Mechanism of reduction of chromium(VI) ion by 2-mercaptosuccinic acid in aqueous solution

Article abstract of DOI:10.1016/S0020-1693(01)00804-0

The kinetics of the reduction of the chromium(VI) ion by 2-mercaptosuccinic acid (thiomalic acid) were studied by rapid scanning stopped flow spectrophotometry. The conditions used were [Cr(VI)]_T=0.20 mM, [MSA]_T=5-90 mM, 3.0≤pH≤5.6 i

Full text of DOI:10.1016/S0020-1693(01)00804-0

www.elsevier.com/locate/ica
Inorganica Chimica Acta 331 (2002) 178–187
Mechanism of reduction of chromium(VI) ion by
2
-mercaptosuccinic acid in aqueous solution
Dwight C. Ramdon, Donovan A. Dixon, Tara P. Dasgupta *
Department of Chemistry, Uni6ersity of the West Indies, Mona Campus, Kingston 7, Jamaica
Received 9 July 2001; accepted 11 October 2001
Dedicated to Professor A.G. Sykes
Abstract
The kinetics of the reduction of the chromium(VI) ion by 2-mercaptosuccinic acid (thiomalic acid) were studied by rapid
scanning stopped flow spectrophotometry. The conditions used were [Cr(VI)] =0.20 mM, [MSA] =5–90 mM, 3.05pH55.6
T
T
in citric acid–phosphate buffer, or 3.35pH55.4 in 0.40 M acetic acid–acetate buffer, 20.05T535.0 °C at I=0.50 M
NaClO ). Spectrophotometric titration at 350 nm indicates the stoichiometry of the reaction to be 1:3. The kinetics of both
(
4
formation and decay of the intermediate chromium(VI) thioester were followed at umax=425 nm and rate expressions, specific
rate constants and corresponding activation parameters were derived from the proposed mechanism. The acetic acid–acetate
buffer was found to catalyze the formation but not the decay rate of the intermediate. The citric acid–phosphate buffer and
dissolved oxygen did not have any significant effect on the reaction rates. The justification of the mechanism was discussed in
terms of standard biological conditions. © 2002 Elsevier Science B.V. All rights reserved.
Keywords: Chromium(VI); 2-Mercaptosuccinic acid; Chromium(VI) thioester; Buffer catalysis; Rate equations; Activation parameters
1
. Introduction
the reduction of Cr(VI) to Cr(III) by these cellular
reductants, appear to be responsible for the DNA
damage in cells [6].
Chromium is used widely in industries as a curing
agent for polysulfide pre-polymers, chrome plating, pig-
ment production, leather tanning [1] and as a synthetic
oxidizing agent in laboratories [2]. In the early 1950s it
was found that workers in chromium based industries
were more susceptible to lung and skin cancers [3] and
many investigations have subsequently been conducted
in an attempt to understand how chromium induces
cancer [4]. Cr(III) is considered to be a trace nutrient in
Generally, thiol containing compounds (–SH func-
tional group present) and ascorbic acid, are the active
reductants towards Cr(VI) ion under physiological con-
ditions [7]. As a result, most of the work done on
Cr(VI) have been confined to physiological pH of 7.4,
with glutathione being extensively used as reductant,
since it has been found to be the most abundant cellular
thiol [6,8,9]. Our previous work in this area was also
centered around physiological pH [10–13]. Some
amount of work has also been done with other thiols
under similar experimental conditions and in acidic
media (pH 1–2) [14–16].
2
−
mammals [5]. Cr(VI) in the form of CrO
can enter
4
2
−
cells by SO4
transport systems and does not interact
with DNA in a harmful way [1]. However, it is the
interaction of Cr(VI) with cell components such as
cysteine, glutathione, penicillamine and ascorbic acid
that have been found to be harmful to DNA molecules
Even though physiological pH is generally stated as
7
.4, there are a few areas within the human body where
[4]. Reactive species such as radicals produced during
acidic conditions exist, for example in the kidneys and
stomach [17]. Thermodynamically, from electrode po-
tential data, Cr(VI) is a stronger oxidizing agent under
acidic conditions [18]. Kinetic studies over an acidic pH
range should then have relevant information. Prelimi-
*
Corresponding author. Tel.: +1809-977-1834; fax: +1809-977-
835.
E-mail address: tara@uwimona.edu.jm (T.P. Dasgupta).
1
0
020-1693/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved.
PII: S0020-1693(01)00804-0

D.C. Ramdon et al. / Inorganica Chimica Acta 331 (2002) 178–187
179
nary reactions of Cr(VI) with various reductants show
that over the pH range 3–6, experimental rate con-
stants change significantly with change in acidity and
this range is therefore rich in kinetic information.
Within this range, reductants containing carboxylate
groups are deprotonated, and reactivity of this func-
tional group may be evaluated. If a charge difference
arises from speciation of the reductant, it is possible to
assess whether or not this is an important factor affect-
ing its overall reducing capabilities.
Taking these factors into account, we have embarked
on an intense study of the reactions of Cr(VI) and some
sulfur containing substrates over the mildly acidic pH
range of 3–6. The dicarboxylic reductant, 2-mercapto-
succinic acid (MSA) was one of the thiols chosen since
there are three species existing in solution with different
charges over this pH range. It may exist as double or
singly protonated form, or as the dianion having the
two carboxylates deprotonated [19]. The –SH func-
tional group of the latter species remains unionized.
This can also be used to evaluate the effect of proton
donors on this particular reaction.
Ps system, generated deionized water used for all exper-
imental work.
