A new roate to cationic half-sandwich ruthenium(II) complexes with chiral cyclopentadienylphosphane ligands

Article abstract of DOI:10.1002/ejic.200300934

Optically active cyclopentadieny-linked phosphane ligands L_n of general formula C₅H₅CH₂CH(R) PR′₂ (R = Ph, R′ = Ph, L₁; R = Ph, R′ = Cy, L₂; R = mesityl, R′ = Ph, L₃)

Full text of DOI:10.1002/ejic.200300934

FULL PAPER
A New Route to Cationic Half-Sandwich Ruthenium(II) Complexes with Chiral
Cyclopentadienylphosphane Ligands
Angelino Doppiu[a] and Albrecht Salzer*[a]
Keywords: Asymmetric catalysis / Constrained geometry ligands / Cyclopentadienyl ligands / P ligands / Ruthenium
Optically active cyclopentadienyl-linked phosphane ligands
the metal through the cyclopentadienyl and the phosphane
Ln of general formula C5H5CH2CH(R)PRЈ2 (R = Ph, RЈ = Ph, moieties forming a chelate, in almost quantitative yields. The
L1; R = Ph, RЈ = Cy, L2; R = mesityl, RЈ = Ph, L3) and
C5H5CH(R)CH2PPh2 (R = Cy, L4; R = Bzl, L5) are synthesised
by ring-opening reactions of 1-R-spiro[2.4]hepta-4,6-dienes
(R = Ph, Cy, mesityl, benzyl), with lithium phosphides LiPRЈ2
complexes are active catalysts in the redox isomerisation of
geraniol affording chiral citronellal but with low enantiose-
lectivity. They also catalyze an aldol-type C−C coupling reac-
tion between 3-buten-2-ol and benzaldehyde affording al-
(Ar = Ph, Cy). Reaction of the ligands Ln with [RuH(η5- most racemic 4-hydroxy-3-methyl-4-phenyl-2-butanone with
C7H11)2][BF4] in acetonitrile at reflux affords the bis(acetoni-
good diastereoselectivity (syn/anti ratio up to 77:23).
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
trile) ruthenium(II
)
complexes [Ru(η5-C5H4(C2H3R)PRЈ2-
κP)(CH3CN)2][BF4], in which the ligands are coordinated to
Introduction
compound have been reported by Trost.[7] Kirchner has
shown that substitution of one acetonitrile ligand in
[Ru(Cp)(AN)3]ϩ
[Ru(Cp)(AN)2(PR3)]ϩ
for
tertiary
phosphanes
PR3
Organometallic ruthenium complexes are becoming in-
creasingly important in homogeneous catalyses such as re-
duction, oxidation, isomerisation, and carbonϪcarbon
bond-formation.[1] The preparation of the necessary ru-
thenium catalysts most often starts from RuCl3·nH2O,
[8a]
also gave a very active catalyst
for the redox isomerisation of allylic alcohols,[8b] and CϪC
bond-formation.[8cϪ8l] Various attempts have been made to
link the Cp and phosphorus moieties in such compounds
through a spacer. In almost all cases, these complexes con-
tain a PPh3 ligand, as the synthesis, starting from
[RuCl2(PPh3)3], gives no alternate choice.[9] Only in one case
could bis(acetonitrile) complexes be isolated by Takahashi
in which the linker connects through an ester group at the
cyclopentadienyl ligand.[10] These complexes were tested in
catalytic allylic amination and alkylation with excellent re-
sults.[11] Prompted by the results of Kirchner and Takah-
ashi, we endeavoured to develop a synthetic route to
[Ru(CpϪP)(AN)2]ϩ-type chiral complexes containing a
phosphane ligand tethered onto the cyclopentadienyl ring
via a two-carbon linker (CpϪP). We have recently reported
a route to such complexes,[12] and we have shown, in collab-
oration with Kirchner, that they react with alkynes dis-
playing unusual reactivity.[13] Here we report on an im-
proved method that affords the desired optically active ru-
thenium() complexes in almost quantitative yields, as well
as a preliminary account of their application in catalytic
organic transformations.
[Ru(H2O)6]2ϩ
,
[RuCl2(PPh3)3],
[RuCl2(COD)]n,
or
[RuCl2(cymene)]n.[2] Analogously, the chemistry of cyclo-
pentadienylruthenium complexes is based on few starting
materials, mainly on the readily available precursors
[Ru(Cp)Cl(PPh3)2] and [Ru(Cp)Cl(CO)2].[2] However, the
selective replacement of either PPh3 or CO has proved diffi-
cult, limiting the synthetic utility. Coordinatively unsatu-
rated complexes and complexes with labile ligands are the
most promising candidates for catalytic applications, since
the availability of a vacant coordination site is a prerequisite
for both stoichiometric and catalytic organic transform-
ations at the transition metal centre. Therefore, more
labile systems, including [Ru(Cp)(CH3CN)3]ϩ,[3] and
[Ru(Cp)Cl(COD)],[4] have been developed. In particular, the
cationic complex [Ru(Cp)(CH3CN)3]ϩ is a promising versa-
tile intermediate since the acetonitrile ligands (AN) are sub-
stitutionally labile and can be replaced by other ligands.[5,6]
This complex can be regarded as a source of the highly
active coordinatively unsaturated fragment [Ru(Cp)]ϩ,
which may well have interesting potential in catalysis. In-
deed, several CϪC bond-forming reactions catalysed by this
Results and Discussion
[a]
Synthesis of the Ligands
Institut für Anorganische Chemie der RWTH Aachen,
Professor Pirlet Strasse 1, 52056, Aachen, Germany
Fax: (internat.) ϩ49-(0)241/809-2288
When we started our research, there was a dearth of gen-
eral procedures to prepare chiral CpϪP ligands by a
E-mail: albrecht.salzer@ac.rwth-aachen.de
2244
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejic.200300934
Eur. J. Inorg. Chem. 2004, 2244Ϫ2252

A New Route to Cationic Half-Sandwich Ruthenium(II) Complexes
FULL PAPER
method that allows their modular construction by indepen-
dent variation of the three components, namely the cyclo-
pentadienyl, the linker, and the phosphane.[9] We have de-
veloped such a route by adapting the method first reported
by Kauffmann,[14] which is based on the ring-opening re-
actions of spiro[2.4]hepta-4,6-dienes.[15] Traditionally, the
spiro compounds have been synthesised by bis-alkylation of
cyclopentadiene with the corresponding dibromoalkane in
the presence of sodium amide or sodium hydride.[16] How-
ever, a more convenient approach to introduce chirality into
these spiroannulated precursors is the use of the corre-
sponding bis(mesylate)s or bis(tosylate)s as alkylating re-
agents. Rieger has used this route to synthesise ethyl-
bridged bis(cyclopentadienyl) ligands starting from styrene
epoxide.[17] The same approach was recently adopted by
Whitby for the synthesis of chiral indenyl-linked phosphane
ligands.[18] Our synthetic strategy is based on the use of chi-
ral vicinal diols as starting materials, as depicted in
Scheme 1. The chiral diols can be obtained in high enanti-
omeric purity by several methods, for example Sharpless
asymmetric dihydroxylation,[19] Jacobsen hydrolytic kinetic
resolution of epoxides,[20] and biological means. They are
easily converted into the bis(methanesulfonate) esters in al-
most quantitative yield by reaction with methanesulfonyl
chloride. Displacement of the methanesulfonate groups by
excess cyclopentadiene/sodium amide affords the spiroan-
nulated dienes in excellent yields (79Ϫ95%). The formation
of the spiro compound involves first a nucleophilic substi-
tution of one mesylate group by the cyclopentadienyl anion,
followed by a fast deprotonation of the resulting substituted
cyclopentadiene, and finally a second nucleophilic substi-
tution of the other mesylate group and ring closure. The
nucleophilic substitution at the chiral carbon is entirely
stereoselective and proceeds with complete inversion of
configuration at the stereogenic centre.[15] The chiral spiro
compounds that have been synthesised by this methodology
are shown in Scheme 2.
Scheme 2
vated benzylic position of the cyclopropane ring affording
the ligands L1Ϫ3. Interestingly, this regioselectivity is the
opposite to that found for the nucleophilic opening of phe-
nyl epoxide and phenyl episulfide with lithium phosphides;
in that context, the attack of the nucleophile proceeds re-
gioselectively at the less-hindered site.[21] The ring-opening
reaction proceeds with complete inversion of the configura-
tion at the stereogenic centre, as is demonstrated by X-ray
crystal structure determinations of some derived metal
complexes.[12,15] The opposite regioselectivity is observed in
the ring-opening reaction of (R)-1-benzylspiro[2.4]hepta-
4,6-diene and (S)-1-cyclohexylspiro[2.4]hepta-4,6-diene,
which lack the activated benzylic position in the cyclopro-
pane ring. In this case, the nucleophilic attack with lithium
diphenylphosphide occurs on the unsubstituted carbon
atom of the cyclopropane ring, affording the ligands L4 and
L5.[22] The same regioselectivity has been observed in the
ring-opening of 2-cyclohexylspiro(cyclopropane-1,1Ј-in-
dene).[18] For convenience of storage, the lithium salts of the
ligands can be reacted with H2O to afford the cyclopen-
tadienyl(alkyl)phosphanes as a mixture of two positional
isomers depending on to which olefinic carbon atom of the
Cp the linker is attached. The ligands are shown in
Scheme 3. The overall yields for three steps (starting from
the diols) are excellent (63 to 88%).
