Bonding in cis- and trans-dichlorobis(dimethylsulfide)platinum(II) studied by vibrational (micro)spectroscopy and force field calculations

Article abstract of DOI:10.1016/S1386-1425(01)00516-9

The vibrational spectra of the square-planar complexes cis- and trans-dichlorobis(dimethylsulfide)platinum(II) have been investigated using Raman and synchrotron infrared microscopy of small samples, and mid- to far-infrared spectroscopy of macroscopic sa

Full text of DOI:10.1016/S1386-1425(01)00516-9

Spectrochimica Acta Part A 58 (2002) 129–139
www.elsevier.com/locate/saa
Bonding in cis- and
trans-dichlorobis(dimethylsulfide)platinum(II) studied by
vibrational (micro)spectroscopy and force field calculations
M. Glerup a,b, G.O. Sørensen a, P. Beichert c, M.S. Johnson a,*
a Department of Chemistry (KLV), Uni6ersity of Copenhagen, Uni6ersitetsparken 5, DK-2100, Copenhagen Ø, Denmark
b Materials Research Department, Risø National Laboratory, Frederiksborg6ej 399, 4000 Roskilde, Denmark
c Alfred Wegener Institute for Polar and Marine Research, Am Handelshafen 12, D-27570 Bremerha6en, Germany
Received 3 March 2001; received in revised form 23 April 2001; accepted 30 April 2001
Abstract
The vibrational spectra of the square-planar complexes cis- and trans-dichlorobis(dimethylsulfide)platinum(II) have
been investigated using Raman and synchrotron infrared microscopy of small samples, and mid- to far-infrared
spectroscopy of macroscopic samples. Synthesis of the compounds produces the cis- and trans-stereoisomers. Using
microscopy we were able to identify single crystals of the trans isomer and amorphous regions containing both cis and
trans isomers. The infrared and Raman data are interpreted using force field calculations. Spectral shifts are discussed
in terms of influence of the cis- and trans-ligands. It is found that vibrational spectroscopy is a sensitive technique
for measuring subtle chemical effects in transition metal complexes which complements and may improve studies
made with X-ray crystallography. © 2002 Elsevier Science B.V. All rights reserved.
Keywords: Synchrotron infrared microscopy; Raman microscopy; Platinum(II) complexes; Valence force field calculation
able ligands. In these systems the strength of a
given metal atom–ligand bond is influenced by
the ligands in the cis and trans positions. This is
known as the cis or trans influence and is related
to the cis and trans effect on the rate of substitu-
tion at a given site in the complex [1]. Measures of
bond strength such as bond length or vibrational
frequency are often used to characterize the ligand
influence. A systematic series of transition metal
complexes has been characterized using X-ray
crystallography [2], including cis-dichlorobis(di-
methylsulfide)platinum(II) [3] and trans-dichloro-
1. Introduction
The principal oxidation states of Pt are II and
IV. The first forms square planar (and occasion-
ally five-coordinate) complexes through five-coor-
dinate transition states, and the latter six-
coordinate complexes [1]. Cis–trans isomerism in
the square planar complexes is known, given suit-
* Corresponding author. Tel.: +45-353-20303; fax: +45-
353-50609.
E-mail address: msj@kiku.dk (M.S. Johnson).
1386-1425/01/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved.
PII: S1386-1425(01)00516-9

130
M. Glerup et al. / Spectrochimica Acta Part A 58 (2002) 129–139
bis(dimethylsulfide)platinum(II) [4,5]. Although
X-ray crystallography is a straightforward way to
approach the subject, the changes in geometry can
be similar to the uncertainties involved in the
method, due to thermal motion, experimental er-
rors and packing effects. In contrast, vibrational
spectroscopy can be a more sensitive and accurate
measure of subtle effects on a given bond, and the
results of the present work indicate that it can be
successfully applied to these systems. While
NMR, Raman and infrared spectroscopy have
been used to investigate the ligand influence [6,7],
this is the first study to use Raman and infrared
microscopy. A straightforward method for deter-
mining the Pearson hard/soft character of metal
cations based on the force constants of the square
planar MCl24− system has been described [8]. The
force constants were optimized based on infrared
and Raman data.
