Photoreduction of IrCl62- complex in alcohol solutions and its reaction with hydroxyalkyl radicals

Article abstract of DOI:10.1002/(SICI)1097-4601(1998)30:10<711::AID-KIN3>3.0.CO;2-Y

The photochemistry of IrCl₆^2- complex in simple alcohols have been studied using laser-flash photolysis. Single electron transfer from the solvent molecule to the light-excited complex has been shown to be the primary photochemical process, Quantum yields of the photoreduction of IrCl₆^2- complex and the rate constants of its reaction with hydroxyalkyl radicals were determined at 200-330 K. Deviations of the rate constants from Debye-Smoluchowski equation for diffusion-controlled reactions are discussed. 1998 John Wiley & Sons, Inc.

Full text of DOI:10.1002/(SICI)1097-4601(1998)30:10<711::AID-KIN3>3.0.CO;2-Y

2؊
Photoreduction of IrCl6
Complex in Alcohol
Solutions and Its Reaction
with Hydroxyalkyl Radicals
E. M. GLEBOV,1 V. F. PLYUSNIN,1 V. P. GRIVIN,1 YU. V. IVANOV,1 N. V. TKACHENKO,2
H. LEMMETYINEN2
1Institute of Chemical Kinetics and Combustion, Novosibirsk 630090, Russia
2Institute of Materials Chemistry, Tampere University of Technology, P.O. Box 589, 33101 Tampere, Finland
Received 21 July 1997; accepted 1 April 1998
2
ABSTRACT: The photochemistry of IrCl6 Ϫ complex in simple alcohols have been studied using
laser-flash photolysis. Single electron transfer from the solvent molecule to the light-excited
complex has been shown to be the primary photochemical process. Quantum yields of the
2
Ϫ
photoreduction of IrCl6 complex and the rate constants of its reaction with hydroxyalkyl
radicals were determined at 200–330 K. Deviations of the rate constants from Debye–Smo-
luchowski equation for diffusion-controlled reactions are discussed. ᭧ 1998 John Wiley & Sons,
Inc. Int J Chem Kinet: 30: 711–719, 1998
2
ϩ
3ϩ
INTRODUCTION
of radical reactions with Cuaq [3–6], Feaq [5],
3
ϩ
2Ϫ
Fe(CN)6 [7], and IrCl6 [8] complexes are mea-
sured by the method of pulse radiolysis. For halide
complexes, the reaction rate constants of alcohol rad-
Catalytic and photocatalytic reactions, involving com-
plex ions of transient metals, such as generation of
molecular hydrogen from water [1] and oxidation of
organic compounds [2], have been intensively studied
for the last years. To develop the quantitative models
of these processes, one should know the rate constants
of fast (close to diffusion-controlled) reactions of or-
ganic radicals with complex ions in various solvents
and in wide temperature range.
There is a lack of reference data on the reactions
between transient metal complexes and alcohol radi-
cals. As a rule, the data available is concerned to water
solutions at a room temperature. So, the rate constants
Ϫ
icals with CuCl2 [9] and FeCl4 [10] in alcohol solu-
tions were determined.
In this article, laser-flash photolysis was used to
determine the rate constants of IrCl6 Ϫ complex reduc-
2
tion by hydroxyalkyl radicals in alcohol solutions at
200–330 K. These reactions are important for study-
ing the photochemical and radiation-chemical pro-
cesses, involving iridium complexes. Particular inter-
est to halide iridium complexes is inspired by both
their application as oxidizers of organic compounds
[11] and the generation of hydrogen by radiation-
chemical decomposition of alcohols in aqueous solu-
tions (actually, colloid iridium is a catalyst of this pro-
cess [12,13]).
Recently, we have demonstrated [14,15] the for-
mation of hydroxyalkyl radicals upon the photolysis
of IrCl6 in methanol and ethanol due to single-
electron transfer from solvent molecule to the excited
complex
Correspondence to: E. M. Glebov
Contract grant Sponsor: Russian Foundation of Fundamental
Research
Contract grant number: grant 96-03-33495
Contract grant Sponsor: Competition Center at the St. Peters-
burg University
2
Ϫ
Contract grant number: grant 94-9.2-108
᭧ 1998 John Wiley & Sons, Inc.
CCC 0538-8066/98/100711-09

712
GLEBOV ET AL.
h␯
R
with cold nitrogen or air flow. The temperature was
automatically maintained with accuracy of 0.5 C.
2
2
IrCl6 Ϫ G
H (IrCl6 Ϫ)* 9
9
:
RH
3Ϫ
ϩ
2
Quantum yield of IrCl6 Ϫ photolysis was measured
ϩ Rи ϩ H
IrCl6
(1)
by two methods: in stationary and flash experiments.
At 295 K the quantum yield of stationary photolysis
in ethanol, 1- and 2-propanol, and 1-butanol was de-
termined from the known value for methanol (0.1
[14]). In flash-photolysis experiments the energy of
each laser pulse was measured. The quantum yield was
determined from the decrease in optical density of the
absorption band of the initial complex with its maxi-
mum at 495 nm.
where Rи
и
и
is the hydroxyalkyl radical ( CH2OH and
CH3 CHOH for methanol and ethanol solutions, re-
spectively). In deoxygenated solutions, the radicals re-
act with the initial complex to form IrCl6 Ϫ and a cor-
responding aldehyde (A) [8]
3
R ϩ IrCl6 Ϫ !: IrCl6 ϩ A ϩ Hϩ
и
(2)
2
3Ϫ
In the presence of oxygen Rи is rapidly trapped to
form the peroxide radical
RESULTS AND DISCUSSION
и
и
R ϩ O2 !: RO2
(3)
2
؊
Mechanism of IrCl6 Photolysis in Alcohols
at 295 K
и does not participate in a
Due to its inactivity RO2
2
Ϫ
subsequent transformations of iridium complexes.
