Kinetics of oxidation of ascorbic acid and 1,4-dihydroxybenzene by semiquinone radical bound to ruthenium(II)

Article abstract of DOI:10.1039/b104585g

In 40% (v/v) MeOH-H₂O media, containing [H⁺] (0.001-0.038 mol dm^-3), the semiquinone (sq) radical, bound to Ru(II) in [Ru(bpy)₂(sq)]⁺ 1, oxidises ascorbic acid (H₂A) to dehydroascorbic acid (A), and 1,4-dihydroxybenzene (H₂Q) to p-benzoquinone (Q); 1 is itself reduced to [Ru(bpy)₂(Hcat)]⁺ 2H. The reactions are centred at sq not Ru(II). The sq/cat couple in 1 is reversible and its E_1/2 increases with increasing [H⁺]; rate of chemical reduction of 1 to 2H increases in parallel. Rate increases also with increasing mole percent of D₂O in the solvent suggesting a preliminary protonation equilibrium producing 1H, in which a H⁺ binds to the π-electron cloud of Ru(II)-bound sq. Under the experimental conditions, the kinetically significant species are 1H, H₂Q, H₂A and HA^-. The kinetic activity of HA^- ion is only ≈200 times more than that of H₂A. This testifies against a purely outer-sphere mechanism and suggests significant electronic interaction between the redox partners. Increased percentage of MeOH in the solvent decreases λ_max for the LMCT band; reaction rate for ascorbic acid decreases in parallel.

Full text of DOI:10.1039/b104585g

View Article Online / Journal Homepage / Table of Contents for this issue
Kinetics of oxidation of ascorbic acid and 1,4-dihydroxybenzene
by semiquinone radical bound to ruthenium(II)
a
b
a
b
Debjani Ghosh, Atindra D. Shukla, Rupendranath Banerjee* and Amitava Das*
a
Department of Chemistry, Jadavpur University, Calcutta 700 032, India.
E-mail: dsthcr28@cal2.vsnl.net.in
b
Silicates and Catalysis Discipline, Central Salt and Marine Chemicals Research Institute,
Bhabnagar 364002, India. E-mail: salt@csir.res.in
Received 24th May 2001, Accepted 12th November 2001
First published as an Advance Article on the web 12th February 2002
ϩ
Ϫ3
In 40% (v/v) MeOH–H O media, containing [H ] (0.001–0.038 mol dm ), the semiquinone (sq) radical, bound
2
ϩ
to Ru() in [Ru(bpy) (sq)] 1, oxidises ascorbic acid (H A) to dehydroascorbic acid (A), and 1,4-dihydroxybenzene
H Q) to p-benzoquinone (Q); 1 is itself reduced to [Ru(bpy) (Hcat)] 2H. The reactions are centred at sq not Ru().
2 2
ϩ
The sq/cat couple in 1 is reversible and its E1/2 increases with increasing [H ]; rate of chemical reduction of 1 to 2H
2
2
ϩ
(
increases in parallel. Rate increases also with increasing mole percent of D O in the solvent suggesting a preliminary
2
ϩ
protonation equilibrium producing 1H, in which a H binds to the π-electron cloud of Ru()-bound sq. Under the
Ϫ
Ϫ
experimental conditions, the kinetically significant species are 1H, H Q, H A and HA . The kinetic activity of HA
2
2
ion is only ≈200 times more than that of H A. This testifies against a purely outer-sphere mechanism and suggests
2
significant electronic interaction between the redox partners. Increased percentage of MeOH in the solvent
decreases λmax for the LMCT band; reaction rate for ascorbic acid decreases in parallel.
The electron and proton transfer on the acceptor side of the
primary photochemical reaction centre in bacterial photo-
synthesis involves the formation of a semiquinone as the first
recrystallised from ethanol. All reported data are at 25.0 ЊC,
Ϫ3
I 0.10 mol dm maintained with NaClO . Crystals of NaClO ؒ
4
4
H O were made by neutralising a 35% HClO solution (made by
2
4
1
step in the formation of ubiquinol. The semiquinone is a stable
dilution of 70% HClO , G.R., E. Merck) with solid NaHCO3
4
state, which remains attached to the protein pocket until the
second reduction of quinol. Semiquinone related ligands
attached to transition metal ions represent models for the
primary photochemical donor–acceptor centre in bacterial
photosynthesis and for the biological transport of iron for some
(G.R., E. Merck) followed by evaporation to incipient crystal-
lisation. All other reagents used were of reagent grade.
