Highly enantioselective Rh-catalyzed hydrogenation based on phosphine-phosphite ligands derived from carbohydrates

Article abstract of DOI:10.1021/jo0157122

A new class of efficient catalysts was developed for the asymmetric hydrogenation of α,β-unsaturated carboxylic acid derivatives by synthesizing a series of novel phosphine-phosphite ligands (4a-d) derived from readily available D-(+)-xylose. Excellent en

Full text of DOI:10.1021/jo0157122

8364
J . Org. Chem. 2001, 66, 8364-8369
High ly En a n tioselective Rh -Ca ta lyzed Hyd r ogen a tion Ba sed on
P h osp h in e-P h osp h ite Liga n d s Der ived fr om Ca r boh yd r a tes
Oscar Pa`mies, Montserrat Die´guez, Gemma Net, Aurora Ruiz,* and Carmen Claver
Departament de Quı´mica Fı´sica i Inorga`nica, Universitat Rovira i Virgili, Pl. Imperial Ta`rraco 1,
43005 Tarragona, Spain
aruiz@quimica.urv.es
Received April 26, 2001
A new class of efficient catalysts was developed for the asymmetric hydrogenation of R,â-unsaturated
carboxylic acid derivatives by synthesizing a series of novel phosphine-phosphite ligands (4a -d )
derived from readily available D-(+)-xylose. Excellent enantioselectivities (>99%) were achieved
under very mild reaction conditions (1 bar H2 and 20 °C). Varying the biphenyl substituents in the
phosphite moiety greatly affected the enantioselectivity in the hydrogenation reactions. The results
also indicate that the sense of enantioselectivity is mainly controlled by the configuration of the
phosphite moiety. 31P{1H} NMR and kinetic studies on intermediates of the catalytic cycle show
that the [Rh(P1-P2)(enamide)]BF4 species is the resting state and that the rate dependence is first
order in rhodium and hydrogen pressure and zero order in enamide concentration.
In tr od u ction
available from nature at a reasonable price and because
they make tedious optical resolution procedures unneces-
sary. Moreover, carbohydrates are particularly advanta-
geous because they are highly functionalized compounds
with several stereogenic centers, which easily allows the
regio- and stereoselective introduction of different func-
tionalities.6 Their modular nature therefore allows the
synthesis of a systematic series of ligands3a-c,4,7 that can
be screened in the search for high activities and enan-
tioselectivities and, at the same time, can provide infor-
mation about the origin of the stereoselectivity of the
reaction.7j
In previous studies, different types of phosphorus
ligands with a xylofuranoside backbone have been ap-
plied to asymmetric hydrogenation with varying degrees
of success. Brunner et al. reported moderate enantiose-
lectivities (up to 35%) with phosphine-phosphinite and
diphosphinite ligands.8 More recently, we reported mod-
erate-to-good enantiomeric excesses with diphosphite (ee
up to 35%)4b and diphosphine (ee up to 91%)7e at room
temperature, but in both cases the activities were low.
Following our interest in using carbohydrates as an
available chiral source for ligands, and bearing in mind
Achiwa’s idea that two different donor sites can a priori
match the intermediates better and so influence their
reactivity and enantioselectivity,9 we have designed a
The asymmetric hydrogenation of functionalized pro-
chiral olefins is one of the most important applications
of asymmetric catalysis. Over many years, the scope of
this reaction has been gradually extended both in reac-
tant structure and catalyst efficiency.1 Chiral bidentate
phosphorus ligands have played a dominant role in the
success of asymmetric hydrogenation.1,2 Many chiral
diphosphines1,2 and diphosphinites3 have therefore been
synthesized as ligands for enantioselective transition-
metal catalyzed hydrogenations. Recent reports on the
use of chiral phosphite4 and phosphoroamidite5 ligands
in asymmetric hydrogenation have demonstrated their
potential utility. Nevertheless, the search for new highly
efficient ligand systems derived from readily available
simple starting materials is still of great importance in
catalysis research.
Chiral ligands from the chiral pool have recently
attracted a great deal of interest because they are easily
* To whom correspondence should be addressed. Fax: +34-
977559563.
(1) (a) Noyori, R. Asymmetric Catalysis in Organic Synthesis;
Wiley: New York, 1994. (b) Catalytic Asymmetric Synthesis; Ojima,
I., Ed.; Wiley: New York, 2000. (c) Comprehensive Asymmetric
Catalysis; J acobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer:
Berlin, 1999; Vol. 1.
(2) (a) Brunner, H.; Zettlmeier, W. Handbook of Enantioselective
Catalysis; VCH: Weinheim, 1993. (b) Togni, A.; Breutel, C.; Schnyder,
A.; Spindler, F.; Landert, H.; Tijani, A. J . Am. Chem. Soc. 1994, 116,
4062. (c) Nettekoven, U.; Kamer, P. C. J .; van Leeuwen, P. W. N. M.;
Widhalm, M.; Speck, A. L.; Lutz, M. J . Org. Chem. 1999, 64, 3996. (d)
Barens, U.; Burk, M. J .; Gerlach, A.; Hems, W. Angew. Chem., Int.
Ed. 2000, 39, 1981.
(3) (a) Selke, R. J . Organomet. Chem. 1989, 370, 249. (b) RajanBabu,
T. V.; Ayers, T. A.; Casalnuovo, A. L. J . Am. Chem. Soc. 1994, 116,
4101. (c) RajanBabu, T. V.; Ayers, T. A.; Halliday, G. A.; You, K. K.;
Calabrese, J . C. J . Org. Chem. 1997, 62, 6012. (d) Chan, A. S. C.; Hu,
W.; Pai, C. C.; Lau C. P. J . Am. Chem. Soc. 1997, 119, 9570. (e) Ayers,
T. A.; RajanBabu, T. V. Process Chem. Pharm. Ind. 1999, 327.
(4) (a) Reetz, M. T.; Neugebauer, T. Angew. Chem., Int. Ed. 1999,
38, 179. (b) Pa`mies, O.; Net, G.; Ruiz, A.; Claver, C. Eur. J . Inorg.
Chem. 2000, 1287. (c) Reetz, M. T.; Mehler, G. Angew. Chem., Int. Ed.
2000, 39, 3889.
(6) Hanessian, S. In Total Synthesis of Natural Products: The
“Chiron” Approach; Pergamon Press: London, 1983; Vol. 3.
(7) For some recent applications, see: (a) Beller, M.; Kraute, J . G.
