intense cooking activities in ZK and
FA as mentioned in Section 3.1.4 .
As for the adipic acid (diC
6
), higher concentrations ( Tabl e 4 )
were found compared to the value of 2 ng/m
3
reported in the
city of Philadelphia ( Ray and McDow, 2005 ) except for 1.04
ng/m
3
observed in ZK during winter. Adipic acid (diC
6
) is pro-
duced by the photooxidation of cyclohexene via ozone and
OH reactions ( Kawamura and Yas ui, 2005 ). Cyclohexene can
be found in motor exhausts revealing the anthropogenic na-
ture of the compound ( Grosjean and Fung, 1984 ).
Additionally, the averag e concentrations of phthalic acid
(PhA) were 3.41 and 6.65 ng/m
3 at ZK and FA respectively.
These values are higher than the 2.4 ng/m
3 reported for a
site in southern Sweden ( Hyder et al., 2012 ). PhA is generally
produced by secondary photochemical reaction with PAHs,
specically naphthalene, but can also be emitted directly from
combustion sources ( Nguyen et al., 2016 ).
As mentioned earlier, PhA and diC
6 are known to be
the oxidation products of compounds emitted by anthro-
pogenic sources while diC
9 is mainly from biogenic activi-
ties. In order to qualitatively evaluate the evolution of the
strength of the anthropogenic versus the biogenic sources of
the diacids, the diC
6
/diC
9
and the PhA/diC
9
ratios can be used
( Kunwar et al., 2019 ; Meng et al., 2013 ). Higher ratios will be ob-
served for samples that are more inuenced by anthropogenic
sources.
The average values of the diC
6
/diC
9
and the PhA/diC
9
ratios
increased in ZK from 0.09 and 0.38 respectively in winter sam-
ples to 0.44 and 0.54 in summer samples. The higher PhA/diC
9
ratio is explained by the lower concentrations of diC
9 in the
summer period which might be related to the lower concen-
trations of one of its precursors such as oleic acid ( Table 2 )
in ZK. This assumption can also stand for the diC
6
/diC
9
ratio
with also higher diC
6
concentrations that might be caused by
an increase in the vehicular emissions during this period.
On the contrary, the ratios decreased in FA from 1.28 and
2.91 in winter to 0.52 and 0.71 in summer for diC
6
/diC
9 and
PhA/diC
9
, respectively. This might be due to a decrease in an-
thropogenic emissions due to the shutdown of the cement
factories in this period (lower PhA and diC
6
concentrations in
summer while diC
9
is almost constant).
All of these ndings regarding SOA showed rstly the com-
plexity of the atmospheric processes and secondly that the
composition of this part of the aerosol strongly depend on
local and seasonal factors affecting the primary emissions
and the photooxidation conditions. Thus, trees, plants, and
vegetation around the site affect the emissions of the Bio-
genic Volatile Organic Compounds leading to variations in
the concentrations of their oxidation products. In addition to
that, meteorological factors and local atmospheric chemistry
(aerosol acidity, NO
x conditions, solar radiation intensity…)
inuence the formation of the SOA to a high extent.
3. Conclusions
The analysis of the organic fraction in PM
2.5
collected over a
one year period in Lebanon in the urban-industrial areas of
Zouk Mikael and Chekka region, in particular Fiaa, revealed
signicant variations between the sites on the concentration
journal of environmental sciences 101 (2021) 98–116 113
levels, the potential sources, and the seasonal variations of the
organic compounds. The most abundant class of compounds
was the fatty acids, that is part of the primary organic aerosols
(POA), emitted mainly from cooking activities.
For most of the POA, ZK recorded higher concentrations
due to the more urbanized, residential and industrial inu-
ence than FA. The petrogenic source at both sites was high-
lighed considering the concentrations of the n-alkanes, CPI,
Wax % and C
max with a low contribution of the primary bio-
genic emissions especially in winter. The fuel combustion in
ZK might be assigned to vehicular emissions, diesel gener-
ators and most importantly the HFO combustion from the
power plant by observing the PAHs concentrations and their
corresponding diagnostic ratios. In addition to the PAHs and
n-alkanes, the hopanes identied in ZK underline the impor-
tance of the road trafc in the region. On the other hand, the
main sources highlighted in Chekka region were a mix of coke
and fuel burning. The variation in the concentrations of ph-
thalates at ZK and FA suggests plastic incineration near the
study areas alongside a contribution from the plastic produc-
tion industries in ZK during summer.
