Trinuclear clusters of osmium containing monodentate, chelating and orthometallated forms of 2,2′-bis(diphenyIphosphino)-1,1′-binaphthyl (BINAP)

Article abstract of DOI:10.1039/a806430j

The trinuclear cluster [Os₃(CO)₁₁{(A)-BINAP}] 1 containing monodentate (R)-BINAP is formed by thermal treatment of [Os₃(CO)₁₂] with 2,2′-bis(diphenylphosphino)-1,1′-binaphthyl, (R)-BINAP. This complex decarbonylates thermally to give the cluster [Os₃(CO)₁₀{(R)-BINAP}] 2, which was shown spectroscopically to contain chelating BINAP. On further thermal decarbonylation orthometallation at a phenyl group takes place to give [Os₃(μ-H)(CO)₈{μ₃-(R)-BINAP - H}] 3. There is no evidence for restricted rotation about the P-Ph bonds within the PPh₂ groups as we have previously observed for [Ru₃(μ-OH)₂(CO)8{μ-(R)-BINAP}]. This restricted rotation is an effect of crowding and appears to be a distinct property of μ-coordination rather than chelation of BINAP. However, both clusters 1 and 2 are fluxional, but as a result of different processes. Cluster 1 exists as two interconverting isomers 1A and 1B (³¹P NMR evidence) which we believe have different conformations within the monodentate (R)-BINAP ligand. Our observations are consistent with restricted rotations about the Ph₂P-naphthyl bonds. The decacarbonyl cluster 2 has C₂ symmetry and shows ¹³C NMR coalescences consistent with a dynamic merry-go-round motion at the two Os(CO)₄ units but not at the Os(CO)₂(BINAP) unit. Single-crystal structures of clusters 1 and 3 confirm these two new modes of BINAP coordination.

Full text of DOI:10.1039/a806430j

View Article Online / Journal Homepage / Table of Contents for this issue
Trinuclear clusters of osmium containing monodentate, chelating
and orthometallated forms of 2,2Ј-bis(diphenylphosphino)-1,1Ј-
binaphthyl (BINAP)
Antony J. Deeming* and Marc Stchedroff
Department of Chemistry, University College London, 20 Gordon Street, London,
UK WC1H OAJ. E-mail: a.j.deeming@ucl.ac.uk
Received 14th August 1998, Accepted 14th September 1998
The trinuclear cluster [Os3(CO)11{(R)-BINAP}] 1 containing monodentate (R)-BINAP is formed by thermal
treatment of [Os3(CO)12] with 2,2Ј-bis(diphenylphosphino)-1,1Ј-binaphthyl, (R)-BINAP. This complex
decarbonylates thermally to give the cluster [Os3(CO)10{(R)-BINAP}] 2, which was shown spectroscopically
to contain chelating BINAP. On further thermal decarbonylation orthometallation at a phenyl group takes place
to give [Os3(µ-H)(CO)8{µ3-(R)-BINAP Ϫ H}] 3. There is no evidence for restricted rotation about the P–Ph bonds
within the PPh2 groups as we have previously observed for [Ru3(µ-OH)2(CO)8{µ-(R)-BINAP}]. This restricted
rotation is an effect of crowding and appears to be a distinct property of µ-coordination rather than chelation of
BINAP. However, both clusters 1 and 2 are fluxional, but as a result of different processes. Cluster 1 exists as two
interconverting isomers 1A and 1B (31P NMR evidence) which we believe have different conformations within the
monodentate (R)-BINAP ligand. Our observations are consistent with restricted rotations about the Ph2P–naphthyl
bonds. The decacarbonyl cluster 2 has C2 symmetry and shows 13C NMR coalescences consistent with a dynamic
merry-go-round motion at the two Os(CO)4 units but not at the Os(CO)2(BINAP) unit. Single-crystal structures of
clusters 1 and 3 confirm these two new modes of BINAP coordination.
Chiral diphosphines as chelating ligands have been extensively
used in transition-metal-catalysed asymmetric reactions of
organic substrates.1 A notable example is the application of
2,2Ј-bis(diphenylphosphino)-1,1Ј-binaphthyl (BINAP)2 with
ruthenium() and rhodium() to enantioselective catalysis,
giving high enantiomeric excesses in many cases.3 Structural
studies have been carried out for a number of complexes in
which BINAP is chelating.4 Even in the cluster [Ru4(µ-H)4-
(CO)10{(S)-BINAP}], formed by direct reaction of [Ru4(µ-H)4-
(CO)12] with (S)-BINAP, the ligand is chelating.5 We have
recently characterized the first example of a bridging BINAP
ligand in the cluster [Ru3(µ-OH)2(CO)8{µ-(R)-BINAP}], which
interestingly is crowded to the extent of having restricted
rotation about two of the four P–Ph bonds.6 We now report our
results on introducing (R)-BINAP into triosmium clusters to
give products with three different modes of coordination.
