A mechanistic study of the oxidation of phenol by OH/NO2/NO3 in aqueous solution

Article abstract of DOI:10.1039/b201652d

The flash photolysis of nitrate anions at λ = 248 nm and peroxydisulfate anions at λ = 351 nm was used to study the oxidation process of phenol by OH/NO₂ and NO₃/NO₂in aqueous solution under different experimental conditio

Full text of DOI:10.1039/b201652d

View Article Online / Journal Homepage / Table of Contents for this issue
A mechanistic study of the oxidation of phenol by OH/NO /NO3
2
in aqueous solution
Paolo Barzaghi and Hartmut Herrmann*
Institut f u¨ r Troposph a¨ renforschung, Permoserstr. 15, Leipzig, Germany. Fax: +49 341 235 2325
Received 14th February 2002, Accepted 7th May 2002
First published as an Advance Article on the web 19th June 2002
The flash photolysis of nitrate anions at l ¼ 248 nm and peroxydisulfate anions at l ¼ 351 nm was used to
2 3 2
study the oxidation process of phenol by OH/NO and NO /NO in aqueous solution under different
experimental conditions. Two different mononitrophenols (isomer ortho- and para-) and a dihydroxy derivative
catechol) were identified as the main reaction products by means of HPLC-DAD-ED technique and their
yields of formation where directly compared with the initial radical concentrations of OH, NO and NO ,
(
2
3
respectively. Product yields were determined for different pH and as a function of the number of laser photolysis
pulses down to single pulse experiments, where yields of 70 nM o-nitrophenol, 60 nM p-nitrophenol and 10 nM
catechol have been obtained starting with [OH]
radicals at pH ¼ 0.5 and T ¼ 298 K. From single pulse experiments under varying conditions a mechanism
involving (i) initial attack of phenol by OH and (ii) by initial attack of NO followed by the reaction of the
0
¼ [NO
2
]
0
¼ 160 nM leading to the formation of 50 nM of NO
3
3
intermediate with NO is suggested. Absolute product yields are determined. For OH/NO the o-/p-
2
2
nitrophenol product ratio is r ¼ 1.5 whereas for the NO
3
/NO
2
reaction sequence it is r ¼ 1.0. Implications for
tropospheric chemistry are discussed.
1
. Introduction
Emissions produced by combustion processes of fossil fuels by
power plants, factories, households, and automobile exhaust
contain precursors of phenols (BTX) as well as phenols. These
1
–3
compounds may affect human health.
phytotoxicity of nitro- and dinitrophenols is well documen-
Furthermore, the
4
,5
ted and appears to be linked with position of the nitro
6
group(s) relative to the phenolic moiety.
The tropospheric oxidation of aromatic hydrocarbons (ben-
zene, toluene, and isomers of xylene) results in the formation
of phenols or cresols as well as ring fragmentation products.
7
,8
Phenols are known to be formed directly in the atmosphere by
the OH-initiated atmospheric oxidation of aromatic hydrocar-
8
–10
bons and product studies
show yields of about 25%.
Many open questions exist with regards to the role and the
fate of the phenolic compounds in the tropospheric aqueous
phase, i.e. the droplets of clouds, fog and rain and the aqueous
aerosol. Phenols may be converted by OH/NO or NO /NO
2
2
3
into nitrophenols in both the gas and the aqueous phase.
Nitrophenols have been found in all of the tropospheric com-
1
1–13
14
fog and high concentrations of
partments, like cloud,
nitrated phenols have been found in precipitation events.
1
5–18
1
9
Annual series have recently become available.
In the present study the reactions between phenol and OH/
NO and NO /NO were studied by a laser flash photolysis/
2
3
2
long path laser absorption (LFP-LPLA) experiment combined
with product studies performed with a HPLC-DAD-ED appa-
ratus (Fig. 1).
Fig. 1 Experimental set-up in the LFP-LPLA/HPLC-DAD-ED
experiment.
2
. Experimental methods
l ¼ 351 nm (Xe/F gas fill, W ¼ 400 mJ) with pulse length
of 10–40 ns. The optical detection system consists of a
helium–neon laser (Mod. 35-2, Spindler and Hoyer) operated
at 632 nm the light of which was passed into a combination
of two dielectrically coated mirrors in White configuration
and folded 32 times through the reaction cell, resulting in a
Photolysis experiment
The flash photolysis set-up consists of a multigas excimer laser
(
2
0
Mod Compex 201, Lambda-Physik) operated at two different
wavelength: l ¼ 248 nm (Kr/F gas fill, W ¼ 700 mJ) and
DOI: 10.1039/b201652d
Phys. Chem. Chem. Phys., 2002, 4, 3669–3675
This journal is # The Owner Societies 2002
3669

View Article Online
total optical pathlength of d ¼ 192 cm. In order to block the
excimer laser radiation a cut-off filter (Schott OG 550) was
placed before the photodiode detector. The electrical output
of the photodiode was amplified and fed to a digital storage
oscilloscope (Mod. Delta Classic, Gould) connected to a com-
puter. The timing of the experiment was controlled by a delay-
pulse generator (Mod. DG535, SRS).
absorption maxima In addition, the identity of the products
was confirmed by independent LC-MS analysis.