For the secondary organic aerosols (SOA), higher concen-
trations were reported during summer at both sites due to
the enhancement of the photochemical reactions and high
temperatures. The α-pinene oxidation products were the most
abundant class in the secondary biogenic organic aerosols.
Even with low concentrations, BSOA compounds can largely
contribute to the SOC. Due to higher vegetation and forests in
the Chekka region, higher concentrations of the BSOA were
noted during both seasons. In addition, industrial activities
have an important inuence on the SOA formation especially
for compounds having their precursors emitted from anthro-
pogenic sources such as phthalic and adipic acids.
Finally, following the high concentrations of the different
organic compounds classes of which some are of particular
interest to the health eld due to their high toxicity, further in-
vestigation would be needed specially source apportionment
studies and health risk assessment for a better air quality
management planning.
Declaration of Competing Interest
The authors declare that they have no known competing -
nancial interests or personal relationships that could have ap-
peared to inuence the work reported in this paper.
Acknowledgments
The authors would like to acknowledge the National Coun-
cil for Scientic Research of Lebanon (CNRS-L) for granting
a doctoral fellowship to Marc Fadel. This project was also
funded by the Research Council and the Faculty of Sciences
of Saint Joseph University of Beirut –Lebanon. The “Unité
de Chimie Environnementale et Interactions sur le Vivant”
(UCEIV-UR4492) participates in the CLIMIBIO project, which is
nancially supported by the Hauts-de-France Region Council,
the French Ministry of Higher Education and Research, and the
European Regional Development Funds.
Supplementary materials
Supplementary material associated with this article can be
found, in the online version, at doi:10.1016/j.jes.2020.07.030 .
Abbas, I., Chaaban, J., Al-Rabaa, A.-R., Shaar, A., Shaar, A., 2019.
Solid Was te Management in Lebanon: Challenges and
Recommendations. Int. J. Environ. Waste Manag. 4 (2), 053–063 .
Abdallah, C., Af, C., Sauvage, S., Borbon, A., Salameh, T. ,
Kfoury, A., et al., 2020. Determination of gaseous and
particulate emission factors from road transport in a Middle
Eastern capital. Tr ans p. Res. D Tran sp. Envrion. 83, 102361 .
Af, C., Chélala, C., Borbon, A., Abboud, M., Gerard, J., Farah, W.,
et al., 2008. SO2 in Beirut: air quality implication and effects of
local emissions and long-range transport. Air Qual. Atmos.
Health 1, 167–178 .
Anderson, J.O., Thundiyil, J.G., Stolbach, A., 2012. Clearing the air:
a review of the effects of particulate matter air pollution on
human health. J. Med. To xic o l . 8, 166–175 .
Andreou, G., Rapsomanikis, S., 2009. Origins of n-alkanes,
carbonyl compounds and molecular biomarkers in
atmospheric ne and coarse particles of Athens, Greece. Sci.
Total Environ 407, 5750–5760 .
Annegarn, H., Marcazzan, G.M., Cereda, E., Marchionni, M.,
Zucchiatti, A., 1992. Source proles by unique ratios (SPUR)
analysis: Determination of source proles from receptor-site
streaker samples. Atmos. Environ. 26, 333–343 .
Atkinson, R., 2008. Our Present Understanding of the Gas-Phase
Atmospheric Degradation of Vo cs. Springer, Netherlands,
Dordrecht .
Baalbaki, R., Nassar, J., Salloum, S., Shihadeh, A.L., Lakkis, I.,
Saliba, N.A., 2018. Comparison of atmospheric polycyclic
aromatic hydrocarbon levels in three urban areas in Lebanon.
Atmos. Environ 179, 260–267 .
Badran, G., Ledoux, F., V erd i n , A., Abbas, I., Roumie, M.,
Genevray, P. , e t a l ., 2020. Toxicity of ne and quasi-ultrane
particles: Focus on the effects of organic extractable and
non-extractable matter fractions. Chemosphere 243, 125440 .