E.; Zapf, A. Angew. Chem., Int. Ed. Engl. 1997, 36, 772. (b) Yonehara,
K.; Hashizume, T.; Mori, K.; Ohe, K.; Uemura, S. Chem. Commun.
1999, 415. (c) Pa`mies, O.; Die´guez, M.; Net, G.; Ruiz, A.; Claver, C.
Organometallics 2000, 19, 1488. (d) Pa`mies, O.; Net, G.; Ruiz, A.;
Claver, C. Tetrahedron: Asymmetry 2000, 11, 1097. (e) Pa`mies, O.; Net,
G.; Ruiz, A.; Claver, Eur. J . Inorg. Chem. 2000, 2011. (f) Pa`mies, O.;
Die´guez, M.; Net, G.; Ruiz, A.; Claver, C. Chem. Commun. 2000, 1607;
(g) Li, W.; Zhang, Z.; Xiao, D.; Zhang, X. J . Org. Chem. 2000, 65, 3489.
(h) Pa`mies, O.; Die´guez, M.; Net, G.; Ruiz, A.; Claver, C. Tetrahedron:
Asymmetry 2000, 11, 4377. (i) Die´guez, M.; Pa`mies, O.; Ruiz, A.;
Castillo´n, S.; Claver, C. Tetrahedron: Asymmetry 2000, 11, 4701. (j)
Die´guez, M.; Pa`mies, O.; Ruiz, A.; Castillo´n, S.; Claver, C. Chem. Eur.
J . 2001, 7, 3086.
(5) van den Berg, M.; Minnaard, A. J .; Schudde, E. P.; van Esch, J .;
de Vries, A. H. M.; de Vries, J . G.; Feringa, B. L. J . Am. Chem. Soc.
2000, 122, 11539.
(8) Brunner, H.; Pieronczyk, W. J . Chem. Res., Synop. 1980, 3, 76.
10.1021/jo0157122 CCC: $20.00 © 2001 American Chemical Society
Published on Web 11/15/2001

Highly Enantioselective Rh-Catalyzed Hydrogenation
J . Org. Chem., Vol. 66, No. 25, 2001 8365
Sch em e 1. Syn th esis of P h osp h in e-P h osp h ite
Liga n d s 4a -d
Sch em e 2. Syn th esis of Rh od iu m (I)-Ca tion ic
Com p lexes 5a -d
Ta ble 1. Selected Sp ectr oscop ic NMR Da ta for
Com p lexes 5a -d a
P1 ) phosphine
P2 ) phosphite
compd
δ (P1)
J P1-Rh
δ (P2)
J P2-Rh
J P1-P2
5a
13.9
18.6
7.5
15.3
19.7
8.8
145.2
148.3
142.0
144.3
147.9
141.9
141.3
140.7
135.1
132.8
143.7
131.4
130.4
141.0
141.3
140.7
260.5
266.8
253.2
260.8
268.2
248.9
262.2
260.6
43.3
45.6
41.6
40.2
43.9
42.1
48.2
49.0
5a (55%)b
5a (45%)b
5b
5b (60%)b
5b (40%)b
5c
14.9
15.2
5d
a
NMR measured in CD2Cl2 at 25 °C, chemical shift (δ) in ppm,
b
coupling constants (J ) in hertz. NMR measured at -80 °C,
relative abundance in brackets.
new family of chiral bidentate phosphine-phosphite
ligands with xylofuranoside backbone. These ligands
combine the advantages of both types of ligand. We also
synthesized their cationic Rh(I) precursors and investi-
gated their use in the highly active and enantioselective
rhodium-catalyzed asymmetric hydrogenation of R,â-
unsaturated carboxylic acid derivatives.10 To the best of
our knowledge, this is the first example of phosphine-
phosphite ligands applied to hydrogenation.11 We also
carried out a kinetic study and investigated the reactivity
of some key intermediates.
The 1H and 13C NMR spectra were as expected for these
xylofuranoside ligands. Two doublets, one for each phos-
phorus moiety, were observed in the 31P NMR spectra.
The J P-P values of 12-33 Hz are similar to those reported
for other phosphine-phosphite ligands.11a Due to the
axial chirality of the biphenyl group, two pairs of signals
were expected for ligands 4a and 4b. However, the
existence of a single doublet for each phosphorus atom
in the variable-temperature 31P NMR suggested rapid
ring inversions (atropoisomerization) in the bisphenol-
phosphorus moieties on the NMR time scale.15
Syn th esis of Rh od iu m (I) Com p lexes. The olefinic
cationic Rh(I) complexes were prepared in high yields by
reacting stoichiometric amounts of chiral phosphine (P1)-
phosphite (P2) ligands 4a -d with [Rh(cod)2]BF4 in dichlo-
romethane (Scheme 2).
Complexes 5a -d were isolated as a yellow, moderately
air-stable powder by adding hexane. The elemental
analysis of C and H matched the stoichiometry [Rh(cod)-
(P1-P2)]n(BF4)n. The FAB mass spectra showed the
highest ion at m/z values that correspond to the cationic
mononuclear species.
Resu lts a n d Discu ssion
Syn t h esis of P h osp h in e-P h osp h it e Liga n d s.
Scheme 1 shows the sequence for the synthesis of the
ligands. The new ligands 4a -d were synthesized very
efficiently in two steps from oxetane 1, which is easily
prepared on a large scale from D-(+)-xylose.12 The key
step is the oxetane ring opening using a slight excess of
potassium diphenylphosphide in DMF to afford phos-
phine-alcohol intermediate 2. The yield was considerably
higher than with the method of Brunner et al., which
prepared 2 from related 5-tosylate derivative.13 Subse-
quently treating 2 with 1 equiv of the corresponding in
situ formed phosphorochloridite 3a -d 14 in the presence
of base provided easy access to the desired phosphine-
phosphite ligands 4a -d . These were isolated in good
yields as air-stable solids.
The 1H NMR spectra of complexes 5a -d were in
agreement with the expected coordinated pattern for the
xylofuranoside protons. Also as expected for these C1-
symmetrical complexes, there were four signals for the
olefinic carbon atoms of the coordinated cyclooctadiene.
However, the eight methylenic protons appeared as a
unique broad signal. This phenomenon has also been
reported for related asymmetric compounds with a
similar ligand backbone.4b,7c,16
(9) Inoguhi, K.; Sakuraba, S.; Achiwa. K. Synlett 1992, 169.