Microanalyses (C, H, N) were performed using a Perkin-
Elmer 4100 elemental analyzer. IR spectra were recorded as
KBr pellets using Perkin-Elmer Spectra GX 2000 spectrometer.
Absorbance and electronic spectra were recorded with a
Shimadzu (1601 PC) spectrophotometer using 1.00 cm quartz
cells. Fast-atom bombardment measurement was carried out
on a VG-ZAB instrument using 3-nitrobenzyl alcohol as
matrix. Electrochemical experiments were performed using a
CH 660A (USA) electrochemical instrument; a conventional
three-electrode cell assembly was used. A saturated Ag/AgCl
reference electrode and platinum working electrode were used.
2
3
enzymes. They may also serve as models for ascorbate oxidase.
Many groups have, therefore, examined the structural charac-
terisation, electrochemistry, spectroscopy, magnetic properties
and electronic structure of transition metal complexes involv-
4–6
ing non-innocent semiquinone related ligands. The members
of the catechol–semiquinone redox series involving bis-
(
bipyridine)ruthenium(/) complexes appear particularly well
characterised structurally and with respect to the metal and
ligand oxidation states. However, there appears no report on the
kinetics of redox reactions of coordinated semiquinone, in spite
of their relevance in understanding the photochemical reaction
centre.
؊
؊
[
Ru(bpy) (sq)]X (X ؍ PF or Cl )
2 6
ϩ
The hexafluorophosphate and chloride salts of [Ru(bpy) (sq)] ,
2
8
7
ϩ
1 were prepared by known methods with some modifications.
The ESR spectra of the complex [Ru(bpy) (sq)] (1, bpy is
2
[
Ru(bpy) Cl ]ؒ2H O (200 mg, 0.385 mmol), 1,2-dihydroxy-
2
,2Ј-bipyridine, sq is semiquinone radical) show that the
2 2 2
benzene (43 mg, 0.385 mmol) and KOH (44 mg, 0.78 mmol)
unpaired electron is located primarily on the semiquinone
ligand and therefore Ru()sq is the proper description of 1. The
redox changes in the semiquinone ligand occur at potentials
well separated from those centred on ruthenium. The complex 1
therefore affords a nice chance to study the kinetics of redox
changes in coordinated semiquinone without complications
due to simultaneous redox changes at the metal centre.
3
were refluxed in ethanol (50 cm ) for 6 h under a dinitrogen
blanket. To this resulting mixture FcPF6 (0.127 mg, 0.385
mmol) was added and stirred at room temperature for 1/2 h.
3
The volume was reduced to ≈5 cm and an aqueous KPF sol-
6
ution was added. The precipitate was filtered off, air dried and
purified by column chromatography using a silica column and
CH CN–water (98 : 2, v/v) solution of NH PF as the eluent.
3
4
6
The first major fraction was collected and concentrated in vacuo
3
Experimental
(≈5 cm ). Excess NH PF was removed by solvent extraction in
4
6
the aqueous layer. The desired compound was collected in the
Materials
CH Cl layer and dried in vacuo. Yield: 155 mg, 66%. Elemental
2
2
All solutions were prepared in double distilled, freshly boiled
analyses: Calc. for C H N O PF Ru: C, 46.82; H, 3.15; N,
26 21 4 2 6
ϩ
water. -Ascorbic acid (H A, G.R., E. Merck) was stored in the
8.14. Found: C, 46.0; H, 3.18; N, 8.10%. FAB MS: m/z 664 (M ,
2
ϩ
Ϫ1
dark. 1,4-Dihydroxybenzene (H Q, quinol, A.R., B.D.H) was
≈5%), 519 (M Ϫ PF , >25%). IR (KBr, cm ): 1610, 1580
2
6
1
220
J. Chem. Soc., Dalton Trans., 2002, 1220–1225
This journal is © The Royal Society of Chemistry 2002
DOI: 10.1039/b104585g

View Article Online
Ϫ
6
(
C᎐C and C᎐N), 1444 (semiquinone streching), 835 (PF ). For
In terms of absorbance, eqn. (5) becomes,
᎐
᎐
the corresponding chloride salt, the crude reaction mixture was
evaporated to dryness and chromatographed on an alumina
column (grade III) with acetonitrile–toluene (60 : 40, v/v) as
eluent. The second major fraction was collected and found to
be the desired product. Yield: 35%. Elemental analyses: Calc.