Bi, C., Chen, Y., Zha o , Z., Li, Q., Zhou, Q., Ye, Z., et al., 2020.
Characteristics, sources and health risks of toxic species
(PCDD/Fs, PAHs and heavy metals) in PM2.5 during fall and
winter in an industrial area. Chemosphere 238, 124620 .
Boonyatumanond, R., Murakami, M., Wattayakorn, G., Togo, A.,
Tak ada , H., 2007. Sources of polycyclic aromatic hydrocarbons
(PAHs) in street dust in a tropical Asian mega-city, Bangkok,
Thailand. Sci. Total Environ 384, 420–432 .
Borgie, M., Dagher, Z., Ledoux, F., V erd i n , A., Cazier, F., Ma r t i n , P. ,
et al., 2015. Comparison between ultrane and ne particulate
matter collected in Lebanon: Chemical characterization, in
vitro cytotoxic effects and metabolizing enzymes gene
expression in human bronchial epithelial cells. Environ. Pol.
205, 250–260 .
Borgie, M., Ledoux, F., D a g h e r, Z., Verdin, A., Cazier, F., Courcot, L.,
et al., 2016. Chemical characteristics of PM2.5–0.3 and PM0.3
and consequence of a dust storm episode at an urban site in
Lebanon. Atmos. Res. 180, 274–286 .
Bray, E.E., Evans, E.D., 1961. Distribution of n-parafns as a clue to
recognition of source beds. Geochim. Cosmochim. Acta. 22,
2–15 .
Carlton, A.G., Wiedinmyer, C., Kroll, J.H., 2009. A review of
Secondary Organic Aerosol (SOA) formation from isoprene.
Atmos. Chem. Phys. 9, 4987–5005 .
114 journal of environmental sciences 101 (2021) 98–116
Cazier, F. , Gene v r ay, P., Dewaele, D., Nouali, H., Verdin, A.,
Ledoux, F., e t a l ., 2016. Characterisation and seasonal
variations of particles in the atmosphere of rural, urban and
industrial areas: Organic compounds. J. Environ. Sci. 44, 45–56 .
Cecinato, A., Guerriero, E., Balducci, C., Muto, V. , 2014. Use of the
PAH ngerprints for identifying pollution sources. Urban Clim
10, 630–643 .
Claeys, M., Graham, B., Vas, G., Wang, W., Vermeylen, R.,
Pashynska, V. , C., J., Guyon, P., Andreae, M.O., Artaxo, P.,
Maenhaut, W., 2004. Formation of Secondary Organic Aerosols
Through Photooxidation of Isoprene. Sci. Mag. 303, 1173–1176 .
Cooper, J.E., Bray, E.E., 1963. A postulated role of fatty acids in
petroleum formation. Geochim. Cosmochim. Acta. 27,
1113–1127 .
Daher, N., Saliba, N.A., Shihadeh, A.L., Jaafar, M., Baalbaki, R.,
Sioutas, C., 2013. Chemical composition of size-resolved
particulate matter at near-freeway and urban background
sites in the greater Beirut area. Atmos. Environ. 80, 96–106 .
De Gouw, J., Jimenez, J.L., 2009. Organic Aerosols in the Earth’s
Atmosphere. Environ. Sci. Technol. 43, 7614–7618 .
Deshmukh, D.K., Mozammel Haque, M., Kawamura, K., Kim, Y. ,
2018. Dicarboxylic acids, oxocarboxylic acids and
α-dicarbonyls in ne aerosols over central Alaska:
Implications for sources and atmospheric processes. Atmos.
Res. 202, 128–139 .
Ding, X., Wang, X.-M., Zheng, M., 2011. The inuence of
temperature and aerosol acidity on biogenic secondary
organic aerosol tracers: Observations at a rural site in the
central Pearl River Delta region, South China. Atmos. Environ
45, 1303–1311 .
DRI, 2003. Analysis of Semi-volatile Organic Compound by
GC/MS. Desert Research Institute, DRI Standard Operating
Procedure., 1–25.