The electrochemical detector was operated by applying a
fixed DC potential to the working electrode of a flow-through
electrolysis cell and recording the current resulting from the
oxidation of analyte as a function of elution time.
The electrochemical detector was used in screen-out mode,
setting the first electrode and the second electrode at 0.80
and 1.15 V, respectively.
The thermostated reaction cell used in the present work con-
sisted of a glass cylinder with an inner diameter of 50 mm and
an inner length of 60 mm with a total volume of 118 ml. The
front windows of the cell are plane plates made of fused silica.
All product studies were carried out in triplicate under vary-
ing experimental conditions (pH, low concentration of nitro-
gen dioxide, without oxygen, in the presence of a radical
scavenger and with different number of pulses) in order to dis-
criminate the different possible pathways.
2
1
More details on this set-up can be found elsewhere.
2
2
Solid-phase extraction
Phenols are very reactive towards various nitrating agents.
Hence, it was checked and found that the system is completely
ineffective in the nitration of phenol in the absence of a laser
pulse or reaction initiator.
Samples were extracted with LiChrolut ENV cartridges
Merck). Conditioning was performed with 3 ml of methanol
(
23
ꢀ
1
and 3 ml of water at a flow-rate of 2 ml min . A variable sam-
ple volume (12–20 ml) was applied at a flow-rate of 1 ml
The formation of nitrate radical in the flash photolysis
experiment of nitrate anions (0.01 M, pH ¼ 0.5 with HClO
4
)
ꢀ1
min . After the preconcentration step the cartridges were
rinsed with 1 ml of water to remove matrix impurities. After-
wards the cartridges were dried by gentle suction. Then the
sample was eluted with 1 ml of acetonitrile at a flow-rate of
24
at 248 nm, occurs as follows:
NO3 þ hnðl ¼ 248 nmÞ þ Hþ ! OH þ NO2
ꢀ
ðR-1Þ
ðR-2Þ
ꢀ
1
OH þ HNO3 ! NO3 þ H2O
0
.5 ml min . Breakthrough volumes for this type of cartridges
were measured by percolating equivalent volume of standard
solution using the method described above. No breakthrough
was observed under the present experimental conditions and
the recovery rates are quantitative (see Table 1).
OH titration
The initial maximal OH-concentration was determined by
ꢀ
reaction of OH to give Cl
2
radical anions following photoly-
Chromatographic separation and detection
sis of nitrate anions (0.014 M) at pH ¼ 0.5 in the presence of
2
5
excess NaCl (0.13 M) at room temperature.
The separation of the compounds was achieved using a HPLC-
UV-ED apparatus configured with a gradient pump (Mod. L-
NO3 þ hnðl ¼ 248nmÞ þ Hþ ! OH þ NO2
ꢀ
ðR-1Þ
ðR-aÞ
ðR-bÞ
ðR-cÞ
6
200A, Merck-Hitachi), an autosampler (Mod AS-4000,
Merck-Hitachi), diode array detector (Mod. L-4500,
Merck-Hitachi) an electrochemical detector (Mod. Coulochem
100A, ESA) equipped with an ultra-analytical cell Mod. 5011
ESA) and an interface module Mod. D-6000 connected to a
ꢀ
ꢀ
OH þ Cl 0 HOCl
a
ꢀ
þ
HOCl þ H 0 Cl þ H2O
5
(
Cl þ Clꢀ 0 Cl2ꢀ
computer. A stainless steel column (250 ꢁ 4 mm I.D.) packed
The flash photolysis set-up consists of a multigas excimer laser
with a regular spherical silica Kromasil (Hyperchrome) with
˚
(
(
Mod LPX 100, Lambda-Physik) operated at l ¼ 248 nm
particle size of 5 mm and porosity of 100 A derivatizated with
octadecyl chain was used as stationary phase. The analysis was
Kr/F gas mixture) with an average energy of about 450 mJ.
The optical detection system consists of a Hg–Xe lamp (150
W) as the analytical light source, the light of which was led into
a combination of two dielectrically coated mirrors in White
performed by an isocratic elution with a mobile phase com-
posed of 70% KH
2
PO
4
25 mM (pH ¼ 4 adjusted with
ꢀ
1
.
H
3
PO ) and 30% CH
4
3
CN at a flow-rate of 0.7 ml min
20
configuration and folded 12 times through the thermostated
The analytes were identified by direct comparison of retention
time and absorption spectrum of the standard compound.
