X-ray diffraction, Raman spectroscopic, and solid-state NMR studies of the group 14 metal-(tetracarbonyl)cobalt complexes Ph3MCo(CO)4 (M = Si, Sn, Pb)

Article abstract of DOI:10.1139/v02-110

Single-crystal X-ray diffraction studies illustrate that the three title compounds are isomorphous, belonging to the triclinic space group P1, with slightly distorted trigonal bipyramidal geometry about cobalt. The solid-state ²⁹Si, ¹¹⁹Sn, and ²⁰⁷Pb cross-polarization magic angle spinning (CP MAS) NMR spectra are presented. The indirect spin-spin coupling constant (J), quadrupolar-dipolar shift (d), direct dipolar coupling constant (D′), anisotropy in spin-spin coupling (ΔJ), and the chemical shift tensor were extracted. A plot of the reduced coupling constant vs. s-electron densities at the nucleus indicates that the Fermi contact term may be dominant for the tin and lead complexes; however, the large ΔJ for all complexes indicate that there are also significant anisotropic terms. Trends in the Raman scattering spectra are also discussed.

Full text of DOI:10.1139/v02-110

813
X-ray diffraction, Raman spectroscopic, and
solid-state NMR studies of the group 14 metal–
(tetracarbonyl)cobalt complexes Ph3MCo(CO)4
(M = Si, Sn, Pb)
J.M. Geller, J.H. Wosnick, I.S. Butler, D.F.R. Gilson, F.G. Morin,
and F. Bélanger-Gariépy
Abstract: Single-crystal X-ray diffraction studies illustrate that the three title compounds are isomorphous, belonging
to the triclinic space group P1, with slightly distorted trigonal bipyramidal geometry about cobalt. The solid-state 29Si,
119Sn, and 207Pb cross-polarization magic angle spinning (CP MAS) NMR spectra are presented. The indirect spin–spin
coupling constant (J), quadrupolar–dipolar shift (d), direct dipolar coupling constant (D>), anisotropy in spin–spin cou-
pling (ꢀJ), and the chemical shift tensor were extracted. A plot of the reduced coupling constant vs. s-electron densi-
ties at the nucleus indicates that the Fermi contact term may be dominant for the tin and lead complexes; however, the
large ꢀJ for all complexes indicate that there are also significant anisotropic terms. Trends in the Raman scattering
spectra are also discussed.
Key words: 29Si, 119Sn, and 207Pb CP MAS NMR, tetracarbonyl cobalt, spin–spin coupling, chemical shift tensor,
quadrupole coupling, Fermi contact, cobalt–group 14.
Résumé : Des études par diffraction des rayons X par des cristaux uniques permettent d’illustrer que les trois
composés mentionnés dans le titre sont isomorphes, appartenant au groupe d’espace triclinique P1, avec une géométrie
bipyramidale trigonale légèrement déformée autour du cobalt. On présente des études de RMN du 29Si, 119Sn et du
207Pb à l’état solide avec polarisation croisée à l’angle magique de rotation. On en a déduit la constante de couplage
spin-spin indirecte (J), le déplacement quadripolaire–dipolaire (d), la constante de couplage dipolaire directe (D>),
l’anisotropie du couplage spin–spin (ꢀJ) ainsi que le tenseur du déplacement chimique. Une courbe de la constante de
couplage réduite en fonction des densités des électrons s au niveau du noyau indique que le terme de contact de Fermi
peut être dominant pour les complexes de l’étain et du plomb; toutefois, les valeurs importantes de ꢀJ pour tous les
complexes indiquent que les termes anisotropiques ont aussi une grande importance. On discute aussi des tendances
dans les spectres de diffusion Raman.
Mots clés : RMN, polarisation croisée, angle magique de rotation 29Si, 119Sn et 207Pb, cobalt tétracarbonyle, couplage
spin–spin, tenseur du déplacement chimique, couplage quadripolaire, contact de Fermi, cobalt groupe 14.
[Traduit par la Rédaction] Geller et al. 820
Introduction
to increased linewidths and low sensitivity. From an NMR
perspective, the 59Co (spin = 7/2) nucleus is attractive due to
its high magnetogyric ratio (ꢁ = 6.3175 × 107 rad T–1 s–1)
and its 100% natural abundance, both of which are helpful
for high receptivity. The major drawback in the study of
cobalt NMR comes from the large quadrupole moment
(0.4 × 10–28 m2), which often leads to rapid relaxation.
Thus, in solution, direct observation of the 59Co nucleus is
limited to compounds with octahedral symmetry, i.e., with
low quadrupolar coupling constants. The spin-lattice relax-
ation times, chemical shifts, and linewidths have been re-
ported for a series of octahedral Co(III) complexes (1).
Cobalt complexes have also been studied in solution by
observing a spin –1/2 nucleus bonded to the cobalt.
Solution-phase NMR studies of 13CO-enriched tetracarbonyl
cobalt – group 14 complexes have illustrated that a
downfield shift occurs for the axial carbonyl with respect to
the radial CO on going down the group, and an axial–radial
exchange process occurs that can be resolved at lower
The study of transition metal complexes by NMR is wide-
spread since the technique can give information about the
electronic environment and bonding, but studies of many
transition metals have been limited because they have nu-
clear spins >1/2 and large quadrupole moments, which lead
Received 14 December 2001. Published on the NRC Research
Press Web site at http://canjchem.nrc.ca on 28 June 2002.
J.M. Geller, J.H. Wosnick, I.S. Butler,1 D.F.R. Gilson,1
and F.G. Morin. Department of Chemistry, McGill
University, 801 Sherbrooke Street West, Montreal, QC H3A
2K6, Canada.