Solutions of MSA buffered in citric acid–phosphate
at pH 2.98 and 5.11 at ionic strength of 0.50 M
(NaClO ) and [MSA] =51 mM, were stored at room
4
T
temperature for 3 days then analysed. Recoveries in
excess of 95% of that expected from the mass of MSA
were obtained. A similar set of solutions was saturated
with ultra-high purity (UHP) argon for 30 min, and
gave recoveries \95% expected MSA concentration
when standardised. Solutions of Cr(VI) and MSA in
citric acid–phosphate buffer stored separately
overnight at 4 °C, gave similar rate constants when run
at the same temperatures as the freshly prepared ones.
Only freshly prepared solutions of MSA were used for
acquiring kinetic traces.
2.2. Stoichiometry
Spectrophotometric titrations were accomplished by
periodic monitoring of the Cr(VI) peak at u=350 and
65 nm until the absorbance change was negligible.
3
MSA is an important chiral multi-functional interme-
diate in organic chemistry and has been widely em-
ployed in the synthesis of various biologically active
sulfur containing compounds such as antileukemic,
spiro[indoline-3,2%-thiazolidine]-2,4%-diones [20], and the
anti-microbial [21], and antitubercular [22], 4-thiazolidi-
nones. More recently, it has been used in the synthesis
of novel polyanionic inhibitor of HIV and other viruses
These were performed in both acetic acid–acetate and
citric acid–phosphate buffers, but did not give repro-
ducible data at pH 3.25, u=350 nm. However, those
conducted at pH 5.36, u=365 nm gave reproducible
data in citric acid–phosphate buffer. A plot of this
steady absorbance against [MSA] /[Cr] gave a reason-
T
T
able break point. Also, an attempt at employing Ell-
man’s reagent to complex with MSA in these solutions
in order to monitor the concentration of the unreacted
portion, was carried out with some success.
Stoichiometric studies were also done by reacting
standard Cr(VI) with excess standard MSA. The reac-
tion was allowed to go to completion and the unreacted
[23].
Recent reports have suggested that the reaction be-
tween Cr(VI) and thiols vary greatly with pH [24]. The
present study therefore attempts to evaluate such inter-
actions utilizing the thiol, MSA, with emphasis on
spectral intermediates, factors governing their forma-
tion and decay rates, as well as activation parameters
involved in these reactions. This can then be used to
extrapolate the kinetic behaviour under physiological
conditions.
MSA was determined by titration with KIO in the
3
presence of excess H SO and KI [27]. CCl was added
2
4
4
to make the end point more visible in the presence of
Cr(III) product formed. Although the concentration of
Cr(VI) used for the reaction was ꢀ5 mM compared to
0
.20 mM in the kinetic studies, the results were never-
theless in good agreement with the spectrophotometric
titration data.
2. Experimental
2.1. Materials
2.3. Kinetics
All reagents were of analytical grade and supplied by
Freshly prepared solutions of MSA and Cr(VI) were
either Aldrich or BDH chemical companies. Na CrO
2
4
made in buffer and the ionic strength adjusted to 0.50
M with NaClO4 as electrolyte. The kinetics of the
reaction was studied in both acetic acid–acetate and
citric acid–phosphate buffers on a HI-TECH Scientific
SF61 DX2 stopped-flow unit, by observing the decay of
the Cr(VI) peak between 350 and 365 nm (depending
on pH), as well as formation and decay of an interme-
diate species which has absorbance maxima at 425 nm.
All reactions were monitored for at least three half-
(
GPR) was obtained by recrystallization of the salt
using de-ionized water to give Na CrO ·4H O, solu-
2
4
2
tions of which were standardised spectrophotometri-
cally [7], and by titrating with ferrous ammonium
sulfate [25], then stored at 4 °C for future use. The
purity of MSA was determined by Ellman’s method
[26] and titration with standard KIO [27]. Distilled
3
water was made from an all glass FiStream cyclon
distiller. Feeding this through a Labconco water Pro™

180
D.C. Ramdon et al. / Inorganica Chimica Acta 331 (2002) 178–187
lives. These solutions were stored at 4 °C between
preparation and time of use, after which they were
equilibrated for 10 min to achieve the desired tempera-
ture prior to acquiring the time-based traces. The tem-
perature of the reaction solutions were maintained at
was formed more rapidly at higher [MSA] and lower
T
pH. Under such conditions, the rapid scanning tech-
nique was employed and the band at 425 nm depicting
the intermediate, appears to overlap with the Cr(VI)
peak at 350 nm to give the tri-phasic trace seen in Fig.
3, inset. The form of the traces at 425 nm remains the
same as that in Fig. 2 inset, but the resulting band has
significant absorbance at 500 nm (75 nm greater than
425 nm). Assuming this peak is symmetrical about 425
nm, then at 350 nm (75 nm less than 425 nm) signifi-
cant absorbance is also expected (taking the base line of
0.05 in account). Therefore, due to this overlap, the
kinetic studies were conducted primarily at 425 nm.
Fig. 3 also shows that there is an intermediate band
below 300 nm. The absorbance seen at 350 nm may not
be representative of only unreacted Cr(VI), but possibly
a combination of the unreacted Cr(VI) and an overlap-
ping portion of the intermediate band within this
region.
9
0.1 °C using a circulating water bath from Poly-
Science. The pH of the reactions were measured after
mixing manually Cr(VI) and thiol solutions in a 1:1
ratio using an expandable ion analyzer from Orion
Research and a glass electrode. Argon saturated solu-
tions were used to evaluate oxygen participation.
Attempts at fitting the formation and decay of the
intermediate at 425 nm to a double exponential equa-
tion over the entire pH range were quite successful.
Traces for the decay of Cr(VI) fitted a triple exponen-
tial equation at low pH, but in each case, two of the
three rate constants obtained from the triple exponen-
tial equation were essentially the same. However, decay
of Cr(VI) at high pH was single exponential with the
rate being identical to the decay of the intermediate
observed at 425 nm. Repetitive scans were analysed
with SPECFIT/32 Global analysis system (2000, ver.3).
The psuedo first-order rate constants obtained were
fitted to rate expressions using Statgraphics non-linear
regressional program (1999, ver. 4).
