Carbon-Sulfur Bond Cleavage in Bis(N-alkyldithiocarbamato)cadmium(II) Complexes: Heterolytic Desulfurization Coupled to Topochemical Proton Transfer

Article abstract of DOI:10.1021/ic035135v

Three bis(N-alkyldithiocarbamato)cadmium(II) complexes [Cd(S ₂CNHR)₂] (1, R = n-C₃H₇; 2, R = n-C₅H₁₁; 3, n-C₁₂H₂₅) were prepared by metathesis of the corresponding lithium salt, Li [S₂CNHR], with cadmium chloride. The crystal structures of 2 and 3 consist of planar molecular units of [Cd(S₂CNHR)₂] connected by intermolecular Cd...S interactions to give a one-dimensional chain. The chains are connected by a network of intermolecular N-H...S hydrogen bonds between the dithiocarbamato nitrogen atom and bridging sulfur atoms in neighboring chains. In solution, the ¹¹³Cd NMR spectrum of 2 is dependent on concentration and temperature, indicative of a dimerization equilibrium mediated by similar Cd...S intermolecular bridging interactions. In the solid state, thermal gravimetric analyses show that all three complexes decompose smoothly via a heterolytic C-S bond cleavage reaction to give the corresponding alkyl isothiocyanate and cadmium sulfide as the primary products, with the formation of primary amine and CS₂ as coproducts. These products can result only from the net transfer of protons between N-alkyldithiocarbamato ligands in the solid state. Thus, the C-S bond cleavage reaction is interpreted in terms of the topochemical arrangement of molecular units in the crystalline state, which provides a pathway for proton transfer between ligands via N-H...S hydrogen bonds. Decomposition was also initiated by addition of a tertiary amine to a solution of [Cd(S ₂CNHR)₂]. This confirms that C-S bond cleavage must be coupled to deprotonation of the -NH group, and explains why dialkylated derivatives [Cd(S₂CNR₂)₂] are inert to this particular mode of C-S bond cleavage. This system thus constitutes an unusual example of heterolytic, nonoxidative C-S bond cleavage that appears to proceed by a topochemical transfer of protons, which has implications for C-S bond cleavage processes in single-source precursors for II-VI semiconductor materials.

Full text of DOI:10.1021/ic035135v

Inorg. Chem. 2004, 43, 3180−3188
Carbon−Sulfur Bond Cleavage in Bis(N-alkyldithiocarbamato)cadmium(II)
Complexes: Heterolytic Desulfurization Coupled to Topochemical
Proton Transfer
Laura H. van Poppel, Thomas L. Groy, and M. Tyler Caudle*
Department of Chemistry and Biochemistry, Arizona State UniVersity, Tempe, Arizona 85287-1604
Received September 30, 2003
Three bis(N-alkyldithiocarbamato)cadmium(II) complexes [Cd(S CNHR)2] (1, R) n-C H ; 2, R) n-C H ; 3, n-C H )
2
3
7
5
11
12 25
were prepared by metathesis of the corresponding lithium salt, Li[S CNHR], with cadmium chloride. The crystal
2
structures of 2 and 3 consist of planar molecular units of [Cd(S CNHR)2] connected by intermolecular Cd‚‚‚S
2
interactions to give a one-dimensional chain. The chains are connected by a network of intermolecular N−H‚‚‚S
hydrogen bonds between the dithiocarbamato nitrogen atom and bridging sulfur atoms in neighboring chains. In
solution, the 113Cd NMR spectrum of 2 is dependent on concentration and temperature, indicative of a dimerization
equilibrium mediated by similar Cd‚‚‚S intermolecular bridging interactions. In the solid state, thermal gravimetric
analyses show that all three complexes decompose smoothly via a heterolytic C−S bond cleavage reaction to give
the corresponding alkyl isothiocyanate and cadmium sulfide as the primary products, with the formation of primary
amine and CS as coproducts. These products can result only from the net transfer of protons between
2
N-alkyldithiocarbamato ligands in the solid state. Thus, the C−S bond cleavage reaction is interpreted in terms of
the topochemical arrangement of molecular units in the crystalline state, which provides a pathway for proton
transfer between ligands via N−H‚‚‚S hydrogen bonds. Decomposition was also initiated by addition of a tertiary
amine to a solution of [Cd(S CNHR)2]. This confirms that C−S bond cleavage must be coupled to deprotonation
2
of the −NH group, and explains why dialkylated derivatives [Cd(S CNR ) ] are inert to this particular mode of C−S
2
2 2
bond cleavage. This system thus constitutes an unusual example of heterolytic, nonoxidative C−S bond cleavage
that appears to proceed by a topochemical transfer of protons, which has implications for C−S bond cleavage
processes in single-source precursors for II−VI semiconductor materials.
Introduction
sublime. Recently, these properties have been exploited to
prepare II-VI semiconductor materials using N,N-dialkyl-
dithiocarbamato complexes as single-source precursors.
Thermolytic desulfurization of dialkyldithiocarbamato com-
plexes of cadmium and zinc has become a method of choice
for the synthesis of metal sulfide semiconductor materials
by chemical vapor deposition.3 In addition, nanoparticulate
clusters of group 12 sulfides are prepared by thermolysis of
[M(S2CNR2)2] precursors in high-boiling solvents.4 Thermal
A considerable body of literature exists on the preparation
and structure of N,N-dialkyldithiocarbamato complexes of
cadmium(II) and zinc(II) ions.1,2 These compounds are air
and moisture stable and sufficiently volatile that they readily
* To whom correspondence should be addressed. E-mail: tcaudle@
asu.edu.
(1) Jian, F.-F.; Xue, T.; Jiao, K.; Zhang, S.-S. Chin. J. Chem. 2003, 21,
50-55. Baba, I.; Farina, Y.; Kassim, K.; Othman, A. H.; Razak, I.
A.; Fun, H. K.; Ng, S. W. Acta Crystallogr., Sect. E 2001, 57, 55-
56. Ivanov, A. V.; Forshling, V.; Kritikos, M.; Antsutkin, O. N. Russ.
J. Coord. Chem. (Transl.) 2000, 26, 53-62. Sreehari, N.; Varghese,
B.; Manoharan, P. T. Inorg. Chem. 1990, 29, 4011-4015. Francetic,
V.; Leban, I. Vestn. SloV. Kem. Drus. 1979, 26, 113-122. Zvonkova,
Z. V.; Khvatkina, A. N.; Ivanova, N. S. Kristallografiya 1967, 12,
1065-1068. Bonamico, M.; Mazzone, G.; Vaciago, A.; Zambonelli,
L. Acta Crystallogr. 1965, 19, 898-909. Bonamico, M.; Dessy, G.;
Mazzone, G.; Mugnoli, A.; Vaciago, A.; Zambonelli, L. Atti Accad.
Naz. Lincei, Cl. Sci. Fis., Mat. Nat., Rend. 1963, 35, 338-347.
(2) Tiekink, E. R. T. Z. Kristallogr. 2000, 215, 445-446.
(3) Rumyantsev, Y. M.; Fainer, N. I.; Kosinova, M. L.; Ayupov, B. M.;
Sysoeva, N. P. J. Phys. IV 1999, 9, 777-784. O’Brien, P.; Walsh, J.
