Formation of an observable intermediate during the reduction of [Co (III)(NH3)5CN]2+ by ?CR1R2(OH) radicals

Article abstract of DOI:10.1016/j.ica.2005.03.030

^?CR¹R²OH, Rⁱ = CH ₃ or H, react with the complex [Co^III(NH₃) ₅CN]²⁺ to form an observable intermediate probably via bonding to the nitrogen of the cyanide. This intermediate isomerizes to form a second intermediate. The second intermediate decomposes into Co ²⁺(aq), 5NH₄⁺, CN^- and R ¹R²CO. The plausible structures of the intermediates are discussed. The radicals ^?CH³, ^?CH ₂CHO, CH(OH)CO2-, CH2C(CH3)2OH,CO2- and CH3O2 are considerably less reactive towards this complex, the formation of intermediates in their presence is not observed.

Full text of DOI:10.1016/j.ica.2005.03.030

Inorganica Chimica Acta 358 (2005) 2821–2826
www.elsevier.com/locate/ica
Note
Formation of an observable intermediate during the reduction
(
III)
2+
of [Co (NH ) CN] by CR R (OH) radicals
Å
1 2
3
5
a a,b
Ronit Herscu-Kluska , Haim Cohen , Dan Meyerstein
a,c
a
Chemistry Department, Ben-Gurion University of the Negev, P.O. Box 653, 84105 Beer Sheva, Israel
b
Nuclear Research Centre Negev, Beer Sheva, Israel
c
Biological Chemistry Department, College of Judea and Samaria, Ariel, Israel
Received 24 January 2005; accepted 14 March 2005
Available online 18 April 2005
Abstract
Å
1
CR R OH, R = CH or H, react with the complex [Co (NH ) CN] to form an observable intermediate probably via bonding
2
i
III
2+
3
3 5
to the nitrogen of the cyanide. This intermediate isomerizes to form a second intermediate. The second intermediate decomposes
+
2
into Co (aq), 5NH
+
ꢀ
Å
1
, CN and R R CO. The plausible structures of the intermediates are discussed. The radicals CH ,
2
Å
3
4
Å
Å
ꢀ
Åꢀ
tion of intermediates in their presence is not observed.
Å
CH CHO, CHðOHÞCO , CH
2
CðCH
3
Þ OH; CO and CH
3
O2 are considerably less reactive towards this complex, the forma-
2
2
2
2
ꢀ
2005 Elsevier B.V. All rights reserved.
Keywords: Pulse radiolysis; Cobalt complex; Alkyl radicals
1
. Introduction
The reduction of [Co (NH ) X] , X = halides and
the procedure described in the literature [7,8]. It was char-
acterized by the typical UV–vis d–d absorption bands, at
440 nm (57.0 M cm ) and 327 nm (53.9 M cm ) in
good agreement with its reported spectrum [8].
All chemicals and gases used in the study were of
A.R. grade, supplied by Aldrich, Fluka, Merck or Max-
ima. The solutions were prepared using distilled water
that was further purified using a millipore Milli-Q sys-
tem. The final resistance was better than 10 MX/cm.
The pH was measured using a Corning 220 pH meter.
UV–vis spectra were measured using a 8452A HP diode-
array spectrophotometer.
III
2+
ꢀ1
ꢀ1
ꢀ1
ꢀ1
3
5
Å
1
2
pseudo halides, by alkyl radicals and by CR R OH rad-
icals proceeds via the inner sphere mechanism [1–4].
III
3þ
Å
1
by the CR R OH
2
Even the reduction of Co ðNH
3
Þ
6
radicals was shown to proceed via the inner sphere
mechanism [5]. However, in none of these reactions
was an intermediate observed [5,6].
It seemed reasonable that the reduction of
2+
III
[Co (NH ) CN] by the same radicals will also proceed
3 5
via the inner sphere mechanism and that intermediates of
1
2
2þ
_
the type ðNH
3
Þ CoAC@NACR R OH , if formed,
The radicals studied were produced using ionizing
5
6
0
might have a longer life-time and therefore be observable.
