Probing the nature of the Co(III) ion in corrins: Comparison of reactions of aquacyanocobyrinic acid heptamethyl ester and aquacyano-stable yellow cobyrinic acid hexamethyl ester with neutral N-donor ligands

Article abstract of DOI:10.1021/ic302386u

Equilibrium constants (log K) for substitution of coordinated H ₂O in aquacyanocobyrinic acid heptamethyl ester (aquacyanocobester, ACCbs) and aquacyano-stable yellow cobyrinic acid hexamethyl ester (aquacyano-stable yellow cobester, ACSYCbs), in which oxidation of the C5 carbon of the corrin interrupts the normal delocalized system of corrins, by neutral N-donor ligands (ammonia, ethanolamine, 2-methoxyethylamine, N-methylimidazole, and 4-methylpyridine) have been determined spectrophotometrically as a function of temperature. Log K values increase with the basicity of the ligand, but a strong compensation effect between ΔH and ΔS values causes a leveling effect. The aliphatic amines with a harder donor atom produce ΔH values that are more negative in their reactions with ACSYCbs than with ACCbs, while the softer, aromatic N donors produce more negative ΔH values with ACCbs than with ACSYCbs. Molecular modeling (DFT, M06L/SVP, and a quantum theory of atoms in molecules analysis of the electron density) shows that complexes of the aliphatic amines with SYCbs produce shorter and stronger Co-N bonds with less ionic character than the Co-N bonds of these ligands with the cobester. Conversely, the Co-N bond to the aromatic N donors is shorter, stronger, and somewhat less ionic in the complexes of the cobester than in those of the SYCbs. Therefore, the distinction between the harder Co(III) in ACSYCbs and softer Co(III) in ACCbs, reported previously for anionic ligands, is maintained for neutral N-donor ligands.

Full text of DOI:10.1021/ic302386u

Article
pubs.acs.org/IC
Probing the Nature of the Co(III) Ion in Corrins: Comparison of
Reactions of Aquacyanocobyrinic Acid Heptamethyl Ester and
Aquacyano-Stable Yellow Cobyrinic Acid Hexamethyl Ester with
Neutral N‑Donor Ligands
Susan M. Chemaly,* Louise Kendall, Monika Nowakowska, Dale Pon, Christopher B. Perry,
and Helder M. Marques*
Molecular Sciences Institute, School of Chemistry, University of the Witwatersrand, PO Wits, Johannesburg, 2050 South Africa
*
S Supporting Information
ABSTRACT: Equilibrium constants (log K) for substitution of coordinated H O in
2
aquacyanocobyrinic acid heptamethyl ester (aquacyanocobester, ACCbs) and
aquacyano-stable yellow cobyrinic acid hexamethyl ester (aquacyano-stable yellow
cobester, ACSYCbs), in which oxidation of the C5 carbon of the corrin interrupts the
normal delocalized system of corrins, by neutral N-donor ligands (ammonia,
ethanolamine, 2-methoxyethylamine, N-methylimidazole, and 4-methylpyridine) have
been determined spectrophotometrically as a function of temperature. Log K values
increase with the basicity of the ligand, but a strong compensation effect between ΔH
and ΔS values causes a leveling effect. The aliphatic amines with a harder donor atom
produce ΔH values that are more negative in their reactions with ACSYCbs than with
ACCbs, while the softer, aromatic N donors produce more negative ΔH values with
ACCbs than with ACSYCbs. Molecular modeling (DFT, M06L/SVP, and a quantum
theory of atoms in molecules analysis of the electron density) shows that complexes of
the aliphatic amines with SYCbs produce shorter and stronger Co−N bonds with less
ionic character than the Co−N bonds of these ligands with the cobester. Conversely, the Co−N bond to the aromatic N donors
is shorter, stronger, and somewhat less ionic in the complexes of the cobester than in those of the SYCbs. Therefore, the
distinction between the harder Co(III) in ACSYCbs and softer Co(III) in ACCbs, reported previously for anionic ligands, is
maintained for neutral N-donor ligands.
1
. INTRODUCTION
electron-withdrawing group such as NO deactivates the axial
25
coordination site toward ligand substitution. This all indicates
that the chemistry of the axial coordination site of Co(III) can
be controlled and modified by the electronic structure of the
corrin ring.
An intriguing aspect of the inorganic chemistry of cobalt corrins
(
(
derivatives of vitamin B , Figure 1) is the lability of Co(III)
12
for example, see refs 1−17). The approximate lability ratio of
Co(III) toward axial ligand substitution in corrin, porphyrin,
9
6
4
14,18
To further examine this, we recently prepared and fully
cobaloxime, and tetra-ammine systems is 10 :10 :10 :1;
1,17
characterized
a derivative of the hexamethyl ester of
thus, the extent of unsaturation of the equatorial ligand seems
to be an important factor in determining the lability of the axial
coordination site. We suggested that this kinetic cis effect is the
consequence of the delocalization of electron density between
the equatorial ligands and the metal, making the metal softer
cobyrinic acid, (5R,6R)-Coα,Coβ-dicyano-5,6-dihydro-5-hy-
droxy-hexamethylcob(III)yrinate-c,6-lactone (which we refer
to as a “stable yellow cobester”, DCSYCbs, Figure 1), originally
2
6,27
reported by others,
inic acid heptamethyl ester, Coα,Coβ-dicyano-heptamethylcob-
III)yrinate (DCCbs, Figure 1), in which the 13-atom, 14 π-e−
from the corresponding dicyanocobyr-
1
4,18,19
and imparting to it a more labile Co(II)-like character.