Du, Y., X u , X., Chu, M., Guo, Y., Wang, J., 2016. Air particulate
matter and cardiovascular disease: the epidemiological,
biomedical and clinical evidence. J. Thorac. Dis. 8, E8–E19 .
El Haddad, I., Marchand, N., Dron, J., Temime-Roussel, B.,
Quivet, E., Wortham, H., et al., 2009. Comprehensive primary
particulate organic characterization of vehicular exhaust
emissions in France. Atmos. Environ. 43, 6190–6198 .
El Haddad, I., Marchand, N., Temime-Roussel, B., Wortham, H.,
Piot, C., Besombes, J.L., et al., 2011a. Insights into the
secondary fraction of the organic aerosol in a Mediterranean
urban area: Marseille. Atmos. Chem. Phys. 11, 2059–2079 .
El Haddad, I., Marchand, N., Wortham, H., Piot, C., Besombes, J.L.,
Cozic, J., et al., 2011b. Primary sources of PM2.5 organic aerosol
in an industrial Mediterranean city, Marseille. Atmos. Chem.
Phys. 11, 2039–2058 .
Feng, J., Li, M., Zhang, P. , G ong, S., Zhong, M., Wu, M., et al., 2013.
Investigation of the sources and seasonal variations of
secondary organic aerosols in PM2.5 in Shanghai with organic
tracers. Atmos. Environ. 79, 614–622 .
Fu, P., Kawamura, K., Kanaya, Y., Wang, Z., 2010. Contributions of
biogenic volatile organic compounds to the formation of
secondary organic aerosols over Mt. Tai , Central East China.
Atmos. Environ. 44, 4817–4826 .
Fu, P.Q., Kawamura, K., Cheng, Y.F., Hatakeyama, S., Takami, A.,
Li, H., et al., 2014. Aircraft measurements of polar organic
tracer compounds in tropospheric particles PM10 over central
China. Atmos. Chem. Phys. 14, 4185–4199 .
Gadi, R., Shivani, Sharma, S.K., Mandal, T.K . , 2019. Source
apportionment and health risk assessment of organic
constituents in ne ambient aerosols (PM2.5): A complete
year study over National Capital Region of India.
Chemosphere 221, 583–596 .
Geron, C.D., Arnts, R.R., 2010. Seasonal monoterpene and
sesquiterpene emissions from Pinus taeda and Pinus
virginiana. Atmos. Environ. 44, 4240–4251 .
Grosjean, D., Fung , K., 1984. Hydrocarbons and Carbonyls in Los
Angeles Air. J Air Pollut. Control Assoc. 34, 537–543 .
Guenther, A., Hewitt, C.N., Erickson, D., Fall, R., Geron, C.,
Graedel, T. , et al. , 1995. A global model of natural volatile
organic compound emissions. J. Geophys. Res. Atmos. 100,
8873–8892 .
Guo, H., 2003. Particle-associated polycyclic aromatic
hydrocarbons in urban air of Hong Kong. Atmos. Environ. 37,
5307–5317 .
Hallquist, M., Wenger, J.C., Baltensperger, U., Rudich, Y.,
Simpson, D., Claeys, M., et al., 2009. . The formation, properties
and impact of secondary organic aerosol: Current and
emerging issues. Atmos. Chem. Phys. 9, 5155–5236 .
Henry, R.C., Lewis, C.W., Hopke, P.K., Williamson, H.J., 1984. Review
of receptor model fundamentals. Atmos. Environ. 18,
1507–1515 .
Hsieh, L.-Y., Kuo, S.-C., Chen, C.-L., Tsai, Y.I. , 2007. Origin of
low-molecular-weight dicarboxylic acids and their
concentration and size distribution variation in suburban
aerosol. Atmos. Environ. 41, 6648–6661 .
Hyder, M., Genberg, J., Sandahl, M., Swietlicki, E., Jönsson, J.
˚
A.,
2012. Yearly trend of dicarboxylic acids in organic aerosols
from south of Sweden and source attribution. Atmos. Environ.
57, 197–204 .
Ion, A.C., Vermeylen, R., Kourtchev, I., Cafmeyer, J., Chi, X.,
Gelencsér, A., et al., 2005. Polar organic compounds in rural
PM2.5 aerosols from K-puszta, Hungary, during a 2003
summer eld campaign: Sources and diel variations. Atmos.
