A matrix isolation and DFT study of the generation and characterization of monomeric vapour phase platinum chlorides

Article abstract of DOI:10.1016/S0009-2614(01)01369-0

Molecular platinum monochloride (PtCl) and platinum dichloride (PtCl₂) have been prepared from platinum atoms and chlorine doped argon in a hollow-cathode sputtering device, matrix isolated in solid argon and characterized using electronic, infrared and X-ray absorption spectroscopies together with high level DFT calculations.

Full text of DOI:10.1016/S0009-2614(01)01369-0

10 January 2002
Chemical Physics Letters 351 (2002) 319–326
www.elsevier.com/locate/cplett
A matrix isolation and DFT study of the generation
and characterization of monomeric vapour phase
platinum chlorides
a,1
a
Adam J. Bridgeman , German Cavigliasso a,b, Neil Harris ,
ꢀ
Nigel A. Young
a,
*
a
Department of Chemistry, The University of Hull, Hull HU6 7RX, UK
Department of Chemistry, The University of Cambridge, Cambridge CB21EW, UK
b
Received 20 September 2001; in final form 14 November 2001
Abstract
Molecular platinum monochloride (PtCl) and platinum dichloride (PtCl2) have been prepared from platinum atoms
and chlorine doped argon in a hollow-cathode sputtering device, matrix isolated in solid argon and characterized using
electronic, infrared and X-ray absorption spectroscopies together with high level DFT calculations. Ó 2002 Elsevier
Science B.V. All rights reserved.
€
1. Introduction
chlorine has also been studied by Schafer and co-
workers [5,6] and at temperatures below 650 °C
the principal platinum containing vapour species is
Pt6Cl12, whilst at higher temperatures, PtCl3 (ca.
700 °C) and then PtCl2 ð700–1000 °CÞ become the
dominant platinum vapour compounds. Lands-
berg and Schaller [7] have also identified the
presence of Pt3Cl3 in the reaction of platinum and
chlorine from vapour pressure measurements.
Makarov et al. [8] in recent mass spectrometric
The information in the literature concerning the
vapour phase composition above heated platinum
chlorides and platinum heated in the presence of
chlorine appears confusing and contradictory.
€
Schafer and co-workers [1–3] first identified
Pt6Cl12 mass spectrometrically in the vapour above
heated platinum(II) chloride and confirmed, using
equilibrium measurements, that Pt6Cl12 is the only
Pt–Cl vapour species above heated b-PtCl2 up to
ca. 600 °C[4]. The reaction of platinum with
€
experiments observed the reverse to Schafer and
reported that when hydrated H2PtCl6 was heated
to only 100–175 °C, PtCl4 was the major platinum
vapour species, and that between 175 and 250 °C
this was replaced by PtCl2. At these temperatures
they noted that the vapour pressure of Pt6Cl12
would be so low as to be unobservable in their
experiments. In all cases Cl2 is also found to be a
* Corresponding author. Fax: +44-1482-466410.
E-mail addresses: a.j.bridgeman@hull.ac.uk (A.J. Bridg-
eman), n.a.young@hull.ac.uk (N.A. Young).
1
Also corresponding author.
0009-2614/02/$ - see front matter Ó 2002 Elsevier Science B.V. All rights reserved.
PII: S0009-2614(01)01369-0

320
A.J. Bridgeman et al. / Chemical Physics Letters 351 (2002) 319–326
major component of the vapour phase due to de-
composition, and Makarov et al. [8] calculate that
at the temperatures of their study the vapour
pressure of Cl2 is about 30 times that of PtCl2.
There appear to be no spectroscopic investigations
of the vapour above heated platinum chlorides.
Due to the uncertainty of the vapour composi-
tion above the heated platinum chlorides we have
prepared monomeric platinum chlorides using the
reaction of chlorine with platinum atoms trapped
in argon matrices. Using this approach and in
conjunction with high-level DFT calculations we
have generated and characterized monomeric PtCl
and PtCl2 for the first time.