This mechanism is typical of the photoreduction of
halide complexes of transient metals in polar organic
solvents [9,10,16]. The present article reports on the
quantum yields of IrCl6 photoreduction and shows
that the photolysis mechanism (1–3) is also charac-
teristic for other simple alcohols.
Optical absorption spectrum of IrCl6 complex con-
sists of the electron transfer bands in both visible and
UV regions, as well as of the d-d bands in the region
of 300–380 nm [20]. Figure 1 shows the temperature
dependence of the absorption spectrum of IrCl6 in
methanol solution. With the temperature decreasing,
the charge transfer bands in the region from 400 to
500 nm narrow down, and their extinction coefficients
increase. This effect was taken into account when
2Ϫ
2
Ϫ
EXPERIMENTAL
2
Solutions of the IrCl6 Ϫ complex were prepared using
Na IrCl 6H O (Aldrich) and optically pure alcohols
ؒ
2
6
2
(Merck). Oxygen was removed by bubbling the solu-
tions with nitrogen for 20 min. Tetraethylthiuramdi-
sulfide was prepared as described in [17]. Optical ab-
sorption spectra were recorded on UNICAM 8735/50
or Specord UV-Vis (Carl Zeiss) spectrophotometers.
2Ϫ
For stationary photolysis of IrCl6 an eximer XeCl
laser (308 nm) was used.
Laser-flash photolysis setup [18] consisted of an
eximer laser for excitation (wavelength 308 nm, pulse
duration 20 ns), a xenon arch lamp with a mono-
chromator as a source of probing light and a mono-
chromator with a photomultiplier in the measuring
channel. The photomultiplier signal was supplied to a
digital Tektronix 7912AD oscilloscope connected
with an IBM type computer. In some experiments the
laser-flash photolysis setup described in ref. [16] was
used. While measuring the temperature dependence of
the reaction rate constants, we have taken into account
a change in the concentration of the initial complex
due to the heat expansion of a solvent. The parameters
of volumetric alcohol expansion were taken from ref.
[19]. In these experiments the samples were placed
into a quartz cryostat with optical windows blown out
2Ϫ
Figure 1 Optical absorption spectrum of IrCl6 complex
in methanol at different temperatures. Curves 1–4 denote
T ϭ 300, 225, 175, and 77 K, respectively.

2
Ϫ
IrCl6 COMPLEX IN ALCOHOL SOLUTIONS
713
Table I Quantum Yields (␾) at 295 K, Preexponential
Factors (␾0), and Activation Energies (Ea(␾)) of the
2
Quantum Yield of IrCl6 Ϫ Photoreduction
Ea(␾), kJ/
mole
0
Alcohol
␾
␾
methanol
ethanol
1-propanol
2-propanol
1-butanol
0.10a
0.13
0.07
0.08
0.06
0.43 Ϯ 0.05
1.00 Ϯ 0.20
3.4 Ϯ 0.3
5.3 Ϯ 0.5
0.53 Ϯ 0.07
0.42 Ϯ 0.05
4.4 Ϯ 0.3
4.7 Ϯ 0.3
a Measured in [14]; the measurement accuracy of the quantum
yield is 10%.
2
Ϫ
for IrCl6 in 1-propanol, 2-propanol, and 1-butanol
have shown the primary photochemical act to be sim-
ilar to that observed for methanol and ethanol. In the
experiments with the solutions containing oxygen, an
instant decrease in the optical density of the absorption
bands of the initial complex is observed (Fig. 2(a),
curve 1). The spectrum of this bleaching coincides
2
with that of IrCl6 Ϫ. The values of quantum yields at
2
Ϫ
Figure 2 Laser-flash photolysis of IrCl6 in alcohols at
room temperature are shown in Table I.
295 K. (a) Kinetic curves in 2-propanol (recording at
For deoxygenated solutions, after a laser pulse, we
490 nm; [IrCl6 ] ϭ 3.5 ϫ 10 4 M; and 1 cm cuvettee):
curve 1, natural content of oxygen, and curve 2, without
oxygen. (b) Dependence of the observed rate constant of
2
Ϫ
Ϫ
2
Ϫ
observe not only the instant bleaching in IrCl6 ab-
sorption bands but also the kinetics of additional de-
crease in the optical density. The value of the final
bleaching is by a factor of two larger than that of the
initial one (just after the laser pulse). The reaction ki-
netics is exponential (Fig. 2(a), curve 2). This process,
however, is of a pseudo-first-order because the effec-
tive rate constant depends linearly on IrCl6 Ϫ concen-
tration (Fig. 2(b)). Thus, in deoxygenated solutions the
hydroxyalkyl radicals react with IrCl6 Ϫ complex. Ta-
ble II lists the values of the rate constant for this re-
action at room temperature.