(
As part of our investigations of this aspect of B12 chemistry
we have been interested for some time in the cis effect of the
corrin macrocycle. It is well established that the electronic
spectra of the cobalamins are dominated by π → π* transitions
of the corrin ring
dependent on the identity of the axial ligand;
delocalized system of normal corrins is interrupted by oxidation
−
of the C5 carbon, leading to a 10 π-e system between N21 and
N24 and an isolated double bond between N21 and C4. We
showed from a variety of spectroscopic data and DFT
calculations that the diminished conjugation of the corrin in
DCSYCbs increases the hardness of Co(III) in the corrin. We
prepared the corresponding aquacyano complexes of the
20−23
and that the spectra are heavily
19,22,24
there is
therefore electronic communication between the axial and
equatorial ligands. We have shown that the C10−Cl bond
length in X-10−ClCbls, in which the C10 H is replaced by Cl,
14
depends strongly on the polarizability of the axial ligand, X.
Received: November 1, 2012
We have also shown that substitution of the C10 H by an
©
XXXX American Chemical Society
A
dx.doi.org/10.1021/ic302386u | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
−
−
−
Figure 1. Structure of, left, the dicyano (X = Y = CN ) and aquacyano (X = H O, Y = CN or X = CN , Y = H O) cobester. Latter exists in solution
2
2
as a mixture of two interconverting diastereomers. Corresponding stable yellow cobesters, right, have an interrupted conjugation between C5 and
1
C6. Lactone bridges C6 and C7 and is directed toward the upper (β) face of the corrin.
cobester and the stable yellow cobester, ACCbs and ACSYCbs,
and showed that they exist as a mixture of interconverting
DCSYCbs, and its aquacyano form, ACSYCbs) were prepared as
1
previously described. NH Cl (Merck), NH EtOH, NH EtOH·HCl,
4
2
2
−
−
NH EtOCH , CH NH 40% solution in H O, CH NH ·HCl, NMeIm,
2 3 3 2 2 3 2
diastereomers (i.e., α-H O,β-CN and α-CN ,β-H O cobester
2
2
−
−
4MePy, 4MeOPy, hydroxylamine·HCl, sodium bicarbonate, and
sodium carbonate (all Sigma Aldrich), CF CH NH ·HCl, 2-
and α-H O,β-CN and α-CN ,β-H O stable yellow cobester)
which can therefore not be isolated. We compared the
thermodynamics (with the anionic ligands N , NO , CN ,
SO3 ) and kinetics (with CN ) of the substitution of H O.
2
2
17
3
2
2
chloroethylamine·HCl, Py, DMAP, and O-methoxylamine·HCl (Al-
drich), and sodium dihydrogen phosphate and disodium hydrogen
phosphate (Saarchem, Unilab) were used as received. Water was
purified using a Millipore RO unit and further purified using a
Millipore Milli-Q unit (18 MΩ cm).
−
−
−
17
3
2
2−
−
2
−
We found that the ligands with a harder donor atom (N in N3
−
and NO ) produce ΔH values that are more negative in their
2
It was difficult to determine the acid dissociation constant of
reactions with ACSYCbs than with ACCbs. If the ligand has a
−
2−
coordinated H O in the cobesters because of competing hydrolysis of
softer donor (C in CN , S in SO ) then ΔH is less positive,
2
3
1
the ester moieties. A reasonable value was obtained for ACCbs (pK =
a
or more negative, for reactions with ACCbs than with
ACSYCbs. Thus, the softer metal in ACCbs has a preference
for softer ligands and the harder metal in ACSYCbs for the
harder ligands.
1
0.56 ± 0.08 at 25.0 °C, see below) as ester hydrolysis only sets in
above ca. pH 11. The value for ACSYCbs is probably lower, but in that
case hydrolysis is rapid and occurs above ca. 10.5. As the pH at which
equilibrium constants were determined was between 9 and 9.2,
There is a significant increase in inertness of Co(III) on
decreasing the extent of the conjugation of the corrin ligand.
ACCbs is more labile than ACSYCbs toward substitution of
correction for hydrolysis of coordinated H O in either ACCbs or
2
ACSYCbs is unnecessary. The values of the pK s of the conjugate acids
a
2
8
of the ligands were from the NIST database.
.3. Methods. The acid dissociation constant of ACCbs was
determined using UV−vis spectroscopy. ACCbs was dissolved in a
−
‡
−1
2
H O by CN with ΔH smaller by some 12 kJ mol . We
2
interpreted this as arising from an earlier transition state in
which bonding between the softer metal of ACCbs and the
ligand is greater than in ACSYCbs with its harder metal center.
2
9
multicomponent buffer (μ = 0.1 M, Na SO ) in an external 50 mL
2
4
thermostatted cell. pH was adjusted by adding negligible volumes of 8
M NaOH between pH 5.5 and 11.5 while the solution was pumped
through a flow cell housed in the thermostatted cell compartment of a
Cary 3E UV−vis spectrophotometer. The dependence of the
absorbance at 354 nm was fitted using standard nonlinear least-
squares methods employing a Newton−Raphson procedure (Figure
S1, Supporting Information).
Our comparison of the thermodynamics of H O substitution
2
on ACCbs and ACSYCbs was limited to anionic ligands. In this
paper we turn our attention to neutral N-donor ligands,
including an imidazole (N-methyl imidazole, NMeIm), four
pyridines, (pyridine, Py; 4-methyl pyridine, 4MePy; 4-
methoxypyridine, 4-MeOPy; dimethylaminopyridine, DMAP),
The concentrations of the cobesters and stable yellow cobesters in
solution were determined by addition of excess KCN to give DCCbs
and four primary amines (NH , ethanolamine (NH EtOH), 2-
3
2
−4
−1
−1 30
−4
(
ε368 3.0 × 10 L mol cm ) and DCSYCbs (ε484 1.0 × 10
L
methoxyethylamine (MeOEtNH ), and 2,2,2-trifluoroethyl-
−1
−1 26
2
mol cm ), respectively.
amine (CF CH NH )). Reactions with hydroxylamine, me-
3
2
2
The equilibrium constants for coordination of the ligands by ACCbs
and ACSYCbs (eq 1) were determined by addition of between 8 and
15 aliquots of a stock solution of the appropriate ligand to a 20−70
μM solution of the cobalt corrin contained in a 10 mm path length
cuvette housed in the thermostatted cell block of a UV−vis
spectrophotometer. All absorbance readings were corrected for
dilution. Solutions were buffered with carbonate (pH 9) or phosphate
thoxylamine, methylamine, and 2-chloroethylamine were also
investigated; these proved to be problematic, for reasons
discussed.