3. Results
3.1. Nature of the reaction
The reaction of Cr(VI) with MSA over the pH range
in citric acid–phosphate, acetic acid–acetate or by
making use of the self-buffering capacity of MSA,
rapidly produces a red–brown intermediate with
umax=425 nm, at the expense of Cr(VI) at 350 nm. Fig.
Fig. 1. Absorbance vs. wavelength for formation of Cr(VI)/MSA
intermediate at cycle time=4 s, T=25.0 °C, pH 5.22 (self-buffer),
[MSA] =15 mM. [Cr(VI)] =0.20 mM.
T
T
1
shows the repetitive scan leading to formation of this
intermediate up to its maximum concentration. Two,
though not very sharp, isosbestic points were main-
tained up to this time at u=327 and 397 nm. The
intermediate then decays to the Cr(III) product with its
characteristic low absorbance (umax=429 and 590 nm,
both at Abs ꢀ0.025) as shown by the repetitive scan in
Fig. 2. There is also a noticeable growth in absorbance
at 260 nm. The inset shows the absorbance vs. time
profile obtained from repetitive scans at u=360 and
425 nm representing Cr(VI) and intermediate, respec-
tively. The pseudo first-order rate constants, kobsF and
kobsD, denote formation and decay of the intermediate,
respectively. Both kobsF and kobsD were obtained by
fitting the acquired traces to double exponential equa-
tions over the entire time of acquisition. At higher
concentrations of Cr(VI) where more Cr(III) is formed
at the end of the reaction, a well defined spectra of the
Cr(III) product can be obtained.
Fig. 2. Absorbance vs. wavelength for the decay of Cr(VI) and
Cr(VI)/MSA intermediate at cycle time=20 s, pH 5.22 (self-buffer),
[
MSA] =15 mM, [Cr(VI)] =0.20 mM. Inset: Absorbance vs. time
T T
A larger maximum concentration of the intermediate
plots at u=425 and 360 nm.

D.C. Ramdon et al. / Inorganica Chimica Acta 331 (2002) 178–187
181
with excess MSA at pH 2.62 gave a green solution as
the final product, showing umax at 423 and 574 nm with
absorbances of 0.177 and 0.162, respectively. A major-
ity of this adsorbed to a Dowex 1-X8, 100–200 mesh
anion column. This was washed with deionized water
then eluted with 2 M NaCl, giving a solution of pH
3
.95 and umax=434 and 604 nm, both with absorbances
of 0.254. Attempts at crystallizing this eluted anionic
portion only produced the NaCl which was formerly
used to elute the anionic Cr(III). The other Cr(III)
species which were not anionic appeared to be only a
very small fraction of the total Cr(III). Some re-
searchers have been able to isolate the disulfide of a few
thiols [28,29]. Attempts at isolating the disulfide of
MSA in this reaction using the same methods were
unsuccessful.
3.4. Effect of buffer
Fig. 3. Spectra showing unreacted chromate, the intermediate at
maximum concentration, and the final Cr(III) product using rapid
scanning stopped flow at pH 3.18 in acetic acid–acetate buffer,
Reports have suggested that Cr(VI) is unstable in
hydrochloric and sulfuric acids, but stable in acetic,
nitric and perchloric acids [30]. Perez-Benito et al.
[
MSA] =15 mM. Inset: corresponding absorbance vs. time traces at
T
[
31,32] have done extensive studies on the participation
u=425 and 350 nm.
of buffers on Cr(VI) reactions with glutathione [31] and
penicillamine [32]. They found that buffers do partici-
pate in thioester formation increasing in catalytic power
in the order acetateBcitrateBphosphate. As a result,
acetic acid was chosen as the buffering media and was
found to be better than citric acid–phosphate for the
Cr(VI)/S(IV) system [33], with no catalytic effects being
observed from the former buffer. The Cr(VI)/MSA
system appears to be different as there is a significant
dependence on the acetate buffer. Table 1 shows the
pseudo first-order rate constants and the catalytic effect
produced by the presence of the acetic acid–acetate
buffer. Fig. 4 shows that kobsF is related linearly to the
3
.2. Stoichiometry
Spectrophotometric titrations performed in both
buffers did not give reproducible data at pH 3.25. It has
been reported that the reactions of thiols differ in
stoichiometric excess of thiol compared to excess oxi-
dant concentration [27]. The nature of the reactions of
thiols are not fully understood in stoichiometric excess
of oxidant. The oxidized form of the thiol is usually the
corresponding disulfide. Depending on the nature of the
oxidant, and if it is in stoichiometric excess, further
oxidation products are possible [27].
concentration of acetic acid–acetate buffer, [AcO] , at
T
Similar titrations performed at pH 5.36 and
fixed pH and total concentration of MSA, [MSA] . A
T
[Cr(VI)] =0.40 mM gave reproducible data in citric
T
similar trend was observed at pH 3.29 and 5.24, with
the least effect of the acetic acid–acetate buffer being
seen at pH 3.29.
acid–phosphate buffer with a break point at ꢀ3.0.
Subsequent complexing of the unreacted MSA from
these solutions with Ellman’s reagent [26] also produced
a break point at ꢀ3.0. Mixing standard solutions of
Cr(VI) and MSA (in the absence of buffer and elec-
trolyte), followed by back titration of the unreacted
Table 1
Effect of acetic acid–acetate buffer concentration, [AcO]
pseudo first-order rate constants, kobsF and kobsD, at 25.0 °C, pH
, on the
T
4
.35, [MSA] =20 mM, [Cr(VI)] =0.20 mM, u=425 nm, I=0.50
T
T
MSA with KIO [27], showed that each mole of Cr(VI)
3
M (NaClO4)
consumes approximately 3.0 mol of MSA. It should be
noted that the concentrations used in the titrimetric
method ([Cr(VI)]$5 mM) are much higher than those
utilized in the kinetic runs ([Cr(VI)]=0.20 mM) even
though there was good agreement between both.
