New developments in the synthesis of heterotopic atropisomeric diphosphines via diastereoselective aryl coupling reactions

Article abstract of DOI:10.1016/j.tetasy.2004.05.014

The new heterotopic atropisomeric diphosphine (R)-5,6-benzo-2,2′- bis(diphenylphosphino)-4′,5′,6′-trimethylbiphenyl has been prepared. The key step of this synthesis is a diastereoselective, intramolecular aryl-aryl coupling reaction via oxidation of a suitable, chiral diarylcuprate. The catalytic properties of the diphosphine in ruthenium promoted hydrogenations of model substrates and in rhodium promoted 1,4-additions of boronic acids to α,β-unsaturated ketones are fully comparable to those of reference ligands such as BINAP. This seems to indicate that C₂-symmetry is not a structural prerequisite for atropisomeric chiral diphosphines to obtain high enantioselectivities in 1,4-addition reactions as well as in hydrogenation reactions.

Full text of DOI:10.1016/j.tetasy.2004.05.014

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 15 (2004) 2253–2261
New developments in the synthesis of heterotopic atropisomeric
diphosphines via diastereoselective aryl coupling reactions
*
*
^
Jonathan Madec, Guillaume Michaud, Jean-Pierre Genet and Angela Marinetti
Laboratoire de Synthꢀese Sꢁelective Organique et Produits Naturels, UMR 7573, ENSCP
11, rue Pierre et Marie Curie, 75231 Paris Cedex 05, France
Received 13 April 2004; accepted 4 May2004
Abstract—The new heterotopic atropisomeric diphosphine (R)-5,6-benzo-2,20-bis(diphenylphosphino)-40,50,60-trimethylbiphenyl has
been prepared. The key step of this synthesis is a diastereoselective, intramolecular aryl–aryl coupling reaction via oxidation of a
suitable, chiral diarylcuprate. The catalytic properties of the diphosphine in ruthenium promoted hydrogenations of model sub-
strates and in rhodium promoted 1,4-additions of boronic acids to a,b-unsaturated ketones are fullycomparable to those of ref-
erence ligands such as BINAP. This seems to indicate that C2-symmetry is not a structural prerequisite for atropisomeric chiral
diphosphines to obtain high enantioselectivities in 1,4-addition reactions as well as in hydrogenation reactions.
Ó 2004 Elsevier Ltd. All rights reserved.
1. Introduction
moiety.7 Despite the lack of C2-symmetry, this ligand
induces high enantioselectivityin the ruthenium-pro-
moted hydrogenations of functionalised carbonyl
derivatives.
The utilityof enantiomericallypure diphosphines with
atropoisomeric structures has been undeniablyestab-
lished bytheir outstanding properties in a number of
significant catalytic processes.1 Nevertheless, the search
for new derivatives in these series still represents an
extremelyactive research field as fine tuning of the
structure of the phosphorus ligands significantlyaffects
their catalytic properties. Thus, for instance, the
enantioselectivities are highlydependent on the steric,
conformational and electronic properties of the chiral
atropisomeric ligands, with an optimal selectivityat-
The keystep for the preparation of MeO–NAPhePHOS
was an intramolecular aryl–aryl coupling reaction,
according to Lipshutz’s method.8;9 The 2,4-pentanediol
tether, first introduced bySugimura and co-workers, 10–13
was used as the chiral auxiliaryto induce high diaste-
reoselectivityin the coupling reaction. The resulting
macrocyclic bis-ether derivative was then converted
successivelyinto the corresponding biphenol and
diphosphine through known methods (Scheme 1).
2–6
tained byusing highlyspecific substrate/ligand pairs.
In this context, we devised recentlythe synthesis of a
new chiral atropisomeric diphosphine, called MeO–
NAPhePHOS, which bears an unsymmetric biaryl
The same approach, which allows the synthesis of highly
modular ligands, has been extended here to the synthesis
of a new unsymmetric diphosphine where naphthyl and
I
OMe
O
O
I
O
O
OH
OH
PPh2
PPh2
MeO
MeO
MeO
(R,R)
(S,R,R)
(R)
(S)-MeO-NAPhePHOS
Scheme 1. Synthetic approach to the unsymmetric diphosphine MeO–NAPhePHOS.
* Corresponding author. E-mail addresses: genet@ext.jussieu.fr; marinet@ext.jussieu.fr
0957-4166/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2004.05.014

2254
J. Madec et al. / Tetrahedron: Asymmetry 15 (2004) 2253–2261
3,4,5-trimethylphenyl units constitute the biaryl frame-
work.
The methyl substituents of 5 should modulate mainly
the steric properties and the dihedral angle of the biaryl
framework, while it should onlyslightlyaffect the elec-
tronic properties, with respect to the previouslypre-
pared MeO–NAPhePHOS ligand.
2. Results and discussion
2.1. Synthesis of the heterotopic atropisomeric diphos-
phine TriMe-NAPhePHOS
As shown in Schemes 2 and 3, the synthesis of the biaryl
framework of 5 follows exactlythe same approach used
for the preparation of MeO–NAPhePHOS. The halo-
genated aromatic fragments, for example, 1-iodo-2-
naphthol and 2-bromo-3,4,5-trimethylphenol, have been
connected to the chiral tether, (R,R)-2,4-pentanediol,
through ether bonds, which are formed bytwo successive
Mitsunobu reactions. Diisopropyl azodicarboxylate
is preferablyused as the reagent. High iyelds of the
desired ethers 17 and 2 are obtained in optimised
conditions.
In designing the first heterotopic atropisomeric diphos-
phine of the series above, we envisioned the use of the
2-naphthyl and 2-methoxyphenyl fragments, which
form the structures of BINAP and MeO–BIPHEP,14
respectively, two of the most known and successful
atropisomeric diphosphines.
