Kinetics of oxidation of glyoxylic acid by [ethylenebis(biguanide)] silver(III) in aqueous media

Article abstract of DOI:10.1016/j.poly.2003.12.006

The complex cation, [ethylenebis(biguanide)]silver(III), [Ag(H ₂L)]³⁺, and its conjugate bases, [Ag(HL)]²⁺ and [AgL]⁺ oxidise glyoxylic acid, CHOCO₂H quantitatively to formic acid and carbon dioxide, themselves being reduced to Ag⁺. Free ethylenebis(biguanide) was recovered in near-quantitative yield. In the investigated pH range (1.00-7.00), the reaction is first order in both [complex] and total glyoxylate, [HGX]_T. The reaction proceeds through four parallel paths: [Ag(H₂L)]³⁺ - HGX (k₁), [Ag(H₂L)]³⁺ - GX^- (k₂), [Ag(HL)]²⁺ - GX^- (k₃), and [AgL]⁺ - GX^- (k₄). The respective rate constants (in 10 ⁴ M^-1s^-1) are, respectively, 7.34, 112, 40.7 and 19.2 at 25.0°C and I = 1.0 M (NaNO₃). Formation of adducts by the reducing species with the metal complex via hydrogen-bonding or electrostriction may be proposed for the low-energy pathway for electron-transfer. A substantial rate retardation in presence of the radical scavenger acrylonitrile suggests successive one-electron changes, Ag ^III → Ag^II → Ag^I.

Full text of DOI:10.1016/j.poly.2003.12.006

Polyhedron 23 (2004) 895–901
www.elsevier.com/locate/poly
Kinetics of oxidation of glyoxylic acid
by [ethylenebis(biguanide)]silver(III) in aqueous media
*
Amit Das, Subrata Mukhopadhyay
Department of Chemistry, Jadavpur University, Calcutta 700 032, India
Received 28 October 2003; accepted 16 December 2003
Abstract
The complex cation, [ethylenebis(biguanide)]silver(III), [Ag(H2L)]3þ, and its conjugate bases, [Ag(HL)]2þ and [AgL]þ oxidise
glyoxylic acid, CHOCO2H quantitatively to formic acid and carbon dioxide, themselves being reduced to Agþ. Free ethylene-
bis(biguanide) was recovered in near-quantitative yield. In the investigated pH range (1.00–7.00), the reaction is first order in both
[complex] and total glyoxylate, [HGX]T. The reaction proceeds through four parallel paths: [Ag(H2L)]3þ – HGX (k1), [Ag(H2L)]3þ
–
GXꢀ (k2), [Ag(HL)]2þ – GXꢀ (k3), and [AgL]þ – GXꢀ (k4). The respective rate constants (in 104 Mꢀ1 sꢀ1) are, respectively, 7.34, 112,
40.7 and 19.2 at 25.0 °C and I ¼ 1:0 M (NaNO3). Formation of adducts by the reducing species with the metal complex via hy-
drogen-bonding or electrostriction may be proposed for the low-energy pathway for electron-transfer. A substantial rate retardation
in presence of the radical scavenger acrylonitrile suggests successive one-electron changes, AgIII ! AgII ! AgI.
Ó 2003 Elsevier Ltd. All rights reserved.
Keywords: Kinetics; Mechanism; Redox; Silver; Glyoxylic acid
1. Introduction
of different silver(III) species connected through protic
equilibria is possible because, in spite of the remarkable
thermodynamic stability, the silver(III) complex oxidises
[9–20] various organic and inorganic species with a va-
riety of electron transfer pathways – inner sphere for
weakly reducing ligands [10,11,13,14,19,20], outer
sphere for weakly basic but strongly reducing ligands
[15,17] and even by Agþ catalysis [9,10,13].
We report herein the kinetics of oxidation of gly-
oxylic acid and its conjugate base by the title silver(III)
complex. Detailed kinetic studies of the oxidation of
glyoxylic acid, one of the smallest carboxylic acids, is of
importance as in metabolic processes such as glycine
catabolism [21] and in plant physiology [22], higher va-
lent metal ions like manganese(III) complexes play an
essential role. Glycine oxidation to CO2 may occur via
glyoxylic acid arising from oxidative deamination of
glycine by the enzyme glycine oxidase [23]. The sil-
ver(III) species, [Ag(OH)4]ꢀ oxidises glycine [24] in
strongly alkaline media to NH3, HCHO and CO2.
