Mechanistic studies on the oxidation of glyoxylic and pyruvic acid by a [Mn4O6]4+ core in aqueous media: Kinetics of oxo-bridge protonation

Article abstract of DOI:10.1002/hlca.200690186

In aqueous media (pH 2.5-6.0), the Mn^IV tetramer [Mn ₄(μ-O)₆(bipy)₆]⁴⁺ (1 ⁴⁺; bipy = 2,2′-bipyridine) oxidizes both glyoxylic and pyruvic acid to formic and acetic acid, respectively, under formation of CO ₂. Kinetics studies suggest that the species 1⁴⁺, its oxo-bridge protonated form [1H]⁵⁺, i.e., [Mn₄(μ-O) ₅(μ-OH)(bipy)₆]⁵⁺, the reducing acids (RH) and their conjugate bases (R^-) all take part in the reaction. The oxo-bridge protonated oxidant [1H]⁵⁺ was found to react much faster than 1⁴⁺. Thereby, the gemdiol forms of the α-oxo acids (especially in the case of glyoxylic acid) are the possible reductants. A one-electron/one-proton electroprotic mechanism operates in the rate-determining step.

Full text of DOI:10.1002/hlca.200690186

Helvetica Chimica Acta – Vol. 89 (2006)
1947
Mechanistic Studies on the Oxidation of Glyoxylic and Pyruvic Acid by a
[Mn4O6]4+ Core in Aqueous Media: Kinetics of Oxo-Bridge Protonation
by Suranjana Das, Jhimli Bhattacharyya, and Subrata Mukhopadhyay*
Department of Chemistry, Jadavpur University, Calcutta 700032, India
(e-mail: smukhopadhyay@chemistry.jdvu.ac.in)
In aqueous media (pH 2.5–6.0), the MnIV tetramer [Mn4(m-O)6
A
oxidizes both glyoxylic and pyruvic acid to formic and acetic acid, respectively, under formation of CO2.
Kinetics studies suggest that the species 14+, its oxo-bridge protonated form [1H]5+, i.e., [Mn4(m-O)5
(m-
G
OH)(bipy)6]5+, the reducing acids (RH) and their conjugate bases (Rꢀ) all take part in the reaction.
The oxo-bridge protonated oxidant [1H]5+ was found to react much faster than 14+. Thereby, the gem-
diol forms of the a-oxo acids (especially in the case of glyoxylic acid) are the possible reductants. A
one-electron/one-proton electroprotic mechanism operates in the rate-determining step.
Introduction. – The unique manganese-oxo aggregate (OEC) present in photosys-
tem II (PS-II) catalyzes the light-driven oxidation of H2O to O2 [1–6] resulting in the
O2-rich atmosphere encountered on Earth. The OEC cycles through five redox states,
S0 – S4, the index of which refer to the number of oxidizing equivalents stored [7][8].
The overall four-electron (4 eꢀ) oxidation of two H2O molecules leading to O2 is asso-
ciated with transfer of four H+: 2 H2O ! O2 +4 H+ +4 eꢀ.
Ligands derived from H2O (O2ꢀ or OHꢀ) are present as bridges between Mn-atoms
along with carboxylato moieties in the catalytic site [9–11]. Successive redox reactions
at the Mn site are definitely associated with a substantial change in the oxo-bridge
basicity [1][9][10], and it is likely that a change in the protonation state of the bridged
metal cluster also occurs [12]. The observed decrease in the exchange coupling between
MnIV in a model tetranuclear system [13][14], resulting from protonation of the oxo
bridges, has important implications in interpreting the changes in magnetic behavior
of the OEC upon S-state advancement and changes in configuration. Oxo-bridge pro-
tonations also cause a substantial increase in the MnꢀMn distance in multinuclear Mn
complexes, and an increase in reduction potential [14–18]. Besides these physical
effects, investigations on chemical aspects resulting from oxo-bridge protonation
have hardly been studied, expect in a few reports where it was observed that catalase
activity of [Mn(salpn)(m-O)]2 (salpn=1,3-bis(salicylideneamineto)propane) is inhib-
ited by a single protonation on the oxo-bridge [18], whereas disproportionation of a
(MnIII)2 complex requires oxo-bridge protonation [19]. It is also of note that oxo-bridge
protonation in multinuclear higher-valent Mn complexes sometimes leads to cluster
breakup [16][20], rendering their redox chemistry H+-coupled. The acid–base chemis-
try resulting from oxo-bridge protonation is also well-studied, along with the kinetic
stabilities of oxo bridged species [15].
ꢀ 2006 Verlag Helvetica Chimica Acta AG, Zürich

