Thioacetate complexes of group 12 metals. Structures of [Ph4P][Zn(SC{O}Me)3(H2O)] and [Ph4P][Cd(SC{O}Me)3]

Article abstract of DOI:10.1039/a906494j

The compounds [Ph₄P][Zn(SC{O}Me)₃(H₂O)] 1 and [Ph₄P][M(SC{O}Me)₃] (M = Cd 2 or Hg 3) were synthesized by treating the appropriate metal salt with Et₃NH⁺MeC{O}S^- and Ph₄PCl in the ratio 1:3:1 in water. The structures of 1 and 2 were determined by single crystal X-ray diffraction methods. In 1 the Zn is bonded to sulfur atoms of three thioacetate ligands and an oxygen atom of the H₂O molecule to have an idealised tetrahedral geometry. The hydrogen atoms of the co-ordinated H₂O molecule are involved in intramolecular O-H ... O hydrogen bonding with two carbonyl oxygen atoms. Approximate mirror symmetry is present in the anion. In the [Cd(SC{O}Me)₃]^- anion in 2 a propeller-like arrangement of the three co-ordinated MeC{O}S^- is found. The three sulfur atoms are strongly bound to Cd giving a pyramidal CdS₃ kernel. The anion is best described as having fac octahedral geometry with an approximate C₃ symmetry. Metal NMR data have been measured for 2 and 3. The reduced temperature cadmium-113 NMR spectrum of a [Cd(SC{O}Me)₃]^--[Cd(SC{O}Ph)₃]^- mixture provides evidence for [Cd(SC{O}Me)_x-(SC{O}Ph)_3-x]^- (x = 3-0), confirming the integrity of the two parent complexes in solution. Cadmium-113 NMR spectra of 2:L mixtures (L = H₂O, DMF or DMSO) show that [Cd(SC{O}Me)₃]^- is a weak acceptor. Mercury-199 NMR spectra of analogous 3:L mixtures show that [Hg(SC{O}Me)₃]^- has, at best, a weaker acceptor ability towards H₂O, DMF and DMSO than its cadmium analogue. The syntheses and metal NMR data show that the acceptor ability of [M(SC{O}Me)₃]^- (M = Zn to Hg) towards H₂O varies with M in the order Zn^II > Cd^II > HgII. The Royal Society of Chemistry 1999.

Full text of DOI:10.1039/a906494j

View Article Online / Journal Homepage / Table of Contents for this issue
Thioacetate complexes of Group 12 metals. Structures of
[Ph4P][Zn(SC{O}Me)3(H2O)] and [Ph4P][Cd(SC{O}Me)3]†
Jeyagowry T. Sampanthar,a Theivanayagam C. Deivaraj,a Jagadese J. Vittal*a and
Philip A. W. Dean*b
a Department of Chemistry, National University of Singapore, Lower Kent Ridge Road,
Singapore 119 260. E-mail: chmjjv@nus.edu.sg
b Department of Chemistry, University of Western Ontario, London, N6A 5B7, Canada.
E-mail: pawdean@julian.uwo.ca
Received 10th August 1999, Accepted 25th October 1999
The compounds [Ph4P][Zn(SC{O}Me)3(H2O)] 1 and [Ph4P][M(SC{O}Me)3] (M = Cd 2 or Hg 3) were synthesized by
treating the appropriate metal salt with Et3NHϩMeC{O}SϪ and Ph4PCl in the ratio 1:3:1 in water. The structures of
1 and 2 were determined by single crystal X-ray diffraction methods. In 1 the Zn is bonded to sulfur atoms of three
thioacetate ligands and an oxygen atom of the H2O molecule to have an idealised tetrahedral geometry. The hydrogen
atoms of the co-ordinated H2O molecule are involved in intramolecular O–H ؒ ؒ ؒ O hydrogen bonding with two
carbonyl oxygen atoms. Approximate mirror symmetry is present in the anion. In the [Cd(SC{O}Me)3]Ϫ anion in
2 a propeller-like arrangement of the three co-ordinated MeC{O}SϪ is found. The three sulfur atoms are strongly
bound to Cd giving a pyramidal CdS3 kernel. The anion is best described as having fac octahedral geometry with
an approximate C3 symmetry. Metal NMR data have been measured for 2 and 3. The reduced temperature cadmium-
113 NMR spectrum of a [Cd(SC{O}Me)3]Ϫ–[Cd(SC{O}Ph)3]Ϫ mixture provides evidence for [Cd(SC{O}Me)x-
(SC{O}Ph)3Ϫx]Ϫ (x = 3–0), confirming the integrity of the two parent complexes in solution. Cadmium-113 NMR
spectra of 2:L mixtures (L = H2O, DMF or DMSO) show that [Cd(SC{O}Me)3]Ϫ is a weak acceptor. Mercury-199
NMR spectra of analogous 3:L mixtures show that [Hg(SC{O}Me)3]Ϫ has, at best, a weaker acceptor ability towards
H2O, DMF and DMSO than its cadmium analogue. The syntheses and metal NMR data show that the acceptor
ability of [M(SC{O}Me)3]Ϫ (M = Zn to Hg) towards H2O varies with M in the order ZnII > CdII > HgII.
