Cis-influence of hydroporphyrin macrocycles on axial ligation equilibria and alkyl exchange reactions of alkylcobalt(III) porphyrin complexes.

Article abstract of DOI:10.1021/ic991304t

Stability constants are reported for the coordination of pyridine and substituted pyridines to the alkylcobalt(III) complexes of octaethylporphyrin (OEP), t-octaethylchlorin (OEC), and ttt-octaethylisobacteriochlorin (OEiBC) in toluene solution. The stability constants correlate with the base strength of the nitrogenous ligand. A cis-influence of the macrocycle saturation level on the stability constants is observed. Stability constants for coordination of a given pyridine ligand to an alkylcobalt(III) complex are roughly 10 times smaller than the stability constants for the corresponding cobalt(II) complex. Analysis of a thermodynamic cycle demonstrates that this leads to decreased stability of the complex with respect to Co-C bond homolysis upon ligand coordination, a base-on effect. Alkyl exchange occurs between cobalt complexes of different tetrapyrroles. Equilibrium data establish that the exchange is nonstatistical and that the Co-C bond is stabilized by increasing the saturation of the tetrapyrrole macrocycle.

Full text of DOI:10.1021/ic991304t

1518
Inorg. Chem. 2000, 39, 1518-1524
Cis-Influence of Hydroporphyrin Macrocycles on Axial Ligation Equilibria and Alkyl
Exchange Reactions of Alkylcobalt(III) Porphyrin Complexes
Alan M. Stolzenberg* and Jack S. Summers
Department of Chemistry, West Virginia University, Morgantown, West Virginia 26506
ReceiVed NoVember 8, 1999
Stability constants are reported for the coordination of pyridine and substituted pyridines to the alkylcobalt(III)
complexes of octaethylporphyrin (OEP), t-octaethylchlorin (OEC), and ttt-octaethylisobacteriochlorin (OEiBC)
in toluene solution. The stability constants correlate with the base strength of the nitrogenous ligand. A cis-
influence of the macrocycle saturation level on the stability constants is observed. Stability constants for coordination
of a given pyridine ligand to an alkylcobalt(III) complex are roughly 10 times smaller than the stability constants
for the corresponding cobalt(II) complex. Analysis of a thermodynamic cycle demonstrates that this leads to
decreased stability of the complex with respect to Co-C bond homolysis upon ligand coordination, a “base-on”
effect. Alkyl exchange occurs between cobalt complexes of different tetrapyrroles. Equilibrium data establish
that the exchange is nonstatistical and that the Co-C bond is stabilized by increasing the saturation of the
tetrapyrrole macrocycle.
Introduction
Coordination equilibria are extremely important in the
chemistry of vitamin B12, a biologically essential cobalt tetra-
pyrrole complex. Two such equilibria markedly affect the Co-C
bond dissociation energies of coenzyme B12 and of B12 model
complexes. Complexation of adenosylcobalamin by apoenzyme
results in a dramatic acceleration of the homolysis of the Co-C
bond by a factor of g1010 relative to the case of free
coenzyme.4,5 The “base-on” effect, which results from coordina-
tion of an axial ligand, accelerates the homolysis of the Co-C
bond by a factor of 102-103.4-11 Although the manner in which
these accelerations occur is not understood, they are thought to
involve steric interactions and distortions induced by coordina-
tion of the axial ligand or of the coenzyme to the enzyme.
An observation seemingly contradictory to the “base-on”
effect, though, is that, for axial bases of similar steric sizes,
increasing the strength of the base increases the apparent strength
of the Co-C bond in (R-phenethyl)bis(dimethylglyoximato)-
cobalt(III) complexes.12,13 It was suggested that more basic
ligands stabilized higher oxidation states and thus favored the
six-coordinate alkyl-Co(III)-base complex over the five-
coordinate Co(II)-base complex.12 Our structural studies of
Co(OEP)(CH3)Py cast doubt upon this explanation.14 The cobalt
ion of the six-coordinate complex appeared to be rather electron
rich. Moreover, the Co-Nax distances suggested that ligand
Metal complexes of hydroporphyrins and other non-porphyrin
tetrapyrroles play central roles as prosthetic groups in the
biochemical pathways of the carbon, nitrogen, and sulfur cycles
and in the metabolism of many anaerobes.1 The absence of
enzymes that contain porphyrin prosthetic groups and are
competent to catalyze these processes is conspicuous and raises
the question of whether non-porphyrin tetrapyrroles are specif-
ically required. Thus, there is considerable interest in delineating
the effects of changes in the structure and saturation level of a
tetrapyrrole macrocycle on the chemistry of its complexes and
in particular on the chemistry of a coordinated metal ion.
As part of our studies of the chemical differences between
porphyrins and hydroporphyrins,1 we investigated the effects
of the macrocycle saturation level on the ligand-binding
equilibria of four-coordinate cobalt(II) and zinc(II) porphyrin
and hydroporphyrin complexes.2 The macrocycle saturation level
exerted a small but real cis-influence on the equilibria. Stability
constants for coordination of a given ligand to OEiBC3
complexes were typically 4 times greater than those for
coordination to OEP complexes and 1.8 times greater than those
for coordination to OEC complexes. For most ligands, the
stability constant was greater for the zinc complexes than for
the cobalt complexes. However, for the weak base 3,5-
dichloropyridine, the stability constant was greater for cobalt
complexes. This reversal was attributed to a relief of strain due
to core expansion that occurs upon ligand coordination to cobalt
complexes.
(4) Finke, R. G.; Hay, B. P. Inorg. Chem. 1984, 23, 3041-3043.
(5) Hay, B. P.; Finke, R. G. J. Am. Chem. Soc. 1986, 108, 4820-4829.
(6) Hay, B. P.; Finke, R. G. J. Am. Chem. Soc. 1987, 109, 8012-8018.
