Sulfur dioxide insertion into the Co-C bond in organocobaloximes: Crystal structure and Co-C bond reactivity study

Article abstract of DOI:10.1021/om0507171

The insertion of sulfur dioxide into the Co-C bond in ArCH ₂Co(dioxime)₂Py forms a mixture of ArCH₂SO ₂Co(dioxime)₂Py, ClCo(dioxime)₂Py, and a neutral trinuclear complex, [(dpgH)₂PyCo^III-μ-SO ₃]₂-Co^II(MeOH)₄. The inserted products 4-CN-C₆H₄CH₂SO₂Co(dpgH) ₂Py and C₆H₅CH₂SO ₂Co(gH)₂Py and the trinuclear complex have been structurally characterized by X-ray for the first time. The insertion of sulfur dioxide into the Co-C bond affects the orientation of the benzyl group.

Full text of DOI:10.1021/om0507171

92
Organometallics 2006, 25, 92-98
Sulfur Dioxide Insertion into the Co-C Bond in
Organocobaloximes: Crystal Structure and Co-C Bond Reactivity†
Study
Preeti Chadha, B. D. Gupta,* and Kingsuk Mahata
Department of Chemistry, Indian Institute of Technology, Kanpur 208016, India
ReceiVed August 19, 2005
The insertion of sulfur dioxide into the Co-C bond in ArCH2Co(dioxime)2Py forms a mixture of
ArCH2SO2Co(dioxime)2Py, ClCo(dioxime)2Py, and a neutral trinuclear complex, [(dpgH)2PyCoIII-µ-SO3]2-
CoII(MeOH)4. The inserted products 4-CN-C6H4CH2SO2Co(dpgH)2Py and C6H5CH2SO2Co(gH)2Py and
the trinuclear complex have been structurally characterized by X-ray for the first time. The insertion of
sulfur dioxide into the Co-C bond affects the orientation of the benzyl group.
occur with great difficulty, and lead to a mixture of products.7
In contrast, the reaction with allylcobaloxime is facile but
becomes complicated with but-3-enyl- and hexenylcobaloximes,
where a mixture of products is formed.7d Reactions have been
carried out under thermal and photochemical conditions with
or without solvent. The reaction at the secondary carbon in cis-
and trans-(4-methylcyclohexyl)cobaloxime indicates inversion
of configuration at the reacting carbon atom.8
Introduction
The recently available crystallographic data on cobalamins
suggests that the structural effects of changes in R are similar
to those found in cobaloximes and sometimes can be related to
their chemical reactivity.1 Since Co-C bond cleavage is the
key step involved in B12-dependent enzymatic or cobaloxime
mediated reactions, the strength of the Co-C bond as a function
of steric and electronic factors with a wide range of axial ligands
in cobaloximes and in related complexes with different chelates
have been systematically investigated.2 The recent results, based
on spectral and structural studies, have shown that in addition
to the trans effect of the axial base, the cis influence, the effect
of the equatorial dioxime on the axial ligands, plays an important
role toward the stability of the Co-C bond.3 Insertion reactions
have been used to test the reactivity of M-C bonds in
organometallic compounds. Insertion of small molecules such
as O2, SO2, CO, etc. into a metal-carbon bond is an important
reaction and occurs with a remarkably wide range of organo-
metallic compounds;4 the metal may be one of the transition or
main-group elements, and carbon is part of a wide variety of
organic ligands. There have been numerous reports on oxygen
insertion5 into Co-C bonds, but the corresponding SO2 insertion
has been relatively little studied.
The insertion of sulfur dioxide into benzylcobaloxime,
PhCH2Co(dmgH)2B, is the most well-studied system because
of the inherently weak Co-C bond. A recent study has shown
that the reaction is not true insertion but is an intermolecular
process in which the organic group and the metal in the insertion
product do not originate from the same molecule of the
organometallic substrate.7,9 Both chain10a and nonchain10b mech-
anisms for the insertion of sulfur dioxide in organocobaloximes
have been described. In general, the reaction is concerted with
the attack of SO2 at the metal and at the organic group, but in
the case of organocobaloximes a dissociative mechanism is
operating, where the key step is the cleavage of the Co-C
bond.10 Despite the appreciable amount of work on the kinetics,
products, and stereochemistry of these reactions, there is no
comprehensive picture of the mechanism. The sulfinato group
(Co-S bond) in the inserted product has been identified by IR
only,7,9 and no crystal structure of the inserted product is known
to date. The PyCo(dioxime)2SO2Co(dioxime)2Py complex, pro-
posed as one of the termination steps, has never been character-
ized. A side product, insoluble in most of the organic solvents
except methanol, formed in all reactions mentioned above has
never been identified. Also, since Co-C bond cleavage is the
The early work on the insertion of SO2 in RCoIII(chelate)2B
complexes (chelate ) salen, saloph, bae, dmgH; R ) alkyl,
aryl)6 has shown that these reactions are very slow (>24 h),
† Dedicated to Dr. E. D. Jemmis, IPC, IISc, Bangalore, India.
* To whom correspondence should be addressed. Tel: +91-512-2597046.
Fax: +91-512-2597436. E-mail: bdg@iitk.ac.in.
(1) (a) Randaccio, L.; Furlan, M.; Geremia, S.; Slouf, M.; Srnova, I.;
Toffoli, D. Inorg. Chem. 2000, 39, 3403. (b) Randaccio, L.; Geremia, S.;
Nardin, G.; Slouf, M.; Srnova, I. Inorg. Chem. 1999, 38, 4087.
(2) (a) Toscano, P. J.; Swider, T. F.; Marzilli, L. G.; Bresciani-Pahor,
N.; Randaccio, L. Inorg. Chem. 1983, 22, 3416 and references therein. (b)
Bresciani-Pahor, N.; Marzilli, L. G.; Randaccio, L.; Summers, M. F.;
Toscano, P. J. Coord. Chem. ReV. 1985, 63, 1. (c) Randaccio, L.; Bresciani-
Pahor, N.; Zangrando, E.; Marzilli, L. G. Chem. Soc. ReV. 1989, 18, 225.
