Hydrogen transfer from formic acid to alkynes catalyzed by a diruthenium complex

Article abstract of DOI:10.1139/v00-169

The diruthenium(0) complex [Ru₂(μ-CO)(CO)₄(μ-dppm)₂] (1) (dppm = Ph₂PCH₂PPh₂), is a catalyst for the transfer hydrogenation, using formic acid as hydrogen donor, of the alkynes PhC≡CPh, PhC≡CMe, EtC≡CEt, and PrC≡CPr but not of the terminal alkynes HC≡CH, PhC≡CH, BuC≡CH, or the alkynes containing one or two electron-withdrawing substituents PhC≡CCO₂Me and MeO₂CC≡CCO₂Me. In the successful reactions, the formic acid is first decomposed to carbon dioxide and hydrogen, which then hydrogenates the alkynes in a slower reaction. In the unsuccessful reactions, the decomposition of formic acid is strongly retarded by the alkyne. In the case with the alkyne PhC≡CH, it is shown that the alkyne reacts with protonated 1 to give first [Ru₂(μ-CPh=CH₂)(CO)₄(μ-dppm) ₂][HCO₂], which then isomerizes to give the catalytically inactive, stable complex [Ru₂(μ-CH=CHPh)(CO)₄(μ-dppm)₂][HCO ₂]. This complex has been structurally characterized and both of the μ-styrenyl complexes are shown to be fluxional in solution.

Full text of DOI:10.1139/v00-169

915
Hydrogen transfer from formic acid to alkynes
catalyzed by a diruthenium complex
Yuan Gao, Michael C. Jennings, and Richard J. Puddephatt
Abstract: The diruthenium(0) complex [Ru2(µ-CO)(CO)4(µ-dppm)2] (1) (dppm = Ph2PCH2PPh2), is a catalyst for the
transfer hydrogenation, using formic acid as hydrogen donor, of the alkynes PhCϵCPh, PhCϵCMe, EtCϵCEt, and
PrCϵCPr but not of the terminal alkynes HCϵCH, PhCϵCH, BuCϵCH, or the alkynes containing one or two electron-
withdrawing substituents PhCϵCCO2Me and MeO2CCϵCCO2Me. In the successful reactions, the formic acid is first
decomposed to carbon dioxide and hydrogen, which then hydrogenates the alkynes in a slower reaction. In the unsuc-
cessful reactions, the decomposition of formic acid is strongly retarded by the alkyne. In the case with the alkyne
PhCϵCH, it is shown that the alkyne reacts with protonated 1 to give first [Ru2(µ-CPh=CH2)(CO)4(µ-dppm)2][HCO2],
which then isomerizes to give the catalytically inactive, stable complex [Ru2(µ-CH=CHPh)(CO)4(µ-dppm)2][HCO2].
This complex has been structurally characterized and both of the µ-styrenyl complexes are shown to be fluxional in so-
lution.
Key words: ruthenium, hydrogenation, catalysis, binuclear.
Résumé : Le complexe de diruthénium(0) [Ru2(µ-CO)(CO)4(µ-dppm)2] (1) (dppm = Ph2PCH2PPh2) est un catalyseur
pour l’hydrogénation par transfert dans laquelle l’acide formique est utilisé comme source d’hydrogène; cette réaction
est efficace pour les alcynes tels que PhCϵCPh, PhCϵCMe, EtCϵCEt et PrCϵCPr, mais elle ne fonctionne pas avec les
alcynes terminaux, tels que HCϵCH, PhCϵCH, BuCϵCH, ou avec les alcynes comportant un ou deux substituants
électroaffinitaires, tels que PhCϵCCO2Me et MeO2CCϵCCO2Me. Dans les réactions qui donnent les résultats espérés,
l’acide formique est initialement décomposé en bioxyde de carbone et en hydrogène qui provoque l’hydrogénation des
alcynes dans une réaction plus lente. Dans les réactions qui ne donnent pas les résultats espérés, la décomposition de
l’acide formique est fortement ralentie par l’alcyne. Dans le cas de l’alcyne PhCϵCH, on a démontré que l’alcyne réa-
git avec le complexe 1 protoné pour donner dans une première étape le [Ru2(µ-CPh=CH2)(CO)4(µ-dppm)2][HCO2] qui
s’isomérise alors pour donner le complexe stable et inactif [Ru2(µ-CH=CHPh)(CO)4(µ-dppm)2][HCO2]. On a caractérisé
la structure de ce complexe et on a démontré que, en solution, les deux complexes µ-styrényles sont en état de fluxion.
Mots clés : ruthénium, hydrogénation, catalyseur, binucléaire.