2
. EXPERIMENTAL SECTION
2
.1. Instrumentation. UV−vis spectra were recorded in 10 mm
(pH 7) at an ionic strength μ = 0.10 M. Stock solutions of NH and
3
quartz cuvettes on a Cary 1E or Cary 300 UV−visible spectropho-
tometer. The temperature of the cell block was maintained (±0.1 °C)
by a water-circulating bath or with a Peltier device. pH was measured
using a Metrohm 720 or 827 pH meter with a Unitrode or Biotrode
electrode calibrated at pH 7.00 and 9.00 with standard buffers.
NH EtOH were prepared from NH Cl and NH EtOH·HCl,
2
4
2
respectively, and adjusted to pH 9 by addition of NaOH. The stock
solution of NH EtOMe was adjusted to pH 9 by addition of HCl. For
2
titrations of ACCbs and ACSYCbs with NH , NH EtOH, and
3
2
NH EtOMe, the ionic strength in the cell increased to 0.22−0.28 M
2
2.2. Reagents. Coα,Coβ-dicyano-heptamethylcob(III)yrinate (di-
and 0.30−0.67, respectively. Stock solutions of NMeIm, 4MePy, and
4MeOPy for reaction with ACCbs were prepared by dissolving the
ligand in buffer; neat NMeIm and 4MePy were used for titrations with
ACSYCbs. Reactions of all ligands were “instantaneous”, except that
cyanocobester, DCCbs, Figure 1A) and its aquacyano form, ACCbs,
and (5R,6R)-Coα,Coβ-dicyano-5,6-dihydro-5-hydroxy-
hexamethylcob(III)yrinate-c,6-lactone (dicyano-stable yellow cobester,
B
dx.doi.org/10.1021/ic302386u | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Table 1. Equilibrium Constants for Substitution of Coordinated H O in ACCbs and ACSYCbs
2
ACCbs
ACSYCbs
1
b
ΔH/kJ mol−1 ΔS/J K−1 mol−1
b
ligand
NH3
T/°C
log K
ΔH/kJ mol− ΔS/J K−1 mol−1 log K25
T/°C
10.0
log K
log K25
10.0
3.22(7)
3.08(5)
2.95(7)
2.84(6)
2.71(3)
2.04(23)
2.00(13)
1.89(13)
1.74(16)
1.69(10)
−41(1)
−85(2)
2.75
3.30(4)
−46(2)
−98(7)
2.94
1
2
2
3
5.0
0.0
5.0
0.0
9.9
6.7
3.2
2.0
7.4
15.0
20.0
25.0
30.0
10.0
15.0
17.5
20.0
25.0
3.18(12)
3.02(14)
2.86(5)
c
c
2.78(15)
2.18(7)
NH EtOH
−24(2)
−44(7)
1.91
−79(3)
−23(9)
1.52
2
1
2
3
3
1.96(16)
1.84(16)
1.75(7)
1.44(12)
1.24(10)
1.11(14)
2.28(35)
1.92(10)
1.59(30)
1.48(13)
1.27(12)
1.31(29)
1.47(33)
1.49(15)
1.52(15)
1.64(33)
0.37(16)
0.40(9)
3
3
0.0
2.5
c
NH EtOMe
11.0
2.40(10)
2.15(8)
1.89(9)
1.68(10)
1.52(7)
4.43(4)
4.32(3)
4.21(3)
4.16(6)
4.05(4)
3.62(3)
3.53(2)
3.41(3)
3.22(2)
3.14(2)
3.54(4)
3.45(3)
3.32(3)
3.23(3)
3.11(3)
−51(2)
−23(1)
−29(2)
−25(1)
−136(8)
5(4)
1.84
4.29
3.36
3.33
10.0
17.5
25.0
32.5
40.0
9.9
−53(4)
23(3)
5(1)
−145(15)
107(11)
24(2)
1.70
1.56
0.38
2
1
2
3
4
7.5
5.0
2.5
0.0
9.9
6.7
3.2
2.0
7.4
NMeIm
1
2
3
3
17.1
19.5
24.4
31.5
10.0
17.5
25.0
32.5
4
MePy
10.0
−33(8)
−19(3)
17.5
25.0
32.5
40.0
0.41(22)
0.44(23)
c
4
MeOPy
10.0
See text
1
2
3
4
7.5
5.0
2.5
0.0
d
d
CF CH NH
2
25.0
25.0
0.4
25.0
25.0
<−2
3
2
NH OCH3
1.93(3)
see text
2
a
These values have been corrected for the pK of the conjugate acid of the ligand. Correction for the pK of coordinated H O is probably
a
a
2
b
c
d
unnecessary (see text). Interpolated value of log K at 25 °C from the values of ΔH and ΔS. Unstable at 40 °C. Estimated; see text.
between ACSYCbs and 4MePy, which was allowed to stand for ∼1 h
nm between about 320 and 570 nm, excluding wavelengths (such as
those close to isosbestic points) where the absorbance change was
small and where the decrease in the signal-to-noise ratio makes values
of log K unreliable. Typically, we only used wavelengths where the
correlation coefficient of the nonlinear fit to the binding isotherm (see
below) was >0.97. K was then found from a weighted average,
before taking readings.
+
+
[
NC−Co(Cor)−OH ] + L ⇌ [NC−Co(Cor)−L] + H O
2
2
(1)
+
+
K = [NC−Co(Cor)−L] /[NC−Co(Cor)−OH ] ·[L]
2
2
weighted by the reciprocal of the variance, 1/σ , where the standard
deviation σ = (standard error of the fit)/n1 for n data points in the
/2
pK −pH
a
[L]neutral = [L]total /(1 + 10
)
(2)
titration.
