Trimethylplatinum(iv) complexes of dithiocarbamato ligands: An experimental NMR study on the barrier to C-N bond rotation

Article abstract of DOI:10.1039/b000264j

The trimethylplatinum(iv) complexes of a number of dithiocarbamato ligands have been prepared. The complexes are dimeric in the solid state and in solution, with the ligands acting in both a bridging and chelating fashion. Restricted rotation about the ligand C-N bonds in solution leads to the formation of four rotomers. The kinetics of the restricted rotation was measured by a variety of dynamic NMR techniques in the slow and intermediate exchange regimes. Two distinct processes are observed, namely the independent rotation about each C-N bond and correlated rotation about both C-N bonds. The free energies of activation, which are strongly dependent on the nature of the ligand substituents, are in the range 65-88 kJ mol"1 at 298 K. The origins of the barrier to rotation and the effects of the nitrogen substituents are discussed. The crystal structures of/ac-[PtMe3(Me2NCS2)]2 and /ac-[PtMe3(Ph(H)NCS2)]2 are reported. The Royal Society of Chemistry 2000.

Full text of DOI:10.1039/b000264j

View Article Online / Journal Homepage / Table of Contents for this issue
Trimethylplatinum(IV) complexes of dithiocarbamato ligands:
an experimental NMR study on the barrier to C–N bond rotation
Peter J. Heard,*a Kenneth Kite,b Julie Søgaard Nielsena and Derek A. Tocherc
a School of Biological and Chemical Sciences, Birkbeck University of London, Gordon House,
29 Gordon Square, London, UK WC1H 0PP. E-mail: p.heard@chem.bbk.ac.uk
b Department of Chemistry, University of Exeter, Exeter, UK EX4 4QD
c Department of Chemistry, University College London, Christopher Ingold Laboratories,
20 Gordon Street, London, UK WC1H 0AJ
Received 12th January 2000, Accepted 8th March 2000
Published on the Web 30th March 2000
The trimethylplatinum() complexes of a number of dithiocarbamato ligands have been prepared. The complexes
are dimeric in the solid state and in solution, with the ligands acting in both a bridging and chelating fashion.
Restricted rotation about the ligand C–N bonds in solution leads to the formation of four rotomers. The kinetics
of the restricted rotation was measured by a variety of dynamic NMR techniques in the slow and intermediate
exchange regimes. Two distinct processes are observed, namely the independent rotation about each C–N bond
and correlated rotation about both C–N bonds. The free energies of activation, which are strongly dependent on
the nature of the ligand substituents, are in the range 65–88 kJ molϪ1 at 298 K. The origins of the barrier to rotation
and the effects of the nitrogen substituents are discussed. The crystal structures of fac-[PtMe3(Me2NCS2)]2 and
fac-[PtMe3(Ph(H)NCS2)]2 are reported.
Recent theoretical studies on the origins of the rotational
Introduction
barrier in amides, based on Atoms in Molecules theory,13
have cast doubt on the validity of the resonance model.
According to the Atoms in Molecules theory the barrier to a
conformational rearrangement depends on the net change
in the attractive (Va) and repulsive (Vr) forces within mol-
ecules.14 The barrier to C–N bond rotation in amides is thus
thought to arise as a result of a large decrease in the attractive
forces at the nitrogen atom as it becomes pyramidal in the
transition state. There is a small reduction in repulsive forces in
the transition state, but this is insufficient to offset the decrease
in Va. Substituents on the carbonyl carbon atom that are able
to donate electron density to the sp2 hybridised nitrogen,
through the σ framework, in the planar ground state, are there-
fore expected to increase the rotational barrier; the decrease in
Va will be greater on going to the transition state. This model
readily explains the relative barrier heights to C–N bond
rotation in formamide and thioformamide. The C(S)H group is
more polarisable than the C(O)H group and can therefore
donate more electron density to nitrogen in the ground state.
Restricted rotation about the amide C–N bond is of interest
because it plays an important role in helping to determine the
secondary and tertiary structures of peptides,1 and the origins
of the barrier have been the subject of a number of recent
theoretical studies.2–6
Classically, resonance structures (Scheme 1) have been used
Scheme 1
to rationalise the barrier to C–N bond rotation in amides and
related systems:7 the greater the contribution of structure (B),
the higher is the barrier to rotation. Factors that lead to the
stabilisation of (B), such as electron donating substituents on
nitrogen or electron withdrawing carbon substituents, should
lead to an increase in the barrier to rotation. One of the main
objections to the resonance model is its apparent failure to pre-
dict the relative barrier heights to rotation in formamide and
thioformamide. As oxygen is more electronegative than sulfur,
structure (B) should be more stable in H2NC(O)H than
H2NC(S)H. Formamide might therefore be expected to exhibit
the higher barrier to C–N bond rotation, but the experimentally
measured barrier is lower.8–11 The observed trend in the activ-
ation energies can in fact be rationalised on the basis of the
resonance model if π basicity is also considered. Oxygen is
more π-basic than sulfur and consequently structure (A)
(Scheme 1) will be more stable in H2NC(O)H than H2NC(S)H.
The barrier to rotation thus depends on the balance between
the electronegativity and the π basicity. Studies on monothio-
carbamates indicate π basicity is dominant,12 suggesting that
the barrier should actually be higher in thioformamide than
formamide.
Furthermore, calculations indicate little change in the C᎐O or
᎐
C᎐S bond lengths on rotation, inconsistent with a resonance
᎐
model.