Indeed, the results corroborate this assumption.
radiation techniques (pulse radiolysis [9] and Co c
source). The pulse radiolysis technique employed the
Varian Linear Electron Accelerator of the Hebrew Uni-
versity, which produces 5 MeV, 200 mA short electron
pulses of 0.5–1.5 ls (delivered dose 6.5–19.5 Gray/
2
. Experimental
6
0
pulse). The second irradiation source was a Co c radi-
ation source (1.1 MeV photons), Noratom 3500 unit,
(delivered dose rate of 6.5 Gray/min).
III
The complex Co pentaaminocyano perchlorate,
Co(NH ) CN](ClO ) , was synthesized according to
[
3
5
4 2
0
020-1693/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2005.03.030

2822
R. Herscu-Kluska et al. / Inorganica Chimica Acta 358 (2005) 2821–2826
Deaeration of the solutions was achieved by bubbling
atoms are converted into the desired aliphatic radicals
which are the only reactive species present in the
solution.
the desired gas through the solution for 20 min. Transfer
and dilution of deaerated solutions were carried out
using the syringes technique [10].
2þ
The yield of Co
method at 620 nm [11].
was determined by a colorimetric
aq
3. Results and discussion
Formaldehyde was determined by a colorimetric
method [12] of its yellow complex with acetyl acetone
2
3.1. Reaction of [Co(NH ) CN] with the hydrated
+
3
5
ꢀ
(
diacetyl-dihydro lutidine) at 412 nm.
Ethane and methane were determined using a Varian
electron e
aq
+
3
the separation of the gases was carried out on a Poropaq
700 gas chromatograph equipped with a FID detector,
In order to determine whether [Co(NH ) CN] ab-
3 5
sorbs light in the UV–vis region, the hydrated electron
ꢀ
00
0
Q 1/8 , 9 column with helium as the carrier gas.
e
has been used as the reducing reagent. Helium sat-
aq
ꢀ
4
urated solutions containing 1 · 10 M of [Co(NH ) -
CN] and 1.0 M of C(CH ) OH in the pH range 3–6
3 3
3
5
2
+
2.1. Production of the a hydroxy and other aliphatic
radicals
(achieved by addition of 0.050 M phosphate buffer) were
irradiated by a short electron pulse from the linear accel-
Å
When water is exposed to ionizing radiation the pro-
cess occurring is [13]
erator. (The radicals CH C(CH ) (OH) produced by H
2
3 2
abstraction from the (CH ) COH by hydroxyl radicals
3 3
ꢀ
[17] do not, react with the complex, see below.) No
unstable intermediates absorbing light in the UV–vis re-
gion were observed. The only final product detected was
e
;c
ꢀ
Å
Å
H
2
O ! e ð2:65Þ; OHð2:65Þ; Hð0:65Þ; H
H ð0:45Þ
in parentheses are given the relative yields in G values;
G value is the number of product species produced per
00 eV energy absorbed by the sample). If the irradiated
water is saturated by N O (0.022 M) and contains
2
O
2
ð0:75Þ;
ð1Þ
aq
2
2þ
Co
and its G value was determined to be 3.0 ± 0.6,
aq
(
which equals the combined yields of the hydrated elec-
trons and hydrogen atoms, thus, proving that the hy-
drated electrons react via:
1
2
1
2
III
½Co ðNH
Þ5CNꢂ2þ þ e
ꢀ
0
.1–1 M of the appropriate alcohol HCR R (OH), the
following reactions occur within less than 1 ls:
3
aq
þ
5H
þ
2þ
ꢀ
þ
!
and that the complex Co(NH ) CN is either very
CoðNH
3
Þ CN ! Co
þ CN þ 5NH
ð7Þ
ꢀ
þ
Å
9
ꢀ1 ꢀ1
s [14]
5
aq
4
N Oþe þH ! OHþN k ¼ 8:9ꢁ10 M
2
aq
2
+
3
5
ð2Þ
unstable with a half life shorter than the detection limit,
1 ls, and/or that it has no absorption band with
Å
Å
1
2
Å
1
2
OH= H þ HCR R ðOHÞ ! CR R ðOHÞ þ H
2
O=H
2
ꢀ1
ꢀ1
e P 100 M cm in the 270–600 nm region. Both these
possibilities are in accord with the expected properties of
the Co(II) complex.
