Theoretical and spectroscopic investigation of coordination compounds from P4S3, copper(I) iodide and W(CO)5

Article abstract of DOI:10.1002/ejic.200901108

The formation of a new type of organometallic-inorganic hybrid coordination polymer from P₄S₃·W(CO)₅ (1) and CuI has been studied. The structure of [P₄S₃·W(CO ₅) (CuI) (3) has been verified by the combination of spectroscopic (IR, Ra-man, ³¹P MAS NMR) and X-ray diffraction methods. Theoretical studies of β-P₄S₃· P₄S ₃·W(CO)₅ (1), 1D-(P₄S₃)(CuI) (2), and. 3D-(P₄S₃)(CuI)₃ (4), which may be considered as poten-tial building blocks of 3, were carried out on the DFT level in the crystalline phase. The comparison of calculated and measured vibration modes of the P₄S₃ cage allows the unequivocal, assignment of the recorded. Raman shifts between 200 and 480 cm^-1 and the determination of the degree of integration of P₄S₃ within the one-dimensional stacks of 2 or copper iodide networks of 3 and 4.

Full text of DOI:10.1002/ejic.200901108

FULL PAPER
DOI: 10.1002/ejic.200901108
Theoretical and Spectroscopic Investigation of Coordination Compounds from
P4S3, Copper(I) Iodide and W(CO)5
Gabor Balázs,[a] Andreas Biegerl,[a] Christian Gröger,[b] Joachim Wachter,*[a]
Richard Weihrich,[a] and Manfred Zabel[a]
Keywords: Phosphorus / Sulfur / Copper / Tungsten / Coordination polymers / Density functional calculations
The formation of a new type of organometallic-inorganic hy-
brid coordination polymer from P4S3·W(CO)5 (1) and CuI has
been studied. The structure of {P4S3·W(CO)5}(CuI) (3) has
tial building blocks of 3, were carried out on the DFT level
in the crystalline phase. The comparison of calculated and
measured vibration modes of the P4S3 cage allows the un-
been verified by the combination of spectroscopic (IR, Ra- equivocal assignment of the recorded Raman shifts between
man, 31P MAS NMR) and X-ray diffraction methods. Theoret-
ical studies of β-P4S3, P4S3·W(CO)5 (1), 1D-(P4S3)(CuI) (2),
and 3D-(P4S3)(CuI)3 (4), which may be considered as poten-
200 and 480 cm–1 and the determination of the degree of in-
tegration of P4S3 within the one-dimensional stacks of 2 or
copper iodide networks of 3 and 4.
of (P4Q4)(CuI)3 it was concluded that there is only weak
interaction between the phosphorus and the metal atoms.
The investigation of the nature of the interaction between
cage molecules and Lewis acidic (LA) metal complex frag-
ments requires theoretical analyses of the bonding proper-
ties of apical and basal coordination isomers of P4Q3·LA
adducts, usually on the basis of gas phase model com-
pounds.[2a,8] If structural data were available discrepancies
between experimental and calculated data were observed.[9]
Another efficient tool for the determination of bonding in-
teractions is vibrational spectroscopy. Comparative Raman
spectroscopy of an inorganic polymer is so far based on the
empirical comparison of vibrational data of the polymer
with those of the free molecule. Typical examples include
As4S4·HgHal2 (Hal = Br, I), which exhibits very weak Hg–
S interactions,[10] and (As4S3)(CuCl)n (n = 1,2), which is
characterized by relatively strong As–S bonds.[11]
Recently, the influence of the condensed phase on DFT
calculations has been described by Hocking, who examined
metal–ligand bond lengths of coordination complexes by
DFT methods in vacuo and condensed phase.[12] In this
work we use the program package CRYSTAL06[13] to con-
sider the influence of packing effects and intermolecular
forces. It allows for a DFT approach on periodic structures
successfully applied to calculate vibrational frequencies of
pyrite type compounds.[14] Hybrid functionals like B3LYP
can be used, too, to improve results for less dense structures
like quartz or silicates significantly.[15] For comparison pur-
poses and in order to get a deeper insight we studied Ra-
man spectra of the series β-P4S3,[16] P4S3·W(CO)5 (1), one-
dimensional (P4S3)(CuI) (2),[4] and 3D-(P4S3)(CuI)3 (4),[4]
which are potential building blocks for the construction of
{P4S3·W(CO)5}(CuI) (3). All spectra except that of 3 are
calculated by DFT methods in the crystalline phase. In or-
Introduction
Tetraphosphorus sesquisulfide is a cage compound with
fascinating coordination properties. The addition of
M(CO)5 fragments (M = Cr, Mo, W) to the apical phospho-
rus is, from a historical point of view, the most intensively
studied reaction type since more than thirty years.[1] The
simultaneous coordination of Lewis-acidic complex frag-
ments or cations at basal and apical phosphorus has been
reported only recently,[2] and even the participation of sul-
fur lone pairs could be realized.[3] The use of P4S3 as build-
ing block in copper(I) halide networks has stimulated a new
kind of “soft” solid-state chemistry, for the formation of
quaternary phases is achieved from solutions by interdif-
fusion techniques. Up to four cage phosphorus atoms are
involved as coordination sites, but not sulfur.[4] The nature
of the CuHal substructures depends on the employed halo-
gen (Cl– or I–). The coordination behavior of CuBr is not
as straightforward and will be reported elsewhere.[5]
Conventional solid-state techniques from the molten ele-
ments, phosphorus and sulfur, and CuI end up with the
formation of (P4Q4)(CuI)3 phases (Q = S, Se), in which β-
P4Q4 cage molecules are embedded in polymeric matrices
of copper iodide.[6,7] No hint has been obtained for the for-
mation of P4Q3 containing phases, while CuCl or CuBr give
rise to the formation of PHal3. From crystallographic data
[a] Institut für Anorganische Chemie der Universität Regensburg,
93040 Regensburg, Germany
Fax: +49-941-943-4439
E-mail: Joachim.Wachter@chemie.uni-regensburg.de
[b] Institut für Biophysik und Physikalische Biochemie der Uni-
versität Regensburg,
93040 Regensburg, Germany
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.200901108.
