Chiral oxo- and oxy-functionalized diphosphane ligands derived from camphor for rhodium(I)-catalyzed enantioselective hydrogenation

Article abstract of DOI:10.1002/1099-0690(200301)2003:1<138::AID-EJOC138>3.0.CO;2-O

The synthesis of two series of diastereomeric oxo- and oxysubstituted diphosphanes 7a-9a and 7b-9b, as well as an analogous nonfunctionalized diphosphane 17, was performed starting from (R)-camphor. The new diphosphanes were used as ligands in the enantioselective rhodium(I)-catalyzed hydrogenation of functionalized olefins - α- and β-de hydroamino acids and their esters - in order to elucidate the effect of the oxo- and oxy-functional groups. The enantioselectivities, ranging from 2-90% ee, and the rates were strongly dependent on the type and relative position of the oxo or oxy substituent in the catalyst. Possible explanations for the effects are given. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003.

Full text of DOI:10.1002/1099-0690(200301)2003:1<138::AID-EJOC138>3.0.CO;2-O

FULL PAPER
Chiral Oxo- and Oxy-Functionalized Diphosphane Ligands Derived from
Camphor for Rhodium( )-Catalyzed Enantioselective Hydrogenation
I
Igor V. Komarov,*[a] Axel Monsees,[b] Anke Spannenberg,[a] Wolfgang Baumann,[a]
Ute Schmidt,[a] Christine Fischer,[a] and Armin Börner*[a,c]
Keywords: Asymmetric synthesis / Camphor / Hydrogenation / Phosphanes / Rhodium
The synthesis of two series of diastereomeric oxo- and oxy-
effect of the oxo- and oxy-functional groups. The enantiose-
substituted diphosphanes 7a−9a and 7b−9b, as well as an
lectivities, ranging from 2−90% ee, and the rates were
analogous nonfunctionalized diphosphane 17, was per- strongly dependent on the type and relative position of the
formed starting from (R)-camphor. The new diphosphanes oxo or oxy substituent in the catalyst. Possible explanations
were used as ligands in the enantioselective rhodium(
I)-cata-
for the effects are given.
lyzed hydrogenation of functionalized olefins − α- and β-de- ( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
hydroamino acids and their esters − in order to elucidate the
Germany, 2003)
Introduction
The vast majority of ligands used in Rh-, Ir- and Ru-
based catalysts for enantioselective hydrogenation and
many other reactions are chiral diphosphanes,[1] which usu-
ally possess no additional functional groups.[2] Their stereo-
differentiating properties are believed to be caused mainly
by steric repulsion. In contrast, some very efficient and se-
lective natural catalysts Ϫ enzymes Ϫ are highly func-
tionalized. Attractive interactions between enzyme and sub-
strate, primarily hydrogen bonds and electrostatic interac-
tions, are responsible for their amazing selectivity and react-
ivity.[3] There are only a few examples of homogeneous
hydrogenation catalysts which exploit additional functional
groups in ligands to achieve high enantioselectivity. For ex-
ample, asymmetric catalytic hydrogenation of prochiral de-
hydroacylamino acids and dehydrodipeptides in the pres-
ence of RhI complexes bearing ligands 1 and 2 was sug-
gested to proceed with high enantioselectivity because of
electrostatic attraction between carboxylate ion and the
protonated amino group in the ligands.[4]
The presence of hydroxy and alkoxy groups in diphos-
phane ligands such as 3,[5]
4 5
(RoPHOS)[6] and
(BASPHOS)[7] were shown to influence Ϫ dramatically in
some cases Ϫ the selectivity and activity of the RhI cata-
lysts. Mechanistic studies provided evidence that these ef-
fects might be caused by secondary interactions of the hy-
droxy groups with the metal centre or the substrate.[8] In
all cases the effects were strongly dependent on the spatial
orientations of the additional functional group(s) and the
substrate employed.
It is obvious that the synthesis of functionalized ligands
and their employment in catalysis not only provides a great
academic challenge but also has unexplored industrial po-
tential. Up to now, however, the design of functionalized
diphosphane ligands has received neither theoretical nor
sufficient empirical support, and so the synthesis of series
of structurally related ligands and their subsequent screen-
ing in metal-catalyzed asymmetric reactions is currently the
most practical approach to find efficient catalysts and elu-
cidate the role of the additional functionality.
[a]
Institut für Organische Katalyseforschung,
Buchbinderstraße 5Ϫ6, 18055 Rostock, Germany
[b]
Project House Catalysis, Degussa AG, Industriepark Höchst,
G 830,
65926 Frankfurt am Main, Germany
[c]
Fachbereich Chemie der Universität Rostock,
Albert-Einstein-Straße 3a, 18059 Rostock, Germany
E-mail: ik214@mail.yahoo.com
armin.boerner@ifok.uni-rostock.de
138
 2003 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim 1434Ϫ193X/03/0101Ϫ0138 $ 20.00ϩ.50/0 Eur. J. Org. Chem. 2003, 138Ϫ150

Chiral Oxo- and Oxy-Functionalized Diphosphane Ligands
FULL PAPER
This study was directed toward the synthesis of a set of
isomeric chiral chelating diphosphane ligands bearing dif-
ferent oxo- and oxy-functionalities. Hydroxy, oxo and ace-
tal groups were chosen, representing functional groups with
different hydrogen bond donor and acceptor properties.
The new chiral diphosphanes were prepared from camphor
and subsequently tested in the RhI-catalyzed enantioselec-
tive hydrogenation of functionalized olefins of academic
and industrial relevance.
Scheme 1
Another problem is the need to use rather drastic condi-
tions to perform nucleophilic substitution of the bromo
substituents at the neopentyl-like positions in 6a and 6b.
This results in side reactions caused by the basic nature of
the nucleophilic phosphide reagent, even if the carbonyl
group is protected. To solve this problem, both isomeric
dibromocamphors were transformed into the correspond-
ing diiodocamphors 10a and 10b (Scheme 2). The weakly
basic and highly nucleophilic iodide ion smoothly substi-
tutes bromide in the dibromocamphors. On the other hand,
the diiodocamphors, possessing good leaving groups, are
much more reactive than the bromo derivatives in reactions
with phosphides. With the above considerations in mind,
the following syntheses of the target functionalized ligands
(7a, 7b; 8a, 8b; 9a, 9b) were performed.
Results and Discussion
Synthesis of New Ligands
The rich chemistry of camphor[9] and its availability from
the ‘‘chiral pool’’ prompted us to use it as a starting mat-
erial for the synthesis of new diphosphanes. Although cam-
phor is widely used in syntheses of chiral auxiliaries,[10] and
in total syntheses of natural compounds,[11,12] its potential
in homogeneous catalysis is largely unexplored, as already
pointed out in a recent paper.[13] Camphor can easily be
converted into many mono- and dibromo-substituted deriv-
atives, which are useful intermediates in ligand synthesis.[14]
Especially attractive for the synthesis of functionalized di-
phosphanes are 9,10- and 8,10-dibromocamphors 6a and
6b, available from camphor by stereospecific bromination/
rearrangement reaction sequences.[15] Thus, the bromome-
thyl groups in 6a and 6b can be advantageously used for
nucleophilic substitution with phosphorus nucleophiles to
obtain 7a and 7b, whereas the carbonyl group can be fur-
ther converted into, for example, the hydroxy group by re-
duction, giving rise to hydroxy derivative of the diphos-
phanes 8a and 8b, or transformed into an acetal group
(9a, 9b).[16]
Scheme 2
The configurations at C-2 in the hydroxy derivatives 8a
and 8b were confirmed by NOE experiments (Figure 1).
Figure 1. Strong positive NOEs between 2-H and endo-6-H
For the synthesis of the hydroxy derivative of the diphos-
phane 8a we also used an alternative strategy, which in-
On the way to the desired ligands we were confronted volved reduction of the carbonyl group in 9,10-diiodo-
with two major obstacles. The first was the Grob-type frag- camphor 10a and protection of the formed hydroxy group
mentation shown for 9,10-dibromocamphor in Scheme 1. prior to the nucleophilic substitution. As in the case of the
This fragmentation is a well-known reaction of α,α-disub- oxo derivatives 7a and 7b, reduction of 10a by LiAlH4 pro-
stituted β-halocarbonyl compounds.[17] This forced us to ceeded with satisfactory exo diastereoselectivity. The hy-
use carbonyl group protection in the early stages of the syn- droxy group in 12 was then protected by SiEt3 as shown
thesis.
recently for 10-iodoisoborneol.[13] Subsequent treatment
Eur. J. Org. Chem. 2003, 138Ϫ150
139

I. V. Komarov, A. Börner et al.
FULL PAPER
with Ph2PLi and deprotection yielded the hydroxy
derivative of the diphosphane 8a (Scheme 3).
Scheme 5
In the 9,10-(PPh2)2-substituted series we analyzed the op-
tical purity of ligand 7a, prepared for the analysis in both
enantiomeric forms, starting from (R)- and (S)-camphor. In
the presence of 1 equiv. of 18, each enantiomer of 7a gave
a
31P{H} NMR spectrum containing two singlets with in-
tegral intensity ratio of approximately 1:1, corresponding to
two palladium complexes. 31P-31P coupling was not ob-
served, probably due to rapid site-site exchange of the li-
gands. A similar effect has been reported for (R,R)-
DIOP.[20] For each enantiomer of 7a, the signals character-
izing the other enantiomer were below the detection limits,
which establishes the high (Ͼ 98% ee) optical purity of the
ligand. It is surprising that the optical purity of the ligand
prepared from (S)-camphor turned out to be higher than
that of the (S)-camphor itself. This is presumably due to the
recrystallization of the intermediate bromocamphors dur-
ing the synthesis of the ligand.
The analysis of the 8,10-(PPh2)2-substituted ligands was
complicated by the additional coordination of the oxo- and
oxy-functionality with Pd. We therefore prepared the oxa-
tricyclic ligand 19, as previously described from 8,10-di-
bromocamphor, again in both enantiomeric forms
(Scheme 6).[14] No crystallization was used during the syn-
thesis, in order to avoid enantiomer enrichment. The optical
purity of 19 was determined without any problems with the
help of the palladium complex 18. Thus, the 31P{H} NMR
spectrum of this compound in the presence of 0.5 equiv. of
18 revealed two pairs of doublets [δ ϭ 36.3 and 8.9 ppm,
J(31P-31P) ϭ 54.1 Hz, and δ ϭ 34.4 and 11.9 ppm, J(31P-
31P) ϭ 51.3 Hz, in a ratio of 1:0.86], which correspond to
two complexes formed between the enantiomerically pure
ligand and the palladium reagent. Signals corresponding to
the complexes derived from the other enantiomer of 19 [δ ϭ
33.8 and 10.6 ppm, J(31P-31P) ϭ 54.1 Hz, and δ ϭ 29.0
and 17.4 ppm, J(31P-31P) ϭ 54.1 Hz] had overall integral
intensities of less than 2%.
Scheme 3
For the synthesis of the analogous unfunctionalized li-
gands we planned to convert the carbonyl groups in the
dibromocamphors into CϭCH2 groups by use of a ‘‘Ti0’’
reagent,[18] followed by hydrogenation of the double bond
and subsequent treatment with Ph2PLi. This was achieved
only when starting from the 9,10-dibromocamphor 6a
(Scheme 4). Ligand 17 was obtained as a mixture of endo
and exo isomers in a ratio of 1:3 (the configurations at C-2
in both isomers were verified by NOE experiments). Treat-
ment of the titanium-based reagent with the 8,10-dibromo-
camphor gave a complex mixture of unidentified products.
Scheme 4
Optical Purities of Ligands
The optical purity of commercially available natural cam-
phor is not perfect: 99.2Ϫ99.5% ee for (R)-camphor and
78.6Ϫ92.8% ee for (S)-camphor.[19] In most cases, high op-
tical purity of ligands is desired for achievement of max-
imum enantioselectivity in catalytic reactions. The optical
purities of some diphosphanes prepared both from 9,10-
and 8,10-dibromocamphors were therefore determined.
The analysis was carried out by 31P{H} NMR spectro-
scopy in the presence of (Ϫ)-bis(µ-chloro)bis[(R)-
dimethyl(α-methylbenzyl)aminato-C2,N]dipalladium()
(18). This method has been widely used for analysis of the
optical purities of chelating diphosphanes.[20] In the case of
C1-symmetrical diphosphane ligands as studied in this
work, each enantiomer of a ligand forms two complexes
with 18 (Scheme 5). Usually all four complexes formed
from a mixture of enantiomers have different 31P{H} NMR
Scheme 6
We confirmed the high optical purity of 19 by an inde-
shifts (singlets or pairs of doublets). The optical purity can pendent method. The enantiomeric diphosphanes were ox-
be determined by integration of the signals.
idized to the corresponding phosphane oxides with H2O2.
140
Eur. J. Org. Chem. 2003, 138Ϫ150