[AuCl4]ꢀ also oxidises glycine [25] to NH3, HCO2H
and CO2 in weakly acidic media via the intermediate
formation of glyoxylic acid. We observed that the title
The trivalent silver species [ethylenebis(bigua-
nide)]silver(III) (Fig. 1) is an authentic cationic complex
of silver [1–3] stable in aqueous solution over a wide pH
range against dechelation and intramolecular redox de-
composition [4]. Stabilisation of the silver(III) center by
the square-planar acyclic tetraaza ligand is so extensive
(log K ¼ 52) [5] that the title complex behaves as a
milder oxidant [6], more so than Agþ [7]. The complex is
freely soluble in aqueous media below pH ca. 4.5 but
ꢁ0.2 mM complex can be made soluble up to pH at least
7.0 if alkali is added carefully with stirring to avoid local
excesses. Moreover, the silver(III) retains its identity in
strongly acidic and in strongly alkaline media, a prop-
erty unknown in other higher valent silver complexes. In
aqueous solutions the complex behaves as a mild dibasic
acid (pKa1 ¼ 3:8, pKa2 ¼ 6:3) originating from ligand-
based protic equilibria [8] and thus the reactivity studies
* Corresponding author. Tel.: +91-33-2867-1181; fax: +91-33-2414-
6266.
E-mail address: subrataju@vsnl.net (S. Mukhopadhyay).
0277-5387/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2003.12.006

896
A. Das, S. Mukhopadhyay / Polyhedron 23 (2004) 895–901
+
3
solution at 1.0 M NaNO3 using a Titrino Autotitrator
(Metrohm, 736 GP) at (25 ꢂ 0.1) °C. Production of CO2
as the gaseous reaction product was qualitatively con-
firmed by GC analysis. The gaseous product formed
during the reaction was collected over a saturated so-
dium chloride solution and analysed by gas chroma-
tography as described earlier [27] using a Chemito
(India) GCHT 8610 equipped with TCD, a Porapack Q
and molecular sieve 13X column. Splitless mode injec-
tion of 500 lL of sample was used. The oven tempera-
ture, injector and detector temperature were kept at 40,
55, and 100 °C, respectively. The carrier hydrogen gas
flow was adjusted to 30 ml minꢀ1. All absorbance versus
time data were recorded with a Shimadzu (1601 PC)
spectrophotometer using 1.00 cm quartz cells. The ki-
netics was monitored in situ in the Ôkinetic modeÕ of the
instrument at 420 nm in the electrically controlled
thermostated (25.0 ꢂ 0.1 °C) cell housing (CPS-240A).
In this mode the change in absorbance is automatically
and continuously recorded. For faster reactions gly-
oxylic acid solution (adjusted to desired pH) was directly
injected into the spectrophotometer cell containing
other components of the reaction mixture kept at the
same pH. The desired concentration of the complex and
reducing agent was achieved after mixing. The solution
pH values were adjusted with HClO4 or NaOH. Al-
though the measured pH is usually defined in terms of
the activity of hydrogen ions, we used the hydrogen ion
concentration by calibrating the pH electrode with an-
alytically prepared solutions as described earlier [28].
H
N
NH
H N
2
2
C
C
N
NH
H C
2
Ag
H C
2
N
C
NH
C
H N
2
N
H
NH
2
Fig. 1. Structure of [ethylenebis(biguanide)]silver(III) complex.
silver(III) complex reacts very slowly with glycine al-
though it reacts with glyoxylic acid at a moderately fast
rate. Under these perspectives it is important to under-
stand the chemistry of Ag(III) – biomolecule interac-
tions. Moreover, the ligand H2L in [Ag(H2L)]3þ is
substitution inert so that AgIII never is released in
solution to open extra, complicating pathways in its
reduction.
2. Experimental
2.1. Materials
The silver(III) complex, abbreviated as [Ag(H2L)]3þ
hereafter, was prepared according to the known proce-
dure [2,3] with a slight modification in the crystallisation
process [9]. Aqueous solution of glyoxylic acid were
prepared by dissolving solid glyoxylic acid monohydrate
(98%, Lancaster) in water and standardised either by
standard alkali using WeslowÕs indicator or spectro-
photometrically [26] at 520 nm (e ¼ 17990 Mꢀ1 cmꢀ1 as
estimated by us, in agreement with the literature [26]
value 17 870 Mꢀ1 cmꢀ1) after converting glyoxylic acid
to 1,5-diphenylformazancarboxylic acid. Both the re-
sults agreed well (within 4%). We found that an aqueous
solution of ꢁ0.5 M glyoxylic acid is stable for at least
one week when kept at low temperature (ꢁ5–10 °C) and
in the dark. Aqueous solutions of NaNO3 or NaClO4
(both of G.R., E. Merk) were standardised by passing
through a Dowex 50W X-8 strong cation exchange resin
in the Hþ from and titrating the liberated acid with
standard NaOH to a phenolphthalein end point. All
measurements were made at 25.0 °C and I ¼ 1:0 M
(NaNO3 or NaClO4). All solutions were prepared in
water, which was deionised and then doubly distilled.
