Ruthenium xantphos complexes in hydrogen transfer processes: Reactivity and mechanistic studies

Article abstract of DOI:10.1039/b813543f

The in situ combination of [Ru(PPh₃)₃(CO)H ₂] with xantphos is catalytically active for the alkylation of alcohols with the ketonitrile ^tBuC(O)CH₂CN in a model oxidation-Knoevenagel-reduction process. The precursor complex [Ru(xantphos)(PPh₃)(CO)H₂] was isolated and reacted with stoichiometric amounts of PhCH₂OH and PhCHO. Under these conditions, the alcohol is decarbonylated to afford [Ru(xantphos)(CO)₂H ₂] and finally [Ru(xantphos)(CO)₃], both of which prove to be less active for catalysis than the starting complex. The reactivity of the xantphos system contrasts with that of [Ru(dppp)(PPh₃)(CO)H ₂], which is catalytically inactive for the Knoevenagel reaction and fails to show any stoichiometric reactivity with alcohols. The Royal Society of Chemistry 2009.

Full text of DOI:10.1039/b813543f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Ruthenium xantphos complexes in hydrogen transfer processes: reactivity and
mechanistic studies†
Araminta E. W. Ledger, Paul A. Slatford, John P. Lowe, Mary F. Mahon, Michael K. Whittlesey* and
Jonathan M. J. Williams*
Received 5th August 2008, Accepted 17th September 2008
First published as an Advance Article on the web 27th November 2008
DOI: 10.1039/b813543f
The in situ combination of [Ru(PPh3)3(CO)H2] with xantphos is catalytically active for the alkylation of
alcohols with the ketonitrile tBuC(O)CH2CN in a model oxidation–Knoevenagel–reduction process.
The precursor complex [Ru(xantphos)(PPh3)(CO)H2] was isolated and reacted with stoichiometric
amounts of PhCH2OH and PhCHO. Under these conditions, the alcohol is decarbonylated to afford
[Ru(xantphos)(CO)2H2] and finally [Ru(xantphos)(CO)3], both of which prove to be less active for
catalysis than the starting complex. The reactivity of the xantphos system contrasts with that of
[Ru(dppp)(PPh3)(CO)H2], which is catalytically inactive for the Knoevenagel reaction and fails to show
any stoichiometric reactivity with alcohols.
complex8 for the formation of C–C bonds from alcohols by indirect
Introduction
Wittig reactions, which proceed via the aldehyde and alkene.
Ruthenium complexes have been widely used as catalysts for
Recently, we reported that the combination of [Ru(PPh3)3-
hydrogen transfer reactions, especially in the reduction of ketones
to secondary alcohols.1 The metal-catalysed transfer of hydrogen
(CO)H2] with bidentate phosphines can afford an effective catalyst
for the alkylation of the ketonitrile 6 with alcohols (Scheme 2).
from alcohols to alkenes has been of significant importance in the
Particularly striking was the high activity found when the added
ligand was xantphos 5.9 The [Ru(PPh3)3(CO)H2]/xantphos com-
formation of new C–C bonds by the borrowing hydrogen strategy
outlined in Scheme 1. A transition metal catalyst temporarily
bination has subsequently proved to be highly effective for other
processes including the conversion of alkynediols into furans10 and
pyrroles,11 as well as the elimination of alcohols from oxime ethers
to give nitriles.12
removes hydrogen from an alcohol 1 to generate the more
electrophilic aldehyde 2. This intermediate aldehyde undergoes
in situ transformation into an alkene 3, which is then reduced to
generate the new C–C bond in product 4 by return of the hydrogen
from the catalyst. The borrowing hydrogen strategy avoids the
requirement for traditional alkylating agents, such as alkyl halides.
Scheme 1 Borrowing hydrogen in the formation of C–C bonds from
alcohols.
Scheme 2 Catalytic C–C bond formation with [Ru(PPh3)3(CO)H2] and
xantphos.