Chiral Oxo- and Oxy-Functionalized Diphosphane Ligands
FULL PAPER
HPLC analysis of the phosphane oxides on a chiral column
showed that both enantiomers were of high optical purity.
We were unable to detect the opposite enantiomer in each
case (Ͼ 98% ee).
Since the analysis of 19 unambiguously showed its high
optical purity, this should also hold for ligands 7bϪ9b. It is
highly unlikely that any racemization would happen during
the synthesis of 8,10-diiodocamphor, acetal formation, nu-
cleophilic substitution with Ph2PLi and hydrolysis of the
acetal [in fact, it did not happen in the 9,10-(PPh2)2-substi-
tuted series of ligands; see the above analysis of 7a]. We thus
assume that the optical purity of 8,10-(PPh2)2-substituted
ligands should be similarly high as that of the ligand 19.
Synthesis and Structure of Precatalysts
The new diphosphanes 7a, 7b, 8a, 8b, 9a, 9b and 17 (L)
were used to prepare RhI complexes of the type
[Rh(L)(COD)]BF4 by mixing [Rh(acac)(COD)] and the li-
gand in THF, followed by addition of 1 equiv. of HBF4. In
all cases the crystalline complexes were formed in moderate
yields. They were fully characterized by spectroscopic
methods. It was found in the case of the nonfunctionalized
ligand 17 that crystallization yielded the complex with only
the exo isomer of the ligand, although a mixture of the endo
and exo isomers had been used for the synthesis.
For RhI chelate complexes with diphosphanes of C1 sym-
metry there are typically two sets of double doublets in the
31P{H} NMR spectra. Relevant chemical shifts and coup-
ling constants are listed in Table 1. Resonances in the region
between δ ϭ 9 and 16 ppm are typical for phosphorus
atoms involved in seven-membered chelate rings with
Rh.[21] Interestingly, the signals of the COD methine groups
were not visible in the 1H NMR spectrum of
[Rh(8b)(COD)]BF4, whereas the CH2 groups appeared as
two very broad signals. This indicates slow (on the NMR
timescale) motion of the COD ligand in the coordination
sphere of Rh.[22]
Figure 2. Crystal structures of cations of [Rh(7a)(COD)]BF4 (a)
and [Rh(7b)(COD)]BF4 (b); hydrogen atoms are omitted for clarity,
the thermal ellipsoids are shown at 30% level of probability;
˚
selected bond lengths [A] and angles [°]: [Rh(COD)(7a)]BF4:
RhϪP(1) 2.327(2), RhϪP(2) 2.343(2) P(1)ϪRhϪP(2) 92.94(9),
C(9)ϪC(10)ϪC(11)ϪC(31) 62.57; [Rh(COD)(7b)]BF4: RhϪP(1)
2.3555(15), RhϪP(2) 2.3341(17), P(1)ϪRhϪP(2) 93.72(7),
C(24)ϪC(23)ϪC(21)ϪC(22) 56.53
Table 1. 31P{H} NMR spectroscopic data for complexes
[Rh(L)(COD)]BF4 (solvent: CD3OD)
Ligand (L) δ [ppm] ∆δ [ppm] J(103Rh-31P) [Hz] J(31P-31P) [Hz]
structures of the cations, as shown in Figure 2. All RhϪP
and RhϪC bond lengths and angles of the examined com-
plexes are closely related to those in other (diphosphane)-
rhodium() precatalysts.[23]
One particular aspect of the crystal structures, namely
the conformation of the seven-membered chelate rings,
needs special attention. The CϪCϪCϪC unit in these rings
is a part of the rigid camphor skeleton, which fixes this
unit in the synclinal conformation (the CϪCϪCϪC torsion
angle is about 60°). In most (diphosphane)Rh precatalysts
7a
8a
9a
17
7b
8b
9b
12.8, 9.3
15.2, 14.3
14.3, 13.5
15.6, 14.3
13.2, 3.7
14.9, 5.3
15.6, 5.2
3.5
0.9
0.8
1.3
9.5
9.6
10.4
142.9
142.3
142.8
142.0
142.5
141.5
142.9
36.1
42.5
42.1
43.0
32.3
31.9
33.3
Crystals suitable for X-ray analysis were obtained for the used for asymmetric hydrogenation, in which this unit is
isomeric pair of the precatalysts [Rh(7b)(COD)]BF4 and also fixed by trans fusion of a five- and a seven-membered
[Rh(7a)(COD)]BF4 by slow crystallization from methanol. ring (e.g., DIOP and its analogues) this conformation is
A single-crystal X-ray structural analysis established the closer to anticlinal, and the CϪCϪCϪC torsion angles are
Eur. J. Org. Chem. 2003, 138Ϫ150
141

I. V. Komarov, A. Börner et al.
FULL PAPER
a beneficial influence on the degree of enantioselection in
the hydrogenation.[26]
Hydrogenation Results
The new precatalysts based on 9,10- and 8,10-(PPh2)2-
substituted ligands were employed for the hydrogenation of
(Z)-2-acetamidocinnamic acid (AH) and its methyl ester
(AMe), as well as of (Z)- and (E)-3-acetamidoacrylic acids
and their methyl esters[27] (Scheme 7). The reactions were
performed under identical conditions: at 1 bar pressure of
hydrogen, 25 °C, in methanol or dichloromethane, with
molar catalyst/substrate ratios of 1:100 or 1:1000.
Figure 3. Conformations around a C(4)ϪC(5) bond of the seven-
membered chelate ring in different RhI precatalysts
usually more than 60° (Figure 3, see also the caption of Fig-
ure 2).[24]
Consequently, unlike most of related (diphosphane)Rh
complexes possessing seven-membered chelate rings in
twist-chair or boat B4ϪB5 conformations,[25] this ring in
complexes [Rh(7a)(COD)]BF4 and [Rh(7b)(COD)]BF4 ad-
opts the boat B3 conformation (Figure 4, λ-chiral in the for-
mer complex and δ-chiral in the latter). As can be seen from
Figure 4, the B3 conformation is characterized by a slightly
distorted chiral alternating pseudoaxial-pseudoequatorial
arrangement of the P-phenyl groups, which expose their
edge or face to the Rh atom depending on whether they are
pseudoaxial or pseudoequatorial (edge-on-face-on model).
As shown in many examples, this arrangement may have
Scheme 7
Striking differences between pairs of isomeric catalysts
in terms of activity and enantioselectivity were observed.
Catalysts derived from ligands 7a, 8a and 9a are highly act-
ive and give moderate or good enantioselectivies in the hy-
drogenation of AH and AMe (Table 2). The hydrogenation
proceeded within a few minutes, with enantioselectivities of
up to 90% ee. A reduction in the catalyst/substrate molar
ratio to 1:1000 reduced the rates of the hydrogenation only
slightly, and with almost no loss of selectivity. The high en-
antioselectivity of these catalysts might be explained in
terms of the edge-on-face-on model discussed above. Chiral
alternating (pseudo)axial-(pseudo)equatorial arrangements
of the Ph substituents around Rh as found in the crystal
structure for one of the precatalysts are likely to be main-
tained in catalytic intermediates, due to the rigidity of the
camphor skeleton, and might have a decisive influence on
the stereochemical outcome of the hydrogenation. Similarly
high enantioselectivities have previously been achieved with
related ligands forming seven-membered chelate rings and
showing alternating edge-to-face disposition of the P-phenyl
groups.[28] It should be pointed out, however, that while in
most known cases the λ-chiral conformation of the precata-
lysts produces (S) isomers of the α-amino acids,[1] our pre-
catalyst [Rh(7a)(COD)]BF4 has λ-chiral conformation (see
Figure 4) but gives (R)-amino acids (Table 2). β-Amino acid
precursors were also reduced with good ees (up to 82% ee).
In general, a change in the solvent from methanol to dichlo-
romethane affected the enantioselectivity. In agreement
with some literature data,[29] (Z) isomers of the β-amino
acid precursors were hydrogenated with better selectivity in
MeOH, the corresponding (E) isomers in CH2Cl2. Close
inspection of the hydrogenation data for complexes bearing
Figure 4. Fragments of the structures of complexes [Rh(7a)-
(COD)]BF4 (a) and [Rh(7b)(COD)]BF4 (b) showing conformations
of the seven-membered rings and position of their substituents Ϫ
phenyl rings and CH3 (CH2) groups of the camphor skeleton; hy-
drogen atoms are omitted for clarity
142
Eur. J. Org. Chem. 2003, 138Ϫ150

Chiral Oxo- and Oxy-Functionalized Diphosphane Ligands
FULL PAPER
Table 2. Hydrogenation results of various substrates with complexes Rh[La(COD)]BF4 where La is a ligand of the 9,10-(PPh2)2-substituted
series (7aϪ9a, 17)[a]
Substrate
Solvent
Ligand (La)
7a
Conv.
[%]
8a
Conv.
[%]
9a
Conv.
[%]
17
Conv.
[%]
Time
[min]
ee
[%]
Time
[min]
ee
[%]
Time
ee
[%]
Time
[min]
ee
[%]
[min]
2.5
4
21
MeOH
MeOH[b]
CH2Cl2
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
2
100
100
100
100
100
100
96
70
(R)
78
(R)
73
(R)
90
(R)
80
(R)
68
(S)
82
(S)
13
(R)
7
(S)
41
(S)
65
(S)
19
1
100
100
93
80
(R)
77
100
100
100
100
100
97
81
(R)
65
(R)
83
(R)
81
(R)
88
(R)
40
(S)
69
(S)
25
(R)
17
(R)
36
(S)
52
(S)
15
1.5
100
79
(R)
7
2
(R)
81
5
15
3
3
100
82
(R)
(R)
90
(R)
88
22
23
24
25
2
2
100
100
87
1.5
8
7
6
(R)
58
(S)
71
300
30
120
80
800
400
600
170
40
350
30
180
30
51
62
61
(S)
77
87
95
(S)
22
(R)
9
(R)
36
(S)
27
(S)
99
80
100
48
100
160
1100
1400
2400
100
48
56
250
1400
1400
220
93
85
75
180
19
25
27
(S)
45
31
19
25
1400
(S)
8
(S)
26
69
71
43
(S)
(S)
(S)
[a]
[b]
Molar catalyst-substrate ratio 1:100 (if otherwise not stated), 1 bar H2, 25 °C. Molar catalyst-substrate ratio 1:1000.
9,10-(PPh2)2-substituted ligands reveals their similarity in ciple for complexes with ligands 7bϪ9b, while in the case
catalytic performance, despite the fact that they possess dif- of isomeric ligands 7aϪ9a it is impossible for steric reasons.
ferent functional groups. Moreover, the rates of hydrogena-
There is spectroscopic evidence for the functional group
tion and the ees were essentially the same with the complex coordination to Rh in solutions of complexes with diphos-
based on the nonfunctionalized ligand 17. Therefore, we as- phanes 7bϪ9b. In the series of 9,10-(PPh2)2-substituted li-
sume that none of the O groups in this series of catalysts gands (7a, 8a, 9a and 17) both phosphorus atoms are char-
participates actively in the enantioselective step.
acterized by similar shifts (∆δ ϭ 0.9Ϫ3.5, Table 1) indicat-
Analogous ligands in the 8,10-(PPh2)2-substituted series ing similar chemical environments around the phosphorus
(7b, 8b and 9b) form catalysts with entirely different and nuclei. In contrast, the 31P NMR shifts of the Rh precata-
inferior hydrogenation properties (Table 3). The highest ees lysts with ligands based on the 8,10-(PPh2)2-substituted
were achieved with the oxo derivative 7b and AH as the camphor skeleton (7b, 8b, 9b) are quite different, ∆δ being
substrate (73% ee), but this value is still lower than that in the range of 9.5Ϫ10.4. One of the signals is essentially
obtained in the case of the corresponding isomeric ligand the same as for the complexes with 9,10-(PPh2)2-substituted
7a (81% ee). Large differences in the ees of up to 80% for ligands, but the other is shifted upfield. This unusual shift
different O groups were observed when the same substrates can be explained by assuming hemilabile coordination[31]
were reduced. In many cases, products with opposite con- of the oxo- and oxy-functionalities to Rh. Because of the
figuration were obtained (compare, for example, the hydro- coordination, a six-membered PϪRhϪO chelate ring is
genation of AH and AMe by complexes based on 7b and formed in addition to the seven-membered PϪRhϪP che-
8b). The differences in selectivity and activity within the late ring. Phosphorus nuclei involved in six-membered rings
8,10-(PPh2)2-substituted ligand series are much larger than are in general characterized by a shift of their 31P NMR
those within the 9,10-(PPh2)2-substituted series. This is a signals to higher field (around 10 ppm).[21,32] The participa-
clear demonstration that the oxo and oxy functionalities tion in both a six-membered and a seven-membered ring
disturb the catalytic action of the metal ion in the case of holds only for the diphenylphosphanyl group at C-10 in
the 8,10-(PPh2)2-substituted ligands. One reason for such a 7bϪ9b.
A
31P-1H
correlation
spectrum
of
disturbance could be coordination of the functional groups [Rh(9b)(COD)]BF4 gave evidence that it is the signal at C-
to the metal centre, shown to influence the hydrogenation 10, which is shifted upfield.
rate and selectivity in related ligands.[8] MMX calcula-
Further support for hemilabile coordination of the O
tions[30] revealed that such coordination is possible in prin- functionalities to Rh in complexes with 8,10-(PPh2)2-substi-
Eur. J. Org. Chem. 2003, 138Ϫ150
143