raise several interesting questions. Is [Cd(SC{O}Ph)3]Ϫ unique
Introduction
in exhibiting polytopal isomerism? What are the structures (and
For the past several years we have been interested in the
stabilities) of [M(SC{O}R)3]Ϫ in solution? Does the commonly
chemistry of thiobenzoates.1–10 The presence of both hard and
found MS3 planar geometry occur only in complexes with the
soft donor sites makes thiocarboxylates, RC{O}SϪ in general,
Ph{O}CSϪ ligand or in complexes of the Group 12 metals with
interesting ligands. Recently, we have reported1 homoleptic
other thiocarboxylate ligands R{O}CSϪ also? As a contribu-
anionic complexes [M(SC{O}Ph)3]Ϫ (M = Mn, Co or Ni) in
tion to answering some of these questions, we have carried out a
which both oxygen and sulfur atoms of the thiobenzoate
study of the thioacetate complexes of the zinc-group elements.
ligands are bonded to the central metal atom in a bidentate
Here we report the synthesis of anionic thioacetato-complexes
fashion. In contrast, the cadmium atoms in [Na{Cd(SC-
of Zn, Cd and Hg and the structures of those of Zn and Cd.
{O}Ph)3}2]Ϫ have trigonal planar geometry with respect to the
The 113Cd and 199Hg NMR chemical shifts have been measured
sulfur atoms.2 In this trinuclear anion the planarity of the CdS3
for the complexes of Cd and Hg in solution, for comparison
kernel was attributed to the presence of the Naϩ, which binds
with those of the thiobenzoato-complexes.3 In addition,
to the oxygen atoms of the carbonyl groups, precluding (or
at least reducing) their bonding to cadmium. However, a
metal NMR data have been obtained for [M(SC{O}Me)3]Ϫ–
[M(SC{O}Ph)3]Ϫ mixtures and for mixtures of [M(SC{O}-
similar trigonal planar MS3 geometry, supported by one or
Me)3]Ϫ and the potential ligands H2O, DMF and DMSO.
more intramolecular M ؒ ؒ ؒ O interactions, was observed3–5 in
Thioacetate complexes of the Group 12 metals are potential
[M(SC{O}Ph)3]Ϫ (M = Zn, Cd or Hg) even when the alkali
candidates as single-source precursors to make metal sulfide
metal cation was absent. For M = Zn to Hg, trigonal planar
materials. Hampden-Smith and co-workers have successfully
MS3 kernels are generally found only within complexes with
used the adducts of M(SC{O}Me)2 (M = Zn or Cd) with
hindered thiolate ligands.11–16 Rhombohedral [Ph4P][Cd(SC-
tmen17 and lutidine (2,6-dimethylpyridine) (Lut)18 to deposit
{O}Ph)3] has a unique structure containing both planar and
the corresponding metal sulfides.
pyramidal CdS3 kernels.9 Unlike those of [M(SC{O}Ph)3]Ϫ
(M = Mn, Co or Ni), the pyramidal MS3 kernels in the
rhombohedral salt exhibit only weak M ؒ ؒ ؒ O interactions.
Experimental
General procedure
The structural studies of [M(SC{O}Ph)3]Ϫ (M = Zn to Hg)
All materials were obtained commercially and used as
received. The compounds monoclinic-[Ph4P][Cd(SC{O}Ph)3]
and [Ph4P][Hg(SC{O}Ph)3] were prepared as described in the
literature.3 All reactions were performed under an atmosphere
of N2(g). All non-aqueous solvents were dried over 3 Å molec-
ular sieves. They and water were degassed with N2(g) or Ar(g)
† Supplementary data available: rotatable 3-D crystal structure diagram
in CHIME format. See http://www.rsc.org/suppdata/dt/1999/4419/
Also available: 113Cd NMR titration data. For direct electronic access
see http://www.rsc.org/suppdata/dt/1999/4419/, otherwise available
from BLDSC (No. SUP 57673, 3 pp.) or the RSC Library. See Instruc-
tions for Authors, 1999, Issue 1 (http://4419.rsc.org/dalton).
J. Chem. Soc., Dalton Trans., 1999, 4419–4423
This journal is © The Royal Society of Chemistry 1999
4419

View Article Online
before the reactions or preparation of NMR samples. The
yields are reported with respect to the metal salts. The prepared
complexes were stored under a N2 atmosphere and at 5 ЊC as
they deteriorate on prolonged standing in the air at room
temperature. The Microanalytical Laboratory at National
University of Singapore performed microanalyses.
interchange using 0.1 M Cd(ClO4)2 (aq) at 294 1 K and neat
HgMe2 at 296 1 K for 113Cd and 199Hg, respectively. Samples
were prepared in 10 mm od NMR tubes under an atmosphere
of argon, with solvents which had been degassed with Ar(g),
and used directly. (In initial experiments, samples were prepared
using ice-cold solvent. However, it was found that, except for
samples containing 3 with added water, the metal NMR spectra
at ambient probe temperature do not change over a period of
hours, indicating that the precaution of cooling the solvent is
unwarranted.) Concentrations are expressed as mass of solute
per volume of solvent. For the chemical shift titrations,
additions of DMSO, DMF or water to solutions in CH2Cl2
were made under a flow of Ar(g) using a 10 µL syringe.
Preparations
[Ph4P][Zn(SC{O}Me)3(H2O)] 1. Thioacetic acid (2.0 mL, 28
mmol) was added dropwise to a stirred aqueous solution of
Et3N (3.9 mL, 28 mmol) to obtain a solution of Et3NHϩ
Me{O}CSϪ. To this was added ZnCl2ؒ1.5H2O (1.226 g, 9 mmol)
in water (5 mL). The resulting solution was stirred for 10 min.