Of course, a number of other aromatic and heteroaro-
matic fragments could be considered for the synthesis of
new biphenols and diphosphines bythe same synthetic
approach. As the second target in this study, we choose
diphosphine 5 where a naphthyl group is associated with
the 2,3,4-trimethylphenyl fragment. The 3,4,5-trimethyl-
phenyl moiety has already been used for the synthesis of
C2-symmetric atropisomeric diphosphines.15
The biaryl coupling reaction has been performed
according to Lipshutz’s method:8;9 after halogen–lith-
ium exchange with sec-BuLi, the mixed cuprate is
formed byaddition of 1.5 equiv of CuCN. Finally, oxi-
dation of the cuprate with molecular oxygen affords the
macrocyclic bis-ether derivative 3 as a single diastereo-
isomer after purification bycolumn chromatography. In
this case as well, the chiral tether induces veryhigh
diastereoselectivity. The configuration of the biaryl
frame obtained byusing ( R,R)-2,4-pentanediol as
starting material, is assumed to be (R) byanalogyto the
stereochemical course of the biaryl coupling reaction in
the synthesis of MeO–NAPhePHOS. In that case the
PPh2
PPh2
5
Br
OH
Br
OH OH
O
OH
O
I
OH
O
I
i.
+
85%
I
i., 90%
(R,R)
(R,S)-1
(S,S)-2
ii.
iii.
81%
OH
OH
O
O
2
70%
(R,S,S)-3
(R)-4
Scheme 2. Synthesis of the unsymmetric diphenol(R)-4. Reagents and conditions: (i) PPh3, i-PrO2CN@NCO2i-Pr (DIAD), THF, 0–25 °C, 48 h; (ii)
sec-BuLi, )78 °C, THF, CuCN, O2, 70% yield; (iii) BBr3, CH2Cl2, )50 °C, 81% yield.
i.
OH
OH
PPh2
PPh2
P(O)Ph2
PPh2
PPh2
P(O)Ph2
+
+
ii.
(R)-4
(R)-5
(R)-6
(R)-7
iii.
Scheme 3. Synthesis of (R)-TriMe–NAPhePHOS, 5. Reagents and conditions: (i) Tf2O, pyridine, CH2Cl2, rt., 12 h, 95% yield; (ii) HPPh2,
NiCl2 (dppe), DABCO, DMF, 100 °C, 3 days: 72% total yield (phosphine + monooxides); (iii) HSiCl3/n-Bu3N, xylene, 140 °C, 18 h, 97% yield.

J. Madec et al. / Tetrahedron: Asymmetry 15 (2004) 2253–2261
2255
absolute configuration of the biaryl moiety had been
established by X-ray crystallography.
ketoesters (entries 4–6) the enantioselectivitylevels at-
tained bythe unsmymetrical ligands MeO–NAPhe-
PHOS and TriMe–NAPhePHOS lie always between
those given byBINAP and MeO–BIPHEP under anal-
ogous conditions.
The chiral auxiliarywas removed byreacting 3 with an
excess BBr3. The biphenol 4 was thus obtained in 81%
yield, with a total 42% yield over four steps. The enan-
tiomeric purity(ee >99%) was determined byHPLC on
a Chiralcel OD-H column, flow rate: 1 mL/min, eluent:
hexane/propan-2-ol (99:1), t ¼ 35.1 min (R-assumed
configuration), t ¼ 43.0 (S).
In previous work, small but significant differences in the
catalytic performances of atropoisomeric ligands have
been tentativelyrelated to small changes in the dihedral
angles of the biaryl framework.3;19 An analogous struc-
ture–enantioselectivityrelationship could roughlyhold
here: the calculated dihedral angle values (h) for MeO–
NAPhePHOS and TriMe–NAPhePHOS in their ruthe-
nium complexes (P*P)Ru(H)Br(MeCOCH2CO2Me) are
of 77.2° and 77.0°, respectively20 (Fig. 1).
The diphenylphosphino groups were then introduced
on the biaryl framework by nickel catalysed coupling of
the bis-triflate of 4 with diphenylphosphine (Scheme 3).16
The diphosphine 5, was obtained in low yield, but a
significant amount of the corresponding diphosphine-
monooxides were recovered after column chromato-
graphy. Even under rigorously inert gas atmosphere this
partial oxidation could not be avoided. However, the
phosphine oxides were quantitativelyreduced to the
These values are veryclose to each other and are
intermediate between those of BINAP (79.5°) and
MeO–BIPHEP (75.7°), which parallels the observed
variations in enantiomeric excesses. Nevertheless, com-
parison of the four ligands based onlyon geometrical
parameters can’t be fullysignificant, as the four ligands
don’t represent a homogeneous series with respect to
their electronic properties.
targeted diphosphine 5 byreduction with an HSiCl
/
3
n-Bu3N mixture, according to the usual procedure.
Finally, the diphosphine (R)-5, called TriMe–NAPhe-
PHOS, was recovered in a total 66% yield from diol 4. The
(S)-enantiomer of the same ligand has been obtained by
the same procedure, starting from (S,S)-2,4-pentanediol.
Evaluation of their relative basicity, as well as of their
electron donating properties towards a transition metal,
has been attempted bycomparing the 1J(31P–77Se)
coupling constants of the diphosphine diselenides21 and
Thus, the preparation of TriMe–NAPhePHOS, 5, af-
fords a new example of the synthesis of unsymmetric
biaryl–diphosphines based on the diastereoselective,
copper mediated, biaryl coupling reaction.
the
m
(CO) stretching frequencies of their
RhCl(CO)(P*P) complexes21, respectively. Results are
shown in Table 2.
1
The JP–Se coupling constants values for MeO–NAPhe-
PHOS fit well with those of BINAP and MeO–BIPHEP,
and suggest a similar basicity.
2.2. Ruthenium promoted hydrogenation reactions
With the unsymmetric ligands (R)-MeO–NAPhePHOS17
and (R)-TriMe–NAPhePHOS in hand, we started a
systematic study of their catalytic properties, in
comparison with those of the C2-symmetric BINAP and
MeO–BIPHEP ligands. We have at first expanded our
previous work on ruthenium promoted hydrogenations
and we have then considered rhodium promoted 1,4-
addition reactions on model substrates.