In eq 1, L specifically refers to the neutral or unprotonated N donor
available for coordination to the metal center; this will depend on the
For cases where log K is relatively small (log K < 4), absorbance
data were fitted using standard nonlinear least-squares methods to a
simple binding isotherm (eq 3) as objective function, with A , A , and
pH of the solution (eq 2) where pK refers to the acid dissociation
a
0
1
constant of the conjugate acid of the ligand. pH was measured at the
start and end of each titration, and the average pH value was used for
this pH correction. Equilibrium constants reported have therefore
been corrected for the effect of pH.
K as the parameters to be optimized and where [L] refers specifically
to neutral, unprotonated L (eq 2). A and A are the absorbance values
0
1
at λ corresponding to 0 and 100% complex formation, respectively.
17
As we mentioned above and discussed elsewhere, both ACCbs
and ACSYCbs exist in solution as an equilibrium mixture of two
diastereomers. They are expected to have similar, but not identical,
equilibrium constants (log K, eq 1) for replacement of coordinated
A = (A + KA [L])/(1 + K[L])
(3)
λ
0
1
When log K is relatively large (>4), the implicit assumption in eq 3
that [L]free ≈ [L]neutral will not be true as a non-negligible fraction of L
H O by L. Spectrophotometeric determination of log K will therefore
2
will be complexed to the metal ion. In such cases [L] is given by a
show a wavelength dependence, which is a function of the relative
free
3
2
extinction coefficients of the two diastereomers at the monitoring
quadratic expression, only one root of which (eq 4) has physical
meaning and where [M]total is the total metal ion concentration.
31
17
wavelength. As we did before, we determined log K at every 1 or 2
C
dx.doi.org/10.1021/ic302386u | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
2
indicative of formation of a stable yellow corrinoid; thus, it
−
a2 ± a − 4a a
2
1
3
[
L]free
=
appears that NH OH does not merely coordinate the metal ion
2
2
a1
but acts as a reductant as well. There is a report of a similar
46
a = K; a = 1 + K[M]
− K[L]neutral ; a = −[L]
reaction between NH OH and B12a itself; in that case, by
2
1
2
total
3
neutral
working rapidly at pH 7, the reduction reaction was sufficiently
slow to allow for an estimation of log K. When we performed
the reaction at pH 6 we did indeed observe spectral changes
clearly indicative of ligand coordination (the γ band shifted
from 354 to 359 nm, gaining 6% in intensity; the D/E band
shifted from 406 to 419 nm; and the αβ bands shifted from 497
and 526 to 515 and 547 nm, respectively), but over the course
of 15 min the spectrum underwent a blue shift with an increase
in intensity at 460 nm, indicative of attack on the corrin ring.
Hence, reductive attack on the corrin prevented a reliable
determination of the log K value.
(4)
Values of K were determined as a function of temperature between 10
and 40 °C. Values of ΔH and ΔS were determined from the slope and
intercept, respectively, of weighted linear least-squares van’t Hoff plots
of ln K against 1/T.
.4. Computational Methods. Density functional theory (DFT)
calculations were carried out using the Gaussian 09 suite of
programs. Models used for calculation of the structure of the ammine,
imidazole, NH EtOH, and 4MePy adducts of ACSYCbs were based on
the crystal structure of dicyano-SYCbs after suitable editing. For all
calculations, the b, d, e, f, and g methyl-ester side chains were replaced
by methyl groups while the a side chain was left in place. Since no
crystal structures are available for any of the compounds of interest,
the β-axial ligand in the crystal structure of dicyano-SYCbs was
replaced by the appropriate ligand as follows. The NH adduct was
modeled by replacing the β-CN ligand with NH . The lowest energy
conformation of the imidazole moiety relative to the corrin ring was
determined by rotating the Co−N
increments in both the positive and the negative directions using the
semiempirical PM6 method. The 4MePy ligand was positioned such
that its orientation relative to the corrin ring was the same as that for
2
33
2
1
Reaction of NH OCH with ACCbs proceeded smoothly at
2
3
1
pH 9.2 with well-defined isosbestic points and log K at 25.0 °C
=
1.93(3). However, reaction between NH OCH and
2
3
3
−
ACSYCbs at pH 9.2 and 7.0 gave complex spectral changes
(Figure S5, Supporting Information). It appears that two
different ligands coordinate the metal center, and there is a slow
loss of intensity of the entire spectrum, consistent with
destruction of the chromophore. It has recently been reported
3
bond through 360° in 5°
Imidazole
34
II
3−
that in a free radical mechanism [Fe (CN) H O] catalyzes
5
2
imidazole. The NH EtOH ligand was positioned over the southeastern
quadrant of the corrin ring when viewed from above; this is known to
be an energy-minimum conformation for many cobalamin deriva-
2
the disproportionation of methoxylamine to NH , N , and
3
2
CH OH by initially coordinating the amine and then
3
47
35
promoting homolysis of the N−O bond. We tentatively
suggest by analogy that NH OCH coordinates to ACSYCbs,
tives. Models used for calculation of the structure of the ammine,
imidazole, NH EtOH, and 4MePy adducts of the cobester were based
2
3
2
3
6
+
on the crystal structure of ImCbl after suitable editing. For all
calculations, the b, d, e, f, and g amide side chains were replaced by
methyl groups while the a and c side chains were left in place and
converted into methyl esters. The α-axial dimethylbenzimidazole
which catalyzes its disproportionation, leading to at least some
formation of the NH3 complex (Figure S6, Supporting
Information). The radical nature of the reaction presumably
is responsible for the spectral bleaching. Because of the
intractability of the reaction of CH ONH with ACSYCbs, we
−
ligand was replaced with CN and the imidazole β axial ligand with the
appropriate ligand. The 4MePy ligand was positioned such that its
orientation relative to the corrin ring was the same as that for
3
2
only determined log K for its reaction with ACCbs at 25 °C.
