A metallacyclic λ5-phosphaalkenyl complex of ruthenium(II): X-ray structure of [Ru {κ2-P(=O)CButC(=O)} (CNBut)2(PPh3)2]

Article abstract of DOI:10.1039/a707987g

The reaction of [Ru(P=CHBu^t)Cl(CO)(PPh₃)₂] 1 or [Ru(P=CHBu^t)Cl(CNBu^t)(CO)(PPh₃)₂] 2b with excess pivalo isonitrile under aerobic conditions provides the novel metallacyclic λ⁵-phosphaalkenyl-P complex [Ru{κ²-P(=O)CBu^tC(=O)}(CNBu^t) ₂(PPh₃)₂], which has been crystallographically characterised.

Full text of DOI:10.1039/a707987g

View Article Online / Journal Homepage / Table of Contents for this issue
5
A metallacyclic l -phosphaalkenyl complex of ruthenium(ii): X-ray structure
2
of [Ru{k -P(NO)CButC(NO)}(CNBut)2(PPh3)2]
Anthony F. Hill,*†a Cameron Jones,*‡b Andrew J. P. White,a David J. Williamsa and James D. E. T.
Wilton-Elya
a Department of Chemistry, Imperial College of Science Technology and Medicine, South Kensington, London, UK SW7 2AY
b Department of Chemistry, University of Wales, Swansea, Singleton Park, Swansea, UK SA2 8PP
The reaction of [Ru(PNCHBut)Cl(CO)(PPh3)2]
1
or
product if an excess of CNBut is used, under aerobic conditions.
Spectroscopic data§ and a crystallographic study (Fig. 1)¶
confirm the identity of the new compound as the novel
[Ru(PNCHBut)Cl(CNBut)(CO)(PPh3)2] 2b with excess piv-
alo isonitrile under aerobic conditions provides the novel
5
2
5
2
metallacyclic l -phosphaalkenyl-P complex [Ru{k -
P(NO)CButC(NO)}(CNBut)2(PPh3)2], which has been
crystallographically characterised.
metallacyclic l -phosphaalkenyl-P complex [Ru{k -P(NO)C-
ButC(NO)}(CNBut)2(PPh3)2] 3. Of note amongst the spectro-
scopic data for 3, is the 31P{1H} NMR resonance for the
phosphaalkenyl centre which appears as a triplet [d 47.0 2J(P2P)
25.2 Hz], to substantially higher field of those observed for 2b
[d 389.8 2J(P2P) 11.7 Hz] or 1 [d 450.4 2J(P2P) 10.0 Hz]. The
phosphoryl group contributes to a strong absorption in the
infrared spectrum at 1198 cm21(Nujol), whilst the acyl group is
apparent at 1644 cm21. All other spectroscopic data are as
expected and unremarkable.
3
The
l -phosphaalkenyl-P
complex
[Ru(PNCHBut)Cl-
(CO)(PPh3)2] 11 is intriguing in that despite effective atomic
number considerations for ruthenium, both the spectroscopic
features and the emerging reactivity profile2–4 point to a
nucleophilic phosphorus centre (A). The RuCl(CO)(PPh3)2
X
The geometry at ruthenium is essentially octahedral with cis-
interligand angles in the range 85.2(2)–96.1(3)°, the exception
being P(1)–Ru–C(2) which is contracted to 66.5(2)° by virtue of
the constraints of chelation. The two ruthenium isonitrile
distances are 1.988(10) and 2.037(10) Å suggesting that the
trans influence of the phosphaalkenyl ligand is comparable to
that of the acyl component of the metallacycle. The principle
structural feature of interest is the metallacycle, the unsaturation
of which is reflected in the coplanarity to within 0.03 Å of the
atoms C(5), C(1), P(1) O(3), C(2), O(4) and Ru, a planarity
which extends to include the CN groups of the isonitriles.