Scheme 3
Synthesis of the Ruthenium Complexes
We have recently reported that reaction of [RuCl2(η3:η3-
C10H16)]2 with the Cp-linked phosphane ligand L1 in a 1:2
ratio in a mixture of acetonitrile and ethanol at room tem-
perature, in the presence of Li2CO3 and KPF6, affords
Scheme 1
The ring-opening reaction of (S)-1-R-spiro[2.4]hepta-4,6- the cationic complex [Ru(η5-C5H4CH2CH(Ph)PPh2-
dienes (R ϭ phenyl, mesityl) with lithium diphenylphos- κP)(AN)2][PF6] in 70Ϫ75% yield.[12] However, apart from
phide or lithium dicyclohexylphosphide proceeds with com- the desired compound, side products were also formed that
plete regiospecificity by attack of the nucleophile at the acti- have not been identified yet. The purification of the desired
Eur. J. Inorg. Chem. 2004, 2244Ϫ2252
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2245

A. Doppiu, A. Salzer
FULL PAPER
complex was accomplished by a cumbersome crystallisation the reaction is carried out on a gram scale, the evolution of
step. Therefore, we sought for a better synthetic route. A the mixture can easily be followed by the colour changes
paper from Cox and Roulet provided inspiration for a new of the solution. Initially, the mixture is brown-yellow and
method.[23] The paper reports: ‘‘Dissolution of complex 1 becomes immediately yellow upon addition of the ligand.
in deoxygenated acetonitrile leads to rapid displacement of Upon refluxing, a red-orange colour appears and lasts for
one molar equiv. of 2,4-dimethylpenta-1,3-diene and forma- a few minutes, eventually converting to an orange-yellow
tion of the cation [Ru(η5-C7H11)(AN)3]ϩ’’. [Bis(η5-2,4-di- colour.
methylpentadienyl)hydridoruthenium] tetrafluoroborate (1)
is known to show a fluxional behaviour, as the hydrido li-
gand is involved in three-centre RuϪHϪC agostic interac-
tions with all terminal methylene groups of the dimethylpe-
ntadienyl ligands (Scheme 4).[24]
Scheme 4
Scheme 6
The species 1a, which is in equilibrium with 1, is highly
reactive towards two-electron ligands, and the η4-C7H12 li-
gand is substitutionally labile, as it is easily replaced by
acetonitrile. We performed the reaction between L1 and
[Ru(η5-C7H11)2H][BF4], and we were delighted to obtain
the pure bis(acetonitrile) complex Ru1 in almost quantitat-
ive yield (Scheme 5).
According to this synthetic procedure, the ruthenium
complexes depicted in Scheme 7 could be obtained in al-
most quantitative yields. They were characterised by ana-
lytical and spectroscopic methods. The phosphorus res-
onates at low field in Ru1؊3 (δ ഠ 88 ppm); the same signal
appears at higher field in Ru4 and Ru5 (δ ഠ 52 ppm). In the
free ligands, the phosphorus shows two signals in the 31P
NMR spectrum due to the two regioisomers at about δ ϭ
0.5 (L1), 18 (L2), Ϫ6 (L3), Ϫ18 (L4), and Ϫ20 ppm (L5). The
large downfield shift upon coordination is characteristic of
the formation of a five-membered chelate ring.[25] In the 1H
NMR spectra, the signals of the CϪH hydrogen of the
linker are low-field shifted in Ru1؊3 (δ ϭ 4.66, 4.19, and
5.42 ppm respectively); for comparison, the hydrogen res-
onates at δ ϭ 1.94 and 2.55 ppm for Ru4 and Ru5 respec-
tively. The coordinated acetonitrile ligands give rise to two
doublets with a long range coupling (5J) with phosphorus
of about 1 Hz. Easy replacement of the acetonitrile ligands
Scheme 5
A possible reaction mechanism that accounts for the for-
mation of Ru1 is presented in Scheme 6. At first, 2 may
be formed, with subsequent coordination of phosphorus to
afford 3. The reaction has been followed by means of a 31P
NMR experiment, revealing the formation of intermediate
species with 31P resonances at about δ ϭ 30 ppm. The coor-
dination of the phosphorus atom seems to be very fast, as
no signals of the free ligands could be observed (δ ϭ 0.4
and 0.9 ppm). No other intermediates could be detected,
and the species convert into the final complex, which shows
1
with CD3CN was confirmed by means of H NMR spec-
troscopy, suggesting that they are substitutionally labile.
Catalytic Applications
With several chiral complexes in hand, we began a pre-
a singlet at δ ϭ 88.7 ppm. The evolution of 3 to 4 involves liminary screening of their catalytic activity in organic
the intramolecular coordination of the cyclopentadiene transformations. Kirchner has reported that
group, which is favoured on thermodynamic grounds by the [Ru(Cp)(AN)2(PPh3)]ϩ, in which the phosphane is not con-
formation of a chelate species. At this stage, an intramol- nected to the Cp, is a highly active catalyst at 57 °C for the
ecular H-transfer between the cyclopentadiene and the di- redox isomerisation of allylic alcohols to give the corre-
methylpentadienyl group may be postulated. It may occur sponding carbonyl compounds.[8b] We tested Ru1 in this re-
with participation of the metal through a hydrido species. action and, in line with his report, we found that it catalyses
The species 5 thus formed contains the labile η4-dimeth- the isomerisation of allylic alcohols (R ϭ H, Me, Ph) at
ylpentadiene ligand that is replaced by acetonitrile afford- room temperature [Equation (1)]; details are given in the
ing the desired compound in a straightforward way. When Exp. Sect.
2246
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2244Ϫ2252

A New Route to Cationic Half-Sandwich Ruthenium(II) Complexes
FULL PAPER
(2)
The conversion depends strongly on the complex used,
the best results being found for Ru1 and Ru4 for the conver-
sion of geraniol. The poor conversion of nerol has already
been observed in other catalytic systems.[27] Ru2 is not active
under these conditions; this fact may be ascribed to the
steric bulkiness or the high Lewis basicity of the PCy2 frag-
ment. Unfortunately, the complexes are not able to induce
high enantioselectivity. Further studies are necessary to de-
termine if this catalytic system could be improved. Never-
theless, the results obtained with our complexes, although
moderate, represent the first application of a ruthenium
catalyst in the isomerisation of geraniol to optically en-
riched citronellal.[28]
The mechanistic rationale for the ruthenium-catalysed re-
dox isomerisation of allylic alcohols postulates the forma-
tion of a ruthenium-enolate complex.[29] We envisioned that
the formed enolate could be trapped by an electrophile, and
indeed we found that an aldol-type product was formed via
a ruthenium-catalysed cross-coupling of 3-buten-2-ol and
benzaldehyde [Equation (3)].
Scheme 7
(1)
Kirchner also found that C1-substituted allylic alcohols
react much more quickly than those featuring substituents
at the C3 carbon; heating was required to effect the conver-
sion of 2,3-disubstituted alcohols, while C1- and C3-disub-
stituted alcohols were not converted. To test the reactivity
of our optically active complexes in the redox isomerisation
of pro-chiral allylic alcohols, we chose geraniol and nerol
as substrates. Isomerisation of diethylgeranylamine for the
production of practically optically pure (ϩ)-citronellal cata-
lysed by a Rhϩ/BINAP complex is a very important indus-
trial process leading to key intermediates for the fragrance
and flavour industry.[26] Comparable success has not been
achieved for the corresponding isomerisation of readily
available geraniol or nerol. To the best of our knowledge,
the hitherto best catalyst system reported is based on a
Rhϩ/BINAP complex, and affords citronellal with 60%
ee.[27] Initially, the reaction was performed in CH2Cl2 at
room temperature and at reflux but only modest or no con-
version into citronellal was obtained, corroborating the
finding of Kirchner that C3-disubstituted alcohols are not
prone to undergo the redox isomerisation. However, when
we performed the reaction in THF at reflux, we were ple-
ased to observe conversion of the alcohol to citronellal
[Equation (2)]. The results are summarised in Table 1.