I. V. Komarov, A. Börner et al.
FULL PAPER
Table 3. Hydrogenation results of various substrates with complexes Rh[Lb(COD)]BF4 where Lb is a ligand of the 8,10-(PPh2)2-substituted
series (7bϪ9b)[a]
Substrate
Solvent
Ligand (Lb)
8b
Conv.
[%]
7b
Conv.
[%]
9b
Time
[min]
ee
[%]
Time
[min]
ee
[%]
Time
[min]
Conv.
[%]
ee
[%]
21
22
23
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
CH2Cl2
MeOH
90
100
100
100
76
43
(S)
49
(S)
73
(S)
64
(S)
28
(R)
1400
1400
20
67
49
(R)
27
(R)
35
(R)
27
(R)
44
(R)
100
100
34
4
(R)
3
150
200
600
1400
55
800
(R)
19
(R)
34
(R)
7
(R)
19
(R)
3
100
100
74
500
100
45
600
800
1600
1800
200
33
60
12
24
25
2500
1400
54
30
4
(S)
1400
13
(S)
Ϫ
CH2Cl2
MeOH
no reaction
1400
Ϫ
30
2
(S)
14
(R)
45
(R)
15
(S)
CH2Cl2
MeOH
1400
1400
8
26
1400
18
2
(S)
18
[a]
Molar catalyst-substrate ratio 1:100, 1 bar H2, 25 °C.
tuted ligands can be derived from their 103Rh NMR spec- This, together with the X-ray data for complexes
tra. By use of a reverse detection technique we determined [Rh(7b)(COD)]BF4 and [Rh(7a)(COD)]BF4, casts some
the 103Rh NMR shifts (relative to Ξ ϭ 3.16 MHz)[33] for doubts on the hemilabile coordination hypothesis. The
complexes [Rh(9a)(COD)]BF4 and [Rh(9b)(COD)]BF4. For other possible reason for the observed functional group ef-
the former, δ ϭ Ϫ157 ppm, while for the latter δ ϭ fect on the rate and selectivity of the catalytic hydrogena-
Ϫ29 ppm was observed. The observation of such a down- tion could be secondary interaction
field shift for the complex [Rh(9b)(COD)]BF4 indicates ligandϪsubstrate interactions in the catalytic intermediates.
changes in the coordination sphere of Rh.[34]
Work is in progress to find evidence for or disprove this pro-
Ϫ noncovalent
At a first glance, the crystal structure data for complexes posal.
[Rh(7b)(COD)]BF4 and [Rh(7a)(COD)]BF4 (Figure 2) ap-
pear to disprove the hemilabile coordination hypothesis. In
both complexes the oxygen atom is well separated from the
Rh (5.899 and 6.025 A for complexes with 7b and 7a, re-
Conclusions
˚
spectively). However, the crystal structures do not say any-
In general, all the catalysts based on 9,10-(PPh2)2-substi-
thing about the behavior of the complexes in solutions, and tuted camphor ligands (7a, 8a, 9a, 17) show the same tend-
(most importantly) about the structure of the true catalytic encies in their catalytic behavior. In contrast, complexes
intermediates. (The latter is, of course, also true for the bearing ligands of the 8,10-(PPh2)2-substituted series (7b,
NMR solution data of the precatalysts.) The hemilabile co- 8b, 9b) differ significantly in their stereodifferentiating
ordination might be more facile in the intermediates, as re- properties. Catalysts based on ligands 7a, 8a and 9a are
ported previously.[8] Thus, the difference in 31P NMR shifts superior to their isomeric counterparts both in their activity
for [Rh(9b)(COD)]BF4 mentioned above disappears on hy- and in their enantioselectivity. Our results provide evidence
drogenation of its COD ligand in CD3OD [presumably with that structurally similar ligands with additional functional
formation of the bis(CD3OD) complex with no hemilabile groups can differ greatly in their catalytic performances. As
coordination] and appears again on addition of the sub- possible reasons for this, coordination of the functional
strate AMe (restoration of the hemilabile coordination in groups to the Rh centre or secondary ligandϪsubstrate in-
the catalystϪsubstrate adduct).
teraction can be suggested. Whatever the reason, the effect
Low-temperature measurements on CD2Cl2 solutions of of the oxo- and oxy-functional groups in diphosphane li-
both [Rh(9a)(COD)]BF4 and [Rh(9b)(COD)]BF4 com- gands can be seriously damaging for catalytic performance
plexes revealed no significant changes in the 31P and of the corresponding hydrogenation catalysts, in terms both
1H NMR spectra (in the aliphatic regions) up to Ϫ90 °C. of activity and of selectivity.
144
Eur. J. Org. Chem. 2003, 138Ϫ150

Chiral Oxo- and Oxy-Functionalized Diphosphane Ligands
FULL PAPER
yield. White crystals, m.p. 115Ϫ116 °C. C10H14I2O (404.03): calcd.
C 29.73, H 3.49; found C 29.77, H 3.37. 1H NMR (400.13 MHz,
CDCl3): δ ϭ 3.28 (d, J ϭ 10.8 Hz, 1 H) and 3.15 (d, J ϭ 10.8 Hz,
1 H, 1-CH2I), 3.17 (d, J ϭ 10.6 Hz, 1 H) and 2.97 (dq, J ϭ 1.0,
10.6 Hz, 1 H, 7-CH2I), 2.42 (d, J ϭ 3.7 Hz, 1 H, 4-CH), 2.40 (dt,
J ϭ 3.7, 17.0 Hz, 1 H) and 2.01 (d, J ϭ 17.0 Hz, 1 H, 3-CH2), 2.14
(tm, 1 H), 1.86Ϫ2.05 (m, 1 H), 1.83 (ddd, J ϭ 3.7, 9.3, 9.3 Hz, 1
H), 1.41 (ddd, J ϭ 3.7, 9.1, 9.1 Hz, 1 H, 5,6-CH2), 1.24 (d, J ϭ
1.0 Hz, 3 H, 7-CH3) ppm. 13C NMR (100.63 MHz, CDCl3): δ ϭ
213.8 (CϭO), 58.7 (C), 52.3 (C), 45.5 (4-CH), 42.6 (CH2), 34.1
(CH2), 26.0 (CH2), 18.6 (CH3), 14.1 (CH2I), Ϫ0.8 (CH2I) ppm.
Experimental Section
General Remarks: All reagents were obtained from Aldrich and
Merck. Solvents were dried and freshly distilled under argon before
use. Reactions involving phosphanes and organometallic com-
pounds were performed under argon by use of standard Schlenk
techniques. Thin layer chromatography was performed on pre-co-
ated TLC plates (silica gel 60 F254, Merck). Flash chromatography
was carried out with silica gel 60 (particle size 0.040Ϫ0.063 mm,
Merck). Melting points were measured on a hotplate and are not
corrected. NMR spectra were recorded at the following frequen-
cies: 400.13 MHz (1H), 100.63 MHz (13C), 161.98 MHz (31P).
103Rh NMR shifts were determined by inverse detected (four-pulse
HMQC) triple-resonance experiments 31P,103Rh{1H}.[35] Each de-
termination was carried out at least twice, with variation of the
pulse frequency and the t1 increment to ensure that the signals in
(1S,4R,7R)-1,7-Bis(iodomethyl)-7-methylspiro[bicyclo[2.2.1]heptane-
2,2Ј-[1,3]dioxolane] (11a): A mixture of 9,10-diiodocamphor (10a,
7.2 g, 17.9 mmol), ethylene glycol (35 mL), benzene (150 mL; cau-
tion, toxic!) and p-toluenesulfonic acid (monohydrate, 150 mg) was
heated at reflux in a DeanϪStark apparatus for ca. 5 d. Every 12 h
a cartridge made of glass wool packed with P2O5 was put in the
apparatus to remove water efficiently. It is very important to protect
the reaction flask from moisture. The reaction was monitored by
TLC (hexane/diethyl ether, 3:1). After the reaction was complete,
the mixture was cooled to room temperature, diluted with diethyl
ether (200 mL) and washed with brine and water (3 ϫ 100 mL),
dried (Na2SO4) and concentrated. The product was purified by col-
umn chromatography (hexane/diethyl ether, 3:1, as an eluent). Pale
the F1 dimension were not folded. Chemical shifts of H and 13C
1
NMR spectra are reported in ppm downfield from TMS as an in-
ternal standard. Chemical shifts of 31P NMR spectra are referenced
to H3PO4 as an external standard. Signals are quoted as s (singlet),
d (doublet), br. (broad) and m (multiplet). Elemental analyses were
performed with a LECO CHNS-932. Hydrogenation experiments
of functionalized olefins were carried out under normal pressure
and isobaric conditions with automatic recording of the gas uptake
(1.0 atm overall pressure over the solution). The experiments were
performed in 15.0 mL of solvent at 25.0 °C. The degrees of conver-
sion of the prochiral dehydroamino acids and the ees of the prod-
ucts were determined by GC. The acids were esterified with tri-
methylsilyl diazomethane before GC measurements: FID; carrier
gas: Ar, 1 mL/min; methyl N-acetylphenylalaninate; fused silica, 10
m, XE-60--valine-tert-butylamide, i.d. 0.2 mm; oven temperature
150 °C. The conversion into methyl 3-N-acetylaminobutanoate and
its ee were determined by GC: Chiraldex β-PM 50 m ϫ 0.25 mm
(Astec), 130 °C. Optical purity analyses of the phosphane oxides
prepared from both enantiomers of 19 were accomplished by
HPLC on a Chiralcel OD-H column (Merck), with hexane/ethanol
(95:5) as eluent (flow 1.0 mL/min).
1
yellow oil, 7.4 g (92% yield). H NMR (400.13 MHz, CDCl3): δ ϭ
3.95Ϫ4.10 (m, 3 H), 3.75Ϫ3.80 (m, 1 H, OCH2CH2), 3.51 (dq, J ϭ
1.2, 9.6 Hz, 1 H), 3.16 (d, J ϭ 9.6 Hz, 1 H, 7-CH2I), 3.36 (d, J ϭ
10.3 Hz, 1 H), 3.17 (d, J ϭ 10.3 Hz, 1 H, 1-CH2I), 2.12 (t, J ϭ
4.4 Hz, 1 H, 4-CH), 2.17 (m, 1 H), 2.01 (ddd, J ϭ 2.2, 4.7, 13.3 Hz,
1 H), 1.70Ϫ1.80 (m, 2 H), 1.48 (d, J ϭ 13.3 Hz, 1 H), 1.34 (m, 1
H, 3-, 5-, 6-CH2), 1.21 (d, J ϭ 1.2 Hz, 3 H, 7-CH3) ppm. 13C NMR
(100.63 MHz, CDCl3): δ ϭ 116.6 (2-C), 64.9, 62.9 (OCH2CH2),
53.7 (C), 52.7 (C), 45.8 (4-CH), 43.8 (CH2), 28.9 (CH2), 25.3 (CH2),
17.7 (CH2I), 17.6 (CH3), 2.5 (CH2I).
(1S,4R,7S)-1,7-Bis(iodomethyl)-7-methylspiro[bicyclo[2.2.1]heptane-
2,2Ј-[1,3]dioxolane] (11b): This compound was prepared analog-
ously to 11a, from 8,10-diiodocamphor (10b, 6.58 g, 16.3 mmol),
ethylene glycol (33 mL), benzene (150 mL), and p-toluenesulfonic
acid monohydrate (150 mg). Pale yellow oil, 6.24 g (85.5% yield).
1H NMR (400.13 MHz, CDCl3): δ ϭ 3.87Ϫ4.10 (m, 4 H),
3.75Ϫ3.85 (m, 1 H, OCH2CH2O, HCHI), 3.37 (d, J ϭ 9.9 Hz, 1
H), 3.10 (d, J ϭ 9.9 Hz, 1 H, CH2I), 3.03 (d, J ϭ 10.3 Hz, 1 H,
HCHI), 2.40 (dd, J ϭ 3.2, 12.5 Hz, 1 H), 2.14 (t, J ϭ 4.6 Hz, 1 H,
4-CH), 1.97 (ddd, J ϭ 2.8, 4.6, 13.5 Hz, 1 H), 1.65Ϫ1.80 (m, 2 H),
1.55 (d, J ϭ 13.5 Hz, 1 H), 1.25Ϫ1.35 (m, 1 H), 1.13 (s, 3 H, 7-
CH3) ppm. 13C NMR (100.63 MHz, CDCl3): δ ϭ 115.6 (2-C), 66.1,
63.9 (OCH2CH2O), 54.4 (C), 53.9 (C), 46.9 (4-CH), 45.5 (CH2),
33.3 (CH2), 25.7 (CH2), 19.5 (CH3), 17.6 (CH2I), 2.7 (CH2I) ppm.
(1S,4R,7R)-1,7-Bis(iodomethyl)-7-methylbicyclo[2.2.1]heptan-2-one
(10a): A solution of 9,10-dibromocamphor (10 g, 32.2 mmol), ob-
tained as described previously,[15] and KI (53.5 g, 322.0 mmol) in
dry DMF (150 mL) was stirred under argon at 110 °C overnight.
A precipitate formed. The mixture was then cooled and diluted
with water (500 mL), and the product was extracted with Et2O
(500 mL). The diethyl ether extract was washed with water (3 ϫ
200 mL) and dried (Na2SO4), and the solvents were evaporated.
The residue was recrystallized from heptane. White crystals, m.p.
133 °C, 10.6 g (81.8%). C10H14I2O (404.03): calcd. C 29.73, H 3.49;
1
found C 29.92, H 3.46. H NMR (400.13 MHz, CDCl3): δ ϭ 3.47
(dq, J ϭ 0.2, 10.1 Hz, 1 H) and 3.33 (d, J ϭ 10.1 Hz, 1 H, 7-CH2I),
3.43 (d, J ϭ 11.3 Hz, 1 H) and 3.17 (d, J ϭ 11.3 Hz, 1 H, 1-CH2I),
2.46 (t, J ϭ 4.0 Hz, 1 H, 4-H), 2.37 (dt, J ϭ 4.0, 18.6 Hz, 1 H) and
1.97 (d, J ϭ 18.6 Hz, 1 H, 3-CH2), 2.12 (t, J ϭ 10.9 Hz, 1 H),
1.85Ϫ1.95 (m, 1 H), 1.43Ϫ1.60 (m, 2 H, 5-CH2 and 6-CH2), 1.04
(s, 3 H, 7-CH3) ppm. 13C NMR (100.63 MHz, CDCl3): δ ϭ 213.6
(CϭO), 59.4 (C), 51.5 (C), 43.8 (4-CH), 41.9 (CH2CϭO), 29.3
(CH2), 25. 9 (CH2), 17.9 (CH3), 15.6 (CH2I), 0.0 (CH2I) ppm. MS
(EI): m/z (%) ϭ 404 (0.12) [Mϩ], 277 (77.8), 235 (12.6), 149 (40.2),
122 (40.8), 121 (100), 108 (21.1), 107 (90.5), 94 (13.3), 93 (66.5), 91
(27.9), 79 (65.2), 53 (24.5), 41 (44.2), 39 (39.6).
Diphenyl({(1S,4R,7S)-1-[(diphenylphosphanyl)methyl]-7-methyl-
spiro[bicyclo[2.2.1]heptane-2,2Ј-[1,3]dioxolan]-7-yl}methyl)-
phosphane (9a): A Ph2PLi solution [prepared from Ph2PCl (4.5 mL,
25.11 mmol) and Li (520 mg, 75.33 mmol) in 40 mL of THF, by
stirring at room temperature for 1 h and then heating at reflux for
2 h] was added at 0 °C (ice bath) whilst stirring (under argon) to a
solution of the diiodo acetal 11a (3.75 g, 8.37 mmol) in THF
(10 mL). The ice bath was removed, and the mixture was heated at
reflux for 12 h. The mixture was cooled to room temperature, water
was added (20 mL), and the product was extracted with diethyl
ether (50 mL). The extract was washed twice with water and con-
centrated, and the product was purified by column chromatography
(eluent hexane/diethyl ether, 3:1). Colorless, amorphous solid, 2.2 g
(54.4% yield). C36H38O2P2 (564.64): calcd. C 76.58, H 6.78; found
(1S,4R,7S)-1,7-Bis(iodomethyl)-7-methylbicyclo[2.2.1]heptan-2-one
(10b): This compound was obtained analogously to 10a, from 8,10-
dibromocamphor, obtained as described previously,[15] in 81.5%
Eur. J. Org. Chem. 2003, 138Ϫ150
145

