Synthesis and reactivity of cationic iridium(I) complexes of cycloocta-1,5-diene and chiral dithioether ligands. Application as catalyst precursors in asymmetric hydrogenation

Article abstract of DOI:10.1039/a704611a

New chiral dithioether compounds (-)-2,2-dimethyl-4,5-bis(isopropylsulfanylmethyl)-1,3-dioxolane (-)-diospr and (+)-2,2-dimethyl-4,5-bis(phenylsulfanylmethyl)-1,3-dioxolane (+)-diosph were prepared from diethyl (+)-L-tartrate. An alternative synthetic method for preparing the previously described bis(methylsulfanylmethyl) dithioether (-)-diosme was devised. By co-ordinating of the dithioethers to different (cycloocta-1,5-diene)iridium(I) compounds chiral cationic complexes [Ir(cod){(-)-diosme}]BF₄ 1, [Ir(cod){(-)-diospr}]BF₄·CH₂Cl₂ 2 and [Ir(cod){(+)-diosph}]BF₄ 3 were synthesized and then studied by ¹H, ¹³C NMR and FAB mass spectrometry. The complexes reacted with CO to give the corresponding binuclear tetracarbonyls [Ir₂(μ-L)₂(CO)₄][BF]₂ 4-6. The dithioether ligands were replaced by PPh₃ in 1-3 providing [Ir(cod)(PPh₃)₂]BF₄. The addition of H₂ to complexes 1 and 2 at -70°C gave cis-dihydridoiridium(III) complexes [IrH₂(cod){(-)-L}]BF₄ 7 and 8 which are in equilibrium in solution with the parent complexes, depending on the temperature. Two possible diastereomers were distinguished for 8 at low temperatures. Complexes 1-3 were active precursors in the asymmetric hydrogenation of different prochiral dehydroamino acid derivatives and itaconic acid, at room temperature under an atmospheric pressure of H₂, and the highest enantiomeric excess obtained was 47%.

Full text of DOI:10.1039/a704611a

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis and reactivity of cationic iridium(I) complexes of
cycloocta-1,5-diene and chiral dithioether ligands. Application as
catalyst precursors in asymmetric hydrogenation†
Montserrat Diéguez, Aránzazu Orejón, Anna M. Masdeu-Bultó, Raouf Echarri, Sergio Castillón,
Carmen Claver* and Aurora Ruiz*
Departament de Química, Facultat de Química, Universitat Rovira i Virgili, Pl. Imperial Tarraco
1, 43005 Tarragona, Spain
New chiral dithioether compounds (Ϫ)-2,2-dimethyl-4,5-bis(isopropylsulfanylmethyl)-1,3-dioxolane (Ϫ)-diospr
and (ϩ)-2,2-dimethyl-4,5-bis(phenylsulfanylmethyl)-1,3-dioxolane (ϩ)-diosph were prepared from diethyl
(ϩ)--tartrate. An alternative synthetic method for preparing the previously described bis(methylsulfanylmethyl)
dithioether (Ϫ)-diosme was devised. By co-ordinating of the dithioethers to different (cycloocta-1,5-
diene)iridium() compounds chiral cationic complexes [Ir(cod){(Ϫ)-diosme}]BF4 1, [Ir(cod){(Ϫ)-
diospr}]BF4ؒCH2Cl2 2 and [Ir(cod){(ϩ)-diosph}]BF4 3 were synthesized and then studied by 1H, 13C NMR and
FAB mass spectrometry. The complexes reacted with CO to give the corresponding binuclear tetracarbonyls
[Ir2(µ-L)2(CO)4][BF]2 4-6. The dithioether ligands were replaced by PPh3 in 1–3 providing [Ir(cod)(PPh3)2]BF4. The
addition of H2 to complexes 1 and 2 at Ϫ70 ЊC gave cis-dihydridoiridium() complexes [IrH2(cod){(Ϫ)-L}]BF4 7
and 8 which are in equilibrium in solution with the parent complexes, depending on the temperature. Two possible
diastereomers were distinguished for 8 at low temperatures. Complexes 1–3 were active precursors in the
asymmetric hydrogenation of different prochiral dehydroamino acid derivatives and itaconic acid, at room
temperature under an atmospheric pressure of H2, and the highest enantiomeric excess obtained was 47%.
Iridium() complexes of cycloocta-1,5-diene (cod) are of inter-
est because co-ordinated cod is readily replaced with other lig-
ands. It can also be hydrogenated with H2 to provide vacant co-
ordination sites around iridium and increase the catalytic activ-
ities of the complexes.1 Phosphorus ligands are mainly used,
but several reports describe results obtained with sulfur com-
pounds as ligands in homogeneous catalysis of reactions like
hydrogenation2–5 or hydroformylation2,6 of olefins. In the last
few years we have been studying the synthesis of new chiral
dithioethers so that they can be applied to the asymmetric
hydroformylation6d,g and hydrogenation of prochiral olefins. To
the best of our knowledge, dithioether chiral ligands have only
been used in the asymmetric hydrogenation of ketones with
palladium complexes.7
P(C6H5)2
P(C6H5)2
binap
R
R
R
R
P
P
P
P
R
R
R
R
duphos
bpe
R = Me, Et, Prn or Pri
Homogeneous asymmetric catalytic hydrogenation is one of
the most important applications of enantioselective catalytic
technologies8–11 and there are considerable economic and eco-
logical reasons for this.12 Recently great progress has been made
in this field and very high enatiomeric excess (e.e.) values have
been achieved. Since the beginning of the 90’s, Burk et al.13 have
been exploring novel electron-rich phospholane ligands (bpe
and duphos derivatives) and RajanBabu et al.14 have explored
electron-rich phosphinite -glucose-derived ligands that give
powerful rhodium catalysts for the enantioselective hydrogen-
ation of dehydroamino acids. They provide the most impressive
results to date as far as intrinsic reactivity and enantioselectivity
are concerned. Nevertheless, Ru–binap [binap = 2,2Ј-bis(di-
phenylphosphino)-1,1Ј-binaphthyl] complexes have been shown
to be more versatile asymmetric hydrogenation catalysts than
rhodium complexes for a series of unsaturated substrates.9a,c
It should be noted that in the synthesis of (S)-naproxen using
Ru–binap complexes both a low reaction temperature and a
high hydrogen pressure are necessary to obtain a high enantio-
meric excess,9c,15 Scheme 1. For the synthesis to be applied in
industry it would be desirable for the conditions to be smoother
O
O
O
OPh
Ph
R′2PO
D-glucose-derived phosphinite
R′ = Ph, C6H3(CF3)2 –3,5, C6H3F2 –3,5, or C6H3Me2 –3,5
OPR′2
H
2 100 atm
Ru-(S)-binap
CO2H
CO2H
MeOH, –20 o
C
MeO
MeO
(S )-naproxen, 97% e.e.