R.; Watson, I. M.; Hart, L.; Silva, S. R. P. J. Cryst. Growth 1996,
167, 133-142. Fainer, N. I.; Kosinova, M. L.; Rumyantsev, Y. M.;
Salman, E. G.; Kuznetsov, F. A. Thin Solid Films 1996, 280, 16-19.
Fainer, N. I.; Rumyantsev, Y. M.; Kosinova, M. L.; Kuznetsov, F. A.
Proc. Electrochem. Soc. 1997, 97-25, 1437-1442. Zeng, D.; Hamp-
den-Smith, M. J.; Alam, T. M.; Rheingold, A. L. Polyhedron 1994,
13, 2715-2730.
3180 Inorganic Chemistry, Vol. 43, No. 10, 2004
10.1021/ic035135v CCC: $27.50 © 2004 American Chemical Society
Published on Web 04/10/2004

Carbon-Sulfur Bond CleaWage
Scheme 1
Scheme 2
Scheme 3
N-alkyldithiocarbimide intermediate, which is also formed
by deprotonation of an N-alkyldithiocarbamato ligand,
Scheme 2.
decomposition of N,N-dialkyldithiocarbamato complexes
probably proceeds via free-radical decomposition of the
dialkyldi-
Group 12 N,N-dialkyldithiocarbamato complexes can be
prepared by the direct reaction of a metal salt with CS2 in
the presence of a secondary amine, Scheme 3.9 However,
our attempts to prepare N-alkyldithiocarbamatocadmium(II)
complexes by substituting a primary amine in Scheme 3
resulted instead in apparent desulfurization of CS2 at room
temperature. As the expected product was bis(N-alkyldithio-
carbamato)cadmium(II), this led to the hypothesis that
desulfurization occurs from an N-alkyldithiocarbamato ligand,
and that the C-S bond is activated by deprotonation of the
-NH group. This contrasts with the behavior of N,N-
dialkyldithiocarbamatocadmium(II) derivatives which cannot
form the dithiocarbimide intermediate and thus show no
evidence for carbon-sulfur bond cleavage under the same
conditions.
thiocarbamato ligand since elimination of sulfide also
requires concomitant elimination of an alkyl group. Free
radical decomposition is consistent with the formation of
disulfide products from the thermal decomposition of N,N-
dialkyldithiocarbamato complexes.5 The C-S bond of N,N-
dialkyldithiocarbamato anions can also be cleaved by
oxidative addition to trivalent group 5 metal ions,6 but this
would appear to be unrelated to their thermolytic cleavage
in the presence of redox-inert group 12 ions.
Given this extensive body of literature on N,N-dialkyl-
dithiocarbamato complexes of Cd2+ and Zn2+, it is surprising
to find a substantial void concerning N-alkyldithiocarbamato
complexes of group 12 ions. In fact, only one example of a
structurally or chemically characterized N-alkyldithiocarba-
mato derivative of cadmium(II) could be located.7 This may
arise in part from the fact that such complexes are predicted
to be considerably less volatile, and thus not suitable as
single-source CVD precursors. However, it may also arise
from different chemical reactivity of the N-alkyldithiocar-
bamato species, which makes them less stable than their N,N-
dialkylated counterparts. The N-alkyldithiocarbamato anion
is formally the product of condensation between HS- and
an alkylisothiocyanate, Scheme 1a. The presence of a
dissociable proton suggests that N-alkyldithiocarbamato
ligands might eliminate sulfide via heterolytic desulfurization,
which should yield the corresponding isothiocyanate. This
is not the case for N,N-dialkyldithiocarbamato derivatives,
Scheme 1b, since this would require a chemically unlikely
heterolytic cleavage of an N-Calkyl bond. There is chemical
precedent for heterolytic C-S bond breaking in the desulfu-
rization of CS2 observed when CS2 is reacted with imido
complexes of zirconium(IV).8 This reaction proceeds via an
In this paper we describe the synthesis, solid state structure,
and solution structure of a new class of bis(N-alkyldithio-
carbamato)cadmium(II) complexes and show that they
decompose via heterolytic desulfurization of the dithiocar-
bamato ligand under relatively mild conditions. Decomposi-
tion generates readily identifiable products by solid state
thermolysis and base-initiated decomposition at room tem-
perature, which both occur via a common heterolytic C-S
activation process involving abstraction of the -NH proton
of the N-alkyldithiocarbamato ligand. This reaction thus
provides considerable new insight into nonoxidative C-S
bond cleavage reactions and their use as a source of sulfide
for the synthesis of important metal sulfide materials.
Experimental Section
n-Pentylamine (C5H11NH2) and dodecylamine (C12H25NH2) were
acquired from Aldrich and used without further purification.
n-Propylamine (C3H7NH2) was acquired from Aldrich and distilled
from CaH2 prior to use. CdCl2 was dried overnight at 120 °C to
remove any water present and then stored under N2. Phenyllithium
was obtained from Acros Organics and stored under N2. Proton
and 13C NMR spectra were measured on a Varian Gemini 300 MHz
spectrometer. 113Cd NMR was measured on a Varian Inova 400
MHz spectrometer. Samples for infrared spectrometry were prepared
as a KBr pellet, and FTIR spectra were collected on a Nicolet Avatar
360 FTIR spectrometer. TG/DTA thermal analyses were performed
on a Setaram TG 92 instrument. Residual gas analysis was measured
using a Stanford Research Systems RGA 200 mass spectrometer.
Powder X-ray diffraction data were acquired using Rigaku D/MAX-
(4) Malik, M. A.; Revaprasadu, N.; O’Brien, P. Chem. Mater. 2001, 13,
913-920. Lazell, M.; O’Brien, P. J. Chem Soc., Chem. Commun. 1999,
2041-2042. Malik, M. A.; O’Brien, P.; Revaprasadu, N. Chem. Mater.
2002, 14, 2004-2010. Sreekumari Nair, P.; Revaprasadu, N.;
Radhakrishnan, T.; Kolawole, G. A. J. Mater. Chem. 2001, 11, 1555-
1556. Trindade, T.; O’Brien, P.; Zhang, X.-M. Chem. Mater. 1997,
9, 523-530.
(5) Barone, G.; Chaplin, T.; Hibbert, T. G.; Kana, A. T.; Mahon, M. F.;
Molloy, K. C.; Worsley, I. D.; Parkin, I. P.; Price, L. S. J. Chem Soc.
Dalton Trans. 2002, 1085-1092. Wieber, M.; Schmidt, E. Z. Anorg.
Allg. Chem. 1988, 556, 179-188.
(6) Gilletti, P. F.; Femec, D. A.; Keen, F. I.; Brown, T. M. Inorg. Chem.
1992, 31, 4008-4012.
(7) Agre, V. M.; Shugam, E. A.; Rukhadze, E. G. Tr. IREA 1967, 30,
369-371.
(8) Zuckerman, R. L.; Bergman, R. G. Organometallics 2000, 19, 4795-
4809.
(9) McCowan, C. S.; van Poppel, L.; Caudle, M. T. Unpublished results.
Inorganic Chemistry, Vol. 43, No. 10, 2004 3181

van Poppel et al.