F. Bélanger-Gariépy. Université de Montréal, Départment
de chimie, C.P. 6128 Succursale Centre-ville, Montréal QC
H3C 3J7, Canada.
1Corresponding authors (e-mail: denis.gilson@mcgill.ca and
ian.butler@mcgill.ca).
Can. J. Chem. 80: 813–820 (2002)
DOI: 10.1139/V02-110
© 2002 NRC Canada

814
Can. J. Chem. Vol. 80, 2002
Fig. 1. ORTEP diagram of Ph3SiCo(CO)4 showing 30% thermal
probability ellipsoids.
temperatures (2). Compounds of the form (C6H5)3MCo(CO)4,
where M is a group 14 element, have been extensively stud-
ied (3–6). The solution 207Pb NMR spectra were studied for
the series R3PbCo(CO)4 (R = methyl, ethyl, phenyl), and the
207Pb chemical shifts display a difference of over 200 ppm
on changing the R group from methyl to phenyl (7).
Solid-state NMR spectroscopy allows for a wider variety
of compounds to be studied since the quadrupolar relaxation
is slower in the solid phase. Similarly, complexes with axial
symmetry are also amenable to examination. Direct excita-
tion of the 59Co central transition (+1/2 to –1/2) in solids has
been reported; 59Co NMR static spectra have been obtained
for cobaltophthalocyanines (8), porphyrins (9, 10), and vita-
min B12 (11). Most recently, the chemical shift tensors and
orientations of the electric field gradients (EFG) were deter-
mined for a series of octahedral Co(III) complexes (12). In a
number of porphyrin systems with axial symmetry, the 59Co
magic angle spinning (MAS) spectra were acquired directly
because the quadrupolar–dipolar broadening was small (9).
Both solid- and solution-state 59Co NMR have been recently
reviewed (13).
the Fermi contact dominance in spin–spin coupling can also
be established for the I–Co spin pairs.
The 59Co nuclear quadrupole coupling constants (ꢂ) of the
triphenyl silicon, tin, and lead tetracarbonyl cobalt com-
plexes have been determined from single-crystal NMR stud-
ies. Differences in dꢃ–pꢃ back-bonding to the group 14
metal (M) were postulated to influence the amount of para-
magnetic shielding experienced by the 59Co nucleus (14).
These quadrupole coupling constants were in good agree-
ment with the pure nuclear quadrupole resonance (NQR) fre-
quencies reported by Brown et al. (15).
Single-crystal X-ray diffraction methods were used to de-
termine accurate molecular geometries for this series of
compounds. The structure of the triphenylsilyl complex has
been reported previously by McIndoe and Nicholson (26).
The approximate C3v symmetry of the carbonyl groups
around the cobalt nucleus led to an early interest in the vi-
brational spectra, with IR spectral studies of the carbonyl
stretching frequencies (5), force constant calculations (6),
and additional calculations by 12C–16O direct CO-factored
methods (27). In this work, the Raman scattering spectra for
the three complexes are reported.
Various compounds can be examined if the 59Co nucleus
is studied indirectly through a spin 1/2 nucleus probe (I).
The interest in I–Co spin pairs stems from the measurement
of indirect spin–spin couplings (JI,Co), which are not ob-
served in analogous solution NMR experiments due to rapid
relaxation and “self-decoupling” of the quadrupolar nucleus.
The 31P–59Co and 15N–59Co spin pairs have been studied by
this method (16–21). Residual dipolar coupling is also mani-
fested in the MAS NMR spectrum of nucleus I by a skewing
of the multiplet resulting from the J coupling between the I
and S nuclei. More specifically, instead of observing 2S + 1
evenly spaced peaks in the multiplet, the peaks are shifted so
that the spacings are uneven. Analysis of spectra exhibiting
quadrupolar-induced dipolar splitting can also yield informa-
tion on the EFG tensor at the quadrupolar nucleus, the mag-
nitude and sign of the quadrupolar coupling constant (ꢂ),
the direct dipolar coupling constant (D), anisotropy in spin–
spin coupling (ꢀJ), and the chemical shift tensor of nucleus
I. A treatment based on first-order perturbation theory has
been developed to interpret the MAS spectra of nucleus I
(22–24).
Results and discussion
X-ray crystallography
Complexes I–III are isomorphous, crystallize in the space
group P1, and have one molecule per asymmetric unit. The
ORTEP (36) diagrams are shown in Figs. 1–3, and details of
the crystal data and structure refinements are given in Ta-
ble 1. There are very few reports of Si–Co bond lengths
studied by X-ray diffraction methods. In the trichlorosilyl
and trifluorosilyl tetracarbonyl cobalt complexes, the Si—Co
bond lengths are 2.254(3) and 2.226(5) Å, respectively (37,
38). These are significantly shorter than the Si—Co bond
lengths for I (2.3799(14) Å) that we obtained, or the value
of 2.3810(7) Å previously reported (26) for the same com-
pound. The Co—Sn bond length in II (2.5950(8) Å) is within
the range reported for the series Cl4–xSn[Co(CO)4]x (x =
1ꢄ4), where the Sn—Co distances increase from 2.477(1) Å
for Cl3SnCo(CO)4 (39) to 2.669(1) Å for Sn[Co(CO)4]4 (39).
The Pb—Co distance in III is 2.667(2) Å; the only other
Pb—Co bond lengths reported are 2.761(5) and 2.738(5) Å
for the two independent Pb—Co distances in Pb[Co(CO)4]4
(40).