1948
Helvetica Chimica Acta – Vol. 89 (2006)
The complex investigated, [Mn4(m-O)6
(bipy)6]4+ (14+; bipy=2,2’-bipyridine), was
G
originally prepared by Girerd and co-workers [16]. On the basis of the oxidation states
of its metal centers, 14+ formally corresponds to an S3 or S4 state of OEC [2], and is sta-
ble in aqueous solution. The one-electron-reduced, mixed-valent (MnIV)3MnIII form of
U
14+ is one of the best EPR spectroscopic models for the S2 state [16]. Mechanistic stud-
ies of the reduction of 14+, thus, appeared interesting.
The present investigation deals with the kinetics and mechanistic aspects of the oxi-
dation of two small carboxylic acids, glyoxylic acid (=oxoacetic acid) and pyruvic acid
(=2-oxopropanoic acid), which are of biological relevance [21–23]. Examples of the
oxidation of such a-oxo carboxylic acids by metal oxidants are not too scarce [24].
Interestingly, these acids and their anions co-exist in equilibrium with their hydrated
geminal-diol forms in aqueous solution [25]1). However, so far, not even a single report
is available in the literature that clearly describes the actual species involved in the
redox process, if not apportioned. An attempt has been made through this investigation
to search the mechanistic pathways of the oxidation of these substrates and to detect
the possible species involved in the reduction of the title Mn complex. In this study
we also gained strong kinetic evidence of oxo-bridge protonation. The reactivity of
the oxo-bridge-protonated oxidant is much higher than that of its conjugate base in oxi-
dizing the title reducing agents. We further note that most of the Mn enzymes, including
PS-II, work in combination with radicals [4], and in this study, we will show that the title
Mn complex reacts with the a-oxo acids via one-electron steps generating acyl radicals.
Results and Discussion. – Equilibrium Measurements. The measured pKa values of
glyoxylic and pyruvic acid are 3.13 and 2.50 in H2O, and 3.40 and 2.75 in 95% D2O,
H
respectively. These values are the averages of at least ten independent measurements,
the experimental uncertainty being well within 0.10 pKa unit. These values compare
well with the literature data [26][27] (Table 1). Intuitively, one would expect that,
due to the electron-pushing effect of the Me group, glyoxylic acid is the stronger acid
than pyruvic acid, just as is the case with formic vs. acetic acid. The observed reversal
in the expected order of acid strength (Table 1) is in line with the predominant hydra-
1
)
The hydration constants of the reducing species as found in [25] are as follows: glyoxylic acid,
1.0”102–1.4”103; glyoxylate, 6.5–260; pyruvic acid, 0.61–3.1; pyruvate, 6.4”10ꢀ2 ꢀ0.21.

Helvetica Chimica Acta – Vol. 89 (2006)
1949
tion of glyoxylic acid in aqueous solution [25], which leads to the gem-diol species
HC(OH)2CO2H; the corresponding hydro form is, at best, weak in the case of pyruvic
acid [25]. Interestingly, we note that progressive replacement of the a-H-atom of acetic
acid (AcOH) increases the acidity of the species thus formed. The pKa values of
CH3CO2H, CH2(OH)CO2H, and HC(OH)2CO2H are 4.57 [28], 3.62 [28], and 3.13
(this work), all values referring to 25.08 and an ionic strength I of 1.0M. It may, thus,
be concluded that dissociation mainly occurs from the hydrated form of glyoxylic
acid, but from the regular (non-hydrated) form of pyruvic acid. The two OH groups
in HC(OH)2CO2H possibly make the bonded C-atom less electropositive, and the
pKa value is, therefore, a strong indication of the diol form.
Table 1. pKa Values of Glyoxylic and Pyruvic Acid in Aqueous Media. T=25.08; ionic strength I in M.
Reducing agent
Glyoxylic acid
H
2O
Ref.
D2O/H2O 95 :5
U
3.13Æ0.10a)
2.91 (I=1.0)
3.20 (I=1.0)
2.98 (I=0.5)
3.46 (I=0)
2.50Æ0.10a)
2.35 (I=0.5)
2.60 (I=0)
t.w.a)
[26a]
[24h]
[26b]
[26c]
t.w., [24i]
[27]
[26c]
3.40Æ0.10a)
Pyruvic acid
2.75Æ0.10a)
a) This work (see text and Exper. Part).
Stoichiometric Analyses. Quantitative (96Æ6%) evolution of CO2 (in accordance
with Eqns. 1 and 2) was found in both the a-oxo acid oxidations, along with HCO2H
and CH3CO2H, respectively, for the redox reaction of glyoxylate and pyruvate. These
stoichiometric results, thus, confirm that the reducing species are, overall, two-elec-
tron-transfer agents. The likely MnII species under the experimental conditions are
[MnII(bipy)] complexes [29]. The UV/VIS spectra of the product solutions were super-
imposable to that of a mixture of Mn(NO3)2 and bipy under the reaction conditions.
G
Thus, the oxidation of glyoxylic and pyruvic acid is represented by:
14+ +4 HC(O)COOH+4 H+ ! 4 Mn2+ +6 bipy+4 CO2 +4 HCOOH+2 H2O (1)
14+ +4 MeC(O)COOH+4 H+ ! 4 Mn2+ +6 bipy+4 CO2 +4 MeCOOH+2 H2O (2)
Kinetics. In the presence of excess reducing agent, the reactions followed excellent
first-order kinetics for at least four half-lives; and any initial drop in absorbance for
faster reactions could be estimated by the observed first-order rate constants k0
A
and UV/VIS measurement in the photometerꢁs kinetics mode. The following changes
in reaction conditions had, within experimental uncertainty, no effect on the observed
first-order rate constants: presence or absence of dissolved O2, diffuse light, change in
monitoring wavelength in the range 380–530 nm. At a fixed reducing-agent concentra-
tion, k0 increased with increasing acidity for the oxidation of both reducing agents