Eur. J. Inorg. Chem. 2010, 1231–1237
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1231

G. Balázs, A. Biegerl, C. Gröger, J. Wachter, R. Weihrich, M. Zabel
FULL PAPER
der to validate the new method the simulated vibrational
frequencies of β-P4S3 and 1 are compared with data ob-
tained from the gas phase and with measured Raman spec-
tra.
Results
Syntheses and Structures
P4S3· W(CO)5 (1): Reaction of a solution of P4S3 in THF
with a slight excess of W(CO)5THF gave yellow 1. Its 31P
NMR spectrum exhibits two resonances which are in agree-
ment with those described by Riess.[1e] The reaction of P4S3
with four to six equivalents of W(CO)5THF gave an insepa-
rable mixture of P4S3{W(CO)5}n (n = 1, 2, 4) adducts, from
which crystals of 1, P4S3{W(CO)5}2 and P4S3{W(CO)5}4
were isolated and characterized by X-ray diffraction analy-
ses. In all cases the W(CO)5 fragments occupy the phos-
phorus sites of the cage.[17]
The molecular structure of 1 is analogous to that of
P4S3·Mo(CO)5.[1a] This means that a W(CO)5 fragment is
fixed at the apical P atom of the P4S3 cage (Figure 1). A
comparison of bond features of the coordinated P4S3 mo-
lecule with those of the free cage reveals that all distances
are less affected than in the case of P4S3·Mo(CO)5. Com-
pared to β-P4S3 the average distances within the cage are
longer by 0.11 Å for d(Papical–S), equal for d(Pbasal–S) and
longer by 0.022 Å for d(P–P).[16] A slight shortening of the
bond W–C5 reflects the weak donor properties of the P4S3
ligand. The packing in the crystal is shown in Figure 2.
Stacks of 1 are pairwise arranged so that the cage units lie
in the crystallographic ab plain, whereas the W(CO)5 units
are linked together in a second plain parallel to the former.
The closest intermolecular contacts within each layer are
between 3.589 and 3.654 Å for d(P···S), what is close to the
sum of van der Wals radii (3.75 Å) of phosphorus and sul-
fur.
Figure 2. Section of the crystal structure of 1; stacks viewed along a.
{P4S3· W(CO)5}(CuI) (3): Layering a solution of CuI in
CH3CN over a solution of 1 in CH2Cl2 (molar ratio 1:1)
gave after four days insoluble golden shining fibers of 3 in
84% yield. The composition of 3 follows from elemental
analysis and has been confirmed by a preliminary X-ray
crystallographic study. The IR spectrum reveals a ν(CO)
absorption pattern at 2080, 2000 and 1940 cm–1, which is
typical of a W(CO)5 fragment.
A crystal structure analysis of 3 has been attempted, but
the quality of the structure solution has been handicapped
by the extremely fine nature of the fibers. Inspite of this
problem the heavy atom skeleton could be determined. The
structure of 3 contains slightly twisted, castellated (CuI)n
chains, which are linked by P4S3 cage molecules each con-
tributing two basal P atoms (Figure 3). The resulting two-
dimensional network is confined by W(CO)5 fragments
which are fixed at the apical phosphorus atoms of each of
the P4S3 cages. Unfortunately, the CO groups could not be
located but their presence is proved by the IR spectrum of
3. Overall, the P4S3 cage serves for the first time as a triden-
tate ligand towards two different Lewis-acidic metal frag-
ments. Further support of the proposed structure arises
from solid-state 31P MAS NMR and Raman spectra (see
below).