Monodentate BINAP is found in [Os3(CO)11{(R)-BINAP}] 1,
chelating BINAP in [Os3(CO)10{(R)-BINAP}] 2, and ortho-
metallated BINAP in the cluster [Os3(µ-H)(CO)11{(R)-
BINAP Ϫ H}] 3. The modes of coordination in 1 and 3 are
unprecedented and were confirmed by crystal structure
determination. Interestingly we were unable to identify an iso-
mer of 2 containing bridging BINAP. This is surprising in view
of the bridging nature of BINAP in [Ru3(µ-OH)2(CO)8{µ-(R)-
BINAP}]6 and in [Os3(µ-H)2(CO)8{µ-(R)-TolBINAP}],7 where
TolBINAP is the 4-tolyl analogue of BINAP.
One of the objectives of this work, still to be realised, is the
incorporation of a chelating chiral diphosphine at one metal
centre or a bridging chiral diphosphine at two metal centres in
clusters and the examination of the reactivity at the remaining
metal atoms to establish whether chiral selectivity can be
transmitted across a cluster and between metal atoms. In this
Results and discussion
paper we report our initial characterization of some BINAP
derivatives of triosmium clusters and are able to report two new
modes of BINAP coordination in addition to chelation and
bridging.
Treatment of [Os3(CO)12] with (R)-BINAP in refluxing hydro-
carbon solvent (heptane or octane) gave three isolable products
which are air and thermally stable and can be separated on
silica. Similar products but in different yields were obtained
J. Chem. Soc., Dalton Trans., 1998, 3819–3824
3819

View Article Online
Fig. 1 Molecular structure of [Os3(CO)11{(R)-BINAP}] 1.
for coordination in any but an equatorial position.8 Axial sites
are crowded as a result of interactions between axial ligands
and, although axial substitution is probably favoured elec-
tronically, it is only observed with small or rod-shaped ligands.9
Relative to the structure of (R)-BINAP in the crystal,6 there
has been a rotation about one naphthyl–PPh2 bond to allow
coordination to Os(1). In free (R)-BINAP the two lone-pairs
are directed inwards and the uncoordinated PPh2 group in 1 has
a similar conformation to that of the PPh2 groups in the crystal
structure of the free ligand.
Table 1 Selected bond lengths (Å) and angles (Њ) for the cluster
[Os3(CO)11{(R)-BINAP}] 1
Os(1)–Os(2)
Os(2)–Os(3)
Os(1)–Os(3)
Os(1)–P(1)
P(1)–C(101)
2.989(2)
2.880(2)
2.915(2)
2.405(6)
1.80(3)
P(1)–C(111)
P(1)–C(201)
P(2)–C(121)
P(2)–C(131)
P(2)–C(211)
1.83(3)
1.88(2)
1.83(2)
1.84(2)
1.80(2)
Os(1)–Os(3)–Os(2)
Os(2)–Os(1)–Os(3)
Os(1)–Os(2)–Os(3)
Os(3)–Os(1)–P(1)
Os(2)–Os(1)–P(1)
Os(1)–P(1)–C(101) 117.2(8)
Os(1)–P(1)–C(111) 114.9(8)
60.00(4)
59.40(4)
60.60(4)
102.8(2)
161.9(2)
Os(1)–P(1)–C(201)
C(101)–P(1)–C(111)
C(101)–P(1)–C(201) 100.0(12)
C(111)–P(1)–C(201) 113.3(12)
C(121)–P(2)–C(131) 100.7(12)
C(121)–P(2)–C(211) 102.2(14)
C(131)–P(2)–C(211) 102.7(11)
110.8(7)
99.4(12)
Variable-temperature NMR spectra of 1 were recorded to
explore possible nonrigidity. Indeed, there are clear changes in
the 31P-{1H} NMR spectrum in CDCl3 with changes in tem-
perature (Fig. 2). The two sharp singlets for free and coordin-
ated 31P nuclei at 27 ЊC broaden considerably at Ϫ20 ЊC and
resolve at Ϫ55 ЊC into four sharp singlets at δ Ϫ10.8 and
5.4 (major isomer A) and at δ Ϫ17.9 and 8.7 (minor isomer B).