I. V. Komarov, A. Börner et al.
C 76.50, H 6.82. 1H NMR (400.13 MHz, C6D6): δ ϭ 7.65Ϫ7.85 (400.13 MHz, C6D6): δ ϭ 7.50Ϫ7.61 (m, 4 H), 7.30Ϫ7.40 (m, 2 H),
FULL PAPER
(m, 8 H), 7.15Ϫ7.35 (m, 12 H, arom.), 3.80 (q, J ϭ 6.9 Hz, 1 H),
3.60Ϫ3.75 (two overlapped q, J ϭ 6.9 Hz, 2 H), 3.49 (q, J ϭ 6.9 Hz,
7.20Ϫ7.30 (m, 2 H), 6.90Ϫ7.10 (m, 12 H, arom.), 2.74 (dd, J ϭ
3.2, 15.0 Hz, 1 H), 2.28 (dd, J ϭ 2.5, 15.0 Hz, 1 H, PϪCH2), 2.17
1 H, OCH2CH2O), 2.87 (d, J ϭ 14.4 Hz, 1 H), 2.40Ϫ2.65 (m, 5 (t, J ϭ 4.2 Hz, 1 H, 4-CH), 2.03 (d, J ϭ 3.2 Hz, 2 H, PCH2), 1.88
H), 2.17 (dm, J ϭ 16.8 Hz, 1 H), 1.72 (s, 3 H, 7-CH3), 1.60Ϫ1.85 (m, 1 H), 1.75 (m, 1 H), 1.42 (d, J ϭ 18.2 Hz, 1 H), 1.40 (m, 2 H),
(m, 2 H), 1.52 (d, J ϭ 13.0 Hz, 1 H), 1.30 (m, 1 H) ppm. 13C NMR
0.83 (m, 1 H, 3-, 5-, 6-CH2), 1.05 (s, 3 H, 7-CH3) ppm. 13C NMR
(100.63 MHz, C6D6): δ ϭ 127Ϫ143 (arom. signals), 117.1 (d, (100.63 MHz, C6D6): δ ϭ 215.5 (s, CϭO), 126Ϫ147 (arom.), 62.4
3JC-P ϭ 3.8 Hz, 2-C), 65.2 (s), 63.1 (s, OϪCH2CH2ϪO), 57.2 (dd,
(dd, JC-P ϭ 5.7, 12.4 Hz, C), 51.6 (dd, JC-P ϭ 2.9, 14.3 Hz, C), 42.5
JC-P ϭ 4.8, 9.5 Hz), 54.0 (dd, JC-P ϭ 4.8, 14.3 Hz, 1,7-C), 45.2 (s, (s, CH2), 40.9 (d, JC-P ϭ 8.6 Hz, 4-CH), 35.0 (d, JC-P ϭ 19.1 Hz,
1
3-CH2), 43.5 (d, JC-P ϭ 11.5 Hz, 4-CH), 35.5 (d, JC-P ϭ 14.3 Hz, CH2), 28.2 (d, JC-P ϭ 13.4 Hz, CH2), 27.1 (s, CH2), 25.6 (d, JC-P ϭ
PϪCH2), 28.8 (d, JC-P ϭ 9.5 Hz), 27.4 (s, 5,6-CH2), 26.6 (d,
18.1 Hz, CH2), 19.2 (dd, JC-P ϭ 4.8, 11.5 Hz, 7-CH3) ppm. 31P
3
1JC-P ϭ 20.0 Hz, PϪCH2), 20.1 (d, JC-P ϭ 12.0 Hz, 7-CH3) ppm. NMR (161.98 MHz, C6D6): δ ϭ Ϫ21.4, Ϫ23.2 ppm.
31P NMR (161.98 MHz, C6D6): δ ϭ Ϫ20.7, Ϫ21.9 ppm.
(1S,2R,4R,7R)-1,7-Bis[(diphenylphosphanyl)methyl]-7-methyl-
Diphenyl({(1S,4R,7R)-1-[(diphenylphosphanyl)methyl]-7-methyl-
spiro[bicyclo[2.2.1]heptane-2,2Ј-[1,3]dioxolan]-7-yl}methyl)-
phosphane (9b): This compound was prepared analogously to the
isomer 9a, from 11b in 55.5% yield, as colorless crystals, m.p. 111
°C. Purification was performed by column chromatography (hex-
bicyclo[2.2.1]heptan-2-ol (8b): The oxo derivative 7b (255 mg,
0.49 mmol) was dissolved in THF (5 mL), and LiAlH4 (50 mg,
1.3 mmol) was added to the solution. The mixture was stirred at
room temperature for 5 h, and water was then added to quench
excess LiAlH4 (the reaction flask was cooled in an ice bath). The
ane/diethyl ether, 5:1, as eluent). C36H38O2P2 (564.64): calcd. C product was extracted with diethyl ether (3 ϫ 20 mL), concentrated
76.58, H 6.78; found C 76.52, H 6.80. 1H NMR (400.13 MHz,
C6D6): δ ϭ 7.65Ϫ7.85 (m, 8 H), 7.10Ϫ7.35 (m, 12 H, arom.), 3.76
and purified by column chromatography (hexane/diethyl ether, 2:1
as an eluent). White crystals (120 mg, 47% yield). 1H NMR
(q, J ϭ 7.0 Hz, 1 H), 3.67 (q, J ϭ 7.0 Hz, 1 H), 3.59 (q, J ϭ 7.0 Hz, (400.13 MHz, C6D6): δ ϭ 7.64 (t, J ϭ 8.1 Hz, 2 H), 7.58 (t, J ϭ
1 H), 3.42 (q, J ϭ 7.0 Hz, 1 H, OCH2CH2O), 3.51 (dd, J ϭ 2.7, 8.1 Hz, 2 H), 7.52 (m, 2 H),7.41 (t, J ϭ 8.1 Hz, 2 H), 6.95Ϫ7.10
14.7 Hz, 1 H), 2.92 (d, J ϭ 14.3 Hz, 1 H), 2.40Ϫ2.50 (m, 4 H), 2.01
(m, 12 H, arom.), 3.88 (dd, J ϭ 3.4, 7.6 Hz, 1 H, 2-CH),3.49 (dd
(dt, J ϭ 3.2, 12.9 Hz, 1 H), 1.70 (m, 1 H), 1.56 (m, 1 H), 1.43 (d, J ϭ 4.0, 14.5 Hz, 1 H, CH2P), 2.65 (d J ϭ 13.5, 1 H, CH2P),
J ϭ 13.3 Hz, 1 H), 1.35 (m, 1 H), 1.24 (s, 3 H, 7-CH3) ppm. 13C 2.15Ϫ2.25 (m, 4 H, OH, 4-CH, CH2P), 1.76 (dq, J ϭ 3.4, 13.3 Hz,
NMR (100.63 MHz, C6D6):
127.0Ϫ130.0 (arom. carbon atoms), 117.2 (s, 2-C), 65.2 (s), 63.0 (s,
OCH2CH2O), 57.4 (dd, JC-P ϭ 4.8, 9.5 Hz, C), 53.9 (dd, JC-P
δ ϭ 141.0Ϫ143.0, 132.0Ϫ135.0, 1 H), 1.46 (dd, J ϭ 4.2, 13.3 Hz, 1 H, 3-CH2), 1.38 (m, 1 H), 1.18
(m, 1 H), 0.89 (m, 1 H), 0.72 (m, 1 H, 5-, 6-CH2), 0.95 (s, 3 H, 7-
ϭ
CH3) ppm. 13C NMR (100.63 MHz, C6D6): δ ϭ 127Ϫ142 (arom.),
3
5.7, 13.4 Hz, C), 45.0 (s, CH2), 43.3 (d, JC-P ϭ 7.2 Hz, 4-CH), 34.5 77.5 (d, JC-P ϭ 4.8 Hz, 2-CH), 54.1 (dd, JC-P ϭ 4.8, 9.5 Hz, C),
1
(d, JC-P ϭ 16.2 Hz, PϪCH2), 28.9 (d, JC-P ϭ 6.7 Hz, CH2), 27.61 52.1 (dd, JC-P ϭ 7.6, 13.4 Hz, C), 42.8 (d, JC-P ϭ 10.5 Hz, 4-CH),
1
3
1
(s, CH2), 26.6 (d, JC-P ϭ 21.0 Hz, PϪCH2), 20.2 (d, JC-P
ϭ
39.9 (s, CH2), 33.9 (d, JC-P ϭ 16.2 Hz, PCH2), 32.9 (s, CH2), 27.6
10.5 Hz, 7-CH3) ppm. 31P NMR (161.98 MHz, C6D6): δ ϭ Ϫ19.5, (s, CH2), 27.5 (d, 1JC-P ϭ 18.1 Hz, PCH2), 19.6 (d, J ϭ 11.5 Hz, 7-
Ϫ21.7 ppm.
CH3) ppm. 31P NMR (161.98 MHz, C6D6): δ ϭ Ϫ19.5, Ϫ23.0 ppm.
(1S,4R,7S)-1,7-Bis[(diphenylphosphanyl)methyl]-7-methylbicyclo-
[2.2.1]heptan-2-one (7a): The acetal 9a (300 mg, 0.53 mmol) was
dissolved in a mixture of concd. HCl (1 mL), water (2 mL) and
THF (5 mL), and the solution was brought to reflux under argon
for 5 h. The mixture was cooled to room temperature, and aq.
NaOH (2.5 , 10 mL) was added carefully. The product was ex-
tracted with diethyl ether (20 mL), and the extract was washed with
water (2 ϫ 20 mL) and concentrated. The phosphane was purified
(1R,2R,4R,7R)-1,7-Bis(iodomethyl)-7-methylbicyclo[2.2.1]heptan-2-
ol (12): 9,10-Diiodocamphor (10a, 1.64 g, 4.06 mmol) was added in
portions at 0Ϫ5 °C (ice bath) under an inert gas to a stirred suspen-
sion of LiAlH4 (300 mg) in THF (20 mL). The mixture was then
stirred for 1.5 h at 0 °C, after which water was carefully added to
quench excess LiAlH4. Diethyl ether (50 mL) was added, and the
precipitate was filtered off and washed with diethyl ether and THF.
The filtrate was concentrated, and the residue was applied to a
by column chromatography (hexane/diethyl ether, 5:2, as eluent). column. The desired product was eluted first by hexane/EtOAc
Colorless crystals (from hexane), m.p. 118 °C (150 mg, 49% yield, (3:1) (1 g, 2.46 mmol, 60.1% yield), followed by 7-(iodomethyl)-1,7-
calculated from the diiodo acetal 11a). C34H34OP2 (520.59): calcd. dimethylbicyclo[2.2.1]heptan-2-ol and then by 2-[2-(iodomethyl)-2-
C 78.44, H 6.58; found C 78.40, H 6.59. 1H NMR (400.13 MHz,
methyl-3-methylenecyclopentyl]ethanol. White crystals, m.p. 80 °C.
C6D6): δ ϭ 7.72Ϫ7.85 (m, 4 H), 7.60Ϫ7.70 (m, 4 H), 7.25Ϫ7.35 C10H16I2O (406.05): calcd. C 29.58, H 3.97; found C 30.29 H, 4.03.
(m, 12 H, arom. protons), 3.05 (dd, J ϭ 3.6, 15.4 Hz, 1 H), 2.56
1H NMR (400.13 MHz, CDCl3): δ ϭ 3.95 [pseudo p (ddd), J ϭ
(dd, J ϭ 2.2, 12.1 Hz, 1 H), 2.48 (s, 2 H), 2.45 (dd, J ϭ 2.2, 12.1 Hz, 4.3 Hz, 1 H, CHOH], 3.47 (d, J ϭ 9.3 Hz, 1 H), 3.17 (d, J ϭ 9.3 Hz,
1 H), 2.31 (dt, J ϭ 4.2, 18.2 Hz, 1 H), 2.17 (t, J ϭ 13.3 Hz, 1 H), 1 H, CH2I), 3.36 (dd, J ϭ 9.9, 1.2 Hz, 1 H), 3.00 (d, J ϭ 9.9 Hz,
1.75 (d, J ϭ 18.2 Hz, 1 H), 1.65 (m, 1 H), 1.51 (m, 1 H), 1.02 (s, 3 1 H, CH2I), 2.30 (t, J ϭ 4.0 Hz, 1 H, 4-H), 2.22 (d, J ϭ 4.2 Hz, 1
H), 0.99 (m, 1 H) ppm. 31C NMR (100.63 MHz, C6D6): δ ϭ 215.4 H, OH), 1.60Ϫ1.75 (m, 4 H, 2 ϫ CH2), 1.35Ϫ1.42 (m, 1 H),
3
(d, JC-P ϭ 3.8 Hz, CϭO), 129Ϫ140 (arom. signals), 62.2 (dd, 1.13Ϫ1.21 (m, 1 H, CH2), 1.24 (d, J ϭ 1.2 Hz, 3 H, CH3) ppm.
J
C-P ϭ 5.7, 12.6 Hz, C), 51.7 (d, JC-P ϭ 15.2 Hz, C), 42.9 (s, CH2), 13C NMR (100.63 MHz, CDCl3): δ ϭ 80.5 (CH2OH), 51.1 (C),
41.3 (d, JC-P ϭ 10.4 Hz, 4-CH), 34.2 (dd, JC-P ϭ 5.7, 19.1 Hz, CH2), 51.8 (C), 47.8 (CH), 38.5 (CH2), 33.2 (CH2), 26.3 (CH2), 18.4
28.2 (d, JC-P ϭ 14.3 Hz, CH2), 27.2 (s, CH2), 25.4 (d, JC-P
ϭ
(CH3), 17.6 (CH2), 9.6 (CH2) ppm.
17.2 Hz, CH2), 19.8 (d, JC-P ϭ 10.5 Hz, CH3) ppm. 31P NMR
(161.98 MHz, C6D6): δ ϭ Ϫ23.49 (d, J ϭ 2.8 Hz), Ϫ23.82 (d, J ϭ
(1S,2R,4R,7S)-1,7-Bis(iodomethyl)-7-methyl-2-(triethylsilyloxy)-
bicyclo[2.2.1]heptane
(13):
Chlorotriethylsilane
(0.985 mL,
2.8 Hz) ppm. IR (KBr): ν˜ ϭ 1739 (CϭO) cmϪ1
.
5.87 mmol) was added to a solution of the alcohol 12 (1.986 g,
4.89 mmol) and imidazole (0.433 g, 6.36 mmol) in DMF (10 mL),
cooled in an ice bath. The resulting solution was stirred for 15 min
(1S,4R,7R)-1,7-Bis[(diphenylphosphanyl)methyl]-7-methylbicyclo-
[2.2.1]heptan-2-one (7b): This compound was obtained analogously
1
to the isomer 7a, from 9b in 23% yield. White crystals. H NMR at 0 °C and then overnight at room temperature. Ice-cold water
146
Eur. J. Org. Chem. 2003, 138Ϫ150

