PULSE RADIOLYSIS 0~ OXALICACID AND OXALATES
749
Pulse Radio ysis of Oxalic Acid and Oxalates by N, Getoff, F. Schworer, V. M. Markovic, K. Sehested,*' and S. 0. Nielsen Danish Atomic Energy Commission, Research Establishment Risb, DK-4OOO Roskilde, Denmark, and Mtrx-Planck-Institut fur Kohlenforschung, Abteilung Strahlenchemie, D-433 Malheim a.d. Ruhr, F.R. Germany (Received ,June 1W,1970)
Publication costs assisted by the Danish Atomic Energy Commission
Pulse radiolyses of oxalic acid and oxalates in aqueous solution were carried out. The following rate constaiits were determined: k(e,,HC&-) = (3.2 f 0.6) X lo9M-I sec-l and Ic(e,,C2042-) = (1.7 f 0.b) x lo7 J8-l sec-l. With CNS- ions as competitive solutes k(OH HzC204) = (1.0 f 0.1) X lo6 M-l seec-l; k(OH HCzQ4-) = (3.2 f 0.1) X lo7 1V-l see-', and k(OH C2042-) = (5.3 i 0.3) X lo6 N-' sea-' were a l s ~obtained. With NzO as a scavenger for ea,- and Hz (45 atm) for OH the absorption spectra of several radicals were obtained. Their absorption maxima are between 230 and 250 nm. Reaction mechanisms and kinetic data of the radicals are discussed.
+
+ +
+
1. Introduction Aqueous oxalic acid bas been studied extensively and used as a reliable radiation dosimeter.2a It has also been used as a scavenger for OH radicals for determination of the pH effect, on the primary products of water r a d i ~ l y s i s . *The ~ ~ ~rate constants of the reactions of OH radicalsY4ea,-,- and I1c atoms5 with oxalic acid have been determined by means of competition reactions under steady-sta6e conditions only. It has been found that oxalic acid is a very good scavenger of ea,-,k[eag-
+ H12C2Qz1(HC:tOg--; C20d2-)] = 2.6 X f010(3.4X lo9; 4.8 X lo7) M - l sec-I
The OH radicals react more slowly with the oxalic acid and with both oxalate forms
k[OH
+ HZC204 (ACzQ4-;
6d142-)] ==
5 X IO6 (3.5 X IO7; 6 X lo6) M-l sec-l The corresponding rate constants for the reactions of H atoms were foixntl to be extremely low
Some preliminary pulse radiolysis experiments with oxalic acid were made by Nielsen, Sehested, Pagsberg (Riso), and Christensen (AB Atomenergi, Sweden), and those experiments initiated the present investigations. The formation and the kinetics of the decay as well as the absorpticnn spectra of the short-lived intermediates of irradiated aqueous oxalic acid and oxalates are still unknown. In order to obtain more insight into this subject, w e carried out pulse radiolysis of oxalic acid and oxa1a;lw at different pH values and under various experiinental conditions. We determined k(e,,-,- -tHGz04-) and k(easC2042-)by following the decay of 'The rate constants for the reactions of OH radicals ' b + i t h . the three forms of oxalic acid were
+
+
determined with CNS- and I- ions as competing species.
2. Experimental Section Preparation of the Solutioizs. The solutions were prepared by dissolution of analytical grade oxalic acid and potassium oxalates (E. Merclr AG, Darmstadt) in triply distilled water. The p B of the solutions was adjusted by means of high-purity perchloric acid (6.I?. Smith, Chemical Corp., Columbus, Ohio). The solutions were saturated by bubbling at room temperature (for 45 min) with high-purity argon or NZQ in 100-ml Summit syringes made of Pyrex. All glassware was baked for several hours a t 470". Pulse Irradiation and Optical Detection. 3 lost of the pulse irradiations were carried out with the 10MeV Linac at Riso. The solutions were irradiated with 0.5-psec electron pulses (dose: 3.5 krads/pulse) in a Suprasil quartz cell (light path length 5.2 em, i d . 1.5 em) at room temperature. Some experiments were carried out with a special pressure cell (light path length 7.1 cm) under 45 atm H2. The description of the cell as well as the technique has been given by Pagsberg, el alB6 The details concerning the pulse irradiation and the optical and electronic equipment are found elsewhere. ti p7
(1) To whom correspondence should be addressed a t Danish iltomic Energy Commission, Research Establishment Riso, DK-4000 Roskilde, Denmark. (2) (a) N. W. Holm and I. G. Draganib, Atompraxis, 14, 11/12, 495 (1968): (b) Z. D. Draganid, I. G. Draganib, and XI. M. Kosanic, J . P h y s . Chem., 68, 2085 (1964). (3) Z. D. Draganid, I. G. Draganib, and M. M. Kosanic, {bid., 70, 1418 (1966). (4) Z. D . Draganib, M. M. Kosanic, and M. T. Nedadovic, ;bid., 71, 2390 (1967). (5) 0. Mi6i6 and I, Draganib, I n t . J . Radial. P h w . Chem., 1, 287 (1969). (6) P. Pagsberg, N. Christensen, J. Ilabani, G . Nilsson, J~E'enger, and S. 0. Nielsen, J . P h y s . Chem., 73, 1029 (1969).