[AcO]T (M)
kobsF (s−1)
k
obsD
(s−1)
0
0
0
0
0
0
.10
.20
.30
.40
.50
.60
0.229
0.311
0.390
0.467
0.537
0.608
0.0948
0.0975
0.0943
0.0942
0.0939
0.0940
3.3. Product studies
The Cr(III) product was found to be anionic based
on ion exchange studies. Mixing unbuffered Cr(VI)
Errors were all less than 1%.

182
D.C. Ramdon et al. / Inorganica Chimica Acta 331 (2002) 178–187
using citric acid–phosphate buffer instead, produced no
clearly defined trend for either kobsF and kobsD, as seen
in Table 2. In fact, the acetate independent rate con-
stant (intercept of Fig. 4) was fairly close to the pseudo
first-order rate constants obtained in citric acid–phos-
phate buffer. It implies that the experimental rate con-
stants obtained in citric acid–phosphate buffer are
fairly reliable, and more accurately represent the true
MSA dependent rate constants. Reactions were never-
theless conducted in both buffers and a comparison of
the second-order rate constants was made.
3
.5. Effect of [MSA] and pH
T
Both pseudo first-order rate constants, kobsF and
Fig. 4. Effect of acetic acid–acetate buffer concentration, [AcO] , on
T
the pseudo first-order rate constant for the formation of the interme-
diate, kobsF, at 25.0 °C, pH 4.35, [MSA] =20 mM, [Cr(VI)] =0.20
kobsD, vary linearly with [MSA] and are unaffected by
T
T
T
the presence of dissolved oxygen as observed from data
collected in argon saturated solutions (Table 3). This
linearity is also maintained at higher pH values as
evident in Fig. 5, but the values for kobsD are of a lower
magnitude when compared to kobsF. Increasing pH
leads to a decrease in the pseudo first-order rate con-
mM, u=425 nm, I=0.50 M (NaClO4).
Table 2
Effect of citric acid–phosphate buffer concentration on the pseudo
first-order rate constants, kobsF and kobsD, at 25.0 °C, pH 4.35–4.63,
[
MSA] =20 mM, [Cr(VI)] =0.20 mM, u=425 nm, I=0.50 M
T
T
(NaClO4)
stants over the range 5.05[MSA] 545 mM, 2.915
T
pH55.60, in citric acid–phosphate buffer (Table 4). A
lowering of the pH in acetic acid–acetate buffer also
produced a similar increase in the pseudo first-order
rate constants. However, the values of kobsF in acetic
acid–acetate buffer were higher than the corresponding
values in citric acid–phosphate (Fig. 6), except at the
lower end of the pH range. Acetic acid showed no
significant catalysis of the reaction, but it appears that
the acetate anion acts as the effective catalyst. The
speciation of MSA in solution over the pH range is
represented by Scheme 1, along with the relevant acid
dissociation constants [34,35]. There is some disagree-
ment concerning these values [19,36,37], but they were
reported by two independent groups and are in fairly
pH
102
phosphate]T
M)
102 [citric
acid]T (M)
k
kobsD (s−1
)
obsF
(s−1)
[
(
4.35 2.50
4.39 5.00
4.48 7.50
4.53 10.0
4.63 12.5
4.63 15.0
1.25
2.50
3.75
5.00
6.25
7.50
0.170
0.162
0.159
0.176
0.179
0.186
0.0703
0.0686
0.0680
0.0618
0.0598
0.0582
The kobsD values showed no dependence on the con-
centration of this buffering system at all pH values.
Conducting the reactions under the same conditions
Table 3
Effect of [MSA]T on pseudo first-order rate constants for the formation, kobsF, and decay, kobsD, of the intermediate at T=25.0 °C, pH 2.96 and
4
.38 in citric acid–phosphate buffer, u=425 nm, I=0.50 M (NaClO ), [Cr(VI)] =0.20 mM
4
T
[
MSA]T (mM)
pH 2.96 a
pH 4.38 a
pH 4.38 b
kobsF (s−1)
kobsD (s−1
)
kobsF (s−1
)
kobsD (s−1
)
kobsF (s−1
)
kobsD (s−1
)
5
0
5
0
5
0
5
0
5
0.300
0.553
0.802
1.07
1.30
1.62
1.86
2.15
2.41
0.0784
0.152
0.218
0.291
0.359
0.409
0.646
0.517
0.607
0.0653
0.116
0.161
0.219
0.270
0.320
0.370
0.431
0.0151
0.0337
0.0560
0.0754
0.0960
0.122
0.0677
0.111
0.0178
0.0400
1
1
2
2
3
3
4
4
0.228
0.276
0.314
0.370
0.435
0.0853
0.107
0.148
0.160
0.166
0.147
0.166
a
Aerobic.
Argon saturated solution.
b

D.C. Ramdon et al. / Inorganica Chimica Acta 331 (2002) 178–187
183
Fig. 5. Pseudo first-order rate constant for the formation of the
intermediate, kobsF, vs. [MSA]T at T=25.0 °C, pH 2.96 and 4.38 in
citric acid–phosphate buffer, u=425 nm, I=0.50 M (NaClO4),
Fig. 6. Effect of pH and buffer on the pseudo first-order rate constant
for the formation of the intermediate, kobsF, at 25.0 °C, [MSA] =20
T
[
Cr(VI)] =0.20 mM.
T
mM, I=0.50 M (NaClO ), u=425 nm, [Cr(VI)] =0.20 mM.