The mol ratio A:B = 1.65:1. We considered initially that these
isomers might have different sites of attachment of BINAP at
the Os3(CO)11 unit, but the IR spectrum of the isomeric mixture
in solution is closely similar to those of other tertiary phos-
phine compounds of the type [Os3(CO)11L] in which the ligand
L is known to be equatorial.8 The simplicity of the spectrum is
only consistent with equatorial coordination for both isomers
1A and 1B present in solution. Hence we conclude that the
isomers both have equatorial coordination but with different
conformational arrangements within the η1-BINAP ligand
which do not affect the ν(CO) IR spectrum. Note that the
largest chemical shift difference between the isomers is for the
free PPh2 group (Fig. 2). Rotation about the Ph2P–naphthalene
bonds leads to the lone-pair pointing inwards or outwards from
the centre of the ligand. In free BINAP in the solid both lone-
pairs point inwards in parallel directions, while one of those in
cluster 1 points inwards as in free BINAP but the coordinated
one, of necessity, points outwards. We believe therefore that
isomers 1A and 1B may have the conformations in–out and
out–out (approximate lone-pair directions) respectively and
that rotation about the Ph2P–naphthalene bond for the
uncoordinated PPh2 group is rapid at room temperature. At
Ϫ55 ЊC this rotation is slow leading to the observed spectra
(Fig. 2). Fig. 3 represents our proposed structures for isomers
1A and 1B. When the lone-pair is pointing outwards, one
phenyl group has a close parallel alignment with the distant
naphthyl group. We are attempting to synthesize [{Os3-
(CO)11}2{(R)-BINAP}] in which the out–out arrangement
when the mixture is saturated with CO gas during the reaction
or when (R)-BINAP is treated with [Os3(CO)10(MeCN)2] in
THF. The FAB mass spectra of the products in a 3-nitrobenzyl
alcohol matrix are consistent with the apparent formulae
[Os3(CO)x{(R)-BINAP}] 1 (x = 11), 2 (x = 10) and 3 (x = 8).
Spectroscopy and structure of [Os3(CO)11{(R)-BINAP}] 1
1
As expected, cluster 1 contains monodentate BINAP. The H
NMR spectrum is complex, consistent with an unsymmetrical
BINAP ligand, and the 31P-{1H} NMR spectrum shows two
moderately broad singlets at δ Ϫ14.1 and 8.2 at room tempera-
ture which sharpen at 40 ЊC. Free BINAP gives a signal at δ
Ϫ14.5 and therefore it is reasonable to suppose that the two
signals for 1 correspond to coordinated (δ 8.2) and free
(δ Ϫ14.1) phosphorus atoms. Monodentate BINAP has not
previously been observed so we have determined the single-
crystal structure of cluster 1, obtained as yellow crystals of the
dichloromethane solvate, 1ؒ0.5CH2Cl2, by evaporation of a
dichloromethane-hexane mixture. Generally we have found that
clusters containing BINAP are difficult to crystallize in a suit-
able form for single-crystal studies and the two structures
reported in this paper are for crystals that are not of ideal
quality. In addition there was the problem that 1ؒ0.5CH2Cl2
formed thin plates. The molecular structure is shown in Fig. 1
and selected bond lengths and angles are in Table 1. The struc-
ture has the monodentate BINAP ligand in an equatorial site
as we predicted from NMR data. There is no evidence in the
many crystal structures determined for clusters of the type
[M3(CO)11L] (M = Ru or Os, L = tertiary phosphine or arsine)
3820
J. Chem. Soc., Dalton Trans., 1998, 3819–3824

View Article Online
Fig. 4 Representation of cluster 1,1-[Os3(CO)10{(R)-BINAP}] 2 show-
ing the five non-equivalent CO ligands, A to E, and a mechanism lead-
ing to the exchange of carbonyl B, C and D without exchange with A
and E.
present in solution or two or more conformations are in rapid
exchange even at Ϫ100 ЊC.
