Functional phosphines XV. Ruthenium complexes containing C 5H8(PR2)2 and Ph2PCH 2CR′2NH2 ligands (R = Me, Ph, OPh; R′ = H, Me): Synthesis and application to homogeneou

Article abstract of DOI:10.1016/j.jorganchem.2004.07.062

Treatment of [Ru(η⁴-C₈H₁₂) {η³-(CH₂)₂CMe}₂] with C ₂ chiral cyclopentane-1,2-diyl-bis(phosphines) trans-1,2-C ₅H₈(PR₂)₂ in

Full text of DOI:10.1016/j.jorganchem.2004.07.062

Journal of Organometallic Chemistry 690 (2005) 1–13
www.elsevier.com/locate/jorganchem
Functional phosphines XV. Ruthenium complexes
containing C5H8(PR2)2 and Ph2PCH2CR02NH2 ligands (R = Me, Ph,
OPh; R0 = H, Me): synthesis and application to homogeneous ˜C@O
hydrogenation and transfer hydrogenation q
*
Lutz Dahlenburg , Christian Kuhnlein
¨
Institut fu¨r Anorganische Chemie der Friedrich-Alexander, Universita¨t Erlangen-Nu¨rnberg, Egerlandstrasse 1, D-91058 Erlangen, Germany
Received 10 May 2004; accepted 19 July 2004
Abstract
Treatment of [Ru(g4-C8H12){g3-(CH2)2CMe}2] with C2 chiral cyclopentane-1,2-diyl-bis(phosphines) trans-1,2-C5H8(PR2)2 in
hexane afforded the chelate complexes [Ru{g3-(CH2)2CMe}2{1,2-C5H8(PR2)2}], where R = Me (2), Ph (3), NC5H10 (4), and OPh
(5). The mixed-ligand compounds ½RuCl2f1; 2-C5H8ðPR2Þ gðPh2PCH2CR02NH2Þꢀ [R0 = H: R = Me (6), Ph (7), OPh (8); R0 = Me:
2
R = Ph (9)] were obtained by reactions of the bis(2-methylallyl) precursors 2, 3, and 5 with methanolic HCl in acetone, followed
by the addition of the required aminophosphine in DMF. The (P\P)2- and (P\N)2-chelated complexes [RuCl2{1,2-C5H8(PMe2)2}2]
(1) and [RuCl2(Ph2PCH2CMe2NH2)2] (10) resulted from RuCl3 Æ 3H2O and 1,2-C5H8(PMe2)2 or Ph2PCH2CMe2NH2 under reduc-
ing conditions. The crystal structures of 1, 3, 4, 6, 7, 9, and 10 were determined by single-crystal X-ray diffraction. Complexes 7, 9,
and 10, activated by KOBu-t, i-PrOH, were used as catalysts for the transfer hydrogenation of acetophenone with i-PrOH as the
hydrogen source. Base modified complex 10 also turned out to be an active catalyst for the direct hydrogenation of the ketone
by H2 under pressure.
ꢀ 2004 Published by Elsevier B.V.
Keywords: Ruthenium; Bis(phosphines); P,N ligands; X-ray structure analysis; Hydrogenation
1. Introduction
excess of strong base, especially potassium alkoxide in
isopropanol, these complexes catalyze the reduction of
ketones by molecular hydrogen with consistently high
enantioselectivities, if appropriate chiral bis(phosphines)
and diamines are used as steering ligands [2]. RuII-cata-
lyzed hydrogenations of ketones are also exceptional
with regard to their generally high chemoselectivity for
˜C@O over ˜C@C— reduction as well as the very large
substrate-to-catalyst ratios (up to 2 · 106) that can be
reached. It has been pointed out that this extraordinary
activity alone can attract industrial attention, because
the catalytic hydrogenation of ketones by the inexpen-
sive, easy to handle, and ideally atom-economic reduc-
tant H2 to afford achiral or racemic alcohols at low
(P,N)2-coordinated transition metal complexes, in
particular Noyoriꢀs outstanding ruthenium(II) com-
pounds [RuX2{bis(phosphine)}(1,2-diamine)], have
been dominating the field of homogeneous ˜C@O
hydrogenation for several years. In the presence of an
q
Part XIV: [1a]; simultaneously Part XVII of ‘‘Chiral Bis(phos-
phines)’’ (Parts XVI and XV: [1b,1c]).
*
Corresponding author. Tel.: +49 9131 8527353; fax: +49 9131
8527387.
E-mail
address:
dahlenburg@chemie.uni-erlangen.de
(L.
Dahlenburg).
0022-328X/$ - see front matter ꢀ 2004 Published by Elsevier B.V.
doi:10.1016/j.jorganchem.2004.07.062

2
L. Dahlenburg, C. Ku¨hnlein / Journal of Organometallic Chemistry 690 (2005) 1–13
catalyst loadings could substitute the traditional
stoichiometric ˜C@O reduction by means of conven-
tional hydride transfer reagents such as NaBH4 or
LiAlH4 [3]. These are more expensive, more difficult to
handle, and produce undesired inorganic hydroxides as
compulsory by-products.
2. Experimental
2.1. General remarks
All manipulations were performed under nitrogen
using standard Schlenk techniques. Solvents were dis-
tilled from the appropriate drying agents prior to use.
IR: Mattson Polaris. NMR: Bruker DPX 300 (300.1
In continuation of our work on ˜C@C— and ˜C@O
hydrogenation catalysts based on RhI, IrI, and IrIII
complexes containing the structurally versatile chiral
cyclopentane-1,2-diyl-bis(phosphine) or b-aminophos-
phine bidentates previously prepared in our group
[1], we here describe the synthesis of some P,N-coordi-
nated RuII compounds derived from such chelate lig-
ands and give a first account of their properties as
catalysts for the reduction of the standard test sub-
strate acetophenone. In two different ways do these
complexes contrast with the advanced Noyori systems,
where the central metal is always coordinated to the
two phosphorus and nitrogen atoms of one bis(phos-
phine) and one diamine ligand: either their coordina-
tion spheres are made up of one chelating
bis(phosphine) and one aminophosphine to form
1
MHz for H, 75.5 MHz for 13C, and 121.5 MHz for
31P) at 20 2 ꢁC with SiMe4 (or the solvent) as internal
or H3PO4 as external standards (downfield positive;
‘‘m’’: deceptively simple multiplets [6]).Mass spectra:
Jeol MS 700. Published procedures were used for the
synthesis of the starting materials 1,2-C5H8(PCl2)2 [7],
1,2-C5H8 (PR2)2 (R = Me, Ph [7], NC5H10, OPh [8]),
Ph2PCH2- CMe2NH2 [1d], [RuCl2(PPh3)3] [9],
[Ru(H)(Cl)(PPh3)3] [10], [Ru(g4-C8H12)Cl2]n [11,12a],
and [Ru(g4-C8H12) {g3-(CH2)2CMe}2] [12a]. RuCl3 Æ
3H2O (Pressure Chemical Co.), KOBu-t (Aldrich),
and Ph2PCH2CH2NH2 (Fluka) were obtained
commercially.
P\P/P\N
derivatives
as
examplified
by
2.2. Metal complexes
½RuCl2f1; 2-C5H8ðPR2Þ gðPh2PCH2CR0 NH2Þꢀ (R = Me,
2
2
Ph, OPh; R0 = H, Me), or their structural motif
features the pairwise P\N coordination of two amino-
phosphine ligands such as, e.g., in [RuCl2(Ph2P-
CH2CMe2NH2)2]. The proven P-modular character
of the trans-1,2-C5H8(PR2)2 ligands, where the P–C-
or P–O-bonded structural components can be inter-
changed systematically and easily [1c], adds more
structural flexibility to the design of RuII (pre)catalysts
than hitherto reported [4]. The benefit of such com-
plexes with great diversity in the structures of their lig-
ands and coordination spheres is that they facilitate
the rational tuning of the catalysts by providing in-
sight into the relations which exist between the cata-
lytic performance and the stereoelectronic properties
of the active metal–ligand template.
One further aspect of the present investigations
was to probe such complexes in cross experiments
as (pre)catalysts for both direct ˜C@O hydrogenation
with molecular H2 as reducing agent [2] and transfer
hydrogenation with isopropanol as the source of H+
and Hꢁ equivalents [5]. As summarized in a current
in-depth paper dealing with this specific topic [3], just
a very few comparative studies have been carried out
until recently on the catalytic application of one and
the same type of complexes in direct as well as trans-
fer hydrogenation. Throughout the exploratory stud-
ies described hereafter, racemic 1,2-C5H8(PR2)2 and
achiral Ph2PCH2CR02NH2 chelating ligands rather
than optically active 1,2-C5H8(PR2)2 bis(phosphines)
[1c] and R2PCH(Ph)CH(Me)NHR aminophosphines
[1d] were employed for the sake of experimental
simplicity.
2.2.1. [RuCl2{1,2-C5H8(PMe2)2}2] (1)
Stirring an ethanol solution (ꢂ20 mL) of 151 mg
(0.72 mmol) of RuCl3 Æ 3H2O and 289 mg (1.52 mmol)
of 1,2-C5H8(PMe2)2 at reflux temperature for 5 h re-
sulted in the deposition of a beige precipitate which
was filtered off, washed with ethanol (3 · 3 mL) und
dried under vacuum. Recrystallization from CH2Cl2/
Et2O at ꢁ20 ꢁC gave the product as orange crystals
composed of equimolar quantities of the {(R,R,R,R)/
(S,S,S,S)} and (R,R,S,S) diastereomers; yield 340 mg
(85%). Anal. Found: C, 39.17; H, 7.01. Calc. for
C18H40Cl2P4Ru (552.35): C, 39.14; H, 7.30%. 1H
NMR (CDCl3): d = 1.23, 1.34 (both X3AA0X0 ꢁ s, 12
3
H each, both PCH3), 1.48–1.53, 1.67–1.75, 2.10–2.21
(all m, 4H each, all CH2), 2.32–2.34 (m, 4H, CH).
13C{1H}NMR (CDCl3): d = 5.16, 5.88 (both AA0X-
P
‘‘qui’’,
J(P,C) = 23.26 Hz each, both PCH3), 9.60,
P
9.71 (both AA0X-‘‘qui’’, J(P,C) = 25.45 Hz each, both
PCH3), 21.06, 21.11 (both s, both C4H2), 29.78, 30.12
P
(both AA0X-‘‘t’’,
J(P,C) = 6.54 Hz each, both
C3,5H2), 47.97 (AA0X-‘‘qui’’, J(P,C) = 27.61 Hz,
C1,2H). 31P{1H} NMR (CDCl3): d = 9.78, 10.43 (both
s of equal intensity; stereoisomers not assigned).
2.2.2. [Ru{g3-(CH2)2CMe}2{1,2-C5H8(PMe2)2}] (2)
A solution-suspension of 192 mg (0.60 mmol) of
[Ru(g4-C8H12){g3-(CH2)2CMe}2] and 155 mg (0.60
mmol) of 1,2-C5H8(PMe2)2 in 10 mL of hexane was
heated at reflux temperature for 5 h. The slightly cloudy
mixture was filtered and the filtrate was evaporated to
dryness. The residue was triturated with 3 mL of