(6)
for C H N O ClRu: C, 56.01; H, 3.77; N, 10.05. Found: C,
26
21
4
2
kA and associated errors were evaluated by least-squares
analysis (ORIGIN 4.0) of the LHS of eqn. (6) against time data
and using known values for A , A , ε and ε .
ϩ
5
6.0; H, 3.80; N, 10.0%. FAB MS: m/z 555 (M , ≈10%), 519
ϩ Ϫ1
(
M Ϫ Cl, ≈35%). IR (KBr, cm ): 1608, 1578 (C᎐C and C᎐N),
o
∝
A
B
1
448 (semiquinone stretching). The λmax for the characteristic
Reaction stoichiometry was determined by spectrophoto-
metric titration (λ, 933.5 nm) of a solution of complex 1 with
different amounts of reducing agent in the presence of excess
HClO in 40% (v/v) MeOH–H O media.
MLCT band of the complex in MeOH–H O mixed solvents of
2
different solvent composition are given below:
4
2
Results and discussion
Reduction product of 1
Kinetics
The complex 1 is stable in acidic MeOH–H O media and its
2
spectrum [Fig. 1A] did not change for at least 24 hours. How-
Most of the kinetics were studied in 40% (v/v) MeOH–H O
using the MeOH-soluble PF6 salt of 1. The Cl salt was used
for reactions in H O and H O–D O media. The Cl salt was
2
Ϫ
Ϫ
Ϫ
2
2
2
otherwise avoided due to its hygroscopic nature.
Kinetics were monitored at 933.5 nm, the λmax of complex 1
in 40% (v/v) MeOH–H O media. At this wavelength, the
2
reducing agents and their oxidation products do not absorb,
while absorbance of 2H, i.e., the reduction product of 1, is very
small. The kinetics were monitored in situ in the thermostatted
cell housing (CPS-240A), following the decrease in absorbance
with time. For ascorbic acid, kinetics were measured both under
the second order and the first order conditions. Reactions of
quinol were followed only under the first order conditions. For
second order conditions, [1] : [R] = 1 : 1 was maintained. Under
the first order conditions, an excess of reducing agent over
complex 1 was used. All the reactions were carried out in the
presence of a large excess of HClO in the range 0.001 to 0.038
4
Ϫ3
mol dm . For a reaction 2A ϩ B = product, under first order
Ϫ3
Fig. 1 Spectra of 0.05 mmol dm solutions of 1 (A) and its mixture
conditions:
Ϫ3
with 0.10 mmol dm ascorbic acid; immediately after mixing (B) and
ϩ
after 2 (C), 4 (D), 7 (E), 12 (F) and 20 (G) min; [H ] 0.01 M, I 0.10 M.
(
1)
ever, the NIR absorption band gradually loses its intensity
when H A or H Q is added. The reducing agents slightly affect
the other spectral regions. A family of spectra changing with
2
2
The integrated form of eqn. (1) in terms of absorbance is
eqn.(2). The absorbance (A )–time (t) data for the reactions
t
time after the addition of H A is shown in Fig. 1. The final
2
7
under the first order conditions were fitted into eqn. (2) using
spectrum [Fig. 1G] closely resembles those reported for [Ru-
bpy) (Hcat)] 2H, the singly protonated form of the catechol
complex [Ru(bpy) (cat)] , (2, cat is the catecholate form of
ϩ
the nonlinear least-squares regression routine of the program
(
2
9
ϩ
package ORIGIN 4.0. The values for k and associated errors
0
2
were thus determined.
the ligand). 2H lacks the LLCT bands characteristic of 2 and
can be easily distinguished from 2. Spectral changes similar to
ln(A Ϫ A ) = Ϫ2k t ϩ ln(A Ϫ A )
(2)
t
∞
0
0
∞
those in Fig. 1 are observed with H Q and we conclude that in
2
HClO4 media both ascorbic acid and 1,4-dihydroxybenzene
reduce 1 to 2H under the present experimental conditions.