Figure 1. Molecular structure of 1. Selected distances [Å]: W(1)–
P(4) 2.448(2), W(1)–C(1) 2.055(10), W(1)–C(2) 2.026(12), W(1)–
C(3) 2.060(11), W(1)–C(4) 2.058(12), W(1)–C(5) 2.031(10), S(1)–
P(3) 2.094(3), S(1)–P(4) 2.098(4), S(2)–P(2) 2.089(3), S(2)–P(4)
2.096(3), S(3)–P(1) 2.095(3), S(3)–P(4) 2.098(4), P(1)–P(2) 2.246(4),
P(1)–P(3) 2.242(4), P(2)–P(3) 2.247(4).
Figure 3. Section of the structure of 3 (CO groups omitted); stacks
along b.
1232
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 1231–1237

Investigation of an Organometallic-Inorganic Hybrid Coordination Polymer
Raman Spectroscopy
Raman spectra of P4S3 (Figure 4) and compounds 1–4
have been measured. Spectra of 1 were calculated from the
gas phase and, including those of P4S3, compared with data
obtained from the condensed phase. An extension of the
latter method on 2 and 4 and an extrapolation of the ob-
tained data on 3 allows to determine the influence of cop-
per or tungsten coordination on the P4S3 cage.
Figure 5. Relative deviations of DFT calculations of Raman fre-
quencies of P4S3 in the gas phase[18] and in the condensed phase.
In the next step DFT calculations on P4S3·W(CO)5 (1)
were performed in the gas phase with the Turbomole[21]
program package using both, the BP86[22] and B3LYP[23]
functionals with the TZVP basis set[24] for all atoms. Quasi-
relativistic core potential was used for the tungsten atom.
The calculated bond lengths are in agreement with the ex-
perimental ones, however they are slightly longer by 0.04 Å
(mean). The vibrational frequencies calculated with the
BP86 functional are in better agreement with the experi-
mental data than those calculated with the B3LYP func-
tional. Therefore, only the results obtained with the BP86
functional obtained without scaling are discussed.
Figure 4. Experimental Raman spectra of P4S3 and 2.
DFT Calculations
The calculated vibrational spectra of 1 can be formally
The normal mode frequencies and the corresponding vi- divided in a region up to 400 cm–1, a second region from
brational assignments of P4S3 have already been calculated 400 to 600 cm–1 and a region around 2000 cm–1. In the first
in the gas phase and compared with Raman frequencies of and third region the vibrational frequencies are relatively
solid β-P4S3 (Table 1).[18–20] Jensen et al. calculated vi- well reproduced by calculations, however in the second re-
brational frequencies of P4S3 in more or less good agree- gion a strong coupling of the different vibration modes of
ment with experimental data by means of Hartree–Fock the P4S3 skeleton with the CO groups of the W(CO)5 unit
(HF) and second-order Moller-Plesset (MP2) methods. The was predicted but no direct experimental evidence for this
best results were obtained with the density functional coupling was found. This strong coupling prevents a clear
theory (DFT) method using the exchange correlation func- assignment of different vibration modes by visualisation of
tion B3LYP with the standard 6-31G* basis set, while HF the atomic displacements.
and MP2 led to relatively big deviations (Table 1).[18] Then
The comparison of calculated (216, 278 and 317 cm–1 as-
we carried out computational studies of frequencies of crys- signed to P–S–Pwag, P–S–Pbend and P–Pstretch, respectively)
talline β-P4S3. The results, which are summarized in Table 1 and experimental Raman frequencies (223, 276 and
and in Figure 5 show that there is still a better agreement 332 cm–1) recorded for 1 (Figure 6, Table 2) allows for an
with experimental Raman shifts.
unambiguous labelling. The frequencies ν1, ν2 and ν3 calcu-
Table 1. Comparison of Raman shifts of P4S3, calculated in the gas phase and in the solid state (CRYSTAL06), with experimental shifts.
Jensen et al.[18]
MP2
Exp. Raman shifts
DFT/B3LYP[a]
Assignment
HF
DFT/B3LYP
252
332
407
514
485
570
241
307
350
451
423
494
228
296
339
411
400
467
219
283
339
417
438
483
211
270
335
413
422
481
P–S–P wag, ν1
P–S–P bend, ν2
P–P stretch, ν3
P–S–P stretch (a), ν4
P–S–P stretch (s), ν5
S–P stretch, ν6
[a] Condensed phase, this work.
