Inverted piano-stool dimers of half-sandwich Ru(II) complexes with (R)-phenylglycine methylester and (S)-phenylalanineamide: An X-ray structural study and preliminary catalytic results

Article abstract of DOI:10.1016/j.jorganchem.2005.07.028

The two new complexes [Ru(η⁶-p-cymene)(κ¹- N-(rac)-phenylglycine methylester)Cl₂] (1) and [Ru(η⁶- p-cymene)(κ²-N,N′-(S)-phenylalanineamido)Cl] (2) have been synthesized by reacting [Ru(p-cymene)Cl₂]₂ with the HCl-free corresponding ligands. The complexes have been fully characterized by spectroscopic and analytical methods, and their solid structures have been determined by single crystal X-ray analysis. Both complexes have a pseudo-tetrahedral geometry: in 1 ruthenium is bound to the nitrogen, to the η⁶-coordinated p-cymene molecule and to two chloride ligands. In 2 the ligand has lost an amide proton binding ruthenium in a N,N′ bidentate fashion, the coordination geometry being completed by a η⁶-coordinated p-cymene molecule and a chloride ligand. The X-ray structure of 1 has revealed the racemization of the ligand, while in 2 the ligand has retained its configuration but, interestingly, the two diasteromers R_RuS_C and S_RuS_C have co-crystallized in the same single crystal. The crystal architecture of 2 is characterized by the presence of two opposite helices of akin diastereoisomers, connected through strong intermolecular hydrogen bonds between the amine and carbonyl groups. ESI-MS of an i-PrOH solution of 2 points out that the dimers are maintained also in solution. Complex 2 is an active catalyst for the homogeneous transfer hydrogenation of acetophenone and cyclohexanone in an i-PrOH/t-BuOK mixture, with TOFs up to 800 and 1000 h^-1, respectively.

Full text of DOI:10.1016/j.jorganchem.2005.07.028

Journal of Organometallic Chemistry 690 (2005) 4602–4610
www.elsevier.com/locate/jorganchem
Inverted piano-stool dimers of half-sandwich Ru(II) complexes
with (R)-phenylglycine methylester and (S)-phenylalanineamide:
An X-ray structural study and preliminary catalytic results
*
Paolo Pelagatti , Alessia Bacchi, Francesca Calbiani, Mauro Carcelli, Lisa Elviri,
Corrado Pelizzi, Dominga Rogolino
`
Dipartimento di Chimica Generale ed Inorganica, Chimica Analitica, Chimica Fisica, Universita degli Studi di Parma,
Parco Area delle Scienze 17/A, 43100 Parma, Italy
Received 6 June 2005; accepted 7 July 2005
Available online 19 August 2005
Abstract
The two new complexes [Ru(g6-p-cymene)(j1-N-(rac)-phenylglycine methylester)Cl2] (1) and [Ru(g6-p-cymene)(j2-N,N0-(S)-phe-
nylalanineamido)Cl] (2) have been synthesized by reacting [Ru(p-cymene)Cl2]2 with the HCl-free corresponding ligands. The com-
plexes have been fully characterized by spectroscopic and analytical methods, and their solid structures have been determined by
single crystal X-ray analysis. Both complexes have a pseudo-tetrahedral geometry: in 1 ruthenium is bound to the nitrogen, to
the g6-coordinated p-cymene molecule and to two chloride ligands. In 2 the ligand has lost an amide proton binding ruthenium
in a N,N0 bidentate fashion, the coordination geometry being completed by a g6-coordinated p-cymene molecule and a chloride
ligand. The X-ray structure of 1 has revealed the racemization of the ligand, while in 2 the ligand has retained its configuration
but, interestingly, the two diasteromers RRuSC and SRuSC have co-crystallized in the same single crystal. The crystal architecture
of 2 is characterized by the presence of two opposite helices of akin diastereoisomers, connected through strong intermolecular
hydrogen bonds between the amine and carbonyl groups. ESI-MS of an i-PrOH solution of 2 points out that the dimers are main-
tained also in solution. Complex 2 is an active catalyst for the homogeneous transfer hydrogenation of acetophenone and cyclohex-
anone in an i-PrOH/t-BuOK mixture, with TOFs up to 800 and 1000 hꢀ1, respectively.
Ó 2005 Elsevier B.V. All rights reserved.
Keywords: Ruthenium; Hydrogen transfer; N ligands; Helical structures; Amino acids
1. Introduction
Homogeneous enantioselective catalytic processes
promoted by transition metal complexes are elegant
ways to obtain important organic molecules with high
optical purity. The search of asymmetric supporting li-
gands able to transfer the chirality to the final organic
product is then a fundamental chapter in catalysis. Ami-
no acids or their derivatives represent a cheap and easily
available class of chiral ligands, and the literature counts
several examples of homogeneous enantioselective pro-
cesses catalyzed by amino acid based metal complexes
[1,4]. However, homogeneous enantioselective catalysts
containing amino acid amides have been much less
The use of a-amino acids or their derivatives as li-
gands in organometallic chemistry is a well defined topic
[1]. Metal complexes with amino acid-type ligands have
found applications mainly in bio-inorganic chemistry,
for example in the synthesis of peptides [2] or bio-inor-
ganic models [3].
*
Corresponding author.
E-mail address: paolo.pelagatti@unipr.it (P. Pelagatti).
