Mechanistic insight from activation parameters for the reaction between co-enzyme B12 and cyanide: Further evidence that heterolytic Co-C bond cleavage is solvent-assisted

Article abstract of DOI:10.1039/b210068a

The potential involvement of the solvent in heterolytic Co-C bond cleavage for vitamin B₁₂ derivatives has been probed by measuring for the first time activation parameters for heterolytic Co-C bond cleavage. If a water molecule is part of the transition state, negative entropies and volumes of activation should be observed. Conversely, if Co-C bond cleavage is a purely dissociative process, these parameters will be positive. Activation parameters have been determined for the reactions of co-enzyme B₁₂ (5′-deoxyadenosylcobalamin, AdoCbl) and the corresponding cobinamide, 5′-deoxyadenosylcobinamide (AdoCbi⁺), with cyanide, since previous studies have shown that these reactions proceed at convenient rates, Co-C heterolytic bond cleavage is rate-determining and the reaction proceeds cleanly (that is, negligible homolytic cleavage). In addition, the kinetics of the reaction of AdoCbl with cyanide in aqueous solution were re-investigated at high concentrations of sodium cyanide (up to 3.00 M) at pH 11.0 and I = 5.0 M (NaClO₄) using UV-visible and ¹H NMR spectroscopy. A significant kinetic saturation was obtained at high CN^- concentrations and the limiting rate constant at 25.0°C was found to be (2.21 ± 0.04) × 10^-2 s^-1, with an overall formation constant for the (β-Ado)(α-CN)Cbl^- intermediate, K_CN/K_Co, of 0.259 ± 0.007 M^-1, ΔH^≠, ΔS^≠ and ΔV^≠ under these conditions were found to be 55 ± 1 kJ mol^-1, -98 ± 3 J K^-1 mol^-1 and -5.7 ± 0.3 cm³ mol^-1, respectively. ΔV^≠ for the reaction of AdoCbi⁺ and CN^- at pH 11.0 and 35.0°C was found to be -4.5 ± 0.4 cm³ mol^-1, while ΔV^≠ for the reaction of AdoCbl and AdoCbi⁺ with tetrabutylammonium cyanide in 92% DMF-8% D₂O at 35°C was found to be -8.2 ± 0.3 and -8.8 ± 0.7 cm³ mol^-1, respectively. The activation parameters data obtained in the present study support the earlier suggestion that the rate determining step for the reaction of AdoCbl and AdoCbi⁺ with CN^- which involves heterolytic cleavage of the Co-C bond of the (β-Ado)(α-CN)Cbl^- and (β-Ado)(α-CN)Cbi intermediates, respectively, is a solvent-assisted, β-elimination process, in which the solvent partially protonates the ribosyl oxygen of the intermediate.

Full text of DOI:10.1039/b210068a

View Article Online / Journal Homepage / Table of Contents for this issue
Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Mechanistic insight from activation parameters for the reaction
between co-enzyme B12 and cyanide: further evidence that
heterolytic Co–C bond cleavage is solvent-assisted†
Mohamed S. A. Hamza,‡a Andrew G. Cregan,b Nicola E. Brasch*b and Rudi van Eldik*a
a Institute for Inorganic Chemistry, University of Erlangen-Nürnberg, Egerlandstr. 1,
91058 Erlangen, Germany
b Research School of Chemistry, The Australian National University, Canberra ACT 0200,
Australia
Received 14th October 2002, Accepted 11th December 2002
First published as an Advance Article on the web 20th January 2003
The potential involvement of the solvent in heterolytic Co–C bond cleavage for vitamin B12 derivatives has been
probed by measuring for the first time activation parameters for heterolytic Co–C bond cleavage. If a water molecule
is part of the transition state, negative entropies and volumes of activation should be observed. Conversely, if Co–C
bond cleavage is a purely dissociative process, these parameters will be positive. Activation parameters have been
determined for the reactions of co-enzyme B12 (5Ј-deoxyadenosylcobalamin, AdoCbl) and the corresponding
cobinamide, 5Ј-deoxyadenosylcobinamide (AdoCbiϩ), with cyanide, since previous studies have shown that these
reactions proceed at convenient rates, Co–C heterolytic bond cleavage is rate-determining and the reaction proceeds
cleanly (that is, negligible homolytic cleavage). In addition, the kinetics of the reaction of AdoCbl with cyanide in
aqueous solution were re-investigated at high concentrations of sodium cyanide (up to 3.00 M) at pH 11.0 and I = 5.0
M (NaClO4) using UV-visible and 1H NMR spectroscopy. A significant kinetic saturation was obtained at high CNϪ
concentrations and the limiting rate constant at 25.0 ЊC was found to be (2.21 0.04) × 10Ϫ2 sϪ1, with an overall
formation constant for the (β-Ado)(α-CN)CblϪ intermediate, KCN/KCo, of 0.259 0.007 MϪ1. ∆H≠, ∆S≠ and ∆V ≠
under these conditions were found to be 55 1 kJ molϪ1, Ϫ98 3 J KϪ1 molϪ1 and Ϫ5.7 0.3 cm3 molϪ1, respectively.
