OH-radical-induced oxidation of methanesulfinic acid. The reactions of the methanesulfonyl radical in the absence and presence of dioxygen

Article abstract of DOI:10.1039/b009631h

Hydroxyl radicals were generated radiolytically in N₂O-saturated solutions. As shown by pulse radiolysis, methanesulfinic acid reacts with^·OH (k = 5.3 × 10⁹ dm³ mol^-1 s^-1) and N₃^· giving rise to an intermediate which has an absorption maximum at 330 nm (Gε ≈ 5.2 × 10^-5 m² mol^-1). This is attributed to the methanesulfonyl radical, CH₃S(O)O^·. Pulse radiolysis with conductometric detection indicates that CH₃S(O)O^· is formed in only ～80% yield. This is confirmed by scavenging experiments with Fe(CN)₆^4-, ascorbate and sulfite, which are oxidised by CH₃S(O)O^· with rate constants of ～2 × 10⁹ dm³ mol^-1 s^-1. Steady-state radiolysis of methanesulfinic acid shows that methanesulfonic acid is the major product (G = 2.1 × 10^-7 mol J^-1). Further products are sulfate (0.7 × 10^-7 mol J^-1), methane (0.3 × 10^-7 mol J^-1), ethane (0.08 × 10^-7 mol J^-1) and dimethyl sulfone (not quantified). It is suggested that an OH-adduct is formed initially which mainly eliminates OH^-, but also decomposes yielding ^·CH₃ and bisulfite. The formation of methanesulfonic acid can be explained by a disproportionation of the methanesulfonyl radicals via recombination and subsequent hydrolysis. In the presence of dioxygen, a chain reaction occurs whereby 2 mol methanesulfonic acid are formed per 1 mol dioxygen consumed. G(methanesulfonic acid) ≈ 250 × 10^-7 mol J^-1 was found to be independent of dose rate (0.011-0.165 Gy s^-1), methanesulfinic acid concentration [(0.1-4) × 10^-3 mol dm^-3] and dioxygen concentration. An efficient chain process was also observed upon electron beam irradiation [G(methanesulfonic acid) ≈ 200 × 10^-7 mol J^-1 at 0.8 Gy per pulse and 75 × 10^-7 mol J^-1 at 9 Gy per pulse (pulse duration 2 μs)]. It is proposed that the oxidation of the substrate by the methylsulfonylperoxyl radical, CH₃S(O₂)OO^·, to give the strongly oxidising CH₃S(O₂)O^· radical, initiates the chain reaction, with the latter propagating the chain by reacting with a substrate molecule to give methanesulfonic acid and the methanesulfonyl radical. Branching and partial removal of the chain-carrying CH₃S(O₂)O^· radicals by H-abstraction from the substrate is suggested as the likely path leading to chain termination. Oxidation of methanesulfinate by ozone (k = 2 × 10⁶ dm³ mol^-1 s^-1) occurs only by O-atom transfer, and an electron transfer that would start a chain reaction was not observed.

Full text of DOI:10.1039/b009631h

View Article Online / Journal Homepage / Table of Contents for this issue
OH-Radical-induced oxidation of methanesulfinic acid.
The reactions of the methanesulfonyl radical in the absence
and presence of dioxygen
a,b
b
a
b,c
2
Roman Flyunt,† Oksana Makogon,† Man Nien Schuchmann, Klaus-Dieter Asmus
a
and Clemens von Sonntag*
a
Max-Planck-Institut für Strahlenchemie, Stiftstrasse 34–36, P.O. Box 101365.
D-45470-Mülheim an der Ruhr, Germany. E-mail: vonsonntag@mpi-muelheim.mpg.de;
Tel: ϩϩ 49-208-306-3529; Fax: ϩϩ49-208-306-3951
Radiation Laboratory, University of Notre Dame, Notre Dame, IN 46556, USA
Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame,
IN 46556, USA
b
c
Received (in Cambridge, UK) 30th November 2000, Accepted 27th February 2001
First published as an Advance Article on the web 30th March 2001
Hydroxyl radicals were generated radiolytically in N O-saturated solutions. As shown by pulse radiolysis,
2
ؒ
9
3
Ϫ1 Ϫ1
ؒ
methanesulfinic acid reacts with OH (k = 5.3 × 10 dm mol s ) and N giving rise to an intermediate which
has an absorption maximum at 330 nm (Gε ≈ 5.2 × 10 m mol ). This is attributed to the methanesulfonyl radical,
CH S(O)O . Pulse radiolysis with conductometric detection indicates that CH S(O)O is formed in only ~80%
3
Ϫ5
2
Ϫ1
ؒ
ؒ
3
3
4Ϫ
6
yield. This is confirmed by scavenging experiments with Fe(CN) , ascorbate and sulfite, which are oxidised by
ؒ
9
3
Ϫ1 Ϫ1
CH S(O)O with rate constants of ~2 × 10 dm mol s . Steady-state radiolysis of methanesulfinic acid shows that
3
Ϫ7
Ϫ1
Ϫ7
Ϫ1
methanesulfonic acid is the major product (G = 2.1 × 10 mol J ). Further products are sulfate (0.7 × 10 mol J ),
methane (0.3 × 10 mol J ), ethane (0.08 × 10 mol J ) and dimethyl sulfone (not quantified). It is suggested that
an OH-adduct is formed initially which mainly eliminates OH , but also decomposes yielding CH and bisulfite.