2.3. Stoichiometric measurements
The stoichiometry was measured both under kinetic
and reverse condition by estimating glyoxylic acid
spectrophotometrically and estimating unreacted AgIII
in condition when initial [AgIII] is in excess over total
glyoxylic acid, [HGX]T (defined as [HGX] + [GXꢀ]).
Excess glyoxylic acid was mixed with the AgIII complex
in the pH range 3.0–5.0. After the reaction mixture
turned colourless indicating complete reduction of
Ag(III) to Ag(I), excess NaCl was added to remove
AgCl. Freshly prepared phenylhydrazine hydrochloride
(1%) was added (3.0 ml) into each mixture. The resulting
solutions were incubated at ꢁ110 °C for 10 min and
then cooled to room temperature (ꢁ25 °C) followed by
successive addition of 2.5 ml each of concentrated HCl
and 1% K3Fe(CN)6. The mixtures were allowed to stand
for ꢁ5 min for the quantitative formation of pink col-
oured formazan derivative of glyoxylic acid [26] that was
measured at 520 nm.
2.2. Physical measurements and kinetics
Reaction mixtures containing excess AgIII over gly-
oxylic acid were used to measure the unreacted [AgIII].
After completion of reactions (as indicated by no further
change in absorbance), measured volumes of dilute
HNO3 were added to reach a final pH within 1.0–2.0
The acid dissociation constant, Ka of the glyoxylic
acid was determined by titrating several aliquots of
different strengths with carbonate free deaerated NaOH

A. Das, S. Mukhopadhyay / Polyhedron 23 (2004) 895–901
897
Table 1
Stochiometric results for the oxidation of glyoxylic acid (HGX) by the silver(III) complex at I ¼ 1:0 M (NaNO3) and T ¼ 25 °C
[Ag(III)]T (mM)
[HGX]T (mM)
pH
[HGX]left (mM)
[Ag(III)]left (mM)
D[HGX]T/D[Ag(III)]T
0.15
0.30
0.40
0.60
0.80
1.00
1.00
1.20
0.30
0.50
3.1
4.2
4.9
5.3
4.6
4.4
3.9
3.5
3.7
0.16
0.24
0.47
0.61
0.79
0.71
0.88
0.94
1.07
0.87
0.95
1.05
0.97
1.07
0.94
0.93
0.20
0.30
0.45
1.00
0.13
0.46
Average ¼ 0.98 ꢂ 0.10
and the remaining [AgIII] was estimated at 380 nm
(e ¼ 1175 Mꢀ1 cmꢀ1) [8]. Both experiments, whether in
kinetic or in reverse condition resulted a 1:1 stoichiom-
etry between the redox partners (Table 1).
perimental temperature. No external buffer was used in
kinetic measurements since buffering (by acetate, for
example) affects the spectra of the complex; specific in-
teraction of buffer with the complex was thus indicated.
Even in the absence of an external buffer, the media pH
before and after the reactions did not change by more
than 0.10 units.
2.4. Product analysis
Generation of formic acid as one of the oxidation
products of the title reaction was tested by the addition
of chromotropic acid [29] in the product mixture after
removing Agþ as AgCl. Formation of CO2 as the gas-
eous product was qualitatively confirmed by GC anal-
ysis. We also observed that at the end of the reaction,
more than 90% of the free ligand (H2L) can be recov-
ered. For this purpose concentrated (1.0–2.0 mM) so-
lutions of the AgIII complex were reacted with
stoichiometric excess glyoxylic acid. Immediately after
completion of reaction (absorbance <0.01 at 420 nm),
all Agþ was precipitated as AgCl and removed by fil-
tration. From the filtrate, the free ligand, ethylene-
bis(biguanide) was quantitatively isolated [30] as the
sparingly soluble [ethylenebis(biguanide)]copper(II)
sulfate, and the amount of Cu(II) thus held was esti-
mated iodometrically after decomposing the Cu(II)
complex.
