Synthesis, structure, DFT calculations, and full vibrational analysis of the prototypical thioformaldehyde complex mer-[W(CO)3(Me2PC2H4PMe 2)(η2-S=CH2)]

Article abstract of DOI:10.1039/b203195g

Reaction of PPN[W(CO)₃(dmpe)(SH)]2 (PPN = Ph₃PNPPh₃, dmpe = Me₂PC₂H₄PMe₂) with aqueous formaldehyde in the presence of trifluoroacetic acid gives the thioformaldehyde complex mer-[W(CO)₃(Me₂PC₂H₄PMe ₂)(η²-S=CH₂)]3 in almost quantitative yield. The isotopomer mer-[W(CO)₃(Me₂PC₂H₄PMe ₂)(η²-S=CD₂)]3-D₂ was obtained analogously. 3 has a slightly distorted pentagonal-bipyramidal structure with one carbonyl group, two P atoms and the C and S atoms of the thioformaldehyde ligand spanning the pentagonal plane. In solution, 3 is in equilibrium with fac-[W(CO)₃(Me₂PC₂H₄PMe ₂)(η²-S=CH₂)]4. The structures and energies of 3 and 4 and their rotamers 3′ and 4′ were calculated by DFT methods leading to a very good agreement with experimental data. FT-IR and FT-Raman spectra of 3 and 3-D₂ were recorded and assigned with the aid of DFT calculations. Calculated vibrational amplitudes emphasize the extensive mixing of modes, particularly in the low-wavenumber region. The HOMO of 3 can approximately be described as a lone-pair at sulfur and is 0.38 eV higher in energy than that of isomer 4. The LUMOs of both isomers are highly delocalized with that of 4 being 0.45 eV lower in energy. On this basis it is expected that electrophiles will add preferentially at the sulfur atom of 3 while nucleophiles if at all will add to the thiocarbonyl carbon of 4.

Full text of DOI:10.1039/b203195g

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis, structure, DFT calculations, and full vibrational
analysis of the prototypical thioformaldehyde complex
mer-[W(CO) (Me PC H PMe )(ꢀ2-S᎐CH )]†
᎐
3
2
2
4
2
2
Wolfdieter A. Schenk,*a Birgit Vedder,‡a Matthias Klüglein,a Damien Moignob
and Wolfgang Kiefer*b
a Institut für Anorganische Chemie, Universität Würzburg, Am Hubland, D-97074 Würzburg,
Germany. E-mail: Wolfdieter.Schenk@mail.uni-wuerzburg.de
b Institut für Physikalische Chemie, Universität Würzburg, Am Hubland, D-97074 Würzburg,
Germany. E-mail: Wolfgang.Kiefer@mail.uni-wuerzburg.de
Received 2nd April 2002, Accepted 2nd July 2002
First published as an Advance Article on the web 17th July 2002
Reaction of PPN[W(CO)3(dmpe)(SH)] 2 (PPN = Ph3PNPPh3, dmpe = Me2PC2H4PMe2) with aqueous formaldehyde
in the presence of trifluoroacetic acid gives the thioformaldehyde complex mer-[W(CO) (Me PC H PMe )(η2-S᎐
᎐
3
2
2
4
2
CH )] 3 in almost quantitative yield. The isotopomer mer-[W(CO) (Me PC H PMe )(η2-S᎐CD )] 3-D was obtained
᎐
2
3
2
2
4
2
2
2
analogously. 3 has a slightly distorted pentagonal-bipyramidal structure with one carbonyl group, two P atoms and
the C and S atoms of the thioformaldehyde ligand spanning the pentagonal plane. In solution, 3 is in equilibrium
with fac-[W(CO) (Me PC H PMe )(η2-S᎐CH )] 4. The structures and energies of 3 and 4 and their rotamers 3Ј and 4Ј
᎐
3
2
2
4
2
2
were calculated by DFT methods leading to a very good agreement with experimental data. FT-IR and FT-Raman
spectra of 3 and 3-D2 were recorded and assigned with the aid of DFT calculations. Calculated vibrational
amplitudes emphasize the extensive mixing of modes, particularly in the low-wavenumber region. The HOMO of 3
can approximately be described as a lone-pair at sulfur and is 0.38 eV higher in energy than that of isomer 4. The
LUMOs of both isomers are highly delocalized with that of 4 being 0.45 eV lower in energy. On this basis it is
expected that electrophiles will add preferentially at the sulfur atom of 3 while nucleophiles if at all will add to the
thiocarbonyl carbon of 4.
in particular density functional theory (DFT), has nowadays
Introduction
become almost routine. Indeed, the development of gradient-
corrected functionals and the use of small-core relativistic
effective core potentials (ECPs)15,16 have made it possible to
predict, often with impressive accuracy, geometries, bond
energies, vibrational spectra, NMR chemical shifts, activation
energies of chemical reactions and other important properties
of transition metal complexes.16–18
Thioaldehydes,2 and in particular thioformaldehyde3 are
typical examples of the so-called “double bond rule”: the
relatively weak π bonding involving 3p orbitals encourages
oligo- and poly-merization reactions to such an extent that
thioaldehydes RHC᎐S can not be isolated as monomers unless
᎐
they are stabilized by bulky substituents R. Transition metal
complexes of thioaldehydes are nevertheless well-known,4 and
even the parent thioformaldehyde has been stabilized as a
ligand in mononuclear complexes of titanium,5 zirconium,6
tantalum,7 rhenium,8 ruthenium,1 osmium,9 cobalt,10 and
rhodium.11 The molecular properties of the isolated molecule
H C᎐S are known from gas-phase12 and matrix data13 whereas
Here we report the synthesis of a new complex of thio-
formaldehyde, mer-[W(CO) (dmpe)(η2-S᎐CH )] (dmpe
=
᎐
3
2
Me2PC2H4PMe2) and its isotopomer, mer-[W(CO)3(dmpe)-
(η2-S᎐CD )], along with an X-ray structure determination and
᎐
2
full vibrational analysis, the latter backed by DFT calculations
which also include rotamers and geometrical isomers.