The first examples of homogeneous catalysts for such chemistry
were reported by Grigg et al. in 1981.2 There have been many
other catalysts reported since,3 especially ruthenium,4 iridium5 and
palladium6 based systems. We have previously reported the use
of an iridium complex7 and an N-heterocyclic carbene ruthenium
In light of the high activity of the in situ generated ruthe-
nium xantphos system, we have now prepared and isolated the
precursor complex [Ru(xantphos)(PPh3)(CO)H2] (18) and studied
its stoichiometric reactivity towards alcohols and aldehydes. The
reactivity of 18 is contrasted with [Ru(dppp)(PPh3)(CO)H2] (19,
dppp = Ph2P(CH2)3PPh2), which proves to be ineffective for
the catalytic reaction in Scheme 2. We now show that whereas
18 oxidises benzyl alcohol, 19 is totally inert. Moreover, under
stoichiometric conditions, decarbonylation of benzyl alcohol by
18 also takes place to afford Ru(II) and Ru(0) species that prove to
be less catalytically active than the starting material. These results
Department of Chemistry, University of Bath, Claverton Down, Bath, UK
BA2 7AY. E-mail: chsmkw@bath.ac.uk, chsjmjw@bath.ac.uk; Fax: +44
1225 386 231; Tel: +44 1225 383 942
† Electronic supplementary information (ESI) available: spectroscopic
data for compounds 7b–k are provided, along with crystallographic data
for complex 22. CCDC reference numbers 678477 (18), 678478 (20) and
678479 (22). For ESI and crystallographic data in CIF or other electronic
format see DOI: 10.1039/b813543f
716 | Dalton Trans., 2009, 716–722
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 1 Activity of [Ru(PPh3)3(CO)H2]–phosphine combinations for
reaction of 1 with 6a
Table 2 Range of alcohols examined in the alkylation of ketonitrile 6
with [Ru(PPh3)3(CO)H2]–xantphosa
Ligandb
Catalyst loading/mol%
Time/h
Conversion (%)
R
Catalyst loading (mol%)b Conversion (%) Yield (%)
None
Dppe
Dppp
Dppf 9
Dppf 9
10
Xantphos 5
11
DPEphos 8
5
5
5
5
0.5
0.5
0.5
0.5
0.5
18
1
1
1
3
3
3
3
3
56
4
Ph (1a)
p-MeOC6H4 (1b) 0.5
o-MeOC6H4 (1c) 0.5
p-FC6H4 (1d)
p-BrC6H4 (1e)
p-O2NC6H4 (1f) 0.5
p-F3CC6H4 (1g)
Furyl (1h)
PhCH2 (1i)
C10H21 (1j)
0.5
100
100
100
100
100
52
100
100
100
100
100
100
78
83
82
89
79
31
83
72
87
85
69
76
<1
74
56
91
100
0
0.5
0.5
0.5
5
2.5
5
5
5
8
a Reaction conditions: benzyl alcohol (1a) 1 equiv., ketonitrile (6) 1 equiv.,
piperidinium acetate 5 mol%, toluene, 120 ◦C, 3 h. b A 1 : 1 ratio of
[Ru(PPh3)3(CO)H2] to ligand was used in all cases.
Cyclopropyl (1k)
IndenylCH2 (1l)
a Reaction conditions: alcohol 1 equiv., ketonitrile (6) 1 equiv., piperi-
dinium acetate 5 mol%, toluene, reflux, 4 h. b A 1 : 1 ratio of
[Ru(PPh3)3(CO)H2] to xantphos was used in all cases.
suggest that the success of 18 in catalysing the reaction shown in
Scheme 2 is not only due to the ability of the catalyst precursor
to react with the alcohol in the first place, but also because of the
fact that 6 is able to react with the intermediate aldehyde before
it is decarbonylated by ruthenium to form less active metal based
species.
a range of alcohols. With benzylic alcohols (1a–e, g, Table 2),
a catalyst loading of 0.5 mol% was sufficient to achieve 100%
conversion in 3 h with correspondingly good isolated yields
with the exception of the p-nitro-substituted benzyl alcohol (1f)
and furfuryl alcohol (1h), although in the latter case, complete
conversion was obtained with a higher catalyst loading. As
complete conversions were seen in many cases after 3 h, the
reactions with 1a, 1b and 1d were run for shorter times (1.5 h,
not shown in Table 2) to establish whether there was an electronic
influence on reaction rate. The results for benzyl alcohol 1a (88%),
p-methoxybenzyl alcohol 1b (95%) and p-fluorobenzyl alcohol 1d
(70%) suggest that the relative ease of oxidation of the alcohols
to the intermediate aldehydes is a limiting factor in the overall
process. Further support for this argument was provided by the
lower reactivity associated with the aliphatic alcohols 1i–l.
The success of xantphos prompted us to investigate other
xanthene based ligands 12–17 (Fig. 1) for the reaction shown
in Scheme 3. Two of the ligands, sixantphos 13 and isopropy-
lxantphos 15 were found to give a somewhat improved reactivity
compared to the parent xantphos (Table 3), although the com-
mercial availability of 5 was the reason that it was used for all
subsequent studies of stoichiometric reactivity. One conclusion
clearly supported by the data in Table 3 is that in contrast to other
catalysis employing xanthene based ligands, there is no obvious
correlation between the reactivity and the reported natural bite
angles of the ligands.16
Results and discussion
Catalytic Activity of [Ru(PPh3)3(CO)H2]–phosphine
As we have previously reported, the alkylation of ketonitrile 6 with
benzyl alcohol 1a (Scheme 3) was catalysed by an in situ mixture of
[Ru(PPh3)3(CO)H2] and a range of bidentate phosphorus ligands
as shown in Table 1. The parent complex [Ru(PPh3)3(CO)H2]
showed some catalytic activity in the absence of any added ligands
(although over quite a long reaction time), however, the addition
of one equivalent of either of the ferrocene based ligands 9 or 10,13
or xantphos 5,14 provided a much more reactive system that gave
good conversions to 7a at only 0.5 mol% catalyst loading. The
subtle nature of the stereoelectronic effects at play in the catalysis
was well illustrated by the fact that the bulky ferrocene derivative
11 as well as the flexible P,O,P ligand DPEphos 815 proved to be
completely inactive. Of equal note was the failure of the more
common chelating phosphines dppe and dppp to give any product
formation.
Table 3 Activity of added xanthene ligands for the reaction in Scheme 3a
Ligandb
Bite angle/◦
Conversion (%)
c
Homoxantphos 12
Sixantphos 13
Thixantphos 14
Xantphos 5
Isopropylxantphos 15
Nixantphos 16
Benzylnixantphos 17
102.0
108.5
109.6
111.4
113.2
114.2
114.1
46
98
77
81
92
3
4
a Reaction conditions: alcohol
1 equiv., ketonitrile (6) 1 equiv.,
Scheme 3 Ligand screening in the benzylation of ketonitrile 6.