L. Dahlenburg, C. Ku¨hnlein / Journal of Organometallic Chemistry 690 (2005) 1–13
3
acetone, filtered off, washed with acetone (3 · 1 mL) and
dried in vacuo. Recrystallization from hexane at
ꢁ20 ꢁC afforded the complex as a diastereomeric mix-
ture of the (K-R,R)/(D-S,S) and (D-R,R)/(K-S,S) pairs
of enantiomers; yield 159 mg (65%) of pale yellow nee-
dles pertinaciously retaining variable amounts of solvent
of crystallization. Anal. Found: C, 51.44; H, 10.15. Calc.
for C17H36P2Ru (402.12): C, 50.60; H, 8.99%. 31P{1H}
NMR (CDCl3): d = 32.31, 36.22 (both s; relative intensi-
ties 1:2; diastereomers not assigned).
2.2.5. [Ru{g3-(CH2)2CMe}2{1,2-C5H8[P(OPh)2]2}]
(5)
From 326 mg (1.02 mmol) of [Ru(g4-C8H12){g3-
(CH2)2CMe}2] and 521 mg (1.04 mmol) of 1,2-
C5H8[P(OPh)2]2: 472 mg (64%) of isomerically pure
5. Anal. Found: C, 62.34; H, 5.87. Calc. for
C37H42O4P2Ru (713.76): C, 62.26; H, 5.93%.
13C{1H} NMR (CDCl3): d = 20.7 (t, J(P,C) = 6.51
Hz, C4H2), 25.24 (s, allyl CH3), 29.14 (br, allyl CH2
trans P), 33.74 (s, C3,5H2), 44.97 (m, C1,2H), 56.80
(t, J(P,C) = 28.70 Hz, allyl CH2 trans P), 100.89 (s,
allyl CMe), 118.72, 118.82, 120.51, 121.72, 128.34,
153.66 (all m, OC6H5). 31P{1H} NMR (CDCl3):
d = 184.13 (s).
The following complexes 3–5 were obtained
analogously.
2.2.3. [Ru{g3-(CH2)2CMe}2{1,2-C5H8(PPh2)2}] (3)
From 275 mg (0.86 mmol) of [Ru(g4-C8H12){g3-
(CH2)2CMe}2] and 386 mg (0.88 mmol) of 1,2-
C5H8(PPh2)2: 432 mg (77%) of (D-R,R)/(K-S,S)-3
(X-ray structure analysis) as a greenish yellow powder
which contained only marginal amounts of the (K-
R,R)/(D-S,S) pair of enantiomers as judged from
31P{1H} NMR. Anal. Found: C, 67.79; H, 6.98. Calc.
for C37H42P2Ru (649.72): C, 68.40; H, 6.52%. 1H
NMR (CDCl3): d = 0.43, 0.48, 1.16, 1.67 (all br, 2H
each, all allyl CH2), 1.73–1.97 (m, 2H, ring CH2), 1.92
(s, 6H, allyl CH3), 2.01–2.16, 2.24–2.40 (both m, 2H
each, both ring CH2), 3.45–3.58 (m, 2H, ring CH),
6.68, 7.06, 7.36, 7.82 (all m, 4/6/6/4 H, all C6H5).
2.2.6. [RuCl2{1,2-C5H8(PMe2)2}(Ph2PCH2CH2NH2)]
(6)
A solution of 134 mg (0.33 mmol) of allyl complex 2
in 5 mL of acetone was stirred with 0.38 mL of 2 M
aqueous HCl, dissolved in 5 mL of methanol, for 2 h
at room temperature. The greenish yellow residue
remaining after removal of all volatile material in va-
cuo was re-dissolved in 2 mL of DMF and treated with
5 mL of a solution of 78 mg (0.34 mmol) of the P,N
ligand in the same solvent. Stirring for 2 h at ambient
conditions followed by evaporation to dryness left a
yellow semi-solid which was dissolved in 4 mL of ace-
tone. Dilution of the filtered mixture with 10 mL of
pentane resulted in the precipitation of the product
as a yellow solid which was re-crystallized at ꢁ20 ꢁC
from a toluene/pentane solvent mixture; yield 74 mg
(38%) of orange crystals containing variable amounts
of toluene of crystallization. Anal. Found: C, 49.70;
H, 6.17; N, 2.08. Calc. for C23H36Cl2NP3Ru (591.44):
C, 46.71; H, 6.14; N, 2.37; for C30H44Cl2NP3R, C7H8
(683.54): C, 52.71; H, 6.49; N, 2.05%. 1H NMR
(CDCl3): d = 0.85, 1.13 (both d, J(P,H) = 8.79 Hz each,
3H each, both PCH3 trans N), 1.29, 1.40 (both dd,
J(P,H) = 8.79/2.19 Hz each, 3H each, both PCH3 trans
P), 1.35–1.54, 1.55–1.79 (both m, 2H each, both C5H8
CH2), 2.04–2.35 (m, 4H, C5H8 CH2 and CH), 2.41–
2.63 (m, 2H, CH2P), 2.85–3.24 (m, 3H, NCH2 and
NH), 3.29 (br, 1H, NH), 7.03–7.80 (m, 10H, C6H5).
13C{1H} NMR (CDCl3): d = 4.16, 7.14, 7.26, 12.61
(all d, J(P,C) = 21.80, 26.16, 24.70, 28.34 Hz, all
PCH3), 20.66, 21.31 (both dd, J(P,C) = 16.00/6.49,
13C{1H} NMR (CDCl3): d = 24.51 (AA0X-‘‘t’’,
P
P
J(P,C) = 10.90 Hz, C3,5H2), 24.82 (br s, CH3), 30.59
(t, J(P,C) = 5.45 Hz, C4H2), 40.11 (AA0X-‘‘t’’,
P
J(P,C) = 18.17 Hz,
C
1,2H), 42.14 (AA0X-‘‘t’’,
J(P,C) = 9.06 Hz, allyl CH2 trans P), 56.99 (t,
J(P,C) = 23.98 Hz, allyl CH2 cis P), 93.99 (s, allyl
CMe), 126.37, 126.78 131.75, 133.24 (all m, C6H5).
31P{1H} NMR (CDCl3): d = 62.65 (s, (K-R,R)/(D-S,S)
form; ꢂ2%), 63.42 (s, (D-R,R)/(K-S,S) form; ꢂ98%).
2.2.4.
[Ru{g3-(CH2)2CMe}2{1,2-C5H8[P(NC5-
H10)2}2]] (4)
From 440 mg (1.37 mmol) of [Ru(g4-C8H12){g3-
(CH2)2CMe}2] and 639 mg (1.38 mmol) of
1,2-C5H8[P(NC5H10)2]2: 402 mg (46%) of grey (D-R,R)/
(K-S,S)-4 (X-ray structure analysis) contaminated by
less than 2% of the (K-R,R)/(D-S,S) pair of enantiomers
(NMR evidence). Anal. Found: C, 57.70; H, 9.84; N,
7.72. Calc. for C33H62N4P2Ru (677.88): C, 58.47; H,
9.22; N, 8.26%. 13C{1H} NMR (CDCl3): d = 23.96 (t,
J(P,C) = 5.09 Hz, C4H2), 24.74, 25.35, 25.84, 26.56 (all
17.44/6.57 Hz, both 1 C5H8C3,5), 28.92 (ABX-‘‘t’’,
P
J(P,C) = 13.09 Hz, C5H8 C4H2), 40.39 (ABX-dd
J(P,C) = 17.37 Hz, PCH2), 45.89 (d, J(P,C) = 16.61
P
s, allyl CH3 and piperidino C3-5H2), 30.39 (AA0X-‘‘t’’,
P
J(P,C) = 11.32 Hz, C5H8 C3,5H2), 31.51 (br, allyl
Hz, NCH2), 49.62, 50.02 (both ABX-‘‘dd’’,
P
P
CH2 trans P), 35.44 (AA0X-‘‘t’’,
J(P,C) = 10.94 Hz,
J(P,C) = 24.69, 25.14 Hz, both 1 C5H8 C1,2H),
125.73–134.57 (C6H5). 31P{1H} NMR (CDCl3): ABX
system with d(PA) = 11.50 (PMe2 trans PPh2),
d(PB) = 42.67 (PPh2), d(PX) = 30.65 (PMe2 trans
NH2), J(PA,PB) = 325.92, J(PA,PX) = 31.44, and
J(PB,PX) = 29.59 Hz.
C5H8 C1,2H), 48.72 (s, piperidino C2,6H2), 62.26 (t,
J(P,C) = 24.34 Hz, allyl CH2 cis-P), 94.68 (s, allyl
CMe). 31P{1H} NMR (CDCl3): d = 162.91 (s, (D-R,R)/
(K-S,S) form; P98%), 164.40 (s, (K-R,R)/(D-S,S) form;
62%).