One problem when considering the spec-
troscopy of systems like the title compound is that
approximately equal amounts of each stereoiso-
mer are produced during synthesis. Microscopy is
a convenient technique for directly analyzing the
precipitate produced during synthesis. In the
present work, we were able to measure the spectra
of single crystals of the trans compound using
microscopy, including the new technique of syn-
chrotron infrared microscopy [9], which employed
a microscope at the infrared beamline at the
MAX-I electron storage ring in Lund, Sweden
[10]. This technique gives a significantly enhanced
signal for microscopy experiments as compared
with experiments performed with a blackbody
light source. Complimentary information was
gathered using Raman microscopy and mid- and
far-infrared spectroscopy, and the results assigned
with the aid of force field calculations. A number
of workers have interpreted the infrared and Ra-
man spectra of similar inorganic complexes using
normal coordinate force field analysis [11,12].
(DMS) were added dropwise to a solution of 2.3 g
K2PtCl4 in 15 ml of water. The precipitate was
allowed to stand overnight at room temperature
and then separated by filtration [4]. Powder and
single crystal X-ray diffraction showed that this
precipitate contains the two isomers cis- and
trans-Cl2(DMS)2 platinum(II) [4,5]. Recrystallisa-
tion was carried out in chloroform. The X-ray
diffraction work showed that while both isomers
can be recrystallised from chloroform, the cis
crystals form with one molecule of chloroform
which is lost when the crystals are dried, resulting
in an amorphous or powdery sample. Below we
present results which are consistent with single
pure trans crystals and amorphous regions which
may be cis or a mixture of cis and trans.
Near infrared Fourier transform (NIR-FT) Ra-
man spectra were recorded on a Bruker model
IFS 66 spectrometer equipped with a FRA 106
FT-Raman module connected via optical fibers to
a Bruker Ramanmicroscope™. The light source
was a Nd:YAG laser with an excitation wave-
length of 1064 nm. The spectrometer was
equipped with a Ge-detector cooled to liquid ni-
trogen temperature. The Raman shift cut-off at
high frequency is 3500 cm−1, and a narrow band
frequency filter removes the spectral features in
the region from 84 to −100 cm−1. A 40×
objective was used in the microscope, giving an
approximate spot size diameter of 30 mm. The
spectral resolution was around 6 cm−1 for the
apodized spectra. All spectra have been post zero-
filled with a factor of eight. The spectra were
measured on nominally perfect crystals, and on
amorphous areas in the sample.
A Bruker infrared microscope was used in com-
bination with the Bruker IFS 120 HR interferom-
eter at the beamline for infrared spectroscopy at
the MAX-I electron storage ring at MAX-Lab in
Lund, Sweden. The facilities have been described
earlier [10,13], and only the part necessary for an
understanding of this experiment is described
here. The microscope was equipped with an MCT
detector, a NaCl beamsplitter and a Cassegranian
telescope geometry. Experiments could be per-
formed either in transmission or reflection at ap-
proximately normal incidence. The interferometer
beam entered the microscope housing via a 25
2. Experimental
The
isomers
of
dichlorobis(dimethyl-
sulfide)platinum(II) were prepared by the follow-
ing method. About 12 ml of dimethylsulfide

M. Glerup et al. / Spectrochimica Acta Part A 58 (2002) 129–139
131
mm diameter KBr window. By combing the field
lens (15 or 36×magnification) with the 20×ocu-
lar, magnifications of 300 and 720 were realized.
Using the smallest microscope aperture of 0.3 mm
the spatial resolution is nominally 8 mm. At this
resolution, given the microscope’s numerical aper-
ture of 0.5, far field microscopy is diffraction
limited for frequencies less than about 1500
cm−1. At 1000 cm−1 the resolution is limited by
diffraction for spatial resolutions less than 12 mm.
In reflection mode the NaCl beamsplitter prevents
use of the microscope below 700 cm−1, a limit
similar to that due to the MCT detector. A video
camera was used to locate single crystals or amor-
phous regions of the sample which was placed on
a first surface mirror.
or Raman active but not both; species Au and Bu
are IR active and Ag and Bg Raman active.
Calculations are based on the transfer of struc-
tural data and force constants from the building
blocks, the PtCl24− ion and the dimethyl sulfide
molecule, found in the literature. This procedure
was followed in a study of the corresponding
trans palladium complex [15] and our calculations
are performed in a similar way. The program
VIBROT was used for the calculations. Docu-
mentation and source code in C written for
UNIX-like platforms is available by ftp [16] to-
gether with the input files for the present
calculations.