[Traduit par la Rédaction] 921
Introduction
Gao et al.
toward the decomposition of formic acid (5), and also to react
easily with terminal alkynes (6), it was considered likely that it
might also be active for the catalytic transfer hydrogenation of
alkynes using formic acid. This article reports that the desired
catalysis is successful and also describes studies of the reaction
mechanism.
There has been increasing interest in using formic acid in
catalytic transfer hydrogenation of multiple bonds, as an alter-
native to the traditional hydrogenation using hydrogen gas (1–
3). Several active catalysts for this transfer hydrogenation have
been identified, and all such catalysts are also active in the de-
composition of formic acid to CO2 and H2 (1–4). All the cata-
lysts studied so far have been mononuclear, and only a few of
these have been studied in terms of reaction mechanisms (3, 4).
Since the binuclear complex [Ru2(µ-CO)(CO)4(µ-dppm)2]
(dppm = Ph2PCH2PPh2) has been shown to have high activity
Results
Hydrogen transfer from formic acid to internal alkynes
In the presence of [Ru2(µ-CO)(CO)4(µ-dppm)2] (1) the re-
action of HCOOH with internal alkynes, containing either
phenyl or alkyl substituents (diphenylacetylene, 1-phenyl-1-
propyne, 3-hexyne, or 4-octyne), in acetone at room temper-
ature gave the pure cis-alkene as shown in eq. [1]. No trace
of trans-alkene, and no further hydrogenation of alkene to
alkane, was detected when the reactions were monitored by
1H NMR. Using a 20:10:1 ratio of formic acid to
diphenylacetylene to complex 1, the reaction to give cis-
stilbene was complete in 1 day. Incomplete conversion of
about 40% was observed when using equimolar amounts of
formic acid and diphenylacetylene under similar conditions.
The conversion rate of alkyne to alkene was found to follow
Received August 8, 2000. Published on the NRC Research
Press Web site at http://canjchem.nrc.ca on July 5, 2001.
Dedicated to Professor B.R. James, in recognition of his
distinguished contributions to homogeneous catalysis, on the
occasion of his 65th birthday.
Y. Gao, M.C. Jennings, and R.J. Puddephatt.1 Department
of Chemistry, University of Western Ontario, London, ON
N6A 5B7, Canada.
1Corresponding author (telephone: (519) 679-2111; fax: (519)
850-2371; e-mail: pudd@uwo.ca).
Can. J. Chem. 79: 915–921 (2001)
© 2001 NRC Canada
DOI: 10.1139/cjc-79-5/6-915

916
Can. J. Chem. Vol. 79, 2001
Scheme 1. Reagents: (i) H+; (ii) HCO2H, H2; (iii) H2, H+; (iv) HCO2 , CO; (v) CO2; (vi) CO, H2; (vii) CO; (viii) H+, CO, H2.
–
the sequence: diphenylacetylene > > 1-phenyl-1-propyne >
3-hexyne ≈ 4-octyne. For example, using a 10:10:1 molar ra-
tio of HCO2H:PhCCMe:complex 1, the conversion was
Hydrogen transfer from formic acid to terminal
alkynes and to internal alkynes with electron-
withdrawing substituents
The terminal alkynes (PhCCH, HCCH, 1-hexyne) or inter-
nal alkynes with electron-withdrawing substituents
(PhCCCO2Me, MeO2CCCCO2Me) strongly inhibited the de-
composition of formic acid to carbon dioxide and hydrogen.
Thus, using H13CO2H to allow detection of 13CO2 by 13C
NMR, no CO2 was detected at room temperature and only a
trace was detected on warming the sample to 50°C. In terms
of the ruthenium complexes present, in each case complexes
2 and 3 were formed first, in the same way as in the absence
of the alkyne. This is consistent with the faster reaction of 1
with formic acid than with the alkynes (5, 6). However, the
other intermediates of Scheme 1, including the important in-
termediate 4, were not observed. Instead, products arising
from reactions of 2 or 3 with the alkynes were observed. In
the cases with the alkynes PhCϵCCO2Me or
MeO2CCϵCCO2Me, the reactions gave complex mixtures
that could not be identified but the terminal alkynes reacted
more selectively, as described below.
about 15% after 1 day.