Ϫ7
Ϫ1
Ϫ7
Ϫ1
Ϫ
ؒ
3
The formation of methanesulfonic acid can be explained by a disproportionation of the methanesulfonyl radicals
via recombination and subsequent hydrolysis. In the presence of dioxygen, a chain reaction occurs whereby 2 mol
Ϫ7
Ϫ1
methanesulfonic acid are formed per 1 mol dioxygen consumed. G(methanesulfonic acid) ≈ 250 × 10 mol J was
Ϫ1
Ϫ3
found to be independent of dose rate (0.011–0.165 Gy s ), methanesulfinic acid concentration [(0.1–4) × 10 mol
Ϫ3
dm ] and dioxygen concentration. An efficient chain process was also observed upon electron beam irradiation
G(methanesulfonic acid) ≈ 200 × 10 mol J at 0.8 Gy per pulse and 75 × 10 mol J at 9 Gy per pulse
Ϫ7
Ϫ1
Ϫ7
Ϫ1
[
(
pulse duration 2 µs)]. It is proposed that the oxidation of the substrate by the methylsulfonylperoxyl radical,
ؒ
ؒ
CH S(O )OO , to give the strongly oxidising CH S(O )O radical, initiates the chain reaction, with the latter
3
2
3
2
propagating the chain by reacting with a substrate molecule to give methanesulfonic acid and the methanesulfonyl
ؒ
radical. Branching and partial removal of the chain-carrying CH S(O )O radicals by H-abstraction from the
3
2
substrate is suggested as the likely path leading to chain termination. Oxidation of methanesulfinate by ozone
6
3
Ϫ1 Ϫ1
(
k = 2 × 10 dm mol s ) occurs only by O-atom transfer, and an electron transfer that would start a chain
reaction was not observed.
The OH-radical-induced free-radical chemistry of sulfur-
has allowed us to study certain of its reactions in more detail
(cf. refs. 3–7).
containing compounds is very different from that observed with
ؒ
their carbon analogues. For example, OH reacts with acetone
In the present study we have extended our investigation of
1
ؒ
only slowly by H-abstraction, whereas its reaction with di-
OH radical reactions to the methanesulfinate ion whose
methyl sulfoxide is very fast, proceeding mainly by addition
carbon analogue, acetate, reacts by H-abstraction; electron
transfer from the carboxylate group is negligible (2.5%). Pulse
2
8
[reaction (1)]; only 7% H-abstraction [reaction (2)] occurs. The
adduct thus formed is very short-lived and rapidly decomposes
by β-fragmentation into methanesulfinic acid and a methyl
radical [reaction (3)]. In fact, this reaction sequence is the most
convenient source of methyl radical in aqueous solution and
radiolysis of methanesulfinic acid in the absence and presence
of dioxygen has already attracted some attention, and in
various aspects we can build upon this interesting study.
9
It will be shown that methanesulfinic acid behaves in many
ؒ
respects like dimethyl sulfoxide, i.e. the OH radical reacts
preferentially by addition and subsequent fragmentation,
giving rise to either methanesulfonyl or methyl radicals. In the
†
On leave from Institute of Physico-Chemistry, National Academy of
Science of the Ukraine, Naukova Street 3a, UA-79053 L’viv.
DOI: 10.1039/b009631h
J. Chem. Soc., Perkin Trans. 2, 2001, 787–792
This journal is © The Royal Society of Chemistry 2001
787

View Article Online
presence of dioxygen, the former initiates a rapid chain
reaction reminiscent of the OH-radical-induced autoxidation
10,11
of the hydrogen sulfite ion.
In contrast, the acetate ion does
not give rise to a chain reaction under otherwise identical
12
conditions.
Experimental
Pulse radiolysis of aqueous solutions of sodium methane-
sulfinate (97%, Lancaster; main impurity, ~3%, methane-
sulfonate as determined by HPIC) in Milli-Q-filtered
(
Millipore) water saturated either with N O or with
2
N O–O (4 : 1 v/v) was carried out with a 2.8 MeV Van de
2
2
Graaff accelerator delivering electron pulses of 0.4 µs duration.
For preparative pulse radiolysis 2 µs pulses were employed.
Intermediates were monitored by optical and conductometric
detection. The pulse radiolysis set-up has been described
Fig. 1 Comparison of the spectra (5 µs after the pulse) of radicals
produced in the pulse radiolysis (~5 Gy per pulse) of an N O-saturated
2
Ϫ3
3
solution of sodium methanesulfinate (10 mol dm ) at pH 7.2 in the
13
Ϫ2
Ϫ3
Ϫ2
3
previously. For dosimetry, N O-saturated 10 mol dm
absence (᭹) and presence of sodium azide (10 mol dm ) at pH 8.5
(᭝).
2
1
4,15
thiocyanate solution was used for optical detection,
Ϫ
Ϫ4
2
Ϫ1
15
taking Gε[(SCN)2 ] = 5.28 × 10 m J at 480 nm. For con-
᝽
Ϫ2
Ϫ3
Ϫ
ؒ
Ϫ
ductometric detection, N O-saturated 10 mol dm dimethyl
eaq ϩ N O ϩ H O → OH ϩ OH ϩ N
(5)
2
2
2
2
2
sulfoxide solution at pH 4 was used as dosimeter, taking
ϩ
Ϫ
Ϫ7
Ϫ1 16
G(H ) = G(CH SO ) = 5.8 × 10
mol J .
At elevated
ؒ
3
2
methanesulfinic acid. The rate constant of OH has been
ؒ
9
9
3
Ϫ1
Ϫ1
methanesulfinate concentrations, the increased OH-scavenging
capacity has been taken into account.