4
L. Dahlenburg, C. Ku¨hnlein / Journal of Organometallic Chemistry 690 (2005) 1–13
Analogous procedures were used for the preparation
of compounds 7–9.
2.2.10. [RuCl2(Ph2PCH2CMe2NH2)2] (10)
A mixture of 150 mg (0.72 mmol) of RuCl3 Æ 3H2O,
150 mg of zinc dust (ꢂ3 equiv.), and 373 mg (1.45 mmol)
of the P,N ligand in 20 mL of THF was heated at reflux
temperature for 3 h. The residue obtained after filtration
over Al2O3, elution with 40 mL of THF, and evapora-
tion of all volatile material was triturated with metha-
nol/diethyl ether (1:1), filtered off, and thoroughly
washed with diethyl ether and methanol; yield 182 mg
(36%) of an orange powder which retained some MeOH
solvent even after prolonged drying under vacuum.
Anal. Found: C, 55.87; H, 6.31; N, 3.61. Calc.
C32H40Cl2N2P2Ru (686.57): C, 55.98; H, 5.87; N, 4.08;
for C32H40Cl2N2P2Ru, C3H4O (742.673): C, 56.60; H,
2.2.7. [RuCl2{1,2-C5H8(PPh2)2}(Ph2PCH2CH2NH2)]
(7)
From 196 mg (0.39 mmol) of 3 and 70 mg (0.31 mmol)
of Ph2PCH2CH2NH2: 151 mg (52%)of 7 as orange-yellow
crystals. Anal. Found: C, 60.37; H, 5.71; N, 1.36. Calc. for
C43H44Cl2NP3Ru (839.67): C, 61.50; H, 5.71; N, 1.67%.
1H NMR (CDCl3): d = 1.58–2.16, 2.26–2.64, 2.67–3.18
(all m, 6/3/3 H, CH2 and CH), 3.54, 4.00 (both br, 1 H
each, both 1NH), 6.67–7.68 (m, 30 H, C6H5). 31P{1H}
NMR (CDCl3): ABX system with d(PA) = 35.31,
d(PB) = 38.64 (both PPh2 in mutual trans position),
d(PX) = 46.29 (PPh2 trans NH2), J(PA,PB) = 315.72,
J(PA,PX) = 29.59, J(PB,PX) = 31.44 Hz.
1
5.97; N, 3.77%. H NMR (CDCl3): d = 1.25 (s, 12 H,
CH3), 2.83 (br, 4H, CH2), 3.76 (br, 4H, NH2), 6.85–
7.30 (m, 20 H, C6H5); (CH3OH) = 3.38 (d, J(H,
H) = 6.00 Hz). 13C {1H} NMR (CDCl3): d = 34.00 (s,
2.2.8. [RuCl2{1,2-C5H8[P(OPh)2]2}(Ph2PCH2CH2-
NH2)] (8)
P
CH3), 46.47 (AA0X-‘‘t’’,
J(P,C) = 25.40 Hz, CH2),
From 362 mg (0.50 mmol) of 5 and 121 mg (0.52 mmol)
of Ph2PCH2CH2NH2 after recrystallization of the crude
product from CH2Cl2/pentane (2:3): 151 mg (52%) of a
an ochre solid containing the mer,trans and fac,cis isomers
of 8 in an approximate 3:2 molar ratio. Anal. Found: C,
57.44; H, 5.67; N, 1.97. Calc. for C43H44Cl2NO4P3Ru
(903.09): C, 57.44; H, 4.91; N, 1.55%. 31P{1H} NMR
(CDCl3); mer,trans-8: ABX system with d(PA) = 35.06
59.69 (s, CMe2), 129.67 (s, phenyl C4), 131.20 (s, phenyl
C3,5), 135.75 (s, phenyl C2,6), 139.68 (AA0X-‘‘t’’,
P
J(P,C) = 39.24 Hz, phenyl C1). 31P{1H} NMR
(CDCl3): d = 56.54 (s).
2.3. Catalytic hydrogenation of acetophenone
2.3.1. General procedures
(PPh2),
d
(PB) = 162.38 (P(OPh)2 trans PPh2),
Solutions of the precatalysts 7, 9, or 10 (typically 2–
5 · 10ꢁ3 M) were prepared in benzene/isopropanol
(1:1; used both for transfer and direct hydrogenation),
neat isopropanol, or benzene (solvents employed in di-
rect hydrogenation experiments). KOBu-t was dissolved
in i-PrOH to make up a 0.02 M solution. Aliquots of
these solutions were then mixed to generate catalyst-
to-base ratios of 1:5 (transfer and direct hydrogenation)
or 1:100 (direct hydrogenation only). After stirring the
mixtures for 30 min at 50 ꢁC under nitrogen, acetophe-
none was added in quantities corresponding to sub-
strate-to-catalyst ratios of 200:1 for transfer
hydrogenation experiments and 2000:1 or 10000:1 for
catalytic runs carried out under the conditions of direct
˜C@O hydrogenation. The final volumes of the reaction
mixtures amounted to ꢂ10–15 mL.
d(PX) = 176.81 (P(OPh)2 trans NH2), J(PA,PB) = 439.31,
J(PA,PX) = 39.77, J(PB,PX) = 41.61 Hz. fac,cis-8: ABX
system with d(PA) = 174.39, d(PB) = 206.12 (both
P(OPh)2), d(PX) = 31.25 (PPh2), J(PA,PB) = 44.81,
J(PA,PX) = 23.06, J(PB,PX) = 18.44 Hz.
2.2.9.
NH2)] (9)
[RuCl2{1,2-C5H8(PPh2)2}(Ph2PCH2CMe2-
From 481 mg (0.74 mmol) of 3 and 190 mg (0.74
mmol) of Ph2PCH2CMe2NH2 at 70 ꢁC: 304 mg (48%)
of 9 as a beige solid which was repeatedly reprecipitated
from a CH2Cl2/hexane and CH2Cl2/Et2O solvent mix-
tures. The product pertinaciously retained variable
amounts of diethyl ether of crystallization. Anal. Found:
C, 63.00; H, 7.39; N, 1.33. Calc. for C45H48Cl2NP3Ru
(867.78): C, 62.28; H, 5.58; N, 1.61; for
C45H48Cl2NP3Ru, 0.50(C4H10O) (904.78); i.e., single
crystals grown from Et2O/toluene/hexane: C, 63.39; H
5.90; N, 1.55; for C45H48Cl2NP3Ru, 2(C4H10O)
(1016.03): C, 62.65; H, 6.75; N, 1.38%. 1H NMR
(CDCl3): d = 0.72, 1.43 (both 2, 3H each, both CH3),
1.87–2.20, 2.43–2.46, 2.62–2.86, 2.91–2.94, 3.15–3.22
(all m, 4/1/1/3/1H, CH2 and CH), 3.92, 4.20 (both br,
1H each, both 1NH), 6.86–7.33 (m, 30H, C6H5).
31P{1H} NMR (CDCl3): ABX system with
d(PA) = 32.50, d(PB) = 37.10 (both PPh2 in mutual trans
position), d(PX) = 44.93 (PPh2 trans NH2), J(PA,PB) =
314.73, J(PA,PX) = 29.60, J(PB,PX) = 31.44 Hz.
In transfer hydrogenations, stirring under an inert
atmosphere was continued at 50 ꢁC. For monitoring
the progress of the reactions, small aliquots were re-
moved at intervals, evaporated to dryness, re-dissolved
in diethyl ether, and filtered over a short silica gel col-
umn. Volatile material was distilled off and the mixtures
1
of products were analyzed by H NMR. Conversions
and product compositions were determined on the basis
of the integrations of the PhC(O)CH3 and PhCH
(OH)CH3 signals.
For direct hydrogenation of the substrate by H2, the
initial catalyst/substrate mixtures were transferred in
small Schlenk tubes equipped with a magnetic stirring