0022-328X/$ - see front matter Ó 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.07.028

P. Pelagatti et al. / Journal of Organometallic Chemistry 690 (2005) 4602–4610
4603
investigated [5]. In this paper, we describe the synthesis,
the characterization and the X-ray structure analysis of
two new half-sandwich ruthenium(II) complexes ob-
tained by using (R)-phenylglycine methyl ester (1) and
(S)-phenylalanineamide (2) as ligands. Furthermore, a
preliminary study on the potentiality of complex 2 as
homogeneous catalyst in the hydrogen transfer reaction
from i-PrOH to ketones is reported.
served racemization of the ligand during the synthesis of
[Ru(g6-benzene)(L-alaMe)Cl2] complex [8], where a
methyl group is attached to the a-carbon atom.
The crystal structure of 1 (Fig. 1) shows that the
coordination geometry of the [RuNCl2(g6-C6)] moiety,
summarized in Table 1, may be described as tetrahedral
by considering the center (Cy) of the g6-p-cymene
aromatic ring as the fourth ligand position. In fact, the
C15
2. Results and discussion
C19
C14
C13
C10
C16
After neutralization, L1 Æ HCl reacts with [Ru(p-cym-
ene)Cl2]2 in a methanol/dichloromethane mixture at
room temperature leading to complex 1 (Scheme 1).
In order to avoid the hydrolysis of the methyl-ester
group of the ligand, the reaction has been conducted un-
der strictly anhydrous conditions. 1 has been isolated as
an orange powder, stable in the solid state as well as in
open air solutions. The complex has a pseudo-tetrahe-
dral geometry, where ruthenium is bound to the amine
group of L1, to two chloride ligands and to a g6-coordi-
nated p-cymene molecule. The exclusion of the ester
group from the coordination sphere is evidenced by
the strong IR stretching band at 1736 cmꢀ1, equivalent
to that of L1 Æ HCl. The stretching signals of the amine
group are in the region 3278–3153 cmꢀ1. In the 1H
NMR spectrum, the most interesting feature is the split-
ting of the NH2 signals (two multiplets at 3.04 and 7.07
ppm, respectively), due to the interaction of the nitrogen
donor with the metal center; the other signals showing
the expected chemical shifts. In order to unequivocally
establish the structure of the complex, well shaped single
crystals of 1 have been grown in methanol. The X-ray
diffraction analysis has confirmed the proposed struc-
ture, although pointing out the racemization of the li-
gand. The metal-promoted activation of protons
attached to a-carbons is a known process which, how-
ever, takes place at high pH and temperatures, usually
[6]. Repeated attempts aimed to detect a Ru-hydride
intermediate by NMR experiments were uninformative.
More likely, the explanation of the occurred racemiza-
tion of L1 resides in the presence of a phenyl ring on
the a-carbon atom, able to stabilize the incipient carban-
ion [7] formed upon proton abstraction promoted by
traces of free t-BuOK. This is supported by the not ob-
C17
C12
Ru
C11
C18
C5
C4
C7
C3
Cl 2
C1
C2
Cl 1
N
C6
C8
O1
O2
C9
Fig. 1. Perspective view and labeling scheme of 1. Thermal ellipsoids
drawn at the 50% probability level.
Table 1
˚
Coordination bond lengths (A) and angles (°) for 1 and 2
1
Ru–N
Ru–(C10–C15)
Ru–Cy
Ru–Cl2
Ru–Cl1
2.165(1)
N–Ru–Cl2
2.156(1)–2.203(1) N–Ru–Cl1
81.69(4)
81.67(4)
87.23(1)
1.64
2.4139(4)
2.4206(4)
Cl2–Ru–Cl1
Cy–Ru–N
Cy–Ru–Cl1
Cy–Ru–Cl2
134
126
129
2
B
A
Ru1–N1
2.058(9)
Ru2–N3
2.053(9)
Ru1–N2
Ru1–(C10–C15)
2.125(8)
2.15(1)–2.20(1)
Ru2–N4
Ru2–(C29–C34)
2.133(9)
2.16(1)–
2.27(1)
2.438(3)
1.70
Ru1–Cl1
Ru1–Cy1
2.435(3)
1.65
Ru2–Cl2
Ru2–Cy2
Ph
O
Ru
1/2[Ru(p-cymene)Cl2]2
Cl
NH2
N1–Ru1–N2
N1–Ru1–Cl1
N2–Ru1–Cl1
Cy1–Ru1–N1
Cy1–Ru1–N2
Cy1–Ru1–Cl1
76.8(3)
86.6(3)
82.8(2)
131
133
128
N3–Ru2–N4
N3–Ru2–Cl2
N4–Ru2–Cl2
Cy2–Ru2–N3
Cy2–Ru2–N4
Cy2–Ru2–Cl2
76.7(3)
87.8(3)
82.6(3)
131
132
128
Cl
Ph
MeOH/CH2Cl2
H2N
OMe
O
L1
MeO
Scheme 1. Synthesis of complex 1.

4604
P. Pelagatti et al. / Journal of Organometallic Chemistry 690 (2005) 4602–4610
Ru–C distances relative to the p-cymene coordination
ular recognition pattern in (p-cymene)RuXYZ
complexes. In particular, this motif has been invoked
to justify the remarkable trend to co-crystallization
shown by piano-stool diastereoisomeric pairs [9], and
this issue will be further considered in discussing the
crystal structure of 2.
HL2 reacts with [Ru(p-cymene)Cl2]2 under the same
experimental conditions as for 1, leading to complex 2
(Scheme 2).
˚
range from 2.170(2) to 2.203(1) A, with Ru–Cy
˚
= 1.665(1)A.
The overall shape of 1 may thus be defined as a piano-
stool geometry. The large steric hindrance of p-cymene
produces a remarkable distortion in the tetrahedral
coordination geometry, by widening the Cy–Ru–N and
Cy–Ru–Cl angles and tightening the N–Ru–Cl ones.