Table 1. Summary of Single Crystal X-ray Diffraction Data for
Compound 2
II diffractometer. Compositional analyses were run in the in-house
facility at the ASU Center for Solid State Science.
formula
C12H24CdN2S4
436.97
orthorhombic
Pca21
Li[S2CNHR] (R ) C3H7, C5H11, C12H25). In a typical prepara-
tion, 0.58 mL (7.1 mmol) of n-propylamine was carefully reacted
with 1 equiv of phenyllithium in anhydrous tetrahydrofuran under
an atmosphere of nitrogen. After stirring for 1 h, 0.43 mL (7.1
mmol) of carbon disulfide was carefully added to this solution. The
reaction was stirred for 6 h, and then, the solvent was removed in
vacuo. The residue was washed with dichloromethane, producing
0.92 g (92%) of a flocculant white powder that gave characterization
data consistent with Li[S2CNHC3H7]. Li[S2CNHC5H11] (yield:
64%) and Li[S2CNHC12H25] (yield: 82%) were prepared by similar
procedure.
fw
cryst syst
space group
color
colorless
14.6353(7)
4.1454(2)
30.3184(15)
1839.39(15)
4
a (Å)
b (Å)
c (Å)
V (Å3)
Z
temp (K)
λ (Å)
300
0.71073
12098
measured reflns
indep reflns
obsd reflns
R
3189
3037 (|Fo| > 4.0σ|Fo|)
1
Li[S2CNHC3H7]. H NMR (ppm, DMSO-d6): 7.89 [1H, br s,
0.0640
Rw
0.1557
NH], 3.27 [2H, m, J(H-H) ) 7 Hz, NCH], 1.43 [2H, m, J(H-H)
GOF
1.198
) 7 Hz, â-CH], 0.79 [3H, t, J(H-H) ) 7 Hz, CH3]. IR (cm-1
,
x23
0.46(8)
KBr): 3251 (s), 2962 (m), 2918 (s), 2847 (s), 1510 (s), 1437 (m),
1159 (w), 1061 (w), 946 (s), 869 (m). Anal. Found (Calcd) %: C,
34.2 (34.1); H, 5.6 (5.7); N, 9.8 (9.9); S, 45.3 (45.4).
[Cd(S2CNHC12H25)2] (3). A solution of 0.26 g (0.96 mmol) of
Li[S2CNHC12H25] in 10 mL of THF was combined with a solution
of 0.090 g (0.48 mmol) of anhydrous CdCl2 in 10 mL of THF,
producing a white precipitate. The solvent was reduced to yield
more white precipitate. Both batches of precipitate were combined
and recrystallized from a THF/toluene solution yielding 0.25 g
1
Li[S2CNHC5H11]. H NMR (ppm, DMSO-d6): 7.85 [1H, m,
NH], 3.31 [2H, m, NCH], 1.42 [2H, m, J(H-H) ) 7 Hz, â-CH],
1.22 (4H, m, CH2), 0.85 [3H, t, J(H-H) ) 7 Hz, CH3]. IR (cm-1
,
KBr): 3251 (s), 2956 (m), 2923 (s), 2858 (s), 1503 (s), 1470 (m),
1383 (m), 1350 (m), 1301 (w), 1142 (w), 1061 (w), 946 (s), 673
(m). Anal. Found (Calcd) %: C, 49.5 (49.6); H, 7.3 (7.1); N, 8.2
(8.3); S, 37.7 (37.9).
Li[S2CNHC12H25]. 1H NMR (ppm, DMSO-d6): δ 7.86 [1H, m,
NH], 3.32 [2H, m, NCH], 1.41 [2H, m, â-CH], 1.23 [18H, br s,
CH2], 0.84 [3H, t, CH3]. IR (cm-1, KBr): 3251 (s), 2962 (m), 2918
(s), 2850 (s), 1508 (s), 1470 (m), 1394 (m), 1361 (m), 1301 (w),
1110 (w), 1061 (w), 946 (s), 667 (m). Anal. Found (Calcd) %: C,
53.6 (53.8); H, 10.0 (10.1); N, 5.1 (5.2); S, 23.8 (23.9).
(81%) of
3 as colorless crystals. Mp (decomposition):
1
135-138 °C. H NMR (ppm, DMSO-d6): 9.88 [1H, t, NH], 3.24
[2H, m, J(H-H) ) 6.5 Hz, NCH], 1.51 [2H, m, â-CH], 1.23 (18H,
m, CH2), 0.86 [3H, t, J(H-H) ) 6.5 Hz, CH3]. 13C NMR (ppm,
DMSO-d6): δ 14.42, CH3; 22.56, CH2CH3; 26.83, CH2; 28.28, CH2;
29.18, CH2; 29.47, 3C, CH2; 29.51, 3C, CH2; 31.75, CH2; 49.93,
CH2NH; 206.64, CS2. IR (cm-1, KBr): 3207 (s), 3011 (w), 2956
(m), 2918 (s), 2847 (m), 1530 (s), 1470 (m), 1399 (m), 1339 (m),
1312 (m), 940 (m), 722 (m), 673 (m). Anal. Found (Calcd) % for
[Cd(S2CNHC12H25)2]: C, 49.3 (49.3); H, 8.2 (8.2); N,4.4 (4.4); S,
20.2 (20.3).
[Cd(S2CNHC3H7)2] (1). A solution of 0.12 g (0.86 mmol) of
Li[S2CNHC3H7] in 10 mL of distilled water was combined with a
solution of 0.08 g (0.43 mmol) of anhydrous CdCl2 in 10 mL of
distilled water to produce a white precipitate. This precipitate was
recrystallized from a THF/toluene solution yielding 0.12 g (75%)
Single-Crystal Diffraction Studies. Single crystals of 2 and 3
were coated with mineral oil and mounted on the tip of a glass
fiber. Single-crystal diffraction data were collected at ambient
temperature on a Bruker SMART APEX system using graphite
monochromated Mo KR radiation (λ ) 0.71073 Å) and corrected
for Lorentz and polarization effects. SHELXS-97 and SHELXL-
97 were used for structure solution and refinement. Refinement of
F2 data was against all reflections. The weighted R and goodness-
of-fit are based on F2, and conventional R factors are based on F,
with F set to zero for negative F2. The threshold expression of F2
> 2σ(F2) was used only for calculating R factors and is not relevant
to the choice of reflections for refinement. The data set for
compound 2 refined to a suitable R-value, and pertinent details of
its data collection and refinement are provided in Table 1.
Compound 3 did not yield an anisotropic refinement of publishable
R-value, but the isotropic refinement was of sufficient quality to
establish atom connectivity and packing.
1
of 1 as colorless crystals. Mp (decomposition): 130-133 °C. H
NMR (ppm, DMSO-d6): 9.86 [1H, t, NH], 3.24 [2H, m, J(H-H)
) 5 Hz, NCH], 1.51 [2H, m, J(H-H) ) 7 Hz, â-CH], 0.84 [3H,
t, J(H-H) ) 7 Hz, CH3]. 13C NMR (ppm, DMSO-d6): δ 11.26,
CH3; 21.12, CH2CH3; 51.67, CH2NH. IR (cm-1, KBr): 3213 (s),
3016 (m), 2962 (m), 2923 (m), 2880 (m), 1530 (s), 1459 (w), 1400
(m), 1395 (m), 1295 (m), 1153 (w), 1066 (w), 965 (m), 891 (w),
667 (m), 563 (w). Anal. Found (Calcd)% for Cd(S2CNHC3H7)2:
C, 25.1 (25.2); H, 4.2 (4.2); N, 7.4 (7.4); S, 33.7 (33.7).
