Chloro(η6-p-cymene)(phosphoramidite)ruthenium(II), a 16-electron fragment stabilized by an η2-aryl-metal interaction, and its use in asymmetric cyclopropanation

Article abstract of DOI:10.1002/hlca.200690168

Chloride abstraction from the half-sandwich complexes [RuCl ₂(η⁶-p-cymene)(P*-κP)] (2a: P=(S_a,R,R)-1a=(1S_a)-[1,1′-binaphthalene]-2, 2′-diyl bis[(1R)-1-phenylethyl)]phosphoramidite; 2b: P=(S _a,R,R)-1b=(1S_a)-[1,1′-binaphthalene]-2, 2′-diyl bis[(1R)-(1-(1-naphthalen-1-yl)ethyl]phosphoramidite) with (Et₃O)[PF₆] or Tl[PF₆] gives the cationic, 18-electron complexes dichloro(η⁶-p-cymene){(1S _a)-[1,1′-binaphthalene]-2,2′-diyl {(1R)-1-[(1,2-n)- phenyl]ethyl}[(1R)-1-phenylethyl]phosphoramidite-κP}ruthenium(II) hexafluorophosphate (3a) and [Ru(S)]-dichloro(η⁶-p-cymene) {(1S_a)-[1,1′-binaphthalene]-2,2′-diyl {(1R)-(1-(1,2-η)-naphthalen-1-yl]ethyl}[(1R)-1-(naphthalen-1-yl)ethyl] -phosphoramidite-κP)ruthenium(II) hexafluorophosphate (3b), which feature the η²-coordination of one aryl substituent of the phosphoramidite ligand, as indicated by ¹H-, ¹³C-, and ³¹P-NMR spectroscopy and confirmed by an X-ray study of 3b. Additionally, the dissociation of p-cymene from 2a and 3a gives dichloro{(1S_a)-[1,1′-binaphthalene]-2,2′-diyl [(1R)-(1-(η⁶-phenyl)ethyl][(1R)-1-phenylethyl] phosphoramidite-κP)ruthenium(II) (4a) and di-μ-chlorobis((1S _a)-[1,1′-binaphthalene]-2,2′-diyl [(1R)-1- (η⁶-phenyl)ethyl][(1R)-1-phenylethyl]phosphoramidite-κP} diruthenium(II) bis(hexafluorophosphate) (5a), respectively, in which one phenyl group of the N-substituents is η⁶-coordinated to the Ru-center. Complexes 3a and 3b catalyze the asymmetric cyclopropanation of α-methylstyrene with ethyl diazoacetate with up to 86 and 87% ee for the cis- and the trans-isomers, respectively.

Full text of DOI:10.1002/hlca.200690168

1696
Helvetica Chimica Acta – Vol. 89 (2006)
Chloro(h6-p-cymene)(phosphoramidite)ruthenium(II), a 16-Electron
Fragment Stabilized by an h2-Aryl–Metal Interaction, and Its Use in
Asymmetric Cyclopropanation
by Dominik Huber, P. G. Anil Kumar, Paul S. Pregosin, Igor S. Mikhel, and Antonio Mezzetti*
Department of Chemistry and Applied Biosciences, ETH Zurich, CH-8093 Zurich
(phone: +41446326121; fax: +41446321310; e-mail: mezzetti@inorg.chem.ethz.ch)
Dedicated to Professor Giambattista Consiglio on the occasion of his 65th birthday
Chloride abstraction from the half-sandwich complexes [RuCl2(h6-p-cymene)(P*-kP)] (2a:
P*=(Sa,R,R)-1a=(1Sa)-[1,1’-binaphthalene]-2,2’-diyl bis[(1R)-1-phenylethyl)]phosphor
P*=(Sa,R,R)-1b=(1Sa)-[1,1’-binaphthalene]-2,2’-diyl bis[(1R)-(1-(1-naphthalen-1-yl)ethyl]phosphor-
amidite) with (Et3
O)[PF6] or Tl[PF6] gives the cationic, 18-electron complexes dichloro(h6-p-
cymene){(1Sa)-[1,1’-binaphthalene]-2,2’-diyl {(1R)-1-[(1,2-h)-phenyl]ethyl}[(1R)-1-phenylethyl]phos-
phoramidite-kP}ruthenium(II) hexafluorophosphate (3a) and [Ru(S)]-dichloro(h6-p-cymene){(1Sa)-
[1,1’-binaphthalene]-2,2’-diyl {(1R)-1-[(1,2-h)-naphthalen-1-yl]ethyl}[(1R)-1-(naphthalen-1-yl)ethyl]-
amidite; 2b:
AHCTREUNG
phosphoramidite-kP)ruthenium(II) hexafluorophosphate (3b), which feature the h2-coordination of
one aryl substituent of the phosphoramidite ligand, as indicated by 1H-, 13C-, and 31P-NMR spectroscopy
and confirmed by an X-ray study of 3b. Additionally, the dissociation of p-cymene from 2a and 3a gives
dichloro{(1Sa)-[1,1’-binaphthalene]-2,2’-diyl [(1R)-(1-(h6-phenyl)ethyl][(1R)-1-phenylethyl]phosphora-
midite-kP)ruthenium(II) (4a) and di-m-chlorobis{(1Sa)-[1,1’-binaphthalene]-2,2’-diyl [(1R)-1-(h6-phe-
nyl)ethyl][(1R)-1-phenylethyl]phosphoramidite-kP}diruthenium(II) bis(hexafluorophosphate) (5a),
G
respectively, in which one phenyl group of the N-substituents is h6-coordinated to the Ru-center. Com-
plexes 3a and 3b catalyze the asymmetric cyclopropanation of a-methylstyrene with ethyl diazoacetate
with up to 86 and 87% ee for the cis- and the trans-isomers, respectively.
Introduction. – After Henri Brunner’s seminal studies of [Mn(Cp)(CO)(PPh3)-
(NO)] (Cp=cyclopenta-2,4-dien-1-yl) [1], pseudotetrahedral complexes containing
chirotopic metal atoms have been extensively investigated over the last 35 years,
with particular emphasis on stereochemical aspects. As configurational stability is
required for these studies, octahedral complexes of d6 ions – mainly MnI [1], ReI [2],
RuII [3a,b] (for seminal papers concerning [RuCl(Cp)(PꢀP*)] complexes, see also
[3c,d], [4a], for application as stoichiometric reagents in organic chemistry, see
[4b,c]), RhIII [5], and IrIII [5] – have been used for the synthesis of substitutionally
inert 18-electron species with stereogenic metal atoms (for the extension of this concept
to formally seven-coordinate d4 metal ions, see [6]). With RuI ions, two major classes of
compounds have been studied: [RuCl(Cp)(PꢀP*)] (PꢀP*=chiral diphosphine) [3]
and [RuX(h6-arene)(LꢀL)]+ [4].
Despite their long history, the application of chiral half-sandwich ruthenium com-
plexes in asymmetric catalysis has been mainly restricted to hydrogenation [7][8] and
Diels–Alder reactions [4c]. For the latter, Kündig and co-workers have developed com-
© 2006 Verlag Helvetica Chimica Acta AG, Zürich

Helvetica Chimica Acta – Vol. 89 (2006)
1697
plexes based on the 16-electron fragment [M(Cp)(PꢀP*)]+ that contain an electron-
poor chiral diphosphine PꢀP* and act as mild and highly selective Lewis acids [9].
On the same lines, Davies and co-workers (selected examples in [10]), Oro and co-
workers (for selected references, see [11]), and Faller and co-workers (selected exam-
ples in [12]) have independently studied complexes of the type [RuCl(h6-p-cyme-
ne)(LꢀL)]+ (LꢀL=PꢀO-, PꢀN-, NꢀN-, and P,S-hybrid ligands), mainly as Diels–
Alder catalysts. A general structural feature of these complexes is the use of a chiral
bidentate ligand (LꢀL*) to control the stereochemistry at the metal during the synthe-
sis (or resolution of diastereoisomeric products), upon ligand substitution, and in cat-
alysis.
We have been investigating the possibility of controlling the absolute configuration
at the metal center by means of a monodentate chiral ligand. Although a few (arene)-
ruthenium complexes containing chiral, monodentate P-donor ligands have been
reported [13], monodentate stereogenic ligands have not been used in ruthenium
half-sandwich precatalysts yet. Conceptually related approaches are the use of hemila-
bile bidentate ligands (e.g., diphosphine monooxides) [12c] or of P-donors that are teth-
ered to the arene or cyclopentadienyl ligand [14][12e,f]. It should be noted that the use
of a monodentate chiral ligand in pseudotetrahedral half-sandwich complexes opens
new potential applications for the large number of recently developed, chiral monoden-
tate phosphoramidite ligands [15a] (for recent applications in asymmetric catalysis, see,
e.g., [15b–j]).
As a model reaction, we chose the asymmetric cyclopropanation of olefins, which
requires the formation of a diastereoisomerically enriched carbene intermediate, e.g.,
[RuCl(=CHR)(h6-p-cymene)(P*)]+ (P*=chiral phosphoramidite). This species has a
precedent in diastereoisomerically pure [Fe(Cp)(=CHMe)(CO)(P*)]+ reported by
Brookhart and co-workers, where P* is a chiral monodentate phosphine [16]. More-
over, Hossain and co-workers have recently reported a diastereo- and enantioselective
stoichiometric carbene transfer from analogous complexes to olefins [17]. In contrast,
examples of catalytic cyclopropanation of olefins catalyzed by iron and ruthenium
half-sandwich complexes are restricted to achiral systems. Examples thereof are
[Fe(Cp)(CO)
C
G
G
A
In preliminary reports, we have shown that [RuCl2(h6-p-cymene)(P*)] (2; p-cym-
ene=1-methyl-4-(1-methylethyl)benzene;
P*=(Sa,R,R)-1a=(1Sa)-[1,1’-binaphtha-
lene]-2,2’-diyl bis[(1R)-1-phenylethyl]phosphoramidite; P*=(Sa,R,R)-1b=(1Sa)-[1,1’-
binaphthalene]-2,2’-diyl bis[(1R)-1-(naphthalen-1-yl)ethyl]phosphoramidite)1) cata-
lyze the cyclopropanation of a-methylstyrene (=(1-methylethenyl)benzene) with
ethyl diazoacetate after chloride abstraction by (Et3O)[PF6] or Tl[PF6] with an enantio-
A
selectivity of up to 86 and 87% ee for the cis- and trans-cyclopropane derivatives,
respectively [22]. The complex formed upon chloride abstraction from 2b shows an
unexpected (h2-arene)–metal interaction [23]. In the present paper, besides a full
account of the products of chloride abstraction from [RuCl2(h6-p-cymene)(P*)], we
1
)
For systematic names, see Exper. Part.