9
ꢀ1 ꢀ1
k
OH ¼ ð0:8–2Þ ꢁ 10 M
¼ ð2–70Þ ꢁ 10 M
s
[14]
[14]
6
ꢀ1 ꢀ1
s
k
H
ð3Þ
When the organic solute is (CH ) SO methyl radicals
are formed via [15]
2+
.2. Reaction of [Co(NH ) CN] with CH OH,
3 5 2
Å
3
Å
Å
CH(CH )OH and C(CH ) OH
3 3 2
3
2
Å
When N O saturated solutions containing (1–
2
Å
OH þ ðCH
3
Þ SO ! ðCH
3
Þ S OðOHÞ
ꢀ4
2+
2
2
10) · 10 M [Co(NH ) CN] at the pH range 3–6
3
5
9
ꢀ1 ꢀ1
k ¼ 7.0 ꢁ 10 M
followed by
s
ð4Þ
ð5Þ
(phosphate buffer 0.050 M) containing (0.10–1.0) M of
the appropriate alcohol (methanol, ethanol or 2-propa-
nol) are irradiated by a short pulse from the linear
accelerator, the formation of unstable intermediates in
the 5–40 ls time range is observed. These intermediates
decompose to the final stable products within 1000–
Å
ðCH Þ SOðOHÞ ! CH þ ðCH Þ SOðOHÞ
3
2
3
3
2
7
ꢀ1
k ¼ 1:5 ꢁ 10 s
4
hydroxymethyl radical, CH OH, with the complex is
000 ls. A typical kinetic plot for the reaction of the
If the (CH ) SO solution is saturated with 80% N O/
3
2
2
Å
2
methylperoxyl radicals via
0% O , then the methyl radicals are transformed into
2
2
presented in Fig. 1.
The kinetics of formation of the unstable intermedi-
ates obeys pseudo first-order rate laws with a linear
dependence on the concentration of the cobalt (III)
complex. The rates are independent on the alcohol con-
centration, pH, phosphate buffer concentration, and
Å
Å
9
ꢀ1 ꢀ1
s
CH þ O ! CH O k ¼ 1:5 ꢁ 10 M
[16] ð6Þ
3
2
3
2
Thus, in less than 1 ls from the end of the electron
pulse (or the absorption of the c radiation) all
hydroxyl radicals, hydrated electrons and hydrogen

R. Herscu-Kluska et al. / Inorganica Chimica Acta 358 (2005) 2821–2826
2823
Fig. 1. Kinetic changes, measured at 290 nm (inset: first-order plot) after a short electron pulse (0.5 lsecond), was delivered to an N
2
O saturated
ꢀ4
2+
CN] and 0.050 M phosphate buffer at pH 3.0. (a) Absorption spectrum of the
solution containing 0.10 M methanol, 3.0 · 10 M [Co(NH
3
)
5
Å
ꢀ4
2+
intermediate formed by the CH
buffer at pH 3.5.
2
OH radicals. N
2
3 5
O saturated solution contained 0.1 M methanol, 3 · 10 M [Co(NH ) CN] , 0.05 M phosphate
wavelength of measurement or pulse intensity. The re-
sults clearly show that the formation of the unstable
intermediates is due to the reaction of the a-hydroxy-
alkyl radical with the cobalt complex:
centrations and the dependence of k8ob on
(
III)
2+
[{Co (NH ) CN} ] was plotted, Fig. 2 (for the
3
5
hydroxymethyl and hydroxyisopropyl radicals). The
slopes of these lines are the specific rate constants of
7
7
reaction (8) k : (9 ± 2) · 10 , (6 ± 1) · 10
and
for CH OH, CH(CH )OH and
III
Co ðNH
Þ5CNꢂ2þ þ CR R ðOHÞ
Å
1
2
8
½
3
7
ꢀ1 ꢀ1
Å
Å
(
6 ± 1) · 10 M
s
2
3
Å
!
