Ambient measurement of nitrotriphenylenes and possibility of nitrotriphenylene formation by atmospheric reaction

Article abstract of DOI:10.1021/es990553d

Environmental nitration of PAHs by nitrogen oxides on airborne particulates and identification and quantification of nitro-PAHs in them have been studied on fluoranthene and pyrene because of their large abundance. However, the direct mutagenic activity of these nitro-PAHs can account for only up to about 20% of the total direct mutagenic activity of airborne particulate samples, and the presence of unknown mutagenic nitro-PAHs was suggested. 1- and 2-Nitrotriphenylenes were first identified and quantified in ambient particulate samples collected in Tokyo. Their mutagenicity and formation by laboratory reactions of triphenylene deposited on a filter with NO₂/NO₃/N₂O₅ were also investigated. The concentrations of 1- and 2- nitrotriphenylenes were much higher than that of 1-nitropyrene in the same samples. 2-Nitrotriphenylene showed over 2 orders of magnitude greater mutagenic activity than 1-nitrotriphenylene in both strains TA98 and YG1024 (18 and 480 revertants/nmol for 1-nitrotriphenylene and 3800 and 101 700 revertants/nmol for 2-nitrotriphenylene in TA98 and YG1024). The mutagenic 2- nitrotriphenylene was expected to contribute largely to the total mutagenic activity of the extracts from airborne particulates. In addition, more effective formation of nitrotriphenylenes by the nighttime reaction of triphenylene in the presence of O₃ in the dark can explain the observation of high nitrotriphenylene concentration in the nighttime samples. Environmental nitration of PAHs by nitrogen oxides on airborne particulates and identification and quantification of nitro-PAHs in them have been studied on fluoranthene and pyrene because of their large abundance. However, the direct mutagenic activity of these nitro-PAHs can account for only up to about 20% of the total direct mutagenic activity of airborne particulate samples, and the presence of unknown mutagenic nitro-PAHs was suggested. 1-and 2-Nitrotriphenylenes were first identified and quantified in ambient particulate samples collected in Tokyo. Their mutagenicity and formation by laboratory reactions of triphenylene deposited on a filter with NO₂/NO₃/N₂O₅ were also investigated. The concentrations of 1- and 2-nitrotriphenylenes were much higher than that of 1-nitropyrene in the same samples. 2-Nitrotriphenylene showed over 2 orders of magnitude greater mutagenic activity than 1-nitrotriphenylene in both strains TA98 and YG1024 (18 and 480 revertants/nmol for 1-nitrotriphenylene and 3800 and 101700 revertants/nmol for 2-nitrotriphenylene in TA98 and YG1024). The mutagenic 2-nitrotriphenylene was expected to contribute largely to the total mutagenic activity of the extracts from airborne particulates. In addition, more effective formation of nitrotriphenylenes by the nighttime reaction of triphenylene in the presence of O₃ in the dark can explain the observation of high nitrotriphenylene concentration in the nighttime samples.

Full text of DOI:10.1021/es990553d

Environ. Sci. Technol. 2000, 34, 1893-1899
CHART 1
Ambient Measurement of
Nitrotriphenylenes and Possibility of
Nitrotriphenylene Formation by
Atmospheric Reaction
S A T O K O I S H I I , ‡
YO S H I H A R U H I S A M A T S U , * , † K O J I I N A Z U , ‡
M O R I O K A D O I , § A N D K E N - I C H I A I K A ‡
soot from burning wood (3, 4). Nitro-PAH formation by the
reaction of parent PAH with nitrogenous species has been
discussed on gas-phase reaction (5, 6) and gas-solid-phase
reaction (7-11).
Department of Environmental Chemistry and Engineering,
Interdisciplinary Graduate School of Science and Engineering,
Tokyo Institute of Technology, 4259 Nagatsuta, Midori-ku,
Yokohama 226-8502, Japan, Department of Community
Environmental Science, National Institute of Public Health,
4-6-1 Shirokanedai, Minato-ku, Tokyo 108-8638, Japan, and
College of Science and Technology, Nihon University, 1-8-4,
Kanda Surugadai, Chiyoda-ku, Tokyo 100-0062, Japan
Environmental nitration of PAHs by nitrogen oxides on
airborne particulates has been studied particularly on
fluoranthene and pyrene because of their large abundance
in ambient particulate matter; nitrofluoranthenes and ni-
tropyrenes have been identified and quantified in airborne
particulate samples (12-14), and their mutagenic activities
were determined using Salmonella typhimurium strains in
the absence of S9 mix (1, 14, 15). However, the direct
mutagenic activity of these nitro-PAHs can account for only
up to about 20% of the total direct mutagenic activity of
airborne particulate samples (12), and the presence of
unknown mutagenic nitro-PAHs was suggested. Recently,
nitro-PAH lactones were reported to contribute to the
ambient mutagenicity (16, 17). 3-Nitrobenzanthrone was
further shown to be a new class of powerful direct-acting
mutagens of atmospheric origin, and it existed not only in
diesel exhaust particles but also in airborne particulates,
suggesting formation by the reaction of benzanthrone with
NO2 and O3 (18).