∆V ≠ for the reaction of AdoCbiϩ and CNϪ at pH 11.0 and 35.0 ЊC was found to be Ϫ4.5 0.4 cm3 molϪ1, while ∆V ≠
for the reaction of AdoCbl and AdoCbiϩ with tetrabutylammonium cyanide in 92%DMF–8%D2O at 35 ЊC was
found to be Ϫ8.2 0.3 and Ϫ8.8 0.7 cm3 molϪ1, respectively. The activation parameters data obtained in the
present study support the earlier suggestion that the rate determining step for the reaction of AdoCbl and AdoCbiϩ
with CNϪ which involves heterolytic cleavage of the Co–C bond of the (β-Ado)(α-CN)CblϪ and (β-Ado)(α-CN)Cbi
intermediates, respectively, is a solvent-assisted, β-elimination process, in which the solvent partially protonates the
ribosyl oxygen of the intermediate.
Introduction
The reaction between β-alkyl derivatives of vitamin B12, alkyl-
cobalamins and cyanide has been known since the B12 deriv-
atives were first isolated and characterised over four decades
ago. The mechanism of the reaction between one of the two
co-enzyme forms of vitamin B12, adenosylcobalamin (5Ј-deoxy-
adenosylcobalamin, co-enzyme B12, AdoCbl, Fig. 1) and cyan-
ide has especially attracted attention in the literature.1 On the
basis of finding one kinetically observable step, no detectable
intermediate(s), a linear dependence of the observed rate con-
stant on the cyanide concentration and significantly negative
activation entropy and activation volume values for the rate-
determining step, we initially suggested that rate-determining
associative attack by cyanide occurs at the β-5Ј-deoxyadenosyl
site rather than at the α-dimethylbenzimidazole site of co-
enzyme B12.1a However, studies on the mechanism of the reac-
tion between alkylcobalamins and cyanide for a wide range of
alkylcobalamins have shown that other alkylcobalamins react
with cyanide to form (β-alkyl)(α-cyano)cobalamin intermedi-
ates, which can, in many cases, undergo heterolytic Co–C cleav-
age and react with a second cyanide to form dicyanocobalamin,
and the corresponding β-alkyl Co–C cleavage products
(Scheme 1).2 In addition, more recent detailed kinetic studies
† Electronic supplementary information (ESI) available: NMR and
kinetic data. See http://www.rsc.org/suppdata/dt/b2/b210068a/
‡ On leave from the Department of Chemistry, Faculty of Science,
Ain Shams University, Cairo, Egypt.
Fig. 1 The structure of adenosylcobalamin. The protons that resonate
in the aromatic region are labeled.
D a l t o n T r a n s . , 2 0 0 3 , 5 9 6 – 6 0 2
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
596

View Article Online
Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Scheme 1 Proposed mechanism for the reaction of alkylcobalamins with cyanide.
has previously been proposed for a number of systems (C–N,3a
Si–N,3b M–C (M = Si, Sn)3c), but to our knowledge this is the
first study in which direct kinetic evidence for the involvement
of the solvent in the formation of the transition state preceding
metal–carbon bond cleavage has been obtained.
