The preparation of the first examples of sulfimide complexes of platinum; the X-ray crystal structure of [Pt(Ph2SNH)4]Cl2

Article abstract of DOI:10.1016/S0277-5387(00)00508-8

The reactivity of S,S-diphenylsulfimide, Ph₂SNH, (1) towards platinum complexes has been investigated. Solutions of 1 and [PtCl₂(PMe₂Ph)₂] show an equilibrium between the starting materials and cis-[PtCl(Ph₂SNH)(PMe₂Ph)₂]Cl; addition of [NH₄][BF₄] generates pure cis-[PtCl(Ph₂SNH)(PMe₂Ph)₂][BF₄] (2). Four equivalents of 1 react with [PPh₄]₂[PtCl₄] over a period of days to give crystalline [Pt(Ph₂SNH)₄]Cl₂ (3); the crystal structure of 3 confirms the expected square planar geometry with significant H-bonding interactions between the cation and the chlorides. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0277-5387(00)00508-8

www.elsevier.nl/locate/poly
Polyhedron 19 (2000) 2077–2081
The preparation of the first examples of sulfimide complexes of
platinum; the X-ray crystal structure of [Pt(Ph2SNH)4]Cl2
P.F. Kelly *, A.C. Macklin, A.M.Z. Slawin, K.W. Waring.
Department of Chemistry, Loughborough Uni6ersity, Loughborough, Leicestershire, LE11 3TU, UK
Received 25 May 2000; accepted 13 July 2000
Abstract
The reactivity of S,S-diphenylsulfimide, Ph2SNH, (1) towards platinum complexes has been investigated. Solutions of 1 and
[PtCl2(PMe2Ph)2] show an equilibrium between the starting materials and cis-[PtCl(Ph2SNH)(PMe2Ph)2]Cl; addition of [NH4][BF4]
generates pure cis-[PtCl(Ph2SNH)(PMe2Ph)2][BF4] (2). Four equivalents of 1 react with [PPh4]2[PtCl4] over a period of days to give
crystalline [Pt(Ph2SNH)4]Cl2 (3); the crystal structure of 3 confirms the expected square planar geometry with significant
H-bonding interactions between the cation and the chlorides. © 2000 Elsevier Science Ltd. All rights reserved.
Keywords: Complexes; Platinum; Sulfimides; Hydrogen bonding
drogen bonding between both sets of triple NꢀH units
on either side of the [Co(Ph2SNH2)6]2+ cation and the
chloride counterions [4]. The interesting ligand proper-
ties of 1 were further illustrated by its reactivity to-
wards Pt(MeCN)2Cl2; now a nucleophilic attack of the
sulfimide upon the coordinated nitrile is seen giving rise
to the novel Ph2SNC(Me)NH ligand [5]. The key fea-
ture of this reaction is that the latter fragment cannot
be formed by simple reaction of the sulfimide with
MeCN in the absence of the platinum centre. Although
the latter reaction was of great interest and has great
potential as a source of new hybrid sulfimide/nitride
fragments as free systems, the unexpected reaction of 1
with MeCN meant that simple coordination chemistry
with Pt went unobserved.
One aspect of our current interests centres upon
making direct contrast between the coordinating ability
of 1 and of underivatised SꢀN species such as S4N4. In
the previous work alluded to above, platinum com-
plexes proved to be amongst the most versatile reagents
for such reactions; thus we have an interest in observing
the simple coordination chemistry of 1 with this metal.
Here we report on the results of the first such studies
and reveal that 1 is not only more reactive towards
simple Pt species than underivatised SꢀN systems but
that the products also exhibit interesting H-bonding
structures that bode well for their use in constructing
extended arrays.
1. Introduction
We have long been interested in the ability of sulfur
(and selenium)–nitrogen species to react with transition
metal centres [1]. In our earlier work on such systems
we tended to concentrate upon the reactivity of SꢀN
species that were either completely devoid of sub-
stituents (eg. S4N4, [S3N3]− etc.) or had only hydrogen
present in addition to the sulfur and nitrogen (e.g.