In this study, we investigated the ambient amount of
nitrotriphenylenes and their formation by the reaction of
triphenylene (1) with NO2 to elucidate the contribution of
nitrotriphenylenes to be a direct active mutagen of airborne
particulate samples (Chart 1). Though triphenylene is one of
the 4-ring PAHs which are much abundant in airborne
particulates, few studies have been carried out because of its
low mutagenic activity. However, according to the reported
high mutagenicity of 2-nitrotriphenylene (2) (19) and the
enhanced nitration of triphenylene by acid catalysts (20), it
is expected that nitrotriphenylene, especially 2-nitrotri-
phenylene, will contribute to the observed mutagenicity of
airborne particulate samples and can be formed by atmo-
spheric reaction. Thus we measured the amount of nitro-
triphenylenes in airborne particulate samples collected both
at daytime and at nighttime and investigated nitrotriphen-
ylene formation in the nitration of triphenylene in the
presence of O3 or under light irradiation.
Environmental nitration of PAHs by nitrogen oxides on
airborne particulates and identification and quantification
of nitro-PAHs in them have been studied on fluoranthene
and pyrene because of their large abundance. However, the
direct mutagenic activity of these nitro-PAHs can account
for only up to about 20% of the total direct mutagenic
activity of airborne particulate samples, and the presence
of unknown mutagenic nitro-PAHs was suggested. 1-
and 2-Nitrotriphenylenes were first identified and quantified
in ambient particulate samples collected in Tokyo. Their
mutagenicity and formation by laboratory reactions of
triphenylene deposited on a filter with NO / NO /N2O were
2
3
5
also investigated. The concentrations of 1- and 2-nitrotri-
phenylenes were much higher than that of 1-nitropyrene
in the same samples. 2-Nitrotriphenylene showed over 2
orders of magnitude greater mutagenic activity than
1-nitrotriphenylene in both strains TA98 and YG1024 (18
and 480 revertants/nmol for 1-nitrotriphenylene and 3800 and
101 700 revertants/nmol for 2-nitrotriphenylene in TA98
and YG1024). The mutagenic 2-nitrotriphenylene was expected
to contribute largely to the total mutagenic activity of
the extracts from airborne particulates. In addition, more
effective formation of nitrotriphenylenes by the nighttime
reaction of triphenylene in the presence of O in the dark
3
can explain the observation of high nitrotriphenylene
concentration in the nighttime samples.
Experimental Section
Introduction
Airborne Particulates Sam pling. Airborne particulate matter
was collected on a quartz fiber filter (Pallflex Product Co.,
2500QATUP) using two high-volume air samplers with an
impactor to remove particles larger than 10 µm in diameter
(70 m3/ h) on the rooftop of a five-story building facing a
traffic-heavy thoroughfare in central Tokyo (Minato-ku)
during December 9-11, 1998. Daytime sample-collection
was made for 9 h from 8 a.m. to 5 p.m., and the nighttime
one was similarly done from 8 p.m. to 5 a.m. The samples
for PAH analysis were extracted from collected airborne
particulate matter with acetonitrile by sonication for 10 min,
and the extract solution was concentrated to an adequate
Some species of nitrated polycyclic aromatic hydrocarbons
(nitro-PAHs) which are ubiquitous in our atmospheric
environment are known to exhibit high mutagenicity in in
vitro test (1) and high carcinogenicity in animal experiments
(2). Nitro-PAH has been attributed largely to direct emissions
from combustion sources such as diesel engine smoke and
* Corresponding author phone: +81-3-3441-7111; fax: +81-3-
3446-4314; e-mail: hisa@iph.go.jp.
† National Institute of Public Health.
‡ Tokyo Institute of Technology.
§
Nihon University.
9
10.1021/es990553d CCC: $19.00
Published on Web 04/15/2000
2000 Am erican Chem ical Society
VOL. 34, NO. 10, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 8 9 3

volume for the analysis. For nitro-PAH analysis, the sample
was extracted with dichloromethane, and the extract solution
was prefractionated by preparative HPLC using the solvent
gradient (n-hexane for 5 min, linear gradient to 5% dichlo-
romethane/ n-hexane for 5 min, linear gradient to dichlo-
romethane for 15 min, and hold for 5 min, and linear gradient
to acetonitrile for 5 min, and finally hold for 10 min). Fractions
eluted from 15 to 27 min were picked up and concentrated
to an adequate volume for the analysis.
Materials. Triphenylene was obtained from Tokyo Kasei
Kogyo Co. Ltd. All solvents used were HPLC-grades from
Kokusan Chemical Co. Ltd. Permical permeation tube, which
was used to evolve NO2, was purchased from Gastek Co. Ltd.
Other chemicals were obtained from Aldrich Co.