The observed IR stretching frequencies of the rhodium
complexes of heterotopic diphosphines are probablynot
fullysignificant and can be hardlycompared to those of
BINAP and MeO–BIPHEP. This is mainlybecause
mixtures of two isomeric, undistinguishable complexes
are formed from unsymmetrical diphosphines and a
single, averaged m (CO) stretching frequencyis observed.
Moreover, the observed frequencies are out of the ex-
pected range, at lower frequencythan that of the MeO–
BIPHEP complex. The unsymmetrical nature of the
ligands might affect the IR spectra and prevent the direct
comparison with complexes containing C2-symmetric
ligands.
Both unsymmetric NAPhePHOS diphosphines, namely
(R)-5 and (R)-MeO–NAPhePHOS, have been engaged in
ruthenium promoted hydrogenations of carbonyl and
olefin derivatives. The usual model b-ketoesters as well as
other functionalised carbonyl derivatives have been used
as substrates. The ruthenium catalyst has been generated
from a mixture of (COD)Ru(2-Me-allyl)2 and the di-
phosphine, byaddition of 2 equiv of HBr, according to
the usual procedure.18 For comparative purposes, the
same hydrogenation reactions have been performed in
parallel experiments byusing ( R)-BINAP and (R)-MeO–
BIPHEP. Results are reported in Table 1.
Finally, the observed trends in enantioselectivities seem
to correlate roughlywith the geometric features of these
ligands, nevertheless the small but significant differences
in their electronic properties could affect their behav-
iours as well.
Ligand 5 affords veryhigh enantiomeric excesses in the
hydrogenation reactions where other atropoisomeric
ligands also perform successfully(entries 1,3,7). In the
case of more challenging substrates such as a-ketoesters
(entry8), olefins (entries 9 and 10) and halogenated b-
2.3. Rhodium promoted 1,4-additions of boronic acids to
enones
For a second series of catalytic tests we considered a
C–C bond forming reaction, the rhodium promoted

2256
J. Madec et al. / Tetrahedron: Asymmetry 15 (2004) 2253–2261
Table 1. Hydrogenation of functionalised carbonyl and olefin derivatives by ruthenium–diphosphine complexes
Ee (config)
EntrySubstrate
Conditions
PPh2
PPh2
PPh2
PPh2
PPh2 MeO
PPh2 MeO
PPh2
PPh2
MeO
O
O
1
4 bar, 50 °C, 24 h, MeOH
99
90
98
90
25
45
99
72
99
93
98
86
32
48
99
85
99
92
99
83
32
51
99
75
99
97
99
78
42
58
99
90
(R)
(S)
(S)
(S)
(S)
(S)
(R)
(R)
OMe
O
O
2
3
4
5
6
7
8
10 bar, 80 °C, 24 h, EtOH
4 bar, 50 °C, 24 h, EtOH
10 bar, 110 °C, 3 h, EtOH
10 bar, 110 °C, 1 h, EtOH
10 bar, 110 °C, 1 h, EtOH
10 bar, 50 °C, 24 h, EtOH
20 bar, 50 °C, 24 h, MeOH
Ph
OEt
O
O
OEt
O
O
Cl
OEt
O
O
F3C
OEt
O
O
C2F5
O
OEt
O
P(OEt)2
O
Ph
CO2Me
CO2Me
9
4 bar, 50 °C, 24 h, MeOH
10 bar, 50 °C, 24 h, EtOH
89
75
90
70
89
69
90
68
(S)
(R)
CO2Me
CO2Me
NHAC
10
Ph
Reactions were conducted on a 1 mmol scale, using 1 mol % of in situ prepared RuBr2[(R)-ligand] as catalyst. All conversions were quantitative,
according to 1H NMR spectroscopy. Enantiomeric excesses (ee) were determined by chiral liquid and gas chromatography.
Figure 1. CAChe MM2 representation of the (MeO–NAPhePHOS)Ru(H)Br-(MeCOCH2CO2Me) complex (unsymmetric atropisomeric biaryl
backbone not represented for clarity, blue ¼ P, red ¼ O, green ¼ Br, white ¼ Ru and H, grey ¼ C).
1,4-addition of boronic acids to unsaturated ketones.22
As shown in the pioneering work of Hayashi and
co-workers,23 atropisomeric diphosphines behave as
highlyefficient chiral auxiliaries in these reactions.

J. Madec et al. / Tetrahedron: Asymmetry 15 (2004) 2253–2261
2257
Table 2. 1J(31P–77Se) coupling values in diphosphines diselenides. Carbonyl stretching frequencies of RhCl(CO)(P*P) complexes
PPh2
PPh2
PPh2
PPh2
PPh2
PPh2
MeO
MeO
PPh2
PPh2
MeO
1J(31P–77Se) (Hz)
m (CO) (cmÀ1
738
2016
739; 742
2009
723; 736
2011
742
2014
)
Cyclohexenone and 3-nonen-2-one was reacted with
arylboronic acids in standard conditions, by using a
dioxane/H2O mixture as the solvent and 3% of the
rhodium complex Rh(acac)(C2H4)2 as the catalyst pre-
cursor (Scheme 4). The rhodium-phosphine complexes
were formed in situ byusing the ( S) configured ligands.
Results are shown in Table 3.
bound to phosphorus and the exact nature of the biaryl
moietyhardlyaffects the stereochemical course of these
reactions. These preliminarydata do not exclude that
more pronounced effects of the biaryl moieties on the
enantioselectivitycould be evidenced bycomparing
atropisomeric diphosphines with stereoelectronic prop-
erties markedlydifferent from each other.
Results in Table 3 are issued from parallel screening of
the four ligands. For comparison, literature data (ees
and isolated yields) are given in brackets. Conversion
rates and enantiomeric excesses obtained with the
unsymmetric ligands are fully comparable to those
obtained with BINAP and MeO–BIPHEP in analogous
conditions. This shows that the C2-symmetry of the
chiral diphosphine is not a prerequisite in these
reactions and that diphosphines with unsymmetric bi-
aryl framework are well suited ligands. This is in
agreement with the Hayashi’s mechanism24 and stereo-
chemical model where the diphosphine is supposed to
chelate the metal in the keyintermediate of the catalytic
cycle (Fig. 2).