1950
Helvetica Chimica Acta – Vol. 89 (2006)
Table 2. Some Representative First-Order Rate Constants for the Oxidation of Glyoxylic and Pyruvic Acid
by the Title Complex. Conditions: [14+]=0.10 mM, T=25.08, I=1.0M (NaNO3). Values in parentheses
were calculated from Eqn. 10 with the rate constants reported in Table 3. Cbipy (=[(bipy)H)+]+[bipy])
and TR (=RH+Rꢀ) in mM.
Glyoxylic acid
Pyruvic acid
pH
TR
Cbipy
103 k0 [sꢀ1
]
pH
TR
Cbipy
104 k0 [sꢀ1
]
2.89
3.07
3.60
4.23
4.48
5.14
5.69
5.96
5.18
5.18
5.19
5.19
5.49
5.43
5.50
5.51
5.24
5.26
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
4.0
6.0
8.0
10
2.0
2.0
2.0
2.0
2.0
2.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
1.0
3.0
10
110 (103)
2.54
3.29
3.73
4.09
4.50
5.33
5.92
4.03
4.00
4.04
4.04
4.14
4.12
4.11
4.10
3.63
3.54
2.0
2.0
2.0
2.0
2.0
2.0
2.0
4.0
6.0
8.0
10
2.0
2.0
2.0
2.0
2.0
2.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
3.0
1.0
10
172 (181)
78.0 (79.3)
32.0 (33.4)
9.91 (10.2)
6.91 (6.47)
2.69 (2.50)
1.77 (1.45)
1.45 (1.52)
5.09 (4.80)
7.71 (7.20)
9.88 (9.50)
12.0 (11.9)
1.95 (1.87)
1.95b) (1.94)
1.96 (1.85)
1.98 (1.84)
4.38c)
44.7 (42.5)
16.9 (17.0)
8.44 (8.20)
4.14 (3.91)
1.65 (1.55)
1.21a) (1.25)
18.5 (18.5)
28.5 (29.4)
34.1 (36.2)
43.6 (45.2)
8.02 (7.44)
8.18 (7.74)
8.24 (7.89)
8.46 (8.04)
30.3c)
30
60
3.0
3.0
30
3.0
3.0
59.7d)
6.29d)
a) 104 k0 =1.24 sꢀ1 in the absence of added bipy. b) 103 k0 =1.91 sꢀ1 in the absence of added bipy. c) I=0.5M
d
(NaNO3). ) I=0.1M (NaNO3).
(Figure). Both reactions show good linear plots of k0 vs. TR, and there was no TR-inde-
pendent term in either redox process2). Interestingly, both the reactions are indepen-
dent on Cbipy
(=[(bipy)H)+]+[bipy]) in the range 1–60 mM (Table 2). Thus, the weak
G
influence of the concentration of the 2,2’-bipyridine ligands on the reaction rates
refutes the importance of the equation equilibrium (Eqn. 3):
[Mn4(m-O)6
A
G
N
The observed dependence of k0 on H+ could not be fitted satisfactorily to any type
of reaction with only one protic equilibrium (Eqns. 4 or 5). Both need to be considered
simultaneously, as outlined in Eqns. 6–9. Thereby, the approximation 1>>K1[H+] was
made, which leads to the rate-law presented in Eqn. 10 (where RH is glyoxylic or pyr-
uvic acid).
k1
5þ
1
4þ þ Hþ
½1HŠ
(4)
G
H
Convention: TR =[RH]+[Rꢀ], where RH are the reducing agents (acids).
2
)

Helvetica Chimica Acta – Vol. 89 (2006)
1951
Figure. Plot of the rate constant k0 vs. pH for glyoxylic acid (a) and pyruvic acid (b). Conditions:
[complex]=0.10 mM, TR =2.0 mM, Cbipy =3.0 mM, T=25.08, I=1.0M (NaNO3). The solid lines were
drawn on the basis of the rate constants reported in Table 3 in combination with Eqn. 10. The
actually observed rate-constant values are represented by circles.
ka
RH
Rꢀ þ Hþ
(5)
(6)
(7)
(8)
(9)
G
H
k1
½1HŠ5þ þ RH
½1HŠ5þ þ Rꢀ
products
products
H
G
H
k2
G
k3
1
4þ þ RH
products
products
H
G
H
k4
1
4þ þ Rꢀ
G

