Mechanistic elucidation of the substitution behavior of alkyl cobaloximes in water and methanol as solvents

Article abstract of DOI:10.1021/ic049761j

The ligand substitution behavior of trans-[Co(Hdmg)₂(R)S] (R = CH₃, PhCH₂; Hdmg = dimethylglyoximate; S = H₂O and/or MeOH) was studied for imidazole, pyrazole, 1,2,4-triazole, N-acetylimidazole, 5-chloro-1-methylimidazole, NO₂^-, Ph₃P, Ph₃As, and Ph₃Sb as entering ligands. In all cases, except for Ph₃As and Ph₃Sb, the observed kinetics shows a linear dependence on the entering nucleophile concentration with no evidence for a back reaction. In the case of Ph₃As and Ph₃Sb as entering nucleophiles, kinetic evidence for a reverse solvolysis reaction is at hand. Activation parameters (ΔH?, ΔS?, and ΔV?) were determined from the temperature and pressure dependence of all studied reactions and support the operation of a dissociatively activated substitution mechanism. The rate and activation parameters show that there is an increase in the dissociative character from a dissociative interchange to a limiting dissociative mechanism that depends on the nature of R and the entering nucleophile. The crystal structure of trans-[Co(en)₂(Me)H₂O]²⁺ was determined by X-ray analysis. The Co-O and Co-C bond lengths were found to be 2.153(6) and 1.995(10) A, respectively. The kinetic and structural data are discussed in reference to a series of earlier studied systems for which data are reported in the literature.

Full text of DOI:10.1021/ic049761j

Inorg. Chem. 2004, 43, 6093−6100
Mechanistic Elucidation of the Substitution Behavior of Alkyl
Cobaloximes in Water and Methanol as Solvents
Basam M. Alzoubi, Gu1nter Liehr, and Rudi van Eldik*
Institute for Inorganic Chemistry, UniVersity of ErlangensNu¨rnberg, Egerlandstr. 1,
91058 Erlangen, Germany
Received February 23, 2004
The ligand substitution behavior of trans-[Co(Hdmg)2(R)S] (R ) CH , PhCH ; Hdmg ) dimethylglyoximate; S )
3
2
H O and/or MeOH) was studied for imidazole, pyrazole, 1,2,4-triazole, N-acetylimidazole, 5-chloro-1-methylimidazole,
2
NO -, Ph3P, Ph3As, and Ph3Sb as entering ligands. In all cases, except for Ph3As and Ph3Sb, the observed
2
kinetics shows a linear dependence on the entering nucleophile concentration with no evidence for a back reaction.
In the case of Ph3As and Ph3Sb as entering nucleophiles, kinetic evidence for a reverse solvolysis reaction is at
q
q
q
hand. Activation parameters (∆H , ∆S , and ∆V ) were determined from the temperature and pressure dependence
of all studied reactions and support the operation of a dissociatively activated substitution mechanism. The rate
and activation parameters show that there is an increase in the dissociative character from a dissociative interchange
to a limiting dissociative mechanism that depends on the nature of R and the entering nucleophile. The crystal
structure of trans-[Co(en)2(Me)H O]2+ was determined by X-ray analysis. The Co−O and Co−C bond lengths were
2
found to be 2.153(6) and 1.995(10) Å, respectively. The kinetic and structural data are discussed in reference to
a series of earlier studied systems for which data are reported in the literature.
Introduction
derivatives, especially aquacobalamin, was investigated as
a reference system in which the cobalt-carbon bond is
substituted by a cobalt-water bond that has no labilization
effect on the Co(III) center. Work in our laboratories has
focused on the substitution behavior of coenzyme B12 and
modified alkylcobalamins with R ) Et, Pr, Me, CF3CH2,
NCCH2, CF3, and CN,1 trans-[Co(en)2(Me)H2O]2+,2 trans-
[Co(en)2(Me)NH3]2+,2b trans-[Co(BQDI)2(PPh3)S]+ (BQDI
) o-benzosemiquinonediimine; S ) MeOH, MeCN),3 [Co-
(LNHPy)(HLNHPy)(R)]+ (HLNHPy ) 2-(2-pyridylethyl)-
amino-3-butanone oxime; R ) Et, Me, CF3CH2, bridging-
CH2),4 and trans-[Co(Hdmg)2(R)S] (Hdmg ) dimethylgly-
oximate; R ) cyclo-C5H9, Et, Me, Ph, PhCH2, CF3CH2; S
) H2O, MeOH).5,6 In these studies, detailed kinetic inves-
Over the past few years, we have developed an active
interest in the systematic tuning of the substitution behavior
of alkylcobalt(III) complexes for various reasons. First of
all, such reactions are relevant to the mechanistic behavior
of coenzyme B12, a bio-organometallic cofactor with a
cobalt-carbon bond. Second, the introduction of a single
cobalt-carbon bond in Co(III) complexes has a profound
influence on their substitution lability. Third, such complexes
allow a systematic adjustment of their reactivity via steric
hindrance and electron donor/acceptor effects of spectator
ligands and the coordinated alkyl or aryl group. Such studies
are of fundamental interest in the elucidation of inorganic/
bioinorganic reaction mechanisms and serve as model
systems for tuning the lability of inert metal complexes.
In this respect, it is interesting to note that many investiga-
tions in the past have focused on the mechanistic behavior
of coenzyme B12 itself, or on modified forms of the
coenzyme, or on closely related model systems with a quite
different chelate arrangement around the Co(III) center. By
way of comparison, the substitution behavior of vitamin B12
(1) Hamza, M. S. A.; van Eldik, R. Dalton Trans. 2004, 1.
(2) (a) Hamza, M. S. A.; Du¨cker-Benfer, C.; van Eldik, R. Inorg. Chem.
2000, 39, 3777. (b) Alzoubi, B. M.; Hamza, M. S. A.; Du¨cker-Benfer,
C.; van Eldik, R. Eur. J. Inorg. Chem. 2002, 968.
(3) Alzoubi, B. M.; Hamza, M. S. A.; Du¨cker-Benfer, C.; van Eldik, R.
Eur. J. Inorg. Chem. 2003, 2972.
