siderably larger than that found in the precursor 1a (10.0 Hz).
The resonances due to the phosphine ligands both show
coupling to the phosphaalkenyl phosphorus, however there is no
resolvable coupling between the phosphine resonances con-
sistent with the absence of an Ru–Au bond.
We gratefully acknowledge the generous loan of ruthenium
salts by Johnson Matthey Chemicals and the Nuffield Founda-
tion and the Engineering and Physical Sciences Research
Council (UK) for financial support and the provision of a
diffractometer.
The formulation was confirmed by a single-crystal X-ray
diffraction analysis,¶ the results of which are summarised in
Fig. 1. The geometry at ruthenium is essentially octahedral with
cis-interligand angles in the range 87.2(3)–93.0(1)°, whilst the
geometry at gold is close to linear [P(1)–Au–P(20) 177.6(1)°].
The two ruthenium chloride distances are 2.481(2) and 2.464(2)
Å indicating that the trans influence of the phosphaalkenyl
ligand is only marginally greater than that of the carbonyl
ligand. Interest focusses on the phosphaalkenyl bridge although
the novelty of 3a leaves us with little precedent for comparison.
It is noteworthy that the coordination at phosphorus is trigonal
[intersubstituent angles in the range 119.7(3)–120.2(3)°],
indeed this planarity extends to include Au, P(1), C(22) and the
equatorial ruthenium coordination plane [maximum deviation
from planarity of 0.085 Å by Cl(66)]. The P(20)–C(21) bond
length of 1.664(9) Å is clearly multiple in nature, and
comparable to that observed [1.640(4) Å] in [Fe{m-PNC(Si-
Me3)2}2(CO)6].3 The two Au–P distances differ only margin-
ally, with that to the phosphaalkenyl ligand [2.320(2) Å] being
the longer of the two. The Ru–P(20) separation of 2.296(2) Å is
however ca. 0.1 Å shorter than those to the phosphines [P(45)
2.397(2), P(26) 2.413(2) Å], indicating a p-acid role for the
phosphaalkenyl ligand when bound to a retrodative metal
centre. Finally the trans disposition of the ruthenium and the
butyl group about the PNC double bond confirms our earlier
suggestion that hydroruthenation of P·CBut to provide 1a
occurs in a regiospecifically trans manner.
The facility of the addition of [AuCl(PPh3)] to 1a bodes well
for the future elaboration of heterobimetallic phosphaalkenyl
complexes by the bridge-assisted methodology. Preliminary
results indicate that the gold acetylide [Au(C·CC6H4Me-
4)(PPh3)] also adds across the Ru–P bond to provide the
ruthenium acetylide complex [Ru{P(AuPPh3)NCHBut}-
(C·CC6H4Me-4)Cl(CO)(PPh3)3] 4a§ (Scheme 2). In a similar
manner both mercury-(i) and -(ii) chloride react with 1 to
provide [Ru{P(HgCl)NPHBut}Cl2(CO)(PPh3)2] 5a, the former
reaction being accompanied by deposition of elemental mer-
cury. Bis(alkynyl)mercurials [Hg(C·CR)2], however, fail to
react with 1a in contrast to [Au(C·CC6H4Me-4)(PPh3)].
Similar reactions, and the products [Ru{P(AuPPh3)NCH-
But}XCl(CS)(PPh3)2] [X = Cl 3b, C·CC6H4Me-4 4b] and
[Ru{P(HgCl)NPHBut}Cl2(CS)(PPh3)2] 5b are also observed
between these reagents and the thiocarbonyl complex
[Ru(PNCHBut)Cl(CS)(PPh3)2] 1b, obtained from the reaction of
[RuHCl(CS)(PPh3)3] and P·CBut.
Footnotes
† E-mail: a.hill@ic.ac.uk
‡ E-mail: c.a.jones@swansea.ac.uk
§ Spectral data: for 3a, IR/cm21 (Nujol) 1950 [n(CO)]; (CH2Cl2) 1960
[n(CO)]. NMR (CD2Cl2, 298 K) 1H: d 0.95 (s, 9 H, Me), 7.22, 7.53, 7.99 (m
3 3, 46 H, Ph + PNCH). 13C{1H}: d 199.4 [d, CO, J(PC) 10.7 Hz], 182.9
[d, PNC, J(PC) 19.6 Hz], 136–127 (Ph), 39.8 [d, CMe3, J(PC) 9.0 Hz], 32.3
[d, CH3, J(PC) 10.7 Hz]. 31P{1H}: d 319.4 [dt(br), PNC, J(PP) 268.5, J(PPA)
ca. 30 Hz], 37.8 [d, AuPPh3, J(PP) 265.8 Hz], 18.7 (d, RuPPh3, J 29.6 Hz).
FABMS: m/z (%) [assignment]; 1379(11) [Ru2Cl2(CO)2(PPh3)4]+,
1248(17) [M 2 HCl]+, 1148(0.5) [M 2 HPNCClCMe3], 1188(6) [H2-
Ru2Cl2(CO)2(PPh3)3]+, 987(11) [M 2 HCl 2 PPh3]+, 887(1) [M 2
HPNCClCMe3 2 PPh3]+, 689(4) [RuCl(CO)(PPh3)2]+, 654(11) [Ru-
(CO)(PPh3)2]+, 626 [HRu(PPh3)2]+, 459(20) [AuPPh3]+, 363(8) [RuPPh3]+,
263(18) [HPPh3]+.