3. Conclusion
Finally, these preliminary catalytic tests show that the
new ligand 5 is an efficient, unsymmetric atropisomeric
diphosphine, whose catalytic properties in ruthenium
promoted hydrogenations and rhodium catalysed
1,4-addition reactions are fullycomparable to those of
the reference ligands BINAP and MeO–BIPHEP.
Fine-tuning of the reaction conditions, including the
choice of the catalyst precursors, is beyond the objec-
tives of the present preliminarywork. It will be per-
formed in the future, on selected and synthetically
relevant substrates.
4. Experimental section
4.1. General methods
P
P
Rh
NMR spectra were recorded on a Bruker AC 200 (at
50 MHz for 13C), on a Bruker AC 300 (at 75 MHz for
13C) or on a Bruker Avance 400 (at 100 MHz for 13C, at
162 MHz for 31P, at 376 MHz for 19F). Chemical shifts
(d) are reported in ppm downfield relative to external
Me4Si, H3PO4 (85%) or CFCl3 (1% in CDCl3). Mass
spectra were recorded on a Ribermag instrument.
Optical rotations were measured on a Perkin–Elmer 241
polarimeter. Melting points (mp) were determined on a
Kofler melting point apparatus. Enantiomeric excesses
O
Ar
Figure 2. Key intermediate in the catalytic cycle of the rhodium-(S)-
TriMe-NAPhePHOS promoted 1,4-addition reactions, according to
the Hayashi’s stereochemical model.
The chiral discrimination should come mainlyfrom the
three dimensional arrangement of the phenyl groups
O
O
Rh(acac)(C2H4)2(3 mol %)
O
Ar
(S)-diphosphine (3 mol %)
or
Ar B(OH)2
+
C5H11
or
C5H11
Scheme 4. Rhodium/diphosphine promoted 1,4-additions of boronic acids to unsaturated ketones.
dioxane/H2O (10:1)
100°C, 5h
Ar
O

2258
J. Madec et al. / Tetrahedron: Asymmetry 15 (2004) 2253–2261
Table 3. Asymmetric 1,4-addition of ArB(OH)2 to enones catalyzed by diphosphine–rhodium complexes
PPh2
PPh2
PPh2
PPh2
PPh2
PPh2
MeO
MeO
PPh2
PPh2
MeO
Substrates
(%) Ee (conv.)
(%) Ee (conv.)
98 (95)
(%) Ee (conv.)
(%) Ee (conv.)
98 (97)
O
PhB(OH)2
98 (95) [97 (64)]23
96 (98) [96 (97)]23
98 (95)
98 (96)
O
O
95 (98)
96 (98)
MeO
B(OH)2
98 (50)
96 (48)
88 (97)
96 (46)
87 (96)
97 (45)
87 (98)
B(OH)2
PhB(OH)2
O
90 (98) [92 (88)]23
C5H11
Reactions were conducted on a 0.4 mmol scale, using 3 mol % of Rh(acac)(C2H4)2 as catalyst. Reactions were performed in dioxane/water (10:1) at
100 °C, for 5 h. Conversions were measured by 1H NMR spectroscopy. Enantiomeric excesses (ee) were determined by chiral HPLC.
were determined byGC on a Lipodex A and Chirasil-L-
Val capillarycolumns or byHPLC analyses on a Waters
600 system, using Daicel chiral columns. All reactions
were carried out under an atmosphere of argon unless
otherwise specified.
(CH–O), 89.5 (C-I), 113.7 (C–Br), 114.8 (CHortho-O),
115.4 (CHortho-O), 124.0, 127.7, 127.9, 128.5 (C), 129.7,
129.8, 131.2, 135.5 (C), 135.7 (C), 136.5 (C), 152.1 (C–
O), 155.5 (C–O) ppm. MS (EI) m=z (81Br, %): 554 (M,
17%), 270 (53%), 41 (100%). Anal. Calcd for
C24H26BrIO2: C, 52.10; H, 4.74. Found: C, 52.26, H,
4.66. ½aŠ ¼ +107 (c 1, CHCl3).
D
4.2. (2S,4S)-4-(a-Iodo-b-naphthyloxy)-2-(2-bromo-3,4,5-
trimethylphenyloxy)pentane 2
4.3. Synthesis of (R,S,S)-3 via oxidative coupling of
(S,S)-2
Has been prepared bytwo successive Mitsunobu’s
reactions. The first step of the synthesis as well as
spectral data for (2R,4S)-4-(a-iodo-b-naphthyloxy)-2-
pentanol 1 have been reported previously. 2-Bromo-
3,4,5-trimethylphenol25 has been obtained in 95% yield
bybromination of the corresponding phenol with N-
bromosuccinimide in DMF at 0 °C. It was engaged then
with 1 in the second Mitsunobu-type reaction: a THF
solution (200 mL) containing 2-bromo-3,4,5-trimethyl-
phenol (10.1 g, 47 mmol) and diisopropylazodicarboxyl-
ate (9 mL, 56 mmol) was cooled to 5 °C. A solution of
alcohol 1 (16.7 g, 47 mmol) and PPh3 (14.7 g, 56 mmol)
in THF was then added slowly. The mixture was stirred
at room temperature for 24 h. After evaporation of the
solvent, cyclohexane (200 mL) was poured into the res-
idue. The insoluble by-products were removed by fil-
tration and the residue was purified bychromatography
on a silica gel column with cyclohexane/ethyl acetate
(90:10 mixture) as the eluent (Rf ¼ 0.8). The bis-ether 2
was isolated in 90% yield (23 g) as a yellow oil. 1H NMR
(200 MHz, CDCl3): d ¼ 1.40 (d, J ¼ 6.2 Hz, Me), 1.46
(d, J ¼ 6.2 Hz, Me), 1.78 (s, Me), 1.82 (s, Me), 2.1 (m,
2H, CH2), 2.28 (s, Me), 4.7–4.8 (m, 1H, CH–O), 4.8–4.9
(m, 1H, CH–O), 6.34 (s, 1H), 6.98 (d, J ¼ 9.0 Hz, 1H),
7.3–7.6 (4H), 8.8 (d, J ¼ 8.5 Hz, 1H); 13C NMR
(50 MHz, CDCl3): d ¼ 15.8 (Me), 20.3 (Me), 20.6 (Me),
20.8 (Me), 21.0 (Me), 45.3 (CH2), 73.4 (CH–O), 73.8
The dihalogenated substrate 2 (19.5 g, 35 mmol) was
dissolved in anhydrous THF (200 mL) and the solution
cooled to )78 °C. An excess sec-BuLi (108 mL, 1.3 M
solution in hexane) was added dropwise and the reaction
mixture was stirred at )78 °C for 1 h. Then, CuCN
(3.2 g, 39 mmol) was added and the reaction mixture was
stirred vigorouslyfor about 2 h at the same temperature.