CaF2 windows were used. The Pt L3-edge X-ray
absorption spectra were collected in fluorescence
mode (Canberra 13 element SSD detector) on
station 9.2 of the Daresbury Laboratory SRS
(operating at 2 GeV with circulating currents of ca.
200 mA) using a Si(2 2 0) double crystal mono-
chromator detuned by ca. 50% to remove har-
monic contamination. Fourteen spectra were
collected for each matrix sample, averaged and
calibrated using the first maximum in the first de-
rivative of the Pt L3-edge of Pt foil (11 564 eV [11]).
Background subtraction was carried out using
P
AXAS [12] by fitting the pre-edge region to a
quadratic polynomial, subtracting this from the
data and approximating the atomic component of
the post-edge region with a high (typically sixth)
order polynomial. This approximation was opti-
mized in order to minimize the low-r features in
the Fourier transforms by an iterative process,
although it should be noted that atomic XAFS
features may be expected in this part of the FT
[13]. Fitting of the experimental data was carried
out with EXCURV98 [14] making use of multiple
scattering curved wave theory, a von-Barth ground
state and a Hedin–Lundqvist exchange potential.
Density functional (DF) calculations were car-
ried out with the ADF program (ADF2000.02)
[15] using the VWN form of the LDA and the
BP86 gradient corrections. STO basis sets of triple-
f quality incorporating frozen cores up to 2p for
chlorine and up to 4f for Pt and the ZORA rela-
tivistic approach [15] were utilized. Single point
calculations were also performed with ZORA plus
spin–orbit coupling. SVFF calculations were car-
ried out using the Wilson GF method within
SOTONVIB [16].
2. Experimental and computational methods
The platinum atoms were generated in a hol-
low-cathode sputtering device based on the design
of Green and Reedy [9] which has been described
in detail previously [10]. The platinum hollow
cathode (platinum foil (Goodfellows, 99.99% 25
lm) inside a hollow copper screw) was mounted in
a water cooled and electrically isolated metal
flange to which the glass vacuum jacket containing
the anode and gas inlet system were connected.
The platinum anode was mounted on an adjust-
able drive (modified Young’s PTFE tap) and held
centrally in the platinum cathode by a thin silica
tube. The flow rate of the discharge gas through
the sputtering device was controlled by PTFE
capillary valves (Young’s) and was sufficient for it
to act as the matrix gas as well. The gas mixtures
were prepared using standard manometric proce-
dures. The sputtering device was operated at ca.
600–700 V and 15 mA by an Ion-Tech DCpower
supply.
IR spectra were recorded using a Bruker IFS66
FTIR spectrometer equipped with KBr (MIR) and
6 lm Mylar (FIR) beamsplitters, and DTGS de-
tectors. The vacuum shroud used CsI windows and
had a base pressure of ca. 2 Â 10À7 mbar before
the APD DE204SL Displex closed cycle helium
cryostat, (base temperature of ca. 10 K) was
turned on. The UV–Vis–NIR spectra were re-
corded using a Varian Cary 5E spectrometer, with
a similar vacuum and pumping regime, except that
3. Results and discussion
The electronic absorption spectrum of the
products in an argon matrix obtained when Pt was
sputtered with neat argon is shown in Fig. 1a. The
spectral features are in very good agreement with
those for Pt atoms isolated in argon matrices
[17,18], with no evidence for the formation of di-
mers or higher order multimers [19] thus indicating
that the hollow cathode device is an excellent

A.J. Bridgeman et al. / Chemical Physics Letters 351 (2002) 319–326
321
the intense band at 27000 cmÀ1 displayed vibra-
tional fine structure with a spacing of ca.