The. Journal of Physical Chemistry, 6101. 76, N o . 6 , 1971
GETOFF,SCHWORER, MARKOVIC, SEHESTED, AND XIELSEN
750 The dosv absorbed in the irradiated samples was determined by means of a M Fe(CN),4- solution saturated with N 2 0 and containing a small amount of air.6 * The absorption of the radiation-induced Fe(CN)63- W L S measured at 420 nm (e4t0 = 1000 M-’ C M - * ) ,anc, ~ the absorbed dose was calculated by the use oi G[FetCN),j”-] = 5.25. The rate corstsnts for the reactions of eaq- and OH radicals w t h both oxalate forms were determined by means of the 3-MeV Van de Graaff accelerator (High Voltage En;* Co , type K) of the Max-Planck-Institut in l\lIiilheini/Ruhr. Electron pulses (1.3 psec) were used. The dose was varied from 0.9 to 3.6 krads/pulse. The first-order decay of the electron absorption was followed either at 600 nm (Photomultiplier, 1P28) or at 720 nm (Photodiode, SD 100, Edgerton, Grier and Germeshaufm). The dose was 0.9 kradlpulse, small enough 1o minimize undesired secondary reactions. Further details about this pulse irradiation hcility have been reported previously. 9,10 d!feasuren,ent of Absorption Spectra. Since oxalic acid and both oxalates absorb light strongly in the ultraviolet region, it was not possible to work with concentrations higher than 10 mM even when a flashed Xe lamp“ ~7msused. This concentration is not high enough to ensure the complete scavenging of radicals. Furthermore the OH radicals also absorb in the same wavelength regioix6 Therefore, a computer program was used ir1 each case to calculate the concentrations of radicals lorxiied at a given reaction mechanism and set of rate Constants at a given time after the electron pulse. The same caiculations gave the necessary corrections for 0 1 9 absorption. In order to ensure the validity cf this method several concentrations of oxalic acid were always used, usually between 1 and 10 mM, as well as matrix solutions containing no solute. The computer program used for solving the differential equations representing the course of the radiolytic reactions’l RB,P based on a fifth-order Runge-Kutta method. The computations were carried out on the Riso computer GIER.12313
3. Results and E9iscussion 3.1. Reactions with Solvated Electrons. The rate constants fer the reactions of eaq- with both oxalate forms were determined by the pseudo-first-order decay of the electron absorption. Oxalate solutions (0.610 ma!) a t pH 5 and 9, containing 10 mM of ethanol as scavenger fior the OH radicals and saturated with 02free argon, were used. The rate constants, k(e,,oxalate), were ralculated from the slopes of the straight lines, obtairned by the plotting of -In OD VS. time. Appropriate matrix corrections were applied. In the case of p l l 5 (ca. 16% HCzO4- and 84% C20d2-) a correction for /r(e8Lcl- C2042-) is about 2% and the correction for eaq- 4- &Of is about 10%. pK1 = 1.25 and pKz = 4.28 for oxalic acid dissociation were
+
+-
T h e Journal of Phusicc!l Chemistry, Val. 76,No. 6 , 1971
used in the calculations. The results obtained are shown in Table I together with literature data. The Table I: Rate Constants for t h e Reaction of ea,- with Oxalic Acid and Oxalates k in M
Rate oonSolute
HzCzOa HCz04CzOa2a
- 1 800 -I------,
stant
This work
MieiC and DraganP
ki
Not determined (3.2 rt 0 . 6 ) X lo9 (1.7 ==I 0 . 5 ) x 107
( 2 . 5 =t0.5) x 1010 (3.4 =k 0 . 4 ) x 109 (4.8ir 0 . 3 ) x 107
k, k8
See ref 5 .
rate constants determined by means of competition reactions under steady-state conditions6 are in reasonable agreement with our data. As was to be expected, the rate constants decrease appreciably with __ the dissociation degree of oxalic acid. It is assumed that the electron Is added preferably to an undissociated carboxylic group and a radicalion is formed
O(1)
COOIJ.