γ-Radiolyses were carried out in a panoramic Co-source at
dose rates varying between 0.011 and 0.165 Gy s . Methane-
determined to be (6 ± 1) × 10 dm mol
5
s
(our value:
.3 × 10 dm mol s , see below); that of H could not be
measured, but it may be as low as those reported for dimethyl
17
9
3
Ϫ1 Ϫ1
ؒ
6
0
Ϫ1
7
3
Ϫ1 Ϫ1 22
sulfoxide (≤10 dm mol s ).
sulfinate and methanesulfonate, as well as sulfate anions, were
separated by ion chromatography (Dionex DX-2010i, 3 × 250
mm AS4-14 column with a 3 × 50 mm AG-14 pre-column,
In the presence of O , e.g. in N O–O (4 : 1)-saturated solu-
2
2
2
ؒ
tions, the solvated electron is still scavenged by N O. While OH
does not react with O , H is readily scavenged [reaction (6)]. In
2
ؒ
2
Ϫ4
Ϫ3
ASRS-I suppressor). The eluent used was an 9 × 10 mol dm
Ϫ4
Ϫ3
ؒ
ؒ
2
Na CO –8.5 × 10 mol dm NaHCO solution at a flow rate
H ϩ O → HO
(6)
2
3
3
2
3
Ϫ1
of 1 cm min . Their retention times were 2.6, 3.6 and 23.5
min, respectively. Hydrogen peroxide cannot be determined
using the sensitive molybdate-activated iodide method,
because iodide plus H O induces the autoxidation of methane-
ؒ
neutral solution, the HO2 radical is largely dissociated, and
Ϫ
ؒ
2
23
18
O2 is the dominating species [pK (HO ) = 4.8].
᝽
a
2
2
sulfinic acid, and so we had to rely on the less sensitive assay
Reactions in the absence of dioxygen
19
with titanium(IV) sulfate. Methane and ethane in the irrad-
iated samples were purged with argon and measured by GC on
a 3.2 m Porapak-9 capillary column. Oxygen concentration in
aqueous solution of sodium methanesulfinate was determined
with an oxygen-sensitive Clark-electrode (TriOx EO 200,
Wissenschaftlich-Technische Werkstätten, Weilheim). Product
yields are considered to be accurate to within ±10%. Some of
the irradiated samples were rotary evaporated, from which less
volatile products were extracted with diethyl ether and analysed
by GC-MS on a Hewlett-Packard 5971A Mass Selective
Detector coupled with an HP 5890 Series II Gas Chromato-
graph. Dimethyl sulfone was thus identified by comparison of
its mass spectrum with that of authentic material.
In the pulse radiolysis of an N O-saturated solution of
methanesulfinate anion at pH 7, an intermediate is formed
which is characterised by an absorption maximum at 330 nm
2
ؒ
(
Fig. 1) and is assigned to the CH S(O)O radical (cf. also
3
Ϫ7
Ϫ1
ؒ
ref. 9). Assuming G(radicals) = 4.8 × 10 mol J (85% of OH,
see below) an absorption coefficient of 1080 ± 100 dm mol
cm at 330 nm is calculated. These spectral characteristics
agree well with those of the earlier study, where the absorption
maximum was reported at 325 nm with ε = 900 ± 100 dm
3
Ϫ1
Ϫ1
3
Ϫ1
Ϫ1
ؒ
9
mol cm , assuming 100% of OH.
2
If its reaction with sulfoxides [cf. reaction (1)] is a good
guide, one must assume that also in the present system OH
ؒ
mainly adds to methanesulfinate [reaction (7)] and that
Ozone stock solutions were prepared by bubbling ozone from
a dioxygen-fed ozonator (Philaqua Philoz 04, Gladbeck) for
some minutes through water, and its ozone content was deter-
H-abstraction at the methyl group probably plays a minor
ؒ
role [reaction (8)]. The OH-adduct may decompose by
Ϫ
2
0,21
3
β-fragmentation in two directions; either it eliminates OH
mined spectrophotometrically taking ε(260 nm)
= 3300 dm
ؒ
Ϫ1
Ϫ1
to give the CH
3
S(O)O radical [reaction (9)], or it may split
mol cm . In all experiments, the methanesulfinic acid was
in large excess over the added ozone concentration (methane-
sulfinic acid consumption <15%).
off a methyl radical, the other product being the bisulfite anion
reaction (10)]. Our data do not allow us to give rate constants
[
Results and discussion
The radical-generating system
Hydroxyl radicals are generated in the radiolysis of water
[reaction (4)]. The radiation-chemical yields (G values) of the
ionizing
Ϫ
ؒ
ؒ
ϩ
H O
2
eaq , OH, H , H , H O , H
(4)
2
2
2
radiation
ؒ
Ϫ
Ϫ7
Ϫ1
primary radicals are G( OH) ≈ G(e ) = 2.9 × 10 mol J ,
for these very fast reactions. Evidence will be presented below
to show that reaction (9) is in fact the major pathway, with
a minor contribution of ~10–20% from reaction (10). The
occurrence of latter pathway is supported by the formation of
aq
ؒ
Ϫ7
Ϫ1
Ϫ7
G(H ) = 0.6 × 10
mol J , and G(H O ) ≈ 0.7 × 10
J . N O is used to convert the solvated electron into OH
reaction (5)]. The OH radical and the H atom will react with
mol
2
2
Ϫ1
ؒ
2
ؒ
ؒ
[
7
88 J. Chem. Soc., Perkin Trans. 2, 2001, 787–792

View Article Online
Table 1 Compilation of rate constants measured in the present study
Table 2 Products and their G values of the γ-radiolysis of sodium
Ϫ3 Ϫ3
methanesulfinate (10 mol dm ) in N O-saturated solutions at pH 6.8
2
3
k/dm
mol
Ϫ1 Ϫ1
7
Ϫ1
Reaction
s
Products
10 G values/mol J
ؒ
Ϫ
2
9
9
9
9
OH ϩ CH SO → products
5.3 × 10
1.8 × 10
2.1 × 10
1.5 × 10
Methanesulfonate
Sulfate
Dimethyl sulfone
Methane
2.1
0.7
3
Ϫ
Ϫ
ؒ
2Ϫ
4
SO4 ϩ CH SO → CH S(O)O ϩ SO
᝽
3
2
3
ؒ
ؒ
4Ϫ
Ϫ
3Ϫ
CH S(O)O ϩ Fe(CN) → CH SO ϩ Fe(CN)
6
Detected, not quantified
0.3
0.08
3
6
3
2
Ϫ
CH S(O)O ϩ ascorbate → CH SO ϩ ascorbate
3
3
2
radical
3
Ethane
ؒ
2Ϫ
3
Ϫ
ؒ
Ϫ
9
9
CH S(O)O ϩ SO → CH SO ϩ SO
CH S(O)O → products
1.7 × 10
3
2
3
ؒ
2
1.8 × 10 (2k)
3
ؒ
Ϫ
anion (pK = 6.9) and the CH S(O)O radical have about the
a
2
same equivalent conductance as the methanesulfinate anion.