The p-cymene is oriented so that the C10–C16 bond to
the methyl group is eclipsed to the Ru–Cl2 bond
(C16–C10–Ru–Cl2 = ꢀ11.6(2)°). The most relevant
conformational degrees of freedom of L1 in 1 are the
rotations around the Ru–NH2 bond and the NH2–CH
bond, which, taken together, determine the orientation
of the aromatic group of the phenylglycine residue rela-
tively to the p-cymene ligand. As regards the former, in 1
the N–H bonds belonging to the mono-hapto –NH2
group are eclipsed with respect to the Ru–Cl bonds
(Cl2–Ru–N–C1 = 126.2(1)°), while the latter is charac-
terized by torsion angle Ru–N–C1–C2 = ꢀ60.5(2)°.
These values show that the ligand aromatic residue is
oriented towards the p-cymene methyl end, giving an
edge-to-face intramolecular contact between the p-cym-
ene aromatic C–H groups and the L1 p electron density
The product is stable in the solid state as well as in
open air solutions. The ligand shows a bidentate behav-
ior, binding the metal through the amine and amide
nitrogens. The complexation occurs with elimination
of a HCl molecule, consequent to the monodeprotona-
tion of the amide nitrogen, and then L1 acts as an anio-
nic N,N0 bidentate ligand. The pseudo-tetrahedral
coordination is completed by a chloride ligand and a
g6-coordinated p-cymene ring. The IR spectrum shows
a strong stretching band of the amide C@O group at
1581 cmꢀ1 [10], while the stretching of the NH bonds
originates an unresolved band in the region 3285–3115
cmꢀ1. The 1H NMR spectrum recorded in CD3OD
shows the presence of two diastereomers (see Section
4); at room temperature they are in a 75:25 ratio. The
formation of pairs of diastereomers is common for
half-sandwich Ru(II) complexes of the three-legged pia-
no-stool type containing chiral bidentate ligands: in
complexes of general formula Ru(arene)(LL0)X
(X = halogen or other unidentate ligand) the formally
monohapto coordination of the p-cymene ring renders
the metal a stereogenic center. Thus, in the presence of
an enantiomerically pure ligand, in the present case hav-
ing an SC configuration, the two diastereomers RRuSC
and SRuSC arise. Interestingly, the X-ray diffraction
analysis carried out on a single crystal of 2 obtained at
5 °C from a dichloromethane solution, showed the pres-
ence of both diastereomers. Contrarily to ligand L1, the
X-ray analysis of 2 indicates that L2 has retained its
configuration at the stereogenic carbon atom. To the
best of our knowledge, this is the first example of an
a-amino (primary)amide Ru(II) complex of the three-
legged piano-stool type structurally characterized. The
presence of two diasteromers in the same crystal is an
interesting feature already observed for other Ru(g6-
arene)(LL0)Cl type complexes, and about 20 examples
˚
(H11ꢁ ꢁ ꢁC2 = 2.70 A). In examining the crystal packing
of 1 it is convenient to refer to the distribution of the
shortest Ruꢁ ꢁ ꢁRu contacts, in order to analyze how
the complex molecules are paired by intermolecular
interactions (Fig. 2).
The two closest Ruꢁ ꢁ ꢁRu distances are related to an
inverted piano-stool arrangement (Fig. 2, A–B pair) as-
˚
sisted by C–Hꢁ ꢁ ꢁCl2(i) contacts (Ruꢁ ꢁ ꢁRu(i) = 5.68 A, C
˚
ꢁ ꢁ ꢁCl = 3.590(2) A, Cy–Ruꢁ ꢁ ꢁRu(i) = 78°, i = ꢀx,ꢀy,
2ꢀz), and to a different centrosymmetric pair (Fig. 2,
A–C pair) based on –NHꢁ ꢁ ꢁCl1(ii) hydrogen bonds
˚
˚
(Ruꢁ ꢁ ꢁRu(ii) = 6.80 A, Nꢁ ꢁ ꢁCl(ii) = 3.517(2) A, N–Hꢁ ꢁ ꢁ
Cl(ii) = 145(2)°, ii = ꢀx,ꢀy,1ꢀz). Brunner has individu-
ated the inverted piano-stool motif as a common molec-
O
Ru
1/2[Ru(p-cymene)
Cl2]2
Cl
HN
O
NH2
H2N
NH2
MeOH/CH2Cl2
–HCl
HL2
Fig. 2. Molecular aggregation in the crystal structure of 1. A–
B = inverted piano-stool dimer; A–C = hydrogen bonded dimer.
Scheme 2. Synthesis of complex 2.

P. Pelagatti et al. / Journal of Organometallic Chemistry 690 (2005) 4602–4610
4605
are reported in the literature [9]; the majority of them
deals with complexes of general formula [Ru(g6-p-cym-
ene)(salicylaldiminate-ligand)Cl], and only in one case
the ligand is N,N0 [11]. This co-crystallization is usually
based on a sort of molecular recognition within inverted
piano-stool dimers. This feature is evidenced in Fig. 3,
where the crystal structure of the two independent dia-
stereomers SRuSC (molecule A, Ru2) and RRuSC (mole-
cule B, Ru1) present in the asymmetric unit of 2 is
reported.