Spectroscopy and structure of [Os3(CO)10{(R)-BINAP}] 2
Unfortunately we were unable to obtain suitable single crystals
for XRD to establish whether BINAP is chelating or bridging in
2. We have compared the ν(CO) IR spectra of 2 with those of
related complexes of the type [Os3(CO)10(diphosphine)] which
are known to contain either chelating or bridging diphos-
phines.10 Isomers are formed in some cases. For example 1,1-
and 1,2-isomers of [Os3(CO)10(Ph2PCH2CH2PPh2)] have been
characterized.11 We also compared IR spectra of 2 with those
of the 1,1- and 1,2-isomers of [Os3(CO)10(PMe2Ph)2].12 Unfortu-
nately the IR spectrum of 2 did not correspond unambiguously
with either spectrum of these isomeric forms, although the
match was closer for the 1,1-isomers. However, the 13C-{1H}
NMR spectra clearly indicate that BINAP is chelating and
the cluster is without doubt the 1,1-isomer. It is predicted that
there will be five different CO ligand environments in ratio
2:2:2:2:2 for both 1,1- and 1,2-isomers of 2 (Fig. 4 shows the
1,1-isomer). Five signals are indeed observed at Ϫ65 ЊC at
δ 203.0, 187.2, 186.1, 174.4 and 169.9 (Fig. 5). The observation
of a 1:2:1 triplet for the signal at δ 203.0 (JPC = 7.9 Hz) (more
clearly resolved at intermediate temperatures) can only be
reconciled with the 1,1-structure with this signal assigned to the
carbonyl ligands A (Fig. 4). For comparison, the 1,1-isomer of
[Os3(CO)10(PMe2Ph)2] gives a 1:2:1 triplet at δ 202.1 (JPC = 9.5
Hz) for the corresponding axial carbonyl ligands of the
Os(CO)2(PMe2Ph)2 group. This assignment of structure for 2 is
confirmed by the pattern of NMR coalescences at higher tem-
peratures. Three of the five CO signals (those at δ 186.1, 174.4
and 169.9) broaden on raising the temperature to Ϫ5 ЊC and are
close to coalescence at 40 ЊC. The signals at δ 203.0 and 187.2
remain sharp over this temperature range. A process involving a
merry-go-round exchange at the Os2(CO)8 unit with intermedi-
ates containing bridging CO will exchange the CO ligands B, C
and D while leaving the CO ligands A and E unaffected as
observed. Fig. 4 shows this process for cluster 2 which we
believe is the only one to fit the observed behaviour simply. We
did not measure the spectrum above 40 ЊC because of thermal
decarbonylation to give cluster 3 at higher temperatures. We
believe there is no way to interpret this NMR behaviour for the
alternative structure of 2, that is 1,2-[Os3(CO)10{(R)-BINAP}].
Fig. 2 31P-{1H} NMR spectra of [Os3(CO)11{(R)-BINAP}] 1 in
CDCl3. The coordinated PPh2 group gives the signals to the left and the
free PPh2 group those to the right. The peak marked with* corresponds
to free (R)-BINAP.
Fig. 3 Conformation of (R)-BINAP in the solid state6 (left) with
proposed conformations leading to isomers 1A and 1B for [Os3(CO)11-
{(R)-BINAP}].
Spectroscopy and structure of [Os3(ì-H)(CO)8{ì-(R)-
BINAP ؊ H}] 3
would be required for the BINAP bridge. This would then give
two parallel alignments between phenyl and naphthyl groups.
We have also examined the 31P-{1H} NMR spectra of free
BINAP over a range of temperatures to see whether similar
conformers to those in cluster 1 are present. There is no evi-
dence for this because the 31P-{1H} 600 MHz NMR spectrum
of free BINAP in a 1:1 CD2Cl2–C2Cl4F2 mixture is a sharp
singlet down to Ϫ100 ЊC. Either there is only one conformer
The cluster
3 was assigned the apparent stoichiometry
[Os3(CO)8{(R)-BINAP}] from its mass spectrum but it is not
surprising that this apparently unsaturated cluster has under-
1
gone internal oxidative addition. The H and 13C-{1H} NMR
spectra are extremely complex but the double doublet signal in
the hydride region of the 1H NMR spectrum at δ Ϫ16.86
J. Chem. Soc., Dalton Trans., 1998, 3819–3824
3821

View Article Online
Table 2 Selected bond lengths (Å) and angles (Њ) for the cluster [Os3-
(µ-H)(CO)8{(R)-BINAP Ϫ H}] 3
Os(1)–Os(2)
Os(2)–Os(3)
Os(1)–Os(3)
Os(2)–P(1)
Os(2)–P(2)
Os(1)–C(101)
Os(3)–C(101)
2.885(2)
3.061(2)
2.799(2)
2.399(9)
2.380(10)
2.35(3)
Os(1)–C(106)
P(1)–C(111)
P(1)–C(201)
P(1)–C(106)
P(2)–C(121)
P(2)–C(131)
P(2)–C(211)
2.45(3)
1.80(5)
1.84(4)
1.86(4)
1.83(4)
1.86(4)
1.87(3)
2.11(3)
P(1)–Os(2)–P(2)
Os(1)–Os(2)–P(1)
Os(3)–Os(2)–P(1)
Os(2)–P(1)–C(201) 112.4(4)
Os(2)–P(1)–C(111) 124(2)
Os(2)–P(1)–C(106) 105.0(12)
93.5(4)
74.4(30)
84.5(2)
Os(2)–P(2)–C(121)
Os(2)–P(2)–C(131)
Os(2)–P(2)–C(211)
P(1)–C(106)–C(101) 119(3)
P(1)–C(106)–C(105) 118(3)
119.0(14)
115.0(12)
107.3(11)
making the cluster a 48-electron species. Although we would
expect there to be three Os–Os bonds, the Os(2) ؒ ؒ ؒ Os(3) dis-
tance of 3.061(2) Å would normally be considered too long for
a bonding interaction.