(3)
It has recently been reported that [RuCl2(PPh3)3] cata-
lyses this reaction.[30] The results obtained with our com-
plexes are detailed in Table 2. The yields are not satisfactory
due to the concurrent isomerisation of the allylic alcohol to
Table 2. Aldol reaction between benzaldehyde and 3-buten-2-ol[a]
Catalyst
Loading (mol %)
Solvent
CH2Cl2
Yield (%)
syn/anti
Ru1
5
1
0.1
5
1
0.1
1
1
1
1
31
25
26
57
80
51
49
47
49
53
69:31
74:26
67:33
76:24
77:23
77:23
71:29
62:38
60:40
63:37
Table 1. Redox isomerisation of geraniol and nerol[a]
THF
Catalyst
Substrate
Product
Conversion
ee
Ru1
Ru1
Ru2
Ru3
Ru4
Ru4
Ru5
geraniol
nerol
geraniol
geraniol
geraniol
nerol
(ϩ)-citronellal
citronellal
Ϫ
(Ϫ)-citronellal
(ϩ)-citronellal
(Ϫ)-citronellal
(ϩ)-citronellal
73%
18%
0%
71%
97%
35%
32%
19%
nd
Ϫ
8%
17%
nd
Ru2
Ru3
Ru4
Ru5
THF
THF
THF
THF
[a]
Reactions carried out at room temperature until complete con-
sumption of the allylic alcohol. Isolated yields based on benzal-
dehyde are reported. The diastereoselectivity was based on 1H
NMR analysis of the crude product. The values given in the table
are the average of at least two runs.
geraniol
13%
[a] The values given in the table are the average of at least two runs.
See the Exp. Sect. for more details.
Eur. J. Inorg. Chem. 2004, 2244Ϫ2252
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2247

A. Doppiu, A. Salzer
FULL PAPER
separated and the aqueous phase was washed with CH2Cl2 (3 ϫ
50 mL). The combined organic extracts were washed successively
with 1  HCl, saturated NaHCO3 and brine, and dried over
MgSO4. The solution was then filtered, and the solvent was re-
moved under reduced pressure. The bis(methanesulfonate) esters
were obtained in almost quantitative yield and they were used with-
out further purification for the preparation of the spiro com-
pounds.
butan-2-one; this may also explain the non-linear depen-
dence of product yield on catalyst loading. The enantiose-
lectivity is also disappointing, and a maximum of 5% ee has
been obtained. However, the diastereoselectivity is appreci-
able, favouring the syn diastereomer. A substantial improve-
ment in both yield and diastereoselectivity could be ob-
tained upon switching from CH2Cl2 to THF, and it is note-
worthy that the catalysts are quite active even at low load-
ing. Further studies to improve the catalyst performance
are intended.
Preparation of the Spiro Compounds: Freshly cracked cyclopen-
tadiene (2 equiv.) was added dropwise to a suspension of NaNH2
(2.5 equiv.) in THF (ca. 25 mL/g). The addition of CpH was ad-
justed so as to maintain a gentle reflux. Upon complete addition,
the mixture was stirred for 1/2 h and subsequently, a solution of
the bis(methanesulfonate) in THF (ca. 15 mL/g) was added drop-
wise. The addition was accompanied by considerable heat evol-
ution. The reaction mixture was stirred at room temperature over-
night. MeOH and H2O were carefully added, the mixture was di-
luted with Et2O and the layers separated. The aqueous phase was
washed with Et2O (3 ϫ 50 mL). The combined organic extracts
were washed with brine, dried over MgSO4, and treated with decol-
ourising charcoal. The solution was finally filtered, and the solvent
removed under reduced pressure. The oil obtained was purified by
filtration through a short column of basic alumina (activity I) with
hexane as eluent. Reproducible elemental analyses could not be
obtained, as the solvent could not be removed completely from the
oily compounds.
Conclusion
A novel synthetic method has been developed that allows
the one-pot synthesis of optically active cationic bis(aceton-
itrile)ruthenium() complexes of chiral cyclopentadienyl-
linked phosphane ligands in very high yields. The method
is general and allows the use of ligands bearing phosphanes
with variable steric and electronic features. Moreover, the
linker, which connects the cyclopentadienyl unit to the
phosphane, can be straightforwardly modified. Considering
the labile nature of the acetonitrile ligands, these complexes
are potential candidates for catalytic implementation. In-
deed, a preliminary study on their reactivity revealed that
the complexes catalyse the isomerisation of geraniol to cit-
ronellal, as well as an aldol-type cross-coupling reaction be-
tween 3-buten-2-ol and benzaldehyde. Although the ee val-
ues obtained are rather low, the method is of interest for
preparing other complexes that may find various appli-
cations in asymmetric catalysis.
Preparation of the Ligands by Ring Opening of the Spiro Compounds
with Lithium Phosphides: The lithium phosphide was prepared by
treating a solution of the secondary phosphane in THF (ca. 10 mL/
g) with nBuLi (1.1 equiv.) at Ϫ78 °C. The deep-orange solution
formed was stirred for 15 minutes at room temperature and then
cooled again to Ϫ78 °C. At this temperature, the spiro compound
(1.1 equiv.) dissolved in THF (1 mL/g) was added to the solution
and the reaction mixture was allowed to reach room temperature
overnight. The solvent was removed under reduced pressure and
the oil left was washed several times with hexane until precipitation
of the lithium salt of the ligand. This was dissolved in THF and
degassed H2O was added. After several extractions with CH2Cl2,
the combined organic phases were dried over MgSO4 and treated
with decolourising charcoal. The solution was filtered through a
short column of basic alumina (activity I), and the solvent removed
under reduced pressure affording the pure ligand as an air-sensitive
oil or solid as a one to one mixture of two regioisomers. The signals
in the 1H NMR spectra are highly overlapped and could not be
attributed. Accordingly, only selected 13C NMR data and the 31P
resonances are given. The ligands were characterised as their de-
rived ruthenium complexes.
Experimental Section
General Remarks: All reactions were carried out under nitrogen
using standard Schlenk techniques. Solvents were dried and deoxy-
genated by standard methods. NMR spectra were recorded on a
Varian Unity 500 spectrometer (500 MHz, 1H; 125 MHz, 13C;
202 MHz, 31P) at room temperature. Chemical shifts are given in
ppm relative to the solvent signal (1H and 13C) or to external
1
H3PO4 (31P). H and 13C{1H} signal assignments were confirmed
by 1H,1H-COSY, APT, 1H,13C-HETCOR, and NOE-difference
experiments. (R)-1-Phenylethane-1,2-diol was obtained by re-
duction of (R)-mandelic acid with LAH.[31] (S)-1-Cyclohexyle-
thane-1,2-diol was obtained by initial reduction of the arene ring
of (S)-mandelic acid with hydrogen using 5% Rh on alumina as
catalyst, followed by reduction of the carboxylic group with
LAH.[32] (R)-3-Phenylpropane-1,2-diol was obtained by reduction
of the corresponding acid with LAH. (R)-1-(2,4,6-Trimethylpheny-
l)ethane-1,2-diol was obtained by Sharpless asymmetric dihydrox-
ylation of the corresponding olefin.[33] [Ru(η5-C7H11)2H][BF4][34]
was prepared according to a published procedure. All other chemi-
cals were purchased and used without further purification.
Synthesis of the Ruthenium Complexes: A flask was charged with
the ligand and 1 in equimolar amounts. CH3CN (ca. 20 mL/g) was
added and the solution refluxed overnight. Then, the solution was
filtered through a short pad of deactivated alumina (activity V) and
the solvents evaporated to dryness. The solid obtained was washed
several times with Et2O and dried under high vacuum.
(S)-1-Phenylspiro[2.4]hepta-4,6-diene (S1): The title compound was
prepared, according to the representative procedure, from 10.0 g
Preparation of the Bis(methanesulfonate) Esters: NEt3 (2.4 equiv.) of (R)-phenyl-1,2-bis(methanesulfonyloxy)ethane as a colourless or
was added to a solution of the diol in CH2Cl2 (ca. 10 mL/g). The
solution was cooled in an ice-water bath and methanesulfonyl chlo-
ride (2.2 equiv.) in CH2Cl2 (5 mL/mL) was added dropwise over
1 h. The mixture was stirred for 1 h at 0 °C and then for 2 h at
room temp. The mixture was poured into 1  HCl. The layers were
pale-yellow oil. Yield: 4.64 g (81%). [α]2D0 ϭ ϩ229 (c ϭ 1, CHCl3).