(4) Alzoubi, B. M.; Hamza, M. S. A.; Felluga, A.; Randaccio, L.; Tauzher,
G.; van Eldik, R. Eur. J. Inorg. Chem. 2003, 3738.
(5) Hamza, M. S. A.; Felluga, A.; Randaccio, L.; Tauzher, G.; van Eldik,
R. Dalton Trans. 2004, 287.
* To whom correspondence should be addressed. E-mail: vaneldik@
chemie.uni-erlangen.de.
(6) Alzoubi, B. M.; Hamza, M. S. A.; Felluga, A.; Randaccio, L.; Tauzher,
G.; van Eldik, R. Eur. J. Inorg. Chem. 2004, 653.
10.1021/ic049761j CCC: $27.50 © 2004 American Chemical Society
Published on Web 08/21/2004
Inorganic Chemistry, Vol. 43, No. 19, 2004 6093

Alzoubi et al.
tigations enabled the determination of rate and activation
parameters, on which basis detailed mechanistic assignments
could be made. For instance, for the reaction of cyanide with
different alkylcobalamins it was found that the nature of the
alkyl group has a marked influence on the thermodynamic
stability constants and the kinetics and mechanism of the
substitution reactions of the axial ligand trans to the alkyl
group.1,7,8
The Co-C bond length depends on three contributing
factors. First, there are cis steric interactions, distortions in
the corrin macrocycle or equatorial ligand toward the axial
trans alkyl ligand resulting in Co-alkyl bond lengthening.
Second, a trans steric influence occurs when the bulkiness
of an axial ligand induces an elongation or weakening of
the trans metal-ligand bond. Third, the trans electronic
influence is associated with how the change in basicity of
the axial ligand affects the Co-alkyl bond length and/or bond
dissociation energy (trans effect) and how changes in the
σ-donor character of the alkyl group affect the Co-N(ax)
bond distance.9
In the present study, we have extended the work on trans-
[Co(Hdmg)2(R)S] to R ) Me and PhCH2 (S ) H2O and/or
MeOH) and to a wide range of different entering nucleophiles
(viz. imidazole, pyrazole, 1,2,4-triazole, N-acetylimidazole,
-
5-chloro-1-methylimidazole, NO2 , Ph3P, Ph3As, and Ph3-
Sb) in order to determine the role of the nucleophilicity of
the entering ligand and the nature of R on the reactivity of
the Co(III) center. These data allow us to present an overall
comparison of steric and electronic effects that control the
reactivity of alkyl and aryl Co(III) complexes, in reference
to a series of earlier studied systems for which data are
reported in the literature.
Experimental Section
Materials. All chemicals were of p.a. grade and used as received.
Imidazole, pyrazole, 1,2,4-triazole, N-acetylimidazole, and meth-
ylhydrazine were purchased from Fluka, 5-chloro-1-methylimida-
zole from Acros, TAPS buffer from Sigma, and Ph3Sb from Strem.
Triphenylphosphine, cobalt(II) nitrate hexahydrate, and sodium
nitrite were purchased from Merck, and Ph3As from Aldrich.
Ultrapure water was used in the kinetic measurements. trans-[Co-
(Hdmg)2(R)H2O] (R ) CH3, PhCH2) and trans-[Co(en)2(Me)H2O]-
S2O6 were prepared as previously reported and were characterized
by elemental analysis and UV-vis and NMR spectroscopy; the
results were in agreement with literature data.18,19 Preparation of
solutions and all measurements were carried out in the dark because
the investigated complexes are light sensitive.
It is well-known that high-pressure techniques can assist
the elucidation of inorganic and bioinorganic reaction mech-
anisms through the determination of activation volumes
obtained from the pressure dependence of the rate constant
and the construction of reaction volume profiles.10,11 In the
case of Co(III) complexes, activation volume data for
complex-formation reactions of trans-[Co(en)2(Me)H2O]2+
with CN-, imidazole,2a and NH32b are similar to data for the
reactions of aquacobalamin with different ligands such as
All the reactions in aqueous solution were studied at pH 9.0 by
using TAPS buffer. The pH of 9.0 was selected to prevent
protonation of the nitrogen donor nucleophiles and formation of
the corresponding hydroxo complexes.
-
N-acetylimidazole,12 HN3, N3 , pyridine and its derivatives,
X-ray Structure Determination. Intensity data were collected
on a Nonius CAD4 Mach3 diffractometer with graphite monochro-
mated Mo KR radiation in the ω/2θ scan mode at room temperature.
The structure of trans-[Co(en)2(Me)H2O]S2O6 was solved by direct
methods using SIR-9720 and refined by full-matrix least-squares
on F2 using SHELXL-97.21 Complete crystallographic details, bond
lengths, bond angles anisotropic temperature factors, and hydrogen
atom coordinates are available as a CIF file as Supporting
Information. Some crucial bond lengths and bond angles are
summarized in Table 1.
and thiourea and its derivatives.13-15 In all these cases, ∆Vq
was found to be in the range between +4 and +10 cm3
mol-1, an indication that the reactions follow an Id mecha-
nism. These values are significantly smaller than those
reported for substitution reactions of cobalt(III) porphyrin
systems, where it was found that ∆Vq values for ligand
substitution on [Co(TMPP)(H2O)2]5+ and [Co(TPPS)(H2O)2]3-,
TMPP ) meso-tetrakis(4-N-methylpyridyl)porphine and
TPPS ) meso-tetrakis(p-sulfonatophenyl)porphine, are +14.4
and +15.4 cm3 mol-1, respectively, and these were accord-
ingly assigned to a limiting D mechanism.16,17
Crystal data: CoC5H21N4S2O7, fw ) 372.3 g mol-1, a ) 19.913-
(10) Å, b ) 12.094(10) Å, c ) 11.754(10) Å, R ) â ) γ ) 90°,
V ) 2831(4) Å3, orthorhombic, space group ) Iba2 (No. 45), Z )
8, Dc ) 1.738 g cm-3, µ(Mo KR) ) 1.5 mm-1, T ) 298 K, R1 )
(7) Hamza, M. S. A.; Zou, X.; Brown, K. L.; van Eldik, R. Inorg. Chem.,
2001, 40, 5440.
(8) Hamza, M. S. A.; Zou, X.; Brown, K. L.; van Eldik, R. J. Chem.
