Behavior and prediction of photochemical degradation of chlorinated polycyclic aromatic hydrocarbons in cyclohexane

Article abstract of DOI:10.1016/j.chemosphere.2007.08.064

The photochemical degradation of 11 chlorinated polycyclic aromatic hydrocarbons (ClPAHs) and the corresponding 5 parent PAHs was examined to simulate the compound's fate on aerosol surfaces. All the ClPAHs and PAHs decayed according to the first-order reaction rate kinetics. The photolysis rates of ClPAHs varied greatly according to the skeleton of PAHs; the rates of chlorophenanthrenes (ClPhes) and 1-chloropyrene were higher than those of corresponding parent PAHs, whereas chlorofluoranthenes, 7-chlorobenz[a]anthracene and 6-chlorobenzo[a]pyrene were more stable under irradiation compared to respective parent PAH. Considering the photoproducts of ClPhes detected, the oxidation could occur immediately at positions of the highest frontier electron density. Finally, the quantitative structure-property relationship models were developed for direct photolysis half-lives and average quantum yields of the ClPAHs and parent PAHs, in which the significant factors affecting photolysis were E_LUMO+1, total energy and surface area, and E_LUMO, E_LUMO - E_HOMO and total energy, respectively.

Full text of DOI:10.1016/j.chemosphere.2007.08.064

Available online at www.sciencedirect.com
Chemosphere 70 (2008) 2110–2117
www.elsevier.com/locate/chemosphere
Behavior and prediction of photochemical degradation of
chlorinated polycyclic aromatic hydrocarbons in cyclohexane
*
Takeshi Ohura , Takashi Amagai, Masakazu Makino
Institute for Environmental Sciences, University of Shizuoka, 52-1 Yada, Shizuoka 422-8526, Japan
Received 24 April 2007; received in revised form 29 August 2007; accepted 29 August 2007
Available online 23 October 2007
Abstract
The photochemical degradation of 11 chlorinated polycyclic aromatic hydrocarbons (ClPAHs) and the corresponding 5 parent PAHs
was examined to simulate the compound’s fate on aerosol surfaces. All the ClPAHs and PAHs decayed according to the first-order reac-
tion rate kinetics. The photolysis rates of ClPAHs varied greatly according to the skeleton of PAHs; the rates of chlorophenanthrenes
(
ClPhes) and 1-chloropyrene were higher than those of corresponding parent PAHs, whereas chlorofluoranthenes, 7-chloro-
benz[a]anthracene and 6-chlorobenzo[a]pyrene were more stable under irradiation compared to respective parent PAH. Considering
the photoproducts of ClPhes detected, the oxidation could occur immediately at positions of the highest frontier electron density. Finally,
the quantitative structure-property relationship models were developed for direct photolysis half-lives and average quantum yields of the
ClPAHs and parent PAHs, in which the significant factors affecting photolysis were ELUMO+1, total energy and surface area, and ELUMO
ELUMO ꢀ E and total energy, respectively.
,
HOMO
ꢀ
2007 Elsevier Ltd. All rights reserved.
Keywords: ClPAH; Photolysis; Half-lives; Quantum yield; QSPR
1
. Introduction
photochemical reaction of PAHs adsorbed to particles
could also occur in an organic layer surrounding the parti-
cle core (McDow et al., 1994; Ohura et al., 2004, 2005).
That is, the direct photolytic pathways of pollutant degra-
dation can be simulated in the laboratory by irradiation of
the compound in an organic solvent. The photostability of
various PAHs has been investigated in organic solvents
such as toluene (Jang and McDow, 1995, 1997), benzene
(Jang and McDow, 1997), and cyclohexane (Lamotte
et al., 1987).
Polycyclic aromatic hydrocarbons (PAHs) are ubiqui-
tous hazardous materials because of their mutagenic and
carcinogenic properties, and are distributed in the environ-
ment through combustion, discharge of fossil fuels, auto-
mobile emissions, and subsequent atmospheric transport
and deposition (Bidleman, 1988; Baek et al., 1991; Mantis
et al., 2005; L o¨ nnermark and Blomqvist, 2006; Pekey et al.,
2
007). Therefore, determination of the environmental fate
of PAHs is important for human risk assessment (Menzie
et al., 1992; Finlayson-Pitts and Pitts, 1997; Sabljic,
Environmental occurrences, behavior and toxicity of
PAH derivatives have been studied by many researchers.
As typical PAH derivatives, nitrated PAHs have been well
studied in the fields of environmental behavior because of
their high level of toxicity (Finlayson-Pitts and Pitts,
2000). Recently, the focus of attention has been on the
effect on the environment of chlorinated polycyclic aro-
matic hydrocarbons (ClPAHs) with more than 3 rings,
which have been found in urban air, snow, tap water,
and sediment (Ohura, 2007). Ohura and co-workers
2
001). The dominant degradation pathway of PAHs asso-
ciated to graphitic particles in the atmosphere is suggested
to be photolysis (Nielsen, 1984; Kamens et al., 1988;
McDow et al., 1994; Finlayson-Pitts and Pitts, 1997). The
*
Corresponding author. Tel.: +81 54 264 5789; fax: +81 54 264 5798.
E-mail address: ooura@smail.u-shizuoka-ken.ac.jp (T. Ohura).
