Gas-phase reaction of phenol with NO3

Article abstract of DOI:10.1021/es001290m

The fast gas-phase reaction of NO₃ radicals with phenol was found to yield 2-nitrophenol as the only relevant nitration product. The yield of this productwas high and independent of the concentration of NO₂ at the concentrations appl

Full text of DOI:10.1021/es001290m

Environ. Sci. Technol. 2001, 35, 1791-1797
nitrate radical NO3 with aromatic compounds in the gas phase
Gas-Phase Reaction of Phenol with
(11). The reaction of phenol with NO3 in gas phase to give
mainly 2-nitrophenol has been recently reported (7), but the
reasons for the exclusive ortho-regiochemistry are still
uncertain. On the contrary, the reaction of phenol with NO3
in liquid phase has been reported to give both 2- and
4-nitrophenol in organic solvents (12) and in water (13).
Understanding the mechanism of the nitration of phenol
both in the gas phase and in the aqueous phase is important
because nitrophenols are found in air and rainwater. These
compounds have been found in rainwater in the microgram
per liter range in the Rhein-Ruhr area in western Germany
(14, 15) and in clouds (16, 17). However, only o-nitrophenol
was found in Los Angeles rain (18). In urban Milano, only
p-nitrophenol was found in rainwater in a summer campaign
in 1998 (19). More recently, air samples were analyzed for
nitrophenols, and the ratio of o-/ p-nitrophenol was found
to vary with the location of the sampling station and the
season. In fact, in Milano p-nitrophenol was always found,
and occasionally also o-nitrophenol was present (20). More
p-nitrophenol is found in air sampled in an open space than
in a tunnel in Wuppertal (21). Moreover, nitrophenols are
known to be present in car exhaust (22), and an o-/ p-
nitrophenol ratio of 1.3 was measured in the exhaust. The
finding of only some of the possible nitrophenols (3 isomers),
dinitrophenols (6 isomers), nitrocresols (9 isomers), and
dinitrocresols (16 isomers) may be used for the identification
of sources and where tropospheric processing has occurred.
Hence, studying the mechanism of the gas-phase nitration
of phenol could complement recent studies on the aqueous-
phase formation of nitrophenols (23) and could explain the
different content in nitrophenols of air and rainwater samples.
NO
3
E Z I O B O L Z A C C H I N I , †
M A U R I Z I O B R U S C H I , † J E N S H J O R T H , ‡
S I M O N E M E I N A R D I , † M A R C O O R L A N D I , †
B R U N O R I N D O N E , * , † A N D
‡
E L I S A R O S E N B O H M
Department of Environmental Sciences, University of
Milano-Bicocca, Piazza della Scienza 1,
I-20126 Milano, Italy, and Commission of the European
Communities, Joint Research Centre, Environment Institute,
21020 Ispra (VA), Italy
The fast gas-phase reaction of NO radicals with phenol
3
was found to yield 2-nitrophenol as the only relevant nitration
product. The yield of this product was high and independent
of the concentration of NO at the concentrations applied.
2
In the presence of ozone, also significant amounts of
4-nitrophenol and p-benzoquinone were formed. The rate
constant of the reaction between NO radicals and phenol
3
was determined to be 5.8 × 10-12 cm3 molecule-1 s-1
.
The selective formation of 2-nitrophenol (2) is suggested
to derive from either the concerted keto-enol tautomerism
in the reaction of a phenoxy radical with NO or the
2
concerted elimination of nitric acid from a cyclohexa-3,5-
diene intermediate.
Experimental Section
Introduction
Gas-Phase Reactions. The reaction was studied in purified
air at 760 Torr in a 480-L Teflon-coated 60 cm diameter
cylindrical chamber equipped with a 81.2 m total beam path
length White type mirror system coupled to a Bruker IFS
113V FT spectrometer for on-line infrared spectroscopy.
Concentrations of reactants and reaction products were
monitored in situ by long path infrared spectroscopy.
Experiments were performed at ambient pressure and
temperature. Calibrations of the IR measurements were
performed either by introducing known volumes of vapor at
known pressures into the chamber or by evaporating known
quantities of the compounds and transferring them with a
stream of air into the chamber. Calibrated gas-phase spectra
of o-nitrophenol and p-nitrophenol were obtained, while
the dinitrophenols were found to have a too low volatility to
be calibrated in the gas phase.