2
3
1H NMR (CDCl3): δ ϭ 2.06 (dd, JH,H ϭ 4.58, JH,H ϭ 8.85 Hz,
2
3
1 H, CH2), 2.32 (dd, JH,H ϭ 4.58, JH,H ϭ 7.93 Hz, 1 H, CH2),
3
3.32 (t, JH,H ϭ 8.24 Hz, 1 H, CHPh), 5.92 (m, 1 H, CH), 6.21 (m,
1 H, CH), 6.45 (m, 1 H, CH), 6.50 (m, 1 H, CH), 7.18Ϫ7.30 (m,
2248
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2244Ϫ2252

A New Route to Cationic Half-Sandwich Ruthenium(II) Complexes
5 H, Ph) ppm. 13C NMR (CDCl3): δ ϭ 17.05 (CH2), 30.85 (CHPh), 1JC,P ϭ 9.6 Hz, CH, Cy), 32.98 (d, JC,P ϭ 8.6 Hz, CH, Cy), 33.24
FULL PAPER
1
1
1
45.55 (Cspiro), 126.37 (CH, Ph), 127.93 (CH, Ph), 128.05 (CH, Ph),
(d, JC,P ϭ 2.9 Hz, CH, Cy), 33.7 (d, JC,P ϭ 2.9 Hz, CH, Cy),
2
2
128.57 (CH), 130.21 (CH), 135.73 (CH), 138.77 (CH), 139.28 (C, 34.79 (d, JC,P ϭ 22.1 Hz, CH2), 35.48 (d, JC,P ϭ 23.0 Hz, CH2),
Ph) ppm. HRMS [Mϩ]: calcd. 168.09390; found 168.09335.
39.37 (d, JC,P ϭ 19.2 Hz, CHPh), 40.37 (d, JC,P ϭ 20.1 Hz,
CHPh), 40.93 (s, CH2, Cp), 43.61 (s, CH2, Cp), 142.50 (s, C, Cp),
1
1
(S)-1-Cyclohexylspiro[2.4]hepta-4,6-diene (S2): The title compound
was prepared, according to the representative procedure, from
20.0 g of (S)-cyclohexyl-1,2-bis(methanesulfonyloxy)ethane as a
colourless or pale-yellow oil. Yield: 9.15 g (79%). [α]2D0 ϭ Ϫ36 (c ϭ
2
142.53 (s, C, Cp), 145.03 (d, JC,P ϭ 12.4 Hz, C, Ph), 147.60 (d,
2JC,P ϭ 13.5 Hz, C, Ph) ppm. 31P NMR (CDCl3): δ ϭ 18.08 (s),
18.75 (s) ppm. HRMS [Mϩ]: calcd. 366.24764; found 366.24738.
1
1, CHCl3). H NMR (CDCl3): δ ϭ 0.90Ϫ2.0 (m, 11 H, Cy), 1.60
(S)-[2-(Cyclopenta-1,3-dienyl)-1-(2,4,6-trimethylphenyl)ethyl]di-
phenylphosphane and (S)-[2-(Cyclopenta-1,4-dienyl)-1-(2,4,6-tri-
2
3
2
(dd, JH,H ϭ 3.96, JH,H ϭ 7.32 Hz, 1 H, CH2), 1.79 (dd, JH,H
ϭ
3.96, 3JH,H ϭ 8.55 Hz, 1 H, CH2), 1.90 (m, 1 H, CH, Cy), 6.09 (m, methylphenyl)ethyl]diphenylphosphane (L3): The title compounds
1 H, CH), 6.29 (m, 1 H, CH), 6.49 (m, 1 H, CH), 6.59 (m, 1 H, were prepared as an equimolar mixture of two regioisomers, ac-
CH) ppm. 13C NMR (CDCl3): δ ϭ 19.18 (CH2, Cy), 26.00 (CH2, cording to the representative procedure, from 3.1 g of S3 and 2.5 g
Cy), 26.22 (CH2), 26.38 (CH2, Cy), 32.91 (CH2, Cy), 33.29 (CH2, of HPPh2 as a colourless oil. Yield: 5.0 g (93%). 13C NMR (CDCl3,
Cy), 35.23 (CHCy), 42.09 (CH, Cy), 42.77 (Cspiro), 127.50 (CH),
129.95 (CH), 135.67 (CH), 139.77 (CH) ppm. HRMS [Mϩ]: calcd.
174.14085; found 174.14067.
selected data): δ ϭ 20.74 (CH3), 21.36 (CH3), 22.49 (CH3), 22.65
2
2
(CH3), 31.47 (d, JC,P ϭ 20.1 Hz, CH2), 32.12 (d, JC,P ϭ 21.1 Hz,
CH2), 39.87 (d, 1JC,P ϭ 17.2 Hz, CHMs), 40.85 (d, 1JC,P ϭ 18.2 Hz,
CHMs), 41.03 (s, CH2, Cp), 43.56 (s, CH2, Cp) ppm. 31P NMR
(CDCl3): δ ϭ Ϫ6.53 (s), Ϫ6.26 (s) ppm. HRMS [Mϩ]: calcd.
397.20852; found 397.20828.
(S)-1-(2,4,6-Trimethylphenyl)spiro[2.4]hepta-4,6-diene (S3): The title
compound was prepared, according to the representative pro-
cedure, from 12.5 g of (R)-(2,4,6-trimethylphenyl)-1,2-bis(methane-
sulfonyloxy)ethane as a pale-yellow oil. Yield: 6.80 g (87%). [α]2D0
ϭ
(R)-[2-Cyclohexyl-2-(cyclopenta-1,3-dienyl)ethyl]diphenylphosphane
and (R)-[2-Cyclohexyl-2-(cyclopenta-1,4-dienyl)ethyl]diphenylphos-
phane (L4): The title compound was prepared as an equimolar mix-
ϩ2 (c ϭ 1, CHCl3). 1H NMR (CDCl3): δ ϭ 2.17 (dd, 2JH,H ϭ 4.27,
2
3
3JH,H ϭ 8.55 Hz, 1 H, CH2), 2.23 (dd, JH,H ϭ 4.27, JH,H
ϭ
3
9.16 Hz, 1 H, CH2), 2.29 (s, 9 H, CH3), 3.14 (t, JH,H ϭ 8.85 Hz, ture of two regioisomers, according to the representative procedure,
1 H, CHMs), 5.85 (m, 1 H, CH), 6.34 (m, 1 H, CH), 6.50 (m, 1 H, from 2.1 g of S2 and 2.0 g of HPPh2 as a colourless solid. Yield:
CH), 6.58 (m, 1 H, CH), 6.83 (s, 2 H, Ms) ppm. 13C NMR (CDCl3):
3.1 g (80%). 13C NMR (CDCl3, selected data): δ ϭ 29.99 (d,
δ ϭ 20.23 (CH2), 20.80 (CH3), 28.61 (CHMs), 45.03 (Cspiro), 128.55 1JC,P ϭ 15.3 Hz, CH2), 31.37 (d, JC,P ϭ 10.55 Hz, CH2), 39.73 (s,
1
3
(CH, Ms), 129.11 (CH), 130.04 (CH), 133.19 (C, Ms), 135.94 (C, CH2, Cp), 40.81 (s, CH2, Cp), 41.76 (d, JC,P ϭ 8.54 Hz, CH, Cy),
Ms), 137.52 (CH), 139.19 (CH) ppm. HRMS [Mϩ]: calcd. 42.16 (d, JC,P ϭ 13.4 Hz, CHCy), 42.89 (d, JC,P ϭ 13.4 Hz,
2
2
3
210.14085; found 210.14114.
CHCy), 43.03 (d, JC,P ϭ 7.79 Hz, CH, Cy), 137.78 (d, JC,P
14.4 Hz, C), 138.92 (d, JC,P ϭ 13.4 Hz, C), 139.22 (d, JC,P
ϭ
ϭ
(R)-1-Benzylspiro[2.4]hepta-4,6-diene (S4): The title compound was
prepared, according to the representative procedure, from 25.4 g of
(S)-benzyl-1,2-bis(methanesulfonyloxy)ethane as a pale-yellow oil.
Yield: 14.3 g (95%). [α]2D0 ϭ Ϫ11 (c ϭ 1, CHCl3). 1H NMR
14.4 Hz, C), 147.19 (d, JC,P ϭ 3.8 Hz, C), 149.85 (d, JC,P ϭ 4.8 Hz,
C) ppm. 31P NMR (CDCl3): δ ϭ Ϫ18.89 (s), Ϫ18.19 (s) ppm.
HRMS [Mϩ]: calcd. 360.20069; found 360.20039.