0-
+
HC~O~-
/
-3-&-OH
!
( 21
COQ-
COO%-
cZo42+ eaq- -% 1
(3)
COOThe first two rate constants, kl and lcz, conform neither to Brgnsted’s acid catalysis law, applied to reactions between acids and solvated electrons,14nor to the influence of the groups adjacent to the -CO- group, as described by Hart, et al. l5 (7) H. C. Christensen, G. Nilsson, P . Pagsberg, and S. 0 . Nielsen, Rev.Sci. Instrum., 40, 786 (1969). (8) J. Rabani and M. S. Matheson, J . Phrjs. Chem., 70, 761 (1966). (9) G. 0. Schenck and F. Schworer, “Jahrbuch des Landesamtes ftlr Forschung des Landes Nordrhein-Westfalen,” Westdeutches Verlag, Koln, 1964, p 435. (10) hi. Getoff and F. Schworer, Radiat. Rea., 41, 1 (1970). (11) H. Fricke and J. K. Thomas, Radiat. Ees. Szcppl., 35, 4 (1964). (12) K. Sehested, 0. L. Rasmussen, and H. Fricke, J . Phya. Chem., 72, 626 (1968). (13) K. Sehested, E. Bjergbalcke, 0. L. Rasmussen, and H. Fricke, J . Chem. Phys., 51, 3159 (1969). (14) J. Rabani, A d e a n . Chem. Ser., No. 50, 242 (1965). (15) E. J. Hart, E. M . Fielden, and M. Anbar, J . Phys. Chem., 71, 3993 (1967).
PULSE RAD:IOLYF,IB OF OXALICACIDAND OXALATES The radical-ions formed according to reactions 1-3 will be called RAl, RA2, and RA3, respectively. S.2. Reactions with OH Radicals. The rate constants for the reactions of OH radicals with oxalic acid and with both oxalate forms were determined by the method developed by Adams, et As competitive solute CIc’S-- ions and in some cases also I- ions were used. The transient, ((2N8)2-, has a strong absorption at 475 nm ( ~ = 6 7600 M-I cm-I) and the Iz- radical ions a t 380 nm.]? A.U thiocyanate solutions as well as mixtures of thiocyanate and oxalate (pH 3 and 9.3) were saturated with nitrow oxide (2 X 1Q-3At) for scavenging of eiLs-. At each p value a t least three different ratios of concentrations were irradiated with 0.9-psec electron pulses. By plotting of the ratios of the optical densities with and without oxalic acid (respective oxalates) as a function of ihe concentration ratios of oxalic acid (respective oxalates) and thiocyanate, straight lines were obtained. From the slopes (8)of the straight lines the corresponding sate constants were calculated, i.e.
k(OH -E oxalate) = S X k(OH
-+ CNS-)
M-’ sec-1
(4)
At pH 0.5 there is about 85% of H2CzO4and 15% of HC204-, whereas at loH 3 only HCzQ4- and at pH > 6 only C2042--ii2 present. Consequently the rate constants, k(OH 13CS04-)and k(OH C202-), were Hzobtained directly, whereas the value for k(0H Cz04)was calculated from the k value a t pH 0.5 and corrected for tho contribution of HC204-. The mean vahies of several determinations are given in Table I4 for k(0H -5- GKS-) = 7 . 5 X 109M-1sec-I
+
+
+
751 difficult to say whether H abstraction or ON addition takes place. However, it is assumed that the overall reaction is
+ OH
H2C204
COO
H20
+ COOH 1
(5)
The reaction with a dissociated carboxylic group is somewhat faster than that with an undissociated one, and most probably electron transfer occurs
HCz04-
+ OH -%-
OH-
+
COO (6)
COO COO
+ OH -% OH- + 1
Cz042-
COO-
(7)
The radicals formed by the reaction of OH radicals with oxalic acid or oxalates (reactions 8-7) are denoted RB1, RB2, and RB3, respectively. 9.3. Reactions with H Atoms. For a study of these reactions solutions of 5 mM oxalic acid at pH 1.5 and 3 were irradiated under hydrogen pressure (45 atm). No appreciable absorption of short-lived species was observed in the optical region between 240 and 300 nm at pH 1.5, which confirms the very low- rate of reactions reported. l5 The addition of H may be regarded as the first step in the reaction with oxalic acid and oxalate ions
+
HzCz04 H -%
C(OWjz
I
(8)
GOOH
C(OH)z
HC204-
+ H --% 1
(9)
COO-Table 11: Rate Chnstants for the Reaction of OH Radicals with Oxalic Acid and Oxalates Ilate
k in M -1 set -1------. _-----
con-
a
Solute
si a n t
HzCzOa HCzO4G102_ .