0
045-6535/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chemosphere.2007.08.064

T. Ohura et al. / Chemosphere 70 (2008) 2110–2117
2111
(
Ohura et al., 2004, 2005; Kitazawa et al., 2006) have
species. The ClPAHs tested in this study are abbreviated as
follows: 9-chlorophenanthrene (9-ClPhe), 3,9-dichlorophe-
detected various ClPAHs associated with particles in urban
air, and discussed the fates and possible sources of emis-
sion. However, the characteristics of ClPAHs such as envi-
ronmental fate remain uncertain in comparison with those
of the parent PAHs. In this study, we first investigated the
photostability of ClPAHs using a chemical model system,
followed by comparison to the corresponding parent
PAHs. The effects of chlorination of PAHs are discussed
on the basis of the photolysis rates and products of each
test compound. Note that we used an inert solvent, cyclo-
hexane, for the investigation of photostability to reduce
the effects of reactivity of the solvent itself. Finally, the
photolysis rate and average quantum yield of each ClPAH
and PAH were predicted via a quantitative structure–prop-
erty relationships (QSPRs) model, which indicated the sig-
nificant molecular descriptors of the photodegradation.
nanthrene (3,9-Cl Phe), 9,10-dichlorophenanthrene (9,10-
2
Cl Phe), 3,9,10-trichlorophenanthrene (3,9,10-Cl Phe),
2
3
3-chlorofluoranthene (3-ClFluor), 8-chlorofluoranthene
(8-ClFluor), 3,4-dichlorofluoranthene (3,4-Cl Fluor), 3,
2
8-dichlorofluoranthene (3,8-Cl Fluor), 1-chloropyrene (1-
2
ClPy), 7-chlorobenz[a]anthracene (7-ClBaA), and 6-chloro-
benzo[a]pyrene (6-ClBaP). The structural formulae of these
ClPAHs are illustrated in Fig. 1. The synthesized ClPAHs
were >95% pure (as determined by GC–MS).
2.2. Photodegradation experiments
Photodegradation experiments were performed with a
turntable photoreactor (Ace Glass Inc., Vineland, NJ)
using a 450 W high-pressure mercury lamp (Ushio Co.,
Ltd., Tokyo, Japan, maximum wavelengths 313, 334, 365,
404, 435, 546, and 578 nm) as the light source, and a quartz
2
. Experimental
immersion well with circulating water. The immersion well
was surrounded by a Pyrex sleeve to filter out high-energy
UV bands (k < 290 nm). Although this setup does not
duplicate the spectral distribution of the actinic flux, it does
reproduce the wavelengths normally encountered in the
atmosphere. The photoreactor was positioned in a water
bath with constant water circulation and the temperature
in the bath was maintained at 25(± 1) ꢁC. The solutions
were irradiated in 13 mm · 100 mm quartz reaction tubes.
The quantum yields (/) for each ClPAH and PAH pho-
tolysis was determined as follows. The photolysis quantum
yield / for a certain wavelength k (nm) is defined as:
2
.1. Chemicals
We selected 5 PAH species as reference substances and
chlorinated each by the procedure of Dewhurst and
Kitchen (1972). The PAHs were selected because of their
atmospheric and toxicological relevance. A solution of
2
0 g of N-chlorosuccinimide (Wako Pure Chemical Indus-
tries, Ltd., Osaka, Japan) in 20 ml of propylene carbonate
PC, Wako Pure Chemical Industries) and a solution of
(
each PAH (ꢁ0.1 M) in 20 ml of PC were mixed and incu-
bated at 100 ꢁC for ꢁ3 h in the dark. The reaction solvent
for each PAH was fractionated by high-pressure liquid
chromatography (HPLC) (column, COSMOSIL 5C18-
AR; eluent, methanol), and the fractions corresponding
to the dominant peaks were isolated and analyzed by gas
ki
i
/ki ¼ Rateki ðIki Þꢀ1
abs
ð1Þ
) would be the degradation rate of
ꢀ
1
ꢀ1
where Rateki (mol l
s
the compound accounted for by the direct photolysis in-
1
abs
ki
chromatography–mass spectrometry (GC–MS) and by H
duced by the k emission line only, and I is the intensity
i
NMR spectroscopy (500 MHz, CDCl ), yielding 11 ClPAH
of radiation absorption by the compound on irradiation at
3
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
9
-chlorophenanthrene
3,9-dichlorophenanthrene
(#3)
9,10-dichlorophenanthrene
(#4)
3,910-trichlorophenanthrene
(#5)
(
#2)
Cl
Cl
Cl
Cl
Cl
Cl
3
-chlorofluoranthene
#7)
8-chlorofluoranthene
3,4-dichlorofluoranthene
(#9)
3,8-dichlorofluoranthene
(#10)
(
(#8)
Cl
Cl
Cl
1
-chloropyrene
#12)
7-chlorobenz[a]anthracene
6-chlorobenzo[a]pyrene
(
(#14)
(#16)
Fig. 1. Chemical structures of the chlorinated polycyclic aromatic hydrocarbons tested (the numbers correspond to those in Table 1).