᎐
2
physical characterization of the complexes of thioformal-
dehyde remained restricted largely to NMR spectroscopy and
X-ray structure determinations.4
Results
Vibrational spectroscopy provides direct information on
the force field of a molecule or ion and thus on the strength
of chemical bonds. However, for larger molecules of low
symmetry it is exceedingly difficult to assign the multitude
of absorptions (or Raman scattering lines) to the respective
normal modes. Selective isotopic substitution is a reliable
tool to identify the participation of an individual group in a
molecular vibration.14 Even then, as the result of the extensive
mixing of normal modes of like symmetry, it can be very
difficult to arrive at an unambiguous assignment of the
vibrational spectrum. The use of quantum chemical methods,
The reaction of pentacarbonyl-hydrogensulfido-tungstate 119
with 1,2-bis(dimethylphosphino)ethane (dmpe) in THF
proceeded with visible gas evolution and formation of the
CO-substitution product 2 in essentially quantitative yield
(Scheme 1). 2 was obtained as a bright yellow, slightly
air-sensitive microcrystalline powder which, according to its
ionic nature, is soluble only in polar organic media such as
THF, acetone, or acetonitrile. A triplet at Ϫ3.78 ppm in the 1H
NMR spectrum, upfield from the SH signal of 1,19 revealed
the presence of the SH ligand. A singlet with tungsten satellites
in the 31P NMR spectrum and three intense CO stretching
absorptions in the IR spectrum are diagnostic of the facial
coordination geometry shown in Scheme 1.
† The coordination chemistry of the C᎐S function, part 16. For part 15
᎐
Addition of aqueous formaldehyde and trifluoroacetic acid
to a solution of 2 in THF resulted in a spontaneous colour
see ref. 1.
‡ Née Wolfsberger.
DOI: 10.1039/b203195g
J. Chem. Soc., Dalton Trans., 2002, 3123–3128
This journal is © The Royal Society of Chemistry 2002
3123

View Article Online
Table
1
Selected bond lengths (Å) and angles (Њ) for mer-
[W(CO)3(dmpe)(η2-S᎐CH2)] 3
᎐
W–S
2.4937(8)
2.4716(7)
2.4992(8)
2.290(3)
2.011(3)
2.021(3)
W–C(13)
S–C(1)
C(11)–O(1)
C(12)–O(2)
C(13)–O(3)
2.031(3)
1.745(3)
1.161(4)
1.142(4)
1.150(4)
W–P(1)
W–P(2)
W–C(1)
W–C(11)
W–C(12)
S–W–C(1)
P(1)–W–S
P(2)–W–S
P(1)–W–P(2)
C(12)–W–P(1)
C(1)–W–C(12)
42.52(8)
161.94(3)
83.25(3)
78.82(3)
78.39(9)
77.18(12)
C(11)–W–C(12)
C(11)–W–C(13)
C(12)–W–C(13)
W–S–C(1)
89.18(13)
178.35(12)
91.45(13)
62.49(11)
74.99(12)
S–C(1)–W
Scheme 1 Synthesis of the thioaldehyde complex 3. Reagents and
conditions: (i) Me2PC2H4PMe2/THF/60 ЊC/45 min. (ii) H2O/CF3CO2H/
CH2O/THF/20 ЊC/5 min.
change to deep yellow. After chromatographic work-up the
thioformaldehyde complex 3 was isolated in excellent yield
as a deep yellow crystalline powder. The isotopomer [W(CO)3-
(dmpe)(η2-S᎐CD )] (3-D ) was obtained in the same way by
᎐
2
2
using
a solution of [D2]formaldehyde in D2O. The IR
spectrum of a solid sample of 3 exhibited three CO stretching
absorptions in the relative intensity ratio medium–strong–
very strong, which is a clear indication of the meridional
arrangement of the CO ligands. Spectra taken from THF
solutions revealed an additional absorption at 1980 cmϪ1
,
hinting at the existence of a second isomer of 3. The 31P NMR
spectra were also in accord with the presence of a mixture
of isomers. The main component gave rise to two doublets
with tungsten satellites caused by two unequal J(183W–31P)
couplings. The minor isomer was represented by a singlet with
tungsten satellites. In the 1H NMR spectra the methylene
group of the major isomer appeared as a doublet of doublets at
δ = 3.98 while the CH2 signal of the minor isomer was visible
as a triplet at δ = 3.91. Both signals were absent in the spectra of
3-D2. The 13C resonance of the CH2 group appeared at δ = 39.7,
in a shift range typical of η2-coordinated thioformaldehyde;8
η1(S)-coordinated thioaldehydes have resonances at δ = 200.20
Thus it was established that in solution, 3 is in equilibrium with
the facial isomer 4 (Scheme 2).
Fig. 1 ORTEP44 diagram of mer-[W(CO)3(dmpe)(η2-S᎐CH2)] (3).
Thermal ellipsoids drawn at the 50% level, hydrogen atoms except those
at C(1) omitted.