Scheme 1
and the ligands more accessible. Iridium systems have been used
less in asymmetric hydrogenation of prochiral olefins.16
By taking the diop [4,5-bis(diphenylphosphinomethyl)-2,2-
dimethyl-1,3-dioxolane] chiral structure as a model17 and by
modifying the known chiral dithiol diosh,18 almost twenty years
ago James and McMillan3 prepared the methyl dithioether
derivative (Ϫ)-diosme. Here, we report an alternative synthetic
method to prepare (Ϫ)-diosme, the synthesis of new more
† Non-SI unit employed: atm = 101 325 Pa.
J. Chem. Soc., Dalton Trans., 1997, Pages 4611–4618
4611

View Article Online
hindered chiral dithioethers (Ϫ)-diospr and (ϩ)-diosph, and
the preparation of cationic iridium() complexes with all three
of these compounds, and their reactivity. We also examine the
asymmetric hydrogenation of prochiral olefinic substrates by
using the complexes as catalyst precursors at room temperature
and atmospheric pressure of hydrogen.
of a triflate group. The compound is stable for a few hours
in air.
Treatment of compound V with NaH and propane-2-thiol in
tetrahydrofuran (thf) affords (Ϫ)-diospr which was isolated as a
colourless liquid. It is not stable in air at room temperature but
at low temperature is stable for several days. It was not possible
to prepare this dithioether from the ditosyl compound IV
previously described.18 Ditriflate V with a better leaving group,
had to be prepared.
Results and Discussion
Synthesis of the dithioethers
The dithioether diosph was prepared from the ditosyl com-
pound IV. A dimethylformamide solution of IV was treated
with benzenethiol in aqueous NaOH under reflux, to give
diosph as a white solid in high yield. It is stable in the solid state.
The dithioethers diospr and diosph were characterised by elem-
ental analyses and 1H and 13C NMR spectroscopy.
The new compounds (Ϫ)-diospr and (ϩ)-diosph were prepared
from diethyl (ϩ)--tartrate, I (Scheme 2). The route to com-
pounds III and IV follows previously described procedures.18
For the preparation of (Ϫ)-diospr the diol III is converted into
the ditriflate V by adding pyridine and triflic anhydride to a
dichloromethane solution of III. The ditriflate was isolated as a
white solid and characterised by elemental analyses and 1H and
13C NMR spectroscopy. The 13C NMR spectrum shows a quad-
ruplet at δ 118.6 (1JCF = 320.2 Hz) which confirms the presence
1
The H NMR spectra in CDCl3 of (Ϫ)-diospr and (ϩ)-
diosph show a singlet due to the methyl protons at δ 1.41 and
1.40 respectively. The four methylenic protons CH2S appear as
an ABX system showing a multiplet at δ 2.80 and 3.20 and the
two methinic protons CH also appear in an ABX system as a
multiplet signal at δ 3.90 and 4.05 respectively. The 13C NMR
spectra also show only one signal for the tertiary and secondary
carbon atoms for both compounds, in agreement with a C2
H
H
O
O
O
O
SH
SH
SR
SR
1
symmetry. The H NMR spectrum of (Ϫ)-diospr also shows a
H
H
septuplet at δ 3.05 (J = 6.7 Hz) corresponding to the methinic
protons of the isopropyl group and two doublets at δ 1.28
(J = 6.7) and 1.31 (J = 6.7 Hz) corresponding to the diastereo-
topic methyl protons of the isopropyl group. In the case of
(ϩ)-diosph the 1H and 13C NMR spectra also show signals from
R = Me, (–)-diosme
R = Pri, (–)-diospr
R = Ph, (+)-diosph
(–)-diosh
O
H
H
HO
HO
OEt
OEt
O
I
Ref. 18
O
H
O
O
OEt
OEt
H
O
II
Ref. 18
H
H
H
H
O
O
Ref. 18
O
O
(CF3SO2)2O
–20 °C
OH
O3SC6H4Me-p
O3SCF3
O3SCF3
OH
O3SC6H4Me-p
O
O
H
H
V (81%)
III
IV
Ref. 3
PriSH
Na(SPh)
MeSH
H
H
H
O
H
O
O
O
SMe
SMe
SPh
SPh
SPri
SPri
O
H
O
H
(+)-diosph
(–)-diospr
(–)-diosme
(85%)
(92%)
(77%)
Scheme 2 Synthetic procedures for the preparation of (Ϫ)-diosme, (Ϫ)-diospr and (ϩ)-diosph
4612
J. Chem. Soc., Dalton Trans., 1997, Pages 4611–4618

View Article Online
Table 1 The NMR spectroscopic data for the dithioethersa
Compound
CMe
CH
CH2
SCH
Me
SMe
Ph
CMe
1H
(Ϫ)-diosme
(Ϫ)-diospr
1.45(s)
1.41(s)
4.05(m)
3.90(m)
2.80(m)
2.80(m)
—
—
2.20(s)
—
—
—
—
—
3.05(sep)b
1.28(d)c
1.31(d)d
—
(ϩ)-diosph
13C
1.40(s)
4.05(m)
3.20(m)
—
—
7.15–7.35(m)
—
(Ϫ)-diosme
(Ϫ)-diospr
(ϩ)-diosph
27.2
27.1
27.3
79.6
79.9
79.0
36.8
33.2
37.0
—
35.3
—
—
23.2
—
16.5
—
—
—
—
109.0
108.9
109.7
126.3, 129.0
129.4, 135.6
a In CDCl3 solvent. Chemical shifts in ppm with SiMe4 as internal standard, coupling constants in Hz; room temperature. Abbreviations: s, singlet;
b 3
c 3
d 3
m, multiplet; d, doublet; sep, septuplet.
J
= 6.7 Hz.
J
= 6.7 Hz.
J = 6.7 Hz.
HH
HH
HH
[Ir(cod)2]BF4
+
(–)-L
(a) CH2Cl2
–cod
R
S
S
R
H
O
O
(b) CH2Cl2
–AgCl
1 [{Ir(µ-Cl)(cod)} ]
+
(–)-L
+
AgBF4
BF4
Ir
–
2
2
H
1 R = Me
2 R = Pri
(c) Me2CO
[Ir(cod)(Me2CO)2]BF4
+ (–)-L
Ph
H
O
(d) H2, CH2Cl2
–C8H16
S
Ir
S
[Ir(cod)2]BF4
+ (+)-diosph
BF4
O
H
Ph
3
Scheme 3 L = Chiral dithioether
the phenyl groups. The NMR signals for diospr and diosph
were assigned using correlation spectroscopy (COSY) and
heteronuclear correlation spectroscopy (HETCOR). The chem-
ical shifts and coupling constants as well as those of diosme are
listed in Table 1.