In the experiments to study yields of nitrophenols, N2O5
was synthesized in the chamber by mixing O3 with an excess
of NO2. Subsequently, the phenol was added and the reaction
was allowed to proceed until only insignificant amounts
(<100 ppb) of N2O5 remained. Initial concentrations were
10-25 ppmV N2O5, 3-55 ppmV NO2, and 15-30 ppmV of
the organic compound. In some experiments, phenol and
NO2 were added first to the chamber and then ozone. The
initial concentrations in this case were 11-51 ppmV NO2,
22-31 ppmV phenol, and 16-70 ppmV ozone.
Phenols are known to be formed in the atmosphere by the
OH-initiated atmospheric oxidation of aromatic hydrocar-
bons, and a number of product studies reported in the
literature show that their yields can be relatively high (1).
Phenols and nitrophenols may be emitted in the exhaust (2,
3) as partial oxidation products of aromatic hydrocarbons.
Phenols can then subsequently be transformed into
nitrophenols, which are known to be phytotoxic (4). The
nitrate radical, NO3, may be the promoter for these reactions.
This radical is formed in the atmosphere by the reaction
between O3 and NO2 (5):
NO2 + O3 f NO3 + O2
NO3 is recognized to play an important role in the nighttime
tropospheric chemistry because of its fast reaction with other
radicals and with several organic trace gases, in particular
those of biogenic origin (6). The reactions of NO3 radicals
with phenol and cresols are fast both in the gas phase (0.4-
1.4 × 10-11 cm3 molecule-1 s-1) (7, 8) and in the aqueous
phase (k ) 1.8 ( 0.3 × 109 M-1 s-1 at pH ) 0 and T ) 298
K) (9). Hence, these reactions are a sink (10) for NO3, and
nitrophenols may be formed by tropospheric gas-phase
reactions or within the tropospheric aqueous phase.
We have recently demonstrated that isomer distribution
in the reaction products and polar effects in the rate constants
may help to elucidate some details of the reactions of the
The rate constant of the reaction of the NO3 radical with
phenol was determined relative to that of its reaction with
2-methyl-2-butene. In these experiments, phenol and 2-meth-
yl-2-butene were added first to the chamber, and then N2O5
was introduced by evaporating solid N2O5, prepared by the
method of Schott and Davidson (24), and bringing it into the
* Corresponding author phone: +39 02 6448 2813; fax: +39 02
6448 2890; e-mail: Bruno.Rindone@unimib.it.
† University of Milano-Bicocca.
‡ Environment Institute.
9
10.1021/es001290m CCC: $20.00
Published on Web 04/04/2001
2001 Am erican Chem ical Society
VOL. 35, NO. 9, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 7 9 1

FIGURE 1. Plots of examples of observed decays of phenol vs 2-methyl-2-butene in the reaction with NO radicals under conditions of
3
high and low NO concentrations.
2
chamber with a stream of air. In this case, the initial
concentrations were 15-18 ppmV of 2-methyl-2-butene and
5-10 ppmV of phenol. N2O5 was in most experiments added
several times during the experimental run in portions of a
few ppmV. Initial NO2 concentrations were 1-3 ppmV in
these experiments. In other experiments, NO2 was added to
provide initial concentrations of 29-59 ppmV in order to
test the influence of [NO2] on the observed rate constant.
Product Studies. The products were measured using two
different analytical methods:
known rate constant. The reference compound was, in this
case, 2-methyl-2-butene, and the rate constant k1 for the
reaction
NO3 + phenol f products
could then be calculated relative to the rate constant, k2, for
the reaction
NO3 + 2-methyl-2-butene f products
(i) FTIR analysis in the reaction chamber.
