Formation of the bidentate [Ph2SNC(Me)N(H)] ligand by metal-assisted sulfimide addition to acetonitrile; X-ray crystal structure of [Pt(Ph2SNH)(Ph2SNC(Me)NH)Cl]Cl.MeCN

Article abstract of DOI:10.1039/a901783f

In contrast to similar reactions of Cu, Co and Pd species, the platinum(II) complex [Pt(MeCN)₂Cl₂] reacts with Ph₂SNH (ratio 1:2) in MeCN via a metal assisted addition of the sulfimide to the acetonitrile to give [Pt(Ph₂SNH)(Ph₂SNC(Me)NH)Cl]Cl^.MeCN, the first example of a complex of the bidentate [Ph₂SNC(Me)N(H)] ligand.

Full text of DOI:10.1039/a901783f

Formation of the bidentate [Ph2SNC(Me)N(H)] ligand by metal-assisted
sulfimide addition to acetonitrile; X-ray crystal structure of
[Pt(Ph2SNH)(Ph2SNC(Me)NH)Cl]Cl·MeCN
Paul F. Kelly* and Alexandra M. Z. Slawin
Department of Chemistry, Loughborough University, Loughborough, Leicestershire, UK LE11 3TU.
E-mail: P.F.Kelly@lboro.ac.uk
Received (in Basel, Switzerland) 4th March 1999, Accepted 1st May 1999
In contrast to similar reactions of Cu, Co and Pd species,
the platinum(ii) complex [Pt(MeCN)2Cl2] reacts with
Ph2SNH (ratio 1:2) in MeCN via a metal assisted addition of
about the reaction and its product. First of all it is clear that
during the course of the reaction one of the sulfimide units has
added to an acetonitrile and completed the metallocycle by
binding to the platinum through a sulfur. Within this metal-
locycle the S–N distance appears to be virtually identical to that
in the unreacted, coordinated sulfimide ligand (1.625 cf. 1.618
Å) indicating that this may be best represented as a double bond.
Of the two C–N bonds one, N(14)–C(13), is significantly
shorter than the other (1.302 cf. 1.358 Å) indicating a higher
bond order. Thus we can approximate the bonding within the
neutral ligand to N(H)NC(Me)NNSPh2 though as with most
sulfimide systems this obviously constitutes something of an
over simplification. Bond distances and angles within the
coordinated unreacted sulfimide unit are not significantly
different to those noted in other complexes of 1. A point of note
about the structure comes with the fact that there are significant
H-bonding interactions between the two N–H bonds and the
free chloride (average of the two H···Cl distances being 2.26 Å).
It is possible that this is a contributory factor in the significant
deviation of the N(2)–Pt–S(1) angle from linear (167.6°). In the
light of this structure the unexpected IR spectrum may be
explained; the strong N–H band comes about thanks to the
hydrogen bonding in the system, while the other bands may be
assigned to C–N stretches and deformations. The structure is
the
sulfimide
to
the
acetonitrile
to
give
[Pt(Ph2SNH)(Ph2SNC(Me)NH)Cl]Cl·MeCN, the first ex-
ample of a complex of the bidentate [Ph2SNC(Me)N(H)]
ligand.
As part of our continuing interest in the coordination chemistry
of sulfur–nitrogen species1 we have recently reported on the
reactivity of S,SA-diphenylsulfimide, Ph2SNH 1, towards a
number of metal centres. For example, during investigations
into its reactivity towards CuCl2 we showed that the N-bound
2:1 complex, trans-[CuCl2(Ph2SNH)2], forms readily via
reaction in MeCN. Though such a product was not unexpected,
it proved to be unusual in its ability to crystallise in either
square-planar or pseudo-tetrahedral geometries.2 Such isomer-
ism appears to be unique for a neutral Cu(ii) species, indicating
that the ligand properties of 1 are not as simple as one might
Ph
Ph
H
S
N
1
1
also backed up by the H and 13C NMR data (the latter, for
imagine. This conclusion is backed up by the observation that
[Co(Ph2SNH2)6]Cl2 exhibits strong, concerted, directed hydro-
gen bonding between both sets of triple N–H units on either side
of the coordination octahedron and the chloride counter ions.3
These results, backed up by the obvious potential of the
sulfimides in general as functionalised ligands, prompted us to
investigate the coordination chemistry of 1 further by address-
ing its ability to coordinate to Pt and Pd centers.