Soc., Dalton Trans. 2002, 3832.
(9) Hansen, L. M.; Kumar, P. N. V. P.; Marynick, D. S. Inorg. Chem.
1994, 33, 728.
22
0.066 for the data Fo > 2σ(Fo), (Rint ) 0.050 for 2335 unique
reflections). The minimum and maximum electron densities on the
final difference Fourier map were -0.51 and 0.89 e Å-3, respec-
tively.
(10) (a) Inorganic High-Pressure Chemistry. Kinetics and mechanisms; van
Eldik, R., Ed.; Elsevier Biomedical: Amsterdam, Netherlands, 1986.
(b) Drljaca, A.; Hubbard, C. D.; van Eldik, R.; Asano, T.; Basilevsky,
M. V.; le Noble, W. J. Chem. ReV. 1998, 98, 2167. (c) High-Pressure
Chemistry: Synthetic, Mechanistic, and Supercritical Applications;
van Eldik, R.; Kla¨rner, F.-G., Eds.; Wiley-VCH: Weinheim, Germany,
2002.
(11) (a) De Vito, D.; Weber, J.; Merbach, A. E. Inorg. Chem. 2004, 43,
858. (b) Grundler, P. V.; Salignac, B.; Cayemittes, S.; Alberto, R.;
Merbach, A. E. Inorg. Chem, 2004, 43, 865.
(12) Hamza, M. S. A. J. Chem. Soc., Dalton Trans., 2002, 2831.
(13) Meier, M.; van Eldik, R. Inorg. Chem. 1993, 32, 2635.
(14) Prinsloo, F. F.; Meier, M.; van Eldik, R. Inorg. Chem. 1994, 33, 900.
(15) Prinsloo, F. F.; Breet, E. L. J.; van Eldik, R. J. Chem. Soc., Dalton
Trans. 1995, 685.
(16) Funahashi, S.; Inamo, M.; Ishihara, K.; Tanaka, M. Inorg. Chem. 1982,
21, 447.
Instrumentation. The pH of the solution was measured using a
Mettler Delta 350 pH meter. The pH meter was calibrated with
(17) Leipoldt, J. G.; van Eldik, R.; Kelm, H. Inorg. Chem. 1983, 22, 4146.
(18) Schrauzer, G. N.; Windgassen, R. J. J. Am. Chem. Soc. 1966, 88, 3738.
(19) Kofod, P.; Harris, P.; Larsen, S. Inorg. Chem. 1997, 36, 2258.
(20) Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.; Burla,
M. C.; Polidori, G.; Camalli, M. SIR-92-Program Package for Solving
Crystal Structures by Direct Methods. J. Appl. Crystallogr. 1994, 27,
435.
(21) Sheldrick, G. M. SHELXL-97; University of Go¨ttingen: Go¨ttingen,
Germany, 1997. (a) Sheldrick, G. M. Acta Crystallogr., Sect. A 1990,
A46, 467-473. (b) Sheldrick, G. M. Acta Crystallogr., Sect. D 1993,
D49, 18-23. (c) Sheldrick, G. M.; Schneider, T. R. Methods Enzymol.
1997, 227, 319-343.
2
2
2
(22) R1 ) ∑(Fo - Fc)/∑Fo; wR2 ) [(∑w(Fo - Fc )2/∑w(Fo )2]1/2
.
6094 Inorganic Chemistry, Vol. 43, No. 19, 2004

Alkyl Cobaloximes in Water and Methanol
Table 1. Bond Lengths (Å) and Bond Angles (deg) with Standard
Deviations for the Complex trans-[Co(en)2(Me)H2O]S2O6
Table 2. UV-Vis Spectral Data for trans-[Co(Hdmg)2(R)(L)] in H2O
and MeOH at Room Temperature
L
λ, nm (10-3ꢀ, M-1 cm-1
)
λ f (nm)
bond distance (Å)
bond angle (deg)
H2Ob
278sh a (4.94), 384 (1.43),
and 440 (1.22)
Co-O1
2.153(6)
1.999(7)
1.995(8)
1.931(7)
1.927(7)
1.995(10)
1.453(12)
1.437(12)
1.553(11)
1.480(12)
1.512(14)
O1-Co-N1
88.2(3)
89.5(3)
90.2(3)
87.8(3)
177.6(3)
84.2(3)
176.3(3)
96.0(3)
89.4(3)
92.5(3)
177.2(3)
90.6(4)
87.3(3)
92.2(3)
92.1(3)
108.4(5)
109.1(6)
Co-N1
Co-N2
Co-N3
Co-N4
Co-C5
N1-C1
N2-C2
N3-C3
N4-C4
C1-C2
O1-Co-N2
O1-Co-N3
O1-Co-N4
O1-Co-C5
N1-Co-N2
N1-Co-N3
N1-Co-N4
N1-Co-C5
N2-Co-N3
N2-Co-N4
N2-Co-C5
N3-Co-N4
N3-Co-C5
N4-Co-C5
Co-N1-C1
Co-N2-C2
imidazoleb
pyrazoleb
282 (5.48), 372 (2.14),
450
450
450
and 416sh (0.85)
280 (5.84), 374 (1.91),
and 420sh (0.87)
1,2,4-triazoleb
282 (5.74), 372 (2.11),
and 416sh (0.78)
N-acetylimidazoleb
372 (2.19) and 420sh (0.82)
358 (3.37) and 414sh (1.06)
422sh(1.04) and 464sh (0.25)
284sh (6.45), 384 (1.70),
and 444 (1.64)
450
450
445
5-chloro-1-methylimidazoleb
- b
NO2
MeOHc
Ph3Pc
370sh (4.15) and 420sh (1.23)
288 (7.08), 372 (2.68),
and 420sh (0.93)
445
445
imidazolec
Ph3Asc,d
Ph3Sbc,d
MeOHe
308sh, 378sh, and 444
316sh, 372sh, and 448
274sh (12.95), 358 (6.64),
and 470 (1.06)
332
335
standard buffer solutions at pH 4.0, 7.0 and 10.0. UV-vis spectra
were recorded on Shimadzu UV-2101 and Hewlett-Packard 8452A
spectrophotometers.