1952
Helvetica Chimica Acta – Vol. 89 (2006)
k0
A
(10)
Eqn. 10, along with the non-linear least-squares values for k1K1, (k2K1Ka +k3), and
k4, as reported in Table 3, reproduced the observed k0 values acceptably well (within
7%).
Table 3. Rate Constants for the Reduction of 14+ by a-Oxo Acids in Aqueous Media. Conditions: Cbipy =3.0
mM, T=25.08, I=1.0M (NaNO3).
Reaction path
H2O
D2O/H2O 95 :5
N
Glyoxylic acid
Pyruvic acid
Glyoxylic acid
Pyruvic acid
(1.60Æ0.05)”104
60Æ5
(1.70Æ0.07)”103
14Æ1
(1.10Æ0.06)”104
50Æ5
(9.00Æ0.07)”102
9.6Æ0.6
a
k1K1 )
b
k2K1Ka +k3 )
b
k4 )
0.68Æ0.04
0.057Æ0.003
0.50Æ0.03
0.045Æ0.003
a) In Mꢀ2 sꢀ1; b) In Mꢀ1 sꢀ1
.
Since long it is known [25] that hydration of the two a-oxo acids and their anions
forms the respective gem-diols, the extent of hydration decreasing in the order glyoxylic
acid>glyoxylate>pyruvic acid>pyruvate. However, as the spread of the reported val-
ues for the hydration of each species is sufficiently high [25] (though the above-men-
tioned order is maintained), no quantitative evaluation of rate parameters was
attempted to confirm the participation of dehydrated or hydrated reducing species.
Table 3 clearly demonstrates that [1H]5+ is kinetically superior than [1]4+ in oxidiz-
ing the acids RH or their dissociated forms Rꢀ. However, it is not possible to conclu-
sively compare the reactivities of RH or Rꢀ in reducing [1H]5+ or [1]4+ as the k2 and k3
paths are ꢂproton-ambiguousꢁ. Having a common view that deprotonated reductants
normally react faster [30] than their neutral conjugate acids, we may assume
10k3 ꢁk4, and it results k2K1 ꢂ4k1K1 (for glyoxylic acid), and k2K1 ꢃ2.5k1K1
A
uvic acid). For k1K1 >k2K1, it follows that k3 ꢂ50 (for glyoxylic acid) or k3 ꢂ10 (for pyr-
uvic acid). A reasonable conclusion regarding the relative contributions of the k2 and k3
paths is, thus, not possible, yet significant contribution of the k2 path in either redox
reaction is anticipated, as the reaction rates were found to increase significantly with
decreasing ionic strength of the reaction medium (Table 2), which indicates reaction
between oppositely charged species. The dominance of the k1 path may be attributed
to H-bonding involving the protonated oxo-bridge and the reducing species that ren-
ders a close proximity of the redox agents, in addition to the higher positive charge
of the protonated species [1H]5+ compared to 14+.
The oxidation rates of glyoxylic and pyruvic acid by periodate were found to
increase with increasing pH. We, thus, suggested that the reactions proceed through
nucleophilic attack of periodate on the C=O group of the a-oxo acids, where the
hydrated forms are non-reactive [24d]. In the present case, however, we found a reverse
trend: increase in rate with decreasing pH. This might be rationalized by the weaker
reactivity of the reducing anions in comparison with their parent acids, along with
the superior reactivity of [1H]5+ over [1]4+. Furthermore, we noted that glycolic acid

Helvetica Chimica Acta – Vol. 89 (2006)
1953
(HOꢀCH2ꢀCO2H) does basically not react with the Mn oxidant3). This observation
supports our assumption that the gem-diol forms of the reducing a-oxo acids are the
actual reactive species.
In deriving the rate law of Eqn. 10, we assumed K1[H+]ꢄ1 (so that TMn ꢅ[14+]).
From this inequality, along with the k1K1 values in Table 3, we can fix a range for K1
between 10ꢀ6 and 10. Though we are unable to more precisely estimate K1, a low
value is anticipated as we got no evidence for any UV/VIS spectroscopic changes of
the tetranuclear Mn complex (even at pH 2.0) recorded at pH 6.0. Reports so far avail-
able [13][31] for the very low basicities of oxo bridges in aqueous solution of a number
of multinuclear MnIV complexes support the approximation4) K1[H+]ꢄ1, as well as
K1 ꢄ1.
Instead of the above description of reaction sequences, where 14+ and [1H]5+ react
with RH and Rꢀ, a sequence of parallel reactions of 14+, [1H]5+, and [1H2]6+
([1H]5+ +H+ Ð [1H2]6+; K2) with Rꢀ may equally be in agreement with the observed
kinetics. The ꢂproton ambiguityꢁ, however, is resolved in the way that for being
[1H2]6+ to be a reactive species, its second-order rate constant, while reacting with
Rꢀ, would have to exceed far diffusion control (calculated data not shown), as
K2 ꢄK1, based on charge considerations. This alternative reaction path is, thus, not con-
sidered.
Table 3 also demonstrates that among the ki rate constants (i=1–4), k1 is the high-
est, k2 possibly also being high, while k4 is rather low. It appears that the [1H]5+ species
kinetically dominates over 14+. In fact, the very high reactivity of protonated oxo-
bridged dinuclear MnIV species has been reported [15], which are quickly reduced to
lower states in the presence of acids, illustrating that the MnIV oxidation level is desta-
bilized upon protonation of the oxo bridge.
Mechanism. In Eqns. 6–9, we presume that an initial rate-determining step is fol-
lowed by rapid reactions wherein the intermediate mixed-valent Mn species
(MnIV)3MnIII are quickly reduced by the excess reducing agents or by the radicals pro-
G
duced from the one-electron oxidation of them. Polymerization of acrylonitrile put for-
ward a definite evidence of the involvement of free radicals in the reaction course.
Glyoxylate radicals produced from one-electron oxidation of glyoxylate have already
been established by EPR measurements [24f], and pyruvate radicals have also been
proposed elsewhere [24j,k]. The follow-up reactions must be rapid for the following
3
)
After 3 h, 0.02M glycolic acid decreased the absorbance of the title Mn oxidant by less than 1% (con-
ditions: H2O (pH 3.0), 25.08, I=1.0M (NaNO3), Cbipy =3.0 mM).
)
The so far known acidity values of protonated oxo bridges in multinuclear MnIV complexes clearly
demonstrate that when the ligands are anionic and have stronger donor ability (as compared to pyr-
idine and related ligands like bipy), this leads to more electron density on the metal centers. There-
fore, the oxo bridges are not required to donate as much electron density to the metal centers and can
4
be involved in proton-acceptor chemistry. For example, the pKa of [(MnIV
)
A
ACHTREUNG
)
ACHTREUNG
A
bridging oxo groups in the latter complex, comparable to that used in our study, are much less ꢂpro-
ton-hungryꢁ than the former. From the inequality K1[H+] !1 at pH ca. 2, an estimated upper limit for
K1 lies around 1, and the lower limit lies at ca. 10ꢀ6, assuming that the maximum k1 is in the range one
expects for diffusion-controlled processes.