2112
T. Ohura et al. / Chemosphere 70 (2008) 2110–2117
the wavelength k , which is expressed in equivalent units,
ous energy levels of the occupied and the unoccupied
molecular orbital, for example EHOMO, ELUMO, E ,
HOMOꢀ1
i
ꢀ1
ꢀ1
abs
Ein l
s
. In the present study, Iki was observed by using
multi channel photo detector (MCPD-3700, Otsuka Elec-
tronics, Co. Ltd., Osaka, Japan). The average quantum
yields /ave for ClPAH photolysis need to take into account
the radiation absorption intensity at the different wave-
lengths, that is:
and ELUMO+1, and the related parameters calculated from
them, ELUMO + EHOMO and ELUMO ꢀ EHOMO, the total
energy (TE), and the molecular surface area (SA). All of
them were calculated for the optimized molecular struc-
ture. Gauss ViewW ver.2.1 (Frisch et al., 2000) was used
for three-dimensional molecular modeling of the com-
pounds tested. The molecular structure was optimized by
Gaussian 98W ver.5.4 (Frisch et al., 1998). The quantum
chemical calculation was performed on the basis of density
functional theory in terms of B3LYP as the exchange and
abs
ki
abs ki
tot
abs
ki
abs
tot
X
X
X
I
I
I
I
Rateki
1
abs
/ave
¼
¼
/ ¼
¼
Rateki
abs
I
I
i
i
ki
tot
i
RatePh
ð2Þ
abs
tot
I
*
the correlation functions and the 3-21 basis set. Next, we
P
where RatePh
¼
iRateki is the experimentally measured
selected some significant descriptors, as judged from the
2
initial degradation rate of each ClPAH on direct
photolysis.
values of r , the square of correlation coefficient, and car-
ried out multiple linear regression analysis (MRA) in terms
of the descriptors to establish a robust prediction model.
Each ClPAH was dissolved in cyclohexane (HPLC
grade, Wako Pure Chemical Industries) at a concentration
of ꢁ0.15 mM. Samples were removed with a pipette, trans-
ferred to amber glass chromatography vials, and analyzed
with an HP 6890 GC-HP 5972 A MS system in scan mode
to identify the photolysis products. The products were ana-
3. Results and discussion
3.1. Photolysis of ClPAHs
lyzed with
0 m · 0.32 mm · 0.25 lm film thickness, J&W Scientific)
capillary column. Helium was used as the carrier gas at a
a DB-5 (5% phenyl-methylpolysiloxane,
The photodegradation of 11 ClPAH congeners and their
corresponding parent PAHs were investigated using
chemical model systems. Decay of the compounds can be
described by the first-order kinetics as follows:
6
5
ꢀ
1
flow-rate of 1.0 ml min . The oven temperature was kept
at 80 ꢁC for 2 min, then increased from 80 ꢁC to 300 ꢁC
ꢀ1
lnðC =C Þ ¼ ꢀkt
ð3Þ
t
0
at a rate of 5 ꢁC min , and kept at 300 ꢁC for 20 min.
The temperature of the injector and GC–MS transfer line
was kept at 300 ꢁC. The MS system was run in the electron
impact ionization mode and the electron energy was 70 eV.
where C and C are the concentrations of the compounds
0
t
at time 0 and at time t, respectively, and k is the photolysis
rate constant. The rate constants (k) and half-lives (t1/2
)
estimated from Eq. (3) for degradation of the compounds
are given in Table 1. The degradability tended to be higher
for relatively high molecular weight (MW) PAHs and ClP-
AHs, which has also been predicted from the QSPR model
of photolysis of PAHs (Chen et al., 2001a,b). As far as the
2
.3. Molecular modeling and statistics
To develop a statistical model, we calculated 8 quantum
chemical and geometrical descriptors as follows: the vari-
Table 1
Photolytic kinetic rates and average quantum yields of ClPAHs and the parent PAHs
a
ꢀ1 b
)
No.
Compound
Abbreviation
MW
k (h
t1/2 (h)
/
ave
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
a
Phenanthrene
Phe
9-ClPhe
3,9-Cl
9,10-Cl
3,9,10-CPhe
Fluor
3-ClFluor
8-ClFluor
3,4-Cl
3,8-Cl
Py
1-ClPy
BaA
187
212
246
246
280
202
236
236
270
270
202
236
228
262
252
286
0.035
0.089
0.082
0.049
0.048
0.031
0.0044
0.018
0.0033
0.0035
0.203
0.315
0.221
0.104
1.67
19.7
7.8
8.5
14.2
14.4
22.4
158
38.1
210
198
3.4
2.2
3.1
6.7
0.4
2.9
1.1Eꢀ03
2.2Eꢀ03
2.1Eꢀ03
1.1Eꢀ03
9.0Eꢀ04
3.7Eꢀ04
6.0Eꢀ05
2.9Eꢀ04
5.1Eꢀ05
4.3Eꢀ05
5.7Eꢀ03
6.3Eꢀ03
5.4Eꢀ03
2.1Eꢀ03
3.5Eꢀ02
7.3Eꢀ03
9-Chlorophenanthrene
3,9-Dichlorophenanthrene
9,10-Dichlorophenanthrene
3,9,10-Trichlorophenanthrene
Fluoranthene
3-Chlorofluoranthene
8-Chlorofluoranthene
3,4-Dichlorofluoranthene
3,8-Dichlorofluoranthene
Pyrene
1-Chloropyrene
Benz[a]anthracene
7-Chlorobenz[a]anthracene
Benzo[a]pyrene
6-Chlorobenzo[a]pyrene
2
Phe
Phe
2
2
Fluor
Fluor
1
1
1
1
1
1
1
2
7-ClBaA
BaP
6-ClBaP
0.243
MW: molecular weight.