In this work, the methyl dithioether (Ϫ)-diosme was prepared
using an alternative route to that described by James and
McMillan.3 This consisted of treating the ditriflate V with NaH
and methanethiol (Scheme 2) and there was no need to prepare
the dithiol derivative (Ϫ)-diosh, thus involving fewer steps and
a greater yield.
+
H
R
S
H
H
H
H
O
H
H
H
H
Ir
H
H
H
H
C2
H
O
S
R
H
H
H
H
Fig. 1 Molecular symmetry of the cationic complexes [Ir(cod)L]ϩ
molecular species. For complex 2 CH2Cl2 is also lost. This sug-
gests that 1–3 are mononuclear complexes. The conductivity of
their acetone solutions at different concentrations gives A
values around 326 in Onsager’s equation (Λe = Λ0 Ϫ Ac ) show-
Synthesis of the dithioether complexes
¹
²
Olefinic complexes. [Ir(cod){(Ϫ)-diosme}]BF4 1, [Ir(cod)-
{(Ϫ)-diospr}]BF4ؒCH2Cl2 2 and [Ir(cod){(ϩ)-diosph}]BF4 3. In
order to obtain chiral complexes, the chiral dithioethers were
co-ordinated to [Ir(cod)]ϩ fragments. Compounds 1 and 2 were
obtained by the reactions (a)–(c) in Scheme 3 from different
starting materials. It was only possible to obtain 3 by bub-
bling H2 through a dichloromethane solution of [Ir(cod)2]BF4
and adding a stoichiometric amount of chiral diosph according
to (d) in Scheme 3. The presence of cyclooctane formed from
the hydrogenation of cycloocta-1,5-diene was observed in solu-
tion by 1H NMR spectroscopy.
ing the mononuclear nature of the complexes (1:1 electrolytes)
in acetone solution.19
Complexes 1–3 exist in solution as cations with a C2 mole-
cular symmetry and with the two alkyl or aryl groups in anti
position respectively as shown in Fig. 1. This follows from the
analysis of NMR data (see Table 2), assigned using COSY and
distortionless enhancement of polarisation transfer (DEPT)
spectra in combination with a 13C᎐1H correlation (HETCOR).
In particular for the signals from the dithioether ligands we can
see that: (i) the two CH᎐O groups are equivalent, one signal
being observed in the 1H and 13C NMR spectra in all cases; (ii)
there is only one signal in 13C NMR spectra for the two second-
ary carbons in the three complexes; for 1 and 2, in the 1H NMR
spectra the diastereotopic methylenic protons appear as two
double doublets which correspond to Hax (Jgem ca. 12.3, Jax-ax ca.
10.3) and Heq (Jgem ca. 12.3 and Jeq-ax = 3.5 Hz) respectively; for
3 these methylenic protons appear as a multiplet at δ 3.5 which
Complexes 1–3 were isolated by adding diethyl ether as yel-
low, orange and red-brown powders respectively. The elemental
analysis matches the stoichiometry [{Ir(cod)L}n][BF4]n for 1
and 3 and [{Ir(cod)L}n][BF4]nؒnCH2Cl2 for 2. The FAB mass
spectra show the heaviest ions at m/z 523 1, 579 2 and 647 3
Ϫ
which correspond to the loss of the BF4 anion from the
J. Chem. Soc., Dalton Trans., 1997, Pages 4611–4618
4613

View Article Online
Table 2 The NMR spectroscopic data for complexes 1–3a
cod
Dithioether
Complex
CH2
CH᎐CH
CMe
CH
CH2
SCH
—
Me
—
SMe
Ph
—
CMe
᎐
1H
1b
1.85(m)
2.00(m)
2.35(m)
1.75(m)
2.20(m)
2.40(m)
1.70(m)
2.30(m)
4.45(m)c
1.45(s)
1.41(s)
1.35(s)
4.30(m)
3.20(dd) Hax
3.35(dd) Heq
2.55 (s)
—
d
e
2f
3
4.15(m)
4.60(m)
4.68(m)
4.20(m)
3.11(dd) Hax
3.45(dd) Heq
3.80(sep)h 1.31(d)i
1.55(d)i
—
—
—
—
g
j
3.90(m)
4.05(m)
3.50(m)
—
—
7.30–7.55 (m)
—
13C
1b
32.2
32.3
29.5
33.6
31.1
31.2
79.4
80.0
76.0
80.8
79.4
79.6
26.8
26.7
26.8
78.4
78.1
78.1
43.0
39.9
40.3
—
—
17.0
—
—
—
110.2
110.1
111.0
2
36.1
—
21.0
22.7
—
3b
—
128.2, 130.3
130.8, 131.1
1
a Chemical shifts in ppm with SiMe4 as internal standard, coupling constant in Hz; room temperature; H and 13C NMR in CDCl3. b In CD2Cl2
d 2
e 2
g 2
solvent. c T = Ϫ40 ЊC; δ 4.35(m) and 4.45(m).
J
= 12.0, 3Jax-ax = 10.0 Hz.
J
= 12.0, 3Jeq-ax = 3.5 Hz. f δ 5.3 (CH2Cl2).
J
= 12.6, 3Jax-ax = 10.7
gem
gem
gem
h 3
i 3
j 2
Hz.
J
= 6.5 Hz.
J
= 6.4 Hz.
J
= 12.6, 3Jeq-ax = 3.5 Hz.
gem
HH
HH
cannot be resolved by changing the temperature; (iii) the two
MeS, PriS and PhS groups are equivalent; thus, in the 13C NMR
spectra there is only one signal for the MeS groups, three for the
Pri groups and four for the PhS groups.
R
S
S
R
H
O
O
Ir
BF4
The 1H NMR spectra show the olefinic proton signals of the
co-ordinated cyclooctadiene ligand as one multiplet for com-
plex 1 and two multiplets for 2 and 3. When the temperature is
decreased to Ϫ40 ЊC, the olefinic protons for 1 also appear as
two multiplets at δ 4.35 and 4.45 as expected for C2 symmetry.
For the endo- and exo-methylenic protons of cyclooctadiene
four signals were expected but only three multiplets were
observed for complexes 1 and 2 and two for 3. The 13C NMR
spectra reveal two different olefinic and methylenic resonances
for the three complexes which correspond to a C2 symmetry.
All the NMR data indicate that only one of two possible anti
diastereomers can be distinguished for complexes 1–3. When
the temperature was changed from Ϫ60 to 40 ЊC no other dia-
stereomers were observed in the 1 NMR spectra.
H
1 R = Me
2 R = Pri
r.t.