(10) The preliminary hydrogenation results were partly reported in
Pa`mies, O.; Die´guez, M.; Net, G.; Ruiz, A.; Claver, C. Chem Commun.
2000, 2383.
(11) Few phosphine-phosphite ligands have been described, al-
though some of them have been successfully applied in asymmetric
hydroformylation and allylic alkylation. (a) Nozaki, K.; Sakai, N.;
Nanno, T.; Higashijima, T.; Mano, S.; Horiuchi, T.; Takaya, H. J . Am.
Chem. Soc. 1997, 119, 4413. (b) Deerenberg, S.; Kamer, P. C. J .; van
Leeuwen, P. W. N. M. Organometallics 2000, 19, 2065. (c) Deerenberg,
S.; Schrekker, H. S.; van Strijdonck, G. P. F.; Kamer, P. C. J ., van
Leeuwen, P. W. N. M.; Fraanje, J .; Goubitz, K. J . Org. Chem. 2000,
65, 4810.
The VT-31P NMR spectra for complexes 5c and 5d ,
which contain a binaphthyl moiety, showed two sharp
double doublets due to the 31P/31P and 31P/103Rh couplings
(Table 1). These results are consistent with the presence
of a single isomer in solution. At room temperature, the
31P NMR spectra for complexes 5a and 5b showed a sharp
double doublet in the phosphite region and a broad
doublet in the phosphine region (Table 1). This suggested
fluxional processes on the NMR time scale, and these
were confirmed by measuring a VT-31P NMR. At 193 K,
the 31P NMR spectra therefore showed two sets of signals
(12) Oxetane
1 was prepared by a minor modification of the
procedure described in: Goueth, P. Y.; Fauvin, M. A.; Mashoudi, M.;
Ramiz, A.; Ronco, G. L.; Villa, P. J . J . Nat. 1994, 6, 3.
(13) Brunner et al. previously synthesized this phosphine in low
yield (10%) from 1,2-O-isopropylidene-5-tosyl-R-xylofuranose. Brunner,
H.; Pieronczyk, W. J . Chem. Res., Synop. 1980, 74; J . Chem. Res.,
Miniprint 1980, 1251.
(15) Pastor, D. D.; Shum, S. P.; Rodebaugh, R. K.; Debellis, A. D.;
Clarke, F. H. Helv. Chim. Acta 1993, 76, 900.
(14) Buisman, G. J . H.; Kamer, P. C. J .; van Leeuwen, P. W. N. M.
Tetrahedron: Asymmetry 1993, 4, 1625.
(16) Pa`mies, O.; Die´guez, M.; Net, G.; Ruiz, A.; Claver, C. J . Chem.
Soc., Dalton Trans. 1999, 3439.

8366 J . Org. Chem., Vol. 66, No. 25, 2001
Pa`mies et al.
Ta ble 2. Asym m etr ic Hyd r ogen a tion of Meth yl
(N)-Aceta m id oa cr yla te 6 a n d Meth yl
the lowest enantioselectivity (Table 2, entry 1). Moreover,
when methanol was used as a solvent, enantioselectivity
decreased slightly over time, which suggests that, due
to the partial decomposition of the ligand in methanol,
the catalyst evolved over time.18 The best catalyst
performance (in terms of activity and enantioselectivity)
was obtained when dichloromethane was used as a
solvent (Table 2, entry 2).
(Z)-(N)-Aceta m id ocin n a m a te 7 w ith [Rh (cod )2]BF 4/4a
As expected, the efficiency of the process was not
affected when preformed catalyst precursor 5b was used
(Table 2, entry 5) or when a 1-fold excess of ligand was
added (Table 2, entry 6). Heller et al. recently showed
that when norbornadiene (nbd) is used as a counter
ligand in the cationic Rh-Duphos asymmetric hydrogena-
tion, the reaction is faster than with related cycloocta-
diene complexes.19 We therefore compared the cod pre-
catalyst 5b with the corresponding nbd complex, but the
activity and enantioselectivity of the two catalyst precur-
sors were similar (Table 2, entry 7 vs 5).
The rest of the ligands were therefore tested under
“standard” conditions, i.e., dichloromethane as a solvent
and a ligand-to-rhodium ratio of 1. Ligand 4a , with the
nonsubstituted biphenyl moiety at the phosphite, re-
sulted in higher activity than ligand 4b, although the
enantioselectivity was lower (Table 2, entry 8 vs 2). The
presence of bulky tert-butyl groups in the ortho positions
of the biphenyl moiety therefore had an extremely
positive effect on enantioselectivity. Interestingly, the
sense of the enantioselectivity was reversed: the (S)-
enantiomer was obtained with ligand 4a whereas the (R)-
enantiomer was obtained with ligand 4b.
Ligands 4c and 4d , which contain a stereogenic bi-
naphthyl moiety, produced a high reaction rate and high
enantioselectivity (Table 2, entries 9 and 10). Thus,
ligand 4c, which has an (R)-binaphthyl moiety, led to an
ee of 97.6% (S), whereas diastereomer 4d , which has an
(S)-binaphthyl moiety, resulted in an ee of 98.3% (R).
This indicates that the sense of the enantiodiscrimination
is predominantly controlled by the conformation of the
binaphthyl at the phosphite moiety.
To verify the generality of these results, we performed
the Rh-catalyzed asymmetric hydrogenation of methyl
(Z)-(N)-acetylaminocinnamate 7 (Table 2, entries 11-14).
As expected, the results followed the same trend as those
of methyl (N)-acetamidoacrylate 6, but the enantiomeric
excesses were somewhat smaller and the reaction rates
were slightly slower. The configuration of the hydroge-
nated product was not affected by the phenyl group in 7.
The catalyst precursor with ligand 4b produced the
highest enantiomeric excess (98.8%; Table 2, entry 12).