᎐
formaldehyde S,S-dioxide, [W(CO) (dppm)(η2-O S᎐CH )]
᎐
3
2
2
(dppm = Ph2PCH2PPh2).23
In order to gain a deeper insight into the bonding situation in
3 and 4 we have carried out some DFT calculations on both
isomers. Four calculated structures representing local minima
are shown in Figs. 2 and 3, calculated bond distances and angles
are compiled in Table 2.
Scheme 2 Equilibrium between 3 and its facial isomer 4.
The structure of 3 was determined by X-ray crystallography;
Fig. 1 shows an ORTEP diagram. The tungsten atom resides
in the centre of a distorted pentagonal bipyramid with the
thioformaldehyde ligand occupying two adjacent sites. The
largest angle deviation within the pentagonal base is associated
with the three-membered W–S–C ring, all other angles between
Fig. 2 Calculated structures and energies of mer-[W(CO)3(dmpe)(η2-
᎐
S᎐CH2)] (3, 3Ј). Values in kJ molϪ1 are relative to the preferred form 3.
neighboring bonds in this plane are in the range of 80
3Њ
The calculations predicted that 4 is slightly lower in energy
(by ca. 2 kJ molϪ1) than 3 while experimentally it is actually
5 kJ molϪ1 higher. Such small differences, however, are well
within the uncertainty of the method. The orientation of the
thioformaldehyde ligand was correctly predicted, i.e. 3Ј is ca. 15
kJ molϪ1 higher in energy than 3. The calculated bond distances
and angles of 3 agree very well with experimental values. The
largest discrepancy is associated with the W–S bond, the
length of which was overestimated by 0.066 Å. Even the trend
in W–P bond lengths, r{W–P(2)} > r{W–P(1)}, was correctly
reproduced. For the facial isomer 4 the rotamer 4Ј is again
(Table 1). The W–P bonds are slightly different in length. The
W–S bond length equals that in the η1-thiobenzaldehyde
complex [W(CO) (S᎐CHPh)],21 and the W–C(1) distance
᎐
5
compares well with the length of metal–carbon bonds in
tungsten–alkene complexes.22 The carbon–sulfur bond is much
longer than in free thioformaldehyde3 and equal in length to
that in the rhenium complex [CpRe(NO)(PPh )(η2-S᎐CH )]ϩ.8
᎐
3
2
Of the two possible orientations of the H C᎐S ligand the
᎐
2
one with the sulfur atom pointing away from the equatorial
carbonyl ligand seems to be preferred. The same geometrical
situation was found in a closely analogous complex of thio-
disfavored, in this case by ca. 20 kJ molϪ1
.
3124
J. Chem. Soc., Dalton Trans., 2002, 3123–3128

View Article Online
Table 2 Calculated bond lengths (Å) and angles (Њ) for [W(CO)3-
(dmpe)(η2-S᎐CH2)]
᎐
3
3Ј
4
4Ј
W–S
W–P(1)
W–P(2)
W–C(1)
W–C(11)
W–C(12)
W–C(13)
S–C(1)
C(11)–O(1)
C(12)–O(2)
C(13)–O(3)
2.560
2.482
2.538
2.309
2.026
2.004
2.028
1.759
1.180
1.179
1.179
2.571
2.490
2.525
2.300
2.026
2.019
2.026
1.753
1.180
1.175
1.180
2.568
2.547
2.541
2.330
1.998
1.997
1.997
1.752
1.181
1.180
1.180
2.574
2.540
2.534
2.332
2.000
2.014
1.994
1.740
1.182
1.175
1.180
Fig. 4 Calculated vibrational amplitudes for ν(W–C) of 3.
S–W–C(1)
P(1)–W–S
P(2)–W–S
P(1)–W–P(2)
C(12)–W–P(1)
C(1)–W–C(12)
C(11)–W–C(12)
C(11)–W–C(13)
C(12)–W–C(13)
W–S–C(1)
42.0
161.4
81.6
79.9
83.5
73.0
90.6
177.6
91.2
61.3
76.7
41.8
160.9
120.0
79.0
83.6
119.1
90.8
177.0
91.5
60.8
41.6
85.3
80.8
79.1
87.5
72.4
98.4
84.1
81.8
61.9
76.5
41.2
89.2
117.7
80.0
92.1
118.2
92.6
88.1
83.9
61.9
76.9
S–C(1)–W
77.4
Fig. 5 Calculated vibrational amplitudes for ν(W–S) of 3.
compared to those of 3 (Table 4). Indeed, the wavenumber of
the calculated symmetric CO stretch of isomer 4 exactly
matched the observed absorption.
Discussion
Fig. 3 Calculated structures and energies of fac-[W(CO)3(dmpe)(η2-
᎐
S᎐CH2)] (4, 4Ј). Values in kJ molϪ1 are relative to the preferred form 3.