In analyzing the kinetic data we have neglected rad-
ical-radical reactions (dimerization and disproportion-
ation [22]) which occur with rate constants of 1.4 ϫ
2
Ϫ
IrCl6 decay in 1-propanol (curve 1), 2-propanol (curve 2),
and 1-butanol (curve 3) on the concentration of the complex.
measuring the temperature dependence of the quantum
yield. Charge transfer bands in the visible region of
IrCl6 spectrum display no photoactivity in both
aqueous [21] and methanol [14] solutions. The irra-
diation of methanol and ethanol solutions of IrCl6 Ϫ in
the region of charge transfer bands overlap with the d-
d band (308 nm) results in the photoreduction with a
subsequent formation of IrCl6 Ϫ complex [14].
2
2
Ϫ
2
2
3
Stationary and laser-flash photolysis experiments
0
2Ϫ
Table II Rate Constants at 295 K (k2), Preexponential Factors (k2 ), and Activation Energies (Ea(k2)) of IrCl6
Reaction with Hydroxyalkyl Radicals, Diffusion Rate Constants (kdiff) at 295 K and Activation Energies (Ediff) of
Diffusion Motion
Ϫ
9
0
Ϫ
11
Ϫ
9,b
c
k2 ϫ 10
M
,
k2 ϫ 10
,
Ea(k2),
kJ/mole
kdiff ϫ 10
M
Ediff
Ϫ
Ϫ1
Ϫ
Ϫ1
Ϫ
Ϫ1
Alcohol
Radical
1s
M
1s
1s
kJ/mole
3.2 Ϯ 0.1a
2.3 Ϯ 0.1a
1.4 Ϯ 0.1
2.0 Ϯ 0.1
1.3 Ϯ 0.1
0.46 Ϯ 0.11
0.65 Ϯ 0.22
6.3 Ϯ 0.5
8.2 Ϯ 0.6
12.1
6.8
3.4
3.8
2.5
12.9
15.1
19.9
23.8
21.7
и
methanol
ethanol
1-propanol
2-propanol
1-butanol
CH2OH
иCHOH
CH3
иCHOH
иCOH
CH3CH2
(CH3)2
CH3(CH2)2
1.2 Ϯ 0.5
0.22 Ϯ 0.05
9.6 Ϯ 0.7
7.2 Ϯ 0.4
иCHOH
a Measured in [14].
b Calculated according to eq. (4).
c Determined from the temperature dependence of kdiff using temperature-dependent viscosities [19].

714
GLEBOV ET AL.
1
1
8
Ϫ1 Ϫ1
109 MϪ sϪ
CH2OH [23] and 3.0 ϫ 10 M s
fusion rate constants at 295 K and the activation en-
ergy of diffusion in alcohols, calculated using the
Smoluchowski–Debye eq. (4) and the viscosity/tem-
perature ratio [19]
for и
и
for CH3 CHOH radicals [24]. Similar values are typ-
ical of the radicals of the other alcohols. When radical
concentration is about 10Ϫ5 M (the maximum value in
и
our experiments), the characteristic time of CH2OH
recombination amounts to 35 ␮s which is by an order
kdiff ϭ 8 RT/3000␩
(4)
of magnitude higher than the time of radical decay in
2
the reaction with IrCl6 Ϫ. The doubling of the initial
The values of the rate constants of the reactions
2
Ϫ
Ϫ2
bleaching in the absorption band of the IrCl6 com-
plex also testifies to a negligible contribution of the
recombination reaction.
between radicals and IrCl6 at room temperature are
2–4 fold smaller than calculated value of the diffu-
sion-controlled rate constant.
When the reaction is limited by diffusion, a de-
crease in the experimental rate constants, compared to
the diffusion rate constants, could be assigned to the
Temperature Dependence of Quantum
2؊
Yields of IrCl6 Photolysis and Rate
Constants of its Reaction with Hydroxyalkyl
Radicals
Ϫ
existence of a spin-statistic factor. The IrCl6 2 ion is a
low-spin complex with electron configuration 5d5 and
electron spin 1/2. The radical also has spin 1/2. There-
fore, the total spin of the system may be either 1 or 0.
Laser-flash photolysis measurements performed
within a wide temperature range (200–320 K) al-
lowed one to determine the temperature dependence
3
The reaction gives rise to the IrCl6 Ϫ complex with 5d6
electron configuration and zero electron spin. The rad-
ical transforms into the molecule with zero spin, i.e.,
the total system spin may be only zero. Thus, the spin-
statistic factor f is 1/4. In this case, the reaction rate
constant is kdiff ϫ f and the activation energy must ap-
proach the activation energy of diffusion motion. Ta-
ble II shows that the activation energy of the reaction
2Ϫ
of the initial quantum yield of IrCl6 photoreduction
in various alcohols and of the rate constants of its re-
action with hydroxyalkyl radicals (Fig. 3). Activation
energy values and preexponential factors for the quan-
tum yield are shown in Table I and those for the rate
constants are given in Table II.
2
Ϫ
Two latter columns in Table II summarize the dif-
between radicals and IrCl6 is from 2 to 3 times
smaller than that of diffusion. So, the experimental
data can not be explained by the value of spin-statistic
factor.