Quantification was performed using calibration curves at dif-
ferent wavelengths where the compounds show relative
reaction cell (V ¼ 122.5 ml), giving a total optical pathlength
of d ¼ 84 cm. Then the light was collected in a monochroma-
tor/photomultiplier system (Zeiss PMQ/PMT 1P 28) with the
wavelength fixed at l ¼ 365 nm. The electrical output was
amplified and fed to a digital storage oscilloscope (Mod.
4
050, Gould) connected to a computer.
Table 1 Solid phase extraction recoveries and correlation coefficient
for the considered compounds
SPE
recovery (%)
UV/DAD calibration
curve corr. coeff.
Nitrite anions analysis
Compound
Phenol
Samples were flashed with different numbers of laser pulse
(1000, 750, 500, 250, 100, 20, 15, 10, 4) under the reaction con-
ditions. The concentration of nitrite anions was measured by
means of ion chromatography (anion column–Super Sep,
Waters) with conductivity detection (Metrohm System 690).
0.99 ꢂ 4
89.9 ꢂ 7
0.91 ꢂ 6
98.2 ꢂ 6
99.1 ꢂ 4
89.3 ꢂ 5
96.1 ꢂ 7
99.2 ꢂ 4
97.5 ꢂ 5
91.1 ꢂ 3
93 ꢂ 6
0.993
0.998
0.989
0.999
0.997
0.997
0.995
0.991
0.997
0.988
0.988
0.997
2
3
4
2
2
1
1
1
1
1
4
-Nitrophenol
-Nitrophenol
-Nitrophenol
,4-Dinitrophenol
,6-Dinitrophenol
,2-Dihydroxybenzene
,3-Dihydroxybenzene
,4-Dihydroxybenzene
,2,3-Trihydroxybenzene
,3,5-Trihydroxybenzene
-Nitrocathecol
Reagents
All the reagents were purchased from Fluka–Sigma–Aldrich
and were p.a. grade (>99%). All solutions were freshly pre-
pared for each experiment using MilliQ (Millipore) grade
water.
92.2 ꢂ 8
3
670
Phys. Chem. Chem. Phys., 2002, 4, 3669–3675

View Article Online
2
6,27
3
. Results and discussion
products.
matography analysis that nitrite anions were accumulating in
It was experimentally determined by ion chro-
2
8
The objective of the present study is to clarify the aqueous
phase product yields of the reaction of NO with phenol. To
this end, NO was produced under only slightly oxidizing con-
3
the reactions solutions (Table 2). It has been suggested that,
in acidic aqueous conditions in the presence of nitrous acid,
nitration of phenol occurs and results in the formation of
mono-nitrophenols (o- and p-nitrophenol). In particular it
was observed that the main product of the reaction was the
para isomer and a nitrosation/oxidation mechanism was sug-
gested where in the first step a nitrosation in the para position
occurs, followed by oxygen mediated oxidation leading to the
formation of 4-nitrophenol.
For this reason experiments were also performed using a sin-
gle laser pulse and the results of these experiments (Table 3)
2 3
can be better used to describe the OH/NO /NO system.
A scheme (Fig. 2) summarising the data of Table 3 could be
developed: The obtained results show that the agreement
between initial radical yields and product concentrations is
generally reasonable.
In the present experimental conditions the two radical spe-
cies involved in the first step of the oxidation process are
OH (E–G) and NO (E, F). When the oxidation processes (E–
3
ditions by the sequence of reactions (R-1) and (R-2) at
pH ¼ 0.5. In this system, due to the nature of nitrate anion
photolysis, OH and NO are initially present.
2
Only at low pH, when considerable amounts of HNO3(aq)
exist, is the NO radical formed and directly observed by time-
3
resolved laser absorption. Both radical concentration measure-
ment as well as product studies were used to characterise the
oxidation reaction. Hence, a direct comparison of concentra-
tion of the nitrate radical produced with the amount of pro-
ducts directly sampled from the reaction cell becomes possible.
The radicals formed in (R-1) and (R-2) react with phenol
5
(
1 ꢁ 10ꢀ M) leading to the product yields of nitro derivatives
as a function of the number of pulses summarised in Table 2.
It is interesting to note that in the reported series of experi-
ments (A–E) the o/p ratio changes with decreasing number of
pulses (Table 2).
3
This appears to be due to secondary processes which could
include (i) photolysis of primary products as well as (ii) reac-
tions of primary products with the abundant radicals OH,
G) occurs the nitration derivatives (2, 3) are the main products
of the overall reaction.