H2, –70 °C
HA
Ir
HB
BF
R
S
S
R
H
O
H
O
The C2-related equivalencies within the dithioether ligands
and the diolefin are consistent with the formulation of these
complexes as square-planar four-co-ordinate cations, similar to
related complexes containing chiral diphosphines, [Ir(cod){(Ϫ)-
chiraphos}]ϩ [(Ϫ)-chiraphos = (2S,3S)-2,3-bis(diphenylphos-
phino)butane] and [Ir(cod){(Ϫ)-norphos}]ϩ {(Ϫ)-norphos =
2,3-bis(diphenylphosphino)bicyclo[2.2.1]heptane}.20
7 R = Me
8 R = Pri
Scheme 4 r.t. = Room temperature
With PPh3. The reaction of the diene complexes 1–3 with
PPh3 in a complex:PPh3 molar ratio of 1:2 displaces the
dithioether ligands and provides the previously prepared com-
plex [Ir(cod)(PPh3)2]BF4.22
Reactivity of olefinic complexes
With carbon monoxide. Bubbling carbon monoxide through
dichloromethane solutions of the diene complexes 1–3 yields
carbonyl complexes [{IrL(CO)2}n][BF4]n (R = Me 4, Pri 5 or Ph
6) which are formed by displacing the diene. The elemental
analyses for 4 and 5 match the stoichiometry proposed.
Complex 6 could not be isolated.
With H2. Hydridoiridium diolefin complexes are intermedi-
ates in the homogeneous hydrogenation of olefins.16a For this
reason we believe that it is interesting to study the reactivity of
the olefinic dithioether complexes synthesized with H2.
The nuclearity of complexes 4 and 5 was established by
measuring their equivalent conductivity in acetone solutions at
different concentrations. Plots of the Onsager equation gave A
values around 890 which are characteristic of 2:1 electrolytes.19
The Fourier-transform-IR spectra of the dichloromethane
solutions of these tetracarbonyl complexes show three stretch-
ing frequencies υ(CO) in the 2100–1960 cmϪ1 region which are
characteristic of dinuclear tetracarbonyl complexes of IrI and
When H2 is bubbled for 30 mins at Ϫ70 ЊC and at atmos-
pheric pressure through CD2Cl2 solutions of olefinic iridium
complexes 1 and 2, dihydrido olefin complexes 7 and 8 are
formed in solution, Scheme 4. In the high-field region of the 1H
NMR spectrum of the CD2Cl2 solution of 7 at Ϫ70 ЊC two
metal hydride resonances at δ Ϫ12.88 and Ϫ13.18 which each
integrate for one proton can be distinguished; one of the reson-
ances is due to HA and the other to HB cis to each other and
trans to the cyclooctadiene and thioether. Two close metal
1
RhI.19c,21 The H NMR spectra of solutions of the complexes
prepared ‘in situ’ show signals which correspond to the co-
ordinated dithioether ligands and the non-co-ordinated
cyclooctadiene.
hydride signals can also be seen at very similar δ for the related
23
complex
cis-[IrH2(cod)(tht)2]ClO4
(tht = tetrahydrothio-
phene). This may be due to the fact that the cod and thioether
4614
J. Chem. Soc., Dalton Trans., 1997, Pages 4611–4618

View Article Online
Table 3 Hydrogenation results with catalytic systems 1–3a
H
CO2R
C
C
Ph
NHCOMe
Entry
Precursor Substrate t/h
Conversion (%) e.e. (%)
VI R = H
1
2
3
4
5
6
7
8
2
3
3
2
3
1
2
3
VI
16
16
48
16
12
24
6
96
99
37(R)
16(R)
13(R)b
11(S)
10(S)
22(S)b
47(S)b
6(S)b
VII R = Me
VI
VII
VIII
VIII
IX
IX
IX
50
91
CO2H
CO2H
CH2
C
H2C
C
NHCOMe
100
100
91
CH2CO2H
VIII
IX
4
100
ligands trans with respect to HA and HB have similar σ-donor
and π-acceptor properties.24
a At 20 ЊC and 1 atm H2. Solvent 6 cm3 CH2Cl2. Substrate:precur-
sor = 40:1. b Determined by polarimetry.
In a CD2Cl2 solution of complex 8 at Ϫ70 ЊC four signals of
two different intensities can be observed (relation 2:1) at δ
Ϫ13.13, Ϫ14.78 and Ϫ13.29 and Ϫ15.00 which may be from
two diastereomeric dihydridoiridium complexes. Two minor
signals of not identified products are also observed. The related
complex [Ir(cod)(diop)]PF6 also adds hydrogen easily to give a
perature under atmospheric pressure of H2. The three iridium
complexes 1–3 lead to active systems for the asymmetric hydro-
genation of prochiral olefins VI–IX. The best results are shown
in Table 3.
25
cis-dihydrido olefin complex cation [IrH2(cod)(diop)]PF6 but
For the hydrogenation of compound VI the most active sys-
tems are 2 and 3 (entries 1, 2) which have similar activity, but
the e.e. is higher (37% R) for precursor 1. All the systems lead to
the (R) enantiomer of N-acetylphenylalanine, as does the
rhodium–(R,R)-diop system.17 It is well known that polar solv-
ents can considerably affect the activity observed in the asym-
metric hydrogenation of prochiral olefins. However, for the
hydrogenation of VI using the catalytic system 1, experiments
in CH2Cl2 (99% conversion in 24 h) and MeOH (6% conversion
in 48 h) reveal that the activity in CH2Cl2 is higher.
The hydrogenation of compound VII with the catalytic sys-
tems 1–3, proceeds very slowly achieving conversions between
6 and 50% in 48 h with low enantioselectivity [11% (R) with 1
and 13% (R) with 3, entry 3]. The steric bulk of the methyl
group probably makes the co-ordination of the prochiral olefin
somewhat difficult. When 2 is used as the catalytic system the
very low activity could be attributed to the methyl group in the
olefin together with the isopropyl group in the catalytic system.
The previously described iridium system [Ir(cod)(nmdpp)-
(PhCN)]ϩ [nmdpp = (Ϫ)-neomenthyldiphenylphosphine] also
gives very low activity when hydrogenating VII.16b,c
Compound VIII was hydrogenated with catalytic systems 2
and 3 (entries 4, 5) at a comparable rate to VI but the enantio-
selectivity is lower. Compound IX is reduced with catalytic
systems 2 and 3 (entries 7, 8) considerably faster than the other
unsaturated compounds. For complex 2 a conversion of 91% is
achieved in 6 h together with an e.e. of 47% (S), which is the
highest e.e. obtained.