3.2. Stoichiometry and the redox reaction
Stoichiometric determination and product analyses
clearly indicate the occurrence of reaction (1) in the
entire redox process:
AgIII þ CHOCO2H þ H2O
! AgI þ HCO2H þ CO2 þ 2Hþ
ð1Þ
3.3. Kinetics
No immediate spectrophotometric change was ob-
served on mixing glyoxylic acid with the silver(III)
complex over the entire range of experimental pH and
total glyoxylate ([HGX]T). However, the absorbance at
k P 420 nm gradually decreases to less than 0.01. We
measured some k0 values in the presence of added (1.0
mM) ethylenebis(biguanide) and found to be identical
with experiments in its absence within reasonable un-
certainties, and we observed no effect of diffused light in
our system. The log10 (absorbance) versus time plots
were found to be linear for more than 90% completion
of reactions and the pseudo-first-order rate constants,
k0, defined by Eq. (2) were obtained from the least-
squares slopes of these plots
3. Results and discussion
3.1. Equilibrium studies and preliminary observations
The pKa of the glyoxylic acid was found to be
3.20 ꢂ 0.10 at 25.0 °C and at I ¼ 1:0 M (NaNO3).
Though this value to some extent differs from the re-
ported [31] value (2.91) under similar condition of
temperature and ionic strength, we used the value 3.20
for pKa for the calculation of rate constants. Most of the
kinetic runs were performed at 420 nm where all other
reaction components except AgIII are transparent.
Within the time span of the reactions, autodecomposion
of the silver(III) complex was insignificant at the ex-
ꢀd½AgðIIIÞꢃ=dt ¼ k0½AgðIIIÞꢃ:
ð2Þ
Any initial drop in absorbance for faster reactions
can be computed from the time of mixing of reactants
(ꢁ4 s) and the observed-first-order rate constants. Av-
erages of k0 values from at least three runs were taken
and individual runs were reproducible within 3%.

898
A. Das, S. Mukhopadhyay / Polyhedron 23 (2004) 895–901
Table 2
Some representative first-order rate constants for oxidation of gly-
oxylic acid by the silver(III) complexa
Detailed data of the kinetics of the process were ob-
tained as a function of pH and [HGX]T and some rep-
resentative data are summarised in Table 2. Over the
entire pH range studied the reaction shows a clear first-
order dependence on [HGX]T and there is no [HGX]T
independent term. The following changes in the reaction
conditions had, within the limits of experimental un-
certainties, no influence on the values of k0: a fivefold
variation in [Ag(III)] from 0.1 to 0.5 mM, occasional
shaking of spectrophotometer cell, presence or absence
of dissolved oxygen and a variation in the monitoring
wavelength in the range 420–500 nm.
pH
[HGX]T (M)
0.10
104k0 (sꢀ1
)
1.00
1.40
2.00
2.26
2.50
2.85
3.00
3.50
4.00
4.35
4.50
5.01
5.62
6.00
6.30
6.51
6.77
7.00
2.02
2.01
2.03
3.55
3.52
6.00
5.98
0.80 ꢂ 0.02
0.96 ꢂ 0.03
1.40 ꢂ 0.03
1.77 ꢂ 0.06
2.20 ꢂ 0.05
3.71 ꢂ 0.06
4.40 ꢂ 0.09
6.09b ꢂ 0.08
6.00 ꢂ 0.08
5.30 ꢂ 0.09
4.90 ꢂ 0.10
4.20 ꢂ 0.11
3.72 ꢂ 0.10
3.34 ꢂ 0.06
3.00 ꢂ 0.06
2.80 ꢂ 0.07
2.46 ꢂ 0.07
2.30 ꢂ 0.07
2.67 ꢂ 0.08
5.70 ꢂ 0.12
11.0 ꢂ 0.22
13.0 ꢂ 0.27
25.2 ꢂ 0.35
6.55 ꢂ 0.12
13.4 ꢂ 0.20
The observed rate constants remarkably depend on
the media ionic strength maintained by NaNO3 or Na-
ClO4 (Table 2). A considerable increase in reaction rate
at pH > 3.0 with decrease in the ionic strength indicates
1
reaction between oppositely charged ions. However at
pH 6 2.0 the reaction rate was found to be practically
independent on the media ionic strength probably due
to involvement of neutral species glyoxylic acid as the
sole reducing component.
0.20
0.40
0.80
0.20
0.40
0.20
0.40
Reactions studied in the presence of 6% (v/v) acry-
lonitrile showed slow polymerisation after ca. 60%
consumption of initial AgIII. The log10 (absorbance)
versus time plots in this condition showed a definite
upward curvature after ca. 50% completion of reactions
probably due to heterogeneity in the reaction media.
However, the first-order rate constants computed from
the initial linear portions of these curves clearly dem-
onstrate much slower reactions in presence of radical
scavengers like acrylonitrile.
a T ¼ 25:0 °C, I ¼ 1:0 M (NaNO3), [complex] ¼ 0.15 mM.
b 104 k0 (sꢀ1) values are 9.25, 14.0 and 19.0 at I ¼ 0:5, 0.2 and 0.1 M
(NaNO3), respectively.