(ii) Samples to be used for subsequent product analysis
were obtained by bubbling known volumes of the reaction
mixture through a 5.3 × 10-4 M sodium hydrogen carbonate
solution. The extraction was performed with 10 mL of
dichloromethane after neutralization with sulfuric acid. HPLC
with UV detection using a HPLC-DAD instrument (Waters
600E system controller and HP 1040 A diode array detector)
was used to detect and quantify all the isomers monitoring
the absorbance at 230 nm. The column that gave the best
performance was a stainless steel column, 250 mm × 4.4
mm i.d., packed with 5-µm particles (100 Å) of silica-based
C18 material, Kromasil (Superchrom). The number of
theoretical plates for phenol was calculated to be about
100 000. The flow was 1 mL min-1. We applied an isocratic
gradient using a mixture of two different solutions: The first
(solution A) was a mixture of an aqueous solution of KH2PO4
(5 × 10-3 M), phosphoric acid was added to adjust the pH
to 4.5 (90 vol % of A) and CH3CN (10 vol %). The second
(solution B) was a mixture of an aqueous solution of KH2PO4
(5 × 10-3 M), phosphoric acid was added to adjust the pH
at 4.5 (25 vol % of B) and CH3CN (75 vol %). The two solutions
in a volume ratio, A/ B ) 55/ 45, were filtered through 0.45-
µm filters (Millipore, Bedford, MA) and degassed with helium
during analysis. All reference compounds were purchased
from Fluka or Aldrich.
from the relation
phenol(t0)
phenol(t)
k
2-methyl-2-butene(t0)
2-methyl-2-butene(t)
ln
)
1 ln
{
}
{
}
k2
The ratio k1/ k2 was measured for relatively low and
relatively high concentrations of NO2, as discussed in the
Experimental Section. The ratio k1/ k2 was found to be 0.62
( 0.12, which gives a k1 value of 5.8 ( 1.1 × 10-12 cm3
molecule-1 s-1, using the value of 9.38 × 10-12 cm3 molecule-1
s-1 for k2 (4) and giving the uncertainty as one standard
deviation on the experimental data. The rate constant was
independent of the [N2O5]/ [NO2] ratio (Figure 1), thus
showing that the reaction of phenol takes place with NO3
radicals and not with N2O5 or with NO2. This value was slightly
higher than that found at 296 ( 2 K (3.92 ( 0.25) (5) and at
298 ( 1 K (3.64 ( 0.14) (28). The reason for this difference
is not obvious; wall losses are not a likely explanation since
the loss of phenol in clean air in the cell was measured to
be about 5% per hour and the reaction between NO3 radicals
and phenol under the conditions applied is completed within
a much shorter time period. The initial concentrations of
phenol applied and the amount of phenol that reacted in the
present experiments were roughly 10 times higher than what
was applied in the work of Atkinson et al. (5), and one might
speculate about uptake of phenol on particles formed. Finally,
of course, different analytical techniques were applied (IR
spectroscopy vs GC-FID), and errors in one or the other
method may also be an explanation for the different results.
Products in Nitration of Phenol by Reaction with NO3
Radicals. High yields of 2-nitrophenol (2) were found in the
gas-phase nitration of phenol (1) initiated by NO3 radicals.
Table 1 shows yields obtained under different conditions
and measured by FTIR. In particular, two experimental
procedures were used:
Calculations. All calculations were performed using the
ab initio Density Functional Theory (DFT) with the hybrid
functional of Becke (25) and Lee et al. (26) and a flexible
Gaussian basis of type 6-31G** (27).