Data are averages of triplicate experiments.
b

T. Ohura et al. / Chemosphere 70 (2008) 2110–2117
2113
parent PAHs are concerned, the photostability was as fol-
O
O
a
lows; BaP < BaA 6 Py < Phe < Fluor. This trend is some-
what consistent with the reactivity of PAH toward
nitrating species (Nielsen, 1984; Minero et al., 2007) and
UV irradiation (Chen et al., 2001a,b). Among the ClPAHs,
the rate constant is largest in 1-ClPy, followed by 6-ClBaP,
O
Clx
Clx
Fluorenone (Flu) derivative
( :x=0, : x=1)
Benzocoumarin (BC) derivative
( :x=0, : x=1)
7
-ClBaA, and 9-ClPhe. Although the order of reactivity
0
0
0
0
0
.007
.006
.005
.004
.003
was somewhat consistent with that of a previous study car-
ried out in toluene (Ohura et al., 2005), the current rate
constants were approximately 2–20 times lower than those
in the previous study. This difference of reactivity could be
due to the hydrogen donor potential of the organic solvents
and the stability of the resulting solvent radicals (Dung and
O’Keefe, 1994; Ohura et al., 2004).
The rate of photolysis of Phe (Cl = 0) is lower compared
to the ClPhe isomers, among which 9-ClPhe was the less
photostable compound. Actually, photostability tended to
increase again for Cl > 1. Also 1-ClPy showed a photodeg-
radation rate 1.6 times higher compared to the parent
PAH. In contrast, among the ClFluor isomers, the photol-
ysis rates decreased with increasing chlorination extent.
Also 7-ClBaA and 6-ClBaP were more stable than the par-
ent PAHs. It suggests that certain polycyclic aromatics
would be stabilized upon chlorination. Indeed, such stabil-
ization of chlorine-substituted aromatics has been observed
in the case of chlorinated dioxins (Kim and O’Keefe, 2000).
Next, to invoke the contribution of different structural
features to explain their different behavior, the photolysis
quantum yields for the ClPAHs and PAHs were also eval-
uated. The average quantum yields (/ave) of each com-
pound upon irradiation in cyclohexane are given in Table
b
#
2
#3
BC
Cl-BC
Cl-Flu
0.002
0.001
0
Flu
0
0
0
0
0
0
.006 #4
#5
.005
.004
.003
.002
.001
Flu
Cl-Flu
Cl-BC
BC
0
0
1
2
3
4
0
1
2
3
4
Time (h)
Fig. 2. Photoproducts of chlorinated phenanthrenes produced during the
irradiation: (a) chemical structures of fluorenone (Flu) and benzocoum-
arin (BC) derivatives; (b) production of Flu and BC derivatives in the
experiment with 9-ClPhe (#2), 3,9-Cl
,9,10-Cl Phe (#5). Open and closed circles, open and closed triangles
2 2
Phe (#3), 9,10-Cl Phe (#4), and
3
3
show Flu and BC, and chlorinated Flu and BC, respectively.
time (Fig. 2b). Similarly, the photolysis of 3,9-Cl Phe (#3)
yielded the chlorinated derivative of BC (Cl–BC) as the
predominant product. On the other hand, 9,10-Cl Phe
#4) and 3,9,10-Cl Phe (#5) yielded Flu and the chlori-
2
2
1
. The order of values calculated were consistent with the
(
3
corresponding photolysis rate constants, as shown in the
nated derivatives (Cl–Flu) as the predominant products,
respectively. As a common future of ClPhes, those substi-
tuted by chlorine at 9 position resulted in production of
mainly BC derivatives. Conversely, when the 9 and 10 posi-
tions on the Phe are occupied by chlorine, the related
ClPhes produce mainly Flu derivatives. Among the photol-
ysis rates of ClPhes, those of 9-ClPhe (#2; k = 0.089, see
highest and lowest value for BaP and 3,8-Cl Fluor,
respectively.
2
3
.2. Photoproducts
The identification of the important photolytic reactions
of pollutants is necessary for the assessment of their hazard
for human health and the environment in general. The
phototoxicity of many PAHs, including the toxicity of pho-
tolytic products, have been observed in various in vitro and
in vivo assays (Arfsten et al., 1996; Yu, 2002; Yan et al.,
Table 1) and 3,9-Cl Phe (#3; 0.082) were approximately
2
2
-fold faster than 9,10-Cl Phe (#4; 0.049) and 3,9,10-
2
Cl Phe (#5; 0.048). Lee et al. (2001) reported that PAHs
3
are susceptible to oxidation at the position in which fron-
tier electron density is high. Therefore we suggest that oxi-
dation of ClPhes through photolysis occurs preferentially
at positions of the highest frontier electron density (9 and
2
004). Therefore, we identified the photolytic products of
ClPAHs using GC–MS analysis.
There was no detection of photoproducts from chlori-
nated derivatives of Fluor, Py, and BaP, suggesting the
rapid elimination of these photoproducts upon photolysis.
Here, the photoproducts of ClPhes were further investi-
gated. Upon ClPhes photolysis, either oxidized aromatics
10) compared to abstraction of chlorine from other
positions.