L. Dahlenburg, C. Ku¨hnlein / Journal of Organometallic Chemistry 690 (2005) 1–13
5
bar to an autoclave. The autoclave was pressurized and
vented several times with H2 (Messer-Griesheim;
99.999%), and finally pressurized to 20 bar and kept at
60 ꢁC for 3 h. Work-up and determination of conver-
sions and product compositions were done as outlined
above for the mixtures of products resulting from trans-
fer hydrogenation.
Tmin = 0.734, Tmax = 0.809), interpolation [13b] (6:
Tmin = 0.844, Tmax = 0.845), or refined [13c]
(7: Tmin = 0.723, Tmax = 0.751; 9 Æ 1/2Et2O: Tmin = 0.804,
Tmax = 0.899) methods. The structures were solved by
direct methods and subsequently refined by full-matrix
least-squares procedures on F2 with allowance for aniso-
tropic thermal motion of all non-hydrogen atoms
employing the WINGX package [14a] with the relevant
programs (SIR-97 [15], SHELXL-97 [16], ORTEP-3 [14b])
implemented therein. Carbon atom C12 of the enve-
lope-shaped cyclopentane backbone of molecule 4
showed the very common flap-like disorder between
two positions with half occupancies of the two sites. 1:
C18H40Cl2P4Ru (552.35); monoclinic, P21/n, a =
2.3.2. Results of transfer hydrogenation experiments
Ph(Me)CO/10/KOBu-t = 200:1:5 in C6H6/i-PrOH
(1:1) at 60 ꢁC for: 0.5 h, ꢂ2%; 1 h, 24%; 2 h, 68%; 3
h, 95% of PhCH(Me)OH (Fig. 8).
Ph(Me)CO/7/KOBu-t = 200:1:5 in C6H6/i-PrOH (1:1)
at 50 ꢁC for: 1 h, 84%; 3 h, 97%; 4 h, 98% of
PhCH(Me)OH (Fig. 9: -j-).
˚
8.0109(8), b = 16.903(3), c = 9.154(2) A, b = 96.23(1)ꢁ,
3
V = 1232.2(4) A , Z = 2, dcalc = 1.513 g cmꢁ3, l(Mo
˚
Ph(Me)CO/9/KOBu-t = 200:1:5 in C6H6/i-PrOH (1:1)
at 50 ꢁC for: 1 h, 29%; 1.5 h, 37%; 2 h, 42%; 2.5 h, 50%; 3
h, 53%; 3.5 h, 56%; 4 h, 58%; 4.5 h, 61%; 5 h, 65% of
PhCH(Me)OH (Fig. 9: ꢁhꢁ).
Ka) = 1.115 mmꢁ1; 2.41ꢁ 6 H 6 23.16ꢁ, 1896 reflections
collected (0 6 h 6 +8, 0 6 k 6 +18, ꢁ10 6 l 6 +10),
1756 unique (Rint = 0.0289); wR = 0.1100 for all data
and 115 parameters, R = 0.0409 for 1529 structure fac-
tors Fo > 4r(Fo). 3: C37H42P2Ru (649.72); monoclinic,
Ph(Me)CO/7/KOBu-t = 2000:1:5 in C6H6/i-PrOH
(1:1) at 50 ꢁC for: 1 h, 19%; 3 h, 50%; 5 h, 65%; 24 h,
85%; 30 h, 86% of PhCH(Me)OH (Fig. 10).
˚
C2/c, a = 17.468(5), b = 13.781(4), c = 13.316(2) A,
3
˚
b = 92.92(3)ꢁ, V = 3201(1) A , Z = 4, dcalc = 1.348 g
2.3.3. Results of direct ˜C@O hydrogenations
Ph(Me)CO/10/KOBu-t = 2000:1:5 in C6H6: 19% of
PhCH(Me)OH.
Ph(Me)CO/10/KOBu-t = 2000:1:5 in C6H6/Me2C-
DOH: 70% of PhCH(Me)OH (no deuterated product
detected).
Ph(Me)CO/10/KOBu-t = 2000:1:5 in i-PrOH: 47% of
PhCH(Me)OH.
Ph(Me)CO/10/KOBu-t = 2000:1:100 in i-PrOH: Ph-
CH(Me)OH formed in quantitative yield.
Ph(Me)CO/10/KOBu-t = 10000:1:100 in i-PrOH: 30%
of PhCH(Me)OH.
cmꢁ3, l(Mo Ka) = 0.614 mmꢁ1; 2.33ꢁ 6 H 6 30.07ꢁ,
4830 reflections collected (0 6 h 6 +24, 0 6 k 6 +19,
ꢁ18 6 l 6 +18), 4692 unique (Rint = 0.0109); wR =
0.1066 for all data and 189 parameters, R = 0.0402 for
3769 structure factors Fo > 4r(Fo). 4: C33H62N4P2Ru
˚
ꢀ
˚
(677.88); triclinic, P1, a = 10.320(3) A, b = 10.6229(8)
˚
A, c = 16.2594(9) A, a = 97.175(5)ꢁ, b = 94.691(11)ꢁ,
3
˚
c = 101.775(13)ꢁ, V = 1720.7(5) A , Z = 2, dcalc = 1.308
g cmꢁ3, l(Mo Ka) = 0.576 mmꢁ1; 2.19ꢁ 6 H 6 28.27ꢁ,
8823 reflections collected (ꢁ13 6 h 6 +13, ꢁ14 6 k
6 +14, 0 6 l 6 +21), 8530 unique (Rint = 0.0159);
wR = 0.0759 for all data and 384 parameters,
R = 0.0304 for 7276 structure factors Fo > 4r(Fo).
6 Æ C7H8: C23H36Cl2NP3Ru, C7H8 (683.54); triclinic,
2.4. X-ray structure determinations
˚
˚
˚
ꢀ
P1, a = 10.594(7) A, b = 11.336(4) A, c = 15.718(6) A,
a = 97.09(5)ꢁ, b = 104.18(6)ꢁ, c = 114.71(6)ꢁ, V =
Single crystals of 1 (0.40 · 0.30 · 0.28 mm), 3
(0.49 · 0.40 · 0.25 mm), 4 (0.50 · 0.13 · 0.13 mm),
6 Æ C7H8 (0.80 · 0.18 · 0.13 mm), 7 (0.30 · 0.10 · 0.03
mm), 9Æ1/2Et2O (0.35 · 0.23 · 0.08 mm), and 10
(0.38 · 0.30 · 0.25 mm) were obtained from the follow-
ing solvents and solvent mixtures: CDCl3 (10), toluene/
pentane (3, 6 Æ C7H8), toluene/acetone (4), CH2Cl2/
Et2O (7), CHCl3/Et2O (1), and Et2O/toluene/hexane
(9 Æ 1/2Et2O). Diffraction measurements were made at
ambient temperature or at ꢁ90 2 ꢁC (6 Æ C7H8, 9 Æ
1/2Et2O) on an Enraf-Nonius CAD-4 MACH 3 diffrac-
tometer using graphite-monochromated Mo Ka radia-
3
˚
1607(1) A , Z = 2, dcalc = 1.413
g
cmꢁ3
,
l(Mo
Ka) = 0.824 mmꢁ1; 2.16ꢁ 6 H 6 28.17ꢁ, 8303 reflections
collected (ꢁ14 6 h 6 0, ꢁ13 6 k 6 +15, ꢁ20 6 l 6
+20), 7884 unique (Rint = 0.0344); wR = 0.1841 for all
data, 324 parameters, and 4 restraints (C atoms of the
toluene molecule of crystallization kept in a common
plane), R = 0.0641 for 6426 structure factors Fo > 4r
ꢀ
(Fo). 7: C43H44Cl2NP3Ru (839.67); triclinic, P1,
˚
a = 10.226(2) A, b = 11.743(3) A, c = 18.637(7) A,
˚
˚
a = 76.81(3)ꢁ, b = 80.03(2)ꢁ, c = 66.09(2)ꢁ, V = 1984(1)
3
A , Z = 2, dcalc = 1.406 g cmꢁ3, l(Mo Ka) = 0.682
˚
tion (k = 0.71073 A): orientation matrices and unit cell
mmꢁ1; 2.19ꢁ 6 H 6 24.07ꢁ, 6504 reflections collected
(ꢁ11 6 h 6 +11, ꢁ13 6 k 6 +13, 0 6 l 6 +21), 6285 un-
ique (Rint = 0.0949); wR = 0.2067 for all data and 451
parameters, R = 0.0950 for 2757 structure factors Fo >
4r(Fo). 9 Æ 1/2Et2O: C45H48Cl2NP3Ru, 0.50 (C4H10O)
˚
parameters from the setting angles of 25 centered medi-
um-angle reflections; collection of the diffraction intensi-
ties by x scans; data either uncorrected for absorption
(3, 4) or corrected for absorption using appropriate
semi-empirical [13a] (1: Tmin = 0.777, Tmax = 0.796; 10:
ꢀ
˚
˚
(904.78); triclinic, P1, a = 12.005(2) A, b = 13.644(1) A,