It is proposed that the intermediate, formed in reac-
Intermediate I
ð8Þ
C(CH ) OH, respectively.
3 2
It is interesting to note that extrapolation to zero
complex concentration results in a large intercept for
all three intermediates, thus pointing out that reaction
III
1
2
2þ
_
tion (8), is ðNH
3
Þ Co –CN–CR R ðOHÞ . It cannot
5
+
be Co(NH ) CN as this intermediate does not absorb
3
5
(8) is an equilibrium process:
in this spectral region (see above). It is suggested that
the carbon centered radical binds to the cyano ligand
as no intermediates are observed in the reactions of
III
2þ
Å
1
2
½Co ðNH Þ CNꢂ þ CR R ðOHÞ¡Intermediate ð8Þ
3
5
3þ
these radicals with CoðNH Þ6 (5). Thus, the transient
3
2
1
.0E+05
.5E+05
could be either:
∴
1
2
2+
.
1
2
2+
{
[(NH
3
)
5
Co-CN CR R OH]
[(NH
3
)
5
Co— C=N-CR R OH] }
[
A]
1.0E+05
.0E+04
y = 6E+07x + 78700
.
2
or [(NH
3
)
5
Co-CN
R = 0.99
5
1
2
2+
CR R OH]
B]
[
0.0E+00
.000
0
0.001
Co (NH CN] (M)
0.002
III
2+
[
3 5
)
In order to determine the specific rates of reaction (8)
for the three radicals, the observed pseudo first-order
rate constants were measured at different complex con-
Fig. 2. Dependence of kobs vs. [complex] for the reaction of the
Å
complex with C(CH
3
)
2
OH at pH 2.5, N
2
O saturated solutions
ꢀ3
containing 1 M alcohol and [x] · 10 M complex.

2824
R. Herscu-Kluska et al. / Inorganica Chimica Acta 358 (2005) 2821–2826
where k can be evaluated from the intercepts in Fig. 2:
ꢀ
obtained for the intermediates formed by the
Å
8
4
4
4
ꢀ1
Å
(
17 ± 5) · 10 , (10 ± 3) · 10 and (8 ± 2) · 10 s for
CH OH, CH(CH )OH and C(CH ) OH, respectively.
CH(CH )OH and C(CH ) OH radicals.
3 3 2
Å
Å
Å
The decomposition of the intermediates has been
2
3
3 2
The contribution due to the second-order bimolecular
radical–radical reactions, reaction (9), is only
measured (Fig. 1) and the kinetics also obeyed a first-or-
der rate law for all three radicals. The rates are indepen-
dent of the concentration of the cobalt (III) complex, the
corresponding alcohol, the buffer, the pH or the pulse
intensity and wavelength of measurement. The specific
rate constants for the decomposition reaction (10):
4
ꢀ1 ꢀ1
ꢃ1 · 10 M
s and as such cannot account for the
much higher values observed.
2Å
CR R ðOHÞ ! products
1
2
9
ꢀ1 ꢀ1
s [17,18]
2
k
9
¼ ð1–3Þ ꢁ 10 M
ð9Þ
III
Co ðNH
Å
1
2
2þ
½
3
Þ CN; CR R ðOHÞꢂ intermediate
5
From these values the equilibria constants K can be
8
! products
ð10Þ
2
2
calculated:
(
(5 ± 2) · 10 ,
(6 ± 2) · 10
and
for CH OH, CH(CH )OH and
2
ꢀ1
Å
Å
3
3
7 ± 3) · 10 M
were measured: (6.0 ± 0.9) · 10 , (5.0 ± 0.9) · 10 and
2
3
Å
3
ꢀ1
for CH OH, CH(CH )OH and
2 3
Å
Å
C(CH ) OH, respectively.