The reaction of 1 with aromatic aldehydes in the presence of
acid was one of the early syntheses of transition metal com-
plexes of thioaldehydes. It was limited to benzaldehydes bearing
electron-releasing substituents in the para position and only
gave complexes of η1(S)-coordinated thioaldehydes.19 Since
the formation of η2(C,S) complexes should be favored with
increasing electron density at the central metal atom — this was
inter alia observed in a series of related dithioester complexes
A comparison of the infrared and Raman spectra of 3 and
3-D2 allowed the ready identification of fundamental modes
involving the three-membered W–S–C ring. Aside from the
ν(CD2) vibrations at 2255 and 2155 cmϪ1, sizeable isotopic
shifts were detected for absorptions at 785 and 422 cmϪ1 which
were shifted to 755 and 406 cmϪ1 in 3-D2, respectively. With
regard to wavenumber and isotopic shift the higher of the two
corresponds very well with the asymmetric ν(C–S) of thiirane
(651 cmϪ1, shifted to 618 cmϪ1 in [D4]thiirane24). On this basis
these two absorptions can confidently be assigned to ν(C–S)
and ν(W–C), respectively. The complete vibrational spectra
were finally compared with DFT calculations of all funda-
mental modes of both isotopomers which led to a quite satis-
factory agreement between measured and calculated data
(Table 3). It should be added here that any assignment which
describes a certain vibration of a larger molecule as originating
from the motion of but two or three atoms is a gross simplifica-
tion. Particularly in the low-wavenumber region extensive coup-
ling occurs. To underscore this we have calculated the ampli-
tudes of some of the vibrations of 3. Figs. 4 and 5 show the
results for ν(W–C) (calculated at 401 cmϪ1) and ν(W–S) (calcu-
lated at 278 cmϪ1), respectively. ν(W–C) represents a largely
unperturbed bond stretching motion, with perhaps some con-
tribution from WCO bending. The ν(W–S) mode, however, is
extensively coupled with twisting and deformation motions of
the backbone of the dmpe ligand.
of the type [W(CO)n(PR1 )5 (S᎐C(R2)(SR3))]25 — we looked
᎐
3
Ϫ n
for appropriate ways to modify reagent 1. Indeed, the lability of
CO ligands on 1 and their facile exchange for tertiary phos-
phines has been noted on several occasions,26,27 and accordingly
the reaction with the bidentate donor dmpe proceeded in excel-
lent yield. The nucleophilicity of 2 is much higher than that of 1
leading to a clean reaction with aqueous formaldehyde to give
3. To the best of our knowledge this is the first mononuclear
thioformaldehyde complex of a Group 6 metal. A binuclear
complex, [Cp W (CO) (µ-S᎐CH )], has been obtained from
᎐
2
2
3
2
[Cp2W2(CO)6(µ-S)] and diazomethane.28
The spectroscopic and structural data of 3 allow some clues
with regard to the ligand properties of side-on coordinated
thioformaldehyde. The fairly high CO stretching vibrations
indicate that H C᎐S is a good π acceptor ligand, comparable to
᎐
2
maleic acid esters,29 but not as good as CS2 30 or SO2.31 As pre-
dicted by the Dewar–Chatt–Duncanson model, a high degree
of charge transfer into the π* orbital leads to a pronounced
lengthening of the C–S bond.32 From the C–S distance in 3
(1.745 Å, Table 1) one can estimate a C–S bond order of 1.35,
taking the C–S bond lengths in thioformaldehyde (1.614 Å)12
and thiirane (1.815 Å)33 as standard values for a double and
single bond, respectively. A similar interpolation based on the
In order to further corroborate the assignment of the add-
itional CO stretching absorption observed in solution, some
fundamental vibrations of 3Ј, 4, and 4Ј were calculated and
J. Chem. Soc., Dalton Trans., 2002, 3123–3128
3125

View Article Online
Table 3 Selected measured and calculated vibrations (cmϪ1) of mer-[W(CO)3(dmpe)(η2-S᎐CH2)] 3 and mer-[W(CO)3(dmpe)(η2-S᎐CD2)] 3-D2 and
᎐
᎐
their assignments
[W(CO)3(dmpe)(η2-S᎐CH2)] (3)
[W(CO)3(dmpe)(η2-S᎐CD2)] (3-D2)
᎐
᎐
R
IR
DFT
R
IR
DFT
Assignment
2257(w)
2157(w)
1991(s)
1905(s)
2255(w)
2155(w)
1990(s)
1915(s)
1850(vs)
2330
2200
1986
1928
1890
ν(CD2)
ν(CD2)
ν(CO)
ν(CO)
ν(CO)
ω(CH2)
ν(CS)
τ(CH2)
ω(CD2)
τ(CD2)
δ(WCO)
1991(s)
1905(s)
1990(s)
1915(s)
1850(vs)
1986
1928
1890
975
800
777
786(m)
777(w)
785(m)
771(w)
754(m)
755(m)
763
720(s)
708(s)
605(s)
585(s)
565(m)
541(m)
534(w)
522(w)
509(m)
459(s)
449(s)
442(w)
416(m)
726
720
611
605
570
537
519
605(s)
585(s)
611
605
τ(CD2)
δ(WCO)
543(m)
534(w)
522(w)
514(m)
460(m)
444(m)
537
521
514(m)
461(s)
449(s)
442(s)
416(m)
511(m)
460(m)
444(m)
471
454
470
454
ν(W–CO)
δ(WCO)
416(m)
422(sh)
440
410
401
326
309
288
278
416(m)
439
410
385
326
309
288
278
421(m)
347(m)
328(s)
307(s)
262(s)
403(m)
346(m)
328(s)
306(s)
261(s)
406(m)
ν(W–C)
ν(W–P)
ν(W–P)
δ(CPC)a
ν(W–S)b
a Strongly coupled with ν(W–S). b Strongly coupled with CH2–CH2 torsion.
Table 4 Comparison of selected calculated vibrations (cmϪ1) of mer-
[W(CO)3(dmpe)(η2-S᎐CH2)] 3, 3Ј and fac-[W(CO)3(dmpe)(η2-S᎐CH2)]
therefore equal in length. While the length of the W–S bond
does not change on going from 3 to 4, we observe a notable
weakening of the W–C(1) bond. This might indicate that
nucleophilic addition and ring opening, if possible at all,
should be more feasible for the facial isomer 4.