1954
Helvetica Chimica Acta – Vol. 89 (2006)
reasons: 1) the reducing agent may be very reactive radical species, in addition to the
pure reducing agents. 2) The most likely one-electron-reduced species of [1]4+ are
highly reactive ([(MnIV)3
N
N
[20]). 3) In the electrochemical reduction of 14+ at 0.50 V (vs. SCE) just before the first
one-electron reduction wave to attempt a single electron reduction, it was not possible
to stop the process at the stage of the one-electron-reduced species. Color fading of the
solution and EPR spectroscopic evidence indicated MnII at the final state [16]. These
observations clearly indicate the immediate collapsing nature of the (MnIV)3MnIII spe-
U
cies. 4) A cluster breakup is expected once one of the Mn-atoms is reduced to the MnII
state [13][32]. 5) Finally, tri-or dinuclear species formed from 14+ should bear two to
four H2O molecules in their coordination sphere. Such aqua complexes are known to
possess a very high kinetic activity towards redox processes [33]. The correct formula-
tion of the one-electron-reduced species of 14+ should have a protonated oxo bridge, as
the basicity of the bridge must increase remarkably when one MnIV is reduced to MnIII
[34]. Of the four Mn centers, one of the two terminal ones is expected to be reduced
first as these two MnIV centers should have a higher formal positive charge than the
two other central ones due to the coordination of four oxo ligands to the two central
MnIV sites.
The MnIV tetramer 14+ is coordinatively saturated, and an inner-sphere attachment
of reducing species is, thus, unlikely. Redox reactions of multinuclear higher-valent oxo-
bridged Mn and Fe complexes are often associated with H+ transfer [32][35–38]. The
oxo bridges in the one-electron-reduced Mn oxidant in this investigation, (MnIV)3MnIII,
G
must have increased basicity, and fast H+ transfer from bulk solvent is expected. The
slow rate limiting the one-electron reduction of the tetramer may, thus, be coupled
with simultaneous H+ transfer to the oxo bridge and, in fact, we found the rate con-
stants k1 to k4 (Table 3) determined in 95% D2O to be significantly lowered [39].
G
This clearly supports the occurrence of H+-coupled electron transfer as a key step in
the overall redox reaction.
Glyoxylate oxidation by high-valent metal centers is generally faster [24] than pyr-
uvate oxidation. The relative strengths of C(O)ꢀCOOH bonds in these two reducing
species may be responsible for this observation. The presence of an electron-donating
Me group in pyruvic acid increases the electron density at the adjacent C=O C-atom,
which reduces similar charge repulsion between the C=O and the COOH C-atoms. The
calculated CꢀC bond distances, however, in glyoxylic and pyruvic acids, estimated with
the aid of AM1, PM3, and ZINDO/1 semiempirical calculations, showed a reverse
result: the bond length in glyoxylic acid was determined as 1.505 (AM1), 1.524
(PM3), and 1.449 Š (ZINDO/1). For pyruvic acid, the values were 1.510 (AM1),
1.530 (PM3), and 1.458 (ZINDO/1). These calculations, when applied to the hydrated
form of glyoxylic acid, results in a bond length of 1.527 (AM1), 1.550 (PM3), and 1.469
Š (ZINDO/1). Similarly, hydrated pyruvic acid gives rise to values of 1.534 (AM1);
1.558 (PM3), and 1.473 Š (ZINDO/1). These results indirectly support the involvement
of hydrated glyoxylic and non-hydrated pyruvic acid as reactive species, respectively.
The lower C(O)ꢀCOOH bond energy in hydrated glyoxylic acid compared to that in
regular pyruvic acid may, thus, be a driving force for faster initial electron transfer lead-
ing to faster overall oxidation of glyoxylic acid. Though we find it difficult to exactly
pinpoint a single act in the rate-determining step of the title redox system, formation

