Kinetics and mechanism of axial ligand substitution of alkyl cobaloximes by substituted pyridines in different solvents

Article abstract of DOI:10.1002/ejic.200300733

Ligand-substitution reactions of trans-RCo(Hdmg)₂S (where R = PhCH₂ or CF₃CH₂, Hdmg = dimethylglyoximate, and S = H₂O or MeOH) were studied for the nucleophiles 4-aminopyridine (4-NH₂Py), p

Full text of DOI:10.1002/ejic.200300733

FULL PAPER
Kinetics and Mechanism of Axial Ligand Substitution of Alkyl Cobaloximes
by Substituted Pyridines in Different Solvents
Basam M. Alzoubi,[a] Mohamed S. A. Hamza,[a,b] Alessandro Felluga,[c] Lucio Randaccio,[c]
Giovanni Tauzher,[c] and Rudi van Eldik*[a]
Keywords: Cobaloximes / Solvent effects / trans effect / Ligand-substitution reactions / Nitrogen heterocycles
Ligand-substitution reactions of trans-RCo(Hdmg)2S (where
R = PhCH2 or CF3CH2, Hdmg = dimethylglyoximate, and S =
tion parameters for the substitution of H2O by Py and 4-
CNPy in the case where R = CF3CH2 were also found. The
H2O or MeOH) were studied for the nucleophiles 4-aminopy- kinetic data and activation parameters show that the dissoci-
ridine (4-NH2Py), pyridine (Py) and 4-cyanopyridine (4-
CNPy). From the pressure and temperature dependence of
ative character of the reaction decreases on changing the R
group from PhCH2 to CF3CH2 and the solvent from MeOH
the substitution of methanol by 4-NH2Py, Py, and 4-CNPy, to H2O.
the activation parameters ∆H϶, ∆S϶, and ∆V϶ were estim-
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
ated for the cases where R = PhCH2 and CF3CH2. The activa- Germany, 2004)
Introduction
cobalamin complexes. Ligand-substitution reactions of vit-
amin B12 and its models in general follow a dissociative (Id
or D) type of mechanism.[16Ϫ28] In the case of the co-
enzyme, evidence was initially reported for an associative
substitution mode, which was, however, recently reinter-
preted and instead assigned to a subsequent CoϪC hetero-
lysis reaction.[29] An interchange mechanism may involve
either associative or dissociative activation (Ia or Id mecha-
nism), according to the relative importance of bond making
and bond breaking in the transition state. Evidence that
usually leads to the assignment of a dissociative mechanism
includes the absence of any relationship between the nature
of the entering nucleophile and the rate constant, an in-
crease in lability with an increase in the electron donating
ability of the alkyl group R in the trans position, and posi-
tive values for both ∆V϶ and ∆S϶.
The study of the chemical and physical properties of or-
ganocobalt compounds, which serve as models for vitamin
B12, is of interest both to inorganic and bioinorganic chem-
ists. Various pseudo-octahedral organometallic complexes
of CoIII, containing equatorial ligands such as diox-
imes,[1Ϫ9] imino groups/oximes[10] and Schiff bases derived
from diamines and salicylaldehyde[11] or acetylacetonate[12]
have been proposed as models for the vitamin B12 co-
enzyme. Some of us recently studied the reaction of cyanide
with various alkylcobalamins (RCbl, where R ϭ CN,
CNCH2, CF3, CF3CH2, CF2H, CH3, BrCH2) and found
that the nature of the alkyl group has an important influ-
ence on the thermodynamic equilibrium constants, and the
kinetics and mechanism of the substitution reactions of the
axial ligand trans to the alkyl group.[13,14]
In general, kinetic studies are performed under pseudo
first order conditions using a large excess of ligand over the
complex. A linear dependence of the observed pseudo first-
order rate constant on the concentration of the entering
nucleophile has been observed in almost in all cases, both
for aquacobalamin and synthetic models in aqueous solu-
tion. The first deviation from linearity was reported for the
reaction of [RCo{(DO)(DOH)pn}H2O]ϩ complexes (R ϭ
Me and Et).[15] Curvature in the plot of kobs versus ligand
concentration generally characterizes multistep processes,
such as a limiting dissociative mechanism (D), which in-
volves a pentacoordinate intermediate, or a mechanism in
which intermediate species are formed in a rapid pre-equi-
librium which precedes the slow substitution step. In the
above-mentioned complexes, linear relationships between
The nature of the axial alkyl group in cobaloximes[2] and
in related [RCo{(DO)(DOH)pn}H2O]ϩ complexes[15] has
an effect on the substitution of the ligand in the trans posi-
tion. This substitution is very rapid and involves a dissoci-
ative activation mode, similar to the substitution of labile
[a]
Institute for Inorganic Chemistry, University of Erlangen-
Nürnberg,
Egerlandstr. 1, 91058 Erlangen, Germany
[b]
Department of Chemistry, Faculty of Science, Ain Shams
University
Cairo, Egypt
[c]
`
Dipartimento di Scienze Chimiche, Universita degli Studi di
Trieste,
34127 Trieste, Italy
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
Eur. J. Inorg. Chem. 2004, 653Ϫ659
DOI: 10.1002/ejic.200300733
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
653

B. M. Alzoubi, M. S. A. Hamza, A. Felluga, L. Randaccio, G. Tauzher, R. van Eldik
FULL PAPER
kobs and ligand concentration were found for the substi- terms of the interchange mechanism presented in Equa-
tution of water by a number of nucleophiles. However, cur- tions (2) and (3). The corresponding rate law is given in
vature was observed for L ϭ imidazole, benzylamine, pyri- Equation (4), where K represents the equilibrium constant
dine and 4-methylpyridine. Since the later four ligands all for precursor formation, and k the interchange rate con-
contain aromatic group, it was tentatively suggested that stant. In these cases, k and K can be separated kinetically
these are involved in forming the intermediate, perhaps so that their values can be discussed individually.
through a π-π interaction with the conjugated system of the
(DO)(DOH)pn equatorial ligand.