Results
Rate Constant. The rate constant of the reaction of NO3
radicals with phenol at 295 ( 2 K was investigated by the
relative rate method, where the decay of phenol in an NO3
radical containing atmosphere is compared to that of a
reference compound that reacts with NO3 radicals with a
9
1 7 9 2 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 9, 2001

TABLE 1. Gas-Phase Nitration of Phenol in Different Conditionsa
N2O
start,
ppm
NO
start,
ppm
NO
after,
ppm
NO
used,
ppm
av
HNO
3
after,
ppm
phenol
start,
ppm
conv
phenol,
ppm
5
2
2
2
NO ,
2-nitrophenol,
%
4-nitrophenol,
%
ratio
2-/4-nitrophenol
2
exp
ppm
1
2
3
4
5
6
7
8
9
12.2
21.8
22.5
23.4
-
-
-
-
-
39.1
56.8
21.7
3.6
23.0
32.6
8.4
40.4
60.8
20
6.3
0.9
31.4
8.5
16.9
14.8
23.1
39.6
46.7
44.1
49.7
19.2
13.2
28.7
27
39.8
58.8
20.8
4.9
12.0
32
8.4
17.3
17.7
7.2
15.2
14.7
18.9
21.1
9.3
14.1
21.8
21.2
14
20
5.6
12.5
10
11.8
17.1
7.2
10.8
13.5
12
29
47
38
65
55
68
47
69
55
25.7
25.9
30.9
22.3
26.2
21.6
30.8
17
31
21
22
24
40
3.82
1.77
3.24
2.14
2.88
1.38
17.7
20.7
a
The table shows initial and final experim ental conditions. In experim ents 1-4, phenol is added after the addition of ozone and NO2 while in
experim ents 5-9 ozone is added last. Initial [N2O5] in experim ents 1-4 is not m easured directly but calculated from the difference between NO2
added to the cham ber and [NO2] found after the reaction with ozone, assum ing that all NO2 has reacted to form N2O5. The estim ated uncertainties
on [NO2], HNO3, and [O3] are 10% and on N2O5 and 2-nitrophenol are 20%. 4-Nitrophenol presents a particular problem as discussed in the text
SCHEME 1
SCHEME 2
(i) NO3 radicals generated by the thermal dissociation of
N2O5 in the presence of excess NO2 were reacted with 1. In
this case (reactions 1-4), only 2 was found and 4-nitrophenol
(3) was not observable by FTIR spectroscopy. At the NO2
levels applied, no dependence of the yields in 2 on the average
concentration of NO2 during the experiments could be
observed. No correction of the observed yields were made
for losses of nitrophenols due to the reaction with NO3, since
this reaction is more than 100 times slower than the reaction
of NO3 radicals with phenol (7).
(ii) A different situation was observed when 1 reacted
with NO3 radicals in the presence of ozone. In this case
(reactions 5-9), significant amounts of 3 were observed.
The predominant formation of 2 in some conditions was
confirmed by off-line analytical procedures. These analyses
showed that, besides 2 and 3, also minor amounts of 2,4-
dinitrophenol (5) and 2,6-dinitrophenol (6) were present in
molar yields of 1-3% each. In all the experiments, p-
benzoquinone (4) was also found present in molar yields of
3-6% (Scheme 1). A comparison between the on-line FTIR
analysis of 2 and the off-line HPLC analysis showed that the
latter underestimates the amount of 2 by more than 20%.
The loss of 2 in the off-line analysis may be due to losses in
the sampling line or to oxidative degradation of 2 in alkaline
solution.
Phenols are known to react with NO3 in aqueous acidic
conditions via a single electron transfer (SET) (29) giving a
cation radical such as 7, which is in prototropic equilibrium
with the phenoxy radical (8) and a nitrate anion (Scheme 2).
Reaction of 8 with NO2 should give both 2 and 3 via tautomers
9 and 10. This behavior is also observed in the reaction of
NO3 with electron-rich aromatics (30). Hence, the interme-
diate formation of 8 explains the observed selectivity for the
orthonitration only if, in gas phase, reaction 9 w 2 is much
faster than reaction 10 w 3. This may be due to the concerted
keto-enol tautomerism 9 w 2. The phenoxy radical 8 may
also derive from the direct hydrogen abstraction from the
phenolic hydroxyl (H-ABS).
Discussion
Regarding the quality of the experimental data on the yields
of nitrophenols, it should be mentioned that the experience
from the present and other experimental studies indicate
that 2-nitrophenol tends to remain in the gas phase under
the experimental condions applied and thus can be measured
accurately by IR spectroscopy. However, we have found that
4-nitrophenol, which has higher melting and boiling points
than 2-nitrophenol, has a stronger tendency to stick on
particles, and thus on-line IR spectroscopy may not be the
best way of measuring this compound. An inspection of the
data in Table 1 does not suggest any evident influence on
initial concentrations on the product yields.
