NMR, IR and DFT studies of phenylplatinum complexes with O-monodentate coordinated silsesquioxanate and auxiliary phosphine ligands

Article abstract of DOI:10.1016/j.jorganchem.2011.09.023

The molecular structure and spectroscopic properties of a series of phenylplatinum complexes containing silsesquioxanate and phosphine ligands with general formula trans-[Pt{O₁₀Si₇(R)₇(OH) ₂}(Ph)(L)₂]

Full text of DOI:10.1016/j.jorganchem.2011.09.023

Journal of Organometallic Chemistry 697 (2012) 23e32
Contents lists available at SciVerse ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
NMR, IR and DFT studies of phenylplatinum complexes with O-monodentate
coordinated silsesquioxanate and auxiliary phosphine ligands
,
b
b
b
c
Neli Mintcheva a , Ivelina Georgieva , Tzvetan Mihaylov , Natasha Trendafilova , Makoto Tanabe ,
*
Kohtaro Osakada c
a Department of Chemistry, University of Mining and Geology, Students’ Town, Sofia 1700, Bulgaria
b Institute of General and Inorganic Chemistry, Bulgarian Academy of Sciences, Acad. G. Bonchev str., Bl. 11, Sofia 1113, Bulgaria
c Chemical Resources Laboratory, Tokyo Institute of Technology, 4259 Nagatsuta, Midori-ku, Yokohama 226-8503, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
The molecular structure and spectroscopic properties of a series of phenylplatinum complexes con-
taining silsesquioxanate and phosphine ligands with general formula trans-[Pt{O10Si7(R)7(OH)2}(Ph)(L)2]
(1: R ¼ cyclo-C5H9, L ¼ PEt3; 2: R ¼ iso-C4H9, L ¼ PEt3; 3: R ¼ CH3, L ¼ PEt3; 4: R ¼ cyclo-C5H9, L ¼ PMe3;
5: R ¼ cyclo-C5H9, L ¼ PMe2Ph; 6: R ¼ cyclo-C5H9, L ¼ PPh2Me; 7: R ¼ cyclo-C5H9, L ¼ PPh3) have been
investigated by DFT/OPW91/6-31G(d) calculations, 1H, 13C, 29Si and 31P NMR and IR spectroscopy. DFT
molecular modeling based on available X-ray and NMR data for complexes 1 and 2 allowed deriving
structure-NMR spectra correlations. It was found that the alkyl substituents (R) attached to Si atoms,
cyclo-C5H9, iso-C4H9 and CH3, slightly influence the geometry and multinuclear NMR parameters of the
complexes in the series studied. The molecular structures of the Pt(II) complexes with R ¼ cyclo-C5H9 (4
e7) were predicted by DFT calculations of their simplified models with R ¼ CH3 (40L70). The geometry
optimizations of 40L70 showed square-planar configuration of Pt(II) center bonded to two trans phos-
phine ligands, a phenyl group and an O-monocoordinated silsesquioxanate. The structures 40L60 are
stabilized by two intramolecular hydrogen bonds similar to 1 and 2. A fast conformer exchange process
A4B and switching of H-bonds in solution of 1e6 were suggested based on (i) the calculated conformer
energies and small barrier of the process, and (ii) the observed single 1H NMR signal at low magnetic
field. The stability of the Pt(II) complexes depends on the nature of the phosphine ligands and decreases
in the order PMe2Ph > PMe3 > PPh2Me > PEt3 > PPh3. The PPh3 ligands attached to Pt(II) in 7 cause the
largest geometry changes and a new set of weaker hydrogen bonds. The comparison of the calculated
NMR and IR parameters with the experimental spectroscopic data reveals good coincidence and thus
confirmed the suggested molecular structures.
Received 5 August 2011
Received in revised form
20 September 2011
Accepted 26 September 2011
Keywords:
Silsesquioxane
Platinum complexes
Monophosphine ligands
DFT calculations
NMR
IR
Ó 2011 Elsevier B.V. All rights reserved.
1. Introduction
Recently our group reported preparation, X-ray and multinu-
clear NMR spectroscopic characterization of a number of platinum
Silsesquioxanes are organosilanol compounds possessing
eOeSieOe bonds network and organic group attached to each Si
atom. Silsequioxanes containing one, two, three or four silanol
groups, eSieOH, belong to the incompletely condensed silsesquiox-
anes and they are defined as mono-, di-, tri-, tetrasilanol silses-
quioxanes [1]. After 90’s when Feher has discovered facile synthesis
of incompletely condensed silsesquioxanes the investigations of their
metal complexes have been rapidly progressing [2]. New compounds
with improved mechanical, thermal and dielectric properties as well
as with novel properties based on surface activity are designed and
they attract much scientific and technological interest [3].
and palladium complexes with trisilanol silsesquioxane [4]. The
crystal structures of two phenylplatinum complexes with silses-
quioxanates trans-[Pt{O10Si7R7(OH)2}(Ph)(PEt3)2] (1: R ¼ cyclo-
C5H9, 2: R ¼ iso-C4H9) were determined [5]. However growth of
single crystals suitable for X-ray structural analysis failed for some
other complexes during the study. In these cases, combination of
NMR and IR spectroscopy and DFT molecular modeling is extremely
valuable for structural determination of the Pt(II)-silsesquioxane
complexes. Recently DFT calculations were successfully used to
analyze the molecular structure and NMR properties of the dinu-
clear platinum complex with silsesquioxanate ligand [6].
In this paper DFT/OPW91/6-31G(d) method is applied to discuss
the molecular structures of the new synthesized complexes trans-
[Pt{O10Si7(cyclo-C5H9)7(OH)2}(Ph)(L)2] (4: L ¼ PMe3, 6: L ¼ PPh2Me,
* Corresponding author. Tel.: þ359 2 8060 311; fax: þ359 2 962 49 40.
E-mail address: nmintcheva@abv.bg (N. Mintcheva).
0022-328X/$ e see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2011.09.023

24
N. Mintcheva et al. / Journal of Organometallic Chemistry 697 (2012) 23e32
7: L ¼ PPh3) containing O-monodentate coordinated silsesquioxa-
nate and auxiliary phosphine ligands. To complete the series of the
Pt(II)-silsesquioxane complexes above, another previously reported
complex trans-[Pt{O10Si7(cyclo-C5H9)7(OH)2}(Ph)(PPhMe2)2] (5)
was also included in the theoretical study [5]. The reliability of the
method used was verified by calculations of two crystallography
characterized complexes trans-[Pt{O10Si7R7(OH)2}(Ph)(PEt3)2] (1:
R ¼ cyclo-C5H9, 2: R ¼ iso-C4H9) [5]. A special attention was paid to
the effect of the organic groups attached to each Si atom in the
silsesquioxanate on the geometrical parameters and NMR chemical
shifts, because most of the DFT calculations are performed for the
model compounds where the bulky cyclo-C5H9 substituents in the
silsesquioxane framework are replaced by the simplest alkyl group,
CH3. 1H, 13C, 31P and 29Si NMR chemical shifts and IR spectra pf the
optimized Pt(II) complexes (1e3, 40e70) are calculated and the
experimental NMR and IR data are assigned. On the basis of the
calculated and experimental spectra comparison, the suggested
molecular structure of the new complexes is confirmed. The effect
of the phosphine ligands PMe3, PEt3, PPhMe2, PPh2Me and PPh3 on
the geometry, stability and spectroscopic properties in a series of
Pt(II) complexes (3, 40e70) was examined as well.
(121.5 MHz, benzene-d6, r.t.):
NMR (100 MHz, benzene-d6, r.t.):
23.14, 23.72, 23.93, 26.85 (1:1:2:2:1) (7C, CH-pentyl),
27.46e29.57 (28C, CH2-pentyl), 122.47 (C6H5 para), 128.63 (C6H5
meta), 138.28 (C6H5 ortho), (C6H5 ipso was not observed). 29Si{1H}
d
ꢀ10.97 ppm, JPtP ¼ 2842 Hz 13C{1H}
d
12.84 (vt, JCP ¼ 18 Hz, PCH3), 23.00,
NMR (79.5 MHz, benzene-d6, 0.02
M
Cr(acac)3, r. t.):
d
ꢀ57.05, ꢀ57.87, ꢀ65.10, ꢀ65.60, ꢀ67.75 (2:1:1:1:2). IR spectrum
(KBr): 3400 (m), 2949 (s), 2865 (s), 1572 (w), 1453 (w), 1420 (w), 1289
(w),1246 (m),1100 (vs), 949 (s), 910 (m), 856 (w), 740 (m), 706 (w), 503
(m) cmꢀ1
.