[Ru(PPh3)3(CO)H2] 0.5 mol%, piperidinium acetate 5 mol%, toluene,
110 ◦C, 3.5 h. b A 1 : 1 ratio of [Ru(PPh3)3(CO)H2] to xanthene ligand
was used in all cases. c See ref. 17.
Subsequent studies showed that the [Ru(PPh3)3(CO)H2]–
xantphos combination could be applied to the coupling of 6 with
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 716–722 | 717
©

View Article Online
Fig. 1 Xanthene ligands tested in the benzylation of ketonitrile 6.
Synthesis and characterisation of [Ru(PPh3)3(CO)H2]–phosphine
precursors
Fig. 3 Molecular structure of 18.23 All hydrogen atoms except Ru–H
are omitted. Thermal ellipsoids are shown at the 30% probability level.
◦
˚
In order to understand the activity associated with
[Ru(PPh3)3(CO)H2] and xantphos, and conversely the low
activity found upon using dppp, both of the precursor complexes
[Ru(xantphos)(PPh3)(CO)H2] 18 and [Ru(dppp)(PPh3)(CO)H2]
1918 were prepared and isolated. In contrast to the sharp room
Selected bond lengths (A) and angles ( ): Ru–P(1) 2.3283(6); Ru–P(2)
2.3950(6); Ru–P(3) 2.2973(6); Ru–C(1) 1.897(3); P(1)–Ru–P(2) 102.76(2);
P(1)–Ru–P(3) 145.06(2); P(2)–Ru–P(3) 107.16(2).
of xantphos and dppp, the xantphos ligand enforces a wide
P(1)–Ru–P(2) angle [102.76(2)◦] at ruthenium, distorting the
octahedral geometry to a greater extent than in 19 [P(1)–Ru–P(2),
92.14(9)◦].18b This distortion is further manifested in the trans-P–
Ru–P angles 145.06(2) and 160.67(9)◦ for 18 and 19, respectively.
1
temperature 31P{ H} NMR spectrum seen for 19, 18 displayed
two broad resonances for the xantphos ligand at d 45.2 and
30.5 ppm and a sharp doublet of doublets at d 58.5 ppm [JPP
=
237.1 Hz, 15.5 Hz] for the PPh3 group.19 Upon cooling to 199 K,
the signal at d 45.2 ppm resolved into a doublet of doublets with
JPP values of 237.1 and 15.5 Hz, the larger splitting arising from
coupling to the trans PPh3 ligand. The two hydride resonances
in the proton spectrum were also somewhat broad at 298 K
(Fig. 2), and although they sharpened to some extent upon
cooling to 199 K, this was not enough to resolve JHP and JHH
values accurately. The coupling constants were determined from
the simulated spectrum, which is shown with the experimental
data in Fig. 2. While we were unable to elucidate the exact reason
for the broadening of the resonances of 18,20 there are a number
of reports that support the flexibility of xantphos.21
Stoichiometric reactivity of benzyl alcohol with 18 and 19
Quite different behaviour was observed upon heating 18 and 19 in
the presence of 10 equivalents of PhCH2OH in toluene at 120 ◦C
1
for 2 h. While in the case of 19, H NMR spectroscopy showed
only unreacted starting material, the xantphos complex 18 was
converted into a ca. 2 : 1 mixture of [Ru(xantphos)(CO)2H2] 20 and
The molecular structure of 18 was determined by X-ray
crystallography as shown in Fig. 3. The chelating xantphos ligand
coordinates to the ruthenium centre in axial and equatorial
positions, with the PPh3 ligand occupying the second axial
position. As expected on the basis of the different bite angles
Scheme 4 Reactivity of [Ru(xantphos)(PPh3)(CO)H2] (18) with benzyl
alcohol.
1
Fig. 2 Simulated22 (lower) and experimental (upper) H NMR spectrum of 18 ([d8]-toluene, 500 MHz, 298 K). Calculated chemical shifts (ppm) and
coupling constants (J): d -6.67 (JHP = 35.0, JHP = 28.0, JHP = 15.5, JHH = 6.6 Hz), d -8.72 (JHP = 77.0, JHP = 33.9 Hz, JHP = 27.3, JHH = 6.6 Hz).
718 | Dalton Trans., 2009, 716–722
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
[Ru(xantphos)(PPh3)(CO)2] 21 (Scheme 4).24 PhCHO, benzene
and H2 were also detected. The dicarbonyl dihydride complex 20
displayed two doublet of doublet of doublet hydride resonances at
d -6.13 and -7.65 ppm; the large JHP coupling constant of 82 Hz
observed on the lower frequency signal showed it to be the one
trans to phosphorus.25 In the IR spectrum, two CO bands of equal
intensity were found at 2007 and 1960 cm-1, consistent with a cis
arrangement of the carbonyl ligands observed for other reported
[Ru(P–P)(CO)2H2] complexes.26 The molecular structure of 20 was
elucidated following an independent synthesis, which involved the
photolysis of [Ru(xantphos)(CO)3] 2227 under a purge of hydrogen
at 0 ◦C. The X-ray structure of 20 (Fig. 4) showed a lengthening of
the Ru–xantphos and Ru–CO distances compared to those found
Table 4 Catalytic activity of xantphos complexes 20–22 for reaction of
1a with 6a
Loading
(mol%)
Catalystb
Time/h Conversion (%)
[Ru(xantphos)(CO)2H2] 20
0.5
0.5
0.5
0.5
55
4
30
[Ru(xantphos)(PPh3)(CO)2] 21 0.5
[Ru(xantphos)(CO)3] 22
0.5
a Reaction conditions: alcohol 1 equiv., ketonitrile (6) 1 equiv., piperi-
dinium acetate 5 mol%, refluxing toluene, 3 h. b Neither of the dppp species
given in ref. 28 showed any catalytic activity.
the formation of 22 by removal of CO, addition of PPh3 or addition
of PPh3/H2 proved unsuccessful.