Figure 4 demonstrates the dependencies of loga-
rithms kobs ϭ k2 and kdiff on the inverse temperature
2
for methanol and 1-butanol solutions of IrCl6 Ϫ. It is
seen that for butanol at low temperature (200 K) kobs
exceeds kdiff , calculated using the Smoluchowski–De-
bye eq. (4) by more than an order of magnitude. The
same holds for the isopropanol solutions.
Superdiffusion values of kobs in isopropanol and 1-
butanol may arise due to the peculiarities of both the
diffusion motion of reagents and the chemical reaction
itself. First, the question arises of the applicability of
the Stokes–Einstein eq. (5) [25] for the diffusion co-
efficient at low temperatures
D ϭ kT/␣␲ r␩
(5)
where r is the hydrodynamic radius of a diffusing mol-
ecule, ␣ ϭ 6 for the most frequent case of stick bound-
ary conditions.
Equation (5) refers to the case of large spherical
particles which diffuse in homogeneous solvent. In
practice, however, this equation is often used to de-
scribe the diffusion of molecules of arbitrary size and
shape. Direct measurements of the self-diffusion co-
Figure 3 Arrhenius plot of the rate constant k2 for the re-
2
Ϫ
и
action of the CH2OH radical with IrCl6 (a) and the quan-
2
Ϫ
tum yield of IrCl6 photoreduction in methanol (b).

2
Ϫ
IrCl6 COMPLEX IN ALCOHOL SOLUTIONS
715
tunneling from the radical to the complex observed
when decreasing temperature. In particular, as the dif-
fusion rate decreases, the effective reaction radius may
increase. This may be verified by measuring the rate
constant of another diffusion-controlled reaction
which has no relation to the electron transfer. For this
purpose, we have used the reaction of the dithiocar-
bamate radicals recombination.
Determination of Diffusion Rate Constant
Using Reaction of Dithiocarbamate
Radicals Recombination
и
ϭ dtcи) arise
Dithiocarbamate radicals (R2NCS2
upon flash photolysis of the solutions of thiuramdi-
sulfide (R2NC(S)S9S(S)CNR2 ϭ tds, in our case
R ϭ C2H5)
Figure 4 Temperature dependencies of the reaction rate
constants in methanol (a) and 1-butanol (b). (1) Rate con-
stant of the diffusion-controlled reaction calculated accord-
ing to Smoluchowsky–Debye eq. (4); (2) Rate constant of
h␯
J 2 dtcи
tds I
R
(6)
krec
2
Ϫ
the reaction between IrCl6 and hydroxyalkyl radical; (3)
Rate constant of the recombination of dithiocarbamate rad-
icals (krec); and (4) (Dot line) Kinetic rate constant of the
These radicals display the optical absorption band with
a maximum at 600 nm (⑀ ϭ 3100 MϪ cmϪ1 [17]) and
in majority of solvents they recombine with a diffusion
rate constant [17].
1
2Ϫ
reaction between IrCl6 and hydroxyalkyl radical calculated
according to eq. 7 with kdiff ϭ krec
.
Figure 4 shows the dependence of the logarithm of
и
the recombination rate constant (krec) of dtc radicals
in methanol and butanol on the inverse temperature.
efficients of alcohol molecules were performed only
at temperatures above 0ЊC [26], where they depend
linearly on the T/␩ ratio. In this case, if the
(3V0/4␲)1/3 value is considered to represent the molec-
ular radius r (V0 is the molecular volume calculated
using the available empirical rules of Edward [27]),
the value of coefficient ␣ in (5) tends to 4. In detail,
␣ is equal to 4.7, 4.2, 4.3, and 4.0 for methanol, eth-
anol, isopropanol, and butanol, respectively. The value
of ␣ ϭ 4 corresponds to the case of slip boundary con-
ditions. This is preferable for the diffusion of particles,
comparable in size with the solvent molecules [25].
However, it is insufficient just to change the ␣ value
in eq. (5) from 6 to 4 in order to explain the experi-
At room temperature in methanol krec ϭ (7.9 Ϯ
0.7) ϫ 109 MϪ sϪ . This value actually coincides with
1
1
the rate constant of the redox quenching of the excited
[Ru(bpy)3 ϩ] complexes by N,N,NЈ,NЈ-tetramethyl-
2
1
1
paraphenylendimanine
kquen ϭ 6.9 ϫ 109 MϪ sϪ
taken in ref. [28] as the reference diffusion-controlled
reaction (a large negative value of Gibbs’ free energy
⌬G0 ϭ Ϫ58.8 kJ/mole and the equality of spin-statis-
tic factor to unity have stimulated the choice). This
holds that for the recombination reaction (6) the spin-
statistic factor is close to unity (due to hyperfine in-
teraction in radicals and due to spin-orbital interaction
in the complex it may increase from 1/4 to 1 [29]).
The value of krec in methanol at room temperature
is smaller than the diffusion limit (Table II). However,
when the temperature is lower than 260 K, krec begins
to exceed kdiff . In butanol and isopropanol over the
entire temperature range (190–320 K), krec exceeds
the diffusion rate constant, calculated using the Smo-
luchowski–Debye eq. (4). The preexponential factors
and the activation energies for krec are listed in Table
III. Note that at room temperature krec is almost the
same for all alcohols although the viscosity changes
substantially (for butanol it is by a factor of five higher
than for methanol).
2
mental data on the reaction of radicals with IrCl6 Ϫ in
isopropanol and butanol.