2.2.2. Synthesis of
trans-[Pt{O10Si7(cyclo-C5H9)7(OH)2}(Ph)(PPh2Me)2] (6)
To
a toluene solution (10 ml) of trans-[PtI(Ph)(PPh2Me)2]
(0.15 mmol, 0.120 g) were added silsesquioxane (c-C5H9)7Si7O9(OH)3
(0.15 mmol, 0.131 g) and Ag2O (0.20 mmol, 0.046 g). The reaction
mixture was stirred at room temperature for 6 days. When the reaction
completed the suspension was filtrated off and the filtrate was
evaporated under reduced pressure to dryness. The product was
extracted with 10 ml hexane, and the solution was filtrated, evapo-
rated and dried in vacuum to give 6in 99% yield (0.230 g). Anal. Calc. for
C67H96O12Si7P2Pt: calc. (found) C 52.02 (52.26) %; H 6.25 (6.23) %. 1H
NMR (300 MHz, benzene-d6, r.t.): 9.0 (s, 2H, OH), 7.44 (8H,P-C6H5-o),
7.00 (12H,P-C6H5-m,p), 6.65 (d, 2H, Pt-C6H5-o), 6.40 (t, 1H, Pt-C6H5-p),
6.25 (t, 2H, Pt-C6H5-m), 2.43 (m, 9H, P-CH3), 2.3e1.4 (m, 56H, CH2-
pentyl), 1.2 (m, 7H, CH-pentyl). 31P{1H} NMR (121.5 MHz, benzene-d6,
2. Experimental Section
2.1. Reagents and instruments
All manipulations of the complexes were carried out using
standard Schlenk techniques under argon or nitrogen atmosphere.
Toluene for reactions was distilled using the Glass Contour purifi-
cation system and stored under nitrogen. The platinum complexes,
trans-[PtI(Ph)(PMe3)2], trans-[PtI(Ph)(PPh2Me)2], trans-[PtI(Ph)
(PPh3)2] and their precursor [PtI(Ph)(cod)] were prepared according
to the literature [7]. The hydroxo-platinum complex trans-
r.t.):
r.t.):
d
13.05 ppm, JPtP ¼ 3098 Hz 13C{1H} NMR (100 MHz, benzene-d6,
13.50 (vt, JCP ¼ 17 Hz, PCH3), 23.01, 23.14, 23.78, 23.90, 26.48
d
(1:1:2:2:1) (7C, CH-pentyl), 27.38e29.74 (28C, CH2-pentyl), 121.05 (Pt-
C6H5 para), 127.03 (Pt-C6H5 meta), 128.64 (P-C6H5 meta), 132.66 (vt,
JPC ¼ 28 Hz, P-C6H5 ipso),133.31 (P-C6H5 ortho),138.88 (Pt-C6H5 ortho),
(Pt-C6H5 ipso was not observed). 29Si{1H} NMR (79.5 MHz, benzene-d6,
0.02 M Cr(acac)3, r. t.):
d
ꢀ56.87, ꢀ57.38, ꢀ64.80, ꢀ65.03, ꢀ67.70
[Pt(OH)(Ph)(PPh3)2]
and
silsesquioxane,
1,3,5,7,9,11,14-
(2:1:1:1:2). IR spectrum (KBr): 3450 (m), 3252 (sh), 2950 (s), 2861 (s),
1572 (w),1451 (w),1437 (w),1292 (w),1247 (m),1110 (vs), 948 (s), 909
heptacyclopentyltricyclo [7.3.3.1(5,11)]hepta-siloxane-endo-3,7,14-
triol, ((cyclo-C5H9)7Si7O9(OH)3) were synthesized as described by
Yoshida et al. [8] and Pescarmona et al. [9]. Ag2O was purchased
from Wako Pure Chemical Ind., Ltd.
(m), 857 (m), 744 (m), 705 (m), 530 (sh), 501 (s) cmꢀ1
.
2.2.3. Synthesis of trans-[Pt{O10Si7(cyclo-C5H9)7(OH)2}(Ph)(PPh3)2]
1H, 13C{1H}, 29Si{1H}, and 31P{1H} NMR spectra were recorded on
Varian Mercury 300 or Bruker Biospin Avance III 400 spectrometers.
Chemical shifts of the signals in 1H and 13C{1H} NMR spectra were
adjusted to the residual peaks of the solvents used. Peak positions in
the 29Si{1H} and 31P{1H} NMR spectra were referenced to external
standard SiMe4 in C6D6 and 85% H3PO4 in C6D6, respectively. IR
absorption spectra were recorded on a Shimadzu FT/IR-8100 spec-
trometer. Elemental analyses were carried out with a Yanaco MT-5
CHN autorecorder at the Center for Advanced Materials Analysis,
Technical Department, Tokyo Institute of Technology.
(7)
To
a benzene solution (2 ml) of trans-[Pt(OH)(Ph)(PPh3)2]
(0.070 mmol, 0.056 g) was added silsesquioxane (c-C5H9)7Si7O9(OH)3
(0.070 mmol, 0.070 g) and the mixture was stirred at room tempe-
rature for 5 days. The solvent was evaporated under reduced pressure,
the residue was washed with hexane (2 times ꢂ 3 ml) and dry in
vacuum to give 7 in 89% yield (0.103 g). 1H NMR (300 MHz, benzene-
d6, r.t.): 7.7 (m,12H, P(C6H5)3), 7.1 (m,18H, P(C6H5)3), 6.5 (t,1H, Pt-C6H5
para), 6.3 (t, 2H, JHH ¼ 8.4 Hz, Pt-C6H5 meta), 4.7 (br, 2H, OH), 2.3e1.4
(m, 56H, CH2-pentyl),1.1 (m, 7H, CH-pentyl). 31P{1H} NMR (121.5MHz,
benzene-d6, r.t.):
benzene-d6, 0.02 M Cr(acac)3, r. t.):
d
24.6 ppm, JPtP ¼ 3225 Hz 29Si{1H} NMR (79.5 MHz,
2.2. Preparation of complexes 4, 6, 7
d
ꢀ49.9, ꢀ57.1, ꢀ57.3, ꢀ60.2, ꢀ67.6
(1:1:2:1:2). A reliable elemental analysis data of 7 and a precise
assignment of 13C NMR spectra were not obtained due to contami-
nation of Ph3PO and other impurities. IR spectrum (KBr): 3400 (m),
2949 (s), 2865 (s),1572 (w),1435 (w),1246 (m),1100 (vs), 910 (m), 737
2.2.1. Synthesis of trans-[Pt{O10Si7(cyclo-C5H9)7(OH)2}(Ph)(PMe3)2]
(4)
To a toluene solution (10 ml) of trans-[PtI(Ph)(PMe3)2] (0.15 mmol,
0.076 g) were added silsesquioxane (c-C5H9)7Si7O9(OH)3 (0.15 mmol,
0.131 g) and Ag2O (0.20 mmol, 0.046 g), then the mixture was stirred at
room temperature for 6 h. The unreacted Ag2O and AgI were filtrated
through celite and the filtrate was evaporated. The crude product was
dissolved in hexane and was filtrated again. The solution was
concentrated to 1.5e2 ml and was stored at ꢀ20 ꢁC for a few days to
give compound 4 as crystals in 65% yield (0.126 g). Anal. Calc. for
C47H88O12Si7P2Pt: calc. (found) C 43.46 (42.76) %; H 6.83 (6.53) %. 1H
NMR (400 MHz, benzene-d6, r.t.): 9.2 (s, 2H, OH), 7.28 (d, 2H, C6H5-o),
6.93 (m, 3H, C6H5-m þ C6H5-p), 2.3e1.4 (m, 56H, CH2-pentyl), 1.2 (m,
7H, CH-pentyl), 1.05 (vt, JHH ¼ 4 Hz, 18H, CH3). 31P{1H} NMR
(m), 706 (m), 523 (m), 502 (m) cmꢀ1
.