Eur. J. Inorg. Chem. 2010, 1231–1237
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1233

G. Balázs, A. Biegerl, C. Gröger, J. Wachter, R. Weihrich, M. Zabel
FULL PAPER
lated in the crystalline phase are of less accuracy, but the
agreement is still good for these relatively large cells com-
pared to SiO2.[15] Furthermore, the comparison of the cal-
culated vibrational modes of 1 with the experimental spec-
trum of P4S3 shows that the first three absorption bands are
only slightly influenced by the coordination of the W(CO)5
fragment. This observation is supported by the structural
parameters (Figure 1). The experimental Raman shift at
367 cm–1 can be attributed to a P–S–P bending coupled
with the ν(P–W) vibration. It is fairly reproduced by BP86
calculations (352 cm–1), while it agrees well in the solid state
(372 cm–1). In the third region of the spectra, the calculated
ν(CO) frequencies at 1974 and 1977 cm–1 are moderately
shifted to higher wave numbers, whereas the calculated ab-
sorption at 2073 cm–1 is only slightly shifted to lower wave
numbers compared to the experimental data [ν(CO) = 1927,
1945, 2081 cm–1]. The increasing number of ν(CO) absorp-
tions may be explained by a decrease of C4v symmetry (lo-
cal symmetry) of the W(CO)5 fragment to C2v (or Cs) in
the gas phase model and packing effects in the crystal.
Figure 7. Section of the structure of 2.[4]
The calculated Raman shifts of 2 agree well with the ex-
perimental data and there is no need to introduce scaling
factors (Table 3). The vibrational modes are assigned by
calculation. A comparison of the measured Raman spectra
of P4S3 and 2 (Figure 4) reveals the existence of two groups
of frequencies. Whereas ν1 and ν2 are only marginally
shifted, what can be ascribed to the fact that the basis of
the coordinated P4S3 cage is not involved in these modes,
the other frequencies are affected by the coordination to
copper sites. The P–P stretching of P4S3 at 339 cm–1 (ν3) is
influenced by coordination and the frequency splits into
two ones because of symmetry mutations of the coordi-
nated P3 basis compared to the free P4S3 basis. The sym-
metric P–P stretching can be observed at 331 cm–1 (ν3a),
whereas the antisymmetric P–P stretching can be identified
at 354 cm–1 (ν3b). As a result, the uncoordinated phos-
phorus atom is able to vibrate easier than the coordinated
basal phosphorus atoms. This is also reflected by a small
increase of these bonds in the crystal structure of
(P4S3)(CuI) by 0.044(1) Å.
Table 3. Experimental and calculated Raman shifts (cm–1) of 2 and
4 in comparison to P4S3.
Figure 6. Experimental Raman spectra of 1 and {P4S3· W(CO)5}-
(CuI).
Freq. no.
2
4
P4S3 (exp.)
calcd.
exp.
calcd.
exp.
Table 2. Experimental and calculated IR wavenumbers and Raman
shifts of 1.
ν1
ν2
ν3a
ν3b
ν4
ν5
ν6
ν7
209
278
328
346
419
450
476
496
215
282
331
354
421
454
477
499
215
259
310
344
433
448
493
215
274
219
283
339
–
417
438
483
–
294
IR wavenumbers (cm–1)
352[a]
438
ν(CO) (KBr)
˜
CRYSTAL06
Turbomole
2081 s
2094
2015
1955
1952
1931
2073
–
1977
1974
–
462
498
1945 vs
1927 sh
[a] Vibrations at 355 and 370 cm–1 cannot clearly be resolved (see
text).
Raman shifts (cm–1)
Freq. no.
CRYSTAL06
Turbomole
exp.
Assignment
Important differences are also observed for the modes
in which the sulfur bridges are involved. Thus, for P4S3 a
symmetric P–S–P stretching (ν4) is observed at 417 cm–1
whereas antisymmetric P–S–P stretching (ν5) is found at
438 cm–1. In 2 these Raman shifts are split into two vi-
brations which arise from participation of both coordinated
ν1
ν2
ν3
ν4
210
267
311
372
216
278
317
352
223
276
332
367
P–S–Pwag
P–S–Pbend
P–Pstretch
P–S–Pbend/P–W
In a further step we performed calculations on copper phosphorus atoms of the basis, while the free basal phos-
iodide coordination compounds with P4S3. As a first exam- phorus atom is hardly incorporated into the stretching
ple we took (P4S3)(CuI) (2), which is a one-dimensional modes. As a result the symmetric P–S–P stretching mode
polymer composed of Cu2I2 four-membered rings bridged appears at 421 cm–1 (ν4) and the antisymmetric P–S–P
by P4S3 units (Figure 7).[4] As an extension of this work we stretching mode at 454 cm–1 (ν5). The stretching of S–Papical
got qualitative information on the shielding trends in 31P is only slightly affected (ν6) by the incorporation of the cage
MAS NMR spectroscopy in 2.
in the polymeric chain when compared to P4S3.