Chiral Oxo- and Oxy-Functionalized Diphosphane Ligands
(20 mL) was added, and the product was extracted with CH2Cl2, dissolved under argon in dry CH2Cl2 (10 mL), and this solution
FULL PAPER
dried with Na2SO4, concentrated and purified by flash chromato-
graphy (hexane/EtOAc, 10:1; Rf ϭ 0.8). Colorless oil (2.38 g,
4.57 mmol), 93.5% yield. 1H NMR (400.13 MHz, CDCl3): δ ϭ 3.91
was added dropwise to a stirred suspension of Zn dust (Ͻ 10 mi-
cron, 98% purity, 5.69 g, 87.1 mmol) in a CH2Br2/THF mixture
(2.04 mL and 20 mL, respectively) at Ϫ40 °C. The mixture was then
(dd, J ϭ 7.1, 3.4 Hz, 1 H, CHOH), 3.48 (d, J ϭ 9.1 Hz, 1 H), 3.12 kept in a refrigerator at 0Ϫ5 °C for 3 d. This reagent was added to
(d, J ϭ 9.1 Hz, 1 H, 1-CH2I), 3.36 (dq, J ϭ 9.1, 1.0 Hz, 1 H), 3.02 a solution of 9,10-dibromocamphor (2 g, 6.45 mmol) in THF
(d, J ϭ 9.1 Hz, 1 H, 7-CH2I), 2.2 (t, J ϭ 3.8 Hz, 1 H, 4-H), 1.6Ϫ1.8 (5 mL), and the resulting mixture was stirred for 2 d at room tem-
(m, 4 H, CH2), 1.34 (m, 1 H), 1.14 (m, 1 H, CH2), 1.23 (d, J ϭ perature. It was then neutralized with sat. aq. NaHCO3 and fil-
1 Hz, 3 H, CH3), 0.98 (t, J ϭ 7.9 Hz, 9 H, SiCH2CH3), 0.64 (q, tered. The deposit was washed on the filter with CH2Cl2. The fil-
J ϭ 7.9 Hz, 6 H, SiCH2CH3) ppm. 13C NMR (100.63 MHz, trate was washed with water (3 ϫ 20 mL), dried with Na2SO4 and
CDCl3): δ ϭ 79.4 (CHOH), 52.8 (C), 51.3 (C), 47.9 (CH), 40.1 concentrated. The crude product was purified by chromatography
(CH2), 33.8 (CH2), 26.0 (CH2), 18.0 (CH2), 8.4 (CH2), 7.0 (eluent hexane/Et2O, 3:1). Colorless oil (1.25 g, 60.5% yield). 1H
(SiCH2CH3), 6.4 (CH3), 5.1 (SiCH2CH3) ppm.
NMR (400.13 MHz, CDCl3): δ ϭ 4.89 (t, J ϭ 2.5 Hz, 1 H, ϭCH),
4.75 (t, J ϭ 2.0 Hz, 1 H, ϭCH), 3.67, 3.49 (ABq, 2 H, CH2Br),
3.64, 3.54 (ABq, 2 H, CH2Br), 2.41 (br. d, J ϭ 16.4 Hz, 1 H), 2.23
(t, J ϭ 4.4 Hz, 1 H), 2.12 (dt, J ϭ 4.0, 13.0 Hz, 1 H), 2.01 (dt, J ϭ
2.0, 16.4 Hz, 1 H), 1.87 (ddd, 1 H), 1.54 (ddd, J ϭ 4.0, 9.7, 13.0 Hz,
1 H), 1.04 (s, 3 H, CH3) ppm. 13C NMR (100.63 MHz, CDCl3):
δ ϭ 155.7, 104.3, 55.8, 53.7, 43.7, 40.7, 35.9, 33.1, 31.9, 26.8, 16.6
ppm.
(1S,2R,4R,7S)-1,7-Bis[(diphenylphosphanyl)methyl]-7-methyl-
bicyclo[2.2.1]heptan-2-ol (8a). Method A: The compound was pre-
pared from 7a and LiAlH4 analogously to 8b, in 89% yield. Method
B:
A solution of Ph2PLi [prepared from Ph2PCl (1.97 mL,
10.97 mmol) and Li (207 mg, 30 mmol) in 30 mL of THF, stirring
at room temperature, 1 h and at reflux, 2 h] was added dropwise at
0Ϫ5 °C (ice bath), whilst stirring under argon, to a solution of the
protected hydroxy diiodide 13 (2.38 g, 4.57 mmol) in THF (10 mL).
The reaction mixture was stirred for 30 min in the ice bath and for
(1R,2R/S,4R,7R)-1,7-Bis(bromomethyl)-2,7-dimethylbicyclo-
[2.2.1]heptane (16): Compound 15 was dissolved in methanol
30 min at room temperature, and then heated at reflux for 1.5 h. (15 mL), Pd on charcoal (10%, 100 mg) was added to the solution,
Degassed water was added, and the product 14 was extracted with
diethyl ether (100 mL). The diethyl ether solution was washed twice
with water and concentrated, and the product was dried in high
vacuum at 50 °C for 4 h. The diphosphane 14 thus obtained was
characterized by NMR, and used for the next step without further
and the double bond was hydrogenated in an autoclave at 55 bar
of hydrogen and room temperature. The uptake of hydrogen
stopped after about 16 h. The mixture was then filtered and con-
centrated, and the product was purified by column chromatography
with hexane as the eluent. The NMR spectroscopic data showed
that the product was a mixture of the endo and exo isomers in a
ratio of 1:3. Colorless oil (146 mg, 48.5% yield). 1H NMR
1
purification. H NMR (400.13 MHz, C6D6): δ ϭ 7.08Ϫ7.40 (m, 8
H), 6.65Ϫ6.95 (m, 12 H, arom.), 3.93 (dd, J ϭ 4.9, 2.6 Hz, 1 H, 2-
H), 2.42 (d, J ϭ 13.5 Hz, 1 H), 2.24 (dd, J ϭ 5.9, 13.5 Hz, 1 H, (400.13 MHz, CDCl3): δ ϭ 3.71 (dd, J ϭ 1.1, 10.3 Hz), 3.52 (d,
PCH2), 1.98 (dd, J ϭ 13.2, 3.6 Hz, 1 H), 1.85 (dd, J ϭ 13.2, 2.2 Hz,
1 H, PCH2), 1.91 (m, 1 H, 4-H), 1.60 (m, 2 H, 3-CH2), 1.32 (m, 2
J ϭ 10.3 Hz, CH2Br, minor isomer), 3.43 (s, CH2Br, minor), 3.57
(dd, J ϭ 1.0, 10.1 Hz), 3.26 (d, J ϭ 10.1 Hz, CH2Br, major), 3.52
H, CH2), 1.22 (s, 3 H, CH3), 1.01 (m, 2 H, CH2), 0.78 (t, J ϭ (d, J ϭ 9.9 Hz), 3.46 (d, J ϭ 9.9 Hz, CH2Br, major), 2.2Ϫ2.3 (over-
8.0 Hz, 9 H, SiCH2CH3), 0.45 (q, J ϭ 8.0 Hz, 6 H, SiCH2CH3) lapped m), 1.9Ϫ2.1 (overlapped m), 1.45Ϫ1.85 (overlapped m),
ppm. 31P NMR (161.98 MHz, C6D6): δ ϭ Ϫ19.9, Ϫ23.5 ppm. The
Et3Si-protected diphosphane 14 was dissolved in degassed THF
(20 mL), tetrabutylammonium fluoride hydrate (3.36 g, 11.4 mmol)
was added, and the solution was stirred overnight at room temper-
1.25Ϫ1.35 (overlapped m), 1.19 (d, J ϭ 0.8 Hz, 7-CH3, minor), 1.16
(d, J ϭ 0.8 Hz, 7-CH3, major), 1.13 (d, J ϭ 7.4 Hz, 2-CH3, major),
0.98 (d, J ϭ 6.9 Hz, 2-CH3, minor), 0.78 (dd, J ϭ 10.8, 3.5 Hz)
ppm. 13C NMR (100.63 MHz, CDCl3): δ (major isomer) ϭ 17.3
ature. Degassed water was added (20 mL), and the product was (CH3), 19.3 (CH3), 26.6 (CH2), 36.1 (CH2), 37.2 (CH2), 37.4 (CH2),
extracted with diethyl ether (100 mL). The diethyl ether solution
was washed with water and concentrated, and the residue was puri-
42.2 (CH), 42.3 (CH2), 46.1 (CH), 53.0 (C), 53.9 (C) ppm; δ (minor
isomer) ϭ 15.9 (CH3), 16.1 (CH3), 25.7 (CH2), 27.8 (CH2), 35.1
fied by column chromatography (under argon; pentane/Et2O, 2:1, (CH2), 37.2 (CH2), 38.4 (CH), 42.2 (CH2), 45.2 (CH), 53.2 (C),
as eluent) to give 8a as a colorless, amorphous solid (1.7 g,
3.25 mmol, 71.1% yield calculated from 12). C34H36OP2 (522.61):
calcd. C 78.14, H 6.94; found C 78.18, H 6.90. 1H NMR
(400.13 MHz, C6D6): δ ϭ 7.5Ϫ7.7 (m, 8 H), 7.0Ϫ7.2 (m, 12 H,
arom.), 3.97 (dd, J ϭ 4.5, 2.0 Hz, 1 H, 2-H), 2.63 and 2.42 (ABq,
J ϭ 13.5 Hz, 2 H, PϪCH2), 2.3 (br. s, 1 H, OH), 2.21 (m, 3 H, 3-
H and PCH2), 1.97 (d, J ϭ 13 Hz, 1 H, 3-CH2), 1.60 (d, s, J ϭ
13 Hz, 4 H, CH3 and 3-CH2), 1.33 (m, 2 H, 6-CH2 and 5-CH2),
0.96 (m, 1 H, 6-CH2), 0.71 (m, 1 H, 5-CH2) ppm. 13C NMR
(100.63 MHz, C6D6): δ ϭ 127Ϫ142 (arom.), 77.8 (d, J ϭ 3 Hz,
CHOH), 54.2 (dd, JC-P ϭ 4.8, 9.5 Hz, C), 52.5 (dd, JC-P ϭ 7.6,
14.3 Hz, C), 43.1 (d, JC-P ϭ 9.5 Hz, 4-CH), 40.2 (s, 3-CH3), 35.6
(d, 1JC-P ϭ 18.1 Hz, PCH2), 33.0 (d, JC-P ϭ 5.7 Hz, 6-CH2), 27.8 (d,
1JC-P ϭ 15.3 Hz, PCH2), 27.6 (s, 5-CH2), 19.5 (d, JC-P ϭ 12.4 Hz, 7-
CH3, assignments were made using H,H- and C,H-COSY) ppm.
31P NMR (161.98 MHz, C6D6): δ ϭ Ϫ20.2, Ϫ23.1 ppm. IR (KBr):
55.0 (C) ppm.
Diphenyl({(1S,2R/S,4R,7S)-1-[(diphenylphosphanyl)methyl]-2,7-
dimethylbicyclo[2.2.1]hept-7-yl}methyl)phosphane (17): A Ph2PLi
solution [prepared from Ph2PCl (0.219 mL, 1.22 mmol) and Li
(25 mg, 3.66 mmol) in 5 mL of THF, by stirring at room temper-
ature for 1 h and then heating at reflux for 2 h] was added at 0 °C
(ice bath) whilst stirring (under argon) to a solution of the dibrom-
ide 16 (146 mg, 0.47 mmol) in THF (3 mL). The ice bath was then
removed, and the mixture was heated at reflux for 12 h. The mix-
ture was cooled to room temperature, water was added (20 mL),
and the product was extracted with diethyl ether (20 mL). The ex-
tract was washed twice with water and concentrated, and the di-
phosphane was purified by column chromatography (eluent hex-
ane/diethyl ether, 10:1). The NMR spectroscopic data showed that
the product was a mixture of the endo and exo isomers in the ratio
1
ν˜ ϭ 3550 (OH) cmϪ1
.
of 1:3. Colorless, amorphous solid, 100 mg (41% yield). H NMR
(400.13 MHz, C6D6): δ ϭ 7.77 (t, J ϭ 7.3 Hz), 7.55Ϫ7.70 (m),
7.10Ϫ7.25 (m, arom.), 2.67 (dd, J ϭ 5.7, 14.0 Hz), 2.34 (d, J ϭ
14.0 Hz, PCH2, major isomer), 2.55 (dd, J ϭ 4.8, 13.2 Hz), 2.48
(1R,4R,7R)-1,7-Bis(bromomethyl)-7-methyl-2-methylenebicyclo-
[2.2.1]heptane (15): Titanium tetrachloride (4.04 g, 21.3 mmol) was
Eur. J. Org. Chem. 2003, 138Ϫ150
147