In this paper, we present the 29Si, 119Sn, and 207Pb
solid-state CP MAS NMR spectra of the triphenyl-substituted
group-14 cobalt(I) tetracarbonyls. These compounds were
chosen because they contain a spin–spin (I–Co) pair with ax-
ial symmetry surrounding the cobalt nucleus. This work fol-
lows from our previous study on triaryl silicon–, tin–, and
lead–managanese pentacarbonyls, where we illustrated that
the Fermi contact term was the dominant mechanism in
spin–spin coupling (25). We now make a comparison be-
tween this 59Co series and the 55Mn series to determine if
The geometry around the Co atom in all three complexes
is a slightly distorted trigonal bipyramidal. For I–III, the
M-Co-COax angles are linear: 179.07(9), 178.9(2), and 179.1(5)°,
respectively. There is some distortion of the equatorial
carbonyls towards the Ph3M ligand and away from the axial
© 2002 NRC Canada

Geller et al.
815
Fig. 3. ORTEP diagram of Ph3PbCo(CO)4 showing 30% thermal
Fig. 2. ORTEP diagram of Ph3SnCo(CO)4 showing 30% thermal
probability ellipsoids.
probability ellipsoids.
Table 1. Crystal structure data and structural refinements.
I (Si)
II (Sn)
III (Pb)
C22H15CoO4Si
430.360
Colourless, needle
0.40 × 0.09 × 0.04
Triclinic
P1
9.859(8)
10.311(5)
10.681(5)
85.64(4)
C22H15CoO4Sn
520.960
Pale yellow, block
0.27 × 0.17 × 0.15
Triclinic
C22H15CoO4Pb
609.470
Pale yellow, block
0.27 × 0.11 × 0.08
Triclinic
Formula
MW (g mol–1)
Crystal color, habit
Crystal Size (mm)
Crystal System
Space group
a (Å)
P1
P1
10.1203(1)
10.5225(1)
11.0554(1)
85.40(1)
10.110(5)
10.500(8)
10.956(4)
84.62(5)
b (Å)
c (Å)
ꢅ (°)
ꢆ (°)
69.25(5)
69.03(6)
69.71(3)
ꢁ (°)
81.57(5)
1004.0(11)
2
1.4235
1.54056
7.484/220(2)
Nonius CAD-4
ꢈ–2ꢉ
81.93(5)
1087.92(1)
2
1.5903
1.54178
15.296/293(2)
Bruker AXS SMART 2K
ꢈ
81.45(6)
1077.6(11)
2
1.8783
1.54056
21.534/293(2)
Nonius CAD-4
ꢈ–2ꢉ
V (Å3)
Z
Dcalcd. (Mg m–3)
ꢇ (Å)
(mm–1)/T (K)
Diffractometer
Scan type
Measured reflections,
Unique reflections, Rint
Observed data, criterion
Absorption correction
Transmission range
ꢉmax (°)
28853
13172
15142
3802, 0.047
2572, I > 2ꢊ(I)
Analyticala
0.2740–0.7528
69.80
4147, 0.0413
3688, I > 2ꢊ(I)
Multiscanb
0.0590–0.2210
72.72
4081, 0.053
3421, I > 2ꢊ(I)
Analyticala
0.0816–0.3022
69.80
Range measured
–12 < h < 12
–12 < k < 12
–13 < l < 13
F2
0.0304/0.0492
0.0516/0.0545
254
–11 < h < 9
–12 < k < 12
–13 < l < 13
F2
0.0403/0.0439
0.1059/0.1086
254
–12 < h < 12
–12 < k < 12
–13 < l < 13
F2
0.0520/0.0631
0.1318/0.1514
254
Refinement on
R1 (obs/all)
wR2 (obs/all)
Refined parameters
S
0.866
1.018
1.002
Extinction coefficient
0.0030(1)
0.0010(2)
0.0052(4)
Note: R1 = ꢋ(*Fo* – *Fc*)/ꢋ*Fo*; wR2 = {ꢋ[w(F o2 – F c2 )2]/ꢋ[w(F o2)2]}1/2; w = 1/ꢊ2(F 2o). S = {ꢋ[w (F o2 – F 2c )2]/(n – p)}1/2; n = number of data; p =
number of refined parameters.
aReference 34.
bReference 35.
© 2002 NRC Canada

816
Can. J. Chem. Vol. 80, 2002
Table 2. Selected bond lengths (Å) and angles (°) determined
Table 3. Raman vibrational frequencies of solid compounds (cm–1).
from X-ray diffraction.
Mode/bond
I (Si)
II (Sn)
III (Pb)
I (Si)
II (Sn)
III (Pb)
85 s
84 s
83 m
Bond lengths (Å)
M—Co
Co—C1
Co—C2
Co—C3
Co—C4
M—C11
M—C21
M—C31
Bond angles (؇)
C1-Co-C2
C1-Co-C3
C1-Co-C4
C2-Co-C3
C2-Co-C4
C3-Co-C4
C1-Co-M
C2-Co-M
112 m
189 w
206 w
239 w
422 m
539 w
619 w
675 w
706 w
1994, 2008 m
2037 s
2093 m
109 s
107 s
2.3799(14)
1.809(3)
1.776(3)
1.786(3)
1.788(3)
1.878(2)
1.876(3)
1.873(3)
2.5953(7)
1.808(6)
1.790(5)
1.790(5)
1.792(5)
2.152(4)
2.139(4)
2.144(4)
2.667(2)
ꢌM–Co
171 m
212 m
237 w
422 m
549 w
617 w
655 m
699 w
1989, 2001 m
2030 s
2087 m
155 m
202 m
221 w
418 m
544 w
615 w
646 m
695 w
1988, 1997 m
2023 s
2080 m
1.813(13)
1.750(15)
1.785(13)
1.798(12)
2.202(10)
2.205(9)
2.217(11)
ꢌCO
e
(1)
(2)
ꢌCO a1
ꢌCO a1
96.67(12)
96.47(12)
95.49(13)
117.20(12)
117.99(12)
121.38(12)
179.07(9)
84.34(9)
97.4(3)
98.0(3)
96.2(3)
116.6(3)
118.3(3)
120.5(2)
179.2(2)
83.2(2)
82.2(2)
83.0(2)
107.9(2)
108.7(2)
109.2(2)
44.7(5)
53.5(4)
–131.6(3)
44.3(2)
38.4(2)
40.1(2)
97.1(6)
97.5(6)
96.9(6)
116.0(6)
119.0(6)
120.3(6)
179.1(5)
83.7(4)
group 14. The small decrease in the average of the ꢌCO fre-
quencies on going from silicon to lead suggests a decrease
in multiple bond character for the M–Co bonds. The ꢌCO
e-mode is split due to crystal effects.