The low concentration of (4) in the experiments E and F,
NO
by-products like nitrite anions and N
2
and NO
3
and (iii) reactions involving nitrate photolysis
.
independently of O concentration, can be explained because,
2
2
in presence of considerable concentration of NO , it is known
2
O
5
The direct photolysis of the primary products, in the present
reaction conditions, could partially influence the reaction
yields and it was found that the isomer o-nitrophenol
from gas-phase studies that the reaction of hydroxycyclohexa-
dienyl radicals (HCHD) with NO is some orders of magnitude
2
2
9
faster than the addition of O2 and the same has been sug-
gested also for the reaction in the aqueous phase by von Sonn-
ꢀ
1
ꢀ1
(
excimer light more efficiently than p-nitrophenol
e
248 nm,298 K ¼ 2000 M
cm ) is photolysed by the laser
3
0
tag’s group.
Furthermore, it was also observed in pulse radiolysis and
ꢀ
1
ꢀ1
(
e
248 nm,298K ¼ 1800 M cm ) and a photolysis rate close
3
1
to 1% per pulse (E ¼ 630 mJ) for 2-nitrophenol was deter-
FT-IR smog chamber experiments that phenoxy radicals as
reactive intermediates almost totally react with nitrogen diox-
ide rather than with oxygen.
mined (Table 2).
Furthermore, it was observed that p-nitrophenol is in excess
concentration in experimental runs A–C, whereas the yield of
o-nitrophenol is nearly constant.
Flashing several times into a non–flowing nitrate anions
solution leads to the accumulation of the nitrate ion photolysis
The interaction of the nitrate radical with phenol in the pre-
sence of low concentrations of aqueous nitrogen dioxide (I)
was also studied in order to evaluate the possible contribution
to the formation of nitro compounds by the nitrate radical
Table 2 Initial maximal radical concentration and the identified reaction products in the flash photolysis experiment
[
OH]max/nM ¼
a
b
2ꢀ
[NO ] /nM
[
NO
2
]
max/nM
[NO
3
]
max/nM
Number
Exp. of pulses 1 pulse
Sum of OH
products /nM converted (%) o/p ratio
c
Total
1 pulse Total 1 pulse Total Prod. conc./nM
A
B
C
D
E
20
15
10
4
41
53
820
795
590
208
160
18
19
18
12
47
360
285
180
48
14
19
21
18
56
280
285
210
72
2: 89 (94) ꢂ 12; 3: 520 ꢂ 48;
2: 100 (105) ꢂ 14; 3: 490 ꢂ 7;
608 ꢂ 25
590 ꢂ 10
74 ꢂ 4
74 ꢂ 2
103 ꢂ 5
90 ꢂ 1
85 ꢂ 7
0.17
0.21
0.21
1.1
59
52
2: 110 (114) ꢂ 22; 3: 500 ꢂ 37; 610 ꢂ 27
2: 97 (99) ꢂ 2; 3: 90 ꢂ 2;
187 ꢂ 2
1
160
47
56
2: 67 ꢂ 8; 3: 59 ꢂ 9; 4: 10 ꢂ 10; 136 ꢂ 9
1.1
a
b
Calculated after OH yield measurement. Direct time-resolved NO
parenthesis are reported the yield of formation corrected for rate of photolysis.
ꢀ1
ꢀ1 47 c
In
3
absorption measurement (e(NO3,635 nm) ¼ 1240 ꢂ 40 l mol cm ).
Table 3 Initial maximal radical concentration and the identified reaction products in the single flash photolysis experiment
[
[NO
OH]max/nM ¼
a
b
Exp. pH
O
2
2
]/nM
[NO
3
]
max/nM
Prod. conc./nM
Sum. of prod./nM o/p ratio Remarks
E
F
G
H
I
0.5 Yes 160
0.5 No 160
3.0 Yes 160
47 ꢂ 6
2: 67 ꢂ 8; 3: 59 ꢂ 8; 4: 10 ꢂ 10;
136 ꢂ 9
1.1 ꢂ 0.1
1.3 ꢂ 0.1
1.6 ꢂ 0.3
—
OH/NO
OH/NO
OH/NO
2
2
2
/NO
/NO
3
66 ꢂ 12
Trace
—
2: 71 ꢂ 13; 3: 57 ꢂ 10; 4: 10 ꢂ 8; 138 ꢂ 11
3
2: 29 ꢂ 13; 3: 20 ꢂ 13; 4: trace;
—
2: trace; 3: trace; 4: n.d.;
59 ꢂ 13
3.2 Yes only NO
3.8 Yes 20
2
—
NO
NO
2
3
124 ꢂ 15
—
Remarks: E–G: nitrate anions photolysis at l ¼ 248 nm; H: nitrate anions photolysis at l ¼ 248 nm in presence of radical scavenger; I: peroxo-
disulfate anions photolysis at l ¼ 351 nm.
a
b
Calculated after OH yield measurement. Direct time-resolved NO
absorption measurement (e(NO3, 635 nm) ¼ 1240 ꢂ 40 Mꢀ1 cm ).