(7) Grate, J. H.; Schrauzer, G. N. J. Am. Chem. Soc. 1979, 101, 4601-
4611.
(8) Schrauzer, G. N.; Grate, J. H. J. Am. Chem. Soc. 1981, 103, 541-
(1) See ref 2 for reviews of the biological roles of metallohydroporphyrin
complexes and comparisons of the properties of hydroporphyrin and
porphyrin complexes.
(2) Summers, J. S.; Stolzenberg, A. M. J. Am. Chem. Soc. 1993, 115,
10559-10567.
(3) Abbreviations: P, any tetrapyrrole dianion; OEP, 2,3,7,8,12,13,17,18-
octaethylporphyrin dianion; OEC, trans-2,3-dihydro-2,3,7,8,12,13,17,
18-octaethylporphyrin (chlorin) dianion; OEiBC, 2,3,7,8-tetrahydro-
2,3,7,8,12,13,17,18-octaethylporphyrin (isobacteriochlorin) dianion
(mixture of tct- and ttt-diastereomers unless otherwise noted); Py,
pyridine; 3,5-Cl2Py, 3,5-dichloropyridine; 4-CNPy, 4-cyanopyridine;
4-Pic, 4-picoline; 4-DMAP, 4-(dimethylamino)pyridine.
546.
(9) Brown, K. L.; Brooks, H. B. Inorg. Chem. 1991, 30, 3420-3430.
(10) Chemaly, S. M.; Pratt, J. M. J. Chem. Soc., Dalton Trans. 1980, 2274-
2281.
(11) Blau, R. J.; Espenson, J. H. J. Am. Chem. Soc. 1985, 107, 3530-
3533.
(12) Ng, F. T. T.; Rempel, G. L.; Halpern, J. J. Am. Chem. Soc. 1982,
104, 621-623.
(13) Ng, F. T. T.; Rempel, G. L.; Mancuso, C.; Halpern, J. Organometallics
1990, 9, 2762-2772.
(14) Summers, J. S.; Petersen, J. L.; Stolzenberg, A. M. J. Am. Chem. Soc.
1994, 116, 7189-7195.
10.1021/ic991304t CCC: $19.00 © 2000 American Chemical Society
Published on Web 03/10/2000

Reactions of Alkylcobalt(III) Porphyrin Complexes
Inorganic Chemistry, Vol. 39, No. 7, 2000 1519
binding was substantially stronger in the five-coordinate Co(II)
complex than in the six-coordinate Co(III) complex.
We report the stability constants for coordination of a series
of pyridine bases to the five-coordinate alkyl-Co(III) complexes
of OEP, OEC, and OEiBC in this paper. As in the case of the
four-coordinate Co(II) complexes, the macrocycle saturation
level exerted a small but real cis-influence on the stability
constants. Moreover, the saturation level also influenced the
Co-C bond dissociation energy. This was established by the
nonstatistical equilibria of the facile alkyl transfer reaction, eq
1, that we discovered during the course of this investigation.
CoIII(P)R + CoII(P*) h CoII(P) + CoIII(P*)R
(1)
Comparisons of the relative magnitudes of the equilibrium
constants for base coordination to four-coordinate Co(II) and
five-coordinate alkyl-Co(III) complexes and their dependencies
on base strengths provide insights into both the apparent
weakening of the Co-C bond upon base coordination (“base-
on” effect) and the apparent strengthening of the Co-C bond
with increasing base strength.
Experimental Section
Figure 1. Plot of (δA - δobs)/(δobs - δAD) for the axial methyl
resonance of Co(OEC)CH3 vs (δA - δobs)/(δobs - δAD) for the axial
methyl resonance of Co(OEP)CH3 in a competitive titration with
DMAP. Error bars indicate the estimated error in the ratio for Co-
(OEC)CH3 (y value) for each data point. The estimated error in the
ratio for Co(OEP)CH3 (x value) is not shown but is comparable in
magnitude to the y error on a percent basis.
Materials. Reactions, recrystallizations, and sample manipulations
were carried out on a vacuum line, under a nitrogen atmosphere using
standard Schlenk techniques, or in a Vacuum/Atmospheres drybox.
Alkylcobalt(III) complexes were handled under subdued lights. All
solvents, reagents, and bases were of reagent grade or better. Toluene
was distilled from sodium, tetrahydrofuran (THF) was distilled from
sodium benzophenone ketyl, and chloroform was distilled from
phosphorus pentoxide. Pyridine and 4-picoline were distilled from
barium oxide, and 3,5-dichloropyridine was sublimed. 4-(Dimethyl-
amino)pyridine (DMAP) and 4-cyanopyridine were used as received.
Alkyl halides were examined for purity and were distilled if necessary.
The free-base compounds H2(OEP),15 H2(OEC),16 and H2(OEiBC)16
and their respective Co(II) complexes17,18 were prepared by literature
methods. The tct- and ttt-stereoisomers of H2(OEiBC) were separated
by chromatography on MgO.19 The Co(II) complexes of the separated
isomers were prepared by the same method as used for the mixture.
Alkylcobalt(III) porphyrin and hydroporphyrin complexes were
prepared by reaction of the respective anionic cobalt(I) complex with
excess alkyl iodide.20 Solutions of the cobalt(II) complex in THF were
reduced by stirring over 1% sodium amalgam (18 h for Co(OEP) and
Co(OEC); 1 h for Co(OEiBC)). After separation from the amalgam by
filtration through a fritted glass disk, the solution was frozen and the
reaction vessel was evacuated. The alkyl iodide was distilled onto the
frozen solution, the vessel was filled with nitrogen, and the solution
was allowed to thaw and then was stirred for 1 h at room temperature.