Owing to the novelty of 3, very little directly comparable
Ln +1M P
Ln M
P CR2
Ln +1M P
CR2
CR2
A
X = CR2, O, NR
B
C
CH2
C
O
P
(Ph3P)2(OC)2Os
C
CF3
(Pri3P)2ClRh
C CMe3
P
D
E
fragment, possessing 15 valence electrons might be expected to
commandeer three electrons from the phosphaalkenyl ligand B
and thereby enforce linearity and attendant electrophilic
character at phosphorus. This is, however, not the case. We have
interpreted this counter-intuitive behaviour with reference to a
similar dichotomy which prevails for nitrosyl ligands bound to
metal centres with high d-occupancies.1 One aspect of the
‘semi-bent’ nature of the phosphaalkenyl ligand in 1 is that the
ruthenium centre readily, though reversibly, accepts two-
electron ligands, resulting in a metal centre which requires bent
(one-electron) phosphaalkenyl coordination. The resulting
nucleophilic nature of the phosphorus centre has been demon-
strated by its reduction to a complex of the unusual fluoro-
phosphine ButCH2PHF ligand via the reaction of the isonitrile
adduct [Ru(PNCHBut)Cl(CNC6H3Me2-2,6)(CO)(PPh3)2] 2a
with HBF4.2 Herein we wish to report a curious transformation
of 1 into a metallacyclic phosphaalkenyl complex 3. This is
3
structural data exists. Two metallacyclic l -phosphaalkenyl
complexes have been structurally characterised (D5 and E),6
P(2)
N(16)
O(3)
C(1)
C(15)
P(1)
3
5
accompanied by oxidation of the phosphorus from l to l (C),
resulting in a rare example of a trigonal phosphorus centre
surrounded by three p-interactive substituents. A notable
feature of 3 is that it is, we believe, the first example of a
Ru
C(2)
C(5)
C(9)
N(10)
O(4)
5
l -phosphaalkenyl-P complex.
P(3)
Whilst complex 1 readily forms a 1:1 adduct 2a with
CNC6H3Me2-2,6, the same reaction with pivalo isonitrile
(CNBut) is somewhat more complex. Under strict control of
reagent stoichiometry and reaction conditions it is possible to
prepare the adduct [Ru(PNCHBut)Cl(CNBut)(CO)(PPh3)2] 2b,
spectroscopic data for which are comparable to those for 2a.§
On occasion however, complex 2b is contaminated with a
second product, which is also formed in low yield from 2b on
standing in solution. This second compound is the exclusive
Fig. 1 Crystal structure of 3
Chem. Commun., 1998
367

View Article Online
3
though perhaps more relevant here are the structural features of
the phosphaalkene complexes [Ru{P(AuPPh3)NCHBut}Cl2-
(CO)(PPh3)2] 43 and [Ru(PMeNCHBut)ClI(CO)(PPh3)2] 54
where the Ru–P separations are 2.296(2) and 2.280(2) Å,
respectively. The Ru–P(1) bond length of 2.350(2) Å in 3 thus
indicates a substantially reduced degree of Ru–P multiple bond
character, consistent with a perhaps surprising decrease in
The chemistry of l phosphaalkenyl ligands has seen
substantial growth in recent times.1–6,9 With the advent of a l
5
example, it will be interesting to see how their respective
coordination chemistries compare.
This work was supported by the EPSRC in the form of a
studentship (to J. D. E. T. W.-E.) and a diffractometer and by the
Nuffield Foundation (to C. J.). A. F. H. gratefully acknowledges
the Leverhulme Trust and the Royal Society for the award of a
Senior Research Fellowship.
5
apparent p-acidity for the l -phosphorus centre, notwithstand-
ing the perturbations associated with chelation. This counter-
intuitive result is possibly due to the co-ordination of a
competitive p-acceptor trans to P(1), whilst the phosphaalkene
ligands in 4 and 5 are trans to p-donor ligands. The P(1)–C(1)
separation of 1.713(11) Å is marginally longer than the PNC
bond lengths of 1.664(9) and 1.657(8) Å, found for 4 and 5,
respectively but falls within the range 1.68–1.72 Å found for
free phosphaalkenes,7 indicating substantial double bond char-
acter. In contrast the C(1)–C(2) bond at 1.54(1) Å is long for a
single Csp2–Csp2 bond length.