I. V. Komarov, A. Börner et al.
FULL PAPER
(dd, J ϭ 5.1, 13.2 Hz, PCH2, minor isomer), 2.29 (m, PCH2, major [Rh(COD)(8a)]BF4: 1H NMR (400.13 MHz, CD3OD/CDCl3, ca.
and minor isomer, 2-CH major and minor, 4-CH major and mi- 1:1): δ ϭ 8.10Ϫ8.25 (m, 4 H), 7.20-7.70 (m, 14 H, arom.), 4.37 (br.
nor), 1.57 (d, J ϭ 7.9 Hz), 1.40Ϫ1.90 (overlapped m), 1.31 (d, J ϭ s, 1 H), 4.15Ϫ4.30 (m, 3 H, ϭCH), 3.34 (dd, J ϭ 4.2, 7.1 Hz, 1 H,
7.4 Hz, 2-CH3 major), 1.28 (s, 7-CH3 major), 1.23 (s, 7-CH3 mi- 2-CH), 2.92 (t, J ϭ 16.8 Hz, 1 H), 2.72 (d, J ϭ 13.8 Hz, 1 H), 2.53
nor), 1.10 (d, J ϭ 6.9 Hz, 2-CH3 minor), 0.80Ϫ1.10 (overlapped (m, 1 H), 2.00Ϫ2.55 (m, 5 H), 1.55Ϫ1.70 (m, 8 H), 1.43 (m, 1 H),
m) ppm. 13C NMR (100.63 MHz, CDCl3): δ (major isomer) ϭ
125Ϫ145 (arom.), 53.1 (d, JC-P ϭ 13.0 Hz, C), 52.4 (d, JC-P
10.3 Hz, C), 43.8 (CH), 41.0 (CH), 38.7 (CH2), 38.3 (CH2), 35.5
0.98 (m, 1 H), 0.77 (s, 3 H, CH3), 0.41 (m, 1 H) ppm. 13C NMR
(100.63 MHz, CD3OD/CDCl3, ca. 1:1): δ ϭ 127Ϫ134 (arom.),
105.2, 104.4, 97.3, 95.4 (ϭCH), 81.3 (d, J ϭ 10.5 Hz, 2-CH), 55.8,
ϭ
1
1
(d, JC-P ϭ 22.0 Hz, PCH2), 28.6 (d, JC-P ϭ 18.0 Hz, PCH2), 27.7 53.1, 47.8 (4-CH), 41.2, 35.8, 32.3 (d, J ϭ 28.3 Hz), 31.6, 30.5, 30.4
(CH2), 20.8 (CH3), 19.7 (d, JC-P ϭ 4.3 Hz, CH3) ppm. 31P NMR (d, J ϭ 18 Hz), 29.8, 29.1, 27.4 (CH2), 21.3 (CH3) ppm. 31P NMR
(161.98 MHz, C6D6): δ (major isomer) ϭ Ϫ19.6, Ϫ22.6 ppm. δ
(minor isomer) Ϫ19.5, Ϫ23.2 ppm.
(161.98 MHz, CD3OD/CDCl3, ca. 1:1): δ ϭ 15.2 (dd, J ϭ 42.5 and
142.3), 14.3 (dd, J ϭ 42.5 and 142.3, chemical shifts and coupling
constants were obtained by g NMR simulation of the observed AB
part of the ABX spin system) ppm. IR (KBr): ν˜ ϭ 3530 [ν(OH)]
(1R,3S,6S,7S)-3-Diphenylphosphanyl-7-[(diphenylphosphanyl)-
methyl]-6-methyl-4-oxatricyclo[4.3.0.03,7]nonane (19): A solution of
Ph2PLi [prepared from Ph2PCl (2.89 mL, 16.1 mmol) and Li
(550 mg) in THF (20 mL), room temperature for 1 h and heating
at reflux for 2 h) was added dropwise at Ϫ78 °C under argon to a
solution of 8,10-dibromocamphor (2 g, 6.45 mmol) in THF (5 mL).
The reaction mixture was stirred at Ϫ78 °C for 0.5 h and 1 h at
room temperature, and then heated at reflux for 5 h. Water (50 mL)
was added, and the product was extracted with diethyl ether
(100 mL), washed with water and concentrated. Column chromato-
graphy (hexane/diethyl ether, 8:1, as eluent) gave the diphosphane
as a white solid, which could be recrystallized from hexane. White
crystals, m.p. 154 °C (1.41 g, 2.71 mmol, 42% yield). C34H34OP2
cmϪ1
.
[Rh(COD)(9a)]BF4: 1H NMR (400.13 MHz, CD3OD): δ ϭ 7.95 (m,
2 H), 7.80 (m, 2 H), 7.0Ϫ7.5 (m, 16 H, arom.), 4.24 (br. s, 1 H),
4.03 (br. s, 1 H), 3.94 (br. s, 1 H), 3.87 (br. s, 1 H, COD CH), 3.54
(m, 2 H), 3.25Ϫ3.40 (m, 3 H), 3.14 (m, 1 H, OCH2CH2), 2.59 (d,
J ϭ 9.8 Hz, 1 H), 2.38 (t, J ϭ 14.8 Hz, 1 H), 2.27 (s, 1 H),
1.50Ϫ2.20 (m, 12 H), 1.08 (m, 1 H), 0.9 (m, 1 H), 0.6 (s, 3 H, CH3)
ppm. 13C NMR (100.63 MHz, CD3OD): δ ϭ 127Ϫ135 (arom.),
116.5 (d, JC-P ϭ 7.6 Hz, OCO), 103.3, 101.9, 99.2, 95.1 (ϭCH),
65.1, 63.3 (OCH2CH2O), 53.9 (C), 52.9 (C), 46.1 (d, JC-P
ϭ
10.5 Hz, 4-CH), 44.3 (3-CH2), 37.7 (m, CH2), 26Ϫ31 (overlapped
CH2), 19.6 (CH3) ppm. 31P NMR (161.98 MHz, CD3OD): δ ϭ 14.3
(dd, J ϭ 42.1 and 142.8 Hz), 13.5 (dd, J ϭ 42.1 and 142.8 Hz) ppm.
1
(520.59): calcd. C 78.44, H 6.58; found C 78.43, H 6.60. H NMR
(400.13 MHz, C6D6): δ ϭ 8.50 (t, J ϭ 8 Hz, 2 H), 8.25 (t, J ϭ 8 Hz,
2 H), 7.54 (t, J ϭ 8 Hz, 2 H), 7.45 (t, J ϭ 8 Hz, 2 H), 7.05Ϫ7.29 (m,
12 H), 3.75 (d, J ϭ 7.8 Hz, 1 H), 3.62 (d, J ϭ 7.8 Hz, 1 H), 2.70
(dd, J ϭ 2.2 and 15 Hz, 1 H), 2.33 (d, J ϭ 15 Hz, 1 H), 2.07 (m, 1
H), 1.75Ϫ1.92 (m, 2 H), 1.55Ϫ1.68 (m, 2 H), 1.35 (m, 2 H), 1.04
(s, 3 H) ppm. 13C NMR (100.63 MHz, C6D6): δ ϭ 127Ϫ141
(arom), 92.9 (dd, JC-P ϭ 3.8, 26.7 Hz, PϪC), 72.1 (OCH2), 58.7 (d,
[Rh(COD)(7b)]BF4: 1H NMR (400.13 MHz, CD3OD): δ ϭ 7.73 (m,
2 H), 7.51 (m, 2 H), 6.8Ϫ7.4 (m, 12 H), 6.59 (m, 2 H), 6.21 (m, 2
H, arom.), 5.05 (br. s, 1 H), 4.32 (br. s, 1 H), 4.12 (br. s, 1 H), 3.87
(br. s, 1 H, COD CH), 2.74 (t, J ϭ 17 Hz, 1 H), 1.2Ϫ2.5 (m, 15
H), 1.48 (s, 3 H, CH3), 0.65 (m, 1 H), 0.33 (m, 1 H) ppm. 13C
3
NMR (100.63 MHz, CD3OD): δ ϭ 217.1 (d, JC-P ϭ 12.4 Hz, Cϭ
1
JC-P ϭ 13.4 Hz, C), 55.2 (C), 44.4 (CH), 41.4 (d, JC-P ϭ 13.4 Hz,
O), 127Ϫ138 (arom.), 105.7, 100.3, 99.4, 90.8 (ϭCH), 58.3 (C), 55.3
PϪCH2), 30.6 (CH2), 28.9 (CH2), 24.9 (d, JC-P ϭ 9.5 Hz, CH2),
13.4 (d, J ϭ 9.5 Hz, CH3) ppm. 31P NMR (161.98 MHz, C6D6):
δ ϭ Ϫ18.2 (d, J ϭ 18 Hz), Ϫ25.6 (d, J ϭ 18 Hz) ppm.
1
(C), 44.1 (d, JC-P ϭ 8.5 Hz, 4-CH), 41.6 (3-CH2), 36.9 (d, JC-P
ϭ
21 Hz, PϪCH2), 34.3, 33.8, 26.0Ϫ28.0 (CH2), 24.8 (CH3) ppm. 31
P
NMR (161.98 MHz, CD3OD): δ ϭ 13.2 (dm, J ϭ 154 Hz), 3.7
(dm, J ϭ 154 Hz) ppm.
General Procedure for the Preparation of the Precatalysts with Li-
gands 7a, 7b, 8a, 8b, 9a, 9b and 17: [Rh(COD)(acac)] (311 mg,
1 mmol) was added to a stirred solution of the diphosphane
(1 mmol) in THF (2 mL). The solution was stirred for 15 min, a
stoichiometric amount of aq. 40% HBF4 was then added, and stir-
ring was continued for another 15 min. The complex was precipit-
ated with diethyl ether (20 mL), dissolved in CH2Cl2 (0.5 mL),
again precipitated by diethyl ether, and dried under vacuum for 5 h
at 50 °C. Yellow powders, containing nonstoichiometric amounts
of THF and diethyl ether, making the elemental analyses of the
complexes incorrect. However, the complexes were fully character-
ized spectroscopically.
[Rh(COD)(8b)]BF4: 1H NMR (400.13 MHz, CD3OD): δ ϭ 8.42 (t,
J ϭ 8.7 Hz, 2 H), 8.06 (t, J ϭ 8.7 Hz, 2 H), 7.81 (t, J ϭ 8.7 Hz, 1
H), 7.55Ϫ7.70 (m, 5 H), 7.47 (t, J ϭ 8.7 Hz, 1 H), 7.25Ϫ7.45 (m,
5 H), 7.10 (br. s, 2 H), 6.90 (t, J ϭ 8.7 Hz, 2 H, arom.), 4.3Ϫ5.5
(br. signals, 4 H, ϭCH), 4.16 (t, J ϭ 14.5 Hz, 1 H, 2-H), 2.85 (d,
J ϭ 14.5 Hz, 1 H), 2.41 (dd, J ϭ 10.3, 15.2 Hz, 1 H), 2.34 (br. s, 5
H), 2.22 (m, 1 H), 2.04 (br. s, 5 H), 1.85 (m, 1 H), 1.71 (m, 1 H),
1.54 (m, 1 H), 0.90 (m, 1 H), 0.54 (td, J ϭ 4.6, 11.9 Hz, 1 H), 0.36
(tt, J ϭ 3.7, 9.5 Hz, 1 H, CH2), 1.92 (s, 3 H, CH3) ppm. 13C NMR
3
(100.63 MHz, CD3OD): δ ϭ 127Ϫ134 (arom.), 80.6 (d, JC-P
ϭ
10.8 Hz, 2-CH), 55.3, 52.3, 48.2 (4-CH), 41.6, 36.2, 33.1 (d, JC-P ϭ
28.3 Hz), 30.9, 30.4, 30.1 (d, J ϭ 18 Hz), 30.0, 29.6, 27.1 (CH2),
20.8 (CH3) ppm. 31P NMR (161.98 MHz, CD3OD): δ ϭ 14.9 (dd,
J ϭ 31.9 and 141.5 Hz), 5.3 (dd, J ϭ 31.9 and 141.5 Hz) ppm.
[Rh(COD)(7a)]BF4: 1H NMR (400.13 MHz, CD3OD):
δ ϭ
8.20Ϫ8.30 m, 2 H), 7.80Ϫ7.90 (m, 4 H), 7.50Ϫ7.80 (m, 14 H,
arom.), 5.14 (br. s, 1 H), 4.61 (br. s, 1 H), 4.50 (br. s, 1 H), 4.21
(br. s, 1 H, COD CH), 3.62 (m, 1 H), 3.05 (tm, J ϭ 12.9 Hz, 1 H), [Rh(COD)(9b)]BF4: 1H NMR (400.13 MHz, CD3OD): δ ϭ 8.36
2.88 (dd, J ϭ 11.9, 15.6 Hz, 1 H), 2.77 (m, 1 H, PϪCH2), (br. m, 2 H), 7.98 (t, J ϭ 7.3 Hz, 2 H), 7.20Ϫ7.70 (m, 12 H), 7.12
2.15Ϫ2.55 (m, 11 H), 1.85Ϫ2.00 (m, 2 H), 1.59 (s, 3 H, CH3), 1.45
(m, 1 H),1.35 (m, 1 H) ppm. 13C NMR (100.63 MHz, CD3OD):
δ ϭ 212.0 (CϭO), 125Ϫ135 (arom), 116.6, 104.8, 98.4, 98.3 (ϭ
(br. m, 2 H), 6.84 (t, J ϭ 7.3 Hz, 2 H, arom.), 5.29 (br. s, 1 H),
4.54 (br. s, 1 H), 4.24 (br. s, 1 H), 3.70 (br. s, 1 H, ϭCHϪ), 3.80
(t, J ϭ 8.2 Hz, 1 H), 3.68 (m, 1 H), 3.59 (q, J ϭ 6.7 Hz, 1 H), 3.48
CH), 47.9 (CH), 41.6, 41.2, 34.2, 32.1, 31.9, 31.3, 30.8, 30.5, 29.9, (q, J ϭ 6.7 Hz, 1 H, OCH2CH2O), 2.83 (m, 1 H), 2.56 (dd, J ϭ
28.6, 26.3, 18.8 (CH3) ppm. 31P NMR (161.98 MHz, CD3OD): δ ϭ
2.5, 12.0 Hz, 1 H), 2.35Ϫ2.50 (m, 3 H), 2.10Ϫ2.25 (m, 3 H), 1.89
12.8 (dd, J ϭ 36.1 and 142.9 Hz), 9.3 (dd, J ϭ 36.1 and 142.9 Hz) (m, 1 H), 1.87 (s, 3 H, CH3), 1.52 (m, 1 H), 1.34 (d, J ϭ 12.0 Hz,
ppm. IR (KBr): ν˜ ϭ 1742 [ν(CϭO)] cmϪ1
.
1 H), 1.23 (m, 1 H), 1.02 (m, 1 H), 0.45 (td, J ϭ 2.5, 11.7 Hz, 1 H)
148
Eur. J. Org. Chem. 2003, 138Ϫ150