Analytical Instrum entation. Fractionation of the samples
were performed with a preparative high-performance liquid
chromatograph (HPLC, Senshu, Model SSC-3100 and a Model
SSC-3000A-II UV detector (λ ) 254 nm) using a column of
Nucleosil 100-1525 (10 mm i.d. × 250 mm)). Gas chro-
matographic/ mass spectrometric (GC/ MS) analysis was
performed using a mass spectrometer (Hewlett-Packard, HP-
5890A series II connected to JEOL JMS-HX110A); ionization
voltage 70 eV; chamber temperature 280-290 °C; accelerating
voltage 10 kV. Nitrotriphenylenes in the airborne particulate
samples were identified by GC/ MS with selected ion moni-
toring (SIM), and synthesized samples were characterized
by operating in the scanning mode (45-500 u). The GC
column used was a DB-5 column (5% phenylmethylsilicone,
0.25 mm i.d. × 30 m, J&W Scientific, Inc.). GC identification
and quantification of nitrotriphenylenes and 1-nitropyrene
was performed with a gas chromatograph equipped with a
nitrogen-phosphorus detector (NPD, Hewlett-Packard, HP-
5890A series II) using a HP-5 column (5% phenylmethyl-
silicone, 0.25 mm i.d. × 30 m, Hewlett-Packard). The
temperature program for GC/ MS and GC/ NPD analysis was
as follows: 70 °C for 1 min, to 300 °C with ramping rate of
7 °C/ min, isothermal for 10 min. HPLC analysis of triphen-
ylene, pyrene, and benzo[e]pyrene was performed with a
Shimadzu LC-9A with a RF-550 spectrofluorometric detector
using a Vydac 201TP54 (4.6 mm i.d. × 250 mm), and the
mobile phase was acetonitrile/ water (8/ 2, v/ v) at 30 °C.
Synthesis of Nitrotriphenylenes. 1-Nitrotriphenylene (3)
and 2-nitrotriphenylene (2) were synthesized according to
the literature (20-22). To an acetic anhydride (Ac2O) solution
of triphenylene ((1), 662 mg, 2.5 mmol in Ac2O of 2.5 mL),
a solution of HNO3 (d ) 1.38, 0.18 mL) in Ac2O (0.84 mL)
added at 0 °C with continuous stirring for 30 min. After stirring
for another 30 min, the mixture was poured onto ice/
dichloromethane, and the organic layer was washed with
water. When most of the solvent was evaporated, 1 g of silica
gel was added, followed by complete evaporation of the
solvent. The crude sample was placed on top of a column
packed with silica gel and eluted with 10% dichloromethane/
n-hexane. Purification was subsequently carried out by
preparative HPLC using 10% dichloromethane/ n-hexane. The
purity of both isomers was above 99.9% on GC/ NPD.
1-Nitrotriphenylene: yellow powder; EI-MS: m/ z (%) ) 273
(M+, 22), 243 (M+ - NO, 100), 226 (M+ - HNO2, 66), 215 (M+
- CNO2, 86); HR-EI-MS C18H11-NO2 calcd 273.0790, found
273.0767. 2-Nitrotriphenylene (2): yellow powder; EI-MS:
m/ z (%) ) 273 (M+, 100), 243 (M+ - NO, 4), 226 (M+ - HNO2,
86), 215 (M+ - CNO2, 33); HR-EI-MS C18H11-NO2 calcd
273.0797, found 273.0797.
FIGURE 1. Gas chromatogram with NPD of the 12/9 nighttime
samples. The identity of the peaks are as follows: 1-NP, 1-nitro-
pyrene; 1-NTP, 1-nitrotriphenylene; and 2-NTP, 2-nitrotriphenylene.
acetyltransferase activity 3-100 times as high as the original
TA98. It was kindly provided by Dr. T. Nohmi, National
Institute of Hygienic Sciences, Tokyo. The samples for the
assay were concentrated by a rotary evaporator and dried
under a stream of nitrogen. They were then dissolved in
dimethyl sulfoxide and subjected to the mutagenicity test.
Assays for synthesized nitrotriphenylenes were performed
five times. For each mixture after the reaction of triphenylene
with nitrogen dioxide, assay was carried out three times.
Laboratory Nitration. One hundred micrograms of
triphenylene dissolved in acetone/ ether (1/ 1, v/ v) was
uniformly deposited on a 40 cm2 piece of Teflon-coated glass
fiber filter (Pallflex Product Co., TX40H120WW) and was air-
dried. The apparatus for the exposure of triphenylene to NO2
was the same as that previously described (9, 11, 18). The
light source for photochemical reaction was a high-pressure
mercury lamp (100 W) with cooling water and a Pyrex filter
to cut off ultraviolet light shorter than 290 nm in wavelength.
After the filter was installed in the reaction vessel, the dry air
containing 10 ppm of NO2 was supplied to the vessel at a
flow rate of 0.7 L/ min. The sample was then irradiated by the
ultraviolet light. In the case of the dark reaction in the
presence of O3, an artificial air containing 10 ppm of NO2
and 5 ppm of O3 was employed. The reaction products were
dissolved in acetonitrile with sonication for 10 min. Tri-
phenylene and nitrotriphenylenes were analyzed by HPLC
with spectrofluorometric detection and GC/ NPD (or GC/
MS), respectively. The reaction products were fractionated
by preparative HPLC to identify the mutagens in them, and
the effluents from the column were fractionated at intervals
of 1 min.