(CDCl3): δ ϭ 1.71 (dd, 2JH,H ϭ 4.27, 3JH,H ϭ 7.32 Hz, 1 H, CH2), (S)-[2-Benzyl-2-(cyclopenta-1,3-dienyl)ethyl]diphenylphosphane and
1.85 (dd, 2JH,H ϭ 4.27, 3JH,H ϭ 8.55 Hz, 1 H, CH2), 2.30 (m, 1 H, (S)-[2-Benzyl-2-(cyclopenta-1,4-dienyl)ethyl]diphenylphosphane (L5):
3
CHBzl), 2.94 (d, JH,H ϭ 7.02 Hz, 2 H, CH2Ph), 6.10 (m, 1 H,
The title compound was prepared as an equal mixture of two re-
CH), 6.33 (m, 1 H, CH), 6.49 (m, 1 H, CH), 6.58 (m, 1 H, CH), gioisomers according to the representative procedure from 6.5 g of
7.1Ϫ7.3 (m, 5 H, Ph) ppm. 13C NMR (CDCl3): δ ϭ 19.46 (CH2),
28.29 (CHBzl), 38.32 (CH2Ph), 43.20 (Cspiro), 126.10 (CH, Ph),
S4 and 6.0 g of HPPh2. Yield: colourless solid, 11.0 g (93%). 13C
NMR (CDCl3, selected data): δ ϭ 33.11 (d, 1JC,P ϭ 12.4 Hz, CH2),
1
2
127.95 (CH, Ph), 128.29 (CH, Ph), 128.29 (CH), 130.68 (CH), 34.41 (d, JC,P ϭ 12.4 Hz, CH2), 39.20 (d, JC,P ϭ 15.45 Hz,
135.41 (CH), 139.64 (CH) 140.74 (C) ppm. HRMS [Mϩ]: calcd. CHBzl), 39.96 (d, JC,P ϭ 15.3 Hz, CHBzl), 40.91 (s, CH2, Cp),
1
3
182.10955; found 182.10938.
41.32 (s, CH2, Cp), 42.42 (d, JC,P ϭ 8.54 Hz, CH2Ph), 43.98 (d,
3JC,P ϭ 9.55 Hz, CH2Ph) ppm. 31P NMR (CDCl3): δ ϭ Ϫ20.56 (s),
Ϫ20.47 (s) ppm. HRMS [Mϩ]: calcd. 368.16939; found 368.16937.
(S)-[2-(Cyclopenta-1,4-dienyl)-1-phenylethyl]diphenylphosphane and
(S)-[2-(Cyclopenta-1,3-dienyl)-1-phenylethyl]diphenylphosphane
(L1): The title compound was prepared as an equal mixture of two
regioisomers according to the representative procedure from 5.0 g
(S)-Bis(acetonitrile)[(η5-2-cyclopentadienyl-1-phenylethyl)diphenyl-
phosphane-κP] ruthenium(II) Tetrafluoroborate (Ru1). This com-
pound was prepared, according to the representative procedure,
of S1 and 5.0 g of HPPh2 as a colourless solid. Yield: 9.2 g (97%).
2
13C NMR (CDCl3, selected data): δ ϭ 33.67 (d, JC,P ϭ 23.1 Hz, from 2.8 g of L1 and 3.0 g of 1 as yellow crystals. The NMR spec-
2
CH2), 34.29 (d, JC,P ϭ 23.0 Hz, CH2), 41.07 (s, CH2, Cp), 43.70
troscopic data are identical to those already reported for [Ru{η5-
1
1
(s, CH2, Cp), 45.28 (d, JC,P ϭ 13.4 Hz, CHPh), 46.17 (d, JC,P
ϭ
C5H4CH2CH(Ph)PPh2-κP}(CH3CN)2][PF6].[12]
Yield:
4.73 g
14.4 Hz, CHPh), 140.83 (s, C, Cp), 140.90 (s, C, Cp), 144.69 (d, (96%). M.p.: 165Ϫ170 °C. [α]2D0 ϭ ϩ8.7 (c ϭ 1, CHCl3). IR (KBr):
2JC,P ϭ 13.5 Hz, C, Ph), 147.20 (d, JC,P ϭ 13.4 Hz, C, Ph) ppm. ν˜ ϭ 2274 cmϪ1, 1600, 1581, 1058, 704. MS (FAB): m/z ϭ 536 [Mϩ],
2
31P NMR (CDCl3): δ ϭ 0.458 (s), 0.886 (s) ppm. HRMS [Mϩ]:
495 [Mϩ Ϫ CH3CN], 454 [Mϩ Ϫ 2CH3CN]. C29H28BF4N2PRu
(623.38): calcd. C 55.87, H 4.53, N 4.49; found C 55.73, H 4.62,
N 4.35.
calcd. 354.15374; found 354.15323.
(S)-Dicyclohexyl[2-(cyclopenta-1,4-dienyl)-1-phenylethyl]phosphane
and (S)-Dicyclohexyl[2-(cyclopenta-1,3-dienyl)-1-phenylethyl]phos-
(S)-Bis(acetonitrile)[dicyclohexyl(η5-2-cyclopentadienyl-1-phenyl-
phane (L2): The title compounds were prepared as an equimolar ethyl)phosphane-κP]ruthenium(II
mixture of two regioisomers, according to the representative pro- compound was prepared, according to the representative pro-
cedure, from 4.7 g of S1 and 5.0 g of HPCy2 as a colourless solid. cedure, from 1.1 g of L2 and 1.13 g of 1 as a yellow solid. Yield:
Yield: 8.6 g (93%). 13C NMR (CDCl3, selected data): δ ϭ 32.83 (d, 1.9 g (100%). M.p.: 195Ϫ200 °C. [α]2D0ϭ ϩ97.5 (c ϭ 1, CH3CN).
) Tetrafluoroborate (Ru2): This
Eur. J. Inorg. Chem. 2004, 2244Ϫ2252
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2249

A. Doppiu, A. Salzer
FULL PAPER
1H NMR (CD2Cl2): δ ϭ 0.52 (m, 1 H, Cy), 1.0Ϫ2.21 (m, 21 H, Cy), 4.90 Hz, CH), 129.13 (d, JC,P ϭ 4.77 Hz, CH), 130.38 (CH), 130.76
2.41 (d, 5JH,P ϭ 0.91 Hz, 3 H, CH3CN), 2.44 (d, 5JH,P ϭ 0.92 Hz, 3 (CH), 132.05 (d, JC,P ϭ 10.55 Hz, CH), 133.29 (d, JC,P ϭ 11.56 Hz,
H, CH3CN), 2.47Ϫ2.67 (m, 2 H, CH2), 3.95 (m, 1 H, Cp), 4.19 (m, CH), 133.53 (C), 133.82 (C) ppm. 31P NMR (CD2Cl2): δ ϭ 52.14
1 H, CHPh), 4.25 (m, 1 H, Cp), 5.05 (m, 1 H, Cp), 5.13 (m, 1 H,
(s, PPh2) ppm. IR (KBr): ν˜ ϭ 2274 cmϪ1, 1629, 1600, 1581, 1058,
Cp), 7.26Ϫ7.41 (m, 5 H, Ph) ppm. 13C NMR (CD2Cl2): δ ϭ 4.45 704. MS (FAB): m/z (%) ϭ 502 (60) [Mϩ Ϫ CH3CN ϩ 1], 461
(CH3, CH3CN), 4.67 (CH3, CH3CN), 26.83 (CH2, Cy), 27.16 (CH2, (100) [Mϩ Ϫ2CH3CN ϩ 1], 87 (100) [BF4Ϫ]. C29H34BF4N2PRu
Cy), 27.40 (CH2, Cy), 27.70 (d, JC,P ϭ 13.4 Hz, CH2, Cy), 28.48 (629.42): calcd. C 55.33, H 5.44, N 4.45; found C 54.93, H 5.69,
2
(d, JC,P ϭ 12.44 Hz, CH2), 28.65 (CH2, Cy), 30.92 (d, JC,P
ϭ
N 4.08.
1
5.78 Hz, CH2, Cy), 33.08 (d, JC,P ϭ 17.34 Hz, CH, Cy), 36.06 (d,
(S)-Bis(acetonitrile)[(2-benzyl-η5-2-cyclopentadienylethyl)di-
phenylphosphane-κP]ruthenium(II) Tetrafluoroborate (Ru5): The title
compound was prepared, according to the representative pro-
cedure, from 2.69 g of L5 and 2.75 g of 1 as a yellow solid. Yield:
4.65 g (100%). [α]2D0ϭ Ϫ10.7 (c ϭ 1, CH3CN). M.p.: 141Ϫ144 °C.