C3-Co-M
C4-Co-M
83.28(9)
81.8(4)
82.9(4)
Solid-state CP MAS NMR
83.89(10)
107.60(11)
108.72(11)
108.87(11)
45.8(2)
A treatment based on first-order pertubation theory has
been developed by Olivieri (23), which is valid in the limit
ꢂ/[4S(2S – 1)ꢌs] << 1, where ꢌs is the Larmor frequency of
the quadrupolar nucleus. Thus, for 59Co (S = 7/2), provided
that ꢂ/84ꢌs << 1, the splittings observed in the solid-state
spectra under MAS can be treated by first-order perturbation
theory. With the assumption that that the J tensor is co-linear
with the internuclear dipole vector, the line splittings are
given by eq. [1],
C11-M-C21
C11-M-C31
C21-M-C31
Co-M-C11-C12
Co-M-C21-C22
Co-M-C31-C32
C2-Co-M-C11
C3-Co-M-C21
C4-Co-M-C31
108.8(4)
110.0(4)
110.5(4)
48.7(12)
54.4(10)
–129.7(8)
39.2(5)
53.4(2)
–129.1(2)
39.03(12)
37.42(11)
39.67(12)
37.0(5)
38.7(5)
[1]
ꢀꢌm = –msJIS + (3ꢂD>/20ꢌs)(3cos2ꢆD – 1
carbonyl group, so that the average M-Co-COeq angles for
the three complexes are 83.8(5), 82.8(5), and 83(1)°, respec-
tively. Such a distortion from a trigonal bipyramidal struc-
ture towards a capped tetrahedron has been previously
discussed (41). The average COeq-Co-COeq angles in the
three complexes are 119(2), 118(2), and 118(2)°, respec-
tively, further confirming the threefold symmetry surround-
ing the Co nucleus. In the three complexes, the ipso-carbon
atoms of the triphenyl groups are rotated 37ꢄ43° (Table 2)
from the ideal staggered position of 60° with respect to the
triangle formed by the three equatorial carbonyl groups. This
difference is attributed to the steric interaction between car-
bonyl carbons and the ortho-carbon atom of the phenyl
rings. The closest intermolecular non-bonded Cortho—Ccarbonyl
distances are between 3.4 and 3.6 Å. The important bond
lengths and angles for complexes I–III are listed in Table 2.
+ ꢍsin2ꢆDcos2ꢅD)[{S(S + 1)
– 3m2s }/{S(2S – 1)}]
where D> = (D – ꢀJ/3) and the second-order quadupolar–
dipolar splitting (d) is 3ꢂD>/20ꢌs. D is the direct dipolar cou-
pling constant (D = ( o/4ꢃ)ꢁsꢁIh2/(4ꢃ(rIS)3) and ꢀJ is the
anisotropy in the spin–spin coupling. ꢅD and ꢆD are, respec-
tively, the azimuthal and polar angles relating the
internuclear vector to the principle axis of the EFG and ꢍ is
the asymmetry in the field gradient. Equation [1] assumes
that the J tensor is co-linear with the internuclear dipole vec-
tor. From NQR studies (15), ꢍ was found to be 0.05 or less,
indicating a near axially symmetric EFG about cobalt for all
three complexes. The 59Co (S = 7/2) resonance frequency at
7.05 T is 71.2 MHz, and the nuclear quadrupole coupling
constants for these complexes range from 101 to 110 MHz
at 298 K (14, 15). Thus, the first-order perturbation treat-
ment is applicable for the three complexes studied here.
The local symmetry about the M–Co bonds (approxi-
mately C3v) is sufficiently high and ꢍ is close enough to zero
that it can be assumed that the angle ꢆD is 0° and the value
of ꢅD is undefined, because it is multiplied by sin2ꢆD, which
is 0. Hence, eq. [1] is reduced to the simple form
Raman spectroscopy
The IR spectra of I–III have been previously reported in
the C–O stretching (5, 6, 27) and far-IR (42) regions, and the
Raman spectra (Table 3) complement these IR data. Only the
tin–cobalt IR stretching frequency has been previously as-
signed (42) and appears at 171 cm–1. Based on this assign-
ment, the ꢌM-Co values for the silicon and lead complexes
have been assigned as shown in Table 3. For the ꢌCO vibra-
tions, there is a decrease in wavenumber on going down
[2]
ꢀꢌm = –msJIS + (3ꢂD>/10ꢌs)[3/4 – 1/7m2s ]
© 2002 NRC Canada

Geller et al.
817
Table 4. NMR data for the silicon, tin, and lead complexes.
Fig. 4. CP MAS NMR spectra of I–III at 7.05 T. The three
spectra were plotted on the same horizontal axis (in ppm) to il-
lustrate the increase in span on going from 29Si to 207Pb. For the
chemical shifts, see Table 4.