example, showing the CH3–C carbons at d 19.0 and 183.6
respectively). One final important feature of the structure comes
from the fact that it constitutes the first example of a sulfimide
unit binding through the sulfur atom as opposed to the
nitrogen.
Many aspects of the latter chemistry prove to be quite
straightforward;4 thus salts of [Pd2X6]22 (X = Cl or Br) form
[PdX3(Ph2SNH)]2 or [Pd(Ph2SNH)4]X2 depending upon reac-
tion stoichiometry while 4 equivalents of 1 react slowly with
[PPh4]2[PtCl4] in CH2Cl2 to give [Pt(Ph2SNH)4]Cl2. There is no
evidence from either X-ray crystallographic or IR spectroscopic
studies that such products contain 1 as anything other than
simple N-bound ligands. During such work, however, we noted
that the product of the reaction of 1 with [Pt(MeCN)2Cl2] (ratio
2:1) in acetonitrile† generated crystalline material the IR
spectrum of which was unusual in that it exhibited a very strong,
broad N–H peak at 3071 cm21 (of the type we have more
usually seen associated with cationic sulfimide complexes) and
extra bands not normally associated with the Ph2SNH ligand (at
1540, 1249 and 807 cm21). As such data were not consistent
with the simple formulation [Pt(Ph2SNH)2Cl2] we investigated
the complex by X-ray crystallography.‡
The structure obtained for 2 (which includes an acetonitrile of
solvation) is shown in Fig. 1. Although two of the four
coordination sites on the platinum contain expected ligands (N-
bound 1 and a chloride) the other two sites are taken up by the
nitrogen and sulfur atoms of a chelating [Ph2SNC(Me)N(H)]
ligand. Close inspection of the latter and the overall metal-
locycle of which it is part reveals a number of important points
Fig. 1 X-Ray crystal structure of 2·MeCN. Selected bond distances (Å) and
angles (°): Pt–S(1) 2.1876(12), S(1)–N(1) 1.625(4), N(1)–C(13) 1.358(6),
C(13)–N(14) 1.302(6), N(14)–Pt 1.989(4), Pt–Cl(1) 2.3191(14), Pt–N(2)
2.012(4), N(2)–S(2) 1.618(4); Pt–S(1)–N(1) 106.7(2), S(1)–N(1)–C(13)
110.8(3), C(13)–N(14)–Pt 119.1(3), N(2)–Pt–S(1) 167.56(14), N(2)–Pt–
Cl(1) 94.36(14), Cl(1)–Pt–S(1) 97.84(5), N(2)–Pt–N(14) 87.8(2), Pt–N(2)–
S(2) 127.5(3).
Chem. Commun., 1999, 1081–1082
1081

A number of aspects of this reaction are noteworthy. One
important feature is that it can be repeated using Pt(MeCN)2Cl2
in CH2Cl2; in this case it is clear that coordinated MeCN ligands
must be reacting as we no longer have the excess of the ligand
in the form of the solvent. Additionally, it should be noted that
1 does not react with MeCN in the absence of the metal centre.
Thus the solid retrieved after removal of the solvent from a
solution of 1 in MeCN after stirring for 1 h at either ambient
temperature, or indeed at reflux temperature, has an identical IR
spectrum to the starting material. The choice of metal is also
crucial; as we have already noted we have seen no evidence of
this type of reaction occurring during reactions involving Co
and Cu and attempts to observe analogous reactions using
Pd(MeCN)2Cl2 or [Pd(MeCN)4][BF4]2 also fail. Thus the
reaction would appear to be specific to Pt, at least amongst the
aforementioned metals. This brings us to the question of the
reaction mechanism. It is possible that the first stage of the
reaction involves coordination of one 1 ligand followed by
reaction with the remaining coordinated MeCN. At present we
regard this as a less likely option as it implies that coordination
to the metal is activating the sulfimide towards attack by MeCN.