᎐
᎐
4, 4Ј and their assignments
3
3Ј
4
4Ј
Assignment
In order to gain some deeper insight into the chemical
reactivity the frontier orbitals of these thioformaldehyde com-
plexes were calculated (Figs. 6 and 7). For both isomers the
1986
1928
1890
800
1991
1940
1889
806
1981
1922
1910
810
1988
1936
1907
824
ν(CO)
ν(CO)
ν(CO)
ν(CS)
401
326
309
278
402
325
307
286
390
318
302
269
382
326
309
273
ν(WC)
ν(WP)
ν(WP)
ν(WS)
C–S stretching frequencies of thioformaldehyde12 (1059 cmϪ1),
3, and thiirane gives a C–S bond order of 1.33. Although the
exact match of the two estimates might be somewhat fortuitous,
it certainly underscores the correlation between structure,
molecular vibrations, and bonding. It might be added here that
ab initio calculations on the model compound [Fe(CO)2-
(PH ) (η2-S᎐CH )] have indeed shown that the metal–ligand
᎐
3
2
2
bond energy essentially arises from the back-bonding contri-
bution.34 The preferred orientation of the thioformaldehyde
ligand in 3 and 4 maximizes the overlap of the HOMO of
the respective metal complex fragment, fac- or mer-[W(CO)3-
(dmpe)], which is polarized towards the better π-accepting CO
ligands,35 and the π* orbital of thioformaldehyde, which has its
largest coefficient at carbon.36 The situation here is thus similar
to that found for the rhenium complex [CpRe(NO)(PPh3)-
(η2-S᎐CH )]ϩ.8
The W–P(1) bond in 3 is marginally shorter than W–P(2).
This means that the structural trans influence37 of the side-on
coordinated thioformaldehyde ligand is only slightly less than
that of the tightly bound CO group. The DFT calculations
which were used here tend to overestimate this effect but
indicate its presence also in the rotamer 3Ј (Table 2). In the
facial isomer 4 both W–P bonds are in equivalent positions and
Fig. 6 Calculated HOMO (left) and LUMO (right) of the preferred
conformation of mer-[W(CO)3(dmpe)(η2-S᎐CH2)] (3).
᎐
LUMO appears highly delocalized and does not give any hint at
a preferred site for nucleophilic attack. The HOMO, however,
consists mainly of a p orbital at the sulfur atom with a strongly
antibonding contribution from one of the occupied d orbitals
at tungsten. On this basis one has to expect a facile electrophilic
attack at sulfur as the preferred mode of reactivity. At least for
uncharged complexes of thioformaldehyde there are many
known examples of electrophilic alkylations.5,6,9–11 The HOMO
of 3 is 0.38 eV higher in energy than that of 4 which means that
the meridional isomer 3 should be somewhat more reactive in
this regard. The LUMO energies follow the opposite trend
᎐
2
3126
J. Chem. Soc., Dalton Trans., 2002, 3123–3128

View Article Online
Table 5 Crystal data and structure refinement for mer-[W(CO)3-
(dmpe)(η2-S᎐CH2)] 3
᎐
Empirical formula
Formula weight
Temperature/K
Wavelength/Å
Crystal system, space group
a/Å
C10H18O3P2SW
464.09
130(2)
0.71073
Monoclinic, P21/n
9.1843(5)
b/Å
c/Å
12.4238(7)
13.6206(7)
β/Њ
90.483(1)
1554.11(15)
4, 1.984
Volume/Å3
Z, Calculated density/Mg mϪ3
Absorption coefficient/mmϪ1
Reflections collected/unique
Final R indices [I > 2σ(I )]
R indices (all data)
7.767
25623/3414
R1 = 0.019, wR2 = 0.044
R1 = 0.021, wR2 = 0.045
Fig. 7 Calculated HOMO (left) and LUMO (right) of the preferred
conformation of fac-[W(CO)3(dmpe)(η2-S᎐CH2)] (4).
᎐
ation and diffuse function was used [D95 ϩ (d)] for P, C, H, S
which again marks isomer 4 as the one for which nucleophilic
additions, if possible at all, should be preferred.
and O atoms (DFT2).42
Preparation of PPN[W(CO)3(dmpe)(SH)], 2
Conclusions
A solution of PPN[W(CO)5(SH)] 1 (0.90 g, 1.00 mmol) and
dmpe (0.20 mL, 0.18 g, 1.20 mmol) in THF (20 cm3) was kept at
60 ЊC for 45 min. The solvent was removed under reduced
pressure and the residue washed repeatedly with Et2O and
hexanes giving 2 as a deep yellow crystalline powder (0.97 g,
98%), mp 34 ЊC (Found: C, 54.49; H, 4.83; N, 1.43; S, 3.23%.
C45H47NO3P4SW requires C, 54.61; H, 4.79; N, 1.42; S, 3.24%).
νmax/cmϪ1 (CO) 1894 (vs), 1799 (s) and 1752 (s) (THF).
δH (CD3CN) Ϫ3.78 [1 H, t, J(HP) 8.7 Hz, WSH], 1.34 [6 H,
d, J(HP) 7.1 Hz, PCH3], 1.42 [6 H, d, J(HP) 6.8 Hz, PCH3],
1.30–1.80(m,4H,PCH2)and7.43–7.68(m,30H,Ph);δC(CD3CN)
12.4 [d, J(CP) 20 Hz, PCH3], 18.4 [d, J(CP) 25 Hz, PCH3], 31.2
[vt, J(CP) ϩ J(CPЈ) 40 Hz, PCH2], 127.4–134.5 (m, Ph), 215.9
[t, J(CP) 6 Hz, CO] and 223.4 [dd, J(CP) 7, 38 Hz, CO]; δP
(CD3CN) 6.8 [s, J(PW) 198 Hz, dmpe] and 20.9 (s, PPN).