More recently, a non-linear dependence of kobs on the
concentration of the entering ligand has been reported for
the reaction of aquacobalamin with substituted pyri-
dines.[16] This was first incorrectly interpreted as evidence
for the operation of a limiting dissociative mechanism, but
later work clearly demonstrated that the observed deviation
was due to the rapid formation of adducts involving the
nucleophiles and cobalamin, preceding the slow substi-
tution reaction according to a dissociative interchange
mechanism.[17] This finding greatly enhanced our interest in
mechanisms associated with the reactions of nucleophiles
that contain an aromatic group because of the connection
with the biochemistry of vitamin B12. Therefore, in order
to obtain a deeper understanding of this phenomenon, we
extended the present study to cobaloximes, which are the
Figure 1. Plots of kobs versus [4-NH2Py] for the reaction with trans-
(CF3CH2)Co(Hdmg)2(MeOH) as a function of temperature. Exper-
imental conditions: [CoIII] ϭ 1 ϫ 10Ϫ3  and temperature 20.0 (A),
25.0 (B), 30.0 (C), 35.0 (D), and 40.0 °C (E)
most widely recognised model systems for the study of vit-
amin B12. We report kinetic data for ligand-substitution re-
actions of RCo(Hdmg)2S with various pyridines, where R ϭ
PhCH2 and CF3CH2, S ϭ H2O and MeOH and L ϭ 4-
NH2Py, Py, and 4-CNPy, in order to study the kinetic trans
effect in alkyl cobalt complexes as a function of nucleophile
concentration, temperature and pressure.
Results and Discussion
The ligand-substitution reactions of trans-RCo(Hdmg)2S
(where R ϭ CF3CH2 and PhCH2) with various nucleophiles
L ϭ 4-NH2Py, Py, and 4-CNPy to form RCo(Hdmg)2L,
according to Equation (1), were studied with MeOH and
H2O as solvents.
RCo(Hdmg)2S ϩ L Ǟ RCo(Hdmg)2L ϩ S
(1)
The results fall into two categories: (i) when the incoming
ligand is Py or 4-NH2Py, a non-linear concentration depen-
dence is obtained for both PhCH2 (S ϭ MeOH) and
CF3CH2 (S ϭ MeOH or H2O) derivatives, (ii) when the
incoming ligand is 4-CNPy, a linear relationship is observed
for both the PhCH2 (S ϭ MeOH) and CF3CH2 (S ϭ
MeOH or H2O) derivatives. The reaction of (4Ϫ10) ϫ 10Ϫ4
 trans-RCo(Hdmg)2(H2O) with an excess of nucleophile
(4-NH2Py and Py) was studied at various temperatures in
MeOH and H2O. The results are shown in Figure 1 (R ϭ
CF3CH2, L ϭ 4-NH2Py in MeOH) and Figure 2 (R ϭ
CF3CH2, L ϭ Py in H2O); see also Figure S1 (R ϭ PhCH2,
L ϭ 4-NH2Py in MeOH) and Figure S2 (R ϭ CF3CH2,
L ϭ Py in MeOH) in the Supporting Information (see foot-
note on the first page of this article). The plots show signifi-
Figure 2. Plots of kobs versus [Py] for the reaction with trans-
(CF3CH2)Co(Hdmg)2(H2O) as a function of temperature. Exper-
imental conditions: [CoIII] ϭ 1 ϫ 10Ϫ3 , pH ϭ 8.0, I ϭ 0.1 
NaClO4 and temperature 20.0 (A), 25.0 (B), 30.0 (C), 35.0 (D), and
40.0 °C (E)
(2)
(3)
kobs ϭ kK[L]/(1 ϩ K[L])
(4)
Figures 3 and S3 (for the latter see the Supporting Infor-
cant curvature at high nucleophile concentration and negli- mation) show plots for kobs versus [L] for the reaction of
gible intercepts, which can be interpreted (see below) in R ϭ CF3CH2 with L ϭ 4-CNPy in MeOH and H2O, and
654
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 653Ϫ659

Axial Ligand Substitution of Alkyl Cobaloximes
FULL PAPER
R ϭ PhCH2 with L ϭ Py and 4-CNPy in MeOH, respec- with 4-NH2Py, Py and 4-CNPy. Similar plots are presented
tively. The linear plots have negligible intercepts, which indi- in Figures S4 and S5 (see the Supporting Information) for
cate that the back reactions do not contribute significantly the reactions between RCo(Hdmg)2H2O and the nucleo-
and that no parallel reaction takes place. This behaviour philes L ϭ 4-NH2Py, 4-CNPy, and Py, where R ϭ PhCH2
can be expressed by the rate law given in Equation (5), from and S ϭ MeOH, and the nucleophiles L ϭ 4-CNPy and Py,
which it follows that the overall second-order rate constant where R ϭ CF3CH2 and S ϭ H2O, respectively. The acti-
(ka) is a composite of k and K, and cannot be separated vation parameters for the substitution of methanol by 4-
easily.