Chiral Oxo- and Oxy-Functionalized Diphosphane Ligands
FULL PAPER
ppm. 13C NMR (100.63 MHz, CD3OD): δ ϭ 128Ϫ137 (arom.),
117.1 (d, JC-P ϭ 12.4 Hz, 2-C), 103.9, 100.4, 98.9, 88.8 (ϭCHϪ),
64.9, 63.8 (OCH2CH2), 55.5, 50.3 (C), 47.9 (d, JC-P ϭ 9.5 Hz, 4-
CH), 43.6, 39.4 (d, JC-P ϭ 21. Hz), 33.9, 33.2, 25.7Ϫ27.9 (CH2),
25.9 (d, JC-P ϭ 16 Hz, CH3) ppm. 31P NMR (161.98 MHz,
CD3OD): δ ϭ 15.6 (dd, J ϭ 33. 3 and 142.9 Hz), 5.2 (dd, J ϭ 33.
3 and 142.9 Hz) ppm.
[1]
Asymmetric Synthesis (Ed.: J. D. Morrison), Academic Press,
Orlando, 1985, vol. 5. R. Noyori, Asymmetric Catalysis in Or-
ganic Synthesis, Wiley, New York, 1994. Comprehensive Asym-
metric Catalysis (Eds.: E. N. Jacobsen, A. Pfaltz, H. Yamam-
oto), Springer, Heidelberg, 1999, vol. IϪIII. Catalytic Asym-
metric Synthesis (Ed.: I. Ojima), Wiley-VCH, New York, 2000.
See, for example: H. Brunner, W. Zettlmeier, Handbook of En-
antioselective Catalysis with Transition Metal Compounds, Li-
gands, VCH Verlagsgesellschaft, Weinheim, 1993, vol. II.
W. P. Janks, Catalysis in Chemistry and Enzymology, McGraw-
Hill, New York, 1969.
M. Yatagai, M. Zama, T. Yamagishi, M. Hida, Chem. Lett.
1983, 1203. I. Yamada, M. Yamaguchi, T. Yamagishi, Tetrahed-
ron: Asymmetry 1996, 7, 3339.
S. Borns, R. Kadyrov, D. Heller, W. Baumann, A. Spannen-
berg, R. Kempe, J. Holz, A. Börner, Eur. J. Inorg. Chem. 1998,
1291Ϫ1295. S. Borns, R. Kadyrov, D. Heller, W. Baumann, J.
Holz, A. Börner, Tetrahedron: Asymmetry 1999, 10,
1425Ϫ1431.
J. Holz, M. Quirmbach, U. Schmidt, D. Heller, R. Stürmer, A.
Börner, J. Org. Chem. 1998, 63, 8031Ϫ8034. W. Li, Z. Zhang,
D. Xiao, X. Zhang, Tetrahedron Lett. 1999, 40, 6701Ϫ6704. Y.-
Y. Yan, T. V. RajanBabu, J. Org. Chem. 2000, 65, 900Ϫ906.
W. Li, Z. Zhang, D. Xiao, X. Zhang, J. Org. Chem. 2000, 65,
3489Ϫ3496. Y.-Y. Yan, T. V. RajanBabu, Org. Lett. 2000, 2,
199Ϫ202.
J. Holz, D. Heller, R. Stürmer, A. Börner, Tetrahedron Lett.
1999, 40, 7059Ϫ7062. T. V. RajanBabu, Y.-Y. Yan, S. Shin, J.
Am. Chem. Soc. 2001, 123, 10207Ϫ10213. J. Holz, R. Stürmer,
U. Schmidt, H.-J. Drexler, D. Heller, H.-P. Krimmer, A. Börner,
Eur. J. Org. Chem. 2001, 4615Ϫ4624.
For reviews on the effect of oxy groups in Rh-catalyzed asym-
metric hydrogenation see: A. Börner, Eur. J. Inorg. Chem. 2001,
327Ϫ337. A. Börner, Chirality 2001, 13, 625Ϫ628. For a more
general review of ambifunctional catalysts, see: G. J. Rowland,
Tetrahedron 2001, 57, 1865Ϫ1882.
3
[2]
[Rh(COD)(17)]BF4: 1H NMR (400.13 MHz, CD3OD):
δ ϭ
[3]
[4]
8.30Ϫ8.50 (m, 4 H), 7.80Ϫ7.90 (m, 4 H), 7.40-7-70 (m, 12 H,
arom.), 4.43 (br. s, 1 H), 4.37 (br. s, 1 H), 4.31 (br. s, 2 H), (COD
CH), 3.00 (m, 1 H), 2.85 (m, 1 H), 2.75 (d, J ϭ 15.0 Hz, 1 H),
1.00Ϫ2.60 (overlapped m, 17 H), 0.96 (d, J ϭ 7.1 Hz, 3 H, 2-CH3),
0.75 (s, 3 H, 7-CH3) ppm. 13C NMR (100.63 MHz, CD3OD): δ ϭ
125Ϫ135 (arom), 113.2, 103.1, 98.6, 98.5 (ϭCH), 47.9 (CH), 43.2
(CH), 41.6, 41.8, 33.4, 32.8, 32.9, 31.6, 30.8, 30.0, 28.9, 28.6, 25.6,
20.3 (CH3), 18.8 (CH3) ppm. 31P NMR (161.98 MHz, CD3OD):
δ ϭ 15.6 (dd, J ϭ 43.0 and 142.0 Hz), 14.3 (dd, J ϭ 43.0 and
142.0 Hz) ppm.
[5]
[6]
Crystal Structure Determinations: Crystals of [Rh(7b)(COD)]BF4
and [Rh(7a)(COD)]BF4 for the X-ray analyses were obtained by
slow crystallization from MeOH solutions of the rhodium com-
plexes. Diffraction data were collected with a STOE-IPDS dif-
fractometer with graphite-monochromated Mo-Kα radiation. The
structures were solved by direct methods (SHELXS-97)[36] and re-
fined by full-matrix, least-squares techniques against F2 (SHELXL-
97).[37] XP (Siemens Analytical X-ray Instruments, Inc.) was used
for structure representations. The non-hydrogen atoms {except the
atoms of disordered groups and solvent molecules of
[Rh(COD)(7b)BF4]} were refined anisotropically. The hydrogen
atoms were placed in theoretical positions and were refined by
using the riding model. Crystal and refinement data of [Rh(COD)-
(7a)]BF4: crystal size ϭ 0.3 ϫ 0.2 ϫ 0.2 mm, space group P212121,
[7]
[8]
[9]
T. Money, Nat. Prod. Rep. 1985, 2, 253Ϫ289.
[10]
˚
J. Seyden-Penne, Chiral Auxiliaries and Ligands in Asymmetric
Synthesis, John Wiley & Sons, New York, 1995.
J. H. Hutchinson, T. Money, S. E. Piper, Can. J. Chem. 1986,
64, 854Ϫ860.
V. Vaillancourt, M. R. Agharahimi, U. N. Sundram, O.
Richou, D. J. Faulkner, K. F. Albizati, J. Org. Chem. 1991,
56, 378Ϫ387.
T. Sell, S. Laschat, I. Dix, P. G. Jones, Eur. J. Org. Chem.
2000, 4119Ϫ4124.
I. V. Komarov, M. V. Gorichko, M. Y. Kornilov, Tetrahedron:
Asymmetry 1997, 8, 435Ϫ445.
W. M. Dadson, M. Lam, T. Money, S. E. Piper, Can. J. Chem.
1983, 61, 343Ϫ346.
Reviews for the synthesis of hydroxy phosphanes see: J. Holz,
M. Quirmbach, A. Börner, Synthesis 1997, 983Ϫ1006. J. Holz,
M. Quirmbach, L. Götze, A. Börner, Rec. Develop. Org. Chem.
1998, 2, 685Ϫ711.
K. B. Becker, C. A. Grob, in The Chemistry of the Double-
bonded Functional Groups. Part II, Suppl. (Ed.: A. S. Patai),
Interscience, New York, 1977.
J. A. Clase, D. L. F. Li, L. Lo, T. Money, Can. J. Chem. 1990,
68, 1829Ϫ1836.
V. Rautenstrauch, M. Lindström, B. Bourdin, J. Currie, E. Ol-
iveros, Helv. Chim. Acta 1993, 76, 607Ϫ615.
E. P. Kyba, S. P. Rines, J. Org. Chem. 1982, 47, 4800Ϫ4801.
P. E. Garrou, Chem. Rev. 1981, 81, 229Ϫ266.
orthorhombic, a ϭ 9.867(2), b ϭ 18.742(4), c ϭ 19.908(4) A, V ϭ
3
3681.5(13) A , Z ϭ 4, ρcalcd. ϭ 1.477 g/cm3, µ (Mo-Kα) ϭ 0.605
˚
[11]
[12]
mmϪ1, T ϭ 200 K, F(000) ϭ 1688, Θ range for data collection ϭ
1.49Ϫ21.13°, index range (h,k,l)
ϭ Ϫ9/9, Ϫ18/19, Ϫ20/19,
reflections collected ϭ 13768, observed reflections ϭ 2988, refined
parameters ϭ 460, R1 [2σ(I)] ϭ 0.0369, R1 (all data) ϭ 0.0532.
Crystal and refinement data of [Rh(COD)(7b)]BF4: crystal size ϭ
[13]
[14]
[15]
[16]
0.4 ϫ 0.3 ϫ 0.2 mm, space group P21, monoclinic, a ϭ 9.583(2),
3
˚
˚
b ϭ 19.029(4), c ϭ 11.457(2) A, β ϭ 106.00(3), V ϭ 2008.3(7) A ,
Z ϭ 2, ρcalcd. ϭ 1.380 g/cm3, µ (Mo-Kα) ϭ 0.557 mmϪ1, T ϭ 200 K,
F(000) ϭ 862, Θ range for data collection ϭ 2.14Ϫ24.34 °, index
range (h,k,l) ϭ Ϫ10/10, Ϫ22/21, Ϫ13/13, reflections collected ϭ
10914, observed reflections ϭ 5122, refined parameters ϭ 469, R1
[2σ(I)] ϭ 0.0433, R1 (all data) ϭ 0.0530. CCDC-186344 ([Rh-
(COD)(7a)]BF4) and -186343 ([Rh(COD)(7b)]BF4) contain the
supplementary crystallographic data for this paper. These data can
be obtained free of charge at www.ccdc.cam.ac.uk/conts/
retrieving.html or from the Cambridge Crystallographic Data
Centre, 12, Union Road, Cambridge CB2 1EZ, UK [Fax: (in-
ternat.) ϩ 44-1223/336-033; E-mail: deposit@ccdc.cam.ac.uk].
[17]
[18]
[19]
[20]
[21]
[22]
[23]
Acknowledgments
I. Toth, B. E. Hanson, Organometallics 1993, 12, 1506Ϫ1513.
M. P. Anderson, L. H. Pignolet, Inorg. Chem. 1981, 20,
4101Ϫ4107.
G. Balavoine, S. Brunie, H. B. Kagan, J. Organomet. Chem.
1980, 187, 125Ϫ139.
The authors are grateful for a grant given by the Alexander von
Humboldt Foundation to I. K. We thank Dr. A. Kless for helpful
discussions. We are indebted to the Degussa AG (Frankfurt) and
the Fonds der Chemischen Industrie for financial support. It is
a pleasure for us to acknowledge skilled technical assistance by
S. Buchholz.
[24]
[25]
R. Kadyrov, A. Börner, R. Selke, Eur. J. Inorg. Chem. 1999,
705Ϫ711.
Eur. J. Org. Chem. 2003, 138Ϫ150
149

