PPh3
CO
CO
Ru
CO
CO
Ru
Ru3(CO)8H(m-H)(PPh3)3 fragment under hydrogen to yield
mononuclear ruthenium complexes under relatively mild condi-
tions. This observation is interesting, given that mixtures of
Ru3(CO)12 and PPh3 have been used in hydroformylation
catalysis and Ru(CO)3(H)2(PPh3) has been implicated as an
intermediate in hydroformylation.5,7 No para-hydrogen en-
hanced products, either tri- or mono-nuclear, were detected on
treating Ru3(CO)12 with p-H2 either in the presence or the
absence of UV irradiation.
Here, we have established some features of the ruthenium
carbonyl/triplenylphosphine/hydrogen system, which shows
some important differences from those with PMe2Ph and PMe3.
Furthermore, we have established the feasibility of using p-H2
to probe photochemical reactions, H2 addition to clusters, and
cluster fragmentations.
Ph3P
OC
CO
Ru PPh3
PPh3
CO
+CO/-PPh3
CO
Ru
Ru
H
CO
Ha
Hb
H
PPh3
Fragmentation
+H2, 20 min
CO
CO
CO
hν,
+H2
CO
Ph3P
1a
2a
CO
Ru(CO)3(PPh3)2
+H2 308 K,
immediate
Isomerisation
Isomerisation
CO
OC
CO
PPh3
CO
PPh3
CO
Ru
H
PPh3
CO
Ru
Ru
PPh3
Ph3P
Ru
H
H
CO
H
Ru
CO
OC
CO
PPh3
H
H
CO
CO
CO
3
1b
2b
Scheme 1
We are grateful to the EPSRC (C. J. S. and spectrometer), the
University of York (J. P. L.), Bruker UK (spectrometer) and the
Royal Society for financial support. We appreciated helpful
discussions with Professor R. N. Perutz, Dr M. K. Whittlesey
and Dr J. M. Lynam. A generous loan of ruthenium trichloride
from Johnson Matthey is also gratefully acknowledged.
perhaps a trigonal twist process similar to that reported for
Ru(CO)(H)2(PPh3)3.9
The interconversions between 2a, 2b and 1a were further
investigated using an authentic sample of 2a. When a benzene-
d6 solution of 2a was heated under 3 atm of p-H2 to 328 K, no
1
signals for 1a were visible in the H NMR spectrum, but the
hydride resonances for 2b were again observed [Fig. (1b)]. An
NMR tube containing a fresh solution of 2a was then placed
under 1 atm of CO, frozen and then filled with 3 atm of p-H2.
When the solution was warmed to 318 K, the enhanced hydride
resonances of 1a and 2b were observed [Fig. 1(c)]. Evidently 2a
readily undergoes replacement of PPh3 by CO. We have
recently adapted an NMR probe to allow a sample to be
irradiated by UV light from a HgXe arc while NMR spectra are
recorded.10 1H NMR spectra recorded during photolysis of a
benzene-d6 solution Ru(CO)3(PPh3)2 under 3 atm of p-H2 at 292
K showed large enhanced hydride signals due to 1a and a small
non-enhanced signal due to 2a. The signals for 1a were not
observed in the absence of UV radiation, confirming that, in this
case, hydrogen exchange between 1a and free H2 is photo-
chemical rather than thermal.
Many phosphine-substituted derivatives of Ru3(CO)12 have
been characterised and studied, but the reactions of these
complexes with hydrogen are less well understood.11 We have
therefore used p-H2 to monitor the reaction of Ru3(CO)9(PPh3)3
with hydrogen. Nevinger et al. used a complex synthetic
procedure to obtain a species believed to be Ru3(CO)11H(m-H),
whose 1H NMR spectrum contained hydride resonances at
d 211.79 and 18.55.12 We recorded the 1H NMR spectrum of
a benzene-d6 solution of Ru3(CO)9(PPh3)3 under p-H2 at 308 K,
and detected enhanced resonances at d 210.20 (dd, JPH 11.2,
JHH 26.0 Hz) and 217.64 (ddd, JPH 21.2,10.4, JHH 26.0 Hz)
for two mutually coupled hydrides in a species 3 [Fig. 1(d)]. The
chemical shifts and couplings to phosphorus suggested that the
former resonance represented a terminal hydride cis to a
phosphine and the latter a hydride bridging two ruthenium
atoms and cis to a phosphine ligand on each ruthenium. A 2D
experiment located the resonances for the corresponding 31P
nuclei at d 44.4 (coupled to both hydrides) and 217.6 (coupled
only to the terminal hydride). From this evidence it appeared
that 3 was the 48-electron cluster Ru3(CO)8H(m-H)(PPh3)3,
with the structure shown in Scheme 1. The third 31P nucleus was
not detected by the 2D experiment because it is not coupled to
either hydride. Complex 3, which is comparable to both
Notes and references
† Selected spectroscopic data for 1a, 1b, 2a, 2b and 3: NMR spectra in C6D6
at 400.13 MHz (1H) and 202.45 MHz (31P) recorded on 5 mm samples in a
5 mm inverse geometry probe. 1a: 1H dH(328 K) 26.67 [Ha, J(PH) 15.9,
J(COH) 27.8, J(HH) 25.4 Hz], 27.34 [Hb, J(PH) 61.6, J(HH) 25.4 Hz].