4
T
close agreement with each other. They also possess
more reasonable errors and were obtained under highly
controlled conditions.
on the participation of metal ions on Cr(VI) reactions
2
+
with a series of thiols. Zn
was found to be a true
catalyst only in acetic acid–acetate buffer for the for-
2
+
2+
3.6. Effect of Zn
mation of thioesters. We found that Zn
significantly
catalyses both the formation and the decay of the
2
+
Perez-Benito et al. [32,38] have done extensive studies
thioester over the range [Zn ]=0–1 mM in acetic
Table 4
Effect of pH on pseudo first-order rate constants for formation, kobsF, and decay, kobsD, of intermediate at 25.0 °C, u=425 nm, [MSA] =20
T
mM, I=0.50 M (NaClO ), [Cr(VI)] =0.20 mM
4
T
Citric acid–phosphate buffer
Acetic acid–acetate buffer
pH
.99
kobsF (s−1)
kobsD (s−1
)
pH
kobsF (s−1
)
kobsD (s−1
)
2
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
5
5
5
5
5
1.03
1.00
0.80
0.261
0.233
0.213
0.195
0.178
0.164
0.147
0.138
0.142
0.116
0.104
0.0916
0.0828
0.0747
0.0690
0.0473
0.0430
0.0374
0.0338
0.0290
0.0252
0.0212
0.0185
3.30
3.31
3.37
3.42
3.46
3.51
3.53
3.60
3.64
3.81
3.91
4.02
4.13
4.28
4.41
4.48
4.55
4.65
4.74
4.95
4.99
5.10
5.24
0.891
0.888
0.869
0.851
0.791
0.806
0.789
0.765
0.754
0.710
0.657
0.627
0.582
0.544
0.491
0.452
0.396
0.361
0.318
0.268
0.227
0.193
0.162
0.236
0.229
0.218
0.212
0.221
0.201
0.189
0.179
0.173
0.161
0.145
0.132
0.118
0.109
0.102
0.0990
0.0794
0.0713
0.0645
0.0591
0.0509
0.0459
0.0418
.05
.17
.31
.43
.55
.65
.78
.87
.82
.03
.13
.25
.34
.44
.59
.73
.84
.94
.03
.25
.35
.47
.56
0.948
0.883
0.806
0.719
0.641
0.555
0.484
0.542
0.391
0.347
0.287
0.249
0.205
0.190
0.136
0.115
0.0939
0.0779
0.0581
0.0492
0.0405
0.0330

184
D.C. Ramdon et al. / Inorganica Chimica Acta 331 (2002) 178–187
acid–acetate buffer. Catalysis of the formation process
was more significant than the corresponding decay.
This catalytic effect is removed when the reactions are
done in the presence of the chelating agent, EDTA,
2
+
provided [EDTA]\[Zn ]. Pseudo first–order rate
constants are unaffected by EDTA in the range
2
+
[
EDTA]=0–4 mM, with no Zn
added. This would
2
+
imply that any contaminating amount of Zn
in the
reaction medium from buffer or electrolyte is negligible
and the catalytic effect seen from the buffer is a genuine
property of acetic acid–acetate. No significant amount
2
+
of catalysis was found by Zn
in citric acid–phos-
phate buffer for either pseudo first-order rate constants
under similar conditions.
Scheme 2. Proposed mechanism for the reaction between Cr(VI) and
MSA.
4. Discussion
product [39]. However, stoichiometric studies on the
Cr(VI)/cysteine system had predicted Cr(VI):cysteine
ratio to be 1:5, because three molecules of cysteine
reduce Cr(VI) and the other two coordinate to the
Cr(III) [40]. It was reported that the Cr(III)–S coordi-
nation is responsible for the increase in absorbance at
Based on the kinetic measurements, Scheme 2 has
been proposed as a possible mechanism, which seems to
be consistent with the observed stoichiometry where 3
mol of MSA reduce 1 mol of Cr(VI). This implies that
the reduced form of MSA may not be coordinated in
the final Cr(III) product, but it is likely that the oxi-
dized form is the one coordinated. Previous reports of
the Cr(VI) and Cr(III) reactions with cysteine have put
forward the argument that the reduced form of the
thiol is coordinated via sulfur to Cr(III) [8], and this
was confirmed by a crystal structure study of the
2
60 nm [8,41]. This large increase in absorbance at 260
nm was in fact observed with the Cr(VI)/MSA system
under self-buffering conditions. From the stoichiomet-
ric ratio of 1:3 obtained, this would not be consistent
with a Cr(III)–S interaction, but we continue to investi-
gate the possibility. One solution is that these bonds are
readily hydrolyzed near physiological pH [42]. Hence,
under the conditions which the stoichiometric determi-
nations were performed, evidence for Cr(III)–S bond-
ing may not be seen.
Some groups have detected disulfides either by
HPLC [43] or by extracting in ethanol, followed by its
precipitation [28]. We have been unable to isolate the
disulfide of MSA, which may be represented as
(
MSA) . This could be on account of the four carboxy-
2
late groups making it highly soluble and enabling it to
function as a tetradentate ligand for Cr(III). If Cr(III)
is coordinated to the disulfide of MSA then an overall
uni-negative charge on the product would result.
The reaction proceeds via the formation of a chromi-
um(VI) thioester intermediate, that can be spectrally
detected with umax=425 nm and denoted as Cr(VI)–
MSA. The Cr(VI)–MSA intermediate then undergoes
slow bi-molecular decay, and is in fact the overall rate
determining step in the reaction. To date, only one
+
−
3
thioester, [Ph As ][p-BrPhSCrO ], has been isolated
4
and characterized by single crystal X-ray diffraction
44]. Similar attempts have been made by the same
[
group using other thiols to isolate thioesters in di-
methylformamide at low temperatures. These were not
as successful [29]. We have made similar unsuccessful
attempts to isolate the Cr(VI)–MSA thioester.
Scheme 1. Acid dissociation of MSA in aqueous solution [19].