Table 1 lists the most relevant geometric features of
the two molecules, which cocrystallize in the acentric po-
lar space group I41. In both cases the ruthenium coordi-
nation is tetrahedral, by considering the chlorine ligand,
the amidic and aminic nitrogen atoms of the bidentate
ligand, and the geometric centre (Cy) of the p-cymene
aromatic ring as the four donors. In both molecules
the coordination to the deprotonated amidic donor is
stronger than to the aminic group. The N,N0 chelation
of L2 to Ru generates a five membered chelation ring,
with bite angles N–Ru–N = 76.8(3)° in A and B, while
the remaining coordination angles on Ru are larger than
the corresponding ones in 1. The opposite absolute con-
figuration of the metal in A and B results in a different
orientation of the Ru–Cl bond relatively to the C–H
bond of the chiral carbon atom (C2 and C21, respec-
tively, for A and B). Both bonds are axially oriented
with respect to the chelation rings, but they are in anti
configuration in B, and syn in A. The p-cymene is always
coordinated with the methyl end pointing towards the
chlorine
ligand
(C16–C11–Ru1–Cl1 = ꢀ20.6(9)°,
˚
˚
C16ꢁ ꢁ ꢁCl1 = 3.55(1) A; C35–C34–Ru2–Cl2 = ꢀ23.5(9)°,
C35ꢁ ꢁ ꢁCl2 = 3.56(1) A). Diastereomers A and B cocrys-
tallize by forming an inverted piano-stool dimer, where-
in A and B are related by a non-crystallographic
operation mimicking a center of inversion, so that the
aromatic p-cymene moieties lie parallel on opposite sides
relatively to the metal atoms. It has been suggested [9]
that the stability of this inverted piano-stool recognition
pattern is the reason of the relevant occurrence of diaste-
reomers cocrystallization for Ru(g6-arene)XYZ com-
plexes, with electronegative X and Y. The inversion
operation is required in order to create contacts between
complementary groups on A and B: C29ꢁ ꢁ ꢁO1 =
3.41(1), C10ꢁ ꢁ ꢁO2 = 3.36(1), N1ꢁ ꢁ ꢁCl2 = 3.490(8), and
˚
N3ꢁ ꢁ ꢁCl1 = 3.481(8) A. This non-crystallographic oper-
ation acts locally between the metal centers of the A–B
dimer so that the two molecules show opposite chirality
on the metals. However, due to the enantiomeric purity
of L2, the symmetry relation cannot be extended to the
entire molecular pair, but it is confined to the central
core, inside the limit defined by the chiral carbons.
The two resulting SRuSC and RRuSC complexes are in
0
˚
close contact, with Ru1ꢁ ꢁ ꢁRu2 = 5.14 A, Cy–Ru–Ru –
Cy0 = 179°, and Cy–Ru–Ru0 = 111°, which are quite typ-
ical values for this kind of pattern [9]. The next shortest
C8
C9
C7
C38
C35
C4
C34
C33
C6
C5
C32
C29
C37
C36
O1
C1
C31
Ru2
C30
Cl 2
C3
C2
N1
A
B
N4
N2
C18
Cl 1
N3
C21
C28
Ru1
C22
C23
C17
C20
O2
C14
C27
C15
C26
C13
C12
C10
C24
C25
C11
C19
C16
Fig. 3. Perspective view and labeling of the two diastereomers SRuSC (molecule A) and RRuSC (molecule B) co-crystallized in 2. Thermal ellipsoids at
the 30% probability level, hydrogens omitted with the exception of those bonded to N and to chiral C atoms. Molecules A and B form an inverted
piano-stool pair.

4606
P. Pelagatti et al. / Journal of Organometallic Chemistry 690 (2005) 4602–4610
˚
(Nꢁ ꢁ ꢁO = 2.84(1) A, N–Hꢁ ꢁ ꢁO = 161(6)°, ii = y + 1,
ꢀx + 1/2,z + 1/4) for A. Both helices A and B are gen-
erated by the 41 symmetry operator, with pitch
˚
c = 14.572(1) A (Fig. 4), but they have opposite hand-
ness, right-handed for B and left-handed for A.
The helix handness is governed by metal configura-
tion, so that all the chlorine ligands stick outside from
the corners of the chains. The intra-chain packing of
L2 is the main difference between the two diastereomeric
helices A and B. Helices interact in the crystal through
the inverted piano-stool dimers located at each 90° turn
of the helix, so that each helix is surrounded by four
helices of opposite type, packed in antiparallel mode.
The formation of 2 has been monitored by ESI-MS in
i-PrOH, dissolving HL2 Æ HCl, [Ru(p-cymene)Cl2]2 and
KOH in a 2:1:2 molar ratio ([Ru] = 10ꢀ3). Within 60
minutes the reaction is completed (disappearance of
the signal belonging to [Ru(p-cymene)Cl2]2) and the
ESI(+) spectrum shows a signal centered at m/z = 399,
attributable to [2 ꢀ Cl]+, and another one at m/z =
435, attributable to [2 + H+]+. Importantly, the dimeric
units seem maintained in i-PrOH solution; in fact, as it is
shown in Fig. 5, the spectrum exhibits two ion signals at
m/z 869 and 907 accounting for the presence of the di-
meric form of 2 as [22 + H+]+ and [22 + K+]+, respec-
tively (cone voltage: 50 V, see Section 4). These
experimental isotope clusters were in agreement with
the theoretical masses and reconstructed singly charged
ESI isotope patterns. These data agree well with the re-
Fig. 4. Helical structure of hydrogen bonded chains in 2. The opposite
handness of the chains formed by A (left handed) and B (right handed)
is evidenced by a dotted line linking the metal atoms. The pair of
molecules A–B in the center of the figure form an inverted piano-stool
dimer (fifth molecules from the top). The packing of the chains is
shown in the inset.
˚
Ruꢁ ꢁ ꢁRu contacts are larger than 6.7 A. Most inter-
esting is the organization of the crystal architecture.