6
L. Dahlenburg, C. Ku¨hnlein / Journal of Organometallic Chemistry 690 (2005) 1–13
˚
A,
c = 15.100(3)
a = 78.20(1)ꢁ,
b = 83.47(1)ꢁ,
at d(CD2Cl2) = ꢁ4.3 (free PPh3), 28.4 (Ph3P = O), and
46.7 ([RuCl2(PPh3)3]), in addition to two singlets at
d = 9.8 and 10.4. The latter two could be assigned une-
quivocally to the two {(R,R/R,R)/(S,S/S,S)} and (R,R/
S,S) diastereomeric forms of the bis(chelate) complex
trans-[RuCl2{1,2-C5H8(PMe2)2}2] (1) by deliberately syn-
thesizing that compound from RuCl3 Æ 3H2O and two
equivalents of the P,P ligand in refluxing ethanol
(Scheme 1). Samples of 1 which were recrystallized from
CHCl3/Et2O likewise contained the {(R,R/R,R)/(S,S/
S,S)} racemate and the (R,R/S,S) meso form in 1:1 molar
ratio, and molecules of the latter were shown to be present
in the specimen chosen for the single-crystal structure
determination (Fig. 1).
3
˚
c = 65.635(8)ꢁ, V = 2204.2(6) A , Z = 2, dcalc = 1.363 g
cmꢁ3, l(Mo Ka) = 0.620 mmꢁ1; 1.94ꢁ 6 H 6 30.06ꢁ,
13508 reflections collected (0 6 h 6 +16, ꢁ17 6 k
6 +19, ꢁ21 6 l 6 +21), 12938 unique (Rint = 0.0365);
wR = 0.1641 for all data and 478 parameters,
R = 0.0582 for 8880 structure factors Fo > 4r(Fo). 10:
C32H40Cl2N2P2Ru (686.57); monoclinic, P21/c, a =
˚
9.927(4), b = 27.187(6), c = 11.753(6) A, b = 92.30(4)ꢁ,
V = 3169(2) A , Z = 4, dcalc = 1.439 g cmꢁ3, l(Mo
3
˚
Ka) = 0.789 mmꢁ-1; 2.05ꢁ 6 H 6 25.47ꢁ, 6212 reflec-
tions collected (0 6 h 6 +12, 0 6 k 6 +32, ꢁ14 6 l
6 +14), 5862 unique (Rint = 0.0441); wR = 0.1896 for
all data and 352 parameters, R = 0.0570 for 4224 struc-
ture factors Fo > 4r(Fo).
The lack of success in selectively replacing two of the
three PPh3 ligands of [Ru(H)(Cl)(PPh3)3] by 1,2-
C5H8(PMe2)2 is reminiscent of an earlier study of the
reaction of [RuCl2(PPh3)3] with the P,P ligands Ph2P(CH2)n-
PPh2 (n = 1-4), from which only for n = 4 a com-
pound [RuCl2{Ph2P(CH2)nPPh2}(PPh3)] with a seven-
membered chelate ring (cf. [Ru(H)(Cl){(R)-binap}(PPh3)]
and [Ru(H)(Cl){(R,R)-1,2-C6H10 (NHPPh2)2}(PPh3)]
[17]) was isolable [18]. The failure to obtain coordinatively
unsaturated complexes for n < 4 was attributed to the de-
crease of the chelate bite angle, which results in sterically
less congested coordination spheres, accessible to further
substitution with formation of the observed coordina-
tively saturated derivatives [RuCl2{bis(phosphine)}2].
A different approach to Noyori type mixed-ligand
complexes utilizes the reaction of diallyl bis(phosphine)
complexes [Ru{g3-(CH2)2CMe}2{bis(phosphine)}] [12]
in acetone with methanolic HCl, followed by treatment
of the resulting solvated intermediates [RuCl2{bis(phos-
phine)}]n(acetone)x with one equivalent of a diamine in
DMF [19]. Following that procedure, the 1,2-
C5H8(PR2)2-substituted diallyl compounds [Ru{g3-
(CH2)2CMe}2{1,2-C5H8(PR2)2}] with R = Me (2), Ph
3. Results and discussion
3.1. Synthesis and characterization of the complexes
Morris and coworkers [17] have shown that the ruthe-
nium complexes trans-[Ru(H)(Cl){(R)-binap} (diamine)]
and trans-[Ru(H)(Cl){(R,R)-1,2-C6H10 (NHPPh2)2}
(diamine)], where diamine = H2NCMe2CMe2NH2,
(R,R)-H2NCH(Ph)CH (Ph)NH2, or (R,R)-1,2-C6H10-
(NH2)2, can be prepared from [Ru(H) (Cl)(PPh3)3] by
sequential substitution of the required bis(phosphines)
and diamines for the monodentate PPh3 ligands. At-
tempts to use this protocol for the synthesis of similar
P\P/P\N derivatives, e.g., trans-[Ru(H)(Cl){1,2-C5H8-
(PMe2)2}(Ph2PCH2CH2NH2)], by treating the tris
(triphenylphosphine) complex first with racemic 1,2-
C5H8(PMe2)2 and then with the P,N ligand failed: 31P
NMR spectra of the product mixtures obtained from
reactions of [Ru(H)(Cl)(PPh3)3] with equimolar quanti-
ties of the bis(phosphine) repeatedly showed resonances
PPh3
PPh3
rac-C5H8(PMe2)2 (1 equiv.)
H
PPh3
Cl
PPh3
Ru
Ru
+
1. THF, 2. CD2Cl2
Ph3P
Cl
Ph3P
Cl
Me
P
Me2
P
Me
P
Me2
P
Me
P
Me2
P
2 Cl
Ru
2 Cl
Ru
2 Cl
Ru
+
P
P
P
P
P
P
Cl
Cl
Cl
Me2
Me2
Me2
Me2
Me2
Me2
(R,R/R,R)/(S,S/S,S)-1
(R,R/S,S)-1
rac-C5H8(PMe2)2
refl. EtOH
(2 equiv.)
RuCl3 x 3H2O
Scheme 1.

L. Dahlenburg, C. Ku¨hnlein / Journal of Organometallic Chemistry 690 (2005) 1–13
7
(PR2)2} (Ph2PCH2CH2NH2)] [R = Me (6), Ph (7), OPh
(8)] and [RuCl2{1,2-C5H8(PPh2)2}(Ph2PCH2CMe2-
NH2)] (9) were conveniently synthesized by combining
2, 3, or 5 first with two equivalents of aqueous HCl in
methanol and then with equimolar amounts of the re-
quired b-aminophosphine in DMF (Scheme 2). Due to
the sensitiveness to acid of the P–N bonds of 4, the
method could not be used for the preparation of a pip-
eridyl-substituted derivative [RuCl2{1,2-C5H8[P(NC5-
H10)2]2}(Ph2PCH2CH2NH2)].
Because of the racemic nature of the different cyclo-
pentane-based P,P ligands the diallyl bis(phosphine)
complexes 2–5 can exist as diastereomeric (K-R,R)/(D-
S,S) and (D-R,R)/(K-S,S) pairs of enantiomers. For
the PMe2-substituted derivative 2, the presence in solu-
tion of the two diastereomeric forms was indeed evident
from the 31P NMR spectra displaying singlet resonances
at d = 32.1 and 36.2 with relative intensities close to 1:2.
In contrast, complexes 3–5 having sterically more
demanding residues in their –R2P donor groups were
shown by NMR spectroscopy to be formed with diaster-
eoselectivities exceeding 98% (see Section 2). For com-
plexes 3 and 4 the predominating stereoisomers could
be assigned as (DR, R)/(K-S,S) by X-ray structure anal-
ysis; Figs. 2 and 3.
Fig. 1. Perspective view of [RuCl2{1,2-C5H8 (PMe2)2}2] (1; meso
˚
form); selected bond lengths [A] and angles [ꢁ]: Ru–Cl1, 2.425(1); Ru1–
P1, 2.328(1); Ru1–P2, 2.331(1). Cl1–Ru1–Cl1_#, 180.0, Cl1–Ru1–P1,
90.50(5); Cl1–Ru1–P1_#, 89.50(5); Cl1–Ru1–P2, 90.60(5), Cl1–Ru1–
P2_#, 89.40(5); P1–Ru1–P1_#, 180.0; P1–Ru1–P2, 84.80(5); P1–Ru1–
P2_#, 95.20; P2–Ru1–P2_#, 180.0. Symmetry transformation used to
generate equivalent atoms _#: ꢁx, ꢁy, ꢁz.
Bis(allyl)ruthenium complexes with mondodentate
and chelating phosphorus ligands have been the subject
of a number of crystallographic studies reported in the
past [12a,20,21]. Their coordination geometries have
been described as distorted tetrahedral with the two
(3), NC5H10 (4), and OPh (5), respectively, were pre-
pared by the addition of one equivalent of the chelating
bis(phosphine) to [Ru(g4-C8H12){g3-(CH2)2CMe}2] in
hexane at reflux temperature. Subsequently, the desired
P\P,P\N-coordinated complexes [RuCl2{1,2-C5H8-
R2
P
R2
P
Ru
Ru
P
P
R2
R2
(
∆
-S,S)
(
Λ
-R,R)
rac-C5H8(PR2)2
(significant portions only for R = Me)
Ru
refl. hexane
R2
P
R2
P
Ru
Ru
P
P
R2
R2
(
∆
-R,R)
Λ
( -S,S)
R = Me (2), Ph (3), NC5H10 (4), OPh (5)
Ph
Ph
R2
P
R'
R2
P
H2
N
R'
1. HClaq (acetone/methanol)
2. Ph2PCH2CR'2NH2 (DMF)
Cl
P
R'
R'
NH2
Ru
Ru
+
P
P
P
Cl
Cl
Cl
R2
Ph2
R2
R' = H:
R = Me (6), Ph (7), OPh (8)
R' = Me: R = Ph (9)
(only for 8)
Scheme 2.