I (29Si)
II (119Sn)
III (207Pb)
26.0(0.2)
26.3(0.2)
57(1)
20.5(0.2)
35.3(0.5)
57(2)
127(1)
193(1)
110(1)
1.85 × 1021
–2(2)
360(5)
138(10)
80(5)
280(5)
110.82
314(4)
950(10)
ꢎiso (solution) (ppm)
ꢎiso (solid) (ppm)
1JM,Co (Hz)
1KM,Co (T2 J–1)
D (Hz)
–8.92 × 1020 –5.39 × 1020
47(2)
—
—
—
—
101.09
ꢄ417(4)
–1580(20)
127(5)
92(4)
52(1)
ꢄ41(5)
132(9)
104.11
–606(3)
–2690(30)
ꢎ11 (ppm)
ꢎ22 (ppm)
ꢎ33 (ppm)
Span (ꢎ11–ꢎ33) (ppm)
ꢂ (MHz)a
D> (Hz)b
ꢀJ (Hz)c
aFrom ref. 14. The frequencies are ±0.05 MHz.
bCalculated from the M–Co bond lengths given in Table 2, errors
calculated using ±5ꢊ.
cStandard deviation propagated from d.
and the second-order quadupolar-dipolar splitting (d) given
by 3ꢂD>/10ꢌs, can be obtained.
shielding range of 207Pb previously obtained (51) from a plot
of ꢎ(207Pb) vs. ꢎ(119Sn) isotropic chemical shifts, where a
slope of 3.0 was obtained. Unexpectedly large chemical
shielding ranges for heavy nuclei have been previously dis-
cussed (52).
The spectrum of III is quite different from the spectra of I
and II, showing an almost symmetric octet. The lines of the
octet are almost evenly spaced, which is indicative of a
small d value. The quadrupolar–dipolar shift term is small
because the effective D> is small. Since the value ꢂ for III
was determined (110.82 MHz), D> can be obtained using
eq. [1], and because the direct dipolar coupling constants
can be calculated from the measured bond lengths, ꢀJ can
be extracted from eq. [2].
The differences between the isotropic solid and solution
chemical shifts are greatest for the lead complex and small-
est for silicon. This trend can be attributed to the larger
range of observed chemical shifts for 207Pb and 119Sn rather
than structural differences between the three complexes.
Very few instances of direct measurement of the spin–spin
couplings of I–Co spin pairs have been reported. The com-
pounds most frequently studied are highly symmetric with
octahedral or tetrahedral symmetry, e.g., Na[Co(CO)4] (53)
and Na3[Co(CN)6] (54). Coupling between 59Co and 13C and
31P has been indirectly measured by T1 relaxation measure-
ments in solution (55, 56) and in the solid state (57). The
one-bond coupling constants to cobalt reported here are rare
because these complexes have large quadrupole coupling
constants, and are less symmetrical than the complexes pre-
viously reported.
The contributions to the spin–spin coupling constant arise
from the diamagnetic spin-orbit (DSO), paramagnetic
spin-orbit (PSO), spin-dipolar term (SD), Fermi contact
(FC), and a spin-dipolar – Fermi contact (SD–FC) cross
term. The DSO and PSO contributions are usually small,
have opposite signs to each other, and tend to cancel, leaving
the latter major terms (58). To compare the spin–spin cou-
pling constants of different nuclei, the reduced coupling con-
The NMR data for I–III, are given in Table 4, and the full
spectra of all of the compounds are given in Fig. 4. In all
cases, only a single chemical shift was observed, which is
consistent with the crystal structures. In the case of the
triphenylsilyl complex, the spectrum has only a centreband
at the spinning speeds used. The (ꢎ11–ꢎ33) span of the spec-
trum is therefore very small, as was observed previously for
the triphenylsilyl(pentacarbonyl)manganese(I) complex (25).
The centreband should exhibit eight lines, but the downfield
lines overlap, which is characteristic of a d/J ratio that is
close to, or greater than, one. Since the lines overlap in the
downfield region, the sign of d is positive.
The 119Sn CP MAS spectrum of the triphenyltin complex
has a much larger span and the components of the chemical
shift tensor could be calculated, using the Herzfeld–Berger
method, from the relative intensities of the spinning side-
bands (28, 43). The span of the spectrum (133 ppm) is
smaller than the values reported (25) for Ph3SnMn(CO)5
(187–209 ppm) and for the less-symmetrical diaryl com-
plexes (44, 45). The centreband shows an overlap of the
downfield lines, characteristic of a d/J ratio close to one.
Since 207Pb spectra usually show a very wide range of
chemical shifts, it is surprising that the span of the triphenyl
lead complex is only 280 ppm. This value is similar to those
obtained for the triphenylplumbyl(pentacarbonyl)manga-
nese(I) complex (25), where the spans ranged from 268 to
329 ppm. It is notable that both the lead–cobalt and the
lead–manganese complexes have smaller spans than the
more symmetrical hexaorganyldiplumbanes, which have
spans ranging from 484 to 996 ppm (46). Comparison of the
spans for the tin and lead complexes showed the expected
increase as the size of the valence p-electron radius (<r–3>np)
increases. According to the expression for paramagnetic
shielding suggested by Ramsey (47ꢄ49), the ratio of the
spans should correlate well with the <r–3>6p/<r–3>5p ratio of
1.4 previously reported (50). The span of the lead complex
is 2.1 times greater than that of the tin complex. Our results
are not inconsistent with the greater-than-expected chemical
© 2002 NRC Canada

818
Can. J. Chem. Vol. 80, 2002
stants (KI,J = 4ꢃ2JI,J/(hꢁI ꢁJ)) are used. For the FC term, KI,J
is proportional to the product of the s-electron densities at
the nuclei and the s-character of the hybrid orbitals used to
form the bond, all divided by an average excitation energy
(ꢀE) (59, 60). Using the s-electron densities given by
Pyykkö and Weisenfeld (61), with a non-relativistic value
for silicon and relativistic values for the heavier elements, a
plot of the reduced coupling constants against the s-elec-
tron densities (*ꢏ(0)*2) for the silicon, tin, and lead nuclei
showed a linear relationship, passing through the origin, for
the triphenyl complexes with manganese pentacarbonyl (25).