S4N4H4). In more recent studies we looked at the
properties of substituent-capped long-chain species such
as Me3SiNSNSNSNSiMe3, though in such cases the
substantial disruption of the latter during reaction
means that only unsubstituted SꢀN ligands are found in
the products [2]. In our most recent work, however, we
have prepared a range of complexes of a class of SꢀN
ligand that bears organic groups and retains them upon
coordination. Specifically we have investigated the reac-
tivity of S,S%-diphenylsulfimide, Ph2SNH (1), towards a
number of metal centres.
These studies have generated a range of unexpected
results, such as the unique isomerism of trans-
[CuCl2(Ph2SNH)2] [3] and the concerted, directed hy-
* Corresponding author. Tel.: +44-1509-222-578; fax: +44-1509-
223-925.
E-mail address: p.f.kelly@lboro.ac.uk (P.F. Kelly).
0277-5387/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0277-5387(00)00508-8

2078
P.F. Kelly et al. / Polyhedron 19 (2000) 2077–2081
2. Experimental
2.1. General
data used (Rint=0.1197). Data were corrected for
Lorentz and polarisation effects and an empirical ab-
sorption correction was applied resulting in transmis-
sion factors ranging from 0.842 to 1.00.
All reactions were performed under an atmosphere of
dry nitrogen using dried, degassed solvents (CH2Cl2
and MeCN from calcium hydride, Et2O from sodium–
benzophenone). S,S%-Diphenylsulfimide was prepared
by our recently reported variation of the literature
method [6]; [PPh4]2[PtCl4] was prepared by addition of
[PPh4]Cl to K2[PtCl4] in water, [PtCl2(PMe2Ph)2] by
addition of PMe2Ph to [PtCl2(COD)].
Structure analysis and refinement: the structure was
solved by direct methods and refined by full-matrix
least-squares against F2 leading to R1=0.0376, wR2=
0.0494 with I\2|(I). The maximum/minimum residual
electron densities in the final DF map were 0.991 and
−1.068 e A−3; calculations were performed using
,
SHELXTL [7].
2.2. Preparations
3. Results and discussion
2.2.1. [PtCl(Ph2SNH)(PMe2Ph)2][BF4] (2)
The unexpected observation that 1 reacts with Pt-
(MeCN)2Cl2 to generate [Pt(Ph2SNH)(Ph2SNC(Me)-
NH)Cl]Cl [5] diverted attention away from the possible
simple coordination chemistry that 1 could show to-
wards platinum. Attention to such chemistry is made
worthwhile for two reasons. Firstly, we have used plat-
inum as one of the main metal centres for reaction with
binary SꢀN species such as S4N4. Contrast between the
reactivity of such species and 1 is thus best achieved by
looking at such complexes. Secondly, we have already
noted a propensity for strong H-bonding from coordi-
nated 1 to counterions; such interactions may well be
utilised in the formation of extended array systems. As
platinum has a well-defined square-planar geometry
one would expect to see a significant contrast between
the H-bonding arrangement in its homoleptic com-
plexes of 1 and the aforementioned arrangement in
[Co(Ph2SNH2)6]Cl2.
A solution of [PtCl2(PMe2Ph)2] (50 mg, 0.09 mmol)
in CH2Cl2 (10 ml) was treated with a solution of 1 (23
mg, 0.11 mmol) in the same volume of CH2Cl2. After
stirring for 1 h the colourless solution was treated with
solid [NH4][BF4] (10 mg, 0.1 mmol) and the resulting
mixture stirred vigorously overnight. Filtration fol-
lowed by reduction of the volume in vacuo and slow
diffusion of diethylether yielded a mass of well-formed
colourless crystals of 2. 31P NMR of solutions before
crystallisation showed an effectively quantitative yield
of the product: AX spectrum, lA −12.7 J(31P–195Pt)
1
1
3651 Hz; lB−20.0 ppm J(31P–195Pt) 3141 Hz. Micro-
analysis: Anal. Found: C, 41.6; H, 1.4; N, 4.1. Calc. for
C28H33BClF4NP2PtS: C, 42.3; H, 1.8; N, 4.2%. IR: in
addition to peaks due to the PMe2Ph ligands, bands
also seen at 3307(m, w(NꢀH)), 1065 (s, br), 517 (m),
284(m, w(PtꢀCl)).
Treatment of a solution of 2 in CH2Cl2 with
[nBu4N]Cl resulted in the rapid establishment of an
equilibrium between 2 and [PtCl2(PMe2Ph)2] (as shown
by 31P NMR).