D.C. Ramdon et al. / Inorganica Chimica Acta 331 (2002) 178–187
185
Table 5
acid–acetate buffer, produced a significant positive in-
tercept, that was not accounted for by Eq. (1), even
though the fits proved to be acceptable. A significant
intercept was not observed when the buffer used was
citric acid–phosphate, and so the positive intercept in
Fig. 4 was interpreted as arising from catalysis by the
acetic acid–acetate buffer. This in contrast to earlier
reports [31,32] which had predicted that acetic acid–
acetate buffer to have the least effect on the formation
process.
Second-order rate constants kF1 and kF2 at different temperatures, in
the two buffers, along with activation parameters for the reaction
between Cr(VI)/MSA under conditions: I=0.50
M (NaClO4),
[
MSA] =20 mM, [Cr(VI)] =0.20 mM
T
T
T (°C)
Citric acid–phosphate
Acetic acid–acetate
kF1
kF2
(M
kF1
(M
kF2
(M
−
1
s−1
−1
s−1
−1
s−1
−1
s−1
)
(M
)
)
)
2
2
3
3
0.0
5.0
0.0
5.0
6691
8492
10492
13192
12.190.5
15.690.8
17.290.7
19.290.6
5192
5991
6992
9792
2491
3391
4392
4892
The rate expression for the decay of the intermediate
was derived in a similar manner as the formation
reaction using Scheme 2.
‡
DH (kJ
37.690.9
−9293
16.090.8
−17493
23.690.8 45.190.3
−13893 −7799
−
mol 1)
k
obsD
‡
DS (J
+
3
+ 2
+
−
1
{kD1[H ] +kD2[H ] KM1+kD3[H ]KM1
K
M2}[MSA]
T
mol
K
=
−
1)
+ 3
+ 2
+
[H ] +[H ] K +[H ]KM1KM2+KM1KM2KM3
M1
(2)
Again, the type of buffer employed in the kinetic study
does not affect the form of the rate expression, but in
this case the second-order rate constants for the decay
are similar in the two buffers (Table 6).
The rate expression for the formation of the Cr(VI)–
MSA thioester (Eq. (1)) was derived from Scheme 2,
and is represented in terms of second-order rate con-
+
stants, kF1 and k , [H ], the concentration of total
F2
The corresponding activation parameters for each
second-order rate constants were calculated from
Eyrings equation [45] and are given in the respective
MSA, [MSA] , and the equilibrium constants KM1, KM2
T
and KM3
.
‡
kobsF
tables. The negative value of DS suggests that the
+
3
+ 2
activated complexes are more organized than the reac-
tants, and appear to proceed via an associative type of
mechanism.
{
k [H ] +k [H ] KM1}[MSA]T
F2
F1
+ 2
=
+
3
+
[
H ] +[H ] K +[H ]KM1KM2+KM1KM2KM3
M1
(
1)
The change in buffer affected only kD3 significantly,
but not kD1 and kD2. This was further enforced by the
2
−
kobsF shows no contribution from [HMSA ] as confi-
rmed by the linearly rearranged form of Eq. (1). The
values obtained from linear fits were in good agreement
with those from non-linear regressional analysis. The
form of the expression remains the same when either
citric acid–phosphate or acetic acid–acetate is the
buffer. However, it is the value of the second-order rate
constants that differ, as shown in Table 5. The depen-
dence of kobsF on [MSA]T when conducted in acetic
fact that the dependence of kobsD on [MSA] showed no
T
definite positive intercept in either buffer. Perez-Benito
et al. have done extensive studies on buffer participa-
tion in reactions of Cr(VI) with thiols, particularly
glutathione [31] and DL-penicillamine [32] over similar
conditions. They too agree that the buffers have negligi-
ble participation in the decay of the thioester intermedi-
ates. The kinetic traces were similar to ours, but their
Table 6
Second-order rate constants, kD1, kD2 and kD3 at different temperatures in two buffers, and with activation parameters for the reaction between
Cr(VI)/MSA under conditions: I=0.50 M (NaClO ), [MSA] =20 mM, [Cr(VI)] =0.20 mM
4
T
T
T (°C)
Citric acid–phosphate buffer
Acetic acid–acetate buffer
−
1
s−1
kD2 (M−1 s−1
kD3 (M−1 s−1
kD1 (M−1 s−1
kD2 (M−1 s−1
kD3 (M−1 s−1
)
kD1 (M
)
)
)
)
)
20.0
25.0
30.0
35.0
15.390.2
19.590.2
24.190.3
28.890.6
4.590.1
5.190.1
5.890.2
7.890.3
0.290.1
0.390.1
0.690.1
0.590.3
17.490.6
20.090.5
2791
4.790.2
6.090.2
7.590.4
8.390.4
0.990.2
0.990.2
1.090.5
1.290.4
3291
DH‡ (kJ mol−1)
‡
3492
−11795
18.790.6
−17492
N/A
N/A
3494
−114910
37.490.5
−11592
N/A
N/A
−1
−1)
DS (J mol
K

186
D.C. Ramdon et al. / Inorganica Chimica Acta 331 (2002) 178–187
proposed mechanism was different from us. The mathe-
matical treatment in deriving the rate expression uti-
lized complex bi-phasic expressions [46] involving an
equilibrium formation of thioester intermediate, fol-
lowed by a uni-molecular decomposition.
Our studies show significant participation of acetic
acid-acetate buffer in the formation of thioester, but
not from citric acid–phosphate buffer. Table 4 indi-
cates that the experimental rate constants are higher
when acetate is used as the buffer. The kF2 values of
Table 5 show a similar trend. The kF1 path is a bit
different since acetic acid does not show significant
catalysis, however at the low pH where kF1 is significant
citric acid shows some tendency to catalyze the forma-
tion process. This is in direct contrast to the findings of
other researchers [31,32], in which they found phos-
phate to be the best catalytic buffering medium. How-
ever, the experimental rate constants in acetate media
in our study are of a similar magnitude to those re-
ported in the literature [38]. The bi-phasic expression is
applicable only at 425 nm and seems to be inadequate
at 350 nm over our experimental conditions. In addi-
tion, the multi-phase nature disappears as pH increases.