Mech a n istic Con sid er a tion s. The commonly ac-
cepted mechanism of asymmetric diphosphine-Rh(I)-
catalyzed hydrogenation, first proposed by Landis and
Halpern, involves the reversible binding of the substrate
to the catalysts, followed by enantio- and rate-determin-
ing activation of H2 and subsequent fast reductive
elimination of the hydrogenated product (Scheme 3).20
In the past few years, this mechanism has proved valid
for other phosphorus ligands. Thus, RajanBabu et al.3b,c,21
% convnc
entry substrate solvent ligand TOFb (time/min) % eed
1
2
3
6
6
6
6
6
6
6
6
6
6
7
7
7
7
MeOH
CH2Cl2
Toluene
THF
4b
4b
4b
4b
4b
4b
4b
4a
4c
4d
4a
4b
4c
4d
85
100 (75)
91 (R)
40 100 (150) >99 (R)
12 100 (600)
64 100 (120)
41 100 (150) >99 (R)
40 100 (150) >99 (R)
43 100 (150) >99 (R)
97 (R)
92 (R)
4
5e
6f
7g
8
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
>1200
100(<5) 88.2 (S)
100 (20) 97.6 (R)
100 (20) 98.3 (S)
100 (10) 84.1 (S)
9
318
330
653
10
11
12
13
14
31 100 (180) 98.8 (R)
212 100 (30) 94.3 (R)
245 100 (30) 91 (S)
a
[Rh(cod)2]BF4 ) 0.01 mmol. 4/Rh )1.1. Substrate/Rh ) 100.
b
P ) 1 atm. CH2Cl2 ) 6 mL. T ) 25 °C. TOF in mol product ×
mol Rh-1 × h-1 measured by GC after 5 min. c Percent conversion
d
measured by GC. Percent enantiomeric excess measured by GC
using an L-Chiralsil-Val column. e Using preformed complex 5b
(0.01 mmol). f 4b/Rh ) 2. Using preformed [Rh(nbd)(4b)]BF4
g
(0.01 mmol).
attributed to different isomers in solution (Table 1). The
formation of different isomers may have been caused by
two diastereoisomers obtained from the atropisomerism
of the bisphenol in the phosphite moiety, different
conformers for the six-membered chelate ring, or a
combination of both. Although several conformers of the
six-membered chelate ring have been observed in similar
complexes with furanoside ligands,4b,7,16 the presence of
a single species in solution for complexes 5c and 5d
indicates that interconversion in the biphenyl is respon-
sible for the different isomers in solution for complexes
5a and 5b. This behavior contrasts with the inhibition
biphenyl interconversion upon coordination to the metal
center that is usually observed with diphosphites and
other phosphine-phosphite ligands.4b,11,17
Asym m etr ic Hyd r ogen a tion . In a first set of experi-
ments, we used the rhodium-catalyzed hydrogenation of
methyl (N)-acetamidoacrylate 6 to scope the potential of
ligands 4a -d for asymmetric catalysis. The reaction
proceeded smoothly at 1 bar of H2 at room temperature.
In general, the catalysts were prepared in situ by adding
the corresponding phosphine-phosphite ligand to [Rh-
(cod)2]BF4 as a catalyst precursor. The results are sum-
marized in Table 2.
The effects of different reaction parameters (e.g.,
solvent, catalyst preparation, and ligand-to-rhodium
ratio) were investigated for the catalytic precursor con-
taining ligand 4b. The results showed that the efficiency
of the process was affected by the nature of the solvent
(Table 2, entries 1-4). Although rhodium-catalyzed
asymmetric hydrogenation reactions are usually per-
formed in methanol, the reaction in this solvent showed
(18) Partial hydrolysis of the phosphite moiety was observed by 31
P
NMR when complex 5a was stirred in methanol-d4 under hydrogen
atmosphere.
(19) Bo¨rner, A.; Heller, D. Tetrahedron Lett. 2001, 42, 223.
(20) Landis, C.; Halpern, J . J . Am. Chem. Soc. 1987, 109, 1746.
(21) RajanBabu, T. V.; Radetich, B.; You, K. K.; Ayers, T. A.;
Casalnuovo, A. L.; Calabrese, J . C. J . Org. Chem. 1999, 64, 3429.
(17) Buisman, G. J . H.; van der Veen, L. A.; Klootwijk, A.; de Lange,
W. G. J .; Kamer, P. C. J .; van Leeuwen, P. W. N. M.; Vogt, D.
Organometallics 1997, 16, 2929.

Highly Enantioselective Rh-Catalyzed Hydrogenation
J . Org. Chem., Vol. 66, No. 25, 2001 8367
Ta ble 3. Hyd r ogen P r essu r e Dep en d en cy on th e
Hyd r ogen a tion Rea ction Ra tea
PH2 (atm)
[5b] (mM)
TOFb
1
2
3
1.67
1.67
1.67
40
74
115
a
b
Conditions: T ) 25 °C, [6]o ) 0.17 M. TOF in mol product ×
mol Rh-1 × h-1 measured by GC after 5 min.
This kinetic study indicates that the rate-determining
step is the activation of hydrogen. This feature matches
with the expected Landis-Halpern mechanism. How-
ever, the so-called “dihydride mechanism”,22 in which the
substrate coordination takes place after the activation
of hydrogen, cannot be excluded because the same kinetic
pattern would be obtained if the addition of hydrogen
were the rate-determining step.
F igu r e 1. Variation of the conversion over time. Conditions:
T ) 25 °C, pH2 ) 1 bar, [5b] ) 1.67 mM, [6]o ) 0.17 M.
To obtain more insight into the sequence of the
catalytic cycle, we studied the species formed under
hydrogenation conditions. First we investigated the
reactivity of catalyst precursor 5b with molecular hydro-
gen. The VT-1H NMR spectra did not show the formation
of hydride species when catalyst precursor 5b was
dissolved in dichoromethane-d2 and pressurized with 1.1
bar of H2. Also, the VT-31P NMR spectra did not show
any new signal, which indicated that catalyst precursor
5b did not react with molecular hydrogen under these
conditions. After dehydroamino acid derivative 6 was
added, two new double doublets at 145.8 (P2) and 16.3
(P1) ppm were observed in the 31P{1H} NMR spectrum.
These new signals were assigned to the cationic rhodium
complex [Rh (4b)(6)]+. The P1-P2 coupling constant is
63 Hz, whereas J {P1-Rh} and J {P2-Rh} are 158 and
236 Hz, respectively. VT-31P{1H} NMR spectra between
+25 and -80 °C showed that there was only one
diastereomer. Also, neither the [Rh(4b)(6)H2]BF4 species
nor the [Rh(4b)(alkyl)H]BF4 species were observed under
hydrogenation conditions.
F igu r e 2. Variation of the rhodium concentration. Condi-
tions: T ) 25 °C, pH2 ) 1 bar, [6]o ) 0.17 M.
Sch em e 3
The NMR results using Rh-phosphine-phosphite
catalyst precursor 5b agreed with the well-established
Landis-Halpern mechanism via Rh enamide species
(Scheme 3). However, this evidence is not conclusive
because species [Rh(4b)(6)H2]BF4 may have been present
in amounts that were below the detection limit of the
NMR equipment.