A third possibility is that 1 reacts with NO3 via an addition-
elimination mechanism such as that depicted in Scheme 3
(31). Here, four regioisomeric adducts 11-14 are formed
with an equilibrium reaction in the ADD mechanism. These
will then add NO2 to give diastereoisomeric cyclohexadienes
15-21. Final elimination of HNO3 will form 2 and 4 from the
ipso adduct (11) or the meta adduct (12); 3-nitrophenol (22)
from the ortho adduct (13) or the para adduct (14).
To better understand the mechanism involved in these
reactions, structural and energetic data for the possible
reaction intermediates were investigated theoretically. Ge-
9
VOL. 35, NO. 9, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 7 9 3

SCHEME 3
ometry optimizations and energy determinations performed
at UB3LYP/ 6-31G** level of theory of the molecules partici-
pating to reactions in Scheme 3, showed that hydrogen
abstraction to give a phenoxy radical 8 is exothermic by 14.94
kcal/ mol. The SET mechanism is believed not to be possible
in gas phase.
Concerning the ADD mechanism, the regiochemistry of
the adduct under equilibrium condition may influence the
distribution of reaction products. Since energy differences
among different adducts (ipso, ortho, meta, and para) is
supposed to be small, and in the same order of magnitude
of conformational energies, an analysis of potential energy
surfaces (PES) in order to locate all possible stable confor-
mations is needed. Analysis of the PES was performed for
the rotation around C-O and O-N bonds [C1-C2-O2-N1
)θ;C1-C2-O1-H1 )æ] (Scheme 4) using the semiempirical
PM3 Hamiltonian in the unrestricted formalism (UHF/ PM3)].
For each point of the PESs all other degrees of freedom were
fully optimized (Figure 2).
(154.76°) (Scheme 4) are consistent with the formation of an
intramolecular seven-membered ring hydrogen bond be-
tween O3 and H1. The energy of the hydrogen bond can be
roughly estimated equal to -3.14 kcal/ mol from the difference
between the energy of this conformer (conformation 13-2 in
Table 2) and the conformation 13-3 in which the hydrogen
bond is not present.
Contrary to what has been suggested earlier (10), ipso-
hydroxynitratocyclohexadienyl adduct is less stable than the
ortho adduct by 8.67 (conformation 11-1) and 4.50 kcal/ mol
(conformation 11-2). Hence, if the stability of ortho adduct
is governing the regiochemistry, 3-nitrophenol (22) should
be formed.
The stability of the intermediate adduct had been used
to understand the preferential formation of 2-methylphenol
in the hydroxylation of toluene with OH in gas phase (28).
However, with 1 the stability of the adduct was unable to
explain the ortho regiochemistry in the gas-phase nitration.
Hence, some other factor should be central for the regiose-
lectivity. A model reaction could be useful for the identifica-
tion of this additional effect. The combined action of ozonized
oxygen and nitrogen dioxide in organic (Kyodai nitration) to
yield ortho and para nitration of phenol was used for this
purpose. The reactant in the Kyodai nitration is the NO3
radical. The reaction was suggested by us to occur through
The stable conformers obtained from the PES analysis
were re-optimized at UB3LYP/ 6-31G** level of theory. The
energies of the first reaction shown in Scheme 3 computed
at this level are shown in Table 2. One of the ortho conformers
is the most stable (∆E ) -10.0 kcal/ mol). For this conformer
the distance O3-H1 (1.857 Å) and the angle O3-H1-O1
9
1 7 9 4 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 9, 2001

FIGURE 2. Potential energy surfaces.
the formation of a charge-transfer complex (CTC) 23 that
evolves to adducts 11-14 (Scheme 5) (12).
NO2 to give an equilibrium mixture (33) of the two mesoforms
of 1,4-cyclohexadienes (15), thus leading to both 2 and 3, to
a diastereoisomeric mixture of the four isomeric of 1,3-
cyclohexadienes (16), and to 12, thus leading to 2. 12 will
react with NO2 to give a diastereoisomeric mixture of the
four isomeric of 1,3-cyclohexadienes (17), thus leading to 2
and to a diastereoisomeric mixture of the four isomeric of
1,3-cyclohexadienes (18), thus leading to 3. Hence, together
with the regioselective addition of NO2, also a regioselective
loss of nitric acid from one of the adducts is necessary to
understand the reasons for the preferred regioselective ortho
nitration of phenol.