8
L. Dahlenburg, C. Ku¨hnlein / Journal of Organometallic Chemistry 690 (2005) 1–13
allylic carbon atoms defining the coordination polyhe-
dron. Notwithstanding that the distribution of the Ru–
C distances in molecules 3 and 4 shows the typical
pattern of shorter bonds to the central than to the termi-
nal allyl carbon atoms, we prefer to describe the molec-
ular structures of the two compounds as distorted
octahedral in order to emphasize the D/K helicity of
the tris(chelate) complexes and to account for the P–
Ru–P bite angles of 86.9ꢁ and 87.4ꢁ. These are similar
to those measured for analogous complexes featuring
five-membered chelate rings but are quite different from
the virtually ideal tetrahedral P–Ru–P angle of 109.9ꢁ
displayed by the monophosphine-coordinated complex
[Ru{g3-(CH2)2CH}(PPh3)2] [20a]. The Ru–P distances
(Figs. 2 and 3) fall within the range reported for closely
related [Ru(allyl){bis(phosphine)}] derivatives [21].
While the reaction sequence outlined by Scheme 2 af-
forded the bis(phosphine)-aminophosphine complexes 6,
7, and 9 as pure mer-P3 stereoisomers, the phenoxy-
substituted derivative 8 was produced as an isomeric
mixture containing the fac- and the mer-P3 forms in
close to 3:2 molar ratio. The meridional or facial
arrangement of the three –PR2 donor groups in the
coordination spheres was easily deduced from the pres-
ence or absence of a large (>300 Hz) trans P,P coupling
constant in the characteristic ABX type 31P{1H} spectra,
but it needed X-ray diffraction analysis to unambigu-
ously assign the mer complexes 6, 7, and 9 as trans
rather than cis ClRu–Cl isomers (Figs. 4–6).
Fig. 2. Perspective view of [Ru{g3-(CH2)2CMe}2{1,2-C5H8(PPh2)2}](3;
(K-S,S) form shown); selected bond lengths [A] and angles [ꢁ]: Ru1–P1,
˚
2.3094(7); Ru–C1, 2.175(3); Ru–C2, 2.228(3); Ru–C3, 2.259(3). P1–
Ru1–P1_#,86.92(4); P1–Ru1–C1, 123.71(9); P1–Ru1–C2, 91.36(9); P1–
Ru1–C3, 155.30(11); P1–Ru1–C1_#,109.61(9); P1–Ru1–C2_#,
91.96(9); P1–Ru1–C3_#, 88.46(9); C1–Ru1–C1_#, 104.3(2); C2–Ru1–
C2_#, 175.4(2), C2–Ru1–C3, 64.5(1); C2–Ru1–C3_#, 112.4(1); C3–
Ru1–C3_#, 105.2(2). Symmetry transformation used to generate equiv-
alent atoms _#:ꢁx, y, ꢁz + 1/2.
The molecular structures reveal distorted octahedral
coordination geometry around the metal centers as ex-
pected. As a consequence of the very different trans-bond
weakening influences generally exerted by P and N donor
groups, the Ru–P distances opposite the Ru–N bonds are
˚
significantly shorter (2.248–2.302 A) than those in the
˚
trans-P–Ru–P moieties (2.327–2.366 A). The Ru–N bond
˚
lengths range from 2.150 to 2.194 A, which is at the lower
end of the spread reported for the metal-to-nitrogen sep-
arations of previously characterized [RuCl2P3N] and
[RuCl2P2N2] complexes containing one or two chelated
NH2 or NHR donor functions [22]. The Ru–Cl bond
˚
lengths are 2.419 and 2.422 A in the structure of the –
PMe2-coordinated complex 6 and range from 2.415 to
Fig. 3. Perspective view of [Ru{g3-(CH2)2CMe}2{1,2-C5H8[P
(NC5H10)2]2}] (4;(D-R,R) form shown); selected bond lengths [A] and
˚
˚
2.442 A in structures 7 and 9 possessing –PPh2-substi-
angles [ꢁ]: Ru1–P1, 2.3245(5); Ru1–P2, 2.3321(8); Ru1–C1, 2.190(2);
Ru1C2, 2.246(2); Ru1–C3, 2.264(2); Ru1–C5, 2.192(2); Ru1–C6,
2.275(2); Ru1–C7, 2.224(2). P1–Ru1P2, 87.42(2); P1–Ru1–C1,
106.11(6); P1–Ru1–C2, 86.95(6); P1–Ru1–C3, 88.30(6); P1–Ru1–C5,
128.03(7); P1–Ru1–C6, 160.88(7); P1–Ru1–C7, 95.97(6); P2–Ru1–C1,
128.74(7); P2–Ru1–C2, 97.23(6); P2–Ru1–C3, 162.34(7); P2–Ru1–C5,
105.57(6); P2–Ru1–C6, 87.56(6); P2–Ru1–C7, 87.51(6); C1–Ru1–C5,
103.61(9); C2–Ru1–C3, 65.42(9), C2–Ru1–C6, 111.99(9); C2–Ru1–C7,
174.55(8); C3–Ru1–C6, 101.69(9); C3–Ru1–C7, 109.98(9); C5–Ru1–
C7, 65.38(9).
tuted chelate ligands. Though they appear to be fairly
unaffected by the higher steric demand made by –PPh2
than by –PMe2 donors in cis positions, increased steric
congestion in the coordination spheres of 7 and 9 is man-
ifested by the significant, albeit not uncommon [22], devi-
ation from linearity of the two Cl–Ru–Cl angles (164.0ꢁ
and 165.6ꢁ, set against 170.4ꢁ in 6).
Previous work of the Morris group and ourselves has
shown that amine complexes of RuII and IrI possessing
CH units adjacent to the amino group, in the presence of
strong base, tend to undergo degradation of their
H2N\NH2 or R2P\NH2 ligands on C–H bond-breaking
phosphorus atoms and the two central carbon atoms of
the allylic ligands at the corners of the tetrahedron or as
pseudo-octahedral with the P donor atoms and the outer

L. Dahlenburg, C. Ku¨hnlein / Journal of Organometallic Chemistry 690 (2005) 1–13
9
Fig. 6. Perspective view of [RuCl2{1,2-C5H8(PPh2)2}(Ph2CH2CMe2-
˚
NH2)] (9); selected bond lengths [A] and angles [ꢁ]: Ru1–Cl1, 2.427(1);
Ru1–Cl2, 2.436(1); Ru1–P1, 2.340(1); Ru1–P2, 2.302(1); Ru1–P3,
2.355(1); Ru1–N1, 2.186(3). Cl1–Ru1–Cl2, 165.58(4); Cl1–Ru1–P1,
84.77(4); Cl1–Ru1–P2, 92.90(4); Cl1–Ru1–P3, 94.13(4); Cl1–Ru1–N1,
84.37(11); Cl2–Ru1–P1, 93.12(4); Cl2–Ru1–P2, 101.02(4); Cl2–Ru1–
P3, 86.72(4); Cl2–Ru1–N1, 81.53(11); P1–Ru1–P2, 83.15(4); P1–Ru1–
P3, 174.95(4); P1–Ru1–N1, 93.78(10); P2–Ru1–P3, 101.84(4); P2–
Ru1–N1, 176.08(10); P3–Ru1–N1, 81.20(10).
Fig. 4. Perspective view of [RuCl2{1,2-C5H8(PMe2)2}(Ph2CH2CH2-
˚
NH2)] (6); selected bond lengths [A] and angles [ꢁ]: Ru1–Cl1, 2.422(2);
Ru1–Cl2, 2.419(2); Ru1–P1, 2.248(2); Ru1–P2, 2.327(3); Ru1–P3,
2.333(3); Ru1–N1, 2.194(4). Cl1–Ru1–Cl2, 170.36(5); Cl1–Ru1–P1,
93.79(7); Cl1–Ru1–P2, 88.16(7); Cl1–Ru1–P3, 88.41(7); Cl1–Ru1–N1,
85.40(12); Cl2–Ru1–P1, 93.45(7); Cl2–Ru1–P2, 86.05(7); Cl2–Ru1–P3,
97.07(7); Cl2–Ru1–N1, 87.46(12); P1–Ru1–P2, 85.63(9); P1–Ru1–P3,
96.62(9); P1–Ru1–N1, 178.74(12); P2–Ru1–P3, 176.02(4); P2–Ru1–
N1, 95.30(14); P3–Ru1–N1, 82.40(14).
pathways formulated as ‘‘dehydrogenation of the diam-
ine ligand’’ [17c] or ‘‘b-elimination of imine fragments
from initially formed amides’’ [1d]. In order to circum-
vent such difficulties ligands that lack hydrogen atoms
a to the amino group were used for mechanistic studies,
e.g., H2NCMe2CMe2NH2 [17c,17d] and Ph2PCH2-
CMe2NH2 [1d]. A dichloro ruthenium complex of the
latter, [RuCl2(Ph2PCH2CMe2NH2)2] (10), was obtained
by reacting RuCl3 Æ 3H2O with two equivalents of the
aminophosphine in THF in the presence of zinc, similar
to the preparation of the N,N-dimethyl isomer
[RuCl2(Ph2PCH2CH2NMe2)2] [24]; Scheme 3.
The coordination geometry of 10 derived from an
X-ray structure analysis corresponds to (OC-6-13) [23]
with the chloro ligands in trans positions and the two
nitrogen and phosphorus atoms cis to each other; Fig.
7. This arrangement is apparently favored over the ster-
ically less crowded geometry in which the diph-
enylphosphino groups are trans, because it puts the
strong trans influence P donors opposite the weaker
trans bond influencing amino substituents. Similar
geometries were previously reported for the structures
of some related complexes, including [RuCl2(Ph2P-
CH2CH2NMe2)2] [24,25a], [RuCl2(Ph2PCH2CH2N-
Fig. 5. Perspective view of [RuCl2{1,2-C5H8(PPh2)2}(Ph2CH2CH2-
˚
NH2)] (7); selected bond lengths [A] and angles [ꢁ]: Ru1–Cl1, 2.442(4);
Ph2
P
Ph2
P
Cl
Ru1–Cl2, 2.415(4); Ru1P1, 2.366(5); Ru1–P2, 2.288(5); Ru1–P3,
2.348(5); Ru1–N1, 2.150(12). Cl1–Ru1–Cl2, 164.0(1); Cl1–Ru1–P1,
93.89(1); Cl1–Ru1–P2, 102.2(1); Cl1–Ru1–P3, 87.0(1); Cl1–Ru1–N1,
82.4(3); Cl2–Ru1–P1, 82.9(1); Cl2–Ru1–P2, 93.1(1); Cl2–Ru1–P3,
94.8(1); Cl2–Ru1–N1, 82.1(3); P1–Ru1–P2, 85.1(2); P1–Ru1–P3,
175.1(2); P1–Ru1–N1, 93.7(3); P2–Ru1–P3, 99.5(2); P2–Ru1–N1,
175.2(4); P3–Ru1–N1, 81.6(3).
Zn / THF
Ph2PCH2CMe2NH2
Ru
Me
Me
Me
Me
RuCl3 x 3H2O
N
N
Cl
H2
H2
(10)
Scheme 3.