Chem. Phys 5, 1805–1814 .
Jaafar, M., Baalbaki, R., Mrad, R., Daher, N., Shihadeh, A.,
Sioutas, C., et al., 2014. Dust episodes in Beirut and their effect
on the chemical composition of coarse and ne particulate
matter. Sci. Total Environ. 496, 75–83 .
Jaoui, M., Lewandowski, M., Kleindienst, T.E., Offenberg, J.H.,
Edney, E.O., 2007. β-caryophyllinic acid: An atmospheric tracer
for β-caryophyllene secondary organic aerosol. Geophys. Res.
Lett 34 .
Ji, Y., Wang, F., Zhang, L., Shan, C., Bai, Z., Sun, Z., et al., 2014. A
comprehensive assessment of human exposure to phthalates
from environmental media and food in Tianjin, China. J.
Hazard. Mater. 279, 133–140 .
Kawamura, K., Yasui, O., 2005. Diurnal changes in the distribution
of dicarboxylic acids, ketocarboxylic acids and dicarbonyls in
the urban Toky o atmosphere. Atmos. Environ. 39, 1945–1960 .
Kfoury, A., Ledoux, F. , E l K h o u ry, B., Nakat, H., Nouali, H., Cazier, F. ,
et al., 2009. A study of the inorganic chemical composition of
atmospheric particulate matter in the region of Chekka, North
Lebanon. Leban. Sci. J. 10, 3–16 .
Khalili, N.R., Scheff, P.A., Holsen, T.M., 1995. PAH source
ngerprints for coke ovens, diesel and, gasoline engines,
highway tunnels, and wood combustion emissions. Atmos.
Environ. 29, 533–542 .
Kim, Y.-H., Kim, K.-H., 2015. A simple methodological validation
of the gas/particle fractionation of polycyclic aromatic
hydrocarbons in ambient air. Sci. Rep. 5, 11679 .
Kleindienst, T.E., Jaoui, M., Lewandowski, M., Offenberg, J.H.,
Lewis, C.W., Bhave, P.V., e t a l ., 2007. . Estimates of the
contributions of biogenic and anthropogenic hydrocarbons to
secondary organic aerosol at a southeastern US location.
Atmos. Environ. 41, 8288–8300 .
Kotianova, P., Puxbaum, H., Bauer, H., Caseiro, A., Marr, I., Cik, G.,
2008. Temporal patterns of n-alkanes at trafc exposed and
suburban sites in Vienna. Atmos. Environ. 42, 2993–3005 .
Kourtchev, I., Warnke, J., Maenhaut, W., Hoffmann, T. , Cla eys , M.,
2008. Polar organic marker compounds in PM2.5 aerosol from
a mixed forest site in western Germany. Chemosphere 73,
1308–1314 .
journal of environmental sciences 101 (2021) 98–116 115
Kroll, J.H., Seinfeld, J.H., 2008. Chemistry of secondary organic
aerosol: Formation and evolution of low-volatility organics in
the atmosphere. Atmos. Environ. 42, 3593–3624 .
Kunwar, B., Kawamura, K., Fujiwara, S., Fu, P., Miyazaki, Y.,
Pokhrel, A., 2019. Dicarboxylic acids, oxocarboxylic acids and
α-dicarbonyls in atmospheric aerosols from Mt. Fuji, Japan:
Implication for primary emission versus secondary formation.
Atmos. Res. 221, 58–71 .
Landkocz, Y., Ledoux, F. , A ndré , V., C a z i e r, F., Gen e v ray, P. ,
Dewaele, D., et al., 2017. Fine and ultrane atmospheric
particulate matter at a multi-inuenced urban site:
Physicochemical characterization, mutagenicity and
cytotoxicity. Environ. Pollut. 221, 130–140 .
Ledoux, F., Kfoury, A., Delmaire, G., Roussel, G., El Zein, A.,
Courcot, D., 2017. Contributions of local and regional
anthropogenic sources of metals in PM2.5 at an urban site in
northern France. Chemosphere 181, 713–724 .