Neither 1 nor 2H aquates or decomposes during the reactions
studied here.
The A –t data, collected under second order conditions ([1] :
R] = 1 : 1) were fitted to a typical second order reaction for
which,
t
[
Ϫd[1]/dt = 2k [1][R]
(3)
The break point in the absorbance versus total concentration
of a reducing agent, T (X = Q for H Q and A for ascorbic acid)
A
x
2
Comparing eqns. (1) and (3), one gets
was observed at [1] : T = 2 : 1 for both H A and H Q. Fig. 2 is a
x
2
2
typical example recorded with H Q. The most likely oxidation
2
k = k [R]
(4)
products of H A and H Q are therefore dehydroascorbic acid
o
A
2
2
(
A) and 1,4-benzoquinone (Q) respectively.
10
For [1] : [R] = 1 : 1, the integrated form for eqn. (3) is eqn. (5)
Kinetics and reaction scheme
With excess [R], A –t data followed first order kinetics in [1] and
yielded an excellent fit to eqn. (2) for more than three half-lives
see Fig. 3). A –t data collected under second order conditions
(
5)
t
(
t
where ao is the initial concentration of both 1 and R. The
amount of complex 1 consumed at any time t is 2x while that
for R is x.
([1] = [R]) yielded equally an good fit to the second order eqn.
(6) (see Fig. 4). The kA and k values have been collected in
o
Tables 1 and 2 respectively.
J. Chem. Soc., Dalton Trans., 2002, 1220–1225
1221

View Article Online
a
Table 1 Second order rate constants, kA, for the reduction of 1 to
2
H by ascorbic acid
Ϫ3
3
Ϫ1 Ϫ1
[
HClO ]/mmol dm
k /dm mol
s
4
A
1
2
5
0
5
0
5
0
4
8
.0
.0
.0
9
9.4
9.2
9
10.4
12.3
13.8
17.5
19.5
21.2
1
1
2
2
3
3
3
a
Ϫ3
Ϫ3
[
complex] 0.10 mmol dm ; T A 0.10 mmol dm ; T 25.0 ЊC; I 0.10 mol
Ϫ3
dm (NaClO ); 40% (v/v) MeOH–H O media. Error in k , 3–5%.
4
2
A
a
Table 2 First order rate constants, k , for the reduction of 1 to 2H
o
Fig. 2 Stoichiometry of the reaction between 1 and 1,4-dihydroxy-
Ϫ3
ϩ
Ϫ3
4
Ϫ1
Ϫ3
Ϫ3
T
X
/mmol dm
[H ]/mmol dm
10 k
/s
o
benzene. [1] 0.10 mmol dm ; [HClO ] 10 mmol dm ; T 25.0 ЊC;
4
Ϫ3
I 0.10 mol dm (NaClO ); 40% (v/v) MeOH–H O media. λ 933.5 nm.
4
2
(
X = A)
0
0
.40
.50
10
44
30
1.0
2
5
0
5
0
5
0
.0
.0
40
49
50
62
77.5
80
84
57.5
61
65
1
1
2
2
3
0
0
0
.60
.70
.80
10
X = Q
1
.00
3.0
0.29
0.415
0.80
1.5
1.65
2.08
2.5
2.8
3.15
2.0
2.3
2.5
2.5
2.9
5
0
5
0
5
0
4
8
.0
1
1
2
2
3
3
3
2
3
5
0
5
0
5
0
.0
.0
.0
15
Fig. 3 A typical first order plot for the oxidation of ascorbic acid.
1
1
2
2
3
Ϫ3
Ϫ3
Ϫ3
[
1] 0.10 mmol dm ; T A 0.50 mmol dm ; [HClO ] 2.0 mmol dm ;
4
Ϫ3
T 25.0 ЊC; I 0.10 mol dm (NaClO ); 40% (v/v) MeOH–H O media.