1234
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 1231–1237

Investigation of an Organometallic-Inorganic Hybrid Coordination Polymer
Figure 8. Section of the structure (left) and the (CuI)n network (right) of 4.[4]
The Raman spectrum of 3 shows a relatively simple 31P MAS NMR Spectroscopy
pattern between 200 and 400 cm–1 (Figure 6). The most
More structural information on the new compounds 1
and 3 may be obtained by solid-state 31P MAS NMR spec-
troscopy. The coordination of W(CO)5 at the apical P atom
of the cage gives rise to a slight downfield shift of 3.7 (1)
or 10.1 ppm (3) compared to P4S3 (Table 4). The spectrum
of 3 exhibits a further group of signals (Figure 10), which
after simulation split into a singlet at –77.8 ppm and two
multiplets at –84.1 ppm (JCu,P = 854 Hz) and –93.5 ppm
(JCu,P = 717 Hz), respectively. The observed pattern indi-
cates a slight inequivalency of the two P atoms bearing cop-
per, probably caused by a slight anisotropy within the crys-
tal structure (Figure 3). Unfortunately, the actual state of
the structure solution does not provide more precise infor-
mation.
striking difference in this region is the shift of ν3 (P–Pstretch
)
from 332 cm–1 in 1 to 360 cm–1, which is accompanied by
an increase in intensity. The weak vibration at 375 cm–1 is
tentatively assigned to ν(P–W), similar to that of 1. The
region between 400 and 500 cm–1 shows similarities to the
spectrum of 1 thus reflecting a similar coupling of the dif-
ferent vibration modes of the P4S3 cage with the CO groups
of the W(CO)5 unit. For these reasons a closer theoretical
examination of the spectrum of 3 was not attempted.
In the following an extension of the calculations from
one- to three-dimensional polymers is demonstrated. As a
representative example we took 3D-(P4S3)(CuI)3, the struc-
ture of which consists of undulated (CuI)n layers which are
linked by tridentate P4S3 ligands (Figure 8).[4] Two atoms
of the P3 basis form columns upon linking two fused six-
membered Cu3I3 ring-subunits of one and the same
(Cu3I3)n layer, while the apical phosphorus atom coordi-
nates to Cu of the next sheet thus forming seven-membered
columns. The Raman spectrum of (P4S3)(CuI)3 is shown in
Figure 9, the comparison of calculated and experimental
data is given in Table 3. It is striking that the frequencies
ν1–ν6 correspond to those found for 2. The frequency at
185 cm–1 may be clearly assigned to a ν(Pap–Cu) vibration,
not present in the spectra of 1–3. The additional frequencies
at 355 and 370 cm–1 are difficult to assign, because the cal-
culated shifts for vibrations of symmetries B1g and B2g,
which are found in the respective region, are very close to
each other.
Table 4. 31P MAS NMR chemical shifts (δ in ppm) and coupling
constants 31P–63/65Cu (J in Hz) of β-P4S3 and compounds 1–4.
Papical
Pbasal
β-P4S3[31]
P4S3·W(CO)5 (1)
(P4S3)(CuI) (2)
81.8 s
85.5 s
113.8 s
92.9 s
82.2 m (1100)
–102.7 s
–110.7 s
[4]
–82.8 m (900), –95.3 s
–77.8 s, –84.1 m (854), –93.5 m (717)
–94.2 m (698), –94.9 s
{P4S3·W(CO)5}(CuI) (3)
(P4S3)(CuI)3[4] (4)[4]
A comparison of 31P MAS NMR spectra of 1, 2 and 3
shows that the chemical shift of the apical phosphorus atom
of the P4S3 cage in 3 is slightly shifted to low field by
W(CO)5 coordination (Table 4). Interestingly, the lowest
chemical shift is found in the 1D-polymer 2 (see below).
The resonances of the basal phosphorus atoms of 2–4 are
shifted to low field between ∆(δ) = 24.9 (3) and 7.8 ppm (4)
with respect to P4S3. The observed trends are contradictory
[25]
[26]
to those observed for P4Se4(CuI)3
or P4S4(CuI)3
in
which the cage molecules are embedded in copper iodide
networks. For both examples significant shielding effects
have been proposed as a consequence of the coordination
of CuI to P sites.
Although the apical P atom of 2 does not coordinate to
copper the corresponding resonance is remarkably shifted
to low field with respect to β-P4S3 (Table 4). A similar shift
of the same order [∆(δ) = 32 ppm] has been found in 1D-
(P4S3)(CuBr).[5] In order to understand this anomaly of
chemical shifts we have carried out a Mulliken population
analysis of the total electron distribution in the solid state
Figure 9. Raman spectrum of 4.
Eur. J. Inorg. Chem. 2010, 1231–1237
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1235

G. Balázs, A. Biegerl, C. Gröger, J. Wachter, R. Weihrich, M. Zabel
FULL PAPER
out in Schlenk tubes of 3.0 cm diameter, if not otherwise stated.