320 cmÀ1, and on the weak band at 19150 cmÀ1
there is the indication of vibrational fine structure
with a similar interval. In addition to these three
bands assigned to charge-transfer transitions in the
major species there is a broad band at 16040 cmÀ1
which appears to consist of two vibrational pro-
gressions of ca. 260 cmÀ1, separated by ca.
350 cmÀ1, which are assigned to a minor PtCln
species. In addition, there are some very sharp and
weak features at 10 850, 10 640, 10 450 and
8215 cmÀ1 that cannot be assigned with certainty.
Therefore, the electronic absorption spectra indi-
cate the formation of one species using 5% Cl2=Ar,
and the formation of this as well as a second,
minor species with 1% Cl2=Ar discharge gas.
(a)
(b)
The matrix infrared spectrum (Fig. 2a) obtained
when the platinum was sputtered with 5% Cl2=Ar
contained one broad band at 430 cmÀ1 with no
(c)
Fig. 1. Electronic absorption spectra of the products matrix
isolated from a Pt hollow-cathode sputtered with (a) neat ar-
gon, (b) argon doped with 1% Cl2, and (c) argon doped with 5%
Cl2 (spectra offset for clarity).
source for matrix isolated atoms. When the argon
discharge gas was doped with 1% Cl2 the spectra
(Fig. 1b) contained new features, some of which
displayed vibrational fine structure, in addition to
those of the Pt atoms or Cl2 at 30 500 cmÀ1. When
the argon was doped with 5% Cl2 (Fig. 1c) there
were fewer new bands in addition to those of Cl2
and only negligible signs of any Pt atoms, indi-
cating the formation of fewer PtCln species than
for the 1% Cl2 doped case. The two bands at ca.
27 000 and 29000 cmÀ1 have the same relative in-
tensity in the spectra using both 1% and 5% Cl2
doped Ar and are assigned to the major PtCln
species formed in both cases. In addition to these
two relatively intense bands there is also a weak
feature at ca. 19150 cmÀ1 in both sets of spectra,
which can also be assigned to this major species. In
the spectra from the 1% Cl2=Ar discharge mixture
Fig. 2. Infrared spectra of the products matrix isolated from a
Pt hollow-cathode sputtered with (a) argon doped with 5% Cl2,
(b) argon doped with 0.5 % Cl2. SVFF calculated spectra for (c)
linear PtCl2 and (d) PtCl.

322
A.J. Bridgeman et al. / Chemical Physics Letters 351 (2002) 319–326
Table 1
Observed and SVFF calculated frequencies (cmÀ1) for PtCl2 and PtCl
Observed
Calculated
180°
150°
130°
m3
m1
m3
m1
m3
m1
m3
m1
Pt35Cl2
Pt35Cl37Cl
Pt37Cl2
430.4
426.8
421.7
–
–
–
430.4
426.8
421.8
400.0
393.8
389.0
430.4
426.8
421.6
403.0
396.8
392.2
430.4
426.8
421.3
406.5
400.3
396.0
Pt35Cl
Pt37Cl
399.1
390.2
399.1
389.9
195Pt used for all calculations.
ꢁÀ1
ꢁÀ1
.
ꢁÀ1
ꢁÀ1
ꢁÀ1
PtCl2: 180° fr ¼ 3:053 mdyne A , frr ¼ 0:2441 mdyne A ; 150° fr ¼ 3:064 mdyne A , frr ¼ 0:2043 mdyne A ; 130° fr ¼ 3:074
ꢁÀ1
mdyne A , frr ¼ 0:1260 mdÀy1ne A
ꢁ
PtCl: fr ¼ 2:783 mdyne A
.
resolvable isotopic fine structure. With a more
dilute discharge mixture (0.5% Cl2=Ar), the band
at 430 cmÀ1 was resolved into a 9:6:1 triplet (Fig.
2b) characteristic of a vibrational mode of a di-
chloride. The isotopic structure from the Pt iso-
topes was not resolved, as to be expected. Using
the outermost isotopic bands (430.4 and
421:7 cmÀ1) in a SVFF calculation gives a bond
angle of 163° for a PtCl2 molecule which, due to
the insensitivity of the sine function near 90°, can
be regarded as indicative of a linear geometry [20].