The solvent was removed under vacuum. The solid was brought into
a drybox and dissolved in a minimal quantity of benzene. The resulting
solution was applied to the top of a column of basic alumina that had
been deactivated with THF (3 mL/100 g of alumina). The product was
eluted with a 4:6 benzene/hexane mixture. The solvent was removed
and the residue recrystallized from toluene. Acetylcobalt(III) porphyrin
and hydroporphyrin complexes were prepared by an analogous
procedure substituting acetyl chloride for the alkyl iodide.
Instrumentation. Absorption spectra were recorded on a Perkin-
Elmer Lambda 6 UV/vis spectrophotometer. The temperature was
controlled to (0.1 °C with a Haake F3-CH circulating bath. Proton
magnetic resonance spectra were obtained on a JEOL-GSX 270 MHz
NMR spectrometer.
Equilibrium Constants for Coordination of Axial Ligands.
Stability constants for coordination of substituted pyridines to Co(OEP)-
CH3 were determined spectrophotometrically. The procedures employed
and the use of the program SQUAD to calculate stability constants
and their associated errors were described previously.2 The stability
constants for coordination of substituted pyridines to other alkylcobalt
1
porphyrin and hydroporphyrin complexes were determined by an H
NMR method. The experiment entailed the titration of a physical
mixture of the complex of interest and Co(OEP)CH3 with the substituted
pyridine. If δA is defined as the chemical shift of a particular nucleus
in the uncomplexed acceptor A, δAD as the chemical shift of the nucleus
in the completely formed complex AD, and δobs as the observed
chemical shift of the nucleus at a specific concentration of donor D
and if A and AD are in fast exchange, then a plot of (δA - δobs)/(δobs
- δAD) for the resonances associated with a particular nucleus in one
metal complex versus the same function for a resonance associated
with the second complex will be linear with slope equal to the ratio of
the equilibrium constants, eq 2. A representative plot obtained for
δA′ - δobs′
δobs′ - δAD′ KAD δobs - δAD
KAD′ δA - δobs
)
(2)
titration of a mixture of Co(OEC)CH3 and Co(OEP)CH3 with DMAP
is shown in Figure 1. Note that eq 2 does not require the concentrations
of the metal complexes or the donor to be known. Thus, it is not
necessary to satisfy the condition [D]0 . [AD] to obtain useful data.
The derivation of eq 2 is described in the Supporting Information.
Equilibrium Constants for Alkyl Exchange Reactions. Alkyl
exchange reactions were conducted by reacting an alkylcobalt(III)
complex of one tetrapyrrole macrocycle with a cobalt(II) complex of
a second tetrapyrrole macrocycle. Reactions were conducted anaero-
bically in toluene-d8, in which the complexes are five-coordinate and
four-coordinate, respectively. Reactions were also conducted in 0.5 M
(15) Sessler, J. L.; Mozaffari, A.; Johnson, M. R. Org. Synth. 1992, 70,
68-78.
(16) Stolzenberg, A. M.; Spreer, L. O.; Holm, R. H. J. Am. Chem. Soc.
1980, 102, 364-370.
(17) Bonnett, R.; Dimsdale, M. J. J. Chem. Soc., Perkin Trans. 1 1972,
2540-2548.
(18) Lahiri, G. K.; Summers, J. S.; Stolzenberg, A. M. Inorg. Chem. 1991,
30, 5049-5052.
(19) Sullivan, E. P., Jr.; Grantham, J. D.; Thomas, C. S.; Strauss, S. H. J.
Am. Chem. Soc. 1991, 113, 5264-5270.
(20) Ogoshi, H.; Watanabe, E.-i.; Koketsu, N.; Yoshida, Z.-i. Bull. Chem.
Soc. Jpn. 1976, 49, 2529-2536.

1520 Inorganic Chemistry, Vol. 39, No. 7, 2000
Stolzenberg and Summers
pyridine-d5 in toluene-d8. Under these conditions, complexation equi-
libria for coordination of pyridine to alkyl-Co(III) complexes of OEP,
OEC, and OEiBC were 97%, 99%, and >99% complete, respectively,
and those of cobalt(II) complexes were >99% complete.
The reaction apparatus consisted of a glass tee connected to a 90°
Teflon barrel valve on each horizontal arm and a 180° Teflon barrel
valve on the vertical arm. One horizontal arm had a tube attached below
the valve. The other had a male 14/20 standard taper joint below the
valve. An NMR tube blown onto a female 14/20 joint was connected
to the male joint. The 180° valve was attached to an O-ring joint, which
was used to connect the apparatus to a vacuum line.
Table 1. 1H NMR Data for Axial R Groups of RCoIII(P)L
Complexesa
δ
R
P
L
R-CH
â-CH
CH3
OEP
OEC
ttt-OEiBC
-4.60
Py -4.73
-3.10
Py -3.30
-1.20
Py -1.38
tct-OEiBC Py -1.38, -1.56b
OEP
OEC
ttt-OEiBC
tct-OEiBC
OEP
OEC
ttt-OEiBC
The apparatus was assembled in a drybox. A small stir bar and
approximately 0.5 mg each of an alkylcobalt(III) complex and a
cobalt(II) complex of different tetrapyrrole macrocycles were placed
in the tube through the body of the valve. The Teflon barrel was
replaced and closed, the apparatus was removed from the drybox,
attached to a vacuum line, and evacuated, and 1.0 g of toluene-d8 was
bulb-to-bulb distilled into the tube. Upon thawing, the solids dissolved
and the resulting solution was stirred for 30 min at room temperature.
The apparatus was then tipped to allow half of the solution to pour
through the two 90° valves into the NMR tube. The valves were closed,
the solution in the NMR tube was degassed by repeated freeze-pump-
thaw cycles, and the NMR tube was sealed off under vacuum. A new
NMR tube was connected to the male 14/20 joint, the apparatus was
reevacuated, and the remaining solution was poured into the NMR tube
and frozen. An aliquot of pyridine-d5 sufficient to make the solution
0.5 M in pyridine was bulb-to-bulb transferred into the NMR tube.
The contents of the NMR tube were subjected to repeated freeze-
pump-thaw cycles, and the NMR tube was sealed off under vacuum.