P4S3 was synthesized by melting red phosphorus and sulfur in a
molar ratio of 5 to 3 under a nitrogen atmosphere. P4S3 was ex-
tracted by toluene and purified by recrystallisation from CH2Cl2.
31P MAS NMR spectra were recorded on a Bruker Advance 300
spectrometer using a double resonance 2.5 mm MAS probe. The
31P resonance was 121.495 MHz. All spectra were acquired at a
MAS rotation frequency of 35 kHz, a 90° pulse length of 2.3 µs
and with a relaxation delay of 450 s. For spectrum simulation the
program DMFIT was used.[27] The Raman spectra were taken by
reflection with an Andor DV401 CCD detector. The excitation of
the microcrystalline samples were carried out with a HeNe laser (λ
= 632.8 nm).
CRYSTAL06 Computational Details: The calculations were per-
formed within the framework of DFT theory as implemented in the
LCAO-code CRYSTAL06.[13,28] Therein, the electronic structure is
calculated from Gaussian type local basis sets. IR and Raman fre-
quencies are calculated from the vibrational spectra at the gammma
point.[29] Exchange and correlation were treated as described by the
B3LYP functional for all results presented in this paper. All elec-
tron basis sets were used for Cu (0.28, 0.4), and O (0.24, 0.5), val-
ence basis sets for P (0.23, 0.49), S (0.22, 0.44), I (0.22, 0.33), and
W (0.25, 0.26) with respective optimized coefficients for outer (sp,
d) functions.[30] The calculations were convergenced to total energy
∆E Ͻ 10–8 H applying a k-points shrinking factors of 4 to 8 and
Anderson mixing (see ref.[13]).
Figure 10. Solid-state 31P MAS NMR spectra of 3. a) Experimental
spectrum; b) simulated spectrum; c–e) simulated spectra showing
individual components. Spinning side bands are marked by the
symbol ϫ; the assignment of P3 and P4 is arbitrary.
with the CRYSTAL 06 program package for P4S3, 2, and 4.
The same basis sets were employed as for the calculation of
the vibrational spectra. Qualitatively it could be shown that
the 3s orbital electron distribution of the apical P atoms is
nearly the same for P4S3 and (P4S3)(CuI)3, but decreases
significantly for 2. This result corresponds perfectly to the
Syntheses
P4S3·W(CO)5: P4S3 (300 mg, 1.36 mmol) was added to a W(CO)5
THF solution (2.045 mmol, c = 28 mmolL–1). The reaction mixture
was stirred at room temperature and the solvent was removed after
20 h. The reaction product was washed three times with small
amounts of toluene and yellow powder of 1 (yield 555 mg, 78%)
trend observed for the 31P chemical shifts of the concerned was obtained. 31P{1H} NMR (300 MHz, C6D6): δ = 85.5 (q, JP,P
= 31, JP,W = 126 Hz), –110.4 (d, JP,P = 31 Hz) ppm. C5O5P4S3W
(543.98): calcd. C 11.04, S 17.68; found C 11.02, S 17.81. IR [KBr,
ν(CO)]: ν = 2083 (s), 2000 (w), 1938 (br) cm–1.
compounds. The origin of this effect is still unclear, for
there are no striking deformations within the P4S3 cage nor
are there significant contacts between stacks of 1D-
(P4S3)(CuI) chains in the crystal.[4]
˜
˜
{P4S3·W(CO)5}(CuI): A solution of CuI (18 mg, 0.092 mmol) in
CH3CN (10 mL, c = 9 mmol L–1) was layered carefully over a solu-
tion of 1 (50 mg, 0.092 mmol) in 20 mL CH2Cl2 (c = 5 mmol· L–1).
After four days golden shining fibers of 3 crystallized. The product
was washed with a CH2Cl2/CH3CN mixture and dried under vac-
uum. Yield 51 mg (75%); 31P MAS NMR (300 MHz): δ = 92.9 (s),
–77.8 (s), –84.1 [m, J(Cu,P) = 854 Hz], –93.5 [m, J(Cu,P) = 717 Hz]
ppm. C5CuIO5P4S3W (734.47): calcd. C 8.18, S 13.10; found C
Conclusions
The new organometallic-inorganic hybrid coordination
polymer {P4S3·W(CO)5}(CuI) (3) has been formed from
P4S3·W(CO)5 (1) and CuI. The structure of 3 has been
solved by the combination of spectroscopic (IR, Raman,
31P MAS NMR) and X-ray diffraction methods. Theoreti-
cal studies of the potential building blocks β-P4S3, 1, 1D-
(P4S3)(CuI) (2) and 3D-(P4S3)(CuI)3 (4) have been carried
out on the DFT level in the crystalline phase. The new
method is able to consider the influence of the solid state
such as packing effects and intermolecular forces and it re-
places calculations of molecular model compounds, which
are often sections of polymers in the gas phase. The ob-
tained results lead to a better agreement of computed Ra-
man frequencies with experimental data.