In general, although hydrogenation with complexes 1–3 takes
place under ambient conditions of pressure and temperature
the e.e. are low. The e.e. obtained with catalytic system 2 is
higher than those with 1 and 3, perhaps due to the greater
hindrance and rigidity of the Pri group. These catalytic systems
are more active but less enantioselective than related rhodium–
chiral thiolate systems previously reported5c–e for the asym-
metric hydrogenation of the prochiral olefins cited. However 1–
3 are more active, and 2 is also more enantioselective, than
related ruthenium–chiral sulfoxide systems for the hydrogen-
ation of itaconic acid.3
the possible diastereomers were not distinguished, just as they
were not for complex 7. In the low-field region of the 1H NMR
spectra for CD2Cl2 solutions of 7 and 8, there are signals cor-
responding to cyclooctadiene and dithioether ligands co-
ordinated in the hydridoiridium() complexes, together with a
small amount of starting material. When both complexes 7 and
8 are warmed hydrogen is partly lost and there is an increase in
the amount of parent complexes. These complexes can be
recovered in high yield by adding diethyl ether at room tem-
perature. This indicates that the dihydridoiridium complexes in
solution are in equilibrium with the parent complexes depend-
ing on the temperature, Scheme 4. The related olefinic dihy-
drido complexes [Ir(cod)L2]PF6 (L = PMePh2, PPh3, ¹Ph2PCH2-
¯
²
CH2PPh2, PBun3 or ¹diop) also have similar behaviour.25
¯
²
The nature of these hydrido ligands was confirmed by meas-
uring T1 using H relaxation rates.26a The hydride resonances
1
have T1 values of 211 and 288 ms for complex 7 and 180 and
200 and 178 and 227 ms for the major and minor diastereoi-
somers of 8 respectively, in CD2Cl2 at Ϫ70 ЊC and 300 MHz.
These values are consistent with classical hydride.26
When diethyl ether is added to CH2Cl2 solutions of cis-
[IrH2(cod){(Ϫ)-diosme}]BF4 7 at Ϫ70 ЊC a yellow powder is
obtained. The elemental analysis matches this stoichiometry.
The Fourier-transform-IR spectrum in KBr and Nujol
mull shows a broad signal at 2013 cmϪ1 which may include the
two asymmetric absorption bands, ν(Ir᎐H), expected for a cis-
dihydridoiridium() compound.27 The situation is similar
for the related complex cis-[IrH2(cod)(dth)]ClO4 (dth = 2,6-
dithiaheptane).23
As far as the reactivity of [Ir(cod){(ϩ)-diosph}]BF4 3 with
H2 is concerned, when this gas is bubbled through a CD2Cl2
solution for 30 min at Ϫ70 ЊC broad signals appear between
δ Ϫ12.5 and Ϫ16.5 in the high-field region of the 1H NMR spec-
trum. When the temperature is gradually increased from Ϫ70 to
15 ЊC the signal at δ Ϫ15.9 disappears and the other resonances
resolve into four narrower signals which may correspond to
different diastereomers of cis-[IrH2(cod){(ϩ)-diosph}]BF4
which are not rigid on the proton NMR time-scale. Likewise,
when it is warmed to 28 ЊC, hydrogen is partly lost, as it is for
complexes 7 and 8. The hydride resonances at Ϫ70, Ϫ40, 0 and
15 ЊC have T1 values in CD2Cl2 of around 300 ms at 300 Hz,
consistent with classical hydride.26
As is well known, hydrogen pressure can have a considerable
effect on the stereoselectivity of the asymmetric hydrogenation
of prochiral olefins.28 Studies at higher pressures are in progress.
Experimental
Catalytic activity
Asymmetric hydrogenation of prochiral olefins. To explore
how the iridium dithioether complexes 1–3 behave as catalyst
precursors, we initially studied the asymmetric hydrogenation
of prochiral olefins Z-α-(acetamido)cinnamic acid VI, methyl
α-(acetamido)cinnamate VII, α-(acetamido)acrylic acid VIII
and itaconic acid (methylenebutanedioic acid) IX at room tem-
Elemental analyses were carried out with a Carlo-Erba micro-
analyzer. Infrared spectra were recorded on a Midac Grams/
1
386 spectrophotometer, H and 13C NMR spectra on a Varian
Gemini 300 MHz spectrometer. Proton T1 studies were per-
formed using the standard inversion recovery 180Њ–τ–90Њ pulse
sequence.26a Fast atom bombardment mass spectrometry was
J. Chem. Soc., Dalton Trans., 1997, Pages 4611–4618
4615

View Article Online
performed on a VG autospect in a 3-nitrobenzyl alcohol matrix.
Conductivities were measured in acetone solutions at several
concentrations in the range 10Ϫ3–10Ϫ5 , with a Philips PW9509
conductimeter. Gas chromatography analyses were performed
with a Hewlett-Packard 5890A instrument (fused-silica capil-
lary column 25 m × 0.25 mm permabond L-Chirasil-Val).
Optical rotations were determined with a Perkin-Elmer 241 MC
polarimeter at the indicated temperature. The specific rotation
is given in deg cm3 gϪ1 dmϪ1 units. All synthesis of iridium
complexes and dithioethers were carried out under nitrogen
using standard Schlenk techniques. Solvents were distilled and
deoxygenated before use. The indium compounds [{Ir(µ-
2.84; S, 15.04%); δH(300 MHz, CDCl3, SiMe4) 1.46 (6 H, s,
CMe2), 4.23 (2 H, m, CH) and 4.55 (4 H, m, CH2); δC(74.5
MHz, CDCl3) 26.2 (CMe2), 73.5 (CH2), 74.1 (CH), 111.9
(CMe2) and 118.6 (q, CF3, JCF = 320.2 Hz).
[Ir(cod){(؊)-diosme}]BF4 1. The compound was prepared by
the following three routes.
(i) Addition of an excess of (Ϫ)-diosme (23 mg, 0.11 mmol)
to a dichloromethane solution (3 cm3) of [Ir(cod)2]BF4 (40 mg,
0.08 mmol) produced an immediate colour change. Subsequent
addition of ether precipitated the desired complex, which was
filtered off, washed with cold ether and vacuum dried (34.3 mg,
70%).
22
Cl)(cod)}2],29 [Ir(cod)2]BF4 and [{Ir(µ-OMe)(cod)}2]30 were
prepared by the general procedures described.
(ii) Adding an excess of (Ϫ)-diosme (40 mg, 0.18 mmol) and
the stoichiometric amount of AgBF4 (23 mg, 0.12 mmol) to a
dichloromethane solution (3 cm3) of [{Ir(µ-Cl)(cod)}2] (40 mg,
0.05 mmol) produced a white precipitate of silver chloride,
which was filtered off through Kieselguhr. Addition of ether to
the filtrate precipitated the required complex, which was filtered
off, washed with cold ether and vacuum dried (38.0 mg, 50%).