4
2
2.8
2.9
2.9
b
3
2
.0
.95
c
a
Ϫ3
Ϫ3
[
complex] 0.10 mmol dm ; T 25.0 ЊC; I 0.10 mol dm (NaClO );
4
b
4
0% (v/v) MeOH–H O media. Repeated with minimum stray light.
Repeated after purging dinitrogen, error in k , 2–5%.
2
c
o
Some reactions were carried out avoiding ambient light as far
as possible by using volumetric flasks coated on the outside
with a black paint and by bringing the spectrophotometer cell
to the light path only at the instance of absorbance measure-
ments during kinetic studies. These measures minimised stray
light but did not change the reaction rate significantly. The reac-
tion rate also changed little when the medium was purged with
purified dinitrogen.
Rates of oxidation of hydroquinone and ascorbic acid by
ϩ
transition metal complexes generally increase if [H ] is
1
1a,b
Ϫ
decreased.
This is because the conjugate base forms, HA
Ϫ
and HQ of the reducing agents are kinetically much more
active than their parent acids. Additionally, for metal cations
as oxidant, the hydrolysed form of the cation dominates at
Fig. 4 A typical second order plot for the oxidation of ascorbic acid.
Ϫ3
Ϫ3
Ϫ3
[
1] 0.1 mmol dm ; T A 0.10 mmol dm ; [HClO ] 20.0 mmol dm ,
4
Ϫ3
T 25.0 ЊC; I 0.10 mol dm (NaClO ); 40% (v/v) MeOH–H O media.
4
2
1
222
J. Chem. Soc., Dalton Trans., 2002, 1220–1225

View Article Online
ϩ
The observed dependence of rate and E1/2 on [H ] indicates
either a preliminary protonation equilibrium or a simultaneous
transfer of an electron and a proton in the rate-determining
step. We studied some kinetics in a H O–D O mixture to dis-
2
2
tinguish between the two possibilities. It is known that the rate
increases significantly with increased percentage of D O in the
2
H O–D O mixture if a preliminary acid–base equilibrium step
2
2
13
14–17
exists, but for electroprotic reactions, the rate decreases.
Such a kinetic isotope effect kH2O/k ≥ 31.1 has been reported
recently by Meyer and co-workers.
2
D O
1
8
In contrast the rate in H O is closely similar to the rate in
2
D O for a simple electron transfer reaction. Fig. 7 shows that in
2
Fig. 5 (A) Cyclic voltammogram of 1 in 40% (v/v) MeOH–H O media
Ϫ3
2
Ϫ3
Ϫ3
with (a) 0.01 mol dm , (b) 0.003 mol dm and (c) 0.001 mol dm
HCl; (B) Cyclic voltammogramm of ferrocene in 40% (v/v) MeOH–
H O media with (a) 0.0 mol dm , (b) 0.003 mol dm and (c) 0.01 mol
Ϫ3
Ϫ3
2
Ϫ3
Ϫ1
dm HCl. Scan rate 200 mV s
.
ϩ
12
reduced [H ] and Pellizzetti noticed that hydrolysed metal
cations are kinetically superior oxidants when the reductant
contains a transferable hydrogen atom, as in the cases of ascor-
bic acid and quinol. In the present study, no indication could be
ϩ
obtained for increased rate on reduction of [H ]. Rather, the
ϩ
ϩ
rate is enhanced at higher [H ]. Increased [H ] also assists the
electroreduction of the sq radical. The sq/cat couple of complex
1
is completely reversible. Its E1/2 value in 40% (v/v) MeOH–
H O is 0.281 V (NHE) at pH 4.6. The potential increases with
2
ϩ
Fig. 7 kA versus mol% of D O in a D O–H O mixture. [H ] 1.0 mmol
2
2
2
decreasing pH (Fig. 5A) and the E1/2 versus pH graph is a
straight line (Fig. 6) with a slope (Ϫ0.061 ± 0.002) very close to
the theoretical value, Ϫ0.059, expected when one proton per
mole of 1 is involved in the electron transfer reaction at 25 ЊC.
Ϫ3
Ϫ3
Ϫ3
dm (A) and 30 mmol dm (B). [1] 0.10 mmol dm ; T A 0.05 mmol
dm ; T 25.0 ЊC; I 0.10 mol dm (NaClO ); 40% (v/v) MeOH–H O
Ϫ3
Ϫ3
4
2
media.
