Oxidation of glyoxal initiated by ?OH in oxygenated aqueous solution

Article abstract of DOI:10.1039/a701468f

The kinetics and mechanism of the oxidation of glyoxal, which is a constituent of cloud water, initiated by ^?OH in oxygenated solution have been investigated using pulse radiolysis with optical and conductivity detection of the transient species, and steady-state radiolysis with spectrophotometric and ion chromatographic analysis of the permanent products. The data obtained are consistent with glyoxal being present in the form of the dihydrate [CH(OH)₂]₂ which is oxidised to glyoxylic acid (pK₂ = 3.4) and hydrogen peroxide via a peroxyl radical ^?O₂C(OH)₂CH(OH)₂ that splits off HO₂^? in a non-rate determining step. The following rate constants have been determined: k{^?OH + [CH(OH)₂]₂} = (1.10 + 0.04) × 10⁹ dm³ mol^-1 s^-1 and k[^?C(OH)₂CH(OH)₂ + O₂] = (1.38 ± 0.11) × 10⁹ dm³ mol^-1 s^-1. It is concluded that oxidation of glyoxal by ?OH in cloud water can proceed by a chain reaction involving H₂O₂.

Full text of DOI:10.1039/a701468f

View Article Online / Journal Homepage / Table of Contents for this issue
Oxidation of glyoxal initiated by ~OH in oxygenated aqueous solution
George V. Buxton,* Treena N. Malone and G. Arthur Salmon
Cookridge Radiation Research Centre, University of L eeds, Cookridge Hospital, L eeds,
UK L S16 6QB
The kinetics and mechanism of the oxidation of glyoxal, which is a constituent of cloud water, initiated by ~OH in oxygenated
solution have been investigated using pulse radiolysis with optical and conductivity detection of the transient species, and steady-
state radiolysis with spectrophotometric and ion chromatographic analysis of the permanent products. The data obtained are
consistent with glyoxal being present in the form of the dihydrate [CH(OH) ] which is oxidised to glyoxylic acid (pK \ 3.4) and
2
2
a
hydrogen peroxide via a peroxyl radical ~O C(OH) CH(OH) that splits o† HO ~ in a non-rate determining step. The following
2
2
2
2
rate constants have been determined: kM~OH ] [CH(OH) ] N \ (1.10 ^ 0.04) ] 109 dm3 mol~1 s~1 and k[~C(OH) CH(OH)
2
2
2
2
]
O ] \ (1.38 ^ 0.11) ] 109 dm3 mol~1 s~1. It is concluded that oxidation of glyoxal by ~OH in cloud water can proceed by a
2
chain reaction involving H O .
2
2
Hydrocarbons have been identiÐed in the atmosphere and are
known to be emitted from human and biological activity.1
Glyoxal, an a-dicarbonyl compound, has been identiÐed2 in
cloud water with a mean concentration of 2 lmol dm~3.3 It is
a product of the oxidation of aromatic compounds, which are
abundant in motor-vehicle emissions and in the combustion
products of biomass; it is also a product of the oxidation of
terpenes, in particular isoprene, which is the predominant
hydrocarbon emitted from forests.4 The hydroxyl radical has
been suggested as an important intermediate in the aqueous-
phase chemistry of the atmosphere which contributes to the
generation of acid precipitation.5 It is ubiquitous in the gas
phase of the atmosphere from which it can enter cloud drop-
lets, and it may be generated within cloud droplets in several
di†erent photochemical and thermal reactions of compounds
that are present there.6
Hydrogen peroxide was analysed by the Ghormley
method13 and glyoxylic acid was determined using ion chro-
matography (Dionex 2000i).14
Results and Discussion
The radiolysis of water is summarised by reaction (1)
H O , 0.27 e~ , 0.06~H, 0.27~OH, 0.045H ,
2
(aq)
2
0.07H O , 0.27H` (1)
2
2
where the numbers are the yields (lmol J~1) of the radiolysis
products.