ks
k6
k,
This work
(1.0 Irt 0.1) x 106 ( 3 . 2 i 0.1) x 107 (5 3 i 0 . 3 ) x 106
DraganiC, et aLa
5 3.5 5
x x x
106 107 106
Since, under our experimental conditions, the rate constants for these three reactions (8-10) could not be determined by direct pulse radiolysis measurements,
See ref 2.
together with literature data.4 These values, as well CNS-) = 6.6 X lo9 J1-l as those based on k(OH sec-1,19 are in fair agreement with steady-state competition measurerrtent~,~ while those based on k(OH CNS-) = 2.8 X IO1> and k(OH I-) = 3.4 X are about four 1 imes :is high. It is well known that the reaction of OH radicals with a -COO13 group is very slow, the adjacent group being very often the preferred point of attack.16 It is
+
+
+
(16) G. E. Adams, J. W.Boag, J. Currant, and B. D. Michael in “Pulse Radiolysis. Proceedings of the International Symposium,” M. Ebert, J. P. Reene, A. J. Swallow, and J. H. Baxendale, Ed., Manchester, April 21-23, 1965, Academic Press, London, 1965, pp 131-14.3. (17) J. H. Baxendale, P. L. T. Bevan, and D. A. Stott, Trans. Faraday Soc., 64, 2389 (1968). (18) C. L . Greenstock, I. 117. Hunt, and M. Ng, ibid., 6 5 , 3279 (1969). (19) G. E. Adams, J. W. Boag, J. Currant, and B. D. Michael in “Pulse Radiolysis. Proceedings of the International Symposium,” M. Ebert, J. P. Keene, A. J. Swallow, and J. H. Baxendale, Ed., Manchester, April 21-23, 1965, Academic Press, London, 1965, pp 117-131.
The Journal of Physical Chemistrg, Vol. 75, N o . 6 , 1971
GETOFF,SCHWORER, MARKOVIC, SEHESTED, AND NXELSEN
752
- -
we took the ]previously determined values for our computations, namely ks kg klo = 1 X lo6 M-' ~ e c - 1 . ~For simplification, we shall call these radicals RC1, RC2, and RC3, respectively. 3.4. Absorption Spectra and Kinetics. The attack of the primary products of water radiolysis on oxalic acid and oxalates leads to the formation of a number of intermediates. It was established that they all absorb i n the optical region from 230 to 350 nm, The radicals will be discussed in groups. The Rrl-Type Radicals. As already mentioned, these species are formed by reactions of eau- with oxalic acid and with both oxalate ions in argon-saturated solutions. Apart from the above-mentioned reactions (1-3 and 5-10) the following processes are assumed eaa-
+ ~ ~ 0 - 1H + H%O le11
H
+ R -*
2k z
=
20
Rz2k12 = 1.6 X 1O'O 111-l sec-I
H -t OM -2H20 kI3 OH 4- OH
2.3 X 1O1O M-l sec-1
=
1.0
x
(11)
(12)
1010M-1sec-16 (13)
-% 13202 2ki4
=
1.1 X 1Olo M - l sec-1
(14)
The compntatbis for this system a t pH 3 (argonsaturated solution) show that reaction 2 is almost completed durinp: thc pulse under our conditions. When the pulse ends, no significant amount of other radicals, except RA2, has been formed. The concentrations of RA2 and OH radicals a t the pulse end, computed for a system of e q I?, 6 , 9 , and 11-14 are given in Table 111.