These catalytic hydrogenation reactions were all slower
than the catalytic decomposition of formic acid by complex
1
1 (5). This reaction could be monitored independently by H
NMR and it was complete in about 20 min under the above
conditions, to give CO2 and H2. It is therefore clear that the
catalytic transfer hydrogenations occur in two steps. First,
the formic acid is decomposed to give hydrogen and then, in
a much slower reaction, catalytic hydrogenation of the
alkyne occurs. Excess formic acid is required because excess
hydrogen is needed for the slow second step. A similar reac-
tion sequence has been proposed for catalytic transfer hydro-
genation using formic acid and the mononuclear catalyst
[RuCl2(PPh3)3], either alone or in combination with the het-
erogeneous hydrogenation catalyst Pd/C (7). Complex 1 was
The reaction of PhCϵCH with complex 2 was relatively
slow and thus was chosen for detailed study. At room tem-
perature, the reaction occurred through several steps, as
monitored by 1H and 31P NMR, to finally give a single prod-
uct after 1 day at room temperature. This compound was
shown to be
a
catalyst for hydrogenation of
diphenylacetylene using hydrogen gas at 1 atm (1 atm =
101.325 kPa) pressure, consistent with the above mecha-
nism, and the hydrogenation is not affected by the presence
or absence of free carbon dioxide.
characterized
crystallographically
as
[Ru2(µ-η1:η2-
CHCHPh)(CO)4(µ-dppm)2][HCOO] (7) and it is logical to
suggest that it is formed from 2 by cis-insertion of
phenylacetylene into the ruthenium hydride bond, with loss
of a carbonyl ligand.
The reactions were monitored at low temperature (–5 to
–10°C) by NMR using H13COOH. The catalytic decomposi-
tion of formic acid occurred by way of the intermediates
shown in Scheme 1, and occurred at a similar rate as in the
absence of alkyne (5). No new intermediates were observed
and complex 1 was regenerated once the catalytic decompo-
sition of formic acid was complete (5). No alkene was de-
tected until the formic acid was completely decomposed to
CO2 and H2. During the subsequent slow hydrogenation
step, the only complex present in detectable quantity was
complex 1. This result is not unexpected since these internal
alkynes without electronegative substituents (eq. [1]) do not
react with complex 1, but it does show that all reaction inter-
mediates are short-lived (6).
The molecular structure of the cation 7 is shown in Fig. 1
and selected interatomic distances and angles are listed in
Table 1. The unit cell contains two independent cations 7 as
well as formate anions and solvent molecules. Table 1 shows
equivalent bond parameters for the two cations 7 in each
row, and there are no major differences. The discussion will
focus on the cation containing atoms Ru(1) and Ru(3) shown
in Fig. 1. Each cation 7 contains a Ru2(µ-dppm)2 group in
which the Ru2P4C2 atoms adopt a slightly twisted, extended
boat conformation. In addition each ruthenium is bound to
two terminal carbonyl ligands and to the bridging styrenyl
group. The α-carbon (C(3)) of the styrenyl group is σ-
© 2001 NRC Canada

Gao et al.
917
Fig. 1. A view of the structure of the cation [Ru2(CO)4(µ-η1,η2-
CH=CHPh)(µ-dppm)2]+. Only the ipso carbon atoms of the
phenyl substituents of the dppm ligands are shown for clarity.
Table 1. Selected bond lengths (Å) and bond angles (°) for com-
plex 7.
Bond lengths (Å)
Ru(1)—C(11A)
Ru(1)—C(12A)
Ru(1)—C(3)
Ru(1)—P(1)
Ru(1)—P(5)
1.87(1)
1.90(1)
2.27(1)
2.378(3) Ru(4)—P(8)
2.387(3) Ru(4)—P(4)
Ru(4)—C(23B)
Ru(4)—C(24B)
Ru(4)—C(13)
1.89(1)
1.90(1)
2.25(1)
2.349(3)
2.382(3)
2.41(1)
2.867(1)
1.87(1)
1.96(1)
2.09(1)
2.368(3)
2.372(3)
1.40(1)
1.48(1)
Ru(1)—C(4)
2.39(1)
2.855(1) Ru(2)—Ru(4)
Ru(4)—C(14)
Ru(1)—Ru(3)
Ru(3)—C(14A)
Ru(3)—C(13A)
Ru(3)—C(3)
Ru(3)—P(3)
Ru(3)—P(7)
1.88(1)
1.92(1)
2.09(1)
2.347(3) Ru(2)—P(6)
2.363(2) Ru(2)—P(2)
Ru(2)—C(22B)
Ru(2)—C(21B)
Ru(2)—C(13)
C(3)—C(4)
C(4)—C(5)
1.38(1)
1.49(1)
C(13)—C(14)
C(14)—C(15)
Bond angles (°)
C(4)-C(3)-Ru(3)
C(4)-C(3)-Ru(1)
Ru(3)-C(3)-Ru(1)
C(3)-C(4)-C(5)
129.4(7)
77.8(6)
81.6(3)
123(1)
C(14)-C(13)-Ru(2) 131.1(9)
C(14)-C(13)-Ru(4)
Ru(2)-C(13)-Ru(4)
79.1(6)
82.8(4)
C(13)-C(14)-C(15) 122(1)
tained two vinyl resonances, each as a doublet of quintets, at
δ = 8.4 and δ = 5.5, assigned to the hydrogen atoms on the α-
and β-carbons [C(3) and C(4) in Fig. 1], respectively, (9).