Equilibrium constants for alkyl exchange were calculated from the
ratios of the integrations of selected 1H resonances of the two
diamagnetic alkylcobalt(III) complexes and the two paramagnetic
cobalt(II) complexes, weighted appropriately for the number of protons
in these resonances. Assignment of the resonances in the spectra of
the mixtures was straightforward, given that the four individual
complexes had been independently prepared and spectroscopically
characterized. Spectra were monitored for several hours to ensure that
equilibrium had been achieved. Reaction solutions involving methyl
complexes appeared to be stable indefinitely. In contrast, isopropyl
complexes decomposed over several hours to give cobalt(II) complexes
and undetermined organic product(s) derived from the alkyl group.
Identical equilibrium constants were obtained when equilibrium was
approached by reacting the alkylcobalt(III) complex of the second
tetrapyrrole and the cobalt(II) complex of the first tetrapyrrole.
C2H5
-3.60 (br m)
-2.52 (br q)
-0.52 (br q)
-5.14 (t)
-4.43 (t)
-3.03 (t)
-0.52, -0.73 (br q) -3.14, -3.19 (t)
i-C3H7
-3.50 (m, 1H)
-5.24 (d, 6H)
-4.20, -4.22 (d, 3H)
-2.97, -3.02 (d, 3H)
-2.83
-1.81
-0.55
-2.13 (br m, 1H)
-0.25 (br m, 1H)
CH3C(O) OEP
OEC
ttt-OEiBC
a In ppm relative to TMS at 21 °C. Solvent is 0.5 M pyridine-d5 in
toluene-d8 if L ) Py and toluene-d8 if no L is specified. b Present in a
41:54 ratio.
Table 2. 1H NMR Data for Meso CH Protons of RCoIII(P)L
Complexesa
R
P
L
δ
CH3
OEP
10.21
10.11
Py
Py
Py
t-OEC
9.65, 9.63, 8.66b
9.67, 9.44, 8.48, 8.38
8.98, 7.92, 7.91, ...c
8.91, 7.83, 7.81, ...c
9.99
ttt-OEiBC
CH3C(O)
OEP
t-OEC
9.40, 9.34, 8.52, 8.47
a In ppm relative to TMS at 21 °C. Solvent is 0.5 M pyridine-d5 in
toluene-d8 if L ) Py and toluene-d8 if no L is specified. b Two nearly
coincident singlets. c Peak obscured by the solvent residual proton peak.
halogen.28 Hydroporphyrin macrocycles suffer oxidative dehy-
drogenation under these conditions and are converted to
porphyrin.16 Thus, we pursued the second method.
Reduction of cobalt(II) hydroporphyrin complexes to their
cobalt(I) anions and alkylation of the anions with alkyl or acyl
halides proceeded readily. The handling characteristics of the
hydroporphyrin complexes suggest that the stability and lability
of their cobalt-carbon bonds are comparable to those of the
corresponding porphyrin complexes. Although no specific
instances of photolability were observed, all work was conducted
under subdued lighting as a general precaution. Rigorous
exclusion of O2 was a necessity. All complexes react readily
with O2 to afford (alkylperoxo)cobalt(III) complexes.24,29 Given
the susceptibility of hydroporphyrin complexes to oxidative
dehydrogenation, they are somewhat less robust than porphyrin
complexes. Particular care was exercised in the isolation and
purification of the hydroporphyrin complexes.
Reduction of the porphyrin macrocycle has several effects
on the 1H NMR spectra of the alkyl- and acylcobalt(III)
complexes, Tables 1 and 2. The axial alkyl and acyl proton
peaks shift downfield and the meso proton peaks shift upfield
with increasing saturation of the macrocycle. The invariant
stretching frequency of 1720-1722 cm-1 for the axial acetyl
Results
1
Synthesis and H NMR Spectroscopy of Alkylcobalt(III)
Tetrapyrrole Complexes. Alkylcobalt(III) porphyrin complexes
have been prepared by two methods: (i) reaction of a CoIII(P)X
complex (where P ) a porphyrin dianion and X ) a halide
anion) with a Grignard or alkyllithium reagent;21-23 (ii) alkyl-
ation of a CoI(P) complex with an alkyl halide or other
electrophile.20,21,24,25 Application of the first method to hydro-
porphyrins appeared problematic. CoIII(P)X complexes are
prepared from CoII(P) complexes by air oxidation in the presence
of HX17,26,27 or in the case of X ) I by oxidation with elemental
(21) Clarke, D. A.; Dolphin, D.; Grigg, R.; Johnson, A. W.; Pinnock, H.
A. J. Chem. Soc. C 1968, 881-885.
(22) Callot, H. J.; Metz, F. J. Chem. Soc., Chem. Commun. 1982, 947-
948.
(23) Callot, H. J.; Metz, F.; Cromer, R. NouV. J. Chim. 1984, 8, 759-763.
(24) Perree-Fauvet, M.; Gaudemer, A.; Boucly, P.; Devynck, J. J. Organo-
met. Chem. 1976, 120, 439-451.
(25) Dolphin, D.; Halko, D. J.; Johnson, E. Inorg. Chem. 1981, 20, 4348-
4351.
(26) Setsune, J.-i.; Saito, Y.; Ishimaru, Y.; Ikeda, M.; Kitao, T. Bull. Chem.
Soc. Jpn. 1992, 65, 639-648.
(27) Johnson, A. W.; Kay, I. T. J. Chem. Soc. 1960, 2979-2983.
(28) Cao, Y.; Petersen, J. L.; Stolzenberg, A. M. Inorg. Chem. 1998, 37,
5173-5179.