Alternative explanations for the discrepancy between the
mechanisms for the reactions between cyanide and AdoCbl,
and between cyanide and other alkylcobalamins (that is,
associative or dissociative) were recently discussed.1g EXAFS
studies indicate AdoCbl is mainly present in solution as a five-
cooordinate complex (i.e., the α axial site is vacant), due to
the large trans influence of the β-5Ј-deoxyadenosyl group on the
α-5,6-dimethylbenzimidazole group.4 This could favour an
associative substitution mechanism in aqueous solution. How-
ever, a more detailed comparison of the second-order rate con-
stants found for the cyanation reaction of a series of modified
RCbl’s (R = n-Pr, Me, CF2H, CH2CN, CF3, CNϪ), showed that
this interpretation cannot be correct.1g
Fig. 2 “Base-on” and “base-off” forms of alkylcobalamins. The base-
off form can exist in a five coordinate (no ligand coordinated at the
α axial site) and six coordinate form (a solvent molecule occupies the
α axial site).
have shown that alkylcobalamins must be converted to their
base-off forms (Fig. 2) prior to reacting with cyanide to form
the mono-cyano intermediate.1g,2d Activation parameter meas-
urements indicate that formation of the base-off alkylcobal-
amin (kBO of Scheme 1) is a dissociative (Id or D) process, as
would be expected for this system.1g,2d
Recently we investigated the mechanism of the reaction
between AdoCbl and cyanide in 92%DMF–8%D2O.1f Under
these conditions, a (β-5Ј-deoxyadenosyl)(α-cyano)cobalamin
intermediate was observable by 1H NMR and UV-visible
spectroscopy and the intermediate was light-sensitive, confirm-
ing that it contains a Co–C bond. A mechanism was proposed
in which fast substitution of CNϪ at the α axial site to
form (β-Ado)(α-CN)CblϪ precedes rate-determining, solvent-
assisted Co–C bond heterolysis. Rapid addition of a second
cyanide and rearrangement gave the products, dicyanocobal-
amin, adenineand1-cyano--erythro-2,3-dihydroxy-4-pentenol.
Thus, under these conditions, Co–C bond cleavage at the
β axial site of the (β-Ado)(α-CN)CblϪ intermediate is the rate-
determining step, rather than addition of the first cyanide as
found in aqueous medium.1a Interestingly, a solvent-assisted
bond cleavage involving transfer of a proton from the solvent
Recently complementary investigations on the reaction
between a model compound for adenosylcobalamin, 5Ј-deoxy-
adenosylcobinamide (AdoCbiϩOHϪ, Fig. 3), and cyanide in
both aqueous solution and in DMF–D2O solvent mixtures were
carried out.5 For this system, it was found that fast addition of
cyanide to form a (β-Ado)(α-CN)cobinamide intermediate pre-
cedes rate-determining, Co–C bond heterolysis, regardless of
the DMF–D2O solvent ratio. Given the similarity in the struc-
tures of AdoCbl and AdoCbiϩ (Fig. 1; R = 5Ј-deoxyadenosyl,
and Fig. 3), it seems unlikely that the mechanism of the
reaction between AdoCbl and cyanide in aqueous solution
would be different from that for AdoCbiϩ ϩ CNϪ. Importantly,
there were also serious discrepancies in the activation par-
ameter data.1g—specifically, with the exception of AdoCbl, the
rate-determining step for formation of the mono-cyano inter-
mediate exhibited dissociative activation parameters for all
D a l t o n T r a n s . , 2 0 0 3 , 5 9 6 – 6 0 2
597

View Article Online
Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
atmosphere. Solution pH was adjusted using NaOH solution.
The pD of the D2O solutions was calculated using the relation-
ship pD = pH ϩ 0.40.6 No correction was made for the Naϩ
concentration.
1H NMR spectra were recorded on an Inova 500 MHz NMR
spectrometer equipped with a 5 mm thermostated (25.0
0.2 ЊC) probe. All solutions were prepared in D2O and
referenced externally to a solution of TSP (3-trimethylsilyl
propionate) in D2O.
All kinetic measurements were carried out under pseudo-first
order conditions; that is, the nucleophile concentration was at
least in ten fold excess. An excellent fit to a single first-order rate
equation was found for all kinetic data for at least 5 half-lives
(OLIS-KINFIT fitting program). UV-visible spectra at ambi-
ent pressures were recorded on a Cary 1E spectrophotometer
equipped with a 8 × 6 cell changer and operated with Win UV
Bio software (version 2.00) or a Shimadzu UV-2101 spectro-
photometer. The cell blocks of both spectrophotometers were
thermostated ( 0.1 ЊC).