Results and Discussion
Identification of Triphenylene and Nitrotriphenylenes in
Airborne Particulates. Triphenylene was determined in
acetonitrile extracts of the ambient daytime and nighttime
samples collected at Minato-ku, Tokyo. Triphenylene and
chrysene of molecular weight 228 were eluted at 14.7 and
15.7 min after injection, respectively. Triphenylene was
successfully separated and quantitatively analyzed using a
column of Vydac 201TP54 (4.6 mm i.d. × 250 mm).
1- and 2-Nitrotriphenylenes in the purified fractions from
airborne particulate samples were analyzed by GC/ NPD. A
gas chromatogram with NPD of the 12/ 9 nighttime sample
is shown in Figure 1. The identification of 1- and 2-nitro-
triphenylenes was based on retention time matching with
authentic 1- and 2-nitrotriphenylenes. To show that the
identification of 1- and 2-nitrotriphenylenes in the ambient
samples by our analytical system was little affected by the
coexisting isomers such as nitrobenz[a]anthracenes and
Mutagenicity Assay. The mutagenicity assay was per-
formed following the method of Ames et al. (23) (Salmonella
typhimurium strain, -S9 mix) with the slight modification
of including a preincubation step (24). The bacterial strains
used were Salmonella typhimurium His- strains TA98 and
YG1024 which is an O-acetyltransferase overproducing
derivatives of TA98 (25). YG1024 exhibits the enhanced
9
1 8 9 4 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 10, 2000

FIGURE 2. Gas chromatogram with NPD showing resolution of 1-
and 2-nitrotriphenylenes (1- and 2-NTPs), 7-nitrobenz[a]anthracene
(7-NBaA), and 6-nitrochrysene (6-NC). 1-Nitropyrene and the isomers
in order of elution are as follows: 1-nitropyrene, 7-nitrobenz[a]-
anthracene, 1-nitrotriphenylene, 6-nitrochrysene, and 2-nitrotri-
phenylene.
FIGURE 4. Mass chromatograms of ambient composite sample
collected in Minato-ku, Tokyo during December 9-11, 1998.
magnitude larger than those of GC/ NPD (0.1 and 0.1 ng,
respectively). The composite sample made by mixing the six
samples analyzed by GC/ NPD was then used. As shown in
Figure 4, 1- and 2-nitrotriphenylenes in the composite sample
were confirmed on the basis of retention time and the
molecular ion, m/ z 273 ([M+]), together with the characteristic
fragment ions, m/ z 243 ([M - NO]+), of authentic 1- and
2-nitrotriphenylenes. 7-Nitrobenz[a]anthracene and 6-ni-
trochrysene were not detected in the composite sample.
Am bient Am ount of Triphenylene and Nitrotriphen-
ylenes. Figure 5(a) shows the ambient concentrations of
triphenylene, pyrene, and benzo[e]pyrene in daytime and
nighttime airborne particulate samples. Triphenylene and
pyrene were present at higher levels in the daytime samples
than in those of the nighttime. Triphenylene was found at
concentrations ranging from 1.1 to 3.0 ng/ m3, and these were
comparable levels to that of pyrene (1.0-2.8 ng/ m3). Benzo-
[e]pyrene also was determined as a reference polycyclic
aromatic hydrocarbon (PAH) due to its high atmospheric
stability. The concentrations of benzo[e]pyrene and pyrene
obtained in this study coincided with those so far reported
in Japan during the wintertime, and no reported value has
been available for atmospheric concentration of triphenylene
(30). On the contrary, the concentrations of both 1- and
2-nitrotriphenylenes were higher in the nighttime samples
than in the daytime ones (Figure 5(b)). The difference be-
tween the daytime and nighttime samples is more pro-
nounced with 1-nitrotriphenylene. The concentrations of 1-
and 2-nitrotriphenylenes were much higher than that of
1-nitropyrene.
FIGURE 3. Mass chromatograms for molecular ions (m/z 273) and
the [M - NO]+ (m/z 243) fragment ions from the GC/MS-SIM analysis,
showing resolution of 1- and 2-nitrotriphenylenes (1- and 2-NTPs),
7-nitrobenz[a]anthracene (7-NBaA), and 6-nitrochrysene (6-NC). The
isomers in order of elution are as follows: 7-nitrobenz[a]anthracene,
1-nitrotriphenylene, 6-nitrochrysene, and 2-nitrotriphenylene.
nitrochrysenes, we then examined the resolution of 1- and
2-nitrotripenylenes from 7-nitrobenz[a]anthracene and 6-ni-
trochrysene by GC/ NPD as shown in Figure 2. Because
7-nitrobenz[a]anthracene in diesel exhaust particulates and
6-nitrochrysene in airborne particulates have been detected
(26, 27) and the preferred positions of nitration in benz[a]-
anthracene and chrysene are predominantly the 7- and
6-position (20, 28, 29), respectively, there is quite a possibility
that they will coexist in the samples. 7-Nitrobenz[a]an-
thracene and 6-nitrochrysene were eluted before the elution
of 1-nitrotriphenylene and 2-nitrotriphenylene, respectively,
indicating successful separation of 1- and 2-nitrotri-
phenylenes and the two isomers.