2.3. Computational Procedure
Full geometry optimizations of the molecular systems (1, 2, 3,
40e70) were carried out without symmetry constraint. Frequency
calculations were performed and the minima on the potential
energy surfaces of the reported structures were characterized by
the absence of negative eigenvalues in the Hessian matrix. The
vibrational modes were analyzed by means of the atom move-
ments, calculated in Cartesian coordinates and by visual inspection

N. Mintcheva et al. / Journal of Organometallic Chemistry 697 (2012) 23e32
25
of the vibrational modes animated with ChemCraft program [10].
3. Results and discussion
Second generation pure DFT functional, OPW91, consisting of
exchange OPTX [11] and correlation PW91 was used [12]. The split-
valence plus polarization basis set, 6-31G(d) was applied for the
ligand atoms (d polarization function for O, Si, C, P atoms) in case of
the geometry optimizations and frequency calculations and 6-
31G(d,p) basis set (p polarization function for H atoms) for NMR
calculations [13]. The Stuttgart RSC 1997 ECP basis set for Pt(II) ion
was used [14]. The Pt (II) atom was described with pseudo-
relativistic effective core potential ECP60MWB where [Kr]4d104f14
3.1. Synthesis of Pt(II)-silsesquioxanate complexes
The reaction of complex trans-[PtI(Ph)(L)2] (L ¼ PMe3, PPh2Me)
and silsesquioxane trisilanol (cyclo-C5H9)7Si7O9(OH)3 in the pres-
ence of Ag2O gives platinum complex with O-monodentate coor-
dinated silsesquioxanate and auxiliary phosphine ligands, trans-[Pt
{O10Si7(cyclo-C5H9)7(OH)2}(Ph)(L)2] (4: L ¼ PMe3, 6: L ¼ PPh2Me) as
shown in eq. (1).
core (60 electrons) was replaced by pseudopotential and the
5s25p65d96s1 outer-core shell was treated explicitly in the valence
space [15]. Previous methodological studies of Pt(II)-histamine
complex have shown that OPW91 functional combined with the
6-311G(d,p) basis set for non-metallic atoms and Stuttgart ECP
basis set for platinum atom yields the most satisfactory agreement
in relation to the geometrical parameters and 1H and 13C NMR
chemical shifts [16].
When a silsesquioxane reacts with platinum complex having PPh3
ligands, trans-[PtI(Ph)(PPh3)2] in the presence of Ag2O, a hetero-
bimetallic PteAg complex [Pt{O11Si7(cyclo-C5H9)7(OH)(AgPPh3)}
(Ph)(PPh3)] was formed as previously reported [5]. Silver-free
complex, trans-[Pt{O10Si7(cyclo-C5H9)7(OH)2}(Ph)(PPh3)2] (7) was
obtained by metathesisreactionofhydroxo-platinumcomplex, trans-
[Pt(OH)(Ph)(PPh3)2] and silsesquioxane in benzene solution at room
temperature for 5 days (Eq.(2)).
The 1H, 13C, 29Si and 31P isotropic shielding constants of the
optimized mononuclear Pt(II) complexes were calculated by using
the gauge independent atomic orbital (GIAO) method [17]. The
absolute isotropic magnetic shielding constants (si) were used to
Heating of the reaction mixture or toluene (benzene) solution of
isolated complexes 4, 6 and 7 causes their decomposition and
formation of the corresponding phosphine oxide. Complex 7 is stable
in the maternal solution for a few days, but is not isolated as a pure
solid due to tendency to contain Ph3PO impurity formed during
purification. Owing to that reason we were not able to obtain reliable
elemental analysis data and assignment of 13C NMR spectrum.
obtain the chemical shifts (di
¼
sTMSesi) by referring to the stan-
dard compounds, tetramethylsilane (TMS) (for Si, C and H atoms)
and H3PO4 (for P atom). Both referred to compounds, TMS and
H3PO4, were calculated at the same OPW91/6-31G(d,p) level of
theory.
3.2. Methodological investigations of
trans-[Pt{O10Si7R7(OH)2}(Ph)(PEt)2](1:R¼ cyclo-C5H9;2:R¼ iso-C4H9)
complexes
All calculations were carried out using Gaussian09 program
package [18]. The effect of the benzene solvent on NMR chemical
shifts of 2 and on the formation energies of 3, 40e70 was computed
using the conductor-like polarizable continuum model (CPCM)
implemented in Gaussian09 [19].
To perform
a reliable molecular modeling of the newly
synthesized Pt complexes and to analyze correctly their

26
N. Mintcheva et al. / Journal of Organometallic Chemistry 697 (2012) 23e32
multinuclear NMR and IR spectra, an appropriate DFT method
should be selected. In the last decades DFT methods have proven to
be powerful and economic alternatives for calculations of structure
and spectroscopic properties of Pt complexes. However responsible
results obtained for one type of Pt complexes can not be guaranteed
for other types. Hence it becomes of great importance to test the
reliability of the computational method employed for each parti-
cular case. The mononuclear Pt(II) complexes 1 and 2 were struc-
turally characterized by X-ray diffraction analysis [5] and they were
used as benchmark examples to estimate the reliability of the DFT/
OPW91 level of calculations as to the geometrical and NMR
parameters. The structures of 1 and 2 are schematic presented in
Fig. 1. Selected calculated bond lengths of optimized geometries of
1 and 2 at OPW91/6-31G(d) level (ECPMWB60 potential for Pt) and
the corresponding experimental crystallographic data are given in
Table S1 in Supplementary materials.
displayed larger bond length changes up to 0.016 Å, which may be
due to the solid state effects. The negligible effect of cyclo-C5H9 and
iso-C4H9 substituents on the geometrical parameters of the phe-
nylplatinum silsesquioxane complexes 1 and 2 gives us the reason
to simplify the molecular system of the complexes by replacing the
bulky substituents in silsesquioxane with the simplest alkyl group
(CH3): trans-[Pt{O10Si7(CH3)7(OH)2}(Ph)(PEt)2] (3). The calculated
bond distances of 3 reveal no significant deviations from the
calculated values of 1 and 2 (Table S1 in Supplementary materials).
Therefore we apply the models with CH3 substituents in our further
study of platinum-silsesquioxane complexes 4e7.
3.3. Geometry optimization of Pt(II)-silsesquioxane complexes with
phosphine ligands (4e7)
The theoretical study of
a series of complexes trans-[Pt
The mean unsigned error (MUE) for the bond lengths is 1.8%
(0.029 Å) and 1.5% (0.026 Å) for 1 and 2, respectively. The
enlargement of the basis set including polarization (p) functions on
all non-metallic atoms does not change the calculated bond
distances. Inclusion of diffuse functions on C, P, O atoms (bonded to
Pt) leads to elongation of PteP, PteO and PteC bond distances by
w0.02 Å and causes larger deviations from the experimental data.
For comparison, the popular B3LYP functional showed larger
distance discrepancies (by 0.04 Å). It should be mentioned that one
reason for the bond length discrepancy in our case is due to
comparison of experimental geometrical parameters obtained in
solid state with those calculated in gas phase. Taking into account
the large molecules under consideration, 175 and 168 atoms for 1
and 2, the OPW91/6-31G(d) method appears reasonable compro-
mise between accuracy and CPU time of calculations and therefore
we applied it together with ECPMWB60 for Pt(II) for structural
characterization of the new platinum silsesquioxane complexes 4, 6
and 7.