[Cd(S2CNHC5H11)2] (2). A solution of 0.24 g (1.42 mmol) of
Li[S2CNHC5H11] in 10 mL of distilled water was combined with a
solution of 0.13 g (0.71 mmol) of anhydrous CdCl2 in 10 mL of
distilled water producing a pale yellow precipitate. This precipitate
was recrystallized from a THF/CH3CN solution yielding 0.27 g
(84%) of
2 as colorless crystals. Mp (decomposition):
Results Section
140-143 °C. 1H NMR (ppm, DMSO-d6): 9.88 [1H, t, J(H-H) )
6 Hz, NH], 3.26 [2H, m, J(H-H) ) 6 Hz, NCH], 1.51 [2H, m,
J(H-H) ) 7 Hz, â-CH], 1.26 (4H, m, CH2), 0.86 [3H, t, J(H-H)
) 7 Hz, CH3]. 13C NMR(ppm, DMSO-d6): δ 13.79, CH3; 21.72,
The metathesis reaction between a lithium N-alkyldithio-
carbamato salt and cadmium chloride in neutral aqueous
solution cleanly produces bis(N-alkyldithiocarbamato)cad-
mium(II) complexes 1-3 in good yield and purity. By
contrast, condensation of the primary amine with carbon
disulfide in the presence of cadmium chloride gave negligible
yields of [Cd(S2CNHR)2]. Instead, the latter reaction yields
cadmium sulfide in essentially quantitative yield based on
CH2CH3; 27.38, CH2; 28.42, CH2; 49.88, CH2NH. IR (cm-1
,
KBr): 3198 (s), 3008 (m), 2995 (s), 2922 (s), 2850 (s), 1529 (s),
1464 (m), 1418 (s), 1339 (m), 1306 (m), 1240 (w), 1142 (w), 1063
(m), 945 (s), 728 (w), 662 (s), 557 (w). Anal. Found (Calcd)% for
Cd(S2CNHC5H11)2: C, 32.5 (32.9); H, 5.3 (5.5); N, 6.0 (6.4); S,
29.0 (29.4).
3182 Inorganic Chemistry, Vol. 43, No. 10, 2004

Carbon-Sulfur Bond CleaWage
Figure 1. Molecular structure of 2 (50% probability ellipsoids). Important
bond lengths (Å): Cd-S(1a), 2.674(2); Cd-S(2a), 2.592(3); Cd-S(1b),
2.672(2); Cd-S(2b), 2.648(3); C(1a)-S(1a), 1.770(10); C(1a)-S(2a), 1.698-
(9); C(1b)-S(1b), 1.749(10); C(1b)-S(2b), 1.686(9); C(1a)-N(1a), 1.302-
(12); C(1b)-N(1b), 1.308(12). Important bond angles (deg): S(1a)-Cd-
S(2a), 69.19(8); S(1b)-Cd-S(2b), 68.73(7); S(1a)-Cd-S(2b), 111.38(8);
S(2a)-Cd-S(1b), 110.70(8); S(1a)-C(1a)-S(2a), 119.1(5); S(1b)-C(1b)-
S(2b), 121.9(5).
CdCl2. The sulfide anion must arise from CS2, and so this
reaction corresponds to net desulfurization of carbon disul-
fide. Preparative control experiments using diethylamine
under identical conditions instead gives quantitative yields
of Cd[S2CNEt2]2, the most well-known of the bis(N,N-
dialkyldithiocarbamato)cadmium(II) complexes. The results
from preparative experiments therefore establish either a clear
difference between the reactivity of primary and secondary
amines with CS2, or a difference between the reactivity of
N-alkyldithiocarbamato and N,N-dialkyldithiocarbamato ligands
coordinated to Cd2+. Since stable N-alkyl- and N,N-dialkyl-
dithiocarbamatolithium salts are easily prepared by reaction
of CS2 with primary lithium amides, this would indicate that
the difference lies in the comparative chemistry of [Cd(S2-
CNHR)2] and [Cd(S2CNR2)2].
Figure 2. Intermolecular interactions in 2. n-Pentyl groups have been
truncated for clarity. Cadmium (green), sulfur (orange), nitrogen (blue),
carbon (white). Shaded lines indicate intermolecular Cd‚‚‚S interactions.
Cd‚‚‚S ) 2.93 Å, Cd‚‚‚Cd ) 4.14 Å. Dashed lines indicate intermolecular
N-H‚‚‚S hydrogen bonding interactions. N‚‚‚S distance ) 3.43 Å.
Compounds 1-3 all form colorless, very thin platelike
crystals that are flaky and difficult to mount for X-ray studies,
but otherwise stable to air and moisture. The small aspect
ratio of the crystals of 2 is a result of the orthorhombic crystal
system with a unit cell of dimensions 14.6 × 4.1 × 30.0 Å3
for 2, Table 1.10 The molecular structure of 2, Figure 1,
consists of a rhombic planar cadmium ion that results from
chelation by two N-n-pentyldithiocarbamato ligands. The two
large S-Cd-S angles are 111°, and the two smaller
S-Cd-S angles are 69°. The four sulfur atoms reside in a
perfect plane as shown by the fact that the in-plane angles
add to exactly 360°. The in-plane Cd-S bond lengths are
typical of dithiocarbamato complexes.11,12 There are three
long Cd-S bonds averaging 2.66 Å, and one that is
somewhat shorter at 2.59 Å. The reason for this distortion
is not obvious. However, the space group Pca21 lacks
inversion symmetry, so the Cd2+ ion cannot lie on an
inversion center, and this may allow one Cd-S distance to
deviate. The dithiocarbamato group is in a planar configu-
ration, which is favored by delocalization of the lone pair
from the nitrogen atom into the -CS2 unit. In the solid state,
the alkyl groups on 2 are disposed in a trans configuration
with respect to each other. However, this likely arises from
packing phenomena since there would appear to be no
obvious stability gained in the trans configuration relative
to the cis configuration.
In the crystalline solid state, complex 2 forms a one-
dimensional polymer via intermolecular Cd‚‚‚S interactions
orthogonal to the intramolecular CdS4 plane, Figure 2. This
gives an infinite linear chain of [Cd(S2CNHC5H11)2] mol-
ecules canted such that the cadmium ion of one molecular
unit is situated directly above a sulfur of the previous unit.
(11) Glinskaya, L. A.; Zemskova, S. M.; Klevtsova, R. F. Zh. Strukt. Khim.
2000, 40, 979-983. Zemskova, S. M.; Glinskaya, L. A.; Klevtsova,
R. F.; Fedotov, M. A.; Larionov, S. V. Zh. Strukt. Khim. 1999, 40,
284-292. Cox, M. J.; Tiekink, E. R. T. Z. Kristallogr. 1999, 214,
670-676. Jian, F.; Wang, Z.; Bai, Z.; You, X.; Fun, H.-K.; Chinnakali,
K. J. Chem. Crystallogr. 1999, 29, 227-231. Jian, F.-F.; Wang, Z.-
X.; Fun, H.-K.; Bai, Z.-P.; You, X.-Z. Acta Crystallogr., Sect. C 1999,
C55, 174-176. O’Brien, P.; Walsh, J. R.; Watson, I. M.; Motevalli,
M.; Henriksen, L. J. Chem. Soc., Dalton Trans. 1996, 2491-2496.