(29) Kendrick, M. J.; Al-Akhdar, W. Inorg. Chem. 1987, 26, 3971-3972.

Reactions of Alkylcobalt(III) Porphyrin Complexes
Inorganic Chemistry, Vol. 39, No. 7, 2000 1521
Table 3. Equilibrium Constants for Coordination of Bases by
Table 4. Ratios of Equilibrium Constants for Coordination of Bases
by Alkylcobalt(III) Complexes Relative to Co(OEP)CH3
Methylcobalt(III) Complexesa
a
log Kb
ratio
basec
pKad Co(OEP)CH3 Co(t-OEC)CH3 Co(ttt-OEiBC)CH3
base
Co(OEC)CH3 Co(ttt-OEiBC)CH3 Co(OEP)C2H5
3,5-Cl2Py 0.67
1.531(26)
1.627(39)
1.759(40)
2.063(37)
2.402(25)
2.06
2.33
2.96
2.47
2.76
3.42
3,5-Cl2Py
Py
4-DMAP
3.40
3.70
3.60
8.70
10.00
10.30
4-CNPy
Py
4-Pic
1.90
5.22
6.02
0.21
a Determined by the NMR method in toluene-d8 at 20 °C. Estimated
error ) 0.1.
4-DMAP 9.71
a At 20.0 ( 0.1 °C in toluene. b Numbers in parentheses are estimated
standard deviations in the least significant digits. c See ref 3 for
abbreviations. d Aqueous for BH+; see Table 1 of ref 2 for sources.
groups suggests that these shifts reflect the diminished π-systems
and smaller ring currents of the hydroporphyrin macrocycles
relative to porphyrins rather than differences in electron densities
at cobalt. The lower symmetry of the hydroporphyrin macro-
cycle has two consequences. First, the peaks due to the now
inequivalent macrocycle protons become sufficiently great in
number and complexity that individual peaks are often difficult
to resolve and assign. Typically, alkylcobalt(III) hydroporphyrin
pyrroline â-H, CH2, and CH3 peaks are found in regions near
4.0, 1.8, and 0.9 ppm and pyrrole CH2 and CH3 peaks are found
near 3.4-4.0 and 1.5-1.8 ppm, respectively. Second, addition
of a proton to a â-carbon of an octaalkylporphyrin creates an
asymmetric center and results in stereoisomerism. The method
used to reduce octaethylporphyrin specifically affords hydro-
porphyrins with trans ethyl substituents on the pyrroline
(reduced) ring. Both faces of the macrocycle are equivalent in
t-OEC, which is a racemic mixture. The OEiBC obtained is a
mixture of the ttt- and tct-diastereomers, where the middle letter
designates the relative stereochemistry of the “inboard” ethyl
substitutents (3 and 7) on adjacent pyrroline rings. The ttt-
diastereomer is a racemic mixture and has equivalent macrocycle
faces. The tct-diasteromer is a meso compound and has in-
equivalent macrocycle faces. Not unexpectedly, the two equiva-
lent methyls of the isopropyl group are diastereotopic in
CoIII(OEP)(i-Pr) and resolved in CoIII(OEC)(i-Pr) and CoIII-
(OEiBC)(i-Pr). Two or three sets of alkyl peaks were resolved
in the spectra of CoIII(OEiBC)CH3 and CoIII(OEiBC)Et prepared
from the mixture of OEiBC diastereomers. Consequently, all
subsequent work was performed with complexes prepared from
the separated diastereomers of H2(OEiBC) in order to simplify
the spectra. In addition, equilibrium measurements were made
only with ttt-Co(OEiBC) complexes to eliminate statistical
effects resulting from multiple species.
Figure 2. Correlations between log K in toluene and ligand basicity
(aqueous pKa of conjugate acid) for CH3Co(OEP), (circles), CH3Co-
(OEC), (diamonds), and CH3Co(OEiBC) (triangles).
stability constants. Consequently, an NMR method that involved
titration of a mixture of the complex of interest and Co(OEP)-
CH3 with the substituted pyridine was used to determine the
stability constants for the complex relative to the corresponding
stability constant for Co(OEP)CH3. As shown in Figure 1, a
plot of (δA - δobs)/(δobs - δAD) for the resonances associated
with a particular nucleus in one metal complex versus the same
function for a resonance associated with a second complex is
linear. The slope of the line is equal to the ratio of the
equilibrium constants. The method has the advantage that the
presence of impurities is readily apparent and does not interfere
with the analysis of the data. Moreover, the concentrations of
complex and pyridine need not be known to high precision to
obtain meaningful results. The significant change in chemical
shift of the axial alkyl protons upon pyridine coordination makes
them particularly well suited to the method. With the exception
of 3,5-Cl2Py and Co(OEP)CH3, equilibria could be driven to
completion to allow direct determination of δAD. In this case,
the δAD values were fitted numerically to give the most linear
plot. The ratios of stability constants determined by this method
are reported in Table 4. Stability constants calculated from the
ratios and the spectrophotometrically determined stability
constants for Co(OEP)CH3 are included in Table 3.
The effect of the variation of the base strength of the axial
ligand on the ligand-binding equilibria of Co(P)CH3 complexes,
eq 3, is shown in Figure 2, which is a plot of log K vs the
aqueous pKa of the ligand’s conjugate acid. The lines represent-
ing the response of the OEP, OEC, and OEiBC equilibria to
ligand basicity are nearly parallel and are displaced toward larger
log K values with progressive saturation of the macrocycle. The
increase in stability constants represents a ∆∆G relative to OEP
complexes of about 0.71-0.77 kcal/mol for t-OEC complexes
and of 1.26-1.37 kcal/mol for ttt-OEiBC complexes. Similar
behavior was observed previously for the ligand-binding
Equilibrium Constants for Coordination of Substituted
Pyridines by Alkylcobalt(III) Tetrapyrrole Complexes. The
stability constants for coordination of substituted pyridines by
Co(OEP)CH3 in toluene solution, eq 3 (with P ) OEP and R
Co(P)R + L h Co(P)(R)L
(3)
) CH3), were determined spectrophotometrically by a nonlinear
least-squares fit of absorption data at multiple wavelengths. The
data are reported in Table 3.