In the present series of compounds, the reduced coupling
constants for the tin and lead complexes also showed a lin-
ear dependence extrapolating to the origin, but the silicon
value was too high (Table 4). This, combined with the large
ꢀJ values for all complexes, suggests that, in addition to the
FC term, there are appreciable anisotropic contributions to
the J tensor most probably from the SD–FC term. Large ꢀJ
values were also observed for the manganese pentacarbonyl
complexes (25) and for other heavy-atom cases, such as
199Hg–31P coupling (62–65). Theoretical calculations of ꢀJ
have been limited to light-atom systems, but even in the case
of 13C–13C coupling in benzene, ethane, ethene, and ethyne
(66, 67) there is appreciable anisotropy, but the FC term is
still the major contribution.
all sidebands were fitted with Lorentzian line shapes using
the program Peakfit (Jandel Scientific) and the errors given
are from the averages of the fitted parameters. Calculation of
the chemical shift tensors were performed with the aid of a
computational package developed by Eichele and
Wasylishen (28).
Ph3SiCo(CO)4 (I)
Compound I was synthesized using the reaction first de-
scribed by Chalk and Harrod (3) and later improved by
McIndoe and Nicholson (26). Co2(CO)8 (2.38 g, 6.95 mmol)
and Ph3SiH (6.79 g, 6.95 mmol) were reacted in hexanes
(30 mL) at room temperature for 90 min. After cooling to
0°C (ice bath), an off-white solid was precipitated. The solid
was isolated by Buchner filtration and washed with cold
hexanes. Recrystallization from hot hexanes yielded clear
crystals of the desired compound (3.71 g, 62%); mp 138–
140°C (lit. (1) 135–140°C). IR ꢌCO (hexanes, cm–1): 2093
(m), 2033 (m), 2005 (s); (CH2Cl2, cm–1): 2094 (s), 2033 (m),
1993 (s). 13C NMR ꢎ: 198.7 (s, CO), 138.2 (s, ipso),
135.4 and 127.9 (s, ortho and meta), 129.6 (s, para). The IR
and 13C NMR spectra were in agreement with previously
published results (26).
Ph3SnCo(CO)4 (II)
Compound II was synthesized according to the method
described by Darensbourg (6). Co2(CO)8 (4.00 g,
11.7 mmol) was dissolved in methanol and the solution was
gently warmed to 50°C and stirred until the pink colour of
[Co2+][Co(CO)4ꢄ]2 was observed and gas evolution ceased.
The reaction mixture was then cooled to 0°C (ice bath) and a
methanolic solution of Ph3SnCl (4.20 g, 10.9 mmol) was
added. The reaction mixture was stirred for 1 h at room tem-
perature and then cooled to 0°C, which resulted in the pre-
cipitation of a white solid. The solid was isolated, washed
with cold methanol, and then dried by vacuum filtration.
Recrystallization from hot hexanes yielded large yellow
crystals (3.97 g, 70%); mp 118–121°C (lit. (5) 119–121°C).
IR ꢌCO (hexanes, cm–1): 2087 (m), 2026 (m), 1997 (s);
(CH2Cl2, cm–1): 2088 (s), 2026 (m), 1994 (s). 13C NMR ꢎ:
Experimental
General
All reactants were used as received from Aldrich or Strem
[Co2(CO)8]. With the exception of THF (which was dried
over sodium with benzophenone indicator), solvents were
not dried before use,. All reactions were performed under ni-
trogen with subsequent work-up under air.
IR spectra were measured (100 scans) in either CH2Cl2 or
hexanes solutions between KBr plates on a Bruker IFS-48
FT-IR spectrometer at 1.0 cm–1 resolution. Raman spectra
were recorded (100 scans) on a Bruker IFS-88 FT-Raman
spectrometer with samples contained in a metal cup at
4.0 cm–1 resolution. Solution-state 13C, 29Si, and 119Sn pro-
ton-decoupled NMR spectra were measured in CDCl3 solu-
tions at room temperature (21°C), on a Varian XL-400
spectrometer operating at a field strength of 9.40 T. The
207Pb spectrum was measured on a JEOL 270 MHz spec-
trometer operating at 6.34 T. Solid-state proton-decoupled
CP MAS spectra were obtained on a Chemagnetics
CMX-300 spectrometer operating at 59.52 MHz (29Si),
111.73 MHz (119Sn), and 63.10 MHz (207Pb). Experimental
conditions I: contact time 2 ms; pulse delay 750 s; 128 tran-
sients. II: contact time 2 ms; pulse delay 20 s; 700 tran-
sients. III: contact time 2 ms; pulse delay 3 s; 18644
transients. Chemical shifts were externally referenced to
198.8 (s, CO), 140.2 (ipso, 1J
= 459, 1J
=
119Sn,C
117Sn,C
3
439 Hz), 128.8 (meta, JSn,C = 53.4 Hz), 136.16 (ortho,
2JSn,C = 41.2 Hz), 129.4 (para, JSn,C = 12.2 Hz).