The realization of anomalous kobs values may be
related to the fact that at low temperature the appli-
cability of the Stokes–Einstein equation is violated
and the microviscosity may be substantially lower than
the macroviscosity. However, the absence of direct
measurements of the diffusion coefficients at t Ͻ
0ЊC does not allow one to estimate the reliability of
this hypothesis.
Another reason for the superdiffusion values of
kobs may be connected with the peculiarities of electron

716
GLEBOV ET AL.
Table III The Rate Constants at 295 K (krec), Preexponential Factor (krec0), and Activation Energy (Ea(krec)) of the
Reaction of Dithiocarbamate Radical Recombination and the Exponent n in the Experimental Dependence krec ca. T/
n
␩
Ϫ
9
0
Ϫ11
krec ϫ 10
M
krec ϫ 10
Ϫ
Ϫ
1
Ϫ
Ϫ1
Alcohol
1s
M
1s
Ea(krec), kJ/mole
n
methanol
ethanol
2-propanol
1-butanol
7.9 Ϯ 0.7
3.3 Ϯ 0.6
6.2 Ϯ 0.6
7.8 Ϯ 0.8
8.8 Ϯ 0.9
1.1 Ϯ 0.3
11 Ϯ 5
6.0 Ϯ 0.2
8.6 Ϯ 0.4
12.5 Ϯ 0.9
10.0 Ϯ 0.6
0.37 Ϯ 0.02
0.50 Ϯ 0.03
0.48 Ϯ 0.04
0.41 Ϯ 0.03
4.6 Ϯ 1.5
Moreover, in all studied alcohols the recombination
constant of dtcи radicals over the entire temperature
range does not obey the Smoluchowski–Debye de-
pendence krec ca. (T/␩). The slope of straight lines in
linear equations ln(krec) Ϫ ln(␩/T) (Fig. 5) for various
alcohols is within the range (Ϫ0.38)–(Ϫ0.50) (the
last column in Table III) and within the experimental
accuracy it differs from the theoretical value of Ϫ1.
Known are the cases [30–38] where the Smolu-
chowski–Debye equation fails for the rate constants
of diffusion-controlled reactions and, accordingly, the
Stokes–Einstein equation fails for the diffusion co-
efficient. As a rule, this holds for the reactions, occur-
ring in highly viscous liquids (␩ ca. 1–1000 P). The
fusion coefficient [37,38] usually attribute to the influ-
ence of a microheterogeneous structure of viscous liq-
uid, leading to non-Fick’s diffusion [38]. As it has
been mentioned, these effects are stronger for binary
mixtures and for solute molecules diffusion compared
to the self-diffusion of solvent molecules [34].
In our experiments the deviations from the Smo-
luchowski–Debye equation are observed for the liq-
uids with low viscosity (alcohols, ␩ ca. 0.2–10 cP).
As a rule, this equation is reported to hold for the
diffusion-controlled reactions in low-viscous nonpolar
solvents [39,40]. The anomalous character of diffusion
constants is typical of the solvents with hydrogen
bonds, such as alcohols [36,41–43].
Direct measurements of diffusion coefficients in
usual (low-viscous) solvents also exemplify the anom-
alous dependencies on viscosity [44]. As in the case
of strongly viscous liquids, the anomalies arise upon
diffusion of small admixtures [45,46]. The diffusion
of various admixtures in 2-methylpentane-2,4-diol has
been studied in [47]. It has been shown that the planar
molecules (benzene, naphthalene, phenantrene, and
pyrene) diffuse much faster than a rather spherical-
shape CBr4 molecule. In this case, the parameter ␣ in
the Stokes-Einstein eq. (5) is temperature-dependent
and its value for the benzene molecule is smaller than
unity already at 295 K. Note that the dtcи radical is
also planar and its diffusion may display similar pe-
culiarities.
Ϫ1/2
observed experimental k ca. ␩
dependence [30–
Ϫ1/2
33], as well as the D ca. ␩
dependence of the dif-
There are also cases of a “normal” behavior of the
diffusion-controlled rate constants in alcohols. Thus,
Gilles and Boyd [48] have established that the recom-
и
bination of (CH3)2 COH radicals, arising upon pulse
radiolysis of isopropanol within the temperature range
Ϫ
1 dependence.
147–300 K,
krec ca. T ϫ ␩
of
obey the
Probably, for the radicals with the structure resembling
that of the solvent molecules, the anomalies of diffu-
sion motion are of minor importance.
Note that the articles on diffusion, mentioned in this
section, are purely empirical and have no theoretical
concepts in common [44]. Thus, the question of the
Figure 5 Dependence of ln(krec/T) on ln(␩) for the recom-
bination of dithiocarbamate radicals in methanol.