Ph3Pe
360sh (14.43)
410
a sh ) shoulder. b trans-[Co(Hdmg)2(Me)(L)] in H2O (pH 9.0, I ) 0.1
M NaClO4 and 0.5 M for NO2- as nucleophile). c trans-[Co(Hdmg)2(Me)(L)]
in MeOH. d No extinction coefficients are specified since only partial
formation of the corresponding complex could be reached due to the
unfavorable equilibrium position (see ratio of rate constants for the forward
and reverse reactions given in Table 3). e trans-[Co(Hdmg)2(PhCH2)(L)]
in MeOH. f Wavelength at which kinetic data were recorded.
Kinetic measurements were carried out on an Applied Photo-
physics 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 ligand concentration was in
at least a 10-fold excess. Measurements at high pressure were
carried out using a homemade high-pressure stopped-flow unit.23
Kinetic data were analyzed with the OLIS KINFIT program. All
instruments used were thermostated to the desired temperature ((0.1
°C).
Results and Discussion
The kinetics of the ligand substitution reactions (eq 1) of
trans-[Co(Hdmg)2(R)S] (R ) CH3, CH2Ph; S ) H2O,
MeOH) were studied for different nucleophiles L to form
trans-[Co(Hdmg)2(R)L]. The UV-vis absorption spectra of
trans-[Co(Hdmg)2(R)S] and trans-[Co(Hdmg)2(R)L] com-
plexes, where L ) imidazole, pyrazole, 1,2,4-triazole,
-
N-acetylimidazole, 5-chloro-1-methylimidazole, NO2 , Ph3P,
Ph3As, and Ph3Sb, are summarized in Table 2.
ka
Figure 1. Plots of kobs versus [5-chloro-methylimidazole] (A), [pyrazole]
(B), and [N-acetylimidazole] (C) for the reaction with 1 × 10-4 M trans-
[Co(Hdmg)2(Me)H2O] at 25.0 °C, pH 9.0, and I ) 0.1 M NaClO4.
trans-[Co(Hdmg)2(R)S] + L y
\
kz
b
trans-[Co(Hdmg)2(R)L]0/- + S (1)
imidazole, pyrazole, 1,2,4-triazole, N-acetylimidazole, 5-chloro-
Figures 1-3 and S1 (Supporting Information) report plots
of the pseudo-first-order rate constant kobs versus [L] for the
reactions with imidazole, pyrazole, 1,2,4-triazole, N-acetyl-
-
1-methylimidazole, and NO2 , respectively (R ) CH3, S )
H2O); 23.6 and 19.6 M-1 s-1 for L ) imidazole and Ph3P,
respectively (R ) CH3, S ) MeOH); and 252 M-1 s-1 for
L ) Ph3P (R ) PhCH2 and S ) MeOH) at 25 °C.
-
imidazole, 5-chloro-1-methylimidazole, NO2 , and Ph3P at
25 °C. The linear plots have negligible intercepts, indicating
that a reverse or parallel reaction does not contribute
significantly under the selected experimental conditions. This
behavior can be expressed by the rate law given in (eq 2)
with a negligible contribution from kb. The second-order rate
constants for substitution of the solvent (ka) were found to
be 70.3, 90.7, 70.6, 48.2, 105.1, and 113.9 M-1 s-1 for L )
kobs ) ka[L] + kb
(2)
Kinetic data for the reaction of trans-[Co(Hdmg)2(R)-
MeOH] with Ph3As and Ph3Sb are reported in Figures 4 and
5, from which it follows that good linear plots with
significant intercepts are obtained within the experimental
error limits. This behavior can also be expressed by the rate
law given in eq 2, where ka and kb represent the rate constants
(23) van Eldik, R.; Gaede, W.; Wieland, S.; Kraft, J.; Spitzer, M.; Palmer,
D. A. ReV. Sci. Instrum. 1993, 64, 1355.
Inorganic Chemistry, Vol. 43, No. 19, 2004 6095

Alzoubi et al.
Figure 4. Plots of kobs versus [Ph3As] for the reaction with trans-[Co-
(Hdmg)2(Me)MeOH] as function of temperature. Experimental conditions:
[CoIII] ) 1 × 10-4 M and T ) 5.0 (A), 10.0 (B), 15.0 (C), 20.0 (D), and
25.0 (E) °C.
Figure 2. Plots of kobs versus [imidazole] for the reaction with 1 × 10-4
M trans-[Co(Hdmg)2(Me)S] where S ) H2O (A) and MeOH (B) at 25.0
°C, pH 9.0, and I ) 0.1 M NaClO4 for aqueous solution.
Figure 3. Plots of kobs versus [Ph3P] for the reaction with (0.5-1) × 10-4
M trans-[Co(Hdmg)2(R)MeOH] where R ) PhCH2 (A) and CH3 (B) at
25.0 °C.
Figure 5. Plots of kobs versus [Ph3Sb] for the reaction with trans-[Co-
(Hdmg)2(Me)MeOH] as function of temperature. Experimental conditions:
[CoIII] ) 1 × 10-4 M and T ) 5.0 (A), 10.0 (B), 15.0 (C), 20.0 (D), and
25.0 (E) °C.
for the forward and reverse reactions in eq 1, respectively.
Direct evidence for a reversible complex-formation reaction
was obtained from the overall spectral changes observed
during the reaction, which increased significantly with
increasing entering nucleophile concentration. Unfortunately,
the solubility of both Ph3As and Ph3Sb in methanol is too
low to enable a direct spectrophotometric determination of
the overall equilibrium constant. In the most favorable case,
the ratio of ka/kb ()12 M-1 at 25 °C, see Table 3) in the
case of Ph3As as entering ligand is such that a 1 M solution
of the ligand will be required to shift the equilibrium over
to the product side, whereas the solubility is limited to only
0.08 M at 25 °C.
Tables 3 and S1-S3 (Supporting Information) summarize
the kinetic data obtained as a function of temperature and
pressure for the series of reactions investigated. In the case
of Ph3As and Ph3Sb as entering nucleophiles, the temperature
and pressure dependence was studied at low and high
nucleophile concentrations in order to determine the effect
on ka and kb, respectively. The values of ka and kb as a
function of temperature and pressure, along with the corre-
sponding activation parameters, are summarized in Table S3
(Supporting Information). Plots of ln ka and ln kb versus
pressure gave good linear relationships as shown in Figure
S2 (Supporting Information) for the reaction of trans-[Co-
(Hdmg)2(R)MeOH] with Ph3As, and the constructed volume
profile is shown in Figure 6. Plots of ln ka versus pressure
gave good linear relationships as shown in Figure S3 for
the reaction of trans-[Co(Hdmg)2(R)H2O] with imidazole,
pyrazole, 1,2,4-triazole, N-acetylimidazole, 5-chloro-1-me-
-
thylimidazole, and NO2 . Similar plots are presented in
Figure S4 (Supporting Information) for the reactions between
trans-[Co(Hdmg)2(R)MeOH] and Ph3P where (R ) CH3 and
PhCH2).