Notes and References
† E-mail: a.hill@ic.ac.uk
‡ E-mail: c.a.jones@swansea.ac.uk
§ Selected data for new complexes [25 °C, IR Nujol(CH2Cl2), NMR
(CDCl3), satisfactory microanalytical data obtained]. 2b: IR: 2148 (2148)
[n(CN)], 1930 (1961) [n(CO)] cm21. NMR: 1H d 0.79 (s, 9 H, PCCCH3),
0.97 (s, 9 H, NCCH3), 7.24–7.95 (m, 31 H, C6H5 + PNCH) 31P{1H} d 389.8
2
2
[t, J(P2P) 11.7 Hz], 24.6 [d, J(P2P) 11.7 Hz]. FABMS: m/z 874 [MH]+,
772 [MH 2 HPNCHBut]+, 744 [MH 2 HPNCHBut 2 CO]+, 709 [MH 2
HPNCHBut 2 Cl 2 CO]+. 3: IR: 2169, 2028 (2171, 2038) [n(CN)], 1644
(1606) [n(CO)] cm21. NMR 1H d 0.61 (s, 9 H, PCCCH3), 0.86 (s, 9 H,
NCCH3), 1.27 (s, 9 H, NCCH3) 7.27–8.09 (m, 30 H, C6H5). 31P{1H} d 47.0
[t, 2J(P2C) 25.2 Hz], 31.2 [dd, 2J(P2P) 25.2, 8.4 Hz]. FABMS: m/z 856 [M
2 CNBut]+, 602 [M 2 CNBut 2 PPh3]+. 4: IR: 2184(sh), 2163 (2179, 2156)
[n(CN)], 2003, 1980(sh) (2021) [n(CO)] cm21. NMR: 1H d 0.66 (s, 9 H,
PCCCH3), 0.96 (s, 9 H, NCCH3), 1.18 (s, 9 H, NCCH3), 7.32–7.76 (m, 31
H, C6H5 + PNCHBut). 31P{1H} d 336.9(s), 33.5(s). FABMS: m/z 921 [M]+,
838 [M 2 CNBut]+, 810 [M 2 CNBut] 2 CO]+, 530 [M 2 CNBut 2
PPh3]+.
It remains for the mechanism to be established whereby 3
forms from 1 or 2b, however we would make the following
points which taken together support the route proposed in
Scheme 1. The reaction proceeds in polar solvent mixtures
suggesting ruthenium–chloride ionisation occurs. This is sup-
ported by the formation and isolation of the salt cis,cis,trans-
[Ru(PNCHBut)(CO)(CNBut)2(PPh3)2]Cl 4§ when 1 is treated
with 2 equiv. of pivaloisonitrile and isolated immediately. The
b-position of vinyl ligands is typically nucleophilic in nature, in
particular for later transition metals, and it seems reasonable to
expect a similar property for phosphavinyl ligands. The
carbonyl ligand will be activated towards nucleophilic attack as
a result of the complex being cationic. Ring closure could
provide the saturated metallacycle shown, and the proton which
is a to both phosphorus and a carbonyl group would be expected
to be acidic. Deprotonation then leads to unsaturation of the
metallacycle. The aerial oxidation of the phosphorus centre is,
in contrast to 1, an endearing feature of which is its remarkable
aerobic stability. Nevertheless, the Ru(CNBut)2(PPh3)2 frag-
ment would be expected to be particularly p-basic, activating
the p-acid phosphorus centre towards oxidation. Notably the
conversion of 4 to 3 is accelerated by addition of a non-
nucleophilic base (DBU). We have so far been unsuccessful in
isolating the intermediates between 4 and 3, however the
alternative route of deprotonation prior to cyclisation seems less
favourable, given that it would produce a 20-valence electron
phosphaalkyne complex of zerovalent ruthenium.8
¶ Crystal data for 3: C52H57N2O2P3Ru·CH2Cl2, M = 1020.9, triclinic,
space group P1 (no. 2), a = 13.616(1), b = 14.629(2), c = 15.473(3) Å,
¯
a = 89.93(1), b = 89.39(1), g = 66.72(1)°, U = 2830.9(7) Å3, Z = 2, Dc
= 1.198 g cm23, m(Cu-Ka) = 42.0 cm21, l = 1.54178 Å, F(000) = 1060.