The origin of the enantioselectivity can be explained
by analyzing the structure of the square-planar enamide
intermediates and their reactivity toward hydrogen.23
The coordination mode of the enamide is based on both
electronic and steric properties, so we investigated whether
it is affected by the different steric properties of ligands
4a and 4b. Thus, when complex 5a was treated with
enamide 6 in the same way as complex 5b (vide supra),
two new double doublets at 146.3 (P2) and 15.1 (P1) ppm
attributed to [Rh(4a )(6)]+ were observed in the 31P{1H}
NMR. The chemical shifts and the coupling constants
(J {P1-P2}) 61.2 Hz, J {P1-Rh}) 154.6 Hz, J {P2-Rh})
237.4 Hz) are similar to those for complex [Rh(4b)(6)]+.
Also, VT-31P{1H} NMR showed that only one diasteroi-
and we4b recently provided experimental support for a
similar pathway using cationic Rh(I) diphosphinite and
diphosphite precursors, respectively. However, the pres-
ence of two different functionalities in these Rh(I) phos-
phine-phosphite catalysts may affect the catalytic se-
quence. To learn more about the catalytic cycle, we made
a kinetic study and investigated the species formed with
catalyst precursor 5b under hydrogenation conditions.
The rate-controlling steps of the hydrogenation reac-
tion of methyl (N)-acetamidoacrylate 6 using catalyst
precursor 5b were determined by a kinetic study in which
we investigated the concentration dependency of the
reaction rates of all the reactants (e.g., substrate and
rhodium concentration and hydrogen pressure). The
substrate concentration dependency was established by
following the conversion over time. The graph of conver-
sion versus time clearly shows a zero-order dependency
on substrate concentration (Figure 1).
To study the effect of rhodium concentration on reac-
tion rate, we varied this concentration between 0.83 and
4.17 mM (Figure 2). The rate (TOF) of hydrogenated
product formation is linearly proportional to the rhodium
concentration, which indicates a first-order dependency.
The hydrogen pressure was varied between 1 and 3
bar. The results in Table 3 show a first-order dependency.
(22) For instance, see: (a) Young, J . F.; Osborn, J . A.; J ardine, F.
H.; Wilkinson, G. Chem. Commun. 1965, 131. (b) Gridnev, I. D.;
Higashi, N.; Asakura, K.; Imamoto, T. J . Am. Chem. Soc. 2000, 122,
7183.
(23) (a) Felgus, S.; Landis, C. R. J . Am. Chem. Soc. 2000, 122, 12714.
(b) Landis, C. R.; Feldgus, S. Angew. Chem., Int. Ed. 2000, 39, 2863
and references therein.

8368 J . Org. Chem., Vol. 66, No. 25, 2001
Pa`mies et al.
purified by column chromatography (eluent: ether, Rf 0.90).
Yield: 590 mg (80%) of a white solid. 31P NMR, δ (CDCl3):
-22.3 (s).
somer was present. This suggests that for our Rh-
phosphine-phosphite, the coordination mode of the
enamide is mainly controlled by the electronic properties
of the ligand. In summary, the formation of the diaster-
eomeric Rh(III)-dihydride complexes, and therefore that
of the hydrogenated product, is mainly controlled by the
steric hindrance of the phosphite moiety, which controls
the rotation of the substrate with respect to the ligand
that follows the oxidative addition of H2. This agrees with
our hydrogenation results, which indicated that the sense
of the enantioselectivity was controlled by the phosphite
moiety (vide supra).
3-(1,1′-Bip h en yl-2,2′-d iyl)p h osp h ite-5-d eoxy-1,2-O-iso-
p r op ylid en e-5-d ip h en ylp h osp h in e-r-D-xylofu r a n ose (4a ).
In situ formed phosphorochloridite 3a (1 mmol) was dissolved
in toluene (8 mL), to which pyridine (0.25 mL, 6 mmol) was
added. 5-Deoxy-1,2-O-isopropylidene-5-diphenylphosphine-R-
D-xylofuranose 2 (0.25 g, 0.7 mmol) was azeotropically dried
with toluene (3 × 1 mL) and dissolved in toluene (8 mL), to
which pyridine (0.25 mL, 6 mmol) was added. This solution
was added in 30 min to the solution of 3a at room temperature.
The reaction mixture was stirred overnight at room temper-
ature, and the pyridine salts were removed by filtration.
Evaporation of the solvent gave a white foam, which was
purified by flash column chromatography (eluent: toluene, Rf
0.23). Yield: 244 mg (61%) of a white solid. Anal. Calcd for
Con clu sion s
C
32H30O6P2: C, 67.13; H, 5.28. Found: C, 67.45; H, 5.42. 31P
A series of novel phosphine-phosphite ligands 4a -d
that contain a furanoside as a simple but highly effective
chiral backbone was reported. These ligands, which are
easily prepared in a few steps from the readily available
D-(+)-xylose, are very effective in the Rh(I)-catalyzed
asymmetric hydrogenation of R,â-unsaturated carboxylic
acid derivatives. The results show that the presence of
bulky tert-butyl groups in the ortho positions of the
biphenyl moiety or the presence of a stereogenic binaph-
thyl moiety had an extremely positive effect on enanti-
oselectivity. The absolute configuration of the major
enantiomer of the hydrogenated product is mainly con-
trolled by the configuration of the phosphite moiety.
Kinetic studies indicates that the rate dependence is first
order in rhodium and hydrogen pressure and zero order
in enamide concentration. NMR studies on the interme-
diates formed under hydrogenation conditions indicate
that the [Rh(P1-P2)(enamide)]BF4 species is the resting
state. The combination of high enantioselectivities and
activities under mild conditions (room temperature, 1 bar
of H2) open up a new class of ligands for asymmetric
hydrogenation.