2
Ϫ
IrCl6 COMPLEX IN ALCOHOL SOLUTIONS
717
reasons of superdiffusion values of the reaction rate
constants, including dithiocarbamate radicals recom-
bination in alcohols needs further studying. However,
these results indicate that the increase of the reaction
radii at low temperature in the reactions of the electron
observed to be low and almost the same for all alco-
hols within the experimental accuracy. The mono-
molecular rate constant of electron transfer in the
2Ϫ
и
IrCl6 . . . R encounter pair may be estimated us-
ing the relationship [50]
2
transfer from radicals to the IrCl6 Ϫ complex is hardly
probable. This is evident from the fact that the rate
constants, exceeding calculated by eq. (4), are also ob-
served for the recombination of radicals, which is a
reaction of another type.
kkin ϭ kmono ϫ V
(8)
where V ϭ 4␲R2⌬ is the volume of the reaction zone;
R is the total radius of partners; and ⌬ is the thickness
of the reaction zone (a distance over which an electron
is transferred). The total radius for both the complex
Determination of Kinetic Rate Constants of
2؊
and radical is R Ϸ 5.6 A (RIr Ϸ 4.3 A [51], Rrad
Ϸ
؉ Rи Reaction
IrCl6
1.4 A). Taking ⌬ ca. 1 A, one may calculate V Ϸ
The contradiction caused by the superdiffusion rate
constant of reaction (2) vanishes if the reaction of di-
thiocarbamate radicals recombination is assumed to be
diffusion-controlled (over the entire temperature range
k2 being smaller than krec). For methanol (Fig. 4(a)) the
activation energies of these two reactions almost co-
incide (Tables II and III) and the values of their rate
constants differ by a factor of two.
1
0.25 MϪ1 and kmono Ϸ 2 ϫ 1010 sϪ . Thus, the electron
tunneling from radical to complex at room temperature
may occur during about 50 ps. For ⌬ ca. 0.1 A this
time decreases to 5 ps.
Comparison of Rate Constants of the
Reactions of Hydroxyalkyl Radicals with
IrCl6 and Other Coordination Compounds
2Ϫ
For the reaction of IrCl6 with isopropanol and
butanol radicals the activation energies are also close
2؊
и
to that of dtc radicals recombination (smaller by ca.
Table V gives the literature data on some known re-
actions of transient metal complexes of with hydroxy-
alkyl radicals at 295 K. There are two points to be
discussed.
3kJ/mole) and at room temperature the rate constant
k2 is smaller than krec by a factor of three and six,
respectively. A parallel decrease in the rate constant
and in the reaction activation energies compared to
krec, assumed as kdiff (kdiff ϭ krec), may be explained by
the fact that the rate of electron transfer in the en-
Firstly, the rate constants of the reactions between
2
Ϫ
и
IrCl6
and hydroxyalkyl radicals ( CH2OH,
и
и
CH3 CHOH, and (CH3)2 COH) in water solutions at
295 K, measured in [8] are twice as obtained in alco-
hol solutions (see Table II). This difference is not
rather comprehensible. It should be noted that he work
[8] doesn’t contain primary kinetic information. So,
one is inclined to think that the values of rate constants
measured in [8] may become over-estimated because
of the influence of competing channels of radicals de-
cay. These channels may be the recombination of rad-
icals or its reaction with the impurities of oxygen.
Secondly, the rate constants of the reaction between
2
Ϫ
. . . Rи, i.e., the rate constant of
counter pair IrCl6
the kinetic stage, kkin , is smaller than the rate constant
of diffusion motion, kdiff . In this case, the observed
rate constant kobs can be written as [49]
kkinkdiff
kobs
ϭ
(7)
kkin ϩ kdiff
Figure 4 shows the temperature dependence of kkin
,
determined using eq. (7). Table IV summarizes the
activation energies and preexponential factors for this
constants. The activation energy of the kinetic stage is
2
Ϫ
radicals and IrCl6 exceed substantially the similar
values for chloride and aquated complexes of Fe(III)
2
Table IV Parameters of Rate Constant kkin for R ϩ IrCl6 Ϫ Reaction: Rate Constants at 295 K (kkin), Preexponential
и
Factor (kkin0) and Activation Energy (Ea(kkin))
Ϫ
Ϫ
Ϫ1
0
Ϫ
Ϫ
Ϫ1
kkin ϫ 10 9, M 1s
kkin ϫ 10 10, M 1s
Ea (kkin), kJ/mole
Radical Rи
и
CH2OH
6.6
7.7
3.4
1.4
9.9 Ϯ 6.5
12.9 Ϯ 12
6.4 Ϯ 4.3
2.0 Ϯ 0.5
6.6 Ϯ 0.9
6.9 Ϯ 2.3
7.2 Ϯ 2.2
6.6 Ϯ 1.0
иCHOH
CH3
иCOH
(CH3)2
иCHOH
CH3CH2CH2

718
GLEBOV ET AL.
Table V Rate Constants of the Reaction between Complexes and Hydroxyalkyl Radicals
Ϫ
Ϫ1
Complex
Radical
Solvent
k2, M 1s
Refs.