When a solution of Ph4PCl (3.369 g, 9 mmol) in water was
added a cream-coloured precipitate was formed. The mixture
was stirred for 1 h, after which CH2Cl2 (8 mL) was added. The
precipitate in the aqueous layer was extracted into the CH2Cl2
layer, which became yellow. The CH2Cl2 layer was separated
and washed with 30 mL portions of deionised water 3 or 4
times to remove the side product, Et3NHCl. Then Et2O was
added to the CH2Cl2 solution until turbidity was noticed, when
the mixture was set aside in a refrigerator overnight, to obtain
yellow crystals. The crystals were decanted, washed with Et2O
and dried in vacuum. Yield: 3.66 g (62.7%). Suitable single
crystals were selected and used for X-ray diffraction experi-
ments. Calc. for C30H31O4PS3Zn: C, 55.60; H, 4.82; P, 4.78; S,
14.84; Zn, 10.09. Found: C, 55.62; H, 4.75; P, 4.68; S, 15.04;
The stability constants for eqn. (4) were found by successive
approximation based on the fit of the fractional populations
calculated from them to eqn. (1)19 (∆δCd (= δCd([Cd(SC-
∆δCd = P(complex)ؒ∆δCd(complex)
(1)
{O}Me)3]Ϫ ϩ L) Ϫ δCd([Cd(SC{O}Me)3]Ϫ); complex = [Cd(SC-
{O}Me)3(L)]Ϫ). This method puts all of the error into the value
of ∆δCd(complex). The approach was validated by the values of
r2 and the reasonableness of the values found for ∆δCd(com-
plex). Regression analyses of the metal chemical shift titration
data vs. P(complex) were carried out using QuattroPro 8.0
(Corel Corporation, Ottawa, Canada, 1997). For the purposes
of these calculations, concentrations in mol per L of solvent
were taken to approximate molarities. Also, because such small
amounts of the prospective ligands L were used, the approxi-
mation was made that these additions caused no general solvent
effects on the chemical shifts.
1
Zn, 10.15%. H NMR: δ 2.34 (s, CH3{O}CS, 9 H), 4.75 (H2O)
and 7.6–7.8 (m, C6H5, 20 H). 13C NMR: δ 36.8 (CH3{O}CS),
1
3
117.9 (Ph C1, J(P–C) = 89.7), 131.3 (Ph C3,5, J(P–C) = 24.2),
134.8 (Ph C2,6, J(P–C) = 9.8), 136.3 (Ph C4, J(P–C) = 2.3 Hz)
2
4
and 212.2 (CH3{O}CS).
Thermogravimetric analysis
[Ph4P][Cd(SC{O}Me)3] 2. The preparation was as for com-
plex 1, except that the cadmium was added as a solution of
CdCl2ؒH2O (1.812 g, 9 mmol) in water (5 mL). Yield of yellow
crystals: 3.2 g (53%). The crystals were suitable for single crystal
X-ray studies. Calc. for C30H29CdO3PS3: C, 53.21; H, 4.32; Cd,
16.60; P, 4.57; S,14.21. Found: C, 53.07; H, 3.93; Cd, 15.46;
Thermogravimetric analysis of complex 1 was carried out using
a SDT 2980 TGA Thermal Analyser with a heating rate
of 20 ЊC minϪ1 in a N2 atmosphere using a sample size about
5–10 mg per run.
1
X-Ray crystallography
P, 4.11; S, 14.44%. H NMR: δ 2.29 (s, CH3{O}CS, 9 H) and
7.6–7.8 (m, C6H5, 20H). 13C NMR: δ 35.4 (CH3{O}CS), 118.0
The diffraction experiments were carried out on a Bruker
SMART CCD diffractometer. The softwares used were:
SMART,20 for collecting frames of data, indexing reflections
and determination of lattice parameters; SAINT,20 for inte-
gration of intensity of reflections and scaling; SADABS,21
for absorption correction; and SHELXTL,22 for space group
and structure determination, least-squares refinements on F2,
graphics and structure reporting. The hydrogen atoms attached
to the oxygen of the H2O molecule in complex 1 were success-
fully located. The positional and common isotropic thermal
parameters were refined in the least-squares cycles. The
absolute structure (Flack) parameter was refined to Ϫ0.03(2)
for 2. A brief summary of the crystallographic data is given
in Table 1.
1
3
(Ph C1, J(P–C) = 89.6), 131.3 (Ph C3,5, J(P–C) = 12.9), 135.0
(Ph C2,6, J(P–C) = 10.4), 136.4 (Ph C4, J(P–C) = 2.0 Hz) and
2
4
213.2 (CH3{O}CS).
[Ph4P][Hg(SC{O}Me)3] 3. This complex was synthesized in
the same way as 1, but on half the scale and using HgCl2
(1.222 g, 4.5 mmol) in 11 mL of MeOH–water (1:10 v/v) as the
source of mercury. Yield of yellow crystals: 4.56 g (66.3%).
Calc. for C30H29HgO3PS3: C, 47.08; H, 3.82; Hg, 26.21; P, 4.05;
S, 12.57. Found: C, 46.78; H, 3.55; Hg, 25.51; P, 3.62; S, 12.47.
1H NMR: δ 2.30 (s, CH3{O}CS, 9 H), 7.6–7.8 (m, C6H5, 20H).
1
13C NMR: δ 36.1 (CH3{O}CS), 117.9 (Ph C1, J(P–C) = 89.1),
3
2
131.3 (Ph C3,5, J(P–C) = 12.8), 134.9 (Ph C2,6, J(P–C) = 9.8),
136.3 (Ph C4, 4J(P–C) = 3 Hz) and 205.9 (CH3{O}CS).