Reaction of trifluoroethylamine, CF CH NH , with ACCbs
imidazole. Once again, the NH EtOH ligand was positioned over the
3
2
2
2
at pH 6.8 first proceeded smoothly with good isosbestic points,
but this was followed by a slow loss of intensity over the entire
spectrum (Figure S7, Supporting Information). Even though
the end point for the titration could not be determined, an
southeastern quadrant of the corrin ring when viewed from above.
Geometry optimizations were performed using the M06L density
37
38
functional and the SVP basis set in combination with the SVP
39,40
density fitting basis set.
(M06L was chosen to allow us to take
advantage of the performance gains afforded using the density fitting
approximate pH-independent value for log K of 0.4 could be
4
1,42
approximation.
) Convergence criteria for SCF calculations and
48
obtained from a plot of A357 vs (A − A357)/[L] (eq 5), where
0
geometry optimizations were Gaussian 09 default values. The
topological properties of the electron density (ρ) were obtained
using Bader’s Quantum Theory of Atoms in Molecules (QTAIM)
the monitoring wavelength of 357 nm is the maximum of the γ
band of the product and A , A , and L are defined as in eq 3.
0
1
43
44
The plot gives a straight line with slope = 1/K and intercept =
A1; the value of K was then corrected for pH using eq 2.
approach as implemented in AIMALL.
Values of log P, the water-octanol partition coefficient, were
45
estimated using ChemDraw Ultra.
A357 = (A − A357)/(K[L]) + A1
(5)
0
3
. RESULTS
For ACSYCbs at the same pH only the slow loss of intensity of
the spectrum was observed (Figure S7, Supporting Informa-
tion). For 0.16 M CF CH NH , the decrease in intensity was
Examples of the absorbance changes accompanying replace-
ment of coordinated H O in ACCbs and ACSYCbs are shown
2
3
2
2
in Figure S2, Supporting Information. The spectroscopic
changes for ACSYCbs are significantly smaller and span a
narrower useable wavelength range than for ACCbs; this results
in somewhat larger errors in log K values. An example of the fits
of the data to eq 3 is shown in Figure S3, Supporting
Information. Typically, about 150 wavelengths were used per
titration with ACCbs and about 100 wavelengths with
ACSYCbs to obtain weighted log K average values. These are
listed in Table 1. Examples of van’t Hoff plots obtained for the
data in this work are shown in Figure S4, Supporting
Information.
sufficiently slow to show that no coordination had occurred.
Assuming that 5% coordination of CF CH NH should be
3
2
2
detected and correcting for pH, a log K value of <−2 is
estimated.
Reaction of 2-chloroethylamine, ClCH CH NH (0.15 M),
2
2
2
with ACSYCbs at pH 9 was slow; it proceeded for ∼90 min
with good isosbestic points and then reversed itself over ∼15 h.
At the same time the pH dropped from 9.00, immediately after
addition of ClCH CH NH , to 6.99, presumably from slow
2
2
2
release of HCl during coordination of ClCH CH NH .
2
2
2
28
Methylamine, CH NH (pK of its conjugate acid = 10.65 ),
3
2
a
Reaction of NH OH with ACCbs at 25 °C, pH = 9, rapidly
is more than an order of magnitude more basic than NH and
2
3
produces a species with a band at around 460 nm which is
any of the other amines studied. ACSYCbs becomes
D
dx.doi.org/10.1021/ic302386u | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
increasingly unstable with increasing basicity (see above).
Table 2. Equilibrium Constants for Substitution of
−
Attempts to react ACSYCbs with CH NH or CH NH ·HCl in
Coordinated H O Trans to CN in Co(III) Corrins
3
2
3
2
2
pH 9 carbonate buffer at an ionic strength μ = 0.10 M, where
a
log K
∼
2% CH NH is present, were unsuccessful because large
3 2
a
28
b
ligand (L)
Py
ligand pKa
ACCbs
ACSYCbs ACCbi
amounts of CH NH (∼0.8 M for CH NH ·HCl and ∼0.2 M
3
2
3
2
4
9
c
50
for CH NH ) were needed to drive the reaction to completion
1
5.24
6.05
6.47
7.00
7.10
9.58
4.62
5.84
8.5
2.19
3.36
3.33
3.95
4.29
<−1
2.3
3.1
3
2
5
0
and the buffer capacity was insufficient to maintain a constant
pH. Increasing the buffer capacity by increasing the ionic
strength of the carbonate buffer to μ = 0.50 M still gave rise to
unacceptable changes in pH: a decrease of 0.65 log units for
CH NH ·HCl and an increase of 0.53 for CH NH . Reaction of
2
4MePy
0.38
3
4MeOPy
Imidazole
N-MeIm
49
51
4
4.15
5
2
5
1.56
4.3
d
50
6
DMAP
4.75
3
2
3
2
ACSYCbs with CH NH in pH 9 carbonate buffer, at ionic
7
NH OCH3
1.93
0.4
3
2
2
c
50
strength μ = 0.50 M (Figure S8, Supporting Information),
showed a good isosbestic point at 460 nm up to a pH of 9.2,
but at pH values greater than 9.35 a progressive shift of this
isosbestic point was observed, consistent with increasing
hydrolysis of the ester moieties of ACSYCbs.
8
CF CH NH
<−2
0.78
50
3
2
2
49
9
NH
NH
NH
NH
EtBr
2.25
2.75
1.91
1.84
2.5
2
5
0
10
11
12
13
9.25
9.52
9.68
10.65
2.94
1.52
1.74
3.15
3
2
2
EtOH
EtOCH
3
5
It has been reported that coordination of Py by ACCbs
CH
3
NH
2
3.4
4
9
a
b
occurs with log K = 2.19 and that coordination of
pK of the ligand’s conjugate acid. ACCbi = aquacyanocobinamide.
a
c
d
dimethylaminopyridine (DMAP) by ACCbi occurs with log
Estimated; see text. 4-Dimethyamino-pyridine
4
9
K = 4.75. When ACSYCbs was reacted with 0.52 M Py at pH
.2, no reaction was observed. Further addition of Py resulted
9
in turbidity of the solution. Assuming that 5% coordination of
Py should be detected, a log K value of <−1 is estimated. Both
DMAP and 4MeOPy react with ACSYCbs extremely slowly.