From the above G value of conductance decrease the decom-
position of the intermediate OH-adduct radical appears to take
place mainly via reaction (9) (70–80%) and to a lesser extent via
reactions (8) and (10) (20–30%).
sulfate and methane/ethane (see below). The latter products
have the methyl radical as precursor.
An identical spectrum is observed when the methanesulfinate
ؒ
anion reacts with N (the latter is produced by the reaction of
3
ؒ
Ϫ
3
ؒ
3
OH with N ) (cf. Fig. 1). The N radical is known generally
In basic solutions (pH 10–10.5), the conductance increase
to undergo electron transfer reactions. Its reaction with the
Ϫ
observed 20 µs after the pulse is equivalent to G(OH Ϫ
methanesulfinate anion is expected to yield quantitatively
Ϫ7
Ϫ1
ؒ
anion) = 5.8 × 10 mol J . This indicates that the intermediate
the CH S(O)O radical. However, in the γ-radiolysis of NaN
3
Ϫ2
3
Ϫ3
Ϫ3
OH-adduct radical decays very largely via reaction (9), since
(
2 × 10 mol dm ) and sodium methanesulfinate (1 × 10
mol dm ) in an N O-saturated solution, the yield of CH S(O)O
is identical to that for the OH-induced reaction, and there is
reason to assume that in this case the reaction of the N3 radical
Ϫ
Ϫ3
ؒ
reaction (10) does not release OH at this pH, but rather the
2
3
ؒ
formation of bisulfite anion actually leads to the consumption
Ϫ
Ϫ
ؒ
of OH . This relatively high G(OH ) to some extent conflicts
with the observations made at acidic pH.
might initially also proceed via addition to the methanesulfinate
ؒ
The CH S(O)O radical is strongly oxidising. Its reaction
3
anion [reaction (11)]. This short-lived adduct subsequently
4Ϫ
ؒ
with Fe(CN)
6
has been followed by absorption build-up of
Fe(CN)6 in the pulse radiolysis of an N O-saturated solution
4Ϫ
undergoes either azide elimination to give the CH S(O)O
3
3Ϫ
2
radical as the main product [reaction (12)], or β-fragmentation
Ϫ2
Ϫ3
Ϫ
of sodium methanesulfinate (10 mol dm ) and Fe(CN)
6
Ϫ4 Ϫ3 3Ϫ
to produce the methyl radical and N SO2 [reaction (13)]. This
3
(
10 mol dm ) at 420 nm, where Fe(CN) has an absorption
6
3 Ϫ1 Ϫ1
maximum (ε = 1020 dm mol cm ) [reaction (14)].
The rate of build-up at this wavelength was measured as a
4Ϫ
function of the Fe(CN)6 concentration, and from these data
Ϫ7
9
3
Ϫ1 Ϫ1
is supported by the finding that methane (G ~ 0.01 × 10 mol
(not shown) the rate constant of 2.1 × 10 dm mol
s
is
Ϫ1
Ϫ7
Ϫ1
J ) and ethane (G ~ 0.06 × 10 mol J ) are formed. The low
yield of methane can be explained by the fact that the H atom is
obtained. Also, this system presents a further approach to
ؒ
determine G[CH S(O)O ]. From the absorbance build-up
3
24
3Ϫ
ؒ
scavenged by the azide anion.
G[Fe(CN)6 ] [corrected for the direct reaction of OH with
ؒ
4Ϫ
ؒ
Ϫ7
Ϫ1
The overall rate constant of OH with the methanesulfinate
anion [reactions (7) and (8)] has been determined by competi-
tion kinetics with thiocyanate. From the slope of the linear
Fe(CN)6 ] G[CH S(O)O ] = 5.0 × 10 mol J is obtained, i.e.
3
ؒ
~82% of OH reacting with methanesulfinate leads to the
formation of the CH S(O)O radical. The same G value was
ؒ
3
ؒ
Ϫ
competition plot the ratio k /k( OH ϩ SCN ) = 0.48 was
obtained when the experiments were carried out under the low-
7,8
obtained. From the established rate constant of the reaction
dose-rate conditions of γ-radiolysis. Similarly, it was found that
ؒ
10
3
Ϫ1
Ϫ1 22
ؒ
of OH with thiocyanate (k = 1.1 × 10 dm mol s ),
the CH S(O)O radical rapidly oxidises ascorbate. The rate con-
3
9
3
Ϫ1 Ϫ1
9
3
Ϫ1 Ϫ1
k7,8 = 5.3 × 10 dm mol
s
is determined (for a compilation
stant k = 1.5 × 10 dm mol
s
was obtained by monitoring
ؒ
ؒ
of rate constants see Table 1). This value is also in good agree-
the build-up of the ascorbate radical (~86% of OH plus H at
9
3
Ϫ1 Ϫ1
9
3
Ϫ1 Ϫ1
ment with the reported rate constant of 6.0 × 10 dm mol
determined by build-up kinetics at 325 nm.
s
360 nm) and sulfite [k = 1.7 × 10 dm mol s , obtained by
9
ؒ
monitoring the decay of the CH S(O)O radical at 360 nm].