10
L. Dahlenburg, C. Ku¨hnlein / Journal of Organometallic Chemistry 690 (2005) 1–13
during which the catalytically active species is formed
from the precatalyst. This period is largely reduced on
changing the catalyst complex from 10 to 7, as shown
by the reaction profiles given for the latter in Figs. 9
(s/c 200) and 10 (s/c 2000). Given that the actual catalyst
results from the precatalyst complexes [RuCl2
(Ph2PCH2CMe2NH2)2] (10) and [RuCl2{1,2-C5H8
(PPh2)2}(Ph2PCH2CH2NH2)] (7) by base-induced
abstraction of HCl from the cis-ClRu–NH2-moiety as
described for both the base-modified solvent-transfer
and direct hydrogenation catalysts [(g6-arene) RuCl-
{H2N\X}]/base (X = alkoxide or amide) 3,5b,5c,5d,26
and [RuX2{bis(phosphine)}(1,2-diamine)] [2b,3,5d,
17,17c,17d,27], respectively, the substantial induction
period observed for 10 can be ascribed to steric shielding
of the amino functions by the adjacent methyl substitu-
Fig. 7. Perspective view of [RuCl2(Ph2CH2CMe2NH2)2] (10); selected
˚
bond lengths [A] and angles [ꢁ]: Ru1–Cl1, 2.428(2); Ru1–Cl2, 2.422(2);
Ru1–P1, 2.253(2); Ru1–P2, 2.264(2); Ru1–N1, 2.175(6); Ru1–N2,
2.194(6). Cl1–Ru1–Cl2, 161.61(7); Cl1–Ru1–P1, 102.01(8); Cl1–Ru1–
P2, 91.19(8); Cl1–Ru1–N1, 94.80(17); Cl1–Ru1–N2, 81.76(17); Cl2–
Ru1–P1, 88.63(8); Cl2–Ru1–P2, 101.36(7); Cl2–Ru1–N1, 81.73(16);
Cl2–Ru1–N2, 86.49(17); P1–Ru1–P2, 101.86(8); P1–Ru1–N1,
82.51(16); P1–Ru1–N2, 173.83(17); P2–Ru1–N1, 174.63(16); P2–
Ru1–N2, 82.83(18); N1–Ru1–N2, 93.0(2).
HR)2] (R = n-Pr [22] CH2Ph [22,25b]), and [RuCl2-
(Ph2PCH2CH2NH2)2] [25b]. The Ru–N distances,
˚
2.175 A opposite the longer (2.264 A) and 2.194 A oppo-
˚
˚
˚
A)
site
the
shorter
(2.253
Ru–P
bond,
reflect the expected trans-compensatory effects; within
experimental error, they appear to be at best slightly
elongated if compared to those of [RuCl2
˚
(Ph2PCH2CH2NH2)2] (2.164 and 2.180 A [25b]), which
is without bulky methyl substituents adjacent to the
amino groups. Similar to the latter, complex 10 features
a considerable distortion of the Cl–Ru–Cl axis (161.6ꢁ)
from linearity.
Fig. 8. Conversion-time profile of the transfer hydrogenation of
acetophenone catalyzed by 10-KOBu-t (1:5) at s/c 200:1; C6H6/i-PrOH
(1:1), T = 60 ꢁC.
3.2. Catalytic hydrogenation of acetophenone
Both P\P/P\N- and (P\N)2-coordinated complexes
were probed for their behavior as catalysts for ˜C@O
reduction under the conditions of direct and transfer
hydrogenation.
With acetophenone as the standard test substrate and
complexes 7, 9, and 10 as hydrogenation catalysts, trans-
fer hydrogenation experiments were carried out at 50–60
ꢁC in isopropanol/benzene (1:1), in the presence of 5
equiv. of KOBu-t as activating base. Fig. 8 shows a typ-
ical reaction profile obtained with complex 10 at a sub-
strate-to-catalyst ratio (s/c) of 200:1. Transformation of
the ketone into the alcohol is seen to follow a sigmoidal
curve with parallel consumption of the substrate and
formation of the product. The sigmoidal type of the con-
version-time curve indicates an inital incubation time
Fig. 9. Conversion-time profiles of the transfer hydrogenation of
acetophenone catalyzed by 7-KOBu-t (1:5) (-j-) and 9-KOBu-t (1:5)
(ꢁhꢁ) at s/c 200:1; C6H6/i-PrOH (1:1), T = 50 ꢁC.

L. Dahlenburg, C. Ku¨hnlein / Journal of Organometallic Chemistry 690 (2005) 1–13
11
could be due to catalyst deactivation by Ru-alkoxide
formation as a result of an acid–base reaction between
a
dihydrido
intermediate
[Ru(H)2(Ph2PCH2C-
Me2NH2)2], likely to be formed as the catalytically ac-
tive species under reaction conditions [2,17,27], and
either of the alcohols Me2CHOH and PhCH(Me)OH.
The formation of catalytically less active alkoxide
complexes has previously been described for ˜C@O
hydrogenations catalyzed by [Ru(H)2{(R)-binap}
(H2NCMe2CMe2NH2)] and attributed to enhanced
Brønsted acidity of the alcohols in solvents of higher
dielectric constant (<5 for benzene but ꢂ18 for isopro-
panol) [17d]. Considerably enhanced catalytic activity
was therefore expected for less O–H acidic isopropanol
solutions of 10. In full agreement, the addition of base
in large excess ([KOBu-t]/[10] = 100 at s/c 10000) re-
sulted in 30% conversion of acetophenone to 1-pheny-
lethanol after 3 h, corresponding to a turnover
frequency of 1000 at p(H2) = 20 bar and T = 60 ꢁC.
It would be of interest to compare these results with
those obtained for the direct hydrogenation of acetophe-
none catalyzed by [RuCl2(Ph2PCH2CH2NH2)2] and
[RuCl2(Ph2PCH2CH(Me)NH2)2], which differ only
gradually from 10 in the lower degree of methyl substi-
tution at the Ca-NH2 position. However hydrogenation
experiments with the former two compounds, which led
to quantitative transformation of Ph(Me)CO and other
ketones to the corresponding product alcohols if run for
12 h at s/c 2500 in the presence of 5 equiv. of added
KOPr-i at 20 ꢁC under ꢂ3.5 bar of H2, were not
conducted in isopropanol but in the neat ketone. Hydro-
genations performed in benzene used 2,2-dimethyl-1-
phenylpropanone as the substrate at s/c 400 [4]. Hence,
it is difficult to draw conclusions on the relative activities
of the three catalytic systems.
Fig. 10. Conversion-time profile of the transfer hydrogenation of
acetophenone catalyzed by 7-KOBu-t (1:5) at s/c 2000:1; C6H6/i-PrOH
(1:1), T = 60 ꢁC.
ents, making NH2 deprotonation by the base more
difficult.
Fig. 9 qualitatively compares the reaction profiles ob-
tained for transfer hydrogenations catalyzed by either
[RuCl2{1,2-C5H8(PPh2)2}(Ph2PCH2CH2NH2)] (7) or
[RuCl2{1,2-C5H8(PPh2)2}(Ph2PCH2CMe2NH2)] (9). The
clear-cut drop in activity observed for 9 is as expected
in that it reflects the hindered accessibility of the reactive
Ru–amide and, respectively, RuH–amine bonds
[3,5d,26] for the hydrogen donor solvent and the ketone
substrate.
Complex 10 was also inspected for its performance as
catalyst of the direct hydrogenation of acetophenone.
Exploratory reactions, which were run for 3 h at s/c
2000, in the presence of 5 equiv. of added KOBu-t, in
benzene solution heated at 60 ꢁC under 20 bar of H2,
resulted in 19% substrate conversion, corresponding to
a turnover frequency (TOF) of 126 hꢁ1. If a solvent sys-
tem composed of benzene/Me2CD OH (1:1) was used
under otherwise identical conditions, the yield of the
1-phenylethanol increased to 70% (TOF 467 hꢁ1). No
deuterated product was detected, proving that the
reaction is a net transfer of hydrogen from the gas and
not from the solvent.
Further work to exploit P\P/P\N- and (P\N)2-coor-
dinated ruthenium complexes with optically active ami-
nophosphine ligands [1d] for catalytic ˜C@O
hydrogenation is underway.
4. Supplementary material
The complex, which (as shown above) is a rather ac-
tive transfer hydrogenation (pre)catalyst giving 95%
substrate conversion after 3 h at s/c/KOBu-t = 200:1:5
in C6H6/i-PrOH (1:1) at 60 ꢁC (see Fig. 9 and Section
2), therefore also turns out as an active (pre)catalyst
for the direct hydrogenation of the ketone under compa-
rable conditions.
Conversion of the substrate to the alcohol dropped to
only 47% (TOF 313 h-1) if the reaction was carried out in
neat isopropanol, notwithstanding that this solvent is
often used as the medium of choice for Ru-catalyzed
˜C@O hydrogenations [2,27]. The reduced activity of
10 in pure i-PrOH compared to C6H6/Me2CHOH (1:1)
Crystallographic data (excluding structure factors)
for the structures reported in this paper have been
deposited at the Cambridge Crystallographic Data
Centre and allocated the deposition numbers CCDC
237806 (3), CCDC 237807 (6 Æ C7H8), CCDC 237808
(1), CCDC 237809 (4), CCDC 237810 (7), CCDC
237812 (9 Æ 1/2Et2O), and CCDC 237813 (10). Copies
of the data can be obtained free of charge at
www.ccdc.cam.ac.uk/conts/retrieving.html or on appli-
cation to CCDC, 12 Union Road, Cambridge CB2
1EZ, UK (fax: (internat.) +44-1223/336-033; e-mail:
deposit@ccdc.cam.ac.uk).