(d) Randaccio, L. Comments Inorg. Chem. 1999, 21, 327.
(6) Abbreviations: salen ) N,N′-ethylenebis(salicylideneiminato), saloph
) N,N′-phenylenebis(salicylideneiminato), bae ) N,N′-ethylenebis(acetyl-
acetonyliminato), dmgH ) dimethylglyoximato, dpgH ) diphenylglyoxi-
mato, gH ) glyoximato.
(7) (a) Cozens, R. J.; Deacon, G. B.; Felder, P. W.; Murray, K. S.; West,
B. O. Aust. J. Chem. 1970, 23, 481. (b) Murray, K. S.; Cozens, R. J.; Deacon,
G. B.; Felder, P. W.; West, B. O. Inorg. Nucl. Chem. Lett. 1968, 4, 705.
(c) Johnson, M. D.; Lewis, G. J. J. Chem. Soc. A 1970, 2153. (d) Cooksey,
C. J.; Dodd, D.; Gatford, C.; Johnson, M. D.; Lewis, G. J.; Titchmarsh, D.
M. J. Chem. Soc., Perkin Trans. 2 1972, 655.
(3) Gupta, B. D.; Mandal, D. Organometallics 2005, 24, 1501 and
references therein.
(4) (a) Jacobson, S. E.; Wojcicki, A. J. Organomet. Chem. 1974, 72,
113. (b) Ca´mpora, J.; Lo´pez, J. A.; Palma, P.; del Rio, D.; Carmona, E.
Inorg. Chem. 2001, 40, 4116. (c) Cooney, J. M.; Depree, C. V.; Main, L.;
Nicholson, B. K. J. Organomet. Chem. 1996, 515, 109.
(5) (a) Gupta, B. D.; Vijaikanth, V.; Singh, V. J. Organomet. Chem.
1998, 570, 1. (b) Gupta, B. D.; Roy, M.; Das, I. J. Organomet. Chem. 1990,
397, 219. (c) Gupta, B. D.; Roy, S. Inorg. Chim. Acta 1985, 108, 261.
(8) Cotton, J. D.; Crisp, G. T. J. Organomet. Chem. 1980, 186, 137.
(9) (a) Gupta, B. D.; Roy, S.; Roy, M. Tetrahedron Lett. 1987, 28, 1219.
(b) Gupta, B. D.; Roy, M.; Oberoi, M.; Dixit, V. J. Organomet. Chem.
1992, 430, 197.
(10) (a) Johnson, M. D.; Crease, A, E. J. Am. Chem. Soc. 1978, 100,
8013. (b) Johnson, M. D.; Derenne, S. J. Organomet. Chem. 1985, 286,
C47.
10.1021/om0507171 CCC: $33.50 © 2006 American Chemical Society
Publication on Web 11/24/2005

SO2 Insertion in Organocobaloximes
Organometallics, Vol. 25, No. 1, 2006 93
Table 1. Elemental Analysis Data for the Compounds 1a-9a
found (calcd)
no.
formula
C
H
N
S
1a
2a
3a
4a
5a
6a
7a
8a
9a
C40H34CoN5O6S
ClC40H33CoN5O6S
C41H33CoN6O6S
C38H32CoN5O6S2
C16H16CoN5O6S
C20H24CoN5O6S
ClC20H23CoN5O6S
C21H23CoN6O6S
C36H32CoN5O6S
62.88 (62.26)
60.23 (59.63)
62.72 (61.81)
58.10 (58.69)
41.67 (41.29)
46.58 (46.07)
42.80 (43.24)
46.60 (46.15)
60.51 (59.92)
4.45 (4.41)
4.13 (4.10)
4.12 (4.15)
4.17 (4.12)
3.48 (3.44)
4.67 (4.61)
4.18 (4.14)
4.16 (4.21)
4.46 (4.44)
9.14 (9.08)
8.59 (8.70)
10.51 (10.55)
8.93 (9.01)
15.28 (15.05)
13.54 (13.44)
12.48 (12.61)
15.55 (15.38)
9.78 (9.71)
4.20 (4.15)
4.04 (3.98)
4.05 (4.02)
8.17 (8.24)
6.83 (6.88)
6.08 (6.14)
5.85 (5.77)
5.92 (5.86)
4.50 (4.44)
Scheme 1
hν
III
III
III
+ SO (g)
8
+
RCH2Co (dioxime)2B
RCH2SO2Co (dioxime)2B ClCo (dioxime)2B
2
0 °C, CH2Cl2
1-10
1a-10a
11
dioxime ) dpgH; B ) Py;
RCH2 ) benzyl (1), 4-chlorobenzyl (2), 4-cyanobenzyl (3), 2-thienylmethyl (4)
dioxime ) gH; B ) Py; RCH2 ) benzyl (5)
dioxime ) gH; B ) 4-tBuPy; RCH2 ) benzyl (6), 4-chlorobenzyl (7), 4-cyanobenzyl (8)
dioxime ) dpgH; B ) Py; RCH2 ) allyl (9), 3-methylallyl (10)
ClCo(dioxime)2B; dioxime ) dpgH; B ) Py (11)
[(dpgH)2PyCoIII-µ-SO3]2CoII(MeOH)4 (12)
136, 154, 289, and 307 are the matrix peaks. Elemental analysis
data for compounds 1a-9a are given in Table 1.
key feature of these reactions and the reactivity of the Co-C
bond depends on the equatorial ligands (cis influence),3 it would
be appropriate to study this reaction by varying the equatorial
dioxime ligand to see if it leads to any change in the product
distribution and also to see if it is possible to correlate the
molecular structure with the spectral data.