Dryoxygen was bubbled through the solution for 2 h at
)78 °C. After warming to room temperature, the reac-
tion mixture was hydrolyzed with a saturated solution of
aqueous ammonium chloride. The organic phase was
extracted with diethyl ether and dried over MgSO4. The
final product was purified bycolumn chromatography
with cyclohexane/ethyl acetate (85:15 mixture) as eluent
(Rf ¼ 0.4). The macrocyclic ether 3 was thus obtained as
a single diastereoisomer in 70% yield (8.9 g), as a col-
1
ourless solid. Mp 55 °C. H NMR (300 MHz, CDCl3):
d ¼ 1.26 (d, J ¼ 6.5 Hz, Me), 1.42 (d, J ¼ 6.4 Hz, Me),
1.72 (m, AB, JAB ¼ 15.3 Hz, J ¼ 4.4 Hz, J ¼ 2.8 Hz, 1H,
CH2), 1.8–2.0 (m, 1H, CH2), 1.91 (s, Me), 2.23 (s, Me),
2.37 (s, Me), 4.6 (m, 2H, CH–O), 6.90 (s, 1H), 7.3–7.4
(4H), 7.8 (2H); 13C NMR (50 MHz, CDCl3): d ¼ 15.6
(Me), 18.3 (Me), 21.0 (Me), 22.2 (Me), 22.8 (Me), 41.5
(CH2), 74.8 (CH–O), 75.4 (CH–O), 117.4 (CHortho-O),
119.1 (CHortho-O), 123.8, 125.5 (C), 125.9, 126.1, 126.2

J. Madec et al. / Tetrahedron: Asymmetry 15 (2004) 2253–2261
2259
(C), 128.0, 128.7, 129.4 (C), 130.2 (C), 133.2 (C), 136.4
(C), 136.5 (C), 154.4 (C–O), 155.0 (C–O) ppm. MS (EI)
m=z (%): 346 (M, 100%), 278 (87%). Anal. Calcd for
C24H26O2: C, 83.20; H, 7.56. Found: C, 82.81; H, 7.57.
100 °C for 45 min. A degassed solution containing the
bis-triflate of 4 (0.82 g, 1.5 mmol) and DABCO (0.68 g,
6.0 mmol) in DMF (10 mL) was added to the nickel
solution. The mixture was heated at 100 °C and further
portions of HPPh2 (150 lL for each portion; total
amount of HPPh2 ¼ 3.5 mmol) were added after 1, 3,
and 8 h. Heating was maintained for further 3 days.
After evaporation of the solvent, the final product was
purified byflash chromatographyon silica gel with a
cyclohexane/ethyl acetate gradient (from 95:5 to 80:20
mixtures) as eluent. The diphosphine 5 and the corres-
ponding monoxides 6 and 7 were obtained in a total 72%
yield.
½aŠ ¼ )120 (c 1, CHCl3).
D
4.4. (R)-5,6-Benzo-40,50,60-trimethyl-2,20-biphenol 4
A solution of boron tribromide in CH2Cl2 (12 mL, 1 M
solution) was added to a cooled solution (0 °C) of 3
(1.4 g, 4 mmol) in CH2Cl2 (10 mL). After stirring for 3 h
at 0 °C, an aqueous solution of HCl (20 mL, 10% solu-
tion) was added at the same temperature. Extraction of
the organic phase with ethyl acetate, drying over MgSO4
and purification bycolumn chromatographywith
cyclohexane/ethyl acetate (70:30) as the eluent gave 4 as
a colourless solid (81% yield, 890 mg). The enantiomeric
puritywas determined byHPLC analysis: Chiralcel OD-
H column, flow rate:1 mL/min, eluent: hexane/propan-2-
ol (99:1), detection at 254 nm, retention times 35.1 min
4.5.1. Diphosphine 5. White solid. Mp 85–88 °C. 1H
NMR (400 MHz, CDCl3): d ¼ 1.34 (s, Me), 2.12 (s,
Me), 2.29 (s, Me), 6.89–7.32 (m, 22H), 7.74 (d,
J ¼ 8.8 Hz, 1H), 7.77 (d, J ¼ 8.8 Hz, 1H);13C NMR
(100 MHz, CDCl3) (selected data): d ¼ 17.6 (Me), 21.2
(Me), 27.0 (Me), 143.0 (C, dd, J ¼ 28.8, Hz,
J ¼ 10.5 Hz), 147.0 (C, dd, J ¼ 30.7 Hz, J ¼ 9.7 Hz);
31P NMR (162 MHz, CDCl3): d ¼ )14.0 and )13.8
(JAB ¼ 19.6 Hz); MS (DCI/NH3) m=z (%): 615
([M + H]þ, 100%); HRMS-DCI/NH3: [M + H]þ calcd for
1
(R) and 43.0 min (S), ee >99%. Mp 184 °C. H NMR
(200 MHz, CDCl3): d ¼ 1.89 (s, Me), 2.20 (s, Me), 2.37
(s, Me), 4.4 (m, 1H, OH), 5.1 (m, 1H, OH), 6.83 (s, 1H),
7.2–7.4 (4H), 7.8 (1H), 7.88 (d, J ¼ 9.1 Hz, 1H); 13C
NMR (50 MHz, CDCl3): d ¼ 15.4 (Me), 17.0 (Me), 21.0
(Me), 113.4 (Cortho-O), 114.8 (CHortho-O), 115.6 (Cortho-O),
117.4 (CHortho-O), 123.8, 124.0, 127.2, 128.0 (C), 128.3,
129.2 (C), 130.7, 133.1 (C), 137.6 (C), 139.1 (C), 151.8
(C–O), 152.0 (C–O) ppm. MS (DCI/NH3) m=z (%): 286
(M + NH4, 100%), 279 (M + H, 87%). Anal. Calcd for
C19H18O2: C, 81.99; H, 6.52. Found: C, 79.09; H, 6.64.
