Kinetic study on the reversible hydride transfer between methylene blue and thionine

Article abstract of DOI:10.1039/b000046i

The kinetics of the hydride transfer reaction between leuco methylene blue and thionine was investigated spectrophotometrically by means of the stopped-flow technique. Leuco methylene blue and leuco thionine were produced by photoreduction of methylene bl

Full text of DOI:10.1039/b000046i

View Article Online / Journal Homepage / Table of Contents for this issue
Kinetic study on the reversible hydride transfer between methylene
blue and thionine
Yingjin Liu, Shunzo Yamamoto,* Yoshimichi Fujiyama and Yoshimi Sueishi
Department of Chemistry, Faculty of Science, Okayama University, 3-1-1 T sushima-naka,
Okayama, 700, Japan
Received 5th January 2000, Accepted 13th March 2000
The kinetics of the hydride transfer reaction between leuco methylene blue and thionine was investigated
spectrophotometrically by means of the stopped-Ñow technique. Leuco methylene blue and leuco thionine
were produced by photoreduction of methylene blue and thionine with triethylamine in ethanol in the drive
syringe of the stopped-Ñow apparatus and their solutions were used as such in the measurement of the reaction
rate. This reaction was found to be reversible, and the rate constants for forward and backward reactions were
obtained. The ratio of k /k is consistent with the value of the equilibrium constant (K) which was obtained
f
b
from the absorbance at 655 nm of methylene blue at equilibrium using the stoichiometric relationships and
conservation conditions. The mechanism of this reversible hydride transfer reaction was discussed.
Kreevoy et al. found that rate constants for hydride transfer
between NAD` analogues could be predicted by MarcusÏ
theory of atom transfer over a wide range of structure and
equilibrium constants.1h7 They pointed out that this result
indicates that the reaction involves no high-energy interme-
diates. For reactions in which the reactants and products have
the same charge type and are structurally similar, one-step
reactions of the type shown in eqn. (1) are predicted;
the detailed kinetics and mechanism for this reaction were not
shown. We have checked brieÑy whether the hydride transfer
occurs between some dyes and leuco dyes, and found that for
T`/MBH and MB`/TH systems, the spectral changes are
large and the reactions were able to be followed. This reaction
was found to be reversible, as shown by eqn. (2) below.
In the present study, we followed the reactions between T`
and MBH and MB` and TH by producing MBH and TH by
means of the photoreduction of MB` and T` with tri-
ethylamine in ethanol in a closed system and by mixing a
solution of MB` (T`) with a solution of TH (MBH). This
paper reports on the detailed kinetics and mechanisms of this
reversible reaction.
A ` ] A H ¢ A H ] A `
(1)
i
j
i
j
Thionine and methylene blue exist as monovalent cations in
the ordinary pH region and their redox reactions are
reversible. Since leuco methylene blue (MBH) and leuco thion-
ine (TH) are very unstable and easily oxidized by coexisting
oxygen in solution, few kinetic studies on the reactions of
these compounds with p-acceptors have been previously
reported. Recently, we have studied the kinetics of the reac-
tions of TH and MBH with some p-benzoquinones8 and ferric
ion (Fe3`)9 by means of a combination of the formation of
MBH and TH by photoreduction of MB` and T` with tri-
ethylamine and ascorbic acid and the stopped-Ñow technique.
By these procedures we can follow the fast reactions of MBH
and TH with di†erent oxidants.
Between some cationic dyes and neutral leuco dyes a
reversible one-step hydride transfer similar to that mentioned
above is expected to occur because in this reaction, reactants
and products have the same charge type. Olson observed
color changes by mixing methylene blue and leuco safranine
bluish solutions and described that this demonstration is
useful for the illustration of reversible redox reactions.10 But
Experimental
Materials
Thionine (T`Cl~) and new methylene blue (Basic Blue 24:
NMB`Cl~) were purchased from Tokyo Chemical Industry.
Methylene blue (MB`Cl~), crystal violet (CV`Cl~), malachite
green (MG`Cl~), leuco crystal violet (CVH) and leuco mala-
chite green (MGH) were purchased from Wako Pure Chemi-
cal Industries. These compounds were used as received. Leuco
thionine (TH), leuco methylene blue (MBH) and leuco new
methylene blue (NMBH) were produced by photoreduction of
these parent dyes as mentioned below. Triethylamine was
used as received. Ethanol was dried over a molecular sieve
3A-1/16 and distilled. Ethanol-d (99.5%) was purchased from
1
Aldrich and used as received.