Condensation of the highly nucleophilic complex [W(CO)3-
(dmpe)(SH)]Ϫ 2 with formaldehyde in the presence of acid
provides an easy access to the thioformaldehyde complex mer-
[W(CO) (Me PC H PMe )(η2-S᎐CH )] 3 and its isotopomer
᎐
3
2
2
4
2
2
mer-[W(CO) (Me PC H PMe )(η2-S᎐CD )] 3-D . A complete
᎐
3
2
2
4
2
2
2
vibrational analysis — facilitated by the absence of large
organic substituents — was undertaken with the aid of
DFT calculations. The results confirm the notion that the
bonding between metal atom and thioformaldehyde ligand
is appropriately described in terms of the Dewar–Chatt–
Duncanson model. Due to extensive π-backbonding the C–S
bond order is reduced to 1.33 which is also in accord with
structural data. Taken together it is demonstrated here that
a
combination of structural, spectroscopic and modern
theoretical methods provides a detailed understanding of
bonding and reactivity of transition metal–organic complexes.
Preparation of mer-[W(CO) (dmpe)(ꢀ2-S᎐CH )], 3 and
᎐
3
2
mer-[W(CO) (dmpe)(ꢀ2-S᎐CD )], 3-D
᎐
3
2
2
Experimental
To a solution of PPN[W(CO)3(dmpe)(SH)] 2 (0.50 g, 0.50
mmol) in THF (10 cm3) was added a 40% solution of formalde-
hyde in water (0.10 cm3, 1.33 mmol) and CF3CO2H (0.04 cm3,
0.06 g, 0.52 mmol). The solvent was removed under reduced
pressure, the residue dissolved in dichloromethane and
chromatographed over silica using dichloromethane–acetone
(40 : 1) as eluent. A broad yellow band was collected, the
solvent evaporated, and the remaining yellow oil washed
repeatedly with hexanes giving 3 as an orange-colored crystal-
line solid (0.21 g, 90%), mp 74 ЊC (dec.) (Found: C, 26.15; H,
3.86; S, 6.45%. C10H18O3P2SW requires C, 25.88; H, 3.91; S,
6.91%). νmax/cmϪ1 (CO) 1996 (w), 1914 (m) and 1869 (s) (THF).
δH (C6D6) 0.90–1.23 (m, 4 H, PCH2) 1.18 [6 H, d, J(HP) 8.4 Hz,
PCH3], 1.41 [6 H, d, J(HP) 8.8 Hz, PCH3] and 3.98 [dd, 2 H,
J(HP) 2.8, 0.5 Hz, SCH2]; δC (C6D6) 18.2 [d, J(CP) 28 Hz,
PCH3], 19.6 [d, J(CP) 28 Hz, PCH3], 27.9 [dd, J(CP) 28, 14 Hz,
PCH2], 31.8 [dd, J(CP) 30, 17 Hz, PCH2], 39.7 (s, SCH2), 197.7
[dd, J(CP) 8, 4 Hz, CO] and 214.0 [dd, J(CP) 13, 8 Hz, CO]; δP
(C6D6) 12.4 [d, J(PP) 15 Hz, J(PW) 222 Hz] and 18.7 [d, J(PP)
15 Hz, J(PW) 205 Hz]. Additional signals are due to the facial
isomer 4: νmax/cmϪ1 (CO) 1980 (w). δH (C6D6) 0.92 [d, J(HP)
8.1 Hz, PCH3] and 3.91 [t, J(HP) 1.1 Hz, SCH2]; δP (C6D6) 4.9
[s, J(PW) 200 Hz].
PPN[W(CO)5(SH)] was prepared according to literature
methods.19 Other chemicals were obtained commercially and
used without further purification.
All experiments were carried out under N2 in dried and
deoxygenated solvents. Reactions were routinely monitored by
IR and NMR spectroscopy. Chromatographic separations were
carried out using silica (Merck, grain size 0.06–0.20 mm) as
stationary phase.
IR spectra were recorded on a Bruker IFS-25 spectrometer
with the samples prepared either as Nujol mulls between KBr
plates or THF solutions in NaCl cells [ν(CO) region]. Raman
spectra were recorded on a Bruker IFS-120 spectrometer
equipped with a FRA-106 Raman module, a CaF2-beam split-
ter and a germanium diode detector. Solid samples were placed
in a 5 mm NMR tube and irradiated with the 1064 nm emission
of a Nd–YAG laser. NMR spectra were recorded on a Jeol
JNM-LA 300 instrument. 1H (300.4 MHz) and 13C (75.45
MHz) chemical shifts are reported relative to internal TMS; 31
P
(121.5 MHz) chemical shifts are relative to external 85% H3PO4
with the deuterium signal of the solvent serving as lock and
internal reference. X-Ray data were collected on a Bruker
Smart-Apex CCD diffractometer using Mo-Kα radiation.
Analyses were carried out by the Microanalytical Laboratory
of the Institut für Anorganische Chemie.
3-D2 was obtained by employing CD2O in D2O. Identical
spectra were obtained except for the missing signals of the
SCH2 group.