Table 111: Computed Yields of RA2 and OH Radicals at, Pulse End for Three Oxdate Solutions (saturated with argon) a t pH 3, Pulse Length 0.5 psec, Dose 3.5 krads ----Computed RA2
values---
c. / o 1 O' H [HCzOa-], mM
1 2.5 5
OH,
Correotion for OH absorption,
P M
PM
%
1.2 2.6 4.1
9.7 9.6 9.4
Max 45 Max 25 Max 17
C'3O - *
The spectrum of RA2 radicals was obtained after correction For OH absorption and is shown in Figure 1. The results obtained with three different oxalate Concentrations agree very well within f 10%. The fate of the RA2 radical was followed in a hydrogen-saturated solution (45 atm) at p H 3. Because of the very low rate constant for the reaction of H atoms with HC,OS- practically only RA2 radicals are formed The Journal of Ph"hysical Chemistry, Vol. Y6,N o . 6,1971
Id]
II
I k' ,oi L-OH
(RA1)
?400H
1 0 2 L 200
I
1
,
,
I
,
,
,
-v---T-
300
250
350 nm
Figure 1. The absorption spectra of intermediates obtained by reaction of ea4- with oxalic acid and both oxalate ions.
under these conditions. The kinetic measurements were carried out at 260 and 280 nm. It was possible to resolve the observed decay into two consecutive processes. At the beginning, for about 80 fisec, first-order decay with a half-life TI,, = 22.5 psec, was ascribed to the reaction
0-
/
C-OH
I coo-
CO
I
-% COO-
+ OH-
1cL5
=
(3.1
f
0.3) X lo4 see-'
(15)
Assuming a protonation of the radical instead of reaction 15, one may calculate a first-order rate constant k = 3 X 10' sec-', which we consider to be too low for a protonation process. The radical formed in (15), however, most probably hydrates before recombination, but our experiments have not given enough information to be able to state which of the two reactions is the rate-determining step. The first-order decay was followed by a second-order decay of the radical ions, lasting from about 100 psec to about 2.5 msec after the end of the pulse
coo-
coo-
(16)
2kie _-
- 3.1 X lo6 em sec-l
€280
On the assumption that no significant decay of the -OOC-6(OH)z radical occurs before completion of reaction 15, the following molar extinction coefficients at 260 and 280 nm were calculated. (20) S. Gordon, E. J. Hart, M. S. Matheson, J. Rabani, and J. K. Thomas, Discuss. Faraday Soc., 36, 193 (1963).
PULSE RADIOLYSISOF OXALICACIDAND OXALATES
and eZa0 =
1
1.65 X IO3 114-1 cm-l
This gives 2kI6
=
5.1 X IO8
sec-l
The spectrum 0 ; the RA3 radical (reaction 3) was obtained from argon -saturated C20d2- solutions (5 and 10 mM) a t p H 9.4. The spectra obtained were corrected for OH and RB3 absorption. The concentrations of these two radicals were computed at a time 10 psec after the pulse end, and the reactions 3, 7 , 10, 12, 13, 14, and also the following equations were taken into consideration.
I
+ e & ---+
10
2kl7
eaq-
eaq-
k18
2.5 X 10'0 M-1 sec-I
Figure 2 . The absorption spectra of intermediates obtained by reaction of O B radicals with oxalate ions.
21
(18)
bination of the RA3 radical only, in aceordanec with the reaction
kIs, = 3 X 1Olo M - I 9ec-I z1
(19)
The computed vadues are given in Table IV. The molar extinction coefficients of RA3 for different wavelengths are given in Figure 1.