(2)
DOI: 10.1039/b000046i
Phys. Chem. Chem. Phys., 2000, 2, 2367È2371
This journal is ( The Owner Societies 2000
2367

View Article Online
Kinetic measurements
MBH (or TH, NMBH) solution was prepared by irradiating
MB` (or T`, NMB`) and triethylamine solution in ethanol,
deaerated by bubbling with nitrogen gas, with visible light
supplied by a 650 W projector lamp. On irradiation, the solu-
tion changed from colored to colorless immediately. This indi-
cates that the leuco dyes were formed. Ethanolic solutions of
dyes and triethylamine and dyes as reaction partners for leuco
dyes were separately charged in drive syringes of a stopped-
Ñow apparatus (Otsuka Electronic stopped-Ñow spectropho-
tometer, RA-401). The two solutions were deaerated by
bubbling with nitrogen gas before they were charged. The
former solution was photoreduced by irradiation with visible
light and a solution of leuco dyes was produced. Mixing was
performed within 1 ms by means of nitrogen gas pressure.
After mixing, rapid-scan spectra were taken every 2 or 10 ms.
For MB`/TH and T`/MBH systems, after mixing, the
changes in the absorbance at 655 nm were monitored. The
runs were repeated several times. During the measurements,
the solution of TH (or MBH) was being irradiated contin-
uously to prevent re-oxidation of the leuco dyes by oxygen
remaining in the solutions. All kinetic measurements were per-
formed at 25.0 ^ 0.1 ¡C.
Fig. 2 Absorption spectra of MB` (ÈÈ) and T` (È È È) in ethanol.
between MB` and TH and between MBH and T`. Fig. 3
shows the rapid formation of T` and the simultaneous
decrease in MB` concentration. Fig. 3 also shows that the
reaction does not proceed to completion, and the reaction
attains equilibrium fairly rapidly. A clean isosbestic point indi-
cates the absence of signiÐcant concentrations of any interme-
diate species.
As is shown in Fig. 2, the absorption by T` at 655 nm
(which is the wavelength of the peak of MB`) is nearly negli-
gible. Time-dependent curves of the absorption at 655 nm
after averaging the repeated runs from both sides of the reac-
tion are provided as supplementary data.¤ The absorbance at
655 nm increased rapidly and attained constant values after
the mixing of an MBH solution with T` solutions with
several concentrations. Although the T` concentrations are in
excess of the MBH concentration, the constant values depend
on the T` concentration. On the other hand, the absorption
at 655 nm decreased rapidly and also attained almost con-
stant values after the mixing of the MB` solution with TH
solutions of several concentrations. These Ðndings indicate
that the reactions from both sides do not proceed to com-
pletion from either side, and these reactions are reversible.
Results
We have checked brieÑy whether hydride transfer occurs
between some dyes and leuco dyes.¤ It was found that the
reaction between NMBH and MB` is very fast and proceeds
almost to completion, and that a back reaction was not
observed. As mentioned above, however, it was found that for
the T`/MBH and MB`/TH systems, the spectral changes are
large and the forward and backward reactions were able to be
followed.
As mentioned above, we produced TH and MBH by photo-
reduction of thionine and methylene blue with triethylamine.
Fig. 1 shows the absorption spectra of MB` taken before and
after photoreduction [(1) and (2)] and the spectrum of an
equivolume mixture of MBH and p-chloranil (CA) solutions
(3). It was reported that some p-benzoquinones oxidized MBH
rapidly and quantitatively.8 Fig. 1 can be explained by the
formation of MBH by the photoreduction and re-production
of MB` by oxidation with CA.
Fig. 2 shows the absorption spectra of MB` and T` in
ethanol. Fig. 3 shows the rapid scan spectra for the reaction
Fig. 1 Absorption spectra of MB` taken (1) before and (2) after pho-
toreduction and (3) absorption spectrum of an equivolume mixture of
MBH and CA.