A colourless prism of dimensions 0.27 3 0.20 3 0.05 mm was used. Data
were measured on a Siemens P4/PC diffractometer with graphite mono-
chromated Cu-Ka radiation (w-scans). 7804 Independent reflections were
measured (2q @ 116°) of which 5605 had ıFoı > 4s(ıFoı) and were
considered to be observed. The structure was solved by direct methods and
all the major occupancy non-hydrogen atoms of the complex were refined
anisotropically by full-matrix least squares based on F2 using absorption-
corrected data to give R1 = 0.082, wR2 = 0.205 for the observed data and
530 parameters. The somewhat high R factors are a consequence of disorder
in the But substituents and the included solvent molecule. CCDC
182/727.
1 R. B. Bedord, A. F. Hill and C. Jones, Angew. Chem., Int. Ed. Engl., 1996,
35, 547.
2 R. B. Bedford, D. E. Hibbs, A. F. Hill, M. B. Hursthouse, K. M. A. Malik
and C. Jones, Chem. Commun., 1996, 1895.
3 R. B. Bedford, A. F. Hill, C. Jones, J. D. E. T. WIlton-Ely, A. J. P. White
and D. J. Williams, Chem. Commun., 1997, 179.
4 R. B. Bedford, A. F. Hill, C. Jones, J. D. E. T. Wilton-Ely, A. J. P. White
and D. J. Williams, J. Chem. Soc., Dalton Trans., 1997, 139; 1113.
5 D. S. Bohle, G. R. Clark, C. E. F. Rickard and W. R. Roper,
J. Organomet. Chem., 1988, 353, 355.
L
L
R
Cl
Ru
Cl
i
P
P
Ru
L
R
RNC
CO
OC
L
1
2b
i
6 P. Binger, J. Haas, A. T. Herrmann, F. Langhauser and C. Kru¨ger, Angew.
Chem., Int. Ed. Engl., 1991, 30, 310.
7 R. Appel, F. Knoll and I. Ruppert, Angew. Chem., Int. Ed. Engl., 1981,
20, 731; A. H. Cowley, R. A. Jones, J. G. Lasch, N. C. Norman,
C. A. Stewart, A. L. Stuart, J. L. Atwood, W. E. Hunter and H.-M. Zhang,
J. Am. Chem. Soc., 1984, 106, 7015.
L
L
R
H
R
+
RNC
+
RNC
RNC
P
P
i
Ru
Ru
RNC
C
O
CO
L
L
ii
4
8 The complex [Ru(P·CC6H2But3)(CO)2(PPh3)2] has however been de-
scribed recently: R. B. Bedford, A. F. Hill, M. D. Francis, C. Jones and
J. D. E. T. Wilton-Ely, Inorg. Chem., 1997, 36, 5142.
9 For a recent review, see: L. Weber, Angew. Chem., Int. Ed. Engl., 1996,
35, 271.
O
L
L
RNC
RNC
iii
RNC
RNC
P
P
Ru
L
R
Ru
L
R
C
O
C
O
3
Scheme 1 L = PPh3, R = But. Reagents: i, CNR; ii, 2HCl; iii, O2.
Received in Cambridge, UK, 6th November 1997; 7/07987G
368
Chem. Commun., 1998