I. V. Komarov, A. Börner et al.
FULL PAPER
[26]
B. D. Vineyard, W. S. Knowles, M. J. Sabacky, G. L. Bachman,
108, 27Ϫ110. J. J. Schneider, Nachr. Chem. 2000, 48, 614Ϫ620.
P. Braunstein, F. Naud, Angew. Chem. 2001, 113, 702Ϫ722;
Angew. Chem. Int. Ed. 2001, 40, 680Ϫ699.
D. A. Slack, I. Greveling, M. C. Baird, Inorg. Chem. 1979,
18, 3125Ϫ3132.
R. J. Goodfellow in Multinuclear NMR (Ed.: J. Mason),
Plenum Press, New York, 1987, p. 521.
D. J. Weinkauff, J. Am. Chem. Soc. 1977, 99, 5946Ϫ5952. H.
Brunner, A. Winter, J. Breu, J. Organomet. Chem. 1998, 553,
285Ϫ306. V. A. Pavlov, E. I. Klabunovskii, Yu. T. Struchkov,
A. A. Voloboev, A. I. Yanovsky, J. Mol. Catal. 1988, 44,
217Ϫ243. D.-R. Hou, J. Reibenspies, T. J. Colacot, K. Burgess,
Chem. Eur. J. 2001, 5391Ϫ5400.
[32]
[33]
[34]
[35]
[27]
G. Zhu, Z. Chen, X. Zhang, J. Org. Chem. 1999, 64,
M. Bühl, W. Baumann, R. Kadyrov, A. Börner, Helv. Chim.
Acta 1999, 82, 811Ϫ820.
6907Ϫ6908. D. Heller, J. Holz, H.-J. Drexler, J. Lang, H.-P.
Krimmer, K. Drauz, A. Börner, J. Org. Chem. 2001, 66,
6816Ϫ6817. D. Heller, H.-J. Drexler, Y. You, W. Baumann, K.
Drauz, H.-P. Krimmer, A. Börner, Chem. Eur. J., in press.
R. Benn, C. Brevard, J. Am. Chem. Soc. 1986, 108, 5622Ϫ5624.
´
R. Benn, A. Rufinska, Magn. Reson. Chem. 1988, 26, 895Ϫ902.
D. Nanz, PhD Thesis, Universität Zürich, 1993.
[28]
[36]
[37]
T. V. RajanBabu, B. Radetich, K. K. You, T. A. Ayers, A. L.
G. M. Sheldrick, SHELXS-97, University of Göttingen, Ger-
Casalnuovo, J. C. Calabrese, J. Org. Chem. 1999, 64,
many, 1997.
3429Ϫ3447.
G. M. Sheldrick, SHELXL-97, University of Göttingen, Ger-
[29]
S. G. Lee, Y. J. Zhang, Org. Lett. 2002, 4, 2429Ϫ2431.
A. Kless, unpublished results.
J. C. Jeffrey, T. B. Rauchfuss, Inorg. Chem. 1979, 18,
2658Ϫ2666. A. Bader, E. Lindner, Coord. Chem. Rev. 1991,
many, 1997.
[30]
Received May 31, 2002
[O02293]
[31]
150
Eur. J. Org. Chem. 2003, 138Ϫ150