3
The magnitude of the doublet coupling constants J (H-H) =
14 Hz confirm that these protons are trans within the vinyl
group, and so confirms the stereochemistry shown crystallo-
graphically. Both vinyl hydrogen resonances show effec-
tively equal coupling to the four phosphorus atoms, with
J (P-H)obs = 7 and 3 Hz for the α- and β-protons, respec-
tively. The phenyl protons of the bridging styrenyl ligand
were observed δ = 6.9 (para), δ = 6.5 (meta), and at δ = 5.7
(ortho). These unusual chemical shifts for aryl protons are
probably due to shielding by the surrounding four phenyl
groups of the dppm ligands. Two broad, unresolved reso-
nances were observed for the CHaHbP2 protons at δ = 4.2
and 3.7 ppm.
At –50°C, complex 7 exhibits an ABCD pattern of peaks
in the 31P NMR spectrum, consistent with the structure es-
tablished in the solid state by X-ray crystallography. The
fluxional process that leads to the equivalence of phosphorus
atoms must include not only the process A (Scheme 2),
which can be considered to arise from an intermediate in
which the plane of the vinyl group is perpendicular to the
metal—metal bond (10–12), but also process B (Scheme 2),
which requires an intermediate in which the plane of the vi-
nyl group is parallel with the metal—metal bond. Process A
is thought to occur through a µ-η1-vinyl intermediate (10–
12), but it is likely that process B occurs by way of an inter-
mediate with a terminal σ-vinyl group. Certainly the conven-
tional ruthenium—alkene π-bond will be lost in the
transition state in which the vinyl group and Ru—Ru bond
are coplanar. At low temperature, four resonances were ob-
served for the CH2P2 protons (one coupled pair at δ = 4.7
and 4.3 and the other at δ = 4.4 and 3.2), indicating that all
are inequivalent. The chemical shifts of the vinyl protons did
not change significantly but the appearance did. In particu-
bonded to Ru(3) (Ru(3)C(3) = 2.09(1) Å) and the C=C unit
is π-bonded to Ru(1) (Ru(1)C(3) = 2.27(1) Å, Ru(1)C(4) =
2.39(1) Å). Both C(3) and C(4) are chiral centers and the
unit cell contains equal numbers of the R,S and S,R enantio-
mers: Fig. 1 shows the S,R enantiomer and the molecule
containing Ru(2)Ru(4) is R,S. The distance C(3)—C(4) =
1.38(1) Å is slightly longer than a free double bond (1.34 Å)
but significantly shorter than a single bond (1.53 Å), indicat-
ing relatively weak back-bonding to the C=C π*-orbital. The
phenyl substituent of the styrenyl group is positioned be-
tween a sheath of phenyl substituents of the dppm ligands.
The angles Ru(3)-C(3)-C(4) = 129.4(7) and C(3)-C(4)-C(5) =
123(1)° are each somewhat greater than the ideal angles of
120° for a double bond and, of course, significantly greater
than the tetrahedral angle. Again this suggests that the car-
bon—carbon bond of the µ-styrenyl group is close to a dou-
ble bond. The distortion of the angle Ru(3)-C(3)-C(4) is
probably to maximize the Ru(1)-C(3-)C(4) π-bonding inter-
action. The distance Ru(1)—Ru(3) = 2.855(1) Å is consis-
tent with a metal—metal single bond. Each ruthenium then
has an 18-electron configuration if Ru(1) carries the positive
charge. The structural features are consistent with those of
other µ-alkenyl complexes (8).
The 31P NMR spectrum of 7 contained only a very broad
singlet at δ = 33 at room temperature and so 7 is clearly
1
fluxional. The H NMR spectrum at room temperature con-
© 2001 NRC Canada

918
Can. J. Chem. Vol. 79, 2001
Scheme 2.
Scheme 3.
dppm ligands. A much smaller motion of the α-phenyl
substituent is required in process B. Following this line of
reasoning, it is likely that process B is also fast for complex
7 (Scheme 2) and that process A is rate-determining. The
proposed fluxionality of 8 is shown in Scheme 3.
The complex 8 was formed rapidly and isomerized to 7
slowly over a period of several hours at room temperature.
Clearly then, 8 is the kinetic product arising from reaction of
PhCCH with 2, and it isomerizes slowly to the thermody-
namic product 7. The isomerization is suggested to occur by
a β-elimination–reinsertion sequence (Scheme 4). The
stereochemistry of the intermediate is uncertain, as is the
mechanism of insertion and β-elimination, which could oc-
cur at one or across both metal centers.