Second-order rate constants (ka) for substitution of trans-
[Co(Hdmg)2(R)S] by the series of entering ligands in Table
3 at 25 °C do not depend strongly on the nature of the
entering ligand either in aqueous or in methanol solution.
The data in Table 4 show a similar trend except for
6096 Inorganic Chemistry, Vol. 43, No. 19, 2004

Alkyl Cobaloximes in Water and Methanol
Table 3. Kinetic Data for the Reaction of trans-[Co(Hdmg)2(R)S], for R ) CH3 and PhCH2, and S ) H2O and MeOH, with Different Nucleophiles
(L)
ka at 25 °C,
∆Hq,
∆Sq,
∆Vq,
La
M-1 s-1
kJ mol-1
J mol-1 K-1
cm3 mol-1
imidazoleb,c
pyrazoleb,c
70.3 ( 0.5
91 ( 2
74 ( 2
79 ( 2
72 ( 3
70 ( 1
89 ( 2
84 ( 3
94 ( 2
74 ( 2
103 ( 4
114 ( 5
89 ( 3
95 ( 3
92 ( 1
+40 ( 8
+56 ( 7
+31 ( 9
+22 ( 3
+92 ( 6
+78 ( 10
+97 ( 8
+29 ( 5
+126 ( 14
+143 ( 18
+77 ( 10
+95 ( 9
+109 ( 3
+3.7 ( 0.2
+4.5 ( 0.4
+4.0 ( 0.1
+3.1 ( 0.2
+2.5 ( 0.1
+2.3 ( 0.1
+12.3 ( 0.4
+7.3 ( 0.1
+10.6 ( 0.5
+15.9 ( 0.2
1,2,4-triazoleb,c
70.6 ( 0.7
48 ( 1
N-acetylimidazoleb,c
5-chloro-1-methylimidazoleb,c
105.1 ( 0.5
113.9 ( 0.8
19.6 ( 0.1
23.6 ( 0.2
20.1 ( 1.1
1.66 ( 0.03 s-1
20.3 ( 1.7
10.24 ( 0.04 s-1
252.0 ( 0.8
-b,c
NO2
Ph3Pd,e
imidazoled,e
d,e
Ph3As(forward reaction)
Ph3As(reverse reaction)
Ph3Sb(forward reaction)
d,e
d,e
d,e
Ph3Sb(reverse reaction)
+11.3 ( 0.4
+15.2 ( 0.3
Ph3Pf,g
a The two superscripts (b-e) given in the first column of the table refer to the experimental conditions selected for the temperature and pressure dependence
studies, respectively (see Supporting Information). b trans-[Co(Hdmg)2(Me)H2O] ) 1 × 10-4 M, imidazole, pyrazole, 1,2,4-triazole, and N-acetylimidazole
-
) 0.01-0.5 M, 5-chloro-1-methylimidazole ) 1.88 × 10-2 to 0.3 M, and NO2 ) 6.25 × 10-2 to 0.5 M, pH 9.0, and I ) 0.1 M NaClO4 and 0.5 M for
NO2- as ligand. c trans-[Co(Hdmg)2(Me)H2O] ) 2 × 10-4 M, imidazole, pyrazole, 1,2,4-triazole, and N-acetylimidazole ) 0.01 M, 5-chloro-1-methylimidazole
) 3.75 × 10-2 M, and NO2- ) 6.25 × 10-2 M, T ) 25.0 °C, pH 9.0, and I ) 0.1 M NaClO4 and 0.5 M for NO2- as ligand. d trans-[Co(Hdmg)2(Me)MeOH]
) 1 × 10-4 M, [imidazole] ) 0.05-0.5 M, [Ph3P] ) (0.5-2.5) × 10-2 M, [Ph3As] and [Ph3Sb] ) (0.5-4.0) × 10-2 M. e trans-[Co(Hdmg)2(Me)MeOH]
) 2 × 10-4 M, [imidazole] ) 0.05 M, [Ph3P] ) 0.01 M, [Ph3As] ) (0.5-4.0) × 10-2 M, T ) 25.0 °C, and [Ph3Sb] ) 0.02 M at 15.0 °C. f trans-
[Co(Hdmg)2(PhCH2)MeOH] ) 5 × 10-5 M and [Ph3P] ) (0.5-2.5) × 10-2 M. g trans-[Co(Hdmg)2(PhCH2)MeOH] ) 5 × 10-5 M and [Ph3P] ) 0.01 M
at 25.0 °C.
precursor formation constant (K) and an interchange rate
constant (k). This will further account for some of the
differences in the values of ka especially in cases where
precursor formation is favored by the nature of the entering
nucleophile and so affects the value of ka.
Substitution reactions of trans-[Co(Hdmg)2(R)S] show a
strong dependence on the nature of the alkyl group. The
second-order rate constant for substitution of S ) methanol
by Ph3P was found to be 19.6 and 252 M-1 s-1 at 25 °C for
R ) CH3 and PhCH2, respectively, and that for substitution
of water by imidazole was found to be 70.3 and 20.9 M-1
s-1 at 25 °C for R ) CH3 and CH2Cl,24 respectively. The
lability of these complexes is affected by the size and
electronic structure of the alkyl group. The second-order rate
constant for substitution of trans-[Co(Hdmg)2(R)H2O] by
Figure 6. Volume profile for the following reaction:
pyrazole, 1,2,4-triazole, and N-acetylimidazole was found
to be 1309, 1200, and 135 M-1 s-1 for R ) Et, 755, 691,
ka
trans-[Co(Hdmg)2(R)MeOH] + Ph3As y\kz
trans-[bCo(Hdmg)2(R)AsPh3] + MeOH
and 119 M-1 s-1 for R ) PhCH2, and 0.36, 0.35, and 0.07
M-1 s-1 for R ) CF3CH2 at 25 °C, respectively, as
summarized in Table 4.25 The second-order rate constants
show that the kinetic trans effect decreases in the order Et
> PhCH2 > CH3 > CH2Cl > CF3CH2 according to the donor
properties of the alkyl group.