Each diastereomer is in fact part of a helix of akin mol-
ecules based on the strong hydrogen bonds N2–
˚
Hꢁ ꢁ ꢁO1(i) (Nꢁ ꢁ ꢁO = 2.78(1) A, N–Hꢁ ꢁ ꢁO = 163(7)°,
i = y + 1/2,1ꢀx,zꢀ1/4) for
B
and N4-Hꢁ ꢁ ꢁO2(ii)
Fig. 5. ESI(+) mass spectra of the pre-catalyst 2 acquired at a cone voltage of 50 V after 60 min. In the inset are shown the ion signals at m/z 869 and
907 attributable to the [2M + H+]+ and [2M + K+]+ adduct ions from the dimeric form of 2, respectively.

P. Pelagatti et al. / Journal of Organometallic Chemistry 690 (2005) 4602–4610
4607
cently observed tendency of neutral a-amino amidate
(Noyoriꢀs catalysts) and acidate half-sandwich Ru(II)
complexes to form dimers in solution. PGSE (pulsed
field gradient spin-echo) NMR technique [12], has evi-
denced that the aggregation tendency is maintained in
several solvents, i-PrOH included.
The not observed enantioselectivity in the acetophe-
none reduction, seems easily ascribable to the contem-
porary presence of both diastereoisomers of complex 2
in solution, although the ratio between the two enantio-
mers of 1-phenylethanol does not correspond to the ra-
tio between the two diastereomers of 2, as established by
1H NMR. This could indicate a labile metal configura-
tion, as demonstrated for amino acidato half-sandwich
complexes by Carmona et al. [16c,17]. The preparation
of Ru(II) complexes containing other amino amide li-
gands designed with the aim of improving the enantiose-
lectivity of the hydrogen transfer reaction, and the
possible understanding of the role of the dimers on the
catalysis, is currently under way in our laboratory.
Chiral vicinal diamines or chiral aminoalcohols are
important classes of supporting ligands for enantioselec-
tive homogeneous catalysis. These ligands have been
extensively used in the Ru(II) catalyzed asymmetric
hydrogen transfer reaction from alcohols to ketones,
leading to exceptionally high enantiomeric excesses
[13]. Phenylalanine amide contains two NH2 groups
connected through a chiral backbone, and then complex
2 can be considered a good candidate as catalyst for the
asymmetric hydrogen transfer reaction. The use of ami-
no acid amides as chiral supporting ligands for hydro-
gen transfer reduction counts a limited number of
recent examples, dealing with proline derivatives [14]
and dipeptide-analogue ligands [15]. Moreover, no
information are available on the structure of the pre-
catalyst since they are generated in situ, usually. With
this in mind, we were particularly interested in testing
complex 2 as catalyst for the hydrogen transfer reaction
from i-PrOH to ketones. We chose two different ketones
as substrate, one prochiral (acetophenone) and one not-
prochiral (cyclohexanone). The catalytic experiments
were conducted in dry i-PrOH, at two different temper-
atures (room temperature or 80 °C), using t-BuOK as
co-catalyst and a Ru:ketones:base = 1:1000:2 molar ra-
tio. The catalytic results are collected in Table 2.
Already at room temperature complex 2 is an active
catalyst in the hydrogenation of acetophenone, leading
to 75% conversion after 6 hours of reaction (entry 1).
When the temperature is raised to 80 °C, after 1 hour
of reaction a 80% conversion has been observed (entry
2). Regrettably, the ees are negligible in both cases. With
cyclohexanone at 80 °C the complete conversion to
cyclohexanol has been obtained after 1 h of reaction (en-
try 3). Although in the case of acetophenone the ee is far
to be satisfactory, the TOFs are significant (up to
1000 hꢀ1 in the case of cyclohexanone). In fact, the pro-
ductivity of 2 is much greater than those of parent com-
plexes, to say those generated with proline-derivatives
[14] and the preformed amino acidato half-sandwich
Ru(II) complexes [16].
3. Conclusions
We have prepared and structurally characterized two
half-sandwich Ru(II) complexes containing amino acid
derivatives. The unusual cocrystallization of the two
diastereomers RRuSC and SRuSC has been observed with
complex 2, having (S)-phenylalanineamide as ligand.
This occurs through an inverted piano-stool dimer motif
and each diastereomers belongs to a helix of akin mole-
cules. The two helixes have opposite handness defined
by the metal configuration, left-handed for the SRuSC
isomer and right-handed for the RRuSC isomer. The di-
meric units seem maintained in i-PrOH solutions, as evi-
denced by ESI-MS analysis. Complex 2 is an active
catalyst in the hydrogen transfer reaction of ketones
(cyclohexanone and acetophenone) in i-PrOH/t-BuOK,
leading to high TOF values.
4. Experimental
4.1. General methods
All reactions were performed under an atmosphere of
dry nitrogen, employing standard Schlenk techniques.
Solvents were dried prior to use and stored over molec-
ular sieves and under nitrogen. Elemental analysis (C,
H, N) were performed by using a Carlo Erba Mod.
EA 1108 apparatus. Infrared spectra were recorded with
a Nicolet 5PCFT-IR spectrophotometer in the range
Table 2
Catalytic data for the hydrogen transfer reactions catalyzed by complex 2a
Entry
Substrate
T (°C)
Time (h)
Conversion (%)
TONb
TOFc
ee (%)
1
2
3
a
Acetophenone
Acetophenone
Cyclohexanone
r.t.
80
80
6
1
1
75
80
100
750
800
1000
125
800
1000
6(R)
6(R)
–
Ru/substrate/t-BuOK molar ratio = 1:1000:2.