In this work we report our findings of sulfur dioxide insertion
in RCH2Co(dioxime)2Py (dioxime ) dpgH, gH) complexes
(Scheme 1). We also report the first crystal structure of a sulfur
dioxide inserted product. The side product (termed as methanol
fraction) has been characterized by X-ray and FAB mass and
is a neutral trinuclear complex, CoIII-CoII-CoIII (12). The
spectral data have been correlated with the molecular structures.
X-ray Structural Determination and Refinement. Orange
crystals were obtained by slow evaporation of the solutions of 3a
and 6a in a dichloromethane, n-hexane, and benzene mixture and
by that of 12 in methanol. Single-crystal X-ray data were collected
at 100 K on a Bruker SMART APEX CCD diffractometer using
graphite-monochromated Mo KR radiation (λ ) 0.710 73 Å). The
linear absorption coefficients, scattering factors for the atoms, and
anomalous dispersion corrections were taken from ref 12. The data
integration and reduction were processed with SAINT13 software.
An empirical absorption correction was applied to the collected
reflections with SADABS14 using XPREP.15 The structure was
solved by direct methods using SIR-9716 and was refined on F2 by
the full-matrix least-squares technique using the SHELXL-9717
program package. All non-hydrogen atoms were refined anisotro-
pically in the structures. The hydrogen atoms of the OH group of
oxime were located on difference maps and were constrained to
those difference map positions. The hydrogen atom positions or
thermal parameters were not refined but were included in the
structure factor calculations. The pertinent crystal data and refine-
ment parameters are compiled in Table 2.
Experimental Section
ClCo(dioxime)2Py (dioxime ) dpgH, gH), RCH2Co(dioxime)2Py
(dioxime ) gH, dpgH; RCH2 ) benzyl, 4-chlorobenzyl, 4-cy-
anobenzyl, 2-thienylmethyl), and RCH2Co(gH)2-4-tBuPy were either
synthesized by the literature procedure11 or were available from
our laboratory from a different study. The synthesis work was
carried out in subdued light and under a blanket of argon or
nitrogen. Silica gel (100-200 mesh) and distilled solvents were
used in all chromatographic separations. 1H and 13C NMR spectra
were recorded on JEOL-JNM LAMBDA 400 spectrometer in
CDCl3. For 1H and 13C NMR spectra, TMS was used as an internal
reference. Elemental analysis was carried out using a Thermoquest
CE instruments CHNS-O elemental analyzer. The FAB-mass
spectra were recorded on a JEOL SX 102/DA-6000 mass spec-
trometer/data system using argon/xenon (6 kV, 10 mA) as the FAB
gas. The accelerating voltage was 10 kV, and the spectra were
recorded at room temperature. m-Nitrobenzyl alcohol (NBA) was
used as the matrix unless specified otherwise. The peaks at m/z
SO2 Insertion: General Procedure. A solution of benzylco-
baloxime (1; 0.09 g, 0.196 mmol) in 10 mL of dry dichloromethane
was purged with a stream of dry nitrogen gas to remove all traces
of oxygen. It was irradiated with a 200 W tungsten lamp placed at
about 10 cm from the reaction vessel. The temperature was
maintained at 0 °C by a Julabo refrigerator circulator, and dry sulfur
(12) International Tables for X-ray Crystallography; Kynoch Press:
Birmingham, England, 1974; Vol. IV.
(13) SAINT+, version 6.02; Bruker AXS, Madison, WI, 1999.
(14) Sheldrick, G. M. SADABS, Empirical Absorption Correction
Program; University of Go¨ttingen, Go¨ttingen, Germany, 1997.
(15) XPREP, version 5.1; Siemens Industrial Automation Inc., Madison,
WI, 1995.
(11) (a) Gupta, B. D.; Roy, S. Inorg. Chim. Acta 1988, 146, 209. (b)
Gupta, B. D.; Qanungo, K. J. Organomet. Chem. 1997, 543, 125. (c) Lo´pez,
C.; Alvarez, S.; Font-Bardia, M.; Solans, X. J. Organomet. Chem. 1991,
414, 245.
(16) Sheldrick, G. M. SHELXTL Reference Manual: Version 5.1; Bruker
AXS: Madison, WI, 1997.
(17) Sheldrick, G. M. SHELXL-97: Program for Crystal Structure
Refinement; University of Go¨ttingen, Go¨ttingen, Germany, 1997.

94 Organometallics, Vol. 25, No. 1, 2006
Table 2. Crystal Data and Structural Refinement Details for 3a, 6a, and 12
Chadha et al.