Pulse radiolysis experiments
Rate of reaction of ~OH with glyoxal. The rate of this reac-
tion was measured in N O-saturated solution over the pH
2
range 2.0È6.0 by competition kinetics using Fe(CN) 4~ as the
6
competitor and measuring the yield of Fe(CN) 3~ by observ-
6
In this paper we report on a study of the oxidation of
glyoxal by ~OH in the presence of oxygen, using the tech-
niques of pulse and steady-state radiolysis. From the evidence
available in the literature7,8 we conclude that glyoxal exists
ing the optical absorbance at a wavelength of 420 nm. Under
these conditions e~ is converted to ~OH during the pulse (0.6
(aq)
ls), so that at pH [ 3, G(~OH) \ 0.55 lmol J~1,15 and ~H is
unreactive on this timescale. In more acidic solutions some
e~ react with H` to form more ~H, but this does not a†ect
primarily in
a
monomeric form as the dihydrate,
[
CH(OH) ] , at the concentrations found in cloud water
(
aq)
2
2
the determination of k .
droplets and used in the present experiments.
2
~
OH ] [CH(OH) ] ] ~C(OH) CH(OH) ] H O
(2)
(3)
2
2
2
2
2
Experimental
The pulse radiolysis apparatus and procedures used here have
~
OH ] Fe(CN) 4~ ] Fe(CN) 3~ ] OH~
6
6
Values of k at the relevant pH were taken from ref. 16. The
3
2
been described in detail elsewhere.9 For optical measurements,
data gave k \ (1.10 ^ 0.04) ] 109 dm3 mol~1 s~1 at 20 ¡C,
dosimetry was carried out using an O -saturated solution of
independent of pH over the range 2.0È6.0. At pH P 4.0, k
2
2
1
0~2 mol dm3 KSCN with Ge[(SCN) ~~] at 475 nm taken to
was also measured by direct observation of the formation of
2
be10 2.59 ] 10~4 m2 J~1 where G is the radiation chemical
yield (mol J~1) and e is the molar absorption coefficient (m2
mol~1). For conductivity measurements, the dosimeter was an
the glyoxal radical and the same value was obtained as by the
competition method. An activation energy for reaction (2) of
12.6 ^ 0.9 kJ mol~1 was obtained from direct measurements
N O-saturated solution containing 10~3 mol dm~3 dimethyl
of k at pH 4.0 over the temperature range 6È46 ¡C. At
2
2
sulfoxide at pH 4.4 for which GK \ 2.02 ] 10~8 S m2 J~1.11
pH \ 4, values of k could only be obtained by the com-
2
Steady-state radiolysis was carried out using a 60Co c-ray
source at a dose rate of 15.7 Gy min~1 as measured by Fricke
dosimetry.12 All experiments were carried out at ambient tem-
perature (18È20 ¡C) unless otherwise stated.
Glyoxal was [98% pure (Fluka) and was used as received.
All other reagents (BDH or Hopkin and Williams) were
AnalaR grade or better. Solutions were made up with distilled
water which had been further puriÐed by Millipore Ðltration
petition method because further absorbance changes occurred
after the glyoxal radical had been formed.
The spectrum of the product radical, assumed to be
~C(OH) CH(OH) , measured in N -saturated solution con-
2
2
2
taining 1 ] 10~3 mol dm~3 glyoxal is shown in Fig. 1. The
spectrum was measured 3 ls after the pulse, i.e. the time when
reaction (2) was complete, and was found to be independent of
pH. As noted above, further changes in absorbance occurred
after the formation of the glyoxal radical below pH 4.0; these
changes are thought to be due to the acid-catalysed dehydra-
tion of the radical but they are not relevant to the present
study.
(
MilliQ). The pH of the solutions was adjusted with perchloric
acid or phosphate bu†er. Oxygen and nitrous oxide (BOC)
were of the highest purity available and were taken straight
from the cylinder.