1698
Helvetica Chimica Acta – Vol. 89 (2006)
describe the products of the dissociation of p-cymene from 2a and 3a, in which the coor-
dination of the dangling aryl moiety of the P* ligand is turned from h2 to h6, as well as
additional catalytic experiments.
Results and Discussion. – [RuCl2(h6-p-cymene)(P*-kP)]. Ligand 1a readily reacts
with [{RuCl2(h6-p-cymene)}2] in CH2Cl2 at room temperature to give [RuCl2(h6-p-cym-
ene)(1a-kP)] (2a) in 93% yield (Scheme). The X-ray structure determination [22] of 2a
shows a relatively undistorted coordination geometry at the Ru center (Table 1, Fig. 1).
Although the interpretation of such a half-sandwich complex as pseudotetrahedral is
practical in terms of stereochemical understanding, the angles at Ru show that the coor-
dination is more accurately described as octahedral. The RuꢀP distance of 2.317(3) Å
falls at the upper end of the range found for similar complexes. Thus, it is longer than in
the cationic phosphoramidite (P) complex [Ru(Cp)(CO)
A
the neutral [RuCl(Cp)((PhO)2PN(Me)N(Me)P(OPh)2)], which contains a bidentate
AHCTREUNG
phosphoramidite ligand (2.2129(6) and 2.1905(6) Å) [25], probably because of the che-
late effect.
Scheme. Synthesis of 2a and 2b
Ligand 1b reacts with [RuCl2(h6-p-cymene)]2 to give [RuCl2(h6-p-cymene)(1b-kP)]
(2b) in 49% yield. A small amount of free ligand 1b (ca. 4% of total) was present in the
CD2Cl2 solutions of isolated 2b, as indicated by the 31P-NMR signal at d 150.0. The 1H-
NMR spectrum indicated that an equivalent amount of [{RuCl2(h6-p-cymene)}2] was
present as impurity, suggesting that the bulky phosphoramidite 1b dissociates from

Helvetica Chimica Acta – Vol. 89 (2006)
1699
Table 1. Bond Distances [Å] and Angles [8] in 2a
RuꢀCl(1)
RuꢀP
2.405(3)
2.317(3)
2.201(12)
2.213(11)
2.170(10)
RuꢀCl(2)
PꢀN(1)
2.387(3)
1.681(7)
2.218(9)
2.197(11)
2.206(10)
RuꢀC(1)
RuꢀC(3)
RuꢀC(5)
RuꢀC(2)
RuꢀC(4)
RuꢀC(6)
Cl(1)ꢀRuꢀCl(2)
Cl(1)ꢀRuꢀC(3)
Cl(1)ꢀRuꢀC(4)
PꢀRuꢀC(5)
85.50(10)
87.8(3)
89.0(3)
93.5(3)
91.84(10)
Cl(2)ꢀRuꢀP
89.06(9)
90.5(3)
89.8(3)
92.7(3)
Cl(2)ꢀRuꢀC(1)
Cl(2)ꢀRuꢀC(2)
PꢀRuꢀC(6)
Cl(1)ꢀRuꢀP
Fig. 1. ORTEP plot of (Sa,R,R)-2a. 30%-Probability ellipsoids. Arbitrary atom numbering.
2b in solution. Indeed, the amount of free ligand 1b increased with time and reached
19% of the total integrated intensity after 7 h in solution. As additional, unidentified
products were formed, whose 31P-NMR signals accounted for 14% of the total inte-
grated intensity, the dissociation reaction was not investigated further.
The above observations indicate that complexes [RuCl2(h6-p-cymene)(P*-kP)] are
relatively unstable, following different decomposition pathways depending on the
nature of the phosphoramidite ligand. Complex 2a preferentially dissociates p-cymene
(see below), whereas 2b loses the phosphoramidite ligand. As the dichloro complexes
2a and 2b are catalytically inactive, we concentrated our efforts on the products of chlo-
ride abstraction, though.
[RuCl(h6-p-cymene)(1,2-h-P*-kP)]PF6. Complex 2a reacted with (Et3
with Tl[PF6]) in CH2Cl2 to give the yellow complex [RuCl(h6-p-cymene)(1,2-h-1a-
A

1700
Helvetica Chimica Acta – Vol. 89 (2006)
kP)][PF6] (3a) featuring a s at d 154.6 in the 31P-NMR spectrum, based on the NMR
spectroscopic studies described below. Complex 3a slowly converts to a new species
both in the solid state and in solution (see below), which precluded X-ray analysis.
The reaction of 2b either with (Et3O)[PF6] or with Tl[PF6] (1.1 equiv.) gave a yellow
G
complex 3b that showed a broadened s at d 168.0 in the 31P-NMR spectrum. Complex
3b analyzes as [RuCl(h6-p-cymene)(1b)][PF6] and is stable in solution for (at least) 3 d
and in the solid state for (at least) 7 weeks. Assuming that the phosphoramidite ligand
acts as a two-electron donor, the formulation of 3a and 3b as [RuCl(h6-p-cymene)-
(P*-kP)][PF6] would imply a 16-electron count and coordinative unsaturation at RuII.
However, five-coordinate, 16-electron RuII complexes of the type [RuX(Cp*)(PiPr2-
ACHTREUNG
AHCTREUNG
clear 2D-NMR measurements indicated that 3a and 3b are mononuclear six-coordinate
complexes in which the coordination sphere of the 16-electron fragment [RuCl(h6-p-
cymene)(P*-kP)]+ is saturated by means of an h2 interaction between the RuII atom
and the phenyl or naphthalenyl ring of one of the ArCH(Me) substituents at the N-
atom (Ar=Ph or Np), which was confirmed by an X-ray study in the case of 3b.
The line widths for several proton resonances in the aromatic region of the room-
temperature 1H- and 13C-NMR spectra of both 3a and 3b indicated a dynamic behavior
on the NMR time scale. Upon cooling the samples to ꢀ208, the signals sharpened suf-
ficiently to measure one-bond and long-range 13C,1H correlations. Under these condi-
tions, six resolved signals were observed for the phenyl C-atoms of one of the
PhCH(Me)N groups of 3a, indicating restricted rotation, whereas the phenyl group
of the other PhCH(Me)N group is freely rotating. An HMBC experiment allowed us
to attribute the 13C-NMR signal of C(1) at d 120.0 in 3a (101.1 in 3b) based on the
three-bond interactions to HꢀC(3) and HꢀC(5), respectively (Fig. 2). Analogously,
HꢀC(4) and HꢀC(6) correlate to C(2) at d 105.8 for 3a (d 96.7 for 3b) (Table 2).
The low-frequency shifts Dd of the C(1) and C(2) signals of 3a indicate that one of
the diastereotopic phenyl rings is h2-bonded to Ru. The magnitudes of the coordination
chemical shifts Dd (24.3 and 22.4 ppm for C(1) and C(2), resp.) of 3a indicate a modest-
to-weak p-olefin-type complex [27]. The second, noncomplexed phenyl ring has 13C-
NMR chemical shifts in the normal aromatic region and shows two equivalent ortho
(and meta) H- and C-atoms, which is indicative of free rotation at this temperature.
The naphthalenyl derivative 3b exhibits analogous features, with the C(1) and C(2)
atoms of the naphthalenyl group resonating at d 101.1 and 96.7, respectively.

Helvetica Chimica Acta – Vol. 89 (2006)
1701
Fig. 2. 13C,1H Long-range correlation from a heteronuclear multibond correlation (HMBC) spectrum
of 3a. For atom numbering, see formula.
13
Table 2. Selected C-NMR Chemical Shift Values d [ppm] for 2a, 3a, 3b, 4a, and 5aa)
2a
3a
3b
4a
5a
C(1)
C(2)
C(3)
C(4)
C(5)
C(6)
144.3
128.2
127.7
126.4
120.0
105.8
136.4
131.3
130.7
134.4
101.1
96.7
106.6
80.6
104.9
95.0
97.8
81.0
106.0
81.6
105.5
99.0
90.8
82.1
131.8
132.6
b
ca. 127.7
ca. 128.2
)
b
)
a) The numbering scheme is given in the formulae. The corresponding signals of 2b were not attributed
because they overlap with those of the ligand naphthalenyl groups. b) The 13C-NMR chemical shift values
are in the region d 128–132.
Molecular-volume data obtained by NMR spectroscopy (pulse-gradient spin-echo
(PGSE) diffusion measurements) (selected examples in [28]) confirm that the cationic
complexes 3a and 3b are mononuclear, as their diffusion coefficients are similar to
those of the neutral dichloro complexes 2a and 2b (Table 3). Additionally, the compar-
ison of the hydrodynamic radii of the anion and cation gives information concerning the
extent of ion pairing in solution. Thus, for 100% ion-pair formation, the rh values for the
cation and anion would be identical [28h]. Considerable ion pairing (>50%) occurs