This could be on account of the spectral bands of the
intermediate having significant overlap with those in
the 350–370 nm range.
Fig. 7. Proposed five-coordinate, seven-membered ring transition
state for the Cr(VI)/MSA thioester intermediate.
occur at the Cr(VI) centre, –OH must be an effective
leaving group. Protonation facilitates cleavage of the
Cr–O bond, and this proton can be supplied by a
–
COOH group or even the –SH group.
Fig. 7 shows the proposed five-coordinate transition
state as a seven-membered ring. The formation of such
large ring systems are not very favourable, but from an
entropy viewpoint, it is more likely to have intramolec-
ular proton donation than intermolecular. At this pH,
+
the free [H ] is low, while that from the bound –
COOH is present in higher concentration. A similar
transition state was proposed by Wetterhahn [47], but
2
−
that involved the use of CrO4 along with HSRCOOH
as a general form of a mercapto–carboxylic acid. These
reactants do not exist at significant concentrations
based on known pK values of carboxylic acid groups.
A comparison of Eq. (1) with Eq. (2) shows that only
the MSA species with protonated carboxylate groups
seem to be effective in forming the thioester. As seen in
Eq. (2) which is the expression for the decay of the
intermediate, the species with deprotonated carboxy-
−
2−
The HCrO4 /CrO4
5
4
equilibrium has a pK value of
−
.9–6.0 [7,48,49], whereas the –COOH/–COO pK5
.6 depending on the substituents [35,36]. Therefore,
over the pH range 4.6–5.9, HCrO4 exists with –
COOH/–COO ; below pH 4.6, HCrO
mainly –COOH, and above pH 5.9, CrO4 is predom-
−
−
−
2
−
4
exists with
lates, HMSA , is active over the pH range in the
reduction of the Cr(VI)–MSA intermediate. Although
not very precise, the values obtained from the fits are,
however, not negligible. Obviously, the kinetically reac-
tive forms of the reductant in both cases are the ones
having the larger number of protons. Although unde-
tected in this work, it is postulated in Scheme 2 that a
Cr(V)–MSA intermediate is formed by the reductive
attack of the thiol on the intermediate. Other one
electron transfer steps involving the reduced form of
thiol and Cr(V) and/or Cr(IV) lead to the final Cr(III)
product.
2
−
−
inant with –COO . From this standpoint, it seems that
the most realistic interaction occurs between HCrO4
−
and –COOH. The concentrations of both species de-
crease with an increase in pH, accounting for the
observed decrease in pseudo first-order rate constants.
We can extend the argument to include the participa-
tion of the acetate anion. In this case, it may act as a
Lewis base and assist in sulfur/hydrogen bond cleavage.
The citrate anion may also act as a Lewis base but is
more bulky and hence less effective in catalysis.
At high pH, the five-coordinate transition state may
also be formed, but in the absence of ionizable car-
boxylic protons the seven-membered ring may not be
readily formed. The five-coordinate transition state
would then exist in steady state. The only other species
capable of donating protons to this transition state is
the thiyl proton from the –SH. An intermolecular
proton transfer is expected, along with the donation of
a single electron, hence there is no formation of the
Cr(VI)–MSA thioester intermediate. The second in-
coming MSA effects the reduction to the Cr(V) species
and explains the mixed order expression seen at neutral
pH [24].
If reactions are carried out at pH 5.87, thioester
formation is expected to be minimal and its decay
becomes identical to the decay of Cr(VI). At pH 6 or
greater, there is no spectral evidence seen for the forma-
tion of the thioester, similar to Cr(VI)/penicillamine
system [32]. A report on the reaction of MSA with
Cr(VI) at neutral pH only suggested kinetic evidence
for the thioester [24]. We believe that the presence of
the neighbouring carboxylic protons may play a signifi-
cant role in thioester formation, consistent with an
−
earlier report [47]. HCrO4 exists over the pH and
concentration ranges studied [48,4], and appears to be
the reactive form of Cr(VI). In order for substitution to

D.C. Ramdon et al. / Inorganica Chimica Acta 331 (2002) 178–187
187
(
b) K. Desai, A.J. Baxi, J. Indian Chem. Soc. 69 (1992) 212.
This dependency on proton transfer especially at
higher pH, may have some implications at physiological
pH in biological systems. It is known that DNA double
helix strands are held together by hydrogen bonds, so it
is possible that under cellular conditions where there
are less proton donors, the reactive Cr(VI) may con-
sume some of these hydrogen bonds holding the DNA
strands together. This will no doubt disrupt their struc-
tures and affect their normal mode of action.
[
[
22] M.P. Dave, J.M. Patel, N.A. Langalia, K.A. Jhaken, J. Indian
Chem. Soc. 61 (1984) 891.
23] A. Leydet, C. Jeantet-Segond, C. Bouchitte, C. Moullet, B.
Boyer, J.P. Roque, M. Witvrouw, J. Este, R. Sneock, G. Andrei,
E.D. Clercq, J. Med. Chem. 40 (1997) 350.
[24] P. O’Brien, G. Wang, P.B. Wyatt, Polyhedron 11 (1992) 3211.
[
25] G.H. Jeffery, J. Basset, J. Mendham, R.C. Denney, Vogel’s
Textbook of Quantitative Chemical Analysis, 5th ed., Longman,
UK, 1989, p. 377.
[
[
[
26] E. Beutler, O. Duron, B.M. Kelly, J. Lab. Clin. Med. 61 (1963)
8
82.
27] P.C. Jocelyn, Biochemistry of the –SH Group, Academic Press,
London, 1972, p. 103, 141.