NH2Py, Py, and 4-CNPy in the case where R ϭ PhCH2
were found to be ∆H϶ ϭ 67
2, 77 1 and 80
2 kJ·molϪ1, ∆S϶ ϭ ϩ28 7, ϩ57 5 and ϩ71 6 J·molϪ1
Ϫ1, and ∆V϶ ϭ ϩ9.8 0.1, ϩ10.3 0.3 and ϩ13.0 0.3
K
cm3 molϪ1, respectively. The activation parameters in the
case where R ϭ CF3CH2 were found to be ∆H϶ ϭ 90 9,
91 5 and 89 3 kJ·molϪ1, ∆S϶ ϭ ϩ49 29, ϩ42 17
and ϩ36 8 J·molϪ1 KϪ1, and ∆V϶ ϭ ϩ7.5 0.3, ϩ8.6
0.1 and ϩ9.4 0.1 cm3 molϪ1, respectively. The activation
parameters for substitution of H2O by Py and 4-CNPy,
where R ϭ CF3CH2, were found to be ∆H϶ ϭ 94 4 and
93 1 kJ·molϪ1, ∆S϶ ϭ ϩ62 14 and ϩ55 4 J·molϪ1
K
Ϫ1, and ∆V϶ ϭ ϩ5.7 0.4 and ϩ7.4 0.1 cm3 molϪ1
,
respectively. These activation parameters, in particular the
values of ∆V϶, support the operation of a dissociative
mechanism.
The discussion of the kinetic data will focus on the ob-
served second-order rate constants for the overall reaction
given in Equation (1), which is a composite of K and k, as
given in Equations (2) and (3). Although for some systems
k and K could be separated kinetically, the activation par-
ameters were always determined for the combined value of
kK, i.e. ka. Little is presently known about the temperature
and pressure dependence of the precursor formation step,
and for that reason the thermodynamic contributions of K
form part of the overall reported parameters. This compli-
cation should be kept in mind when interpreting the overall
rate and activation parameters.
Figure 3. Plots of kobs versus [4-CNPy] for the reaction with trans-
(CF3CH2)Co(Hdmg)2S where S ϭ MeOH (A) and H2O (B) as a
function of temperature. Experimental conditions: [CoIII] ϭ 1 ϫ
10Ϫ3 , (pH ϭ 8.0, I ϭ 0.1  NaClO4 for S ϭ H2O) and tempera-
ture 25.0 °C
kobs ϭ ka[L] ϭ kK[L]
(5)
Tables 1, 2, 3, and 4 summarize the kinetic data obtained
at different temperatures and pressures for the series of
reactions investigated in this study. Plots of ln(ka) versus
pressure give good linear relationships as shown in Figure 4
for the reactions of trans-(CF3CH2)Co(Hdmg)2(MeOH)
The reactivity of the cobaloxime complexes shows a
strong dependence on the nature of the alkyl substituent R.
The lability of these complexes was affected by the elec-
tronic and steric properties of the alkyl group. The second-
order rate constants for the reaction of trans-
(PhCH2)Co(Hdmg)2(MeOH) with 4-NH2Py, Py, and 4-
CNPy were found to be 356, 182, and 296 Ϫ1 sϪ1 at 25.0
°C, respectively. However, those obtained for the reaction
with trans-(CF3CH2)Co(Hdmg)2(MeOH) were 0.330, 0.147
and 0.153 Ϫ1 sϪ1 at 25.0 °C, respectively. The second-order
Table 1. Kinetic data for the reaction of trans-
RCo(Hdmg)2(MeOH) with 4-NH2Py as a function of temperature
R
T, °C
k, sϪ1
K, Ϫ1
kK, Ϫ1 sϪ1
[a]
PhCH2
5.0
10.0
127
4
0.38 0.02 48
3
7
187 10 0.47 0.03 88
15.0
20.0
25.0
228 22 0.62 0.07 141 21
458 39 0.52 0.06 238 34
1017 135 0.35 0.03 356 56
rate
constants
for
the
reaction
of
trans-
(CF3CH2)Co(Hdmg)2(H2O) with Py and 4-CNPy in H2O
were found to be 0.317 and 0.214 Ϫ1 sϪ1 at 25.0 °C, respec-
tively. This trend shows that the kinetic trans effect is
greater for R ϭ PhCH2 than for R ϭ CF3CH2, in accord-
ance with the donor properties of the alkyl group. For the
∆H϶, kJ·molϪ1
∆S϶, J·molϪ1 KϪ1
67
ϩ28
2
7
CF3CH2 [b] 20.0
0.24 0.02 1.13 0.16 0.27 0.04
0.61 0.05 0.54 0.07 0.33 0.05
0.88 0.03 0.94 0.05 0.83 0.05
1.50 0.07 0.96 0.09 1.44 0.15
2.60 0.16 1.02 0.12 2.65 0.35
25.0
30.0
35.0
40.0
complexes
trans-[RCo(LNHpy)(HLNHpy)]ϩ,
where
HLNHpy is the tridentate 2-(2-pyridylethyl)amino-3-but-
anone oxime ligand and LNHpyϪ its conjugate base, the
second-order rate constants for substitution of 2-(pyridyl-
ethyl) were found to be 19.1, 0.25, 2.2 ϫ 10Ϫ2, and 1.7 ϫ
10Ϫ2 Ϫ1 sϪ1 for R ϭ Et, Me, CF3CH2, and ϪCH2Ϫ,
respectively.[30] The reaction between alkylcobalamins
∆H϶, kJ·molϪ1
90
9
∆S϶, J·molϪ1 KϪ1
ϩ49 29
[a]
[PhCH2Co(Hdmg)2MeOH] ϭ 4 ϫ 10Ϫ4 , [4-NH2Py] ϭ
0.01Ϫ0.64  in MeOH. [b] [CF3CH2Co(Hdmg)2MeOH] ϭ 1 ϫ 10Ϫ3
, [4-NH2Py] ϭ 0.02Ϫ1.5  in MeOH.