Baggio, R.; Garland, M. T.; Perec, M. Acta Crystallogr., Sect. C 1996,
C52, 823-826. Glinskaya, L. A.; Zemskova, S. M.; Klevtsova, R. F.;
Larionov, S. V.; Gromilov, S. A. Polyhedron 1992, 11, 2951-2956.
Casas, J. S.; Sanchez, A.; Bravo, J.; Garcia-Fontan, S.; Castellano, E.
E.; Jones, M. M. Inorg. Chim. Acta 1989, 158, 119-126. Agre, V.
M.; Shugam, E. A. Kristallogr. 1972, 17, 303-308. Agre, V. M.;
Shugam, E. A.; Rukhadze, E. G. Zh. Strukt. Khim. 1966, 7, 897-
898. Malissa, H.; Kolbe-Rohde, H. Talanta 1961, 8, 841-845. Albano,
V.; Domenicano, A.; Vaciago, A. Gazz. Chim. Ital. 1966, 96, 922-
934.
(10) The atoms C(5a), C(6a), C(5b), and C(6b) were refined isotropically,
since all four atoms became nonpositive definite when refined
anisotropically. The anomolous thermal behavior in these atoms arises
from the presence of two enantiomorphs cocrystallizing in an acentric
space group Pca21, which made it difficult to accurately model
empirical absorption corrections for the irregular crystal dimensions.
The large ratio of predicted to reported transmission factors, (Tmax
/
Tmin exp/(Tmax/Tmin ) 1.42, is also a result of the difficulty in
)
)
rep
determining absorption corrections for these crystal dimensions. A
TWIN refinement gave a Flack parameter x ) 0.46(8), indicating that
the crystal is composed of 54(8)% of the molecule as illustrated and
46(8)% of the inverted form.
(12) Dee, C. M.; Tiekink, E. R. T. Z. Kristallogr. 2002, 217, 85-86.
Inorganic Chemistry, Vol. 43, No. 10, 2004 3183

van Poppel et al.
Figure 3. PLUTO diagram of 3 derived from isotropic refinement of X-ray diffraction data.
Scheme 4
This results in a nominal six-coordinate environment of sulfur
atoms around each cadmium ion with two long axial Cd‚‚‚S
interactions at 2.93 Å. Packing diagrams of 2 also reveal
the existence of close contacts between the sulfur atom of
one molecule and a carbamato nitrogen atom of a neighbor-
ing molecule, Figure 2. The N‚‚‚S distance for these
interactions is about 3.4 Å. This value is within the range of
N-H‚‚‚S hydrogen bonding interactions, which typically
have N‚‚‚S distances between 3.3 and 3.6 Å.13 These
hydrogen bonding interactions serve two functions in the
solid state structure of 2. First, they connect one stack to
another and thereby support the formation of the three-
dimensional crystal structure. Second, the CdS4 planes in
one stack are oriented normal to the CdS4 planes in the
neighboring stack. In this way, each molecule of [Cd(S2-
CNHR)2] is hydrogen bonded to two different molecules of
[Cd(S2CNHR)2] in the neighboring stack. The result is that
each molecule of [Cd(S2CNHR)2] is joined to one above it
by two Cd‚‚‚S interactions and two N-H‚‚‚S hydrogen
bonds. These important structural features of 2 appear to be
general for this class of compounds. Compound 3 consists
of a molecular unit similar to that of 2, Figure 3, and was
solved in a similarly flattened P21/c unit cell. Compound 3
also exhibits an identical network of intermolecular Cd‚‚‚S
and N-H‚‚‚S interactions as 2, although the crystalline
structure of 3 is slightly different from 2 in regards to the
disposition of the N-alkyl groups.
(3251 cm-1) in the corresponding N-alkyldithiocarbama-
tolithium salts. The decreased N-H frequency observed in
the cadmium complexes supports the hypothesis that the
N-H‚‚‚S interaction arises from a weak hydrogen bond. The
dithiocarbamato group in 1-3 also gives an asymmetric
NCS2 stretching frequency at 1530 cm-1, but the lithium
dithiocarbamate salts, Li[S2CNHR], give stretching frequen-
cies at 23 cm-1 lower in energy at 1507 cm-1. This would
seem to suggest that the average C-S bond order increases
upon coordination to the cadmium ion. However, shifts in
the vibrational energies of the dithiocarbamato group may
also be influenced by the extensive intermolecular contacts
in the solid state.
The polymeric solid state structure of 1-3 is consistent
with the general observation that the compounds are highly
insoluble in noncoordinating solvents but are soluble in
coordinating solvents such as tetrahydrofuran. At low
concentrations in THF, the complexes exhibit a single 113Cd
NMR signal at approximately 270 ppm versus external
aqueous cadmium sulfate (δ ) 5 ppm). This signal arises
from the solvated molecular unit [Cd(S2CNHR)2(THF)2].
However, the chemical shift is dependent on the concentra-
tion of the complex, Figure 4, which we interpret to arise
from dimerization of the molecular unit, Scheme 4. The
probable structure of the dimer, involving intermolecular
Cd‚‚‚S interactions, is consistent with the dimeric structure
of bis(diethyldithiocarbamato)cadmium(II),2,12 and with the
polymeric structure of 2 and 3 in the solid state. The
concentration dependence results from a rapid chemical
exchange rate between the monomer and dimer so that the
113Cd chemical shift is a concentration-weighted average of
the chemical shifts of the monomer and dimer. As the
concentration of the cadmium N-alkyldithiocarbamate com-
plex is increased, the 113Cd NMR signal shifts downfield,
suggesting that the Cd nucleus becomes less shielded upon
dimerization. The observation that the dimer resonates
downfield of the monomer implies considerable ionic
The presence of hydrogen bonding to the -NH group is
also supported by infrared spectroscopy. The infrared spectra
of 1-3 give a single sharp absorption peak near 3206 cm-1
that is assigned to an N-H stretching mode. The N-H
stretching vibration occurs 45 cm-1 toward higher energy
(13) Lynch, D. E.; McClenaghan, I.; Light, M. E.; Coles, S. J. Cryst. Eng.
2002, 5, 79-94. Lynch, D. E.; McClenaghan, I. Acta Crystallogr.,
Sect. E 2001, 57, 26-27. Lynch, D. E.; McClenaghan, I. Acta
Crystallogr., Sect. E 2001, 57, 11-12. Lynch, D. E.; McClenaghan,
I. Acta Crystallogr., Sect. C 2000, 56, 587. Lynch, D. E.; McClena-
ghan, I. Acta Crystallogr., Sect. C 2003, 59, 53-56.
3184 Inorganic Chemistry, Vol. 43, No. 10, 2004

Carbon-Sulfur Bond CleaWage
∆H° - T∆S° into eq 1 gives the temperature dependence of
the chemical shift in terms of the standard enthalpy and
entropy of dimerization, eq 2.14 Fitting this expression to
the data in Figure 5 gives ∆H° ) 28(2) kJ/mol and ∆S° )
160(8) J/mol‚K. The unfavorable enthalpy for dimerization
means that the Cd-OTHF interaction is stronger than the
Cd‚‚‚S interaction, although this is almost entirely offset by
a favorable entropy term at 298 K. This suggests that the
entropy associated with liberation of solvent molecules
provides a substantial contribution to the free energy of
dimerization in solution.