4-CN-C6H4CH2SO2Co(dpgH)2Py‚
2C6H6 (3a)
[C6H5CH2SO2Co(gH)2(4-tBuPy)]2‚
2CH2Cl2‚C6H6‚H2O (6a)
[(dpgH)2PyCoIII-µ-SO3]2-
CoII(MeOH)4 (12)
empirical formula
formula wt
temp (K)
radiation, λ (Å)
cryst syst
space group
unit cell dimens
a (Å)
C53H45CoN6O6S
952.94
100(2)
Mo KR, 0.710 73
triclinic
P1h
C48H64Cl4Co2N10O13S2
1312.87
100(2)
Mo KR, 0.710 73
triclinic
C70H70Co3N10O18S2
1580.28
100(2)
Mo KR, 0.710 73
triclinic
P1h
P1h
13.285(2)
13.854(2)
14.562(2)
93.985(2)
90.190(3)
115.726(2)
2407.0(6)
10.9123(8)
16.2234(12)
17.2569(13)
101.9630(10)
90.1680(10)
91.1730(10)
2988.0(4)
10.0213(10)
11.1220(11)
17.4143(16)
89.046(2)
76.351(2)
67.463(2)
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
1736.2(3)
Z
2
2
1
F(calcd), Mg/m3
1.315
0.456
992
0.31 × 0.25 × 0.22
-17 e h e 8,
-17 e k e 18,
-18 e l e 19
15 642
1.459
0.871
1360
0.27 × 0.20 × 0.19
-14 e h e 14,
-19 e k e 21,
-22 e l e 15
20 188
1.511
0.848
817
0.38 × 0.28 × 0.23
-9 e h e 13,
-13 e k e 14,
-23 e l e 21
11 703
µ (mm-1
F(000)
)
cryst size (mm3)
index ranges
no. of rflns collected
no. of indep rflns
refinement method
GOF on F2
11 311
14 273
8280
full-matrix least squares on F2
1.053
full-matrix least squares on F2
1.035
full-matrix least squares on F2
1.025
final R indices (I > 2σ(I))
R1 ) 0.1093
wR2 ) 0.2955
R1 ) 0.1491
wR2 ) 0.3214
11 311/62/604
R1 ) 0.0744
wR2 ) 0.1825
R1 ) 0.1011
wR2 ) 0.1975
14 273/0/718
R1 ) 0.0774
wR2 ) 0.2094
R1 ) 0.1007
wR2 ) 0.2259
8280/0/468
R indices (all data)
no. of data/restraints/params
Table 3. Results of SO2 Insertion into Organocobaloximes 1-10
R in RCH2Co(dioxime)2Py
solvent
CH2Cl2
products
yield, %
benzyl (1)a
C6H5CH2SO2Co(dpgH)2Py (1a)
ClCo(dpgH)2Py (11)
4-Cl-C6H4CH2SO2Co(dpgH)2Py (2a)
ClCo(dpgH)2Py (11)
4-Cl-C6H4CH2SO2Co(dpgH)2Py (2a)
ClCo(dpgH)2Py (11)
4-Cl-C6H4CH2SO2Co(dpgH)2Py (2a)
4-CN-C6H4CH2SO2Co(dpgH)2Py (3a)
ClCo(dpgH)2Py (11)
23
46
24
48
30
20
40
23
48
24
23
30
40
4-chlorobenzyl (2)
CH2Cl2
MeCN/CHCl3 (3:1)
THF
CH2Cl2
4-cyanobenzyl (3)
2-thienylmethyl (4)
CH2Cl2
2-thienylmethylSO2Co(dpgH)2Py (4a)
ClCo(dpgH)2Py (11)
benzyl (5)
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
C6H5CH2SO2Co(gH)2Py (6a)
benzyl (6)b
C6H5CH2SO2Co(gH)2(4-tBuPy) (6a)
4-chlorobenzyl (7)b
4-cyanobenzyl (8)b
allyl (9)
4-Cl-C6H4CH2SO2Co(gH)2(4-tBuPy) (7a)
4-CN-C6H4CH2SO2Co(gH)2(4-tBuPy) (8a)
CH2dCHCH2SO2Co(dpgH)2Py (9a)
ClCo(dpgH)2Py (11)
45
30
20
45
c
3-methylallyl (10)
CH2Cl2
CH3CHdCHCH2Co(dpgH)2Py (10a)
ClCo(dpgH)2Py (11)
46
a A similar ratio of products is formed in chloroform. b Base ) 4-tBuPy. c Pure product could not be isolated; 10-15% of 12 is formed in most of the
reactions.
dioxide gas18 was bubbled through the solution under a positive
ClCo(dpgH)2Py (11) eluted out first with 5% ethyl acetate, followed
by the inserted product with 10% ethyl acetate (1a). The rest of
the compound was eluted out with methanol (12).
In gH complexes, the column details are same, except that
the first fraction is the inserted product and is eluted out with
60% ethyl acetate. The rest of the compound was eluted out with
methanol.
pressure of nitrogen.9 The reaction was continued until all of the
starting cobaloxime had disappeared. The progress of the reaction
was monitored with TLC on silica gel with a mixture of ethyl
acetate and dichloromethane as the eluent. At the end of the reaction
the solvent was removed and the products were separated using
column chromatography.
Column Details. In dpgH complexes the reaction mixture, a
brown powder, dissolved in a minimum amount of dichloromethane
was loaded on the silica gel column pre-eluted with dichlo-
romethane. The polarity was increased slowly with ethyl acetate.
Results and Discussion
Benzylcobaloxime (1) reacts with sulfur dioxide gas under
photolytic conditions to give a mixture of C6H5CH2SO2-
CoIII(dpgH)2Py (1a) and chlorocobaloxime (11) in 23% and 46%
yields, respectively. A smooth reaction occurs and is complete
in about 1.5 h. A similar reaction occurs with substituted benzyl
and thienylmethyl cobaloximes (2-4) (Table 3). Either chlo-
(18) Dry sulfur dioxide gas was generated from sodium sulfite and dilute
sulfuric acid and passed through traps of sulfuric acid and anhydrous calcium
chloride. The solution of organocobaloxime was kept over molecular sieves
during the reaction.

SO2 Insertion in Organocobaloximes
Organometallics, Vol. 25, No. 1, 2006 95
Table 4. 1H NMR Data (ppm) for Complexes 1-9 and 1a-9aa
pyridine
no.