7.97, S 12.73. IR [KBr, ν(CO)]: ν = 2080 (vs), 2000 (s), 1940 (br)
˜
˜
cm–1.
Crystal Structure Analyses: 1: Data were collected on an Oxford
Diffraction Gemini Ultra diffractometer (Cu-Kα radiation) at
123 K. The structure was solved by direct methods and refined by
full-matrix least-squares (SHELXL97 program) with all reflections.
3
¯
Yellow needles, 0.17ϫ0.09ϫ0.08 mm , triclinic, P1, a = 6.950(3),
b = 8.998(4), c = 11.559(6) Å, α = 87.2(0), β = 82.8(0), γ = 81.13(0)°.
V = 708.26(6) Å3, Z = 2, ρcalcd. = 2.551 g/cm–3, θ = 4.98–62.20°, µ
= 23.635 mm–1, 6629 measured reflections, 2191 independent re-
flections (Rint = 0.0304), 2039 observed reflections [IϾ2σ(I)], 163
refined parameters, R1 (all data) = 0.0420, wR2 = 0.1091, residual
electron density 2.688/–1.231 e·Å–3.
CCDC-757374 (for 1) contains the crystallographic data (excluding
structure factors). These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.cam.
ac.uk/data_request/cif.
Experimental Section
General: All manipulations were carried out under nitrogen by
Schlenk techniques. Usually, the diffusion experiments were carried
1236
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 1231–1237

Investigation of an Organometallic-Inorganic Hybrid Coordination Polymer
[13] R. Dovesi, V. R. Saunders, C. Roetti, R. Orlando, C. M. Zicov-
Supporting Information (see also the footnote on the first page of
this article): Atomic displacements of the key vibrational modes of
P4S3, 1, 2 and 4.
ich-Wilson, F. Pascale, B. Civalleri, K. Doll, N. M. Harrison,
I. J. Bush, P. D’Arco, M. Llunell, CRYSTAL06, user’s guide,
Torino, 2006.
[14] M. Meier, R. Weihrich, Chem. Phys. Lett. 2008, 461, 38.
[15] C. M. Zicovich-Wilson, F. Pascale, C. Roetti, V. R. Saunders,
R. Orlando, R. Dovesi, J. Comput. Chem. 2004, 25, 1873.
[16] a) T. K. Chattopadhyay, M. May, H. G. von Schnering, G. S.
Pawley, Z. Kristallogr. 1983, 165, 47; b) U. Müller, H. Gruber,
Z. Kristallogr. 1997, 212, 662.
[17] A. Biegerl, J. Wachter, unpublished results.
[18] J. O. Jensen, D. Zeroka, A. Banerjee, THEOCHEM 2000, 505,
31.
[19] M. Gardner, J. Chem. Soc., Dalton Trans. 1973, 691.
[20] M. Ystenes, W. Brockner, F. Menzel, Vib. Spectrosc. 1993, 5,
195.
[21] a) R. Ahlrichs, M. Bär, M. Häser, H. Horn, C. Kölmel, Chem.
Phys. Lett. 1989, 162, 165; b) O. Treutler, R. Ahlrichs, J. Chem.
Phys. 1995, 102, 346; c) http://www.turbomole-gmbh.com/.
[22] a) A. D. Becke, Phys. Rev. A 1988, 38, 3098; b) S. H. Vosko, L.
Wilk, M. Nusair, Can. J. Phys. 1980, 58, 1200; c) J. P. Perdew,
Phys. Rev. B 1986, 33, 8822; Erratum: J. P. Perdew, Phys. Rev.
B 1986, 34, 7406.
[23] a) P. A. M. Dirac, Proc. R. Soc. London, Ser. A 1929, 123, 714;
b) J. C. Slater, Phys. Rev. 1951, 81, 385; c) S. H. Vosko, L. Wilk,
M. Nusair, Can. J. Phys. 1980, 58, 1200; d) A. D. Becke, Phys.
Rev. A 1988, 38, 3098; e) C. Lee, W. Yang, R. G. Parr, Phys.
Rev. B 1988, 37, 785; f) A. D. Becke, J. Chem. Phys. 1993, 98,
5648.
[24] a) A. Schäfer, C. Huber, R. Ahlrich, J. Chem. Phys. 1994, 100,
5829; b) K. Eichkorn, F. Weigend, O. Treutler, R. Ahlrichs,
Theor. Chem. Acc. 1997, 97, 119.
[25] G. Brunklaus, J. C. C. Chan, H. Eckert, S. Reiser, T. Nilges, A.
Pfitzner, Phys. Chem. Chem. Phys. 2003, 5, 3768.
[26] S. Reiser, G. Brunklaus, J. H. Hong, J. C. C. Chan, H. Eckert,
A. Pfitzner, Chem. Eur. J. 2002, 8, 4228.
[27] D. Massiot, F. Fayon, M. Capron, I. King, S. Le Calvé, B.