12
L. Dahlenburg, C. Ku¨hnlein / Journal of Organometallic Chemistry 690 (2005) 1–13
(c) C.E. Davies, I.M. Gardiner, J.C. Green, M.L.H. Green, N.J.
Acknowledgements
Hazel, P.D. Grebenik, V.S.B. Mtetwa, K. Prout, J. Chem. Soc.
Dalton Trans. (1985) 669–683.
Support of this work by the Deutsche Forschungsg-
emeinschaft (Bonn, SFB 583) is gratefully acknowl-
edged. We are also indebted to Mrs. S. Hoffmann for
her skilful assistance.
[8] L. Dahlenburg, C. Becker, J. Ho¨ck, S. Mertel, J. Organomet.
Chem. 564 (1998) 155–166.
[9] P.S. Hallman, T.A. Stephenson, G. Wikinson, Inorg. Synth. 12
(1970) 236–240.
[10] M. Jime´nez-Tenorio, M.C. Puerta, P. Valerga, Inorg. Chem. 33
(1994) 3515–3520.
[11] M.O. Albers, T.V. Ashworth, H.E. Oosthuizen, E. Singleton,
Inorg. Synth. 26 (1989) 68–77.
References
ˆ
[12] (a) J.P. Genet, C. Pinel, V. Ratovelomanana-Vidal, S. Mallart,
X. Pfister, M.C. Cana˜o De Andrade, J.A. Laffitte, Tetrahedron:
Asymmetry 5 (1994) 665–674;
[1] (a) L. Dahlenburg, R. Go¨tz, Inorg. Chim. Acta 357 (2004) 2875–
2880;
(b) L. Dahlenburg, A. Wuhr, Tetrahedron Lett. 44 (2003) 9279–
¨
ˆ
(b) J.P. Genet, C. Pinel, V. Ratovelomanana-Vidal, S. Mallart,
X. Pfister, L. Bischoff, M.C. Cana˜o De Andrade, S. Darses, C.
Galopin, J.A. Laffitte, Tetrahedron: Asymmetry 5 (1994) 675–
690.
9281;
(c) L. Dahlenburg, Eur. J. Inorg. Chem. (2003) 2733–2747;
(d) L. Dahlenburg, R. Go¨tz, Eur. J. Inorg. Chem. (2004) 888–905.
[2] (a) For leading overviews on enantioselective hydrogenation of
organic carbonyl compounds, see: R. Noyori, T. Ohkuma, Pure
Appl. Chem. 71 (1999) 1493–1501;
[13] (a) A.C.T. North, D.C. Phillips, F.S. Mathews, Acta Crystallogr.
Sect. A 24 (1968) 351–359;
(b) , For interpolation procedure, see:C.W. Dwiggins Jr., Acta
Crystallogr. Sect. A 31 (1975) 146–148, as well as Int. Tab. Vol. II
(1959) p. 302, Table 5.3.6 B, and Int. Tab. Vol. C (1992) p. 523,
Table 6.3.3.3;
(b) R. Noyori, T. Ohkuma, Angew. Chem. 113 (2001) 40–75;
Angew. Chem. Int. Ed. 40 (2001) 40–74;
(c) R. Noyori, M. Koizumi, D. Ishii, T. Ohkuma, Pure Appl.
Chem. 73 (2001) 227–232;
(c) N. Walker, D. Stuart, Acta Crystallogr. Sect. A 39 (1983) 159–
166.
(d) R. Noyori, Angew. Chem. 114 (2002) 2108–2123;
Angew. Chem. Int. Ed. 41 (2002) 2008–2022 (2001 Nobel Lecture
in Chemistry);
[14] (a) L.J. Farrugia, J. Appl. Crystallogr. 32 (1999) 837–838;
(b) L.J. Farrugia, J. Appl. Crystallogr. 30 (1997) 565.
[15] A. Altomare, M.C. Burla, M. Camalli, G.L. Cascarano, C.
Giacovazzo, A. Guagliardi, A.G.G. Moliterni, G. Polidori, R.
Spagna, J. Appl. Crystallogr. 32 (1999) 115–119.
[16] G.M. Sheldrick, SHELXL-97 (Release 97-2), Universita¨t Go¨ttin-
gen, 1998.
(e) J.-P. Genet, Acc. Chem. Res. 36 (2003) 908–918;
(f) D.G. Genov, D.J. Ager, Angew. Chem. 116 (2004) 2876–2879;
Angew. Chem. Int. Ed. 43 (2004) 2816–2819, and literature cited
therein.
[3] V. Rautenstrauch, X. Hoang-Cong, R. Churlaud, K. Abdur-
Rashid, R.H. Morris, Chem. Eur. J. 9 (2003) 4954–4967.
[4] While this work was in progress, we became aware of a patent
application likewise aiming at the design of Ru-centered ˜C@O
[17] (a) K. Abdur-Rashid, A.J. Lough, R.H. Morris, Organometallics
19 (2000) 2655–2657;
(b) K. Abdur-Rashid, A.J. Lough, Organometallics 20 (2001)
1047–1049;
hydrogenation (pre)catalysts with
a greater diversity in the
(c) K. Abdur-Rashid, M. Faatz, A.J. Lough, R.H. Morris, J. Am.
Chem. Soc. 123 (2001) 7473–7474;
structures of their ligands and coordination spheres: V. Rau-
tenstrauch, R. Challand, R. Churlaud, R.H. Morris, K. Abdur-
Rashid, E. Brazi, H. Mimoun (Firmenich, S.A., Switz) PCT Int.
Appl. WO 2002022526 A2 (2002) 42 pp. [Chem. Abstr. 136 (2003)
264828].
(d) K. Abdur-Rashid, S.E. Clapham, A. Hadzovic, J.N. Harvey,
A.J. Lough, R.H. Morris, J. Am. Chem. Soc. 124 (2002) 15104–
15118.
[18] C.W. Jung, P.E. Garrou, P.R. Hoffman, K.G. Caulton, Inorg.
Chem. 23 (1984) 726–729.
[5] For leading overviews on ˜C@O transfer hydrogenation, see: (a)
G. Zassinovich, G. Mestroni, S. Gladiali, Chem. Rev. 91 (1992)
1051–1069;
[19] H. Doucet, T. Ohkuma, K. Murata, T. Yokozawa, M. Kozawa,
E. Katayama, A.F. England, T. Ikariya, R. Noyori, Angew.
Chem. 110 (1998) 1792–1796;
(b) R. Noyori, S. Hashiguchi, Acc. Chem. Res. 30 (1997) 97–102;
(c) M.J. Palmer, M. Wills, Tetrahedron: Asymmetry 10 (1999)
2045–2061;
Angew. Chem. Int. Ed. 37 (1998) 1703–1707.
[20] (a) A.E. Smith, Inorg. Chem. 11 (1972) 2306–2310;
(b) R.A. Marsh, J. Howard, P. Woodward, J. Chem. Soc. Dalton
Trans. (1973) 778–783.
(d) R. Noyori, M. Yamakawa, S. Hashiguchi, J. Org. Chem. 66
(2001) 7931–7944;
(e) K. Everaere, A. Mortreux, J.-F. Carpentier, Adv. Synth.
Catal. 345 (2003) 67–77;
[21] (a) F. Hapiot, F. Agbossou, C. Meliet, A. Mortreux, G.M.
Rosair, A.J. Welch, New. J. Chem. (Nouv. J. Chim.) 21 (1997)
1161–1172;
(f) D. Cuervo, M.P. Gamasa, J. Gimeno, Chem. Eur. J. 10 (2004)
425–432, and references cited therein;
(b) K.S. MacFarlane, S.J. Rettig, Z. Liu, B.R. James, J.
Organomet. Chem. 557 (1998) 213–219;
(g) P. Brandt, P. Roth, P.G. Andersson, J. Org. Chem. 69 (2004),
ASAP Web Release Date: 01-Jul-2004; and references cited
therein.
(c) C. Six, B. Gabor, H. Go¨rls, P. Mynott, P. Philipps, W.
Leitner, Organometallics 18 (1999) 3316–3326;
(d) G. Fries, K. Ilg, M. Pfeiffer, D. Stalke, H. Werner, Eur. J.
Inorg. Chem. (2000) 2597–2601;
[6] (a) R.K. Harris, Can. J. Chem. 42 (1964) 2275–2281;
(b) D.H. Redfield, J.H. Nelson, L.W. Cary, Inorg. Nucl. Chem.
Lett. 10 (1974) 727–733;
(e) H. Werner, G. Fries, B. Weberndo¨rfer, J. Organomet. Chem.
607 (2000) 182–193;
(c) A.W. Verstuyft, J.H. Nelson, L.W. Cary, Inorg. Chem. 15
(1973) 732–734.
(f) D.C. Smith Jr., J. Cadoret, L. Jafarpour, E.D. Stevens,
S.P. Nolan, Can. J. Chem. (Rev. Can. Chim.) 79 (2001) 626–
631.
[7] (a) D.L. Allen, V.C. Gibson, M.L.H. Green, J.F. Skinner, J.
Bashkin, P.D. Grebenik, J. Chem. Soc. Chem. Commun. (1983)
895–896;
[22] M.S. Rahman, P.D. Prince, J.W. Steed, K.K. Hii, Organometal-
lics 21 (2002) 4927–4933.
(b) M.L.H. Green (British Petroleum Co. PLC, UK), Eur. Pat.
Appl. EP 117156 A1 (1984) 32 pp. [Chem. Abstr. 102 (1985) 6809];

L. Dahlenburg, C. Ku¨hnlein / Journal of Organometallic Chemistry 690 (2005) 1–13
13
[23] (a) T.E. Sloan, Top. Stereochem. 12 (1981) 1–36, [Top. Inorg.
Organomet. Stereochem.];
(b) U. Bunzli-Trepp, Handbuch fur die systematische Nomenkl-
(b) D.G.I. Petra, P.C.I. Kamer, P.W.N.M. van Leeuwen, K.
Goubitz, A.M. van Loon, J.G. de Vries, H.E. Schoemaker, Eur.
J. Inorg. Chem. (1999) 2335–2341;
¨
¨
atur der Organischen Chemie, Metallorganischen Chemie und
Koordinationschemie, Logos Verlag, Berlin, 2001, Sect. A.6.4.
[24] J.-Y. Shen, C. Slugovc, P. Wiede, K. Mereiter, R. Schmid, K.
Kirchner, Inorg. Chim. Acta 268 (1998) 69–76.
(c) D.A. Alonso, P. Brandt, S.J.M. Nordin, P.G. Andersson, J.
Am. Chem. Soc. 121 (1999) 9580–9588;
(d) M. Yamakawa, H. Ito, R. Noyori, J. Am. Chem. Soc. 122
(2000) 1466–1478;
[25] (a) Z.J. Guo, A. Habtemariam, P.J. Sadler, B.R. James, Inorg.
Chim. Acta 173 (1998) 1–7;
(e) K. Everaere, A. Mortreux, M. Bulliard, J. Brussee, A. van der
Gen, G. Nowogrocki, J.-F. Carpenier, Eur. J. Org. Chem. (2001)
275–291;
(b) R. Morris, A. Habtemariam, Z. Guo, S. Parsons, P.J. Sadler,
Inorg. Chim. Acta 339 (2002) 551–559.
`
(f) O. Pamies, J.-E. Ba¨ckvall, Chem. Eur. J. 7 (2001) 5052–
5058.
[26] (a) K.-J. Haack, S. Hashiguchi, A. Fujii, T. Ikariya, R. Noyori,
Angew. Chem. 109 (1997) 297–300;
Angew. Chem. Int. Ed. Engl. 36 (1997) 285–288;
[27] C.A. Sandoval, T. Ohkuma, K. Muna˜iz, R. Noyori, J. Am.
Chem. Soc. 125 (2003) 13490–13503.