3.1. PtCl24− ions
The far infrared (FIR) Fourier transformed
spectrum, 100–600 cm−1, has been measured on
a Bruker IFS120 HR with a Globar lamp as the
light source and equipped with a deuterium
triglycine sulfate (DTGS) detector working at
room temperature. A 3.5 mm Mylar beam splitter
was used. The interferrogram was zero filled with
a factor of two and no apodization was used. The
spectrum was obtained with 32 scans and a spec-
tral resolution of 2 cm−1. Two mg of sample and
200 mg of polyethylene were pressed in standard
KBr press.
Force constant calculations for this ion were
described by Jones [17], and later reinvestigated
by Goggin et al. [18]. The ion has D4h symmetry
,
with a PtꢀCl bond length of 1.33 A. The mean
atomic weight of Pt, 195.08 U, was used.
The redundancy among the four ClꢀPtꢀCl in-
ternal bending coordinates can be eliminated in
the symmetry coordinates. The four out of plane
coordinates are defined following Wilson, Decius
and Cross [14] as CliCljPt–Clk with cyclic combi-
nation of i, j, k from the cyclic atomic numbering
of the Cl atoms. They include two redundancies
that can be removed in forming the symmetry
coordinates. Alternatively, the calculations can be
performed directly using the valence coordinates,
since the VIBROT program incorporates a proce-
dure for the automatic removal of redundancies
described by Sørensen [19]. The force constants
and frequencies are given in Table 1. The number
of digits given for the force constants does not
reflect their uncertainty, but are included to en-
sure that our calculations can be reproduced. The
frequencies reproduce the assignment of Jones
[17].
The mid infrared (MIR) spectrum, 400−3500
cm−1, has been measured on a Perkin Elmer
1760X spectrometer equipped with a DTGS de-
tector. Three mg of the cis–trans complex were
pressed into a tablet in 200 mg of potassium
bromide. The spectrum was obtained with 32
scans and spectral resolution of 2 cm−1
.
3. Theoretical
The vibrational spectrum was calculated using
the Wilson FG matrix method [14]. The com-
pounds have 57 normal modes. For the cis
configuration the symmetry group is C1 (i.e. no
symmetry); IR and Raman transitions are allowed
for all normal modes. The trans isomer (symmetry
group C2h) has an inversion center. For groups
with such a center the rule of mutual exclusion
states that a given normal mode can be either IR
3.2. Dimethyl sulfide ions
Our calculations were based on the work by
Tranquielle et al. [20]. For the structure with C2v
,
symmetry we have used CꢀH distances 1.093 A,

132
M. Glerup et al. / Spectrochimica Acta Part A 58 (2002) 129–139
Table 1
The force constants obtained by the VIBROT
program shown in Table 2 resulted from a fit to
the observed frequencies of DMS, D3-DMS and
D6-DMS given in [20]. Using the approach of
Tranquille et al. [20], the CH3-groups were treated
in a lower symmetry than C3v. A distinction is
made by the subscript p or o; a p-force constant
involves an internal coordinate that is symmetric
with respect to the plane of the CSC atoms. The
subscript o refers to the corresponding pair of
anti-symmetrized internal coordinates. Atoms H
and H¦ belong to different methyl groups in
equivalent positions.
Force constants and calculated frequencies for [PtCl4]2−
Vibration
Force constanta
Symmetry and
calculated
frequencies, cm−1
f(PtCl)
1.772243
0.074990
0.294338
1.215976
A1g: 328
B1g: 305
B2g: 173
A2u: 147
B2u: 112
Eg: 313, 165
f(PtCl,PtCl%)
f(PtCl,PtCl¦)b
f(ClPtCl%)
f(PtCl, ClPtCl¦) 0.060000
f(ClPtCl%,
Cl%PtCl¦)
f(outo)
0.189535
0.351920
–
a
2
,
,
Units are mdyn per A, mdynA per rad and mdyn per rad,
mdyn=10−8 N.
b ClꢀPtꢀCl¦ colinear.
3.3. The trans complex
The trans complex has C2h symmetry. This
structure is similar to the corresponding Pd com-
plex [15] as shown by a recent X-ray investigation
[5]. Drawings of the structure of the cis and trans
complexes are given in Fig. 1, based on the X-ray
study. To within the uncertainties of that work,
the structure of DMS is very similar in the Pt and
Pd complexes. This means that the CSC planes of
the two DMS ligands are perpendicular to the
plane of PtCl2S2. The PtꢀCl and PtꢀS distances
Table 2
Force constants and calculated frequencies for DMS
Mode
Force
Symmetry and Calculated
Frequencies, cm−1
Constanta
f(CHp)
4.833425 A1: 2997, 2928, 1446,
1339, 1038, 695, 278
f(CHo)
f(CH, CH%)
f(CS)
f(CS, CS%)
f(SCHp)
f(SCHo)
4.762826
0.049455
3.258269
−0.089515 A2: 2979, 1438, 949, 162
,
were found to be 2.297 and 2.309 A, respectively,
0.489094
0.706632
0.067889
and for the PtꢀSꢀC angles we have used the value
106.8° that is the average value for the four
PtꢀSꢀC angles. Finally the ClꢀPtꢀS angles were
set to 92.4 and 87.6° in such a way that the ClꢀC
distances are slightly changed from the values for
90° angles.