Lelieveld, J., Pozzer, A., Pöschl, U., Fnais, M., Haines, A., Münzel, T.,
2020. Loss of life expectancy from air pollution compared to
other risk factors: a worldwide perspective. Cardiovasc. Res.
Li, M., McDow, S.R., Tollerud, D.J., Mazurek, M.A., 2006. Seasonal
abundance of organic molecular markers in urban particulate
matter from Philadelphia, PA . Atmos. Environ 40, 2260–2273 .
Li, W., Peng, Y., Bai , Z., 2010. Distributions and sources of
n-alkanes in PM2.5 at urban, industrial and coastal sites in
Tianjin, China. J. Environ. Sci. 22, 1551–1557 .
Lu, S., Kang, L., Liao, S., Ma, S., Zhou, L., Chen, D., e t al., 2018.
Phthalates in PM2.5 from Shenzhen, China and human
exposure assessment factored their bioaccessibility in lung.
Chemosphere 202, 726–732 .
Manoli, E., Kouras, A., Samara, C., 2004. Prole Analysis of
Ambient and Source Emitted Particle-Bound Polycyclic
Aromatic Hydrocarbons from Three Sites in Northern Greece.
Chemosphere 56, 867–878 .
Masclet, P., Bresson, M.A., Mouvier, G., 1987. Polycyclic aromatic
hydrocarbons emitted by power stations, and inuence of
combustion conditions. Fuel 66, 556–562 .
Mastral, A., Callén, M., Lopez, J., Murillo, R., Garcia, T., N ava r ro , M.,
2003. Critical review on atmospheric PA H. Assessment of
reported data in the Mediterranean basin. Fuel Process.
Technol. 80, 183–193 .
Mastral, A.M., Callén, M.S., 2000. A Review on Polycyclic Aromatic
Hydrocarbon (PAH) Emissions from Energy Generation.
Environ. Sci. Technol. 34, 3051–3057 .
Melki, P., 2017. Health impact of airborne particulate matter in
Northern Lebanon: from a pilot epidemiological study to
physico-chemical characterization and toxicological effects
assessment.
Melki, P.N., Ledoux, F., Aou a d , S., Billet, S., El Khoury, B.,
Landkocz, Y., e t a l . , 2017. Physicochemical characteristics,
mutagenicity and genotoxicity of airborne particles under
industrial and rural inuences in Northern Lebanon. Environ.
Sci. Pollut. Res. Int. 24, 18782–18797 .
Meng, J., Wang, G., Li, J., Cheng, C., Cao, J., 2013. Atmospheric
oxalic acid and related secondary organic aerosols in Qinghai
Lake, a continental background site in Tibet Plateau. Atmos.
Environ. 79, 582–589 .
Mikuška, P., K
ˇ
r
˚
umal, K., Ve
ˇ
ce
ˇ
ra, Z., 2015. Characterization of
organic compounds in the PM2.5 aerosols in winter in an
industrial urban area. Atmos. Environ. 105, 97–108 .
Nguyen, L., Kawamura, K., Ono, K., Ram, S., Engling, G., Lee, C.-T.,
et al., 2016. Comprehensive PM2.5 Organic Molecular
Composition and Stable Carbon Isotope Ratios of Biomass
Burning Smoke at Sonla, Vi etnam: Fingerprint of Biomass
Burning Components. Aerosol Air Qual. Res 16 .
Ormeño, E., Fernandez, C., Bousquet-Mélou, A., Greff, S., Morin, E.,
Robles, C., et al., 2007. Monoterpene and sesquiterpene
emissions of three Mediterranean species through calcareous
and siliceous soils in natural conditions. Atmos. Environ. 41,
629–639 .
Pei, X.Q., Song, M., Guo, M., Mo, F.F. , S h e n, X.Y., 2013. Concentration
and risk assessment of phthalates present in indoor air from
newly decorated apartments. Atmos. Environ. 68, 17–23 .
Pindado, O., Pérez, R.M., García, S., Sánchez, M., Galán, P. ,
Fernández, M., 2009. Characterization and sources assignation
of PM2.5 organic aerosol in a rural area of Spain. Atmos.
Environ. 43, 2796–2803 .