1702
Helvetica Chimica Acta – Vol. 89 (2006)
Table 3. Diffusion Coefficients [·10ꢀ10 m2 sꢀ1] and Hydrodynamic Radii [Å] in CD2Cl2
)
a
D
rh
1a
10.06
5.3
6.2
6.6
5.4
8.2
6.3
5.7
6.5
6.6
4.5
[RuCl2(L)(1a-kP)] (2a)
[RuCl(L)((1,2-h)-1a-kP)][PF6] (3a)
8.66
8.12
9.83
6.55
8.45
9.44
8.21
8.18
11.80
cation
anion
cation
anion
[Ru2Cl2(h6-1a-kP)2][PF6]2 (5a)
1b
[RuCl2(L)(1b-kP)] (2b)
[RuCl(L)((1,2-h)-1b-kP)][PF6] (3b)
cation
anion
a) L is h6-p-cymene.
between the complex cation 3a and [PF6]ꢀ (at 2 mM concentration in CD2Cl2), as indi-
cated by the hydrodynamic radius rh of 5.4 Å observed for the [PF6]ꢀ counterion, which
is much larger than for freely diffusing hexafluorophosphate (2.6 Å) [28d]. The ion
pairing in the naphthalenyl analog 3b is less pronounced (rh(PF6)=4.5 Å).
A
Finally, it should be noted that the pseudotetrahedral Ru-atom of 3a and 3b is ster-
eogenic, whereas this is not the case in 2a and 2b. The 31P-, 1H-, and 13C-NMR spectra
indicated that 3a and 3b are formed as a single diastereoisomer. The absolute configu-
ration at the Ru center of 3b was unambiguously determined by an X-ray study.
X-Ray Structure of rac-3b. After several unsuccessful attempts to crystallize 3b con-
taining enantiomerically pure (Sa,R,R)-1b, high-quality crystals were obtained with the
racemic complex rac-3b, obtained by mixing (Sa,R,R)-3b and (Ra,S,S)-3b in a 1:1 ratio.
The crystal was made up of pairs of discrete cations (SRu,Sa,R,R)-3b and (RRu,Ra,S,S)-3b
(related by an inversion center), and of [PF6]ꢀ anions with normal nonbonded distances.
With ligand (Sa,R,R)-1b, the configuration at the Ru center is (S) (Fig. 3) [29]. Besides
the chloro ligand and the P-atom of 1b, the Ru center is coordinated in a h6 fashion to
p-cymene and in a h2 fashion to the naphthalenyl group of one ArCH(Me)N moiety of
the phosphoramidite to give an 18-electron complex.
The most interesting structural feature is obviously the h2 coordination of the naph-
thalenyl moiety to the Ru center, as indicated by the RuꢀC(1) and RuꢀC(2) distances
of 2.379(2) and 2.386(2) Å, respectively (Table 4). As an effect of coordination to Ru,
the C(1)ꢀC(2) distance of 1.407(3) Å is longer than the corresponding separation in the
noncoordinated naphthalenyl moiety (C(15)ꢀC(16), 1.375(3) Å). Additionally, the pat-
tern of the alternating CꢀC bond distances suggests extensive loss of aromaticity of the
h2-coordinated ring as compared to the noncoordinated one. The distal ring is nearly
not affected, most distances being the same within experimental error. As a conse-
quence of the Ruꢀh2-aryl bond, the C(1) atom is pyramidalized to some extent, as indi-
cated by the sum of the C(2)ꢀC(1)ꢀC(10), C(2)ꢀC(1)ꢀC(11), and C(10)ꢀC(1)ꢀC(11)
angles of 353.68. However, nonbonded interactions between the naphthalenyl ring and
the cymene ligand probably contribute to this disortion, as suggested by the shortest
contact of 3.114(4) Å between C(3) and C(53).

Helvetica Chimica Acta – Vol. 89 (2006)
1703
Fig. 3. ORTEP plot of (SRu,Sa,R,R)-3b. 30%-Probability ellipsoids. Arbitrary atom numbering.
Table 4. Bond Distances [Å] and Angles [8] in 3b
RuꢀCl(1)
RuꢀC(1)
2.3926(6)
2.379(2)
2.258(2)
2.208(2)
2.221(2)
1.407(3)
1.442(3)
1.472(3)
1.334(4)
1.433(4)
1.417(3)
1.401(3)
RuꢀP(1)
2.2783(5)
2.386(2)
2.216(2)
2.306(2)
2.298(2)
1.375(3)
1.409(4)
1.432(2)
1.355(4)
1.419(4)
1.421(3)
1.418(3)
RuꢀC(2)
RuꢀC(46)
RuꢀC(48)
RuꢀC(54)
C(1)ꢀC(2)
C(2)ꢀC(3)
C(1)ꢀC(10)
C(3)ꢀC(4)
C(4)ꢀC(5)
C(5)ꢀC(10)
C(9)ꢀC(10)
RuꢀC(47)
RuꢀC(49)
RuꢀC(53)
C(15)ꢀC(16)
C(16)ꢀC(17)
C(15)ꢀC(24)
C(17)ꢀC(18)
C(18)ꢀC(19)
C(19)ꢀC(24)
C(23)ꢀC(24)
Cl(1)ꢀRuꢀC(1)
Cl(1)ꢀRuꢀP(1)
P(1)ꢀRuꢀC(1)
P(1)ꢀRuꢀC(47)
C(1)ꢀRuꢀC(54)
93.11(5)
81.066(19)
72.05(5)
90.82(6)
97.44(8)
Cl(1)ꢀRuꢀC(2)
Cl(1)ꢀRuꢀC(49)
P(1)ꢀRuꢀC(2)
C(53)ꢀRu-Cl(1)
C(2)ꢀRuꢀC(53)
84.04(6)
88.88(6)
103.25(6)
113.11(6)
85.14(8)
At difference with other h2 interactions in Ru complexes reported so far [30–32],
the bonding in 3b involves a ‘dangling’ aryl group and not the chelate ring of a diphos-
phine. To the best of our knowledge, the only other example of h2 interaction between a
Ru-atom and a dangling aryl, besides 3b, has been found in a heptanuclear Ru cluster in
which a phenyl group at a m5-bridging alkyne loosely coordinates to a Ru-atom (2.42(1)

1704
Helvetica Chimica Acta – Vol. 89 (2006)
and 2.48(1) Å) [33]. Compared to these values, the RuꢀC(1) and RuꢀC(2) distances of
2.379(2) and 2.386(2) Å in 3b are considerably shorter, but longer than in (olefin)ruthe-
nium(II) complexes (average below 2.2 Å) [34] or in ruthenium(II) complexes contain-
ing the atropisomeric biarene diphosphines [2’-(diphenylphosphino)[1,1’-binaphtha-
len]-2-yl]diphenylphosphine oxide (2.255(4) and 2.346(4) Å) [31] and (6,6’-dime-
thoxy[1,1’-biphenyl]-2,2’-diyl)bis[diisopropylphosphine] (2.299(5) and 2.366(5) Å)
[30b]. However, the latter Ru–aryl interactions are enhanced by the chelate effect as
they involve a biaryl diphosphine bridge.
Although there is increasing circumstantial evidence that the coordination of pend-
ant aromatic rings may play an important role in transition-metal catalysts by stabiliz-
ing metal complexes with a low electron count, thoroughly documented examples are
still rare. Interestingly, an h2 interaction between ligand 1a and the metal has been pre-
dicted by calculation for nickel hydrovinylation catalysts, in which it stabilizes the hy-
dride alkene intermediate [NiH(styrene)(1,2-h-1a-kP)] [35]. Our results may help
explain the remarkable influence of the formal secondary-amine appendage (N-sub-
stituent) observed in the same reaction [36]. An analogous interaction involving a
binaphthalene group has been claimed for cationic (monophosphine)palladium(II)
complexes [37][38], but it is still controversial whether the coordination mode is h2
or h1 [39]. On similar lines, the cyclometalation of the phosphoramidite at the phen-
ethyl Me group has been observed in iridium(I) complexes [40][41], which offers an
alternative possibility of turning a ‘monodentate’ phosphoramidite into a chelating
ligand.
Loss of p-Cymene. As mentioned above, the derivatives containing ligand 1a (2a,
and 3a), are prone to lose p-cymene both in solution and in the solid state. Complex
2a slowly dissociates p-cymene upon standing for several days in CD2Cl2. The 31P-
NMR spectra of the reaction solution indicate that dichloro{(1Sa)-[1,1’-binaphtha-
lene]-2,2’-diyl [(1R)-1-(h6-phenyl)ethyl][(1R)-1-phenylethyl]phosphoramidite-kP}ru-
thenium(II)1) (4a), featuring a s at d 150.9, is slowly formed. At room temperature,
quantitative conversion of 2a to 4a required about one month, after which pure 4a
was isolated by evaporation of the solvent under reduced pressure. Besides MALDI
1
mass spectrometry and elemental analysis, 31P-, H-, 13C-NMR spectroscopy, 31P,1H-
1
1
HMQC, 13C,1H-HMQC, H,1H-COSY, and H,1H-NOESY 2D-NMR measurements
(Exper. Part and Table 2) were used to characterize 4a with a procedure analogous to
that described for 5a (see below).
Similarly, on standing in CD2Cl2 solution, 3a is converted to 5a, which features a s at
d 150.4 in the 31P-NMR spectrum. The reaction is quantitative after 16 d at room tem-
perature or within 17 h at 408. The process also takes place in the solid state, albeit on a

Helvetica Chimica Acta – Vol. 89 (2006)
1705
longer time scale (2–3 months, room temperature, under Ar). Attempts to obtain pure
5a by repeated recrystallizations of 3a gave samples that were spectroscopically but not
analytically pure. Finally, 3a was heated in the solid state under high vacuum (<5·10ꢀ3
mbar) at 1508 for 77 h. Mass spectroscopy and elemental analysis indicate the empirical
formula [RuCl(1a)][PF6] for 5a. The same combination of NMR multinuclear and
PGSE methods used for 3a and 3b indicated that 5a is the binuclear species
[Ru2Cl2(h6-1a-kP)2]2+, in which the phenyl ring of one PhCH(Me)N moiety binds the
1
Ru-atom in an h6 fashion. H- and 13C-NMR Data confirmed the absence of the h6-
p-cymene and that only one phenyl ring is freely rotating.
The 31P, 1H-HMQC correlation allowed the assignement of the different
PhCH(Me)N methine H- and C-atoms of 5a and, thus, with the help of NOEs, to con-
nect these to the corresponding coordinated and free phenyl rings. As expected, a set of
six 13C-NMR signals from one of the PhCH(Me)N phenyl rings is displaced to lower
frequency (Table 2, Fig. 4), thus confirming an h6 complexation to Ru. Distinct patterns
are observed for the two PhCH(Me)N methine H-atoms. The one belonging to the h6-
complexed moiety appears as a dq (d 4.14, 3J(P,H)=42.7 Hz), whereas that of the dan-
gling PhCH(Me)N group is observed as a m (d 4.70, 3J(P,H)=7.0 Hz). These very dif-
3
ferent J(P,H) values reflect the different conformations of the PhCH(Me)N groups.
The PGSE measurements for the cation 5a in CD2Cl2 show a large hydrodynamic
radius rh of 8.2 Å, which is consistent with a binuclear structure [28a,b] (Table 3).
Again, we note very substantial ion pairing between cation 5a and [PF6]ꢀ in that the
rh value 6.3 Å is much larger than the hydrodynamic radius of [PF6]ꢀ in MeOH (2.6
Å) [28d].
In contrast to 3a, 3b remained substantially unchanged for 3 d in CD2Cl2 at room
temperature. After 12 days in solution, a new s at d 153.1 in the 31P-NMR spectrum
of the solution indicated a small amount (7% of total intensity) of an unidentified prod-
uct, which was not further investigated. The fact that p-cymene is less susceptible of dis-
sociation in 3b than in 3a may indicate that the h6 coordination is less favorable for
naphthalenyl than for phenyl. This would explain why 2b dissociates the phosphorami-
dite ligand rather than p-cymene, as well as the fact that an excess of 1b is required in
catalytic cyclopropanation, as discussed below.
Asymmetric Cyclopropanation. Complex 3a, formed in situ by treating 2a with a
chloride scavenger ((Et3O)[PF6] or Tl[PF6]), cyclopropanates styrene in the presence
G
of ethyl diazoacetate as carbene source to give a nearly 1:1 racemic mixture of cis-
and trans-cyclopropanes (Table 5, Run 1) [22]. Racemic cyclopropanes are formed
from a-methylstyrene, too, with low cis selectivity (Run 4). The low level of asymmetric
induction is probably not an effect of p-cymene dissociation from 3a to give 5a, as 5a
cyclopropanates a-methylstyrene with a low but significant level of enantioselectivity
(26 and 21% ee for the cis (1R,2S) and trans (1R,2R) isomers, resp.; Run 5).
Preliminary tests indicated substantial asymmetric induction with the more encum-
bered naphthalenyl-containing ligand 1b, but a number of parameters had to be opti-
mized. Eventually, an in situ preparation of 3b was developed, in which [{RuCl2(h6-
p-cymene)}2] was treated with an excess of 1b (2 equiv. vs. Ru) to suppress ligand dis-
sociation, followed by addition of the chloride scavenger (either (Et3
A
Tl[PF6], 1 equiv. vs. Ru). With this protocol, styrene is reproducibly cyclopropanated