Since the benchmark Pt(II) complexes differ in organic substi-
tuents bonded to each Si atom of silsesquioxane, cyclo-C5H9 in (1)
and iso-C4H9 in (2), the substituents effect on the geometry
parameters, especially on the PteL bond lengths (L ¼ O, P, C) was
estimated. A comparative analysis showed small calculated bond
length changes up to 0.007 Å. Slight increasing of the calculated
PteP1, PteP2 and PteO bond lengths was observed when the iso-
C4H9 groups were replaced with more bulky cyclo-C5H9 (2.340,
2.337, 2.190 Å for 2; 2.342, 2.340, 2.200 Å for 1) and this trend well
reproduces the increasing of experimental bond distances PteO
and PteP in the order 2 < 1. The crystallographic data for 1 and 2
{O10Si7(cyclo-C5H9)7(OH)2}(Ph)(L)2] (1: L ¼ PEt3, 4: L ¼ PMe3, 5:
L ¼ PPhMe2, 6: L ¼ PPh2Me, 7: L ¼ PPh3) was carried out by model
compounds, where the cyclo-C5H9 substituents were replaced by
CH3 groups and they are denoted as follows: trans-[Pt
{O10Si7(CH3)7(OH)2}(Ph)(L)2] (3: L ¼ PEt3, 40: L ¼ PMe3, 50:
L ¼ PPhMe2, 60: L ¼ PPh2Me, 70: L ¼ PPh3). The optimized geome-
tries of the model complexes, 40, 50, 60 and 70 provided detailed and
valuable information that can be used to suggest the molecular
structures of their analogues 4, 5, 6 and 7 since suitable single
crystals for structure characterization were not obtained. Selected
calculated bond lengths of 40e70 at OPW91/6-31G(d) level are
collected in Table S2 in Supplementary materials. In the optimized
compounds the Pt(II) center has square-planar configuration and it
is bonded to two phosphine ligands at trans position, phenyl ring
and O-monocoordinated silsesquioxane, as it is shown in Fig. 1.
In all optimized complexes the Ph group bonded to Pt atom is
directed perpendicular to the coordination plane, in agreement
with the crystallographycally determined structures of 1 and 2 [5].
The stability of the conformational isomers of 50 and 60 depending
on the location of Me and Ph groups in the phosphine ligands,
PMe2Ph and PPh2Me respectively was examined. The most stable
conformer of 50 contains the Ph group located on the side of the
phenyl ligand bonded to Pt and two Me groups are situated close to
the silsesquioxane moiety, as shown in Fig. 2(a). In the most stable
conformer of 60 the Me group attached to P atom is close to the
silsesquioxane skeleton, Fig. 2(b).
The electronic and steric effects of the phosphorus substituents
cause slight changes in the bond distances around the Pt(II) center.
Within the studied complexes there is not well defined trend in the
PteL bond lengths (L ¼ C, P, O). The shortest PteP distances are
found for 50 (2.313 Å) and they increase up to 2.345 Å in the order
50 < 40 < 60 < 70 < 3. At the same time, the PteC1 bond lengths in 50
and 60 are larger (2.01 Å) than those of 3, 40 and 70 (2.00 Å). The
shortest PteO1 bond length is observed in 70 (2.155 Å), while the
longest one (2.197 Å) is found in 3. Thus, the PteO bond distances
increase in the following order 70 <40 < 50 < 60 < 3.
The SieO distances in the silsesquioxane fragment of 3, 40, 50, 60
vary between 1.63 and 1.68 Å. Significant deviations from these
average values are observed for 70 where the Si1eO1 bond is
shortened (1.596 Å) and the Si1eO2 and Si1eO8 bonds are elon-
gated (1.691 Å and 1.706 Å) (see the behavior of 29Si NMR signals of
7 in Section 3.5). Such changes indicate deformation of the silses-
quioxane fragment probably because of the steric hindrance in the
molecule, Fig. 2(c). Two intramolecular hydrogen bonds,
O7eH7,,,O1 and O6eH6,,,O7 (Fig. 1) are observed in the opti-
mized structures 3, 40, 50, 60. The shorter H7,,,O1 distances
(1.647e1.734 Å) than H6,,,O7 ones (1.824e1.876 Å) reveal
a stronger hydrogen bond between the coordinated oxygen and the
OH group in agreement with the experimental crystallographic
Fig. 1. Geometry of trans-[Pt{O10Si7(R)7(OH)2}(Ph)(L)2] (1: R ¼ cyclo-C5H9, L ¼ PEt3; 2:
R ¼ iso-C4H9, L ¼ PEt3; 3: R ¼ CH3, L ¼ PEt3; 40: R ¼ CH3, L ¼ PMe3; 50: R ¼ CH3,
L ¼ PPhMe2; 60: R ¼ CH3, L ¼ PPh2Me).

N. Mintcheva et al. / Journal of Organometallic Chemistry 697 (2012) 23e32
27
Fig. 2. OPW91/6-31G(d) optimized geometries of trans-[Pt{O10Si7(CH3)7(OH)2}(Ph)(L)2] (50: L ¼ PPhMe2) (a), (60: L ¼ PPh2Me) (b) and (70: L ¼ PPh3) (c).
data of 1 and 2 [5]. In the series studied both H,,,O distances
increase in the order 40 < 50 < 60 < 3 << 70. Among the Pt(II)
complexes studied the structure of 70 shows a materially difference,
Fig. 2(c). The largest H6,,,O7 bond length (1.966 Å) is obtained in
70. The H7,,,O1 distance (3.746 Å) is much longer than the sum of
the van der Waals radii of H and O atoms (2.72 Å) [20] and therefore
the hydrogen bonding breaks off. A new H-bond is formed between
OH group and an oxygen atom from the eSieOeSie network,
where H7,,,O8 bond length is 2.284 Å. Obviously the bulk phenyl
groups of the PPh3 ligands induce a reorganization in the silses-
quioxane skeleton including changes of the SieO bonds and
formation of a new set of weak hydrogen bonds. Such changes lead
to lower stability of the complex 7 in comparison with the other
species in the series.
The calculated NMR data confirmed the assignment of the
experimental NMR spectra of 1 and 2 reported in [5] with some
exceptions, which are discussed below.
The observed 13C signals of 1, 2 and 3 at three regions 7e13 ppm,
23e30 ppm and 120e140 ppm were assigned to methyl and
methylene carbons of the PEt3 ligands, to carbons of iso-C4H9 or
cyclo-C5H9 groups and to phenyl carbons, respectively. The calcu-
lated chemical shifts reproduce very well the trend of the experi-
mental 13C NMR data of 1 and 2. In low magnetic field (
d
112e157 ppm) six signals for phenyl carbons in the order (1 Cpara, 2
identical Cmeta, 2 identical Cortho and 1 Cipso) were calculated. Similar
NMR results were obtained for the dinuclear platinum silses-
quioxane complex [6]. Recently Kakeya et al. studied diaryl Pt(II)
13
complexes and reported a
C
signal with well solved coupling
ipso
constants in the lowest magnetic field position [21]. However in the
case of Pt(II)-phenyl silsesquioxane complexes we observed an
intensive downfield signal for Cortho (w138 ppm) followed by weak
signal (w131 ppm) with coupling constant JCP ¼ 10 Hz for Cipso. The
latter was assigned based on the coupling patterns confirmed by
the calculated coupling constants JCiP ¼ 10.3av and JCiP ¼ 9.0av for 2
and 3, respectively (Table S3). According to the calculations we are
able to make a correct assignment of the signals of Cmeta and Cpara
atoms. The peak at 127.85 ppm for 2 should be assigned to Cmeta and
3.4. 1H, 13C, 31P and 29Si NMR calculations of the
trans-[Pt{O10Si7R7(OH)2}(Ph)(PEt)2](1:R¼ cyclo-C5H9;2:R¼ iso-C4H9)
complexes
The multinuclear NMR parameters of complexes 1e3 were
calculated with DFT/OPW91/6-31G(d,p) method and the results are
presented in Table S3 in Supplementary materials. The same
method was successfully used for calculations of 1H, 13C, 31P and
29Si chemical shifts of the dinuclear Pt(II) complex with silses-
quioxanate ligand reported recently [6]. The experimental NMR
spectra of the studied complexes were measured in benzene-d6. To
approach the experiment, 1H, 13C, 31P and 29Si chemical shifts of 2
were calculated in benzene solution as well (Table S3).
The NMR data differences obtained in solution and in gas phase
are up to 0.5 ppm, which indicate insignificant solvent effect on the
observed chemical shifts. Hence, we accept that comparison of the
calculated 1H, 13C, 31P and 29Si NMR data in gas phase with the
experimental chemical shifts obtained in solution is reasonable.
the signals at 122.38 and 122.36 ppm for
para,respectively.
In the 1H NMR spectra of 1 and 2 the low magnetic field position
1 and 2 e to
C
of the signals at 8.53 and 8.84 ppm, respectively were assigned to
the hydroxy hydrogens involved in two intramolecular bondings.