The conclusions of the Atoms in Molecules analysis have
been challenged on theoretical grounds15 and recent experi-
mental work aimed at testing the model was inconclusive.16
Dithiocarbamates are important and versatile compounds
that find widespread use as fungicides, bactericides, for the
vulcanisation of rubber and in the petrochemical industry.17
Metal complexes of dithiocarbamates are used to prepare metal
sulfide thin films by metal–organic chemical vapour deposition
(MOCVD)18 and in regenerative solar cell technology to
improve light-to-electrical energy conversion.19 Like amides,
dithiocarbamates exhibit restricted C–N bond rotation20 and
provide another experimental model for probing the origins of
the rotational barrier. We report here the results of an experi-
mental dynamic NMR study on C–N bond rotation in tri-
methylplatinum() complexes of dithiocarbamates. These
DOI: 10.1039/b000264j
J. Chem. Soc., Dalton Trans., 2000, 1349–1356
This journal is © The Royal Society of Chemistry 2000
1349

View Article Online
complexes were chosen because they are stable and provide
suitable NMR handles on the stereodynamics.
automation programs T1IR and T1IRPG) and analysed using
the program T1CALS.25 Selective inversion experiments were
carried out using our program SOFTPULV, which generates
the pulse sequence D1–180Њ–U-BURP256–VD–90Њ– free induc-
tion decay. The soft U-BURP256 pulse26 was centred on the
peak to be inverted. The relaxation delay, D1, was 10 s for the
1H experiments and 2 s for the 195Pt experiments. The VD list
contained typically 16–20 delays. Exchange rates were extracted
from the longitudinal magnetisations using the program
CIFIT.27 Rate constants obtained from the dynamic NMR
spectra were used to calculate the Eyring activation parameters;
errors quoted are those defined by Binsch and Kessler.28
Infrared spectra were recorded as pressed KBr discs on a
Nicolet Magna 205 FT-IR spectrometer, operating in the range
4000–400 cmϪ1. Mass spectra (LSIMS) were obtained at the
London School of Pharmacy, on a VG Analytical ZAB-SE4F
Instrument. Elemental analyses were performed at University
College.
Experimental
Materials
Starting materials were purchased from Aldrich Chemical
Company and used without further purification. Trimethyl-
platinum() sulfate was prepared by our published method.21
The dithiocarbamate ligands were all prepared similarly, as
illustrated by the procedure for ethylammonium N-ethyl-
dithiocarbamate.
Ethylammonium N-ethyldithiocarbamate. The reaction was
carried out under an atmosphere of dry, oxygen-free nitrogen
using standard Schlenk techniques. 20 cm3 of a 2.0 mol dmϪ3
methanolic solution of ethylamine were added dropwise to a
diethyl ether solution of carbon disulfide (1.5 cm3 in ca. 5 cm3
of Et2O). The resulting solution was stirred for ca. 1 h. The
product was precipitated by the addition of 100 cm3 of diethyl
ether and isolated by filtration. No further purification was
necessary. Yield 2.86 g (86%).
Crystallography
Single crystals of the complexes [PtMe3(Me2NCS2)]2 2 and
[PtMe3(Ph(H)NCS2)]2 6 were obtained as described above and
mounted on glass fibres. Geometric and intensity data were
obtained using an automatic four-circle Nicolet R3mV dif-
fractometer, using the ω–2θ technique at 293(2) K. Three
standard reflections, remeasured every 97 scans, showed no sig-
nificant loss of intensity during the data collection of either
sample. Data were corrected using routine procedures and
empirical absorption corrections applied (Ψ scan method). The
structures were solved by direct methods and refined to con-
vergence by least squares (SHELXL 93).29 Crystal data, collec-
tion parameters, and refinement parameters are reported in
Table 2.
Complexes. All procedures were performed in the dark, to
prevent the photoreduction of the trimethylplatinum() group.
In a typical experiment, an aqueous solution of the appropriate
ligand (in small excess) was added to a stirred aqueous solution
of trimethylplatinum() sulfate. A dense white precipitate of
the complex formed immediately. The product was isolated by
filtration, dried in air, and crystallised at ambient temper-
ature from a benzene–light petroleum (bp 60–80 ЊC) mixture.
Analytical data are reported in Table 1.
CCDC reference number 186/1892.
Physical methods
See http://www.rsc.org/suppdata/dt/b0/b000264j/ for crystal-
lographic files in .cif format.
Hydrogen-1 NMR spectra were recorded on either a Bruker
AMX400 or AMX600 Fourier transform spectrometer, operat-
ing at 400.13 and 600.13 MHz, respectively; chemical shifts are
quoted relative to tetramethylsilane as an internal standard.
Solution 195Pt NMR spectra were recorded on the same instru-
ments, operating at 85.75 and 128.63 MHz, respectively;
chemical shifts are quoted relative to the absolute frequency
scale, Ξ(195Pt) = 21.4 MHz. Probe temperatures, controlled by
standard B-VT 2000 units, were checked periodically against a
standard sample of ethylene glycol, and are considered accurate
to 1 K. Probe temperatures were allowed to equilibrate for ca.
15 min before the acquisition of each spectrum. Band shapes
were analysed (non-iteratively) using the program DNMR 5.22
The solid-state CPMAS 195Pt NMR spectrum of complex 7
was obtained by Dr A. E. Aliev on a Bruker MSL300 Fourier
transform spectrometer, operating at 64.4 MHz.