ꢀ1 47
3
Phys. Chem. Chem. Phys., 2002, 4, 3669–3675
3671

View Article Online
may occur via addition/elimination (R-3), H-abstraction of
the phenolic H-atom (R-4) or electron transfer (R-5) (Fig.
32
). In all three cases, intermediates are converted into nitro-
3
phenols by an addition of one molecule of nitrogen dioxide in
a second reaction step.
Furthermore, the possible contribution of a direct nitration
2
by NO was investigated (H). It was verified by addition of a
ꢀ4
OH-radical scavenger (tert-butanol, 5.5 ꢁ 10 M) that a pos-
sible direct reaction between undissociated phenol and aqu-
eous NO
amounts.
2
did not yield nitrated compounds in detectable
In Fig. 3 an overview of the possible products formation
pathways occurring under the conditions of this study is given.
From gas-phase studies, it is known that phenol reacts with
NO (R-3, R-3a) leading to the formation mainly of 2-nitro-
3
9 22
phenol (2). In a related work on phenol nitration by means
of a combined action of ozonized oxygen and nitrogen dioxide
(
Kyodai nitration) in different nonprotic solvents under nona-
cidic conditions the formation of a charge-transfer-complex
CTC) in the first step of the reaction has been suggested.
Fig. 2 Summarising scheme comparing the radical yields and the
product yields with the possible reaction pathways.
(
The CTC recombines to give four different NO -radical-
3
adducts which are converted into nitrophenols by an addition
of a molecule of NO and a concerted elimination of a mole-
2
itself. In this case the nitrate radical was produced by photoly-
sis of peroxodisulfate anions (0.03 M) at l ¼ 351 nm in the
presence of large excess of nitrate anions (0.1 M):
3
cule of HNO .
2
In a recent kinetic study it was found that the direct forma-
4
tion of (5) via an H-abstraction pathway by nitrate radical (R-
4
) could contribute about one half of the measured rate con-
S2O8 þ hnðl ¼ 351 nmÞ ! 2SO4ꢀ
2
ꢀ
ðR-dÞ
ðR-eÞ
9
ꢀ1 ꢀ1
s at 298 K). The present
product study and a former work on radiolysis of aqueous
stant (k2nd ¼ 1.8 ꢂ 0.3 ꢁ 10 M
SO4ꢀ þ NO3ꢀ ! NO3 þ SO42ꢀ
3
3
phenol suggest that the oxidation process is strongly pH-
dependent. In very acidic conditions (pH ¼ 0.5), independent
of which mechanism is involved in the first reaction step, the
key intermediate is the phenoxyl radical (Fig. 4).
In these experiments no formation of nitro derivatives was
observed and it can clearly be concluded that nitrate radicals
are involved only in the initial attack towards phenol. This
Fig. 3 Reactions involved in phenol oxidation and nitration in the flash photolysis experiment.
Phys. Chem. Chem. Phys., 2002, 4, 3669–3675
3
672

View Article Online
Fig. 4 Proposed oxidation mechanism of phenol under strongly acidic aqueous conditions.
The formed phenoxyl radical could undergo nitration by an
addition of one molecule of nitrogen dioxide forming a keto-
intermediate (8a, b) that by rapid proton transfers with water
ging from 6 to 17% of the total amount of the identified pro-
ducts.
The long exposure times in lamp-photolysis experiments (at
least 1 h) could promote secondary reactions and explain the
formation of products deriving from further oxidation, i.e.
the formation of 4-nitrocatechol and 2-hydroxy-p-benzoqui-
none was also observed among the products (Table 4).
Apparently, the mono-nitrophenols are ‘‘early’’ products in
flash photolysis experiments like in the present study.
Furthermore, higher pHs may enhance the formation of sec-
ondary oxidation products. At higher pH conditions the OH-
3
4
gives (2) and (3) respectively.
The o/p ratio obtained under acidic conditions (E, F), where
the ratio is close to unity (o/p ¼ 1.1), is a further indication
that the oxidation process is linked to the formation of phe-
noxyl radical. From the possible resonance forms (5a, b, c) a
theoretical o/p ratio of 2 could be expected. However, it is
observed experimentally in the self-recombination of phenoxyl
3
5
radicals that the reactivity of the ortho-mesomeric forms (5a,
c) was lower than the reactivity of the p-mesomer. Hence, a
lower o/p ratio results which is in agreement with the findings
obtained here.
radical adduct could react more effectively with O
ing to dihydroxy-derivatives, and in the presence of NO
2
(R-9), lead-
(R-
2
8a, b) by nitration or further reaction with molecular oxygen
to yield quinones.