Eur. J. Inorg. Chem. 2004, 653Ϫ659
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
655

B. M. Alzoubi, M. S. A. Hamza, A. Felluga, L. Randaccio, G. Tauzher, R. van Eldik
FULL PAPER
Table 2. Kinetic data for the reaction of trans-RCo(Hdmg)2(S) with L in MeOH and H2O as a function of temperature
ka, Ϫ1 sϪ1
R ϭ PhCH2, S ϭ MeOH
L ϭ Py[a] L ϭ 4-CNPy[b]
R ϭ CF3CH2,L ϭ 4-CNPy
T, °C
S ϭ MeOH[c]
S ϭ H2O[d]
5.0
10.0
15.0
20.0
18.3 0.1
33.9 0.3
28
49
97
173
296
2
2
1
2
3
64
108
182
1
1
1
0.073 0.001
0.153 0.001
0.254 0.001
0.464 0.003
0.820 0.004
0.112 0.001
0.214 0.003
0.393 0.004
0.739 0.004
1.390 0.020
25.0
30.0
35.0
40.0
∆H϶, kJ·molϪ1
∆S϶, J·molϪ1 KϪ1
77
ϩ57
1
5
80
ϩ71
2
6
89
3
93
1
ϩ36
8
ϩ55
4
[a]
[b]
[c]
[PhCH2Co(Hdmg)2MeOH] ϭ 4 ϫ 10Ϫ4 , [Py] ϭ 0.01 .
[PhCH2Co(Hdmg)2MeOH] ϭ 4 ϫ 10Ϫ4 , [4-CNPy] ϭ 0.01 .
[CF3CH2Co(Hdmg)2MeOH] ϭ 4 ϫ 10Ϫ4 , [4-CNPy] ϭ 0.08 . [CF3CH2Co(Hdmg)2H2O] ϭ 1 ϫ 10Ϫ3 , [4-CNPy] ϭ 0.08 , pH ϭ
[d]
8 and I ϭ 0.1  NaClO4.
Table 3. Kinetic data for the reaction of trans-(CF3CH2)Co(Hdmg)2(S) with Py in MeOH and H2O as a function of temperature
T, °C
k, sϪ1
K, Ϫ1
kK, Ϫ1 sϪ1
S ϭ MeOH[a]
20.0
25.0
30.0
35.0
0.88 0.05
1.13 0.05
2.04 0.14
3.36 0.40
9.82 0.89
0.07 0.01
0.13 0.01
0.12 0.01
0.12 0.02
0.08 0.01
0.062 0.010
0.147 0.013
0.245 0.026
0.403 0.083
0.786 0.121
40.0
∆H϶, kJ·molϪ1
∆S϶, J·molϪ1 KϪ1
91
5
ϩ42 17
S ϭ H2O[b]
20.0
25.0
30.0
35.0
0.33 0.01
0.96 0.08
1.71 0.25
2.20 0.19
3.40 0.08
0.57 0.03
0.33 0.04
0.34 0.07
0.54 0.07
0.68 0.03
0.188 0.011
0.317 0.047
0.581 0.147
1.188 0.185
2.312 0.117
40.0
∆H϶, kJ·molϪ1
∆S϶, J·molϪ1 KϪ1
94
4
ϩ62 14
[a]
[b]
[CF3CH2Co(Hdmg)2MeOH] ϭ 1 ϫ 10Ϫ3  and [Py] ϭ 0.04Ϫ4.0 . [CF3CH2Co(Hdmg)2(H2O)] ϭ 1 ϫ 10Ϫ3 , [Py] ϭ 0.04Ϫ1.28
, pH ϭ 8 and I ϭ 0.1  NaClO4.
(RCb1) and cyanide showed that the kinetic trans effect de- higher than that found for the corresponding reaction with
creases in the order (R ϭ ) Pr Ͼ Me (ca. 104 Ϫ1sϪ1) Ͼ cysteine,[34] and this was ascribed to metal-to-ligand π
CF3CH2 Ͼ CF2H (103 Ϫ1 Ϫ1) Ͼ CF3 Ͼ NCCH2 Ͼ CN bonding. The range of rate constants for axial ligand bind-
s
(0.1 Ϫ1 sϪ1).[14] The substitution of H2O by CNϪ for vari- ing to methylaquacobaloxime are 4.3Ϫ7.5 (RNH2),
ous cobinamides (RCbi) increases in the order (R ϭ) 12.8Ϫ27.3 (RSϪ), 49.6Ϫ55.2 (RSH), and 92Ϫ200 (X-py)
H2O ϭ DMBz Ͻ OHϪ Ͻ CNϪ Ͻ CH2ϭCH, Me, and Et, Ϫ1 sϪ1 [35] That is, the π-acceptor ligands (X-Py, RSϪ and
.
with the corresponding rate constants ranging from about RSH) react more rapidly than pure σ donors (RNH2). In
103 to about 108 Ϫ1 sϪ1 [31]
.