31P NMR reveals that an equimolar mixture of 1 and
cis-PtCl2(PMe2Ph)2 in CH2Cl2 at room temperature
contains both the platinum starting material and a
complex identified by an AX spectrum. Both these
features of the spectrum remain, albeit in varying ra-
tios, even when the relative amounts of the reagents are
altered; even a large excess of 1 does not remove all
cis-PtCl2(PMe2Ph)2 from the system. This observation
suggested the presence of an equilibrium between 1,
cis-PtCl2(PMe2Ph)2 and a new coordination complex
and this was confirmed by removing free chloride from
the system. Thus, treatment with [NH4][L], where L=
[PF6]− or [BF4]−, results in the complete conversion of
the platinum starting material through to the product.
Though both salts may be precipitated by addition of
ether, the [PF6]− case does not crystallise well via slow
diffusion-only an oil is obtained. In the case of the
[BF4]− salt, however, slow diffusion of ether leads to
colourless crystals of the product, which we have char-
acterised as cis-[PtCl(Ph2SNH)(PMe2Ph)2][BF4] (2) by a
combination of NMR, IR and micro-analysis. The key
evidence comes from the fact that an AX pattern is
observed in the 31P NMR spectrum; this rules out a
2.2.2. [Pt(Ph2SNH)4]Cl2 (3)
A solution of [PPh4]2[PtCl4] (0.1 g, 0.1 mmol) in
CH2Cl2 (20 ml) was treated with solid 1 (0.1 g, 0.5
mmol); after stirring the solution was left to stand.
Overnight a small crop of pale yellow crystals ap-
peared; the mixture was allowed to stand for one week
during which time the amount of crystalline material
increased. At this point the solution was decanted from
the crystals which were washed with CH2Cl2 and dried
in vacuo. Yield: 83 mg, 79% based upon Pt. Microanal-
ysis: Anal. Found C, 53.2; H, 3.9; N, 4.7. Calc. For
C48H44N4S4PtCl2 C, 53.8; H, 4.1; N 5.2%.
2.3. X-ray crystallography
Data were collected for 3 (using a Bruker SMART
diffractometer with graphite monochromated Mo Ka
radiation) using small slices: 11 178 data collected, 3293

P.F. Kelly et al. / Polyhedron 19 (2000) 2077–2081
2079
trans structure, while the 195Pt–31P coupling constants
of the phosphorus atoms are consistent with P trans to
Cl and to N. Comparison of the latter value (3141 Hz)
with that for P trans to N(H) in the [S2N2H]− complex
[Pt(S2N2H)(PMe2Ph)2]+ (3240 Hz) serves to confirm
that the ligand is N-bound (noting that in the afore-
mentioned case the value for P trans to S is significantly
different: 2595 Hz) [8]. By analogy with the latter case
therefore we can be confident therefore that the sulfi-
mide has coordinated through the nitrogen atom.
Attempts to use X-ray crystallography to confirm the
structure find problems brought about by the small size
and poor quality of the crystals and by disorder within
the anions1. For this reason we do not formally present
the X-ray data here though it should be noted that the
solution is accurate enough to confirm the connectivity
as shown in the structure in Scheme 1.
why the [BF4]− salt should crystallise so readily while
the [PF6]− analogue generates only an oil.
The above reaction allows us to make some impor-
tant comparisons between the coordinating ability of 1
and that of species such as S4N4. The latter is inert to
PtCl2(PMe2Ph)2 at ambient temperatures and only re-
acts upon either thermolysis [10] or photolysis [11]. In
both cases the reaction probably proceeds via smaller
anions such as [S3N3]−; the sodium salt of the latter
reacts to give Pt(S2N2)(PMe2Ph)2, a product which also
forms when S4N4H4 reacts in the presence of the base
DBU [12]. Thus in order for underivatised SꢀN species
to react with PtCl2(PMe2Ph)2 a driving force such as
salt elimination, photochemical disruption or simple
brute force of high temperatures has to be employed.
Clearly, therefore, 1 has a much stronger affinity for the
platinum centre as it reacts, albeit in the equilibrium
shown in Scheme 1, at ambient temperatures.