The DFT calculations were performed using Gaussian 9838
and Becke’s 1988 exchange functional39 in combination with
the Perdew–Wang 91 gradient-corrected correlation functional
(BPW91).40 The Los Alamos effective core potential plus
double zeta (LANL2DZ)41 was employed for tungsten, whereas
the Dunning–Huzinaga full double zeta basis set with polariz-
Structure determination of mer-[W(CO) (dmpe)(ꢀ2-S᎐CH )] 3
᎐
3
2
The crystal data for 3 are summarized in Table 5. The structure
was solved using the direct methods provided within SHELXS-
J. Chem. Soc., Dalton Trans., 2002, 3123–3128
3127

View Article Online
9743 and refined by full-matrix least squares using SHELXL-
97.43
CCDC reference number 160647.
See http://www.rsc.org/suppdata/dt/b2/b203195g/ for crystal-
lographic data in CIF or other electronic format.
Lipkowitz and D. B. Boyd, VCH, New York, 1996, vol. 8,
pp. 145–202.
18 D. Moigno, W. Kiefer, J. Gil-Rubio and H. Werner, J. Organomet.
Chem., 2000, 612, 125–132; D. Moigno, W. Kiefer, B.
Callejas-Gaspar, J. Gil-Rubio and H. Werner, New J. Chem., 2001,
25, 1389–1397.
19 R. G. W. Gingerich and R. J. Angelici, J. Am. Chem. Soc., 1979, 101,
5604–5608.
20 W. A. Schenk, T. Stur and E. Dombrowski, J. Organomet. Chem.,
1994, 472, 257–273.
21 H. Fischer, K. H. Fluck and C. Troll, Chem. Ber., 1992, 125,
2675–2680.
22 F. W. Grevels, M. Lindemann, R. Benn, R. Goddard and C. Krüger,
Z. Naturforsch., Teil B, 1980, 35, 1298–1309; H. Berke, G. Huttner,
C. Sontag and L. Zsolnai, Z. Naturforsch., Teil B, 1985, 40, 799–807;
F. W. Grevels, J. Jacke, P. Betz, C. Krüger and Y. H. Tsay,
Organometallics, 1989, 8, 293–298.
23 W. A. Schenk, K. Nielsen, N. I. Burzlaff and M. Hagel, Inorg.
Chem., 2002, 41, 1079–1085.
24 W. D. Allen, J. E. Bertie, M. V. Falk, B. A. Hess, G. B. Mast,
D. A. Othen, L. J. Schaad and H. F. Schaefer, J. Phys. Chem., 1986,
84, 4211–4227.
25 W. A. Schenk, D. Rüb and C. Burschka, J. Organomet. Chem., 1987,
328, 287–303.
Acknowledgements
This work was generously supported by the Deutsche
Forschungsgemeinschaft (SFB 347, Projects B3 and C2) and
the Fonds der Chemischen Industrie.
References
1 Part 15: N. Kuhnert, N. Burzlaff, E. Dombrowski and W. A. Schenk,
Z. Naturforsch., Teil B, 2002, 57, 259–274.
2 W. M. McGregor and D. C. Sherrington, Chem. Soc. Rev., 1993, 22,
199–204.
3 H. W. Kroto, Chem. Soc. Rev., 1982, 11, 435–491; J. Fabian,
R. Mayer, P. Cársky and R. Zahradnik, Z. Chem., 1985, 25, 50–61.
4 H. Fischer, R. Stumpf and G. Roth, Adv. Organomet. Chem., 1999,
43, 125–196.
5 J. W. Park, L. M. Henling, W. P. Schaefer and R. H. Grubbs,
Organometallics, 1990, 9, 1650–1656.
6 S. L. Buchwald, R. B. Nielsen and J. C. Dewan, J. Am. Chem. Soc.,
1987, 109, 1590–1591.
26 D. J. Darensbourg, A. Rokicki and R. Kudaroski, Organometallics,
1982, 1, 1161–1166.
27 D. J. Darensbourg, K. M. Sanchez and J. Reibenspies, Inorg. Chem.,
1988, 27, 3636–3643.
7 J. E. Nelson, G. Parkin and J. E. Bercaw, Organometallics, 1992, 11,
2181–2189.
28 M. Herberhold, W. Jellen and H. H. Murray, J. Organomet. Chem.,
1984, 270, 65–70.
29 W. A. Schenk and H. Müller, Chem. Ber., 1982, 115, 3618–3630.
30 W. A. Schenk, T. Schwietzke and H. Müller, J. Organomet. Chem.,
1982, 232, C41–C47.
31 W. A. Schenk and F.-E. Baumann, J. Organomet. Chem., 1983, 256,
261–276.
32 D. M. P. Mingos, J. Organomet. Chem., 2001, 635, 1–8; G. Frenking,
J. Organomet. Chem., 2001, 635, 9–23; U. Pidun and G. Frenking,
Organometallics, 1995, 14, 5325–5336; G. Frenking and N. Fröhlich,
Chem. Rev., 2000, 100, 717–774.
33 K. Okiye, C. Hirose, D. G. Lister and J. Sheridan, Chem. Phys. Lett.,
1974, 24, 111–113.
34 M. Rosi, A. Sgamellotti, F. Tarantelli and C. Floriani, Inorg. Chem.,
1988, 27, 69–73.
35 T. A. Albright, J. K. Burdett and M. H. Whangbo, Orbital
Interactions in Chemistry, Wiley, New York, 1985, p. 277–294.
36 V. P. Rao, J. Chandrasekhar and V. Ramamurthy, J. Chem. Soc.,
Perkin Trans. 2, 1988, 647–659.