C43H36P2: 615.2371; found: 615.2363; ½aŠ ¼ +43 (c 1,
D
CHCl3).
4.5.2. Diphosphine oxide 6 or 7. White solid. Mp 102–
1
105 °C. H NMR (300 MHz, CDCl3): d ¼ 1.01 (s, Me),
2.09 (s, Me), 2.32 (s, Me), 6.90–7.69 (m, 24H); 13C NMR
(75 MHz, CDCl3) (selected data): d ¼ 16.2 (Me), 16.7
(Me), 21.1 (Me); 31P NMR (121 MHz, CDCl3):
d ¼ )15.0 and 26.9.
½aŠ ¼ +24 (c 1, CHCl3).
D
4.5. (R)-5,6-Benzo-2,20-bis(diphenylphosphino)-40,50,60-
trimethylbiphenyl 5
The bis-triflate of 4 was prepared byadding slowlya
trifluoromethanesulfonic
4.5.3. Diphosphine oxide 6 or 7. White solid. Mp 108–
1
solution
of
anhydride
112 °C. H NMR (300 MHz, CDCl3): d ¼ 1.47 (s, Me),
(9.75 mmol) in CH2Cl2 (65 mL) to a cooled solution
containing diol 4 (0.90 g, 3.2 mmol) and pyridine
(0.79 mL, 9.75 mmol) in CH2Cl2 (65 mL). After the
addition was complete the reaction mixture was allowed
to warm slowlyto room temperature and stirred for
16 h. After evaporation of the solvent, the final product
was purified byfiltration on a short neutral alumina
column with cyclohexane/ethyl acetate (90:10 mixture)
as the eluent (Rf ¼ 0.7). The bis-triflate of 4 was ob-
1.98 (s, Me), 2.19 (s, Me), 6.70–7.80 (m, 24H); 13C NMR
(75 MHz, CDCl3) (selected data): d ¼ 15.8 (Me), 17.9
(Me), 30.2 (Me); 31P NMR (121 MHz, CDCl3):
d ¼ )15.4 and 27.9.
4.6. Reduction of the phosphine oxides 6 and 7
To a solution of the two phosphine oxides 6 and 7
(642 mg, 1.018 mmol) in dryxylene (7 mL) were added
n-tributylamine (2.91 mL, 12.22 mmol) and trichlorosi-
lane (1.03 mL, 10.18 mmol). The resulting mixture was
heated at 140 °C overnight. After cooling to room
temperature, 10 mL of degassed 1 N aqueous HCl
were added dropwise and the mixture was stirred for
30 min. DryCH 2Cl2 (30 mL) was then added, the
organic layer was washed with degassed 4 N aqueous
NaOH (10 mL), degassed distilled water (10 mL),
degassed brine (10 mL) and concentrated under vacuum.
The crude product was purified bycolumn chromato-
graphy with degassed cyclohexane/ethyl acetate (70:30)
mixture as eluent to afford 5 in 97% yield (610 mg) as a
white solid.
1
tained in 95% yield as a white solid. Mp 135 °C. H
NMR (200 MHz, CDCl3): d ¼ 1.97 (s, Me), 2.31 (s,
Me), 2.46 (s, Me), 7.19 (s, 1H), 7.4–7.6 (4H), 7.97 (dd,
J ¼ 8 Hz, J ¼ 1Hz, 1H), 8.04 (d, J ¼ 9.3 Hz, 1H); 13C
NMR (50 MHz, CDCl3): d ¼ 15.8 (Me), 18.1 (Me), 20.9
(Me), 119.0, 119.6, 123.4 (C), 125.7 (C), 126.3, 127.0,
127.8, 128.3, 131.2, 132.3 (C), 132.7 (C), 136.2 (C), 139.2
(C), 139.4 (C), 144.8 (C–O), 145.1 (C–O) ppm. MS (EI)
m=z (%): 542 (M, 20%), 276 (40%), 260 (50%), 69
(100%).
A solution of NiCl2 (dppe) (160 mg, 0.3 mmol) in
anhydrous DMF (5 mL) was degassed. HPPh2 (150 lL,
0.87 mmol) was added and the mixture was heated at

2260
J. Madec et al. / Tetrahedron: Asymmetry 15 (2004) 2253–2261
4.7. Diphosphine selenides
experiments several glass tubes were introduced in the
same autoclave. Conversions were determined by 1H
NMR. Enantiomeric excesses and absolute configura-
tions were determined byGC or HPLC, bycomparison
with known samples.
To a mixture of a diphosphine (0.03 mmol) and selenium
powder (50 mg, excess) was added degassed chloroform
(2 mL). The mixture was stirred and heated at reflux for
5 h and then cooled to room temperature. After filtra-
tion on Celite eluting with chloroform, the solvents were
removed under reduced pressure and a yellow pale solid
was obtained.
Methyl 3-hydroxybutyrate. GCanalysis : Lipodex A
column, flow 1 mL/min, initial temperature 35 °C
(30 min), rate 1 °C/min, final temperature 100 °C.
Retention times 45.3 (S), 48.0 (R).