Figure 4. 113Cd NMR shift for 2 dissolved in THF as a function of the
concentration of 2. Chemical shift is referenced against an internal capillary
containing 3 M CdSO4 in D2O (δ ) 5 ppm). Solid line is a fit of eq 1 to
the data with K ) 2500 M.
δ )
4a0
[THF]2
1 +
e[(∆S°/R)-(∆H°/RT)]
-
8a0
[THF]2
1 +
e[(∆S°/R)-(∆H°/RT)]
x
1 -
δmonomer
+
4a0
[THF]2
(
)
e[(∆S°/R)-(∆H°/RT)]
4a0
[THF]2
1 +
e[(∆S°/R)-(∆H°/RT)]
-
Figure 5. 113Cd NMR shift for 2 dissolved in THF as a function of
temperature. Chemical shift is referenced against an internal capillary
containing 3 M CdSO4 in D2O (δ ) 5 ppm). [2] ) 0.185 M. Solid line is
a fit to eq 2 to with ∆H° ) 28(2) kJ/mol and ∆S° ) 160(8) J/mol.
8a0
[THF]2
1 +
e[(∆S°/R)-(∆H°/RT)]
x
δdimer (2)
4a0
(
)
e[(∆S°/R)-(∆H°/RT)]
[THF]2
character to the intermolecular Cd‚‚‚S interaction and would
be consistent with the long intermolecular Cd‚‚‚S bond
observed in the X-ray structure. The binding isotherm relating
observed chemical shift, δ, to the total concentration of 2,
a0, is given in eq 1.14 It gives an acceptable fit to the data in
Figure 4, yielding an estimate for Kdim ) 2500 M, where
Kdim ) [dimer][THF]2/[monomer]2, and δmonomer and δdimer
are the chemical shifts of the monomer and dimer, respec-
tively.
Upon heating to 500 °C, all three compounds give a yellow
residue whose X-ray powder diffraction pattern is consistent
with cadmium sulfide. Heating the complexes to 195 °C for
1 h in sealed tubes permitted collection of the volatile
thermolysis products. GC/mass spectra of these products
showed the presence of the alkyl isothiocyanate15 in which
the alkyl group was derived from the dithiocarbamato ligand.
Residual gas analysis shows that CS2 is also eliminated,
which implies elimination of the primary amine as well. The
NMR spectrum of the involatile products shows broad peaks
consistent with amine that is probably coordinated with the
solid CdS under constant volume conditions of the sealed
tube experiments. However, these data do not permit us to
entirely rule out the presence of some corresponding N,N′-
dialkylthiourea in the involatile products. Symmetric N,N′-
disubstituted ureas and thioureas are formed in thermal
decompositions of certain N-alkyl- or N-arylcarbamato
compounds,16 but these have been shown to be secondary
products arising from condensation of an alkyl isothiocyanate
and amine, which are the primary products. Furthermore,
symmetric N,N′-dialkylureas and N,N′-dialkylthioureas re-
versibly dissociate under thermal conditions to give alkyl
isocyanates or alkyl isothiocyanates.16 Therefore, we con-
clude that the alkyl isothiocyanate and corresponding amine
4a0Kdim
[THF]2
8a0Kdim
[THF]2
1 +
-
1 +
x
δ ) 1 -
δmonomer +
4a0Kdim
[THF]2
(
)
4a0Kdim
[THF]2
8a0Kdim
1 +
-
1 +
[THF]2
x
δdimer (1)
4a0Kdim
[THF]2
(
)
The 113Cd chemical shift is also dependent on the absolute
temperature, as shown in Figure 5. As the temperature is
lowered from ambient to 190 K, the 113Cd NMR signal moves
upfield, which means that the dimerization equilibrium is
shifted toward the left as the temperature is decreased.
Inserting the Gibbs free energy expression -RT ln Kdim
)
(15) Table is deposited in Supporting Information.
(16) Daly, N. J.; Ziolkowski, F. Int. J. Chem. Kinet. 1980, 12, 241-252.
(14) Derivation of this equation is deposited in Supporting Information.
Drozdov, N. S. Zh. Obshch. Khim. 1936, 6, 1368-1374.
Inorganic Chemistry, Vol. 43, No. 10, 2004 3185

van Poppel et al.
represent primary products of the thermal decomposition of
[Cd(S2CNHR)2], although we do not preclude the possibility
that a fraction may be condensed as an involatile N,N′-
dialkylthiourea under the applied thermolysis conditions.
Thermolysis of 1 under constant pressure conditions occurs
in a single resolvable step at 140-200 °C, Figure 6a, leaving
38(1)% of the original mass, as expected for stoichiometric
formation CdS from 1. Compound 2 shows two resolvable
thermal steps in the TGA and DTA profiles, and these occur
at 110-150 and 180-240 °C, Figure 6b. If the temperature
of the sample is poised at 110 °C for 30 min, a thermally
stable material is formed corresponding to 62(2)% of the
original mass. The IR and proton NMR spectra of this
material show that the only residual organic material consists
of unreacted dithiocarbamato ligand. This intermediate must
therefore correspond to a material of composition CdSn(S2-
CNHR)2(1-n), where n ) 0.56. This indicates that, under
constant pressure conditions of the TGA experiment, all
neutral organic species are volatilized. Heating this same
sample to 500 °C reproducibly gives a new material
containing 39(2)% of the initial mass, giving n ) 0.91. For
complete decomposition to CdS, n ) 1, so this final material
corresponds to decomposition of 91% of the original dithio-
carbamato ligand in 2. Heating compound 3 from 15 to 500
°C gives two weight loss steps occurring at 115-150 and
200-300 °C. These are each associated with unique DTA
endotherms. A third DTA endotherm occurs at 110°C but is
not associated with any loss in weight and probably corre-
sponds to a solid-solid phase change. We ascribe the
stepwise weight loss in Figure 6c to initial decomposition
of a fraction of the N-dodecyldithiocarbamato ligand, giving
a thermally stable intermediate consisting of 85(3)% of the
original mass, so that n ) 0.19. This intermediate showed
only the N-dodecyldithiocarbamato ligand by IR and proton
NMR spectroscopy. Heating this sample to 500 °C gives a
final material with 26(2)% of the original mass, giving n )
0.93. This indicates that the product stabilizes after decom-
position of 93% of the dithiocarbamato ligand. The ther-
molysis data for 1-3 are summarized in Table 2.
The products formed in the thermal studies require that
C-S bond activation be coupled to proton transfer. However,
it is not obvious from solid state thermolysis experiments
whether -NH proton transfer is required to initiate C-S
bond cleavage or whether it occurs after C-S bond cleavage.
To address this question, we examined the reaction of 1-3
with a Brønsted base in solution. Figure 7a shows the proton
NMR spectrum of 2 in DMSO-d6, which exhibits a splitting
pattern for the R-CH2 protons at 3.26 ppm (M) that is
consistent with coupling to the â-CH2 protons at 1.51 ppm
(X) and the -NH proton at 9.88 ppm (A, not shown).
Simulation of the R-CH2 signal using a first-order AM2X2
coupling scheme gives JMA ) 5.5 Hz and JMX ) 7.5 Hz.