The equilibria constants K can also be calculated
III
from the dependence of the 1/[{Co (NH ) CN} ] val-
(4.0 ± 0.7) · 10 s
3
2
Å
8
C(CH ) OH, respectively. The fact that the decomposi-
3 2
2+
3
5
tion of the complex radical intermediates obeys first-or-
der kinetics is quite surprising. The only possibility that
can explain it is that the equilibrium process is followed
ues on 1/OD , Fig. 3. A linear dependence is observed
inf
2
and K can be calculated, the values (4.4 ± 1.0) · 10 ,
8
2
2
ꢀ1
4
ꢀ1
(7.3 ± 1.0) · 10 and (5.3 ± 1.0) · 10 M are the values
obtained for CH OH, CH(CH )OH and C(CH ) OH,
by an intramolecular fast step, k11 > 2 · 10 s to yield
a more stable intermediate, Intermediate II, reaction
(11):
Å
Å
Å
2
3
3 2
respectively. These values are in good agreement with
those obtained from the direct kinetic measurements
and are more accurate as the estimated error is much
smaller.
III
Å
Intermediate I
1
2
2þ
½Co ðNH Þ CN; CR R ðOHÞꢂ
3
5
III
Å
Intermediate II
1
2
2þ 0
The values of K are not large and indicate that only
8
! ð½Co ðNH Þ CN; CR R ðOHÞꢂ Þ
ð11Þ
3
5
ꢃ10% of the radicals are transformed into the complex–
radical intermediate via reaction (8) and that ꢃ90% of
the hydroxyaliphatic radicals are still present in the irra-
diated solutions.
The second intermediate decomposes to the final
products. This conclusion is derived from the fact that
the bleaching of absorption occurs in the ꢃ2000 ls time
range. If the absorption is that of the intermediate which
is formed via reaction (8), then the dominant species in
the solution will be the hydroxyalkyl radical (ꢃ90%, see
above) and as the bimolecular decomposition, reaction
The spectra of the three intermediates in the UV–vis re-
gion 270–600 nm have been measured and one is given for
the hydroxymethyl intermediate, Fig. 1(a). One absorp-
tion band is observed at the UV but its maximum absorp-
tion peak could not be determined, kmax < 270 nm. The
extinction coefficient at 270 nm could be determined using
(
9), for the three a-hydroxyalkyl radicals is diffusion
9
ꢀ1 ꢀ1
s , i.e., the decompo-
ꢀ
1
Å
controlled 2k = (1–3) · 10 M
the value of 440 M for K for CH OH (see above) and
8
9
2
ꢀ1
ꢀ1
sition will obey second-order rate laws and terminate
within less than a millisecond which is not the experi-
mental observation. Furthermore, the fact that there is
no change observed in the spectrum between the first
and the second intermediate indicates that both interme-
diates have very similar spectra in the UV–vis range.
found to be very large e(270 nm) > 10000 M cm
.
This value indicates that it stems from an allowed elec-
tronic transition, probably a LMCT transition. Similar
spectra, with similar molar absorption coefficients, were
III
Å
1
2
2+
Thus, [Co (NH ) CN, CR R (OH)] – Intermediate
3
5
I is undergoing a chemical conversion to produce a sec-
ond intermediate: Intermediate II (with a very similar
UV–vis spectrum as no absorption change is observed
4.00E+03
3.50E+03
3.00E+03
2.50E+03
2.00E+03
1.50E+03
1.00E+03
5.00E+02
0.00E+00
y = 200x-730
R = 0.96
2
III
Å
1
2
2+
0
spectroscopically) ([Co (NH ) CN, CR R (OH)] ) –
3
5
and this second intermediate, which is more stable,
undergoes a first-order intramolecular electron transfer
step followed by fast hydrolysis of the divalent labile co-
balt (II) and protonation of the ammine ligands to pro-
duce the final stable products:
0
5
10
15
20
25
1
/O.D
Fig. 3. Dependence of 1/[complex] vs. 1/OD for the reactions of the
Å
III
ð½Co ðNH
Å
Intermediate II
1
2
2þ 0
complex with CH
1
2
OH at pH 2.5, N
2
O saturated solutions containing
Þ CN; CR R ðOHÞꢂ Þ ! products ð10aÞ
3
5
ꢀ3
M alcohol and [x] · 10 M complex.