N-acetylimidazole as entering ligand (see further in Discus-
sion). This is typical for a process that is dissociatively
activated, either via a dissociative interchange (Id) or limiting
dissociative (D) mechanism, in which bond breakage with
the leaving group largely controls the rate and nature of the
substitution mechanism. Similarly, the rate constants for the
backward reaction were found to be 1.7 and 10.2 s-1 at 25
°C for the displacement of Ph3As and Ph3Sb, respectively,
indicating that the release of the nucleophile does depend
on its nucleophilicity, again in line with a dissociatively
activated mechanism controlled by bond cleavage with the
leaving group. The most important factors that account for
the increase in kb from Ph3P to Ph3As and Ph3Sb are the
basicity and π-acceptor ability of these nucleophiles that
decrease in this order. A more detailed discussion of the
activation parameters given below reveals that a dissociative
interchange mechanism (Id) is favored, in terms of which
the overall second-order rate constant ka is a composite of a
The solvent has a significant effect on the rate con-
stant of the ligand substitution reaction. The second-order
rate constant for substitution of the solvent in trans-
[Co(Hdmg)2(R)S] by imidazole was found be 70.3 and 23.6
M-1 s-1 at 25 °C for S ) H2O and MeOH, respectively.
The slower reaction in methanol is ascribed to solvent effects
and the fact that we are dealing with a totally different
leaving group. The values of the second rate constant ka in
MeOH are again independent of the nature of the entering
nucleophile.
(24) Dreos-Garlatti, R.; Tauzher, G.; Costa, G. Inorg. Chim. Acta 1983,
70, 83.
(25) Hamza, M. S. A.; Felluga, A.; Randaccio, L.; Tauzher, G.; van Eldik,
R. In preparation.
Inorganic Chemistry, Vol. 43, No. 19, 2004 6097

Alzoubi et al.
Table 4. Comparison of Rate and Activation Parameters for the Reaction of trans-[Co(chel)2(R)H2O] and Aquacobalamin with Imidazole, Pyrazole,
1,2,4-Triazole, and N-Acetylimidazole as Entering Nucleophiles
ligand
∆Hq, kJ mol-1
∆Sq, J K-1 mol-1
∆Vq, cm3 mol-1
ka, M-1 s-1
trans-[Co(Hdmg)2(Me)H2O]a
+40 ( 8
imidazole
pyrazole
1,2,4-triazole
N-acetylimidazole
74 ( 2
79 ( 2
72 ( 3
70 ( 1
+3.7 ( 0.2
+4.5 ( 0.4
+4.0 ( 0.1
+3.1 ( 0.2
70.3 ( 0.5
90.7 ( 2.0
70.6 ( 0.7
48.2 ( 1.0
+56 ( 7
+31 ( 9
+22 ( 3
trans-[Co(Hdmg)2(CH3CH2)H2O]25,a
pyrazole
1,2,4-triazole
N-acetylimidazole
69 ( 0.6
65 ( 1
72 ( 0.6
+46 ( 2
+31 ( 3
+36 ( 2
+4.6 ( 0.3
+4.9 ( 0.3
+4.1 ( 0.3
1309 ( 14
1200 ( 7
135 ( 2
trans-[Co(Hdmg)2(PhCH2)H2O]25,a
pyrazole
1,2,4-triazole
N-acetylimidazole
70 ( 2
66 ( 1
72 ( 1
+46 ( 8
+31 ( 3
+38 ( 5
+6.2 ( 0.5
+7.5 ( 0.4
+4.3 ( 0.3
755 ( 16
691 (13
119 ( 4
trans-[Co(Hdmg)2(CF3CH2)H2O]25,a
pyrazole
1,2,4-triazole
N-acetylimidazole
83 ( 1
86 ( 2
87 ( 2
+23 ( 3
+37 ( 6
+26 ( 5
+4.4 ( 0.3
+2.3 ( 0.3
+3.4 ( 0.3
0.358 ( 0.006
0.348 ( 0.002
0.069 ( 0.001
trans-[Co(en)2(Me)H2O]2+ 1,29
-22 ( 7
imidazolea
53 ( 2
67 ( 6
59 ( 2
72 ( 4
+4.7 ( 0.1
198 ( 13
284 ( 10
318 ( 5
24 ( 3
pyrazoleb
+27 ( 19
1,2,4-triazoleb
N-acetylimidazoleb
+1 ( 6
+23 ( 14
aquacobalamin12,30a
imidazole
pyrazole
1,2,4-triazole
N-acetylimidazole
21.2 ( 0.7
35.0 ( 0.8
24.0 ( 0.4
6.6 ( 1.6
85 ( 4
90 ( 2
108 ( 12
+69 ( 13
+83 ( 6
+135 ( 40
+7.2 ( 0.3
a The second rate constants were determined at 25.0 °C. b The second-order rate constants were calculated from the activation parameters for 25.0 °C.