It is noteworthy that complex 7 is stable in solution in the
presence of excess PhCCH, hydrogen gas, acids (HBF4,
HCO2H), or bases (Et3N, potassium t-butoxide). The strong
inhibition of the catalytic decomposition of formic acid by
phenylacetylene, and absence of catalytic transfer hydroge-
nation of phenylacetylene, can therefore be attributed to for-
mation of the stable complex 7, which is not an active
catalyst. It has not been possible to characterize the products
obtained when using the alkynes HCCH, BuCCH,
MeO2CCCCO2Me, or PhCCCO2Me, since intractable mix-
tures were obtained, but it is likely that they also give stable
alkenyl or alkyne complexes that are not active catalysts.
lar, the β-H resonance now appeared as a doublet of dou-
blets with J (H-H) = 14 Hz and J (P-H) = 10 Hz, thus
showing coupling to only one phosphorus atom. The α-H
resonance gave a complex unresolved multiplet, indicating
nonequivalent couplings to several phosphorus atoms. These
data are all easily rationalized in terms of the static structure
determined crystallographically.
3
When the reaction of 1 with formic acid and phenyl-
acetylene was monitored by NMR, a reaction intermediate
was detected and characterized as [Ru2(µ-η1:η2-
CPhCH2)(CO)4(µ-dppm)2][HCOO] (8). Complex 8 displays
an AA′BB′ pattern of peaks in the 31P NMR spectrum with
δ (P) = 25.0 and 26.8 ppm. The vinyl protons appeared as
1
broad singlets in the H NMR spectrum at δ 4.8 and 5.5; the
absence of resolved coupling between them indicates that
these are geminal protons and this was confirmed by a 1H13C
HSQC experiment which showed that both vinyl protons are
attached to the same carbon atom at δ (C) = 111.5. The
phenyl protons of the styrenyl group appeared at δ = 6.3
(ortho), 6.6 (meta), and 6.8 (para). The 31P NMR spectrum
broadened but did not split further at –90°C. Since the static
structure is expected to give an ABCD pattern of peaks in
the 31P NMR spectrum, it is clear that the fluxionality differs
from that of complex 7. In particular, the two processes anal-
ogous to A and B of Scheme 2 must have different rates, one
being fast and the other slow on the NMR time scale; it is
likely that B is rapid but that A is slow. The distinction can
Discussion
In catalytic transfer hydrogenation using formic acid, the
donor must transfer two hydrogen atoms to the metal center
(with loss of CO2) to give metal hydride bonds, and the sub-
strate must coordinate to the metal and accept the hydrogen
atoms, typically by insertion followed by reductive elimina-
tion. However, within this framework there are many possi-
ble reaction sequences (1–3). One important classification is
based on whether the hydrides react rapidly with substrate as
they are formed, as is implied by the term transfer hydroge-
nation, or are first eliminated as free dihydrogen. The second
case is then equivalent to hydrogenation with hydrogen gas,
but with the hydrogen generated in situ from the formic acid
(13). There is evidence for both cases in catalytic transfer
hydrogenation using formic acid and mononuclear catalysts
1
be made on the basis of the CH2P2 resonances in the H
NMR spectrum. In each process, the styrenyl group stays on
the same side of the Ru2(µ-dppm)2 unit, so each CH2 group
will have nonequivalent protons CHaHb. However, process A
does not lead to equivalent dppm ligands, whereas process B
does. Hence, if it is process A that is fast, four resonances
(each 1H) are expected whereas, if process B is fast, two
resonances (each 2H) are expected for the CH2P2 protons.
The spectrum contains two resonances for the CH2P2 pro-
tons at δ = 3.2 (2H) and 3.4 (2H), indicating that it is process
B that is fast. It is likely that steric hindrance in complex 8
prevents process A, which requires the phenyl substituent of
the styrenyl group to swing by the phenyl substituents of the
© 2001 NRC Canada

Gao et al.
919
Scheme 4.
Scheme 5.
(1–3, 7). In the present case, the successful hydrogenations
all occur by the mechanism in which formic acid is catalyti-
cally decomposed to carbon dioxide and hydrogen, followed
by the slower catalytic hydrogenation of the alkyne.
In the successful catalytic reactions, no intermediates
were detected in which the alkyne or its hydrogenation prod-
ucts were coordinated to ruthenium. This was the case when
using any of HCOOH, HCOOH:Et3N (5:2), or H2 as the hy-
drogen source. It is likely that weak binding of the alkyne to
the diruthenium complex is a prerequisite to successful ca-
talysis, since those alkynes known to bind strongly (6) were
not hydrogenated. On the other hand, alkenes are not hydro-
genated by this catalyst system and they do not coordinate
either. It seems there is a balance such that both substrates
that bind strongly (terminal alkynes, alkynes with electon-
withdrawing substituents) and those that cannot coordinate
at all (alkenes) are not hydrogenated, and that substrates that
are hydrogenated can probably bind transiently to the
diruthenium center (internal alkynes without electron-
withdrawing groups).