NMR, δ (CDCl3): -18.3 (d, 1P, J P-P ) 17 Hz), 148.2 (d, 1P,
1
J P-P ) 17 Hz). H NMR, δ (CDCl3): 1.22 (s, 3H, CMe2), 1.30
3
2
(s, 3H, CMe2), 2.38 (dd, 1H, H-5′, J 5′-4 ) 8.1 Hz, J 5′-5 ) 13.2
3
2
Hz), 2.57 (dd, 1H, H-5, J 5-4 ) 6.6 Hz, J 5-5′ ) 13.2 Hz), 4.12
3
(m, 1H, H-4), 4.62 (d, 1H, H-2, J 2-1 ) 3.2 Hz), 4.73 (dd, 1H,
3
3
3
H-3, J 3-4 ) 2.2 Hz, J 3-P ) 9.6 Hz), 5.87 (d, 1H, H-1, J 1-2
)
3.2 Hz), 7.0-7.5 (m, 13H, Ar). 13C NMR, δ (CDCl3): 26.2
(CMe2), 26.4 (CMe2), 27.1 (d, C-5, J C-P ) 14.8 Hz), 77.5 (dd,
C-4, J C-P ) 4.5 Hz, J C-P ) 19.4 Hz), 78.1 (dd, C-3, J C-P ) 5.2
Hz, J C-P ) 11.4 Hz), 84.6 (d, C-2, J C-P ) 2.3 Hz), 105.1 (C-1),
111.8 (CMe2), 122.0, 125.4, 128.5, 128.6, 128.7, 128.9, 129.3,
130.0, 130.1, 132.8, 133.9 (Ar).
3-(3,3′,5,5′-Tetr a -ter t-bu tyl-1,1′-bip h en yl-2,2′-d iyl)p h os-
ph ite-5-deoxy-1,2-O-isopr opyliden e-5-diph en ylph osph in e-
r-D-xylofu r a n ose (4b). Treatment of in situ formed phos-
phorochloridite 3b (1 mmol) and 5-deoxy-1,2-O-isopropylidene-
5-diphenylphosphine-R-D-xylofuranose 2 (0.25 g, 0.7 mmol) as
described for compound 4a afforded compound 5b, which was
purified by flash chromatography (eluent: toluene, Rf 0.50).
Yield: 435 mg (78%) of a white solid. Anal. Calcd for
C
48H62O6P2: C, 72.34; H, 7.84. Found: C, 71.98; H, 7.94. 31P
NMR, δ (CDCl3): -21.5 (d, 1P, J P-P ) 11.9 Hz), 144.7 (d, 1P,
J P-P ) 11.9 Hz). 1H NMR, δ (CDCl3): 1.11 (s, 3H, CMe2), 1.23
(s, 3H, CMe2), 1.33 (s, 9H, CH3, t-Bu), 1.35 (s, 9H, CH3, t-Bu),
1.43 (s, 9H, CH3, t-Bu), 1.51 (s, 9H, CH3, t-Bu), 2.32 (ddd, 1H,
3
2
H-5′, J 5′-4 ) 7.2 Hz, J 5′-5 ) 13.5 Hz, J 5′-P ) 5.2 Hz), 2.52
(ddd, 1H, H-5, 3J 5-4 ) 7.5 Hz, 2J 5-5′ ) 13.5 Hz, J 5-P ) 1.8 Hz),
4.02 (m, 1H, H-4), 4.12 (d, 1H, H-2, J 2-1 ) 3.6 Hz), 4.62 (dd,
Exp er im en ta l Section
3
Gen er a l Rem a r k s. All reactions and purifications were
carried out under dry argon atmosphere in oven-dried glass-
ware. Solvents were purified by standard procedures. Com-
pounds [Rh(cod)2]BF4,24 oxetane 1,12 and phosphorochloridites
3a -d 14 were prepared by previously described methods.
Elemental analyses were performed on a Carlo Erba EA-1108
instrument. 1H, 13C{1H}, and 31P{1H} NMR spectra were
recorded on a Varian Gemini 300 MHz spectrometer. Chemical
shifts were relative to SiMe4 (1H and 13C) as internal standard
or H3PO4 (31P) as external standard. All assignments in NMR
spectra were determined by COSY and HETCOR spectra.
1H, H-3, 3J 3-4 ) 2.6 Hz, 3J 3-P ) 8.4 Hz), 5.65 (d, 1H, H-1, 3J 1-2
) 3.6 Hz), 7.1-7.5 (m, 9H, Ar). 13C NMR, δ (CDCl3): 26.2
(CMe2), 26.4 (CMe2), 27.4 (d, C-5, J C-P ) 14.8 Hz), 31.2 (CH3,
t-Bu), 31.4 (CH3, t-Bu), 31.5 (CH3, t-Bu), 34.5 (C, t-Bu), 35.3
(C, t-Bu), 77.4 (dd, C-4, J C-P ) 5.1 Hz, J C-P ) 18.4 Hz), 77.8
(d, C-3, J C-P ) 5.8 Hz), 84.3 (d, C-2, J C-P ) 2.3 Hz), 104.3 (C-
1), 111.4 (CMe2), 124.3, 126.6, 128.4, 128.6, 128.8, 132.6, 132.8,
133.0, 133.2, 138.3, 138.6, 140.1, 140.4, 145.6, 146.8 (Ar).
3-[(R)-1,1′-Bin a p h th yl-2,2′-d iyl)p h osp h ite-5-d eoxy-1,2-
O-isop r op ylid e n e -5-d ip h e n ylp h osp h in e -r-D-xylofu r a -
n ose (4c). Treatment of in situ formed phosphorochloridite
3c (1 mmol) and 5-deoxy-1,2-O-isopropylidene-5-diphenylphos-
phine-R-D-xylofuranose 2 (0.25 g, 0.7 mmol) as described for
compound 4a afforded compound 5c, which was purified by
flash chromatography (eluent: toluene, Rf 0.17). Yield: 279
mg (59%) of a white solid. Anal. Calcd for C40H34O6P2: C, 71.42;
H, 5.09. Found: C, 72.02; H, 4.89. 31P NMR, δ (CDCl3): -22.5
Hydrogenation reactions were performed in
a previously
described hydrogen vacuum line.25 Gas chromatographic analy-
ses were run on a Hewlett-Packard HP 5890A instrument
(split/splitless injector, Permabond L-Chirasil-Val, 25 m col-
umn, internal diameter 0.25 mm, carrier gas: 150 kPa He,
F.I.D. detector) equipped with a Hewlett-Packard HP 3396
series II integrator.