aromatic
allyl/others
CH2 (s)
dpgH/gH
R (d)
â
γ (t)/tBu (s)
O-H‚‚‚O
1
1a
2
2a
3
3a
4
4a
5
5a
6
6a
7
7a
8
6.92-7.35
6.94-7.69
6.92-7.27
7.15-7.32
6.92-7.27
7.12-7.50
6.83-7.26
6.77-7.77
7.06-7.10
7.19-7.33
6.98-7.19
7.20-7.27
6.99
3.44
4.47
3.35
4.34
3.30
4.47
3.63
4.66
3.01
4.31
2.91
4.29
2.87
4.24
2.82
4.36
2.97
3.91 (d)
2.94 (d)
b
b
b
b
b
b
b
b
7.24
7.55
7.16
7.55
7.24
7.55
7.25
7.63
8.85
8.84
8.83
8.79
8.89
8.84
8.84
8.84
8.54
8.35
8.32
8.19
8.36
8.17
8.30
8.24
8.97
8.76
8.88
7.40 (t)
b
7.79
7.91
7.81
7.91
7.82
7.95
7.79
8.00
7.75
7.75
1.19
1.20
1.25
1.21
1.25
1.25
7.84
7.91
7.82
18.73
7.41 (t)
7.49 (t)
7.41 (t)
b
7.39 (t)
b
18.71
18.48
18.68
18.67
17.67
7.35 (t)
b
7.23
7.24
7.29
7.26
7.29
7.55
c
17.65
-
17.70
7.16-7.20
7.08-7.30
7.35-7.41
17.63
8a
9
9a
10
5.31 (dd), 6.25 (m)
5.26 (dd), 5.95 (m)
1.34 (d), 5.78 (m)
6.93-7.56
7.19-7.32
7.04-7.26
7.47 (t)
7.43 (t)
18.45
18.71
a Ligand: s ) singlet; m ) multiplet; dd ) doublet of doublets; t ) triplet. b Merge with aromatic protons. c Merge with dpgH protons.
rocobaloxime is not formed or its amount is reduced when the
reaction is carried out in THF or MeCN/CHCl3 (3:1), as the
case may be. The reaction in methanol forms only a trace
amount of inserted product. The reactions carried out in
nonhalogenated solvents also result in loss of some amount of
dioxime ligand.
In the gH complex 5, the reaction forms 5a and no chloro-
cobaloxime is formed. The solubility of 5a is found to be very
poor in chloroform, but it is improved when 4-tBuPy is used as
the base instead of pyridine. A slight increase in yield is also
noticed. Allylcobaloxime (9) reacts in a similar fashion and
forms the corresponding inserted product 9a and chloroco-
baloxime (11). However, the reaction is much more complicated
with (3-methylallyl)cobaloxime (10). A mixture of products is
formed, and the inserted product 10a is only a small fraction
of the total product.
In every reaction, an additional product, insoluble in most
organic solvents except methanol, is also formed. It has proved
very difficult to characterize this product by 1H NMR. Its crystal
structure shows it to be a CoIII-CoII-CoIII complex (12;
discussed later).
Chlorocobaloxime (11) is formed by the abstraction of a
halogen atom from dichloromethane by the intermediate co-
baloxime(II) generated during photolysis. This has been con-
firmed by the reaction of 2 with sulfur dioxide in THF, where
no chlorocobaloxime is formed. In a blank reaction, it is found
that benzylcobaloxime (1) in dichloromethane under identical
photolytic conditions but in the absence of SO2 is stable and
does not abstract halogen from dichloromethane.
All the reactions described in this paper are free radical in
nature.9 The radical nature of these reactions is apparent from
the characteristics of the reaction: (a) a concentration-dependent
induction period; (b) no insertion at 0 °C without irradiation;
(c) reactions are inhibited by galvinoxyl, a radical trap. The
key step is the homolytic cleavage of the Co-C bond and the
stability/reactivity of cobaloxime(II) generated during photolysis.
The formation of chlorocobaloxime indicates that the abstraction
of halogen from the solvent is more facile in CoII(dpgH)2Py as
compared to CoII(gH)2Py, where no such product is formed.
This may seem to be surprising and may contradict the earlier
observations of Halpern19 that the rate of halogen abstraction
from alkyl halide by cobaloxime(II) is insensitive to the variation
in dioxime unit. However, keeping into account the facts that
(a) benzylcobaloxime (1) does not abstract halogen from the
solvent CH2Cl2 under irradiation in the absence of SO2 and (b)
no halogen abstraction by cobaloxime(II) occurs in the oxygen
insertion reaction5 in benzylcobaloxime, the halogen abstraction
process must be getting catalyzed in the presence of the substrate
sulfur dioxide. This may have some relevance to cobalamins
in which the Co-C bond is apparently activated greatly by the
coenzyme interaction with apoenzyme, by binding of the
resulting holoenzyme to the substrate, or by both of these
processes.20
The characteristics of the reaction show that the mechanism
of the reaction is similar to that reported earlier for PhCH2Co-
(dmgH)2Py with SO2.9 However, we have not been able to
isolate and characterize the product, PyCo(dioxime)2SO2Co-
(dioxime)2Py, involved in the termination step. However, the
side product, the methanol fraction, has been identified and is
discussed later. The crystal structures of 3a and 6a clearly show
a Co-S bond (discussed later).
Spectroscopy. UV, IR, NMR, and X-ray crystallography have
been the primary means of characterizing the SO2-inserted
products. The Co-C CT band, observed in the parent co-
baloxime, disappears in the inserted complex.9 The ν(Co-SO2)
band in the infrared spectra of the inserted product appears at
ν
sym (1060-1070 cm-1) and νassym (1230-1245 cm-1), a range
observed for the S-bonded sulfinate group in many similar
complexes.7,9
1H and 13C NMR Spectra. 1H NMR spectra are easily
assigned on the basis of chemical shifts. The signals are assigned
according to their relative intensities and are consistent with
the related complexes previously described. 13C chemical shifts
of similar complexes have not been reported previously. We
have therefore assigned these values on the basis of the
previously reported 13C NMR values in the cobaloximes.21 1
H
NMR data for compounds 1-9 and 1a-9a are given in Table
4, and 13C NMR data for compounds 2, 6, 7, 2a, 6a, and 9a are
given in Table 5.
(19) Schneider, P. W.; Phelan, P. F.; Halpern, J. J. Am. Chem. Soc. 1968,
91, 77.
(20) (a) Toraya, T. Cell. Mol. Life Sci. 2000, 57, 106. (b) Masuda, J.;
Shibata, N.; Morimoto, Y.; Toraya, T.; Yasuoka, N. Structure (London)
2000, 8, 775. (c) Brown, K. L. Chem. ReV. 2005, 105, 2075.