In addition, the purified fractions from the airborne
particulate samples were analyzed by GC/ MS-SIM. Mass
chromatograms for molecular ions, m/ z 273 ([M+]), together
with the characteristic fragment ions, m/ z 243 ([M - NO]+),
also indicated resolution of 1- and 2-nitrotriphenylenes from
the isomeric 7-nitrobenz[a]anthracene and 6-nitrochrysene
on a 30-m DB-5 column (Figure 3). No identification of the
six ambient daytime and nighttime samples was performed
since the detect limits of GC-MS-SIM for 1- and 2-nitro-
triphenylenes (10 and 10 ng, respectively) were 2 orders of
Ambient concentrations depend on the level of emissions
and meteorological conditions. However the result that
triphenylene, pyrene, and benzo[e]pyrene were present at
higher levels in the daytime than in the nighttime indicates
that the concentration of these PAHs sampled in this work
depended on the level of emissions strongly, because these
PAHs are emitted from automobile engine (31) and traffic is
usually heavy in the daytime. If 1- and 2-nitrotriphenylenes
as well as PAH were directly emitted mainly from a diesel
engine, they should be present at higher levels in the daytime
samples than in the nighttime ones. The high concentration
of 1- and 2-nitrotriphenylenes in the nighttime samples is
probably due to atmospheric reaction of triphenylene to form
1- and 2-nitrotriphenylenes. The most reactive PAHs such
as pyrene toward electrophilic nitration produced their
electrophilic nitration products under ambient high NOx
9
VOL. 34, NO. 10, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 8 9 5

FIGURE 5. Ambient concentration of triphenylene (TP), benzo[e]pyrene (BeP), pyrene (PY), 1- and 2-nitrotriphenylenes (1- and 2-NTPs),
and 1-nitropyrene (1-NP) in daytime and nighttime airborne particulate samples and concentration ratio of NTPs to BeP: (a) TP, BeP, and
PY, (b) 1-NTP, 2-NTP, and 1-NP, and (c) 1-NTP/BeP and 2-NTP/BeP.
determined because it is a well-known nitroarene in airborne
particulates. The greater mutagenic activity and higher
concentration of 2-nitrotriphenylene in the airborne par-
TABLE 1. Direct-Acting Mutagenic Activities (Revertants/nmol)
of Nitrotriphenylenes
ticulate samples than those of 1-nitropyrene clearly indicate
that 2-nitrotriphenylene is an important contributor to the
total mutagenic activity of the sample of airborne particulate
matter.
Salmonella typhimurium strain
(revertants/nmol ( SD) (-S9 mix)
TA98
YG1024
1-Nitrotriphenylene (3)
2-Nitrotriphenylene (2)
18 ( 6
3800 ( 250
480 ( 55
101 700 ( 4160
Form ation of Nitrotriphenylenes by Laboratory Nitra-
tion of Triphenylene. To understand the atmospheric
formation of nitrotriphenylene, a portion of 100 µg of
triphenylene deposited on a filter was exposed to 10 ppm
NO2 in the dark with O3 or with light irradiation for 1-30 h.
Figure 6 shows the amounts of nitrotriphenylenes in the
reaction products and the remaining triphenylene plotted
against reaction time. The reaction of triphenylene with NO2
in the dark without O3 did not occur. When triphenylene was
exposed to NO2-O3 in the dark (Figure 6(a)), 1- and
2-nitrotriphenylenes were immediately formed, and their
amount increased with the decrease in the amount of
remaining triphenylene. After the reaction for 8 h, nine kinds
of dinitrotriphenylenes were also formed probably due to
sequential nitration of nitrotriphenylenes. At the reaction
time of 19 h, the maximum yields of 1- and 2-nitrotriphen-
ylenes (9% and 15%, respectively) were attained, and 78% of
triphenylene was degraded. The amounts of dinitrotri-
phenylenes increased gradually until 24 h and then slowly
decreased. The maximum total yield reached 14%. On the
other hand, in the photoreaction (Figure 6(b)), the amount
of 2-nitrotriphenylene was small with a yield of up to 2% and
seemed to be constant after the reaction for 8 h, although
the degradation profile of triphenylene was almost the same
in both the dark reaction and the photoreaction. 1-Nitro-
triphenylene and dinitrotriphenylenes were not formed in
any significant amounts under light irradiation. It is clearly
shown that the nitration of triphenylene occurs more actively
in the presence O3 in the dark than in the absence of O3 with
light irradiation.
conditions and using filters spiked with deuterated PAH (32).