When ACSYCbs was reacted with 0.062 M DMAP and 0.79 M
4
MeOPy, respectively, at pH 9.2 the reaction was still
proceeding after 24 h for DMAP and 48 h for 4MeOPy. This
made it impractical to determine log K for the equilibria of
ACSYCbs with DMAP and 4MeOPy.
4
. DISCUSSION
Values of log K (T = 25 °C, eq 2) for substitution of axially
coordinated H O in ACCbs, interpolated from the van’t Hoff
2
plots where values as a function of temperature are available, or
from a determination at 25 °C only (Table 1) for the ligands
studied vary in the order L = NMeIm (4.29) > 4-MePy (3.36)
≈
4-MeOPy (3.33) > NH (2.75) > NH OCH (1.93) ≈
3
2
3
NH EtOH (1.91) > NH EtOCH (1.84) > CF CH NH (0.4).
2
2
3
3
2
2
For ACSYCbs, L = NH (2.94) > NH EtOCH (1.70) >
3
2
3
NMeIm (1.56) ≈ NH EtOH (1.52) > 4MePy (0.38) >
2
CF CH NH (<−2) (Table 1). Listed in Table 2 for purposes
3
2
2
of comparison and discussion are the log K values from Table 1,
5
,49−51
log K values for ACCbs obtained by others,
as well as
available values for the related aquacyanocobinamide (ACCbi).
The values of log K for coordination of a given ligand by
ACSYCbs are generally lower than those for its coordination by
Figure 2. Dependence of log K for substitution of H O trans to CN−
by aromatic amines (top) and aliphatic amines (bottom) on the
basicity of the ligand. Values given are for ACCbs (■), ACCbi (▲),
and ACSYCbs (○). Ligands are identified by the numbering given in
2
ACCbs, except for NH where the converse it true (log K =
3
2.75 and 2.94, respectively). The ligands fall into two classes:
those with an aromatic N donor (the pyridines and the azoles)
and those with an aliphatic N donor. The value of log K clearly
increases with the basicity of the aromatic N donor, but that
dependence is much less pronounced in the case of the
aliphatic N-donor ligands (Figure 2).
Table 2.
zero; thus, the reactions are enthalpically driven. For ACSYCbs,
the aliphatic N-donor ligands, with a harder donor atom, NH3,
NH EtOH, and NH EtOCH , also have large and negative ΔH
One of the limitations of comparing equilibrium constants at
a single temperature is that there is often a compensation effect
between ΔH and ΔS for the reaction; this has a leveling off
2
2
3
and ΔS values. In contrast, the aromatic N donors, with a softer
donor atom, NMeIm and 4MePy, have positive ΔH and ΔS
values, and the reactions are entropically driven.
17
effect on log K values. Figure S9, Supporting Information,
shows that a strong correlation does indeed exist, and that as
ΔH becomes more negative, so does ΔS.
Values of ΔS for reactions in solution are not always easy to
For the ligand substitution reactions of ACCbs, all ligands
have relatively large and negative ΔH values and large and
negative ΔS values, except for NMeIm, where ΔS is close to
interpret. Substitution of H O in ACCbs and ACSYCbs by
2
neutral ligands (L) should not give rise to significant changes in+
solvation of the complexes, as both [NC−Co(Cor)−OH2]
E
dx.doi.org/10.1021/ic302386u | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 3. Axial Bond Lengths and Topological Properties of the Electron Density at Their Bond Critical Points of Models of the
a
Cobester and Stable Yellow Cobester with Axial Ligand L
L
Co−L /Å
ρ(Co−L)
|V|/G
Co−CN /Å
ρ(Co−CN)
ACCbs
|V|/G
log K
ΔH /kJ mol−
ΔS/J K−1 mol−1
NH3
2.112
2.139
2.089
2.128
0.06455
0.06185
0.06634
0.06201
0.99112
0.99402
0.98828
0.98330
1.852
1.854
1.855
1.855
0.13453
0.13399
0.13368
0.13377
1.43234
1.42918
1.42904
1.43132
2.75
1.90
4.29
3.39
−41
−24
−23
−29
−85
−44
5
NH EtOH
2
imidazole
4Mepy
−33
ACSYCbs
NH3
2.088
2.119
2.141
2.188
0.06921
0.06544
0.05927
0.05467
1.00424
1.00119
0.97785
0.97668
1.856
1.859
1.848
1.849
0.13279
0.13178
0.13517
0.13508
1.41187
1.40691
1.42504
1.42628
2.94
1.52
1.56
0.38
−46
−79
23
−98
−236
107
24
NH EtOH
2
imidazole
4Mepy
5
aThe electron density ρ (in au) at the bond critical point is a measure of the strength of the bond;.53−55 1 au of ρ = 6.7483 e Å . The ratio of the
potential V and kinetic G energy densities at the bond critical point can be used to characterize a bond. For covalent interactions |V|/G > 2; for ionic
−3
5
6
−1
interactions, |V|/G < 1; interactions of intermediate character have 1 < |V|/G < 2. The values V and G are in au (1 au = 627.5095 kcal mol ).
+
and [NC−Co(Cor)−L] have the same charge, unlike for
anionic ligands where charge neutralization occurs in the
product. For all the charged ligands previously studied entropy
Conversely Co(III) in the cobester rather than in the stable
yellow cobester has greater affinity for the two softer aromatic
N-donor ligands and results in shorter Co−N bonds.
Therefore, the distinction between the harder Co(III) in
ACSYCbs and softer Co(III) in ACCbs, seen for anionic
1
7
increases, as expected, presumably as a result of less solvent
ordering on product formation. Solvent effects on the neutral
ligands appear not to be important since the more hydrophilic
ligands (NH , log P = −1.32; NH EtOH, −1.30; NH EtOCH ,
17
ligands, is maintained for neutral N-donor ligands.