ϩ
The potential shift with [H ] is not due to any change in junc-
the present reactions of ascorbic acid the rate increases almost
exponentially with increasing mol% of D O. The observ-
ation strongly suggests a preliminary protonation equilibrium
(Scheme 1) and refutes both a rate-determining electroprotic
reaction and a simple electron transfer reaction.
tion potential. We noted that the potential for the reversible
2
ferrocene/ferrocenium couple (Fig. 5B) and sq/q couple of 1
ϩ
(
Fig. 5A) did not change with change in [H ].
Scheme 1
I1Q and I2Q represent preliminary adducts formed by two
11a
forms of the complex. The pK for H Q is 8.0 and we assume
a
2
H Q to be the exclusive form of 1,4-dihydroxybenzene in the
2
ϩ
experimental range of [H ].
The scheme leads to
(
7)
Fig. 6 Plot of E1/2 (NHE) for the sq/cat couple versus pH.
J. Chem. Soc., Dalton Trans., 2002, 1220–1225
1223

View Article Online
Where T Q (≅ [H Q]) is the total concentration of hydroquinone
2
(
9)
and the stoichiometric factor is 2.
ϩ
Provided that, (1 ϩ K T ) ӷ {K [H ](1 ϩ K T )} and
1
Q
Q
H
2Q
Q
ϩ
that k K Ӷ k K K [H ], the expression reduces to:
1Q
1Q
2Q 2Q
H
Ϫ
where T A (≅ [H A] ϩ [HA ]) is the total concentration of the
ascorbic acid. Provided that, 1 ӷ {(K K K T ϩ [H ](K ϩ
2
ϩ
4A
aA
H
A
H
K K T )}, eqn. (9) reduces to eqn. (10):
2A
H
A
(
8)
ϩ
k = (k K K K ϩ k K K [H ])T
A
(10)
0
4A 4A aA
H
2A 2A
H
ϩ
The k values for different T at fixed [H ] fit satisfactorily to
eqn. (6) (see Fig. 8) and yields K1Q = (1026 ± 44) mol dm and
0
Q
Ϫ1
3
Again, according to eqn. (4)
k = k T
(11)
(12)
0
A
A
hence, from eqns. (10) and (11)
ϩ
k = (k K K K ϩ k K K [H ])
A
4A 4A aA
H
2A 2A
H
The second order rate constants, k , increased linearly (r =
.989) with [H ]; The slope = (311 ± 42) and intercept = (7.7 ±
A
ϩ
0
0
=
.9), values lead to. k K K = (311 ± 42) and k K K
2A
2A
H
4A 4A
H
4
6
Ϫ2 Ϫ1
(7.7 ± 0.9) × 10 dm mol
The results demonstrate overwhelming kinetic dominance of
H over 1, in the oxidation of ascorbic acid and quinol. This
s .
1
arises from the greater oxidising power of 1H as indicated by
ϩ
the higher E1/2 value of 1 at higher [H ].
11a
It has been shown that for a purely outer-sphere reaction,
Ϫ
the ratio of rate constants in the HA path to the H A path is
2
4
2
≈
10. In the present case the k K /k K ratio is only ≈2 × 10
4A
4A 2A 2A
which suggests that the reactions are not purely outer-sphere.
As 1H is substitutionally inert, we do not see any possibility for
the formation of an inner-sphere adduct. However, significant
electronic interaction is possible via a H-bond between 1H and
Ϫ3
Ϫ3
Fig. 8 ko versus T Q plot; [1] 0.10 mmol dm , [HClO ] 15 mmol dm
;
4
Ϫ3
T 25.0 ЊC; I 0.10 mol dm (NaClO ); 40% (v/v) MeOH–H O media.
4
2
Ϫ
the reducing agents, viz., H Q, H A and HA .