TriMe–NAPhePHOS selenide: 31P NMR (CDCl3)
d ¼ 33.7 and 37.3 ppm
MeO–NAPhePHOS selenide: 31P NMR (CDCl3)
Ethyl 3-hydroxy-3-phenylpropionate. GCanalysis : Lipo-
dex A column, flow 1 mL/min, temperature 110 °C.
Retention times 74.0 (S), 75.4 (R).
d ¼ 33.6 and 34.7 ppm
Ethyl 3-hydroxy-4-methylvalerate. GCanalysis : Lipodex
A column, flow 0.5 mL/min, initial temperature 40 °C
(15 min), rate 0.5 °C/min, final temperature 70 °C.
Retention times 93.4 (S), 94.9 (R).
4.8. RhCl(CO)(P*P)] complexes
To a mixture of [ClRh(CO)2]2 (0.03 mmol) and a di-
phosphine ligand (0.06 mmol) was added degassed
CH2Cl2 (1.5 mL). The mixture was stirred at room
temperature for 1 h. The solvents were removed under
reduced pressure and an orange crystalline solid was
obtained.
Ethyl 4-chloro-3-hydroxybutyrate. GCanalysis : Lipodex
A column, flow 0.5 mL/min, temperature 70 °C. Reten-
tion times 74.5 (S), 77.8 (R).
Ethyl 4,4,4-trifluoro-3-hydroxypropionate. GCanalysis :
Lipodex A column, flow 1 mL/min, initial temperature
50 °C (15 min), rate 2 °C/min, final temperature 100 °C.
Retention times 30.8 (S), 31.1 (R).
TriMe–NAPhePHOS gave a 1:1 mixture of two isomeric
complexes. Complex A (tentative assignment): 31P NMR
(CDCl3) at d ¼ 23.2 (JP–Rh ¼ 128 Hz, JP–P ¼ 45 Hz), 45.0
(JP–Rh ¼ 162 Hz). Complex B (tentative assignment): 31
P
Ethyl 4,4,5,5,5-pentafluoro-3-hydroxypropionate. GC
analysis: Lipodex A column, flow: 0.5 mL/min, initial
temperature 40 °C (45 min), rate 0.5 °C/min, final tem-
perature 90 °C. Retention times 82.3 (R), 83.4 (S).
NMR
(CDCl3)
at
d ¼ 24.4
(JP–Rh ¼ 129 Hz,
JP–P ¼ 45 Hz), 46.1 (JP–Rh ¼ 164 Hz).
The 31P NMR spectrum of the MeO–NAPhePHOS-
rhodium complex showed broad signals at about 24 and
44 ppm.
Diethyl 2-hydroxypropylphosphonate. GCanalysis : Chi-
ralsil-L-Val column, flow 1.4 mL/min, temperature
110 °C. Retention times 56.2 (R), 58.8 (S).
4.9. Computational methods
Ethyl 2-hydroxy-3-mꢁethylbutyrate. GCanalysis : Lipodex
A column, flow 1 mL/min, temperature 45 °C. retention
times 39.3 (S), 40.2 (R).
The molecular geometries of ruthenium complexes
incorporating diphosphine ligands were optimized
with the molecular mechanics program CAChe
(Computer Aided Chemistry), using the force field
parameters MM2. Calculation type: structure optimi-
zation, optimization method: conjugate gradient for no
more than 3000 updates or until convergence to 10À4
Methyl mandelate. GCanalysis : Lipodex A column, flow
1 mL/min, temperature 100 °C. retention times 31.2 (S),
33.3 (R).
Dimethyl succinate. HPLCanalysis : Chiralcel OD-H
column, eluent hexane/propan-2-ol (98:2), flow
0.8 mL/min, k ¼ 215 nm. Retention times 10.5 (S), 20.8
(R).
kcal molÀ1
.
4.10. Asymmetric hydrogenation. Typical procedure
N-Acetamidophenylalanine methyl ester. HPLCanalysis :
Chiralcel OD-H column, eluent hexane/propan-2-ol
(90:10), flow 1 mL/min, k ¼ 254 nm. Retention times
10.8 (S), 13.4 (R).
Hydrogenation experiments were performed in a glass
tube, under magnetic stirring. All reactions have been
made at a 1 mmol scale, with 1% ruthenium catalyst,
which was prepared from (COD)Ru(2-methylallyl)2
(3.2 mg) and the chiral diphosphine 5 (7.4 mg, 1.2 equiv),
byaddition of 2.2 equiv of aqueous HBr (0.16–0.18 M)
in acetone. After evaporation of the solvent, the crude
residue was taken up in degassed MeOH (2 mL) or
EtOH (2 mL), substrate was added and the glass tube
was placed in a stainless steel autoclave, under H2 at a
given pressure and temperature (see Table 1). In parallel
4.11. Rhodium promoted 1,4-additions of boronic acids to
unsaturated ketones: typical procedure
To a mixture of degassed Rh(acac)(C2H4)2 (3 mol %), a
diphosphine ligand (3.3 mol %) and arylboronic acid
(5 equiv) at room temperature were successivelyadded

J. Madec et al. / Tetrahedron: Asymmetry 15 (2004) 2253–2261
2261
5. Duprat de Paule, S.; Jeulin, S.; Ratovelomanana-Vidal,
dioxane (1 mL), water (0.1 mL) and the enone substrate
(0.4 mmol). The homogeneous mixture was heated at
100 °C for 5 h and then cooled to room temperature.
After filtration on Celite (ethyl acetate), the organic
layer was washed with saturated aqueous NaHCO3 and
brine. It was dried then over MgSO4 and evaporated
under reduced pressure. Chromatographyover silica gel
(cyclohexane/AcOEt ¼ 90:10) afforded the correspond-
ing 1,4-addition product. Enantiomeric excesses were
determined byHPLC analysis.
^
V.; Genet, J.-P.; Champion, N.; Dellis, P. Eur. J. Org.
Chem. 2003, 1931.
6. Jeulin, S.; Duprat de Paule, S.; Ratovelomanana-Vidal,
^
V.; Genet, J.-P.; Champion, N.; Dellis, P. Angew. Chem.,
Int. Ed. 2004, 43, 320–325.