(21) (a) Gupta, B. D.; Yamuna, R.; Singh, V.; Tiwari, U.; Barclay, T.;
Cordes, W. J. Organomet. Chem. 2001, 627, 80. (b) Gupta, B. D.; Qanungo,
K.; J. Organomet. Chem. 1997, 543, 125.

96 Organometallics, Vol. 25, No. 1, 2006
Table 5. 13C NMR Data for Complexes 2, 6, 7, 2a, 6a, and 7a
Chadha et al.
pyridine
no.
C-N
R
â
γ
aromatic + dpgH
others
CH2Co
2
2a
6
6a
7
7a
151.13
154.29
137.89
141.34
137.93
141.40
150.14
150.28
149.25
149.12
149.27
149.10
125.59
126.21
122.72
123.39
122.81
123.42
138.05
143.43
162.55
164.25
161.03
167.78
127.74, 128.22, 129.01, 129.56, 129.75, 130.19
128.04, 128.33, 129.35, 129.62, 129.80, 132.62, 151.40
124.83, 126.78, 127.63, 128.05, 128.59, 145.60
127.98, 128.10, 129.21, 131.28, 114.87
121.53, 127.79, 129.73, 130.16
32.66
63.10
34.84
63.59
32.85
62.46
30.14, 34.26
30.11, 35.11
30.18, 32.85
30.11, 35.12
127.58, 128.27, 132.56, 149.75
Sulfur dioxide insertion has affected the chemical shifts of
the three groups RCH2, dioxime, and axial pyridine in the H
(b) Pyγ. We have recently proposed that δ∆Pyγ can be taken
as the measure of the trans effect. The downfield shift in both
1H and 13C NMR in the inserted complexes clearly points to
the lower trans effect of the SO2CH2Ar group as compared to
CH2Ar.
1
NMR spectra of 1a-8a (Table 4). RCH2 shifts downfield by
1.0-1.50 ppm in 1a-8a as compared to the case for the parent
cobaloxime 1-8; the downfield shift is more in gH complexes
as compared to dpgH. Pyridine R-protons are shifted upfield
(0.01-0.19 ppm), and the gH protons are shifted downfield by
0.3-0.4 ppm. No comment can be made on the dpgH protons,
as these overlap with the benzyl aromatic protons.
(c) gH. The downfield shift in 5a-8a as compared to the
shifts for 5-8 is due to the perpendicular orientation of the
benzyl group in the former (see crystal structure details).
1
Methanol Fraction. H NMR could not characterize the
There is a drastic chemical shift in the 13C NMR spectra of
2a, 6a, and 7a (Table 5). For example, CH2 is shifted downfield
by 30 ppm and Pyγ and CdN are shifted by 3-6 ppm with
respect to the parent cobaloxime. The chemical shifts of CH2
in 2a and 6a are almost identical, indicating that the nature of
dioxime has no effect on the CH2 shift. In contrast, the 13C signal
in the parent cobaloxime 6 appears 2.2 ppm downfield with
respect to its value in 2.
methanol fraction obtained from the column separation of the
reaction mixture. The NMR spectrum showed the presence of
dpgH and pyridine, but surprisingly, the axial benzyl group was
missing. Slow evaporation led to the formation of red crystals.
The X-ray analysis of these crystals showed the composition
as Co3S2N10O18C70H70. The Diamond diagram of the molecular
structure for [(dpgH)2PyCoIII-µ-SO3]2CoII(MeOH)4 (12) along
with a selected numbering scheme is shown in Figure 1. Crystal
refinement details are given in Table 2, and selected bond
lengths and bond angles are given in Table 6.
Now, the question arises “Why does dioxime has an effect
on CH2 in the parent cobaloxime but not in the inserted product,
and which factors cause such a large shift?”
The crystal structure shows this to be a neutral trinuclear
complex having a crystallographic center of inversion occupied
by the central Co(2) atom. Similar such structures have recently
been reported.24 All three Co centers are in distorted-octahedral
coordination environments. The central Co(2) is in the +2
oxidation state, and the terminal Co atoms are in the +3 state.
In the terminal Co atoms the dpgH ligand occupies four
equatorial positions and the two axial positions are occupied
by sulfur of the bridging SO32- anion and a nitrogen of pyridine.
Six oxygen atomsstwo from the bridging sulfinato-O and four
from methanolssurround the central Co(2) atom. The EPR
spectrum recorded at room temperature and at 120 K in the
solid state and in solution (CH2Cl2 + MeCN) gives only broad
signals and no hyperfine splitting (Supporting Information).
This is in accordance with the recent finding in similar tri-
nuclear complexes, where the EPR signals were shown to
disappear above 30 K due to fast spin-lattice relaxation of the
high-spin Co(II) complex.24a The crystal structure is also
supported by FAB mass and elemental analysis (Anal. Calcd
for Co3S2N10O18C70H70: C, 53.20; H, 4.43; N, 8.86; S, 4.05.
Found: C, 51.23; H, 3.86; N, 8.74; S, 3.92). In the FAB mass
spectrum we have not been able to see the molecular ion peak
but the major peaks at m/z 616, 538, and 835 correspond to
[M1]+ ((dpgH)2Co pyridine), [M1 - pyridine]+, and [M - M1
- 4MeOH]+ respectively.
In general, the chemical shift of CH2 in benzylcobaloximes
depends mainly upon two factors: the ring current22 of dioxime
and the substituent on the CH2 group. These two factors put
together are responsible for cobalt anisotropy.23 The ring current
in the dioxime, in turn, is affected by the substituents on the
dioxime and by the ring current of the benzyl group (π-π
interaction with dioxime). This is reflected in the δ(1H) CH2
values of these complexes; for example, it appears downfield
in 2 as compared to 6 because of lower ring current and higher
cobalt anisotropy in the dpgH complexes3 (∆δC-N, a measure
of the dioxime ring current, supports this view; compare ∆δC-N
values in 2a and 6a). However, in the case of inserted products
1a-8a the change in chemical shift in CH2 is due to the
substituent SO2 group only. The effect of the benzyl group on
the dioxime ring current is minimized because of its perpen-
dicular orientation; the crystal structure shows that the benzyl
group is too far away from the dioxime to have any through-
space interaction. Further support that a through-space interac-
tion does not occur comes from 13C NMR spectra. Since it
operates mainly through-bond, a 30 ppm downfield shift can
arise only due to a direct bond between CH2 and SO2 group.