1JC,P ϭ 13.32 Hz, CH, Cy), 56.54 (d, JC,P ϭ 18.22 Hz, CHPh),
1
57.62 (CH, Cp), 68.80 (CH, Cp), 77.86 (d, JC,P ϭ 7.66 Hz, CH,
Cp), 90.97 (d, JC,P ϭ 3.89 Hz, CH, Cp), 102.88 (d, JC,P ϭ 3.77 Hz,
Cipso, Cp), 127.44 (C, CN), 127.82 (CH), 128.21 (d, JC,P ϭ 2.89 Hz,
CH), 128.48 (C, CN), 129.21 (CH), 138.47 (d, JC,P ϭ 8.67 Hz, C)
ppm. 31P NMR (CD2Cl2): δ ϭ 88.78 (s, PCy2) ppm. IR (KBr): ν˜ ϭ
2274 cmϪ1, 1600, 1058, 704. MS (FAB): m/z (%) ϭ 549 (15) [Mϩ
ϩ 1], 508 (85) [Mϩ Ϫ CH3CN], 466 (100) [Mϩ Ϫ 2CH3CN], 87
(100) [BF4Ϫ]. C29H40BF4N2PRu (635.47): calcd. C 54.80, H 6.36, N
4.41; found C 55.18, H 6.60, N 3.91.
5
1H NMR (CD2Cl2): δ ϭ 1.99 (d, JH,P ϭ 1.22 Hz, 3 H, CH3CN),
5
2.18 (d, JH,P ϭ 1.53 Hz, 3 H, CH3CN), 2.55 (m, 1 H, CHBzl),
2
3
2.82 (dd, JH,H ϭ 13.43, JH,H ϭ 8.24 Hz, 1 H, CH2Ph), 2.95 (dd,
2JH,H ϭ 13.43, 3JH,H ϭ 6.72 Hz, 1 H, CH2Ph), 3.02Ϫ3.18 (m, 2 H,
CH2), 4.12 (m, 1 H, Cp), 4.14 (m, 1 H, Cp), 5.06 (m, 1 H, Cp),
5.25 (m, 1 H, Cp), 7.01Ϫ7.74 (m, 15 H, Ph) ppm. 13C NMR
(CD2Cl2): δ ϭ 3.78 (CH3, CH3CN), 4.16 (CH3, CH3CN), 38.60 (d,
(S)-Bis(acetonitrile){[η5-2-cyclopentadienyl-1-(2,4,6-trimethyl-
phenyl)ethyl]diphenylphosphane-κP}ruthenium(II) Tetrafluoroborate
(Ru3): This compound was prepared, according to the representa-
tive procedure, from 1.4 g of L3 and 1.33 g of 1 as a yellow solid.
Yield: 2.3 g (99%). M.p.: 160Ϫ165 °C. [α]2D0ϭ Ϫ12.4 (c ϭ 1,
CH3CN). 1H NMR (CD2Cl2): δ ϭ 1.40 (s, 3 H, CH3), 1.76 (d,
5JH,P ϭ 1.22 Hz, 3 H, CH3CN), 2.15 (s, 3 H, CH3), 2.20 (s, 3 H,
CH3), 2.37 (d, 5JH,P ϭ 0.92 Hz, 3 H, CH3CN), 2.42Ϫ2.79 (m, 2 H,
CH2), 4.30 (m, 1 H, Cp), 4.45 (m, 1 H, Cp), 5.18 (m, 1 H, Cp),
5.31 (m, 1 H, Cp), 5.42 (m, 1 H, CHMs), 6.48 (s, 1 H, Ms), 6.76
(s, 1 H, Ms), 7.20Ϫ7.50 (m, 10 H, Ph) ppm. 13C NMR (CD2Cl2):
δ ϭ 3.29 (CH3, CH3CN), 4.48 (CH3, CH3CN), 20.80 (CH3), 22.22
3
2JC,P ϭ 6.66 Hz, CHBzl), 41.12 (d, JC,P ϭ 21.11 Hz, CH2Ph),
50.36 (d, 1JC,P ϭ 33.54 Hz, CH2), 57.87 (CH, Cp), 67.78 (CH, Cp),
80.47 (d, JC,P ϭ 8.67 Hz, CH, Cp), 88.39 (d, JC,P ϭ 5.78 Hz, CH,
Cp), 112.42 (d, JC,P ϭ 4.77 Hz, Cipso, Cp), 126.69 (C, CN), 126.86
(CH), 127.26 (C, CN), 128.76 CH), 129.13 (CH), 129.21 (CH),
129.33 (CH), 129.40 (CH), 130.56 (d, JC,P ϭ 5.78 Hz, CH), 130.83
(d, JC,P ϭ 1.88 Hz, CH), 132.18 (d, JC,P ϭ 9.67 Hz, CH), 132.99
(C), 133.26 (d, JC,P ϭ 11.43 Hz, CH), 133.62 (C), 139.77 (C) ppm.
31P NMR (CD2Cl2): δ ϭ 52.87 (s, PPh2) ppm. IR (KBr): ν˜ ϭ 2277
cmϪ1, 1630, 1602, 1058, 744, 699. MS (FAB): m/z ϭ 510 (30) [Mϩ
Ϫ CH3CN ϩ 1], 469 (100) [Mϩ Ϫ 2CH3CN ϩ 1], 87 (100) [BF4Ϫ].
C30H30BF4N2PRu (637.40): calcd. C 56.53, H 4.74, N 4.40; found
C 56.65, H 4.67, N 4.31.
2
(CH3), 22.63 (CH3), 30.16 (d, JC,P ϭ 10.55 Hz, CH2), 60.71 (CH,
Cp), 64.28 (d, 1JC,P ϭ 23.12 Hz, CHMs), 65.37 (CH, Cp), 82.82 (d,
J
C,P ϭ 7.66 Hz, CH, Cp), 87.19 (d, JC,P ϭ 5.78 Hz, CH, Cp), 107.97
Procedure for the Isomerisation of Allylic Alcohols: In a representa-
tive experiment, an NMR tube was charged under nitrogen with
the catalyst (1 mol %), CD2Cl2 was added (0.5 mL), followed by
the allylic alcohol (prop-2-en-1-ol, but-3-en-2-ol, or 1-phenylprop-
2-en-1-ol). The tube was capped and vigorously shaken. The reac-
(d, JC,P ϭ 4.77 Hz, Cipso, Cp), 126.04 (C, CN), 127.63 (C, CN),
128.70 (d, JC,P ϭ 9.55 Hz, CH), 128.48 (d, JC,P ϭ 8.67 Hz, CH),
129.57 (CH), 129.82 (CH), 130.90 (C), 131.16 (CH), 131.43 (CH),
132.29 (d, JC,P ϭ 8.67 Hz, C), 132.91 (d, JC,P ϭ 9.67 Hz, CH),
135.50 (C), 136.14 (C), 136.57 (d, JC,P ϭ 11.56 Hz, CH), 137.25
(C), 138.98 (C) ppm. 31P NMR (CD2Cl2): δ ϭ 86.78 (s, PPh2) ppm.
IR (KBr): ν˜ ϭ 2274 cmϪ1, 1629, 1600, 1581, 1058, 704. MS (FAB):
m/z (%) ϭ 579 (20) [Mϩ ϩ 1], 538 (60) [Mϩ Ϫ CH3CN ϩ 1], 497
(100) [Mϩ Ϫ 2CH3CN ϩ 1], 87 (100%) [BF4Ϫ]. C32H34BF4N2PRu
(665.45): calcd. C 57.75, H 5.15, N 4.21; found C 57.70, H 5.38,
N 4.29.
1
tion was monitored by H and 13C{1H} NMR spectroscopy. Typi-
cally, the reaction was completed (no olefinic protons were de-
tected) after three hours at room temperature affording exclusively
the carbonyl compounds with only traces of unidentified side-prod-
ucts.
Procedure for the Isomerisation of Geraniol and Nerol: A flask was
charged with a catalytic amount of the ruthenium complex (5
mol %) and THF (10 mL) and the allylic alcohol (2 mL) were then
added. The reaction mixture was refluxed for 24 hours. The solvent
was then removed under reduced pressure and the residue was fil-
tered through a short pad of silica gel eluting with Et2O. The oil
left after evaporation of the solvent was bulb-to-bulb distilled
(100Ϫ110 °C/10 mbar) affording pure citronellal. The conversion
(S)-Bis(acetonitrile)[(2-cyclohexyl-η5-2-cyclopentadienylethyl)di-
phenylphosphane-κP]ruthenium(II) Tetrafluoroborate (Ru4): The title
compound was prepared, according to the representative pro-
cedure, from 1.1 g of L4 and 1.15 g of 1 as a yellow solid. Yield:
1.8 g (95%). M.p.: 142Ϫ145 °C. [α]2D0ϭ ϩ4 (c ϭ 1, CH3CN). 1H
NMR (CD2Cl2): δ ϭ 0.80Ϫ1.86 (m, 11 H, Cy), 1.94 (m, 1 H,
5
5
CHCy), 1.97 (d, JH,P ϭ 0.91 Hz, 3 H, CH3CN), 2.21 (d, JH,P
ϭ
1
was determined on the crude reaction product by H NMR spec-
1.22 Hz, 3 H, CH3CN), 2.92 (td, JH,P ϭ 13.42, JH,H ϭ 6.10 Hz, 1
H, CH2), 3.18 (td, JH,P ϭ 13.12, JH,H ϭ 4.88 Hz, 1 H, CH2), 4.02
(m, 1 H, Cp), 4.27 (m, 1 H, Cp), 5.12 (m, 1 H, Cp), 5.24 (m, 1 H,
Cp), 7.39Ϫ7.56 (m, 10 H, Ph) ppm. 13C NMR (CD2Cl2): δ ϭ 3.69
troscopy. The ee was determined by comparison of the optical ro-
tation (neat) with that reported in the literature.[27]
Procedure for the Catalytic Aldol Reaction: A flask was charged
(CH3, CH3CN), 4.12 (CH3, CH3CN), 26.37 (CH2, Cy), 31.54 (CH2, with a catalytic amount of the ruthenium complex (5 to 0.1 mol %).