3
In the pulse radiolysis of slightly acidic N O-saturated
The final products and their G values of the γ-radiolysis of
2
solutions of methanesulfinate [pH 4.3–4.7; pK (methanesulfinic
sodium methanesulfinate in N O-saturated solution are listed in
Table 2.
a
2
2
acid) = 2.35] the conductance change observed 20 µs after the
ϩ
pulse is negative, corresponding to the consumption of G(H ϩ
The formation of the major product methanesulfonate indi-
Ϫ7
Ϫ1
ؒ
anion) = 4.0 × 10 mol J . This decrease of conductance can
cates that the CH S(O)O radical decays mainly by bimolecular
3
Ϫ
be explained by the release of OH in reaction (9) and the
disproportionation or recombination and hydrolysis of the
mixed anhydride [reactions (15)–(18)].
ϩ
subsequent loss of H upon neutralisation. At the same
time the methanesulfinate anion is replaced by the neutral
The formation of methane, ethane and dimethyl sulfone pro-
vides direct evidence for the formation of the methyl radical
suggested in reaction (10). The small number of methyl radicals
ؒ
CH S(O)O radical. On the other hand, no change in conduct-
3
ance is associated with reactions (8) and (10), since the bisulfite
J. Chem. Soc., Perkin Trans. 2, 2001, 787–792
789

View Article Online
can undergo either hydrogen abstraction with the starting
Table 3 Products and their G values in the γ-radiolysis (dose rate 0.011
Ϫ1 a
Gy s ) and pulse radiolysis (6 Gy per pulse) of sodium methane-
material to give methane, or recombination to give ethane
Ϫ3
Ϫ3
ؒ
sulfinate (10 mol dm ) in N
.8
2 2
O–O (4 : 1 v/v)-saturated solution at pH
[
reaction (19)], or cross-termination with CH S(O)O , the
3
6
major radical in this system. The potential products are di-
methyl sulfone [reaction (20), detected] and methyl meth-
anesulfinate [reaction (21), not found].
7
Ϫ1
1
0 G values/mol J
a
Products
γ-Radiolysis
Pulse radiolysis
Methanesulfinate consumption
Methanesulfonate
270
250
125
2
83
Oxygen consumption
b
Hydrogen peroxide
Sulfate
Formaldehyde
2.2
2.2
1.2
a
b
Ϫ7
Ϫ1
For data on other conditions see text. Includes 0.75 × 10 mol J
formed in the radiolysis of water.
Ϫ7
Ϫ1
The fact that sulfate (0.7 × 10 mol J ) instead of bisulfite
or sulfite [reaction (10)] was observed as a final product can be
9
ؒ
tration. It was suggested by Sehested and Holcman that the
explained by the rapid oxidation of sulfite by the CH S(O)O
3
ؒ
9
3
Ϫ1 Ϫ1
methylsulfonylperoxyl radical [CH S(O) OO ] initiates a
radical [reactions (22) and (23), k = 1.7 × 10 dm mol s , see
3 2
chain reaction by rapidly oxidising the substrate to re-form the
Ϫ1
above].
ؒ
8
3
CH S(O)O radical [reaction (25), k = 6.2 × 10 dm mol
3
25
Ϫ1
ؒ
2Ϫ
3
Ϫ
ؒ
Ϫ
s ]. Indeed, the sulfonylperoxyl radical has been considered as
27
CH S(O)O ϩ SO → CH S(O)O ϩ SO
3
(22)
3
3
one of the most reactive peroxyl radicals. The propagating
ؒ
ؒ
ؒ
ؒ
Ϫ
3
radicals would thus be CH S(O)O and CH S(O) OO . The
intermediate product, the methaneperoxymonosulfonic acid
formed in reaction (25), was suggested to subsequently react
3 3 2
CH S(O)O ϩ SO ϩ H O →
3
2
Ϫ
2Ϫ
ϩ
CH S(O)O ϩ SO ϩ 2 H (23)
3
4
ؒ
Altogether two CH S(O)O radicals will be reduced by one
3
ؒ
Ϫ7
Ϫ1
sulfite. This will leave G[CH S(O)O ] = 4.2 × 10 mol J (i.e.
3
7
5%) for bimolecular self-termination. The yield found for
methanesulfonate fits this material balance. Thus we can
ؒ
conclude from the above product analysis that ~88% of OH
with methanesulfinic acid to give two molecules of the final
reacting with methanesulfinate leads to the formation of
3
3
ؒ
product, methanesulfonic acid [reaction (26), k ≈ 5 × 10 dm
26
Ϫ1 Ϫ1 9
the CH S(O)O radical and ~12% to the methyl radical and
3
mol s ].
bisulfite, leaving very little room for reaction (8).