This split R-CH2 signal is characteristic of the N-alkyldithio-
carbamato ligand in all three complexes we studied. When
triethylamine is added to the solution of 2, we observe slow
precipitation of cadmium sulfide, indicating that C-S bond
cleavage has occurred. In the proton NMR spectrum, the
signal at 3.26 ppm is decreased, and it is replaced by a new
Figure 6. Thermal gravimetric profiles (black) and differential thermal
analysis profiles (red) for decomposition of 1 (a), 2 (b), and 3 (c) under
N2, T ) 15- 500 °C, 10 °C/min.
Table 2. Thermolysis Data for 1-3
thermolysis
temp (°C)
compd
% wta
n
nominal compositionb
1
2
140-200
110-150
180-240
115-150
200-300
38(1)c
62(2)d
39(2)c
85(3)e
26(2)c
1
CdS
0.56
0.91
0.19
0.93
Cd5S3(S2CNHC5H11)4
Cd10S9(S2CNHC5H11)2
Cd5S(S2CNHC12H25)8
Cd10S9(S2CNHC5H11)2
3
a For remaining solid residue. b Smallest stoichiometric unit. c After
heating to 500 °C. d After poising temperature at 110 °C for 30 min. e After
poising temperature at 115 °C for 30 min.
triplet signal centered at 3.64 ppm, Figure 7b. This change
is associated with a decrease in intensity of the -NH signal
at 9.88 ppm (not shown), and concomitant loss of the JMA
coupling to the R-CH2 protons. These changes indicate
formation of n-pentylisothiocyanate in situ. This was con-
firmed by a solution infrared spectrum of the product, which
exhibits the strong isothiocyanate band at 2100 cm-1. There
is no evidence from IR or NMR spectroscopy for formation
of the corresponding N,N′-dialkylthiourea. Control experi-
ments show that reaction of C5H11NH2 with C5H11NCS in
DMSO-d6 does give N,N′-di-n-pentylthiourea. Furthermore,
even upon complete decomposition, approximately half of
the original N-alkyldithiocarbamato ligand remains intact.
These observations indicate that primary amine is not
eliminated in the solution decomposition reaction, suggesting
that the second equivalent of dithiocarbamato anion is a
spectator ligand that is not decomposed in the base-catalyzed
3186 Inorganic Chemistry, Vol. 43, No. 10, 2004

Carbon-Sulfur Bond CleaWage
bond approximations. Figure 2 shows that the intermolecular
Cd(1)‚‚‚S(1a) vector is very nearly perpendicular to the CdS4
plane, as shown by the C(1a)-S(1a)‚‚‚Cd(1) and Cd(1)-
S(1b)‚‚‚Cd(1) angles of 96° and 95°, respectively. As a result,
the Cd‚‚‚S interaction must have a minimal contribution from
the filled sp2 orbital on sulfur, which is in the CdS4 plane,
and must occur primarily via one of the delocalized π-type
ligand orbitals orthogonal to the CdS4 plane. This has the
effect of polarizing the electron density toward the bridging
sulfur atom, S(1a) or S(1b), and thereby decreasing the
C-Sbridging bond order. This is consistent with the longer
C(1a)-S(1a) and C(1b)-S(1b) distances (1.76 Å) compared
with the nonbridging C(1a)-S(2a) and C(1b)-S(2b) dis-
tances (1.68 Å). The hydrogen bonding occurs to the bridging
sulfur atoms S(1a) and S(1b) and has an N-H‚‚‚S vector
that is almost perpendicular to the CdS4 plane. This means
that the hydrogen bonding must also involve the delocalized
π-orbitals of the ligand and probably contributes also to the
longer C-Sbridging bond distance. These compounds therefore
fit generally with the trend showing that noncovalent
interactions to sulfur tend to have a more orthogonal
geometry than corresponding interactions with oxygen
atoms.19
The intermolecular interactions observed in 2 and 3, and
their apparent impact on the C-S bond order, suggest a
connection between the specific topochemical arrangement
observed in the crystal and the type of decomposition we
observe in neat 1-3. The thermal decomposition reaction
requires proton transfer from one dithiocarbamato ligand to
another. The network of N-H‚‚‚S hydrogen bonds suggests
a potential pathway for proton transfer between dithiocar-
bamato ligands. This idea is supported by the observation
that decomposition can be initiated directly in the solid state
at modest temperatures, so that it must involve a minimum
of molecular rearrangement. Such a reaction will be favored
by a preorganized and concerted proton transfer pathway
directed along the N-H‚‚‚S vector. The intermolecular
Cd‚‚‚S interactions may also assist in the activation of the
C-Sbridging bond by decreasing the bond order as described
above. Scheme 5 shows a proposed mechanism for the initial
decomposition steps in 1-3 that is consistent with our
experimental evidence. Interactions between one [Cd(S2-
CNHR)2] unit and the neighboring stack are shown in gray.
Concerted proton transfer along this particular hydrogen
bonding pathway forms a tautomeric dithiocarbimido inter-
mediate X, which undergoes spontaneous C-S bond break-
ing to give the alkyl isothiocyanate and the intermediate Y.
Compound Y rearranges to Z via either intermolecular or
Figure 7. (a) 300 MHz proton NMR spectrum (DMSO-d6) of 2. The
expanded inset shows the methylene region. (b) Spectrum of a sample of
2 after addition of excess triethylamine and aging for 1 h. The expanded
inset shows changes in the methylene region. The excised portions at 0.9
and 2.5 ppm are where the triethylamine protons resonate.
reaction. This was demonstrated by adding triethylamine to
a solution containing Li[S2CNHR] and CdCl2 in a 1:1 ratio,
which results in decomposition of all of the RHNCS2- ligand
and precipitation of [Et3NH+][Cl-]. This experiment shows
that only 1 equiv of the dithiocarbamato complex is des-
ulfurized to isothiocyanate for every CdS produced, and that
inert spectator anions can substitute for the second equivalent
of dithiocarbamato ligand. While the fate of the organic
products is different for the thermal and base-initiated
decomposition reactions, both reactions are consistent with
a chemical mechanism whereby reversible proton transfer
precedes irreversible C-S bond breaking. In both cases,
proton transfer is a critical element for the activation of the
C-S bond in this system.
Discussion Section
The most pertinent features in the structure of bis(N-
alkyldithiocarbamato)cadmium(II) complexes 1-3 are the
intermolecular Cd‚‚‚S and N-H‚‚‚S interactions. The
Cd‚‚‚S interaction is reminiscent of that observed in dimeric
bis(N,N-dialkyldithiocarbamato)cadmium(II) complexes,2,12
although the geometry about cadmium is distorted square
pyramidal in the latter cases. Each [Cd(S2CNHR)2] molecular
unit in the crystal is bridged to two others in the same chain
by Cd‚‚‚S interactions, and any two adjacent molecular units
in the same chain are bridged to each other by N-H‚‚‚S
hydrogen bonding interactions with molecular units in the
neighboring chain. These weak but complementary individual
interactions result in a stronger net intermolecular interaction
than is possible when the -NH proton is substituted by
another alkyl group. Although N-H‚‚‚S hydrogen bonds are
considerably weaker than corresponding N-H‚‚‚O hydrogen
bonds, their ability to support a polymeric structure in the
solid state structure does suggest some practical significance
in crystal engineering.13 Hydrogen bonds to metal-coordi-
nated sulfur atoms are also commonly observed in sulfur
metalloproteins such as rubredoxins17 and ferrodoxins,18
although the preorganization afforded by the protein’s
secondary and tertiary structures probably provides much of
the stability for these interactions.