(iii) An acetone solution (10 cm3) of [Ir(cod)(Me2CO)2]BF4
was prepared by treating [{Ir(µ-OMe)(cod)}2] (66 mg, 0.10
mmol) with a solution of tetrafluorobic acid (54%) in diethyl
ether (60 cm3, 0.20 mmol). The resulting solution was stirred for
30 min and was added to a solution of (Ϫ)-diosme (44 mg, 0.22
mmol) in acetone (5 cm3). The orange solution formed was
stirred for 30 min. The desired complex was precipitated by
adding ether and then filtered off, washed with cold ether and
vacuum dried, (79.0 mg, 81%) (Found: C, 32.98; H, 4.98; S,
10.35. Calc. for C17H30BF4IrO2S2: C, 33.49; H, 4.96; S, 10.52%);
m/z 523 (Mϩ).
Syntheses
diosme. A suspension of NaH (2.30 g, 95 mmol) in paraffin,
cleaned twice in hexane, in thf (14 cm3) was cooled to Ϫ78 ЊC
and methanethiol (2 cm3, 36 mmol) at Ϫ78 ЊC was added. The
resulting solution was stirred and the temperature increased to
0 ЊC. After 5 min the solution was cooled and a solution of
compound V (1 g, 2 mmol) in thf (9 cm3) added. After 45 min
the solvent was evaporated and water (100 cm3) was added to
the residue which was extracted with dichloromethane (3 × 50
cm3). The extract was then dried and concentrated. The residue
was purified by column chromatography (hexane–ethyl acetate
5:1) and the required compound was obtained (0.44 g, 85%) as
a colourless liquid: [α]2D3 = Ϫ6.06 (c 3.2 in CHCl3) (Found: C,
49.00; H, 8.06; S, 2.91. Calc. for C9H18O2S2: C, 48.60; H, 8.16; S,
2.88%).
diospr. A solution of propane-2-thiol (0.2 cm3, 2.30 mmol) in
thf (2 cm3) was added to a suspension of NaH (78 mg, 3.25
mmol) in paraffin, cleaned twice in hexane, in thf (1 cm3). The
resulting solution was stirred for 20 min. Then a solution of
compound V (330 mg, 0.77 mmol) in thf (3 cm3) was added.
After 45 min the solvent was evaporated and water (50 cm3) was
added to the residue which was extracted with dichloromethane
(3 × 50 cm3). The extract was then dried and concentrated. The
residue was purified by column chromatography (hexane–ethyl
acetate 20:1) and the required compound was obtained (162
mg, 77%) as a colourless liquid: [α]2D3 = Ϫ7.58 (c 3.2 in CHCl3)
(Found: C, 55.51; H, 9.9; S, 22.49. Calc. for C13H26O2S2: C,
56.07; H, 9.4; S, 23.02%).
[Ir(cod){(؊)-diospr}]BF4ؒCH2Cl2 2. The compound was
prepared by the following two routes.
(i) Addition of an excess of (Ϫ)-diospr (33 mg, 0.12 mmol) to
a dichloromethane solution (3 cm3) of [Ir(cod)2]BF4 (40 mg,
0.08 mmol) produced an immediate colour change. Subsequent
addition of ether precipitated the desired complex, which was
filtered off, washed with cold ether and vacuum dried (42.0 mg,
79%).
(ii) Addition of an excess of (Ϫ)-diospr (54 mg, 0.19 mmol)
and the stoichiometric amount of AgBF4 (23 mg, 0.12 mmol) to
a dichloromethane solution (3 cm3) of [{Ir(µ-Cl)(cod)}2] (40 mg,
0.05 mmol) produced a white precipitate of silver chloride,
which was filtered off through Kieselguhr. Addition of ether to
the filtrate precipitated the required complex which was filtered
off, washed with cold ether and vacuum dried (43.0 mg, 54%)
(Found: C, 35.46; H, 5.25; S, 8.65. Calc. for C21H38BF4IrO2S2:
C, 35.69; H, 5.44; S, 8.66%); m/z 579 (Mϩ).
diosph. Benzenethiol (3.3 cm3, 32.3 mmol) was added to a
solution of NaOH (1.27 g, 32 mmol) in water (12 cm3). The
resulting solution was stirred for 2 h. Then a solution of
compound IV (5 g, 10.63 mmol) in dimethylformamide was
added. The resulting solution was stirred under reflux for 24 h.
The solvent was evaporated and water (100 cm3) was added to
the residue which was extracted with diethyl ether (3 × 50 cm3).
The extract was then dried and concentrated. The residue was
purified by column chromatography (hexane–ethyl acetate
20:1) and the required compound was obtained (3.4 g, 92%) as
a white solid: [α]2D3 = ϩ45 (c 3.2 in CHCl3) (Found: C, 65.39; H,
6.44; S, 18.00. Calc. for C19H22O2S2: C, 65.86; H, 6.39; S,
18.50%).
[Ir(cod){(؉)-diosph}]BF4 3. An excess of (ϩ)-diosph (41 mg,
0.12 mmol) was added to a dichloromethane solution (3 cm3) of
[Ir(cod)2]BF4 (40 mg, 0.08 mmol). Then hydrogen was bubbled
through for 15 min producing a change in colour. Subsequent
addition of ether precipitated the desired complex, which was
filtered off, washed with cold ether and vacuum dried (40.0 mg,
67%) (Found: C, 32.98; H, 4.98; S, 10.35. Calc. for C27H34BF4-
IrO2S2: C, 33.49; H, 4.96; S, 10.52%); m/z 647 (Mϩ).
Compound V. Pyridine (0.76 cm3, 9.5 mmol) was added to a
solution of compound III (0.57 g, 3.5 mmol) in dichlorometh-
ane (21 cm3). The resulting solution was stirred for 10 min.
Then it was cooled to Ϫ20 ЊC and triflic anhydride (CF3SO2)2O
(1.4 cm3, 8.3 mmol) was slowly added. After 25 min TLC
(hexane–ethyl acetate 3:2) showed that the reaction was com-
plete. The solvents were evaporated under vacuum and the resi-
due was purified by column chromatography (hexane–ethyl
acetate 5:1) and the required compound was obtained (1.68 g,
81%) as a white solid: [α]2D3 = Ϫ8.13 (c 3.2 in CHCl3) (Found:
C, 24.86; H, 2.91; S, 15.53. Calc. for C9H12F6O8S2: C, 25.36; H,
[Ir2{(ì-(؊)-diosme}2(CO)4][BF4]2 4. Carbon monoxide was
bubbled through a dichloromethane solution (6 cm3) of the
complex [Ir(cod){(Ϫ)-diosme}]BF4 (40 mg, 0.065 mmol). After
5 min the starting solution lightened. Cold ether was added to
give the desired compound, which was filtered off, washed with
cold ether and vacuum dried (13.7 mg, 60%) (Found: C, 24.20;
H, 3.40; S, 10.70. Calc. for C11H18BF4IrO4S2: C, 23.70; H, 3.25;
S, 11.50%); νmax/cmϪ1 (CO) 2067s, 2021s and 1997s.