Ϫ2
6
Ϫ1
2
2
k K K = (20.5 ± 1.2) mol dm s . Also, as expected from
2
Q
2Q
H
In this context, it is noticed with interest that in spite of the
insignificant kinetic activity of 1, the proposed adduct, I1Q
ϩ
Ϫ3
eqn. (8), the k versus [H ] plot (T = 1.0 mmol dm ) is a
0
Q
,
straight line (r = 0.97) with negligible intercept. The assump-
tions leading to eqn. (8) thus appear justified. The slope of the
formed between 1 and H Q, is moderately stable. We anticipate
2
that I1Q is stabilised by interaction between the OH group of
Ϫ3
3
Ϫ1 Ϫ1
line [(9.23 ± 1.0) × 10 dm mol
s
] yields k K K = (18.7
II
2Q
2Q
H
H Q and the π-electron cloud on the ring of Ru -bound sq in
Ϫ2
6
Ϫ1
2
19
±
1.0) mol dm s
.
1
.
The π-electron density of the sq ring in 1H is greatly
The simplest explanation for the kinetics observed with
ascorbic acid is shown in Scheme 2.
depleted and the adduct should be weak.
Product formation
In the reaction described above, 1 is ultimately reduced to
ϩ
[
Ru(bpy) (Hcat)] . The complex 1 is a one electron oxidant but
2
H Q and ascorbic acid are two electron reductants. Therefore,
2
the immediate products of the first act of electron transfer
ϩ
ؒ
should be [Ru(bpy) (Hcat)] and the radical species HQ or
2
ؒ
HA . We proposed above that the proton in 1H attaches itself
to the π-electron cloud of the Ru -bound sq radical. Addition
II
of an electron to 1H should form the catecholate anion, when
Ϫ
the proton attached to the π-electron cloud shifts to the O site
ؒ
and thus forms complex 2. In the next step, the radical, HQ or
Scheme 2
ؒ
HA , reacts rapidly with another molecule of 1H according to
eqn. (13), for example:
Here, I2A and I4A represent preliminary adducts.
In 1H, most probably the π-electron cloud on sq in 1 provides
ؒ
ϩ
ϩ
electrostatic attraction to the proton. Such a mode of pro-
1H ϩ HQ
[Ru(bpy)
2
(catH)] ϩ Q ϩ H
(13)
Q ϩ QH ,
Ϫ1 Ϫ1
19
tonation is closely similar to that proposed for the second
ؒ
Ϫ
ؒ
Ϫ
ϩ
protonation of the free p-benzoquinone radical.
In Scheme 2, the reaction between 1 and HA has been
An analogous reaction, Q ϩ Q ϩ 2H
2
Ϫ
7
3
proceeds with a rate constant of 8 × 10 dm mol
s
at pH
19
excluded because of the nature of the dependence of rate on
7.4. The reaction (13) is thus expected to be very fast.
Rate constant kA decreases with increased percentage of
MeOH, in MeOH–H O reaction media (Table 3). It may be an
ϩ
ϩ
[
H ]. This path demands decreased rate with increased [H ].
ϩ
Ϫ
But, actually rate increased with [H ]. Again, HA is well
2
known to be kinetically much more active than H A; if the
reaction between 1 and HA were insignificant, then, that
effect of the expected decrease in KH with increased percentage
of the less-polar component in the solvent. The increased per-
2
Ϫ
20
between 1 and H A should also be insignificant. On this basis,
centage of MeOH probably also renders 1H a weaker oxidant.
The trend of change in λmax of the MLCT band with solvent
composition indicates this possibility.
2
we excluded also the reaction between 1 and H A from Scheme
2
2
. Scheme 2 leads to
1
224 J. Chem. Soc., Dalton Trans., 2002, 1220–1225

View Article Online
Table 3 Effect of solvent composition in the media on the second
References
a
order rate constant, k , for the reduction of 1 to 2H by ascorbic acid
A
1
J. Lavergne, C. Matthews and N. Ginet, Biochemistry, 1999, 38, 4542.
3
Ϫ1 Ϫ1
%(v/v) MeOH
k /dm mol
s
2 M. S. Graige, M. L. Paddock, J. M. Bruce, G. Fener and M. Y.
A
Okamura, J. Am. Chem. Soc., 1996, 118, 9005.
3
(a) S. D. Pell, R. B. Salmonsen, A. Abelleira and M. J. Clarke, Inorg.
Chem., 1984, 23, 385; (b) S. E. Jones, L. E. Leon and D. T. Sawyer,
Inorg. Chem., 1982, 21, 360.