We previously reported the cluster [Ru3(µ-H)(CO)9{µ-(R)-
BINAP Ϫ H}]6 which has significantly different spectroscopic
characteristics from those of cluster 3 but this is not surprising
since the Ru cluster has an extra CO ligand. Most likely the
η1 :η2 bridging nature of the orthometallated phenyl in 3 is
replaced by an η1 coordinated phenyl in the Ru compound
without any η2 contact. No suitable crystals for XRD work
were formed and therefore this cannot be confirmed.
Experimental
Materials and spectra
The clusters [Os3(CO)12] and [Os3(CO)10(MeCN)2] were pre-
pared by published methods.13,14 The (R)-BINAP was used as
obtained from Aldrich. Octane, heptane, and decane used were
dried over sodium wire. Solvents THF and dichloromethane
were dried and re-distilled. 300 and 400 MHz spectra were
obtained on a Bruker AC300 and Varian VXR400 spectro-
meters at UCL respectively. 600 MHz spectra were obtained on
the University of London Intercollegiate Research Service
using a Bruker AMX600 spectrometer at Queen Mary and
Westfield College. 31P NMR spectra (ref. H3PO4) were obtained
on a Bruker AMX400 spectrometer (University of London
Intercollegiate Research Service at King’s College London). IR
spectra were obtained on a Nicolet 280 FT-IR spectrometer,
and FAB MS (3-nitrobenzyl alcohol matrix) on a ZAB mass
spectrometer.
Fig. 5 13C-{1H} NMR spectra of 1,1-[Os3(CO)10{(R)-BINAP}] 2
showing assignment of signals with reference to Fig. 4, the coalescence
of signals B, C and D, and the triplet pattern for signal A.
(JPH = 8.4 and 17.3 Hz) is consistent with an orthometallation
product. An examination of the aromatic region seems to indi-
cate metallation at a phenyl group. The 31P-{1H} NMR spec-
trum (1H-decoupled in the aromatic but not the hydride region)
shows a double doublet for each of the non-equivalent phos-
phorus nuclei as expected.
A single-crystal structure determination established the
arrangement of the orthometallated BINAP ligand at the tri-
osmium unit. Generally poor crystals were obtained from a
methanol–dichloromethane solution but an adequate crystal
was found for study. The molecular structure is shown in Fig. 6
and selected bond lengths and angles are in Table 2. Unlike the
precursor cluster 2 which has 1,1-diequatorial BINAP coordin-
ation, cluster 3 has 1,1-axial-equatorial coordination of the
BINAP at Os(2) as well as a bridging coordination of the
orthometallated phenyl group at Os(1) and Os(3). Fig. 6 shows
that the mode of µ3-bridging in this ligand system requires the
BINAP to have axial–equatorial coordination. Thus Os(2) is
bonded to P(1) and P(2) and orthometallation has occurred at
Os(3) which carries the σ-Os(3)–C(101) bond, while Os(1) has
an η2-contact to C(101) and C(106) of the orthometallated
phenyl ring. The geometry of the η1 :η2-phenyl is unexceptional
with the σ-Os(3)–C(101) bond being significantly shorter than
the Os–C distances in the η2-contact. The hydride ligand was
not located but CO ligand positions show that it is bridging
Os(2) and Os(3), which is the longest Os–Os contact [3.061(2)
Å] compared with Os(1)–Os(2) [2.885(2) Å] and Os(1)–Os(3)
[2.799(2) Å]. The hydride must lie close to the intersection of
the C(22)–Os(2) and C(33)–Os(3) vectors, slightly above the Os3
plane. The orthometallated ligand is a seven-electron donor
Direct thermal reaction of [Os3(CO)12] with BINAP in heptane
A solution of BINAP (0.040 g) and [Os3(CO)12] (0.060 g) in
heptane (25 cm3) was refluxed under nitrogen for 3.5 h. The
solution became darker after 1.5 h and faint orange after 3.5 h.