The lability of organo cobaloximes is much higher than
that of inorganic cobaloximes due to the alkyl group in the
axial position that significantly increases the length of the
Co-OH2 bond in the ground state and increases its pKa value.
pKa values were found to be 10.9626 and 12.6827 for trans-
[Co(Hdmg)2(CF3CH2)H2O] and trans-[Co(Hdmg)2(Me)H2O],
respectively. The activation enthalpy values for substitution
of trans-[Co(Hdmg)2(R)H2O] by pyrazole, 1,2,4-triazole, and
N-acetylimidazole for R ) PhCH2, CH3, and CF3CH2 were
found to be in the ranges 66-72,25 72-79, and 83-86 kJ
mol-1,25 respectively, as summarized in Table 4. The
activation entropies in Table 3 are in general significantly
positive and in line with a dissociatively activated reaction
mechanism. A plot of ∆Hq versus ∆Sq gives a good linear
relationship for the reaction of trans-[Co(Hdmg)2(Me)S] with
Figure 7. Correlation between ∆Hq and ∆Sq for the second-order rate
constant for substitution of H2O in trans-[Co(Hdmg)2(Me)H2O] by N-
acetylimidazole (1), 1,2,4-triazole (2), imidazole (3), pyrazole (4), NO2
imidazole, pyrazole, 1,2,4-triazole, N-acetylimidazole, 5-chloro-
1-methylimidazole, and NO2 in water as shown in Figure
-
-
7, and for the reaction with imidazole, Ph3P, Ph3As, and Ph3-
Sb in MeOH as shown in Figure S5 (Supporting Informa-
tion). The good compensation correlation between ∆Hq and
∆Sq suggests that an isokinetic relationship exists, which
supports the claim that a single mechanism operates for the
series of nucleophiles.28
The activation volumes for substitution of coordinated
water in trans-[Co(Hdmg)2(Me)H2O] by imidazole, pyrazole,
1,2,4-triazole, N-acetylimidazole, 5-chloro-1-methylimida-
(5), and 5-chloro-1-methylimidaziole (6).
-
zole, and NO2 were found to be between +2.3 and +4.5
cm3 mol-1. The relatively small positive values suggest that
the reaction follows an Id mechanism in which the entering
nucleophile is weakly bound in the transition state. The
substitution of coordinated water by py and 4-CNpy in trans-
[Co(Hdmg)2(CF3CH2)H2O] proceeds according to the same
mechanism since ∆Vq was found to be +5.7 and +7.4 cm3
mol-1, respectively.6 Similarly, ∆Vq values between +3.6
and +9.7 cm3 mol-1 were reported for displacement of water
in trans-[Co(en)2(Me)H2O]2+ 2,29 and in aquacobalamin14,30
(26) Brown, K. L.; Lyles, D.; Pencovici, M.; Kallen, R. G. J. Am. Chem.
Soc. 1975, 97, 7338.
(27) Brown, K. L.; Chernoff, D.; Keljo, D. J.; Kallen, R. G. J. Am. Chem.
Soc. 1972, 94, 6697.
(28) Espenson, J. H. Chemical Kinetics and Reaction Mechanisms;
McGraw-Hill: New York, 1995, p 164.
(29) Hamza, M. S. A. J. Coord. Chem. 2003, 56, 553.
6098 Inorganic Chemistry, Vol. 43, No. 19, 2004

Alkyl Cobaloximes in Water and Methanol
by different nucleophiles. ∆Vq values for the substitution of
water by N-acetylimidazole in trans-[Co(Hdmg)2(Me)H2O]
and aquacobalamin12 were found to be +3.1 and +7.2 cm3
mol-1, respectively. Similar data for ligand substitution
reactions of trans-[Co(Hdmg)2(R)H2O] by 1,2,4-triazole for
25
R ) CF3CH2,25 CH3, and PhCH2 were found to be +2.3,
+3.2, and +7.5 cm3 mol-1, respectively, as summarized in
Table 4. These values all favor a dissociative interchange
mechanism that consists of rapid precursor formation and a
rate-determining interchange reaction. The consequence is
that the overall second-order rate constant ka is a composite
value of the precursor formation constant (K) and the
interchange rate constant (k), such that ka ) kK as mentioned
above. This naturally complicates the interpretation of ka and
its activation parameters since it was not possible to separate
k and K (and their thermodynamic parameters) kinetically
in the studied reactions, as could be done in other cases.6,13
As mentioned before, activation volumes for ligand
substitution reactions on [Co(TMPP)(H2O)2]5+ and [Co(TPPS)-
(H2O)2]3- (TMPP ) meso-tetrakis(4-N-methylpyridyl)por-
phine and TPPS ) meso-tetrakis(p-sulfonatophenyl)porphine)
are +14.4 and +15.4 cm3 mol-1, respectively,16,17 and were
assigned to a limiting D mechanism. The activation volumes
for the substitution of methanol in trans-[Co(Hdmg)2(R)-
MeOH] by Ph3P for R ) CH3 and PhCH2 were also found
to be more positive, viz. +12.3 and +15.2 cm3 mol-1,
respectively, which could suggest the operation of a limiting
dissociative mechanism. However, this trend could also be
due to the larger partial molar volume of methanol as
compared to water, since this will affect the magnitude of
∆Vq for a dissociatively activated process. By way of
comparison, it is interesting to note that the reaction of trans-
[Co(Hdmg)2(CH3)MeOH] with imidazole is characterized by
a significantly smaller activation volume than that found for
Ph3P, which suggests the operation of a dissociative inter-
change mechanism. The reason for this apparent difference
in mechanism may be related to the formation of a precursor
complex with imidazole, which would then favor a dissocia-
tive interchange mechanism in which precursor formation
will contribute to the overall activation parameters found for
ka since ka ) kK.
In one system, it was possible to construct a volume profile
for the reaction between trans-[Co(Hdmg)2(Me)MeOH] and
Ph3As (Figure 6), from which the dissociative character of
the substitution process is clearly seen. The larger ∆Vq for
the back reaction is related to the larger partial molar volume
of the leaving ligand Ph3As, which also accounts for the
overall negative reaction volume. Although these larger
activation volumes tend to suggest the operation of a limiting
D mechanism, this suggestion is speculative due to the
uncertain contribution of the larger partial molar volumes
of the leaving groups for both the forward and back reactions
as outlined above. In the case of a dissociative interchange
Id mechanism, the volume of activation for the forward
reaction will include contributions arising from precursor
formation and the interchange reaction step since ka ) kK
Figure 8. Crystal structure of the cation trans-[Co(en)2(Me)H2O]2+
.