˚
˚
in 18 [20: Ru–P 2.3805(9); Ru–P 2.4146(8) A; Ru–CO, 1.958(5) A;
˚
˚
18: Ru–P 2.3283(6); Ru–P 2.3950(6) A; Ru-CO 1.897(3) A] and a
widening of the trans-P–Ru–L angle from 145.06(2)◦ for L = PPh3
in 18 to 156.18(10)◦ for L = CO in 20.
The ruthenium complexes 20–22 were tested for their activity
in the benzylation of ketonitrile 6, but all three were found to be
less active than 18 (Table 4). Thus, although none of these three
ruthenium complexes can therefore be associated with the overall
catalytic Knoevenagel reaction, their formation reveals a relatively
facile deactivation pathway that is available to 18, at least under
stoichiometric conditions.
Reactivity of 18 and 19 with benzaldehyde
The observation of free benzaldehyde and benzene upon reaction
of 18 with benzyl alcohol prompted us to probe the reaction
with benzaldehyde itself. In the presence of 5 equivalents of
PhCHO (toluene, 120 ◦C, 1 h), 18 yielded a mixture of products.
Both PhCH2OH and benzene were observed in the proton NMR
1
spectrum, while a 31P{ H} spectrum showed a small amount of
the ruthenium(0) species 21, a significant amount of 22, and small
amounts of a new ruthenium containing complex 23 (Scheme 5).
1
Complex 20 was not observed. H and 13C NMR spectroscopy
revealed that 23 contained a single hydride (d -6.14 ppm, dd,
2JH–P = 121.3 Hz, 2JH–P = 25.8 Hz), one xantphos ligand and two
carbonyl groups. The identity of the remaining sixth ligand could
not be elucidated; an acyl group was excluded as no high frequency
13C resonance was detected, even when 13C labelling experiments
were employed.30 Efforts to isolate 23 from the reaction mixture
or prepare it independently by either thermolysis or photolysis
of 22 in the presence of PhCHO proved unsuccessful. Given the
presence of the benzene in the reaction mixture, it seems plausible
to suggest that 23 may correspond to the phenyl hydride complex
[Ru(xantphos)(CO)2(Ph)H] (Scheme 5, X = Ph), which would be
Fig. 4 Molecular structure of 20.29 All hydrogen atoms (except Ru–H)
are omitted. Thermal ellipsoids are shown at the 30% probability level.
◦
˚
Selected bond lengths (A) and angles ( ): Ru–P(1) 2.4146(8); Ru–P(2)
2.3805(9); Ru–C(1) 1.958(5); Ru–C(2) 1.878(4); P(1)–Ru–P(2) 102.20(3);
P(1)–Ru–C(1) 91.89(10); P(1)–Ru–C(2) 92.86(10); P(2)–Ru–C(1)
95.98(10); P(2)–Ru–C(2) 156.18(10); C(1)–Ru–C(2) 101.92(15).
While the formation of benzaldehyde from PhCH2OH is con-
sistent with the ability of 18 to catalyse the Knoevenagel reaction
in Scheme 2, the formation of 20, benzene and H2 shows that
the ruthenium complex can also bring about the decarbonylation
of alcohols under non-catalytic conditions. Indeed, when 18 was
heated with Ph13CH2OH, both [Ru(xantphos)(13CO)(CO)H2] and
[Ru(xantphos)(13CO)2H2] were formed (some 13CO labelled 18
and Ph13CHO was also detected). Complex 20 was also capable
of decarbonylation chemistry; reaction of the complex with
5 equivalents of PhCH2OH gave the tricarbonyl species 22 as
the final ruthenium containing product of the reaction. While
further studies showed that 18 would decarbonylate a range of
aromatic alcohols, including Ph(CH2)nOH (n = 1, 2, 3) and
1,2-C6H4(CH2OH)2, efforts to make the process catalytic were
prevented by ease with which 22 was formed and its lack of
reactivity towards either alcohols or aldehydes. Attempts to inhibit
Scheme 5 Ruthenium products arising from reaction of [Ru(xantphos)-
(PPh3)(CO)H2] (18) with benzaldehyde.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 716–722 | 719
©

View Article Online
complex 21 implies that there must also be a pathway involving
the reversible dechelation of the xantphos ligand (this is needed
to get to Ru(0) and retain the PPh3 ligand); it may be that the
dechelation process (which is assumed to be rapid)20 accounts
for the broadened NMR signals seen for 18. The formation of
the other ruthenium(0) complex 22 as the ultimate product of
decarbonylation also requires the xantphos ligand to dechelate.
Conclusions
Xanthene based phosphine ligands have been shown to afford
ruthenium dihydride complexes, which are catalytically active for
the alkylation of alcohols by the ketonitrile 6. The high reactivity of
the xantphos complex [Ru(xantphos)(PPh3)(CO)H2] 18 contrasts
with that of the dppp analogue 19, which proves to be totally un-
reactive. While we cannot say precisely why the two ligands impart
such different properties, that they do so is revealed very simply
in the differing stoichiometric behaviour with PhCH2OH. In
the absence of 6, 18 decarbonylates benzyl alcohol to afford
[Ru(xantphos)(CO)2H2] and finally [Ru(xantphos)(CO)3]. While
the decarbonylation chemistry is not involved in the pathway
for the Knoevenagel reaction, it provides us with a ‘marker’ for
catalytic activity.