Although the position of the central component at
426:8 cmÀ1 arising from the Pt35Cl37Cl isotopomer
allows for an estimate of 393:8 cmÀ1 to be made
for the position of the symmetric stretching mode,
this will be too weak to be observed in the spectra.
The data from these calculations are given in Table
1, and the SVFF calculated spectrum for linear
PtCl2 is shown in Fig. 2c. Therefore, the bands at
430 cmÀ1 are assigned to the asymmetric stretch-
ing mode of linear PtCl2. In addition to this set of
bands, weaker features were also present in the
spectra. Of these, the two at 399.1 and 390:2 cmÀ1
have a frequency and intensity ratio characteristic
of a vibrational mode of a monochloride. The
SVFF calculated spectrum of PtCl is shown in Fig.
2d and confirms the assignment. Even weaker
features at 408.9 and 406:0 cmÀ1 are most likely
due to a matrix site effect of the bands at 430 cmÀ1
as is often observed for metal dichlorides in argon
matrices [20].
Table 2 lists the DFT calculated frequencies and
IR intensities for PtCl, PtCl2, PtCl3, PtCl4, and cis
and trans-PtCl2ðNH3Þ complexes. For the am-
2
Table 2
DFT calculated 195Pt–35Cl frequencies (cmÀ1) and IR intensities (km molÀ1) for PtCl, PtCl2, PtCl3, PtCl4 and cis and trans-
PtCl2ðNH3Þ
2
ms
mas
d
PtCl
2P
C
C2v
386 (12)
399 (5)
–
1v
PtCl2 (bent)
PtCl2 (linear)
PtCl3
1A1
3RÀg
2B2
3A2g
415 (27)
419 (57)
382 (26)
368 (290) eu
58 (2)
79 (0)
D
1h
C2v
389 (0)
307 (0), 341 (18)
359 (0) a1g, 333 (0) b1g
60 (1) b1, 91 (3) b2, 90 (1) b2
36 (0) b2u, 104 (1) a2u, 166 (0)
b2g; 167ð0Þ
PtCl4
D4h
a
cis-PtCl2ðNH3Þ2
1A1
1Ag
C2v
C2h
344 (24)
317 (0)
331 (24)
330 (52)
146 (0)
a
trans-PtCl2ðNH3Þ2
131 (11)
a The experimental Pt–Cl frequencies are 326 ðms) and 318 ðmasÞ cmÀ1 for cis-PtCl2ðNH3Þ and 327 ðmsÞ cmÀ1 for trans-PtCl2ðNH3Þ
2
2
[21,22,30].

A.J. Bridgeman et al. / Chemical Physics Letters 351 (2002) 319–326
323
mine complexes, the agreement with experimental
IR and Raman frequencies [21,22,30] is very good,
suggesting that the results for the simpler halide
molecules should be reliable. The experimental
band at 430:4 cmÀ1 is consistent with the values
À
3
1
for PtCl2 in either the Rg or A1 state. However,
the position of the DFT calculated frequencies for
the symmetric modes are in much better agreement
with the linear rather than bent geometry for
PtCl2. The calculated frequency for PtCl at
33 cmÀ1 lower than that for mas of PtCl2, together
with a consistent offset between observed and
calculated values, confirms the assignment of the
doublet at 399.1 and 390:2 cmÀ1 to PtCl. The
values for PtCl3 and PtCl4 appear to be too low
with PtCl3 expected to exhibit two bands. The
experimental and calculated vibrational spectra
thus indicate that PtCl2 is formed exclusively in a
5% Cl2=Ar discharge whereas lower chlorine con-
centrations results in both PtCl and PtCl2 being
formed, in good agreement with the electronic
absorption data.