the ligand-substitution reactions of trans-Me-
The pKBHϩ values for 4-NH2Py, Py, and 4-CNPy are Co(Hdmg)2(H2O), the higher reactivity of the substituted
9.17,[32] 5.23,[32] and 1.7[33] (estimated), respectively. Both in pyridines (as compared to neutral or anionic thiolates)
water and methanol, the second-order rate constants for the stand in contrast to the opposite reactivity order for these
ligation of the pyridines are only slightly affected by their same nucleophiles with various unsaturated carbon electro-
basicity. Such a weak sensitivity of the kK values is consist- philes.[35] The ratio of the second-order rate constants for
ent with an Id mechanism, which is controlled by the break- the reaction of pyridine and 4-CNpy with alkylcobaloxime
ing of the CoϪS bond. Analogously, the substitution rate (R ϭ CF3CH2) is 1.48, and that for R ϭ Me is 1.24. The
constants of aquacobalamin by a variety of substituted second-order rate constant in the case of R ϭ Me is higher
pyridine ligands were found to be independent of their than for R ϭ CF3CH2, which is due to the stronger labiliz-
pKBHϩ values.[25]
ation by R ϭ Me.
The second-order rate constants for substitution of H2O
The second-order rate constants for the substitution of
in trans-(CF3CH2)Co(Hdmg)2(H2O) by Py and 4-CNPy are H2O in trans-RCo(Hdmg)2(H2O) by pyridine were found to
656
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 653Ϫ659

Axial Ligand Substitution of Alkyl Cobaloximes
Table 4. Kinetic data for the reaction of trans-RCo(Hdmg)2S with 4-NH2Py, Py, and 4-CNPy as a function of pressure
FULL PAPER
Pressure, MPa
kobs, sϪ1
4-NH2Py
Py
4-CNPy
R ϭ PhCH2, S ϭ MeOH[a]
10
50
90
2.28 0.07
1.94 0.05
1.63 0.04
1.37 0.04
ϩ9.8 0.1
0.98 0.03
0.83 0.01
0.68 0.01
0.58 0.01
ϩ10.3 0.3
1.28 0.03
1.01 0.05
0.81 0.01
0.66 0.06
ϩ13.0 0.3
130
∆V϶, cm3 molϪ1
R ϭ CF3CH2, S ϭ MeOH[b]
10
50
90
0.150 0.006
0.136 0.007
0.119 0.002
0.107 0.003
ϩ7.5 0.3
0.107 0.001
0.093 0.002
0.082 0.001
0.072 0.001
ϩ8.6 0.1
0.116 0.006
0.100 0.006
0.087 0.001
0.075 0.004
ϩ9.4 0.1
130
∆V϶, cm3 molϪ1
R ϭ CF3CH2, S ϭ H2O[c]
10
50
90
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
0.135 0.002
0.127 0.003
0.115 0.002
0.104 0.002
ϩ5.7 0.4
0.114 0.003
0.101 0.004
0.091 0.002
0.081 0.002
ϩ7.4 0.1
130
∆V϶, cm3 molϪ1
[a]
[b]
[PhCH2Co(Hdmg)2MeOH] ϭ (4Ϫ7) ϫ 10Ϫ4 , {[4-NH2Py], [Py] ϭ 0.04  and at 5.0 °C}, {[4-CNPy] ϭ 0.0225  and at 10.0 °C}.
[CF3CH2Co(Hdmg)2MeOH] ϭ 2 ϫ 10Ϫ3, [NH2Py] ϭ 0.08 , {[Py], [4-CNPy] ϭ 0.16 }, and at 40.0 °C. [c] [CF3CH2Co(Hdmg)2H2O] ϭ
2 ϫ 10Ϫ3, [Py] and [4-CNPy] ϭ 0.08 , pH ϭ 8, at 40.0 °C and I ϭ 0.1  NaClO4.
to be 1.12 ϫ 10Ϫ1, 4.79 ϫ 10Ϫ2, and 4.16 ϫ 10Ϫ3 Ϫ1 sϪ1
for S ϭ EtOH, MeOH, and H2O, respectively.[37]
Ligand-substitution reactions of trans-RCo(Hdmg)2H2O,
for R ϭ PhCH2 by Py and 4-CNPy in methanol, and also
for R ϭ CF3CH2 by 4-CNPy in H2O and MeOH, are
characterised by a linear dependence of the observed
pseudo first-order rate constant on the concentration of the
entering nucleophile as shown in Figures 3 and S3. In terms
of the suggested dissociative interchange mechanism, this
means that precursor formation is weak and does not show
up in the kinetic data even at high concentrations of the
entering ligand. The curvature observed in the nucleophile
concentration dependence of kobs for the substitution reac-
tions of trans-RCo(Hdmg)2(H2O) (R ϭ PhCH2 and
CF3CH2) by 4-NH2Py and Py suggests that an interchange
mechanism involving a rapid pre-association of the complex
with the incoming ligand is operating. The precursor-for-
mation constant decreases when the electron donor ability
of the alkyl group in the trans position is increased. It was
found that the precursor-formation constants for the substi-
tution of MeOH by 4-NH2Py are 0.35 and 0.54 Ϫ1 at 25.0
°C for R ϭ PhCH2 and CF3CH2, respectively. It is also
apparent that the precursor-formation constant depends on
the nature of the nucleophile. The π-electron density on the
Py ring increases in the order 4-CNPy Ͻ Py Ͻ 4-NH2Py,[38]
which is also the order of increasing tendency for precur-
sor formation.