f(SCHo, SCH%o)
f(SCHp, SCHo) −0.103401 B1: 2995, 2928, 1438,
1312, 899, 740
f(SCHp, SCH¦p)
f(SCHo, SCH¦o)
f(HCHp)
0.100757
0.005842
0.503122
Most of the force constants were transferred
unchanged from the values found for the Pd-com-
plex [15]. In Table 3 we show the changes and
additions we have introduced in order to fit the
lower frequencies of the Pt-core as compared with
the Pd-compound. Observed and calculated Ra-
man frequencies are given in Table 4; all the
calculated IR and Raman frequencies given in
Table 5 and Table 6.
f(HCHo)
f(CSC)
f(CH,CS)
f(CS, SCHp)
f(CS, SCHo)
f(CH, SCH)
f(tors: CSCH)
0.523348 B2: 2979, 1440, 972, 170
0.949485
−0.344100
0.280018
0.302584
−0.038552
0.015856
a
2
,
,
Units are mdyn per A, mdynA per rad and mdyn per rad.
,
CS distances of 1.802 A, tetrahedral angles for the
3.4. The cis complex
methyl groups and a CSC angle of 98.90°. For the
hydrogens in the CSC-plane the HCSC chains
have a trans configuration.
An X-ray investigation of the structure of the

M. Glerup et al. / Spectrochimica Acta Part A 58 (2002) 129–139
133
PtCl2S2 ‘plane’ whereas the other DMS molecule
has one of the S atoms nearly in the PtCl2S2
‘plane’. We have used the structural parameters
determined by the X-ray investigation [3]. In the
first attempt, the valence force constants of the
trans complex were transferred without any
changes. This gave rise to a Pt-Cl band placed at
375 cm−1. Vibrations from bonds between elec-
tron rich atoms have a high polarisability and are
typically intense in a Raman spectrum. No bands
are observed in the Raman spectrum at wavenum-
bers higher than 360 cm−1, therefore, two force
constants were modified to decrease the vibra-
tional frequency of the PtꢀCl symmetrical stretch
band. The f(PtꢀCl) force constant was decreased
,
to 1.8 mdyn per A which is nearly the initial value
of the f(PtꢀCl) force constant in the PtCl24− ion.
The interaction force constants f(PtCl, PtS),
f(PtCl, PtCl), f(PtS, PtS) were increased to 0.13
mdyn per rad, giving a lower frequency for the
PtꢀCl symmetrical stretch vibration; all calculated
frequencies are given in Table 6.
Fig. 1. Geometries of the cis and trans isomers [5].
4. Results and discussion
cis complex has been presented by Horn et al. [3].
The symmetry is C1 with the PtCl2S2 almost situ-
ated in a plane. For one of the DMS molecules
the SCS plane is almost perpendicular to the
A far infrared spectrum was recorded for a
mixture of the complexes in a polyethylene pellet.
Results are shown for the frequency region 100–
800 cm−1 in Fig. 2, together with two different
Table 3
Force constants fitted for the Ag and Bg frequencies of the trans complexa
Mode
Specie
Initial value
Final value
f(PtCl)
f(PtS)
(PtCl4)
–
–
(PtCl4)
(PtCl4)
(PtCl4)
(DMS)
(DMS)
–
1.772243
–
–
0.294338
1.215976
0.189535
3.258269
0.949485
–
2.044776
1.520539
0.074990
0.294338
1.864919
0.189535
3.074018
1.044587
0.012000
0.351920
f(PtCl, PtS)
f(PtS, PtS%)=f(PtCl, PtCl%)
f(ClPtS)
f(ClPtS, SPtCl%)=f(ClPtS, S%PtCl)
f(CS)
f(CSC)
f(ClPtSC)
f(outo)
(PtCl4)
0.351920
a
2
,
,
Units are mdyn per A, mdynA per rad and mdyn per rad.