(17) Bougault, C. M.; Eidsness, M. K.; Prestegard, J. H. Biochemistry 2003,
42, 4357-4372. Sieker, L. C.; Holmes, M.; Le Trong, I.; Turley, S.;
Liu, M. Y.; LeGall, J.; Stenkamp, R. E. J. Biol. Inorg. Chem. 2000,
5, 505-513. Dauter, Z.; Wilson, K. S.; Sieker, L. C.; Moulis, J.-M.;
Meyer, J. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 8836-8840.
(18) Backes, G.; Mino, Y.; Loehr, T. M.; Meyer, T. E.; Cusanovich, M.
A.; Sweeney, W. V.; Adman, E. T.; Sanders-Loehr, J. J. Am. Chem.
Soc. 1991, 113, 2055-2064. Fukuyama, K.; Nagahara, Y.; Tsukihara,
T.; Katsube, Y.; Hase, T.; Matsubara, H. J. Mol. Biol. 1988, 199, 183-
193. Tsukihira, T.; Fukuyama, K.; Nakamura, M.; Katsube, Y.; Tanaka,
N.; Kakudo, M.; Wada, K.; Hase, T.; Matsubara, H. J. Biochem.
(Tokyo) 1981, 90, 1763-1773.
The coordination geometry about the bridging sulfur atoms
is highly distorted from that predicted on the basis of valence
(19) Platts, J. A.; Howard, S. T.; Bracke, B. R. F. J. Am. Chem. Soc. 1996,
118, 2726-2733.
Inorganic Chemistry, Vol. 43, No. 10, 2004 3187

van Poppel et al.
Scheme 5
Scheme 6
intramolecular proton transfer. This gives the zwitterionic
versible cleavage of the C-S bond. Although it is not clear
whether intermolecular interactions play a role in the
decomposition reaction in solution, formation of Cd‚‚‚S
adducts in solution, inferred from the 113Cd NMR experi-
ments, suggests that those interactions may provide a degree
of C-S bond activation. Such interactions may also assist
in Cd-S bond formation and nucleation of cadmium sulfide.
+
-
RNH2 CS2 species that rapidly cleaves the N-CS2 bond
to eliminate CS2 and RNH2.20 It is important that protonation
of the N-alkyldithiocarbamato ligand leads to irreversible
N-CS2 cleavage since this minimizes the problem of
N‚‚‚H recombination and resulting unproductive proton
transfer. The large lattice energy of CdS also provides a
considerable thermodynamic driving force for the overall
reaction, and is consistent with prior reports on the use of
thiophilic metal ions as desulfurizing agents for thiocarbam-
ates.21
This desulfurization reaction gives consistent stoichiomet-
ric yields of products specific to heterolytic C-S bond
cleavage, suggesting no competing homolytic or oxidative
pathways. Furthermore, identical products are obtained upon
thermolysis or base-initiated cleavage in nitrogen or in air,
where oxygen would be expected to scavenge free radical
intermediates and yield oxygenated products. The reaction
thus appears to be distinct from reports on thermolysis of
N,N-dialkyldithiocarbamato complexes of other redox-inert
metal ions,5 and from the oxidative addition of the C-S bond
observed in some redox-active species.6 This work suggests
that N-alkyldithiocarbamato complexes have potential as
precursors for the preparation of metal sulfide materials, and
have a number of advantages over N,N-dialkyldithiocarba-
mato complexes that are presently employed. We have shown
that the N-alkyldithiocarbamatocadmium(II) complexes de-
compose according to a well-defined chemical process to
yield characterized products under moderate conditions.
Furthermore, the decomposition can be initiated chemically
by addition of base, and can thereby be controlled in ways
that the thermolysis of N,N-dialkyldithiocarbamato com-
plexes cannot. The topochemical aspects of this system are
also related to the general problem of chemical reactivity in
condensed phases or on surfaces, where the static molecular
orientation is important to chemical reactivity. As a result,
the chemical mechanism for this unusual heterolytic C-S
bond cleavage reaction is of considerable interest, and is the
subject of our continuing work in this area.
Decomposition is initiated for 1-3 in a narrow temperature
range between 110 and 140 °C. However, complete decom-
position of 2 and 3 requires higher temperatures than for 1,
Table 2. This may be reflective of the difficulty of ac-
complishing intramolecular proton transfer in the later stages
of the decomposition process, where the long-range hydrogen
bond network is broken down. This is apt to be more
problematic for the bulkier dithiocarbamato ligands, and so
explains why 1 decomposes quantitatively to CdS by 280
°C whereas 2 and 3 show residual ligand even at 500 °C.
In solution, triethylamine probably serves to deprotonate
an N-alkyldithiocarbamato ligand to give the dithiocarbimido
complex, Scheme 6, which is followed by C-S bond
cleavage to give the alkyl isothiocyanate, CdS, and unreacted
carbamato salt. This scheme is supported by previous work
showing that group 10 N-alkylcarbamato complexes are
deprotonated by strong bases,22 although they are not reported
to undergo subsequent C-S bond cleavage in solution. The
C-S bond cleavage step is most reminiscent of the pathway
taken upon insertion of CS2 into group 4 metal amide bonds,
which also result in C-S bond cleavage of a putative
dithiocarbimido intermediate.8 Scheme 6 explains why direct
condensation of a primary amine, CS2, and CdCl2 does not
give [Cd(S2CNHR)2] complexes. Any N-alkyldithiocarbama-
tocadmium(II) complexes that do initially form are depro-
tonated by unreacted primary amine, culminating in irre-
Acknowledgment. The authors gratefully acknowledge
Arizona State University and the National Science Founda-
tion (Grant CHE 9985266) for financial support of this
project.
(20) Joris, S. J.; Aspila, K. I.; Chakrabati, C. L. J. Phys. Chem. 1970, 74,
860-865.
(21) Barba, N. A.; Kentenaru, K. F. IzV. Vyssh. Uchebn. ZaVed., Khim.
Khim. Tekhnol. 1975, 18, 1720-1722.
(22) Hadjikostas, C. C.; Katsoulos, G. A.; Sigalas, M. P.; Tsipis, C. A.
Polyhedron 1989, 8, 2637-2639. Tsipis, C. A.; Meleziadis, I. J.;
Kessissoglou, D. P.; Katsoulos, G. A. Inorg. Chim. Acta 1984, 90,
L19-L22. Hadjikostas, C. C.; Katsoulos, G. A.; Sigalas, M. P.; Tsipis,
C. A. Inorg. Chim. Acta 1989, 163, 173-176. Tian, Y.-P.; Duan, C.-
Y.; Lu, Z.-L.; You, X.-Z. Polyhedron 1996, 9, 1495-1502.
(23) Flack, H. D. Acta Crystallogr., Sect. A. 1983, 39, 876-881.
Supporting Information Available: Crystallographic data files
for 2 (CIF format) and 3 (TXT format), derivation of eqs 1 and 2,
and table of mass spectrometry data. This material is available free
of charge via the Internet at http://pubs.acs.org.
IC035135V
3188 Inorganic Chemistry, Vol. 43, No. 10, 2004