Results
Complexes 1–7 were prepared as described above and isolated
as air-stable, crystalline solids. Infrared spectra show three
bands in the region 3000–2800 cmϪ1, characteristic of a fac
octahedral [PtMe3]ϩ moiety.21 The C–S and C–N stretching
modes of the dithiocarbamato ligands are also readily identi-
fied;30 the positions of the C–N stretching bands indicate an
intermediate bond order between one and two.30 Bands show
little dependence on the nature of the ligand. Liquid secondary
ion mass spectrometry (LSIMS) was performed on all com-
plexes and, with the exception of 2 and 4, which completely
fragmented on ionisation, good spectra were obtained. The
highest observed mass/charge peaks correspond to species of
general formula [PtMe3(dtc)]2 (dtc = dithiocarbamate), indicat-
Hydrogen-1 two-dimensional exchange (EXSY) spectra were
obtained using the Bruker automation program NOESYTP,
which generates the pulse sequence D1–90Њ–D0–90Њ–D8–90Њ–
free induction decay. Spectra were recorded typically with 2048
and 512 words of data in f2 and f1, respectively. The initial delay,
D0, was 3 µs and relaxation delay, D1, was 5.0 s. The mixing
time, D8, was varied according to the complex under invest-
igation and the experimental temperature. Exchange rates were
extracted from the resulting two-dimensional intensity matrix
using the program D2DNMR.23 The 195Pt EXSY NMR spec-
trum of complex 7 was acquired at 333 K on the Bruker
AMX400 instrument (see above) using the NOESYTP pro-
gram, which was modified to enable 1H decoupling in f1 and f2.
The relaxation delay, D1, was 2 s, the initial variable delay, D0,
was 3 µs, and the mixing time, D8, was 0.1 s. The spectrum was
acquired with 256 words of data in f1 and 4096 words of data in
f2, and a spectral width of ca. 350 ppm in both domains.
Hydrogen-1 and 195Pt spin–lattice relaxation data were
obtained using the inversion-recovery technique24 (Bruker
1350
J. Chem. Soc., Dalton Trans., 2000, 1349–1356

View Article Online
Table 1 Analytical data for complexes 1–7
Infrared datac/cmϪ1
Analysisd (%)
C
Mass spectral
Complex
Yielda
75
datab
ν(N–H)
ν(C–H)
ν(C–N)
ν(C–S)
ν(Pt–C)
H
N
1
665 [M ϩ H]ϩ
3344
3259
3177
1609
2965
2899
2805
1609
848
615
596
15.19(14.45)
3.36(3.34)
4.07(4.22)
649 [M Ϫ CH3]ϩ
619 [M Ϫ 3CH3]ϩ
2
3
61
54
2947
2896
2800
2964
2895
2806
2800
2960
2891
2803
2961
2892
2804
2954
2897
2853
2956
2895
2806
1530
1516
968
990
20.29(19.99)
24.78(24.73)
4.13(4.20)
4.96(4.94)
3.86(3.89)
3.53(3.61)
776 [M]ϩ
555
761 [M Ϫ CH3]ϩ
731 [M Ϫ 3CH3]ϩ
4
5
6
7
85
57
60
71
3267
3266
3248
1517
1518
1527
1489
968
619
556
534
588
549
17.42(17.34)
20.13(19.99)
30.04(29.40)
31.30(31.27)
3.71(3.79)
4.17(4.20)
3.74(3.71)
4.03(4.06)
3.96(4.04)
3.81(3.89)
3.29(3.43)
3.24(3.32)
705 [M Ϫ CH3]ϩ
675 [M Ϫ 3CH3]ϩ
630 [M Ϫ 6CH3]ϩ
816 [M]ϩ
969
615
976
801 [M Ϫ CH3]ϩ
771 [M Ϫ 3CH3]ϩ
844 [M]ϩ
974
620
829 [M Ϫ CH3]ϩ
799 [M Ϫ 3CH3]ϩ
a Percentage yield relative to [Pt(CH3)3]2SO4ؒ4H2O. b LSIMS technique. c Recorded as pressed KBr discs; not all bands observed. d Calculated figures
in parentheses.
Table 2 X-Ray dataa for complexes 2 and 6
2
6
Formula
Mr
C12H30N2Pt2S4
720.80
C20H30N2Pt2S4
816.68
Point group
Space group
Radiation
a; b; c/Å
Orthorhombic
Pccn
Mo-Kα (λ = 0.71073 Å)
10.106(2); 10.656(2); 19.655(4)
Monoclinic
P21/c
Mo-Kα (λ = 0.71073 Å)
10.611(2); 17.725(4); 14.373(3)
90.0; 94.37(3)
2695.4(10)
2.013
α; β; γ/Њ
U/Å3
2116.6(7)
2.262
13.592
Dc/mg mϪ3
µ/mmϪ1
10.687
Data collection range/Њ
Reflections collected
2.78 < θ < 25.06
3550
1866 (0.100)
0.0522
0.1322
0.0774
2.57 < θ < 25.06
4750
4485 (0.0295)
0.0590
0.1316
0.1179
Unique (Rint
)
R [I > 2σ(I)]
wR2 [I > 2σ(I)]
R (all data)
wR2 (all data)
0.1846
0.2103
a Estimated standard deviation in parentheses.
ing the presence of dimeric species. The observed isotope
distribution patterns are in accordance with those calculated for
the proposed species. Analytical data are reported in Table 1.
bond angles is ≈360Њ). Selected bond distances and angles are
reported in Table 3.
NMR studies
Crystal structures of complexes 2 and 6
Hydrogen-1 and 195Pt NMR spectroscopic data indicate that
the trimethylplatinum() dithiocarbamato complexes, 1–7, are
also dimeric in solution. The dimers presumably retain the C2-
symmetric configuration observed in the solid state. There is no
evidence of monomeric or any other dimeric species in solution.