134
M. Glerup et al. / Spectrochimica Acta Part A 58 (2002) 129–139
Table 4
the trans peaks observed in the dotted spectrum
are also observed in the solid spectrum, c.f. Fig. 2.
The force field calculation verifies the above
statement. The Pt-S-Cl vibration (237.7cm−1) is
only IR allowed whereas the same vibration for
the cis compound (235.5 cm−1) is IR and Raman
active. In the Raman spectrum (c.f. Fig. 2) we do
not observe any bands at approximately 235 cm−1
for the trans compound (dotted spectrum) but the
spectrum measured on the amorphous region
(solid spectrum) contains a band at 235 cm−1. The
IR spectrum also contains a strong vibrational
band at 235 cm−1, corresponding to the sum of
the PtꢀSꢀC band from the cis and trans complex.
The force field calculation predicts that the trans
compound (dotted spectrum) should not have any
active Raman vibrations in the frequency region
296–318 cm,−1 whereas the cis compound should
have one allowed Raman transition, in accordance
with the experimental results.
The Raman spectrum of the trans compound
was measured on samples with different crystal
orientations. The main difference is the intensity
ratios between the peaks placed at 683 and 725
cm−1 and the peaks placed at 330 cm−1 and 343
cm−1. This is expected since the laser light is
slightly polarized and the crystal orientation
changes from one experiment to the next.
Fig. 3 shows the MIR spectrum for the cis and
trans isomers measured using a KBr pellet. A
Raman spectrum is shown in Fig. 3 as a reference;
all the Raman spectra are alike in this frequency
region, in accordance with the force field calcula-
tions; e.g. Tables 5 and 6.
Raman frequencies in cm−1 for the trans complexa
Assignment
Observed Calculated
Observed
−Calculated
Ag
CHa
CHa
CHs
HCHa
HCHa
SCHs
SCHa
SCHa
CSs
PtSs
PtCls
CSC
2999
?
2919
1419
?
2997.3
2979.5
2927.7
1446.4
1440.3
1338.4
1038.9
978.8
684.3
346.1
327.8
296.4
–
–
–
–
–
–
–
–
1322
1032
989
683.1b
−1.2
−2.1
2.1
−0.5
−0.1
0.1
–
344.0b
329.9b
295.9b
204.1b
169.6b
?
ClPtS
CH3(tors)
PtSC
204.2
169.5
139.6
Bg
Cha
Cha
CHs
HCHa
HCHa
SCHs
SCHa
SCHa
Csa
PtSC
CH3(tors)
Pt-SCC(tors)
2999
2995.4
2979.4
2927.9
1438.7
1437.9
1311.0
951.6
900.1
723.6
205.8
154.8
26.8
–
–
–
–
–
–
–
?
2919
1419
?
1322
951
?
–
724.8b
1.2
−1.7
–
204.1b
?
?
–
a The difference between the observed and the calculated
vibrational frequency is given in the fourth column for the
vibrations used to optimize the fit.
b Included in the fit.
The spatial resolution of the IR-synchrotron
microscopy experiment was higher than in the
Raman experiment, but the frequency range was
limited to the medium infrared region above 700
cm−1 by the beamsplitter and detector in the
microscope. The force field calculations indicate
that the cis and trans isomers have very similar
spectra (Tables 5 and 6) in this frequency region.
The single crystals were seen to have large flat
faces following recrystallization; their size was
from 2 to 60 microns measured using a video
camera attached to the infrared microscope. Spec-
tra were measured of single crystals and amor-
Raman microscopy spectra. The Raman spectra
were recorded on an amorphous region of the
sample, and on a single crystal. Force field calcula-
tions only predict significant differences for fre-
quencies below 400 cm−1 (Tables 5 and 6), in
accordance with the experimental observations.
In Fig. 2 more bands are seen in the solid curve
than in the dotted curve. The lower symmetry of
the cis compound results in more allowed vibra-
tional transitions, leading us to the conclusion that
the single crystal is the trans isomer (dotted spec-
trum). The amorphous region most likely consists
of a mixture of the cis and trans complexes since

M. Glerup et al. / Spectrochimica Acta Part A 58 (2002) 129–139
135
phous regions, and variations are seen from one
sample to the next as shown in Fig. 4.
In Fig. 4a, the CH stretching band of the
dimethyl sulfide ligand is displayed. The contour
of this band changes depending on the crystal.