1706
Helvetica Chimica Acta – Vol. 89 (2006)
Fig. 4. One bond 13C,1H correlation showing the five aromatic protons of the h6-bound phenyl ring
plus the two CH(Me)N protons in 5a. For atom numbering, see formula.
with an enantioselectivity as high as 77 and 68% ee for the cis and trans isomers, respec-
tively, with ligand 1b, albeit with no diastereoselectivity (Run 2).
The best results were obtained with 3b and a-methylstyrene, a 1,1-disubstituted ole-
fin, which is cyclopropanated with 86 and 87% ee for the cis and trans isomers, respec-
tively (Run 6). As catalyst 3a gives racemic products (Run 4), enantioselection requires
bulky residues at the N-substituent of the ligand (the naphthalenyl group of 1b). This
effect is reinforced with an a-substituted styrene such as a-methylstyrene, which was
chosen, therefore, as substrate for further studies. As the catalyst performance is inde-
pendent of the chloride scavenger used, only runs obtained with (Et
reported in Table 5. However, the presence of one chloro ligand is required to achieve
high enantioselectivity, as activation of 2b with 2 equiv. of (Et3O)[PF6], which probably
3O)[PF6] are
AHCTREUNG
causes abstraction of both chlorides, decreases both the yield and the enantioselectivity
(Run 8).
To analyze the effect of the relative configuration at the atropisomeric 1,1’-binaph-
thalene moiety and at the N-substituents, the diastereoisomeric ligand (Sa,S,S)-1b’ was
used instead of (Sa,R,R)-1b in selected catalytic reactions. The spectroscopic data
(including a complete NMR structural analysis as for 3b) indicate that the diaster-
eoisomer pairs of complexes 2b/2b’ and 3b/3b’ have a similar structure (see Exper.

Helvetica Chimica Acta – Vol. 89 (2006)
Table 5. Catalytic Cyclopropanationa)
1707
Run
Catalyst
R
Conv. [%]
Yield [%]
cis/trans
ee [%] (abs. config.)
cis trans
8 (1S,2R)
77 (1R,2S)
5 (1S,2R)
rac
1
2
3
4
5
6
7
8
9
10
11
12
13
3a
3b
3b’
3a
5a
H
H
H
38
6
26
5
41:59
45 :55
47:53
60 :40
59 :41
56 :44
66 :34
61:39
57:43
56 :44
57:43
59 :41
57:43
3 (1R,2R)
68 (1R,2R)
13 (1S,2S)
rac
28
31
19
20
47
12
25
23
20
17
24
20
31
17
19
47
7
23
21
14
13
23
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
26 (1R,2S)
86 (1R,2S)
9 (1S,2R)
19 (1R,2S)
82 (1R,2S)
83 (1R,2S)
48 (1R,2S)
73 (1R,2S)
86 (1S,2R)
21 (1R,2R)
87 (1R,2R)
7 (1S,2S)
19 (1R,2R)
83 (1R,2R)
84 (1R,2R)
46 (1R,2R)
73 (1R,2R)
87 (1S,2S)
3b
3b’
2bb)
3bc)
3bd)
3be)
3bf)
3b’’
a) See Exper. Part for catalyst preparation. Reaction conditions: ethyl diazoacetate (0.48 mmol, 1 equiv.
vs. olefin) in CH2Cl2 (1 ml) was added over 6 h to a CH2Cl2 soln. of the olefin (0.48 mmol) and the catalyst
(24 mmol, 5 mol-%). The total reaction time was 20 h, unless otherwise stated. b) 2.1 equiv. of (Et3
O)[PF6]
A
were added. c) The total reaction time was 44 h. d) Ethyl diazoacetate was added in one portion. e) Iso-
lated complex 3b was used. f) Isolated complex 3b plus 1 equiv. of 1b were used.
Part). With both substrates tested in asymmetric cyclopropanation, the (Sa,R,R)-dia-
stereoisomer 1b is the matched one, and the (Sa,S,S)-diastereoisomer 1b’ the mis-
matched one, as the enantioselectivity for the cis isomer collapses from 77% ee (sty-
rene) and 86% ee (a-methylstyrene) to nearly racemic on changing from 1b to 1b’
(Runs 2 vs. 3 and 6 v s. 7). With a-methylstyrene, the change is larger than with styrene,
and the sense of induction is reversed. The same trend is observed for the enantiomeric
excess of the trans isomer, whereas the diastereoselectivity changes to a low extent. The
sharp decrease of the enantioselectivity to nearly racemic on inversion of the relative
configuration indicates that both types of stereogenic units, i.e., that of the N,N-bis(1-
arylethyl) moiety and that of the 1,1’-binaphthalene framework, contribute to enantio-
selection to a similar extent, with minor variations depending on the substrate. In con-
trast, the chiral N,N-bis(1-phenylethyl) moiety dominates stereoselection in the copper-
catalyzed conjugate addition of diethylzinc to cyclohexenone, in which the matched
(Sa,R,R)-1a and mismatched (Sa,S,S)-1a ligands give >98 and 75% ee, respectively
[42], and in the enantioselective allylic amination of achiral allylic esters with an iri-
dium–1a complex, where (Ra,R,R)-1a and (Sa,R,R)-1a gave up to 97 and 75% ee,
respectively [43].
The issue of the stability of catalyst 3b was briefly studied with respect to the disso-
ciation of the phosphoramidite. When isolated 3b is used, lower yield and reduced

1708
Helvetica Chimica Acta – Vol. 89 (2006)
enantioselectivity (48 and 46%) are observed (Run 11). The enantioselectivity is parti-
ally restored (73% for both isomers, Run 12) upon addition of 1 equiv. of ligand 1b to
isolated 3b. The 31P-NMR spectra of the resulting solutions display the signal of 3b and
that of the free ligand 1b in a ca. 1:1 intensity ratio, indicating that only one ligand mol-
ecule binds to the Ru center. These observations suggest dissociation of 1b from 3b in
solution, a reaction that has been observed by NMR spectroscopy for CD2Cl2 solutions
of 2b, but has not been clearly established with cationic 3b (see above).
The good cis selectivity observed with the achiral half-sandwich complexes
[Fe(Cp)(CO)2
(THF)]+ [18] and [RuCl(Cp)(PPh3)2] [21] was the rationale for the
G
choice of chiral analogues as enantioselective, highly cis-diastereoselective cyclopropa-
nation catalysts, which are still rare [44–46] as compared to trans-selective ones (for a
recent comprehensive review, see [47a]; see also [47b]). The results with [RuCl(h6-p-
cymene)(1,2-h-P*-kP)] do not match this expectation, though, as the diastereoselectiv-
ity is very modest. The best ligand/substrate combination is 1b/a-methylstyrene, which
combines the highest enantioselectivity for both diastereoisomers with one of the high-
N
est cis/trans ratios (56 :44; Run 6). With both ligands, the formation of the trans isomer
is slightly favored in the case of styrene, but the cis isomer prevails with a-methylstyr-
ene (Runs 1 vs. 4, 2 v s. 6, and 3 v s. 7). This diastereoselectivity trend is opposite to that
observed for [Fe(Cp)(CO)2
(THF)]+, which shows decreasing selectivity on going from
N
styrene (cis/trans ratio of 85 :15) to a- and p-methylstyrene (both 60 :40), 2-methoxy-
propene (55 :45), and ethyl vinyl ether (45 :55) [18]. The explanation was that elec-
tron-releasing substituents on the olefin slow down the collapse of the short-lived g-car-
bocationic intermediate, giving time for a conformational rearrangement, as originally
proposed by Brookhart and Liu for the analogous stoichiometric reaction [48]. The fact
that we observed the opposite trend hints to substantial mechanistic differences
between [Fe(Cp)(CO)
2(THF)]+ and [RuCl(h6-p-cymene)(1,2h-P*-kP)]+. Finally, it
G
should be noted that the related half-sandwich complex [Ru(h5-Me5C5)(MeCN)(Pꢀ
N)]+ (PꢀN=2-[2-(diphenylphosphino)phenyl]-4,5-dihydro-4-isopropyloxazole) is nei-
ther diastereoselective (cis/trans ratio 46 :54) nor enantioselective [11c].
The catalytic activity of [RuCl(h6-p-cymene)(1,2h-P*-kP)]+ is generally low. Lon-
ger reaction times (44 instead of 24 h) marginally increased the overall cyclopropane
yield from 19 to 23% (Run 9). Monitoring the reaction course by gas chromatography
showed a small but steady increase of the product concentration, which suggests that
the catalyst stays active during the entire reaction time. After 20 h, the 1H-NMR spec-
trum of a reaction solution with ligand 1b and a-methylstyrene shows that, in addition
to the cyclopropanation products (9%), maleate and fumarate were formed (ca. 46 and
8% yield, resp.). However, as unreacted ethyl diazoacetate was still present in the reac-
tion solution (ca. 27% of starting amount), the most probable explanation of the low
cyclopropane yield is the intrinsic low activity of the catalyst and not the competing
homocoupling reaction of carbene to maleate and fumarate. This conclusion is also sup-
ported by the observation that the addition of ethyl diazoacetate in one portion does
not affect significantly the reaction yield (Run 10).
Although no clear trend can be distinguished, the increased productivity of both
catalysts with a-methylstyrene as compared to styrene (Runs 1 vs. 4 and 2 v s. 6) is prob-
ably an electronic effect. In fact, as (carbene)ruthenium complexes are relatively weak
electrophiles [49], an electron-rich olefin such as a-methylstyrene is expected to react