Kinetic measurements under high pressure were carried out
using either a home-made high pressure unit connected with the
Shimadzu spectrophotometer or a high pressure stopped-flow
instrument.7 At least three kinetic runs were recorded under all
conditions, and the reported rate constants represent the mean
values.
Fig. 3 The structure of 5Ј-deoxyadenosylcobinamide, AdoCbiϩ. The
cobinamide derivatives are model compounds of cobalamins in which
the α-ribazole-3Ј-phosphate of the cobalamin has been cleaved by
hydrolysis of the P–O bond.
alkylcobalamins. This led us to question our interpretation
of the AdoCbl ϩ CNϪ kinetic data in our earlier study,1a and
UV-visible spectra obtained immediately upon mixing AdoCbl
with CNϪ up to 3.0 M CNϪ have subsequently shown that
the reaction between AdoCbl and CNϪ also proceeds via a
mechanism of the type shown in Scheme 1.1g
All reported errors represent one standard deviation of the
mean value. Data fits were carried out using the non-linear least
squares fitting program Origin, version 3.5.
In this study the kinetics of the reaction between AdoCbl and
CNϪ have been re-investigated in aqueous solution using a
much larger range of cyanide concentrations than in our earlier
study.1a These results allow us to directly compare rate and
equilibrium constants for the reactions between AdoCbl and
cyanide and AdoCbiϩ and cyanide in aqueous solution, thus
providing an excellent means of further testing our proposed
general mechanism for the reaction between alkylcobalamins
(including adenosylcobalamin) and alkylcobinamides and
cyanide. In addition, activation parameters have been deter-
mined for the first time for the Co–C heterolytic bond cleavage
process of an alkylcobalamin ((β-Ado)(α-CN)CblϪ) and an
alkylcobinamide ((β-Ado)(α-CN)Cbi) in aqueous solution,
under conditions where the Co–C bond heterolysis is rate-
determining and there is no interference (or minimal inter-
ference) from the pre-equilibrium formation of the mono-cyano
intermediate. Importantly, the determination of activation
parameters for Co–C heterolysis provides us with a novel way
to further probe the suggested involvement of the solvent in the
transition state of the Co–C bond cleavage process. Activation
parameters for the Co–C bond cleavage step of the cyanation
of AdoCbl and AdoCbiϩ have also been determined in
92%DMF–8%D2O, and show how the solvent controls the rate
and mechanism of this step.
Kinetic measurements on the reaction between AdoCbl and
NaCN
The reaction between AdoCbl and CNϪ was studied at [NaCN]
= 0.150–3.00 M, pH 12.0, I = 5.0 M (NaClO4). Each measure-
ment was done in triplicate. In a typical experiment, 0.800 mL
of a NaCN solution (0.300–2.40 M NaCN) and 0.800 mL of
a ∼1.00 × 10Ϫ4 M AdoCbl solution were mixed in a tandem
cuvette and the absorbance monitored at 367 nm. For NaCN ≥
1.40 M, 10.0 µL of ca. 1.30 × 10Ϫ2 M AdoCbl solution was
injected into 3.00 mL of a NaCN solution and the absorbance
monitored at 367 nm. Similar procedures were used to deter-
mine rate constants for the reaction between AdoCbl and CNϪ
in D2O (pD 12.0, I = 5.0 M (NaClO4)).
1H NMR Spectroscopy measurements on the reaction between
AdoCbl and NaCN
Experiments were carried out at pD 12.0, I = 5.0 M (NaClO4).
∼12 mg of solid AdoCbl was dissolved in a NaCN solution
1
([NaCN] = 0–3.00 M), filtered, (45 µm filter) and a H NMR
spectrum immediately recorded.
Results and discussion
Kinetic measurements
Experimental
Adenosylcobalamin (AdoCbl, >98%) was purchased from
Aldrich and sodium perchlorate (≥98%) from BDH. Tetra-
butylammonium perchlorate (≥99%) tetrabutylammonium
cyanide ( ∼97%) and dimethylformamide were from Fluka.