The changes are similar to those observed in cis-
and trans-Pt(II) bis(glycino) complexes [21]. The
synchrotron is a polarized light source (75% in the
Table 5
Compilation of vibrational frequencies obtained from the force field calculation for trans-Pt(DMS)2Cl2
trans-Pt(DMS)2Cl2
Au
IR
Bu
IR
Ag
Bg
Raman
Raman
CHa
CHa
CHs
2997.3
2979.5
2927.7
1446.4
1440.3
1338.4
1038.8
978.4
684.1
347.4
318.5
290.6
237.7
187.0
158.7
112.2
CHa
CHa
CHs
HCHa
HCHa
SCHs
SCHa
SCHa
Csa
2995.4
2979.4
2927.9
1438.7
1437.9
1311.0
951.6
900.6
726.9
227.8
158.2
132.2
81.7
CHa
CHa
CHs
2997.3
2979.5
2927.7
1446.4
1440.3
1338.4
1038.9
978.8
684.3
346.1
327.8
296.4
CHa
CHa
CHs
HCHa
HCHa
SCHs
SCHa
SCHa
CSa
PtSC
CH3(tors)
Pt-SCC(tors)
–
2995.4
2979.4
2927.9
1438.7
1437.9
1311.0
951.6
900.1
723.6
205.8
154.8
26.8
HCHa
HCHa
SCHs
SCHa
SCHa
CSs
PtCla
CSC
PtSC
HCHa
HCHa
SCHs
SCHa
SCHa
CSs
PtSs
PtCls
CSC
PtSC
CH3(tors)
Pt-op
ClS-op
Pt-SCC(tors)
–
PtSa
ClPtS
CH3(tors)
PtSC
204.2
169.5
139.6
–
–
–
–
ClPtS
CH3(tors)
ClPtS
20.4
–
–
–
–
–
–
Table 6
Compilation of vibrational frequencies obtained from the force field calculation for cis-Pt(DMS)2Cl2
cis-Pt(DMS)2Cl2
CHa1
CHa2
Cha%1
CHa%2
CHa%2
CHa%1
CHa2
2997.5
2997.4
2995.5
2995.4
2979.5
2979.5
2979.4
2979.4
2927.9
2927.9
2927.8
2927.7
1446.4
1446.3
1440.2
1440.2
1438.6
1438.6
1437.8
HCHa%1
SCHs1
SCHs2
SCHs%1
SCHs%2
SCHa1
SCHa2
SCHa%1
SCHa%2
SCHa1
SCHa2
SCHa%
SCHa%2
CSa1
1437.8
1338.6
1338.5
1311.3
1311.2
1040.3
1039.9
979.8
979.4
952.4
952.2
902.4
901.1
729.0
726.1
684.1
682.9
351.3
348.7
PtSCw2
PtCla
CSC2
CSC1
PtSCr1
ClPtS
PtSCr2
PtSa
PtSCw1
CH3-1(tors)
CH3%2(tors)
CHs-2(tors)
CH3%1(tors)
ClPtS
Pt-op
ClPtCl
ClS-op
Pt-SCC1(tors)
Pt-SCC2(tors)
322.9
301.3
296.3
275.6
235.5
224.3
217.4
206.0
195.2
162.9
159.2
157.8
153.9
137.9
117.2
113.5
95.9
CHa1
CHs%2
CHs%1
CHs1
CHs2
HCHa2
HCHa1
HCHa%2
HCHa%1
HCHa2
HCHa1
HCHa%2
CSa2
CSs2
CSs1
PtCls
23.5
23.0
PtSs

136
M. Glerup et al. / Spectrochimica Acta Part A 58 (2002) 129–139
Fig. 2. FIR spectrum of the isomeric compound Pt(DMS)2Cl2 pressed in polyethylene, shown as a transmittance spectrum (thick
curve). The dotted curve is the NIR-FT-Raman spectrum of the trans-Pt(DMS)2Cl2 compound (single crystal) and the thin curve
is the spectrum of cis–trans-Pt(DMS)2Cl2 (amorphous region).
Fig. 3. The MIR transmission spectrum of cis–trans-Pt(DMS)2Cl2 pressed as a KBr pellet, shown as a transmittance spectrum (thin
curve). The thick curve is the NIR-FT-Raman spectrum; only one spectrum is shown since in this frequency region it is not possible
to differentiate between the two isomers.

M. Glerup et al. / Spectrochimica Acta Part A 58 (2002) 129–139
137
sity variations cannot be assigned as arising only
from varying concentrations of cis or trans isomers.