Helvetica Chimica Acta – Vol. 89 (2006)
1955
of a hydrated acyl radical (RCC (OH)2) accompanied by one-electron reduction of the
oxidant, and simultaneous H+ transfer to the oxo bridge of the one-electron-reduced
oxidant, may define the rate-determining step. Formation of acyl radicals by one-elec-
tron oxidants like MnIII [40] or VV [24] has been proposed earlier.
The initial decarboxylation of glyoxylic acid to oxalic acid, followed by the oxida-
tion of oxalic acid to CO2, seems unlikely, as we verified that oxalic acid reacts with
14+ at a rate at least ten times slower than the oxidation of glyoxylic acid under compa-
rable conditions. Again, it is known that acetaldehyde, on oxidation by a metal ion in
mineral acid, yields both formic and acetic acid [41][42]. However, in the present inves-
tigation, we found only acetic acid as the oxidation product of pyruvic acid. Initial
decarboxylation of pyruvic acid, followed by oxidation of acetaldehyde, is, thus,
ruled out. Moreover, we noted that acetaldehyde does not react with the Mn oxidant
when used in concentrations similar to those used for the present reducing agents.
A Junior Research Fellowship awarded to S. D. by the Council of Scientific and Industrial Research,
New Delhi, India, is gratefully acknowledged.
Experimental Part
Materials. The complex salt hydrate [Mn4(m-O)6(bipy)6][ClO4]4 ·2 H2O was prepared according to a
U
literature procedure [16]. Note that one H2O molecule is easily lost [16], which explains why the elemen-
tal analysis of the material matched with that calculated for the monohydrate (anal. calc. for
C60H50Cl4Mn4N12O23: C 43.16, H 2.99, N 10.07; found C 43.39, H 3.03, N 10.00); the tetramer 14+ used
in all the experiments, thus, appears to be the sufficiently pure monohydrate. The preparation, standard-
ization, and storage of glyoxylic acid, pyruvic acid, and sodium nitrate were described earlier [24k][37].
2,2’-bipyridine (bipy; Sigma) was used as received. D2O (99.9%) was purchased from Sigma or Merck; all
G
other chemicals were of reagent grade. Doubly distilled, deionized and freshly distilled H2O was used
throughout.
Equilibrium Measurements. The dissociation constants Ka of glyoxylic and pyruvic acid were deter-
mined by titration with CO2-free NaOH in both H2O and D2
(NaNO3)) using a Metrohm 736-GP-Titrino autotitrator, as described earlier [38].
A
Stoichiometric Measurements. The generation of gaseous CO2 as oxidation product of both glyoxylic
and pyruvic acid was quantitatively established by GC analysis, as described earlier [43]. Formation of
formic and acetic acid, respectively, from the oxidation of glyoxylic and pyruvic acid was secured with
the aid of chromotropic acid and the La(NO3)3/I2 test [44]. Mn2+ and bipy did not interfere.
Physical Measurements and Kinetics. UV/VIS Absorbance spectra were recorded on a Shimadzu
1601-PC spectrophotometer using 1-cm quartz cells. The kinetics were monitored in situ in the ꢂkinetics
modeꢁ of the instrument, and in a thermostated (25.0Æ0.18) cell housing (CPS-240A) at 420 nm, where
all the reaction partners (except the Mn oxidant; e=7500 Mꢀ1 cmꢀ1) are transparent. A few runs were also
performed at other wavelengths: 380, 450, and 500 nm. The ionic strength was normally maintained at
1.0M with NaNO3. Excess of the 2,2’-bipyridine concentration, Cbipy
(=[(bipy)H)+]+[bipy]), was in the
U
range 1–60 mM, and acted as a good buffer5) in the pH range studied (2.5–6.0), as measured with an
Orion-Ross combined-electrode system (Model 81-02), before and after the reaction. The electrode
5
)
Excess bipy present in soln. nicely buffers the reaction within 0.03 pH units above pH 3.2. Below pH
3.0, an increase in pH of ca. 0.20–0.26 was observed at the end of the reactions. This might be due to
the removal of H+ from the reaction medium by the released bipy. Plots of log10
(absorbance) vs. time
N
in this low-pH region were found to be slightly curved upwards after ca. 60–70% completion of the
reaction. The first-order rate constants in such situations were computed from the slope of the initial
linear plots.