2
Cy), 33.44 (CH2, Cy), 42.57 (d, JC,P ϭ 20.10 Hz, CHCy), 42.72
The solvent was added (5 or 10 mL), followed by benzaldehyde (5
(d, JC,P ϭ 4.77 Hz, CH, Cy), 47.94 (d, JC,P ϭ 33.54 Hz, CH2), or 10 mmol) and then 3-buten-2-ol (10 or 20 mmol). The reaction
57.48 (CH, Cp), 68.61 (CH, Cp), 79.98 (d, JC,P ϭ 7.79 Hz, CH, mixture was stirred at room temperature until complete consump-
Cp), 88.62 (d, JC,P ϭ 5.65 Hz, CH, Cp), 112.82 (d, JC,P ϭ 4.77 Hz, tion of the allylic alcohol (checked by TLC). The solvent was then
Cipso, Cp), 126.56 (C, CN), 127.28 (C, CN), 129.05 (d, JC,P removed under reduced pressure and the residue was purified by
3
1
ϭ
2250
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2244Ϫ2252

A New Route to Cationic Half-Sandwich Ruthenium(II) Complexes
FULL PAPER
[8b]
1999, 18, 3843Ϫ3850.
C. Slugovc, E. Rüba, R. Schimd, K.
flash column chromatography (hexane/ethyl acetate, 7:3). The syn/
anti ratio was determined by 1H NMR spectroscopy. The enanti-
oselectivities were determined by HPLC analysis with a Chiralpak
AS column (eluent: hexane/2-propanol, 95:5; flow rate: 0.8 mL/
min; column temperature: 30 °C; retention times for the syn dia-
stereoisomer: 15 min and 22 min).
[8c]
Kirchner, Organometallics 1999, 18, 4230Ϫ4233.
C. M.
Standfest-Hauser, K. Mereiter, R. Schmid, K. Kirchner, Dalton
[8d]
Trans. 2003, 11, 2329Ϫ2334.
C. M. Standfest-Hauser, K.
Mereiter, R. Schmid, K. Kirchner, Organometallics 2002, 21,
[8e]
4891Ϫ4893.
E. Rüba, K. Mereiter, R. Schmid, V. N. Sapu-
nov, K. Kirchner, H. Schottenberger, M. J. Calhorda, L. F.
Veiros, Chem. Eur. J. 2002, 8, 3948Ϫ3961. [8f] E. Rüba, K. Mer-
eiter, R. Schmid, K. Kirchner, E. Bustelo, M. C. Puerta, P.
[8g]
Valerga, Organometallics 2002, 21, 2912Ϫ2920.
E. Becker,
Acknowledgments
C. Slugovc, E. Rüba, C. Standfest-Hauser, K. Mereiter, R.
Schmid, K. Kirchner, J. Organomet. Chem. 2002, 649, 55Ϫ63.
The generous financial support by the Deutsche Forschungs-
gemeinschaft DFG within the Collaborative Research Centre (SFB,
380) ‘‘Asymmetric Syntheses by Chemical and Biological Means’’
and by the ‘‘Fonds der Chemischen Industrie’’ is gratefully
acknowledged. We thank Prof. Carsten Bolm for a generous gift of
(R)-1-(2,4,6-trimethylphenyl)ethane-1,2-diol.
[8h]
E. Rüba, K. Mereiter, R. Schmid, K. Kirchner, H. Schot-
[8i]
tenberger, J. Organomet. Chem. 2001, 637Ϫ639, 70Ϫ74.
E.
Rüba, K. Mereiter, R. Schmid, K. Kirchner, Chem. Commun.
[8l]
2001, 19, 1996Ϫ1997.
K. Mauthner, K. M. Soldouzi, K.
Mereiter, R. Schmid, K. Kirchner, Organometallics 1999, 18,
4681Ϫ4683.
[9]
H. Butenschön, Chem. Rev. 2000, 100, 1527Ϫ1564.
[1] [1a]
[10] [10a] N. Dodo, Y. Matsushima, M. Uno, K. Onitsuka, S. Takah-
ashi, J. Chem. Soc., Dalton Trans. 2000, 35Ϫ41. [10b] Y. Matsu-
shima, N. Komatsuzaki, Y. Ajioka, M. Yamamoto, H. Kikuchi,
Y. Takata, N. Dodo, K. Onitsuka, M. Uno, S. Takahashi, Bull.
Chem. Soc. Jpn. 2001, 74, 527Ϫ537.
S.-I. Murahashi, T. Naota, J. Synth. Org. Chem. Jpn. 1988,
46, 930Ϫ956. [1b] T. Naota, H. Takaya, S.-I. Murahashi, Chem.
[1c]
Rev. 1998, 98, 2599Ϫ2660.
B. M. Trost, F. D. Toste, A. B.
[1d]
Pinkerton, Chem. Rev. 2001, 101, 2067Ϫ2096.
B. M. Trost,
[1e]
Acc. Chem. Res. 2002, 35, 695Ϫ705.
V. Ritleng, C. Sirlin,
[11]
M. Pfeffer, Chem. Rev. 2002, 102, 1731Ϫ1769.
Y. Matsushima, K. Onitsuka, T. Kondo, T. Mitsudo, S. Takah-
ashi, J. Am. Chem. Soc. 2001, 123, 10405Ϫ10406.
A. Doppiu, U. Englert, A. Salzer, Inorg. Chim. Acta 2003,
[2]
[2a] M. I. Bruce, M. A. Bennett, T. W. Matheson, Comprehensive
Organometallic Chemistry (Eds.: G. Wilkinson, F. G. A. Stone,
[12]
[2b]
E. W. Abel), Pergamon Press: Oxford, 1982.
M. I Bruce,
350, 435Ϫ441.
M. A. Bennett, A. F. Hill, Comprehensive Organometallic
[13]
E. Becker, K. Mereiter, M. Puchberger, R. Schmid, K.
Kirchner, A. Doppiu, A. Salzer, Organometallics 2003, 22,
Chemistry II (Eds.: G. Wilkinson, F. G. A. Stone, E. W. Abel),
[2c]
Pergamon Press: Oxford, 1995.
U. Kölle, Encyclopedia of
3164Ϫ3170.
Inorganic Chemistry (Ed.: R. King), Wiley Chichester, 1994,
[14] [14a]
T. Kauffmann, J. Ennen, H. Lhotak, A. Rensing, F. Stein-
7, 3533Ϫ3545.
seifer, A. Woltermann, Angew. Chem. 1980, 92, 328Ϫ329; An-
[3] [3a]
[3b]
T. P. Gill, K. R. Mann, Organometallics 1982, 1, 485Ϫ488.
B. M. Trost, C. M. Older, Organometallics 2002, 21,
[14b]
gew. Chem. Int. Ed. Engl. 1980, 19, 301Ϫ302.
T.
Kauffmann, K. Berghus, A. Rensing, Chem. Ber. 1985, 118,
2544Ϫ2547.
4507Ϫ4516.
[4]
[5]
[6]
M. O. Albers, D. J. Robinson, A. Shaver, E. Singleton, Or-
ganometallics 1986, 5, 2199Ϫ2205.
C. Slugov, E. Rüba, R. Schmid, K. Kirchner, K. Mereiter,
Monatsh. Chem. 2000, 131, 1241Ϫ1248.
W. Luginbühl, P. Zbinden, P. A. Pittet, T. Armbruster, H. B.
Bürgi, A. E. Merbach, A. Ludi, Inorg. Chem. 1991, 30,
2350Ϫ2355.
[15] [15a]
S. Ciruelos, U. Englert, A. Salzer, C. Bolm, A. Maischak,
Organometallics 2000, 19, 2240Ϫ2242. [15b] S. Ciruelos, A. Dop-
piu, U. Englert, A. Salzer, J. Organomet. Chem. 2002, 663,
183Ϫ191.