Fig. 3 Rapid-scan spectra for reactions between (a) MBH and T`
and (b) MB` and TH at 25 ¡C in ethanol. Cycle time, (a) 20 ms and
(b) 8 ms.
¤ Available as electronic supplementary information. See http://
www.rsc.org/suppdata/cp/b0/b000046i/.
2368
Phys. Chem. Chem. Phys., 2000, 2, 2367È2371

View Article Online
Table 1 Rate constants for forward and backward reactions between
MBH and T`
Kreevoy et al. determined the second-order rate constants
for reversible reactions with the aid of three equations.1 The
Ðrst one is conventional for the description of pseudo-Ðrst-
order reactions. This equation could only be used for reac-
tions which left no signiÐcant fraction of the limiting reagent
unconsumed in the end. Reactions which come to equilibrium
with signiÐcant fractions of both reactants still remaining,
were described by the second equation if the initial concentra-
tions of the reactants were equal, and by the third equation if
they were unequal (in this case the spectroscopically inactive
reactant was used in at least 10-fold excess). In the present
study, the concentrations of reactants were not equal and that
of one of the reactants was not in excess. Therefore, we can
not use the equations presented by Kreevoy et al.
105[MBH] /M
105[T`] /M
10~5 k /M~1 s~1
0
0
f
1.6
1.6
1.6
1.6
2.3
3.3
4.9
7.3
1.08
1.02
1.12
1.02
Average 1.06 ^ 0.05
105[MB`] /M
105[TH] /M
10~5 k /M~1 s~1
0
0
b
1.8
1.8
1.8
1.8
0.76
1.1
1.5
9.29
8.34
7.93
1.9
7.80
Average 8.3 ^ 0.7
For the reversible second-order reaction between MBH and
T`, the following relation is obtained:
k /k \ 0.13 ^ 0.02
f
b
x
ab(x ] x) [ (a ] b)x x
e
e
e
ln
\ k t (3)
f
2ab [ (a ] b)x
ab(x [ x)
obtained (Fig. 4), and are listed in Table 2. Both rate constants
e
e
(k and k ) can be calculated by the procedure shown in the
where a and b are the initial concentrations of MBH and T`
f
b
Appendix. These values are listed in Table 2. The results are
consistent with those obtained by the Ðrst method.
and x and x are the concentrations of MB` after the time t
e
and at equilibrium. This equation becomes identical to the
The advantages of the second method can be summarized
as follows: (i) we can obtain the values of k and k by a single
second equation of Kreevoy et al. if a \ b and to the third
equation if a @ b. If the value of the left-hand side of eqn. (3)
at time t is Y (t), a linear relationship between Y (t) and t is
expected. However, there are some dead-times in the measure-
ments by the stopped-Ñow method and the dead-times seem
to be di†erent in each run. Therefore, we rewrite eqn. (3) to the
following equation
f
b
run from the one side of the reaction. (ii) Since a \ x/x \ (A
e
t
[ A )/(A [ A ), we can apply this method by knowing the
0
=
0
change in the absorbance, even when we do not know the
concentrations of the species (we do not know the molar
absorption coefficient). The value of x can be calculated using
e
a, b and r. This method was proposed previously by Zeng et
*Y \ Y (t) [ Y (t ) \ k (t [ t )
(4)
al.11,12 for the thermokinetics of reversible reactions where the
rates of heat production by the reactions were measured. This
method can be applied when the linear response of the extent
of reaction can be observed to equilibrium. Since both
0
f
0
where t is the time when the measurement was begun. From
the slopes of the straight lines between *Y and (t [ t ) for
0
0
reactions between MBH and T` and for the reverse reaction
from MB` and TH, the values of k and k were obtained (the
f
b
linear relations between *Y and (t [ t ) are shown as supple-
0
mentary data¤). The values of k and k are listed in Table 1.
f
b
The use of eqn. (3) requires that concentrations are calculated
from absorbances, and this was done using the molar absorp-
tion coefficient (e \ 92 000 M~1 cm~1 for the absorption of
MB` at 655 nm in ethanol). In the case that the concentra-
tions are not known (molar absorption coefficients are
unknown or the absorption bands of reactants and products
are largely overlapped), the following equation is useful (the
derivation of this equation is given in the Appendix).