CCDC reference number 186/1705.
See http://www.rsc.org/suppdata/dt/1999/4419/ for crystallo-
graphic files in .cif format.
NMR Spectra
Proton and 13C-{1H} NMR spectra were recorded at 298 K on a
Bruker ACF 300 MHz spectrometer. Samples were made up in
5 mm od NMR tubes with CD2Cl2 or CDCl3 as solvent and
TMS as internal reference. Metal NMR spectra were obtained
using a Varian XL-300 spectrometer system, operating at 66.53
and 53.72 MHz for 113Cd and 199Hg, respectively. Proton
decoupling was continuous for 199Hg but inverse-gated for
113Cd, which has a negative Overhauser effect. Spectra were
obtained with the field pre-shimmed, but without a field-
frequency lock (field drift is negligible). Temperatures were
measured using a calibrated thermocouple in a stationary
dummy sample. External referencing was carried out by sample
Results and discussion
All three complexes were prepared by a general route from
the metal salts and deprotonated Me{O}CSϪ in the ratio of
1:3, eqn. (2), and isolated as the Ph4Pϩ salts. Only the zinc()
compound picked up a molecule of H2O, eqn. (3).
H2O
(2)
M2ϩ ϩ 3 Me{O}CSϪ
[M(SC{O}Me)3]Ϫ
[Zn(SC{O}Me)3]Ϫ ϩ H2O → [Zn(SC{O}Me)3(H2O)]Ϫ (3)
4420
J. Chem. Soc., Dalton Trans., 1999, 4419–4423

View Article Online
Table 1 Crystallographic data for [Ph4P][Zn(SC{O}Me)3(H2O)] 1 and
[Ph4P][Cd(SC{O}Me)3] 2
Table 3 Some formation constants, K, and 113Cd NMR complexation
a
shifts, ∆δCd([Cd(SC{O}Me)3(L)]Ϫ) for L:2 mixtures in CH2Cl2
1
2
L
T/K
K/mol LϪ1 ∆δCd([Cd(SC{O}Me)3(L)]Ϫ)
Chemical formula
Formula weight
T/K
C30H31O4PS3Zn
648.07
296
C30H29CdO3PS3
677.08
296
H2O
DMSO
DMF
294b
178d
178f
1.0
5.0
1.0
173 2c
119 3e
94 5g
Radiation, wavelength/Å
Crystal system
Space group
a/Å
b/Å
c/Å
Mo-Kα, 0.71073
Monoclinic
P21/n
11.6160(3)
19.2122(3)
14.3978(3)
93.281(1)
3207.9(1)
4
Mo-Kα, 0.71073
Monoclinic
P21
10.7195(1)
13.0561(2)
11.2199(2)
100.701(1)
1542.97(4)
2
a See text. Cadmium concentration is 0.1 mol per L of solvent. b Water
is insoluble in the solution when H2O:2 = ca. 2:1; it is insufficiently
soluble at 178 K to collect sufficient data for evaluation of K and
∆δCd([Cd(SC{O}Me)3(L)]Ϫ). c r2 = 0.9997. d Values of ∆δCd are too small
at 294 K for evaluation of K and ∆δCd([Cd(SC{O}Me)3(DMSO)]Ϫ);
e.g. 7 ppm when DMSO:2 = 5.0:1. e r2 = 0.998. f Values of ∆δCd are too
small at 294 K for evaluation of K and ∆δCd([Cd(SC{O}Me)3(DMF)]Ϫ);
e.g. 5 ppm when DMSO:2 = 5.0:1. g r2 = 0.995.
β/Њ
V/Å3
Z
µ/mmϪ1
1.042
0.991
Reflections measured
Independent reflections
Final R indices [I > 2σ(I)]
R1
14196
6666 (Rint = 0.0292)
9847
6857 (Rint = 0.0128)
determined whether this is planar or pyramidal. The relative
linewidths indicate that the limiting no-exchange spectrum has
not been reached at 178 K. A similar mixing experiment for
[Hg(SC{O}Me)3]Ϫ and [Hg(SC{O}Ph)3]Ϫ was not definitive. A
0.0556
0.1015
0.0271
0.0656
wR2
₁
single resonance was observed at 296 K, δHg(∆ν ) = Ϫ721 (≈130
₂

Table
2
Metal NMR data for thioacetato- and thiobenzoato-
complexes of cadmium and mercurya
Hz), but at 178 and at ≈168 K (for a partially frozen solution)
only two broad signals were observed with δHg ≈ Ϫ646 and
Ϫ629, with the former being the more intense. It seems that
even at the freezing point the slow exchange region is barely
attained. The rate of exchange of S-donor ligands bound to
Hg is typically faster than the rate of ligand exchange in the
corresponding cadmium complexes.24,25
₁
₁
δCd(∆δ (approx.)/
δHg(∆δ (approx.)/
₂

₂

Complex
T/K
Hz)b
Hz)c
[Cd(SC{O}Me)3]Ϫ d
[Cd(SC{O}Ph)3]Ϫ e
[Hg(SC{O}Me)3]Ϫ d
[Hg(SC{O}Ph)3]Ϫ e
295
178
295f
178
294
178
296f
178
309(45)
303(90)
287(7)
Acceptor behaviour of [Zn(SC{O}Me)3]Ϫ is shown by the
isolation of complex 1. The complexation shifts, ∆δCd (= δcd([Cd-
(SC{O}Me)3]Ϫ ϩ L) Ϫ δCd([Cd(SC{O}Me)3]Ϫ)), that are found
when the ligands H2O, DMSO or DMF are added incre-
mentally to solutions of 2 in CH2Cl2 (SUP 57673), indicate that
[Cd(SC{O}Me)3]Ϫ behaves as a weak acceptor towards these
ligands. The data can be analysed (Experimental section)
in terms of eqn. (4) to give the formation constants, and
280(15)
Ϫ689(170)
Ϫ636(210)
Ϫ757(190)
Ϫ624(≈500)
a This work unless otherwise specified. b Relative to external 0.1 M
Cd(ClO4)2(aq.). c Relative to external HgMe2. d Ph4Pϩ salt. e Ph4Asϩ salt.
f Ref. 3.