7.0x10-4
6.0x10-4
5.0x10-4
4.0x10-4
A plot of k0 versus pH has a bell shape (Fig. 2) at
a particular [HGX]T. Such a pH dependence indicates
the involvement of two or more acid–base equilibria
[32]. Considering the values for the acid dissociation
3.0x10-4
k
2.0x10-4
1
One of the referees suggested that the rate constants might have
been changed if LiClO4 is used instead of Naþ salts for maintaining the
ionic strength of the reaction medium. We performed several exper-
iments in presence of varying amount of LiClO4 keeping I ¼ 1:0 M.
The first-order rate constants thus obtained were found to be
substantially higher (15–25%, depending on pH of reaction) than
those measured using NaNO3 or NaClO4. The redox partners of the
title reaction are oppositely charged and thus conventional metal ion
catalysis (M. Gupta, S.K. Saha, P. Banerjee, J. Chem. Soc., Perkin
Trans. II (1988) 1781) can be ruled out. A possible reason for the rate
enhancement in presence of Liþ ion may be explained by very high
approximate hydration number as estimated by transference data
(F.A. Cotton, G. Wilkinson, C.A. Murillo and M. Bochmann,
Advanced Inorganic Chemistry (Ch. 3), 6th ed., Wiley, New York,
1999, p. 102) of Liþ (25.3) than Naþ (16.6) that results in a major
lowering of the effective concentration of solvent water which in turn
increases the activity of the redox partners. A similar situation has also
been verified with the oxidation of azide by the title silver(III) complex.
We also note that the effect of LiClO4 for the rate enhancement is
marginal when the reactive form of the reducing agent is uncharged.
We will explore in future the finer details of these effects.
1.0x10-4
0.0
1
2
3
4
pH
5
6
7
Fig. 2. Dependence of k0 on pH at T ¼ 25:0 °C, [Ag(III)]T ¼ 0.15 mM,
[HGX]T ¼ 0.10 M, I ¼ 1:0 M (NaNO3). The solid line represents the
fit of Eq. (10).
constants of the complex in aqueous solution [8]
(Ka1 and Ka2, which are 1:58 ꢄ 10ꢀ4 and 5:01 ꢄ 10ꢀ7 M,
respectively) and Ka ( ¼ 6:31 ꢄ 10ꢀ4 M), the mea-
sured acid dissociation constant of glyoxylic acid un-
der experimental conditions, Scheme 1 in Eqs. (3)–(9)
provides a reasonable explanation for the kinetic
observations.

A. Das, S. Mukhopadhyay / Polyhedron 23 (2004) 895–901
899
[Hþ] terms, viz. k2Ka and k3Ka1Ka respectively, until v2
for the fitting becomes acceptably small (of the order of
10ꢀ16). The rate constants k2 and k3 evaluated in this
way agree quite well (within 5%) with the values ob-
tained from Eqs. (11) and (12). This is another support
for the proposed scheme of reactions.
3.4. Mechanism
The observed sequence of rate constants k1 < k2 >
k3 > k4 clearly demonstrates superior kinetic reactivity
of protonated oxidants and deprotonated reductant
over their counterparts – a trend normally followed in
majority of redox reactions [16–19,33–37]. Among the
rate constants evaluated ðk1–k4Þ, we find k1 is the
smallest indicating negligible reactivity of HGX to-
wards the title AgIII complex. We therefore did not in-
clude in the scheme the reaction between [Ag(HL)]2þ
and HGX. Moreover, the acid dissociation constants of
[Ag(H2L)]3þ and HGX are quite distant and simulta-
neous equilibrium generation of both [Ag(HL)]2þ and
HGX is not appreciable.
Scheme 1.
Scheme 1 leads to the rate law
3
k0½Hþꢃ =ð½HGXꢃT a1a2Þ
3
2
¼ k1½Hþꢃ þ k2Ka½Hþꢃ þ k3Ka1Ka½Hþꢃ þ k4Ka1Ka2Ka;
The title silver(III) complex is a weak oxidant (E1=2
AgIII/AgII ¼ 0.05 V) versus saturated calomel electrode
(SCE) [6,38] but the thermodynamics of two electron
oxidation of glyoxylic acid is much favourable, esti-
ð10Þ
2
2
where a1 ¼ ½Hþꢃ =ð½Hþꢃ þ Ka1½Hþꢃ þ Ka1Ka2Þ is the
fraction of the silver(III) complex present as
[Ag(H2L)]3þ and a2 ¼ ½Hþꢃ=ðKa þ ½HþꢃÞ is the fraction
of glyoxylic acid present as HGX.
2
mated to be around 0.3 V versus NHE. Rate retar-
dation in presence of acrylonitrile indicates generation
of AgII and polymerisation initiating radicals, viz.