3
2
2
3
4
5
−
0.67), which should be more solvated than the more
ASSOCIATED CONTENT
■
hydrophobic ligands (NMeIm, log P = −0.03; 4MePy, 1.14),
give rise to negative ΔS values, the opposite of what would have
been expected, as loss of solvated water from the ligand on
coordination should proceed with ΔS > 0.
For the aliphatic amines, reactions with ACSYCbs produce
more negative ΔH values than for ACCbs but for the two softer
ligands, NMeIm and 4MePy, the opposite is the case. DFT
models of the structure of the β-L,α-CN isomers of the
cobester and the stable yellow cobester, L = NH , NH EtOH,
imidazole, and 4MePy (imidazole rather than N-MeIm was
used for simplicity; substitution on N has little effect on the
*
S
Supporting Information
Spectrophotometric titration of ACCbs between pH 5.5 and
1.5; examples of the spectroscopic changes accompanying the
displacement of H O by NH in ACCbs and ACSYCbs, their
1
2
3
fits to a binding isotherm, and van’t Hoff plots; spectrophoto-
metric changes on titration of ACSYCbs with NH EtOH,
2
NH OCH , CF CH NH , and CH NH ; correlation between
−
2
3
3
2
2
3
2
ΔH and ΔS for the ligand substitution reactions of coordinated
3
2
H O in ACCbs; and illustrations of DFT models of the
2
cobester and the stable yellow cobester with various ligands.
This material is available free of charge via the Internet at
http://pubs.acs.org.
δ
donor properties of N , pK = 7.00 and 7.10, respectively), were
examined. They are illustrated in Figure S10, Supporting
α
a
Information, and the results are given in Table 3.
AUTHOR INFORMATION
■
*
The presence of a lactone in the stable yellow cobester
certainly increases the steric crowding of the β face, and there is
a close contact between an H on the ligand and the lactone O
Corresponding Author
E-mail: Susan.Chemaly@wits.ac.za (S.M.C.); Helder.
Marques@wits.ac.za (H.M.M.).
(Figure S10, Supporting Information). This, however, does not
appear to unduly influence the ability of the metal to coordinate
the ligand. For example, modeling shows that the Co−N ligand
Notes
The authors declare no competing financial interest.
to NH EtOH, which gives the closest steric contact with the
2
lactone, is shorter in its complex with the stable yellow cobester
than with the cobester itself (Table 3). Therefore, electronic
factors appear to override these steric factors.
ACKNOWLEDGMENTS
■
The financial assistance of the Department of Science and
Technology, the National Research Foundation, Pretoria,
through SARChI, the South African Research Chairs Initiative,
and the University of the Witwatersrand, Johannesburg, is
gratefully acknowledged.
The NH and NH EtOH complexes of the stable yellow
3
2
cobester have shorter and stronger (larger ρ values) Co−N
bonds than those of the cobester. These stronger bonds have
less ionic character (although the difference is very small) and
cause a small elongation of the trans Co−CN bond. By
contrast, the Co−N bond to the aromatic N donors is shorter,
stronger, and somewhat less ionic in the complexes of the
cobester than in those of the stable yellow cobester. The
modeling therefore rationalizes the experimental ΔH values and
suggests that the reason why reactions of the aliphatic amines
with ACSYCbs produce more negative ΔH values than the
reactions with ACCbs is a greater affinity of the metal for the
harder N-donor ligand which leads to a shorter Co−N bond.
REFERENCES
■
(
1) Chemaly, S. M.; Brown, K. L.; Fernandes, M. A.; Munro, O. Q.;
Grimmer, C.; Marques, H. M. Inorg. Chem. 2011, 50, 8700−8718.
(
(
5
(
2) Randall, W. C.; Alberty, R. A. Biochemistry 1967, 6, 1520−1525.
3) Reenstra, W. W.; Jencks, W. P. J. Am. Chem. Soc. 1979, 101,
780−5791.
4) Balt, S.; van Herk, A. M. Transition Met. Chem. 1983, 8, 152−154.
(5) Baldwin, D. A.; Betterton, E. A.; Pratt, J. M. J. Chem. Soc., Dalton
Trans. 1983, 2217−2222.
F
dx.doi.org/10.1021/ic302386u | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(
1
(
6) Marques, H. M.; Brown, K. L.; Jacobsen, D. W. J. Biol. Chem.
988, 263, 12378−12383.
7) Marques, H. M.; Egan, T. J.; Marsh, J. H.; Mellor, J. R.; Munro,
O. Q. Inorg. Chim. Acta 1989, 166, 249−255.
8) Stochel, G.; van Eldik, R.; Kunkely, H.; Vogler, A. Inorg. Chem.
(36) Hannibal, L.; Bunge, S. D.; van Eldik, R.; Jacobsen, D. W.;
Kratky, C.; Gruber, K.; Brasch, N. E. Inorg. Chem. 2007, 46 (9), 3613−
3618.
(37) Zhao, Y.; Truhlar, D. G. J. Chem. Phys. 2006, 125, 194101.
(38) Schaefer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97,
2571.
(39) Eichkorn, K.; Treutler, O.; Ohm, H.; Ahser, M.; Ahlrichs, R.
Chem. Phys. Lett. 1995, 240, 283.
(40) Eichkorn, K.; Weigend, F.; Treutler, O.; Ahlrichs, R. Theor.
Chem. Acc. 1997, 97, 119.
(41) Dunlap, B. I. J. Chem. Phys. 1983, 78, 3140−3142.
(42) Dunlap, B. I. J. Mol. Struct. (THEOCHEM) 2000, 529, 37−40.
(43) Bader, R. F. Atoms in Molecules: A Quantum Theory; Oxford
University Press: Oxford, 1990.