X-ray absorption fine structure (XAFS) spec-
troscopy provides local structural information
around the platinum. A 5% Cl2=Ar discharge gas
was used to ensure only one PtCln species was
formed and to minimize the proportion of unre-
acted Pt atoms as XAFS provides an average en-
vironment of all the Pt species present. Fig. 3
shows the XANES data for K2PtCl6 and K2PtCl4
diluted in BN at 25 °Cas well as the matrix iso-
lated sputtered products using both neat Ar and
5% Cl2 doped Ar. Although the edge position and
edge shifts of XANES data are often useful in
oxidation state determination, this is less useful for
L3-edge data because the edge region is dominated
by the intense ‘white line’ arising from 2p65dn to
2p21=22p33=25dnþ1 transitions. The position of this
white line will be affected by the change in the
energies of the 2p core states and the 5d valence
orbitals in the different coordination geometries as
much as the change in the formal oxidation state.
In the past we have noted that the Ni K-edge
position for linear nickel dihalides in cryogenic
matrices is higher than that for either square pla-
nar, tetrahedral or octahedral model compounds
[23]. However, the intensity of the white line is very
informative as it is related to the number of holes
Fig. 3. Pt L3 edge XANES spectra for (a) K2PtCl6 diluted in
BN at 298 K, (b) K2PtCl4 diluted in BN at 298 K, (c) products
matrix isolated from a Pt hollow-cathode sputtered with 5% Cl2
doped Ar and (d) products matrix isolated from a Pt hollow-
cathode sputtered with neat Ar.
in the 5d valence orbitals and hence oxidation state
of the platinum [24], and the spectra of K2PtCl6
and K2PtCl4 (Fig. 3a,b) clearly show that the in-
tensity of the white line for K2PtCl4 is less than
that of K2PtCl6. The intensity of the white line in
the spectrum of the products when Pt was sput-
tered with 5% Cl2 (Fig. 3c) is very similar to that of
K2PtCl4, confirming the formation of a PtII com-
pound. The lack of any significant Pt atom con-
centration is shown by comparison with the
spectrum obtained from the products sputtered
with neat Ar (Fig. 3d).
The EXAFS part of the spectrum can be used to
obtain inter- and intra-molecular distances from
the Pt. In order to confirm the accuracy of the Pt–
Cl distances, Pt L3-edge data were also collected
for K2PtCl4 and K2PtCl6 which yielded Pt–Cl
ꢁ
distances of 2.31(2) A for K2PtCl4 and 2.32(2) for
K2PtCl6 in excellent agreement with a very recent
EXAFS study [25] as well as single crystal X-ray

324
A.J. Bridgeman et al. / Chemical Physics Letters 351 (2002) 319–326
Fig. 4. Pt L3-edge EXAFS (a) and Fourier transform (b) of the products matrix isolated from a Pt hollow-cathode sputtered with 5%
Cl2 doped Ar. Solid line, experimental data, dotted line, curved wave multiple scattering theory for linear PtCl2.
ꢁ
ꢁ
data [26,27]. The Pt L3-edge data of the products
when Pt was sputtered with 5% Cl2=Ar are shown
in Fig. 4 and Table 3. The data is fairly noisy due
to the limited Ar penetration depth at these ener-
gies (ca. 60 lm), which meant that it had to be
interaction of 2.19(2) A. This is 0.12 A shorter
than that observed in either K2PtCl4 or K2PtCl6,
and confirms that the spectrum is indeed arising
from a low coordination number compound, as we
have observed a similar shortening of metal–
halogen bond lengths in matrix isolated NiBr2 [23].