The solution was cooled to 5 ЊC for one hour to allow the excess
of [Os3(CO)12] to crystallize. After filtration through a thin layer
of silica, the solvent was removed from the filtrate under
reduced pressure. The residual solid was separated by prepar-
ative TLC (2 mm, Merck 1045), eluting with a dichloro-
methane–hexane mixture (3:7 by volume). Three bands yielded
yellow 1 (10%), orange 2 (15%) and yellow 3 (14%) respectively.
Cluster 1: IR (cyclohexane), ν(CO)/cmϪ1: 2066vs, 2030vs, 2009s,
1992vw, 1976vs, 1961s, 1942w. 1H NMR (CDCl3, 20 ЊC): δ 6.15
(t), 6.45 (d), 6.5 to 7.8 (overlapping multiplets), 8.40 (dt). 31P-
{1H} NMR (CDCl3, 20 ЊC): δ Ϫ14.1 (s) and 8.2 (s) at 40 ЊC and
δ 5.4 (s), Ϫ10.8 (s) (isomer 1A), 8.7 (s), Ϫ17.9 (s) (isomer 1B) at
Ϫ55 ЊC. FAB MS: observed centre of isotopic envelope 1504,
calculated for parent molecular ion 1500. Cluster 2: IR (cyclo-
hexane), ν(CO)/cmϪ1: 2095m, 2049s, 2015s, 2010vs, 1999m,
1988m, 1971w, 1904w. 1H NMR (CDCl3, 20 ЊC ): δ 6.40 (d), 6.85
3822
J. Chem. Soc., Dalton Trans., 1998, 3819–3824

View Article Online
Fig. 6 Molecular structure of [Os3(µ-H)(CO)8{(R)-BINAP Ϫ H}] 3.
Table 3 Crystal data and structure solution and refinement para-
meters for clusters 1ؒ0.5CH2Cl2 and 3a
(d), 6.92 (td), 7.0 to 7.3 (overlapping m), 7.35 (td), 7.45 (dd),
7.84 (d), 7.89 (d). 13C-{1H} NMR (100 MHz, CDCl3, 20 ЊC ):
δ 203.0, 187.2, 186.1, 174.4, 169.9 (CO-signals), 141 to 125 (aryl
signals, many overlapping). 31P-{1H} NMR (CDCl3, 20 ЊC)
δ Ϫ5.8 (s). FAB MS: observed centre of isotopic envelope 1473,
calculated for parent molecular ion 1472. Cluster 3: IR (cyclo-
hexane) ν(CO)/cmϪ1: 2108m, 2055s, 2037s, 2020vs, 2000m,
1992s, 1979s, 1963w, 1958w. 1H NMR (CDCl3, 20 ЊC): δ 8.46 (d),
7.7 to 7.2 (overlapping m), 7.10 (dt), 6.93 (m), 6.8 to 6.6 (over-
lapping m), 6.40 (d), 6.19 (dd), Ϫ16.86 (dd, JPH = 8.4, 17.3 Hz).
13C-{1H} NMR (CDCl3, 20 ЊC, ): δ 184.7, 183.8, 178.8, 178.1
(CO-signals), 151 to 122 (complex aryl signals). 31P-{1H} NMR
(CDCl3, 20 ЊC): δ 4.18 and 2.13. FAB MS: observed centre of
isotopic envelope 1417, calculated for parent molecular ion
1417 (Found: C, 45.35; H, 2.7; P, 4.0. C52H32O8Os3P2 requires C,
44.1; H, 2.3; P, 4.35%). A similar reaction in refluxing octane
(1.5 h) gave 1 (12%), 2 (24%) and 3 (36%) and a reaction in
refluxing decane (45 min) with CO bubbling through the
reaction mixture. Thermolysis of [Os3(CO)11(BINAP)] 1 in
toluene-d8 in a 10 mm NMR tube at 105 ЊC over 5.5 hours gave
conversion to clusters 2 and 3.