The spectroscopic method was not as reliable for analogous
measurements with cobalt hydroporphyrin complexes. The
absorption bands of five- and six-coordinate cobalt(III) hydro-
porphyrin complexes were sufficiently broad that the presence
of significant quantities of unalkylated cobalt(II) complexes, of
macrocycle-oxidized alkylcobalt(III) complexes, or of (alkyl-
peroxo) cobalt(III) complexes might not be readily apparent.
These impurities contribute to the observed UV/vis spectra and
could therefore interfere with an accurate determination of the

1522 Inorganic Chemistry, Vol. 39, No. 7, 2000
Stolzenberg and Summers
Table 5. Equilibrium Constants for Alkyl Exchange Reactionsa
formation constant for conversion of Co(TPP)C2H5 to Co(TPP)-
(C2H5)Py in CH2Cl2 as K ) 8.9.30 This value is in reasonable
agreement with the equilibrium constant determined here for
Co(OEP)C2H5 in toluene.
K
P
P*
R
Py absent
Py presentb
OEP
t-OEC
OEP
t-OEC
ttt-OEiBC
t-OEC
CH3
CH3
i-C3H7
1.98c
1.55c
0.96c
4.90d
1.10c
Our results establish that the equilibrium constants for the
ligand-binding reactions represented by eq 3 increase with
increasing saturation of the tetrapyrrole macrocycle and with
increasing base strength (pKa of the conjugate acid) of the ligand.
Equilibrium constants decrease with increasing steric bulk of
the trans alkyl group, as expected.9,31-33 For a given ligand,
the equilibrium constants for ligand binding to the methyl-
cobalt(III) tetrapyrrole complexes are about a factor of 10
smaller than those for the corresponding Co(II) tetrapyrrole,
which in turn are smaller than those for Zn(II) tetrapyrroles. In
addition, equilibrium constants for the Co(OEP)CH3 complexes
are about a factor of 10 larger than those for methylcobinamides
and are roughly a factor of 100 smaller than those for
methylcobaloximes.34 The increased hydrogen bonding of
pyridine bases in water (the solvent used in the measurements
with cobinamides and cobaloximes) relative to toluene would
be expected to decrease the equilibrium constants for the
porphyrin. Thus, cobalt porphyrins are better quantitative models
of coenzyme B12 axial ligation equilibria than are cobaloximes.
a Determined in toluene-d8 solution at 20 °C. Equilibrium constants
for reactions defined in eq 1 and 4. b [Pyridine] ) 0.5 M. c Estimated
error ) 0.1. d Estimated error < 0.5.
equilibria of four-coordinate CoII(P) and ZnII(P) complexes,
although the ∆∆G values for t-OEC complexes appeared to be
somewhat smaller in those cases.2
The relative stability constants for coordination of pyridine-
d5 to five-coordinate Co(OEP)CH3 and four-coordinate Co(OEP)
were determined by the NMR method in CDCl3. The ratio
obtained was the same, within error, as the ratio determined in
toluene, either by the NMR method or by the spectrophotometric
method. The stability constant for coordination of pyridine to
Co(OEP) in toluene solution was determined to be roughly 13
times that in CHCl3.2 Thus, the stability constant for coordination
of pyridine to Co(OEP)CH3 has a similar solvent dependence.
The NMR method also was applied to a mixture of Co(OEP)-
CH3 and Co(OEP)C2H5 to examine the effect of the bulk of the
axial alkyl group on the stability constant for coordination of
pyridine. The ratio of stability constants was determined to be
4.8, with the methyl complex having the larger stability constant.
Thus, log K for coordination of pyridine to Co(OEP)C2H5 is
1.08.
Equilibrium Constants for Alkyl Ligand Exchange Reac-
tions. One experiment that examined the competition between
CoIII(OEP)CH3 and a CoIII(OEC)CH3 for pyridine coordination
was inadvertently performed with a sample of the hydropor-
phyrin complex that contained a substantial amount of the
unmethylated CoII(OEC) complex as an impurity. Within less
than 0.5 h of mixing, peaks due to paramagnetic CoII(OEP)
appeared in the 1H NMR spectrum. Concurrently, the peaks due
to CoII(OEC) and CoIII(OEP)CH3 decreased and those due to
CoIII(OEC)CH3 increased in intensity. The unexpectedly rapid
transfer of the methyl group between cobalt tetrapyrrole
complexes led us to investigate alkyl exchange reactions.
The alkyl exchange reactions of alkylcobalt(III) and cobalt(II)
tetrapyrrole complexes are of considerable scope. Reactions
proceed both in toluene, in which the complexes, respectively,
are five-coordinate and four-coordinate, eq 1, and in pyridine
in toluene, in which the complexes, respectively, are six-
coordinate and five-coordinate, eq 4. Exchange is not limited
A definitive explanation could not be given for the cis-
influence of the macrocycle saturation level on log K for
coordination of bases to Co(II) and Zn(II) tetrapyrrole com-
plexes.2 Solvation and π-bonding effects were eliminated as
possible sources. Measurements of the pKa values for depro-
tonation of the three macrocycles showed that they were similar.
However, several lines of evidence suggested that these pKa’s
are not good indicators of the σ-donor strengths of the
tetrapyrroles. The possibility remains that the cis-influence
results from a decreased σ-donor strength upon increasing
tetrapyrrole saturation. The origin of the cis-influence on ligand
binding in alkylcobalt(III) tetrapyrrole complexes is likely to
be similar to that for Co(II) and Zn(II) tetrapyrrole complexes.