Fig. 4(b) is from 1280 to 1330 cm−1, showing a
clear change in peak location from 1313 to 1319
cm−1, and from 1294 to 1296 cm−1. Fig. 4(c)
shows changes in the frequency region of 910–1070
cm−1. The force field fitting process was based on
the low frequency bands (Table 4), therefore, the
possibility of assigning bands in the mid infrared
region is limited.
plane of the storage ring), and it is possible that
crystal orientation may also play a role in the
intensity variations seen in Fig. 2. Thus the inten-
It is remarkable that the frequency of the CH
stretching vibrations 2919 and 2999 cm−1, shown
in Figs. 3 and 4 and Tables 4–6 are high compared
with the typical sp3 CH stretching frequency. We
cannot give a fundamental explanation for the
band position changes in the infrared microscopy
spectra. The bands are not due to artifacts since
they are also observed in the MIR spectrum of the
sample measured as a powder.
Despite the analysis of many samples and re-
peated searches, pure cis crystals were not observed
in our experiment. Perhaps the trans compound
forms crystals more readily due to the effect of the
center of inversion on its charge distribution.
5. Conclusion
The Raman data show that the low frequency
region of the spectrum is especially useful for
assigning the vibrational modes of transition metal
complexes, since this is the spectral region corre-
sponding to ligand-metal vibration. At the beam-
line for infrared microscopy being planned for the
MAX-III storage ring, currently under construc-
tion in Lund Sweden, we will have the special
detectors and a modified microscope for investigat-
ing this key spectral region. As opposed to black-
body light sources, the intensity of light from the
storage ring increases with decreasing wavenumber,
presenting a significant advantage for diffraction
limited microscopy, even at wavelengths/diffrac-
tion limited spot sizes extending through the far
infrared to the millimeter wave region. This will
give us the opportunity to expand this study and
make more direct correlations to the low frequency
Raman spectra.
Fig. 4. Spectra of two crystals of the isomeric compound
Pt(DMS)Cl2. The same crystal is on top in each figure. (a)
displays the frequency region 2950–3050 cm−1, showing how
the contour of the CH stretching band of the dimethylsulfide
ligand changes depending on the crystal, (b) is from 1280–
1330 cm−1, showing a clear change in peak location from
1313 to 1319 cm−1, and from 1294 to 1296 cm−1, (c) shows
changes in the frequency region of 910–1070 cm−1
.

138
M. Glerup et al. / Spectrochimica Acta Part A 58 (2002) 129–139
In this study we were able to distinguish be-
isomer, we can conclude that a cis DMS influence
plus a trans Cl− influence has about the same
effect as a cis Cl− plus a trans DMS. This is
consistent with the result of the X-ray studies
indicating that DMS has both a greater cis and a
greater trans influence on the PtꢀS bond than
Cl−.
tween the cis and the trans isomers due to symme-
try, and because of the force field calculation we
could verify our initial conclusion and assign
some of the vibrational bands.
X-ray crystallography can resolve bond lengths
to less than 1 in 102, while vibrational spec-
troscopy resolves frequency to 1 in 103 or 104. It is
worthwhile to compare these two techniques in
their ability to examine subtle bonding effects.
With reference to Fig. 1, we can now analyze
the cis and trans ligand influence. In the cis com-
pound the PtꢀCl bond is trans to DMS and cis to
both Cl− and DMS. In the trans compound the
same bond is trans to Cl− and cis to two DMS
molecules. The fact that we had to lower the force
constant for the PtꢀCl stretch vibration for the cis
compound to get a better match to the observed
data implies that the Cl− ion is more strongly
bound in the trans complex. Since it is known that
the cis influence has little effect on the PtꢀCl bond
length, the results suggest that the trans influence
of chloride is higher than that of DMS. Evaluat-
ing the trans influence using the magnitude of the
force constants has been discussed by T.G. Apple-
ton et al. and L.I. Elding et al. [6,7]. The present
work confirms that it is a valuable tool.
Acknowledgements
One of the authors (M. Glerup) would like to
thank the Danish Research Academy for general
funding. The authors are grateful to Haldor
Topsøe A/S and to the Danish Natural Research
Council for funding the NIR-FT-Raman instru-
ment within the material science program. The
,
authors would like to thank Ake Oskarsson of the
Inorganic Chemistry Department, Lund Univer-
sity for synthesizing the compounds used in this
study. Daniel Christensen of the Department of
Chemistry, University of Copenhagen provided
valuable assistance with recording the Raman and
infrared spectra. The experiments would not have
been possible without the assistance of Bengt
Nelander and the staff of the MAX-Lab storage
ring in Lund Sweden. In addition, we gratefully
acknowledge Professor Otto Schrems of the Al-
fred Wegener Institute in Bremerhaven Germany
for lending us an infrared microscope, and Ole
Faurskov Nielsen of the University of Copen-
hagen for valuable discussions.