1
1
1
\ exp(k *t)
]
[exp(k *t) [ 1]
(5)
1
1
1 [ a
1 [ a
r
t`*t
t
Here, r \ k /k x and a \ x/x , (a and a
2 e t`*t
are the values of
1
e
t
a at the time t and at the time t ] *t). At a Ðxed time interval
(*t), a plot of 1/(1 [ a ) vs. 1/(1 [ a ) should be linear with
t`*t
t
a slope of exp(k *t) and an intercept of (1/r)[exp(k *t) [ 1].
1
1
The values of a were obtained by a \ (A [ A )/(A [ A ),
Fig. 4 Plots of 1/a
vs. 1/a for the reaction between MBH and T`
t
t
0
=
0
t`*t
where A , A and A are the absorbances at 655 nm at times
in ethanol. Initial concentrations; [MBH] \ 1.6 ] 10~5 M,
0
t
=
0
0, t and O. From the slope and intercept, k and r can be
[T`] \ 3.3 ] 10~5 M, *t \ 21 (L), 30 (…), 39 (K) and 48 ms (=).
0
1
Table 2 Values of parameters involved in eqn. (5) and rate constants for forward and backward reactions between MBH and T`
105[MBH] /M
105[T`] /M
*t/ms
k /s~1
r
105x /M
105k /M~1 s~1
10~5 k /M~1 s~1
10~5 k /M~1 s~1
0
0
1
e
2
f
b
1.6
1.6
1.6
1.6
1.6
1.6
1.6
2.3
3.3
3.3
3.3
3.3
4.9
7.3
27
21
30
39
48
18
12
12.6
14.6
14.5
14.6
14.7
18.2
20.8
[3.49
[3.22
[3.37
[3.19
[3.13
[3.29
[3.73
0.57
0.60
0.63
0.59
0.58
0.69
0.83
[6.33
[7.56
[6.83
[7.76
[8.10
[8.02
[6.72
1.35
1.12
1.20
1.10
1.08
1.08
1.08
7.62
8.68
8.03
8.86
9.17
9.09
7.80
8.5(0.6)
Average 1.14(0.10)
k /k \ 0.13 ^ 0.02
f
b
Phys. Chem. Chem. Phys., 2000, 2, 2367È2371
2369

View Article Online
Table 3 The values of absorbance at 655 nm at equilibrium and
equilibrium constant between MBH and T` (MB` and TH)
Since a hydride or a proton is transferred in a rate-
determining step in the above mechanism, a kinetic isotope
e†ect is expected. We tried to produce leuco methylene blue-d
105[MBH] /M
105[T`] /M
Abs= at 655 nm
K
0
0
(MBD) by using ethanol-d instead of ethanol as solvent and
1
1.6
1.6
1.6
1.6
2.3
3.3
4.9
7.3
0.47
0.55
0.66
0.77
0.13
0.12
0.14
0.14
measured the rate of the reaction between MBD and T`. The
values of k and K obtained for MBD were 0.71 ] 105 M~1
f
s~1 and 0.13. The ratio of kH/kD was about 1.55, while KH/KD
f
f
is nearly unity. The isotope e†ect observed for k is not high
but is not negligible. It is higher than would be expected for a
105[MB`] /M
105[TH] /M
Abs= at 655 nm
K
f
0
0
1.8
1.8
1.8
1.8
0.76
1.1
1.5
0.98
0.76
0.63
0.46
È
secondary or a solvent isotope e†ect. This suggests that the
hydride (or proton) transfer is the rate-determining step.
0.10
0.18
0.16
A high isotope e†ect will be expected for the near symmetri-
cal transition state such as in the present reaction. Smaller
isotope e†ects may show that the hydrogen atom which is
involved in the leuco dye comes partly from the OÈH (OÈD)
bond of ethanol, but partly from CÈH bonds. The possibility
of the latter is supported by the fact that one of the products
observed is acetaldehyde in the photoreduction of methylene
blue in ethanol.15 If the content of MBD is c in leuco methy-
lene blue formed in ethanol-d , kD obtained in ethanol-d is
1.9
Average \ 0.14 ^ 0.03
methods can be applied to the present case, the usefulness of
the second method was conÐrmed for determining the rate
constants for forward and backward reactions by a single run
followed spectrophotometrically.