Ravindra, K., Sokhi, R., Vangrieken, R., 2008. Atmospheric
polycyclic aromatic hydrocarbons: Source attribution,
emission factors and regulation. Atmos. Environ. 42,
2895–2921 .
Ray, J., McDow, S.R., 2005. Dicarboxylic acid concentration trends
and sampling artifacts. Atmos. Environ. 39, 7906–7919 .
Riffault, V., A r n dt, J., Marris, H., Mbengue, S., Setyan, A.,
Alleman, L.Y., et al., 2015. Fine and Ultrane Particles in the
Vicinity of Industrial Activities: A Review. Crit. Rev. Environ.
Sci. Technol. 45, 2305–2356 .
Robinson, A.L., Subramanian, R., Donahue, N.M.,
Bernardo-Bricker, A., Rogge, W. F. , 2006. Source Apportionment
of Molecular Markers and Organic Aerosol. 3. Food Cooking
Emissions. Environ. Sci. Technol 40, 7820–7827 .
Rogge, W.F., Hildemann, L.M., Mazurek, M.A., Cass, G.R., 1996.
Mathematical modeling of atmospheric ne
particle-associated primary organic compound
concentrations. J. Geophys. Res 101 (19), 379 .
Rogge, W.F., Hildemann, L.M., Mazurek, M.A., Cass, G.R.,
Simoneit, B.R.T., 1991. Sources of ne organic aerosol. 1.
Charbroilers and meat cooking operations. Environ. Sci.
Technol 25, 1112–1125 .
Rogge, W.F., Hildemann, L.M., Mazurek, M.A., Cass, G.R.,
Simoneit, B.R.T., 1993a. Sources of ne organic aerosol. 2.
Noncatalyst and catalyst-equipped automobiles and
heavy-duty diesel trucks. Environ. Sci.Technol 27, 636–651 .
Rogge, W.F., Hildemann, L.M., Mazurek, M.A., Cass, G.R.,
Simoneit, B.R.T., 1993b. Sources of ne organic aerosol. 4.
Particulate abrasion products from leaf surfaces of urban
plants. Environ. Sci. Technol 27, 2700–2711 .
Ruehl, C.R., Ham, W.A., Kleeman, M.J., 2011. Temperature-induced
volatility of molecular markers in ambient airborne
particulate matter. Atmos. Chem. Phys. 11, 67–76 .
Rutter, A.P., Snyder, D.C., Stone, E.A., Shelton, B., DeMinter, J.,
Schauer, J.J., 2014. Preliminary assessment of the
anthropogenic and biogenic contributions to secondary
organic aerosols at two industrial cities in the upper Midwest.
Atmos. Environ. 84, 307–313 .
Salameh, T. , Sauvage, S., Af, C., Borbon, A., Léonardis, T. , Brioude,
J., et al., 2015 . Exploring the seasonal NMHC distribution in an
urban area of the Middle East during ECOCEM campaigns:
Ver y high loadings dominated by local emissions and
dynamics. Environ. Chem.
Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 2002.
Measurement of Emissions from Air Pollution Sources. 4.
C1 −C27 Organic Compounds from Cooking with Seed Oils.
Environ. Sci. Technol 36, 567–575 .
Seinfeld, J.H., Pandis, S.N., 2016. Atmospheric Chemistry and
Physics: From Air Pollution to Climate Change, 3r. John Wiley
& Sons.
Simoneit, B.R.T., 1989. Organic matter of the troposphere —V:
Application of molecular marker analysis to biogenic
emissions into the troposphere for source reconciliations. J.
Atmos. Chem. 8, 251–275 .
Simoneit, B.R.T., 1999. A review of biomarker compounds as
source indicators and tracers for air pollution. Environ. Sci.
Pollut. Res. 6, 159–169 .
116 journal of environmental sciences 101 (2021) 98–116
Simoneit, B.R.T., 2002. Biomass burning —a review of organic
tracers for smoke from incomplete combustion. J. Appl.
Geochem. 17, 129–162 .
Simoneit, B.R.T., Medeiros, P.M., Didyk, B.M., 2005. Combustion
Products of Plastics as Indicators for Refuse Burning in the
Atmosphere. Environ. Sci. Technol. 39, 6961–6970 .