TON = mole product/mole catalyst.
TOF = (mole product/mole catalyst) hꢀ1
b
c
.

4608
P. Pelagatti et al. / Journal of Organometallic Chemistry 690 (2005) 4602–4610
4000–400 cmꢀ1, by using KBr disks. H NMR spectra
were obtained at 300 K on a Bruker 300 FT spectrome-
ter, by using SiMe4 as internal standard. The GC anal-
yses have been performed by means of a Dani HP
3800 flame-ionization instrument, equipped with a CP
Chirasil Dex CB capillary column. The GC conditions
were: oven temperature from 90 to 190 °C with an incre-
ment of 15 °C/min.; retention times: acetophenone = 4.9
min, (R)-1-phenylethanol = 5.7 min, (S)-1-phenyletha-
nol = 5.8 min. The ESI-MS spectra were acquired with
a Quattro LC triple quadrupole instrument (Micromass,
Manchester, UK) equipped with an electrospray inter-
face (Micromass). ESI-MS analysis were performed by
operating the mass spectrometer in positive ion (PI)
mode acquiring mass spectra over the scan range m/z
100–1300 using a step size of 0.1 Da and a scan time
of 1.2 s. The operating parameters of interface were as
follows: source temperature 70 °C, desolvation tempera-
ture 70 °C, ES (+) voltage 3.0 kV, cone voltage 50 V, rf
lens 0.3 V.
5.2. [Ru(g6-p-cymene)(j2-N,N0-L2)Cl] Æ 1/2H2O (2)
1
130 mg (0.6 mmol) of HL2 Æ HCl and 67 mg (0.6
mmol) of t-BuOK were dissolved in 30 ml of methanol.
A dichloromethane solution (15 ml) of [Ru(p-cymene)-
Cl2]2 (200 mg, 0.3 mmol) was then added and the
resulting orange solution was stirred at room tempera-
ture for four hours. The solvents were then removed in
vacuum and the residue was treated with dichlorometh-
ane filtering the KCl off. The remaining red light solu-
tion is refrigerated at 5 °C obtaining red prismatic
crystals. After filtration, a further crop of solid (orange
powder) was obtained by treating the mother liquor
with diethyl ether. Yield: 230 mg (75%). M.p.: 227–229
°C. Anal. calc. for C19H25ClN2ORu Æ 1/2H2O: C,
51.52; H, 5.92; N, 6.32. Found: C, 51.83; H, 5.78; N,
1
6.56%. H NMR (CD3OD): major diastereomer d 1.23
3
(d, 3H, p-cymene, JHH = 7Hz), 1.32 (d, 3H, p-cymene,
3JHH = 7Hz), 2.03 (sbr, 1H, NH), 2.09 (s, 3H, p-cym-
ene), 2.43 (m, 2H, CH2), 2.70 (m, 1H, p-cymene), 3.07
2
The hydrochloride forms of enantiomerically pure
(R)-phenylglycine methylester (L1 Æ HCl) and S-phenyl-
alanine amide (HL2 Æ HCl) were purchased by Aldrich,
while [Ru(p-cymene)Cl2]2 was synthesized following a
literature reported method [18]. Acetophenone (Aldrich)
and cyclohexanone (CarloErba) have been utilized as
received.
(dd, 1H, CH(Bz)NH2, JHH = 4.3Hz), 5.55 (d, 1H, p-
3
3
cymene, JHH = 6Hz), 5.65 (d, 1H, p-cymene, JHH
=
3
5.8 Hz), 6.11 (d, 1H, p-cymene, JHH = 5.8Hz), 6.50
(d, 1H, p-cymene, 3JHH = 6Hz), 7.32–7.09 (m, 7H, Ph +
NH2); minor diastereomer d 1.16 (d, 3H, p-cymene,
3JHH = 6.9Hz), 5.20 (dbr, 2H, p-cymene, 3JHH = 5.7Hz),
5.42 (dbr, 2H, p-cymene, 3JHH = 4.7Hz). IR: 3285–3115
(NH), 1581 (C@O amide).
5. Preparation of the Ru(II) complexes
6. X-ray structures
5.1. [Ru(g6-p-cymene)(j1-N-L1)Cl2] (1)
Single crystals of 1 and 2 were mounted on glass fibers
and X-ray diffraction data were collected on a Bruker-
Siemens SMART AXS 1000 equipped with CCD detec-
tor, using graphite monochromated Mo Ka radiation
198 mg (0.9 mmol) of L1 Æ HCl and 100 mg (0.9
mmol) of t-BuOK were dissolved in 30 ml of methanol
at room temperature; n-pentane was added and KCl
was filtered-off. A dichloromethane solution (15 ml)
of [Ru(p-cymene)Cl2]2 (300 mg, 0.5 mmol) was then
added and the resulting orange solution was stirred
at room temperature for four hours. An orange solid
precipitated. This was filtered-off, washed with diethyl
ether and dried in vacuum. A further crop of solid
was recovered from the refrigerated mother liquor.