3.3. Prediction of photolysis of ClPAHs
(
fluorenone (Flu) and benzocoumarin (BC)) or their chlori-
nated derivatives (Cl–Flu and Cl–BC) were tentatively
identified as the major photoproducts (Fig. 2a). For 9-
ClPhe (compound #2), BC was produced preferentially in
comparison to Flu, levels of which were shown as relative
intensity to internal standard (Perylene), and built up over
To establish quantitative structure–property relation-
ships (QSPRs), a model based on the descriptors derived
from quantum computation has been used to predict the
direct photolysis of various organic compounds (Chen
et al., 1996, 2000, 2001a,b; Sabljic and Peijnenburg, 2001;

2114
T. Ohura et al. / Chemosphere 70 (2008) 2110–2117
Niu et al., 2006). Here, we have developed a QSPR model
to estimate the photolytic half-lives and average quantum
yields of ClPAHs and PAHs, and then used quantum
chemical descriptors with multiple linear regression analy-
sis (MRA) to identify the dominant effects on the
photodegradation.
The values of the selected molecular descriptors of each
ClPAH and corresponding parent PAH used are summa-
rized in Table 2. To estimate logt1/2 values from those
log /ave ¼ 119ELUMO ꢀ 71:9ðELUMO ꢀ EHOMO
0:00109TE þ 14:0
Þ
ꢀ
ð5Þ
The descriptor ELUMO ꢀ E
is one of the significant
factors affecting log/ave of the ClPAHs and PAHs, and an
HOMO
increase of those values could lead to decrease of log/ave
.
ꢀ
Chen et al. (2000) also reported that large ELUMO
EHOMO is one of the significant descriptors to control the
photolysis quantum yields of PAHs. In addition, Pearson
descriptors using MRA, ELUMO
,
ELUMO ꢀ EHOMO,
(1986) reported that absolute hardness can be expressed
ELUMO+1, total energy (TE), and surface area (SA) were
selected (Table 3), and the analytical QSPR equation for
logt1/2 can be obtained and defined as follows:
as ꢀ(E
ꢀ EHOMO)/2, which is regarded as a measure
LUMO
of energy stabilization in a chemical system, indicating that
chemical structures of molecules tend to be more stable at
log t1=2 ¼ 125ELUMO þ 24:6ðELUMO ꢀ EHOMOÞ þ 312ELUMOþ1
higher values of ELUMO ꢀ E
(Niu et al., 2007).
HOMO
ꢀ
0:00457TE þ 0:00557SA ꢀ 6:90
ð4Þ
The descriptors ELUMO+1, TE and SA were inter-corre-
lated with each other (Table 4), suggesting that these param-
eters could play an important role in predicting logt1/2 as
shown in Eq. (4). On the other hand, there was little correla-
Furthermore, average quantum yields (/ave) of ClPAHs
and PAHs were also predicted from those descriptors using
MRA. ELUMO, ELUMO ꢀ EHOMO, and total energy (TE)
were selected as the significant descriptors (Table 3), and
the QSPR equation for log/ave can be defined as follows:
tion between the three parameters and ELUMO ꢀ E
HOMO
(also ELUMO). However, to construct an effective model
for predicting log/ave, ELUMO ꢀ EHOMO and ELUMO should
Table 2
The quantum chemical descriptors for ClPAHs and PAHs used in this study
a
No.
Descriptors
HOMO (eV)
E
E
LUMO
E
LUMO + EHOMO
E
LUMO ꢀ EHOMO
E
HOMOꢀ1
E
LUMO+1
Total energy
(hartree)
Surface area
( )
2
(eV)
(eV)
(eV)
(eV)
(eV)
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
a
ꢀ0.212
ꢀ0.218
ꢀ0.221
ꢀ0.223
ꢀ0.226
ꢀ0.213
ꢀ0.218
ꢀ0.218
ꢀ0.219
ꢀ0.222
ꢀ0.197
ꢀ0.202
ꢀ0.197
ꢀ0.202
ꢀ0.189
ꢀ0.194
ꢀ0.0375
ꢀ0.0477
ꢀ0.0549
ꢀ0.0555
ꢀ0.0625
ꢀ0.0657
ꢀ0.0739
ꢀ0.0724
ꢀ0.0808
ꢀ0.0803
ꢀ0.0555
ꢀ0.0633
ꢀ0.0580
ꢀ0.0671
ꢀ0.0650
ꢀ0.0733
ꢀ0.249
ꢀ0.265
ꢀ0.276
ꢀ0.278
ꢀ0.289
ꢀ0.279
ꢀ0.292
ꢀ0.291
ꢀ0.300
ꢀ0.302
ꢀ0.253
ꢀ0.265
ꢀ0.255
ꢀ0.269
ꢀ0.254
ꢀ0.267
0.174
0.170
0.167
0.167
0.164
0.148
0.144
0.146
0.139
0.141
0.141
0.138
0.139
0.135
0.124
0.120
ꢀ0.223
ꢀ0.230
ꢀ0.239
ꢀ0.234
ꢀ0.243
ꢀ0.219
ꢀ0.223
ꢀ0.224
ꢀ0.226
ꢀ0.230
ꢀ0.229
ꢀ0.237
ꢀ0.220
ꢀ0.226
ꢀ0.224
ꢀ0.230
ꢀ0.0312
ꢀ0.0387
ꢀ0.0477
ꢀ0.0440
ꢀ0.0527
ꢀ0.0187
ꢀ0.0259
ꢀ0.0278
ꢀ0.0332
ꢀ0.0341
ꢀ0.0238
ꢀ0.0321
ꢀ0.0339
ꢀ0.0403
ꢀ0.0322
ꢀ0.0381
ꢀ540
ꢀ999
280
305
334
326
354
289
316
323
339
349
276
306
336
357
329
353
ꢀ1460
ꢀ1460
ꢀ1920
ꢀ616
ꢀ1080
ꢀ1080
ꢀ1540
ꢀ1540
ꢀ616
1
1
1
1
1
1
1
ꢀ1080
ꢀ693
ꢀ1150
ꢀ769
ꢀ1230
The numbers correspond to those in Table 1.