In our analysis of the ligand-binding equilibria of Co(II) and
Zn(II) tetrapyrroles, we developed the tool of plotting log K
values for binding of a particular base to a complex vs log K
2
values for binding of the base to Zn(OEP), log KZn(OEP)
.
Zn(OEP) is a more suitable reference acid than Haq+. The slope
of a line in such plots represented the sensitivity of ∆G for
ligand binding to the basicity of the axial ligand. It depended
strongly on the identity of the metal but was essentially
independent of the identity of the macrocycle. Thus, the slope
was identified with the acidity of the metal in the specific
environment of the coordination complex. The intercept of a
line represented ∆G for coordination of a hypothetical ligand
which has ∆G ) 0 for coordination to Zn(OEP) when all
reactants and products are at unit activity. The intercept
depended strongly on the identities of both the metal and the
macrocycle. To the extent that the structural changes occuring
upon ligand coordination are independent of the identity of the
ligand (a reasonable assumption for para-substituted pyridines),
the contribution of the change in strain energy to the energetics
CoIII(P)(R)Py + CoII(P*)Py h CoII(P)Py + CoIII(P*)(R)Py
(4)
to methyl groups. Ethyl and isopropyl groups also transfer.
Although the kinetics of the reaction were not investigated, our
observations suggest that the rate of transfer becomes somewhat
slower as the bulk of the alkyl group increases. The exchange
reactions proceed to equilibrium. The equilibrium constants,
Table 5, are not statistical (i.e., equal to 1.0). Cobalt octaethyl-
hydroporphyrin complexes have a slightly greater affinity for a
methyl group than Co(OEP).
(30) Kadish, K. M.; Han, B. C.; Endo, A. Inorg. Chem. 1991, 30, 4502-
4506.
(31) Brown, K. L.; Satyanarayana, S. Inorg. Chim. Acta 1992, 201, 113-
119.
(32) Pailes, W. H.; Hogenkamp, H. P. C. Biochemistry 1968, 7, 4160-
Discussion
4166.
(33) Baldwin, D. A.; Betterson, E. A.; Chemaly, S. M.; Pratt, J. M. J. Chem.
Soc., Dalton Trans. 1985, 1613-1618.
(34) Garr, C. D.; Sirovatka, J. M.; Finke, R. G. Inorg. Chem. 1996, 35,
5912-5922.
The data reported in this paper for the ligand-binding equilbria
of alkyl-Co(III) porphyrin complexes represent the majority
of such data in the literature. One previous report gave the

Reactions of Alkylcobalt(III) Porphyrin Complexes
Inorganic Chemistry, Vol. 39, No. 7, 2000 1523
of the reaction will be independent of the base strength of the
ligand. Thus, the change in strain contributes to the intercept,
not the slope. The relief of strain due to core expansion that
occurs upon ligand coordination to the cobalt complexes was
considered responsible both for the larger intercepts of the Co(II)
complexes relative to the Zn(II) complexes and for the inverted
order log KCo(II) > log KZn(II) for the weakest bases.
complexes are lacking because it is unusual to find a ligand
system that stabilizes both a true four-coordinate Co(II) complex
and a true five-coordinate alkyl-Co(III) complex. In most cases,
at least one form dimerizes or is strongly associated with a base
or solvent. The ligand-binding equilibrium constants measured
in these cases are really for displacement of one coordinated
base by a second base. Measurements for adenosylcobinamide
and cob(II)inamide in the coordinating solvent ethylene glycol
show that pyridine and imidazole bases coordinate more strongly
to the cobalt(II) form.36
Analysis of log KMeCo(P) vs log KZn(OEP) plots reveals that the
slopes for the alkylcobalt(III) complexes are roughly 0.36. This
is essentially indistinguishable from the slopes for the Co(II)
complexes, within the errors of the determination. The intercepts
are respectively 0.7, 1.3, and 1.7 for the OEP, OEC, and OEiBC
complexes. These are from 0.3 to 0.6 smaller than the intercepts
for the Co(II) complexes and from 0.5 to 1.5 larger than those
for the Zn(II) complexes. Less strain appears to be released upon
coordination of an axial ligand to the methylcobalt(III) com-
plexes than to the Co(II) complexes. This is consistent with the
smaller core expansion and decreased alkyl to porphyrin distance
that was observed upon base coordination to CH3Co(OEP).14
Implications for the Base-On Effect. The effect of coordi-
nation of an axial ligand on the Co-C bond homolysis can be
analyzed by means of a thermodynamic cycle consisting of eqs
3, 5, 6, and 7. The homolysis of the six-coordinate complex,
eq 7, is equivalent to the reverse of the process in eq 3 followed
by the processes in eqs 5 and 6. Consequently, the ∆G° values
are related by eq 8. Substitution and rearrangement of eq 8 show
that ∆∆G° for bond homolysis depends on the log of the ratio
of the equilibrium constants, eq 9, where K3 and K6 are the
respective equilibrium constants for eqs 3 and 6.
The above analysis and our results contradict the rationaliza-
tion advanced to explain the observation that increasing the
strength of the base increases the apparent strength of the Co-C
bond in cobaloxime complexes.12,13 In point of fact, more basic
ligands stabilize the Co(II) complex to a greater extent than
they stabilize the Co(III) complex. The apparent increase in
stability of alkylcobaloxime complexes with increasing base
strength could be explained, though, if the sensitivity of KCo(III)
to the pKa of the base is greater than that of KCo(II). This is not
unreasonable because stronger bases may be better able to
compete against the strong trans-effect of the alkyl ligand than
weak bases. If so, increasing the pKa of the base would decrease
the log(KCo(III)/KCo(II)) term in eq 9, which in turn would decrease
the destabilization of the six-coordinate alkyl complex relative
to the five-coordinate alkyl complex. If the process in eq 7 was
studied by itself, one would conclude that stronger bases stabilize
the Co-C bond rather than recognize that they destabilize it
less. The slopes of the log KCo(III) and log KCo(II) vs pKa plots
for OEP complexes are indistinguishable within error. However,
this does not rule out a significant difference for cobaloxime
complexes, which have much larger KCo(III) values. Regardless
of the explanation, it is not clear that the apparent increase in
stability of alkylcobaloxime complexes with increasing base
strength is relevant to B12 or other alkylcobalt complexes.