1ؒ0.5CH2Cl2
3
Formula
M
Crystal system
Space group
a/Å
b/Å
c/Å
C55.5H33ClO11Os3P2
1543.81
Monoclinic
P21
15.610(4)
10.957(2)
16.186(4)
108.97(2)
2618.0(11)
2
74.34
ω-2θ
5–50
0,0,Ϫ18 to 18,13,18
4958
4773
571
0.0783
C52H32O8Os3P2
1417.32
Orthorhombic
P212121
10.661(4)
21.26(2)
24.21(2)
90
5489(6)
4
70.32
ω
5–50
β/Њ
U/Å3
Z
µ(Mo-Kα)/cmϪ1
Scan mode
2θ range/Њ
hkl range
0,0,0 to 12,25,28
5266
5266
Total data
Unique data
Parameters in refinement
R (all data)b
[I > 2σ(I)]
wR2 (all data)c
[I > 2σ(I)]
Goodness of fit
Abs. struct. parameter
321
0.1131
0.0891
0.273
0.241
1.075
0.0595
0.161
0.136
1.068
Reaction of [Os3(CO)10(MeCN)2] with BINAP
Ϫ0.02(3)
0.02(5)
The cluster [Os3(CO)10(MeCN)2] (0.10 g) and an equimolar
quantity of (R)-BINAP were refluxed in dry THF (25 cm3) for
30 min. The products isolated by TLC on silica were 2 (50%)
and minor quantities of 1 and 3 which were only characterized
by their IR spectra. Cluster 1 is probably formed from traces of
[Os3(CO)11(MeCN)] in the osmium starting material.
a Both structures: direct methods structure solution, graphite-
monochromated Mo-Kα radiation (λ = 0.71073 Å), three standard
reflections every 97, no decay, data corrected for absorption empirically
by ψ-scan method, maximum and minimum transmission 0.986 and
0.501 for 1 and 0.930 and 0.728 for 3, full-matrix least-squares refine-
¹
4
ment of F2. b R = Σ|Fo| Ϫ |Fc|/Σ|Fo|. c wR2 = [Σw(Fo2 Ϫ Fc2)2/ΣwFo ]²
where w = 1/[σ2(Fo)2 ϩ (0.0578p)2 ϩ 54.72p] for 1 and w = 1/[σ2(Fo)2 ϩ
(0.0866p)2 ϩ 478.5p] for 3, where p = (Fo2 ϩ 2Fc2)/3.
Crystallography
Yellow crystals of 1ؒ0.5CH2Cl2 and 3 were grown from hexane–
dichloromethane mixtures. X-Ray data were collected at
room temperature using a Nicolet R3v/m diffractometer with
crystals mounted in air. Details of the crystal structure
determinations are given in Table 3. Structures were solved by
direct methods using SHELXTL PLUS15 and refined using
SHELXL 93.16 All non-H atoms of 1ؒ0.5CH2Cl2 except for
CH2Cl2 were refined anisotropically and the dichloromethane
C and Cl atoms were filled into final cycles of refinement.
Only the Os, P and O atoms of 3 were refined anisotrop-
ically. H-atoms in both crystals were included in idealised
positions, riding upon their respective carbon atoms, with C–H
distance fixed at 0.96 Å and isotropic thermal parameters at
0.08 Å3.
CCDC reference number 186/1160.
J. Chem. Soc., Dalton Trans., 1998, 3819–3824
3823

View Article Online
J. Chem. Soc., Chem. Commun., 1989, 1208; J. B. Hoke, L. S. Hollis
and E. W. Stern, J. Organomet. Chem., 1993, 455, 193; D. D. Patak,
H. Adams, N. A. Bailey, P. J. King and C. White, J. Organomet.
Chem., 1994, 479, 237; K. Tani, T. Yamagata, Y. Tatsumo,
Y. Yamagata, K. Tomita, S. Akutagawa, H. Kumobayashi and
S. Otsuka, Angew. Chem., Int. Ed. Engl., 1985, 24, 217; H. Taninaga,
K. Sakai and T. Tsubomura, J. Chem. Soc., Chem. Commun., 1995,
2273; H. Kawano, T. Ikariya, Y. Ishii, T. Kodama, M. Saburai,
S. Yoshikawa, Y. Uchida and S. Akutagawa, Bull. Chem. Soc. Jpn.,
1992, 65, 1595; H. Kawano, Y. Ishii, T. Kodama, M. Suburai and
Y. Uchida, Chem. Lett., 1987, 1311; K. Mashima, T. Hino and
H. Takaya, J. Chem. Soc., Dalton Trans., 1992, 2099.
See http://www.rsc.org/suppdata/dt/1998/3819/ for crystallo-
graphic files in .cif format.