Equilibrium was established starting with known concen-
trations of MB` and TH or T` and MBH. The concentration
of MB` at equilibrium was determined from the absorbance
of MB` at 655 nm. The concentrations of the other three
species could then be evaluated by using the stoichiometric
relationships and conservation conditions. The equilibrium
constant, K, was evaluated from its deÐning equation;
1
f
1
expressed by
kD \ ck ] (1 [ c)kH
(6)
f
D
f
where k is the value of k for pure MBD. Therefore, the ratio
D
f
of kD/kH is given as follows;
f
f
kD
k
kH
f
D
f
\ c
] (1 [ c)
(7)
kH
f
[TH] [MB`]
If c ^ 0.5 (it is assumed that only less than half of the hydro-
gen atom involved in leuco dye comes from the OÈD bond in
ethanol-d ), the isotope e†ect for k larger than 3.4 was
e
e
e
K \
(6)
[T`] [MBH]
e
The value of K is shown in Table 3.
1
f
f
obtained using the value obtained for kH/kD (1.55).
f
As mentioned above, a clean isosbestic point was obtained
and this shows the absence of signiÐcant concentrations of
any intermediate species. Although we can not distinguish the
direct hydride transfer mechanism from the stepwise electronÈ
protonÈelectron transfer mechanism, we tentatively propose
the following simple mechanism:
Discussion
As is shown in Tables 1 and 2 the values of k and k , evalu-
f
b
ated by the two methods, are in good agreement with each
other. Further, the value of K directly obtained (shown in
Table 3) is in good agreement with the ratios of k /k . These
f
b
consistent results show the validity of the evaluation of the
rate constants with the second method.
MBH ] T` ¢ (MBGÉ É ÉT`) (fast)
(MBHÉ É ÉT`) ¢ (MB`É É ÉTH) (slow)
(MB`É É ÉTH) ¢ (MB` ] TH (fast)
The magnitudes of k and k reÑect the di†erences in the
f
b
reactivities of MBH and TH and in those of T` and MB`.
From the data of the rate constants for oxidations of MBH
and TH by p-benzoquinones (BQÏs) and Fe3` presented in the
previous papers, the reactivity of MBH is lower than that of
TH (the ratio of rate constants for MBH and TH, k(MBH)/
k(TH) \ 0.29 (2-methoxy-p-BQ), 0.29 (2,5-dimethyl-p-BQ),
0.36 (2,6-dimethyl-p-BQ), 0.39 (2,5-di-tert-butyl-p-BQ) and
0.35 (Fe3`)). Similar values were obtained for di†erent oxi-
dants. This shows that the value of about 0.3 is a measure of
the di†erence in the reactivity of MBH and TH for oxidation.
On the other hand, the reactivity of T` and MB` as electron
or hydride acceptors depends on the redox potential E0(A/
A~). From the values of E0(A/A~) for T` and MB` ([0.11
for T` and [0.05 for MB`),13 it is clear that the reactivity of
MB` is higher than that of T`. Since the reactivities of TH
and MB` are higher than those of MBH and T`, respectively,
the value of k obtained was smaller than that of k , and an
Here, (MBHÉ É ÉT`) and (MB`É É ÉTH) are complexes involving
pÈp interaction via the aromatic rings. This mechanism is the
same as that proposed by Kreevoy et al. for hydride transfer
between NAD` analogues.1
Appendix
The bimolecular reversible reaction between MBH ] T` has
the rate law
d[MB`]
\ k [MBH][T`] [ k [MB`][TH]
(A1)
f
b
dt
where, k and k are the rate constants for the forward and
f
b
backward reactions. If the initial concentrations of MBH and
T` are a and b and those of MB` and TH are 0, and the
concentrations of MB` after time t is x, then the concentra-
tions of reactants and products at time t are given as follows;
f
b
equilibrium constant less than unity was obtained.
For the hydride transfer between NAD` analogues,
Kreevoy et al.1 postulated the direct hydride transfer mecha-
nism and Ohno et al.14 suggested the stepwise electronÈ
protonÈelectron transfer (or electronÈhydrogen transfer)
mechanism. For the nearly symmetrical reversible nature of
the present reaction, the potential energy diagram of the reac-
tion must be nearly symmetrical, and therefore in the stepwise
mechanism the rate-determining step must be the step of
proton-transfer located at the centre and the asymmetrical
electron-hydrogen transfer mechanism is eliminated.