The Kyodai nitration of phenols was demonstrated to be
a radical process by the low sensitivity to the electronic
contribution of the nuclear substituent, as shown by the low
Hammett F value of ) -0.72 using Hammett’s σ of the
substituent. A fast formation of a CTC followed by recom-
bination to give radical adducts such as 11-14 in the case
of 1 could explain this behavior. A similar mechanism should
occur in the gas-phase nitration of phenols with NO3 in the
presence of NO2. Here in fact the available rate constants are
5.8 × 10-12 for phenol and 10.7 × 10-12 for p-cresol (7). These
numbers suggest a Hammett F value smaller than that found
in the Kyodai nitration (nearly 0), thus suggesting that a
similar radical mechanism has to be occurring for the two
groups of aromatic substrates.
However, in gas-phase experiments where only 2 was
found, some effect directing the reaction toward the ortho
regiochemistry in the nitration reaction should be operating.
The exclusive ortho regiochemistry is the result of a selective
reaction of NO2 at the position ortho to the phenolic
hydrogen.
The first point is to understand the reason of the preferred
•-
formation of adduct 11. In the CTC, the NO3 fragment is
likely to be hydrogen bonded to the positive oxygen in the
[C6H5OH]•+ fragment. Hence, a concerted collapse of the CTC
to form selectively 11 under kinetic control (Scheme 6) could
drive the reaction toward the ortho regiochemistry. Moreover,
the fact that the yield of 2 is independent from the
concentration of NO2 suggests again that the collapse of the
CTC to 11-14 is rate-determining and that the reaction
between the adducts and NO2 are much faster than the
corresponding reactions with O2 at the NO2 concentrations
applied in this study.
Hence, the reaction could occur via the intermediate
formation of a phenoxy radical (8) if the keto-enol tau-
tomerism 9 w 2 is concerted. Alternatively, the exclusive
formation of 2 may be understood if the ipso adduct (11) or
the meta adduct (12), although not the most stable adducts,
are the intermediates in this reaction. In fact, 2 cannot derive
from 13 or 14.
Concerning the product-determining steps, it should be
noted that in the gas-phase elimination reaction may occur
only via the E2 mechanism, and this requires a strict
stereochemical relationship among the two groups to be
eliminated (H+ and NO3-). As shown in Scheme 6, only the
cis and trans form of the 1-hydroxy-1-nitrato-2-nitro-3,5-
cyclohexadienes (16) may undergo a concerted elimination
Formation of 11 or 12 is not the only condition to
understand the regioselectivity. In fact, 11 will react with
9
VOL. 35, NO. 9, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 7 9 5

SCHEME 5
TABLE 2. Energy (kcal mol-1) of the Reaction NO + Phenol
3
f Radical Adducts 11-14
SCHEME 6
CTC (Scheme 3). Another possibility is that, for some reason,
under the conditions of the experiments where ozone is
present, less 4-nitrophenol is adsorbed on particles and thus
the compound can be observed in the gas phase.
SCHEME 4
Atmospheric Implications
The reaction of NO3 radicals with phenol and cresols are fast
(0.4-1.4 × 10-11 cm3 molecule-1 s-1) and thus a potentially
important atmospheric sink for these at nighttime and a
potentially relevant source of the corresponding nitroderiva-
tives. A high yield of nitrophenol is found, which does not
show an evident dependence on the NO2 concentration under
the experimental conditions applied. If this insensitivity to
[NOx] will be valid also at typical atmospheric levels, and
thus if equally high yields will be found under ambient
conditions, cannot be determined from the present experi-
mental results but will require further studies (e.g., in large
simulation chambers).
Chamber studies indicate that 2 will be the main nitration
product of the gas-phase reactions of phenol with NO3 and
with OH radicals. This, combined with information about
the selectivity of aqueous-phase nitration reactions with
respect to the formation of isomers, may allow us to draw
conclusions about the origin of nitrophenols in the environ-
ment (e.g., in rainwater) based on the ratio between isomers.