Alonso, J.-O. Durand, B. Bujoli, Z. Gan, G. Hoatson, Magn.
Reson. Chem. 2002, 40, 70.
Acknowledgments
This work was supported by the Deutsche Forschungsgemeinschaft
(DFG) (Wa 486/11-1 and We 4284/1-1). We gratefully acknowledge
continuous support by Prof. Dr. M. Scheer. R. W. thanks Prof. R.
Dovesi for fruitful discussions. We also thank Dipl. Chem. C.
Güntner for technical assistance in recording the Raman spectra.
[1] a) A. W. Cordes, R. D. Joyner, R. D. Shores, E. D. Dill, Inorg.
Chem. 1974, 13, 132; b) R. Jefferson, H. F. Klein, J. F. Nixon,
J. Chem. Soc. C 1969, 536; c) M. Di Vaira, M. Peruzzini, P.
Stoppioni, J. Organomet. Chem. 1983, 258, 373; d) M. Di Vaira,
M. Peruzzini, P. Stoppioni, Inorg. Chem. 1983, 22, 2196; e) J. G.
Riess, ACS Symp., Ser. 1983, 232, 17.
[2] a) E. Guidoboni, I. de los Rios, A. Ienco, L. Marvelli, C. Me-
alli, A. Romerosa, R. Rossi, M. Peruzzini, Inorg. Chem. 2002,
41, 659; b) M. Di Vaira, I. de los Rios, F. Mani, M. Peruzzini,
P. Stoppioni, Eur. J. Inorg. Chem. 2004, 293; c) I. de los Rios,
F. Mani, M. Peruzzini, P. Stoppioni, Inorg. Chem. 2004, 43,
164; d) P. Barbaro, M. Di Vaira, M. Peruzzini, S. S. Con-
stantini, P. Stoppioni, Chem. Eur. J. 2007, 13, 6682.
[3] a) A. Adolf, M. Gonsior, I. Krossing, J. Am. Chem. Soc. 2002,
124, 7111; b) I. Raabe, S. Antonijevic, I. Krossing, Chem. Eur.
J. 2007, 13, 7510.
[4] A. Biegerl, E. Brunner, C. Gröger, M. Scheer, J. Wachter, M.
Zabel, Chem. Eur. J. 2007, 13, 9270.
[5] A. Biegerl, C. Gröger, H. R. Kalbitzer, J. Wachter, M. Zabel,
Z. Anorg. Allg. Chem., in press.
[6] S. Reiser, G. Brunklaus, J. H. Hong, J. C. C. Chan, H. Eckert,
A. Pfitzner, Chem. Eur. J. 2002, 8, 4228.
[7] a) A. Pfitzner, S. Reiser, H.-J. Deiseroth, Z. Anorg. Allg. Chem.
1999, 625, 2196; b) A. Pfitzner, S. Reiser, Inorg. Chem. 1999,
38, 2451.
[28] M. D. Towler, M. Causa, A. Zupan, Comput. Phys. Commun.
1996, 98, 181.
[29] F. Pascale, C. M. Zicovich-Wilson, F. Lopez Gejo, B. Civalleri,
R. Orlando, R. Dovesi, J. Comput. Chem. 2004, 25, 888.
[30] Crystal basis set library: http://www.crystal.unito.it/Basis_Sets/
Ptable.html.
[31] a) R. K. Harris, P. J. Wilkes, P. T. Wood, J. D. Woollins, J.
Chem. Soc., Dalton Trans. 1989, 809; b) H. Eckert, C. S. Liang,
G. D. Stucky, J. Phys. Chem. 1989, 93, 452; c) I. Raabe, S. An-
tonijevic, I. Krossing, Chem. Eur. J. 2007, 13, 7510.
Received: November 17, 2009
[8] J. D. Head, K. A. R. Mitchell, L. Noodleman, N. L. Paddock,
Can. J. Chem. 1977, 55, 669.
[9] C. Aubauer, E. Irran, T. M. Klapötke, W. Schnick, A. Schulz,
J. Senker, Inorg. Chem. 2001, 40, 4956.
[10] a) M. F. Bräu, A. Pfitzner, Angew. Chem. 2003, 118, 4576–4578;
Angew. Chem. Int. Ed. 2006, 45, 4464; b) M. F. Bräu, A.
Pfitzner, Z. Anorg. Allg. Chem. 2007, 633, 935.
[11] P. Schwarz, J. Wachter, M. Zabel, Eur. J. Inorg. Chem. 2008,
5460.
[12] R. K. Hocking, R. J. Deeth, T. W. Hambley, Inorg. Chem. 2007,
46, 8238.
Published Online: February 4, 2010
Eur. J. Inorg. Chem. 2010, 1231–1237
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1237