Since triphenylene is very unreactive toward electrophilic
nitration, 1- and 2-nitrotriphenylenes will not be formed
during collection (33). Figure 5(c) shows the concentration
ratio of 1- and 2-nitrotriphenylenes to benzo[e]pyrene in the
daytime and nighttime samples. The ratios of 1-nitrotri-
phenylene/ benzo[e]pyrene and 2-nitrotriphenylene/ benzo-
[e]pyrene were much higher in the nighttime samples than
in the daytime ones. This fact leads to the conclusion that
1- and 2-nitrotriphenylenes are at least partly formed by
nitration of triphenylene in the atmosphere and that the
formation of nitrotriphenylenes occurs easily during the
nighttime.
Mutagenic Activity of Nitrotriphenylenes. Table 1 lists
the mutagenic activity of 1- and 2-nitrotriphenylenes in two
Salmonella typhimurium strains TA98 and YG1024 in the
absence of S9 mix. The strain YG1024 that exhibits the
enhanced O-acetyltransferase activity is extremely sensitive
to nitroarenes and aromatic amines (25). We determined the
activity of 1- and 2-nitrotriphenylenes in strain YG1024 for
the first time. The activities of 1- and 2-nitrotriphenylenes
in strain TA98 determined in this study were 18 and 3800
revertants/ nmol, respectively, and these values are lower than
those reported by Greibrokk et al. (19). The mutagenicity of
2-nitrotriphenylene was over 2 orders of magnitude greater
than that of 1-nitrotriphenylene in both strains TA98 and
YG1024. 1-Nitropyrene has always been analyzed when the
direct mutagenic activity of airborne particulate samples was
9
1 8 9 6 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 10, 2000

FIGURE 6. Formation of nitrotriphenylene (nitro TP) by the exposure of 100 µg of triphenylene (TP) to 10 ppm NO : (a) dark reaction (with
2
O ) and (b) photoreaction (2) 1-nitrotriphenylene; (b) 2-nitrotriphenylene; (9) dinitrotriphenylenes; and (4) triphenylene. Dinitrotriphenylenes
3
were estimated by using authentic 1-nitrotriphenylene on the assumption that molar sensitivity of all dinitrotriphenylenes for GC/NPD is
consistent with that of monotriphenylenes.
The nitration of PAH in the presence of O3 in the dark
occurs mostly in the nighttime because the NO3 radical and
N2O5 are formed at night in polluted ambient air from the
reaction sequence
NO2 + O3 f NO3 + O2
NO3 + NO2 a N2O5
(1)
(2)
Under our experimental conditions, N2O5 and the NO3 radical
are most probably the reactive species because N2O5, NO3
radical, and NO2 are in equilibrium (34). In the reaction of
gas-phase PAH such as fluoranthene and naphthalene, an
addition-elimination sequence involving the NO3 radical
species has been proposed (5). We reported that 3-nitro-
benzanthrone is a new class of powerful direct-acting
mutagens of atmospheric origin and would be formed by
the reaction of benzanthrone in NO2-O3 in the atmospheric
environment (18). From the result of the variation in isomer
distribution on the conditions such as polarity of the organic
solvent and the temperature employed, the reaction of
benzanthrone in NO2-O3 would take place via the competi-
tion between the homolytic process involving the NO3 radical
and the heterolytic process involving NO2+ ion (35). We think
that the nitrotriphenylene formation by the dark reaction
with O3 is also different from the reaction involving the NO3
radical which is proposed in the reaction of gas-phase PAH.
Under photolysis conditions also, the NO3 radical and
N2O5 are formed since O3 is formed from the following
reactions
FIGURE 7. Mutagenic activities of the reaction products by the
exposure of 100 µg of triphenylene to 10 ppm NO as a function of
2
reaction time. The mutagenicity tests were performed on strains
TA98 (a) and YG1024 (b) in the absence S9 mix: (b) dark reaction
(with O ); (2) dark reaction (without O ); and (0) photoreaction.
3
3
Mutagenic Activity of Nitration Products of Triphen-
ylene. Dinitrotriphenylenes as well as mononitrotriphen-
ylenes were formed by the dark reaction of triphenylene with
NO2 in the presence of O3 in our study. Consequently, the
contribution of products such as dinitrotriphenylenes other
than mutagenic 2-nitrotriphenylene to the mutagenic activi-
ties of the whole reaction products should be understood.
In Figure 7, the change with time of mutagenic activities of
the reaction products is shown for two Salmonella typh-
imurium strains TA98 and YG1024 in the absence of S9 mix.
The change with time of mutagenic activities of the reaction
products for strain TA98 (a) and YG1024 (b) was similar and
reached the maximum at 24 h in agreement with the change
with time of the total nitrotriphenylene amount as shown in
Figure 6. The mutagenic activities for strain YG1024 were
about 15 times higher than those for strain TA98. This also
agrees that main mutagenic products were nitrotriphenylenes
since YG1024 is more sensitive to nitroarenes than TA98.
While the mutagenic activities of the products in the dark
reaction under NO2 in the absence of O3 was a little low,
those in the presence of O3 increased linearly until the
reaction time of 24 h and slowly decreased with an increase
in exposure time. The mutagenic activities of the photo-
reaction products were lower than those of the dark reaction
products in the presence of O3 for both strains and reached
a constant value after increasing linearly at 1-8 h.