Two immediate questions come to mind as an exten-
sion to this reaction and the structure of 2, namely (i)
can more than one unit of 1 be coordinated to the
platinum and (ii) is it possible to generate the trans
isomer of 2? On the first point, there is no evidence
that, at ambient temperatures at least, a second
molecule of 1 can react with 2 to generate a complex of
the type [Pt(Ph2SNH)2(PMe2Ph)2]2+. While it is possi-
ble that steric influences play a part in precluding
formation of the latter, the isolation of a homoleptic
sulfimide complex noted below would make this a less
likely explanation. On the second point we have so far
produced no evidence for the formation of the other
potential isomer of 2. Heating in toluene — a tech-
nique which can be used to generate the trans isomer of
[PtCl2(PMe2Ph)2] from the cis — for example [9] does
not appear to induce isomerisation. It may well be that
again, steric considerations are paramount and mitigate
against formation of the trans-phosphine isomer; for
example, it could be argued that in the cis arrangement
the bulky sulfimide unit has close contact with only one
of the phosphines whereas it would interact with both
in a trans structure.
Although the above reaction appears to be limited to
addition of one unit of 1 as ligand, we can in the
correct circumstances generate a homoleptic platinum
complex of 1. We have previously shown that
[PPh4]2[PtCl4] forms when K2[PtCl4] reacts with
[PPh4]Cl in water [13]. The presence of the phospho-
nium cation solublises the [PtCl4]2− anion in CH2Cl2
and addition of four equivalents of 1 results in a slow
reaction which eventually deposits [Pt(Ph2SNH)4]Cl2 (3)
as a well-formed pale yellow crystalline product after a
few days standing. The product has been characterised
by micro-analysis and IR spectroscopy; in the latter
case the spectrum is quite distinctly similar to that of
the copper analogue [Cu(Ph2SNH)4]Cl2 [6]. Thus it
shows a very strong and broad NꢀH stretch at 3056
cm−1 (3081 cm−1 for the Cu analogue) and a strong
SꢀN stretch at 934 cm−1 (which comes at 940 cm−1 in
the copper case). The NꢀH stretch is particularly perti-
nent — we have noted that this stretch is only strong
when 1 is present on a cation and there is thus H-bond-
ing to the anion; in neutral species it can often be hard
If a solution of 2 in CH2Cl2 is treated with one
equivalent of [nBu4N]Cl, 31P NMR reveals the presence
of both 1 and 2. This observation confirms that the
initial reaction of 1 with [PtCl2(PMe2Ph)2] must proceed
via the kind of equilibrium shown in Scheme 1. It is
only through removal of the free chloride as [NH4]Cl
that the reaction can be driven to completion. That the
choice of anion should have such a dramatic effect
upon the ability of the product to crystallise is some-
what surprising; there would appear no obvious reason
1 Triclinic, a=15.724(5), b=17.407(6), c=14.64(1) A; h=
,
3
,
(
112.31(3), i=97.537, k=90.86(3)°; V=3466(3) A ; space group P1.
Due to the small size and poor quality of the crystal the refinement
did not converge well. The [BF4]− ions show signs of disorder but
this could not be resolved into discrete sites.
Scheme 1. The equilibrium present in solutions of 1 and PtCl2-
(PMe2Ph)2 and the formation of 2.

2080
P.F. Kelly et al. / Polyhedron 19 (2000) 2077–2081
This reaction also provides a contrast between the
reactivity of 1 and S4N4. In the latter case we have
observed substitution of one of the chlorides of the
[PtCl4]2− followed by a redox reaction to give a Pt(IV)
complex of [S4N4]2− [13]. Although a redox reaction is
not an obvious option for 1, the fact that no more than
one chloride is substituted by the S4N4 (even when in
excess) clearly provides further evidence of relative
reactivity of the two species.
In conclusion we can say that this work clearly
indicates that 1 has a powerful affinity for platinum
centres; although the option to bind via sulfur is there,
it prefers to coordinate through the nitrogen and is
significantly more reactive than ‘organo-free’ sulfur ni-
trides such as S4N4. The presence of strong, concerted
H-bonding interactions within 3 confirms the potential
of sulfimide ligands such as 1 to act as a new class of
initiator of extended array systems.