In earlier work on phenol oxidation by means broad-band
CW-lamp photolysis of nitrate anions
3
6,37
it was observed that
In acidified aqueous solution the dihydroxycyclohexadienyl
radical formed after OH-addition (R-6) undergoes a fast unim-
olecular acid–base catalysed elimination of water, with a first
the main products of the reaction are the OH-oxidation pro-
ducts (Table 4) where the catechol (4) is the most abundant
compound and the nitrated compounds (o- and p-nitrophenol
and p-nitrocatechol) were only minor reaction products, ran-
6
ꢀ1
38
order rate constant of 1.6 ꢂ 0.6 ꢁ 10
s
yielding the more resonance-stabilised phenoxyl radical.
(pH ¼ 1.3–1.9)
Table 4 Product study results in phenol oxidation experiments by means of nitrate anion photolysis with lamp and laser technique. The values are
expressed like percentage of reacted phenol
Irradiation
time
Identified products/
phenol reacted
pH
a 37
a 37
a 37
Niessen et al.
Niessen et al.
Niessen et al.
1 h
1 h
5.2 31
5.2 32
4
2
13
9.8
7
—
—
—
2.2
—
—
—
0.6
4.8
—
—
—
12
6.4
3
72
4.6
3.4
9.6
—
—
2.8
3.4
14.7
—
3.4
<0.3
—
4.8 64
6.6 44
1.8 83
1 h
1 h
3.6 14.1
5.5 51.6
0.5 3.7
3.0 Trace
—
—
—
—
11.3
3.3
a 36
Machado et al.
This work
This work
b
20 ns
20 ns
—
—
21.6
14.7
24.5 49.8 (86.4 )
b
10.1 24.9 (80.7 )
—
a
b
Sets of results for other experimental conditions are available in the referenced paper. Considering the contribution of direct photolysis of phe-
nol (e ¼ 200 Mꢀ cm at 248 nm) assuming F ¼ 1.
1
ꢀ1
Phys. Chem. Chem. Phys., 2002, 4, 3669–3675
3673

View Article Online
2
Fig. 6 Evolution of the model system for the reaction OH/NO /
NO
K.
3
+ phenol (E, Table 3) under acidic conditions (pH ¼ 0.5) at 298
Fig. 5 Summary of the reaction intermediates involved in the OH-
addition mechanism.
4. Summary and conclusion
In this work a laser photolysis/long-path laser absorption
apparatus (LFP-LPLA) was coupled to an analytic HPLC-
DAD-ED-based technique to study the phenol oxidation pro-
ducts in the presence of (i) OH/NO and (ii) NO /NO .
It was shown that in acidic conditions phenol undergoes
nitration and the main products of the reaction are the two
expected mono-nitrophenols (o- and p-nitrophenol). It appears
that the more important intermediate is the phenoxyl radical
which further reacts with a molecule of nitrogen dioxide form-
In the experiment at pH ¼ 3 (Table 3, G) the yields of for-
mation of nitrophenols are similar for lamp and laser experi-
ments and a o/p ratio shifted to a value close to 2 was
obtained, suggesting that the reaction proceeds via a different
mechanism.
2
3
2
2
When only OH and NO are present, two mechanisms might
be involved in the nitration processes, (i) an addition–addi-
tion–elimination mechanism (‘ADD-ADD-EL’, R-6, R-8)
and (ii) an H-abstraction reaction (H-ABS) from the phenolic
group. In the first case the OH radical addition could give four
different regioisomeric forms (Fig. 5).
ing the nitrophenolic compounds. It is shown that NO is only
3
involved in the first step of the reaction.
The present study together with kinetic
studies on tropospheric multiphase systems indicates that
nitration of phenolic compounds is effectively performed in a
two-step mechanism in tropospheric aqueous systems. When
2
4,44
and modelling
4
5
The electrophilic behaviour of the OH radical leads to a pre-
ferential attack in ortho and para positions (7, 7a) with respect
3
9
the phenolic moiety, forming a dihydroxycyclohexadienyl
radical (R-6) and a further addition of a NO molecule in
initiated by radicals such as OH at daytime and NO during
3
2
the night intermediates (substituted hydroxy-cyclohexadienyl
and/or phenoxyl radicals) are formed which could stabilize
by reaction with aqueous NO2 to yield nitrophenols. These
processes might be competitive with the gas-phase production
of nitrophenolic compounds and possibly even to their direct
the ortho position with respect the OH having entered giving
m-nitrophenol. However, in all the experiment carried out in
the present work the presence of m-nitrophenol in detectable
amounts could not be confirmed.
4
6
Regarding the H-abstraction mechanism by OH, leading to
phenoxyl radical formation, it was theoretically calculated for
a phenolic model compound that this pathway could contri-
emission from sources.
As stated earlier, nitrophenols are potentially harmful and
have been identified in all phases and it has been suggested that
some compounds are actually predominantly formed in solu-
4
0
bute only about 5%.