4a, IR/cm21 (Nujol) 2094 [n(C·C)], 1953 [n(CO)], 817 [d(C6H4)];
(CH2Cl2) 2095 [n(C·C)], 1954 [n(CO)]. NMR (CD2Cl2, 298 K) 1H: d 0.98
[s, 9 H, C(CH3)3], 2.25 [s, 3 H, C6H4CH3], 6.79, 6.88 [(A B)2, 4 H, C6H4,
J(AB) 7.9 Hz], 7.55, 8.15 (m 3 2, 45 H, Ph), 7.64 [d, 1 H, PNCH, J(PH)
23.1 Hz]. 13C{1H}: d 200.4 [d, CO, J(PC) 10.7 Hz], 186.3 [s(br), PNC],
135.6–126.8 (Ph and RuC·C), 117.5 [d, RuC·C, J(PC) 19.4 Hz], 39.5 [d,
CMe3, J(PC) 14.3 Hz], 32.3 [d, C(CH3)3, J(PC) 11.1 Hz], 20.9 (C6H4CH3).
31P{1H}: d 312.8 [dt, PNC, J(PP) 227.5, J(PPA) 27.7 Hz], 40.0 [d, AuPPh3,
J(PP) 227.5 Hz], 26.4 (d, RuPPh3, 27.7 Hz). FABMS: m/z (%)
[assignment]; 1499(2) [M + nba 2 H2O]+, 1470(2) [M + nba 2 CO 2 H]+,
1364(0.3) [M]+, 1329(2) [M 2 Cl]+, 1248(4) [M 2 HCCC7H7]+, 1118(24)
[M 2 HCCC7H7 2 HPCHBut 2 CO]+. 5a, IR/cm21 (Nujol) 1981(sh), 1967
[n(CO)], 1260m, 1027m, 803m; (CH2Cl2) 1980 [n(CO)]. NMR (CD2Cl2,
298 K) 1H: d 0.85 [d, 9 H, CH3, J(PH) 2.2 Hz], 7.40–7.97 (m 3 4, 31 H,
PNCH and Ph). 13C{1H}: d 197.3 (m, CO), 178.2 [d, PNC, J(PC) 26.7 Hz],
127.2–134.8 (Ph), 40.6 [d, CMe3, J(PC) 14.3 Hz], 30.4 [d, CH3, J(PC) 14.2
Hz]. 31P{1H}: 257.9 [t, PNC, J(PP) 33.3, J(HgP) 7757 Hz], 18.2 (d, PPh3,
27.7 Hz). FABMS: m/z (%) [assignment]; 1025(1) [M 2 Cl]+, 689(14)
[RuCl(CO)(PPh3)2]+, 654(7) [Ru(CO)(PPh3)2]+, 625(7) [Ru(PPh3)2]+,
363(19) [RuPPh3]+.
¶ Crystal data for 3a: C60H55AuCl2OP4Ru·2CH2Cl2, M 1454.7,
monoclinic, space group P21/n, 10.276(3), b = 29.300(7),
=
a
=
c = 20.772(6) Å, b = 101.50(2)°, U = 6129(3) Å3, Z = 4, Dc = 1.577
g cm23, m(Mo-Ka) = 30.4 cm21, l = 0.710 73 Å, F(000) = 2896. A
yellow prism of dimensions 0.83 3 0.67 3 0.27 mm was used. Data were
measured on a Siemens P4/PC diffractometer with graphite monochromated
Mo-Ka radiation (w-scans). 7949 Independent reflections were measured
(2q @ 45°) of which 5644 had ıFoı > 4s(ıFoı) and were considered to be
observed. The structure was solved by direct methods and the non-hydrogen
atoms were refined anisotropically by full-matrix least squares based on F2
using absorption-corrected data to give R1 = 0.047, wR2 = 0.098 for the
observed data and 578 parameters. Atomic coordinates, bond lengths and
angles, and thermal parameters have been deposited at the Cambridge
Crystallographic Data Centre (CCDC). See Information for Authors, Issue
No. 1. Any request to the CCDC for this material should quote the full
literature citation and the reference number 182/308.
The chemistry described for the complexes 1 illustrates their
unusual nature which makes them distinct from other phospha-
alkenyl complexes for which the effective atomic number rule
holds domain. For such complexes, the linear three-electron
(electrophilic at P) or bent one-electron (nucleophilic at P) role
of the phosphaalkenyl ligand is distinctly dichotomous. We
have previously suggested2 that the complexes 1 represent a
special case: linearisation of the M = PNC linkage, and the
attendant reduction of the nucleophilicity of the phosphorus is
not apparently favoured by the 15-electron ‘RuCl(CE)(PPh3)2’
fragment. Such behaviour is typical of formally isoelectronic
nitrosyl complexes of the late-transition metals. This leads in
the unique case of 1 to the juxtaposition of a nucleophilic
phosphaalkenyl phosphorus adjacent to a coordinatively un-
saturated ruthenium centre, i.e. a 1,2-dipole predisposed to the
1,2-addition of dipolar reagents.
References
1 L. Weber, Angew. Chem., Int. Ed. Engl., 1996, 35, 271 and references
therein.
2 R. B. Bedford, A. F. Hill and C. Jones, Angew. Chem., Int. Ed. Engl.,
1996, 35, 547.
3 A. M. Arif, A. H. Cowley and S. Quashie, J. Chem. Soc., Chem.
Commun., 1985, 428.
4 E. Niecke, H.-J. Metternich, M. Nieger, D. Gudat, P. Wenderoth,
W. Malisch, C. Hahner and W. Reich, Chem. Ber., 1993, 126, 1299.
Received, 14th October 1996; Com. 6/06987H
180
Chem. Commun., 1997