Helvetica Chimica Acta – Vol. 89 (2006)
1709
faster than styrene with the putative (carbene)ruthenium. We have previously noted
that electron-rich olefins are more prone to carbene transfer from [RuCl(PNNP)]+ cat-
alysts (PNNP=chiral tetradentate ligand with a P2N2 donor set) [46b]. Furthermore,
for each substrate, yields decrease with increasing steric bulk of the ligand, that is, on
going from 1a to 1b.
Conclusion. – The coordination chemistry described above clearly shows that phos-
phoramidite ligands can behave as two-, four-, or eight-electron donors. Thus, it cannot
be considered a general fact that they are ‘monodentate chiral ligands’, and care should
be taken when this assumption is made without supporting evidence. The application to
catalytic cyclopropanation indicates that the h2-aryl interaction offers an additional
possibility to stabilize catalytically active 16-electron fragments. Provided that a
bulky phosphoramidite ligand is used, highly enantioselective carbene transfer is ach-
ieved with a-methylstyrene, an example of 1,1-disubstituted olefins, which are more dif-
ficult to cyclopropanate than styrene. However, despite the secondary h2- and h6-aryl
interactions, bulky phosphoramidites tend to dissociate from the metal, which suggests
that the thermodynamic chelate effect is less efficient than with diphosphines.
We thank Sebastian Gischig and Isabelle Haller for the X-ray structure determinations and the Swiss
National Science Foundation for financial support to D. H. (grant 200020-101357).
Experimental Part
General. Reactions with air- or moisture-sensitive materials were carried out under Ar by using
Schlenk techniques, or in a glove box under purified N2. (ꢀ)-Bis[(1R)-1-phenylethyl]amine hydrochlo-
ride, (ꢀ)-bis[(1R)-1-(naphthalen-1-yl)ethyl]amine hydrochloride, (ꢀ)-bis[(1S)-1-(naphthalen-1-yl)-
ethyl]amine hydrochloride, ethyl diazoacetate, a-methylstyrene, and decane were obtained from Aldrich.
(ꢀ)-(1S)-[1,1’-Binaphthalene]-2,2’-diol, styrene, dodecane, triethyloxonium hexafluorophosphate, (ꢀ)-
(R)-a-phellandrene, and phosphorus trichloride were purchased from Fluka, thallium hexafluorophos-
phate from Strem or Alfa Aesar, and ruthenium(III) chloride trihydrate from Pressure Chemicals.
Ethyl diazoacetate was distilled and stored over 4-Å molecular sieves. All other commercially available
reagents were used without further purification. Solvents were purified by standard procedures: CH2Cl2
was distilled from CaH2. [RuCl2(h6-p-cymene)]2 [50], (1Sa)-[1,1’-binaphthalene]-2,2’-diyl bis[(1R)-1-phe-
nylethyl]phosphoramidite
f][1,3,2]dioxaphosphepin-4-amine; 1a), (1Sa)-[1,1’-binaphthalene]-2,2’-diyl bis[(1R)-1-(naphthalen-1-
yl)ethyl]phosphoramidite (=(11bSa)-N,N-Bis[(1R)-1-(naphthalen-1-yl)ethyl]dinaphtho[2,1-d:1’,2’-
f][1,3,2]dioxaphosphepin-4-amine; 1b) and (1Sa)-[1,1’-binaphthalene]-2,2’-diyl bis[(1S)-1-(naphthalen-
1-yl)ethyl]phosphoramidite (=(11bSa)-N,N-Bis[(1S)-1-(naphthalen-1-yl)ethyl]dinaphtho[2,1-d:1’,2’-
(=(11bSa)-N,N-Bis-[(1R)-1-phenylethyl]dinaphthol[2,1-d:1’,2’-
f][1,3,2]dioxaphosphepin-4-amine; 1b’), were prepared according to literature procedures [42][51].
1
[a]D: Perkin-Elmer-341 polarimeter; 1-dm cell; in CHCl3, unless otherwise stated. H- (500 MHz), 31P-
(202.5 MHz), and 13C-NMR (125.8 MHz) Spectra: Bruker-Avance spectrometer; CD2Cl2 solns., unless
otherwise stated; chemical shifts d in ppm downfield of SiMe4; d(P) referenced externally to 85%
H3PO4 (d 0.0); coupling constants J in Hz. Mass spectra were measured by the MS service (Laboratorium
für Organische Chemie, ETH Zürich). ESI-MS: Micromass-Auto-SpecUltima mass spectrometer, at 70
eV. HR-MALDI-MS: IonSpec-Ultima-HR-MALDI-FT-ICR mass spectrometer, at 4.7 Tesla; DCTB
(trans-2-{3-[4-(tert-butyl)phenyl]-2-methylprop-2-enylidene}malononitrile) matrix. Elemental analyses
were carried out by the Laboratory of Microelemental Analysis (Laboratorium für Organische Chemie,
ETH Zürich).
{(11bSa)-N,N-Bis[(1R)-1-phenylethyl]dinaphtho[2,1-d:1’,2’-f][1,3,2]dioxaphosphepin-4-amine-kP4}-
A
N

1710
Helvetica Chimica Acta – Vol. 89 (2006)
mmol) and 1a (750 mg, 1.39 mmol, 1.1 equiv.) were dissolved in CH2Cl2 (20 ml). The resulting soln. was
stirred at r.t. for 1 h. iPrOH (15 ml) was added, and CH2Cl2 was evaporated. The precipitate was filtered
off and dried in vacuo: 2a (992 mg, 93%). Orange solid. [a]2D0 =+65 (c=0.125). 1H-NMR: 1.09 (d, J=7.0,
3 H, Me2
N
G
(cym)); 2.74 (sept., J=7.0, Me
AHCTREUNG
(cym)); 4.98–5.11 (m, 2 CHN); 5.16 (d, J=6.1, 1 arom. H (cym)); 5.22 (d, J=6.1, 1 arom. H (cym));
6.59–7.96 (m, 22 arom. H). 31P-NMR: 142.3 (s). MALDI-MS: 846 (8, [RuCl2(h6-p-cymene)(1a)]+), 810
(2, [RuCl(h6-p-cymene)(1a)]+), 712 (3, [RuCl2(1a)]+), 676 (100, [RuCl(1a)]+), 641 (5, [Ru(1a)]+). Anal.
calc. for C46H44Cl2NO2PRu (845.80): C 65.32, H 5.24, N 1.66; found: C 65.44, H 5.40, N 1.67.
{(11bSa)-N,N-Bis[(1R)-1-(naphthalen-1-yl)ethyl]dinaphtho[2,1-d:1’,2’-f][1,3,2]dioxaphosphepin-4-
amine-kP4}dichloro[h6-1-methyl-4-(1-methylethyl)benzene]ruthenium (2b). As described for 2a, from
[{RuCl2(h6-p-cymene)}2] (196 mg, 0.320 mmol) and 1b (451 mg, 0.705 mmol, 1.1 equiv.) in CH2Cl2 (10
ml): 2b (260 mg, 49%). Orange solid. [a]2D0 =+155 (c=0.125). 1H-NMR (ꢀ408): 1.13 (d, J=6.9, 3 H, Me2-
A
ACHTREUNG
AHCTREUNG
J=5.5, 1 arom. H (cym)); 4.93 (d, J=5.7, 1 arom. H (cym)); 4.93–5.01 (m, 1 arom. H); 5.40 (d, J=5.5, 1
arom. H (cym)); 5.53 (dq, 3J(H,H)=7.0, 3J(H,P)=19.4, 1 CHN); 6.47–6.65 (m, 1 CHN); 6.81–8.16 (m,
24 arom. H); 8.41–8.55 (m, 1 arom. H). 31P-NMR: 146.6 (s, 96%, complex 2b); 150.0 (s, 4%, free ligand
1b). MALDI-MS: 910 (7, [RuCl(h6-p-cymene)(1b)]+), 776 (100, [RuCl(1b)]+), 741 (11, [Ru(1b)]+), 640
(64, [1b]+). Anal. calc. for C54H48Cl2NO2PRu (945.91): C 68.57, H 5.11, N 1.48; found: C 68.62, H 5.27,
N 1.51.
{(11bSa)-N,N-Bis[(1S)-1-(naphthalen-1-yl)ethyl]dinaphtho[2,1-d:1’,2’-f][1,3,2]dioxaphosphepin-4-
amine-kP4}dichloro[h6-1-methyl-4-(1-methylethyl)benzene]ruthenium (2b’). As described for 2b, from
[RuCl2(h6-p-cymene)]2 (65 mg, 0.107 mmol) and 1b’ (150 mg, 0.234 mmol, 1.1 equiv.) in CH2Cl2 (10
1
ml): 2b’ (165 mg, 82%). Orange solid. H-NMR (ꢀ408): 1.13 (d, J=7.0, 3 H, Me2
A
J=7.0, 3 H, Me2
ACHTREUNG
3 H, ArCH(Me)); 2.83 (sept., J=6.8, Me
AHCTREUNG
J=5.9, 1 arom. H (cym)); 4.93 (d, J=5.7, 1 arom. H (cym)); 4.98 (t, J=7.5, 1 arom. H (naph)); 5.59
(d, J=5.9, 1 arom. H (cym)); 5.85 (t, J=7.5, 1 arom. H (naph)); 5.89–5.96 (m, 1 CHN); 5.95 (t,
J=7.5, 1 arom. H (naph)); 6.13–6.21 (m, 1 CHN); 6.27 (d, J=6.5, 1 arom. H (naph)); 6.37 (d, J=7.0,
1 arom. H (naph)); 6.61 (t, J=7.5, 1 arom. H (naph)); 6.83 (d, J=8.0, 1 arom. H (naph)); 6.97 (d,
J=8.0, 1 arom. H (naph)); 7.23 (d, J=8.0, 1 arom. H (naph)); 7.29–7.62 (m, 11 arom. H (incl. naph));
7.76 (d, J=8.5, 1 arom. H); 7.80 (d, J=9.0, 1 arom. H); 7.98 (d, J=8.0, 1 arom. H); 8.03 (t, J=9.0, 1
arom. H (naph)); 8.11 (d, J=9.0, 1 arom. H); 8.41 (d, J=9.0, 1 arom. H); 31P-NMR: 162.6 (s).
Chloro[h6-1-methyl-4-(1-methylethyl)benzene]{(11bSa)-N-{(1R)-1-[(1,2-h)-phenyl]ethyl}-N-[(1R)-
1-dinaphtho[2,1-d:1’,2’-f][1,3,2]dioxaphosphepin-4-amine-kP4}ruthenium(1+)
Hexafluorophos-
phate(1ꢀ) (3a). Complex 2a (200 mg, 0.236 mmol) was treated with Tl[PF6] (87 mg, 0.248 mmol, 1.05
i
equiv.) in CH2Cl2 (12 ml). After stirring at r.t. for 21 h, TlCl was filtered off, PrOH (17 ml) was
added, and CH2Cl2 was evaporated. The precipitate was filtered off and dried in vacuo: 3a (201 mg,
89%). Orange solid. The integration of the 1H- and 31P-NMR spectra (3a, 87%; 5a, 13%) indicated
that the sample contained some 5a as impurity. [a]2D0 =+99 (c=0.105). 1H-NMR (ꢀ208): 0.90 (d,
J=7.0, 3 H, Ph
1.48 (d, J=7.5, 3 H, Phfree
J=5.9, 1 arom. H (cym)); 3.27–3.35 (m, Ph
arom. H (cym)); 4.78 (quint., 3J(H,H)=7.2, 3J(H,P)=7.2, Phfree
A
G
ACHTREUNG
(cym)); 6.41 (d, J=6.5, 1 arom. H of (C=C)coord); 7.03–8.23 (m, 21 arom. H). 31P-NMR: 154.6 (s, 93%,
complex 3a); 150.4 (s, 7%, complex 5a). ESI-MS: 774 (100, [MꢀHCl]+), 676 (20, [RuCl(1a)]+). The ele-
mental analysis of a sample containing 3a (87%) and 5a (13%, as determined by integration of the 31P-
NMR spectrum) gave: Anal. calc. for C46H44ClF6NO2P2Ru (955.31; 87%)/C72H60Cl2F12N2O4P4Ru2
(1642.18; 13%): C 57.16, H 4.51, N 1.50; found: C 57.19, H 4.54, N 1.50.
[Ru(S)]-Chloro[h6-1-methyl-4-(1-methylethyl)benzene]{(11bSa)-N-{(1R)-1-[(1,2-h)-naphthalen-1-
yl]ethyl}-N-[(1R)-1-(naphtalen-1-yl)ethyl]dinaphtho[2,1-d:1’,2’-f][1,3,2]dioxaphosphepin-4-amine-kP4}-
A