Reactions in the presence of dioxygen
Pulse radiolysis observations. In the pulse radiolysis of an
N O–O -saturated solution of sodium methanesulfinate, the
2
2
ؒ
absorption of the CH S(O)O radical at 330 nm decays rapidly
3
Methaneperoxymonosulfonic acid is analogous to peroxy-
during the first 5–10 µs. This is then followed by a much slower
decay (t1/2 about 100 µs). As reported in the earlier study, we
9
monosulfuric acid (Caro’s acid). We have in the present context
Ϫ5
Ϫ3
looked at the reaction of Caro’s acid (2 × 10 mol dm ) with
likewise observe that, depending on the ratio of dioxygen to
methanesulfinic acid, a residual absorption of the methanesul-
fonyl radical remains after the initial fast decay. The amount of
initial absorbance decrease is only 10% at a methanesulfinate
Ϫ4
Ϫ3
methanesulfinic acid (1 × 10 mol dm ) and found that the
reaction is completed in less than 2 min (not taking into
account the time required for chromatography). This gives
support to the above value of the rate constant suggested for
Ϫ3
Ϫ3
concentration of 1 × 10 mol dm and a dioxygen concen-
9
Ϫ4
Ϫ3
reaction (26).
tration of 1.3 × 10 mol dm . This amount increases to 80%
at the lower methanesulfinate concentration of 4 × 10 mol
dm and higher dioxygen concentration of 6.5 × 10 mol
dm . From the kinetics of the initial fast decay, the rate
Ϫ4
Ϫ3
Ϫ4
Products of the chain reaction. In the present study, the
products of the chain oxidation of methanesulfinic acid and
their yields have been determined under the low dose rate of
γ-radiolysis and the high dose rate of pulsed electrons (cf.
Table 3).
Ϫ3
constant of the addition of dioxygen to the methanesulfonyl
8
3
Ϫ1 Ϫ1
radical [reaction (24)] is estimated to be ~ 8 × 10 dm mol s ,
3
9
9
in fair agreement with the published value of 1.2 × 10 dm
mol s .
Ϫ1 Ϫ1
In N O–O (4 : 1)-saturated solutions, the dioxygen concen-
2
2
Ϫ4
Ϫ3
tration is 2.6 × 10 mol dm . As long as the chain reaction
Ϫ7
proceeds, dioxygen is consumed with a G value of 125 × 10
Ϫ1
mol J (Fig. 2) and methanesulfonic acid is formed with
Ϫ7
Ϫ1
G = 250 × 10 mol J (cf. inset in Fig. 2). A noticeable break
in the further formation of methanesulfonic acid is observed
after the depletion of dioxygen. As can be seen from Fig. 2, the
overall stoichiometry for the formation of the major product,
methanesulfonate, can be described by reaction (27).
Following the initial fast decay of the absorbance of the
ؒ
CH S(O)O radical, which can be attributed to reaction (24),
3
the subsequent evolution of the remaining absorbance at 330
nm suggests the existence of an “equilibrium” concentration of
ؒ
the CH S(O)O radical. The assumption of a reversibility of the
2 CH
3
S(O)OH ϩ O
2
→ 2 CH
3
S(O) OH
2
(27)
3
dioxygen addition reaction (24), which has been observed in
2
5,26
the case of certain thiyl radicals,
cannot, however, explain
This chain reaction is characterised by the fact that the chain
length is independent of the concentration of the two major
the above observations that the “equilibrium” concentration of
the CH S(O)O radical does not only depend on the dioxygen
concentration, but also on the methanesulfinic acid concen-
ؒ
Ϫ4
components, methanesulfinic acid (varied from 1 × 10 to
3
Ϫ3
Ϫ3
Ϫ4
4 × 10 mol dm ) and dioxygen (varied from 2.6 × 10 to
7
90
J. Chem. Soc., Perkin Trans. 2, 2001, 787–792

View Article Online
Fig. 2 γ-Radiolysis of N O–O -saturated aqueous solutions of meth-
Fig. 4 γ-Radiolysis of neutral N O–O (4 : 1)-saturated aqueous
2 2
Ϫ1
2
2
Ϫ3
Ϫ3
anesulfinic acid (1 × 10 mol dm ). Consumption of O as a function
solution of methanesulfinate at 0.011 Gy s . G(sulfate) as a function
2
of dose. Inset: Consumption of methanesulfinic acid (᭝) and formation
of methanesulfonic acid (᭺) as functions of dose.
of the substrate concentration.
close to diffusion controlled (see Table 1), cannot compete
effectively with its propagation. Thus the chain would continue
to run until one of the components, dioxygen or methane-
sulfinic acid, is exhausted. The branching [reactions (9) and
(
10)] occurs at the initiation stage and does not impact the reac-
9
tion chain postulated [reactions (24) and (25)]. Moreover,
taking into account the rates of bimolecular termination of all
the radicals involved, assumed to be diffusion controlled, it can
be readily shown by a computer simulation that these do not
alter the situation substantially. Thus, the simple mechanism
9
suggested by Sehested and Holcman cannot explain the
limited extent of the chain reaction determined by our product
studies.
An efficient chain breaking process is required to limit the
Ϫ7
Ϫ1
chain length, e.g. to G ~250 × 10 mol J , i.e. a turnover of
50, as under our experimental conditions. We propose that, as
~
Fig. 3 Pulse radiolysis of neutral N O–O (4 : 1)-saturated aqueous
2
2
Ϫ3
Ϫ3
28–30
solutions of methanesulfinate (1 × 10 mol dm ). G(methanesulfonic
acid) as a function of dose rate (Gy per pulse). Inset: G(methane-
sulfonic acid) as a function of the methanesulfinate concentration,
in the peroxyl-radical-induced oxidation of organic sulfides,
ؒ
the CH S(O) OO peroxyl radical reacts with the methane-
3
2
sulfinate ion by an overall two-electron oxidation [equivalent to
6
Gy per pulse.
an oxygen-atom transfer, reaction (28), but possibly proceeding
ؒ
via an adduct intermediate] to yield the CH S(O) O oxyl-type
3
2
Ϫ3
Ϫ3
1
.3 × 10 mol dm ). The chain length is also independent of
radical. This is then proposed to be followed by an electron
Ϫ1
the dose rate in the range of 0.011 to 0.165 Gy s (γ-radiolysis).