Table 3
Model fitting results
Variable
logt1/2 (Pred.)
log/ave (Pred.)
c
c
Coefficient
SE
P
Coefficient
SE
P
E
E
E
LUMO
125
24.6
312
ꢀ0.00457
0.0557
ꢀ6.90
82.4
17.7
112
0.0020
0.020
4.01
1.6Eꢀ01
1.9Eꢀ01
2.0Eꢀ02
4.9Eꢀ02
2.1Eꢀ02
1.2Eꢀ01
119
ꢀ71.9
14.0
9.13
2.0Eꢀ06
4.4Eꢀ06
LUMO ꢀ EHOMO
LUMO+1
a
TE
SA
Constant
ꢀ0.00109
0.00030
1.91
3.8Eꢀ03
9.4Eꢀ06
b
14.0
a
TE: total energy.
b
c
SA: surface area.
SE: standard error.

T. Ohura et al. / Chemosphere 70 (2008) 2110–2117
2115
Table 4
Correlation coefficients between certain quantum chemical descriptors
a
b
E
LUMO
ELUMO ꢀ EHOMO
E
LUMO+1
TE
SA
E
E
E
LUMO
1.000
0.679
LUMO ꢀ EHOMO
1.000
LUMO+1
a
ꢀ0.167
ꢀ0.354
ꢀ0.142
ꢀ0.329
1.000
0.734
ꢀ0.643
TE
SA
a
0.426
1.000
b
ꢀ0.556
ꢀ0.724
1.000
TE: total energy.
SA: surface area.
b
be further added to the prediction parameters as shown in
Eq. (5). Therefore, t1/2 is mainly influenced by the geometri-
cal characteristics, such as molecular surface area, of the
reaction molecules, and /ave may be influenced by the geo-
metrical characteristics as well as the energies of molecular
orbitals of the molecules.
(Table 5). The relationship for logt1/2 values showed signif-
icant correlation (R = 0.951, Fig. 3a). Similarity, the pre-
dicted quantum yields were also closed to the
experimental ones (R = 0.939, Fig. 3b). It can be concluded
that those models can be used to make predictions for
other ClPAH congeners, and the prediction may give an
initial estimation of photodegradability of the atmospheric
ClPAH congeners.
In addition, the predicted logt1/2 and log/ave values
were compared to the corresponding observed values
Table 5
Comparison between experimental and predicted logt1/2 or log/ave values of the ClPAHs and PAHs
a
b
b
No.
logt1/2 (Obs.)
logt1/2 (Pred.)
Diff.
log/ave (Obs.)
log/ave (Pred.)
Diff.
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
a
1.30
0.893
0.927
1.15
1.16
1.35
2.20
1.58
2.32
2.30
0.534
0.342
0.496
0.823
ꢀ0.381
0.456
1.06
0.838
0.747
1.37
1.39
1.63
1.83
1.88
1.96
2.41
0.433
0.544
0.576
0.645
ꢀ0.148
0.288
0.24
0.06
0.18
ꢀ0.22
ꢀ0.23
ꢀ0.28
0.37
ꢀ2.97
ꢀ2.65
ꢀ2.68
ꢀ2.98
ꢀ3.05
ꢀ3.43
ꢀ4.22
ꢀ3.53
ꢀ4.29
ꢀ4.37
ꢀ2.24
ꢀ2.20
ꢀ2.27
ꢀ2.69
ꢀ1.46
ꢀ2.14
ꢀ2.43
ꢀ2.83
ꢀ2.95
ꢀ3.06
ꢀ3.16
ꢀ3.79
ꢀ3.99
ꢀ3.98
ꢀ3.94
ꢀ4.07
ꢀ2.14
ꢀ2.34
ꢀ2.15
ꢀ2.44
ꢀ1.83
ꢀ2.07
ꢀ0.54
0.18
0.27
0.09
0.12
0.36
ꢀ0.23
ꢀ0.30
0.44
0.37
ꢀ0.35
ꢀ0.30
ꢀ0.11
0.14
ꢀ0.12
ꢀ0.24
0.37
1
1
1
1
1
1
1
ꢀ0.11
0.10
ꢀ0.20
ꢀ0.08
0.18
ꢀ0.23
0.17
ꢀ0.07
The numbers correspond to those in Table 1.
Difference between the observed and predicted values.
b
2
1
0
.5
2
a
b
1
0
-
-
-
-
1.5
2.0
2.5
3.0
9
15
8
7
13 16
11
6
.5
1
5
12
1
14
2
4
1
3
2
14
4
3
5
1
2 13
.5
0
1
6
1
-3.5
-4.0
-4.5
1
9
7
10
6
R = 0.951
P = 8.33E-05
8
R = 0.939
P = 7.45E-06
1
5
-
0.5
-4.5
-4.0
-3.5
-3.0
-2.5
-2.0
-1.5
-
0.5
0
0.5
1
1.5
2
2.5
logt1/2 (Obs.)
logφave (Obs.)