When the Cl occupation number was included in
ꢁÀ1
truncated at ca. 13 A . The FT of the EXAFS
data reveals one significant peak due to a Pt–Cl
Table 3
Refined XAFS parametersa for K2PtCl4, K2PtCl6 and matrix isolated products when Pt cathode is sputtered with 5% Cl2=Ar
2r2
Ef
(eV)
FIh
Ri
e
f
g
rPt–Cl
ꢁ
(A)
ꢁ2
ðA Þ
K2PtCl4 (298 K)
X-ray diffractionb
Pt L3-edge XAFSc
2.309
2.309(3)
0.0068(3)
)10.8
2.72
25.80
KPtCl6 (298 K)
X-ray diffractiond
Pt L3-edge XAFSc
2.311
2.322(4)
0.0061(4)
)11.3(8)
)7.6(2)
4.73
17.3
31.88
58.07
Pt products sputtered with 5% Cl/Ar (10 K)
Pt L3-edge XAFS 2.193(97)
0.0059(12)
a Standard deviation in parentheses.
b Ref. [26].
c This work.
d Ref. [27].
e Estimated systematic errors in XAFS bond lengths are Æ1:5% for well-defined co-ordination shells.
f 2r2 is the Debye–Waller factor.
g Ef is a single refined parameter to reflect differences in the theoretical and experimental Fermi levels.
ÂÀ
Á
Ã2
R
P
 R
h FI ¼
vTi À vEi ki3
:
i
Ã
i R ¼
jvT À vEjk3 dk= jvEjk3 dk  100%:

A.J. Bridgeman et al. / Chemical Physics Letters 351 (2002) 319–326
325
the refinement, a value of 2.3 was obtained. Whilst
this is a little large for PtCl2, coordination num-
bers are notoriously hard to determine accurately
using EXAFS, especially for noisy data. Features
in the FTs at approximately twice the Pt–Cl dis-
tance would confirm a linear geometry [23,25,28],
and the multiple scattering pathways giving rise to
these are included in the final refinement shown in
Fig. 4. Whilst there is the probability of such a
feature in the data, the inherent noise makes it very
hard to identify unambiguously.
The DFT calculated structures for PtCl, PtCl2,
PtCl3, PtCl4 and cis and trans-PtCl2ðNH3Þ2 are
given in Table 4. The BP86 results for the ammine
complexes show extremely good agreement with
crystallographically determined geometries [29],
with a slight improvement upon recently reported
calculations [30], suggesting again that the com-
putational method is reliable for these heavy metal
complexes. The calculated bond lengths for PtCln
show an increase with oxidation state and coor-
dination number. The best agreement with the
EXAFS values is for linear PtCl2, although the
values for bent PtCl2 and PtCl are very similar.
The bond lengths in the T-shaped PtCl3 and
square planar PtCl4 molecules are somewhat
longer.
geometry for PdCl2. Wesendrup and Schwerdtfe-
ger [32] predict a linear geometry for PtF2. The
XANES and EXAFS data and the DFT structures
and energies are thus consistent with the vibra-
tional and electronic absorption data and suggest
formation of linear PtCl2 when Pt foil is sputtered
with a 5% Cl2=Ar gas mixture.
The linear d8PtCl2 molecule is predicted to have
3RÀg ground state for the dg4rg2p2g configuration
identical to that predicted for NiCl2 [33]. The bent
molecule has a 1A1 ground state which can be
thought of as arising from a Renner–Teller split-
4
1
1
ting of the low-spin Pg state of dgr2gpg2. The A1
state has double occupation of the resulting a2
orbital with the b2 component of the pg level
2
empty. The ða2Þ configuration greatly reduces the
out-of-plane p bonding but the in-plane bonding is
enhanced through d–p hybridization. Through
mixing with Pt 6py, the dyz orbital becomes
strongly directed towards the ligands so that the
anti-bonding b2 orbital is destabilized. The bond-
ing is thus increased even though the out-of-plane
component is reduced. The square planar geome-
try of PtCl24À leads to a single strongly anti-bond-
ing d-orbital and hence to a large LFSE for the
low spin d8 configuration. The bent PtCl2 geome-
try can be considered as one half of a square pla-
nar molecule and similarly leads to a single
strongly anti-bonding d-orbital and high LFSE.