Experimental
All manipulations were operated under a dry nitrogen at-
mosphere using either standard Schlenk techniques or a
glove box. Acetone was dried over 3 Å molecular sieves, to-
luene was dried by distillation from sodium benzophenone,
and CH2Cl2 was distilled before use from CaH2. The
HCOOH (96%) and H13COOH (95%, 99 C13 atom%) con-
tained 4 to 5% water and were used as purchased. [Ru2(µ-
CO)(CO)4(µ-dppm)2] (1) was synthesized according to liter-
ature procedure (15). H, 13C, and 31P NMR spectra were re-
corded using Varian Inova 400 or Gemini 300 spectrometers.
1
The mechanism of hydrogenation is necessarily specula-
tive. The only difference in observed intermediates, com-
pared to those observed in the catalytic decomposition of
formic acid in the absence of alkynes, was that the hydride
derivatives 5 and 6 were not observed. It is easy to under-
stand how the coordinatively unsaturated hydride 6 might re-
act with alkynes, and one possible mechanism based on this
observation is shown in Scheme 5. Since none of the possi-
ble intermediates 9–11 of Scheme 5 is directly observed, the
structures are very tentative. However, the overall mecha-
nism involving oxidative addition of hydrogen with CO loss
to give 5 and 6, followed by rapid alkyne coordination (com-
plex 9), cis-insertion to give a vinyl derivative (complexes
10, 11), and reductive elimination of alkene with CO addi-
tion to regenerate the resting state complex 1, is consistent
with the observations. There are numerous instances of re-
lated mechanistic proposals with mononuclear ruthenium
complexes, but the extension to binuclear catalysts is new
(13, 14).
Hydrogen transfer studies from HCOOH to alkynes
To a saturated solution of [Ru2(µ-CO)(CO)4(µ-dppm)2]
(1) (1.1 × 10–3 mmol) in acetone-d6 (0.5 mL, 2.2 mM) in an
NMR tube was added PhCCPh (1.98 mg, 1.1 × 10–2 mmol)
and the tube was sealed with a septum. HCOOH (1.00 µL,
2.2 × 10–2 mmol) was then injected through the septum by
syringe. The decay of formic acid and formation of cis-
stilbene were monitored as a function of time by comparing
the integrals of the formyl and vinyl hydrogen resonances
1
with the integral of the CH2P2 resonance of 1 in the H
NMR spectrum. The decomposition of HCO2H to H2 and
CO2 was complete in 20 min, while the hydrogenation of
PhCCPh to cis-stilbene was complete in 24 h at 20°C.
Reactions with diphenylacetylene, 1-phenyl-1-propyne, 3-
hexyne, and 4-octyne were carried out in a similar way. For
comparison of rates, the ratio of alkyne:formic acid:1 was
maintained at 10:10:1. Alkene products were characterized
1
by their H NMR spectra in acetone-d6 as follows: cis-
stilbene (16): δ (1H) = 6.6 (s, =CH); cis-1-phenylpropene
© 2001 NRC Canada

920
Can. J. Chem. Vol. 79, 2001
Table 2. Crystal data and structure refinement for
(17): δ (1H) = 7.3 (m, 5H, Ph), 6.42 (dq, J (H-H) = 13 Hz, J
(H-H) = 2 Hz, PhCH=), 5.77 (dq, J (H-H) = 13 Hz, J (H-H) =
7 Hz, MeCH=), 1.85 (dd, J (H-H) = 7 Hz, J (H-H) = 2 Hz,
Me); cis-3-hexene (18): δ (1H) = 5.3 (m, =CH0, 2.03 (m,
CH2), 0.9 (t, J (H-H) = 6 Hz, Me); cis-4-octene (19): δ (1H) =
5.3 (m, =CH), 2.03 (m, CH2)], 1.38 (m, CH2), 0.9 (t, J (H-
H) = 6 Hz, Me).
In the similar reaction using PhCCH (3 µL, 0.027 mmol),
the complete decomposition of HCOOH (6.5 µL,
0.130 mmol) took 2 weeks to reach completion and no hy-
drogenation product (styrene) was detected. The only ruthe-
nium complex present at the final stage was complex 7. The
7[HCO2]·1.5CH2Cl2·0.5C5H12·0.5 H2O.