Fig. 1. The molecular structure of [Pt(Ph2SNH)4]Cl2 (3).
to detect. Ultimate confirmation of the structure of 3
comes through X-ray crystallography. Fig. 1 shows its
molecular structure and confirms the expected square-
planar geometry; it also confirms that the ligands are all
N-bound (an important point given the thiophilic na-
ture platinum — note the presence of a PtꢀS bound
sulfimide-type unit, albeit as part of a metallocycle, in
[Pt(Ph2SNH)(Ph2SNC(Me)NH)Cl]Cl [5]). Comparison
with the structure of [Cu(Ph2SNH)4]Cl2 [6] reveals an
effectively identical SꢀN bond distance within the lig-
ands though the average angle at the nitrogens is less
(122.6 cf. 125.3°) in 3. Perhaps the most important
observation is that, as in the Cu case, the ligands
arrange themselves in such a way that the two pairs of
cis sulfurs extend on to opposite sides of the MN4
plane; as a result each pair of NꢀH groups can co-oper-
ate in H-bonding to a chloride. The overall average
4. Supplementary material
Crystallographic data for the structural analysis has
been deposited with the Cambridge Crystallographic
Data Centre, CCDC No. 143910 for compound 3.
Copies of this information may be obtained free of
charge from The Director, CCDC, 12 Union Road,
Cambridge, CB2 1EZ, UK (fax: +44-1223-336033;
e-mail: deposit@ccdc.cam.ac.uk or www: http://
www.ccdc.cam.ac.uk).
Acknowledgements
,
HꢀCl bond distance is thus only 2.25 A, with NꢀHꢀCl
We are grateful to Johnson Matthey for loans of
precious metals.
angles of 168°. As mentioned earlier such cooperative
H-bonding seems to be a feature of complexes of 1; in
the case of 3 Table 1 it clearly raises the possibility that
insertion of the appropriate bridging H-bonding anion
could result in the formation of extended structures.
Thus one may envisage metathesis reactions to intro-
duce anions such as fumarate or terepthalate into the
system which could then link the cations into extended
arrays. Work towards this is underway.
References
[1] P.F. Kelly, J.D. Woollins, Chem. Soc. Rev. (1992) 245, and
references therein.
[2] P.F. Kelly, A.M.Z. Slawin, A. Soriano-Rama, J. Chem. Soc.,
Dalton Trans. (1996) 53.
[3] P.F. Kelly, A.M.Z. Slawin, K.W. Waring, J. Chem. Soc., Dalton
Trans. (1997) 2853.
[4] P.F. Kelly, A.M.Z. Slawin, K.W. Waring, Inorg. Chem. Com-
mun 1 (1998) 249.
[5] P.F. Kelly, A.M.Z. Slawin, Chem. Commun. (1999) 1081.
[6] P.F. Kelly, S.M. Man, A.M.Z. Slawin, K.W. Waring, Polyhe-
dron 18 (1999) 3173.
[7] Siemens SHELXTL, Revision 5.03, Siemens Analytical X-ray,
Madison, WI, USA, 1995.
[8] R. Jones, C.P. Warrens, D.J. Williams, J.D. Woollins, J. Chem.
Soc., Dalton Trans. (1987) 907.
[9] F.R. Hartley, Chemistry of Platinum and Palladium, Applied
Science, London, 1973.
Table 1
,
Selected bond distances (A) and angles (°) in complex 3
PtꢀN(1)
PtꢀN(2)
N(1)ꢀS(1)
N(2)ꢀS(2)
2.042(5)
2.041(5)
1.589(5)
1.580(5)
N(1)ꢀPtꢀN(2)
N(1)ꢀPtꢀN(2A)
PtꢀN(1)ꢀS(1)
PtꢀN(2)ꢀS(2)
92.2(2)
87.8(2)
120.6(3)
124.6(3)

P.F. Kelly et al. / Polyhedron 19 (2000) 2077–2081
2081
[10] J.M. Jolliffe, P.F. Kelly, J.D. Woollins, J. Chem. Soc., Dalton
Trans. (1989) 2179.
[12] R. Jones, P.F. Kelly, D.J. Williams, J.D. Woollins, Polyhedron 6
(1987) 1541.
[11] C.W. Allen, P.F. Kelly, J.D. Woollins, J. Chem. Soc., Dalton
Trans. (1991) 1343.
[13] P.S. Belton, V.C. Ginn, P.F. Kelly, J.D. Woollins, J. Chem. Soc.,
Dalton Trans. (1992) 1135.
.