Red crystals suitable for X-ray analysis were obtained
by re-crystallization in methanol. Yield: 240 mg
(78%). M.p.: 168–170 °C. Anal. calc. for C19H23Cl2-
NO2Ru: C, 48.41; H, 5.34; N, 2.97. Found: C, 48.57;
˚
(k = 0.71069 A). Data collection details are: crystal to
detector distance = 5.0 cm, 2424 frames collected (com-
plete sphere mode), time per frame = 30 s, oscillation D
U = 0.300°. Crystal decay resulted negligible in both
cases. Data reduction was performed up to d = 0.70
˚
and 0.90 A for 1 and 2, respectively, by the SAINT package
[19] and data were corrected for absorption effects by the
SADABS [20] procedure (Tmax = 1.000, Tmin = 0.788 for 1
and Tmax = 1.000, Tmin = 0.848 for 2). The phase prob-
lem was solved by direct methods [21] and refined by
full-matrix least-squares on all F2 [22], implemented in
the WINGX package [23]. Anisotropic displacement
parameters were refined for all non-hydrogen atoms,
while hydrogen atoms were introduced in calculated
positions, except for amine and amide hydrogens, which
were located from Fourier maps and refined isotropi-
cally. A partially occupied (50%) water molecule com-
pletes the asymmetric unit contents for 2. Absolute
structure for 2 was assessed by Flackꢀs parame-
ter = ꢀ0.11(7). Final maps were featureless. Use of the
1
H, 5.32; N, 2.93%. H NMR (CDCl3): d 1.27 (d, 3H,
3
p-cymene, JHH = 7Hz), 1.32 (d, 3H, p-cymene,
3JHH = 7Hz), 2.16 (s, 1H, CH(Ph)NH2), 2.23 (s, 3H,
p-cymene), 3.04 (m, 1H, NH), 2.93 (m, 1H, p-cymene),
3.77 (s, 3H, C(O)OCH3), 5.27 (d, 1H, p-cymene,
3
3JHH = 6Hz), 5.33 (d, 1H, p-cymene, JHH = 6Hz),
5.48 (m, 2H, p-cymene), 7.42 (m, 5H, Ph), 7.72 (m,
1H, NH). IR: 3278-3219-3153 m(NH2), 1736 m(C@O es-
ter), 1260 m(C@O ester).

P. Pelagatti et al. / Journal of Organometallic Chemistry 690 (2005) 4602–4610
4609
Table 3
Crystal data and structure refinement for 1 and 2
Identification code
1
2
Empirical formula
Formula weight
Wavelength (A)
Crystal system
Space group
C19H25Cl2NO2Ru
471.37
0.71069
[C19H25ClN2ORu]2 Æ 0.5H2O
876.86
0.71069
Tetragonal
I41
˚
Triclinic
ꢀ
P1
Unit cell dimensions
˚
a (A)
9.7100(9)
9.7898(9)
10.772(1)
23.701(2)
14.572(1)
˚
b (A)
˚
c (A)
a (°)
b (°)
c (°)
97.595(2)
97.369(2)
104.982(2)
966.28(15)
3
˚
Volume (A )
8186(1)
Z
2
8
Dcalc (Mg/m3)
1.620
1.100
1.423
0.905
Absorption coefficient (mmꢀ1
)
F(000)
480
3592
h Range for data collection (°)
Reflections collected
1.94–30.38
13585
1.21–23.29
36794
Data/restraints/parameters
Goodness-of-fit on F2
5254/0/238
1.038
5917/9/438
1.163
Final R indices [I > 2r(I)]
R indices (all data)
Largest DF maximum/minimum (e A
R1 = 0.0209, wR2 = 0.0534
R1 = 0.0244, wR2 = 0.0549
0.481/ꢀ0.455
R1 = 0.0359, wR2 = 0.1001
R1 = 0.0520, wR2 = 0.1182
0.909/ꢀ0.306
ꢀ3
˚
)
Cambridge Crystallographic Database [24] facilities was
made for structure discussion. Data collection and
refinement results are summarized in Table 3.
References
[1] K. Severin, R. Bergs, W. Beck, Angew. Chem. Int. Ed. 37 (1998)
1634.
[2] (a) R. Kra¨mer, M. Maurus, K. Polborn, K. Su¨nkel, C. Robl, W.
Beck, Chem. Eur. J. 2 (1996) 1518;
7. Hydrogen transfer reactions
(b) K. Haas, W. Beck, Eur. J. Inorg. Chem. (2001) 2485.
[3] R. Fish, G. Jaouen, Organometallics 22 (2003) 2166.
[4] (a) F. Joo´, E. Tro´csa´nyi, J. Organomet. Chem. 231 (1982) 63;
(b) D.A. Laidler, D.J. Molner, J. Organomet. Chem. 270 (1984)
121;
0.05 mmol of 2 was dissolved in 5 ml of i-PrOH and
the solution thermostated at the desired temperature
(room temperature or 80 °C). An i-PrOH solution (5
ml) of the ketone (50 mmol) was added and after a hour
an i-PrOH solution (5 ml) of t-BuOK (0.1 mmol, 11 mg)
was added. After an additional hour of reaction a small
portion of the reactant solution was withdrawn,
quenched with water, extracted with diethyl ether, eluted
through a short silica column with diethyl ether and fi-
nally analyzed by GC.
(c) M. Nakagawa, H. Nakao, K. Watanabe, Chem. Lett. (1985)
391;
(d) S. Colonna, A. Manfredi, M. Spadoni, L. Casella, M. Gullotti,
J. Chem. Soc., Perkin Trans. (1987) 71;
(e) A. Mori, H. Abe, S. Inoue, Appl. Organomet. Chem. 9 (1995)
189;
(f) A. Saitoh, K. Achiwa, K. Tanaka, T. Morimoto, J. Org.
Chem. 65 (2000) 4227.
[5] (a) A. Mori, H. Ohno, H. Nitta, K. Tanaka, S. Inoue, Synlett
(1991) 563;
(b) M.I. Burguete, M. Collado, J. Escorihuela, F. Galindo, E.
´
Garcıa-Verdugo, S.V. Luis, M.J. Vicent, Tetrahedron Lett. 44
8. Supplementary material
(2003) 6891;
(c) A.V. Malkov, J.B. Hand, P. Kocˇovsky´, Chem. Commun.