иCOH
CuCl2
(CH3)2
2-propanol
Ethanol
H2O
(2–3) ϫ 108
(6 Ϯ 1) ϫ 108
6.0 ϫ 109
[9]
[10]
[8]
[8]
[8]
[8]
[8]
[7]
[7]
[7]
[7]
[5]
[3–6]
[5,6]
[3,5,6]
[5]
Ϫ
иCHOH
FeCl4
CH3
2Ϫ
2Ϫ
2Ϫ
2Ϫ
2Ϫ
и
IrCl6
IrCl6
IrCl6
IrCl6
IrCl6
CH2OH
CH3
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
H2O
4.5 ϫ 109
иCHOH
ca. 2 ϫ 109
4.7 ϫ 109
и
CH2CH2OH
иCOH
(CH3)2
1.2 ϫ 109
и
CH2C(CH3)2OH
3
3
3
3
ϩ
ϩ
ϩ
ϩ
4 ϫ 109
и
Fe(CN)6
Fe(CN)6
Fe(CN)6
Fe(CN)6
CH2OH
иCHOH
CH3
5.3 ϫ 109
иCHOH
CH3CH2
(CH3)2
CH3
3.7 ϫ 109
4.7 ϫ 109
иCOH
3
ϩ
иCHOH
Feaq
Cuaq
Cuaq
Cuaq
Cuaq
3.8 ϫ 108
2
2
2
2
ϩ
ϩ
ϩ
ϩ
(1.1–1.9) ϫ 108
(9 Ϯ 2) ϫ 107
(4.5–5.2) ϫ 107
3.2 ϫ 106
и
CH2OH
иCHOH
CH3
иCOH
(CH3)2
и
CH2(CH3)COH
H2O
The authors wish to thank Dr. V.V. Korolev and Dr. A.D.
Milov for helpful discussions.
and Cu(II) which may be due to the different reaction
mechanisms. The research into pulse radiolysis of wa-
ter-alcohol solutions, containing Cuaq complex, re-
2ϩ
port the appearance of intermediate absorption, related
ϩ
to the formation of the Cu2
. . . Rиradical complex
BIBLIOGRAPHY
[5], i.e., the reaction occurs via the inner-sphere mech-
anism. The rate constants for IrCl6 are close to the
typical values for the Fe(CN)6 complex. Both com-
plexes display stable coordination sphere and the re-
action with radicals occurs via the mechanism of out-
ersphere electron transfer.
2
Ϫ
1. I. A. Potapov, M. V. Rozenkievich, and Yu A Sakhar-
ovskii, Koord. Khim., 7, 229 (1981).
2. A. I. Kryukov and S. Ya. Kuchmii, Fotokhimiya Kom-
pleksov Perekhodnykh Metallov (Photochemistry of
Transition Metal Complexes), Naukova Dumka, Kiev,
1989, p. 240.
3
ϩ
3. H. Cohen and D. Meyerstein, J. Am. Chem. Soc., 94,
6944 (1972).
4. G. Ferraundi, Inorg. Chem., 17, 2506 (1978).
5. G. V. Buxton and J. C. Green, J. Chem. Soc. Faraday
Trans. I, 74, 697 (1978).
6. M. Freiberg, W. A. Mulac, K. H. Schmidt, and D. Mey-
erstein, J. Chem. Soc. Faraday Trans. I, 76, 1838
(1980).
7. G. E. Adams and R. L. Wilson, Trans. Faraday Soc.,
65, 2981 (1969).
CONCLUSION
2
Ϫ
Thus, for five simple alcohols the IrCl6 photore-
duction is shown to be due to electron transfer from a
solvent molecule to the excited complex. The quantum
yield and its temperature characteristics are measured.
The hydroxyalkyl radicals, arising from photoreduc-
2
Ϫ
tion, disappear in the reaction with the initial IrCl6
8. S. Steenken and P. Neta, J. Amer. Chem. Soc., 104, 1244
(1982).
9. N. P. Gritsan, V. F. Plyusnin, and N. M. Bazhin, Teor.
Eksper. Khim., 21, 39 (1986).
10. I. V. Khmelinski, K. A. Amosov, V. F. Plyusnin, and
V. P. Grivin, Koord. Khim., 16, 1373 (1990).
11. K. K. Sen Gupta and U. Chatterjee, J. Inorg. Nucl.
Chem., 43, 2491 (1981).
12. G. Mills and A. Henglein, Radiat. Phys. Chem., 26, 385
(1985).
complex. Rate constants and activation energies of
these reactions are slightly smaller than those for the
reaction of dithiocarbamate radical recombination
taken as the diffusion-controlled reaction. This makes
it possible to determine the parameters of electron
transfer stage in a contact radical-complex pair. Ac-
cording to estimations, the electron transfer occurs
during the time which is shorter than 100 ps.
13. G. Mills and A. Henglein, Radiat. Phys. Chem., 26, 391
(1985).
The work was supported by Russian Foundation of Funda-
mental Research (grant 96-03-33495) and the Competition
Center at the St. -Petersburg University (grant 94-9.2-108).
14. E. M. Glebov, V. F. Plyusnin, N. I. Sorokin, V. P. Gri-
vin, A. B. Venediktov, and H. Lemmetyinen, J. Pho-
tochem. Photobiol. A:Chem., 90, 31 (1995).

2
Ϫ
IrCl6 COMPLEX IN ALCOHOL SOLUTIONS
719
15. E. M. Glebov, V. F. Plyusnin, V. L. Vyazovkin, and
A. B. Venediktov, J. Photochem. Photobiol. A:Chem.,
107, 93 (1997).
16. V. F. Grivin, I. V. Khmelinski V. F. Plyusnin, I. I. Bli-
nov, and K. P. Balashev, J. Photochem. Photobiol.
A:Chem., 51, 167 (1990).
31. T. Sawai and W. Hamill, J. Chem. Phys., 56, 5524
(1972).
32. J. Fuller, N. Petelski, D. Ruppel, and M. Tomlinson, J.
Phys. Chem., 74, 3066 (1970).