Two chemical shifts at 8.66/6.72 ppm and 8.79/6.78 ppm for 1 and
2, respectively were computed due to different H-bonding
strengths of the two hydroxyl groups as predicted from the
experimental and calculated structures, Table S1. The average NMR
values (7.69 ppm for 1 and 7.79 ppm for 2) well reproduce the

28
N. Mintcheva et al. / Journal of Organometallic Chemistry 697 (2012) 23e32
experimental 1H signals. The downfield shift of the signal for 2
corresponds with the stronger H-bonding in this complex. Gene-
rally in the 1H NMR spectra of the studied complexes 1e6 a single
OH hydrogen peak was observed at low magnetic field position.
Such signal may due to (i) a rapid exchange of the two hydroxy
hydrogens in solution by dynamic processes between the two
isomers A and B, Fig. 3 or (ii) a complex structure with symmetri-
cally bifurcated hydrogen bonds between the coordinated oxygen
and the two SieOeH groups [5].
Another possible explanation is a transfer of the Pt atom to the
other OH groups, which would lead to the position isomers A, C and
D shown in Fig. 3. All possible isomers are optimized and the
relative energies are given in Fig. 3. The isomers A and B with
switching hydrogen bonds (O7eH7$$$O1 and O6eH6$$$O7 for A
and O7eH7$$$O6 and O6eH6$$$O1 for B) are very close in energy
(difference 1.1 kcal/mol for complex 2 and 0.01 kcal/mol for 40
complex). Thus from energetic point of view both conformers may
exist in solution. To estimate the possibility of a conformational
transition A 4 B, the transition state structure of the process was
calculated. For the simplest 40 complex in gas phase the calculations
predicted a low energy barrier of 4.1 kcal/mol. Approximately the
same energy barrier value was calculated in toluene solution,
4.2 kcal/mol. On the basis of these values a transfer A 4 B could be
suggested. The transition state structure (first order saddle point)
has symmetrically bifurcated hydrogen bonds, as shown in Fig. 3.
The calculations predicted another minimum energy structure with
slightly higher energy (4 kcal/mol) and with equivalent bifurcated
H-bonds and OH hydrogen chemical shifts. Both OH$$$ O hydrogen
bonds are weaker than those in A or B, and their calculated 1H NMR
optimized structures are indistinguishable and reveal identical 1H
NMR chemical shifts. Such isomers may exist in solution but the
conversion between them is much less probable.
The dynamic behaviour of the Pt-silsesquioxanes in solution was
studied by variable-temperature 1H NMR spectroscopy. It revealed that
the signal is shifted downfield with decreasing of temperature and no
signal splitting was observed [5]. Obviously rapid exchange between
forms A and B takes place even at ꢀ80 ꢁC in toluene solution. The
downfield 1H NMR shifts ( 8.53 ppm (at 25 ꢁC) and
d d 9.2 ppm
(at ꢀ80 ꢁC)) could be explained by decrease of the electron density
around the hydrogen nucleus due to increasing H-bond strength. At
low temperature the H,,,O distances decrease due to reduced pop-
ulations of the excited vibrational levels. At the same time the OeH
bond length increases and the electric field exerted by O at H decreases,
leading to a downfield shift about 0.7 ꢂ 10ꢀ2 ppm per ꢁC [22].
In the range 7.8O6.9 ppm five peaks for the phenyl 1H atoms of
1 and 2 were calculated and the results are in a very good agree-
ment with the experimental ones, Table 3. In the experimental 1H
NMR spectra the meta- and para-phenyl hydrogen signals are
overlapped and observed at 6.94 ppm, while the signals for both
ortho-phenyl hydrogens are split at room temperature as the
calculations predicted.
The methodological study of the geometry and multinuclear 13C,
29Si, 31P and 1H NMR data of 1 and 2 revealed that the OPW91/6-
31G(d)//6-31G(d,p) level combined with ECPMWB60 for Pt(II) can
be successfully applied to predict the structure of the Pt(II)-
silsesquioxane complexes 4, 5, 6 and 7. The R-substituents (cyclo-
C5H9, iso-C4H9 and CH3) at silsesquioxane fragment have shown an
insignificant effect on the 1H, 13C, 31P and 29Si NMR data (up to
w2 ppm) concerning silsesquioxane skeleton and phenyl and
auxiliary phosphine ligands. This result gives us a reason to
continue the NMR study of complexes 3e7 using their model
analogues and replacing cyclo-C5H9 with CH3 residue.
signal is at higher magnetic field (d w6 ppm) in comparison with
the signals of 1e6. The position isomers A, C and D where Pt(II) is
bonded to O1, O6 and O7 atoms of the silsesquioxane showed also
small relative energies (0.0, 1.2 and 1.6 kcal/mol) (Fig. 3). The
Fig. 3. OPW91/6-31G(d) optimized structures of the isomers with different H-bondings A, B and TS for A4B conversion; position isomers A, C and D, where Pt atom is attached to
the possible O atoms for trans-[Pt{O10Si7(R)7(OH)2}(Ph)(L)2] (2: R ¼ iso-C4H9, L ¼ PEt3; 40: R ¼ Me, L ¼ PMe3*).
DE is a relative energy of isomers in kcal/mol.

N. Mintcheva et al. / Journal of Organometallic Chemistry 697 (2012) 23e32
29
Table 1
Selected calculated 1H, 13C, 31P and 29Si NMR chemical shifts (ppm) and nuclear spinespin coupling constants
J
(Hz) of the model compounds
trans-[Pt{O10Si7(CH3)7(OH)2}(Ph)(L)2] (40: L ¼ PMe3, 50: L ¼ PMe2Ph, 60: L ¼ PPh2Me, 70: L ¼ PPh3) at OPW91/6-31G(d,p) level and experimental NMR data for the complexes
trans-[Pt{O10Si7(cyclo-C5H9)7(OH)2}(Ph)(L)2] (4: L ¼ PMe3, 5: L ¼ PMe2Ph, 6: L ¼ PPh2Me, 7: L ¼ PPh3).
calc
calc
calc
calc
Atom
1exp
a
3calc
4exp
40
5exp
a
50
6exp
60
7exp
70
13C
PCH3
PCH2CH3
PCH2CH3
e
e
12.84
e
13.96
e
11.8
e
13.87
e
13.50
e
15.54
e
e
e
e
e
7.64
10.88, JPC¼ 4.0av
18.14, JPC ¼ 18.6av
e
12.98, JPC ¼ 16.4
P-C6H5 para
P-C6H5 meta
P-C6H5 ortho
P-C6H5 ipso
Pt-C6H5 para
Pt-C6H5 meta
Pt-C6H5 ortho
Pt-C6H5 ipso
29Si
e
e
e
129.75
128.21
120.07av
119.61av
121.73av
b
120.47av
119.54av
123.5av
125.86av
112.32
118.66
126.45
151.89
120.35av
119.90av
125.02av
126.11av
112.35
119.31
125.55
156.09
128.64
133.31
132.66
121.05
127.03
138.88
b
131.37, JCoP ¼ 5.7
134.26, JCiP ¼ 27.3 126.69av
122.38
Overlapped
137.61
130.95, JCiP ¼ 10.2 125.84, JCiP ¼ 9.0
113.37
119.16
125.84, JCoP ¼ 2.4
122.47
128.63
138.28
b
113.75
119.31
125.45
155.08
122.09
127.30
138.37
113.49
118.32
126.94
129.06, JCiP ¼ 10.2 151.27
eOeSi6eOe
eOeSi2eOe
eOeSi4eOe
eOeSi7eOe
eSi1eOePt
eSi3eOH
eSi5eOH
1H
ꢀ67.68
ꢀ67.68
ꢀ64.69
ꢀ64.35
ꢀ57.67
ꢀ56.65
ꢀ56.65
ꢀ58.92
ꢀ52.50
ꢀ52.31
ꢀ51.99
ꢀ42.66
ꢀ43.46
ꢀ39.68
ꢀ67.75
ꢀ67.75
ꢀ65.60
ꢀ65.10
ꢀ57.87
ꢀ57.05
ꢀ57.05
ꢀ57.42
ꢀ53.19
ꢀ52.50
ꢀ52.06
ꢀ42.95
ꢀ42.79
ꢀ40.44
ꢀ67.60
ꢀ67.60
ꢀ64.87
ꢀ64.59
ꢀ57.30
ꢀ56.75
ꢀ56.75
ꢀ58.42
ꢀ52.60
ꢀ52.58
ꢀ52.16
ꢀ43.07
ꢀ42.95
ꢀ41.10
ꢀ67.70
ꢀ67.70
ꢀ65.03
ꢀ64.80
ꢀ57.38
ꢀ56.87
ꢀ56.87
ꢀ58.58
ꢀ52.05
ꢀ52.02
ꢀ51.97
ꢀ41.46
ꢀ43.22
ꢀ39.97
ꢀ67.6
ꢀ67.6
ꢀ57.3
ꢀ57.3
ꢀ60.2
ꢀ57.1
ꢀ49.9
ꢀ52.23
ꢀ52.16
51.11
ꢀ51.07
ꢀ51.61
ꢀ41.28
ꢀ38.94
OH
8.53
7.33av
9.2
8.05av
9.1
8.64av
9.0
7.65av
4.7
4.14av
PteC6H5 ortho 7.57,7.41
PteC6H5 meta 6.94
7.57,7.50
7.17,7.15
6.84
7.28
6.93
6.93
1.05
e
7.52, 7.65 6.6
7.18, 7.16 6.72
6.14, 7.71 6.65
6.57, 7.15 6.25
6.99, 7.19 Overlapped 6.70, 7.93
6.45, 6.60 6.3
6.23, 6.95
6.38
PteC6H5 para
PeCH3
6.94
6.88
1.15
e
b
6.74
6.40
2.43
7.45
6.98
6.98
6.34
6.5
Overlapped
7.42
1.43
2.26
PeC6H5 ortho
PeC6H5 meta
PeC6H5 para
31P
e
e
7.47av
7.44av
7.42av
7.72av
7.44av
7.43av
7.8
7.0
7.0
7.82av
7.48av
7.50av
7.06
6.99
e(P(R)3)2
16.45
16.25av
ꢀ10.97
ꢀ3.30av
ꢀ1.29
8.03av
13.05
21.97av
24.60
39.15av
JPtP ¼ 2862
JPtP ¼ 2842
JPtP ¼ 2956
JPtP ¼ 3098
JPtP ¼ 3225
a
data from ref. [5].