(a) PyR. Dioxime ring current affects the chemical shift of
PyR. Since gH complexes have a higher ring current than the
dpgH complexes, it always appears upfield (0.3-0.7 ppm) in
the former (compare 1-4 with 5-8 and 1a-4a with 5a-8a).
However, the difference in chemical shift is not that large
(0.01-0.19 ppm upfield shift) when we compare values in 1-8
vs those in 1a-8a. This seems justified, since cobalt anisotropy
works differently in XCo(dioxime)2Py and RCo(dioxime)2Py
complexes3 and ArCH2SO2Co(dioxime)2Py behaves like inor-
ganic cobaloximes.
We are unable to offer any explanation for the formation of
the trinuclear complex, but efforts are underway to develop a
general procedure for its synthesis.
(23) (a) Charland, J. P.; Zangrando, E.; Bresciani-Pahor, N.; Randaccio,
L. Marzilli, L. G. Inorg. Chem. 1993, 32, 4256. (b) Moore, S. J.; Marzilli,
L. G. Inorg. Chem. 1998, 37, 5329.
(24) (a) Rat, M.; de Sousa, R. A.; Tomas, A.; Frapart, Y.; Tuchagues,
J.-P.; Artaud, I. Eur. J. Inorg. Chem. 2003, 759. (b) Fukuhara, C.; Asato,
E.; Shimoji, T.; Katsura, M.; Matsumoto, K.; Ooi, S. J. Chem. Soc., Dalton
Trans. 1987, 1305. (c) House, D. A.; McKee, V.; Steel, P. Inorg. Chem.
1986, 25, 4884. (d) Golobic, A.; Stefane, B.; Polanc, S. Polyhedron 1999,
18, 3661.
(22) (a) Lo´pez, C.; Alvarez, S.; Solans, X.; Font-Altaba, M. Inorg. Chim.
Acta 1986, 111, L19. (b) Lo´pez, C.; Alvarez, S.; Solans, X.; Font-Altaba,
M. Inorg. Chem. 1986, 25, 2962. (c) Gilaberte, J. M.; Lo´pez, C.; Alvarez,
S.; Font-Altaba, M.; Solans, X. New J. Chem. 1993, 17, 193.

SO2 Insertion in Organocobaloximes
Organometallics, Vol. 25, No. 1, 2006 97
Figure 1. Structure of [(dpgH)2PyCoIII-µ-SO3]2CoII(MeOH)4 (12).
Table 6. Selected Bond Lengths (Å) and Bond Angles (deg)
in 12
Co1-S1
Co1-N5
Co2-O6
Co2-O8
2.2373(12)
2.044(4)
2.054(3)
2.091(3)
Co2-O9
S1-O5
S1-O6
S1-O7
2.076(3)
1.471(3)
1.496(3)
1.461(3)
O5-S1-Co1
O6-S1-Co1
O7-S1-Co1
109.43(15)
106.14(13)
108.89(14)
O6-Co2-O8
O9-Co2-O8
89.47(13)
89.93(14)
d (Å)
R (deg)
0.0086(1)
7.69
τ (twist) (deg)
72.15
X-ray Crystal Structure. A slow evaporation of solvent from
the solution of 3a and 6a (dichloromethane, benzene, and
hexane) in the refrigerator resulted in the formation of orange
crystals. The X-ray analysis of these crystals showed the
Figure 3. Structure of [C6H5CH2SO2Co(gH)2(4-tBuPy)]2 (6a).
composition as [4-CNC6H4CH2(SO2)Co(dpgH)2Py]‚2C6H6 (3a)
and [C6H5CH2(SO2)Co(gH)2(4-tBuPy)]2‚2CH2Cl2‚C6H6‚H2O (6a).
The crystal structure of 6a shows two molecules in the
asymmetric unit of the unit cell. Since there is a structural
variation in these molecules, these are numbered as 6a-I and
6a-II and the crystal data are given separately.
Diamond diagrams of the molecular structures for 3a and
for 6a-I and 6a-II along with selected numbering schemes are
shown in Figures 2 and 3, respectively. Selected bond lengths
Figure 2. Structure of 4-CN-C6H4CH2SO2Co(dpgH)2Py (3a).

98 Organometallics, Vol. 25, No. 1, 2006
Chadha et al.
Table 7. Selected Bond Lengths (Å) and Bond Angles (deg)
and Structural Data for 3a and 6a
major difference lies in the orientation of the benzyl group. The
benzyl group is perpendicular to the dioxime plane in all three
cases. In 3a and 6a-I the planes of the benzyl group and
pyridine/tBuPy ring are at angles of 29.61 and 31.04°, but in
6a-II the angle is 75.33°, which places it over toward the O-H‚
‚‚O plane.28 This is very unusual in cobaloxime chemistry, since
the axial ligands are usually perpendicular to the O-H‚‚‚O
plane. This may have resulted because of a large deviation in
the pyridine twist angle and high R value (11°) in 6a-II as
compared to those in 3a and 6a-I.