[MBH] \ (a [ x), [T`] \ (b [ x),
[MB`] \ x and [TH] \ x.
The rate of production of MB` is then given by
dx
dt
\ k (a [ x)(b [ x) [ k x2
(A2)
f
b
2370
Phys. Chem. Chem. Phys., 2000, 2, 2367È2371

View Article Online
If x is the concentration of MB` at equilibrium, when the net
rate of the reaction is zero,
both rate constants (k and k ) can be calculated from eqn.
(A8) and (A9).
e
f
b
k (a [ x )(b [ x ) [ k x2 \ 0
(A3)
f
e
e
b e
From eqn. (A2) and (A3) we can prove that
dx
dt
\ k (x [ x) ] k (x [ x)2
(A4)
References
1
e
2 e
1
R. M. G. Roberts, D. Ostovic and M. M. Kreevoy, Faraday
Discuss. Chem. Soc., 1982, 74, 257.
where k \ k (a ] b [ 2x ) ] 2k x and k \ k [ k . Inte-
1
f
e
b e
2
f
b
grating eqn. (A4), we obtain
[k ] k (x [ x)]k
2
D. Ostovic, R. M. G. Roberts and M. M. Kreevoy, J. Am. Chem.
Soc., 1983, 105, 7629.
1
1
2
e
e
3
4
M. M. Kreevoy and I. H. Lee, J. Am. Chem. Soc., 1984, 106, 2550.
D. Ostovic, I. H. Lee, R. M. G. Roberts and M. M. Kreevoy, J.
Org. Chem., 1985, 50, 4206.
ln
\ k t
(A5)
1
(k ] k x )(x [ x)
2 e
e
Let r \ k /k x and a \ x/x , and considering the experimen-
5
6
7
8
9
M. M. Kreevoy and A. T. Kotcherar, J. Am. Chem. Soc., 1990,
112, 3579.
1
2 e
e
tal values a (at time t) and a
(at time t ] *t), we have
t
1
t`*t
Y. Kim, D. G. Truhlar and M. M. Kreevoy, J. Am. Chem. Soc.,
1991, 113, 7837.
1
1
\ exp(k *t)
]
[exp(k *t) [ 1] (A6)
I. H. Lee, E. H. Jeovrg and M. M. Kreevoy, J. Am. Chem. Soc.,
1997, 119, 2722.
1
1
1 [ a
1 [ a
r
t`*t
t
Using k , k and r, x , k and k are expressed as follows:
S. Yamamoto, Y. Fujiyama, M. Shiozaki, Y. Sueishi and N.
Nishimura, J. Phys. Org. Chem., 1995, 8, 805.
Y. J. Liu, S. Yamamoto and Y. Sueishi, J. Phys. Org. Chem., 1999,
12, 194.
1
2
e
f
b
ab r ] 2
a ] b r ] 1
x \
(A7)
(A8)
(A9)
e
10 E. S. Olson, J. Chem. Educ., 1977, 54, 366.
11 J. S. Liu, X. C. Zeng, A. M. Tian and Y. Deng, T hermochim. Acta,
1993, 36, 43.
k ] 2k x
a ] b
k [ (a ] b [ 2x )k
1
2 e
k \
f
12 Y. J. Liu, S. Yamamoto, S. Q. Cheng, Y. Q. Zhang and X. C.
Zeng, J. Sichuan Univ., 1999, 36, 287.
13 H.-J. Timpe and S. Neuenfeld, J. Chem. Soc., Faraday T rans.,
1992, 88, 2329.
1
e 2
k \
b
a ] b
14 A. Ohno, T. Shio, H. Yamamoto and S. Oka, J. Am. Chem. Soc.,
1981, 103, 2045.
From Table 2 and eqn. (A7), we can calculate the Ðnal reac-
tion extent x and the value of k using k \ k /rx . Finally,
15 Y. Usui, Bull. Chem. Soc. Jpn., 1965, 38, 206.
e
2
2
1
e
Phys. Chem. Chem. Phys., 2000, 2, 2367È2371
2371