Helvetica Chimica Acta – Vol. 89 (2006)
1711
(150 mg, 0.234 mmol, 1.1 equiv.) were dissolved in CH2Cl2 (12 ml). After stirring at r.t. for 1 h, Tl[PF6]
i
was added, and the resulting soln. was stirred at r.t. for 21 h. Then, TlCl was filtered off, PrOH (18
ml) was added, and CH2Cl2 was evaporated. The precipitate was filtered off and dried in vacuo: 3b
(203 mg, 89%). Red solid. [a]2D0 =+381 (c=0.125). 1H-NMR (ꢀ208): 0.08 (s, Me (cym)); 1.18 (d,
J=7.0, 3 H, Me2
1.73 (d, J=6.7, 3 H, naph
(cym)); 3.52–3.58 (m, naph
(cym)); 5.05 (dq, 3J(H,P)=25.3, 3J(H,H)=6.7, naphfree
CHN); 5.71 (d, J=6.1, 1 arom. H (cym));
A
G
ACHTREUNG
freeCH(Me)); 2.62 (d, J=5.5, 1 arom. H (cym)); 2.90 (sept., J=7.0, Me2
AHCTREUNG
N
6.75–8.20 (m, 25 arom. H); 9.24 (d, J=8.5, 1 arom. H). 31P-NMR: 168.0 (s). ESI-MS: 910 (100,
[RuCl(h6-p-cymene)(1b)]+), 875 (3, [Ru(h6-p-cymene)(1b)]+). Anal. calc. for C54H48ClF6NO2P2Ru
(1055.43): C 61.45, H 4.58, N 1.33; found: C 61.31, H 4.66, N 1.35.
Chloro[h6-1-methyl-4-(1-methylethyl)benzene]{(11bSa)-N-{(1S)-1-[(1,2-h)-naphthalen-1-yl]ethyl}-N-
[(1S)-1-(naphthalen-1-yl)ethyl]dinaphtho[2,1-d:1’,2’-f][1,3,2]dioxaphosphepin-4-amine-kP4}ruthenium(1+)
Hexafluorophosphate (1ꢀ) (3b’). Complex 2b’ (30 mg, 0.032 mmol) was treated with Tl[PF6] (12 mg,
0.035 mmol, 1.1 equiv.) in CD2Cl2 (1 ml). After stirring at r.t. for 19 h, TlCl was filtered off. The resulting
soln. of 3b’ was characterized spectroscopically. 1H-NMR (ꢀ608): 0.30 (s, Me (cym)); 0.82 (br. s, naphcoord
-
A
CH (cym)); 1.44 (br. s, naphfree
G
N
(cym)); 2.58 (d, J=5.9, 1 arom. H (cym)); 2.90 (sept., J=7.0, Me
H (cym)); 4.53 (d, J=6.0, 1 arom. H (cym)); 4.58 (dq, 3J(H,H)=6.4, 3J(H,P)=20.8, naphcoord
5.30–5.40 (m, naphfreeCHN); 5.65 (d, J=5.7, 1 arom. H (cym)); 6.51 (d, J=5.5, 1 arom. H (naph));
6.73–8.31 (m, 24 arom. H (incl. naph)); 8.72 (d, J=8.0, 1 arom. H (naph)). 31P-NMR: 151.2 (s).
Dichloro{(11bSa)-N-[(1R)-1-(h6-phenyl)ethyl]-N-[(1R)-1-phenylethyl]dinaphtho[2,1-d:1’,2’-f][1,3,2]-
ACHTREUNG
AHCTREUNG
AHCTREUNG
ACHTREUNG
with Et
N
N
dried in air and then in vacuo: 4a (133 mg, 93%). b) A soln. of [{Ru(p-cymene)Cl2}2] (122.4 mg, 0.2 mmol)
and the ligand 1a (215.8 mg, 0.4 mmol) in CHCl3 (30 ml) was stirred under reflux for 7 d. Then, the sol-
vent was evaporated up to ca. 2 ml, and the complex 4a was precipitated with Et
thus obtained was washed twice with Et2O and purified by flash chromatography (SiO2; CH2Cl2/MeOH
98 :2; Rf 0.15): 4a (125 mg, 44%). [a]2D0 =+267 (c=0.4, CH2Cl2). 1H-NMR: 1.10 (d, J=7.3, Phfree
CH(Me)); 1.87 (d, J=6.5, Ph
coordCH(Me)); 4.23 (d, J=6.0, 1 Ho (Phcoord)); 4.24 (dq, 3J(H,H)=6.5,
3J(H,P)=35.8, Phcoord
CHN); 4.83 (quint., J(H,H)=6.8, J(H,P)=6.8, Phfree
U
AHCTREUNG
-
A
ACHTREUNG
3
3
N
G
(Phcoord)); 5.76–5.80 (m, 1 Hm (Phcoord)); 6.22–6.26 (m, 1 Hm (Phcoord)); 6.24–6.29 (m, 1 Hp (Phcoord));
7.16–7.57 (m, 11 arom. H); 7.70 (d, J=8.9, 1 arom. H); 7.97 (d, J=8.2, 1 arom. H); 8.02 (s, 2 arom.
H); 8.05 (d, J=8.4, 1 arom. H); 8.18 (d, J=8.9, 1 arom. H). 13C-NMR: 19.5 (Phfree
A
-
ACHTREUNG
(Phcoord)); 95.0 (2J(C,P)=21.2, Cp (Phcoord)); 97.8 (Cm (Phcoord)); 104.9 (2J(C,P)=8.3, Cm (Phcoord)); 106.6
(J(C,P)=1.4, Cipso (Phcoord)); 121.2 (Ar); 122.3 (J(C,P)=1.7, Ar); 123.4 (J(C,P)=2.9, Ar); 124.5
(J(C,P)=4.0, Ar); 125.8, 126.0, 126.4, 126.9 (Ar); 127.5 (J(C,P)=2.1, Ar); 128.4, 128.6, 128.7, 128.8,
129.2, 130.0 (Ar); 131.1 (J(C,P)=1.4, Ar); 132.0, 132.1, 132.3 (Ar); 133.2 (J(C,P)=1.7, Ar); 140.5
(3J(C,P)=2.3, Cipso (Ph)); 148.5 (2J(C,P)=15.2, POC); 149.0 (2J(C,P)=7.3, POC). 31P-NMR: 150.9 (s).
MALDI-MS: 676 (81, [RuCl(1a)]+), 641 (25, [Ru(1a)]+). Anal. calc. for C36H30Cl2NO2PRu (711.59): C
60.76, H 4.25, N 1.97; found: C 60.74, H 4.30, N 2.01.
Di-m-chlorobis{(11bSa)-N-[(1R)-1-(h6-phenyl)ethyl]-N-[(1R)-1-phenylethyl]dinaphtho[2,1-d:1’,2’-f]-
A
Schlenk vessel was charged with 3a (62 mg, 0.065 mmol) under Ar. The vessel was evacuated (high vac-
uum, <5×10ꢀ3 mbar), and the solid was heated in an oil bath at 1508 for 77 h: 5a (quant.). Orange solid.
[a]2D0 =+189 (c=0.125). 1H-NMR: 1.09 (d, J=7.2, 6 H, Phfree
ACHTREUNG
AHCTREUNG
3
H)=7.0, J(H,P)=7.0, 2 H, Phfree
ACHTREUNG