[Cd(SC{O}Me)3]Ϫ ϩ L
[Cd(SC{O}Me)3(L)]Ϫ (4)
The 1H and 13C NMR spectra in solution of these com-
pounds are consistent with the proposed formula. The presence
of the water molecule in 1 was indicated in the initial weight
loss observed in TG in the temperature range 70–100 ЊC
(observed 3.0%, calculated 2.8%).
the complexation 113Cd NMR chemical shifts, ∆δCd ([Cd(SC-
{O}Me)3L]Ϫ), of [Cd(SC{O}Me)3L]Ϫ. Table 2 presents a
summary of the results. We regard the values in Table 2 to be
approximate, because of the approximations that were made in
their derivations (Experimental section). However, the fits to
the formation of a 1:1 complex from an unassociated [Cd(SC-
{O}Me)3]Ϫ are good enough to rule out appreciable formation
of any 2:1 (L:2) complex or appreciable aggregation of the
precursor [Cd(SC{O}Me)3]Ϫ. The chemical shift and com-
plexation shift, ∆δCd, of 2 in pure DMSO at 294 K are ≈387
and ≈78. Those of 2 in pure DMF are 337 (28) at 294 K and
≈376 (≈71) at 213 K, with an extrapolated value for ∆δCd
of ≈90 at 178 K. The remarkable agreement between the
last value and that calculated for [Cd(SC{O}Me)3(DMF)]Ϫ
in CH2Cl2 from complexation shifts (Table 3) is probably
fortuitous.
Metal NMR studies
The metal chemical shifts are given in Table 2, with data for
the corresponding thiobenzoato-complexes for comparison. At
ambient probe temperature the metal resonances of the thio-
acetato-complexes are deshielded relative to the thiobenzoato
complexes for both 113Cd and 199Hg. The temperature sensi-
tivities of the chemical shifts, dδ/dT, are small and positive for
both cadmium complexes, but larger in magnitude and negative
for both mercury complexes. For [Hg(SC{O}Ph)3]Ϫ both the
sign23 and the magnitude of dδ/dT, Ϫ1.1 ppm KϪ1, are note-
worthy. Possibly conformational changes occur in this anion as
the temperature is changed.
Mercury-199 NMR shows that [Hg(SC{O}Me)3]Ϫ (as 3) is
an insignificant acceptor of DMSO and DMF. Even at 178 K
values of ∆δHg are so small that they could be due to changes in
the composition of the solvent when the potential donor is
added. For example, with L:Hg = 10.0:1, ∆δHg = 3 for both L =
DMSO and DMF. Clearly negligible complexation is occurring.
It is reasonable to expect that ∆δHg([Hg(SC{O}Me)3(L)]Ϫ) will
be larger than ∆δCd([Cd(SC{O}Me)3(L)]Ϫ).26 Addition of water
to a solution of complex 3, 0.1 mol per L of CH2Cl2, at 294 K
produces shielding. The value of ∆δHg ≈ Ϫ11 when [H2O]/[3] =
2.0. Unfortunately, δHg for 3–water mixtures changes slowly
with time, so that values of K and ∆δHg(complex) could not be
obtained. Even so it seems evident that [Hg(SC{O}Me)3-
(OH2)]Ϫ is less stable than its cadmium counterpart.
Theintegrityofboth[Cd(SC{O}Me)3]Ϫ and[Cd(SC{O}Ph)3]Ϫ
in solution is shown by a mixing experiment. At 294 K a 1:1
solution of these two ions (each 0.1 mol per L of solvent) shows
₁
a single broad resonance, δCd = 298, ∆ν ≈ 180 Hz. At reduced
₂

temperature fine structure becomes evident, and at 178 K four
resonances are observed, with approximately binomial inten-
₁
sities and δ(approx ∆ν ) of ≈304 (300), 297.6 (150), 289.2 (90)
₂

and 279.5 (50 Hz). These four signals can be assigned to
[Cd(SC{O}Me)x(SC{O}Ph)3Ϫx]Ϫ, with x = 3, 2, 1, or 0, respec-
tively. Evidently the complexes remain triligated in solution.