4
5
6
7
8
0
0
0
0
9
8
6.5
5
3.9
4
5
S. Arzberger, J. Soper, O. P. Anderson, A. Cour and M. Wicholas,
Inorg. Chem., 1999, 38, 157.
J. H. Rodrignez, D. E. Wheeler and J. K. McCusker, J. Am. Chem.
Soc., 1998, 120, 12051.
4.7
a
Ϫ3
Ϫ3
[
complex] 0.10 mmol dm ; T A 0.10 mmol dm ; [HClO ] 0.01 mol
4
Ϫ3 Ϫ3
dm ; T 25.0 ЊC; I 0.10 mol dm (NaClO ); 40% (v/v) MeOH–H O
6 C. G. Pierpont, Inorg. Chem., 1997, 36, 19.
7 H. Masui, A. B. P. Lever and P. R. Auburn, Inorg. Chem., 1991, 30, 2402.
8
9
4
2
media; error in k , 3–6%.
A
H. Haga, E. S. Dadsworth and A. B. P. Lever, Inorg. Chem., 1986, 25,
4
47.
Microcal Origin 4.0, Microcal Software Inc., Northampton, USA,
995.
1
Conclusion
1
0 (a) K. J. Laidler, Chemical Kinetics, 3rd edn., Harper & Row, New
York, 1987, p. 32; (b) C. Capellos and B. H. J. Bielski, Kinetic
System, Wiley-interscience, New York, 1972.
1 (a) A. McAuley, L. Spencer and P. R. West, Can. J. Chem., 1985, 63,
1198; (b) D. Dixon, T. P. Dasgupta and N. P. Sadler, J. Chem. Soc.,
Dalton Trans., 1997, 1903.
Reduction of sq radical bound to Ru() involves (a) proton-
ation of sq in Ru()-sq to generate the thermodynamically
stronger, and the kinetically dominant, oxidant [Ru(bpy) -
1
2
2ϩ
(
Hsq)] , 1H, and (b) formation of a dead end adduct I1Q
between 1 and H Q. We could gather no evidence for simul-
12 E. Pellizzetti, J. Chem. Soc., Dalton Trans., 1980, 484.
13 (a) K. Wiberg, Chem. Rev., 1955, 55, 713; (b) R. P. Bell, Acid–base
catalysis, Oxford University Press, London, 1941, p. 145; (c) E. L.
Purlee, J. Am. Chem. Soc., 1959, 81, 263.
2
taneous transfer of a proton and an electron in the rate-
determining step though such an electroprotic mechanism has
been postulated for the reduction of quinone to semiquinone to
quinol in photosystem I.
1
4 R. A. Binstead and T. J. Meyer, J. Am. Chem. Soc., 1987, 109, 3287.
5 W. J. Albery, in Proton Transfer Reactions, eds. E. Caldin and
V. Gold, Wiley-Interscience, New York 1975, ch. 9.
6 W. J. Albery and M. H. Davies, J. Chem. Soc., Faraday Trans., 1972,
68, 167.
7 U. Roy Choudhury, S. Banerjee and R. Banerjee, J. Chem. Soc.,
Dalton Trans., 2000, 589.
1
1
1
Acknowledgements
This work has been carried out with financial assistance from
the Council of Scientific and Industrial Research (CSIR) and
the Department of Science and Technology, New Delhi. This
and the award of SRF to D. G. and A. D. S. by CSIR is acknow-
ledged. Authors (A. D. S. and A. D.) are thankful to Dr R. V.
Jasra and Dr P. K. Ghosh of CSMCRI for their keen interest in
the work.
18 M. V. Huynh, P. S. White and T. Meyer, Angew Chem., Int. Ed.,
000, 39, 4104.
2
1
2
9 (a) V. A. Roginsky, L. M. Pisarenko, W. Bors and C. Michel,
J. Chem. Soc., Perkin Trans. 2, 1999, 871; (b) S. Isaaca and
R. van Eldik, J. Chem. Soc., Perkin Trans. 2, 1997, 1465.
0 A. Frost and R. G. Pearson, Kinetics and Mechanism, John Wiley
& Sons, Inc., USA, 1961, 2nd edn., p. 135.
J. Chem. Soc., Dalton Trans., 2002, 1220–1225
1225