Dithiocarbamates are able to stabilise metal centres in a num-
ber of ways.31 The trimethylplatinum() cation usually forms
32–36
dimeric complexes of the type fac-[PtMe3L]2
with anionic
donors, such as acetylacetonates. Mass spectral data indicate
that the trimethylplatinum() dithiocarbamato complexes, 1–7,
also form dimeric species (see above). The crystal structures of
2 and 6 (Fig. 1) were obtained to confirm the presence of dimers
in the solid state. The structures show clearly that the ligands
act in both a chelating and bridging fashion, giving C2-
symmetric dimers; the nature of the ligand substituents does
not affect the molecular structure. In both complexes the
ligand C–N distances are intermediate between single and
double bonds and the nitrogen atoms are planar (sum of the
Complexes 1–3. The static 1H NMR spectra of the complexes
of the symmetrically substituted dithiocarbamato ligands, 1–3,
display signals due to the dithiocarbamato substituents and the
platinum–methyl groups. The two dithiocarbamato ligands are
related by a C2-axis and are indistinguishable in the NMR
experiment, but because there is no plane of symmetry along
the C–N bonds the two substituents on nitrogen are inequiv-
alent. The NMR spectra of the complexes thus display two
J. Chem. Soc., Dalton Trans., 2000, 1349–1356
1351

View Article Online
Fig. 2 400 MHz 1H NMR spectrum of [PtMe3(Me2NCS2)]2 2 at
298 K.
Fig. 3 128 MHz 195Pt NMR spectrum of complex 7 in CDCl3
at 298 K.
reported in Table 4. The static 195Pt NMR spectra of complexes
1–3 each display a single resonance (Table 5).
Restricted rotation about the C–N bonds of the dithio-
carbamato ligands leads to the formation of four degenerate
rotomers in solution; these are depicted in Scheme 2. Rotation
about the C–N bonds exchanges the dithiocarbamato N sub-
Fig. 1 Molecular structures of complexes 2 and 6, showing the atom
1
stituents and can readily be monitored by H dynamic NMR
labelling schemes.
spectroscopy. The complexes display signs of irreversible
decomposition at elevated temperatures, frustrating accurate
Table 3 Selected bond lengths (Å) and angles (Њ) for complexes 2
and 6a
1
analysis of the H DNMR line shapes. The exchange kinetics
was therefore measured in the slow exchange regime by either
two-dimensional exchange spectroscopy (EXSY)23,37 or selec-
tive inversion experiments,38–41 as appropriate. Eyring activation
parameters are reported in Table 6.
2
6
Pt(1)–S(1)
Pt(1)–S(2)
Pt(1)–C(1)
Pt(1)–C(2)
Pt(1)–C(3)
S(1)–C(4)
S(2)–C(4)
N(1)–C(4)
N(1)–C(5)
N(1)–C(6)
2.574(4)
2.490(4)
2.10(2)
2.09(2)
2.09(2)
1.77(2)
1.71(2)
1.34(2)
1.49(3)
1.47(2)
Pt(1)–S(1)
Pt(1)–S(2)
Pt(1)–C(1)
Pt(1)–C(2)
Pt(1)–C(3)
S(1)–C(7)
S(2)–C(7)
N(1)–C(7)
N(2)–C(14)
N(1)–C(8)
2.517(5)
2.489(5)
2.09(2)
2.08(2)
2.05(3)
1.79(2)
1.71(2)
1.33(3)
1.35(3)
1.43(2)
Complexes 4–7. Restricted rotation about the C–N bonds
in the complexes of the unsymmetrically substituted dithio-
carbamato ligands, 4–7, also leads to the formation of four
rotomers (Scheme 2), three of which are chemically distinct.
This is seen clearly in the 195Pt NMR spectra of the complexes.
The two platinum atoms are equivalent in rotomers (I) and (III)
(Scheme 2) and each therefore gives a single signal. Rotomers
(I) and (III) are distinguished readily on the basis of their
relative intensities; (I) is the more abundant species (see below).
In the degenerate pair of rotomers, (IIa)/(IIb), the C2-symmetry
of the molecule is broken and the two platinum atoms are
chemically distinct, giving two equally intense lines, with 195Pt
satellites. The observation of 195Pt satellites is conclusive evi-
dence that the complexes exist as dimers in solution. The 195Pt
NMR spectrum of [PtMe3(Ph(Me)NCS2)]2 7 is shown in Fig. 3
and 195Pt NMR data are reported in Table 5.
S(1)–Pt(1)–S(2)
S(1)–C(4)–S(2)
C(1)–Pt(1)–S(1)
C(2)–Pt(1)–S(2)
C(4)–N(1)–C(6)
C(5)–N(1)–C(6)
71.78(14)
115.1(10)
172.3(5)
173.0(6)
122(2)
S(1)–Pt(1)–S(2)
S(3)–Pt(2)–S(4)
S(1)–C(7)–S(2)
C(1)–Pt(1)–S(1)
C(2)–Pt(1)–S(2)
C(7)–N(1)–C(8)
71.9(2)
72.0(2)
114.5(11)
170.1(8)
169.9(8)
132(2)
118.1(14)
a See Fig. 1 for atom labelling scheme; estimated standard deviations
given in parentheses.