Experimental
General considerations
Scheme 6 Proposed reaction pathways of the xantphos complex 18 for
catalytic Knoevenagel chemistry and stoichiometric decarbonylation.
Manipulations were carried out using standard Schlenk or
glovebox techniques under argon. Solvents were purified using
either an Innovative Technologies solvent system (THF, hexane)
or distilled under nitrogen from sodium benzophenone ketyl
(toluene, benzene) or Mg–I2 (EtOH). NMR spectra were recorded
on Bruker Avance 300, 400 and 500 MHz NMR spectrometers,
formed following decarbonylation of the presumed acyl precursor
[Ru(xantphos)(CO)(COPh)H].
While the dppp complex 19 also proved to be reactive towards 5
1
equivalents of PhCHO, a 31P{ H} NMR spectrum of the reaction
mixture revealed the formation of several low intensity Ru–
phosphine containing products, along with a large amount of
1
and referenced as follows: CDCl3 (1H, d 7.27 ppm; 13C{ H} d
1
1
1
77.2 ppm), toluene (1H, d 2.09 ppm; 13C{ H}, d 21.3 ppm). 31P{ H}
NMR chemical shifts were referenced externally to 85% H3PO4
(d 0.0 ppm). IR spectra were recorded on a Nicolet Nexus FTIR
spectrometer. Mass spectrometry of organic products was per-
formed by the EPSRC National Mass Spectrometry Service Cen-
tre, Swansea, UK, while mass spectra of the ruthenium complexes
were recorded in the Department of Chemistry, University of Bath
using a micrOTOF electrospray time-of-flight (ESI-TOF) mass
spectrometer (Bruker Daltonik GmbH) coupled to an Agilent
1200 LC system (Agilent Technologies). Elemental analyses were
performed by Elemental Microanalysis Ltd, Okehampton, Devon,
UK. [Ru(PPh3)3(CO)H2]32 and [Ru(dppp)(PPh3)(CO)H2]18a were
prepared according to literature procedures.
free PPh3. While H NMR showed signals for free PhCH2OH,
no major ruthenium hydride containing species could be detected.
Proposed mechanism for the catalyst action in the alkylation of 6
In light of the information on the deactivation chemistry of 18,
a plausible mechanistic pathway for the Knoevenagel reaction
of 1a with 18 as the catalytic precurs◦or is shown in Scheme 6.
Under the reaction conditions of 120 C, loss of hydrogen from
18 would be accessible,31 allowing the incoming benzyl alcohol
to co-ordinate. Oxidative addition of the O–H bond followed
by b-hydride elimination would afford benzaldehyde. While the
benzaldehyde could act as a ligand (as shown for the putative
intermediate complex 25), the need for piperidinium acetate in
the conversion of benzaldehyde and 6 into alkene suggests that
this process may occur without any involvement of the ruthenium
centre. The alkene could return to the ruthenium, where upon
hydrogenation would generate the final organic product 7a and
regenerate a catalytically active ruthenium fragment. Under the
stoichiometric conditions described above, in which the aldehyde
would not be consumed by ketonitrile, loss of hydrogen from 25
could take place followed by oxidative addition into the C–H
bond of the aldehyde, decarbonylation of the resulting aldehyde (to
afford 23), elimination of benzene and re-addition of H2 to yield
the dicarbonyl species 20. The formation of the xantphos–PPh3
Representative catalytic procedure: 4,4-dimethyl-3-oxo-2-
benzylpentanenitrile (7a). To a solution of benzyl alcohol 1a
(310 mL, 3 mmol) and 4,4-dimethyl-3-oxopentanenitrile 6 (376 mg,
3 mmol) in toluene (3.0 mL) in a carousel reaction tube was added
[Ru(PPh3)3(CO)H2] (13.9 mg, 0.015 mmol), xantphos 5 (8.6 mg,
0.015 mmol) and piperidinium acetate (21.8 mg, 0.15 mmol). The
reaction mixture was heated to 110 ◦C in a pre-heated carousel
reaction station and stirred for 3 h. After cooling, the solvent
was removed in vacuo and the crude product was purified by
column chromatography on silica using 19 : 1 petroleum ether
(bp 40–60 ◦C)–ether as the eluent, giving the title compound
720 | Dalton Trans., 2009, 716–722
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
(509 mg, 79% isolated yield) as a colourless oil. Found: C 78.01,
H 7.95, N 6.48. C14H17NO requires C 78.10, H 7.41, N 6.06%).
1H NMR (CDCl3, 300 MHz, 298 K) d/ppm: 7.25–7.09 (m, 5H,
1960 (s). Elemental analysis calcd (%) for C41H34O3P2Ru (737.71):
C 66.75, H 4.65; found: C 66.66, H 4.63.
3
2
ArH), 3.96 (app. t, JHH = 7.6 Hz, 1H, –CH–), 3.11 (dd, JHH
=
Acknowledgements
13.6 Hz, 3JHH = 7.6 Hz, 1H, –CHH–), 3.03 (dd, 2JHH = 13.6 Hz,
1
3JHH = 7.6 Hz, 1H, –CHH–), 0.99 (s, 9H, –C(CH3)3). 13C{ H}
We thank the EPSRC for funding and Professor P. W. N. M. van
Leeuwen for the gift of ligands 12–15.