J. Chem. Soc., Faraday T rans., 1997, 93(16), 2889È2891
2889

View Article Online
Fig. 1 Spectrum of the product from reaction (2) in N O-saturated
2
solutions containing 1 ] 10~3 mol dm~3 glyoxal at pH 2.5 (=) and
Fig. 4 Comparison of the pH dependence of the second-order rate
constant for the decay of the observed product of reaction (5) (K)
with that for HO ~/O ~~ (ÈÈÈ) from ref. 19 and in this work from
pH 4.0 (K). The spectra were taken 3 ls after the pulse. Inset: FAC-
SIMILE Ðt (ÈÈÈ) for the absorbance change (…) due to reactions
2
2
(
2) and (5) at pH 3.0.
the formate/formic acid system (…)
Reaction of the glyoxal radical with oxygen. The reaction of
~
C(OH) CH(OH) with oxygen was measured in solutions
2
2
containing 1 ] 10~3 mol dm~3 glyoxal, 6.85 ] 10~4 mol
ditions the expected reactions are (2), (4) and (5), with
dm~3 O and 1.14 ] 10~2 mol dm~3 N O. Under these con-
G(HO ~) \ 0.06 lmol J~1 and G(RO ~) \ 0.55 lmol J~1 so
2
2
2
2
that RO ~ comprises ca. 90% of the initial products.
2
~
H ] O ] HO ~
(4)
(5)
2
2
~
C(OH) CH(OH) ] O ] RO ~
2
2
2
2
The rate constant for reaction (5) was obtained from the rate
of decay of the absorbance of ~C(OH) CH(OH) at a wave-
2
2
length of 280 nm. Because the growth of absorbance due to
reaction (2) overlapped the beginning of this decay, k was
5
calculated using FACSIMILE curve-Ðtting software17 as illus-
trated in the inset to Fig. 1. Measurements made at pH 3 and
4
gave k \ (1.38 ^ 0.11) ] 109 dm3 mol~1 s~1; at higher pH,
5
there was little change in absorbance with time so that k
could not be determined under those conditions.
5
Fig. 2 shows the sum of the absorption spectra of the pro-
ducts of reactions (4) and (5) at di†erent pH. The shapes and
pH dependence of these spectra are very similar to those of
HO ~ and O ~~ (see Fig. 2 inset) suggesting that RO ~ rapidly
Fig. 2 Spectrum of the observed product of reaction (5) measured in
solutions containing 10~3 mol dm~3 glyoxal, 6.85 ] 10~4 mol dm~3
2
2
2
dissociates to form HO ~/O ~~ and glyoxylic acid as in reac-
2
2
O at pH 2.0 (=), pH 3.0 (L), pH 4.0 (·), pH 5.3 (K), pH 5.7 (…) and
2
pH 6.9 (|). Spectra of HO ~ and O ~~ are shown in the inset.
2
2
Fig. 3 Dependence of the molar absorption coefficient of the
observed product of reaction (5) (K) at a wavelength of 240 nm on
pH. The expected dependence calculated for HO ~~/O ~~ is shown
Fig. 5 Oscilloscope traces associated with conductivity changes due
to reactions (4) and (5) measured at pH 3.0 (L) and pH 4.0 (=) in
solutions containing 1 ] 10~3 mol dm~3 glyoxal, 6.85 ] 10~4 mol
dm~3 O and 1.14 ] 10~2 mol dm~3 N O
2 2
ÈÈÈ). The formate/formic acid data (…) are shown for comparison.
(
2
2
2890
J. Chem. Soc., Faraday T rans., 1997, V ol. 93

View Article Online
Table 1 Measured and calculated conductivity changes at pH 3.0
and 4.0
dihydrate since only one type of radical was detected as the
product of reaction (2) (see Fig. 1). Thus free radical oxidation
of glyoxal in cloud water will be a source of HO ~ and hence
hydrogen peroxide. Both of these species are in turn sources of
pH
; G j
/S m2 J~1
; G j
i i calc.