not observed.
b
3.5. Theoretical and experimental NMR study of Pt(II)-
silsesquioxane complexes with phosphine ligands (4e7)
The 31P NMR signals for all complexes in the series are observed as
a singlet flanked by Pt satellites in a wide range of the chemical shifts.
The coupling constants of 4, 6 and 7 are in the typical range for square-
planar Pt(II) complexes having the phosphine ligands at trans position
(JPtP ¼ 2842 Hz (4), JPtP ¼ 3098 (6) Hz and JPtP ¼ 3225 Hz (7)). The
calculated 31P chemical shifts (ꢀ3.30, 8.03, 16.25, 21.97 and 39.15 ppm
for 40, 50, 3, 60, and70, respectively) generally well reproduce the trend of
the experimental signals at ꢀ10.97, ꢀ1.29, 13.05, 16.45 and 24.60 ppm
for the complexes 4, 5, 6, 1 and 7. It was found that the downfield shift
correlates with the electron density decrease on P (estimated by
increasing positive Mulliken charge of P atom). The presence of
electron-donating CH3 groups in the phosphine ligands increases the
electron density on P atoms, while the electron-withdrawing phenyl
rings decrease it and shifts the 31P chemical shifts to low field.
The experimental multinuclear NMR data of 4e7 were assigned
according to the calculations for the model compounds 40e70,
Table 1.
29Si NMR spectra of
4
and
6
show five signals
at ꢀ67.75; ꢀ65.60; ꢀ65.10; ꢀ57.87; ꢀ57.05 and ꢀ67.70; ꢀ65.03;
ꢀ64.80; ꢀ57.38; ꢀ56.87, respectively. The intensity ratio 2:1:1:1:2 is
typical for the studied series Pt(II)-silsesquioxane complexes (1e6)
and reveals their similar structures. Because of the rapid exchange of
the hydrogen atoms within the intramolecular hydrogen bonds, the
silicon atoms of the pairs (Si3 and Si5) and (Si2 and Si6) (Fig.1) become
equivalent and give rise to the corresponding signals. The calculations
are in accord with the interpretation that the downfield chemical
shifts are due to silicon atoms of silanol groups engaged in hydrogen
bonding and coordinated to Pt, and the upfield signals correspond to Si
atoms from the silsesquioxane skeleton.
In 1H NMR spectra of 4 and 6, a single signal was observed at low
magnetic field (9.2 and 9.0 ppm for 4 and 6, respectively) and it was
assigned to OH hydrogens similar to the other complexes 1, 2 and 5,
while in the spectrum of 7 this signal appears at 4.7 ppm. As we
discussed above a new set of weaker hydrogen bonds is formed in 7,
leading to upfield shiftof thesignal. Inthe series ofcomplexes studied
a correlation between the calculated O-H.O bond distances and the
1H NMR signal position was found. As can be seen the chemical shift
(4.7 < 8.5 < 9.0 < 9.1 < 9.2 ppm) moves downfield in the case of
stronger hydrogen bond in the complexes 7 < 1 < 6 < 5 < 4.
29
The calculated Si chemical shifts of 70 are very helpful for
assignment of 29Si NMR spectrum of 7. As can be seen from the
geometrical parameters significant rearrangement in the silses-
quioxane moiety occurs and it reflects on the 29Si NMR signals.
Although bad quality of the spectrum obtained the following peaks
could be distinguished: ꢀ67.6 (Si 2, Si 6); ꢀ60.2 (Si 1); ꢀ57.3 (Si 4, Si
7), ꢀ57.1 (Si3) and ꢀ49.9 (Si5). The upfield shift of Si1 signal could
be explained with higher electronic density of Si1 atom (q(Si1)
2.28 a.u.) than that in the other complexes 1e6 (q(Si1) w 2.29 a.u.).
In the complex 70 the O1 is not involved in a hydrogen bond and the
Si1-O1 bond length is shortened in contrast to that of complexes
1e6.
3.6. Formation energy of complexes 3e7
In order to characterize the stability of the series platinum
complexes their formation energies (DEf) were calculated using the
reaction equations given in Table 2.

30
N. Mintcheva et al. / Journal of Organometallic Chemistry 697 (2012) 23e32
Table 2
DFT/OPW91/6-31G(d) formation energies
DEf (kcal/mol) of the series Pt(II) complexes in benzene.
product
Reaction
DEf
(3)
2 trans-[PtI(Ph)(PEt3)2] þ 2(CH3)7Si7O9(OH)3 þ Ag2O / 2 trans-[Pt{O10Si7(CH3)7(OH)2}(Ph)(PEt3)2] þ 2AgI þ H2O
2 trans-[PtI(Ph)(PMe3)2] þ 2(CH3)7Si7O9(OH)3 þ Ag2O / 2 trans-[Pt{O10Si7(CH3)7(OH)2}(Ph)(PMe3)2] þ 2AgI þ H2O
2 trans-[PtI(Ph)(PMe2Ph)2] þ 2(CH3)7Si7O9(OH)3 þ Ag2O / 2 trans-[Pt{O10Si7(CH3)7(OH)2}(Ph)(PMe2Ph)2] þ 2AgI þ H2O
2 trans-[PtI(Ph)(PPh2Me)2] þ 2(CH3)7Si7O9(OH)3 þ Ag2O / 2 trans-[Pt{O10Si7(CH3)7(OH)2}(Ph)(PPh2Me)2] þ 2AgI þ H2O
trans-[Pt(OH)(Ph)(PPh3)2] þ (CH3)7Si7O9(OH)3 / trans-[Pt{O10Si7(CH3)7(OH)2}(Ph)(PPh3)2] þ H2O
0.82
(40)
(50)
(60)
(70)
ꢀ0.72
ꢀ5.86
ꢀ0.07
24.31
Similar to the experimental conditions the reagents and products
intense bands in the region 3190-2860 cmꢀ1 and obviously they
are overlapped.