33. J. Mayer, M. Szadkowski-Nicze, and J. Kroh, Radiat.
Phys. Chem., 32, 519 (1988).
34. N. Kato, T. Miyazaki, K. Fueki. S. Miyata, and Yu.
Kawai, J. Phys. Chem., 88, 1445 (1984).
35. L. L. Koroli, V. A. Kuzmin, and I. V. Khudyakov, Int.
J. Chem. Kinet., 16, 379 (1984).
36. B. Cercek, Int. J. Radiat. Phys. Chem., 7, 223 (1975).
37. V. I. Popov, S. A. Dzyuba, and V. M. Moralev, Zh. Fiz.
Khim., 62, 472 (1988).
38. O. S. Mogutnov, S. A. Dzyuba, and Yu. D. Tsvetkov,
Zh. Fiz. Khim., 68, 1235 (1994).
39. H.-H. Schuh and H. Fischer, Int. J. Chem. Kinet., 8, 341
(1976).
40. Kh. S. Bagdasaryan, Yu. I. Kiryukhin, and Z. A. Sinit-
syna, Khim. Phyz., 1, 1666 (1982).
41. W. R. Ware and J. S. Novros, J. Phys. Chem., 70, 3264
(1966).
42. H.-H. Schuh and H. Fischer, Helv. Chim. Acta, 61, 2130
(1978).
43. J. Saltiel, P. T. Shannon, O. C. Zafiriou, and A. K.
Uriate, J. Am. Chem. Soc., 102, 6799 (1980).
44. R. C. Reid, J. M. Prausnitz, and T. K. Sherwood, The
Properties of Gases and Liquids, McGraw-Hill, New
York, 1977, pp. 468–511.
45. T. G. Hiss and E. L. Gussler, AIChE J., 19, 698 (1973).
46. W. Hayduk, R. Castacieda, H. Bromfield, and R. R. Per-
ras, AIChE J., 19, 859 (1973).
47. H. J. V. Tyrrell and P. J. Watkiss, J. Chem. Soc. Fara-
day Trans. I, 75, 1417 (1979).
48. L. Gilles and A. W. Boyd, Can. J. Chem., 54, 531
(1976).
49. S. G. Entelis and R. P. Tiger: Kinetika Reaktssi v Rast-
vorah. Kolichestvennyi Uchet Vliyaniya Sredy (Kinetics
of Reactions in Solutions. Qualitative Inclusion of the
Influence of Media), Khimiya, Moskow, 1973, p. 34.
50. V. M. Berdnikov and A. B. Doctorov, Chem. Phys., 69,
205 (1982).
51. R. A. Lidin, L. L. Andreev, and V. A. Molochko, Sprov-
ochnik po Neorganicheskoi Khimii (Handbook on In-
organic Chemistry), Khimiya, Moskow, 1987, p. 151.
17. V. F. Plyusnin, E. P. Kuznetsova, G. A. Bogdanchikov,
V. P. Grivin, V. N. Kirichenko, and S. V. Larionov, J.
Photochem. Photobiol. A:Chem., 68, 299 (1992).
18. H. Lemmetyinen, R. Ovaskanien, K. Nieminen, K. Vas-
konen, and I. Sychtchikova, J. Chem. Soc. Perkin
Trans., 2, 113 (1992).
19. B. P. Nikolskii (ed.), Spravochnik Khimika (Handbook
of Chemist), Vol. 1, Khimiya, Moskow-Leningrad,
1966, pp. 568–571, 990–1000.
20. C. K. Jorgensen, Mol. Phys., 2, 309 (1959).
21. L. Moggi, G. Varani, M. F. Manfrin, and V. Balzani,
Inorg. Chim. Acta, 4, 335 (1970).
22. A. K. Pikaev, Sovremennaya Radiatsionnaya Khimiya.
Radioliz Gazov i Zhidkostei (Modern Radiation Chem-
istry. Radiolysis of Gases and Liquids), Nauka, Mos-
kow, 1986, p. 350.
23. D. W. Johnson and G. A. Salmon, Can. J. Chem., 55,
2030 (1977).
24. D. W. Johnson and G. A. Salmon, J. Chem. Soc. Far-
aday Trans. I, 71, 583 (1975).
25. C. H. Bamford, C. F. H. Tipper, and R. G. Compton
(eds.), Comprehensive Chemical Kinetics, Vol.25: S. A.
Rice (ed.): Diffusion-Limited Reactions, Elsevier, Am-
sterdam-Oxford-New York-Tokyo, 1985, p. 404.
26. H. Borchers, H. Hausen, K.-H. Hellwege, K. Schafer,
and E. Schmidt (eds.), Landolt-Bornstein Zahlenwerte
und Funktionen aus Physic, Chemie, Astronomie, Geo-
physic und Technick, II Band, 5 Teil, Transporthano-
mene I (Viskositat und Diffusion), Springer-Verlag, Ber-
lin-Heidelberg-New York, 1969, pp. 208–214, 591–
592.
27. J. T. Edward, J. Chem. Ed., 47, 261 (1970).
28. J. E. Baggott, J. Phys. Chem., 87, 5223 (1983).
29. R. Z. Sagdeev, K. M. Salikhov, and Yu.N. Molin, Us-
pekhi Khimii, 46, 569 (1977).
30. J. A. Leone and W. H. Hamill, J. Chem. Phys., 49, 5294
(1968).