/S m2 J~1
2
i i meas.
3
4
.0
.0
5.32 ] 10~9
1.62 ] 10~8
5.61 ] 10~9
1.60 ] 10~8
~OH via reactions (7), (10), (11) and (12),
O ~~ ] O ] H O ] ~OH ] OH~ ] 2O
(10)
(11)
(12)
2
3
2
2
H O ] hl ] 2~OH
2
2
tions (6) and (7),
H O ] FeII ] FeIII ] ~OH ] OH~
2
2
~
O C(OH) CH(OH) ] HOOCCH(OH) ] HO ~
(6)
(7)
2
2
2
2
2
with the result that oxidation of glyoxal in cloud water initi-
HO ~ H O ~~ ] H`
ated by ~OH generates glyoxylic acid (pK \ 3.4) in a chain
reaction without any diminution in the oxidising capacity of
the atmosphere. Further oxidation of glyoxylic acid to oxalic
2
2
a
Further evidence that the observed absorbing product is
HO ~/O ~~ rather than RO ~ is provided by the data in Fig. 3
acid (pK \ 1.23) is expected to occur by the same mecha-
2
2
2
where the dependence of the molar absorption coefficient of
the product at a wavelength of 240 nm on pH is compared
with that of HO ~/O ~~. Also shown in Fig. 3 are data for
a1
nism. The concentrations of glyoxylic and oxalic acids in
atmospheric water droplets are typically in the range 0.1È20
lmol dm~3,24 of the same order of magnitude as dissolved
iron. Moreover, it has been shown under these conditions24
that the photolysis of iron(III) oxalato complexes could be a
2
2
solutions where glyoxal was replaced by formate/formic acid,
which is known to produce HO ~/O ~~.16,18 Supporting evi-
2
2
dence is also provided by the data in Fig. 4 where the pH-
dependent second-order rate constants for the decay of the
observed product compare well with those for self reaction of
HO ~/O ~~.19
major source of H O in cloud, fog and rain water.
2 2
Acknowledgement is made to the Commission of the Euro-
pean Union for Ðnancial support under contract ENV4-CT95-
2
2
Conductivity changes associated with reactions (4) and (5)
are illustrated in Fig. 5. The magnitude of the change in con-
ductivity with pH is consistent with dissociation of a proton
from a species with a pK less than 4.0. Taking the pK of
0175.
a
a
References
glyoxylic acid to be 3.4 at an ionic strength of 10~3 mol
dm~3, based on the value20 of 3.18 at an ionic strength of 0.1
mol dm~3, and the pK of HO ~ \ 4.8,21 one can calculate
1
P. Warneck, T he Chemistry of the Natural Atmosphere, Int.
Geophys. Series, Academic Press, London, 1988.
V. A. Isidorov, I. G. Zenkevich and B. V. Io†e, Atmos. Environ.,
1985, 19, 1.
a
2
2
the expected change in conductivity, taking j \ 3.15
H`
]
10~2 S m2 mol~1, and j \ 4.1 ] 10~3 S m2 mol~1 for gly-
i
3
4
5
T. M. Olson and M. R. Ho†man, J. Phys. Chem., 1988, 92, 533.
R. A. Rasmussen, J. Air Pollut. Control Assoc., 1972, 22, 537.
G. V. Buxton, S. McGowan, G. A. Salmon, J. E. Williams and
N. D. Wood, Atmos. Environ., 1996, 30, 2483.
Heterogeneous and L iquid Phase Processes. L aboratory Studies
Related to Aerosols and Clouds, ed. P. Warneck, Springer-Verlag,
Berlin, 1996.
oxylate (the value for acetate ion22) and O ~~. The calculated
2
values are compared with the measured values in Table 1; the
agreement is good at each pH. The rate of growth of the con-
ductivity signal was limited by the response time of the mea-
suring device (ca. 5 ls), indicating that changes resulting from
reactions (6) and (7) occurred on a shorter timescale.