In fact, the ratio 2-nitrophenol/ 4-nitrophenol of 1.5 has been
found in engine exhaust (22). Moreover, this ratio both in air
and in rain samples has been found to be highly variable (18,
20, 21). The different physicochemical characteristics of these
isomeric nitrophenols may play a role in their atmospheric
behaviors. The influence of ozone on the yield of the
4-nitrophenol isomer is mechanistically interesting but is
not likely to be relevant under atmospheric conditions, where
ozone concentrations are typically 2-3 orders of magnitude
lower than in the laboratory experiments.
of nitric acid in a E2-type elimination reaction, and this will
give selectively 2. The conclusion is that the initial CTC
formation, its collapse to 11, and the subsequent addition
of NO2 are equilibrium reactions Hence, the irreversible loss
of nitric acid will drive the reaction toward 2. In conclusion,
the selective formation of 2 is suggested to derive from either
the concerted keto-enol tautomerism in the reaction of a
phenoxy radical with NO2 or in the concerted elimination of
nitric acid from a 1,2-dihydrobenzene intermediate.
In the experiments where some ozone was present when
1 was introduced in the reaction chamber (reactions 4-9),
both nitrophenols were formed. We do not see any evident
reason for this behavior. A hypothetical explanation could
be that ozone assists in the abstraction of a hydrogen atom
or a proton from the 4-position in the intermediate 10 and
thus facilitates the formation of the nitrophenols. Alterna-
tively, if the ADD mechanism is occurring, ozone could inhibit
the concerted recombination of the two fragments in the
9
1 7 9 6 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 9, 2001

(18) Kawamura, K.; Kaplan, I. R. Organic Compounds in Rainwater.
In Organic Chemistry of the Atmosphere; Hansen, L. D., Eatough,
D. J., Eds.; CRC Press: Boca Raton, 1991; p 233.
(19) Belloli, R.; Barletta, B.; Bolzacchini, E.; Meinardi, S.; Orlandi,
M.; Rindone, B. J. Chromatogr. A 1999, 846, 277-281.
(20) Belloli, R.; Bolzacchini, E.; Clerici, L.; Rindone, B.; Sesana, G.;
Librando, V. Atmos. Environ. (submitted for publication).
(21) Kurtenbach, R. Personal communication.
(22) Tremp, J.; Mattrel, P.; Fingler, S.; Giger, W. Water, Air, Soil Pollut.
1993, 68, 113-123.
(23) Uptake and Nitration of Aromatic Compounds in the Tropo-
spheric Aqueous Phase, EC Contract Research No. ENV4-CT97-
0411, Final Report.
Acknowledgments
Financial support of this work from the European Com-
munity for Scientific Research (ENV4-CT97-0411) is grate-
fully acknowledged. We thank also the Italian CNR for
financial support.
Literature Cited
(1) Atkinson, R. J. Phys. Chem. Ref. Data 1994, Monogr. No. 2.
(2) Unregulated Motor Vehicle Exhaust Gas Components; Volk-
swagen AG Report; 1989.
(3) Tremp, J.; Mattrel, P.; Fingler, S.; Giger, W. Water Air Soil Pollut.
1993, 68, 113-123.
(4) (a) Shea, P. J.; Weber, J. B.; Overcash, M. R. Residue Rev. 1983,
87, 1. (b) Schafer, W. E.; Scho¨ nherr. Ecotoxicol. Environ. Saf.
1985, 10, 239.
(5) Finlayson-Pitts, B. J.; Pitts, J. N., Jr. Atmospheric Chemistry:
Fundamentals and Experimental Techniques; Academic Press:
New York, 2000.
(6) Atkinson, R. J. Phys. Chem. Ref. Data 1991, 20, 459-507.
(7) Atkinson, R.; Aschmann, S. M.; Arey, J. Environ. Sci. Technol.
1992, 26, 1397-1403.
(8) Bolzacchini, E.; Hjorth, J.; Meinardi, S.; Orlandi, M.; Rindone,
B.; Rosenbohm, E., Eurotrac-2 Symposium Chemical Mechanism
Determination Subproject, Karlsruhe, September 23-25, 1998.