Only at very much higher dose rates of pulse radiolysis do the
chain lengths become noticeably shorter (Fig. 3), although at a
dose of 0.8 Gy per µs pulse the yield of methanesulfonic acid
still is not much lower than the value obtained under the low
dose rate of γ-radiolysis (Table 3).
transfer from the methanesulfinate ion to the strongly oxidising
ؒ
ؒ
CH S(O) O [reaction (29)], whereby the CH S(O)O radical is
3
2
3
regenerated. The difference between our extended mechanism
6
and that proposed by Sehested and Holcman is that the
ؒ
propagating radicals in our case include the CH S(O) O ,
3
2
ؒ
ؒ
besides the CH S(O)O and CH S(O) OO radicals.
3
3
2
Owing to the fact that the methanesulfinic acid used in our
experiments contains ~3% methanesulfonic acid, and that these
two compounds elute close together in our ion chromatography
analyses, it was difficult to evaluate the formation of methane-
sulfonic acid in the presence of very high methanesulfinic
acid concentrations. However, sulfate has a sufficiently long
retention time, and a study of its formation as a function of a
larger variation of the substrate concentration could be made.
As can be seen from Fig. 4, its yield increases noticeably at
higher methanesulfinic acid concentrations.
The nature of the chain reaction. From their pulse radiolysis
9
data, Sehested and Holcman convincingly concluded that
the steady-state concentration of the methanesulfonyl radicals
depends on the [dioxygen] : [methanesulfinic acid] ratio. Under
our experimental conditions the steady-state concentration of
the methanesulfonyl radicals is rather high. Accordingly, reac-
ؒ
2
The CH S(O) O radical can be viewed as possessing a
3
ؒ
tion (25), combined with the rapid addition of dioxygen (k
=
reactivity intermediate between OH (good H-abstractor) and
SO4 (good electron abstractor). Hence, parallel to the chain-
propagating step [reaction (29)], a branching reaction, in the
form of H-abstraction by the CH S(O) O radical to give the
carbon-centred radical and methanesulfonic acid [reaction
(30)], can take place. The alkylperoxyl radical originating from
the carbon-centered radical in its reaction with dioxygen
24
9
3
Ϫ1 Ϫ1
Ϫ
1
.2 × 10 dm mol s ), is expected to lead to a very fast chain
᝽
reaction. Indeed, our results show that even under the very high
dose rate of the pulse radiolysis experiments the methane-
sulfonic acid yields are still very high (Fig. 3). In fact, such a
chain reaction would proceed so fast that the bimolecular
termination of the methanesulfonyl radical, although its rate is
ؒ
3
2
J. Chem. Soc., Perkin Trans. 2, 2001, 787–792
791

View Article Online
[
reaction (31)] has a much lower oxidising power than the
4 R. O. C. Norman and B. C. Gilbert, Adv. Phys. Org. Chem., 1967, 5,
3.
D. Veltwisch and K.-D. Asmus, J. Chem. Soc., Perkin Trans. 2, 1982,
143.
D. Veltwisch and K.-D. Asmus, J. Chem. Soc., Perkin Trans. 2, 1982,
147.
5
chain-carrying S-centered peroxyl and oxyl radicals in this sys-
tem (for a study on the reduction potentials of peroxyl radicals
see ref. 31). The relatively inert OOCH S(O)O radical may
thus undergo termination reactions rather than propagating the
chain. The chain length of about 50 suggests a branching ratio
k /k in the order of 10 .
5
6
1
ؒ
–
2
1
7 P. Ulanski, G. Merényi, J. Lind, R. Wagner and C. von Sonntag,
J. Chem. Soc., Perkin Trans. 2, 1999, 673.
2
29
30
8
W.-F. Wang, M. N. Schuchmann, H.-P. Schuchmann and C. von
Sonntag, Chem. Eur. J., 2001, 7, 791.
A more detailed description of this complex chain reaction
would also require an explanation for the increase in sulfate
yields with increasing substrate concentration (see Fig. 4).
Here, one might tentatively suggest that the OOCH S(O)O
radical [cf. reaction (31)], which normally decays by bimolec-
ular termination, may, at sufficiently high substrate concen-
tration, also play a role as a chain carrier by oxidising the
substrate. The bona fide hydroperoxide resulting from this
mechanism would readily decompose into formaldehyde and
9
K. Sehested and J. Holcman, Radiat. Phys. Chem., 1996, 47,
3
57.
ؒ
Ϫ
2
10 G. V. Buxton, G. A. Salmon and N. D. Wood, in Phys.-Chem. Behav.
Atmos. Pollut. [Proc. Eur. Symp.], ed. G. Restelli and G. Angeletti,
Kluwer, Dordrecht, 1990, p. 245.
1
1 G. V. Buxton, S. McGowan, G. A. Salmon, J. E. William and N. D.
Woods, Atmos. Environ., 1996, 30, 2483.
2 M. N. Schuchmann, H. Zegota and C. von Sonntag,
Z. Naturforsch., B: Chem. Sci., 1985, 40, 215.
1
SO (→sulfate). The decomposition of similar hydroperoxides
3
13 C. von Sonntag and H.-P. Schuchmann, Methods Enzymol., 1994,
32–34
is well documented.
Another hypothesis postulates loss of
233, 3.
ؒ
ؒ
ؒ
ؒ
CH radicals from an RO –methanesulfinic acid adduct [RO
14 R. H. Schuler, L. K. Patterson and E. Janata, J. Phys. Chem., 1980,
84, 2088.
3
2
2
being a peroxyl radical of relatively low reactivity, e.g., CH OO
and OOCH S(O)O ], rearrangement and hydrolysis of the
sulfate monoester [reactions (32)–(34)]. Evidence for the exist-
ence of similar adducts of RO2 to sulfides has been reported.
3
ؒ
Ϫ
15 G. V. Buxton and C. R. Stuart, J. Chem. Soc., Faraday Trans., 1995,
2
9
1, 279.