The preference for the linear or bent geometry thus
rides on the familiar balance between the electron
pairing energy and the bonding.
Linear PtCl2 is predicted to be 59 kJ molÀ1
more stable at the LDA level and 64 kJ molÀ1 at
the BP86 level than the bent form. Inclusion of
spin–orbit coupling in the ZORA calculations de-
creases these values somewhat to 37 and
34 kJ molÀ1, respectively. Siegbahn [31] predicts a
linear geometry for PdF2 and a highly bent (98°)
Although experimental evidence shows the iso-
lation of PtCl and PtCl2, it is noteworthy that
Table 4
DFT calculated structures for PtCl, PtCl2, PtCl3, PtCl4 and cis and trans-PtCl2ðNH3Þ2
ꢁ
Pt–Cl bond lengths (A)
Symmetry
Bond angle (°)
LDA
BP86
LDA
BP86
PtCl
2P
C
C2v
2.151
2.144
2.148
2.181
2.183
2.187
–
–
1v
PtCl2
PtCl2
PtCl3
1A1
3RÀg
2B2
130
180
100 (Â2)
160
90
130
180
101 (Â2)
157
90
D
1h
C2v
2.228 (Â2)
2.216
2.266
2.238 (Â2)
2.221
2.275
PtCl4
cis-PtCl2ðNH3Þ2
3A2g
1A1
1Ag
D4h
C2v
C2h
a
2.273
2.292
2.315
2.337
96
178
96
178
a
trans-PtCl2ðNH3Þ
2
a The experimental Pt–Cl bond lengths of cis and trans-PtCl2ðNH3Þ are 2.33 and 2.32 A, respectively [29].
ꢁ
2

326
A.J. Bridgeman et al. / Chemical Physics Letters 351 (2002) 319–326
€
[3] H. Schafer, U. Wiese, H. Rabeneck, Z. Anorg. Allg. Chem.
513 (1984) 157.
PtCl3 and PtCl4 are actually calculated to be more
stable than PtCl2. Decomposition of PtCl3 to
PtCl2 þ 1=2Cl2 and PtCl4 to PtCl2 þ Cl2 are pre-
dicted to require 55 and 150 kJ molÀ1, respec-
tively. Decomposition of PtCl2 to PtCl þ 1=2Cl2 is
predicted to require 240 kJ molÀ1. The DF calcu-
lations thus predict that linear PtCl2 is formed as
the kinetic product in the reaction of Pt atoms and
Cl2 under these experimental conditions.
€
[4] H. Schafer, Z. Anorg. Allg. Chem. 415 (1975) 217.
[5] U. Wosiewitz, H. Schafer, J. Less-Common Met. 32 (1973)
€
389.
[6] H. Schafer, U. Wosiewitz, Z. Anorg. Allg. Chem. 415
€
(1975) 202.
[7] A. Landsberg, J.L. Schaller, J. Less-Common Met. 23
(1971) 195.
[8] E.V. Makarov, V.A. Bryukvin, V.K. Tagirov, Russ. J.
Non-Ferrous Met. 38 (1997) 35.
[9] D.W. Green, G.T. Reedy, J. Chem. Phys. 69 (1978) 544.
[10] I.R. Beattie, P.J. Jones, K.R. Millington, A.D. Willson, J.
Chem. Soc., Dalton Trans. (1988) 2759.
4. Conclusions
[11] X-ray data booklet, Center for X-ray Optics and Advanced
Light Source, Lawrence Berkeley National Laboratory,
University of California, Berkeley, CA USA, 2001.
[12] N. Binsted, PAXAS, Program for the Analysis of X-Ray
Absorption Spectra, University of Southampton, 1988.
[13] W.E. O’Grady, X. Qian, D.E. Ramaker, J. Phys. Chem. B
101 (1997) 5624.