Figure 4. Plots of ln(k) versus pressure for the reaction between
trans-(CF3CH2)Co(Hdmg)2(MeOH) and L, where L ϭ Py (A), 4-
CNPy (B) and 4-NH2Py (C). Experimental conditions: [CoIII] ϭ 2
ϫ 10Ϫ3 , [NH2Py] ϭ 0.08 , [Py] and [4-CNPy] ϭ 0.16 , at
40.0 °C
be 1.49 ϫ 104 (isopropyl), 1.17 ϫ 103 (ethyl), 1.16 ϫ 102
(methyl), 1.22 ϫ 103 (phenylethyl), 1.64 ϫ 102 (3-cyanopro-
pyl), 1.65 ϫ 102 (2-methoxyethyl), 2.34 ϫ 101 (1-propenyl),
4.2 (iodomethyl), 3.17 ϫ 10Ϫ1 (2,2,2-trifluoroethyl), 1.06 ϫ
10Ϫ1 (chloromethyl), and 5.59 ϫ 10Ϫ2 (cyanomethyl) Ϫ1
sϪ1 at 25 °C, respectively.[36] The second-order rate con-
stants for substitution of methanol by pyridine in the com-
It is generally known from the literature that a system
plex trans-(RSO2)Co(Hdmg)2(MeOH) were found to be must be sufficiently electron-rich in order that bonding
4.79 0.01 and 3.56 0.11 Ϫ1 sϪ1 for R ϭ p-tolyl and through the π-electrons of pyridine can occur.[39] This π-
methyl, respectively. Substitution of the solvent by pyridine interaction results in the effective binding of the nucleophile
in the complex (CH3C6H4SO2)Co(Hdmg)2S does have an close to the metal centre during precursor formation, which
effect on the second-order rate constants, which were found is followed by an interchange of ligands controlled by
Eur. J. Inorg. Chem. 2004, 653Ϫ659
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
657

B. M. Alzoubi, M. S. A. Hamza, A. Felluga, L. Randaccio, G. Tauzher, R. van Eldik
FULL PAPER
breakage of the CoϪOH2 bond by an Id mechanism.[25] The the complex trans-(PhCH2)Co(Hdmg)2(H2O) were found to
pre-equilibrium constants for the reaction
of be ϩ9.8, ϩ10.3 and ϩ13 cm3 molϪ1, respectively. This is
[MeCo{(DO)(DOH)pn}H2O]ϩ with L ϭ imidazole, ben- consistent with an enhanced dissociative (limiting D)
zylamine, Py and 4-MePy at 25 °C were found to be 0.26, character in the mechanism on going from the CF3CH2 de-
0.12, 1.00, and 1.81 Ϫ1, respectively.[15] The pre-equilib- rivative to the PhCH2 derivative. Interestingly, ∆V϶ for sub-
rium constants for the reaction of aquacobalamin with Py, stitution of H2O by Py and 4-CNPy (R ϭ CH2CF3 in H2O)
4-MePy, and 2,4-DMePy were found to be 1.1, 2.5, and 4.9 were found to be ϩ5.7 and ϩ7.4 cm3 molϪ1, respectively.

Ϫ1, respectively.[25]
In these cases the small positive values of ∆V϶ suggest that
By way of comparison, the lability of trans- the mechanism is of an Id nature. Thus the substitution
RRh(Hdmg)2(H2O) is affected by the steric and electronic mechanism changes from limiting D to Id along the series
effects of the alkyl group R, and in all cases the reaction from R ϭ PhCH2 to R ϭ CF3CH2, on decreasing the ba-
rate is almost independent of the nucleophilicity of the in- sicity of the nucleophiles, and on changing the solvent from
coming ligand.[40,41] There is a definite trend in the ∆V϶ MeOH to H2O. This mechanistic changeover is clearly seen
data: the values are significantly positive for the faster sub- in the size of ∆V϶, and can be directly correlated with the
stitution reactions (R ϭ CH3) but significantly negative for donor properties of R, which in turn control the magnitude
the slower substitution reactions (R ϭ CF3CH2).[42] This of the rate constants. The greater donating properties of
trend correlates well with the trans-influence of the R PhCH2 result in a limiting dissociative process as charac-
groups in the ground state, as observed for a series of pyri- terized by the significantly positive ∆V϶. Similarly, the
dine (py) complexes: the RhϪPy bond length decreases in weaker donor properties of CF3CH2 favour more a dissoci-
the order CH3 [2.220(3) A] Ͼ ClCH2 [2.178(3) A] Ͼ ative interchange mechanism, as shown by the ∆V϶ data.
CF3CH2 [2.145(3) A].
˚
˚
[40,41]
˚
A similar trend is expected for The activation parameters for substitution of aquacobala-
the RhϪO bond in the corresponding aqua complexes: fas- min by thiourea, substituted thiourea and pyridine for the
ter reactions are thus associated with the stronger trans in- forward and reverse reactions conclusively confirm that the
fluence of the methyl group, and slower reactions with the mechanism of the substitution is a dissociative interchange.
weaker trans influence of ClCH2 and CF3CH2. Further- For instance, the volumes of activation for both the forward
more, the nature of the R group directly determines the and reversible reactions have values between ϩ6 and ϩ10
substitution mechanism. The strong donor properties of cm3 molϪ1 [16,17]
CH3 result in a dissociative substitution process (Id), We conclude that all the available data underline the op-
.