NMR (CDCl3, 75.5 MHz, 298 K) d/ppm: 205.2 (CO), 136.4
(ArC1), 129.3 (ArC3), 129.1 (ArC2), 127.8 (ArC4), 117.3 (–CN),
45.7 (–C(CH3)3), 38.9 (–CH–), 36.2 (–CH2–), 25.8 (–C(CH3)3). IR
(Nujol, n/cm-1): 2242 (s) n(CN), 1716 (s) n(CO). HRMS (ESI):
m/z = 233.1649 (calcd 233.1648 for [M - NH4]+).
Notes and references
1 G. Zassinovich, G. Mestroni and S. Gladiali, Chem. Rev., 1992, 92,
1051; M. J. Palmer and M. Wills, Tetrahedron: Asymmetry, 1999, 10,
2045; S. Gladiali and E. Alberico, Chem. Soc. Rev., 2006, 35, 226.
2 R. Grigg, T. R. B. Mitchell, S. Sutthivaiyakit and N. Tongpenyai,
Tetrahedron Lett., 1981, 22, 4107.
3 For reviews of C–C bond formation from alcohols by borrowing
hydrogen see: G. Guillena, D. J. Ramo´n and M. Yus, Angew. Chem.,
Int. Ed., 2007, 46, 2358; M. H. S. A. Hamid, P. A. Slatford and J. M. J.
Williams, Adv. Synth. Catal., 2007, 349, 1555.
4 Examples of Ru catalysts: C. S. Cho, B. T. Kim, T.-J. Kim and S. C.
Shim, J. Org. Chem., 2001, 66, 9020; K. Motokura, D. Nishimura, K.
Mori, T. Mizugaki, K. Ebitani and K. Kaneda, J. Am. Chem. Soc.,
2004, 126, 5662; R. Mart´ınez, G. J. Brand, D. J. Ramo´n and M. Yus,
Tetrahedron Lett., 2005, 46, 3683.
5 Examples of Ir catalysts: K. Fujita and R. Yamaguchi, Synlett, 2005,
560; C. Lo¨fberg, R. Grigg, A. Keep, A. Derrick, V. Sridharan and
C. Kilner, Chem. Commun., 2006, 5000; C. Lo¨fberg, R. Grigg, M. A.
Whittaker, A. Keep and A. Derrick, J. Org. Chem., 2006, 71, 8023; M.
Morita, Y. Obora and Y. Ishii, Chem. Commun., 2007, 2850.
6 Examples of Pd catalysts;C. S. Cho, J. Mol. Catal. A: Chem., 2005, 240,
55; M. S. Kwon, N. Kim, S. H. Seo, I. S. Park, R. K. Cheedrala and J.
Park, Angew. Chem., Int. Ed., 2005, 44, 6913; Y. M. A. Yamada and Y.
Uozumi, Org. Lett., 2006, 8, 1375.
[Ru(xantphos)(PPh3)(CO)H2] (18). To
a
solution of
[Ru(PPh3)3(CO)H2] (0.25 g, 0.27 mmol) in toluene (10 mL)
was added xantphos (0.19 g, 0.33 mmol) and the mixture heated
to 120 ◦C for 3 h. Removal of solvent gave a red oily residue, which
was washed with EtOH (2 ¥ 10 mL) and hexane (1 ¥ 10 mL).
The resulting solid was recrystallised from benzene–hexane to
give 18 as an orange solid (0.121 g, 46% yield). Selected 1H NMR
2
([d8]-toluene, 500 MHz, 298 K) d/ppm: -6.67 (dddd, JHP
=
2
2
2
35.0 Hz, JHP = 28.0 Hz, JHP = 15.5 Hz, JHH = 6.6 Hz, 1H,
RuH), -8.72 (ddd, 2JHP = 77.0 Hz, 2JHP = 33.9 Hz, 2JHP = 27.3 Hz,
2JHH = 6.6 Hz, 1H, RuH); 31P{ H} ([d8]-toluene, 202 MHz, 203 K)
1
2
2
d/ppm: 30.5 (br m), 45.2 (dd, JPP = 239.1, JPP = 15.3 Hz),
58.5 (dd, JPP = 237.2 Hz, JPP = 15.5 Hz). Selected 13C{ H}
([d8]-toluene, 125 MHz, 298 K) d/ppm: 205.7 (m, CO). IR
(KBr) n/cm-1: 1946 (s) n(CO). Elemental analysis calcd (%)
for C58H49O2P3Ru (971.96): C 71.66, H 5.08; found: C 71.14, H
5.29; ESI-TOF MS: [M + H - H2]+ m/z = 971.1917 (theoretical
971.1921).
2
2
1
7 M. G. Edwards and J. M. J. Williams, Angew. Chem., Int. Ed., 2002, 41,
4740.
8 M. G. Edwards, R. F. R. Jazzar, B. M. Paine, D. J. Shermer, M. K.
Whittlesey, J. M. J. Williams and D. D. Edney, Chem. Commun., 2004,
90; see also: S. Burling, B. M. Paine, D. Nama, V. S. Brown, M. F.
Mahon, T. J. Prior, P. S. Pregosin, M. K. Whittlesey and J. M. J.
Williams, J. Am. Chem. Soc., 2007, 129, 1987.