28] S. Gangopadhyay, M. Ali, A. Dutta, P. Banerjee, J. Chem. Soc.,
Dalton Trans. (1994) 841.
29] W. Mazurek, P.J. Nichols, B.O. West, Polyhedron 10 (1991) 753.
30] G.P. Haight Jr, D.C. Richardson, N.H. Coburn, Inorg. Chem. 3
(1964) 1777.
Acknowledgements
We thank the Board for Graduate Studies and Re-
search, University of the West Indies, for a scholarship
[
[
(DCR).
[
[
[
31] J.F. Perez-Benito, D. Lamrhari, C. Arias, J. Phys. Chem. 94
(
1994) 12621.
32] J.F. Perez-Benito, D. Lamrhari, C. Arias, Can. J. Chem. 72
1994) 1637.
33] D.C. Ramdon, D.A. Dixon, T.P. Dasgupta, Inorg. React. Mech.
(2001) 313.
References
(
[
[
[
[
1] D. Stearns, K.E. Wetterhahn, NATO ASI Ser. 2 26 (1997) 55.
2] F.H. Westheimer, Chem. Rev. 45 (1949) 419.
3] R. Sanz, J. Occup. Med. 31 (1989) 1013.
4] M. Cieslak-Golonka, Polyhedron 15 (1996) 3667 and references
therein.
2
[
[
34] L.J. Porter, D.D. Perrin, Aust. J. Chem. 22 (1969) 267.
35] M.C. Aragoni, M. Arca, G. Crisponi, F. Christiani, F. Asia,
V.M. Nurchi, Talanta 43 (1996) 1357.
36] A.E. Martell, R.M. Smith, Critical Stability Constants: other
Organic Ligands, vol. 3, Plenum Press, New York, 1977, p. 140.
37] T. Kiss, P. Buglyo, G. Micera, D. Sana, J. Chem. Soc., Dalton
Trans. (1993) 1849.
[
[
[
[
[
[
[
5] W. Mertz, Phys. Rev. 49 (1969) 163.
6] P.H. Connett, K.E. Wetterhahn, Struct. Bond. 54 (1983) 93.
7] P.H. Connett, K.E. Wetterhahn, J. Am. Chem. Soc. 107 (1985)
4
282.
38] J.F. Perez-Benito, N. Saiz, E. Amat, J. Mol. Catal. A: Chem. 135
[
[
8] P.A. Lay, A. Levina, Inorg. Chem. 35 (1996) 7709.
9] S.L. Brauer, A.S. Hneihen, J.S. McBride, K.E. Wetterhahn,
Inorg. Chem. 52 (1996) 373.
(
1998) 1.
39] P. DeMeester, D.J. Hodgson, H.C. Freeman, C.J. Moore, Inorg.
Chem. 16 (1977) 1494.
40] D.W.J. Kwong, D.E. Pennington, Inorg. Chem. 23 (1984) 2528.
41] T.M. Santos, J. Pedrosa de Jesus, P. O’Brien, Polyhedron 11
[
[
[
[
[
10] D.A. Dixon, N.P. Sadler, T.P. Dasgupta, J. Chem. Soc., Dalton
Trans. (1993) 3489.
11] D.A. Dixon, N.P. Sadler, T.P. Dasgupta, J. Chem. Soc., Dalton
Trans. (1995) 2267.
12] D.A. Dixon, N.P. Sadler, T.P. Dasgupta, J. Chem. Soc., Dalton
Trans. (1997) 1903.
13] D.A. Dixon, N.P. Sadler, T.P. Dasgupta, Inorg. React. Mech. 1
[
[
(
1992) 1687.
[
42] (a) P. O’Brien, J. Pedrosa de Jesus, T.M. Santos, Inorg. Chim.
Acta 131 (1987) 5;
(
b) C.J. Wechler, E. Deutsch, Inorg. Chem. 12 (1973) 2682.
43] A.J. Holmes, D. Lyn, H. Williams, J. Chem. Soc., Perkin Trans.
(2000) 1639.
44] W. Mazurek, G.D. Fallon, P.J. Nichols, B.O. West, Polyhedron
(1990) 770.
45] R.G. Wilkins, Kinetics and Mechanism of Reactions of Transi-
tion Metal Complexes, 2nd ed., VCH Publishing Co., Cam-
bridge, 1991, p. 88.
[
[
[
(
1998) 41.
14] A. McAuley, J.P. McCann, J. Chem. Soc., Dalton Trans. (1975)
83.
2
7
9
[
[
[
15] A. McAuley, M. Olatunji, Can. J. Chem. 55 (1977) 3328.
16] A. McAuley, M. Olatunji, Can. J. Chem. 55 (1977) 3335.
17] X. Fang, F. Hua, Q. Fernando, Chem. Res. Toxicol. 9 (1996)
2
84.
[
[
[
[
18] D.R. Lide, CRC Handbook of Physics and Chemistry, 72nd ed.,
CRC Press, Boston, 1991–1992, pp. 8–18.
19] A. Dutta, B. Saha, M. Ali, P. Banerjee, J. Chem. Res. (M) (1997)
[46] J.H. Espenson, Chemical Kinetics and Reaction Mechanisms,
McGraw-Hill, New York, 1995, p. 77.
[47] P.H. Connett, K.E. Wetterhahn, J. Am. Chem. Soc. 108 (1986)
1842.
[48] M.M. Sena, I.S. Scarminio, K.E. Collins, C.H. Collins, Talanta
53 (2000) 453.
[49] J.J. Cruywagen, J.B.B. Heyns, E.A. Rohwer, Polyhedron 17
(1998) 174.
1
216.
20] M. Rajopadhye, F.D. Popp, J. Heterocyclic Chem. 24 (1987)
637.
21] (a) B.B. Vashi, D.S. Mehta, V.H. Shah, Indian J. Chem. 34B
1995) 802;
1
(