Table IV : Computed Yields of RA3, RB3, and OH Radicals 10 psec after the Pulse for Two Concentrations of Oxalate Solutions (saturatecl with argon) at pH 9.4, Pulse Length 0.5 psec, Dose 3.5 ltrads -- Computed TaIues-----
7--
mM
5 19
606
coo-,
coo -,
I
EB3
I" &l
fiM
2.7 3.8
1.4 2.2
0% I"M
2.0
1.7
350 nm
(17)
+ OH -2-o
R.43 C002
300
250
21
+ H -2.-El* +- 0
[Cz042-1,
--?----
200
2kI7 = 9.0 X lo9 M-' sec-I
eaq-
i
Total correction for OH and RB3 absorption,
%
Max 50 Max 35
l _ l _ _
The species formed a t pH 9.4 in argon-saturated Cz042- solutions follow approximately second-order decay with slope X = $3 X 1Q4cm sec-' at the beginning and X == 6 [IO4 cm sec-l after a few hundred microseconds, These slops's, however, are followed for a relatively small change in optical densities (less than 2 half-lives in both cases), and they are considered less accurate than the other kinetic data given in this paper. Eere, both radicals, RA3 and RB3, are present at the beginning The slxond radical decays much faster than the first (see later) and has a much smaller molar extinction coefficient a t 280 nm (Figure 2). Therefore the second slope probably corresponds to the recom-
CO02-
2
I
coo-
2'.2p_ products 2k20 5 2 X lo8&[-I sec-I
(20)
An attempt has been made to determine the absorption spectrum of RA1 radicals a t pH 2 (16% € G O 4 and 84% HC204-). In this case the corrections for OH and RA2 radicals are too large (SO-SO%). The spectrum is, however, presented in Figure 1 for comparison. The 6 values are probably too low, and the spectrum may be identical with that of the RA2 radical An alcohol as OH scavenger might have improved the detwminations. The RB-Type Radicals. These radicals are formed as a result of the attack of OH radicals on oxalic acid and both oxalate ions. The determination of the absorption spectrum of the RB1 radical is difficult. Bowever, in writing reactions 5 and 6 it was assumed that RB1 and RB2 are identical. The absorption spectrum of RB2 \vas obtained for solutions containing various concentrations (1 -10 mM) of HC2Os- in N20-saturated solutions a i pR 3. I n this case there is competition between XzO and HGZ0,- for eaq-. Rfost of the eaq-, however, are transformed into OH radicals according to the reaction eaq-
+ x20-% x2+ OH + OHkzl
=
5.6 X lo9 n4-l sec-'
22
(21)
Therefore, only a small amount of RA2 radicals is formed. The corrections for RA2 and OH radicals (21) M. 5. Matheson and J . Rabani, J. P h y s , Chem., 69, 1324 (1965). (22) J. P. Keene, Radiat. Res., 2 2 , 1 (1964).
The Journal of Physical Chemistry, Vol. 76, N o . 0 , 1971
GETOFF,SCHWORER, MARKOVIC, SEHESTED, AND NIELSEN
754 were computed for the system of eq 2, 6, 9, 11-14, and 2 1. The results are given in Table V. The absorption spectrum of the RB2 radical (Figure 2 ) was obtained from various concentrations of HCz04corrected according t O the data in Table V. Here again very good agreement was obtained for several different concentrations. Only mean values are presented. 'They were obtained with about f10% standard deviation.
sorption (minimum on curve A, Figure 3) is shifted to shorter times with an increase of the oxalate concentration (Figure 3, curve B). The reaction of radical RB2 with oxalic acid, yielding radical RC1, is suggested as an explanation of the yield of dihydroxytartaric acid and COz measured in deaerated solutions of oxalic acid at p H 1 and 3 . 2 3
A t pN 1
COOH
COO
Table V : Computed Yields of R.B2, RA2, and OH Radicals a t Pulse End for Several Concentrations of XgO-Saturated Oxalate Solutions at pH 3, Pulse Length 0.5 psec, Dose 3.5 kradc; ,----Computed COCI FtH2
I
1
PJ4
OH,
Correction for OH 5nd RA2 absorption,
%
filv
3 6.4 10.1 13 4
1 2 5 10
OH
GOO-,
COOH,
[ECzO4-], mM
values--RA2
1.5 0.3
---____^_I
The kinetics of these radicals are complex (Figure 3). At pH 3 the CBI-I radicals are practically consumed within about 20 paec. Figure 3 shows for two oxalate concentrations the second-order deca,y, starting after a few hundred microseconds. At the beginning an increase of optical absorption is observed, followed by a subsequent decrease, which ind LCBtes the transformation of radicals with lower molar extinr: tion (coefficient into radicals with higher coefficient. The position of the maximum of the ab-
(16) Figure 3 gives 2k16 - - - 1.6 X lo5 cm sec-* E270
With €270 5 2100 AI-l cm-1 (see section 3.4) the following rate constant was calculated 2kI8 5 3.4 X lo8 M - * see-l