R. Herscu-Kluska et al. / Inorganica Chimica Acta 358 (2005) 2821–2826
2825
A reasonable assumption is that the first intramolec-
ular step, reaction (11), is an isomerization reaction of
the intermediate. Such a reaction might be accompanied
by only minor changes in its absorption spectrum as is
observed experimentally.
minecyano complex. As the attack of the a hydroxyali-
phatic radicals is shown to be on the cyano ligand,
this observation suggests that these radicals have an in-
creased reactivity towards the cyano group. Further-
2þ
more, the fact that no Co
was formed indicates
aq
The final products have been determined using steady-
6
that the ammine ligands are not susceptible to an attack
by the aliphatic radicals or for electron transfer step to
the central trivalent cobalt via the outer sphere
mechanism.
0
state irradiations (20 min in the Co c source), of
N O saturated solutions containing [Co(NH ) CN]
2
+
2
3 5
ꢀ
3
1
buffer 0.050 M) and 1.0 M of the appropriate alcohol,
.0 · 10 M at the pH range 3–6 (addition of phosphate
and found to be (for all three reacting a-hydroxyalkyl
3.4. Nature of the intermediates observed
2
þ
aq
radicals) GðCo
Þ ¼ 6:0 ꢄ 0:9. In the case of the
Å
CH OH radicals, formaldehyde was determined to be
2
The short lifetime of the intermediates inhibits use of
other analytical tools to determine the nature of the
intermediates. It is suggested that the first intermediate
formed is [A] and not [B]. This suggestion is based on
the following argument: steric hindrance is expected to
inhibit fast reactions of the radicals with the carbon
atom of the cyano ligand.
the second final product with the same yield:
G = 6.0 ± 0.6. Thus, the yields of the final products
equal those of the a-hydroxyalkyl radicals formed by
the irradiation. This corroborates the suggestion
that the decomposition reaction of the second intermedi-
ate is
III
ð½Co ðNH
Å
1
2
2þ 0
3
Þ CN; CR R ðOHÞꢂ Þ
5
ðIntermediate IIÞ
þ
5H
2þ
1
2
þ
4
!
Co
þ CR R ¼ O þ 5NH
ð12Þ
aq
2
+
3
radicals
.3. Reactivity of [Co(NH ) CN] with other aliphatic
3 5
It was decided to try and check the reactivity of the
trivalent cobalt pentaamminecyano complex with other
aliphatic radicals. For this purpose, N O saturated solu-
2
ꢀ
4
2+
tions containing (1–10) · 10 M [Co(NH ) CN] at
3
5
pH 3–6 (phosphate buffer 0.050 M) and 0.50 M tert-
butanol or 0.30 M of DMSO (dimethyl sulfoxide) or
0
0
.10 M glycolic acid, or 0.050 M sodium formate or
.10 M ethylene glycol were irradiated in the linear
accelerator. The corresponding radicals have been pro-
Å
Å
Å
ꢀ
,
2
duced:
Å
CH C(CH ) OH,
2
CH3,
CHðOHÞCO
3 2
ꢀ
Å
CH CHO and CO . When the solution containing
2
2
DMSO was saturated with 80%N O/20%O , the
2
2
:
CH
all these radicals, no intermediates were observed. No
3
O
radical is the reacting species produced. For
2
2
effect of the [Co(NH ) CN] complex concentration
+
3
5
on the kinetics of disappearance of the radicals was ob-
served. The reactions obeyed second-order rate laws cor-
responding to the bimolecular radical reaction (9).