in the reactions were solvated in benzene, by using continuum
solvation model, CPCM. The calculations revealed that the stability of
the complexes depends on the nature of phosphine ligands and it
decreases in the following order: 50 > 40 > 60 > 3 > 70, where the
The observed intense broad band centered at 1110 cmꢀ1
involves a number of IR bands (calc.1082,1046, 976 cmꢀ1 with large
intensities) and they are assigned to n(SieO) and d(SiOH) modes.
auxiliary ligands are as follow PMe2Ph
>
PMe3
>
PPh2Me
The strong band at 948 cmꢀ1 in the IR spectrum of 6 covers several
vibrations with different intensities. The most intense bands are
calculated at 925 cmꢀ1 and 924 cmꢀ1 and they are attributed to
> PEt3 > PPh3. According to
D
Ef data 50 is the most stable complex,
followed by 40, 60 and 3 complexes with relatively similar stability,
whereas 70 is least stable. Most probably the formation of two intra-
molecular hydrogen bonds in 1e6 stabilize the molecule, while the
lack of such bonding and repulsion of the bulky PPh3 ligands lead to
facile decomposition of 7. The results are in accord with the experi-
mentally observed stability of the complexes in solution and in solid
state.
d
(CCH) þ
(SieO(Pt)) þ
of the hydrogen-bonded silanol groups appears at lower frequen-
cies (857 cmꢀ1) as compared to
(SieO(Pt)) band and this trend is
in agreement with the larger SieO(H) bond lengths. A new band at
n
(C-C) vibrations of cyclo-C5H9 substituents and to
n
d
(PCH) vibrations, respectively. The (SieO(H)) band
n
n
705 cmꢀ1 in the IR spectrum of 6 is assigned to
d(SiOH). It is
observed in the range 702e708 cmꢀ1 for the other complexes (1, 4,
5, 7). This band is related to the type and strength of the SiOH$$$O
hydrogen intramolecular bonds and it is different in the ligand,
which forms dimmer structure by intermolecular H-bonds [23].
The calculated band at 496 cmꢀ1 with medium intensity is assigned
3.7. Vibrational analysis, theoretical and experimental study
Selected bands observed in the IR spectra of complexes 1, 4e7 and
their assignments according to the calculations are shown in Table 3.
The IR spectrum of trans-[Pt{O10Si7(cyclo-C5H9)7(OH)2}(Ph)
(PPh2Me)2] (6) was calculated and compared to the experimental
one, Fig. 4.
to n
(PteO) vibration and corresponds to the shoulder at 530 cmꢀ1
in the experimental spectrum of 6. The main peak observed at
501 cmꢀ1 (calc. 480 cmꢀ1) is due to torsion of the silsesquioxane
skeleton, t(SiOSiO). The same band position was found in the IR
spectrum of silsesquioxane [24].
In compliance with the calculations the broad band at
3451 cmꢀ1 (calc. 3464 cmꢀ1) is due to the
n(O6-H6) vibration of
The medium intensity band at 744 cmꢀ1 for 6 arises from the
stretching CeP vibration of the phosphine ligands. This signal is
observed in the range 737e741 cmꢀ1 for the other Pt-phosphine
complexes (Table 3). According to the calculations the medium
OH group engaged in the H-bond. The stretching vibration of
the second OH group (O7-H7) is observed at 3252 cmꢀ1 as
a shoulder owing to the stronger O1$$$H7 bonding and larger
O7-H7 bond length. The calculations suggested the
band to appear at lower frequency (2999 cmꢀ1
because of a coupling between (O7-H7) and (CeH) or method
deficiency. In the experimental IR spectra of 6 the bands at 2950
and 2861 cmꢀ1 were attributed to
(CeH) vibrations of cyclo-
C5H9 residue. The calculations predicted the (CeH) frequencies
n
(O7-H7)
band at 419 cmꢀ1 is assigned to the
n(PteP) vibration that was
)
probably
observed at 430 cmꢀ1 for 6 and in the range 418e448 cmꢀ1 for the
n
n
other complexes.
In summary the appearance of
and the (PteO) band at w 530 band is most probably related to the
O-monocoordinated silsequioxane to Pt(II).
d
(SiOH) band at w702e708 cmꢀ1
n
n
n
for cyclo-C5H9, phenyl and methyl groups to appear as low
Table 3
Selected calculated IR frequencies (cmꢀ1) of complex 6 compared to the experimental IR data of complexes 1, 4e7.
Band assignment
Frequency
6 calc
6 exp
1 exp
4 exp
5 exp
7 exp
n
n
(OeH)
(CeH)
3464 (2999)
3110
2997
1082
1046
976
925
924
916
870
729
702
655
496
481s
419
3451br,m 3252 sh
2950s
2861s
3200br,m
2949s
2867s
3400br,m
2949s
2865s
3250br,m
2950s
2865s
3400br,m
2949s
2865s
n(SieO) þ
d
(SiOH)
1110vs
1150vs
1100vs
1120vs
1100vs
d
d
(SiOH)
Overlapped (988sh)
948s
Overlapped
922s
Overlapped
949s
Overlapped
945s
Overlapped
940sh
(CCH) þ
n(CeC)
n
d
(SieO(Pt)) þ
d
(PCH)
(PCH) þ t(HCCH)ph
909 m
857 m
744 m
705 m
vw
530sh
501s
430 m
878 m
763 m
739 m
708 m
vw
520sh
505s
448
910 m
856 m
740 m
706 m
vw
520sh
503s
440
907 m
845
741 m
702 m
vw
534
503s
418
910 m
n
n
(SieO(H)) þ
(CeP)
d
(SiOH)
737 m
706 m
vw
523
502s
447
d
(SiOH)
n
n
(PteC)
(PteO)
t(SieOeSieO)
(PteP)
n
br, broad; sh, shoulder; vs, very strong; s, strong; vw, very weak; m, medium.

N. Mintcheva et al. / Journal of Organometallic Chemistry 697 (2012) 23e32
31
Fig. 4. Experimental (a) and calculated (b) (at DFT/OPW91/6-31G(d) level) IR spectra of trans-[Pt{O10Si7(cyclo-C5H9)7(OH)2}(Ph)(PPh2Me)2] (6).
4. Conclusions
PMe2Ph > PMe3 > PPh2Me > PEt3 > PPh3. The complexes 1, 2, 4e6
with O-monocoordinated silsesquioxane and two hydrogen bonds
O7eH7$$$O1 and O6eH6$$$O7, are characterized by the presence of
29Si NMR resonances with intensity ratio 2:1:1:1:2 and the downfield
29Si NMR signals are due to the silicon centers of SieOH and Si-OPt.
Changes in 29Si NMR spectrum and in the intensity ratio of the
signals observed for 7 are an indication of different chemical envi-
ronment of each silicon atom of silsesquioxane framework refined by
molecular modeling of the complex 70. The comparison of the
calculated NMR and IR parameters with the experimental spectro-
scopic data reveals good coincidence and thus confirmed the sug-
gested molecular structures.
Three new Pt(II)-silsesquioxane complexes (4, 6, 7) were prepared
and a series of Pt(II) complexes with different auxiliary mono-
phosphine
ligands,
trans-[Pt{O10Si7(cyclo-C5H9)7(OH)2}
(Ph)(L)2] (1: L ¼ PEt3, 4: L ¼ PMe3, 5: L ¼ PPhMe2, 6: L ¼ PPh2Me, 7:
L ¼ PPh3) was accomplished and discussed. The molecular structures
of Pt(II)-silsesquioxane complexes (4e7) were predicted by DFT/
OPW91/6-31G(d) calculations (ECPMWB60 for Pt(II)) of their
models (40e70), where the cyclo-C5H9 groups in silsesquioxanate
ligand are substituted by simpler CH3 groups. The method used was
verified by structure calculations of two complexes trans-[Pt
{O10Si7R7(OH)2}(Ph)(PEt3)2] (1: R ¼ cyclo-C5H9, 2: R ¼ iso-C4H9) with
known X-ray structures. Based on the combined experimental and
calculated 1H, 13C, 31P and 29Si NMR chemical shifts of 1 and 2,
a correlation structure - NMR parameters was derived and further
used for elucidation of the structures and NMR properties of analo-
gous Pt(II)-silsesquioxane complexes (4e7). It was found that the
alkyl substituents (cyclo-C5H9, iso-C4H9 and CH3) attached to Si atoms
slightly influence the geometry of trans-[Pt{O10Si7(R)7(OH)2}(Ph)(-
PEt3)2] (R ¼ cyclo-C5H9, iso-C4H9 and CH3) as well as the multinuclear
NMR parameters concerning silsesquioxane skeleton, phenyl and
auxiliary phosphine ligands. The geometry optimizations of 40e70
predicted square-planar configuration of the Pt(II) center which is
bonded to two trans phosphine ligands, phenyl ring and O-mono-
coordinated silsesquioxane. The structures 40e60 are stabilized by
two intramolecular hydrogen bondings in agreement with the
hydroxy 1H NMR signal at low magnetic field. A new set of weaker
hydrogen bonds is formed in 70 causing upfield shift of the 1H NMR
signal and low stability of the complex. The observed single 1H NMR
signal in case of two different hydrogen bonds in 4e6 was explained
by fast conformer transition A 4 B and switching of H-bonds. The
conformer exchange process was verified by calculations, predicting
close conformer energies and small energy barrier of the process.