1
(d, 1P, J P-P ) 33 Hz), 150.7 (d, 1P, J P-P ) 33 Hz). H NMR, δ
5-Deoxy-1,2-O-isop r op ylid en e-5-d ip h en ylp h osp h in e-r-
D-xylofu r a n ose (2). Potassium diphenylphosphine (2.2 mmol)
was added to a solution of oxetane 1 (0.35 g, 2 mmol) in
dimethylformamide (8 mL). The mixture was stirred overnight
at room temperature. Water (50 mL) was then added, and the
product was extracted with dichloromethane (3 × 50 mL). The
organic phase was then evaporated, and the residue was
(CDCl3): 1.30 (s, 3H, CMe2), 1.32 (s, 3H, CMe2), 2.37 (ddd, 1H,
3
2
H-5′, J 5′-4 ) 8.4 Hz, J 5′-5 ) 13.5 Hz, J 5′-P ) 5.1 Hz), 2.55
(ddd, 1H, H-5, 3J 5-4 ) 6.6 Hz, 2J 5-5′ ) 13.5 Hz, J 5-P ) 3.6 Hz),
4.11 (m, 1H, H-4), 4.72 (d, 1H, H-2, J 2-1 ) 3.6 Hz), 4.76 (dd,
3
3
3
1H, H-3, J 3-4 ) 2.4 Hz, J 3-P ) 10.1 Hz), 5.88 (d, 1H, H-1,
3J 1-2 ) 3.6 Hz), 7.0-7.5 (m, 15H, Ar). 13C NMR, δ (CDCl3):
26.4 (CMe2), 26.5 (CMe2), 26.9 (d, C-5, J C-P ) 14.3 Hz), 77.6
(dd, C-4, J C-P ) 5.0, 20.6 Hz), 78.3 (dd, C-3, J C-P ) 5.1, 15.9
Hz), 84.6 (d, C-2, J C-P ) 3.4 Hz), 104.4 (C-1), 112.0 (CMe2),
121.4, 121.7, 125.0, 125.2, 126.3, 127.0, 128.4, 128.5, 128.6,
129.0, 129.9, 130.4, 132.5, 132.7, 132.9, 133.2 (Ar).
(24) Green, M.; Kuc, T. A.; Taylor, S. H. J . Chem. Soc. A 1971, 2334.
(25) Cativiela, C.; Ferna´ndez, J .; Mayoral, J . A.; Mele´ndez, E.; Uso´n,
R.; Oro, L. A.; Ferna´ndez, M. J . J . Mol. Catal. 1992, 16, 19.

Highly Enantioselective Rh-Catalyzed Hydrogenation
J . Org. Chem., Vol. 66, No. 25, 2001 8369
3-[(S)-1,1′-Bin a p h th yl-2,2′-d iyl)p h osp h ite-5-d eoxy-1,2-
O-isop r op ylid e n e -5-d ip h e n ylp h osp h in e -r-D-xylofu r a -
n ose (4d ). Treatment of in situ formed phosphorochloridite
3d (1 mmol) and 5-deoxy-1,2-O-isopropylidene-5-diphenylphos-
phine-R-D-xylofuranose 2 (0.25 g, 0.7 mmol) as described for
compound 4a afforded compound 5d , which was purified by
flash chromatography (eluent: toluene, Rf 0.19). Yield: 212
mg (45%) of a white solid. Anal. Calcd for C40H34O6P2: C, 71.42;
H, 5.09. Found: C, 71.79; H, 4.88. 31P NMR, δ (CDCl3): -22.5
(d, 1P, J P-P ) 15.5 Hz), 148.4 (d, 1P, J P-P ) 15.5 Hz). 1H NMR,
δ (CDCl3): 1.19 (s, 3H, CMe2), 1.36 (s, 3H, CMe2), 2.45 (ddd,
(dd, 1P, P1, J P1-P2 ) 43.9 Hz, J P1-Rh ) 147.9 Hz), 130.4 (dd,
1P, P2, J P2-P1 ) 43.9 Hz, J P2-Rh ) 268.2 Hz). Minor isomer:
8.8 (dd, 1P, P1, J P1-P2 ) 42.1 Hz, J P1-Rh ) 141.9 Hz), 141.0
(bd, 1P, P2, J P2-Rh ) 248.9 Hz).
[Rh (4c)(cod )]BF 4 (5c). Yield: 45 mg (94%). Anal. Calcd
for C48H46O6P2BF4Rh: C, 59.43; H, 4.74. Found: C, 59.68; H,
4.85. 31P NMR, δ (CD2Cl2): 14.9 (dd, 1P, P1, J P1-P2 ) 48.2 Hz,
J P1-Rh ) 141.3 Hz), 141.3 (dd, 1P, P2, J P1-P2 ) 48.2 Hz, J P2-Rh
1
) 262.2 Hz). H NMR, δ (CD2Cl2): 1.19 (s, 3H, CH3), 1.39 (s,
3H, CH3), 1.9-2.6 (m, 8H, CH2), 3.12 (m, 1H, H-5′), 3.50 (m,
3
1H, H-5), 4.47 (d, 1H, H-2, J 2-1 ) 3.6 Hz), 4.71 (m, 1H, H-4),
3
2
1H, H-5′, J 5′-4 ) 8.4 Hz, J 5′-5 ) 13.5 Hz, J 5′-P ) 1.8 Hz),
4.83 (m, 1H, CHd), 4.95 (m, 1H, CHd), 5.28 (m, 1H, CHd),
2.63 (ddd, 1H, H-5, 3J 5-4 ) 6.6 Hz, 2J 5-5′ ) 13.5 Hz, J 5-P ) 2.1
5.36 (dd, 1H, H-3, 3J 3-4 ) 2.1 Hz, J 3-P ) 13.9 Hz), 5.72 (d, 1H,
3
3
Hz), 4.15 (m, 1H, H-4), 4.45 (d, 1H, H-2, J 2-1 ) 3.3 Hz), 4.68
H-1, J 1-2 ) 3.6 Hz), 5.80 (m, 1H, CHd), 7.0-8.2 (m, 22H,
(dd, 1H, H-3, 3J 3-4 ) 2.4 Hz, 3J 3-P ) 9.0 Hz), 5.79 (d, 1H, H-1,
3J 1-2 ) 3.3 Hz), 7.0-7.5 (m, 15H, Ar). 13C NMR, δ (CDCl3):
26.0 (CMe2), 26.3 (CMe2), 27.1 (d, C-5, J C-P ) 14.8 Hz), 77.4
Ar).
[Rh (4d )(cod )]BF 4 (5d ). Yield: 43 mg (90%). Anal. Calcd
for C48H46O6P2BF4Rh: C, 59.43; H, 4.74. Found: C, 59.14; H,
5.01. 31P NMR, δ (CD2Cl2): 15.2 (dd, 1P, P1, J P1-P2 ) 49.0 Hz,
J P1-Rh ) 140.7 Hz), 140.7 (dd, 1P, P2, J P1-P2 ) 49.0 Hz, J P2-Rh
(dd, C-4, J C-P ) 5.6 Hz, J C-P ) 19.4 Hz), 78.2 (dd, C-3, J C-P
)
5.1 Hz, J C-P ) 11.4 Hz), 84.5 (d, C-2, J C-P ) 2.9 Hz), 104.4
(C-1), 111.7 (CMe2), 121.7, 125.0, 125.3, 126.3, 126.4, 128.4,
128.5, 128.6, 128.8, 128.9, 130.2, 130.5, 132.7, 132.9, 134.0 (Ar).