HGXþ and GXꢀ. A competitive decay of glyoxylate
radicals occurs via fast consumption by AgII leading to
products and initiation of acrylonitrile polymerisation –
it explains lowering in reaction rate in presence of
acrylonitrile. Generation of glyoxylate radicals by one-
electron oxidation of glyoxylic acid is well-established
by EPR spectroscopy [39]. However, we observed that
the whole reaction course was EPR silent. We infer
that the radicals produced from one-electron oxida-
tion of glyoxylic acid react in a fast manner with
AgII before they can escape into bulk solvent, thus
evading detection. Unfortunately one electron potential
data for HGX/HGXþ couple as well as its self-ex-
change rate is unavailable. Thus we are unable to con-
clude the nature of electron transfer, inner or outer
sphere.
In the lower pH range (up to pH 2.5), the second
dissociation of the dibasic complex may be neglected
and reactions (8) and (9) are unnecessary to reproduce
the observed k0 values. On the other hand at pH P 6.0
we can assume only [Ag(HL)]2þ and [AgL]þ as the
contributing silver species reacting with GXꢀ. The rate-
laws in these extremes are, respectively,
k0ðKa þ ½HþꢃÞ=½HGXꢃ ¼ k1½Hþꢃ þ k2Ka;
ð11Þ
ð12Þ
T
k0ðKa2 þ ½HþꢃÞ=½HGXꢃ ¼ k3½Hþꢃ þ k4Ka2:
T
The second-order rate constants (in 104
M )
ꢀ1 sꢀ1
evaluated from these linear plots (left-hand sides of Eqs.
(11) and (12) versus [Hþ]) are, respectively, k1 ¼ 7:34,
k2 ¼ 112, k3 ¼ 40:7 and k4 ¼ 19:2. They reproduce the
observed first-order rate constants satisfactorily (within
10%) using Eq. (10). We also tried to construct a global
fit of the left-hand side of Eq. (10) with [Hþ]; a third-
order polynomial fit of the left-hand side of Eq. (10)
with [Hþ] resulted a very close (within 2%) value of k1 as
found by Eq. (11) and less close agreement for the k2
value. However, the constant term k4Ka1Ka2Ka yielded a
negative value that could not be made positive even
varying k1 and k2 values by more than 50%. We thus
used a constant value of the term k4Ka1Ka2Ka (using the
k4 values as obtained from Eq. (12) and known Ka1, Ka2
and Ka values) and k1, and then started iteration to
obtain the best values of the coefficients of the [Hþ]2 and
Up to pH 2.5, where only [Ag(H2L)]3þ is the kineti-
cally reactive oxidant, preequilibrium formation of any
adducts between [Ag(H2L)]3þ and HGX or GXꢀ is at
best poor, estimated to be within a maximum of the
order of 10ꢀ1
M
ꢀ1. Such a low value does not confirm a
2
Private communication with Dr. Philip N. Ross, Senior Scientist,
Lawrence Berkeley National Laboratory. Glyoxylic acid can thermo-
dynamically be considered as half formic acid and half oxalic acid. The
redox potential should therefore be exactly half way between that of
formic acid (0.14 V) and oxalic acid (0.49 V).

900
A. Das, S. Mukhopadhyay / Polyhedron 23 (2004) 895–901
strong inner sphere attachment of glyoxylic acid with
the metal center. Formation of a strong inner sphere
adduct by glyoxylate with [Mn(pd)2(H2O)2]þ (Hpd ¼
pentane 2,4 dione) has been proposed [40] and the for-
mation constants of glyoxylate complexes of tripositive
rare-earth-metal cations are also high [41]. The in-plane
coordinatively saturated AgIII in the title complex makes
the situation a different one in aqueous solution where
the replaceable water molecules attach only weakly with
the axial positions of the AgIII center. Replacement of
water molecule by glyoxylate is thus expected to be weak
at best. The estimated low value for the formation of
adducts in the present study may be assigned to elec-
trostriction [42–45] or any kind of it aided by hydrogen
bonding [46,47] in the adducts.
A kinetically indistinguishable scheme where the
adducts formed by electrostriction or aided by hydro-
gen bonding on the basal plane may be proposed.
However, this proposition enforces a large separation
between the oxidant and the reductant species that in-
hibits electron transfer within the adduct due to incor-
rect approach, as it were. Occurrence of electron transfer
rather is justified when the reductant approaches the
metal center along the axial site of the latter. A com-
paratively strong adduct formation between the complex
and the oxalic acid or bioxalate ion [18] in comparison
to glyoxylate indicate a much stronger binding with the
more potentially complexing oxalate species inherited by
their bidentacy.