9 P. A. Slatford, M. K. Whittlesey and J. M. J. Williams, Tetrahedron
Lett., 2006, 47, 6787; P. J. Black, G. Cami-Kobeci, M. G. Edwards,
P. A. Slatford, M. K. Whittlesey and J. M. J. Williams, Org. Biomol.
Chem., 2006, 4, 116.
10 S. J. Pridmore, P. A. Slatford and J. M. J. Williams, Tetrahedron Lett.,
2007, 48, 5111.
11 S. J. Pridmore, P. A. Slatford, A. Daniel, M. K. Whittlesey and J. M. J.
Williams, Tetrahedron Lett., 2007, 48, 5115.
[Ru(xantphos)(CO)3] (22). [Ru3(CO)12] (0.1 g, 0.16 mmol) and
xantphos (0.27 g, 0.47 mmol) were dissolved in toluene (20 mL)
and the solution transferred to a 100 mL stainless-steel autoclave.
The sample was placed under 25 atm CO and heated at 100 ◦C
for 72 h. The solution was then cooled to room temperature, the
pressure released and the yellow solution transferred to a Schlenk
tube. Removal of the solvent afforded a yellow-orange oil, which
was crystallized from benzene–hexane to give 22 as crystalline
1
yellow blocks. (0.25 g, 70% yield). 31P{ H} ([d8]-toluene, 162 MHz,
1
298 K) d/ppm: 27.9 (s). Selected 13C{ H} ([d8]-toluene, 100 MHz,
12 N. Anand, N. A. Owston, A. J. Parker, P. A. Slatford and J. M. J.
Williams, Tetrahedron Lett., 2007, 48, 7761.
13 A. R. Elsagir, F. Gassner, H. Gorls and E. Dinjus, J. Organomet. Chem.,
2000, 597, 139.
298 K) d/ppm: 215.4 (t, 2JCP = 3.6 Hz, CO). IR (Nujol) n/cm-1:
n(CO) 2007 (s), 1921 (s), 1920 (s). Elemental analysis calcd (%) for
C42H32O4P2Ru (763.70): C 66.05, H 4.22; found: C 66.11, H 4.28.
14 P. C. J. Kamer, P. W. N. M. van Leeuwen and J. N. H. Reek, Acc. Chem.
Res., 2001, 34, 895; Z. Freixa and P. W. N. M. van Leeuwen, Dalton
Trans., 2003, 1890.
15 G. Dube, D. Selent and R. Taube, Z. Angew. Chem., 1985, 25, 154.
16 P. Dierkes and P. W. N. M. van Leeuwen, J. Chem. Soc., Dalton Trans.,
1999, 1519; P. W. N. M. van Leeuwen, P. C. J. Kamer, J. N. H. Reek and
P. Dierkes, Chem. Rev., 2000, 100, 2741.
17 L. A. van der Veen, P. K. Keeven, G. C. Schoemaker, J. N. H. Reek,
P. C. J. Kamer, P. W. N. M. van Leeuwen, M. Lutz and A. L. Spek,
Organometallics, 2000, 19, 872; J. J. Carbo´, F. Maseras, C. Bo and
P. W. N. M. van Leeuwen, J. Am. Chem. Soc., 2001, 123, 7630.
18 C. W. Jung and P. E. Garrou, Organometallics, 1982, 2, 658; H. Kawano,
R. Tanaka, T. Fujikawa, K. Hiraki and M. Onishi, Chem. Lett., 1999,
401.
[Ru(xantphos)(CO)2H2] (20). A THF (10 mL) solution of ◦22
(0.05 g, 0.065 mmol) was photolysed (300 W Xe arc) at ca. 0 C
under a steady flow of H2 for 0.5 h. In this time, the orange
solution turned a pale straw colour. The solvent was removed
under vacuum to leave a pale yellow residue, which was washed
with EtOH (10 mL) and hexane (2 ¥ 10 mL). The resulting solid
was recrystallised from THF–EtOH to give 20 as small, pale yellow
crystals (0.037 g, 76% yield). Selected 1H NMR ([d8]-toluene,
2
2
500 MHz, 298 K) d/ppm: -6.13 (ddd, JHP = 26.8 Hz, JHP
=
2
2
17.6 Hz, JHH = 6.5 Hz, 1H, RuH), -7.65 (ddd, JHP = 83.1 Hz,
1
2JHP = 30.5 Hz, 2JHH = 6.5 Hz, 1H, RuH). 31P{ H} ([d8]-toluene,
19 Jung and Garrou reported for a variety of different P–P that the PPh3
resonance in [Ru(P–P)(PPh3)(CO)H2] appears at d ca. 59 ppm in the
202 MHz, 298 K) d/ppm: 24.6 (d, 2JPP = 21.5 Hz), 35.1 (d, 2JPP
=
1
31P{ H} NMR spectrum (ref. 18a). The 31P signals in 18 are assigned
1
21.5 Hz). Selected 13C{ H} ([d8]-toluene, 125 MHz, 298 K) d/ppm:
on this basis.
203.3 (dd, 2JCP = 81.9 Hz, 2JCP = 7.1 Hz, CO), 201.5 (dd, 2JCP
=
20 Heating 18 with 3 equiv. of P(p-tolyl)3 at 70 ◦C for 15 h failed to show
any evidence for phosphine exchange or coordination.
11.2 Hz, 2JCP = 8.5 Hz, CO). IR (Nujol) n/cm-1: n(CO) 2007 (s),
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 716–722 | 721
©

View Article Online
21 M. A. Zuideveld, B. H. G. Swennenhuis, M. D. K. Boele, Y. Guari,
G. P. F. van Strijdonck, J. N. H. Reek, P. C. J. Kamer, K. Goubitz, J.
Fraanje, M. Lutz, A. L. Spek and P. W. N. M. van Leeuwen, J. Chem.