equally intense sets of signals in the dithiocarbamato region
and three equally intense signals, with 195Pt satellites (2JPtH ≈ 70
Hz), in the platinum–methyl region. The 1H NMR spectrum of
[PtMe3(Me2NCS2)]2 2 is shown in Fig. 2 and 1H NMR data are
The 1H NMR spectra of complexes 4–7 display three sets of
resonances of differing intensities, due to rotomers (I), (IIa)/
(IIb), and (III). The spectra of all four complexes are analogous
and the example of complex 7, [PtMe3(Ph(Me)NCS2)]2, will
1352
J. Chem. Soc., Dalton Trans., 2000, 1349–1356

View Article Online
Table 4 Hydrogen-1 NMR dataa for complexes 1–7
δ ligand
R1
R2
b
Complex
Rotamer
Population(%)
δ Pt–CH3
1c
2
0.71(73); 0.80(71); 1.13(76)
0.94(≈70); 0.94(≈70); 1.22(76)
0.89(73); 0.91(71); 1.20(75)
8.62
9.05
3.34
3.22
3
1.27;d 3.47;d,e
3.90;d,e
1.36;d 3.56;d,e
3.92d,e
7.00g
4f
(I)
67
15/15
0.87(74); 0.90(71); 1.18(75)
0.86(74); 0.93(73); 0.99(72); 1.19 (76); 1.27(76)
3.19g
(IIa)/(IIb)
3.03g
7.37g
3.19g
6.99g
(III)
(I)
(IIa)/(IIb)
3
70
13/13
3.01g
7.40g
5f
0.87(73); 0.92(75); 1.22(75)
0.85; 0.90; 0.91; 0.98; 1.21
1.28;d 3.57;d–f
3.72d,e,g
1.27;d 3.42;d,e,g
3.64d,e,g
1.18;d 3.60;d,e,g
3.72d,e,g
—
6.82g
6.82g
7.21g
(III)
(I)
(IIa)/(IIb)
4
76
11/11
7.26g
8.48
8.48
8.85
8.88
3.45
3.56
3.59
3.77
6f
7
0.99(≈70); 0.99(≈70); 1.30(72)
0.03(74); 0.84(71); 0.95(72); 0.98; 1.18(75); 1.35(75)
7.2–7.7
7.2–7.7
7.2–7.7
7.2–7.7
7.1–7.6
7.1–7.6
7.1–7.6
7.1–7.6
(III)
(I)
(IIa)/(IIb)
3
46
20/20
0.02; 0.79(71); 1.21
0.98(69); 1.01(64); 1.30(76)
Ϫ0.02(73); 0.70(70); 0.97(69); 1.06(73); 1.09(76); 1.35(76)
(III)
14
0.07(74); 0.66(71); 1.14(76)
b 2
a Spectra recorded at 298 K in (CDCl2)2, exceptc,f; chemical shifts quoted relative to tetramethylsilane; not all signals resolved.
J
/Hz given in
PtH
d 3
e 2
g 3
parentheses. c Recorded in (CD3)2CO at 298 K.
J
≈ 7 Hz.
J
≈ 14 Hz. f Recorded at 263 K.
J
≈ 5 Hz.
HCCH
HCH
HCNH
Scheme 2 The four rotomers of complexes 1–7 and the possible interconversion pathways between them. Note k1 = k3 and k4 = k6.
serve to illustrate the analysis of the problem. The ambient
temperature H NMR spectrum of 7 displays signals in three
regions, namely the platinum–methyl region (ca. δ Ϫ0.5 to 1.5),
the ligand-methyl region (ca. δ 3.4–3.8), and the aromatic
region (ca. δ 7.1–7.6).
The platinum–methyl region displays twelve signals, with
195Pt satellites (2JPtH ≈ 70 Hz). Rotomers (I) and (III) each give
three equally intense Pt–Me signals, and are easily distinguished
on the basis of their relative chemical shifts; one of the
equatorial Pt–Me groups of rotomer (III) lies within the shield-
ing current of the dithiocarbamato phenyl ring and is con-
sequently shifted to lower frequency. All six Pt–Me groups are
inequivalent in (IIa)/(IIb), giving six equally intense resonances,
one of which is shifted to lower frequency by the shielding
current of one of the phenyl rings. The ligand-methyl region
displays four signals: rotomers (I) and (II) each give a single
signal, while (IIa)/(IIb) give a pair of equally intense signals.
The relative abundance of the rotomers (Table 4), determined
by integration of the N–Me resonances, is in the order
(I) > (IIa) = (IIb) > (III). The aromatic region of the spectrum
is highly complex owing to the overlap of the N-phenyl reson-
ances of the different rotomers, frustrating a full assignment. It
is difficult to distinguish unambiguously between rotomers (I)
and (III) and (IIa) and (IIb) in complexes 4 and 5 (i.e. when
no phenyl group is present); the assignments given in Tables 4
and 5 may be reversed. This does not affect the analysis of the
1
J. Chem. Soc., Dalton Trans., 2000, 1349–1356
1353

View Article Online
stereodynamics. Hydrogen-1 NMR data for complexes 4–7 are
reported in Table 4.
Correlated rotation about both C–N bonds in (II),
exchanging rotomers (IIa) and (IIb), is also possible. In prin-
ciple this can also be followed by 1H EXSY experiments, but the
signals due to (IIa)/(IIb) are not sufficiently well separated to
enable EXSY cross-peaks between them to be characterised
properly. Kinetic data were therefore sought by 195Pt exchange
NMR spectroscopy. The acquisition of an informative two-
dimensional 195Pt EXSY spectrum was not immediately suc-
cessful. Several experiments had to be performed (at higher
temperatures) with reduced mixing times before a suitable spec-
trum was obtained. The 195Pt EXSY spectrum of complex 7
shows cross-peaks between the signals due to (IIa)/(IIb). Quan-
titative data were not obtained, but from the relative cross-peak
intensities it is clear that the rate of magnetisation transfer,
On warming, reversible band shape changes were observed in
all regions of the spectrum, indicating the onset of C–N bond
rotation at a measurable rate. The N–Me signals provide the
most suitable spectroscopic handle, so were used for the
measurement of the exchange kinetics. The complex displays
signs of irreversible decomposition at higher temperatures, so
rates were obtained from magnetisation transfer experiments
(EXSY). The EXSY spectrum of 7 (Fig. 4) displays cross-
peaks due to exchange between pairs of rotomers, namely
(I)
(IIa)/(IIb) and (IIa)/(IIb)
(III) as a consequence
of rotation around the C–N bond of one of the dithiocarb-
amato ligands. Cross-peaks are also observed between rotomers
(I) and (III), due to magnetisation transfers that result either
from the correlated rotation of both dithiocarbamato C–N
bonds or the stepwise rotation about first one C–N bond, then
the other. For an equilibrium process of the type shown in
Scheme 3, the stepwise rate constant, kAC, is given by eqn. (1)
k[(IIa)
(I)
(IIb)], is of the same order of magnitude as for the
(III) process, indicating that correlated rotation about
both C–N bonds is occurring.