A hollow-cathode sputtering device is an ex-
cellent source of matrix isolated atoms when used
with neat argon, and new species can be formed
when the argon is doped with chlorine. The use of
matrix isolation electronic absorption spectros-
copy, infrared spectroscopy and X-ray absorption
spectroscopy in conjunction with high level DFT
calculations has resulted in the first observation
and characterization of molecular monomeric
platinum monochloride and platinum dichloride.
This should assist in the spectroscopic character-
ization of the vapour above heated platinum
chlorides.
[14] N. Binsted, EXCURV98, CCLRC Daresbury Laboratory
Computer Program, 1998.
[15] G. te Velde, F.M. Bickelhaupt, E.J. Baerends, C. Fonseca
Guerra, S.J.A. van Gisbergen, J.G. Snijders, T. Ziegler, J
Comput. Chem. 22 (2001) 931, and references therein.
[16] I.R. Beattie, N. Cheetham, M. Gardner, D.E. Rogers, J.
Chem. Soc. A (1971) 2240.
[17] D.H.W. Carstens, W. Brashear, D.R. Eslinger, D.M.
Gruen, Appl. Spectrosc. 26 (1972) 184.
€
[18] W. Klotzbucher, G.A. Ozin, Inorg. Chem. 15 (1976) 292.
[19] K. Jansson, R. Scullman, J. Mol. Spectrosc. 61 (1976) 299.
[20] I.R. Beattie, P.J. Jones, N.A. Young, Chem. Phys. Lett.
177 (1991) 579.
Acknowledgements
[21] P.J. Hendra, Spectra. Chim. Acta 23A (1967) 1275.
[22] L.A. Degen, A.J. Rowlands, Spectra. Chim. Acta 47A
(1991) 1263.
We gratefully acknowledge: the gift of the
sputtering device from Prof. Ian Beattie; the fi-
nancial support of the EPSRC(studentships to
GCand NH, SRS beamtime), the University of
Hull, the Cambridge Overseas Trust and Selwyn
College (Cambridge); the Director of Daresbury
Laboratory for access to the synchrotron; and the
Computational Chemistry Working Party for ac-
cess to computational facilities in the Rutherford
Appleton Laboratory.
[23] N.A. Young, J. Chem. Soc., Dalton Trans. (1996) 249.
[24] A.L. Ankudinov, J.J. Rehr, S.R. Bare, Chem. Phys. Lett.
316 (2000) 495.
ꢀ
ꢀ
[25] R. Ayala, E.S. Marcos, S. Diaz-Moreno, V.A. Sole, A.
~
ꢀ
Munoz-Paez, J. Phys. Chem. B 105 (2001) 7588.
[26] D.H. Templeton, L.K. Templeton, Acta Crystallogr. A 41
(1985) 365.
[27] R. Restori, D. Schwarzenbach, Acta Crystallogr. A 52
(1996) 369.
[28] A. van der Gaauw, O.M. Wilkin, N.A. Young, J. Chem.
Soc., Dalton Trans. (1999) 2405.
[29] G.H.W. Milburn, M.R. Truter, J. Chem. Soc. A (1966)
1609.
[30] R. Wysokinski, D. Michalska, J. Comput. Chem. 22 (2001)
901.
References
[31] P.E.M. Siegbahn, Theor. Chim. Acta 87 (1994) 441.
[32] R. Wesendrup, P. Schwerdtfeger, Inorg. Chem. 40 (2001)
3351.
€
[1] H. Schafer, U. Wiese, K. Rinke, K. Brendel, Angew.
Chem. Int. Ed. Engl. 6 (1967) 253.
€
[2] U. Wiese, H. Schafer, H.G. v. Schnering, C. Brendel, K.
Rinke, Angew. Chem. Int. Ed. Engl. 9 (1970) 158.
[33] A.J. Bridgeman, J. Chem. Soc. Dalton Trans. (1996) 2601.