3a
6a-I
6a-II
Co-S1
2.2258(18)
1.546(8)
1.482(8)
1.656(12)
1.975(5)
116.7(4)
111.99(85)
178.49(31)
2.485(5)
2.463(6)
0.0147(2)
1.50
2.2406(11)
1.470(3)
1.462(3)
1.811(4)
2.011(3)
108.93(13)
111.78(26)
178.85(19)
2.496(4)
2.489(5)
-0.0091(5)
-3.88
2.2416(11)
1.467(3)
1.460(3)
1.821(4)
2.013(3)
107.59(13)
112.37(28)
175.58(19)
2.523(4)
2.507(4)
0.0353(5)
11.10
S1-O5
S1-O6
S-CH2
Co-N5
Co-S1-CH2(ax)
S-CH2-C
Co-N-C(ax)
O1-O4
The insertion of sulfur dioxide seems to affect the orientation
of the benzyl group; for example, a comparison of the molecular
structure of 6a with that of the parent complex BnCo(gH)2(4-
tBuPy) (6)29 shows that the orientation of the benzyl group in
6 is on the dioxime plane, as found in other benzylcobaloxime
structures also,21a whereas it is almost perpendicular to the
O2-O3
d (Å)
R (deg)
τ (twist) (deg)
88.51
83.67
79.57
and bond angles are given in Table 7. These are the first crystal
structures of SO2-inserted cobaloxime complexes.
1
dioxime plane in 6a. This has greatly affected the H NMR
All of the crystal structures clearly show a cobalt-sulfur bond
with sulfur in a tetrahedral geometry. The geometry around each
cobalt atom is distorted octahedral, with four nitrogen atoms
of the dioxime (dpgH, gH) in the equatorial plane and pyridine/
tBuPy and SO2CH2R groups in axial positions. The CH2R group
is perpendicular to the equatorial N4 plane.
shifts in the molecule, as discussed earlier. A similar comparison
with 3a cannot be made, since there is no crystal structure
known for any benzyl dpgH complex.
Conclusions
The deviation of the cobalt atom from the mean equatorial
N4 plane (d) is 0.0147(2) for 3a, -0.0091(5) for 6a-I, and
0.0353(5) for 6a-II with dihedral angles25 of 1.50, 3.88, and
11.10°, respectively. The opposite sign of d suggests that the
deviation is toward the base in 3a and 6a-II and toward the
RCH2SO2 group in 6a-I. The pyridine plane is almost parallel
to the C-C bond of the dioxime, as expected, with twist angles26
(τ) of 88.51, 83.67, and 79.57°, respectively. The Co-S/Co-N
bond lengths are 2.2258(18)/1.975(5), 2.2406(11)/2.011(3), and
2.2416(11)/2.013(3) Å in 3a, 6a-I, and 6a-II. S-CH2 bond
lengths are 1.656(12), 1.810(7), and 1.820(6) Å. The Co-S-C
bond angles are 116.7(4)° for 3a, 108.93(13)° for 6a-I and
107.59(13)° for 6a-II.
A comparison of crystal structure details in 3a, 6a-I, and
6a-II show some interesting differences. The shorter Co-S and
Co-N(axial) bond distances in 3a as compared to the values
in 6a-I and 6a-II might be due to the lower ring current in the
dioxime in 3a. The bond angle of 116.7(4)° supports this
argument and may further suggest a higher sp2 character of the
Co-S bond in 3a. The shortening of the Co-N(py) bond in 3a
is in line with the reported data in CH3Co(dpgH)2Py (2.052
Å),11c CH3Co(gH)2Py (2.064 Å),27a and ClCo(dpgH)2Py (1.965
Å).27b PhCH2SO2Co(dpgH)2Py behaves like an inorganic co-
baloxime.
The insertion of sulfur dioxide into the Co-C bond in
ArCH2Co(dioxime)2B is facile, but abstraction of halogen from
the solvent is a competing reaction. In addition, a neutral
trinuclear complex is also formed. X-ray crystallography has
structurally characterized the inserted product and the trinuclear
complex for the first time. The insertion of sulfur dioxide affects
the orientation of the benzyl group.
Acknowledgment. The work has been supported by a grant
from the CSIR (No. 01(1949)/04/EMR-II), New Delhi, India.
Supporting Information Available: Figures giving 1H and 13
C
NMR spectra for 2a and 6a, FAB mass spectrum for 12 and CIF
files giving crystal data for 3a, 6a, and 12. This material is avail-
able free of charge via the Internet at http://pubs.acs.org. The
CIF files have also been deposited with the Cambridge Crystal-
lographic Data Centre. CCDC numbers for 3a, 6a, and 12 are
278661, 278659, and 278660, respectively. Copies of the data can
be obtained free of charge from the Director, CCDC, 12 Union
Road, Cambridge CB2 1EX, U.K. (fax, +44-1223-336033; web,
http://www.ccdc.cam.ac.uk/).
OM0507171
(27) (a) Bresciani-Pahor, N.; Randaccio, L.; Zangrando, E.; Toscano, P.
J. Inorg. Chim. Acta 1985, 96, 193. (b) Lo´pez, C.; Alvarez, S.; Anguilo´,
M.; Solans, X.; Font-Altaba, M. Inorg. Chim. Acta 1987, 127, 153.
(28) These are the angles between the plane containing the S(SO2)-
C(CH2) and the six carbons of the phenyl with respect to the plane formed
by the pyridine nitrogen and the four carbons of pyridine. If we consider
the planes containing Co-N(Py)-C(para carbon of Py) and Co-S(SO2)-
C(CH2)-C(C1 of phenyl)-C(C4 of phenyl), then the angles have the values
6.77, 6.20, and 82.35° for 3a, 6a-I and 6a-II.
The structure of 6a-II differs from those of 3a and 6a-I. The
(25) The dihedral angle, R (butterfly bending angle), is the angle between
two dioxime planes.
(26) τ (known as the twist angle) is the angle between two virtual planes.
One plane is formed by considering the middle point of the C-C bond of
two oxime units that pass through cobalt and the pyridine/tBuPy nitrogen
and the other containing the pyridine nitrogen and the four carbons of
pyridine.
(29) Gupta, B. D.; Chadha, P. Unpublished data.