Modelling calculations (Table 5 and Fig. 6) performed with
1
1
tion-phase processes. In view of the existing literature and
the present study most of the identified load of 4- nitrophenol
is actually formed in such processes whereas gas-phase nitra-
4
1–43
Gepasi 3.21
products study carried out at higher pH (pH ¼ 3) where
NO is not formed anymore, because of the lower concentra-
and applied to the reaction conditions and
9
3
tion of phenol will predominantly yield 2-nitrophenol.
tion of the undissociated nitric acid, indicate that the forma-
tion of nitro products by a sequential reaction of OH and
NO (R-6, R-8) could account for about 25% of the observed
Tropospheric nitrophenol formation therefore is a good
example of the possible effects of sequential multiphase pro-
cesses involving not only gas-phase processes but phase trans-
fer and solution phase conversions as well.
2
nitro compounds.
Table 5 Reactions included in the modelling calculations for the reaction at pH ¼ 0.5 and T ¼ 298 K (E)
ꢀ
1
ꢀ1
Reaction
Rate constant k/L mol
s
Reference
7
48
Jiang et al., 1992
Wagner et al., 1980
OH + HNO
OH + NO
3
! NO
3
+ H
2
O
5.3 ꢁ 10
10
27
2
! HOONO
1.2 ꢁ 10
+
2ꢀ
3ꢀ
7
49
2
NO2 (aq) + H
2
O ! 2 H + NO + NO
7 ꢁ 10
Schwartz et al., 1984
50
Adams et al., 1965
1
9
9
0
OH + phenol ! OH–Products
OH–Products + NO
! Products
NO
+ phenol ! Products
1.8 ꢁ 10
8.2 ꢁ 10
1.8 ꢁ 10
2
Estimated
Umschlag et al., 2002
24
3
3
674
Phys. Chem. Chem. Phys., 2002, 4, 3669–3675

View Article Online
2
6
J. Mack and J. R. Bolton, J. Photochem. Photobiol. A: Chem.,
1
References
999, 128, 1.
1
E. Sawicki, in Air Pollution and Cancer in Man, ed. U. Mohr, D.
Schm a¨ hl and L. Tomatis, IARC Scientific Publication, Lyon,
977, vol. 16, p. 127.
J. Tsai, H. B. H. Peng, D. Z. Lee and C. C. Lee, Environ. Int, 1995,
1, 47.
I. B. Escher, M. Snozzi and R. P. Schwarzenbach, J. Am. Chem.
Soc., 1996, 118, 3071.
M. Natangelo, S. Mangiapan, R. Bagnati, E. Benfenati and R.
Fanelli, Chemosphere, 1999, 38, 1495.
G. Rippen, E. Zietz, T. Frank, T. Knacker and W. Kl o¨ pffer,
Environ. Technol. Lett., 1987, 8, 475.
P. Gramatica, V. Consonni, A. Santagostino, P. Galvani and E.
Bolzacchini, Air Pollution Research Report, 1999, 67, EUR
27 I. Wagner, H. Strelow and G. Busse, Z. Phys. Chem., 1980, 123, 1.
28 U. Al-Obaidi and R. B. Moodie, J. Chem. Soc., Perkin Trans. 2,
1985, 467.
29 R. Knispel, R. Koch, M. Siere and C. Zetzsch, Ber. Bunsen-Ges.
Phys. Chem., 1990, 94, 1375.
30 G. Merga, H.-P. Schuchmann, B. S. Madhava Rao and C. von
Sonntag, J. Chem. Soc., Perkin Trans. 2, 1996, 1097, and refer-
ences therein.
31 J. Platz, O. J. Nielsen, T. J. Wallington, J. C. Ball, M. D. Hurley,
A. M. Straccia, W. F. Schneider and J. Sehested, J. Phys. Chem.
A, 1998, 102, 7964.
32 H. Herrmann and R. Zellner, in N-Centered Radicals, ed. Z. B.
Alfassi, John Wiley & Sons, New York, 1998, p. 29.
33 Z. B. Alfassi, in General Aspects of the Chemistry of Radicals, ed.
Z. B. Alfassi, John Wiley & Sons, New York, 1999, p. 166.
34 E. J. Land and M. Ebert, Trans. Faraday Soc., 1967, 63, 1181.
35 R. Louw and D. Santoro, Environ. Sci. Technol., 1999, 33, 3281.
36 F. Machado and P. Boule, J. Photochem. Photobiol. A: Chem.,
1995, 86, 73.
37 R. Niessen, D. Lenoir and P. Boule, Chemosphere, 1988, 17, 1977.
38 M. Roder, L. Wojnarovits and G. Foldiak, Magy. Kem. Foly.,
1998, 104, 354.