Fig. 3. Plots of observed versus predicted logt1/2 values (a) and log/ave (b) for ClPAHs and the parent PAHs (the numbers correspond to those in
Table 1).

2
116
T. Ohura et al. / Chemosphere 70 (2008) 2110–2117
Frisch, M.J., Trucks, G.W., Schlegel, H.B., Scuseria, G.E., Robb, M.A.,
4
. Conclusion
Cheeseman, J.R., Zakrzewski, V.G., Montgomery Jr., J.A., Strat-
mann, R.E., Burant, J.C., Dapprich, S., Millam, J.M., Daniels, A.D.,
For ClPAHs, the photodegradability varied greatly
¨
Kudin, K.N., Strain, M.C., Farkas, O., Tomasi, J., Barone, V., Cossi,
according to the skeleton of PAHs. As far as relatively
high-molecular weight PAHs are concerned, such as Fluor,
BaA, and BaP, the chloroderivatives were more stable
toward photolysis. The photoproducts analysis of ClPhes
suggested that oxidation occurs at positions of the highest
frontier electron density. Furthermore, the QSPR models
for direct photolysis half-lives and average quantum yields
of 11 ClPAHs and 5 parent PAHs in cyclohexane solution
under irradiation by a high-pressure mercury lamp was
developed, and shown to have good predictive abilities.
The significant factors affecting photolytic half-lives of
the compounds were ELUMO+1, total energy, and surface
area values, while ELUMO, ELUMO ꢀ EHOMO, and total
energy control the photolysis quantum yields.
M., Cammi, R., Mennucci, B., Pomelli, C., Adamo, C., Clifford, S.,
Ochterski, J., Petersson, G.A., Ayala, P.Y., Cui, Q., Morokuma, K.,
Salvador, P., Dannenberg, J.J., Malick, D.K., Rabuck, A.D., Ragh-
avachari, K., Foresman, J.B., Cioslowski, J., Ortiz, J.V., Baboul, A.G.,
Stefanov, B.B., Liu, G., Liashenko, A., Piskorz, P., Kom a´ romi, I.,
Gomperts, R., Martin, R.L., Fox, D.J., Keith, T., Al-Laham, M.A.,
Peng, C.Y., Nanayakkara, A., Challacombe, M., Gill, P.M.W.,
Johnson, B., Chen, W., Wong, M.W., Andres, J.L., Gonzalez, C.,
Head-Gordon, M., Replogle, E.S., Pople, J.A., 1998. Gaussian 98.
Gaussian, Inc., Pittsburgh, PA.
Jang, M., McDow, S.R., 1995. Benz[a]anthracene photodegradation in the
presence of known organic constituents of atmospheric aerosols.
Environ. Sci. Technol. 29, 2654–2660.
Jang, M., McDow, S.R., 1997. Products of benz[a]anthracene photodeg-
radation in the presence of known organic constituents of atmospheric
aerosols. Environ. Sci. Technol. 31, 1046–1053.
Kamens, R.M., Guo, Z., Fulcher, J.N., Bell, D.A., 1988. Influrnce of
humidity, sunlight, and temperature on the daytime decay of polyar-
omatic hydrocarbons on atmospheric soot particles. Environ. Sci.
Technol. 22, 103–108.
Acknowledgements
This work was supported, in part, by a Grant-in-aid for
Young Scientists (B) from the Japan Society for the Pro-
motion of Science (Project No. 14780416, to T.O).
Kim, M., O’Keefe, P.W., 2000. Photodegradation of polychlorinated
dibenzo-p-dioxins and debenzofurans in aqueous solutions and in
organic solvents. Chemosphere 41, 793–800.
Kitazawa, A., Amagai, T., Ohura, T., 2006. Temporal trends and
relationships of particulate chlorinated polycyclic aromatic hydrocar-
bons and their parent compounds in urban air. Environ. Sci. Technol.
References
4
0, 4592–4598.
Arfsten, D.P., Schaeffer, D.J., Mulveny, D.C., 1996. The effects of near
ultraviolet radiation on the toxic effects of polycyclic aromatic
hydrocarbons in animals and plants: A review. Ecotoxicol. Environ.
Safe. 33, 1–24.
Baek, S.O., Field, R.A., Goldstone, M.E., Kirk, P.W., Lester, J.N., 1991.
A review of atmospheric polycyclic aromatic hydrocarbons: Sources,
fate and behavior. Water Air Soil Pollut. 60, 279–300.
Bidleman, T.F., 1988. Atmospheric processes, wet and dry deposition of
organic compounds are controlled by their vapor-particle partitioning.
Environ. Sci. Technol. 22, 361–367.
Chen, J.W., Kong, L.R., Zhu, C.M., Huang, Q.G., Wang, L.S., 1996.
Correlation between photolysis rate constants of polycyclic aromatic
hydrocarbons and frontier molecular orbital energy. Chemosphere 33,
Lamotte, M., Pereyre, J., Joussot-Dubien, J., Lapouyade, R., 1987. The
photolysis of pyrene and perylene in cyclohexane liquid solution from
highly excited electronic states. J. Photochem. 38, 177–188.
Lee, B.-D., Iso, M., Hosomi, M., 2001. Prediction of Fenton oxidation
positions in polycyclic aromatic hydrocarbons by frontier electron
density. Chemosphere 42, 431–435.
L o¨ nnermark, A., Blomqvist, P., 2006. Emissions from an automobile fire.