CAPS buffer (≥98%) was provided by Sigma. All solutions were
made in D2O at a total ionic strength of 5.0 M (NaClO4) except
otherwise stated. Distilled water was purified through an
ultra-pure water system.
In order to obtain additional support for the mechanism of the
reaction between AdoCbl and CNϪ in aqueous solution as pro-
posed in Scheme 1, the dependence of the observed rate con-
stant of the reaction between AdoCbl and CNϪ was studied as
a function of CNϪ concentration by UV-visible spectroscopy,
up to CNϪ concentrations of 3.00 M. At each CNϪ concen-
tration, the data gave an excellent fit to a first-order rate
equation. The results are presented in Fig. 4.
All pH measurements were carried out at 25.0 ЊC using an
Orion Model 710A pH meter in conjunction with an Orion
9101 BN glass electrode and an Orion 900200 double junction
reference electrode, or a Mettler Delta 350 pH meter with a
combined glass electrode. The electrodes were standardised
using BDH 6.98 and 10.00 buffer solutions. Measurements in
alkaline solution were carried out under a nitrogen or argon
Curvature is observed, typical for saturation kinetics in which
a rapid pre-equilibrium precedes rate-determining Co–C bond
cleavage as proposed in Scheme 1. (Note that for [CNϪ] ≤
0.500 M, the plot is essentially linear, as found in our earlier
study.1a) We have recently provided UV-visible spectro-
scopic evidence that the intermediate is (β-Ado)(α-CN)CblϪ,1g
since the intermediate absorbs strongly in the 580–620 nm
D a l t o n T r a n s . , 2 0 0 3 , 5 9 6 – 6 0 2
598

View Article Online
Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Fig. 4 Plot of observed rate constant (kobs) versus CNϪ concentration
for the reaction between AdoCbl and CNϪ at 25.0 ЊC ([AdoCbl]
∼5 × 10Ϫ5 M, [NaCN] = 0.100–3.00 M, H2O, 0.100 M CAPS,
pH 11.0, I = 5.0 M (NaClO4)). The best fit of the data to eqn. (1) gave
Fig. 5 Plot of 1H NMR chemical shift versus [CNϪ] for the B4
nucleotide signal of {AdoCbl ϩ (β-Ado)(α-CN)CblϪ} (D2O, 0.100 M
CAPS, pD 11.0, 25.0 ЊC, I = 5.0 M (NaClO4)). The data has been fitted
k2 = (2.21 0.04) × 10Ϫ2 sϪ1 and KCN/KCo = 0.259 0.007 MϪ1
.
to eqn. (3) in the text, giving KCN/KCoЈ = 0.10 0.03 MϪ1, δAdoCbl
6.29 0.01 ppm and δAdoCNCbl = 8.0 0.3 ppm.
=
region, which is characteristic of (β-alkyl)(α-CN)cobalamin
and (β-alkyl)(α-CN)cobinamide complexes.1f,1g,2a,2d,5 The inter-
mediate also decomposes in light (i.e., it contains a light-sensi-
tive Co–C bond.1b)
The rate law for a mechanism of the type shown in Scheme 1
is
reaction mixture is monitored at longer times, the peak areas of
these signals decrease until no longer observable, while new
peaks grow in at a comparable rate. The new peaks can be
attributed to the products adenine (7.97, 8.12 ppm9a) the
B2, B4, B7, R1 and C10 resonances of dicyanocobalamin
(8.29, 7.44, 7.34, 6.31(d) and 5.76 ppm9), and the cyanohydrin
1-cyano--erythro-2,3-dihydroxy-4-pentenol (5.31, 5.33, 5.35,
5.38, 5.39, 5.41 ppm and a multiplet at 5.9–6.0 ppm9a).