If this were the case one might suspect that [Pt(Ph2SNH)4]Cl2
(from 1 + [PPh4]2[PtCl4] in CH2Cl2) would be prone to reaction
with this solvent; in fact it can be dissolved in boiling MeCN
and, after precipitation by cooling, shows no change in its IR
spectrum. In addition it should be remembered that 2 exhibits
one unreacted sulfimide ligand. The other possible route
involves nucleophilic addition of the sulfimide to the nitrilic
carbon followed by proton transfer and coordination of the
sulfur via loss of the remaining MeCN. This is conceptually a
very simple route, the only drawback coming from the fact that
it would require the sulfimide/acetonitrile addition to be faster
or more efficient than simple substitution. Some justification for
this comes with the observation that platinum appears to be less
effective (or at least slower) at binding 1 than lighter metals.
Thus while conversion of [Pd2X6]22 to [Pd(Ph2SNH)4]X2 takes
place in a matter of minutes, [Pt(Ph2SNH)4]Cl2 only forms from
[PPh4]2[PtCl4] over the course of many days. In addition, the
[Pt(PMe2Ph)2(Ph2SNH)Cl]+ cation forms a solution equilib-
rium with Pt(PMe2Ph)2Cl2 upon addition of Cl2.4 We thus
favour the second of the two reaction mechanisms; this
conclusion is backed up by the fact that nucleophilic addition to
coordinated nitriles is well documented.5 It should also be noted
that preliminary work indicates that this reaction can be
mirrored for other nitriles such as propionitrile and benzoni-
trile.
new imine-substituted sulfimide systems; work towards this end
is underway.
The authors acknowledge Johnson Matthey for loans of
precious metals.
Notes and references
† A solution of [PtCl2(MeCN)2] (61 mg, 0.18 mmol) in MeCN (20 ml) was
treated with a solution of 1 (71 mg, 0.35 mmol) in the same solvent (5 ml)
added over a period of 2 min with stirring. Upon continuous stirring the
resulting yellow solution suddenly precipitated a pale yellow solid (time
required for precipitation to start varied from experiment to experiment,
ranging from < 1 to 10 min; in all cases, however, once started precipitation
was very rapid). The mixture was stirred for a further 30 min then allowed
to settle whereupon the solvent was decanted, the solid washed with cold
MeCN (10 ml) and then Et2O (10 ml) and dried in vacuo to give a pale
yellow product. Yield 92 mg, 74%.
Recrystallisation from hot MeCN gave 60 mg (48% based upon original
amount of Pd used) of well formed crystalline material, though the solution
retained a significant yellow colour even after cooling indicating that some
reaction/decomposition had occurred during the heating process, hence the
lowering of yield. There was no significant difference in the IR spectra of
the crude and recrystallised material. X-Ray crystallography confirmed the
formulation as [Pt(Ph2SNH)(Ph2SNC(Me)NH)Cl]Cl with no solvated
MeCN present, though this X-ray data is not presented in full here. Found:
C, 42.6; H, 3.3; N, 5.9%. Calc. for C26H25Cl2N3PtS2: C, 44.0; H, 3.6; N,
5.9%. 1H NMR (CDCl3) d 10.4 (1H, s, NH), 7.99 (8H, m, CH), 7.63 (2H,
m, CH), 7.59 (4H, m, CH), 7.50 (6H, m, CH), 2.50 (3H, s, CH3); 13C NMR
(CDCl3) d 183.6 (NCN), 138.2, 135.7, 134.0, 132.1, 130.0, 129.9, 128.0
(phenyl groups), 19.0 (CH3).