The problems of acquiring an informative 195Pt EXSY spec-
trum were thought to arise because of rapid spin–lattice relax-
ation. Rates of relaxation were therefore measured using the
standard inversion-recovery technique24 and found to be very
rapid. Platinum-195 relaxation is usually dominated by either
the spin-rotation (SR) mechanism or chemical shift anisotropy
(CSA) mechanism42,43 and relaxation rates were measured
under various conditions to try to distinguish them. Data,
Scheme 3
k1k2
kAC
=
(1)
k
Ϫ1 ϩ kϪ2
where k1 = k[(I) → (II)] and k2 = k[(II) → (III)]. The experi-
mentally measured rates, k[(I) → (III)], are greater than the
rates calculated for a stepwise process at all temperatures
measured. Thus, although a significant contribution from the
stepwise exchange to the observed magnetisation transfers is
likely, data clearly show that correlated rotation about both
dithiocarbamato C–N bonds is occurring, interchanging
rotomer (I) and (III).
Table 5 Platinum-195 NMR dataa for complexes 1–7
Complex
T/K
δ (I)
δ (IIa)/(IIb)b
δ (III)
1c
2
298
298
298
233
233
233
298
1416
1577
1564
1623
1626
1632
1584
1601d
3
4
1609, 1598 (80)
1609, 1607 (80)
1600, 1587 (76)
1575, 1569 (87)
1587
1596
1582
1559
5
6
7
a Chemical shifts quoted relative to the absolute frequency scale
c
Ξ(195Pt) = 21.4 MHz; data recorded in CDCl3 except ; see Scheme 2
b 2
for rotomer labelling.
J
/Hz given in parentheses. c Recorded
PtPt
1
in (CD3)2CO. d Solid-state CPMAS NMR data, δ11 989, δ22 1502,
δ33 2306, ∆σ = Ϫ1060 ppm, η = 0.73 (CSA tensors assigned according
to Haeberlen’s convention).48
Fig. 4 400 MHz H two-dimensional exchange (EXSY) spectrum of
[PtMe3(Ph(Me)NCS2)]2 7 at 323 K, showing the N–Me region. Cross-
peaks are observed between rotomers (I) and (IIa)/(IIb), (IIa)/(IIb) and
(III), and (I) and (III).
Table 6 Activation parametersa for complexes 1–7
Complex
Processb
∆H‡/kJ molϪ1
∆S‡/J KϪ1 molϪ1
∆G‡/kJ molϪ1
1c
2
75.01(3.35)
83.63(1.98)
123.81(9.89)
82.01(0.77)
55.51(0.80)
73.29(1.03)
86.81(2.37)
82.51(2.21)
83.73(2.81)
12.02(10.76)
71.43(0.14)
81.54(0.19)
82.89(0.56)
72.27(0.05)
66.36(0.01)
65.61(0.01)
86.45(0.29)
87.24(0.27)
87.15(0.35)
7.0(6.0)
d
3
e
e
e
4
12.02(2.41)
Ϫ36.40(2.63)
25.78(3.39)
1.22(6.98)
Ϫ15.88(6.51)
Ϫ11.49(8.27)
5
6
7f
(I) → (IIa)/(IIb)
(IIa)/(IIb) → (III)
(I) → (III)
a Errors given in parentheses; ∆G‡ quoted at 298 K. Data measured on (CDCl2)2 solutions of the complexes exceptc. b See Scheme 2 for labelling.
c Data measured in (CD3)2CO. d Data not reported (unreliable due to narrow temperature range). e Spectra were simulated using equal rate constants
for the four independent rate processes (see Scheme 2). f No quantitative data obtained for the process (IIa) → (IIb) (see text).
1354
J. Chem. Soc., Dalton Trans., 2000, 1349–1356

View Article Online
Table 7 Spin–lattice relaxation timesa for complex 7
dithiocarbamato ligands and the C(S) atom of the other [the
S ؒ ؒ ؒ C distance ≈3.28 Å (cf. sum of van der Waals radii =
3.55 Å)].
Rotomer
The resonance model has been called into question. Recent
calculations using Atoms in Molecules theory indicate that the
barrier to C–N bond rotation results from the loss of attractive
potential, Va, at N as a consequence of the C–N bond lengthen-
ing on rotation.2–6 Electron donating substituents on nitrogen
will increase the attractive potential at nitrogen in the ground
state and are therefore expected to increase the barrier to
rotation. The loss of attractive potential is offset slightly by a
small reduction in the repulsive interactions between the nitro-
gen substituents in the (pyramidal) transition state. The net
effect of the nitrogen substituents on the rotational barriers will
depend on the balance between their electronic and steric
effects. The effect of hydrogen substitution on the barrier can
readily be rationalised using this model. With small substitu-
ents, such as hydrogen, the reduction in repulsive forces on
pyramidalisation will, presumably, be minimal and the barrier
height will depend (essentially) on the change in attractive
potential only. Since hydrogen is a poorer σ donor than alkyl
(or aryl) groups, the attractive potential at N in the ground state
will decrease on substitution with H and the barrier to rotation
will be lowered. A more detailed rationalisation of the observed
trend in the free energies is problematic because the net effects
of the different nitrogen substituents are not precisely known.