4
Ph3PbCo(CO)4 (III)
Compound III was synthesized in an analogous manner to
complex II from Co2(CO)8 (4.00 g, 8.19 mmol) and Ph3PbCl
(5.07 g, 10.7 mmol), as described by Patmore and Graham
(5). Upon recrystallization from hexanes, light-yellow
block-shaped crystals were obtained (4.69 g, 72% yield); mp
102°C (lit. (6) 100–102°C). IR ꢌCO (hexanes, cm–1): 2079
(s), 2020 (m), 1993 (s); (CH2Cl2, cm–1): 2082 (s), 2020 (m),
1
1994 (s). 13C NMR ꢎ: 199.4 (s, CO), 152.9 (ipso, JPb,C
=
tetramethylsilane (29Si), tetramethyltin
(
119Sn), and
3
tetramethyllead (207Pb), respectively. Powdered samples
were contained in 7.5-mm diameter zirconia pencil-type ro-
tors. FIDs on the CMX-300 were zero-filled to 8K data
points. Spinning rates ranging from 2ꢄ4 kHz were used to
identify isotropic chemical shifts. The chemical shifts of
207Pb resonances can be temperature-dependent, but the cen-
tre band position did not vary with spinning speed and,
therefore, no heating effects occurred. The resonance lines in
314 Hz), 129.8 (meta, JPb,C = 80.5 Hz), 136.5 (ortho,
4
2JPb,C = 72.1 Hz), 128.9 (para, JPb,C = 19.2 Hz).
X-ray crystallography
All complexes were recrystallized by dissolving in boiling
hexanes and then cooling to –30°C, where, after several
days, X-ray quality crystals were obtained. Details of the
crystal data and structure refinements are given in Table 1.
© 2002 NRC Canada

Geller et al.
819
20. R. Gobetto, R.K. Harris, and D.C. Apperley. J. Magn. Reson.
96, 119 (1992).
21. D. Rentsch, R. Hany, and W.V. Philipsborn. Magn. Reson.
Chem. 35, 832 (1997).
22. A.C. Olivieri, L. Frydman, and L.E. Diaz. J. Magn. Reson. 75,
50 (1987).
23. A.C. Olivieri. J. Magn. Reson. 81, 201 (1989).
24. R.K. Harris and A.C. Olivieri. Prog. Nucl. Magn. Reson.
Spectrosc. 24, 534 (1992).
Data for I and III were processed using a locally modified
version of the NRC-2 program (29) and data for II was
processed by SAINT (30).2 All structures were solved by di-
rect methods using SHELXS-97 (31) and refined using
SHELXL-96 (32). Non-hydrogen atoms were refined with
anisotropic thermal parameters. The hydrogen atoms were
included in calculated positions. Scattering factors were
from the usual sources (33). I was run at low temperature
(220 K) because it was prone to decomposition under X-rays
at room temperature.
25. D. Christendat, I.S. Butler, D.F.R. Gilson, and F.G. Morin.
Can. J. Chem. 77, 1892 (1999).
26. J.S. McIndoe and B.K. Nicholson. J. Organomet. Chem. 577,
181 (1999).
27. C. Kaya. J. Organomet. Chem. 2, 586 (1999).
28. HBA, version 1.44. K. Eichele and R.E. Wasylishen.
Dalhousie University, Halifax. 2001.
29. F. Ahmed, S.R. Hall, M.E. Pippy, and C.P. Huber. NRC crys-
tallographic computer Programs for the IBM/360. Accession
Nos. 133–147. J. Appl. Cryst. 6, 309 (1973).
30. SAINT Release 6.06. Integration software for single crystal
data. Bruker AXS Inc., Madison, WI. 1999.
Acknowledgments
This work was supported by grants from the Natural Sci-
ences and Engineering Research Council of Canada
(NSERC) and Fonds pour la formation de chercheurs et
l’aide à la recherche (Québec) (FCAR).We thank Dr. A.-M.
Lebuis for assistance in examining the crystal structures and
Professor R.E. Wasylishen for the HBA program.
31. G.M. Sheldrick. SHELXS-97. Program for the solution
of crystal structures. University of Gottingen, Germany.
1997.
32. G.M. Sheldrick. SHELXL-96. Program for the refinement
of crystal structures. University of Gottingen, Germany.
1997.
33. International tables for crystallography. Vol. C. 4. Kluwer
Acacdemic Publishers, Dordrecht. 1992. Tables 4.2.6.8 and 6.1.1
34. E.J. Gabe, Y. Le Page, J.-P. Charland, F.L. Lee, and P.S.
White. J. Appl. Cryst. 22, 384 (1989).
35. G.M. Sheldrick. SADABS. Bruker area detector absorption
corrections. Bruker AXS Inc., Madison, WI. 1996.
36. C.K. Johnson. ORTEPII. Report ORNL-5138. Oak Ridge
National Laboratory, Tennessee, U.S.A.
37. W.T. Robinson and J.A. Ibers. Inorg. Chem. 6, 1209 (1967).
38. K. Emerson, P.R. Ireland, and W.T. Robinson. Inorg. Chem. 9,
436 (1970).
39. P.Z. Klufers. Naturforsch. Teil B, 46, 187 (1991).
40. J.S. Leigh and K.H. Whitmire. Acta Crystallogr. Sect. C, 46,
732 (1990).
41. S.A. Jackson, O. Eisenstein, J.D. Martin, A.C. Albeniz, and
R.H. Crabtree. Organometallics, 10, 3062 (1991).
42. L.F. Wuyts and G.P. van der Kelen. Spectrochim. Acta, 32A,
689 (1976).
43. J. Herzfeld and A.E. Berger. J. Chem. Phys. 73, 6021 (1980).
44. R.K. Harris, S.E. Lawrence, S.-W. Oh, and V.G. Kumar Das. J.
Mol. Struct. 347, 309 (1995).