6
7
8
9
A. R. Fratzke and P. J. Reilly, Int. J. Chem. Kinet., 1986, 18, 757.
A. R. Fratzke and P. J. Reilly, Int. J. Chem. Kinet., 1986, 18, 775.
J. Grimshaw, J. R. Langan and G. A. Salmon, J. Chem. Soc.,
Faraday T rans., 1994, 90, 75.
Steady-state radiolysis experiments
These were carried out to identify the permanent products of
the oxidation of glyoxal by ~OH in the presence of oxygen.
Analysis of the irradiated solution for hydrogen peroxide
showed that G(H O ) \ (3.57 ^ 0.01) ] 10~7 mol J~1. This is
1
0
G. V. Buxton and C. R. Stuart, J. Chem. Soc., Faraday T rans.,
1
995, 91, 279.
11 D. Weltwisch, E. Janata and K-D. Asmus, J. Chem. Soc., Perkin
T rans. 2, 1979, 146.
J. W. T. Spinks and R. J. Woods, An Introduction to Radiation
Chemistry, Wiley, New York, 2nd edn., 1976.
2
2
1
2
very close to the expected result of [G(~OH) ] G(~H)]/2 ] 0.07
mol J~1 [see reaction (1)] if HO ~ is produced in reaction (6),
2
13 A. O. Allen, C. J. Hochanadel, J. A. Ghormley and T. W. Davis,
which will be followed by reactions (8) and (9),
J. Phys. Chem., 1952, 56, 575.
4 S. A. Murray, PhD thesis, University of Leeds, 1997.
1
HO ~ ] HO ~ ] H O ] O
(8)
(9)
2
2
2 2
2
15 R. H. Schuler, A. L. Hartzell and B. Behar, J. Phys. Chem., 1981,
85, 192.
HO ~ ] O ~~ ] H O ] H O ] O ] OH~
2
2
2
2 2
2
16 G. V. Buxton, C. L. Greenstock, W. P. Helman and A. B. Ross, J.
Phys. Chem. Ref. Data, 1988, 17, 513.
When catalase was added to the irradiated solution imme-
diately before analysis no oxidising product was detected by
the Ghormley method. This result is also consistent with
H O being one of the Ðnal oxidation products. The other
1
1
1
7
8
9
A. R. Curtis and W. P. Sweetenham, FACSIMILE/CHECKMAT
users manual, UKAEA, 1976, AERER 12805.
G. V. Buxton, R. M. Sellers and D. R. McCracken, J. Chem. Soc.,
Faraday T rans. 1, 1976, 72, 1464.
2
2
expected product, glyoxylic acid, was detected by ion chroma-
B. H. J. Bielski, Photochem. Photobiol., 1978, 28, 645.
tography with a yield of (4.7 ^ 0.1) ] 10~7 mol J~1, which is
5% lower than expected. Nevertheless, this result provides
further support for the proposed mechanism.
20 Stability Constants Supplement No. 1, Spec. Publ. 25, The Chemi-
1
cal Society, London, 1971.
21
22
23
D. Behar, G. Czapski, J. Rabani, L. M. Dorfman and H. A.
Schwarz, J. Phys. Chem., 1970, 74, 3209.
R. A. Robinson and R. H. Stokes, Electrolyte Solutions, Butter-
worths, London, 1959.
Conclusions
C. von Sonntag, T he Chemical Basis of Radiation Biology, Taylor
and Francis, London, 1987.
All of the data presented above are consistent with the
product of reaction (5) rapidly dissociating to glyoxylic acid
and HO ~. This is characteristic of a-hydroxy-peroxy
radicals23 and consistent with glyoxal being in the form of the
24 Y. Zuo and J. Hoigne, Environ. Sci. T echnol., 1992, 26, 1014.
Paper 7/01468F; Received 3rd March, 1997
2
J. Chem. Soc., Faraday T rans., 1997, V ol. 93
2891