½AgðH2LÞ3þꢃ=Tc
¼ fraction of the silverðIIIÞcomplex
present as ½AgðH2LÞ3þꢃ
2
2
¼ ½Hþꢃ =ð½Hþꢃ þ Ka1½Hþꢃ þ Ka1Ka2Þ
¼ a1 ðsayÞ:
Thus, using (A.1) and (A.2) it yields
½AgðH2LÞ3þꢃ ¼ a1Tc; ½AgðHLÞ2þꢃ ¼ Ka1a1Tc
and
2
½AgLþꢃ ¼ Ka1Ka2a1Tc=½Hþꢃ :
From the glyoxylic acid dissociation equilibrium
Ka ¼ ½GXꢀꢃ½Hþꢃ=½HGXꢃ;
it appears
½HGXꢃ=ð½HGXꢃ þ ½GXꢀꢃÞ
¼ fraction of glyoxylic acid present as
½HGXꢃ ¼ ½Hþꢃ=ðKa þ ½HþꢃÞ ¼ a2 ðsayÞ:
Thus,
½HGXꢃ ¼ a2½HGXꢃ
and
T
½GXꢀꢃ ¼ Kaa2½HGXꢃ =½Hþꢃ:
Now, for the proposed scheme (see text), rate of the
T
reaction becomes
Rate ¼ k1a1a2Tc½HGXꢃ þ k2Kaa1a2Tc½HGXꢃ =½Hþꢃ
T
T
2
þ k2KaKa1a1a2Tc½HGXꢃ =½Hþꢃ
T
Acknowledgements
3
þ k4KaKa1Ka2a1a2½HGXꢃ =½Hþꢃ ;
T
The award of a Junior Research Fellowship to Amit
Das by University Grants Commission, New Delhi and
financial assistance from Jadavpur University Research
Grant are gratefully acknowledged.
whence, it becomes,
3
3
2
k0½Hþꢃ =ð½HGXꢃ a1a2Þ ¼ k1½Hþꢃ þ k2Ka½Hþꢃ
T
þ k3Ka1Ka½Hþꢃ þ k4Ka1Ka2Ka:
Appendix A. Derivation of Eq. (10)
References
The acid dissociation constants for the title silver(III)
complex are
[1] P. Ray, Nature (London) 151 (1943) 643.
[2] P. Ray, K. Chakravarty, J. Indian Chem. Soc. 21 (1944) 47.
[3] P. Ray, Inorg. Synth. 6 (1960) 74.
2þ
Ka1 ¼ ½AgðHLÞ ꢃ½Hþꢃ=½AgðH2LÞ3þꢃ;
ðA:1Þ
ðA:2Þ
[4] R. Banerjee, D. Banerjea, Indian J. Chem. 17A (1979) 246.
[5] D. Sen, N.N. Ghosh, P. Ray, J. Indian Chem. Soc. 27 (1950) 619.
[6] K. Das, R. Banerjee, A. Das, S. Dasgupta, Bull. Electrochem. 5
(1989) 477.
Ka2 ¼ ½AgLþꢃ½Hþꢃ=½AgðHLÞ2þꢃ:
The analytical concentration of the silver(III) complex,
Tc, may be written as
[7] W.M. Latimer, The Oxidation States of the Elements and their
Potentials in Aqueous Solutions, second ed., Prentice-Hall,
Englewood Cliffs, NJ, 1952, p. 176.
2þ
Tc ¼ ½AgðH2LÞ3þꢃ þ ½AgðHLÞ ꢃ þ ½AgLþꢃ;
[8] S. Mukhopadhyay, R. Banerjee, Polyhedron 13 (1994) 53.
[9] R. Banerjee, K. Das, A. Das, S. Dasgupta, Inorg. Chem. 28 (1989)
585.
Substituting [Ag(HL)2þ
]
and [AgLþ] in terms of
[Ag(H2L)3þ] and rearranging, it results
[10] R. Banerjee, A. Das, S. Dasgupta, J. Chem. Soc., Dalton Trans.
(1989) 1645.
3þ
2
Tc ¼ ½AgðH2LÞ ꢃð1 þ Ka1=½Hþꢃ þ Ka1Ka2=½Hþꢃ Þ;
whence we may write,
[11] R. Banerjee, A. Das, S. Dasgupta, J. Chem. Soc., Dalton Trans.
(1990) 1207.

A. Das, S. Mukhopadhyay / Polyhedron 23 (2004) 895–901
901
[12] R. Banerjee, A. Das, S. Dasgupta, J. Chem. Soc., Dalton Trans.
(1990) 2271.
[29] F. Feigl, Spot Tests in Organic Analysis, 5th ed., Elsevier,
London, 1956, pp. 331, 342.