[Ir2{ì-(؊)-diospr}2(CO)4][BF4]2 5. Carbon monoxide was
bubbled through a dichloromethane solution (6 cm3) of the
4616
J. Chem. Soc., Dalton Trans., 1997, Pages 4611–4618

View Article Online
complex [Ir(cod){(Ϫ)-diospr}]BF4 (40 mg, 0.060 mmol). After
5 min the starting solution lightened. Addition of cold ether
gave the desired compound, which was filtered off, washed with
cold ether and vacuum dried (28.1 mg, 61%) (Found: C, 26.02;
H, 3.69; S, 8.96. Calc. for C15H26BF4IrO4S2ؒ0.5CH2Cl2: C,
25.79; H, 3.75; S, 9.18%); νmax/cmϪ1 (CO) 2079s, 2021s and
1996s.
References
1 C. S. Chin and B. Lee, J. Chem. Soc., Dalton Trans., 1991, 1323 and
refs. therein.
2 C. Claver and A. M. Masdeu, Trends Organomet. Chem., 1994, 1,
549 and refs. therein.
3 B. R. James and R. S. McMillan, Can. J. Chem., 1977, 55, 3927.
4 J. Solé, C. Bo, C. Claver and A. Ruiz, J. Mol. Catal., 1990, 61, 163;
C. K. Lai, A. A. Naiini and C. H. Brubaker, jun., Inorg. Chim. Acta,
1989, 164, 205; A. A. Naiini, H. M. Ali and C. H. Brubaker, jun.,
J. Mol. Catal., 1991, 67, 47 and refs. therein.
5 (a) Ph. Kalck, R. Poilblanc, R. P. Martin, A. Rovera and A. Gaset,
J. Organomet. Chem., 1980, 195, C9; (b) H. Schumann, H. Hemling,
N. Goren and J. Blum, J. Organomet. Chem., 1995, 485, 209; (c) M.
Eisen, P. Weitz, S. Shtelzer, J. Blum, H. Schumann, B. Gorella and
F. H. Görlitz, Inorg. Chim. Acta, 1991, 188, 167; (d) H. Schumann,
B. Gorella, M. Eisen and J. Blum, J. Organomet. Chem., 1991, 412,
251; (e) M. Eisen, J. Blum, H. Schumann and B. Gorella, J. Mol.
Catal., 1989, 56, 329.
6 (a) Ph. Kalck, in Organometallics in Organic Synthesis, eds. A. de
Meijere and H. tom Dieck, Springer, Heidelberg, 1987, pp. 297–320;
(b) S. Gladiali, J. C. Bayón and C. Claver, Tetrahedron Asymmetry,
1995, 6, 1453 and refs. therein; (c) F. Agbossou, J. F. Carpentier and
A. Mortreux, Chem. Rev., 1995, 85, 2485; (d) C. Claver, S. Castillón,
N. Ruiz, G. Delogu, D. Fabbri and S. Gladiali, J. Chem. Soc., Chem.
Commun., 1993, 1833; (e) A. M. Masdeu, A. Orejón, A. Ruiz,
S. Castillón and C. Claver, J. Mol. Catal., 1994, 94, 149; ( f )
A. Castellanos-Paéz, S. Castillón and C. Claver, J. Organomet.
Chem., 1997, 539, 1; (g) A. Aaliti, N. Ruiz, J. Forniés, A. Ruiz,
C. Claver, C. J. Cardin, D. Fabbri and S. Gladiali, J. Organomet.
Chem., in the press.
[Ir2{(ì-(؉)-diosph}2(CO)4][BF4]2 6. Carbon monoxide was
bubbled through a dichloromethane solution (6 cm3) of the
complex [Ir(cod){(ϩ)-diosph}]BF4 (40 mg, 0.055 mmol). After
5 min the starting solution lightened. The final compound
could not be isolated. νmax/cmϪ1 (CO) 2054m, 2020s and
1994s.
[IrH2(cod){(؊)-diosme}]BF4 7. Hydrogen was bubbled
through a brown-orange solution of [Ir(cod){(Ϫ)-diosme}]BF4
(40 mg, 0.065 mmol) in CD2Cl2 (0.4 cm3) at Ϫ70 ЊC for 30 min.
Addition of diethyl ether at Ϫ70 ЊC gave a yellow powder
(Found: C, 33.38; H, 5.26; S, 10.26. Calc. for C17H32BF4IrO2S2:
C, 33.39; H, 5.27; S, 10.48%); δH(300 MHz, CD2Cl2, Ϫ70 ЊC)
Ϫ12.88 (1 H, s) and Ϫ13.18 (1 H, s); νmax/cmϪ1 (Ir᎐H) 2013 (br).
[IrH2(cod){(؊)-diospr}]BF4 8. Hydrogen was bubbled
through a brown-orange solution of [Ir(cod){(Ϫ)-diospr}]BF4
(40 mg, 0.060 mmol) in CD2Cl2 (0.4 cm3) at Ϫ70 ЊC for 30 min.
The solution was then transferred to an NMR spectrometer
7 F. Fache, P. Gamez, F. Nour and M. Lemaire, J. Mol. Catal., 1993,
85, 131.
1
1
tube and the H NMR spectrum was recorded (see text for H
NMR data and characterisation).
8 H. B. Kagan, in Comprehensive Organometallic Chemistry, eds.
G. Wilkinson, F. G. A. Stone and E. W. Abel, Pergamon, Oxford,
1982, vol. 8, ch. 53; H. Takaya, T. Onta and R. Noyori, in Catalytic
Asymmetric Synthesis, ed. I. Ojima, VCH, New York, 1993, ch. 1;
R. Noyori, in Asymmetric Catalysis in Organic Synthesis, Wiley,
New York, 1994, ch. 2, p. 16; J. M. Brown, Chem. Soc. Rev., 1993, 25;
H. Brunner, in Top. Stereochem., 1988, 18, 129; J. M. Brown, in
Insights into Speciality Inorganic Chemicals, ed. D. Thompson, The
Royal Society of Chemistry, Cambridge, 1995, ch. 6.
9 (a) S. Akutagawa, in Asymmetric Hydrogenation with Ru᎐BINAP:
Chirality in Industry, eds. A. N. Collins, G. N. Sheldrake and
J. Crosby, Wiley, Chichester, 1992, p. 325; (b) S. Akutagawa, in
Asymmetric Hydrogenation with Ru᎐BINAP: Catalysis of Organic
Reactions, eds. M. Scaros and M. L. Prunier, Marcel Dekker, New
York, 1994, p. 135; (c) S. Akutagawa, Appl. Catal. A, Gen., 1995,
128, 171.