Furthermore, the similarity of the chemical shifts for [Cd-
(SC{O}R)3]Ϫ (R = Me or Ph or mixtures of these) points to the
same MS3 skeletal geometry in solution, although it cannot be
J. Chem. Soc., Dalton Trans., 1999, 4419–4423
4421

View Article Online
Table 4 Selected bond distances (Å) and angles (Њ) for the anion in
complex 1
Zn(1)–S(1)
Zn(1)–S(2)
Zn(1)–S(3)
Zn(1)–O(4)
C(1)–S(1)
C(1)–O(1)
C(3)–S(2)
2.310(1)
2.345(1)
2.333(1)
2.084(3)
1.732(5)
1.224(4)
1.699(4)
C(3)–O(2)
C(5)–S(3)
C(5)–O(3)
C(1)–C(2)
C(3)–C(4)
C(5)–C(6)
1.233(4)
1.733(5)
1.226(5)
1.506(6)
1.504(6)
1.508(6)
S(1)–Zn(1)–S(2)
S(2)–Zn(1)–S(3)
S(3)–Zn(1)–S(1)
O(4)–Zn(1)–S(1)
O(4)–Zn(1)–S(2)
O(4)–Zn(1)–S(3)
Zn(1)–S(1)–C(1)
Zn(1)–S(2)–C(3)
113.01(4)
116.27(4)
115.28(4)
105.69(9)
100.32(9)
103.92(9)
89.5(1)
Zn(1)–S(3)–C(5)
S(1)–C(1)–O(1)
S(2)–C(3)–O(2)
S(3)–C(5)–O(3)
S(1)–C(1)–C(2)
S(2)–C(3)–C(4)
S(3)–C(5)–C(6)
105.9(2)
122.7(3)
126.6(3)
125.2(4)
117.4(4)
114.6(3)
115.6(4)
109.0(2)
Fig. 1 A perspective view of the [Zn(SC{O}Me)3(H2O)]Ϫ anion.
Hydrogen bonding distances (Å) and angles (Њ): O2 ؒ ؒ ؒ O4 2.663(5);
O3 ؒ ؒ ؒ O4 2.614(5); O4–H1 1.29(4) and O4–H2 0.84(4); O4–H1 ؒ ؒ ؒ O2
150(3) and O4–H2 ؒ ؒ ؒ O3 166(4).
Table 5 Selected bond distances (Å) and angles (Њ) for the anion in
complex 2
Cd(1)–S(1)
Cd(1)–S(2)
Cd(1)–S(3)
Cd(1)–O(1)
Cd(1)–O(2)
Cd(1)–O(3)
C(1)–S(1)
2.5334(9)
2.5510(8)
2.5687(8)
2.689(3)
2.555(3)
2.562(3)
1.694(4)
1.713(4)
C(5)–S(3)
C(1)–O(1)
C(3)–O(2)
C(5)–O(3)
C(1)–C(2)
C(3)–C(4)
C(5)–C(6)
1.721(3)
1.217(5)
1.228(4)
1.263(4)
1.502(6)
1.514(5)
1.505(5)
C(3)–S(2)
S(1)–Cd(1)–S(2)
S(2)–Cd(1)–S(3)
S(3)–Cd(1)–S(1)
O(1)–Cd(1)–O(2)
O(2)–Cd(1)–O(3)
O(3)–Cd(1)–O(1)
O(1)–Cd(1)–S(1)
O(2)–Cd(1)–S(2)
O(3)–Cd(1)–S(3)
115.40(4)
112.52(3)
115.36(3)
97.9(1)
77.10(8)
79.7(1)
58.35(8)
60.66(6)
61.24(5)
S(1)–C(1)–O(1)
S(2)–C(3)–O(2)
S(3)–C(5)–O(3)
Cd(1)–S(1)–C(1)
Cd(1)–O(1)–C(1)
Cd(1)–S(2)–C(3)
Cd(1)–O(2)–C(3)
Cd(1)–S(3)–C(5)
Cd(1)–O(3)–C(5)
121.4(3)
121.6(3)
121.5(3)
87.7(1)
92.0(2)
83.6(1)
94.0(2)
83.6(1)
93.7(2)
Fig. 2 An ORTEP28 diagram of the anion [Cd(SC{O}Me)3]Ϫ with
50% probability thermal ellipsoids and the numbering scheme. The
hydrogen atoms are omitted for clarity.
(tmen)]17 and [Zn(SC{O}Me)2(Lut)2]18 by Hampden-Smith
and co-workers. However, the bond distances observed in 1 fall
into the range of 2.304–2.376 Å found3 for [Zn(SC{O}Ph)3]Ϫ.
The Zn–O distance, 2.084(3) Å, is longer than the correspond-
ing distance, 2.031(6) Å, found in [Zn(SC{O}Ph)2(OH2)2].27
The S–Zn–S angles, 113.01(4)–116.27(4)Њ, are higher than the
ideal tetrahedral angle while the S–Zn–O angles, 100.32(9)–
The evidence from the syntheses, together with the metal
NMR measurements, shows that acceptor ability for
[M(SC{O}Me)3]Ϫ (M = Zn, Cd or Hg), with the oxygen donors
studied here, is in the order Zn > Cd > Hg. Evidently the order
of “hardness” of the metal centres, ZnII > CdII > HgII is
unaffected by co-ordination of the “soft” thiobenzoate sulfur
atoms.
105.69(9)Њ, are lower than the ideal value, 109Њ28Ј. The C᎐O
᎐
distances are equal, and unaffected by involvement in hydrogen
bonding, or lack of it. There is a weak interaction between
Zn(1) and the carbonyl oxygen atom, O(1), which is not
involved in hydrogen bonding. The distance Zn(1) ؒ ؒ ؒ O(1), at
2.698(3) Å, is less than the sum of the van der Waals radii,
2.90 Å.29 This existence of the weak interaction is supported by
the occurrence of a relatively small C–S–Zn angle, C(1)–S(1)–
Zn(1) 89.5(1)Њ, for the ligand concerned (the corresponding
angles in the other two ligands are 109.0(2) and 105.9(2)Њ).