[30] P. Ray, Chem. Rev. 61 (1961) 313.
[13] R. Banerjee, R. Das, S. Mukhopadhyay, J. Chem. Soc., Dalton
Trans. (1992) 1317.
[31] A.E. Martell, R.M. Smith, in: Critical Stability Constants, vol. 3,
Plenum Press, New York, 1997, p. 65.
[14] S. Dasgupta, E. Herlinger, W. Linert, J. Chem. Soc., Dalton
Trans. (1993) 567.
[32] R.G. Wilkins, The Study of Kinetics and Mechanism of Reactions
of Transition Metal Complexes, Allyn & Bacon, Boston, MA,
1974 (Ch. 1), p. 460.
[15] R. Banerjee, Proc. Indian Acad. Sci. (Chem. Sci.) 106 (1994) 655.
[16] S.P. Ghosh, M.C. Ghosh, E.S. Gould, Inorg. Chim. Acta 225
(1994) 83.
[33] S.K. Ghosh, R.N. Bose, E.S. Gould, Inorg. Chem. 26 (1987)
2688.
[17] S. Mukhopadhyay, R. Banerjee, J. Chem. Soc., Dalton Trans.
(1994) 1349.
[34] K.D. Asmus, M. Bonifacic, P. Toffel, P. OÕNeill, D.S. Frohlinde,
S. Steenken, J. Chem. Soc., Faraday Trans. 1 (1978) 1820.
[35] K. Lemma, A.M. Sargeson, L.I. Elding, J. Chem. Soc., Dalton
Trans. (2000) 1167.
[18] S. Mukhopadhyay, S. Kundu, R. Banerjee, Proc. Indian Acad.
Sci. (Chem. Sci.) 107 (1995) 403.
[19] P. Bandopadhyay, S. Mukhopadhyay, Polyhedron 21 (2002) 1893.
[20] P. Bandopadhyay, B.B. Dhar, J. Bhattacharyya, S. Mukhopad-
hyay, Eur. J. Inorg. Chem. (2003) 4308.
[36] T. Shi, J. Berglund, L.I. Elding, Inorg. Chem. 35 (1996) 3498.
[37] J. Berglund, L.I. Elding, Inorg. Chem. 34 (1995) 513.
[38] B. Chakravarty, J. Inorg. Nucl. Chem. 41 (1979) 757.
[39] B. Neumann, O. Steinbock, S.C. Muller, N.S. Dalal, J. Phys.
Chem. 100 (1996) 12342.
[21] A. White, Principles of Biochemistry, 2nd ed., McGraw-Hill, New
York, 1954, p. 548.
[22] E.H. Rodd, in: Chemistry of Carbon Compounds, Part B, vol. 1,
Elsevier, Amsterdam, 1952, p. 850.
[40] R. Banerjee, R. Das, A.K. Chakraburtty, J. Chem. Soc., Dalton
Trans. (1990) 3277.
[23] D.M. Greenberg, Chemical Pathways of Metabolism, vol. 2, 2nd
ed., Academic Press, New York, 1961, p. 85.
[24] Y. Sun, L.J. Kirschenbaum, I. Kouadio, J. Chem. Soc., Dalton
Trans. (1991) 2311.
[41] J.E. Powell, Y. Suzuki, Inorg. Chem. 3 (1964) 690.
[42] M.G. Evans, G.H. Nancollas, Trans. Faraday Soc. 49 (1953)
363.
[25] J. Zou, Z. Guo, J.A. Parkinson, Y. Chen, P.J. Sadler, J. Chem.
Soc., Chem. Commun. (1999) 1359.
[43] C. Postmus, E.L. King, J. Phys. Chem. 59 (1955) 1208.
[44] F. Basolo, R.G. Pearson, Mechanism of Inorganic Reactions, 2nd
ed., Wiley, New York, 1967, p. 34.
[26] D.N. Kramer, N. Klein, R.A. Baselice, Anal. Chem. 30 (1959)
250.
[45] M. Grant, R.B. Jordan, Inorg. Chem. 20 (1981) 55.
[46] C.D. Hubbard, A. Gerhard, R. van Eldik, J. Chem. Soc., Dalton
Trans. (2001) 1069.
[27] K. Dutta, S. Bhattacharjee, B. Choudhuri, S. Mukhopadhyay, J.
Environ. Monit. 4 (2002) 754.
[28] S. Banerjee, U. Roy Choudhury, B.C. Ray, R. Banerjee, S.
Mukhopadhyay, Anal. Lett. 34 (2001) 2797.
[47] M.C. Ghosh, J.W. Reed, R.N. Bose, E.S. Gould, Inorg. Chem. 33
(1994) 73.