Empirical formula
Formula weight
Temperature (K)
Wavelength (Å)
Crystal system
Space group
C67H60.75Cl3O6.50P4Ru2
1403.78
200(2)
0.71073
Triclinic
P-1
Cell dimensions
a (Å)
b (Å)
c (Å)
α (°)
β (°)
γ (°)
Volume (Å3)
Z
Density (calculated) (mg m–3)
Absorption coefficient (mm–1)
F(000)
Crystal size (mm3)
θ range for data collection (°)
Refl., ind. refl.
14.8925(9)
21.6229(15)
23.2355(14)
109.723(3)
101.557(3)
100.677(3)
6636.0(7)
alkynes
HCCH,
BuCCH,
PhCCCO2Me,
and
MeO2CCCCO2Me also retarded the decomposition of formic
acid by complex 1 and failed to give the alkenes expected to
be formed by catalytic hydrogenation.
4
Synthesis of [Ru2(CO)4(µ-η1,η2-CH=CHPh)(µ-
dppm)2][HCOO] 7
1.405
0.722
2858
To a saturated solution of [Ru2(µ-CO)(CO)4(µ-dppm)2] in
acetone-d6 (0.5 mL, 2.2 mM) in a septum-sealed NMR tube
was added PhCCH (3 µL, 0.027 mmol) and HCOOH
(6.5 µL, 0.130 mmol) by syringe. The solution was set aside
for 18 h at room temperature to give complex 7 as the only
product as determined by NMR. The same product was
formed by heating to 45°C for 3 h, and using CD2Cl2 as sol-
vent. Crystals were obtained from CD2Cl2 by slow diffusion
of pentane. The crystals were formed with solvent occluded
(see structure determination), which was partially lost on
drying, thus making it difficult to obtain good analytical
0.46 × 0.14 × 0.11
2.58–30.15
72 876, 34 486 (R(int)
= 0.0980)
Absorption correction
Scalepack
Max and min transmission
Data / restraints / parameters
Goodness-of-fit on F2
0.9248 and 0.7324
34 486 / 22 / 1493
1.014
Final R indices [I > 2σ(I)]
R1 = 0.1131, wR2 =
0.2807
1.591 and –0.769
1
data. H NMR (acetone-d6, 20°C) δ: 8.4 (s, 1H, HCOO], 8.2
Largest diff. peak and hole (e Å–3)
(m, 1H, PhCH-CH), 5.4 (d quin, 1H, 3J (H-H) = 14 Hz, J (P-
H) = 3 Hz, PhCH-CH), 5.7 (d, 2H, 3J (H-H) = 8 Hz,
PhCHCH), 6.5 (t, 2H, J (H-H) = 8 Hz, PhCHCH), 6.8 (t,
1H, J (H-H) = 8 Hz, PhCHCH), 5.05 (br s, 2H, P-CH-P),
cessive difference Fouriers. All non-hydrogen atoms were
refined with anisotropic thermal parameters. One formate
anion was poorly behaved and the C—O bond distances
were fixed. The hydrogen atoms were calculated geometri-
cally and were riding on their respective carbon atoms.
There were several disordered solvent molecules in the lat-
tice, and these were modeled isotropically and with partial
occupancies. The crystal data and refinement parameters are
listed in Table 2. The crystal was of poor quality, with disor-
dered anion and solvent molecules, but the structure of the
cations was well-defined.
3.65 (br s, P-CH-P). 31P NMR δ: 33 (br s, dppm). H NMR
1
(acetone-d6, –50°C) δ: 8.6 (s, 1H, HCOO), 8.4 (m, 1H,
PhCH-CH), 5.4 (dd, 1H, 3J (H-H) = 14 Hz, 3J (P-H) =
10 Hz, PhCH-CH), 5.6 (d, 2H, J (H-H) = 7 Hz, PhCHCH),
6.44 (t, 2H, J (H-H) = 7 Hz, PhCHCH), 6.8 (t, 1H, J (H-H) =
7 Hz, PhCHCH), 4.7, 4.4, 4.3, 3.2 (m, each 1H, P-CH-P).
31P NMR δ: 49.3 (ddd, J (Pa-Pb) = 56 Hz, J (Pa-Pc) =
237 Hz, J (Pa-Pd) = 30 Hz, Pa), 35.1 (ddd, J (Pa-Pb) = 56 Hz,
J (Pb-Pc) = 38 Hz, J (Pb-Pd) = 267 Hz, Pb), 32.6 (ddd, J (Pa-
Pc) = 237 Hz, J (Pb-Pc) = 30 Hz, J (Pc-Pd) = 73 Hz, Pc), 30.9
(ddd, J (Pa-Pd) = 30 Hz, J (Pb-Pd) = 267 Hz, J (Pc-Pd) =
73 Hz, Pd).
Acknowledgments
Structure determination
We thank the Natural Sciences and Engineering Research
Council of Canada (NSERC) for financial support.