1956
Helvetica Chimica Acta – Vol. 89 (2006)
was calibrated to read ꢀlog10 [H+] with the help of a calibration curve constructed by plotting the pH-
meter reading against ꢀlog10[H+] [45]. For reactions in D2
C
REFERENCES
[1] K. N. Ferreira, T. M. Iverson, K. Maghlaoui, J. Barber, S. Iwata, Science 2004, 303, 1831; J. Yano, J.
Kern, K-D. Irrgang, M. J. Latimer, U. Bergmann, J. Biesiadka, B. Loll, K. Sauer, J. Messinger, A.
Zouni, V. K. Yachandra, Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 12047.
[2] S. Mukhopadhyay, S. K. Mandal, S. Bhaduri, W. H. Armstrong, Chem. Rev. 2004, 104, 3981.
[3] ꢂPhotosynthetic Water Oxidation: Special Dedicated Issueꢁ, Ed. J. Nugent, Biochim. Biophys. Acta
2001, 1503, 1.
[4] ꢂManganese Redox Enzymesꢁ, Ed. V. L. Pecoraro, J. Wiley & Sons, New York, 1992.
[5] J. Messinger, W. Lubitz, Phys. Chem. Chem. Phys. 2004, 6, E11.
[6] H. Ishikita, B. Loll, J. Biesiadka, W. Saenger, E.-W. Knapp, Biochemistry 2005, 44, 4118.
[7] P. Joliot, G. Barbieri, R. Chabaud, Photochem. Photobiol. 1969, 10, 309.
[8] B. Kok, B. Forbush, M. McGloin, Photochem. Photobiol. 1970, 11, 457; M. Haumann, C. Müller, P.
Liebisch, L. Iuzzolino, J. Dittmer, M. Grabolle, T. Neisius, W. Meyer-Klaucke, H. Dau, Biochemistry
2005, 44, 1894.
[9] P. E. M. Siegbahn, Inorg. Chem. 2000, 39, 2923.
[10] C. W. Hoganson, G. T. Babcock, Science 1997, 277, 1953.
[11] V. K. Yachandra, V. J. DeRose, M. J. Latimer, I. Mukerji, K. Sauer, M. P. Klein, Science 1993, 260,
675.
[12] J. P. Collman, J. T. McDevitt, G. T. Yee, C. R. Leidner, L. G. McCullough, W. A. Little, J. B. Tor-
rance, Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 4581; J. B. Vincent, G. Christou, Inorg. Chim. Acta
1987, 136, L41; C. Meinke, V. A. SolØ, P. Pospisil, H. Dau, Biochemistry 2000, 39, 7033.
[13] C. E. DubØ, D. W. Wright, S. Pal, P. J. Bonitatebus Jr., W. H. Armstrong, J. Am. Chem. Soc. 1998,
120, 3704.
[14] K. S. Hagen, T. D. Westmoreland, M. J. Scott, W. H. Armstrong, J. Am. Chem. Soc. 1989, 111, 1907.
[15] M. J. Baldwin, T. M. Stemmler, P. J. Riggs-Gelasco, M. L. Kirk, J. E. Penner-Hahn, V. L. Pecoraro, J.
Am. Chem. Soc. 1994, 116, 11349.
[16] C. Philouze, G. Blondin, J. J. Girerd, J. Guilhem, C. Pascard, D. Lexa, J. Am. Chem. Soc. 1994, 116,
8557.
[17] C. E. DubØ, D. W. Wright, W. H. Armstrong, J. Am. Chem. Soc. 1996, 118, 10910.
[18] E. J. Larson, P. J. Riggs, M. L. Kirk, J. E. Penner-Hahn, V. L. Pecoraro, J. Chem. Soc., Chem. Com-
mun. 1992, 102.
[19] K. Wieghardt, U. Bossek, B. Nuber, J. Weiss, J. Bonvoisin, M. Corbella, S. E. Vitols, J. J. Girerd, J.
Am. Chem. Soc. 1988, 110, 7398.
[20] G. Blondin, R. Davydov, C. Philouze, M. F. Charlot, S. Styring, B. Škermark, J. J. Girerd, A. Bous-
sac, J. Chem. Soc., Dalton Trans. 1997, 4069.
[21] A. White, in ꢂPrinciples of Biochemistryꢁ, 2nd edn., McGraw-Hill, New York, 1954, p. 458; D. L. Nel-
son, M. M. Cox, in ꢂLehninger Principles of Biochemistryꢁ, 3rd edn., Macmillan Press, Ltd., UK,
2003, Chapt. 16, p. 567; ꢂPlant Biochemistryꢁ, Eds., P. M. Dey, J. B. Harborne, Academic Press,
New York, 1997, pp. 122 and 125; L. Stryer, ꢂBiochemistryꢁ, 4th edn., W. H. Freeman & Co., New
York, 1995, pp. 509, 523.
[22] E. H. Rodd, in ꢂChemistry of Carbon Compoundsꢁ, Elsevier, Amsterdam, 1952, Part B, Vol. 1, p.
850.
[23] E. S. G. Barron, in ꢂTrends in Physiology and Biochemistryꢁ, Academic Press, New York, 1952, p.
471; U. Urzffla, P. J. Kersten, R. Vicuna, Appl. Environ. Microbiol. 1998, 64, 68; S. W. Ragsdale,
Chem. Rev. 2003, 103, 2333.