According to the calculated formation energy of the series Pt(II)
complexes 3, 40e70, the stability of the complexes depends on the
nature of phosphine ligands and decreases in the order
Acknowledgements
This work was financially supported by Bulgarian Science Fund,
Grants DO-02-233 and DCVP-02-2. All calculations are performed
on the MADARA cluster of the Bulgarian Academy of Sciences.
Appendix. Supplementary material
Supplementary material associated with this article can be found,
in the online version, at doi:10.1016/j.jorganchem.2011.09.023.
References
[1] (a) R. Murugavel, A. Voigt, M.G. Walawalkar, H.W. Roesky, Chem. Rev. 96
(1996) 2205;
(b) H.C.L. Abbenhuis, Chem. Eur. J. 6 (2000) 25;
(c) R. Duchateau, Chem. Rev. 102 (2002) 3525;
(d) R. Duchateau, T.W. Dijkstra, R.A. van Santen, G.P.A. Yap, Chem. Eur. J. 10
(2004) 3979;
(e) R.W.J.M. Hanssen, R.A. van Santen, H.C.L. Abbenhuis, Eur. J. Inorg. Chem.
(2004) 675.
[2] (a) F.J. Feher, T.A. Budzichowski, R.L. Blanski, K.L. Weller, J.W. Ziller, Organo-
metallics 10 (1991) 2526;
(b) F.J. Feher, T.A. Budzichowski, Polyhedron 14 (1995) 3239;
(c) H.C.L. Abbenhuis, A.D. Burrows, H. Kooijman, M. Lutz, M.T. Palmer, R.A. van
Santen, A.L. Spek, Chem. Commun. (1998) 2627;
(d) V. Lorenz, A. Fischer, S. Giebmann, J.W. Gilje, Y. Gun’ko, K. Jacob,

32
N. Mintcheva et al. / Journal of Organometallic Chemistry 697 (2012) 23e32
F.T. Edelmann, Coord. Chem. Rev. 206e207 (2000) 321;
(e) K. Wada, N. Itayama, N. Watanabe, M. Bundo, T. Kondo, T. Mitsudo,
Organometallics 23 (2004) 5824;
(f) V. Lorenz, F.T. Edelmann, Z. Anorg. Allg. Chem. 630 (2004) 1147;
(g) E. Lucenti, F.J. Feher, J.W. Ziller, Organometallics 26 (2007) 75.
[14] (a) M. Dolg, U. Wedig, H. Stoll, H. Preuss, J. Chem. Phys. 86 (1987) 866;
(b) M. Dolg, H. Stoll, H. Preuss, R.M. Pitzer, J. Phys. Chem. 97 (1993) 5852.
[15] D. Andrae, U. Haeussermann, M. Dolg, H. Stoll, H. Preuss, Theor. Chim. Acta 77
(1990) 123.
[16] W.P. Oziminski, P. Garnuszek, E. Bednarek, J.Cz. Dobrowolski, Inorg. Chim.
Acta 360 (2007) 1902.
[3] (a) N. Maxim, P.C.M.M. Magusin, P.J. Kooyman, J.H.M.C. van Wolput, R.A. van
Santen, H.C.L. Abbenhuis, Chem. Mater. 13 (2001) 2958;
(b) A.R. Esker, B.A. Vastine, J. Deng, M.K. Ferguson, J.R. Morris, S.K. Satija,
B.D. Viers, Polym. Prepr 45 (1) (2004) 644;
(c) J.M. Mabry, A. Vij, B.D. Viers, W.W. Grabow, D. Marchant, S.T. Iacono,
P.N. Ruth, I. Vij, ACS Symp. Ser. 964 (2007) Science and Technology of Silicones
and Silicone-Modified Materials; (Chapter 18), p.290.
[4] (a) N. Mintcheva, M. Tanabe, K. Osakada, Organometallics 26 (2007) 1402;
(b) M. Tanabe, K. Mutou, N. Mintcheva, K. Osakada, Organometallics 27 (2008)
519;
(c) M. Tanabe, K. Mutou, N. Mintcheva, K. Osakada, J. Organomet. Chem. 696
(2011) 1211;
(d) N. Mintcheva, M. Tanabe, K. Osakada, Organometallics 30 (2011) 187.
[5] N. Mintcheva, M. Tanabe, K. Osakada, Organometallics 25 (2006) 3776.
[6] N. Mintcheva, M. Tanabe, K. Osakada, I. Georgieva, T. Mihailov, N. Trendafilova,
J. Organomet. Chem. 695 (2010) 1738.
[17] (a) K. Wolinski, J.F. Hinton, P. Pulay, J. Am. Chem. Soc. 112 (1990) 8251;
(b) G. Magyarfalvi, P. Pulay, J. Chem. Phys. 119 (2003) 1350.
[18] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb,
J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson,
H. Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino,
G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers,
K.N. Kudin, V.N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari,
A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J.M. Millam,
M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts,
R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski,
R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador,
J.J. Dannenberg, S. Dapprich, A.D. Daniels, O. Farkas, J.B. Foresman, J.V. Ortiz,
J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.02. Gaussian, Inc, Wallingford
CT, 2009.
[7] (a) C.R. Kistner, J.H. Hutchinson, J.R. Dolve, J.C. Storlie, Inorg. Chem. 2 (1963)
1255;
(b) H.C. Clark, L.E. Manzer, J. Organomet. Chem. 59 (1973) 411.
[8] T. Yoshida, T. Okano, S. Otuska, J. Chem. Soc. Dalton Trans. (1976) 993.
[9] P.P. Pescarmona, J.C. van der Waal, T. Maschmeyer, Eur. J. Inorg. Chem. (2004)
978.
[19] (a) V. Barone, M. Cossi, J. Phys. Chem. A 102 (1998) 1995;
(b) M. Cossi, N. Rega, G. Scalmani, V. Barone, J. Comput. Chem. 24 (2003) 669.
[20] A. Bondi, J. Phys. Chem. 68 (1964) 441.
[21] M. Kakeya, M. Tanabe, Y. Makamura, K. Osakada, J. Organomet. Chem. 694
(2009) 2270.
[10] http://www.chemcraftprog.com.
[22] (a) N. Wachter-Jurcsak, C.A. Detmer, Org. Lett. 1 (5) (1999) 795;
(b) M. Garcia-Viloca, A. González-Lafont, J.M. Lluch, Org. Lett. 3 (4) (2001) 589.
[23] L. Hongzhi Liu, K. Shin-ichi, T. Ryoji, O. Hiroyuki, U. Masafumi, J.Organomet.
Chem. 693 (2008) 1301.
[11] N.C. Handy, A.J. Cohen, Mol. Phys. 99 (2001) 403.
[12] J.P. Perdew, in: P. Ziesche, H. Eschig (Eds.), Electronic Structure of Solids ’91,
Akademie Verlag, Berlin, 1991, p. 11.
[13] G.A. Petersson, A. Bennett, T.G. Tensfeldt, M.A. Al-Laham, W.A. Shirley,
J. Mantzaris, J. Chem. Phys. 89 (1988) 2193.
[24] I.E. Davidova, L.A. Gribov, I.V. Maslov, V. Dufaud, G.P. Niccolai, F. Bayard,
J.M. Basset, J. Mol. Struct. 443 (1998) 89.