1
1
1
6 H.-P. Schuchmann, D. J. Deeble, G. O. Phillips and C. von Sonntag,
ؒ
29
Radiat. Phys. Chem., 1991, 37, 157.
7 R. H. Schuler, A. L. Hartzell and B. Behar, J. Phys. Chem., 1981, 85,
192.
Oxidation of methanesulfinic acid by ozone
8 A. O. Allen, C. J. Hochanadel, J. A. Ghormley and T. W. Davis,
J. Phys. Chem., 1952, 56, 575.
With reducing compounds, ozone can undergo both O-atom
transfer as well as electron transfer. One such system involves
tertiary amines, which give rise mainly to aminoxides but to a
small extent (~10%) may also undergo electron transfer. Using
buten-3-ol (k = 7.9 × 10 dm mol s ) as a competitor, we
determined the rate constant of methanesulfinate ion with
ozone to be 2 × 10 dm mol s . This value is comparable to
the rate constants reported for bisulfite, which range between
19 G. M. Eisenberg, Ind. Eng. Chem., Anal. Ed., 1943, 15, 327.
2
0 L. Forni, D. Bahnemann and E. J. Hart, J. Phys. Chem., 1982, 86,
35
255.
4
3
Ϫ1 Ϫ1 36
37
21 E. J. Hart, K. Sehested and J. Holcman, Anal. Chem., 1983, 55,
4
6.
2
2 G. V. Buxton, C. L. Greenstock, W. P. Helman and A. B. Ross,
6
3
Ϫ1 Ϫ1
J. Phys. Chem. Ref. Data, 1988, 17, 513.
23 B. H. J. Bielski, D. E. Cabelli, R. L. Arudi and A. B. Ross, J. Phys.
Chem. Ref. Data, 1985, 14, 1041.
5
6
3
Ϫ1 Ϫ1 38
3
.7 × 10 and 2 × 10 dm mol s . We observed that 1 mol
2
4 D. J. Deeble, B. J. Parsons and G. R. A. Johnson, Radiat. Phys.
Chem., 1990, 36, 487.
5 M. Tamba, G. Simone and M. Quintiliani, Int. J. Radiat. Biol., 1986,
methanesulfonic acid is formed per mol of ozone, i.e. the reac-
tion proceeds presumably only by O-atom transfer. If electron
transfer had occurred, even at a very small percentage, the high
rate of the chain reaction discussed above would have resulted
in a much higher methanesulfonic acid yield. In the O-atom
transfer reactions of ozone the dioxygen is formed in its singlet
excited state [O ( ∆ )], and also in the reaction of ozone with
methanesulfinic acid O ( ∆ ) is formed in 100% yield. If the
reduction potential of bisulfite can be used as a reference for
methanesulfinate, O ( ∆ ) is also not capable of one-electron
2
5
0, 595.
26 X. Zhang, N. Zhang, H.-P. Schuchmann and C. von Sonntag,
J. Phys. Chem., 1994, 98, 6541.
2
7 Y. Razskazovskii, D. Becker and M. D. Sevilla, in S-Centered
Radicals, ed. B. Z. Alfassi, John Wiley & Sons Ltd., 1999,
p. 245.
1
2
g
1
39
2
g
2
8 C. Schöneich, A. Aced and K.-D. Asmus, J. Am. Chem. Soc., 1991,
40
1
13, 375.
1
2
g
29 M. Bonifacic and I. Stefanic, Nukleonika, 2000, 45, 39.
30 R. Flyunt, O. Makogon, C. Schöneich and K.-D. Asmus,
Nukleonika, 2000, 45, 45.
oxidation of methanesulfinate to induce the chain reaction, as
indeed confirmed by our results.
3
1 G. Merényi, J. Lind and L. Engman, J. Chem. Soc., Perkin Trans. 2,
994, 2551.
1
Acknowledgements
32 P. Ulanski, E. Bothe, J. M. Rosiak and C. von Sonntag, J. Chem.
Soc., Perkin Trans. 2, 1996, 5.
We would like to thank Dr H.-P. Schuchmann for valuable
discussions. Part of this work has been supported by the
Office of Basic Energy Sciences of the US Department of
Energy (document NDRL-4282 of the Notre Dame Radiation
Laboratory).
3
3 B. Höbel and C. von Sonntag, J. Chem. Soc., Perkin Trans. 2, 1998,
509.
34 A. Leitzke, E. Reisz, R. Flyunt and C. von Sonntag, J. Chem. Soc.,
Perkin Trans. 2, in the press.
3
3
3
5 F. Muñoz and C. von Sonntag, J. Chem. Soc., Perkin Trans. 2, 2000,
029.
6 P. Dowideit and C. von Sonntag, Environ. Sci. Technol., 1998, 32,
112.
2
1
References
7 F. Muñoz and C. von Sonntag, J. Chem. Soc., Perkin Trans. 2, 2000,
1
2
3
H. Zegota, M. N. Schuchmann, D. Schulz and C. von Sonntag,
Z. Naturforsch., B: Chem. Sci., 1986, 41, 1015.
D. Veltwisch, E. Janata and K.-D. Asmus, J. Chem. Soc., Perkin
Trans. 2, 1980, 146.
661.
38 P. Neta, R. E. Huie and A. B. Ross, J. Phys. Chem. Ref. Data, 1988,
17, 1027.
39 F. Muñoz, E. Mvula, S. Braslavsky and C. von Sonntag,
unpublished results.
W. T. Dixon, R. O. C. Norman and A. L. Buley, J. Chem. Soc., 1964,
3
625.
40 P. Wardman, J. Phys. Chem. Ref. Data, 1989, 18, 1637.
7
92
J. Chem. Soc., Perkin Trans. 2, 2001, 787–792