This value is lower than 5.1 X lo8 calculated earlier (see section 3.4) probably due to the interference of reaction 22 on the decay. Before complete decay of the radical. RB2 its reaction with RC1 also starts to contributez3 COO
C(0H)z
I + I COOH COOH
0
05
10
15 msec
Figure 5. Mired- and second-order decay at h 270 nm of radicals formed in NzO-saturated oxalate solutions at pH 3: A, (0) 2.5 mM HC&4-; B, ( 0 )10 mM HCsOa-. The J O U Tof~Phu$szca,l ~ Chemistry, Vol. 76,No. 6,1971
(22)
This reaction may explain the first part of the curves in Figure 3. The RC1 radical is the hydrated form of the CO.CO0- radical ion and may be assumed to have the same molar extinction coefficient. Hence the RB2 radical is transformed into a radical with about twice the molar extinction coefficient at 270 nm. The second-order decay, then, corresponds to the recombination of RC1 radicals
Max 38 Max 26 Max 26 Max 26
4.3 3.3
0.14 0.3 0.7 1.3
C(OH)z
+ COOH I 3-2CO2+ I COOH COOH I
-
2C02
GHO
+ H20 + 1
cool3
(23)
The influence of oxalic acid concentration on the first part of the curves in Figure 3 supports reaction 22. This reaction is quickly masked by reactions 16 and 23, and thus the accurate determination of lcZzis prevented. An estimate, however, can be made, giving an upper limit to lc32 of a few times lo6M-' see-l. The rate constant for the recombination of RB2 radicals seems to be much lower than 10' M-' see-I, and this reaction can probably be neglected in the presence of oxalic acid in concentrations above I mM. By radiolysis of a Cz042- solution saturated with P I T 2 0 a t pH 9.4, only RB3 radicals are produced. Corrections for the OH radicals have been computed as before (23) V. Markovic, K. Sehesded, and E. Bjergbakke, unpublished work.
PULSE RADIOLYBIS OF OXALIC ACIDAND OXALATES for the system of eq 7, 10, 12, 13, 14, and 22. For a 5 m N Cz042-solution {saturated with N20 the following values were calculated a t 10 psee after the pulse end: [OH] = 5.12 plM and [RB3] = 3.34 p M . From the measured absorption spectrum of RB3, corrected for OH radicals, the corresponding molar extinction coefficienis, shown in !Figure 2, were calculated. The correction for the OB absorption is in this case below
25%. The kinetics of decay showed no indication of the deviation from the second order that was observed at p H 3. This probably means that the RB3 radical hardly reacts with oxalate. A pure second-order decay of the GO4--radical ion was observed
eo0 2
1
COO-
% products
(24)
with
and 2lcZ.i = 9.6 X los M-I sec-l
It might be of interest to compare this value with the corresponding data for the COZ- radical ion and Cot-) = 1 X the COOH radical. For 2k(CO2lo9 M--I sec-l 2 4 we calculate 2k/€280 = 4.2 X lo6 cm sec-’ and for 2k(COOH COOH) = 4 X 109 M-I see-l, 2k/czso = 3 X lo6 cm S ~ C - ’ . ~ ~
+
+
755 From this comparison it can be concluded that reaction 25
coo
I
--tCO2+COOH (25) CQOH does not take place at pH 3, and probably not a t p H 9. The RC-Type Radicals. The reactions of oxalic acid and both oxalates with H atoms, as discussed , ~ thereabove, are very slow (IC 2 lo6 M-I ~ e c - 9 and fore the determination of the corresponding absorption spectra is very difficult. No appreciable absorption could be detected in 5 mM oxalic acid under 45 a t m of H2 at pH 1.5. 5.6. Peroxy Radicals. A few series of pulse radiolysis experiments were carried out with oxygen present and with various concentrations of oxalic acid and oxalates a t different p H values. The absorption spectra of the short-lived species are rather complex. Further investigations of these intermediates are necessary in order to distinguish between the various peroxy radicals. Aclcnowledgments. We wish to thank 0. L. Rasmussen for computer calculations. The help given by the accelerator staff a t Riso and the technical nssistance by P. L. Genske and nil. Will(: are gratefully acknowledged. Two of us (N. G. and V. M.) thank the Danish Atomic Energy Commission for the fellowships granted. (24) J. E’. Keene, Y. Raef, and A. J. Swallow in “Pulse Radiolysis. Proceedings of the International Symposium,” M . Ebert, J. P. Keene, A. 6.Swallow, and J. H. Baxendale, Ed., Manchester, April 21.-23, 1965, Academic Press, London, 1965, pp 99-106. (25) N. Getoff, 13. D. Michael, and E. J. Hart, unpublished results.
The Journal of Physical Chemistry, Vol. 76, N o . 6, 1971