(44) Keith, T. A. AIMAll (Version 11.12.19), 08.05.04; TK Gristmill
Software: Overland Park, KS, 2011; http://aim.tkgristmill.com.
(45) ChemDraw Ultra, 6.0.1; CambridgeSoft.com: Cambridge, MA,
2000.
(
1
(
(
989, 28, 4314−4318.
9) Stochel, G.; van Eldik, R. Inorg. Chem. 1990, 29, 2075−2077.
10) Marques, H. M. J. Chem. Soc., Dalton Trans. 1991, 6, 1437−
1
442.
(
11) Marques, H. M.; Bradley, J. C.; Campbell, L. A. J. Chem. Soc.,
Dalton Trans. 1992, 2019−2027.
12) Waddington, M. D.; Finke, R. G. J. Am. Chem. Soc. 1993, 115,
629−4640.
13) Marques, H. M.; Knapton, L. J. Chem. Soc., Dalton Trans. 1997,
0, 3827−3833.
14) Brown, K. L.; Cheng, S.; Zou, X.; Zubkowski, J. D.; Valente, E.
J.; Knapton, L.; Marques, H. M. Inorg. Chem. 1997, 36, 3666−3675.
15) Hamza, M. S. A.; Zou, X.; Brown, K. L.; van Eldik, R. Inorg.
Chem. 2001, 40, 5440−5447.
16) Hamza, M. S. A.; Elawady, M. A.; Marques, H. M. S. Afr. J.
Chem. 2008, 61, 68−73.
17) Chemaly, S. M.; Florczak, M.; Dirr, H.; Marques, H. M. Inorg.
(
4
(
2
(
(
(46) Toerien, J. G. MSc Dissertation, University of the
Witwatersrand, Johannesburg, 1986.
(
(
2
(
47) Gutier
011, 50, 8817−8825.
48) Newton, T. W.; Arcand, G. M. J. Am. Chem. Soc. 1953, 75,
́
rez, M. M.; Olabe, J. A.; Amorebieta, V. T. Inorg. Chem.
(
Chem. 2011, 50, 8719−8727.
(
(
E. J.; Marques, H. M. Eur. J. Inorg. Chem. 2003, 2095−2107.
(
2
(
(
449−2453.
18) Knapton, L.; Marques, H. M. Dalton Trans. 2005, 5, 889−895.
19) Perry, C. B.; Fernandes, M. A.; Brown, K. L.; Zou, X.; Valente,
49) Hamza, M. S. A. J. Inorg. Biochem. 1998, 69, 269−274.
50) Hamza, M. S. A.; Pratt, J. M. J. Chem. Soc., Dalton Trans. 1994,
1
(
1
(
1
(
377−1382.
51) Hamza, M. S. A.; Pratt, J. M. J. Chem. Soc., Dalton Trans. 1994,
373−1376.
52) Hamza, M.; Pratt, J. J. Chem. Soc., Dalton Trans. 1994, 1367−
371.
53) Espinosa, E.; Souhassou, M.; Lachekar, H.; Lecomte, C. Acta
20) Day, P. Coord. Chem. Rev. 1967, 2, 99−108.
(
21) Giannotti, C. In B ; Dolphin, D., Ed.; Wiley: New York, 1982;
12
pp 393−430.
(
22) Pratt, J. M. In Chemistry and Biochemistry of B ; Banerjee, R.,
12
Ed.; John Wiley & Sons: New York, 1999; pp 113−164.
(
23) Navizet, I.; Perry, C. B.; Govender, P. P.; Marques, H. M. J. Phys.
Crystallogr., Sect. B 1999, 55, 563−572.
Chem. B. 2012, 116, 8836−8846.
(
(
54) Grabowski, S. J. J. Phys. Chem. A 2000, 105, 5551−5557.
55) Bader, R. F. W.; Matta, C. F. J. Phys. Chem. A 2004, 108, 8385−
(
(
24) Perry, C. B.; Marques, H. M. S. Afr. J. Chem. 2005, 58, 9−15.
25) Marques, H. M.; Knapton, L.; Zou, X.; Brown, K. L. J. Chem.
8
394.
Soc., Dalton Trans. 2002, 3195−3200.
26) Gossauer, A.; Gruning, B.; Ernst, L.; Becker, W.; Sheldrick, W. S.
Angew. Chem., Int. Ed. 1977, 16, 481−482.
27) Gruning, B.; Holze, G.; Jenny, T.; Nesvadba, P.; Gossauer, A.;
Ernst, L.; Sheldrick, W. S. Helv. Chim. Acta 1985, 68, 1754−1770.
28) Martell, A. E.; Smith, R. M.; Motekaites, R. J. NIST Criticially
(56) Espinosa, E.; Alkorta, I.; Elguero, J.; Molins, E. J. Chem. Phys.
(
̈
2
002, 117, 5529−5542.
(
̈
(
Selected Stability Constants of Metal Complexes, V. 8.0; NIST:
Gaithersburg, MD, 2004.
(
3
(
29) Munro, O. Q.; Marques, H. M. Inorg. Chem. 1996, 35, 3752−
767.
30) Markwell, A. J.; Pratt, J. M.; Shaikjee, S. S.; Toerien, J. G. J.
Chem. Soc., Dalton Trans. 1987, 1349−1357.
(
(
31) Johnson, G. D.; Bowen, R. E. J. Am. Chem. Soc. 1965, 87, 1655.
32) Marques, H. M.; Munro, O. Q.; Crawcour, M. L. Inorg. Chim.
Acta 1992, 196, 221−229.
(
33) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J., J. A.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
̈
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision A.02; Gaussian, Inc.: Wallingford, CT, 2009.
(
34) Stewart, J. P. J. Mol. Model. 2007, 13, 1173−213.
(
35) Marques, H. M.; Brown, K. L. Coord. Chem. Rev. 1999, 190−
1
92, 127−153.
G
dx.doi.org/10.1021/ic302386u | Inorg. Chem. XXXX, XXX, XXX−XXX