dP(328 K) 20.3 (s). dC(328 K); 199.5 d [J(PC) 8.4 Hz], 197.6 [d J(PC) 4.6
Hz]. 1b: dH(308 K) 26.68 [J(PH) 23 Hz], dP(328 K) 55.1 (s). 2a: dH(295 K)
26.35 [H, J(PH) 23.2 Hz]. dP(295 K) 57.7 (s). dC(295 K) 202.0 [t, J(PC)
8.3 Hz]. 2b: dH(328 K) 26.27 [Hc, J(PH) 26.4, 19.1, J(HH) 27.1 Hz],
27.52 [H, J(PH) 75.1, 33.0, J(HH) 27.1 Hz]. dP(328 K) 50.2 d [J(PP) 32
Hz], 41.6 [d, J(PP) 32 Hz]. 3: dH(300 K) 210.20 [H, J(PH) 11.2, J(HH 26.0
Hz], 217.64 [H, J(PH) 21.2, 10.4, J(HH) 26.0 Hz]. dP (308 K) 44.4 (s) and
31.2 (s).
1 C. R. Bowers and D. P. Weitekamp, J. Am. Chem. Soc., 1987, 109, 5541;
R. Eisenberg, Acc. Chem. Res., 1991, 24, 110; J. Natterer and J. Bargon,
Prog. Nucl. Magn. Reson. Spectrosc., 1997, 31, 293.
2 B. Duckett and R. Eisenberg. J. Am. Chem. Soc., 1993, 115, 5292; P. D.
Morran, S. A. Colebrooke, S. B. Duckett, J. A. B. Lohmann and R.
Eisenberg. J. Chem. Soc., Dalton Trans., 1998, 3363.
3 S. B. Duckett, R. J. Mawby and M. G. Partridge, Chem. Commun., 1996,
383.
4 N. Ahmad, J. J. Levison, S. D. Robinson and M. F. Uttley. Inorg. Synth.,
1974, 15, 50; M. I. Bruce, G. Shaw and F. G. A. Stone. J. Chem. Soc.,
Dalton Trans., 1972, 2094.
5 P. Kalck, Y. Peres and J. Jenck. Adv. Organomet. Chem. 1991, 32,
121.
6 R. Whyman, J. Organomet. Chem., 1973, 56, 339.
7 E. M. Gordon and R. Eisenberg, J. Organomet. Chem., 1986, 306,
C53.
8 S. B. Duckett, G. K. Barlow, M. G. Partridge and B. A. Messerle,
J. Chem. Soc., Dalton Trans. 1995, 20, 1427; S. Hasnip, S. B. Duckett,
D. R. Taylor and M. J. Taylor. Chem. Commun., 1998, 923.
9 G. E. Ball, and B. E. Mann, J. Chem. Soc., Chem. Commun., 1992, 561.
We note that although we have no direct evidence that 1a and 1b
interconvert, this is the case for the PMe3, AsMe2Ph and PMe2Ph
analogues of 2.
10 This will be reported fully elsewhere, however, we note that two UV
transmitting liquid light guides were employed in conjunction with a
modified narrow-bore probe that was used in a wide-bore magnet.
11 M. I. Bruce, M. J. Liddell, O. bin Shawkataly, C. A. Hughes, B. W.
Skelton and A. H. White, J. Organomet. Chem., 1988, 347, 207.
12 L. R. Nevinger, J. B. Keister and J. Maher, Organometallics, 1990, 9,
1900.
Ru3(CO)11H(m-H) and Os3(CO)10H(m-H)L (L
= PPh3 or
CD3CN),13 is presumably formed by a simple substitution of
CO by H2.
13 S. Aime, R. Gobetto and D. Canet, J. Am. Chem. Soc., 1998, 120, 6770;
A. J. Deeming, Adv. Orgnaomet. Chem., 1986, 26, 1.
After 20 min at 308 K, resonances for isomer 1a of
Ru(CO)3H2(PPh3) had appeared in the 1H NMR spectrum of the
solution. Consequently, either unreacted Ru3(CO)9(PPh3)3 or
Communication 9/03321A
1224
Chem. Commun., 1999, 1223–1224