39 N. V. Raghavan and S. J. Steenken, J. Am. Chem. Soc., 1980, 102,
3495.
1
2
3
4
5
6
2
18765, 295.
7
8
9
B. Klotz, I. Barnes, K. H. Becker and B. T. Golding, J. Chem.
Soc., Faraday Trans., 1997, 93, 1507.
T. Berndt, O. B o¨ ge and H. Herrmann, Chem. Phys. Lett., 1999,
3
14, 435.
R. Atkinson, S. M. Aschmann and J. Arey, Environ. Sci. Technol.,
992, 26, 1397.
1
1
1
0
1
R. Atkinson, Atmos. Environ., 2000, 34, 2063.
J. L u¨ ttke, V. Scheer, K. Levsen, G. W u¨ nsch, J. N. Cape, K. J.
Hargreaves, R. L. Storeton–West, K. Acker, W. Wieprecht and
B. Jones, Atmos. Environ., 1997, 31, 2637.
1
1
2
3
J. L u¨ ttke and K. Levsen, Atmos. Environ., 1997, 31, 2649.
J. L u¨ ttke, K. Levsen, K. Acker, W. Wieprecht andM o¨ ller, Int. J.
Environ. Anal. Chem., 1999, 74, 69.
H. Richartz, A. Reischl, F. Trautner and O. Hutzinger, Atmos.
Environ., 1990, 24, 3067.
C. Leuenberger, M. P. Ligocki and J. F. Pankow, Environ. Sci.
Technol., 1985, 19, 1053.
C. Leuenberger, J. Czuczwa, J. Tremp and W. Giger, Chemo-
sphere, 1988, 17, 511.
40 M. J. Lundqvist and L. A. Eriksson, J. Phys. Chem. B, 2000, 104,
848.
41 P. Mendes, Comput. Appl. Biosci., 1993, 9, 563.
42 P. Mendes, Trends Biochem. Sci., 1997, 22, 361.
43 P. Mendes and D. B. Kell, Bioinformatics, 1998, 14, 869.
44 H. Herrmann, B. Ervens, H.-W. Jacobi, R. Wolke, P. Nowacki
and R. Zellner, J. Atmos. Chem., 2000, 36, 231.
45 H. Herrmann, M. Exner, H.-W. Jacobi, G. Raabe, A. Reese and
R. Zellner, Faraday Discuss., 1995, 100, 129.
14
15
16
1
1
7
8
K. Kawamura and I. R. Kaplan, Atmos. Environ., 1986, 20, 115.
K. Levsen, S. Behnert, B. Prieß, M. Svoboda, H.-D. Winkeler and
J. Zietlow, Chemosphere, 1990, 21, 1037.
W. Sch u¨ ssler and L. Nitschke, Chemosphere, 2001, 42, 277.
J. U. White, J. Opt. Soc. Am., 1946, 42, 285.
M. Exner, H. Herrmann and R. Zellner, Ber. Bunsen-Ges. Phys.
Chem., 1992, 96, 470.
B. Barletta, E. Bolzacchini, S. Meinardi, M. Orlandi and B. Rin-
done, Environ. Sci. Technol., 2000, 34, 2224.
M. Ouertani, P. Girard and H. B. Kagan, Tetrahedron Lett., 1982,
46 J. Tremp, P. Mattrel, S. Fingler and W. Giger, Water, Air Soil
Pollut., 1993, 68, 113.
47 K. Sehested, T. Løgager, J. Holcman and O. J. Nielsen, in Pro-
ceedings of EUROTRAC Symposium ’94, ed. P. M. Borrell, P.
Borrell, T. Cvita sˇ and W. Seiler, SPB Academic Publishing, The
Hague, 1994, p. 999.
48 P.-Y. Jiang, Y. Katsumura, K. Ishigura and Y. Yoshida, Inorg.
Chem., 1992, 31, 5135.
2 2
49 S. E. Schwartz and J. I. Teasley, in SO and NO Oxidation
19
20
21
2
2
2
2
2
3
4
5
Mechanisms: Atmospheric Considerations, Acid Precipitation Ser-
ies, ed. J. G. Calvert, Butterworth, Stoneham, MA, 1984, vol. 3,
p. 173.
23, 4315.
Th. Umschlag, R. Zellner and H. Herrmann, Phys. Chem. Chem.
Phys., 2002, 4, 2975.
H.-W. Jacobi, H. Herrmann and R. Zellner, Ber. Bunsen-Ges.
Phys. Chem., 1997, 101, 1909.
50 G. E. Adams, J. W. Boag, J. Currant and B. D. Michael, Pulse
Radiolysis, ed. M. Ebert, J. P. Keene, A. J. Swallow and J. H.
Boxendale, Academic Press, New York, 1965, p. 131.
Phys. Chem. Chem. Phys., 2002, 4, 3669–3675
3675