P r ep a r a tion of Rh od iu m Ca tion ic P r ecu r sor s. In a
general procedure, phosphine-phosphite ligand (0.05 mmol)
was added to a solution of [Rh(cod)2]BF4 (20.2 mg, 0.05 mmol)
in dichloromethane (2 mL). After 5 min, the desired products
were obtained by precipitation with hexane as yellow solids.
[Rh (4a )(cod )]BF 4 (5a ). Yield: 41 mg (95%). Anal. Calcd
for C40H42O6P2BF4Rh: C, 55.22; H, 4.83. Found: C, 55.53; H,
4.96. 31P NMR, δ (CD2Cl2): 13.9 (bd, 1P, P1, J P1-Rh ) 145.2
Hz), 135.1 (dd, 1P, P2, J P2-P1 ) 43.3 Hz, J P2-Rh ) 260.5 Hz).
1H NMR, δ (CD2Cl2): 1.12 (s, 3H, CH3), 1.35 (s, 3H, CH3), 2.0-
2.5 (m, 8H, CH2, cod), 2.98 (m, 2H, H-5′, H-5), 3.41 (d, 1H,
1
) 260.6 Hz). H NMR, δ (CD2Cl2): 1.17 (s, 3H, CH3), 1.42 (s,
3H, CH3), 1.8-2.6 (m, 8H, CH2), 3.10 (m, 1H, H-5′), 3.39 (m,
3
1H, H-5), 4.37 (d, 1H, H-2, J 2-1 ) 3.2 Hz), 4.64 (m, 1H, H-4),
4.79 (m, 1H, CHd), 4.89 (m, 1H, CHd), 5.31 (m, 1H, CHd),
5.37 (dd, 1H, H-3, 3J 3-4 ) 2.2 Hz, J 3-P ) 14.4 Hz), 5.63 (d, 1H,
3
H-1, J 1-2 ) 3.2 Hz), 5.69 (m, 1H, CHd), 7.2-8.2 (m, 22H,
Ar).
Asym m etr ic Hyd r ogen a tion Rea ction s. In a typical run,
a Schlenk flask was filled with a dichloromethane solution (6
mL) of substrate (1 mmol), [Rh(cod)2]BF4 (4.95 mg, 0.01 mmol),
and ligand (0.011 mmol). This was then purged three times
with H2 and vacuum. The reaction mixture was then shaken
under H2 (1 atm) at 298 K. To remove the catalyst, the solution
was placed on a short silica gel column and eluted with CH2-
Cl2. Conversion and enantiomeric excesses were determined
by gas chromatography.
3
H-2, J 2-1 ) 3.6 Hz), 3.79 (m, 1H, H-4), 4.54 (m, 1H, CHd,
cod), 5.01 (m, 1H, H-3), 5.17 (m, 1H, CHd, cod), 5.48 (m, 1H,
H-1), 5.69(m, 1H, CHd, cod), 5.75 (m, 1H, CHd, cod), 7.0-8.0
(m, 18H, Ar). 31P NMR, δ (CD2Cl2, 193K): Major isomer: 18.6
(dd, 1P, P1, J P1-P2 ) 45.6 Hz, J P1-Rh ) 148.3 Hz), 132.8 (dd,
1P, P2, J P2-P1 ) 45.6 Hz, J P2-Rh ) 266.8 Hz). Minor isomer:
7.5 (dd, 1P, P1, J P1-P2 ) 41.6 Hz, J P1-Rh ) 142.0 Hz), 143.7
(dd, 1P, P2, J P2-P1 ) 41.6 Hz, J P2-Rh ) 253.2 Hz).
In Situ NMR Ch a r a cter iza tion Exp er im en ts. In
a
typical experiment, a sapphire tube (Φ ) 10 mm) was filled
under argon with a solution of [Rh(cod)(4b)]BF4 (0.02 mmol)
and methyl N-acetamidoacrylate 6 (0.30 mmol) in dichlo-
romethane-d2 (1.5 mL). The tube was purged twice and
pressurized to 1.2 bar of H2. The reaction was followed under
H2 pressure.
[Rh (4b)(cod )]BF 4 (5b). Yield: 49 mg (91%). Anal. Calcd
for C56H74O6P2BF4Rh: C, 61.46; H, 6.78. Found: C, 61.74; H,
6.42. 31P NMR, δ (CD2Cl2): 15.3 (bd, 1P, P1, J P1-Rh ) 144.3
Hz), 131.4 (dd, 1P, P2, J P2-P1 ) 40.2 Hz, J P2-Rh ) 260.8 Hz).
1H NMR, δ (CD2Cl2): 0.90 (s, 3H, CH3), 0.97 (s, 3H, CH3), 1.21
(s, 9H, CH3, t-Bu), 1.23 (s, 9H, CH3, t-Bu), 1.46 (s, 9H, CH3,
t-Bu), 1.55 (s, 9H, CH3, t-Bu), 2.0-2.5 (m, 8H, CH2, cod), 2.93
Ack n ow led gm en t. We thank the Spanish Ministe-
rio de Educacio´n, Cultura y Deporte and the Generalitat
de Catalunya (CIRIT) for their financial support (PB97-
0407-CO5-01) and for awarding a research grant (to
O.P.). We thank Prof. P. W. N. M. van Leeuwen for his
suggestions and comments.
3
(m, 2H, H-5′, H-5), 3.30 (d, 1H, H-2, J 2-1 ) 3.9 Hz), 3.82 (m,
3
1H, H-4), 4.42 (m, 1H, CHd, cod), 4.60 (dd, 1H, H-3, J 3-4
)
2.1 Hz, J 3-P ) 9.9 Hz), 4.71 (m, 1H, CHd, cod), 5.18 (m, 1H,
H-1), 5.41(m, 1H, CHd, cod), 5.53 (m, 1H, CHd, cod), 7.0-7.7
(m, 14H, Ar). 31P NMR, δ (CD2Cl2, 193K): Major isomer: 19.7
J O0157122