Crystals of [(CO)2Ru(µ-dppm)2(µ-CHCHPh)Ru(CO)2]-
[HCOO]·1.5CH2Cl2·0.5C5H12·0.5H2O were grown from
CH2Cl2–pentane. A yellow block was mounted on a glass fi-
bre at dry ice temperature. Data were collected at 200 K us-
ing a Nonius Kappa-CCD diffractometer using COLLECT
software (20). The unit cell parameters were calculated and
refined from the full data set. Crystal cell refinement and
data reduction was carried out using the Nonius DENZO
package. The data were scaled using SCALEPACK (21) and
no other absorption corrections were applied. The
SHELXTL 5.101 (22) program package was used to solve
the structure by direct methods and refinement was by suc-
References
1. R.A.W. Johnstone and A.H. Wilby. Chem. Rev. 85, 129
(1985).
2. G. Zassinovich, G. Mestroni, and S. Gladiali. Chem. Rev. 92,
1051 (1992).
3. (a) W. Leitner, J.M. Brown, and H. Brunner. J. Am. Chem.
Soc. 115, 152 (1993); (b) R. Noyori and S. Hashiguchi. Acc.
Chem. Res. 30, 97 (1997); (c) K. Matsumura, S. Hashiguchi,
T. Ikariya, and R. Noyori. J. Am. Chem. Soc. 119, 8738
© 2001 NRC Canada

Gao et al.
(1996); (d) J. Yu and J.B. Spencer. Chem. Commun. 1935
921
12. J.R. Shapley, S.I. Richter, M. Tachikawa, and J.B. Keister. J.
Organomet. Chem. 94, C43 (1975).
(1998); Chem. Eur. J. 5, 2237 (1999).
4. Y. Sasson and J. Blum. J. Org. Chem. 40, 1887 (1975).
5. Y. Gao, J. Kuncheria, G.P.A Yap, and R.J. Puddephatt. Chem.
Commun. 2365 (1998).
6. (a) J. Kuncheria, H.A. Mirza, J.J. Vittal, and R.J. Puddephatt.
Inorg. Chem. Comm. 2, 197 (1999); (b) J. Kuncheria, H.A.
Mirza, J.J. Vittal, and R.J. Puddephatt. J. Organomet. Chem.
593, 77 (2000); (c) H.A. Mirza, J.J. Vittal, and R.J.
Puddephatt. J. Coord. Chem. 37, 131 (1996); (d) H.A. Mirza,
J.J. Vittal, and R.J. Puddephatt. Organometallics, 13, 3063
(1994).
13. B.R. James. Homogeneous hydrogenation. Wiley, New York.
1973; (b) B.R. James. Catal. Today, 37, 209 (1997).
14. R.B. Bedford and C.S.J. Cazin. J. Organomet. Chem. 598, 20
(2000).
15. H.A. Mirza, J.J. Vittal, and R.J. Puddephatt. Inorg. Chem. 32,
1327 (1993).
16. (a) L. Breau, N.K. Sharma, I.R. Butler, and T. Durst. Can. J.
Chem. 69, 185 (1991); (b) G. Bosser and J. Paris. J. Chem.
Soc. Perkin Trans. 2, 2057 (1992).
17. L. Luan, J.S. Song, and R.M. Bullock. J. Org. Chem. 60, 7170
(1995).
7. (a) Y. Watanabe, T. Ohta, and Y. Tsuji. Bull. Chem. Soc. Jpn.
55, 2441 (1982); (b) B.T. Khai and A. Arcelli. J. Organomet.
Chem. 309, C63 (1986).
18. O.R. Brown and P.H. Middleton. J. Chem. Soc. Perkin Trans.
2, 955 (1984).
8. A.G. Orpen. J. Chem. Soc. Dalton Trans. 1427 (1983).
9. K.J. Edwards, J.S. Field, R.J. Haines, B.D. Homann, M.W.
Stewart, J. Sundermeyer, and S.F. Woollam. J. Chem. Soc.
Dalton Trans. 4171 (1996).
10. A.F. Dyke, S.A.R. Knox, M.J. Morris, and P.J. Naish. J. Chem.
Soc. Dalton Trans. 1417 (1983).
19. D.J. Hastings and A.C. Weedon. Can. J. Chem. 69, 1171
(1991).
20. COLLECT. Nonius, Delft. 1998.
21. SCALEPACK. Nonius, Delft. 1998.
22. G.M. Sheldrick. SHELXTL Version 5.101. Madison, Wiscon-
sin. 2000.
11. M.D. Fryzuk, T. Jones, and F.W.B. Einstein. Organometallics,
3, 185 (1984).
© 2001 NRC Canada