[16] [16a]
C. F. Wilcox, R. R. Craig, J. Am. Chem. Soc. 1961, 83,
[16b]
3866Ϫ3871.
A. I. D’yachenko, L. G. Menchikov, O. M.
Nefedov, Bull. Acad. Sci. USSR, Div. Chem. Sci. (Engl.
Transl.) 1985, 649Ϫ651.
[7] [7a]
B. M. Trost, P. L. Fraisse, Z. T. Ball, Angew. Chem. 2002,
114, 1101Ϫ1103; Angew. Chem. Int. Ed. 2002, 41, 1059Ϫ1162.
[17] [17a] B. Rieger, G. Jany, R. Fawzi, M. Steimann, Organometallics
[7b]
B. M. Trost, H. C. Shen, Angew. Chem. 2001, 113,
[17b]
1994, 13, 647Ϫ653.
B. Rieger, G. Jany, M. Steimann, R.
[7c]
2375Ϫ2378; Angew. Chem. Int. Ed. 2001, 40, 2313Ϫ2317.
Fawzi, Z. Naturforsch., Teil B 1994, 49, 451Ϫ458.
D. C. Brookings, S. A. Harrison, R. J. Whitby, B. Crombie, R.
V. H. Jones, Organometallics 2001, 20, 4574Ϫ4583.
H. C. Kolb, M. S. Van Nieuwenhze, K. B. Sharpless, Chem.
Rev. 1994, 94, 2483Ϫ2547.
S. E. Schaus, B. D. Brandes, J. F. Larrow, M. Tokunaga, K. B.
Hansen, A. E. Gould, M. E. Furrow, E. N. Jacobsen, J. Am.
Chem. Soc. 2002, 124, 1307Ϫ1315.
B. M. Trost, J.-P. Surivet, Angew. Chem. 2001, 113, 1516Ϫ1519;
[18]
[19]
[20]
[7d]
Angew. Chem. Int. Ed. 2001, 40, 1468Ϫ1472.
B. M. Trost,
A. B. Pinkerton, Angew. Chem. 2000, 112, 368Ϫ370; Angew.
Chem. Int. Ed. 2000, 39, 360Ϫ362. [7e] B. M. Trost, A. B. Pink-
erton, M. Seidel, J. Am. Chem. Soc. 2001, 123, 12466Ϫ12476.
[7f] B. M. Trost, A. B. Pinkerton, F. D. Toste, M. Sperrle, J. Am.
[7g]
Chem. Soc. 2001, 123, 12504Ϫ12509.
B. M. Trost, M. T.
[7h]
Rudd, J. Am. Chem. Soc. 2002, 124, 4178Ϫ4179.
B. M.
[21] [21a]
E. N. Tsvetkov, N. A. Bondarenko, I. G. Malakhova, M.
Trost, A. B. Pinkerton, J. Am. Chem. Soc. 1999, 121,
[21b]
[7i]
I. Kabachnik, Synthesis 1986, 198Ϫ205.
H. B. Kagan, M.
10842Ϫ10843.
1999, 121, 9728Ϫ9729.
Toste, J. Am. Chem. Soc. 2000, 122, 5877Ϫ5878.
Trost, F. D. Toste, H. Sheng, J. Am. Chem. Soc. 2000, 122,
B. M. Trost, F. D. Toste, J. Am. Chem. Soc.
Tahar, J.-C. Fiaud, Tetrahedron Lett. 1991, 32, 5959Ϫ5962. [21c]
P. Pellon, Tetrahedron Lett. 1992, 33, 4451Ϫ4452.
Hauptmann, P. J. Fagan, W. Marshall, Organometallics 1999,
[7l]
B. M. Trost, R. E. Brown, F. D.
[21d]
[7m]
E.
B. M.
[7n]
18, 2061Ϫ2073.
2379Ϫ2380.
B. M. Trost, F. D. Toste, J. Am. Chem. Soc.
[22]
[7o]
In addition to NMR spectroscopic data, the regio- and stereo-
chemistry in the ring opening of (S)-1-cyclohexylspi-
ro[2.4]hepta-4,6-diene has also been confirmed by an X-ray
structural determination of a metal complex with the derived
cyclopentadienylphosphane ligand: U. Englert, personal com-
munication.
2000, 122, 714Ϫ715.
B. M. Trost, A. B. Pinkerton, J. Am.
Chem. Soc. 2000, 122, 8081Ϫ8082. [7p] B. M. Trost, A. B. Pink-
erton, D. Kremzow, J. Am. Chem. Soc. 2000, 122,
12007Ϫ12008. [7q] B. M. Trost, J.-P. Surivet, F. D. Toste, J. Am.
[7r]
Chem. Soc. 2001, 123, 2897Ϫ2898.
B. M. Trost, M. Mach-
[7s]
acek, M. J. Schnaderbeck, Org. Lett. 2000, 2, 1761Ϫ1764.
[23]
[24]
[7t]
T. Lumini, D. N. Cox, R. Roulet, K. Schenk, J. Organomet.
B. M. Trost, H. C. Shen, Org. Lett. 2000, 2, 2523Ϫ2525.
B.
Chem. 1992, 434, 363Ϫ385.
M. Trost, F. D. Toste, Tetrahedron Lett. 1999, 40, 7739Ϫ7743.
[8] [8a]
E. Rüba, W. Simanko, K. Mauthner, K. M. Soldouzi, C.
Slugovc, K. Mereiter, R. Schmid, K. Kirchner, Organometallics
D. N. Cox, R. Roulet, J. Chem. Soc., Chem. Commun. 1989,
175Ϫ176.
Eur. J. Inorg. Chem. 2004, 2244Ϫ2252
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2251

A. Doppiu, A. Salzer
FULL PAPER
[25]
[30c]
I. Lee, F. Dahan, A. Maisonnat, R. Poilblanc, Organometallics
2003, 5, 657Ϫ660.
M. Wang, X.-F. Yang, C.-J. Li, Eur. J.
1994, 13, 2743Ϫ2750.
Org. Chem. 2003, 998Ϫ1003.
[26] [26a]
[26b]
[31]
S. Otsuka, K. Tani, Synthesis 1991, 665Ϫ680.
R.
J. M. Brown, B. A. Murrer, J. Chem. Soc., Perkin Trans. 2
1982, 489Ϫ497.
Noyori, Asymmetric Catalysis in Organic Synthesis, Chapter 3,
[26c]
[32]
[33]
J. Wiley & Sons, New York, 1994.
S. Akutagawa, K. Tani,
D. P. Riley, R. E. Shumate, J. Org. Chem. 1980, 45, 5187Ϫ5193.
G. Miao, B. E. Rossiter, J. Org. Chem. 1995, 60, 8424Ϫ8427.
Catalytic Asymmetric Synthesis, Chapter 3 (Ed.: I. Ojima),
Wiley-VCH, New York, 2000.
C. Chapuis, M. Barthe, J.-Y. de Saint Laumer, Helv. Chim. Acta
2001, 84, 230Ϫ242.
For a discussion of the difficulties encountered in catalyzing
[34] [34a]
A. Bauer, U. Englert, S. Geyser, F. Podewils, A. Salzer,
Organometallics 2000, 19, 5471Ϫ5476. [34b] A. Salzer, M. Käser,
A. Hosang, A. Bauer, U. Effertz, E. Klinkhammer, F. Podewils,
Synthetic Methods of Organometallic and Inorganic Chemistry
(Eds.: W. A. Herrmann, G. Brauer), Georg Thieme Verlag:
[27]
[28]
[28a]
the isomerisation of allylic alcohols, see:
C. Bianchini, A.
[28b]
Meli, W. Oberhauser, New J. Chem. 2001, 25, 11Ϫ12.
R.
Stuttgart, 2000; 9, 45Ϫ58. [34c]A. Salzer, A. Bauer, S. Geyser,
[34d]
´
´
Uma, C. Crevisy, R. Gree, Chem. Rev. 2003, 103, 27Ϫ52.
F. Podewils, Inorg. Synth. 2003, 34, in press.
A. Salzer, F.
[29] [29a] R. C. van der Drift, E. Bouwman, E. Drent, J. Organomet.
Podewils, S. Geyser, Eur. Patent. Appl. 110429.8, Int. Patent.
Appl. PCT/IB01/00784.
[29b]
Chem. 2002, 650, 1Ϫ24.
B. M. Trost, R. J. Kulawiec, J.
Am. Chem. Soc. 1993, 115, 2027Ϫ2036.
Received December 15, 2003
Early View Article
Published Online April 7, 2004
[30] [30a] M. Wang, C.-J. Li, Tetrahedron Lett. 2002, 43, 3589Ϫ3591.
[30b]
X.-F. Yang, M. Wang, R. S. Varma, C.-J. Li, Org. Lett.
2252
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 2244Ϫ2252