The effect of metal co-ordination on the C–N rotation
kinetics cannot be determined. Assuming that there are no
unfavourable steric interactions between the nitrogen substitu-
ents and the metal moiety (which appears to be the case from
the crystal structures), any co-ordination effect will presumably
be constant for all complexes. Preliminary studies by our group
on (free) selenothiocarbamate ligands and their trimethyl-
platinum() complexes, [PtMe3(R1R2NC(S)Se)]2, suggest that
co-ordination has only a minimal effect on the energetics. This
is in contrast to recent work on the tricarbonylrhenium()
complexes of pyridine-2,6-bis(N,N-dialkylcarbothioyl amide,
C5H3N(C(S)NR2)2-2,6) (L), [ReX(CO)3(L)] (X = Cl, Br, or I),
which show that co-ordination of carbothioyl group to the
metal centre reduces the barrier to C–N bond rotation
significantly.47
Field/T
6.3
T/K
Solvent
(I)
(IIa)/(IIb)
(IIa)/(IIb)
(III)
303
323
303
323
303
323
303
323
(CDCl2)2
CDCl3
(CDCl2)2
CDCl3
(CDCl2)2
CDCl3
(CDCl2)2
CDCl3
(CDCl2)2
CDCl3
(CDCl2)2
CDCl3
(CDCl2)2
CDCl3
(CDCl2)2
CDCl3
215
552
346
683
110
266
154
326
66
179
98
230
43
241
607
377
740
110
297
170
376
77
210
111
257
47
211
554
339
686
98
259
145
343
64
176
95
227
42
251
639
400
744
126
309
180
405
79
215
117
269
51
9.4
11.8
14.1
120
65
159
129
70
172
117
63
153
143
76
183
a Spin–lattice relaxation times (×10Ϫ3 s); errors ca. 5%; see Scheme 2
for rotomer labelling.
collected in Table 7, are consistent with the CSA mechanism
being dominant. This is supported by the solid-state 195Pt
CPMAS NMR spectrum of complex 7, which indicates a large
chemical shielding anisotropy: ∆σ = Ϫ1060 ppm (Table 5).
Small contributions from other mechanisms are also
apparent.42,43
Reliable rate data for the C–N bond rotations were measured
for complexes 4–7 by band shape analysis or magnetisation
transfer experiments. Data were used to calculate the Eyring
activation parameters, which are reported in Table 6.
Discussion
The trimethylplatinum() dithiocarbamato complexes, 1–7,
exist as dimers in the solid state and in solution. In solution,
restricted rotation about the dithiocarbamato C–N bonds leads
to the formation of four rotomers. The kinetics of C–N bond
rotation was measured by DNMR spectroscopy and activation
data are reported in Table 6. The free energies of activation, the
most reliable measure of the energetics, show a dependence on
the nitrogen substituents of the dithiocarbamato ligand, par-
ticularly when an alkyl or aryl group is substituted by hydrogen;
substitution leads to a significant decrease in ∆G‡ (at 298 K).
Complex 1, [PtMe3(H2NCS2)]2, might therefore be predicted to
have the lowest barrier to C–N bond rotation. The higher than
expected barrier in 1 (Table 6) probably results because of the
more polar solvent system [complex 1 was studied in acetone
because of its poor solubility]. Polar solvents, such as acetone
and DMSO, have been shown to increase barriers to C–N bond
rotation.44–46 The solvent effect arises because there is a large
decrease in the dipole moment on rotation and consequently
polar solvents destabilise the transition state.44–46
Classically, a resonance model has been used to rationalise
the barrier to C–N bond rotation in amides and related systems,
such as dithiocarbamates.7 The barrier is presumed to arise
because of a contribution from (B) (Scheme 1) to the overall
structure. The amount of nitrogen π donation will be affected
by the substituents on N; electron-donating substituents should
increase π donation and hence the barrier to rotation. Thus the
rotational barrier would be expected to increase when H is sub-
stituted for an alkyl group. This is the case (Table 6). Correlated
rotation around both C–N bonds (see above) provides some
evidence of delocalisation. For the correlated process to occur
there must be communication between the two dithiocarb-
amato ligands in the dimer. This is thought most likely to occur
either via the platinum centres or as a result of through space
interactions between the non-bridging S atom of one of the
Conclusion
The results obtained on the restricted C–N bond rotations in
the dithiocarbamato complexes, 1–7, do not provide any con-
clusive experimental evidence in support of either model (see
above), but some evidence of delocalisation is provided by the
correlated rotation about the C–N bonds of both dithiocarb-
amato ligands. A detailed theoretical study on origin of the
barrier and the effects of the nitrogen substituents is currently
underway in our group; results will be published shortly.
Acknowledgements
Birkbeck University of London is acknowledged for a student-
ship (J. S. N.), and we are grateful to Johnson-Matthey for the
loan of platinum. Drs G. Steadman (University of Wales,
Swansea) and P. King (Birkbeck) are acknowledged for helpful
discussions.
References
1 A. L. Lehninger, D. L. Nelson and M. M. Cox, Principles of
Biochemistry, Worth, New York, 1993.
2 K. B. Wiberg and K. E. Laidig, J. Am. Chem. Soc., 1987, 109, 5935.
3 K. B. Wiberg and C. M. Breneman, J. Am. Chem. Soc., 1992, 144,
831.