NO2 + hν f NO + O
O + O2 + M f O3 + M
(3)
(4)
Furthermore, it is also possible that HONO will be generated
from such a high NO2 concentration and could photolyze,
producing OH radicals. The formation of nitro-PAH from
the OH radical-initiated reactions of gas-phase PAH has been
well documented (5). The photoreaction of triphenylene in
our experimental system showed high positional selectivity
at the 2-position, though the amount of formed 2-nitro-
triphenylene was small. These results indicate that a different
nitration mechanism is operative under dark conditions with
NO2 and O3 and under photolysis conditions in our systems
since the usual nitration of triphenylene with HNO3 in acetic
anhydride gave a small excess of a nitro substituted product
at the 1-position which is more susceptible to attack by an
To separate some direct-acting mutagens formed by the
dark reaction in the presence of O3 and photoreaction of
triphenylene with NO2 for 24 h, samples extracted from the
+
electrophile such as NO2 ion (20, 21).
9
VOL. 34, NO. 10, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 8 9 7

FIGURE 8. Preparative HPLC separation of the reaction products and mutagenic activities of each HPLC fraction tested with TA98 and
YG 1024 in the absence of S9 mix: (a) dark reaction (with O ) 24 h and (b) photoreaction 24 h (9) TA98 and (0) YG1024.
3
These results indicate that 2-nitrotriphenylene can be
largely attributed to the mutagenic activity of the dark reaction
products in the presence of O3 and dinitrotriphenylenes are
not. Additionally also in the photoreaction, 2-nitrotriphen-
ylene would be largely attributed to the mutagenic activity
of the reaction products because Fr.21, which contains
mononitrotriphenylenes, was the most active fraction.
The heterogeneous reactions studied herein indicate the
possibility for atmospheric formation of mutagenic 2-nitro-
triphenylene during the daytime and nighttime. The higher
concentration of nitrotriphenylenes in the nighttime samples
than in the daytime ones can be attributed to the effective
formation of nitrotriphenylenes by the dark reaction of
triphenylene with NO2 and O3. It is presumed impossible to
extrapolate the laboratory nitration of triphenylene with ppm
concentrations of reactants to the ambient conditions, since
ambient NO2 and O3 concentrations in Tokyo Minato-ku
were an average of 0.046 and 0.011 ppm, respectively, in
FIGURE 9. Gas chromatograms of the crude sample and HPLC
fractions from 21 to 24 in the dark reaction product in the presence
December, 1996. However, the heterogeneous reactions
studied herein should be regarded as an important process
for atmospheric formation of mutagenic 2-nitrotriphenylene.
of O for 24 h.
3
reaction products were fractionated by preparative HPLC.
All of the effluents were assayed for mutagenic activities for
Acknowledgments
Salmonella typhimurium strains TA98 and YG1024 in the
We thank Ms. Oosone, College Science and Technology,
absence of S9 mix. Fr.18-23 and Fr.25, which correspond to
Nihon University for technical support and also thank Ms.
the range of 17-22 and 24 min in the retention time in
Nakazato of the JEOL Co. for operating the mass spectrom-
chromatogram of the dark reaction products in the presence
eter. This work was supported by Grants-in-Aid for Envi-
of O3, showed high mutagenic activity as shown in Figure
ronmental Research from the Environment Agency.
8(a). Fr.21 was the most active fraction both in the dark
reaction and the photoreaction.
Figure 9 shows the gas chromatograms of the crude sample
and Fr.21-24 of the products by the reaction in NO2-O3 for
Literature Cited
(1) Rosenkranz, H. S.; Mermelstein, R. Mutat. Res. 1983, 114, 217-
24 h. The 12 peaks were observed in the gas chromatograms
of the crude sample. The compounds corresponding to two
peaks in the most active Fr.21 were identified as 1- and
2-nitrotriphenylenes on the basis of retention time and the
molecular ion together with the characteristic fragment ions
of authentic 1- and 2-nitrotriphenylenes. The compounds
corresponding to the nine peaks in relatively slightly active
Fr.23 and highly active Fr.24 were identified as dinitrotri-
phenylenes by detecting the molecular ion, m/ z 318 ([M+]),
together with the characteristic fragment ions, m/ z 288 ([M
- NO]+), 276 ([M - NO2]+), and 275 ([M - HNO2]+).
267.
(2) Tokiwa, H.; Ohnishi, Y. CRC Crit. Rev. Toxicol. 1986, 17, 23-60.
(3) Rosenkranz, H. S. Mutat. Res 1982, 101, 217-267.
(4) Gibson, T. L. Mutat. Res 1983, 122, 115-121.
(5) Atkinson, R.; Arey, J. Environ. Health Perspect. 1994, 102, 117-
126.
(6) Sasaki, J.; Aschmann, S. M.; Kwok, E. S. C.; Atkinson, R.; Arey,
J. Environ. Sci. Technol. 1997, 31, 3173-3179.
(7) Jager, J.; Hanus, V. J. Hyg. Epodrmiol. Microbiol., Immunol. 1980,
24, 1-12.