1712
Helvetica Chimica Acta – Vol. 89 (2006)
(Phcoord)); 7.03–8.29 (m, 34 arom. H). 31P-NMR: 150.4 (s). ESI-MS: 1352 (5, [Ru2Cl2(1a)2]+), 1317 (7,
[Ru2Cl(1a)2]+), 676 (31, [RuCl(1a)]+), 641 (13, [Ru(1a)]+). Anal. calc. for C72H60Cl2F12N2O4P4Ru2
(1642.18): C 52.66, H 3.68, N 1.71; found: C 52.87, H 3.87, N 1.78.
X-Ray Structure Determination of 2a2). Orange crystals of 2a were grown from iPrOH. Crystal data:
C46H44Cl2NO2PRu, monoclinic, P21, 0.30×0.06×0.03 mm, a=12.3128(11), b=13.9824(12),
c=12.3910(11) Å, b=112.163(2)8, V=1975.6(3) Å3, Z=2, F(000)=872,
D ,
calc =1.422 Mg cmꢀ3
m=0.612 mmꢀ1, MoKa (l 0.71073 Å). Data were collected at r.t. with a Bruker-AXS-Smart-Apex plat-
form in the q range 2.30–20.818. The structure was solved with SHELXTL by direct methods. Of the
7619 measured reflections (ꢀ11ꢁhꢁ2, ꢀ13ꢁkꢁ13, ꢀ12ꢁlꢁ12), 4121 unique reflections were used
in the refinement (full-matrix least squares on F2 with anisotropic displacement parameters). The abso-
lute structure parameter was refined to ꢀ0.03(5). R1 =0.0557 (3578 data with Fo >2s(Fo)), wR2 =0.1091
(all 4121 data). Max. and min. difference peaks were +0.417 and ꢀ0.598 eÅꢀ3
.
X-Ray Structure Determination of 3b2). Red crystals of 3b were grown by layering hexane over a
CH2Cl2 soln. of rac-3b (obtained by mixing (Sa,R,R)-1b and (Ra,S,S)-1b in a 1:1 ratio). Crystal data:
C
55.5H48Cl4F6NO2P2Ru, triclinic, Pꢀ1, 0.76×0.49×0.28 mm, a=13.3040(8), b=13.7074(8),
c=15.1350(9) Å, a=98.822(1), b=96.636(1), g=11.659(1)8, V=2489.9(3) Å3, Z=2, F(000)=1200,
calc =1.574 g cmꢀ3, m=0.661 mmꢀ1. Data were collected at r.t. on a Bruker-AXS-Smart-Apex platform
D
in the q range 1.64–28.288. The structure was solved with SHELXTL by direct methods. Of the 25914
measured reflections (ꢀ17ꢁhꢁ17, ꢀ18ꢁkꢁ18, ꢀ20ꢁlꢁ20), 12262 unique reflections were used in
the refinement (full-matrix least squares on F2 with anisotropic displacement parameters). R1 =0.0405
(11071 data with Fo >4s(Fo)), wR2 =0.1112 (all data). Max. and min. difference peaks were +1.501
and ꢀ0.865 eÅꢀ3
.
NMR PGSE Diffusion Measurements. Pulsed-gradient spin-echo (PGSE) diffusion measurements
were carried out by using the stimulated spin-echo sequence [28]. The diffusion coefficient D was derived
from Eqn. 1 where G=gradient strength, D=delay between the midpoints of the gradients, and d=gra-
dient length.
ꢀ
ꢁ
ꢀ
ꢁ
I
I0
d
3
2
ln
¼ ꢀðgdÞ G2 D ꢀ
D
(1)
All diffusion measurements were performed on a Bruker-Avance 400-MHz spectrometer equipped
with a microprocessor-controlled gradient unit and a multinuclear inverse probe with an actively shielded
Z-gradient coil. The shape of the gradient pulse was rectangular, its length 1.75 ms, and its strength varied
automatically in the course of the experiment. The time between midpoints of the gradients (D) was
chosen as 167.75 ms. The measurements were carried out without sample spinning and in the absence
of external airflow. The sample temp. (300 K for r.t. measurements) was controlled by a digital BVT-
3000 variable-temp. unit. The error coefficients for the D values based on our experience is ꢂ0.06.
The viscosity used in the Stokes–Einstein relation is that of pure CH2Cl2, 0.41 [28h]. The 1H,19F-
HOESY-NMR measurements were carried out with a doubly tuned TXI probe. A mixing time of 800
ms was used, and 32 scans for each of the 1024 t1 increments were recorded.
Standard Catalytic Run. Complex 2a (24 mmol) and (Et3O)[PF6] or Tl[PF6] (26 mmol, 1.1 equiv.) were
G
dissolved in CH2Cl2 (1 ml) and stirred at r.t. for 17 h. TlCl was filtered off with a syringe filter. With ligand
1b, the catalyst was prepared in situ from [{RuCl2(h6-p-cymene)}2] (12 mmol), 1b (48 mmol), and the hal-
ide scavenger (26 mmol, 1.1 equiv.). The internal standard and the olefin (0.48 mmol) were added to the
soln. of the catalyst (24 mmol, 5 mol-%). Ethyl diazoacetate (0.48 mmol, 1 equiv. vs. olefin) in CH2Cl2 (1
ml) was added over 6 h by a syringe pump. The soln., which was protected from light, was stirred for addi-
tional 14 h at r.t. and then analyzed by GC. The total reaction time was 20 h at r.t. Each experiment was at
2
)
CCDC-291266 and CCDC-29166 contain the supplementary crystallographic data for 2a and 3b,
resp. These data can be obtained free of charge via http://www.ccdc.cam.ac.uk/data_request/cif
from the Cambridge Crystallographic Data Centre.

Helvetica Chimica Acta – Vol. 89 (2006)
1713
least reproduced once, and deviations between two to five experiments were usually less than 2%. A con-
trol reaction without the catalyst indicated that there was no formation of the cyclopropane derivatives
under the conditions used.
Styrene. Olefin conversion and yields of cis and trans products were determined by GC analysis with
decane as internal standard. Samples for GC analyses were prepared by filtration over a plug of alumina
to remove the catalyst. Achiral GC analysis: Optima column (25 m), He carrier (100 kPa); temp. pro-
gram: 508 isotherm for 5 min, then to 2008 at 58 minꢀ1. tR [min]: styrene 8.5, decane (internal standard)
12.8, ethyl cis-2-phenylcyclopropanecarboxylate 26.5, and ethyl trans-2-phenylcyclopropanecarboxylate
27.9. Chiral GC analysis: the enantiomeric excesses (ee) of the cis and trans products were determined
with a Supelco-Beta-Dex-120 column, 1.4 ml He minꢀ1; temp. program: 1208 isotherm; tR [min]: cis-(1S,
2R) 52.8, cis-(1R,2S) 55.5, trans-(1R,2R) 62.7, and trans-(1S,2S) 64.6. The abs. configuration was deter-
mined for cis- and trans-ethyl 2-phenylcyclopropanecarboxylate by comparison of the sign of the optical
rotation with literature values [52].
a-Methylstyrene. As for styrene, with dodecane as internal standard. Achiral GC analysis: as for sty-
rene; tR [min]: a-methylstyrene 12.0, dodecane 19.6, ethyl cis-2-methyl-2-phenylcyclopropanecarboxylate
26.3, ethyl trans-2-methyl-2-phenylcyclopropanecarboxylate 27.6. Chiral GC analysis: as for styrene; tR
[min]: cis-(1R,2S) 40.6, cis-(1S,2R) 42.8, trans-(1S,2S) 51.5, and trans-(1R,2R) 52.7. The abs. configuration
was determined for ethyl cis- and trans-2-methyl-2-phenylcyclopropanecarboxylate by comparison of the
sign of the optical rotation with literature values [53].
REFERENCES
[1] H. Brunner, Z. Anorg. Allg. Chem. 1969, 368, 120; H. Brunner, Adv. Organomet. Chem. 1980, 18,
151; H. Brunner, Angew. Chem., Int. Ed. 1999, 38, 1195.
[2] J. A. Gladysz, B. J. Boone, Angew. Chem., Int. Ed. 1997, 36, 551.
[3] a) G. Consiglio, F. Morandini, Chem. Rev. 1987, 87, 761; b) F. Morandini, G. Consiglio, V. Lucchini,
Organometallics 1985, 4, 1202; c) F. Morandini, G. Consiglio, B. Straub, G. Ciani, A. Sironi, J. Chem.
Soc., Dalton Trans. 1983, 2293; d) G. Consiglio, F. Morandini, J. Organomet. Chem. 1986, 310, C66.
[4] a) M. A. Bennett, Coord. Chem. Rev. 1997, 166, 225; b) F. C. Pigge, J. J. Consiglio, Curr. Org. Chem.
2001, 5, 757; c) C. Ganter, Chem. Soc. Rev. 2003, 32, 130.
[5] H. Brunner, A. Kollnberger, A. Mehmood, T. Tsuno, M. Zabel, J. Organomet. Chem. 2004, 689,
4244; H. Brunner, E. Grau, M. Zabel, Organometallics 2004, 23, 3788; H. Brunner, A. Kollnberger,
M. Zabel, Polyhedron 2003, 22, 2639.
[6] H. Brunner, J. Klankermayer, M. Zabel, Organometallics 2002, 21, 5746.
[7] R. Noyori, S. Hashiguchi, Acc. Chem. Res. 1997, 30, 97; M. Yamakawa, H. Ito, R. Noyori, J. Am.
Chem. Soc. 2000, 122, 1466; R. Noyori, Angew. Chem., Int. Ed. 2002, 41, 2008; S. Burk, G. Franciò,
W. Leitner, Chem. Commun. 2005, 3460; M. T. Reetz, X. G. Li, J. Am. Chem. Soc. 2006, 128, 1044.
[8] M. Ito, M. Hirakawa, K. Murata, T. Ikariya, Organometallics 2001, 20, 379.
[9] E. P. Kündig, C. M. Saudan, V. Alezra, F. Viton, G. Bernardinelli, Angew. Chem., Int. Ed. 2001, 40,
4481; F. Viton, G. Bernardinelli, E. P. Kündig, J. Am. Chem. Soc. 2002, 124, 4968.
[10] D. L. Davies, J. Fawcett, S. A. Garratt, D. R. Russell, Chem. Commun. 1997, 1351; D. L. Davies, J.
Fawcett, S. A. Garratt, D. R. Russell, Organometallics 2001, 20, 3029; A. J. Davenport, D. L. Davies,
J. Fawcett, D. R., Dalton Trans. 2004, 1481.
[11] a) D. Carmona, C. Cativiela, S. Elipe, F. J. Lahoz, M. P. Lamata, L. R. de Viu, L. A. Oro, C. Vega, F.
Viguri, Chem. Commun. 1997, 2351; b) D. Carmona, M. P. Lamata, F. Viguri, R. Rodriguez, L. A.
Oro, F. J. Lahoz, A. I. Balana, T. Tejero, P. Merino, J. Am. Chem. Soc. 2005, 127, 13386; c) M.
Lasa, P. Lopez, C. Cativiela, D. Carmona, L. A. Oro, J. Mol. Catal. A 2005, 234, 129.
[12] a) J. W. Faller, B. P. Patel, M. A. Albrizzio, M. Curtis, Organometallics 1999, 18, 3096; b) J. W. Faller,
B. J. Grimmond, D. G. D’Alliessi, J. Am. Chem. Soc. 2001, 123, 2525; c) J. W. Faller, B. J. Grimmond,
Organometallics 2001, 20, 2454; d) J. W. Faller, A. R. Lavoie, B. J. Grimmond, Organometallics
2002, 21, 1662; e) J. W. Faller, P. P. Fontaine, Organometallics 2005, 24, 4132; f) J. W. Faller, D. G.
D’Alliessi, Organometallics 2003, 22, 2749.