(9) Bolzacchini, E.; Umschlag, Th.; Herrmann, H.; Meinardi, S.;
Rindone, B. 6th FECS Conference on Chemistry and the
Environment, Copenhagen, Denmark, August 26-28, 1998.
(10) Colville, R. N.; Chularton, T. W.; Gallager, M.-W.; Wicks, A. J.;,
Downer, R. M.; Tyler, B. J.; Storeton West, K. J.; Fowler, D.;
Cape, J. N.; Dollard, G. J.; Davis, T. J.; Jones, B. M. R.; Penkett,
S. A.; Bendy, B. J.; Burgess, R. A. Atmos. Environ. 1994, 28, 397.
(11) Bolzacchini, E.; Meinardi, S.; Orlandi, M.; Rindone, B.; Hjorth,
J.; Restelli, G. Environ. Sci. Technol. 1999, 33, 461-468.
(12) Barletta, B.; Bolzacchini, E.; Meinardi, S.; Orlandi, M.; Rindone,
B. Environ. Sci. Technol. 2000, 34, 2224-2230.
(24) Schott, G., Davidson, N. J. Am. Chem. Soc. 1958, 80, 1841-
1853.
(25) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
(26) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785-789.
(27) Frisch, M. J.; Trucks, G. W.; Schlege, H. B.; Gill, P. M. W.; Johnson,
B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T. A.; Petersson, G.
A.; Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.;
Zakrzewski, V. G.; Ortiz, J. V.; Foresman, J. B.; Cioslowski, J.;
Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.; Peng, C. Y.;
Ayala, P. Y.; Chen, W.; Wong, M. W.; Andres, J. L.; Replogle, E.
S.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Binkley, J. S.; Defrees,
D. J.; Baker, J.; Stewart, J. P.; Head-Gordon, M.; Gonzales, G.;
Pople, J. A. Gaussian ‘94; Gaussian, Inc.: Pittsburgh, PA, 1995.
(28) Atkinson, R. J. Phys. Chem. Ref. Data 1991, 20, 459-407.
(29) Atkinson, R.; Aschmann, S. M.; Winer, A. M. Environ. Sci Technol.
1987, 21, 1123-1125.
(30) Del Giacco, T.; Baciocchi, E.; Steenken, S. J. Phys. Chem. 1993,
97, 5451-5456.
(31) Herrmann, H.; Zellner, R. Reactions of NO3 in Aqueous Solution.
In Chemistry of N-Centered Radicals; Alfassi, Z., Ed.; Wiley: New
York, 1998; pp 291-343.
(32) Ito, O.; Akhido, S.; Iino, M. J. Org. Chem. 1989, 54, 2436-2440.
(33) (a)Andino, J. M.; Smith, J. N.; Flagan, R. C.; Goddard, W. A., III;
Seinfeld, J. H. J. Phys. Chem. 1996, 100, 10967-10980. (b) Lay,
T. H.; Bozzelli, J. W.; Seinfeld, J. H. J. Phys. Chem. 1996, 100,
6543-6554.
(13) Palm, W.-U.; Behnke, W.; Zetzsch, C. Presented at the EC
Workshop “Chemical Mechanisms of Atmospheric Processes”,
Copenhagen, 1998.
(14) Levsen, K.; Behnert, S.; Priess, B.; Svoboda, M.; Winkeler, H. D.;
Zietlow, J. Chemosphere 1990, 21, 1037-1061.
(34) Bohn, B.; Zetzsch, C. Phys. Chem. Chem. Phys. 1999, 1, 5097-
5107.
(15) Lu¨ ttke, J.; Levsen, K.; Scheer, V.; Wu¨ nsch, G.; Cape, J. N.;
Hargreaves, K. J.; Storeton-West, R. L.; Acker, K.; Wieprecht, W.;
Jones, B. V. Atmos. Environ. 1997, 16, 2637-2648.
(16) Geissler, A.; Scho¨ ler, H. F. Vom Wasser 1993, 80, 357.
(17) Geissler, A.; Scho¨ ler, H. F. Water Res. 1994, 28, 2047.
Received for review May 23, 2000. Revised manuscript re-
ceived October 27, 2000. Accepted December 11, 2000.
ES001290M
9
VOL. 35, NO. 9, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 1 7 9 7