If this Zn ؒ ؒ ؒ O interaction is taken into account the co-
ordination geometry of ZnII is high distorted trigonal bipyr-
amidal. The interplanar angle between the Zn(1)S(2)C(3)-
C(4)O(2)O(4) plane (plane 2 with deviation, 0.094 Å) and the
Zn(1)S(3)C(5)C(6)O(3)O(4) plane (plane 3 with deviation,
0.041 Å) at the Zn(1)–O(4) fold is 60.2(1)Њ. The plane contain-
ing atoms Zn(1)S(1)O(1)C(1)C(2)O(4) (plane 1, deviation,
0.053 Å) bisects planes 2 and 3 at the angles 61.2(1) and 60.2(1)Њ
respectively. In other words, there is an approximate non-
crystallographic mirror symmetry present in the anion along
plane 1.
Structures of the complexes 1 and 2
The crystal structures of compounds 1 and 2 reveal that
they consist of discrete mononuclear anions and cations.
The structures of the anions in 1 and 2 are given in Figs. 1 and
2. Selected bond angles and lengths are given in Tables 4
and 5. The cations are unexceptional and warrant no further
discussion.
In the [Zn(SC{O}Me)3(H2O)] Ϫ anion of 1 the ZnII is bonded
to three thioacetate ligands through sulfur atoms, and to the
oxygen atom of an H2O molecule giving a distorted tetrahedral
geometry at the metal center. The hydrogen atom of the water
molecule is involved in intramolecular O–H ؒ ؒ ؒ O hydrogen
bonding to the oxygen atoms, O(2) and O(3), of two carbonyl
groups. Each hydrogen bond completes a stable six-membered
ring.
The three Zn–S bond distances in complex 1, 2.310(1),
2.333(1) and 2.345(1) Å, are not equal. The S(1) atom of the
thioacetate group which is not involved in hydrogen bonding is
involved in a shorter bond to ZnII than the other two. The Zn–S
distances observed are somewhat higher than the range, 2.294–
2.312 Å, observed in the neutral adducts [Zn(SC{O}Me)2-
In the anion in complex 2 the CdS3 kernel is not planar or
near-planar, as it is in the majority of [M(SC{O}R)3]Ϫ anions
of the Group 12 elements (Table 6). The central CdII is bonded
strongly to the sulfur atoms of the three CH3C{O}SϪ ligands
4422
J. Chem. Soc., Dalton Trans., 1999, 4419–4423

View Article Online
change of R from Ph to Me, since [Hg(SC{O}Ph)3]Ϫ is known
Table 6 Deviations of M from the S3 plane for various compounds
with a (near-)planar HgS3 skeleton only.3,4 Unfortunately,
we have not yet been able to obtain crystals of the relatively
unstable compound 3 that are suitable for single crystal X-ray
analysis. In this case too, the metal NMR results point to a
common geometry for the skeletons of the [Hg(SC{O}R)3]Ϫ
anions in solution.
Deviation
in Å
Compound
ΣS–M–S/Њ Ref.
[Ph4P][Zn(SC{O}Ph)3]
[Ph4As][Cd(SC{O}Ph)3]
[Ph4P][Hg(SC{O}Ph)3]
[Me4N][Na(Cd{SC{O}Ph}3)2]
[Ph4P][Cd(SC{O}Ph)3], m form
r form
Ϫ0.06
Ϫ0.01
Ϫ0.05
0.33
359.8
360.0
359.9
361.0
354.2
359.5a
333.3b
341.3c
359.5
358.5
355.8
344.6
343.3
3
3
3
2
9
9
9
9
0.36
Acknowledgements
0.11a
0.77b
0.65c
0.10
J. J. V. would like to thank the National University of Singapore
for the research grant (Grant No. RP970618), and P. A. W. D.
thanks the Natural Sciences and Engineering Research Council
of Canada for financial support.
[Me4N][Hg(SC{O}Ph)3]
[Et3NH][Cd(SC{O}Ph)3]d
4
5
5
0.18
0.30
0.54
0.61
[Ph4P][Zn(SC{O}Me)3(H2O)]
[Ph4P][Cd(SC{O}Me)3]
This work
This work
References
a Data for the Type B anion. b Data for the Type A anion. c Data for Type
1 R. Devy, J. J. Vittal and P. A. W. Dean, Inorg. Chem., 1998, 37, 6939.
2 J. J. Vittal and P. A. W. Dean, Inorg. Chem., 1993, 32, 791.
3 J. J. Vittal and P. A. W. Dean, Inorg. Chem., 1996, 35, 3089.
4 J. J. Vittal and P. A. W. Dean, Acta Crystallogr., Sect. C, 1997, 53,
409.
5 J. J. Vittal and P. A. W. Dean, Acta Crystallogr., Sect. C, 1998, 54,
319.
6 T. R. Burnett, P. A. W. Dean and J. J. Vittal, Inorg. Chem., 1994, 72,
1127.
7 J. J. Vittal and P. A. W. Dean, Acta Crystallogr., Sect. C, 1997, 52,
1180.
8 J. J. Vittal and P. A. W. Dean, Polyhedron, 1998, 17, 1937.
9 P. A. W. Dean, J. J. Vittal, D. C. Craig and M. L. Scudder, Inorg.
Chem., 1998, 37, 1661.
10 J. T. Sampanthar, J. J. Vittal and P. A. W. Dean, J. Chem. Soc.,
Dalton Trans., 1999, 1993.
11 E. S. Gruff and S. A. Koch, J. Am. Chem. Soc., 1990, 112, 1245.
12 R. A. Santos, E. F. Gruff, S. A. Koch and G. S. Harbison, J. Am.
Chem. Soc., 1991, 113, 469.