An analysis of the geometry of 22 Pt(II) com-
plexes obtained by X-ray crystallography was not
able to differentiate between the trans effect of
DMS and Cl− since the differences in bond length
,
(1% of a bond distance of ca 2.3 A) are compara-
ble to the errors inherent to the technique. The
techniques used in this paper are able to identify
the particular eigen-frequency involved with a
given vibration and although more work is neces-
sary, the results show that vibrational spec-
troscopy is inherently a sensitive indicator of
subtle variations in the chemical bonding of these
systems.
References
[1] F.A. Cotton, G. Wilkinson, Advanced Inorganic Chem-
istry, Fifth ed., Wiley, New York, 1988.
[2] K. Lo¨vqvist, A Crystallographic Study of Platinum(II)
Complexes, Ph.D. Thesis, University of Lund, Sweden,
1996, and references therein.
The work of Løvqvist and Giveen in the group
of Oskarsson [2,5] indicates that the cis influence
is more important for PtꢀS bonds than PtꢀCl
bonds. From X-ray crystallography it is known
that sulfur has a greater cis influence on the PtꢀS
bond than chloride, and that S also has a greater
trans influence on the PtꢀS bond than chlorine.
Since the DMS ligand has a cis Cl− in each
[3] G.W. Horn, R. Kumar, A.W. Maverick, F.R. Fronczek,
S.F. Watks, Acta Cryst. C46 (1990) 135.
[4] E.G. Cox, H. Saenger and W. Wardlaw, J. Chem. Soc.
(1934) 182.
[5] D. Giveen, Packing characteristics of centrosymmetric Pt-
and Pd- complexes; The crystal structures of trans-
dichlorobis(dimethylsulfide) platinum(II) and trans-
dichlorobis(methylphenylsulfide) palladium(II), M.Sc.
Thesis, University of Lund, Sweden, 2000.

M. Glerup et al. / Spectrochimica Acta Part A 58 (2002) 129–139
139
[6] T.G. Appleton, H.C. Clark, L.E. Manzer, Coord. Chem.
Rev. 10 (1973) 335.
[13] M.S. Johnson, B. Nelander, Il Nuovo Cimento 20D
(1998) 449.
8
[7] L.I. Elding, O. Gro¨ning, Inorg. Chem. 17 (1978) 1872.
[14] E.B. Wilson, J.C. Decius, P.C. Cross, Molecular Vibra-
tions, McGraw Hill, New York, 1955.
[8] H.O. Desseyn, S.P. Perlepes, A.C. Fabretti, App. Spect.
45 (1991) 131.
[15] M. Tranquielle, M.T. Forel, J. Mol. Struct. 25 (1974) 413.
[16] G.O. Sørensen, anonymous@kl5axp.ki.ku.dk, 2001.
[17] L.H. Jones, Coord. Chem. Rev. 1 (1966) 351.
[18] P.L. Goggin, J. Mink, J. Chem. Soc. (Dalton) (1974)
1479.
[19] G.O. Sørensen, J. Mol. Spec. 35 (1970) 489.
[20] M. Tranquielle, P. Labarbe, M. Fouassier, M.T. Forel, J.
Mol. Struct. 8 (1971) 273.
[9] G.P. Williams, P. Dumas, Infrared Microspectroscopy
with Synchrotron Radiation, Accelerator-Based Infrared
Sources and Applications, Proceedings of SPIE 3153,
1997.
[10] M.S. Johnson, P. Beichert, O. Schrems, B. Nelander,
Asian Chem. Lett. 4 (2000) 45.
[11] T.R. Cundari, W. Fu, E.W. Moody, L.L. Slavin, L.A.
Snyder, S.O. Sommerer, T.R. Klinckman, J. Phys. Chem.
100 (46) (1996) 18057.
[12] L. Chong-De, L.-J. Jiang, W.-X. Tang, Spectrochim. Acta
49A (3) (1993) 339.
[21] K. Nakamoto, Infrared and Raman Spectra of Inorganic
and Coordination Compounds, Fourth ed., Wiley, New
York, 1986, p. 235.