Soc., Dalton Trans., 2002, 2308; K. A. Len˜ero, M. Kranenburg, Y.
Guari, P. C. J. Kamer, P. W. N. M. van Leeuwen, S. Sabo-Etienne and
B. Chaudret, Inorg. Chem., 2003, 42, 2859.
but efforts to either generate this complex in higher yield or separate it
were unsuccessful.
26 M. K. Whittlesey, R. N. Perutz, I. G. Virrels and M. W. George,
Organometallics, 1996, 16, 268; T. Gottschalk-Gaudig, K. Folting and
K. G. Caulton, Inorg. Chem., 1999, 38, 5241; D. Schott, C. J. Sleigh, J. P.
Lowe, S. B. Duckett, R. J. Mawby and M. G. Partridge, Inorg. Chem.,
2002, 41, 2960; D. Schott, P. Callaghan, J. P. Dunne, S. B. Duckett,
C. Godard, J. M. Goicoechea, J. N. Harvey, J. P. Lowe, R. J. Mawby,
G. Mu¨ller, R. N. Perutz, R. Poli and M. K. Whittlesey, Dalton Trans.,
2004, 3218.
22 P. H. M. Budzelaar, g-NMR, version 4, Cherwell Scientific Publishing
Ltd., Oxford, UK, 1995–1997.
23 Crystal Data for 18, C58H49O2P3Ru, M = 971.95, T = 150 K, l =
¯
˚
0.71073 A, triclinic, space group P1 (no. 2), a = 11.0940(1), b =
˚
12.9620(1), c = 18.1840(2) A, a = 92.599(1), b = 92.909(1), g =
27 The X-ray crystal structure† of 22 is reported in the ESI.
28 The dppp complexes, [Ru(dppp)(CO)2H2] and [Ru(dppp)(CO)3], were
prepared via analogous procedures to those used for 20 and 22.
29 Crystal Data for 20, C41H34O3P2Ru, M = 737.69, T = 150 K,
◦
113.854(1) , U = 2382.20(4) A , Z = 2, Dc 1.355 g cm-3, m =
3
˚
0.473 mm-1, F(000) = 1004, crystal size 0.20 ¥ 0.12 ¥ 0.07 mm,
unique reflections = 10 813 [Rint = 0.0523], observed I > 2sI = 9114,
data/restraints/parameters = 10 813/0/632, R1 = 0.0399 wR2 = 0.07
(observed data), R1 = 0.0549 wR2 = 0.080 (all data), max. peak/hole
¯
˚
l = 0.71073 A, triclinic, space group P1 (no. 2), a = 10.0620(2),
˚
b = 10.6710(2), c = 18.1510(4) A, a = 73.413(1), b = 73.881(1),
◦
0.466 and -0.579 e A , software used, SHELXS, SHELXL and
g = 73.441(1) , U = 1749.00(6) A , Z = 2, Dc 1.401 g cm-3, m =
-3
3
˚
˚
ORTEX. Hydrides, located and refined at 1.6 A from Ru1. 50 : 50
0.577 mm-1, F(000) = 756, crystal size 0.08 ¥ 0.08 ¥ 0.06 mm,
unique reflections = 7924 [Rint = 0.0703], observed I > 2sI = 5643,
data/restraints/parameters = 7924/2/435, R1 = 0.0478 wR2 = 0.0800
(observed data), R1 = 0.0841 wR2 = 0.0909 (all data), max. peak/hole
˚
disorder in phenyl carbons C36–C40.
24 This compound was identified by an independent synthesis using the
method of Duckett and co-workers (D. Blazina, J. P. Dunne, S. Aiken,
S. B. Duckett, C. Elkington, J. E. McGrady, R. Poli, S. J. Walton, M. S.
Anwar, J. A. Jones and H. E. Carteret, Dalton Trans., 2006, 2072) by
-3
˚
0.510 and -1.125 e A , software used, SHELXS, SHELXL and
˚
ORTEX. Hydrides, located and refined at 1.6 A from Ru1.
1
heating [Ru(PPh3)2(CO)3] with xantphos. 31P{ H} ([d8]-toluene, 298 K)
30 M. P. Waugh, R. J. Mawby, A. J. Reid, R. J. Carter and C. D. Reynolds,
Inorg. Chim. Acta, 1995, 240, 263.
31 H–D exchange between the hydride ligands of 18 and D2 (1 atm)
was apparent by 1H NMR spectroscopy after just 30 min at
50 ◦C.
2
2
d/ppm: 34.4 (d, J(P,P) = 84.3 Hz), 54.8 (d, J(P,P) = 84.3 Hz); IR
(Nujol) n/cm-1: n(CO) 1909 (s), 2003 (s). ESI-TOF MS: [M + H - CO]+
m/z = 971.1990 (theoretical 971.1990).
25 A small amount of a second species, which displayed a doublet hydride
resonance at d -7.63 ppm [2J(H,P) = 45.7 Hz] and a singlet 31P
resonance at d 29.2 ppm, was always detected in the formation of 20,
32 N. Ahmad, J. J. Levison, S. D. Robinson and M. F. Uttley, Inorg. Synth.,
1974, 15, 48.
722 | Dalton Trans., 2009, 716–722
This journal is
The Royal Society of Chemistry 2009
©