(2003) 1948.
[6] (a) J.A. McClarin, L.A. Dressel, J.I. Legg, J. Am. Chem. Soc. 98
(1976) 4150;
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Center, CCDC No. 270629 for compound 1 and
CCDC No. 270628 for compound 2.
(b) P. Norman, D.A. Phipps, Inorg. Chim. Acta 17 (1976) L19–
L20;
(c) Y. Ilan, Inorg. Chem. 26 (1987) 2454.
[7] S.-Y. Han, Y.-A. Kim, Tetrahedron 60 (2004) 2447.
[8] W.S. Sheldrick, S. Heeb, Inorg. Chim. Acta 168 (1990) 93.
[9] (a) H. Brunner, M. Weber, M. Zabel, Coord. Chem. Rew. 242
(2003) 3;
Acknowledgment
`
We thank the C.I.M. (Centro Interfacolta di Misure,
‘‘Giuseppe Casnati’’) of the University of Parma for the
instrument facilities.
(b) H. Brunner, M. Weber, M. Zabel, T. Zwack, Angew. Chem.
Int. Ed. 42 (2003) 1859.

4610
P. Pelagatti et al. / Journal of Organometallic Chemistry 690 (2005) 4602–4610
[10] R. Kra¨mer, M. Maurus, R. Bergs, K. Polborn, K. Sunkel, B.
¨
[16] (a) T. Ohta, S. Nakahara, Y. Shigemura, K. Hattori, I. Furukawa,
Chem. Lett. (1998) 491;
Wagner, W. Beck, Chem. Ber. 126 (1993) 1969.
[11] H. Brunner, T. Neuhierl, B. Nuber, Eur. J. Inorg. Chem. (1998)
(b) T. Ohta, S. Nakahara, Y. Shigemura, K. Hattori, I. Furukawa,
Appl. Organomet. Chem. 15 (2001) 699;
1877.
[12] D. Zuccaccia, E. Clot, A. Macchioni, New J. Chem. 29 (2005)
(c) D. Carmona, F.J. Lahoz, R. Atencio, L.A. Oro, M.P. Lamata,
´
´
´
430.
F. Viguri, E.S. Jose, C. Vega, J. Reyes, F. Joo, A. Katho, Chem.
[13] (a) S. Hashiguchi, A. Fujii, J. Takehara, T. Ikariya, R. Noyori, J.
Am. Chem. Soc. 117 (1995) 7562;
Eur. J. 5 (1999) 1544;
(d) A. Katho´, D. Carmona, F. Viguri, C.D. Remacha, J. Kova´cs,
´
(b) R. Noyori, S. Hashiguchi, Acc. Chem. Res. 30 (1997) 97;
(c) K. Matsumara, S. Hashiguchi, T. Ikariya, R. Noyori, J. Am.
Chem. Soc. 119 (1997) 8738;
(d) K. Murata, T. Ikariya, R. Noyori, J. Org. Chem. 64 (1999)
2186;
F. Joo, L.A. Oro, J. Organomet. Chem. 593–594 (2000) 299.
[17] (a) D. Carmona, M.P. Lamata, F. Viguri, I. Dobrinovich, F.J.
Lahoz, L.A. Oro, Adv. Synth. Catal. 344 (2002) 499;
(b) D. Carmona, M.P. Lamata, L.A. Oro, Eur. J. Inorg. Chem.
(2002) 2239.
(e) M.J. Palmer, M. Wills, Tetrahedron: Asymmetry 10 (1999)
2045.
[18] M.A. Bennett, T.N. Huang, T.W. Matheson, A.K. Smith, Inorg.
Synth. 21 (1982) 74.
[14] (a) J.W. Faller, A.R. Lavoie, Orgnometallics 20 (2001) 5245;
(b) H.Y. Rhyoo, H.J. Park, Y.K. Chung, Chem. Commun. (2001)
2064;
[19] SAINT: SAX, Area Detector Integration, Siemens Analytical
instruments Inc., Madison, WI, USA.
[20] G. Sheldrick, SADABS: Siemens Area Detector Absorption Correc-
tion Software, University of Goettingen, Germany, 1996.
[21] A. Altomare, M.C. Burla, M. Camalli, G. Cascarano, C.
Giacovazzo, A. Guagliardi, A.G. Moliterni, G. Polidori, R.
Spagna, SIR97: A New Program For Solving And Refining Crystal
Structures, Istituto di Ricerca per lo Sviluppo di Metodologie
Cristallografiche, C.N.R., Bari, Italy, 1997.
(c) H.Y. Rhyoo, Y.A. Yoon, H.J. Park, Y.K. Chung, Tetrahedron
Lett. 42 (2001) 5045;
(d) H.Y. Rhyoo, H.J. Park, W.H. Suh, Y.K. Chung, Tetrahedron
Lett. 43 (2002) 269.
[15] (a) I.M. Pastor, P. Va¨stila¨, A. Adolfsson, Chem. Commun. (2002)
2046;
(b) I.M. Pastor, P. Va¨stila¨, A. Adolfsson, Chem. Eur. J. 9 (2003)
4031;
[22] G. Sheldrick, SHELXL97. Program for Structure Refinement,
University of Goettingen, Germany, 1997.
(c) A. Bøgevig, I.M. Pastor, H. Adolfsson, Chem. Eur. J. 10
(2004) 294.
[23] J.L. Farrugia, J. Appl. Crystallogr. 32 (1999) 837.
[24] F.H. Allen, O. Kennard, R. Taylor, Acc. Chem. Res. 16 (1983) 146.