^
7. Michaud, G.; Bulliard, M.; Ricard, L.; Genet, J.-P.;
Marinetti, A. Chem. Eur. J. 2002, 8, 3327.
8. Lipshutz, B. H.; Liu, Z.-P.; Kayser, F. Tetrahedron Lett.
1994, 35, 5567.
9. Lipshutz, B. H.; Kaiser, Z.-P.; Liu, Z.-P. Angew. Chem.,
Int. Ed. Engl. 1994, 33, 1842.
10. Sugimura, T.; Yamada, H.; Inoue, S.; Tai, A. Tetrahedron:
Asymmetry 1997, 8, 649–655.
11. Sugimura, T.; Tei, T.; Mori, A.; Okuyama, T.; Tai, A.
J. Am. Chem. Soc. 2000, 122, 2128–2129.
12. Sugimura, T.; Hagiya, Y.; Sato, Y.; Tei, T.; Tai, A.;
Okuyama, T. Org. Lett. 2001, 3, 37–40.
3-Phenylcyclohexanone. HPLCanalysis : Chiralpak AS-
H column, eluent hexane/propan-2-ol (98:2), flow
1.0 mL/min, k ¼ 215 nm. Retention times 18.6 (R), 20.8
(S).
4-Phenylnonan-2-one. HPLCanalysis : Chiralcel OJ col-
umn, eluent hexane/propan-2-ol (99:1), flow 1.0 mL/min,
k ¼ 215 nm. Retention times 8.1 and 8.7.
13. Tei, T.; Sato, Y.; Hagiya, K.; Okuyama, T.; Sugimura, T.
J. Org. Chem. 2002, 67, 6593–6598.
€
14. Schmid, R.; Foricher, J.; Cereghetti, M.; Schonholzer, P.
Helv. Chim. Acta 1991, 74, 370.
15. Henschke, J. P.; Burk, M. J.; Malan, C. G.; Herzberg, D.;
Peterson, J. A.; Wildsmith, A. J.; Cobley, C. J.; Casy, G.
Adv. Synth. Catal. 2003, 345, 300.
16. Cai, D.; Payack, J. F.; Bender, D. R.; Hughes, D. L.;
Verhoeven, T. R.; Reider, P. J. J. Org. Chem. 1994, 59,
7180.
17. The sample used in this work has been prepared starting
from 1-iodo-2-naphthol, 2-iodo-3-methoxyphenol and
(R,R)-2,4-pentanediol, according to Scheme 1. The experi-
mental procedure for the synthesis of 2-iodo-3-methoxy-
phenol has been improved bychanging anisol into the
corresponding methoxymethyl ether before the lithiation–
iodination step. Despite the additional protection–depro-
tection steps, the new procedure affords a significantly
higher total yield with respect to the previously published
direct lithiation–iodination of anisol. For model proce-
dures see: Winkle, M. R.; Ronald, R. C. J. Org. Chem.
1982, 47, 2101–2108.
3-(3-Methoxyphenyl)cyclohexanone. HPLCanalysis
:
Chiralcel OJ column, eluent hexane/propan-2-ol (90:10),
flow 1.0 mL/min, k ¼ 215 nm. Retention times 13.7 and
15.6.
3-(1-Naphtyl)cyclohexanone. HPLCanalysis : Chiralpak
AD column, eluent hexane/propan-2-ol (95/5), flow
1.0 mL/min, k ¼ 215 nm. Retention times 10.7 and 13.8.
Acknowledgements
CNRS and PPG-Sipsyare gratefullyacknowledged for
doctoral fellowships (J.M. and G.M.). We thank Dr. A.
ꢁ
Solladie-Cavallo (Department of Fine Organic Chem-
istry, ECPM, Strasbourg, France) for her permission to
use molecular modelling facilities and Dr. R. Schmid
(Hoffmann-LaRoche) for a gift of (R)- and (S)-MeO–
BIPHEP.
^
18. Genet, J.-P.; Pinel, C.; Ratovelomanana-Vidal, V.; Mall-
art, S.; Pfister, X.; Bischoff, L.; Cano de Andrade, M.;
Darses, S.; Galopin, C.; Laffitte, J.-A. Tetrahedron:
Asymmetry 1994, 5, 675.
19. Zhang, Z.; Qian, H.; Longmire, J.; Zhang, Z. J. Org.
Chem. 2000, 65, 6223–6226.
20. Dihedral angles have been calculated byusing the CAChe
MM2 molecular modeling software.
21. (a) Allen, D. W.; Taylor, B. F. J. Chem. Soc., Dalton
References and notes
ꢁ
Trans. 1982, 51; (b) Suarez, A.; Mendez-Rojas, M. A.;
1. For recent reviews, see: (a) McCarthy, M.; Guiry, P. J.
Tetrahedron 2001, 57, 3809–3844; (b) Tang, W.; Zhang, X.
Chem. Rev. 2003, 103, 3029–3069.
Pizzano, A. Organometallics 2002, 21, 4611.
22. Hayashi, T.; Yamasaki, K. Chem. Rev. 2003, 103, 2829–
2844.
ꢀ
2. Benincori, T.; Piccolo, O.; Rizzo, S.; Sannicolo, F. J. Org.
23. Takaya, Y.; Ogasawara, M.; Hayashi, T.; Sakai, M.;
Miyaura, N. J. Am. Chem. Soc. 1998, 120, 5579–5580.
24. Hayashi, T.; Takahashi, M.; Takaya, Y.; Ogasawara, M.
J. Am. Chem. Soc. 2002, 124, 5052–5058.
25. Fischer, A.; Henderson, G. N. Can. J. Chem. 1983, 61,
145–1052.
Chem. 2000, 65, 8340–8347.
3. Wu, S.; Wang, W.; Tang, W.; Lin, M.; Zhang, X. Org.
Lett. 2002, 4, 4495–4497.
4. Qiu, L.; Qi, J.; Pai, C.-C.; Chan, S.; Zhou, Z.; Choi, M. C.
K.; Chan, A. S. C. Org. Lett. 2002, 4, 4599–4602.