A similar reaction occurred when PtCl2 was dissolved in hot MeCN
(effectively forming the same starting material in situ) and then 1 added. In
this case recrystallisation gave product with one MeCN of crystallisation
{i.e. [Pt(Ph2SNH)(Ph2SNC(Me)NH)Cl]Cl·MeCN} as shown by X-ray
crystallography and microanalysis (Found: C, 44.5; H, 3.9; N, 7.6%; Calc.
for C28H28Cl2N4PtS2: C, 44.8; H, 3.8; N, 7.5%). It is not clear why the two
recrystallisations gave solvated and unsolvated materials.
IR spectroscopy confirmed that an identical product is formed if the
Pt(MeCN)2Cl2 reaction is performed in CH2Cl2.
‡ Crystal data for 2: all measurements were made on a Siemens SMART
diffractometer with graphite monochromated Mo-Ka radiation. Data were
collected using small slices: 12830 data collected. C28H28Cl2N4PtS2, M =
750.65, monoclinic, space group P21/n, a = 13.223(1), b = 10.941(1), c =
20.912(1) Å, b = 100.302(1)°, U = 2976.5(1) Å3, Z = 4, T = 293 K.
m(Mo-Ka) = 5.11 mm21, l = 0.71073 Å, F(000) = 1472. R1 = 0.0268,
wR2
= 0.0486 [I > 2s(I)], 4267 independent reflections (4217 ob-
served).
CCDC 182/1247. See http://www.rsc.org/suppdata/cc/1999/1081/ for
crystallographic files in .cif format.
1 P. F. Kelly, A. M. Z. Slawin and A. Soriano-Rama, J. Chem. Soc., Dalton
Trans., 1996, 53 and references therein.
2 P. F. Kelly, A. M. Z. Slawin and K. W. Waring, J. Chem. Soc., Dalton
Trans., 1997, 2853.
3 P. F. Kelly, A. M. Z. Slawin and K. W. Waring, Inorg. Chem. Commun,
1998, 1, 249.
4 P. F. Kelly, A. C. Macklin, A. M. Z. Slawin, K. W. Waring and R. Yates,
in preparation.
5 B. N. Storhoff and H. C. Lewis, Coord. Chem. Rev., 1977, 23, 1; S. J.
Bryan, P. G. Huggett, K. Wade, J. A. Daniels and J. R. Jennings, Coord.
Chem. Rev., 1982, 44, 149.
6 T. Chivers, K. McGregor and M. Parvez, J. Chem. Soc., Chem. Commun.,
1993, 1021.
7 N. Feeder, R. J. Less, J. M. Rawson and J. N. B. Smith, J. Chem. Soc.,
Dalton Trans., 1998, 4091.
8 3D Search and Research Using the Cambridge Structural Database, F. H.
Allen and O. Kennard, Chem.-Des. Automat. News, 1993, 8, 1; 31.
9 J. P. Chupp, D. J. Dahm and K. L. Leschinsky, J. Heterocycl. Chem.,
1975, 12, 485.
The metallocycle formed in this reaction is a very rare
example of a transition metal based MNCNS cyclic system;
indeed there would only appear to be one other example of such
a unit characterised by X-ray crystallography, namely that
found in [Pt{PhSNC(MeC6H4)NNC(MeC6H4)NSPh}(PPh3)],6
though here it is a part of a tridentate ligand system (the X-ray
structure of recently formed [Pt{N(H)C(Ph)N(H)S}(dppe)]+
was not reported, though the species does indeed contain the
cyclic PtNCNS unit7). In fact we can extend this further and say
that with the exception of cases where X = S (the very well
known and much studied dithiadiazolyls) the only other
structure of the type cyclo-XNCNS to be found in a search of the
Cambridge Structural Database8 is one in which X = O.9
To conclude, we can say that the reaction of Pt(MeCN)2Cl2
with 1 provides an efficient, high yield route to a very rare
structural system, via a metal-assisted activation unprecedented
for a sulfimide. It is quite feasible that the neutral ligand thus
formed will prove to be labile, providing an efficient route to
Communication 9/01783F
1082
Chem. Commun., 1999, 1081–1082