4 L. M. Cameron and K. E. Laidig, Can. J. Chem., 1993, 71, 872.
5 K. B. Wiberg and P. R. Rablen, J. Am. Chem. Soc., 1995, 117, 2201.
6 K. E. Laidig and L. M. Cameron, J. Am. Chem. Soc., 1996, 118,
1737.
J. Chem. Soc., Dalton Trans., 2000, 1349–1356
1355

View Article Online
7 L. Pauling, The Nature of the Chemical bond, Cornell University
Press, Cornell, 1960.
8 J. Sandström, J. Phys. Chem., 1967, 71, 2318.
9 R. C. Neuman, Jr. and L. B. Young, J. Phys. Chem., 1965, 69, 2570.
10 A. Loewenstein, R. Melera, P. Rigny and W. Walter, J. Phys. Chem.,
1964, 68, 1597.
11 W. E. Stewart and T. H. Siddall III, Chem. Rev., 1970, 70, 517.
12 B. J. McCormick, R. Bereman and D. Baird, Coord. Chem. Rev.,
1984, 54, 99.
28 G. Binsch and H. Kessler, Angew. Chem., Int. Ed. Engl., 1980, 19,
411.
29 G. M. Sheldrick, SHELXL 93, University of Göttingen, 1993.
30 K. Nakamoto, J. Fujita, R. A. Condrate and Y. Morimoto, J. Chem.
Phys., 1963, 39, 423.
31 F. A. Cotton and G. Wilkinson, Advanced Inorganic Chemistry, 5th
edn., Wiley, New York, 1995.
32 R. B. King (Editor), Encyclopaedia of Inorganic Chemistry, Wiley,
New York, 1994, vol. 5, p. 2581.
13 R. F. W. Bader, Atoms in molecules: a quantum theory, Oxford
University Press, Oxford, 1990.
14 R. F. W. Bader, J. R. Cheeseman, K. E. Laidig, K. B. Wiberg and
C. Breneman, J. Am. Chem. Soc., 1990, 112, 6530.
15 C. J. Perrin, J. Am. Chem. Soc., 1991, 113, 2865.
16 A. D. Bain, P. Hazendonk and P. Couture, Can. J. Chem., 1999, 77,
1340.
33 A. C. Hazell and M. R. Truter, Chem. Ind. (London), 1959, 564.
34 A. C. Hazell and M. R. Truter, Proc. R. Soc. London, Ser. A, 1960,
252, 218.
35 A. G. Swallow and M. R. Truter, Proc. R. Soc. London, Ser. A, 1960,
252, 205.
36 R. N. Hargreaves and M. R. Truter, J. Chem. Soc. A, 1969,
2282.
17 F. Ullman, Encyclopedia of Industrial Chemistry, 5th edn., Wiley,
New York, 1985, vol. A9, p. 1.
18 M. Motevalli, P. O’Brien, J. R. Walsh and I. M. Watson, Polyhedron,
1996, 15, 2801.
19 R. Argazzi, C. A. Bignozzi, G. M. Hasselman and G. J. Meyer,
Inorg. Chem., 1998, 37, 4533.
37 C. L. Perrin and T. J. Dwyer, Chem. Rev., 1990, 90, 935.
38 A. D. Bain and J. A. Cramer, J. Phys. Chem., 1993, 97, 2884.
39 A. D. Bain and J. A. Cramer, J. Magn. Reson., 1993, 103A, 217.
40 S. Forsen and R. A. Hoffman, J. Phys. Chem., 1963, 39, 2892.
41 R. E. Hoffman and S. Forsen, Prog. Nucl. Magn. Reson. Spectrosc.,
1966, 1, 15.
20 R. C. Fay, Coord. Chem. Rev., 1996, 154, 99.
21 E. W. Abel, P. J. Heard, K. Kite, K. G. Orrell and A. F. Psaila,
J. Chem. Soc., Dalton Trans., 1996, 1233.
42 R. K. Harris and B. E. Mann, NMR and the Periodic Table,
Academic Press, London, 1978.
43 P. S. Pregosin, Ann. Rep. N.M.R. Spectrosc., 1986, 17, 285.
44 K. B. Wiberg, P. R. Rablen, D. J. Rush and T. A. Keith, J. Am. Chem.
Soc., 1995, 117, 4261.
22 D. S. Stephenson and G. Binsch, DNMR 5, Quantum Chemistry
Program Exchange, Indian University.
ˇ
23 E. W. Abel, T. P. J. Coston, K. G. Orrell, V. Sik and D. Stephenson,
45 P. R. Rablen, D. A. Miller, V. R. Bullock, P. H. Hutchinson and
J. A. Gorman, J. Am. Chem. Soc., 1999, 121, 218.
46 P. R. Rablen, S. A. Pearlman and D. A. Miller, J. Am. Chem. Soc.,
1999, 121, 227.
J. Magn. Reson., 1986, 70, 34.
24 R. K. Harris, Nuclear Magnetic Resonance Spectroscopy, Wiley,
New York, 1986.
25 A. D. Bain, T1CALS, McMaster University, 1995.
26 H. Geen and R. Freeman, J. Magn. Reson., 1991, 93, 93.
27 A. D. Bain and J. A. Cramer, J. Magn. Reson., 1996, 74, 819.
ˇ
47 K. G. Orrell, A. G. Osborne, J. O. Prince, V. Sik and D. K. Vellianitis,
Eur. J. Inorg. Chem., 2000, 383.
48 U. Haeberlen, Adv. Magn. Reson. (suppl. 1), 1976.
1356
J. Chem. Soc., Dalton Trans., 2000, 1349–1356