characterised by the significantly positive ∆V϶ value, eration of a limiting dissociative (D) mechanism for substi-
whereas the weaker donor properties of CF3CH2 favour a tution of methanol by 4-CNPy when R ϭ PhCH2. How-
more associative substitution mechanism (Ia), as indicated ever, when R ϭ CF3CH2, the mechanism is Id for substi-
by the significantly negative value of ∆V϶.[42] Thus, in the tution of the solvent by 4-CNPy in methanol or H2O. The
transition state the degree of bond breakage (RhϪOH2) non-linear concentration dependence must therefore be due
and bond formation (RhϪL) is controlled by the donor to the operation of an Id mechanism, and involve the for-
properties of R.
mation of a precursor complex. The strong precursor for-
The ∆H϶ values for substitution reactions of trans- mation is ascribed to an interaction of the π system of the
RCo(Hdmg)2(H2O) (for R ϭ PhCH2 and CF3CH2) by pyri- pyridine ligand with the coordination sphere of the com-
dine derivatives in H2O and methanol are relatively inde- plex. This π interaction results in the effective binding of
pendent of the nature of the incoming ligand. It is expected the entering nucleophile close to the metal centre during
that the reactions involve a dissociative activation process, precursor formation, which is followed by an interchange
since the ligands are loosely bound in the transition state. of ligands controlled by the breakage of the CoϪsolvent
The activation enthalpy for the substitution reaction with bond by an Id mechanism.
R ϭ CF3CH2 is higher than for R ϭ PhCH2 due to the
strong CoϪOH2 bond in the ground state. Hence, ∆H϶ val-
ues appear to be affected by both the electron-withdrawing
Experimental Section
and steric effects of the R group. The influence of these
factors on ∆S϶, however, cannot be rationalised. The en-
Materials: All chemicals were of p. a. grade and used as received.
tropy values should be slightly more positive since the tran-
sition state should be less ordered than the initial state, ow-
ing to the relief of steric strain. The organocobaloximes are
about 105 times more labile than the corresponding nitro
and iodo-aqua complexes. This very high rate enhancement
is due to both the decrease in the activation enthalpy and
the increase in the activation entropy.[43]
4-Aminopyridine and 4-cyanopyridine were purchased from Acros,
tricine buffer from Sigma and pyridine from Baker. Ultra pure
water was used in the kinetic measurements. trans-
RCo(Hdmg)2H2O (R ϭ CF3CH2, PhCH2) were prepared as pre-
viously reported.[44] Because alkyl cobaloximes are light sensitive,
their solutions were prepared and handled away from light.
Instrumentation: The pH of the solution was measured using a
Mettler Delta 350 pH meter. The pH meter was calibrated with
standard buffer solutions at pH 4 and 7. UV/Vis spectra were re-
corded on Shimadzu UV-2101 and Hewlett Packard 8452A spec-
The values of ∆V϶ obtained for R ϭ CF3CH2 in meth-
anol were found to be ϩ7.5, ϩ8.6, and ϩ9.4 cm3 molϪ1
,
respectively. The values of ∆V϶ for substitution of the sol-
vent by pyridine derivatives (4-NH2Py, Py, and 4-CNPy) in trophotometers.
658  2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.eurjic.org
Eur. J. Inorg. Chem. 2004, 653Ϫ659

Axial Ligand Substitution of Alkyl Cobaloximes
FULL PAPER
[18]
[19]
[20]
[21]
[22]
[23]
B. M. Alzoubi, M. S. A. Hamza, C. Dücker-Benfer, R. van
Eldik, Eur. J. Inorg. Chem. 2002, 968.
M. S. A. Hamza, C. Dücker-Benfer, R. van Eldik, Inorg. Chem.
2000, 39, 3777.
G. Stochel, R. van Eldik, H. Kunkely, A. Vogler, Inorg. Chem.
1989, 28, 4314.
A. G. Gergan, N. E. Brasch, R. van Eldik, Inorg. Chem. 2001,
40, 1430.
F. Nome, J. H. Fendler, J. Chem. Soc., Dalton Trans. 1976,
1212.
H. M. Marques, J. C. Bradley, L. A. Campbell, J. Chem. Soc.,
Dalton Trans. 1992, 2019.
H. M. Marques, J. Chem. Soc., Dalton Trans. 1991, 339.
F. F. Prinsloo, M. Meier, R. van Eldik, Inorg. Chem. 1994,
33, 900.
W. W. Reenstra, W. P. Jencks, J. Am. Chem. Soc. 1979, 101,
5780.
F. F. Prinsloo, E. L. J. Breet, R. van Eldik, J. Chem. Soc., Dal-
ton Trans. 1995, 685.
M. S. A. Hamza, J. Chem. Soc., Dalton Trans. 2002, 2831.
Kinetic measurements were carried out on an Applied Photophys-
ics SX 18MV stopped-flow instrument coupled to an online data
acquisition system. At least eight kinetic runs were recorded under
all conditions, and the reported rate constants represent the mean
values. All kinetic measurements were carried out under pseudo-
first order conditions, i.e. the entering nucleophile concentration
was in at least a tenfold excess. Measurements at high pressure were
carried out using a home-made high pressure stopped-flow unit.[45]
Kinetic data were analyzed with the OLIS KINFIT program. All
instruments used were thermostatted to the desired temperature
(
0.1 °C).
[24]
[25]
[26]
[27]
[28]
Acknowledgments
The authors gratefully acknowledge financial support from CNR
(Rome), MIUR (Rome), PRIN No. MMO3185591, the Deutsche
Forschungsgemeinschaft and the Max-Buchner Research Foun-
dation (to R. v. E.).