Acknowledgements
We thank the EPSRC for the provision of the X-ray diffract-
ometer and the University of London Central Research Fund
for support.
References
5 T. Yamagata, K. Tani, Y. Tatsuno and T. Saito, J. Chem. Soc., Chem.
Commun., 1988, 466.
6 A. J. Deeming, D. M. Speel and M. Stchedroff, Organometallics,
1997, 16, 6004.
7 A. J. Deeming, M. Stchedroff and J. W. Steed, unpublished work.
8 A. J. Deeming, Comprehensive Organometallic Chemistry II, ed.
E. W. Abel, F. G. A. Stone and G. Wilkinson, Pergamon, Elsevier
Science Ltd, Oxford, 1995, vol. 7, p. 683.
9 P. A. Dawson, B. F. G. Johnson, J. Lewis, J. Puga, P. R. Raithby and
M. J. Rosales, J. Chem. Soc., Dalton Trans., 1982, 233.
10 A. J. Deeming, S. Donovan-Mtunzi and S. E. Kabir, J. Organomet.
Chem., 1984, 276, C65.
11 A. J. Deeming, S. Donovan-Mtunzi, K. I. Hardcastle, S. E. Kabir,
K. Henrick and M. McPartlin, J. Chem. Soc., Dalton Trans., 1988,
579.
12 A. J. Deeming, S. Donovan-Mtunzi, S. E. Kabir and P. J. Manning,
J. Chem. Soc., Dalton Trans., 1985, 1037.
13 B. F. G. Johnson and J. Lewis, Inorg. Synth., 1972, 13, 93.
14 J. N. Nicholls and M. D. Vargas, Inorg. Synth., 1989, 26, 289.
15 G. M. Sheldrick, SHELXTL PLUS, University of Göttingen,
released by Nicolet Instruments Corporation, 1987.
16 G. M. Sheldrick, SHELXL 93, Program for refinement of crystal
structures, University of Göttingen, 1993.
1 R. Noyori, Asymmetric Catalysis in Organic Synthesis, Wiley,
New York, 1994; R. Noyori, Acta Chem. Scand., 1996, 50, 380;
R. Noyori, in (IUPAC) Stereocontrolled Organic Synthesis, ed.
B. M. Trost, Blackwell, Oxford, 1994, p. 1; G. Procter, Asymmetric
Synthesis, Oxford University Press, 1996; R. Noyari and H. Takaya,
Acc. Chem. Res., 1990, 23, 345.
2 A. Miyashita, A. Yasuda, H. Takaya, K. Toriumi, T. Ito, T. Souchi
and R. Noyori, J. Am. Chem. Soc., 1980, 102, 7932; H. Takaya,
S. Akutagawa and R. Noyori, Org. Synth., 1988, 67, 20.
3 M. Kitamura, M. Tokunaga and R. Noyori, J. Org. Chem., 1992, 57,
4053; H. Takaya, T. Ohta, S. Inoue, M. Tokunaga, M. Kitamura and
R. Noyari, Org. Synth., 1993, 72, 74; X. Wang and B. Bosnich,
Organometallics, 1994, 13, 4131.
4 K. Toriumi, T. Ito, H. Takaya, T. Souchi and R. Noyari, Acta
Crystallogr., Sect. B, 1982, 38, 807; F. Ozawa, A. Kubo,
Y. Matsumoto, T. Hayashi, E. Nishioka, K. Yanagi and
K. Moriguchi, Organometallics, 1993, 12, 4188; N. W. Alcock, J. M.
Brown and J. J. Perez-Torrente, Tetrahedron Lett., 1992, 33, 389;
J. M. Brown, J. J. Perez-Torrente and N. W. Alcock, Organo-
metallics, 1995, 14, 1195; P. S. Pregosin, H. Ruegger, R. Salzmann,
A. Albinati , F. Lianzi and R. W. Kunz, Organometallics, 1994, 13,
83; P. S. Pregosin, H. Ruegger, R. Salzmann, A. Albinati, F. Lianzi
and R. W. Kunz, Organometallics, 1994, 13, 5040; T. Ohta,
H. Takaya and R. Noyori, Inorg. Chem., 1988, 27, 566; M. T. Ashby,
M. A. Khan and J. Halpern, Organometallics, 1991, 10, 2011;
K. Mashima, K. Kusano, T. Ohta, R. Noyori and H. Takaya,
Paper 8/06430J
3824
J. Chem. Soc., Dalton Trans., 1998, 3819–3824