Chemosphere 62, 1043–1056.
Mantis, J., Chaloulakou, A., Samara, C., 2005. PM10-bound polycyclic
aromatic hydrocarbons (PAHs) in the Greater Area of Athens, Greece.
Chemosphere 59, 593–604.
McDow, S.R., Sun, O.-R., Vartiainen, M., Hong, Y.-S., Yao, Y.-L.,
Fister, T., Yao, R.-Q., Kamens, R.M., 1994. Effect of composition and
state of organic components on polycyclic aromatic hydrocarbon
decay in atmospheric aerosols. Environ. Sci. Technol. 28, 2147–2153.
Menzie, C.A., Potocki, B.B., Santodonato, J., 1992. Exposure to
carcinogenic PAHs in the environment. Environ. Sci. Technol. 26,
1278–1284.
1143–1150.
Chen, J., Peijnenburg, W.J.G.M., Quan, X., Yang, F., 2000. Quantitative
structure-property relationships for direct photolysis quantum yields
of selected polycyclic aromatic hydrocarbons. Sci. Total Environ. 246,
11–20.
Chen, J., Quan, X., Schramm, K.-W., Kettrup, A., Yang, F., 2001a.
Quantitative structure-property relationships (QSPRs) on direct pho-
tolysis of PCDDs. Chemosphere 45, 151–159.
Minero, C., Bono, F., Rubertelli, F., Pavino, D., Maurino, V., Pelizzetti,
E., Vione, D., 2007. On the effect of pH in aromatic photonitration
upon nitrate photolysis. Chemosphere 66, 650–656.
Chen, J., Quan, X., Yan, Y., Yang, F., Peijnenburg, W.J.G.M., 2001b.
Quantitative structure-property relationship studies on direct photol-
ysis of selected polycyclic aromatic hydrocarbons in atmospheric
aerosol. Chemosphere 42, 263–270.
Dewhurst, F., Kitchen, D.A., 1972. Synthesis and properties of 6-substi-
tuted benzo[a]pyrene derivatives. J. Chem. Soc. Perk. T 2, 710–712.
Dung, M.H., O’Keefe, P.W., 1994. Comparative rates of photolysis of
polychlorinated dibenzofurans in organic solvents and in aqueous
solutions. Environ. Sci. Technol. 28, 549–554.
Finlayson-Pitts, B.J., Pitts Jr., J.N., 1997. Tropospheric air pollution:
Ozone, airborne toxics, polycyclic aromatic hydrocarbons, and parti-
cles. Nature 276, 1045–1052.
Finlayson-Pitts, B.J., Pitts Jr., J.N., 2000. Chemistry of the Upper and
Lower Atmosphere. Academic Press, San Diego.
Nielsen, T., 1984. Reactivity of polycyclic aromatic hydrocarbons toward
nitrating species. Environ. Sci. Technol. 18, 157–163.
Niu, J.F., Shen, Z.Y., Yang, Z.F., Long, X.X., Yu, G., 2006. Quantitative
structure-property relationships on photodegradation of polybromi-
nated diphenyl ethers. Chemosphere 64, 658–665.
Niu, J., Wang, L., Yang, Z., 2007. QSPRs on photodegradation half-lives
of atmospheric chlorinated polycyclic aromatic hydrocarbons associ-
ated with particulates. Ecotoxicol. Environ. Safe. 66, 272–277.
Ohura, T., 2007. Environmantal behavior, sources, and effects of
chlorinated polycyclic aromatic hydrocarbons. The Scientific World
J. 7, 372–380.
Ohura, T., Kitazawa, A., Amagai, T., 2004. Seasonal variability of 1-
chloropyreneon a atmospheric particles and photostability in toluene.
Chemosphere 57, 831–837.
Frisch, M.J., Nielsen, A.B., Holder, A.J., 2000. Gauss ViewW User’s
Reference. Gaussian, Inc., Pittsburgh, PA.
Ohura, T., Kitazawa, A., Amagai, T., Makino, M., 2005. Occurrence,
profiles, and photostabilities of chlorinated polycyclic aromatic

T. Ohura et al. / Chemosphere 70 (2008) 2110–2117
2117
hydrocarbons associated with particulates in urban air. Environ. Sci.
Technol. 39, 85–91.
Pearson, P.G., 1986. Absolute electronegativity and hardness correlated
with molecular orbital theory. Proc. Natl. Acad. Sci. 83, 8440–8441.
Pekey, B., Karakaßs, D., Ayberk, S., 2007. Atmospheric deposition of
polycyclic aromatic hydrocarbons to Izmit Bay, Turkey. Chemosphere
Sabljic, A., Peijnenburg, W., 2001. Recommendations on modeling
lifetime and degradability of organic compounds in air, soil and water
systems. Pure Appl. Chem. 73, 1331–1348.
Yan, J., Wang, L., Fu, P.P., Yu, H., 2004. Photomutagenicity of 16
polycyclic aromatic hydrocarbons from the US EPA priority pollutant
list. Mutat. Res. 557, 99–108.
6
7, 537–547.
Yu, H., 2002. Environmental carcinogenic polycyclic aromatic hydrocar-
bons: Photochemistry and phototoxicity. J. Environ. Sci. Heal. C 20,
149–183.
Sabljic, A., 2001. QSAR models for estimating properties of persistent
organic pollutants required in evaluation of their environmental fate
and risk. Chemosphere 43, 363–375.