(8) Pitts, Jr., J. N.; Sweetman, J. A.; Zielinska, B.; Atkinson, R.; Winer,
A. M.; Harger, W. P. Environ. Sci. Technol. 1985, 19, 1115-112
9
1 8 9 8 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 10, 2000

(9) Hisamatsu, Y.; Nishimura, T.; Tanabe, K.; Matsushita, H. Mutat.
Res. 1982, 172, 19-27.
(10) Fan, Z.; Chen, D.; Birla, P.; Kamens, R. M. Atomos. Environ.
1995, 29, 1171-1181.
(11) Inazu, K.; Kobayashi, T.; Hisamatsu, Y. Chemosphere 1997, 35,
607-622.
(12) Arey, J.; Zielinska, B.; Harger, W. P.; Atkinson, R.; Winer, A. M.
Mutat. Res. 1988, 207, 45-51.
(13) Hayakawa, K.; Murahashi, T.; Butoh, M.; Miyazaki. M. Environ.
Sci. Technol. 1995, 29, 928-932.
(14) Zhou, J. L.; Fileman, T. W.; Evans, S.; Donkin, P.; Llewellyn, C.;
Readman, J. W.; Mantoura, R. F. C.; Rowland, S. J. Marine
Pollution Bul. 1998, 36, 587-597.
(15) Shane, B. S.; Squadrito, G. L.; Church, D. F.; Pryor, W. A. Environ.
Mol. Mutagen. 1991, 17, 130-136.
(16) Arey, J.; Harger, W. Helmig, H.; Atkinson, R. Mutat. Res. 1992,
281, 67-76.
(24) Yahagi, T.; Nagano, M.; Seino, Y. Mutat. Res. 1977, 121-130.
(25) Watanabe, M.; Ishidate, M., Jr.; Nohmi, T. Mutat. Res. 1990,
234, 337-348.
(26) Paputa-Peck, M. C.; Marano, R. S.; Schuetzlte, D.; Riley, T. L.;
Hampton, C. V.; Prater, T. J.; Skewes, L. M.; Jensen, T. E.; Ruehle,
P. H.; Bosch, L. C. Duncan, W. P. Anal. Chem. 1983, 55, 1946-
1954.
(27) Murahashi, T.; Hayakawa, K. Anal. Chim. Acta 1997, 343, 251-
257.
(28) Newman, M.; Lilie, K. J. Org. Chem. 1979, 44, 1347-1348.
(29) Dewar, M. J. S.; Mole, T. Warford, E. W. T. J. Chem. Soc. 1956,
3576-3580.
(30) Uraki, Y. Abstract of papers; 39th Annual Meeting of the Japan
Society of Atmospheric Environment, Sapporo, Japan Society
of Atmospheric Environment: Tokyo 1998; Abstract 2H1145.
(31) Sjogren, M.; Li, H.; Banner, C.; Rafter, J.; Westerholm, R.; Rannug,
(17) Sasaki, J.; Arey, J.; Harger, W. P. Environ. Sci. Technol. 1995, 29,
1324-1335.
U.; Chem. Res. Toxicol. 1996, 9, 197-207.
(32) Arey, J.; Zielinska, B.; Atkinson, R.; Winer, A. M. Environ. Sci.
Technol. 1988, 22, 457-462.
(33) Nielsen, T. Environ. Sci. Technol. 1984, 18, 157-163.
(34) Tuazon, E. C.; Sanhueza, E.; Atkinson, R.; Carter, P. L.; Winer,
A. M.; Pitts, J. N., Jr. J. Phys. Chem. 1984, 88, 3095-3098.
(35) Enya, T.; Suzuki, H.; Hisamatsu, Y. Bull. Chem. Soc. Jpn. 1998,
71, 2221-2228.
(18) Enya, T.; Suzuki, H.; Watanabe, T.; Hirayama, T.; Hisamatsu,
Environ. Sci. Technol. 1997, 31, 2772-2776.
(19) Greibrokk, T.; Lofroth, G.; Nilsson, L.; Toftgard, R.; Carlstedt-
Duke, J.; Gustafsson, J.-A. Toxicity of nitroaromatic compounds;
Rickert, D. E., Ed.; Hemisphere: Washington, DC, Chapter 11.
(20) Radner, F. Acta Chem. Scand. 1983, B37, 65-67.
(21) Barker, C. C.; Emmerson, R. G.; Periam, J. D. J. Chem. Soc. 1955,
4482-4485.
(22) Svendsen, H.; Ronnigsen, H.-P.; Sydnes, L. K.; Gribrokk, T. Acta
Chem. Scand. 1983, B37, 833-844.
(23) Ames, B. N.; McCann, J.; Yamasaki, E. Mutat. Res. 1975, 31,
347-364.
Received for review May 13, 1999. Revised manuscript re-
ceived September 7, 1999. Accepted February 22, 2000.
ES990553D
9
VOL. 34, NO. 10, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 8 9 9