Helvetica Chimica Acta – Vol. 89 (2006)
1957
[24] a) K. K. Sengupta, J. Indian Chem. Soc. 1964, 41, 423; b) W. J. Albery, R. P. Bell, A. L. Powel,
Trans. Farad. Soc. 1965, 61, 1194; c) K. K. Sengupta, T. Sarkar, Tetrahedron 1975, 31, 123; d)
L. Maros, I. Molnµr-Perl, L. KçvØr, J. Chem. Soc., Perkin Trans. 2. 1976, 1337; e) K. K. Sengupta,
H. R. Chatterjee, Inorg. Chem. 1978, 17, 2429; f) B. Neumann, O. Steinbock, S. C. Müller, N. S.
Dalal, J. Phys. Chem. 1996, 100, 12342; g) R. Banerjee, R. Das, A. K. Chakraborty, J. Chem.
Soc., Dalton Trans. 1990, 3277; h) P. Manikyamba, React. Kinet. Catal. Lett. 2003, 78, 169; i) A.
Das, S. Mukhopadhyay, Polyhedron 2004, 23, 895; j) A. Das, S. Mukhopadhyay, Transition Met.
Chem. 2004, 29, 797; k) B. B. Dhar, R. Mukherjee, S. Mukhopadhyay, R. Banerjee, Eur. J.
Inorg. Chem. 2004, 4854; l) S. Das, G. S. Chaubey, M. K. Mahanti, Kinet. Catal. 2002, 43, 794.
[25] H. Strehlow, Z. Elektrochem. 1962, 66, 392; Y. I. Turꢁyan, Croat. Chem. Acta 1998, 71, 727; P. E.
Sorensen, K. Bruhn, F. Lindelov, Acta Chem. Scand. 1974, A28, 162; J. Kuta, P. Valenta, Collect.
Czech. Chem. Commun. 1963, 28, 1593; A. R. Rendina, J. D. Hermes, W. W. Cleland, Biochemistry
1984, 23, 5148; M.-L. Ahrens, Ber. Bunsenges. Phys. Chem. 1968, 72, 691; Y. I. Turꢁyan, Croat.
Chem. Acta 1999, 72, 13; M. Xie, A. Feng, Huaxue Tongbao 1986, 10, 40; H. Patting, H. Strehlow,
Ber. Bunsenges. Phys. Chem. 1969, 73, 534; J. Damitio, G. Smith, J. E. Meany, Y. Pocker, J. Am.
Chem. Soc. 1992, 114, 3081.
[26] a) A. Hilton, D. L. Leussing, J. Am. Chem. Soc. 1971, 93, 6831; b) D. L. Leussing, E. M. Hanna, J.
Am. Chem. Soc. 1966, 88, 696; c) G. ꢃjelund, I. Wadsç, Acta Chem. Scand. 1967, 21, 1408.
[27] D. L. Leussing, E. M. Hanna, J. Am. Chem. Soc. 1966, 88, 693.
[28] R. M. Smith, A. E. Martell, in ꢂCritical Stability Constantsꢁ, Plenum Press, New York, 1976, Vol. 3,
pp. 3, 24.
[29] R. M. Smith, A. E. Martell, in ꢂCritical Stability Constantsꢁ, Plenum Press, New York, 1976, Vol. 2, p.
234.
[30] S. K. Ghosh, R. N. Bose, E. S. Gould, Inorg. Chem. 1987, 26, 2688; K. Lemma, A. M. Sargeson, L. I.
Elding, J. Chem. Soc., Dalton Trans. 2000, 1167; T. Shi, J. Berglund, L. I. Elding, Inorg. Chem. 1996,
35, 3498.
[31] H. H. Thorp, J. E. Sarneski, G. W. Brudvig, R. H. Crabtree, J. Am. Chem. Soc. 1989, 111, 9249.
[32] M. C. Ghosh, J. W. Reed, R. N. Bose, E. S. Gould, Inorg. Chem. 1994, 33, 73.
[33] S. Chaudhuri, S. Mukhopadhyay, R. Banerjee, J. Chem. Soc., Dalton Trans. 1995, 621; S. Kundu, A.
K. Bhattacharya, R. Banerjee, J. Chem. Soc., Dalton Trans. 1996, 3951; A. K. Bhattyacharya, A. B.
Mondal, R. Banerjee, J. Chem. Soc., Dalton Trans. 1997, 2351; R. Banerjee, B. Mondal, S. Kundu, J.
Chem. Soc., Dalton Trans. 1997, 4341.
[34] W. Rüttinger, G. C. Dismukes, Chem. Rev. 1997, 97, 1; J. S. Vrettos, J. Limburg, G. W. Brudvig, Bio-
chim. Biophys. Acta 2001, 1503, 229.
[35] S. Banerjee, U. Roy Choudhury, R. Banerjee, S. Mukhopadhyay, J. Chem. Soc., Dalton Trans. 2002,
2047; M. J. Baldwin, V. L Pocoraro, J. Am. Chem. Soc. 1996, 118, 11325; C. Carra, N. Iordanova, S.
Hammes-Schiffer, J. Am. Chem. Soc. 2003, 125, 10429.
[36] J. Bhattacharyya, S. Mukhopadhyay, Transition Met. Chem. 2006, 31, 256.
[37] J. Bhattacharyya, K. Dutta, S. Mukhopadhyay, Dalton Trans. 2004, 2910.
[38] A. Das, S. Mukhopadhyay, Helv. Chim. Acta 2005, 88, 2561; J. Bhattacharyya, S. Mukhopadhyay,
Helv. Chim. Acta 2005, 88, 2661.
[39] W. H. Albery, in ꢂProton Transfer Reactionsꢁ, Ed. E. Caldin, V. Gold, J. Wiley & Sons, New York,
1975, Chapt. 9; J. A. Gilbert, S. W. Gersten, T. J. Meyer, J. Am. Chem. Soc. 1982, 104, 6872; R.
A. Binstead, M. E. McGuire, A. Dovletoglou, W. K. Seok, L. E. Roecker, T. J. Meyer, J. Am.
Chem. Soc. 1992, 114, 173; I. J. Rhile, J. M. Mayer, J. Am. Chem. Soc. 2004, 126, 12718; M. Sjçdin,
S. Styring, H. Wolpher, Y. Xu, L. Sun, L. Hammarstrçm, J. Am. Chem. Soc. 2005, 127, 3855.
[40] A. Y. Drummond, W. A. Waters, J. Chem. Soc. 1955, 497; T. Watanabe, S. Katayama, M. Enoki, Y.
Honda, M. Kuwahara, Eur. J. Biochem. 2000, 267, 4222; V. I. Bunik, C. Sievers, Eur. J. Biochem.
2002, 269, 5004.
[41] J. Shorter, C. N. Hinshelwood, J. Chem. Soc. 1950, 3276.
[42] J. Shorter, J. Chem. Soc. 1950, 3425.
[43] K. Dutta, S. Bhattacharjee, B. Chaudhuri, S. Mukhopadhyay, J. Environ. Monit. 2002, 4, 754.
[44] F. Feigl, in ꢂSpot Test in Organic Analysisꢁ, 5th edn., Elsevier, London, 1956, p. 331.

1958
Helvetica Chimica Acta – Vol. 89 (2006)
[45] S. Mukhopadhyay, R. Banerjee, J. Chem. Soc., Dalton Trans. 1994, 1349; S. Banerjee, U. Roy
Choudhury, B. C. Ray, R. Banerjee, S. Mukhopadhyay, Anal. Lett. 2001, 34, 2797; H. N. Irving,
M. G. Miles, L. D. Pettit, Anal. Chim. Acta 1967, 38, 475.
[46] P. K. Glasoe, F. A. Long, J. Phys. Chem. 1960, 64, 188; P. Salomaa, L. L. Schaleger, F. A. Long, J.
Am. Chem. Soc. 1964, 86, 1.
Received April 25, 2006