For the methyl radicals the yield of ethane and meth-
ane which are the gaseous products from methyl radical
reactions via reaction (9) and via hydrogen atom
abstraction from DMSO, was the same in the presence
2
+
and absence of the [Co(NH ) CN] complex. Further-
3
5
more, when identical solutions were irradiated in a
6
0
2þ
Co c-source, no Co
These observations indicate that these aliphatic radicals
do not react (or have low reactivity,
k < 1.5 · 10 M ) with the cobalt (III) pentaam-
was formed as a final product.
aq
Fig. 4. Two possible configurations for the intermediate and its
isomer.
6
ꢀ1 ꢀ1
s

2
826
R. Herscu-Kluska et al. / Inorganica Chimica Acta 358 (2005) 2821–2826
[2] K.R. Olson, N.V. Brezniak, M.Z. Hoffman, Inorg. Chem. 13
The product of the isomerization reaction (11) is ex-
(
3] K.R. Olson, M.Z. Hoffman, J. Chem. Soc., Chem. Commun.
1974) 117.
pected to be similar to the first transient. Theoretical cal-
culations using the semi empirical MM2 [19] program
offer such a transient couple. The calculations resemble
the behavior of water by using continuous phase with
the dielectric constant of water. The calculations lead
to two similar transients, Fig. 4, with a small energy
difference.
The first intermediate is the ‘‘OPEN INTERMEDI-
ATE’’ and the isomerization product of this intermedi-
ate is the ‘‘CYCLIC INTERMEDIATE’’, which is
more stable by ca. 4.9 kcal/mol.
[
[
(1974) 938.
4] A. Haim, H. Taube, J. Am. Chem. Soc. 85 (1963) 595.
[5] H. Cohen, D. Meyerstein, J. Chem. Soc., Dalton Trans. (1977)
1056.
[
6] D.G. Kelley, J.H. Espenson, A. Bakac, Inorg. Chem. 29 (1990)
4
996.
7] K.A. Hofmann, S. Reinsch, Z. Anorg. Chem. 16 (1898) 277.
8] H. Siebert, Z. Anorg. Chem. 327 (1964) 63.
[
[
[9] G. Scholes, P. Shaw, R.L. Wilson, M. Ebert, in: M. Ebert, J.P.
Keene, A.J. Swallaw, J.H. Baxendale (Eds.), Pulse Radiolysis,
Academic Press, New York, 1965.
[
[
10] J.K. Thomas, S. Gordon, E.J. Hart, J. Phys. Chem. 68 (1964)
1
11] D. Katakis, A.O. Allen, J. Phys. Chem. 68 (1964) 1359.
524.
Acknowledgment
[12] T. Nash, Nature 170 (1952) 976.
[
[
[
[
13] M.S. Matheson, L.M. Dorfman, Pulse Radiolysis, MIT Press,
Cambridge, MA, 1969.
This study was supported in part by a grant from the
Israel Atomic Energy Committee. D.M. expresses his
thanks to Mrs. Irene Evens for her continuous interest
and support.
14] G.V. Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, J.
Phys. Chem. Ref. Data 17 (1988) 513.
15] D. Veitwitch, E. Janata, K.D. Asmus, J. Chem. Soc., Perkin
Trans. 2 (1980) 146.
16] J. Rabani, W.A. Mulac, M.S. Matheson, J. Phys. Chem. 81 (1977)
9
9.
[
[
17] M. Simic, P. Neta, E. Hayon, J. Phys. Chem. 73 (1969) 3794.
18] P. Neta, J. Grodkowski, A.B. Ross, J. Phys. Chem. Ref. Data 25
References
(
1996) 709.
[
1] D.C. Campano, E. Kantrowvitz, M.Z. Hoffman, M.S. Weinberg,
J. Phys. Chem. 78 (1974) 686.
[19] MM2 Zindo, CAChe, Version 3, CAChe Scientific Inc.,
Macintosh.