1714
Helvetica Chimica Acta – Vol. 89 (2006)
[13] P. Salvadori, P. Pertici, F. Marchetti, R. Lazzaroni, G. Vitulli, M. A. Bennett, J. Organomet. Chem.
1989, 370, 155; P. Pertici, E. Pitzalis, F. Marchetti, C. Rosini, P. Salvadori, J. Organomet. Chem. 1994,
466, 221; C. G. Arena, S. Calamia, F. Faraone, C. Graiff, A. Tiripicchio, J. Chem. Soc., Dalton Trans.
2000, 3149.
[14] H. Butenschön, Chem. Rev. 2000, 100, 1527; B. M. Trost, B. Vidal, M. Thommen, Chem.–Eur. J.
1999, 5, 1055.
[15] a) W. Tang, X. Zhang, Chem. Rev. 2003, 103, 3029; b) F. Lagasse, H. B. Kagan, Chem. Pharm. Bull.
2000, 48, 315; c) W.-J. Shi, Q. Zang, J.-H. Xie, S.-F. Zhu, G.-H. Hou, Q.-L. Zhou, J. Am. Chem. Soc.
2006, 128, 2780; d) L. Panella, A. M. Aleixandre, G. J. Kruidhof, J. Robertus, B. L. Feringa, J. G. de
Vries, A. J. Minnaard, J. Org. Chem. 2006, 71, 2026; e) I. S. Mikhel, G. Bernardinelli, A. Alexakis,
Inorg. Chim. Acta 2006, 359, 1826; f) M. d’Augustin, L. Palais, A. Alexakis, Angew. Chem., Int. Ed.
2005, 44, 1376; g) A. Alexakis, V. Albrow, K. Biswas, M. d’Augustin, O. Prieto, S. Woodward, Chem.
Commun. 2005, 2843; h) R. B. C. Jagt, J. G. de Vries, B. L. Feringa, A. J. Minnaard, Org. Lett. 2005,
7, 2433; i) V. N. Tsarev, S. E. Lyubimov, A. A. Shiryaev, S. V. Zheglov, O. G. Bondarev, V. A. Davan-
kov, A. A. Kabro, S. K. Moiseev, V. N. Kalinin, K. N. Gavrilov, Eur. J. Org. Chem. 2004, 2214; j) K.
Tissot-Croset, D. Polet, A. Alexakis, Angew. Chem., Int. Ed. 2004, 43, 2426.
[16] M. Brookhart, Y. M. Liu, E. W. Goldman, D. A. Timmers, G. D. Williams, J. Am. Chem. Soc. 1991,
113, 927.
[17] R. D. Theys, M. M. Hossain, Tetrahedron Lett. 1995, 36, 5113; Q. Wang, F. H. Fösterling, M. M. Hos-
sain, Organometallics 2002, 21, 2596, and ref. cit. therein; Q. W. Wang, F. H. Forsterling, M. M. Hos-
sain, J. Organomet. Chem. 2005, 690, 6238.
[18] W. J. Seitz, A. K. Saha, M. M. Hossain, Organometallics 1993, 12, 2604.
[19] F. Simal, A. Demonceau, A. F. Noels, D. R. T. Knowles, S. O’Leary, P. Maitlis, O. Gusev, J. Organo-
met. Chem. 1998, 558, 163; F. Simal, D. Jan, A. Demonceau, A. F. Noels, Tetrahedron Lett. 1999, 40,
1653.
[20] Y. Matsushima, H. Kikuchi, M. Uno, S. Takahashi, Bull. Chem. Soc. Jpn. 1999, 72, 2475.
[21] W. Baratta, W. A. Herrmann, R. M. Kratzer, P. Rigo, Organometallics 2000, 19, 3664.
[22] D. Huber, A. Mezzetti, Tetrahedron: Asymmetry 2004, 15, 2193.
[23] D. Huber, P. G. A. Kumar, P. S. Pregosin, A. Mezzetti, Organometallics 2005, 24, 5221.
[24] H. Nazakawa, K. Kawamura, K. Kubo, K. Miyoshi, Organometallics 1999, 18, 2961.
[25] J. Shen, E. D. Stevens, S. P. Nolan, Organometallics 1998, 17, 3000.
[26] T. J. Johnson, K. Folting, W. E. Streib, J. D. Martin, J. C. Huffman, S. A. Jackson, O. Eisenstein, K. G.
Caulton, Inorg. Chem. 1995, 34, 488.
[27] B. E. Mann, B. F. Taylor, ‘13C-NMR Data for Organometallic Compounds’, Academic Press, Lon-
don, 1981.
[28] a) M. Valentini, H. Rüegger, P. S. Pregosin, Helv. Chim. Acta 2001, 84, 2833; b) B. Binotti, A. Mac-
chioni, C. Zuccaccia, D. Zuccaccia, Comments Inorg. Chem. 2002, 23, 417; c) P. S. Pregosin, E. Mar-
tinez-Viviente, P. G. A. Kumar, Dalton Trans. 2003, 4007; d) P. G. A. Kumar, P. S. Pregosin, J. M.
Goicoechea, M. K. Whittlesey, Organometallics 2003, 22, 2956; e) E. Martinez-Viviente, P. S. Prego-
sin, Inorg. Chem. 2003, 42, 2209; f) P. G. A. Kumar, P. S. Pregosin, M. Vallet, G. Bernardinelli, R. F.
Jazzar, F. Viton, E. P. Kündig, Organometallics 2004, 23, 5410; g) P. G. A. Kumar, P. S. Pregosin, T.
M. Schmid, G. Consiglio, Magn. Reson. Chem. 2004, 42, 795; h) P. S. Pregosin, P. G. A. Kumar, I.
Fernandez, Chem. Rev. 2005, 105, 2977.
[29] K. Stanley, M. C. Baird, J. Am. Chem. Soc. 1975, 97, 6598.
[30] a) C. J. den Reijer, P. Dotta, P. S. Pregosin, A. Albinati, Can. J. Chem. 2001, 79, 693; b) N. Feiken, P.
S. Pregosin, G. Trabesinger, M. Scalone, Organometallics 1997, 16, 537.
[31] P. W. Cyr, S. J. Rettig, B. O. Patrick, B. R. James, Organometallics 2002, 21, 4672.
[32] S. Doherty, J. Knight, R. K. Rath, W. Clegg, R. W. Harrington, C. R. Newman, R. Campbell, H.
Amin, Organometallics 2005, 24, 2633.
[33] J. A. Cabeza, I. del Rio, P. Garcia-Alvarez, D. Miguel, Organometallics 2005, 24, 665.
[34] A. G. Orpen, L. Brammer, F. A. Allen, O. Kennard, D. G. Watson, R. Taylor, J. Chem. Soc., Dalton
Trans. 1989, S1.
[35] M. Hölscher, G. Franciò, W. Leitner, Organometallics 2004, 23, 5606.

Helvetica Chimica Acta – Vol. 89 (2006)
1715
[36] R. Kumareswaran, M. Nandi, T. V. RajanBabu, Org. Lett. 2003, 5, 4345.
[37] T. Hayashi, Acc. Chem. Res. 2000, 33, 354.
[38] P. Kocovsky, S. Vyskocil, I. Cisarova, J. Sejbal, I. Tislerova, M. Smrcina, G. C. Lloyd-Jones, S. C. Ste-
phen, C. P. Butts, M. Murray, V. Langer, J. Am. Chem. Soc. 1999, 121, 7714.
[39] P. G. A. Kumar, P. Dotta, R. Hermatschweiler, P. S. Pregosin, A. Albinati, S. Rizzato, Organometal-
lics 2005, 24, 1306.
[40] C. A. Kiener, C. Shu, C. Incarvito, J. F. Hartwig, J. Am. Chem. Soc. 2003, 125, 14272.
[41] a) C. Welter, A. Dahnz, B. Brunner, S. Streiff, P. Dubon, G. Helmchen, Org. Lett. 2005, 7, 1239; b) S.
Streiff, C. Welter, M. Schelwies, G. Lipowsky, N. Miller, G. Helmchen, Chem. Commun. 2005, 2957.
[42] L. A. Arnold, R. Imbos, A. Mandoli, A. H. M. de Vries, R. Naasz, B. L. Feringa, Tetrahedron 2000,
56, 2865.
[43] T. Ohmura, J. F. Hartwig, J. Am. Chem. Soc. 2002, 124, 15164.
[44] T. Uchida, R. Irie, T. Katsuki, Synlett 1999, 1793; T. Uchida, R. Irie, T. Katsuki, Tetrahedron 2000, 56,
3501.
[45] T. Niimi, T. Uchida, R. Irie, T. Katsuki, Adv. Synth. Catal. 2001, 343, 79; T. Niimi, T. Uchida, R. Irie,
T. Katsuki, Tetrahedron Lett. 2000, 41, 3647; T. Fukuda, T. Katsuki, Tetrahedron 1997, 53, 7201.
[46] a) S. Bachmann, M. Furler, A. Mezzetti, Organometallics 2001, 20, 2102; b) S. Bachmann, A. Mez-
zetti, Helv. Chim. Acta 2001, 84, 3063; c) C. Bonaccorsi, S. Bachmann, A. Mezzetti, Tetrahedron:
Asymmetry 2003, 14, 845; d) C. Bonaccorsi, A. Mezzetti, Organometallics 2005, 24, 4953.
[47] a) M. P. Doyle, M. A. McKervey, T. Ye, ‘Modern Catalytic Methods for Organic Synthesis with
Diazo Compounds: From Cyclopropanes to Ylides’, Wiley, New York, 1998, p. 183; b) ‘Comprehen-
sive Asymmetric Catalysis’, Eds. E. N. Jacobsen, A. Pfaltz, H. Yamamoto, Springer, Berlin, 1999, p.
513–603.
[48] M. Brookhart, Y. M. Liu, J. Am. Chem. Soc. 1991, 113, 939.
[49] G. Maas, Chem. Soc. Rev. 2004, 33, 183.
[50] M. A. Bennett, A. K. Smith, J. Chem. Soc., Dalton Trans. 1974, 233.
[51] J. G. Boiteau, A. J. Minnaard, B. L. Feringa, J. Org. Chem. 2003, 68, 9481.
[52] D. A. Evans, K. A. Woerpel, M. M. Hinman, M. M. Faul, J. Am. Chem. Soc. 1991, 113, 726.
[53] A. Berkessel, P. Kaiser, J. Lex, Chem.–Eur. J. 2003, 9, 4746.
Received March 24, 2006