Catalytic hydrogenations
The reactions under 1 atm of H2 were performed in a previously
described hydrogen-vacuum line.31 In a typical run, substrate
(100 mg) and catalyst precursor, dissolved in dichloromethane
(6 cm3), were shaken under H2 (1 atm) at 293 K. After the
required time the solvent was removed. The extent of conver-
sion was measured by 1H NMR spectroscopy.
Work-up of the hydrogenation product. The following pro-
cedures was used to isolate the hydrogenation product. A; for
N-acetylalanine, the residue was dissolved in water and separ-
ated from the insoluble catalyst by filtration. Evaporation to
dryness afforded the product. B; for methylsuccinic acid, N-
acetylphenylalanine and N-acetylphenylalanine methyl ester,
the residue was dissolved in 0.5  NaOH and separated from
the insoluble catalyst by filtration. The filtrate was acidified
with dilute HCl, extracted with ether, and washed with a little
water. The ether phase was dried over sodium sulfate and evap-
orated to dryness. C; for N-acetylphenylalanine, N-acetylphen-
ylanine methyl ester and N-acetylalanine, gas chromatography
analyses were performed with a Hewlett-Packard 5890A
instrument (fused-silica capillary column 25 m × 0.25 mm,
permabond L-Chirasil-Val) before treating the sample as
described.32 A 0.5 g amount of the residue was heated for 1 h at
100 ЊC with 6  HCl (10 cm3). Then the solvent was evaporated
and PriOH (10 cm3) in 6% HCl was added. The resulting sol-
ution was stirred at 90 ЊC during 1.5 h. The reagent was evapor-
ated and the residue dissolved in dichloromethane (2.5 cm3)
and pentafluoropropionic anhydride (0.3 cm3). This solution
was stirred for 1 h at room temperature. Then the solvent was
evaporated and the residue dissolved in acetone (0.3 cm3) and
analysed by gas chromatography.
10 Asymmetric Synthesis, ed. J. D. Morrison, Academic Press, New
York, 1985, vol. 5; R. Noyori, Science, 1990, 245, 1194; W. S.
Knowles, Acc. Chem. Res., 1983, 16, 106; Asymmetric Catalysis,
Maryinus Nijhoff, Boston, MA, 1984; J. M. Brown, Nature
(London), 1991, 350, 191.
11 W. S. Knowles, J. Chem. Educ., 1986, 63, 222.
12 W. A. Nugent, T. V. RajanBabu and M. J. Burk, Science, 1993, 259,
479; G. M. Ramos Tombo and D. Bellus, Angew Chem., Int. Ed.
Engl., 1991, 30, 1193; J. Alberecht and U. Nagel, Angew. Chem., Int.
Ed. Engl., 1996, 35, 407.
13 M. J. Burk, J. E. Feaster and R. L. Harlow, Organometallics, 1990, 9,
2653; M. J. Burk, J. E. Feaster, W. A. Nugent and R. L. Harlow,
J. Am. Chem. Soc., 1993, 115, 10 125.
14 T. V. RajanBabu, T. A. Ayers and A. L. Caslnuovo, J. Am. Chem.
Soc., 1994, 116, 4101.
15 T. Ohta, H. Takaya, M. Kitamura, K. Nagai and R. Noyori, J. Org.
Chem., 1987, 52, 3174.
16 (a) P. A. Chaloner, M. A. Esteruelas, F. Joó and L. A. Oro, in
Homogeneous Hydrogenation, Kluwer, Dordrecht, 1994, ch. 4; (b)
L. A. Oro, J. A. Cabeza, C. Cativiela, M. D. Díaz de Villegas and
E. Meléndez, J. Chem. Soc., Chem. Commun., 1983, 1383; (c) J. A.
Cabeza, C. Cativiela, M. D. Díaz de Villegas and L. A. Oro,
J. Chem. Soc., Perkin Trans. 1, 1988, 1881.
17 H. B. Kagan and T. P. Dang, J. Am. Chem. Soc., 1972, 94,
6429.
18 M. Carmack and C. J. Kelley, J. Org. Chem., 1969, 33, 2171.
19 R. D. Feltham and R. G. Hayter, J. Chem. Soc., 1964, 4587;
R. Uson, J. Gimeno, J. Fornies and F. Martínez, Inorg. Chim. Acta,
1981, 50, 173; A. Ruiz, C. Claver, J. C. Rodríguez, M. Aguiló,
X. Solans and M. Font-Altaba, J. Chem. Soc., Dalton Trans., 1984,
2665.
Acknowledgements
We thank the Ministerio de Educación y Ciencia and the Gen-
eralitat de Catalunya for financial support (QFN-95-4725-C03-
2; CICYT-CIRIT).
20 W. J. Hälg, L. R. Öhrström, H. Rüeggh, L. M. Venanzi, T. Gerfin
and V. Gramlich, Helv. Chim. Acta, 1993, 76, 788.
J. Chem. Soc., Dalton Trans., 1997, Pages 4611–4618
4617

View Article Online
27 K. Nakamoto, in Infrared and Raman Spectra of Inorganic and
Coordination Compounds, Wiley, New York, 1978.
28 I. Ojima, T. Kogure and N. Yoda, J. Org. Chem., 1980, 45, 4728.
29 J. L. Herde, J. C. Lambert and C. V. Senoff, Inorg. Synth., 1974, 15,
18.
30 J. Chatt and J. M. Davidson, J. Chem. Soc. A, 1965, 843.
31 C. Cativiela, J. Fernandez, J. A. Mayoral, E. Melendez, R. Usón,
L. A. Oro and M. J. Fernandez, J. Mol. Catal., 1992, 16, 19.
32 H. Frank, C. J. Nickolson and E. Bayer, J. Chromatogr. Sci., 1977,
15, 174.
21 Ph. Kalck and R. Poilblanc, Inorg. Chem., 1975, 14, 2779; J. C.
Rodriguez, A. Ruiz and C. Claver, Transition Met. Chem., 1984, 9,
237.
22 M. Green, T. A. Kuc and H. Taylor, J. Chem. Soc. A, 1971, 2334.
23 J. C. Rodriguez, C. Claver and A. Ruiz, J. Organomet Chem., 1985,
293, 115.
24 D. Sellmann, H.-P. Neumer and F. Knoch, Inorg. Chim. Acta, 1991,
190, 61.
25 R. H. Crabtree, H. Felkin, T. Fillebeen-Khan and G. E. Morris,
J. Organomet. Chem., 1979, 168, 183.
26 (a) D. G. Hamilton and R. H. Crabtree, J. Am. Chem. Soc., 1988,
110, 4126; (b) J. C. Lee, jun., W. Yao and R. H. Crabtree, Inorg.
Chem., 1996, 35, 695.
Received 1st July 1997; Paper 7/04611A
4618
J. Chem. Soc., Dalton Trans., 1997, Pages 4611–4618