Increasing base strength does not affect the base-on homolysis
rate constant of adenosylcobinamide but increases the heterolysis
rate constant.37
RCoIII(P) + L h CoII(P) + R• + L
CoII(P) + R• + L h CoII(P)L + R•
(5)
(6)
RCoIII(P)L h CoII(P) L + R•
∆G°7 ) -∆G°3 + ∆G°5 + ∆G°6
(7)
(8)
Alkyl Exchange. Alkyl transfer reactions have been reported
previously for the cases of alkylcobalt(III) corrinoids and
cobalt(II) corrinoids,38,39 alkylcobalt(III) macrocycles (including
cobaloximes) and cobalt(II) macrocycles,40-43 and alkyliron(III)
porphyrins and iron(II) porphyrins.44 For corrinoids, exchange
has only been demonstrated for methyl groups. A wider range
of alkyl groups have been exchanged for cobalt macrocycles.
An SH2 transfer mechanism was suggested for these complexes.
In contrast, a free-radical mechanism was proposed for the iron
porphyrin complexes. Equilibrium constants for exchange of
methyl groups between cobalt complexes containing different
equatorial ligands are generally small but are not statistical (i.e.,
K ) 1).44-46
∆∆G°6-coord-5-coord ) -∆G°7 - ∆G°5 )
2.303RT log(K3/K6) (9)
Our data establish that the equilibrium constant for coordina-
tion of an axial ligand to a Co(II) tetrapyrrole is greater than
that for coordination to the corresponding alkyl-Co(III) tetra-
pyrrole. In this situation, eq 9 shows that ∆G for Co-C bond
homolysis decreases upon ligand coordination. In other words,
the “base-on” effect for cobalt tetrapyrroles simply reflects a
greater stabilization by the base of the product of homolysis
(Co(II) complex) than of the reactant (alkyl-Co(III) complex).
Given that the transition state for homolysis is known to be
very product-like,35 stronger coordination in the Co(II) complex
should also stabilize the transition state for homolysis relative
to the reactant and thereby increase the rate of homolysis.
Although comparable data are not widely available for other
alkylcobalt(III) complexes, there is no compelling reason to
expect that the equilibrium constant for coordination to the
alkylcobalt(III) complex, KCo(III), will be larger than that for
coordination to the corresponding unalkylated cobalt(II) com-
plex, KCo(II). In fact, the commonality of certain structural
features for five- and six-coordinate alkylcobalt complexes14
suggests that the relative magnitudes of KCo(III) and KCo(II)
observed for cobalt tetrapyrroles could be general. Data for other
(36) Sirovatka, J. M.; Finke, R. G. Inorg. Chem. 1999, 38, 1697-1707.
(37) Garr, C. D.; Sirovatka, J. M.; Finke, R. G. J. Am. Chem. Soc. 1996,
118, 11142-11154.
(38) Kra¨utler, B. HelV. Chim. Acta 1987, 70, 1268-1278.
(39) Brown, K. L.; Zou, X. Inorg. Chem. 1992, 31, 2541-2547.
(40) Johnson, M. D. Acc. Chem. Res. 1983, 16, 343-349.
(41) Dodd, D.; Johnson, M. D. J. Chem. Soc., Chem. Commun. 1971, 21,
1371-1372.
(42) Mestroni, G.; Cocevar, C.; Costa, G. Gazz. Chim. Ital. 1973, 103,
273-285.
(43) Dodd, D.; Johnson, M. D.; Lockman, B. L. J. Am. Chem. Soc. 1977,
99, 3664-3673.
(44) Song, B. H.; Goff, H. M. Inorg. Chim. Acta 1994, 226, 231-235.
(45) Van Den Bergen, A.; West, B. O. J. Organomet. Chem. 1974, 64,
125-134.
(46) Endicott, J. F.; Balakrishnan, K. P.; Wong, C.-L. J. Am. Chem. Soc.
1980, 102, 5519-5526.
(35) Halpern, J. Polyhedron 1988, 7, 1483-1490.

1524 Inorganic Chemistry, Vol. 39, No. 7, 2000
Stolzenberg and Summers
The results reported here are the first demonstration of alkyl
exchange for cobalt porphyrins and hydroporphyrins. The
reaction occurs for both base-on six-coordinate and base-off five-
coordinate complexes. The stability of the Co-C bond increased
with increasing saturation of the macrocycle for methyl com-
plexes. Exchange also occurred for alkyl groups larger than
methyl. Interestingly, exchange was slower for larger alkyl
groups. In addition, the exchange reactions of cobalt porphyrins
were more rapid than those of iron porphyrins, which are
believed to have the weaker metal-carbon bonds. Both of these
observations seem inconsistent with a free-radical mechanism
for the exchange. An investigation of the mechanisms for alkyl
exchange reactions of cobalt porphyrins will be reported in a
subsequent paper.47
Acknowledgment. This research was supported by the
National Institutes of Health (Grant GM 33882).
Supporting Information Available: Text and equations presenting
the derivation of eq 2 for analyzing data from the competitive NMR
titration method. This material is available free of charge via the Internet
at http://pubs.acs.org.
IC991304T
(47) Cao, Y.; Stolzenberg, A. M. Manuscript in preparation.