[29] [29a] N. E. Brasch, M. S. A. Hamza, R. van Eldik, Inorg. Chem.
[29b]
1997, 36, 3216.
M. S. A. Hamza, A. G. Cregan, N. E.
Brasch, R. van Eldik, J. Chem. Soc. 2003, 596.
B. M. Alzoubi, M. S. A. Hamza, A. Felluga, L. Randaccio, G.
Tauzher, R. van Eldik, Eur. J. Inorg. Chem. 2003, 3738Ϫ3744.
D. A. Baldwin, E. A. Betterton, J. M. Pratt, S. Afr. J. Chem.
1982, 35, 173.
A. Albert, E. P. Serjeant, The Determination of Ionization Con-
stants, Chapman and Hall, London, 1984.
P. Moore, R. G. Wilkins, J. Chem. Soc. 1964, 3454.
M. S. A. Hamza, A. Felluga, L. Randaccio, G. Tauzher, R. van
Eldik, Dalton Trans. submitted for publication.
K. L. Brown, D. Chernoff, D. J. Keljo, R. G. Kallen, J. Am.
Chem. Soc. 1972, 94, 6697.
K. L. Brown, D. Lyles, M. Pencovici, R. G. Kallen, J. Am.
Chem. Soc. 1975, 97, 7338.
[1]
[30]
[31]
[32]
N. Bresciani-Pahor, M. Forcolin, L. G. Marzilli, L. Randaccio,
M. F. Summers, P. J. Toscano, Coord. Chem. Rev. 1985, 63, 1.
[2]
A. L. Crumbliss, W. K. Wilmarth, J. Am. Chem. Soc. 1970,
92, 2593.
[3]
K. L. Brown, S. Satyanarayana, J. Am. Chem. Soc. 1992, 114,
5674.
K. L. Brown, R. G. Kallen, J. Am. Chem. Soc. 1972, 94, 1894.
L. G. Marzilli, P. J. Toscano, L. Randaccio, N. Bresciani-Pahor,
M. Calligaris, J. Am. Chem. Soc. 1979, 101, 6754.
L. Randaccio, N. Bresciani-Pahor, P. J. Toscano, L. G. Marzilli,
J. Am. Chem. Soc. 1980, 102, 7372.
N. Bresciani-Pahor, L. Randaccio, P. J. Toscano, A. San-
dercock, L. G. Marzilli, J. Chem. Soc., Dalton Trans. 1982, 129.
M. F. Summers, P. J. Toscano, N. Bresciani-Pahor, G. Nardin,
L. Randaccio, L. G. Marzilli, J. Am. Chem. Soc. 1983, 105,
6259.
N. Bresciani-Pahor, L. G. Marzilli, L. Randaccio, P. J. Toscano,
E. Zangrando, J. Chem. Soc., Chem. Commun. 1984, 1508.
G. Costa, G. Mestroni, E. de Savorgnani, Inorg. Chim. Acta
[4]
[33]
[34]
[5]
[6]
[35]
[36]
[7]
[8]
[37]
[38]
J. M. Palmer, E. Deutsch, Inorg. Chem. 1975, 14, 17.
E. S. Gould, Mechanism and Structure in Organic Chemistry,
Holt International, London, 1970.
[9]
[39]
[40]
[41]
K. Schofield, Hetero-Aromatic Nitrogen Compounds: Pyrroles
and Pyridines, Butterworths, London, 1967, 2213.
N. B. Pahor, R. Dreos-Garlatti, S. Geremia, L. Randaccio, G.
Tauzher, E. Zangrando, Inorg. Chem. 1990, 29, 3437.
L. Randaccio, S. Geremia, R. Dreos-Garlatti, G. Tauzher, F.
Asano, G. Pellizer, Inorg. Chim. Acta 1992, 194, 1.
[10]
1969, 3, 323.
[11]
G. Costa, G. Mestroni, L. Stefani, J. Organomet. Chem. 1967,
7, 493.
[12]
[42] [42a]
G. Costa, G. Mestroni, G. Tauzher, L. Stefani, J. Organomet.
Chem. 1966, 6, 181.
M. S. A. Hamza, X. Zou, K. L. Brown, R. van Eldik, Inorg.
Chem. 2001, 40, 5440.
M. S. A. Hamza, X. Zou, K. L. Brown, R. van Eldik, J. Chem.
Soc., Dalton Trans. 2002, 3832.
G. Tauzher, R. Dreos, G. Costa, M. Green, J. Chem. Soc.,
R. van Eldik, A. E. Merbach, Comments, Inorg. Chem.
[42b]
1992, 12, 341.
Angew. Chem. Int. Ed. Engl. 1995, 34, 2245.
C. Dücker-Benfer, R. Dreos, R. van Eldik,
[13]
[43]
[44]
[45]
R. Dreos-Garlatti, G. Tauzher, G. Costa, M. Green, Inorg.
Chim. Acta 1981, 50, 95.
G. N. Schrauzer, R. J. Windgassen, J. Am. Chem. Soc. 1966,
88, 3738.
R. van Eldik, W. Gaede, S. Wieland, J. Kraft, M. Spitzer, D.
A. Palmer, Rev. Sci. Instrum. 1993, 64, 1355.
Received October 14, 2003
[14]
[15]
Chem. Commun. 1973, 413.
[16]
G. Stochel, R. van Eldik, Inorg. Chem. 1990, 29, 2075.
[17]
M. Meier, R. van Eldik, Inorg. Chem. 1993, 32, 2635.
Eur. J. Inorg. Chem. 2004, 653Ϫ659
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
659