37 T. G. Appleton, H. C. Clark and L. E. Manzer, Coord. Chem. Rev.,
1973, 10, 335–422; E. M. Shustorovich, M. A. Porai-Koshits and
Y. A. Buslaev, Coord. Chem. Rev., 1975, 17, 1–98; J. K. Burdett and
T. A. Albright, Inorg. Chem., 1979, 18, 2112–2120.
38 M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. M. W. Gill,
B. G. Johnson, M. A. Robb, J. R. Cheeseman, T. Keith,
G. A. Petersson, J. A. Montgomery, K. Rhaghavachari, M. A. Al-
Laham, V. G. Zakrzewski, J. V. Ortiz, J. B. Forseman, J. Ciolowski,
B. B. Stefanov, A. Nanayakkara, M. Challacombe, C. Y. Peng,
P. Y. Ayala, W. Chen, M. W. Wong, J. L. Andres, E. S. Replogle,
R. Gomberts, R. L. Martin, D. J. Fox, J. S. Brinkley, D. J. Defrees,
J. Baker, J. P. Stewart, M. Head-Gordon, C. Gonzalez and
J. A. Pople, Gaussian 98, Revision A7, Gaussian Inc., Pittsburg,
1998.
39 A. D. Becke, Phys. Rev. A, 1988, 38, 3098–3100.
40 J. P. Perdew and Y. Wang, Phys. Rev. B, 1992, 45, 13244–13249.
41 P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 270–283;
W. R. Wadt and P. J. Hay, J. Chem. Phys., 1985, 82, 284–298;
P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 299–310.
42 T. H. Dunning and P. J. Hay, in Modern Theoretical Chemistry, ed.
H. F. Schaefer III, Plenum Press, New York, 1976, pp. 1–27.
43 G. M. Sheldrick, SHELXS-97 and SHELXL-97, University of
Göttingen, Germany, 1997.
8 W. E. Buhro, A. T. Patton, C. E. Strouse, J. A. Gladysz,
F. B. McCormick and M. C. Etter, J. Am. Chem. Soc., 1983, 105,
1056–1058; W. E. Buhro, M. C. Etter, S. Georgiou, J. A. Gladysz and
F. B. McCormick, Organometallics, 1987, 6, 1150–1156.
9 T. J. Collins and W. R. Roper, J. Chem. Soc., Chem. Commun., 1977,
901–902; T. J. Collins and W. R. Roper, J. Organomet. Chem., 1978,
159, 73–89.
10 L. Hofmann and H. Werner, J. Organomet. Chem., 1983, 255,
C41–C45; L. Hofmann and H. Werner, Chem. Ber., 1985, 118,
4229–4238.
11 W. Paul and H. Werner, Angew. Chem., 1983, 95, 333–334; W. Paul
and H. Werner, Angew. Chem., Int. Ed. Engl., 1983, 22, 316–317;
H. Werner, W. Paul, W. Knaup, J. Wolf, G. Müller and J. Riede,
J. Organomet. Chem., 1988, 358, 95–121; H. Werner, M. Steinmetz,
K. Peters and H. G. von Schnering, Eur. J. Inorg. Chem., 1998,
1605–1617.
12 D. J. Clouthier and D. C. Moule, Top. Curr. Chem., 1989, 150,
167–247; D. R. Johnson, F. X. Powell and W. H. Kirchhoff, J. Mol.
Spectrosc., 1971, 39, 136–145; P. H. Turner, L. Halonen and
I. M. Mills, J. Mol. Spectrosc., 1981, 88, 402–419; S. Carter and
N. C. Handy, J. Mol. Spectrosc., 1998, 192, 263–267.
13 M. Torres, I. Safarik, A. Clement and O. P. Strausz, Can. J. Chem.,
1982, 60, 1187–1191; O. Watanabe, E. Suzuki and F. Watari, Bull.
Chem. Soc. Jpn., 1991, 64, 1389–1391.
14 K. Nakamoto, Infrared and Raman spectra of inorganic and
coordination compounds, 4th edn., Wiley, New York, 1986.
15 M. Ernzerhof, J. P. Perdew and K. Burke, Top. Curr. Chem., 1996,
180, 1–30; E. Engel and R. M. Dreizler, Top. Curr. Chem., 1996, 181,
1–80; A. Berces and T. Ziegler, Top. Curr. Chem., 1996, 182, 41–85;
J. Labanowski and J. Andzelm, Density Functional Methods in
Chemistry, Springer, Heidelberg, 1991; R. G. Parr and W. Yang,
Density Functional Theory of Atoms and Molecules, Oxford
University Press, New York, 1988; L. Szasz, Pseudopotential Theory
of Atoms and Molecules, Wiley, New York, 1986; M. Krauss and
W. J. Stevens, Annu. Rev. Phys. Chem., 1984, 35, 357–385.
16 T. Ziegler, Chem. Rev., 1991, 91, 651–667.
17 G. Frenking, I. Antes, M. Boehme, S. Dapprich, A. W. Ehlers,
V. Jonas, A. Neuhaus, M. Otto, R. Stegmann, A. Veldkamp and
S. F. Vyboishchikov, in Reviews in Computational Chemistry, eds.
K. B. Lipkowitz and D. B. Boyd, VCH, New York, 1996, vol. 8, pp.
63–144; T. R. Cundari, M. T. Benson, M. L. Lutz and S. O.
Sommerer, in Reviews in Computational Chemistry, eds. K. B.
44 M. N. Burnett and C. K. Johnson, ORTEP, Report ORNL-6895,
Oak Ridge National Laboratory, Oak Ridge, TN, 1996.
3128
J. Chem. Soc., Dalton Trans., 2002, 3123–3128