Simoneit, B.R.T., Sheng, G., Chen, X., Fu, J., Zhang, J., Xu, Y., 1991.
Molecular marker study of extractable organic matter in
aerosols from urban areas of China. Atmos. Environ. 25,
2111–2129 .
Tan , J., Guo, S., Ma, Y., D u a n , J., Cheng, Y., H e , K., et al., 2011.
Characteristics of particulate PAHs during a typical haze
episode in Guangzhou, China. Atmos. Res. 102, 91–98 .
Tobiszewski, M., Namie
´
snik, J., 2012. PA H diagnostic ratios for the
identication of pollution emission sources. Environ. Pollut.
162, 110–119 .
USEPA, 2010. AP-42: Compilation of Air Pollutant Emission
Factors; Stationary Point and Area Sources, Fifth Edition. US
EPA edition.
Wak ed , A., Af, C., 2012. Emissions of air pollutants from road
transport in Lebanon and other countries in the Middle East
region. Atmos. Environ. 61, 446–452 .
Wak ed , A., Af, C., Brioude, J., Formenti, P., Chevaillier, S.,
Haddad, I.E., et al., 2013. Composition and Source
Apportionment of Organic Aerosol in Beirut, Lebanon, During
Winter 2012. Aerosol Sci. Technol 47, 1258–1266 .
Wak ed , A., Af, C., Formenti, P., Chevaillier, S., El-Haddad, I.,
Doussin, J.-F., et al., 2014. Characterization of organic tracer
compounds in PM2.5 at a semi-urban site in Beirut, Lebanon.
Atmos. Res. 143, 85–94 .
Wak ed , A., Af, C., Seigneur, C., 2012. An atmospheric emission
inventory of anthropogenic and biogenic sources for Lebanon.
Atmos. Environ. 50, 88–96 .
Wan g, J., Ho, S.S.H., Ma, S., Cao, J., Dai, W. , L i u , S., et al., 2016.
Characterization of PM2.5 in Guangzhou, China: uses of
organic markers for supporting source apportionment. Sci.
Total Environ. 550, 961–971 .
WHO, 2013. Health effects of particulate matter: policy
implications for countries in eastern Europe, Caucasus and
central Asia.
Wu, D., Wang, Z., Chen, J., Kong, S., Fu, X., Deng, H., et al., 2014.
Polycyclic aromatic hydrocarbons (PAHs) in atmospheric
PM2.5 and PM10 at a coal-based industrial city: Implication
for PAH control at industrial agglomeration regions, China.
Atmos. Res. 149, 217–229 .
Xing, Y.- F. , X u, Y.-H., Shi, M.-H., Lian, Y. - X ., 2016. The impact of
PM2.5 on the human respiratory system. J. Thorac. Dis. 8,
E69–E74 .
Yammine, P., Kfoury, A., El Beyrouthy, M., Nouali, H., El-Nakat, H.,
Ledoux, F., e t a l ., 2010. A preliminary evaluation of the
inorganic chemical composition of atmospheric TSP in the
Selaata region, North Lebanon. Leban. Sci. J. 11, 13–29 .
Yan g, H.-H., Lee, W.-J., Chen, S.-J., Lai, S.-O., 1998. PAH emission
from various industrial stacks. J. Hazard. Mater. 60, 159–174 .
Yu, Q., Chen, J., Qin, W., Cheng, S., Zhang, Y., Ahm a d , M., et al.,
2019. Characteristics and secondary formation of
water-soluble organic acids in PM1, PM2.5 and PM10 in Beijing
during haze episodes. Sci. Total Environ. 669, 175–184 .
Zaheer, J., Jeon, J., Lee, S.-B., Kim, J.S., 2018. Effect of Particulate
Matter on Human Health, Prevention, and Imaging Using PET
or SPECT. Prog. Med. Phys 29, 81 .
Zhu, W., Luo, L., Cheng, Z., Yan, N., Lou, S., Ma, Y., 2018.
Characteristics and contributions of biogenic secondary
organic aerosol tracers to PM2.5 in Shanghai, China. Atmos.
Pollut. Res. 9, 179–188 .