Observed rate constants for the reaction between AdoCbl
kobs = k2(KCN/KCo)[CNϪ]/(1 ϩ (KCN/KCo)[CNϪ]) (1)
where KCN/KCo is the overall formation constant for the (β-Ado)-
(α-CN)CblϪ intermediate, as defined in Scheme 1. The best fit
of the data in Fig. 4 to eqn. (1) gave k2 = (2.21 0.04) × 10Ϫ2 sϪ1
and KCN/KCo = 0.259 0.007 MϪ1 (H2O, pH 11.0, I = 5.0 M
(NaClO4), 25.0 ЊC). In an earlier paper, the equilibrium con-
stant ratio for formation of (β-Ado)(α-CN)CblϪ, KCN/KCo, was
determined in H2O, by measuring UV-visible spectra of solu-
tions of AdoCbl upon immediate addition of NaCN, as a func-
1
and CNϪ can also be determined from H NMR peak areas,
although this was not done in this study since we have
previously found that more accurate data for the reaction
between B12 derivatives and cyanide is obtained by UV-visible
spectroscopy.1f,5
Fig. 5 gives the observed chemical shift for the B4 signal8 of
AdoCbl (actually a mixture of AdoCbl and (β-Ado)(α-CN)-
CblϪ) as a function of CNϪ concentration. Assuming that
AdoCbl and the intermediate, (β-Ado)(α-CN)CblϪ, are in fast
exchange on the 1H NMR time scale; that is
tion of CNϪ concentration, giving KCN/KCo = 0.6
0.1 MϪ1
(pH 11.0, 25 ЊC).1g As expected, this value of KCN/KCo is some-
what different and has a larger associated error since the ionic
strength was not controlled, whereas the measurements in this
study have been carried out maintaining the total ionic strength
at 5.0 M (NaClO4).
(2)
The kinetics of the reaction between AdoCbl and CNϪ were
also studied in D2O using the same methodology, giving k2 =
(1.93 0.04) × 10Ϫ2 sϪ1 and KCN/KCo = 0.25 0.01 MϪ1 (D2O,
pD 11.0, I = 5.0 M (NaClO4), 25.0 ЊC). The corresponding plot
of observed rate constant versus cyanide concentration and the
best fit of the data to eqn. (1) are given in Fig. A in the Support-
ing Information. These data are useful since they allow us to
compare these results with those from the {AdoCbiϩ ϩ CNϪ}
system (see later). In addition, the equilibrium constant ratio
KCN/KCo can also be determined in D2O using 1H NMR
spectroscopy (see below), thus providing a further method to
check the proposed mechanism.
then
δobs = (δAdoCbl ϩ (KCN/KCoЈ)[CNϪ]δAdoCNCbl) /
(1 ϩ (KCN/KCoЈ)[CNϪ]) (3)
where δobs is the observed chemical shift at a specific CNϪ
concentration, δAdoCbl and δAdoCNCbl are the chemical shifts of
AdoCbl and (β-Ado)(α-CN)CblϪ respectively, and KCN/KCoЈ is
the overall formation constant of the intermediate determined
1
by H NMR spectroscopy corresponding to eqn. (2). The best
fit of the data in Fig. 5 to eqn. (3) gives KCN/KCoЈ = 0.10
0.03 MϪ1, as shown by the solid line in Fig. 5. Alternatively, the
data can be fitted assuming KCN/KCoЈ = 0.25 MϪ1; that is, the
value of KCN/KCo obtained by UV-visible spectroscopy. Using
this value of KCN/KCoЈ also results in a very good fit of the
experimental data (see Fig. B of the Supporting Information).
This indicates that the value of KCN/KCoЈ obtained by 1H NMR
spectroscopy should be regarded as an estimate only. The
same conclusion is reached when examining the standard devi-
ations of the values in Fig. 5, which are large. The results are
summarized along with other available data in Table 1.
Determination of KCN/KCo by 1H NMR spectroscopy
The reaction between AdoCbl and CNϪ can also be followed by
1H NMR spectroscopy. Indeed, this method has previously
been used by us to study the reaction between AdoCbl and CNϪ
in 92%DMF–8%D2O.1f Upon dissolving AdoCbl in D2O, 8
signals are observed in the aromatic region attributable to the
A2, A8 and A11 proton resonances of the 5Ј-deoxyadenosyl
ligand, the C10 proton of the corrin ring, and the B2, B4, B7
and R1 resonances of the nucleotide (Fig. 1).8 If NaCN is
added to the solution, the peaks in the aromatic region are
shifted downfield (and broaden), with their chemical shifts
being dependent on the specific CNϪ concentration. When the
Activation parameter measurements
In order to further probe the possible involvement of the sol-
vent in formation of the transition state for Co–C heterolysis of
D a l t o n T r a n s . , 2 0 0 3 , 5 9 6 – 6 0 2
599