Table 5. Structural Parameters for Complexes of the Type
trans-[Co(chel)2(X)H2O], Where chel Represents the Equatorial Chelate
Ligand
X
Co-O (Å)
Co-X(Å)
O-Co-X (deg)
trans-[Co(en)2(X)H2O]2+
CH3
CH3
CH3
CH3
2.153(6)
1.995(10)
177.6(3)
trans-[Co((DO)(DOH)pn)(X)H2O]+
34
2.14(2)
trans-[Co((EMO)(EMOH)pn)(X)H2O]+
2.102(3) 1.997(4)
trans-[Co(Hdmg)2(X)H2O]
1.99(4)
31
33
2.058(3)
2.056(5)
1.992(4)
1.980(5)
1.955(11)
1.990(5)
2.044(7)
1.906(5)
1.881(9)
2.364(2)
178.0(2)
172.4(3)
32
CH2CMe3
CN32
32
NO2
Br32
180.0
and ∆Vq(ka) ) ∆Vq(k) + ∆V(K). Due to our inability to
separate k and K kinetically in the present case, no further
details can be included in the volume profile.
The labilization of the ligand in the position trans to the
cobalt-carbon bond can in many cases be observed in the
ground state as a result of the trans influence. In the present
study, it was possible to crystallize trans-[Co(en)2(Me)H2O]-
S2O6 of which the structure, determined by X-ray analysis,
is shown in Figure 8. The crystal consists of organometallic
cations and dithionate anions. The structure of the complex
is a distorted octahedron, the cobalt atom and the four
nitrogen atoms are located in the equatorial plane, and the
water molecule and methyl group occupy the axial positions
as shown in Figure 8. The Co-O and Co-C bond lengths
in trans-[Co(en)2(Me)H2O]2+ were found to be 2.153(6) and
1.995(10) Å, respectively, and the Co-N bond lengths for
N(1), N(2), N(3), and N(4) were found to be 1.999(7), 1.995-
(8), 1.931(7), and 1.927(7) Å, respectively. The significantly
larger Co-N(1) and Co-N(2) distances can be ascribed to
the influence of hydrogen bonding to the oxygen atoms of
2-
the S2O6 anion, which is located close to these amines in
the crystal structure.
A comparison of the crystal structure data with closely
related complexes is given in Table 5. The bond lengths
depend on the nature of alkyl group and the equatorial
ligands, the hybridization state of the carbon bond attached
to the cobalt center, and steric interactions between the axial
alkyl and the equatorial ligands (steric cis influence).31,32 In
addition, the overall charge on the complexes changes from
(30) Marques, H. M.; Egan, T. J.; Marsh, J. H.; Mellor, J. R.; Munro, O.
(31) Marzilli, L. G.; Bresciani-Pahor, N.; Randacio, L.; Zangrando, E.;
Q. Inorg. Chim. Acta 1989, 166, 249.
Finke, R. G.; Myers, S. A. Inorg. Chim. Acta 1985, 107, 139.
Inorganic Chemistry, Vol. 43, No. 19, 2004 6099

Alzoubi et al.
2+ to zero, which will also affect the metal-ligand bond
lengths and complicate the following comparison. The X-ray
structure data show that the bond lengths of Co-O are 2.058-
(3),33 2.102(3),31 2.14(2),34 and 2.153(6) Å for trans-[Co-
(Hdmg)2(Me)H2O], trans-[Co((EMO)(EMOH)pn)(Me)H2O]+,
trans-[Co((DO)(DOH)pn)(Me)H2O]+, and trans-[Co(en)2-
(Me)H2O]2+, respectively, where (DO)(DOH)pn ) diacetyl-
monoxime-diacetyl monoximato-propane-1,3-diyldiimino and
(EMO)(EMOH)pn ) 3,9-dimethyl-2,10-diethyl-1,4,8,11-
tetraazaundeca-1,3,8,10-tetraen-11-ol-1-olato. These data show
that the Co-OH2 bond is stronger in the ground state in the
case of trans-[Co(Hdmg)2(Me)H2O]. The Co-O bond length
increases from 1.955(4) Å for X ) Br to 2.058(3) Å for X
) CH3 in the complex trans-[Co(Hdmg)2(X)H2O] as shown
in the Table 5.32 A comparison of the data in Table 5
indicates that several kinds of structural changes arise from
changes in the electron-donating ability of R and from the
interactions between the alkyl group and the equatorial
moiety, and the nature of the equatorial ligand. The Co-C
bond lengthens from 1.990 to 2.044 Å as the bulk of the
alkyl group increases from CH3 to CH2CMe3 in the case of
the dimethylglyoximate complex, and clearly depends on the
steric interaction between the axial alkyl and the equatorial
ligands.31 In addition, the hybridization of the carbon atom
affects the Co-C bond length, which was found to be 2.00,
1.96, and 1.90 Å for sp3, sp2, and sp hybridization,
respectively.32 Distortion arising from the interaction of the
axial alkyl group with the equatorial ligand can be ascribed
to (a) changes in the C-Co-N(eq) bond angles; (b) widening
of the Co-CH2-R′ angles and shortening of the Co-R bond
lengths; (c) lengthening of the Co-C bond; (d) deformation
of the equatorial ligand unit.32
It can be concluded that the variety of chelates, R groups,
entering ligands, and solvents selected in this and our earlier
studies1-6 enables a systematic tuning of the lability of alkyl
cobalt(III) complexes, which results in a systematic tuning
of the ligand displacement process in terms of a dissociative
interchange (Id) and a limiting dissociative (D) mechanism.
Acknowledgment. The authors gratefully acknowledge
financial support from the Deutsche Forschungsgemeinschaft
(SFB 583 “Redox-active metal complexes”), Fonds der
Chemischen Industrie, and the Max-Buchner Forschungs-
stiftung. We thank Alessandro Felluga, Lucio Randaccio, and
Giovanni Tauzher, Universita` degli Studi di Trieste, Italy,
for providing samples of the complexes used in this study.
(32) Bresciani-Pahor, N.; Forcolin, M.; Marzilli, L. G.; Randaccio, L.
Summers, M. F.; Toscano, P. J. Coord. Chem. ReV. 1985, 63, 1.
(33) McFadden, D. L.; McPhail, A. T. J. Chem. Soc., Dalton Trans. 1974,
363.
(34) Bu¨ckner, S.; Calligaris, M.; Nardin, G.; Randaccio, L. Inorg. Chim.
Acta 1969, 3, 278.
Supporting Information Available: Crystallographic data in
CIF format. Additional tables and figures. This material is available
free of charge via the Internet at http://pubs.acs.org.
IC049761J
6100 Inorganic Chemistry, Vol. 43, No. 19, 2004