45. D. Christendat, J.H. Wosnick, I.S. Butler, D.F.R. Gilson,
A.-M. Lebuis, and F.G. Morin. J. Mol. Struct. 559, 235 (2001).
46. A. Sebald and R.K. Harris. Organometallics, 9, 2096 (1990).
47. N.F. Ramsey. Phys Rev. 77, 567 (1950).
48. N.F. Ramsey. Phys Rev. 78, 699 (1950).
49. N.F. Ramsey. Phys Rev. 83, 540 (1951).
50. R.G. Barnes and W.V. Smith. Phys. Rev. 93, 95 (1954).
51. J.D. Kennedy, W. McFarlane, and G.S. Payne. J. Chem. Soc.
Dalton Trans. 2332 (1977).
52. C.J. Jameson. In Multinuclear NMR spectroscopy. Edited by
J. Mason. Plenum, New York. 1987. pp. 64–65.
References
1. C.W. Kirby, C.M. Puranda, and W.P. Power. J. Phys. Chem.
100, 14618 (1996).
2. D.L. Lichtenberger and T.L. Brown. J. Am. Chem. Soc. 99,
8187 (1977).
3. A.J. Chalk and J.F. Harrod. J. Am. Chem. Soc. 87, 1133
(1965).
4. F. Hein, P. Kleinert, and W. Jehn. Naturwissenschaften, 44, 34
(1957).
5. D.J. Patmore and W.A.G. Graham. Inorg. Chem. 5, 981
(1967).
6. D.J. Darensbourg. Inorg. Chim. Acta, 4, 597 (1970).
7. K.H. Pannell, J.M. Rozell, S. Cortez, and R. Kapoor.
Organometallics, 9, 1322 (1990).
8. A. Medek, V. Frydman, and L. Frydman. J. Phys. Chem. A,
103, 4830 (1999).
9. A. Medek, V. Frydman, and L. Frydman. J. Phys. Chem. B,
101, 8959 (1997).
10. X. Xiao-Ping and S.C.F. Au-Yeung. J. Am. Chem. Soc. 122,
6468 (2000).
11. A. Medek and L. Frydman. J. Am. Chem. Soc. 122, 684
(2000).
12. C.W. Kirby and W.P. Power. Can. J. Chem. 79, 296 (2001).
13. J.C.C. Chan and S.C.F. Au-Yeung. Ann. Rep. NMR Spec. 41,
1 (2000).
14. H.W. Spiess and R.K. Sheline. J. Chem. Phys. 53, 3036
(1970).
15. T.L. Brown, P.A. Edwards, C.B. Harris, and J.L. Kirsch. Inorg.
Chem. 4, 763 (1969).
16. R.W. Schurko, R.E. Wasylishen, and J.H. Nelson. J. Phys
Chem. 100, 8057 (1996).
17. F. Asaro, R. Gobetto, L. Lorenza, and G. Pellizer. Chem. Phys.
Lett. 300, 414 (1999).
18. R.W. Schurko, R.E. Wasylishen, S.J. Moore, and L.G. Marzilli.
Can. J. Chem. 77, 1973 (1999).
19. R.W. Schurko and R.E. Wasylishen. J. Phys. Chem. A, 104,
3410 (2000).
2 Crystallographic information has also been deposited with the Cambridge Crystallographic Data Centre (CCDC Nos. 169557, 169556,
169558 for I–III, respectively). Copies of the data can be obtained free of charge via www.ccdc.cam.ac.uk/conts/retrieving.html (or from
the Cambridge Crystallographic Data Centre, 12, Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033; or de-
posit@ccdc.cam.ac.uk).
© 2002 NRC Canada

820
Can. J. Chem. Vol. 80, 2002
53. E.A.C. Lucken, K. Noack, and D.F. Williams. J. Chem. Soc.
A, 148 (1967).
54. V.P. Chacko and R.G. Bryant. J. Magn. Reson. 57, 79 (1984).
55. P. Kofod, P. Harris, and S. Larsen. Inorg. Chem. 36, 2258
(1997).
61. P. Pyykkö and L. Weisenfeld. Mol. Phys. 43, 557 (1981).
62. M.D. Lumsden, K. Eichele, R.E. Wasylishen, T.S. Cameron,
and J.F. Britten. J. Am. Chem. Soc. 116, 11129 (1994).
63. M.D. Lumsden, R.E. Wasylishen, and J.F. Britten. J. Phys.
Chem. 99, 16602 (1995).
56. A. Fioretta, L. Lorenza, and G. Pellizer. Phys. Chem. Chem.
Phys. 1, 4981 (1999).
57. C. Sizun, P. Kempgens, J. Raya, K. Elbayed, P. Granger, and
J. Rose. J. Organomet. Chem. 604, 27 (2000).
58. J.A. Gonzalez, G.A. Aucar, M.C. Ruiz, and R.H. Contreras.
Int. J. Quantum Chem. 66, 823 (1997).
59. C. Juan and H.S. Gutowsky. J. Chem. Phys. 37, 2198 (1962).
60. C.J. Jameson. In Multinuclear NMR spectroscopy. Edited by
J. Mason. Plenum, New York. 1987. Chap. 4.
64. G. Wu and R.E. Wasylishen. J. Phys. Chem. 97, 7863 (1993).
65. S. Koeker, J.V. Hanna, R.E. Wasylishen, E.W. Ainscough, and
A.M. Brodi. J. Magn. Reson. 135, 208 (1994).
66. J. Kaski, P. Lantto, J. Vaara, and J. Jokisaari. J. Am. Chem.
Soc. 120, 3993 (1998).
67. J. Kaski, J. Vaara, and J. Jokisaari. J. Am. Chem. Soc. 118,
8879 (1996).
© 2002 NRC Canada