13 M. Bochmann, K. Webb, M. Harman and M. B. Hursthouse,
Angew. Chem., Int. Ed. Engl., 1990, 29, 638.
14 E. S. Gruff and S. A. Koch, J. Am. Chem. Soc., 1989, 111, 8762.
15 M. Bochmann, G. Bwembya, R. Grinter, J. Lu, K. J. Webb,
D. J. Williamson, M. B. Hursthouse and M. Mazid, Inorg. Chem.,
1993, 21, 532.
C anion. d Data for two independent ‘ion pairs’.
and weakly to the oxygen atoms of the carbonyl groups in
such a way as to have, roughly, a fac octahedral geometry. The
disposition of the three carbonyl groups on the opposite side of
the Cd to the S3 plane results in the anion having a propeller-
like geometry with an approximate C3 point group symmetry,
as found also for thiobenzoate complexes of Co, Ni and Mn,1
and the type A and C anions of rhombohedral [Ph4P][Cd-
(SC{O}Ph)3].9 Overall, the geometry is most reminiscent of that
of the type C anions in rhombohedral [Ph4P][Cd(SC{O}Ph)3].9
It is interesting that the compound 2 has crystallised in a
polar space group (P21) as observed for the thiobenzoate com-
plexes having (some) anions with fac geometry.1,9 The Cd–S
distances in 2, 2.5334(9), 2.5510(8) and 2.5687(8) Å, are in the
order Cd(1)–S(1) > Cd(1)–S(2) > Cd(1)–S(3). They fall into
the range established for [Cd(SC{O}Ph)3]Ϫ,2,3,5,9 as do the
Cd–O distances, 2.689(3), 2.555(3) and 2.562(3) Å. However,
the Cd–S distances are longer than those observed in
[Cd(SC{O}Me)2(tmen)]17 and [Cd(SC{O}Me)2(Lut)2].18 The
Cd ؒ ؒ ؒ O distances of [Cd(SC{O}Me)3]Ϫ are less than the sum
of the van der Waals radii, 3.10 Å.29 The C–S and C–O dis-
tances are in the same order as the Cd–S distances. In the
present work the shortest C–S distance noted is 1.694 Å for
2 and 1.699 Å for 1. The changes in the bond parameters may
be attributed to the methyl group substitution.
(Near) trigonal planar is the geometry most commonly
observed for the MS3 skeletons of the tris(thiobenzoato) anions
of the Group 12 metals (Table 6). The central metal atom Cd(1)
in complex 2 does not have trigonal planar geometry with
respect to the sulfur atoms; it is 0.612 Å away from the S3 plane.
Also, the sum of the S–Cd–S angles, 343.28Њ, is far below the
expected angle of 360Њ (Table 6). Interestingly, the only other
two discrete thiocarboxylato-complexes of Zn to Hg which
have pyramidal MS3 kernels also have Cd as the central atom.9
Nonetheless, 2 is unique among salts of thiocarboxylato-
complexes of Zn or Cd in containing only anions with a pyr-
amidal MS3 skeleton, suggesting a definite role for the R group
of R{O}CSϪ. It is also interesting that the 113Cd NMR results
point to a uniformity of MS3 skeleton for [Cd(SC{O}R)3]Ϫ
(R = Me or Ph, or mixtures of these) in solution, in contrast to
the results for the solid state.
16 A. K. Duhme and H. Z. Strasdeit, Z. Anorg. Allg. Chem., 1999, 625,
6.
17 M. D. Nyman, M. J. Hampden-Smith and E. N. Duesler, Adv. Mater.
CVD, 1996, 2, 171.
18 M. D. Nyman, M. J. Hampden-Smith and E. N. Duesler, Inorg.
Chem., 1997, 36, 2218.
19 R. Foster and C. A. Fyfe, Prog. NMR Spectrosc., 1969, 4, 1.
20 SMART & SAINT Software Reference Manuals, Version 4.0,
Siemens Energy & Automation, Inc., Analytical Instrumentation,
Madison, WI, 1996.
21 G. M. Sheldrick, SADABS, a software for empirical absorption
correction, University of Göttingen, 1996.
22 SHELXTL Reference Manual, Version 5.03, Siemens Energy &
Automation, Inc., Analytical Instrumentation, Madison, WI, 1996.
23 C. J. Jameson and J. Mason, in Multinuclear NMR, ed. J. Mason,
Plenum, New York, 1987, ch. 4, p. 81
24 G. K. Carson, P. A. W. Dean and M. J. Stillman, Inorg. Chim. Acta,
1981, 56, 59; 1985, 108, 71.
25 G. K. Carson and P. A. W. Dean, Inorg. Chim. Acta, 1982, 66, l57.
26 R. J. Goodfellow, in Multinuclear NMR, ed. J. Mason, Plenum,
New York, 1987, ch. 21.
27 M. Bonamico, G. Dessy, V. Fares and L. Scaramuzza, J. Chem. Soc.,
Dalton Trans., 1976, 67.
28 C. K. Johnson, ORTEP II, Report ORNL-5138, Oak Ridge
National Laboratory, Oak Ridge, TN, 1976.
29 A. Bondi, J. Chem. Phys., 1969, 68, 441.
It would be interesting to have a structural analysis of
complex 3. If its anion had a pyramidal HgS3 skeleton, pyr-
amidalisation could be correlated more definitely with the
Paper 9/06494J
J. Chem. Soc., Dalton Trans., 1999, 4419–4423
4423